Datasets:

BioLaySumm
/

BioLaySumm2025-PLOS

Dataset card Data Studio Files Files and versions Community

Split (3)

train · 24.8k rows

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
Leptospirosis is one of the most important neglected tropical bacterial diseases in Latin America and the Caribbean . However , very little is known about the circulating etiological agents of leptospirosis in this region . In this study , we describe the serological and molecular features of leptospires isolated from 104 leptospirosis patients in Guadeloupe ( n = 85 ) and Martinique ( n = 19 ) and six rats captured in Guadeloupe , between 2004 and 2012 . Strains were studied by serogrouping , PFGE , MLVA , and sequencing 16SrRNA and secY . DNA extracts from blood samples collected from 36 patients in Martinique were also used for molecular typing of leptospires via PCR . Phylogenetic analyses revealed thirteen different genotypes clustered into five main clades that corresponded to the species: L . interrogans , L . kirschneri , L . borgpetersenii , L . noguchi , and L . santarosai . We also identified L . kmetyi in at least two patients with acute leptospirosis . This is the first time , to our knowledge , that this species has been identified in humans . The most prevalent genotypes were associated with L . interrogans serovars Icterohaemorrhagiae and Copenhageni , L . kirschneri serovar Bogvere , and L . borgpetersenii serovar Arborea . We were unable to identify nine strains at the serovar level and comparison of genotyping results to the MLST database revealed new secY alleles . The overall serovar distribution in the French West Indies was unique compared to the neighboring islands . Typing of leptospiral isolates also suggested the existence of previously undescribed serovars . Leptospirosis is an emerging zoonosis with a worldwide distribution . The World Health Organization ( WHO ) estimates that there are over 1 , 700 , 000 severe cases of leptospirosis worldwide , with an higher incidence in impoverished populations in developing countries and tropical regions [1]–[3] . The disease is transmitted during direct contact with animal reservoirs or , more frequently , water and soil contaminated with their urine [4] . Leptospirosis is found in rural regions because of the higher risk of exposure to animal reservoirs [5] , [6] and also in urban slums where inadequate sanitation provides the conditions for rat-borne transmission of the disease [7] , [8] . Outbreaks may occur after heavy seasonal rainfall and extreme climatic events such as tropical storms and hurricanes [8] , [9] . Leptospirosis causes a broad spectrum of symptoms from subclinical infection to multiple organ failure with a mortality rate of 10 to 50% [3] . Leptospirosis is one of the most important neglected tropical bacterial diseases in Latin America and the Caribbean [10] . This includes the French West Indies which consist of the Caribbean islands of Guadeloupe and Martinique , both French overseas departments . Climate-related changes , such as El Niño events , and periods of heavy rainfall may influence the incidence of leptospirosis in this region [11] , [12] . The mean incidence of leptospirosis in the French West Indies is higher than 10/100 , 000 inhabitants peaking at ∼39/100 , 000 inhabitants in Guadeloupe in 2011 , which is 100 times higher than that in Mainland France ( data from the National Reference Center for Leptospirosis , Institut Pasteur ) . Pathogenic Leptospira encompass nine species with more than 300 serovars which are the etiological agents of leptospirosis [13] . The taxonomy of the genus Leptospira has been both complex and controversial . Leptospiral serovars are defined by the cross agglutinin absorption test ( CAAT ) which uses polyclonal antibodies against the lipopolysaccharides ( LPS ) [4] , [14] . However , this test is fastidious to perform and it is restricted to a few reference laboratories . Although the term serogroup has no taxonomic value , it has been used to define a group of antigenically related serovars which can be identified by microscopic agglutination test ( MAT ) . Advancement in molecular techniques has allowed the speciation of members of the genus Leptospira . A significant outcome of the genetic classification scheme was the finding that a serovar may belong to different species [14] . With the emergence of molecular typing methods , it appears that the concept of a “serovar” is no longer fully satisfactory as it may fail to define epidemiologically important strains or genotypes . Despite its medical significance , the isolation of clinical Leptospira strains is rare due to the fastidious growth in culture of this species , and poor awareness of the disease . Detailed characterization of Leptospira isolates is important for understanding the epidemiology of leptospirosis . Leptospira serovars can be prevalent in a particular geographical area and/or associated with a restricted number of animal reservoirs . Local Leptospira isolates can serve as antigens for the serodiagnosis of leptospirosis . The diverse distributions of Leptospira serovars and genotypes may have implications for vaccine design and efficacy . The main function of the National Reference Center ( NRC ) for Leptospirosis , which is also a WHO Collaborating Center , at the Institut Pasteur of Paris is the surveillance of human leptospirosis . This includes the collection of diagnostic data from laboratories around the country , including French overseas territories . The NRC is the only laboratory in France that can confirm leptospirosis diagnosis by means of the MAT , which remains the gold standard for the serological diagnosis of leptospirosis , with an extended panel of antigens . The NRC also identifies clinical isolates both from mainland France and from French overseas territories . As part of a long-term typing project , more than three-quarters of all the Leptospira isolates received from Guadeloupe and Martinique were systematically fingerprinted so as to identify the strains circulating in this region the Americas . The Leptospira cultures from human patients analyzed in this study were previously isolated by the University Hospitals of Pointe à Pitre ( Guadeloupe ) and Fort de France ( Martinique ) and the National Reference Center for Leptospirosis ( Institut Pasteur ) as part of the national surveillance of leptospirosis . The strains and DNAs derived from these cultures were analyzed anonymously for this research study . All serum samples were initially sampled for diagnostic purpose , and archived at the National Reference Center for Leptospirosis ( Institut Pasteur ) . All sera are de-linked from the patients from whom they originated and analyzed anonymously if used in any research study . The study protocol was approved by the ethical committees of the University Hospitals of Pointe à Pitre ( Guadeloupe ) and Fort de France ( Martinique ) and the CNIL ( Commission Nationale Informatique et Liberté ) . This study was part of a protocol approved by the Institut Pasteur and the French Ministry for Education & Research French Ministry for Education & Research ( protocol # AC- DC-2010-1197 ) . Rats are listed as invasive mammals on the French West Indies [15] and authorization ( through arrêté préfectoral ) are regularly published , in agreement with the « Fédération Départementale des Groupements de Défense contre les Ennemis des Cultures » for their captures . Rats were captured in 2002–2003 during a study on schistosomiasis [16] , which was supported by the CNRS ( PNDBE ) , the MENRT ( PRFMMIP 95 ) and the French Ministry of Ecology and Sustainable Development ( Contract CV 02000071 , MEDD , programme “Ecosystemes Tropicaux” ) . Capture and euthanasia of rats was performed by Dr . A . Theron ( University of Perpignan ) who was accredited to experiment on rodent ( authorization n° C 66 . 11 . 02 by the Préfecture des Pyrénées Orientales ) . Studies on rats were performed in accordance with the European Union legislation ( Directive 86/609/EEC ) . Serum samples from suspected cases of leptospirosis were subjected to the microscopic agglutination test ( MAT ) at the National Reference Center for Leptospirosis ( NRC ) at the Institut Pasteur ( Paris , France ) . MAT was performed using 24 leptospiral antigens ( Table S1 ) . A high agglutination titer of the serum with one particular serogroup is taken to identify the presumptive serogroup of the infecting bacterium . For patients presenting symptoms during the first week of infection , total genomic DNA was extracted from plasma collected into EDTA tubes and tested for the presence of pathogenic Leptospira by real-time PCR [17] . Molecular Biology Grade Water ( EUROBIO , Les Ulis , France ) was used for PCR . Reactions with no template DNA were included as negative controls in each PCR experiment . For patients testing positive by PCR , acute and , if possible , convalescent serum samples were collected for serological testing . Leptospirosis cases were defined as having clinical signs and symptoms consistent with leptospirosis and a single MAT titer ≥1/400 for a pathogenic serogroup or detection of pathogenic Leptospira by PCR or culture . A total of 104 clinical isolates of Leptospira isolated from patients in Guadeloupe ( 85 ) and Martinique ( 19 ) between 2004 and 2012 were studied . Leptospira was cultured by inoculating plasma prepared from heparinized blood from patients into EMJH liquid medium [18] at the University Hospitals of Pointe à Pitre ( Guadeloupe ) and Fort de France ( Martinique ) . Leptospires positive cultures were then sent to the NRC for Leptospirosis ( Institut Pasteur , Paris , France ) for typing . Six isolates collected from kidney tissues of Rattus rattus captured in Guadeloupe ( mangrove area of Morne à L'eau ) in 2002–2003 ( André Théron , University of Perpignan ) were also included in the study . Reference strains from the collection maintained by the NRC for Leptospirosis were used for comparisons ( Table S2 ) . DNA was also isolated from the blood of 36 additional patients from Martinique who tested positive for leptospirosis by PCR during the study . The microscopic agglutination test ( MAT ) was used for antigenic characterization of Leptospira isolates , with a standard battery of rabbit antisera against reference serovars representing the 24 serogroups as previously described [18] . Mice monoclonal antibodies F70 C14-10 and F70 C24-20 ( WHO/FAO/OIE and National Collaborating Centre for Reference and Research on Leptospirosis , Royal Tropical Institute , Amsterdam , The Netherlands ) , which react against the serovars Icterohaemorrhagiae and Copenhageni respectively , were also used for some strains as previously described [19] . Genomic DNA was extracted from EMJH cultures or from human plasma ( see above ) . DNA was amplified using Taq polymerase ( GE Healthcare ) under standard conditions . For species identification , the rrs gene was amplified with the primers LA ( 5′-GGCGGCGCGTCTTAAACATG-3′ ) and LB ( 5′-TTCCCCCCATTGAGCAAGATT-3′ ) , and when necessary , by nested primers LC ( 5′-CAAGTCAAGCGGAGTAGCAA-3′ ) and RS4 ( 5′- TCTTAACTGCTGCCTCCCGT-3′ ) [20] , [21] . Part of the secY gene was amplified with the primers F ( 5′-ATGCCGATCATTTTTGCTTC-3′ ) and R ( 5′-CCGTCCCTTAATTTTAGACTTCTTC-3′ ) [22] . Sequencing was performed at the Genotyping of Pathogens and Public Health Platform ( Institut Pasteur , Paris , France ) . All molecular epidemiological data were stored and analysed with Bionumerics software ( Version 6 . 5; Applied-Maths , Belgium ) . Genotyping was also performed by multiple-locus variable-number tandem repeat analysis ( MLVA ) using the loci VNTR4 , VNTR7 , and VNTR10 as described by Salaun et al . [23] . In the absence of PCR products , a second round of nested PCR amplification was performed with the inner primers NP 4A ( 5′-TTGGAGCGCAATCTCTTTTT-3′ ) and NP4B ( 5′- TGAGGATACCCCATTTTTACCTT-3′ ) , NP7A ( 5′-GATGGGCGGAGAAAAGTGTA-3′ ) and NP7B ( 5′-TGGATCGGTATTTTGGTTCA- 3′ ) , NP10A ( 5′-ATTCCAAAACTCAGCCCTCA-3′ ) and NP10B ( 5′- TGATGGGATTACCGGAAGAA-3′ ) . For pulsed-field gel electrophoresis ( PFGE ) , cells were embedded in agarose plugs as previously described [24] , and the DNA in the plugs digested with NotI . PFGE was performed in a contour-clamped homogeneous electric field DRII apparatus ( Bio-Rad Laboratories , Richmond , CA ) . Restriction fragments were resolved with ramping from 5 to 60 s for 50 h , 1 to 30 s for 40 h , or from 1 to 70 s for 36 h at 6 V/cm . Nucleotide sequences have been deposited with GenBank under accession numbers JX827500 - JX827597 . Guadeloupe and Martinique are islands situated in the Caribbean archipelago and are 100 miles apart . Guadeloupe and Martinique share common geological environments ( although Grande Terre in Guadeloupe is composed of limestone , the islands are mainly volcanic ) and are 1 , 705 and 1 , 100 km2 , respectively . They have similar population sizes ( <400 , 000 inhabitants ) and levels of urbanization . The islands are among the most highly developed islands in the Caribbean and their economies depend largely on tourism and agriculture ( sugar cane and bananas ) . The climate is tropical with two distinct seasons , the dry season from December to May and the rainy season from June to November . Over the last five years ( 2007–2011 ) , the annual incidence of leptospirosis has ranged from <12 per 100 , 000 inhabitants in Martinique ( 2007 ) to >41 per 100 , 000 inhabitants in Guadeloupe ( 2011 ) ( data from the NRC for Leptospirosis , Institut Pasteur , France ) , which is among the highest reported in the Caribbean ( <2 per 100 , 000 inhabitants in Trinidad and Tobago [25] , [26] and <13/100 , 000 inhabitants in Barbados [27] ) . In 2011 , the total number of cases was 165 in Guadeloupe and 142 in Martinique , which is two to three-fold more than in 2007 . Most of the infections were during the rainy season from August to November ( around 70% of all cases in 2011 ) . Detection of antibodies in patient sera by MAT has shown that the most prevalent Leptospira serogroup in the French West Indies is Icterohaemorrhagiae ( <25% in Martinique and <37% in Guadeloupe ) . The other serogroups each account for less than 12% of cases and include serogroups Ballum ( <5% in Martinique and <12% in Guadeloupe ) , Sejroe ( <7% in Martinique and <5% in Guadeloupe ) , and Canicola ( <7% in Martinique and <9% in Guadeloupe ) ( data from the NRC for Leptospirosis , Institut Pasteur , France ) . Results of identification of strains sent to the NRC for Leptospirosis ( Institut Pasteur ) for serogroup and genotype identification are shown in Table 1 ( Table 1 ) . Genomic DNA from 36 acute-phase blood samples that were positive for pathogenic Leptospira by PCR at the University Hospital of Fort de France were also included in this study ( Table 2 ) . The geographical distribution of the isolates was as follows: 91 strains isolated in Guadeloupe from 2003 to 2012 ( including 6 rat isolates ) and 55 strains isolated in Martinique from 2011 to 2012 ( including DNA from 36 patients ) ( Tables 1 and 2 ) . Serogrouping of isolates was first performed with rabbit antisera against reference serovars . The most frequent serogroups were Icterohaemorrhagiae ( 58% ) and Ballum ( 25% ) , consistent with the findings obtained by MAT with human serum samples ( see above ) . Other serogroups detected include Mini ( 9 isolates ) , Tarassovi ( 4 isolates ) , Australis ( 1 isolate ) , and Celledoni ( 1 isolate ) . Four isolates scored negative ( no agglutination ) with the antisera raised against the 24 serogroups . A selection of isolates from the serogroup Icterohaemorrhagiae were subsequently typed to serovar level by MAT with monoclonal antibodies ( MAbs ) against the serovars Icterohaemorrhagiae and Copenhageni: both serovars were present among the clinical isolates ( data not shown ) . Molecular typing was then performed by sequencing the 16S rRNA gene ( rrs ) in genomic DNA from 110 cultures and 36 acute-blood samples [20] . All the samples corresponded to one of five pathogenic species: L . interrogans ( 44 samples ) , L . kirschneri ( 36 samples ) , L . borgpetersenii ( 38 samples ) , L . noguchi ( 3 samples ) , and L . santarosai ( 22 samples ) ( Tables 1 and 2 ) . Two samples were phylogenetically related to L . kmetyi ( Figure 1 ) . The sequences of their 279-nucleotide 16S rRNA PCR products were identical with two mismatches ( 99% nucleotide identity ) to the corresponding variable region of the 16S rRNA sequence of the L . kmetyi reference strain . These L . kmetyi-positive cases showed MAT cross reaction with the saprophyte serovar Patoc ( Table 2 ) . The serovar Patoc , which is non-pathogenic , was included in our analysis because it has cross-reactivity with pathogenic serogroups and can be indicative of an infection . The last sample ( 201102109 ) was related to both L . kmetyi ( 273/279 nucleotides ) and L . kirschneri ( 272/279 nucleotides ) . This DNA may therefore correspond to a variant of L . kmetyi or L . kirschneri . PFGE has long been the gold standard method for genotyping Leptospira strains [28] , [29] . PFGE analysis of NotI-digested genomic DNA revealed at least thirteen distinct patterns for the typed isolates ( Figure 2 ) . For each strain , serovar designation was attributed by comparing the patterns with those of reference strains belonging to the identified serogroup and species ( Table S2 ) . For example , patterns of isolates identified as belonging to the species L . santarosai and serogroup Mini were compared with reference serovars that belong to the L . santarosai serogroup Mini ( i . e . serovars Beye , Georgia , Szwajizak , and Tabaquite ) . In this case , the “Mini” isolates displayed a PFGE pattern which was similar ( less than three band differences ) to the type strain of serovar Tabaquite ( Figure 2 ) . The “Tarassovi” isolates displayed unique PFGE patterns which were different from the PFGE patterns of the reference strains of L . borgpetersenii and L . santarosai serogroup Tarassovi ( serovars Kisuba , Tarassovi , Kanana , Guidae , Tunis , Yunxian , Atchafalaya , Atlantae , Bravo , Chagres , Darien , Navet , Rama , and Sulzerae ) . The PFGE profile of the “Australis” isolate was similar to serovar Bajan and distinct to the other reference strains from L . noguchi serogroup Australis ( serovars Rushan , Peruviana , and Nicaragua ) . For the “Celledoni” isolate , none of the reference serovars within this serogroup belong to the species L . santarosai . The PFGE patterns of the “Icterohaemorrhagiae” strains , which were subdivided into L . interrogans or L . kirschneri , were identical to the patterns obtained from L . interrogans serovars Icterohaemorrhagiae and Copenhageni , known to be indistinguishable by PFGE and other molecular typing techniques , and L . kirschneri serovar Bogvere ( less than three band differences were observed ) . The “Icterohaemorrhagiae” strains that were isolated from different patients over an eight-year period ( 2004–2012 ) and those from rats all presented indistinguishable PFGE patterns . The “Ballum” isolates displayed a pattern that was similar to that displayed by L . borgpetersenii serovars Ballum , Castellonis , Guangdong , Arborea , and Soccoestomes [30] ( Figure 2 ) . MLVA ( Multi Locus VNTR Analysis ) is a simple and rapid PCR-based method for the identification of most of the serovars of L . interrogans and L . kirschneri [23] . The L . interrogans and L . kirschneri isolates from the French West Indies had a MLVA pattern with VNTR-4 , VNTR-7 , and VNTR-10 identical to the serovars Icterohaemorrhagiae and Bogvere type strains , respectively . This is in agreement with the clusters determined by PFGE ( Table 1 ) , further confirming the identity of the serovars Icterohaemorrhagiae/Copenhageni and Bogvere . Strains from species L . borgpetersenii , L . noguchi , L . santarosai , and L . kmetyi could not be typed by this method because of the absence of one or more of the VNTR loci . The secY housekeeping gene [22] was also amplified from DNA extracts and sequenced . No PCR products were obtained for DNA from the L . kmetyi strains ( here designated as genotype F ) . This was presumably due to mismatching between the PCR primers and the target gene ( due to DNA sequence divergence ) , preventing PCR amplification [31] . The phylogenetic tree constructed with the secY nucleotide sequences is shown in the Figure 3 ( Figure 3 ) . Our 143 sequences ( not including the 3 L . kmetyi strains ) segregate into five main clades that correspond to the species identified by 16S rRNA sequencing . Thirteen different genotypes were observed and genotypes A ( 42 isolates ) , B ( 35 isolates ) , and C ( 33 isolates ) were the most prevalent . The secY alleles A , B , and C were associated with serovars Icterohaemorrhagiae/Copenhageni , Bogvere , and Arborea/Castellonis/Ballum/Guangdong/Soccoestomes , respectively . The remaining 32 strains were distributed into nine groups ( D , E , G , H , I , J , K , L , and M ) , including six new alleles not found in the database published by Nalam et al . [32] . Thus , there were thirteen groups in total , and most were present on both Guadeloupe and Martinique . However , some genotypes were found only among isolates from Guadeloupe ( group G with 9 isolates ) or Martinique ( groups H with 3 samples , I with 6 samples , and J with 2 isolates ) . Clusters A and B contained both , clinical and rat isolates . Leptospirosis is endemic in the French West Indies . The first human cases were first documented in 1932 in Guadeloupe [33] and 1938 in Martinique [34] . The annual incidence of leptospirosis in the French West Indies was estimated to be approximately 10 cases per 100 , 000 inhabitants in the 1990s . The incidence of leptospirosis during 2002–2004 was affected by the El Nino phenomenon , which resulted in increases in rainfall and the number of cases in Guadeloupe [35] . A prospective study of patients with acute febrile illness in Martinique and Guadeloupe ( InVS , CIRE Antilles-Guyane ) in 2011 improved the surveillance of leptospirosis . This increased awareness could explain the record incidence in 2011 , peaking at <39 cases per 100 , 000 inhabitants in both Guadeloupe and Martinique . Although the use of PCR diagnostic testing is becoming more common in the French West Indies , the diagnosis of leptospirosis is mostly dependent on MAT , which can identify the presumptive serogroup of the infecting bacterium . MAT has been used to show that the most frequent serogroups in Guadeloupe are Icterohaemorrhagiae and Ballum , followed by Sejroe and Canicola [36] ( data from the NRC for Leptospirosis ) . Serogroups Cynopteri , Tarassovi , Panama , Grippotyphosa and Autumnalis are less common . The sensitivity of MAT is low during the acute stage of disease [37] and , because of paradoxical reactions and cross-reactions between serogroups , the accuracy of MAT in identifying the infecting serovar or serogroup can also be poor [38] , [39] , limiting its epidemiological value . In this study , MAT serological data from culture-positive patients were reviewed retrospectively , allowing the identification of a total of 36 patients with MAT data for serum samples ( data not shown ) . It was possible to infer the serogroup identity of infecting leptospires from the MAT results for 26 of these 36 patients ( 72% ) . Similarly , only a small proportion of PCR-positive samples were correctly identified by MAT ( Table 2 ) . This further confirms that only the isolation of Leptospira from patients allowed definitive identification of the infecting serovar and is therefore essential for the study of the epidemiology of the disease . We determined 16S rRNA sequences to identify the isolates to the species level , and then used serogrouping , PFGE , secY sequences , and MLVA to sub-type the species . For most of the isolates ( 101/110 ) , the PFGE patterns were mostly consistent with those of known serovars: i . e . serovars Bogvere , Tabaquite , Bajan , and Icterohaemorrhagiae or Copenhageni . For the serogroup Ballum , the PFGE patterns of the reference isolates for serovars Ballum , Castellonis , Guangdong , Arborea , and Soccoestomes were all similar , with fewer than three band differences [30] . Serovar Arborea was previously identified by CAAT as the major serovar from the serogroup Ballum in the Caribbean island Barbados [40] , suggesting that our strains may belong to serovar Arborea . For the remaining five isolates which were serogrouped ( Tarassovi and Celledoni ) , comparison of PFGE patterns with reference strains was inconclusive for the serovar . Finally , for four isolates the rabbit antisera used did not lead to agglutination such that comparison with the reference strains was not possible . Surprisingly , none of the isolates in the last ten years from the French West Indies were identified as belonging to serogroups Canicola or Sejroe , although up to14% of MAT-positive sera correspond to these two serogroups . Similar findings were reported in Barbados [40] . This may be due to cross-reactions between serogroups in MAT and/or difficulties in isolating these strains from patients ( for example patients not hospitalized because of less severe symptoms or the strains fail to grown in EMJH medium ) . Typing by PCR-based methods for amplification of 16S rRNA , secY , and VNTR loci can be used directly on biological samples , thus avoiding culturing of the pathogen . The bacterial load in blood during the acute phase ranges from 102 to 106 Leptospira/ml . The Leptospira count decreases with time , and can be detected for up to 15 days [41] . Thus , if the bacterial load is low , it may be necessary to use nested-PCR for amplification of the target sequences . The classification according to secY sequences was in good agreement with the groupings determined by PFGE and MLVA , further confirming our previous data on clinical isolates from Mayotte [42] . Sequencing of secY in DNA extracted from the clinical isolates and blood samples allowed a simple and rapid first-line screening and the identification of the presumptive serovar . A total of thirteen genotypes were found in our study , a large proportion of strains ( 75% ) being of only three genotypes associated with serovars Icterohaemorrhagiae/Copenhageni , Bogvere , and Arborea . The secY sequences from the L . santarosai isolates showed the highest nucleotide diversity . Six genotypes were not found in the MLST database and may therefore be specific to the French West Indies . Further characterization of these isolates should include the use of the CAAT , which requires the preparation of antibodies against the strain of interest , for definitive identification of the serovar . We also detected the appearance of strains related to the pathogenic species L . kmetyi in Martinique in at least two patients , one of which was probably exposed during canyoneering activities in the tropical forest ( Hochedez et al . , submitted ) . To our knowledge , L . kmetyi which was first isolated from soil in Malaysia [43] , has never been isolated from a patient with leptospirosis . Further studies are needed to determine the serological and molecular features of these strains and their distribution in the French West Indies . The distribution of the predominant pathogenic leptospiral serovars differed between Guadeloupe and Martinique . Serovars Bogvere , Arborea , and Icterohaemorrhagiae/Copenhageni made up 35 , 31 , and 23% respectively of all Leptospira isolates in Guadeloupe since 2004 . In Martinique , serovar Icterohaemorrhagiae is the most frequent ( 35% ) , followed by Arborea ( 9% ) and Bogvere ( 6% ) . In the Caribbean island of Barbados , 140 miles from Martinique , serovars Arborea ( 14% ) and Icterohaemorrhagiae ( 26% ) similarly cause many human infections , but the serovar Bogvere [38] , [40] , [44] , which was first isolated in Jamaica [45] does not . Serovar Tabaquite ( serogroup Mini ) , which was found in Guadeloupe , was first isolated from a patient in Trinidad [46] . Serovar Bim ( serogroup Autumnalis ) is the most frequently isolated serovar in Barbados ( 75% of all isolates ) [40] , was not isolated in the French West Indies . This suggests that some strains circulate throughout the Caribbean islands but others are highly prevalent only in restricted areas . This may be related to the distribution of the animal reservoirs for the different serovar in these islands . Leptospira can colonize or infect renal tubules of a wide variety of wild and domesticated mammals . In the Caribbean , numerous mammalian species including rodents , opossums , mongoose , bats , pigs , cattle , and dogs have been demonstrated to be hosts of pathogenic Leptospira species [47] . Isolates from the serogroup Icterohaemorrhagiae , including serovars Icterohaemorrhagiae and Bogvere , have been isolated from the kidneys of rats , mice , and mongoose and isolates from the serogroup Ballum were isolated from rats and mice , suggesting predominantly rodent-borne transmission of the disease . Serovar Arborea was reported to be prevalent in both humans and animals in Barbados [40] , [48] . Serovar Bajan was originally isolated from toads and frogs in Barbados [49] . In our study , human and rat isolates from Guadeloupe and belonging to serovars Icterohaemorrhagiae/Copenhageni , Bogvere , and Arborea all showed identical genotypes , consistent with rats being responsible for the transmission of the disease . The identification of the circulating etiological agents of leptospirosis in the French West Indies will help establish appropriate control and prevention measures in this area where the disease is endemic . For example , the reference technique , MAT , requires a panel of live antigens representing a broad range of serogroups . The use of local isolates in the panel of antigens may maximize the chances of detecting an immune response to the infecting bacterium . At the NRC for Leptospirosis ( Institut Pasteur ) , the initial panel of 18 antigens , which already included strains representative of the serogroups Icterohaemorrhagiae , Ballum , Australis and Tarassovi , was thus expanded to include local isolates from serogroups Celledoni and Mini for serum samples originating from the French West Indies . Knowledge of leptospiral epidemiology may also be useful for the development of a whole bacterial vaccine against leptospirosis . Vaccines currently available for use in animals and , in a few countries , in humans generally consist of one , two or more locally prevalent serovars . In France , a human vaccine containing only serovar Icterohaemorrhagiae has been used since 1981 [50] . However , we report here that only one-third of the infections in the French West Indies are due to serovar Icterohaemorrhagiae ( not including serovar Bogvere from the serogroup Icterohaemorrhagiae ) , and the corresponding figure for Barbabos is 22 . 5% [38] . Immunity is restricted to antigenically related serovars , so the vaccine used in France may not be effective against the majority of strains circulating in the French West Indies . Further studies should include the analysis of the influence of serovar and strain genetic background on the clinical presentation and outcome of the disease . It would also be valuable to investigate the reasons for differences in the distributions of Leptospira serovars in the Caribbean islands .	Leptospirosis is an emerging zoonotic disease caused by infection with pathogenic strains of Leptospira . Isolation of Leptospira strains is rare , making it difficult to assess their distribution worldwide . In this study , we characterized cultures of Leptospira obtained from more than one hundred leptospirosis patients from the French West Indies by serology and various molecular typing methods to identify the strains circulating in this endemic region . Typing of leptospiral isolates showed that causative agents of leptospirosis in the French West Indies are mainly from the serogroups Icterohaemorrhagiae and Ballum , but we also identified new genotypes . We also found that the distribution of the predominant pathogenic leptospiral serovars differed between the Caribbean islands . A better understanding of the epidemiology of leptospirosis will improve our knowledge in the distribution of this emerging neglected tropical disease worldwide . The identification of the circulating etiological agents of leptospirosis in the French West Indies will also help establish appropriate control and prevention measures in this area where the disease is endemic .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "medicine", "infectious", "diseases", "neglected", "tropical", "diseases", "leptospirosis" ]	2013	Serovar Diversity of Pathogenic Leptospira Circulating in the French West Indies
The vertebrate neural plate contains distinct domains of gene expression , prefiguring the future brain areas . In this study , we draw an extended expression map of the rostral neural plate that reveals discrete domains inside the presumptive posterior forebrain . We show , by fate mapping , that these well-defined cell populations will develop into specific diencephalic regions . To address whether these early subterritories are already committed to restricted identities , we began to analyse the consequences of ablation and transplantation of these specific cell populations . We found that precursors of the prethalamus are already specified and irreplaceable at late gastrula stage , because ablation of these cells results in loss of prethalamic markers . Moreover , when transplanted into the ectopic environment of the presumptive hindbrain , these cells still pursue their prethalamic differentiation program . Finally , transplantation of these precursors , in the rostral-most neural epithelium , induces changes in cell identity in the surrounding host forebrain . This cell–non-autonomous property led us to propose that these committed prethalamic precursors may play an instructive role in the regionalization of the developing diencephalon . The vertebrate brain is divided caudal to rostral into the hind- , mid- , and forebrain . These territories are further developing into highly specialised structures . In the case of the hindbrain , subdivisions are easy to detect morphologically , because rhombomeres are marked by visible borders during the course of development [1] . By contrast , the partition of the embryonic forebrain is not as obvious . The optic recess creates a border between the dorso-rostral telencephalon and the diencephalon . The latter is traditionally split into five different domains: the ventrally located hypothalamus , the prethalamus ( or ventral thalamus ) , the thalamus ( or dorsal thalamus ) , the dorsally positioned epithalamus , and the caudalmost pretectum [2] . The prosomeric model by Puelles and Rubenstein [3] suggests that the forebrain develops from a set of compartments called prosomeres . Boundaries between individual prosomeres are not morphologically visible . It is hypothesized , based on gene expression profiles , that the pretectum , the thalamus , and the prethalamus develop from different prosomeres [3] . Inside the embryonic diencephalon , morphological landmarks distinguish the forming hypothalamus and epithalamus . The zona limitans intrathalamica ( ZLI ) separates the prethalamus from the thalamus and pretectum , and has been shown recently to be required for the maintenance of prethalamic and thalamic identities [1 , 4] . The molecular and cellular events leading to this complex organization are very poorly understood . Until recently , the neural plate itself was considered as a uniform epithelium of unspecified neural precursors . However , the detection of local signalling centres inside the forming neural plate [2] indicates that , as the neural plate forms , at least some cell populations are strictly committed to distinct fates [5 , 6] . An increasing number of genes found to be expressed in restricted areas inside the rostral neural plate raise the possibility that a whole set of forebrain fate decisions are already taken inside this nascent neuroepithelium . To gain knowledge of early forebrain development , fate map studies are of great interest . Fate maps of the rostral neural plate have been generated in frog , chick , and mouse [7–10] . These studies show that precursors of a given forebrain region are located in a very defined area of the neural plate , implying very little or no cell mixing during early brain development . In zebrafish and medaka , fate maps were generated at early , mid , and late gastrulation stages [11–14] . These fish neural plate maps establish the broad location of forebrain , midbrain , and hindbrain precursors during neural plate maturation without reaching a resolution revealing the location of different diencephalic areas inside the neural plate . Being predominantly interested in exploring cell-fate specification inside the presumptive diencephalon , we set out to generate detailed expression and fate maps of the diencephalic anlage in the mature neural plate ( bud stage in zebrafish ) . We constructed our expression map by double in situ hybridization using a vast panel of forebrain markers . In addition , we established a refined fate map of the anterior neural plate using a laser uncaging technique . The results show a remarkable segregation of each major diencephalic domain . Finally , we show , using ablation and transplantation approaches , that one such domain , the presumptive prethalamus , is already strictly committed and irreplaceable inside the neural plate , and exerts some control upon fate decision of its surroundings . All together , our observations strongly suggest the existence of a novel neural plate “mid-forebrain” compartment , regulating early steps of forebrain partitioning . To date , studies of gene expression in the vertebrate anterior neural plate have highlighted the presence of , from rostral to caudal , a horseshoe-shaped telencephalic domain , followed by a well-defined eye field , which in turn abuts a diencephalic anlage ( see [2] for review ) . Our previous work suggested that the ablation of at least one restricted population of diencephalon precursors ( row 6/7 [6] ) could lead to defects in posterior forebrain development , opening the possibility that some irreversible cell-fate decisions may be taken inside the posterior forebrain territory at that time [6] . To elucidate the diencephalic organization at such an early stage of development , we first established a rostral neural plate gene expression map . We found that , at bud stage , diencephalic markers like arx [15] , fezl [16] , and barhl2 [17] are located directly posterior to the eye field ( Figure 1A and 1B ) . Inside that domain , flh [18] is exclusively expressed laterally ( domain II in Figure 1C , and unpublished data ) , overlapping with the telencephalic territory; whereas arx is expressed in the medial part of the barhl2 expression domain at the same developmental stage ( domain I in Figure 1B and 1C ) . In cells posterior to the ones expressing arx , transcripts of irx7 [19] are detected ( domain III in Figure 1D ) . Although irx7 has been described to be expressed in the midbrain and hindbrain , an overlap of the rostral expression domain with that of the forebrain marker pax6a [20] shows that the anterior-most irx7-expressing cells are located within the presumptive forebrain territory ( unpublished data ) . Other genes that we found to be expressed in the posterior diencephalic domain III are wnt8b ( Figure 1E ) and foxb1 . 2 ( Figure 1F and 1G ) , which are expressed in an overlapping pattern with irx7 ( Figure 1F ) and are located posterior to fezl ( Figure 1G ) . The resulting expression map of early forebrain markers shows a subdivision of the forebrain into six different domains ( Figure 1H ) , which expands the previously described map of the rostral neural plate , revealing subdomains inside the diencephalic anlage . Besides the hypothalamic anlage ( domain IV in Figure 1 ) the diencephalon is further partitioned into the anterior medial domain I ( arx , fezl , and barhl2 ) , the anterior lateral domain II ( flh and barhl2 ) , and the posterior domain III ( irx7 , foxb1 . 2 , and wnt8b ) . Sagittal ( Figure 1I and 1J ) and transverse ( Figure 1K and 1L ) sections of embryos showing fezl ( blue ) and foxb1 . 2 ( red ) expression indicate that , by bud stage , hypothalamic cells are positioned underneath the eye field . This can be interpreted in two ways . Either the neural plate has already begun to keel and the presence of a seemingly continuous retinal field is due to invisible ventricular limits; or , alternatively , the neural plate is not strictly a single pseudo-stratified epithelium , and the hypothalamic precursors are lying , and moving , underneath a single eye field . This second interpretation is supported by previously published fate maps in axolotl [21] , Xenopus [10] , and chick [22] showing an eye field uninterrupted by hypothalamic precursors; and by a very recent study , showing by computer-aided cell-tracking experiments [23] that , contrary to what has been previously suggested [13] , the eye field is not split by hypothalamic precursors in zebrafish . To assess whether the different diencephalic subdomains found in our expression map give rise to specific areas of the developing forebrain , we set out to perform fate-mapping experiments . To achieve a high level of precision in labelling neural plate cells , we needed visible landmarks inside the rostral neural plate . We made use of the green fluorescent protein ( GFP ) transgenic line her5pac:egfp [24] , which shows a very robust GFP expression in a V-shaped territory covering the anlagen of the midbrain and rostral hindbrain at bud stage . Tallafuss and Bally-Cuif [24] showed that the most rostrally positioned GFP-expressing cells contribute exclusively to the developing midbrain . By combination with the forebrain marker pax6a , we confirm that her5pac:egfp is expressed directly posterior to the presumptive forebrain inside the neural plate ( Figure 2A–2D ) . Testing the location of the different diencephalic expression domains in combination with the GFP marker , we see a clear gap between the anterior diencephalic markers such as barhl2 and GFP ( Figure 2E and 2F ) . Inside this gap , the expression of the posterior marker irx7 is found ( Figure 2G and 2H ) . The mild V-shape expression domain of the anterior diencephalon anlage at early bud stage ( Figure 2E ) has straightened by late bud stage ( Figure 2F ) . Conversely , the anlage of the midbrain still remains V-shaped , revealing a triangular posterior diencephalic territory at that stage . These shape changes are accompanied by the initiation of keel formation ( neurulation ) , which positions the rostro-alar midbrain area on top of the posterio-basal diencephalon during early somitogenesis ( Figure 2D ) . As a consequence , at later stages of embryogenesis , the rostral part of the dorsal midbrain ( optic tectum ) appears to lie dorsal to the caudal-most ventral diencephalon ( anterior tegmentum ) . Our expression map of the presumptive forebrain shows that cells located in the diencephalic territory differ already by their molecular constituents . To assess whether these molecular subdivisions match territories fated to become well-defined parts of the developing diencephalon , we performed fate-mapping studies of these specific cell populations , using the her5pac:egfp transgenic fish ( permitting a high level of accuracy in cell targeting ) . We injected a caged form of fluorescein into one-cell stage her5pac:egfp transgenic embryos . Once the embryos reached bud stage , the fluorescein in 6–10 cells , within arbitrary domains , were uncaged using a laser beam ( Figure 3A; see Materials and Methods ) . A cross section of the neural plate just after uncaging shows that cells have been labelled all along the z-axis ( Figure 3B ) , indicating that our setup allows precision along the xy-axes , but not control of depth inside the neural plate . When uncaging broad areas of the diencephalic neural plate , we observe that cells inducing a specific early diencephalic marker tend to keep expressing it specifically through somitogenesis . Indeed , when we carefully label the barhl2 expression domain at bud stage , fluorescein is detected in a broad forebrain domain at prim 5 , which closely resembles the expression pattern of barhl2 in the diencephalon at the same stage ( Figure 3C and 3D ) . Because we wanted to establish a more precise fate map of the diencephalon , we decided to label smaller areas of the neural plate using an arbitrary grid . Figure 4A shows a more abstract version of the expression map shown in Figure 1 , highlighting domains I to IV . The correlation of the chosen grid and the expression map is represented in Figure 4B in which we also added an additional ( orange ) domain to address the border between the diencephalon and the telencephalon . After a set of preliminary data using the grid on Figure 4B , we refined our arbitrary domains to 12 colour-coded areas ( Figure 4C ) . For a better understanding of the results , we indicated different areas of the rostral brain in a prim5 embryo displaying shh expression ( Figure 4D ) . The outcome of the uncaging experiments is shown in Figure 4E–4P′; the colour of the frames corresponds to the colour-coded grid in Figure 4C . Forebrain cells of the anterior midline populate reproducibly the developing hypothalamus ( Figure 4E ) and the eye field ( not shown in our dissected brains ) , whereas cells posterior to those give rise to cells in the posterior tuberculum and anterior tegmental domains ( Figure 4G–4H′ ) . At the interphase between these two domains , we found a medial cell population displaying an unexpected behaviour . These cells , located inside the anterior diencephalic anlage revealed by the expression map , give rise to progenies that spread into a more-alar region than their caudal and rostral midline neighbours ( Figure 4F ) . This domain corresponds to the ventral-most part of the presumptive ZLI ( co-localises with the shh-expressing cells , Figure 4F ) . Cells of the dark red–labelled region in Figure4C become the prethalamus ( Figure 4I ) . Close to this area , towards the edge of the neural plate ( orange and light blue area ) , precursors of the dorsal telencephalon ( Figure 4J ) or dorsal telencephalon and epithalamus ( Figure 4K ) are found . When we labelled cells in a slightly more medial position ( dark blue ) , the fluorescein was detected in a region spanning the dorsal telencephalon and the epithalamus , and abutting the dorsal pretectum ( Figure 4L ) . Cells of the bright red and dark green areas are destined to become part of the thalamus ( Figure 4M–4N′ ) . And cells positioned in the neural plate very close to our GFP landmark give rise to the most-caudal forebrain cells ( pretectum and anterior tegmentum , Figure 4O–4P′ ) . If one compares the relative distribution of the subdomains at bud stage and at prim5 ( Figure 5A and 5B ) , one can predict that a simple rostral shift of the midline accompanying the closure of the plate into a keel may be sufficient to create most of the embryonic brain pattern observed at prim5 . To investigate whether such a shift takes place and if so , at which stage of development this movement happens , we used confocal microscopy to create time-lapse movies allowing to monitor the movement of neural plate cells from bud stage to mid-somitogenesis . We recorded her5pac:egfp transgenic embryos in which a proportion of the nuclei were labelled with red fluorescent protein ( RFP; see Materials and Methods ) . With this technique , we were able to follow cells of the diencephalic territory during the course of development ( Video S1 ) . After three-dimensional ( 3D ) reconstruction ( Figure 5C and 5D ) , the movement of specific nuclei could be tracked over the course of the recorded time ( Figure 5E ) . We found that between bud and 5-somite stage ( when the rostral neural plate is just completing closure ) , substantial movement can be observed in the ectoderm . Looking at the trajectories of the nuclei ( Figure 5E ) , one sees that cells of the medial neural populations ( basal plate ) move anteriorly ( green arrow ) , while lateral cells ( alar plate ) move either first towards the midline and then anterior ( posterior alar forebrain , yellow arrow ) or diagonally towards the anterior if they belong to the anterior alar forebrain ( red arrow ) . These results therefore strongly suggest that the final location of the different diencephalic areas found in prim5 brains is taken during the beginning of somitogenesis , as the neural plate is “keeling . ” When we label cells of the midline ( basal plate ) in the neural plate , we observe often a more dorsal–positioned progeny of the cells then we first expected . A dramatic example of such observation is the presence of basal plate cells inside the ZLI ( Figure 4F ) . In our uncaging experiments , we are not able to define the medio-lateral extent of the basal plate as there is no visible boundary between basal and alar neural plate . To find the border between basal and alar plate , we set out to mark solely basal-derived cells . It is known that , in zebrafish , cells of the basal plate originate close to the shield at 50% epiboly [25] . In contrast , cells of the alar forebrain are positioned in the animal pole region at the same stage . Therefore , we transplanted cells at 50% epiboly on top of the shield ( Figure 5F and 5G ) and selected at bud stage for embryos in which the transplanted cells were nicely spread along the midline ( Figure 5H ) . By that approach , we labelled cells along the basal plate , similar to the uncaging experiment , but this time , because of the specific origin and movement of these basal cells during gastrulation , we were sure that just cells of the basal plate were marked . We let these embryos develop up to prim5 ( Figure 5I ) or prim22 ( Figure 5J–5L ) and compared the distribution of the transplanted cells to the expression of shh . Cells from the basal plate contribute to a bigger part of the rostral forebrain then we expected . In comparison to more caudal parts of the brain and spinal cord , basal plate cells can be found at much more dorsal positions in the rostral forebrain ( up to the optic recess in its rostral-most portion ) , probably due to a high level of proliferation of these cells in the rostral central nervous system ( CNS ) . Basal cells also contribute to the ventral part of the ZLI and thalamus ( yellow arrows in Figure 5I and 5L ) . Just the tip of the ZLI ( white arrows in Figure 5I and 5L ) is always deprived of basal cells . In the hindbrain , very few cells can be found dorsal to the basal–alar border ( arrowhead in Figure 5K ) , which are likely to reveal directed neuronal migration into the alar plate . Our fate-map experiments show that different forebrain territories , marked by the differential combination of gene expression , give rise to specific parts of the forebrain in later stages of development . This finding led us to address whether any of these territories is already functionally specified and irreplaceable . We focussed our effort on the domain forming the future prethalamus ( dark red domain in Figure 4 ) . Cells of this area were ablated by mechanically removing them from the neural plate ( see Materials and Methods ) . To control the specificity of our ablations , we performed RT-PCR on RNA extracted from prethalamic cells ( 30 embryos ablated ) . We tested for presence of transcripts of the anterior diencephalic marker barhl2 , the telencephalic marker foxg1 [26] , and the midbrain marker pax2a . Although barhl2 is readily amplified , neither of the two others is detected ( unpublished data ) . We observed that half of the treated embryos show a reduction or loss of the marker barhl2 ( n = 8 , Figure 6A–6D ) 2 h after suction of the cells . Another set of ablated embryos , fixed at the 10–12-somite stage , shows a complete ( 3/11 ) or partial ( 4/11 ) loss of the prethalamus marker arx ( Figure 6E–6G ) . Finally , 19 ablated embryos were left to develop to prim5 stage . About a third of these show a dramatic reduction of arx ( n = 6/19 , Figure 6H–6J ) . A loss or severe reduction could also be observed with other prethalamic markers such as dlx2a ( Figure 6K and 6L ) . The ablation of the prethalamic precursors does not lead to a loss of thalamic identity ( based on foxb1 . 2 expression at prim5 stage; Figure 6M and 6N ) . These data demonstrate that ablation of the prethalamic anlage impairs the formation of the prethalamus , therefore strongly suggesting that this area is indeed specified and irreplaceable at that stage . Having shown that prethalamic identity is already specified at bud stage , we assessed whether this commitment was sufficient to drive the development of these cells into differentiating progenitors . To address this , we tested whether these cells keep their identity and express later prethalamic markers in an ectopic location ( Figure 7 ) . Cells of the prethalamic anlage were therefore transplanted into ectopic regions of a host neural plate ( Figure 7A ) . Around 75% of the transplants ( n = 28 ) showed expression of the prethalamic markers dlx2a ( Figure 7B–7D and 7H–7J ) , lhx5 ( Figure 7E–7G and 7K–7P ) , and arx ( Figure S1 ) . Because , from late somitogenesis onwards , these markers begin to be expressed in telencephalic progenitors , we analysed , in the same embryos , the expression of bona fide dorsal telencephalic markers tbr1 [27] ( Figure 7B–7D and 7H–7J ) and emx3 [28] ( Figure S1 ) , and show that the transplants acquired diencephalic , but not telencephalic characteristics . When cells are transplanted into the presumptive hindbrain region of the host neural plate ( n = 8; Figure 7B–7G ) we found expression of both dlx2a ( Figure 7B–7D ) and lhx5 ( Figure 7E–7G ) inside the clone and no ectopic expression of the telencephalic markers tbr1 or emx3 ( see inset of Figures 7B and S1 ) . The absence of irx1b transcript inside the clones ( arrow in Figure 7F ) shows that these cells do not express the surrounding hindbrain characteristics . This finding indicates that the transplants develop as prethalamus , showing that these precursors , specified at bud stage , keep their identity independent of their location . In these cases , the transplants form round and compact structures , which are sometimes excluded from the hindbrain ( Figure S1 ) . In contrast to these results , donor cells located just rostral to the prethalamic precursors never express prethalamic markers when transplanted in ectopic neural plate areas ( n = 7; unpublished data ) . Contrasting with this observation , when cells are transplanted in more rostral regions ( Figure 7H–7P ) , the clones are much more dispersed and donor cells are interspersed with the surrounding tissue ( n = 7 , Figure 7H–7J and unpublished data ) . In these cases , ectopic expression of our prethalamic markers is observed both in and outside of the clones ( n = 5 , Figure 7H–7M ) , sometimes splitting the expression of the telencephalic marker tbr1 ( arrow in Figure 7I ) . Ectopic expression of prethalamic markers could be observed anterior to the endogenous prethalamus , inside the telencephalon ( Figure 7H–7J ) , or posterior , inside the thalamic area ( Figure 7K–7P ) . As shown for the more caudal grafts , the rostral transplants do not express typical markers of the surrounding host tissue even when positioned inside the thalamus ( arrow in Figure 7O ) . Given the described role of the ZLI and its signalling component shh in maintenance of prethalamic identity [1 , 4] , we tested whether shh signalling may play a role in the ectopic expression of prethalamic markers of our transplants . We treated embryos with the chemical cyclopamine , directly after transplantation . Even in embryos impaired for shh signalling , not only the endogenous lhx5 expression was still detectable , but also the transplanted cells were still able to express this prethalamic marker ( Figure S1G–S1I ) . The ability of the presumptive prethalamic cells to maintain their identity is therefore independent of the possible expression of shh by some of these precursors . All together , these results unambiguously show that , in the neural plate , the presumptive prethalamic territory is strictly specified and is able to differentiate into an embryonic prethalamus in absence of any extrinsic contribution . Moreover , some prethalamic precursors are able to influence their environment , suggesting that at least a portion of these have the ability to impose prethalamic identity to presumptive forebrain cells . Our current knowledge of the early diencephalic development is very limited . Studies done in fish , frog , chick , and mouse support the idea that , within the prospective forebrain , telencephalon and eyes are specified in regions of no or low Wnt activity , whereas posterior diencephalic fates are promoted by Wnt signalling [29–33] . However , the molecular , cellular , and temporal regulations leading to the acquisition of specific diencephalic identities ( such as prethalamic , epithalamic , hypothalamic , thalamic , and pretectal ) have yet to be elucidated . We established an expression map for genes expressed in the anterior neural plate at bud stage . At such an early stage of development , we were able to find distinct expression territories within the diencephalon anlage . This result is the first experimental evidence showing that as soon as a diencephalic identity can be detected molecularly inside the neural plate , it already contains at least four distinct domains of gene expression . One cannot exclude the possibility that there may even be further discrete subdivisions . In this respect , it is worth noting that the boundary of expression of some of the markers is dynamic . This could create a further molecular difference at a slightly later stage in development , by establishing differential temporal exposure to various proteins . Our fate-map results show that the progeny of each labelled cell group develops in a continuous clone and does not show cell mixing with neighbouring regions . Clones could be allocated to distinct parts of the diencephalon , thus allowing us to predict which cells of the neural plate will develop into the prethalamus , thalamus , pretectum , and their basal counterparts . Forebrain fate maps done in mouse [7] , chick [8 , 9] , and frog [10] showed clonal distributions that suggest a similar lack of cell mixing . Knowing that a lot of morphological changes take place between bud and prim5 stage , such absence of cell mixing points to a considerable level of cohesion in cell movements . Indeed , our time-lapse imaging of the closing neural plate ( Figure 5 and Video S1 ) revealed that the relative positions of the mapped prim5 forebrain territories are established during the formation of the keel , by a series of coordinated movements . In fact , most of the mapped neural plate territories adopted their relative positions by the 5-somite stage ( completion of keeling ) . Superimposing our expression and fate-map data , specific combinations of gene expression can be associated to specific fates . Interesting is the observation that some gene expression in the neural plate seems to be kept in the same cells up to prim5 stage , as shown for the example of barhl2 ( Figure 3C and 3D ) . In contrast , our work demonstrates that the neural plate flh-expressing cells ( domain II ) , previously thought to include mostly the pineal gland precursors [18] , form a territory contributing to the dorsal telencephalon , the roof of prosomere 3 , and the epithalamus . Just very few cells of this early flh expression domain in the neural plate participate to form the pineal gland itself , although flh is clearly mostly expressed in the epiphysis at prim5 [18] . The fate map also reveals two origins for the ZLI precursors ( defined by expression of shh [1] ) : a basal cell population forms the ventral part of the peak , and an alar cell population is contributing to the more dorsal ZLI . Our results support the model by which the ZLI forms at the interphase between an already specified presumptive prethalamus and thalamus . Although our data do not support the existence of a wedge-shaped precursor for the ZLI in zebrafish ( see [1] for review ) , it does not disprove the existence of a ZLI compartment , although in fish , such a segment would have to be much narrower than in chick . The basal contribution to the ZLI is strongly supported by our transplants of basal plate cells . These transplants also highlight basal participation to extensive parts of the anterior forebrain , such as the pre-optic area and the ventral portion of the thalamus . Our clonal analysis further shows that one is able to distinguish between prethalamic or thalamic precursors in the neural plate , pointing to the possible presence of a compartment boundary between these two populations at bud stage . This supports the presence of compartments suggested by the prosomeric model [3] . Our ablation and transplantation data indeed further suggest that the prethalamic territory acts as a compartment , with rostral and caudal boundaries , supporting the idea proposed in the latest prosomeric model [3] , of a prethalamus acting as a boundary separating the secondary prosencephalon from the rest of the forebrain . It also indicates that such prosomeres may form during or just prior to neurulation . Our ablation experiments show that the prethalamic precursors are irreplaceable and therefore are acquiring a unique cell identity between mid and late gastrulation . Detection of residual prethalamic identity at different stages following ablation uncovers some degree of recovery in a small proportion of the prim5 brains . This is likely due to a rescue of the area during the course of development in embryos in which prethalamic precursors were only partially ablated . More importantly , the fact that we still see a dramatic effect in a good proportion of the prim5 embryos indicates that the area cannot be rescued if all prethalamic precursors are removed . This finding indicates the presence of specific factors defining prethalamic identity . A handful of genes are to date known to be expressed in the anterior diencephalic neural plate . The combinatorial expression of these candidates may lead to the establishment of prethalamic identity . Future extensive expression profiling and loss of function experiments will elucidate the genetic components of such fate commitment . Very recent publications show that Fez and Fez-like are key proteins for prethalamic identity [34 , 35] . However , loss of function of these genes does not completely abolish induction of the prethalamic territory , indicating that other factors are required . It has been shown that the expression of shh in the ZLI is important for maturation of the prethalamus [1 , 4 , 36] . One could argue that by ablation of the prethalamic anlage at bud stage , we partially ablate precursor cells of the ZLI , which are positioned close by and therefore may impair proper development of the prethalamus . However , we know that patterning genes such as fezl and lhx5 are still expressed in the prethalamic anlage in the absence of shh expression in the ZLI ( S . Scholpp and C . Houart , unpublished data , and Figure S1 ) . Moreover , we do not observe a loss of shh expression in domain I–ablated embryos ( unpublished data ) . The prethalamic identity is therefore lost in ablated embryos , despite the presence of shh , due to a loss of neural plate cells , which uniquely acquired a specific prethalamic competence . Our transplantation experiments further confirm prethalamic fate commitment . Several hours following transplantation , prethalamic cells are still expressing markers specific to their fate . Strikingly , the transplanted cells not only keep the expression of the transcripts already present at the time of ablation , but are also able to pursue prethalamic differentiation . Indeed , these cells are able to express dlx2a , whose transcription starts to be induced in the wild-type diencephalon 6–7 h after bud stage . As mentioned earlier , shh plays a crucial role in the development of the prethalamus . Therefore , one could speculate that shh signalling may be involved in the establishment of prethalamic identity and activation of prethalamic markers in our transplantation experiments . We excluded this possibility because both endogenous lhx5 transcripts in the developing prethalamus and ectopic expression in the prethalamic transplants are observed in embryos with impaired Hh signalling ( Figure S1G–S1I ) . This finding demonstrates that the early steps in prethalamus development are not dependent upon Hh signalling and , therefore , that other factors are essential to establish the prethalamic fate within these cells of the neural plate . When cells were transplanted away from their origin into the presumptive hindbrain , we observed a cell-autonomous induction of prethalamic markers and no induction of telencephalic markers . The result is rather more complex when transplants are placed in the presumptive forebrain . In these transplants , we observe cell–non-autonomous expression of the prethalamic transcripts dlx2a and lhx5 . This surprising result suggests that some prethalamic precursors not only maintain their identity , but are also able to influence surrounding forebrain tissues . The prethalamic cells may secrete one or a combination of signalling factors , which can be interpreted by neighbouring forebrain tissue , but not by hindbrain cells . In our operated embryos , only a fraction of the transplanted domain is generally able to induce cell–non-autonomous changes , suggesting that a specific subset of the prethalamic precursors contains the signalling property . This signal is able to impose prethalamic identity to presumptive forebrain cells . In the wild-type situation , it is therefore likely that the action of this ( these ) presumptive signal ( s ) is restricted to a narrow region by formation of strict boundaries rostral and caudal to the prethalamic precursors . Here again , Hh is a potential candidate because Vieira and Martinez [37] observe some similar effects when transplanting quail Hh-expressing diencephalic basal plate into chick alar forebrain areas . However , if involved , it cannot be the only player , because our transplants ectopically induce lhx5 expression , which is detected , even in wild type , much prior to ZLI formation and has been shown to be Hh independent ( S . Scholpp , I . Foucher , N . Staudt , D . Peukert , A . Lumsden , and C . Houart , unpublished results; Figure S1 ) . It has been shown that signalling centres inside the rostral neural plate are important for the proper development of the forebrain . At the anterior neural border ( ANB ) WNT-inhibitors guarantee low wnt values in the rostral forebrain important for the development of the eye and telencephalon [30 , 38] . Expression of fgf3 and fgf8 is also described in the ANB [4 , 36] , and FGF8 has been shown to be required for development of the rostral forebrain [39–41] . Finally , the midbrain–hindbrain boundary regulates the formation of the forebrain–midbrain border [42] . In this study , we show that precursors of the rostral diencephalon are strictly committed to their fate during gastrulation , may play a role as a key transition area , and are able to influence the surrounding forebrain territory . Future experiments addressing gene functions in these cells will identify the proteins responsible for their unique properties . In situ hybridization and immunohistochemical stainings were performed according to standard protocols [4 , 20] The expression patterns of the following genes were visualized: arx [15] , barhl2 [17] , dlx2a [43] , fezl [16] , flh [44] , foxb1 . 2 [45] , irx1 [46] , irx7 [19] , lhx5 [47] , shh [4] , six3 [48] , pax6a [20] , tbr1 [27] , and wnt8b [49] . The GFP in embryos of the her5pac:egfp transgenic line was detected by using an anti-GFP antibody ( Torrey Pines Biolabs , Houston , Texas , United States ) and as a secondary antibody Alexa 488 ( Molecular Probes , Invitrogen , Paisley , United Kingdom ) . The biotin in transplanted cells of Figure 7 was visualized with streptavidin 488 ( Molecular Probes ) . Photos were taken with a Nikon eclipse E800 microscope and figures made in Adobe Photoshop CS ( Adobe Systems , Uxbridge , United Kingdom ) . To label specific groups of cells in embryos of the her5pac:egfp transgenic line , 1% caged fluorescein ( Molecular Probes ) was injected in one-cell stage embryos . These embryos were incubated in the dark until bud stage . They were checked for their GFP expression and mounted in 5% methylcellulose . In cells of specific territories in the diencephalic neural plate , the caged fluorescein was converted via a laser beam ( 565 nm ) to its fluorescent form . Subsequently , embryos were kept at 28 °C in the dark until they developed up to prim5 stage . Then embryos were fixed in 4% PFA for 3 h at room temperature or at 4 °C overnight . The uncaged form of the dye was visualized by using an anti-fluorescein antibody coupled with alkaline phosphatase ( Roche , Basel , Switzerland ) , in some cases following in situ hybridization of marker genes . A plasmid containing nuclear RFP ( kindly provided by Megason and Fraser ) was injected into her5pac:egfp transgenic embryos at one to four cell stage to get a mosaic distribution of the RFP-labelled nuclei . At bud stage , the embryos were embedded and oriented in 0 . 8% low–melting point agarose in Danieau medium . The agarose was overlaid with medium , and a time-lapse movie was made using a confocal microscope ( Nikon eclipse C1; Nikon , Tokyo , Japan ) . Every 5 min , pictures were taken every 5 μm along the z-axis ( 170 μm in total ) over a period of 10 h . Subsequently , the data were modified , and individual cells of the neural plate were tracked using Imaris 4 . 2 ( Bitplane AG , Zurich , Switzerland ) software . Specific cells of the neural plate at bud stage were ablated by sucking these cells out with a manual syringe [6] ( Sutter Instrument Company , Novato , California , United States ) . For transplantation , these cells were then placed in the neural plate of wild-type or her5pac:egfp host embryos . Partly embryos were treated with cyclopamine ( 100 μM; Toronto Research Chemicals , North York , Canada ) . The consequences of these manipulations were then revealed by detection of expression domains of various forebrain markers . To follow the development of the basal plate , labelled cells were transplanted on top of the shield at 50% epiboly stage [26] . Once embryos reached bud stage , embryos showing a midline distribution of transplanted cells were further incubated up to 22 hours postfertilization ( hpf ) or 30 hpf .	During the earliest stages of development , the brain is first formed as a simple sheet of cells called the neural plate . Although the plate looks homogenous , it contains distinct domains that can be identified by differential gene expression . These domains correspond to distinct future brain areas . In this study , we examined gene expression patterns in an area of the neural plate that later forms the forebrain to show that well-defined cell populations will develop into specific forebrain regions , such as the prethalamus , thalamus , hypothalamus , and epithalamus . We then tested whether these early neural plate subterritories are fully committed to a particular forebrain identity . We found that precursors of the prethalamus are not replaceable by other neighbouring cells , because ablation of these cells results in loss of prethalamus development . Moreover , when prethalamus precursors were moved into the environment of the presumptive hindbrain , the cells still pursued their prethalamic differentiation program . Finally , when the prethalamic precursors were moved to areas of the future forebrain , they transformed the surrounding host forebrain . We propose that the committed prethalamic precursors play an instructive role in the regionalization of the developing forebrain .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "developmental", "biology", "danio", "(zebrafish)", "eukaryotes", "vertebrates", "teleost", "fishes", "neuroscience", "animals" ]	2007	The Prethalamus Is Established during Gastrulation and Influences Diencephalic Regionalization
Ghana is affected by regular cholera epidemics and an annual average of 3 , 066 cases since 2000 . In 2014 , Ghana experienced one of its largest cholera outbreaks within a decade with more than 20 , 000 notified infections . In order to attribute this rise in cases to a newly emerging strain or to multiple simultaneous outbreaks involving multi-clonal strains , outbreak isolates were characterized , subtyped and compared to previous epidemics in 2011 and 2012 . Serotypes , biotypes , antibiotic susceptibilities were determined for 92 Vibrio cholerae isolates collected in 2011 , 2012 and 2014 from Southern Ghana . For a subgroup of 45 isolates pulsed-field gel electrophoresis , multilocus sequence typing and multilocus-variable tandem repeat analysis ( MLVA ) were performed . Eighty-nine isolates ( 97% ) were identified as ctxB ( classical type ) positive V . cholerae O1 biotype El Tor and three ( 3% ) isolates were cholera toxin negative non-O1/non-O139 V . cholerae . Among the selected isolates only sulfamethoxazole/trimethoprim resistance was detectable in 2011 , while 95% of all 2014 isolates showed resistance towards sulfamethoxazole/trimethoprim , ampicillin and reduced susceptibility to ciprofloxacin . MLVA achieved the highest subtype discrimination , revealing 22 genotypes with one major outbreak cluster in each of the three outbreak years . Apart from those clusters genetically distant genotypes circulate during each annual epidemic . This analysis suggests different endemic reservoirs of V . cholerae in Ghana with distinct annual outbreak clusters accompanied by the occurrence of genetically distant genotypes . Preventive measures for cholera transmission should focus on aquatic reservoirs . Rapidly emerging multidrug resistance must be monitored closely . The World Health Organization ( WHO ) estimates that 3–5 million annual cases of cholera occur worldwide , resulting in 100 , 000–120 , 000 deaths [1] . Since its introduction to Africa during the current seventh cholera pandemic in 1970 , Vibrio cholerae caused regular vast epidemics across the continent with cumulative 3 , 762 , 902 case-notifications to WHO by 2013 [2] . In 2013 alone , 22 African countries reported 56 , 329 cholera cases . However , these numbers are considered to be substantially underestimated due to poorly functioning national epidemiological and laboratory surveillance systems , which are not able to detect the majority of mild disease presentations . Furthermore , systematic underreporting is common to avoid economic and political damage [3 , 4] . The first cholera case has been reported from Ghana in 1970 [5] . With 90% of the population at risk for cholera transmission , Ghana always ranks among the most affected countries on the African continent [4] . Ghana notified an annual average of 3 , 066 ( range: 50–10 , 628 ) cholera cases between 2000 and 2013 with an overall case fatality rate of 1 . 7% [2] . In the year 2014 , Ghana experienced an exceptionally large cholera outbreak with 28 , 975 infections notified to the World Health Organization between June and November [6] . Causes of this sudden increase in case numbers might be very diverse and calls for investigation in a timely manner so as to implement preventive measures accordingly . This recent epidemic might be explained by the extrinsic introduction of a new V . cholerae strain into a susceptible population , as seen for cholera outbreaks in Zimbabwe and Haiti in the years 2009 and 2011 respectively [7 , 8] . Alternatively , multiple , endemic , genetically non-related strains might be responsible for concomitant epidemics , as reported from Kenya [9] . In the latter scenario V . cholerae might persist in aquatic reservoirs and regular epidemics are triggered by low hygienic conditions and climatic factors such as increased rainfall or flooding [10–12] . In recent years , a number of molecular typing tools have been developed to elucidate the source and evolution of V . cholerae outbreaks [13] . This study aims to describe the 2014 cholera epidemic in Ghana and uses molecular subtyping techniques to detect responsible newly emerging or multi-clonal strains , which are then compared to strains that circulated during the 2011 and 2012 epidemics . Results will advise public health authorities whether to focus on monitoring of endemic environmental reservoirs or rather on surveillance of mobile populations , with cross-border epidemiological collaborations to prevent importation of V . cholerae . Within the Ghana Health Service , the Disease Surveillance Service supported by the National Public Health & Reference Laboratory ( NPHRL ) conducts cholera surveillance in Ghana . A case of cholera is defined according to the WHO standard case definition: If cholera is not known to be present in the area , a case of cholera is considered in a patient ≥5 years with severe dehydration or death from acute watery diarrhea , while during a cholera epidemic every patient aged ≥5 years with acute watery diarrhea and/or vomiting is considered as a case . Standardized line lists with suspected cholera cases are provided on a weekly basis by the District Health Management Teams to the Central Disease Surveillance Service in Accra , which collates information on name , place of residence , sex , age , disease onset , disease outcome and hospitalization of cases . A small subset of specimens is usually tested in peripheral laboratories by various methods and results are notified in the line lists . For laboratory confirmation , district and regional laboratories are encouraged to send suspected cholera stool samples to the NPHRL , where culture and serotype identification is performed . For the purposes of this study , all suspected cholera cases in the surveillance database from six Southern Ghanaian regions ( Greater Accra , Central , Western , Eastern , Ashanti , Volta ) from the year 2014 were extracted and described by sex , age and place of residence . Continuous variables were summarized as means with standard deviation ( median with interquartile range for non-normally distributed variables ) and dichotomous or categorical variables were summarized as proportions/percentages . Missing values were excluded from the analysis , thus the denominators for some comparisons differ . Surveillance data were computerized using Excel ( Microsoft , USA ) and statistical analysis was performed with Stata v . 12 . 1 ( Statacorp , Texas , USA ) . For spatio-temporal visualisation case data were imported into Arc GIS 10 . 0 ( ESRI: ArcGis Desktop: Release 10 . 2011 ) . Suspected cases were linked to the respective notifying districts . For the spatio-temporal visualisation , case data/district were then aggregated in six temporal groups , each covering five outbreak weeks . A total of 20 , 185 cases of cholera were reported to the Ghanaian Disease Surveillance Center from the Ashanti- , Central- , Eastern- , Greater Accra- , Western- and Volta regions in 2014 . The date of disease onset was reported for 20 , 120 cases , the remaining 65 cases were removed from the dataset for all other analyses . The earliest reported onset date was the 20th of May 2014 and the latest was the 11th of December 2014 , with a peak number of 2 , 853 cases in the 35th calendar week ( 25–31 August; Fig 1 ) . Age was reported for 19 , 863 cases and distributed with a median age of 26 years and an interquartile range ( IQR ) of 20–35 years . The median age of cases did not change during the course of the outbreak . The majority of cases was male ( 58 . 4%; n = 11 , 796 ) , and median age was not markedly different between males ( 26 years; IQR 20–35 ) and females ( 25 years; IQR 19–35 ) . The case fatality rate ( CFR ) was 0 . 8% ( 165 deaths ) with a higher median age among deceased of 34 years ( IQR 24–47 ) . Laboratory testing was performed in regional laboratories with rapid diagnostic tests from different suppliers for 496 out of 20 , 120 ( 2 . 5% ) suspected cases with a positivity rate of 53% ( 264/496 ) . Spatio-temporal analysis traced back the first cases during the first five weeks of the outbreak to four districts in the Eastern and Greater Accra regions ( Fig 2 ) . During the peak of the outbreak ( outbreak week 6–20 ) the majority of cases were centred around the city of Accra , spreading in northward-direction to the Ashanti region and to the East along the coast . During the last 10 weeks of the outbreak , only three districts within the Ashanti and the Volta region reported cases . Antimicrobial susceptibility testing identified mono-resistance towards sulfamethoxazole/trimethoprim in 100% ( 18/18 ) of isolates from 2011 ( Table 1 ) . The resistance pattern changed in 2012 with 75% ( 9/12 ) of isolates expressing resistance against sulfamethoxazole/trimethoprim , nalidixic acid in combination to reduced susceptibility against ampicillin and ciprofloxacin . The same resistance profile was detected in 95% ( 59/62 ) of isolates in 2014 . No resistance against chloramphenicol , gentamycin or tetracycline was observed for any of the isolates . The selected 92 V . cholerae isolates consisted of 88 ( 95 . 7% ) V . cholerae O1 biotype El Tor serotype Ogawa , 1 ( 1 . 1% ) V . cholerae O1 biotype EL Tor serotype Inaba and 3 ( 3 . 3% ) non-01/non-O139 V . cholerae . Apart from the non-O1/non-O139 V . cholerae strains with no cholera toxin production , all isolates produced the classical type cholera toxin . Out of these 92 isolates , 11 isolates from 2011 , eight isolates from 2012 and 26 isolates from 2014 were genotyped by MLST , PFGE and MLVA ( S1 Fig ) . All V . cholerae O1 belonged to the sequence type ( ST ) 69 , while the three non-O1/non-O139 isolates presented with new alleles ( isolate 059: metE allele 108; isolate 160: metE allele 109; and isolate 169: metE allele 110 and pyrC allele 85 ) and could not be allocated to any known ST . Theses strains were assigned new ST numbers ( isolate 059: ST211 , isolate 160: ST212 and isolate 169: ST213; corresponding to IDs 196 , 197 and 198 in the V . cholerae pubmlst database: http://pubmlst . org/vcholerae/ ) , which are closest related to the already known ST40 and ST39 . As shown in Fig 3 , using a cutoff of 95% , PFGE divided all strains into three main clusters and 11 pulsotypes ( Fig 3 ) . All 2011 isolates clustered together ( cluster A ) as 4 different pulsotypes . Isolates from 2012 belonged to 4 different pulsotypes dispersed in clusters A , B and C . All V . cholerae O1 from 2014 showed the same pulsotype and were grouped in cluster B . Finally , the non-O1/non-O139 isolates collected in 2014 showed different pulsotypes and were grouped outside clusters A , B or C . Neither regional nor district specific pulsotypes had been detected . MLVA differentiated the isolates into three clonal complexes ( CC1-CC3 ) and 22 genotypes ( Fig 4 ) . The majority of 22 out of 26 isolates from 2014 clustered within CC1 together with one isolate from 2012 . Distinct from the CC1 were the three non-O1/non-O139 and one O1 V . cholerae isolate ( Isolate 146 ) from Ledzekuku-Krowor District ( Greater Accra Region ) . For 2012 , five out of eight isolates grouped together in CC2 , while the other three isolates were genetically distinct , within or close to CC1 and CC3 . Similarly , 10 of the 11 isolates from 2011 were classified into CC3 , and one strain clustered outside with three alleles difference to the other CC3 strains . Analogous to the PFGE distribution , subtypes did not cluster by region or by district . The Ghanaian Cholera outbreak in 2014 was caused by Vibrio cholerae O1 biotype El Tor carrying the classical cholera toxin . Molecular subtyping data of three outbreak years illustrate major annual outbreak clusters with co-circulating genetically distant genotypes , which might hint to an endemic reservoir of V . cholerae in Ghana . Public health authorities must be vigilant and take steps to prevent cholera transmission through aquatic reservoirs , particularly within urban agglomerations during the start of the rainy season . Considering the rapidly emerging multidrug resistance among V . cholerae isolates , laboratories are encouraged to monitor antimicrobial susceptibility closely .	The bacterium Vibrio cholerae is mainly transmitted faecal-orally via human-to-human contact or via environmental water sources in which V . cholerae is able to persist . West Africa , including Ghana , is regularly affected by Cholera epidemics , in particular during rainy seasons . In 2014 , Ghana experienced an exceptionally large outbreak with over 20 , 000 cases , which raised questions about newly emerging V . cholerae strains in this region . In this study , we described the duration , the geographical spread and demographics of the outbreak using data from the National Ghanaian Surveillance system . Further , we characterized outbreak isolates from the outbreak years 2011 , 2012 and 2014 by three different subtyping methods . These analyses revealed strains with different genetic background and increasing antibiotic resistance circulating during each outbreak year . These data suggest that V . cholerae has an endemic reservoir in the environment and selection pressure results in a highly heterogeneous population of V . cholerae with a few strains evolving into pathogenic clones during each outbreak period . Public health authorities must be vigilant to prevent cholera transmission through aquatic reservoirs , particularly within urban agglomerations during the start of the rainy season . The rapidly emerging antibiotic resistance has to be monitored closely .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "antimicrobials", "medicine", "and", "health", "sciences", "toxins", "pathology", "and", "laboratory", "medicine", "pathogens", "vibrio", "drugs", "tropical", "diseases", "microbiology", "geographical", "locations", "toxicology", "toxic", "agents", "bacterial", "diseases", "vibrio", "cholerae", "antibiotics", "neglected", "tropical", "diseases", "pharmacology", "bacteria", "bacterial", "pathogens", "africa", "infectious", "diseases", "cholera", "antimicrobial", "resistance", "medical", "microbiology", "epidemiology", "microbial", "pathogens", "people", "and", "places", "ghana", "el", "tor", "disease", "surveillance", "microbial", "control", "biology", "and", "life", "sciences", "organisms" ]	2016	Molecular Epidemiology and Antibiotic Susceptibility of Vibrio cholerae Associated with a Large Cholera Outbreak in Ghana in 2014
As genetic information is transmitted through successive generations , it passes between pluripotent cells in the early embryo and germ cells in the developing foetus and adult animal . Tex19 . 1 encodes a protein of unknown function , whose expression is restricted to germ cells and pluripotent cells . During male spermatogenesis , Tex19 . 1 expression is highest in mitotic spermatogonia and diminishes as these cells differentiate and progress through meiosis . In pluripotent stem cells , Tex19 . 1 expression is also downregulated upon differentiation . However , it is not clear whether Tex19 . 1 has an essential function in germ cells or pluripotent stem cells , or what that function might be . To analyse the potential role of Tex19 . 1 in pluripotency or germ cell function we have generated Tex19 . 1−/− knockout mice and analysed the Tex19 . 1−/− mutant phenotype . Adult Tex19 . 1−/− knockout males exhibit impaired spermatogenesis . Immunostaining and histological analysis revealed defects in meiotic chromosome synapsis , the persistence of DNA double-strand breaks during meiosis , and a loss of post-meiotic germ cells in the testis . Furthermore , expression of a class of endogenous retroviruses is upregulated during meiosis in the Tex19 . 1−/− testes . Increased transposition of endogenous retroviruses in the germline of Tex19 . 1−/− mutant mice , and the concomitant increase in DNA damage , may be sufficient to disrupt the normal processes of recombination and chromosome synapsis during meiosis and cause defects in spermatogenesis . Our results suggest that Tex19 . 1 is part of a specialised mechanism that operates in the germline to repress transposable genetic elements and maintain genomic stability through successive generations . The germ cells of sexually reproducing organisms have a unique role in generating genetic diversity and transmitting genetic information from one generation to the next . Establishment of the germline in mammals involves the induction of germ cells from pluripotent epiblast cells through the action of extra-embryonic ectoderm-derived bone morphogenetic proteins and occurs comparatively late in development , commencing around day 6 . 25 days post coitum ( dpc ) in mice [1]–[4] . At around 12 . 5 dpc -13 . 5 dpc the sexually dimorphic germ cells become committed to develop along either a male or a female pathway and start to initiate sex-specific differentiation [5] . Although there are numerous differences in the differentiation of the germline and in the timing and regulation of meiosis between the sexes , the fundamental events of meiosis that increase genetic diversity and reduce ploidy of the gametes are common to both . The main group of genes that have been shown to be required for mammalian meiosis are those involved in the recombination and synapsis of homologous chromosomes . Mice carrying loss-of-function mutations in these genes , such as Atm , Dmc1 , γH2AX , Mlh1 , Msh5 , Rec8 , Rad51 , Smc1β , Spo11 , Sycp1 , Sycp2 , Sycp3 , Syce2 and Tex14 , typically exhibit defects in chromosome synapsis in both sexes , although male and female germ cells can exhibit different responses to these defects [6]–[8] . A second group of genes that are required for progression through meiosis are those involved in repression of transposable genetic elements . Retrotransposons , for example long interspersed repeats ( LINEs ) , short interspersed repeats ( SINEs ) , and endogenous retroviruses such as intracisternal A-particles ( IAPs ) are the major class of transposable genetic elements in mammals and comprise around 37 . 5% of the mouse genome [9] . To allow new transposition events to propagate through subsequent generations , retrotransposons have evolved to be active in the germline . Accordingly , germ cells appear to have evolved mechanisms to reduce the mutational load of retrotransposon activity . Mutations in genes involved in mediating DNA methylation-dependent transcriptional repression of retrotransposons cause increased expression of retrotransposons and defects in chromosome synapsis during meiosis in either male or female germ cells [10] , [11] . For example Dnmt3L is a catalytically inactive member of the DNA methyltransferase family that is expressed in foetal germ cells but is absent by 6 days post partum ( dpp ) [11] . Male mice null for this gene do not methylate dispersed repeat DNA during foetal germ cell development , and express LINEs and a class of endogenous retroviruses known as intracisternal A-particles ( IAPs ) in the germline [11] . Dnmt3L mutant male mice also exhibit meiotic abnormalities that result in a loss of post-meiotic germ cells in the testis [11] . The Dnmt3L mutant phenotype suggests that epigenetic changes that occur in foetal germ cells can cause meiotic defects later in germ cell development . A second example is provided by the murine piwi-related genes , which encode germline-specific proteins that are associated with a class of small germline-specific piwi-interacting RNAs ( piRNAs ) and are also required to repress retrotransposons during spermatogenesis [12]–[14] . Mili and Miwi2 both appear to be involved in de novo methylation of LINE and IAP elements during germ cell development in male embryos , and both mutants exhibit reduced DNA methylation and increased expression of LINE and IAP element in the testis , and defects in chromosome synapsis during meiosis in male germ cells [12]–[14] . The mechanism by which increased expression of retrotransposons results in the defects in chromosome synapsis seen in these mutant mice is unknown , but the phenotype of these mutant mice suggests that repression of transposable genetic elements is required to allow germ cells to progress through meiosis . A set of testis expressed ( Tex ) genes has been identified in a subtractive hybridisation screen for genes expressed in spermatogonia but not somatic tissue [15] . One of these genes , Tex19 . 1 ( AAH53492 . 1 ) , was found in a screen for potential RNA-targets of the germline-specific RNA binding protein Dazl by immunoprecipitation and microarray analysis [16] . Further unpublished work in this laboratory and work recently published by Kuntz et al . [17] confirmed an earlier report that Tex19 . 1 is a “pluripotent cell expressed gene” [18] . Humans and primates possess a single Tex19 gene in their genome , but in rodents a recent duplication has produced a two-gene family arranged as divergently transcribed genes separated by 29kb of DNA [17] . While expression of murine Tex19 . 1 is restricted to pluripotent stem cells and developing germ cells , Tex19 . 2 is expressed in the testis somatic tissues and does not appear to be restricted to germ cells or pluripotent stem cells in mice [17] . The expression pattern of Tex19 . 1 suggests that the protein could have an important role in pluripotency or germ cell function . Since the sequence of Tex19 . 1 gives no clue to the biochemical function of this protein we decided to take a genetic approach to determining the function of this gene in the germline . In this paper we report that targeted deletion of Tex19 . 1 in mice results in upregulation of endogenous retrovirus expression in testicular germ cells , perturbed chromosome synapsis during meiosis , and impaired spermatogenesis . Tex19 . 1 knockout mice were generated by replacing the Tex19 . 1 open reading frame with a neomycin selection cassette by homologous recombination in E14 embryonic stem cells [19] . Homologous regions were cloned by PCR from E14 embryonic stem cell genomic DNA using primers listed in Supplementary Table S1 . The Tex19 . 1 targeting vector was linearised , electroporated into E14 embryonic stem cells , and neomycin-resistant clones screened for the desired integration event by PCR and Southern blot . Tex19 . 1+/− ES cells were used to generate Tex19 . 1−/− knockout mice by blastocyst injection and breeding as described [19] . Mice were genotyped by multiplex PCR using primers listed in Supplementary Table S1 . Phenotypic analysis was performed on mice with a 129/Ola x CD1 mixed genetic background . Generation and analysis of Tex19 . 1−/− knockout mice was performed under a UK Home Office project licence with approval from an institutional ethics committee . Anti-Tex19 . 1 antibodies were raised in rabbits using the synthetic peptide 78ESEQEPGPEQDAWRG92 ( Eurogentec ) . This peptide was designed to be specific to Tex19 . 1 and is not present in the Tex19 . 2 protein sequence . Antibodies were affinity purified from sera using the immunising peptide immobilised on a Sulfolink column ( Pierce ) according to manufacturer's instructions . Testes were recovered from mice , fixed at 4°C overnight in 4% paraformaldehyde in phosphate buffered saline ( PBS ) and embedded in paraffin wax . 6 µm-thick sections were dewaxed in xylene , rehydrated , and antigen retrieval performed by boiling slides for 15 minutes in 0 . 01 M sodium citrate , pH 6 . 0 . Sections were blocked , incubated with rabbit anti-Tex19 . 1 primary antibody at 1∶50 , and bound antibody detected using the DAKOvision ABC diaminobenzidine ( DAB ) kit as described by the manufacturer ( DakoCytomation ) . For peptide competition , anti-Tex19 . 1 antibodies were pre-incubated with 5 nM immunising peptide . For immunostaining of cultured cells , cells were fixed for 30 minutes at room temperature with 3 . 7% formaldehyde in PBS , then blocked with PBS containing 5% serum and 0 . 01% Tween-20 . Cells were incubated with rabbit anti-Tex19 . 1 primary antibody at 1∶100 , then fluorescently labelled secondary antibodies at 1 µg/mL ( Invitrogen ) . DNA was counterstained with 2 µg/mL DAPI . E14 embryonic stem cells or 13 dpp postnatal testes were lysed in cytoplasmic lysis buffer ( 10 mM Hepes pH 7 . 6 , 3 mM MgCl2 , 40 mM KCl , 50 mM β-glycerophosphate , 5% glycerol , 0 . 5% Igepal CA-630 , 2 mM NaF , 1 mM Na3VO4 , 2 mM DTT and protease inhibitors ) for 5 minutes on ice and the whole cell lysate centrifuged for 5 minutes at 1000g at 4°C . The nuclear pellet was resuspended in Laemmli buffer , boiled for 5 minutes and sonicated to disrupt genomic DNA . The cytosolic supernatant was mixed with Laemmli buffer and boiled for 5 minutes . Equivalent proportions of each fraction were separated by SDS-PAGE then Western blotted . Testis was homogenised in Laemmli buffer , boiled for 5 min and sonicated to disrupt genomic DNA . Western blotting was performed using standard procedures [20] . Tex19 . 1 was detected with rabbit anti-Tex19 . 1 polyclonal antibodies used at a 1∶200 dilution , mouse anti-Gapdh antibodies ( Abcam ) were used at 1∶1000 , mouse anti-HP1α antibodies ( Chemicon ) at 1∶2500 , and rabbit anti-histone H3 antibodies ( Abcam ) at 1∶20000 . Peroxidase-conjugated secondary antibodies and enhanced chemiluminescence were used to detect primary antibodies . Non-radioactive Southern blots were performed using a digoxigenin-labeled DNA probe generated using primers listed in Supplementary Table S1 , and alkaline phosphatase-conjugated anti-digoxigenin antibodies , essentially as described by the manufacturer ( Roche ) . Testis RNA was isolated with Trizol ( Invitrogen ) according to the manufacturer's protocol and reverse transcription performed with Superscript III ( Invitrogen ) on 1 µg RNA per reaction using oligo dT primer . Primers for RT-PCR are listed in Supplementary Table S1 . For quantitative PCR ( qPCR ) , random-primed cDNA was generated from total RNA using Superscript III ( Invitrogen ) . qPCR was performed using SYBR Green PCR System ( Applied Biosystems ) and a PTC-200 thermal cycler equipped with a Chromo4 continuous fluorescence detector and Opticon Monitor software ( MJ Research ) . Primers for qPCR are listed in Supplementary Table S1 . Five technical replicates were performed for each biological sample , and the relative changes in gene expression determined using the ΔΔ−2Ct method as described [21] . As Tex19 . 1 is expressed in the germ cells in the testis the Sertoli cell marker Sdmg1 [22] was used to normalise cDNAs prepared from different animals to reduce the probability that the cDNAs were being normalised to a transcript whose level could be influenced by loss of Tex19 . 1 . Both testes from each adult animal ( 6–36 weeks old ) were weighed , and the mean testis weight was used for statistical comparison . For sperm count one epididymis from each animal was homogenised in 1 mL 1% sodium citrate and incubated for 5 minutes at room temperature to allow the debris to settle . Sperm in the supernatant was then counted with a hemocytometer . Testes were fixed for 4–6 hours in Bouin's solution ( Sigma-Aldrich ) at room temperature , then embedded in wax . For histological analysis 6 µm sections were dewaxed with xylene , rehydrated , then stained with hematoxylin and eosin . Immunostaining of chromosome spreads from meiotic spermatocytes was performed essentially as described [23] . Briefly , testes were homogenized in PBS and 0 . 1 mL of cells were incubated in 0 . 5 mL 5% sucrose on a microscope slide for 1 hour . Cells were lysed with 0 . 1 mL 0 . 05% Triton-X-100 for 10 minutes , and fixed with 0 . 8 mL of fixing solution ( 2% paraformaldehyde , 0 . 02% SDS in PBS ) for 1 hour . The slides were then washed , blocked with 5% serum , 0 . 1% Tween in PBS and incubated with primary antibodies for 1 hour . Mouse anti-Sycp3 antibodies ( Abcam ) were used at a 1∶2000 dilution , rabbit anti-Sycp1 antibodies ( Abcam ) at 1∶250 , rabbit anti-γH2AX antibodies ( Upstate Biotechnology ) at 1∶200 , and mouse anti-Rad51 antibodies ( Upstate Biotechnology ) at 1∶125 . Fluorescently labelled secondary antibodies were used at 1 µg/mL ( Invitrogen ) , and DNA was stained with 2 µg/mL DAPI . Chromosome spreads for metaphase I analysis were prepared as described in [24] . Briefly , testes were incubated for 20 minutes in 1% sodium citrate , minced with scissors , and the cells harvested by centrifugation . Cells were then washed and resuspended in fixing solution ( 3∶1 methanol∶glacial acetic acid ) , dropped onto slides , and the resulting chromosome spreads were stained with Giemsa solution . 100 metaphase I spreads were scored per animal , and two animals scored for each genotype . A 460 bp fragment of the MMERVK10C endogenous retrovirus was amplified by RT-PCR from Tex19 . 1−/− mutant testes using primers listed in Supplementary Table S1 and cloned into pBluescript II SK+ ( Stratagene ) . Sense and anti-sense digoxigenin-labelled riboprobes were generated using T3 and T7 RNA polymerase according to the supplier's instructions ( Roche ) . In situ hybridisation on 6 µm wax sections of Bouin's-fixed testis tissue was performed essentially as described [25] using 100 ng/mL digoxigenin-labelled probe and a hybridisation temperature of 50°C . Bound probe was detected with alkaline phosphatase-conjugated anti-digoxigenin antibodies ( Roche ) and BCIP/NBT precipitating stain ( Vector Labs ) , then sections counterstained with nuclear fast red according to manufacturer's instructions . The antisense MMERVK10C digoxigenin-labelled riboprobe was also used for non-radioactive Northern blotting of 1 µg testis RNA as described [22] . Spermatogenesis in the adult testis involves the differentiation of a small pool of spermatogonial stem cells into large numbers of mature sperm . Within the testis spermatogenesis takes place in the seminiferous tubules where the mitotic spermatogonia reside at the outermost edge of the tubule , and progressive stages of differentiation are found as layers of meiotic spermatocytes then haploid spermatids located more and more centrally towards the lumen of the tubule [26] . Published RT-PCR expression data for Tex19 . 1 in purified spermatogenic cell populations suggests that Tex19 . 1 expression is highest in mitotic spermatogonia , decreases as the spermatocytes progress through meiosis , and is present at low levels in round spermatids [27] . To establish the expression pattern of Tex19 . 1 protein in male testis during spermatogenesis we raised anti-peptide antibodies to Tex19 . 1 . Immunohistochemistry on mouse testis shows strong cytoplasmic expression of Tex19 . 1 in spermatogonia that is downregulated as these cells differentiate and progress through meiosis ( Figure 1A–C ) . We were able to detect cytoplasmic Tex19 . 1 protein in some meiotic spermatocytes ( Figure 1C ) , but not in others ( Figure 1B ) suggesting that Tex19 . 1 protein expression is switched off as the germ cells proceed through meiosis . The expression of Tex19 . 1 protein in spermatogonia and spermatocytes is consistent with that observed for Tex19 . 1 mRNA by in situ hybridisation ( Figure S1A–C ) . Our finding that Tex19 . 1 is present in the cytoplasm of spermatogonia and early spermatocytes in the adult testis is not consistent with the published nuclear localisation of Tex19 . 1 protein in embryonic stem cells [17] . We confirmed that our anti-Tex19 . 1 antibody detects Tex19 . 1 and not a cross-reacting antigen by blocking the anti-Tex19 . 1 immunohistochemistry signal by competition with the immunising peptide , and by immunohistochemistry on Tex19 . 1−/− knockout testes ( Figure S1D–F ) . We were also able to detect a predominantly cytoplasmic subcellular localisation of Tex19 . 1 by immunostaining germ cells isolated from 14 . 5 dpc embryonic testes ( Figure S1G ) . Again the cytoplasmic anti-Tex19 . 1 immunostaining could be competed with the immunising peptide , and was absent in germ cells from Tex19 . 1−/− knockout embryos ( Figure S1H , I ) . These data suggest that the anti-Tex19 . 1 antibody used in this present study specifically recognises endogenous Tex19 . 1 in germ cells , and that at least some Tex19 . 1 is present in the cytoplasm of spermatogonia and early spermatocytes in adult mouse testes . To investigate whether the discrepancy between the cytoplasmic localisation of Tex19 . 1 in germ cells presented in this study and the nuclear localisation of Tex19 . 1 in embryonic stem cells described previously [17] is caused by the difference between the cell types studied we performed immunostaining for Tex19 . 1 on embryonic stem cells . In contrast to the previous report [17] , we found that Tex19 . 1 is predominantly cytoplasmic in embryonic stem cells ( Figure 1D–F ) . To exclude the possibility that our anti-Tex19 . 1 antibody is unable to detect a nuclear population of Tex19 . 1 due to loss or masking of the epitope during the immunohistochemical procedures we biochemically fractionated 13 dpp prepubertal testes and embryonic stem cells into nuclear and cytoplasmic fractions . On Western blots the 42 kDa Tex19 . 1 band was barely detectable in the nuclear fraction , but was easily detectable in equivalent loadings of the whole cell lysate and cytoplasmic fractions ( Figure 1G , H ) . The observed size of the anti-Tex19 . 1 band in the Western blots ( 42 kDa ) correlates well with the predicted molecular weight of Tex19 . 1 ( 40 . 4 kDa ) . This 42 kDa band appears to be endogenous Tex19 . 1 as it is not present in testes from Tex19 . 1−/− knockout animals ( Figure 2E ) . The biochemical fractionation of embryonic stem cells and testes therefore confirms the predominantly cytoplasmic subcellular localisation of Tex19 . 1 that we have observed by immunohistochemistry and immunostaining . Taken together the data presented here strongly suggests that Tex19 . 1 is a predominantly cytoplasmic protein in embryonic stem cells and germ cells . Germ cells in many species possess specialised cytoplasmic structures termed nuage that are implicated in RNA metabolism . Although Tex19 . 1 appears to be a predominantly cytoplasmic germ cell protein , the subcellular localisation and cell-type distribution of Tex19 . 1 appears to be distinct from the nuage component Tdrd1 [28] ( Figure S2 ) . Thus Tex19 . 1 does not appear to be a novel component of nuage . As the subcellular localisation of Tex19 . 1 does not provide any major insight into what the cellular function of this protein might be , and the Tex19 . 1 protein sequence does not contain any functional domains to illuminate the potential biochemical function of this protein , we decided to take a genetic approach to analyse the function of Tex19 . 1 in the germline . In order to investigate the function of Tex19 . 1 in germ cell development we generated Tex19 . 1−/− knockout mice . The Tex19 . 1 open reading frame was replaced with a neomycin-resistance cassette by homologous recombination in embryonic stem cells ( Figure 2A ) , and the targeted deletion confirmed by Southern blotting ( Figure 2B ) . The heterozygous Tex19 . 1+/− embryonic stem cells were used to generate chimaeric mice by blastocyst injection , and the Tex19 . 1− mutant allele bred to homozygosity ( Figure 2C ) . Tex19 . 1−/− homozygous pups were born from heterozygous crosses at a sub-Mendelian frequency ( 72 wild-type , 131 heterozygous , 40 homozygous pups born from heterozygous matings , significant deviation from expected Mendelian 1∶2∶1 ratio , χ2-test p<0 . 01 ) . The low rate of recovery of Tex19 . 1−/− homozygous animals at birth indicates that some Tex19 . 1−/− homozygous embryos are lost during embryonic development . To confirm that the Tex19 . 1− mutant allele removes Tex19 . 1 mRNA and protein we performed RT-PCR on Tex19 . 1−/− testis cDNA with Tex19 . 1-specific primers ( Figure 2D ) and Western blotting on Tex19 . 1−/− testis protein extract with anti-Tex19 . 1 antibodies ( Figure 2E ) . Both methods show that Tex19 . 1 is not expressed in the testes of Tex19 . 1−/− homozygous mice . We conclude that the Tex19 . 1− allele that we have produced is a null allele and that Tex19 . 1 function is ablated in the Tex19 . 1−/− homozygous mice . The surviving Tex19 . 1−/− knockout mice are apparently healthy , with overtly normal morphology and behaviour . However both male and female Tex19 . 1−/− knockout mice have reduced fertility . Tex19 . 1−/− knockout females have a mean litter size of 5 . 0±2 . 2 SD , n = 9 compared to a litter size of 10 . 6±2 . 9 , n = 23 for Tex19 . 1+/− heterozygous females ( Student's t-test p<0 . 01 ) . The reduced fertility in Tex19 . 1−/− homozygous females is consistent with expression of Tex19 . 1 in embryonic ovaries [17] , and a detailed analysis of the cause of the subfertility in the Tex19 . 1−/− knockout female mice will be published elsewhere . Similarly , Tex19 . 1−/− knockout male mice are also severely subfertile when test-mated with wild-type female mice . Although one out of the eleven Tex19 . 1−/− knockout males tested for fertility was able to sire offspring , the remaining Tex19 . 1−/− knockout males were infertile . The sterile Tex19 . 1−/− knockout male mice were apparently able to mate with the wild-type females to produce a copulation plug , but these females did not give birth to any pups . Tex19 . 1−/− knockout male mice have smaller testes ( Figure 3A , B ) , with the median testis weights of adult animals reduced from 111 mg in Tex19 . 1+/+ wild type and Tex19 . 1+/− heterozygous mice to 42 . 5 mg in Tex19 . 1−/− knockout littermates ( Mann Whitney U-test , p<0 . 01 ) . Furthermore , the median epididymal sperm count is reduced to 1 . 3×105 in Tex19 . 1−/− knockout mice from 1 . 3×107 in wild type and heterozygous littermates ( Mann Whitney U-test , p<0 . 01 , Figure 3C ) suggesting that spermatogenesis is defective in the Tex19 . 1−/− knockout testes . We have not been able to detect any difference in testis weight or sperm count between Tex19 . 1+/+ wild-type and Tex19 . 1+/− heterozygous animals . The extent of the spermatogenesis defect in the Tex19 . 1−/− knockout males varied between individual animals . When sperm was counted , a strong reduction was observed for most of the animals , but for the single fertile animal the sperm count was close to normal levels ( Figure 3C ) . A similar variation in testis weight was also evident amongst Tex19 . 1−/− knockout animals ( Figure 3B ) . This phenotypic variation was not influenced by the age of the mice at the time of analysis . Age-matched adult mice were analysed at 6 weeks , 3 months , 6 months and 9 months during the course of this study , and there appeared to be no correlation between the severity of the phenotype and the age at which the adult mice were examined . Rather phenotypic variation was observed in adult mice at all ages ( R . O . and I . R . A . , data not shown ) . The outbred component of the genetic background of these mice may contribute to this variation . We investigated the spermatogenesis defect in Tex19 . 1−/− knockout mice further by examining the testis histology in these animals . We did not detect any overt differences in testis histology between Tex19 . 1+/+ wild-type and Tex19 . 1+/− heterozygous animals . However , Tex19 . 1−/− knockout testes have considerably narrower seminiferous tubules than their wild-type or heterozygous littermates due to a reduction in the number of post-meiotic germ cells ( Figure 4A , B ) . This phenotype was also subject to some heterogeneity . In animals with a more severe phenotype all postmeiotic cell-types were missing and the most advanced meiotic cells were in pachytene stage ( Figure 4C , D ) . In animals with a less severe phenotype a proportion of cells were able to complete meiosis and haploid cells could be detected , although often in comparatively low numbers ( Figure 4C , E ) . Like the testis weight and sperm count phenotypes described in the previous section , there appeared to be no correlation between the severity of the testis histology phenotype and the age at which the adult mice were analysed ( from 6 weeks to 9 months ) . To test whether the reduction in the number of post-meiotic germ cells in Tex19 . 1−/− knockout testes arises from a decrease in the spermatogonial mitotic divisions or apoptosis of differentiating germ cells , we counted the number of B-type spermatogonia , early meiotic cells and apoptotic cells in testis sections from Tex19 . 1+/− heterozygous and Tex19 . 1−/− knockout animals . B-spermatogonia and early meiotic cells were identified by their location and histological appearance in the seminiferous tubules [26] , and apoptotic cells were identified using the TUNEL assay to label fragmented chromatin . Whereas the number of B-type spermatogonia and early meiotic cells did not differ between Tex19 . 1+/− heterozygous and Tex19 . 1−/− knockout testes ( R . O . , data not shown ) , TUNEL staining showed an increase in the number of dying cells in adult Tex19 . 1−/− knockout testis ( Figure S3A , B , N ) . In more severe Tex19 . 1−/− knockout seminiferous tubules , TUNEL-positive cells were found within or next to layers of meiotic germ cells ( Figure S3C ) , but even at high magnification the nuclear morphology of these TUNEL-positive cells was not distinct enough to allow their developmental stage to be unambiguously identified ( Figure S3H–J ) . In less severe Tex19 . 1−/− knockout seminiferous tubules , TUNEL-positive cells could also be found between the layers of meiotic germ cells and post-meiotic round spermatids ( Figure S3D ) . At higher magnification , some of these TUNEL-positive cells could be identified as metaphase I spermatocytes ( Figure S3E–G ) . In order to further define the point during spermatogenesis when the Tex19 . 1−/− knockout cells are dying we examined the synchronous first wave of spermatogenesis that occurs in prepubertal mice . The first wave of spermatogenic germ cells initiates meiosis at around 10 dpp in the prepubertal testis , and progresses through the pachytene stage of meiosis from around 14 to 20 dpp to produce the first post-meiotic round spermatids around 21 dpp , and mature sperm at around 31 dpp [29] , [30] . Analysis of apoptosis ( Figure S3N ) and testis histology ( Figure S4 ) at various stages of prepubertal testis development revealed no overt differences in testis histology and no statistically significant increase in apoptosis at 16 dpp in Tex19 . 1−/− knockout testes . However by 19–22 dpp , a reduction in the number of meiotic and post-meiotic germ cells and an increase in the frequency of cell death are both evident in Tex19 . 1−/− knockout testes ( Figures S3N , S4 ) . In 22 dpp testes , clusters of TUNEL-positive cells can be seen within the layer of pachytene germ cells that line the lumen of the seminiferous tubule suggesting that at least some apoptosis is occurring at the pachytene stage of meiosis ( Figure S3K–M ) . The high level of apoptosis in the Tex19 . 1−/− knockout testes increases by 29–31 dpp to the level seen in adult testes ( Figure S3N ) . This data suggests that the reduction in the number of post-meiotic germ cells and increased levels of apoptosis seen in the adult Tex19 . 1−/− knockout testes is at least partly due to some Tex19 . 1−/− knockout germ cells initiating apoptosis during the pachytene stage of meiosis , and some Tex19 . 1−/− knockout germ cells initiating apoptosis during metaphase I . Although the vast majority of the Tex19 . 1−/− null testes examined contained differentiating germ cells , two of thirty analysed knockout animals had an extremely severe phenotype with one testis that completely lacked germ cells . One of these agametic testes was isolated from a 31 dpp prepubertal mouse ( Figure 4F ) suggesting that this extreme phenotype is indicative of defects occuring during embryonic or early post-natal germ cell development rather than a progressive loss of spermatogonial stem cells in an ageing adult testis . However , as only a small number of testes exhibited this phenotype , we were not able to study this extreme phenotype further and instead focused on the meiotic phenotype evident in the vast majority of the Tex19 . 1−/− mutant testes . We next attempted to determine the cause of the increased apoptosis in Tex19 . 1−/− null testes . Defects in homologous chromosome synapsis or homologous recombination during meiotic prophase can cause apoptosis in late pachytene spermatocytes [31] , [32] . Therefore we used immunocytochemistry on meiotic chromosome spreads to analyse chromosome synapsis and homologous recombination in Tex19 . 1−/− knockout testes . In order to analyse chromosome synapsis , meiotic chromosome spreads were stained using Sycp3 as a marker for lateral elements of meiotic chromosomes and Sycp1 as a marker for synapsed homologous chromosomes [6] . In wild-type pachytene cells the autosomal chromosome axes stain completely for both markers , whereas the X and Y sex chromosomes remain largely asynapsed with only a small area of Sycp1 staining in the pseudo-autosomal region ( Figure 5A ) . In contrast , about half the pachytene cells in Tex19 . 1−/− homozygotes have Sycp3-stained autosomal chromosomal axes that lack Sycp1 staining ( Figure 5B–D , I ) . The asynapsed chromosomes in Tex19 . 1−/− knockout cells did not appear to be arranged in homologous pairs ( Figure 5D ) . However it is not clear whether the asynapsed chromosomes have never paired in Tex19 . 1−/− knockout spermatocytes , or have paired but have subsequently fallen apart . In some of the incompletely synapsed Tex19 . 1−/− knockout cells , some chromosomes appeared to form chains linked by regions of apparent non-homologous synapsis ( Figure 5C , asterisk ) . Incompletely synapsed pachytene cells comprise less than 1% of spreads from Tex19 . 1+/+ wild-type or Tex19 . 1+/− heterozygotous testes ( Figure 5I ) . Thus Tex19 . 1−/− knockout animals exhibit defects in homologous chromosome synapsis during male meiosis . During meiotic prophase , homologous recombination starts prior to homologous chromosome pairing and synapsis [33] . As progression of homologous recombination and chromosome synapsis are interdependent on each other [6] , [7] , we investigated whether the chromosome synapsis defect in Tex19 . 1−/− knockout spermatocytes was a consequence of an earlier defect in the initiation of homologous recombination . The appearance of DNA double strand breaks and the formation of early recombination foci during meiotic prophase can be detected by immunostaining for the phosphorylated histone γH2AX and the recombinase enzyme Rad51 respectively [33] , [34] . γH2AX staining is normally present on chromatin during the leptotene and zygotene stages of early meiotic prophase . As synapsis proceeds during zygotene , the DNA double strand breaks are resolved , resulting in γH2AX staining disappearing from the autosomal chromosomes , but not the sex chromosomes . In normal Tex19 . 1+/+ wild-type pachytene cells , chromosome synapsis is complete and only the sex chromosomes stain for γH2AX ( Figure 5E ) . However , the incompletely synapsed pachytene cells in Tex19 . 1−/− knockout testes , exhibit strong diffuse γH2AX staining ( Figure 5F ) . This γH2AX staining is localised to the regions of the chromosome spreads that contain the unsynapsed chromosomes ( Figure 5F ) . Similarly , immunostaining for the early recombination foci marker Rad51 , which largely disappears from autosomal chromosomes as synapsis proceeds , suggests that Rad51 foci are formed in Tex19 . 1−/− knockout spermatocytes , but are not resolved or matured on the unsynapsed chromosomes ( Figure 5G , H ) . Thus the formation of DNA double strand breaks and the assembly of early recombination foci both appear to be occurring in Tex19 . 1−/− knockout spermatocytes . This suggests that the defect in meiotic chromosome synapsis that we have observed in Tex19 . 1−/− spermatocytes does not appear to be a secondary consequence of impaired initiation of homologous recombination . Rather , the presence of DNA double strand breaks and early recombination foci in the unsynapsed regions of the incompletely synapsed Tex19 . 1−/− pachytene spermatocytes is consistent with impaired chromosome synapsis . Furthermore , the presence of DNA double strand breaks and early recombination foci in unsynapsed regions of incompletely synapsed Tex19 . 1−/− pachytene spermatocytes indicates that the unsynapsed chromosomes arise from a failure to initiate synapsis rather than premature desynapsis . The unsynapsed chromosomes in the incompletely synapsed pachytene Tex19 . 1−/− knockout cells are presumably sufficient to trigger apoptosis at the pachytene checkpoint [31] , [32] , and would account for the increased levels of cell death seen in pachytene stage meiotic germ cells in Tex19 . 1−/− knockout testes ( Figure S3 ) . Although incompletely synapsed pachytene cells could explain the increased levels of cell death in the pachytene meiotic germ cells in Tex19 . 1−/− knockout testes , the presence of apoptotic metaphase I spermatocytes in these animals suggests that there may be an additional defect later in spermatogenesis to account for cell death at the metaphase I stage . Around half of the Tex19 . 1−/− knockout pachytene cells did not appear to have any overt defects in chromosome synapsis ( Figure 5I ) , and would therefore presumably be able to progress to metaphase I and continue through spermatogenesis . To investigate whether there might be additional defects in chromosome behaviour at later stages of meiosis in Tex19 . 1−/− knockout spermatocytes , we prepared and analysed meiotic metaphase I chromosome spreads . During metaphase I of meiosis , homologous chromosomes are held together as bivalents by chiasmata ( Figure 5J ) . 94% of the metaphase I spreads from Tex19 . 1+/− heterozygous testes contained only bivalent metaphase I chromosomes , 5% contained univalent sex chromosomes , and 1% contained univalent autosomes . However , only 34% of the metaphase I spreads from Tex19 . 1−/− knockout testes contained only bivalent metaphase I chromosomes , while 56% of the spreads contained univalent sex chromosomes , and 33% contained univalent autosomes ( Figure 5K ) . 23% of the Tex19 . 1−/− knockout metaphase I spreads feature univalent autosomes and univalent sex chromosomes . Thus Tex19 . 1−/− knockout testes contain increased numbers of univalent chromosomes at meiotic metaphase I that could potentially trigger apoptosis at the metaphase I checkpoint [35] , [36] and account for the apoptotic metaphase I cells seen in Tex19 . 1−/− knockout testes . Furthermore , the presence of univalent chromosomes in metaphase I spreads from Tex19 . 1−/− knockout testes is indicative of a defect in the formation or maintenance of chiasmata in post-pachytene spermatocytes . Meiotic defects similar to those present in the Tex19 . 1−/− knockout testes have been observed in various different mouse mutants that carry defects in genes encoding components of meiotic chromosomes , the meiotic recombination machinery , or the synaptonemal complex [6] , [7] . However , we have been unable to detect any Tex19 . 1 protein physically associated with meiotic chromosomes by immunostaining ( R . O . , data not shown ) , and our finding that Tex19 . 1 is predominantly localised to the cytoplasm rather than the nucleus suggests that Tex19 . 1 is unlikely to be a component of meiotic chromosomes or the synaptonemal complex . We therefore reasoned that the meiotic defects present in the Tex19 . 1−/− knockout testes are unlikely to be a direct effect of Tex19 . 1 on meiotic chromosome structure or function but rather may be an indirect consequence of changes in meiotic gene expression . In order to detect changes in gene expression in the testis of Tex19 . 1−/− knockout mice , we performed microarray analysis using an Illumina MouseWG-6 v1 . 1 Whole Genome Gene Expression Beadchip containing 48 , 318 different probes . To exclude potential differences in transcript levels due to the loss of post-meiotic germ cells in the Tex19 . 1−/− testes we performed this analysis on testes from 16 dpp prepubertal mice during the first synchronous wave of spermatogenesis . At this stage of testis development , some germ cells are already in the pachytene stage of meiosis , but no obvious changes in cell composition were apparent between Tex19 . 1+/+ wild type and Tex19 . 1−/− knockout testes ( Figure S4 ) . RNAs from two different 16 dpp Tex19 . 1−/− knockout testes were compared with Tex19 . 1+/+ wild type or Tex19 . 1+/− heterozygous littermates , and transcripts that had consistent and greater than three fold changes in relative gene expression between the two groups of animals were identified . The Mouse Genome Database ( http://www . informatics . jax . org ) currently lists 97 mutations that are known to give rise to meiotic arrest during spermatogenesis [8] . These male meiotic arrest genes include genes that encode components of meiotic chromosomes , the meiotic recombination machinery and the synaptonemal complex such as Atm , Dmc1 , γH2AX , Mlh1 , Msh5 , Rec8 , Rad51 , Smc1β , Spo11 , Sycp1 , Sycp2 , Sycp3 , Syce2 and Tex14 . None of the male meiotic arrest genes listed in the Mouse Genome Database showed a consistent change in expression level in Tex19 . 1−/− knockout testes compared to littermate controls ( I . R . A . , data not shown ) . However , analysis of the microarray data suggested that the class II LTR-retrotransposon MMERVK10C [37] is upregulated by around four-fold in the testis RNA from each of the 16 dpp Tex19 . 1−/− knockout animals relative to their littermate controls ( I . R . A . , data not shown ) . The mouse genome contains around 16 approximately full-length copies of the MMERVK10C sequence in the genome , and a further 1200 fragments of the MMERVK10C endogenous retrovirus . Increased retrotransposon expression has been proposed to be responsible for impaired chromosome synapsis and meiotic defects during spermatogenesis in Dnmt3L , Miwi2 and Mili mutant mice [11]–[14] . As overexpression of the MMERVK10C retrotransposons could similarly be responsible for the meiotic defects seen in Tex19 . 1−/− mutant mice we sought to determine whether MMERVK10C expression is indeed upregulated in the testis in the absence of Tex19 . 1 . The levels of MMERVK10C expression in testis cDNA from two Tex19 . 1−/− knockout animals relative to their Tex19 . 1+/+ wild-type littermates were each tested by quantitative PCR ( Figure 6A ) . The Sertoli cell marker Sdmg1 [22] was used to normalise cDNAs from different animals . Although there was no significant change in the expression of the ubiquitously expressed β-actin gene , or the germ cell marker Dazl [38] , expression of the MMERVK10C endogenous retrovirus was increased by a factor of approximately four-fold in both Tex19 . 1−/− knockout animals ( Student's t-test , p<0 . 01 ) ( Figure 6A ) . Expression of LINE , SINE or IAP retrotransposons showed no significant change in the absence of Tex19 . 1 ( Figure 6A ) . To further validate the potential upregulation of MMERVK10C transcripts in the Tex19 . 1−/− knockout mice we performed Northern blots on testis RNA from the same two 16 dpp Tex19 . 1−/− knockout animals and their Tex19 . 1+/+ wild type littermates . Using a probe derived from the env gene of the MMERVK10C endogenous retrovirus we were able to detect a predominant 3 . 2 kb MMERVK10C env transcript in mouse testes , and some weaker MMERVK10C env transcripts at around 4 . 5 kb and 7 . 5 kb ( Figure 6B ) . The Northern blot profile for MMERVK10C env transcripts is comparable to that of env-containing transcripts from HERV-K endogenous retroviruses in human teratocarcinoma cell lines [39] . Northern blotting confirmed that the predominant 3 . 2 kb MMERVK10C env transcript is consistently more abundant in testes from Tex19 . 1−/− knockout animals than in testes from their wild-type littermates at 16 dpp ( Figure 6B ) . In order to determine which cell types are accumulating MMERVK10C transcripts in the Tex19 . 1−/− knockout testes we performed in situ hybridisation on testis sections using a MMERVK10C env probe ( Figure 6C–N ) . In Tex19 . 1+/+ wild-type and Tex19 . 1+/− heterozygous testes at 16 dpp , low levels of MMERVK10C env transcripts were present in some meiotic spermatocytes ( Figure 6C , G ) . However , MMERVK10C transcripts were generally more abundant in Tex19 . 1−/− knockout testes than in testes from Tex19 . 1+/+ wild-type or Tex19 . 1+/− heterozygous littermates at 16 dpp ( Figure 6D , H ) . The increased levels of MMERVK10C env transcript in 16 dpp Tex19 . 1−/− testes appeared to be largely due to the presence of strongly expressing cells located towards the centre of the tubules where meiotic spermatocytes are present ( Figure 6K , L ) . Similarly in adult animals MMERVK10C env transcripts were upregulated in meiotic germ cells in the testes from adult Tex19 . 1−/− knockout animals relative to their heterozygous littermates ( Figure 6E , F , I , J ) . A total of nine different Tex19 . 1−/− knockout animals at various ages were assayed for MMERVK10C expression in the testes by in situ hybridisation , and MMERVK10C expression in Tex19 . 1−/− knockout testes was consistently higher that in Tex19 . 1+/+ or Tex19 . 1+/− littermate controls . No in situ hybridisation signals were detected on testis sections using a sense MMERVK10C control probe ( Figure 6M , N ) . Taken together , the quantitative PCR , Northern blotting and in situ hybridisation data all suggest that transcripts from the MMERVK10C endogenous retrovirus are upregulated in the meiotic spermatocytes of Tex19 . 1−/− knockout testes . The upregulation of retrotransposons in Dnmt3L , Mili and Miwi2 mutant mice is associated with defects in de novo DNA methylation of IAP and LINE elements in the male germline , which presumably allows increased transcription of these elements during spermatogenesis [11]–[14] . In order to investigate whether the upregulation of MMERVK10C retrotransposons in Tex19 . 1−/− knockout testis was caused by a similar mechanism , we investigated the DNA methylation status of CpG dinucleotides in MMERVK10C elements by bisulphite sequencing MMERVK10C elements from 16 dpp prepubertal Tex19 . 1−/− knockout testes . The MMERVK10C element includes a weak CpG island overlapping the LTR and 5′untranslated region ( Figure S5 ) . As promoters with weak CpG islands are good candidates for regulation by DNA methylation [40] , we examined DNA methylation at CpG dinucleotides within this region . Sequence analysis of 30 independent clones from each of Tex19 . 1+/+ wild-type , Tex19 . 1+/− heterozygous and Tex19 . 1−/− homozygous 16 dpp testes showed that CpG dinucleotides in this region of the MMERVK10C element are predominantly methylated in the testis at this age ( Figure S5 ) . The MMERVK10C element was also methylated to a similarly high level in liver taken from the same animals as a somatic tissue control ( Figure S5 ) . The prepubertal testis is composed of approximately equal numbers of germ cells and somatic cells at 16 dpp [29] , [30] , therefore around half the clones analysed by bisulphite sequencing are likely to be derived from testicular germ cells and around half from testicular somatic cells . As all of the 16 dpp testis clones represented highly methylated DNA sequences ( Figure S5 ) , the MMERVK10C element appears to be highly methylated in both the germ cell and somatic cell compartments of Tex19 . 1+/+ wild-type , Tex19 . 1+/− heterozygous and Tex19 . 1−/− homozygous 16 dpp testes . Although we have been unable to find any evidence that the methylation status of MMERVK10C elements in the testis changes in the absence of Tex19 . 1 ( Figure S5 ) , we cannot exclude the possibility that the absence of Tex19 . 1 causes reduced DNA methylation in a subset of MMERVK10C elements in the genome , or in a subset of germ cells in 16 dpp testes . If only a subset of germ cells have altered DNA methylation at MMERVK10C elements in Tex19 . 1−/− mutant testes then we estimate that this subset would need to represent less than 25% of the germ cell population to be below our detection limit in this assay ( χ2-test , p<0 . 05 ) . Nevertheless , our observations that loss of Tex19 . 1 causes the upregulation of MMERVK10C retrotransposon elements in the testis , but not IAP or LINE elements , combined with the absence of a detectable change in DNA methylation levels in MMERVK10C elements in Tex19 . 1−/− knockout testes , suggests that Tex19 . 1-mediated repression of retrotransposons may involve a mechanism that is distinct from Dnmt3L/Miwi2/Mili-mediated repression of retrotransposons . Thus we conclude that Tex19 . 1 is part of a novel genetic pathway that represses retrotransposons in the male germline . This study describes the functional consequences of deleting the pluripotency-associated Tex19 . 1 gene in mice . Our data shows that loss of Tex19 . 1 causes impaired spermatogenesis and defects in chromosome synapsis during meiosis . Mutations in genes that are involved in various aspects of meiotic chromosome behaviour such as the initiation of recombination between homologous chromosomes , or the assembly of the synaptonemal complex , all typically cause defective chromosome synapsis during meiosis , and apoptosis in the male germline [6] , [7] . However , although there is some similarity between these phenotypes and the Tex19 . 1 mutant phenotype , we have been unable to detect any localisation of Tex19 . 1 to meiotic chromosomes by immunostaining testis sections or testis chromosome spreads ( R . O . , data not shown ) . Indeed our data suggest that Tex19 . 1 is a predominantly cytoplasmic protein and is therefore unlikely to play a direct role in meiotic chromosome behaviour . Thus , although Tex19 . 1 mutant mice exhibit defects in chromosome pairing during meiosis , we do not believe that Tex19 . 1 is a component of meiotic chromosomes and favour the interpretation Tex19 . 1 is influencing meiotic chromosome behaviour indirectly . Our finding that Tex19 . 1 is a predominantly cytoplasmic protein in germ cells and embryonic stem cells contradicts a previous study suggesting that Tex19 . 1 is a nuclear protein in embryonic stem cells [17] . The reason for the discrepancy between these studies is not yet clear . Kuntz et al . [17] raised monoclonal antibodies to Tex19 . 1 and observed nuclear staining with those antibodies in embryonic stem cells and pre-implantation embryos . The Tex19 . 1 peptide used by Kuntz et al . [17] to raise the monoclonal anti-Tex19 . 1 antibody is located C-terminally to the peptide that we have used to raise the anti-Tex19 . 1 antibodies in this study . Both peptides , and indeed the entire Tex19 . 1 open reading frame , lie within a single exon . In our study we have shown that Tex19 . 1 is predominantly cytoplasmic in embryonic stem cells by immunostaining and by Western blotting of subcellular fractions . We have also shown that Tex19 . 1 has a predominantly cytoplasmic localisation in germ cells by immunostaining germ cells isolated from embryonic testes , by immunohistochemistry on wax sections of adult testis and by Western blotting of subcellular fractions from prepubertal testes . Furthermore we have demonstrated the specificity of our antibody in the assays that we use by immunostaining and Western blotting on material from Tex19 . 1−/− knockout animals . As the cytoplasmic anti-Tex19 . 1 staining patterns that we present in this paper are lost in Tex19 . 1−/− knockout animals , at least some of the Tex19 . 1 protein that is present in germ cells and embryonic stem cells is cytoplasmic . However we cannot exclude the possibility that the two different antibodies raised in these two studies recognise mutually exclusive isoforms of Tex19 . 1 that have different subcellular localisations . Alternatively , the discrepancy between our study and the study by Kuntz et al . [17] could be caused by procedural differences , or by cross-reaction of anti-Tex19 . 1 antibodies with an unrelated antigen . Tex19 . 1−/− null male mice showed some phenotypic variation between individuals ranging from completely agametic testes to fertility . This phenotypic variability may be partly due to the genetic heterogeneity in the outbred component of the genetic background used for this study . However , as some germ cells are more severely affected by the loss of Tex19 . 1 than other germ cells in the same animal , there is also some phenotypic variability in the absence of genetic variation . Furthermore , our finding that loss of Tex19 . 1 can impair spermatogenesis even in this heterogeneous genetic background suggests that mutations in the single human homologue , TEX19 , could contribute to fertility problems in human populations . The human TEX19 gene contains two premature stop codons in the open reading frame that truncates the Tex19 protein from 351 residues in mouse to 164 residues in human [17] . The first premature stop codon in the human TEX19 gene is conserved in other primates suggesting that the C-terminal region of Tex19 is dispensable for function in primates [17] . The significance of this major difference in structure between human and mouse is at present unclear given our current level of understanding of the mechanisms underlying the phenotype in mouse . The Tex19 genomic locus has undergone a duplication event in rodents to generate two closely related divergently transcribed genes [17] . The mutation that we have engineered removes the entire Tex19 . 1 open reading frame , but leaves Tex19 . 2 intact . Therefore Tex19 . 2 could potentially provide some functional redundancy with Tex19 . 1 . Although Tex19 . 1 and Tex19 . 2 are reported to be expressed in testicular germ cells and testicular somatic cells respectively [17] , there appears to be a moderate upregulation of Tex19 . 2 in Tex19 . 1−/− knockout testes as judged by quantitative RT-PCR ( I . R . A . , data not shown ) . It is not clear at present whether this upregulation of Tex19 . 2 occurs in the germ cells or somatic cells of the testis , but any upregulation of Tex19 . 2 that is occurring does not seem to be able to fully compensate for loss of Tex19 . 1 . Nevertheless , deletion of the entire Tex19 locus may be required to rule out the possibility of some functional redundancy between these genes and may reveal additional functions for Tex19 . 1 in the germline . This study demonstrates that Tex19 . 1 has a function in progression through meiosis in the male germline . Characterisation of the meiotic defect in Tex19 . 1−/− knockout spermatocytes indicates that homologous recombination is being initiated in the Tex19 . 1−/− knockout spermatocytes but that , for some chromosomes , synapsis does not occur . As homologous recombination and chromosome synapsis progress interdependently during meiosis , it is possible that the chromosome synapsis defect that we describe in Tex19 . 1−/− knockout spermatocytes is a secondary consequence of a defect in the progression of homologous recombination , or a secondary consequence of defects in the pairing between homologous chromosomes that normally precedes chromosome synapsis [7] . Further work is needed to dissect the molecular basis of the Tex19 . 1 chromosome synapsis defect in more detail , and to understand if and how the upregulation of MMERVK10C retrotransposons that we detect in Tex19 . 1−/− spermatocytes causes these defects in meiotic chromosome synapsis . The Tex19 . 1 mutant phenotype bears some resemblance to the Dnmt3L , Miwi2 and Mili mutant phenotypes in that they all exhibit defects in meiotic chromosome synapsis and increased expression of retrotransposons in the germline [11]–[14] . However it is not yet clear whether there is a direct causal relationship between these two events . The increase in retrotransposon expression does not appear to be caused by defects in meiotic chromosome synapsis [11] , [13] , but it is not clear whether or how the increase in retrotransposon expression causes the defects in meiotic chromosome synapsis in any of these mutant mice . Increased transposition of mobile genetic elements could introduce quantitative , qualitative , or temporal changes in the DNA double strand breaks normally present during early meiotic prophase that could interfere with the homologous recombination events that normally precede and initiate chromosome pairing . Support for this model comes from the observation that mutating genes involved in piRNA function in flies activates the DNA damage signalling pathway [41] , [42] . Alternatively , it is possible that repression of retrotransposons is important for the fidelity of homolog pairing and synapsis during meiosis , and that increased expression of these repetitive elements either interferes with homolog recognition and synapsis , or promotes pairing between non-homologous chromosomes . A third possibility is that proteins encoded by the MMERVK10C endogenous retrovirus mediate the defects in meiotic chromosome synapsis by interfering with host cell proteins involved in meiotic chromosome behaviour or regulation of the meiotic cell cycle . In this regard it is important to note that transgenic mice expressing the rec protein derived from the HERVK human endogenous retrovirus exhibit defects in spermatogenesis [43] . Lastly , there may not be a direct causal relationship between retrotransposon de-repression and chromosome asynapsis . Rather the Tex19 . 1 , Dnmt3L , Miwi2 and Mili mutants may all cause defects in meiotic chromosome structure that lead to both retrotransposon de-repression and defective chromosome synapsis . Clearly further work is needed to clarify the molecular mechanism underlying the chromosome synapsis defect in the Tex19 . 1 mutant mice presented here , and in the Dnmt3L , Miwi2 and Mili mutant mice [11]–[13] . However , this study provides further evidence demonstrating a correlation between de-repression of retrotransposons and impaired chromosome synapsis during mouse meiosis . Although there are gross similarities between the Tex19 . 1 mutant phenotype and the Dnmt3L , Miwi2 or Mili mutant phenotypes , there are also important differences . Dnmt3L , Miwi2 and Mili are all required to repress LINE and IAP retrotransposons in the germline , and these three genes appear to converge on DNA methylation and transcriptional repression of these sequences in the genome [11]–[14] . However , repression of LINE and IAP retrotransposons is not perturbed in Tex19 . 1−/− knockout testes suggesting that Tex19 . 1 is not involved in the transcriptional repression of LINE or IAP elements . Rather our data shows that transcripts from the MMERVK10C class of endogenous retroviruses accumulate in the germ cells in the absence of Tex19 . 1 . These differences between the Tex19 . 1 mutant phenotype and the Dnmt3L , Miwi2 and Mili mutant phenotypes may reflect the existence of multiple mechanisms with different specificities to repress retrotransposons in the germline . The Tex19 . 1 mutant phenotype is characterised by the accumulation of MMERVK10C retrotransposon transcripts , but the molecular basis for this phenotype is not yet clear . The upregulation of MMERVK10C transcripts could be caused by changes acting at any level of gene expression from the initiation of transcription to mRNA turnover . We have not been able to find any difference in the level of DNA methylation at MMERVK10C elements in Tex19 . 1 mutant testes . This provides further evidence that Tex19 . 1 belongs to a different genetic pathway than Miwi2 , Mili and Dnmt3L for repression of retrotransposons in the germline . However , we cannot exclude the possibility that DNA methylation may be altered in a subset of MMERVK10C elements in a subset of germ cells in the Tex19 . 1 mutant testes , and that this subset of elements is responsible for the upregulation of MMERVK10C transcripts that we describe in the Tex19 . 1 mutant testes . An alternative model to explain the upregulation of MMERVK10C elements in Tex19 . 1 mutant testes is that Tex19 . 1 could be a transcriptional repressor of MMERVK10C elements . The nuclear localisation of Tex19 . 1 reported by Kuntz et al . [17] would be consistent with this type of mechanism operating . However , although we cannot exclude the possibility that some Tex19 . 1 acts in the nucleus in the germ cells in the adult testes , our finding that Tex19 . 1 is predominantly cytoplasmic in these cells would be more consistent with Tex19 . 1 acting to regulate gene expression at a post-transcriptional level . We are able to detect MMERVK10C transcripts in wild-type testes ( Figure 6B , G ) suggesting that some MMERVK10C transcripts must escape DNA methylation or transcriptional repression , and that post-transcriptional regulation of MMERVK10C mRNA may play a role in repressing the activity of this retrotransposon . The upregulation of MMERVK10C transcripts in Tex19 . 1 mutant testes does not appear to be the result of changes in RNA splicing as the MMERVK10C isoforms present in Tex19 . 1 mutant testes do not appear to be qualitatively different from those present in wild-type testes . However , the accumulation of MMERVK10C transcripts in Tex19 . 1 knockout testes would be consistent with Tex19 . 1 promoting degradation of MMERVK10C mRNA . Investigation into the biochemical function of Tex19 . 1 should provide a ready test of these models and generate some insight into the molecular mechanism of Tex19 . 1-dependent repression of MMERVK10C endogenous retroviruses . Repression of retrotranposons in the mammalian germline requires mechanisms to distinguish retrotransposons from endogenous genes to allow repression to be targeted to the correct loci . piRNAs , a group of small RNAs that are physically associated with the piwi class of proteins , are abundant in male germ cells and some piRNAs have sequence homology to various classes of retrotransposon [14] , [44]–[46] . The sequence homology between some piRNA molecules and retrotransposons is presumably used to target DNA methylation to retrotransposons rather than endogenous genes . Although there is good genetic evidence that the piwi class of proteins is involved in transcriptional repression of retrotransposons [12]–[14] , there is also good biochemical evidence that piwi proteins and piRNAs are physically associated with the translational machinery in male germ cells [46] , [47] , suggesting a role in translation or mRNA turnover . Thus piRNA-mediated repression of retrotransposons may be working at multiple levels of gene expression in male germ cells . It will be informative to investigate whether the Tex19 . 1 pathway for repression of retrotransposons that we describe here also utilises piRNAs to target repression to MMERVK10C elements . One of the interesting aspects of the Tex19 . 1 phenotype is that although the MMERVK10C subclass of retrotransposons is upregulated in Tex19 . 1 mutant testes , LINE , SINE and IAP retrotransposons are not . It is not clear how Tex19 . 1 determines specificity for the MMERVK10C element . Notably , IAP elements belong to the same subclass of endogenous retroviruses as MMERVK10C elements ( class II LTR retrotransposons ) but are not upregulated in Tex19 . 1 mutant testes . Sequences within the MMERVK10C promoter or transcript could be involved in targeting Tex19 . 1 activity to this element . Alternatively Tex19 . 1 may have the potential to regulate a wider range of retrotransposons than we have been able to identify here , but alternative mechanisms to repress retrotransposon expression during spermatogenesis , such as DNA methylation , may limit the phenotypic effects of losing Tex19 . 1 to a subset of its potential targets . Furthermore , as Tex19 . 1 expression is not restricted to spermatogenesis but also occurs in primordial germ cells , oocytes and pluripotent stem cells , it will be of interest to determine if Tex19 . 1 is involved in repressing MMERVK10C elements and other classes of retrotransposons in these cell types . In addition to its role in the germline , Tex19 . 1 is also expressed in pluripotent cells . Like germ cells , pluripotent cells are viable targets for retrotransposon activity as any new transposition events could be propagated through successive generations . Therefore pluripotent cells presumably also need to modulate retrotransposon activity to ensure that the mutational load on the genome is not too high . Our finding that Tex19 . 1−/− homozygotes are born at a sub-Mendelian frequency is consistent with a role for Tex19 . 1 in pluripotent cells in early embryonic development . Further work is required to determine whether the loss of Tex19 . 1−/− homozygotes during embryogenesis is caused by defects in pluripotent cells , and whether pluripotent cells upregulate retrotransposon expression in Tex19 . 1−/− knockout embryos . The ongoing battle between retrotransposons and the host genome has important consequences for evolution , and for genetic disease . Retrotransposons that can successfully evade genome defences in germ cells and pluripotent cells will be selected for during evolution , whereas germ cells and pluripotent cells are under selective pressure to keep the mutational load on the genome at sustainable levels . The striking differences in the relative abundance of different classes of retrotransposable elements between the mouse and human genomes suggest that this conflict is ongoing during mammalian evolution [9] . Although low levels of mutation and retrotransposition in the germline are required to generate the genetic variation essential for evolution , high levels of mutation or retrotransposition are deleterious to the survival of a species . In humans , endogenous retroviruses with intact coding sequences comprise a very small proportion of the genome [48] , yet intact endogenous retroviral particles are found in human pluripotent stem cells , and in testicular germ cell tumours where the expression of endogenous retroviral proteins has been suggested to contribute to tumourigenesis [39] , [43] , [49] . Furthermore , a number of human genetic diseases are associated with de novo mutagenic retrotransposition events that disrupt the function of endogenous human genes [50] , [51] . Our data suggests that Tex19 . 1 is part of a mechanism that protects the genome from the deleterious effects of retrotransposon activity in the germline , and thereby helps to maintain genomic stability through successive generations .	The germ cells—eggs in females and sperm in males—are responsible for passing genetic information from one generation to the next . As any genetic changes that arise in the germ cells can be transmitted to the next generation , germ cells are a prime target for the activity of mobile genetic elements . Mobile genetic elements make up around 40% of a mammalian genome , and many of these elements are derived from retroviruses that have infected germ cells , or early embryonic precursors to germ cells , and have integrated into the genome . Here , we characterise the function of Tex19 . 1 , a gene whose expression is restricted to germ cells and the pluripotent cells that are early embryonic precursors to germ cells . We show that when Tex19 . 1 is deleted from mice , germ cells have problems progressing through meiosis , and sperm production is impaired . Furthermore , we show that , in the absence of Tex19 . 1 , endogenous retroviruses are activated in male germ cells attempting to go through meiosis . Our results suggest that Tex19 . 1 is part of a specialised mechanism that guards against mutagenic endogenous retrovirus activity in germ cells and pluripotent cells and thus helps to maintain the integrity and stability of the genome through successive generations .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "developmental", "biology/cell", "differentiation", "developmental", "biology/germ", "cells", "genetics", "and", "genomics/gene", "function", "developmental", "biology/stem", "cells" ]	2008	Deletion of the Pluripotency-Associated Tex19.1 Gene Causes Activation of Endogenous Retroviruses and Defective Spermatogenesis in Mice
Gene duplication is an important evolutionary mechanism that can result in functional divergence in paralogs due to neo-functionalization or sub-functionalization . Consistent with functional divergence after gene duplication , recent studies have shown accelerated evolution in retained paralogs . However , little is known in general about the impact of this accelerated evolution on the molecular functions of retained paralogs . For example , do new functions typically involve changes in enzymatic activities , or changes in protein regulation ? Here we study the evolution of posttranslational regulation by examining the evolution of important regulatory sequences ( short linear motifs ) in retained duplicates created by the whole-genome duplication in budding yeast . To do so , we identified short linear motifs whose evolutionary constraint has relaxed after gene duplication with a likelihood-ratio test that can account for heterogeneity in the evolutionary process by using a non-central chi-squared null distribution . We find that short linear motifs are more likely to show changes in evolutionary constraints in retained duplicates compared to single-copy genes . We examine changes in constraints on known regulatory sequences and show that for the Rck1/Rck2 , Fkh1/Fkh2 , Ace2/Swi5 paralogs , they are associated with previously characterized differences in posttranslational regulation . Finally , we experimentally confirm our prediction that for the Ace2/Swi5 paralogs , Cbk1 regulated localization was lost along the lineage leading to SWI5 after gene duplication . Our analysis suggests that changes in posttranslational regulation mediated by short regulatory motifs systematically contribute to functional divergence after gene duplication . Gene duplication is thought to be one of the major sources of evolutionary innovation ( reviewed in [1] ) . Several molecular mechanisms of functional change have been proposed: 1 ) changes at the transcriptional level can alter the expression of the paralogous copy [2]–[5] , 2 ) changes at the enzymatic level can alter the activity or specificity of the protein [1] , [6] , 3 ) changes at the posttranslational level can modify the regulation or localization of the protein [7]–[9] , and 4 ) changes within the splicing sites can change the isoforms produced at each loci [10] , [11] . Studies on genome-wide mRNA expression patterns have established that transcriptional changes are one of the major contributors of functional differences within duplicated genes [12]–[14] . However , whether functional divergence occurs predominantly by changes in gene regulation or by changes within the amino acid coding sequence of the proteins are still unclear [15] . Coding sequences of paralogous genes show increased evolutionary rates after duplication [16] , [17] , consistent with the hypothesis that changes within the amino acid coding sequences are also important contributors to functional divergence . However , because some functional features in proteins comprise a small number of amino acids , statistical studies comparing evolutionary rates of whole proteins do not provide mechanistic explanations for changes in function [18] . For example , many proteins contain short linear motifs ( SLiMs ) such as phosphorylation sites , localization signals and interaction motifs , and these motifs are only 2-15 amino acids long [19] . For instance , the cell-cycle regulator Sic1 is a disordered protein with several phosphorylation and protein binding sites that comprise less than 20% of the protein [20] . Computational identification of short linear motifs is an important challenge , often relying on experimental data [21] , [22] . However , recently we [23] and others [24] have shown that they can be systematically identified in fast evolving disordered regions because they tend to be preferentially conserved . Nevertheless , most short linear motifs in disordered regions probably remain uncharacterized [23] . Therefore , analyses on whole proteins may underestimate the level of functional divergence after gene duplication because changes in constraints in short linear motifs may lead to regulatory changes and therefore functional divergence [8] . Recently , several studies have investigated specific types of posttranslational regulatory changes [8] , [25]–[27] ( reviewed in [28] ) , such as differences in patterns of phosphorylation between paralogs [9] or differences in localization in paralogous proteins [7] , and have shown that regulatory changes can also contribute to functional divergence . However , these regulatory changes can also be attributed in part to trans-regulatory changes ( changes in proteins that control posttranslational regulation ) . Identification of changes in the protein regulatory sequences would allow us to determine cis-regulatory divergence ( changes within duplicated proteins ) , and provide amino acid level mechanistic explanations for protein regulatory changes after duplication [29] . Formally , functional divergence in amino acid sequences after gene duplication has been divided into two types of evolution [30] . The first ( type I ) describes so-called “changes in constraint” where the rate of evolution in a site or region changes after duplication , and remains different in one of the paralogous clades . The second ( type II ) describes a burst of rapid evolution immediately after gene duplication , and then a restoration of similar levels of constraint in the two paralogous lineages . Several statistical methodologies have been developed to identify sites or regions in proteins that fall into these classes [31] , [32] . These approaches have largely focused on identifying sites in globular regions of proteins for which large numbers of homologues can be accurately aligned [33] . These approaches often use likelihood-ratio tests based on advanced probabilistic models of phylogeny and amino acid substitution to compare the rates of evolution in individual sites [34] or groups of sites [31] , [32] to the rest of the protein . For example , previous applications of these methods have identified possible positions in the globular domain of carbonic anhydrase III that are responsible for posttranslational addition of glutathione [35] . In principle , these methods could be applied to identify changes in short linear motifs within disordered regions that contribute to posttranslational regulatory change . However , because real protein evolution can be more complicated than even the most sophisticated models [36] and real protein alignments include non-biological sources of heterogeneity ( such as alignment errors and missing data ) , the likelihood-ratio test can falsely identify type I functional divergence [32] . One strategy to tackle these issues is to estimate the rejection rate of the likelihood-ratio test using empirical data , for example using permutation tests [37] . However , the distribution of the likelihood-ratio test statistic must be obtained through permutations performed for every protein and therefore may be too laborious for genome-wide studies . We set out to study the change in selective constraints in short linear motifs within disordered regions after the whole-genome duplication ( WGD ) in budding yeast by asking whether the rates of evolution of these segments significantly differed after the whole-genome duplication event . We first developed a statistical method to correct the p-value distributions of likelihood-ratio tests and show how this approach can be applied to predicted short linear motifs . We then show that the turnover of predicted motifs within retained paralogs is faster than in genes whose paralogs were lost after duplication ( which we refer to as single-copy genes or proteins ) and that , for these putative short linear motifs , correlated loss of selective constraints appear to be common , consistent with changes in function specific to one of the two paralogs . Finally , we identify examples of experimentally verified motifs present in one paralog that are unlikely to be present in the other copy , and verify our prediction of changes in subcellular localization for one of these examples ( Ace2 and Swi5 ) . Our results show that a view of molecular evolution with amino acid resolving power can allow us to propose specific hypotheses about the functional divergences between paralogs . We have previously shown that short linear motifs can be predicted based on their conservation relative to their surrounding regions [23] . We sought to detect regulatory divergence in proteins by looking for statistical signals of lineage-specific evolutionary rate changes in predicted short linear motifs in multiple sequence alignments . Likelihood-ratio tests have previously been used to detect differences in rate of evolution of full-length yeast proteins after the whole-genome duplication [16] . We sought to perform essentially the same test to identify short linear motifs whose rate of evolution changed significantly after gene duplication . To do so , we first predicted short linear motifs within proteins of species that have diverged prior to the yeast whole-genome duplication ( see Methods ) and mapped the location of the predicted short linear motifs to the genes post-duplication ( Fig . 1A ) . Using a likelihood-ratio test [38] , we tested whether two rates of evolution ( one for the post-duplication clade and one for the remainder of the phylogenetic tree ) explain the data significantly better than one single rate of evolution common to the whole tree ( see Methods ) . This test is performed once for genes that reverted to single-copy , and twice in retained duplicates ( one for each post-WGD protein ) . Previous efforts to identify changes in evolutionary rate have shown that the likelihood-ratio test statistic often deviates from the expected chi-squared null distribution even when there is truly no change in rate of evolution [37] , [39] . Indeed , when we performed simulations of molecular evolution with no changes in rate of evolution specific to the short linear motifs ( Fig . 1B–C , see Methods ) , but included realistic aspects of the evolutionary process ( such as rate heterogeneity , insertions and deletions , etc . ) , we found that the likelihood ratio test falsely identified increased rates of evolution after gene duplication ( Fig . 1D , black circles , Text S1 ) . We hypothesized that the increased rate of false rejections was because the additional evolutionary rate parameter in the alternative hypothesis ( that is supposed to capture the change in selective constraints ) can also model some of the background heterogeneity in evolutionary rate ( due to alignment errors , non-stationary and non-homogeneous evolution , etc . ) . Under assumptions that 1 ) the majority of the tests performed are truly null , and that 2 ) the deviation of the real data from the models assumed by the test is consistent over the columns of the multiple sequence alignment , the distribution of the likelihood-ratio test follows a non-central chi-squared distribution with a data-dependent non-central parameter ( see Methods ) . This non-central parameter ( the expected increase in the test statistic from ‘fitting’ some of the heterogeneous background process using the likelihood ratio test ) is the product of the Kullback-Leibler ( KL ) divergence DKL , ( the “fit” or the expected log-likelihood ratio of the alternative hypothesis over the null hypothesis given the data see Methods ) and the number of data points used to compute the likelihood-ratio test . Larger KL divergence means larger deviation of the background distribution from the null model assumed by the test . To use this in practice , we first estimate a non-central parameter using sequence data generated by a background heterogeneous evolution process and then use the non-central chi-squared distribution to obtain p-values for our test ( see Methods ) . Extensive simulations on full length proteins with non-stationary and non-homogeneous evolution , including alignment errors , showed that this approach works as expected and yields uniform p-values ( see Text S1 ) . We applied this approach to our ‘realistic’ simulation ( Fig . 1C for an example protein ) by calculating a KL divergence parameter for each protein ( see Methods ) and obtained p-values for each likelihood-ratio test ( for each short linear motif ) in that protein . This procedure reduced the false-rejection rate ( Fig . 1D , white circles ) and p-values were nearly uniform . Having confirmed that our approach to detect type I functional divergence could be applied on short linear motifs , we then analyzed our set of protein alignments . After correction for multiple testing , we identified 159 short linear motifs with significantly different rates of evolution after gene duplication at a false discovery rate of 5% ( see Methods , S1 Table ) . This corresponds to 1 . 2% of the motifs identified in single-copy genes ( 67/5825 significant motifs , Fig . 2A ) and 9 . 8% of the identified motifs in retained duplicates ( 92/942 significant motifs , Fig . 2B ) . Because motifs in retained duplicates are tested twice ( once per branch ) , changes in constraints are approximately 4 . 5 times more frequent in retained duplicates versus single-copy proteins ( 5 . 26% vs 1 . 15% of LRTs , p-value <10-20 , Fisher's exact test ) . Our previous ‘realistic’ simulation had no intended site-specific changes in constraints . Despite this , our pipeline ( including the non-central correction ) identified 0 . 059% of the motifs in simulated single-copy proteins ( 4/6753 significant motifs ) and 0 . 55% of the motifs in simulated retained paralogs ( 6/1083 significant motifs ) to have significantly different rates of evolution after false-discovery rate correction . Using these values as our estimate of false positives due to possible computational artifacts ( such as misalignments ) or due to incorrect non-central parameter estimation for the null distribution of the likelihood-ratio test statistic , we expect that 5 motifs in duplicates and 3 motifs in single-copy genes are artifacts . Therefore , although the false positive rate due to artifacts in retained duplicates is significantly higher than in single-copy genes , the increased proportion of motifs identified with changes in constraints in duplicates cannot be explained by these computational artifacts . As another negative control , we also looked at whether the flanking regions of the putative short linear motifs ( five amino acids on each side of the motifs ) showed changes in constraints after gene duplication . After correction for multiple testing , only two flanking regions were identified as having significantly different rates of evolution after gene duplication . Given that these identified changes in constraints on the flanking regions are consistent with our false positive rate , this result indicates that the type I functional divergence we identify in predicted short linear motifs is specific to the motifs and not due to some local change in constraint . Most of the motifs with changes in constraints in duplicates only occurred in one of the two copies ( 85/92 motifs retained in duplicates ) , consistent with the idea of sub-/neo-functionalization after gene duplication through posttranslational regulatory changes [8] ( Fig . 2B ) . One hypothesis as to the fate of paralogous proteins is the duplication-degeneration-complementation ( DDC ) model [2] which explains the preservation of paralogous proteins by the neutral generation of sub-functionalized copies of proteins . Under this hypothesis , one might expect that both paralogous proteins would show signs of relaxed evolution , but that specific functional regions of each protein showing relaxation in selective constraints would be complementary , such that they partition the functional regions in the ancestral protein . We sought to test whether signs of the DDC model could be detected at the posttranslational regulatory level and found 20 paralog pairs where more than one short sequence was detected as having different rate of evolution after gene duplication ( see Methods ) . Of these , seven showed reciprocal changes in constraints on their motifs , which is consistent with degeneration and complementarity at the posttranslational regulatory level as predicted by the DDC model . Despite some evidence for complementarity , the majority of paralogs ( 13/20 ) with more than a single change in constraints appeared to have a lineage bias in their posttranslational regulatory changes . We tested this using the set of 20 paralog pairs described above and asked whether the motifs were more likely to have correlated evolution than expected by chance . To do so , we randomly permutated the changes in constraints across paralogous pairs to establish the null expectation of random assortment and counted the lineage differences in changes in constraints ( see Methods ) . We ensured that the lineage bias was not caused by technical issues , such as large-scale alignment errors or bipartite motifs being predicted as two motifs by the phylo-HMM , by grouping motifs when they were within 35 amino acids of each other for this test ( see Methods ) . This analysis revealed a lineage bias in changes in constraints for regulatory sequences ( p-value = 0 . 01106 , one-tailed non-parametric permutation test , Fig . 3 ) . Therefore , proteins that change function after duplication may typically change multiple short linear motifs in concert , consistent with the idea that multiple regulatory mechanisms often work together to control protein function . For example , multisite phosphorylation from individual or multiple kinases can form intricate regulatory modules on single proteins ( reviewed in [40] ) and these clusters of phosphorylation sites have been found to be frequently conserved through evolution [41]–[43] and have been shown to turnover [44] . The increase in resolving power obtained by analysing short linear motifs allowed us to determine whether specific regions within the paralogous proteins differed in their selective constraints . We wanted to test if this amino acid level analysis could also allow us to detect signatures of functional divergence even when the rate of evolution of the whole protein after duplication did not appear to be different than the pre-WGD clade . Using similar methodologies as previous studies [16] , we found that 57% of the paralog pairs showed no evidence of significant increase in rate of evolution of the whole protein in either of the two lineages . This value is slightly higher than that obtained previously ( 44% [16] ) , which we attribute to either a different gene set or methodology , or to the non-central correction that we applied . Nevertheless , we then searched within these proteins for motifs with significant changes in constraints . Doing so , we identified 37 motifs in 28 paralogous pairs , and 46 motifs in 43 single-copy proteins . This indicates that an analysis of evolutionary rate differences using higher resolving power of functional sequences within proteins can identify additional sources of functional divergences than analyses at the whole protein level . If changes in posttranslational regulation are important for functional divergence after gene duplication , we expect the changes in constraints in short linear motifs that we detected to point to functional differences between paralogous proteins . A previous study investigated changes in localization after gene duplication by taking advantage of the systematic green fluorescently-tagged protein collection in budding yeast [7] , [45] and categorized paralog pairs as having different or similar subcellular localization . We sought to test if motifs present in paralog pairs with different subcellular localizations were more likely to turnover after gene duplication . Motifs with changes in constraints were more than twice as likely to appear in proteins with detected changes in localization ( 26/209 motifs with changes in constraints in proteins with different localization vs 12/197 in proteins with similar localization , p-value = 0 . 032 , permutation test ) , providing evidence that proteins with changes in localization are more likely to have evolved differences in short linear motifs . We were concerned that this result could be primarily driven by the fact that paralog pairs with changes in localization had significantly higher rates of evolution [7] , for example if our non-central correction was not adequate . However , we only found a modest increase in rate of evolution for the paralogs with changes in constraints on motifs and this increase was not significant ( two-tailed p-value = 0 . 093 , Mann-Whitney U test on Dn estimated previously [7] ) . Considering that we have more power to detect changes in constraints in more rapidly evolving proteins , this further suggests that our non-central correction has controlled for the overall protein rate of evolution . We next tested if the changes in constraints we predicted corresponded to interpretable differences in posttranslational regulation by analyzing experimentally characterized motifs ( same set as in [23] ) that overlapped with segments predicted to have a change in constraint in paralogous proteins . In addition , we also wanted to confirm that the differences were not specific to S . cerevisiae by looking at the presence or absence of motifs in the other species we analyzed . Of these , the paralog pair Rck1/Rck2 contained two predicted motifs that were found to have significant changes in constraints in the Rck1 protein . Interestingly , both motifs are involved in Hog1 signaling [46] , [47] . Consistent with our predictions , Rck2 is known to be regulated by Hog1 , while Rck1 is thought not to be regulated by Hog1 [47] . However , while our algorithm identified that the motif required for Hog1 binding in Rck2 was evolving more rapidly in Rck1 , it is clear that Rck1 preserved some of the critical residues required for binding to Hog1 , yet its binding activity to Hog1 has been shown to be poor [47] . This suggests that: 1 ) the protein ancestral to Rck1/Rck2 is likely to also be regulated by Hog1 , and 2 ) that Rck1 is likely to be regulated in a different manner , having lost or changed critical regulatory sequences after the duplication event ( Fig . 4A ) . Another clear example where experimentally characterized regulation of one paralog appears to have been lost in the other following gene duplication is in the Fkh2/Fkh1 paralogous pair of transcription factors . While both proteins play a role in cell-cycle progression , they are known to have non-redundant functions [48] . For example , Fkh2 , but not Fkh1 , associates with Mcm1 [49] . Another important function of the Fkh2 protein that is absent in Fkh1 is its ability to recruit the transcriptional co-activator Ndd1 . This interaction is mediated by at least two adjacent Cdk1 phosphorylation sites [50] , one of which is found to have significant changes in constraints in the Fkh1 lineage . The other phosphorylation site is not predicted by our motif prediction algorithm but is also likely to have changed constraints . We speculate that the ancestral protein to Fkh1/Fkh2 may also have bound Ndd1 in a Cdk1-dependent manner , but Fkh1's regulation appears to have changed , possibly to accommodate new functional roles ( Fig . 4B ) . A third example could be found in the Ace2/Swi5 paralog pair , important cell-cycle regulated proteins known to localize differently in budding yeast [51] . These two proteins have been extensively characterized , with several major posttranslational regulatory sequences identified [52] , [53] . Two of these have significant p-values in our analysis , suggesting that changes in constraints occurred within the Swi5 lineage . One of these is the Cbk1-regulated nuclear export signal , known to give Ace2 its daughter-cell specific nuclear localization [52] , and another is a putative Cbk1-binding motif [23] ( Fig . 4C ) . In Ace2 , Cbk1 phosphorylation prevents nuclear export and Cbk1 is only active in daughter cells [52] . Therefore , we hypothesize that the ancestral protein to the Ace2/Swi5 paralog pair was also regulated by Cbk1 to provide daughter-cell specific nuclear localization , but that loss of these important signals allowed Swi5 to localize to both mother and daughter cells' nuclei . To confirm our sequence-based predictions about evolutionary divergence , we focused on the Swi5/Ace2 paralog pair . It has previously been shown that these motifs in the extant S . cerevisiae proteins control the differential localization pattern of the paralogs [52] . Because the ancestral protein likely contained critical regulatory motifs , we hypothesized that it was also regulated by Cbk1 , and localized asymmetrically in the daughter cell ( Fig . 4C ) . We therefore wanted to assess whether the localization before and after the gene duplication was consistent with our sequence analysis . To test this , we cloned and replaced the S . cerevisiae endogenous SWI5 gene with GFP-tagged Swi5/Ace2 homologs from multiple species that diverged before and after the whole-genome duplication and quantitatively assayed their localization pattern using fluorescence microscopy ( Fig . 5A , see Methods and Text S1 ) . Upon visual inspection , consistent with our predictions , both single-copy genes localized in an Ace2-like pattern with clear daughter specific localization ( Fig . 5B ) . To quantitatively compare the localization asymmetry of the retained duplicates and the single-copy proteins , we manually quantified the nuclear fluorescence ( see Methods ) and computed the difference between fluorescence intensity in bud and mother cells , and used this as measure of asymmetry . While we could not reject the null hypothesis of symmetry in bud and mother cell localization for Swi5 , the single-copy proteins and Ace2 showed statistically significant asymmetry , consistent with our visual inspections ( Fig . 5C , p-value <0 . 05 ) . The most parsimonious explanation for these results is that the ancestral protein also showed asymmetrical nuclear localization . Interestingly , we noted that the quantitative measure of asymmetry for the single-copy proteins was not as extreme as the post-duplicate Ace2 ( Fig . 5C ) . We also observed several cells with clear mother cell GFP localization just as observed for Swi5 ( e . g . , Fig . 5B , L . wal panel , top cell , blue arrow ) . This suggests that the single-copy genes may actually represent a mixture of the Ace2 and Swi5 localization patterns , and may be more consistent with sub-functionalization of the ancestral function , as opposed to the simple lineage specific losses predicted based on sequence analysis alone ( see Discussion ) . To confirm our prediction that the changes in regulation were not specific to the S . cerevisiae lineage and occurred during the period of rapid diversification immediately following the whole-genome duplication , we also examined the corresponding genes from C . glabrata ( a budding yeast species that diverged from S . cerevisiae after the whole genome duplication ) and found similar patterns of localization to S . cerevisiae . This supports our prediction that the change in localization in the two paralogs most likely occurred shortly after the gene duplication event ( Fig . 5B , C ) and rules out the possibility that the changes we observe are simply due to a problem with expressing foreign proteins in S . cerevisiae . Although we cannot rule out more complicated artifacts due to the expression of heterologous proteins , because we observe consistent localization in two proteins that diverged before and two proteins that diverged after the gene duplication , we consider such artifacts unlikely . Although our results only provide indirect evidence for the role of the motifs in the localization of the heterologous proteins we tested , we believe that , along with the experimental evidence for the mutations on the motifs that was performed previously by [52] , that these experiments support our prediction that the asymmetric localization pattern of Ace2 was present in the single-copy ancestral protein , and this asymmetry was lost after the gene duplication in Swi5 due to losses of specific posttranslational regulatory sequences . In this study , we have analyzed the evolution of short linear motifs in protein disordered regions after gene duplication and found that regulatory change is likely to contribute to functional divergence in paralogous genes . An important outstanding question in this analysis is whether the functional changes we identify are adaptive . Previous studies have shown adaptation due to specific changes in posttranslational regulation [54] , however general molecular mechanisms for these adaptive posttranslational regulatory changes are still under study . The resolution of adaptive conflicts has been suggested as a model for adaptation of paralogous copies of multifunctional genes after duplication [4] and differential patterns of posttranslational regulation could be an example of resolved ‘multifunctionality’ . For example , in our analysis of the Ace2 and Swi5 paralogous pair , we observed that the asymmetry of the single-copy proteins was reduced when compared to the post-duplicate Ace2 ( Fig . 5C ) . Although we cannot rule out that these single-copy proteins have other mechanisms within these species that confer daughter specific localization ( as we use a heterologous system to test for their localization ) , we believe that this observation may instead be due to a Swi5-specific motif . Indeed , the characterized nuclear localization signal ( NLS ) of Swi5 [55] was not predicted in our analysis , most likely due to its proximity to the DNA-binding domain , or to the weak conservation of the residues associated with the NLS in the Candida species . This NLS of 20 amino acids spans 50 alignment columns within our alignment , and upon close inspection appears to show that the single-copy protein contains high sequence similarity to the Swi5 NLS and that the Ace2 protein and proteins from Candida have a more dissimilar one , suggesting that they might not be functionally homologous ( S3 Figure ) . This hypothesis is consistent with the predominantly Ace2-like localization pattern of the orthologous protein in the Candida clade [56] . We speculate that this NLS is responsible for the Swi5-like pattern of localization in both Swi5 and the single-copy protein . Given that Swi5 is known to enter the nucleus slightly before Ace2 and becomes degraded before Ace2 exits the daughter-cell nucleus [51] , [57] , the observed pattern for the single-copy protein is consistent with first localizing to both mother and bud nucleus as Swi5 , and subsequent nuclear export from the mother cell . We propose that the differential localization pattern of the Ace2/Swi5 paralogs is a repartitioning of localization of the ancestral protein due to sub-functionalization of the short linear motifs present in the ancestral protein . In this study , we have identified several putative motifs that have changed constraints within proteins after the whole-genome duplication in budding yeasts . Our methodology to identify changes in evolutionary rate in very small motifs relies on a correction to the distribution of the likelihood-ratio test statistic to control for possible ‘protein level’ background heterogeneous evolution that can be encountered . These ‘protein level’ effects , such as changes in protein expression levels [58] and divergence due to changes in essentiality or gene function [59] , [60] , have been shown to be major issues in evaluating correlated changes in evolutionary rates between interacting proteins [61] , [62] . These effects are likely to be encountered in our set of paralogous proteins . Therefore , we ensured that the identification of divergent short linear motifs is unlikely to be caused by these “protein level” effects by correcting the null distribution of the likelihood-ratio test to take account of the whole protein's deviation to the null model assumed by the test . Other methodologies have been previously proposed to empirically obtain the distribution of the likelihood-ratio test statistic [37] . Our approach is similar; however we only estimate one parameter ( the non-central parameter ) because in our case it sufficiently describes the null distribution . Both approaches ( empirically-derived null distribution and estimation of the non-central parameter ) have the caveat that they rely on having several data points ( in our case alignment columns ) that are assumed to be null distributed . An additional constraint of our approach is that it requires that the null distributed data evolves under a shared and constant background heterogeneous evolutionary process to obtain the KL divergence . Therefore , it cannot accurately produce an adequate null distribution under cases where recombination has occurred in a gene , for example . Nevertheless , this approach can be simpler and faster than the permutation tests when performed on genome-wide data where we expect a small proportion of tests to reveal functional divergence . We believe that the non-central chi-squared null-distribution can be applied to other important tests in molecular evolution where genome-scale data are available and where the assumptions of the chi-squared distribution of the likelihood-ratio test statistic are violated; however this is still under study . Our study on short linear motifs reveals that posttranslational regulatory evolution is widespread after gene duplication . However , an important limitation of our study is that it cannot identify novel regulatory sequences that have appeared along any lineage or that occur within structured regions , in part due to the way motifs are predicted . Additional genomic sequences such as population data or from additional post-WGD species may allow further analyses of functional changes in the budding yeast after gene-duplication . These types of analyses are likely to uncover even more functional variations between paralogous proteins than were suggested by protein-wide and motif-wide analyses . Nevertheless , our results are consistent with several results suggested by other studies [8] , [9]: posttranslational regulatory change may underlie an important number of observed functional differences between paralogous proteins . This appears analogous to the models of functional divergence after gene duplication due to transcriptional regulatory change [2] . These parallels between transcriptional and post-translational regulatory evolution [29] suggest that transcription factor binding sites in non-coding DNA are analogous to SLiMs in proteins . In the former , the rapid transcriptional regulatory evolution is facilitated by the rapid evolution and lack of strong constraints on non-coding DNA . In the case of post-translational regulatory evolution , because SLiMs are typically found in protein disordered regions which evolve rapidly due to lack of structural constraints , changes in motifs in disordered regions may be a general means to facilitate functional evolution [63] . We based the orthology assignment on the data from the Fungal Orthogroups Repository [64] because it contained both sequences from Candida species and budding yeasts . Protein sequences and orthology assignment from six Candida yeast species [Candida tropicalis , Candida albicans , Candida parapsilosis , Candida lusitaniae , Candida guilliermondii , Debaryomyces hansenii] were obtained from the Fungal Orthogroups Repository . When several protein sequences from the Fungal Orthogroups Repository were mapped to a single budding yeast orthology group , only the most similar sequence as assessed by blast scores was chosen . The six Candida genes , along with the Saccharomyces cerevisiae gene , were supplemented with protein sequences and orthology assignment from 19 additional related budding yeast species [Saccharomyces mikatae , Saccharomyces bayanus var . uvarum , Saccharomyces kudriavzevii , Candida glabrata , Kazachstania Africana , Kazachstania naganishii , Naumovozyma castellii , Naumovozyma dairenensis , Tetrapisispora blattae , Tetrapisispora phaffii , Vanderwaltozyma polyspora , Zygosaccharomyces rouxii , Torulaspora delbrueckii , Kluyveroymces lactis , Eremothecium gossypii , Eremothecium cymbalariae , Lachancea kluyveri , Lachancea thermotolerans , Lachancea waltii] that were obtained from the Yeast Gene Order Browser [65] . By basing our orthology assignment on the species that have not undergone a whole-genome duplication , our single-copy genes do not include singletons ( newly arisen genes after the whole-genome duplication ) , and our set of retained duplicates do not include small-scale duplicates ( duplications that arose after the whole-genome duplication ) . In total , 452 alignments of retained duplicates and 3566 alignments of single-copy proteins were used in our analysis . Protein sequences were then aligned using MAFFT v6 . 864b with the —auto flag at default settings [66] . We sought to predict small functional regions that could be labeled as short linear motifs . Because we were interested in functional segments that could be identified before the whole-genome duplication [67] , we first removed from the multiple sequence alignment the sets of proteins from species that had undergone the whole-genome duplication and predicted short linear motifs within the remaining species ( which we refer to as the ‘pre-WGD clade’ ) . To identify short linear motifs , we used a phylogenetic hidden Markov model ( phylo-HMM ) [23] . Briefly , this method identifies highly conserved short amino acid sequences within disordered regions of proteins . The unstructured regions are predicted by DISOPRED2 [68] , filtered for coiled coils using pFilt [69] and for repetitive regions using the SEG algorithm [70] . We also use the phylo-HMM to filter out large conserved regions as we consider them likely to be structural regions . In a previous study , the phylo-HMM approach identified 104 of 352 known motifs with a false positive rate of 1 in 9000 amino acids [23] . In addition to the heuristics used in [23] , we now also assume that a scaling factor of rates of evolution within the conserved state is sampled from a discretized Gamma distribution with eight categories [71] with a fixed alpha and beta parameter of 0 . 6 , which was chosen as a heuristic that allowed predictions of large conserved regions ( >35aa ) interspersed by a few fast evolving columns . We now obtain the rates of evolution through a Newton-Raphson procedure , and used a window size of 31 alignment columns for the calculation of the background rate . Because the phylo-HMM tends to classify single insertion/deletion events as slow evolving regions , motifs are trimmed on either end to remove regions that are over 50% gaps or are filtered out if the prediction itself contains over 50% gaps . Flanking regions of the predicted conserved segments consisted of five alignment columns on each side . We sought to systematically identify short linear motifs that evolve at a different rate after the whole-genome duplication . To do so , each predicted motif from the pre-WGD clade was mapped back into the complete alignment . Each predicted motif was then analyzed using the PAML package [72] by a likelihood-ratio test that compares the null hypothesis ( H0 ) that motifs before and after the whole-genome duplication are evolving at the same rate , to a model ( H1 ) with two distinct rates [38] ( PAML program: AAML , clock = 2 , cleandata = 0 , fix_omega = 0 , ncatG = 8 ) . Likelihood-ratio tests have been previously used to study the evolution of the yeast paralogs generated in the WGD [16] . Our test differs from this previous application of the likelihood-ratio test , because we compared the evolutionary rate on each paralogous clade ( post-WGD_1 and post-WGD_2 ) to the evolutionary rate on the lineages that diverged before the whole-genome duplication ( pre-WGD ) one at a time . Formally , the likelihood-ratio test ( LRT ) is:where the data corresponds to the motif segment within the multiple sequence alignment , and αclade represents the rate for corresponding clades . In this model [38] , α is a scaling factor by which the estimated branch lengths are multiplied , and one of the rates always defaults to 1 . Therefore , under the null hypothesis H0 , the single rate is equal to 1 , while the alternative hypothesis H1 allows one of the two rates to be different than 1 and it is estimated by maximum likelihood . Because these models are nested , under the null hypothesis H0 , the distribution of the likelihood-ratio test statistic ( LRT ) follows the chi-squared distribution with degrees of freedom equal to 1 [73] ( see the next Methods section for the correction to the chi-squared distribution performed when assumptions of the test are violated ) . Although it is in principle possible using this test to find short linear motifs that evolve either slower or faster than the proteins in which they are found , because short linear motifs are predicted on the basis of their conservation in the pre-WGD clade , we only expect to identify motifs with faster rates of evolution after the whole-genome duplication . We estimated the false discovery rate using a slight modification of the procedure described in [74] to obtain a threshold for significant p-values . We modified this approach because when applying the LRT described above to our alignments of the yeast proteome , we observed a large number of tests resulting in LRTs of exactly zero ( thus having a p-value of 1 , e . g . Fig . 1D ) , many of which correspond to motifs where no information can be inferred about their rate of evolution . For example , in our real data , for 284/498 of these LRTs of exactly zero , we observed no amino acid differences in the multiple alignments and therefore have no power to estimate a change in evolutionary rate . Because we observed that p-values between 0 . 6 and 0 . 95 appeared uniform as expected for the distribution of truly null p-values , we used this range only to estimate the false discovery rate ( FDR ) . We counted 1836 p-values between 0 . 6 and 0 . 95 out of a total of 7709 tests . If we assume that all these p-values correspond to truly null hypotheses , then we can estimate the proportion of null tests ( π0 ) by 1836/ ( 7709* ( 0 . 95-0 . 6 ) ) = 0 . 6804 . The FDR at p-value threshold t is therefore estimated as: We considered p-values as significant where this FDR is lower than 0 . 05 . Increased evolutionary rate after gene duplication is frequently observed in entire proteins [17] . We reasoned that short linear motifs within these proteins may also show the same changes in protein-level selective constraints . Furthermore , because mutations may not be homogeneous over the phylogeny ( e . g . , due to lineage specific changes in GC content ) , proteins might show biases in their substitution process that are not accounted for by the models assumed in the LRT . Because we were interested in short linear motif evolution , we wished to test for additional changes in motifs using the heterogeneity of protein evolution as the “background” . In this case , we can still compute the LRT statistic , but the test statistic no longer follows the standard chi-squared null distribution because the heterogeneity in rates and patterns of protein evolution can be ‘fit’ using the additional parameter in the alternative hypothesis . This biases the test to reject the null hypothesis and leads to detection of false positives . A permutation test has been proposed for this case [37] however , in our case , this test must be performed for each individual predicted motif , and these permutation tests may lack power for genome-wide analyses . We therefore devised another strategy by which we can approximate the distribution of the LRT statistic under a heterogeneous background process in protein evolution . We assume that evolution of each alignment column is independent and is possibly evolving under a heterogeneous background process after the whole genome duplication event . This heterogeneity that affects the whole protein could be due , for example , to changes in expression level , lineage-specific changes in GC content or alignment errors . The likelihood of the data generated under this scenario can be computed under the alternative hypothesis H1 where there has been a change in constraints P ( data\|H1 ) , or under the ‘null hypothesis’ where evolutionary rate has remained constant , Q ( data\|H0 ) . We note that H1 can capture only some of the true heterogeneity in the data using the additional rate parameter , and the null model H0 captures even less . If θ is a parameter space and β the possible values of those parameters , then there may exist sets of values β* in the parameter space of the alternative hypothesis θH1 that captures some of this heterogeneity and that cannot be captured by the values β0 in the parameter space of the null hypothesis ( θH0 ) . Although this heterogeneous background process does not produce data following a generative process with parameters and values β* , we only seek the extra ‘fit’ obtained from the parameter space θH1 that cannot be captured by the parameter space θH0 . This fit can be summarized by the expectation of the log-likelihood-ratio of the two models , where the expectation is taken using the probabilities P , which is the Kullback-Leibler ( KL ) divergence DKL ( P\|\|Q ) . This measures the additional amount of deviation of the possibly heterogeneous background captured by the alternative hypothesis relative to the null hypothesis . In practice , we cannot necessarily parameterize the heterogeneity in the background evolutionary process , for example if it is due to alignment errors ( i . e . it is difficult to estimate β* or how data is generated from this heterogeneous process ) . Nevertheless , the distribution of the likelihood-ratio test statistic ( LRT ) when we test the alternative hypothesis H1 vs H0 ( by maximizing the ‘fit’ ) , is related to the KL divergence as follows . Given that the data used to compute the LRT are truly drawn from P , the distribution of the likelihood-ratio test statistic converges to a data-dependent non-central chi-squared distribution , χ2 ( k , λ ) , parametrized by the “non-centrality parameter” λ and the degrees of freedom k . The non-centrality parameter is given by λ = 2 L DKL ( P\|\|Q ) , where L is the number of data points used in the LRT [75] . To estimate DKL ( P\|\|Q ) , we note that the mean of the LRT when data is drawn from P must be equal to the mean of the non-central chi-squared , which is given by k+λ . Therefore , where Xi is the data at an alignment column i , k is 1 in our case and L in our case is the number of alignment columns . Under the assumption of independence between alignment columns , DKL can be estimated from the whole alignment using a single likelihood-ratio test , which we believe is reliable since L is the number of alignment columns in the whole protein and is typically large , and we assume that the background process operates uniformly over the alignment columns . Therefore , we let and use: We note that because the motif is small in comparison to the whole protein ( which we use to estimate P ) , its contribution to the calculation of DKL is small and unlikely to affect the results . While the expectation of the likelihood-ratio test statistic ( E[LRT] ) is always greater or equal to the degrees of freedom k , the obtained likelihood-ratio test statistic for a single protein LRTprotein may be smaller than k , especially when DKL is small . In these cases , we assume that DKL is equal to the parameter estimated for proteome-wide ( species ) evolution ( see below ) . We note that P = Q implies β* = β0 , which indicates that the data has no source of background heterogeneity that is better captured by the alternative hypothesis than by the null hypothesis . In that case , DKL is zero and this approach simplifies to the standard chi-squared distribution . Further , although it is possible to formulate a likelihood-ratio test with estimated β* as the values of the parameters of the null hypothesis ( akin to modeling more complex evolutionary processes in the test ) , there are several advantages of modeling the extra ‘fit’ instead . First , it is a single value , and second , it is directional ( such that rejection of the null hypothesis occurs when values of the parameters are farther from β0 than from β* ) . This estimate of the non-centrality parameter gives us a new null distribution for the LRT statistic for the predicted motifs in each protein . Since these motifs are short segments chosen from the entire alignment , we can compute the probability of having observed an LRT statistic as extreme ( or more ) in a short segment , given the length of the motif and the null distribution estimate for that protein . Therefore , the p-value for each motif , m , is given by the non-central chi-squared with 1 degree of freedom and non-centrality λm . where Lm is the length of the short linear motif . A closed-form solution exists , which we used , for the cumulative distribution of the non-central chi-squared with one degree of freedom: Where erf is the error function , LRTm is the LRT statistic computed ( by PAML ) for the motif m , and λm is as above . In more general cases ( i . e . k>1 ) , this computation can be performed using several algorithms ( see e . g . [76] ) . We also noticed that the species used in our study appeared to evolve in a manner that differed from the single rate of evolution null hypothesis ( H0 ) , even for single-copy proteins . To correct for this additional source of heterogeneity , we estimated another DKL parameter using the whole proteome to rule out any effect on the short linear motifs that could be explained simply by species-level evolution . This DKL parameter was estimated to be 0 . 014552523 . We therefore obtained two DKL parameters for each motif , and because we wanted to correct for rate differences which could be explained by genome-wide deviation or the individual protein's deviation , we chose the larger parameter while computing the p-values . This chooses the larger p-value , for which we believe no additional multiple-testing correction needs to be performed ( in that we believe we are still performing only one test per motif ) and allows us to perform a likelihood-ratio test using the standard tools for molecular clock hypothesis testing . Importantly , this global correction means our p-values are always more conservative than the significance values obtained using the standard central chi-squared distribution . To simulate more ‘realistic’ protein evolution ( Fig . 1 ) , we use a similar simulation program as in [23] . We evolve sequences to closely mirror our protein alignments by using every protein in our analysis as a template for a simulated protein . First , AAML is used on every protein alignment to obtain protein-specific branch lengths for the phylogenetic tree ( we use the species tree for all proteins ) . The root sequence is one of the sequences of the alignment ( we chose the protein sequence of median length ) , and a site-specific rate of evolution for each amino acid is inferred by the phylogenetic hidden Markov model , which we use as a scaling factor to evolve the root according to the branch lengths obtained by AAML . Indels are generated as in [23] but site specific rates are propagated to indels , such that insertions have the same rate of evolution as the amino acid positions that created it . To ensure that the sequences were as realistic as possible , we also use two amino acid substitution models: one for ordered regions , and one for disordered regions . These two models differ by their equilibrium , or stationary frequencies , of the 20 amino acids , which is estimated based on DISOPRED2 predictions on the S . cerevisiae proteome . The exchangeabilities of amino acid pairs was estimated as a whole on closely related species as in [23] . Because the rate matrix is a product of the stationary frequencies with the exchangeabilities of amino acids [77] , the substitution matrix for disordered and ordered regions will tend to create amino acids found in disordered and ordered regions , respectively . These stationary frequencies of amino acids are also used in the production of insertions . We assigned ordered or disordered regions in the root sequence , and propagated them across the phylogenetic tree . Finally , to ensure that some motifs can be predicted , we do not allow indels within regions that have been predicted as motifs in the ancestor . Our simulated proteins are then evolved according to estimated phylogenetic trees with two different substitution processes ( and therefore two different stationary frequencies of amino acids ) , and with indels . Importantly , we do not include any site specific changes in constraints . After alignment by MAFFT , the full pipeline used to predict short linear motifs and calculate the likelihood-ratio test is then used on the full set of simulated proteins . In principle , none of the motifs are intended to have lineage-specific changes in constraints . However , in practice , computational artifacts may occur during the simulation ( such as misalignments , deletions of motifs within a clade , mispredictions of short linear motifs ) and these can cause signatures of type I functional divergence . Deletions causing a motif to be removed in one of the lineage are computational artifacts of the simulation because they are unintended; however they also would represent genuine changes in constraints on the motif . However , misalignments and mispredictions of short linear motifs are actual computational artifacts that can also occur within our data . Using this set of simulated proteins , it is therefore possible to conservatively assess how many of the predicted changes in constraint can be explained by these computational artifacts or by incorrect non-central parameter estimation for the null distribution of the likelihood-ratio test statistic . We define correlated evolution to be a tendency for changes in constraints on several functional sequences to occur within only one of the two paralogous proteins . Our test for correlated evolution cumulated the number of conserved segments with changes in constraints within each of the paralogs and asked whether the changes occurred more in a particular direction than expected by chance . Under the null hypothesis , the expected difference in the number of motifs changing in one direction minus the other on one protein should be zero . The sum of all the differences is used as the final test statistic , for which a p-value was obtained by a non-parametric permutation test . To correct for the possibility that the phylo-HMM mistakenly separated a functional fragment as two motifs due to rapid evolution between the regions , we counted multiple motifs that were close to each other ( within 35aa ) and that had accelerated evolution on the same lineage as a single motif for the purpose of this test . We wished to test that the localization of Ace2/Swi5 homologous proteins differed by quantifying the intensity of the green fluorescent protein with respect to bud or mother nuclei . We chose to quantify solely the nuclear intensity as these proteins are transcription factors known to shuttle to the nucleus during the cell cycle , and show distinct patterns of nuclear localization [51] . To obtain normalized fluorescence intensity , images were analyzed by manually quantifying the cell and nuclear median green fluorescence . Cell size in pixel count was also quantified in this manner and was used to identify the daughter cells . The difference in fluorescence intensity between the bud and mother cell was used as the index of asymmetry . Cells where the median fluorescence intensity observed was over 240 were discarded as they were potentially too saturated to obtain reliable measures . Statistical significance was calculated using a Z-test . To determine statistical significance when testing for association between changes in constraints and localization differences as determined by [7] , we asked whether the observed fold increase in rate of motif changes in constraints was higher than random permutations of the ‘different’ and ‘similar’ labels of localization . The updated phylo-HMM and simulation programs can be found at http://www . moseslab . csb . utoronto . ca/phylo_HMM/data . php	How a protein is controlled is intimately linked to its function . Therefore , evolution can drive the functional divergence of proteins by tweaking their regulation , even if enzymatic capacities are preserved . Changes in posttranslational regulation ( protein phosphorylation , degradation , subcellular localization , etc . ) could therefore represent key mechanisms in functional divergence and lead to different phenotypic outcomes . Since disordered protein regions contain sites of protein modification and interaction ( known as short linear motifs ) and evolve rapidly relative to domains encoding enzymatic functions , these regions are good candidates to harbour sequence changes that underlie changes in function . In this study , we develop a statistical framework to identify changes in rate of evolution specific to protein regulatory sequences and identify hundreds of short linear motifs in disordered regions that are likely to have diverged after the whole-genome duplication in budding yeast . We show that these divergent motifs are much more frequent in paralogs than in single-copy proteins , and that they are more frequent in duplicate pairs that have functionally diverged . Our analysis suggests that changes in short linear motifs in disordered protein regions could be important molecular mechanisms of functional divergence after gene duplication .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "phylogenetic", "analysis", "evolutionary", "modeling", "biology", "and", "life", "sciences", "comparative", "genomics", "molecular", "evolution", "computational", "biology", "evolutionary", "biology", "gene", "duplication", "molecular", "biology", "techniques", "molecular", "biology", "molecular", "biology", "assays", "and", "analysis", "techniques" ]	2014	Detecting Functional Divergence after Gene Duplication through Evolutionary Changes in Posttranslational Regulatory Sequences
The London Declaration ( 2012 ) was formulated to support and focus the control and elimination of ten neglected tropical diseases ( NTDs ) , with targets for 2020 as formulated by the WHO Roadmap . Five NTDs ( lymphatic filariasis , onchocerciasis , schistosomiasis , soil-transmitted helminths and trachoma ) are to be controlled by preventive chemotherapy ( PCT ) , and four ( Chagas’ disease , human African trypanosomiasis , leprosy and visceral leishmaniasis ) by innovative and intensified disease management ( IDM ) . Guinea worm , virtually eradicated , is not considered here . We aim to estimate the global health impact of meeting these targets in terms of averted morbidity , mortality , and disability adjusted life years ( DALYs ) . The Global Burden of Disease ( GBD ) 2010 study provides prevalence and burden estimates for all nine NTDs in 1990 and 2010 , by country , age and sex , which were taken as the basis for our calculations . Estimates for other years were obtained by interpolating between 1990 ( or the start-year of large-scale control efforts ) and 2010 , and further extrapolating until 2030 , such that the 2020 targets were met . The NTD disease manifestations considered in the GBD study were analyzed as either reversible or irreversible . Health impacts were assessed by comparing the results of achieving the targets with the counterfactual , construed as the health burden had the 1990 ( or 2010 if higher ) situation continued unabated . Our calculations show that meeting the targets will lead to about 600 million averted DALYs in the period 2011–2030 , nearly equally distributed between PCT and IDM-NTDs , with the health gain amongst PCT-NTDs mostly ( 96% ) due to averted disability and amongst IDM-NTDs largely ( 95% ) from averted mortality . These health gains include about 150 million averted irreversible disease manifestations ( e . g . blindness ) and 5 million averted deaths . Control of soil-transmitted helminths accounts for one third of all averted DALYs . We conclude that the projected health impact of the London Declaration justifies the required efforts . Neglected tropical diseases ( NTDs ) are considered a special category of infectious diseases , distinct from the major killers HIV , tuberculosis and malaria , which have been the main focus of attention and funding for developing countries over the past decades . NTDs are largely confined to ( sub ) tropical resource-constrained regions , where they cause substantial morbidity , disability and even mortality , as documented by the recent Global Burden of Disease ( GBD ) estimates [1–4] , and consequently have high socioeconomic impact [5 , 6] . Most NTDs are either curable or preventable , but in practice there exist major barriers to the effective implementation of control . Fortunately , international commitment to NTD control has rapidly increased in recent years . In 2012 , the World Health Organization ( WHO ) formulated a ‘Roadmap’ towards ambitious control and elimination targets [7] . By endorsing the London Declaration on NTDs , several private and public sector organizations committed to meet those targets [8] . For five NTDs—lymphatic filariasis ( LF ) , onchocerciasis , schistosomiasis , soil-transmitted helminths ( STH ) and trachoma—the primary control strategy is preventive chemotherapy ( PCT ) . For four other NTDs—Chagas’ disease , human African trypanosomiasis ( HAT ) , leprosy and visceral leishmaniasis ( VL ) –control programs rely on case detection with innovative and intensified disease management ( IDM ) , sometimes in combination with other measures such as vector control . Guinea worm ( dracunculiasis ) is confined to just a few residual foci in Africa and close to being eradicated . For LF , trachoma , HAT and leprosy the target is elimination by 2020 , and for the others it is currently control [7 , 9] . The London Declaration was formulated to accelerate progress towards the WHO Roadmap targets by sustaining or expanding existing drug donation initiatives; providing funding to support NTD programs , strengthen drug distribution , and research and development; and enhancing collaboration and coordination on NTDs at ( inter ) national levels [8] . To further motivate and justify these efforts it is important to know their expected health gains . We therefore aim to estimate the global health impact of meeting the WHO Roadmap targets in terms of averted morbidity and mortality , expressed in years lived with disability ( YLD ) , years of life lost ( YLL ) , and disability adjusted life years ( DALYs ) . YLD reflects the number of prevalent cases of each considered disease manifestation multiplied by a disease-specific disability weight between 0 ( perfect health ) and 1 ( equivalent to death ) , whereas YLL reflects the number of deaths times a standard life expectancy at the age of death in years . The number of DALYs is the sum of both measures ( DALYs = YLD + YLL ) . Two datasets were used in our calculations . First , the GBD-2010 estimates regarding NTDs were made available to us by the Institute for Health Metrics and Evaluation ( IHME ) , Seattle , USA [3 , 10] . Second , UNPOP demographic data and projections were obtained from the website of United Nations Department of Economic and Social affairs [11] . The GBD-2010 data consist of three burden estimates: prevalent cases , years lived with disability ( YLD ) and years of life lost ( YLL ) . These estimates were available for 1990 and 2010 , per country , age group and sex . Prevalent cases were provided per disease manifestation ( sequela ) , whereas YLD and YLL were only provided as totals per NTD . Table 1 gives an overview of all 31 sequelae considered in the GBD calculations for the London Declaration NTDs . Guinea worm was not included in the GBD study and is therefore not considered here . For STH , burden estimates were available for ascariasis , hookworm disease and trichuriasis separately . Background documents justifying and describing the underlying assumptions of the GBD estimates , including disability weights , were also kindly made available to us . GBD estimates were structured according to the following age groups: 0–6 days , 7–27 days , 28–364 days , 1–4 years , 5–9 years , … , 75–79 years , and 80+ years . We combined the four youngest age groups into a 0–4 years group . For irreversible sequelae ( see below ) , the number of prevalent cases was redistributed into 1-year age groups , using a smoothing method that minimizes the squared differences between successive years , under the constraint that 5-years totals equal the available data . The demographic data were already available in 1-year age groups . The GBD estimates of the number of prevalent cases for all 31 sequelae and 5 causes of death ( HAT , VL , STH-ascariasis , Chagas’ disease and schistosomiasis ) in 1990 and 2010 were taken as the basis for our calculations . Estimates for other years were obtained by interpolating between 1990 and 2010 , and further extrapolated until 2030 , under the assumption that the 2020 WHO Roadmap targets were met and sustained beyond 2020 . Health impacts were assessed by comparing the results of achieving the targets with the counterfactual , construed as the health burden had the 1990 situation continued unabated . Prevalent cases ( both remaining and counterfactual ) were translated to YLD and YLL , and summed to arrive at DALYs . The health impact of reaching the targets was expressed as DALYs averted over the decades 2011–2020 and 2021–2030 . All calculations were carried out in duplicate in Microsoft Excel , and verified using R . All results ( totals and country-specific values ) , underlying calculations and assumptions are available as an open-access web-based dissemination tool ( https://erasmusmcmgz . shinyapps . io/dissemination/ ) . A detailed step-wise explanation of our methodology is given below . Sequelae were first categorized as either reversible or irreversible ( Table 1 ) , depending on whether treatment of the underlying infection would remove the sequelae in a relatively short time , say , within a couple of years at most . For all reversible sequelae , interventions were considered to affect their prevalence , while for irreversible sequelae this was their incidence . Linear interpolation ( at the log-scale for irreversible sequelae ) was carried out between 1990 ( or the start-year of large-scale control efforts ) and 2010 for prevalence rates ( i . e . the number of prevalent cases divided by population size ) per sequela , country , age group and sex . Absolute numbers were then calculated from these interpolated prevalence rates , using the demographic UNPOP data . For 2020 ( and beyond ) , WHO Roadmap targets were interpreted in terms of prevalence ( for reversible sequelae ) or incidence ( for irreversible sequelae ) levels , based on discussions with—mostly WHO—disease experts ( Table 2 ) . Trends in incidence and prevalence during the intervening years ( usually 2010–2020 ) were obtained through linear interpolation between the 2010 levels ( GBD data ) and the interpreted targets . We then translated the calculated trends into absolute numbers of remaining cases using UNPOP projections for the period 2011–2030 , and compared this with the counterfactual situation of no additional control efforts , to assess the impact of meeting the targets . The counterfactual was construed as the health burden that would have been expected had the 1990 epidemiological situation ( i . e . disease incidence or prevalence ) continued unabated . Whenever the 2010 prevalence of a sequela exceeded that of 1990 , we took 2010 as the counterfactual . Interpolation for irreversible sequelae , such as blindness as a result of onchocerciasis , was carried out at the level of incidence , because even after elimination of infection these sequelae will persist until the death of the last patient . The annual incidence density λ ( a , t ) at age a and calendar time t of an irreversible condition is given by the following equation ∂s ( a , t ) ∂a+∂s ( a , t ) ∂t=−λ ( a , t ) ⋅s ( a , t ) +[1−s ( a , t ) ]⋅μ ( a , t ) ⋅s ( a , t ) where s ( a , t ) denotes the susceptible fraction ( i . e . 1 –prevalence ) of the population and μ ( a , t ) the excess mortality rate among those affected . In a stable endemic situation ( i . e . without cohort effect , thus ∂s ( a , t ) /∂t = 0 ) and without excess mortality ( i . e . μ ( a , t ) = 0 ) , λ ( a , t ) can be obtained from a single cross-sectional survey by taking the differences in the logarithmic age profile of the fraction susceptible . However , because the cross-sectional age profiles of GBD 1990 and 2010 for each sequela differed , we annualized the differences ( on a logarithmic scale ) in these profiles to obtain an estimate for ∂s ( a , t ) /∂t . We further assumed the excess mortality rate to be independent of age and calendar time , and have a pre-set value μ* = 0 . 0 , 0 . 05 or 0 . 10 , dependent on the severity of the sequela ( Table 1 ) . The value of μ* was chosen after consultation of the disease experts and crudely reflected the mortality rates as used in the GBD calculations . The resulting incidences were calculated back to prevalences ( of remaining cases ) by ‘exposing’ cohorts to the derived age and time-specific incidence densities and excess mortality rates . Predicted prevalent cases for each sequela were then translated to YLD , using two matrices of multiplication factors ( one for the year 1990 and one for 2010 ) that we had derived from the GBD data as follows . Whenever an NTD had one sequela ( e . g . trachoma ) , the GBD YLDs in 1990 and 2010 were divided by the number of prevalent cases in the same year to arrive at country , age and sex-specific multiplication factors that capture disability weights , the underlying case-mix ( e . g . severe vs . mild disability , where applicable ) , and correction of burden estimates for co-morbidity , as used in the GBD 2010 study [2] . For NTDs with multiple sequelae ( e . g . onchocerciasis ) we followed the same procedure , but using a weight for each sequela based on an estimate of the average disability weight using GBD documentation ( Table 1 ) , because the YLD data provided by the GBD study did not separate the contributions of different sequelae . We treated all multiplication factors as constants . Remaining cases after 2010 were multiplied by the factors in the 2010-matrix , and for 1990–2010 an interpolation of the multiplication factors in both matrices was used . For counterfactual cases we used the multiplication factors in the 1990-matrix , or both matrices when 2010 was used as counterfactual ( i . e . similar to the approach for remaining cases ) . Regarding our mortality calculations , we first translated GBD YLLs in 1990 and 2010 to actual country , age , and sex cause-specific mortality rates , using the age and sex-specific residual life expectancies as applied in the GBD study [1] . For HAT , VL and ascariasis , where mortality is closely linked ( in time ) to infection prevalence , these rates were treated as prevalent cases ( of reversible sequelae ) as described above and back-calculated to YLLs . For Chagas’ disease and schistosomiasis , where mortality is closely linked to late sequelae , we followed a different procedure . Similar to the calculation of YLDs for NTDs with multiple sequelae , we related YLLs in 1990 and 2010 to prevalent cases of selected sequelae , using a weight representing their lethality . For schistosomiasis , mortality was related to hematemesis ( weight = 50 ) , ascites ( 1 . 0 ) and schistosomiasis infestation ( 0 . 01 ) . For Chagas’ disease , these were heart failure ( 10 ) and chronic heart disease ( 1 . 0 ) . Using the above method , some irreversible sequelae—in particular for Chagas’ disease and LF—showed for some countries values of λ ( a , t ) < 0 , due to unrealistic fast declines in the GBD prevalence estimates between 1990 and 2010 . Here , we chose alternative prevalences , but still within the confidence limit ( Cl ) provided by the GBD study , as follows . We reduced the GBD 1990 ‘Mean’ prevalence to 0 . 25 ‘Mean’ + 0 . 75 ‘Lower Cl’ , and we increased the GBD 2010 ‘Mean’ prevalence to 0 . 25 ‘Mean’ + 0 . 75 ‘Upper Cl’ . The GBD 2010 estimates for leprosy appeared to be mistakenly based on overall leprosy new case detection ( incident cases ) instead of prevalence of ( irreversible ) cases with leprosy grade 2 disability , on which the disability weights are based . We therefore performed a recalculation to arrive at grade 2 disability prevalences as follows . First , we took from the WHO-published global leprosy data for 2010 the proportion of newly detected cases with grade 2 disability , which was 6% [12] . Secondly , prevalence of leprosy cases with grade 2 disability in virtual birth cohorts was accrued at a rate determined by this incidence density , while assuming a steady-state until 1990 and a linear decreasing incidence to 2010 . We further assumed that excess mortality due to leprosy is negligible ( μ* = 0 . 0 ) . These prevalence values constituted the ‘GBD data’ on which our calculations were based . Fig 1 shows the global trends in remaining and averted DALYs , distinguished into YLD of reversible and irreversible sequelae and YLL . According to the original GBD 2010 data ( dark-colored bars ) , the health burden of onchocerciasis , STH , Chagas’ disease , HAT and VL has clearly decreased over the period 1990 to 2010 . For LF , schistosomiasis and leprosy , the absolute burden has increased , but not as fast as would be expected from the counterfactual . Thus , for these NTDs , the relative burden has decreased , when correcting for population growth . Only for trachoma ( and in some countries for schistosomiasis ) , the GBD-estimated burden has increased faster than would be expected from the demographic trends over the period 1990–2010 . Meeting the 2020 targets will lead to a substantial health-impact for all NTDs ( Fig 1 ) . It is clearly visible that reversible sequelae ( green ) are disappearing faster than irreversible sequelae ( brown ) . This makes the health impact of reaching the targets for LF , trachoma and leprosy over the first two decades somewhat less spectacular compared to that for the other NTDs , of which the burden is mainly caused by reversible sequelae or death . Another important factor determining the overall health impact is population growth and other demographic developments , as expressed by the counterfactual . NTDs that are prevalent in Asia ( LF , STH , leprosy and VL ) show a slower rise of the counterfactual compared to the NTDs mainly confined to Africa ( onchocerciasis , trachoma and HAT ) or South America ( Chagas’ disease ) . Overall , meeting the targets of London Declaration NTDs will avert about 600 million DALYs in the two decades after 2010 , nearly equally distributed between PCT and IDM-NTDs , with the former mostly ( 96% ) attributable to averted disability , whereas the latter largely ( 95% ) results from averted premature death ( Fig 2 ) . These health gains include about 150 million averted irreversible disease manifestations , in particular chronic heart disease due to Chagas’ disease , bladder pathology due to schistosomiasis , and hydrocele and lymphedema due to LF ( Table 3 ) . In addition , approximately 5 million deaths are averted , mainly from VL and HAT , and to a lesser extent Chagas’ disease ( Table 4 ) . Of the 600 million DALYs overall averted in the period 2011–2030 , in the ideal situation of meeting the WHO Roadmap targets of London Declaration NTDs , about 30 million will be realized in the year 2020 , increasing to 40 million in the year 2030 . This is of the same order of magnitude as the current annual health burden of any of the ‘big three’ infectious diseases , HIV/AIDS , TB and malaria , which accounted for about 80 , 50 and 80 million DALYs , respectively , in 2010 [3] . Clearly , for these three infections elimination is a more remote perspective than for the nine NTDs targeted by the London Declaration . Thus , the ongoing efforts to control the big three seem to justify similar investments in NTD control . In addition , it can be expected that for several of these NTDs control efforts will lead to a cessation of transmission over vast regions , after which further control can be discontinued and investments wound down adding to the value of this investment for future generations . STH accounts for one-third ( 34% ) of the averted DALYs , almost entirely due to avoided disability . This perhaps surprising finding can be easily explained by the wide-spread distribution of STH [13] . Importantly , approximately half ( 46% ) of the averted STH-burden would be realized in China . This brings to the fore the sensitivity of our results to the choice of counterfactual . That is , our assumption that the situation of 1990 would continue unabated may be questioned for several countries , including China , which have experienced unprecedented economic and social development over the past decades [14] . For example , the health impact for STH would be about halved if the situation in 2010 were used as the counterfactual , as can roughly be concluded from Fig 1 , but such a drastic correction would certainly not be reasonable for many endemic countries in Africa and Southern Asia . On the other hand , socioeconomic development may also have facilitated the spread of NTDs , in particular schistosomiasis , of which large outbreaks followed the construction of dams and irrigation schemes [15] . HAT perhaps follows more erratic patterns , reflecting e . g . civil unrest , war and also ecological circumstances [16] , so that the year 1990 may not be representative of the actual counterfactual over 1990–2020 . Trachoma and schistosomiasis showed large increases in GBD prevalence from 1990 to 2010 , which may well reflect an underestimation of the 1990 burden . Consequently , this may have led to underestimating both the counterfactual and the health impact . Another potential source of underestimation of the health impact for some NTDs may be that the largest gains are achieved in the initial years of programs , followed by a slow down towards the target year , as it becomes harder to reach the more marginalized populations . Furthermore , by using a fixed excess mortality rate μ* for irreversible sequelae ( where applicable ) we may have somewhat overestimated health impacts for these sequelae as treatment is likely to improve over time . However , since the remaining cases get older at the same time , possibly experiencing a higher mortality , we may have introduced some underestimation as well . Clearly , by using a uniform methodology we have introduced ( perhaps occasionally substantial ) under or overestimation of NTD and country-specific results , but we are confident that the overall bias in our estimated health impact of reaching the targets will be small . Almost half ( 44% ) of the overall health impact is attributable to averted deaths , in particular from visceral leishmaniasis and HAT , and to a lesser extent Chagas’ disease , followed by schistosomiasis and STH ( ascariasis ) . In our calculations , we followed the GBD accounting philosophy which assigns all DALYs ( i . e . residual life expectancy at the age of death ) resulting from a death to the year in which it occurred [1] , whereas DALYs attributable to morbidity are accrued during the years that individuals suffer [2] . Moreover , remaining life expectancies were based on the demography of Japan , according to the fundamental concept that all people are entitled to the best life expectancy in the world , irrespective of e . g . country of residence and socioeconomic status . Clearly , other methodologies might have distributed health gains differently over time . Our calculations depend strongly on the estimates made in the GBD study [1–3] . These estimates are notably uncertain for NTDs , given the paucity of data on their geographic spread and control . Most GBD 1990 and 2010 estimates for NTDs show very wide confidence intervals , often ± 50% the mean , but sometimes with an upper confidence limit up to 5 times the mean . As a consequence , our predictions ( all based on GBD point estimates ) are subject to at least a similar degree of uncertainty . Also , the GBD disability weights used are still under heavy debate , such as the relatively low value for blindness as compared to itching [17] . Furthermore , our calculations are confined to the 31 sequelae considered in the GBD study , and discussions continue about whether additional sequelae need to be considered . In particular , the choice not to include so-called subtle morbidities , such as impaired cognitive development due to STH and schistosomiasis , or poor mental health from stigma and discrimination due to the disfigurements caused by LF and leprosy , is considered an important omission by many [13 , 18–21] . Our results also depend upon the interpretation and formulation of the WHO Roadmap targets [7 , 9] , which occasionally are ambiguous . Consulting disease experts at WHO has resulted in agreement about interpretations for most NTDs , even though sometimes the targets were considered too general or utopic . In addition to the intrinsic value of averting human suffering and death , this health impact of reaching the targets will also give rise to major economic and societal improvements , such as increased productivity and avoided ( often catastrophic ) out of pocket payments for treatment and care , which can be assigned monetary values . In particular , the currently ignored subtle morbidities are likely responsible for major societal impacts . We realize that the targets are ambitious , and may for instance be jeopardized by challenges in drug distribution , disease surveillance and health care access . Also , systematic non-compliance in mass-drug administration , population groups currently not eligible for treatment , and development of drug or insecticide resistance could be serious threats , as demonstrated in a recent collection of studies by the NTD Modelling Consortium focusing on the question whether we are on track to reaching the goals [22] . Furthermore , even if the targets are reached by 2020 it is essential that control and surveillance are continued to avoid rebounding effects , certainly for those NTDs where elimination of transmission cannot be expected . In conclusion , NTDs together constitute a major health burden , comparable to any of the three major infectious diseases HIV/AIDS , TB , and malaria . Achieving internationally agreed targets of NTD control and elimination will bring about major gains in health and reductions in human suffering . Much of this will be achieved by avoiding morbidity rather than mortality as many of the parasites involved , such as soil transmitted helminths , rarely kill their hosts . This also implies that our impact assessment depends on the valuation of health states as used by GBD , a valuation that inevitably is somewhat subjective and open to debate . We did not consider the costs involved in reaching these targets , but a recent assessment demonstrated that these are relatively modest [23] , indicating that the cost-effectiveness of interventions to control NTDs will likely be high . One thing is certain however: as NTDs are disorders that disproportionately affect the poor , their control will considerably improve global equity .	Neglected tropical diseases ( NTDs ) are a group of infectious diseases that occur mostly in poor , warm countries . NTDs are caused by various bacteria and parasites , such as worms . They can either be cured or prevented through drugs and other interventions , such as control of insects that spread the infection . The London Declaration is a statement by various organizations , including the World Health Organization ( WHO ) and pharmaceutical companies that donate the necessary drugs . The declaration endorses targets for disease reductions by 2020 , as recently formulated in the WHO Roadmap , to be achieved by rigorous application of available interventions . We explore how much health can be gained if these targets are indeed achieved . We estimate that in such case 5 million deaths can be averted before 2030 and also that huge reductions in ill-health and disability can be realized . Over the period 2011–2030 , a total health gain would be accomplished of about 600 million disability adjusted life years ( DALYs ) averted . DALYs are a measure of disease burden , consisting of life years lost and years lived with disability . This enormous health gain seems to justify similar investments as for e . g . HIV or malaria control .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "death", "rates", "medicine", "and", "health", "sciences", "disabilities", "demography", "tropical", "diseases", "parasitic", "diseases", "bacterial", "diseases", "global", "health", "neglected", "tropical", "diseases", "population", "biology", "infectious", "disease", "control", "public", "and", "occupational", "health", "infectious", "diseases", "protozoan", "infections", "people", "and", "places", "population", "metrics", "helminth", "infections", "schistosomiasis", "chagas", "disease", "leprosy", "biology", "and", "life", "sciences" ]	2016	Concerted Efforts to Control or Eliminate Neglected Tropical Diseases: How Much Health Will Be Gained?
We determine knotting probabilities and typical sizes of knots in double-stranded DNA for chains of up to half a million base pairs with computer simulations of a coarse-grained bead-stick model: Single trefoil knots and composite knots which include at least one trefoil as a prime factor are shown to be common in DNA chains exceeding 250 , 000 base pairs , assuming physiologically relevant salt conditions . The analysis is motivated by the emergence of DNA nanopore sequencing technology , as knots are a potential cause of erroneous nucleotide reads in nanopore sequencing devices and may severely limit read lengths in the foreseeable future . Even though our coarse-grained model is only based on experimental knotting probabilities of short DNA strands , it reproduces the correct persistence length of DNA . This indicates that knots are not only a fine gauge for structural properties , but a promising tool for the design of polymer models . Entanglements in molecular cords like polymers or semi-flexible biopolymers like double-stranded DNA ( dsDNA ) often lead to knotted chain conformations . The significance of DNA knots has been discussed in biological contexts [1 , 2 , 3] , as well as in technological settings: Recent studies [4 , 5] investigated how knots in DNA change the translocation of long DNA molecules through nanopores . DNA translocation dynamics are of practical importance in the context of DNA nanopore sequencing , where a single molecule of either single- or double-stranded DNA is electrophoretically driven through a nano-scale pore across an impermeable thin membrane . The DNA's translocation through the nanopore alters the amplitude of the electrochemical current by perturbing the charge transport along the pore [6 , 7] . In a common approach , the chain's nucleotide sequence is read by directly analysing the time-dependence of the current signal [8] . Mathematically , knots are only well-defined in closed curves . Nevertheless , a physical definition is often applied to open strings [9 , 10]: Ends are connected in a systematic manner before the knot type is analysed ( see the Methods section ) . Knots are categorised by their minimal number of crossings in planar projection . The simplest ( non-trivial ) knot is the so-called trefoil , which has three crossings and can be obtained by closing an overhand knot . Intriguingly , some of our intuitive understanding of macroscopic knots carries over to the nano-scale . About 50 years ago , Frisch and Wasserman conjectured [11] that any molecular cord will eventually be knotted as the chain length increases . This conjecture was later proven for certain classes of lattice polygons [12] , but does not state how polymer length and polymer properties are related to knotting probability . Polymers in globular states [13–15 , 9] and DNA in viral capsids [16 , 17 , 3 , 18] are known to be highly knotted , whereas unconstrained polymers [15 , 19 , 9] or DNA in good solvent conditions are less prone to self-entanglements . The probability of knotting in dsDNA was first measured in the early 1990s by gel electrophoresis [20 , 21] . Strand sizes of up to ≈10 , 000 base pairs were considered and depending on salt conditions , the probability of observing knots was at most a few percent . Consequently , self-entanglements and knots have mostly been ignored in the context of nanopore sequencing . Reference [20] also describes for the first time a coarse-grained model for DNA which is partly based on knotting probabilities . Very recently , a ground breaking study ( Plesa et al . , Nature Nanotechnology in press ) has pushed these boundaries even further and estimated knotting probabilities for significantly larger strands in high salt concentrations by analysing translocation events in solid-state nanopores . The focus of this work is two-fold . First , we introduce a coarse-grained model of dsDNA which is solely based on experimentally determined knotting probabilities . This model is then used to analyse the statistics of DNA knotting and determine typical knot sizes , motivated by the emergence of DNA nanopore sequencing technology . This analysis naturally precedes further experimental or theoretical investigations designed to address the problem of how to avoid or control knots in DNA nanopore sequencing devices . Employing the aforementioned coarse-grained model , the abundance and size of knots in chains exceeding half a million base pairs is studied for physiologically relevant salt concentrations of 0 . 15M NaCl . Although it is well known that knots are likely to form in long polymer chains , no quantitative estimation of DNA knotting probabilities is available for chains beyond ≈50 , 000 base pairs [20 , 21 , 22] . To estimate knotting probabilities for DNA chains of up to half a million base pairs , model parameters are chosen so that predicted knotting probabilities of short DNA chains match knotting probabilities from electrophoresis experiments: DNA is modelled as a semi-flexible chain of impermeable spherical beads of diameter d≈4 . 5nm , which corresponds to ≈13 base pairs ( bp ) . An intrinsic stiffness controls the bending of the chain and the effective DNA diameter d subsumes excluded volume effects as well as screened electrostatic interactions [20] . The simplicity of the model is the key to deriving optimal model parameters and obtaining statistical estimates for chain lengths which are relevant in the context of future applications of nanopore sequencing technology . Mathematical details of the model , as well as the rationale for choosing particular values of the model parameters are discussed in the Methods section . Intriguingly , the DNA model predicts a persistence length of ≈50nm in excellent agreement with experimental findings [23] . We stress that experimentally measured knotting probabilities of short dsDNA chains are the only input to our model ( see the Methods section ) . This observation is non-trivial , as it implies that metric properties of DNA can be predicted from purely ( non-metric ) topological properties , which has never been demonstrated before . Hence , basing a simple real chain model with stiffness on knotting probabilities allows for the description of physical properties of the chain . The model is then employed to predict knotting probabilities of long dsDNA chains as well as typical sizes of DNA knots . This is to be contrasted with the approach in [20] , which in addition to knotting probabilities also requires the persistence length of DNA as an input parameter . Although the dsDNA model in [20] is coarser , and introduces a sequence of impermeable cylinders to model DNA , both approaches derive similar values for the effective diameter of DNA . In comparison , simple random walk models of DNA which lack excluded volume interactions [24] tend to overestimate the occurrence of knots . E . g . , our real chain model predicts that a chain of 150 , 000 base pairs is knotted in roughly 40% of all cases , whereas random walks of 500 segments ( assuming a Kuhn length of 300 base pairs ) exhibit knots in ≈80% of all cases if the same closure is applied . In Fig 1 , a typical trefoil ( light green ) in a DNA chain of ≈13 , 000bp ( represented by a coarse-grained model chain of N = 1 , 000 beads ) is displayed in relation to characteristic nanopore sizes . In Fig 2 computed probabilities for observing knots under physiological salt conditions are shown for DNA strands of up to ≈525 , 000bp ( N = 40 , 000 , S1 Table , supporting information ) . At this length , ≈88% of all chains already contain at least one knot . Remarkably , more than ≈68% contain complex knots with more than three crossings or composite knots . The transition from a mostly unknotted to a mostly knotted ensemble of DNA chains is indicated by the base pair count B0 at which the probability of obtaining an unknotted conformation is 1/e≈0 . 37 [25] . B0≈250 , 000bp ( N≈19 , 000 beads ) also characterizes the regime where knots with higher crossing number ( ≥4 ) become more abundant than trefoil knots . Intriguingly , Fig 3 indicates that more complex entanglements are mainly made up of composite knots which include trefoil knots as prime factors . Beyond 300 , 000bp ( N≈22 , 850 ) the triple trefoil knot and even the 31#41 composite knot occur more often than the figure-eight knot 41 , and formation of prime knots with more than four crossings is very unlikely ( S1 Fig ) . Hence , probabilities of composite knots are not mere product probabilities of the constituent prime factors , reflecting the non-local structure of emerging polymer entanglements . Note that even though the Alexander polynomials of the analysed composite knots ( Fig 3 ) are identical to the Alexander polynomials of specific prime knots with eight crossings ( e . g . 31#31 and 820 share the same polynomial ) , the influence of prime knots with eight crossings on observed knotting probabilities is expected to be negligible , since all prime knots with seven crossings already have vanishingly small probabilities . In addition to estimating the mere abundance of knotted DNA chains , typical knot sizes can be determined as well: To identify the knotted region , a chain is trimmed from both ends until the remaining part becomes unknotted ( see the Methods section ) . In Fig 4 , the size of a trefoil knot refers to the contour length of the knotted region , and its distribution is shown for various DNA lengths . Intriguingly , the most likely size of a trefoil knot is around 3 , 000bp ( N≈230 ) , independent of strand size . This observation as well as computed distributions of trefoil contour lengths are in excellent agreement with recent simulation results [22] . In [22] , typical knot sizes have been determined for a similar coarse-grained model , and for a range of model parameters which can be mapped onto dsDNA at various salt concentrations . Note that DNA models based on random walks predict smaller knots: The maximum of the size distribution is at around 7 segments , corresponding to 2100 base pairs [26] . As opposed to its most likely value , the expectation value of the knot size increases with DNA length ( Fig 4 ) . To estimate a trefoil's geometrical extent , the radius of gyration ⟨Rg2⟩ of the trefoil contour is computed , and its diameter is then taken to be 2⋅⟨Rg2⟩ . The inset in Fig 4 displays the distribution of trefoil diameters for a DNA chain of ≈13 , 000bp ( N = 1000 ) , and the most likely value is ≈200nm . The trefoil displayed in Fig 1 ( light green ) consists of ≈4 , 000 base pairs ( N≈300 ) and has a diameter ≈160nm . It may thus be regarded as a typical representative , even if longer DNA chains are considered . As our DNA model reproduces the correct persistence length of DNA by adjusting model parameters to match experimentally measured knotting probabilities of short dsDNA chains , it can be inferred that knots may be employed as a tool in polymer physics: Given sufficient experimental data on knot statistics of a particular polymer species , the parameters of any suitable polymer model may be chosen so that theoretical knotting probabilities match the experimental ones . As knot statistics are a means to quantify global topological properties of polymer chains , the resulting model is expected to thoroughly reproduce polymer entanglements . Whether such a procedure leads to a good imitation of polymer behaviour and physically consistent results depends , among other things , on the selection of a proper set of adjustable model parameters . The successful derivation of the persistence length of dsDNA indicates that at least for physiological salt concentrations , our DNA model is capable of representing basic physical properties of DNA strands . Other salt concentrations will be tested in future investigations . Simulations of this model indicate that DNA molecules beyond 250 , 000 base pairs are likely knotted and contain composite knots . Our analysis of knotting in dsDNA for physiologically relevant salt concentrations of 0 . 15M NaCl estimates lower bounds for DNA knotting in nanopore sequencing devices which adapt double-stranded DNA for sequencing: Most nanopore setups operate in high salt [27] , which increases the likelihood of knotted conformations [20] . Furthermore , while most nanopore sequencing techniques keep DNA in its single-stranded form [28] , very successful alternatives [29 , 30] approach the problem by ligating adaptors to the ends of dsDNA , which subsequently help to thread the dsDNA into the pore , and afterwards control the translocation of a single DNA strand . More specifically , it was demonstrated [31 , 29] that DNA polymerase can slow down and control the transport of dsDNA through a biological nanopore: A polymerase molecule is anchored at the entrance of the nanopore , which splits the double-stranded DNA and drives the translocation of a single DNA strand . With this measure , sequence information can be obtained more reliably , at least for DNA strands of up to ≈4 , 500 base pairs [30] . If translocation is driven by DNA polymerase , every step in sequential movement takes several milliseconds [31] . This timespan should be long enough to equilibrate the knotted region or at least a substantial part of it , whereas diffusion along the contour of the DNA can probably be neglected [32] . The momentary size of a threaded trefoil can thus be estimated from ensemble statistics to be 2⋅⟨Rg2⟩ ( see the inset in Fig 4 ) , and pulling a trefoil knot through the pore should not tighten it mechanically . However , knots with high crossing number or composite knots may behave differently , in which case 2⋅⟨Rg2⟩ may not be a proper measure of knot size . If a knot is located close to the nanopore , the ion flux may be perturbed , and the magnitude of the perturbation is probably related to its geometrical extent . Therefore , knotty problems may even occur for chain lengths which are within reach of current technology . The presence of complex and composite knots in long DNA chains might lead to a blockage of the nanopore's entrance . Though a blockage can potentially be avoided for simple knots as discussed in [4 , 5] , knots of any crossing number might significantly alter the ion transport along the pore or even the translocation dynamics of the DNA . As soon as knots become abundant in long DNA chains , interpretation of the current signal and discrimination of individual nucleotides may be prone to errors , even if DNA molecules can still be threaded through nanopores at reasonable rates , since the fingerprint of the DNA's nucleotide sequence sensitively depends on the resulting ion flux and DNA translocation dynamics . It is a very hard problem to ascertain how the time-dependent electrochemical current changes in the presence of a DNA knot . A quantitative analysis clearly goes beyond the scope of this paper . We hope that our work will stimulate further experimental and theoretical investigations of the aforementioned issues . The vision that one day , a nanopore sequencing device could read a significant portion of a chromosome from just a single DNA molecule will have to include an idea of how to avoid knots in long DNA chains [33 , 34] . We employ a discrete worm-like chain ( Kratky-Porod ) model [35] with hard sphere interactions between beads and fixed bond lengths . The bending potential is given by U/kBT=−g∑icos ( θi ) with the θi , i = 1 , … , N – 1 , being the angles between adjacent bond vectors . The computational model is fully determined in terms of the number of beads N and stiffness parameter g ≥ 0 . Knotting probabilities of Kratky-Porod chains with excluded volume interactions depend on g in a non-trivial manner [36] , whereas the knotting of ideal Kratky-Porod chains monotonously decreases with stiffness . Screened electrostatic interactions are absorbed in an effective hard sphere diameter d , which depends on the salt concentration [20] . In dsDNA the distance between adjacent base pairs is 0 . 34nm . A DNA strand of B base pairs is thus modelled as a chain of N = B · 0 . 34nm/d beads . In previous experimental studies [20 , 21] , DNA knotting probabilities have been obtained by gel electrophoresis for dsDNA molecules with a length of 5 . 6kbp , 8 . 6kbp [21] and 10kbp [20] for different salt concentrations . Even though DNA had been cyclized before the knot type was determined , knots formed when DNA was still in a linear state . Therefore , experimental knotting probabilities are more likely to reflect probabilities in linear DNA . Either way , probabilities for knots in loops and knots in open chains are very similar , as has been demonstrated for random walks in [10] and for self-avoiding chains in [15] . DNA lengths from these experiments can be converted to chain lengths of the computational model for a given d . To obtain an optimal set of parameters ( g , d ) to model dsDNA under physiologically relevant salt concentrations of 0 . 15M NaCl , knotting probabilities are computed for 16 × 16 = 256 points of an equispaced grid with boundary points N = 250 , 1 , 000 , g = 6 . 5 , 14 . Each chain is simulated with Markov chain Monte Carlo ( MCMC ) methods , applying generalized MOS ( inversion , reflection and interchange ) [37] , crank-shaft and pivot moves [38] . Typical MCMC errors are two orders of magnitude smaller than corresponding experimental errors [20 , 21] and neglected in subsequent analysis . With a non-parametric regression in R [39] ( library method loess ) , a surface is fitted to the grid of simulated knotting probabilities . The comparison of the interpolated knotting probabilities with the experimental results for a salt concentration of 0 . 15M NaCl defines a smooth ( least squares ) error function E ( g , d ) , which is minimized with the aid of the Levenberg-Marquardt algorithm ( R [39] library method nls . lm ) : The minimum ( g , d ) of the error function E is interpreted as an optimal parameter set , yielding g≈11 . 673 and d≈4 . 465 . Even though the physical diameter D of dsDNA is only about 2nm , the effective diameter d also accounts for the influence of screened Coulomb interactions in addition to excluded volume . Thus , d is in general larger than D and would approach D for high salt concentrations ( for which electrostatic interactions of dsDNA are completely screened ) [20] . Production runs ( Figs 2–4 and S1 ) employ this parameter set to predict knotting probabilities for dsDNA strands of up to ≈525 , 000 base pairs ( computed knotting probabilities and MCMC errors are documented in S1 Table , S2 Table and S3 Table ) . For the ( ideal ) Kratky-Porod chain , the functional dependence of the persistence length lp ( g , d ) is given by lp ( g , d ) = −d/ln ( coth ( g ) − 1/g ) [40] , yielding ≈49 . 85nm . As a topological knot is necessarily a closed space curve , the open DNA chain has to be closed prior to knot detection . Here , we join the ends of the chain by first extending them away from the centre of mass , and then connecting them by the legs of a triangle , so that the additionally constructed line segments do not interfere with the original chain volume [41] . For each closed curve , the Alexander polynomial is evaluated and used to identify the knot type [10] . Note that in principle , the implementation of the closure may create additional entanglements and knots . In practice , this effect only plays a minor role when calculating ensemble averages . Different closures result in almost identical knotting probabilities as was demonstrated for random walks in [10] and for self-avoiding chains in [15] . To determine the knotted region of a trefoil knot as shown in Fig 4 , a chain is first trimmed bead by bead from one end and subsequently analysed until the remaining part becomes unknotted . The same procedure is then applied starting from the other terminus . The remaining beads define the contour length of the knot , and its radius of gyration ⟨Rg2⟩ is computed to describe the knot’s physical extent as 2⋅⟨Rg2⟩ .	We develop a coarse-grained model of double-stranded DNA which is solely based on experimentally determined knotting probabilities of short DNA strands . Our analysis is motivated by the emergence of DNA nanopore sequencing technology . The main advantage of nanopore sequencing in comparison to next-generation devices is its capability to sequence rather long DNA strands in a single run , currently up to ≈10 , 000 base pairs . Unfortunately , long DNA strands easily self-entangle into knotted conformations , and sequencing knotted DNA with nanopores may be subject to error . In our manuscript , the typical extent and likelihood of DNA knots is computed for DNA chains of up to half a million base pairs , and we estimate the abundance of complex and composite knots in relation to DNA length . Our analysis indicates that DNA knots may be a formidable roadblock for the development of devices which support substantially longer read lengths . We also show that structural properties of DNA , like its resistance to bending , are intimately linked to the molecule's tendency to form knots . We demonstrate how this connection can be utilized to introduce mathematical models of DNA which account for the molecule's overall statistical properties .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "sequencing", "techniques", "engineering", "and", "technology", "dna", "electrophoresis", "mathematical", "models", "materials", "science", "nanotechnology", "molecular", "biology", "techniques", "dna", "macromolecules", "materials", "by", "structure", "research", "and", "analysis", "methods", "polymers", "polymer", "chemistry", "random", "walk", "electrophoretic", "techniques", "mathematical", "and", "statistical", "techniques", "materials", "by", "attribute", "chemistry", "molecular", "biology", "nucleotide", "sequencing", "biochemistry", "biochemical", "simulations", "nucleic", "acids", "nanomaterials", "genetics", "biology", "and", "life", "sciences", "physical", "sciences", "computational", "biology", "dna", "sequencing" ]	2016	A Monte Carlo Study of Knots in Long Double-Stranded DNA Chains
Whole genome duplication has shaped eukaryotic evolutionary history and has been associated with drastic environmental change and species radiation . While the most common fate of WGD duplicates is a return to single copy , retained duplicates have been found enriched for highly interacting genes . This pattern has been explained by a neutral process of subfunctionalization and more recently , dosage balance selection . However , much about the relationship between environmental change , WGD and adaptation remains unknown . Here , we study the duplicate retention pattern postWGD , by letting virtual cells adapt to environmental changes . The virtual cells have structured genomes that encode a regulatory network and simple metabolism . Populations are under selection for homeostasis and evolve by point mutations , small indels and WGD . After populations had initially adapted fully to fluctuating resource conditions re-adaptation to a broad range of novel environments was studied by tracking mutations in the line of descent . WGD was established in a minority ( ≈30% ) of lineages , yet , these were significantly more successful at re-adaptation . Unexpectedly , WGD lineages conserved more seemingly redundant genes , yet had higher per gene mutation rates . While WGD duplicates of all functional classes were significantly over-retained compared to a model of neutral losses , duplicate retention was clearly biased towards highly connected TFs . Importantly , no subfunctionalization occurred in conserved pairs , strongly suggesting that dosage balance shaped retention . Meanwhile , singles diverged significantly . WGD , therefore , is a powerful mechanism to cope with environmental change , allowing conservation of a core machinery , while adapting the peripheral network to accommodate change . Eukaryotic genomes differ up to an astonishing 200000 fold in the amount of their DNA , by far the widest range within all domains of life [1] . In eukaryotic evolution large changes in genome size have heralded major transitions , starting with the radiation from a common ancestor of the eukaryotic supergroups within a short evolutionary timespan [2] , [3] . Subsequent dramatic radiations of animals in the Cambrian explosion and flowering plants have also been preceded by extensive increases in genome size [4] , [5] . But even within narrow taxonomic bounds remarkable levels of genome size variability exist , such as the seven fold difference within the Brachionus plicatilis species complex [6] . What are the evolutionary mechanisms underlying this flexibility in genome size and how does it affect the dynamics of eukaryotic evolutionary history ? Ever since Ohno first proposed that the genome of the vertebrate ancestor had undergone two rounds ( 2R ) of duplication [7] , evidence of the pervasiveness of WGD in eukaryotic evolution has been mounting . The 2R hypothesis itself has been strongly backed by recent phylogenetic studies [8] , [9] . Similarly , species radiations of angiosperms [10] , teleost fish [11] and yeasts [12] have all been associated with rounds of WGD . Especially in plants , the transition to polyploidy appears to be remarkably frequent . Therefore , in addition to all flowering plants being of paleopolyploid descent [10] , [13] , it is estimated that up to a third of all extant plant species underwent polyploidization since their most recent speciation [14] . Recognizing the ubiquity of WGD in eukaryotic evolution , it becomes crucial to understand the mechanisms that lead to their fixation in evolving populations . Data on plants suggest that changing environmental conditions can give rise to the establishment of polyploid lineages . For example , polyploid incidence is increased in harsher and newly arisen environments such as the arctic [15] and on islands created by volcanic activity [16] or at the ecological limits of non-polyploid parent species ( reviewed extensively in [17] ) . An extensively studied case of ancient WGD that happened in the ancestor of S . cerevisiae was shown to potentially yield a direct adaptive benefit when a novel , glucose rich environment arose [18] , [19] . However , direct adaptive benefits may not play a role in other historic cases of WGD which instead may be better explained by a general increase in evolvability . This may be the reason why a burst of WGDs in plants appears to coincide with the K-T boundary event , explaining the success of these lineages in overcoming the drastic change in climate conditions [20] , [21] . Most duplicates that arose from an ancient WGD event will have typically returned to a single copy state , thereby eroding the signal of WGD [12] . Remaining ohnolog ( duplicates arising from WGD ) fractions , ranging from 16% in yeast [22] to more than 50% in P . tetraurelia [23] , have been studied to gain insights in the potential adaptive benefits of WGD and evolutionary forces that play a role in post WGD genome evolution . In general , duplicate retention post WGD is not equal for all gene classes . A pattern found across species is an over-retention of transcription factors ( TFs ) and signaling genes in duplicate [24] , [25] , [26] , [27] , [28] , [23] . Neutral loss of subfunctions in both copies , for example losing different subset of target genes for TFs could enforce this retention [29] and has indeed been observed for Arabidopsis ohnologs [30] , [31] . However , a characteristic reciprocal relationship between the retention of duplicates resulting from WGD and small scale duplication ( SSD ) can not be easily explained by subfunctionalization . For example , TFs have been overretained post WGD , while underretained post SSD [32] , [33] , [34] , [28] , [35] , [26] . This pattern would , however , be predicted by the gene balance theory , because the two modes of duplication affect the balance between interacting gene products differently . Whereas a WGD should generally retain the balance between highly interacting genes , SSD most likely disrupts this balance by only increasing the dosage of a few genes [34] , [26] , [36] , [28] . This suggests that dosage balance selection could drive retention of duplicates post WGD [28] , [37] at least on short evolutionary timescales . Transient retention due to dosage balance selection can increase the chance that duplicates subfunctionalize or even neofunctionalize , further increasing the likelihood of duplicate retention [27] . How gene balance constraints affect gene divergence and loss remains , however , poorly understood . One important reason is that adaptive and neutral genome evolution post WGD can produce mixed conservation patterns [27] . In short , we lack a comprehensive mechanistic understanding of the causes and consequences of WGD when populations adapt to environmental change as well as its impact on long term genome evolution . Here we have taken an integrated modeling approach to study conditions for and consequences of fixation of WGD in populations that adapt to an environmental change . Within our Virtual Cell model , we tracked mutations and patterns of genome conservation along the line of descent . WGD was modeled as an ongoing mutation , alongside small scale duplications , deletions and rearrangements , as well as point mutations . Lineages arising from identical ancestral populations alternatively evolved with and without WGD , allowing for a direct comparison of the two modes of evolution . Our results show that fixation of a WGD increases the likelihood that a population will readapt successfully to a novel environmental condition . Surprisingly , the ancestral gene content of WGD lineages declines more slowly than that of lineages without WGD , while per gene mutation rates were higher in WGD lineages . At the same time , we found that ohnologs were over-retained relative to expectations based on random losses . This effect was strongest for TFs . In agreement with predictions from the gene balance hypothesis we found that TFs with many outgoing interactions were most likely to remain in duplicate . Because very little subfunctionalization was detected in these TFs we concluded that selection for dosage balance caused the over-retention pattern . Hence , a relatively simple , biologically inspired model can explain the association between WGD and environmental change as well as the overarching pattern of biased gene retention that is found in an expanding body of phylogenetic studies of paleopolyploidy . Initial adaptation times varied widely and those that reached high fitness within 15000 generations almost always involved one or more WGDs ( Fig . 2A ) . In contrast , re-adaptation times for lineages after environmental change were much shorter ( Fig . 2B , D–E ) . More than reached high fitness within 1000 generations . This was surprising , because at the start of re-adaptation , fitnessess dropped on average below the level of randomly initialization starting populations . In addition , in successfully re-adapting lineages WGD events became fixed in a minority of lineages , being particularly rare in rapidly re-adapting lineages ( Fig . 2B inset , F ) . The cases with rapid re-adaptations suggest that mutational paths to new phenotypes can be very short , requiring very little change at the genomic level . Notwithstanding the near absence of WGD in rapidly re-adapting lineages , fixation of WGD in the line of descent improved the overall success rate of re-adaptation from to ( Fig . 2C ) . Even though WGD-mutants were generated continuously in the population throughout the evolutionary experiments , very few WGDs were ultimately accepted in the line of descent of the final population . Accepted WGDs occured almost exclusively ( in of cases ) within 500 generations of the environmental change . The much shorter time scale of genomic expansion relative to the timescale of full re-adaptation is in agreement with our previous work on the Virtual Cell model , showing that early evolution of large genomes generally resulted in better long term evolvability [38] . To study the adaptation process following WGD in more detail we analyzed the evolution of gene content after the environmental change was applied . For all populations the environmental change took place 1000 generations after a fitness was first recorded in the population . At that time the genome was typically several fold larger than the minimum genome size reached towards the end of the simulation as a result of long term streamlining ( Fig . 3 inset ) . Our previous work on the Virtual Cell model showed that streamlining reduces mutational load by the removal of redundant genes and a focussing cellular function into a small set of essential genes [38] , explaining how a relatively large proportion of ancestral gene content is lost during the re-adaptation to the novel environment ( Fig . 3 ) . The conservation of gene content was measured as the fraction of genes in the ancestor , alive during the environmental change , that was maintained in subsequent descendants . Duplicates that arose from WGD and SSD later in evolution were not included in counts of ancestral gene content . As expected , continued neutral evolution in the control set led to drastic streamlining and turnover of the genome , resulting in the loss of approximately two thirds of the original gene content ( Fig . 3: gray shaded area ) . Re-adaptation to environmental change led to even larger changes in gene content , as expected . However , in contrast to our expectation that WGD copies are functionally redundant , a larger fraction of ancestral gene content was conserved in WGD lineages than in non-WGD lineages for more than 5000 generations after environmental change ( Fig . 3 ) . This was despite the fact that the per gene deletion rate remains constant with differences in genome size ( see Methods ) . Also , on the long run , the average conserved fraction in WGD lineages , although dropping below that of non WGD lineages , always remained above half the conserved fraction in non-WGD lineages . This shows that at least some fraction of the ancestral content was selectively retained in duplicate . To find an explanation for the difference in gene content conservation between WGD and non-WGD lineages we analyzed the fixation of different mutation types . The frequency of accepted deletions in neutral , WGD and non-WGD lineages were very similar , although slightly lower for WGD lineages over the whole simulation interval , compared to non WGD and neutrally evolving lineages . However , the fraction deleted per event , for mutations that were accepted , was much smaller for WGD lineages than for neutrally evolving and non-WGD lineages ( Fig . 4A; ) despite this fraction being equal in the background mutations for all three categories . The result was that smaller fractions of the genome were lost per generation in WGD lineages ( Fig . 4A inset; ) . In contrast , the rate at which point mutations were accepted was significantly higher in WGD lineages compared to non-WGD lineages ( Fig . 4B; ) , while for the latter , this rate was again very similar to that in the neutral control set ( ) . This suggested that individual genes diverged much faster in WGD lineages than non-WGD lineages . In summary , WGD appeared to promote the conservation of duplicated genes , while at the same time enabled genes to diverge more and change their function . A process that fits with these two characteristics is subfunctionalization . To investigate whether the contrast between gene content conservation and higher gene function divergence in WGD lineages could be explained by a subfunctionalization process we focused our subsequent analysis on the fates and divergence of the ohnologs . We performed random deletion simulation to find the expected pair retention fractions for TFs , enzymes and pumps , separately . For every evolutionary simulation in our test set a random deletion simulation was performed that had the genome configurations of the common ancestor at the time of environmental change in the evolutionary run as its starting point . In the random deletion run , equal amounts of deletions per gene class were performed to those found in the line of descent in the evolutionary run , but selection was omitted . The random deletion runs were pooled in the same way as the evolutionary runs to make comparisons . As shown in Fig . 5 the expected fractions of ohnologs after randomly selecting genes for deletion are much lower than in the evolutionary simulations . Over-retention is highly significant in the case of TFs ( ) and detectable in enzymes and pumps ( ; ) . Despite the difference in the strength of the bias between TFs and enzymes , the fraction of these respective gene types that is conserved as ohnologs is very similar toward the end of the simulation ( ) , although the fraction has stabilized for TFs , while it is still declining for enzymes . This can be understood by the fact that the rate of deletions is much higher for TFs than for enzymes , resulting in a shift towards higher fractions of enzymes and lower fractions of TFs in the late , streamlined descendants ( Fig . S1 ) . Thus , even though TFs were on the whole more likely to be removed from the network by streamlining , the TFs that were conserved at long evolutionary timescales were much more likely to remain in the genome as ohnologs . In the next section we will test the hypothesis that TF connectivity is the determining factor for the retention of TF ohnologs . In the neutral control set continuous streamlining is responsible for the pattern of increasing TF outdegree ( Fig . 6A gray shaded ) . TFs with a relatively high outdegree remained in the genome at the expense of more sparsely connected TFs , which was also true for WGD and non-WGD simulations . Despite going through environmental change , the evolutionary pattern of non-WGD ( cyan ) lineages is very similar to the neutrally evolving controls . For the WGD lineages , the connectivity of retained genes was calculated separately for ohnologs ( red ) and singles ( yellow ) , revealing a marked difference in their evolved connectivity ( ) . Significantly higher connectivities of ohnologs compared to those of conserved genes in non-WGD ( ) and neutrally evolving lineages ( ) suggests that ancestral connectivity influences duplicate retention post-WGD . At the same time , singles in WGD lineages had significantly lower connectivities , both compared to the ohnologs and the conserved genes in non-WGD ( ) and neutral lineages ( ) . These results raised the possibility that the observed biased retention of TF ohnologs was a side effect of the conservation of highly interacting genes . To test this , we performed additional random deletion experiments . Now , instead of having an equal probability for each TF to be deleted , deletion probability was made dependent on the ancestral TF connectivity . The probabilities were determined by looking at the distribution of connectivities in the ancestral network and determining the fractions of conserved genes per connectivity bin , at subsequent points in evolutionary time ( see Methods for details ) . Performing such a simulation on the ancestral networks produced connectivity changes over time that were highly comparable to the evolution of connectivity in the evolutionary runs . Importantly , however , adding connectivity bias to the random deletion experiment did not change the result that ohnologs were over-retained in the evolutionary simulations ( Fig . S2 ) . We therefore concluded that conservation of highly connected TFs alone could not explain the over-retention of TF ohnologs . Continuing our investigation of the role that subfunctionalization may have in the conservation of gene content and high levels of divergence at the gene level in WGD lineages we investigated the functional divergence of TF ohnologs . If both ohnologs would diverge in function at the same time , they would no longer be able to fully compensate for each other's loss , making the conservation of both more likely . Functional divergence of a TF could happen if its binding site ( BS ) changes and it starts to regulate a different gene set . In general , BS divergence of ancestral genes was substantial on a short timescale , even in neutrally evolving lineages . Later , however the initial divergence was largely undone ( Fig . 6B ) . This reversal of initial divergence can be attributed to the long term genome streamlining that is expected to remove redundant and non-functional genes [38] , expected to be enriched in highly diverged genes . Compared to lineages that did not undergo an environmental change , BSs of re-adapting lineages initially diverged much more initially , highlighting the fast pace of evolutionary change immediately after the environmental change ( Fig . 6B ) . Interestingly , WGD lineages had significantly higher levels of BS divergence compared to non-WGD lineages on the shorter timescale , both in the ohnologs ( ) and singles category ( ) . Subsequently , the remaining ohnologs showed a drastic reduction of the level of divergence , eventually reaching a BS conservation level above that of conserved genes of non-WGD lineages ( ) . The sharp reduction in average BS divergence indicated that fast diverging ohnologs were overwhelmingly lost , while ohnologs that , on the other hand , did not mutate away from their ancestral BS were conserved at long evolutionary timescales . When one of an ohnolog pair is deleted in the course of evolution the remaining gene is subsequently categorized as a single . It is therefore not surprising that the singles category had a final level of BS divergence that was much higher than that of ohnologs ( ) , receiving an influx of highly diverged genes from the ohnolog category . However , their long term conservation may in fact be best explained by their diverged role in the network . It suggests an interesting dual character for the dynamics of post WGD genome evolution . On the one hand , selection acted to conserve highly interacting genes most strongly , both in their copy number and their interaction partners . At the same time , genes of lower connectivity may be returned to a single copy status and diverge in their role within the gene network . The latter process may be particularly important for adaptation in a new environment . Together , our results show that the conservation duplicate pairs and interaction partners of highly connected genes is compatible with the gene balance hypothesis , while subfunctionalization did not play a significant role in pair retention in our model . Instead , most functional divergence was observed in genes conserved as singles after WGD . Apparently these were more free to evolve and adapt to the new environment than ‘singles’ in non-WGD lineages , in congruence with the overall higher rates of divergence in WGD lineages compared to non-WGD lineages ( Fig . 4 ) . In this study we have taken an open-ended approach to studying the relationship between drastic changes in the environment and the occurrence of WGD in the line of descent . Evolving lineages could potentially follow many different evolutionary paths to re-adaption as a result of mutations at multiple scales and a complex genotype to phenotype map . WGD , despite being an ongoing mutation , was observed exclusively in lineages that were still ill-adapted to the prevailing environment , as was the case early during the initial adaptation phase and shortly after an environmental change ( e . g . Fig . 2D ) . This mirrors phylogenetic studies linking WGD to environmental and other types of drastic intracellular change [16] , [15] , [20] , [19] , [39] . One or more WGDs occurred during the initial adaptation phase in almost all lineages that would eventually obtain high fitness . In contrast , a minority of lineages ( ) fixed a WGD following environmental change , while almost no WGDs were observed when re-adaptation was very rapid . This indicates that some of the imposed environmental changes were more easily met by a relatively minor recalibration of the pre-evolved regulatory circuits , despite causing severe initial drops in fitness . Nevertheless , successful re-adaptation was more prevalent in lineages with WGD , and consequently larger genomes . This corroborates our previous research , showing that large genome increases early during adaptation benefit long term adaptation [38] and is in accordance with a similar inference drawn by Van de Peer and co-workers [21] , [20] based on parallel paleopolyploidy events in plants and frequent species radiation in the wake of WGD [40] , [33] , [41] , [5] . A particular case in point of long term evolvability due to WGD is the evolution of novel signaling and developmental pathways in vertebrates [42] , [43] , [24] . In addition to the long term benefits , in most lineages immediate positive fitness effects also played a role in establishing WGD ( Fig . S3 ) . Moreover , WGD was more frequent after particular types of environmental change , most notably when enzyme degradation rates increased ( Fig . S4 ) . This again parallels observations from the phylogenetic record and experiments . For example , there is strong evidence that the ancient WGD in yeast had an immediate benefit in the context of newly evolved fruiting plants [18] , [19] . Moreover , short term fitness advantages appear to play a role in establishing polyploid lineages in founder populations within newly arisen environments [16] , [15] , [44] , [17] . Strong genome streamlining occurred in all simulations , irrespective of environmental change and the fixation of WGD , indicating that maintenance of large genomes comes at a considerable mutational cost [44] , [38] , [45] . However , WGDs create “irremediable complexity” [46] , [28] , [47] , enforcing the maintenance of larger genomes , which would put lineages that evolve to equal fitness without WGD at an advantage . This may explain the relatively low fraction of WGD lineages in our experiments and could be the reason that , although polyploids are widespread among current plant species , their long term survival rate tends to be lower than that of non-polyploids [14] . Summarizing , our simple Virtual Cell model shows a pattern of occurrence of WGD very similar to that in the expanding record of established WGD events in extant organisms . We conclude that it is a generic property of the evolutionary process irrespective of particular evolutionary contingencies and most biochemical constraints . Our results highlight the intricate interplay of short and long term adaptive evolution as well as neutrality and irremediable complexity in shaping the gene content . This is moreover apparent from the duplicate retention pattern , as discussed below . The fractions of ancestral genes that remain in WGD pairs were higher in all functional categories compared to a neutral expectation based on random deletions . Duplicate retention post WGD was strongly biased towards highly interacting genes , a pattern that has been reported for many paleopolyploid species [30] , [48] , [23] , [28] , [43] , [26] , [25] . However over-retention of pairs , in particular in TFs , was much higher than expected from a biased retention of highly connected genes . The maintenance of duplicate pairs therefore needs another explanation and suggests a form of irremediable complexity . The two main explanations being subfunctionalization and , as recognized more recently , dosage balance selection . We found no evidence that subfunctionalization played a role in WGD pair retention withing our model , as the duplicates remained very similar . This is in contrast to what has been reported in various cases of duplicate retention [29] , [30] , [31] , [49] , [50] and the hypothesis that it was the main cause of genome complexification in eukaryotes [51] . There is in fact ample evidence that sub- and neofunctionalization play an important role in cementing the retained duplicates in the genomes of real organisms [30] , [27] and promote innovation [43] , [24] , [52] , although evidence exists that competitive interference between the paralogs may impose a significant obstacle to neutral loss of subfunctions [53] . The lack of subfunctionalization in our model can be explained as follows . Subfunctionalization of regulatory interactions would require that TFs can conserve binding interactions with a subset of ancestral sites , while at the same time losing some other sites . As such fine grained alterations of binding motifs was not possible within the current model due to the discreteness of the binding motifs and hence regulatory interactions , it presented a hard case scenario for subfunctionalization . The fact that we still observed over-retention , most prominently in TFs , again suggests the relevance of the dosage mediated retention mechanism . Another indication that dosage effects were important in the evolutionary dynamics was the observation that high protein degradation rates triggered fixation of adaptive WGD . Dosage balance selection was proposed to account for the inverse relationship between retention of duplicates post WGD and post SSD [37] , [34] , [28] , [54] , [55] , [26] . Originally , dosage balance selection is expected to affect proteins that are part of larger protein complexes . For complex assembly it is assumed that the relative dosage of the constituents is required to stay within narrow bounds , to prevent the accumulation of incomplete complexes [32] , [28] , [37] . Therefore single deletions of a duplicate will mostly not be tolerated after WGD , preventing the return to single copy of subunits of large complexes . Interestingly , our results show that resistance to the deletion of a member of a WGD pair was high , even in the absence of protein complex assembly or physical protein interactions , but that it was still a function of the number of its interactions . This indicates that dosage balance drove the retention . Although weaker than TFs , enzymes pairs were also significantly over-retained post WGD in our simulations . Biased retention of enzyme duplicates has also been reported for the latest of P . tetraurelia's three successive WGDs [23] . Curiously however , enzymes were significantly under-retained from the earlier WGD events . Initially , stoichiometric constraints likely impose dosage balance selection on enzymes in metabolic pathways [56] . However , over longer evolutionary timescales , the enzymatic pathways may acquire compensating expression level changes that free the enzyme duplicates of dosage balance constraints , allowing them to be deleted . Indeed , looking at the trend within the fraction of enzymes found in pairs in our simulations ( Fig . 5 ) , the decline phase is longer than for TFs and may have continued with longer simulation times , explaining the varying levels of retention at different evolutionary timescales . Summarizing , in our simulations dosage sometimes played an important role in establishing adaptive WGD as well as driving the retention of duplicate pairs , conserving core regulatory interactions in the absence of subfunctionalization . This raises the question how novel functions could evolve within our simulations , without significant divergence of conserved ohnologs . The answer appears to be provided by the behavior of the singles in WGD lineages . They were changing much faster than duplicates and also notably faster than genes retained in non-WGD lineages ( Fig . 6B ) . This opens the possibility that the adaptive success of WGD lineages was in part due to more sparsely connected TFs ( Fig . 6A ) that were not essential for fitness and were therefore more free to evolve . These are expected to be in relative abundance immediately after a WGD . This scenario can , moreover , explain the result that even though genome conservation was higher in WGD lineages , individual genes appeared to diverge faster from the ancestral state . Thus , enhanced evolvability of WGD lineages was not primarily a consequence of ‘freeing’ redundant paralogs to adopt new functions , but most importantly due to unhindered evolution of non-paralogous genes to adapt to novel conditions . An important aspect of polyploidization that was left out of our modeling is the variety of mechanisms that can lead to polyploidization . WGD in the current model happened exclusively through autopolyploidization , causing a strict duplication of the genetic material . In contrast , hybridization between individuals from substantially diverged subpopulations can give rise to important phenomena such as biased fractionation patterns [57] , [58] and hybrid fitness [59] , [60] . We envision that incorporating these mechanisms into the current model could give insight into the adaptive consequences of hybridization events and help recognize the type of ancient polyploidization events by observing characteristic patterns of genome evolution . Concluding , our model highlights how the interplay between short and long term adaptive and neutral processes underlies the presence of WGD and post-WGD gene retention and its role in genome complexification . Although we did not set out to model this property explicitly , dosage effects and selection for retaining balanced gene expression readily emerged in the Virtual Cell model , underlining its importance as a generic property of evolution , shaping the content of genomes . In addition , WGD appears to enable the divergence of singly retained ancestral genes , which may endow WGD lineages with long term adaptive benefits . From a broader perspective , our results suggest that WGDs had a defining role in enabling the innovations in eukaryotic lineages , while preserving the hallmarks of their ancestors . The evolutionary simulations were run in two stages . In the first stage 100 populations were randomly initialized and independently evolved until 1000 generations after they passed the high fitness cutoff , continuing to a maximum of 15000 generations . All populations in this batch were evolved under the same standard environmental conditions ( the same as those in [38] ) . From the populations that evolved to a high fitness 10 were randomly selected to go to the next stage . In this stage all ten populations were initially cloned 80 times and each cloned populations uniquely assigned to 1 out of 80 novel environmental conditions , per seed population . After applying the environmental change , populations evolved a further 15000 generations . As a control for the effect of environmental change , evolution was continued without environmental change for all populations from the batch of 100 simulations that evolved to high fitness , including the 10 selected seed populations . The new environments were made by changing the values of five parameters of the system relative to their standard values . These parameters separately control the degradation rate of enzymes , permeability of the resource molecule ( A ) , the internal target concentrations for homeostasis in X and A and the metabolic conversion rate of A to X . For all 5 parameters low and high conditions were chosen by making them a factor 2 to 4 different from the standard environment , depending on the severity of the effect on populations in test simulations ( see Table S1 ) . For some parameters , too large changes resulted in non-viable conditions for most populations , constraining the change we could effectively apply in our simulations . Finally , a systematic set of 80 environmental changes was constructed by making all combinations where exactly three parameters differ from their value in the standard condition . Exact counts of mutations can be traced in the line of descent . To do this , one individual of the final population is selected and its ancestors traced back to the start of the simulation . Mutation events in the line of descent are converted to rates by averaging mutations over 1000 generation intervals . The ancestral gene content conservation was measured using the ancestor trace . The gene content of the ancestor that was alive when the environment was changed is the point of reference . The genomes of individuals in the line of descent was overlapped with the reference and what remains of the ancestral gene content was expressed as a fraction . In the case of WGD or other types of duplication of ancestral genes , only one ( random ) copy is considered the original , while the other copy does not count towards the ancestral content . After WGD the genes that originated at the WGD event were traced in the line of descent . Again , duplicates that arose subsequent to this reference point were not counted towards ancestral WGD genes . The WGD genes that had remained in the genome were then divided into singles and ohnologs . Duplicate retention fractions are the intact ohnologs divided by the total amount of ancestral WGD genes still present in the genome . The background expectations were calculated by counting the losses per gene category of ancestral WGD genes along the line of descent and then resimulating the losses , starting from the first ancestor with the WGD . In the first variant of the random deletion simulations , the deletion counts were performed entirely randomly per gene category on the ancestral content . In the second variant , the probabilities of selecting a gene for deletion were made dependent on their connectivity . The scaling was calculated directly from the original evolutionary experiments , by assigning genes ( TFs ) to bins according to their relative connectivity in the ancestral WGD network and finding the likelihood of deletion in each connectivity bin . This resulted in comparable connectivity curves for the connectivity adjusted random deletion experiments and the evolutionary simulations . Significance of reported differences between WGD , non-WGD and neutrally evolving lineages were all determined by Mann-Whitney rank sum tests , ranking the scores of individual runs within each of the subsets . In the case of mutation rates , the ranking was done over the full evolutionary time interval , while in the case of ohnolog conservation , TF connectivity , and BS divergence ranking was performed on the scores of the last time point , unless stated otherwise in the main text .	The evolution of eukaryotes is characterized by drastic changes in their genome content . Genome expansions have often occurred by duplication of the entire genome . It is generally not know whether organisms gain any adaptive advantage from these mutations . However , they appear to become fixed in response to environmental change . Many interesting whole genome duplications happened long ago in eukaryotic evolutionary history during periods of turbulent genome and species evolution . Genomic data analysis alone cannot resolve the evolutionary mechanisms and consequences of whole genome duplication . Here , we modeled evolution with whole genome duplications in a Virtual Cell model . Simulating populations that undergo a range of different environmental changes we found that next to often increasing fitness directly , whole genome duplications made lineages more evolvable and hence more able to adapt to harsh new environments . Although most duplicates are deleted in subsequent evolution , genes with many interaction partners were retained preferentially , increasing regulatory complexity . Interestingly however , we found that innovation happened most likely in the more loosely connected and less essential genes .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "genome", "evolution", "evolutionary", "modeling", "biology", "and", "life", "sciences", "computational", "biology", "evolutionary", "biology" ]	2014	A Synergism between Adaptive Effects and Evolvability Drives Whole Genome Duplication to Fixation
It is estimated that India has more deaths from rabies than any other country . However , existing estimates are indirect and rely on non-representative studies . We examined rabies deaths in the ongoing Million Death Study ( MDS ) , a representative survey of over 122 , 000 deaths in India that uses enhanced types of verbal autopsy . We estimated the age-specific mortality rates of symptomatically identifiable furious rabies and its geographic and demographic distributions . A total of 140 deaths in our sample were caused by rabies , suggesting that in 2005 there were 12 , 700 ( 99% CI 10 , 000 to 15 , 500 ) symptomatically identifiable furious rabies deaths in India . Most rabies deaths were in males ( 62% ) , in rural areas ( 91% ) , and in children below the age of 15 years ( 50% ) . The overall rabies mortality rate was 1 . 1 deaths per 100 , 000 population ( 99%CI 0 . 9 to 1 . 4 ) . One third of the national rabies deaths were found in Uttar Pradesh ( 4 , 300 ) and nearly three quarters ( 8 , 900 ) were in 7 central and south-eastern states: Chhattisgarh , Uttar Pradesh , Odisha , Andhra Pradesh , Bihar , Assam , and Madhya Pradesh . Rabies remains an avoidable cause of death in India . As verbal autopsy is not likely to identify atypical or paralytic forms of rabies , our figure of 12 , 700 deaths due to classic and clinically identifiable furious rabies underestimates the total number of deaths due to this virus . The concentrated geographic distribution of rabies in India suggests that a significant reduction in the number of deaths or potentially even elimination of rabies deaths is possible . Rabies has been recognized for many millennia in India , long before Aristotle recognized the disease in the Graeco-Roman era [1] . The ancient Vedic text “Sushruta Samhita” contains graphic descriptions of rabies in animals and in humans: “If the patient becomes exceedingly frightened at the sight or mention of the very name of water , he should be understood to have been afflicted with Jala-trsisa ( hydrophobia ) and be deemed to have been doomed” [2] . Several indirect estimates [3]–[4] have suggested that modern India has more rabid dog bites and human rabies deaths than any other country . In 2002 , the World Health Organization ( WHO ) estimated that rabies caused 30 , 000 human deaths per year in India , which accounted for approximately 60% of the estimated global total of rabies deaths [5] . A non-representative survey based on case detection of rabies , and verbal autopsies of identified furious rabies cases , estimated about 17 , 000 human rabies deaths for the whole country [3] . This total was further expanded by 20% to account for paralytic and atypical forms and resulted in the widely quoted final figure of just over 20 , 000 rabies deaths per year . In 2004 , a dog-bite probability model was used to re-evaluate the burden of rabies in Africa and Asia . This method also yielded an estimate of about 20 , 000 human deaths from rabies in India [4] . All these estimates are much higher than the Government of India's official reported deaths in the range of 244 to 556 per year between 2000 and 2009 [6] based on routine hospital surveillance which is likely to miss many rabies deaths . The official Government of India reports of rabies deaths from hospitals are underestimates for several reasons . Most deaths in India occur at home , in rural areas , outside medical care , and there are very large numbers of stray dogs throughout India which frequently bite humans [7]–[9] . In many states , a lack of community access to education about post-exposure rabies prophylaxis and adherence to traditional beliefs about the disease are likely to increase the risk of developing rabies after exposure . Laboratory confirmation of rabies in humans or animals in India is rarely possible . Typical signs and symptoms of classic “furious” rabies are striking and uniquely characteristic and are therefore well recognized by both medical staff and lay people . However , paralytic “dumb” rabies and atypical presentations may easily be misdiagnosed as other neurological entities [10]–[13] . Effective dog rabies control , and possibly elimination , is achievable in India [14]–[15]; however , data on the prevalence of the disease and its distribution across the states are required to raise public awareness , give direction to control programmes , and to establish a basis against which to measure the success of future efforts to reduce rabies transmission or deaths . Here , we provide an estimate of national and regional human rabies mortality based on a nationally representative direct survey of over 122 , 000 deaths in India . We focus on understanding the geographical , age , and gender distributions of rabies deaths . Following each 10-yearly census , the Registrar General of India ( RGI ) divides India into approximately one million units , each containing about 1 , 000 people . In 1993 , the RGI randomly selected 6 , 671 of these units from the 1991 census , from all 28 states and 7 union territories of India , to be included in its Sample Registration System ( SRS ) . The SRS is representative of India at the rural/urban stratum for the major states of India . Each unit has about 150 households ( totaling 1 . 1 million households and approximately 6 . 3 million people ) , which are monitored for vital events on a monthly basis by a part-time enumerator and every 6 months by a full-time surveyor . The Million Death Study ( MDS ) seeks to assign causes to all deaths in the selected SRS areas for the period from 2001 to 2014 [16]–[21] . Verbal autopsy is a tool used to ascertain cause of death based on a structured interview with the relatives or close associates of the dead , in areas where medical certification of the cause of death is lacking . As part of the MDS , an enhanced type of verbal autopsy , using both an open-ended narrative and close-ended questions [16] , [22] ( termed RHIME: Routine , Reliable , Representative and Re-sampled Household Investigation of Mortality with Medical Evaluation ) , was administered by trained RGI surveyors for each identified death starting from 2001 . Two of 130 trained physicians independently reviewed each completed RHIME and assigned a single cause of death using the International Classification of Diseases 10th revision ( ICD-10 ) [23] and specific guidelines developed for the MDS [24] . Differences in coding were resolved by anonymous reconciliation of initial codes , and if needed , by a third , senior physician who adjudicated the final cause of death . Details of the methods , validation and preliminary results for various conditions have been reported elsewhere [16]–[19] , [25] . About 5% of deaths in the MDS sample were randomly re-sampled and subsequently independently re-interviewed by teams other than the SRS staff . From the MDS data available ( 2001–2003 ) , we identified all deaths in which at least one physician had coded rabies ( ICD-10 code A82 ) or dog bite ( ICD-10 code W54 ) as the cause of death . All non-English narratives were translated into English and data were extracted in a standardized fashion . Based on a preceding history of exposure to a dog [or other mammal] bite combined with symptoms such as altered behavior , hydrophobia , psychosis/delirium/confusion , and fever , the causes of deaths were classified as either rabies or not rabies by the authors . We further characterized the rabies deaths by gender , age , urban or rural location , and region . To account for sampling design , the age-specific proportions were weighted according to the SRS sampling fractions in the rural and urban parts of each state [18] , [20] , [26] , although such sampling made little difference to the estimated national totals . Using methods described previously , the proportion of deaths coded as rabies was applied to the United Nations ( UN ) Population Division estimates of deaths in India in 2005 [27] to generate rabies specific death totals and rates for India and its major states . SRS enrolment is on a voluntary basis , and its confidentiality and consent procedures are defined as part of the Registration of Births and Deaths Act , 1969 . Oral consent was obtained in the first SRS sample frame . The new SRS sample obtains written consent at baseline . Families are free to withdraw from the study , but the compliance is close to 100% . The study poses no or minimal risks to enrolled subjects . All personal identifiers present in the raw data are anonymized before analysis . The study has been approved by the review boards of the Post-Graduate Institute of Medical Education and Research , St . Michael's Hospital and the Indian Council of Medical Research . A total of 95 of the 122 , 429 surveyed deaths in 2001–3 were coded as rabies by at least one physician . An additional 59 cases were coded as dog bite . Following central review of the details of each of these dog bite deaths , 45 were re-classified as rabies , arriving at a total of 140 . The majority of rabies deaths occurred in rural areas ( 91% ) and few occurred in health care facilities ( 16% ) ( Table 1 ) . About 97% of rabies deaths were the result of dog bites and the remaining 3% were from cat and wild mammal bites . The median time from a bite to death was 8 weeks ( range 1 week to 4 years ) . Hydrophobia was described in 22% of rabies deaths and other neuropsychiatric symptoms , such as altered behavior ( 49% ) , psychosis/delirium/confusion ( 21% ) , restlessness ( 14% ) , barking/cough ( 18% ) , and dysphagia ( 6% ) were also mentioned in the narratives . Among the treatment histories of patients detected by our survey , 65% ( 91/140 ) had not sought any hospital treatment . While we are not able to infer the specific nature of treatment sought , 34% ( 48/140 ) received one or more injections after their most recent bite . However , only one patient completed a course of 14 injections , which constitutes complete treatment with the rabies vaccine most commonly used in India at the time of our study . Most of the remaining 47 patients received only 1–3 injections , though 5 patients received 4–10 injections ( Table 1 ) . Projection of the 2001–3 survey deaths from rabies to 2005 UN death totals , yields 12 , 700 ( 99% CI 10 , 000 to 15 , 500 ) symptomatically identifiable furious rabies deaths in India ( Table 2 ) . Approximately 62% of all rabies deaths in India in 2005 were in males and 50% were in children under 15 years . The overall rabies mortality rate was 1 . 1 deaths per 100 , 000 population ( 99% CI 0 . 9 to 1 . 4 ) , with the highest rates being in children under 5 years and in the elderly age 70 years or older . Rabies deaths were not evenly distributed throughout the country . One third of all rabies deaths were found in Uttar Pradesh ( 4 , 300 ) and nearly three quarters ( 8 , 900 ) were in 7 central and south-eastern states: Chhattisgarh , Uttar Pradesh , Odisha , Andhra Pradesh , Bihar , Assam and Madhya Pradesh . Among larger states , the highest rates of rabies death per 100 , 000 population were in Chhattisgarh ( 3 . 5 ) , Uttar Pradesh ( 2 . 3 ) , and Odisha ( 1 . 9 ) . ( Figure 1 and Table S1 ) . No rabies deaths were reported in study areas from the following states: Kerala , Jammu & Kashmir , Jharkhand , Manipur , Meghalaya , Nagaland , Sikkim , Mizoram , Andaman & Nicobar Islands , Lakshadweep , Chandigarh , Dadra & Nagar Haveli and Daman & Diu . Together , these states represent approximately 7% of India's population . Of the 5% ( n = 3275 ) MDS sample deaths randomly chosen for independent re-sampling and re-administration and coding of the VA , 2 were originally coded as rabies . Both of these deaths were again identified as rabies in the re-sampling process and there were no other rabies deaths This study is the first to provide an estimate of deaths from symptomatically identifiable furious rabies based on a representative sample of Indian deaths and to report the geographic , age and gender distributions of these deaths . While the MDS was not designed specifically to identify rabies deaths , its large size , and representative sampling make it suitable for identifying deaths due to relatively rare conditions and subsequently generating reliable estimation of population based rates . Our figure of 12 , 700 ( 99% CI 10 , 000 to 15 , 500 ) human deaths from rabies in 2005 is within the uncertainty ranges of a recent indirect estimate by Sudarshan and colleagues of 17 , 137 ( 95% CI 14 , 109–20 , 165 ) prior to the addition of 20% to account for paralytic/atypical forms of the disease [3] . While the Sudarshan study also used verbal autopsies , it relied on case finding in communities located near large medical centers followed by interviews of people in the communities in which the cases originated and thus cannot be considered a truly nationally representative sample . Similarly , the derivation of 19 , 713 ( 95% CI 4 , 192–39 , 733 ) human deaths using a dog-bite probability model is based on several assumptions [4] , most notably that the epidemiology of canine rabies in India , where very few dogs are tested for rabies , is similar to that in Africa . To our knowledge , there have been no nationally representative studies of canine rabies in India . Despite these methodological challenges , the three studies together suggest a range of rabies deaths between 13 , 000–20 , 000 deaths . Although we did not report any rabies deaths in a small number of states ( which represent less than 7% of India's population and total deaths ) , routine government hospital data [6] and medically certified causes of death from urban areas [28] from 1998 to 2004 , would add only about an additional 100 to 500 rabies deaths from these states ( Figure S1 ) . Thus , the inclusion or exclusion of these states does not alter our national estimate of 12 , 700 deaths and lies well within the 99% confidence range of our estimates ( 10 , 000–15 , 500 ) . To further compare our rabies mortality estimates with other estimates , we plotted the proportional mortality from rabies for each of the years from 2001–2003 of the MDS and the estimated proportional mortality of rabies from various government surveys and other published studies over a two-decade period ( Figure 2 ) . This figure shows that our estimate of proportional mortality for rabies ( 1 . 3 per 1000 deaths ) is consistent with other data sources and also with the apparent steady decrease in rabies as a cause of death in India starting in the early 1990s . Figure 2 also suggests a crude cyclical pattern of deaths . The demographic characteristics of our estimates were generally similar to those reported by other epidemiological studies in India . Sixty two per cent were males ( compared to 71% [28] , 72% [29] , and 66% [9] ) and 50% were children less than 15 years old ( compared to 35% [28] , 28% [29] , and 54% [9] ) . While the MDS was not designed to examine rabies treatment , we were nonetheless able to extract useful information from the narratives . The completely treated cases probably received the Semple-type rabies vaccine that was still widely used in India during the study period ( 2001–03 ) [30] . The partially treated cases might have received Semple or cell culture rabies vaccine , tetanus toxoid , antibiotics , another drug , or a traditional remedy . Since treatment information was contained only in the narrative , we are not able to comment on the timing or specific contents of the injections received by the deceased . The most important limitation in our study is the potential for misclassification of rabies deaths as other causes of death . Some rabies deaths were in fact misclassified as being directly due to dog bite , but central review enabled correction of this misclassification . Death with dramatic neurological symptoms ( including the pathognomonic symptom of hydrophobia ) occurring weeks or months after a dog bite would seem to be a distinctive event that would readily be detected by verbal autopsy . However , it is well recognized that not all human patients develop typical furious rabies [31]–[32] and some may die after a short illness , before the signs are recognized or the history of an animal bite is elicited and others may have a long incubation period , in exceptional cases up to about 20 years [33] . Verbal autopsy is unlikely to be able to identify such cases . Furthermore , an unknown proportion of human rabies victims in India develop more insidious paralytic or atypical features without hydrophobia or alternating excitation and lucidity , making it unlikely that rabies will be identified as the cause of death by their family , neighbors or medical staff [10]–[11] , [34]–[35] . Paralytic rabies most often resembles other encephalomyelitides or Guillain-Barré syndrome/Landry's paralysis , but many other atypical presentations of rabies have been reported [12] , [36]–[39] . The proportion of rabies cases presenting with paralytic or atypical symptoms is unknown , although estimates of “less than a fifth” [40] or one third [41] have been suggested but with little , if any supporting evidence . In the MDS , approximately 8 . 7% of captured deaths were deemed to be due to unspecified or ill-defined causes . We do not believe it likely that deaths due to typical rabies are included in this group . While it is possible that there are atypical cases of rabies included in this group , we believe that this number would be very small . Since verbal autopsy is unlikely to identify paralytic or atypical rabies deaths , our estimates presented in this study are restricted to typical , clinically identifiable classic furious rabies . Furthermore , human rabies cases often cluster geographically around a particular rabid dog that bites multiple people . The SRS was not specifically designed to identify such clustered events , and our results might therefore be under-estimating the true rabies mortality rate . Finally , the most recent data available for analysis from the MDS is from deaths that occurred in 2001–2003 . While it would have been preferable to have utilized more recent data , no other more recent nationally representative source of comparable data exists . MDS data collection is continuing and we will update our analysis , including for time trends , when newer data are available . We estimate that there were 12 , 700 deaths due to symptomatically identifiable furious rabies in India in 2005 . It is very important to note that this figure underestimates the total number of deaths due to rabies since paralytic and atypical cases would not have been detected by verbal autopsy . This study is the first to estimate rabies mortality based upon a nationally representative sample of deaths rather than modeling or from extrapolation from selected focal surveillance . Thus we provide previously unavailable regional and demographic information about human rabies deaths that can help to focus both human and canine rabies control programmes in the country and act as a baseline that can be used as comparison for future estimates of rabies mortality . Elimination of the canine reservoir of rabies is not likely in India at anytime in the near future . However , the concentrated geographic distribution of rabies in India suggests that a significant reduction in the number of human deaths or potentially even elimination of rabies deaths is possible and this study serves as a baseline against which future gains may be measured .	Rabies , a disease of antiquity , has been partially controlled in many countries and eliminated in a few . However , according to the World Health Organization , rabies continues to kill thousands of people in India each year , more than in any other country . We used an enhanced type of verbal autopsy ( a structured interview of the relatives or close associates of the dead by non-medical staff with central medical coding by at least two doctors ) to identify the causes of over 122 , 000 deaths in a large scale , representative sample in India in 2001–03 . Using these data , we estimate that in 2005 approximately 12 , 700 people died from symptomatically identifiable furious rabies . Because verbal autopsy is not able to identify atypical presentations of rabies , our figure underestimates the actual number of rabies deaths in India . The majority of rabies deaths occurred in males , in rural areas , in children below the age of 15 years , and in a few states . The concentrated geographic distribution of rabies in India suggests that targeting with preventive campaigns including vaccination of animals and post exposure vaccination of humans might achieve a significant reduction in the number of deaths or potentially even elimination of deaths from this disease .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "medicine", "social", "and", "behavioral", "sciences", "science", "policy", "biology", "veterinary", "science" ]	2012	Deaths from Symptomatically Identifiable Furious Rabies in India: A Nationally Representative Mortality Survey
The factors contributing to chronic Chagas' heart disease remain unknown . High nitric oxide ( NO ) levels have been shown to be associated with cardiomyopathy severity in patients . Further , NO produced via inducible nitric oxide synthase ( iNOS/NOS2 ) is proposed to play a role in Trypanosoma cruzi control . However , the participation of iNOS/NOS2 and NO in T . cruzi control and heart injury has been questioned . Here , using chronically infected rhesus monkeys and iNOS/NOS2-deficient ( Nos2−/− ) mice we explored the participation of iNOS/NOS2-derived NO in heart injury in T . cruzi infection . Rhesus monkeys and C57BL/6 and Nos2−/− mice were infected with the Colombian T . cruzi strain . Parasite DNA was detected by polymerase chain reaction , T . cruzi antigens and iNOS/NOS2+ cells were immunohistochemically detected in heart sections and NO levels in serum were determined by Griess reagent . Heart injury was assessed by electrocardiogram ( ECG ) , echocardiogram ( ECHO ) , creatine kinase heart isoenzyme ( CK-MB ) activity levels in serum and connexin 43 ( Cx43 ) expression in the cardiac tissue . Chronically infected monkeys presented conduction abnormalities , cardiac inflammation and fibrosis , which resembled the spectrum of human chronic chagasic cardiomyopathy ( CCC ) . Importantly , chronic myocarditis was associated with parasite persistence . Moreover , Cx43 loss and increased CK-MB activity levels were primarily correlated with iNOS/NOS2+ cells infiltrating the cardiac tissue and NO levels in serum . Studies in Nos2−/− mice reinforced that the iNOS/NOS2-NO pathway plays a pivotal role in T . cruzi-elicited cardiomyocyte injury and in conduction abnormalities that were associated with Cx43 loss in the cardiac tissue . T . cruzi-infected rhesus monkeys reproduce features of CCC . Moreover , our data support that in T . cruzi infection persistent parasite-triggered iNOS/NOS2 in the cardiac tissue and NO overproduction might contribute to CCC severity , mainly disturbing of the molecular pathway involved in electrical synchrony . These findings open a new avenue for therapeutic tools in Chagas' heart disease . Chagas disease , which is caused by the protozoan parasite Trypanosoma cruzi , afflicts 8–15 million individuals in endemic areas of Latin America and several hundred thousand people in other countries as a result of migration . Although vector transmission has been controlled , there are still more than 50 , 000 new cases of Chagas disease each year [1] , [2] . Despite high parasitism , which usually declines at immunity onset , the clinical signs are usually mild in the acute infection . After decades , most of the infected individuals remain in the asymptomatic indeterminate form , and ∼30% of the patients present arrhythmias and heart failure due to end-stage dilated chronic chagasic cardiomyopathy ( CCC ) , which is associated with inflammation , myocytosis and fibrosis [2] . The pathophysiological factors influencing the clinical outcome of Chagas disease remain unclear [2] . Due to the scarcity of the T . cruzi parasite , CCC has been associated with autoimmune recognition of heart tissue by T-cell-enriched inflammation [3] . There is a consensus , however , that parasite persistence and/or a parasite-driven deregulated immune response operates in CCC [4] , [5] . In this context , high nitric oxide ( NO ) levels have been shown to be associated with the severity of CCC in chronic chagasic patients [6] . Nitric oxide is an important cytotoxic and cytostatic factor in cell-mediated immunity to intracellular pathogens [7] . Excessive NO , however , may cause host injury , including a reduction of myocardial contractibility [8] . Nitric oxide is formed from L-arginine by isoforms of NO synthase ( NOS ) : the constitutive isoforms , neuronal NOS ( nNOS/NOS1 ) and endothelial NOS ( eNOS/NOS3 ) , and the cytokine-inducible NOS ( iNOS/NOS2 ) [8] . In T . cruzi infection , in vitro and in vivo evidence support that NO plays a pivotal role as a first line of parasite growth control [9] . Nevertheless , iNOS/NOS2-derived NO takes part in ventricular dilation and systolic dysfunction in T . cruzi-elicited acute myocarditis [10] . In chagasic patients , high NO levels have been shown to be associated with the severity of CCC [6] . Further , iNOS/NOS2 has been shown to contribute to T . cruzi control in acute infection [10] , [11] . This role , however , was challenged in a study with iNOS/NOS2-defcient ( Nos2tm1Lau ) infected mice showing that iNOS/NOS2 is not required for control of T . cruzi growth [12] . Moreover , iNOS/NOS2 participation in pathology has been questioned by study of gene polymorphism at promoter region in patients [13] . Therefore , there are doubts about the role played by iNOS/NOS2 and NO in T . cruzi infection . Adopting the model of nonhuman primate rhesus monkeys ( Macaca mulatta ) chronically infected with T . cruzi that reproduced several clinical , parasitological and immunological features of Chagas disease [14] , we descriptively investigated the involvement of iNOS/NOS2 and NO in Chagas' heart disease . Further , promptly , iNOS/NOS2-deficient mice were used to add insights on the participation of iNOS/NOS2-derived NO in T . cruzi-elicited heart injury . This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the Brazilian National Council of Animal Experimentation ( http://www . cobea . org . br/ ) and the Federal Law 11 . 794 ( October 8 , 2008 ) . The Institutional Committee for Animal Ethics of Fiocruz ( CEUA/Fiocruz , Licenses 161/03 and 004/09 ) and the Biosafety National Committee ( CQB/CTNBio , License 105/99 ) approved all the procedures used in this study . Seven male rhesus monkeys ( Macaca mulatta , 26±1 . 7 years old ) chronically infected with T . cruzi ( monkeys #42 , #64 , #68 , #90 , #95 , #99 , #103 ) were individually caged in the nonhuman primate units ( Double L Group Ltd . , USA ) of the Nonhuman Primates Breeding Service ( SCPRIM ) , of the Laboratory of Animals Breeding Center at Fiocruz ( CECAL/Fiocruz , Rio de Janeiro , Brazil ) . The monkeys were provided with water ad libitum and fed a commercial chow ( Nuvilab Primates 6030 , Nuvital , Brazil ) that was supplemented daily with fruits , eggs and vegetables . Temperature , humidity and light/dark cycles were standardized . Two noninfected age-matched male monkeys ( #81 and #94 ) were analyzed as controls . Four noninfected male monkeys ( L17 , L21 , M31 , N31 ) from our colony were used as controls for heart function evaluations . Metacyclic trypomastigotes of the Colombian T . cruzi strain were used to infect the monkeys subcutaneously in the arm [15] . These animals were studied during the acute and early chronic infection [15] . In this follow up study , the chronically infected monkeys were analyzed at 16–19 years postinfection ( ypi ) and followed for 48 months . The animals did not receive etiological treatment for T . cruzi . Examinations and procedures were performed under anesthesia with 10 mg/Kg ketamine chloride ( Vetaset , Fort Dodge , Iowa , USA ) intramuscularly according to the Guide for the Care and Use of Laboratory Animals ( NIH Publication No . 85-23 , revised 1996 ) . Blood was obtained by puncture of the femoral vein with appropriate tubes ( Vacutainer , Becton & Dickinson , USA ) . Prior to necropsy , monkeys were sedated with ketamine chloride and euthanized by exsanguination under deep plane of sodium thiopental ( Thiopentax , Cristalia , Belo Horizonte , MG , Brazil ) . Samples of the heart and all major organs were taken for histological studies and polymerase chain reaction ( PCR ) for parasite kDNA and genomic DNA detection . For the histopathology study , heart samples taken from five infected monkeys , which were sacrificed at the parasitemia peak ( 41 days postinfection ( dpi ) , monkeys #37 and #67 ) , when parasitemia was negative ( 70 dpi , monkey #77 , and 76 dpi , monkey #93 ) and 3 years postinfection ( monkey #45 ) used in a previous work [15] were added to our study . Five- to seven-week-old female C57BL/6 ( H-2b ) and iNOS-deficient ( Nos2−/−; B6 . 129P2-Nos2tm1Lau/J ) mice , resulted of seven backcrossings of the original Nos2-deficient lineage in C57BL/6 mice , were obtained from the animal facilities of Fiocruz and were maintained in specific pathogen free conditions . The mice were infected intraperitoneally with 100 blood trypomastigotes of the Colombian strain [16] . The classic 12-lead human electrocardiogram ( ECG ) system was used to analyze rhesus monkeys . Tracings were made at 25 mm/s at a voltage of 1 mV standardized to 1 cm ( ECG-6 , Ecafix , Brazil ) [14] . Two-dimensional and M-mode echocardiogram ( ECHO ) was performed on a regular basis and recorded with a multi-image camera ( Ultrasound Scanner EUB-555 , Hitachi , Japan ) . The ventricular function was assessed in the M-mode , by calculating the ejection fraction , and in the bidimensional mode , by semiquantitatively analyzing the global systolic function [14] . Electrocardiogram , ECHO and radiology ( chest , esophagus , colon ) were performed at 12-month intervals . Mice were intraperitoneally tranquilized with diazepam ( 20 mg/Kg ) , and transducers were placed under the skin for DII derivation . Electrocardiogram traces were recorded during two minutes using Power Lab 2/20 ( PanLab Instruments , Spain ) , analyzed with Scope Software for Windows v3 . 6 . 10 ( PanLab Instruments , USA ) [16] and independently analyzed by two investigators . Tissue fragments were fixed in 10% formalin , processed , and embedded in paraffin . Tissue sections ( 5 µm ) were stained with hematoxylin and eosin ( H&E ) and immunohistochemically ( IHS ) stained . To evaluate collagen deposits , heart sections were stained with Siriusrot F3B ( Chroma Gessellschaft , Germany ) in a saturated aqueous solution of picric acid and fast green . The proportion of collagen-positive areas was evaluated with the digital morphometric apparatus and analyzed with AnalySIS AUTO Software ( Soft Imaging System , USA ) . A polyclonal anti-iNOS/NO2 antibody was obtained from Cayman Chemical ( #160862 , USA ) , and an anti-connexin 43α1 ( Cx43 ) polyclonal antibody was purchased from Sigma ( #C6219 , USA ) . In addition , a polyclonal antibody recognizing T . cruzi antigens was produced in rabbits ( LBI/IOC-Fiocruz , Brazil ) . A biotinylated anti-rabbit and peroxidase-streptavidin complex was purchased from Amersham ( England ) . Antibodies and reagents were utilized in compliance with the manufacturers' instructions . Serial 5-µm sections were subjected to standardized IHS [16] . The material was counterstained with Mayer's hematoxylin . The T . cruzi- , iNOS/NOS2- or Cx43-positive areas in 25 fields ( 12 . 5 mm2 ) per section , in 3 sections per heart , were evaluated with a digital morphometric apparatus . For the quantitative studies of Cx43 expression only areas with preserved myocardial cells were analyzed . The images were analyzed with AnalySIS AUTO Software ( Soft Imaging System , USA ) . The areas that expressed the molecule of interest were integrated with the areas that did not express the molecule of interest , and the data were presented as the percentage of the positive area . Two or three different tissue samples per studied organ were processed separately for DNA extraction , and the purified DNA was PCR amplified using T . cruzi-specific kDNA minicircle primers [14] or analyzed by real time quantitative PCR ( qPCR ) using primers directed to the nuclear satellite DNA [17] . DNA integrity and the possible presence of PCR inhibitors were checked by amplification of the human β-globin sequence [14] . Samples showing no amplification for β-globin were retested after a new DNA extraction . For the real time PCR assays heart and spleen samples ( of one to two cm3 ) kept in liquid nitrogen were sectioned using a cryostat in conditions to avoid DNA cross-contamination and the studied area calculated using the AnalySIS AUTO Software ( Soft Imaging System , USA ) . As negative and positive controls , respectively , heart tissues from noninfected C57BL/6 mice or mice at 50 dpi with the Colombian T . cruzi strain were used [16] . The qPCR experiments were performed using the ABI Prism 7500 Fast ( Applied Biosystems , USA ) , in a final volume of 20 µL containing 2 µL DNA samples and 10 µl GoTaq qPCR Master Mix ( Promega , USA ) . Primers Cruzi 1 and Cruzi 2 were used for the T . cruzi nuclear satellite target [17] . The cycling conditions were as follows: 95°C for 10 minutes , followed by 40 cycles at 95°C for 15 seconds and 58°C for 1 minute . After amplification , the specificity of these primers was confirmed through melting curve analysis of the generated amplicons , revealing a solely melting temperature ( Tm ) for the amplified fragment . Each sample submitted to qPCR analysis was performed in triplicates and the results were expressed as mean values . Parasitic load quantification was obtained by absolute quantification of T . cruzi DNA , following normalization of the heart and spleen analyzed areas . The standard curve was generated by a 1∶10 serial dilution of DNA extracted from T . cruzi Colombian epimastigotes culture stocks , ranging from 106 to 10 parasite equivalents . The levels of total IgG and IgM were determined by kinetic nephelometry ( Beckman Coulter Array 360 , Beckman Coulter , USA ) in accordance with the manufacturer's instructions . Specific anti-T . cruzi IgG antibodies were measured by ELISA ( EIE-Chagas , BioManguinhos , Fiocruz , Brazil ) in accordance with the manufacturer's instructions . Nitrate and nitrite were determined in plasma samples from noninfected and infected monkeys using Griess reagent and vanadium chloride III with a standard curve of 0 . 8–100 µM NaNO2 and NaNO3 [18] . The activity of the CK-MB isoenzyme was measured in serum with a commercial kit ( Labtest , Brazil ) according to the manufacturer's recommendations . The optical density at 340 nm ( Microplate Reader Benchmark , Bio-Rad , USA ) was recorded every 2 minutes for 15 minutes [16] . Data are expressed as arithmetic mean ± SD . Student's t test was adopted to analyze the statistical significance of the apparent differences . All statistical tests were performed with SPSS 8 . 0 software . Differences were considered statistically significant at p<0 . 05 . The general characteristics of the studied monkeys are shown in Table S1 . Electrical conduction abnormalities found in acute and chronically T . cruzi-infected monkeys are summarized in Table 1 . Interestingly , the mild electrical abnormalities that we detected during the acute T . cruzi infection disappeared at the 4th month postinfection [15] . In this follow up study , during the chronic infection ( 18–23 ypi ) , three monkeys ( #42 , #64 and #68 ) showed transient electrical alterations , and three animals ( #90 , #95 and #103 ) presented significant electrical conduction abnormalities . The ECG patterns were normal in the 4 analyses of one monkey ( #99 ) . In infected monkey #90 , multiform ventricular extrasystoles were observed at 16 ypi , atrial extrasystoles were observed at 17 ypi , an incomplete left bundle branch block ( LBBB ) was observed at 20 ypi , and T-wave inversion was observed in all 4 analyses . In monkey #103 , first-degree atrioventricular ( AV ) conduction disturbance and T-wave inversion were seen at 17 ypi . In monkey #95 , LBBB was seen in the 4 exams , and the T-wave inversion was more accentuated and right QRS axis deviation at 20 ypi . Representative ECG registers of chronically T . cruzi-infected are shown in Figure S1 . As shown in Table S2 , only monkey #95 showed ECHO abnormality with asynchronic interventricular septum motility and a decreased left ventricular ejection fraction ( 72 . 4% and 53 . 2% at 16 ypi and 17 ypi , respectively ) . Altogether , monkeys #42 , #64 , #68 and #99 were considered non-cardiopatic and monkeys #90 , #95 and #103 were considered cardiopatic . All T . cruzi-infected monkeys showed normal radiological exams of the chest , esophagus and colon at the end-point of this study ( data not shown ) . Therefore , monkeys #42 , #64 , #68 and #99 presented the indeterminate form of the chronic Chagas disease . In the acute infection , intense heart parasitism was associated with pronounced inflammatory infiltrates at 41 dpi ( monkeys #37 and #67 ) . This effect subsided at 76dpi ( monkey #93 ) , which coincided with parasitemia control . Inflammation was focal or absent at 3 ypi ( monkey #45 ) [15] . In the chronic infection , none of the analyzed organs showed macroscopic alterations . In addition , no ventricular conduction system alteration was observed in the analyzed areas of the fibers composing the bundle of His in chronically T . cruzi-infected rhesus monkeys ( data not shown ) . The main histopathological alterations found in the cardiac tissue are represented in the Figure S2 . In monkeys with electrical abnormalities ( #90 , #95 and #103 ) and decreased left ventricular ejection fraction ( #95 ) , histopathological alterations were heterogeneous with areas with normal aspect ( Figure S2E ) , but mostly areas with a mild or intense multifocal myocarditis ( more pronounced in the left ventricle ) , with lymphocytes , macrophages , and an apparent lack of parasites , was associated with the disrupted spaces between cardiomyocytes and hypertrophy of the myocardial fibers ( Figure S2D , Figure S2F ) . Further , monkeys with the indeterminate form of Chagas disease presented focal myocarditis ( Figure S2A , Figure S2B ) or complete absence of cardiac inflammation or myocardial fiber injury ( Figure S2C ) . Conversely , noninfected controls ( Figure S2G ) consistently did not have inflammation or myocardial fiber injury . Compared with noninfected controls ( Figure 1A , Figure 1G ) , myocardial fibrosis with collagen deposits was detected in the early ( 41 dpi ) acute infection ( Figure 1B , Figure 1G ) . Fibrosis was residual in the early ( 3 ypi ) chronic infection ( Figure 1C , Figure 1G ) and characterized the indeterminate form of the disease . In the late ( 20 ypi ) chronic infection , the monkey that had transient alterations in clinical score ( #64 ) presented intense fibrosis ( Figure 1D , Figure 1G ) , while in the monkey with normal clinical score ( #99 ) heart histological aspects and collagen deposits resembling those of noninfected controls ( Figure 1E , Figure 1G ) . In the cardiopatic monkeys ( #90 , #95 and #103 ) , fibrosis ( Figure 1F , Figure 1G ) was related to the severity of electrical abnormalities . Collagen deposits varying from slight to severe were frequently associated with inflammatory foci ( Figure 1H–K ) . In areas of severe inflammation ( Figure 1H ) , fibrosis was abundant ( Figure 1I ) , and there was substitution of cardiomyocytes by mesenchymal cells ( Figure 1J ) and dense bundles of interstitial matrix that extended from the subepicardium into the subjacent myocardium ( Figure 1K ) . Therefore , in six out of seven T . cruzi-infected rhesus monkeys , clinical and histological alterations were compatible with chronic Chagas disease . Interestingly , three of these monkeys reproduced major aspects of CCC in patients [2] , [19] . Exhaustive microscopic examinations of paraffin heart tissue sections failed to detect T . cruzi pseudocysts or isolated amastigotes in all of the chronically infected monkeys . Immunohistochemistry experiments were performed to detect T . cruzi antigens , and extramyocytic antigens were seen as red amorphous spots ( Figure 2A ) in six out of seven infected animals ( the exception was monkey #99 ) . T . cruzi antigens were commonly surrounded by mononuclear cell infiltrates ( Figure 2A , black arrows ) ; however , inflammatory cells were also seen associated to apparently noninfected fibers ( Figure 2A , white arrow heads ) . In addition , blood samples and fragments of spleen and cardiac tissue were submitted to conventional PCR amplification targeting T . cruzi kDNA minicircles [14] . In the 1st analysis ( 16–19 ypi ) of blood , all T . cruzi-infected animals showed a PCR signal at ∼330 bp for kDNA , whereas the noninfected monkeys ( #81 and #94 ) did not show a signal ( Figure 2B ) . In the 4th analysis ( 20–23 ypi ) , a PCR signal at ∼330 bp was detected in the blood of six infected rhesus monkeys ( Figure 2C ) . Furthermore , T . cruzi kDNA was identified in the cardiac septum , left ( Figure 2D ) and right ventricles , left and right atria , and aorta of all infected monkeys except monkey #99 . All of the infected monkeys were parasitologically positive for T . cruzi at 16–19 ypi [14] . At 20 ypi , however , attempts to identify T . cruzi kDNA in the blood and heart of monkey #99 were consistently negative ( Figure 2C , Figure 2D ) . Considering that low parasitism could contribute to these results , we performed additional studies using two or three tissue samples per analyzed monkey . Real time qPCR for the conserved repetitive nuclear satellite DNA sequences revealed low parasitism in heart samples of all T . cruzi-infected rhesus monkeys except monkey #103 . Conversely , real time qPCR revealed low parasitism in spleen samples of all T . cruzi-infected monkeys ( Figure 2E , Figure 2F ) . Heart and spleen samples of noninfected controls were repeatedly negative ( Figure 2E ) . The β-globin sequence was detected as a control for DNA integrity and PCR inhibition in all analyzed samples ( data not shown ) . Taken together , these results support parasite persistence in indeterminate ( #42 , #64 , #68 , #99 ) and cardiopatic ( #90 , #95 , #103 ) chronically T . cruzi-infected rhesus monkeys . Because antibody detection is a criterion for persistent infection [20] , we analyzed the kinetics of the anti-T . cruzi antibody in a group of infected monkeys that were cardiopatic ( #90 , #95 and #103 ) and a monkey that was indeterminate ( #99 ) at 20 ypi . Figure 2G shows that all monkeys were seronegative prior to infection . The anti-T . cruzi antibody was only detected in one monkey ( #103 ) when parasitemia was positive ( 18 dpi ) . When parasitemia was controlled ( 59 dpi ) [15] all of the infected monkeys presented anti-T . cruzi antibodies . At the end-point of our study , anti-T . cruzi antibodies were detected in cardiopatic monkeys ( #90 , #95 , #103 ) , but not in indeterminate monkey #99 . Importantly , there were no alterations in the levels of total IgM ( Figure S3A ) and IgG ( Figure S3B ) in serum of T . cruzi-infected monkeys in comparison with noninfected controls at the end-point of this study ( 20–23 ypi ) . Because parasite antigens were scarce in the cardiac tissues of monkeys with the indeterminate and cardiac forms of Chagas disease , we evaluated the cardiac tissue for the expression of iNOS/NOS2 , which is an enzyme potentially involved in an important parasite control pathway [9] , [11] . Analysis of serial heart tissue sections revealed that most of the iNOS/NOS2+ cells infiltrating the myocardial interstitium of infected animals ( Figure 3A–D ) were CD68+ macrophages ( data not shown ) . There was a significant increase in the number of iNOS/NOS2+ cells in the myocardium of monkey #95 that presented high parasitism detected by qPCR for genomic DNA ( Figure 3A–D , Figure 3E ) , whereas iNOS/NOS2+ cells were scarce or absent in noninfected animals ( Figure 3E ) . In chronic Chagas disease , severity of CCC correlated with high NO levels in serum [6] . Because an increased number of iNOS/NOS2+ cells in the cardiac tissue of infected monkeys was related to the severity of electrical abnormalities , we decided to study NO levels in the serum of infected monkeys . A high NO concentration was only detected in the monkey ( #95 ) with severe CCC ( Figure 3F ) . In this monkey , parasite control was achieved at 2 months postinfection , at which point antibodies were detected ( Figure 2G ) , in absence of significant NO concentration in serum ( Figure 3G ) . A high NO concentration was only detected after 6 months of infection , and the NO level was persistently elevated at 16 and 20 ypi ( Figure 3G ) . In the case of noninfected monkeys #81 and #94 the low numbers of iNOS/NO2+ cells in the cardiac tissue was parallel to low NO levels in serum ( Figure 3E , Figure 3F ) . In the cardiopatic T . cruzi-infected monkey #95 the high number of iNOS/NO2+ cells in the cardiac tissue paralleled the high NO concentration in serum ( Figure 3E , Figure 3F ) . However , there was no significant correlation ( r2 = 0 . 5206 , p = 0 . 105 ) between the number of iNOS/NO2+ cells in the cardiac tissue and NO concentration in serum of T . cruzi-infected rhesus monkeys . Overexpression of iNOS/NOS2 and NO has been shown to be associated with heart injury in noninfectious conditions [8] . To test the possible consequences of iNOS/NOS2 overexpression in the heart tissue of T . cruzi-infected monkeys , we analyzed the expression of Cx43 . Connexin 43 is the major gap junction protein in the heart , and Cx43 is primarily responsible for the electrical synchrony of cardiomyocytes [21] . Connexin 43 distribution was homogeneous in the intercalated disks of cardiomyocytes of noninfected controls ( Figure 4A ) . Further , the Cx43 pattern and stained area in the indeterminate T . cruzi-infected monkeys 64 ( Figure 4B ) and #99 ( Figure 4C ) were preserved , and the Cx43 staining resembled the noninfected controls ( Figure 4G ) . Conversely , Cx43 was detected as disorganized patches scattered throughout the tissue of cardiopatic infected monkeys #90 ( Figure 4D ) , #95 ( Figure 4E , Figure 4F ) and #103 ( data not shown ) . In addition , Cx43 loss was associated with inflammatory infiltrates ( Figure 4E , Figure 4F ) . Although not considered for the quantitative study , in cardiac areas where myocardial cells were substituted by mesenchymal cells Cx43 loss was , obviously , more pronounced ( Figure 4F ) . Monkey #95 , which showed the more severe form of CCC and a high number of iNOS/NOS2+ cells in the heart , also exhibited a marked Cx43 loss ( Figure 4E , Figure 4F , Figure 4G ) . Interestingly , when the percentage of Cx43-stained area in the cardiac tissue was evaluated considering the clinical score , significant Cx43 loss was detected in cardiopatic T . cruzi-infected monkeys ( 0 . 92±0 . 12% of stained area ) when compared with indeterminate ( 2 . 25±0 . 31% of stained area; p<0 . 05 ) and noninfected ( 2 . 06±0 . 18% of stained area; p<0 . 05 ) rhesus monkeys . Increased CK-MB activity levels in serum , which is a marker of myocardial cell injury [22] , was detected in cardiopatic monkeys #95 and #103 ( Figure 4H ) . In the infected monkeys with normal ECG patterns ( #64 and #99 ) , the CK-MB activity levels resembled the levels in the noninfected controls ( #81 and #94 ) . A positive correlation was seen between the Cx43 loss in the cardiac tissue and the CK-MB activity levels in serum of the studied rhesus monkeys ( r2 = 0 . 6355; p<0 . 05 ) . Importantly , there was a positive correlation between the number of iNOS/NOS2+ cells in the heart tissue and the CK-MB activity level in the serum ( r2 = 0 . 8263; p<0 . 05 ) of chronically T . cruzi-infected rhesus monkeys ( Figure 4I ) . To add insights on the participation of iNOS/NOS2-derived NO in T . cruzi-elicited cardiomyopathy , we studied the effect of the infection of iNOS/NOS2-deficient ( Nos2tm1Lau/J ) mice in different aspects of heart injury . After 40 days of infection , in comparison with noninfected controls wild-type C57BL/6 mice present increased NO levels in serum ( Figure 5A ) , which are not detected in Nos2−/− mice ( 4 . 4±0 . 17 µM in noninfected vs . 4 . 3±2 . 1 µM in T . cruzi-infected ) . In addition , there was an increased number of iNOS/NOS2+ cells in the cardiac tissue of T . cruzi-infected C57BL/6 mice ( Figure 5B ) . Interestingly , a significant increase ( p<0 . 05 ) in the number of parasite nests was seen in the cardiac tissue of Nos2−/− mice ( Figure 5C ) . Further , a similar intensity of myocardial inflammation ( Figure 5D ) and frequency of ICAM-1+ and VCAM-1+ blood vessels ( data not shown ) were detected in wild-type and Nos2−/− mice . Representative ECG registers at 40 dpi are shown in the Figure S4 . Absence of iNOS/NOS2 led to sinus bradyarrhythmia in noninfected controls compared with noninfected wild-type C57BL/6 mice . Further , T . cruzi-infected C57BL/6 and Nos2−/−presented increased PR interval , resulting in first-degree AV block ( Table 2 ) . Importantly , T . cruzi-infected mice showed concomitantly two types of arrhythmia , sinus bradycardia and AV block . Significant increase in QTc and higher QRS scores were detected in T . cruzi-infected C57BL/6 mice compared with noninfected mice ( Table 2 ) . Interestingly , changes in QRS scores were significantly decreased in infected Nos2−/− mice compared with infected C57BL/6 mice ( Table 2 , p<0 . 001 ) . Altogether , T . cruzi infection resulted in more frequent first- and second- degree atrioventricular block type 2∶1 in infected C57BL/6 mice compared with infected Nos2−/− mice ( Table 2 ) . Considering the importance of myocardial cell integrity for electrical conduction and the participation of Cx43 in this process ( 21 ) , we analyzed the CK-MB levels in serum and Cx43expression in the cardiac fibers of iNOS/NOS2 deficient T . cruzi-infected mice . Importantly , when compared with infected C57BL/6 mice , Nos2−/− mice had a significant decrease in CK-MB activity levels in serum ( Figure 5E ) , although myocardial cell lesion was not completely abrogated . Additionally , absence of iNOS/NOS2 resulted in inhibition of Cx43 loss in the cardiac tissue of T . cruzi-infected mice ( Figure 5F ) . Taken together , these data reveal decreased cardiomyocyte lesion and more preserved heart conduction and ventricle polarization in T . cruzi-infected mice with iNOS/NOS2 deficiency . In the present study , the infection of nonhuman primate rhesus monkeys with the Colombian T . cruzi strain reproduced clinical and histopathological aspects of the chronic indeterminate and cardiac forms of Chagas disease , including parasite persistence in the cardiac tissue . Moreover , overexpression of iNOS/NOS2 cells in the cardiac tissue and systemic high NO levels were directly related to cardiomyocyte lesion and heart injury , including electrical abnormalities . Studies in Nos2−/− mice corroborated the participation of iNOS/NOS2 in parasite dissemination control and , moreover , revealed the participation of the iNOS/NOS2-derived NO in disturbance of the major molecular pathway involved in electrical synchrony in T . cruzi infection . The reproduction of several features of CCC , which is the main clinical manifestation of Chagas disease , in T . cruzi-infected rhesus monkeys allows the adoption of this model to test vaccines and new trypanocide chemotherapies . In addition , T . cruzi-infected rhesus monkeys can be used to investigate mechanisms of CCC physiopathology [1] . Although the outbred nature and small number of monkeys can be viewed as a limitation of the present studies , these animals reproduced the clinical spectrum ( indeterminate and cardiac form ) seen in Chagas disease [2] , [19] . Moreover , the chronically infected monkeys reproduced the most typical chagasic electrical conduction abnormalities [2] . Although no increase in the cardiothoracic index was noticed in infected monkeys , asynchronic interventricular septum motility was detected in monkey #95 , which showed a more severe stage of CCC . Electrical conduction alterations were associated with the intensity of myocardial fibrosis , which was spatially related to inflammation in an apparent lack of parasite and resembled the features of Chagas disease [2] , [19] . Myocardial fibrosis was directly related to the 12-lead ECG QRS scoring and the severity of Chagas' heart disease [23] . Fibrosis , which is one of the most important features of CCC [19] , shows a progressive evolution in T . cruzi-infected mice [24] . Fibrosis is directed by inflammatory processes that provoke chemokine-driven accumulation of mesenchymal cells [25] . In the present studies , collagen deposition occurred early in the acute infection ( 41–76 dpi ) concomitantly with inflammation . Interestingly , inflammation resolution and collagen degradation can occur without specific treatment . This finding was seen in monkey #45 , which presented the indeterminate form of Chagas disease at 3 ypi . Monkey #99 did not exhibit cardiac inflammation or fibrosis; however , we could not prove whether fibrosis was established and later remodeled . The predominance of type III , pro-III , and pro-IV collagens in the heart tissue of individuals with chronic T . cruzi infection [24] may favor reversibility of fibrosis because these collagens have a high turnover . In T . cruzi infection , extracellular matrix deposition in the cardiac tissue can be remodeled if inflammation subsides , which has been observed after etiological chemotherapy with benznidazole in the chronic infection [24] and after modulation of inflammation with a partial CC-chemokine receptor antagonist in the acute and chronic infection [26] , [27] . These data suggest parasites and inflammation as triggers of fibrogenic factors in CCC , which reinforces the idea that therapeutic strategies targeting parasites and inflammation , possibly combined , may be beneficial in remodeling fibrosis and restoring heart function in Chagas disease . Observations of chronic myocarditis and fibrosis in the apparent absence of T . cruzi have suggested that autoimmunity is a central mechanism for CCC pathogenesis [3] . In a classic study , however , amastigote forms of T . cruzi were detected inside myocardial cells in all analyzed CCC patients [19] . Furthermore , a PCR signal for T . cruzi kDNA has been observed in the hearts of CCC patients , but it was not detected in seropositive nonCCC patients [28] . In indeterminate and cardiopatic T . cruzi-infected monkeys , parasites ( antigen , kDNA and nuclear satellite DNA ) persisted in the heart and spleen . Although in some tissue samples parasite kDNA was not detected in repeated analyses , the study of different fragments of heart and spleen and the use of more sensitive assay resulted in detection of parasite persistence in all chronically T . cruzi-infected rhesus monkeys . Considering the short life span of parasite DNA in host tissues [29] , PCR signals in the heart and spleen of infected monkeys constitute real proof of T . cruzi persistence . The presence of anti-T . cruzi antibodies in cardiopatic monkeys coincided with parasite persistence in the blood , heart and spleen . In monkey #99 , however , anti-T . cruzi antibodies were restricted to the acute infection and coincided with parasite control . Detection of a specific immune response in chronically infected individuals reflects continuous antigenic stimulus by persistent parasites . The consistent absence of detection of specific antibodies [15] , T . cruzi kDNA in peripheral blood and heart in monkey #99 , which had no electrical abnormalities , cardiac inflammation or fibrosis at 20 ypi in the absence of treatment , led us to consider the possibility of spontaneous cure in this monkey [19] . However , the analysis of different fragments of spleen and heart revealed the presence of T . cruzi genomic DNA in both tissues of monkey #99 , supporting the persistence of parasite in these tissues in this monkey with the indeterminate form of Chagas disease . Therefore , the low parasitism restricted to focal areas in different organs may explain the difficulties to reveal parasite and antibodies presence in this monkey . Although there was no correlation between the quantity of antigens and the intensity of myocarditis , the detection of T . cruzi DNA in the cardiac tissue of infected monkeys supports the idea that persisting parasites trigger detrimental inflammation that can act on cardiomyocytes . Heart inflammation is a major factor that contributes to an increased risk of death in CCC compared with other heart conditions [30] . The components of inflammatory infiltrates contributing to this picture remain unclear . In noninfectious human diseases and murine models of cardiac pathologies , iNOS/NOS2 and NO were shown to be both protective and detrimental for heart physiology , and these effects may be dependent on NO concentration [8] , [31] . Therefore , the increased number of iNOS/NOS2+ cells in the heart tissue of T . cruzi-infected cardiopatic monkeys led us to search for cardiomyocyte lesion . Connexin 43 was severely depleted in cardiopatic monkey #95 , which presented a high number of iNOS/NOS2+ cells infiltrating the cardiac tissue . T . cruzi infection of human cardiomyocytes in vitro [32] and mice in vivo [27] , [33] has also been shown to cause Cx43 loss . This effect was hampered in infected TNFR1-deficient T . cruzi-infected mice , which supports that TNF/TNFR1 signaling is involved in Cx43 loss [33] . Tumor necrosis factor , particularly associated with IFNγ , is an iNOS/NOS2 inducer that results in NO production [7] , [8] . In this context , we detected IFNγ in the serum of monkey #95 ( 150 pg/mL ) and TNF in the serum of monkey #103 ( 28 pg/mL ) and high IFNγ and TNF levels in PMA-stimulated peripheral blood cells of monkeys #103 and #95 , but not in noninfected controls or infected monkeys with the indeterminate form of Chagas disease ( our unpublished data ) . Although consistent with the idea that the cardiopatic monkeys #103 and #95 overproduce inflammatory cytokines that may stimulate NO production , the low number of responsive animals hampered any definitive conclusion . Thus , the mechanism by which high iNOS/NOS2 is induced in cardiac tissue following T . cruzi infection remains unknown . The present study was the first to show that in CCC the numbers of iNOS/NOS2+ cells in cardiac tissue are associated with Cx43 loss and , more clearly , with increased CK-MB activity levels in serum , markers of cardiomyocyte injury [21] , [22] . Interestingly , iNOS/NOS2 upregulation in an inflammatory milieu results in increased NO levels in the cardiac tissue in autoimmune myocarditis [34] . In addition , NO is induced in macrophages and cardiomyocytes by T . cruzi and inflammatory cytokines and chemokines , which require the iNOS/L-arginine pathway [35] . The severity of CCC is associated with high NO levels in chagasic patients [6] . Nitric oxide has also been shown to be involved in heart denervation in acute T . cruzi infection in rats [31] . Therefore , persistent T . cruzi in the cardiac tissue might sustain continuous iNOS/NOS2 expression and a large supply of NO in a tissue that normally experiences low and tightly controlled levels of this mediator [7] , [36] . In consequence , the increased expression of iNOS/NOS2 and supply of NO could lead to cardiomyocyte lesion and heart injury . To test this idea , we adopted a Nos2−/− murine model . Initially , T . cruzi infection of C57BL/6 mice ( the Nos2−/− genetic background ) led to increased systemic NO production and , particularly , iNOS/NOS2+ cells in the cardiac tissue , which reproduced our findings in chronically infected monkeys . In addition , iNOS/NOS2+ cells have previously been detected in the cardiac tissue of acutely infected mice [9] and dogs [37] . Therefore , the present data show that independent of the host and the phase of infection , T . cruzi infection enhances NO in serum and iNOS/NOS2+ cells in the cardiac tissue . The present data also refute the result showing that iNOS/NOS2 is not required for T . cruzi control [12] , but support previous findings that iNOS/NOS2 is essential for T . cruzi control in the cardiac tissue [9] , [11] . Paradoxically , NO might be detrimental in T . cruzi infection because it depresses lymphocyte functions , which could promote parasite survival [7] . Although this idea demands further experimental support , if this is the case the detrimental effect of NO on the immune response may explain parasite persistence in chronically infected individuals , including patients [6] and our experimental models . Interestingly , iNOS/NOS2 absence ( in Nos2−/− ) abolished NO production in T . cruzi-infected mice , which showed that iNOS/NOS2 is essential for NO overproduction . In addition , this result corroborated studies showing that there is no compensation mechanism increasing other NOS isoforms in Chagas disease [12] . Furthermore , iNOS/NOS2 and locally produced NO are not involved in myocarditis formation , which suggests that the nature rather than the intensity of heart inflammation is determinant of the Chagas' heart disease outcome [27] . Furthermore , T . cruzi infection of Nos2−/− mice demonstrated a role for iNOS/NOS2 in myocardial cell lesion and connectivity loss , which supported our findings in chronically infected monkeys . Interestingly , recent proposal brings support for Chagas disease to be considered a junctionopathy [38] . Therefore , iNOS/NOS2-derived NO may be a direct or indirect critical trigger of the molecular pathway leading to myocardial cell connectivity loss . Further , T . cruzi may directly lead to myocardial cell lesion , as revealed by increased CK-MB activity levels in serum and Cx43 loss , in infected mice . In this context , absence of iNOS/NOS2 led to significant decrease in CK-MB activity levels in serum , placing iNOS/NOS2-derived NO as important myocardiotoxic agent , but did not completely abolished myocardial cell lesion that persisted in presence of high cardiac tissue parasitism in acutely T . cruzi-infected Nos2−/− mice . In chronically infected rhesus monkeys , focal persistence of T . cruzi ( revealed by detection of antigen+ spots and of low amounts of parasite DNA ) may contribute to maintain iNOS/NOS2 induction and local NO production leading to myocardial cell lesion . A previous study showed that iNOS/NOS2-derived NO was associated with right ventricular dilation and systolic dysfunction in acute murine T . cruzi-elicited myocarditis [9] . Although NOS/NOS2 deficiency led to bradyarrhythmia in noninfected controls , T . cruzi-infected Nos2−/− mice presented lower frequency of AVB1 and AVB2 than T . cruzi-infected C57BL/6 mice . More importantly , the higher QRS scores detected in CCC patients [2] , 23 , which were reproduced in infected monkeys and C57BL/6 mice , were significantly decreased in T . cruzi-infected Nos2−/− mice , which implicated iNOS/NOS2 and NO in QRS score increases independent of the host . Therefore , the present data support that the iNOS/NOS2-NO pathway participates in T . cruzi-induced myocardial cell lesion and heart injury and suggest that this pathway should be explored as a therapeutic target in CCC . Considering observations in chronic Chagas disease [6] , it is a plausible proposal as the pivotal role of NO in T . cruzi control was restricted to acute infection [39] . Further studies are needed to determine whether inhibition of iNOS/NOS2 will be therapeutically useful in chronic Chagas disease , a condition of vast overproduction of NO .	Chagas disease , a neglected tropical disease caused by the protozoan Trypanosoma cruzi , afflicts from 8 to 15 million people in the Latin America . Chronic chagasic cardiomyopathy ( CCC ) is the most frequent manifestation of Chagas disease . Currently , patient management only mitigates CCC symptoms . The pathogenic factors leading to CCC remain unknown; therefore their comprehension may contribute to develop more efficient therapies . In patients , high nitric oxide ( NO ) levels have been associated with CCC severity . In T . cruzi-infected mice , NO , mainly produced via inducible nitric oxide synthase ( iNOS/NOS2 ) , is proposed to work in parasite control . However , the participation of iNOS/NOS2 and NO in T . cruzi control and heart injury has been questioned . Here , infected rhesus monkeys and iNOS/NOS2-deficient mice were used to explore the participation of iNOS/NOS2-derived NO in heart injury in T . cruzi infection . Chronically infected monkeys presented electrical abnormalities , myocarditis and fibrosis , resembling the spectrum of human CCC . Moreover , cardiomyocyte lesion correlated with iNOS/NOS2+ cells infiltrating the cardiac tissue . Our findings support that parasite-driven iNOS/NOS2+ cells accumulation in the cardiac tissue and NO overproduction contribute to cardiomyopathy severity , mainly disturbing the pathway involved in electrical synchrony in T . cruzi infection .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "immunopathology", "immunology", "biology" ]	2012	Inducible Nitric Oxide Synthase in Heart Tissue and Nitric Oxide in Serum of Trypanosoma cruzi-Infected Rhesus Monkeys: Association with Heart Injury
To escape CD8+ T-cell immunity , human cytomegalovirus ( HCMV ) US11 redirects MHC-I for rapid ER-associated proteolytic degradation ( ERAD ) . In humans , classical MHC-I molecules are encoded by the highly polymorphic HLA-A , -B and -C gene loci . While HLA-C resists US11 degradation , the specificity for HLA-A and HLA-B products has not been systematically studied . In this study we analyzed the MHC-I peptide ligands in HCMV-infected cells . A US11-dependent loss of HLA-A ligands was observed , but not of HLA-B . We revealed a general ability of HLA-B to assemble with β2m and exit from the ER in the presence of US11 . Surprisingly , a low-complexity region between the signal peptide sequence and the Ig-like domain of US11 , was necessary to form a stable interaction with assembled MHC-I and , moreover , this region was also responsible for changing the pool of HLA-B ligands . Our data suggest a two-pronged strategy by US11 to escape CD8+ T-cell immunity , firstly , by degrading HLA-A molecules , and secondly , by manipulating the HLA-B ligandome . Human cytomegalovirus ( HCMV ) represents a prototypic β-herpesvirus persisting throughout life in its host with periodic phases of latency and reactivation of productive infection . Despite rare cases of clinical disease in healthy individuals , HCMV has a permanent impact on immune cells , e . g . resulting in the extraordinary expansion of both CD8+ memory T-cells and memory-like NK cells [1 , 2] . As demonstrated in CMV animal models , protection from CMV disease is strongly dependent on MHC class I ( MHC-I ) restricted CD8+ T-cell responses [3] . MHC-I molecules are dimers formed by the membrane attached heavy chain ( HC ) and the soluble beta-2-microglobulin ( β2m ) . Upon assembly in the ER the heterodimeric MHC-I molecule is recruited to the peptide loading complex ( PLC ) , composed of the transporter associated with antigen processing ( TAP ) and the chaperones tapasin , ERp57 and calreticulin [4 , 5] . In the course of loading an optimal peptide , the trimolecular MHC-I complex is released for transport through the secretory pathway and surface expression . The US6 gene family region of HCMV encodes for several immunoevasins that target the MHC-I antigen presentation pathway at different stages of the protracted HCMV replication cycle [6] . US3 blocks maturation of MHC-I and interferes with tapasin function , US2 and US11 target MHC-I for rapid proteasomal degradation , and US6 inhibits peptide translocation by TAP [7–13] . Of the reported HCMV encoded MHC-I inhibitors , so far a crystal structure exists only for US2 in complex with HLA-A02:01 [14] . Similar to US2 , US11 binds to MHC-I with its ectodomain , but the contact site on MHC-I has not been defined [15 , 16] . The lack of insight into the contacts sites is mirrored by the poor understanding of the HLA allotype specificity of US11 . Whereas varying effects of US11 on HLA-B and -C allotypes have been reported [17–20] , consistent downregulation of HLA-A allotypes was observed [19–21] . HLA-A02:01 has been an instrumental substrate to elucidate ER associated degradation ( ERAD ) pathways activated by US11 . Upon binding to MHC-I , a glutamine in the transmembrane segment of US11 recruits Derlin-1 [15 , 22] . Derlin-1 is a crucial component of an ERAD pathway also including the recently identified E3 ligase TMEM129 and the E2 ligase Ube2J2 , required for ubiquitination and subsequent degradation of MHC-I [23 , 24] . Whereas US11 itself is not degraded by this pathway , low MHC-I expression exposes US11 to an alternative degradation pathway including HRD1 and SEL1L [24 , 25] . The genes of the US6 family most likely evolved in a cytomegalovirus ancestor prior to the split of Old World monkeys and hominoids , as homologs of these genes are found also in chimpanzee and rhesus CMV ( rhCMV ) [26 , 27] . In the context of a rhCMV vaccine vector [28] , in addition to degradation of MHC-I , it was observed that the rhCMV US11 homolog Rh189 is able to suppress the CD8+ T-cell responses to canonical epitopes [29] . Although the data are not formally published , Früh and Picker mention in a recent overview publication that substitution of Rh189 with HCMV US11 in a rhCMV mutant maintains the ability of the virus to block canonical CD8+ T-cell responses [28] , suggesting that in addition to MHC-I degradation , US11 proteins execute further manipulation of MHC-I antigen presentation . When studying the MHC-I ligandome in HCMV-infected cells , we noticed a striking differential effect of US11 on HLA class I locus products . Ligands from HLA-B and -C molecules remained largely unaltered in the presence of US11 , while HLA-A ligands were efficiently eliminated . Further investigations reported here revealed that US11 targets HLA-B by manipulating the quality of the peptide ligands . This illustrates the specific role of a single MHC-I immunoevasin that not randomly degrades MHC-I molecules , but has evolved to exert HLA locus-specific functions . To gain deeper insight into the identity and amount of HCMV peptides presented by MHC-I we undertook MHC-I ligandome analysis . MRC-5 fibroblasts were infected with HCMV AD169VarL strain derived mutants lacking either the MHC-I inhibitor region US2-US6 ( ΔUS2-6 ) or US2-US6 plus US11 ( ΔUS2-6/US11 ) , leaving the US7-US10 region intact as demonstrated by the expression of the neighboring gene US10 in ΔUS2-6/US11 infected cells ( S1 Fig ) . At 48 h post-infection cells were harvested and MHC-I complexes were isolated by the pan-MHC-I-reactive monoclonal antibody W6/32 . Identification and relative quantification of HLA ligands was performed by LC-MS/MS . Much to our surprise , a clear difference in the ability of US11 to target HLA-A and HLA-B molecules was observed . About 90% of all peptide ligands significantly down-regulated by US11 ( ca 22% of total pool; Fig 1A , blue dots ) was derived from HLA-A02:01 and A29:02 molecules ( Fig 1B , left panel ) . Unexpectedly , some peptides ( ca 11% of the total pool ) were found to be increased in cells infected with the ΔUS2-6 mutant ( Fig 1A , red dots ) . The majority ( ca 80% ) of these peptides could be attributed to HLA-B44:02 ( Fig 1B , right panel ) . No major changes were observed for the ligandome of HLA-B07:02 . To reassure that our observations were not distorted by ligands from uninfected cells possibly present in the culture , we analyzed the anchor residues of identified HCMV-derived peptides to predict their HLA-I specificities . The distribution of these viral peptides showed the same HLA-I pattern as the total ligandome , but more pronounced; US11 almost completely abolished viral ligands of HLA-A , but not of HLA-B . Again , HLA-B44:02 ligands increased in the presence of US11 ( S2B Fig ) . Since the protocol for ligandome analysis does not differentiate between surface and intracellular MHC-I molecules we next conducted flow cytometry analysis of MRC-5 fibroblasts mock treated or infected with the ΔUS2-6 or ΔUS2-6/US11 deletion viruses , to determine the surface levels of HLA-A and HLA-B using available specific antibodies for HLA-A02:01 , B07:02 and B44:02 . Whereas less than two-fold reduction for HLA-B surface expression was observed ( Fig 1C and 1D ) , HLA-A02:01 was reduced 3-fold in the presence of US11 as compared to infection with the virus lacking US11 . Therefore , also on the surface on HCMV-infected fibroblasts , a stronger effect of US11 was measured for the HLA-A allotype A02:01 than for the HLA-B allotypes B07:02 and B44:02 . Stronger regulation of HLA-A02 was also observed on ARPE19 epithelial cells infected with the TB40 derived ΔUS2-6 BAC virus . Compared to mock-treated cells HLA-A02 was downregulated 5-fold and HLA-B07 1 . 5-fold ( S1E Fig ) , indicating that different regulation of HLA-A02:01 and HLA-B07:02 by US11 is not a fibroblast adapted function of HCMV . We wondered whether the strong resistance of HLA-B07:02 would also hold true in the context of non-infected cells and therefore N-terminally HA-tagged ( HA-tagged molecules are indicated with ~ in all following figures ) HLA-A02:01 , A03:01 , B07:02 and CD99 were cloned into a lentiviral vector . HeLa cells were first transduced and selected to express the MHC-I molecules or the control protein CD99 and subsequently transduced with a lentivirus encoding US11 in front of an IRES and EGFP sequence . Cell surface expression of HA-tagged MHC-I and CD99 was analyzed on EGFP positive cells in comparison to EGFP negative cells at 0 , 24 and 48 h after transduction . Remarkably , in this context , similarly to the control protein CD99 , HLA-B07:02 appeared completely unaffected by US11 , but not HLA-A02:01 and A03:01 ( Fig 1E ) , demonstrating that resistance of HLA-B07:02 against US11 is independent of other viral or virally induced proteins . In infected cells , US11 may work in concert with other proteins and could explain the stronger effect of US11 on HLA-B in infected MRC-5 cells . HCMV also reorganizes the secretory pathway in infected cells [30] , which could further influence trafficking and surface expression of MHC-I allotypes in the presence of US11 . To gain insight into the breadth of the resistance of HLA-B locus products against US11 , we cloned various HLA-A and -B sequences with an N-terminal HA-tag und analyzed the cell surface level in the absence and presence of US11 after transient transfection of HeLa cells . Usage of the CMV major IE promoter for MHC-I expression leads to only low level of MHC-I regulation ( probably due to overload of the ER and insufficient levels of ERAD-associated proteins required for US11 function ) and we therefore proceeded with a different vector harboring a less potent promoter ( spleen focus-forming virus ( SFFV ) U3 promoter ) to be able to measure an adequate level of US11-mediated downregulation . Using this approach , all HLA-A allotypes were strongly downregulated by US11 ( Fig 2A and 2B ) , while no significant reduction of HLA-B cell surface expression was noted . However , HLA-B44:02 , from which more diverse peptide ligands were eluted in the presence of US11 in infected cells ( Fig 1B , right panel ) , displayed an induced level of surface expression in US11-expressing cells . Altogether , this suggests that the observed resistance towards US11-mediated downregulation is a general feature of HLA-B locus products . Next , pulse-chase experiments were conducted to measure the stability of MHC-I molecules in the absence and presence of US11 . As expected , a clear destabilization of HLA-A02:01 and A03:01 could be observed . Surprisingly , also the stability of HLA-B HCs was strongly reduced in the presence of US11 ( Fig 2C and S4 Fig , using shorter labeling and chase times ) . However , in contrast to HLA-A , HLA-B allotypes were able to dimerize with β2m in the presence of US11 and maturation of these molecules could be observed as a slightly slower migrating HC band ( likely due to glycan modifications ) at 45 min of chase ( Fig 2C ) . This observation suggests , that different from HLA-A , HLA-B molecules are able to exit the ER and accumulate at the cell surface in the presence of US11 . We therefore analyzed the steady-state level of EndoH ( Endoglycosidase H; digests Asn-linked glycans that have not been subjected to modification in the Golgi apparatus ) resistant molecules by Western blot and indeed observed that although the total level of HLA-B was reduced in the presence of US11 , the EndoH resistant molecules were largely unchanged compared to control cells . This was not the case for HLA-A , as a complete loss of EndoH resistant molecules was observed ( Fig 2D ) . We asked whether HLA-B surface expression in the presence of US11 could be more readily recognized by an inhibitory MHC-I receptor such as LIR1 . Indeed , LIR1-Fc incubated with HeLa cells treated as described above showed a clear binding to HLA-B07:02 expressing cells despite co-transfection of US11 , whereas this was not the case for HLA02:01 and -A03:01 expressing cells ( Fig 2E ) . In conclusion , HLA-A and also HLA-B molecules are targets for US11-mediated degradation . However , different from HLA-A , a fraction of the HLA-B molecules can escape degradation and be expressed on the cell surface . In the pulse-chase experiment we observed an apparent co-immunoprecipitation of US11 with all studied types of MHC-I molecules ( S3 and S4 Figs ) . This suggested that although HLA-B molecules escape downregulation by US11 , they are still bound by US11 efficiently . To depict this more clearly , we took advantage of the fact that MHC-I molecules expressed transiently from a strong promoter ( CMV IE promoter ) remain largely stable in the presence of US11 . Under these conditions we assessed binding of US11 to HLA-A02:01 , A03:01 , B07:02 and B15:03 ( Fig 3A ) . With US11 and MHC-I being strongly overexpressed at saturating levels , US11 bound similarly to all MHC-I molecules . We next assessed the possibility that the cytosolic tail of HLA-A allotypes , which is three residues ( CysLysVal ) longer than that of HLA-B allotypes , could be decisive for differential US11 regulation . The C-terminal Val has previously been described to be important for US11-mediated degradation [31] . To this end , we constructed an HLA-A03:01 mutant without these residues ( A3ΔCKV ) and compared it to the reciprocal HLA-B07:02 mutant ( B7+CKV ) . The residues CysLysVal had only small and non-significant effects on surface expression in US11 expressing cells as measured by flow cytometry ( Fig 3B and 3C ) . Even though HLA-B07 was more efficiently downregulated , when expressed with C-terminal CysLysVal residues , downregulation was significantly different from HLA-A03:01 . Therefore , the C-terminal CysLysVal residues are not the major determinant for the difference in US11-mediated regulation . We next scrutinized the role of β2m in the process of US11-mediated degradation , since the HLA-B molecules HLA-B8 and -B5 were reported to possess a higher affinity for β2m compared to HLA-A1 and -A2 [32] . In FO-1 cells MHC-I is not expressed on the surface due to lack of β2m . Co-transfection of a plasmid encoding β2m rescued cell surface expression of MHC-I ( Fig 3D ) . Similarly to HeLa cells , in FO-1 cells transfected with a β2m encoding plasmid , HLA-A03:01 was downregulated by US11 , whereas HLA-B07:02 was not ( Fig 3D ) . Also the stability of HLA-B07:02 was higher compared to A03:01 , and this was not influenced by β2m ( Fig 3E ) , implying that the resistance of HLA-B against US11 is conferred at the stage of unassembled HC . In conclusion , although US11 binding to HLA-B is preserved , downregulation of HLA-B is much less efficient and this is only partly due to the shorter cytosolic tail of HLA-B alloforms . In our previous studies of the PLC composition in HCMV-infected cells [33] , co-immunoprecipitation experiments suggested that US11 interacts with MHC-I as part of the the PLC . This was unexpected , as the concept for MHC-I targeting by US11 has been a rapid degradation facilitated by ERAD . However , in light of our observation , that HLA-B molecules display intrinsic resistance against US11-mediated degradation , we supposed that US11 pursues two different strategies when interacting with HLA-A and HLA-B , respectively . To re-investigate our earlier findings , MRC-5 fibroblasts were treated with siRNA targeting US11 or control siRNA and subsequently infected with the ΔUS2-6 HCMV deletion mutant . At 24 h post-infection a co-immunoprecipitation experiment was performed using W6/32 or anti-ERp57 antibodies . As expected , a band corresponding to the size of US11 was found in complex with both MHC-I and ERp57 , which was not present in US11 siRNA treated cells ( Fig 4A ) . To confirm the identity of the co-immunoprecipitated protein , MRC5 cells were mock treated or infected with the HCMV deletion mutants ΔUS2-6 and ΔUS2-6/US11 and a re-immunoprecipitation was performed . After dissociation of proteins immunoprecipitated with an anti-ERp57 antibody , an anti-US11 antiserum was applied . A protein , the size of US11 , was co-immunoprecipitated from the ΔUS2-6 sample , but not from the ΔUS2-6/US11 sample ( Fig 4B ) , strongly indicating that US11 interacts with the PLC in HCMV infected cells . Therefore , our data imply , that US11 binds both unassembled MHC-I HCs and , in addition , MHC-I assembled in the PLC . Possibly , US11 binds to a structure of MHC-I that is not changed during the transition from unassembled to assembled form , e . g . in the alpha-3 domain . Alternatively , US11 interacts with MHC-I in different manners , one that can target the unassembled MHC-I for degradation and another that leads to a prolonged interaction with assembled MHC-I in the PLC . In this regard we found it interesting , that US11 has a predicted low-complexity region ( LCR; amino acids 28–42 [34] ) N-terminal of the Ig-like domain . Since LCRs tend to be engaged in protein-protein interactions [35] , we set out to investigate the importance of this domain for US11 function and interaction with MHC-I . To this end , we deleted the N-terminus ( amino acids 20–44 of the full-length protein ) with the LCR from US11 wild-type and from a US11 Gln192Ala ( US11Q/A ) mutant ( schematic illustration in Fig 5A ) . The Gln192Ala mutation prevents the recruitment of Derlin-1 and subsequent MHC-I degradation , which leads to retention of MHC-I in the ER [22] . Therefore , this modification of US11 allows for interaction studies , circumventing the issue of losing substrates due to degradation . The HA-tagged US11 versions were stably transduced into HeLa cells and the cells were first checked for steady-state level expression of US11 and MHC-I . In the analysis we included HeLa cells expressing US6 and US3 to control for strong block of MHC-I peptide loading and retention , respectively . Lysates from these cells were subjected to Western blot analysis . The total level of MHC-I was strongly reduced in both US11 and ΔLCRUS11 expressing cells , suggesting that the US11 LCR is dispensable for degradation of MHC-I ( Fig 5B ) . The expected rescue of MHC-I expression was detected in cells expressing the US11 Q192A mutants . Flow cytometry analysis revealed that surface expression of MHC-I was downregulated both by US11 and ΔLCRUS11 to the same extent as by US6 ( Fig 5C ) . Expression of the US11 Q192A mutants lead to lower level of MHC-I downregulation , also when compared to US3 expressing cells , indicating that the retention is not as strong as for US3 . To analyze interactions between US11 , MHC-I and the PLC in more detail , we next performed co-immunoprecipitation experiments using metabolically labeled cells . ( Fig 5D; immunoprecipitation using an anti-transferrin receptor control antibody is shown in S8 Fig ) . Whereas the LCR was dispensable for US11-dependent degradation of MHC-I molecules in the context of these stable cell lines ( compare MHC-I HC levels in Fig 5D , lanes 11 , 12 , and 13 ) , MHC-I ER retention caused by the Q192A mutation ( compare the level of EndoH sensitive MHC-I HC in lanes 11 , 14 and 15 ) and stabilization of MHC-I in the PLC ( compare the level of MHC-I HC in lanes 6 , 9 and 10 as well as in lanes 16 , 19 , and 20 ) was dependent on the LCR , as the mentioned effects was only observed by the full-lengh US11Q/A mutant , but not by ΔLCRUS11Q/A . This difference in function correlated with co-immunoprecipitation of the US11 mutants: US11Q/A co-immunoprecipitated with W6/32 ( lane 14 ) , anti-tapasin ( lane 9 ) and anti-ERp57 antibodies ( lane 19 ) . Most convincingly , a weak co-immunoprecipitation of wildtype US11 was observed with anti-tapasin ( lane 7 ) and anti-ERp57 antibodies ( lane 17 ) , despite very low MHC-I level in these samples , while no co-immunoprecipitation of ΔLCRUS11Q/A was observed by any of the PLC or MHC-I reactive antibodies . In contrast , ΔLCRUS11Q/A could be detected in association with unassembled HCs when using the mAb HC10 for immunoprecipitation , which predominantly binds to free HCs ( Fig 5E , lane 9; of note , HC10 also immunoprecipitates HLA-A68:02 HC [36 , 37] ) . This is consistent with the finding that US11 lacking the LCR is still able to mediate degradation of MHC-I HCs . Moreover , we observed that the affinity of US11 for fully assembled MHC-I , which are recognized by the mAb W6/32 , was strongly reduced ( compare US11 co-immunoprecipitation in Fig 5E , lanes 5 and 8; long exposure in S9 Fig ) , and this was even more pronounced for the ΔLCRUS11 mutant , demonstrating that US11 interacts with unassembled and assembled MHC-I in different manners . In conclusion , in stably transduced cells US11 interacts with unassembled MHC-I HCs and redirect them for degradation independently of the N-terminal LCR . If , however , the recruitment of ERAD is prohibited by the US11 Q192A mutation , US11 remains in a complex with assembled MHC-I molecules that are bound to the PLC and retained in the ER . This ability of US11 is dependent on the LCR , since in ΔLCRUS11Q/A expressing cells , MHC-I molecules matured as in control cells . The findings are summarized in a schematic Table in the supplementary material ( S10 Fig ) . An additional observation from this experiment that appeared contradictory , was the lack of resistant MHC-I molecules in cells stably expressing US11 . However , in contrast to low MHC-I expression levels in these HeLa cells , MHC-I is massively induced upon HCMV infection [33] . To obtain expression levels comparable to infected cells , we treated the US11-expressing cells with IFNγ . Under these conditions MHC-I was more readily detectable even in the presence of US11 ( Fig 6A , lane 4 ) . In control HeLa cells , the lower MHC-I HC band ( red asterisk ) appeared stronger after IFNγ induction than the upper band ( blue asterisk ) , whereas in US11-expressing cells the lower band was weaker than the upper one . This strongly suggests that the lower MHC-I molecule is more sensitive to degradation . To determine the identity of the MHC-I HC bands , HA-tagged versions of the single HLA-A , -B and -C molecules expressed in HeLa cells ( HLA-A68:02 , B15:03 , and C12:03 ) [38] , were expressed by transient transfection and subsequently their SDS-PAGE separation properties were determined after immunoprecipitation ( Fig 6B ) . The obtained pattern suggests that the lower US11 sensitive MHC-I HC corresponds to HLA-A68:02 . The upper band that was more resistant to US11 appeared as a smear and could comprise both HLA-B15:03 and C12:03 . In conclusion , under conditions of a high US11/MHC-I ratio , US11 selectivity is less pronounced . Elevated levels of MHC-I leads to a distinct preference of US11 for degradation of HLA-A . Our data shows that US11 binds to MHC-I HCs and redirects them for proteasomal degradation independently of the LCR . We asked what could be the purpose of the LCR in complex with assembled MHC-I . The PLC not only maintains empty or suboptimally loaded assembled MHC-I molecules in a peptide-receptive state , but also selects peptides with stabilizing properties [39–41] . To clarify the role of US11 in the PLC , we next investigated whether US11 can influence peptide selection . To this end , the MHC-I ligandome of HeLa cells overexpressing US11Q/A , ΔLCRUS11Q/A , or US3 was compared to non-transduced control cells . US3 was included as a control because it retains MHC-I in the ER and was reported to interact with the PLC [10 , 13 , 42] . The ligandome was determined by LC-MS/MS after isolation of MHC-I by the mAb W6/32 and elution of peptides [43] . The stable cell lines were analyzed in two replicates , the results of which clustered tightly . For better binding prediction , only 9-mer peptides were used for further analysis and assigned as ligands of HLA-A68:02 or B15:03 if their affinities were predicted to be <500 nM by NetMHC3 . 4 [44] . If a ligand was classified as a binder to both MHC-I molecules , it was assigned as a ligand to the one for which a higher affinity was predicted . Very similar amounts of ligands were found for HLA-A68:02 and HLA-B15:03 in control cells ( 42–45% each of all 9-mers , Table 1 and Fig 7A ) . In transduced cells ( US11Q/A , ΔLCRUS11Q/A , US3 ) , however , the percentage of HLA-B15:03 derived ligands decreased ( 28–33% of all 9-mers ) , pointing to an advantage for HLA-A68:02 expression or loading in the transduced cells . Changes in MHC-I antigen presentation upon lentiviral transduction has been observed also by others [45] . In US11Q/A expressing cells the overall amount of 9-mer ligands was strongly reduced compared to the other samples ( Table 1 ) , suggesting that US11Q/A impaired proper peptide loading . Of note , this was not the case for HeLa cells expressing ~ΔLCRUS11Q/A or US3 . To gain a more detailed view of the effect of ~US11Q/A on MHC-I peptide loading , the usage of ligand anchor residues was compared between HeLa cell lines . HLA-B15:03 prefers ligands with a lysine or glutamine at position 2 ( P2 ) and a tyrosine or phenylalanine at the C-terminal position ( P9 ) . While no change in the usage of the P9 anchor residues was observed between cells ( S11 Fig , right panel ) , the usage frequency of lysine and glutamine at P2 was inversed in US11Q/A expressing HeLa cells ( Figs 7B and S11 , right panel ) . In these cells lysine was found at P2 in 15–20% of the HLA-B15:03 ligands and glutamine in 37–39% , whereas in wild-type HeLa and in the other transduced cell lines , including the ~ΔLCRUS11Q/A expressing cells , lysine was the most common residue and glutamine was less frequently used , 33–41% and 23–28% , respectively . We did not detect changes in the ligandome of HLA-A68:02 ( S11 Fig , left panel ) . These data suggest that the LCR of US11 interferes with peptide loading of distinct MHC-I molecules . To analyze whether US11 affects the MHC-I ligandome also in HCMV infected cells , we used the data sets from Fig 1 . However , no clear changes in the HLA-B07:02 and B44:02 ligandomes were observed when we compared the cells infected with ΔUS2-6 ( US11pos ) and ΔUS2-6/US11 ( US11neg ) ( S12 Fig ) . These HLA-B allotypes are very strict in their usage of P2 anchor residues with a high frequency of proline at P2 in B07:02 peptides and glutamic acid at P2 in B44:02 peptides and might therefore be more resistant to US11-mediated peptide modification . To better visualize possible changes , the ligandomes were divided into pools of common and unique peptides in the ΔUS2-6 ( US11pos ) and ΔUS2-6/US11 ( US11neg ) samples ( S13A Fig ) . Using this setting , we found that unique HLA-B07:02 peptides in the ΔUS2-6 ( US11pos ) pool varied at P1 , but not much at P3 ( Figs 7C and S13B ) compared to the common pool . Interestingly , unique HLA-B07:02 peptides in the ΔUS2-6/US11 ( US11neg ) pool showed an opposite effect at P1 , emphasizing that the effect is US11-specific . Regarding HLA-B44:02 , only three unique peptides could be defined in the ΔUS2-6/US11 ( US11neg ) pool ( S13A Fig ) and therefore changes in the ligandome could not be strengthened by this group of peptides . However , we observed that the commonly used proline at P4 [46] was reduced by US11 and instead usage of glutamate was increased . As only a few HLA-A peptides were detected in the ΔUS2-6 ( US11pos ) unique pool ( 18 and 15 for A02:01 and A29:02 , respectively ) this data set could not be applied conclusively for analysis of peptide modification . The increased frequency of glutamate at P2 in HLA-A29:02 ligands of the ΔUS2-6 ( US11pos ) unique pool is , nonetheless , an interesting observation ( S14 Fig , right panel ) . To assess the effect of US11 on a larger panel of various MHC-I molecules , we elaborated a convenient and fast flow cytometry based assay system measuring MHC-I cell surface disposition after transient transfection of HeLa cells ( Fig 2A and 2B ) . MHC-I molecules were designed to express an inert HA-epitope tag at the N-terminus to overcome the need for allotype-specific antibodies . In this way , we measured unrestricted expression of HLA-B allotypes on the cell surface , whereas HLA-A allotypes were strongly reduced . Unexpectedly , we observed in pulse-chase experiments that US11 substantially affected HLA-B expression in the early secretory pathway ( Fig 4C ) , despite the unchanged density on the cell surface . However , at variance with HLA-A locus products , HLA-B alloforms were able to dimerize with β2m and mature in the presence of US11 . HLA-B molecules thus escaped US11 and accumulated on the surface to the same extent as in US11-negative cells . The functional expression of HLA-B07:02 in the presence of US11 could be further demonstrated by binding to LIR1 . Furthermore , we found that the level of polymorphic MHC-I synthesized in the ER strongly affects the efficacy and selectivity of US11-mediated degradation . In HCMV-infected cells transcription and biosynthesis of MHC-I is highly upregulated [33] . This could explain why US11 does not reach the critical level required to degrade HLA-B , while HLA-A is still efficiently recruited to ERAD , as we observed also in IFNγ-induced HeLa cells stably expressing US11 . This underlines the relevance of the strict regulation of US11 expresssion via the HRD1-dependent autoregulatory loop [24 , 25]; in the absence of MHC-I substrates US11 itself is targeted to ERAD degradation , controlled by HRD1 . We have begun to address the molecular basis of HLA-B resistance . The shorter cytosolic tail of HLA-B alleles confers some level of resistance , as described previously for US11 [31] and also for HIV-1 encoded Nef [56] , but was not a pivotal factor for US11 in our experimental setup . The results obtained with β2m-deficient FO-1 cells showed that β2m is not critically involved in resistance against degradation . Indeed , this demonstrated that this resistance should be an intrinsic property of HLA-B HCs , which is now an object of further investigation . US11 possesses an LCR sequence between its signal peptide and the Ig-like luminal domain . This region is 15 amino acids long ( residues 28–42 ) and contains seven proline residues . Such structurally undefined regions often function as multiprotein interaction hubs , e . g . found in chaperons [35] . Furthermore , LCRs may have advantages for faster adaptation and evolution [57] . Our analysis revealed that the LCR of US11 is required for several features of US11 that are possibly interconnected . Firstly , the ability of US11 to interact with folded heterodimers of MHC-I HC and β2m , as defined by recognition by the mAb W6/32 [58] , is dependent on the LCR . Secondly , we found that US11 interacts with the PLC most likely via binding to MHC-I , since low MHC-I expression levels resulted in low US11 co-immunoprecipitation with the PLC . Again , this interaction was dependent on the US11 LCR , confirming that only folded heterodimeric MHC-I molecules interact with the PLC [5 , 59] . US11 with a Q192A mutation is not able to forward MHC-I to the ERAD pathway . As a consequence MHC-I is retained in the ER [15 , 22] . This requires a stable interaction between US11 and assembled MHC-I heterodimers that involves the LCR of US11 , because deletion of the LCR rescued MHC-I transport through the secretory pathway in the context of the US11Q/A mutant . However , the ability of US11 to forward MHC-I HC for ERAD degradation is not affected by the deletion of the LCR . In cells stably expressing ΔLCRUS11 a strong reduction of MHC-I was observed , in accordance with the finding that ΔLCRUS11Q/A is still able to stably interact with unassembled MHC-I HCs . This indicates that US11 can initiate retrotranslocation and degradation of MHC-I without its LCR at a stage before heterodimeric MHC-I molecules assemble . The targeting of unassembled MHC-I in β2m deficient cells has been observed previously [60] . The most remarkable feature of the US11 LCR , however , was its ability to manipulate HLA-B15:03 peptide ligands . The usage of the N-terminal P2 position of the ligand anchor residue was changed in a way that the frequently appearing lysine was strongly reduced and the generally less used glutamine emerged most frequently in the presence of US11 . Control cells , ΔLCRUS11Q/A cells or cells expressing US3 , which also binds to and retains MHC-I heterodimers , did not exhibit this effect on HLA-B15:03 , indicating that it is a specific feature of the US11 LCR that interferes with peptide loading . We observed this change only for the N-terminal anchor residues and not for the C-terminal . We were not able to define any similar changes in the ligandome of HLA-A68:02 . The HLA-B molecules HLA-B07:02 and B44:02 present in MRC-5 fibroblast do not allow for measurable modifications at P2 , since the P2 residue is strongly fixed for these molecules ( by proline and glutamate , respectively ) . However , unique HLA-B07:02 and HLA-B44:02 ligands in MRC-5 cells infected with the ΔUS2-6 HCMV mutant virus expressing US11 , displayed changes in the neighboring P1 and P4 , respectively , suggesting that US11 influences the peptide selection for a broad range of HLA-B allotypes . The recently resolved structure of the PLC [5] provides a molecular basis to model manipulation of HLA-B by US11 . The MHC-I peptide binding groove is deeply buried in the PLC with the F-pocked that binds the C-terminal peptide anchor residue pointing inwards into the center of the PLC . The opposite side of the MHC-I HC is the only region of MHC-I still accessible for further protein interactions . This interface is also used by HCMV encoded US2 and adenovirus encoded E3-19K as demonstrated in resolved crystal structures [14 , 61] . If US11 also binds to this particular surface , which is likely , since US11 interacts with MHC-I during its processing through the PLC , the LCR could be in the vicinity of MHC-I residues contributing to the formation of pockets that fix the N-terminal part of the peptide . Whereas HLA-B allotypes are at the forefront in studies determining protective and sensitizing MHC-I in HIV and HCV infections [62 , 63] , such observations have not been made for HCMV or other herpes viruses . This goes well together with the suggestion that HLA-A is more important to control co-evolving DNA viruses [64] . The differential targeting of HLA-A and -B by US11 underlines this view and implies that complete block of antigen presentation by HLA-A is crucial for the virus to cope with highly specific CD8+ T-cells . Unlike HLA-A , a large fraction of HLA-B allotypes contains the Bw4 motif recognized by inhibitory KIRs ( Killer cell immunoglobulin-like receptors ) on NK cells [65] . Therefore , the costs for allowing a reduced level of HLA-B surface expression , yet , with a modified peptide repertoire , might be tolerated by HCMV , in order to dampen NK cell activation . Future studies will provide more insight into the mechanism how the US11 LCR alters the quality of MHC-I peptide ligands and the functional ramifications of this alteration . It is conceivable that this feature of US11 could confound CD8+ T-cell recognition of HCMV infected target cells . The initial priming of CD8+ T precursors is believed to occur via cross-presentation [66 , 67] , i . e . by non-infected dendritic cells in the absence of US11 . Thus , the quality of the MHC-I presented peptides might differ significantly between productively HCMV-infected cells , in which US11 is actively expressed and professional APC priming the CD8+ T-cells . Whether US11 will impact the formation of memory cells and memory inflation is not predictable . However , it will be of great interest to learn whether Rh189 ( RhCMV US11 homolog ) -induced non-canonical CD8+ T-cell restricted epitopes [29] are also dependent on the N-terminal part of the protein and could be a result of manipulation of peptide loading . Although an LCR is not predicted in the N-terminus of Rh189 , it still contains some conserved residues ( S15 Fig ) , possibly important for interaction with assembled MHC-I . MRC-5 fibroblasts ( ECACC 05090501; HLA-A02:01 , A29:02 , B07:02 , B44:02 , C05:01 , C07:02 ) , HeLa ( ATCC CCL-2; HLA-A68:02 , B15:03 , C12:03; ATCC CCL-2 ) , and US6-HA-HeLa cells [68] , ARPE-19 ( ATCC CRL-2302 ) , the melanoma cell line FO-1 [69] and HEK293T ( ATCC CRL‐11268 ) cells were grown in DMEM supplemented with 10% FCS , penicillin and streptomycin . HeLa cells were tranfected with Superfect ( Qiagen ) and FO-1 cells with Jetprime ( Polyplus Transfection ) . Small interfering RNA ( siRNA ) targeting US11 ( ACACUUGAAUCACUGCCACCCCC ) was purchased from Riboxx . Knock-down experiments were performed using Lipofectamin RNAiMax Reagent ( Invitrogen ) . The recombinant HCMV mutants ΔUS2-6/US11 and ΔUS2-US11 was generated according to a previously published procedure [70] using the BAC-cloned AD169varL genome pAD169 [71] as parental BAC . Briefly , PCR fragments was generated using the primer pair KL-DeltaUS11-Kana1 CAAAAAGTCTGGTGAGTCGTTTCCGAGCGACTCGAGATGCACTCCGCTTCAGTCTATATACCAGTGAATTCGAGCTCGGTAC and KL-DeltaUS11-Kana2 TAAGACAGCCTTACAGCTTTTGAGTCTAGACAGGGTAACAGCCTTCCCTTGTAAGACAGAGACCATGATTACGCCAAGCTCC for the ΔUS2-6/US11 mutant and the primer pair KL-DeltaUS7-Kana1 ACCTTTTGTGCATACGGTTTATATATGACCATCCACGCTTATAACGAACCTAACAGTTTACCAGTGAATTCGAGCTCGGTAC and KL-DeltaUS11-Kana2 TAAGACAGCCTTACAGCTTTTGAGTCTAGACAGGGTAACAGCCTTCCCTTGTAAGACAGAGACCATGATTACGCCAAGCTCC for the ΔUS2-US11 mutant and the plasmid pSLFRTKn [72] as template DNA . The PCR fragment containing a kanamycin resistance gene was inserted into the parental BAC by homologous recombination in E . coli . Correct mutagenesis was confirmed by Southern blot and PCR analysis . Recombinant HCMVs including TB40/E ΔUS2-6 [73] were reconstituted from HCMV BAC DNA by Superfect ( Qiagen ) transfection into permissive MRC-5 fibroblasts . Virus titers were determined by standard plaque assay . Production of lentiviruses was performed as described previously [38] . At 48 h post transfection the supernatant was collected and filtered through a 45 μm filter prior to transduction of HeLa cells by centrifugal enhancement . When selected , the cells were cultivated in normal medium for 3–4 days before treatment with 5 μg/ml puromycin ( Sigma ) . The following antibodies were applied: W6/32 ( anti-pan-HLA-A , B , C assembled with β2m and peptide , [58] ) , BB7 . 2 ( anti-HLA-A2 [74] ) , BB7 . 1 ( anti-HLA-B7 [74] ) , TT4-A20 ( anti-HLA-B44 [75] ) , HC10 and HCA2 recognizing free HLA-B/C and HLA-A heavy chains , respectively [76] , anti-CD71 ( immunotech ) , anti-CD85j ( LIR1; Miltenyi ) mouse and rabbit anti-HA antibodies ( Sigma ) , anti-ERp57 ( Millipore ) , APC-coupled anti-mouse antibodies ( BD Pharmingen ) . Polyclonal anti-tapasin and anti-US11 anti-sera were raised by immunization of rabbits ( Genscript ) with synthetic peptides ( aa 418–428 and 90–103 , respectively ) . The tapasin signal peptide sequence was amplified in front of a human influenza hemagglutinin ( HA ) –tag and cloned into XhoI and PstI of pIRES-EGFP ( Tpn-SP-pIRES-EGFP; CMV IE promoter ) . HLA-A02:01 , HLA-B07:02 , HLA-C07:02 , CD99 were amplified from cDNA prepared from MRC-5 cells and HLA-A68:02 and HLA-B15:03 from cDNA from HeLa cells . HLA-B44:02 and HLA-B44:05 were described previously [38] . The cDNA clones for HLA-A01:01 ( NM_001242758 , BC003069 ) , HLA-A03:01 ( NM_002116 ) and HLA-B08:01 ( AK292226 , BC091497 ) have been purchased from Source Bioscience , Nottingham , UK . HLA-A29:02 ( IMGT/HLA database ) was synthesized as a gBlock ( Integrated DNA Technologies , Inc . ) gene fragment . Irrespective of their source , all MHC-I sequences were used as a template for further amplification using specific primers ( Table 2 ) . PCR products were digested with PstI or NsiI and BamHI restriction enzymes and cloned into Tpn-SP-pIRES-EGFP . Sequenced inserts were subsequently subcloned into the puc2CL6IP ( pUC-IP , with spleen focus-forming virus U3 promoter ) lentiviral vector [38] using the restriction sites NheI and BamHI . HA-HLA-C12:03 was purchased in pcDNA3 . 1 from Biocat and subcloned into puc2CL6IP . US11 and US3 cDNA was amplified from AD169 HCMV DNA and cloned into pIRES-EGFP via NheI and BamH or into puc2CL6IP and puc2CL6-IRES-EGFP ( pUC-EGFP ) lentiviral vector using the same enzymes . Point mutation in US11 was inserted using the QuickChange II XL Site-Directed Mutagenesis Kit ( Agilent ) following the protocol described by the manufacturer . MHC-I ligands were isolated by standard immunoaffinity purification using the mAb W6/32 as described previously [43] . LC-MS/MS analysis of MHC-I ligand extracts using nanoflow HPLC ( RSLCnano , Thermo Fisher ) on a 50 μm × 25 cm PepMap RSLC column ( Thermo Fisher ) with a gradient ranging from 2 . 4 to 32 . 0% acetonitrile over the course of 90 min . Eluted peptides were analyzed in an online-coupled LTQ Orbitrap XL mass spectrometer ( Thermo Fisher ) using a top 5 CID ( collision-induced dissociation ) method . The procedure for label-free quantification ( LFQ ) of relative HLA ligand abundances was performed as follows: total injected peptide amounts of paired samples were normalized and LC-MS/MS analysis was performed in five technical replicates for each sample . For normalization , the relative amounts of substance in paired samples were determined by calculating the summed area of peptide identifications in dose-finding LC-MS/MS runs and the samples were adjusted accordingly by dilution . Relative quantification of HLA ligands was performed by calculating the area under the curve of the corresponding precursor extracted ion chromatograms ( XIC ) using ProteomeDiscoverer 1 . 4 ( Thermo Fisher ) . For Volcano plots , the ratios of the mean areas of the individual peptides in the five LFQ-MS runs of each sample were calculated and two-tailed t-tests implementing Benjamini-Hochberg correction were performed using an in-house R script ( v3 . 2 ) . Data processing and spectral annotation was performed as described previously [77] . In brief , the Mascot search engine ( Mascot 2 . 2 . 04; Matrix Science , London , UK ) was used to search the human proteome as comprised in the Swiss-Prot database ( 20 , 279 reviewed protein sequences , September 27th 2013 ) without enzymatic restriction . Oxidized methionine was allowed as a dynamic modification . The false discovery rate was estimated using the Percolator algorithm [78] and set to 5% . Peptide lengths were limited to 8–12 amino acids for HLA class I . Protein inference was disabled , allowing for multiple protein annotations of peptides . HLA annotation was performed using NetMHC [44] ( v3 . 4 ) , annotating peptides with IC50 scores below 500 nM as ligands of the corresponding HLA allotype . In cases of multiple possible annotations , the HLA allotype yielding the lowest IC50 score was selected . MRC-5 and ARPE-19 cells were detached with Accutase ( Sigma ) , treated with FcR blocking reagent as recommended by manufacturer ( Miltenyi ) and stained with antibodies diluted in 3% FCS/PBS . Cells were washed in 3% FCS/PBS supplemented with DAPI and fixed in 4% paraformaldehyde . Cells were analyzed by FACS Canto II ( Becton Dickinson ) . For analysis of MHC-I expression after transient transfection of HeLa cells US11 variants or a control pIRES-EGFP plasmid together with HA-tagged HLA-alleles or control molecules in puc2CL6IP were co-transfected using SuperFect ( Qiagen ) . At 20 h post-transfection , cells were detached with trypsin and measured as described above . LIR-1 binding was analyzed by incubating the cells with recombinant protein [79] and subsequently incubating the cells with an APC-coupled anti-CD85j ( LIR-1 ) antibody . Acquired data was analyzed by FlowJo ( v10 . 1 , Tree Star Inc . ) . For statistical analyses Mann-Whitney U-test or one-way ANOVA followed by Tukey’s or Dunnett´s multiple comparison test were performed using the GraphPad Prism 6 Software . A p-value <0 . 05 was considered significant ( * , p<0 , 05; , p<0 , 005; * , p<0 , 0005 ) . Immunoprecipitation was performed as described previously [38] . Briefly , cells grown in 6-well plates were washed with PBS and metabolically labeled ( Easytag Express [35S]-Met/Cys protein labeling , Perkin Elmer ) with 100 Ci/ml for various times . Cells were lysed in digitonin lysis buffer ( 140 mM NaCl , 20 mM Tris [pH 7 . 6] , 5 mM MgCl2 , and 1% digitonin ( Calbiochem ) ) and cleared from membrane debris at 13 , 000 rpm for 30 min at 4°C . For analysis of lysates with several antibodies , identical lysates were pooled then split up into equal aliquots . Lysates were incubated with antibodies for 2 h at 4°C in an overhead tumbler before immune complexes were retrieved by protein A- or G-sepharose ( GE Healthcare ) . Sepharose pellets were washed four times with increasing NaCl concentrations ( 0 . 15 to 0 . 5 M in lysis buffer containing 0 . 2% detergent ) . For a re-immunoprecipitation the washed beads were subsequently incubated with a lysisbuffer supplemented with 1% Igepal ( Sigma ) and 1% SDS at 95° C for five minutes . The lysisbuffer was diluted to reach a final concentration of 0 . 1% SDS and a subsequent immunoprecipitation was performed . Endoglycosidase H ( New England Biolabs ) treatment was performed as recommended by the manufacturer . Prior to loading onto a SDS-PAGE iImmune complexes were dissociated at 95°C for 5 min in a DTT ( 40 mM ) containing sample buffer . Fixed and dried gels were exposed overnight to a phosphor screen , scanned by Typhoon FLA 7000 ( GE Healthcare ) . For better visualization of the results contrast and light were adjusted . Where mentioned in the figure legend a short or long exposure to x-ray film was used for autoradiography . For Western blot analysis , equal amount of cells were washed in PBS and lysed in lysis buffer ( 50 mM Tris-HCl , pH 7 . 5 , 150 mM NaCl , 1% Igepal , and Complete protease inhibitor ( Roche ) ) . The proteins were separated by SDS-PAGE and transferred to nitrocellulose filter . After incubation with primary antibody a peroxidase coupled secondary antibody was used and chemiluminescence was detected using a LI-COR Blot Scanner . Mock treated or infected MRC5 cells were grown in technical replicates in a 6-well format . Cells were lysed at 24 h post-infection and total RNA was extracted using NucleoSpin kit RNA II ( Macherey-Nagel ) and 600ng was reverse transcribed using a QuantiTect reverse transcription kit ( Qiagen ) and further used for both semi-quantitative and quantitative RT-PCRs . Quantitative RT-PCRs were performed with a QuantiTect SYBR green PCR kit ( Qiagen ) . CT values were normalized to actin ( ΔCT ) and plotted relative to the ΔCT values of the mock treated control cells . For the semi-quantitave analysis the following primer pairs were used: US11-ctrl3’ tggtccgaaaacatccaggg and US11-ctrl5’ ttcgatgaacctccgccctt; US10-ctrl’3 aaccgcatatcaggaggaggga and US10-ctrl’5 tcacgtgcggctgtgttattca , UL40-1 gcagctagcgccgccaccatgaacaaat and UL40-2 cgaggatcctcaagcctttttcaaggcg . For the qRT-PCR we used the primers: qUS10-1 acgacggggaaaatcacgaa and qUS10-2 cagagtagtttcggggtcgg; actin beta primers ( Qiagen , Hs_ACTB_1_SG QuantiTect Primer ) .	The human immune system can cover the presentation of a wide array of pathogen derived antigens owing to the three extraordinary polymorphic MHC class I ( MHC-I ) gene loci , called HLA-A , -B and -C in humans . Studying the HLA peptide ligands of human cytomegalovirus ( HCMV ) infected cells , we realized that the HCMV encoded glycoprotein US11 targeted different HLA gene products in distinct manners . More than 20 years ago the first HCMV encoded MHC-I inhibitors were identified , including US11 , targeting MHC-I for proteasomal degradation . Here , we describe that the prime target for US11-mediated degradation is HLA-A , whereas HLA-B can resist degradation . Our further mechanistic analysis revealed that US11 uses various domains for distinct functions . Remarkably , the ability of US11 to interact with assembled MHC-I and modify peptide loading of degradation-resistant HLA-B was dependent on a low-complexity region ( LCR ) located between the signal peptide and the immunoglobulin-like domain of US11 . To redirect MHC-I for proteasomal degradation the LCR was dispensable . These findings now raise the intriguing question why US11 has evolved to target HLA-A and -B differentially . Possibly , HLA-B molecules are spared in order to dampen NK cell attack against infected cells .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "blood", "cells", "flow", "cytometry", "medicine", "and", "health", "sciences", "immune", "cells", "pathology", "and", "laboratory", "medicine", "hela", "cells", "gene", "regulation", "pathogens", "biological", "cultures", "immunology", "microbiology", "fibroblasts", "viruses", "immunoprecipitation", "dna", "viruses", "connective", "tissue", "cells", "cell", "cultures", "co-immunoprecipitation", "cytotoxic", "t", "cells", "herpesviruses", "research", "and", "analysis", "methods", "human", "cytomegalovirus", "small", "interfering", "rnas", "spectrum", "analysis", "techniques", "white", "blood", "cells", "animal", "cells", "medical", "microbiology", "gene", "expression", "microbial", "pathogens", "connective", "tissue", "biological", "tissue", "t", "cells", "cell", "lines", "precipitation", "techniques", "spectrophotometry", "cytophotometry", "biochemistry", "rna", "cell", "biology", "nucleic", "acids", "anatomy", "viral", "pathogens", "genetics", "biology", "and", "life", "sciences", "cellular", "types", "cultured", "tumor", "cells", "non-coding", "rna", "organisms" ]	2019	HLA-B locus products resist degradation by the human cytomegalovirus immunoevasin US11
Courtship is pivotal for successful mating . However , courtship is challenging for the Cryptococcus neoformans species complex , comprised of opportunistic fungal pathogens , as the majority of isolates are α mating type . In the absence of mating partners of the opposite mating type , C . deneoformans can undergo unisexual reproduction , during which a yeast-to-hyphal morphological transition occurs . Hyphal growth during unisexual reproduction is a quantitative trait , which reflects a strain’s ability to undergo unisexual reproduction . In this study , we determined whether unisexual reproduction confers an ecological benefit by promoting foraging for mating partners . Through competitive mating assays using strains with different abilities to produce hyphae , we showed that unisexual reproduction potential did not enhance competition for mating partners of the same mating type , but when cells of the opposite mating type were present , cells with enhanced hyphal growth were more competitive for mating partners of either the same or opposite mating type . Enhanced mating competition was also observed in a strain with increased hyphal production that lacks the mating repressor gene GPA3 , which contributes to the pheromone response . Hyphal growth in unisexual strains also enables contact between adjacent colonies and enhances mating efficiency during mating confrontation assays . The pheromone response pathway activation positively correlated with unisexual reproduction hyphal growth during bisexual mating and exogenous pheromone promoted bisexual cell fusion . Despite the benefit in competing for mating partners , unisexual reproduction conferred a fitness cost . Taken together , these findings suggest C . deneoformans employs hyphal growth to facilitate contact between colonies at long distances and utilizes pheromone sensing to enhance mating competition . Successful courtship is key to the evolutionary success of sexual organisms , and many species have evolved distinct strategies to locate and choose a mating partner . For example , primates and humans utilize aggression to secure a mating partner [1]; male hummingbirds apply acoustic control using tail feathers during high-speed dives to court females [2]; male Drosophila vibrate their wings to generate different songs to trigger mating responses in females [3]; male tree-hole frogs also adopt acoustic strategies taking advantage of tree trunk cavities to attract females [4]; and female pipefish display a temporal striped pattern ornament to woo male partners [5] . These examples demonstrate that complex eukaryotic organisms can employ visual , vocal , or mechanical tactics to secure a mate and transmit their genetic traits to the next generation . In eukaryotic fungal systems , mating often involves a morphological transition . Saccharomyces cerevisiae yeast cells undergo polarized growth and form shmoo projections in preparation for cell fusion during mating [6] . In filamentous fungi , including both ascomycetes and basidiomycetes , sexual reproduction involves the formation of a fruiting body ( perithecium or basidium , respectively ) [7] . Candida albicans , an ascomycete , undergoes a white-opaque switch to initiate mating [8] . Despite their divergent sexual strategies , these morphological transitions are all controlled by the pheromone response pathway [9] . During yeast mating , physical agglutination of yeast cells does not promote courtship , but rather a gradient of pheromone signals is crucial for successful cell-cell fusion during early mating [10 , 11] . Similarly , in Schizosaccharomyces pombe , local pheromone signals and a spatially focal pheromone response dictate cell-cell pairing and fusion position during early mating processes [12 , 13] . In C . albicans , overexpression of the pheromone response MAP kinase pathway components can enhance mating efficiency [14] . These studies establish that the pheromone response pathway plays a critical role in promoting fungal mating efficiency . The opportunistic human fungal pathogen Cryptococcus deneoformans undergoes a yeast-to-hyphal morphological transition upon mating induction [15] . This species has two modes of sexual reproduction: bisexual reproduction between cells of opposite mating types and unisexual reproduction involving cells of only one mating type [15–17] . Cell fusion between MATa and MATα cells during bisexual reproduction , and between two MATα cells during unisexual reproduction , triggers hyphal development [18] . This morphological transition is orchestrated by the pheromone response pathway [18 , 19] . However , recent studies have shown that hyphal growth during unisexual reproduction can also occur independent of cell fusion and the pheromone response pathway [20–23] , and that pheromone-independent hyphal development is dependent upon the calcineurin pathway [20 , 24] . Because the majority of identified natural and clinical C . neoformans isolates are of the α mating type , unisexual reproduction likely has significant ecological impacts on the Cryptococcus species complex population structure and diversity [25–27] . The limited abundance of MATa cells in natural environments restricts outcrossing and in the absence of a-α mating , unisexual reproduction has been shown to reverse Muller’s rachet and offset the low abundance of MATa cells to avoid an evolutionary dead end [28] . Unisexual reproduction can also generate genotypic and phenotypic diversity de novo [29] . Interestingly , population genetics studies have revealed that genome recombination occurs frequently among environmental isolates [30–32] , even those that are exclusively α mating type , providing evidence that unisexual reproduction involving fusion of MATα cells of distinct genotypes allows meiotic recombination in nature . Despite these evolutionary benefits , cell fusion-independent solo-unisexual reproduction also occurs and because this pathway involves genetically identical genomes , it does not contribute to genome reshuffling or recombination . Similar to pseudohyphal differentiation in S . cerevisiae , C . deneoformans hyphal growth during unisexual reproduction has an ecological benefit in promoting foraging for nutrients and habitat exploration in the surrounding environments [33 , 34] . In this study , we address whether the ability to undergo unisexual reproduction has an additional ecological benefit in promoting foraging for mating partners to facilitate outcrossing and enable recombination in nature . During C . deneoformans solo-unisexual reproduction , cells undergo the yeast-to-hyphal morphological transition independent of cell fusion and nuclei diploidized through endoreplication [16 , 23] . The hyphal growth is a quantitative trait associated with unisexual reproduction that can be used to determine a strain’s ability to undergo unisexual reproduction [35] . Although solo-unisexual reproduction occurs independently of cell-cell fusion , cells can fuse with partners of both the same or opposite mating type at varying frequencies [16 , 23] . To test whether the ability to undergo unisexual reproduction impacts competition for mating partners during outcrossing , we performed mating competition experiments employing three MATα and three MATa C . deneoformans strains with different degrees of unisexual reproduction potential based on their abilities to produce hyphae ( Fig 1A ) [35] . Among these strains , several were F2 progeny derived from crosses between the environmental MATa isolate NIH433 and the clinical MATα isolate NIH12 including a high hyphal ( HH ) strain XL190α , an intermediate hyphal ( MH ) strain XL280α , a low hyphal strain XL187a , and a no hyphal ( NH ) strain JEC20a [15 , 16 , 36–38] . LH strain JEC21α and MH strain XL280a are congenic strains of JEC20a and XL280α , respectively , derived through 10 rounds of backcrossing ( S1 Fig ) [36 , 38 , 39] . For each mating competition experiment , cells of three strains with different hyphal growth carrying dominant , selectable drug resistance markers were mixed , spot-inoculated , and incubated on V8 agar media for 4 days ( Fig 1B ) . Cells were recovered on YPD medium to obtain colony forming units ( CFU ) , and on YPD medium supplemented with different two-drug combinations to determine the cell fusion frequencies . Cell fusion frequencies were compared between different pairs of strains within the same competition mating mixture to determine whether the ability to undergo unisexual reproduction confers benefits in competition for mating partners to facilitate outcrossing ( Fig 1B ) . Prior to the mating competition experiments , cell fusion frequencies were compared between different hyphal strains . During α-α cell fusion , the MATα MH strain displayed a significantly higher cell fusion frequency ( 5 cell fusion events per million CFU ) compared to the HH and LH strains ( 0 . 013 and 0 . 019 cell fusion events per million CFU , respectively ) , in which cell fusion rarely occurred ( Fig 1C ) . This suggests that the ability to undergo more robust hyphal growth is not strictly correlated with α-α cell fusion efficiency . In contrast , during a-α cell fusion , hyphal growth positively correlated with a-α cell fusion efficiency . MH-HH strains had a cell fusion frequency ( 53 cell fusion events per thousand CFU ) about 109 times higher than LH-MH strains , which in turn had a cell fusion frequency ( 0 . 49 cell fusion events per thousand CFU ) about 26 times higher than NH-LH strains ( 0 . 019 cell fusion events per thousand CFU ) ( Fig 1D ) . In all of the strains tested , a-α cell fusion occurred at a much higher level compared to α-α cell fusion , similar to previous findings [16 , 23] . Cell fusion has been previously shown to be dispensable for solo-unisexual reproduction [19 , 23] , which can account for the observed poor correlation between hyphal growth and α-α cell fusion frequency . Thus , we hypothesize that increased hyphal growth may not provide an advantage in competing for mating partners of the same mating type . Indeed , when we performed the unisexual mating competition assay mixing the HH , MH , and LH cells , we observed that HH and LH cells yielded the most fusion products with a cell fusion frequency of 1 . 3 cell fusion events per million CFU that is not significantly different from cell fusion frequencies involved MH cells ( Fig 1E ) that exhibited the highest cell fusion frequency ( Fig 1C ) . These findings indicate that neither α-α cell fusion frequency nor hyphal growth can be used to predict mating partner preference during unisexual reproduction , which supports the hypothesis that the ability to undergo unisexual reproduction does not promote competition for mating partners of the same mating type . To test whether the propensity for unisexual reproduction plays a role in competing for mating partners of the opposite mating type , mating competition assays were conducted for a given MATa isolate between two MATα strains of different hyphal growth phenotypes ( Fig 2A ) . Interestingly , cells capable of producing more hyphae always had a significantly higher cell fusion frequency with MATa cells compared to cells with lower hyphal growth potential ( Fig 2A , S1 Table ) . For example , in the presence of MH MATa cells , HH MATα cells fused with MATa cells 24 times more efficiently than LH MATα cells ( green bar , circles vs squares ) and 8 . 1 times more efficiently than MH MATα cells ( red bar , circles vs triangles ) , and MH MATα cells fused with MATa cells 5 . 8 times more efficiently than LH MATα cells ( yellow bar , triangles vs squares; S1 Table ) . These results suggest that increased hyphal growth correlates with competition for mating partners of the opposite mating type during bisexual reproduction . It was also noted that the mating competition advantage decreased for each competition pair ( 24 , 14 . 5 , and 8 . 9 fold differences for HH vs LH , 8 . 1 , 6 . 5 , and 5 . 3 fold differences for HH vs MH , and 5 . 8 , 4 . 6 , and 1 . 7 fold differences for MH vs LH ) with the decreasing hyphal phenotype of the MATa cells ( Fig 2A and S1 Table ) , suggesting that increased hyphal growth of MATa cells can also promote cell fusion . Besides the observation that hyphal growth enhanced competition for mating partners of the opposite mating type , the presence of higher hyphal MATα cells also stimulated a-α cell fusion . MH MATa and HH MATα cells fused at a frequency of 52 cell fusion events per thousand CFU in the presence of MH MATα cells compared to 10 cell fusion events per thousand CFU in the presence of LH MATα cells ( 5 . 2-fold , circles in red bar vs circles in green bar in Fig 2A a MH group ) ( Dark yellow-shaded cells in S1 Table ) . MH MATa and MH MATα cells fused at a frequency of 6 . 4 cell fusion events per thousand CFU in the presence of HH MATα cells compared to 3 . 4 cell fusion events per thousand CFU in the presence of LH MATα cells ( 1 . 9-fold , triangles in red bar vs triangles in yellow bar in Fig 2A a MH group ) ( Dark blue-shaded cells in S1 Table ) . Similar trends were observed during competition for LH MATa cells in that the presence of MH MATα cells increased LH MATa and HH MATα cell fusion frequency by 4 . 5-fold compared to the presence of LH MATα cells ( circles in red bar vs circles in green bar in Fig 2A a LH group; Medium yellow-shaded cells in S1 Table ) , and the presence of HH MATα cells increased LH MATa and MH MATα cell fusion frequency by 2 . 2-fold compared to the presence of LH MATα cells ( Triangles in red bar vs triangles in yellow bar in Fig 2A a LH group; Medium blue-shaded cells in S1 Table ) . However , cell fusion frequencies between MATa cells and LH MATα cells were comparable in the presence of HH or MH MATα cells ( 0 . 41 or 0 . 59 cell fusion events per thousand CFU for MH MATa cells indicated by squares in Fig 2A a MH group , and 0 . 02 or 0 . 02 cell fusion events per thousand CFU for LH MATa cells indicated by squares in Fig 2A a LH group ) ( Dark and medium green-shaded cells in S1 Table ) . Interestingly , no enhancement of cell fusion frequency by high hyphal MATα cells was observed during competition for NH MATa cells ( Fig 2A a NH group; light color-shaded cells in S1 Table ) . Notably , the enhancement of cell fusion frequency by higher hyphal MATα cells did not occur when either MATa NH or MATα LH cells were involved in a-α cell fusion , suggesting that strains with poor unisexual reproduction potential have a disadvantage in competing for mating partners of the opposite mating type . The ability to undergo unisexual reproduction also correlated with cell fusion between cells of the same mating type when a MATa partner is present . In the presence of MH or LH MATa cells , HH and MH MATα cells fused at higher frequencies ( 26 and 27 cell fusion events per million CFU , respectively ) compared to HH and LH MATα cells ( 8 . 1 and 3 . 4 cell fusion events per million CFU , respectively ) , and in the presence of MH , or LH , or NH MATa cells , HH and LH MATα cells fused at higher frequencies ( 8 . 1 , 3 . 4 , and 2 . 8 cell fusion events per million CFU , respectively ) compared to MH and LH MATα cells ( 0 . 71 , 0 . 28 , and 0 . 14 cell fusion events per million CFU , respectively ) ( Fig 2B ) , suggesting that increased hyphal growth correlated with enhanced α-α cell fusion frequency in the presence of MATa cells . We also observed a trend where α-α cell fusion frequencies ( HH and MH , HH and LH , and MH and LH ) decreased with reduced hyphal MATa cells ( Fig 2B ) , suggesting that the presence of more robust hyphal MATa cells can further enhance α-α cell fusion . In summary , strains with robust hyphal production have an advantage in competing for mating partners of the opposite mating type , and also for mating partners of the same mating type when cells of the opposite mating type are present . The pheromone response pathway plays an important role in the yeast-to-hyphal morphological transition during C . deneoformans sexual reproduction . This signaling cascade is controlled by G proteins and RGS proteins , including the Gα protein Gpa3 which represses hyphal growth during mating [40–43] . To further examine the impact of the ability to undergo unisexual reproduction during mating competition , we generated strains enhanced for hyphal production by deleting the GPA3 gene in the LH strain JEC21α . gpa3Δ mutants exhibited significantly increased hyphal growth during both unisexual and bisexual reproduction compared to the parental strain ( Fig 3A ) . Next , mating competition assays were conducted using the enhanced hyphal ( EH ) strain JEC21α gpa3Δ to test its ability to compete for mating partners . Similar to the observation in HH , MH , and LH strains , enhanced hyphal production did not increase cell fusion between MATα cells but did increase cell fusion frequency by 4-fold between MATa and MATα cells compared to the parental LH strain ( Fig 3B and 3C ) . However , the increase is not statistically significant due to the low cell fusion frequencies between strains of low hyphal background . Unisexual mating competition assays were performed to compare the abilities of LH and EH MATα cells to fuse with MH MATα cells . In the control assay , cell fusion frequencies were comparable between cells of all three strain combinations ( MH with LH-NAT , MH with LH-NEO , and LH-NAT with LH-NEO ) ( Fig 3D ) . In the assay mixing LH , MH , and EH cells , EH cells fused with MH cells at a significantly higher frequency of 2 . 8 cell fusion events per million CFU compared to LH cells ( 85-fold ) ( Fig 3E ) , suggesting that deletion of GPA3 increases competitiveness for mating partners of the same mating type . In the mating competition during bisexual reproduction , no advantage was observed in the fusion of NH MATa cells with either EH or the parental LH MATα cells ( Fig 3G ) . However , a significant 2 . 9-fold increase was observed in total a-α cell fusion events during mating competitions for NH MATa cells between EH and LH MATα cells compared to control mating competitions for the same MATa cells between LH-NAT and LH-NEO MATα cells ( Fig 3F and 3G ) , indicating that presence of cells with enhanced ability to undergo unisexual reproduction allows both EH and LH MATα cells to fuse with MATa mating partners more efficiently during bisexual reproduction . A significant 2 . 9-fold increase was observed in α-α cell fusion between EH and LH MATα cells in the presence of NH MATa cells compared to cell fusion between LH-NAT and LH-NEO MATα cells ( Fig 3H ) , suggesting that in the presence of MATa cells , GPA3 deletion also enhances competition for mating partners of the same mating type . Overall , this analysis of the enhanced hyphal growth strain JEC21α gpa3Δ provides additional support for models in which increased unisexual reproduction potential enhances competition for mating partners . Unisexual reproduction provides evolutionary and ecological benefits for C . deneoformans by generating aneuploid progeny with phenotypic diversity and by promoting habitat exploration through hyphal growth [29 , 34] . Here we further show that unisexual cells have an advantage in competing for mating partners within the same colony . We tested whether hyphal growth during unisexual reproduction confers benefits in foraging for mating partners . Both long-term and short-term foraging for mating partner experiments suggested that hyphal growth promoted foraging for mating . In a six-week mating confrontation experiment , hyphae of different MATα unisexual reproduction backgrounds marked with NEO grew towards the same MATa cells marked with HYG that were 4 mm apart ( Fig 4A and 4B ) . When competing for either the same MATa or MATα cells ( except LH MATα cells ) , although not all pairwise comparisons by t-test were significant due to the lack of contact when strains of no or low hyphal growth were involved , a significant trend by one-way ANOVA was observed that isolates with more hyphal growth yielded more double drug resistant colonies than isolates with reduced hyphal growth ( Fig 4C and S2 Table , seven group one-way ANOVA analyses are listed above pairwise Welch’s t-test analyses ) . In a seven-day mini-colony mating experiment , colonies derived from single cells produced hyphae that allowed contact with adjacent colonies of the opposite mating type ( S2A Fig ) . Similar to the long-term mating confrontation experiment , a significant trend by one-way ANOVA was observed in which isolates with more hyphal growth had an advantage in forming double drug resistant colonies ( S2B Fig and S3 Table ) . Although pairwise comparisons by t-test showed significant differences between crosses involved NH and LH cells that were not observed in the confrontation experiment , this discrepancy is due to differences in the experimental setup where cells were inoculated at 0 . 4 cm apart during confrontation , whereas randomly plated on agar media during mini-colony mating experiment , where colony contact is enabled by both hyphal growth and chance . Overall , these results suggest that hyphal growth during unisexual reproduction can facilitate contact between mating partners in adjacent environments . Elevated pheromone response pathway activation and increased response to pheromones are critical to successful courtship during mating in S . cerevisiae and C . albicans [10 , 11 , 14] . S . cerevisiae utilizes the α-factor protease Bar1 and a-factor barrier Afb1 to discriminate mating partners with different pheromone levels and drive evolution towards higher pheromone production for efficient mating [44–47] . Pheromones also stimulate mating and the yeast-to-hyphal morphological transition during C . neoformans bisexual reproduction [48] . To determine the role of the pheromone response pathway during C . deneoformans mating competition , expression levels were examined for the genes encoding the pheromones MFα and MFa , the pheromone receptors Ste3α and Ste3a , the MAP kinase Cpk1 , the transcription factors Mat2 and Znf2 , and the plasma membrane fusion protein Prm1 [23 , 40 , 49] . Pheromone response pathway activation did not correlate with the hyphal growth phenotype in MATa or MATα strains . In MATa strains , all of the pheromone response pathway genes were significantly upregulated in the MH strain XL280a compared to the LH and NH strains , but only MFa and PRM1 were significantly upregulated in the LH strain XL187a compared to the NH strain JEC20a ( S3 Fig ) . In MATα strains , all of the pheromone response pathway genes were significantly upregulated in the HH and MH strains compared to the LH strain JEC21α , but the MH strain XL280α had a significantly higher pheromone pathway activation compared to the HH strain XL190α ( S3 Fig ) . The pheromone pathway was significantly upregulated in the EH strain JEC21α gpa3Δ compared to the parental LH strain JEC21α ( S3 Fig ) . These expression analyses suggest that pheromone response pathway activation per se is not sufficient to explain the ability to undergo unisexual reproduction and its association with competition for mating partners . It was previously shown that the cell fusion protein Prm1 is not required for unisexual reproduction [23] , and certain environmental factors , such as copper and glucosamine , can induce hyphal growth independently of the pheromone response pathway [20 , 21] . In a recent study on the quorum sensing peptide Qsp1 , deletion of pheromone and pheromone receptor genes had little impact on hyphal growth during unisexual reproduction [22] , further indicating the polygenic nature of unisexual reproduction . Despite the incongruent association of the pheromone response pathway and the ability to undergo unisexual reproduction , pheromone response pathway activation was positively correlated with the hyphal growth during bisexual reproduction . The α pheromone gene MFα , both a and α pheromone receptor genes STE3α and STE3a , and the plasma membrane fusion gene PRM1 showed significant correlation with the hyphal growth phenotype ( Fig 5A ) . Although MFa expression was lower in the cross between a MH and α HH strains compared to a LH and α MH strains , the significant upregulation of the MFα , STE3α , and STE3a may compensate for the overall pheromone response activation ( Fig 5A ) . The gene expression patterns of two transcription factors Mat2 and Znf2 that regulate yeast-to-hyphal morphological transition and mating significantly correlated with the hyphal growth except for the cross between a MH and α HH strains . Nonetheless , these two genes were expressed at higher levels compared to the cross between a LH and α MH strains ( Fig 5A ) . The expression of the MAP Kinase CPK1 gene poorly correlated with hyphal growth ( Fig 5A ) , which is likely due to post-translational control of the MAP kinase through phosphorylation , which can relieve a requirement for expression level upregulation during pathway activation . Overall , the pheromone response pathway activation is largely congruent with the hyphal growth phenotype suggesting that in the presence of cells of the opposite mating type , unisexual cells are capable of upregulating the pheromone response pathway in both MATa and MATα cells to compete for mating partners . To validate that pheromone contributes to mating competitiveness , we tested whether synthetic α pheromone promotes a-α cell fusion . Indeed , exogenous α pheromone promoted cell fusion between a NH and α LH cells in a dose dependent manner ( Fig 5B ) . However , the enhancement of cell fusion frequency by pheromone is limited , compared to the 2788-fold increase of cell fusion frequency between a MH and α HH cells over a NH and α LH cells ( Fig 1D ) that coincided with a 34 . 5-fold increase in α pheromone expression ( Fig 5A ) . Interestingly , the mild increase in cell fusion by exogenous pheromone is not observed in the cross between a NH and α EH cells ( S4A Fig ) , which is likely due to the saturation of Ste3a by the 1184-fold higher α pheromone expressed by α EH cells ( S3 Fig ) . The less than two-fold increase in cell fusion provided by 5 μM exogenous pheromone suggested that changes in α pheromone alone are not able to tip the balance and drive the entire pheromone response pathway towards a stronger increase in mating efficiency . In support , exogenous supply of 500 nM pheromone provided limited impact in cell fusion between cells of higher hyphal growth phenotypes ( S4A Fig ) ; and when hyphal growth was suppressed under nutrient rich conditions , exogenous pheromone had little to no impact on cell fusion ( S4B Fig ) . In response to the pheromone signal , S . cerevisiae undergoes filamentous growth to enhance the probabilities of cells finding a mating partner [50] . Here we observed that the a EH colonies responded to 5 μM α pheromone peptide and produced abundant hyphae ( Fig 5C ) , similar to previous report [51] , suggesting that pheromone can promote hyphal growth and increase the contact opportunities between adjacent colonies within the same environment . Upregulation of the pheromone response pathway enhances mating efficiency; however , this upregulation can result in a fitness cost in S . cerevisiae and C . albicans [14 , 52] . In yeast , a short-term experimental evolution experiment showed that mutations abrogating expression of 23 genes involved in mating conferred a fitness benefit during yeast growth when functions of these genes are not required [52] . In C . albicans , cells undergo a white-opaque switch and upregulate the pheromone response pathway , which results in a fitness cost for the opaque cells [14] . Given that the pheromone response pathway is activated at a higher level in C . deneoformans strains with more hyphal growth during bisexual reproduction ( Fig 5A ) , we investigated whether unisexual reproduction confers a fitness cost . Growth curve analyses in YPD liquid medium were performed using an automated Tecan Sunrise absorbance reader to determine the fitness of different strains . The cultures were agitated with vigorous shaking for one minute bihourly before each OD600 measurement . The minimum agitation allows for differentiation of growth curve kinetics among the strains tested , which would otherwise be indistinguishable when grown on solid YPD medium . Compared to the low hyphal growth strains JEC21α and JEC20a , gpa3Δ mutants exhibited a growth defect in nutrient rich media , suggesting that deletion of GPA3 resulted in a fitness cost ( S5 Fig ) . However , this fitness cost was not due to the yeast-to-hyphal morphological transition as hyphal growth was not present . Interestingly , deletion of the GPA3 gene in the sister species C . neoformans did not induce hyphal growth or cause a growth defect compared to the non-hyphal strain KN99a ( S5 Fig ) . It has been shown that the pheromone response pathway activation by GPA3 gene deletion is lower in KN99a than in JEC21 or JEC20 , indicating that deletion of GPA3 is not sufficient to rewire cellular responses to induce unisexual reproduction in C . neoformans [40] . We next performed fitness competition experiments . After 10 days of incubation of equally mixed cells on YPD and V8 agar medium , cells were collected and plated on selection media to determine colony forming units for each competition strain . Hyphal growth was observed on both YPD and V8 media when a MH cells were present , and both yeast cells and hyphae were collected to compare fitness ( S6 Fig ) . During competition between two LH strains , cells were recovered at about 1 to 1 ratio after 10-day incubation on both YPD and V8 media both in the absence or in the presence of a cells ( Fig 6A ) . In contrast , LH strain outcompeted EH strain when cells were incubated on V8 medium or when a cells were present ( Fig 6B ) . On YPD medium in the absence of a cells , the EH strain that displayed poor growth in liquid media ( S5 Fig ) outcompeted the parental LH strain ( Fig 6B ) , which is likely due to differential cellular responses under different growth conditions . However , this competition advantage was reversed when a cells were present or when mixed cells were incubated on V8 medium ( Fig 6B ) , suggesting that the presence cells of the opposite mating type or the mating inducing environment can elicit a fitness cost . This competition disadvantage for the EH strain on V8 medium was further exacerbated when a cells were present during competition ( Fig 6B ) . These fitness competition assays demonstrate that gpa3Δ mutation enhances mating competition at a cost of growth fitness , and this fitness cost is likely due to the energy required in the expression of the pheromone response pathway genes . It has been reported that long-term passage on rich media in the lab often diminishes the hyphal growth phenotype of MATa strains , which further suggests that there is a fitness cost associated with the ability to undergo unisexual reproduction [53] . Interestingly , in S . cerevisiae , the [SWI+] prion state promotes outcrossing efficiency due to a defect in HO expression and mating type switching , which also leads to a fitness cost [54] . Taken together , fungal species have evolved different strategies in promoting mating in nature accompanied with a fitness tradeoff . Sexual reproduction plays a pivotal role in shaping fungal population structure and diversity in nature . However , studies on how fungi secure a mating partner in nature for successful mating are limited . In this study , we aimed to characterize the ecological and evolutionary benefits of unisexual reproduction in C . deneoformans . Similar to the landmark study by Jackson and Hartwell showing that higher pheromone production promotes courtship in S . cerevisiae [11] , we showed that strains with higher potential for unisexual reproduction are more competitive for mating partners of both the same and the opposite mating types when cells of the opposite mating type are present , and the pheromone response pathway activation is positively correlated with the hyphal growth phenotype ( Fig 7 ) . More interestingly , in addition to pheromone sensing , unisexual cells employ hyphal growth to increase contact opportunities between colonies at relatively long distances . However , this mating competition advantage results in a fitness cost for unisexual cells during mitotic growth under mating-inducing conditions . The strains involved in this study were all derived from natural and clinical isolates under laboratory conditions , suggesting that the ability to undergo unisexual reproduction is likely to span a broad range in the environment . The majority of natural and clinical isolates of the Cryptococcus species complex are found to be of the α mating type , which accounts for 99% of C . neoformans isolates [25–27] . In a survey of C . deneoformans environmental distribution around the Mediterranean basin , 27% of isolates are MATa isolates , which were all recovered in Greece , suggesting certain environmental niches harbor MATa cells [27] . Genomic and genetic evidence also suggest that recombination is prevalent among these environmental isolates , and some isolates are isolated from a single Eucalytus hollow , which underscores that mating occurs in nature [30–32] . Although sexual structures of Cryptococcus species have yet to be documented in nature , plant material-based media such as V8 , on which we conducted the mating competition assays , can readily induce sexual reproduction under laboratory conditions , suggesting that the mating competition we observed could happen in its environmental niche . We hypothesize that in the presence of MATa cells sparsely distributed in the environment , undirected hyphal growth first enables unisexual MATα cells to forage for mating partners over a much larger surface area than is available to cells within a much more compact budding yeast colony . Next , as MATα hyphae come into the proximity of rare MATa cells , pheromone response pathway activation in both MATα and MATa cells can further enhance mating competition . This mating competition advantage could promote outcrossing and provide an evolutionary advantage by facilitating genome reshuffling via meiotic recombination in a pathogenic yeast species . Strains used in this study are listed in S4 Table . Strains with different hyphal growth phenotypes , XL190α , XL280α , JEC21α , XL280a , XL187a , and JEC20a , were selected to represent high , intermediate , and low hyphal strains [16 , 35 , 38 , 39] . Yeast cells were grown at 30°C on Yeast extract Peptone Dextrose ( YPD ) medium . Strains harboring dominant selectable markers were grown on YPD medium supplemented with 100 μg/mL nourseothricin ( NAT ) , 200 μg/mL G418 ( NEO ) , or 200 μg/mL hygromycin ( HYG ) for selection . Mating assays were performed on 5% V8 juice agar medium ( pH = 7 . 0 ) or Murashige and Skoog ( MS ) medium minus sucrose ( Sigma-Aldrich ) in the dark at room temperature for the designated time period . NAT ( pAI3 ) or G418/NEO ( pJAF1 ) resistant expression constructs were introduced into XL190α , XL280α , and JEC21α , and a HYG ( pJAF15 ) resistant expression construct was introduced into XL280α , XL280a , XL187a , and JEC20a ectopically via biolistic transformation as previously described [55–57] . To generate deletion mutants for GPA3 , a deletion construct consisting of the 5’ upstream and 3’ downstream regions of GPA3 gene flanking the NEO cassette was generated by overlap PCR as previously described [58] . The GPA3 deletion construct was introduced into the strain JEC21α via biolistic transformation . Transformants were selected on YPD medium supplemented with G418 , and gene replacement by homologous recombination was confirmed by PCR . Primers used to generate these deletion constructs are listed in S5 Table . Cells were grown on V8 agar medium for seven days or three weeks in the dark at room temperature to allow hyphal formation . Hyphal growth on the edge of mating patches was imaged using a Nikon Eclipse E400 microscope equipped with a Nikon DXM1200F camera . For each competitive mating assay , cells were grown overnight in YPD liquid medium at 30°C and adjusted to OD600 = 0 . 5 in sterile H2O , and then equal volumes of cells marked with different dominant drug resistant markers were mixed and spot inoculated ( 50 μl ) on V8 agar medium . The mating plates were incubated for four days in the dark at room temperature . The cells were harvested and plated in serial dilution on YPD medium and YPD medium supplemented with different two drug combinations ( NAT and NEO , NAT and HYG , or NEO and HYG ) . The cells were incubated for three to five days at room temperature and colony forming units were counted . Cell fusion frequencies were determined as double drug resistant CFU/total CFU . The complete competitive mating experimental design is listed in S6 Table . Each mating competition was performed in biological triplicate . To investigate whether hyphal growth enables cells foraging for mating partners , we performed long-term confrontation and short-term mini-colony mating experiments . For the confrontation mating experiment , HYG resistant XL280a , XL187a , and JEC20a were streaked and grown on V8 medium to form a line of cells , and then NEO resistant XL190α , XL280α , and JEC21α , and JEC21α gpa3Δ::NEO were spot inoculated 4 mm apart in parallel along the MATa cells . Unisexual hyphae grew towards cells of the opposite mating type for six weeks . Then the cells were collected and plated on YPD medium supplemented with HYG and NEO . After incubation at room temperature for three to five days , total double drug resistant colony forming units were counted to determine whether hyphal growth conferred an advantage in foraging for mating partners . Each confrontation mating experiment was performed in biological quintuplicate . For the mini-colony mating experiment , the aforementioned HYG resistant MATa strains and NEO resistant MATα strains were grown overnight in YPD liquid medium and adjusted to OD600 = 0 . 008 . For each mating pair , 100 μl of MATa cells and 100 μl of MATα cells were mixed and plated on V8 agar medium to form evenly spaced mini colonies . Unisexual hyphae facilitate contact between adjacent colonies after growing for seven days . Then the cells were collected and plated on YPD medium and YPD medium supplemented with HYG and NEO . The cells were incubated for three to five days at room temperature and colony forming units were counted . Cell fusion frequencies were determined as double drug resistant CFU/total CFU . Each mating was performed in biological triplicate . To examine pheromone response pathway activation , qRT PCR experiments were performed on RNA extracted from cells incubated on V8 agar medium for 36 hours as previously described [23] . In brief , XL190α , XL280α , JEC21α , JEC21α gpa3Δ::NEO , XL280a , XL187a , and JEC20a were grown overnight in YPD liquid medium and adjusted to OD600 = 2 in sterile H2O . Then cells of individual strains and an equal-volume mixtures of cells for crosses between XL190α and XL280a , XL280α and XL187a , JEC21α and JEC20a , and JEC21α gpa3Δ::NEO and JEC20a were spotted ( 250 μl ) on V8 medium and incubated for 36 hours . Cell patches of individual strains and of mixture of a and α strains were scraped off the medium and transferred into Eppendorf tubes then flash frozen in liquid nitrogen . RNA was extracted using TRIzol reagent ( Thermo ) following the manufacturer’s instructions . RNA was treated with Turbo DNAse ( Ambion ) , and single-stranded cDNA was synthesized by AffinityScript RT-RNAse ( Stratagene ) . cDNA synthesized without the RT/RNAse block enzyme mixture was used to control for genomic DNA contamination . The relative expression levels of targeted genes were measured by qRT PCR using Brilliant III ultra-fast SYBR green QPCR mix ( Stratagene ) in an Applied Biosystems 7500 Real-Time PCR system . A “no template control” was used to analyze the resulting melting curves to exclude primer artifacts for each target gene . Gene expression levels were normalized to the endogenous reference gene GPD1 using the comparative ΔΔCt method . Primers used for qRT-PCR are listed in S5 Table . For each target gene and each sample , technical triplicate and biological triplicate were performed . To address whether pheromone promotes mating competition and hyphal growth , carboxyl farnesylated and methylated α pheromone peptide ( QEAHPGGMTLC ) ( synthesized at GenScript , USA ) was tested for its impact on mating and hyphal growth . α pheromone peptide was dissolved in methanol at the concentration of 50 μM and 10-fold serial dilutions in methanol were prepared as stock solutions . For the mating assay , HYG marked MATa and NEO marked MATα cells were prepared and mixed as mentioned above for the mating competition assays . α pheromone peptide was added to mixed a NH ( CF926 ) and α LH ( CF759 ) cells at the concentrations of 0 , 500 pM , 5 nM , 50 nM , 500 nM , and 5 μM , and the mixed cells were spot-inoculated on the V8 media and incubated in the dark at room temperature for four days . Cells were then harvested and plated on both YPD and YPD media supplemented with NEO and HYG to determine cell fusion frequency . Same mating assays were carried out for crosses between a MH ( CF978 ) and α HH ( CF914 ) , a LH ( CF931 ) and α MH ( CF752 ) , a NH ( CF926 ) and α LH ( CF759 ) , and a NH ( CF926 ) and α EH ( CF1314 ) both in the absence and in the presence of 500 nM α pheromone peptide on both YPD and V8 media . To test the impact of α pheromone peptide on hyphal growth , 5 μM α pheromone peptide and methanol were dropped onto MS media and allowed to dry . a EH ( YPH86 ) cells were grown overnight and washed with H2O twice , and then inoculated onto the MS plate with dried α pheromone peptide and methanol droplets at a different spot . Cells were then microscopically manipulated and transferred to the α pheromone peptide and methanol spots . Colony hyphal growth was monitored daily and imaged after incubation in the dark at room temperature for 72 hours . Competition experiments were performed to compare fitness between two low hyphal strains and between low and enhanced hyphal strains both in the absence and in the presence of the a MH cells . Overnight cultures of JEC21α NAT , JEC21α NEO , JEC21α gpa3Δ::NEO , and XL280a were washed with H2O twice and cell densities were determined with a hemocytometer . For each competition experiment , 10 μl of H2O containing 100 , 000 cells of each strain was spotted on either YPD or V8 agar medium and incubated in the dark at room temperature for 10 days . Cells were collected and plated on YPD medium supplemented with NAT or G418 to determine colony forming units . Cell mixtures were plated before incubation to control for equal mixing . Each competition was performed in triplicate . Fitness was determined by calculating the percentile of the recovered CFU of each strain out of the total recovered dominant drug resistant CFU . To determine the growth fitness of different unisexual strains , KN99a , KN99a gpa3Δ::NEO , JEC20a , JEC20a gpa3Δ::NEO , JEC21α , and JEC21α gpa3Δ::NEO were grown overnight in YPD liquid medium and washed twice in H2O . 10 , 000 cells for each strain were resuspended in 200 μl YPD liquid medium and incubated in a 96-well plate ( Corning ) at 30°C with vigorous shaking for 1 min bihourly . OD600 readings were measured bi-hourly after shaking using an automated Tecan Sunrise absorbance reader . Each sample was tested in quintuplicate . All statistical analyses were performed using the Graphpad Prism 7 program . Welch’s t-test was performed for each pairwise comparison , and one-way ANOVA was performed for each group analysis with a p value lower than 0 . 05 considered statistically significant ( * indicates 0 . 01<p≤0 . 05 , indicates 0 . 001<p≤0 . 01 , * indicates 0 . 0001<p≤0 . 001 , and **** indicates p≤0 . 0001 ) .	Sexual reproduction plays a pivotal role in shaping fungal population structure and diversity in nature . The global human fungal pathogen Cryptococcus neoformans species complex evolved distinct sexual cycles: bisexual reproduction between mating partners of the opposite mating types , and unisexual reproduction with only one mating type . During both sexual cycles , cells undergo a yeast-to-hyphal morphological transition and nuclei diploidize through either cell-cell fusion followed by nuclear fusion during bisexual reproduction or endoreplication during unisexual reproduction . Despite the complex sexual life cycle , the majority of Cryptococcal isolates are α mating type . Albeit the scarcity of MATa cells in the environment , meiotic recombination is prevalent . To decipher this conundrum , we ask whether there is an underlying mechanism in which Cryptococcus species increase their mating opportunities . In this study , we showed that the undirected hyphal growth during unisexual reproduction enables MATα cells to forage for mating partners over a larger surface area , and when MATα hyphae come into close proximity of rare MATa cells , pheromone response pathway activation in both MATα and MATa cells can further enhance mating . This mating enhancement could promote outcrossing and facilitate genome reshuffling via meiotic recombination .	[ "Abstract", "Introduction", "Results", "and", "discussion", "Conclusion", "Materials", "and", "methods" ]	[ "cell", "physiology", "cryptococcus", "neoformans", "medicine", "and", "health", "sciences", "cryptococcus", "pathology", "and", "laboratory", "medicine", "pathogens", "microbiology", "social", "sciences", "fungi", "model", "organisms", "animal", "behavior", "experimental", "organism", "systems", "zoology", "fungal", "pathogens", "research", "and", "analysis", "methods", "saccharomyces", "foraging", "mycology", "animal", "studies", "behavior", "medical", "microbiology", "gene", "expression", "microbial", "pathogens", "candida", "albicans", "yeast", "biochemistry", "candida", "psychology", "eukaryota", "cell", "biology", "genetics", "biology", "and", "life", "sciences", "pheromones", "saccharomyces", "cerevisiae", "yeast", "and", "fungal", "models", "organisms", "cell", "fusion" ]	2019	Unisexual reproduction promotes competition for mating partners in the global human fungal pathogen Cryptococcus deneoformans
The chemotactic response of cells to graded fields of chemical cues is a complex process that requires the coordination of several intracellular activities . Fundamental steps to obtain a front vs . back differentiation in the cell are the localized distribution of internal molecules and the amplification of the external signal . The goal of this work is to develop a mathematical and computational model for the quantitative study of such phenomena in the context of axon chemotactic pathfinding in neural development . In order to perform turning decisions , axons develop front-back polarization in their distal structure , the growth cone . Starting from the recent experimental findings of the biased redistribution of receptors on the growth cone membrane , driven by the interaction with the cytoskeleton , we propose a model to investigate the significance of this process . Our main contribution is to quantitatively demonstrate that the autocatalytic loop involving receptors , cytoplasmic species and cytoskeleton is adequate to give rise to the chemotactic behavior of neural cells . We assess the fact that spatial bias in receptors is a precursory key event for chemotactic response , establishing the necessity of a tight link between upstream gradient sensing and downstream cytoskeleton dynamics . We analyze further crosslinked effects and , among others , the contribution to polarization of internal enzymatic reactions , which entail the production of molecules with a one-to-more factor . The model shows that the enzymatic efficiency of such reactions must overcome a threshold in order to give rise to a sufficient amplification , another fundamental precursory step for obtaining polarization . Eventually , we address the characteristic behavior of the attraction/repulsion of axons subjected to the same cue , providing a quantitative indicator of the parameters which more critically determine this nontrivial chemotactic response . Growth cones are 3D hand-shaped structures , which dynamically change their conformation . Filopodia protrude from the GC membrane , continuously extending and retracting to explore the environment and to create adhesion to the substratum . These phenomena are highly complex and take place at a time frequency faster than the one we are interested into , typically seconds vs . a few minutes . For the purposes of the present model , we neglect fine local details of the shape and we represent GCs as 2D disk-like structures with diameter of . This simplification models the average shape assumed by GCs in their state prior to actual motion ( the “pausing state” cited in [19] ) . The same assumption has been made by other authors interested in GC mathematical modelling , see [16] , [33] , [34] . We refer as a paradigm to the in vitro chemotactic assay for neural cells , where GCs are exposed to steady graded concentrations of a chemical ligand released by a pipette ( see , e . g . , [35] , [36] ) . Observe that this is essentially a two–dimensional situation . Polar coordinates are used , the origin of the axes being positioned in the center of the GC . The angle denotes the azimuthal coordinate on the membrane and its origin is set along the direction connecting the GC center with the pointwise source , which we always suppose to lay on the right hand side of the cell ( see Fig . 1 ) . We denote by the radius of the GC . In the model , we deal with membrane species ( receptors , cyclases and calcium channels ) and cytosolic species ( G–proteins , cyclic AMP , cyclic GMP , kinases and their corresponding precursors ) . To connect membrane and cytosolic concentrations ( and vice versa ) , we use the dimensional corrective factor ( or its inverse ) . We use the notation to represent concentration of a certain species; moreover , we denote by the superscript the bound or active form of the molecules . The nomenclature for the species is reported in Tab . 1 . In this section , we propose an extended version of the model to study the bifunctional response to a guidance cue , a phenomenon known to interest the response to netrins [27] . Engagements of receptor complexes is known to control the specificity and the polarity of the response of the neuron to the guidance cue . Here we will not deal with the details of the dynamics of the formation of the DCC–UNC5 complex , and we will always consider such a dynamics at the equilibrium . From the modelling point of view , this amounts to prescribe a priori the percentage of DCC receptors forming a complex with UNC5 ( on this issue , see also the discussion in the Methods Section ) . Setting in the tests various percentages , we analyze a wide spectrum of situations . To describe the DCC–UNC5 complex , we follow here the idea proposed in [28] , and namely that such a complex leads to cGMP synthesis , regulated by 12-hydroxyperoxyeicosatetraenoic acid ( HPETE ) via direct activation of guanylate cyclase [39] . Enhancement of the cGMP level causes to calcium channel closure [40] . This fact is at the origin of the significant decrease of inward calcium flux in UNC5–overexpressing GCs , repelled by netrin . Due to the substantial similarity with the sole DCC pathway , many mechanisms are modeled here in the same way . The binding reaction for the DCC–UNC5 complex ( for clarity indicated just as UNC ) readsThe ligated complex induces dissociation of a G–protein ( probably of Rho type , see [12] ) , that we denote here by , givingThis process leads to the formation of HPETE . Not being yet completely explained , we consider here a second order reaction to occur ( see the Methods Section for a more detailed discussion of this aspect ) HPETE activates soluble guanylate–cyclase ( ) which , in turn , catalyzes the synthesis of cGMP from the guanosine triphosphate ( GTP ) substrateFormation of the guanosine–dependent kinase protein ( PKG ) enzymatically enhances closure of calcium channels We first demonstrate that the model correctly achieves polarization and reaches a steady state condition , where front–back differentiation is established . In Fig . 4 , we plot the concentration profiles as a function of time , obtained from simulations carried out till . In Fig . 5 , we plot for the different species the abscissa of the barycenter of the molecules as a function of time . Since at the initial time the distribution of molecules is homogeneous , the displacement of their spatial barycenter from represents an index of the intensity of the polarization . In Fig . 6 , we plot the concentrations after as a function of . Significant polarization of the receptors takes place in tenth of minutes; polarization is inherited by all the internal species . An interesting behavior is shown by calcium channels , which undergo in all sectors a first phase of opening , reaching a fairly similar maximum value , followed by closure , more pronounced in the rear side . This mechanism might represent a sort of LEGI , global “inhibition” being constituted by collective opening and local “activation” by differential closure . Note that this is to be intended only as a qualitative interpretation ( see also the discussion in [11] ) . The position of the barycenter of bound receptors presents a sigmoid behavior , which is characteristic of autocatalyzing processes: a first phase of relatively slow accumulation ( lag time ) followed by a quick growth till a steady state . This is in agreement with the experimental result of [19 , Fig . 1e] . In Fig . 7 , we plot the concentration profiles as a function of time for a simulation with a longer integration interval . A steady state is definitely reached . In [19] , an experiment is reported , where a chelator is used to subtract the calcium available in the cytosol , obtaining a suppression in the asymmetric relocalization of receptors . To perform an in silico investigation of this experiment , we have studied the effect of the variation of the feedback coefficient . Fig . 8 ( left ) shows that significant receptor relocalization is obtained only above a nonzero threshold of . More importantly , the model predicts that the lack of receptor relocalization implies absence of chemotactic response and not only a weaker , but still existing , response . Fig . 8 ( right ) quantifies how the asymmetry in DCC receptors localization is reflected in downstream differential opening of the calcium channels . This indicates that a sufficient active relocalization of receptors is an upstream enhancing event needed to produce chemotactic response . In a second investigation , we have considered the effect of amplification due to chemical kinetics . In the signalling pathway , we may identify –loosely speaking– two families of reactions . The first family consists of stoichiometric reactions , as for example ligand–receptor or G–cyclase binding , involving one–to–one molecule synthesis . The second family consists instead of enzymatic reactions , which , involving a one–to–more molecule production , drive an internal amplification processes . We focus our attention on this latter family , taking as a representative case the cAMP production catalyzed by cyclase . We perform different simulations with a decreasing kinetic constant , which quantitatively modulates how many molecules of cAMP are produced starting from an available molecule of activated cyclase . In Fig . 9 , we plot the position of the barycenter of the bound receptors as a function of . This result suggests that polarization is also indissolubly crosslinked with internal amplification . A parametric constraint appears: below a non–zero threshold in the enzymatic ( that is , amplifying ) efficiency of the pathway , no polarization occurs . The above investigations show that both feedback and amplification are precursor events for the achievement of polarization . Their respective actions are necessary and concurring contributions . The results discussed above were obtained setting in the model the diffusion and feedback coefficients as indicated in Tab . 2 . While the diffusion coefficient can be measured as a well defined physical parameter , it is much more difficult to quantify the feedback parameter . This fact has a strong implication , since the ratio of the two parameters influences the dominating behavior of the system . To fix the ideas , we perform the tests varying the feedback coefficient , and we keep constant the diffusion coefficients , ( that we denote here for simplicity as ) . The results of the simulations are shown in Fig . 10 , where the abscissa of the barycenter ( in ) of the bound receptors is plotted as a function of . They suggest that under a certain threshold , diffusion overwhelms drift , leading to an unbiased receptor distribution on the membrane , that is , as if the external field were uniform . We use the model to study the time scales that characterize the process on the front and back sides , respectively . To perform this mathematical analysis , we consider a simplified version of the DCC model , neglecting diffusion and feedback terms . By doing so , we yield a system of ordinary differential equations , decoupled sector by sector . In this study , we prescribe a-priori an asymmetric receptor distribution to describe the polarized situation reached after a sufficient time of exposure to the cue . In particular , we start from the steady state distribution of receptors obtained from the simulation of the DCC model with . We compute in each sector the eigenvalues of the Jacobian matrix of the system in correspondence to its steady state . All the eigenvalues are real negative , indicating that the steady state is an attractive point . Based on the principal component of the corresponding eigenvector , we associate a chemical species with each eigenvalue . Then , using standard tracking techniques [41] , we follow the variation of each eigenvalue along the GC perimeter . In Fig . 11 , we plot the modulus of the eigenvalue associated with each species as a function of the angle . The eigenvalue associated with the slowest process on the front side ( ) appears to be connected to PKA , while on the back side ( ) it appears to be connected to . Observe that all the eigenvalues undergo a variation along the angle , even if for some of them this is not apparent in the logarithmic scale , required to appreciate the different relative behaviors . The graph shows the strong variation of the eigenvalue connected with . Moreover , the general trend of reduction of the absolute values passing from the front side to the back side indicates that the front dynamics is faster than the back one . We use the coupled DCC–UNC5 model to study the response of the system to the ( bifunctional ) netrin cue when the DCC–UNC5 complex is formed . Denoting by #DCC and #UNC the number of receptors on the membrane belonging to the two populations , respectively , we introduce the quantitywhich represents the fraction of DCC–UNC5 receptors . Neurons which display a chemoattractive response when exposed to a netrin gradient are typically characterized by low values of . Neurons genetically manipulated to overexpress the UNC5–type receptors , which display a chemorepulsive response when exposed to the same netrin gradient , are typically characterized by values of near . As a representative situation of the first case , we set . In Fig . 12 , we plot the concentration contours of the main species as a function of angular position and time . This model suggests the following explanation of the attractive behavior: DCC receptors migrate toward the source , while DCC–UNC5 ones migrate away from it , causing a differential opening of calcium channels on front vs . back . Then , we consider the dual case , setting . In Fig . 13 , we plot the concentration contours of the main species as a function of angular position and time . The model suggests a dual behavior with respect to the situation with : DCC–UNC5 receptors migrate toward the front , while DCC receptors migrate toward the back , giving rise to the repulsive response . We have proposed a mathematical model to study the polarization phenomenon triggered by the exposure of GCs to an external cue , taking as a paradigm the in vitro chemotactic assay . The key hypothesis is that symmetry breaking occurs as early as at the level of transmembrane receptors , which undergo a biased distribution after exposure to the cue . This finding appears in the recent work [19] , ( see also the analysis of [42] ) , where experimental evidence of such a process is provided and an idea is proposed of an autocatalyzing loop connecting receptors and downstream actin dynamics . Our main contribution is to quantitatively demonstrate via the mathematical model how such a loop is able to achieve GC “front–back” polarization . More precisely , we assess the fact that spatial bias in receptors is a precursor necessary event for chemotactic response , so that upstream gradient sensing and downstream cytoskeleton dynamics cannot be decoupled . Moreover , we analyze further crosslinked effects and , namely , the contribution to polarization of internal enzymatic reactions , which entail activation of a one–to–more production of molecules . The model shows that the enzymatic efficiency of such reactions must overcome a threshold in order to produce a sufficient amplification , which is another fundamental precursor step for obtaining polarization . A simplified version of the model is used to provide preliminary indications about time scales in the front and back processes , via eigenvalue tracking . The findings suggest that the two processes take place with different speeds , the front one being faster than the rear one . This element could play a significant role , even if further investigations should be carried out , A preliminary analysis of the role of diffusion vs . convection is also sketched , establishing the nature of a strongly advection–dominated system . Eventually , we have proposed an extension of the model to address a peculiar behavior arising when two families of receptors interact to produce the response to the same cue . The study of this case allows to propose some ideas on the mechanism of chemorepulsion , as the synergistic interaction of pathways , that contribute to better understand this less studied phenomenon . The present model demands for improvement . We have run simulations considering a time dependent gradient of ligand . The laboratory experiments ( among the very few on this issue for neural cells ) reported in [19 , Fig . 1g] , show reversibility of the distribution of the receptors , upon removal of the gradient . Our model tested under these conditions tends instead to maintain the receptor asymmetric configuration and removal of the cue itself is demanded . This fact has led us to conjecture the need for introducing inhibitory effects on receptors . At present , we have indeed assumed that the GC has a pool of receptors capable at each time to bind the ligand and we have not considered their physiological cycle , comprising deactivation by phosphorylation and recycling back in the membrane ( as done for example in [29] ) . Attempts have been made to consider such processes , establishing a life–time for each receptor , after which the receptor is removed and substituted by another one without bias in its angular position on the membrane ( keeping in this way the total number of receptors constant in time ) . Including these effects in the model does not seem to significantly change the results , but deeper investigations on these ideas should be carried out , possibly supported by the availability of further information from the biochemical viewpoint ( as done in the context of nonneural cells , for example , in [43] ) . To derive initial conditions for the mathematical model , which are not readily available from experiments , we start assuming the total number of molecules per cell ( indicated with the subscript ) of the following species to be conserved during the integration time , leading to , for all ( 7a ) ( 7b ) ( 7c ) ( 7d ) ( 7e ) ( 7f ) ( 7g ) Total values used in the simulations are reported in Tab . 3 . Moreover , we assume that: Using assumptions ( 1 ) , ( 2 ) and ( 3 ) , we recover from the total values the corresponding concentrations at the left hand sides of Eqs . ( 7a–g ) at . Values are in Tab . 4 . As for the boundary conditions , since the GC is modeled as a circular structure , concentrations and fluxes must coincide at and . Thus , for each species , we impose the periodic boundary conditions Several kinetic constants entering the model are available from literature references for the same reactions we are dealing with or for very similar reactions . The value of the other parameters has been estimated , based on the following considerations: Tab . 5 summarizes the value of the kinetic constants used in the model . We conclude this discussion exploring the overall influence of the parameters over the model predictions . To do this , we consider the DCC model and we perform a set of 300 trial tests prescribing a random variation of all kinetic constants . We monitor the displacement of the barycenter of the bound DCC receptors after 2 h . The results are shown in Fig . 15 , where on the axis we report the total parameter variation computed as , being the total number of parameters , the value of the parameter and its perturbed value . A percentage of the tests show a perturbation in the barycenter displacement lower than the . Both DCC and DCC–UNC5 models constitute nonlinear time dependent diffusion–advection–reaction systems of partial differential equations . Their numerical approximation is a very challenging task: to start with , we partition the GC perimeter into angular sectors ( see Fig . 1 ) and we discretize the spatial derivative operators using finite differences , with a node collocated at the center of each sector . Observe that conservation relations like Eqs . ( 7 ) do not hold sector by sector , but they rather apply to the integral on the angle ( which represents the number of molecules per cell ) . Attention must be paid to the fact that the feedback terms in Eqs . ( 4a ) and ( 5a ) dominate the diffusive terms . To avoid spurious oscillations , upwind finite differences or a sufficiently fine discretization should be adopted [45] . Once finite difference discretization is carried out , a system of coupled nonlinear ordinary differential equations is obtained . Due to the different speeds in the reaction dynamics ( refer for this issue also to the Results and Discussion ) , these systems are very stiff and require an implicit time integrator . We have adopted the ode15s Matlab routine with adaptive choice of the time integration step . Numerical evaluation of the Jacobian matrix has been used for the linearization . The Matlab software package developed by the authors can be made available upon request .	The ability of cells to respond to chemical signals present in the environment is of upmost importance for life . In the developing embryo , cells crawl along graded fields of chemical cues to aggregate into organized patterns . This process is an example of chemotaxis . It is a complex phenomenon , where external signals are transduced into internal chemical pathways leading to directional movement . Differential reorganization of the internal structure is called polarization , and it involves regulatory proteins as well as cytoskeletal elements . In this work , we propose a mathematical and computational model for the quantitative study of chemotactic pathfinding in neural cells . Our starting point is the recent finding that , for such cells , an early polarization event is the redistribution on the membrane of cue–ligated receptors , transported by the cytoskeletal structures , which act as a sort of conveyor belt . We show that this proposed mechanism , connecting in a closed loop cue sensing and cytoskeleton dynamics , is qualitatively and quantitatively adequate to produce polarization . We also investigate the role of the internal biochemical chain in producing signal amplification and its tight interlacing with polarization . An extension of the model is used to study chemotactic behaviors as the attractive/repulsive response of axons exposed to the same cue .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "computer", "science/applications", "cell", "biology/cell", "signaling", "mathematics", "neuroscience/neurodevelopment", "computational", "biology", "chemical", "biology/chemical", "biology", "of", "the", "cell" ]	2009	Autocatalytic Loop, Amplification and Diffusion: A Mathematical and Computational Model of Cell Polarization in Neural Chemotaxis
Salivary hyaluronidases have been described in a few bloodsucking arthropods . However , very little is known about the presence of this enzyme in various bloodsucking insects and no data are available on its effect on transmitted microorganisms . Here , we studied hyaluronidase activity in thirteen bloodsucking insects belonging to four different orders . In addition , we assessed the effect of hyaluronidase coinoculation on the outcome of Leishmania major infection in BALB/c mice . High hyaluronidase activity was detected in several Diptera tested , namely deer fly Chrysops viduatus , blackflies Odagmia ornata and Eusimilium latipes , mosquito Culex quinquefasciatus , biting midge Culicoides kibunensis and sand fly Phlebotomus papatasi . Lower activity was detected in cat flea Ctenocephalides felis . No activity was found in kissing bug Rhodnius prolixus , mosquitoes Anopheles stephensi and Aedes aegypti , tse-tse fly Glossina fuscipes , stable fly Stomoxys calcitrans and human louse Pediculus humanus . Hyaluronidases of different insects vary substantially in their molecular weight , the structure of the molecule and the sensitivity to reducing conditions or sodium dodecyl sulphate . Hyaluronidase exacerbates skin lesions caused by Leishmania major; more severe lesions developed in mice where L . major promastigotes were coinjected with hyaluronidase . High hyaluronidase activities seem to be essential for insects with pool-feeding mode , where they facilitate the enlargement of the feeding lesion and serve as a spreading factor for other pharmacologically active compounds present in saliva . As this enzyme is present in all Phlebotomus and Lutzomyia species studied to date , it seems to be one of the factors responsible for enhancing activity present in sand fly saliva . We propose that salivary hyaluronidase may facilitate the spread of other vector-borne microorganisms , especially those transmitted by insects with high hyaluronidase activity , namely blackflies ( Simuliidae ) , biting midges ( Ceratopogonidae ) and horse flies ( Tabanidae ) . Hyaluronidases are a family of enzymes that degrade hyaluronan ( HA ) and several other glycosaminoglycan constituents of the extracellular matrix of vertebrates ( for review see [1] ) . In insects , hyaluronidases are well-known from venoms of Hymenoptera and represent clinically important allergens of honey-bees , wasps and hornets [2]–[4] . Hyaluronidases were found also in cDNA libraries of salivary glands ( sialomes ) of various bloodsucking insects [5]–[8] and the enzyme activity was found in saliva of three groups of Diptera , namely sand flies , blackflies , and horse flies [9] , [10] . Salivary hyaluronidases of parasitic insects may have diverse effects on the host . They play an important role in blood meal acquisition; by degrading HA abundant in host skin , hyaluronidases increase tissue permeability for other salivary components that serve as antihaemostatic , vasodilatory or anti-inflammatory agents [5] , [9] . This is why hyaluronidases are frequently called “spreading factors” [11] . The enzyme activity facilitates the enlargement of the feeding lesion and the insect acquires the blood meal more rapidly . In addition , HA fragments were shown to have immunomodulatory properties; they affect maturation and migration of dendritic cells , induction of iNOS and chemokine secretion by macrophages and proliferation of activated T cells ( reviewed in [12] ) . As blood sucking insects represent the most important vectors of infectious diseases , local immunomodulation of the vertebrate host may positively enhance the infection . Leishmaniasis is one of the most prevalent vector-borne diseases . It is initiated by the intradermal inoculation of Leishmania promastigotes during the bite of an infected sand fly ( Diptera: Phlebotominae ) . As shown first by Titus and Ribeiro [13] saliva of the sand fly vector exacerbates the initial phase of Leishmania infections in terms of parasite burden and size of the cutaneous lesion . Sand fly saliva was described to contain an array of pharmacologically active compounds affecting host hemostasis and immune mechanisms ( reviewed in [14] , [15] ) but the information about molecules responsible for the exacerbating effect is still very limited . Morris et al . [16] showed that maxadilan , a well-known vasodilator of the New World vector Lutzomyia longipalpis , exacerbates Leishmania infection to the same degree as whole saliva . Maxadilan inhibits splenocyte proliferation induced in vitro and delayed type hypersensitivity in mice [17] and it also has several inhibitory effects on macrophages and monocytes that would support Leishmania survival in the host [18] . However , this important peptide was not found in Old World vectors of genus Phlebotomus ( www . ncbi . nih . gov ) , including P . papatasi where exacerbating effect of saliva was repeatedly demonstrated [19] , [20] . The vasodilatory activity of P . papatasi was instead ascribed to adenosine and AMP present in saliva of this sand fly [21] . In the present work , we studied hyaluronidase activity in bloodsucking insects of four different orders . In addition , we assessed the effect of hyaluronidase coinoculation on the outcome of Leishmania major skin lesions and spreading into draining lymph nodes . Samples used are summarized in Table 1 . The insects originated from laboratory colonies or were collected in the wild . Salivary glands were dissected out in Tris buffer ( 20 mM Tris , 150 mM NaCl , pH 7 . 8 ) and stored in batches ( usually 20 glands in 20 µl of Tris buffer ) at −70°C . Where dissection of salivary glands was not feasible , whole bodies ( Ctenocephalides flea , Culicoides midge ) or the thoracic parts containing salivary glands ( Pediculus louse ) were used at protein concentration 20 µg/µl . Salivary gland extracts ( SGE ) or body extracts ( BE ) were obtained by disruption of tissue by three freeze-thaw cycles in liquid nitrogen , homogenization and centrifugation at 12 , 000 g for 5 min . Protein concentration was determined by Bradford assay using bovine serum albumin in Tris buffer as a standard . Enzyme activity was detected by the dot method on 10% polyacrylamide gels with copolymerized hyaluronic acid ( HA , potassium salt , from human umbilical cord , ICN Pharmaceutical , CA ) . Gels were prepared using 0 . 1 M acetate , pH 5 . 5 , containing 0 . 1 M NaCl , 0 . 05% Tween-20 and 0 . 002% HA . This method was previously proved as sensitive and reproducible [10] . Preliminary experiment with selected salivary extracts revealed that Phlebotomus papatasi and Culex pipiens samples were positive at pH 4 . 5 , 5 . 5 , 6 . 5 and 7 . 5 while Aedes aegypti , Anopheles stephensi and Glossina fuscipes samples were consistently negative ( Fig . S1 ) . Therefore pH 5 . 5 was chosen for this assay as this is about the pH optimum known for salivary hyaluronidases of various Diptera [9] , [10] . Insect samples ( 2 µl volume ) were dotted on the gel and sheep testicular hyaluronidase , ( Sigma , 1 µg in 1 µl ) was used as a control . Incubation was carried out for 24 hrs at 37°C in a moist chamber . The gels were then washed in water , soaked in 50% formamide for 30 min and stained in Stains-all ( Sigma ) solution ( 100 µg/ml in 50% formamide ) for 24 hrs in the dark . After a rinse in distilled water the gels were scanned and photographed . To determine whether the enzyme activity was specific for cleaving HA , we tested positive samples also with another component of extracellular matrix , chondroitin sulfate . The method was performed as described above , only HA was replaced by 0 . 002% chondroitin sulfate ( Sigma ) . Electrophoresis ( SDS PAGE ) was carried out on 10% slab gels ( 0 . 75 mm thick ) using Mini-Protean II apparatus ( Biorad ) and constant voltage 150 V . Substrate gels were copolymerized with 0 . 002% HA . As the hyaluronidase activities and band patterns varied among insects , different loads were used per lane in order to obtain bands of equal intensity . Following electrophoresis , gels were rinsed 2×20 min in 0 . 1 M Tris , pH 7 . 8 , 20 min in 0 . 1 M acetate buffer , pH 5 . 5 ( both with 1% Triton X-100 to wash out SDS ) and then incubated in 0 . 1 M acetate buffer ( without detergent ) for 120 min at 37°C . After rinsing in water the gels were stained with Stains-all as described above . Hyaluronidase activity was visible as a pink band on a dark blue background . Experiments on mice were done in accordance with Czech Act No . 246/1992 and approved by IACUC of the Fac . Sci . , Charles University in Prague . A mouse ear infection model [19] was used to assess the effect of hyaluronidase coinoculation on the outcome of Leishmania infection . Leishmania major clone LV561 ( MHOM/IL/67/LRC-L137 Jericho II ) was cultured on blood agar from defibrinated rabbit blood , supplemented with 50 µg/ml gentamicin . Female BALB/c mice ( Charles River Deutschland , Sulzfeld , Germany ) were used at the age of 8 weeks . Ether-anaesthetized mice were inoculated in the ear dermis with 104 or 105 L . major stationary-phase promastigotes ( subculture 1 ) in 5 µl sterile saline . The inoculum also contained bovine testicular hyaluronidase ( Sigma ) in an amount equivalent to 2 or 10 “optimal salivary glands” of Phlebotomus papatasi [10] , i . e . 0 . 4 and 2 . 0 relative turbidity reducing units , respectively . Bovine testicular hyaluronidase belongs to the same enzyme class as the sand fly salivary hyaluronidases [10] and shares sequence homology with the enzyme of L . longipalpis [5] . Control animals were inoculated with parasites in sterile saline only . Sixty mice ( 10 for each of six groups ) were used for Q-PCR and another 48 ( 8 for each group ) for lesion monitoring . The size of skin lesions was measured weekly using a Vernier caliper gauge . Lesions were monitored for 6 weeks post infection: the area was calculated from two perpendicular measurements as an ellipse area , and its appearance ( degree of ulceration ) was assessed using an arbitrary scale from 1 to 5 ( 1 - low induration , 2 - high induration , 3 - small ulcer , 4 - large ulcer , 5 - perforated ear pinna ) . Independently in both parasite doses ( 104 and 105 ) , the significance of the hyaluronidase effect was tested using nonparametric Kruskal-Wallis ANOVA and post hoc comparisons of mean ranks using Statistica 7 routines [22] . The tests were performed separately for weeks 3 , 4 , 5 , and 6 post-infection; the size of a lesion was calculated as its area weighted by the degree of ulceration . Mice were sacrificed 24 hrs post inoculation ( p . i . ) as the preliminary experiment revealed that lymph nodes of mice dissected 24 hours p . i . gave more consistent results than those dissected 48 hours p . i . ( Fig . S2 ) . Parasite numbers in draining retromaxillar lymph nodes were determined by quantitative PCR ( Q-PCR ) as described earlier [23] . Briefly , dissected lymph nodes were stored in 10 µl saline at −70°C . Total DNA was isolated from homogenised samples using High Pure PCR Template Preparation Kit ( Roche ) ; kinetoplast DNA was targeted using primers described elsewhere [24] . The relative effectiveness of three hyaluronidase doses ( equivalent to 0 , 2 , and 10 P . papatasi salivary glands ) with both infection doses ( 104 and 105 parasites ) was evaluated by analysis of variance ( Statistica v . 7 . 1 , factorial and one-way ANOVA ) . The dot method on gels with copolymerized HA and chondroitin sulfate was used to study the presence of hyaluronidase activity and its substrate specificity . The highest hydrolysis of HA was observed in SGE of deer fly Chrysops viduatus . Pronounced hydrolysis was found in SGEs of blackflies Odagmia ornata and Eusimulium latipes , mosquito Culex quinquefasciatus , sand fly Phlebotomus papatasi and whole body extract of biting midge Culicoides kibunensis ( syn . C . cubitalis ) . Lower activity was detected in BE of cat flea Ctenocephalides felis ( Fig . 1 ) . On the other hand , no detectable hydrolysis of HA occurred in SGEs of kissing bug Rhodnius prolixus , mosquitoes Anopheles stephensi and Aedes aegypti , tse-tse fly Glossina fuscipes , stable fly Stomoxys calcitrans and in thoracic extracts of human louse Pediculus humanus ( Fig . 1 ) . Positive samples were then tested also for chondroitin sulfate hydrolysis ( Fig . 2 ) . High activity was observed in Culex quinquefasciatus and Culicoides kibunensis , in other samples the hydrolysis of chondroitin sulfate was either moderate ( Chrysops viduatus ) or low ( Phlebotomus papatasi , Ctenocephalides felis ) ( Fig . 2 ) ; clearly , HA is the preferred substrate for the enzymes of these three insects . Seven samples positive in the dot method were analyzed by zymography to reveal the apparent molecular weight ( MW ) of hyaluronidases . The MW of the enzymes differed among various insects ( Figs . 3 and 4 ) . Under nonreducing conditions hyaluronidases were detected as major diffuse bands ( Fig . 3 ) . The SGE activity in Phlebotomus papatasi had a MW about 70 kDa while those in both blackfly species tested , Eusimulium latipes , and Odagmia ornata , about 40 kDa . In BE of Culicoides kibunensis , the major band of about 35 kDa was accompanied with a minor one of 70 kDa , supposedly a dimer . Chrysops viduatus SGE revealed one major band with estimated MW of 50 kDa . In BE of flea Ctenocephalides felis , three enzyme bands were detected , the most prominent one of about 52 kDa ( Fig . 3 ) . Under reducing conditions , SDS PAGE revealed sharper enzyme bands allowing more precise assignment of corresponding MW ( Fig . 4 ) . In sand fly P . papatasi , both blackfly species and deer fly Chrysops viduatus , hyaluronidase activity was observed within the same MW ranges as under nonreducing conditions ( 70 , 40 kDa , and 50 kDa , respectively ) . In Culicoides kibunensis and Ctenocephalides felis hyaluronidase activity was not detectable under reducing conditions ( Fig . 4 ) . No hyaluronidase activity was detected in Culex quinquefasciatus SGE under either zymography conditions used , reducing and nonreducing . An additional experiment was performed to explain the contradictory results from the dot method and zymography; SGE of C . quinquefasciatus was dotted on the gel with copolymerized HA with and without the presence of SDS . Hydrolysis was observed only in the sample without SDS ( Fig . 5 ) . Next we examined whether hyaluronidase altered the course of Leishmania major infection in BALB/c mice . We used intradermal inoculation into the ear and the disease burden was calculated from weekly measuring the lesion size . As shown in Fig . 6 , mice coinjected with parasites and hyaluronidase developed bigger lesions . In all groups of mice , the onset of lesion development was at three weeks p . i . Thereafter , the lesions grew faster in coinoculated groups . The experiment was terminated six weeks post infection when , in some animals , ulcerating lesion spread over the majority of ear pinna . In mice inoculated by higher parasite numbers ( 105 ) , both hyaluronidase treatments produced similar effects ( Fig . 6A ) . In mice with an inoculation dose one order of magnitude lower ( 104 ) , the effect of hyaluronidase was concentration-dependent: lesions were bigger in mice coinoculated with hyaluronidase activity equivalent of 10 P . papatasi salivary glands than in those coinoculated with equivalent of 2 glands ( Fig . 6B ) . In both parasite numbers ( 104 and 105 ) over all considered weeks ( 3 to 6 ) post-inoculation , Kruskal- Wallis ANOVA showed significant differences among hyaluronidase treatments ( p always≤0 . 025 ) , with only one exception in week 3 of 104 parasites treatment ( p = 0 . 23 ) . Consequently , the post-hoc comparison of treatments tests confirmed the significant difference between controls ( no hyaluronidase ) and corresponding inoculated hyaluronidase doses ( 2 or 10 glands equivalents ) . We also tested the difference between the 2 and 10 gland equivalent doses: however , despite the common trends apparent in Fig . 6 indicating that there may be a systematic difference between 2 and 10 gland equivalents doses , the post-hoc comparison of treatments test did not prove it in any case but in week 5 of the 104 parasites treatment . We also examined whether hyaluronidase affected Leishmania major load in draining lymph nodes of BALB/c mice one day p . i . Using Q-PCR , no significant differences were observed among control and experimental groups of mice at both parasite doses ( 104 or 105 L . major ) tested ( F ( 2 , 54 ) = 0 . 043; p = 0 . 96 ) ( Fig . S3 ) . Parasitic insects utilize two strategies for finding blood: solenophagy ( or vessel feeding ) and telmophagy ( or pool feeding ) . In solenophagic approach , the feeding fascicle cannulates a blood vessel , while in the pool-feeding mode the mouth part stylets slash through the skin , and the insect sips blood that oozes out from the hemorrhage . In our experiments , pronounced hyaluronidase activity was found in black flies , biting midges , sand flies and deer flies . All these insects belong to parasitic Diptera with pool-feeding mode of blood meal acquisition . The activity was detected also in cat flea ( Ctenocephalides felis , Siphonaptera ) and in Culex quinquefasciatus mosquito ( Diptera ) . Although these two species belong to different insect orders , they are both vessel feeders . In contrast , no activity was detected in other vessel-feeding insects: human lice , kissing bugs , Anopheles and Aedes mosquitoes , tsetse flies , and stable flies . Hyaluronidase activity was previously detected in the saliva of various sand fly species [9] , [10] as well as in the saliva of the black fly Simulium vittatum [9] and horse fly Tabanus yao [8] . Sequences predicted to code for hyaluronidases were found in the salivary transcriptomes of the mosquito Culex quinquefasciatus [6] and the biting midge Culicoides sonorensis [7] . Herein , we demonstrated that Culex quinquefasciatus and Culicoides kibunensis possess hyaluronidase activity and , in parallel experiments , we detected hyaluronidase activity in saliva of two other species of biting midges Culicoides sonorensis and C . nubeculosus ( Volfova et al . , unpublished ) . Therefore , we showed that in biting midges and in Culex quinquefasciatus , the transcripts coding for putative hyaluronidases are translated into functional enzymes . To determine whether the enzyme activity was specific for cleaving HA , we also tested another component of mammalian extracellular matrix , chondroitin sulfate . All hyaluronidase-positive samples tested cleaved chondroitin sulfate , which would indicate that insect hyaluronidases fall into the same class as mammalian hyaluronidases ( E . C . 3 . 2 . 1 . 35 according to IUBMB Enzyme Nomenclature ) [25] . Indeed , sequence analysis of transcripts putatively coding for hyaluronidase enzymes reveals their homology to mammalian enzymes [6] , [7] , [9] . While HA was found as the preferred substrate for most samples tested , very high hydrolysis of chondroitin sulfate was found in Culex quinquefasciatus SGE . This mosquito species differs from other samples tested also in other aspects . In zymography assay , salivary hyaluronidase of Culex quinquefasciatus was irreversibly sensitive to denaturation effect of SDS while enzymes of other insects tested refolded and regained activity after removal of the denaturating agent . Further work is needed to understand the differences in the molecular structure and substrate specificity of hyaluronidases from Culex mosquitoes versus other bloodsucking insects . In addition , in other mosquitoes studied , Anopheles darlingi [26] , funestus [27] and gambiae [28] and Aedes aegypti [29] and albopictus [30] , neither hyaluronidase activity nor hyaluronidase gene was found in salivary transcriptomes . Sequences homologous to hyaluronidase were , however , found by genome sequencing in Anopheles gambiae [31] and Aedes aegypti [32] . As revealed by zymography , hyaluronidases of different insect species tested vary substantially in MW and the structure of the molecule . Putative oligomers were seen in Culicoides kibunensis . Oligomeric forms have been found frequently among mammalian hyaluronidases . In sand flies , oligomers or dimers were found in Lutzomyia longipalpis , Phlebotomus papatasi , and P . sergenti [10] . Multiple bands observed by zymography in Ctenocephalides felis whole body extract could , however , represent multiple hyaluronidase enzymes . In Culicoides and Ctenocephalides , reducing conditions affected the stability of the enzymes; 2-mercaptoethanol inhibited hyaluronidase activity . In Culicoides midges , the sensitivity of hyaluronidase activity to reducing conditions was confirmed by experiments with pure saliva of laboratory bred Culicoides sonorensis and C . nubeculosus ( Volfova et al . , unpublished ) . This implies that reduction-sensitive residues are either important for the function of the active site of the enzyme , or steric relations in the molecule . On the other hand , hyaluronidases of other insects tested , namely Phlebotomus papatasi , Eusimulium latipes , Odagmia ornata , and Chrysops viduatus , remained active under reducing conditions . Addition of 2-mercaptoethanol did not result in differences in the apparent MW , suggesting that the enzymes consist of a single polypeptide chain . These results correspond with previous observations on sand flies; sand fly hyaluronidases strikingly differed in structure and sensitivity to reducing conditions , even among various species of the genus Phlebotomus [10] . We showed that hyaluronidase is a common constituent of saliva of bloodsucking insects . It seems to be essential for insects with pool-feeding mode , where it facilitates the enlargement of the feeding lesion , serving as a spreading factor for other pharmacologically active compounds present in saliva . Very little is known , however , about the possible role of salivary gland hyaluronidase in allergic reactions which occur in some patients after repeated bites of bloodsucking Diptera . In Hymenoptera , venom hyaluronidase is largely responsible for the cross-reactivity of venoms with sera of allergic patients [4] . In several patients , coexistent anaphylaxis to Hymenoptera sting and Diptera bite was described [33] and hyaluronidase is a candidate allergen responsible for this type of crossreactions . In experiments of Sabbah et al . [34] , [35] , IgE of allergic patients recognized shared proteins within MW range 44–50 kDa between wasp venom and total extracts of mosquito and horse fly . Unfortunately , these interesting data are difficult to assess given the incomplete identitification of the mosquito and horse fly species tested . A mouse ear infection model was used to assess the effect of hyaluronidase coinoculation on the outcome of Leishmania major infection . The activity of sand fly enzyme was mimicked by commercially available bovine hyaluronidase . More severe lesions developed in mice where L . major promastigotes were coinjected with hyaluronidase . Even the lower dose of the enzyme corresponding with the activity produced by 1–2 sand fly females resulted in significant differences against the control mice where parasites alone were injected . It would be worth testing if differences observed in lesion size are mainly due to number of parasites or to inflammatory response to coinoculated hyaluronidase . In contrast , there was neither more rapid onset of lesions , nor faster dissemination of Leishmania in the lymph node . Parasite numbers in draining lymph nodes collected 24 and 48 hrs p . i . were similar in all experimental groups . Although hyaluronidase activity exacerbated Leishmania lesions in the skin , it did not support its visceralization . However , we can not exclude the possibility that consequences of hyaluronidase for parasite visceralization are not immediate and thus could not be detected in the present study . The way by which hyaluronidase enhances the establishment of Leishmania is unknown , but we suggest that it is due to HA fragments generated by hyaluronidase activity in the host skin . HA occurs in two main forms: the high MW ( HMW ) polymers and the low MW ( LMW ) fragments . HMW HA is a common component of vertebrate extracellular matrix . LMW HA fragments are generated under inflammatory conditions by endogenous or bacterial hyaluronidases [36] , or non-enzymatically by free radicals [37] . HA fragments have diverse immunomodulatory properties; they affect DC maturation , T cell proliferation , cytokine , and chemokine synthesis by lymphocytes and macrophages ( reviewed in [12] ) . Thus , following injury or infection , HA fragments have been implicated as both endogenous and exogenous triggers of repair and/or defense mechanisms [38] , [39] and might truly represent a “danger signal” [40] . Leishmania parasites , however , may profit from the local increase of HA fragments . Specifically , endothelial cells were shown to respond to LMW HA by IL-8 production [38] that results in neutrophil recruitment . As neutrophil granulocytes were indicated as Trojan horses enabling Leishmania silent entry into macrophages [41] their accumulation at the site of sand fly bite might promote infection establishment . In conclusion , we demonstrated that hyaluronidase promotes Leishmania establishment in murine skin . As this enzyme is present in all Phlebotomus and Lutzomyia species studied to date [10] it seems to be one of the factors responsible for enhancing activity present in saliva of the New-World as well as the Old-World sand flies . We propose that hyaluronidase , in concert with other insect-derived molecules , may facilitate the spread of other vector-borne diseases , especially those transmitted by vectors with high hyaluronidase activity in saliva , namely blackflies , biting midges , deer flies and horse flies .	Hyaluronidases are enzymes degrading the extracellular matrix of vertebrates . Bloodsucking insects use them to cleave the skin of the host , enlarge the feeding lesion and acquire the blood meal . In addition , resulting fragments of extracellular matrix modulate local immune response of the host , which may positively affect transmission of vector-borne diseases , including leishmaniasis . Leishmaniases are diseases with a wide spectrum of clinical forms , from a relatively mild cutaneous affection to life-threatening visceral disease . Their causative agents , protozoans of the genus Leishmania , are transmitted by phlebotomine sand flies . Sand fly saliva was described to enhance Leishmania infection , but the information about molecules responsible for this exacerbating effect is still very limited . In the present work we demonstrated hyaluronidase activity in salivary glands of various Diptera and in fleas . In addition , we showed that hyaluronidase exacerbates Leishmania lesions in mice and propose that salivary hyaluronidase may facilitate the spread of other vector-borne microorganisms .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "infectious", "diseases/protozoal", "infections", "infectious", "diseases/neglected", "tropical", "diseases", "infectious", "diseases/skin", "infections" ]	2008	Hyaluronidase of Bloodsucking Insects and Its Enhancing Effect on Leishmania Infection in Mice
The liver is the central organ for detoxification of xenobiotics in the body . In pharmacokinetic modeling , hepatic metabolization capacity is typically quantified as hepatic clearance computed as degradation in well-stirred compartments . This is an accurate mechanistic description once a quasi-equilibrium between blood and surrounding tissue is established . However , this model structure cannot be used to simulate spatio-temporal distribution during the first instants after drug injection . In this paper , we introduce a new spatially resolved model to simulate first pass perfusion of compounds within the naive liver . The model is based on vascular structures obtained from computed tomography as well as physiologically based mass transfer descriptions obtained from pharmacokinetic modeling . The physiological architecture of hepatic tissue in our model is governed by both vascular geometry and the composition of the connecting hepatic tissue . In particular , we here consider locally distributed mass flow in liver tissue instead of considering well-stirred compartments . Experimentally , the model structure corresponds to an isolated perfused liver and provides an ideal platform to address first pass effects and questions of hepatic heterogeneity . The model was evaluated for three exemplary compounds covering key aspects of perfusion , distribution and metabolization within the liver . As pathophysiological states we considered the influence of steatosis and carbon tetrachloride-induced liver necrosis on total hepatic distribution and metabolic capacity . Notably , we found that our computational predictions are in qualitative agreement with previously published experimental data . The simulation results provide an unprecedented level of detail in compound concentration profiles during first pass perfusion , both spatio-temporally in liver tissue itself and temporally in the outflowing blood . We expect our model to be the foundation of further spatially resolved models of the liver in the future . The liver is the main site of metabolization and detoxification of xenobiotics in the body of mammals . Compounds delivered by blood flow through the portal vein and the hepatic artery are continuously processed within hepatic cells , such that foreign and potentially harmful compounds can be cleared from the blood . Metabolization by enzyme-catalyzed biotransformation produces chemical alterations of the original compounds , thereby enabling their elimination . A second , complementary process is the active secretion to the bile duct from which the compound is further transported to the gastrointestinal tract . In pharmacology and medicine , plasma clearance is used to quantify the rate by which a compound is eliminated from the body [1] . Plasma clearance describes the overall detoxification capacity of the organism and summarizes individual clearance rates from all eliminating organs with the largest contribution coming from the kidney and the liver . While renal clearance can be measured by urinary secretion , a quantification of liver detoxification capacity is difficult since the different physiological functions cannot be assessed directly . In particular , the relative contributions of the different underlying physiological functions such as metabolization or biliary secretion cannot be adequately differentiated since the liver is frequently rather considered as a net systemic sink . While hepatic turnover can be indirectly quantified with drugs following a known pharmacokinetic profile , the local , time-resolved distribution of compounds within the whole organ can generally not be analyzed even with distinguished measurement techniques . This holds in particular for the first pass of drug perfusion in a liver directly after drug administration , when hepatic tissue is exposed to a novel xenobiotic for the first time . Due to the pivotal role of the liver in drug pharmacokinetics and detoxification , several models quantifying hepatic clearance have been developed before [2] . These include tube models for representative hepatic sinusoids ( single tubes , in parallel or in series ) [3] , dispersion liver models [4] , fractal liver models [5] , circulatory models [6] , [7] including zonal models [8] , [9] , and distribution-based models describing statistical variations in transition times [10] . For a more detailed overview , we refer to [11] . Generally , PK models are well suited to investigate distribution and clearance of compounds in the body . In compartmental PK modeling , a limited number of rather generic compartments is usually used to simulate plasma concentration and drug clearance . Following a complementary approach , physiologically based pharmacokinetic ( PBPK ) models describe biological processes at a large level of detail based on prior physiological knowledge [12] . This involves amongst others organ volumes and blood flow rates , such that physiological mechanisms underlying absorption , distribution , metabolization and excretion of compounds can be explicitly described . While most approaches consider intravenous or oral application of therapeutic compounds , PK models describing further uptake routes such as inhalation have also been developed [13] , [14] . Organs in PBPK models are divided in several subcompartments . So-called distribution models are used to describe the mass transfer between these subcompartments which are quantified based on physicochemical compound information such as lipophilicity or the molecular weight . Basic PBPK models can be extended to include enzyme-mediated metabolization or active transport across membranes [15] . Coming from the field of toxicology [16] , PBPK models are nowadays routinely used in drug discovery and development [17] . They are for example applied in pediatric scaling [18] , model-based risk assessment [19] , as well as for multiscale modeling by integrating detailed models of intracellular signaling [20] or metabolic networks [21] into the cellular subcompartment , thus allowing for the analysis of cellular behavior within a whole-body context [22] . Notably , each organ in PBPK modeling is generally represented by few well-stirred subcompartments , thus allowing a quantitative description of drug pharmacokinetics once an equilibrium between the vascular system and the surrounding tissue has been reached . However , a spatially resolved description of drug perfusion in the whole organ covering particularly the first instants after drug administration is impossible due to the inherent well-stirred assumption . To mechanistically describe first pass perfusion , we here present a spatio-temporal model of drug distribution and metabolization in a mouse liver . The model represents liver lobes at the spatial length scale of lobuli such that physiological and molecular processes can be simulated in great detail . Our combined spatially resolved model ( cf . overview in Figure 1 ) involves three main building blocks . These comprise physiological vascular structures , an organ-scale perfusion model describing blood flow and advection of compounds , and finally pharmacokinetic models translated to the spatially resolved tissue . Geometrically accurate models of murine hepatic vascular structures were obtained by using in vivo micro-CT imaging [23] . The mass balance within the tissue was quantified based on equations coming from PBPK modeling . Our combined model was inspired by [24] , where a lobule-scale perfusion model in more physical detail and also allowing for deformation of the porous medium is introduced . A model for cardiac perfusion using very similar modeling techniques is presented in [25] . It considers multiple geometric scales , but no draining vascular system and no metabolization . Our spatially resolved model covers several scales of biological organization displayed at varying levels of resolution . The scales range from the organ level to the cellular space where metabolization and molecular transport take place . The vascular systems form the scaffolds which links the hepatic in-flow to the sinusoidal space and thereby to the lobulus level . The model considers one supplying and one draining vascular system ( denoted by SVS and DVS , respectively ) , with a homogenized hepatic space ( denoted by HHS ) in between as a tissue representation . The HHS includes in particular the sinusoids which are not explicitly resolved as vascular structures . In the latter , blood flow is represented by a fluid transport model [26] . Microcirculation and microanatomy [27] are only considered in effective form . While more detailed perfusion and metabolization models on the lobular scale [28] , [29] or the tissue-level [2] have been developed before , the representation of the HHS as a porous medium [30] is sufficient for our needs . Experimentally , the combined model corresponds to an isolated perfused liver [31] , [32] , since recirculation of blood is not considered . The resolution of the model allows calculating local concentration profiles within the tissue which can for example be used to address heterogeneous phenomena such as spatially varying distributions of lipid droplets in steatotic livers . Spatial heterogeneity of pharmacokinetic parameters such as metabolic capacity can be taken into account . Likewise local exposure profiles of toxic compounds can be simulated such that off-target effects of drug therapy can be analyzed at a high level of resolution . Applications of spatially resolved perfusion and metabolization modeling include optimized design of therapeutic treatment where spatio-temporal perfusion effects are of relevance , e . g . hypothermic machine perfusion [33] or islet cell transplantations [34] . Moreover , such models can be used for planning drug delivery for which spatially heterogeneous distribution is an important property and crucial for administration design itself . Two major examples are intrahepatic injection of chemotherapeutics or radionuclides ( see e . g . [35] ) , in particular in combination with optimization of catheter placement [36] , and targeted drug delivery [37] where drugs are injected in bound form and released at the desired location by mild hyperthermia . Likewise , our model may support data processing and interpretation in imaging or diagnostics . We expect the spatially resolved model presented here to be the foundation stone of further mechanistic models describing the spatial organization of the liver in an unprecedented level of physiological detail . The animal experiment was reviewed and approved by the local authorities ( NRW LANUV , permit number 10576G1 ) according to German animal protection laws . In the methods to be presented , we follow a geometric view from coarse to fine , i . e . from ( 1 ) the body scale ( providing organ input and output ) via ( 2 ) the vascular structures on the organ scale ( perfusion only ) to ( 3 ) the tissue scale ( perfusion and metabolization ) . Models for steatosis and -induced liver necrosis are subsequently presented to demonstrate how our spatially resolved simulations can be used for the analysis of pathological states influencing drug distribution in the liver . Finally , some aspects of computational resolution are addressed . Our spatially resolved model distinguishes between the supplying vascular tree , the HHS and a draining vascular tree which are considered in series ( Figure 2 ) . For reasons of simplicity , the supplying vascular system comprises both the portal vein and the hepatic artery . Since an isolated perfused liver was considered here which explicitly excludes recirculation through the body , the respective contributions of the portal vein and the hepatic artery were not distinguished and only the total blood inflow was taken into account . The draining vascular system represents the hepatic vein . Bile ducts were not considered in our model , since their geometric structure could not be resolved experimentally . In our combined model , the HHS is composed of several subspaces in analogy to the liver compartment in PBPK models . The latter is divided in four subcompartments , i . e . red blood cells ( rbc ) , plasma ( pls ) , interstitium ( int ) , and liver cells ( cell ) . Those four subcompartments are also considered as subspaces of the HHS , in addition a fifth remaining subspace ( rest ) is taken into account . This remaining subspace comprises all those parts inside the liver that are not considered for perfusion , metabolization and active transport , in particular the larger and explicitly resolved vascular structures and the bile ducts . The plasma subspace only refers to the blood plasma . For notational convenience , the sinusoidal subspace ( sin ) is defined as the combination of red blood cells and plasma subspace , thus representing a whole-blood compartment . The sinusoidal subspace is subject to advection , thereby reflecting blood flow through the tissue . The actual metabolization takes place only in the cellular subspace and is part of the PBPK equations that also model the exchange between the HHS subspaces . The vascular trees are resolved separately and considered for pure advection . The volume fractions of the subspaces relative to the overall liver volume , also needed for the mass balance in the compartment models , are denoted by . For our simulations in a mouse liver , we use the values ( 1 ) Volume fractions were obtained by setting to cover the fraction of the vascular volume inside the segmented liver volume , both determined from the experimental image data as described below . The values for the other subspaces from [38] were then scaled accordingly by so that all five volume fractions sum up to . From Equation 1 , we immediately obtain . For the perfusion part in our model , we address how molar concentrations of compounds change due to advection through the vascular systems and the sinusoidal HHS subspace . The exchange with the remaining HHS subspaces and cellular metabolization are considered as a separate contribution to our combined model . The body scale determines the total perfusion of the liver in mice [38] . The blood flow into and out of the root edges of the supplying and draining vascular system , respectively , is the connection of the HHS to the surrounding organism . More precisely , inflowing and outflowing molar concentrations of the compounds of interest are the main model input and output quantities . The HHS is modeled as a porous medium representing the effective behavior of the whole liver volume , with the sinusoidal subspace being perfused according to Darcy's law [51] . The perfusion is split between the red blood cells and plasma subspaces ( see Figure 2 ) proportional to their respective volumes . Using appropriate flow source and sink terms in the HHS corresponding to the exchange with the vascular systems , flow velocities for 3D advection in the HHS are obtained . The advection of concentrations can then be calculated from the given flow velocities using appropriate discretizations described in this section . Technical details about the discretization and implementation in the model are presented in Section 2 in Text S1 . Finally , the pointwise exchange in the spatially resolved model between different HHS subspaces and the actual metabolization are modeled by equations coming from PBPK modeling . The final spatially resolved model can also be used for analysis of pathophysiological states of the liver which have not been taken into account during model establishment itself . We here considered the case of steatosis leading to changes in intracellular lipid content as well as carbon tetrachloride ( ) -induced liver necrosis affecting the spatial composition of the organ . Describing pathophysiological changes in spatially heterogeneous patterns is a key strength of our approach . A comparison of the simulation results with experimental data allows to evaluate model validity , thereby providing targeted suggestions for model extensions and modifications . Steatosis is a common liver disease often caused by obesity or alcohol abuse [57] . It is characterized by lipid accumulations in the cellular subspace [58] , the influence of which can be structurally represented in the model . We here analyzed to what extent steatosis affects hepatic clearance following changes in intrahepatic drug distribution . For our simulations we consider data reported from rats in [59 , Table 8] , namely steatosis extents of about and ( mean SD ) in the left lateral and median liver lobe , respectively , obtained after two weeks of a specific diet . We proceed assuming that similar steatosis patterns can also exist in mice . Let be the ratio of lipid accumulation per liver volume at position , corresponding to the steatosis percentages in [59 , Table 8] . Using the lobe decomposition of our mouse liver dataset ( cf . Figure 3 ) , we consider two states of steatosis . First , we use a homogeneous lipid accumulation throughout the liver . This value is obtained as the average for in the left lateral lobe and in the remaining lobes as the left lateral lobe in our case has of the total volume . Second , we assign a pseudo-randomly varying value uniformly distributed in ( left lateral lobe ) and ( remaining lobes ) to obtain a spatially heterogeneous steatosis pattern . To avoid unphysiologically large local variations , we generated random numbers [60] on a grid four times coarser than the computational resolution and interpolated multilinearly at the nodes actually used . The two steatotic cases are visualized in Figure 4 . In this setting , corresponds to the healthy state [12] . We quantify the impact of steatosis ( an increased intracellular volume fraction of lipids ) by changing in cellular partition coefficient in the PBPK model . Any other effects of steatosis are explicitly omitted here for the sake of simplicity of this proof of concept for a spatial heterogeneity . The cellular partition coefficient is calculated according to the formula [12] ( 10 ) with a constant specific for the respective compound and ( Table 1 ) . The values are substituted in Equations 7 and 8a/8b . As varies spatially , also and thus intracompartemental exchange and metabolization vary accordingly . This is in contrast to commonly used PBPK models that , due to their compartmental organ representation , cannot distinguish between the homogeneous and heterogeneous case as they only use one constant value of . As a second example for pathophysiological changes in the liver , we consider the case of -induced hepatic injury . Administration of in animal models is a frequently used experimental protocol to investigate the processes underlying toxic liver damage [61] . Inducing hepatic injury by leads to necrotic cell death in the pericentral zone , similar to acetaminophen overdoses [62] . We analyzed the impact of pericentral necrosis on hepatic metabolization capacity . In our spatially resolved model , -induced necrosis was represented by replacing the cellular space by interstitial space in the necrotic volume . The latter was set to be the of the liver volume closest to the DVS terminal edges ( see Figure 4 ) , where the percentage is based on the area analysis in [63 , Table 1] , observed one day after administration . Our basis for choosing computational resolutions is the actual size of lobuli in mice . From a cross-section area of , a radius of ( assuming a regular hexagonal shape ) , both from [63 , Table 1] , and assuming the same elongation ( length divided by diameter ) of as for human lobuli [64 , Chapter 2 . 5] , a mouse lobulus has a volume of approximately , the total liver volume of thus corresponds to lobuli . By definition , a lobule is the volume drained by one terminal edge of the hepatic vein , so a fully resolved vascular tree has approximately that many leaf nodes . As a first application example without metabolization , we considered the distribution of the tracer CFDA SE within the liver . Since adequate pharmacokinetic data for CFDA SE were not available for mice , a PBPK model could not be validated in detail . Instead , only the basic physicochemical parameters ( and ) were estimated and subsequently used to calculate the parameters quantifying passive mass transfer in the PBPK model ( Table 1 ) . The pharmacokinetic behavior of CFDA SE was described by passive exchange as given in Equation 7 . We considered an intravenous dose of per kg body mass [68] corresponding to an inflowing concentration of for a duration of for a mouse . Note that the concentration of the compound in the inflowing blood encompasses the corresponding equilibrium concentrations in the red blood cells and the plasma , respectively . The model for CFDA SE was in particular used as a proof of concept for the general performance of the spatially resolved model . We could show with this model that overall mass conservation is satisfied , see Table 1 in Text S1 . Since metabolization of CFDA SE was not considered here , concentrations of CFDA SE in the in- and the outflow alone could be used for this essential step in model validation . The outflow curves for the spatially resolved model ( Figure 6 ) show two effects , a temporal delay and a more smeared-out form of the peak from the spatially resolved simulation compared to the PBPK compartment simulation . The reasons for these observations become clearer when considering the temporal development of the concentrations in the four hepatic subspaces . The spatially resolved model no longer considers mean concentrations in well-stirred compartments but rather calculates heterogeneous distributions of these compounds . Likewise , the transition times needed to flow from the supplying to the draining vascular geometry are heterogeneous due to the different routes taken . We next visualized the total CFDA SE concentration in the HHS ( Figure 7 ) obtained as the weighted average of the concentrations in the different subspaces , ( 11 ) Note that this is the quantity one observes in general for CT or MRI contrast agents by 3D imaging . In Figure 7 and in a Video S3 , the different phases of the first pass of drug perfusion and distribution are shown . Also , the subsequent wash-out of the compound can be observed once the incoming pulse has passed through the liver . Notably , our spatially resolved model describes drug passage as a continuous process which may be used to complement experimental image data at discrete time points . Finally , we simulated steatotic cases where lipid accumulation in the cellular space of the liver influences the distribution behavior of compounds . In particular , we considered whether our spatially resolved simulations may be useful to support diagnostics and imaging . Concentration changes of CFDA SE due to spatially homogeneous and spatially heterogeneous states of steatosis are shown in Figure 8 . As a pharmacokinetic application including intracellular metabolization , we next considered the distribution and metabolization of the sedative midazolam . For model establishment and parameter identification , we used previously published PK data [66] for mice obtaining an intravenous dose of per kg body weight . Metabolization of midazolam by CYP3A was quantified by using gene expression data as a proxy for tissue-specific protein abundance within a whole-body context [15] . This also involves a specific quantification of the hepatic metabolization capacity which is an essential prerequisite for the consecutive parametrization of mass transfer in the HHS . The PBPK model of midazolam was pre-parametrized with the physicochemical compound parameters molecular weight , fraction unbound and lipophilicity . Subsequently , the compound parameters as well as the metabolization parameters were fine-tuned with respect to the experimental PK data [66] ( Table 1 ) . The simulated plasma time curves obtained with the thus established PBPK model are in good agreement with the experimental PK data in mice ( Figure 5 ) . For the midazolam PBPK model in Figure 5 , a concordance correlation coefficient was found , see also Figure 3 in Text S1 . We next used the model parameters identified in the midazolam PBPK model for the spatially resolved model . As before , mass transfer of midazolam within the liver was described by passive exchange between the sinusoidal and interstitial subspace as well as the interstitial and cellular subspace as given in Equation 7 . In addition , a nonlinear cellular metabolization according to Equation 8b was considered in this model . Values for the parameters in the equations are listed in Table 1 . We considered a dose of per kg body mass , corresponding to an inflowing concentration of for a duration of . Outflow concentration time curves from the draining vascular system for the healthy state are shown in Figure 9 . The total molar amounts ( concentrations integrated over the whole liver ) of compounds contained in the red blood cells , plasma , interstitial and cellular subspaces are plotted in Figure 9 . In the simulations , we again observe a delayed and more smeared-out peak in the spatially resolved model . After simulated time , our model predicts a metabolization of approximately of the injected midazolam ( healthy state ) , the rest having flown out from the model or still being present in the HHS and vascular systems . For midazolam metabolization , we also considered steatosis and -induced liver necrosis . In the homogeneous and heterogeneous steatotic state , an increase of the metabolization compared to the healthy state by and , respectively , can be observed , again after simulated time . For liver necrosis following intoxication [61] our simulation predicts a decrease of of the metabolized midazolam amount after . Finally we considered a model for the antibiotic spiramycin for which experimental data for an isolated liver were available in the literature [32] . For model establishment and parameter identification , we again used previously published PK data [67] for mice obtaining an oral dose of per kg body weight of spiramycin . Intravenous PK data are generally necessary for PBPK model development in order to identify systemic clearance capacity and distribution behavior without overlaying processes in the gastro-intestinal tract during oral absorption . Since intravenous PK data , however , were not available for mice , intravenous monkey PK data [69] were used for establishment of the fundamental model structure ( Figure 1 in Text S1 ) . We considered a linear metabolization term and pre-parametrized the distribution model with the physicochemical compound parameters ( MW , , ) . Based on the structure of the intravenous PBPK model , we then established a model for oral administration of spiramycin in mice [32] . Subsequently the model parameters were adjusted with respect to the experimental data [67] ( Table 1 ) . As before for midazolam , the spiramycin PBPK model provides a quantitative description of hepatic clearance capacity . The simulation time curves with the mouse PBPK model for intravenous spiramycin administration are in good agreement with the experimental plasma concentrations ( Figure 5 ) . For the PBPK model for spiramycin , we obtained a concordance correlation coefficient , see also Figure 3 in Text S1 . Based on the validated mouse PBPK model for spiramycin we parametrized the spatially resolved model which is structurally identical to that of midazolam , except for the ( now linear ) metabolization kinetics . The spatially resolved model was then used to simulate experimental data for administration of spiramycin in an isolated liver [32] . The model structure of our spatially resolved model corresponds entirely to the experimental setup of the ex vivo assay , the availability of such highly specific data provided the opportunity to further validate our model . In the experiments [32] , perfusion was performed using a buffer not containing red blood cells . The volume fractions from Equation 1 were hence changed to and . Moreover , a total perfusion of was used , which changes the flow velocities in our model and requires using a smaller time step ( ) . Passive exchange between plasma , interstitial , and cellular subspaces was again modeled as in Equation 7 , mass transfer involving red blood cells , however , was set to zero to take into account the specific experimental setup [32] . Due to the unphysiologically high flow rate , the local effective permeability parameters between interstitial and cellular space were adapted to and . An inflowing spiramycin concentration of for a duration of minutes was used as inflow condition reproducing the inflowing concentration profile in the experimental setup [32] . For a comparison to the experimental data reported in Figure 2 ( wild-type ) in [32] , the outflowing rate of spiramycin was computed and plotted in Figure 10 , again for the healthy state and the two steatotic states described above . Comparing experimental outflow concentrations and those simulated using the spatio-temporal model for the healthy reference case , a concordance correlation coefficient is obtained . Complementarily , volume renderings were generated at different time points after the end of the inflow for minutes ( Figure 10 ) and show the spatial distribution of the spiramycin concentration immediately . This comparison illustrates very nicely how our spatially resolved model can be used to relate macroscopic observations in the plasma to distribution processes at the tissue scale . We here present a spatially resolved model which describes the perfusion , distribution , and metabolization of compounds within the liver . The model structure is based on mass transfer equations obtained from PBPK modeling and vascular structures generated from micro-CT imaging . Our model excludes in particular any recirculation through the body such that metabolization and distribution of compounds can be considered without any overlaying effects . After the end of the initial administrations , a bi-phasic behavior can be observed which is initially governed by the distribution within the tissue and a slow release afterwards . Note that wash-out after the end of injection is additionally determined by advection in the blood flow . Comparing outflowing concentrations from our spatially resolved simulations to those from PBPK compartment models showed a temporal delay , both for CFDA SE and midazolam ( Figures 6 and 9 ) . This is because the compound now needs to pass sequentially through the supplying vascular system , the homogenized hepatic space and the draining vascular system . Different paths through the liver model require different transit times , hence the peaks are more smeared-out in the spatially resolved simulations . This is further emphasized by the temporal development of the concentrations in the four hepatic subspaces for the CFDA SE simulations ( Figure 6 ) . The spatially resolved model no longer considers mean concentrations in well-stirred compartments but rather calculates heterogeneous distributions of the concentrations . Likewise , the transition times needed for the compounds to flow from the supplying to the draining vascular systems are heterogeneous due to the different routes taken . This shows the general performance of the spatially resolved model where mass flows follows the physiological architecture of hepatic tissue governed both by vascular geometry and the composition of the connecting hepatic tissue . While this temporal delay only plays a role during first pass perfusion or similar sudden incidents , results from the spatially resolved model can nevertheless be used to revise PBPK model parameters by comparison with targeted experimental data [32] . Previous approaches already described macroscopic effects such as transit time distribution [10] , [11] , this can also be reproduced using our model . In addition , our approach provides a mechanistic interpretation and visualization of the underlying processes . Our model allows for example a physiology-based description of the liver , thus providing more insight into drug distribution and underlying clearance processes . Likewise , in contrast to fractal models [11] translating the vascular branching to effective pharmacokinetics parameters , we consider the actual anatomical geometry of the organ and its vascular structures . A highly resolved representation is indispensable for models that can also describe individual , potentially heterogeneous , pathologies of the liver . One major drawback , however , of the spatially resolved model is the highly increased computational effort required to run the simulations , see Table 1 in Text S1 . To initially validate our spatially resolved model , we compared simulation results for spiramycin to experimental data obtained ex vivo with an isolated liver . The outflow concentrations simulated using the spatially resolved model and the experimental measurements in [32] are not in full agreement . Note , however , that the simulations of the isolated perfused liver are actually a prediction , since the original equations in the PBPK model were initially adjusted with respect to in vivo PK data [67] . In the light of this workflow it should be noted that the PBPK model represents only an intermediate step before the final spatially resolved model is ultimately established . It is only in this subsequent step that the liver model is integrated in the spatially resolved model , in this case to simulate ex vivo data from an isolated perfused liver [32] . Our approach hence extrapolates in vivo results obtained in a whole-body context to ex vivo data generated in an isolated liver as such supporting a structural transfer of knowledge . Hence , the setup of an isolated perfused liver is a suitable test case . The drawback of this prediction approach is the necessity of integrating experimental data coming from different sources which may partly explain the deviations in the stationary phase during the first 15 minutes during the onset of perfusion . While deviations between experimental data and simulated concentrations can be attributed to large experimental standard deviations or limitations of in silico to ex vivo transferability , a general agreement between the spatially resolved model and experimental data can be observed ( Figure 6 ) . In particular , the clearance rate after the interruption of the spiramycin inflow is in good agreement with the experimental data . This illustrates how our spatially resolved model can be used to relate macroscopic observations in the plasma to distribution processes at the lobulus scale . When applying the combined model to the case of steatosis it was found that already a spatially homogeneous change in the tissue composition leads to spatially heterogeneous differences in the distribution ( Figure 8 ) . The observed behavior showing the effect of an increased intracellular lipid content is actually a zonation effect on the length scale between terminal edges of the supplying and the draining vascular system . As discussed above , the qualitative result and the overall clearance are correct even though the flow distance between the two vascular trees is not the real hepatic lobule size . It was also found that the increased lipid content of the cells leads to longer intracellular retention times since the bound and thus immobile drug fraction increases . In turn , this leads to higher metabolization rates since lipid binding protects the compound from a fast wash-out due to increased retention times in steatotic livers . Differences between spatially homogeneous and heterogeneous steatotic states were also analyzed ( Figure 8 ) . It was found that the difference in lipid accumulation between different lobes and within the lobes had an observable influence on the concentrations as retention times in the cellular subspace are longer in case of higher lipid accumulation . This heterogeneous effect is only visible in spatially resolved modeling ( Figure 9 ) . The model thus provides a mechanistic description of pathophysiological states of the liver and can moreover distinguish between different spatial patterns of the pathology . Let us point out that both the temporal delay of the outflowing peak and differences between different steatotic states are inherent properties of the spatio-temporal model that the original compartmental PBPK model cannot describe . In contrast , the spatially resolved model can capture these effects in a qualitatively plausible way . Our model predicts increased metabolization in steatotic livers , but decreased metabolization following -induced liver necrosis . The simulations hence provide testable predictions which can be compared to previously published experimental data [70] , [71] . For steatotic livers , drug lipophilicity has been related to intrinsic elimination clearance in rats with nonalcoholic steatohepatitis ( NASH ) and control rats , respectively [70] . From this study , an increased clearance of approximately can be estimated for midazolam ( ) in steatotic animals . Even though this relationship has been established in rat livers perfused in situ and cannot be translated directly to our model , it nevertheless confirms qualitative validity of our simulations , since our model predicts an increased metabolization between and . For more detailed comparisons , simulations and experimental measurements would need to be performed for the same experimental setup and in particular in the same species . However , the significantly increased clearance found experimentally in steatotic animals already points to the necessity of a more refined diseased model of steatosis since lipid accumulation alone is obviously not sufficient to explain the observed decrease in metabolic capacity . Possible model extension include , amongst others , previously discussed changes in microcirculation [72] and intracompartmental permeability [70] . For -induced necrosis , our computational findings are also qualitatively validated by experimental observations , where a decreased metabolization of midazolam after pretreatment has been found in rats [71] . Comparing the experimental findings with our current model structure indicates that decreased cytochrome levels in -treated animals need to be considered as future model extensions . Here , our spatially resolved model could in particular be used to differentiate the contributions of enzymatic depletion and volumetric extension of necrosis on the decrease of metabolic capacity . Despite the performance of the newly developed spatially resolved model , several limitations need to be addressed , which represent excellent opportunities for future model refinement . On the technical side , a more detailed geometric vascular model and flow simulation [73] , not only using constant velocity in each cylinder could be considered . However , all this will drastically increase computational costs with little benefit as the intravascular flow patterns are largely irrelevant for what happens in the HHS . Deformations of the organ as in [24] , [74] could also be taken into account . Likewise , changes of the effective blood viscosity [75] besides those due to the Fhrus-Lindqvist effect [44] could also be considered in the model . The hepatic artery as the second supplying vascular system with other inflow concentration could become part of the model if its geometry and the local mixing of blood provided by portal vein and hepatic artery is known for the concrete situation considered [76] . This would also allow for more realistic flow velocities in the SVS . More generally , perfusion heterogeneity could also be considered as well as geometric scales of the perfusion [25] . A more detailed sensitivity analysis than merely one with respect to the vascular geometry ( Figure 7 and Table 1 in Text S1 ) should be performed . For this purpose , known variations as well as measurement uncertainty of both PBPK model parameters and physiological/geometrical data need to be quantified , see e . g . [77] . For a physiologically relevant simulation output , such a sensitivity analysis will require substantial experimental and computational effort and should be part of a future study . Other implementations of the advection-PBPK simulation in the HHS should be investigated as well as the influence of computational resolution on the results . Comparing such fundamentally different implementations , however , is beyond the scope of this article . When considering other metabolization processes , additional compounds , e . g . products formed by the metabolization or compounds only stored in the cellular HHS subspace can easily be included in the model . The exchange across membranes , in Equation 7 , can also be extended easily by active or other nonlinear processes . As discussed above , comparing our computational simulations of pathophysiological states of the liver to experimental data [70] , [71] suggests several model extensions . For steatosis , these include , but are not limited to , a significant increase in liver weight as observed in [59 , Table 6] as well as changes and spatial variations in the effective permeability in Equation 4 and the volume fractions , , and , as a significant decrease of functional capillary density ( sinusoidal length per area ) was reported [59 , Figure 17] . Sinusoidal flow velocities , however , were not observed to change significantly [59 , Figure 16] . Other studies indicate that a change in the microcirculation should be taken into account in a more realistic model of steatosis , see [72] . Moreover , changes of the intracompartemental permeability [70 , Figure 4A] as well as the activity of drug metabolizing enzymes due to steatosis as discussed in [78] may affect the cellular metabolization of compounds . For -induced liver necrosis , changes in cytochrome levels [71] need to be considered in addition to necrotic changes in organ geometry . Here , our spatially-resolved model together with targeted liver histology could be used to differentiate between the different contributions to the decrease in metabolic capacity . Such integrative studies will allow further systematic analyses including iterative model testing and refinement in the future . More general pathological situations can be considered if one has solid knowledge of their spatial heterogeneity and their influence on the model parameters . In case of drugs being administered , also temporal changes of the parameters are possible and can be included in our model . A sensitivity analysis of the spatially resolved model with respect to such parameter perturbations could help to quantify their influence on the heterogeneity of drug distribution . The model in general is not specific for mice , so it can be applied to other species provided the geometry information and PBPK parameters are available . Possibly other connectivity patterns between larger vascular structures and sinusoids depending on the species [79] ( or , closely related in the simulation , diffusion of compounds through vascular walls ) need to be taken into account . The vascular tree geometries used in the model are easily exchanged if more detailed experimentally [80] or algorithmically [42] determined data is available . Similarly , more detailed information about the geometric shape of lobuli ( as in [81] for human livers ) could be taken into account . In particular , in vivo imaging with a slightly higher level of detail than used here will allow running simulations for patient-specific vascular geometries , thus providing great promises for imaging and diagnostics in the future . Corrosion casts [82] , or other types of ex vivo specimens , also scanned in micro-CT , provide higher resolution as time and high radiation doses are not an issue , but obviously do not permit in vivo imaging . Even higher resolution could be obtained by extracting vascular geometries from optical microscopy images of histological serial sections [80] . This , however , requires a tremendous experimental and image processing effort and again is not applicable in vivo . As discussed above , possible zonation effects are qualitatively correctly observed at the length scale between the two incomplete vascular trees in our model rather than the actual length scale of hepatic lobuli . For correct observations in lobuli , our organ-scale simulations should be complemented by sinusoid-scale [83] or lobule-scale simulations in a multi-scale framework [21] , [84] . Since the model can deal with pathological states of the liver and in particular spatially heterogeneous such states , their influence on the intrahepatic distribution of compounds could thereby be simulated pointing to future applications of spatio-temporal modeling in diagnostics . Here , comparison of our continuous simulations with new MRI or CT based image data could support the detection of pathological deviations . Predicting contrast agent distributions may help optimize time points for imaging after injection , benefiting from the much higher temporal and spatial resolution which our simulations can provide . The comparison of simulated and measured contrast agent distributions could therefore be used to identify changes in physiological parameters such that pathologies can be diagnosed . The possibility to simulate heterogeneous distributions provides also important applications for the prediction of toxic side effects . The spatially resolved model allows a location-specific prediction of exposure profiles within the liver . PBPK models have been linked before to models at the cellular scale to predict toxicity responses within hepatic metabolism in response to paracetamol [21] . Together with the spatially resolved model , this can now be used to simultaneously simulate intralobular exposure profiles and the specific cellular response . This allows an in silico prediction of toxic side-effects following the drug administration during the first pass perfusion . Simulations of spatial heterogeneity can also be used to describe local zonation effects within an whole-organ context . PBPK models have been used before to describe genotype-specific differences in hepatic drug uptake [19] and intracellular metabolization [20] . Since the corresponding equations are also used in the spatially resolved model , it also becomes possible to describe first pass effects in a genotype-specific way . Our spatially resolved model could be used for a wide range of technical and medical applications . It could for example be used for hypothermic machine perfusion [33] of livers to be transplanted for which mere static cold storage is ineffective . In this case , recirculation by a perfusion device needs to be considered , for which the influence on relevant compound concentrations can be described based on the existing PBPK models . Moreover , the model could be used to improve treatment planning for islet cell transplantations [34] . Here , mainly the perfusion simulation is needed to predict the distribution of a concentration of cells ( not solutes ) injected in the portal vein . Similarly , the model could help to improve intrahepatic injection of compounds , as discussed in the introduction . Another application could be optimization of targeted drug delivery [37] where drugs are injected in bound form and released at the desired location by mild hyperthermia induced by focused ultrasound . For this purpose , the model has to be combined with a heat transfer simulation [85] . For in vivo modeling within an organism context , more complicated full-body recirculation needs to be taken into account . This in turn requires our model to be integrated in whole-body simulations , regardless whether or not other organs are implemented at a comparable level of detail . Since our model has a spatially resolved internal state and ( depending on the exchange and metabolization kinetics ) may behave non-linearly , a transfer function approach [86] is not immediately applicable . Including recirculation in combination with our spatially resolved model will allow to mechanistically describe the distribution kinetics of fast acting drugs shortly after administration , similarly as it is has been done before with other circulatory models [11] . Extending such earlier approaches , our model will additionally use CT-based vascular trees within the liver . While the spiramycin simulations above show general agreement with experimental results in [32] , this is just a first step towards an exhaustive validation of our approach . Starting points for the important step of model validation in future studies could be comparing simulated and experimentally measured outflow concentrations similar to what was done in [32] or time-resolved imaging of the distribution of tracers ( at least imaged on some slices; see e . g . [87] ) for comparison to simulation results as in Figure 7 . For the latter , also mean transit times [87] estimated from the results in Figure 6 or from Equation 6 in Text S1 could be used for comparison to experimental results . The setting of a compound not entering the cellular subspace , such as in [88] for MRI contrast agent in rats , could be a starting point with a simpler model . In both cases , the ex vivo setting potentially allows for artificially low and thus slow total perfusion , possibly enhancing CT or MR imaging at multiple time points . Much higher spatial resolution at a single time point could be obtained from histological whole-slide scans for which registration [89] and analysis [90] techniques are available . More generally , validation combined with a parameter sensitivity analysis could also help to narrow down parameter ranges where the model predicts physiologically realistic behavior . In this regard , our model could be used for experimental planning to estimate the required spatial and temporal resolution for imaging . Likewise , the number of animal sacrifices could be minimized by specific design of experiments . The model could furthermore be used to quantify the contribution of first pass effects to the overall bioavailability and the experimental variability . As outlined above for steatosis and -induced liver necrosis , our model can be used in combination with targeted experimental data to iteratively investigate pathological changes in liver physiology . Validation or falsification of computational predictions can thereby support mechanistic insights in underlying processes such that overall model structure can be adjusted accordingly . Due to the large level of detail included in our model , such modifications can be directly assigned to specific pathophysiological changes . It is thus possible to test hypotheses about the behavior of pathological livers or to analyze pharmacokinetic effects such as zonation [65] . To this end , PK data , which are ideally sampled densely in time both in the portal vein and in the hepatic vein need to be compared to specific simulation results . Experimentally , one could for example use isolated , pathological livers from genetically modified mice strains or use PBPK models to correlate plasma PK data in these animals with exposure profiles in the liver . Verifying or falsifying these in silico results can then , in turn , trigger further model refinement . We here present a novel method for spatially resolved simulations of first pass perfusion in the liver based on mass balance equations from physiologically based pharmacokinetic modeling as well as vascular geometries obtained by in vivo imaging . The spatio-temporal description of blood flow through the vascular systems in combination with distribution models used in pharmacokinetic modeling allows a mechanistic yet local description of compound perfusion within the tissue . Our combined model is capable of representing spatial parameter heterogeneity , so that the local impact of pathophysiological changes within the liver can be analyzed . The model was used in the present study to investigate spatio-temporal effects of first pass perfusion for exemplary drugs . Two pathophysiological states , steatosis and -induced necrosis , were considered and were found to influence the distribution and metabolization of the compounds . Future applications of the model include optimized design of therapeutic treatments where spatially heterogeneous distributions or spatio-temporal perfusion effects are of relevance , e . g . targeted drug delivery , islet cell transplantations , or catheter placement for intrahepatic injections . We expect the spatially resolved model to be the foundation for further physiologically highly detailed modeling which will help to address specific spatial aspects of pharmacokinetics in the future .	The liver continuously removes xenobiotic compounds from the blood in the mammalian body . Most computational models represent the liver as composed of few well-stirred subcompartments so that a spatially resolved simulation of hepatic perfusion and compound distribution right after drug administration is currently not available . To mechanistically describe the local distribution of compounds in liver tissue during first pass perfusion , we here present a computational model which combines micro-CT based vascular structures with mass transfer descriptions used in physiologically based pharmacokinetic modeling . In the resulting spatio-temporal model , hepatic mass transfer is governed by the physiological architecture and the composition of the connecting hepatic tissue , such that hepatic heterogeneity and spatial distribution can be described mechanistically . The performance of our model is shown for exemplary compounds addressing key aspects of distribution and metabolization of drugs within a mouse liver . We furthermore investigate the impact of steatosis and carbon tetrachloride-induced liver necrosis . Notably , we find that our computational predictions are in qualitative agreement with previous experimental results in animal models . In the future , our spatially resolved model will be extended by including additional physiological information and by taking into account recirculation through the body .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "medicine", "computer", "science", "mathematical", "computing", "liver", "diseases", "mathematics", "materials", "science", "gastroenterology", "and", "hepatology", "drugs", "and", "devices", "drug", "distribution", "porous", "materials", "computerized", "simulations", "drug", "metabolism", "material", "by", "attribute", "pharmacokinetics" ]	2014	Spatio-Temporal Simulation of First Pass Drug Perfusion in the Liver
Worldwide more than 2 billion people are infected with helminths , predominantly in developing countries . Co-infections with viruses such as human immunodeficiency virus ( HIV ) are common due to the geographical overlap of these pathogens . Helminth and viral infections induce antagonistic cytokine responses in their hosts . Helminths shift the immune system to a type 2-dominated immune response , while viral infections skew the cytokine response towards a type 1 immune response . Moreover , chronic helminth infections are often associated with a generalized suppression of the immune system leading to prolonged parasite survival , and also to a reduced defence against unrelated pathogens . To test whether helminths affect the outcome of a viral infection we set up a filarial/retrovirus co-infection model in C57BL/6 mice . Although Friend virus ( FV ) infection altered the L . sigmodontis-specific immunoglobulin response towards a type I associated IgG2 isotype in co-infected mice , control of L . sigmodontis infection was not affected by a FV-superinfection . However , reciprocal control of FV infection was clearly impaired by concurrent L . sigmodontis infection . Spleen weight as an indicator of pathology and viral loads in spleen , lymph nodes ( LN ) and bone marrow ( BM ) were increased in L . sigmodontis/FV-co-infected mice compared to only FV-infected mice . Numbers of FV-specific CD8+ T cells as well as cytokine production by CD4+ and CD8+ cells were alike in co-infected and FV-infected mice . Increased viral loads in co-infected mice were associated with reduced titres of neutralising FV-specific IgG2b and IgG2c antibodies . In summary our findings suggest that helminth infection interfered with the control of retroviral infection by dampening the virus-specific neutralising antibody response . One third of the world population is infected with helminths [1] . Helminth endemic areas overlap with high-risk areas for viral infections in the developing countries [1–3] . Interestingly , antagonistic immune responses control helminth and viral infections . Viral infections skew the immunity towards a proinflammatory type 1 immune response , while helminths strongly polarize towards a type 2 cytokine response [4] . In addition helminths are known for their capacity to dampen the immune response directed against them . Helminth-induced immunosuppression is mediated by the induction of regulatory cell types such as regulatory T ( Treg ) and B cells , negative regulatory receptors such as cytotoxic T lymphocyte associated protein-4 [5 , 6] , B and T lymphocyte attenuator [7] and programmed death-1 [8] , and anti-inflammatory cytokines such as interleukin-10 and transforming growth factor-β [9] . This helminth-mediated downregulation of immunity affects immune responses to unrelated ‘third party’ antigens and thus can be detrimental for the host in terms of vaccinations or co-infections [9 , 10] . Indeed , several human studies suggest that helminth co-infections worsen the outcome of a virus infection [10] . For instance , individuals infected with soil-transmitted helminths , filarial nematodes or water-borne schistosomes , were more susceptible to infection by the human immunodeficiency virus ( HIV ) [11 , 12] , hepatitis C virus [13] and human papillomavirus [14] and/or suffered from increased pathology [13 , 15] . Drug-induced deworming decreased HIV loads [11 , 16–18] in some studies , while anthelminthic treatments had no beneficial effect on HIV infection in other studies [19 , 20] . The mechanism underlying helminth-mediated suppression of virus control was not identified in these studies . To analyse helminth-virus co-infections we set up a co-infection model by first infecting C57BL/6 mice with Litomosoides sigmodontis , followed by infection with FV . Infections of mice with FV are used to study immunity against retrovirus infections [21 , 22] . FV is a murine retroviral complex that consists of the apathogenic replication-competent Friend murine Leukaemia Virus ( F-MuLV ) and the pathogenic but replication-defective Spleen Focus Forming Virus ( SFFV ) . Initial replication of FV takes place in infected erythroid progenitor cells followed by the infection of B cells , T cells , and monocytes/granulocytes [23] . Susceptible BALB/c mice suffer from splenomegaly and succumb to infection due to erythroleukaemia within a few weeks . Resistant C57BL/6 mice generate an immune response to protect them sufficiently from lethal leukaemia , but are not able to eradicate the virus completely and a life long persistence of the virus develops [24] . Similar to infections with HIV , this retroviral complex is controlled by B cell responses , virus-specific cytotoxic T lymphocytes and CD4+ T cells [21 , 25] . CD4+ T cells either act as effector cells or most likely as T helper cells for CD8+ T cells and the production of high-affinity antibodies by B cells [26] . Neutralising antibodies are crucial for the control of FV during acute infection and for vaccine-induced protection against FV [27–30] . Infections of mice with L . sigmodontis are commonly used as a model for human filariasis displaying all features of type 2 immune responses [21] and immunomodulation observed in infected humans [31 , 32] . Infective third stage larvae are transmitted by blood-sucking mites , Ornithonyssus bacoti , and migrate via the lymphatic system to the pleural cavity . There , they moult to fourth-stage larvae ( L4 ) within 10 days and to immature adults within 28 days . In susceptible BALB/c mice mature adults mate and females release their offspring , named microfilariae , into the circulation by day 60 post infection ( p . i . ) . C57BL/6 mice are semi-susceptible to infection with L . sigmodontis . This mouse strain is also efficiently infected , but the parasites do not reproduce and are eradicated within 60 days [33] . We previously reported that L . sigmodontis infection suppressed B and T cell responses to unrelated antigens in both , BALB/c [34 , 35] and C57BL/6 mice [36 , 37] . Strikingly , the suppression of bystander immune responses did not require an acute L . sigmodontis infection . Suppressed B cell responses were still observed 16 weeks after the release of microfilariae into the peripheral circulation had stopped and thus most likely after the eradication of L . sigmodontis female adults [34] . Thus , C57BL/6 and BALB/c mice offer a suitable model to study helminth-mediated immune suppression . The current study had to be performed in semi-susceptible C57BL/6 mice because FV does not induce measurable immune responses in BALB/c mice and rapidly kills infected animals . Here , we show a more severe splenomegaly and enhanced viral loads in L . sigmodontis/FV-co-infected mice compared to FV infection alone . L . sigmodontis infection neither changed the numbers of FV-specific CD8+ T cells nor the cytokine response by CD4+ and CD8+ T cells . Likewise , the phenotype of CD4+ T cells and the number of Foxp3+ Treg were similar in co-infected and FV-infected mice . However , L . sigmodontis infection resulted in significantly reduced FV-specific IgG2b/c titres and FV-neutralising Ig responses . On the other hand , FV infection altered the L . sigmodontis-specific humoral immune response without having an effect on the worm burden . Thus , our results suggest that L . sigmodontis-induced interference with the FV-specific humoral immune response contributed to the impaired virus control in helminth/retrovirus co-infected mice . Animal experimentation was conducted at the animal facility of the Bernhard Nocht Institute for Tropical Medicine in agreement with the German animal protection law . The experimental protocols have been reviewed and approved by the responsible federal health Authorities of the State of Hamburg , Germany , the "Behörde für Gesundheit und Verbraucherschutz" permission number 44/14 . Female C57BL/6 mice were purchased from Harlan and kept in individually ventilated cages . The L . sigmodontis life cycle was maintained in infected cotton rats as described before [37] . Eight to 10 week old C57BL/6 mice were naturally infected by exposure to infected mites ( Ornithonyssus bacoti ) that had been infected 14 days earlier . Mice from different groups ( L . sigmodontis-infected and co-infected ) were placed anesthetised on the same sawdust with infected mites . The FV stock used in these experiments was a FV complex containing B-tropic F-MuLV and polycythemia-inducing SFFV [38] . The stock was prepared as described before [39] . Mice were infected intravenously with 1 . 5 x 104 SFFU of FV spleen homogenate . Mice were sacrificed by deep CO2 narcosis at day 20 post FV infection . Worms were counted after flushing the thoracic cavity with 10 ml cold PBS . Spleen , BM and LN ( popliteal , inguinal , cervical , and axillary ) cells were isolated . The infectious centre assays were performed as described previously [40] . In the vice versa experiments mice were first infected for 24 days with FV and then superinfected with L . sigmodontis for a further period of 25 days . For IFN-γ and IL-4 production , cells were restimulated for 5 h with 2 μg/mL anti-CD28 , 10 μg/mL plate-bound anti-CD3 and Brefeldin A . Cells were stained with Live/Dead Fixable Blue Dead Cell Stain Kit ( Life Technologies ) or Fixable Viability Dye eFluor® 450 ( Affymetrix eBioscience ) according to the manufacturers’ instructions . For surface staining , cells were stained with anti-mouse CD4-Alexa Fluor ( AF ) 680 ( clone: RM4-5 ) , anti-mouse CD8-AF488 , -Allophycocyanin , -PE-Cy7 ( clone: 53–6 . 7 ) , anti-mouse CD43 PE-Cy5 or PerCP ( clone: 1B11 ) for 30 min on ice . For detection of Db-GagL-specific CD8+ T cells , cells were stained with PE-labelled MHC class I H2-Db tetramers ( Tet ) specific for FV GagL peptide . For intracellular expression , cells were stained with anti-mouse IL-4 PE/Cy7 ( clone: 11B11 ) and INF-γ Texas Red or AF488 ( clone XMG1 . 2 ) , or anti-mouse/anti-rat Foxp3 ( clone: FJK-16s ) -staining Set ( Affymetrix eBioscience ) according to the manufacturer's instructions . Gating strategy is shown in supplementary Fig 1 ( S1 Fig ) . Antibodies were purchased from BioLegend or Affymetrix eBioscience . Samples were measured on a LSRII Flow Cytometer ( Becton Dickinson ) and analysed using FlowJo software ( TreeStar ) . ELISA plates were coated overnight with 5 μg/mL F-MuLV antigen or 4 μg/mL L . sigmodontis Antigen ( LsAg ) in PBS . FV-specific IgG2 was measured and calculated as described before [37] . FV-specific isotypes were defined as the highest serum dilution in a serial dilution ( 1:100 to 1:6 . 400 for IgG2b and 1:1000 to 1:64 . 000 for IgG2c ) resulting in an OD450nm above the doubled background . The background OD450nm of the diluent ( 0 . 1% BSA in PBS ) was always below 0 . 1 . For detection of LsAg-specific Ig , sera were either titrated or diluted in a fixed serum concentration as described before [34] . For the analysis of FV-neutralising antibodies , heat-inactivated sera were mixed with an equal volume of 0 . 2 M β-mercaptoethanol ( ME ) , incubated for 30 minutes at 37°C , and then serially diluted ( 1:10 to 1:640 ) with PBS 0 . 01 M β-ME . Sera were mixed with purified F-MuLV and guinea pig complement ( Sigma-Aldrich ) . After 1 h incubation at 37°C samples were added to M . dunni cells which were plated in 24-well plates the day before at a density of 7 . 5 x 103 cells per well . Infectious centre assays were performed as described previously [40] . Foci were counted and dilutions that resulted in a reduction of foci number by 75% or more were considered neutralising . The titre of sera that did not inhibit FV infection in a 1:10 dilution was defined as 1 . LsAg was prepared by homogenization of agile male and female worms isolated from infected BALB/c mice , followed by centrifugation at 10 . 000 x g for 30 min at 4°C . The supernatant was passed through a 0 . 22-mm filter and then stored at -80°C until use . Blood was obtained from naïve , L . sigmodontis-infected , FV-infected and co-infected mice at day 18 by cardiac puncture and allowed to coagulate for at least 1 h at RT . Serum was collected after centrifugation at 10 . 000 x g for 10 min and stored at -20°C . Recipient C57BL/6 mice were infected with FV as described . At day 3 p . i . 250–300 μl serum ( derived from naïve , L . sigmodontis-infected , FV-infected , or co-infected mice ) were intraperitoneally injected . Mice were sacrificed at day 7 post FV infection and viral loads were determined in the BM . Samples were tested for Gaussian distribution and students t test ( unpaired ) or Mann-Whitney test were performed to compare 2 groups . 1-way Anova with Bonferroni post-test or Kruskal-Wallis with Dunn`s multiple comparison test were performed to compare more than 2 groups . Prism software was used for statistical analysis ( GraphPad Software ) . P-values ≤ 0 . 05 were considered statistically significant . To analyse helminth-virus co-infections we first infected C57BL/6 mice with L . sigmodontis by exposing the animals to infected mites . 14 days later , when L4 were present in the pleural cavity , the mice were superinfected with a high dose of FV . The time point of L . sigmodontis infection was chosen since the IgG response to a model antigen and the proliferation of ovalbumin-specific CD4+ TCR transgenic T cells is diminished in day 14 L . sigmodontis-infected mice [36 , 37] . Spleen weight and viral loads in spleen , LN and BM were monitored at day 20 post FV infection ( day 34 post L . sigmodontis infection ) ( Fig 1A ) . Spleen weight ( Fig 1B ) was increased at day 20 p . i . in FV-infected mice and even more pronouncedly in co-infected mice . L . sigmodontis infection alone did not alter the spleen weight nor the number of spleen cells compared to naïve mice ( Fig 1B and 1C ) . Increased splenomegaly was reflected by increased numbers of splenocytes at day 20 p . i . in co-infected mice compared to FV-infected mice ( Fig 1C ) , while numbers of LN and BM cells were similar between the groups ( S2 Fig ) . Viral loads in spleen , LN and BM were higher in L . sigmodontis/FV-co-infected mice than in FV-infected mice at day 20 p . i . ( Fig 1D ) . Increased FV loads were still observed in co-infected mice at day 35 p . i , although viral loads had declined in co-infected and FV-infected mice at this later time point versus viral loads observed at day 20 p . i . ( S2 Fig ) . In summary , concurrent L . sigmodontis infection compromised the control of FV infection . Our findings are consistent with two murine studies analysing the outcome of intestinal nematode infections on the course of viral infections . A pre-existing Trichinella spiralis infection suppressed control of murine norovirus replication [41] and reactivation of latent γ-herpesvirus was shown in mice co-infected with Heligmosomoides polygyrus [42] . Furthermore , a recently published human study showed an increased risk of acquiring HIV seroconversion in individuals suffering from lymphatic filariasis [12] . To elucidate the underlying mechanism we compared the FV-specific immune responses in only FV-infected versus L . sigmodontis/FV-co-infected mice . Since CD8+ T lymphocytes are essential for the control of FV replication during acute infection [43] we first quantified FV-specific CD8+ T cells by staining with a class I Tet specific for the dominant FV epitope DbGagL . As expected , Tet+ CD8+ T cells were absent in lymphoid organs from naïve and L . sigmodontis-infected mice ( S3 Fig ) , but increased to a similar level in spleen , BM and LN of FV-infected and L . sigmodontis/FV-co-infected mice ( Fig 2A ) . To analyse the quality of the CD8+ T cell response we measured IFN-γ expression in CD8+ T cells by intracellular cytokine staining . CD8+ splenocytes and LN cells from naïve and L . sigmodontis-infected mice expressed only low levels of IFN-γ while FV-infected and L . sigmodontis/FV-co-infected mice had increased frequencies of IFN-γ+ CD8+ T cells ( Fig 2B ) . Thereby the increase of IFN-γ+ expression was more pronounced in CD8+ T cells from the spleen than in the LN . Overall , the IFN-γ+ response from L . sigmodontis/FV-co-infected mice resembled that of only FV-infected mice , suggesting that co-infection did not impair the quantity and quality of the CD8+ T cell response to FV infection . Osborne et al . demonstrated reduced numbers and impaired function of virus-specific CD8+ T cells in Trichinella spiralis-norovirus co-infected mice that correlated with increased viral loads [41] . In our study we observed similar numbers of Tet+ CD8+ T cells and no differences in the IFN-γ+ response of CD8+ T cells in general . We analysed the cytokine response of all CD8+ T cells since Tet- T cells contribute to the protective immune response against FV as well . Due to this technical difference and a lack of FV-specific peptides for restimulation we cannot exclude that some aspects of the FV-specific CD8+ T cell response were impaired in co-infected mice . In a previous study we observed an impaired CD8+ T cell response to a vaccine against the liver stage of Plasmodium berghei in L . sigmodontis-infected BALB/c mice [35] . Implementation of a more potent vaccine regime using live Salmonella thereby restored the induction of plasmodium-specific CD8+ T cells in L . sigmodontis-infected mice [35] indicating that L . sigmodontis infection might dampen CD8+ T cell responses under certain circumstances . However , strong stimuli such as prime boost immunizations or virus infections might overcome the helminth-induced suppression of CD8+ T cell responses . We have previously shown that infection with L . sigmodontis suppressed the expansion of CD4+ T cells recognizing helminth-unrelated antigens , such as ovalbumin or keyhole limpet hemocyanin [34 , 36 , 37] . During FV infection protective CD4+ T cells might either act as T helper cells or exhibit anti-viral cytotoxicity against FV infected cells [26] . Therefore we compared the cytokine response , activation and phenotype of CD4+ T cells in spleen and LN during helminth/FV-co-infection . Expression of IFN-γ ( Fig 3A ) in CD4+ T cells reflected the cytokine response observed for CD8+ T cells ( Fig 2B ) . CD4+ T cells from FV-infected and co-infected mice expressed higher levels of IFN-γ than naïve or L . sigmodontis-infected mice . Again the expression of cytokines was more pronounced in spleen cells than in LN cells as observed for CD8+ T cells . The expression of IL-4 , a cytokine that is expressed by follicular T helper cells and Th2 cells , revealed differences between LN and splenic CD4+ T cells ( Fig 3C ) . In the lymph nodes , CD4+ T cells from L . sigmodontis-infected mice showed the highest expression of IL-4 , while frequencies were significantly lower in CD4+ T cells from naïve , FV-infected and co-infected mice . We recorded no statistically significant differences in the expression of IL-4 in CD4+ T cells derived from the spleen . Collectively , the analysis of the CD4+ and CD8+ T cell cytokine responses in LN and spleen revealed a similar IFN-γ expression in FV-infected and co-infected mice . In contrast , the IL-4 expression was suppressed in FV-infected and co-infected mice compared to L . sigmodontis-infected mice , selectively in the LN but not in the spleen . However , since IL-4 does not alter the course of FV infection [23] , a contribution of this cytokine to impaired virus control in co-infected mice is unlikely . In summary the obtained cytokine data rather suggest that FV-induced Th1 polarization outcompeted the pre-existing helminth-induced Th2 polarization as observed for Schistosoma mansoni/lymphocytic choriomeningitis virus co-infection [44] . Next , we measured the activation of CD4+ T cells by flow cytometry . L . sigmodontis infection alone did not induce an activation of CD4+ T cells compared to naïve mice , while CD4+ T cells from only FV-infected and L . sigmodontis/FV-co-infected mice displayed a significant upregulation of CD43 . The magnitude of the CD4+ T cell activation was similar between the two FV-infected groups ( Fig 4A ) . The analysis of CD44 expression and downregulation of CD62L as additional activation markers , as well as the upregulation of Ki-67 as an indicator for T cell proliferation revealed similar results ( S4 Fig ) . The induction of Foxp3+ Tregs favours pathogen survival in both viral and helminth infections [45–47] . In L . sigmodontis and FV infections , the expansion of Tregs is restricted to the site of infection [46 , 48 , 49] . To analyse whether L . sigmodontis and FV synergistically alter Treg responses , we measured percentages and numbers of Foxp3 in CD4+ T cells in infected lymph nodes and spleens . Both frequencies and absolute numbers of Foxp3+ CD4+ Treg were similar in naïve and only L . sigmodontis-infected mice in spleen ( Fig 4B and 4C ) and LN ( Fig 4D and 4E ) . Infection with FV induced an expansion of Foxp3+ CD4+ T cells in both organs , irrespective of pre-existing L . sigmodontis infection ( Fig 4B–4E ) . We have previously shown that Treg numbers increase during FV infection and interfere with the immune control of the virus during the late phase of the acute infection [39 , 46] . Since the expansion of Treg was not exaggerated in L . sigmodontis/FV-co-infected mice compared to FV infection alone , Foxp3+ Treg most likely did not contribute to an impaired control of FV in co-infected mice . In line with this finding , the proliferation of ovalbumin-specific T cells and Ig responses to model antigens were not restored after depletion of Treg in L . sigmodontis-infected mice [34 , 37] . Recovery from acute FV infection is based on the concerted action of CD4+ T cells , CD8+ T cells and antibodies [21] . Since CD8+ and CD4+ T cell responses were similar in FV-infected and co-infected mice , we next analysed the humoral response against the FV helper virus F-MuLV . Protective humoral responses during FV infection are mainly mediated by antibodies of the IgG2 subclass [50] and a beneficial effect of neutralising antibodies in the control of FV was described before [27–29 , 51–53] . Thus , we measured F-MuLV-specific IgG in the serum of FV-infected and L . sigmodontis/FV-co-infected mice by ELISA . FV-specific IgG1 was not detectable at day 20 p . i . Interestingly , co-infected mice had significantly reduced titres of FV-specific IgG2b and IgG2c compared to only FV-infected mice at day 20 p . i . ( Fig 5A and 5B ) . To assess whether this quantitative difference correlated with qualitative differences in the antibody responses , we compared the FV neutralisation capacity of immune sera from day 20 FV-infected and co-infected mice using a complement-dependent neutralisation assay . Sera from co-infected mice had a significantly reduced virus-neutralising capacity compared to FV-infected mice ( Fig 5C ) . To test the clinical relevance of the reduced IgG2 titre and the in vitro neutralisation [28 , 29] we performed serum transfer experiments . Mice , that received sera from FV-infected animals displayed statistically significant lower viral loads in the BM than mice receiving control sera from naïve or L . sigmodontis-infected ( Fig 5D ) animals . Transfer of immune sera from L . sigmodontis/FV-co-infected mice , by contrast , did not statistically significant reduce the viral load , reinforcing the notion that impaired control of FV in co-infected mice was due to reduced antibody titres . However , viral loads were diminished by trend in mice receiving sera from co-infected mice compared to control sera indicating that the reduced amount of FV-specific antibodies still present in the immune serum of L . sigmodontis/FV-co-infected mice is nonetheless beneficial for the host . We did not address the mechanism leading to diminished anti-FV antibody responses in the current study . However , the observation that L . sigmodontis/FV-co-infected mice display decreased FV-specific antibody responses confirm our previous studies showing that infection with L . sigmodontis drastically reduced IgG responses to a model antigen immunization in BALB/c [34] and C57BL/6 mice [36 , 37] . Regarding the mechanism , we showed that reduced IgG responses to model antigen immunization were associated with reduced numbers and frequencies of model antigen-induced follicular T helper cells ( Tfh ) [34] , that are required for a class switch and production of neutralising high affinity antibodies . It was not possible to distinguish FV-specific and L . sigmodontis-specific Tfh using common markers such as CD44 , PD-1 and CXCR5 , in the current study , since Tfh are long-lived [54] and infections with L . sigmodontis and FV both induce germinal centre reactions in the spleen [34 , 52] . Since our cytokine data suggest a FV-induced skewing of the immune response towards a proinflammatory cytokine response we analysed whether the FV-infection has an impact on the humoral immune response and eradication of L . sigmodontis . We found diminished Th2-associated LsAg-specific IgG1 titres in co-infected mice , while Th1-associated LsAg-specific IgG2c and IgG3 titres were clearly increased in co-infected mice in comparison to L . sigmodontis-infected mice ( Fig 6A ) . Thus , FV infection polarizes the humoral immune response against L . sigmodontis to a more proinflammatory response . Despite the differences in the L . sigmodontis-specific humoral immune responses we recorded equal numbers of young adults at day 34 post L . sigmodontis infection , i . e . day 20 post FV infection , in L . sigmodontis-infected and co-infected mice ( Fig 6B and 6C ) . However , since a higher prevalence of intestinal nematodes has been described in HIV+ pregnant women [55] , we further analysed a possible impact of a pre-existing FV-infection on the worm burden . To this end , mice were first infected for 24 days with FV followed by infection with L . sigmodontis ( Fig 6D ) . Despite the pre-existing FV infection , the worm burden was alike in co-infected mice compared to L . sigmodontis-infected mice ( Fig 6E ) . In summary , infection with FV , albeit inducing a polarization of the LsAg-specific humoral immune response , did not have an impact on the worm burden . By contrast , we show that the altered course of a retroviral infection in the presence of a filarial nematode is linked with reduced virus-specific antibody levels . Due to their asymptomatic nature many helminth species might impair the course of a viral infection without being diagnosed . In this context , our data highlight the importance of deworming programs or the development of vaccines against helminths in developing countries where the incidence of HIV/filarial co-infections is high .	The coincidental infection of a host with two different pathogens is widespread in low-income countries . Regions where helminth infections are endemic strongly overlap with areas where the incidence of viral infections such as HIV is high . HIV is a major public health issue causing more than 1 million deaths per year . To analyse the impact of a pre-existing helminth infection on a viral infection we established a helminth/retrovirus co-infection mouse model . Mice that were first infected with Litomosoides sigmodontis and subsequently with a murine retrovirus showed a more severe course of virus infection , i . e . exaggerated splenomegaly and higher viral loads . Since different lymphocytes such as B and T cells contribute to viral control we analysed the cellular and humoral immune response . While T cell responses were similar in co-infected and virus-infected mice , we observed reduced titres of virus-specific antibodies in co-infected mice . Our results suggest that helminth infection interfered with viral control by dampening the virus-specific antibody response . The viral infection itself altered the humoral immune response against L . sigmodontis without changing the worm burden . In summary , our data highlight the importance of deworming programs or vaccines against helminths in developing countries where the incidence of helminth/HIV co-infections is high .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "and", "Discussion" ]	[ "blood", "cells", "innate", "immune", "system", "medicine", "and", "health", "sciences", "immune", "cells", "immune", "physiology", "cytokines", "viral", "transmission", "and", "infection", "spleen", "immunology", "microbiology", "parasitic", "diseases", "nematode", "infections", "developmental", "biology", "molecular", "development", "cytotoxic", "t", "cells", "viral", "load", "white", "blood", "cells", "animal", "cells", "t", "cells", "immune", "response", "immune", "system", "helminth", "infections", "cell", "biology", "virology", "physiology", "biology", "and", "life", "sciences", "cellular", "types" ]	2016	Filariae-Retrovirus Co-infection in Mice is Associated with Suppressed Virus-Specific IgG Immune Response and Higher Viral Loads
Human immunodeficiency virus type 1 ( HIV-1 ) invades the central nervous system ( CNS ) shortly after systemic infection and can result in the subsequent development of HIV-1–associated dementia ( HAD ) in a subset of infected individuals . Genetically compartmentalized virus in the CNS is associated with HAD , suggesting autonomous viral replication as a factor in the disease process . We examined the source of compartmentalized HIV-1 in the CNS of subjects with HIV-1–associated neurological disease and in asymptomatic subjects who were initiating antiretroviral therapy . The heteroduplex tracking assay ( HTA ) , targeting the variable regions of env , was used to determine which HIV-1 genetic variants in the cerebrospinal fluid ( CSF ) were compartmentalized and which variants were shared with the blood plasma . We then measured the viral decay kinetics of individual variants after the initiation of antiretroviral therapy . Compartmentalized HIV-1 variants in the CSF of asymptomatic subjects decayed rapidly after the initiation of antiretroviral therapy , with a mean half-life of 1 . 57 days . Rapid viral decay was also measured for CSF-compartmentalized variants in four HAD subjects ( t1/2 mean = 2 . 27 days ) . However , slow viral decay was measured for CSF-compartmentalized variants from an additional four subjects with neurological disease ( t1/2 range = 9 . 85 days to no initial decay ) . The slow decay detected for CSF-compartmentalized variants was not associated with poor CNS drug penetration , drug resistant virus in the CSF , or the presence of X4 virus genotypes . We found that the slow decay measured for CSF-compartmentalized variants in subjects with neurological disease was correlated with low peripheral CD4 cell count and reduced CSF pleocytosis . We propose a model in which infiltrating macrophages replace CD4+ T cells as the primary source of productive viral replication in the CNS to maintain high viral loads in the CSF in a substantial subset of subjects with HAD . Human immunodeficiency virus type 1 ( HIV-1 ) -associated dementia ( HAD ) is a severe neurological disease that affects a subset of HIV-1-infected individuals [1] , [2] . HIV-1 infection of the central nervous system ( CNS ) occurs shortly after peripheral infection , most likely through the trafficking of infected lymphocytes and monocytes across the blood-brain barrier ( BBB ) [3] , [4] . Once HIV-1 crosses the BBB it can infect perivascular macrophages and brain-resident microglia , and some studies have shown that neurotropic viruses preferentially infect macrophages [5]–[8] . HIV-1 may persist in the CNS during therapy due to the insufficient CNS penetration of some antiretroviral drugs [2] , [9]–[11] . HIV-1 variants have been detected at autopsy in the brains of HAD subjects , and these brain-derived variants are genetically distinct from virus detected in the peripheral blood [7] , [12]–[15] . A principal impediment to studying viral evolution in the CNS is that direct sampling of HIV-1 in brain tissue is usually possible only once , at biopsy or autopsy . To examine viral populations in the CNS over the course of HIV-1 infection we have relied upon repeated sampling of virus in the cerebrospinal fluid ( CSF ) . Previous studies have shown that virus detected in the CSF originates from both local CNS tissue and the peripheral blood [16]–[19] , indicating that the CSF may act as a site of mixing of virus present in the brain and the periphery . In addition , genetic compartmentalization has been reported between blood plasma and CSF viral variants [20]–[23] . We previously examined the cellular sources of HIV-1 in the CNS by utilizing the heteroduplex tracking assay ( HTA ) to measure viral decay rates in HIV-1-infected subjects initiating antiretroviral therapy [24] . In this study we reported that the subset of compartmentalized virus detected in the CSF of four asymptomatic subjects decayed rapidly after the initiation of therapy , suggesting that the compartmentalized virus is coming from a short-lived cell type , such as CD4+ T cells [24] . The population dynamics of systemic HIV-1 replication have been studied extensively [25]–[27] , but the extent of viral replication in specific cell types in the CNS over the course of disease is not yet known . The use of antiretroviral drugs to prevent HIV-1 infection of uninfected cells provides a tool for “viewing” the rate of decay for cell-free virus and virally-infected cells . HIV-1 decay in peripheral blood after the initiation of highly active antiretroviral therapy ( HAART ) occurs in at least two phases [25] , [27] . The first phase of decay is rapid and has been proposed to represent the turnover of cell-free virions and productively infected CD4+ T cells [25]–[28] . The second phase is slower and may reflect the decay of long-lived infected cells , possibly latently infected resting CD4+ T cells and cells of the monocyte lineage [25] , [27]–[29] , and the release of virions from follicular dendritic cells [28] , [30] . Recently , a study using the integrase inhibitor raltegravir reported altered HIV-1 decay kinetics and a reduction of the second viral decay phase [31] , suggesting integration as a rate limiting step of infection in a subset of cells . The implications of these data on measured viral decay rates remain to be clarified; however , the reduction in the second phase of HIV-1 decay may indicate that longer-lived HIV-1-infected cells contribute less to total viral load than previously thought , but it does not preclude the possibility that the second phase of HIV-1 decay may reflect the turnover of long-lived cells [31] , [32] . In this study , we characterized the lifespan of the cellular source of compartmentalized HIV-1 in the CNS of subjects with and without symptomatic neurological disease by calculating viral decay rates during the initiation of antiretroviral therapy . The heteroduplex tracking assay ( HTA ) [33] , [34] was used to distinguish between HIV-1 genetic variants in the CSF that were either compartmentalized to the CSF or equilibrated with the peripheral blood . HTA has been used in previous studies to differentiate between HIV-1 genetic variants in separate anatomical compartments [22] , [24] , [35] , [36] and HIV-1 evolutionary variants [37]–[42] , including drug resistance mutations [43] , [44] . The HTA is a useful tool for resolving and quantifying complex viral populations based on their genotype , and is able to detect HIV-1 variants that comprise as little as 1–3% of the total viral population . We targeted the variable regions of the env gene for HTA analysis of our subject population in order to resolve multiple HIV-1 genetic variants . In this study we confirm rapid viral decay in the CSF of asymptomatic subjects initiating HAART , and we report reduced rates of viral decay of compartmentalized virus in the CSF in a subset of neurologically symptomatic subjects initiating antiretroviral therapy . These results suggest a shift in the cell type that produces the bulk of the virus in the CSF late in disease as part of the process of viral pathogenesis in the CNS . Our analysis included 11 asymptomatic subjects ( 7 new subjects , 4 subjects reported in [24] ) , 1 subject with minor cognitive motor disorder ( MCMD ) , and 7 subjects with HIV-1–associated dementia ( HAD; see Table 1 ) . In general , subjects with HAD have higher viral load in the CSF [17] , [45] , [46] and increased HIV-1 compartmentalization in the CSF [21] , [22] . To assess compartmentalization we measured the relative abundance of HIV-1 variants in the blood plasma and CSF as resolved by the heteroduplex tracking assay ( HTA ) , then calculated the percent difference values between the two viral populations ( see Table 2 ) . We found that the CSF and plasma viral populations were different for subjects with HIV-associated neurological disease ( average = 67% different; range = 36–88% different ) compared to the asymptomatic subjects ( average = 42% different; range = 10–78% different ) . This difference approached statistical significance in spite of the small sample size ( p = 0 . 054 using a two-tailed Mann-Whitney test ) , and this trend of increased viral compartmentalization in the CSF with HAD is consistent with the difference seen in a larger cross-sectional analysis [21] . We next used the HTA to follow differential decay of shared and compartmentalized variants when subjects initiated therapy . In this study , the subjects had an average reduction of 91% of the virus in the blood , and 88% of the virus in the CSF , over the period of sampling for HTA analysis ( Table 2 ) . The HTA is a useful tool for sampling complex viral populations , and is sensitive enough to detect minor variants within the population . We utilized HTAs targeting the hyper-variable regions V1/V2 and V4/V5 of the env gene to detect and measure the decay of individual HIV-1 variants in the cerebrospinal fluid and plasma of subjects initiating HAART . The HTA that was the most reproducible ( V1/V2 or V4/V5 ) was used for the final decay and half-life calculations . The half-lives for the different variants in the blood for four of these subjects have been reported previously [47] . The V1/V2 and V4/V5 HTA analyses for the seven new asymptomatic subjects revealed rapid HIV-1 decay for both compartmentalized and shared variants detected in the CSF ( see Figure 1 ) . The decay of individual variants was organized into two groups for half-life analysis: decay of CSF-compartmentalized variants and decay of variants shared between the blood and the CSF . The HTA gels for the longitudinal samples from the seven new asymptomatic subjects are shown in Figure 1A , and graphs representing the viral decay are shown in Figure 1B . In this analysis , viral variants that decay more slowly will make up an increasing percentage of the total viral population over the course of therapy . However , if all variants decay at the same rate then the relative percentages will remain the same over time . HIV-1 half-lives for plasma and CSF variants were calculated based on the slopes of the decay curves ( summarized in Table 2 ) . Based on data generated from the seven new asymptomatic subjects analyzed in this study , half-lives calculated for the total plasma viral load decay were short ( t1/2 mean = 1 . 46 days; t1/2 range = 0 . 58–2 . 27 days ) , and total CSF viral load half-lives were short ( t1/2 mean = 1 . 5 days; t1/2 range = 0 . 77–2 . 04 days ) . These half-lives are similar to the data reported for 4 asymptomatic subjects that were previously studied [24] . Although some asymptomatic subjects have large percent difference values between the blood and CSF viral populations , not all of the variants detected in the CSF met the criteria for compartmentalization . Viral variants in the CSF were considered compartmentalized if they were unique to the CSF or they were present in a substantially higher concentration in the CSF compared to the plasma . CSF-compartmentalized variants were detected in asymptomatic subjects 5005 , 4014 , and 4022 . To increase our sample size we included the half-life data from the four asymptomatic subjects reported in ref . [24] in our analysis of CSF-compartmentalized decay . Including these additional four subjects ( n = 7 total asymptomatic subjects with some compartmentalized virus: 3 new subjects and 4 previously reported subjects ) , we found that the half-lives for CSF-compartmentalized variants in these subjects were short , with a mean of 1 . 57 days ( t1/2 range = 0 . 75–2 . 75 days; see below ) . These data indicate that CSF-compartmentalized virus in asymptomatic subjects is most likely originating from a short-lived cell type , such as a CD4+ T cell . The reported half-life of a productively infected CD4+ T cell is approximately 2 days [28] , which coincides with our average measured half-life of 1 . 57 days in these subjects . We expanded our analysis of viral decay to HIV-1-infected subjects who were diagnosed with either MCMD or HAD to address the hypothesis that CSF-compartmentalized variants in these subjects originate from longer-lived cells . Viral decay in the CSF of eight subjects with neurological disease was analyzed using HTAs targeting the V1/V2 and V4/V5 regions of env . The HTA analyses for the eight subjects with HIV-associated neurological disease showed either rapid or slow viral decay among the subjects . The longitudinal HTA gels for each neurologically symptomatic subject are shown in Figure 2A , and the graphs of viral decay are shown in Figure 2B and 2C . Similar to the asymptomatic subject decay analysis , individual variants were grouped as either CSF-compartmentalized variants or variants shared between the blood and the CSF for the decay analysis . Total plasma viral load decay was rapid for all subjects with neurological disease , with a mean half-life of 2 . 11 days ( t1/2 range = 1 . 42–2 . 91 days; summarized in Table 2 and Figure 3 ) . We measured rapid viral decay for CSF-compartmentalized variants after the initiation of HAART for four subjects with HAD ( 4033 , 5003 , 7036 , 4051; t1/2 mean = 2 . 27 days; t1/2 range = 1 . 23–3 . 67 days; Figure 2B and Figure 3; summarized in Table 2 ) , similar to asymptomatic subjects . In contrast , prolonged viral decay was measured for CSF-compartmentalized variants for the other four subjects with neurological disease ( 4013 , 5002 , 4059 , 7115; t1/2 range = 9 . 85 days to no initial decay ) , with three subjects displaying biphasic decay . CSF-compartmentalized variants for subjects 4013 , 5002 , and 7115 displayed a biphasic decay ( see Figure 2C ) , where the first phase of viral decay was slow ( 4013 t1/2 = 28 . 5 days; 7115 t1/2 = no initial decay; 5002 t1/2 = no initial decay ) , and the second phase was faster ( 4013 t1/2 = 3 . 9 days; 7115 t1/2 = 6 . 4 days; 5002 t1/2 = 4 . 24 days ) . Figure 3 and Table 2 report the half-lives calculated for both phases of decay . Subject 4059 displayed only a slower decay rate for the CSF-compartmentalized variants ( t1/2 = 9 . 85 days ) . Total CSF viral load decay was similar to the decay rates measured for CSF-compartmentalized variants for all subjects with neurological disease . This is due to the fact that most of the virus in the CSF was compartmentalized in these HAD subjects . The decay of the small amounts of shared variants fluctuated in these subjects from decreasing with a rate similar to the virus in plasma to decreasing with a slow rate similar to that of the CSF-compartmentalized variants ( see Table 2 ) . For each CSF sample time point the CSF white blood cell ( WBC ) count was measured to determine if any subjects had CSF pleocytosis ( defined as >5 cells/µl; [48] , [49] ) . We found that all four subjects with rapid CSF-compartmentalized variant decay either had high CSF WBC levels at entry ( 4033 = 28 cells/µl; 5003 = 46 cells/µl; 7036 = 240 cells/µl; 4051 = 12 cells/µl ) , or the CSF WBC levels increased while on therapy . Conversely , the four subjects with neurological disease that displayed slower CSF-compartmentalized variant decay either had extremely low levels of CSF WBCs at entry ( 4013 = 10 cells/µl; 5002 = 66 cells/µl; 4059 = 1 cells/µl; 7115 = 12 cells/µl ) , or the CSF WBC levels decreased to low levels after the initiation of antiretroviral therapy . We examined the CSF WBC levels of these two groups in more detail by calculating the CSF WBC average for each subject from baseline through the first 14 days of antiretroviral therapy . The subjects with rapid CSF-compartmentalized variant decay had higher CSF WBC averages , while subjects with slower or biphasic CSF-compartmentalized variant decay had lower CSF WBC averages ( see Table 1 ) , and this difference was statistically significant ( p = 0 . 029 using a two-tailed Mann-Whitney test ) . It has been reported that HIV-1-infected subjects with CD4 counts below 50 cells/µl have reduced CSF pleocytosis [49] . We also examined whether the viral decay rates measured by HTA were correlated with the degree of immunodeficiency by analyzing CD4 counts for each group of subjects . The four subjects with rapid CSF-compartmentalized variant decay had significantly higher baseline CD4 counts ( see Table 1 ) compared to the four subjects with slower CSF-compartmentalized variant decay ( p = 0 . 006 using a two-tailed unpaired t-test ) . Thus , in subjects with HIV-1–associated neurological disease , viral decay rates are associated with the degree of immunodeficiency and CSF pleocytosis . We did not detect an association between CSF pleocytosis and rapid viral decay in the CSF for asymptomatic subjects . The CSF WBC average was calculated for each subject as stated above , and the range extended from 0 cells/µl up to 20 cells/µl ( Table 1 ) . All variants detected in the CSF of asymptomatic subjects decayed rapidly upon the initiation of antiretroviral therapy; however , we found that the presence of CSF-compartmentalized variants was associated with higher average CSF WBC levels . All four of the asymptomatic subjects that did not have compartmentalized virus had low average CSF WBC counts ( 4012 , 4030 , 4023 , 4021 ) , while the three asymptomatic subjects that had detectable CSF-compartmentalized variants also had higher average CSF WBC levels ( 5005 , 4022 , 4014; see Table 1 ) . Thus in the asymptomatic subjects the presence of pleocytosis may be associated with an early inflammatory response to increased levels of autonomously replicating virus . Some antiretroviral drugs have poor penetration into the CNS [50] . In order to determine whether the differential decay we detected by HTA was associated with poor CNS drug penetration , we calculated the CNS Penetration Effectiveness ( CPE ) rank [50] for the drug regimens that each of the 15 subjects were receiving at the time of sample collection ( see Table 1 ) . Drugs that have poor penetration into the CNS were assigned a rank of 0 , intermediate penetration was assigned a rank of 0 . 5 , and high penetration was assigned a rank of 1 [50] . The four subjects that showed a longer viral half-life by HTA analysis had CPE ranks ranging from 2 . 0 to 2 . 5 , while the other subjects that displayed rapid viral decay had CPE ranks from 1 . 5 ( 5 subjects ) to 3 . 5 ( 1 subject ) . All subjects with neurological disease had CPE ranks above 2 . 0 except for subject 5003 ( CPE rank = 1 . 5 ) . A previous study reported that CPE ranks below 2 . 0 were associated with a significant ( 88% ) increase in the ability to detect virus in the CSF , and higher CSF viral loads were associated with low CPE ranks [50] . All of the subjects with longer viral half-lives had CPE ranks of 2 . 0 or above , suggesting that the slower HIV-1 decay we detected by HTA was not associated with poor CNS drug penetration . Alternatively , there could be infected cells located in parenchymal compartments that are less accessible to drugs , but this seems unlikely because the virus still has access to the CSF . We also investigated the possibility that slower decay was a result of drug resistance mutations present in the viral population in the CSF . Drug resistance mutations were measured for CSF samples of subjects 4013 , 5002 , 4059 , and 7115 . The resistance test was conducted for time points after the initiation of drug selection to allow for enrichment of any potential drug resistant variants . Subjects 4013 , 5002 , and 4059 showed no evidence of resistance mutations in reverse transcriptase ( RT ) or protease that confers resistance to antiretroviral drugs ( data not shown ) . Subject 7115 had the resistance mutation K103N in RT , which confers resistance to non-nucleoside RT inhibitors ( NNRTI ) . However , at the time of this study , subject 7115 was not taking an NNRTI , and was instead on a drug regimen that included zidovudine , lamivudine , and lopinavir . Therefore , there is no evidence that drug resistance played a role in the slower viral decay detected by HTA in these four subjects . Using the biotin-V3 HTA procedure , we also examined whether slower viral decay was associated with V3 sequence differences . The biotin-HTA is a modification of the original HTA method that incorporates a biotin tag into the probe to allow direct sequencing of the query strand isolated from the gel [51] . This newly developed HTA procedure resolves minor variants in the gel , and then allows the recovery and sequence analysis of both major and minor HIV-1 V3 variants from complex viral populations [51] . Following V3 PCR amplification and HTA analysis , we excised the gel fragments containing the V3 heteroduplexes , purified the query DNA strand using streptavidin-coated magnetic Dynabeads® , and directly sequenced the subsequent V3 PCR products [51] . The migration patterns for the V3 heteroduplexes and the inferred V3 amino acid sequence obtained for the heteroduplex in each gel band are shown in Figure 4 . The biotin-V3 HTA procedure was conducted on plasma samples from all subjects at the first time point collected , and CSF samples were analyzed for subjects with HIV-associated dementia . No significant V3 sequence differences were detected between asymptomatic and symptomatic subjects , or between subjects with rapid versus slow decay by HTA ( Figure 4B ) . Only one subject ( 4014 ) had V3 sequences that were X4-like by the Position-Specific Scoring Matrix ( PSSM ) method [52] of predicting co-receptor usage based on genotype . We did note that two subjects with slower decay by HTA had compartmentalized V3 variants detected in the CSF viral population that were much more R5-like by sequence compared to the V3 sequence variants detected in the plasma viral population . However , R5-like V3 sequences were also detected in the CSF for HAD subjects with rapid viral decay , indicating that V3 sequence differences and co-receptor usage are not responsible for the differential decay detected by HTA . There are several lines of evidence that support the idea that HIV-1 can replicate in the central nervous system ( CNS ) . HIV-1-infected macrophages and microglia have been detected in the brains of subjects with HIV-1–associated dementia ( HAD ) at autopsy [6] , [53] , [54] . In addition , genetically distinct HIV-1 variants , different from those in the peripheral blood , are seen in the CNS of subjects with HAD [7] , [12]–[15] . These inferences can be extended using CSF as a surrogate for the CNS where genetic compartmentalization can be detected when comparing blood and CSF viral variants [20]–[23] , and bulk virus in the CSF of subjects initiating HAART can decay with different kinetics compared to virus in the blood [16] , [19] , [55] . Furthermore , it appears that this independent replication is relevant , if not causal , of HIV-associated neuropathogenesis . The extent of compartmentalization in the CSF , as measured by the heteroduplex tracking assay , increases in subjects with HAD , suggesting more sustained autonomous replication is associated with the neurological disease state [21] , [22] . Also , slow decay of virus in the CSF compared to the blood is associated with subjects with neurological disease , especially HAD subjects , suggestive of virus being produced from a different cellular source [16] , [19] , [55] . In addition to viral genetic compartmentalization there are other markers of neuropathogenesis in HIV-1-infected individuals , such as CSF neopterin [56] , [57] , CSF light-chain neurofilament protein [57]–[59] , and CSF chemokine levels [60]–[64] . In the current work we have attempted to combine the observations of viral genetic compartmentalization and differential decay in subjects initiating HAART by comparing the rates of decay of variants shared between the CSF and the blood versus those variants that were compartmentalized in the CSF . The goal of this work was to examine the link between compartmentalized virus as a marker for autonomous replication in the CNS and the production of virus in the CNS by long-lived cells . We used heteroduplex tracking assays ( HTAs ) targeting the variable regions of env to identify CSF-compartmentalized variants and variants shared between the CSF and blood plasma , and then measured the viral decay kinetics of these two distinct classes of viral variants after the initiation of antiretroviral therapy for asymptomatic and neurologically symptomatic subjects . We found that plasma HIV-1 variants decayed rapidly for both neurologically asymptomatic and symptomatic subjects , indicating that short-lived cells , presumably activated CD4+ T cells , are the predominant source of virus in the periphery during all disease stages . Additionally , shared and compartmentalized variants in the CSF of seven asymptomatic subjects decayed rapidly , with a mean half-life of 1 . 35 and 1 . 57 days , respectively . These decay rates are consistent with our previous study of four asymptomatic subjects [24] . HIV-1 viral load decays in the peripheral blood with the same half-life as a productively infected CD4+ T cell ( approximately 2 days; [25] , [27] , [28] ) , so it is most likely that CSF-shared and compartmentalized virus in asymptomatic subjects is originating from a short-lived cell type , such as a CD4+ T cell . The level of HIV-1 compartmentalization in the CSF in these asymptomatic subjects varied , and we noted that there was a trend of increased CSF pleocytosis in the asymptomatic subjects with greater compartmentalization . We also examined HIV-1 decay in subjects with neurological disease that were starting HAART . Rapid viral decay was measured for CSF-compartmentalized variants after the initiation of HAART for four HAD subjects ( t1/2 mean = 2 . 27 days ) , while slow viral decay was measured for CSF-compartmentalized variants from the other four subjects with neurological disease ( t1/2 range = 9 . 85 days to no initial decay ) . It is known that HIV-1 may persist in the CNS during antiretroviral therapy due to insufficient CNS penetration of some antiretroviral drugs [2] , [9]–[11] . We determined that the slow decay detected for CSF-compartmentalized variants was not associated with poor CNS drug penetration , the presence of drug resistant virus in the CSF , or the detection of X4-like virus genotypes . It has been suggested that HIV-1 produced by long-lived cell lineages such as macrophages , microglia , and resting CD4+ T cells most likely decays with a half-life of 14 days or greater [27]–[29] . The longer half-lives we detected suggest that compartmentalized HIV-1 in the CSF of some neurologically symptomatic subjects may be originating from a long-lived cell type . While slower HIV-1 decay was detected for half of the subjects with neurological disease , compartmentalized variants in the CSF of some subjects decayed rapidly . Further analysis revealed that the differential decay measured for CSF-compartmentalized variants in subjects with neurological disease was correlated with the degree of CSF pleocytosis . Four of the eight subjects with HIV-associated neurological disease displayed rapid CSF-compartmentalized variant decay , and this was correlated with higher CSF WBC levels ( moderate to severe pleocytosis ) . The compartmentalized variants detected in the CSF of the four other subjects showed slow or biphasic decay after the initiation of HAART , and this was associated with lower CSF WBC levels ( no or mild pleocytosis ) . Additionally , the subjects with rapid CSF-compartmentalized variant decay had significantly higher CD4 counts than subjects with slow compartmentalized variant decay , indicating that subjects with slow decay of CSF-compartmentalized virus have increased immunodeficiency . We suggest that more profound immunodeficiency results in fewer lymphocytes trafficking into the CNS , which is consistent with the decreased CSF WBC counts for the subjects with slow decay . HIV-1 infection can be associated with CSF pleocytosis in neurologically symptomatic subjects , asymptomatic subjects , and individuals lacking any CNS opportunistic infections [48] . Additionally , some studies have shown that CSF WBC levels are correlated with CSF HIV-1 RNA concentrations [49] , [65] , [66] , and CSF pleocytosis has been shown to decrease after the initiation of antiretroviral therapy [48] . In this current study we found an association between the extent of immunodeficiency , CSF pleocytosis and rapid HIV-1 decay kinetics for compartmentalized variants in the CSF of neurologically symptomatic subjects , although the strength of the interpretation is somewhat limited by our small sample size . Taken together , we have developed a model of HIV-1 infection in the CNS in the context of neurological disease ( Figure 5 ) . The model has several features that incorporate viral genetic compartmentalization , CSF pleocytosis , and viral decay rates in the CSF as a measure of the virus-producing cell . First , the majority of the virus detected in the CSF of a subset of asymptomatic subjects is imported from the peripheral blood ( Figure 5A ) . HIV-1-infected CD4+ T cells in the peripheral blood release virus that is detectable in the blood plasma and the CSF and that decays rapidly upon the start of antiretroviral therapy , representing the relatively fast turnover of uninfected CD4+ T cells . HIV-1-infected CD4+ T cells in the peripheral blood can migrate from the periphery into the CNS and secrete virus in the CNS that is genetically similar to virus in the peripheral blood . No or only mild pleocytosis was detected for this group of asymptomatic subjects , and we suggest this represents minimal inflammation in the CNS . It is possible that some CNS HIV-1 variants are independently replicating at a low level in these asymptomatic subjects , but we were not able to detect these genetic variants above the background of virus recently imported from the periphery . In these subjects virus decays with the half life of peripheral T cells , the presumed source of the virus . A second pattern exists for the other asymptomatic subjects and also for a subset of the neurologically symptomatic subjects . There is increased compartmentalization of HIV-1 in this subset of asymptomatic subjects , and the majority of virus detected in the CSF is compartmentalized in HIV-1-infected individuals with severe neurological disease . In addition , both of these groups have increased pleocytosis . We found that CSF-compartmentalized variants decayed rapidly upon the initiation of antiretroviral therapy in these remaining asymptomatic subjects and in this subset of four subjects with HIV-1–associated dementia . It is possible that compartmentalized variants detected in these subjects are produced by long-lived cells in the CNS; however the majority of the compartmentalized virus is produced by a short-lived cell type . We propose that compartmentalized virus may be maintained by long-lived cells in the CNS and that this virus is amplified by short-lived trafficking CD4+ T cells to detectable levels in the CSF for asymptomatic subjects , and to high titers in the CSF of HAD subjects ( Figure 5B ) . The elevated level of pleocytosis is indicative of an inflammatory response , most likely to the autonomously replicating virus . Increased levels of CSF white blood cells may account for the influx of T cells that could be the source of the short-lived cells that are amplifying the compartmentalized virus . We would expect that most of the infiltrating T cells are HIV-specific , although some lymphocytes may be migrating into the CNS due to a general inflammatory environment . The asymptomatic subjects in this group have the hallmarks of viral pathogenesis associated with neurological disease and may be at risk for transition to HAD . Third , we detected slow decay of compartmentalized variants in the CSF for the four remaining subjects with neurological disease . These subjects shared the feature of viral genetic compartmentalization but did not show high levels of pleocytosis . Additionally , this subject group had the lowest blood CD4+ T cell counts ( Table 1 ) , indicating a state of increased immunodeficiency . We suggest that these subjects have more profound immunodeficiency , which would allow even more extensive viral replication and compartmentalization in the CNS ( Figure 5C ) . Increased immunodeficiency would result in reduced trafficking of CD4+ T cells into the CNS , so these cells would no longer be present to amplify virus from local CNS tissue , consistent with the reduced pleocytosis in this group . The slow decay rate of virus in the CSF in the absence of inflammatory cells suggests that compartmentalized HIV-1 in the CNS of these HAD subjects is originating from a long-lived cell type , such as perivascular macrophages and/or microglia in the CNS . Virus is unlikely to be coming from T cells that are persisting in the absence of immune-mediated killing since there is still rapid viral decay in the peripheral blood . The CSF viral loads of all four subjects displaying slow decay were high , similar to subjects with rapid viral decay , suggesting that a large amount of compartmentalized virus is being produced by longer-lived cells in the CNS . This may suggest that peripheral , uninfected monocytes may migrate into the brain parenchyma and differentiate into perivascular macrophages to levels that can sustain high viral loads in the CSF . An influx of monocytes into the CNS could also allow the entrance of peripherally-infected monocytes , which would explain the slower decay we detected for shared variants in the CSF of these subjects . Our studies support a model where increasing levels of autonomous viral replication in the CNS first induces an inflammatory state that then progresses to neurologic disease with increasing immunodeficiency . More profound immunodeficiency ultimately reveals long-lived cells that are able to maintain independent replication of virus in the CNS . Several env gene markers have been described in viral sequences taken at autopsy and linked to the ability of HIV-1 to infect macrophages [12] , [13] . The CSF provides an alternative window on these viral sequences where the evolution of the virus and its properties can be followed over time and into the disease state . Viral genetic compartmentalization and other markers of CNS inflammation could also play an important role in defining subjects at risk of progression to neuropathogenesis in the absence of therapeutic intervention . This study was conducted according to the principles expressed in the Declaration of Helsinki . The study was approved by the Institutional Review Board of the University of California at San Francisco . All subjects provided written informed consent for the collection of samples and subsequent analysis . The samples from study subjects used for variant decay analysis were collected during previous studies carried out at the University of California at San Francisco . All subjects used in this study were HIV-1-infected subjects that were initiating highly-active antiretroviral therapy . Subjects 4012 , 4013 , 4014 , 5002 , 5003 , and 5005 were recruited from a study examining antiretroviral therapy responses in the CSF , and are described in more detail in ref . [67] . Serial blood plasma and cerebrospinal fluid ( CSF ) samples were collected at baseline prior to the start of therapy and at varying intervals thereafter . Plasma and CSF HIV-1 RNA concentrations were determined using the Amplicor HIV Monitor kit ( Roche ) . CSF white blood cell counts were measured by routine methods in the San Francisco General Clinical Laboratory . Drug resistance mutations were analyzed for CSF samples of subjects 4013 , 5002 , 4059 , and 7115 using the TRUGENE® HIV-1 Genotyping Test Resistance Report using GuideLines™ Rules 12 . 0 ( Bayer HealthCare ) . Viral RNA isolation , RT–PCR , and HTA procedures were conducted as previously described [24] , [39]–[41] . Briefly , viral RNA was isolated from blood plasma and CSF samples ( 140 µl ) using the QIAmp Viral RNA kit ( Qiagen ) . Prior to RNA isolation , all CSF samples were centrifuged at 2 , 500 rpm for 5 minutes to remove any contaminating cellular debris . Samples with viral RNA levels less than 10 , 000 copies/ml were pelleted ( 0 . 5–1 . 0 ml ) by centrifugation at 25 , 000×g for 1 . 5 hours prior to RNA isolation to increase template number and improve sampling . Reverse transcription and PCR amplification of the V1/V2 , V3 , and V4/V5 regions of env were conducted with 5 µl of purified RNA ( from 60 µl column elution volume ) using primers that have been previously described for V1/V2 [39] , [41] , V3 [51]; and V4/V5 [41] and using the Qiagen One-Step RT-PCR kit ( Qiagen ) as per manufacturer's instructions . Heteroduplex annealing reactions were conducted as previously described [39] , [40] . The heteroduplexes were separated by 6% native polyacrylamide gel electrophoresis for V1/V2 and V4/V5 HTA [24] , [39] , and by 12% PAGE for biotin-V3 HTA [51] . The HTA probes used in these studies have been previously reported: V1/V2 Ba-L probe [39] , [41] , V1/V2 JRFL probe [39] , [41] , V4/V5 NL4-3 probe [24] , V4/V5 YU2 probe [41] , and the V3 Mut-1 probe [51] . The HTA gels were dried under vacuum , and bands were visualized by autoradiography . For the biotin-V3 HTA procedure , the desired labeled bands were excised from the dried gels , the DNA was purified from the gel , and the V3 sequence was obtained as previously described [51] . Duplicate RT-PCR products were analyzed by HTA for each sample to validate sampling and ensure reproducibility of the HTA pattern at each time point . Any time points where the HTA pattern between the two replicates differed significantly ( >20% ) were not used in the data analysis . Percent difference values between plasma and CSF viral populations were calculated as previously described [39] , [41] . The dried HTA gels were exposed to a PhosphorImager screen , and the relative abundance of each detected viral variant ( heteroduplex ) was calculated using ImageQuant software ( Molecular Dynamics ) . The variant RNA concentration was calculated by multiplying the relative abundance of each individual variant by the total HIV-1 RNA concentration for that sample . Variants in the CSF were considered compartmentalized by HTA if they were either unique to the CSF or if they had a substantially higher copy number in the CSF compared to the plasma . Compartmentalized variant half-lives were calculated using the time points when the viral load initially dropped after the start of antiretroviral therapy .	Infection of the central nervous system ( CNS ) with human immunodeficiency virus type 1 ( HIV-1 ) can lead to the development of HIV-1–associated dementia , a severe neurological disease that results in cognitive and motor impairment . Individuals that are chronically infected with HIV-1 sometimes display unique viral variants in their cerebrospinal fluid ( CSF ) that are not detected in the blood virus population , termed CSF-compartmentalized variants . The cell type that produces CSF-compartmentalized virus throughout the course of infection has not been determined . We used a sensitive assay to detect compartmentalized variants in the CSF of subjects with and without neurological disease , and then measured the decay kinetics of compartmentalized virus when subjects were starting antiretroviral therapy . We found that compartmentalized virus decays rapidly in asymptomatic subjects . Additionally , we detected differential decay ( i . e . rapid or slow ) in subjects with neurological disease , and this was associated with the number of white blood cells in the CSF . Our data supports a model of HIV-1 infection in the CNS where compartmentalized virus is produced by a long-lived cell type ( slow decay ) , and this virus can be amplified by short-lived cells ( rapid decay ) that traffic into the CNS , but is increasingly produced from long-lived cells in the immunodeficient state .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "neurological", "disorders/infectious", "diseases", "of", "the", "nervous", "system", "virology/immunodeficiency", "viruses" ]	2009	Compartmentalized Human Immunodeficiency Virus Type 1 Originates from Long-Lived Cells in Some Subjects with HIV-1–Associated Dementia
The reported incidence of dengue fever increased dramatically in recent years in China . This study aimed to investigate and to assess the effectiveness of intervention implemented in a dengue outbreak in Ningbo City , Zhejiang Province , China . Data of a dengue outbreak were collected in Ningbo City in China by a field epidemiological survey according to a strict protocol and case definition . Serum specimens of all cases were collected for diagnosis and the virological characteristics were detected by using polymerase chain reaction ( PCR ) and gene sequencing . Vector surveillance was implemented during the outbreak to collect the larva and adult mosquito densities to calculate the Breteau Index ( BI ) and human biting rate ( HBR ) , respectively . Data of monthly BI and light-trap density in 2018 were built to calculate the seasonality of the vector . A transmission mathematical model was developed to dynamic the incidence of the disease . The parameters of the model were estimated by the data of the outbreak and vector surveillance data in 2018 . The effectiveness of the interventions implemented during the outbreak was assessed by the data and the modelling . From 11 August to 8 September , 2018 , a dengue outbreak was reported with 27 confirmed cases in a population of 5536-people community ( community A ) of Ningbo City . Whole E gene sequences were obtained from 24 cases and were confirmed as dengue virus type 1 ( DENV-1 ) . The transmission source of the outbreak was origin from community B where a dengue case having the same E gene sequence was onset on 30 July . Aedes albopictus was the only vector species in the area . The value of BI and HBR was 57 . 5 and 12 per person per hour respectively on 18 August , 2018 and decreased dramatically after interventions . The transmission model fitted well ( χ2 = 6 . 324 , P = 0 . 388 ) with the reported cases data . With no intervention , the total simulated number of the cases would be 1728 with a total attack rate ( TAR ) of 31 . 21% ( 95%CI: 29 . 99%– 32 . 43% ) . Case isolation and larva control ( LC ) have almost the same TAR and duration of outbreak ( DO ) as no intervention . Different levels of reducing HBR ( rHBR ) had different effectiveness with TARs ranging from 1 . 05% to 31 . 21% and DOs ranging from 27 days to 102 days . Adult vector control ( AVC ) had a very low TAR and DO . “LC+AVC” had a similar TAR and DO as that of AVC . “rHBR100%+LC” , “rHBR100%+AVC” , “rHBR100%+LC+AVC” and “rHBR100%+LC+AVC+Iso” had the same effectiveness . Without intervention , DENV-1 could be transmitted rapidly within a short period of time and leads to high attack rate in community in China . AVC or rHBR should be recommended as primary interventions to control rapid transmission of the dengue virus at the early stage of an outbreak . Dengue Fever is a mosquito-borne infectious disease which is caused by 4 distinct serotypes of dengue virus ( DENV-1 , DENV-2 , DENV-3 , and DENV-4 ) [1 , 2] . The disease is transmitted by female Aedes mosquitoes and has led to heavy disease burden in many countries [3 , 4] . One recent research indicates 390 million dengue infections per year ( 95% credible interval 284–528 million ) worldwide [3] . Another study estimates that 3 . 9 billion people in 128 countries are at risk of dengue infection [5] . The virus has become a leading cause of illness and death in tropics and subtropics . More important issues show that the virus has spread much wilder in recent decades . As reported by World Health Organization ( WHO ) , only 9 countries had experienced severe dengue epidemics before 1970 . But now , the disease is endemic in more than 100 countries in the WHO regions [1] . Dengue fever has been absent in China for about 30 years [2 , 6 , 7] . The reported incidence of the disease increased dramatically in recent years and cases have expanded from the coastal provinces of southern China and provinces adjacent to Southeast Asia to the central ( Henan Province ) and northern part ( Shandong Province ) of China [2] . The climate of most areas lying in tropical and subtropical zone is suitable for the reproduction of the vector and therefore provides a suitable environment for the disease transmission . When the virus imports in a desirable season , the transmission risk would be high . The rapid expansion of virus is an alert to conduct a risk assessment so as to understand the virus characteristics clearly . It could be better to investigate the transmission chain from an imported case to secondary cases in a non-epidemic area where there was no dengue transmission before than in epidemic area . Therefore , to understand a dengue outbreak in non-epidemic area has become an essential public health issue for the disease control and prevention . From 2005 to 2014 , no indigenous case was recorded in Ningbo City , Zhejiang Province , China . However , four indigenous cases occurred in the city in 2015 . Then on 18 August , 2018 , two suspected dengue cases were reported in two hospitals in Ningbo City . Serum specimens were collected and tested by polymerase chain reaction ( PCR ) method , two cases were confirmed as DENV-1 dengue by the laboratory of Ningbo Municipal Center for Disease Control and Prevention ( CDC ) . The first case was a 6-year old female who developed fever on 11 August . The second one was a 38-year old male whose onset date was on 15 August and his initial symptom was also fever . The two index cases were both connected to community A and had no travel history within 14 days before symptoms onset . These findings revealed that dengue virus had been transmitted among people in the community . By performing a standardized case-finding and field epidemiological investigation , the outbreak source was confirmed as a 65-year old male case living in community B where is about 500 meters far from community A whose onset date was on 30 July . He had the same E gene sequence as the two cases . Therefore , the outbreak provides a typical field to research the disease transmission in a non-epidemic area . Mathematical model is commonly adopted to explore the transmission of an infectious disease . The basic dengue model is the simple vector-host transmission model in which the host population is represented by a Susceptible–Infectious–Removed ( SIR ) model and the vector is assumed to remain infectious until death ( SI model ) [8 , 9] . In other researches , internal and external incubation were also considered and the vector and host transmission models were changed to Susceptible–Exposed–Infectious–Removed ( SEIR ) model and SEI model [10–12] . This study aimed to investigate a dengue outbreak in a non-epidemic area in Ningbo City in China by combining mathematical models we developed , to inspect the epidemiological and virological characteristics of the transmission , and to assess the effectiveness of intervention implemented in the outbreak . This effort of outbreak control and investigation was part of Ningbo Municipal CDC’s routine responsibility; therefore , institutional review and informed consent were not required for this study . All data analyzed were anonymized . We performed a time series study in dengue cases reported in Ningbo City from January 2005 to November 2018 . We performed a field epidemiological and modelling study to investigate the epidemiological and virological characteristics of a dengue outbreak , to simulate the incidence of the transmission , and to assess the effectiveness of intervention implemented in the outbreak from August to September 2018 in the city . To explore the seasonality patterns of the vector , a surveillance to investigate the density of the vector was also conducted from January to December 2018 . Ningbo City ( 28°51ʹ to 30°33ʹN , 120°55ʹ to 122°16ʹE ) , locating in the middle section of east coastline and having a population of more than 8 million , is a large city in Zhejiang Province , China . The city has an area of 3730 square kilometers and includes six districts , two counties , and two county-level cities . The ten subareas include 704 communities and 2519 villages . The climate is subtropical monsoon with a yearly average temperature of 16 . 2°C and a yearly rainfall of 1300 mm to 1400 mm . The rainfall occurs mostly from May to September . According to the dengue surveillance data from 2005 to 2018 , Ningbo City is a non-epidemic area in China during the past decade . Aedes albopictus was the only vector species in the city . The study community ( community A ) , which has a population of 5536 and an area of about 150 thousand square meters , is a relative separate community in Ningbo City . Case-finding and epidemiological investigation was conducted among all households in the community , the possible clinics , and hospitals based on the following case definition: Human biting rate ( HBR ) and larvae density were conducted during the outbreak . In our study , HBR was monitored every day in three surveillance sites locating in the community using the human-baited double net ( HDN ) trap from August 18 , 2018 to the end of the outbreak . One local volunteer was employed and rested inside the small bed net ( length × width × height: 1 . 2m × 1 . 2m × 2 . 0m ) and was consequently fully protected from mosquitoes during the survey . A larger bed net ( length × width × height: 1 . 8m × 1 . 8m × 1 . 5m ) was hung over the smaller net and was raised 50 cm above the ground . Both nets were protected from the elements by plastic-sheeting roofs but were not treated with any insecticide . One trained person helped to capture the mosquitoes for 30 minutes in and out the larger bed net per hour during high activity time . The vector species were identified in laboratory and the number of captured mosquitoes was recorded in each survey site to calculate the HBR ( per person per hour ) . We used Breteau Index ( BI ) to assess the larvae density of the vector . An investigation of BI was conducted from 18 August to 14 September in the community . More than 100 households were surveyed each time by screening all possible habitats and larvae or pupae of the vector were collected and numbered to bring back to the laboratory of Ningbo Municipal CDC to bleed them until they emerged to adult mosquitoes for species classification . Therefore , BI was calculated per 100 households in each day of the community . To identify the density and the seasonality of the vector in the city , the density of adult and larvae vector was monitored each month in 2018 from January to December . The adult density was monitored by using light-trap density . More than four surveillance sites were selected in each subarea of Ningbo City , among which two were selected in urban area such as parks and two were in villages’ households . A light-trap was hung about 1 . 5m above the ground in each mosquito capturing site from one hour before sunset to one hour after sunrise of the next day . Aedes mosquitoes were sorted and identified according to morphological characteristics . The light-trap density of the vector was calculated per light-trap per night consequently . BI was also monitored each month in four communities or villages . The investigation method was the same as the one used during the outbreak . A blood specimen was collected from each case and was sent to the laboratory of Ningbo Municipal CDC by cold chain and stored at -80°C freezer immediately in preparation for identification using PCR and gene sequencing . Nucleic acids were extracted using TGuide S32 Magnetic Viral DNA&RNA Kit ( TIANGEN BIOTECH [BEIJING] Co . , Ltd , Beijing , China ) , according to the manufacturer’s instructions . RT-PCR was used to detect the universal and specific primer . The primer sequence and the test procedure were performed according to “Diagnosis for dengue fever ( WS 216–2018 ) ” announced by National Health Commission of the People’s Republic of China . The E gene of dengue virus was amplified using one Step RNA PCR Kit ( TaKaRa [Dalian] Co . , Ltd , Dalian , China ) , according to the manufacturer’s instructions . The PCR procedures were performed as the protocol of E gene amplification which was deposited in protocols . io ( dx . doi . org/10 . 17504/protocols . io . zpuf5nw ) . The purified products amplified by PCR were used for gene sequencing directly . The sequencing was performed by Sunny Co . , Ltd , Shanghai , China . The phylogenetic trees for the E genes were constructed using the neighbor-joining method provided by MEGA 5 . 2 software . Nucleotides and amino acids similarity were compared by MegAlign software . Case isolation and treatment for all cases was implemented immediately on 18 August , 2018 . From 19 August , outbreak control strategy was implemented including household survey , case-finding , health promotion , environmental cleaning , adult vector control by applying insecticides as residual spraying and space spraying in the morning and at dusk , and surveillance of BI and HBR every day during the outbreak . For case isolation , all cases ( including suspected cases , clinically diagnosed cases , and confirmed cases ) were quarantined individually in hospitals until 5 days after illness onset date or until all the symptoms disappeared after 24 hours for those whose symptoms lasted longer than 5 days . The public health professionals performed active case finding in household survey every day according to the case definition . Several ordinary differential equation ( ODE ) models were developed to simulate the transmission of the virus [11–13] . In this study , the transmission model was developed based our previous study[14] . In the model , individuals were divided into the following six compartments: Sp , susceptible; Ep , exposed; Ip , infectious; Ap , asymptomatic; Rp , removed . Vectors were divided into the following three compartments: Sm , susceptible; Em , exposed; Im , infectious . Therefore , total number of the vectors ( Nm ) was calculated by adding up the three mosquito compartments . The human and vector compartments interacted with each other according to Fig 1 . The model ( Model 1 ) was therefore shown as follows: dSpdt=−βmpSpIm dEpdt=βmpSpIm−ωpEp dIpdt= ( 1−q ) ωpEp−γIp dApdt=qωpEp−γ′Ap dRpdt=γIp+γ′Ap dSmdt=ac ( Nm−nIm ) −βpmSm ( Ap+Ip ) −bSm dEmdt=βpmSm ( Ap+Ip ) − ( b+ωm ) Em dImdt=acnIm+ωmEm−bIm Nm=Sm+Em+Im In the equations , 11 parameters ( βmp , βpm , ωp , ωm , q , γ , γ’ , a , c , n , and b ) are included . The definitions of the parameters are summarized in Table 1 . According to the published researches , seasonality could be simulated by trigonometric function . Therefore , we assumed that the function of parameter c was as follows: c=cos[2π ( t−τ ) T] In the model , τ and T refer to simulation delay of the initial time in the whole season and the duration of the season , respectively . Model 1 was therefore changed to Model 2 which includes the seasonality of vectors and was shown as follows: dSpdt=−βmpSpIm dEpdt=βmpSpIm−ωpEp dIpdt= ( 1−q ) ωpEp−γIp dApdt=qωpEp−γ′Ap dRpdt=γIp+γ′Ap dSmdt=acos[2π ( t−τ ) T] ( Nm−nIm ) −βpmSm ( Ap+Ip ) −bSm dEmdt=βpmSm ( Ap+Ip ) − ( b+ωm ) Em dImdt=acos[2π ( t−τ ) T]nIm+ωmEm−bIm Nm=Sm+Em+Im Case isolation was implemented when the case was confirmed . With the intervention , the contact between the case and mosquitos was halted . However , asymptomatic infection is commonly hard to find out because there is no symptom of the infection for the surveillance . Therefore , in this condition , the transmission still exists from asymptomatic individuals to the vectors and from the vectors to human . The model ( Model 3 ) of case isolation is as follows: dSpdt=−βmpSpIm dEpdt=βmpSpIm−ωpEp dIpdt= ( 1−q ) ωpEp−γIp dApdt=qωpEp−γ′Ap dRpdt=γIp+γ′Ap dSmdt=acos[2π ( t−τ ) T] ( Nm−nIm ) −βpmSmAp−bSm dEmdt=βpmSmAp− ( b+ωm ) Em dImdt=acos[2π ( t−τ ) T]nIm+ωmEm−bIm Nm=Sm+Em+Im To reducing the HBR ( rHBR ) during the outbreak , bed net or mosquito repellents were encouraged to be used in the community . We assumed that the transmission relative rate from mosquitos to human and from human to mosquitos would be multiplied by an effective coefficient θ when the HBR was reduced by different levels ( 10% to 100% ) . Therefore , θ would be set as 0 . 9 , 0 . 8 , 0 . 7 , … , and 0 . 0 consequently . In this study , adult vector control ( AVC ) and larvae control ( LC ) were simulated . For AVC , adult vector were controlled by applying insecticides as space spraying during the outbreak . In this condition , the density of adult vector would decline sharply following an exponential model: Nm=j0ejt In the exponential model , j0 and j refer to the baseline ( t = 0 ) and the changing rate coefficient of the density . The reduction of the density is actually resulted from the changing of lifetime parameter b ( bt = b0 + j ) . The model ( Model 4 ) of AVC is as follows: dSpdt=−βmpSpIm dEpdt=βmpSpIm−ωpEp dIpdt= ( 1−q ) ωpEp−γIp dApdt=qωpEp−γ′Ap dRpdt=γIp+γ′Ap dSmdt=acos[2π ( t−τ ) T] ( Nm−nIm ) −βpmSm ( Ap+Ip ) − ( b0+j ) Sm dEmdt=βpmSm ( Ap+Ip ) − ( b0+j+ωm ) Em dImdt=acos[2π ( t−τ ) T]nIm+ωmEm− ( b0+j ) Im For LC , environmental management and modification , disposing of solid waste properly and removing artificial man-made habitats , covering , emptying and cleaning of domestic water storage containers , were adopted to prevent mosquitoes from accessing egg-laying habitats . Under this circumstance , the birth rate from egg to adult vector would decline sharply following an exponential model: at=a0ext In the exponential model , a0 and x refer to the baseline ( t = 0 ) and the changing rate coefficient of the birth rate . The model ( Model 5 ) of LC is as follows: dSpdt=−βmpSpIm dEpdt=βmpSpIm−ωpEp dIpdt= ( 1−q ) ωpEp−γIp dApdt=qωpEp−γ′Ap dRpdt=γIp+γ′Ap dSmdt=a0extcos[2π ( t−τ ) T] ( Nm−nIm ) −βpmSm ( Ap+Ip ) −bSm dEmdt=βpmSm ( Ap+Ip ) − ( b+ωm ) Em dImdt=a0extcos[2π ( t−τ ) T]nIm+ωmEm−bIm In addition , vector control activity was launched to improve community participation and mobilization for sustained vector control; to carry out the monitoring and surveillance of the vector as a routine work . There were 15 parameters , βmp , βpm , ωp , ωm , q , γ , γ’ , a , c , n , b , τ , T , j and x , in the models ( Table 1 ) . Symptoms of infection usually begin 4–8 days after the mosquito bite [11] . Here we simulated 6 days of the incubation period in our model , thus ωp = 0 . 1667 . After entering the mosquito in the blood meal , the virus will require an additional 8–12 days incubation before it can then be transmitted to another human [13] . Here we simulated 10 days in our model , thus ωm = 0 . 1000 . The ratio of symptomatic compared to asymptomatic infection is 2 . 2:1 [15] , thus q = 0 . 6875 . The infectious period is ranging from 3 days to 14 days [12 , 13] . Here we simulated 7 days in our model , thus γ = γ’ = 0 . 1429 . Aedes albopictus appears to be an efficient vertical transmitter of DENV-1 [16] . Vertical infection rates ( the percentage of offspring vertically infected ) of individual positive families are ranging from 1 . 4% to 17 . 4% [16] . Here we simulated 10 . 0% in our model , thus n = 0 . 1000 . The mosquito remains infected for the remainder of its life , which might be from 4 days to 50 days [13] . Since the outbreak occurred in August , we simulated 14 days in our model , thus b = 0 . 0714 . Since we simulated the transmission in August when the density of the vector was at the peak , which means that the vector has a balance population and the birth rate is equal to the death rate , thus a = b = 0 . 0714 . However , the vector population decreased after August and the two parameters were adjusted by the seasonality equation accordingly . Parameters βmp and βpm were calculated using the Models 1–5 to fit the reported cases data . The seasonality relative parameter τ and T were confirmed according to the onset date of the first case , the duration of the season , and curve fitting to the light-trap captured density and BI surveillance data in 2018 . Thus c was simulated consequently . Adult mosquito control parameter j was estimated by the exponential model fitted with the HBR surveillance data during the outbreak . Larvae control parameter x was estimated by the exponential model fitted with the BI surveillance data during the outbreak . To quantify the significance of the curve fitting , the chi-square test was employed by using a significance level of 0 . 05 , which means that the model fits the reported data well when P > 0 . 05 . The value of chi-square was calculated by the following equations: χ2= ( Ar−Ts ) 2Ts In the equation , Ar and Ts refer to reported data and simulated data , respectively . Absolute effectiveness ( AE ) and relative effectiveness of the interventions were calculated by the following equations: AE=TAR1−TAR2 RE=TAR1−TAR2TAR1×100% In the equations , TAR1 and TAR2 refer to the total attack rate ( TAR ) without and with intervention , respectively . A two-step simulation method was adopted . The first simulation step was to model the transmission source from community B to community A . The transmission source from community B is an imported case to community A . The importation can be simulated by using an impulse function according to our published research [17] . In this step , transmission source is modeled by the following impulse function: transmissionsource=impulse ( n , t0 , ti ) In the function , n , t0 and ti refer to number of dengue cases , start time of the simulation , and the interval of the transmission source . Considering that there was only one dengue case in community B during the outbreak , we simulated that a dengue case impulse into community A on 30 July , 2018 . Therefore , n = 1 , t0 = 0 , ti = ∞ . The second step simulation was to model the transmission in community A using Models 1–5 . Berkeley Madonna 8 . 3 . 18 ( developed by Robert Macey and George Oster of the University of California at Berkeley ) was employed for model simulation and least root mean square was adopted to judge the goodness of fit of curve fitting between the simulation models and the reported data . The simulation methods were the same as the previously published researches [17–21] . The chi-square test , which was adopted to quantify the significance of the goodness of fit , was performed by SPSS 13 . 0 ( IBM Corp . , Armonk , NY , USA ) . Since 8 parameters ( ωp , ωm , q , γ , γ’ , a , n , and b ) were estimated by references , there might be some uncertainty about them which might impact the results of models we built . In our study , sensitivity analysis was performed by varying the 8 parameters based on the methods we adopted in our previous research [18] . The ranges of the 8 parameters , which were shown in Table 1 , were split into 1000 values and simulated separately . The prevalence of each simulation were used to calculate the mean prevalence , mean–sd , and mean + sd . There were totally 124 dengue cases reported in Ningbo City from January 2005 to November 2018 , of which 52 . 42% ( 65/124 ) were imported cases . No indigenous case was reported during 2005 to 2014 . However , 4 indigenous cases were confirmed in 2015 . After that year , 55 more indigenous cases were confirmed till the end of 2018 ( Fig 2 ) . There was a dengue outbreak happened in 2018 , with totally 27 indigenous cases confirmed by PCR method . The index case was reported on 11 August , 2018 . Cases increased gradually and reached a peak on 19 August . As an integrated control intervention was implemented focusing on case isolation , adult and larva mosquito control , the incidence decreased accordingly . The last reported case was on 8 September , 2018 . The total attack rate ( TAR ) of the outbreak was 0 . 49% ( Fig 3 ) . Among these cases , 12 were male while 15 female cases with age range 6–86 years . The age and sex distribution of the cases was shown in Table 2 . By performing a standardized case-finding and field epidemiological investigation , the outbreak source was confirmed as a 65-year old male case living in community B where is about 500 meters far from community A . The case developed symptoms on 30 July , 2018 and only visited the hospital on 3 August and was confirmed as a dengue fever case by PCR method . The case had visited community A several times after disease onset and before hospital diagnosis . Of the 28 cases in communities A and B , specimens of 25 cases were obtained to detect the whole E gene sequences ( S1 Table ) . The length of the sequence was 1485bp which coded 495 amino acids . Compared with the gene sequences in the GenBank database , the gene type of all 25 cases were all DENV-1 . Phylogenetic analysis revealed that 25 virus strains in the outbreak highly aggregated into a branch which belonged to Genotypes I ( GI ) and were closely related to the virus in community B and the seven strains in Asia countries including China ( YunNan , TaiZhou , and Guangdong ) , Myanmar , Malaysia , and Singapore , were closely related to the prototype strain Myanmar/2015/MG894863 which was origin from Myanmar and export to Taiwan Province , China in 2015 , however they were much far from the prototype strain Hawaii/1944/KM204119 ( Fig 4 ) . The names and the GenBank accession numbers of the 25 viruses were shown in S1 Table . The all 25 isolates shared high levels of nucleotide identity and amino acid similarity with the seven GI reference strains: 98 . 0%– 99 . 7% and 96 . 6%– 98 . 2% , respectively , while moderate levels with the prototype strain Hawaii/1944/KM204119: 94 . 5% –94 . 8% and 92 . 1% –92 . 5% , respectively and low levels with the prototype strains of Genotypes II–IV ( G II–IV ) . With both the epidemiological evidence and the gene sequencing results , we concluded that the case that was found in community B was the transmission source of the outbreak in community A . According to the surveillance data of adult vector in 2018 , the median monthly density of the vector was 18 . 30 per light-trap per night with a range of 0 . 00 to 29 . 49 . The peak light-trap density of the vector occurred in July . No statistically significance was observed by the curve fitting ( χ2 = 7 . 996 , P = 0 . 434 ) , thus the seasonality of the density ( Fig 5A ) could be simulated by the following sine function: Density=26 . 56sin[2π ( t−2 ) 18] The median monthly BI was 22 . 08 , with a range of 0 . 00 to 33 . 22 . The peak light-trap density of the vector occurred in June . The seasonality of BI ( Fig 5B ) could be simulated well ( χ2 = 9 . 46 , P = 0 . 305 ) by the following sine function: BI=32 . 79sin[2π ( t−2 ) 18] Consequently , the seasonality of the vector population dynamic could be simulated by a trigonometric function with a cycle of 18 months . From Fig 5 , we can see that the model simulated 10 months ( 306 days ) in a year , thus , T = 712 . As the illness onset date was on 11 August , 2018 , the simulation delay of the initial time in the whole season is 170 days , thus τ = 170 . By fitting with the reported cases data ( S1 Fig ) , Model 1 ran well ( χ2 = 6 . 324 , P = 0 . 388 ) and reported the optimal values of βmp and βpm which were 1 . 4000 and 1 . 3613 , respectively . The results of sensitivity analysis showed that the model is not sensitive to the parameters ωm , a , and n . Our model is slightly sensitive to the parameter ωp , but sensitive to q , γ , γ’ , and b ( S2 Fig ) . On 18 August , 2018 , BI and HBR were 57 . 5 and 12 per person per hour . After that , BI and HBR decreased dramatically because of the implementation of the vector control interventions . BI could be simulated by the following exponential model: BI=57 . 5e−0 . 6200t HBR could be simulated by the following exponential model: HBR=12e−1 . 1310t The Malthusian models and surveillance data were shown in Fig 6 . According to the simulation model with no intervention , the total number of the cases would be 1728 with a TAR of 31 . 21% ( 95%CI: 29 . 99%– 32 . 43% ) . This result revealed that the AE and RE of the integrated outbreak control strategy implemented by Ningbo Municipal CDC were 30 . 72% and 98 . 44% , respectively . As shown in Table 3 , case isolation and LC implemented from 11 August has almost the same TAR and DO as no intervention . Different levels of rHBR had different effectiveness with TARs ranging from 1 . 05% to 31 . 21% and DOs ranging from 27 days to 102 days . The TARs could be simulated by the following logistic differential equation model: dTARdt=−2 . 9380TAR ( 1−TAR31 . 22 ) The logistic model fitted the data well ( χ2 = 0 . 199 , P = 1 . 000 ) and showed that the TAR began decreased quickly when rHBR was higher than 82 . 8% but decreased slowly when rHBR was higher than 91 . 8% ( Fig 7 ) . AVC had a very low TAR and DO . Combined-intervention strategy that consists of LC and AVC had a similar TAR and DO as that of AVC . Combined-intervention strategy that consists of “rHBR100%+AVC” , “rHBR100%+LC” , “rHBR100%+LC+AVC” , and “rHBR100%+LC+AVC+Iso” had the same effectiveness ( TAR , DO , AE , and RE ) and had a similar epidemic curve as reported data ( Fig 3 ) . Although four serotypes of dengue virus were isolated in China [2] , DENV-1 has become the predominant serotype since the 1990s [2 , 22 , 23] . Our gene sequencing findings also showed that the same serotype in the outbreak of community A in Ningbo City . This revealed that DENV-1 may still be the target virus in the future outbreak control . The virus is transmitted by female mosquitoes mainly by the species Ae . aegypti and , to a lesser extent , by Ae . albopictus . However , the published findings of the disease outbreaks showed that the virus was spread mostly through Ae . albopictus in China , especially in Guangdong Province [2 , 6] . The outbreak in Ningbo City revealed that the virus is expanding its transmission area through Ae . albopictus in China , and that it is an urgent issue to look insight of the transmissibility of the vector and to assess the vector specific countermeasures for disease control and prevention . The surveillance of BI and light-trap density of the vector showed that the seasonality of vector is obvious with a peak from June to August in the city . The high density of the vector provides a high receptivity and therefore a high probability for the transmission . This is also the case in Ningbo City on 18 August , 2018 . These findings of seasonality of vector density and the incidence were similar to transmission pattern of dengue in China , with indigenous cases mainly reported from July to November with the peak transmission mostly in the hot and humid seasons [2] . In our study , the ODE models were employed to fit the epidemic curves of the outbreak in a large community , the results of the Chi square test showed high good-of-fitness of our models to the reported data , suggesting that the models is suitable for this study and can be used to simulate the incidence of the outbreak and to assess the effectiveness of the countermeasures . The results of sensitivity analysis showed that our model is sensitive to q , γ , γ’ , and b . Therefore , more field epidemiological investigation is needed to explore the proportion of asymptomatic infections , the duration from illness onset to recovery , the infectious period , and the mortality rate of the vector . Our model shows that dengue virus could lead to a high attack rate ( higher that 30% ) within two months without any interventions . This finding suggests that the transmissibility of the virus might be high during the outbreak although the value of the transmissibility was not quantified . Therefore , in future control of dengue transmission , it calls for a high sensitivity and specificity of the surveillance systems including the diagnosis in hospitals or clinics , specimen collection and testing by PCR method or gene sequencing , and field epidemiological investigation . When an outbreak is confirmed and reported to the local CDC , our models could be employed to forecast the TAR and DO to simulate the future transmission scenarios and enable public health department to perform timely countermeasures . Our study showed that the high transmissibility led to the rapid transmission of virus in the community in a short period although the density of the vector decreased after August . The high transmissibility also led to the control difficulty . Under this condition , LC would not be effective because there is a biological delay from larva to adult vector , and during that delay , the virus would have been transmitted widely among the population . But , to our knowledge , LC has a long term effectiveness . In the low or moderate transmissibility scenarios , LC would be valuable . In addition , high proportion of the asymptomatic infection leads to the disease transmission even though symptomatic individuals are totally isolated . Therefore , case isolation is not the primary intervention during an outbreak . Fortunately , our study showed that the rHBR has distinctive effectiveness . However , because of the high transmissibility of the disease , the effectiveness is obvious when HBR is reduced to 17 . 2% of the initial value of the outbreak and would reach a satisfying effectiveness when reduced to 8 . 2% . In reality , HBR is hard to reduce down to 0% in short time . Although AVC is not the most effective among the single intervention , its TAR and DO were close to rHBR100% . Therefore , rHBR and LC are strongly recommended during the outbreak . On 18 August , 2018 ( the outbreak reported date ) , the HBR was 12 per person per hour . But after intervention implemented , HBR was reduced to 4 per person per hour ( down to 33% ) and 1 per person per hour ( down to 8 . 33% ) in the following two days . These surveillance data are similar to the simulated data . To our knowledge , when conducting AVC , the HBR will decrease . However , our outbreak did not provide a background for our investigation . During our outbreak control , AVC was conducted immediately covering the whole community , thus the procedure of interaction between AVC and HBR was shortened in a short time ( a day ) . The results of our models also showed that the TAR and DO of rHBR ( 100% ) , AVC , and rHBR ( 100% ) +AVC were almost the same ( Table 3 ) . Therefore , the interaction mechanism between AVC and rHBR remains unknown . More researches are needed to quantify the interaction . There are several limitations in our study . Firstly , our simulation model has dynamics the dengue incidence using the ODE model based on a small scale outbreak in a community . The small population based modelling might lead to some uncertainty . Fortunately , our previously research found that the simulation results would be stable when the population is larger than 2000 people [24] . In addition , the transmission among the population was clear by the virological- and epidemiological-based evidence . We got gene sequences data of about 89% confirmed cases during the outbreak . The vector surveillance data during the outbreak and in the year of 2018 also provided a solid grounding in the foundations of model . These efforts have improved the reliability of our modelling and effectiveness assessment of intervention . Secondly , the effect of the Jongdari typhoon ( 23 Jul—4 Aug ) was not taken into account , but this potentially may influence the number of dengue confirmed cases . Without intervention , DENV-1 could transmit rapidly within a short period of time and lead to high attack rate in community in China . The ODE models can be used to simulate the incidence dynamic of dengue outbreak and to assess the effectiveness of the countermeasures . AVC or rHBR should be recommended as primary interventions to control the dengue virus transmission rapidly at the early stage of an outbreak . Bed net or mosquito repellents were encouraged to be used in the community to reduce HBR , and space spraying of insecticides was recommended to control adult vector during the outbreak .	Dengue has led to heavy disease burden in China . The reported incidence of the disease increased dramatically in recent years and cases have expanded from southern to central and northern part of China . In this study , the findings include that DENV-1 can transmit rapidly with a short period of time and leads to high attack rate in community , and that rHBR or AVC should be recommended as primary interventions to control rapid transmission of dengue virus at the early stage of an outbreak . Therefore , dengue outbreak is at high risk in many areas in China because of the potential high receptivity ( widely distribution of Ae . albopictus ) and vulnerability ( high frequency of the importation ) of the transmission . The high transmissibility of the virus makes it hard and urgent to control the outbreak . Delayed intervention ( larvae control or case isolation ) is hard to show its effectiveness and the interventions without delay are strongly recommended . Bed net or mosquito repellents were encouraged to use in the community to reduce HBR , and space spraying of insecticides were recommended to control adult vector during the outbreak .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "dengue", "virus", "invertebrates", "medicine", "and", "health", "sciences", "pathology", "and", "laboratory", "medicine", "pathogens", "microbiology", "animals", "simulation", "and", "modeling", "viruses", "developmental", "biology", "rna", "viruses", "molecular", "biology", "techniques", "infectious", "disease", "control", "insect", "vectors", "research", "and", "analysis", "methods", "curve", "fitting", "infectious", "diseases", "mathematical", "functions", "artificial", "gene", "amplification", "and", "extension", "medical", "microbiology", "epidemiology", "mathematical", "and", "statistical", "techniques", "life", "cycles", "microbial", "pathogens", "molecular", "biology", "disease", "vectors", "insects", "arthropoda", "mosquitoes", "eukaryota", "flaviviruses", "polymerase", "chain", "reaction", "viral", "pathogens", "biology", "and", "life", "sciences", "species", "interactions", "larvae", "organisms" ]	2019	Incidence dynamics and investigation of key interventions in a dengue outbreak in Ningbo City, China
The identification of protein binding sites in promoter sequences is a key problem to understand and control regulation in biochemistry and biotechnological processes . We use a computational method to analyze promoters from a given genome . Our approach is based on a physical model at the mesoscopic level of protein-DNA interaction based on the influence of DNA local conformation on the dynamics of a general particle along the chain . Following the proposed model , the joined dynamics of the protein particle and the DNA portion of interest , only characterized by its base pair sequence , is simulated . The simulation output is analyzed by generating and analyzing the Free Energy Landscape of the system . In order to prove the capacity of prediction of our computational method we have analyzed nine promoters of Anabaena PCC 7120 . We are able to identify the transcription starting site of each of the promoters as the most populated macrostate in the dynamics . The developed procedure allows also to characterize promoter macrostates in terms of thermo-statistical magnitudes ( free energy and entropy ) , with valuable biological implications . Our results agree with independent previous experimental results . Thus , our methods appear as a powerful complementary tool for identifying protein binding sites in promoter sequences . Transcriptional regulation is the main mechanism for gene control in prokaryotes . In order to adapt optimal protein expression to nutritional and environmental conditions , a cascade of transcriptional regulators works as signal transducers determining the accessibility of RNA polymerase to bacterial promoters . In the last years , high throughput approaches have been confirmed as powerful tools for a better understanding of the regulatory networks that govern key aspects of cell physiology , such as the mechanisms leading to pathogenesis or the acclimation to xenobiotics and hostile environments , among others [1]–[4] . However , successful transcriptome sequencing requires the generation of comprehensive transcriptome profiles that rely on the isolation of a sufficiently large number of reads to detect those biologically relevant transcripts , that represent a relatively small proportion of the cDNA library [5] . Moreover , those procedures are time consuming and , in many cases , the budget for sequencing costs constrains the total number of reads that can be obtained [6] , [7] . Therefore , computational methods emerge as valuable complementary approaches for prediction or further validation of high throughput results [8] , [9] . Mostly , a statistical approach to the study of sequences is adopted , leading to a general lack of methods based on the physical mechanism of protein-DNA interactions . A possibility to tackle the problem is the microscopic study of protein-DNA interaction [10]–[12] , but this approach demands huge computer facilities and it is restricted to few base pairs up to the date . In this sense , coarse-grained models arise as powerful tools to model biological systems , speeding up the computation and allowing to get a deeper insight in the physical interactions [13] , [14] . Adopting this strategy , we develop a coarse-grained model that allows for the analysis of promoter sequences and the identification and characterization of protein binding sites , likely related to transcriptional activity in the genome of the nitrogen-fixing cyanobacterium Anabaena PCC 7120 . Cyanobacteria are the only prokaryotes able to perform oxygenic photosynthesis , being key contributors to fixation . The ability of some cyanobacterial strains to fix atmospheric nitrogen or the formation of harmful blooms by toxigenic species , among other properties , evidence their ecological relevance [15] . Besides , cyanobacteria are an excellent model for the study of multicellularity in prokaryotes [16] and potential sources for novel drugs derived from their secondary metabolites [17] . The genome of Anabaena PCC 7120 contains 7 , 211 , 789 base pairs ( bp ) and 6 , 223 genes organized in a 6 , 413 , 771 bp chromosome and 6 plasmids [18] . Anabaena PCC 7120 has been used for long time as a model for the study of prokaryotic cell differentiation and nitrogen fixation [19] . More recently , the experimental definition of a genome wide map of transcriptional start sites ( TSSs ) of Anabaena together with the analysis of transcriptome variations resulting from the adaptation to nitrogen stress have provided a holistic picture of this complex process [20] . The problem of protein-DNA recognition is a widely debated issue , yet far to be fully understood . In this sense , it has been widely reported how the physical properties of the DNA chain result in key functional consequences in this process . DNA local structure highly influences some transcription factors ( TFs ) binding [21]–[23] . Thermal stability and bubble formation ( i . e . local long-lived transient openings in the DNA strands ) has also been extensively reported to correlate with several DNA functions , such as the recombination rate , single nucleotide polymorphism , DNA replication or gene transcription [24]–[27] . In this regard , the relation between bubble formation and the location of protein binding sites , is a lengthly , controversial debate , greatly nourished by the study of Peyrard-Bishop-Daxouis ( PBD ) model [28] , [29] . This mesoscopic model was initially intended to reproduce the DNA melting transition , though it has been widely used afterwards for studying bubble formation on DNA promoters , likely correlated with biological relevant sites in the sequence , such as the TSS or the TATA box [30]–[35] . Despite the lack of consensus on whether PBD model is suitable for predicting protein binding sites [36]–[39] , strong evidence supports this idea , showing clear correlation between regions with high propensity to form bubbles , and the presence of binding sites of DNA-interacting proteins such as RNA polymerase , [30]–[32] , [40] or some TFs [33] , [34] , [41] , [42] . Even more , succeeding revisions of this model showed clear relation between flexibility profiles and location of TSSs [43] . Grounded on these evidences , we propose a physical model for protein-DNA interaction in promoters [44] , based on the coupling of a generic particle with the sequence-dependent bubble formation . This simple model is combined with a suitable analysis method [45] allowing the detection of biologically relevant sites , namely TSSs , on promoters of a prokaryote genome . In order to prove the capacity of prediction of the computational methods developed in [44] and [45] for identifying the TSSs of a promoter , we have analyzed the result of simulating the dynamics of nine promoters of Anabaena PCC 7120 . We have analyzed the simulations outputs and built systematically the relevant macrostates of the system . In every case , our analysis algorithm finds the TSS as one of these states , yielding in addition thermodynamic parameters ( e . g . free energy , entropy ) that allow their physical characterization and thus further biological discussion . In this regard , our method arises as a complementary tool that , from physical principles , finds protein binding sites ( we focus on TSSs ) and characterizes them , allowing to discuss the strength -in terms of RNA production- of such sites , something not achievable by statistical methods . Remarkably , in this case the base pair sequence is the only previous information required . Thus , our numerical outcomes are independent numerical predictions to be confronted with previous or future experimental results . We base our model on a modification of the PBD model [28]–[31] , [35] to include the interaction with a generic particle as a sliding protein coupled with the sequence . PBD model reduces the complexity of DNA to a set of units that represent the base pairs of the chain ( see Fig . 1 ) . The only degrees of freedom are the coordinates which stand for the opening of each base pair . The total Hamiltonian of the model accounts for two phenomenological interactions , the intra-base and the inter-base potentials , , where is the linear momentum of the base pair and its reduced mass . The potential describes the inter-base pair or stacking interactions . The election is the anharmonic potential [28] whose elastic constant is for small openings but drops to for large . The parameter sets the length scale for this behavior . The original PBD model uses Morse potential for the intra-base pair interaction . Nevertheless , a successful modification includes an entropic barrier which accounts for solvent interactions with open base pairs [35] , [46] , [47] . This modification sharpens the thermal denaturation and stabilizes the bubbles , reproducing in a more realistic way the experiments [35] , [46] , [47] . We include this effect adding a gaussian barrier [35] , thus . Sequence dependence is introduced only in this potential term as the interaction is stronger if the base pair is C-G than if it is A-T ( see Text S1 for the complete set of parameters ) . Sequence-dependence can be also introduced in the stacking potential parameters , a modification that accounts for flexibility properties of the DNA chain [40] , [43] , [48] . Inspired on the one-dimensional diffusion stage of DNA-interacting proteins [49] , we include a new degree of freedom to the traditional PBD model . This new degree of freedom consists on a brownian particle that moves along the DNA chain ( see Fig . 1 for a schematic representation of the total system ) interacting with it through a phenomenological potential which depends on , the coordinate of the Brownian particle along the DNA molecule , and the DNA instantaneous configuration ( 1 ) This potential creates a classical field composed by a sum of gaussian wells centered at each base ( ) and whose amplitude depends on the opening of the base pair . The term allows a linear dependence for low saturating the interaction for large in order to avoid self-trapping . In this sense , the particle interacts more intensely with open regions of the sequence . In addition , the base pairs are also affected by the particle , so that they will be more likely to be opened if the particle is within its range of interaction . The model introduces only three new parameters , as the longitudinal scale over which the particle slides is adimensional ( ) . The interaction intensity and width are set so that bubbles span around base pairs , an adequate value for the kind of processes studied here [50] . The parameter saturates the interaction around , typical value for open base-pairs [50]–[52] . The model is simulated by integrating numerically the Langevin equations for the chain base pairs and the particle using the stochastic Runge-Kutta algorithm of fourth order [53] ( see Text S1 for explicit formulation of the equations of motion ) . Each of the DNA sequences we study is simulated in five different realizations , each one covering , with a preheating time of . For sequences up to base pairs , these times are enough to ensure equilibrium and ergodicity . In addition , since one-dimensional diffusion times of binding proteins are in the range of milliseconds , our simulation times are reasonable from a biological perspective . The simulation temperature is . We use periodic boundary conditions for the diffusing particle and fixed boundary conditions for the sequence , adding base pair clamps at the end of each sequence to provide “hard-boundaries” and avoid undesirable end effects . Relevant observables from the trajectories can be obtained , mainly the base pairs mean position , where is the number of realizations and the simulation time of each realization , and the particle's trajectory histogram . The large dimensionality of the system requires a method to reduce the number of coordinates while keeping the relevant information of study . PCA [54] is one the most popular methods to reduce systematically the dimensionality of a complex system . PCA performs a linear transformation by diagonalizing the covariance matrix , and thus removing all internal correlations . It has been proved that , by ordering the eigenvalues decreasingly , the few first principal components contain most of the fluctuations of the system , and thus can be chosen as convenient reaction coordinates [35] , [55] , [56] . We project the base pair trajectories into the first five eigenspaces , describing thus the system in terms of the first five principal components and the particle trajectory . With this choice we keep over the of the fluctuations . The Conformational Markov Network has been proven to be a useful and powerful tool to analyze trajectories from high dimensional systems , such as those from Molecular Dynamics simulations [45] , [57]–[59] . This representation is obtained by discretizing the conformational space explored by the system in order to build a complex network . Each node in the network represents a discretized region of the conformational space , a conformational microstate , weighted according to the fraction of trajectory visiting such microstate . The links of the network coincide with the observed transitions between microstates , and are thus directed and weighted . We build the Conformational Markov Network of our system by considering the posible positions of the particle along the chain , and binning each of the five principal components into bins . Typically , the Conformational Markov Network is formed by a large number of nodes which prevent a direct interpretation of the results . In order to extract relevant information about the physical states of the system and its relevance in the dynamics , we split the network into its basins of attraction , i . e . regions in which the probability fluxes ( ) converge to a common state ( attractor ) of the network . To do so , we apply the stochastic steepest descent algorithm , developed in [45] , building a coarse grained representation of the former network . From this basin network , the Free Energy Landscape ( FEL ) can be represented as a hierarchical tree diagram ( dendrogram or disconnectivity graph ) [60] , [61] , by assigning to each node a free energy according to its weight where is the weight of the heaviest basin . This magnitude is used as a control parameter , increasing it step by step from the weightiest node , so that new nodes arise , together with their links ( see Text S1 for a more explicit exposition of the algorithm ) . The disconnectivity graph represents each basin of attraction hierarchically ordered according to its free energy , while the connections among them stand for the barriers needed to jump from to another ( see below and Text S1 for plots of the disconnectivity graphs or dendrograms ) . We define now the macrostates of the system by clustering every basin separated by a free energy barrier lower than , as the system transits among them within short waiting times . In fact , we can check how they represent qualitatively similar physical configurations . Each macrostate has an assigned weight . We want to calculate free energy differences between specific and non-specific states . The basin network contains a huge number of low populated states , see [35] , that constitute transitionary states between well defined attractors of the system . Physically , they are short-lived transitionary states where the particle diffuses until it binds to a target site . We determine these non-specific states as every basin with a population and calculate free energy differences between specific and non-specific states as , where is the total weight of all non-specific states . In addition , we define the entropy of a macrostate as . Up to our knowledge , most works concerning PBD model limit themselves to the study of short promoter sequences , without justifying the study of this region alone , or how would the model behave in coding regions . In order to cover this gap , we have simulated the behavior of three complete genes from Anabaena PCC 7120 . We use here the PBD model without including the interacting particle , as we wish just to check in which regions from a whole gene bubbles form more easily . The results allow us to compare the occurrence and intensities of the fluctuations detected in the promoter and the coding regions , validating our further analyses restricted to the promoter sequences . Figure 2 shows the first four PCA eigenvectors for the analyzed genes with the promoter and codifying regions highlighted . Very localized eigenvectors indicate strong fluctuations in the region of maximal amplitude . As we can see in Fig . 2 , the first eigenvector is delocalized , with small amplitude , accounting for the overall fluctuations of the whole sequence . Nevertheless , the three next eigenvectors are highly localized in specific spots of the sequence . Remarkably , these sites appear in the promoter sequence . Thus , when considering a complete gene within PBD model , most of the system fluctuations occur in the promoter sequence; this is , bubbles form with higher probability there , while the codifying region remains on average closed . This reveals the role of the DNA sequence in the DNA dynamics , and its influence on the DNA-protein interaction problems , supporting strongly that some binding sites in the promoter sequence can be characterized as regions where bubbles form easily , enhancing protein interaction . We have used the complete model ( chain and particle ) to analyze nine promoter sequences comprising to base pairs . In addition , we have chosen promoters with different features , five with a single well characterized TSS ( alr0750 , argC , conR , furA and nifB ) , while four of them exhibit multiple TSSs ( furB , ntcA , petF and petH ) [62]–[69] . Figure 3 shows the base pair opening profile for each promoter sequence with the TSSs highlighted . The particle trajectory histograms are also plotted . In any case , a peak appears close to the TSS , meaning that , on average , bubbles form with high probability around it . In turn , the particle is attracted by this site , as it dwells with high probability around the TSS . As it has been pointed out in several studies , the PBD model by itself has been successfully used to analyze promoter sequences , finding protein binding sites where bubbles form with high probability , so allowing the identification of TSSs or the TATA-box [30] , [32] . Nonetheless , introducing this additional degree of freedom appears as a key feature for our purposes . We are mimicking an hypothetical searching mechanism that indeed affects the dynamics of the system . In the PBD model alone , opening events appear as rare excitations of the unique ground state , where the whole chain is closed . The particle enhances chain opening , stabilizing the bubbles , that last for longer times ( around two orders of magnitude longer ) , enriching the free energy landscape . In addition , bubbles span over a larger number of base pairs , typically around , which is a consistent number if we attend to those that form in transcriptional processes [51] , [52] . It is also remarkable that the opening probability is not strictly related with the A-T content of the local sequence . Although it is clear that long A-T stretches form “softer” regions in the sequence that can open easier , this intuitive argument does not necessarily applies always . The interplay between the sequence and the dynamics is much more complex . The nonlinearity in the Hamiltonian , the long-range cooperativity of the model and the disorder of the sequence revealed in its heterogeneity affects directly the equilibrium and dynamical behavior of the model , being essential to understand the actual breathing dynamics of DNA , as it has been pointed out in previous studies [30] , [31] , [40] . Interestingly , besides the peaks centered on the TSSs , other regions exhibit high probability to form bubbles . Many of these peaks correspond to typical regulation sites of bacteria , such as those located at or from the TSS , also claimed to be related with bubble formation [30] , [40] . These regions appear thus as candidates for possible binding sites of other TFs that are known to be influenced by the physical properties of the DNA chain . Nonetheless , we focus our discussion just on the TSS , as they have been systematically identified in the genome of Anabaena PCC 1720 . In order to analyze the sequences in a more systematic way we apply the FEL analysis described in the methods section . This algorithm allows us to define the most relevant states in the dynamics characterizing them from a quantitative point of view . So far , we have shown which regions in the promoter sequences exhibit a higher probability to form bubbles and to be visited by the particle . Nonetheless , these magnitudes give just qualitative information , as the average do not inform about the importance of opening events in the system . The real interest of our model and method is the possibility of giving quantitative measures about the “strength” of the different sites in the sequences , specially interesting in those promoters with several TSSs . Each site can be characterized by the thermodynamical magnitudes calculated from the FEL landscape analysis . We present together the data extracted from the simulation and analysis methods in Table 1 . For each of the nine analyzed sequences we show the weight , free energy difference with respect to the non-specific states and the entropy of the TSSs state , all previously defined . We include also other non-identified states in case they appear relevant in the dynamics . Most populated states suppose most stable states , giving rise to high free energies differences . The entropy is the multiplicity of such macro states . Even if the free energy is high , a low entropy would indicate that this macro state is made up of few , yet very populated , basins , physically meaning that the state is very localized ( narrow bubbles ) . The opposite case would indicate that the algorithm finds many , less populated basins that represent the same macrostate . This duality could indicate different regulation behaviors that are further addressed in the Discussion section . To illustrate the FEL , Fig . 4 shows the free energy dendrograms of three chosen promoters ( see Text S1 for the six remaining dendrograms ) . For the sake of clarity , we do not show the region corresponding to non-specific basins ( where , defined above ) . The position of each basin on the vertical axis informs about its stability , while their hierarchical arrangement about the barrier needed to jump between each state . The dendrogram or disconnectivity graphs provides thus valuable and intuitive information about the thermodynamic and kinetic properties of the FEL of each promoter . Groups of basins separated by barriers lower than are highlighted by a color circle , defining the macrostates of the system according to the criterion detailed in Methods section . We plot together the physical state associated with it , showing also the fraction of trajectory they occupy . Such states correspond to a large bubble located on the target site , with the particle centered there . In most cases , the most populated macrostate , and thus the most stable one , coincides with an excitation in the TSS region . Other non identified sites also suppose very populated macrostates , suggesting the possibility of additional regulation sites as it is discussed in next section . Our method arises thus as a powerful tool to complement experimental results , providing additional physical information about the relative importance of these sites in regulation processes . In this work , we propose the use of a coarse-grained model for protein-DNA interaction to analyze promoter sequences , allowing the detection and characterization of protein-binding sites ( we focus on the TSS ) . The proposed model is based on physical principles and inspired on a relatively simple idea: certain DNA-interacting proteins ( as RNA polymerase ) couple their binding to DNA bubble dynamics . Due to this , we base our model on a PDB representation of the DNA chain -having been proven to reproduce DNA bubble dynamics successfully- and couple it to an additional degree of freedom representing the protein . In the framework of this model and by using a free energy landscape analysis , we have studied promoters of Anabaena PCC7120 , allowing the detection and characterization of the TSSs . Upon genome analysis and TSSs detection , high-throughput approaches , such as proteomics , are commonly used , resulting in an enormous amount of data in a relatively short period of time . However , analysis of raw data to end up in genome annotation or TSSs mapping is a demanding , time-consuming task , necessary for taking advantage of this information that may delay a more detailed analysis of specific issues . Among the large variety of these methods ( see [70] , [71] for review of most existing methods ) a great amount of valuable information is obtained , resulting in highly efficient analysis of genome that , nonetheless , generally lacks a base on the physical mechanism of protein-DNA interaction . In this sense , our model and analysis method adopt a different strategy , not willing to compete in time performance with statistical-based techniques , but allowing a deeper understanding on the driving processes of protein binding . As a consequence of that , we are able not only to identify the TSSs , but also to characterize them in terms of physical magnitudes , allowing discussions about the strength of each site . The nine promoters of cyanobacterium Anabaena PCC7120 studied in this work have been chosen in order to make the most of our model , without forgetting about its limitations . The genome of Anabaena PCC 7120 is well-known and the positions of TSSs have been defined under different metabolic conditions [72] . Firstly , it is remarkable how the different TSSs in the analyzed genes coincide with relevant states in the dynamics of the model , characterized as the heavier basins . In order to relate the information obtained with possible biological interpretation , we have analyzed a set of genes exhibiting several TSSs and whose regulation has been well characterized [67] , [68] , [73]–[77] . This choice allows us to assess directly the potential relation between the binding free energy values displayed in Table 1 for each of the located sites , and the relative strength of different TSSs associated to the same gene . Among them , it is worth to mention the case of the ntcA promoter . The average opening shown in Fig . 3 reveals how the three existing TSSs in this base pairs sequence [78] are clearly identified , agreeing also as sites which the particle visits with high probability . As displayed in Table 1 , the relative free energy ( with respect to the NS states ) of the three TSSs is quite different . Indeed these values are in very good agreement with the occurrence and behavior of the three TSSs experimentally determined [78] , [78]–[80] . TSS2 , located at position , produces a constitutive transcript regardless of the culture conditions , while TSS1 ( position ) is only used in the absence of nitrogen . Finally TSS3 ( position ) is also active under all conditions , but its use is highly induced under nitrogen deprivation . Table 1 displays a remarkably low free energy for TSS1 , indicating that the presence of this macrostate is low in the dynamics , suggesting that its expression might be enhanced under more restrictive conditions . On the other hand , TSS2 and TSS3 appear as strong binding sites , covering both a large fraction of the total dynamics . These values are in good agreement with the ntcA transcription level at these sites under the correspondent conditions of nitrogen availability . FurB , petF or petH show also consistent results . The TSSs of the three promotores are clearly identified , coinciding with the experimental positions [66] , [72] , [81] . Determination of TSSs for FurB promoter using the primer extension technique unravels revealing two TSSs at positions and from the ATG , both with similar intensities ( [66] ) . Our in silico analysis is in good concordance with such conclusions , as we find two major macrostates with very similar weight ( and ) with an excitation just on these positions . The resulting profiles when the promoters of petF and petH are analyzed also display several preferred macrostates . Primer extension assays revealed a single TSS for the petF gene located at 100 bp upstream the translation start site [82] . More recently , high throughput analysis showed two TSSs for petF , at and , bp , in a better agreement with our predictions . Transcription of petH , encoding ferredoxin-NADP+ reductase takes place from a constitutive promoter at bp from the ATG and a NtcA activated promoter ( TSS at bp ) . According to the proposed model , both TSSs are found as relevant macrostates in the basin network , although not as high peaks in Fig . 3 . Indeed , the constitutive TSS ( ) exhibits a higher probability ( ) than the non-constitutive one ( Table 1 ) , indicating that the model is consistent with the experimental observations . Concerning the five remaining promoters , high peaks are found around their single TSS , coinciding with the most ( or one of the most ) populated macrostates as we have defined them ( Table 1 ) . The case of conR is where our model works worse , as a significantly more relevant state appears in the dynamics . It should be noted that most experimentally determined TSSs have been obtained under standard culture conditions or under nitrogen deprivation , and the existence of additional TSSs under different conditions -impossible to account explicitly in our model- cannot be discarded . In addition , it must be noted that the model is not considering exclusively DNA-RNA polymerase interaction , but the influence of DNA breathing dynamics on protein binding . In such sense , additional binding sites for other proteins which are influenced by mechanical changes in the DNA conformation may also be detected . We have compared our numerical results to the existing experimental ones on TSSs positions and intensities . Nonetheless , it is important to note that our method identifies additional relevant regions of the promoters that have not been experimentally probed yet . We shall mention the cases of promoters furA , conR or nifB where very populated macrostates appear aside from the discussed TSSs . Although we do not exclude the possibility of false positives , these macrostates may be related with unknown regulatory regions . Thus , our results suggest further experiments to search possible new relevant activity regions . Moreover , additional TSSs might appear if studied under different culture conditions , revealing the complexity of transcriptome profiles even in the case of simple organisms such as bacteria . To finish , we have already mentioned studies discussing the influence of bubble formation on certain DNA-binding proteins aside from RNA-polymerase [33] , [34] , [41] , [42] . Being our model based on general physical features , additional macrostates found through our method might indicate the existence of binding sites for further regulatory proteins which participate in transcriptome processes of Anabaena PCC 7120 . Anabaena PCC 7120 has been shown to be an ideal experimental system to probe our numerical method . As it has been displayed , our results agree current experimental knowledge and propose possible new relevant activity regions . However , the model can be applied to the study of promoter sequences in many other organisms . Being the identification of protein binding sites in promoter sequences a key problem to understand and control regulation in biochemical and biotechnological processes , our methods appears as a powerful complementary tool in this scientific endeavor .	Binding of specific proteins to particular sites in the DNA sequence is a fundamental issue for gene regulation in molecular biology and genetic engineering . A deep understanding of cell physiology requires the analysis of a plethora of genes involving characterization of their promoter architectures that determine their regulation and gene transcription . In order to locate the promoter elements of a given gene , experimental determination of its transcription start site ( TSS ) is required . This is an expensive , time-consuming task that , depending on our requirements , could be simplified using computational analysis as a first approach . Nevertheless , most computational methods lack a physical basis on the protein-DNA interaction mechanism . We adopt here this strategy , by using a simple model for protein-DNA interaction to find TSS in a bunch of cyanobacteria promoters . We make use of physical tools to characterize these TSS and to relate them with biological properties as the relative strength of the promoter . Our study shows how a model based on a coarse-grained description of a biomolecule can give valuable insight on its biological function .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "physics", "statistical", "mechanics", "biology", "and", "life", "sciences", "physical", "sciences", "computational", "biology", "biophysics", "biophysical", "simulations" ]	2014	Mesoscopic Model and Free Energy Landscape for Protein-DNA Binding Sites: Analysis of Cyanobacterial Promoters
In adults , motion perception is mediated by an extensive network of occipital , parietal , temporal , and insular cortical areas . Little is known about the neural substrate of visual motion in infants , although behavioural studies suggest that motion perception is rudimentary at birth and matures steadily over the first few years . Here , by measuring Blood Oxygenated Level Dependent ( BOLD ) responses to flow versus random-motion stimuli , we demonstrate that the major cortical areas serving motion processing in adults are operative by 7 wk of age . Resting-state correlations demonstrate adult-like functional connectivity between the motion-selective associative areas , but not between primary cortex and temporo-occipital and posterior-insular cortices . Taken together , the results suggest that the development of motion perception may be limited by slow maturation of the subcortical input and of the cortico-cortical connections . In addition they support the existence of independent input to primary ( V1 ) and temporo-occipital ( V5/MT+ ) cortices very early in life . The infant visual brain is immature at birth . To date there is no direct functional evidence from awake infants showing how the various cortical areas of human visual cortex develop . The available evidence suggests that infants are capable of discriminating motion-direction soon after birth [1–3] , and that sensitivity to global-motion continues to mature slowly over the first 4–7 y in humans and 2–3 y in monkey [4–7] . The protracted development ( after an early emergence ) of global-motion sensitivity is attributed to late maturation of higher-level motion areas , such as temporo-occipital complex ( V5/MT+ ) [2 , 8–12] . Previously it was hypothesized that the visual cortex develops in a hierarchical fashion with higher-order areas developing later , driven by feed-forward projections from previously developed lower-order cortical areas [9 , 13] . There is also evidence that some aspects of motion processing , such as the control of opto-kinetic eye movements ( OKN ) , are limited by the development of subcortical relative to cortical function [14] . Brisk monocular OKN responses can be elicited in newborn infants , but only by motion in the temporal-to-nasal direction . Early development of subcortical mechanisms ( probably of the Nucleus of the Optical Tectum ) may mediate the eye-following response in this direction , while the directional sensitivity in the nasal-to-temporal direction , emerging later at about 10 wk , may be mediated by cortical mechanisms [11 , 15] . Even more peripheral factors , such as photoreceptor efficiency or myelination , may provide important constraints on some functional development . For example [16–18] chromatic and achromatic contrast sensitivities are limited by retinal immaturities , rather than by immaturity of cortical processing . This , together with evidence from infants who suffered from visual deprivation [19] , suggests that the visual cortex may mediate vision very early after birth , provided that the incoming input is mature and transmits reliable visual information . The subcortical input to associative visual cortex can undergo strong reorganization during development . In adult monkey and human , V5/MT+ input originates mainly from cortico-cortical connections and from independent konio-cellular LGN-Pulvinar projections that bypass V1 [20–22] . However , in the first few post-natal weeks , the major inputs to MT+ in marmoset monkey are a disynaptic connection from the Retino-Pulvinar projections ( from the medial portion of the Inferior Pulvinar ) [23] . This input may mediate the directional response of MT+ neurons observed very early postnatally . The present study examines whether the cortical mechanisms of motion processing are functional in very early infancy , and in particular aims to compare the selectivity of MT+ and primary visual cortex ( V1 ) to flow motion to highlight a possible differential development . To date there is no direct evidence about the functional development of the various cortical areas of human visual cortex or of their Blood Oxygenated Level Dependent ( BOLD ) response selectivity from awake infants , although a few studies have shown that it is feasible to record BOLD acoustic responses [24–26] , or BOLD flash responses during deep anaesthesia or sleep [27–31] . Reliable BOLD activation to flashes in calcarine sulcus has been reported , but many studies found a reduction of the BOLD response to visual stimulation respect to no stimulation , particularly evident after the eighth week of age [28 , 30] . The origin of this negative , or more generally delayed BOLD response , is still under debate , and the issue is further complicated by the effect of the use of anaesthesia or sedation that have been shown to modulate BOLD response in human and animal models . In awake infants , studies using Near Infrared Spectroscopy ( NIRS ) [13 , 32–36] , which measures signals of similar origin to the MRI-BOLD response , show evidence for positive responses in both young ( around 8 wk of age ) and older ( 4–6 mo of age ) infants to visual stimulation . The discrepancy between MRI- and NIRS-BOLD responses is still unresolved and may reflect either the state of sedation , the sleepiness of infant or different types of visual stimulation . Interestingly , at 6 mo NIRS BOLD amplitude is reduced in response to homogenous flashes and increases in response to a structured high contrast pattern [34] , suggesting that the type of visual stimulation matters . Here , by measuring BOLD responses to coherent versus random flow-motion stimuli in cooperative infants , we demonstrate that the major circuits mediating motion perception are operative very early , by 7 wk of age . We found a selective response to coherent flow-motion in the temporo-occipital area , cuneus , posterior parietal and posterior associative insular cortex , with similar activation and localization as adults for visual motion [37–39] and vection perception [39–44] . Previous works have shown that resting-state connectivity networks are well segregated in newborn and even in pre-term infants [45–48] and mature rapidly in the first 2 y of age [48–50] . Here we demonstrate adult-like functional connectivity between many motion selective associative areas , but not between primary visual cortex and temporo-occipital ( putative MT+ ) and posterior-insular cortices . The results localize for the first time cortical areas with high selectivity to visual stimuli in the first weeks of life , revealing an unexpected early maturation of the cortical system for motion processing . The flow-selective area in the Cuneus/Pre-Cuneus was labelled in only seven infants , while it was always present in all except one adult; in another subject , the V6 and PIVC/PIC ROIs were not significant at α = 0 . 05 . To assess the congruency of ROI localization , we built an anatomical template by brain segmentation using dedicated tissue probability maps ( illustrated in S4 Fig ) and we performed a fixed-effect GLM multi-subject analysis ( see Methods ) . Fig 4 ( top panels ) shows the results mapped on the same anatomical template of S4 Fig for infants and in Talairach atlas for adults , at the same statistical threshold of q < 0 . 05 ( false discovery rate [FDR] corrected and without mask ) . In adults , the occipital foci associated both with dorsal and with ventral pathways [53 , 54] showed a stronger response to coherent motion , with the exception of V1/V2 areas . A negative response was also labelled in the posterior insular cortex of the left hemisphere ( slice at z = 8 ) , while the right PIVC/PIC became labelled lowering the threshold at p < 0 . 05 ( lower panels ) . This result is consistent both with the large variation in phase of the PIVC/PIC responses ( see Fig 3 and S3 Fig ) and with the large variation in the localization between subjects ( see Table 1 ) . The variation in phase was to be expected given that the BOLD amplitude of PIVC/PIC is modulated by the strength in vection perception [39–43]: subjects with stronger vection illusion might show stronger negative responses . Similarly , also the large scatter in position of the PIVC/PIC region has been already reported [43] . Both factors indicate that the multi-subject GLM may not be the most suitable technique to locate this area . Nevertheless it is reassuring that when decreasing the threshold to p < 0 . 05 ( bottom panels ) only one clear additional cluster in the posterior insular cortex became labelled , reinforcing the suggestion that PIVC/PIC localization can vary considerably across subjects . In infant multi-subject GLM , no activity was significantly labelled for V1/V2; all the ventral areas along the fusiform and the lingual gyri had negative responses , while the adult responses were positive , indicating a late development of direction selectivity for the ventral pathways . In contrast , the dorsal areas , like V3/V3A , LO , TOS , V6/V6A , and MT+ , were clearly labelled , and showed a preference for the coherent flow motion . In the right hemisphere ( slice at ζ = 16 ) a negative response corresponding to the location of the PIVC/PIC was labelled . As in adults , decreasing the threshold at p < 0 . 05 ( bottom panels ) also the PIVC/PIC in the left hemisphere became detectable , suggesting that , as in adults , for infants , the localization variability of this area is high . There were also other negative and positive clusters in the temporal lobe ( see for example slice at ζ = 0 ) that can be also observed in adults , but whose function and circuitry are still unknown . The similarity in the localization pattern between infants and adults strongly suggests that the development of visual primary and dorsal associative cortex for motion processing is quite well advanced by 7 wk of age . In nine out of twelve infants , we were also able to record resting-state activity during spontaneous sleep . Fig 5 shows the correlation of the resting-state activity between each pair of the ROIs of Fig 2 , for all possible combinations in infants and adults . In most regions of infants and adults the correlation was strong and significantly non-zero ( indicated by stars in the individual cells in Fig 5 ) . Significant correlations reflecting functional connectivity were present in both adults and infants between homologue areas of the two hemispheres , including PIVC/PIC , V6 and MT+ . The correlation matrixes were similar in adults and infants , corroborating the similarity of the BOLD responses with two main exceptions . In infants , the correlation between bilateral V1 activity and MT+ was significant ( although just marginally for the right hemisphere ) , but it was positive , while in adults the correlation was strong and negative , as previously reported [55 , 56] . Similarly , the PIVC/PIC activity in infants showed a significant positive correlation , while in adults it was negative . Our results show that visual motion cortical areas have a high degree of functional specialization in very young infants: direction selectivity is well established in many dorsal cortical regions at 7 wk of age . To distinguish between coherent and random flow-motion , the neuronal mechanisms must be selective to motion direction and also integrate local-motion signals along complex trajectories . Direction selectivity in adult primates occurs initially in V1 , while large spatial integration along complex trajectories occurs in V6 and V5/MT+ , and propagates to higher visual cortices [42] . Given that 1-mo-olds ( or younger ) infants show defensive motor responses such as blinking and avoidance head movements in response to large-field radial expansion patterns [1 , 3] , and that motion direction selectivity has been demonstrated behaviourally [2] , it is reasonable to assume that selectivity for motion direction emerges around 7 wk of age or earlier in many regions of the dorsal pathways , including MT+ and V6 , both crucially important for motion perception . This suggests that cortical processing of motion is more mature than has been proposed at this age [8 , 11 , 15] , with a relatively stronger response of MT+ compared with occipital areas in infants . Interestingly , V5/MT+ is situated ventro-laterally , just posterior to the meeting point of the ascending limb of the inferior temporal sulcus and the lateral occipital sulcus , and corresponds , on individual subjects , almost precisely with Flechsig's Field 16 [57] , one of the areas that is myelinated at birth , corroborating the suggestion that V5/MT+ should be considered as a primary sensory area with an early maturation [58] . An earlier maturation of the V5/MT+ is also consistent with ERP evidence at 5 mo of age of a more lateralized selectivity to coherent motion in infants with respect to adults [59] , indicating that the delayed maturation of more occipital areas ( with respect to V5/MT+ ) is protracted over several months . Considering that in adults there is evidence that direction selectivity is computed in MT+ rather than being inherited from inputs of earlier cortical regions [22 , 60] , it is feasible that the MT+ directional selectivity BOLD response at 7 wk of age is also computed locally , and it is independent from V1 input . The neuronal processing of the ventral and dorsal visual pathways are quite distinct and mediate different visual functions [53 , 54] . Ventral area , like V4 and fusiform gyrus , showed preference for coherent motion in adult , although these areas are not considered crucial for the perception of motion direction and probably inherit the direction selectivity from dorsal areas . Interestingly in infants , these areas prefer incoherent motion as illustrated in Fig 4 . These results suggest that the early development of cortical specialization might be a prerogative of the dorsal pathway and visual motion analysis , corroborating the evidence of a later maturation of the ventral stream in children [61] sub-serving object and face recognition [62 , 63] . Overall , our data suggest that both primary and dorsal associative cortices develop direction selectivity at a similar stage . The motion selectivity of V6- , Pre-Cuneus and associative vestibular cortex has been related to the illusory percept of body-motion induced by large flow field ( vection ) [39–41 , 43] . The similar activity of these regions in infants and adults suggests that infants may have a sense of vection and hence a sense of body position . The continuous stimulation of the semi-circular canals in the womb may induce early development of the vestibular system , so the few weeks of visual experience may be sufficient to endorse the vestibular-visual integration necessary for vection perception . While this is consistent with early development of the vestibulo-ocular response in newborn infants [64] , it is surprising that visuo-vestibular cortex is selective to motion so early , as optimal multisensory integration develops slowly in humans [65–67] . Our data show that the hemodynamic response of infants is delayed with respect to adults , similar to that reported at similar ages in response to acoustic [25] and motor/tactile stimulation [51] . The delay of the response is of the order of few seconds , consistent with the delay observed by Arichi [51] , and cannot explain the phenomenon of Negative BOLD corresponding to 180 degree of phase shift and delay on the order of tens of seconds . It is worthwhile noting that other laboratories have observed positive BOLD visual responses in this age window , and negative BOLD responses for older infant [30] , reinforcing the suggestion that the Negative BOLD phenomenon is related to the functional maturation of the cortex . However , the fact that NIRS studies in awake infants consistently find a positive BOLD responses at all ages [13 , 32–36] suggests that the MRI-BOLD negative response may not result from synaptogenesis or the cerebral metabolic rate for oxygen , but rather from the alertness state of the infant . In addition , the positive versus negative BOLD response could be biased by the use of flash stimuli versus black background , rather than a spatial pattern whose contrast is modulated over time . This is supported by the finding that the phase of V6 and MT+ became negative in response to the flow versus blank ( S3 Fig ) , and blank stimuli engage usually little attention in infants . Interestingly , a recent NIRS study showed a negative BOLD response to flashes and a positive BOLD response to high contrast flickering patters in 6-mo-old infants [34] . It is well known that flash stimuli are strongly subject to suppression during blinks , and the suppression may be even stronger when the eyes are shut or during sleep . Repeating our study with older collaborative and awake infants could provide important clues to resolve the origin of negative MRI-BOLD response in infants , although the difficulty in recording stable signal increases rapidly with age because of the increased motility of older infants . During post-natal development , there is a continuous refinement of anatomical connections in the mammalian visual system , which are initially diffuse and then progressively pruned to increase target selectivity . Between 0 and 4 mo , synaptic density increases rapidly [68 , 69] , and myelination intensifies along the visual pathway . Diffusion Tensor Imaging ( DTI ) studies suggest that at birth all major fiber systems are in place , despite low anisotropy value [70] and incomplete myelination [71] . The functional connectivity observed in our data showed an adult-like correlation between inter-hemispheric ROIs and between associative cortices , in agreement with the anatomical DTI findings of well-established fiber connections . However , connectivity between V1 and MT+ and PIVC/PIC were different than in adults . In infants , the optic radiations are diffuse ( see for example Fig 3 in [72] ) and may project not only to V1 but also to other visual cortices . In addition , other transient projections not originating from LGN may be functional in young infants . These diffuse afferent projections to the cortex might mediate the strong BOLD response observed here in all areas , without implicating adult-like cortico-cortical connections , which at this age are highly immature [69] . This model would also explain the weak ( although significant ) correlation between MT+ and V1 , given that these areas would receive independent V1 inputs to generate the strong BOLD response selective to motion . It would also reinforce the suggestion that , in infants , BOLD directional selectivity is computed in MT+ rather than being inherited from inputs of earlier cortical regions . However , some caution is required in comparing adult and infant functional connectivity given that the infant , but not the adult , data were acquired during sleep . Several studies show that functional connectivity during sleep and relaxation [73] may be different , but the differences are not significant for the sensory-visual resting state networks [74] . Resting-state functional connectivity appears to be invariant even under anaesthesia [75] . Therefore , we do not foresee that different depths of state of alertness would affect the outcome of our studies . In addition , most of the functional connectivity strengths are equal in infants and adults , in agreement with previous results [48]: only V1-MT+ was very different . It would be rather strange for sleep to affect so selectively only this connectivity . As in human , in cat and monkey , the putative corresponding area of human MT+ is well developed at birth , with a developmental time-course similar to primary visual cortex [57 , 58] . V5/MT+ neurons of newborn monkey show directional selectivity [8] . Interestingly , in monkey it has been suggested that the fast maturation of V5/MT+ is mediated by strong and direct retino-pulvinar-cortical projections , which are later pruned during development [23] . In adult humans the LGN-MT+ projections , which bypass V1 , probably overtake the functional role of this direct retinal-pulvinar input [20–22] , helping to explain several motion abilities retained after lesion of V1 ( such as “blindsight” ) . The existence of a direct retino-pulvinar input to MT+ also in human infants would explain the weak functional connectivity between MT+ and V1 . However , other explanations cannot be excluded , including delayed development of feedback projections , which may have an overall suppressive effect , explaining the negative correlation in adults . The suppression of V1 activity by feedback connections would also be consistent with the negative BOLD response to motion ( against noise ) of V1 in adults , and the smaller negative ( not significant ) modulation in infants , suggesting an overall delayed development of V1 circuitry respect to MT+ . Whatever the anatomical connections between the various areas in infants , the extensive , well-developed network of visual associative areas selective to motion at 7 wk , demonstrated by this study , suggests that direction selectivity , the most fundamental property for motion perception , emerges very rapidly , and simultaneously in many dorsal cortices . This developmental pattern may be consistent with the early suggestion [76] that direction selectivity development is limited by peripheral factors such as myelination and speed of neuronal signal transmission . Timing of arrival of spike input to cortex is crucial for motion perception . Jitter in temporal delay of the order of milliseconds is sufficient to disrupt or even invert motion [77]; disorganization of input spike trains or high levels of noise may be sufficient to impede the formation of direction selectivity at the cortical level . As input motion signals become more reliable and organized , the cortex may be capable of developing the complex neuronal circuitry for direction selectivity and trajectory integration only within a very limited time window . Similar fast cortical development has been observed previously for stereo acuity or chromatic contrast ( for review see [76] ) , and also for these functions , the limiting factors may be peripheral . If the present suggestion of independent emergence of direction selectivity in MT+ is confirmed by additional evidence , we should revise the widely accepted idea of a slow , uniform , and progressive maturation of the various cortical hierarchy levels [9] and of cortical motion responses [59] . We propose an alternative model in which all dorsal cortical regions have equal potential for fast maturation and for developing appropriate circuitry , once the input begins to transmit reliable neural information . The study was approved by Ethics Review Board of Fondazione Stella Maris and Regional Paediatric Committee ( Meyer Paediatric Hospital ) . Written informed consent was obtained by all subjects or their parents before the experiment . Sixteen ( five females and eleven males ) healthy , full-term , awake infants , mean age 7 . 7 ± 1 . 2 wk ( range = 6 . 6–10 . 4w ) , were scanned by a 1 . 5T MR scanner ( GE Healthcare , USA ) . Data from three subjects were not included in the analysis because of movement artifacts or deep sleep , and for one subject we obtained only the anatomical dataset , leaving twelve subjects with full data . All infants were assessed by an expert paediatric neurologist by means of the Hammersmith Neurological exam [78] and a battery of visual function tests , including fixation to white and black targets and the ocular following response . The infants were again assessed at 3 mo with a neurological follow-up that confirmed normal development . After the MR exam and before any data analysis , an expert child neuro-radiologist performed also an examination of the anatomical scans to reveal possible anomalies and all infant brains were referred as normal . The MR protocol comprised acquisition of a three-dimensional ( 3-D ) T1w FSPGR sequence ( TR/TE = 12 . 28/5 . 14 , isotropic voxel = 1x1x1mm3 ) , and an fMRI session of three different series ( GRE-EPI , TR/TE = 3000/50 , FA = 90° , FOV = 240 X 240 mm , matrix = 96 X 96 , slice thickness = 3 mm ) . fMRI experiments included two series of 84 time points ( 4’12” duration , block design , six periods of alternating coherent flow motion versus a blank or a random motion condition , each lasting 21 s ) . In nine subjects ( age range = 6 . 6–8 wk ) a resting state fMRI series ( 120 time points , 6’00” duration , no stimulus presentation ) was successfully acquired during the spontaneous sleep . All functional series were preceded by four dummy time points to allow signal stabilization . Stimuli were generated in Matlab ( TheMathWorks ) and displayed on LCD goggles ( Resonance Technology ) positioned inside the head coil 10 to 15 cm from the infant eye , giving a visual field of about 27 X 20 degrees . Infant eyes were refracted with retinoscopy at a distance of 87 cm ( the virtual image of the goggles is greater than 1 m ) . All infants were in the normal range between 0 and +2D: given the viewing distance , no additional correction was introduced . Fixation of infant gaze was monitored by an infrared camera installed within the goggles ( sample frequency 60 Hz , see final segments of S1 Movie for fixation example ) . Usually infants scanned the stimulus very attentively and this , together with the real-time movement signals , gave an indication of their level of wakefulness . Infants entered the magnet lying down on the scanner table , swaddled in a sheet by an expert neonatal nurse to calm the baby and to reduce movement . One operator ( MCM , SAC , or LB ) accompanied the child into the magnet , taking the sphinx position , surrounding the body of the child with her arms and wrapping his/her head with her hands in order to maintain a physical contact , to facilitate calming and to reduce motion . The operator was in constant communication with the staff at the acquisition terminal . Infant ears were protected by cotton wool padding at the auditus of the external auditory canal , and sound-attenuating headphones . In order to reduce the stress of the babies , we adopted specific strategies in accordance with the character and routines of each individual child . Some preferred to use a pacifier to relax and to sleep , especially for anatomical and resting state fMRI scanning . If the infant moved or changed position the operator could reposition the goggles by tilting them . In these cases , the recording was continued , but the series was cut offline ( see details below ) . Importantly , the operator could not see the visual stimulus delivered , or if the goggles were switched off for the resting state scan . During the anatomical scan , the goggles were switched off and nine of the infants fell asleep , allowing us to acquire the resting-state scan . The same MR protocol described for children was repeated in nine healthy adults ( six females and three males , mean age 35 y ) . In adult subjects , resting state scans were acquired by asking to the subjects to relax and stay still with eyes closed . The visual motion stimuli comprised 100 dots , half black and half white , of 1 . 3 degrees diameter , moving at constant speed ( 5 deg/s ) with limited lifetime of 10 frames ( at 60 Hz about 160 ms ) . For coherent motion the trajectory of the dots changed gradually from expansion , inward-spiral , rotation , outward-spiral , contraction , then repeating the cycle again ( see S1 Movie , initial segments ) . The full cycle lasted 2 s . The random motion was constructed using the same dot velocities shuffled randomly over the dots , so it had matched local motion power . The dots covered all the visual field except the central 2 degrees . Dot density was kept constant in all displays , and collision between dots was not allowed . The mean luminance was 20 cd/m2 and contrast 0 . 85 ( for further information , see [37] ) . Three-dimensional T1-weighted images of each infants were visually inspected to select good quality data in order to create a specific infant template for this study . Only one infant’s data ( subject DE ) was discarded because of the poor grade of anatomical images; all the remaining twelve 3-D datasets were selected and analysed with the SPM8 package with the Diffeomorphic Anatomical Registration Through Exponentiated Lie algebra ( DARTEL ) algorithm [79] . DARTEL uses diffeomorphic warping to obtain a study-specific template , with an optimal inter-subject realignment and an improved co-registration of small structures , such as those of infant brains . A pre-conditional step for the use of this algorithm is the segmentation of brain tissues using standard tissues probability maps . For infants , we used the segmented partial estimate volumes for grey matter , white matter , and cerebral spinal fluid derived by the 36 MRI dataset of 3-mo-old infants available at http://jerlab . psych . sc . edu/NeurodevelopmentalMRIDatabase/index . html as tissue probability maps [80] . The obtained atlas is shown in S4 Fig . Data analyses were performed with BrainVoyager ( BV , Brain Innovation ) . First , for each functional series , motion spikes or periods of heavy movement that could not be compensated for were eliminated , and the resulting separate intervals were considered as independent shorter samples . Infant motion was estimated by calculating the six rigid body parameters ( three for translation and three for rotation ) across time . Time series with head motion greater than 4 mm ( translation ) or 5° ( rotation ) were excluded . For each subject and stimulus , the mean and maxima values of the six rigid body parameters registered in the survived ( and used ) time courses are reported in S2 Table . The recorded eye movements were analysed to verify fixation and alertness of the infant; intervals of sleep or jerk movements that could not be adequately compensated were discarded from analysis , typically for the duration of half period of stimulation . For all infants , we were able to select at least half duration of the complete recorded run ( on average , 4 . 9 ± 0 . 8 periods for coherent versus random flow motion , 4 . 8 ± 0 . 9 periods for coherent flow motion versus blank , and 110 ± 10 data points for resting state stimulus ) for further analysis , with the remaining periods discarded due to head motion or sleep . Data preprocessing included mean intensity adjustment to compensate for interscan intensity differences , temporal interpolation and re-sampled to compensate for slice dependent time differences ( sinc function ) , 3-D motion correction ( sinc interpolation ) , and high-pass temporal filtering ( GLM-Fourier approach , two cycles ) . Functional data were co-registered on the three-dimensional anatomical T1-weighted images by using an affine alignment with the standard BV nine parameters ( three for translation , three for rotation , three for FOV scale ) . Infant anatomical datasets were in turn transformed into the AC-PC coordinate system by applying a rigid transformation ( 6 parameters; 3 for translation and 3 for rotation ) , whilst adult data were transformed into the standard Talairach space . For each subject , BOLD responses were analysed using a GLM modelling the regressor of interest ( by convolving a box-car function for each stimulation block with a gamma variate function for the hemodynamic response ) and six spurious movement-regressors ( outputs of the 3-D motion correction procedure ) . The first stimulus ( flow-motion versus blank ) was used to select a bilateral region with very strong response , located along the left and right calcarine sulci ( p < 0 . 001 in infant and p < 10−10 in adults ) . Many other strong activities in the occipital pole were present but not analysed . The signal registered in this ROI ( labelled V1-seed ) was used to calculate correlation maps by cross-correlating the V1-seed ROI signal with all other brain voxels , using temporal delays in the range 0 to 21 s , given the unknown hemodynamic delay of infant BOLD responses . S2 Fig shows the maps of four different infants computed for positive and negative correlation at zero delay and at p < 0 . 05 . The low conservative threshold of p < 0 . 05 marked about 40% of all voxels for delay = 0 s and additional 26% for delay = 3 s . Computing the maps for the remaining delays in the range of 3 to 21 s added only about 30% new voxels . Repeating the same analysis only on the selected ROIs ( see below the selection procedure ) , we obtained that 80% of the voxels were labelled at delay = 0 s and 94% combining delay 0 s and 3 s . Given the high proportion of labelled voxels , in the analysis presented here we applied a mask obtained by the union of those computed at 0 and 3 s delays . Both for infants and adults , the GLM analysis of the coherent versus random motion was performed within this mask , and all positive or negative foci at a threshold of p ≤ 0 . 05 were labelled ( see Table 1 for statistical thresholds of the ROIs ) . These ROIs were located by a neuro-radiologist ( with more than 20 y of experience with newborn infants ) along the major infant sulci and gyri . In adults , we restricted the analysis to the major regions of interest selective to motion [37–39] that could be easily traced in infants . Table 1 reported the maximum foci localization in millimetres in the AC-PC coordinate system for infants and in Talairach space for adults , and the respective Z-score for areas that were identified in all adults and in most infants . For multiple foci areas , the coordinates corresponded to the positions of the centre of mass of the aggregate ROIs . In particular , for the occipital activities we reported the union of the bilateral activations of V1 ( V1-seed ) ( reporting the average coordinates of the two foci ) ; an area located dorsally just posterior of the parietal-occipital sulcus , which in adults correspond to the union of V6 and V6A; an area positioned posteriorly in Pre-Cuneus/Cuneus , close to the parieto-occipital sulcus and anterior of the most extreme periphery of V1/V2 . This area could be the human equivalent of the pro-striate [81] , not to be confused with the Pre-Cuneus selective motion area reported by Cardin et al [39 , 82] . Many other occipital activation sites were labelled in response to coherent versus flow motion in the mask , but given the difficulties in distinguishing between them and locating them in all the sulci , we decided not to report or analyse them in detail . The area MT+ was the easiest to locate along the inferior temporal sulcus , both in adults and infants . In adults , this ROI includes MT and MST . An area showing a differential response to coherent versus incoherent motion was labelled in the fundus of the most apical portion of the posterior insula , corresponding to the PIC+ complex of the visual-vestibular network [39–41 , 43 , 44] . The complex comprises two different foci named PIVC and PIC respectively , that can be differentiated on the selectivity to vestibular stimulation . We labelled this area as PIVC/PIC given that we used only visual and not vestibular stimulation . For each ROI , and for both the first ( coherent flow-motion versus blank ) and the second stimulus ( coherent versus random flow-motion ) , we extracted the time-courses averaged across periods of repetition , and then across subjects . We also evaluated the Signal to Noise ratio ( S/N ) and the phase of the response , performing an FFT on the extracted time-course before averaging . S/N was defined as the amplitude of the ratio of the fundamental harmonic and the root-mean-squared amplitude of the two frequencies closest to the fundamental [83] . The phase of the response is a good measure of the hemodynamic delay . In adults , a standard hemodynamic model corresponds to a phase value of 64 deg in our plots . The relationship between the phases of the responses across different areas and single subjects is reported in S3 Fig , both for infants and adults . To evaluate statistical significance between the adult and infant phases of the responses for each ROIs , we calculated the distances of each vector data point from the resultant of the vectorial average and its standard deviation . Then assuming a normal spread , we calculated the variation in phase associated with the predicted 2-D Normal dispersion of data around the average . Results of average , SD and t-tests are reported in S1 Table . In order to consider the infants as a homogenous group , inter-subjects alignment was performed . In particular , the obtained atlas and the volumetric dataset of each subject were co-registered ( by affine transformation ) to one single subject ( MG ) with optimal quality of the anatomical image . The same transformations , one for each subject , were in turn applied to the respective functional data . One subject ( DE ) was discarded from this kind of analysis for poor quality of three-dimensional anatomical dataset . The 11 co-registered functional datasets were used for a multi-subjects analysis of the infants group , by using a fixed–effect ( FFX ) GLM-based analysis and a statistical threshold corrected for False Discovery Rate ( FDR ) q < 0 . 05 and a minimum cluster size of 81 mm3 , corresponding to three functional voxels ( Fig 5 , left panel ) . Similarly a fixed–effect ( FFX ) GLM-based analysis at the same statistical threshold of q ( FDR ) < 0 . 05 and a minimum cluster size of 135 mm3 was performed also on adult data , after co-registration in Talairach space ( Fig 5 , right panel ) . In both infant and adult multi-subjects GLM analysis , no mask was applied . In agreement with previous work that show a great variability of the PIC+ complex location and selectivity between subjects [43] , the statistical threshold of the multi-subject GLM had to be reduced to p < 0 . 05 uncorrected before activity of similar cluster-size could be identified at the expected PIVC/PIC anatomical location in infants and in adults . Resting state series were used to study the correlation between the ROIs described above . For each subject and for each pair of ROIs , the cross correlation of the signals extracted in the resting state was calculated and the value reported in the correlation matrix . Note that given the high-pass temporal filtering the functional connectivity maps could also show a negative correlation . The striking negative correlation between V1 and MT+ in adults that is often reported also depends on the state of the subject ( fixation versus eyes-closed ) [56] . Note that the resting state in infants , but not in adults , was acquired during sleep . Two mean correlation matrices were obtained averaging the single-subject matrices for infants and adults , respectively . In order to achieve a value of significance for each value in the mean correlation matrices , for each group of subjects , the signals of all single subjects were normalized and concatenated , obtaining the signal of a super-subject for each ROI . The calculation of cross correlation of these signals between each pair of ROIs corresponds to the average across subjects of the correlation matrices and gives the significance of the values in terms of p-value . Statistical analysis ( two-tailed t-test ) was performed between correlation matrices of the two groups of subjects .	While it is known that the visual brain is immature at birth , there is little firm information about the developmental timeline of the visual system in humans . Despite this , it is commonly assumed that the cortex matures slowly , with primary visual areas developing first , followed by higher associative regions . Here we use fMRI in very young infants to show that this isn’t the case . Adults are highly sensitive to moving objects , and to the spurious flow projected on their retinas while they move in the environment . Flow perception is mediated by an extensive network of areas involving primary and associative visual areas , but also vestibular associative cortices that mediate the perception of body motion ( vection ) . Our data demonstrate that this complex network of higher associative areas is established and well developed by 7 wk of age , including the vestibular associative cortex . Interestingly , the maturation of the primary visual cortex lags behind the higher associative cortex; this suggests the existence of independent cortical inputs to the primary and the associative cortex at this stage of development , explaining why infants do not yet perceive motion with the same sensitivity as adults .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[]	2015	BOLD Response Selective to Flow-Motion in Very Young Infants
Inflammasomes are cytosolic multi-protein complexes that detect infection or cellular damage and activate the Caspase-1 ( CASP1 ) protease . The NAIP5/NLRC4 inflammasome detects bacterial flagellin and is essential for resistance to the flagellated intracellular bacterium Legionella pneumophila . The effectors required downstream of NAIP5/NLRC4 to restrict bacterial replication remain unclear . Upon NAIP5/NLRC4 activation , CASP1 cleaves and activates the pore-forming protein Gasdermin-D ( GSDMD ) and the effector caspase-7 ( CASP7 ) . However , Casp1–/– ( and Casp1/11–/– ) mice are only partially susceptible to L . pneumophila and do not phenocopy Nlrc4–/–mice , because NAIP5/NLRC4 also activates CASP8 for restriction of L . pneumophila infection . Here we show that CASP8 promotes the activation of CASP7 and that Casp7/1/11–/– and Casp8/1/11–/– mice recapitulate the full susceptibility of Nlrc4–/– mice . Gsdmd–/– mice exhibit only mild susceptibility to L . pneumophila , but Gsdmd–/–Casp7–/– mice are as susceptible as the Nlrc4–/– mice . These results demonstrate that GSDMD and CASP7 are the key substrates downstream of NAIP5/NLRC4/CASP1/8 required for resistance to L . pneumophila . Inflammasomes are multi-protein complexes that assemble in the cytosol of infected or damaged cells and initiate host defense by functioning as a platform for the recruitment and activation of caspase proteases [1 , 2] . The NAIP/NLRC4 family of inflammasomes is especially well-characterized and has been shown to be activated upon detection of specific bacterial proteins , such as flagellin , via direct binding to various NAIP family members [3] . Ligand-activated NAIPs recruit and co-oligomerize with NLRC4 , which in turn recruits and activates Caspase-1 directly , or indirectly via the adaptor protein ASC . Recently it was shown that ASC can also recruit and activate Caspase-8 downstream of NAIP/NLRC4 [4 , 5] . The NAIP5/NLRC4 inflammasome was originally discovered as an essential host component to restrict intracellular replication of several bacterial pathogens , including Legionella pneumophila , the causative agent of a severe pneumonia called Legionnaires’ Disease [6 , 7] . NAIP5 binds directly to L . pneumophila flagellin [8–10] , resulting in NLRC4 and CASP1/8 activation . Flagellin-deficient L . pneumophila evade detection by NAIP5/NLRC4 [11 , 12] and NAIP5-deficient or NLRC4-deficient cells or mice also fail to detect or restrict the intracellular replication of flagellated L . pneumophila [13–15] , but the underlying effectors downstream of NAIP5/NLRC4 required for resistance to L . pneumophila remain unclear . Caspase-1 cleaves dozens of host proteins [16–18] , but two key substrates suggested to participate in host defense are Gasdermin-D ( GSDMD ) ( reviewed in [19] ) and the pro-inflammatory cytokines interleukin-1β ( IL-1β ) and IL-18 . Cleaved Gasdermin-D oligomerizes and inserts into the plasma membrane to form large pores [20 , 21] , leading to release of IL-1β/-18 , as well as to a characteristic form of cell death called pyroptosis . In vitro , pyroptotic cell death , but not IL-1β/-18 , is believed to restrict the intracellular replication of L . pneumophila in macrophages , presumably by elimination of the intracellular niche required for bacterial replication . However , CASP1-deficient macrophages are only partially susceptible to L . pneumophila [4 , 22] and the CASP8 substrates that contribute to inflammasome-mediated host defense remain unclear . Caspase-11 ( CASP11 ) and Caspase-7 ( CASP7 ) are additional caspases previously implicated in resistance to L . pneumophila . CASP11 detects L . pneumophila lipopolysaccharide ( LPS ) and triggers GSDMD cleavage to activate pyroptosis independent of NAIP5/NLRC4 activation . Although CASP11 is activated by L . pneumophila [23–26] , CASP11 does not appear to play a major role in restricting bacterial replication in bone marrow macrophages , as Casp11–/–macrophages fully restrict L . pneumophila replication , and Nlrc4–/–macrophages , which still harbor functional CASP11 , are fully permissive [27] . The lack of a discernable role for CASP11 in restricting L . pneumophila is likely due to a requirement for ‘priming’ signals to induce CASP11 expression , as well as to redundancy with the NAIP5/NLRC4 inflammasome . CASP7 has also been reported to be activated downstream of flagellin detection and CASP1 activation by the NAIP5/NLRC4 inflammasome [28] . NAIP5/NLRC4-dependent CASP7 activation was reported to require CASP1 and , consistent with previous work [17] , CASP7 was suggested to be cleaved directly by CASP1 . In fact , Casp7–/–macrophages were reported to phenocopy the susceptibility of Casp1–/–macrophages , and CASP7 was thus proposed to be required for CASP1-dependent resistance to L . pneumophila [28] . Although GSDMD was not known at the time of this work , in retrospect it is surprising Casp7–/–cells would recapitulate the susceptibility of Casp1–/–cells given the clear role for GSDMD as a direct CASP1 substrate that is sufficient to mediate pyroptosis . Moreover , given that NAIP5/NLRC4 activates CASP8 [4 , 5 , 29] , and that CASP8 cleaves CASP7 [30] , it is surprising that CASP1 would be required for CASP7 activation . Here we sought to define the key effectors downstream of the NAIP5/NLRC4 inflammasome that are required to restrict bacterial replication . Consistent with prior studies , we find that both CASP1 and CASP8 are activated by the NAIP5/NLRC4 inflammasome . Importantly , we find that mice doubly deficient in both enzymes fully recapitulate the susceptibility of NAIP5/NLRC4-deficient mice . We further find that CASP7 activation downstream of NAIP5/NLRC4 is mediated by CASP8 in addition to CASP1 . Thus , mice singly deficient in CASP1 , CASP7 or GSDMD are not fully susceptible to L . pneumophila , whereas Casp7/1/11–/–and Gsdmd/Casp7–/–mice phenocopy the full susceptibility of Nlrc4–/–mice to L . pneumophila . Taken together our results identify CASP7 and GSDMD as the key mediators downstream of NAIP5/NLRC4 inflammasome activation . We have previously shown that CASP8 is activated in response to L . pneumophila infection when we silence GSDMD or in the absence of CASP1 [4] . To confirm these data , we infected macrophages deficient in GSDMD ( Gsdmd–/– ) with L . pneumophila and measured CASP8 activation using western blot and a substrate that detects CASP8 activity . We found that infection with wild type bacteria and fliI mutants ( that express cytosolic flagellin , but do not assembly the flagellum ) , but not with flaA mutants , triggers robust CASP8 activation in Casp1/11–/–and Gsdmd–/–macrophages ( Fig 1A and 1B ) . CASP8 did not appear to be robustly activated in C57BL/6 or Casp11–/–macrophages , possibly because these macrophages undergo rapid pyroptosis . In addition , CASP8 activation was not observed in Asc/Casp1/11–/–macrophages ( Fig 1A and 1B ) . These data support previous findings indicating that CASP8 is activated in inflammasomes , particularly when CASP1 or GSDMD is absent , in a process that requires ASC [4 , 5 , 31] . Next , we tested the role of CASP8 in restricting L . pneumophila replication in macrophages in the absence of CASP1/11 . We have previously shown that Nlrc4–/–macrophages are more permissive than Casp1/11–/–[22] , implying that CASP1 is not the sole caspase activated by NAIP5/NLRC4 . Thus , to discern a role for CASP8 , we compared macrophages deficient in Casp8/1/11/Ripk3 with those deficient in Casp1/11/Ripk3 . Experiments were conducted in a Ripk3 mutant background because Casp8-deficiency is embryonically lethal except in the absence of Ripk3 [32] . We found that whereas Casp1/11/Ripk3–/–macrophages are slightly more permissive to L . pneumophila replication than the C57BL/6 , the Casp8/1/11/Ripk3–/–cells are highly susceptible , similar to Nlrc4–/–macrophages . This was shown using both very low MOI ( MOI = 0 . 015 ) and high MOI ( MOI = 10 ) infections ( Fig 1C and 1D ) . We also assessed bacterial replication in macrophages using a L . pneumophila strain stably expressing the Photorhabdus luminescens luxCDABE ( lux ) operon as described previously [33] . We generated a JR32 strain of L . pneumophila expressing the lux operon and detected robust bacterial replication in Casp8/1/11/Ripk3–/–and Nlrc4–/–macrophages ( Fig 1E ) . In contrast , C57BL/6 and Casp1/11/Ripk3–/–macrophages were restrictive to bacterial replication ( Fig 1E ) . This was observed using wild type JR32 L . pneumophila ( Fig 1E ) and fliI mutants ( Fig 1F ) . As expected , isogenic flaA mutants expressing the lux operon robustly replicate in all macrophages used ( Fig 1G ) . We also tested the importance of CASP8 for bacterial growth restriction using the wild type Lp02 strain of L . pneumophila and obtained comparable results ( S1A Fig ) . As expected , flagellin mutants of Lp02 L . pneumophila effectively replicated in all macrophages used ( S1B Fig ) , whereas dotA mutants were defective for intracellular replication ( S1C Fig ) . CASP7 was previously proposed to require CASP1 for activation downstream of NAIP5/NLRC4 [28] . However , since CASP7 is also known to be a substrate of CASP8 [30 , 34] , we decided to re-assess CASP7 activation in CASP1-deficient cells where CASP8 is still active . We infected Casp1/11–/–macrophages with wild type L . pneumophila ( or with flaA mutants as control ) and assessed CASP7 processing by western blot using an antibody specific for CASP7 p18 . We found that CASP7 is processed in response to flagellin in Casp1/11–/–macrophages ( Fig 2A ) . In Casp1/11–/–cells , CASP7 processing in response to flagellin requires CASP8 because we detect no CASP7 p18 in Casp8/1/11/Ripk3–/–macrophages ( Fig 2A ) . In this experiment , Casp7/1/11–/–macrophages were used to confirm the specificity of the anti-CASP7 p18 antibody ( Fig 2A ) . Next , we used the same macrophages to test if CASP7 is required for CASP8 activation . By assessing CASP8 activation by western blot and by the chemiluminescent substrate , we found that CASP8 is activated in Casp1/11–/–and in Casp7/1/11–/–macrophages , but not in Asc/Casp1/11–/–or Casp8/1/11/Ripk3–/–macrophages ( Fig 2B and 2C ) . These data indicate that , as expected , CASP7 is not required for CASP8 activation in the absence of CASP1/11 . Experiments performed with immortalized macrophages confirmed that ASC but not CASP7 is important for CASP8 activation in the absence of CASP1/11 ( S2 Fig ) . We and others have previously shown that L . pneumophila infection triggers pore formation and pyroptosis that is dependent on CASP1 and CASP11 [11 , 12 , 25 , 35] . We also showed that in the absence of CASP1 and CASP11 , L . pneumophila induces pore formation dependent on flagellin , ASC and CASP8 [4] . Thus , we tested if CASP7 accounts for pore formation downstream of CASP8 . We measured pore formation in real time by assessing the uptake of propidium iodide into the nuclei of permeabilized macrophages as described [25] . We confirmed that pore formation occurs in Casp1/11–/–macrophages in response to L . pneumophila infection ( Fig 3A and 3B ) . The Caspase-1/11-independent pore formation is abolished in Asc/Casp1/11–/–and Casp8/1/11/Ripk3–/–macrophages and is reduced in Casp7/1/11–/–macrophages ( Fig 3B ) . As previously demonstrated , Fig 3B shows that pore formation in wild type C57BL/6 cells is very robust because it occurs through CASP1 and CASP11 [25] . We also measured LDH release after infection as a readout for membrane leakage . As suggested by the pore formation assay , we found that LDH release occurs in C57BL/6 and Casp1/11–/–but not in Asc/Casp1/11–/–or Casp8/1/11/Ripk3–/–macrophages ( Fig 3C ) . Although reduced , we still detected cell death in Casp7/1/11–/–cells in response to infection ( Fig 3C ) . We measured CASP3 activation by western blot and confirmed that CASP3 is activated in response to flagellin in Casp7/1/11–/–cells , but not in C57BL/6 , Casp1/11–/–and Nlrc4–/–macrophages ( Fig 3D ) . These data suggest that CASP3 does not play a significant role downstream of NAIP5/NLRC4 inflammasome as previously reported [13–15] , but can account to explain the modest levels of cell death detected in Casp7/1/11–/–cells . Collectively , these data suggest that in the absence of CASP1/11 , macrophages are still able to respond to flagellin and trigger pore formation and pyroptosis via ASC and CASP8 and partially via CASP7 . Next , we tested the importance of CASP7 for restriction of L . pneumophila replication in macrophages . To address this we used macrophages from Casp7–/–single mutants and also from mice triple deficient for CASP7/1/11 ( Casp7/1/11–/– ) . In contrast to a previous report [28] , we found that Casp7–/–macrophages efficiently restrict the replication of wild type L . pneumophila as measured by luciferase and CFU ( Fig 4A and 4B ) . In contrast , Casp7/1/11–/–macrophages are highly permissive for bacterial replication and phenocopy cells deficient for Nlrc4–/–and Casp8/1/11/Ripk3–/– ( Fig 4A and 4B ) . These data support a role of CASP8 and CASP7 for restriction of bacterial replication in the absence of CASP1/11 . As expected , flaA mutants evaded NAIP5/NLRC4 and effectively replicated in all macrophages regardless of the genotype ( Fig 4C and 4D ) . We performed additional experiments comparing the replication of wild type L . pneumophila and isogenic flaA mutants in each macrophage genotype and found that flaA replicated significantly better than flagellin-positive bacteria in C57BL/6 , Casp1/11–/– , Casp7–/–macrophages ( S3A–S3C Fig ) . The flaA mutants phenocopy the wild type bacteria in Nlrc4–/–and Casp8/1/11/Ripk3–/– ( S3D and S3E Fig ) and flaA replicated slightly better than wild type bacteria in Casp7/1/11–/–macrophages ( S3F Fig ) . These data indicate that NLRC4 , CASP1 , CASP8 and CASP7 are important for flagellin-mediated restriction of bacterial replication . We have previously shown that Nlrc4–/–mice are more susceptible than Casp1/11–/–mice in vivo [22 , 36] . Thus , we tested if CASP7 accounts for the CASP1/11-independent mechanisms of restriction of bacterial replication . To test this , we infected mice with L . pneumophila and measured CFU in the lungs after 48 and 96hs . Strikingly , we found that Casp7/1/11–/–mice are as susceptible as the Nlrc4–/–mice and significantly more susceptible than Casp1/11–/–mice ( Fig 5A ) . As expected the C57BL/6 and heterozygote control Casp7+/–/1+/–/11+/–are highly restrictive ( Fig 5A ) . Next , we crossed the Casp7/1/11–/–x Casp1/11–/–to generate Casp7+/–/1–/–/11–/–progeny and infected these mice together with Casp7/1/11–/–mice . We confirmed that Casp7/1/11–/–mice are more susceptible than the Casp7+/–/1–/–/11–/–mice , suggesting that CASP7 accounts for restriction of L . pneumophila replication in the absence of CASP1/11 ( Fig 5B ) . Our data show that CASP7 operates downstream of CASP8 for restriction of L . pneumophila replication in Casp1/11–/–macrophages . Thus , we investigated if CASP7 is also activated in wild type macrophages . To test this we measured CASP7 cleavage in response to infection of C57BL/6 and Casp1/11–/–macrophages . We detected CASP7 processing both in C57BL/6 and Casp1/11–/–macrophages infected with wild type L . pneumophila ( Fig 6A ) . In contrast , we only detected processing and activation of CASP8 in the Casp1/11–/–macrophages ( Fig 6B and 6C ) . These data suggest that CASP7 can be activated in conditions where CASP8 is inactive , possibly by CASP1 as previously reported [17 , 28] . To test this hypothesis we performed experiments in macrophages that are deficient in CASP8 and sufficient in CASP1 and confirmed that CASP7 is cleaved in Casp8/Ripk3–/–cells ( Fig 6D ) . In this experiment we detected a weak CASP7 cleavage in response to flaA and in Nlrc4–/–cells , suggesting the participation of another pathway for CASP7 activation . We next performed growth curves in macrophages that are CASP1 positive and CASP8 negative to test if CASP1 activation is sufficient to trigger restriction of L . pneumophila replication . We found that Casp8/Ripk3–/–macrophages are as restrictive as C57BL/6 ( and Ripk3–/– ) macrophages ( Fig 6E ) . As expected , Casp1/Ripk3–/–macrophages exhibit intermediate susceptibility and Casp8/1/Ripk3–/– , Casp8/1/11/Ripk3–/–macrophages are as susceptible as Nlrc4–/–macrophages ( Fig 6E ) . The flaA mutants replicated in all macrophages tested , as expected ( Fig 6F ) . These data confirm that CASP1 activation is sufficient to restrict L . pneumophila replication via the NAIP5/NLRC4 inflammasome despite the deficiency in CASP8 . In addition to CASP7 , GSDMD is also downstream of CASP1 . We tested by western blot if GSDMD is activated in response to L . pneumophila and found that GSDMD is cleaved in response to wild type L . pneumophila but not flaA mutants after 6 hs of infection ( S4 Fig ) . Gsdmd–/–macrophages were included to control antibody specificity . GSDMD activation was found in C57BL/6 and Asc–/–macrophages but not in Casp1/11–/–and Nlrc4–/– ( S4 Fig ) , indicating that GSDMD is cleaved by CASP1 independent of CASP8 via the NAIP5/NLRC4 inflammasome . We further generated mice double deficient in GSDMD and CASP8 to investigate if the CASP8-independent , CASP1-mediated restriction of L . pneumophila replication requires GSDMD . We found that Gsdmd/Casp8/Ripk3–/–cells are as restrictive as the Gsdmd/Ripk3–/–and C57BL/6 macrophages ( Fig 6G and 6H ) . These data indicate that CASP1 activation triggers restriction of L . pneumophila replication in the absence of CASP8 and GSDMD , consistent with the participation of CASP7 downstream of CASP1 . Therefore , we reasoned that in wild type macrophages , CASP1 activates both GSDMD and CASP7 when the NAIP5/NLRC4 inflammasome is activated and that GSDMD and CASP7 can both mediate restriction of L . pneumophila independently of each other . The involvement of both CASP7 and GSDMD would explain why Casp7–/–or Gsdmd–/–singly deficient cells are able to restrict L . pneumophila replication in macrophages ( Figs 4A , 4B and 6G ) . To test this hypothesis , we generated Casp7/Gsdmd–/–double deficient mice and tested the requirement of these molecules downstream of NAIP5/NLRC4 inflammasome . Initially , we tested pore formation in response to wild type L . pneumophila and flaA mutants and found that Casp7/Gsdmd–/– , Casp7/1/11–/–and Nlrc4–/–macrophages induced low pore formation in comparison to Casp7–/– , Gsdmd–/–and C57BL/6 macrophages ( Fig 7A–7C ) . Of note , the pore formation assay performed with live L . pneumophila in presence of CASP11 may be difficult to interpret because LPS effectively triggers CASP11-mediated pore formation [25 , 27] . Therefore , we tested pore formation and LDH release in response to flagellin using FlaTox , a reagent that selectively activates NAIP5/NLRC4 without the confounding activation of CASP11 [37] . We found that cytosolic flagellin triggers a response that is CASP1-dependent and requires both CASP7 and GSDMD . This can be observed by pore formation ( Fig 7D ) and LDH release ( Fig 7E ) . To evaluate the effect of CASP7 and GSDMD for restriction of L . pneumophila replication we performed macrophage infections and found that Casp7/Gsdmd–/–cells are significantly more permissive than the single KOs Casp7–/–and Gsdmd–/– ( Fig 7F and 7G ) . Casp7/Gsdmd–/–macrophages phenocopied Casp7/1/11–/–and Nlrc4–/–for restriction of L . pneumophila replication ( Fig 7F ) . We tested bacterial infection in vivo and found that the Casp7/Gsdmd–/–mice are as susceptible as the Nlrc4–/–mice ( Fig 7H ) . In contrast , the Casp7–/–and Gsdmd–/–mice were significantly more restrictive , while still slightly more susceptible than C57BL/6 , suggesting that these molecules are important for host resistance in vivo . Together , these data support a model that both CASP7 and GSDMD are required for restriction of L . pneumophila replication downstream of CASP1 . When CASP1 or GSDMD is missing , CASP8 is activated in the NLRC4 inflammasome and triggers activation of CASP7 , which also accounts for restriction of L . pneumophila replication in macrophages and in vivo ( Fig 8 ) . CASP1 activation downstream of the NAIP5/NLRC4 inflammasome is critical for restriction of L . pneumophila replication in macrophages and in the lungs of infected mice [25 , 27] , but the CASP1 substrates required for restriction of bacterial replication are still obscure . Previous work showed CASP7 is activated downstream of CASP1 [17 , 28] and is required for restriction of L . pneumophila replication via the NAIP5/NLRC4 inflammasome [28] . Although these data provide a direct link between CASP1 and CASP7 , our experiments performed with macrophages and mice deficient only in CASP7 do not suggest an essential role of CASP7 for restriction of L . pneumophila replication in vivo . Indeed , we detected only a modest ( less than one log ) increase in bacterial loads in Casp7–/–compared to C57BL/6 macrophages during L . pneumophila growth curves in macrophages ( Figs 4A , 4B , 7F and 7G ) . In this study we sought to identify the redundant pathways that operate downstream of CASP1 for host resistance to infection . Our data support the hypothesis that redundancy between CASP7 and GSDMD explains the minor phenotypes of singly deficient Gsdmd–/–and Casp7–/–cells for restriction of L . pneumophila infection . This hypothesis is strongly supported by our data indicating that Gsdmd/Casp7–/–double-deficient mice are highly susceptible to L . pneumophila and phenocopy the Nlrc4–/–mice ( Fig 7 ) . We and others have previously shown that CASP8 is also activated by the NAIP5/NLRC4 inflammasome , particularly when CASP1 or GSDMD is inhibited or missing [4 , 5 , 31] . As with CASP1 , the CASP8 substrates required for bacterial restriction have been unclear . Interestingly , our data support a model indicating that CASP7 also operates downstream of CASP8 when the NAIP5/NLRC4 inflammasome is activated by cytosolic flagellin . We speculate that the kinetics of CASP8 activation in this inflammasome is slow compared with the quick and robust induction of pyroptosis that occurs via CASP1 and GSDMD . Therefore , CASP8 activation is preferentially detected in the absence of CASP1 and GSDMD . In these conditions , CASP7 is also activated downstream of CASP8 and our data using the Casp7/1/11–/– ( and the Casp8/1/11–/– ) mice and their macrophages strongly support the role of CASP7 downstream of CASP8 in the NAIP5/NLRC4 inflammasome . Together , our data indicate that CASP1 and CASP8 are the primary caspases activated by the NAIP5/NLRC4 inflammasome . Downstream of CASP1/8 we have now identified CASP7 and GSDMD as the key substrates required for restriction of bacterial replication ( Fig 8 ) . Our data therefore suggest that there is considerable redundancy built into the signaling outputs of the NAIP5/NLRC4 inflammasome: NAIP5/NLRC4 activates both CASP1 and CASP8; CASP1 can activate both GSDMD and CASP7; and CASP8 can activate CASP7 [38] . We speculate that this redundancy may be a mechanism for hosts to ensure responses even to pathogens that inhibit specific arms of the response . The identification of the critical substrates involved in the restriction of bacterial infection via the NAIP/NLRC4 inflammasomes provides important information for our understanding of the biology of these important platforms that operate for host protection against pathogenic bacteria . In response to Yersinia spp . infection , CASP8 was reported to trigger GSDMD activation [38 , 39] . Interestingly , it was recently reported that the GSDMD-mediated proinflammatory function of CASP8 is counteracted by CASP3-dependent cleavage and inactivation of GSDMD [40] , suggesting that CASP3 suppresses GSDMD-mediated cell lysis during CASP8-induced apoptosis . We speculate that this may explain why we did not detect a CASP8-mediated GSDMD cleavage in Casp1/11–/–macrophages ( S4 Fig ) and a robust pore formation and LDH release in Casp7/1/11–/– ( Figs 3 and 7B ) . A question that arises from our studies is how CASP7 and GSDMD operate to restrict L . pneumophila replication . GSDMD is known to trigger pore formation and pyroptosis [20 , 21] , a process that results in host cell death and thereby likely eliminates the intracellular replicative niche . In addition , pyroptosis has been proposed to result in formation of pore-induced intracellular traps ( PITs ) in macrophages infected with intracellular bacteria . The formation of PITs may help sequester bacteria and lead to their clearance by efferocytosis [41] . Our data demonstrate that CASP7 also promotes pore formation and host cell death . Although CASP7 appears to be involved in induction of cell death , the CASP7 substrates operating to trigger pore formation are still unclear . However , our data showing that CASP7 deficiency exerts a significant effect on bacterial replication even on a Gsdmd–/–background implies that CASP7 substrates other than GSDMD contribute to host defense [28 , 42] . It will be of interest to identify these substrates and their mechanisms of action in future studies . Mice used in this study were breed and maintained in institutional animal facilities at FMRP/USP or at UC Berkeley . Mice used were C57BL/6 ( Jax 000664 ) , Casp7–/– ( Jax 006237 ) , Nlrc4–/–[43] , Casp1/11–/–[44] , Asc–/–[45] , Casp11–/–[46] , Asc/Casp1/11–/–[4] , Casp8–/–[5] , Gsdmd–/–[5] , Casp1–/–[5] , Ripk3–/–mice were originally from Xiaodong Wang [47] and backcrossed to C57BL/6 by Astar Winoto . Mice deficient in more than one gene not described above were generated in this study by intercrossing a F1 progeny of the parental strains . The care of the mice was in compliance with the institutional guidelines on ethics in animal experiments; approved by CETEA ( Comissão de Ética em Experimentação Animal da Faculdade de Medicina de Ribeirão Preto , approved protocol number 218/2014 ) . CETEA follow the Brazilian national guidelines recommended by CONCEA ( Conselho Nacional de Controle em Experimentação Animal ) . Animal experiments at UC Berkeley were approved by the institutional animal care and use committee . Mice were euthanized by CO2 asphyxiation with cervical dislocation as a secondary method . Bone marrow-derived macrophages were obtained as previously described [48] . Briefly , mice were euthanized and bone marrow cells were obtained from femurs and tibias . The cells were cultivated in RPMI 1640 ( Gibco , Thermo Fisher Scientific , Massachussetts , USA ) supplemented with 10–20% Fetal Bovine Serum ( FBS ) ( Gibco ) and 30% L929-Cell Conditioned Medium ( LCCM ) and 2 mM L-glutamine ( Sigma-Aldrich ) for 7 days , at 37°C , 5% CO2 . In some experiments , instead of 20% LCCM it was used 10% of a conditional medium from 3T3 cells stably expressing mouse MCSF as a source of macrophage colony stimulation factor . Cells were detached with cold PBS , resuspended in RPMI 1640 supplemented with 10% FBS ( R10 ) and plated as indicated . For all in vitro experiments , the plates were centrifuged at 300 x g for 5 min , room temperature , after cell plating and infection , to ensure homogeneous adherence of cells and infection synchronization , respectively . Incubation of non-infected and infected cells was done at 37°C , 5% CO2 . Legionella pneumophila strains used were JR32 and isogenic mutants for flaA and fliI as previously described [12 , 22] . For some experiments , L . pneumophila strains stably expressing the Photorhabdus luminescens luxCDABE operon were used . Bacteria were cultured in Charcoal-Yeast Extract Agar ( CYE , 10 g/L 4-morpholinepropanesulfonic acid [MOPS] , 10 g/L Yeast extract , 15 g/L technical agar , 2 g/L activated charcoal , supplemented with 0 . 4 g/L L-cysteine and 0 . 135 g/L Fe ( NO3 ) 3 ) at 35–37°C , for 4 days from frozen stocks . Single colonies were streaked on fresh plates and allowed to grow for another 2 days . For in vitro infections , bacteria grown on solid plates were resuspended in autoclaved distilled water and diluted on RPMI as indicated . For in vivo infections , bacteria grown on solid plates were resuspended in autoclaved distilled water and diluted in Phosphate-buffered saline ( PBS ) as needed . Experiments to quantify bacterial CFU in macrophages were made in 24-well plates . A total of 2 x 105 macrophages were plated per well in R10 and incubated overnight . The medium was replaced with the bacterial suspension in R10 with the indicated multiplicities of infection ( MOIs ) for 1hr before being replaced again by fresh R10 media . At the indicated time points , the supernatants were collected , cells were lysed with autoclaved distilled water and the lysate was added to the supernatants . Dilutions were plated on CYE and incubated for 4 days for counting of colony-forming units ( CFU ) . Experiments to measure bacterial replication using a luminescence-based replication assays were made as previously described [33] . Briefly , 105 macrophages/well were plated on white 96-well plates and incubated overnight . The medium was replaced with the bacterial suspension in R10 with an MOI of 0 . 01 or 10 . At the indicated time points , luminescence emission was measured at 470 nm with a Spectra-L plate reader ( Molecular Devices , California , USA ) . 5 x 105 macrophages/well were plated on 24-well plates in R10 and incubated overnight . The medium was replaced with the bacterial suspension ( estimated to reach an MOI of 10 ) in RPMI without Phenol Red ( 3 . 5 g/L HEPES , 2 g/L NaHCO3 , 10 . 4 g/L RPMI without Phenol Red , 1% glutamine , pH 7 . 2 ) and incubated for 7h . The supernatant was collected and LDH release was measured using CytoTox 96® Non-Radioactive Cytotoxicity Assay ( Promega , Winsconsin , USA ) following the manufacturer’s instructions . For estimation of pore formation , 105 macrophages/well were plated on black , clear bottom 96-well plates in R10 and incubated overnight . The medium was replaced with the bacterial suspension ( estimated to reach an MOI of 10 ) in RPMI without Phenol Red and low on NaHCO3 ( 2 g/L HEPES , 0 . 38 g/L NaHCO3 10 . 4 g/L RPMI without Phenol Red , 1% glutamine , 2% FBS , pH 7 . 2 ) , 1:1000 rabbit anti-Legionella antibody and 6 μL/mL propidium iodide ( PI ) . PI was excited at 538 nm and emission was measured at 617 nm with a SpectraMax plate reader ( Molecular Devices , California , USA ) . Triton-X100 1 . 3% was used as a positive control and for normalization . A total of 105 cells/well were plated on white 96-well plates in R10 and incubated overnight . The medium was replaced with the bacterial suspension in RPMI without Phenol Red and low on NaHCO3 ( 2 g/L HEPES , 0 . 38 g/L NaHCO3 10 . 4 g/L RPMI without Phenol Red , 1% glutamine , 2% FBS , pH 7 . 2 ) with an MOI of 10 and the plate was incubated for 8h . The supernatants were collected and Caspase-8 activation was measured using Caspase-Glo 8 Assay ( Promega , Winsconsin , USA ) following the manufacturer’s instructions . To measure caspase and GSDMD cleavage by western blot , 106 cells/well were plated on 24-well plates ( for CASP7 and CASP3 ) or 48-well plates ( for CASP8 and GSDMD ) in R10 and incubated overnight . The medium was replaced with the bacterial suspension in R10 with an MOI of 10 and the plate was incubated for the indicated times . The supernatants were discarded ( for CASP7 and CASP3 ) or collected ( for CASP8 and GSDMD ) and cells were lysed with 50 μL of RIPA supplemented with protease inhibitor ( Complete Protease Inhibitor Cocktail , Roche , Basel , Switzerland ) . For CASP8 and GSDMD , lysates were added to the supernatants . Samples were immediately sonicated for 10 min and frozen at -80°C until analysed . A total of 50 μg of protein from each sample were run on a 15% acrylamide gel , transferred onto a nitrocellulose membrane and the membranes were incubated overnight , at 4°C under mild agitation with Anti-cleaved CASP7 antibody ( rabbit ) ( Cell Signaling Technologies , Massachussetts , USA ) diluted 1:1000 in 5% BSA in TBS 1X with 0 . 01% Tween; or Caspase-8 ( D35G2 ) Rabbit mAb ( Cell Signaling Technologies , Massachussetts , USA ) diluted 1:1000 in 5% non-fat dry milk in TBS 1X with 0 . 01% Tween; Caspase-3 antibody ( #9662 , Cell Signaling Technologies , Massachussetts , USA ) diluted 1:1000 in 5% non-fat dry milk in TBS 1X with 0 . 01% Tween; or a rat monoclonal antibody against GSDMD ( GN20-13 , Genentech ) diluted 1:1000 in 5% non-fat dry milk in TBS 1X with 0 . 01% Tween . Actin was stained with rabbit anti-α-actin ( #A2066 , Sigma-Aldrich , Missouri , USA ) diluted 1:5000 in 5% non-fat dry milk in TBS 1X with 0 . 01% Tween . The membranes were incubated for 1h with goat anti-rabbit or anti-rat secondary antibodies ( Sigma-Aldrich , Missouri , USA ) and analyzed using ECL™ Prime Western Blotting System ( GE Healthcare , Illinois , EUA ) and an Amersham Imager 600 ( GE Healthcare , Illinois , EUA ) . Bands were quantified using ImageJ . All mice were matched by sex and age ( all were at least 8 weeks old at the time of infection ) and were in a C57BL/6 mouse genetic background . For the in vivo experiments , approximately 5–7 mice per group were used , as indicated in the figures . Mice were infected intranasally with 105 bacteria contained in 40 μL of PBS . The animals were anesthetized with ketamine ( 50mg/kg ) and xylazine ( 10mg/kg ) intraperitoneally and infected . At the indicated time points , the lungs were harvested and macerated for 30 seconds in 5 mL of autoclaved distilled water using a tissue homogenizer ( Power Gen 125; Thermo Scientific ) . Dilutions were plated on CYE + 10 μg/mL of streptomycin and plates were incubated for 4 days at 37°C for CFU counting . The data were plotted and analyzed using GraphPad Prism 5 . 0 software . The statistical significance was calculated using the Student’s t-test or analysis of variance ( ANOVA ) . Differences were considered statistically significant when P was <0 . 05 , as indicated by an asterisk in the figures .	Inflammasomes are multi-protein complexes that detect infection and other stimuli and activate the Caspase-1 ( CASP1 ) protease . The effectors required downstream of NAIP5/NLRC4 to restrict bacterial replication remain unclear . Active CASP1 cleaves and activates the pore-forming protein gasdermin D ( GSDMD ) to induce inflammation and cell death . We have previously shown that CASP8 is activated by the NAIP5/NLRC4 inflammasome independently of CASP1 and functions to restrict replication of the intracellular bacterium Legionella pneumophila . Here , we show that CASP7 is activated downstream of CASP8 and is required for CASP8-dependent restriction of L . pneumophila replication in macrophages and in vivo . In addition , CASP7 is also activated by CASP1 . Taken together , these results imply that CASP7 and GSDMD are the two key caspase substrates downstream of NAIP5/NLRC4 . In support of this hypothesis , we found that mice double deficient in CASP7 and GSDMD are more susceptible than the single knockouts and are as susceptible as the Nlrc4 deficient mice for restriction of L . pneumophila replication in vivo . Collectively , our data indicate that GSDMD and CASP7 are activated by CASP1 and induce cell death and restriction of bacterial infection . Therefore , GSDMD and multiple caspases ( CASP1 , CASP7 and CASP8 ) operate downstream of the NAIP5/NLRC4 inflammasome for restriction of infection by pathogenic bacteria .	[ "Abstract", "Introduction", "Results", "Discussion", "Material", "and", "methods" ]	[ "blood", "cells", "bacteriology", "cell", "death", "medicine", "and", "health", "sciences", "immune", "cells", "pathology", "and", "laboratory", "medicine", "intracellular", "pathogens", "pathogens", "immunology", "cell", "processes", "microbiology", "electromagnetic", "radiation", "bacterial", "diseases", "inflammasomes", "bacteria", "bacterial", "pathogens", "immune", "system", "proteins", "infectious", "diseases", "legionella", "pneumophila", "white", "blood", "cells", "microbial", "physiology", "animal", "cells", "legionella", "medical", "microbiology", "proteins", "microbial", "pathogens", "luminescence", "physics", "biochemistry", "bacterial", "physiology", "cell", "biology", "flagellin", "biology", "and", "life", "sciences", "cellular", "types", "physical", "sciences", "macrophages", "organisms" ]	2019	Gasdermin-D and Caspase-7 are the key Caspase-1/8 substrates downstream of the NAIP5/NLRC4 inflammasome required for restriction of Legionella pneumophila
Epidemiological studies have shown that one of the strongest risk factors for prostate cancer is a family history of the disease , suggesting that inherited factors play a major role in prostate cancer susceptibility . Germline mutations in BRCA2 predispose to breast and ovarian cancer with its predominant tumour suppressor function thought to be the repair of DNA double-strand breaks . BRCA2 has also been implicated in prostate cancer etiology , but it is unclear the impact that mutations in this gene have on prostate tumourigenesis . Here we have undertaken a genetic analysis in the mouse to determine the role of Brca2 in the adult prostate . We show that deletion of Brca2 specifically in prostate epithelia results in focal hyperplasia and low-grade prostate intraepithelial neoplasia ( PIN ) in animals over 12 months of age . Simultaneous deletion of Brca2 and the tumour suppressor Trp53 in prostate epithelia gave rise to focal hyperplasia and atypical cells at 6 months , leading to high-grade PIN in animals from 12 months . Epithelial cells in these lesions show an increase in DNA damage and have higher levels of proliferation , but also elevated apoptosis . Castration of Brca2;Trp53 mutant animals led to regression of PIN lesions , but atypical cells persisted that continued to proliferate and express nuclear androgen receptor . This study provides evidence that Brca2 can act as a tumour suppressor in the prostate , and the model we describe should prove useful in the development of new therapeutic approaches . Prostate cancer is the most common cancer in men in developed countries , with a rising incidence of the disease . However , the etiology of this malignancy is still unclear . Prostate cancer progresses through a pathologically defined series of steps involving increasing grades of PIN , invasive adenocarcinoma and metastatic cancer [1] . Androgens are crucial for normal prostate function , and act as pro-survival and proliferation factors in cancer cells . As such , prostate cancer is sensitive to androgen levels and androgen depletion therapy via chemical or surgical castration is an initial step in treatment , typically resulting in tumour regression . However , the cancer normally re-grows and develops as a castration-independent tumour . Epidemiological studies have shown that one of the strongest risk factors for prostate cancer is a family history of the disease , suggesting that inherited factors play a major role in prostate cancer susceptibility [2] , [3] . Approximately 10% of prostate cancers are thought to be hereditary , and this number increases with early on-set disease . In spite of this , little is known about the mechanisms of tumourigenesis of inherited prostate cancer . Prostate cancer frequently clusters in families that have breast cancer , indicating a genetic link between these two diseases [4]–[6] . Germline mutations in BRCA2 predispose to both breast and ovarian cancer making it a good candidate gene for prostate cancer etiology . There is an increased risk of prostate cancer in individuals carrying a mutation in BRCA2 , particularly early-onset disease [7]–[10] . The Breast Cancer Linkage Consortium found a significant relative risk of 4 . 65 for prostate cancer in male carriers of a deleterious BRCA2 mutation that rose to 7 . 33 in men under 65 years of age [7] . Consistent with this , analysis of men with early-onset disease indicates that BRCA2 carriers account for between 0 . 8–2% of prostate cancer cases , compared with the prevalence of 0 . 1% BRCA2 mutations in the general population [11] , [12] . In addition , BRCA2 mutation carriers have been associated with aggressive prostate cancer [13]–[16] . BRCA2 is thought to act as a tumour suppressor , with tumours arising from BRCA2 mutations frequently demonstrating loss-of-heterozygosity with loss of the wild-type allele . BRCA2 plays an important role in the repair of DNA double-strand breaks ( DSB ) through homologous recombination ( HR ) [17] . When there is a second identical DNA copy ( i . e . the sister chromatid after replication ) HR is the primary method of repair and is a relatively error-free DNA repair pathway . After DNA damage , BRCA2 directly interacts with the recombinase RAD51 , a process that is essential for HR-mediated repair of DSBs [18] . When HR is defective or no sister chromatid is available the error-prone methods of single-strand annealing and non-homologous end joining are used for DNA repair [19] . BRCA2-deficient cells form chromosomal aberrations spontaneously in culture and are more sensitive to certain DNA damaging agents [19]–[21] . Hence , loss of BRCA2 is thought to principally lead to tumour progression by the failure to repair DNA by HR , leading to genomic instability . Mouse models have shown a direct in vivo tumour suppressor role for Brca2 in the mammary gland and have demonstrated a synergistic tumour suppressor activity with Trp53 . However , Brca2 heterozygous animals do not show a predisposition to tumour formation and Brca2 null mice result in embryonic lethality [18] , [22] , [23] . To circumvent this prenatal lethality , the Cre-LoxP system has been used to conditionally delete Brca2 in a tissue-specific manner . Deletion of Brca2 from the mouse mammary epithelium either fails to produce mammary-gland tumours or results in mammary-gland tumour formation with long latency ( 1 . 4–1 . 6 years ) [24]–[26] . Tumour latency was reduced in Brca2 mutant mice that were Trp53 heterozygous [24] . In addition , mice with conditional inactivation of Brca2 and Trp53 developed mammary tumours with high penetrance at 6 months [25] . To understand the role of Brca2 in prostate cancer we have used a prostate–specific Cre line and a conditional Brca2 allele to delete Brca2 in adult mouse prostate epithelia . We show that loss of Brca2 in the prostate results in focal hyperplasia and low-grade ( LG ) PIN . Mice with conditional deletion of Brca2 and Trp53 have a high incidence of high-grade ( HG ) PIN , which contain cells with elevated DNA damage . PIN lesions in Brca2;Trp53 homozygous mutant prostates persist and continue to proliferate after androgen depletion . This work confirms the role of Brca2 as a tumour suppressor in the prostate and provides a model to test potential therapeutics in Brca2-deficient prostate neoplasia . To investigate the role of Brca2 in the prostate we deleted Brca2 from the adult mouse prostate epithelia . To achieve this we mated mice carrying a Brca2 allele that has exon 11 flanked by loxP sites ( Brca2F/F ) to transgenic mice carrying Cre recombinase under the control of a prostate-specific composite rat probasin promoter , PBCre4 [25] , [27] . This Cre line has been used successfully to delete tumour suppressor genes and activate oncogenes to drive prostate neoplasia and tumour progression [28]–[30] . Deletion of this Brca2 conditional allele results in the loss of a Rad51-interacting domain , and consequently , homozygous germline deletion leads to embryonic lethality [25] . Cohorts of male control ( Brca2F/F ) , Brca2 heterozygous ( Brca2F/+;PBCre4 ) and Brca2 mutant ( Brca2F/F;PBCre4 ) animals were generated and analysed for tumour progression at 6 months , 10–14 months and 15–20 months of age . None of the Brca2F/+;PBCre4 prostates had any observable morphological differences compared to control prostates at any time point analysed ( Figure 1 and Table 1 ) . Focal hyperplasia that contained atypical cells was first observed in Brca2F/F;PBCre4 prostates at 10–14 months and was also present in these animals at 15–20 months ( Table 1 ) . In addition , at 15–20 months a significant number of Brca2 homozygous mutant prostates had focal LG PIN in their lumen compared to control animals ( 6/17 compared to 0/28; Z-test p = 0 . 0001 ) ( Figure 1A and Table 1 ) . LG PIN lesions formed characteristic tufting and cribiform patterns ( Figure 1B ) . These focal lesions contained multiple atypical cells that had prominent nucleoli and hyperchromasia . Hyperplasia and LG PIN lesions were present in all four prostatic lobes of Brca2 mutants . The tumour suppressor TP53 is frequently mutated in BRCA2 cancers and studies in the mouse have shown a genetic interaction between Brca2 and Trp53 [24] , [31] , [32] . To test if Brca2 and Trp53 cooperate in the prostate we deleted both of these genes in the prostate epithelia using the PBCre4 transgene . Cohorts of male control ( Brca2F/F;Trp53F/F ) , Brca2 homozygous and Trp53 heterozygous ( Brca2F/F;Trp53F/+;PBCre4 ) and Brca2 and Trp53 double homozygous ( Brca2F/F;Trp53F/F;PBCre4 ) animals were generated and analysed for tumour progression . In addition to hyperplasia and LG PIN observed in Brca2 mutants , deletion of Brca2 and Trp53 resulted in the formation of HG PIN lesions . At 10–14 months , Brca2F/F;Trp53F/+;PBCre4 animals had focal LG PIN and hyperplasia ( Figure 2A and Table 1 ) . By 15–20 months , LG PIN was still present and a significant number of animals had focal HG PIN compared to control animals ( 6/15 compared to 0/28; Z-test p = 0 . 0017 ) . However , the frequency of HG PIN was significantly higher in Brca2F/F;Trp53F/F;PBCre4 animals compared to Brca2F/F;Trp53F/+;PBCre4 animals at this age ( 27/32 compared to 6/15; Z-test p = 0 . 0058 ) ( Figure 2A and Table 1 ) . In these animals , focal areas of hyperplasia consisting of atypical cells were present as early as 6 months . At 10–14 months , LG PIN was present and a significant number of Brca2F/F;Trp53F/F;PBCre4 animals had HG PIN compared to control animals ( 7/15 compared to 0/11; Z-test p = 0 . 0276 ) . Hyperplasia and LG PIN lesions were similar to those found in Brca2F/F;PBCre4 mutants . Frequently multiple ducts of each lobe had HG PIN , which were present in proximal and distal regions of the prostate and consisted of many atypical cells filling the lumen . Atypical cells were unorganised with poor orientation , severe nuclear pleomorphism and abnormal nuclear to cytoplasm ratios ( Figure 2B ) . Mitotic figures , apoptotic bodies and areas of necrosis were also present within HG PIN lesions ( Figure 2B ) . In some cases , epithelial cells of the lumen protrude into the adjacent stroma and the smooth muscle surrounding the ducts was no longer continuous but was broken up ( Figure 2B ) . These areas contained atypical smooth muscle cells and desmoplasia in the surrounding stroma . HG PIN lesions were predominantly seen in the anterior prostate ( AP ) and dorsal prostate ( DP ) of Brca2;Trp53 homozygous mutant animals , with a small number observed in lateral and ventral lobes . Deletion of LoxP flanked Brca2 and Trp53 alleles by the PBCre4 transgene in the prostate was confirmed by PCR analysis on micro dissected tissue ( Figure S1 ) . As the predominant tumour suppressor function of BRCA2 is thought to be the repair of DNA DSBs , we assessed the level of spontaneous DNA damage in Brca2 and Brca2;Trp53 mutant prostates . An early response to DNA damage is the phosphorylation of histone H2AX ( γH2AX ) [33] . Areas of hyperplasia and PIN in Brca2 mutant and Brca2;Trp53 mutant prostates contained cells that were positive for γH2AX , which were not present in control prostates ( Figure 3A ) . While Brca2F/F;PBCre4 and Brca2F/F;Trp53F/+;PBCre4 LG PIN lesions had individual or small groups of γH2AX positive cells , Brca2;Trp53 homozygous mutant prostates had large focal areas with many positive cells . These areas of γH2AX correlated with focal HG PIN lesions and were predominantly present in the AP and DP . Deletion of Brca2 frequently results in increased levels of cellular apoptosis , presumably as a result of increased constitutive DNA damage . To determine the level of apoptosis in prostate epithelia after deletion of Brca2 and Trp53 we used the TUNEL assay , which has been used to identify apoptosis in Brca2 null neural tissue [32] . Brca2 mutant prostates showed a 3 fold increase in TUNEL positive cells in areas of hyperplasia and LG PIN , compared to control prostates that had few apoptotic cells ( 0 . 3% TUNEL positive cells vs 0 . 1% in control ) ( Figure 3B ) . A 4 fold increase in apoptosis was observed in Brca2F/F;Trp53F/+;PBCre4 PIN foci ( 0 . 4% TUNEL positive cells vs 0 . 1% in control ) . Notably , there was a 20 fold increase in apoptotic cells in areas of HG PIN in Brca2;Trp53 homozygous mutant prostates ( 2% TUNEL positive cells vs 0 . 1% in control ) ( Figure 3B ) . An anti-Caspase-3 antibody and histological analysis confirmed that TUNEL positive cells were apoptotic and not the result of labelling damaged DNA ( data not shown ) . Ki-67 is a marker of proliferating cells and a prognostic indicator in prostate cancer [34] . Analysis of Ki-67 showed a low number of proliferating cells in control prostates that increased 6 fold in areas of hyperplasia and LG PIN in Brca2F/F;PBCre4 mutant prostates ( 2 . 2% Ki-67 positive cells vs 0 . 4% in control ) ( Figure 4A ) . Levels of proliferation were 9 fold higher in Brca2F/F;Trp53F/+;PBCre4 PIN lesions ( 3 . 5% Ki-67 positive cells vs 0 . 4% in control ) , and dramatically increased by 30 fold in Brca2F/F;Trp53F/F;PBCre4 HG PIN lesions ( 12% Ki-67 positive cells vs 0 . 4% in control ) ( Figure 4A ) . Levels of the androgen receptor ( AR ) expressed in the luminal epithelium usually increase in the nucleus during human prostate carcinoma progression [35] . Brca2F/F;PBCre4 and Brca2F/F;Trp53F/+;PBCre4 mutant prostates had increased AR expression in the cytoplasm and nucleus of epithelial cells that correlated with regions of LG PIN ( Figure 4B ) . The level of AR increased significantly more in Brca2F/F;Trp53F/F;PBCre4 HG PIN lesions where it was found predominantly in the nucleus of luminal epithelial cells ( Figure 4B ) . The human and mouse prostate comprise of basal , luminal and rare neuroendocrine epithelial cells . In addition , intermediate or transit amplifying ( TA ) cells are observed in human prostates [36] . Brca2;Trp53 homozygous mutant HG PIN lesions frequently contained large groups of p63-expressing cells , a marker of basal cells ( Figure 4C ) . Instead of their normal flat shape and position basal to the luminal cells , some p63-expressing cells were rounder and in a position near the lumen of the prostate . Sections fluorescently double labelled with p63 and the basal cell cytokeratin CK5 confirmed the presence of clusters of aberrant basal cells that protrude into the lumen ( Figure 4D ) . Brca2;Trp53 homozygous mutant HG PIN lesions were then double labelled with CK5 and CK8 , a marker of differentiated luminal cells . Areas of neoplasia showed an increase in CK8-expressing luminal cells that were often adjacent to a population of expanded CK5-expressing basal cells ( Figure 4D ) . These regions occasionally had cells that were labelled with both CK8 and CK5 , similar to human TA cells that co-express basal and luminal markers , which are not seen in control mouse prostates ( Figure 4D ) [36] . Androgen ablation is the standard treatment for human prostate cancer . To assess the response of neoplasias formed after deletion of Brca2 and Trp53 to androgen ablation , we surgically castrated animals at 16 months when HG PIN lesions have already formed and analysed them 4 days post-castration . Castration of control animals resulted in normal prostate regression , with a reduction in lumen size ( Figure 5A ) . Brca2F/F;Trp53F/F;PBCre4 animals still contained focal areas of neoplasia with atypical cells following androgen-depletion ( 8/8 mutant animals ) ( Figure 5A ) . However , we did not observe ducts with filled lumens as seen in areas of PIN in non-castrated mutant animals . Following castration of control male mice , AR is expressed at low levels predominantly throughout the cytoplasm of luminal and stromal cells ( Figure 5B ) . Castration of Brca2F/F;Trp53F/F;PBCre4 animals resulted in AR expression in a diffuse pattern throughout most prostate luminal and stromal cells ( Figure 5B ) . However , AR is expressed at higher levels in the nucleus of a small number of cells in the PIN lesions of mutant animals post-castration ( Figure 5B ) . A reduction in the level of circulating androgens results in apoptosis of the AR-expressing luminal epithelial cells , normally leading to initial prostate tumour regression . As expected , luminal cells of control castrated animals contained apoptotic cells throughout the prostate post-castration ( Figure 5C ) . Castrated Brca2;Trp53 homozygous mutants contained cells in all prostatic lobes undergoing programmed cell death , including apoptotic cells in PIN lesions ( 2 . 2% TUNEL positive cells vs 1 . 8% in castrated control , p = 0 . 49 ) ( Figure 5C ) . Interestingly , Ki-67 staining showed an 18 fold increase in proliferating cells in areas of neoplasia in Brca2;Trp53 mutants that persisted after castration , compared to control castrated animals ( 5 . 5% Ki-67 positive cells vs 0 . 3% in castrated control ) ( Figure 5D ) . Studies on human carriers of deleterious BRCA2 mutations have implicated this gene in prostate cancer etiology , but its function in this malignancy is unclear . We have undertaken a genetic analysis of Brca2 function in the adult mouse prostate to define its role in prostate cancer and to create an in vivo model of Brca2-dependent prostate disease progression . Our study has demonstrated that loss of Brca2 in the mouse prostate epithelium results in hyperplasia and LG PIN . These lesions have an increase in the number of cells with DNA damage and apoptotic cells , which could be the result of the impairment of DNA repair pathways . This demonstrates not only that Brca2 can play a role in the initiation of prostate neoplasia but also that other factors are required for prostate tumour progression . Deletion of Brca2 and Trp53 in mouse prostate epithelia resulted in a shorter latency and increased frequency of prostate neoplasia compared to deletion of Brca2 alone ( Figure 6 ) . Moreover , the severity of neoplasia increased in Brca2;Trp53 mutants , with the formation of hyperplasia and LG PIN at initial stages followed by a high incidence of multi-focal , proliferative HG PIN lesions with progressive cellular atypia ( Figure 6 ) . HG PIN lesions that form after deletion of Brca2 and Trp53 contained many cells with DNA damage , indicating increased genomic instability [17] . Multi-focal lesions are a common feature of human prostate cancer and may be due to defects in the DNA damage response [37] . The formation of HG PIN lesions in Brca2;Trp53 mutant prostates may reflect the loss of key regulatory p53-dependent functions in response to DNA damage controlling cell-cycle checkpoints , apoptosis and senescence [38] . This demonstrates a co-operative tumour suppressor function of Brca2 and Trp53 in the prostate similar to the mammary gland [24]–[26] . A recent study investigating whether TP53 and BRCA2 are frequently mutated together in human prostate cancer found that TP53 overexpression could not distinguish BRCA2 carriers with prostate cancer from a control group of prostate cancer cases [39] . However , this study was limited by the small number of BRCA2 cases and inability to detect TP53 mutations that do not stabilize the protein , which are frequently detected in tumours with impaired homologous recombination [40] . Deletion of Brca2 in other tissues frequently leads to an increase in apoptosis , which is partially or fully rescued upon loss of Trp53 [31] , [32] , [41] . However , we see more apoptotic cells in Brca2;Trp53 mutant HG PIN lesions than in Brca2 mutant LG PIN lesions . The increase in apoptosis in HG lesions could be due to the rapid accumulation of additional mutations by proliferating cells , which causes catastrophic amounts of DNA damage . This suggests that while some Brca2-null cells may be rescued from apoptosis after loss of Trp53 , other cells in Brca2;Trp53 deficient HG neoplastic lesions undergo p53-independent cell death . Several different DNA damage-induced p53-independent mechanisms of apoptosis have been reported in different cell types [42] . AR expression is increased in the nucleus of cells in Brca2;Trp53 HG PIN lesions suggesting they are androgen sensitive . Consistent with this , castration of Brca2F/F;Trp53F/F;PBCre4 animals led to regression of PIN and a reduction of cells within the lumen . This suggests that BRCA2-driven prostate cancer would initially respond to conventional androgen ablation . However , atypical cells persist in Brca2;Trp53 HG PIN lesions that continue to proliferate , indicating these lesions may be able to re-grow and become castration-resistant . Interestingly , some cells in Brca2;Trp53 mutant PIN lesions expressed AR at higher levels in the nucleus after castration , indicative of active AR signalling . Castration-resistant human prostate cancer growth commonly remains AR-dependent and is thought to occur through several mechanisms including AR amplification , AR mutation , changes in AR co-regulators and growth factor activation [43] . The presence of nuclear AR in castrated Brca2;Trp53 mutant prostates may indicate the regulation of proliferation by this factor after androgen depletion . We often observed an increase in p63 positive basal cells , the presence of TA-like cells and an adjacent expansion of luminal cells in Brca2;Trp53 mutant HG PIN lesions . This suggests increased proliferation of the basal cell population , with maintenance of differentiation into luminal cells . Similarly , deletion of the tumour suppressor Pten with the PBCre4 transgene results in tumour formation with an increase in basal cells that contain a progenitor cell sub-population , the presence of TA-like cells and luminal cell differentiation [44] . The increase in basal progenitor cell population could represent an expansion of cancer-initiating cells , indicating that HG PIN lesions in Brca2;Trp53 mutant prostates may originate from these cells . Several other murine models of prostate cancer that utilize the PBCre4 transgene display increased and aberrant p63 expression during early stages of cancer progression [29] , [30] . This transgene is expressed in both the basal and luminal cells of the mouse prostate [44] . In contrast , deletion of Pten using a PSA-driven Cre only expressed in luminal cells results in cancer without an expansion in p63 cells [45] . However , this model has slower kinetics than the Pten; PBCre4 model and tumours are initiated from cells in the luminal epithelial compartment . Frequently during human prostate tumour progression there is an increase in TA/progenitor cells , which has led to the proposal that these cells could be tumour-initiating cells [36] , [44] , [46] . Taken together , these data suggest that an increase in the basal progenitor cells could be a common early event in prostate neoplasia but may be dependent on the origin of the cancer-initiating cell . Although we frequently observed HG PIN lesions in Brca2;Trp53 mutant prostates no invasive carcinoma formed . In contrast , deletion of Brca2 and Trp53 in the mouse mammary gland results in invasive carcinoma at 6 months , and consistent with this , human carriers of BRCA2 mutations have a high-risk of breast cancer . The lack of prostate carcinoma in our mutants may reflect the relatively low penetrance of prostate cancer in human BRCA2 mutation carriers and suggests there is only a subset of BRCA2 carriers that develop aggressive forms of the disease [13]–[16] . This subset may be dependent on additional genetic modifiers or environmental factors that influence the risk of individuals carrying a BRCA2 mutation forming prostate tumours [47] . It is possible that human carriers of deleterious BRCA2 mutations frequently form HG PIN lesions similar to our Brca2;Trp53 model but that never progress to carcinoma and therefore go undetected . Ongoing work into genetic modifiers of BRCA2 may identify which subgroups of patients with BRCA2 mutations are more at risk of developing aggressive forms of the disease . Variations in genetic background can have a modifying effect on prostate tumour development in mice with tumour suppressor deletions [48] . Due to the complex nature of mouse breeding we were not able to investigate the effects of genetic background on the prostate lesions observed in Brca2;Trp53 mutant mice in this study . Although a change in genetic background may alter the frequency of tumour phenotype , we only observed PIN in mutant animals suggesting these lesions are an effect of Brca2 and Trp53 loss . This murine study has demonstrated that deletion of the tumour suppressor Brca2 results in LG PIN , with the additional loss of a second tumour suppressor Trp53 leading to HG PIN . Other mouse models of tumour suppressor gene loss result in varying degrees of prostate tumour progression . Mice with prostate-specific homozygous Pten deletion progress to invasive carcinoma and metastasis [28] . The further loss of Trp53 in this model results in a shorter latency to invasive carcinoma [49] . Deletion of Nkx3 . 1 , a gene involved in prostate epithelial cell differentiation , leads to mice that develop epithelial hyperplasia and dysplastic lesions that resemble human PIN , but do not progress to invasive carcinoma [50] , [51] . Our pre-invasive model could be used in the future to test the response to potential therapeutic agents and combination therapies . For example , a recent synthetic lethal approach using PARP inhibitors has been used successfully to specifically induce cytotoxicity in HR-deficient cells [52] . Promising phase I clinical data in BRCA2 carriers with a PARP inhibitor has shown antitumour activity , including resolution of bone metastases in one patient with prostate cancer [53] . These Brca2 mutant mice may provide a useful model to examine cellular responses , such as apoptosis , to combinations of therapies for optimisation of treatment . In addition , this study demonstrates that Brca2 acts as a tumour suppressor and can interact genetically with Trp53 deficiency in the prostate preventing DNA damage accumulation and neoplasia progression . Animals were handled in strict accordance with UK Home Office regulations . Brca2F/F ( targeting exon 11 ) mice and Trp53F/F ( targeting exons 2–10 ) mice [25] and ARR2PBCre transgenic mice , PBCre4 [27] , have been previously described . The animals were bred on a mixed genetic background . Histological phenotype of samples was assessed on haematoxylin and eosin stained sections . Serial sections were then stained for immunohistochemical analysis . Histological assessment was based on published guidelines and assisted by a pathologist [54] , [55] . PIN lesions noted as LG were equivalent to PIN I-II and those noted as HG were equivalent to PIN III-IV in Park et al [54] . The two-sample Z-test was performed to test if there is a significant difference between groups of animals . Ki-67 or TUNEL staining was performed by immunohistochemistry on sections and cells stained with nuclear brown DAB chromogen were counted as positive . Cells from at least 4 high power fields were counted per animal , which totalled more than 900 cells per animal . Five animals of each genotype were analysed . Randomly selected fields were counted for control analysis and sections corresponding to histologically identified areas of hyperplasia and PIN were counted for mutant animals . All values are significant with p<0 . 05 using Student t-test unless otherwise stated . Antibody stains were done on paraffin sections as previously described [56] . The following antibodies were used: Ki-67 ( TEC-3 , Dako , 1∶200 with amplification ) , AR ( PG-21 , Upstate , 1∶250 with amplification ) , p63 ( 4A4 , Santa Cruz Biotechnology , 1∶200 with amplification , 1∶50 ) , γH2AX ( JBW301 , Upstate , 1∶800 with amplification ) , CK5 ( PRB-160P , Covance , 1∶50 ) , CK8 ( MMS-162P , Covance , 1∶200 ) . The ABC elite vector kit was used for amplification using biotinlyated secondary antibodies ( Vector Laboratories ) and the DAB substrate ( Dako ) . Secondary fluorescent antibodies were obtained from Molecular Probes and were used at a 1∶1000 dilution . TUNEL analysis was carried out using the ApopTag apoptosis detection kit ( Chemicon ) .	In Western countries , prostate cancer is the most common male cancer and the second biggest cause of cancer-related deaths in men . Men with a familial history of either breast or ovarian cancer have an elevated predisposition to prostate cancer , suggesting there is a genetic element to this disease . Indeed , the inheritance of a mutated form of the breast cancer susceptibility gene BRCA2 has been linked to the development of prostate cancer , although the precise role that BRCA2 dysfunction plays in the development of prostate cancer is unclear . To address this , we have generated an animal model in which the mouse Brca2 gene is specifically deleted in the adult prostate . These mice develop precancerous prostate lesions , which progress in severity and incidence with the loss-of-function of an additional tumour suppressor , Trp53 . Importantly , blocking male steroidal hormone production by castration leads to partial regression of the prostate lesions , however cells continue to proliferate after androgen withdrawal . This suggests human BRCA2 mutant prostate tumours , like the majority of prostate cancers , will respond to hormone therapy , but will relapse , as frequently occurs in this disease . In summary , our model suggests that BRCA2 acts as a tumour suppressor in the prostate and provides a pre-invasive model to test novel therapeutics .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "oncology/prostate", "cancer", "genetics", "and", "genomics/cancer", "genetics" ]	2010	Brca2 and Trp53 Deficiency Cooperate in the Progression of Mouse Prostate Tumourigenesis
Toll-like receptor 3 ( TLR3 ) senses dsRNA intermediates produced during RNA virus replication to activate innate immune signaling pathways through adaptor protein TRIF . Many viruses have evolved strategies to block TLR3-mediated interferon signaling via targeting TRIF . Here we studied how hepatitis C virus ( HCV ) antagonizes the TLR3-mediated interferon signaling . We found that HCV-encoded NS4B protein inhibited TLR3-mediated interferon signaling by down-regulating TRIF protein level . Mechanism studies indicated that the downregulation of TRIF by NS4B was dependent on caspase8 . NS4B transfection or HCV infection can activate caspase8 to promote TRIF degradation , leading to suppression of TLR3-mediated interferon signaling . Knockout of caspase8 can prevent TRIF degradation triggered by NS4B , thereby enhancing the TLR3-mediated interferon signaling activation in response to HCV infection . In conclusion , our work revealed a new mechanism for HCV to evade innate immune response by blocking the TLR3-mediated interferon signaling via NS4B-induced TRIF degradation . Hepatitis C virus ( HCV ) is an enveloped , single-stranded RNA virus belonging to the Flaviviridae family . HCV has a 9 . 6-kb RNA genome and encodes a large polyprotein of over 3000 amino acids which is cleaved into structural proteins ( core , E1 and E2 ) and nonstructural proteins ( p7 , NS2 , NS3 , NS4A , NS4B , NS5A and NS5B ) . HCV infects approximately 170 million people worldwide , about 80% of whom develop into persistent infection . Persistent HCV infection leads to severe liver diseases , such as liver cirrhosis and hepatocellular carcinoma [1] . No vaccine is available for preventing HCV infection . Interferon ( IFN ) plus ribavirin , the traditional therapy to treat chronic hepatitis C , is not always effective and has strong side effect . Recently developed direct-acting antiviral agents ( DAA ) , including NS3 protease inhibitors , NS5A inhibitors and NS5B nucleotide inhibitors have greatly improved curing efficiency . However , the impact of these highly effective DAAs on global control of HCV infection remains to be seen in the long run as drug-resistant mutations , severe liver disease progression in DAA-cured patients and other newly emerging problems arise [2] . Therefore , HCV infection is still a big threat to human public health . The innate immune system is the first line of host defense against invading viral pathogens , which is initiated by host pattern recognition receptors ( PRRs ) that recognize specific molecular structures known as pathogen-associated molecular pattern ( PAMP ) residing in invading pathogens or produced during pathogen replication [3] . There are three major classes of PRRs: Toll-like receptors ( TLRs ) , RIG-I–like receptors ( RLRs ) and NOD-like receptors ( NLRs ) [4–6] . RLRs , cytosolic RNA helicases that recognize double-stranded RNA ( dsRNA ) or single-stranded RNA ( ssRNA ) with a triphosphate 5’ end , consist of three members RIG-I , MDA5 and LGP2 [7] . Previous study showed that the in vitro synthesized HCV 3’-untranslated regions ( UTR ) RNA can be recognized by RIG-I to trigger IFN response if transfected into hepatocytes , suggesting that the 3’-UTR RNA may contain HCV PAMP [8–10] . Using a hepatic cell line in which HCV infection induces strong IFN response , we recently demonstrated that MDA5 plays a predominant role in sensing HCV PAMP during HCV infection [11] . Furthermore , we showed that LGP2 , another RLR member , is essential for HCV infection-induced IFN signaling , likely facilitating MDA5’s recognition of HCV PAMP [12] . To establish persistent infection , HCV has evolved multiple mechanisms to regulate and evade innate immunity [7 , 13] . HCV NS3/4A serine protease can cleave MAVS , a critical adaptor protein in the RLR-mediated IFN activation , to block RLR-mediated signaling [14] . We and others showed that HCV NS4B protein can also inhibit RLR-mediated IFN activation by targeting STING , an adaptor protein facilitating IRF3 phosphorylation by TBK1 [15–17] . Despite extensive research in how HCV activates and evades RLR-mediated IFN signaling , less has been done for HCV evasion of TLR3-mediated IFN signaling . TLR3 is expressed mainly in early endosome and senses dsRNA which is produced during RNA virus replication [18 , 19] . TIR-domain-containing adaptor protein including IFN-β ( TRIF ) is the sole adaptor protein of TLR3-mediated pathway , and is often targeted by viruses to evade host innate immunity [20–25] . Early studies showed that HCV NS3/4A serine protease targeted TRIF for its cleavage [24 , 26] . Despite these studies , the molecular mechanism underlying HCV blocking TLR3 signaling pathway remains to be explored . In this study we investigated how HCV antagonizes the TLR3-mediated interferon signaling . We found that HCV-encoded NS4B inhibited TLR3-mediated interferon signaling by promoting TRIF protein degradation in a caspase8-dependent manner . Our work revealed a new mechanism for HCV to evade host innate immunity . We and others previously showed that HCV encoded-NS4B protein can inhibit the RLR-mediated IFN signaling by targeting STING [15–17] . In this study we aimed to determine whether NS4B had any effects on the TLR3-mediated IFN signaling . A previous study showed that the addition of poly ( I:C ) to culture medium can activate the TLR3-mediated signaling in PH5CH8 cells , non-neoplastic hepatocytes transformed with large T antigen [27] . Therefore , we first used PH5CH8 cells to analyze the effect of NS4B on the TLR3-mediated signaling . Poly ( I:C ) was either added directly to the culture medium ( M-pIC ) or transfected ( T-pIC ) into PH5CH8 cells that were firstly transfected with a reporter plasmid expressing IFN-β promoter-driven luciferase and a plasmid expressing NS4B . In this setting , M-pIC and T-pIC would activate TLR3- and RLR-mediated interferon signaling respectively . In addition , HCV-encoded NS3/4A , known to cleave MAVS to block the RLR-mediated interferon signaling [28] , was included as a control . As shown in Fig 1A and 1B , consistent with previous studies , both NS4B and NS3/4A decreased T-pIC-induced IFN-β response . However , NS4B but not NS3/4A decreased M-pIC-induced IFN-β response , suggesting that NS4B may have an inhibitory effect on the TLR3-mediated IFN signaling . HEK293T cells lack TLR3 expression and thus are defective in the TLR3-mediated interferon signaling [29] . To investigate the effect of NS4B or NS3/4A on the TLR3-mediated interferon signaling in HEK293T cells , HEK293T cells were transfected with plasmids expressing TLR3 , IFN-luciferase reporter and NS4B or NS3/4A , followed by poly ( I:C ) transfection ( T-pIC ) or treatment ( M-pIC ) . Consistent with the observation in PH5CH8 cells , while both NS3/4A and NS4B decreased T-pIC-induced IFN-β response ( Fig 1C ) , only NS4B suppressed M-pIC-induced IFN-β response in the TLR3-reconsitutted HEK293T cells ( Fig 1D ) . Altogether , these results suggested that NS4B may suppress the TLR3-mediated IFN signaling . STING is a transmembrane protein in the ER , and has a cytoplasmic domain that binds ligands to activate IFN signaling [30] . It has been reported that STING interacted with TRIF directly to trigger innate immune response to microbial infection [31] . We and others have shown that HCV NS4B protein can interact with STING to disrupt the RLR-mediated signaling [15–17] . To investigate whether the NS4B-STING interaction was also involved in NS4B suppression of the TLR3-mediated signaling , PH5CH8 cells were transfected with STING-specific siRNA followed by either poly ( I:C ) transfection ( T-pIC ) or treatment ( M-pIC ) . Knockdown of STING expression was confirmed by RT-qPCR ( Fig 2A ) and Western blot ( Fig 2B ) . As shown in Fig 2C , STING knockdown significantly decreased T-pIC-induced IFN-β response but had no effect on M-pIC-induced IFN-β response , suggesting that STING was not involved in the NS4B-inhibited TLR3 signaling pathway . Many viruses target TRIF to block TLR3 signaling [20–25] , and early studies showed that HCV NS3/4A protease can cleave TRIF to shut down the TLR3-mediated signaling [24] . To examine a possibility that NS4B may target TRIF , we transfected HEK293T cells with plasmids expressing the IFN-β promoter-luciferase reporter , NS4B or NS3/4A , and TRIF or MAVS . As shown in Fig 3A , NS4B decreased both TRIF- and MAVS-mediated IFN-β response , while NS3/4A only decreased MAVS-mediated IFN-β response . Next we examined the effects of NS4B or NS3/4A on the protein levels of TRIF and MAVS . As shown in Fig 3B , NS4B had no obvious effect on MAVS protein expression as previously reported [15] , but reduced TRIF protein level . In contrast , NS3/4A induced the cleavage of MAVS as previously reported [28] , but had no effect on TRIF protein level . Furthermore , we transfected HEK293T cells with increasing dose of plasmids expressing NS4B and plasmids expressing TRIF or MAVS . As shown in Fig 3C , NS4B reduced the TRIF protein level in a dose-dependent manner but had no obvious effect on MAVS protein level . Next we examined the effect of NS4B on endogenous TRIF protein expression . HEK293T ( Fig 3D ) and PH5CH8 ( Fig 3E ) cells were transfected with plasmids expressing NS4B or NS3/4A respectively , and the endogenous TRIF protein level was analyzed by Western blot . As shown in Fig 3D and 3E , NS4B decreased endogenous TRIF protein expression in both HEK293T and PH5CH8 cells . Next we examined the TRIF protein expression in HCV-infected hepatocytes . Huh7 cells were infected with HCVcc at a multiplicity of infection ( MOI ) of 5 , and the mRNA and protein levels of TRIF were determined at day 1 , 2 , 3 and 4 post-infection by RT-qRCR or Western blotting respectively . As shown in Fig 3F , HCV infection decreased endogenous TRIF protein level . In contrast , TRIF mRNA level was slightly increased in the HCV infected cells ( S1A Fig ) . Altogether these results suggested that reduction of TRIF expression by NS4B likely occurred at a translational or post-translational level . We noted that the TRIF protein level increased over time in the mock infected cells ( Fig 3F ) , probably resulting from the influences of cell density or nutrients in the culture medium on TRIF . Next we investigated potential mechanisms involved in the NS4B-induced TRIF protein degradation . HEK293T cells transfected with the NS4B-expressing plasmid were treated with different inhibitors of protein degradation pathways , including chloroquine ( lysosome inhibitor ) , MG132 ( proteasome inhibitor ) and Z-VAD-FMK ( caspases inhibitor ) , and the TRIF protein level was determined by Western blot . As shown in Fig 4A , the NS4B-triggered TRIF protein degradation was partially rescued by Z-VAD-FMK but not by chloroquine or MG132 , suggesting that the NS4B-induced reduction of TRIF protein level may involve caspase-mediated protein degradation . TRIF has been shown to undergo caspase8 or caspase9-dependent cleavage [32] . To determine whether caspase8 or caspase9 played a role in the reduction of TRIF protein level induced by NS4B , we constructed caspase8- or caspase9-knockout HEK293T cells using CRISPR-Cas9 technology ( S2 Fig ) . HEK293T cells stably expressing caspase9-sgRNA ( #4 ) or caspase8-sgRNAs ( #1 and #3 ) displayed decent efficiency in reducing caspase9 or caspase8 expression , and thus were chosen for the further study ( S2 Fig ) . These caspase9- or caspase8-knockout cells as well as control HEK293T cells expressing sgEGFP were transfected with plasmids expressing Flag-tagged NS4B , and the endogenous TRIF expression was determined by Western blot . As shown in Fig 4B , caspase9 knockout had no apparent effect on the NS4B-triggered TRIF protein degradation . In contrast , caspase8 knockout rescued the NS4B-triggered TRIF protein degradation ( Fig 4C ) , suggesting that caspase8 was critical for the TRIF protein degradation induced by NS4B . It was reported that two amino acid residues ( 281D and 289D ) of TRIF are critical for the FasL-triggered TRIF cleavage by caspase8 [32] , therefore we next examined the effect of point mutation of these two residues on NS4B-triggered TRIF degradation . Aspartic acid was mutated to glutamic acid at these two residues in TRIF ( TRIF-281E289E ) . Next , plasmids expressing TRIF-wt or TRIF-281E289E were co-transfected with plasmids expressing NS4B or NS3/4A into HEK293T cells , and TRIF expression level was determined by Western blot . As shown in Fig 4D , while the wild-type TRIF protein expression was reduced by NS4B , the mutant TRIF protein expression level remained unchanged , suggesting that 281D and 289D are important for NS4B-triggered TRIF degradation . Although primary human hepatocytes ( PHH ) express TLR3 , hepatoma-derived Huh7 cells , the sole human hepatic cell line that supports efficient HCV infection in vitro and has been widely used for studying HCV infection and replication in cell culture , do not express a detectable level of TLR3 and thus are defective in the TLR3-mediated IFN signaling [27] . To study how NS4B-triggered TRIF degradation affects the TLR3-mediated signaling in response to HCV infection , we constructed Huh7 cells stably expressing TLR3 by lentiviral transduction . The expression of TLR3 was verified by Western blot ( S3A Fig ) . Huh7-TLR3 cells were treated by poly ( I:C ) in culture medium ( M-pIC ) , and the IFN-β and MxA mRNA levels were determined by RT-qPCR . As shown in S3B and S3C Fig , M-pIC induced IFN-β and MxA production in Huh7-TLR3 cells , but not in Huh7 cells , suggesting that the TLR3-mediated IFN signaling was restored in Huh7-TLR3 cells . Next we infected Huh7-TLR3 cells with HCVcc at MOI of 5 , and determined the endogenous TRIF protein level and caspase8 activation by Western blot . Consistent with the observations in HCV-infected Huh7 cells ( Fig 3F ) , HCV infection decreased endogenous TRIF protein level in Huh7-TLR3 cells ( Fig 5A ) . In addition , HCV infection resulted in cleavage of pro-caspase8 , a hallmark of caspase8 activation ( Fig 5A ) . To investigate whether HCV infection induces IFN response in Huh7-TLR3 cells , we infected Huh7-TLR3 and control Huh7-vec cells with HCVcc at an MOI of 5 , and determined IFN-β and ISGs mRNA levels by RT-qPCR . As shown in Fig 5B–5E , the presence of TLR3 significantly augmented the HCV infection-induced IFN signaling in Huh7-TLR3 cells . These results suggested that HCV infection still activates the TLR3-mediated IFN signaling while reducing the critical adaptor TRIF protein level in Huh7-TLR3 cells , possibly due to incomplete inhibition of TRIF by NS4B . Next we assessed potential effects of the RLR-mediated signaling on TLR3-mediated IFN signaling in HCV-infected Huh7-TLR3 cells . To do so , we knocked out the expression of MAVS , a critical adaptor protein in the RLR-mediated IFN signaling using the CRISPR-Cas9 technology ( Fig 6A ) . The knockout of MAVS significantly diminished the IFN-β expression triggered by transfection of HCV 3’-UTR RNA or poly ( I:C ) ( T-pIC ) , two ligands known to activate RIG-I- or MDA5-mediated IFN signaling respectively [11] ( Fig 6B ) , suggesting that the RLR signaling was impaired in this MAVS-knockout cell . Nevertheless , this cell line remains fully responsive to the M-pIC-induced TLR3 signaling ( S4 Fig ) . We then infected Huh7-TLR3-sgMAVS-#1 cells and control Huh7-TLR3-sgEGFP cells with HCVcc at an MOI of 5 . The IFN-β and ISGs mRNA levels were determined by RT-qPCR . As shown in Fig 6C , RLR signaling blockade by knocking out the adaptor protein MAVS had no significant effect on the IFN-β induction in the Huh7-TLR3 cells , suggesting that the activation of the TLR3-mediated IFN signaling during HCV infection does not involve the RLR signaling . There was a slight decrease of MxA and ISG56 mRNA in the MAVS-knockout cells on day 4 post-infection ( Fig 6D and 6E ) , which may result from amplification of IFN signaling contributed by the RLR signaling at late time points given that both RIG-I and MDA5 are themselves interferon-stimulated genes and can be up-regulated by the TLR3-mediated signaling . To assess the contribution of NS4B/caspase8-mediated TRIF degradation to suppression of TLR3 signaling during HCV infection , we knocked down caspase8 expression in Huh7-TLR3 cells . Huh7-TLR3 cells were transduced with lentivirus expressing caspase8-specific sgRNAs ( #1 and #3 ) that efficiently reduced caspase8 expression in HEK293T cells ( Figs 4C and S2B ) . The knockout of caspase8 in the transduced Huh7-TLR3 cells was verified by Western blot ( S5 Fig ) . These cells and control Huh7-TLR3 cells ( sgEGFP ) were infected with HCVcc . The endogenous TRIF protein level and caspase8 activation were analyzed by Western blot , and IFN-β and ISGs mRNA levels were determined by RT-qPCR . As shown in Fig 7A , HCV infection activated caspase8 and reduced TRIF protein level in the control cells , but had much less effect on TRIF protein level in the caspase8-knockout cells . Importantly , restoration of TRIF protein level in the caspase8-knockout cells was accompanied by the enhanced IFN-β , MxA and ISG56 mRNA levels ( Fig 7B–7D ) . HCV mRNA levels were comparable among the three cells in the first 3 days after infection , but were lower in the caspase8-knockout cells on day 4 and 5 post-infection ( Fig 7E ) , possibly due to more active antiviral IFN signaling in these cells . To further confirm the role of caspase8 in the NS4B-mediated TRIF degradation and HCV infection-induced IFN signaling , we subcloned Huh7-TLR3-sgcaspase8-#1 cells and obtained a single clone of caspase8-knockout cell ( Huh7-TLR3-sgcaspase8-#1-c4 ) . The caspase8 knockout was verified by genomic sequencing and Western blot ( S6A Fig ) . We then infected Huh7-TLR3-sgcaspase8-#1-c4 and control cells with HCVcc at an MOI of 5 , and determined the TRIF expression by Western blot as well as the IFN-β and ISGs mRNA levels by RT-qPCR . Consistent with the observations in Fig 7 , caspase8 knockout restored TRIF protein levels ( S6B Fig ) and enhanced IFN signaling in the HCV-infected Huh7-TLR3 cells ( S6C–S6F Fig ) . Altogether , these results suggested that NS4B promoted caspase8-dependent TRIF degradation to suppress the TLR3-mediated IFN signaling during HCV infection . To establish chronic infection , HCV has evolved multiple strategies to counteract IFN signaling . It has been demonstrated that HCV-encoded NS3/4A serine protease cleaves MAVS and TRIF to shut down the IFN signaling mediated by RLRs and TLR3 [24 , 28] . Growing evidence showed that HCV employs additional strategies to disrupt the RLR- and TLR3-mediated IFN signaling . We and others previously reported that HCV NS4B can block RLR-mediated interferon signaling by targeting STING [15–17] . In this study , we provided several lines of evidence to demonstrate that NS4B can also disrupt the TLR3-mediated signaling by targeting the adaptor protein TRIF for its degradation . First , NS4B transfection blocks IFN signaling activated by extracellular poly ( I:C ) , known to be recognized by TLR3 ( Fig 1 ) . Second , the disruption of TLR3-mediated signaling by NS4B was associated with the reduction of TRIF at the protein level ( Fig 3A–3E ) , but not at the mRNA level ( S1A Fig ) . Third , the NS4B-triggered TRIF protein degradation and blockade of TLR3-mediated IFN signaling were recapitulated in HCV infected cells ( Figs 3F , 5A and 7 ) , suggesting that interference of the TLR3-mediated IFN signaling by NS4B indeed takes place in the context of HCV infection . The RLR-mediated IFN signaling in HCV-infected hepatocytes has been extensively studied , but less has been done so for the TLR3-mediated IFN signaling during HCV infection . We showed that HCV infection does trigger the TLR3-mediated IFN signaling ( Fig 5 ) . The magnitude of IFN activation in the HCV-infected Huh7-TLR3 cells is relatively low ( less than 10-fold induction ) , likely due to the suppression of NS4B . When this NS4B-mediated suppression is relieved by knocking out caspase8 , the IFN induction level can be further enhanced ( Figs 7 and S6 ) . These data demonstrated the importance of the TLR3 signaling in innate immune response against HCV infection and also highlighted the necessity for HCV to control this innate immune pathway . Furthermore , we showed that the RLR signaling blockade had no significant effect on the TLR3-mediated IFN activation during HCV infection ( Fig 6 ) . In addition , we showed that the enhancement of TLR3-mediated IFN signaling by knocking out caspase8 and preventing TRIF degradation was not affected by the RLR signaling blockade ( S7 Fig ) . These results suggest that the TLR3 and the RLR signalings are likely two independent host innate immune responses to HCV infection , which the virus must find ways to evade simultaneously . We explored potential molecular mechanisms underlying down-regulation of TRIF protein level by NS4B . Unlike NS3/4A , NS4B does not have an enzymatic activity to cleave TRIF . Co-immunoprecipitation assays showed that NS4B did not directly interact with TRIF . Our results showed that NS4B-triggered TRIF degradation was caspase-dependent . Z-VAD-FMK , a pan-caspase inhibitor partially restored the TRIF protein expression in the presence of NS4B ( Fig 4A ) . Our results showed that knockout of caspase8 but not caspase9 blocked the NS4B-induced TRIF degradation ( Figs 4C and 7A ) and enhanced IFN signaling in response to HCV infection ( Fig 7 ) . Furthermore , we showed that HCV infection indeed activates caspase8 in the cells particularly at late time points of infection , coincide with the onset of TRIF degradation and reduction in the interferon signaling ( Figs 5 and 7 ) . Two amino acid residues 281D and 289D in TRIF that were previously reported to be critical for the FasL-induced TRIF cleavage by caspase8 [32] seemed to be also important for NS4B-triggered TRIF degradation ( Fig 4D ) , suggesting that NS4B-triggered TRIF degradation may share the similar molecular mechanism with FasL-induced TRIF cleavage by caspase8 . NS4B is an ER membrane-associated protein and can induce morphological changes of ER membrane during active HCV genome replication , leading to ER stress [33 , 34] . It has been reported that ER stress induces FADD oligomerization , which in turn interacts and activates caspase8 through its death effector domain ( DED ) [35] . Therefore , it is conceivable to speculate that NS4B may cause ER stress to activate caspase8 . More research will be needed to test this hypothesis and to understand how the caspase8 activation eventually leads to the TRIF degradation . Previous studies reported that the activation of TLR3 signaling also leads to apoptosis in a manner that requires the involvement of TRIF [36 , 37] . Our finding that NS4B promotes the TRIF degradation raises a possibility that NS4B may also counter apoptosis of HCV-infected hepatocytes , which may contribute to hepatocyte proliferation and liver regeneration , an important prerequisite for development of hepatocellular carcinoma . More studies will be needed to evaluate the potential role of NS4B in antagonizing apoptosis of host hepatocytes . Previous studies demonstrated that TRIF was targeted by HCV-encoded NS3/4A protease for its proteolysis in cell free system and in HEK293 and Huh7 cells [24 , 26] . Interestingly , our results showed that HCV NS3/4A serine protease did not induce the cleavage or degradation of TRIF in HEK293T and PH5CH8 cells . Of notes , our results were consistent with some other groups’ finding that NS3/4A is incapable of cleaving TRIF in PH5CH8 cells and human primary hepatocytes [38–40] . This difference may result from possible different subcellular localization of TRIF and/or NS3/4A in the different cells as well as technical difficulties in detecting low level of TRIF proteins as suggested in other literature [41] . In addition , NS3/4A may act together with NS4B to degrade TRIF in the context of HCV infection . More research will be needed to address this issue . In summary , we found that HCV NS4B protein induces caspase8-dependent TRIF degradation to block TLR3 signaling . Our work revealed a new strategy for HCV to evade innate immune response and should help understand molecular mechanisms underlying persistent HCV infection . HEK293T , Huh7 and their derivative cells were maintained in complete Dulbecco’s modified Eagle’s medium ( DMEM ) ( Invitrogen , Carlsbad , CA , USA ) supplemented with 10% fetal bovine serum , 10 mM HEPES , 2 mM L-glutamine , 100 U of penicillin/ml , and 100 mg of streptomycin/ml . PH5CH8 cells were maintained in DMEM/F12 ( 1:1 ) supplemented with 2 mM L-glutamine , 100 ng/ml epidermoid growth factor ( Toyobo , Osaka Japan ) , 10 μg/ml insulin ( Sigma-Aldrich , St . Louis , MO , USA ) , 5 μg/ml linoleic acid ( Sigma-Aldrich ) , 106 nM hydrocortisone ( Sigma-Aldrich ) , 107 nM selenium ( Sigma-Aldrich ) , 5μg/ml transferrin ( Sigma-Aldrich ) , 100 ng/ml prolactin ( Sigma-Aldrich ) , and 2% fetal bovine serum . All cells were cultured in humidified air containing 5% CO2 at 37°C . HCVcc preparation was as previously described [42] . 8 x 104 PH5CH8 or HEK293T cells seeded in 48-well plates overnight were transfected with 20 ng/well plasmid expressing IFN-β-Luciferase reporter [11] , 20 ng/well of plasmid expressing CMV promoter-driven Renilla luciferase and 300 ng/well plasmid expressing NS4B or NS3/4A . One day after transfection , the cells were either transfected with 400 ng/well poly ( I:C ) ( Invivogen , San Diego , CA , USA ) for 16 h or treated with 50 μg/ml poly ( I:C ) in culture medium for 6 h . Cell lysates were assayed for the luciferase activity using the Dual-Luciferase Reporter Assay System ( Promega ) . The protocols and sequences of primers for quantifying HCV RNA , human IFN-β , MxA , ISG56 , STING and Actin were described previously [15 , 42] . The sequences of primers for quantifying TRIF mRNA were F: 5′- ATCTGGGAGTGTTCGTCCAG-3′; R: 5′- CCAGACTGTGTCATCCCCTT-3′ . The protocol was as described previously [15] . Antibodies against Flag and β-actin were obtained from Abmart ( Shanghai , China ) . HCV NS3 and core monoclonal antibodies were generated by our laboratory [43] . The monoclonal antibodies against TRIF , caspase8 and caspase9 were obtained from Cell Signaling Technology ( Danvers , MA , USA ) . The anti-MAVS monoclonal antibody , Goat-anti Mouse HRP antibody and Goat-anti Rabbit HRP antibody were obtained from Santa Cruz Biotechnology ( Heidelberg , Germany ) . The protein levels were quantified by Image J from at least two independent experiments , normalized against internal Actin control and expressed as values relative to the control or mock infection . The statistical analysis of protein quantification was shown in S8 Fig . The siRNA-based knockdown protocol was as described previously [15] . Briefly , PH5CH8 cells seeded in 12-well plates were transfected with 1 μg of siRNA plasmids for one day , followed by culturing in DMEM containing 3 μg/ml puromycin for another day . After the puromycin selection , the cells were either transfected with 400 ng/well poly ( I:C ) for 16 h or treated with 50 μg/ml poly ( I:C ) in culture medium for 6 h . The cells were then analyzed for IFN-β mRNA level by RT-qPCR assay . Lentiviruses expressing TLR3 were generated by co-transfecting HEK293T cells with pLVX-TLR3-IRES , packaging vector psPAX2 and envelop vector pMD2 . G using Lipofectamine 2000 ( Invitrogen ) . Culture supernatants containing lentiviruses were harvested at 48 h post-transfection , passed through a 0 . 45 μM-pore-size filter , and used to infect Huh7 cells . Huh7 cells seeded in 6-well plate were infected with 1 . 2 ml lentivirus expressing TLR3 to generate Huh7-TLR3 stable cell line . The TLR3 expression in Huh7-TLR3 cells was verified by Western blot . The sequences of 5 sgRNAs targeting caspase9 were #1: 5′-GCAGGCAGCTGATCATAGATC-3′; #2: 5′-GCTTCGTTTCTGCGAACTAAC-3′; #3: 5′-GCTCTTGAGAGTTTGAG-3′; #4: 5′-GCTGAGCATGGAGCCCTG-3′; #5: 5′-GACTCACGGCAGAAGTTC-3′ . Four sgRNAs targeting caspase8 were #1: 5′-GCTCAGGAACTTGAGGG-3′; #2: 5′-GAATGTAGTCCAGGCTC-3′; #3: 5′-GCCTGGACTACATTCCGCAA-3′; #4: 5′-GCTCTTCCGAATTAATAGAC-3′ . Three sgRNAs targeting MAVS were #1: 5’-GGTTCCCTGAGAGTGTGC- 3’; #2: 5’-GTGAGCTAGTTGATCTCG-3’; #3: 5’-GCACACTCTCAGGGAAC-3’ . To generate lentiviruses expressing sgRNA , HEK293T cells seeded at a density of 1x 106 cells per well in 6-well plates a day ago were transfected with 1 . 3 μg of VSV-G expressing plasmid , 2 . 5 μg of Pcmv-dR8 . 91 plasmid and 2 . 5 μg of lenti-caspase8-sgRNAs or lenti-caspase9-sgRNAs or lenti-MAVS-sgRNAs using Lipofectamine 2000 . Culture supernatants containing lentiviruses were harvested at 48 h post-transfection , passed through a 0 . 45 μM-pore-size filter . HEK293T cells seeded in 6-well plate at a density of 5 x 105 per well were infected by 1 . 2ml lentiviruses expressing caspase9-sgRNAs to generate HEK293T-sgcaspase9 stable cell line . HEK293T or Huh7-TLR3 cells seeded in 6-well plate at a density of 5 x 105 per well were infected with 1 . 2ml lentivirus expressing caspase8-sgRNAs to generate HEK293T-sgcaspase8 or Huh7-TLR3-sgcaspase8 stable cell lines . Huh7-TLR3 cells seeded in 6-well plate at a density of 5 x 105 per well were infected with 1 . 2ml lentivirus expressing MAVS-sgRNAs to generate Huh7-TLR3-sgMAVS stable cell lines . The knockout of caspase9 , caspase8 or MAVS in the transduced cells was determined by Western blot . To subclone caspase8-knockout Huh7-TLR3 cells , Huh7-TLR3-sgcaspase8-#1 cells were diluted and plated in 96-well plates with a density of 0 . 9 cell per well . The cells were grown in complete DMEM supplemented with filtered culture supernatants from Huh7 cell . The knockout was verified by genomic sequencing and Western blot . Statistical analysis was performed using GraphPad Prism 5 software . Student’s t test was used for analyzing the difference between two groups , and One-way analysis of variance ( ANOVA ) followed by Tukey post hoc test was used for analyzing the differences among groups of more than three . ns , P>0 . 05; *P<0 . 05 .	We previously reported that hepatitis C virus ( HCV ) NS4B inhibits the RIG-I–like receptors ( RLR ) -mediated interferon signaling . Here we studied whether NS4B antagonizes the TLR3-mediated interferon signaling in the HCV infection . We found that NS4B protein inhibited TLR3-mediated interferon signaling by down-regulating TRIF protein level , a mechanism that is different from the inhibition of RIG-I-mediated signaling by NS4B . Further studies demonstrated that NS4B can activate caspase8 to promote TRIF degradation , leading to suppression of TLR3-mediated interferon signaling . Knockout of caspase8 can prevent the NS4B-induced TRIF degradation , and thus enhance the TLR3-mediated interferon signaling activation in response to HCV infection . In conclusion , our work revealed a new mechanism for HCV to inhibit the TLR3-mediated interferon signaling by NS4B-induced TRIF degradation .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "medicine", "and", "health", "sciences", "pathology", "and", "laboratory", "medicine", "hepacivirus", "pathogens", "metabolic", "processes", "immunology", "microbiology", "immune", "receptor", "signaling", "viruses", "signal", "inhibition", "rna", "viruses", "membrane", "receptor", "signaling", "proteolysis", "immune", "system", "proteins", "contractile", "proteins", "actins", "proteins", "medical", "microbiology", "gene", "expression", "microbial", "pathogens", "hepatitis", "c", "virus", "hepatitis", "viruses", "metabolism", "toll-like", "receptors", "biochemistry", "signal", "transduction", "cytoskeletal", "proteins", "cell", "biology", "flaviviruses", "viral", "pathogens", "interferons", "genetics", "biology", "and", "life", "sciences", "immune", "receptors", "cell", "signaling", "organisms" ]	2018	Hepatitis C virus NS4B induces the degradation of TRIF to inhibit TLR3-mediated interferon signaling pathway
Trypanosoma cruzi calreticulin ( TcCRT ) is a virulence factor that binds complement C1 , thus inhibiting the activation of the classical complement pathway and generating pro-phagocytic signals that increase parasite infectivity . In a previous work , we characterized a clonal cell line lacking one TcCRT allele ( TcCRT+/− ) and another overexpressing it ( TcCRT+ ) , both derived from the attenuated TCC T . cruzi strain . The TcCRT+/− mutant was highly susceptible to killing by the complement machinery and presented a remarkable reduced propagation and differentiation rate both in vitro and in vivo . In this report , we have extended these studies to assess , in a mouse model of disease , the virulence , immunogenicity and safety of the mutant as an experimental vaccine . Balb/c mice were inoculated with TcCRT+/− parasites and followed-up during a 6-month period . Mutant parasites were not detected by sensitive techniques , even after mice immune suppression . Total anti-T . cruzi IgG levels were undetectable in TcCRT+/− inoculated mice and the genetic alteration was stable after long-term infection and it did not revert back to wild type form . Most importantly , immunization with TcCRT+/− parasites induces a highly protective response after challenge with a virulent T . cruzi strain , as evidenced by lower parasite density , mortality , spleen index and tissue inflammatory response . TcCRT+/− clones are restricted in two important properties conferred by TcCRT and indirectly by C1q: their ability to evade the host immune response and their virulence . Therefore , deletion of one copy of the TcCRT gene in the attenuated TCC strain generated a safe and irreversibly gene-deleted live attenuated parasite with high immunoprotective properties . Our results also contribute to endorse the important role of TcCRT as a T . cruzi virulence factor . Chagas disease is a neglected tropical ailment caused by the flagellate protozoan Trypanosoma cruzi . It is estimated that 12–20 million people are infected worldwide causing 10–50 , 000 deaths/year [1] . Vector control strategies were not entirely successful mainly due to the inaccessibility and the vast distances that separate endemic areas . Transmission , despite the spraying of insecticides , has been increasing in parts of Argentina , Venezuela and Brazil [2] . In addition , the cases of Chagas disease have raised in many parts of South America and have spread globally because of immigration into non-endemic areas in developed countries [3] , [4] , [5] . Drugs used for treatment have serious adverse effects and do not cure the chronic stage [6] . However , vaccination to protect the 40–100 million individuals at risk of acquiring this serious disease has not been well developed or entered in human trials . Considering T . cruzi complexity , with a genome of more than 12 , 000 genes and four distinct life stages , DNA and peptide vaccination for Chagas disease is insufficient and has , so far , not been reported to induce sterile immunity after challenge [7] . Currently , there is an increased interest in the development of irreversibly gene-deleted live attenuated parasites , as a possible mechanism to reduce the risk of reversion to virulence . There is considerable evidence in genetically modified organisms such as Toxoplasma , Plasmodium and Leishmania , which argues for the usefulness and effectiveness of these parasites as promising immunogens [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] . In T . cruzi , unfortunately , there are so far only five studies of vaccination using genetically attenuated strains [17] , [18] , [19] , [20] , [21] . The advantages of using this kind of immunogens are: ( 1 ) They can provide the full spectrum of relevant native epitopes and immune stimulating molecules , such as Toll-like receptors organized together , which would generate a high immunogenicity , unlike other types of vaccines that offer only a restricted spectrum of immunogens . ( 2 ) They can be manipulated to develop multiple genetic modifications . ( 3 ) They undergo antigen processing and presentation as in the case of virulent infection . ( 4 ) They generate , after inoculation , a strong and long lasting protective response compared with other experimental T . cruzi vaccines [7] . ( 5 ) They can be grown in axenic conditions with a lower economic production cost than other vaccine strategies [22] . The TCC wild type strain does not produce considerable tissue lesions or bloodstream parasite levels detectable by fresh blood mounts in rats [23] . Immunization with TCC provided , after a virulent challenge , a strong immune protection against virulent T . cruzi infections [24] , [25] , also evidenced when the challenge was performed using 17 isolates of T . cruzi obtained in an extensive endemic area of the Province of Salta , Argentina [26] . A strong control of parasitemia and tissue damage was observed in mice challenged a year after immunization [27] , [28] . The protective effect of TCC was extended to field experiments in guinea pigs [25] and dogs [29] . Unfortunately , the TCC attenuation is genetically undefined and the possibility of reversion to the virulent phenotype cannot be excluded . In order to add a safety mechanism to prevent this reversion , in a previous work , we generated and characterized a TCC clonal cell line that lacks a TcCRT allele ( TcCRT+/– ) and another clone overexpressing it ( TcCRT+ ) . TcCRT is a T . cruzi virulence factor , that after being translocated from the endoplasmic reticulum ( ER ) to the area of flagellum emergence , can hijack the complement C1 component , inhibiting the activation of the classical and lectin complement pathways at their earliest stages [30] , [31] , [32] and producing pro-phagocytic signals increasing parasite infectivity [33] . Recently , an important role of TcCRT in the C1-dependent T . cruzi infectivity of human placenta explants has been determined in one of our laboratories , thus providing a plausible mechanism for congenital transmission of this infection [34] . In our previous work , we determined that the TcCRT+/– mutant contained about 6-fold less TcCRT polypeptide than wild type parasites [35] . Moreover , parasites overexpressing TcCRT contained about 2-fold more TcCRT polypeptide than wild type parasites . Consequently , monoallelic mutant parasites were significantly more susceptible to killing by the complement machinery . On the contrary , TcCRT+ parasites showed higher levels of resistance to killing by the classical and lectin but not by the alternative complement activation pathways . The involvement of surface TcCRT in depleting C1 was confirmed through restoration of serum killing activity by addition of exogenous C1 . In axenic cultures , a reduced propagation rate of TcCRT+/– parasites was observed . Moreover , TcCRT+/– parasites presented a reduced rate of differentiation in in vitro and in vivo assays [35] . The previous studies led us to the objective of this report , to detect whether the TcCRT monoallelic deletion caused changes in the infectivity and immunoprotective behavior of the attenuated TCC strain . All animal protocols adhered to the National Institutes of Health ( NIH ) ‘‘Guide for the care and use of laboratory animals’’ and were approved by the Animal Ethics Committee of the School of Health Sciences , National University of Salta ( N° 014-2011 ) [36] . A T . cruzi clone derived from the attenuated TCC strain [37] , designated here as wild type , was used . Also , we used a clonal cell line lacking one TcCRT allele ( TcCRT+/− ) and a recombinant T . cruzi clone that overexpresses the TcCRT polypeptide ( TcCRT+ ) [35] . Epimastigotes were grown at 28°C in liver infusion-tryptose medium ( LIT ) supplemented with 10% fetal bovine serum decomplemented at 56°C for 60 min . , 20 µg hemin ( Sigma , St . Louis , MO , USA ) , 100 IU of penicillin and 100 µg/ml streptomycin . To obtain metacyclic trypomastigotes , epimastigote forms were allowed to differentiate by adding 10% w/v triatomine gut homogenate to the cultures [38] . The percentage of metacyclic forms was recorded daily in a Neubauer chamber . In addition , we used the Tulahuén strain and a highly infective isolate recently characterized [39] . Hemocultures were performed by seeding 200 µl of heparinized blood into 2 ml of LIT under sterile conditions; the cultures were incubated at 28°C and scanned for motile parasites under an inverted microscope on days 15 , 30 , 45 , and 60 . PCR for T . cruzi detection was also performed . Briefly , 700 µl of blood from each inoculated animal was processed . Kinetoplast DNA was amplified using primers 121 ( 5′-AAATAATGTACGGGTGAGATGCATGA-3′ ) and 122 ( 5′-GTTCGATTGGGGTTGGTGTAATATA-3′ ) . Sample storage , DNA extraction , amplification , electrophoresis and staining were performed as previously described [40] . To assess the stability of the mutation , we used TcCRT+/– and TCC wild type parasites recovered from hemocultures performed on nude mice on day 90 post-infection ( p . i . ) after immunosuppression with cyclophosphamide . These parasites were grown and expanded in LIT medium . Genomic DNA was purified using the phenol–chloroform method . Diagnostic PCR analysis confirmed sequences corresponding to TcCRT and HYG gene . Primers used were: Pair 1 , to amplify the entire TcCRT CDS ( 1 . 2 Kb ) , CRT1 ( 5'-GCCAGATATCATGAGGAGAAATGACATAAA-3' ) which anneals into the TcCRT initiation codon and CRT2 ( 5'-TCCTCTCGAGTCAAAACTTTCCCCACCGAA-3' ) , for the stop codon . Pair 2 , to amplify the CDS of HYG gene ( 0 . 96 kb ) , H1 ( 5'-CGTCTGTCGAGAAGTTTCTG-3' ) which anneals into the HYG initiation codon and H2 ( 5'-GAAGTACTCGCCGATAGTG-3' ) for the stop codon . Pair 3 , CRT 7 ( 5'-CCTTCCGATGGCATTAGC-3' ) which anneals upstream of TcCRT gene plus primer H4 ( 5'-CTCGCTCCAGTCAATGACC-3' ) for the HYG sequence ( 1 . 4 kb ) . Pair 4 , CRT93 ( 5'-ATTCCAAACAACATTGCCGT-3' ) which anneals downstream of TcCRT gene plus H6 ( 5'-GGACCGATGGCTGTGTAGAAGTACTCGCCGATAGTGG-3' ) for the HYG sequence ( 1 . 4 kb ) . Total Immunoglobulin G antibodies against T . cruzi were measured by Enzyme-linked Immunosorbent Assay ( ELISA ) using T . cruzi epimastigote homogenates ( 2 µg/well ) as antigens . Dilutions of sera , anti-mouse IgG as a secondary antibody ( Sigma , St . Louis , MO , USA ) and conjugate were 1/100; 1/2 , 500 and 1/16 , 000 respectively . The antibody concentration was expressed as the optical density at 490-nm wavelength . Male Balb/c inbred or athymic nude ( nu/nu ) immunodeficient mice ( about 1 month old ) were inoculated intra-peritoneally ( i . p . ) with 5×105 metacyclic TCC TcCRT+/–; TcCRT+ and wild type trypomastigotes . Balb/c mice were subjected to PCR ( 15 , 30 , 90 , 180 and 220 days p . i . ) , hemoculture ( 15 , 30 , 90 and 220 days p . i . ) and serological determination of antibody levels ( 20 , 47 , 60 , 90 and 165 days p . i . ) as described above . Nude mice were examined by PCR and hemoculture on day 15 , 30 and 90 p . i . To improve the detection of latent infections , the last sample of both , Balb/c and nude mice , were obtained after immunosuppressive treatment with cyclophosphamide . The immunosuppression regimen is based on 5 , 250 mg/kg cyclophosphamide doses administered during 10 days . Samples were collected 10 days after the last dose . To test whether mutant T . cruzi clones induced immunological protection , groups of 6 Balb/c mice , about 1 month old , were inoculated i . p . with 5×105 metacyclic TCC TcCRT+/–; TcCRT+ and wild type trypomastigotes . A control group was inoculated with 100 µl of PBS ( day 0 ) . On day 15 a boost similar to the initial inoculation was administered . On day 30 , antibody levels from immunized mice were determined and , on day 120 , all groups were challenged with 104 blood trypomastigotes of a highly virulent T . cruzi TcVI isolate , recently characterized [39] . Blood was drawn from the tail tip of mice , under slight ether anesthesia using heparinized , calibrated capillary tubes . Ten microliters of blood were placed between slide and cover slip and the number of parasites per 100 fields was recorded microscopically ( 40X ) twice a week . Then , the number of parasites per 100 fields ( parasitemia ) was recorded from fresh blood mounts under microscope ( 40X ) . Finally , on day 60 post-challenge , surviving animals were sacrificed , spleen index and the presence of histological damage was measured in tissue samples . Tissue samples from heart and quadriceps muscle were fixed in 10% formaldehyde and processed using routine histological techniques . Serial histological hematoxylin-eosin-stained sections ( 3–5 µm thick ) were studied . We searched for lymphocytic infiltrates in areas averaging 53 mm2 for heart and 38 mm2 for quadriceps muscle , scanning at least three sections per organ . Quantification of the inflammatory response was scored blindly as severe ( +++: presence of foci containing numerous inflammatory cells covering at least half of the sections surface ) , moderate ( ++: large inflammatory foci covering up to ¼ of the section surface ) , slight ( +: presence of small and isolated inflammatory foci ) or absent ( –: no presence of foci or inflammatory cells ) . Body and spleen weight were determined to calculate the spleen index ( spleen index = spleen weight X 100/body weight ) as an indirect effect of infection severity . The Mann-Whitney U tests and one-way variance analysis ( ANOVA ) of the GraphPad Prism version 5 . 0 software were used . Values are expressed as means ± standard error of mean of at least three separate experiments . P values equal to or minor that 0 . 05 were considered as significant . To determine whether the TCC mutant parasites were capable of infecting and survive for long periods of time in the host , we monitored their in vivo infectivity and persistence . TCC TcCRT+/–; TcCRT+ and wild type epimastigotes were transformed into metacyclic trypomastigotes and inoculated ( 5×105 ) in Balb/c and nude mice . Since the T . cruzi TCC strain is attenuated it is not possible to detect circulating parasites in blood samples by fresh blood mounts . Therefore , infection was detected by more sensitive methods ( hemoculture and PCR ) . No positive hemocultures were obtained from any immunocompetent Balb/c mice inoculated with the three clones at any of the evaluated time points ( Table 1 ) . However , PCR determinations showed different infection patterns . No positive PCR was detected in TcCRT+/– immunocompetent inoculated mice throughout the follow-up , beyond 200 days , and even after immunosuppression . In contrast , positive reactions were found in all animals infected with wild type parasites . After immunosuppression , 2/3 of wild type inoculated mice were positive . All mice infected with TcCRT+ parasites presented a behavior similar to the wild type strain ( Table 1 ) . Thus , the attenuated TCC T . cruzi strain could be rendered even less virulent than wild type via the targeted deletion of one TcCRT allele . Using nude mice we detected an increased rate of infection by PCR and hemoculture in all three experimental groups . No differences were found between the strains at any time p . i . As expected , after immunosuppression we detected a high mortality in all groups ( Table 1 ) . We determined serum antibody levels in BALB/c mice infected with the three parasite clones on acute and chronic stages of infection and disease developmenRTt . TcC+/– infected mice showed undetectable antibody levels ( p = 0 . 01 ) contrasting with mice inoculated with both wild type and TcCRT+ parasites . In fact , the TcCRT+/– values were comparable to those obtained from the PBS-inoculated , negative controls . No differences between TcCRT+ and wild type were found ( p = 0 . 84 ) . As previously described [27] , mice inoculated with the Tulahuén virulent strain ( positive control ) showed about six-fold higher antibody levels than those obtained with any of the TCC strains ( Fig . 1 ) . To exclude the possibilities of cross-contamination , reversion of the genetic mutation or TcCRT locus instability , we determined whether the parasites isolated from hemocultures after long term infection in mice corresponded to mutant parasites . Genomic DNA was extracted from TcCRT+/– and wild type parasites grown on hemocultures at day 90 p . i . ( Table 1 ) . We amplified sequences corresponding to the TcCRT coding sequence ( CDS ) and the hygromycin phosphotransferase ( HYG ) marker gene . The sizes of amplified fragments in the DNA of the recovered parasites corresponded to those predicted for the replacement of TcCRT by the HYG gene ( Fig . 2 ) . Additionally , the antibiotic resistance of TcCRT+/– and wild type parasites were tested . Only TcCRT+/– parasites survived in the presence of 300 µg/ml Hygromicin B mediated by the HYG resistance gene at the deleted TcCRT allele ( data not shown ) . Thus , this evidence showed that TcCRT+/– parasites conserved the targeted allele introduced by homologous recombination , that there is no cross-contamination and that the locus remained stable throughout the infection cycle in the mammalian host . To assess the immunoprotective capacity of mutant parasites against a subsequent reinfection with virulent parasites , groups of six BALB/c mice were inoculated with 5×105 metacyclic trypomastigotes of each of the three clones plus a naive , sham-preinoculated control group . At day 15 a booster similar to the initial inoculation was administered . To determine whether this immunization regimen induces an immune response , blood samples were taken during the immunization phase on day 30 post-priming . After 120 days , these mice together with controls were challenged with 10 , 000 bloodstream trypomastigotes of a virulent T . cruzi isolate [39] . The protective response generated by immunizing with TcCRT+/– and wild type was significantly higher than in the non-immunized group ( p = 0 . 0001 ) . Mice immunized with TcCRT+/– and wild type parasites showed , after challenge , reduced levels of circulating parasites in peripheral blood , ranging between 0–3 parasites per 100 microscopic fields throughout follow-up , demonstrating the protection afforded by immunization . Parasitemia curves between wild type and TcCRT+/– immunized groups are not significantly different ( p = 0 . 22 ) . These results showed that deletion of a TcCRT allele does not modify the protective response induced by TCC wild type parasites . In contrast , mice immunized with TcCRT+ did not afford protection ( Fig 3A ) . Non-immunized control mice presented high parasitemia with peaks between days 13 and 16 , at a time when there was 50% mortality . As expected , these mice showed high parasitemia before death , thus explaining the wide dispersion of the data at that time ( Fig 3A ) . In contrast , in the remaining experimental groups no mortality was recorded . TcCRT+/– immunized and boosted mice showed undetectable specific anti-T . cruzi antibody levels ( similar to those obtained in the non-immunized , negative controls ) compared to the levels found in both wild type and TcCRT+ ( p = 0 . 004 ) and clearly different from those obtained from mice infected with Tulahuén parasites ( Fig 3B ) . Autopsies were performed on mice 4 months after priming with TCC TcCRT+/–; TcCRT+ or wild type trypomastigotes and 2 months after a virulent T . cruzi challenge . Non-immunized , wild type and TcCRT+ mice presented severe inflammatory response throughout the heart tissue , however , this response was extensively reduced in TcCRT+/– immunized mice ( p = 0 . 002 ) ( Fig . 4A ) , thus confirming the protective effect conferred by previous immunization with these parasites . The same effect was observed in muscle tissue: non immunized , wild type and TcCRT+ mice presented moderate to slight cellular damage that was reduced in TcCRT+/– immunized mice ( p = 0 . 0007 ) ( Fig . 4B ) . Splenomegaly is a macroscopic manifestation of the expansion of B- and T-lymphoid cell populations produced by the infection of mice with T . cruzi [41] . Thus , the spleen index represents an indirect effect of infection severity . Spleen index at day 60 post-challenge was significantly decreased in TcCRT+/– and wild type immunized mice compared to that in the non-immunized controls ( p = 0 . 02 for both cases ) . However , TcCRT+ immunized mice presented no differences ( p = 0 . 14 ) with non-immunized controls ( Fig . 4G ) . In a previous work , we have characterized a mutant cell line that lacks a TcCRT allele ( TcCRT+/– ) , with bases on the attenuated TCC T . cruzi strain . We showed that TcCRT+/– epimastigotes contained about 6-fold less TcCRT polypeptide than wild type parasites . Moreover , they were significantly susceptible to killing by the complement machinery and presented a reduced in vitro propagation and differentiation rate . In addition , we generated another clonal cell line that over-expresses TcCRT ( TcCRT+ ) and showed high resistance levels to complement attack [35] . Furthermore , it was not possible to generate biallelic TcCRT–/– null mutant clones , perhaps a reflection of the essential character of the TcCRT protein for parasite survival . TCC wild type infection is hardly detected in immunocompetent animal models due to the attenuation of this strain . The use of highly sensitive methods such as immunosupression regimens followed by PCR and hemoculture is usually required . When inoculated in Balb/c mice , and during a 6-month follow-up period , mutant TcCRT+/– parasites were not detected by either of these techniques , even after immunosupression ( Table 1 ) . TcCRT is highly immunogenic in different animal species [42] . Most humans infected with T . cruzi possess anti-TcCRT antibodies [43] . However , levels of specific antibodies in TcCRT+/– inoculated mice were even more reduced as compared to mice inoculated with wild type or TcCRT+ and , as described [27] , with the highly infective Tulahuén strain ( Fig . 1 ) . The increased virulence attenuation of TcCRT+/– in mice could probably be related to increased complement susceptibility and to the deposition of C1q on the parasite surface , configuring a strategy called "apoptotic mimicry" . In infective trypomastigotes , TcCRT is translocated from the ER to the area of flagellum emergence where it could hijack C1q resulting in an increased affinity for host cells [32] , [33] , [44] . Previous reports affirm that the C1q binding on the T . cruzi trypomastigote surface increases parasite infectivity [45] and thus , any disruption of TcCRT/C1q interaction may result in a reduction of infectivity both , in vitro and in vivo [33] , [34] . Furthermore , apoptotic mammalian cells express surface ligands with high C1q affinities , among them , the calreticulin orthologue . C1q-coating over apoptotic cells produces pro-phagocytic “eat me” signals that promote clearance of apoptotic bodies conducted by phagocytic cells [46] , [47] . One of our laboratories [33] , proposed that T . cruzi expressing TcCRT mimic the “eat me” signals , promoting C1q coating , phagocytic cell chemotaxis and increasing parasite infectivity in the early stages of infection . In our work , the TcCRT allele deletion and synthesis reduction [35] , possibly generated a lower capacity to capture C1 thereby inducing lower pro-phagocytic signals and reduced infectivity of phagocytic cells in the early stages of infection . In a negative feedback , the limited invasion of phagocityc cells would help TcCRT+/– parasites to stay free , for a longer period of time , and exposed to the complement lytic action in the bloodstream system of the host . These properties may have contributed to the important TcCRT+/– infectivity attenuation ( Table 1 ) . In addition , antibodies aggregated to the T . cruzi surface antigens ( including the anti-TcCRT antibodies ) through their Fc regions have a high affinity for C1q [33] . Thus the apparent paradox that C1q-fixing antibodies , rather than preventing parasite replication , contribute to increase their infectivity , is explained . Thus , pretreatment with anti-TcCRT ( Fab' ) 2 fragments ( which lack the Fc fragment of C1q binding ) produces the disruption of TcCRT/C1q with serious negative impact on the in vivo e in vitro infectivity [33] . As expected , the TcCRT+/– attenuated line did not produce detectable specific anti-T . cruzi antibodies ( Fig . 1 ) probably causing a limited C1q deposit on the parasite surface which , in turn , would contribute to diminish phagocytic signals and hence parasite infectivity . In contrast , mice inoculated with TcCRT+ and wild type showed an increased level in antibody titers compared to TcCRT+/– , which would generate a denser C1q coating . This phenomenon may explain the divergences in infectivity of TcCRT+/– , TcCRT+ and wild type parasites . We were unable to recover infecting parasites from immunocompetent mice by hemoculture . However , we could detect parasite DNA by PCR in those mice infected with TCC wild type and TcCRT+ parasites ( Table 1 ) . This is probably a consequence of both a lower density of circulating parasites and a greater PCR sensitivity for detection of T . cruzi in mouse blood ( X 20 ) as compared with hemoculture [27] . In agreement with our hypothesis , mice inoculated with TcCRT+ and wild type parasites infected a high percentage of mice , although without detectable differences between these groups . It is unclear whether TcCRT+/– parasites did infect . However , the possibility that infection occurs is favored by the fact that an adaptive protective status was verified when the animals were challenged 4 months after a primary infection . Since only marginal antibody levels were occasionally detected , protection maybe cellular rather than humoral , issues now under investigation in our laboratories . Using immunedeficient nu/nu mice , infections caused by the three parasite populations could be detected in a high proportion of mice and even in hemocultures ( Table 1 ) . These results confirm previous studies from our laboratory , showing that the TCC wild type strain infects immature or immunocompromised animals [24] , [48] . These results suggest that although TcCRT+/– infectivity is attenuated , the suppression of host immunity allows the replication and persistence of these parasites in animals . A similar behavior was observed in the dhfr-ts ( dihydrofolate reductase-thymidylate synthase ) single mutant , also developed on the TCC T . cruzi strain . This mutant showed a reduced infectivity in immunocompetent mice and as in this work , no mutant parasites could be recovered from hemocultures [20] . The virulence reduction in genetically modified parasites in mice models has previously been reported for genes Tc52 [49] and oligopeptidase B [50] . In our laboratory , this phenomenon was observed working with mutant gp72 genes [19] , cub ( calmodulin-ubiquitin ) [17] , lyt1 [18] and dhfr-ts [20] . We have extensively studied the TCC T . cruzi strain as a live attenuated experimental vaccine [28] , [29] , [51] . The molecular basis of the TCC attenuation is unknown . Thus , we incorporated a rational attenuation mechanism ( targeted gene deletion ) as a safety device to eliminate the possibility of reversion to a virulent phenotype . In this regard , we tried to rule out the possible reversion of the TcCRT+/– genetic modifications during the chronic stage of the disease in mice . We recovered TcCRT+/– parasites from nude mice at day 90 p . i . ( Table 1 ) and detected sequences corresponding to the TcCRT locus engineering . Thus , the TcCRT+/– mutation is genetically stable in chronically infected mice and there is no reversion to the TCC wild type genotype ( Fig . 2 ) . Furthermore , the same experiment ruled out strain cross contamination during handling in the laboratory . Moreover , we tested whether the TcCRT+/– attenuation affects the protective capacity of the TCC wild type parasites against a virulent challenge . Our results suggest that the deletion of one TcCRT allele did not change the already reported immunoprotection induced by TCC wild type parasites [28] . Using TcCRT+/– immunized mice we did not obtain , after a virulent challenge , a sterilizing protective response , although , we achieved low parasite density , mortality ( Fig . 3A ) and a significantly reduced tissue inflammatory response ( Fig . 4A–F ) and spleen index ( Fig . 4G ) . Infection with the parental TCC clone has been shown to be protective , in spite of the fact that it generates inflammatory foci in cardiac tissue [23] . When we inactivated one of the TcCRT alleles , a significant decrease in local inflammation is recorded , perhaps a reflection of an impaired virulence . Certainly , the most important fact was that the protective response was achieved at the cost of a possible primary infection with attenuated TcCRT+/– parasites which could not be detected by our most sensitive methods during a six month follow-up and even after immunosuppression of the infected mice ( Table 1 ) . It is crucial for vaccinating parasites not to persist in the organism and to discontinue the transmission cycle in peridomestic animals from endemic areas . This could impact over the Chagaś disease infection incidence . In a previous work [33] , mice immunization with TcCRT induced the generation of specific anti-TcCRT antibodies resulting in increased parasitemia of the T . cruzi-challenged mice . Most likely , as mentioned above , immunization with TcCRT induces C1q binding anti-TcCRT antibodies thus increasing the parasite infectivity in the challenged animals . According to this hypothesis , TcCRT+ immunized mice showed higher levels of specific anti-T . cruzi antibodies ( Fig 3B ) inducing an elevated parasitemia after challenge ( Fig 3A ) . On the contrary , the TcCRT+/– attenuated line did not produce detectable antibodies ( Fig 3B ) or parasitemia post-challenge ( Fig 3A ) . Wild type TCC parasites are not detectable by direct blood examination , however , they could be detected by PCR in cyclophosphamide treated chronically infected mice ( Table 1 ) or after hemoculture recovery . In this regard , the attenuated biological behavior of the TcCRT+/– mutants is interesting because if employed as live immunogens , an eventual ( natural or induced ) immunosuppression of the host should not produce the reactivation of the vaccinating parasites . Inoculation and eventual infection with TcCRT+/– parasites did not induce detectable antibodies levels ( Fig . 1 and 3B ) . However , protection from T . cruzi infection is considered at present to be mediated primarily by cytotoxic T cells [52] and not by antibodies . In summary , our results show that TcCRT+/– clones were restricted in two important properties conferred by TcCRT and indirectly by C1q: the ability to evade the host immune response , and their virulence status . Therefore , deletion of one copy of the TcCRT gene in the attenuated TCC strain resulted in the generation of a safe and irreversibly gene-deleted live attenuated parasite with high experimental immunoprotective properties .	Trypanosoma cruzi is a protozoan parasite which infects 9 million people in Latin America . Currently there is no vaccine to prevent this disease . Therefore , different approaches or alternatives are urgently needed to identify new protective immunogens . Live vaccines are likely to be most effective in inducing protection; however , safety issues associated with their use have been raised . Hence , we genetically manipulated an attenuated strain of T . cruzi as a safety device to rule out the possibility of reversion to the virulent phenotype . The genetically modified parasites were highly susceptible to killing by the complement machinery and presented a reduced propagation and differentiation rate . We have extended these studies to assess , the virulence , immunogenicity and safety of the mutant as an experimental vaccine . Accordingly , we show that genetically modified parasites present attenuated virulence in mice . The genetic alteration was stable and , after long term infection , it did not revert back to wild type form . Furthermore , after challenge with a virulent T . cruzi strain , mutant immunization induces a highly protective response evidenced by significantly lowered parasite density , mortality , spleen weight index and tissue inflammatory response . Our study provides new insights into the host-pathogen interactions and into the use and evaluation of irreversibly gene-deleted live attenuated parasites to protect against Chagas disease .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "parastic", "protozoans", "gene", "regulation", "genetics", "molecular", "genetics", "protozoology", "biology", "microbiology" ]	2014	A Monoallelic Deletion of the TcCRT Gene Increases the Attenuation of a Cultured Trypanosoma cruzi Strain, Protecting against an In Vivo Virulent Challenge
Gastrointestinal ( GI ) mucosal dysfunction predicts and likely contributes to non-infectious comorbidities and mortality in HIV infection and persists despite antiretroviral therapy . However , the mechanisms underlying this dysfunction remain incompletely understood . Neutrophils are important for containment of pathogens but can also contribute to tissue damage due to their release of reactive oxygen species and other potentially harmful effector molecules . Here we used a flow cytometry approach to investigate increased neutrophil lifespan as a mechanism for GI neutrophil accumulation in chronic , treated HIV infection and a potential role for gastrointestinal dysbiosis . We report that increased neutrophil survival contributes to neutrophil accumulation in colorectal biopsy tissue , thus implicating neutrophil lifespan as a new therapeutic target for mucosal inflammation in HIV infection . Additionally , we characterized the intestinal microbiome of colorectal biopsies using 16S rRNA sequencing . We found that a reduced Lactobacillus: Prevotella ratio associated with neutrophil survival , suggesting that intestinal bacteria may contribute to GI neutrophil accumulation in treated HIV infection . Finally , we provide evidence that Lactobacillus species uniquely decrease neutrophil survival and neutrophil frequency in vitro , which could have important therapeutic implications for reducing neutrophil-driven inflammation in HIV and other chronic inflammatory conditions . Gastrointestinal ( GI ) mucosal damage and immune dysfunction drive chronic inflammation and microbial translocation in HIV infection , which predict and likely contribute to non-infectious comorbidities and mortality[1–5] . Although long-term antiretroviral therapy ( ART ) partially restores mucosal damage , a degree of mucosal immune dysfunction and inflammation persists and is associated with morbidities and mortality[6–8] . Improving the understanding of this persistent mucosal dysfunction and inflammation during ART is a major hurdle for the development of targeted therapies that may promote health and decrease morbidities and mortality in HIV-infected individuals . Neutrophils , the most abundant immune cell , are the first responders to most infections and are crucial in the immune response to bacterial and fungal pathogens[9] . However , the role of neutrophils in HIV infection is not well understood . Imaging studies assessing myeloperoxidase ( MPO ) , an enzyme produced and secreted by neutrophils , suggest that neutrophils accumulate in the GI tract in treated and untreated HIV infection , yet the frequency and functionality of accumulated neutrophils has yet to be examined[10] . A better understanding of the mechanisms involved in neutrophil accumulation in the GI in HIV infection is necessary to develop new strategies to alleviate GI inflammation in HIV infection . While neutrophils are critical in protection from infections , aberrant neutrophil responses can also be harmful . Neutrophil infiltration in the colonic mucosa is one of the distinguishing characteristics of acute inflammation present in inflammatory bowel disease and correlates with disease severity[11–13] . Neutrophil lifespan is tightly regulated in order to limit unintended damage to tissues by secreted reactive oxygen species and granular enzymes meant to degrade extracellular matrix and disrupt tight junctions[14] . Increased neutrophil lifespan is observed under inflammatory conditions and has been attributed to both direct interaction with microbes and the release of cytokines from other immune cells[15] [16 , 17] . In HIV-infected individuals , studies have reported that the delicate balance of healthy bacterial communities is perturbed , resulting in microbial dysbiosis [18–22] . Importantly , dysbiosis remained evident in individuals on ART and associated with disease progression [19] . However , recent studies demonstrated that these previous results were likely confounded by sexual orientation , for which the study designs did not adequately match or control [23 , 24] . Taking this into account , another recent study matched controls based on sexual orientation and found both men who have sex with men ( MSM ) -specific alterations and HIV-specific alterations in MSM and women[25] . Therefore , it is evident that some combination of infection and sexual orientation alters the microbiome in infected individuals . However , to what degree infection itself and lifestyle parameters contribute to microbial alterations in infected individuals requires further study . Additionally , a recent study of experimental dysbiosis induced by antibiotics in rhesus macaques did not lead to increased disease progression in untreated SIV-infected animals , suggesting that dysbiosis may be co-associated with disease progression rather than causative [26] . However , further studies are required in HIV-infected and uninfected human populations to determine the contributions of microbiome dysbiosis in the context of untreated and treated HIV infection . The effects of microbiome alterations on intestinal neutrophils have not been previously assessed , and given that Toll-like receptor ( TLR ) activation and cytokine stimulation regulate neutrophil survival , changes in the microbiome composition could alter neutrophil lifespan . In this study , we hypothesized that HIV-infected individuals would have increased neutrophil frequencies in the lower GI with reduced neutrophil apoptosis . We further hypothesized that different bacterial species would differentially contribute to alterations in neutrophil apoptosis . Previous studies reporting neutrophil infiltration in HIV infection and the nonhuman primate model of SIV have relied solely on MPO staining of neutrophils measured via microscopy[10 , 27] . Because MPO secretion is increased upon neutrophil activation , it is unclear if increased MPO in the tissues represents increased neutrophil frequency or increased neutrophil activation . In addition , in some cases of inflammation , tissue macrophages also produce MPO and stain positive for the enzyme[28] . For these reasons , we developed a multicolor flow cytometry-based approach that distinguishes neutrophils in the context of other leukocytes in order to obtain a more quantitative measure of neutrophil frequencies in the GI during HIV infection . Using this panel , we were able to identify neutrophils in blood ( Supplementary Fig 1A in S1 Supporting Information ) and fresh GI tissue ( Supplementary Fig 1B in S1 Supporting Information ) to assess the frequency of neutrophils as the percentage of total live CD45+ cells . Colorectal biopsies from a total of 40 HIV-infected , ART-suppressed individuals and 35 HIV-uninfected individuals were collected for this study ( Supplementary Fig 2 in S1 Supporting Information ) . Of these colorectal biopsy samples , neutrophil frequencies were assessed by flow cytometry in real-time in 23 HIV-infected , ART-suppressed and 25 HIV-uninfected participants . Table 1 describes relevant participant demographic information including age , sex , sexual orientation , race/ethnicity , CD4+ T cell count , and time since HIV diagnosis for this subset of individuals . We found increased neutrophil frequencies in the GI of HIV+ individuals compared to uninfected controls ( Fig 1A ) . Importantly , this increase remained when assessed using a multivariate analysis adjusting for age , race , sex , and sexual orientation ( Fig 1B ) . This increase in neutrophils is specific to the GI tract , as we observed no increase in neutrophil frequency in the blood of HIV-infected , ART-suppressed individuals ( Fig 1C ) . These data are the first to demonstrate neutrophils are increased in the GI tract relative to other leukocytes in HIV-infected individuals despite suppressive ART . One potential mechanism for elevated frequencies of neutrophils in the colon in HIV is prolonged neutrophil survival . Neutrophil clearance is an important mechanism for tissue homeostasis , and neutrophils are generally short-lived , with findings from studies investigating neutrophil lifespan in vivo ranging from 8 hours to 5 days[29] . The least inflammatory mechanism of neutrophil clearance from tissues is caspase-3 mediated apoptosis followed by engulfment by macrophages[30–32] . Neutrophils undergoing apoptosis demonstrate reduced surface CD16 expression[33] and those with reduced CD16 expression also demonstrate reduced functionality[34] . Therefore , in order to evaluate neutrophil survival , we measured non-apoptotic , functional neutrophils as those expressing high levels of CD16 and low levels of active Caspase-3 in leukocytes isolated from colorectal biopsies ( Fig 2A ) . Indeed , we found that the frequency of these surviving , functional neutrophils was increased in biopsies from HIV+ individuals compared to uninfected controls ( Fig 2B ) , which was significant in a multivariate analysis adjusted for age , race , sex , and sexual orientation ( Fig 2C ) . While neutrophil infiltration in inflammatory bowel disease has been attributed in part to delayed neutrophil apoptosis[35] , these are the first data demonstrating this as a potential mechanism in ongoing intestinal inflammation in HIV infection . Importantly , we observed no differences in total neutrophils or neutrophil lifespan based on sex or sexual orientation in both an unadjusted analysis and after adjustment for HIV status ( Supplementary Fig 3 in S1 Supporting Information ) . Alterations to the intestinal microbiome in HIV-infected individuals are well-documented[36] . Given the extensive evidence that microbes and microbial ligands impact neutrophil survival through both direct interactions and by influencing cytokine release by other immune cells[15 , 16] , we sought to determine whether alterations in microbial composition associated with differential neutrophil frequency , function and survival . To do this , we assessed the mucosal microbiome composition by 16S rRNA gene sequencing of colorectal biopsies collected from all 40 HIV-infected , ART-suppressed individuals and all 35 HIV-uninfected individuals . We focused on bacterial composition at both the family and genus taxonomic levels ( Fig 3A and 3B ) and observed a modest significant association between overall microbial composition at the genus level and HIV status when HIV-infected , ART-suppressed individuals were compared to uninfected controls ( p=0 . 041 as determined by MiRKAT analysis ) . Importantly , this association remained when adjusted for age , race , sex , and sexual orientation , as can be visualized by principle component analyses of the adjusted relative abundances ( p=0 . 035 , Fig 3C ) . Few individual genera significantly associated with HIV status in either the unadjusted ( Supplementary Fig 4A in S1 Supporting Information ) or adjusted analyses ( Supplementary Fig 4B in S1 Supporting Information ) when the false discovery rate was taken into account ( q-value<0 . 05 ) , which may be due to low sample size . Importantly , recent studies have demonstrated that the dysbiosis previously attributed to HIV infection may actually be a result of sexual risk behaviors , as men who have sex with men ( MSM ) had an increased abundance of Prevotella , independent of HIV status [23 , 25] . This highlights the importance of investigating confounding demographic factors in comparisons of HIV infected and uninfected populations . Therefore , we further investigated the microbial composition of the men in this cohort based on sexual orientation at both the family and genus taxonomic levels ( Fig 4A and Fig 4B ) . We observed a significant association between microbial composition at the genus level and sexual orientation as categorized by MSM or non-MSM ( p=0 . 002 , Fig 4C as determined by MiRKAT analysis ) . This association remained when adjusted for age , race , and HIV status ( p=0 . 020 ) . Several alterations in bacterial taxa were associated with sexual orientation in both the adjusted and unadjusted analyses , including a loss of bacteria in the Bacteroides genus and the Barnesiellaceae family and an increase in bacteria of the Streptococcus genus ( Supplementary Fig 5 in S1 Supporting Information ) . The Prevotella genus only significantly associated with sexual orientation prior to adjustment for HIV status . In order to examine the potential impact of bacterial composition on neutrophils , we assessed associations between all taxa found in at least 25% of individuals and colorectal neutrophils . We found no associations between individual taxa and total neutrophils or neutrophil survival in an unadjusted analysis or following adjustment for age , race , sex , sexual orientation , and HIV status once the false discovery rate threshold was applied ( S2 Supporting Data ) . It is possible that our sample size is too small to detect significant associations across all taxa when adjusting for multiple comparisons . Therefore , we next focused on individual bacteria that we hypothesized could impact neutrophil survival , particularly in HIV-infected populations . Specific bacteria altered in certain HIV-infected populations have been shown to impact mucosal immune cells [18 , 37] . Of note , the Prevotella genus had the highest average relative abundance in our study population , and increased Prevotella in one HIV-infected population associated with increased mucosal T cell and dendritic cell activation [18] . In a follow-up study , the authors further reported the ability of Prevotella species to activate leukocytes in vitro by demonstrating that myeloid dendritic cells stimulated with Prevotella stercorea and Prevotella copri produced increased cyctokine[38] . Although the authors did not take into account sexual orientation in these studies , another recent study correlated Prevotella with T cell activation within an MSM population[25] . It remains unclear if the link between Prevotella and immune activation can be applied to other populations of HIV-infected individuals . It has been particularly difficult to assess these associations in non-MSM , HIV-infected men due to the insufficient number of available samples [25] , which unfortunately were also unavailable for our study . However , given in vivo and in vitro evidence that Prevotella species are able to activate leukocytes , we hypothesized that Prevotella could impact neutrophil survival . In vitro assessment using a reporter cell line demonstrated that Prevotella copri activated NF-kb through a TLR-4 dependent mechanism [39] and NF-kb activation is known to drive survival factors in neutrophils[40] . Additionally , LPS has been shown to increase neutrophil survival both through direct TLR-4 activation on neutrophils and by inducing the release of TNF-α and IL-1β from monocytes[15] . Contrarily , reductions in Lactobacillus have been shown to impact gut health and immune function [21 , 41] , and Lactobacillus species have been demonstrated to induce apoptosis in epithelial cells and myeloid cells[42] . Therefore , we hypothesized that alterations in these genera may impact neutrophil survival in the GI of HIV-infected , ART-suppressed individuals , and we sought to assess the relationships between these bacteria and neutrophils . While we observed no significant differences in the relative abundances of bacteria in the Prevotella or Lactobacillus genera ( Fig 5A ) , we observed a significant difference in the ratio of Lactobacillus:Prevotella between HIV-infected , ART-suppressed and uninfected individuals , suggesting an altered balance of these genera ( Fig 5B ) . Additionally , the Lactobacillus:Prevotella ratio remained significantly altered following adjustment for age , race , sex , and sexual orientation ( Fig 5C ) . Importantly , the Lactobacillus:Prevotella ratio correlated with neutrophil survival in leukocytes isolated from colorectal biopsies from the same individual ( Fig 5D ) , suggesting that an alteration in bacterial composition may impact neutrophil survival in vivo . Given the recent interest in balance analyses of microbiome data , we performed a selbal analysis to determine the microbial signature predictive of increased neutrophil lifespan ( Supplementary Fig 6 in S1 Supporting Information ) . The balance we identified as most closely associated with neutrophil lifespan included three genera in the numerator and ten genera in the denominator groups . Interestingly , Prevotella was among the bacteria in the numerator group , which includes bacteria that may be positively contributing to increased neutrophil lifespan . Additionally , Lactobacillus was among the bacteria in the denominator group , or those that may negatively affect neutrophil lifespan . These results suggest that these bacteria may have a role within global microbiome composition changes in altering neutrophil lifespan . However , these analyses should be interpreted with caution given that this method selects bacteria without applying statistical inference . Further study is needed to confirm these relationships in vivo with a larger sample size and in vitro as culturing methods and commercial strains become available . Given our observation that HIV-infected individuals in our study had significantly altered bacterial community composition in vivo and previously published data suggesting that bacterial ligands can impact neutrophil lifespan , we examined the ability of bacterial ligands and various whole bacteria to impact neutrophil survival in an in vitro culture system with whole blood . These experiments were done using whole blood samples from both HIV-infected and uninfected individuals in order to assess the potential impact of confounding HIV infection on the ability of bacteria to alter neutrophils . In accordance with previous studies , we observed a significantly increased frequency of surviving neutrophils in whole blood after a 20-hour incubation with TLR-4 and TLR-2 agonists compared to unstimulated controls ( Supplementary Fig 7A in S1 Supporting Information ) . Isolated neutrophils incubated with TLR-4 and TLR-2 agonists also demonstrated an increased frequency of surviving neutrophils after a 20-hour incubation that did not reach statistical significance , and to a much lesser extent than that observed in whole blood ( Supplementary Fig 7B in S1 Supporting Information ) . These data suggest that both direct interactions with neutrophils and soluble factors released by other leukocytes impact neutrophil survival in the presence of microbial antigens . We next examined the ability of various bacteria previously reported to be altered in HIV-infected individuals [37] to impact neutrophil survival in vitro . All bacterial species significantly increased neutrophil survival after incubation with whole blood , with the exception of Lactobacillus species ( Fig 6A and Fig 6B ) . LPS purified from Escherichia coli ( E . coli ) was used as a positive control due to its ability to significantly impact neutrophil survival in the previous experiment . Importantly , Lactobacillus plantarum and Lactobacillus rhamnosus decreased neutrophil survival . Additionally , we found that both Lactobacillus species decreased total neutrophil frequencies and that the increased neutrophil survival observed upon stimulation by non-Lactobacillus bacteria resulted in sustained neutrophil frequencies ( Fig 6C ) . This suggests that the non-Lactobacillus bacteria shown to reduce active Caspase-3 expression leads to increased neutrophil survival , rather than causing a different form of cell death such as necrosis . In these experiments , the whole blood neutrophils did not respond differently to any stimulation condition based on HIV status . Additionally , given that these were blood neutrophils , we did not expect other demographic factors such as sexual orientation to have an impact and therefore did not collect such data on these individuals . Also , because Ruminococcus bromii and Bacteroides fragilis increased neutrophil survival similarly to the Prevotella species , we assessed the Lactobacillus:Ruminococcus and Lactobacillus:Bacteroides ratios in vivo and found no differences based on HIV status and no association with neutrophil survival ( Supplementary Fig 8 in S1 Supporting Information ) . Given that bacteria in vivo are unlikely to be spatially separated and would therefore interact with cells simultaneously , we assessed the ability of Lactobacillus to reduce neutrophil survival in the presence of LPS in a subset of individuals . We used the same LPS purified from E . coli as was used in previous experiments to test the ability of Lactobacillus to override strong signals of neutrophil survival . We observed that L . plantarum incubated with whole blood in the presence of LPS reduced neutrophil survival similarly to that of L . plantarum alone ( Fig 6D and Fig 6E ) . L . plantarum is commonly used in therapeutic studies with probiotics and often found in available probiotic supplements . It has also been investigated alone and in supplements with other bacteria for its ability to reduce inflammation in SIV and HIV[43 , 44] . Therefore , this observation has important implications for the potential use of Lactobacillus to therapeutically target neutrophil survival , as it suggests that Lactobacillus found in common probiotics could potentially override survival signals induced by other microbes or microbial molecules in the environment . We additionally assessed the effect of different enteric bacteria on the survival of isolated neutrophils and observed much lower survival among isolated neutrophils compared to those in whole blood after 20 hours of incubation in media alone ( Supplementary Fig 9 in S1 Supporting Information ) . No significant effects of bacterial stimulation were observed after correction for multiple comparisons . It is likely that the isolation procedure activated apoptosis pathways that could not be reversed by subsequent stimulation . We therefore cannot conclude to what extent the bacteria may alter homeostatic apoptosis of isolated neutrophils without this confounding activation upon isolation . Further studies will be necessary to better determine the effects of other leukocytes in the alteration of neutrophil survival by different bacteria . Neutrophils have been suggested to contribute to intestinal inflammation in HIV infection , however the causes and consequences of neutrophil accumulation in the intestines during infection are not well understood . Here we demonstrate that neutrophil lifespan is altered in the GI in treated HIV infection and report data suggesting a potential link between the intestinal microbiome and neutrophil accumulation . While numerous studies have linked reduced neutrophil apoptosis to disease severity in IBD , these are the first data suggesting that increased neutrophil lifespan contributes to GI neutrophil accumulation in HIV Infection . Further , we report that mucosal bacteria have differential effects on neutrophil survival in vivo and in vitro , suggesting that an altered microbiome resulting from a combination of HIV infection and lifestyle , such as sexual orientation , may contribute to neutrophil accumulation and inflammation in the GI through effects on neutrophil apoptosis . We assessed the alterations to the microbiome in this cohort by 16S rRNA gene sequencing of colon biopsies , and we report that microbial composition was modestly associated with HIV status and robustly associated with sexual orientation in men . Taken together , these data suggest that both HIV infection and sexual orientation may contribute to observed alterations in colon microbial composition in HIV-infected individuals in this study . Interestingly , we observed some alterations in bacterial abundances based on HIV status in the colorectal biopsies that differ from the results of previously published studies in HIV infection . Specifically , decreased Bradyrhizobium associated with HIV status in our cohort ( although this was not significant using a multiple comparisons approach ) , while a previous study reported increased Bradyrhizobium in the duodenum of HIV-infected individuals[45] . However , this previously reported increase was only observed in individuals with abnormal blood CD4+ T cell counts , suggesting that differences in treatment and disease progression likely contribute to differences in observed microbiome alterations . Additionally , we observed that decreased Peptoniphilus associated significantly with HIV status following adjustment for sexual orientation , while a previous study reported an increase in the rectum in ART-treated , HIV-infected individuals[46] . This could be due to differences between colon and rectal microbial composition as well as differences in cohort demographics . For instance , the authors note that Peptoniphilus is reduced in the penile microbiota following circumcision and is found in vaginal and genitourinary tract infections[47 , 48] , suggesting that cohorts may have different abundances of this bacteria in the lower GI depending on circumcision rates and genital tract microbial composition . Additionally , we observed a positive association between Bilophila and HIV status following adjustment for sexual orientation , while a previous study reported reduced Bilophila in the stool in HIV-infected individuals in China[49] . However , the Chinese cohort examined included individuals with prior antibiotic use as well as untreated individuals and they did not control or match for sexual orientation . These differences highlight the variability in reported microbiome alterations that could result from different GI sample types and assessing cohorts from different geographic areas with different antibiotic usage , treatment status , and disease progression . We further examined Prevotella and Lactobacillus , two genera we hypothesized may impact neutrophil apoptosis based on previously published studies . In this study , although Prevotella was among the top genera that associated with HIV status , this association was not significant by a multiple comparisons approach . Further , after we adjusted for demographics and sexual orientation , Prevotella no longer associated with HIV status , supporting recently published studies linking Prevotella enrichment to MSM sexual orientation rather than HIV status[23] . In accordance with that study , we observed a significant association between Prevotella and MSM in this cohort; however , this association did not remain significant after adjusting for HIV status . Several studies have indicated that Lactobacillus is depleted in HIV-infected individuals[45 , 50] and a higher abundance of gut Lactobacillales was associated with reduced microbial translocation , increased CD4+ T cells in the periphery and gut , and less immune activation in HIV-infected , ART-treated individuals[51] . In this study , Lactobacillus did not specifically associate significantly with HIV status after accounting for multiple comparisons . However , we observed a significant difference between the Lactobacillus:Prevotella ratios in HIV-infected and uninfected individuals , suggesting an altered balance of these bacteria . Additionally , we observed an association between this ratio and neutrophil survival in matched colon biopsies , providing evidence that microbial composition may impact neutrophil survival in vivo . Further , Prevotella demonstrated the ability to increase neutrophil survival in vitro , although this ability was not specific to Prevotella , as the Rumincoccus and Bacteroides species similarly increased neutrophil survival . However , in combination with the in vivo ratio and selbal balance analyses indicating that Prevotella abundance associates with neutrophil survival , these data provide evidence of a role for Prevotella in increased GI neutrophil survival within the context of the greater microbial community . Neutrophil survival and Prevotella alterations coexist in several inflammatory conditions , including bacterial vaginosis[52 , 53] , rheumatoid arthritis[54–56] , and periodontitis[57 , 58] , suggesting that this link warrants further investigation in larger , matched cohorts of HIV-infected individuals . Finally , the selbal balance analysis indicates that other bacteria , including Roseburia and Treponema , may positively associate with neutrophil lifespan in vivo . Future studies should assess the effects of different bacterial balances and combinations on neutrophils to better understand the role of altered bacterial abundances in neutrophil lifespan in inflammatory conditions . Importantly , what is considered to be a healthy microbiome varies geographically , and it has been argued that this is likely due to dietary differences[59 , 60] . We did not assess diet in these individuals , which could have impacted our ability to detect differences in Prevotella and Lactobacillus between the infected and uninfected individuals in this cohort , as both genera have a demonstrated link to diet[61] . Future studies are needed to better understand how such important lifestyle and demographic factors may impact the relationship between the microbiome and neutrophil frequency and lifespan in vivo . Given our observation that HIV-infected individuals in our study population had altered microbial composition , we performed in vitro assessments of the effects of various enteric bacteria on neutrophils . These experiments revealed the unique ability of Lactobacillus species to increase neutrophil apoptosis , which has not been previously reported and has important implications for potential therapeutic intervention in HIV and other diseases of intestinal inflammation . Importantly , the ability of bacteria to alter neutrophil lifespan in vitro was not affected by the HIV status of the individual , suggesting that these interactions could occur in the context of HIV infection but are not specific to HIV-infected individuals . Anti-inflammatory effects of various Lactobacillus species are well described , and have been attributed to several factors: 1 ) the ability of superoxide dismutase secreted by Lactobacillus to neutralize reactive oxygen species; 2 ) the inhibition of the NF-κB pathway leading to a reduction in pro-inflammatory cytokines and chemokines; and 3 ) the expansion of regulatory T cells [62–64] . Indeed , the increase in neutrophil apoptosis we report in the presence of Lactobacillus may be caused by NF-κB inhibition , which is known to drive the production of survival factors in neutrophils and be an important regulator of apoptosis[40] . These data also suggest that the previously reported ability of Lactobacillus to reduce intestinal inflammation in vivo may be a consequence of increased neutrophil apoptosis and reduced neutrophil accumulation[65] [66 , 67] . In HIV infection , increased microbial translocation and dysbiosis may also result in increased neutrophil recruitment in addition to increased neutrophil lifespan . As such , IL-8 , a potent neutrophil chemokine and a regulator of neutrophil survival , is increased in the colorectal mucosa of HIV infected individuals relative to uninfected controls[68 , 69] . Additionally , studies investigating neutrophil lifespan have demonstrated that neutrophil apoptosis is regulated by both cytokines released by monocytes and by direct interaction in response to TLR stimulation , and it is likely that both contribute to alterations in neutrophil apoptosis induced by bacteria[15] . The relative contribution of direct interactions with neutrophils and secreted factors from other leukocytes should be further assessed . Finally , bacteria produce molecules in vivo that could additionally impact neutrophil apoptosis . For example , short chain fatty acids have been demonstrated to inhibit NF-κB activation and attenuate antimicrobial and inflammatory neutrophil responses to LPS[61] . Therefore , the ability of microbial products to alter neutrophil apoptosis and neutrophil-driven inflammation in vivo is an important area of future research . It is important to point out that this study has several limitations . Due to the requirement that samples be processed and analyzed fresh to assess neutrophils , there were geographic and time constraints that inhibited our ability to recruit more individuals for this study or narrow the focus of our recruitment . As such , we have a relatively small sample size and were unable to recruit controls specifically matched for sexual orientation , age , race , and other demographic characteristics . While we applied the appropriate statistical corrections to account for differences in these characteristics between the HIV-infected and control groups , additional studies are necessary to further assess the relationship between neutrophils and the microbiome in HIV infection in vivo , particularly in the context of sexual orientation . Likely due to the relatively small sample size , we were unable to provide evidence that microbiome alterations associate with GI neutrophils within HIV-infected individuals only , and our in vivo analyses associating microbiome alterations and mucosal neutrophil lifespan included both HIV-infected and HIV-uninfected individuals given the skewed distribution of the variables within each population . A larger cohort with a wider distribution of data is necessary in order to fully assess the contribution of microbial changes to neutrophil alterations in the context of HIV infection in vivo . Additionally , we were unable to assess neutrophils or the neutrophil/microbiome relationship in untreated individuals because most individuals that are aware of their status in the geographic areas of this study are on treatment . However , we believe that these data are clinically relevant given that individuals are now treated immediately upon diagnosis . Future studies in areas where there are more untreated individuals could be beneficial to further assess neutrophils in HIV infection and their role in GI inflammation in HIV . Neutrophil apoptosis has emerged as a therapeutic target for the resolution of acute and chronic inflammation in the lungs , the intestines , and arthritis but no strategies are approved for use in humans to-date[70] . Here , we provide evidence that this may also be a therapeutic target to reduce intestinal inflammation in HIV-infected individuals by demonstrating increased GI neutrophil survival in HIV infection . Further , the ability of Lactobacillus to uniquely reduce neutrophil survival and neutrophil frequency suggests that ongoing studies investigating Lactobacillus-containing probiotics should additionally assess neutrophil accumulation as a potential mechanism for any observed alterations in intestinal inflammation . Finally , these data lead to new avenues of research whereby commensal bacteria , their surface antigens , and their products should be further assessed for their therapeutic ability to reduce neutrophil accumulation and tissue damage in HIV infection and other inflammatory conditions . HIV+ and HIV- study participants were recruited through either the University of Washington Center for AIDS Research , University of California San Francisco SCOPE cohort , Northwestern University , or the University of Washington AIDS Clinical Trials Unit . Biopsies were obtained by either colonoscopy or rectosigmoidoscopy . Blood samples for bacterial stimulations were collected in EDTA from HIV+ and HIV- participants recruited through the University of Washington Center for AIDS research . All HIV+ participants were on potent combination antiretroviral therapy at time of biopsy or blood draw with no detectable plasma viral load . The appropriate Institutional Review Boards approved all protocols and informed written consent was obtained from all participants . Table 1 describes relevant participant demographics information including age , sex , sexual orientation , ethnicity , CD4+ T cell count , and time since HIV diagnosis for the individuals that had biopsies that could be processed , stained in real-time , and met the threshold criteria ( described below ) for accurate neutrophil measurements by flow cytometry . Demographic characteristics for additional biopsy samples that did not meet these criteria but were able to be assessed by 16S sequencing are included in Supplementary Table 1 in S1 Supporting Information . Blood samples from additional donors were obtained from the University of Washington AIDS Clinical Trials Unit for in vitro experiments ( as depicted in Fig 6 and Supplementary Figs 7 and 9 in S1 Supporting Information ) , and HIV status was the only demographic data made available for those individuals . The appropriate Institutional Review Boards approved all protocols and informed written consent was obtained from all participants . Approval numbers at each institution are as follows: Northwestern University: STU00200953: Rectal Biopsies; University of Washington/Harborview Medical Center: STUDY00002763; University of California , San Francisco: 10-01218 . All participants were adults >18 years of age . Gut biopsies were enzymatically digested with media ( RPMI 1640 with 2 . 05mM L-glutamate , 100U/ml Penicillin , 100μg/ml Streptomycin [all from GE Healthcare , Logan , UT] ) supplemented with Liberase ( 40 μg/ml , Sigma-Aldrich , St . Louis , MO ) and DNAse ( 4 μg/ml , Sigma-Aldrich ) for 1 hour at 37°C with vigorous stirring , ground through a 70-μm cell strainer into a single cell suspension , and then analyzed by flow cytometry . Neutrophils were isolated by lysing the red blood cells in whole blood with ACK lysing buffer ( ThermoFisher Scientific ) and then labeling the leukocytes with CD15 microbeads ( Miltenyi Biotec , Bergisch Gladbach , Germany ) . The labeled cells were then loaded onto a MACS Column , which was placed into a magnetic MACS Separator ( both from Miltenyi Biotec ) . The retained CD15+ cells were then washed with ice-cold buffer ( PBS with 0 . 5% BSA and 2 mM EDTA ) . The column was then removed from the separator and the labeled cells were eluted in ice-cold buffer . Cells were counted and used immediately in bacterial or TLR stimulation experiments . Single cell isolations from biopsies were analyzed by flow cytometry immediately after isolation . Biopsies and blood or isolated neutrophils from the bacteria stimulations were stained using the following surface antigen mouse anti-human antibodies with clone denoted in ( ) , from Becton Dickinson , and Co . ( BD ) Biosciences ( Franklin , NJ ) unless otherwise stated: CD45 PE-CF594 ( HI30 ) , CD11b APC-Cy7 ( ICRF44 ) , CD66b PE ( Biolegend , G10F5 ) , CD49d PE/Cy5 ( Biolegend , 9F10 ) , CD20 Brilliant Violet 570 ( Biolegend , 2H7 ) , CD3 Brilliant Violet 570 ( Biolegend , UCHT1 ) , CD16 BV605 ( 3G8 ) , CD15 BV650 ( HI98 ) , and CD14 BV786 ( M5E2 ) . Following surface staining , cells were permeabilized using Cytofix/Cytoperm ( BD Biosciences ) . Intracellular active Caspase-3 was stained using a v450-conjugated rabbit anti-human antibody ( BD Biosciences , C92-605 ) . Stained samples were fixed in 1% paraformaldehyde and collected on an LSR II ( BD Biosciences , La Jolla , California ) . Analysis was performed in FlowJo ( version 9 . 7 . 6 , Treestar Inc . , Ashland , Oregon ) . Samples with less than 100 events in the neutrophil gate were not included in analyses due to an inability to ensure adequate fluorescence separation of populations and therefore accurate gating of the neutrophil cell population . Genomic DNA was extracted from colon tissue biopsies using the QIAamp PowerFecal DNA Kit ( QIAGEN , Valencia , CA ) . DNA for 16S rRNA sequencing was processed following the Earth Microbiome Project protocols ( http://press . igsb . anl . gov/earthmicrobiome/protocols-and-standards/16s/ ) with the following modifications . During the library preparation , each DNA sample was amplified in triplicate using the FailSafe PCR System ( Epicentre , WI ) and the 515FB-806RB primer pair to generate a 400 bp amplicon from the V4 variable regions of the 16S rRNA gene . The triplicates reactions were pooled , quantified using Qubit dsDNA High Sensitivity Assay Kit ( ThermoFisher Scientific , Waltham , MA ) , and visualized using a LabChip GX ( PerkinElmer , MA ) . Using the concentration of the 400 bp peak , 0 . 4 ng of each library was pooled into a single sample . The ~400 bp amplicon from the pooled sample was isolated using a BluePippen System ( Sage Science , MA ) , cleaned using AMPure XP Beads ( Beckman Coulter , IN ) and quantified using the KAPA Library Quantification Kit ( KAPA Biosystems , MA ) . Sequencing was carried out as detailed in the EMP protocol; specifically , 7 pM of the pooled library with 30% PhiX phage as a control was sequenced using a 300-cycle Illumina MiSeq Kit . 16S rRNA gene sequence data was analyzed using the QIIME 2 software package[71] . Sequences were classified using the Naïve Bayes classifier trained on Greengenes 13_8 and binned into operational taxonomic units ( OTUs ) at 99% sequence similarity[72] . OTUs were then classified using Greengenes 13_8 and converted to relative abundance at the family and genus taxonomic levels for visualization and statistical analyses . Taxonomy plots were created in RStudio Version 1 . 1 . 422 using the phyloseq package[73] . Prevotella stercorea ( DSMZ #18206 , Braunschweig Germany ) , Prevotella copri ( DSMZ #18205 ) , Bacteroides fragilis ( ATCC #25285 , Manassas , Virginia ) , and Ruminicoccus bromii ( ATCC #27255 ) were all grown in anaerobically in chopped meat broth ( Hardy Diagnostics , Santa Maria , CA ) supplemented with 1% trace minerals ( ATCC ) , 1% vitamin supplements ( ATCC ) , 0 . 05% Tween 89 ( Sigma-Aldrich , Saint Louis , MO ) , 29 . 7 mM acetic acid ( Sigma-Aldrich ) , 8 . 1 mM proprionic acid ( Sigma-Aldrich ) , and 4 . 4 mM butyric acid ( Sigma-Aldrich ) . Acinetobacter junii ( ATCC #17908 ) was grown aerobically in nutrient agar ( BD ) . Lactobacillus plantarum ( ATCC #14917 ) and Lactobacillus rhamnosus ( ATCC #53103 ) were grown aerobically in MRS broth ( BD ) . All bacteria were counted using CountBright Absolute Counting Beads ( Thermo Fisher Scientific , Waltham , MA ) and Syto 9 dye ( Thermo Fisher Scientific ) on the LSR II . Bacteria were frozen as dry cell pellets until reconstituted in PBS for use in stimulations . For stimulations , bacteria were added at 2 . 5 bacteria per leukocyte to 100 μl of whole blood or 500 , 000 isolated neutrophils in 1 ml R10 media ( RPMI 1640 with 2 . 05mM L-glutamate , 100U/ml Penicillin , 100μg/ml Streptomycin , and 10% fetal bovine serum [all from GE Healthcare , Logan , UT] ) and incubated aerobically for 20 hours at 37°C . Following the incubation , the supernatant was removed and the blood and isolated neutrophils were washed before being analyzed by flow cytometry . Differences in neutrophil frequencies and active Caspase-3 expression between infected and uninfected individuals were determined by Mann-Whitney test . In an effort to control for imbalances in potential confounders they were included as covariates in subsequent multivariate models . Adjusted multivariate analyses were conducted by regressing log-transformed neutrophil frequencies and active Caspase-3 expression on infection status , age , race ( white vs . non-white ) , sex and sexual orientation . Note that sex and sexual orientation were treated as a single trinary variable , coded using two dummy variables , with female sex as the referent category . A paired one-way ANOVA was used to assess differences between groups stimulated with different bacteria or TLR agonists in the in vitro experiments followed by a Dunnett’s or Tukey’s post-hoc analysis for multiple comparisons . p-values reported are adjusted p-values from the post-hoc analysis . Correlations were assessed using the Spearman’s rank correlation analysis . Taxon counts at the family or genus level were center-log ratio ( CLR ) [74] transformed to accommodate compositionality and encourage normality . Taxa present in fewer than 25% of samples were omitted . Community level ( beta-diversity ) analyses were conducted using MiRKAT[75] , which is a generalization of PERMANOVA[76] , under Euclidean distance . For the adjusted analyses of associations with total neutrophil and neutrophil survival we included age , race , sex/sexual preference , and HIV status as covariates . For the adjusted analyses of associations with HIV status , we included age , race and sex/sexual preference as covariates . For the adjusted analyses based on MSM status we included age , race , and HIV status . Associations with individual taxa were determined by regressing the abundance of each taxon on the variable of interest and additional covariates as described above followed by false discovery rate[77] control for multiple testing . In addition to community level and individual taxon level analyses , we used the previously published selbal approach [78] to explore whether balances of particular bacterial taxa could predict neutrophil survival . We considered the percentage of active Caspase-3 low , CD16 high neutrophils as the outcome and used the selbal procedure as described to search for a global balance of taxa related to the outcome . No covariates were included .	HIV infection results in chronic immune activation that leads to increased risk of other diseases and premature death , and this has been linked to gastrointestinal tract ( GI ) damage in infected individuals . In this study , we investigated neutrophils , a cell involved in the immune response to pathogens , in colorectal tissue of HIV-infected individuals receiving treatment . Because neutrophils use methods to contain pathogens that can also damage tissue and have been implicated in tissue damage in other GI diseases , it has been proposed that they contribute to GI damage in HIV infection . However , the role of neutrophils in GI damage in HIV has not been well studied . This study quantifies neutrophils in relation to other white blood cells in the GI tissues in HIV infection and demonstrates that they are increased in the GI in infected individuals . Additionally , we present evidence that neutrophils in the GI in HIV-infected individuals have a longer lifespan , which represents one potential reason for their increased frequency . Finally , we present data that different bacteria that naturally reside in the GI can alter neutrophil lifespan and that changes in the relative abundances of these bacteria in HIV infection may be contributing to increased neutrophil lifespan .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "blood", "cells", "urology", "hiv", "infections", "medicine", "and", "health", "sciences", "immune", "cells", "gut", "bacteria", "pathology", "and", "laboratory", "medicine", "biopsy", "microbiome", "pathogens", "immunology", "microbiology", "surgical", "and", "invasive", "medical", "procedures", "retroviruses", "viruses", "immunodeficiency", "viruses", "rna", "viruses", "prevotella", "bacteria", "neutrophils", "microbial", "genomics", "infectious", "diseases", "white", "blood", "cells", "animal", "cells", "medical", "microbiology", "hiv", "microbial", "pathogens", "lactobacillus", "cell", "biology", "viral", "pathogens", "genitourinary", "infections", "genetics", "biology", "and", "life", "sciences", "cellular", "types", "viral", "diseases", "genomics", "lentivirus", "organisms" ]	2019	Increased mucosal neutrophil survival is associated with altered microbiota in HIV infection
Terminating protein translation accurately and efficiently is critical for both protein fidelity and ribosome recycling for continued translation . The three bacterial release factors ( RFs ) play key roles: RF1 and 2 recognize stop codons and terminate translation; and RF3 promotes disassociation of bound release factors . Probing release factors mutations with reporter constructs containing programmed frameshifting sequences or premature stop codons had revealed a propensity for readthrough or frameshifting at these specific sites , but their effects on translation genome-wide have not been examined . We performed ribosome profiling on a set of isogenic strains with well-characterized release factor mutations to determine how they alter translation globally . Consistent with their known defects , strains with increasingly severe release factor defects exhibit increasingly severe accumulation of ribosomes over stop codons , indicative of an increased duration of the termination/release phase of translation . Release factor mutant strains also exhibit increased occupancy in the region following the stop codon at a significant number of genes . Our global analysis revealed that , as expected , translation termination is generally efficient and accurate , but that at a significant number of genes ( ≥ 50 ) the ribosome signature after the stop codon is suggestive of translation past the stop codon . Even native E . coli K-12 exhibits the ribosome signature suggestive of protein extension , especially at UGA codons , which rely exclusively on the reduced function RF2 variant of the K-12 strain for termination . Deletion of RF3 increases the severity of the defect . We unambiguously demonstrate readthrough and frameshifting protein extensions and their further accumulation in mutant strains for a few select cases . In addition to enhancing recoding , ribosome accumulation over stop codons disrupts attenuation control of biosynthetic operons , and may alter expression of some overlapping genes . Together , these functional alterations may either augment the protein repertoire or produce deleterious proteins . Ribosomes translate the genetic information in the mRNA to a linear sequence of amino acids in the polypeptide chain through a process consisting of initiation , elongation , termination and recycling . During initiation , the 30S subunit of the bacterial ribosome and various initiation factors assemble at the initiation codon on the mRNA . Elongation commences after the 50S subunit of the ribosome joins the complex . Cognate aminoacyl tRNAs , together with elongation factors decode the mRNA sequentially , binding first at the acceptor site ( A site ) , followed by movement to the P site after amino acid transfer to the polypeptide chain at the peptidyl-transferase center . Several layers of error correction minimize the misincorporation of non-cognate amino acids [1 , 2] . Termination is signaled when a stop codon ( UAA , UAG , UGA ) enters the A site of the ribosome , where it is recognized either by release factor ( RF ) 1 or 2 [3] . RF1 or RF2 hydrolyze the polypeptide chain to terminate translation , and are then dissociated from the ribosome by RF3 during recycling [4] . Peptide release is a high fidelity process ( error frequency of approximately 10−5 ) , ensuring that stop codon recognition precedes peptide release [5–7] . Finally , the ribosome is dissociated to its 30S and 50S subunits by elongation factor EF-G , and ribosome recycling factor ( RRF ) [8–10] . There is increasing structural and biochemical understanding of the three bacterial release factors . RF1 and RF2 are structural mimics of an aminoacyl tRNA and both are essential in native E . coli K-12 [11 , 12] . They bind in the A site using conserved protein motifs in Domain 2 to recognize the 2nd and 3rd positions of the stop codons ( RF1: UAA and UAG; RF2: UAA and UGA ) [13] . Their universally conserved GGQ amino acid motif then reaches into the peptidyl transferase center to release the peptide chain by catalyzing its hydrolysis from the tRNA [12 , 13] . Methylation of RF1 and RF2 by PrmC at their GGQ motif enhances release factor activity [14] . Interestingly , E . coli K-12 strains have an RF2 variant with Thr at position 246 rather than the canonical Ala246 or Ser246 [15] . All other bacteria , including other E . coli lineages , have Ala or Ser at position 246 [16] . In this work , we call the E . coli K-12 RF2 variant , RF2K-12 , and the E . coli B variant with Ala246 , RF2B . RF2K-12 is discrepant from RF2B in its properties . First , RF2K-12 has reduced ability to catalyze hydrolysis of the peptide bond to terminate translation relative to RF2B [15] . Second , RF2K-12 but not RF2B is almost completely dependent on methylation for activity [14] . Third , UAA decoding is done primarily by RF1 in strains with RF2K-12 , but by both RF1 and RF2 in strains with RF2B [14 , 15] . Indeed , RF1 is non-essential in an E . coli K-12 strain with the RF2B variant [17 , 18] . Finally , because the level of RF2 is tuned to need via an internal UGA frameshifting event necessary to produce the full-length protein , RF2K-12 is present at a higher level than RF2B as expected from its lower activity [14 , 19 , 20] . RF3 , a non-essential release factor of E . coli , promotes dissociation of RF1 and RF2 from the ribosome [21 , 22] . This reaction occurs slowly in the absence of RF3 [4 , 23] . RF3 is a homologue of EF-G , a GTPase translocation factor that catalyzes movement of tRNA on the ribosome and ribosome dissociation [24] . The current idea is that RF3-GTP binds to the ribosome in the same location as EF-G , and similarly induces inter-subunit rotation , which creates a steric clash with the bound RFs , promoting dissociation of RF1 and RF2 [25 , 26] . The role of the release factors in translation termination fidelity has been explored in vivo by measuring by release of small artificial peptides or in vitro by measuring frameshifting or stop codon readthrough in a synthetic constructs containing a known frameshifting site or premature stop codon [16 , 27–32] . These assays showed that RF2B and methylated RF1 and RF2 terminated translation better than RF2K-12 , or unmethylated RF1 and RF2 [9 , 14 , 16 , 33–35] . Additionally , temperature sensitive mutations within RF1 and RF2 increase stop codon readthrough on in vitro constructs [29 , 31 , 32 , 36 , 37] . Suppressors of these temperature sensitive mutations mapped to the prfC ( RF3 ) locus [38] . Further studies indicate that strains lacking RF3 have increased stop codon readthrough of lacZ reporter constructs and increased frameshifting over the known prfB ( RF2 ) frameshifting sequence [9 , 33–35 , 39] . Overexpression of RF3 was shown to decrease frameshifting over the prfB frameshifting sequence [34 , 39] . These studies and RF3 overexpression studies indicated that cooperative interactions between RF1/2 and RF3 improved termination efficiency [38] . In vitro studies also implicate RF2 and RF3 in post-peptidyl transfer quality control ( post PT-QC ) , a mechanism for selectively terminating translation of polypeptides that have misincorporated amino acids , and phenotypes suggestive of post PT-QC were found in vitro and in vivo [34 , 40] , but were not reproduced in K-12 strains with RF2B [35] . Taken together , the work thus far indicates how release factor mutations alter translation termination at specific reporter constructs or known frameshifting sites . However , the effects of these mutations on translation termination have not been studied on a global scale or at physiologically relevant native gene loci . To elucidate the global effects of these mutations and observe how they perturb the translatome , we used ribosome profiling to examine the behavior of ribosomes at stop codons , and compare the extent of recoding and readthrough events genome-wide in native E . coli K-12 , with that in K-12 RF2B cells and in both strains lacking RF3 . We also examined changes in protein expression globally among the strains . We find significant differences in the duration of termination/release dependent upon strain background , identify new recoding events and reveal the impact of altered termination on genes whose expression is regulated by transcription-translation coupling . We examined the effect of altered release factors on the growth of E . coli MG1655 , the prototypical K-12 strain used in our studies . Previous results indicated inconsistent phenotypes for MG1655 and BW25113 , another K-12 strain [34 , 35 , 41] . Here , we determine the growth rates of MG1655 ( K-12 RF2K-12 ) and isogenic single and double mutant release factor derivatives ( Table 1 ) . At 37°C in MOPS-complete glucose medium supplemented with all amino acids , K-12 RF2K-12 ΔRF3 has a significantly slower growth rate than K-12 RF2K-12 ( doubling times of 36 min and 28 min , respectively ) , but the slow growth phenotype of ΔRF3 is rescued by RF2B ( Fig 1A ) . This suggests that the enhanced activity of RF2B compensates for the ΔRF3 defects that result in reduced growth rate . K-12 RF2B and K-12 RF2K-12 have indistinguishable growth rates . Our previous global phenotyping screen of the E . coli BW25113 single gene deletion library indicated that ΔRF3 was quite cold-sensitive at 16°C and 20°C [42] . ΔRF3 strains in other backgrounds , both K-12 and non K-12 derived , also grow slowly at 25°C , a phenotype that is reversed by RF2BΔRF3 [35] . We therefore characterized the growth rate of our isogenic E . coli K-12 MG1655 strains across an expanded temperature range . Consistent with previous results [35] , K-12 RF2K-12 ΔRF3 exhibits severe cold sensitivity at 20°C , which was reversed in the K-12 RF2B ΔRF3 strain ( Fig 1B ) . Surprisingly , however , at 15°C , the RF2B variant did not rescue ΔRF3 ( Fig 1B ) . These results motivated us to examine strain viability at each temperature ( Fig 1C ) . We find that although it is slow growing , K-12 RF2K-12 ΔRF3 maintains viability at all temperatures tested . In contrast , K-12 RF2B ΔRF3 exhibited a near 5-fold decrease in viability at 15°C . Thus , at very low temperature the RF2B variant is more deleterious than RF2K-12 variant when paired with RF3 deletion ( see Discussion ) . We performed ribosome profiling for K-12 RF2K-12 and K-12 RF2B strains , with and without RF3 . All ribosome profiling experiments were performed at 37°C in MOPS- complete glucose medium because it contains a balanced complement of amino acids ( see Methods ) . As a proxy for duration of ribosome termination and release , we quantified the extent of ribosome occupancy at stop codons . The density of footprints at any codon is related to the dwell time of the ribosome at that position [43] . Thus , higher stop codon occupancy is indicative of increased duration of termination/release . We note that this metric is a composite measurement that minimally consists of the rates of: release factor binding , polypeptide chain termination , release factor release , and ribosome recycling . Our initial analysis used a dataset of approximately 1200 well-expressed genes aligned at the stop codons of their open reading frames ( ORFs ) , and is comprised of 947 UAA stop codons , recognized by both RF1 and RF2 , and 231 UGA codons recognized only by RF2 ( Fig 2 , see Methods ) . For additional analyses , we expanded the dataset to 3390 genes so that we could also study the lower abundant UAG stop codons , as well as all of the 4-base stop codons ( S1 Fig , see Methods ) . We used our 1200 gene dataset to compare global stop-codon occupancy by native K-12 RF2K-12 and K-12 RF2K-12 ΔRF3 cells , as well as to compare K-12 RF2B cells and K-12 RF2B ΔRF3 cells . We found that K-12 RF2K-12 ΔRF3 cells have a global 2 . 5-fold increase in ribosome occupancy at stop codons relative to native K-12 RF2K-12 , indicating a significant defect in termination/release ( Fig 2A ) . K-12 RF2B ΔRF3 cells also have a global increase in ribosome occupancy at stop codons relative to K-12 RF2B ( Fig 2B ) , but the magnitude of this increase is less ( 1 . 4-fold ) . It is likely that the more active RF2B variant limits the effect of ΔRF3 . Changes in occupancy are also evident in ribosome footprint density plots of single genes ( S2 Fig ) . Because global stop codon occupancy could be driven by the behavior at the highly abundant UAA stop codons , we next compared the behavior at each of the three stop codons . We first compared K-12 RF2K-12ΔRF3 and K-12 RF2B ΔRF3 at all three stop codons . Changes in occupancy of the ΔRF3 strains at all three stop codons , including the low abundance UAG stop codon recognized only by RF1 , mirrors their change in global occupancy ( Fig 2C and 2D and S1B Fig ) . Occupancies over UAG are very similar to those over UAA , although the low number of UAG stops increases the noise in the data ( S1B Fig ) . The defects of translation termination/release seen at all stop codons for strains lacking RF3 is consistent with previous studies showing that RF3 is important for release of both RF1 and RF2 [22] . Notably , relative to K-12 RF2K-12 , the K-12 RF2K-12ΔRF3 strain has a very slight increase in ribosome occupancy ~20–30 bp upstream of the stop , at the position expected for a second ribosome ( Fig 2A ) . The height of this shadow peak increases in highly translated genes ( S3 Fig ) , suggestive of ribosome pileup . However , further experiments are necessary to definitively establish this point . In summary , deleting RF3 results in a general defect in termination/release at all three stop codons , but the magnitude of this effect is much smaller in K-12 RF2B than in the native K-12 RF2K-12 strain . We next compared the stop codon occupancy of K-12 RF2K-12 with that of K-12 RF2B both globally ( Fig 2A and 2B ) and at specific stop codons ( Fig 2C and 2D , S1B Fig ) . Both strains have similar stop codon occupancy globally , as well as at the UAA and UAG stop codons ( Fig 2A–2C , S1B Fig ) . However , for UGA codons , recognized solely by RF2 , there is a general trend towards decreased occupancy in K-12 RF2B versus K-12 RF2K-12 strain , consistent with observations that the RF2B protein is more efficient at mediating termination at UGA than RF2K-12 variant ( Fig 2C and 2D , S1B Fig ) [15 , 16] . The base following the stop codon , called the fourth base , is known to impact the efficiency of stop codon termination and is recognized by both RF1 and RF2 [39 , 44 , 45] . We created a metagene plot for each 4-base stop codon to examine the occupancy preferences of each of our strains . While these plots generally have more noise than the 3-base stop plots due to a smaller dataset for each 4-base stop codon , particularly on the lower abundant UAG and UGA stops , we can observe general trends of occupancy within each specific stop codon group ( S1 Fig ) . Despite noise , behavior of strains at the 4-base codons generally mirrored their behavior at the 3-base codons , with some additional patterns . The K-12 RF2B strain has the most severe decrease in stop codon occupancy relative to K-12 RF2K-12 at nearly all 4-base UGA stop codons ( S1 Fig ) . Strikingly , at the UGAA stop , K-12 RF2B ΔRF3 has ribosome occupancy equivalent to that of native K-12 RF2K-12 , rather than the slightly elevated level characteristic at other stops . Together , these results suggest that RF2B variant is significantly more effective than the RF2K-12 protein at UGAA stops ( see Discussion ) . We also used the ribosome profiling data to examine release factor expression in each strain . Previous experiments using reporter constructs indicated that the rate of frameshifting at the internal prfB UGA stop codon used to change expression of RF2 , increased in a ΔRF3 strain [34 , 39] . We show that this result also is true at the endogenous prfB locus by calculating the ratio of ribosome occupancy of the larger second frameshifted ORF ( encoding full length RF2 ) relative to that of the smaller first ORF . Native K-12 RF2K-12 cells exhibit ~28% frameshifting at this locus and this value increased to ~47% in the K-12 RF2K-12 ΔRF3 strain ( Fig 3 ) . K-12 RF2B cells appear to have slightly less frameshifting ( 21% ) than K-12 RF2K-12 cells . K-12 RF2BΔRF3 cells do not exhibit the increased frameshifting as seen in RF2K-12ΔRF3 cells , but instead exhibit a similar amount of frameshifting as K-12 RF2K-12 . Concomitantly , we measured the overall expression levels of RF2 , finding general agreement with our calculated rate of frameshifting . K-12 RF2K-12 cells have a higher level of RF2 than K-12 RF2B cells ( Table 2 ) , consistent with the previously reported ~40% reduction of RF2 in K-12 RF2B calculated by quantitative western blot [14] . Additionally , as expected from increased frameshifting , K-12 RF2K-12 ΔRF3 cells have increased expression of RF2 relative to K-12 RF2K-12 cells ( Table 2 ) . K-12 RF2K-12 ΔRF3 cells also have increased expression of RF1 relative to K-12 RF2K-12 cells . The fact that K-12 RF2K-12 ΔRF3 cells exhibit severe termination/release defects despite increased release factor expression suggests that increased expression only partially mitigates this phenotype . Neither RF2 nor RF1 expression increases in K-12 RF2BΔRF3 relative to K-12 RF2B , presumably because the higher activity of the RF2B protein relative to RF2K12 prevents the signals triggering enhanced release factor expression . Extended stop codon occupancy could have the downstream consequence of facilitating either stop codon readthrough or of frameshifting , collectively called here recoding events . Our finding of increased programmed frameshifting at the prfB locus in certain strain backgrounds is consistent with that idea . Recoding events would result in an increase in ribosome density after the annotated stop codon . We therefore quantified the average ribosome density downstream of the open reading frame ( ORF ) relative to the average density within the ORF itself . We call this metric relative post-ORF ribosome occupancy ( RPOR ) . We analyzed RPOR for the almost 1600 genes that are well separated from their immediate downstream genes , i . e . having ≥ 65 nucleotides between the stop codon of the upstream gene and the start codon of the downstream gene ( Fig 4A ) . The distance constraint is necessary to enable us to unambiguously examine ribosomes past the stop codon of the gene without interference from translation of the downstream gene . A plot of the cumulative distribution of RPOR values indicates that most genes have very low RPOR values across all strains . Indeed , nearly 60% of genes had an RPOR value under 0 . 1 , indicative of very few ribosomes in the post-ORF region ( Fig 4B ) . Thus , as expected , translation termination has high fidelity and is generally efficient [1 , 7] . However , K-12 RF2K-12ΔRF3 trends towards higher RPORs than K-12 RF2K-12 across the entire range of RPOR values ( Fig 4B and S4A Fig ) , indicating a potential for globally reduced termination efficiency across the translatome . This shift is quite pronounced for those genes with the highest RPOR values: K-12 RF2K-12ΔRF3 has nearly twice as many genes with RPOR > 1 . 0 compared to K-12 RF2K-12 ( Fig 4C ) . Although the RPOR values of K-12 RF2BΔRF3 are not nearly as elevated as those of K-12K-12ΔRF3 ( Fig 4B ) , replicate experiments suggest that they are elevated relative to the native K-12 RF2K-12 strain ( S4B Fig ) . Finally , the RPOR values of K-12 RF2B are very similar to those of K-12 RF2K-12 . Our most complete K-12 RF2B dataset showed a very slight decrease in RPOR values relative to K-12 RF2K-12 cells , but a smaller dataset did not exhibit this trend ( S5 Fig ) . Ribosome profiling in other organisms , including Saccharomyces cerevisae and Drosophila melanogaster , is capable of producing protected ribosome fragments with reading frame information more precisely allowing for identification of recoding events [46 , 47] . However , bacterial ribosome footprints are generated with MNase , and reading frame information is lost because of the sequence specificities of this nuclease [48] . These specificities do not allow for perfect cutting of mRNA around the ribosome , resulting in variable ribosome footprint sizes , depending on sequence context , without reading frame information [49] . Therefore , we used a secondary criterion to identify those ORFs whose translation is likely extended into the post-ORF region as a result of readthrough or frameshifting . If the ribosomes found in the post-ORF region are actually translating , then translation should terminate after the ribosomes encounter a post-ORF stop codon in the translating reading frame . Therefore , we searched the data for ORFs exhibiting a reduction in ribosome density after a stop codon present in any frame of the post-ORF region relative to its density prior to that stop codon ( Fig 5 ) . We hand-annotated the 100 genes with the highest relative post-ORF occupancy in both native K-12 RF2K-12 and K-12 RF2K-12ΔRF3 strains ( 121 total; S1–S4 Tables ) to identify cases where there was a decrease in ribosome density coincident with a stop codon located in any of the three possible post-ORF reading frames . Using this criterion we classified 43 ORFs as likely recoding events ( S1 Table ) . An additional 41 ORFs were classified as possible recoding events based on appropriate post-ORF termination with the addition of confounding factors such as low reads or other sequence elements that could contribute to high RPOR values ( S2 Table ) . The vast majority of genes annotated as likely recoding events had a larger RPOR value in K-12 RF2K-12ΔRF3 than in native K-12 RF2K-12 cells ( S1 Table ) . Interestingly , UGA codons were significantly over-represented ( p-value = 1x10-4 ) in both likely and possible recoding events , occurring at nearly double their expected frequency , while UAA codons were de-enriched ( S5 Table ) . We classified the 37 ORFs without reduced ribosome density after post-ORF stop codons as non-recoding events ( S3 Table ) . We additionally found 4 cases of exceptionally high RPOR , which appear to stem from mis-annotations ( ydcM , yeaP , wbbK , yebW ) ( S4 Table ) and 12 ORFs that had too few reads to classify . Taken together , our data strongly suggest that native K-12 RF2K-12 has some recoding events and that deletion of RF3 in the K-12 background enhances recoding , likely due to poor termination/release efficiency . We tested whether our screening criterion , reduced ribosome density after a stop codon in the post-ORF region , identifies genes with a recoding event for three individual genes: nudL and panZ identified in this work , and pheL , a known frameshifter ( S1 Table ) [50] . We individually expressed the three genes on a multi-copy plasmid in all strain backgrounds; they all contained an N-terminus FLAG-tag , and the two newly identified putative recoding events were additionally tagged with streptavidin on the C-terminus of the suspected extension frames . We tested for protein extension products with quantitative Western blotting . This revealed that all three genes exhibited extended proteins in native K-12 RF2K-12 , and that two have increased extended product in K-12 RF2K-12ΔRF3 ( Fig 6B , S6 Fig ) . The nudL post-ORF region ribosome density decreased significantly after two closely spaced stop codons , suggesting either readthrough ( 0 frame ) or a -1 frameshift ( Fig 6A ) . By constructing C-terminal streptavidin tags in both frames , we were able to show that the protein extension product resulted from stop codon readthrough in the 0 frame ( S7 Fig ) . The extension product is approximately 6-fold more prevalent in K-12 RF2K-12ΔRF3 than in K-12 RF2K-12 ( Fig 6B ) and is undetectable in strains with RF2B even when they lack RF3 ( S8A Fig ) . Ribosome density in the post-ORF region of panZ is complex with many possible points of termination ( Fig 6C ) . We tested whether there was extension in the -1 frame , as predicted by the major decrease in ribosome density after the first stop codon in the post-ORF region . We observe a large amount of the -1 frameshift product in all strains , accounting for ~35% of the total PanZ production ( Fig 6D , S8B Fig ) . The high level of PanZ extension product may result from unknown cis element ( s ) that leads to frameshifting . The three well-established frameshifting events in native K-12 E . coli are RF2 ( encoded by prfB ) , PheL , and DnaX [51] . We discussed the frameshift used to control the amount of full-length RF2 above ( Fig 3 ) . Here we describe our studies of PheL , the other established locus where a frameshift event produces an extended product . Our ribosome profiling data showed decreased ribosome density after a stop in the 0 frame ( S6A Fig ) , in addition to the previously identified +1 frameshift , which ends in the downstream pheA gene [52] . Using N-terminal FLAG-tagged PheL , we see both products in K-12 RF2K-12 , with enhanced production in K-12 RF2K-12 ΔRF3 cells ( S6B Fig ) . However , we find no extended FLAG-PheL products in K-12 RF2B or K-12 RF2BΔRF3 cells ( S6B Fig ) . Interestingly , native K-12 RF2K-12 cells have an unusually large ribosome density over the pheL UGAA four-base stop codon ( S6A Fig ) . This enhanced density is almost completely eliminated in both K-12 RF2B and K-12 RF2BΔRF3 cells ( S6A Fig ) , likely because UGAA signals rapid termination in cells with RF2B protein variant . Genes in E . coli are densely packed in the chromosome , with approximately 15% of adjacent ORF pairs having overlapping stop and start codons [53] . In some cases , this overlap has been shown to promote translational coupling , possibly by enabling upstream ribosomes to influence downstream ORF translation by unwinding mRNA structure or affecting ribosome dissociation-reinitiation cycle [54–58] . We asked whether the relative translation level of downstream genes ( normalized to that of upstream genes ) increased in K-12 RF2K-12ΔRF3 as compared to native K-12 RF2K-12 for adjacent ORF pairs with overlapping stop and start codons . We find that when the upstream gene has higher translation than the downstream gene in native K-12 RF2K-12 cells , deleting RF3 can increase translation of the downstream gene ( S9A Fig ) . The aberrantly high translation was rescued in K-12 RF2B ΔRF3 , and K-12 RF2B is similar to native K-12 RF2K-12 ( S9B and S9C Fig ) . Accumulation of ribosomes at the upstream stop codon may promote unwinding of the mRNA structure at the translation initiation region of the downstream gene to increase its translation level , or alternatively , may be an example of readthrough . Expression of many E . coli biosynthetic genes is controlled by regulated transcription termination , or attenuation . Attenuation is mediated by two competing RNA stem-loop structures , one signaling transcription termination and the other allowing transcription to continue [59] . Ribosome occupancy in the leader peptide mRNA determines the ratio of the two stem-loop structures . Enhanced ribosome occupancy in the leader peptide , indicative of a deficiency of the amino acid produced by the biosynthetic operon , leads to enhanced transcriptional readthrough , thus ensuring adequate production of the limiting amino acid . We asked whether altered ribosome occupancy at leader peptide stop codons in the K-12 RF2K-12ΔRF3 strain altered attenuation . A comparison of ribosome footprint density of native K-12 RF2K-12 with that of K-12 RF2K-12ΔRF3 indicated that biosynthetic genes under the control of a leader peptide appeared to be down-regulated in K-12 RF2K-12ΔRF3 relative to native K-12 RF2K-12 in rich media ( Fig 7A ) and less so , if at all , in minimal media ( Fig 7B ) . We quantified the attenuation of two leader peptide operons , ivbL and trpL , in both MOPS-complete glucose , a rich medium , and MOPS-glucose , a minimal medium , using two chromosomally integrated lacZ reporters for each operon . Reporter #1 measured expression of the leader peptide to quantify the rate of transcription of the operon; and reporter #2 measured expression of downstream genes ( Fig 7C ) . Relative to native K-12 RF2K-12 , K-12 RF2K-12 ΔRF3 cells have reduced expression of trp genes in both rich and minimal medium , while ivbL has reduced expression in rich medium ( Fig 7D ) . As reporter #1 allows us to normalize for rate of transcription , we conclude that this decrease in expression is the result of increased attenuation due to increased ribosome occupancy over the stop codon in K-12 RF2K-12 ΔRF3 cells ( Fig 7E ) . We also performed these experiments for the hisGDCBHAFI operon but were unable to measure transcription , as reporter #1 was toxic . Results with reporter #2 indicated that expression of his was decreased in K-12 RF2K-12 ΔRF3 cells in rich but not minimal medium ( S10 Fig ) . We believe that this also represents attenuation because expression of the ivlBN and trpEDCBA leader peptides were comparable in K-12 RF2K-12 and RF2K-12ΔRF3 strains , suggesting that the lack of RF3 may not alter transcription from the leader peptide promoter . The results of expression data from our ribosome profiling experiments , coupled with confirmatory lacZ fusion experiments indicate that altered ribosome density over the stop codon alters the outcome of the RNA structure competition , such that cells with enhanced stop codon occupancy have increased transcription termination . Most likely , increased occupancy over the stop codon shifts the equilibrium between the readthrough and termination stem-loop structures to one favoring the termination stem-loop . These findings are consistent with and extend an early study showing that some temperature sensitive RF2 strains increased transcription termination in the trp operon [60] . In that study , enhanced termination was observed only at UGA stops , but in our study , possibly because of a stronger perturbation , we observe increased termination also at the UAG stop in the ivbL locus . These results indicate how the defects of release factor mutations can alter downstream gene expression depending on transcription-translation coupling . It is critically important for organisms to terminate the translation of proteins accurately and in an appropriate time frame . Release factors are central to this process . In this work , we compared the accuracy of translation termination and its downstream consequences at a global scale in native E . coli K-12 cells , which have a reduced function RF2 protein ( RF2K-12 ) with K-12 cells harboring a fully functional RF2 ( RF2B ) , and examined the consequences of the deletion of RF3 in both strain backgrounds . Our results provide the first picture of the genome wide consequences of stressing translation on ribosome behavior and release factor synthesis in a panel of isogenic strains with increasingly deficient release factor activity . We used ribosome occupancy over stop codons as our in vivo composite metric for translation termination and ribosome release . In native K-12 RF2K-12 cells , the average occupancy of ribosomes at stop codons is about 2-fold higher than that in the coding regions , reflecting the time required from stop codon recognition to ribosome release ( Fig 2A ) . The faster rate of peptide hydrolysis by the RF2B variant relative to RF2K-12 [16] is manifest as a reduction in ribosome stop codon occupancy in K-12 RF2B at UGA stop codons , which are fully dependent on RF2 for polypeptide release ( Fig 2D ) . This effect is particularly strong at the UGAA codon which has relatively weak binding to RF2 ( S1F Fig ) [39 , 44 , 61] . Weak association of RF2 with a stop codon may allow immediate dissociation of RF2 following hydrolysis , which is especially rapid in strains containing the more active RF2B variant , thus further decreasing the time to ribosome dissociation . On the other hand , both K-12 RF2B and K12 RF2K-12 strains have similar stop codon occupancies at UAA codons . UAA is decoded by both RF1 and RF2 , with previous studies indicating that UAA is predominantly recognized by RF1 in strains with RF2K-12 , but that RF2 plays a major role in recognizing UAA codons in strains with RF2B [14 , 15 , 30] . Thus , the higher activity of RF2B as compared to RF2K-12 is manifest only at the UGA codon , recognized solely by RF2 . Deletion of RF3 globally perturbs stop codon occupancy by ribosomes , both for K-12 RF2B and K-12 RF2K-12 strains . K-12 RF2B ΔRF3 exhibits about a 1 . 4-fold increase in occupancy at all stop codons relative to K-12 RF2B ( Fig 2B–2D ) , indicating that the absence of RF3-mediated release factor dissociation visibly increases the dwell time of ribosomes at stop codons . Because of the composite nature of our measurement , we cannot say whether the effect we see is commensurate with the expectation from in vitro studies that loss of RF3 decreases the rate of release factor dissociation by as much as 500-fold or whether our results indicate that EF-G and RRF may partially compensate for RF3 , as has been suggested [4 , 62 , 63] . For K-12 RF2K-12 ΔRF3 cells , the global perturbation of stop codon occupancy by ribosomes is more severe , exhibiting an almost 2 . 5-fold increase in ribosome occupancy at all stop codons relative to native K-12 cells ( Fig 2A ) , a 1 . 5-fold increase in expression of both RF1 and RF2 ( Table 2 ) , and ≥20% increase in doubling time ( Fig 1A ) . As the level of RF2 is adjusted by an internal frameshift , the rate of frameshifting changes in concert with the change in protein levels ( Fig 3 ) . RF1 and RF2 are sub-stoichiometric with respect to ribosomes: RF1 is 100-fold less abundant; and RF2 is 12-fold less abundant than the number of ribosomes in K-12 RF2K-12 cells [37 , 43 , 64] . This level of release factors is normally sufficient for efficient termination because release factors act catalytically on the small subpopulation of terminating ribosomes , and are then rapidly released from the ribosome by RF3 action [9 , 33] . However , in the absence of RF3 , the much slower dissociation of RFs increases the dwell time of RFs at stop codons , which in turn leads to a decrease in the cellular concentration of free RFs . This effect is exacerbated when it is coupled with the defective peptide release RF2K-12 protein , leading to even longer dwell time of release factors and further depletion of pool of free RFs . The cell responds by increasing expression of RF1 and RF2 in the K-12 RF2K-12ΔRF3 strain , but this is insufficient to counteract sequestration of RFs as termination is globally slowed at all stop codons , including UAG , which is recognized solely by RF1 ( Fig 2 , S1 Fig ) . Our studies stressed translation in K-12 RF2K-12 by removing RF3 . Previous studies indicated that perturbing translation in K-12 RF2K-12 by inactivating the pseudouridine synthase , rluD , or the RF methyltransferase , prmC , were ameliorated by the RF2B variant [14 , 65] . In K-12 E . coli deletion of these modification enzymes leads to a very severe growth defect , which is lethal in the case of K-12 ΔprmC . Therefore the relatively gentler perturbation by ΔRF3 in K-12 allows us to identify these changes in ribosomal occupancy that would be difficult or impossible to identify in K-12 deletions of rluD or prmC . Deletions of prmC or rluD are viable in backgrounds containing RF2B and they would likely exhibit the similar molecular signature as the K-12 RF2K-12ΔRF3 strain , enhanced expression of RF1 and 2 coupled with a significant enhancement in ribosome occupancy at stop codons . Two lines of evidence suggest that the enhanced ribosome occupancy seen at stop codons in cells with reduced RF2 function ( e . g . RF2K-12 ) and lacking RF3 leads to translation readthrough and/or frameshifting at a significant number of genes . First , for 121 ORFs with the highest post-ORF ribosome occupancy , ≥30% of the genes have a decrease in ribosomal density after a stop codon in the post-ORF region ( S1 Table ) . This signature is indicative of bona fide translation termination at those stop codons . Second , we rigorously confirmed recoding events in three cases . For the 2 new cases , we visualized the extension products by performing Western blotting on proteins that had distinct tags at both their N-terminus and at the C-terminus of the putative protein extension ( Fig 6 and S6 Fig ) . These experiments reveal that 2 of the 3 , nudL and pheL show readthrough of the UGA stop codon , notable because in bacteria readthrough events are rare relative to frameshifting events [51] . We favor the idea that near-cognate tRNAs for cysteine and tryptophan known to decode UGA stop codons [66 , 67] account for readthrough , as the sequence element required for using selenocysteine to decode UGA [68–70] is not present at these loci . Interestingly , PheL exhibits the previously documented +1 frameshift [52] , in addition to the stop codon readthrough indicating that during their extended occupancy over the stop codon , both readthrough and frameshifting events are possible . Programmed recoding can expand the repertoire of gene products and is utilized for gene regulation in bacteria , viruses , yeast and higher eukaryotes [51 , 71] . While most of the recoding previously found in E . coli produces functional proteins , as is the case for PrfB and DnaX , a functional role for +1 PheL frameshift seen in K-12 E . coli has not been found [52] . We believe that the bulk of the suspected recoding we found in E . coli simply reflects the limitations of the reduced function RF2K-12 protein , which terminates poorly at UGA stops relative to RF2B , a phenotype that is exacerbated by the absence of RF3 . This idea is consistent with over-representation of UGA codons amongst ORFs likely to exhibit the greatest amount of recoding and with the strain specific behavior of the readthrough products of NudL and PheL and +1 frameshift product of PheL ( S6 Table , S6 and S8 Figs ) . The NudL and PheL extended products increase about 5-fold in K-12 RF2K-12ΔRF3 relative to K-12 RF2K-12 but were not present in K-12 RF2B . Remarkably , they are also absent in K-12 RF2BΔRF3 . In these two cases , ΔRF3 alone may not be sufficient to drive recoding and might require a second perturbation coming from the weaker RF2K-12 variant . It is likely that many of the other UGA stop codons with very high post-ORF ribosome occupancy will show the same patterns as NudL and PheL . While we examined recoding only for ORFs with very high post-ORF ribosome occupancy , the shift towards higher occupancy in K-12 RF2K-12ΔRF3 strains occurs in majority of the 1600 genes examined ( Fig 4B ) , potentially producing defective protein products , albeit at a very low level . Why might the defects in translation termination of ΔRF3 in a K-12 background lead to increased recoding ? As a consequence of release factor sequestration , the effective concentration of free RF1/2 is reduced , reducing termination efficiency and increasing the likelihood of recoding . During the increased time spend at stop codons , conformational flexibility in the complex may allow transient slippage and/or partial binding of an amino-acyl tRNA , which would be followed by binding of EF-G and subsequent elongation . The chances for EF-G binding are increased when cells lack RF3 because both proteins bind to the same position on the ribosome [13] . Recent studies of the kinetics of intragenic frameshifting reveal that it is accompanied by slower EF-G catalyzed ribosome translocation and a change in the fluctuating conformational states [72–74] . Because of the altered ribosome conformation at the stop codon in the absence of RF3 , we speculate that such conditions may also occur during slippage at the stop codon , with its frequency determined by sequences surrounding the stop codon . Given the sensitivity of E . coli K-12 to perturbations in translation termination , we wonder why this lineage does not have the more conserved , fully functional RF2B protein . There are several possibilities . First , the alterations in gene expression for K-12 RF2K-12 relative to K-12 RF2B , in the presence or absence of RF3 , may in aggregate be beneficial to the host , resulting in maintenance of the allele . Second , the lab strains derived from E . coli K-12 may not been under strong enough selective pressure to select for enhanced function release factor mutants . In support of this idea , when E . coli K-12 RF2K-12 is exposed to extreme translational stress by deleting ΔrluD ( rRNA pseudouridine synthase ) , it acquires suppressors at the RF2 locus [65] . Finally , K-12 RF2K-12 may have acquired a compensatory mutation ( s ) enabling the reduced function allele to perform relatively well . Similar compensatory mutations have been documented in studies of antibiotic resistance [75] . If so , one might imagine that the compensatory mutations might be incompatible with the fully functional allele in some circumstances . The loss of viability of K-12 RF2B ΔRF3 but not K-12 RF2K-12 ΔRF3 at very low temperatures is consistent with possibility that contemporary K-12 strains have acquired compensatory mutations , which increase their fitness . All experiments were performed in E . coli K-12 strain MG1655 or its derivatives . Our strain name and genotype of each strain in K-12 E . coli are as follows; K-12 RF2K-12: MG1655 prfB [E . coli K-12] , K-12 RF2K-12ΔRF3: MG1655 prfB [E . coli K-12] ΔprfC::frt , K-12 RF2B: MG1655 prfB [E . coli B] , K-12 RF2BΔRF3: MG1655 prfB [E . coli B] ΔprfC::frt . The K-12 RF2K-12ΔRF3 strain was constructed by transducing the prfC::kanR locus from the KEIO collection into MG1655 and its kanR cassette was flipped out using pCP20 [76] . The K-12 RF2B strain was constructed by transducing the prfB ( RF2B ) locus with kanR downstream into MG1655 , The original prfB [E . coli B]: kanR construct was a gift of Lei Wang at The Salk Institute for Biological Studies . The linked kanR marker was removed by λ Red assisted homologous recombination to obtain K-12 RF2B without markers [77] . K-12 RF2BΔRF3 was constructed using standard P1 transduction of the KEIO locus into the K-12 RF2B background . Cultures for growth curve experiments were grown in rich defined liquid medium , MOPS-complete glucose overnight at 37°C then diluted to OD420 0 . 005 in 35mL of fresh MOPS-complete glucose medium . Cultures were grown in a shaking water-bath incubator at 37°C with cell density measured by optical density at 420nm wavelength every 30 minutes until stationary phase . Cell densities over OD420 0 . 5 were diluted to ensure we obtained values within the linear range of the spectrophotometer . The spotting assay of cold sensitivity was performed by culturing cells to mid-exponential growth in LB medium at 37°C and making serial 1 in 10 dilutions up to 10−6 . We spot plated 3uL of each dilution onto LB plates in duplicate and incubated at 37°C , 30°C , 25°C , 20°C , and 15°C . Plates were removed from the incubators and imaged when individual colonies were visible . Cells for the cold viability experiments were prepared in the same manner and dilution series were plated on LB plates by glass beads in triplicate . Plates were incubated in the same temperature series and viable cells were counted once colonies were visible . Ribosome profiling was performed as previously described by Oh et al [49] . All strains; K-12 RF2K-12 , K-12 RF2K-12ΔRF3 , K-12 RF2B and K-12 RF2BΔRF3 were grown at 37°C in MOPS-complete glucose , a rich defined media , as a liquid culture [78] . Previous ribosome profiling experiments have shown that ribosome pausing increases over serine codons during exponential growth in Luria broth due to serine being the first catabolized amino after sugar is utilized [78] . Use of MOPS-complete glucose eliminates these pauses and provides a more stable growth medium for E . coli during our experiments . Profiling was also performed in K-12 RF2K-12 and K-12 RF2K-12ΔRF3 strains in MOPS-glucose , a minimal defined media , for the results seen in Fig 7B . Cells from each strain were rapidly harvested by filtration and lysate was produced by pulverization of liquid nitrogen cooled samples . From this lysate ribosome footprints were created using MNase and total RNA was extracted for a simultaneous RNA-seq library production . Ribosome footprints between 24-32nt were isolated from the initial polyacrylamide sizing gel . Total RNA purified samples for RNA-seq were fragmented by alkaline fragmentation and 20-35nt fragments were isolated from the sizing gel . After ligation to Linker-1 the fragments were converted to a cDNA library and subsequent library preparations were performed as previously described [46 , 49] . Libraries were sequenced on an Illuminia HiSeq 2000 or HiSeq4000 . Raw and processed data are available on the NCBI Gene Expression Omnibus ( GEO ) under the accession number GSE88725 . Generated sequencing reads were analyzed as previously described [46 , 49] . Reads were initially aligned to a genome file containing E . coli rRNA and tRNA sequences to computationally subtract rRNA and tRNA reads . Then the remaining unaligned reads were aligned to the E . coli genome NC_00913 . 2 , reads with more than two mismatches were excluded as were reads aligning to multiple portions of the genome . Aligned reads between 20bp and 40bp in length were trimmed by 10bp on each side and center mapped on the genome as previously described [49] . Wiggle files were generated using the same read lengths and center mapping as described above and counts were normalized per million reads and visualized in IGV . To analyze post-ORF occupancies as outlined in Fig 4 , we also aligned ribosome profiling reads to the 3’ end in accordance with a recent publication to increase reading frame information [79] . We found both center alignment and 3’ alignments of fragments led to the same conclusion . Gene expression for each gene was calculated in Plastid , a Python library for deep sequencing genomics , by masking occupancy at the first and last 5 codons within the annotated ORF , and normalized for read-depth and gene length by using RPKM ( reads per kilobase of transcript per million mapped reads ) [80] . Expression between two strains was only compared when genes each had at least 100 counts . Metagene analysis was performed by normalizing 70-100nt upstream of the stop codon within each gene then plotting the median for each position . Genes with less than 1 counts in this window and genes less than 50nt away from the downstream coding frame were excluded from the metagene analysis in Fig 2 ( note: S1 Fig differs slightly with a lack of read count filter to include more genes ) . Signals from biological replicates of different samples were averaged and standard deviations were calculated respectively . Genome wide calculations of ribosomes in the post-ORF occupancy were performed for all genes with an intergenic region of 65bp or greater that had average mRNA and ribosome footprint densities of 0 . 1 or greater ( overlapping genes were excluded ) . Post-ORF occupancy was determined by calculating the average density in the window 20-60bp downstream of the annotated stop codon . The start point of 20bp was chosen in order to eliminate the impact of ribosomes over the stop codon contributing to post-ORF counts . This window was utilized for all genes regardless if possible extensions were shorter or longer than that window . The average post-ORF occupancy of a gene was divided by the average occupancy of gene , which was calculated as described above , to obtain the relative post-ORF ribosome occupancy metric ( RPOR ) . The RPOR roughly estimates the fraction of recoding occurring at each gene . For the cumulative distribution analysis all RPOR values of 0 . 0 were removed because we are unable to resolve between zero recoding events and insufficient read depth , the presence of zeros also falsely inflated the K-S statistic . The pFLAG-MAC plasmid background was used to construct inducible N-terminal-FLAG , C-terminal streptavidin tagged plasmids for nudL and panZ while only the N-terminal-FLAG was used for pheL . We cloned in the gene locus from directly after the start codon extending past the annotated stop codon using restriction digest cloning with HindIII and EcoRI . Utilizing our ribosome profiling data we were able to identify the possible extension and cloned in a sufficient segment of each 3’-UTR , using predicted stop codons to place the C-terminal streptavidin tag . Each plasmid sequence was confirmed by Sanger sequencing and TSS transformed into K-12 RF2K-12 , K-12 RF2K-12ΔRF3 , K-12 RF2B and K-12 RF2B ΔRF3 strains . Strains containing our pFLAG-MAC constructs were cultured to mid-exponential growth and induced with 1mM IPTG for 2 hours and total cell protein was harvested using TCA precipitation . NudL and PanZ proteins were resolved on 12% Bis-Tris gels with MOPS running buffer . PheL protein was resolved on 16% Tricine-SDS gels with Tricine-SDS running buffer . Proteins were transferred onto nitrocellulose membranes via wet transfer then blocked using Li-Cor Odyssey PBS blocking buffer and incubated with primary antibodies α-FLAG ( Sigma-F3165 ) and α-SurA at 1:5000 dilutions and α-streptavidin ( abcam , ab76950 ) at a 1:2500 dilution overnight at 4°C . Membranes were washed with 1x PBS before incubation with Li-Cor α-mouse and α-rabbit secondary antibodies at 1:10000 dilutions and again washed before imaging . Membranes were imaged on a Li-Cor infrared imager . The 700nm and 800nm channels of the Li-Cor image allowed us to visualized both FLAG and streptavidin on the same membrane . Quantifications were performed using Image-J software and SurA as a loading control . Two reporter constructs were assembled for each biosynthetic locus; hisL , trpL and ivbL using an attλ integratable plasmid . The leader peptide reporter ( #1 ) was constructed by cloning in the entire 5’UTR and leader peptide without its stop codon upstream of lacZ creating a translational fusion . The downstream reporter ( #2 ) was constructed by cloning the entire 5’UTR , leader peptide , intergenic region and a small segment of the next gene downstream into the same plasmid backbone upstream of lacZ . Once all six plasmids were generated and validated by Sanger sequencing , we integrated them into both K-12 and K-12 RF2K-12 ΔRF3 cells . Integration was performed using CRIM helper plasmid pINT-ts and single integrants were confirmed by diagnostic PCR [81] . After multiple attempts we found the hisL leader peptide reporter to be toxic to the cell , likely due to the high production of the lacZ-fusion . Cultures containing reporter plasmids were grown in either MOPS complete-glucose or MOPS-glucose , a defined minimal medium , overnight at 37°C and dilutions were performed to OD420 of 0 . 005 prior to the experiment in their respective mediums in triplicate . Once cultures had reached exponential growth samples of 500uL and 1mL were taken simultaneously to measure β-galactosidase ( β-gal ) activity and OD420 at three timepoints . β-gal assay samples were immediately added to a tube containing 500uL Z-buffer , 1 . 25uL beta-mercaptoethanol , 30uL 0 . 1% SDS and 40uL chloroform then vortexed before being placed on ice . Once collection was completed all β-gal sample tubes were incubated at 28°C prior to induction with 200uL 4mg/mL ONPG . Development was stopped with 500uL of 1M sodium bicarbonate and samples were centrifuged for 5 minutes at max speed to remove cellular debris . Samples were then transferred into a 96-well plate to measure OD420 in a Varioskan plate reader . Cell density and OD420 of developed β-gal samples were used to calculate Miller units of activity .	Proteins are the cellular workhorses , performing essentially all of the functions required for cell and organismal survival . But , it takes a great deal of energy to make proteins , making it critical that proteins are made accurately and in the proper time frame . After a ribosome synthesizes a protein , release factors catalyze the accurate and timely release of the finished protein from the ribosome , a process called termination . Ribosomes are then recycled and start the next protein . We utilized ribosome profiling , a method that allows us to follow the position of every ribosome that is making a protein , to globally investigate and strengthen insights on termination fidelity for cells with and without mutant release factors . We find that as we decrease release factor function , the time to terminate/release a protein increases across the genome . We observe that the accuracy of terminating a protein at the correct place decreases on a global scale . Using this metric we identify genes with inherently low termination efficiency and confirm two novel events resulting in extended protein products . In addition we find that beyond disrupting accurate protein synthesis , release factor mutations can alter expression of genes involved in the production of key amino acids .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "chemical", "compounds", "operons", "carbohydrates", "organic", "compounds", "glucose", "genome", "analysis", "translation", "termination", "molecular", "biology", "techniques", "dna", "cellular", "structures", "and", "organelles", "genetic", "footprinting", "research", "and", "analysis", "methods", "gene", "expression", "chemistry", "molecular", "biology", "ribosomes", "biochemistry", "genetic", "fingerprinting", "and", "footprinting", "cell", "biology", "organic", "chemistry", "nucleic", "acids", "protein", "translation", "genetics", "monosaccharides", "biology", "and", "life", "sciences", "physical", "sciences", "genomics", "gene", "prediction", "computational", "biology" ]	2017	Global analysis of translation termination in E. coli
Post-exposure prophylaxis ( PEP ) against rabies infection consists of a combination of passive immunisation with plasma-derived human or equine immune globulins and active immunisation with vaccine delivered shortly after exposure . Since anti-rabies immune globulins are expensive and scarce , there is a need for cheaper alternatives that can be produced more consistently . Previously , we generated potent virus-neutralising VHH , also called Nanobodies , against the rabies glycoprotein that are effectively preventing lethal disease in an in vivo mouse model . The VHH domain is the smallest antigen-binding functional fragment of camelid heavy chain-only antibodies that can be manufactured in microbial expression systems . In the current study we evaluated the efficacy of half-life extended anti-rabies VHH in combination with vaccine for PEP in an intranasal rabies infection model in mice . The PEP combination therapy of systemic anti-rabies VHH and intramuscular vaccine significantly delayed the onset of disease compared to treatment with anti-rabies VHH alone , prolonged median survival time ( 35 versus 14 days ) and decreased mortality ( 60% versus 19% survival rate ) , when treated 24 hours after rabies virus challenge . Vaccine alone was unable to rescue mice from lethal disease . As reported also for immune globulins , some interference of anti-rabies VHH with the antigenicity of the vaccine was observed , but this did not impede the synergistic effect . Post exposure treatment with vaccine and human anti-rabies immune globulins was unable to protect mice from lethal challenge . Anti-rabies VHH and vaccine act synergistically to protect mice after rabies virus exposure , which further validates the possible use of anti-rabies VHH for rabies PEP . Rabies virus ultimately causes an aggressive and lethal infection in the brain of humans and other mammals . Rabies virus is a model neurotropic RNA virus that belongs to the family Rhabdoviridae , Genus Lyssavirus [1;2] . The virus is transmitted through the saliva of an infected animal by biting or scratching . Once the virus enters peripheral nerves or neurons , it quickly replicates in the neuronal cytoplasm and progeny virus is transported through the neuronal network by crossing tight interneuronal synapses , eventually giving rise to encephalitis [3;4] . Each year , an estimated 59000 people die from rabies and about 29 million receive post-exposure prophylaxis ( PEP ) after close contact with a suspected animal [5] . Passive antibody therapy with anti-rabies immune globulins ( RIG ) plays a major role in rabies post-exposure prophylaxis after high risk exposure [6] . Together with thorough wound cleansing , it is the first line of defence against the virus , and prophylaxis without RIG is associated with treatment failure [7;8] . Pioneering studies on the effects of anti-rabies serum date back to the late 1800s and early 1900s , and since 1954 the World Health Organisation ( WHO ) recommends the use of RIG in combination with vaccination for rabies post-exposure prophylaxis [9] . Treatment with RIG and vaccine should be initiated as soon as possible after potential infection , with additional vaccine administrations in the following weeks to activate a full-blown and lasting immune response . Passive immunization with RIG serves to immediately neutralize the virus and close the gap between viral exposure and the vaccine-induced immune response [7] . In this regime , initial protection is offered by RIG , which is then gradually replaced by vaccine-induced antibodies mounted between day 0 and 7–14 , providing continued protection to patients [10] . Rabies antibodies can be either from equine ( ERIG ) or human ( HRIG ) origin . Due to adverse effects , such as serum sickness , equine antibodies are now used under the form of pepsin-digested Fab fragments , but if available , HRIG is still preferred over ERIG [9] . The production of HRIG , however , requires sufficient numbers of immune donors and gives rise to the typical problems associated with biological products of human origin , such as the transmission of infectious agents [9] . The worldwide shortage and the high costs makes these products poorly available to developing countries , where rabies is endemic [7;9] , the reason why the WHO recommends to develop alternatives [11] . VHH or Nanobodies ( a trade-name by Ablynx ) are the smallest functional fragments ( 15 kDa ) of heavy chain-only antibodies naturally occurring in Camelidae , and represent the antigen-binding variable domain . By nature VHH are hydrophilic and do not require hydrophobic interactions with a light chain , which allows high solubility , physicochemical stability and high-yield production in Escherichia coli , yeast or mammalian expression systems . The single domain nature and the small size of VHH also allows for easy formatting by genetic fusion into multimeric and multispecific constructs [12–14] . Previously , we generated a potent neutralizing anti-rabies VHH recognising two epitopes on the rabies glycoprotein , fused to an anti-albumin VHH to extend its serum half-life ( HLE ) . The Rab-E8/H7-ALB11 was able to neutralize the virus at picomolar doses [15] . Post exposure treatment with anti-rabies VHH at 24 hours after intranasal virus challenge could significantly delay disease onset in mice , and depending on the dose , could rescue part of the mice from lethal disease [15] . The main aim of this study was to examine whether the combined treatment with anti-rabies VHH and vaccine ( Rabipur , Novartis ) after exposure to rabies virus has added value compared to single treatment with either compound in the intranasal rabies virus challenge model , which is very well suited to study intervention strategies for prevention and prophylaxis [16] . Via the intranasal route the virus can directly access the brain via the olfactory epithelium , which results in a highly reproducible infection [17] . First disease signs appear at 7 days , which rapidly progress the following 2 days , requiring euthanasia at 8–9 days post inoculation ( DPI ) . This model was recently also proposed as a valuable alternative to intracranial inoculation for rabies vaccine potency testing [18] . The typically short incubation period of this model ( 6 . 07 ± 0 . 59 days ) is ideal to study the potentially beneficial effect of the combined passive ( VHH ) and active ( vaccine ) immunisation on disease outcome . Our results show that anti-rabies VHH and vaccine act synergistically to protect mice after rabies virus exposure , which further validates the possible use of anti-rabies VHH for rabies PEP . VHH directed against the rabies virus glycoprotein G were described previously [19] . Briefly , llamas were vaccinated using the inactivated rabies Human Diploid Cell Vaccine ( HDCV , Sanofi , France ) and RNA was extracted from peripheral blood lymphocytes . VHH genes were amplified from a cDNA library . Anti-rabies VHH were selected by panning phage libraries on plates coated with the native G protein . Multivalent VHH constructs were generated by the fusion of monovalent VHH into multimeric VHH constructs using flexible glycine-serine ( GS ) linkers [20] . In this study , we used the half-life extended VHH ( HLE Rab-E8/H7-ALB11 ) , containing two different VHH against the rabies virus spike protein and an anti-albumin VHH ( ALB11 ) for half-life extension , and the non-HLE Rab-E8/H7 [15] . VHH was produced and kindly provided by Ablynx ( Zwijnaarde , Belgium ) . Human rabies immune globulins ( HRIG ) ( Berirab , CSL Behring GmbH , Germany ) are gammaglobulins purified from plasma of vaccinated human donors . Rabipur ( Purified Chicken Embryo Cell Vaccine , Novartis , Belgium ) was reconstituted according to the manufacturer’s instructions and was administered via intraperitoneal or intramuscular injection . The vaccine contains at least 2 . 5 antigenic units ( AU ) /ml . It contains the inactivated Flury LEP strain produced on purified chick embryo cells . Challenge Virus Standard ( CVS ) -11 is a virulent classical rabies virus obtained from the American Type Culture Collection ( ATCC reference VR959 ) and was grown in baby hamster kidney ( BHK ) -21 cells . For virus inoculation in mice , a dose of 102 . 5 50% cell culture infectious doses ( CCID50 ) was used . Six-to-eight weeks old female Swiss outbred mice ( Charles River , France ) were used . Mice were kept in filter top cages , water and feed provided ad libitum and exposed to a natural day/night light cycle . Intranasal ( IN ) inoculation procedures are described in detail by Rosseels et al . [16] . The intranasal inoculation of rabies virus is an excellent technique to study antiviral treatment in the brain , since it leaves the brain mechanically intact , in contrast to intracranial inoculation , and yields a highly reproducible brain infection and disease outcome with little variation in the median survival time . This inoculation route has been used before for the evaluation of post exposure prophylaxis of rabies in mice [21] . For intraperitoneal ( IP ) or intramuscular ( IM ) injections maximum volumes of respectively 1000 and 100 μl were respected ( 50 μl per site in case of IM injections ) . Prior to intramuscular or intranasal administrations , mice were briefly anesthetized using isoflurane gas ( IsoFlo , Abbott laboratories Ltd . , United Kingdom ) , as described by Rosseels et al . ( 2011 ) [16] . Three retro-orbital bleedings were performed under isoflurane anaesthesia during the 28 day immunization period . Mice were observed daily for signs of disease until 35 days post virus inoculation . Mice develop a typical disease pattern , which progresses as follows: isolation from the group ( score 1 ) , slow/less vivid movement ( score 2 ) , paresis in paws ( score 3 ) , uncoordinated movement ( score 4 ) , absence of spontaneous movement ( score 5 ) , no response to stimuli ( score 6 ) and the end-stage characterized by mice burying their heads in cage bedding and slow breathing ( score 7 ) . The score per mouse ranges thus from 0 ( no disease ) to 7 ( severe nervous disease ) . Disease progression was represented by plotting the daily score in function of the days post inoculation ( DPI ) . The incubation period was defined as the period between virus inoculation and the first appearance of disease signs . In our experience , mice with a disease score of 6 or more die within 24 hours . Therefore , mice were euthanized by cervical dislocation when they reached a score of ≥ 6 . Results were expressed with Kaplan-Meier survival curves . Rabies virus infection in the brain was confirmed using real-time reverse transcriptase polymerase chain reaction ( RT-qPCR ) as described by Suin et al . [22] , and by the fluorescent antigen test ( FAT ) , performed according to the Manual of Diagnostic Tests and Vaccines for Terrestrial Animals ( Office International des Epizooties , 2008 ) . The viral RNA load in the brain of mice was determined using RT-qPCR , as previously described [15;22] . Briefly , the brain was homogenized and RNA was extracted according to the manufacturer’s instructions ( RNeasy kit , Qiagen , Hilden , Germany ) . Ribosomal 18S was used as a reference gene for standardization and delta cycle thresholds ( Δ Cq ) values were calculated using the following formula: Δ Cq = Cqref−Cq , with Cqref equal to 45 , the number of cycles in this program . The virus-neutralizing titer of serum , antibody and VHH preparations was determined with the Rapid Fluorescent Focus Inhibition Test ( RFFIT ) , according to the Manual of Diagnostic Tests and Vaccine for Terrestrial Animals ( Office International des Epizooties , 2008 ) . The neutralizing potency is expressed in international units ( IU ) /ml in reference to "The Second International Standard for Anti-Rabies Immunoglobulin" , purchased from the United Kingdom National Institute for Biological Standards and Control . GraphPad Prism was used for statistical analyses of in vivo data . Differences in survival times were tested using the Log-Rank test with a Bonferroni post-test , differences in Δ Cq values were tested using a Student’s t-test after normalization to the house-keeping gene . Differences in antibody titers were also tested using a Student’s t-test . All experimental procedures were approved by the Ethical Commission of the WIV-ISP and CODA-CERVA ( advice number 070515–05 ) and were performed according to the EU Directive 2010/63/EU for animal experiments . To validate the protective effect of rabies vaccine ( Rabipur , Novartis ) in the intranasal rabies mouse model , mice were vaccinated with two intramuscular vaccine doses ( 0 . 25 AU/mouse ) , with a 3-day interval , following the schedule also used later on for PEP . This vaccination schedule is schematically represented in Table 1 . Mice received a viral challenge 25 days after the last vaccine , allowing sufficient time for the development of an immune response . The mounting of the humoral immune response in the blood after vaccination was monitored by assessing the rabies neutralization activity in vitro ( RFFIT ) in blood collected at different time points . In Fig 1 , it is shown that mice that received the vaccine had detectable antibody titers eight days after the first dose ( day -20 ) , ( mean 7 . 22 ± 3 . 28 IU/ml , range 3 . 73–12 . 62 IU/ml ) , which were well above the generally accepted protective threshold of 0 . 5 IU/ml . Antibody titers continued to increase until 28 days later ( day 0 , mean 11 . 47 ± 4 . 77 IU/ml , range 6 . 01–14 . 81 IU/ml ) . To verify if the efficacy of the vaccine would be affected by the simultaneous administration of anti-rabies VHH , an interference phenomenon which is well known for anti-rabies immune globulins [23] , a group of mice received besides the vaccine also a single dose of anti-rabies VHH in the same pre-exposure setting . In this regime , the first vaccination ( day-28 ) was accompanied by anti-rabies VHH ( Rab-E8/H7-ALB11 ) at a dose of 1 . 5 mg/mouse ( corresponding to 60 mg/kg , 392600 IU/kg ) , at the moment of the first vaccination ( day -28 ) . Vaccine ( IM ) and VHH ( IP ) were administered at separate sites . As reference groups mice were treated with anti-rabies VHH alone , or left untreated . In mice , the half-life of the anti-albumin VHH is approximately 1 . 5 days , hence the anti-rabies VHH will be removed from the circulation at the moment of viral challenge . Fig 1 shows that the rabies neutralization titers of mice that were injected with anti-rabies VHH , whether or not in combination with vaccination , were high 3 days after VHH administration ( day -25 , mean 88 . 28 ± 58 . 05 IU/ml , range 0 . 61–149 . 18 IU/ml ) . As expected , anti-rabies VHH titers rapidly declined over time with the clearance of the VHH from the blood ( day -20 , mean 9 . 43 ± 6 . 04 IU/ml , range 0 . 16–15 . 57 IU/ml ) and no detectable titers ( < 0 . 5 IU/ml ) on day 0 . Mice that received both vaccine and anti-rabies VHH had a mean titer of 5 . 69 ± 3 . 03 IU/ml ( range 1 . 73–9 . 37 IU/ml ) at day -20 , similar to mice that received vaccine alone , while at day 0 , antibody titers were significantly ( p<0 . 005 ) lower in the vaccine + VHH group ( mean 5 . 15 ± 3 . 38 IU/ml , range 0 . 37–10 . 03 IU/ml ) , compared to the vaccine only group . Mice were challenged by intranasal virus inoculation 4 weeks after the start of the vaccination ( day 0 ) . Fig 2 shows the survival curves of the vaccinated and control mice . Despite the fact that all mice had high neutralizing antibody titers at the time of challenge , only 50% was protected from disease and survived the challenge . In the remaining mice disease progression was delayed ( median survival time 27 days versus 9 days in control group ) . Disease signs in vaccinated mice were different compared to control mice , which typically develop signs of depression , such as unresponsiveness to stimuli and isolation from the group ( S1 Video ) . The vaccinated animals remained responsive to stimuli and aware of the environment , while developing ascending paresis , starting at the hind limbs , that gradually evolved into paralysis . Eventually , mice had to be euthanized because of severe paresis and paralysis ( S2 Video ) . The survival of mice that received the combination regime 4 weeks before viral challenge was substantially reduced compared to the mice that received only the vaccine ( 11% versus 50% ) . The median survival time of these mice was not significantly different from the control group ( 10 days versus 9 . 5 days ) , despite the presence of relatively high neutralizing antibody titers at the moment of challenge . As expected , survival rates of mice that received anti-rabies VHH were comparable to the control group . The presence of the anti-rabies VHH in the circulation hence seems to reduce the vaccine efficacy . This may indicate that in absence of the virus , the binding of the anti-rabies VHH to the vaccine may interfere with the induction of an effective humoral immune response . In previous in vivo studies , post-exposure treatment with the anti-rabies VHH one day after virus challenge was shown to provide protection from disease and death in a dose-dependent manner [15] . The same set-up was used to examine the efficacy of the combination of vaccine with a single anti-rabies VHH dose after exposure to the virus , which is the main indication for the use of vaccine in humans . Two different experiments were conducted . In a first experiment mice were treated with IP administered anti-rabies VHH ( Rab-E8/H7-ALB11 , 1 , 5 mg = 7852 IU/mouse ) and IM administered vaccine ( 0 . 25 AU/mouse ) , twenty-four hours after challenge with a lethal rabies dose . A second vaccine dose was administered 3 days after the first . This treatment was than compared to treatment with anti-rabies VHH at the same dose or the vaccine regimen alone . The anti-rabies VHH dose was the lowest effective dose in post-exposure treatment in previous studies [15] . Similar to the pre-exposure set-up , vaccinated mice received a second vaccination 3 days after the first dose . In the second experiment , the same vaccination schedule was applied , but instead of anti-rabies VHH , mice were treated with human rabies immune globulins ( HRIG , Berirab , IP , 1 ml/mouse = 121 . 50 IU/mouse ) at 24h after virus challenge . This is the highest volume and dose of the commercial HRIG product which could be administered to mice . Control mice were treated with HRIG alone . A schematic overview of both experiments can be found in Table 2 . The survival curves of the different treatment groups in the post-exposure prophylaxis setting are depicted in Figs 3 and 4 . In the post-exposure setting , the combination of vaccination with anti-rabies VHH rescued 60% of mice ( Fig 3 ) , significantly better than the treatment with anti-rabies VHH alone which rescued only 19% of mice . The vaccine by itself in the post-exposure setting did not provide any protection , and disease was similar to the control group . The median survival time was significantly longer after the combined treatment ( >35 days ) , compared to treatment with anti-rabies VHH ( 14 days , p<0 . 01 ) only , vaccine only ( 7 days , p<0 . 001 ) or the control group ( 8 days , p<0 . 001 ) . Mice that were treated with the combination of vaccine and HRIG did not survive challenge , similar to mice treated with HRIG alone . The median survival time of mice treated with vaccine and HRIG was 9 days and treatment with HRIG alone resulted in a median survival time of 10 days . The viral RNA load in the brain of mice was also assessed ( Fig 5 ) . Mice that received the PEP with vaccine and anti-rabies VHH had significantly lower viral RNA loads than control mice or mice treated with anti-rabies VHH only ( Fig 5 ) . Together these data show that in the post-exposure setting anti-rabies VHH acts synergistically with a standard vaccination regime to protect mice from disease after virus exposure . Post exposure prophylaxis ( PEP ) for rabies consists of a combination of passive ( human or equine immune globulins ) and active immunisation ( vaccine ) soon after exposure . Anti-rabies immune globulins are expensive , scarce and often not available or affordable for people in developing countries , that are typically most at risk [24;25] . Also in Western countries , RIG are increasingly difficult to procure [26] . Cheaper and easier-to-produce alternatives are needed . Previously , we developed anti-rabies VHH ( Nanobody ) capable of neutralizing virus at picomolar doses in vitro [15] . We also showed that post-exposure treatment with anti-rabies VHH only is capable of prolonging the incubation period of the disease in a dose dependent manner . In the current study , we evaluated whether post exposure treatment with the combination of anti-rabies VHH ( half-life extended Rab E8/H7-ALB11 ) and vaccine ( Rabipur , Novartis ) is better than single treatment with anti-rabies VHH or vaccine only . The combined treatment was tested using an intranasal challenge model of mice . Treatment was initiated at 24 hours after challenge . In humans , rabies can have incubation periods as short as 4–6 days , especially if the virus is deposited in highly innervated facial tissues , as is often the case in children [27] . Failure of classic PEP is described for several cases , often with short incubation periods or when highly innervated tissues were infected , which allows quick entry of the virus in nerves [28–30] . In order for PEP to be effective , it is believed that the virus needs to be intercepted by passive or active immune effectors before invasion of the central nervous system [28] . In case of a short incubation period , with rapid invasion of the nervous system , PEP cannot intercept the virus in time to prevent brain infection . Compared to anti-rabies VHH or vaccine alone , the combination therapy in a post-exposure setting significantly delayed the onset of disease , prolonged median survival time and decreased mortality . Sixty per cent of mice treated with anti-rabies VHH and vaccine survived the infection , in contrast to 0% with vaccine only and 19% with anti-rabies VHH only . This is in agreement with the observations from Servat et al . , who also showed that PEP with vaccine only was unable to prevent lethal disease [31] . Post-exposure treatment with anti-rabies VHH only proved more effective than vaccine only . This partial protection is in line with studies previously described by our group [15] . We assume that the synergy between vaccine and VHH lies in the fact that anti-rabies VHH can immediately delay the spread of the virus and prolong the incubation period , which allows more ( sufficient ) time for the active immune response to mount and control the infection in part of the mice . Indeed , treatment with VHH prolongs the incubation period from six to ten days , and the earliest antibody and cellular immune response can be expected as soon as seven days after intramuscular vaccination with an inactivated rabies vaccine [32] . This hypothesis also explains the limited efficacy of the combined treatment with vaccine and HRIG . Indeed , in the current and a previous study [15] , we found that administration of HRIG to mice after lethal challenge merely prolongs the median survival time by one or two days . This limited prolongation of the incubation period is probably not long enough to mount an effective immune response , able to control the virus infection before it becomes lethal . Our results indicate that an active antibody response was induced in all survivor mice , corresponding to low residual levels of viral RNA ( ΔCq ≤10 ) in the brain at the endpoint measurement ( 35 DPI ) . Pre-exposure treatment with vaccine ( IM ) and VHH ( IP ) seemed to partially reduce the immunogenicity of the vaccine , a phenomenon that is also described for the combination of RIG and vaccine [23;33;34] . Mice which received anti-rabies VHH in conjunction with vaccine developed significantly lower antibody titers 4 weeks later and were significantly less well protected against virus challenge . Indeed , whereas mice receiving vaccine only had a 50% survival rate and a delayed disease progression , only 11% of the mice treated with vaccine and anti-rabies VHH survived infection and no delay could be observed . These results were confirmed in independent experiments in which a pre-incubated mix of rabies virus and VHH was administered simultaneously at the same site ( S1 Fig ) . Antibodies can interfere with active immunization via different mechanisms . Most of the described mechanisms are Fc dependent , like inhibition of the B-cell responses by binding to the Fc-receptor , cross-linking of the B-cell receptor and the complement system , or antigen removal by macrophages [35] . Only humoral , and not cellular , immune responses seem to be affected by the presence of specific antibodies [36] . Since the used anti-rabies VHH is not a full antibody and lacks the Fc domain , it is unlikely that these mechanisms are involved [37] . The half-life extended anti-rabies VHH can interact with the neonatal Fc receptor through the intermediate of albumin , but it remains an unlikely mechanism since the non-HLE anti-rabies VHH , lacking an albumin-binding VHH component showed similar reduction of the vaccine efficacy ( S1 Fig ) . Therefore a likely mechanism could be epitope masking . By binding to the surface glycoproteins of the inactivated vaccine virus , the anti-rabies VHH might shield recognition of the epitopes by the immune system [36] . The fact that the combination of anti-rabies VHH with vaccine still proved superior in PEP , argues for the relative importance of immediate passive immunisation in PEP , especially when the virus has easy access to nerves or neuronal cells . Pre-exposure vaccination offered only partial protection upon intranasal virus challenge ( 50% survivors ) . Half of the mice that were actively immunized with ( inactivated ) vaccine , both at 28 and 25 days before challenge , still developed lethal brain infection . This incomplete protection , even with high antigenic doses ( 2 x 0 . 25 AU/mouse ) , is also described by other researchers , using similar models [18] . Nevertheless , the applied vaccine schedule resulted in clear seroconversion of all mice , with virus-neutralizing serum titers well above the protective threshold of 0 . 5 IU/ml ( range 6 . 01–18 . 04 IU/ml ) at the moment of challenge . Moreover , the challenge occurred at four weeks after the first vaccine administration , at the moment when the peak serological response can be expected [38;39] . The height of the neutralizing antibody titer in vaccinated mice did not correspond to the level of protection upon challenge . Some mice with titers up to 20 IU/ml still developed lethal disease . The incomplete protection in the post-exposure setting may be explained by the aggressive nature of the used intranasal challenge model , in which virus is inoculated directly on a site that contains a high concentration of olfactory neuronal cells , providing a direct portal of entry to the central nervous system . In earlier studies we found spread of the virus in the olfactory bulbs of the brain already at the first day after inoculation [15] . Once inside the central nervous system , the virus is protected from several systemic immune effectors , which may limit the protection by the vaccine [40;41] . We therefore assume that the mice that survived the challenge after preventive vaccination or PEP with anti-rabies VHH and vaccine were able to develop a cellular immune response , capable of controlling the infection in the brain . The intranasal challenge model is our preferred experimental model because of the high reproducibility , practicability , safety and animal wellbeing issues [16] . It may be that in an infection model with a longer incubation period and a more pronounced phase of peripheral virus replication in non-neuronal cells , preventive vaccination would be more effective , since vaccine-induced antibodies might be more effective to intercept virus spread between non-neuronal and neuronal cells . In our hands , intramuscular inoculation of rabies virus requires unnaturally high levels of virus in the inoculum ( >105−6 CCID50 ) and yields variable inter-assay results , limiting its use for experimental comparison of intervention strategies [16] . Another remarkable finding was the different clinical picture observed depending on the vaccination status of the mouse prior to virus challenge . Naïve mice typically showed signs of depression , such as isolation from the group , inactivity and unresponsiveness to stimulation . In contrast , pre-immunised mice remained alert and vivid , but developed ascending paresis , resulting in paralysis of all limbs , requiring euthanasia . Vaccinated mice developed disease after a longer incubation period ( 13 . 7 instead of 9 days ) and had a longer morbidity period ( 3 instead of 1 . 5 days ) , which resulted in a longer median survival time ( 27 instead of 9 days ) , compared to naïve mice . They also had lower viral loads in the brain at the peak of disease . The vaccine-induced immune response thus had a clear effect on pathogenesis and symptomatology . Iwasaki et al . also found that the host immune response has a clear impact on the development of , what they refer to as , either “encephalitic” or “paralytic” disease in mice . Rabies virus challenge in immunocompetent mice resulted in “paralytic disease” , with relatively low viral loads and a high extent of inflammation and damage in the brain . The same challenge in cyclophosphamide-treated mice resulted in the absence of an immune response and “encephalitic disease” , with severe general depression , only minor paralysis , high viral loads , and less neuronal cell damage [42] . In our study , the pre-immunized mice developed a disease pattern similar to the immunocompetent mice of Iwasaki et al . , whereas the naïve mice evolved comparably to the cyclophosphamide-treated mice . In human cases , the average survival time of paralytic rabies is twice as long , compared to the encephalitic ( furious ) form [42] . Patients with paralytic rabies typically remain fully conscious , while developing ascending motor weakness [43] . Also in dogs , paralytic rabies is associated with reduced viral load and more prominent inflammation [44] . Our observations further add to the evidence that paralytic rabies may be caused by an immuno ( patho ) logical response of the host to the virus infection . In humans , passive immunisation with anti-rabies antibodies is expected to bridge the immunity gap between virus exposure and onset of the active antibody production induced by the vaccine . The half-life extension of the anti-rabies VHH is based on the addition an anti-albumin VHH component . In mice , addition of anti-albumin VHH extends the half-life to 0 . 5–1 . 9 days [15] , while in humans it is extended up to 10–20 days [45] . It would therefore be feasible to formulate and dose anti-rabies VHH for humans to obtain protective levels ( > 0 . 50 IU/ml ) in the blood for 14 days , which would be sufficient for the active immune response to take over . Compared to ( human ) rabies immune globulins ( 150 IU/ml ) , VHH can be produced and formulated at very high potencies ( >6000 IU/ml ) . WHO recommends that rabies immune globulins are administered locally into the wound , however , due to the limited potency per ml of the rabies immune globulins , this is not possible for small wounds or injuries to nose , fingers or toes as it can cause compartment syndrome . VHH formulations containing high potencies per ml could overcome this problem and would be more suited for infiltration of the whole dose into small body parts . These results provide evidence for the possible use of anti-rabies VHH together with vaccine for post exposure prophylaxis of rabies . Early treatment with anti-rabies VHH can delay the incubation period of the disease , which allows more time for the vaccine-induced immunity to control the infection . The ease of production and high thermal stability of VHH are important advantages over the currently used anti-rabies immune globulins .	Rabies is an infectious disease causing 59 , 000 deaths and millions are exposed each year worldwide . Post-exposure prophylaxis ( PEP ) against rabies consists of a combination of passive ( immune globulins ) and active immunisation ( vaccine ) directly after viral exposure . Currently used plasma-derived anti-rabies immune globulins are expensive and scarce , urging the development of alternatives . Nanobodies or VHH are the smallest antigen-binding fragments of camelid heavy chain antibodies and are easy to produce with intrinsic good thermal stability and solubility . Combined treatment with anti-rabies VHH and vaccine gave significantly better protection than either compound alone in an intranasal rabies challenge model in mice , which validates the potential use of anti-rabies VHH as replacement of immune globulins in PEP .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "viral", "vaccines", "medicine", "and", "health", "sciences", "immune", "physiology", "pathology", "and", "laboratory", "medicine", "pathogens", "immunology", "tropical", "diseases", "microbiology", "vaccines", "preventive", "medicine", "animal", "models", "viruses", "rabies", "model", "organisms", "rna", "viruses", "neglected", "tropical", "diseases", "vaccination", "and", "immunization", "antibodies", "research", "and", "analysis", "methods", "rabies", "virus", "public", "and", "occupational", "health", "immune", "system", "proteins", "infectious", "diseases", "zoonoses", "proteins", "medical", "microbiology", "microbial", "pathogens", "mouse", "models", "globulins", "biochemistry", "lyssavirus", "virology", "viral", "pathogens", "physiology", "biology", "and", "life", "sciences", "viral", "diseases", "organisms" ]	2016	Post-exposure Treatment with Anti-rabies VHH and Vaccine Significantly Improves Protection of Mice from Lethal Rabies Infection
Viruses are masters of evolution due to high frequency mutations and genetic recombination . In spite of the significance of viral RNA recombination that promotes the emergence of drug-resistant virus strains , the role of host and environmental factors in RNA recombination is poorly understood . Here we report that the host Met22p/Hal2p bisphosphate-3′-nucleotidase regulates the frequency of viral RNA recombination and the efficiency of viral replication . Based on Tomato bushy stunt virus ( TBSV ) and yeast as a model host , we demonstrate that deletion of MET22 in yeast or knockdown of AHL , SAL1 and FRY1 nucleotidases/phosphatases in plants leads to increased TBSV recombination and replication . Using a cell-free TBSV recombination/replication assay , we show that the substrate of the above nucleotidases , namely 3′-phosphoadenosine-5′-phosphate pAp , inhibits the activity of the Xrn1p 5′-3′ ribonuclease , a known suppressor of TBSV recombination . Inhibition of the activity of the nucleotidases by LiCl and NaCl also leads to increased TBSV recombination , demonstrating that environmental factors could also affect viral RNA recombination . Thus , host factors in combination with environmental factors likely affect virus evolution and adaptation . Viruses with RNA genomes are abundant pathogens of plants and animals . Many RNA viruses have ultrafast replication cycles and rapid evolution , leading to the continuous emergence of new strains and variants . In addition to high frequency mutations and genome reassortments for multicomponent RNA viruses [1] , [2] , [3] , [4] , RNA recombination is one of the major driving forces in RNA virus evolution , helping viruses to invade new hosts , develop resistance against drugs and other antivirals and form more virulent strains [3] , [5] , [6] , [7] . Another benefit of RNA recombination is the increased fitness of viruses in some hosts [8] . However , natural selection pressure on the recombinant viruses could be a significant limiting force during their spread , keeping recombinant and parent viruses close to an “evolutionary optimum” level . RNA recombination also functions in the repair of viral RNA molecules by utilizing truncated/damaged viral RNA molecules [9] , [10] , [11] , [12] . The repair function of RNA recombination might compensate viruses for the high mutation rate , which could introduce detrimental mutations into the viral genomes , reducing the fitness of clonal viral populations [3] , [13] . Thus , RNA recombination can also be regarded as a guardian of the viral genome , and its second function is to increase genome variability . Viral RNA recombination leads to the joining of two or more noncontiguous segments of the same RNA or two separate RNAs together [14] . Recombination is thought to be a frequent event during the infectious cycles of some RNA viruses [5] , [6] . Most RNA recombination is based on template switching by the viral polymerase , as documented by in vitro approaches for a number of viruses [15] , [16] , [17] , [18] , [19] , [20] . In spite of the high frequency RNA recombination for some viruses , the detection of recombinant viral RNAs could be challenging since most of the recombinants are likely poorly adapted to their environment and therefore recombinant viral RNAs are eliminated rapidly from viral populations . Comparison of viral RNA genomes , however , reveals that recombination has shaped the evolution of many RNA viruses [6] . Studies on the viral replication proteins have revealed their roles in RNA recombination events and led to template-switching recombination model as the most widespread mechanism during recombination events [14] , [21] , [22] . Moreover , sequences/structures in the viral RNA could act as hot- or cold-spots in promoting or inhibiting viral RNA recombination , respectively [14] , [23] , [24] , [25] , [26] . Altogether , RNA recombination seems to be a dynamic and probabilistic event that shapes the population of viruses by contributing to virus variability , but also serving as a genome repair mechanism to maintain the infectivity of RNA viruses [3] , [14] . In spite of our increasing knowledge about viral RNA recombination over the last two decades that contributed new insights into the roles of viral proteins and the viral RNA in RNA recombination [14] , the roles of host proteins and environmental factors are poorly understood . Genome-wide screens of ∼5 , 500 yeast knock out and knock down strains and proteomics approaches with 4 , 100 purified yeast proteins involving Tomato bushy stunt virus ( TBSV ) , a tombusvirus infecting a wide range of plants , and yeast model host have revealed that several dozens of host genes could affect viral RNA recombination either directly or indirectly [27] , [28] , [29] , [30] , [31] . One of the most critical host factors identified is the cytosolic Xrn1p 5′-3′exoribonuclease ( Xrn4 in plants ) . This exoribonuclease can reduce the frequency of TBSV recombination by efficiently degrading RNA recombination intermediates [32] , [33] , [34] . Other ribonucleases , which are components of the RNA silencing pathway , also affect recombination of a fungal RNA virus , suggesting that ribonucleases might affect the evolution of a range of RNA viruses [35] , [36] . In this work , we tested the role of the previously identified MET22/HAL2 gene , whose deletion increased TBSV RNA recombination in yeast [30] . MET22 codes for a bisphosphate-3′-nucleotidase in the sulfate assimilation pathway involved in methionine biosynthesis and it affects salt tolerance [37] . Met22p removes the 3′ phosphate from 3′-phosphoadenosine-5′-phosphate ( pAp ) , thus producing AMP , as well as hydrolyzing 3′-phosphoadenosine 5′-phosphosulfate ( pApS ) [38] , [39] and is also active on other biphosphorylated nucleotides ( pNp ) [40] . Our working model was that deletion of MET22 might promote TBSV RNA recombination by leading to increased cytosolic level of pAp [41] and subsequent pAp-mediated inhibition of the ribonuclease activity of Xrn1p in yeast cells [37] . The reduced Xrn1p activity would then lead to increased TBSV recombination due to the elevated levels of short RNA recombination intermediates that are not degraded by Xrn1p efficiently in the presence of pAp [32] , [33] , [34] . To test this model , we complemented met22Δ yeast with Met22p mutants defective in bisphosphate-3′-nucleotidase function , which did not suppress TBSV recombinant RNA accumulation , suggesting that the enzymatic function of Met22p is important to inhibit TBSV recombination . In addition , inhibition of Xrn1p exoribonuclease via pAp in a cell-free TBSV replication assay demonstrated increased accumulation of TBSV recombination products as well as enhanced level of partial degradation products of TBSV replicon ( rep ) RNA , which are intermediates in RNA recombination [34] . Inhibition of the Met22p activity with either LiCl or NaCl also increased TBSV recombination , suggesting that environmental factors , such as salt stress , could affect viral RNA recombination . To confirm that Met22p affects TBSV recombinant ( rec ) RNA accumulation , we expressed Met22p from the weak galactose-regulatable GALS promoter from its original chromosomal location in BY4741 yeast ( Gals-met22 ) that also carried the plasmids for launching TBSV repRNA accumulation [42] , [43] . Culturing yeast for 22 hours in a media containing galactose led to suppression of recRNA accumulation by ∼15-fold when compared to Gals-met22 yeast cultured in a media containing glucose that represses the GALS promoter ( compare lanes 1 and 5 , Fig . 1 ) [44] . Interestingly , the accumulation of partly degraded TBSV repRNAs , named degRNAs [32] , [34] , was also suppressed by ∼5-fold in Gals-met22 yeast grown for 22 hr in the presence of galactose . These degRNAs represent 5′-truncated TBSV repRNAs ( shown schematically in Fig . 1 ) [30] , [34] . The accumulation levels of recRNAs and degRNAs in Gals-met22 yeast grown for 22 hr in the presence of galactose were only a little bit higher than the levels of recRNAs and degRNAs observed in the wt BY4741 yeast expressing Met22p from its original promoter ( compare lanes 5 and 6 , Fig . 1 ) , suggesting that MET22 is responsible for affecting the generation and accumulation of TBSV recRNAs and degRNAs in yeast cells . To test if the bisphosphate-3′-nucleotidase activity of Met22p is important for TBSV recRNA accumulation , we complemented met22Δ yeast with various Met22p mutants expressed from plasmids as shown in Fig . 2A . We found that expression of Met22p with mutations in the critical signature motif ( MutA , Fig . 2A ) or C-term truncated version of Met22p removing the metal-binding site required for binding to the essential Mg2+ ion ( MutD ) [38] resulted in lack of complementation , thus high TBSV recRNA level , in met22Δ yeast when compared with the expression of the wt Met22p ( lanes 5–6 and 11–12 versus 15–16 , Fig . 2B ) . In contrast , Met22p with mutations within a nonessential N-terminal segment ( MutB , Fig . 2A ) was able to efficiently suppress TBSV recRNA accumulation in met22Δ yeast ( lanes 7–8 , Fig . 2B ) . Altogether , these complementation data suggest that the bisphosphate-3′-nucleotidase function of Met22p is important for the RNA recombination suppressor activity of Met22p in yeast . The major function of Met22p bisphosphate-3′-nucleotidase in yeast cells is the removal of pAp and pApS products of the sulfate assimilation pathway , which are known inhibitors of Xrn1p 5′-3′ exoribonuclease [37] , [39] . Interestingly , Xrn1p has been shown to decrease the stability of TBSV RNAs and suppress TBSV RNA recombination [30] , [32] , [33] , [34] . Therefore , it is possible that deletion of MET22 might promote TBSV RNA recombination by leading to an ∼80-fold increase in accumulation of pAp [41] and subsequent pAp-mediated inhibition of the ribonuclease activity of Xrn1p in yeast cells [37] . To test this model , we estimated the half-life of TBSV RNAs in met22Δ yeast . Indeed , the stability of TBSV repRNA increased by ∼3-fold in met22Δ when compared with the wt yeast ( Fig . 3A , lanes 6–10 versus 1–5 ) . The increased half-life for TBSV repRNA is in agreement with the possible inhibition of Xrn1p activity . In addition , the double-deletion ( met22Δ xrn1Δ ) yeast supported increased level of recRNA accumulation ( by 26-fold , Fig . 3B , lanes 1–3 ) when compared with BY4741 ( see Fig . 1 , lane 6 ) , similar to the high recombination rate in single-deletion met22Δ yeast or in xrn1Δ yeast ( Fig . 3B ) . The profile of TBSV degRNAs accumulating in these yeasts suggest that the double-deletion strain is more similar to xrn1Δ than to met22Δ yeasts since met22Δ xrn1Δ yeast strain accumulates mostly the longer degRNA1 product ( Fig . 3B ) . It is proposed that the degRNA1 product is due to a cleavage by an endoribonuclease [34] . On the contrary , met22Δ yeast accumulates mostly the shorter degRNA2 product , suggesting that a limited 5′-to-3′ degradation of degRNA1 by the incompletely inhibited Xrn1p nuclease still takes place in met22Δ yeast to give rise to degRNA2 . Also , over-expression of Met22p in xrn1Δ yeast did not result in decreased level of recRNA accumulation ( not shown ) , unlike when Met22p was expressed in the met22Δ yeast strain ( Fig . 2B , lanes 15–16 ) . Altogether , these data support the model that MET22 and XRN1 are both inhibitors of TBSV recombination and they act in the same pathway . Since it has been documented that salt-stress inhibits the activity of Met22p [37] , [45] , we tested the accumulation of recRNAs in BY4741 yeast treated first with various amounts of LiCl . We found that 20 and 40mM LiCl increased TBSV recRNA levels by ∼20 and ∼80-fold for DI-72 repRNA ( Fig . 4A–B ) and by up to 120-fold for the recombinogenic DI-AU-FP repRNA ( Fig . S1 ) . Since the accumulation of degRNA2 also increased remarkably in the LiCl-treated yeast ( Fig . 4A–B ) , it is likely that the observed effect of LiCl is due to its inhibition of the Met22p-Xrn1p pathway . To obtain evidence that the above LiCl treatment indeed affects the activity of cellular 5′-3′ exoribonucleases , such as Xrn1p ( cytosolic ) and Rat1p ( nucleus ) , we tested the accumulation of the nondegraded ITS1 region of pre-ribosomal RNA ( Fig . S2 ) [37] . As expected , LiCl treatment increased the accumulation of pre-ribosomal RNA carrying the ITS1 region by ∼7-fold ( Fig . S2 ) , which is indicative of reduced level of Xrn1p and Rat1p nuclease activities in yeast cells . Second , we tested the effect of NaCl treatment of yeast cells and found ∼7-fold increase for TBSV recRNA levels ( Fig . 4A , C ) . The accumulation of degRNA2 also increased by ∼25-fold in the NaCl-treated yeast ( Fig . 4A , C ) , suggesting that degradation of TBSV RNAs is decreased by NaCl due to inhibition of the Met22p-Xrn1p pathway . One of the advantages of studying viral RNA recombination and replication with TBSV is the availability of a yeast-based cell-free ( CFE ) assay capable of supporting the in vitro assembly of the viral replicase complex , including one full replication cycle of the TBSV repRNA [31] , [46] , [47] . Inhibition of the endogenous Met22p and Xrn1p present in the CFE obtained from wt BY4741 yeast by 60 mM LiCl and 5 mM pAp led to ∼3-fold increase of both repRNA and recRNA accumulation ( Fig . 5 , lanes 1 versus 4 ) . The amount of degRNA also increased by ∼2-fold . However , adding purified recombinant Xrn1p to the above CFE containing the inhibitors , led to ∼10-fold inhibition of recRNA accumulation ( Fig . 5 , lanes 4 versus 6 ) , while the accumulation of repRNA and degRNA decreased by ∼4 and ∼5-fold , respectively . When compared with the control sample , 60 mM LiCl and 5 mM pAp inhibitors did decrease the suppressor activity of the exogenous Xrn1p by ∼3-fold in TBSV recombination and replication ( compare lanes 5 with 2 , Fig . 5 ) . Altogether , the in vitro data strongly support the role of LiCl and pAp in TBSV RNA recombination and replication by inhibiting the recombination suppressor activity of Xrn1p . To test if plants have a pathway similar to Met22/Xrn1 pathway in yeast that can affect recombination of TBSV , first we used LiCl and pAp inhibitors in Nicotiana benthamiana protoplasts electroporated with DI-ΔRI repRNA that lacks the 5′ terminal 169 nt from the wt DI-72 repRNA and can recombine efficiently in plant protoplasts and yeast [31] , [32] , [33] . We found that LiCl treatment increased TBSV recRNA and degRNA accumulation by ∼3- and ∼1 . 5-fold , respectively , after 24 hours of incubation ( Fig . 6 , lanes 5 versus 1 ) , while pAp treatment alone had no significant effect on TBSV recRNA and degRNA accumulation ( lane 6 ) . However , the largest stimulatory effect on TBSV recRNA and degRNA accumulation was obtained by the combined use of LiCl and pAp , leading to ∼4 . 5- and 2 . 5-fold increase , respectively ( lane 10 ) . Overall , the data from protoplasts suggest that plant cells also have a Met22/Xrn1-like pathway that is inhibited by LiCl and pAp , thus resulting in increased level of TBSV recombination . To examine if a plant nucleotidase analog of the yeast MET22 gene can also affect TBSV RNA recombination , first , we expressed the Arabidopsis AHL nucleotidase/phosphatase gene [48] in met22Δ yeast . Interestingly , AtAHL reduced the accumulation of TBSV recRNA and degRNA by 5- and 10-fold , respectively ( lanes 13–14 , Fig . 2B ) , confirming that a plant analog of the yeast MET22 gene can also suppress TBSV recombination . To test if silencing of the AHL gene in N . benthamiana could influence TBSV recombination , we agroinfiltrated N . benthamiana leaves with plasmids expressing Cucumber necrosis virus ( CNV ) , which can be used as a helper tombusvirus , and the highly recombinogenic TBSV DI-AU-FP RNA after knocking down the level of NbAHL mRNA via gene silencing ( Fig . 7B ) . The accumulation of TBSV recRNAs was increased by ∼3-fold in the agroinfiltrated leaves of the NbAHL knockdown plants ( Fig . 7A ) , which is less than ∼8-fold increase observed in XRN4 ( the homolog of the yeast XRN1 ) [32] knockdown plants . Knocking down the expression of NbAHL did not affect the growth of N . benthamiana , while the XRN4 knockdown plants showed some stunting ( Fig . 7C ) . To further test TBSV recombination in AHL knockdown plants , we used agroinfiltration with plasmids expressing DI-ΔRI repRNA in combination with the CNV helper virus to launch replication in the silenced leaves . Subsequent analysis of TBSV RNA levels revealed that the levels of recRNAs and degRNAs were increased by ∼22- and ∼9-fold , respectively ( Fig . 7D ) . Thus , two different TBSV repRNAs showed high frequency recombination in plants silenced for AHL nucleotidase/phosphatase , confirming that plant AHL plays a comparable role in TBSV recombination to the yeast MET22 nucleotidase . Since there are at least three Met22-like nucleotidases in Arabidopsis , such as AHL , SAL1 and FRY1 [49] , [50] , we decided to knockdown the expression levels of all three genes simultaneously . The accumulation of the CNV helper virus increased by ∼3-fold in the triple-nucleotidase gene knockdown plants , which died ∼2–3 days faster than the control plants after co-agroinfiltration with plasmids expressing both CNV helper virus and the TBSV DI-AU-FP repRNA ( Fig . 8A ) . The uninoculated triple knockdown plants showed slight stunting , but the individual leaves were actually larger than the leaves of the control plants treated with the “empty\ silencing vector ( Fig . 8B ) . The accumulation of TBSV recRNAs and degRNAs increased ∼3- and 4-fold , respectively , in the DI-AU-FP repRNA inoculated leaves of triple gene knockdown plants ( Fig . 8C ) , suggesting that recombination was comparable in the AHL-knock down and the triple-nucleotidase gene knockdown plants . It is likely that there is still some residual nucleotidase activity in the triple gene knockdown plants , therefore , we infiltrated 200 mM LiCl to leaves to further inhibit the nucleotidase activity in the triple-nucleotidase gene knockdown N . benthamiana plants . Interestingly , we observed ∼8-fold increase in TBSV recRNA and ∼4-fold increase in degRNA accumulation in the triple gene knockdown plants infiltrated with LiCl when compared to triple gene knock down plants infiltrated with water control ( Fig . 8D , compare lanes 9–17 with 18–23 ) . When compared to the control nonsilenced and untreated plants replicating CNV helper and the TBSV DI-ΔRI repRNA , the accumulation of TBSV recRNAs and degRNAs increased by ∼15- and 17-fold , respectively , in the triple gene knockdown plants infiltrated with LiCl ( Fig . 8D , compare lanes 9–17 with 1–8 ) . The accumulation of the CNV helper virus also increased by ∼5-fold in the triple gene knockdown plants infiltrated with LiCl ( Fig . 8D ) . Overall , these data strongly support the role of plant AHL , SAL1 and FRY1 nucleotidases/phosphatases in tombusvirus recombination , replication and viral RNA degradation . Viral RNA recombination plays a major role in virus evolution [3] , [5] , [6] . In spite of the possible significance , we know little about the roles of host and environmental factors in viral RNA recombination [51] . In this work , using TBSV and yeast as a model host , we demonstrate that Met22/Xrn1 pathway and environmental factors affecting this pathway , namely salt-stress caused by LiCl and NaCl , plays a role in viral RNA recombination . In vitro experiments with a cell-free extract from yeast revealed that the combined use of LiCl , an inhibitor of Met22p bisphosphate-3′-nucleotidase , and pAp , an inhibitor of Xrn1p 5′-3′ exoribonuclease , could promote TBSV RNA recombination ( Fig . 5 ) . Since there is only a single cycle of RNA replication in the CFE , the fact that RNA recombinants accumulate at a detectable level in vitro suggests that LiCl and pAp are potent inducers of viral RNA recombination . These compounds also reduce the complete degradation of the viral RNA and increase the accumulation of the original repRNA by inhibiting the activity of the Xrn1p ribonuclease in the CFE . Xrn1p ribonuclease is a major enzyme controlling degradation of the tombusvirus RNA , which is uncapped at the 5′ end [52] . Thus , inhibition of the activity of the 5′-3′ exoribonuclease leads to increased levels of partially degraded TBSV RNA products , which then could affect ( i ) the frequency of RNA recombination by serving as intermediate templates during recombination events [30] , [31] , [34] , ( ii ) facilitate the formation of defective interfering RNAs [52] , and ( iii ) possibly alter the fitness of viral populations . Genetic experiments in yeast model host also supported that MET22 affects viral RNA recombination via XRN1 . For example , deletion of MET22 increased the half-life/stability of the TBSV RNA by three fold , suggesting that the activity of the Xrn1p ribonuclease , the major factor involved in TBSV RNA degradation in yeast [32] , [33] , [34] , is inhibited via the pAp substrate of Met22p [37] , [39] . Moreover , the double deletion ( met22Δ xrn1Δ ) strain behaved as the single deletion ( xrn1Δ ) strain in the TBSV recombination assay ( Fig . 3 ) . Also , the profile of the partially degraded viral RNA products in the single and double deletion strains was similar ( Fig . 3 ) . Therefore , we propose that Met22p is a suppressor of TBSV recombination via its regulatory function of Xrn1p activity . We also provide evidence that comparable pathway to the Met22/Xrn1 pathway of yeast also regulates TBSV RNA recombination in plant cells . Addition of LiCl to the N . benthamiana protoplast media or the combined use of LiCl and pAp both resulted in increased TBSV RNA recombination and led to higher levels of partially degraded TBSV RNAs ( Fig . 6 ) . Also , expression of the Ahl nucleotidase/phosphatase from Arabidopsis , a yeast MET22 analog , suppressed TBSV RNA recombination and decreased the accumulation of the partially degraded viral RNAs in met22Δ strain . Silencing of the expression of AHL gene alone , or triple-gene silencing of AHL , SAL1 and FRY1 nucleotidase/phosphatases in N . benthamiana increased the level of TBSV RNA recombinants in plant leaves . The most pronounced increase in accumulation of TBSV RNA recombinants was seen in the triple-gene silenced plants treated with LiCl , suggesting that the combined effect of genetic and environmental factors could be critical in regulation of the rate of viral RNA recombination . Based on the known biochemical functions of Met22p , as well as the presented in vitro and in vivo results on TBSV recombination , we propose that Met22p ( AHL , SAL1 and FRY1 nucleotidases in N . benthamiana ) regulates TBSV RNA recombination and degradation of TBSV repRNA via affecting pAp level in cells ( Fig . 9 ) . Moreover , Met22p also affects the level of other biphosphorylated nucleotides ( pNps , such as pCp , pGp , pTp , pUp and pIp ) generated by various pathways in cell [40] , raising the possibility that other pNps might also affect viral RNA recombination . In the presence of active Met22p in cells , pAp/pNp level is low , thus allowing high activity of Xrn1p ribonuclease , which in turn , removes the partially degraded TBSV RNA products from cells [32] , [34] . The partially degraded TBSV RNA products are generated by an unidentified endoribonuclease ( s ) and rapidly degraded by Xrn1p exoribonuclease in the parental BY4741 yeast [34] . However , as shown earlier , the accumulation of the partially degraded TBSV RNA products in xrn1Δ yeast leads to high frequency recombination due to the use of these degRNAs in the recombination events by the viral replicase [32] , [34] . We show that deletion of MET22 likely prevents the rapid and complete degradation of repRNAs due to the accumulation of pAp substrate of Met22p , which inhibits the activity of Xrn1p 5′-3′ exoribonuclease [37] , [39] . Therefore , the high level of TBSV degRNAs in met22Δ cells will promote high frequency RNA recombination as well as decrease the rate of their degradation . In addition , environmental factors , such as LiCl and NaCl causing salt-stress , could affect TBSV RNA recombination by inhibiting the activity of Met22p in yeast cells ( Fig . 9 ) . Overall , our results suggest that environmental and host factors could play an important role in viral RNA recombination and evolution . S . cerevisiae strains BY4741 ( MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0 ) , met22Δ , gcn4Δ , and xrn1Δ were obtained from Open Biosystems . To express wt and mutated Met22p proteins in met22Δ yeast , we made pYES-Met22 plasmid expressing a 6xHis-Met22p ( tag is present at the N- terminus ) under the control of a GAL1 promoter . The cDNA of MET22 was amplified with primers #2177 and #2178 ( Table S1 ) and cloned at the KpnI/XhoI sites of pYES2/NT . To obtain pYES-MetA and pYES-MetB , expressing mutants MetA and MetB ( Fig . 2A ) , the MET22 sequence was amplified with primers #2584/#2178 and #2585/#2178 , respectively , and the PCR products were cloned to either BamHI/XhoI linarized pYES-Met22 or EcoRI/XhoI linarized pYES-Met22 . pYES-MetC and pYES-MetD , expressing mutants MetC and MetD , were obtained by PCR with primers #2586/#2177 and #2587/#2177 , respectively , and the PCR products were digested with KpnI/XhoI and cloned into KpnI/XhoI linarized pYES plasmid . For generating the expression vector pYES-AHLAt , the cDNA sequence of AHL was generated from Arabidopsis total RNA extract using RT-PCR with primers #2588/#2589 and then the RT-PCR product was cloned to the BamHI/XhoI site of pYES2/NT . Yeast strain met22Δxrn1Δ was obtained by homologous recombination of yeast strain xrn1Δ ( Open Biosystems ) with the PCR product of primers #2581/#2590 and pFA6-hphNT1 [44] as a template for hphNT1 selection . Correct deletion of the MET22 gene was verified by PCR with primers #2501/#2591 . Yeast strain Gals-met22 was obtained based on homologous recombination . Briefly , the BY4741 yeast strain was transformed with the PCR product made by using primers #2581/#2583 and pYM-N30 [44] as a template for kanMX4 selection . Correct integration of the GALS sequence in strain Gals-met22 strain was verified with primers KanB ( Open biosystems ) /#2591 . Yeast strains were co-transformed with indicated plasmids or PCR products by using the lithium acetate/ssDNA/polyethylene glycol method [53] , and transformants were selected by complementation of auxotrophic markers . The yeast strain Gals-met22 was co-transformed with pGBK-His33/DI72/CUP1 and pGAD-His92/CUP1 [27] , [31] . The transformed yeast strains were grown at 29°C in SC-UHL ( synthetic complete media without uracil , histidine and leucine ) with 2% glucose containing 50 µM CuSO4 until reaching cell density of 0 . 6 OD600 . Then yeast cultures were resuspended in SC-UHL with galactose medium containing 50 µM CuSO4 . Yeast was grown at 29°C for indicated timepoints before collecting for Northern analyses . For the complementation study ( Fig . 2 ) , yeast was co-transformed with pGBK-His33/DI72/CUP1 , pGAD-His92/CUP1 and the indicated Met22p mutants in pYES . The transformed yeast strains were pre-grown at 29°C in SC-UHL with 2% glucose until reaching cell density of 1 . 0 OD600 . Then yeast cultures were diluted to 0 . 1 OD600 in SC-UHL with galactose medium containing 50 µM CuSO4 to launch TBSV repRNA replication and to co-express the Met22p mutants . Yeast was grown at 29°C for 24 hours before sample collection for analyses . The statistical analysis was performed using AVEDEV ( average of absolute deviations of data points ) program in Microsoft Excel ( version 2008 for Mac ) based on 12–24 independent samples . For the RNA stability studies ( Fig . 3A ) , yeast strains BY4741 and met22Δ were transformed with pYC2-DI72 [43] . The transformed yeast strains were grown at 29°C in SC-U ( synthetic complete without uracil ) with 2% galactose . After 20 h , the cultures were re-suspended in SC-U supplemented with 2% glucose and collected after indicated time-points . For the analysis of TBSV repRNA replication and recombination in met22Δxrn1Δ , xrn1Δ and met22Δ yeast strains , they were co-transformed with pGBK-His33/CUP1 [28] , pGAD-His92/CUP1 and pYC2-DI72 . The transformed yeast strains were pre-grown at 29°C in SC-UHL supplemented with 2% glucose until reaching cell density of 1 . 0 OD600 . The pre-grown yeast cultures were diluted to 0 . 1 OD600 in SC-ULH medium supplemented with 2% galactose and 50 µM CuSO4 to launch TBSV repRNA replication . After 6h at 23°C , the cultures were collected and re-suspended in SC-UHL supplemented with glucose and 50 µM CuSO4 . The yeast cultures were grown for additional 18 hours at 23°C before sample collection for analyses . For the LiCl treatment ( Fig . 4 ) , yeast strain BY4741 was co-transformed with pGBK-His33/CUP1 , pGAD-His92/CUP1 and pYC2-DI72 or pYC2-DI-AU-FP , respectively . For the NaCl treatment ( Fig . 4 ) , we used yeast strain gcn4Δ , because it has been shown that NaCl treatment increases indirectly the activity of GCN4 transcription factor , which then upregulates the expression of MET22 [45] . The transformed yeast strains were pre-grown at 29°C in SC-UHL supplemented with 2% glucose until reaching cell density of 1 . 0 OD600 . The pre-grown yeast cultures were diluted to 0 . 1 OD600 in SC-UHLM+A ( synthetic complete without uracil , histidine , leucine , methionine , supplemented with 5g/l ammonium sulphate ) with 2% galactose and 50 µM CuSO4 to launch TBSV repRNA replication and the indicated amount of LiCl or NaCl ( Fig . 4 , and S1 ) . After 6h culturing , yeast cells were collected and re-suspended in SC-UHLM+A supplemented with glucose , 50 µM CuSO4 and LiCl or NaCl . The yeast cultures were grown for additional 18 hours before sample collection for analyses . T7 transcripts of DI-ΔRI and CNV RNAs were obtained as described previously [34] . Isolation , electroporation and culturing N . benthamiana protoplasts was done as described [54] . Briefly , 6 µg DI-ΔRI RNA together with 5 µg CNV genomic RNA transcripts were used for co-electroporation of 30×105 N . benthamiana protoplasts in 1ml electroporation buffer [54] . After electroporation , protoplasts were resuspended in 5 ml SP ( 34 . 2 g/l sucrose , 0 . 58 g/l MES , 72 . 8 g/l mannitol , pH 5 . 8 ) with indicated amounts of LiCl and pAp ( adenosine 3′-5′ biphosphate , Sigma ) . Protoplasts were incubated in the dark for 20 h at room temperature followed by RNA extraction and Northern blot analysis as described previously [54] . The virus-induced gene silencing ( VIGS ) assay was described previously [31] , [55] , [56] . The N . benthamiana sequences were obtained by a BLAST search using the Arabidopsis thaliana AHL , SAL1 and FRY1 gene sequences in Solanaceae Genomics Resource mRNAs database from the J . Craig Venter Institute . NbAHL correspondents to clone EB432053 , NbSAL1 to TA11598 and NbSAL2 to BP135480 , respectively . The VIGS vectors , pTRV2-AHLNb , pTRV2-Sal1Nb and pTRV2-FRY1Nb , were obtained by RT-PCR with primers #2191/#9192 for NbAHL , #2935/#2937 for NbSAL1 and primers #2932/#2933 for NbFRY1 . Nine days after the VIGS treatment ( using one or all three VIGS vectors together with pTRV1 ) [31] , [56] , the level of NbAHL mRNA was determined by RT-PCR with primers #2940/#2941 . We used the level of tubulin mRNA as a control by RT-PCR using primers #2859/#2860 . Subsequently , the silenced leaves were co-agroinfiltrated with pGD-CNV and pGD-ΔRI , or pGD-CNV and pGD-DI-AU-FP . One day after agroinfiltration , selected leaves were infiltrated with 200mM LiCl . Leave samples were collected of the agroinfiltrated leaves four days after agroinfiltration , followed by total RNA extraction and Northern blot analysis as described [32] . TAP-tagged purified Xrn1p [34] or Xrn1p-His ( from A . W . Jonson ) [57] was added to yeast cell-free extract , which were programmed with 1 µg of DI-72 or DI-ΔRI repRNAs as described [47] . The in vitro replication assay was performed for 4 hours at 25°C . Total RNA was extracted and loaded on a 5% polyacrylamide gel ( PAGE ) containing 8 M urea . The 32P-labeled bands were imaged with a Typhoon 9400 Imager ( GE Healthcare ) [47] . TBSV RNA replication and recombination was analyzed using total RNA obtained from yeast or plants . Northern blot analysis were performed as described previously [33] . Briefly , for detection of DI-72 repRNA and its derivatives , including recRNAs , we prepared a 32P-labeled region III+IV probe with T7 transcription using PCR amplified DNA obtained with primers #2754 and #2755 and pYC-DI72 as template . Northern blots were imaged with Typhoon ( GE Healthcare ) and analyzed by the ImageQuant program . Quantification was performed and the recRNA2 or degRNA2 levels were calculated in comparison to the amount of repRNA in each sample . Also , the repRNA measurements were normalized based on the ribosomal RNA levels in each sample . To detect the CNV genomic RNA , we made a 32P-labeled complementary RNA probe with T7 transcription from PCR products obtained with primers #312/#22 ( 3′ end ) or #1660/#20 ( 5′-3′ end ) . An RNA probe was obtained for detection of region I of DI-72 ( using primers #20/#15 in PCR ) . To detect MET22 mRNA , we made a 32P-labeled complementary RNA probe obtained by T7 transcription from a PCR product using primers #2177/#2200 .	Viral RNA recombination plays a major role in virus evolution . Yet , we know little about the roles of host and environmental factors in viral RNA recombination . In this work , using TBSV and yeast as a model host , we show that MET22 nucleotidase suppresses viral RNA recombination . In vitro experiments with a cell-free extract from yeast revealed that the substrate of Met22p bisphosphate-3′-nucleotidase , namely pAp , could promote TBSV RNA recombination via inhibiting the activity of the Xrn1p 5′-3′ ribonuclease , a known suppressor of viral RNA recombination . Altogether , we demonstrate that the Met22/Xrn1 pathway and environmental factors affecting this pathway , namely salt-stress caused by LiCl and NaCl , play a role in viral RNA recombination . The authors also provide evidence that a similar pathway affects TBSV recombination in plants as well . The most pronounced increase in accumulation of TBSV RNA recombinants was seen in the triple-nucleotidase gene silenced plants treated with LiCl , suggesting that the combined effect of genetic and environmental factors could be critical in regulation of the rate of viral RNA recombination .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "virology/viral", "replication", "and", "gene", "regulation", "virology/virus", "evolution", "and", "symbiosis", "virology" ]	2010	The Combined Effect of Environmental and Host Factors on the Emergence of Viral RNA Recombinants
Body size in Drosophila larvae , like in other animals , is controlled by nutrition . Nutrient restriction leads to catabolic responses in the majority of tissues , but the Drosophila mitotic imaginal discs continue growing . The nature of these differential control mechanisms that spare distinct tissues from starvation are poorly understood . Here , we reveal that the Ret-like receptor tyrosine kinase ( RTK ) , Stitcher ( Stit ) , is required for cell growth and proliferation through the PI3K-I/TORC1 pathway in the Drosophila wing disc . Both Stit and insulin receptor ( InR ) signaling activate PI3K-I and drive cellular proliferation and tissue growth . However , whereas optimal growth requires signaling from both InR and Stit , catabolic changes manifested by autophagy only occur when both signaling pathways are compromised . The combined activities of Stit and InR in ectodermal epithelial tissues provide an RTK-mediated , two-tiered reaction threshold to varying nutritional conditions that promote epithelial organ growth even at low levels of InR signaling . Cellular and organ growth ( anabolism ) in animals is regulated by complex interactions of nutritional and hormonal cues . As accumulation of cell mass usually precedes cell division , cellular growth is intimately coupled to proliferation and net organ growth . In all eukaryotes studied , the evolutionarily conserved protein complex TORC1 ( target of rapamycin complex 1 ) integrates nutritional and hormonal cues and translates this information into cellular growth and proliferation . When sufficient ATP and amino acids are present , TOR kinase directly phosphorylates S6K and 4E-BP and stabilizes Myc to promote the activity of the protein translation machinery , thereby permitting protein production and cell growth [1] , [2] . The amino-acid-sensing machinery is located on a late endosomal compartment , where a small GTPase heterodimer consisting of RagA/B/GTR1 and RagC/D/GTR2 , together with Rheb , is required to stimulate TORC1 activity upon amino acid stimulation [3]–[5] . In animals , complex hormonal regulation is layered upon the permissive cellular nutrient sensing to ensure coordinated tissue growth . Results from genetically tractable models revealed that the Insulin/Insulin Growth Factor ( IGF ) ligands and receptors are the principal organ growth regulators coupled to nutrition [1] , [6] , [7] . Drosophila Insulin-like peptides ( dILPs ) signal through an evolutionarily conserved growth promoting pathway initiated by binding of the adaptor proteins , Chico and Lnk , to the intracellular domain of the InR [8] , [9] . Subsequent recruitment of the Phosphatidylinositol-kinase class I ( PI3K-I ) leads to recruitment of Akt and PDK1 kinases to the plasma membrane through their respective phosphatidylinositol 3 , 4 , 5 trisphosphate ( PIP3 ) -lipid binding pleckstrin homology ( PH ) domains . PDK1 phosphorylates and activates Akt [10]–[12] , which in turn activates TORC1 by inhibiting its negative regulators PRAS40 and TSC1/TSC2 [13]–[16] . Suboptimal nutrient conditions during both animal development and homeostasis can be compensated for by the differential control of growth and catabolism in different organs . The molecular mechanisms underlying these tissue-specific responses in vivo are only beginning to be elucidated . Shortage of amino acids is sensed by the larval fat body resulting in the activation of autophagy and of an unknown relay signal reducing systemic dILP levels and growth [1] , [17]–[20] . Prolonged starvation halts the growth of most polyploid larval tissues including the gut , fat body , and epidermis [21] . Strikingly , the imaginal tissues that will make up the adult fly during metamorphosis continue to grow even when amino acid levels in the hemolymph drop [21] , [22] . A recent elegant study revealed that the continued cycling of neuroblasts in the Drosophila brain , when insulin signaling is low , is supported by growth signaling from the RTK , ALK ( Anaplastic Lymphoma Kinase ) . ALK bypasses the TORC1 requirements for growth but still acts through the direct TORC1 targets , 4E-BP and S6K [22] , [23] . Stitcher ( Stit ) is a Ret-like , receptor tyrosine kinase ( RTK ) activated upon epidermal wounding in Drosophila embryos . Stit signaling induces the transcriptional activation of genes involved in barrier repair and the coordinated cytoskeletal rearrangements leading to epidermal wound closure [24] . Here , we report that Stit also promotes growth in the Drosophila epithelial imaginal wing discs . We show that in these organs , Stit induces growth in parallel to InR , but also suppresses autophagy upon low InR signaling through the canonical PI3K-I/TORC1 pathway . Thus , the Stit and InR RTKs ensure a two-threshold response to TORC1 activity in proliferating epithelial tissues , increasing their repertoire of reactions to nutrient stress . stit mutants die as pupae with melanized abdomens , and Stit protein was detected in several developing imaginal epithelial tissues ( the wing , thorax , leg discs , and abdominal histoblasts ) , but was absent in eye-antennal disc ( Figure S1A , B and unpublished data ) [24] . To uncover a potential Stit function in epithelial tissue development , we focused on the wing imaginal discs . The larval wing discs generate a dorsal and a ventral epithelial cell layer , which appose each other during pupation to form the adult wing blade . We first made FRT stitExel9056 and stitEx266 mutant clones in the future dorsal surface of the wing using apterous ( ap ) -GAL4 to drive UAS-FLPase . This resulted in strong upwardly bent wings with no apparent defects in vein patterning or hair orientation . This phenotype suggested that Stit controls the balanced growth of the dorsal and ventral wing compartments ( Figure 1A ) . As stit mutant cells showed residual immunoreactivity during larval stages ( Figure S1A ) , we additionally used a transgene encoding a kinase defective variant of Stit ( StitKD ) to acutely disrupt endogenous Stit as well as stit RNAi-expressing trangenes ( stit-IR ) . Broadly expressed StitKD by daughterless ( da ) -GAL4 caused pupal lethality resembling the stit mutant phenotype . Expression of StitKD or stit-IR in the dorsal portion of the wing using either ap-GAL4 or MS1096-GAL4 reproducibly caused upwardly bent wings , but the bending was more severe than that caused by the mutant clones ( Figures 1A and 2A ) . Similarly , expression of StitKD and stit-IR using engrailed ( en ) -GAL4 resulted in bending of the posterior wing part , suggesting that Stit coordinates shape and growth in the entire tissue ( Figure 1D and not shown ) . Overexpression of wild-type stit ( ap>stit ) could rescue the wing phenotypes caused by either StitKD or stit-IR ( unpublished data ) , suggesting that both StitKD and stit-IR can act as potent and specific inhibitors of Stit function . In conclusion , clones of stit cells and compartmentalized stit inactivation in the wing resulted in growth defects leading to tissue shape alterations . To determine the cellular defects leading to wing shape changes upon Stit disruption , we first analyzed stit mutant clones in adult wings . We generated wings composed of forked ( f ) -marked wild-type cells or f-marked stit mutant cells opposing unmarked minute cells to assess the growth and competitiveness of stit mutant cells . In this setting , wild type almost completely outgrew the minute clones . We found a small , but significant reduction in both the size ( wing cell area ) and total number of cells in stit clones compared to wild-type clones ( Figures 1B and S1C ) . The overall size of wings containing stit mutant clones was also slightly reduced compared to wings composed of f-marked wild type cells . Similarly , stit inactivation by expression of stit-IRv1091 or stit-IRD158 in the posterior compartment by en-GAL4 resulted in a 20%–30% reduction in cell numbers , a significant cell size reduction and an overall size reduction of the posterior wing compartment ( Figure 1D , E ) . This indicated that Stit controls cell growth and/or proliferation in the wing imaginal discs . To directly assess imaginal epithelial growth , we co-expressed GFP with StitKD or stit-IR in the dorsal compartment using ap-GAL4 . We labeled discs with anti-DE-Cadherin and determined cell numbers and their ratios in the juxtaposed dorsal and ventral epithelial layers at 20–40 h after pupal formation ( APF ) ( Figures 1C and S1D ) . Wing discs from ap>GFP control pupae had a D/V cell ratio close to the expected value 1 . ap>StitKD , however , showed a D/V cell ratio of 0 . 67 , indicating a 33% reduction of cells within the dorsal compartment compared to the ventral or compared to the wild-type control ratio . The shape of pupal wing cells expressing StitKD was different than control wing cells . They were shorter than their ventral counterparts , or conversely their ventral counterparts appeared taller ( Figure S1D , X–Z section ) . This D/V cell shape difference was not observed in the control and suggests a compensatory mechanism , where either the dorsal cells stretch to cover the larger ventral area or conversely the ventral cells compact to accommodate a reduced dorsal surface . Thus , stit mutant clones or localized expression of StitKD and stit-IR caused a decrease in cell numbers at pupal stages and a reduction in both cell size and number in the adult wing . We graded the impact of the genetic manipulations to cell numbers in the order: StitKD>stit-IR>stit null mutant . This suggested a prominent Stit protein perdurance in the null mutant clones in the wing and prompted us to focus our analysis on the phenotypes generated by the stit RNAi and the dominant negative transgenes . We first assessed whether the reduction in cell numbers was due to an increase in cell death or a decrease in cell proliferation . Expression of stit-IR and StitKD in the dorsal compartment of wing discs did not lead to increase in cell death as assessed by TUNEL labeling or Caspase 3 staining ( Figure S1G ) . Moreover , concurrent expression of p35 , a baculoviral caspase inhibitor , together with stit-IR or StitKD did not ameliorate the wing bending caused by stit inactivation ( unpublished data ) , suggesting that Stit does not influence apoptosis . To examine its role in cell proliferation , we inactivated stit in clones using hs-flp;Act>CD2>GAL4 ( AFG4 ) to drive StitKD and GFP expression . In parallel , we used the same approach to express InRDN or RagADN as positive controls . We determined the number of GFP positive cells in individual clones 48 h after clone induction . We plotted the percentage of clones within defined size intervals and calculated the mean cell doubling time ( CDT ) for each genotype . The cell numbers in GFP expressing control clones produced a bell-shaped distribution , where the majority ( 40% ) of clones were comprised of 11–15 cells . This resulted in a CDT of 12 . 9 h ( Figure 1F ) , in close agreement with previous findings [25] . The expression of InRDN or RagADN shifted this bell-shaped curve to the left , indicating an increase in the frequency of clones comprised of fewer cells and a corresponding increase in mean CDT to 14 . 2 and 17 . 3 h , respectively . Similarly , StitKD expression reduced the number of cells per clone , resulting in a mean CDT of 14 . 9 h , suggesting that Stit is required for epithelial cell proliferation . To pinpoint a potential function of Stit in cell proliferation , we labeled ap>StitKD or ap>stit-IR discs with anti-phosphohistone H3 ( PH3 ) , EdU ( a BrdU analog ) , and anti-dGeminin . We observed a strong reduction in the labeling of all three markers in ap>StitKD or ap>stit-IR discs compared to the control discs ( Figures 1G , 1H , S1E and unpublished data ) . Surprisingly , the 3-fold decrease of anti-PH3 positive cells was not confined in the dorsal compartment but was evident in the entire wing pouch of ap>StitKD . The EdU and Geminin stainings also showed a general reduction in labeling cells upon stit inactivation compared to control discs ( Figure S1E ) . This generalized decrease of proliferative markers in the wing discs in response to localized Stit inactivation is in congruence with a nonautonomous mechanism that coordinates wing growth in response to local perturbations [26] . We labeled discs with the G2/M phase marker Cyclin B . Cyclin B expression is dynamic during the cell cycle , accumulating from the end of S-phase , through G2 , abating during mitosis . ap>StitKD expressing wing discs showed increased Cyclin B levels in the ventral compartment , suggesting a compensatory prolongation or an arrest in G2 therein ( Figure S1F ) . Thus , apart from the global reduction of mitotic markers and the reduction of the wing size upon Stit inactivation , we did not detect any selective block in the cell cycle . Presumably , the decrease in dorsal proliferation occurs early and continuously during development , while the ventral compensation follows in response . We conclude that stit is required for normal levels of cell proliferation during wing development . Rather than playing a direct role controlling the cell cycle , Stit is more likely required for cellular growth . Inactivation of stit in the dorsal compartment either by MS1096>stit-IR or by ap>StitKD generated strong upwardly bent wings resembling the defects caused by the disruption of the InR and other growth regulators ( Figures 1A , 2A and Table S1 ) [27] , [28] . ap-GAL4-driven InRDN caused a strong reduction in wing size and bending primarily around the margin ( Figure 2A ) . This contrasted the bending of the entire wing blade generated by ap>StitKD expression ( Figure 1A ) . The difference in wing bending at the margin caused by InRDN versus the bending of the entire wing blade caused by StitKD suggests a spatial control of Stit and InR activation in the discs . Co-expression of StitKD and InRDN in the dorsal compartment further reduced wing size and increased the bending compared to the defects caused by InRDN alone ( Figure 2A ) . These results suggest that Stit and InR function in parallel to control wing growth . To further examine the effect of inactivating both receptors on wing cell proliferation , we generated hs-flp;Act>CD2>GAL4 clones expressing both StitKD and InRDN and determined the number of cells per clone . The fraction of clones composed of fewer cells increased , leading to a marked increase in CDT ( 17 . 5 h ) compared to the effect caused by inactivation of either receptor alone . This suggests that both Stit and InR are required for optimal tissue growth ( Figure 1F ) . We therefore examined potential genetic interactions of stit with mutations affecting the PI3K-I/TORC1 signaling cassette . Overexpression of Stit by MS1096>stit causes severe crumpling of the whole blade ( Figure 2B ) . Milder overexpression in the dorsal surface by ap>stit at 18°C caused a gentle bend of the wing downwards , indicative of an overgrowth of the dorsal surface ( Figure S2A ) . A similar downward bending was generated by overexpression of the growth activator Rheb in the dorsal compartment using MS1096>Rheb ( Figure 2A ) . The wing overgrowth phenotypes caused by the overexpression of Stit in the dorsal compartment were accompanied by an increase of BrdU and anti-PH3 in the dorsal compartment ( Figure S2B ) . Thus , the changes in wing shape provide a sensitive assay for Stit function in tissue growth . We tested whether the defects caused by stit inactivation or overexpression can be modified by an array of loss-of-function alleles and overexpression constructs of genes regulating cell death , cell cycle control , and growth ( Table S1 ) . We did not detect any interactions with mutations affecting cell cycle progression . However , MS1096-GAL4-driven co-expression of stit-IR together with activating components of the PI3K-I/TORC1 pathway—PI3K-I , Akt , and Rheb—suppressed the stit bent wing phenotype ( Figure 2A ) . Similarly , simultaneous inactivation of stit and PTEN ( PTEN-IR ) gave a flatter wing , suppressing the effect of stit knock down ( Table S1 ) . Furthermore , expression of an activated form of dS6K kinase ( the Drosophila ortholog of p70 S6K , a direct TORC1 target ) was sufficient to suppress the stit inactivation wing-bending phenotype , indicating that increased TORC1 signaling at any level of the intracellular pathway is sufficient to compensate for a lack of stit . Conversely , the wing phenotype resulting from stit overexpression was suppressed by expression of PTEN , a negative regulator of the pathway and could be enhanced by increasing levels of positive regulators or effectors of the TORC1 pathway , PI3K-I , Akt , Rheb , and dS6K ( Figure 2B , Table S1 ) . This implied that Stit activates either the PI3K-I pathway or a novel TORC1 regulatory pathway . To investigate the postulated role of Stit in cellular growth , we turned to the larval fat body , which is composed of endoreplicating cells growing without cellular division and thereby facilitates the analysis of cellular growth . Stit is not expressed in the fat body and its inactivation by StitKD expression in this tissue did not lead to any discernible cell growth phenotype ( Figure S3B , C ) . To investigate if Stit can induce PI3K-I activation and cellular growth , we overexpressed stit and UAS-RFP in clones in larvae expressing the GFP-PH ( tGPH ) reporter , which is recruited by PIP3 at the plasma membrane , thus reflecting PI3K-I activity . In parallel , we generated clones overexpressing a membrane targeted PI3K-I ( PI3K-CaaX ) and StitKD as positive and negative controls ( Figures 3 and S3 ) . As expected , clonal overexpression of PI3K-CaaX caused a pronounced cell overgrowth and induced membrane accumulation of GFP , while StitKD had no effects ( Figures 3A and S3B , C ) . stit overexpressing cells showed an accumulation of the GFP-PH signal but not significant overgrowth compared to their wild-type neighbors ( Figure 3B ) . Upon 24 h starvation , however , PI3K-CaaX- or stit-expressing cells were clearly larger and maintained a strong cortical GFP-PH signal compared to their neighbors ( Figure 3A , B ) . This analysis suggests that Stit , like InR signaling , can activate PI3K-I and spare fat body cells from starvation-induced size reduction [19] , [20] . To investigate the potential regulatory role of Stit in starvation-induced autophagy , a TORC1-regulated process , we assessed the accumulation of a Cherry-tagged Atg8a autophagy reporter in clones overexpressing Stit . A starvation period of 5 h induced a punctate Ch::Atg8a accumulation in wild-type cells . This increase in Ch::Atg8a was not evident in cells overexpressing either stit or PI3K-I ( Figure 3C , 3D , and 3G ) . Similarly , the developmental wave of programmed autophagy ( P . A . ) observed in fed larvae just before pupation was blocked as efficiently by Stit as by PI3K-I ( Figure 3E , 3F , and 3G ) [20] . Thus , Stit overexpression can spare fat body cells from starvation-induced cell size reduction and autophagy . These observations implicate TORC1 as a downstream effector of Stit . To test this hypothesis we reared larvae overexpressing stit or PI3K-CaaX with 50 µM rapamycin ( a potent inhibitor of TORC1 activity ) and investigated their ability to block Ch::Atg8a accumulation . Rapamycin treatment for 24 h abolished the inhibitory effect of stit or PI3K-I on Ch::Atg8a accumulation upon starvation ( Figure 3E , 3F , and 3G ) , thus indicating that Stit-mediated suppression of autophagy requires TORC1 activation . To assay TORC1 activity , we examined the phosphorylation of the translational repressor 4E-BP by TORC1 , upon stit overexpression . We generated clones of cells in the fat body of fed and starved larvae overexpressing stit , RagADN , or PI3K-CaaX together with GFP and labeled the tissues with an antibody against p-4E-BP ( Figures 4 and S4A ) [22] . RagADN expression , known to lower TORC1 activity , caused a reduction in p-4E-BP labeling in clones of fed larvae compared to neighboring cells ( Figures 4K and S4A ) [3] . Conversely , cells expressing PI3K-CaaX showed a stronger p-4E-BP signal than surrounding cells upon starvation ( Figure 4A , K ) . As such , anti-p-4E-BP reflects the nutritional status and TORC1 activity levels of the cell . Clonal Stit overexpression in fed larvae did not greatly affect p-4E-BP signal intensity ( Figure 4K ) but , like PI3K-CaaX , caused a robust increase of the signal in stit-expressing cells compared to wild-type neighbors following starvation ( Figure 4A , B , and K ) . This indicates that stit , like InR , activation is sufficient to activate TORC1-mediated p-4E-BP phosphorylation and to protect cells against the starvation-induced drop in TORC1 activity . Next , we investigated whether dS6K , another well-characterized TORC1 effector , can be phosphorylated by Stit overexpression upon starvation . We analyzed fat body cell clones overexpressing RagADN or stit with a p-dS6K antibody . RagADN expression strongly reduced the cytoplasmic punctate p-dS6K staining observed in neighboring cells lacking RagADN expression ( Figure S4B and G ) . To assess the specificity of the cytoplasmic p-dS6K labeling , we overexpressed dS6K in fat body cell clones and subjected the larvae to a 5-h starvation period . We found that the p-dS6K signal was selectively enhanced in cells overexpressing dS6K compared to surrounding cells with endogenous dS6K levels ( Figure S4C and G ) . This indicated that the p-dS6K antibody can faithfully recognize the cytoplasmic punctate accumulations of p-dS6K . To investigate the nature of these puncta , we labeled larvae expressing the late endosomal/lysosomal marker , GFP-Lamp1 , with anti-p-dS6K . During the P . A . wave that clears the larval tissues at pupation GFP-Lamp1 puncta become enlarged and decorated by p-dS6K . This p-dS6K localization supports the observation that a major site of TORC1 activity is on late endosomal/lysosomal structures ( Figure S4E ) [5] . Clonal overexpression of stit in the fat body ( Figure S4D and G ) also resulted in the increase of p-dS6K puncta in starved larvae . This analysis indicates that Stit can induce the accumulation p-dS6K and p-4E-BP , two well-characterized TORC1 targets upon starvation . We then monitored p-4E-BP accumulation , cell growth , and the autophagy reporter in clones of fat body cells expressing stit to ask if Stit acts through the conventional PI3K-I-TORC1 pathway to support growth and suppress starvation-induced autophagy . As previously established , clonal expression of PI3KDN , Akt-IR , and RagADN in fat body cells all led to reduction of cell growth and p4E-BP levels and entry into autophagy ( Figure 4C , E , and G ) [3] , [19] . Also , expression of TorTED led to a strong cell size reduction , rendered cells equal to neighbors in respect to p4E-BP levels , while it induced a high Ch::Atg8a accumulation ( Figure 4I ) . Co-expression of either PI3KDN , Akt-IR , RagADN , and TorTED together with stit reversed the effect of Stit in suppressing autophagy as well as its effect in sustaining growth and p-4E-BP levels upon starvation ( Figure 4B , D , F , H , J , and K ) . Collectively , the analysis of stit overexpression in endoreplicating tissues indicates that Stit can signal through the conventional PI3K-I/TORC1 pathway to sustain TORC1 signaling levels and block autophagy upon starvation . To address whether Stit controls the growth of proliferating epithelial wing discs through the PI3K-I pathway , we first expressed PI3K-CaaX in clones using hs-flp;Act>CD2>GAL4 ( AFG4 ) and recorded the recruitment of the GFP-PH probe at the cell cortex . We detected a marked increase of the GFP-PH signal in the PI3K-CaaX cells , marked with RFP compared to adjacent nonexpressing cells under starvation conditions . This indicated that GFP-PH reliably reflects PI3K-I activation upon starvation in the wing ( Figure 5A ) . Similarly to PI3K-CaaX , overexpression of stit by MS1096-GAL4 resulted in an increase of the GFP-PH signal along the membranes of the Stit overexpressing cells compared to their neighbors expressing endogenous levels ( Figure 5B ) . This indicated that Stit could activate the PI3K-I pathway in the wing discs . However , neither the clonal inactivation of Stit nor interference with InR signaling by the expression of the dominant negative constructs was sufficient to induce a change in the intensity or the localization of the PI3K-I activity reporter ( Figure 5C , D ) . By contrast , the concurrent expression of both StitKD and InRDN using the same driver lead to a marked decrease in the GFP-PH intensity in the cells expressing both constructs compared to their neighbors ( Figure 5E ) . This was most evident at the interface of cells expressing both StitKD and InRDN . This suggests that Stit and InR co-operate to activate PI3K-I in the wing . We further investigated the interplay of Stit and InR during wing growth by monitoring TORC1-dependent dS6K phosphorylation . We first asked whether in situ staining with the p-dS6K antibody provides a reliable readout for the detection of dS6K activation in the wing discs . We expressed dS6K in the dorsal compartment of the wing discs and stained for p-dS6K . We detected an increase of the p-dS6K signal selectively in the basal side of the epithelial cells expressing dS6K ( Figure S5A ) . Conversely , dS6K inactivation by expression dS6K-IR in the posterior compartment of wing discs resulted in the reduction of the pdS6K basal signal in the posterior cells ( Figure S5B ) . This prompted us to use the p-dS6K antibody for the in situ analysis of dS6K activation upon localized inactivation of Stit and members of the InR pathway in the wing discs . We expressed RagADN , raptor-IR ( an RNAi construct directed against the TORC1 component raptor ) , InRDN , stit-IR , and StitKD together with GFP in the dorsal wing compartment and assessed their effects on dS6K phosphorylation . As expected , InRDN , RagADN , or raptor-IR lead to a marked reduction in the intensity of p-dS6K labeling compared to the ventral compartment ( Figure 6A , B , C , D , and L ) . Conversely , Rheb expression resulted in an increase in the intensity of p-dS6K labeling ( Figure 6J , L ) . This indicates that the detected dS6K phosphorylation was responsive to the level of TORC1 activity in the compartment . The expression of StitKD or stit-IR with the same driver resulted in a reduction of p-dS6K labeling to similar levels as the inactivation of InR pathway components ( Figure 6D , E , F , and L ) . In accord with the in situ analysis of p-dS6K , we detected a modest but reproducible reduction of p-dS6K levels in Western blots of wing disc protein extracts deriving from stit mutant larvae compared to wild-type controls ( Figure 6M ) . The reduction in dS6K phosphorylation upon Stit or InR pathway component inactivation is in close agreement with the adult wing morphology defects caused by the same constructs . This indicates that Stit and InR in parallel control epithelial tissue growth by inducing dS6K phosphorylation . To address possible intersection points of InR and Stit signaling , we overexpressed Rheb and inactivated Stit in the same cells using ap-GAL4 . UAS-Rheb expression ameliorated the decrease in p-dS6K labeling caused by Stit inactivation ( Figure 6F , K , and L ) . The restoration of p-dS6K staining intensity to 80% of wild-type together with the strong suppression of the adult wing-bending phenotype caused by Rheb expression in Stit-deficient cells indicate that Stit acts through Rheb to activate TORC1 during wing development ( Figures 6K , L and 2A ) . We concluded that Stit and InR collectively activate PI3K-I and regulate TORC1 levels in the wing imaginal disc to control tissue growth . InR and Stit control TORC1 activity during wing growth while Stit can block autophagy in endoreplicating larval tissues in response to starvation . To directly monitor starvation-induced autophagy in the wing , we used the Ch::Atg8a reporter under the control of its endogenous promoter . We first expressed either GFP alone , or TORTED , or RagADN or InRDN or PTEN together with GFP using ptc-GAL4 . While GFP alone had no effect on the induction of autophagy ( Figures S6A and 7G ) , both TORTED and PTEN caused a marked increase in the number of Ch::Atg8a puncta forming within the GFP marked expression domain ( Figure 7A , B , and G ) , indicating that the PI3K-I/TORC1 axis functions to suppress autophagy in the wing . Surprisingly , RagADN and InRDN or InR-IR showed little or no accumulation of Ch::Atg8a puncta in the ptc expression domain ( Figures 7C , G and S6B ) despite each causing a substantial decrease in the levels of TOR target p-dS6K ( Figure 6 and unpublished data ) . This argues that p-dS6K phosphorylation and the activation of autophagy markers respond to different levels of InR and TORC1 activity in the wing . This contrasts the analysis of TORC1 activation in the fat body , where InR signaling is the dominant receptor activating PI3K-I . In this tissue , InR mutant cells or cells expressing InR-IR or InRDN both restrict growth and activate the catabolic process of autophagy ( Figure S3E , F ) [19] . Like the InR , Stit inactivation by either StitKD or stit-IR using ptc-GAL4 ( Figure 7D , G and unpublished data ) did not increase Ch::Atg8a puncta formation . This analysis indicates that although inactivation of TORC1 alone leads to autophagy , inactivation of either of its upstream receptors is not sufficient to induce the catabolic response in epithelial proliferating cells . We hypothesized that InR and Stit signaling may cooperatively drive TORC1 activation , and hence only a reduction of both pathways would mimic the effect of TORTED on autophagy . We introduced either StitKD InRDN or Stit-IR InR-IR or StitKD InR-IR into the ptc-GAL4 Ch::Atg8a background to examine the effect of inactivating both receptors on autophagy ( Figures 7E , G and S6 ) . Each manipulation resulted in a striking increase in the levels of Ch::Atg8a puncta comparable to the one caused by TORTED expression . Finally we challenged ptc>StitKD animals with prolonged nutrient restriction on low-energy food . This regime delays development by 2–3 d but is sufficient to support larval development to adulthood . Remarkably , constant nutrient restriction gave an increase in autophagy specifically within the ptc domain upon reduction of Stit signaling ( Figure 7F , G ) . We conclude that Stit and InR are interchangeably required to sustain TORC1 activity and to prevent a catabolic switch in wing discs ( Figure 8 ) . Thereby , they endow proliferating imaginal epithelial tissues with a two-tiered control of growth and autophagy . Animals modulate organ-specific growth according to their developmental stage and homeostatic needs . This is particularly evident during nutrient starvation , when organisms respond by relocating stored energy resources and by recycling of cellular material . The starvation response is , however , not equal in every organ in regards to growth ( anabolism ) and shrinkage ( catabolism ) and the molecular mechanisms involved in these different responses are poorly understood . Several findings suggest that variations of insulin-signalling-mediated growth are at play in different organs . In adult Drosophila , the size of the gut is dynamically regulated depending on food availability and InR signaling [29] . Another recent study revealed that growth can be controlled by an alternative RTK when Insulin signaling is reduced in the developing larval Drosophila brain [22] . When larvae were cultivated under nutrient-restricted conditions and InR signaling was low , neuroblasts ( NBs ) were still able to proliferate . This “brain sparing” is dependent on the ALK receptor in neuronal lineages and its ligand Jelly Belly , which is expressed in glia [30] . Importantly , under normal feeding conditions , ALK is essential for NB development , showing that the ALK signaling pathway is not merely a general backup for the InR upon low nutrition; it rather promotes the growth of specific NB lineages at low InR signaling levels . In this respect , Stit , like ALK , supports growth under variable nutrient conditions , promoting proliferation in epithelial tissues upon low InR signaling but is also necessary for optimal epithelial growth under normal conditions . Curiously , the signaling pathways supporting growth and proliferation downstream of ALK and Stit appear to be different . TSC1/2 , Rheb , and TORC1 were dispensable for ALK function in growth and proliferation , while the direct TORC1 downstream targets S6K and 4E-BP were required . Stit , in contrast to ALK , utilizes the classical PI3K-I/TORC1 pathway to drive growth . First , Stit is sufficient to drive PI3K-I activation and suppress autophagy in starved fat body cells . Second , the suppression of autophagy is rapamycin-sensitive and hence TORC1 dependent . Third , Stit and InR cooperate to control PI3K-I activity and autophagy suppression in the wing , and finally , PI3K-I and Rheb overexpression can rescue Stit inactivation phenotypes . As ALK does not signal through TORC1 , it is hence unlikely to regulate autophagy . Thus , apart from the insulin receptor , Stit provides the first example of an RTK that negatively regulates autophagy . Simultaneous reduction of the signaling activity of both Insulin and Stit receptors , or prolonged starvation together with reduced Stit activity , leads to the induction of autophagy in the wing . We propose that the simultaneous inactivation of Stit and InR reduces PI3K-I activity and TORC1 signaling to a critically low level , beyond the limit of our p-dS6K detection range . This reveals a mechanism where Stit or InR signaling prevent TORC1 activity from dropping below a threshold where it can no longer suppress autophagy . Thus , while anabolic growth can vary in response to RTK signaling , autophagy and catabolism in proliferating epithelia are strictly inhibited by signaling from either Stit or InR . The cooperative functions of Stit and InR provide a novel failsafe mechanism , allowing TORC1 activity to variably modulate growth under fluctuating nutrient conditions without incurring a transition to catabolism ( Figure 8 ) . As Stit is selectively expressed in imaginal discs giving rise to adult epithelial organs , it may function to safeguard the growth of these tissues during conditions of low nutrient availability , at the expense of nonexpressing tissues . The product of the mammalian Ret oncogene and Stit share several distinctive features . Their amino acid sequences are 42% identical and 64% homologous within the kinase domain . Both Stit and Ret are composed of an extracellular region with a Cadherin domain , a transmembrane stretch , and an intracellular tyrosine kinase domain . Apart from Stit the fly genome encodes a second Ret paralog ( dRet ) predominantly expressed in neurons . Although the signals that activate the Drosophila Ret-like proteins remain unknown , Stit is activated upon epidermal wounding to initiate re-epithelialization and barrier repair . Mammalian Ret is activated by GDNF to instruct epithelial morphogenesis in the uteric duct of the kidney [31] and Ret-activating mutations have been implicated in a variety of human cancers including epithelial cancers ( breast and lung ) and multiple endocrine neoplasia ( men2 ) [32]–[35] . More recently , overexpression of an activated form of Drosophila Ret that mimics the mutation that leads to men2 has been used to identify potential Ret signal transducers and drugs that interfere with its aberrant activation [36] . Our analysis reveals the physiological role of Stitcher in epithelial tissue growth and proliferation and strengthens the notion that Stit and Ret share the same functions in controlling PI3K-I and TORC1 activity in epithelial tissues . The newly identified function of Stit in sparing proliferating epithelial organs from starvation-induced autophagy raises the question of whether Ret activation may suppress autophagy as well . Since oncogenic Ret mutations promote cancer growth in part by activation of the TORC1 pathway [36] , [37] , our findings suggest that aberrant Ret signaling may suppress autophagy during cancerous growth , potentially providing an advantageous mechanism or driving force for the growth of Ret-expressing tumors . Flies were cultivated at 25°C on our standard lab fly medium consisting of , per liter , 32 . 7 g dried potato powder , 60 g sucrose , 27 . 3 g dry yeast , 7 . 3 g agar , 4 . 55 ml propionic acid , and 2 g nipagin , giving a final concentration of 15 . 3 g/l protein and 6 g/l sugar . Low-energy food consisted of 35 g dried potato powder , 10 g glucose , and 8 g agar per liter . For starvation experiments , larvae were transferred to PBS Agar ( 1% ) for defined periods . Rapamycin ( Santa Cruz Biotech , sc-3504A resuspended in Methanol ) was diluted to 100 uM in 20% Sucrose PBS and mixed to a paste 1∶1 with dry yeast , which was added to PBS agar vials to which larvae were placed for defined periods . The fly stocks w1118 , ap-GAL4/CyO , MS1096-GAL4 [38] , en-GAL4 , da-GAL4 , UAS-InRDN ( K1409A ) , UAS-InR-CA ( R418P ) , UAS-InR-IR , UAS-Rheb , UAS-PTEN , UAS-PTEN-IR , raptor-IR , UAS-Akt , UAS-PI3K-CaaX , UAS-PI3K-I , UAS-dS6K alleles ( wild type and constitutive active , S6KTE , S6KSTDE , or S6KSTDETE , which are intrinsically active due multiple serine/threonine to acidic amino acid substitutions [39] ) , UAS-TORTED , UAS-CycB , UAS-CycE . L , CycEAR95 , dap4 , UAS-FLP , FRT82B Ubi-GFP , UAS-GFP , hs-flp; FRT82B , yw hs-flp;Dr/Tm3 , Sb ( 1 ) , y , w , hs-flp; Act>CD2>GAL4; UAS-GFP , and UAS-p35 were from Bloomington , while UAS-stit-IR ( 1091 and 8401 ) and UAS-dS6K-IR ( 18126 ) were from VDRC . UAS-StitKD , UAS-stit , stit-α-gfp , stitex266 , and stitEXEL9056 were described previously [24] . The pWIZ RNAi vector was used to generate the UAS-stit-IRD158 that targets the third exon of stit . Other stocks included hs-flp , UAS-RFP ( kind gift from D . Hipfner ) , UAS-RagAT16N , designated RagADN , act>CD2>GAL4/CyO;tGFP-PH ( gift from Stephen M . Cohen ) [40] , pmCh::Atg8a/CyO [41] , yw FLPf36a; FRT82B UbqGFP 83f+ 87D M ( 3 ) 95A/Tm6b , Tb ( 1 ) ( gift from Dr . Fernando Roch ) , and hsp70-Flp; UAS-Dicer; r4-mCh::Atg8a , act>CD2>GAL4 , and UAS-GFPnls ( generously provided by Thomas Neufeld ) . The last stock has a leaky heat shock promoter requiring no heat shock , otherwise 20 or 75 min heat shock in a 37°C water bath was applied for fat body or wing clones , respectively . hs-flp; act >CD2>GAL4–generated clones are denoted as AFG4 . Larval fat body and/or discs were dissected from larvae , fixed in 4% formaldehyde/PBS ( either Sigma F1635 or polysciences #18814 ultrapure ) for 20 min , and labeled immediately afterward following standard protocols ( PBSBT containing 0 . 5% BSA ) . Primary anti-sera included rabbit anti-GFP ( Invitrogen , 1/300 ) , rabbit anti-p-dS6K and p-4E-BP ( Cell signaling #9209 and #2855 , each diluted 1/100 ) , rabbit anti-PH3 ( sigma , 1/200 ) , rabbit anti-CycB ( D . Glover , 1/750 ) , rat anti-dGem ( H . Richardson , 1/300 ) , mouse anti-Fas3 ( 1/100 ) , and rat anti-DE-Cad ( 1/30 ) from DSHB and guinea-pig anti-Stit ( 1/5 , 000 ) [24] . Secondary antibodies were from Jackson Immunolab and Molecular probes . TUNEL and EdU assay kits were from Roche and performed as detailed therein . Samples were mounted in Vectashield H-1000 ( vectorlabs ) for imaging on either Zeiss LSM510 , 710 , or 780 confocal microscopes or Zeiss Axioplan2 . Western blotting was performed using HRP-conjugated antibodies ( Jackson Immunolabs ) . Rabbit p-dS6K ( as above ) was used at 1/1 , 000 , mouse anti-tubulin ( sigma , T5168 ) was used at 1/100 , 000 , while rabbit anti-dS6K ( kind gift of T . Neufeld ) was used at 1/1 , 000 . Twenty larval wing discs per well were run on 10% mini-PROTEAN TGX gels ( Biorad ) . Wild-type animals were starved for 48 h on PBS agar . Densitometric analysis was performed using ImageJ and Image Studio Lite Western Blot Analysis Software ( LI-COR ) . Image analysis was performed using ImageJ . For DE-Cad pupal wing labeling and PH3 mitotic cell counting in the wing , a cell counting grid was applied to a projection of the confocal stack to aid manual counting . In the case of pupal wing cell quantification , projections of the dorsal or ventral apical cell surfaces immediately overlaying each other were analyzed . GAL4-driven expression of GFP or RFP allowed distinction of cells of interest in all cases . For intensity measurements ( e . g . , p-dS6K ) , thin ( 10 µm ) projections of stacks were analyzed . In brief , the threshold function was applied to images and the dorsal and ventral regions ( or in the fat body cells expressing GFP versus their immediate neighbors ) were manually highlighted and the intensities measured using the analyzed particles function . Care was taken to ensure that control cells or regions of tissues from different samples and genotypes gave similar background values to which test cells or regions were compared . From this the intensity relative to area was determined for both background and test regions . For AFG4 clones , the number of GFP-positive cells within a clone was manually counted from confocal stacks spanning the entire disc . CDT was calculated using the formula ( log2/logn ) h , where n = median number of cells per clone and h = age of the clone [25] . The adult wing hair measurements were aided by the image J . In Figure S1C , the posterior region is defined as the area from L4 to posterior wing margin and the anterior region is defined as the area from L3 to anterior wing margin . Error bars in all figures represent standard deviation from the mean . All p values were generated by student's t test .	Growth of organs , or anabolism , is tightly controlled by nutritional and hormonal cues such as insulin-like peptides that also suppress autophagy through their receptors and downstream growth pathway . Starvation conditions induce growth arrest and catabolism ( involving autophagy ) in some tissues while sparing the growth of other prioritized organs . The mechanism behind this tissue-specific regulation of growth versus catabolism is largely unknown . In this study , we show that Stitcher , a Drosophila Ret-oncogene-like growth factor receptor , controls epithelial tissue growth . Stitcher , working in parallel with the Insulin receptor , endows epithelial organs , such as imaginal wing discs , with resistance to low nutrient and insulin conditions by suppressing autophagy and , at the same time , promotes cell division and growth in these tissues . Thus , Stitcher and the Insulin receptor work together to allow a two-threshold response to starvation in epithelial tissues . In cancer , this pathway is almost invariably constitutively stimulated , and so we postulate that oncogenic mutations of Ret promote tumor growth partly by counteracting the tumor suppressive effects of autophagy .	[ "Abstract", "Introduction", "Results", "Discussion", "Experimental", "Procedures" ]	[ "animal", "models", "developmental", "biology", "signal", "transduction", "drosophila", "melanogaster", "animal", "genetics", "model", "organisms", "organism", "development", "cell", "division", "cell", "growth", "molecular", "development", "genetics", "biology", "molecular", "cell", "biology", "morphogenesis", "genetics", "of", "disease", "cell", "differentiation", "gene", "function" ]	2013	Two-Tiered Control of Epithelial Growth and Autophagy by the Insulin Receptor and the Ret-Like Receptor, Stitcher
Based on spatiotemporal clustering of human dengue virus ( DENV ) infections , transmission is thought to occur at fine spatiotemporal scales by horizontal transfer of virus between humans and mosquito vectors . To define the dimensions of local transmission and quantify the factors that support it , we examined relationships between infected humans and Aedes aegypti in Thai villages . Geographic cluster investigations of 100-meter radius were conducted around DENV-positive and DENV-negative febrile “index” cases ( positive and negative clusters , respectively ) from a longitudinal cohort study in rural Thailand . Child contacts and Ae . aegypti from cluster houses were assessed for DENV infection . Spatiotemporal , demographic , and entomological parameters were evaluated . In positive clusters , the DENV infection rate among child contacts was 35 . 3% in index houses , 29 . 9% in houses within 20 meters , and decreased with distance from the index house to 6 . 2% in houses 80–100 meters away ( p<0 . 001 ) . Significantly more Ae . aegypti were DENV-infectious ( i . e . , DENV-positive in head/thorax ) in positive clusters ( 23/1755; 1 . 3% ) than negative clusters ( 1/1548; 0 . 1% ) . In positive clusters , 8 . 2% of mosquitoes were DENV-infectious in index houses , 4 . 2% in other houses with DENV-infected children , and 0 . 4% in houses without infected children ( p<0 . 001 ) . The DENV infection rate in contacts was 47 . 4% in houses with infectious mosquitoes , 28 . 7% in other houses in the same cluster , and 10 . 8% in positive clusters without infectious mosquitoes ( p<0 . 001 ) . Ae . aegypti pupae and adult females were more numerous only in houses containing infectious mosquitoes . Human and mosquito infections are positively associated at the level of individual houses and neighboring residences . Certain houses with high transmission risk contribute disproportionately to DENV spread to neighboring houses . Small groups of houses with elevated transmission risk are consistent with over-dispersion of transmission ( i . e . , at a given point in time , people/mosquitoes from a small portion of houses are responsible for the majority of transmission ) . Dengue is the most widespread mosquito-borne viral disease with 3 . 6 billion people at risk of infection world-wide each year [1] . Aedes aegypti is the principal mosquito vector of dengue virus ( DENV ) . Indirect transmission occurs by horizontal transfer of virus between humans and female Ae . aegypti [2] . A key component of understanding DENV transmission dynamics is to understand the spatial and temporal scale at which human-mosquito encounters and virus transmission occur . DENV infection in humans has been shown to have substantial spatial and temporal variation at relatively small scales . Cohort studies in rural Thailand , where dengue is hyperendemic , indicate that dengue epidemiology and clinical presentation can differ dramatically between children in close geographic and temporal proximity at the level of a school and village [3]–[5] . Clusters of human DENV infections have also been detected in and around individual households [6]–[8] . Much of this fine scale spatiotemporal heterogeneity has been thought to be due , at least in part , to the behavior of the mosquito vector . Flight patterns and feeding behavior of the female Ae . aegypti , which have been studied extensively , indicate that this is a relatively sedentary species that feeds frequently and almost exclusively on human blood [9]–[15] . Much less is known about the interactions between humans and Ae . aegypti in natural settings that result in DENV transmission . Results from our combined longitudinal cohort and geographic cluster study in Kamphaeng Phet , Thailand are consistent with focal DENV transmission occurring at a fine scale [5] , [6] . Within 100 meters of a house with a DENV-infected child ( as detected by school absence-based surveillance ) , the likelihood of another house with a DENV-infected child decreased with increasing distance from the original infected child's house . In the current report , we present additional data from the geographic cluster component of our larger cohort/cluster study that more specifically defines the dimensions of local transmission and quantifies the factors that support it . We detected a positive association between DENV infection in children and female Ae . aegypti at fine geographic and temporal scales . Our results add new details to the understanding of focal DENV transmission that can be used to further inform dengue surveillance and prevention strategies , and provide currently missing data for the construction , parameterization and validation of mathematical and simulation models of DENV transmission and control . The study protocol was approved by the Institutional Review Boards of the Thai Ministry of Public Health ( MOPH ) , Walter Reed Army Institute of Research ( WRAIR ) , University of Massachusetts Medical School ( UMMS ) , University of California at Davis ( UCD ) and San Diego State University ( SDSU ) . Written informed consent was obtained from the parents of study participants and assent was obtained from study participants older than seven years . Our study methodology was previously described [5] , [6] . Briefly , the geographic cluster study presented here was part of a larger combined longitudinal cohort and geographic cluster study conducted from 2004 to 2007 among children living in Muang district , Kamphaeng Phet province in north-central Thailand . Children came from 11 schools and 32 villages consisting of >8 , 445 houses . Demographics of house residents and house spatial coordinates were geo-coded into a Geographic Information System ( GIS ) database ( MapInfo [2000] version 6·0; MapInfo Corporation ) . Geographic cluster investigations were initiated by “index” cases selected from a longitudinal cohort of approximately 2000 primary school children . Active school absence-based surveillance was used to detect symptomatic DENV infection in the cohort from June to November of each study year [5] . Cohort children who were DENV-positive by semi-nested reverse transcriptase polymerase chain reaction ( RT-PCR ) [16] from an acute blood sample drawn within three days of illness onset served as an “index” case to initiate a positive cluster investigation around the index case house . Cohort children who were dengue PCR-negative from an acute illness blood sample served as an “index” case for a negative ( i . e . , control ) cluster investigation . In each geographic cluster , ten to 25 child contacts aged six months to 15 years living within 100 meters of the index case were enrolled regardless of clinical status . The child contacts were evaluated at days 0 ( i . e . , the same day as cluster initiation ) , 5 , 10 , and 15 by temperature measurement and symptom questionnaire . Blood samples were collected on days 0 and 15 . Paired day 0 and 15 blood samples from child contacts were tested by both dengue PCR and an in-house dengue/Japanese encephalitis IgM/IgG capture EIA [17] . Dengue EIA-positive results were categorized as “recent dengue” ( RD ) if IgM was negative but IgG was positive with a declining titer between days 0 and 15 [18] , “enrollment seroconversion” ( ES ) if IgM was positive on both days 0 and 15 , and “post-enrollment seroconversion” ( PES ) if IgM was negative on day 0 but positive on day 15 . Based on estimated antibody kinetics and human incubation period [19] , [20] , the approximate interval between infection and day 0 blood collection for RD infections was thought to be about 3 weeks or more , ES infections up to about 2 weeks , and PES infections to be several days . Day 15 PCR-positive infections were thought to have occurred at or soon after cluster initiation . On day 1 of each cluster investigation , adult Ae . aegypti were collected using backpack aspirators from inside and within the immediate vicinity of each house within a cluster . Ae . aegypti larvae and pupae were collected from water-holding containers [21] . After mosquito collections were completed , a pyrethrin mixture insecticide spray ( BP-300: Pyronyl oil concentrate OR-3610A , Prentiss Inc . ) was applied by ultralow volume aerosol inside and around each house to kill adult mosquitoes with the intention of terminating local DENV transmission [22] . Temephos was applied to artificial water holding containers to kill immature mosquitoes . On day 7 , the Thai Ministry of Public Health ( MOPH ) sprayed deltamethrin or permethrin 10% in and around each house in a cluster according to their standard procedures . Female Ae . aegypti were processed so that individual , serotype-specific rates for DENV-infectious mosquitoes could be detected . Mosquito abdomens were removed so that only those females that had virus particles in the head or thorax ( i . e . , disseminated infections with presumably infective salivary glands ) were identified . Individual heads and thoraces were stored at −70°C in the field laboratory and transported weekly on dry ice to the Armed Forces Research Institute of Medical Sciences ( AFRIMS ) laboratory in Bangkok . At the AFRIMS laboratory , heads and thoraces of individual mosquitoes were ground and suspended in 100 µL of RPMI with 1% L-glutamine and 10% heat-inactivated FBS . Ten mosquito suspensions were pooled by combining 14 µL from each individual suspension . Pools were then tested by dengue PCR and each individual sample from a PCR-positive pool was tested by using 14 µL of the individual suspension diluted times ten [23] , [24] . Data were analyzed using SPSS ( SPSS for Windows version 19 ) . Demographic , environmental and entomological parameters were analyzed at the cluster and house levels . Student's t-test or analysis of variance ( ANOVA ) was used to determine differences in continuous variables including distances between houses . Chi-square or Fisher's exact test was used for proportions . A mixed-effects logistic regression model was used to analyze the probability of infection of cluster contacts , while accounting for the nesting of observations within cluster investigations . Of 805 child contacts enrolled in 50 positive cluster investigations , 129 ( 16 . 0% ) had evidence of DENV infection; 119 ( 14 . 8% ) were dengue EIA-positive on day 0 and/or 15 of which 40 were PCR-positive on day 0 , and an additional 10 ( 1 . 2% ) were DENV-positive only by PCR on day 15 . In comparison , nine ( 1 . 1% ) of 794 enrolled child contacts in 53 negative clusters had evidence of DENV infection; seven ( 0 . 9% ) were dengue EIA-positive of which three were PCR-positive on day 0 , and an additional two ( 0 . 3% ) were DENV-positive by day 15 PCR alone [5] . Within positive clusters , the percentage of enrolled contacts that were dengue EIA-positive varied significantly according to distance from the index case house . The DENV infection rate among contacts from the same house as a positive index case was 35 . 3% . If the child contact lived in a different house but within 20 meters of the index case house , the infection rate was 29 . 9% . The infection rate decreased with increasing distance from the index case house , down to 6 . 2% when the contact lived 80–100 meters away . The inverse relationship between DENV infection rate among contacts and distance from the index case house was significant ( Chi square , p<0 . 001 ) ( Figure 1 ) . A mixed-effects logistic regression model confirmed that this association remained significant after controlling for age and gender ( Table 1 ) . Of the 119 dengue EIA-positive child contacts in the positive clusters , 15 ( 12 . 6% ) were categorized as having RD infection , 41 ( 34 . 5% ) as ES infection , and 63 ( 52 . 9% ) as PES infection . RD , ES and PES infections in the positive clusters all tended to decrease as the distance from the index case house increased ( Figure 2 ) . DENV infections based solely on a day 15 PCR-positive result did not appear to decrease with increasing distance from the index case house; however , the number of these cases was low ( Figure 2 ) . Twenty-three ( 1 . 3% ) of 1755 female Ae . aegypti were dengue PCR-positive in positive clusters , while one ( 0 . 1% ) of 1548 was PCR-positive in negative clusters ( p<0 . 001 ) . Considering only those houses with index cases or enrolled child contacts , all 19 DENV-infectious female Ae . aegypti collected from these houses were in positive clusters . These 19 mosquitoes came from 16 different houses in 14 different positive clusters ( Table 2 ) . All four DENV serotypes were represented . Houses with DENV-infectious mosquitoes had significantly more Ae . aegypti pupae and total female Ae . aegypti mosquitoes than houses without infectious mosquitoes ( Table 3 ) . Two of the houses with infectious mosquitoes ( houses 1 and 3 ) contained the largest and second largest number ( 164 and 129 ) of Ae . aegypti pupae collected from any house in the entire study ( Table 2 ) . There were 17 DENV infections in children from the 16 houses with DENV-infectious mosquitoes; 9 DENV infections were in child contacts and 8 were in index cases . Within the houses with infectious mosquitoes , the serotype , when available , of DENV in infected children was identical to that in the infectious mosquito from the same house ( Table 2 ) . DENV infection in children was positively associated with the presence of DENV-infectious mosquitoes in the house . The DENV infection rate among child contacts in houses with infectious mosquitoes was 47 . 4% compared to 28 . 7% in houses from the same cluster but without infectious mosquitoes , and 10 . 8% in houses from other positive clusters ( Fisher's exact , p<0 . 001; Table 3 ) . Conversely , the DENV infection rate among female Ae . aegypti from houses with a positive child index case was 8 . 2% ( Table 4 ) . Excluding index case houses , the rate of infectious mosquitoes from positive cluster houses with a DENV-infected child was 4 . 2% . This rate was only 0 . 4% when no child was infected in a positive cluster house ( Fisher's exact , p<0 . 001 ) . Within the 100-meter radius of positive clusters , almost all DENV-infectious Ae . aegypti were collected from index case houses or from houses within 40 meters of the index case house ( Figure 3 ) . This negative correlation of infectious mosquitoes with distance from the index case house was most likely due to the positive association between houses containing infected children and infectious mosquitoes . Within an individual positive cluster , houses with DENV-infectious mosquitoes tended to be closer to houses containing DENV-infected children than to all houses . Figure 4 shows the mean distance between each of the 16 houses with infectious mosquitoes and other houses in their respective clusters . In three houses ( #10 , 11 and 12 ) , the only infected children in the cluster were in the houses with the infectious mosquito ( es ) . Of the 13 possible comparisons , 12 houses with infectious mosquitoes were closer to houses with infected children than to all houses within the cluster . On average , houses with infected children were closer to houses with infectious mosquitoes than to houses with no infectious mosquitoes within their respective positive clusters ( p = 0 . 028 ) . This study demonstrates a positive association between DENV-infectious Ae . aegypti and DENV-infected children living in the same and neighboring houses . Spatiotemporal clustering of DENV infection in children and mosquitoes was detected at a fine scale , consistent with focal aggregation well within a 100-meter radius area . Houses with infectious mosquitoes had an especially high risk ( 47 . 4% ) of human DENV infection along with elevated measurements of mosquito density; neighboring houses also had elevated risk of human infection . Our results are consistent with the notion that houses with high DENV transmission risk contribute disproportionately to virus amplification and spread . Infections followed a pattern of over-dispersion , which has been reported for other infectious diseases to include indirectly transmitted , mosquito-borne infections [25]–[28] . At a given point in time , people and mosquitoes in a relatively small portion of houses were responsible for the majority of DENV transmission . Our results are the first to demonstrate a direct relationship between DENV infection in humans and mosquitoes at very fine spatiotemporal scales in the natural setting . Other researchers have reported heterogeneity of human DENV infection across space and time [4] , [6] , [8] , [29] . Many entomological studies have shown the limited flight range and preferential and frequent human feeding behavior of Ae . aegypti that would be expected to enhance DENV transmission [11] , [13]–[15] , [22] , [30] . Prior studies of DENV infections in mosquitoes tended to focus on mosquitoes collected in or around houses of people with dengue-like illness [31] , [32]; and when these studies were done across communities , infected mosquitoes were not explicitly linked to human infection [33] . Perhaps because of the difficulty in collecting adult Ae . aegypti , there has been relatively little research done on mosquito DENV infections in relation to human infection dynamics . Our study expands on this picture by showing that human and mosquito infections are positively associated with each other at small geographic and temporal scales . The strongest association was at the level of the individual house . We did not directly evaluate the role of human adults in DENV transmission . It is possible that spatiotemporal dynamics of DENV transmission is different in adults and children , perhaps due to age-specific differences in existing immunity , the rate at which they are bitten [34] , or in their movement patterns and exposure to daytime-biting Ae . aegypti [35] . We would not expect our overall conclusions to change , however , because both adults and children would have contributed to our findings whether or not adults were separately evaluated . Fine scale spatial aggregation of DENV transmission may persist for three weeks or longer . Given the estimated time of infection of RD , ES and PES infections and because all of these categories of infection appeared to show focal aggregation within the ≤100-meter radius of the clusters , the spatial pattern we detected could have been present for greater than three weeks . This pattern , which is similar to what was observed for DENV-infected Ae . aegypti in households in Mexico [12] , may have persisted for a longer period if not truncated by the vector control interventions instituted on day 1 ( by the study team ) and day 7 ( by the MOPH ) of the cluster investigations . The lack of focal aggregation among day 15 PCR-positive child contacts supports this notion , although the small number of those infections may have been insufficient for a meaningful analysis . Our testing method favored identification of PCR-positive mosquitoes that were infectious . The DENV incubation period in mosquitoes from the time that they imbibe an infectious blood meal to the time they become infectious ( i . e . , extrinsic incubation period ) typically lasts for 10–14 days under environmental conditions like those in Kamphaeng Phet , Thailand [2] , [36] . This implies that DENV-infectious mosquitoes in our study fed on an infected human considerably earlier than the time of cluster initiation and , thus , the transmission chain in houses with infectious mosquitoes had been taking place for some time before the “index” case was detected and the cluster investigation initiated . Consequently , as with infected children in the clusters , focal aggregation of infectious mosquitoes within the clusters may have been going on for two weeks or longer prior to initiation of each cluster investigation . So although “index” cases were used to initiate cluster investigations , they were not necessarily the first infection to occur within the cluster . Again , because vector control measures were instituted on day 1 and 7 and no further entomological collections were performed afterwards , we were not able to determine how long the focal pattern of DENV infection in mosquitoes would have persisted . We speculate that the duration of these focal areas of higher risk is limited more by the availability of susceptible humans than by susceptible mosquitoes . Future studies could investigate the required duration of interventions , which may need to be continued for one month or more . Significantly more Ae . aegypti pupae and adult females were collected from houses containing infectious mosquitoes than from those without . In addition , the risk of DENV infection in children was high in houses with infectious mosquitoes and , notably , remained elevated in neighboring houses . The higher entomological indices , however , were detected only in houses that actually contained infectious mosquitoes . These findings indicate that certain individual houses with high DENV transmission risk may disproportionately contribute to virus transmission within neighboring houses , likely due to local human and mosquito movement . Our study did not specifically evaluate when these elevated entomological measurements began or how long they persisted . They could have been present for some time prior to detection . Therefore , even in clusters with high DENV transmission , there may be individual houses that are responsible for the bulk of the transmission risk . Dengue management interventions that fail to include these individual , high-risk houses may have less impact than expected on reducing overall DENV spread . Similarly , surveillance programs that average measurements or indices of risk over a large area may fail to detect individual high-risk houses that disproportionately contribute to persistence and expansion of local transmission [25] . Fine scale spatiotemporal clustering of human-mosquito DENV transmission supports the hypothesis that DENV spread to more distant locations is driven by human movement [35] . Whether DENV is successfully transmitted at those distant locations is likely related to a suite of factors including susceptibility of the local human population , mosquito vector density and infection status , vector competence , degree of human-vector contact , and intrinsic virus factors . Locations with high levels of human movement and potential for high interaction between people and mosquitoes merit additional investigation . These components of transmission may need to be factored into dengue surveillance and control efforts more than is currently being done [37] . Results from our study have implications for strategies to prevent DENV transmission . Transmission models that address DENV spread and the impact of vaccines alone or in combination with vector control need to account for the spatiotemporal scale and dynamics of DENV transmission . Depending on the questions being asked , these models and the interpretation of surveillance data that feed into them will need to account for the presence of high-risk hotspots of human-vector virus exchange that have a high impact on DENV spread to surrounding areas [27] , [28] , [38] . These efforts should be integrated into an overall multifaceted strategy that takes into account DENV spread by movement of viremic humans among focal areas of concentrated , high levels of transmission .	Dengue is the leading cause of mosquito-borne viral infections globally . An improved understanding of the spatial and temporal distribution of dengue virus ( DENV ) transmission between humans and the principal vector , Aedes aegypti , can enhance prevention programs . Human DENV infection is known to occur at very fine spatiotemporal scales . We sought to link and quantify human DENV infections with infectious mosquitoes at these fine scales by conducting geographic cluster investigations around febrile children with and without DENV infection . We found that DENV infection in children was positively associated with houses in which infectious mosquitoes were captured . These houses also had more Ae . aegypti pupae and adult female mosquitoes than neighboring houses . However , the neighboring houses still had elevated rates of human DENV infection . Our results indicate that certain houses with high risk of DENV transmission contribute disproportionately to DENV amplification and spread to surrounding houses . At a given point in time , people and mosquitoes from a small portion of houses are responsible for the majority of DENV transmission .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "medicine", "public", "health", "and", "epidemiology", "viral", "transmission", "and", "infection", "dengue", "microbiology", "viral", "vectors", "neglected", "tropical", "diseases", "infectious", "diseases", "dengue", "fever", "biology", "vectors", "and", "hosts", "public", "health", "virology", "viral", "diseases" ]	2012	Fine Scale Spatiotemporal Clustering of Dengue Virus Transmission in Children and Aedes aegypti in Rural Thai Villages
Asymmetric strand segregation has been proposed as a mechanism to minimize effective mutation rates in epithelial tissues . Under asymmetric strand segregation , the double-stranded molecule that contains the oldest DNA strand is preferentially targeted to the somatic stem cell after each round of DNA replication . This oldest DNA strand is expected to have fewer errors than younger strands because some of the errors that arise on daughter strands during their synthesis fail to be repaired . Empirical findings suggest the possibility of asymmetric strand segregation in a subset of mammalian cell lineages , indicating that it may indeed function to increase genetic fidelity . However , the implications of asymmetric strand segregation for the fidelity of epigenetic information remain unexplored . Here , I explore the impact of strand-segregation dynamics on epigenetic fidelity using a mathematical-modelling approach that draws on the known molecular mechanisms of DNA methylation and existing rate estimates from empirical methylation data . I find that , for a wide range of starting methylation densities , asymmetric—but not symmetric—strand segregation leads to systematic increases in methylation levels if parent strands are subject to de novo methylation events . I found that epigenetic fidelity can be compromised when enhanced genetic fidelity is achieved through asymmetric strand segregation . Strand segregation dynamics could thus explain the increased DNA methylation densities that are observed in structured cellular populations during aging and in disease . Cairns proposed [1] that asymmetric strand segregation could help to minimize effective mutation rates in epithelial cells , which undergo frequent division and thus are highly susceptible to mutation . Under Cairns's model , after each round of DNA replication , the double-stranded molecule that contains the oldest DNA strand is preferentially targeted to the daughter cell that will be a somatic stem cell . The oldest DNA strands are expected to contain fewer errors than are daughter strands because some of the errors that arise on daughter strands during their synthesis fail to be repaired . Empirical findings suggest the possibility of asymmetric strand segregation in some [2]–[5] — but not all [6] , [7] — mammalian cell lineages . A few reports have discussed possible epigenetic causes and consequences of asymmetric strand segregation [8]–[15] . Klar [8] reported that epigenetic differences between DNA strands encode developmental asymmetries in fission yeast , and , more recently , suggested that breakdown of strand asymmetry could lead to disease in humans [10] . Merok et al . [11] noted that asymmetric strand segregation , which they report for cultured mammalian cells , could have consequences for the integrity of information encoded in epigenetic modifications of DNA . Cairns suggested that epigenetic changes to older strands could help to mark the stem cells that preferentially retain them [13] , and Rando [14] proposed that epigenetic modifications , including DNA methylation , could provide information that would distinguish among DNA strands of different ages . Here , I use a population-epigenetic model of an epithelial crypt to investigate in detail the potential consequences of asymmetric strand segregation for the fidelity of epigenetic information . I compared the dynamics of mean and oldest-strand methylation densities under asymmetric and symmetric strand segregation ( Figure 1 ) . Three key observations held for both high ( Figure 2 ) and low ( Figure 3 ) initial methylation densities: ( i ) when de novo methylation events were permitted to occur on both parent and daughter strands , asymmetric strand segregation resulted in population-mean and oldest-strand methylation densities that increased monotonically ( upper curves , Figures 2 and 3 ) ; ( ii ) when de novo methylation events occured on both parent and daughter strands , symmetric segregation yielded population-mean and oldest-strand methylation densities that , although dynamic , remained very near the predicted equilibrium ( middle curves , Figures 2 and 3 ) ; ( iii ) when de novo methylation events were limited to the daughter strand , population-mean and oldest-strand methylation densities under both asymmetric and symmetric segregation remained very close to starting values ( dotted lines , Figures 2 and 3 ) . Thus , for a wide range of starting methylation densities , asymmetric — but not symmetric — strand segregation leads to systematic increases in methylation levels , if parent strands are subject to de novo methylation events . The population-epigenetic model I develop here reveals that asymmetric strand segregation in somatic stem cells could lead to monotonic increases in DNA methylation densities in structured cellular populations . These increases are predicted to occur when de novo methylation occurs on parent as well as daughter strands , but not when de novo methylation events are limited to the daughter strand . The predictions of my model are made using empirical estimates of methylation rates in differentiated cells , for which substantial amounts of data are available . Further work will be necessary directly to ascertain methylation rates in somatic stem cells . Nevertheless , the essential findings of my study are consistent across a broad range of parameter values ( see , for instance , Figures 2 and 3 ) , suggesting that these results will hold even if methylation densities and rates differ appreciably between differentiated and somatic stem cells . The accumulation of aberrant methylation predicted by my model may have different time courses depending on the biological properties of a given lineage of somatic stem cells , and on the initial methylation density of a given locus . When somatic stem cell division always gives rise to one stem cell and one differentiated cell , as I model here , substantial increases in DNA methylation densities can occur over just a few cell divisions ( Figures 2 ) . When somatic stem cell division sometimes gives rise to one stem cell and one differentiated cell , and sometimes to two somatic stem cells , somewhat lower rates of increase could occur . The rate of increase will also depend on the initial DNA methylation density ( compare , for instance , Figures 2 and 3 ) . Lorincz et al . [16] found that progression to dense methylation is especially likely for genomic regions that have already attained intermediate methylation densities . In light of this finding , it seems plausible that even slow or transient increases in DNA methylation could raise methylation densities to a threshold sufficient to trigger more substantial increases . What might be the functional implications of the increased DNA methylation densities predicted under asymmetric strand segregation ? The accumulation of methyl groups on a long-lived DNA strand could serve as a signal to guide asymmetric strand segregation itself [17] , or to distinguish stem cells from differentiated cells [13] . My findings could also help to explain the positive correlation observed between age and methylation density in endometrial [18] and intestinal [19] tissues . Both of these are rapidly-dividing tissues of the sort initially predicted by Cairns [1] , and reported by some groups [4] , to undergo asymmetric strand segregation . In contrast , slowly-dividing cells , such as those in the hematopoetic lineage , have constant methylation densities [20]–[23] and have been reported not to undergo asymmetric strand segregation [6] . Thus , the systematic increases in DNA methylation densities predicted here may be specific to the rapidly-dividing lineages Cairns initially discussed [1] . My results may also have implications for the etiology of cancer in humans . Several epithelial cancers are associated with reductions in epigenetic fidelity , including the accumulation of aberrant methylation and abnormal gene silencing [24] , [25] . Barrett's esophagus illustrates the potential relevance of these findings . The esophageal epithelium in Barrett's esophagus contains abnormal intestinal crypt-like structures , and is characterized by abrupt increases in DNA methylation densities and consequent silencing of loci critical to cell-cycle regulation [26] . Thus , it is possible that directional change in epigenetic information may be a cost of the increased genetic fidelity achieved through asymmetric strand segregation , with implications for human disease . I developed a simplified model of an epithelial crypt with which to track methylation dynamics ( Figure 1 ) . Each crypt consists of one somatic stem cell , and four differentiated cells . At each round of stem-cell division , one terminally differentiated cell is produced , and one stem cell is produced . The top-most of the terminally differentiated cells is sloughed off at the epithelial surface . Segregation of the oldest DNA strand always to the stem cell characterizes asymmetric strand segregation ( Figure 1a ) ; segregation of the oldest DNA strand at random to the stem and terminally differentiated cells characterizes symmetric segregation ( Figure 1B ) . ( 1 ) I developed vector ( 1 ) to record the methylation density of each of the 10 individual strands of the five DNA molecules in the epithelial crypts ( Figure 1 ) . The initial state of the epithelial crypt is given by . The first element of ( 1 ) , “1” , is used to implement the assumption that individual methyl groups are not lost once they are incorporated . Elements two through eleven represent the methylation densities of the ten individual strands of DNA in the five double-stranded molecules of the crypt cells . These elements are best considered in pairs . Elements two and three represent the methylation states of the parent and daughter strands in the founding somatic stem cell . In particular , the second element gives the methylation density of the oldest parent strand at the start of the simulation . This strand is assumed to have started with methylation density , as calculated under our previous model [31] from the chosen parameter values for maintenance and de novo methylation events . It then acquired additional methyl groups through parent strand de novo methylation events occurring at rate , for those cases where is greater than 0 . The third element gives the initial methylation density of the daughter strand in the founding somatic stem cell , as calculated from the starting methylation density of the parent strand , , using our earlier model [31] . Elements four and five represent methylation states of the parent and daughter strands in the first of the four differentiated cells , and so on up to elements ten and eleven , which give the methylation states of the parent and daughter strands in the differentiated cell closest to the epithelial surface . Because I start by modelling the establishment of the crypt , the strands represented by vector positions three through eleven have not yet been synthesized at the start of the simulation , and therefore have methylation densities of zero . Methylation densities of strands that do not yet exist are excluded from the calculation of population-mean densities . ( 2 ) I developed matrix ( 2 ) to model the occurrence of methylation events on individual strands of DNA molecules , and to track their progression through the simplified epithelial crypt described above . The first row of matrix is a placeholder , and is used to regenerate the “1” that is the first element of vector . The second row of the matrix is used to calculate the updated methylation density of the oldest DNA strand . The third row of the matrix is used to simulate methylation events on the daughter strand that has just been produced through replication of the oldest DNA strand . This density is determined by the maintenance methylation rate , , and the daughter-strand de novo methylation rate , . The fourth row of the matrix is used to simulate methylation events on the parent strand in the newest differentiated cell , and records methylation events that occur through parent-strand de novo methylation , at rate . The fifth row of the matrix is used to simulate methylation events on the daughter strand in the newest differentiated cell , and records methylation events that occur by daughter-strand de novo methylation , at rate . The sixth through eleventh rows of the matrix are used to simulate the progression of existing DNA molecules through the crypt . As noted above , we assume that methyl groups are never removed from a strand once they have been added , and are added to a strand either during the round in which it is synthesized , or during subsequent rounds in which it serves as a parent strand in DNA replication . Rows 6 , 8 and 10 simulate the movements of strands that are parents in their respective cells; rows 7 , 9 and 11 simulate the movements of strands that are daughters in their respective cells . Upon replication and cell division , cells containing the various DNA molecules advance one cell-position toward the epithelial surface . Both asymmetric and symmetric strand segregation are modelled using matrix , with slight differences in the treatment of the resulting updated . To model a single round of DNA replication and cell division under asymmetric strand segregation , I multiplied matrix by vector . To investigate methylation trajectories over multiple rounds of DNA replication and cell division , I multiplied matrix by vector recursively up to 500 times . This large number of divisions would be unreasonable for many tissues , but is likely appropriate for the rapidly-dividing cells of the endometrium [18] and intestinal epithelium [38] over a period of several years [39] . To model a single round of DNA replication and cell division under symmetric strand segregation , I used a similar approach , multiplying matrix by vector , but included for each round a random draw of a number between 0 and 1 . When the random number was less than or equal to 0 . 5 , I retained the vector that resulted from this initial multiplication , simulating retention of the oldest DNA strand in the somatic stem cell . When the random number was greater than 0 . 5 , I simulated the export of the oldest strand to the differentiated daughter cell by exchanging the methylation densities given in vector positions two and three with those given in vector positions four and five . I repeated this process of multiplication , random number selection , and vector rearrangement for up to 500 rounds of DNA replication , methylation events , and cell division . Several authors have investigated the rate of maintenance methylation , , yielding estimates that range from 0 . 95 to 0 . 999 [31] , [40]–[43] ( Fu et al . , Manuscript in Preparation ) . Here , I assumed . Comparatively few studies have investigated the rate of de novo methylation . The estimates that do exist exhibit substantial variation . Pfeifer et al . [40] estimated a de novo methylation rate of 0 . 05 , Laird et al . [41] estimated a rate of 0 . 17 , under the assumption that de novo events are limited to the daughter strand . Results from Genereux et al . [31] suggest that part of the variation in these estimates of the de novo methylation rate may be attributable to bona fide biological variation among CpG cytosine sites , and perhaps among loci . For de novo methylation rates , I first chose parameter values that yield an expected equilibrium methylation density of 0 . 8 under the model described previously [31] , and assuming . To meet this condition , the sum of parent , and daughter , , de novo rates must be 0 . 1 . To accommodate uncertainties about the strandedness of de novo methylation events , I explored two cases: de novo events occurring on the daughter strand only ( , ) , and de novo events occurring at equal rates on the parent and daughter strands ( , ) . These values were used to generate the results in Figure 2 . To investigate whether or not the change in methylation density observed when starting with was limited to scenarios with high initial densities , I also conducted simulations spanning a range of initial values . For the simulation shown in Figure 3 , I started with a methylation density of , and assumed the maintenance rate , , to be 0 . 975 . To meet these conditions under our previous model [31] , the sum of parent , , and daughter , , de novo rates must be 0 . 0028 . Here , as before , I considered parameter values that included de novo methylation events on the parent strand ( ) , and parameter values that limited de novo methylation events to the daughter strand ( ) .	Through my investigations of the fidelity of epigenetic inheritance , I became intrigued by the interplay of genetic and epigenetic fidelities . Cairns proposed in 1975 that the lifetime risk of epithelial cancers would be reduced if chromosomes containing the oldest DNA strands were selectively segregated to somatic stem cells . I wondered about the implications of such asymmetric strand segregation for the fidelity of epigenetic information . To address this issue , I modelled the partitioning of DNA molecules after replication , with special attention to the molecule that contained the oldest strand . I found that the enhanced genetic fidelity that may be achieved through asymmetric strand segregation could , under some scenarios , compromise epigenetic fidelity . I am excited to pursue these studies as they apply to epigenetic changes observed to occur during aging and in human diseases , including several cancers .	[ "Abstract", "Introduction", "Results", "Discussion", "Models" ]	[ "developmental", "biology/aging", "cell", "biology/developmental", "molecular", "mechanisms", "evolutionary", "biology", "genetics", "and", "genomics/epigenetics", "genetics", "and", "genomics/cancer", "genetics" ]	2009	Asymmetric Strand Segregation: Epigenetic Costs of Genetic Fidelity?
Transcriptional silencing by heritable cytosine-5 methylation is an ancient strategy to repress transposable elements . It was previously thought that mammals possess four DNA methyltransferase paralogs—Dnmt1 , Dnmt3a , Dnmt3b and Dnmt3l—that establish and maintain cytosine-5 methylation . Here we identify a fifth paralog , Dnmt3c , that is essential for retrotransposon methylation and repression in the mouse male germline . From a phenotype-based forward genetics screen , we isolated a mutant mouse called ‘rahu’ , which displays severe defects in double-strand-break repair and homologous chromosome synapsis during male meiosis , resulting in sterility . rahu is an allele of a transcription unit ( Gm14490 , renamed Dnmt3c ) that was previously mis-annotated as a Dnmt3-family pseudogene . Dnmt3c encodes a cytosine methyltransferase homolog , and Dnmt3crahu mutants harbor a non-synonymous mutation of a conserved residue within one of its cytosine methyltransferase motifs , similar to a mutation in human DNMT3B observed in patients with immunodeficiency , centromeric instability and facial anomalies syndrome . The rahu mutation lies at a potential dimerization interface and near the potential DNA binding interface , suggesting that it compromises protein-protein and/or protein-DNA interactions required for normal DNMT3C function . Dnmt3crahu mutant males fail to establish normal methylation within LINE and LTR retrotransposon sequences in the germline and accumulate higher levels of transposon-derived transcripts and proteins , particularly from distinct L1 and ERVK retrotransposon families . Phylogenetic analysis indicates that Dnmt3c arose during rodent evolution by tandem duplication of Dnmt3b , after the divergence of the Dipodoidea and Muroidea superfamilies . These findings provide insight into the evolutionary dynamics and functional specialization of the transposon suppression machinery critical for mammalian sexual reproduction and epigenetic regulation . Transposable elements have been described as ‘dark energy’ that acts both as a creative force , by giving rise to new genes and regulatory elements , and as a threat , by disrupting genome architecture [1] . Transposons occupy roughly 46% and 38% of the human and mouse genomes , respectively , and retrotransposons are the predominant class , including some that continue to be retrotransposition-competent [2–4] . Genomes have co-evolved with their retrotransposons and thus have multiple defense mechanisms to restrain transposon activity [4] . This restraint is of utmost importance in the germline , where retrotransposon activity not only facilitates vertical transmission , but also threatens genome integrity and germ cell viability . The primary means of retrotransposon suppression is transcriptional silencing via cytosine-5 methylation [1 , 5] . In the mouse male germline , after a developmentally programmed methyl-cytosine erasure , retrotransposon methylation is re-established in prospermatogonia prior to birth [6–9] . Shortly after birth , spermatogenesis initiates and spermatogonia enter a meiotic cell cycle , which encompasses an extended prophase . During meiotic prophase , the spermatocyte genome experiences developmentally programmed DNA double-strand breaks ( DSBs ) that are subsequently repaired by homologous recombination . Contemporaneously , homologous chromosomes align and build a proteinaceous structure called the synaptonemal complex ( SC ) , which holds recombining chromosomes together ( synapsis ) . Meiosis is completed by two successive divisions , forming haploid spermatids that enter the final developmental phase of spermiogenesis , culminating with the production of sperm [10] . Failure to methylate retrotransposons leads to abnormal retrotransposon expression , spermatogenic arrest in meiotic prophase , and sterility [11 , 12] . Mammals possess three known enzymes that can catalyze DNA cytosine-5 methylation: DNMT1 , DNMT3A , and DNMT3B . DNMT1 maintains DNA methylation by acting on hemimethylated DNA during replication , and de novo methylation is established by DNMT3A and DNMT3B [13–15] . The mammalian DNMT3 family also includes DNMT3L , a catalytically inactive adaptor that lacks sequence motifs essential for cytosine-5 methylation [16–18] . DNMT3L interacts with and stimulates the activity of DNMT3A and DNMT3B [17 , 19–23] . DNMT3L has also been implicated in the recognition of target sequences via interaction with histone H3 [24 , 25] . Retrotransposon methylation in the germline is established by the concerted activities of DNMT3A , DNMT3B , and DNMT3L [7 , 17 , 26 , 27] . Whereas mouse Dnmt3a and Dnmt3b are essential for development [15] , Dnmt3l-deficient males are viable but sterile [17 , 18] . They fail to methylate retrotransposons and accumulate retrotransposon-derived transcripts in both spermatogonia and spermatocytes . Abnormal retrotransposon expression is accompanied by defects in chromosome synapsis , inability to complete meiotic recombination , meiotic prophase arrest , and a complete absence of spermatids [11 , 17 , 27–29] . Here , we identify a new , fourth member of the Dnmt3 family in mice . Using a forward genetics approach , we isolated a male-sterile mutant that we named rahu , for “recombination-affected with hypogonadism from under-populated testes” ( Rahu is a harbinger of misfortune and darkness in Vedic mythology ) . rahu mapped to a missense mutation in the predicted Dnmt3 pseudogene Gm14490 . Re-named Dnmt3c , this gene is required for the methylation and repression of retrotransposons in the male germline . Dnmt3c was also independently discovered by Bourc’his and colleagues [30]; importantly , our results validate and expand their findings on the developmental role of Dnmt3c . We propose that Dnmt3c encodes a de novo DNA methyltransferase that diversified from DNMT3B during evolution of the Muroidea and that became functionally specialized for germline defense against retrotransposon activity . We performed a random mutagenesis screen to recover mutants with autosomal recessive defects in meiosis . Male mice of the C57BL/6J ( B6 ) strain were mutagenized with the alkylating agent N-ethyl-N-nitrosourea ( ENU ) [31–33] , then a three-generation breeding scheme was carried out to “homozygose” recessive mutations , including outcrossing to FVB/NJ ( FVB ) mice to facilitate downstream genetic mapping ( Fig 1A ) [32 , 34] . First , ENU-mutagenized B6 males were crossed to wild-type FVB females to generate founder males ( F1 ) that are potential carriers of a mutation of interest . Then , each F1 male was crossed to wild-type FVB females to generate second-generation females ( G2 ) ; if the F1 male was heterozygous for a mutation of interest , half of his daughters should be carriers . Finally , G2 daughters were crossed back to their F1 father to generate third-generation males ( G3 ) , of which one eighth should be homozygous for the mutation . We screened juvenile ( 15–19 days post-partum ( dpp ) ) G3 males for meiotic defects , focusing on recombination or its interdependent process , chromosome synapsis . To this end , we immunostained squashed spermatocyte nuclei for well-established meiotic markers whose deviations from wild-type patterns are diagnostic of specific defects: SYCP3 , a component of axial elements and of lateral elements of the SC [10 , 35 , 36] , and phosphorylated histone H2AX ( γH2AX ) , which forms in response to meiotic DSBs [37] . As axial elements begin to form during the leptotene stage of meiotic prophase in wild type , SYCP3 staining appears as dots or short lines; axial elements elongate during zygonema and the first stretches of tripartite SC appear; SC juxtaposes autosomes all along their lengths in pachynema; and the SC begins to disassemble in diplonema ( Fig 1B ) . Most DSBs are formed in leptonema and zygonema , resulting in florid staining for γH2AX at these stages , but this signal disappears from autosomes over the course of zygonema and pachynema as chromosomes synapse and DSBs are repaired by recombination ( Fig 1B ) . γH2AX also accumulates in a recombination-independent manner in the sex body , a heterochromatic domain encompassing the sex chromosomes [37–40] ( clearly visible in pachynema and diplonema in Fig 1B ) . Mutants with defects in recombination and/or SC formation deviate from these wild-type patterns in diagnostic ways [10 , 36] . To streamline the screening process , we first used the SYCP3 staining pattern to classify meiotic prophase spermatocytes as either early prophase-like ( leptonema or early zygonema ) or later prophase-like ( late zygonema , pachynema or diplonema ) . The γH2AX staining pattern was then evaluated to determine whether DSBs were formed and repaired properly and whether sex body formation appeared normal . From F1 founder lines screened in this manner , we isolated a mutant line with SYCP3 and γH2AX patterns symptomatic of defects in meiotic DSB repair and/or synapsis ( Fig 1B and 1C ) . Four of 28 G3 males from this line displayed a mutant phenotype in which spermatocytes with SYCP3 staining characteristic of early prophase I were abundant but later stages were absent or greatly depleted , indicating a block to meiotic progression ( Fig 1B and 1C and S1 Table ) . Early prophase-like spermatocytes showed nucleus-wide γH2AX staining similar to wild type , consistent with formation of meiotic DSBs ( Fig 1B ) . However , the mutants also had elevated numbers of abnormal spermatocytes displaying nucleus-wide γH2AX along with longer tracks of SYCP3 staining that were consistent with varying degrees of synapsis ( Fig 1B and 1C ) . This pattern is a hallmark of recombination- and synapsis-defective mutants such as Msh5-/- and Sycp1-/- [36 , 37 , 41–43] . Adult male rahu mutants were sterile , with none of the four animals we tested siring progeny . The mutant males displayed pronounced hypogonadism , with a 67% reduction in testes-to-body-weight ratio compared to littermates ( mean ratios were 0 . 20% for rahu mutants and 0 . 62% for wild-type and heterozygous animals; p<0 . 01 , one-sided Student’s t-test; Fig 2A ) . In histological analysis of testis sections , seminiferous tubules from adult rahu mutants contained only early spermatogenic cells and completely lacked post-meiotic germ cells ( Fig 2B ) . The most developmentally advanced cell types visible appeared apoptotic ( Fig 2B ) , and increased apoptosis was confirmed by TUNEL staining ( Fig 2C ) . The patterns suggest that spermatocyte arrest and apoptosis occur during mid-pachynema , possibly at stage IV of the seminiferous epithelial cycle , which is typical of mutants unable to properly synapse their chromosomes , repair meiotic DSBs , and/or transcriptionally silence their sex chromosomes [43–46] . Stage IV arrest is also observed in Dnmt3l mutants , which lack germline DNA methylation [17 , 28] . In crosses with rahu heterozygous males , rahu homozygous females were fertile with an average litter size of 8 . 3 ( 19 litters from 10 dams ) , similar to rahu heterozygous females with an average litter size of 8 . 1 ( 29 litters from 10 dams ) . The rahu allele segregated in an expected Mendelian ratio: 26% wild type , 48% heterozygotes , and 26% mutant homozygotes from heterozygous dams ( n = 119 ) ; 49% heterozygotes and 51% mutant homozygotes from homozygous mutant dams ( n = 101 ) . Homozygous mutants survived to adulthood and no morphological abnormalities have been apparent . We conclude that rahu does not severely impair female meiosis , embryonic development , or adult somatic functions . We mapped the likely rahu causative mutation using a genetic polymorphism-based positional cloning strategy [34 , 47] . Because the mutagenized B6 mice were outcrossed to FVB mice ( Fig 1A ) , ENU-induced mutations should be linked to DNA sequence polymorphisms from the B6 background , so G3 males homozygous for recessive phenotype-causing mutations should also be homozygous for linked B6 polymorphisms . Furthermore , mutant mice from the same line should be homozygous for at least some of the same linked B6 polymorphisms , whereas related but phenotypically normal mice should not . We therefore searched for genomic regions of B6 single-nucleotide polymorphism ( SNP ) homozygosity that are shared between mutant rahu mice and not shared with phenotypically normal mice from the rahu line . We first coarsely mapped such regions by hybridization of genomic DNA from seven rahu mutants to a mouse SNP genotyping microarray . This yielded a single cluster of SNPs spanning 33 . 58 Mbp flanked by heterozygous SNPs gnf02 . 126 . 027 ( Chr2:127 , 800 , 747 ) and rs3664408 ( Chr2:161 , 380 , 222 ) ( Fig 3A ) . Next , whole-exome sequencing of mutants identified seven un-annotated homozygous DNA sequence variants within the 33 . 58-Mbp mapped region ( Fig 3A ) . To determine which was the likely causal mutation , we manually genotyped sequence variants within the 33 . 58-Mbp region in both mutant and phenotypically normal mice , targeting strain polymorphisms as well as the presumptive ENU-induced lesions themselves ( Fig 3A ) . Presumptive ENU-induced lesions that were homozygous in phenotypically normal mice or that were heterozygous in meiosis-deficient mutants could be excluded as candidates . We applied this strategy to ~100 additional G3 and G4 mice , which allowed us to winnow the phenotype-causing mutation to within a 17 . 43-Mbp region flanked by the sequence change in the Rrbp1 gene ( Chr2:143 , 947 , 738 ) and SNP rs3664408 ( Chr2:161 , 380 , 222 ) ( Fig 3A ) . This smaller region contained only one novel sequence variant , within a gene model named Gm14490 ( NCBI Gene ID: 668932 and Ensembl Gene ID: ENSMUSG00000082079 ) . The rahu lesion in Gm14490 is an A to G nucleotide transition at position Chr2:153 , 727 , 342 ( Fig 3B ) . Surprisingly , however , Gm14490 was annotated as a pseudogene . It was first identified as a paralog of Dnmt3b , but the absence of expressed sequence tags and the presence of stop codons in the available gene build led previous researchers to conclude that Gm14490 was a pseudogene [48] . However , more recent gene builds predicted a gene structure , encompassing introns with an open reading frame , that is matched by testis transcripts ( Fig 3B–3D ) . Mouse ENCODE RNA-sequencing data from adult testis revealed splice junctions within 5 bp of the boundaries for all exons except exon 5 , suggesting that the predominant Gm14490 transcript isoform ( s ) in adult testis does not contain exon 5 ( Fig 3B and 3C ) . In the adult mouse , Gm14490 is expressed in testis , with little or no expression in most somatic tissues other than brain ( Fig 3D and 3E ) . Gm14490 is predicted to yield a 2 , 218-nucleotide transcript ( Ensembl Transcript ID: ENSMUST00000119996 ) containing 19 exons . The rahu point mutation is located in exon 18 ( Fig 3B ) . The available expression data and the manifestation of a phenotype led us to surmise that Gm14490 is not a pseudogene . To test this hypothesis and to confirm that the identified point mutation is causative for the rahu phenotype , we generated targeted endonuclease-mediated ( em ) alleles of Gm14490 by CRISPR/Cas9-mediated genome editing using a guide RNA ( gRNA ) targeted to exon 4 ( Fig 3B ) . We analyzed four frameshift-generating alleles ( em1 , em2 , em3 , em4 ) and one in-frame deletion allele ( em5 ) , which results in a single-amino-acid deletion . We expected that frameshift mutations in an exon near the N-terminus of Gm14490 are likely to lead to loss of function , whereas the single-amino-acid deletion would not . As predicted , young adult males carrying two copies of frameshift alleles had diminutive testes , similar to rahu mutants ( Fig 2A ) . The in-frame deletion allele em5 alone did not confer hypogonadism ( homozygote mean = 0 . 55% , heterozygote mean = 0 . 57% ) . The em2 homozygote showed reduced testes-to-body-weight ratios compared to compound heterozygotes also carrying the in-frame deletion em5 allele ( 73% reduction; em2 homozygote = 0 . 15% , em2/em5 compound heterozygote mean = 0 . 57% ) , as did the em2/em3 compound heterozygote ( 64% reduction; em2/em3 = 0 . 20% , em2/em5 and em3/em5 compound heterozygotes mean = 0 . 55% ) . Frameshift alleles did not complement the hypogonadism phenotype when crossed to rahu ( Fig 2A ) . rahu/em1 and rahu/em4 compound heterozygotes had significantly reduced testes-to-body-weight ratios compared to rahu/em5 compound heterozygotes ( 70% and 69% reduction , respectively; rahu/em1 mean = 0 . 17% , rahu/em4 mean = 0 . 17% , rahu/em5 mean = 0 . 58%; p<0 . 01 , one-sided Student’s t-test ) . Furthermore , adults with two frameshift alleles had depleted seminiferous tubules lacking post-meiotic germ cells , and the frameshift alleles did not complement the spermatocyte arrest phenotype when crossed to rahu ( Fig 2D and 2E ) . Crosses between em1/rahu compound heterozygous females and em1/+ heterozygous males gave litters ( average size 9 . 3 pups; six litters from four dams ) . Also , em2/em2 mutant females bred with em2/+ heterozygote carrier males produced progeny ( average litter size of six pups; two litters from two dams ) . Thus , similar to rahu , the frameshift mutations do not cause infertility in females . We conclude that rahu is allelic to the Gm14490 frameshift mutations and that rahu is likely to be a null or near-null for function of the gene . We further conclude that Gm14490 is not a pseudogene and that it is essential during meiotic prophase in spermatocytes . On the basis of sequence homology to Dnmt3b and functional data shown below , we refer to Gm14490 henceforth as Dnmt3c , also in keeping with a recent independent study [30] . Dnmt3c is predicted to encode a 739-aa protein with 77% overall similarity to the DNA cytosine methyltransferase DNMT3B ( EMBOSS Water tool , Smith-Waterman algorithm [49–51] ) . It contains a cysteine-rich ATRX–DNMT3–DNMT3L ( ADD ) domain with 98 . 3% sequence similarity to that of DNMT3B , and a 96 . 8% similar DNA cytosine methyltransferase domain ( Fig 4A and S2 Table ) . DNMT3C contains matches to both the sequence and arrangement of the six highly conserved cytosine methyltransferase domain motifs ( I , IV , VI , VIII , IX and X ) that are characteristic of active DNA methyltransferases ( Fig 4A and S2 Table ) [13 , 52] . Among the mammalian DNMT3 family members , DNMT3C shares most similarity with DNMT3B ( Fig 4B ) , although it lacks a clear match to the Pro-Trp-Trp-Pro ( PWWP ) domain in DNMT3B ( Fig 4A ) [53] . The rahu mutation causes a glutamic acid to glycine substitution at position 692 within motif IX ( Fig 4A and 4C ) . These data suggest that Dnmt3c encodes a novel DNA methyltransferase and that the Dnmt3crahu mutant phenotype is a consequence of perturbing its methylation function in the male germline . In a crystal structure of the carboxy-terminal domains of DNMT3A and DNMT3L , these peptides form a tetrameric complex with two DNMT3A and DNMT3L dimers , which further dimerize through DNMT3A-DNMT3A interaction [19] . Mutating residues at the DNMT3A-DNMT3A interface abolishes activity [54] . Homology-based modeling of DNMT3C places E692 near the potential dimerization interface ( Fig 4D and 4E ) , as well as near the inferred DNA recognition region [19] . Thus , the E692G mutation in DNMT3C may interfere with protein-protein interactions and/or protein-DNA interactions that are required for normal DNMT3C activity . A fundamental role of DNA methylation in the germline is to silence retrotransposons [27] , so we examined the expression of long interspersed nuclear element-1 ( L1 ) and intracisternal A particle ( IAP ) retrotransposon families , which are known to be active in the germline [5 , 55–57] . Because the mutants undergo spermatogenic arrest , we examined animals at 14 dpp . At this age , testes predominantly contain spermatogonial stem cells and early meiotic cells , with the most advanced stage being pachynema [58] . Dnmt3crahu mutants displayed increased expression of both L1 ( ~9-fold ) and IAP ( ~3-fold ) transcripts as assessed by quantitative RT-PCR ( Fig 5A ) ( mean fold change of 8 . 5 using L1 ORFs primers and 8 . 7 using L1 ORF2 primers; mean fold change of 3 . 6 using IAP Gag primers and 2 . 1 using IAP 3′ LTR primers ) . We confirmed these findings by immunostaining testis sections for proteins encoded by L1 and IAP retrotransposons . In seminiferous tubules of heterozygotes , L1-encoded ORF1p was detectable at low levels and IAP-encoded Gag was barely detectable . However , both proteins accumulated to substantially higher levels in Dnmt3crahu mutant testes ( Fig 5B ) . Tubules of mutant juveniles contained L1 ORF1p in spermatocytes and IAP Gag in spermatogonia . Tubules from mutant adults contained prominent immunofluorescence signal for both proteins in spermatocytes . Thus , disruption of Dnmt3c derepresses expression of L1 and IAP retrotransposons in the male germline . Next we assayed transposon DNA methylation directly by methylation-sensitive digestion and Southern blotting . Genomic DNA from juvenile Dnmt3crahu mutants and wild-type littermates was digested with HpaII , for which DNA cleavage is blocked by CpG methylation within its recognition site , or with its methylation-insensitive isoschizomer MspI as a control . Digested DNA was separated on agarose gels and hybridized on Southern blots to a probe derived from the 5′ UTR of L1_MdA ( A family of L1 ) sequences . DNA purified from somatic tissue ( tail ) was highly resistant to cleavage by HpaII , whether purified from homozyogous mutants or a wild-type littermate ( Fig 5C ) . This indicates that Dnmt3c is dispensable for DNA methylation in the soma , at least for L1Md_A elements . Testis DNA was also almost completely resistant to cleavage with HpaII when purified from wild type , but substantially less so when purified from Dnmt3crahu homozygous mutants ( Fig 5C ) . We conclude that Dnmt3c is needed to establish normal levels of DNA methylation within L1Md_A elements , specifically in the germline . Taken together , these findings support a germline-specific function for Dnmt3c in retrotransposon DNA methylation and transcriptional repression . To more thoroughly assess the contribution of Dnmt3c in repressing retrotransposons , we performed RNA-seq on whole-testis samples from the same 14-dpp-old Dnmt3crahu mutant and heterozygous littermates analyzed by RT-PCR . The expression levels of distinct LINE and LTR families were up-regulated in Dnmt3crahu mutants , but SINE elements appeared to be changed little , if at all ( Fig 6A ) . Because of variable expression between littermates of the same genotype , we analyzed the fold change of median expression values between mutants and heterozygotes ( Fig 6B ) . Retrotransposons belonging to the L1 and ERVK superfamilies showed the strongest derepression . Specifically , L1 families L1Md_Gf , L1Md_T , L1_MdA and L1_Mm were up-regulated 12 . 1- , 9 . 6- , 6 . 4- and 2-fold , respectively . ERVK families IAPEZ-int , IAPLTR1_Mm , MMERVK10C-int , RLTR10C and IAPA_MM-int were up-regulated 10 . 1- to 4 . 5-fold . Two ERVK elements , IAPLTR3 and IAPLTR3-int were down-regulated in Dnmt3crahu mutants ( by 2 . 8- and 4 . 1-fold respectively ) . The most affected L1 families in Dnmt3crahu mutants are the same as those derepressed in Dnmt3l mutants , namely , young L1 families of the A , T and Gf subtypes [29 , 59] . Comparison of the expression fold change in Dnmt3l–/–knockouts with Dnmt3crahu mutants showed a striking overlap in effects for both the L1 and the LTR families ( Fig 6C ) . This overlap is not an indirect effect of down-regulated Dnmt3l expression in Dnmt3crahu mutants , as RNA-seq coverage for neither Dnmt3l nor Dnmt3c was significantly changed in Dnmt3crahu mutants ( Dnmt3l fold change 1 . 2 , Dnmt3c fold change 1 . 0 ) . To determine the global contribution of Dnmt3c to retrotransposon methylation , we performed whole-genome bisulfite sequencing ( WGBS ) on whole-testis samples from six 12-dpp-old Dnmt3crahu mutant and wild-type littermates . CpGs within genes , CpG islands and SINE elements were mostly non-differentially methylated in Dnmt3crahu mutants ( Fig 7A ) . An increase in the proportion of differentially methylated CpGs , specifically hypomethylated CpGs , was observed within LINE and LTR elements ( 16 . 04% and 6 . 62% hypomethylated CpGs , respectively ) ( Fig 7A ) . Meta-plots averaging over specific element types showed that LINE elements were on average hypomethylated near their 5′ ends , while LTR elements were hypomethylated on average across their entire bodies ( Fig 7B ) . Comparison of the mean methylation levels of individual retrotransposon families showed that LINE elements belonging to the L1 superfamily and LTR elements belonging to the ERVK and ERV1 superfamilies were significantly hypomethylated in Dnmt3crahu mutants ( Fig 7C ) . In addition , one LTR element belonging to the ERVL-MaLR family , MLT1H2-int , was significantly changed . Consistent with a role for Dnmt3c-dependent DNA methylation in retrotransposon transcriptional repression , the specific retrotransposon families that were derepressed in Dnmt3crahu mutants were also differentially methylated . For example , differentially expressed L1 families L1Md_Gf , L1Md_T , L1_MdA , and L1_Mm were significantly hypomethylated by 17 . 4% , 10 . 9% , 6 . 6% , and 2% , respectively ( p <0 . 01 , two-sided Student’s t-test ) ( Fig 7C ) . Differentially expressed ERVK families IAPLTR1_Mm and RLTR10C were significantly hypomethylated by 9 . 8% and 19 . 1% , respectively ( p <0 . 01 , two-sided Student’s t-test ) , and IAPEZ-int , MMERVK10C-int , and IAPA_MM-int were hypomethylated by 7 . 6% , 8 . 6% and 12 . 6% , respectively ( p-value of 0 . 021 , 0 . 025 , and 0 . 020 , respectively; two-sided Student’s t-test ) ( Fig 7C , S1 Fig and S3 Table ) . Dnmt3c is located directly adjacent to Dnmt3b in the mouse genome , and the two genes share 50 . 4% DNA sequence identity ( EMBOSS Water tool , Smith-Waterman algorithm [49–51] ) . Dot-plot analysis of the genomic region encompassing both genes showed that the sequence identity shared between Dnmt3c and Dnmt3b extends over long stretches ( up to 436 bp of 100% sequence identity and 2 , 805 bp of >95% sequence identity; appearing as horizontal lines in Fig 8A ) . Sequence similarity begins ~3 , 300 bp upstream of the annotated start of Dnmt3c , which matches the intronic region between Dnmt3b exons two and three . Similarity extends ~100 bp beyond the last annotated exon of Dnmt3c and matches the 3′ UTR of Dnmt3b , suggesting that these regions encode functional elements that have constrained sequence divergence ( Fig 8B ) . Dnmt3c and Dnmt3b are remarkably similar with respect to their exon organization ( Fig 8B ) , with stretches of sequence identity encompassing all Dnmt3c exons , except exons two and five . The sequence similarity in Dnmt3c is most extensive near its 3′ end ( exons eight to nineteen ) , which encodes the ADD domain and methyltransferase motifs . As expected from the absence of the PWWP domain in DNMT3C , two of the three exons encoding this domain in Dnmt3b do not have obvious matches in the corresponding part of the genomic sequence of Dnmt3c ( Fig 8B ) . Although Dnmt3c and Dnmt3b share extensive nucleotide identity , they are functionally specialized: whereas Dnmt3b establishes methylation both in somatic cells during embryogenesis and in the germline [7 , 15 , 26] , Dnmt3c appears to function solely in the germline . Also , mouse and human DNMT3B protein sequences cluster separately from mouse DNMT3C ( Fig 4B ) , suggesting that Dnmt3c and Dnmt3b have been evolving independently within the mouse lineage . These properties are consistent with a local gene duplication event followed by functional diversification . Dot-plot comparisons of the mouse genomic region containing Dnmt3c and Dnmt3b with homologous regions in rat and human showed that the duplication is also present in rat ( appearing as a continuous central diagonal plus two off-center partial diagonals indicated by the red arrows in Fig 8C ) but is absent in human ( appearing as two offset diagonals marked by red arrowheads in Fig 8C ) . To determine the ancestry of this duplication event , we analyzed species from related rodent families ( Fig 8C and 8D ) [60] ( UCSC Genome Browser ) . Dot-plot comparisons showed clear evidence of the Dnmt3c and Dnmt3b duplicate pair in Upper Galilee Mountains blind mole rat , Chinese hamster , and prairie vole , but not in lesser Egyptian jerboa or more distantly related rodents ( Fig 8C and 8D ) . These findings indicate that the tandem duplication of Dnmt3b occurred between 55 and 45 million years ago , after the divergence of the Dipodoidea and Muroidea rodent superfamilies , but before the divergence of the Cricetidae and Muridae families [61] . This study illustrates the utility of forward genetic screens in identifying novel meiotic genes in mouse and reports the identification of Dnmt3c , a new gene essential for mouse spermatogenesis and genome regulation . We find that Dnmt3c functions in the methylation and subsequent repression of retrotransposons in the male germline , and that it arose by tandem duplication of the Dnmt3b gene . While this work was in progress , analogous findings were reported independently [30] , and results from both studies agree well . Uniquely , we have isolated a novel point-mutated allele of Dnmt3c ( rahu ) that harbors a non-synonymous mutation of a conserved residue , and performed expression and epigenetic profiling of this mutant . The presence of the six highly conserved cytosine methyltransferase motifs in DNMT3C and the methylation defect observed in Dnmt3crahu mutants , which harbor a mutation within one motif ( E692G ) , suggest that DNMT3C is enzymatically active . Indeed , DNMT3C methylates cytosines in vitro , and expression of Dnmt3c in Dnmt1 Dnmt3a Dnmt3b triple knockout embryonic stem cells leads to a gain in DNA methylation [30] . We found that homology-based modeling of the DNMT3C methyltransferase domain using the DNMT3A crystal structure places the E692 amino acid at a potential dimerization interface , as well as near the potential DNA binding interface [19] . Thus , the rahu mutation may compromise protein-protein interactions and/or protein-DNA interactions of DNMT3C . Intriguingly , a mutation affecting the corresponding region in DNMT3B is observed in patients suffering from immunodeficiency , centromeric instability and facial anomalies ( ICF ) syndrome , an autosomal recessive disease caused by mutations in DNMT3B [15 , 62] . The patient-derived mutation results in a three-amino-acid ( serine-threonine-proline ) insertion immediately downstream of the corresponding glutamic acid that is mutated in Dnmt3crahu mutants ( DNMT3B Uniprot VAR_011502 ) . Expression of recombinant mutant DNMT3B harboring this insertion in cell lines indicated that the mutation does not severely compromise protein stability . Rather , nuclear localization patterns were abnormal , suggesting that these residues may function in targeting DNMT3B to specific genomic regions [63] . We speculate that the E692G substitution in Dnmt3crahu mutants may similarly interfere with DNMT3C localization to target sequences . In mouse , Dnmt3c and Dnmt3l mutants have similar meiotic phenotypes , including male infertility due to spermatocyte arrest apparently occurring at epithelial stage IV ( [17 , 18 , 27 , 28 , 30] and this study ) . Both genes are required for methylating and silencing retrotransposons in the male germline , and , consistent with previous findings , comparison of RNA-seq data from Dnmt3crahu mutants with published Dnmt3l–/–data suggests considerable overlap in the transposon families that they target [27 , 29 , 30] . Dnmt3l is also required for methylation at imprinted loci [7 , 17 , 18 , 26] , but we did not observe embryonic defects characteristic of loss of methylation at maternally imprinted loci: Dnmt3c mutant females produced healthy litters of expected size , and their offspring survived to adulthood without discernible abnormalities , consistent with the findings of Barau et al . Among paternally imprinted loci , Barau et al . observed hypomethylation in Dnmt3c mutants only at the Rasgfr1 imprinting control region . They also showed that Dnmt3l mutants were globally hypomethylated , including at intragenic and intergenic regions . In comparison , hypomethylation in Dnmt3c mutants is primarily restricted to L1 , ERVK , and ERV1 retrotransposons ( [30] and this study ) . Taken together , these results suggest that Dnmt3c provides a more specialized contribution than Dnmt3l to the germline methylation landscape . The majority of differentially methylated regions in Dnmt3c mutants overlap with transposons ( [30] and this study ) , so it is likely that the SC and γH2AX defects in Dnmt3crahu mutants are indirect effects of transposon hypomethylation . Loss of transposon methylation in Dnmt3l mutants leads to abnormal levels of meiotic DSBs within transposon sequences , which in turn are thought to lead to deleterious non-allelic recombination events , culminating in meiotic arrest [29] . Similarly , transposon hypomethylation in Dnmt3crahu mutants may result in an abnormal meiotic DSB landscape . A non-exclusive alternative is that the Dnmt3crahu meiotic recombination defect may be linked to retrotransposon derepression , for example , via accumulation of DNA damage induced by a transposon-encoded endonuclease activity . This idea is supported by the presence of SPO11-independent DSBs in mice lacking Maelstrom , a piRNA pathway component required for transposon methylation and repression [64 , 65] . Maelstrom mutant mice that also lack SPO11 , whose catalytic activity is required for meiotic DSBs , show extensive immunofluorescence staining for DSB markers , consistent with damage that is mechanistically distinct from that induced during developmentally programmed meiotic recombination events . Yet another possibility is that transposon hypomethylation in Dnmt3crahu mutants perturbs the expression of neighboring meiotic genes . Comparative genomic analyses suggest that Dnmt3c arose in the Muroidea phylogenetic lineage by duplication of , and subsequent divergence from , Dnmt3b ( [30] and this study ) . It is conceivable that Dnmt3c neo-functionalized in response to selective pressure imposed by an increase in the retrotransposon load within the genome . An alternative hypothesis is that in organisms that lack Dnmt3c , its function is performed by a Dnmt3b isoform or by a yet-to-be discovered DNMT3 paralog . Given that the duplication is specific to muroid rodents and that Dnmt3c was previously mis-annotated as a pseudogene , its discovery exemplifies the power of forward genetic approaches . Moreover , the rapid evolution of meiotic proteins and the diversity of meiotic strategies adopted across different taxa necessitate organism-specific approaches . With advances in genomics facilitating the molecular characterization of phenotype-causing lesions identified in forward genetic screens , this approach will continue to be fruitful in furthering our understanding of gametogenesis . All experiments conformed to regulatory standards and were approved by the Memorial Sloan Kettering Cancer Center ( MSKCC ) Institutional Animal Care and Use Committee . Male mice of the C57BL/6J background were mutagenized by three weekly injections of 100 μg ENU/g of body weight , then outcrossed to FVB/NJ females . Wild-type mice of both inbred strains were purchased from The Jackson Laboratory . The mutagenesis and three-generation breeding scheme to generate homozygous mutant offspring were conducted as described elsewhere [31 , 34] ( Fig 1A ) . To minimize the chance of repeated recovery of the same ENU-induced mutation , no more than ten F1 founder males were screened from each mutagenized male . Each F1 founder male was used to generate ≥six G2 females and ~24 G3 males . For screening , testes from G3 males were dissected , snap-frozen in liquid nitrogen , and stored at -80° . Males were screened for meiotic defects at ≥15 dpp , by which age spermatocytes in the relevant stages of meiotic prophase I are abundant [58] . An upper age limit of 19 dpp was imposed to avoid the need for weaning . For a given line , spermatocyte squash preparation and immunostaining were carried out only after testes had been obtained from ~24 G3 males . This side-by-side analysis facilitated comparisons of phenotypes between mice . One testis per mouse was used to generate squash preparations of spermatocyte nuclei and immunostained with anti-SYCP3 and anti-γH2AX antibodies as described below . The second testis was reserved for later DNA extraction if needed . Based on the extent of axial element and SC formation , SYCP3-positive spermatocyte nuclei were classified as either early prophase-like ( equivalent to leptonema or early zygonema in wild type ) or late prophase-like ( late zygonema , pachynema , or diplonema ) . The γH2AX staining pattern was then evaluated . For each immunostained squash preparation , we aimed to evaluate ~20 early prophase-like cells and ~50 late prophase-like cells , if present . Priority for subsequent mapping and further analysis was given to lines that yielded at least two G3 males with similar spermatogenesis-defective phenotypes , derived from two or more G2 females . Genotyping of Dnmt3crahu animals was done by PCR amplification using rahu F and rahu R primers ( S4 Table ) , followed by digestion of the amplified product with Hpy188I ( NEB ) . The rahu mutation ( A to G ) creates a novel Hpy188I restriction site . Endonuclease-mediated alleles were generated at the MSKCC Mouse Genetics Core Facility using CRISPR/Cas9 . A guide RNA ( target sequence 5′-CATCTGTGAGGTCAATGATG ) was designed to target predicted exon 4 of Gm14490 ( NCBI Gene ID: 668932 and Ensembl Gene ID: ENSMUSG00000082079 ) and used for editing as described [66] . Using the T7 promoter in the pU6T7 plasmid , the gRNA was synthesized by in vitro transcription and polyadenylated , then 100 ng/μl of gRNA and 50 ng/μl of Cas9 mRNA were co-injected into the pronuclei of CBA × B6 F2 hybrid zygotes using conventional techniques [67] . Founder mice were tested for presence of mutated alleles by PCR amplification of exon 4 using Gm14490 F1 and R1 primers ( S4 Table ) , followed by T7 endonuclease I ( NEB ) digestion . Mis-targeting of CRISPR/Cas9 to Dnmt3b was considered unlikely as the gRNA has 10 ( 50% ) mismatches relative to the homologous region of Dnmt3b ( in exon 7 of Ensembl transcript ENSMUST00000109774 ( alignment of Gm14490 exon 4 and Dnmt3b exon 7 using EMBOSS Water tool , Smith-Waterman algorithm [49–51] ) . Also , this Dnmt3b region lacks the protospacer adjacent motif ( PAM ) . Nonetheless , we screened mice directly by PCR ( Dnmt3b F1 and R1 primers; S4 Table ) and T7 endonuclease I assay at the relevant region in Dnmt3b to rule out presence of induced mutations . Animals that were positive for Gm14490 mutation and negative for Dnmt3b mutation were selected for further analysis . We deduced the mutation spectrum of founder Dnmt3cem mice by PCR amplification of the targeted region from tail-tip DNA ( Gm14490 F1 and R2 primers; S4 Table ) followed by Sanger sequencing ( Seq1; S4 Table ) . Sequence traces were analyzed using TIDE [68] , CRISP-ID [69] , and Poly Peak Parser [70] . Dnmt3cem founder males mosaic for frame-shift mutations were bred to mutant Dnmt3crahu females to generate compound heterozygotes carrying both the Dnmt3crahu allele and a Dnmt3cem allele . Dnmt3cem-carrying founder mice were also interbred to generate homozygotes or compound heterozygotes carrying two distinct Dnmt3cem alleles . Genotyping of Dnmt3cem animals was done by PCR amplification of the targeted region followed by Sanger sequencing . PCR amplification was done with either Gm14490 F1 and R2 primers followed by sequencing with Seq1 , or with Gm14490 F2 and R2 primers followed by sequencing with Seq2 ( S4 Table ) . Sequence traces were analyzed to determine the mutation spectrum as described above . Dnmt3cem animals were also genotyped for the homologous region in Dnmt3b by PCR amplification with Dnmt3b F2 and R1 primers followed by sequencing with Seq1 ( S4 Table ) . All genome coordinates are from mouse genome assembly GRCm38/mm10 unless indicated otherwise . For genetic mapping , the screen breeding scheme ( Fig 1A ) was expanded: additional G2 females were generated and crossed to their F1 sire , and were identified as mutation carriers if they birthed G3 males displaying the Dnmt3crahu phenotype . Breeding of G2 carriers to the F1 founder was continued to accrue additional homozygous mutants . The Dnmt3crahu phenotype was coarsely mapped by microarray-based genome-wide SNP genotyping using the Illumina Mouse Medium Density Linkage Panel . To obtain genomic DNA , testes or tail biopsies were incubated in 200 μl of DirectPCR lysis reagent ( Viagen ) containing 1 μl of proteinase K ( >600 mAU/ml , Qiagen ) for 24 hr at 55° . DNA was subsequently RNase A-treated , phenol:chloroform-extracted , and ethanol-precipitated . Microarray analysis was performed at the Genetic Analysis Facility , The Centre for Applied Genomics , The Hospital for Sick Children , Toronto , ON , Canada . For bioinformatics analysis , 720 SNPs out of 1449 SNPs total on the linkage panel were selected based on the following criteria: autosomal location , allelic variation between B6 and FVB backgrounds , and heterozygosity in the F1 founder . For fine-mapping by manual genotyping of variants , genotyping was done by PCR amplification followed either by Sanger sequencing or by digestion with an appropriate restriction enzyme . For whole-exome sequencing , DNA from three phenotypically mutant G3 mice was prepared as for microarray analysis and pooled into a single sample . Because the mutant mice should share the phenotype-causing mutation ( s ) , we expected this pooling approach to boost the reliability of mutation detection . Whole-exome sequencing was performed at the MSKCC Integrated Genomics Operation . Exome capture was performed using SureSelectXT kit ( Agilent Technologies ) and SureSelect Mouse All Exon baits ( Agilent Technologies ) . An average of 100 million 75-bp paired reads were generated . Read adapters were trimmed using FASTX-Toolkit version 0 . 0 . 13 ( http://hannonlab . cshl . edu/fastx_toolkit/ ) and read pairs were recreated after trimming using a custom Python script . Reads were aligned to mouse genome assembly GRCm38/mm10 using Burrows Wheeler Aligner-MEM software version 0 . 7 . 5a [71] with default settings , and duplicate reads were removed using Picard tools version 1 . 104 ( https://broadinstitute . github . io/picard/ ) . A minimum mapping quality filter of 30 was applied using SAMtools version 0 . 1 . 19 [72] . Genome Analysis Toolkit version 2 . 8-1-g932cd3a ( Broad Institute; [73–75] ) was used to locally realign reads with RealignerTargetCreator and IndelRealigner , to recalibrate base quality scores using BaseRecalibrator , and to identify variants using UnifiedGenotyper with the following settings: mbq 17; dcov 500; stand_call_conf 30; stand_emit_conf 30 . Variants were annotated using ANNOVAR software [76] . To obtain a list of potential phenotype-causing lesions , variants were filtered further to only include those that 1 ) had a minimum sequencing depth of six reads , 2 ) were called as homozygous , and 3 ) were not known archived SNPs ( i . e . , they lacked a reference SNP ID number ) . The positions of variants within the 33 . 5-Mbp mapped region that we identified using this strategy are as follows: Chr2:129 , 515 , 815 in F830045P16Rik; Chr2:130 , 422 , 084 in Pced1a; Chr2:130 , 946 , 117 in Atrn; Chr2:137 , 046 , 739 in Slx4ip; Chr2:140 , 158 , 720 in Esf1; Chr2:143 , 947 , 738 in Rrbp1; and Chr2:153 , 727 , 342 in Gm14490 . ENCODE long-RNA sequencing data used are from release 3 . We acknowledge the ENCODE Consortium [77] and the ENCODE production laboratory of Thomas Gingeras ( Cold Spring Harbor Laboratory ) for generating the datasets . GEO accession numbers are as follows: Testis GSM900193 , Cortex GSM1000563 , Frontal lobe GSM1000562 , Cerebellum GSM1000567 , Ovary GSM900183 , Lung GSM900196 , Large intestine GSM900189 , Adrenal gland GSM900188 , Colon GSM900198 , Stomach GSM900185 , Duodenum GSM900187 , Small intestine GSM900186 , Heart GSM900199 , Kidney GSM900194 , Liver GSM900195 , Mammary gland GSM900184 , Spleen GSM900197 , Thymus GSM900192 . For histology , testes isolated from adult or juvenile mice were immersed overnight in 4% paraformaldehyde ( PFA ) at 4° with gentle agitation , followed by two 5-min washes in water at room temperature . Fixed testes were stored in 70% ethanol for up to 5 days . Testes were embedded in paraffin , then 5-μm-thick sections were cut and mounted on slides . The periodic acid Schiff ( PAS ) staining , immunohistochemical TUNEL assay , and immunofluorescent staining were performed by the MSKCC Molecular Cytology core facility . Slides were stained with PAS and counterstained with hematoxylin using the Autostainer XL ( Leica Microsystems ) automated stainer . The TUNEL assay was performed using the Discovery XT processor ( Ventana Medical Systems ) . Slides were manually deparaffinized in xylene , re-hydrated in a series of alcohol dilutions ( 100% , 95% and 70% ) and tap water , placed in Discovery XT , treated with proteinase K ( 20 μg/ml in 1× phosphate-buffered saline ( PBS ) ) for 12 min , and then incubated with terminal deoxynucleotidyl transferase ( Roche ) and biotin-dUTP ( Roche ) labeling mix for 1 hr . The detection was performed with DAB detection kit ( Ventana Medical Systems ) according to manufacturer’s instructions . Slides were counterstained with hematoxylin and mounted with coverslips with Permount ( Fisher Scientific ) . The immunofluorescent staining was performed using Discovery XT . Slides were deparaffinized with EZPrep buffer ( Ventana Medical Systems ) and antigen retrieval was performed with CC1 buffer ( Ventana Medical Systems ) . Slides were blocked for 30 min with Background Buster solution ( Innovex ) , followed by avidin-biotin blocking ( Ventana Medical Systems ) for 8 min . Slides were incubated with primary antibody for 5 hr , followed by 60 min incubation with biotinylated goat anti-rabbit ( 1:200 , Vector Labs ) . The detection was performed with Streptavidin-HRP D ( part of DABMap kit , Ventana Medical Systems ) , followed by incubation with Tyramide Alexa Fluor 488 ( Invitrogen ) prepared according to manufacturer’s instructions . After staining , slides were counterstained with 5 μg/ml 4′ , 6-diamidino-2-phenylindole ( DAPI ) ( Sigma ) for 10 min and mounted with coverslips with Mowiol . PAS-stained and TUNEL slides were digitized using Pannoramic Flash 250 ( 3DHistech ) with 2× lens . Images were produced and analyzed using the Pannoramic Viewer software . Immunofluorescence images were produced using a LSM 880 ( Zeiss ) with 40× lens . Spermatocyte squashes were prepared as described [78] , with few modifications . Isolated testes with the tunica albuginea removed were minced and suspended in 2% PFA in 1× PBS . Fixation was allowed for approximately 10 sec and spermatocytes were spotted onto slides . A coverslip was pressed down onto the spermatocytes to squash them , and the preparation was snap-frozen in liquid nitrogen . Slides were removed from liquid nitrogen and the coverslip was pried off , followed by three 5-min washes in 1× PBS with gentle agitation . Washed slides were rinsed in water , air dried and stored at -80° . For immunofluorescence , slides were thawed in 1× PBS for 5 min with gentle agitation and stained as described [79] . Slides were incubated with blocking buffer ( 1× PBS with 0 . 2% gelatin from cold-water fish skin , 0 . 05% TWEEN-20 , 0 . 2% BSA ) with gentle agitation for 30 min . Slides were stained with primary antibodies overnight at 4° , washed three times for 5 min in blocking buffer with gentle agitation , incubated with appropriate Alexa Fluor secondary antibodies ( 1:100; Invitrogen ) for 30 min at room temperature , then washed three times for 5 min in blocking buffer . All antibodies were diluted in blocking buffer . Slides were rinsed in water and cover slips were mounted using mounting medium containing DAPI ( Vectashield ) . Stained slides were stored at 4° for up to 5 days . Immunostained slides were imaged on a Marianas Workstation ( Intelligent Imaging Innovations; Zeiss Axio Observer inverted epifluorescent microscope with a complementary metal-oxide semiconductor camera ) using a 63× oil-immersion objective . Primary antibodies and dilutions used are as follows: mouse anti-SYCP3 ( SCP-3 ( D-1 ) , 1:100 , Santa Cruz , sc-74569 ) , rabbit anti-γH2AX ( p-Histone H2A . X ( ser 139 ) , 1:750 , Santa Cruz , sc-101696 ) , rabbit anti-ORF1p ( 1:1000 , gift from A . Bortvin ) , rabbit anti-IAP Gag ( 1:5000 , gift from B . R . Cullen ) . DNMT3C protein sequence was obtained by translating the 2 , 218-bp Ensembl transcript ENSMUST00000119996 ( release 87 ) cDNA sequence . An additional two nucleotides ( AG ) were added to the predicted transcript end to create a stop codon , resulting in a 2 , 220-bp transcript . DNMT3B protein sequence was obtained by translating the 4 , 320-bp Dnmt3b-001 Ensembl transcript ENSMUST00000109774 . 8 ( release 87 ) coding sequence; this translation has 100% sequence identity to M . musculus DNMT3B with accession number O88509 . ADD and PWWP domains were predicted using the NCBI Conserved Domain Database Search tool ( accession numbers cd11728 and cd05835 , respectively ) [80] . To determine the cytosine methyltransferase motif locations in DNMT3C and DNMT3B , the following sequences were aligned by Clustal Omega alignment method using MegAlign Pro software ( DNA STAR , Lasergene ) : M . musculus DNMT3C sequence determined as described above , M . musculus DNMT3B ( accession number O88509 ) , H . sapiens DNMT3B ( accession number Q9UBC3 ) , M . musculus DNMT3A ( accession number O88508 ) , H . sapiens DNMT3A ( accession number Q9Y6K1 ) , M . musculus DNMT3L ( accession number Q9CWR8 ) , H . sapiens DNMT3L ( accession number Q9UJW3 ) , M . musculus DNMT1 ( accession number P13864 ) , H . sapiens DNMT1 ( accession number P26358 ) , and H . parahaemolyticus HhaI ( accession number P05102 ) . Then the six highly conserved motifs ( I , IV , VI , VIII , IX , X ) were annotated as defined for HhaI [52] . The cytosine methyltransferase domain was annotated as the start of Motif I to the end of Motif X . The start and end locations for the domains and motifs are listed in S2 Table . Clustal Omega alignments from MegAlign Pro were used to produce a tree using BioNJ algorithm [81] , and the figure was prepared using FigTree version 1 . 4 . 3 ( http://tree . bio . ed . ac . uk/software/figtree ) . The model of DNMT3C was generated with Phyre2 protein structure prediction tool . The protein was modeled from template 2QRV ( carboxy-terminal domain of DNMT3A , 77% identity ) , with a confidence score of 100% . To extract genomic DNA , one half of a single testis or equivalent mg of tail tissue was incubated in 200 μl of DirectPCR lysis reagent ( Viagen ) containing 1 μl of proteinase K solution ( >600 mAU/ml , Qiagen ) for 24 hr at 55° . DNA was subsequently RNase A-treated , phenol:chloroform-extracted twice , and ethanol-precipitated . ~1 μg of DNA was digested for 4 hr at 37° with the methylation-sensitive HpaII restriction enzyme ( NEB ) or its methylation-insensitive isoschizomer MspI ( NEB ) . ~250 ng of digested DNA was electrophoresed on a 0 . 9% agarose gel and transferred as described [82] to an Amersham Hybond-XL membrane ( GE Healthcare ) . The L1 5′UTR probe has been described elsewhere [64] and corresponds to nucleotides 515–1628 of the L1 sequence ( GenBank accession number M13002 ) . The probe was random-priming labeled with [α32P]-dCTP using High Prime premixed solution ( Roche ) . Hybridizations were carried out overnight at 65° . For RNA expression analysis , six littermates ( three homozygous mutant and three heterozygous mice born from a cross between a Dnmt3crahu/+ male and a Dnmt3crahu/rahu female ) aged 14 dpp were analyzed . Procedures involving commercial kits were performed according to manufacturers’ instructions . Total RNA from one half of a single testis , after removing the tunica albuginea , was extracted using the RNeasy Plus Mini Kit containing a genomic DNA eliminator column ( Qiagen ) . The RNase-Free DNase Set ( Qiagen ) was used to perform an on-column DNase digestion during RNA purification , as well as an in-solution DNase digestion after RNA purification , followed by RNA cleanup . For RT-PCR , 1–3 μg of RNA was used with random hexamer primers to synthesize cDNA using the SuperScript III First-Strand Synthesis System ( Invitrogen ) . cDNA was diluted five-fold or more for PCRs . Quantitative PCR was carried out using LightCycler 480 SYBR Green I Master ( Roche ) for detecting products on a LightCycler 480 II Real-Time PCR instrument ( Roche ) . All reactions were done at 60° annealing temperature , with an extension time of 20 sec for L1 ORF2 , IAP 3′ LTR and IAP Gag primers , and 40 sec for L1 ORFs primers ( S4 Table ) . All reactions were done in triplicate and accompanied by control reactions using cDNA synthesized without reverse transcriptase ( -RT controls ) . The success of reactions was confirmed by analysis of amplification curves , melting curves , and electrophoresis of representative amplification products on an agarose gel . LightCycler 480 Software was used to quantify products by absolute quantification analysis using the second derivative maximum method . All crossing point ( Cp ) values were normalized to the mean Cp value obtained for triplicate Actb reactions , to get relative values . The mean of relative values for triplicate reactions was used to obtain mean relative values . The mean relative value represents the relative amount of product in any given mouse . To obtain fold change values , the mean relative value for each mouse was normalized to the mean of that obtained for three heterozygous mice . RNA sequencing ( RNA-seq ) was performed at the Integrated Genomics Operation of MSKCC . 1 μg of total RNA underwent ribosomal depletion and library preparation using the TruSeq Stranded Total RNA LT kit ( Illumina ) with 6 cycles of post-ligation PCR . Sequencing was performed using Illumina HiSeq 2500 ( 2×100-bp paired-end reads ) using the TruSeq SBS Kit v3 ( Illumina ) . On average , 61 million paired reads were generated per sample . Resulting RNAseq fastq files were aligned using STAR version 2 . 4 . 0f1 [83] to the mouse genome ( GRCm38/mm10 ) using Gencode M11 transcriptome assembly [84] for junction points . Coding genes and transposable elements were quantified using TEtoolkit [85] to annotate both uniquely and ambiguously mapped reads . Gencode annotation and Repbase database [86] for repetitive sequences and transposable elements were used during the quantification . Differentially expressed genes were calculated using DESeq2 [87] on the counts . For plotting , counts of transposable elements were normalized to all of the annotated reads including coding genes as counts per million ( CPM ) . Dnmt3l RNA-seq data are published ( GEO GSE57747 [29] ) . Expression values for 20-dpp-old Dnmt3l mutant and wild type , provided by the authors , were used to calculate retrotransposon expression fold change and matched to our RNAseq data by transposable element name . WGBS was performed with six animals from two litters ( two homozygous mutants and two wild-type mice from one litter , and one homozygous mutant and one wild-type mouse from a second litter ) aged 12 dpp . Genomic DNA was extracted from whole testis by incubating a single testis in 200 μl of DirectPCR lysis reagent ( Viagen ) containing 1 μl of proteinase K solution ( >600 mAU/ml , Qiagen ) for 24 hr at 55° . DNA was subsequently RNase A-treated , phenol:chloroform-extracted , and ethanol-precipitated . Whole-genome bisulfite libraries were prepared and sequenced at the New York Genome Center ( NYGC ) using a tagmentation-based protocol developed by NYGC and J . Greally . In brief , 100 ng of genomic DNA was fragmented using tagmentation by Tn5 transposase and purified by Silane bead cleanup ( Dynabeads MyOne Silane , Thermo Fisher Scientific ) . End filling was performed using dATP , dGTP , dTTP and methylated dCTP to protect added cytosines from bisulfite treatment . End-repaired DNA was purified by SPRI bead cleanup ( Beckman ) , and subjected to bisulfite treatment and cleanup using EZ DNA Methylation-Gold MagPrep kit ( Zymo Research ) . Illumina sequencing adapters ( standard i5 adapter and custom i7 adapter ) were added using PCR amplification . Finally , size selection of 300-bp to 800-bp fragments was performed using SPRI bead cleanup ( Beckman ) . Libraries were sequenced on the Illumina X10 platform ( v3 chemistry ) using 2×150 cycles with standard Illumina read 1 primer and custom read 2 and i7 index primers to generate >375 million paired reads . Control library generated from Kineococcus radiotolerans was spiked at 20% to enhance library complexity . Adapter sequence were first N-masked from raw FASTQ files using cutadapt v1 . 9 . 1 [88] ( http://cutadapt . readthedocs . io/en/stable/index . html ) . Short-read alignment was performed with bwa-meth [89] ( https://github . com/brentp/bwa-meth ) to mouse genome assembly GRCm38/mm10 . We modified bwa-meth’s default minimum longest match length for a read ( 0 . 44read-length ) to greater than 30 bp ( 0 . 2read-length ) . The resulting alignment files were marked for duplicates using Picard v2 . 4 . 1 ( http://broadinstitute . github . io/picard ) . Methylation levels were calculated using MethylDackel v0 . 1 . 13 ( https://github . com/dpryan79/MethylDackel ) at cytosines excluding quality control failed , supplemental , duplicate , and MAPQ less than 20 reads ( which excluded multi-mapped reads ) . Additionally , bases with quality less than 20 or within the first 11 bases sequenced on either read pair were also excluded . For all downstream analyses , CpGs with minimum coverage of 5 reads were used . methylKit [90] was used to call differentially methylated CpGs with >25% differential methylation ( up or down ) and SLIM ( Sliding Linear Model [91] ) -adjusted p <0 . 01 . Differentially methylated CpGs were annotated using Repbase [86] for LINE , LTR , and SINE elements that have minimum 95% coverage of the consensus sequence . Refseq [92] was used to annotate genic regions and UCSC genome browser ( http://genome . ucsc . edu ) was used to annotate CpG islands . Methylation profile plots were made using elements that have minimum 95% coverage of the consensus sequence and contain minimum ten CpGs across the body of the element . Dot plots were made using custom Perl code available at http://pagelab . wi . mit . edu/material-request . html ( David Page lab , Whitehead Institute , Cambridge , Massachusetts ) . A summary of sequences used is provided in S5 Table . All sequences were screened and masked for species-specific repetitive sequences prior to analysis using RepeatMasker software [93] with default settings . Two nucleotides ( AG ) were added to the annotated end of the Gm14490 transcript to create a stop codon . Beaver , guinea pig , and rabbit dot plots were vertically reflected to maintain a consistent gene order of Commd7 to Mapre1 from left to right . Gene models were created based on Ensembl ( version 87 ) exon annotations . To make the species cladogram , artificial sequences were aligned to resemble published data [60] ( UCSC Genome Browser; http://genome . ucsc . edu ) and the figure was prepared using FigTree version 1 . 4 . 3 . RNA-seq data and WGBS data are available at Gene Expression Omnibus ( GEO ) with the accession number: GSE100960 .	Half of human genomes are made up of transposons , which are mobile genetic elements that pose a constant threat to genome stability . As a defense strategy , transposons are methylated to prevent their expression and restrain their mobility . We have generated a mutant mouse , called ‘rahu’ , that fails to methylate transposons in germ cells , suffers an increase in transposon expression and is sterile . rahu mice carry a mutation in a new gene , Dnmt3c , which appeared during rodent evolution through gene duplication 45–55 million years ago and is an essential component of the germline defense system against transposons in male mice .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "spermatocytes", "retrotransposons", "germ", "cells", "genetic", "elements", "sequence", "motif", "analysis", "epigenetics", "dna", "mammalian", "genomics", "dna", "methylation", "chromatin", "sperm", "research", "and", "analysis", "methods", "sequence", "analysis", "chromosome", "biology", "bioinformatics", "animal", "cells", "gene", "expression", "chromatin", "modification", "dna", "modification", "animal", "genomics", "biochemistry", "dna", "sequence", "analysis", "cell", "biology", "nucleic", "acids", "phenotypes", "database", "and", "informatics", "methods", "genetics", "transposable", "elements", "biology", "and", "life", "sciences", "cellular", "types", "genomics", "mobile", "genetic", "elements" ]	2017	rahu is a mutant allele of Dnmt3c, encoding a DNA methyltransferase homolog required for meiosis and transposon repression in the mouse male germline
Ankyrin repeat proteins are elastic materials that unfold and refold sequentially , repeat by repeat , under force . Herein we use atomistic molecular dynamics to compare the mechanical properties of the 7-ankyrin-repeat oncoprotein Gankyrin in isolation and in complex with its binding partner S6-C . We show that the bound S6-C greatly increases the resistance of Gankyrin to mechanical stress . The effect is specific to those repeats of Gankyrin directly in contact with S6-C , and the mechanical ‘hot spots’ of the interaction map to the same repeats as the thermodynamic hot spots . A consequence of stepwise nature of unfolding and the localized nature of ligand binding is that it impacts on all aspects of the protein's mechanical behavior , including the order of repeat unfolding , the diversity of unfolding pathways accessed , the nature of partially unfolded intermediates , the forces required and the work transferred to the system to unfold the whole protein and its parts . Stepwise unfolding thus provides the means to buffer repeat proteins and their binding partners from mechanical stress in the cell . Our results illustrate how ligand binding can control the mechanical response of proteins . The data also point to a cellular mechano-switching mechanism whereby binding between two partner macromolecules is regulated by mechanical stress . Tandem repeat proteins , also known as solenoid proteins , are a special class of proteins comprising tandem arrays of small structural motifs ( 20–40 residues ) that pack in a roughly linear fashion to produce elongated , superhelical architectures , thereby presenting extended surfaces that act as scaffolds for molecular recognition . Examples include ankyrin , tetratricopeptide , HEAT and leucine rich repeats . Their structures are characterized by short-range interactions between residues either within a repeat or in adjacent repeats . As such they contrast with globular proteins , which are stabilized by many sequence-distant interactions that frequently result in complex topologies , and with polyproteins like titin , in which independently folded domains are covalently linked in tandem arrays but without significant non-covalent interactions between the individual domains . It is thought that the lack of sequence-distant contacts affords repeat proteins a high degree of flexibility and elasticity [1] , [2] , and atomic force microscopy ( AFM ) studies have identified certain unique properties that underlie this spring-like mechanical behavior [3]–[5] , [6] . However , the relationship between spring and scaffold functions of repeat proteins is not understood and requires a determination of the mechanics of these proteins upon ligand binding . Gankyrin is an oncoprotein that is overexpressed in hepatocellular carcinomas [7] . It belongs to the ankyrin repeat family of proteins , which are involved in numerous protein-protein interactions and which have been postulated to be the spring elements in mechanotransduction [8] . Each ankyrin repeat forms a β-turn followed by two antiparallel α-helices and a loop . Gankyrin binds the S6 ATPase subunit of the 19S regulatory particle of the 26S proteasome and it enhances the degradation of the tumour suppressors pRb and p53 . The interaction of Gankyrin with S6 C-terminal domain ( S6-C ) is typical of repeat protein molecular recognition in that the whole length of Gankyrin is used to create an extended surface for binding [9] ( Figure 1 ) . All but the C-terminal ankyrin repeat of Gankyrin ( repeat seven ) make contacts with S6-C . The interaction involves complementary charged residues on the two proteins that form several positively and negatively charged patches along the elongated interface . The latter comprises residues from the β-turns and the N-terminal helices of repeats 1–6 of Gankyrin . An important process in mechanotransduction is mechano-switching . For example , force has the potential to partially unfold proteins , shutting off or triggering biochemical reactions by disrupting binding motifs or exposing cryptic binding sites . Force modulates a protein's free energy surface; a small force does not necessarily abolish the native minimum but may cause the breakage of non-covalent bonds or , conversely , activate catch bonds that bind more tightly with force [10] . Likewise , binding can affect the mechanical response of proteins to external stress , as shown for Dihydrofolate Reductase [11] , Im9 [12] and protein G [13] , [14] . Here we use atomistic molecular dynamics simulations to compare the mechanics of Gankyrin in isolation and in complex with S6-C . The results show how the ligand affects every aspect of Gankyrin's behavior , including the order of repeat unfolding and the nature of partially unfolded intermediate states , the forces required and the work transferred to the system to unfold the whole protein and its parts . Gankyrin shows a unique behavior that may be prototypical of repeat proteins . In the absence of the S6-C ligand , there are many different mechanical unfolding pathways accessible to Gankyrin with a preference for a C- to N-terminus mechanism . In contrast , in the presence of S6-C the central repeats of Gankyrin , which make the most contact with the ligand , are stabilized and consequently the number of accessible unfolding pathways is reduced . Thus , the outermost repeats of Gankyrin unfold at low forces whereas the central repeats preserve their structure and binding interactions and only rupture at higher forces . The different force regimes under which Gank unfolds when bound to S6-C might have relevance in the mechanical control of protein-protein interactions in the cell and this behavior could be exploited in future to design new protein mechano-switches . Molecular dynamics simulations were performed using the CHARMM program [15] , [16] with the charmm19 united atom force field . The effect of the aqueous environment was modeled with the fast analytical continuum treatment of solvation ( FACTS ) algorithm [17] . An implicit solvent model was preferred to explicit ones because an atomistic representation of even a single hydration layer of an extended protein conformation would require an immense computational effort; moreover , implicit solvents relax instantaneously , which reduces artifacts when the protein is pulled fast . The previous application of this methodology to Gank [6] gave results that agreed with experimental measurements , which validates our choice of implicit solvation model . Initial atomistic models for Gank and Gank-S6-C complex were obtained from pdb structures 1qym [18] and 2dwz [9] , respectively . The residues of Gank in both systems were renumbered starting from the first resolved residue , which is Cys4 ( i . e . , Cys 4 in the pdb file is named Cys 1 in the present manuscript , and the following residues are renamed accordingly ) . The number of resolved residues in 1qym is 223 , whereas 226 residues of Gank were resolved in 2dwz . The three extra residues are at the C-terminus of Gank in the complex and they have been retained in the simulations . Langevin dynamics at 300 K with a friction coefficient of 1 ps−1 and an integration timestep of 2 fs was used . After 100 ps of equilibration at 300 K , the Gank constructs were subjected to steered molecular dynamics , whereby a spring with elastic constant set to 20 pN/Å was attached to the N-terminal main-chain nitrogen and to the C-terminal carbonyl carbon and its equilibrium distance was increased at constant velocity to simulate pulling . This resulted in an equal force being applied to both terminal atoms , along the vector joining them and in the direction of increasing distance . Pulling speeds of 0 . 05 Å/ps and 0 . 01 Å/ps were used . At pulling speed of 0 . 01 Å/ps , twelve independent simulations of isolated Gank and twelve of the Gank-S6-C complex were performed with different initial velocities starting from conformations sampled during equilibration . Each simulation was 80 ns long , which allowed for the complete stretching of Gank . Conformations were saved every 4 ps for analysis . At pulling speed of 0 . 05 Å/ps 24 and 16 simulations were performed for the Gank-S6-C complex and for the isolated Gank , respectively . Each simulation was 16 ns long and a snapshot was saved every 1 ps for analysis . The distribution of the force measured along the simulations at 0 . 01 Å/ps resembles a normal distribution with a long tail in the large force regions ( Figure 2A , top ) . The force distribution for the isolated Gank is narrower and peaks at lower values than does the distribution for the complex Gank-S6-C . The main part of the distribution resembles a Gaussian , possibly due to the large number of force-field parameters contributing to it . Thus , this part of the distribution was fitted to a normal distribution and the deviation between the actual distribution and the fit was determined ( Figure 2A bottom ) . Force peaks in the force-extension profiles ( Figure 3 ) reaching values where the deviation is the largest were retained as unambiguous peaks ( i . e . values larger than 200 pN and 250 pN for the Gank and the Gank-S6-C simulations , respectively ) . Lower force peaks are not easily separable from the noise and were therefore discarded . Similar quasi-normal distributions were obtained at pulling speeds of 0 . 05 Å/ps , however , the number of force peaks above the noise level was very small and the force peak analysis was therefore not carried out . The work transferred [19] , [20] by the external force to unfold the protein was measured by integrating along the simulations , where is the external force and is the change in the vector connecting the two atoms where the force is applied ( the N and C termini of the protein ) . This procedure is equivalent to measuring the area under the force-extension profiles ( Figure 3 ) . Because the data were saved at discrete time points , operatively we defined the work transferred by the external force up to time as: ( 1 ) where the sum runs along the saved time frames of the simulation , is the elapsed simulation time at frame n , is the average force measured in two successive frames and is the change in the end-to-end vector ( connecting the backbone C atom at the C-terminus of Gank to the N atom at the N terminus ) between two successive frames . Contributions from the individual ankyrin repeats are readily obtained by decomposing the vector into its repeat components ( see also Figure 2B ) : ( 2 ) where is the end-to-end vector of repeat r at frame i and is the corresponding change in two successive frames . The individual repeat contributions of the work are then: ( 3 ) A few examples of the behavior of in selected runs are shown in Figure S1 . We would like to emphasize that , whereas the assignment of force peaks to particular repeat unfolding events may sometimes be problematic because of the broadness of the peaks and noisy nature of the force extension profiles , the repeat component of the work is in contrast a well-defined and robust quantity that we can use instead to report on the forces needed to unfold the repeats via the integration step in Eq . 3 . The CαCβ-atom root-mean-square deviation ( RMSD ) of each ankyrin repeat of Gank from the native x-ray conformation was monitored along the simulations . Typically , the RMSD of a single repeat remains smaller than 5 Å for a certain time and then it displays an abrupt rise due to repeat unfolding , quickly reaching values of 20–25 Å when fully extended . Thus , in each simulation and for each repeat , the unfolding midpoint was set as the time point when the RMSD reached 10 Å . Different unfolding RMSD thresholds ( from 7 Å to 15 Å ) did not significantly change the average repeat unfolding time and the statistics of the repeat unfolding sequences at either unfolding speeds . On the 0 . 01 Å/ps pulling speed simulations a contact analysis has been performed . Pairs of residues defined a native contact when the distance between their Cα atoms was smaller than 8 Å in the native structure . Only those pairs more than three residues apart in the sequence were considered . The lifetime of a contact was defined as the last time the contact was observed along the simulations . The identification of non-native contacts is based on the concept of recurrence , defined in S . I . The definition of recurrent contacts translates broadly to those contacts that could be observed consistently in at least a continuous 0 . 4-ns-long stretch of trajectory . Non-native contacts are recurrent contacts that are not present in the native state . This technique is used to provide approximations to probability density distributions from a discrete sample of data points . Each point of the set is replaced by a Gaussian function with a fixed width . The total approximated probability density distribution for the data set is then the sum of all the Gaussian functions after normalization . The width of the Gaussian function is chosen on the basis of the expected error on the data values . The technique is used in place of a histogram of the data points , because it provides smooth distributions . The residues of Gank involved in binding to S6-C were defined as those that had at least one atom at a distance of less than 5 Å from any S6-C atom . The native binding residues were defined as those that were involved in binding in more than 80% of the equilibration trajectory . The residue binding lifetime was defined as the last time residues were observed to be involved in binding along the simulations , averaged over the different runs . Gank ( pdbid 1qym [18] ) and S6-C-bound Gank ( pdbid 2dwz [9] ) were first equilibrated and then subjected to constant speed steered molecular dynamics simulations using the program CHARMM [15] , [16] with the united atom force field ( param19 ) and FACTS [17] as model for implicit treatment of solvent ( see “Materials and Methods” for further details about the simulations ) . During the pulling of Gank to the fully extended state the force exerted on the protein ends was recorded . In AFM experiments on polyproteins the peaks in the force-extension profiles generally correspond to the unfolding of single monomeric units [21] , [22] and the increase in contour length between peaks corresponds to the unfolding length of a monomer; for ankyrin-repeat proteins [3] , [5] , [23] , the peaks in the force-extension profiles generally correspond to the unfolding of single repeats and the increase in contour length between peaks corresponds to the unfolding length of a single repeat ( 105–120 Å ) , although contour length extensions of 40–70 Å corresponding to the unfolding of half a repeat have also been observed in the cases of Gank [6] , AnkyrinR [3] , and NI6C [5] ) . In the present simulations , only the lower pulling speed produced force-extension profiles in which a sufficient number of clear force peaks could be identified above the noise level ( see “Materials and Methods” for further details about identification of force peaks ) . Representative force-extension profiles at the low pulling speed are shown in Figure 3A–D . The regions of the force-extension profile preceding the peaks were fitted separately to a worm-like chain ( WLC ) model with fixed persistence length of 3 . 8 Å , as in our earlier work [6] . The contour length and the breaking force for each peak were also measured . Representative snapshots from one of the simulation of the complexed Gank provide a pictorial description of the unfolding process ( Figure 4 ) . Analysis of the histogram of the contour length extensions shows that , for the uncomplexed Gank , the peaks are separated by distances that are multiples of L = 56 . 8 Å ( Figure 5 , top ) , corresponding to the unfolding length of half an ankyrin repeat [3] , [5] , [23]; a 2 L periodicity is also apparent . In the case of the Gank-S6-C complex , the 2 L periodicity is more evident and the peaks are most frequently separated by distances that are multiples of 2 L ( Figure 5 , bottom ) . Thus , the unfolding events in the uncomplexed Gank frequently involve half of a repeat ( usually one of the two helices comprising the ankyrin motif ) , whereas in the Gank-S6-C complex the unfolding events more frequently involve a whole repeat . Our simulations are therefore in agreement with our previous AFM experiments on uncomplexed Gank . As well as the differences in the step size of repeat unfolding described above , differences in the unfolding forces of uncomplexed versus complexed Gank , are also observed ( Figure 3E ) . In uncomplexed Gank the force peaks are typically between 100 pN and 250 pN , and the majority of the peaks are observed at total extensions of less than 400 Å . In the case of the Gank-S6-C complex , the force peaks lie in the range of 200 pN and 400 pN and with total extensions frequently larger than 300 Å . The scarcity of force peaks at total extensions below 300 Å in the complex and above 400 Å in the uncomplexed Gank is not due to a lack of unfolding events in these ranges but rather it is due to the forces peaks lying below the noise level . Furthermore , in uncomplexed Gank , the force is roughly independent of contour length . In contrast , in the case of the Gank-S6-C complex those few peaks observed at total extensions below 300 Å , corresponding to the unfolding of repeats 7 and 6 ( see below ) , have forces similar to those found in uncomplexed Gank , whereas peaks at total extensions above 300 Å show larger forces . Figure 3E shows that for both uncomplexed and complexed Gank the force peaks cluster at values of the extension that are separated approximately by the contour length of a single ankyrin repeat ( ∼110 Å ) , as discussed above ) , although the distribution is not sharp and intermediate values of the extension are also observed . The work transferred by the external force to unfold Gank was measured along the simulations ( see “Materials and Methods” for a detailed description of this calculation ) , together with the repeat components of this work ( Figure 6 ) . These calculations do not require the identification of the force peaks , and consequently they could be performed at both of the pulling speeds used here . At the pulling speed of 0 . 01 Å/ps , significantly more work was needed to completely stretch ( i . e . after 64 ns of simulations ) the complexed Gank ( 2100±100 kBT ) than the uncomplexed Gank ( 1400±100 kBT ) . Similarly at 0 . 05 Å/ps pulling speed , the work transferred for complete stretching was 3200±200 kBT and 2300±100 kBT for complexed and uncomplexed Gank , respectively . These values are of the same order of magnitude as values measured experimentally for other biological macromolecules [24] , [25] . The larger work transferred at the higher pulling speed is expected due to the larger fraction of energy irreversibly dissipated in the faster process . At both pulling speeds , all of the repeats , with the exception of repeat 6 , required significantly more work to unfold when Gank was bound to S6-C ( t-test at 1%-significance level , Figure 6 , and Table S1 and Table S2 ) . Repeat 1 showed the largest change upon complexation , with about twice the work required for unfolding in the complex compared with isolated Gank ( Table S1 and Table S2 ) . The unexpected increase in the work required to unfold repeat 7 , which does not directly contact S6-C , may be due to the larger length of this repeat ( by 3 residues ) in the complexed Gank construct compared with the uncomplexed Gank construct . The ankyrin repeats of Gank unfolded sequentially , one after the other , as judged by the RMSD from the native structure . Successive repeat unfolding events were evenly spaced in time for both uncomplexed and complexed Gank ( Figure 7 ) . The presence of S6-C bound to Gank introduced delays in unfolding , which were more pronounced for the later repeat unfolding events . The precise sequence of unfolding of the repeats carries important information about the distribution of mechanical stability across the repeat array . For isolated Gank the most common unfolding sequence was from the C-terminus to the N-terminus although several other unfolding pathways are also observed ( Table 1 top and Figure S2 top ) . Such behavior is expected for repeat proteins because the sequence similarity between the repeats means that they all have similar stabilities , which introduces degeneracy into the unfolding mechanism . The presence of bound S6-C had a dramatic impact on the sequence of unfolding ( Figure S2 bottom ) ; for example , repeat 1 ( r1 ) is rarely the last repeat to unfold in complexed Gank . Overall , the key feature of the different behavior is the narrower spectrum of unfolding positions available to repeats r1 , r2 , r5 and r6 in complexed Gank . A quantitative comparison of the unfolding sequences in the complexed versus uncomplexed Gank can be performed using contingency tables built by counting the number of simulation runs where repeat r1 has a certain position in the unfolding sequence in complexed versus uncomplexed Gank ( Table 1 ) . An exact Fisher's test on this contingency table shows that complexed Gank has a significantly different unfolding sequence compared with uncomplexed Gank ( confidence level 0 . 006% ) . Similarly significant differences ( maximum confidence level 2 . 3% ) are obtained using repeats r2 , r5 or r6 . We built these contingency tables by pooling all the simulations at both pulling speeds as the same kind of test revealed that the unfolding sequences were not significantly dependent on the pulling speed . The change in the sequence of repeat unfolding events and the time delays in the later unfolding events results in an increase in the average unfolding time of repeats r1–r4 in the complex compared with isolated Gank ( Tables S3 and S4 ) . In summary , the data show that S6-C binding significantly modulates the forced unfolding of Gank by specifically stabilizing the ligand-contacting repeats , which results in prolonged lifetimes of these repeats and less diversity of unfolding pathways accessible to the protein . A more detailed analysis of the force-induced unfolding process is obtained by monitoring the lifetime of native contacts in Gank ( Figure 8 ) . The lifetime of intra-repeat contacts ( Figure 8 , the α-helical contacts and the β-turn contacts within the black and the green ellipses , respectively ) confirms the directional character of the unfolding process , which proceeds roughly from the C- to the N-terminus , and the larger resistance of the complexed Gank compared with the isolated Gank . A prominent difference between the two sets of simulations is the early loss of inter-repeat contacts between β-turn regions ( within the blue ellipses in Figure 8 ) observed for repeat pairs r2–r3 , r3–r4 and r4–r5 in isolated Gank and not in the Gank-S6-C complex . In contrast , inter-repeat helix-helix contacts ( red ellipses ) tend to be lost just before one of the two repeats of the pair unfolds . The non-native contacts observed in the simulations are mostly local in nature , i . e . , occurring between residues close in sequence , and they are slightly more frequent in uncomplexed Gank than in the complex ( Figure S3 ) . A closer inspection of the data reveals that they involve the force-induced deformations of the repeat array , whereby one repeat rotates with respect to the adjacent repeat around the long axis of the protein . The simulations of the Gank-S6-C complex show how specific features of the interaction between the two proteins are affected by the mechanical stress exerted on Gank . The binding surface in the unperturbed complex comprises residues from repeats r1–r6 of Gank . Most of the contacting residues are in the β-turn and the N-terminal α-helix of each repeat . The loss of contacts between the two proteins during the simulations occurs gradually and follows the sequence of unfolding of the repeats themselves ( Figure 9A ) . The contacts that fail last are those involving residues in repeats r2 , r3 and r4 , more specifically the β-turn of repeat 4 , the β-turn and N-terminal α-helix of repeat 3 and the N-terminal α-helix of repeat r2 ( Figure 9B and C and Table 2 ) . The large variability in the lifetime of binding contacts in repeats r2 and r4 is related to the variability in the observed unfolding sequence in the different simulation runs ( Figure S2 ) , whereby one of the two repeats is most often the last to unfold . Within repeats 4–6 , the lifetimes of the binding contacts involving residues in the β-turn are slightly larger than those in the N-terminal α-helix ( Figure 9A ) . Binding contacts involving residues in repeat 3 tend to survive the unfolding of the repeat , specifically residues Ile76 and Trp71 in the N-terminal α-helix of repeat 3 . These interactions appear to have a non-specific hydrophobic character that does not require a perfectly structured repeat to be maintained . Ligand binding typically increases the thermodynamic stability of proteins , whereas mechanical stabilization upon ligand binding has been observed for some proteins [11] , [13] , [14] , [26]–[28] but not for others [12] , [29] , [30] . This distinction arises because mechanical unfolding as measured by force microscopy using a constant pulling speed is not an equilibrium process and therefore mechanical stability reflects the height of the energy barrier , i . e . the difference in stability between the native state and the mechanical transition state . Moreover , force-induced unfolding may often be very different from chemical-induced unfolding because of the fundamental difference between the two reaction coordinates and the different nature of the unfolded states in the two reactions . A ligand will enhance the mechanical stability of a protein only if it preferentially stabilizes the native state over the mechanical unfolding transition state , an approach that has been used to rationally tune mechanical stability [14] . The situation is more complicated for repeat proteins , however , because mechanical unfolding proceeds in a stepwise manner and so there is a series of transition states rather than a single one . We show here that ligand binding to a repeat protein has a discrete effect on its mechanical stability that is localized to specific repeats and consequently it dramatically alters the protein's mechanical behavior . The bound ligand increases the resistance of Gank to mechanical stress by delaying its complete unfolding and by requiring larger forces and consequently a larger energy to unfold it . We speculate that the ligand offers a supplemental support frame ( or “sink” ) to the protein that helps it to withstand the external force by distributing the stress over a larger molecular volume . The high pulling speeds used in simulations have been shown in some cases to mask the differences between two systems under comparison , due to the high noise level [31] . However , the differences we observed between complexed and uncomplexd Gank are insensitive to changes in the pulling speed and they are significant even at the high pulling speeds of the simulations; we predict therefore that significant differences should be observed experimentally in force-extension profiles measured by AFM . Indeed , our simulations predict up to two-fold difference in the forces observed for complexed versus uncomplexed Gank independent of pulling speed . And since the unfolding force peaks of isolated ankyrins are experimentally detectable [3] , [5] , [6] , so should be the detection of higher force peaks for upon complexation . We found previously , and confirmed in the present study , that when subjected to an external pulling force the predominant unfolding pathway of Gank is from the C-terminus to the N-terminus [6] , although a variety of other unfolding pathways are also accessible . This behavior is profoundly altered in the presence of the ligand . One consequence of ligand binding is that the N-terminal repeat ( r1 ) rarely unfolds last , which also explains the large change in the r1 component of transferred work: indeed , unfolding of a repeat with a folded neighbor ( r2 in this case ) requires extra energy due to the inter-repeat interactions . More generally , we observe that ligand binding greatly reduces the number of unfolding routes accessible to Gank , and this can be rationalized as follows: The sequence similarity between the repeats means that they all have similar stabilities and therefore unfolding can proceed through many different pathways . Indeed multiple pathways have also been observed in chemical-induced unfolding of repeat proteins ( e . g . [32] ) ; Ligand binding stabilizes repeats r2–r5 ( the mechanical hot spots ) , making the stability distribution across the repeat array more uneven and resulting in energetic bias and less diversity of unfolding pathways . The bound ligand induced a more regular and cooperative mechanical unfolding behavior at the repeat level , in the sense that isolated Gank showed a wide variety of stretching patterns involving both partially unfolded repeats , whereby only one of the two helices is unfolded , and the unfolding of more than one repeat at a time ( Figure 5 , top panel ) ; in contrast , the S6-C-bound Gank tended to unfold more “sharply” repeat by repeat ( Figure 5 , lower panel ) , similar to the consensus ankyrin repeat protein NI3C [6] , [23] . S6-C binding has the effect of reducing the importance of the cross talk between repeats by compensating for the loss of inter-repeat stability upon unfolding with the stabilization of each individual repeat involved in binding ( particularly the β-turns regions as shown in Figure 8 ) . The simulations helped to identify those residues of Gank whose contacts with S6-C are the most resilient to pulling ( Figure 9B , C and Table 2 ) . These residues are Arg38 , Trp43 , Ser46 , Ala47 , Asp67 , Asp68 , Ala69 , Trp71 , Ile76 , Ser79 , Asn100 , Gln101 , Asn102 , Cys104 ( according to our numbering scheme ) . We can compare these mechanical hot spots with the thermodynamic hot spots investigated by pull-down experiments [9] . Pull-downs were used to test mutations at positions Arg38 , Lys113 , Asp36 , Asp68 and Glu159 ( according to our numbering scheme ) , but only mutation of Arg38 was found to disrupt S6-C binding sufficiently to prevent pull down . The other residues did not appear to make as big a contribution to the interaction , which agrees with our finding that they do not have a large contact lifetime with S6C in our simulations . Contrary to the pull-down results , we found that Asp68 had a high contact lifetime . This may be a consequence of the fact that both neighboring residues Ala69 and Asp67 have large contact lifetimes , rather than its role in binding . The mechanical unfolding of repeat proteins is characterized by relatively small force peaks; therefore , the force at rupture is determined with a much larger relative error than in less compliant systems , making it more difficult to detect a trend in the height of the force peak as a function of the unfolding event as predicted by [33] . Nevertheless , the simulations do clearly show two distinct force regimes of complexed Gank: up to 300 Å total extension , corresponding most frequently to the unfolding of repeats 7 and 6 , the forces were similar to those observed in uncomplexed Gank . Above 300 Å total extension , corresponding most frequently to the unfolding of repeats 1–5 , the unfolding forces were much higher than those at lower extensions . This property of repeat proteins might have an important functional role in mechanotransduction , by preventing the complete dissociation of a protein complex below a certain force threshold . A mechano-switch can be thought of as a switch between two shapes of the free energy that rules thermodynamics , kinetics and the modulation of the latter with a force . Repeat proteins composed of units with unfolded contour lengths ranging between ∼100–150 Å , such as ankyrin repeat proteins and HEAT repeat proteins , could function as mechano-switches because their repeats are progressively unfolded only with increasing forces [33] , thereby allowing mechanical stress to modulate the protein structure and its binding properties . Moreover , the elastic behavior of the individual repeats [6] could contribute to the stability of repeat protein complexes by allowing refolding of individual repeats before complete dissociation of the protein from its ligand . The different force regimes under which Gank unfolds when bound to S6-C might have relevance in the mechanical control of protein-protein interactions in the cell and this behavior could be exploited in future to design new protein mechano-switches . The characteristic stepwise mechanical unfolding of repeat proteins is at the very heart of their compliance and an essential component of their elastic behavior [1] , [2] . These properties implicate repeat proteins in mechanical signal transduction . For example: repeat-protein stretching and contraction motions to regulate the activity of a bound enzyme [34]; ankyrin nanosprings to operate the gating of ion channels in hair cells [35]; HEAT-repeat carrier proteins that wrap around their cargoes to transport them between the cytoplasm and the nucleus [36] . The sensitivity of repeat-protein mechanics to ligand binding , as revealed in the present work , suggests the possibility of a further regulatory mechanism whereby elasticity is dictated by ligand binding .	Here we use molecular dynamics simulation to compare the mechanical properties of the 7-ankyrin-repeat oncoprotein Gankyrin in isolation and in complex with binding partner S6-C . Tandem repeat proteins like Gankyrin comprise tandem arrays of small structural motifs that pack linearly to produce elongated architectures . They are elastic , mechanically weak molecules and they unfold and refold repeat by repeat under force . We show that S6-C binding greatly increases the resistance of Gankyrin to mechanical stress . The enhanced mechanical stability is specific to those ankyrin repeats in contact with S6-C , and the localized nature of the effect results in fundamental changes in the way the protein responds to force . Thus , the forced unfolding of isolated Gankryin involves a diverse set of pathways with a preference for a C- to N-terminus unfolding mechanism whereas this diversity is reduced upon complex formation with the central repeats , which are those most tightly bound to the ligand , tending to unfold last . Our study shows how stepwise unfolding can buffer repeat proteins and their binding partners from mechanical stress in the cell . It also points to a mechano-switching mechanism whereby binding between two partner macromolecules is regulated by mechanical stress .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "biology", "computational", "biology", "biophysics" ]	2013	Effects of Ligand Binding on the Mechanical Properties of Ankyrin Repeat Protein Gankyrin
The chromosomal program of meiotic prophase , comprising events such as laying down of meiotic cohesins , synapsis between homologs , and homologous recombination , must be preceded and enabled by the regulated induction of meiotic prophase genes . This gene regulatory program is poorly understood , particularly in organisms with a segregated germline . We characterized the gene regulatory program of meiotic prophase as it occurs in the mouse fetal ovary . By profiling gene expression in the mouse fetal ovary in mutants with whole tissue and single-cell techniques , we identified 104 genes expressed specifically in pre-meiotic to pachytene germ cells . We characterized the regulation of these genes by 1 ) retinoic acid ( RA ) , which induces meiosis , 2 ) Dazl , which is required for germ cell competence to respond to RA , and 3 ) Stra8 , a downstream target of RA required for the chromosomal program of meiotic prophase . Initial induction of practically all identified meiotic prophase genes requires Dazl . In the presence of Dazl , RA induces at least two pathways: one Stra8-independent , and one Stra8-dependent . Genes vary in their induction by Stra8 , spanning fully Stra8-independent , partially Stra8-independent , and fully Stra8-dependent . Thus , Stra8 regulates the entirety of the chromosomal program but plays a more nuanced role in governing the gene expression program . We propose that Stra8-independent gene expression enables the stockpiling of selected meiotic structural proteins prior to the commencement of the chromosomal program . Unexpectedly , we discovered that Stra8 is required for prompt down-regulation of itself and Rec8 . Germ cells that have expressed and down-regulated Stra8 are refractory to further Stra8 expression . Negative feedback of Stra8 , and subsequent resistance to further Stra8 expression , may ensure a single , restricted pulse of Stra8 expression . Collectively , our findings reveal a gene regulatory logic by which germ cells prepare for the chromosomal program of meiotic prophase , and ensure that it is induced only once . In sexually reproducing organisms , germ cells undergo meiosis , a specialized cell division program that produces haploid gametes . The reductive segregation of chromosomes depends upon a complex series of chromosomal events that occur during meiotic prophase . This chromosomal program must be supported by expression of a large suite of genes . A genome-wide description of this gene expression program , and how it is regulated , has not been available for mammals or other animals with specialized sex cells , or germ cells . Indeed , the best existing model for such a gene regulatory program is that of budding yeast [1–4] . The chromosomal program of meiotic prophase , including events such as laying down of meiotic cohesins , synapsis between homologs , and homologous recombination , has been the subject of intense study [5–7] . Investigations of these processes in mammals have relied principally upon identifying mouse orthologs of proteins that have demonstrated meiotic functions in lower eukaryotes , and that are well conserved among sexually reproducing species [8 , 9] . However , not all proteins involved in the meiotic chromosomal processes are well conserved among eukaryotes , and identifying these exceptions has proven challenging [10] . Identification of a gene set specific to mammalian meiotic prophase would provide an orthogonal means of discovering poorly conserved or even novel proteins involved in the chromosomal program of mammalian meiotic prophase . Studies in a mammalian system are also required if we are to understand how the gene expression program of mammalian meiotic prophase is regulated; the regulation of meiotic initiation is poorly conserved . For instance , between mouse and budding yeast , the regulatory logic of meiotic initiation appears similar , but the molecular identities of the regulators are not conserved [11] . In both mouse ovarian and testicular germ cells , meiosis is initiated by retinoic acid ( RA ) [12–14] , a signaling molecule restricted to chordates [15] . RA induces Stra8 , a vertebrate-specific gene that encodes a putative helix-loop-helix-containing transcription factor [13 , 14 , 16–20] . Stra8 is required for all chromosomal events of meiotic prophase assayed , including cohesion , synapsis , and recombination , as well as the preceding meiotic DNA replication [12 , 20] . In mouse fetal ovarian germ cells , induction of Stra8 by RA requires the germ-cell-expressed competence factor Dazl [21] . Dazl , which encodes an RNA binding protein expressed in postmigratory XX and XY germ cells , is required for germ cells to gain competence to respond to developmental cues , including RA [22] . Thus far , the roles of RA , Stra8 , and Dazl have largely been assayed with respect to the chromosomal program of meiotic prophase; their potential roles in regulating the gene expression program have not been examined systematically . We sought to elucidate the gene regulatory program of meiotic prophase . We used the mouse fetal ovary as a model for two reasons . First , germ cells in the fetal ovary initiate and progress through meiotic prophase with greater synchrony than in the postnatal or adult testis . All germ cells in the fetal ovary initiate meiosis around embryonic day 13 . 5 , progress through meiotic prophase during subsequent fetal development , and arrest at diplotene of meiotic prophase before birth . Initiation and progression of meiotic prophase occurs in an anterior-to-posterior wave: expression of Stra8 and meiotic prophase genes begins in the anterior portion of the fetal ovary before extending towards the posterior [23 , 24] . We therefore took advantage of the relative synchrony of cell state over time and space to finely dissect initiation and progression of meiotic prophase . Second , the roles of Dazl , RA , and Stra8 in meiotic initiation are well defined in the fetal ovary . To determine how Dazl , RA , and Stra8 regulate the gene expression program , we profiled expression in wild-type and mutant animals . We used whole-gonad , genome-wide transcriptome profiling , to obtain a global description of gene expression , and followed up with targeted single-cell , single-transcript measurements to precisely quantify elements of regulatory control at the level of individual germ cells . We identified a set of 104 genes specific to meiotic prophase , as assayed in fetal ovarian germ cells . We characterized how Dazl , RA , and Stra8 regulate this gene expression program , thus complementing our previous understanding of how they regulate the chromosomal program . From these data , we discerned two elements of gene regulatory logic centered on Stra8 , a key inducer of the chromosomal program . Initial induction of genes requires Stra8-independent and Stra8-dependent pathways . After gene induction , Stra8 is required for subsequent down-regulation of its own expression . We propose that these elements of gene regulatory logic account for how germ cells prepare for and ensure a single induction of the chromosomal program of meiotic prophase . To identify and catalog the gene expression program of meiotic prophase as it occurs in the fetal ovary , we performed genome-wide transcriptome profiling by RNA-seq on whole fetal ovaries at embryonic days 12 . 5 , 14 . 5 , and 16 . 5 ( E12 . 5 , E14 . 5 , and E16 . 5 ) . At these time points , ovarian germ cells are in pre-meiotic , leptotene ( early meiotic prophase ) , and pachytene ( mid-late meiotic prophase ) stages , respectively [25 , 26] . At each time point , we assayed expression in gonads of wild-type and germ-cell-depleted ( KitW/KitWv ) mice [27] , in order to identify germ-cell-dependent genes . During this embryonic period , testicular germ cells do not initiate meiosis , but instead enter and remain in mitotic G0/G1 arrest [28] . We therefore also profiled expression from wild-type fetal testes , at the same time points , to identify ovary-enriched genes . We defined “meiotic prophase genes” as those meeting the criteria summarized in Fig 1A . Since genes involved in meiotic prophase should be expressed leading up to or during prophase , we required that genes be expressed in wild-type ovaries at one or more of the three time points ( E12 . 5 , E14 . 5 , or E16 . 5 ) at greater than 5 Fragments Per Kilobase of transcript per Million mapped reads ( FPKM ) . We found 10 , 366 genes to be expressed at this level in the fetal ovary ( Fig 1A ) . Additionally , since only germ cells might be expected to express genes required for meiotic prophase , we also required that gene expression be at least 2-fold higher in wild-type than in germ-cell-depleted ovaries . A total of 526 genes met both criteria ( Fig 1A , S1 Table ) . There was one conspicuous absence: Rec8 , a meiotic cohesin , was not germ-cell-enriched . We verified by single-molecule fluorescent in situ hybridization that Rec8 was indeed expressed in ovarian somatic cells as well as germ cells ( S1 Fig ) , and added it to the 526 genes . Using k-means clustering , we identified several distinct gene expression profiles ( Fig 1B ) . Of the 527 genes , about 46% , including Stra8 , were up-regulated between E12 . 5 and E14 . 5 , then down-regulated by E16 . 5 , suggestive of functions restricted to early meiotic prophase . About 22% of the 527 genes were up-regulated between E12 . 5 and E14 . 5 and remained elevated at E16 . 5; these genes include Sycp3 , which encodes a synaptonemal complex protein . About 11% of genes were not up-regulated until E16 . 5 , suggestive of functions later in meiotic prophase . The remaining 21% of the 527 genes were highly expressed at E12 . 5 and progressively down-regulated by E16 . 5 . These include many pluripotency markers , including Pou5f1 ( Oct4 ) , Nanog , and Sox2 , and reflect the down-regulation of a pluripotency program as germ cells enter meiosis [29–31] . To winnow this list of 527 ovarian germ cell genes down to those functioning in meiotic prophase , we required two additional criteria ( Fig 1A ) . Since testicular germ cells do not embark on meiosis until well after birth , we required that gene expression be at least 2-fold higher in fetal ovary than in fetal testis; 360 genes met this additional criterion . Since genes with meiotic functions should be up-regulated as germ cells enter and progress through meiosis , we also required that genes be at least 2-fold up-regulated between E12 . 5 and E14 . 5 or E16 . 5; 104 genes satisfied this as well as the earlier criteria . We refer to this final set of 104 genes as the gene expression program of meiotic prophase . Of these 104 genes , 54 have previously been implicated in meiotic prophase by independent , lower-throughput methods . For 33 of these 104 genes , loss-of-function mutants have been examined for fertility defects; defects in meiotic prophase or fertility were reported for 32 of the 33 genes tested in this manner ( S1 Text ) . For 21 of the remaining 71 genes , detailed descriptions of RNA or protein expression patterns are publicly available , and all are consistent with functions in meiotic prophase . Thus , among 104 genes implicated in meiotic prophase through our systematic whole-genome RNA-seq analysis , 53 ( of 54 genes tested ) are substantiated by prior studies . These findings suggest that many of the remaining 50 ( of 104 ) genes are novel and uncharacterized genes involved in meiotic prophase , representing a great opportunity for future study . Review of the published literature indicates that our RNA-seq analysis captured most meiotic prophase genes that are expressed specifically in meiotic germ cells . Of 21 genes for which mutant germ cells have been reported to arrest at leptotene , zygotene , or pachytene stages of meiotic prophase ( as cataloged by Handel and Schimenti , 2010 ) , 14 are represented in our list of 104 genes . The seven genes with meiotic prophase arrest phenotypes that we failed to identify by RNA-seq analysis are either ubiquitously expressed ( such as Cyclin-dependent kinase 2 , Cdk2 ) or are expressed in both ovarian and testicular germ cells ( such as Piwi-like RNA-mediated gene silencing 2 , Piwil2 ) . The design of our study would preclude our identifying genes that are expressed in both ovarian germ cells and somatic cells , or that are expressed at substantial levels in fetal testes . We next sought to determine how the meiotic prophase genes are activated . Stra8 was previously shown to be required for meiotic initiation , as primarily assayed by the meiotic chromosomal program . Dazl and RA are germ-cell-intrinsic and -extrinsic factors , respectively , required for induction of Stra8 and initiation of the chromosomal program . The roles of these factors in regulating the program of gene expression are largely unknown . We first determined whether Dazl and Stra8 regulate meiotic prophase genes by examining gene expression by RNA-seq in whole E14 . 5 Dazl-deficient ( Dazl-/- ) and Stra8-deficient ( Stra8 -/- ) ovaries as compared to corresponding homozygous wild-type controls ( Fig 2 , S2 Table ) . Dazl is required for germ cells to acquire competence to respond to RA . Dazl-expressing germ cells respond to RA by expressing Stra8 and initiating meiosis . Dazl is also more broadly required for the processes of gametogenesis , which encompass meiosis , the sex-specific cellular differentiation events of oogenesis and spermatogenesis , and the down-regulation of pluripotency markers [21 , 22] . Given Dazl’s broad role in competence for gametogenesis , we predicted that Dazl would be required for induction of the meiotic prophase gene expression program . Indeed , we found that expression of practically all meiotic prophase genes ( 100 of 104 ) was significantly diminished if not eliminated in Dazl-deficient ovaries ( Fig 2 , S2 Table ) . The remaining four genes were expressed at < 5 FPKM in both wild-type and Dazl-deficient ovaries . Stra8 is required for the chromosomal program of meiotic prophase , including loading of meiotic cohesins , such as REC8 , and assembly of the synaptonemal complex proteins , including SYCP3 . However , although the REC8 and SYCP3 proteins do not localize to chromosomal axes in Stra8-deficient germ cells , the proteins are nevertheless produced [20] . In fact , Rec8 expression can be induced in testicular germ cells by RA in the absence of Stra8 function [32] . These results suggest that while Stra8 might regulate the entirety of the chromosomal program , it might have a more limited role in governing the gene expression program . We aimed to clarify the extent to which Stra8 regulates the meiotic gene expression program . We found that expression of the 100 Dazl-dependent genes ranged across a wide spectrum of Stra8-dependency . For slightly over half of the 100 genes , including Dmc1 , which is required to repair meiotic double-strand breaks , expression appeared to be fully dependent on Stra8 . Expression of these genes was reduced in Stra8-deficient ovaries to levels as low as in the Dazl-deficient ovary ( Fig 2 , S2 Table ) . Expression of the remaining genes appeared to be partially dependent on , or in a few cases , largely independent of Stra8 . Some genes , such as Sycp3 , were expressed at lower levels in Stra8-deficient ovaries than in wild-type ovaries , but still at higher levels than in Dazl-deficient ovaries . At the Stra8-independent extreme of the spectrum is Rec8 , whose levels were not only undiminished in Stra8-deficient ovaries , but in fact were modestly increased . Thus , RNA-seq analyses of whole Dazl-deficient and Stra8-deficient ovaries suggest a model of gene induction whereby Dazl is required for induction of the meiotic prophase gene expression program via at least two pathways: a Stra8-independent pathway , and a Stra8-dependent pathway . Whole-gonad RNA-seq analysis provides genome-wide breadth in characterizing the gene expression program of meiotic prophase . However , because this method averages across a population that includes a diversity of both germ cells and somatic cells , our observations may not accurately reflect events in individual germ cells . Specifically , we wondered whether our observation that some genes appeared partially Stra8-independent by RNA-seq actually reflected a partial reduction in gene expression in all Stra8-deficient cells . If so , this would indicate that Stra8-dependent and Stra8-independent pathways act additively in individual germ cells . Alternatively , our RNA-seq observation could be explained by a subset of Stra8-deficient germ cells retaining wild-type levels of gene expression , with other Stra8-deficient germ cells having greatly reduced levels of gene expression . Distinguishing between these two scenarios required measurement of gene expression with single-cell resolution . We used single molecule fluorescence in situ hybridization ( smFISH ) to quantify gene expression in single cells in situ . smFISH involves multiple short fluorescently-labeled oligonucleotide probes that collectively bind along the same target transcript to detect and localize each target mRNA molecule as a punctate signal [33] ( Fig 3A ) . These punctate signals can be quantified to determine the number of transcripts per cell volume ( transcript density ) ( Fig 3B ) . We selected 13 genes , spanning a spectrum of Stra8-dependencies as measured by RNAseq in E14 . 5 ovaries ( Fig 2 ) , for examination by smFISH in germ cells of E14 . 5 wild-type , Stra8-deficient , and Dazl-deficient ovaries . Selected genes include the meiotic-specific cohesins Rec8 , Smc1b , and Stag3; the synaptonemal complex proteins Sycp1 , Sycp2 , and Sycp3; Dmc1 and Msh5 , which are involved in double-strand break repair; and Hormad1 , which promotes homolog alignment and synaptonemal complex formation . We also included Mei1 and M1ap , which exhibit defects in meiotic prophase when mutated , and Gm1564 and Ugt8a , which are presently uncharacterized . Our smFISH studies of individual germ cells confirmed that all 13 genes were Dazl-dependent , and that they ranged across a spectrum of Stra8-dependence . For conceptual simplification and ease of discussion , we will describe this spectrum as comprising three classes: Class 1 genes—fully Stra8-independent , Class 2 genes—partially Stra8-independent , and Class 3 genes—fully Stra8-dependent ( Fig 3C–3E ) . Rec8 fell into Class 1 , fully independent of Stra8 expression ( Fig 3C and 3F ) . Expression of Rec8 in Stra8-deficient germ cells , as a population , was in fact slightly higher than in wild type , an observation we later explored . Dmc1 , Msh5 , Hormad1 , Mei1 , and M1ap fell into Class 3 , fully dependent on Stra8 ( Fig 3E and 3H , S2A Fig ) . Their expression in Stra8-deficient germ cells was reduced ( compared to wild-type germ cells ) to the same degree as in Dazl-deficient germ cells . Sycp3 , Sycp2 , Sycp1 , Stag3 , Smc1b , Gm1564 , and Ugt8a fell into Class 2 , partially independent of Stra8 expression ( Fig 3D and 3G , S2A Fig ) . Their expression in Stra8-deficient germ cells was , as a population , significantly lower than in wild type , but significantly higher than in Dazl-deficient germ cells . We always observed a unimodal distribution of gene expression , which is consistent with gene expression being reduced in each germ cell . The direction and relative magnitude of gene expression changes as measured by smFISH and RNA-seq are consistent for all 13 genes ( S2B Fig ) . What is the role of RA in regulating the meiotic prophase genes ? It was previously shown that RA induces Stra8 expression in fetal ovarian germ cells [13 , 14] . Therefore , Stra8-dependent induction of Class 2 and 3 genes would depend , indirectly , on RA . Does RA also regulate the Stra8-independent induction of Class 1 and 2 genes ? We previously showed that RA induces Rec8 in the absence of Stra8 [32] , and we now demonstrate , quantitatively , the full independence of Rec8 expression from Stra8 . By extension , we hypothesized that RA is responsible for Stra8-independent induction of not just Rec8 , a Class 1 gene , but also of the Class 2 genes . An ideal test of this hypothesis would be to eliminate RA in vivo in the fetal ovary . This was not technically feasible , so we instead sought evidence of RA regulation by analyzing gene expression in hundreds of individual germ cells , and using endogenous variation in expression of an RA-induced gene in these hundreds of germ cells as a read out of cell response to RA . If variation in expression of an RA-induced gene reflects the individual cell’s response to RA , then expression of two RA-induced genes across hundreds of individual germ cells should be positively correlated . To test this , we examined variation in expression of the two known independently RA-induced genes , Stra8 and Rec8 . We found that Rec8 transcript density is indeed positively correlated with Stra8 transcript density in germ cells of E14 . 5 fetal ovaries ( Fig 3I ) . We then employed variation in the level of Rec8 expression as a quantifiable read out of RA response , so as to determine if Stra8-independent expression of genes is due to RA . If so , then expression of the gene , in the absence of Stra8 , should be correlated with that of Rec8 . We quantified expression of each Class 2 gene alongside Rec8 , in hundreds of individual Stra8-deficient germ cells at E14 . 5 . Expression of Sycp3 , Sycp2 , Sycp1 , Stag3 , Gm1564 , and Ugt8a is positively correlated with Rec8 expression ( Fig 3J , S3 Fig , S3 Table ) . As expected , for Class 3 genes , which are fully Stra8-dependent , residual expression in the absence of Stra8 did not correlate with Rec8 expression ( Fig 3K , S3 Fig , S3 Table ) . These results are consistent with the Stra8-independent pathway being regulated by RA , either directly or indirectly . Our model of gene regulation inferred from E14 . 5 fetal ovaries led us to predict two consequences for gene expression over time . First , we reasoned that , for Class 2 genes , both Stra8-independent and Stra8-dependent pathways might be required to attain maximal levels of gene expression . If so , expression of Class 2 genes in Stra8-deficient germ cells would not reach peak wild-type levels , even after an extended period of time . Second , we considered the possibility that RA-dependent , Stra8-independent induction of Class 1 and 2 genes might function to induce genes that are required early in meiotic prophase , in anticipation of Stra8 . If so , Class 1 and 2 genes might be induced in parallel with Stra8 , and before Class 3 genes . To determine the temporal dynamics of gene expression with fine resolution , we took advantage of previous observations that fetal ovarian germ cells initiate and progress through meiotic prophase in an anterior-to-posterior wave [23 , 24] . Stra8 , Dmc1 , and Sycp3 expression have been observed to be induced first in germ cells in the anterior portion of the fetal ovary , and only later in the posterior . Therefore , measuring gene expression as a function of anterior-posterior position in addition to time should provide finer resolution of events than would time alone . To formally test the hypothesis that the anterior-posterior axis is a proxy for time , we compared expression changes of 527 germ-cell-enriched genes over time ( between E12 . 5 and E13 . 5 , anterior third of ovaries only ) , and over space ( between posterior and anterior thirds of E13 . 5 ovaries ) . We found that gene expression changes over both time and space were indeed highly correlated ( S4 Fig , S4 Table ) , validating our spatiotemporal approach . We assayed gene expression over a spatiotemporal axis , using Stra8 expression in wild-type germ cells as a reference , as follows . We measured the transcript density of Stra8 in individual germ cells at E11 . 5 , E12 . 5 , E13 . 5 , E14 . 0 , E14 . 5 , E15 . 0 , E15 . 5 , and E16 . 5 . For each time point , we calculated the average transcript density along the longitudinal axis , from the posterior pole ( germ cells at least advanced state ) to the anterior pole ( germ cells at most advanced state ) ( Fig 4A ) . We then joined these average expression traces from consecutive time points to create a continuous trace of average transcript density along a spatiotemporal axis ( Fig 4B ) . Using this approach , we quantified the Stra8 pulse of expression in the wild-type ovary , which was previously observed semi-quantitatively , by whole-mount in situ hybridization [23] . We applied this spatiotemporal analysis to characterize expression dynamics of the subset of 13 meiotic prophase genes in wild-type , Stra8-deficient , and Dazl-deficient germ cells . First , we asked if Class 2 genes indeed required both Stra8-independent and Stra8-dependent pathways to attain maximal levels of gene expression . We found that , in Stra8-deficient germ cells , Class 2 genes failed to reach expression levels seen in wild type even when given an additional one to two days after expression peaks in wild type . For example , expression of Sycp3 in Stra8-deficient germ cells had begun to decline by E16 . 5 , without having reached the peak levels of expression achieved ( at E15 . 5 ) in wild-type germ cells ( Fig 4B , S5 Fig ) . Thus , the Stra8-independent pathway is crucial to ensure full expression of Class 2 genes . Second , we asked if Stra8-independent induction of Class 1 and 2 genes might enable early gene expression . We found that induction of four Class 1 and 2 genes–Rec8 , Stag3 , Smc1b , and Gm1564 –indeed occurred early , and contemporaneous with Stra8 . Half-maximal expression of these genes preceded or coincided with half-maximal expression of Stra8 ( Fig 4D , S5 Fig ) . In contrast , all five of the Class 3 genes tested reached half-maximal expression after Stra8 had done so . Thus , the Stra8-independent pathway is able to induce expression of some Class 1 and 2 genes in parallel with Stra8 . Spatiotemporal analysis of Rec8 expression in Stra8-deficient germ cells unexpectedly revealed that in the absence of Stra8 , germ cells expressed Rec8 at modestly higher levels ( Figs 3F and 4E ) , and Rec8 expression persisted for at least a day longer than in wild type . Therefore , a Stra8-dependent process is required for the subsequent down-regulation of Rec8 . By our measurements , Rec8 and Stra8 are induced and subsequently down-regulated with nearly identical dynamics . Therefore , we wondered if Stra8 down-regulation also requires Stra8 function . To measure Stra8 promoter activity in the Stra8-deficient germ cells , we measured expression of a lacZ reporter knocked into the endogenous Stra8 locus [20] . We compared this to lacZ expression in Stra8 heterozygotes , where one functional copy of Stra8 is present . As with expression of Rec8 , expression of lacZ in the homozygous Stra8 knockout persisted for at least a day longer than in the heterozygous Stra8 mouse ( Fig 4F , S6 Fig ) . Thus , we have identified down-regulation of Stra8 and Rec8 as a novel Stra8-dependent event ( Fig 4G ) . The observation that Stra8 expression is rapidly down-regulated after its initial induction led us to wonder if , in addition to down-regulating Stra8 , germ cells become refractory to subsequent induction of Stra8 by RA . If so , wild-type germ cells that have expressed Stra8 once should not be able to express Stra8 again , even if they were provided with a second ( exogenous ) dose of RA . To test this prediction , we administered exogenous RA to pregnant mice at E15 . 5 , by which time most germ cells have down-regulated Stra8 . We then measured expression of Stra8 a day later , at E16 . 5 ( Fig 5 ) . Stra8 expression was not increased in ovarian germ cells of fetuses that received RA , compared to fetuses that did not receive RA . As a control , we tested if RA was able to induce Stra8 in E15 . 5 testicular germ cells . Since testicular germ cells do not ordinarily express Stra8 until after birth , we expected that they would be able to induce Stra8 expression if exposed to RA before birth . We found that at E16 . 5 , Stra8 expression was induced about 20-fold in testicular germ cells of fetuses that received RA , compared to fetuses that did not receive RA . Thus , fetal ovarian germ cells that have down-regulated Stra8 expression are refractory to re-expressing Stra8 when exposed to RA 24 hours after initial down-regulation of Stra8 . Earlier , we measured expression of Rec8 and Sycp3 in hundreds of individual E14 . 5 germ cells that lacked Stra8 , and found that Rec8 and Sycp3 levels were positively correlated , implying their co-regulation by a Stra8-independent pathway ( Fig 3J ) . We were initially surprised to find , upon performing the same analysis in E14 . 5 wild-type germ cells ( Fig 6A ) , that Rec8 and Sycp3 levels were not correlated in any simple fashion when Stra8 was present . We reasoned that these differences between wild-type and Stra8-deficient germ cells should be due to the two Stra8-dependent processes described earlier—partial induction of Sycp3 , and down-regulation of Rec8 . Our earlier results suggested that induction of Sycp3 occurs first , followed by down-regulation of Rec8 . To corroborate these understandings , we set out to finely dissect the time course of Rec8 and Sycp3 expression by separately analyzing germ cells in posterior-to-anterior quarters of the E14 . 5 wild-type ovary ( Fig 6B ) . In the posteriormost quarter of the ovary , where germ cells are least differentiated , Rec8 and Sycp3 levels were positively correlated ( Fig 6B ) , suggesting that the dominant process at this very early stage is induction of both Rec8 and Sycp3 . In subsequent stages , Sycp3 continues to be induced , but Rec8 is down-regulated . For example , in the middle quarters of the ovary , where germ cells are more differentiated , many germ cells had increased Sycp3 levels , but Rec8 expression was decreased , particularly in those cells with high expression of Sycp3 . In the anteriormost quarter of the ovary , where germ cell differentiation is most advanced , all germ cells had high Sycp3 levels , but most had very low levels of Rec8 expression ( Fig 6B ) . In summary , fetal ovarian germ cells progressed from a state of low Sycp3/low Rec8 expression , to a state of high Sycp3/high Rec8 expression , and finally to a state of high Sycp3/low Rec8 expression ( Fig 6C ) , reflecting initial induction of both Sycp3 and Rec8 , and subsequent down-regulation of Rec8 . As a consequence of down-regulation of Rec8 ( but not Sycp3 ) in the most advanced germ cells , expression of Rec8 and expression of Sycp3 were no longer correlated . Thus , the conclusions arising from our earlier analyses were corroborated by positionally informed , single-cell correlation analysis in the E14 . 5 wild-type ovary . At the onset of meiotic prophase , meiotic prophase genes are induced by Dazl , RA , and Stra8 , organized in two branching pathways ( Fig 7A ) . Dazl is required for the induction of nearly all genes expressed specifically during meiotic prophase . While the results highlight Dazl’s crucial role in the gene expression program for meiotic prophase , the mechanism by which it enables gene expression remains unclear . Downstream of Dazl , gene induction occurs via a Stra8-independent pathway as well as a Stra8-dependent pathway; these pathways function both separately and additively . Expression of some genes requires only the Stra8-independent pathway ( Class 1 ) , while other genes require both Stra8-independent and Stra8-dependent pathways ( Class 2 ) , and yet other genes are fully Stra8-dependent ( Class 3 ) . Both the Stra8-dependent and Stra8-independent pathways are regulated by RA . These genetic insights lead us to two speculative hypotheses regarding molecular mechanisms of meiotic prophase gene regulation by STRA8 and RA receptors ( RARs ) : ( 1 ) The Stra8-dependent pathway is mediated directly by the STRA8 protein , a putative basic helix-loop-helix transcription factor , and ( 2 ) the Stra8-independent pathway is mediated directly by RARs . Transcriptome data from both whole gonads ( this study ) and sorted germ cells [34] support this possibility: fetal ovarian germ cells initiating meiosis express all three RARs ( RAR alpha , beta , and gamma ) and their heterodimeric partners , the retinoid X receptors ( RXR alpha , beta , and gamma ) . Potential redundancies among the RARs and RXRs complicate genetic interrogation of the roles of the RARs . The possible roles of STRA8 and RARs in directly regulating gene expression can be tested by chromatin-immunoprecipitation-sequencing ( ChIP-seq ) of RARs and STRA8 in germ cells that are initiating meiosis . We predict that Class 1 genes will be bound by RARs but not STRA8 , Class 3 genes will be bound by STRA8 but not RARs , and Class 2 genes will be bound by both RARs and STRA8 . A ChIP-Seq study in embryonic stem cells identified RAR binding of both the Stra8 and Rec8 promoter regions [35] . Of course , it is also possible that the Stra8-independent and Stra8-dependent pathways are mediated indirectly , by germ-cell-expressed factors that have not yet been implicated in meiotic initiation . The branching regulatory model described here is reminiscent of a motif termed a feed forward loop ( FFL ) , which has been shown to generate a temporal order of gene activation [36] . An FFL comprises an upstream regulator , in this case RA , which regulates a downstream regulator , in this case Stra8 . Both the upstream regulator , RA , and the downstream regulator , Stra8 , regulate multiple downstream targets , in this case the meiotic prophase genes . Genes respond to input from either the upstream or downstream regulator , or both . Modulating the activation strengths of upstream versus downstream regulators can generate a temporal order of gene activation: genes with greater input from the upstream regulator are activated earlier , and genes with greater input from the downstream regulator are activated later . Consistent with such an outcome , we observe that a subset of Class 1 and 2 genes , which are fully or partially induced by the Stra8-independent pathway , are expressed earlier than Class 3 genes and with timing of induction close to that of Stra8 induction . We propose that Class 1 and 2 genes may be induced earlier so as to prepare cells for the meiotic chromosomal events triggered by Stra8 . Indeed , we observe that Class 1 and 2 genes include almost all known meiotic cohesins and synaptonemal complex proteins , which structurally associate with meiotic chromosomes and may therefore be required early , and in sufficient quantities to satisfy the stoichiometric requirements of the chromosomal program . Early expression of cohesin and synaptonemal complex proteins , prior to initiation of the chromosomal program , may be a common feature of both sexes and across species . In mouse testicular germ cells , the synaptonemal complex protein genes Sycp1 , Sycp2 , and Sycp3 are expressed as early as in mitotic spermatogonia [37] . In the C . elegans gonad and the D . melanogaster ovary , Rec8 and synaptonemal complex proteins respectively are also expressed during the amplifying mitotic divisions preceding meiosis [38–40] . Induction of Class 2 genes by a combination of Stra8-independent and Stra8-dependent pathways may also contribute to fine-tuning expression levels of meiotic prophase genes . Precise regulation of gene dosage has been shown to be important for meiotic chromosomal processes . In the mouse , heterozygous loss of function for either one of the cohesins Smc1b and Rec8 perturbs formation of the synaptonemal complex and affects synapsis and recombination between homologs [41] . Therefore , although the Stra8-independent pathway is sufficient for partial gene expression , the two pathways in combination may serve to optimize levels of gene expression and chromosomal function of the Class 2 genes . Subsequent to the initial induction of Stra8 and Rec8 , their expression declines rapidly . We discovered that this down-regulation depends on Stra8 . It remains to be determined whether this occurs directly , via Stra8 activity as a putative transcriptional regulator , or indirectly , as a consequence of progression of cell state . In either case , we propose that Stra8-dependent down-regulation of Stra8 and Rec8 may serve to limit gene expression to their appropriate window of function . In particular , Stra8-dependent down-regulation of itself represents a negative feedback loop that prevents prolonged induction of the chromosomal program of meiotic prophase . In addition , we found that ovarian germ cells that have down-regulated Stra8 are refractory to re-expressing Stra8 even in the presence of exogenous RA , which may prevent re-initiation of the chromosomal program . In yeast , an analogous negative feedback loop is postulated to restrict supernumerary rounds of DNA replication and nuclear division . IME1 , a transcription factor that initiates the yeast meiotic transcriptional program , induces IME2 , which restricts expression of IME1 and destabilizes IME1 protein [3 , 42] . Absence of IME2 results in prolonged IME1 expression and additional rounds of DNA synthesis and nuclear division [43] . Based on similarities between ovarian and testicular germ cells , it is likely that the gene regulatory program as inferred from fetal ovarian germ cells is shared , at least in part , between the sexes . In both sexes , RA induction of Stra8 has been shown to be required for initiation of the chromosomal program of meiotic prophase [12 , 14 , 20] . RA and Stra8 could therefore also regulate gene expression in the male . In testicular germ cells entering meiosis , Stra8 is also rapidly induced , at pre-leptotene , and then rapidly down-regulated , by leptotene [17] , suggesting the possibility that there is also negative feedback on Stra8 expression . However , several aspects of regulation in the male remain unclear . For instance , in the male , RA-STRA8 signaling regulates not only meiotic initiation but also spermatogonial differentiation [44] . In fetal ovarian germ cells , competence to respond to RA and initiate meiosis requires Dazl [21]; in testicular germ cells , Dazl’s role in meiotic competence remains unknown . Thus , studies of adult testicular germ cells deficient for Dazl or Stra8 will be required to determine if Dazl and Stra8 govern the gene regulatory program of meiotic prophase in the adult testis in a manner similar to that reported here for the fetal ovary . Our findings have practical implications for in vitro derivation of germ cells and gametes . First , our results provide a blueprint to guide efforts in recapitulating the gene regulatory program of meiotic prophase in vitro . Second , our findings substantiate previous criticisms against taking expression of meiotic genes as sufficient evidence of meiosis [45 , 46] . By explicitly interrogating the regulation of the gene expression program and the chromosomal program by Dazl and Stra8 , we showed that the two programs are regulated distinctly . Specifically , the chromosomal program of meiotic prophase requires Stra8 function , but a subset of the gene expression program is induced independently of Stra8 . Our findings thus highlight gene expression as a preparatory phase for the chromosomal program , and underscore the insufficiency of meiotic gene expression as an assay for meiotic progression . Rather , both the gene regulatory program and the chromosomal program are essential for successful meiosis . All experiments involving mice were performed in accordance with the guidelines of the Massachusetts Institute of Technology ( MIT ) Division of Comparative Medicine , which is overseen by MIT's Institutional Animal Care and Use Committee ( IACUC ) . The animal care program at MIT/Whitehead Institute is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care , International ( AAALAC ) , meeting or exceeding the standards of AAALAC as detailed in the Guide for the Care and Use of Laboratory Animals . The MIT IACUC approved this research ( no . 0714-074-17 ) . Germ-cell-depleted ( KitW/KitWv ) and homozygous wild-type control ( Kit+/Kit+ ) were generated by crossing C57BL/6J-KitWv/Kit+ ( The Jackson Laboratory ) males to WBB6F1/J-KitW/Kit+ ( The Jackson Laboratory ) females [47] . KitW and KitWv alleles were genotyped as previously described [48 , 49] . Stra8-deficient ( Stra8-/- ) , Dazl-deficient ( Dazl-/- ) and homozygous wild-type control embryos were generated by heterozygote matings of Dazltm1Hjc [50] and Stra8tm1Dcp [20] mice respectively . Dazltm1Hjc , Stra8tm1Dcp , and wild-type mice used are of C57BL/6 background . Dazltm1Hjc and Stra8tm1Dcp alleles were genotyped as previously described [20 , 50] . 500 mg/kg of body weight all-trans RA ( Sigma-Aldrich , St Louis , MO ) , dissolved at 30 mg/ml of corn oil , was administered to pregnant mice via gavage . Timed matings were set up by housing female mice with male mice overnight . Noon of the day when a vaginal plug was evident was considered E0 . 5 . For RNA-seq analysis , embryonic gonads were dissected away from mesonephroi . For RNA-seq from germ-cell-depleted ( KitW/KitWv ) and homozygous wild-type control ( Kit+/Kit+ ) gonads , Stra8-deficient ( Stra8-/- ) , Dazl-deficient ( Dazl-/- ) and homozygous wild-type control gonads , whole gonads were processed for sequencing . For RNA-seq from E12 . 5 and E13 . 5 anterior and posterior portions of the ovary , ovaries were dissected into thirds , and the anterior and posterior thirds were processed for sequencing . For smFISH , embryonic gonads were dissected with mesonephroi intact to provide anterior-posterior orientation . For embryos E13 . 5 and older , the sex of tissues was determined by scoring the presence or absence of testicular cords . For E11 . 5 and E12 . 5 embryos , sex was determined by PCR as previously described [23] . For all RNA-seq experiments , total RNA ( ~1 ug ) was extracted from embryonic gonads using Trizol ( Invitrogen ) according to the manufacturer’s protocol , and hemoglobin transcripts were selectively removed from total RNA using GLOBINclear ( Invitrogen , Carlsbad , CA ) according to the manufacturer’s protocol . For KitW/KitWv and Kit+/Kit+ embryonic gonads and E12 . 5 and E13 . 5 embryonic ovary thirds , libraries were prepared using the Illumina mRNA-Seq Sample Preparation Kit ( Illumina , San Diego , CA ) according to the manufacturer’s protocol . Libraries were sequenced on the Illumina Genome Analyzer II platform to obtain 36-base-pair single reads . For E14 . 5 Dazl-deficient , Stra8-deficient , and wild-type control ovaries , libraries were prepared using the Illumina TruSeq RNA Sample Preparation Kit . Libraries were multiplexed and sequenced on the Illumina HiSeq 2000 platform to obtain 40-base-pair single reads . RNA-seq data for KitW/KitWv and Kit+/Kit+ , E12 . 5 and E13 . 5 embryonic ovary thirds , and E14 . 5 Dazl-deficient , Stra8-deficient , and wild-type control ovaries have been deposited in NCBI GEO under accession number GSE70361 and NCBI SRA under accession numbers SRP058992 , SRP059594 , SRP059601 , and SRP059599 . RNA-seq on gonads from KitW/KitWv and Kit+/Kit+ embryos was performed on two biological replicates for each condition . RNA-seq on anterior and posterior thirds of ovaries from E12 . 5 and E13 . 5 wild-type embryos was performed on two and three replicate pools respectively , where each pool consisted of ovary thirds from eight embryos . RNA-seq on E14 . 5 Dazl-deficient and Stra8-deficient ovaries was performed on three biological replicates each , with paired homozygous wild-type controls . Reads were aligned to the mouse genome ( mm9 ) using TopHat [51] , allowing only unique alignments ( option—g1 ) . We counted reads mapping to the Refseq annotated gene set using htseq-count [52] . Fold-changes and FDR-corrected p values , q , for differentially expressed genes were calculated using edgeR , using tagwise-dispersions and normalizing for library complexity [53] . FPKMs were calculated using Cufflinks [54] . K-means clustering was performed using Cluster 3 . 0 on log transformed and mean centered FPKMs , using the Pearson correlation as the similarity metric [55] , and visualized using Treeview [56] . Non-coding genes were excluded from analyses . Probe design , synthesis , and coupling were as previously described [33 , 57] . Probes sequences are provided in S2 Text . Gonads were fixed in 4% paraformaldehyde ( PFA ) /PBS for 2 hours at 4°C , incubated overnight in 30% sucrose/4% PFA/PBS at 4°C , then embedded in O . C . T . compound ( Sakura Finetek , Torrance , CA ) . Frozen blocks were sectioned at 8 μm thickness , fixed in 4% formaldehyde at room temperature for 15 minutes , rinsed in PBS , and dehydrated overnight in 70% ethanol at 4°C . The hybridization procedure was performed as previously described [33 , 57] . FITC-coupled anti-SSEA-1 antibody ( BD 560127 ) ( BD Biosciences , Franklin Lakes , NJ ) was added to the hybridization step at 1:30 to identify germ cells . In all experiments , germ cells were identified by either smFISH for Dazl or Oct4 , in combination with SSEA1 immunostaining , and/or DAPI nuclear staining . Counting of individual mRNA particles , image stitching , and data analysis was performed using custom Matlab software as previously described [33 , 57] . To depict distributions of transcript densities for each group , we pooled biological replicates in one violin plot . Comparison of groups was performed by comparing means of at least two biological replicates from at least two litters using the two-sample t-test . To depict correlations between pairs of genes in individual cells , we show one representative biological replicate , but calculate the Spearman correlation coefficient for each biological replicate . To depict average transcript densities over space and time , we pooled biological replicates . At each time point , we determined average transcript densities from the posterior to anterior of the ovary for 100 windows of size 0 . 2 of the total length of the ovary . The average transcript density traces of consecutive time points were joined together from posterior to anterior . Using average transcript density traces of Stra8 as a guide , we overlapped some time points by shifting along the x-axis in order to maximize overlap between the average expression traces for Stra8 . We determined shifts using Stra8 expression , and applied the same shifts to spatiotemporal plots for all other genes .	The formation of haploid gametes from diploid germ cells requires a specialized reductive cell division known as meiosis . This reductive division is enabled by chromosomal events that occur during meiotic prophase , including synapsis and crossing-over of homologous chromosomes . These chromosomal events involve meiosis-specific genes that must be expressed before they act during meiosis . Using gene expression profiling , we identified a set of mammalian meiosis-specific genes . To understand how expression of these genes is controlled , we examined their expression in the absence of known regulators of the chromosomal events: 1 ) retinoic acid ( RA ) , which induces meiosis , 2 ) Dazl , which is required for germ cell competence to respond to RA , and 3 ) Stra8 , which is induced by RA and is required for the chromosomal events of meiotic prophase . We uncover two key features of gene regulation . First , while the genes require RA and Dazl to be expressed , they vary in their dependence on Stra8 , thus creating a regulatory hierarchy . Genes induced independently of Stra8 , and thus early in this hierarchy , may encode proteins that are stockpiled in anticipation of the chromosomal events . Second , Stra8 induces its own down-regulation , and may thus prevent repeated induction of meiosis in a single germ cell .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[]	2015	A Gene Regulatory Program for Meiotic Prophase in the Fetal Ovary
The rK39 recombinant protein is derived from a specific antigen produced by the Leishmania donovani complex , and has been used in the last two decades for the serodiagnosis of visceral leishmaniasis . We present here a systematic review and meta-analysis of studies evaluating serologic assays to diagnose visceral leishmaniasis to determine the accuracy of rK39 antigen in comparison to the use of other antigen preparations . A systematic review with meta-analysis of the literature was performed to compare the rK39 strip-test and ELISA formats against serological tests using promastigote antigens derived from whole or soluble parasites for Direct Aglutination Test ( DAT ) , Indirect Immunofluorescence test ( IFAT ) and ELISA with a promastigote antigen preparation ( p-ELISA ) . Gold standard diagnosis was defined by the demonstration of amastigotes on hematological specimens . A database search was performed on Medline , Lilacs , Scopus , Isi Web of Science , and Cochrane Library . Quality of data was assessed using the QUADAS questionnaire . A search of the electronic databases found 352 papers of which only 14 fulfilled the selection criteria . Three evaluated the rK39 ELISA , while 13 evaluated the rK39 immunochromatographic strip test . The summarized sensitivity for the rK39-ELISA was 92% followed by IFAT 88% and p-ELISA 87% . The summarized specificity for the three diagnostic tests was 81% , 90% , and 77% . Studies comparing the rK39 strip test with DAT found a similar sensitivity of 94% , although the DAT had a slightly higher specificity . The rK39 strip test was more sensitive and specific than the IFAT and p-ELISA . We did not detect any difference in the sensitivity and specificity between strips produced by different manufacturers . The rK39 protein used either in a strip test or in an ELISA , and the DAT are the best choices for implementation of rapid , easy and efficient test for serodiagnosis of VL . Visceral Leishmaniasis ( VL ) is a neglected tropical disease for which a simple and quick diagnostic test is available , but not yet widely implemented in rural areas [1] . Although several different serodiagnostic test formats exist , many of them have not been validated in prospective field studies [2] . In addition , only a limited number of well conducted trials comparing the different types of tests have been published [2] . This lack of data has limited the routine use of these tests in many settings . The optimal test for serologic diagnosis is one that is easy to use , cheap to make , and has both a high sensitivity and specificity . The use of crude antigens for the serodiagnosis of tropical diseases is often limited by the difficulty in producing large quantities of the antigen , and therefore can be difficult to make in a standardized manner . However , once a useful diagnostic antigen has been identified , modern molecular biology techniques allow the antigen to be manufactured as a recombinant protein in a standardized manner with an advantage over crude lysate undefined antigen used in DAT . In the recent past , numerous recombinant antigens have become available for use in serodiagnostic testing of leishmaniasis [3] . Molecular diagnostic assays for leishmaniasis using polymerase chain reaction ( PCR ) targeting multi-copy genes , ( e . g . , rRNA , kinetoplastDNA ( kDNA ) minicircles ) have been developed . Sensitivity and specificity of these tests depend upon the region targeted , with a recent study finding a peripheral blood PCR assay having an overall sensitivity of 98 . 5% [4] . However , nucleic acid testing is currently difficult to perform in many clinical labs in the developing world . The currently available serodiagnostic tests for VL have been based on four major formats: Direct agglutination ( DAT ) , indirect immunofluorescence ( IFAT ) , ELISA and immunocromatography [3] , [5] , [6] . The DAT and IFAT classically utilize whole promastigotes to screen for antibodies , while the p-ELISA uses a crude lysate of promastigotes . Immunochromatographic tests , and a newer ELISA have been developed using the recombinant protein rK39 , which is a kinesin-like gene found in Leishmania chagasi . Recently , the WHO Special Program for Research and Training in Tropical Disease ( TDR ) evaluated five different immunochromatographic tests utilizing either rK39 or rKE16 , a recombinant protein developed from the kinesin gene of a Leishmania donovoni isolate [7] . Testing was performed in East Africa , Brazil and on the Indian subcontinent , and sensitivities ranged from 36 . 8–100% and specificities from 90 . 8–100% . No test was the clear winner across all regions and conditions . In addition , since comparisons between different test types are limited , it remains unclear even which type of serologic test is most optimal for use in the diagnosis of VL . Meta-analyses are excellent tools to evaluate multiple studies with similar goals but varying methods [8] . We performed a meta-analysis of studies evaluating serodiagnostic tests for VL to determine the accuracy of rK39 antigen based serodiagnostic tests in comparison to other available serodiagnostic tests for the diagnosis of VL . This is a systematic review of the literature using the Cochrane recommendations to compare the sensitivity and specificity of serodiagnostic tests using rK39 antigen in the ELISA and strip test formats against crude lysate Leishmania antigen used in an ELISA or whole promastigote parasite as antigen used in DAT and IFAT . Three hundred fifty-two papers have been published to date evaluating the rK39 antigen as diagnostic tool for serodiagnosis of VL . Within these reports , significant discrepancies in sensitivity and specificity are found , likely due to a multitude of reasons including: variability in testing methods , testing performed in different geographic regions , different sources for the tests , and lack of homogeneity of the studied population . Applying the selection criteria described below , we included 14 studies in this meta-analysis . Inclusion criteria were: a ) have a full description of accuracy of the diagnostic test , b ) that were performed on human specimens , c ) include a rK39 antigen based test , d ) direct demonstration of Leishmania parasites as the confirmatory diagnostic method for VL . Exclusion criteria were: a ) insufficient primary and/or secondary data information of ( lack information about sensitivity and specificity ) , b ) lack of inclusion of a control group , c ) inclusion of co-infections with HIV , d ) inclusion of subjects that were receiving or had received VL treatment , e ) full text article unavailable , f ) written in a language other than English , Spanish and Portuguese . Figure 1 describes the process utilized for selection of studies for the meta-analysis . Data were obtained by a search of the published literature cited in Medline , Lilacs , Scopus , Isis Web of Science , Cochrane Library accessed on June 1 , 2011 . Our search strategy utilized the following key words: rK39 antigen , Visceral Leishmaniasis , parasitology , accuracy test and human . We also utilized the suggested synonymous terms for our search . i ) Initial screening of manuscripts was performed by selecting those with titles related to our subject of study ii ) Each manuscript selected by title then had its abstract reviewed for inclusion or exclusion based upon the variables reported in the results . iii ) Selected studies the had the full text reviewed to collect existing primary variables of true positive and false positive , true negative and false negative data presented in each study that allowed re-calculation of sensitivities and specificities , likelihood , predictive values and confidence intervals . Also , careful reviews of the references in each selected manuscript were performed to rescue those not selected by search strategy . This literature review was performed by two independent reviewers . If the reviewers were in consensus , the manuscript was accepted for the meta-analysis , if there was disagreement , a third reviewer was consulted . The QUADAS ( Quality in Diagnostic Accuracy Studies ) protocol was used to evaluate the quality of each study [9] . Receiver operating characteristic ( ROC ) plot were constructed using the R Statistical Software package . A total of 5 , 548 individuals were included into the meta-analysis from the 14 included studies: 2 , 097 cases of VL and 3 , 451 controls . The distribution of studies according to region was as follows: Brazil ( n = 5 ) ; Indian ( n = 4 ) ; Nepal ( n = 2 ) ; Tunis , Italy , and Kuwait ( n = 1 each ) . The manuscripts were published during the period between 1998 to 2011 . Thirteen studies reported results of strip tests: Sundar 1998 [10] , Sundar 2007 [11] , Saghrouni 2009 [12] , Romero 2009 [13] , Mathur 2005 [14] , Mandal 2008 [15] , Iqbal 2002 [16] , de Assis 2008 [17] , de Assis 2011 [18] , Chappuis 2003 [19] , Carvalho 2003 [20] , Brandonisio 2002 [21] , and Boelaert 2004 [22] . Three reported results with the rk39-ELISA: de Assis 2008 [17] , Pedras 2008 [23] and Romero 2009 [13] . IFAT results were reported from seven studies Boelaert 2004 [22] , Brandonisio 2002 [21] , de Assis 2008 [17] , Iqbal 2002 [16] , Pedras 2008 [23] , Romero 2009 [13] and Saghrouni 2009 [12]; ELISA with crude promastigote lysate results were reported from six studies: Carvalho 2003 [20] , de Assis 2008 [17] , Mathur 2005 [14] , Pedras 2008 [23] , Romero 2009 [13] , and Mandal 2008 [15]; and finally DAT results were reported from six studies: Boelaert 2004 [22] , Chappuis 2003 [19] , Pedras 2008 [23] , Sundar 2007 [11] , de Assis 2011 [18] , and Mandal 2008 [15] . A total of 12 studies fulfilled equal or greater than 10 questions from the 14 QUADAS questions . The studies of Brandonisio 2002 [21] and Mandal 2008 [15] fulfilled 8 and 7 questions respectively . The summarized sensitivity of the 13 studies evaluating the rK39 antigen strip test was 92% [91 . 49–92 . 92] and the summarized specificity was 95% [94 . 30–95 . 48] . The likelihood ratio of a positive test ( LR+ ) was found to be 18 . 042 [18 . 03–18 . 06] and the likelihood ratio of a negative test ( LR− ) was found to be 0 . 082 [−0 . 13–0 . 29] . These 13 studies were stratified according to which technique were compared . i ) Five studies compared the strip test vs p-ELISA: Carvalho 2003 [20] , de Assis 2008 [17] , Mathur 2005 [14] , Romero 2009 [13] and Mandal 2008 [15]; ii ) five compared the strip test vs DAT: Boelaert 2004 [22] , Chappuis 2003 [19] , Sundar 2007 [11] , de Assis 2011 [18] , Mandal 2008 [15]; and iii ) six compared the strip test vs IFAT: Boelaert 2004 [22] , Brandonisio 2002 [21] , de Assis 2008 [17] , Iqbal 2002 [16] , Romero 2009 [13] and Saghrouni 2009 [12] . Figure 2 demonstrates the ROC plot for comparison of sensitivity and specificity of the strip test and other test formats . The ROC plot for the strip test vs p-ELISA demonstrating a sensitivity , specificity and a positive likelihood ratio ( LR+ ) of 93% , 96% , and 21 . 16 for the strip test and 88% , 79% , and 4 . 14 for p-ELISA , respectively . The study by Romero et al had sensitivity and specificity values that were quite different from the other studies ( sensitivity of 50% and specificity of 77% ) . The exclusion of the Romero study for ROC analysis increased the specificity of strip test 99% and ( LR+ ) to 68 . 77 , but did not alter the sensitivity , and increased the sensitivity to 91% , specificity 88% , ( LR+ ) to 7 . 81 for the p-ELISA . ROC plots for the strip test vs . DAT demonstrated that the sensitivities and specificities were very close for both tests ( 94 . 23% , 89 . 97% and LR+ 9 . 39 for DAT and 94 . 48% and 88 . 75% and LR+ 8 . 40 for the strip test ) . The ROC plot comparison of the strip test vs . IFAT revealed a better performance of strip test than IFAT with a sensitivity , specificity and LR+ of 87% , 95% and 16 . 89 for the strip test vs 84% , 92% and 10 . 01 for the IFAT . Only two studies compared the strip test vs . the rK39-ELISA: de Assis 2011 [18] and Romero 2009 [13] . The summarized sensitivity of the tests were similar ( 93% for both ) and the specificity of each test was 95% and 80% respectively . This resulted in a higher LR+ for the strip test than the rK39-ELISA ( 19 . 24 vs . 4 . 61 respectively ) . Only three studies evaluated the rK39 antigen used in the ELISA format: de Assis 2011 [18] , Pedras 2008 [23] and Romero 2009 [13] . The pooled characteristics of the studies were: sensitivity of 92% , specificity of 80% and LR+ of 4 . 73 . ROC plot comparison of rK39-ELISA , p-ELISA and IFAT is presented in figure 3 . Summarized sensitivity results for the three tests are 92% , 87% and 88% and summarized specificity results are 81% , 77% and 90% respectively . However , evaluation of the Romero study [13] noted discrepancies between the raw data and in the sensitivity and confidence intervals ( Table 1 ) . The ROC plot constructed excluding Romero's study is presented in figure 2B . From the thirteen studies that utilized the strip test format , nine used the Kalazar Detect™ Test for VL manufactured by Inbios ( Seattle , USA ) : Sundar 2007 [11] , Saghrouni 2009 [12] , Romero 2009 [13] , Mathur 2005 [14] , Mandal 2008 [15] , Chappuis 2003 [19] , Carvalho 2003 [20] , Brandonisio 2002 [21] , and Boelaert 2004 [22] . Two studies used DiaMed-IT's Leish: de Assis 2008 [17] , de Assis 2011 [18]; one study used the Leishmania rapid test strip-test from Intersep: Iqbal 2002 [16]; and one study the recombinant K39 strip test from Arista Biologicals: Sundar 1998 [10] . The best performances in a single study for a strip test were demonstrated with the recombinant K39 strip test from Arista Biologicals , followed by the strip test from Inbios . However , there was considerable variability in the results from the studies conducted with the Inbios strip test . Some of the variability in sensitivity and specificity has been shown to be associated with the geographic region where the test is being performed [7] . The best receiver operating characteristic ( ROC ) plot for the Inbios strip test was demonstrated by the results of Brandonisio 2002 [21] who evaluated the test in Italy ( Figure 4 ) . Three studies evaluated a p-ELISA using L . chagasi promastigotes as the source of antigen: Carvalho 2003 [20] de Assis 2008 [17] , and Pedras , 2008 [23]; two studies used L . infantum: Mathur 2005 [14] , Mandal 2008 [15]: and one study L . amazonensis: Romero 2009 [13] . The ROC plot was better for the studies that used L . infantum with the relative order of L . infantum>L . chagasi>L . amazonensis . Cut off dilution for a positive test on DAT varies from 1∶800 to 1∶6400 . The ROC plot demonstrated that the order of dilutions was 1∶800>1∶1600>1∶3200>1∶6400 . For IFAT the cut off dilution varies from 1∶32 to 1∶400 , ROC plot demonstrated that the order of dilutions was 1∶400>1∶160>1∶80>1∶40>1∶32 . This meta-analysis utilized rigorous selection criteria to select published papers that compared four different serodiagnositic tests currently being utilized in endemic areas for serodiagnosis of VL . We found that recombinant K39 protein used either in a strip test or ELISA format and the DAT using whole promastigote antigen are the most accurate tests for serodiagnosis of VL . During the past 70 years where various strategies have been used for the serodiagnosis of VL , there have been waves of enthusiasm favoring one technique over others [24] , [25] . The formol-gel test ( FGT ) based on gelling and opacification of the serum from a patient with VL in the presence of formaldehyde was the only bedside test available in earlier days of last century to confirm a diagnosis of VL in patients suspected to have Kala-azar [22] . Early on it was discovered that high levels of specific antibodies against L . donovani spp . and a polyclonal B cell activation classified VL as one of the few diseases resembling multiple myeloma because of the high levels of globulins found in patients' serum [26] . This property of VL allowed for the development of serodiagnostic tests such as the DAT . The DAT easily can detect high titers up to or higher 1/64 . 000 dilution folds in a serum of most VL patients [27] , and there are many comparative studies that have demonstrated that DAT is a good option to help physicians in endemic areas to confirm the diagnosis of VL in patients with suggestive symptoms , and when examination for Leishmania parasites in a bone marrow or spleen aspirate biopsies is not available [5] , [6] , [28] , [29] , [30] . Why has the DAT not been developed into an affordable commercial test available for use in endemic areas ? The WHO attempted to push the development of such a test in the early 1990s , but was not successful due to the following reasons: 1 ) an essential part of the test is the growth of parasites , and the variability in the techniques and preservation of the antigen can lead to variation in test results and 2 ) high levels of cross reactivity with other trypanosomatides exist [6] , [31] , [32] , [33] , [34] , [35] . The IFAT for VL is a very sensitive test , but far from being a diagnostic tool in endemic areas , as it has similar limitations in terms of antigen preparation , and in addition requires the use of special microscope equipment [36] . On the other hand , ELISA has been extensively used for serological diagnosis in VL . However , the sensitivity and specificity of this test is variable depending upon the antigen chosen for use [29] , [35] , [37] . More recently , third generation tests for the serodiagnosis of VL have utilized two basic formats: ELISA and immunochromatography [2] , [36] . A 2003 comparison of 11 defined recombinant or synthetic proteins with soluble Leishmania antigen ( SLA ) in an ELISA format revealed that the best antigen was rK39 with 100% sensitivity and 97% specificity in the diagnosis of VL [38] . The recombinant protein K39 ( rK39 ) is a repeated 39-amino acid sequence derived from a gene cloned from L . chagasi and expressed in E . coli . The protein is related to the kinesin family of proteins , has a high epitope density , and is present in high amount in the amastigotes forms of L . donovani complex [39] . rK39 was first demonstrated to be an indicator of disease in L . chagasi infected patients [40] . More recently , the rK39 antigen has been developed into a strip test for the serodiagnosis of VL [10] , [19] , [41] , [42] , [43] . When comparing the rK39 strip test and the DAT , several issues arise . The DAT has the ability to detect low levels of antibodies due to the mosaic of antigens present in the extract . This sensitivity can come at a cost in specificity , as some of these antigens are cross reactive , and therefore careful attention needs to be placed on determining the cut-off values for a positive test [44] . Strip tests also have limitations , as with other serologic tests , patients can have antibody present for months after cure of disease , and also the tests can detect antibodies in the sera of asymptomatic patients [24] . False negative results have been reported and may vary from location to location [36] . In this review , 154 out of 1884 stored samples from clinically ill patients with positive amastigotes in bone marrow aspirate did not react with the strip test . This finding could potentially be explained by variations in sample storage protocols [45] . A previous meta-analysis comparing the DAT and strip test including 30 papers evaluating the DAT 13 evaluating the rK39 strip found that tests are comparable . The DAT was found to be 1% most sensitive and 2% more specific than strip test , but this analysis did not report ROC comparison of sensitivity and specificity because not all studies selected performed both techniques [46] . One important criteria of our meta-analysis study was the inclusion of only studies that compared both methods in each individual study . In addition our meta-analysis included comparison of ELISA and IFAT . In our study the best performance of the strip test were from those studies carried out in Europe , different from Chappuis's meta-analysis which found that sensitivity seemed higher and more homogenous in studies carried out in South Asia [46] . A newly developed assay based on the detection of antibodies to the rk28 fusion protein reported a very promising sensitivity and specificity ( 96% and 98% respectively ) of ELISA to detect anti-Leishmania antibody in sera from VL patients . However , both the rK39 and rK28 antigens demonstrated similar areas under the ROC curves [47] , [48] . Meta-analysis is an important tool that drives direction for best evidences in medicine by comparing studies done in different places , environments and populations as long as the same question was used by different investigators . But , unfortunately when we focus the meta-analysis on sensitivities and specificities of serological tests the heterogeneity introduced by variations in diagnostic thresholds becomes an important limitation of this analytic tool [49] . We suggest that investigators in endemic areas should consider using both rK39 strip test and DAT prior to initiating anti-Leishmania treatment when demonstration of the parasite in bone marrow or spleen aspirate biopsies is not available .	Visceral Leishmaniasis ( VL ) is a neglected tropical disease for which serodiagnostic tests are available , but not yet widely implemented in rural areas . The rK39 recombinant protein is derived from a kinesin-like protein of parasites belonging to the Leishmania donovani complex , and has been used in the last two decades for the serodiagnosis of VL . We present here a systematic review and meta-analysis of studies evaluating serologic assays ( rK39 strip-test , rK39 ELISA , Direct Agglutination Test [DAT] , Indirect Immunofluorescence test [IFAT] and ELISA with a promastigote antigen preparation [p-ELISA] ) to diagnose VL to determine the accuracy of rK39 antigen in comparison to the use of other antigen preparations . Fourteen papers fulfilled the inclusion and exclusion selection criteria . The summarized sensitivity for the rK39-ELISA was 92% followed by IFAT 88% and p-ELISA 87% . The summarized specificity for the three diagnostic tests was 81% , 90% , and 77% . Studies comparing the rK39 strip test with DAT found a similar sensitivity ( 94% ) and specificity ( 89% ) . However , the rK39 strip test was more specific than the IFAT and p-ELISA . In conclusion , we found the rK39 protein used either in a strip test or in an ELISA is a good choice for the serodiagnosis of VL .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "medicine", "infectious", "diseases", "clinical", "laboratory", "sciences", "diagnostic", "medicine", "parasitic", "diseases" ]	2012	Comparative Study of rK39 Leishmania Antigen for Serodiagnosis of Visceral Leishmaniasis: Systematic Review with Meta-Analysis
In 2014 , the United States experienced an epidemic of acute flaccid myelitis ( AFM ) cases in children coincident with a nationwide outbreak of enterovirus D68 ( EV-D68 ) respiratory disease . Up to half of the 2014 AFM patients had EV-D68 RNA detected by RT-PCR in their respiratory secretions , although EV-D68 was only detected in cerebrospinal fluid ( CSF ) from one 2014 AFM patient . Given previously described molecular and epidemiologic associations between EV-D68 and AFM , we sought to develop an animal model by screening seven EV-D68 strains for the ability to induce neurological disease in neonatal mice . We found that four EV-D68 strains from the 2014 outbreak ( out of five tested ) produced a paralytic disease in mice resembling human AFM . The remaining 2014 strain , as well as 1962 prototype EV-D68 strains Fermon and Rhyne , did not produce , or rarely produced , paralysis in mice . In-depth examination of the paralysis caused by a representative 2014 strain , MO/14-18947 , revealed infectious virus , virion particles , and viral genome in the spinal cords of paralyzed mice . Paralysis was elicited in mice following intramuscular , intracerebral , intraperitoneal , and intranasal infection , in descending frequency , and was associated with infection and loss of motor neurons in the anterior horns of spinal cord segments corresponding to paralyzed limbs . Virus isolated from spinal cords of infected mice transmitted disease when injected into naïve mice , fulfilling Koch’s postulates in this model . Finally , we found that EV-D68 immune sera , but not normal mouse sera , protected mice from development of paralysis and death when administered prior to viral challenge . These studies establish an experimental model to study EV-D68-induced myelitis and to better understand disease pathogenesis and develop potential therapies . Enterovirus D68 ( EV-D68 ) was first identified in 1962 after it was isolated from four children in California with acute respiratory illnesses [1] . Enteroviruses are typically spread through fecal-oral transmission , and are associated with diarrheal illnesses , undifferentiated fever with rash , hand-foot-and-mouse disease , meningitis , and encephalitis [2] . EV-D68 , however , possesses several properties more similar to rhinoviruses than conventional enteroviruses , including optimal replication in the cooler temperatures of the upper respiratory tract ( 33°C ) , acid sensitivity , and spread primarily via respiratory , rather than fecal-oral , transmission [1 , 3 , 4] . Passive surveillance data from the National Enterovirus Surveillance System ( NESS ) indicated that EV-D68 has been a rare cause of respiratory illness , with only 26 documented cases of EV-D68 in the United States ( US ) from 1970–2005 [5] . However , within the last decade , EV-D68 outbreaks have become more common worldwide [6 , 7] , and in the second half of 2014 , the US experienced an unprecedented EV-D68 respiratory disease outbreak with over 1150 cases reported nationwide [8–10] . This number is almost certainly an underestimate , as only the most severe cases underwent pathogen identification , and rapid tools for laboratory confirmation of EV-D68 in the clinical setting were not widely available until mid-2015 [11] . Coincident with the US EV-D68 respiratory outbreak , physicians began reporting an increased number of cases of acute flaccid paralysis with a striking resemblance to poliomyelitis [12–16] . The paralysis occurred primarily in young children ( median age ~7 ) [17–20] . Many children experienced prodromal symptoms of fever and upper respiratory illness before the onset of limb weakness [14 , 16–21] . Magnetic resonance imaging ( MRI ) showed signal abnormalities in the anterior horns of the spinal cord , the location of motor neurons innervating upper and lower limbs [14 , 16 , 19–21] . The CDC established guidelines for reporting acute flaccid myelitis ( AFM ) , defined as acute limb weakness with characteristic spinal cord imaging abnormalities on MRI occurring in children on or after August 2014 . Under these guidelines , 120 confirmed cases of AFM from 34 states were documented in 2014 [17] . Sporadic cases meeting the CDC definition , from 0 to 7 per month , continued to be reported in 2015 , and in 2016 the CDC reported 132 confirmed cases of AFM from 37 states . [17] . Case-control study data from the 2014 cluster of AFM cases in Colorado showed that the odds of a child with AFM having had a concomitant EV-D68 respiratory infection were over 10 times greater than for controls with acute respiratory disease [22] . EV-D68 RNA was detected in respiratory secretions as the predominant pathogen in about half of the affected children tested in 2014 , although EV-D68 RNA was amplified from cerebrospinal fluid from only one 2014 AFM patient to date [18 , 19] . There was no statistical correlation of AFM with infection by any other pathogen in respiratory samples [22] , and in-depth metagenomic sequencing of CSF failed to reveal an alternative infectious etiology [18] . Given the lack of consistent CSF isolation , a potential causal role of EV-D68 in AFM has not been formally established . Several additional pieces of evidence , however , suggest that EV-D68 has the capacity to produce central nervous system ( CNS ) disease . Other viruses in the Enterovirus genus , such as polioviruses , enterovirus 70 , and enterovirus 71 , are established causes of acute flaccid paralysis . EV-D68 has also been described as a cause of neurological disease in two previous case reports [5 , 23] . The first involved a young adult who developed acute flaccid paralysis with EV-D68 detected in CSF in 2005 [5] . The second involved a 5-year-old boy who developed fatal EV-D68 meningomyeloencephalitis with neuron loss in motor nuclei of the brain and cervical spinal cord in 2008 [23] . Changes in the viral genome can also potentially contribute to virulence and spread . EV-D68 strains have undergone significant evolutionary shifts since their original isolation [7 , 18] . The Fermon and Rhyne strains , obtained from respiratory swabs of children in California in 1962 , are considered the prototype EV-D68 strains [1] . During the mid-1990’s , the prototype lineage separated into two clades , designated A and C [7] . Clade B further separated from clade C in the mid-2000s [7] . A clade named B1 then separated from clade B around 2010 [18] . Most of the EV-D68 respiratory cases that occurred in the 2014 epidemic were caused by a single lineage of clade B1 viruses related to strains previously seen circulating in the US , Asia , and southern Europe from 2011–2013 [9] . A minority of cases were caused by clade A and C members [9] . Of the AFM patients analyzed in the United States , up to half had respiratory sputum samples positive by reverse transcription polymerase chain reaction ( RT-PCR ) for EV-D68 , with increasing likelihood of EV-D68 detection if the sample was collected closer to the onset of prodromal febrile/respiratory symptoms [18 , 19] . Metagenomic sequencing of available EV-D68 positive respiratory samples in one study from California and Colorado mapped all AFM-associated EV-D68 strains to clade B1 [18] . Because non-clade B1 strains were less prevalent , however , it is possible this study was unable to detect an association between AFM and other clades . In order to develop an animal model of EV-D68-associated neuropathogenesis , we screened both 2014 outbreak and prototype 1962 EV-D68 strains for the ability produce paralysis in mice . The 2014 strains tested were isolated from clinical respiratory samples from the 2014 outbreak and represented clades A , B , and B1 . Five EV-D68 strains from the 2014 outbreak ( clade A strain KY/14-18953; clade B strains IL/14-18952 and CA/14-4231; B1 strains MO/14-18947 and CA/14-4232 ) and two prototype strains ( Fermon and Rhyne ) were screened for the ability to cause neurological disease in two day-old outbred Swiss Webster mice ( Fig 1A and 1B ) . Intracerebral injection of viral strains was chosen to investigate neurovirulence and neurotropism by bypassing potential barriers to viral central nervous system ( CNS ) entry . Strains were injected at the highest available viral titer to maximize the possibility of eliciting disease ( see Methods ) . Mice were monitored daily for up to 4 weeks or until death occurred . Following injection , four out of the five contemporary strains tested ( KY/14-18953 , IL/14-18952 , MO/14-18947 , and CA/14-4232 ) induced paralytic disease in 33–100% of mice ( Fig 1B ) . Paralysis occurred in these mice between dpi 3 –dpi 9 for all strains . Mortality ( 5–70% ) varied by strain , and most of the mice that died had paralysis ( 31 out of 38 deaths , 80% ) . Example images of mice injected with one of these strains , MO/14-18947 , can be seen in Fig 1C and S1–S3 Movies . One contemporary strain , CA/14-4231 , failed to induce paralysis or death despite its close phylogenetic relationship to IL/14-18952 ( Fig 1A and 1B ) . The degree of limb involvement ranged from monoparesis to quadriparesis . Paralysis in very young mice ( postnatal days 5–7 ) could be identified as a reduction of movement in one or more limbs , leading to a reduced ability to crawl and turn when evaluated on a flat surface ( Fig 1C , left panel; S1 Movie ) . In older , more mobile mice ( ~8 days old or more ) , paralysis could be assessed along a continuum from mild loss of motor function , as exemplified by toe or knuckle walking , to the complete inability to use the limb for ambulation ( Fig 1C , center and right panels; S2 and S3 Movies ) . Affected limbs appeared to hang in unnatural positions and developed atrophy over time . Sensation in affected limbs remained grossly intact as determined by response of vocalization and attempt to move away from the noxious stimuli ( toe pinch test ) [24] . Compared to the contemporary strains , mice injected with the prototype strains , Rhyne and Fermon , had fewer signs of morbidity or mortality . One mouse in the Rhyne group ( n = 1 out of 18 , 6% ) developed a transient right hindlimb weakness that was obvious only during ambulation ( Fig 1B ) . Signs of weakness began on dpi 5 but disappeared by dpi 9 . The mouse was able to sit with this limb in proper position while at rest , and the limb did not show signs of atrophy . Rhyne has been reported to cause a myositis [1] , and this may have accounted for the weakness noted . Mice injected with either Fermon ( n = 24 ) or rhabdomyosarcoma ( RD ) cell culture media ( n = 36 ) showed no evidence of paralysis ( Fig 2B ) . Two mice , one in the Fermon group and one in the Rhyne group , died early ( dpi 2–4 ) , although they did not show signs of neurological disease or paralysis prior to death . The cause of death in these two mice was unclear . One strain from clade B1 , MO/14-18947 , was chosen for further in-depth analysis of the paralysis phenotype . MO/14-18947 was chosen because it belongs to clade B1 , the predominant circulating EV-D68 clade in 2014 . In mice infected with MO/14-18947 , the onset of paralysis occurred most frequently between dpi 3–5 , although onset occasionally occurred as late as dpi 9 ( Fig 1D ) . Mice that developed earlier paralysis were more likely to die than mice that developed paralysis at a later age , with an overall death rate in paralyzed mice of 33% . Mice dying of infection had more severe paralysis , as defined by the average number of limbs affected , compared to mice surviving infection ( Fig 1E ) . Following intracerebral injection of the virus , forelimbs were most commonly affected at paralysis onset ( 52% ) , although some mice presented with hindlimb paralysis ( 30% ) or both forelimbs and hindlimbs affected ( 18% ) . Up to 48% of mice experienced disease progression to other limbs after onset of initial paralysis ( Fig 1E ) . The majority ( 72% ) of mice surviving to dpi 28 showed no motor recovery as quantified by the number of limbs affected; the others ( 28% ) recovered between dpi 9 to 15 . Recovery occurred most often in mice with milder disease ( e . g . only one limb affected ) ( Fig 1E ) . Mice injected with MO/14-18947 generated neutralizing antibody titers against EV-D68 , regardless of the presence or absence of paralysis . Sera of mice tested at dpi 12 ( n = 12 ) had antibody titers ranging from 1:40 to 1:1 , 280 . Neutralizing antibody titer increased in MO/14-18947-infected mice tested at dpi 28 ( n = 11 ) , with a titer range of 1:640 to 1:>10 , 240 ( Fig 1F ) . Neutralizing titers in MO/14-18947 mice were compared to those in non-paralytogenic strains in order to determine whether these strains were able to evoke a serological response in the host ( Fig 1F ) . Fermon frequently failed to generate a detectable antibody response in infected mice . Only two mice out of 10 mice tested at dpi 12 developed neutralizing antibodies with titers ranging from 1:20–1:80 , whereas none of the Fermon-infected mice tested at dpi 28 ( n = 18 ) had detectable neutralizing titers . In contrast , mice infected with Rhyne ( n = 7–10 ) and CA/14-4231 ( n = 8–10 ) both generated a sustained neutralizing antibody response against their respective strains , similar to those seen in mice infected with MO/14-18947 ( Fig 1F ) . Mice injected with media control did not have detectable anti-EV-D68 antibody titers at either dpi 12 ( n = 8 ) or dpi 28 ( n = 8 ) . To confirm EV-D68 infection of these strains in mice , viral titers were examined in the skeletal muscle after intramuscular inoculation . All strains examined , except Fermon , replicated to equivalent viral titers in the muscle regardless of the ability to induce paralysis . Fermon did not replicate and did not cause paralysis . The ability to infect skeletal muscle corresponded to the ability to produce neutralizing antibodies ( S1 Table ) . Viral growth in spinal cords from groups of paralyzed mice injected intracerebrally with EV-D68 MO/14-18947 , as well as from litters of mice tested before the typical onset of paralysis ( dpi 0 and dpi 2 ) , was quantified by TCID50 assay ( Fig 2A ) . Infectious virus was not detected in the spinal cords of any mice on dpi 0 , but mean viral titer increased progressively in the spinal cords of mice tested on dpi 2 and dpi 4 . The mean spinal cord titer then remained >1000 TCID50 in mice tested through dpi 8 , after which it dropped steadily and became undetectable by dpi 12 . Two-step quantitative RT-PCR ( qRT-PCR ) for EV-D68 utilizing primers targeting the VP1 capsid gene was used to detect EV-D68 RNA extracted from whole spinal tissue from mice from each time point post-infection ( 20 ) . Results of the qRT-PCR analysis paralleled the results of the TCID50 assay , although titer by qRT-PCR remained detectable at dpi 12 , indicating a longer time of detectability of viral genome as compared to infectious particles in tissue ( Fig 2B ) . Infectious virus was detected in the brains of dpi 0 mice by TCID50 assay approximately an hour after injection ( average 102+/-1 . 05 TCID50/mL ) . However , by dpi 2 the virus could only be detected in brains from 2 out of 11 mice by TCID50 assay , and from dpi 4 to 12 no virus was detected in brain of any paralyzed animals . Infectious virus remained below the limits of quantification by TCID50 assay in sera from all paralyzed mice . Pathological examination of mice infected intracerebrally with MO/14-18947 revealed marked injury and loss of motor neuron populations in the anterior horn of the spinal cord corresponding to the ipsilateral affected limb ( Fig 3 ) . Fig 3 illustrates a typical case with marked injury and loss of the motor neuron population in the anterior horn ipsilateral to the affected right limb as indicated by loss of choline acetyltransferase ( ChAT ) staining , a specific marker of spinal cord anterior horn motor neurons , and NeuN staining , a general marker of neurons ( Fig 3A ) . In contrast , the motor neuron population corresponding to the unaffected left limb appeared intact ( Fig 3A ) . Examination of a consecutive spinal cord section stained for EV-D68 VP2 capsid protein revealed viral antigen within the few remaining motor neurons on the affected side ( Fig 3B–3D ) . No staining was seen on the unaffected side ( Fig 3B ) or in media injected control mice ( S1 Fig ) . To demonstrate the presence of virus at an earlier time point following infection ( during which the motor neuron population was relatively intact ) , spinal cords from dpi 3 mice from a litter injected intracerebrally with MO/14-18947 that had not yet begun to show signs of paralysis were examined . EV-D68 VP2 antigen was consistently detected within motor neurons in these mice by immunostaining ( Fig 3E ) . Unfortunately , direct co-localization of EV-D68 antigen and ChAT was not obtainable due to inability to find compatible primary antibodies from different host species for co-labeling , so consecutive sections ( 10 um apart ) were used for these staining experiments . Transmission electron microscopy ( TEM ) images of the cervical spinal cord anterior horn from a dpi 4 MO/14-18947 injected mouse with forelimb paralysis confirmed the presence of dying cells , consistent in location and morphology with motor neurons , filled with cytoplasmic clusters of ~30 nm particles morphologically consistent with enteroviruses ( Fig 4A and 4B ) [25] . In order to determine whether routes of EV-D68 infection other than intracerebral injection could produce paralytic disease , mice were infected intramuscularly , intranasally , or intraperitoneally with the MO/14-18947 strain . Intramuscular injection of EV-D68 into the left hindlimb produced paralysis in 100% ( n = 18 out of 18 , 100% ) of injected animals . Paralysis onset occurred from dpi 2 to 3 in the injected hindlimb before often progressing to the contralateral hindlimb and then forelimb ( s ) ( Fig 5A , S4 Movie ) . EV-D68 RNA in the spinal cord was detected by RT-PCR in additional mice ( n = 5 out of 5 , 100% ) examined on dpi 3 following left hindlimb injection , with an average estimated copy number per spinal cord of 105 . 3±0 . 7 . Examination of the spinal cords from dpi 3 mice injected into the left hindlimb revealed infection of motor neurons and viral antigen consistent with signs of paralysis ( Fig 5B ) . An additional group of mice ( n = 8 out of 8 ) injected with MO/14-18947 in the right forelimb also developed paralysis starting in the injected limb between dpi 2–4 , consistent with the onset of paralysis seen in the hindlimb injections . In contrast , intramuscular injection with RD control media failed to produce paralysis ( n = 0 out of 14 , 0% ) . Although rare , paralysis following intranasal infection was observed in 2 of 73 mice ( 2 . 7% ) of mice showing signs of paralytic disease . Onset of paralysis occurred in these mice between dpi 8 and 10 . One mouse developed paralysis in the right forelimb only ( Fig 5C; S5 Movie ) ; the other had mild left forelimb paralysis and appeared generally weak in all limbs . The mouse with paralysis onset at dpi 10 was sacrificed on dpi 12 ( Fig 5C ) , and examined for the presence of EV-D68 RNA in its spinal cord tissue by RT-PCR . It was found to be positive for EV-D68 in its spinal cord tissue with an estimated EV-D68 genome copy number of 104 . 0 , comparable to viral titers found in mice after intracerebral injection . The other mouse that developed signs of paralysis after intranasal infection was sacrificed for histological examination on dpi 8 and showed viral antigen in the motor neurons of the cervical spinal cord ( Fig 5D ) . Intraperitoneal injection of MO/14-18947 produced disease in only 1 out of 22 ( n = 4 . 5% ) mice . Paralysis in this mouse occurred in the right rear leg on dpi 5 . To determine whether EV-D68 was the direct cause of the paralytic disease in mice by fulfilling Koch’s postulates [26] , spinal cord lysate from a dpi 4 mouse with signs of paralysis following intracerebral injection of MO/14-18947 was cultured in rhabdomyosarcoma ( RD ) cells . Significant cytopathic effect ( CPE ) was noted within 3 days in the inoculated RD cell culture . The spinal cord-passaged cell culture lysate was passed through a 0 . 22 μM filter syringe and injected intracerebrally into the brains of naïve mice ( Fig 6A ) . 38% ( n = 9 out of 24 ) injected with the cell culture lysate developed paralytic disease between dpi 3 and 8 . This rate of paralysis is consistent with that seen in previous experiments ( e . g . , Fig 1B ) , especially when considering that the inoculum was nearly 100-fold lower than that used in other experiments ( TCID50 104/mL compared with TCID50 106/mL ) . Analysis of 3 mice that developed early severe paralysis revealed high spinal cord viral titers by TCID50 ( Fig 6B ) . Metagenomic deep sequencing followed by SURPI ( sequence-based ultra-rapid pathogen identification ) analysis confirmed the presence of EV-D68 strain MO/14-18947 RNA in the original spinal cord lysate , the RD cell culture lysate , and the spinal cord tissue of the mice injected with the spinal cord-passaged cell culture lysate ( Fig 6B; Tables A-D in S1 Text ) [27] . No paralysis or other signs of disease were seen in mice ( n = 12 ) injected with RD cell culture lysates that had been inoculated with normal mouse spinal cord . To evaluate the role of immune sera in protection against disease , 1 day-old mice were injected intraperitoneally with either pooled immune sera against MO/14-18947 ( neutralizing antibody titers: 1:320–640 ) or pooled control normal mouse sera ( neutralizing antibody titer undetectable ) and then challenged 24 hrs later with intracerebral injection of MO/14-18947 ( 2 . 3 x 106 TCID50/mL ) . 57% of mice ( n = 12 out of 21 ) receiving normal mouse sera developed paralysis compared to only 4 . 5% of mice ( n = 1 out of 22 ) treated with EV-D68 immune sera ( p < 0 . 0002 by Fisher’s Exact Test ) . There were no deaths in the immune sera treated group and 18% mortality in mice receiving normal mouse sera ( p = 0 . 05 by Fisher’s Exact Test ) . We have described a mouse model of spinal cord infection and paralysis caused by clinical isolates of EV-D68 . Of five 2014 EV-D68 strains tested , four strains induced a paralytic disease in mice following intracerebral injection . An in-depth characterization of one of these strains , clade B1 MO/14-18947 , revealed that this paralysis replicated key features of human AFM including a lower motor neuron pattern of paralysis with a predilection for the upper limbs , limited motor recovery over time , and no sensory or cerebral involvement [13 , 16 , 19 , 20] . MRI studies and electromyography of AFM patients suggest loss of motor neurons without damage to sensory pathways ( 12 , 17 , 24 ) . Consistent with these findings , paralyzed mice exhibited no gross loss of sensory function , and viral antigen was detected almost exclusively in cells consistent with motor neurons as determined by ChAT staining , morphology , and anatomical location within the spinal cord . EV-D68 appears to have a specific tropism for spinal cord motor neurons as demonstrated by the absence of significant growth in brain and the restricted pattern of antigen distribution and neuronal injury in the spinal cord . The kinetics of viral growth closely paralleled the observed development of paralysis , suggesting that direct viral injury , rather than a post-infectious immune-mediated process is the most likely mechanism of neuronal cell loss and subsequent paralysis . Fulfillment of Koch’s postulates supports a causal role of EV-D68 infection in the development of paralytic disease in this model . We also established that EV-D68 neuroinvasion could occur by several alternative routes in addition to producing disease after intracerebral injection . Although intranasal infection rarely produced paralytic disease , mice that developed signs of paralysis exhibited spinal cord infection similar to that seen following other routes of infection . The rarity of paralytic disease following intranasal infection ( ~3% ) of EV-D68 is consistent with the low incidence of AFM after EV-D68 respiratory infection in humans ( likely <1% ) . Surprisingly , intramuscular infection produced paralysis even more consistently than intracerebral infection . As EV-D68 appears to have specific tropism for motor neurons , we hypothesize that intramuscular infection may be a more efficient and direct pathway to motor neuron infection than intracerebral injection . The exact pathways and mechanism of spread for EV-D68 after virus inoculation at specific sites ( intramuscular , intracerebral , or intranasal ) remain to be fully elucidated . In the current study , mice injected intramuscularly in either the forelimb or the hindlimb initially developed paralysis in the inoculated limb . Similarly , the intranasally inoculated mice and the majority of the intracerebrally inoculated mice initially developed forelimb paralysis . This pattern of paralysis onset after different routes of inoculation is most consistent with viruses that spread along neural pathways . Neural spread typically involves initial infection of and injury to the segments of the spinal cord containing neurons innervating the site of inoculation [28] . In contrast , viremic spread characteristically results in uniform involvement of motor neurons innervating forelimbs and hindlimbs , regardless of the site of inoculation [28] . The rarity of paralysis following intraperitoneal inoculation ( a proxy for intravenous inoculation ) and the lack of infectious virus in sera of paralyzed mice in this model argue against a critical role for viremic spread in neuropathogenesis . Notably , EV-D68 viremia has only been rarely identified in human AFM patients ( 1 in 25 , or 4% , as reported in one study ) [18] . Ultimately , several mechanisms may facilitate spread ( neural or viremic ) depending on the route of infection , as have been described for polioviruses [29] . To date , only EV-D68 strains from clade B1 have been isolated from AFM patients [18] . A previous study identified six polymorphisms confined to AFM-associated clade B1 strains that could represent genetic changes associated with neurovirulence [18] . However , in the current study , strains from multiple clades ( A , B , and B1 ) produced paralysis in neonatal mice ( a 2014 clade C strain was not available for testing ) ( S2 Fig ) . These data indicate that the clade B1-specific polymorphisms and sequence homology alone do not fully explain paralysis in this mouse model . In addition , the observed molecular epidemiologic association of clade B1 with AFM may be due to the high overall prevalence of this clade circulating in the population in 2014 [18] . Unexpectedly , one contemporary clade B strain , CA/14-4231 , failed to induce paralysis , despite its close phylogenetic relationship to a neurovirulent clade B strain , IL/14-18952 ( S2 Fig ) . Failure to induce paralysis was not a result of the inability to infect mice ( S1 Table ) , suggesting that CA/14-4231 may lack genetic sequences critical for neurovirulence . Further comparative analyses using infectious clones will likely to be needed to establish the determinants of neurovirulence and host range of EV-D68 in mice . It would also be informative to test EV-D68 strains from Europe , North America , and Africa associated with smaller outbreaks in 2009/2010 ( clades A , B , and C ) , but these were not available for the current study [7 , 30] . Furthermore , population factors , such as spread in a large immunologically naïve population and host-specific genetic susceptibility , also cannot be fully ruled out as contributors to AFM . The previous finding of a sibling pair , both infected by identical strains of a clade B1 EV-D68 strain , yet only one developing AFM [18 , 19] , points to the potential importance of such host-related factors in viral neuropathogenesis . Finally , a mouse model is an important first step in the in vivo screening of potential drug and vaccine therapies against EV-D68 , for which there are currently no established treatments . In the current study , treatment of naïve neonatal mice with EV-D68 immune sera containing anti-EV-D68 antibodies , but not normal mouse sera , prevented paralysis and death after viral challenge . Additional mouse studies testing specific antiviral agents or delaying passive antibody administration until later in the disease course are currently in progress . The results presented here suggest that immunomodulatory strategies such as vaccination or the use of EV-D68 hyperimmune sera may be potentially effective strategies for treatment or prevention of EV-D68 associated neurological disease . Our findings are of particular relevance given the recent surge in AFM cases in 2016 [17] . All studies were done in accordance with the University of Colorado IACUC and Animal Use Committee ( B-34716 ( 03 ) 1E ) . Mice were cared for in adherence to the NIH Guide to the Care and Use of Laboratory Mice . Mouse pups exhibiting paralysis were euthanized if unable to nurse . Mice were anaesthetized with inhaled isoflurane before tissue collection or perfusion . All EV-D68 viral strains were obtained from the American Type Culture Collection ( ATCC ) or from the California Department of Public Health ( courtesy of Shigeo Yagi ) . Viral stocks were grown in RD cells ( ATCC ) at 33°C and 5% CO2 until most cells were dead or dying . Cells debris was removed from RD grown stocks by ultracentrifugation . Titers of viral stocks were determined by TCID50 assay as calculated by the Kärber method . The TCID50/mL titers for each strain used in this paper are as follows: Fermon– 8x106 , Rhyne– 1x107 , MO/14-18947 ( clade B1 ) – 5x106 , CA/14-4232 ( clade B1 ) – 1x105 , IL/14-18952 ( clade B ) – 5x107 , CA/14-4231 ( clade B ) – 3x107 , KY/14-18953 ( clade A ) – 2x106 . The pure culture stock used in the Koch’s postulate experiment was grown from whole spinal cord lysate on RD cells . The stock was passed through a 0 . 22-micron sterile syringe filter before injection . All experiments were performed on NIH Swiss Webster mouse pups of both sexes from Envigo ( Indianapolis , IN ) . Mouse litters were randomly assigned to experimental groups . Unless otherwise specified , virus infections were performed on two day-old mice . For intracerebral infection , mice were injected with ~20 μL via insulin syringe ( 29 G needle ) with undiluted virus or RD control media stock into the right hemisphere just anterior to lamboid suture . For intramuscular infection , mice were injected with ~20 μL via insulin syringe ( 29 G needle ) with undiluted virus into the medial aspect of the left hindlimb . For intraperitoneal infection , mice were injected with ~20 μL via insulin syringe ( 29 G needle ) with undiluted virus into the peritoneal cavity . For intranasal infection , a total of 40 μL of undiluted virus was micro-pipetted onto the noses of a post-natal day 2 pups in two boluses of 20 μL with a 30-minute interval between exposures . Mice were examined daily for signs of paralysis . For passive transfer experiments , 100 μL of pooled sera was given by intraperitoneal injection via insulin syringe on post-natal day 1 to mice randomized between two litters . Immune sera was pooled from dpi 28 mice ( n = 12 ) previously injected intracerebrally with US/MO/14-18947 . Control sera were pooled from dpi 28 mice ( n = 12 ) previously injected intracerebrally with RD control media . The passive antibody transfer experiment data is displayed as a combination of two replicate experiments . Mice were randomized between control and experimental conditions for each replicate . In all studies , no mice were excluded from analyses . Serum was collected for TCID50 analysis and neutralizing antibody titer analysis . After collection , whole blood was placed immediately on ice , spun at 14 , 000 rpm for 10 minutes at 4°C , and then the serum was removed and placed in a fresh tube . Serum was stored at -20°C if used for antibody neutralization or at -80°C if used for TCID50 analysis . Brains were removed intact and placed in a BeadBug tissue homogenizer tubes ( Benchmark Scientific , Edison , NJ ) with 3 . 0 mm beads and brought to a standard volume of 1 mL with PBS . Spinal cords were removed intact as previously described and placed in a BeadBug tissue homogenizer tubes with 3 . 0 mm beads and brought to a standard volume of 0 . 3 mL with PBS [32] . Muscle tissue ( anterior and posterior lower leg muscles ) was removed and weighed , and it was then brought to a standard volume of 0 . 3 mL of PBS . Tissues were then mechanically lysed in the BeadBug microtube homogenizer ( Benchmark Scientific , Edison , NJ ) . 60 μL of each tissue was removed for TCID50 analysis . Titer of virus in each tissue was determined by TCID50 assay and final viral titer was calculated using the Kärber method . Lack of growth below the detectable limit was graphed as a titer of zero . For RT-PCR , total RNA from the remaining spinal cord sample was extracted using a Qiagen RNeasy Plus Micro Kit with RNA carrier ( Hilden , GER ) to facilitate RNA pull-down from this small tissue . Total RNA from each spinal cord sample was used to make cDNA . Each spinal cord sample was converted to cDNA using a Bio-Rad iScript RT supermix ( Berkley , CA ) . Equivalent volumes of cDNA were used for PCR with degenerate primers targeting the VP1 gene ( 20 ) ( PCR protocol: 95°C 3 min , 40x cycles of: 95°C 10 sec , 53°C 30 sec , 72°C 30 sec , melt curves: 65–95°C in 0 . 5°C steps ) . Samples were compared to a plasmid standard curve [11] . Temperature melt curves temperature and slope were used to assess the quality of each sample . The starting genome copy number for each dilution of the standard curve was estimated using spectrophotometer data . The starting genome copy number per μL was correlated to the resulting cycle threshold ( Ct ) value at each dilution in the standard curve , and the standard curve equation was calculated using Excel ( Microsoft , Redmond , WA ) . Mice were anaesthetized with isoflurane before perfusion . Mice were perfused intracardially with 4% Na-periodate-lysine-paraformaldehyde fixative ( PLP , 0 . 01M sodium periodate , 0 . 075M lysine , 2 . 0% paraformaldehyde , 0 . 037M phosphate ) before tissue removal . Spinal cords were post-fixed in PLP at 4°C overnight and then transferred to a solution of 30% sucrose in PBS ( pH 7 . 4 ) until saturated . Spinal cords were then kept in a 1:1 mixture of a 20% sucrose solution in PBS and Optimal Cutting Temperature ( O . C . T . ) compound for 1–2 days at 4°C before being placed into embedding cups filled with O . C . T . compound and rapidly frozen . Ten μm thick serial frozen sections were cut on a cryostat . Sections were then incubated with primary antibodies overnight: rabbit anti-ChAT ( 1:100; #ab178850 , Abcam , Cambridge , UK ) , rabbit anti-EV-D68 VP2 ( 1:100; #GTX132314 , GeneTex , Irvine , CA ) , and mouse anti-Neun ( 1:50; #MAB377 , Millipore , Billerica , MA ) . Alexa Fluor goat-anti-rabbit 488 or goat-anti-mouse 546 ( 1:1000; Molecular Probes , Eugene , OR ) were used for secondary antibody labeling . Nuclei were labeled with Hoechst 33342 ( 10 ng/mL; Invitrogen , Carlsbad , CA ) . Sections imaged at 100X were taken as two separate images in a single plane and stitched with FIJI [33 , 34] . The 200X and 600X images were imaged as Z-series and flatted as brightest point projections . All images were taken on an FV-1000 confocal microscope ( Olympus , Center Valley , PA ) and brightened linearly for publication with FIJI . Paralyzed mice were perfused intracardially with ice-cold 0 . 1 M phosphate buffer followed by Karnovsky’s fixative ( 2 . 5% glutaraldehyde , 2% paraformaldehyde , in 0 . 1 M phosphate buffer at pH 7 . 4 ) until rigid . Extracted spinal cords were post-fixed at 4°C overnight . Cervical and lumbar regions were dissected from the spinal cords and sent to the Electron Microscopy Core for further processing . Spinal cord regions were sliced on a vibratome into 200 μm-thick sections for resin embedding . Ultra-thin sections ( 65 nm ) were cut on a Reichert Ultracut E from a small trapezoid positioned over the anterior horns and were picked up on Formvar-coated slot grids ( EMS ) and then stained with uranyl acetate and lead citrate . Sections were imaged on an FEI Technai G2 transmission electron microscope ( Hillsboro , OR ) with an AMT digital camera ( Woburn , MA ) . RD cells were grown to form a sub-confluent cell layer in a 96-well plate . Serum samples were heat inactivated for 30 minutes at 56°C . To perform the virus neutralization assay , stock virus ( US/MO/14-18947 ) at a concentration of 100-TCID50 and serum dilutions starting at a 1:10 in eleven 2-fold steps were mixed and incubated for 1 hour at 37°C . Diluent and serum were mixed 1:1 and used as a control . After incubation , dilution mixtures were inoculated into separate wells of the cell plate , and the plate was placed in a 33°C 5% CO2 incubator . Cells in the plate were observed for evidence of CPE for 2 weeks . The reciprocal of the highest dilution that inhibited viral CPE was taken as the neutralizing titer . Sera that failed to protect at 1:10 were considered to have an undetectable titer that was graphed as zero . See “Tissue Processing and Analysis” for information on anesthesia and tissue collection . Nucleic acid was extracted from 50 μL of spinal cords lysate with the Direct-zol RNA Kit ( Zymo Research , Irvine , CA ) or 50 μL virus culture using the automated Qiagen EZ1 robotic instrument ( Qiagen , Valencia , CA ) , followed by treatment with Turbo DNase , then randomly reverse transcribed to cDNA with random hexamer primers . NGS libraries were constructed by using the Nextera XT kit ( Illumina , San Diego , CA ) , followed by 161-bp , single-end sequencing on an Illumina MiSeq instrument . Metagenomic next-generation sequencing data were analyzed using the SURPI ( "sequence-based ultrarapid pathogen identification" ) computational pipeline [27] , which identifies viruses , bacteria , fungi , and parasites by computational subtraction of human host sequences followed by nucleotide and translated nucleotide ( protein ) alignment of remaining reads to all microbial sequences present in the National Center for Biotechnology Information ( NCBI ) GenBank database ( as of December 2015 ) . Raw SURPI outputs consisting of aligned microbial reads were taxonomically classified to the appropriate rank ( family , genus , or species ) by use of an in-house developed LCA ( lowest common ancestor ) algorithm incorporating the SNAP nucleotide aligner ( v0 . 15 . 4 ) [35] . Summary read count tables were generated for viruses , bacteria , and eukaryotic organisms including fungi and parasites outside of the phylum Chordata or kingdom Viridiplantae ( "non-chordate eukaryotes" ) . Reads aligning to enterovirus D68 ( EV-D68 ) were automatically mapped using SURPI to the US/MO/14-18947 genome ( GenBank accession KM851225 . 1 ) for determination of percent genomic coverage achieved . Statistical analyses were done using GraphPad Prism ver 6 ( Carlsbad , CA ) . Comparisons of categorical data were done with a Fisher’s exact test . Data were considered significant at p ≤ 0 . 05 for all statistical tests . Re-sequenced EV-D68 genomes corresponding to the culture stocks of the EV-D68 strains used for infection and recovered viral RNA from passaged spinal cord lysates have been deposited in the National Center for Biotechnology Information ( NCBI ) GenBank database under the following accession numbers ( KU844178-KU844181 ) . Deep sequencing data corresponding to viral culture supernatants and the spinal cord lysates used for EV-D68 genome assembly and metagenomic analyses have been deposited in the NCBI Sequence Read Archive ( accession number SRP055445 ) . The assembled genomes and associated deep sequencing data can also be found as part of NCBI BioProject PRJNA266569 .	Reports of polio-like paralysis , referred to as acute flaccid myelitis ( AFM ) , have recently emerged in association with infections caused by enterovirus D68 ( EV-D68 ) . In the second half of 2014 , 120 cases of AFM , mostly in young children , were reported during a nationwide outbreak of EV-D68 respiratory disease . The number of AFM cases has risen again in 2016 . Although epidemiological evidence between EV-D68 infection and AFM is accumulating , a causal link has not been definitely established . Here we demonstrate that strains of EV-D68 recovered during the 2014 epidemic can cause a paralytic illness in mice that resembles human AFM . Evidence that EV-D68 causes paralysis in this mouse model include: ( 1 ) loss of spinal cord motor neurons innervating paralyzed limbs , ( 2 ) detection of virus in the spinal cord and , specifically , motor neurons , ( 3 ) transmission of neurological disease when injecting virus isolated from spinal cords of paralyzed mice into naïve mice , thus fulfilling Koch’s postulates , and ( 4 ) the ability to prevent AFM by pre-administering serum containing EV-D68 antibodies from previously infected mice . This experimental mouse model can be used to better understand the pathogenesis of EV-D68-induced CNS disease and to facilitate the development of potential therapies .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "medicine", "and", "health", "sciences", "respiratory", "infections", "nervous", "system", "neuroscience", "pulmonology", "motor", "neurons", "animal", "models", "model", "organisms", "experimental", "organism", "systems", "mammalian", "genomics", "research", "and", "analysis", "methods", "sequence", "analysis", "spinal", "cord", "infectious", "diseases", "sequence", "alignment", "animal", "cells", "bioinformatics", "enterovirus", "infection", "mouse", "models", "animal", "genomics", "cellular", "neuroscience", "dna", "sequence", "analysis", "neuroanatomy", "anatomy", "cell", "biology", "database", "and", "informatics", "methods", "neurons", "genetics", "biology", "and", "life", "sciences", "cellular", "types", "viral", "diseases", "genomics" ]	2017	A mouse model of paralytic myelitis caused by enterovirus D68
FtsZ , the primary protein of the bacterial Z ring guiding cell division , has been recently shown to engage in intriguing treadmilling dynamics along the circumference of the division plane . When coreconstituted in vitro with FtsA , one of its natural membrane anchors , on flat supported membranes , these proteins assemble into dynamic chiral vortices compatible with treadmilling of curved polar filaments . Replacing FtsA by a membrane-targeting sequence ( mts ) to FtsZ , we have discovered conditions for the formation of dynamic rings , showing that the phenomenon is intrinsic to FtsZ . Ring formation is only observed for a narrow range of protein concentrations at the bilayer , which is highly modulated by free Mg2+ and depends upon guanosine triphosphate ( GTP ) hydrolysis . Interestingly , the direction of rotation can be reversed by switching the mts from the C-terminus to the N-terminus of the protein , implying that the filament attachment must have a perpendicular component to both curvature and polarity . Remarkably , this chirality switch concurs with previously shown inward or outward membrane deformations by the respective FtsZ mutants . Our results lead us to suggest an intrinsic helicity of FtsZ filaments with more than one direction of curvature , supporting earlier hypotheses and experimental evidence . For the fundamental task of cell and organelle division , life has evolved various strategies , many of which are based on ringlike contractile structures assembling from within a compartment to induce binary fission . The exact mechanisms of contraction of these rings are , however , poorly understood due to the plethora of different molecules involved , concealing presumably simple fundamental motifs . The bacterial FtsZ ring is a good example of such a structure . The FtsZ protein , a self-assembling GTPase present in the cytoplasm of most bacteria , is a tubulin homologue and the main component of the E . coli divisome , the molecular machinery driving cytokinesis [1 , 2] . It interacts with additional anchor proteins to form a dynamic ring at the cytoplasmic membrane , which acts as a scaffold to recruit the remaining regulating elements of the divisome and cell wall–building machinery [3] . Intriguingly , FtsZ filaments were recently found to treadmill circumferentially around the division plane , guiding cell wall synthesis [4 , 5] . In spite of this compositional complexity , the primary role of FtsZ in the contractile ring of many organisms is unquestioned . However , based on the structural analysis of the protein and its assemblies , so far , no decisive model has surfaced for how exactly structural changes within monomers and filaments could be transmitted into large-scale contractile forces [6] . Guanosine triphosphate ( GTP ) -dependent structural changes of FtsZ monomers and membrane-attached filaments toward greater curvatures have been proposed , but evidence has been lacking for how a continuously shrinking membrane orifice could be engineered from them . This is mainly due to the fact that although the closure dynamics of Z rings could be observed in vivo [7 , 8] , there is yet no direct proof that purified Z rings may actively proceed to closure through all stages of increasing curvature [3 , 9] . For membrane attachment , FtsZ requires either FtsA or ZipA . Together they form the so-called proto-ring , the first molecular assembly of the divisome [2 , 10] . FtsA is an amphitropic protein that associates to the membrane by an ATP-linked process mediated by a short amphipathic helix [11] . The bitopic membrane protein ZipA contains a short N-terminal region facing toward the periplasmic space , a transmembrane region , and a C-terminal FtsZ-interacting domain connected by a flexible linker region [12] . FtsZ binds the proto-ring tethering elements through its C-terminal end , which is also the interaction region for FtsZ-regulating proteins such as MinC , inhibiting FtsZ polymerization and hence FtsZ ring formation at undesired locations . Thus , the C-terminal region of FtsZ acts as a central hub , integrating signals that modulate divisome assembly in E . coli [2] . The membrane anchor proteins can be bypassed by an FtsZ chimera—FtsZ–yellow fluorescent protein ( YFP ) –membrane-targeting sequence ( mts ) —through replacing the FtsZ central hub with a YFP and an amphipathic helix to provide autonomous membrane attachment . This mutant was found to become internalized and accumulated in narrow regions of tubular liposomes , forming ringlike structures . Polymerization of this chimeric FtsZ protein at the external face of liposomes has also shown to induce inward or outward deformations , depending on the location of the mts ( C-terminus or N-terminus ) [13 , 14] . Recently , the coreconstitution of FtsZ and FtsA on supported bilayers has revealed that FtsA promotes the self-organization of FtsZ fibers into dynamic patterns , giving rise to coordinated streams and swirling rings with preferential directions , due to treadmilling dynamics [15] . In contrast , this dynamic behavior was not observed when FtsZ was tethered to the membrane through ZipA or when the membrane-targeted FtsZ variant was used . Interestingly , FtsZ–FtsA dynamic vortices showed no apparent change in size and curvature , suggesting that the functional role of the energy-consuming circumferential movement of FtsZ in a ring of apparently conserved radius was a secondary one: i . e . , the sequential spatial targeting of downstream enzymes for cell wall assembly . There was further evidence by recent in vivo studies demonstrating the circumferential treadmilling dynamics of FtsZ in dividing cells and directly connecting it to spatial targeting of peptidoglycan synthesis [4 , 5] . However , in spite of a wave of new papers targeting the exact role of FtsZ in ring architecture , the current models of how exactly the Z ring is constricting over time , and how the large accessible range of FtsZ filament curvatures may be exploited in this process , are still disappointingly vague . To study the intrinsic role of FtsZ in the formation of dynamic patterns , we thus revisited the in vitro reconstitution of the membrane-targeted chimeric FtsZ variant , FtsZ-YFP-mts , to flat supported membranes . We further addressed the question whether the pronounced spatial dynamics in FtsZ vortices on the membrane could constitute a direct , rather than indirect , spatial cue in the constriction process of the Z ring . To this end , we compared FtsZ mutants that showed different deformation phenotypes when targeted to free-standing membranes . Our results can be summarized as follows: The protein chimera FtsZ-YFP-mts ( 0 . 5 μM ) in its GDP-bound form ( corresponding to a nonassembled state , according to sedimentation velocity , S1 Fig ) did not form visible structures on a supported lipid membrane , as revealed by a total internal reflection fluorescence microscope ( TIRFM ) . We have found that FtsZ-YFP-mts under assembly-promoting conditions ( 4 mM GTP , 5 mM Mg2+ ) formed filaments on supported lipid bilayers ( SLBs ) , which self-organize with time into dynamic ringlike structures ( S1 Movie ) . The assembly of the dynamic rings is a time-dependent phenomenon . After several minutes of GTP addition—during which highly dynamic short filaments were observed to attach , detach , and diffuse on the surface—longer curved filaments appeared to grow directionally ( Fig 1A panel 5:00 ) . At this stage , intrinsic motion drives filament–filament interactions to create small and dim closed circular structures ( Fig 1A panel 15:00 ) . These structures tend to be highly unstable: Closed filaments were able to open , to fuse with adjacent filaments , or to close back . At later times , closed circular strands turned into thicker ringlike structures ( Fig 1A panel 30:00 ) . To understand the formation of circular structures and how they evolve into stable and thicker rings , we resolved and tracked individual filaments ( consisting of several protofilaments ) before stable ring formation . Here , we identified nucleation sites where clockwise chiral growth leads to the formation of circular structures ( Fig 1B , S2 Movie ) . Despite the fact that growth seems to be a discontinuous phenomenon limited by the accessibility of soluble protein , we can estimate a mean growth rate ( slope p in the kymograph ) which was found to be around 60 nm/s . Strikingly , these images also showed filament flexibility ( panels 7–12 ) and breakage ( panels 15–18 ) resulting in the formation of short free fragments . Such fragments were found to glide and “explore” the surface via treadmilling ( Fig 1C , S3 Movie ) , indicating that this process fuels filament–filament interactions and therefore ulterior formation of closed circular structures . The kymograph in Fig 1C showed a representative example of one filament growing in the leading edge ( p-line ) and shrinking in the trailing edge ( d-line ) . A coarse estimation of the velocity of displacement of this filament was about 55 nm/s . We further investigated the impact of protein concentration on the stability and dynamics of FtsZ vortex formation in the presence of GTP ( 4 mM ) and Mg2+ ( 5 mM ) . Below 0 . 2 μM , no FtsZ filaments could be detected ( S2 Fig ) . Interestingly , increasing the protein concentration to around 1 . 0 μM resulted in the formation of abundant three-dimensional polymer networks on the membrane , and no dynamic FtsZ rings were observed ( S2 Fig , [15] ) . These results showed that the self-organization behavior of membrane-targeted FtsZ polymers was critically dependent on total protein concentration . Next , we compared the kinetics of protein binding to the membrane at 0 . 2 μM and 0 . 5 μM of protein ( Fig 2A ) , under conditions previously used to detect the swirling rings ( see Fig 1 ) . Upon the addition of GTP ( 4 mM ) and Mg2+ ( 5 mM ) , a similar membrane adsorption rate ( Fig 2A ) and the parallel appearance of short and highly dynamic filaments ( Fig 2B ) were initially found for the two protein concentrations . Remarkably , the transition from short filaments to rudimentary circular structures ( gray area in Fig 2A ) also occurred at similar times in both cases . After a lag time of around 10 min , the adsorption rate was found to be significantly slower at 0 . 2 μM than at 0 . 5 μM , suggesting that the kinetics of ring stabilization and widening of the structures was concentration dependent ( Fig 2A ) . These differences , found at elapsed times greater than 10 min , also correlated with the fact that the morphology of the rings observed at a protein concentration of 0 . 2 μM after 45 min incubation ( Fig 2B , bottom right panel ) was similar to the ones obtained after a lag time of 20 min when 0 . 5 μM protein was used ( Fig 2B , upper mid panel ) . The morphological similarity found at these two time points ( denoted as 2 and 3 in Fig 2A ) occurred at a similar protein coverage of the membrane , suggesting that protein surface density , rather than bulk concentration , is the key parameter determining the nature of the network that assembles on the membrane . The correlation between the morphologies at time points 2 and 3 of Fig 2A was further established by determining the average diameter of the formed rings to be about 1 μm at both protein concentrations ( Fig 2B ) . This suggests that although the adsorption rates were different at these time points , the proteins condensed into similar structures . Then , we monitored the impact of GTP concentration ( 0 . 04 , 0 . 4 , and 4 mM ) on the formation of swirling vortices at fixed protein ( 0 . 2 μM ) and Mg2+ ( 5 mM ) concentration . Surprisingly , at the lowest GTP assayed ( 0 . 04 mM ) , a highly ordered mesh of static filaments was found at t = 0 . These filaments retained a certain degree of curvature and behaved as a nematic phase that entirely covered the membrane area ( Fig 3 , left panel ) . A similar behavior was also observed at intermediate GTP concentration ( 0 . 4 mM ) ( S3 Fig ) . Notably , the surface mean intensity is 3-fold ( approximately 1 , 500 ) increased , compared to the minimal density to form rings ( Fig 2 ) , suggesting that the parallel arrangement of filaments correlates with a high-density regime of protein . Furthermore , aligned filaments showed no significant change after 10 min , in contrast to dynamic rings ( S3 Fig ) . It is known that Mg2+ favors self-association and assembly of FtsZ both in solution and at membranes [16 , 17] . Therefore , one possibility could be that the free Mg2+ controls the surface protein density , rather than the total GTP concentration , since GTP is known to bind Mg2+ with affinity in the millimolar range [18] . To examine this alternative , we repeated the self-organization assays at 0 . 04 mM GTP in the presence of 1 mM free Mg2+ , which resulted in the formation of chiral vortices ( Fig 3 , right panel ) . Interestingly , the emergence of chiral vortices was found to correlate with a significantly lower mean surface protein density than the one measured at 0 . 04 mM GTP and 5 mM Mg2+ that resulted in the dense packing of static polymers ( Fig 3 , left panel ) . These findings show that free Mg2+ controls the concentration of GTP FtsZ-YFP-mts polymers at the membrane and then the self-assembly of the FtsZ filaments in the membrane . After formation , single rings reach a quasi-steady state as rotating vortices , meaning that the light intensity along their perimeter shows a nearly periodic time dependence . These rotating structures formed by the membrane-targeted FtsZ-YFP-mts ( mts C-terminal ) consistently showed a chiral clockwise rotation ( Fig 4A and S4 Movie ) . The directional ring dynamics were confirmed by the positive slope of kymographs generated along the ring circumference . Quantifying the slope of the kymograph ( see Materials and methods and S4 Fig ) of N = 60 rings , we calculated the velocity distribution with a mean velocity of 34 nm/s or 3 . 9° sec−1 for rings of about 500 nm radius ( Fig 4D ) . Interestingly , the rotational velocities measured here are in good agreement with those reported in vivo ( 30 nm/s ) , in spite of the significantly reduced complexity of the reconstituted system [4 , 5] . We next sought to understand whether there was any relationship between the structural features of the protein and the obviously chiral dynamics of our FtsZ mutants . Hence , we made use of a previously established chimera variant that was shown to have opposite effects on deformable membranes [14 , 19] . In the presence of GTP , FtsZ-YFP-mts is able to induce inward ( concave ) deformations on lipid vesicles . Strikingly , when the mts sequence is switched to the N-terminus , an outward ( convex ) deformation is observed [14] . To study the role of the position of the mts in our dynamic vortices , we carried out similar self-organization assays using an FtsZ chimera in which the membrane attachment was located at the opposite—the N-terminal—end ( mts-H-FtsZ-YFP ) . Upon addition of GTP and Mg2+ , defined dynamic rings were observed ( Fig 4B ) . Strikingly , now the FtsZ swirls appeared to rotate counterclockwise ( S4 Movie ) , a feature that was confirmed by the negative slope of the kymographs ( Fig 4B ) . As before , we measured the slope of kymographs for N = 50 different rings to calculate the velocity distribution with a mean of about 25 nm/s or 2 . 8° sec−1 for a ring of 500 nm radius . This velocity is slower than that of the FtsZ-YFP-mts vortices ( 34 nm/s ) ( Fig 4D ) . These observations show that the positioning of the mts determines the direction of polymerization , as it does for membrane binding and transformation . The fact that the N-terminal mts mutant—without a protein spacer between the FtsZ and the membrane attachment—results in the same qualitative dynamic behavior , although being inverted in chirality , also refutes potential speculations that YFP may take over a necessary ( sterical ) role of FtsA [15] . To further investigate how exactly GTP hydrolysis influences the formation of collective streams , we carried out similar self-organization assays using a variant of the FtsZ chimera with no GTPase activity ( S5 Fig ) , in which the Threonine at position 108 was replaced by an Alanine ( FtsZ[T108A]-YFP-mts ) . Well-defined rings similar in size to the ones found with FtsZ-YFP-mts could be observed upon the addition of GTP and Mg2+ ( Fig 4C ) . Interestingly , these rings did not seem to treadmill and rotate ( S4 Movie ) , as evidenced by the lack of clear patterns in the kymographs generated to track polymer dynamics ( Fig 4C ) . Interestingly , FtsZ[T108A]-YFP-mts rings grow from nucleation points in a less dynamic manner compared to FtsZ-YFP-mts ( S5 Movie ) . From these results , we conclude that GTPase activity is not required for the formation but for the quasi-steady-state rotational dynamics of the ring patterns , suggesting that GTPase activity particularly promotes filament destabilization in the trailing edge . Treadmilling can be explained by an imbalance between growth and shrinkage at the two opposite ends of the polar filament . Since treadmilling is obviously GTP-turnover dependent , and the growth into ringlike structures by capturing preformed diffusing filaments is not , the critical requirement for treadmilling seems to be the destabilization and shrinkage at the trailing edge . In order to directly visualize the destabilization dependent on nucleotide state , we developed a single molecule assay using FtsZ-YFP-mts incubated with fluorescently labeled nanobodies ( green fluorescent protein [GFP]-Booster-Atto647N , see Materials and methods ) ( Fig 5A ) to investigate the protein turnover at the membrane , implying that faster disassembly suggests higher destabilization . By measuring the probability of protein detachment as a function of time ( S6 Fig ) , we could calculate the mean residence time of single FtsZ subunits within the filaments on the membrane . Using this analysis , we found that the mean residence time of FtsZ-YFP-mts in fragments forming dynamic rings was tr=11 . 5s ( Fig 5B ) , in good agreement with previous fluorescence recovery after photobleaching ( FRAP ) studies with native FtsZ [20 , 21] . This residence time turns out to be significantly faster than for the GTP hydrolysis–deficient mutant FtsZ[T108A]-YFP-mts ( S6 Fig , similar to the photobleaching time scale contribution approximately 32 s ) . By considering the rotational speed as measured in Fig 4D and this residence time , we reason that rings are assembled by multiple filaments that treadmill in a synchronized manner , with a mean length of 〈l〉=v-tres=390nm ( 78 monomers ) . In comparison , in vitro assembly of native FtsZ showed shorter filaments that were , on average , 120 to 200 nm long ( 30–50 monomers ) [21 , 22] . In addition , we measured the residence time without GTP ( GDP form ) with 1 mM free Mg2+ . Assuming that the residence time of a polymer of n-monomers scales to the power of n trpol~ ( trmon ) n , one can estimate that the residence time associated to one monomer is approximately 1 s , which agrees with our results ( approximately 0 . 8 s ) in GDP form at 1 mM free Mg2+ . Our single molecule experiments also allowed us to directly elucidate the impact of free Mg2+ and , in particular , its obvious role in the formation of the high-density FtsZ mesh . It was found that the protein release from the membrane upon GTP addition was slower at 5 mM ( tr = 15 . 1 s ) than at 1 mM free Mg2+ ( Fig 5B ) . These findings represent compelling evidence that the formation of a high-density mesh of filaments is linked to the slow detachment of protein , at least when compared to the dynamic rings at lower free Mg2+ ( 1 mM ) . Also , the residence time of FtsZ in GDP form at 5 mM free Mg2+ was increased ( tr = 1 . 72 s ) compared to 1 mM . This general increase in the residence time implies that lateral interactions ( bundling ) , favored by free Mg2+ , promote larger and more crosslinked polymeric species with higher membrane affinity and less susceptibility to destabilization . In our minimalistic in vitro reconstitution study , we found that polymers of an artificially membrane-targeted variant of FtsZ autonomously and without the presence of FtsA self-organize on a supported bilayer upon addition of GTP and Mg2+ to form chiral ringlike dynamic patterns ( Fig 1 ) , displaying a clockwise or counterclockwise protein movement , dependent on whether the membrane attachment was enforced through the C-terminal or N-terminal end of the protein , respectively ( Fig 4A and 4B ) . The mts in both cases was taken from MinD , one of the elements of the site-selection MinCDE complex , which allows FtsZ to be peripherally attached to the membrane . We thus showed that the ability of FtsZ to create dynamic patterns is an intrinsic property ( Fig 1A ) rather than a specific interaction with a specific protein anchor . Instead , the formation of dynamic FtsZ ring structures in vitro is highly linked to ( i ) the surface protein density and ( ii ) GTPase activity , destabilizing surface-bound filaments and thus being key for treadmilling [5] . We found that the most decisive factor for the emergence of dynamic vortices of FtsZ on membranes is the overall surface coverage by protein monomers and filaments ( Fig 6A ) , which varies over time upon protein adsorption to the membrane ( Fig 2 ) and is controlled by free Mg2+ concentration ( Fig 3 ) . Dynamic vortices appear primarily in an intermediate density regime ( surface mean intensity: 450–1 , 000 arbitrary units [A . U . ] ) and isotropic bundles in a high-density regime ( surface mean intensity > 1 , 000 A . U . ) . Transitions from highly dynamic vortices to isotropic bundles occur upon an increase in lateral contacts that arrest treadmilling filaments , increasing their mean effective length ( as in the case of ZipA [15] ) . This was clearly observed in our single molecule assay ( Fig 5B ) that shows a slower turnover in the situation of dense isotropic bundles , i . e . , longer filaments . Along these lines , the increase in lateral interactions at high free Mg2+ also explains the rapid formation of filaments at 5 mM free Mg2+ ( Fig 3 ) , since larger FtsZ assemblies bind to the membrane and interact with each other more frequently ( Fig 5B ) . Presumably , the main reason why Loose and Mitchison failed to observe dynamic vortices in the case of the FtsZ-YFP-mts [15] is because of the high protein concentration used in their experiments ( 1 . 5 uM ) . Our results are compatible with previous atomic force microscopy analysis of static structures formed by FtsZ polymers on mica as a function of protein concentration at the surface [23] . The protein concentration–dependent formation of dynamic FtsZ patterns also nicely correlates with a recent theoretical study suggesting that protein density at the membrane controls the formation of vortex patterns on membranes in a phase-like behavior . According to this , independent curved polar filaments showing chiral motion and repulsion can self-assemble into vortex or ringlike structures in an intermediate density regime . While at low protein densities filaments travel independently , at the high-density regime they form isotropic networks and jammed bundles [24] . By gradually increasing protein concentration on the membrane , we were able to investigate the initial formation phase of dynamic rings . At low density ( surface mean intensity < 450 A . U . ) , curved and polar filaments initially emerge from nucleation points , which presumably are small attached filaments above a critical length . Intriguingly , the overall adsorption rates to the membrane at this stage are similar for protein concentrations of 0 . 2 μM and 0 . 5 μM ( Fig 2A ) . Upon sufficiently high membrane coverage of nucleating filaments after the initial phase , protein binding from solution begins to scale with total ( i . e . , bulk ) protein concentration . Whether nucleators are directly formed on the membrane after GTP addition or whether short polymers are formed in solution and an increase in affinity with growth brings them to the membrane cannot be determined based on our data and will be the topic of further investigation . We found that filaments growing from nucleators are prone to fragmentation , resulting in free fragments that may stay connected to the membrane . There , they assemble with other attached filaments by diffusion and directional treadmilling , which ultimately results in closed rings in which the treadmilling continues . Our experiments demonstrate that treadmilling , particularly via destabilization of the trailing edge , is highly regulated by GTPase activity . When GTPase activity is switched off ( FtsZ[T108A]-YFP-mts ) , rings seemed to grow only from nucleation points and do not treadmill , at least on time scales found for FtsZ-YFP-mts ( Fig 4C ) . Moreover , the residence time of single FtsZ[T108A]-YFP-mts subunits in the filaments is comparable with the photobleaching control , implying that protein turnover is almost nonexistent ( S6 Fig ) . It has been suggested that a kinetic and structural polarity at monomeric level and a GTP/GDP gradient are requirements for robust treadmilling [25] . As seen from our experiments of the initial vortex growth phase , a GTP/GDP gradient along the treadmilling direction is likely to result from the preferential addition of GTP subunits to the existing filaments at the polar front and a more likely GTP turnover toward the “older” tail . In light of the role of GTPase activity for the formation of dynamic vortices , the measured velocity of the FtsZ + FtsA vortices reported by Loose and Mitchison is about 3-fold faster ( 108 nm/s ) compared to our FtsZ-YFP-mts rings ( 34 nm/s ) . In addition , these authors reported a higher GTPase activity of the FtsZ + FtsA compared to the FtsZ-YFP-mts [15] . Nonetheless , it is not clear how variables such as GTPase activity and attachment strength influence the speed of rotation . For instance , our mutant mts-H-FtsZ-YFP has shown a considerable decrease in rotation speed compared to the FtsZ-YFP-mts . Remarkably , to observe mts-H-FtsZ-YFP dynamic rings , we had to increase the bulk concentration to 1 . 25 uM . This may be due to a reduced affinity for membranes , affecting the overall dynamics . The most remarkable outcome of this study is the clear dependence of vortex chirality on the positioning of the membrane anchor , which , in turn , has severe effects on the topology of membrane deformation by FtsZ . Chirality is inverted by switching the membrane anchor from the C-terminus ( clockwise ) to the N-terminus ( counterclockwise ) . Intriguingly , these two different mutants cause concave ( C-terminal ) or convex ( N-terminal ) deformations when bound to deformable liposomes [14] . To explain these different deformations , Erickson and colleagues have previously depicted FtsZ filaments as arc segments with a direction of membrane attachment either parallel or antiparallel to the vector of curvature . In order to support the here-observed chiral treadmilling of curved rings on planar membranes , however , attachment through the preferential binding face of the filaments needs to have a perpendicular component to both the ring curvature and filament polarity . Fig 6B shows a curved filament with a C-terminal ( clear gray ) and N-terminal ( dark gray ) face perpendicular to the curvature of the filament . Note that the mts is represented here with one parallel component to the curvature , as suggested by Erickson , and one perpendicular component to accommodate flat membrane binding . In this flat representation of a curved FtsZ filament , treadmilling is explained by a polar growth at the leading edge and a destabilization mechanism , driven by GTP hydrolysis , toward the GDP-enriched region at the trailing edge [25] . Nevertheless , due to the fact that curved structures can be either attached along their axis of apparent curvature ( in vivo , Osawa and Erickson [14] ) or perpendicular to their axis of primary curvature ( as reported here ) , we have to conclude that either the membrane attachment of the filament is immensely flexible or—and this is more likely , based on previous structural investigations—that the filament does not have a single but rather more than one direction of curvature , like a helix or a twisted arc [19 , 26] . Indeed , a very similar geometry with more than one curvature direction has recently been reported for endosomal sorting complex required for transport III ( ESCRTIII ) filaments ( nicely reviewed by Chiaruttini and Roux [27] ) . This is particularly intriguing as , among many other roles in eukaryotic and prokaryotic cells connected with membrane abscission , ESCRT is the alternative system to FtsZ with respect to cell division in Archaea [28 , 29] . Therefore , in light of evidence showing that FtsZ forms helical structures in vivo [30–33] and in vitro [19] , we here suggest an alternative structural model to the one depicted in 6B ( Fig 6C ) . We propose that an FtsZ filament with more than one main direction of curvature , such as a helix , would much more elegantly accommodate the combination of inward/outward deformations and chiral treadmilling for the opposite mts mutants . Such a corkscrew-like FtsZ filament can be simply described by an intrinsic radius and a pitch , in which the latter would reflect on the attachment direction ( Fig 6C ) . On a deformable surface ( deflated liposome ) , the growing filament would either pull up ( left ) or push down ( right ) the surface due to the respective pitch . The interplay between the elastic response of the membrane ( increased membrane tension ) and local changes in the helix radius due to GTP hydrolysis [34 , 35] could explain the stabilization of higher curvature or smaller radii regions ( Fig 6C ) . On a nondeformable surface ( SLB ) , however , since the surface is not resilient , the filament would experience a strain , get destabilized , and eventually break upon growth . Thus , together with the very recent studies showing the linkage between treadmilling of FtsZ polymers and peptidoglycan synthesis in E . coli [4] and B . subtilis [5] cells , our findings shed new light on the interplay between FtsZ structure and treadmilling dynamics but may also hint to a direct mechanical link of these to bacterial division . The minimal system we used unambiguously shows that the observed chiral vortices are the result of intrinsic GTP-linked FtsZ polymerization dynamics on the membrane without the need of additional complex interactions with FtsA and ATP , pointing to a fascinating archetypal feature of this important structural protein . The reduced number of components allowed us to selectively determine the influence of key factors—e . g . , the surface density of FtsZ—on the self-organization behavior , thus contributing to a much better mechanistic understanding of FtsZ’s dynamic architecture and its potential physiological implications . Mutations in ftsZ-YFP-mts were constructed using site-directed mutagenesis . FW oligonucleotide was designed using the NEBaseChanger–Substitution ( http://nebasechanger . neb . com/ ) . RV oligonucleotide was obtained from the reverser complement sequence of the FW oligonucleotide . Oligonucleotides 5’- GGTGGTGGTGCCGGTACAGGT-3’ and 5’- ACCTGTACCGGCACCACCACC-3’ were used to generate FtsZ-YFP-mts-T108A mutant by replacing a Thr in position 108 by an Ala . Briefly , ftsZ-YFP-mts was first amplified using the FW and RV oligonucleotides in different PCR reactions , testing 3 different temperatures: 54°C , 58 . 5°C , and 65°C . In a second PCR reaction , the PCR products from the FW and RV oligonucleotides were mixed; also , 3 different temperatures were tested: 54°C , 58 . 5°C , and 65°C . After digestion with DpnI , the 3 PCR products were used to transform CH3-Blue competent cells . Five colonies were picked for DNA extraction and selected for sequencing . The FtsZ-YFP-mts , FtsZ[T108A]-YFP-mts , and mts-H-FtsZ-YFP were purified as previously described [36] . Briefly , the protein was expressed from a pET-11b expression vector and transformed into E . coli strain BL21 . ON overexpression was performed at 20°C for the proteins FtsZ-YFP-mts and FtsZ[T108A]-YFP-mts and at 37°C for protein mts-H-FtsZ-YFP . Cells were lysed by sonication and separated by centrifugation . Then , protein was precipitated from the supernatant , adding 30% ammonium sulphate and incubating the mixture for 20 min on ice ( slow shaking ) . After centrifugation and resuspension of the pellet , the protein was purified by anion exchange chromatography using a 5× 5 ml Hi-Trap Q-Sepharose column ( GE Healthcare , 17515601 ) . Purity of the protein was confirmed by SDS-PAGE and mass spectrometry . E . coli polar lipid extract ( Avanti , AL , United States ) , initially dissolved in chloroform , was desiccated under a gas nitrogen stream . Chloroform traces were further removed by 1-h vacuum . This lipid film was hydrated with SLB-buffer ( 50 mM Tris-HCl , 150 mM KCl , pH 7 . 5 ) to reach a lipid concentration of 4 mg ml−1 . After 10 min sonication , SUVs were obtained . Glass coverslips #1 . 5 ( Menzel , Germany ) were cleaned in air plasma . Then , a plastic chamber was attached on a cleaned glass coverslip using ultraviolet-curable glue ( Norland Optical Adhesive 63 ) . SLBs were obtained by the method of vesicle fusion from SUVs on a glass surface as described elsewhere [37] . The SUV dispersion was diluted in SLB buffer ( 50 mM Tris-HCl at pH 7 . 5 , 150 mM KCl ) to 0 . 5 mg ml−1 , of which 75μl was added to the reaction chamber . Adding CaCl2 to a final concentration of 3 mM promoted vesicle fusion and the formation of a lipid bilayer on the glass . The samples were incubated at 38°C for 20 min and then washed with prewarmed SLB buffer to remove nonfused vesicles . Lastly , a washing step with the reaction buffer ( 50 mM Tris-HCl at pH 7 . 5 , 150 mM KCl and MgCl2 [5 mM or 1 mM] ) was carried out in the sample . FtsZ-YFP-mts or FtsZ-YFP-mts mutants were added to the reaction buffer above the supported lipid membrane in the chamber . The final volume of the samples was approximately 200μl . FtsZ-YFP-mts was added to a final concentration of 0 . 5 μM or 0 . 2 μM . Polymerization was induced by adding GTP to a final concentration between 0 . 04 and 4 mM , as indicated in the text . All experiments were performed on a WF1 GE DeltaVision Elite TIRFM ( GE Healthcare Life Sciences , Germany ) equipped with an OLYMPUS 100× TIRF objective ( NA 1 . 49 ) . The UltimateFocus feature of DeltaVision Elite maintains the focus plane constant in time . FtsZ-YFP-mts was excited with a 488 nm diode laser ( 10 mW , before objective ) Fluorescence imaging is performed using a standard FITC filter set . Images were acquired with a PCO sCMOS 5 . 5 camera ( PCO , Germany ) controlled by the softWoRx Software ( GE Healthcare Life Sciences , Germany ) . For time-lapse experiments , images were acquired every 3 or 10 s , with a 0 . 05 s exposure time , with light illumination shuttered between acquisitions . Image analysis was carried out in MATLAB 2015 ( MATLAB and Image Processing and Computer Vision Toolbox Release 2015a , The MathWorks , Inc . , Natick , Massachusetts , USA ) and processing with Fiji/ImageJ ( Rasband , W . S . , ImageJ , US National Institutes of Health , Bethesda , http://rsb . info . nih . gov/ij/ , 1997–2007 ) . Images correspond to an average of 5–10 frames from a time-series experiment . For the kymograph analysis , time-series acquisitions were filtered using a standard mean filter and were drift corrected ( Image J ) . A MATLAB script allows the user to define a ring by providing 2 coordinates . Every ring is automatically fitted to a circle with radius r . Then , 3 trajectories corresponding to 3 concentric circles having radii r , r+1 , and r−1 pixels are determined . At this point , the script will read the time-series data and calculate a kymograph for each time point and trajectory . The final kymograph corresponds to the average of the 3 different trajectories . To automatically calculate the slope , we first smooth the kymograph with a Savitzky-Golay filter of order 2 and enhance its contrast using a contrast-limited-adaptive-histogram-equalization ( CLAHE ) routine ( MATLAB ) . Next , using Fourier analysis , we find the characteristic frequency for the patterns on the kymograph ( S4 Fig ) . Finally , the slope corresponds to the change in phase at this frequency . Quality criteria are properly chosen to reject low-quality regions over the kymograph . To synchronize time-lapse acquisitions , the initial frame ( time 0 ) was defined when surface mean intensity was around 200 A . U . FtsZ-YFP-mts previously incubated 1:1 with the nanobody GFP-Booster Atto647N ( ChromoTek , Germany ) for at least 1 h at 4°C under agitation . To filter out nonbound nanobody , we centrifuge our protein in a 30 KDa Amicon unit . The GFP-Booster Atto647N was excited with a 640 nm diode laser ( 30 mW , before objective ) . Single molecule imaging was performed using a standard Cy5 filter set . After 10 min of GTP addition , a significant number of spots in the single molecule channel ( Atto647N ) were observed and imaged at a rate of 1 fps or 3 fps with 0 . 3 s exposure time . To improve imaging conditions , we added 10 nM protocatechuate-dioxygenase ( PCD ) and 2 mM 3 , 4-protocatechuicacid ( PCA ) as an oxygen-scavenging system . To determine the position of every single molecule and calculate its residence time , we employed a MATLAB routine designed by Weimann and Ganzinger [38] . Briefly , a bandpass filter was used to remove low- and high-frequency noise . Then , single molecule positions with intensity above a user-defined threshold were determined by their brightness-weighted centroid . The detection algorithm is highly efficient for detecting particles with a signal-to-noise ratio above 1 . 5 . The user-defined threshold was chosen to detect the largest number of spots and kept constant for all experiments . Single molecules were tracked among consecutive frames in an area given by a radius of 10 pixels ( pixel size = 0 . 042 μm ) . Thus , the residence time is defined as the time that the particle stays in this area before its signal vanishes . To calculate the mean residence time , we calculated the probability as a function of time t to obtain a loss of signal event at times ≤ t ( cumulative probability , S6A Fig ) . We fitted to a double exponential function Ae-kt + Be-kpt , in which k refers to the inverse of the mean residence time , and kp corresponds to the photobleaching rate . A and B are constrained , since the photobleaching contribution is limited to be between 0 . 2–0 . 25 in the fitting routine for all conditions ( MATLAB ) . The photobleaching rate was calculated as kp = 0 . 031s-1 using a single exponential fit shown in S6B Fig . Events shorter than 2 frames are below the accuracy of our method and were not included in the statistics . The cumulative probability was measured for 5 different experiments having a total number of events ( N ) in each experiment . For GDP forms: 5 mM Mg2+ , N varies in the range of 3 , 000–5 , 300 events ( 1 fps ) and 660–8 , 800 events ( 3 fps ) ; 1 mM Mg2+ , N = 3 , 000–11 , 000 ( 3 fps ) . For GTP forms: 0 . 04 mM GTP , N = 1 , 300–3 , 000 ( 1 fps ) ; and 4 mM GTP , N = 1 , 200–6 , 700 . In the case of FtsZ*[T108A]-YFP-mts at 4 mM GTP , N = 180–800 .	FtsZ is a tubulin homologue and the primary protein of the bacterial Z ring that guides cell division . In vivo , but also in reconstituted systems , FtsZ shows an intriguing treadmilling dynamic along circular tracks of approximately 1 micrometer in diameter . In cells , this treadmilling along the circumference of the division site is suggested to dynamically guide peptidoglycan—and thus new cell wall—synthesis . In vitro , when reconstituted along with its membrane adaptor FtsA on flat supported membranes , FtsZ self-organizes into similarly treadmilling vortices as observed in vivo but with a clear chirality . With the aim of thoroughly investigating these dynamics , revealing the origin of chirality , and potentially relating it to a membrane-transforming ability of FtsZ , we reconstituted different membrane-targeted mutants of FtsZ on flat membranes . In this minimized system , we found that dynamic ring formation is an intrinsic feature of FtsZ without the need of any other protein . However , self-organization into dynamic treadmilling only occurs within a specific protein , cation , and guanosine triphosphate ( GTP ) concentration range . Our work led us to propose that the observed chirality of FtsZ treadmilling may be explained by an inherent helical character of the filaments with more than one direction of curvature .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "fluorescence", "imaging", "glycosylamines", "classical", "mechanics", "enzymes", "condensed", "matter", "physics", "enzymology", "guanosine", "geometry", "luminescent", "proteins", "mathematics", "yellow", "fluorescent", "protein", "materials", "science", "damage", "mechanics", "macromolecules", "materials", "by", "structure", "research", "and", "analysis", "methods", "polymers", "polymer", "chemistry", "imaging", "techniques", "proteins", "deformation", "guanosine", "triphosphatase", "chemistry", "physics", "biochemistry", "hydrolases", "curvature", "nucleosides", "biology", "and", "life", "sciences", "physical", "sciences", "glycobiology", "nucleation" ]	2018	Treadmilling analysis reveals new insights into dynamic FtsZ ring architecture
The reovirus fusion-associated small transmembrane ( FAST ) proteins function as virus-encoded cellular fusogens , mediating efficient cell–cell rather than virus–cell membrane fusion . With ectodomains of only ∼20–40 residues , it is unclear how such diminutive viral fusion proteins mediate the initial stages ( i . e . membrane contact and close membrane apposition ) of the fusion reaction that precede actual membrane merger . We now show that the FAST proteins lack specific receptor-binding activity , and in their natural biological context of promoting cell–cell fusion , rely on cadherins to promote close membrane apposition . The FAST proteins , however , are not specifically reliant on cadherin engagement to mediate membrane apposition as indicated by their ability to efficiently utilize other adhesins in the fusion reaction . Results further indicate that surrogate adhesion proteins that bridge membranes as close as 13 nm apart enhance FAST protein-induced cell–cell fusion , but active actin remodelling is required for maximal fusion activity . The FAST proteins are the first example of membrane fusion proteins that have specifically evolved to function as opportunistic fusogens , designed to exploit and convert naturally occurring adhesion sites into fusion sites . The capacity of surrogate , non-cognate adhesins and active actin remodelling to enhance the cell–cell fusion activity of the FAST proteins are features perfectly suited to the structural and functional evolution of these fusogens as the minimal fusion component of a virus-encoded cellular fusion machine . These results also provide a basis for reconciling the rudimentary structure of the FAST proteins with their capacity to fuse cellular membranes . By nature of their route of entry into cells , enveloped viruses possess proteins dedicated to the regulation and execution of membrane fusion between the viral envelope and target cell membrane . A multi-step process defines what may be a universal pathway to membrane fusion , involving membrane contact ( attachment and enforced close apposition ) , lipid mixing ( hemifusion ) , and content mixing ( pore formation and stabilization ) [1]–[3] . Extensive analyses suggest the structural transition of enveloped virus fusion protein complexes from a metastable pre-fusion conformation to a lower energy post-fusion structure provides the energy to drive the multi-step fusion process [4] , [5] . Although details of the protein structural rearrangements that accompany membrane fusion have emerged , the precise relationships between structural interactions within components of the fusion machinery and the different steps in the fusion reaction remain unclear . In spite of considerable diversity in the architecture of the enveloped virus fusion protein complexes , recent studies reveal a remarkable conservation in the relationships between structural remodelling of these protein complexes and the process of membrane merger [6]–[9] . The emerging paradigm predicts that triggered rearrangements in the fusion protein complex result in exposure and membrane insertion of a fusion peptide , followed by folding back of the extended structure and hairpin formation that presumably drives membrane apposition and merger [10] , [11] . A second unifying principle is that the enveloped viruses use protein complexes of varying complexity that function autonomously to co-ordinately regulate progression through all stages of the multi-step fusion process . In the simplest situation , the flaviviruses and rhabdoviruses use multiple copies of a single trimeric glycoprotein for the entire fusion reaction [12] , [13] . In all other viruses , the activities responsible for membrane attachment and membrane fusion are segregated into different polypeptides or separate multimeric proteins that nonetheless function together as cognate components of an autonomous membrane fusion machine . For example , the ortho- , retro- , filo- , and coronaviruses assign the initial membrane contact and latter membrane merger stages of the process to separate polypeptide subunits within a homotrimeric protein complex [14]–[17] . A slightly different situation occurs in the alphaviruses , where the receptor binding E2 glycoprotein initially forms a heterodimer with the E1 membrane fusion polypeptide; a low pH trigger and insertion of the fusion peptide into target membranes converts E1 to a functional , homotrimeric fusion protein [18] . Even the herpesviruses and the majority of the paramyxoviruses , that utilize separate multimeric proteins for the membrane contact and membrane merger steps of fusion reaction , couple membrane binding to membrane fusion via transient lateral interactions that are believed to be involved in triggering the structural transition of the fusion protein [8] , [19] . Coordinating the membrane attachment and membrane fusion stages of the fusion reaction using an autonomous fusion machine reflects the need for enveloped viruses to spatially and temporally regulate fusion of the virus envelope with a suitable target cell membrane . The concept that all viral fusion proteins function as components of autonomous , metastable fusion machines utilizing extensive structural rearrangements to drive membrane fusion is challenged by the reovirus fusion-associated small transmembrane ( FAST ) proteins . The FAST proteins are an unusual family of membrane fusion proteins encoded by the fusogenic orthoreoviruses , a diverse group of nonenveloped viruses [20] . At only 95–140 residues in size , the FAST proteins are the smallest known proteins capable of inducing biological membrane fusion . Unlike enveloped virus fusion proteins , the FAST proteins are nonstructural viral proteins and therefore not involved in virus entry into cells [21] . Following their expression inside virus-infected or transfected cells , the FAST proteins traffic to the plasma membrane where their sole defined function is to disseminate the virus infection by mediating membrane fusion with adjacent uninfected cells [22] . The unusual biological role of the FAST proteins is reflected in their unique structural features . There are currently three members of the FAST protein family named according to their approximate molecular masses ( p10 , p14 and p15 ) , all of which are single-pass transmembrane proteins that assume an Nexoplasmic/Ccytoplasmic membrane topology [20] . Unlike all other fusion machines whose topology orients the majority of the protein on the proximal ( i . e . contacting ) side of the membrane , as much or more of the mass of the FAST proteins is localized on the distal side of the membrane resulting in small N-terminal ectodomains of only ∼20–40 residues [21] , [23] , [24] . The FAST proteins also exhibit considerable diversity in their repertoires and arrangement of structural motifs . For example , the p14 and p15 ectodomains contain an essential N-terminal myristate moiety that is lacking in p10 , while the p10 and p14 ectodomains have hydrophobic patches that share some similarity to the fusion peptide loops found in class II enveloped virus fusion proteins [25]–[27]; p15 lacks this motif in its ectodomain but has a similar motif in its endodomain [24] . The surprising structural features and diversity of the FAST proteins have not been reconciled with existing models of protein-mediated membrane fusion . The purified p14 FAST protein , when reconstituted into liposome membranes , mediates liposome-cell fusion and liposome-liposome lipid-mixing [28] , suggesting the FAST proteins , like the enveloped virus fusion protein complexes , can function as autonomous membrane fusion machines . The FAST protein ectodomains , however , lack the structural complexity typical of most viral fusion proteins and appear to be incapable of using hairpin formation to drive membrane apposition and fusion . If , and how , the FAST proteins mediate the earliest stages of the fusion reaction ( i . e . membrane attachment and close apposition ) is therefore not apparent . We now show that although the FAST proteins have retained within their rudimentary structures the activities required to mediate the actual merger of closely apposed membranes , they rely on non-cognate , surrogate receptor-binding proteins to mediate membrane attachment for enhanced cell–cell fusion activity . Furthermore , maximal cell–cell fusion activity requires active actin remodelling . The use of surrogate , non-viral adhesion factors is perfectly suited to the evolution of the FAST proteins as virus-encoded cellular fusogens , and has important implications on mechanistic models of FAST protein-mediated membrane fusion reaction . Based on the capacity of the purified p14 FAST protein to induce liposome-cell and liposome-liposome fusion [28] , we speculated the FAST proteins are responsible for all stages of the membrane fusion reaction , including the earliest stage of membrane attachment . To directly test this hypothesis , quantitative liposome-cell binding assays were performed . Surprisingly , titration analysis of p14-liposome binding to target cells indicated low-level , non-saturable adherence of p14-liposomes at lipid concentrations exceeding 1 . 6 mM ( Fig . 1 ) . As previously reported [28] , p14-liposomes adhered better to target cells than liposomes lacking the p14 FAST protein . However , the percent binding efficiency of p14-liposomes ranged from only 1 . 9–3 . 3% of the input liposomes . This low-affinity adherence did not reflect a preponderance of liposomes lacking p14 , since flow cytometry revealed the majority of liposomes contained p14 , with an average protein density of 6–7×103 p14 molecules per 400 nm liposome [28] . Furthermore , there was no evidence of receptor saturation contributing to the low binding efficiency , as indicated by the progressive increase in bound liposomes with increasing doses of input liposomes ( Fig . 1 ) . We conclude that p14 lacks specific high-affinity receptor-binding activity . Therefore , while a low level of non-specific adherence is sufficient to facilitate at least some FAST protein-mediated liposome-cell fusion , these results raised the question of whether surrogate adhesion factors might be essential for , or could enhance , the normal biological functioning of the FAST proteins as cell–cell fusogens . Since the FAST proteins function as virus-encoded cellular fusogens , we reasoned they could have evolved to specifically rely on cell adhesion proteins to mediate the earliest steps in the fusion reaction . In view of the implications of a multi-component fusion complex on mechanistic models of FAST protein function , we therefore sought to define the relationship , if any , between the FAST proteins and cellular adhesion factors . Cadherins presented as a likely candidate to provide an adhesive activity that could influence FAST protein-mediated cell–cell fusion , since this ubiquitous family of membrane glycoproteins is involved in the formation of virtually all types of homotypic cell–cell adhesion [29] . The broad distribution of cadherin junctions is compatible with the promiscuous cell–cell fusion activity of the FAST proteins . To initially explore whether cadherins might be involved in the FAST protein-mediated cell–cell fusion reaction , p14-transfected cells were briefly cultured under low calcium conditions prior to , and during , syncytiogenesis . Calcium depletion disrupts homotypic cadherin interactions , resulting in rapid dissociation of cadherin-dependent adhesions without affecting integrin-mediated cell attachment to the substratum [30] . As qualitatively observed by light microscopy of Giemsa-stained monolayers ( Fig . 2A ) , and as quantified using a standard syncytial assay [23] ( Fig . 2B ) , low calcium conditions inhibited p14-induced syncytium formation by 75∼90% in both fibroblast ( QM5 and HT-1080 ) and epithelial ( MDCK ) cells . The ability of two other members of the FAST protein family , p10 and p15 , to fuse QM5 fibroblasts was similarly inhibited by ∼80–90% under low calcium conditions ( Fig . 2B ) . Most importantly , the fusion activity of the vesicular stomatitis virus ( VSV ) G and influenza HA viral fusion proteins was unaffected by similar low calcium conditions ( Fig . 2B ) ; since these enveloped virus fusion proteins have their own receptor-binding activity , their cell–cell fusion activity should be independent of calcium-mediated cadherin contacts . The inhibition of FAST protein-mediated syncytiogenesis under low calcium conditions was therefore unlikely to be due to a generalized inhibitory effect of calcium depletion on cell–cell fusion . To determine whether the actual FAST protein-induced membrane fusion reaction , not just syncytiogenesis , was also calcium-dependent , a quantitative pore formation assay [31] was adapted to assess the p14-induced fusion reaction . Two independent populations of cells labelled with either green fluorescent protein or calcein red-orange were seeded together , transfected with a p14 expression plasmid , co-cultured in the absence or presence of calcium to allow cell–cell fusion to proceed , then trypsin-treated to generate a single cell suspension . The number of co-fluorescent cells was then quantified by flow cytometry . Under normal calcium conditions , p14-induced membrane fusion was easily detectible by the increase in the number of co-fluorescent cells relative to the low background level of co-fluorescent cells observed in vector-transfected cells ( Fig . 3A ) . Under low calcium conditions , the transfer of the soluble fluorescent markers in p14-transfected cell monolayers was reduced to near background levels . Quantifying the numbers of co-fluorescent cells in the dot plots from p14-transfected cells indicated that low calcium conditions inhibited the pore formation/expansion stage of membrane fusion by ∼80% ( Fig . 3B ) . Identical results were obtained by quantifying the extent of the increase in green fluorescence of the gated red cells by Overton subtractions ( Fig . 3B ) . These results were highly reproducible over four independent experiments , confirming that both FAST protein-induced syncytium formation and the actual membrane fusion reaction itself are calcium-dependent . If the FAST proteins are relying on cadherins to generate fusion sites , then p14 in the plasma membrane should co-localize at sites of cadherin-mediated adhesion . Immunofluorescence microscopy of transfected QM5 cells revealed the obvious concentration of N-cadherin at sites of cell–cell contact ( Fig . 4A ) . In contrast , p14 was broadly distributed in transfected cells and on the cell surface , with extensive regions of p14 staining that did not overlap with cadherins , suggesting p14 does not specifically localize with cadherins . This conclusion was supported by previous radioimmunoprecipitation studies that did not reveal stable p14-cadherin interactions [23] , [25] , [32] . There was , however , clear overlap of a percentage of p14 near regions of intense N-cadherin staining ( Fig . 4A ) . Therefore , while p14 does not appear to specifically co-localize with cadherins , the apparently stochastic localization of p14 at sites of cadherin junctions would allow p14 to exploit these junctions for cell–cell fusion . We also noted a correlation between cadherin status in different cell types and the extent of p14-mediated syncytium formation; this correlation did not apply to the VSV G protein , whose fusion activity is independent of cadherin interactions ( Fig . 2B ) . The VSV G protein is a low pH-activated viral fusion protein that can gradually induce cell–cell fusion when transiently over-expressed in transfected cells in the absence of a triggering acid treatment , due presumably to either gradual acidification of the medium or pH activation of the G protein in the exocytic pathway [33] . Under these conditions , which were chosen since they closely mirror the progressive , untriggered cell–cell fusion mediated by the FAST proteins , VSV G induced equivalent levels of cell–cell fusion by 24 h post-transfection in both cadherin-containing QM5 cells and in cadherin-deficient L cells , as shown qualitatively in Giemsa-stained monolayers ( Fig . 5A , panels c and d ) and quantitatively by counting syncytial nuclei per field ( Fig . 5B ) . In contrast , p14-induced syncytiogenesis was dramatically different in these two cell types . At 8 h post-transfection , p14 induced extensive syncytium in QM5 cells , as shown qualitatively ( Fig . 5A , panel a ) and quantitatively ( Fig . 5B ) . A similar situation applied to the cadherin-containing HT1080 cells , which induced extensive syncytium formation by 9 h post-transfection ( Fig . 4B ) . There was no evidence of cell–cell fusion in the p14-transfected L cells at this early timepoint ( Fig . 5B ) , but cell–cell fusion did eventually occur in the L cells , becoming detectible by 17–20 h post-transfection . Even by 24 h post-transfection , however , p14-induced syncytiogenesis in the L cells ( Fig . 5A , panel b ) was still only ∼20% of that obtained in QM5 cells at 8 h post-transfection . Therefore , although the cadherin-deficient L cells did support p14-induced syncytiogenesis , cell–cell fusion in the L cells was markedly reduced in both the rate and extent of syncytium formation compared to the cadherin-containing QM5 fibroblasts . Since numerous differences aside from cadherin expression could influence p14-induced syncytiogenesis in QM5 and L cells , two complementary approaches were pursued to more directly examine the influence of cadherins on FAST protein-mediated cell–cell fusion . First , siRNAs were used to silence N-cadherin expression in HT-1080 human fibroblast cells . N-cadherin expression was decreased by >70% in cells transfected with siRNAs directed against human N-cadherin relative to cells transfected with control siRNAs ( Fig . 4B ) . Decreased cadherin expression coincided with a ∼75% decrease in p14-mediated syncytiogenesis , effectively reducing the level of cell–cell fusion to that observed when low extracellular calcium levels were used to disrupt cadherin interactions in control siRNA-transfected HT-1080 cells ( Fig . 4B ) . Second , if reducing cadherin interactions inhibits FAST protein-mediated syncytiogenesis , then would increasing cadherin contacts have the opposite effect ? To examine this question , p14-induced syncytium formation was examined in the cadherin-deficient L cells and in EL cells , which are L cells stably expressing E-cadherin [34] . Introduction of E-cadherin into L cell fibroblasts resulted in both a noticeable increase in cell–cell contact ( Fig . 6A ) , and a reproducible 30–40% increase in p14-induced syncytium formation ( Fig . 6B ) . Furthermore , this increase in cell–cell fusion was eliminated when EL cells were cultured under low calcium conditions , suggesting it was directly due to trans-cadherin interactions . The results obtained in the L cells and EL cells , in conjunction with the decrease in FAST protein-induced cell–cell fusion following siRNA knockdown of cadherin expression in HT-1080 cells , clearly indicated that in the absence of their own receptor binding capability , the FAST proteins can exploit cadherin junctions to provide the initial membrane attachment stage of the fusion reaction . However , some level of p14-induced cell–cell fusion persisted under low calcium conditions in fibroblast and epithelial cells ( Fig . 2B ) , in cadherin siRNA-depleted HT1080 cells ( Fig . 4B ) , in L-cells devoid of cadherin ( Fig . 5A ) , and in fusion of p14-liposomes to target cells [28] , suggesting cadherins enhance , but are not required , for FAST protein-induced syncytiogenesis . The apparent lack of specific interactions between p14 and cadherins , coupled with the enhancing though non-essential role of cadherins in the cell–cell fusion reaction mediated by the FAST proteins , suggested cadherins do not represent the cognate membrane attachment component of a bipartite FAST protein fusion complex . Rather , we predicted the FAST proteins evolved as independent membrane fusion proteins that seconded the close membrane apposition stage of the fusion reaction to surrogate , non-cognate adhesins . To test this hypothesis , we examined whether adhesion factors other than cadherins could exert a similar stimulatory effect on the cell–cell fusion activity of the FAST proteins . The uncleaved precursor of the influenza virus HA fusion protein , HAO , is fusion-inactive but retains its ability to bind sialic acid . Furthermore , the fusion activity of the cleaved HA protein was unaffected by disrupting cadherin interactions ( Fig . 2B ) . HAO-sialic acid interactions could therefore conceivably substitute for cadherin-mediated cell–cell adhesion under low calcium conditions to enhance FAST protein-induced cell–cell fusion . To explore this possibility , we analyzed p14-induced syncytiogenesis in QM5 cells stably expressing the fusion-inactive influenza HA0 protein . As shown ( Fig . 7A ) , the presence of HAO resulted in a substantial increase in the cell–cell fusion activity of p14 under calcium conditions that disrupt cadherin interactions . The stimulatory effect of HAO on p14-induced cell–cell fusion was ablated using non-immune horse serum ( Fig . 7A ) , which contains α2-macroglobulin and other components that inhibit HAO binding to its sialic acid receptor [35] , supporting the conclusion that HAO-sialic acid interactions can effectively substitute for cadherin interactions to enhance the p14-induced cell–cell fusion reaction . QM5-HA cells transfected with a non-fusogenic mutant of p14 , p14-G2A [23] , or with vector alone exhibited no syncytiogenesis ( Fig . 7A ) , confirming that HA0 was not contributing to the fusion reaction beyond providing cell–cell adhesion via its sialic acid binding activity . Furthermore , surface immunofluorescence microscopy revealed that a proportion of p14 co-localized with HAO , including at sites of cell–cell contact , and this colocalization was maintained under low calcium conditions ( Fig . 7B ) . These results supported the hypothesis that the FAST proteins have evolved to retain the minimal activity required to bring about fusion of closely apposed membranes , and rely on non-cognate adhesins to mediate the initial membrane contact phase of the fusion reaction . Several features of the surrogate adhesion results suggested that additional factors might be contributing to the syncytiogenic activity of the FAST proteins . First , the enhancing effect of HAO on p14-induced syncytiogenesis did not fully compensate for the loss of cadherin interactions under low calcium conditions ( Fig . 7A ) . Second , although the addition of cadherins increased the susceptibility of L cells to p14-induced fusion , syncytiogenesis in the EL cells was still substantially less than that obtained in QM5 cells . While several explanations could explain these anomalies ( e . g . differences in the surface expression of cadherins , FAST proteins , and/or HAO in the different cell lines ) , one possibility was intracellular events that accompany cadherin interactions . The formation of trans-cadherin complexes triggers a cascade of downstream events that convert the weak , focal interactions mediated by individual cadherin pairs to stronger , more extended regions of adhesive contact , a process referred to as “active” adhesion that is intimately dependent on active actin remodelling [36] , [37] . During active adhesion , as occurs in MDCK cells ( Fig . 8 , panels a and b ) , cadherins and F-actin concentrate at extended regions of close cell–cell contact [38] . In contrast , F-actin was not concentrated at sites of cell–cell contact in cadherin-deficient L cells , forming instead extensive networks of actin fibres ( Fig . 8 , panels c and d ) , and these cells did not form extended adhesion contacts , a phenotype we refer to as “no adhesion” . Interestingly , while ectopic expression of E-cadherin in EL cells did result in regions of focal cell–cell contact ( Fig . 8 , panel e ) , these focal cadherin junctions did not develop into extended regions of close cell–cell contact . Moreover , the actin cytoskeleton in EL cells retained the architecture observed in L cells , forming an extensive array of stress fibres throughout the cells with little indication of F-actin concentration at sites of cell contact ( Fig . 8 , panels e and f ) . This phenotype , where cadherins are engaged but actin is not remodelled to form extended adhesive junctions , has been called “passive adhesion” [37] . There was therefore a correlation between the adhesion properties of the different cell types and their susceptibility to FAST protein-induced cell–cell fusion . To more clearly assess the relationship between active adhesion and syncytium formation induced by the FAST proteins , we sought to generate the three adhesion phenotypes within a single cell type , thereby avoiding potential complications due to possible cadherin-independent differences in different cell lines . Previous studies report that a combination of low calcium conditions and the actin depolymerising agent cytochalasin D ( cytoD ) can be used to generate the active , passive and no adhesion phenotypes [39] . Following disruption of cadherin contacts by calcium-depletion , reversion to normal calcium conditions in the presence of cytoD allows cadherin engagement but prevents actin polymerization and the formation of the extended regions of cell–cell contact characteristic of active adhesion . A similar procedure was followed to generate these different types of adhesion within the fusion permissive QM5 cells . We first determined the concentration of cytoD that would interfere with actin polymerization in QM5 cells and the formation of new extended regions of cell–cell contact while having minimal inhibitory effects on p14-induced syncytium formation . Low doses of cytoD ( 0 . 1–0 . 5 µg/ml ) resulted in the partial redistribution of filamentous actin into cytoplasmic actin aggregates , with minimal effects on stable cadherin junctions ( Fig . 9A , panel b ) and p14-induced cell–cell fusion ( Fig . 9B ) . The observed 20–30% decrease in syncytium formation in cells treated with low doses of cytoD presumably reflected the previously reported ability of cytoD to disrupt recently formed , unstable cell–cell contacts and prevent the formation of new regions of extended intercellular junctions [38] . The slightly altered distribution in cell surface fluorescence of p14 following cytoD treatment ( Fig . 9C ) might also have contributed to the modest decline in cell–cell fusion . Increased concentrations of cytoD ( above 1 µg/ml ) resulted in extensive disruption of the actin cytoskeleton and the formation of cytoplasmic actin aggregates ( Fig . 9A , panel c ) , and inhibited p14-induced syncytium formation by >80% ( Fig . 9B ) . We therefore chose 0 . 1 µg/ml of cytoD to inhibit the formation of extended cell–cell contacts following calcium depletion and repletion in order to generate the passive adhesion phenotype . Transfected QM5 cells were treated with calcium-free medium for 30 minutes to disrupt cadherin complexes just prior to the onset of p14-induced syncytium formation , and then incubated for a 2–4 h under three different culture conditions to allow cell–cell fusion to progress . First , continued incubation under low calcium conditions to maintain disruption of cadherin-dependent cellular contacts generated the no adhesion phenotype , as evidenced by the loss of extended regions of cadherin-mediated cell–cell contact ( Fig . 10A ) . As previously shown ( Fig . 2B ) , low calcium conditions that generated the no adhesion phenotype inhibited p14-induced cell–cell fusion by ∼80% ( Fig . 10D ) . Second , incubating cells previously cultured in the absence of calcium under normal calcium conditions resulted in the rapid restoration of active adhesion , with cadherin interactions mediating the formation of extended adhesion sites containing both actin and cadherins ( Fig . 10B ) . Syncytiogenesis in these cells was fully restored to the levels observed in cells that were never incubated under low calcium conditions ( Fig . 10D ) . Third , performing the calcium switch in the presence of low concentrations ( 0 . 1 µg/ml ) of cytoD allowed cadherin contacts to reform but inhibited complete actin remodelling . Under these treatment conditions , cells formed punctate intercellular cadherin contacts ( arrows in Fig . 10C ) , but F-actin was not extensively co-localized in these adhesions and extended cellular junctions did not form , indicative of passive adhesion . This transition of cells from the no-adhesion to passive-adhesion phenotype only partially restored p14-induced syncytium formation to ∼50% of the maximal level ( i . e . that observed in cells that were not cultured under low calcium conditions prior to treatment with cytoD ) . This level of cell–cell fusion closely paralleled the relative fusion efficiency in the QM5-HAO cells under low versus normal calcium conditions ( Fig . 7A ) , conditions that mimic the passive versus active adhesion phenotypes generated by the calcium-switch experiments . Therefore , cells forming active adhesions consistently supported p14-mediated fusion better than cells forming passive adhesions , which in turn fused more efficiently than cells lacking even focal cadherin contacts . Previous studies revealed that the purified p14 FAST protein is both necessary and sufficient to induce liposome-cell and liposome-liposome fusion [28] , suggesting the FAST proteins are autonomous fusion machines responsible for all stages of the membrane fusion reaction . Our present results , however , necessitate a refinement of this general conclusion for the process of FAST protein-mediated cell–cell fusion . We propose that the reovirus FAST proteins contain within their rudimentary structures all of the activities necessary to efficiently mediate the fusion of closely apposed membranes . However , in their natural biological context as cell–cell fusogens , the FAST proteins exploit cellular adhesion factors and active actin remodelling for maximal membrane fusion activity . This is the first such example of a membrane fusion machine comprised of a viral fusion protein that has specifically evolved to utilize surrogate , non-cognate adhesion factors . Two significant implications emerge from this bipartite model . First , the model provides new insights into the form-fits-function evolution of the FAST proteins as virus-encoded cellular fusogens . Second , the use of surrogate , non-cognate adhesins and active actin remodelling provides a means to rationalize the simple structure of the FAST proteins with their role as cellular fusogens . A prerequisite for all membrane fusion events is the initial tethering of the two membranes to be fused . Enveloped viruses have evolved to utilize a viral adhesin , a component of the viral fusion complex , to mediate cell attachment . Under certain situations , where the viral adhesin is insufficient to provide membrane attachment , surrogate adhesins can provide this activity . For example , the E2 adhesin of the Sindbis virus E1/E2 fusion complex lacks sialic acid binding activity and does not promote virion attachment to red blood cell target membranes . However , co-expression of the influenza virus HAO protein mediates red blood cell attachment to cells expressing E1/E2 and results in efficient cell–cell fusion [3] . There are also instances where , in the absence of suitable receptors on target membranes , the fusion component of the enveloped virus fusion complex can function to promote membrane attachment . This is best exemplified by certain paramyxoviruses whose F protein can function in the absence of the HN attachment component [40] , and by fusion of some enveloped viruses to protein-free target liposomes , where membrane attachment is presumably mediated by low pH-triggered exposure of the fusion peptide and insertion into the target membrane [41]–[43] . The above examples underscore the importance of membrane attachment as a prelude to subsequent membrane fusion , and it was therefore not unanticipated that cell–cell fusion mediated by the FAST proteins would also be reliant on membrane attachment . The surprising observation was the discovery that the FAST proteins lack their own adhesion capacity ( Fig . 1 ) and in their natural biological context as cell–cell fusogens , have specifically evolved to use surrogate adhesion factors . Results obtained by calcium depletion ( Fig . 2 ) , siRNA knockdown ( Fig . 4 ) , and ectopic expression of cadherins in cadherin-deficient cells ( Fig . 6 ) all indicated that cadherins can serve as surrogate adhesins to increase the efficiency of FAST protein-mediated cell–cell fusion . Furthermore , the role of cadherins in enhancing the function of the FAST proteins likely does not reflect a generalized effect of such cellular junctions on cell–cell fusion , as evident by the lack of any adverse effects of disrupting cadherin interactions on syncytiogenesis mediated by two different classes of enveloped virus fusion proteins ( Fig . 2 ) , and by the ability of the VSV G protein to induce syncytium formation equally well in both cadherin-containing and cadherin-deficient cell types ( Fig 5 ) . While cadherins can clearly serve as surrogate adhesins for the FAST proteins , there is no evidence that the FAST proteins specifically interact or co-localize with cadherins , and p14 functions to induce liposome-cell fusion with no requirement for cadherins in the donor membrane [28] . Cadherins are therefore not a cognate component of a supramolecular FAST protein fusion machine . The observation that cells are still susceptible to p14-induced syncytium formation in the absence of cadherin-mediated contacts ( e . g . cadherin-deficient L cells or in cells whose cadherin contacts are disrupted by calcium depletion or siRNA knockdown ) further suggests that other cellular adhesion factors can substitute for cadherins . Nectins , a group of calcium-independent cell adhesion molecules that act upstream of cadherins [44] , exhibit weaker interactions than cadherins but occur over a similar distance ( i . e . ∼20–25 nm ) [45] , [46] , suggesting they could provide opportunities for the FAST proteins to initiate fusion , albeit with decreased efficiency . Furthermore , HAO receptor binding effectively substituted for cadherin contacts , preserving p14-induced syncytiogenesis to an even greater extent than cadherin-mediated passive adhesion ( compare the percent increase in fusion under low versus normal calcium conditions in Figs . 7 and 10 ) . We therefore conclude that the FAST proteins serve as the fusion component of a functionally bipartite fusion machine that is reliant on surrogate , non-cognate adhesion factors to mediate the earliest stages of the fusion reaction . This conclusion further implies that the FAST proteins are not stabilized in a metastable pre-fusion conformation by interactions with their adhesion factors , nor are such interactions involved in triggering the fusion reaction . This is in contrast to the situation with the enveloped virus fusion machines , where spatial relationships between the binding and fusion components frequently influence the folding , stability or triggering of the pre-fusion complex and/or coordinated progression through the fusion reaction [7] , [8] , [10] . In this biphasic model of FAST protein function , the membrane attachment and membrane merger stages represent two distinct , uncoupled phases . The first phase is mediated by cellular adhesins that do not directly interact with the FAST protein fusogens , which have evolved to function as opportunistic fusogens , retaining within their rudimentary structures all that is needed to complete the second phase by converting naturally occurring adhesion sites into fusion sites . In addition to the benefits conferred by surrogate adhesion proteins , active adhesion was required for maximal levels of FAST protein-induced cell–cell fusion . Support for this conclusion derives from the reduced cell–cell fusion observed under three different conditions that generated the passive adhesion phenotype; cadherin engagement in the presence of low concentrations of cytoD to partially inhibit actin remodelling ( Fig . 10 ) , ectopic expression of cadherins in cadherin-deficient L cells , which did not result in cytoskeletal remodelling ( Fig . 8 ) , and the use of HAO as the surrogate adhesin , which does not trigger actin rearrangements ( Fig . 7 ) . In contrast , cadherin-mediated actin remodelling did not contribute to the efficacy of cell–cell fusion mediated by either HA or VSV G; these enveloped virus fusion proteins have evolved to function as autonomous fusion machines and both were unaffected by calcium conditions that disrupt cadherin interactions ( Fig . 2 ) . The VSV G protein was also equally effective at inducing cell–cell fusion in cadherin-containing QM5 cells and cadherin-deficient L cells ( Fig . 5 ) . Although cadherin-mediated actin remodelling was not involved in cell–cell fusion mediated by these enveloped virus fusion proteins , the actin cytoskeleton can affect virus-cell and/or cell–cell fusion when actin dynamics are altered by manipulating the activity of the Rho family GTPases that regulate cytoskeletal structure [47]–[49] . Cytoskeletal remodelling also contributes to extended alignment of the apposing membranes and trafficking of pre-fusion exocytic vesicles to the site of fusion during Drosophila myoblast fusion [50]–[52] . As discussed below , numerous changes in the environment of the two contacting membranes that accompany the transition from passive to active adhesion could exert an influence on the FAST protein fusion reaction . We suggest a model of FAST protein-mediated cell fusion that integrates the unusual structural and functional properties of these fusogens with the enhancing , though non-essential , role of surrogate adhesion factors ( Fig . 11 ) . Initial tethering of the two membranes is mediated by either non-specific adhesion of liposomes to target cells , or in the case of cell–cell fusion by weak nectin interactions , passive cadherin engagement or other surrogate adhesins ( e . g . HAO ) . This membrane attachment stage would provide contact sites with interbilayer distances of ∼13–25 nm ( Fig . 11 , a and b ) [29] , [45] . These distances are considerably larger than the ∼1 . 5 nm distance that p14 projects from the membrane in which it resides , as estimated by atomic force microscopy measurements under aqueous conditions and the NMR structure of the p14 ectodomain [25] , [32] . In the case of the enveloped viruses , the fusion protein itself is believed to be responsible for breaching this intermembrane distance . A proposed unifying principle for viral protein-mediated membrane fusion involves refolding of the fusion protein from its metastable pre-fusion conformation to its hairpin-like , post-fusion minimal energy state , with mechanical energy serving to pull the membranes into close proximity [11] , [53] . Considering the structural limitations of the FAST protein ectodomains , we previously suggested that the FAST proteins are unlikely to adhere to this unifying principle [24] , [25] , [28] . The present results now provide some alternative possibilities as to how close membrane apposition might be achieved . In the case of passive adhesion , stochastic out-of-plane fluctuations of the membrane ( Fig . 11 , d ) or actin-driven membrane oscillations ( Fig . 11 , e ) as the two apposed membranes “probe” each other could transiently reduce the interbilayer separation to the critical repulsive range of <2–3 nm [4] , [54] , allowing the FAST proteins to exert their opportunistic fusogenic activity . Active adhesion and the formation of adherens junctions ( Fig . 11 , c ) would increase the probability that membranes reach this critical distance by strengthening weak trans-cadherin interactions via lateral clustering of cadherins , by extending the surface area of close cell–cell contact , or by leading to the formation of gap junctions ( Fig . 11 , f ) that reduce intermembrane distances to 2–4 nm [38] , [55] , [56] . Together , these effects increase the likelihood that suitably stable adhesion sites would exist in close proximity to regions of the plasma membrane containing adequate quantities of the FAST protein needed for fusion . The actin remodelling that accompanies active adhesion could also disrupt cortical actin and/or promote displacement of cellular membrane proteins from the fusion site , both of which can inhibit membrane fusion [5] , [57] . While the FAST proteins are unlikely to use hairpin formation to promote close membrane apposition , we do not exclude the possibility that other dynamic structural changes in the FAST proteins could contribute to this process . For example , reversible solvent exposure of hydrophobic residues in the small ectodomain ( e . g . amino acids in the ectodomain hydrophobic patch or the N-terminal myristic acid ) could alter the hydration layer between membranes while residues in the larger endodomain might contribute to actin remodelling and reductions in intermembrane distances . Studies are currently underway to explore these possibilities . This biphasic fusion reaction mediated by a viral fusion protein reliant on surrogate non-viral adhesion factors is perfectly suited to the role of the FAST proteins as viral-encoded cellular fusion proteins . As non-structural viral proteins not involved in virus entry , the FAST proteins are not subject to the same spatial and temporal imperatives that dictate the functioning of enveloped virus fusion proteins . Gradual accumulation of the FAST proteins in the plasma membrane of reovirus infected cells , governed by their protein expression from sub-optimal translation start sites and by protein degradation , is all that is required to coordinate the rate of syncytium formation with the virus replication cycle [26] , 58 . By exploiting generic adhesion molecules and naturally occurring adhesion junctions , the FAST proteins have the capacity to fuse a diversity of cell types , providing the fusogenic reoviruses with access to the replication machinery of multiple cell types during a single round of replication , leading to rapid localized dissemination of the infection [22] . Seconding the membrane attachment phase of the fusion reaction to surrogate adhesins would also reduce the genetic commitment on the part of the virus , no doubt contributing to the evolution of this remarkable group of fusogenic nonenveloped viruses within the confines of the limited coding capacity of the reovirus genome . Vero and QM5 cells were maintained as previously described [23] . MDCK and HT-1080 cells were maintained in minimal essential medium ( MEM ) supplemented with 10% fetal bovine serum ( FBS ) . L cells and EL cells were maintained in MEM supplemented with 5% FBS with EL cells also receiving 500 µg/ml G418 to maintain selective pressure [34] . E- and N-cadherin mAbs were from BD Transduction Labs . Goat anti-rabbit and anti-mouse F ( Ab ) 2 H+L chain Alexa Fluor 488- and 555-conjugated secondary antibodies and phalloidin were from Molecular Probes . Influenza HA ( H1 [WSN] ) in pCAGGS and rabbit anti-HA antiserum were a gift from Dr . Earl Brown ( University of Ottawa ) . The actin-disrupting drug cytoD was from Sigma . The p14 , p15 , p10 , p14-G2A and HA epitope-tagged p14 ( p14-2HAN ) cDNA clones , and the p14 polyclonal and p14 anti-ectodomain antisera were previously described [21] , [25] , [27] , [28] . VSV G protein ( Indiana strain ) in a eukaryotic expression vector was a gift from Dr . Patrick Lee . Fluorescent liposomes and p14-containing proteoliposomes were prepared exactly as previously described [28] . Liposomes ( 0 . 4–1 . 6 mM ) were incubated with QM5 fibroblasts on ice for 1 h , then removed and cells were washed with Hank's balanced salt solution ( HBSS ) . The cells were then resuspended in 10 mM EDTA in phosphate-buffered saline ( PBS ) and bound liposomes were quantified by fluorimetry as previously described [28] . The quantity of lipid molecules bound was calculated using a standard curve comparing fluorescence intensity to phospholipid concentrations . Cadherin-mediated cell–cell contacts were disrupted by washing cells with PBS followed by a 1 min incubation with PBS +0 . 5 mM EDTA , just prior to the onset of syncytiogenesis in the transfected cells . Cells were then washed with PBS and incubated for the duration of the experiment with either MEM or S-MEM ( calcium free MEM , Invitrogen ) supplemented with 10% dialysed FBS . Transfected cells in 12-well cluster plates were fixed with methanol at various times post-transfection based on the extent of syncytium formation in cells incubated under control conditions ( e . g . in normal calcium media ) . Cell–cell fusion was quantified by determining the average number of syncytial nuclei present in five random microscopic fields of Giemsa-stained monolayers , as described previously [23] . Generally , cells were fixed when syncytia in the control wells had progressed to ∼50–250 syncytial nuclei per microscopic field ( 200× magnification ) . This level of cell–cell fusion was determined to give the most accurate and reproducible results . For samples with considerably less than an average of 50 syncytial nuclei per field , 10–20 random fields were counted to enhance accuracy . Results are reported as the percent fusion relative to the indicated control treatment , set at 100% . A population of QM5 cells was labelled with 20 µM calcein red-orange AM ( Molecular Probes ) , mixed 1∶1 with a second population of cells stably expressing EGFP ( Clontech ) and co-cultured overnight , then transfected with either p14 or control empty vector ( pcDNA3 ) . Cells were subjected to calcium depletion at 3 h post-transfection , just prior to the onset of syncytiogenesis , and incubated in either MEM or S-MEM for an additional 3 h . Cells were trypsinized , resuspended in PBS , and analyzed by flow cytometry ( FACSCalibur ( Becton Dickinson ) ) using appropriate filter sets and Cell Quest software . A minimum of 300 , 000 events were recorded , and all data were analyzed using FSC Express 2 . 0 ( De Novo Software ) . HT-1080 cells were transfected with N-cadherin or control siRNA oligonucleotides ( Dharmacon Research Inc . ) at a final concentration of 10 nM using INTERFERin transfection reagent ( Polyplus Transfection ) . At 24 h post-transfection , cells were trypsinized and reseeded , cultured for 20 h , then transfected with p14 cDNA using Lipofectamine ( Invitrogen ) . At 3–4 h post-p14 transfection , some wells were depleted of extracellular calcium to disrupt cadherin-mediated contacts as described below . Cells were fixed with methanol at 9 h post-p14 transfection and fusion was quantified by syncytial indexing as described above . Cell lysates were prepared from a parallel experiment and used for Western blot analysis using anti-N cadherin and anti-actin antibodies , HRP-conjugated secondary antibody , and ECL ( Amersham Biosciences ) according to the manufacturer's instructions , as previously described [32] . Images were captured and quantified using a Typhoon imaging system ( Amersham ) and ImageQuant software ( GE Healthcare ) . Cells grown on gelatin-coated glass coverslips were fixed with 3 . 7% formaldehyde ( 20 min ) and permeabilized with 0 . 1% Triton X-100 in PBS ( 20 min ) . For surface immunofluorescence , cells were stained as below at 4°C in HBSS prior to fixation with 3 . 7% formaldehyde . Actin was stained ( 20 min ) with Alexa Fluor 488- or 555-conjugated phalloidin . N- and E-cadherin , HA , p14 and p14-2HAN were detected by incubating cells for 1 h with the appropriate primary antibody followed by fluorophore-conjugated secondary antibodies for 45 min . Cells were mounted with fluorescent mounting medium ( Dako ) , images were acquired with LSM imaging software on a Zeiss LSM510 META laser scanning confocal microscope using the 488 nm argon laser for Alexa Fluor 488 or the 548 nm HeNe laser for Alexa Fluor 555 . Images were captured with the 63× or 100× Plan APOCHROMAT ( 1 . 4 NA ) objective lenses and processed in Adobe Photoshop version 6 . 0 using only linear adjustments . Three different adhesion phenotypes ( active , passive , no adhesion ) were generated in p14-transfected QM5 cells using a modified protocol [39] . Active adhesion was obtained by culturing cells in growth medium containing normal calcium levels with or without 0 . 1 µg/ml cytoD , or by culturing cells for 30 min in medium lacking calcium followed by incubation in growth media with normal calcium . No adhesion was generated by calcium depletion followed by continued incubation in calcium-free media in the presence and absence of 0 . 1 µg/ml cytoD . Passive adhesion was generated by first subjecting cells to calcium depletion to disrupt cellular junctions , followed by incubation in normal calcium containing medium containing 0 . 1 µg/ml cytoD . These conditions allowed cadherin-mediated contacts to form , but inhibited actin remodelling and the development of extended junction formation . Cells under all three conditions were fixed 2–4 h after the calcium switch and processed either for fluorescent microscopy or for quantification of fusion as described above . QM5 cells stably expressing influenza HAO ( QM5-HAO ) were selected using G418 ( Gibco ) , and HAO expression was confirmed by immunostaining . Cleavage of the HA0 precursor to its fusion-active HA form was accomplished by treatment with 10 µg/ml of trypsin for 5 min in HBSS . Cells were then washed and incubated in growth media containing 10% FBS for 10 min to inhibit residual trypsin activity . When appropriate , calcium was depleted using the calcium switch assay described above , followed by incubation for 20 min in MEM or S-MEM+10% dFBS to allow HAO or HA receptor interactions to form in low calcium conditions . Fusion was triggered with MEM or S-MEM at pH 4 . 8 containing 10 mM citrate buffer for 1 min . Cells were then transferred to MEM or S-MEM with 10% dFBS to allow syncytia to progress ( 20–40 min ) , then fixed with methanol , Giemsa-stained and syncytia were quantified as described above . For VSV-G fusion , cells were transiently transfected and cell–cell fusion was allowed to gradually progress without a specific low pH treatment to activate fusion , as previously reported [32] . This situation more closely mirrors the untriggered cell–cell fusion mediated by the FAST proteins .	Much of our current understanding of how proteins mediate membrane fusion derives from the study of enveloped virus fusion proteins . These fusion protein complexes function autonomously to co-ordinately regulate virus–cell attachment and subsequent membrane merger . In contrast , the reovirus Fusion-Associated Small Transmembrane ( FAST ) proteins are the only example of virus-encoded cellular fusogens , specifically designed to mediate cell–cell rather than virus–cell membrane fusion . In view of their small size , it was unclear if , or how , the FAST proteins are responsible for promoting the membrane attachment and close apposition stages of the fusion reaction . We now show that the FAST proteins have specifically evolved to function as the fusion component in a biphasic cell–cell fusion reaction , where the membrane attachment and membrane merger stages represent two distinct , uncoupled phases . Exploiting cadherins as surrogate adhesins , the FAST proteins have retained within their rudimentary structures the minimal determinants required to convert pre-existing adherens junctions into sites of cell–cell membrane fusion . These results raise the interesting possibility that other , yet to be identified cellular fusion proteins may resemble the FAST proteins , using separate adhesins and less complex fusion proteins in a similar biphasic membrane fusion reaction .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "virology", "cell", "biology/membranes", "and", "sorting", "infectious", "diseases/viral", "infections", "biochemistry/macromolecular", "assemblies", "and", "machines", "virology/host", "invasion", "and", "cell", "entry", "cell", "biology/cell", "adhesion", "cell", "biology/cytoskeleton" ]	2008	A Virus-Encoded Cell–Cell Fusion Machine Dependent on Surrogate Adhesins
Chemicals that are highly prevalent in our environment , such as phthalates and pesticides , have been linked to problems associated with reproductive health . However , rapid assessment of their impact on reproductive health and understanding how they cause such deleterious effects , remain challenging due to their fast-growing numbers and the limitations of various current toxicity assessment model systems . Here , we performed a high-throughput screen in C . elegans to identify chemicals inducing aneuploidy as a result of impaired germline function . We screened 46 chemicals that are widely present in our environment , but for which effects in the germline remain poorly understood . These included pesticides , phthalates , and chemicals used in hydraulic fracturing and crude oil processing . Of the 46 chemicals tested , 41% exhibited levels of aneuploidy higher than those detected for bisphenol A ( BPA ) , an endocrine disruptor shown to affect meiosis , at concentrations correlating well with mammalian reproductive endpoints . We further examined three candidates eliciting aneuploidy: dibutyl phthalate ( DBP ) , a likely endocrine disruptor and frequently used plasticizer , and the pesticides 2- ( thiocyanomethylthio ) benzothiazole ( TCMTB ) and permethrin . Exposure to these chemicals resulted in increased embryonic lethality , elevated DNA double-strand break ( DSB ) formation , activation of p53/CEP-1-dependent germ cell apoptosis , chromosomal abnormalities in oocytes at diakinesis , impaired chromosome segregation during early embryogenesis , and germline-specific alterations in gene expression . This study indicates that this high-throughput screening system is highly reliable for the identification of environmental chemicals inducing aneuploidy , and provides new insights into the impact of exposure to three widely used chemicals on meiosis and germline function . Man-made environmental chemicals such as phthalates , bisphenols , and pesticides , continue to increase in numbers , and some of them have been linked to reproductive problems [1–5] . However , rapidly identifying chemicals that impact reproductive health and understanding how they interfere with meiosis remains challenging . This is partly due to the fact that meiosis is not easily recapitulated in a tissue culture setting and that female mammalian meiosis can span from several months in mice to decades in humans . Failure to achieve accurate chromosome segregation during meiosis causes aneuploidy and can lead to infertility , stillbirths , miscarriages and birth defects [6 , 7] . Thus , high-throughput screens to assess the impact of environmental chemicals on reproductive health have been in high demand . C . elegans is a genetically and molecularly tractable model organism that provides many advantages for the study of meiosis and its use in high-throughput screens , including sharing a high degree of conservation of its genes and biochemical pathways with humans , carrying a well defined and characterized germline , a rapid life cycle ( it develops from an egg into an adult in approximately 3 days at 20°C ) and low maintenance costs [8–14] . Using chemotherapeutic agents and environmental compounds from the ToxCast Phase I library with comprehensive mammalian in vivo end point data ( ToxRef database ) , we previously demonstrated that a Pxol-1::gfp transcriptional reporter strain in C . elegans can be used to identify chemicals inducing embryonic aneuploidy . Moreover , we showed that this approach is highly predictive of mammalian reproductive toxicity ( balanced accuracy rate of 70%; this value corresponds to the average of sensitivity ( correct identification of true positives ) and specificity ( correct identification of true negatives ) ) [15] . Here , for the first time , we successfully combined the use of this strain with sorting of live worms based on fluorescence intensity with a large object flow cytometry system , the COPAS Biosort ( Union Biometrica ) , in a high-throughput screen . We screened a library of 46 chemicals consisting of pesticides , phthalates , and chemicals used in hydraulic fracturing and crude oil processing , selected based on their widespread presence in the environment and yet not well understood effects on the germline . Nineteen of these chemicals led to a GFP signal fold ratio over vehicle alone that was higher than the levels detected for bisphenol A ( BPA ) exposure , a widely used plasticizer and endocrine disruptor previously shown to affect meiosis in worms and mammals [1 , 16–20] . Three of these chemicals , dibutyl phthalate ( DBP ) , permethrin and 2- ( thiocyanomethylthio ) benzothiazole ( TCMTB ) , all high production volume chemicals , were evaluated further to validate the screening platform and gain insight into how they interfere with events in the germline . DBP is a phthalate ester widely used as either a solvent or plasticizer . Therefore , it is found in a variety of items such as personal care products , plastic food wraps , the enteric-coatings of solid oral drug products , adhesives , and printing inks . DBP is highly prevalent in the environment and its estimated daily intake for the general population is 7–10 μg/kg/day [21–23] . Higher levels have been detected in human urine , follicular fluid , and serum in different occupationally exposed groups [21 , 24] . In vitro and animal studies have shown that DBP disrupts the reproductive system resulting in inhibition of ovarian antral follicle growth and viability , altered gene expression in ovaries , testicular malformation and dysfunction , inhibition of spermatogenesis , and altered androgen signaling in males [25–28] . In humans , a significant inverse relationship has been observed between levels of DBP metabolites ( i . e . mBP ) in prenatal urine and male anogenital distance [29] . Permethrin , a synthetic pyrethroid insecticide , is commonly used for crop protection and the treatment of head lice and scabies , since it is considered to have low toxicity compared to other insecticides [30] . Permethrin enters the body through skin , inhalation , and oral uptake . This chemical has been detected in adults and children [31–34] and its major metabolites , cis-permethrin and trans-permethrin , were detected at levels of 34 and 36 pg/ml in human cord serum , respectively [35] . Permethrin has been shown to alter gene expression and induce breaks in genes associated with leukemia and lymphoma in peripheral blood mononuclear cells exposed in vitro [36] . Finally , TCMTB is widely used as a wood preservative , an antimicrobial chemical for water systems , as a preservative for paper products , leather products , paints and wallpaper , and also as a pesticide in seed treatment for barley , cotton , corn , oats , rice and wheat [37] . While data on human intake is scarce , inhalation has been associated with testicular cancer in rodents ( TCMTB risk assessment for the reregistration eligibility decision ( RED ) document , 2006 ) . Importantly , the effects on meiosis from exposures to DBP , permethrin and TCMTB remain poorly understood . Here we identified a set of chemical exposures affecting the germline and resulting in aneuploidy by using a high-throughput screening strategy in C . elegans . Subsequent analysis of three chemicals identified in this screen , DBP , permethrin , and TCMTB , provides new insights into their effects during meiosis and early embryogenesis . Exposure to these chemicals resulted in elevated DNA double-strand break ( DSB ) formation during meiosis and activation of a DNA damage checkpoint as indicated by elevated phosphorylated CHK-1 ( pCHK-1 ) signal and p53/CEP-1-dependent germ cell apoptosis . Oocytes in diakinesis exhibited defects such as chromosome fragments suggestive of impaired meiotic DSB repair . Live cell imaging revealed chromosome segregation defects and spindle abnormalities during early embryogenesis . Finally , germline-specific expression for conserved DSB formation , repair and DNA damage checkpoint signaling genes is altered following the chemical exposures . These results show that this high-throughput screening platform can be successfully applied to rapidly and reliably identify chemicals affecting germline function and suggest that DBP , permethrin , and TCMTB interfere with maintenance of genomic integrity during meiosis and achieving accurate chromosome segregation . Our high-throughput screening platform takes advantage of two key features of C . elegans: their transparency and the rarity of males ( X0<0 . 2% of offspring of self-fertilizing XX hermaphrodites; [38] ) , to identify chemical exposures that impair chromosome segregation in the germline and result in a high incidence of males . Increased X chromosome nondisjunction is detected using the reporter strain that has GFP expression controlled by a male-specific promoter ( Pxol-1::GFP ) [39 , 40] , and carrying a collagen gene mutation ( col-121 ( nx3 ) ) that increases cuticle permeability without affecting the worm’s life cycle [41] . The col-121 collagen gene mutation allows us to reduce the chemical concentrations to ≤100 μM , which are concentrations that circumvent lethality , are frequently used in chemical screens in C . elegans , and correlate well with mammalian reproductive endpoints ( [42]; Fig 1A and Materials and Methods ) . Briefly , we synchronized animals at the L1 larval stage by hypochlorite treatment and placed them on NGM plates with E . coli OP50 bacteria for food until the last larval stage ( L4 ) . Worms were exposed for 24 hours in 24-well plates starting at the L4 stage , when their gonads are fully formed , in liquid ( M9 buffer ) with OP50 . Importantly , use of live bacteria is not detrimental resulting in low false-positive and–negative rates ( [15 , 42 , 43]; S1A Fig ) . Gravid exposed mothers were then screened for increased incidence of GFP+ eggs ( destined to become males ) by comparison to vehicle alone ( 0 . 1% DMSO ) using the COPAS Biosort ( Union Biometrica ) , which allows for rapid sorting of live worms based on fluorescent intensities . We screened more than 5 , 000 animals in triplicate biological repeats for each chemical exposure . The 46 chemicals encompassed pesticides , phthalates , and chemicals used in hydraulic fracturing and crude oil processing ( S1 Table ) . GFP positive signal above background from each chemical exposure was calculated as fold increase over DMSO ( Fig 1B and S1 Table; also see Materials and Methods ) . We identified nineteen chemicals showing higher fold increase over DMSO than BPA , an endocrine disruptor shown to affect meiosis leading to increased chromosome nondisjunction in worms and mammals ( Fig 1B ) . Three of these are chemicals used in crude oil processing ( mercury , benzene and xylene ) , with mercury scoring highest from among all chemicals tested . Reprotoxic effects have been previously reported for mercury , benzene and xylene [44 , 45] . Mercury bioaccumulates and can cause pathophysiological changes in the hypothalamus pituitary gland that may alter follicle-stimulating hormone ( FSH ) and preovulatory luteinizing hormone ( LH ) release thereby affecting reproductive function [44] . Women with occupational exposure to the hydrocarbons benzene and xylene have been shown to have reduced LH and mid-luteal phase pregnanediol 3-glucuronide ( pd3G ) as well as increased follicular-phase pd3G , which can cause reproductive abnormalities [45 , 46] . Nine of the chemicals were pesticides including thiabendazole , triasulfuron , and pendimethalin , which previous in vivo and in vitro studies have shown to be genotoxic and can result in mammalian germ cell aneuploidy [15 , 47–51] . Piperonyl butoxide , another pesticide in the list , has been shown to induce decreased female reproductive organ weight and histopathological changes in the ovary , uterus and vagina in rats likely due to its anti-estrogenic activity [52] . Four were phthalates ( DEP , BBP , DBP , and DEHP ) , which are commonly added as solvents , additives and stabilizers to personal care products and medications and have adverse effects on reproductive and developmental health in humans ( reviewed in [53] ) . Exposure to phthalates has also been shown to disturb sperm function [54 , 55] , increase the percentage of pyriform sperm heads [56] , increase DNA damage in sperm [57–60] , impair mouse primordial follicle assembly in vitro [61] and reduce oocyte quality , embryonic developmental competency as well as alter expression of ovarian and pre-implantation embryonic genes in mice [62] . Finally , epidemiological studies suggest that exposure to isopropanol and ethylene glycol , chemicals used in hydraulic fracturing , may have a negative impact on human reproductive health [63] . The observed increase in X chromosome nondisjunction detected for all of these chemicals suggests effects in germline functions that will require further investigation . For further validation of this high-throughput screening strategy , we selected a phthalate , DBP , and two pesticides , TCMTB and permethrin . These were selected given that they elicited elevated levels of X chromosome nondisjunction and their reprotoxicity is less understood , allowing us to also gain more insight into their effects in germline function ( see below ) . To assess whether DBP , permethrin and TCMTB affect chromosome segregation in general ( including autosomes ) , and determine the dose of exposure for subsequent studies , we exposed worms to various concentrations of these chemicals ( 1 , 10 , 100 and 500 μM ) for 24 hours starting at late L4 , as in the high-throughput screen , and scored the number of eggs laid ( brood size ) , embryonic lethality , and larval lethality . A decreased brood size and increased embryonic lethality can be due in part to defects during meiosis leading to errors in autosomal chromosome segregation and the consequent formation of aneuploid gametes in C . elegans [9 , 38 , 64] . We observed approximately a 50% reduction in the mean numbers of eggs laid on plates , which is indicative of increased sterility , for worms exposed to 500 μM DBP , 500 μM permethrin and both 100 μM and 500 μM TCMTB , compared to vehicle alone ( Fig 2A ) . We also observed significantly increased embryonic lethality for exposures starting at 100 μM for DBP and permethrin ( P<0 . 001; two-tailed Mann-Whitney test , C . I . 95% ) and 10 μM for TCMTB ( P<0 . 05 ) ( Fig 2A ) . Furthermore , we observed higher larval lethality among the progeny of worms exposed to 500 μM TCMTB ( P<0 . 05 ) . These data further support and extend the results of our high-throughput screen suggesting that all three chemicals affect chromosome segregation in general and not limited to the X chromosome . Moreover , based on this analysis , doses of 100 μM for DBP and permethrin , and 10 μM for TCMTB , which result in embryonic lethality without significantly reducing the brood size or causing larval lethality , were used for all subsequent analysis . To determine whether the increased chromosome nondisjunction is due in part to defects during meiosis , we examined DAPI-stained gonads from worms following exposures . In C . elegans , nuclei are positioned in a spatial and temporal gradient along the germline facilitating the identification of alterations in chromosome organization at specific meiotic stages [9] . We observed an increase in the number of gonads with gaps ( areas with a reduced density of nuclei ) in worms exposed to DBP compared to vehicle alone ( 30 . 2% , n = 53 , and 10 . 9% , n = 55 , respectively ) , as well as the presence of nuclei with DAPI-bright chromatin forming aggregates and nuclei with DAPI-bright and collapsed chromatin in a leptotene/zygotene-like organization at late pachytene in the gonads of worms exposed to all three chemicals compared to vehicle ( aggregates: DBP: 20 . 8% , n = 53; permethrin: 21 . 7% , n = 46; TCMTB: 33 . 3% , n = 60; and DMSO: 7 . 3% , n = 55; leptotene/zygotene-like organization: DBP: 24 . 5% , n = 53; permethrin: 47 . 8% , n = 46; TCMTB: 31 . 7% , n = 60; and DMSO: 5 . 5% , n = 55 ) ( Fig 2B and 2C ) . However , these defects are not due to overt impairments to early stages of meiotic progression or chromosome synapsis ( S2A–S2E Fig ) . This is evidenced by normal localization of phosphorylated SUN-1 ( SUN-1 S8 ) , where SUN-1 corresponds to a conserved inner nuclear envelope protein with CHK-2- and PLK-2-dependent phosphorylation , with a signal appearing upon entrance into meiosis at the leptotene/zygotene stage and persisting on nuclei until mid-pachytene [65] . This is further supported by the normal localization of SYP-1 , a structural component of the central region of the synaptonemal complex , observed associating with nuclei upon entrance into meiosis , forming full tracks between homologs at pachytene and starting to disassemble by late pachytene [66] . Taken together , these results show evidence of sterility , embryonic lethality and larval lethality as well as chromosome defects during pachytene after DBP , permethrin and TCMTB exposure , but these are not due to defects in meiotic progression or chromosome synapsis . Nuclei with DAPI-bright and collapsed chromatin at late pachytene have been previously correlated with germ cells undergoing apoptosis [67] . To determine whether the chemical exposures are causing increased germ cell apoptosis , we scored germline nuclei undergoing apoptosis ( germ cell corpses , also referred to as apoptotic bodies ) by acridine orange staining as in [39] . In wild type , animals exhibit less than three apoptotic bodies in late pachytene reflecting regular physiological apoptosis [64 , 68] ( Fig 3A ) . We observed a significant 2- to 3-fold increase in the levels of germ cell corpses in late pachytene following all three chemical exposures compared to vehicle alone ( Fig 3B and 3C ) . Moreover , the elevated germ cell apoptosis was observed in a dose-dependent manner , starting at 100 μM for DBP and permethrin , and 10 μM for TCMTB ( S3 Fig ) . In C . elegans , physiological germ cell apoptosis does not depend on p53/CEP-1 , which responds to genotoxic stress as a result of the activation of a DNA damage checkpoint [69] . Analysis of germ cell apoptosis levels in cep-1;col-121 worms revealed the elevated apoptosis following all three chemical exposures was p53/CEP-1-dependent ( Fig 3C ) . Activation of a DNA damage checkpoint was further supported by the increased signal detected in pachytene nuclei for a checkpoint kinase involved in DNA damage sensing , phosphorylated CHK-1 [70] , following each chemical exposure ( Fig 3D ) . Therefore , DBP , permethrin , and TCMTB exposures lead to activation of a DNA damage checkpoint resulting in increased p53/CEP-1-dependent germ cell apoptosis to clear affected nuclei . The pachytene DNA damage checkpoint can be activated by the presence of unrepaired DSBs or aberrant recombination intermediates [64] . To examine this further , we quantified levels of RAD-51 foci as in [67] . RAD-51 binds to 3’ ssDNA ends at DSBs to promote strand invasion/exchange during DSB repair [71] . In vehicle-exposed gonads , like in wild type , low levels of RAD-51 foci were observed in nuclei at the premeiotic tip ( zones 1–2 ) undergoing mitosis , as well as upon entrance into meiosis at transition zone where leptotene/zygotene stage nuclei are located ( zone 3 ) . Levels continued to rise throughout pachytene , peaking by mid-pachytene ( zone 5 ) , and then decreased by late pachytene ( zone 7 ) as DSB repair progressed ( Fig 4A–4C ) . In contrast , levels of RAD-51 foci were elevated specifically during meiosis for all three chemical exposures ( Fig 4B and 4C ) . Levels of RAD-51 foci were indistinguishable from vehicle alone throughout the mitotic zone , but were higher than vehicle alone during pachytene . Moreover , analysis of col-121 worms depleted of SPO-11 , the protein required for meiotic DSB formation , further confirmed that the elevated levels of RAD-51 foci were SPO-11-dependent and therefore , meiotic-specific and not due to damage from the chemicals to the chromosomes ( S4A–S4D Fig ) . To determine whether the elevated levels of RAD-51 foci may be due in part to elevated DSB levels , we quantified RAD-51 foci in rad-54;col-121 double mutants following chemical exposures . In a rad-54 mutant , DSBs are formed and RAD-51 associates with DSB repair sites , but further repair is blocked essentially “trapping” DSB-bound RAD-51 and allowing for quantification of the total number of DSBs [72] . Levels of RAD-51 foci were significantly higher during meiosis for all three exposures compared to vehicle alone ( Fig 4D ) . Taken together , these results suggest that exposures to DBP , permethrin and TCMTB result in elevated meiotic DSB levels and impaired DSB repair leading to activation of a DNA damage checkpoint and p53-dependent increased germ cell apoptosis . To determine whether exposures to DBP , permethrin and TCMTB might also result in defects at late prophase I , we examined chromosome morphology in oocytes at late diakinesis . In C . elegans , the six pairs of attached homologs ( bivalents ) are detected as six DAPI-stained bodies during diakinesis ( Fig 5B ) . With high resolution microscopy we analyzed the -1 and -2 oocytes at diakinesis , which correspond to the two last oocytes proximal to the spermatheca ( Fig 4A ) . We observed increased numbers of oocytes carrying chromosomes exhibiting a frayed morphology , chromatin bridges and chromosome fragments in germlines exposed to all three chemicals compared to vehicle alone ( Fig 5A and 5B ) . This suggests that despite activation of p53-dependent apoptosis in late pachytene , some nuclei that failed to undergo normal DSB repair are progressing into late diakinesis . To determine if DBP , permethrin and TCMTB exposures also impact early embryogenesis , we examined the first embryonic cell division by live imaging using transgenic worms carrying H2B::mCherry; γ-tubulin::GFP and the col-121 ( nx3 ) mutation . We detected the presence of lagging chromosomes ( congression failure ) and spindle abnormalities in the embryos examined from all three chemical exposures , and evidence of chromatin bridges in the metaphase to anaphase transition from DBP and TCMTB exposures compared to vehicle alone ( Fig 5C and 5D ) . These results suggest that these chemical exposures affect meiosis as well as early embryogenesis , supporting the elevated chromosome nondisjunction detected by high-throughput screening following these treatments . Given the effects on DSB formation , repair and DNA damage checkpoint activation observed during meiosis following all three chemical exposures , we next examined whether these defects arise from alterations in expression of DSB repair and DNA damage response genes . We examined mRNA levels by quantitative RT-PCR for 15 critical and conserved genes involved in these processes ( Fig 6A and 6B ) . We used glp-1;col-121 double mutants that develop as wild type at 15°C , but grow into adults lacking a germline when shifted to 25°C [73] , to distinguish changes in gene expression occurring in the soma from those taking place in the germline . All three chemical exposures led to a significant increase in chk-1 expression at 15°C ( P<0 . 05 ) , but not at 25°C ( Fig 6A and 6B ) , indicating a germline-specific change in gene expression and correlating with the increased pCHK-1 foci we detected in the germline ( Fig 3D ) . Moreover , DBP exposure resulted in germline-specific up regulation of spo-11 and down regulation of mre-11 ( P<0 . 001 and P<0 . 05 , respectively ) , which are factors involved in DSB formation and repair , as well as upregulation of prmt-5 ( P<0 . 01 ) , involved in regulation of DNA damage-induced apoptosis [74 , 75] . Taken together , these results suggest that germline-specific alterations in the expression of genes involved in DSB formation , repair and response may contribute in part to the defects in maintaining genomic integrity and achieving accurate chromosome segregation observed in the germline following these chemical exposures . We showed that a high-throughput screening strategy can be successfully applied to identify environmental chemicals causing aneuploidy using the nematode C . elegans . Due to the increasing number of chemicals being introduced into the environment and their broad uses , strategies for rapidly assessing toxicity , in a manner predictive of their effects on human health , are in high demand . C . elegans is a metazoan system that offers various advantages for this type of analysis including low maintenance costs , a rapid life cycle and a high degree of conservation of its genes and biochemical pathways with humans [9 , 76–80] . This nematode is also being successfully used in high-throughput screens of compounds and genes impacting pathways related to human disease [11 , 13 , 81–83] . C . elegans is also an ideal model system specifically for studying the effects of chemical exposures on the germline since genes and pathways involved in regulating key processes such as germ stem cell renewal and differentiation , meiosis , ovulation and embryogenesis are conserved between C . elegans and humans . Moreover , its germline is well characterized and amenable to studies using genetic , biochemical , and molecular biology tools combined with powerful cytological approaches [8 , 12 , 14] . Our high-throughput screen identified several different classes of chemicals that are leading to increased chromosome nondisjunction and follow up studies will further explore how they are affecting the germline . Along those lines , here we also provided new insights into the effects of DBP , permethrin and TCMTB on germline functions . DBP is metabolized by esterases to form mono-n-butyl phthalate ( mBP ) once it enters the body , while permethrin undergoes hydrolysis and oxidation in the liver by carboxylesterases and cytochrome P450 to conjugated and unconjugated cis/trans 3- ( 2 , 2-dichlorovinyl ) -2 , 2-dimethylcyclopropane carboxylic acid ( CVA ) with their plasma levels reaching peak values in 5–7 hours [84–86] , and TCMTB is converted into cyanide , 2-mercaptobenzothiazole ( 2-MBT ) by cytochrome P450 in the liver [87] . In our study , we exposed worms for 24 hours to 100 μM or 27 . 8 μg/ml of DBP , which our dose-response studies showed impaired chromosome segregation with low overall toxicity , as determined by both growth and behavior of the worms . Measurement of the internal concentration from whole worm extracts by isotopic dilution mass spectrometric analysis revealed internal levels of 8 . 9 μg/ml for DBP , and 2 . 2 μg/ml for mBP , confirming that this chemical reaches internal circulation ( Fig 7 ) . A study of a small group of women undergoing IVF in the USA detected a median value of 1 . 46 ng/ml of mBP in follicular fluid [88] while a larger study of 110 women undergoing IVF in China detected median values for mBP of 2 . 05 ng/ml in follicular fluid and 102 . 30 ng/ml in urine and maximum values for DBP of 415 ng/ml in follicular fluid and 2 . 32 μg/ml in urine [89] . Studies showing the relationship between prenatal phthalate exposure and anogenital distance ( AGD ) as an outcome of reproductive toxicity detected geometric means for mBP of 67 . 62 ng/ml in urine from 196 women in Sweden and 15 . 04 ng/ml in urine from 380 women in the USA [29 , 90] . This suggests that the levels which resulted in germline defects in C . elegans are within the range relevant to human exposures . In this study , worms were exposed for 24 hours to external doses of 100 μM permethrin and 10 μM TCMTB , corresponding to 39 . 13 μg/ml and 2 . 38 μg/ml , respectively . Measurement of the internal concentrations from whole worm extracts revealed internal levels of 5 . 6 μg/ml for cis- and trans-permethrin combined , and 0 . 5 μg/ml for 2-methylthio benzothiazole ( 2-MeS BTH ) , showing that these chemicals also reach internal circulation ( Fig 7 ) . TCMTB is apparently metabolized very rapidly in worms ( akin to mammals [87] ) and we were unable to detect it in our extracts . We also did not detect either benzothiazole ( BTH ) or 2-hydroxy benzothiazole ( 2-OH BTH ) metabolites . Unfortunately , data on the concentrations of these metabolites in non-blood tissue or organs in either mammalian models or humans is scarce to nonexistent , thus limiting comparisons regarding exposure levels . However , the internal levels detected for DBP and mBP , coupled with the effects on germline functions detected by our high-throughput screen and follow up studies for all three chemicals , suggest that C . elegans offers the necessary sensitivity to detect effects at environmentally relevant doses of exposure . Here we showed that all three chemicals resulted in elevated meiotic DSB levels , impaired DSB repair , activation of p53-dependent germ cell apoptosis and elevated phosphorylated CHK-1 signal during late pachytene , along with chromosomal abnormalities in oocytes at late diakinesis and impaired chromosome segregation during early embryogenesis . In vitro studies of the effects of DBP exposure on Sertoli cell culture showed increased apoptosis stemming from inhibiting the PI3K/AKT and mTOR pathways which promote the proliferation and survival of sperm and the maintenance of testicular homeostasis [26] . Antral follicles isolated from female mice and exposed to DBP exhibited increased expression of the cyclin-dependent kinase inhibitors Cdkn1a and Cdkn2a and pro-apoptotic factors Bax and Bid along with down regulation of cyclin Ccnd2 resulting in growth inhibition and follicular death [28] . Permethrin has been shown to cause DNA damage in mitotic cells where exposure of peripheral blood mononuclear cells ( PBMCs ) induced breaks in the KMT2A and IGH genes which can be driver mutations for lymphoma and leukemia along with increased aneuploidy [36] . TCMTB has been reported as exhibiting relatively low toxicity upon either oral or dermal uptake since it is rapidly metabolized into 2-MBT in the body and excreted through the urinary tract , however it has been considered highly toxic via the inhalation route and resulted in increased incidence of testicular interstitial cell adenomas in male rats [87] , ( TCMTB risk assessment for the reregistration eligibility decision ( RED ) document , 2006 ) . To our knowledge , the germline-specific upregulation of chk-1 following exposures to DBP , permethrin and TCMTB , and the downregulation of mre-11 along with the upregulation of spo-11 and prmt-5 following DBP exposure , have not been previously reported . This altered gene expression profile is congruent with the elevated meiotic DSB formation , altered meiotic DSB repair and activation of a DNA damage checkpoint observed during meiosis . SPO-11 is the topoisomerase-like conserved protein that catalyzes meiotic DSBs so that the elevated levels of DSB formation observed in DBP exposed worms may be due in part to deregulation of spo-11 expression . prmt-5 encodes for the ortholog of human PRMT5 , a protein arginine methyltransferase involved in the post-translational modification of a variety of proteins including histones and G protein-coupled receptors , thereby regulating transcription and signaling [91] . Interestingly , PRMT5 has been proposed to regulate the target gene specificity of p53 in mammals [74] and to negatively regulate apoptotic signaling in response to DNA damage in C . elegans by repressing p53/CEP-1 transcriptional activity through downregulation of cbp-1/p300 , which encodes for a cofactor of CEP-1 [75] . Alternatively , upregulation of prmt-5 may result in alterations in transcription or chromatin accessibility , for example via its role in histone regulation , which in turn could also contribute to elevated DSB formation and/or altered repair . Finally , mre-11 encodes for a member of the MRX/N ( Mre11 , Rad50 , Xrs2/Nbs1 ) complex required for meiotic DSB formation and resection [92–97] . The elevated levels of DSBs and RAD-51 foci detected following DBP exposure suggest that downregulation of mre-11 may not be interfering with DSB formation and/or end resection in this case . Finally , although all three chemical exposures led to elevated germline-specific chk-1 gene expression , additional studies will be required to determine how permethrin and TCMTB affect DSB formation and repair during meiosis . Taken together , our results demonstrate that a high-throughput screening platform can be used in C . elegans to successfully identify environmental chemicals affecting the germline . Moreover , our findings revealed the effects of DBP , permethrin , and TCMTB exposures on the germline and potential mechanisms by which DBP affects germline functions , broadening our understanding of the potential effects of environmental toxicants on human reproductive health . C . elegans strains were cultured at 20°C under standard conditions as described in [98] . The following mutations and chromosome rearrangements were used in this study: LGI , cep-1 ( lg12501 ) , rad-54 ( ok615 ) ; LGIII , glp-1 ( bn18 ) ; LGIV , col-121 ( nx3 ) , him-8 ( e1489 ) ; LGV , yIs34[Pxol-1::GFP , rol-6] . To ensure the GFP signals detected in the COPAS Biosort stemmed only from gravid worms and not debris , we first synchronized worms at the L1 stage by hypochlorite treatment as in [99] and sorted either Pxol-1::gfp; col-121 or Pxol-1::gfp; col-121; him-8 ( e1489 ) worms at different developmental stages ( L1 through young adults ) . Presence of the him-8 mutation results in a high incidence of male progeny ( 36 . 7%; [38] ) due to increased X chromosome nondisjunction . The use of both strains allowed us to exclusively gate adult worms and establish the threshold for worms carrying GFP+ embryos above background . Thus , only an adult population from age-matched animals underwent screening ( S1B–S1D Fig ) . Age-matched embryos were obtained from gravid worms following sodium hypochloride treatment and subjected to overnight starvation [99] . Thoroughly washed age-matched L1-stage worms were grown on regular NGM plates up to the L4 stage ( 5 , 000 to 6 , 000 L1 worms were grown on each 100 mm plate ) . Between twenty to thirty thousand L4-stage animals were resuspended in M9 buffer with freshly cultured OP50 bacteria ( OD600 = 24 ) . 300 worms in 250 μl of M9 with OP50 bacteria were dispensed , along with each chemical , into individual wells in 24-well plates . All chemicals , except TCDD , were purchased from Sigma Aldrich ( St . Louis , MO ) and dissolved in DMSO at 0 . 1 M except for polyacrylamide and hydroxyethyl cellulose , which were dissolved in water respectively at 20 mg/ml and 33 mg/ml for solubility reasons . TCDD was purchased from AccuStandard ( New Haven , CT ) . Final DMSO and chemical concentrations were 0 . 1% and 100 μM , respectively , except for dicofol , mancozeb , parathion-methyl , phosalone , pyridaben , TCMTB , arsenic oxide , and mercury , which were further diluted 10-fold to circumvent lethality . Chlorpyrifos-methyl and TCDD were used at 1 μM and 100 nM , respectively , for the same reason . After 24 hours , the exposed worms were transferred to 1 . 5 ml tubes , washed five times with M9 , and utilized for subsequent experiments and analysis through the COPAS Biosort ( Union Biometrica , Holliston , MA ) . Time-of-flight ( Tof ) and GFP peak height were used as reading parameters in the COPAS Biosort . Three independent biological repeats , encompassing a total of more than 5 , 000 worms for each chemical exposure , were run through the COPAS Biosort . Fold-increase GFP+ signal over DMSO was calculated for each biological repeat and then an average of the fold increase was calculated ( Fig 1B ) . Age-matched worms were exposed to either vehicle alone ( DMSO ) or each chemical in liquid for 24 hours as described above . After 24 hours , the exposed worms were washed five times with M9 , and transferred to regular NGM plates to score their embryonic lethality , larval lethality and sterility . Worms were moved every 24 hours to new NGM plates ( this was done for three consecutive days ) . The total number of fertilized eggs laid , hatched , and the number of progeny that reached adulthood were scored . Germ cell corpses were scored as in [39] , utilizing a Leica DM5000B fluorescence microscope . The germlines of more than 30 worms from at least two independent biological repeats were scored for each chemical exposure . Statistical comparisons between groups were performed using the two-tailed Mann-Whitney test , 95% C . I . Whole mount preparation of dissected gonads and immunostainings were performed as in [67] . Primary antibodies were used at the following dilutions: goat α-SYP-1 ( 1:3 , 000; [100] ) , rabbit α-pCHK-1 ( 1:100; Santa Cruz ) , guinea pig α-pSUN-1 Ser8-pi ( 1:700; [65] ) , and rabbit α-RAD-51 ( 1:10 , 000; Novus Biological ( SDI ) ) . The following secondary antibodies from Jackson ImmunoResearch Laboratories ( West Grove , PA ) were used at a 1:200 dilution: α-rabbit Cy3 , and at a 1:500 dilution: α-goat Alexa 647 , α-rabbit Alexa 488 , and α-guinea pig Alexa 488 . Vectashield from Vector Laboratories ( Burlingame , CA ) was used as a mounting media and anti-fading agent . Immunofluorescence images were collected at 0 . 2 μm intervals with an IX-70 microscope ( Olympus , Waltham , MA ) and a cooled CCD camera ( CH350; Roper Scientific ) controlled by the Delta Vision system ( Applied Precision , Pittsburgh , PA ) . Images were subjected to deconvolution by using the SoftWoRx 3 . 3 . 6 software ( Applied Precision ) . Quantitative analysis of RAD-51 foci for all seven zones composing the germline was performed as in [67] . The average number of nuclei scored per zone ( n ) from 3 to 6 gonads for each chemical-treated group was as follows , ± standard deviation: For the col-121 line: zone 1 ( n = 80 . 3±4 . 3 ) , zone 2 ( n = 93 . 5±7 . 4 ) , zone 3 ( n = 97 . 0±9 . 3 ) , zone 4 ( n = 86 . 3±6 . 6 ) , zone 5 ( n = 59 . 3±8 . 5 ) , zone 6 ( n = 56 . 5±1 . 7 ) , zone 7 ( n = 57 . 8±10 . 7 ) . For the rad-54;col-121 line: zone 1 ( n = 56 . 5±5 . 8 ) , zone 2 ( n = 67 . 0±6 . 2 ) , zone 3 ( n = 59 . 8±5 . 2 ) , zone 4 ( n = 46 . 0±6 . 6 ) , zone 5 ( n = 39 . 8±7 . 8 ) , zone 6 ( n = 31 . 5±5 . 4 ) , zone 7 ( n = 26 . 3±2 . 5 ) . Statistical comparisons were performed using the two-tailed Mann-Whitney test , 95% C . I . Feeding RNAi experiments were performed at 20°C in col-121 mutants as described in [101] with the following modifications: three L4-stage animals were placed on each RNAi plate and F3 generation L4-stage worms were used for chemical exposures at 25°C . HT115 bacteria expressing empty pL4440 vector was used as the control RNAi . Strong RNAi knockdown of spo-11 results in oocytes with 12 DAPI-stained bodies due to the lack of meiotic DSBs and subsequent crossovers leading to 6 unattached pairs of homologs . We verified that 100% ( n>22 ) of the oocytes for each chemical exposure exhibited 12 DAPI-stained bodies . The effectiveness of RNAi was also confirmed by RT-PCR from at least four individual worms subjected to RNAi . Expression of gpd-1 ( GAPDH ) transcript was used as a control . Three samples of 20 ( 15°C ) to 30 animals ( 25°C ) each were collected in 100 μl of Trizol ( Invitrogen ) and RNA was extracted according to the manufacturer’s instructions . The extracted RNA was subjected to reverse transcription using iScript ( Biorad ) and quantitative real time PCR was performed using SsoFast EvaGreen supermix ( Biorad ) according to the manufacturer’s instructions . Each sample was run in triplicate . Cq numbers were normalized to gpd-1 , then the normalized values from DBP , permethrin , and TCMTB treated samples were statistically compared with the normalized values from vehicle ( DMSO ) treated samples . Bars in graphs show mean values normalized to DMSO ± SEM . Statistical comparisons were performed using the unpaired two tailed t-test , 95% C . I . Strain CV639 ( H2B::mCherry; γ-tubulin::GFP;col-121 ( nx3 ) ) was used for live imaging and worms were immobilized with 0 . 01% levamisole on 3% agarose pads . Images were captured with a 60X objective every 10 seconds on an IX-70 microscope ( Olympus , Waltham , MA ) and a cooled CCD camera ( CH350; Roper Scientific ) controlled by the DeltaVision system ( Applied Precision , Pittsburgh , PA ) . After exposure to each chemical as described above , worms were washed 10 times in M9 buffer and frozen with minimal M9 in liquid nitrogen . The worm pellet was resuspended in lysis buffer [0 . 5 M sucrose , 25 mM HEPES ( pH7 . 6 ) , 5 mM EDTA , 0 . 5% CHAPS , 0 . 5% DOC ( Deoxychloric acid ) ] . Samples were then sonicated at 4°C for 10 cycles ( 1 minute on and 1 minute off per cycle ) with a Bioruptor Plus 300 ( Diagenode , Belgium ) . DBP was extracted from worms/lysates using hexane and analyzed using gas chromatography-mass spectrometry ( GC-MS ) . A Thermo trace 1310 GC and a HP-5MS capillary column ( 30 m×0 . 25 mm×0 . 25 μm ) interfaced with ISQ single quadrupole mass spectrometer ( Waltham , MA , USA ) was used for the analysis . Procedural blanks and matrix spikes were included for quality control purposes along with the analysis of control/vehicle and treated C . elegans . The trace level DBP found in the procedural blank was subtracted from sample values to report the final concentration . The matrix spike recovery was 91 . 6% . mBP ( the metabolite of DBP ) in worms was analyzed using a method described for urine earlier , with some modifications [102] . Briefly , the worm lysates were enzymatically ( β-glucuronidase ) deconjugated followed by extraction using a solid-phase extraction ( SPE ) method with a solvent mixture of acetonitrile and ethyl acetate . An API 4500 electrospray QTRAP mass spectrometer ( ESI-MS/MS; Applied Biosystems , AB Sciex , Framingham , MA , USA ) operated in the negative mode of ionization interfaced with an Agilent 1260 HPLC ( Agilent Technologies Inc . , Santa Clara , CA ) was used for the analysis of mBP . Quantification of mBP was achieved by an isotopic dilution method . Permethrin was extracted from worm lysates using a similar protocol to that applied for DBP analysis . A 1:2 ratio of hexane and dichloromethane solvent mixture was used for the extraction and analysis was performed using an Agilent single quadrupole GC-MS under electron ionization mode . The matrix spike recoveries for both cis- and trans-permethrin were 112% and 101% , respectively . For TCMTB analysis , worm lysate was spiked with 40 ng of D4-Benzothiazole ( internal standard ) and maintained at room temperature for equilibration ( 15 min ) . Methanol and acetone ( 1:1 ratio ) solvent mixture was used for the extraction of target chemicals . The extracts were centrifuged and filtered through 0 . 2 μM nylon membrane filters and transferred into HPLC amber vials . A Shimadzu Prominence Modular HPLC system ( LC-20 AD UFLC; Shimadzu Corporation , Kyoto , Japan ) equipped with an Agilent Zorbax SB-Aq column ( 2 . 1 mm X 150 mm , 3 . 5 mm; Santa Clara , CA , USA ) serially connected with an AB SCIEX 3200 triple quadrupole mass spectrometer was used for the identification and quantification of TCMTB and its metabolites ( Benzothiazole ( BTH ) , 2-methylthio benzothiazole ( 2-MeS BTH ) and 2-hydroxy benzothiazole ( 2-OH BTH ) ) under the positive electrospray ionization mode . Although we screened for three major possible metabolites of TCMTB , we could detect only 2-MeS BTH in the treated worms . The gradient mobile phase ( A: acetonitrile and B: water that contains 0 . 1% formic acid ) was eluted at a flow rate of 300 μL/min for the effective separation of target chemicals .	The ever-increasing number of new chemicals introduced into our environment poses a significant problem for risk assessment . In addition , assessing the direct impact of toxicants on human meiosis remains challenging . We successfully utilized a high-throughput platform in the nematode C . elegans , a genetically tractable model organism which shares a high degree of gene conservation with humans , to identify chemicals that affect the germline leading to aneuploidy . We assessed chemicals that are highly prevalent in the environment in worms carrying a fluorescent reporter construct allowing for the identification of X chromosome nondisjunction combined with a mutation increasing cuticle permeability for analysis of low doses of exposure . Follow up analysis of three chemicals: DBP , permethrin and TCMTB , further validated the use of this strategy . Exposure to these chemicals resulted in elevated levels of DNA double-strand breaks , activation of a DNA damage checkpoint , chromosome morphology defects in late meiotic prophase I as well as impaired early embryogenesis and germline-specific changes in gene expression . Our results support the use of this high-throughput screening system to identify environmental chemicals inducing aneuploidy , and provide new insights into the effects of exposure to DBP , permethrin , and TCMTB on meiosis and germline function .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "meiosis", "invertebrates", "cell", "death", "medicine", "and", "health", "sciences", "reproductive", "system", "gonads", "caenorhabditis", "cell", "cycle", "and", "cell", "division", "cell", "processes", "animals", "germ", "cells", "animal", "models", "dna", "damage", "caenorhabditis", "elegans", "oocytes", "model", "organisms", "experimental", "organism", "systems", "molecular", "biology", "techniques", "dna", "research", "and", "analysis", "methods", "chromosome", "biology", "animal", "cells", "animal", "studies", "high", "throughput", "screening", "molecular", "biology", "molecular", "biology", "assays", "and", "analysis", "techniques", "biochemistry", "eukaryota", "cell", "biology", "anatomy", "ova", "nucleic", "acids", "apoptosis", "genetics", "nematoda", "biology", "and", "life", "sciences", "cellular", "types", "organisms", "genital", "anatomy" ]	2019	Assessing effects of germline exposure to environmental toxicants by high-throughput screening in C. elegans
According to the ‘ceRNA hypothesis’ , microRNAs ( miRNAs ) may act as mediators of an effective positive interaction between long coding or non-coding RNA molecules , carrying significant potential implications for a variety of biological processes . Here , inspired by recent work providing a quantitative description of small regulatory elements as information-conveying channels , we characterize the effectiveness of miRNA-mediated regulation in terms of the optimal information flow achievable between modulator ( transcription factors ) and target nodes ( long RNAs ) . Our findings show that , while a sufficiently large degree of target derepression is needed to activate miRNA-mediated transmission , ( a ) in case of differential mechanisms of complex processing and/or transcriptional capabilities , regulation by a post-transcriptional miRNA-channel can outperform that achieved through direct transcriptional control; moreover , ( b ) in the presence of large populations of weakly interacting miRNA molecules the extra noise coming from titration disappears , allowing the miRNA-channel to process information as effectively as the direct channel . These observations establish the limits of miRNA-mediated post-transcriptional cross-talk and suggest that , besides providing a degree of noise buffering , this type of control may be effectively employed in cells both as a failsafe mechanism and as a preferential fine tuner of gene expression , pointing to the specific situations in which each of these functionalities is maximized . The problem of tuning protein expression levels is central for eukaryotic cell functionality . A variety of molecular mechanisms are implemented to guarantee , on one hand , that protein copy numbers stay within a range that is optimal in the given conditions and , on the other , that shifts in expression levels can be achieved efficiently whenever necessary [1–3] ( whereby ‘efficiency’ here encompasses both a dynamical characterization , in terms of the times required to shift , and a static one , in terms of moving as precisely as possible from one functional range to another ) . Quantifying and comparing their effectiveness in different conditions is an important step to both deepen our fundamental understanding of regulatory circuits and to get case-by-case functional insight about why a specific biochemical network has been selected over the others . As the major direct regulators of gene expression , transcription factors ( TFs ) are most immediately identified as the key potential modulators of protein levels [4] . In a somewhat simplified picture , one may imagine that a change in amount of a TF can induce a change in the expression level of the corresponding gene , and that the ability to regulate the latter ( the output node ) via the former ( the input node ) can be assessed by how strongly the two levels correlate . The effectiveness of a regulatory element is however limited by the stochasticity of intracellular processes , from the TF-DNA binding dynamics to translation [5] . A convenient framework to analyze how noise constrains regulation is provided by information theory [6 , 7] . In particular , the simplest situation in which a single TF modulates the expression of a single protein can be characterized analytically under the assumption that the noise affecting the input-output channel is sufficiently small . The mutual information between modulator and target –a convenient quantity through which regulatory effectiveness can be characterized– depends on the distribution of modulator levels and can be maximized over it . Remarkably , in at least one case this maximum has been found to be almost saturated by the actual information flow measured in a living system ( for more details see [8 , 9] ) . In other terms , for sufficiently small noise levels in the channel that links TFs to their functional products , one may quantify the optimal regulatory performance achievable in terms of the maximum number of bits of mutual information that can be exchanged between modulator and target . Several control mechanisms however act at the post-transcriptional level [10–12] . Among these , regulation by small regulatory RNAs like eukaryotic microRNAs ( miRNAs ) has attracted considerable attention over the past few years [13–15] . In short , miRNAs are small non-coding RNA molecules encoded by nuclear DNA , that can inhibit translation or catalyze degradation of mRNAs when bound to them via protein-mediated base-pairing . miRNAs appear to be crucial in an increasing number of situations ranging from development to disease [16–18] . Their function however can differ significantly from case to case . For instance , they have been well characterized as noise buffering agents in protein expression [19 , 20] or as key signaling molecules in stress response [21] . Recently , though , investigations of the miRNA-mediated post-transcriptional regulatory ( PTR ) network have hinted at a possibly more subtle and complex role . It is indeed now clear [22–27] that the miRNA-RNA network describing the potential couplings stretches across a major fraction of the transcriptome , with a large heterogeneity both in the number of miRNA targets and in the number of miRNA regulators for a given mRNA . The competition effects that may emerge in such conditions suggest that miRNAs may act as channels through which perturbations in the levels of one RNA could be transmitted to other RNA species sharing the same miRNA regulator ( s ) . Such a scenario has been termed the ‘ceRNA effect’ , whereby ceRNA stands for ‘competing endogenous RNA’ [28] . In view of its considerable regulatory and therapeutic implications , the ceRNA effect has been extensively analyzed both theoretically and experimentally [29–44] . The apparent ubiquity of potentially cross-talking ceRNAs however raises a number of fundamental questions about the effectiveness of “regulation via competition” per se . Although hundreds of targets are predicted for a single miRNA , observations show that only few of them are sensitive to changes in miRNA expression levels . Most targets are likely to provide a global buffering mechanism through which miRNA levels are overall stabilized [28 , 29] . Effective competition between miRNA targets requires that the ratio of miRNA molecules to the number of target sites lies in a specific range , so that the relative abundance of miRNA and RNA species must be tightly regulated for the ceRNA mechanism to operate [29–34] . On the other hand , the magnitude of the ceRNA effect is tunable by the miRNA binding and mRNA loss rates [33 , 34 , 42] . The performance of a regulatory element , however , does not only depend on kinetic parameters , but also on the range of variability ( and possibly on the distribution itself ) of modulator levels ( e . g . TFs ) [8 , 9] . The maximal regulatory effectiveness of a given genetic circuit –quantifying how precisely the output level can be determined by the input level– can therefore generically be obtained by solving an optimization problem over the distribution of inputs . This type of approach provides an upper bound to the effectiveness of a regulatory mechanism as well as indications concerning which parameters , noise sources and/or interactions most hamper its performance . It would be especially important to understand in which conditions the degree of control of the output variable ( i . e . the ceRNA/protein level ) that can be accomplished through post-transcriptional miRNA-mediated cross-talk may exceed that obtainable by different regulatory mechanisms . In this work we characterize the maximal regulatory power achievable by miRNA-mediated control and compare it with that of a direct , TF-based transcriptional unit [45] . In principle , since fluctuations can be reduced by increasing the number of molecules , an ( almost ) arbitrary amount of information can be transmitted through a biochemical network . However , cells have to face the burden of macromolecular synthesis [46–48] . Optimality is therefore the result of a trade-off between the benefits of reduced fluctuations and the drawbacks of the associated metabolic costs . For this reason , we start by fixing a maximal rate of transcription ( or , alternatively , the maximal number of output molecules ) so as to have a simple but reasonable framework to characterize and compare the capacities of the different regulatory channels . Next , we quantify how an input signal is processed by the transcriptional ( TF-based ) and post-transcriptional ( miRNA-mediated ) regulatory elements by characterizing the response in the output ceRNA’s expression levels . In such a setting , information flow is hampered by intrinsic noise if the target gene is weakly derepressed by the activation of its competitor . Otherwise , target derepression appears to have a strong impact on a regulatory element’s capacity . Upon varying the magnitude of derepression by tuning the kinetic parameters , we then show that in certain regimes miRNA-mediated regulation can indeed outperform direct control of gene expression . Finally , we argue that the presence of miRNA molecules in large copy numbers notably reduces the level of intrinsic noise on weakly targeted transcripts . In this case , the mutual regulation of ceRNA molecules by miRNA-mediated channels may become a primary mechanism to finely tune gene expression . Besides providing a quantitative characterization of the maximal regulatory power achievable through miRNA-based post-transcriptional control , these results provide important hints on the circumstances in which regulation by small RNAs may function as the main tuner of gene expression in cells . We consider ( see Fig 1 ) a system formed by two ceRNA species ( whose levels are labeled m1 and m2 , respectively ) and one miRNA species ( with level labeled μ ) , whose transcriptions are activated by a single TF each ( with levels labeled , respectively , f1 , f2 and fμ ) . Both ceRNAs are in turn targeted by the miRNA . miRNA-ceRNA complexes ( levels labeled ci with i = 1 , 2 ) assemble and disassemble at rates k i ± , respectively , whereas complexes can be degraded both stoichiometrically ( i . e . without miRNA recycling ) at rates σi and catalytically ( i . e . with miRNA recycling ) at rates κi [49] . In addition , ceRNA and miRNA molecules degrade ( resp . synthesize ) at rates di ( resp . bi ) and δ ( resp . β ) , respectively . Steps leading to the formation of the RNA-induced silencing complex ( RISC ) , allowing for the miRNA-ceRNA binding , are neglected for simplicity . TF levels are treated as externally controlled parameters . Denoting the TF-DNA binding/unbinding rates by kin and kout , respectively ( for simplicity , these parameters are taken to be the same for all involved TFs ) , the TF binding sites’ fractional occupancies nℓ ( 0 ≤ nℓ ≤ 1 , ℓ ∈ {1 , 2 , μ} ) obey the dynamics d n ℓ d t = k in ( 1 - n ℓ ) f ℓ h - k out n ℓ , ( 1 ) according to which transcriptional activation requires the cooperative binding of h TF molecules for each RNA species involved . In general , the occupancies by different TFs equilibrate on different timescales [50] . However it is often assumed that the transcriptional on/off dynamics is much faster than transcription itself [51 , 52] . As a consequence , each nℓ can be fixed at its ‘equilibrium’ value . n ¯ ℓ = k in f ℓ h k in f ℓ h + k out . ( 2 ) Every process in the above scheme contributes to the overall level of noise . We represent the mass-action kinetics of the system through the set of coupled Langevin processes ( i = 1 , 2 ) [41 , 53] d m i d t = - d i m i + b i n ¯ i - k i + μ m i + k - c i + ξ i - ξ i + + ξ i - , d c i d t = k i + μ m i - ( k i - + κ i + σ i ) c i + ξ i σ + ξ i + - ξ i - - ξ i κ , d μ d t = - δ μ + β n ¯ μ - ∑ i k i + μ m i + ∑ i ( k i - + κ i ) c i + ξ μ - ∑ i ξ i + + ∑ i ξ i - + ∑ i ξ i κ , ( 3 ) where the mutually independent random ( Poisson ) ‘forces’ ξi , ξ i ± , ξμ , ξ i κ and ξ i σ denote , respectively , the intrinsic noise in ceRNA levels ( due to random synthesis and degradation events ) , in the association/dissociation processes of complexes , in the miRNA level , in the catalytic complex decay and in the stoichiometric complex decay . Each of the above noise terms has zero mean , while correlations are given by ⟨ ξ i ( t ) ξ i ( t ′ ) ⟩ = ( d i m ¯ i + b i n ¯ i ) δ ( t - t ′ ) , ( 4 ) ⟨ ξ i + ( t ) ξ i + ( t ′ ) ⟩ = k i + m ¯ i μ ¯ δ ( t - t ′ ) , ( 5 ) ⟨ ξ i - ( t ) ξ i - ( t ′ ) ⟩ = k i - c ¯ i δ ( t - t ′ ) , ( 6 ) ⟨ ξ μ ( t ) ξ μ ( t ′ ) ⟩ = ( δ μ ¯ + β n ¯ μ ) δ ( t - t ′ ) , ( 7 ) ⟨ ξ i κ ( t ) ξ i κ ( t ′ ) ⟩ = κ i c ¯ i δ ( t - t ′ ) , ( 8 ) ⟨ ξ i σ ( t ) ξ i σ ( t ′ ) ⟩ = σ i c ¯ i δ ( t - t ′ ) , ( 9 ) where we introduced the steady state molecule numbers m ¯ i = b i n ¯ i + k i - c ¯ i d i + k i + μ ¯ , ( 10 ) μ ¯ = β n ¯ μ + ∑ i ( k i - + κ i ) c ¯ i δ + ∑ i k i + m ¯ i , ( 11 ) c ¯ i = k i + μ ¯ m ¯ i σ i + k i - + κ i . ( 12 ) We shall be interested in the fluctuations of molecular levels around the steady state that arise due to intrinsic noise sources . The stochastic dynamics of the system can be simulated via the Gillespie algorithm ( GA , see Materials and Methods ) . Fig 2A shows typical GA results for m ¯ 1 , m ¯ 2 and μ ¯ , with the corresponding Fano factors ( FFs ) shown in Fig 2B , as functions of f1 . Numerical results are matched against analytical estimates obtained by the linear noise approximation ( see Materials and Methods ) . In particular , in Fig 2 one sees that an increase of m ¯ 1 is accompanied by a concomitant increase of m ¯ 2 and by a decrease of the average number of free miRNA molecules . This is an instance of miRNA-mediated ceRNA cross-talk . Indeed , upon up-regulating m1 by injecting the corresponding TF ( i . e . by increasing f1 ) , the level of free miRNAs will decrease as more and more molecules will be actively repressing ceRNA1 , causing in turn an up-regulation of ceRNA2 . f1 will thus positively correlate with m2 . One easily sees that steady-state ceRNA levels depend on μ ¯ through a sigmoidal function , namely [42] m ¯ i = b i n ¯ i d i F i [ μ ¯ ] , F i [ μ ¯ ] = μ 0 , i μ 0 , i + μ ¯ . ( 13 ) The constant μ 0 , i = d i k i + ( 1 + k i - σ i + κ i ) acts as a ‘soft’ threshold for the miRNA level , allowing to distinguish three situations: if μ ¯ ⪡ μ 0 , i , ceRNAi is free or unrepressed: spontaneous ceRNA degradation dominates over miRNA-mediated decay channels , so that , effectively , the ceRNA level is weakly sensitive to small changes in μ ¯; if μ ¯ ≫ μ 0 , i , ceRNAi is bound or repressed: miRNA-mediated ceRNA decay dominates over spontaneous ceRNA degradation but , again , the ceRNA level is weakly sensitive to small changes in μ ¯ as most ceRNAs are bound in complexes with the miRNA; if μ ¯ ≃ μ 0 , i , ceRNAi is susceptible to μ: spontaneous and miRNA-mediated decay channels have comparable weights and the ceRNA level is very sentitive to small changes in μ ¯ . The behaviour of the FFs ( see Fig 2B ) emphasizes how noise patterns change in the different regimes . In the displayed example , ceRNA1 and ceRNA2 become susceptible for ln ( f1 ) ≃ 2 . 5 and ln ( f1 ) ≃ 2 . 1 , respectively . Indeed , one observes that the corresponding FFs peak close to these values , in accordance with the observation that stochastic fluctuations are enhanced when the rates of substrate supply are adequately balanced in a stoichiometrically coupled system [34 , 54 , 55] . The FF for ceRNA2 appears to approach one for large values of f1 , as expected for the pure Poisson birth/death process that characterizes the free regime [56] . On the other hand , for very small but nonzero mean fractional occupancy n ¯ 1 ( corresponding to small values of f1 ) , m1 will with high probability only take on the values 0 or 1 , as a transcribed molecule will quickly undergo degradation or sequestration in a complex . In such a situation , the mean and variance of m1 will be calculated by summing up zeros and ones over time , leading to a FF equal to one . Target prediction algorithms suggest that miRNA binding affinities vary significantly among their targets [57] . Such heterogeneities indeed are mapped to the miRNA binding kinetics and are shown to influence the susceptibility of the targets to the miRNA molecules [36] . Moreover , the level of complementarity between the regulator and the target seems to be decisive for the selection of a decay channel ( catalytic or stoichiometric ) for the miRNA-ceRNA complex [15 , 27] . One may therefore expect that the effectiveness of miRNA-mediated post-transcriptional control depends strongly on the kinetic parameters characterizing the network . Here in particular , we are going to investigate how the capacities of these regulatory elements are affected by changes in ( i ) miRNA-ceRNA binding kinetics , ( ii ) miRNA recycling rates , and ( iii ) effective transcription rates of all RNA species involved . In order to contrast the performances of miRNA- and TF-channels we shall start by analyzing their respective responses to the same input signal . More precisely , the input variable fj will be varied from 0 to a value fmax defined by the condition nj ( fmax ) = 0 . 99 ( i . e . from a situation in which the promoter is always free to one in which it is essentially always occupied ) . The quality of miRNA-target interaction influences the binding kinetics and may be decisive for the activation of the target decay channel [15 , 27 , 63 , 64] . Estimations of the miRNA-ceRNA binding affinities are experimentally challenging . However , computational methods predict a considerable degree of heterogeneity across different miRNA-ceRNA pairs [57] . Remarkably , for the majority of cases reported in the literature , the predicted miRNA-binding energies of the RNAs on the ‘input side’ of the channel are lower than those of the RNAs on the ‘output side’ , in line with the optimal conditions highlighted by our model . For example , binding affinities between the long non-coding RNA linc-MD1 and its regulatory miRNAs ( playing a central role in skeletal muscle cell differentiation ) are significantly lower than those characterizing miRNA interactions with linc-MD1’s competitors , namely the MAML1 and MEF2C mRNAs , as predicted by miRanda algorithm [39 , 64] . Likewise , the circular RNA CDR1 has been found to contain around 70 binding sites with high complementarity for miR-7 , corresponding to a strong effective coupling through which it can regulate the expression of miR-7’s target genes [37 , 65 , 66] . Finally , the high sequence homology of pseudogenes ( long non-coding RNA genes developed from protein-coding genes but unable to produce proteins ) with their parental gene allows them to compete for a large number of shared miRNAs [28 , 38 , 67] . Our study also points to the potential relevance of miRNA-ceRNA complex decay channels . It is known that , in case of sufficient complementarity , miRNA molecules can function as siRNAs and cleave their targets after binding them [15 , 27] . Such targets decay catalytically and are therefore effectively degraded by the miRNAs . Our model predicts that miRNA-mediated control may be the preferred regulatory mechanism in presence of kinetic heterogeneities at the level of miRNA recycling rates . In particular , targets that undergo catalytic degradation may be efficiently derepressed by their competitors . This type of scenario has been observed in experiments concerning the bacterial small RNA Qrr [61] . Qrr represses its targets by distinct mechanisms . For instance , luxR is repressed catalytically , luxM stoichiometrically , while luxO is silenced through translational repression . luxM and luxO are however able to derepress LuxR in the presence of Qrr . In the light of our results , these observations may therefore point to a higher than expected role for the ceRNA effect in vivo , especially in cases in which heterogeneities in kinetic parameters are thought to be strong [34 , 59 , 62 , 68 , 69] . Non-coding RNA molecules , and miRNAs specifically , are increasingly associated to regulatory functions . Besides making it mandatory to characterize the specific role of ncRNAs on a case-by-case basis , especially for situations like disease or differentiation , this fact also raises the question of what ingredients can make miRNAs a preferred tool to regulate the level of a target RNA over , for instance , the target’s TF . A possible answer lies in the noise-buffering role that miRNAs can play , which is especially evident in genetic circuitries like incoherent feed-forward loops [19 , 20] . By reducing relative fluctuations in the output level , miRNAs can confer robustness to gene expression profiles . However the so-called ‘ceRNA hypothesis’ opens the way to the possibility that their regulatory functions are carried out at a broader , though more subtle , level . In short , according to the ceRNA scenario miRNAs can mediate an effective positive interaction between their target RNAs driven by the targets’ competition to bind them . In this sense , miRNAs can be seen as a sort of ‘channel of communication’ between RNAs through which RNA levels can be altered and noise can be processed ( both buffered and amplified ) . Previous work [42] has shown that the ceRNA effect may generate both highly plastic and highly selective ceRNA-ceRNA couplings , thereby representing a potentially powerful mechanism to implement gene regulation at the post-transcriptional level . Although predicted theoretically , the extent and relevance of ceRNA effect in vivo is poorly understood . On one hand , considerable evidence points to the ceRNA effect playing a major role in certain dis-regulated or transient cellular states . For instance , it has been shown that the expression of the tumor-suppressor gene PTEN can be regulated by its miRNA-mediated competitors VAPA , CNOT6L , SERINC1 or ZNF460 [70] . Furthermore , many pseudogenes have been found to compete with their parental genes for a shared pool of common microRNAs , thus regulating their expression as competitive endogenous RNA [28 , 38 , 67 , 71] . Such mechanisms seem to be of particular relevance in cancer . For instance , murine models engineered to overexpress the pseudogenes of the proto-oncogene BRAF develop an aggressive malignancy resembling human B cell lymphoma since , by functioning as ceRNAs , they elevate BRAF expression both in vitro and in vivo [38] . Likewise , the long noncoding RNA linc-MD1 has been shown to regulate the skeletal muscle cell differentiation clock by sponging miRNAs from its competitors , thereby enacting a ceRNA mechanism . In particular , MAML1 and MEF2C ( coding for transcription factors that activate muscle-specific gene expression ) compete with linc-MD1 for miR-133 and miR-135 respectively [39] . Taken together , the available evidence indicates that miRNA activity depends on the miRNA:target ratio , on miRNA target site abundance and on miRNA binding affinities . Further analyses of high throughput datasets confirm this observation [30 , 36] . One may therefore question how said factors may influence miRNA-mediated post-transcriptional control . The problem however arises of quantifying the degree of control that can be exerted through miRNAs . Taking the ‘channel’ analogy more strictly ( as done before for simpler regulatory elements [9 , 45 , 52] ) , one may resort to information theoretic concepts and tools to characterize precisely how well miRNAs can process fluctuations coming from the modulator nodes and transfer them to the target nodes . This issue goes beyond noise buffering , specifically including the ability to respond to large changes in the mean levels as well as to changes in the structure of fluctuations . As the properties of a channel are conveniently encoded in the mutual information between the input and output nodes , asking how well a channel can function amounts to asking what is the channel’s capacity , i . e . the maximum value of the input-output ( or modulator-target ) mutual information achievable through that channel . This work aimed precisely at quantifying the effectiveness of microRNA-mediated post-transcriptional control of gene expression by computing the capacity of the corresponding regulatory channel and comparing it to that of direct , TF-driven transcriptional regulation . Evidently , multiple factors can influence the flow of information across nodes in a biochemical network , starting from the intrinsic noisiness of each reaction step . Our basic challenge was therefore understanding in which circumstances miRNA-mediated control can outperform the TF-based one , thereby obtaining insight on why the ceRNA effect appears to be so often employed by cells in situations where accurate tuning and/or shifts of expression levels are required . We have therefore considered , along the lines of [33 , 34 , 41–44] , a mathematical model of the ceRNA effect and characterized its steady state in terms of both mean molecular levels and regulatory capacities of the miRNA-mediated and TF-based channels via stochastic simulations . We have first considered how the two channels process inputs ( the TF levels f1 and f2 ) that vary in the same range . We have shown that , while the capacity of the TF-channel depends monotonously on each miRNA-ceRNA binding rate and is largest when the target is unrepressed by miRNAs ( as might have been expected ) , the capacity of the post-transcriptional channel is maximal in a specific range of values of the miRNA-ceRNA binding rates . In agreement with [33] , we found that miRNA-channel’s efficiency is tunable to optimality by the binding kinetics . Furthermore , our model suggests that both capacities decrease as the miRNA recycling rates increase , confirming previous indications obtained by different analytical techniques [42] . Consistently with the scenario observed experimentally for the bacterial small RNA Qrr [61] , our model finally suggests that catalytically regulated targets are weakly capable of competing for miRNAs but might be significantly derepressed by their competitors . In addition , post-transcriptional miRNA-mediated information processing was shown to be characterized by a threshold behaviour as a function of the AOV . In other terms , no information can be transmitted across the channel unless the target’s degree of derepression is sufficiently large . This implies that the regulatory effectiveness of the channel is well encoded by the degree of target derepression when the latter is sufficiently high , in which case it is possible to identify regimes in which post-transcriptional regulation is more accurate than transcriptional control . To get a deeper insight on the origin of the observed threshold behaviour one must however go beyond the AOV and consider more carefully how the miRNA-ceRNA binding noise affects the overall picture . After showing that miRNA-ceRNA binding noise is indeed at the origin of the threshold behaviour that limits the miRNA-channel capacity , we have uncovered the rather remarkable property that in presence of large but weakly interacting miRNA populations the ceRNA effect can regulate gene expression as effectively as the target’s modulator node itself . The present work has focused on a small genetic circuit made up of a single miRNA species and two target RNA species at steady state . Previous work has however shown that cross-talk is possible even during transients [44] . Going beyond stationarity is therefore likely to bring to light new scenarios where miRNA-mediated regulation plays possibly a yet more prominent role . On the other hand in a typical eukaryotic cell there are thousands ceRNAs , hundreds miRNAs and a rich structure of conserved targeting patterns [72] . Moreover , cells might be interested in tightly controlling not only each output individually but also particular combinations of output levels ( which might be required e . g . for the efficient operation of metabolic pathways ) . In such a scenario , miRNA-mediated control could represent a powerful mechanism to increase robustness and flexibility in specific directions of the output space . In view of this , it would be important to consider a more general multi-source network coding problem in which a large number of transcription processes are seen as mutually independent information sources , and each of the information sources is multicast to sets of output nodes through the effective network of miRNA-mediated cross-talk interactions . The information-theoretic scheme employed in this work is easily generalized to deal with more complex networked situations . Novel insight might finally shed light on the partly controversial picture unveiled by recent experiments addressing the relevance of the ceRNA effect in vivo [30 , 35–39] . Numerical simulations have been carried out using the Gillespie algorithm ( GA ) , a standard stochastic method to analyze the time evolution of a system of chemical reactions which is exact for spatially homogeneous systems [73] . In short , based on the reaction rates , GA calculates when the next reaction will occur and what reaction it will be , and then modifies the amount of each molecular species in the system according to the process that took place . If we denote the probability of reaction r to occur in the time interval ( τ , τ + dτ ) by P ( r , τ ) dτ , the algorithm proceeds through the following steps: Initiate the number of reactants in the system and the termination time; Generate a random pair ( r , τ ) according to P ( r , τ ) ; Using the pair ( r , τ ) just generated , advance time by τ and change number of species involved in reaction r accordingly; Read out the molecular population values . If the termination time is reached , stop the simulations , otherwise return to Step 2 . After a long run , independently on the initial setup , the system of chemical reactants will come to the equilibrium state . The mathematical model of ceRNA competition can be solved numerically in the so-called linear noise approximation . Letting x = ( m1 , m2 , μ , c1 , c2 ) stand for the vector of molecular levels , the kinetic mass action Eq ( 3 ) can be re-cast in compact form as d x d t = g ( x ) + η , ( 25 ) where the vector g encodes for the deterministic part of the dynamics , while the vector η represents the aggregate noise terms . Each element of η has zero mean , and we shall denote its correlations by 〈ηa ( t ) ηb ( t′ ) 〉 = Γab δ ( t − t′ ) . Denoting by x ¯ the steady state , small deviations from it ( i . e . δ x = x ( t ) - x ¯ ) relax , in the linear regime , according to d d t δ x = A δ x + η , ( 26 ) where A = d g d x \| x = x ¯ . In this approximation , the correlation matrix Cab = 〈δxa δxb〉 is given by [74] C a b = - ∑ p , q , r , s B a p B b r Γ q s λ p + λ r ( B - 1 ) p q ( B - 1 ) r s , ( 27 ) where λ’s and B’s are , respectively , eigenvalues and eigenvectors of the matrix A . Table 1 reports the values of the kinetic parameters ( or of their range of variability ) employed in the different figures . As k 1 + , k 2 + , κ 1 , κ 2 , b 1 , b 2 , b μ and dμ are varied in wide ranges in order to test how channel capacities depend on them , our choice was mainly guided by the need to focus the analysis on regimes where ceRNA cross-talk is established ( so that the miRNA-mediated channel can actually convey information ) .	The discovery of RNA interference has revolutionized the decades’ old view of RNAs as mere intermediaries between DNA and proteins in the gene expression workflow . MicroRNAs ( or miRNAs ) , in particular , have been shown to be able to both stabilize the protein output by buffering transcriptional noise and to create an effective positive interaction between the levels of their target RNAs through a simple competition mechanism known as ‘ceRNA effect’ . With miRNAs commonly targeting multiple species of RNAs , the potential implication is that RNAs could regulate each other through extended miRNA-mediated interaction networks . Such cross-talk is certainly active in many specific cases ( like cell differentiation ) , but it’s unclear whether the degree of regulation of gene expression achievable through post-transcriptional miRNA-mediated coupling can effectively overcome the one obtained through other mechanisms , e . g . by direct transcriptional control via DNA-binding factors . This work quantifies the maximal post-transcriptional regulatory power achievable by miRNA-mediated cross-talk , characterizing the circumstances in which indirect control outperforms direct one . The emerging scenario suggests that , in addition to its widely recognized noise-buffering role , miRNA-mediated control may indeed act as a master regulator of gene expression .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "channel", "capacity", "gene", "regulation", "regulatory", "proteins", "dna-binding", "proteins", "dna", "transcription", "micrornas", "post-transcriptional", "gene", "regulation", "transcription", "factors", "computer", "and", "information", "sciences", "transcriptional", "control", "proteins", "gene", "expression", "biochemistry", "rna", "nucleic", "acids", "genetics", "information", "theory", "biology", "and", "life", "sciences", "non-coding", "rna" ]	2016	Probing the Limits to MicroRNA-Mediated Control of Gene Expression
In Enterococcus faecalis , sex pheromone-mediated transfer of antibiotic resistance plasmids can occur under unfavorable conditions , for example , when inducing pheromone concentrations are low and inhibiting pheromone concentrations are high . To better understand this paradox , we adapted fluorescence in situ hybridization chain reaction ( HCR ) methodology for simultaneous quantification of multiple E . faecalis transcripts at the single cell level . We present direct evidence for variability in the minimum period , maximum response level , and duration of response of individual cells to a specific inducing condition . Tracking of induction patterns of single cells temporally using a fluorescent reporter supported HCR findings . It also revealed subpopulations of rapid responders , even under low inducing pheromone concentrations where the overall response of the entire population was slow . The strong , rapid induction of small numbers of cells in cultures exposed to low pheromone concentrations is in agreement with predictions of a stochastic model of the enterococcal pheromone response . The previously documented complex regulatory circuitry controlling the pheromone response likely contributes to stochastic variation in this system . In addition to increasing our basic understanding of the biology of a horizontal gene transfer system regulated by cell-cell signaling , demonstration of the stochastic nature of the pheromone response also impacts any future efforts to develop therapeutic agents targeting the system . Quantitative single cell analysis using HCR also has great potential to elucidate important bacterial regulatory mechanisms not previously amenable to study at the single cell level , and to accelerate the pace of functional genomic studies . Enterococci are major contributors to the current antibiotic resistance crisis [1] . They are among the most common agents of antibiotic-resistant nosocomial infections , and their conjugative mobile genetic elements contribute to rapid intra- and intergenic horizontal transfer of resistance determinants [2–7] . Transfer of the tetracycline-resistance conjugative plasmid pCF10 between E . faecalis cells is controlled by two antagonistic signaling peptides ( Fig 1A ) [8] . A secreted pheromone cCF10 ( C; sequence LVTLVFV , originally termed “clumping-inducing” since it induces formation of visible cell aggregates ) is produced by plasmid-free recipients and a secreted inhibitor peptide iCF10 ( I; sequence AITLIFI ) is encoded by the prgQ gene in the pCF10 plasmid [9–12] . In the absence of C , the prgQ promoter ( PQ ) is repressed and basal transcription from the PQ promoter terminates approximately 400 nt from the transcription start site at IRS1 , ( an inverted repeat sequence and terminator ) , resulting in a “short Q” ( QS ) RNA . Induction of conjugation occurs when C is imported into plasmid carrying cells , where it binds to the master transcription regulator PrgX [13] and prevents repression of transcription from PQ by PrgX . Increased levels of prgQ RNAs override counter-transcript-mediated attenuation at IRS1 to produce “long Q” transcripts ( QL ) that can extend through the entire prgQ operon ( Fig 1A and S1 Fig ) [8 , 14] . QL RNAs are virtually undetectable in uninduced cells , but upon exposure to C they increase in a dose-dependent fashion [15 , 16] and result in expression of downstream conjugation genes . Induction of the prgQ operon increases the production of the inhibitor peptide I which competes with C for binding to PrgX; PrgX-I complexes increase PrgX repression of PQ ( Fig 1A and 1B ) [12 , 17 , 18] . While the basal levels of I produced from QS transcripts help prevent spurious induction in the absence of recipients , the increase in I following induction is essential for the rapid shut-down of the response [16] . The identification of multiple layers of positive and negative feedback loops operating in the pCF10 system and quantitative analysis of prgQ expression in populations of donor cells suggested that the system could function as a bistable switch [19 , 20] . Conjugative transfer of pCF10 in E . faecalis is population composition dependent . At low donor densities relative to recipients , the conjugation operon is induced at high levels and results in high conjugation frequencies . In contrast , high donor densities relative to recipients result in decreased induction of the conjugation operon and reduced conjugation frequencies [16] . This calibrated response by the donor population likely increases their fitness by reducing energy expenditure and potential deleterious effects of induction of conjugation on donor viability [21 , 22] when there is already a large population of donors carrying the plasmid . However , a reduced response to induction limits the opportunity for plasmid transfer to new recipients , which may be beneficial to both the plasmid and the new host . Hence , we hypothesized that the stochasticity in the enterococcal pheromone response allows for induction in a sub-population of donors even when the pheromone concentrations are insufficient to induce high levels of conjugation . Stochasticity will be reflected in the heterogeneity of the donor response to induction by C and will lead to a small number of cells responding to a very low concentration of C that is below the threshold for induction for the majority of cells . Many past studies have characterized the pheromone response at the population level and revealed the dynamics of the response [8] , but these studies were unable to detect heterogeneity of induced expression in individual cells that could have functional relevance [23] . In this study , we exposed populations of cells to pheromones and examined the induced response within single cells using direct transcript labeling and a fluorescent reporter system . These experimental methods , combined with mathematical modeling , demonstrated that the response is stochastic and heterogeneous . The stochastic nature of pheromone induction may directly account for the heterogeneity in the response that was observed and the occurrence of conjugation in the presence of low inducing pheromone concentrations and high inhibiting pheromone concentrations . A unique feature of cell-cell signaling in the enterococcal sex pheromone systems is the involvement of two antagonistic peptide signals , the chromosomally-encoded plasmid transfer inducing peptide C ( mate-sensing signal ) , and the plasmid-encoded transfer inhibiting peptide I ( self-sensing signal ) . We have shown that extracellular levels of I produced by uninduced donor cultures accumulate in a density-dependent fashion , functioning as a classic quorum sensing signal of donor population density [12] . C-mediated induction is drastically reduced at high donor density , as determined by quantitative analysis of transcription in donor cultures . However , the notion that I can completely shut down induction is called into question by experiments looking at the conjugation efficiency of donor cultures exposed to various concentrations of C and high concentrations of I ( Fig 2 ) . In this experiment , donor cells were incubated with synthetic peptides for 30 minutes and then mixed with recipients for 10 minutes prior to plating on medium selective for transconjugants . This short mating time is only sufficient to allow the donors induced during the pre-incubation to undergo one round of conjugation . Under these very stringent mating conditions , fewer than 1 transconjugant per 106 donors was produced when no exogenous peptides were added . At the other extreme , the frequency rose to nearly 10 , 000- fold ( to nearly 1 transconjugant per 10 donors ) when the donors were pre-exposed to a saturating concentration of 50 ng ml-1 C . Exposure to intermediate C concentrations in the 0 . 3 to 2 . 5 ng ml-1 range ( more closely resembling the levels of C naturally produced by recipient cells [12] ) resulted in a jump to levels of transfer about 1 , 000-fold above uninduced cells , but well below the maximum . Most striking was the observation that induction with 2 . 5 ng ml-1 C in the presence of a great excess of I ( 50 ng ml-1 ) still increased transfer by a factor of 10 . These data suggest that a very small number of donors may be induced to transfer under conditions where the vast majority of the population is uninduced . Confirmation that these results reflected a variable transcriptional response among this small subpopulation required analysis of the pheromone response at the single cell level . Fluorescence in situ hybridization chain reaction ( HCR ) was recently developed to detect specific transcripts in a range of eukaryotic organisms [24] and in microbial symbionts of termites and the bobtail squid [25 , 26] . We adapted the technology to quantitatively analyze the dynamics and variability in the levels of pheromone-inducible transcripts of E . faecalis at the single cell level ( S2 Fig ) . Initially , we used an E . faecalis strain carrying pBK2 , which contains a pheromone-inducible lacZ reporter for QL expression , along with the native prgX and prgQ genes and promoters in their native configurations ( Fig 1B ) . We used HCR probes against lacZ mRNA ( paired with HCR amplifiers ) to assay transcript levels ( S1 Table ) . The lacZ reporter allowed for rapid independent confirmation of the induction state of cultures . Simultaneous labeling of transcripts from the constitutively expressed chromosomal gene ptsI was also performed by HCR ( S1 Table ) . The fluorescent HCR signal of lacZ mRNA in cells induced for 30 min with a high concentration of C was very strong in most cells ( Fig 3A ) compared to that of uninduced cells ( Fig 3B ) . Virtually all cells , both induced and uninduced , labeled strongly with cell envelope and nucleoid stains , as well as with the HCR probes for ptsI mRNA , suggesting that our labeling protocol ( S3 Fig ) effectively permeabilized cells to the HCR probes without lysis . The rare green fluorescent particles observed in preparations of uninduced cells were of low intensity and generally localized outside of cells ( Fig 3B and S4 Fig ) . These results indicated the feasibility of using HCR for more detailed quantitative single-cell analysis of multiple transcripts in E . faecalis . Cells carrying a pheromone-inducible lacZ reporter ( the pBK2 plasmid ) ( Fig 1B ) were induced with a high concentration of C , fixed at a range of time points , and assayed for lacZ expression by HCR ( Fig 4A ) . Notably , lacZ mRNA was detected within 15 minutes after C addition , and reached maximum levels within 30 to 60 minutes . By 120 minutes , the number of cells expressing lacZ was significantly reduced ( Fig 4A ) . These results reflect the temporal response that has been observed in the past using population- based methods such as qRT-PCR , RNAseq , and microarray [16 , 27] . The induction process was also examined using an inducible GFP reporter ( pCIE-GFP; Fig 1B ) , and the onset of induction was similarly observed . GFP fluorescence was seen approximately 60 min after C addition ( Fig 4B and S1 Movie ) . The delayed response compared to HCR is likely caused by the time required for protein folding after translation . Due to the inherent stability of the GFP protein , a decrease in fluorescence was not observed even though the analogously induced lacZ transcript level had decreased as seen with HCR . Using the GFP construct further allowed for tracking gene expression in individual cells over time using time-lapse microscopy which is not possible using HCR as it requires fixed cells . Hence , the two methods were complimentary in revealing the population dynamics of induction by C . Both methods showed heterogeneity in that some cells became induced earlier than others and that the signal intensities were not the same in every induced cell . We quantified the fluorescent intensity of HCR or GFP signals of individual cells at various time points following C addition ( Fig 4C and 4D and S5 , S6 , S7 and S8 Figs ) . An uninduced cell population ( 0 ng ml-1 C ) was used to establish a strict threshold which was then used to categorize cells as induced or uninduced at different time points ( Fig 4C and 4D ) . Among the uninduced population , a very small number of cells ( 0 . 3% ) had a high HCR fluorescence intensity , likely due to artifacts of the labeling method . After removing those outliers , the maximum intensity of the uninduced population was used as the threshold . The intensity of induced cells was normalized to the threshold intensity . Using a strict threshold helped distinguish genuine response from noise , but may also have resulted in an elevated number of false negative cells . With both GFP and HCR , the signal intensity of individual cells at any given time point varied widely ( Fig 4C and 4D ) . As time progressed the fraction of cells which were induced increased . We had previously developed an ordinary differential equation ( ODE ) -based mathematical model describing the induction process [16] . In light of new knowledge on the oligomeric states of apo-PrgX and PrgX/peptide complexes [28] , availability of kinetic parameters , and a greater mechanistic understanding of its interactions with DNA , the model was refined ( S1 Fig ) . A stochastic model based on our updated model was developed using Hy3S [29] which consists of 15 species and 38 biochemical reactions and interactions involved in the induction of pCF10 ( S1 Text and S2 Table ) . The QL transcript level in response to C was simulated in 10 , 000 individual cells ( Fig 4E ) . The simulated time dynamics of QL behaved similarly to experimentally observed HCR and GFP profiles . The fractions of induced cells increase over time and after 150 minutes the population of highly induced cells shows a reduction indicating response shut-down . We varied the concentrations of C from 0 . 625 ng ml-1 to 10 ng ml-1 and used HCR to examine the fraction of cells induced over time ( Fig 5A ) . From past studies , we expected the population-averaged QL transcript level to increase to a higher level with increasing C and decrease rapidly after reaching a peak [19] . Using HCR we saw the fraction of cells induced increase with increasing C concentrations ( Fig 5A ) . We further designed probes to the prgB transcript and directly probed the dynamics of the QL transcript of conjugation in cells harboring the wild type pCF10 plasmid ( Figs 5B and 1A and S1 Table ) . The induction profile of prgB from pCF10 was similar in magnitude and time course to lacZ from pBK2 ( Fig 5A and 5B ) . The faster increase and subsequent decrease in induction observed with pCF10 likely relates to increased import of the C and I peptides due to the presence of the pCF10 PrgZ peptide binding protein [30] . However , the overall similarity suggests that pBK2 contains the key regulatory components required for heterogeneous induction of pCF10 . Additionally , flow cytometry analysis of the lacZ and prgB HCR fluorescence of our samples showed a similar induction profile ( S9 Fig , S3 Table and S1 Methods ) . The heterogeneous response to varying concentrations of C is not only reflected in the fraction of cells induced ( i . e . having a fluorescence level above the threshold ) , but also in the transcript level among those induced . Variation in transcript level can be seen in the plot of HCR signal intensity distribution of the induced cells at 30 minutes post-induction ( Fig 5C ) . Using qRT-PCR measurement ( a population averaging assay ) , we had previously reported induction at 1 ng ml-1 C [16] . Using HCR we followed induction on the single cell level and saw that even at C concentrations as low as 1 . 25 ng ml-1 a few cells were induced to levels seen in populations exposed to higher C concentrations ( Fig 5C ) . Samples of these rare , highly induced cells were readily visualized ( Fig 5D ) and this small fraction of highly induced cells at 1 . 25 and 2 . 5 ng ml-1 of C subsequently subsided over a similar time duration to that observed with higher C concentrations ( Fig 5E and 5F ) . The observation of a few highly induced cells at low pheromone concentrations where the vast majority of cells remained uninduced suggests that stochasticity is at play in this system . We next tracked the induction at different C concentrations using the GFP reporter and time lapse microscopy . The time point that each cell became induced ( the first time point at which GFP fluorescence was observed ) after exposure to C was determined and plotted ( Fig 6A ) . At a high inducer concentration of 50 ng ml-1 , the earliest induction occurred around 40 minutes after C addition , while the median induction time was around 90 minutes . At later time points , when the majority of cells had been induced , the frequency of newly induced cells decreased . At lower C concentrations of 2 . 5 and 5 ng ml-1 , the median time to initial induction shifted to longer time points , and even at 160 minutes cells continued to be induced at a relatively high rate . Importantly , even though the population median induction time was delayed in low versus high C concentrations , the time at which the first cells became induced was similar irrespective of C concentration ( around 40–80 minutes as determined by GFP expression ) ( Fig 6A ) . When we used the stochastic model to simulate the response of cells to varying concentrations of C , the model ( Fig 6B ) predicted the behavior observed experimentally using GFP ( Fig 6A ) . Further , in the presence of high concentrations of I ( 50 ng ml-1 ) , reflecting the conditions that occur at high donor densities , low C concentrations ( 2 . 5 ng ml-1 ) still result in early responding cells at a low frequency ( Fig 6A and 6B ) . Stochasticity in the induction response resulted in early induction of some donor cells , even under low C conditions unfavorable to induction ( Fig 6 ) , and allowed a small subpopulation of cells to undergo conjugation while the vast majority of cells remained in an uninduced state ( Fig 2 ) . The occurrence of rare conjugation events in a bacterial community with a high density of plasmid-containing cells may help spread the plasmid to recipient cells which may carry other critical traits . We hypothesize that stochasticity in molecular interactions in the pheromone response circuit result in heterogeneity and stochasticity in the population response to induction . At the single donor cell level , the concentrations of C and I are critical determinants of the induction state . With extracellular concentrations in the nM range , the intracellular levels of C and I are a few molecules per cell , as are PrgX complexes with C and I [12] . There are about 5 copies of the pCF10 plasmid per cell , and each copy contains a single functional operator target for PrgX regulation . Therefore , the intracellular levels of the critical molecular species controlling the pCF10 pheromone response are in the range of a small number of molecules per cell . The induction of the conjugation system is driven by a series of molecular events ( Fig 1 ) involving import of the peptides C and I , their binding with PrgX to form PrgX-peptide tetramers , and their interaction with the operator site which regulates the access of RNA polymerase to the PQ promoter [28] . The probability of each molecular event depends upon the concentrations of the molecular entities involved . The low intracellular levels of these entities can magnify stochastic behavior , causing individual cells to have different numbers of each of these molecules per cell and thus different outcomes . Even under unfavorable induction conditions of low C and high I , the probability of a PrgX-C complex displacing the repressing PrgX or PrgX-I complexes from the operator ( and leading to induction ) is low but non-zero . Hence , stochasticity can lead to rare induction even under unfavorable conditions . Since the intracellular concentrations of the peptides are one of the major sources of stochasticity , the rate of peptide import by the donor could significantly impact the induction and conjugation response . In the case of wild type donors , this process is mediated by the cooperative functions of the plasmid-encoded PrgZ pheromone binding protein and the chromosomal oligopeptide permease [30] . Thus , examination of peptide import at the single cell level could provide new insights into the induction process . A heterogeneous pheromone response may benefit the bacterial community by balancing the benefits of dissemination of potentially beneficial plasmid genes with fitness costs due to expression of the large set of genes required for conjugative transfer [21 , 22] . Our results may contribute to an improved understanding of the evolutionary significance of stochastic variation in the enterococcal pheromone response and other microbial communication systems . In future work , it would be worthwhile to examine the transcription of additional genes in the distal segments of the long pheromone-inducible operon ( Fig 1A ) and quantify actual conjugation events at the individual cell level . Variability in the induction of downstream genes among the cells induced for expression of proximal conjugation genes would suggest the possibility that only some individuals in an “induced” population subset actually function in plasmid transfer . In this division-of-labor model , the remaining induced donors could contribute to cooperative behaviors such as formation of cell aggregates , and to the rapid shut down of the pheromone response by production of I . A better understanding of the stochastic behavior of this system may also benefit the development of novel therapeutic approaches for the prevention or treatment of opportunistic infections by multi-drug resistant enterococci . It might be possible to manipulate pheromone signaling to alter the population balance of commensal and resistant pathogenic strains as a therapeutic strategy [31] . However , stochastic variation affecting antibiotic resistance plasmid transfer should bring caution to such strategies . Any therapeutic agents that target bacteria responding to the normal peptide signals may affect the majority of the population , but fail to eradicate unresponsive bacteria , analogous to the function of persister cells in biofilm-associated antibiotic resistance [32] . Finally , with the ability to determine the transcript level of any gene using specific probes , HCR should facilitate quantitative analysis of the expression dynamics of many genes at the single cell level for many different microorganisms in addition to enterococci . The strains and plasmids used are listed in S4 Table . All bacterial strains in this study were derived from E . faecalis strain OG1 [33] and the endogenous conjugative plasmid pCF10[34 , 35] . OG1RF is an OG1 derivative with rifampicin and fusidic acid resistance [34 , 36] . OG1Sp is an OG1 derivative with spectinomycin resistance [37] . JRC104 is an OG1RF derivative that does not produce the cCF10 peptide due to a nonsense point mutation in the cCF10 encoding ccfA gene [37] . pBK2 contains the pCF10 prgX-Q regulatory region with a lacZ fusion downstream of IRS1 [38] , while pCIE-GFP was similarly constructed but with a gfp fusion [39] . E . faecalis strains were grown statically at 37°C in M9 medium containing 3 g l-1 yeast extract , 10 g l-1 casamino acids , 36 g l-1 glucose , 0 . 12 g l-1 MgSO4 , and 0 . 011 g l-1 CaCl2 . M9 medium was supplemented with 20 μg ml-1 chloramphenicol in overnight cultures of JRC104+pCIE-GFP to ensure maintenance of the plasmid . Antibiotic concentrations used in the mating experiment were 250 μg ml-1 for spectinomycin , 200 μg ml-1 for rifampicin , and 25 μg ml-1 for fusidic acid . E . faecalis cultures of donor ( OG1Sp+pCF10 ) and recipient cells ( JRC104 ) were grown overnight at 37°C in 3 ml of M9 medium . Overnight cultures were centrifuged , washed twice with 1 ml KPBS containing 2 mM EDTA , and diluted 1:5 in fresh M9 medium . The cultures were incubated for 1 hour at 37°C and various concentrations of the C pheromone ( cCF10 ) and I inhibitor ( iCF10 ) were added to the donor culture followed by 30 minute incubation at 37°C . The donors and recipients were then mixed in 1:1 ratio and mating was carried out at 37°C for 10 minutes . Serial dilutions of these samples were plated on selective THB agar medium containing 30 g l-1 Bacto Todd Hewitt Broth ( Becton , Dickinson and Company ) and 15 g l-1 agar to enumerate the donors ( spectinomycin and chloramphenicol resistant ) , recipients ( rifampicin and fusidic acid resistant ) , and transconjugants ( rifampicin , fusidic acid , and chloramphenicol resistant ) . Overnight cultures were sub-cultured 1:10 in M9 medium and grown to early exponential phase ( ≈3 h to OD600 ≈ 1 . 2 ) . Cultures were induced with between 0 . 625 ng ml-1 and 10 ng ml-1 cCF10 ( C ) peptide and cells were harvested at times 0 to 180 minutes after C addition . Fixation of cells began immediately upon harvest when they were mixed 1:1 with EM-grade 8% paraformaldehyde ( PFA; 4% final concentration ) and fixed for >20 h at 4°C . Following fixation , cells were isolated by centrifugation at 13 , 000×g for 5 minutes and subsequently resuspended in KPBS with trace RNaseOUT ( Invitrogen ) . Nucleic acid probes and hairpin amplifier sequences used in this study were obtained from Molecular Instruments ( www . molecularinstruments . org ) , ( S1 Table ) . To label transcripts ( ptsI , lacZ , and prgB ) in E . faecalis , established HCR methodology [24] was adapted to label 20 μl aliquots of cells in suspension at a time [25 , 26 , 40–43] . Briefly , cells were permeabilized , DNA probes were hybridized to transcripts of interest , fluorescent amplifier hairpins were hybridized to the bound probes , and subsequently cells were counterstained and mounted for microscopy . Between each re-suspension step , cells were pelleted by centrifugation at 13 , 000×g for 2 minutes and re-suspended in the subsequent buffer or reagent . Our HCR protocol was as follows . For permeabilization , cells were suspended in 20 μl of permeabilization buffer containing 5 mg ml-1 lysozyme ( Sigma-Aldrich ) , 0 . 1 M Tris/ HCl , 0 . 05 M EDTA , and trace RNaseOUT and incubated at 37°C for 3 h . After permeabilization , cells were suspended in 20 μl KPBS to wash and then prehybridized in 20 μl of the probe hybridization buffer ( Molecular Instruments ) for 30 minutes at 45°C . Next , cells were suspended in 20 μl of probe hybridization buffer preheated to 45°C and containing 2 nM of each probe to the transcripts of interest ( 5 to 6 probes per transcript ) then incubated at 45°C for >20 h . Following probe hybridization , cells were washed twice in 20 μl wash buffer ( Molecular Instruments ) for 30 minutes each at 45°C . Cells were then suspended in 20 μl amplification buffer ( Molecular Instruments ) for 30 minutes at room temperature for pre-amplification . Meanwhile required amplifier hairpins were heated in individual tubes to 95°C for 90 seconds in a thermocycler and then cooled to room temperature in a dark drawer for approximately 30 minutes . After pre-amplification , cells were suspended in 20 μl of amplification buffer containing 60 nM of each hairpin as appropriate and incubated in the dark at room temperature for >20 h . Probe sequences and amplifier details for each transcript of interest can be found in S1 Table . Following amplification , cells were washed twice in 20 μl 5× sodium chloride sodium citrate , Tween 20 ( 5× SSCT ) ( Molecular Instruments ) in the dark for 30 minutes each at room temperature . To counterstain , cells were suspended in solutions of Hoechst 33342 nucleic acid stain ( Thermo Fisher ) and Alexa Fluor 647: wheat germ agglutinin ( WGA ) conjugate to label the cell envelope ( Invitrogen ) . Cells were washed in KPBS , suspended in ddH2O , and then 10 μl of each suspension was applied to 22×22 mm No . 1 . 5 coverslips ( Gold-Seal ) to dry , and then mounted in hardening Prolong Diamond Antifade Mountant ( Molecular Probes by Life Technologies ) . Mountant was allowed to harden for >48 h at 4°C before imaging . Images shown in Fig 3 were taken using a Zeiss Axio Observer . Z1 confocal microscope equipped with an LSM 800-based Airyscan super-resolution detector system ( Zeiss ) . Confocal images were acquired through a 63× , 1 . 40- numerical aperture ( NA ) objective ( Zeiss ) in Airyscan mode . Images in Fig 3A were acquired as z stacks at 0 . 15- μm intervals , deconvolved through Airyscan processing , flattened using a maximum intensity projection with Ortho Display , and presented for publication as a Min/Max projection using Zen software ( version 2 . 1 , Zeiss ) . Images in Fig 3B were acquired in Airyscan mode at a single z plane , deconvolved , and similarly presented for publication as a Min/Max projection using Zen software . In this Min/Max projection , the Alexa Fluor 488 channel ( corresponding to HCR labeled lacZ transcripts ) appears to have some small bright puncta that appear markedly different from the HCR signal and rarely overlap with cells . While these puncta appear bright , their actual intensity is minimal in comparison to true HCR signal intensity as demonstrated in S4 Fig . Images shown in Figs 4A and 5D were taken using a Nikon E-800 microscope equipped with a spinning disc BD CARV II confocal image adapter ( BD Biosciences ) . A Cascade 1k EMCCD camera ( Photometrics ) was used to acquire images as wide-field z stacks at 0 . 2- μm intervals through a 100× , 1 . 45- NA objective ( Nikon Instruments ) . Image z stacks were deconvolved using Huygens Professional software ( version 4 . 5 . 0p8 , Scientific Volume Imaging ) . The images shown here are cropped ImageJ ( version 1 . 49m , NIH ) maximum- intensity projections of the deconvolved z stacks with background subtracted using a rolling ball radius of 50 . 0 pixels . Images that were used for image analysis ( Figs 4C , 5A , 5B , 5C and 5E and S3 , S5 , S6 , S7 and S8 Figs ) were taken using an Olympus IX83- P2ZF inverted microscope equipped with an X-Cite 120LED light source ( Excelitas Technologies ) for fluorescent excitation . Emission filters were 409 nm , 506 nm , and 562 nm for the Hoechst 33342 stain , Alexa Fluor 488 , and Alexa Fluor 546 , respectively . A Hamamatsu C11440 Orca-Flash 4 . 0 CMOS camera was used to acquire images as wide-field z stacks at 0 . 24- μm intervals through a 60× , 1 . 42 NA objective ( Olympus ) . Image z stacks were deconvolved using Huygens Professional software , then flattened using a maximum- intensity projection , and subjected to background subtraction using a rolling ball radius of 50 . 0 pixels in ImageJ ( version 1 . 49m , NIH ) before image analysis in Matlab ( version 2015b , Mathworks ) as described below . In brief , the blue fluorescence channel ( the reference channel corresponding to Hoechst nucleic acid labeling ) was used as a proxy to define the pixel locations of individual cells and co-localized fluorescent overlap from HCR labeled transcripts was quantified ( S5 and S6 Figs ) . The overall image analysis scheme can be found in the Supplementary Information . Images corresponding to each fluorescent channel ( that had been deconvolved and then flattened by maximum intensity projection and subjected to background subtraction ) were imported into Matlab ( version 2015b , Mathworks ) in 16 bit TIFF format . To identify cell positions , the blue fluorescence ( reference channel ) images were binarized using Otsu’s method [44] for thresholding . The internal Regionprops function was used to identify and characterize the sizes and pixel locations of objects in the image . From there , objects less than 3 pixels or greater than 30 pixels were filtered out from the analysis using find and ismember functions . S6 Fig shows that this filtering strategy was effective in identifying cells . Objects between 3 and 30 pixels in size were analyzed as cells in subsequent analysis . The PixelList property from the Regionprops function ( called on the blue reference channel ) was used to define the pixels corresponding to each cell . Using these coordinates , the HCR intensity value corresponding to each cell was calculated by taking the mean intensity of the pixels corresponding to each cell . Graphs were created using Matlab and Mathematica ( version 11 . 0 . 1 . 0 , Wolfram ) . 100 μl of an overnight culture of JRC104+pCIE-GFP diluted 1:3 in M9 medium was added to a poly-D-Lysine coated glass bottom 96 well plate ( No . 1 . 5 , MatTek Corporation ) . The cells were incubated at 37°C for 1 h , and then washed with KPBS to remove loose cells . M9 media containing Hoechst 33342 and cCF10 at concentrations of 0 , 2 . 5 , 5 , or 50 ng ml-1 was then added to the cells . For experiments involving iCF10 , 50 ng ml-1 iCF10 was added to the cells in addition to cCF10 . Cells were imaged on a Zeiss Axio Observer Z1 inverted confocal microscope ( Zeiss Zen 2 . 0 ) using the 405 nm and 488 nm lasers for excitation of Hoechst and GFP , respectively . A 63× , 1 . 4 NA objective was used . Images were taken every 10 minutes from 40 to 150 minutes after addition of cCF10 . Two fields of view were imaged for each cCF10 concentration and ten z-stacks were taken at 0 . 26 μm intervals for each image . Images were imported into ImageJ and background subtracted using a radius of 4 pixels ( ImageJ version 1 . 50a , NIH ) . For the blue fluorescence channel ( the reference channel corresponding to Hoechst fluorescence ) , a maximum intensity z-projection was used to flatten the z-stacks ( ImageJ ) . For the GFP fluorescence channel , the Sum Slices z-projection was used to flatten the z-stacks ( ImageJ ) . These processed images were then imported into Mathematica for further analysis . To identify cell positions , the blue fluorescence ( reference channel ) images from the first time point were binarized and then distance transform and maximum detection functions were applied to create cell location markers . Then , using the markers previously found as positional information , a watershed transformation was applied to the binarized images and the SelectComponents function was used to identify any object found by the watershed algorithm between 3 pixels and 2 . 5x the mean object size . The ComponentsMeasurements function was then used to find the centroid coordinates and the equivalent disk radius of all identified cells . Using these coordinates as starting points , the algorithm ImageFeatureTrack was applied to all images within the time series to track the cells through different images and counteract microscopic drift . Any cells which were unable to be tracked by the algorithm through all time points were removed from further analysis . To identify the GFP fluorescence intensity for each cell over time , the centroid location and equivalent disk radius for each cell in each image were used to create a mask . When applied to the GFP fluorescence images , the mask set all pixels equal to zero except for the area surrounding a specific cell . The GFP intensity of the cell at that time was then calculated by taking the mean intensity of the non-zero pixels . This process was repeated for each cell in each image at each time point . Using this method , the GFP intensity for each cell over time was determined . Graphs were created using Mathematica . An overview of this analysis strategy and images demonstrating effective automated identification of cells can be seen in S7 and S8 Figs . The stochastic mathematical model was developed using the Hybrid Jump/Continuous Markov Stochastic Simulator ( HyJCMSS ) , part of the Hybrid Stochastic Simulation for Supercomputers ( Hy3S ) suite [29] . The model consisted of 15 species and 38 biochemical reactions involved in the induction of expression of QL ( S2 Table ) . The cell volume was assumed to be 10−15 L and was modeled to increase exponentially until cell division that occurred every 40 ± 4 minutes . Upon cell division , the number of proteins and mRNA molecules was halved . Initial conditions for all the species were obtained by solving equations S1-S15 ( S1 Text ) for steady-state when the extracellular concentration of cCF10 ( C ) was set to 0 ng ml-1 C . For each simulation , 10 , 000 trials ( corresponding to 10 , 000 cells ) were carried out .	Within a given niche , expression levels of individual cells ( and resulting functional behaviors ) may differ substantially from the mean of the population due to stochasticity or microenvironment heterogeneity . Quantification of bacterial gene expression at the single cell level provides a more informative picture of microbial communities . In Enterococcus faecalis , intercellular communication via peptide pheromones controls conjugation-mediated transfer of antibiotic resistance , adaptation to stress , and formation of biofilms . Population-level studies have shown that induction of conjugation by peptides is tightly controlled , but the extent of single cell variation in the process has not been explored . We analyzed induction in single cells by direct transcript labeling and GFP reporter expression and show that the response is heterogeneous and stochastic . Mathematical simulations modeled the components of the system and predicted this response . Importantly we show that conjugation can occur in response to low inducer peptide concentrations and in the presence of high inhibitory peptide concentrations . In both of these cases , some cells respond at a similar early time . Stochasticity in the response could explain the occurrence of induction and conjugation in these scenarios that might seem to disfavor plasmid transfer and permit plasmid dissemination while minimizing fitness costs of induction to donor populations .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "fluorescence", "imaging", "medicine", "and", "health", "sciences", "pathology", "and", "laboratory", "medicine", "pathogens", "plasmids", "microbiology", "antibiotic", "resistance", "genetic", "elements", "enterococcus", "forms", "of", "dna", "dna", "pharmacology", "bacteria", "bacterial", "pathogens", "research", "and", "analysis", "methods", "specimen", "preparation", "and", "treatment", "staining", "antimicrobial", "resistance", "imaging", "techniques", "proteins", "medical", "microbiology", "microbial", "pathogens", "enterococcus", "faecalis", "biochemistry", "cell", "staining", "nucleic", "acids", "post-translational", "modification", "genetics", "microbial", "control", "biology", "and", "life", "sciences", "pheromones", "genomics", "mobile", "genetic", "elements", "image", "analysis", "signal", "peptides", "organisms" ]	2017	Stochasticity in the enterococcal sex pheromone response revealed by quantitative analysis of transcription in single cells
Recent insights suggest that non-specific and/or promiscuous enzymes are common and active across life . Understanding the role of such enzymes is an important open question in biology . Here we develop a genome-wide method , PROPER , that uses a permissive PSI-BLAST approach to predict promiscuous activities of metabolic genes . Enzyme promiscuity is typically studied experimentally using multicopy suppression , in which over-expression of a promiscuous ‘replacer’ gene rescues lethality caused by inactivation of a ‘target’ gene . We use PROPER to predict multicopy suppression in Escherichia coli , achieving highly significant overlap with published cases ( hypergeometric p = 4 . 4e-13 ) . We then validate three novel predicted target-replacer gene pairs in new multicopy suppression experiments . We next go beyond PROPER and develop a network-based approach , GEM-PROPER , that integrates PROPER with genome-scale metabolic modeling to predict promiscuous replacements via alternative metabolic pathways . GEM-PROPER predicts a new indirect replacer ( thiG ) for an essential enzyme ( pdxB ) in production of pyridoxal 5’-phosphate ( the active form of Vitamin B6 ) , which we validate experimentally via multicopy suppression . We perform a structural analysis of thiG to determine its potential promiscuous active site , which we validate experimentally by inactivating the pertaining residues and showing a loss of replacer activity . Thus , this study is a successful example where a computational investigation leads to a network-based identification of an indirect promiscuous replacement of a key metabolic enzyme , which would have been extremely difficult to identify directly . Enzymes have traditionally been associated with discrete activities [1] . Clear-cut annotations populate well-known enzyme databases such as KEGG and Uniprot , and foster an implicit assumption that gene activities are specific . However , recent advancements in our understanding of evolution and enzyme activity have cast this view into question , and suggest instead that non-optimized and/or promiscuously active enzymes are frequent and active across life [2] . ‘Generalist’ enzymes ( i . e . , those carrying more than one function ) have been shown to be abundant , to play different biological roles than ‘specialist’ enzymes , and to behave differently than specialist enzymes during switches in media conditions [3] . Promiscuous enzyme activities ( i . e . , those that are only active with low affinities/activities ) have been shown to play adaptive biological roles ( e . g . , [4] and [5] ) , and to often arise through neutral mutations that are not detrimental to the primary enzymatic activity [6] . In particular , antibiotic resistance may emerge initially through the action of over-expressed , promiscuous genes [7 , 8] . Because of recent conceptual and modeling advances , there is significant interest in being able to predict and exploit enzyme promiscuity [9–12] . Tools have been developed that mine molecular signatures or three-dimensional catalytic domains of enzymes to predict whether enzymes are promiscuous [13–15] , and these methods have been used , e . g . , to design likely retrosynthetic pathways [9 , 11] . While these methods achieve important goals , they have not been aimed at evaluating the functional implications of existing promiscuous activities on a genome-wide , network scale . A notable recent effort did probe the global relevance of promiscuous enzyme activities , but the study used ‘as is’ data collected from enzyme databases , which , while having the advantage of being fully experimentally verified , include many non-biologically-relevant instances and exclude many biologically relevant ones [16] . Another recent study used knowledge gaps identified in a genome-scale metabolic model of E . coli to identify candidate isozymes , which display substrate promiscuity [17] . In that study , a ‘target’ enzyme that is found to be nonessential in vitro despite being essential in silico is knocked out , after which potential isozymes that are found to be upregulated are cumulatively knocked out until the cell can no longer survive . This method thus identifies enzymes with a secondary isozyme function that , when the cell overexpresses them , can compensate for the loss of the ‘target’ . Interestingly , in one case they consider , the target is in fact essential , and adaptive evolution is required for the potential isozyme to be expressed enough to compensate for its loss . Searching for many more such low activity promiscuous functions in an unsupervised , genome-wide way is the focus of this present study . To accomplish this , we utilize an unsupervised PSI-BLAST based method for assessing potential secondary functions of genes in E . coli . We predict promiscuous ‘replacer’ functions that may compensate for primary ‘target’ functions in other genes if they are altered or lost , and compare these predictions to known cases in which an over-expressed gene can take on a secondary role , as shown in an assay generally referred to as multicopy suppression [18] . Multicopy suppression reveals ‘replacer’ genes whose over-expression suppresses the lethality caused by knocking out conditionally essential ‘target’ genes . It is thus an elegant assay for uncovering promiscuous gene functions . We find that our method for predicting promiscuous functions successfully recapitulates known multicopy suppression events [8 , 19] and predicts new ones , several of which we validate in vitro . We next develop a genome-scale metabolic modeling ( GEM ) based approach to predict and experimentally test cases of multicopy suppression in which functional promiscuous replacement happens ‘indirectly’ through a bypassing pathway or reaction , rather than by directly replacing the function of the target gene . For this , we focus on the gene target pdxB , for which multiple promiscuous replacers have been reported in the past [8] . PdxB is a key enzyme for producing pyridoxal 5’-phosphate ( the active form of Vitamin B6 ) , an essential nutrient in E . coli . Our experiment proceeded in 4 steps: First , we predicted that the gene thiG harbors a promiscuous activity that metabolically bypasses pdxB . Second , we validated this prediction with a new in vitro individual multicopy suppression experiment . Third , we performed a detailed structural analysis of ThiG and hypothesized which residues perform the promiscuous function . And Fourth , we mutated this active site and confirmed loss of pdxB replacement activity . Thus , we begin at the level of a genome-wide screen for promiscuous functions , and proceed to identify and validate a promiscuous pathway for production of an essential nutrient in E . coli , which could have important functional implications for this model organism . We aimed to predict promiscuous functions of genes in a systematic and unsupervised manner , and specifically to do it on a genome-wide scale . We started by doing permissive PSI-BLAST based searches for distant gene similarities across RAST , a large consistent database of metabolic reactions and compounds that includes thousands of microbes [20] . This search allowed us to assign putative secondary functions for each initial “root” gene in the search ( see Methods and Fig 1 ) . We did this across metabolic genes in E . coli , which resulted in a set of phylogenetic trees , one for each root gene , that include up to thousands of genes from other bacteria along with their associated functions . Using these trees , we searched for instances in which a gene in one of the trees had an assigned function different from the one that the RAST database assigns to the root gene . We restricted our search to genes whose primary functions are metabolic , as this is our focus of interest . Since our tree building method was more permissive than typical BLAST or PSI-BLAST , we considered these predicted functions as potential secondary functionalities that might only be effective at high expression levels or with slight mutations . We entered these cases in a large gene-by-function matrix . To facilitate testing and validation , we used our promiscuity matrix to predict pairs of genes that correspond to those identified through an in vitro assay called Multicopy suppression [19] . Multicopy suppression tests ‘replacer’ genes for promiscuous activities that either directly ( catalyze similar reactions ) or indirectly ( catalyze distinct reactions ) replace the function of a ‘target’ gene and preserve growth . It proceeds in four steps: ( 1 ) a ‘target’ gene is identified in the bacterium of interest ( target genes must be essential on a certain test medium [usually M9] , but not on a rich medium ) ; ( 2 ) a strain is created in which the target gene is inactivated ( due to the condition in step 1 , this strain is not viable on the test medium ) ; ( 3 ) a library of native genes ( hereafter called ‘replacers’ ) is individually over-expressed in the organism on high-copy plasmids; and ( 4 ) transformed strains are grown on the test medium , and survivors are noted as having identified successful target-replacer gene pairs . Importantly , the replacer genes are native to the genome of the organism , but their natural expression dosage is not sufficient to promiscuously support growth . Hence , replacer genes discovered through multicopy suppression are hypothesized to possess secondary , low-affinity/low-activity functions that compensate for the function of the target . We predict target-replacer pairs in two ways: direct replacements , and indirect replacements . To predict direct replacements , we search our matrix for instances in which one ‘replacer’ gene in E . coli has a secondary function assigned through our method , which corresponds exactly to the function of some other ‘target’ gene that is also found in E . coli . We call this straightforward method the enzyme PROmiscuity PrEdictoR , or PROPER . To predict indirect replacements , we use the added insight of a GEM of E . coli ( obtained from SEED: [21] ) . The SEED E . coli GEM was used , as opposed to one of the highly curated E . coli GEMS such as [22] , because of our requirement that metabolic reactions from the SEED database can be seamlessly added to the GEM to test their effect on cellular fluxes ( which would first require a large reconciliation of nomenclature if using a non-SEED GEM ) . Specifically , we search for replacers that , through a promiscuous function identified in our matrix , metabolically bypass the need for the target . These predictions were done in three steps: ( 1 ) we knocked out the target gene in a GEM ( this could only be done for genes that were conditionally lethal on M9 in silico ) ; ( 2 ) we searched in our gene similarity trees for any E . coli genes that harbor non-E . coli promiscuous functions; ( 3 ) we added each of these promiscuous functions in turn to the GEM , and looked for any that rescued in silico growth . This GEM-based PROmiscuity PrEdictoR ( GEM-PROPER ) thus identified indirect replacer genes that act through metabolism ( see panel 3 of Fig 1 ) . In all , PROPER predicted 2811 direct target-replacer pairs in E . coli , encompassing 794 metabolic target genes and 753 metabolic replacer genes ( see S1 Table ) . GEM-PROPER predicted a total of 98 indirect target-replacer pairs in E . coli ( see full list in S4 Table ) . Since direct replacers span a large portion of metabolism , we focus on these predictions for our large-scale analyses ( i . e . , the first several sections of the results ) . Afterwards , we shift our focus to GEM-PROPER , and present a detailed analysis of a novel indirect predictor that we validated . The number of direct replacers per target follows an exponentially decaying distribution ( see S1 Fig ) , implying that promiscuous gene functions cluster around key target functions . The target genes with the most putative replacers are involved in fatty acid degradation or synthesis [fadD and fadK ( degradation ) ; fabD ( synthesis ) ] , or transport processes [malF ( maltose/maltodextrin transport ) ; potC , ydcV , potB , and ydcU ( spermidine/putrescine transport ) ; and cysW ( sulfate transport ) ] . Fatty acid degradation and synthesis processes are known to involve enzymes with many potential specificities depending on only small alterations [23] . To validate our direct predictions , we collected results from two studies of multicopy suppression in E . coli , which between them had discovered 48 instances of multicopy suppression among 21 target genes [8 , 19] . Both of these assays looked for multicopy suppressors of targets that were essential on M9 medium but not in a rich medium , and one used a combinatorial assay that assessed every gene in E . coli as a potential replacer [19] . All but 3 of the target genes for which replacers had been found were metabolic , supporting our focus on metabolism . In all , we predicted a total of 63 replacers for the 21 targets , versus the 48 found previously by multicopy suppression . The replacers we predicted overlapped with those found previously by 8 genes . Notably , for each of the 21 target genes , there were over a thousand metabolic genes that could serve as potential replacers; therefore , the chance of achieving this level of overlap between our predictions and those found experimentally through random guesses is extremely low ( p = 4 . 5e-15 in a hypergeometric test ) . As another control test , we performed a BLAST search of each target gene identified in Patrick et . al . against all metabolic genes in E . coli , and kept only those above the default threshold as potential replacers . This yielded 122 predicted replacers for the 21 target genes , among which none matched results from Patrick et . al . , and only two overlapped with predictions made by PROPER ( neither of those turned out true in our own experiments ) . This emphasizes the added value of PROPER over simple BLAST searches for finding find such low affinity promiscuous functions . We also considered it possible that some of our target-replacer predictions are true , but cannot be verified using multicopy suppression because the target gene is not conditionally lethal ( which is a necessary condition for doing a multicopy suppression experiment–see Fig 1 , panel 4 ) . As a quick test of this , we compared our direct predictions to the isozyme sets aspC/tyrB , argD/astC/gabT/puuE , and gltA/prpC , which were discovered in [17] as described in the introduction . Indeed , all 8 pairwise combinations of genes from within any of these 3 isozyme sets comes up as a reciprocal target-replacer pair in our direct predictions , meaning that either gene in each pair can play the role of the target or the replacer . That all of these pairs would come up in our predictions and , given that , that they would all be reciprocal are both highly unlikely by chance ( p = 3e-19 and p = 3e-5 in hypergeometric tests , denoting that all pairs from [17] ( a ) overlap our predictions vs . a background of all-vs . -all pairings of our predicted targets and replacers , and ( b ) are reciprocal , assuming the same likelihood for any pair ) . This suggests that promiscuous functions that have enough activity to be considered isozymes ( or , more specifically , to be defined so through the experimental pipeline in [17] ) may tend to come up in our predictions as reciprocal , and suggests a complimentary association between our methods and those in the aforementioned work . The random gene insertion method used by Patrick et al . to screen for potential target-replacer pairs [19] is demonstrably not comprehensive ( e . g . , 7 replacers were found for pdxB in [8]; none of these were reported in the Patrick study for pdxB , although pdxB was in the set of replacers that Patrick et al . tested ) . Hence , we next asked how many of our novel predicted target-replacer pairs ( those that had not been previously found experimentally ) may be true . To test this we developed a single-replacer multicopy suppression assay analogous to that used to validate individual results in the initial multicopy suppression study by Patrick [19] . This assay involves plating target-replacer strains ( i . e . , strains with the target knocked out , and the replacer over-expressed on a plasmid ) on M9 medium , and observing how long it takes for colonies to appear . We judged successful target-replacer pairs as those that grew more robustly than an empty plasmid control ( see Methods , as well as Supplementary methods in S1 Text for a fuller explanation of this assay ) . Using these criteria , we verified 7 of the 8 target-replacer pairs that were identified in Patrick et . al . and predicted by PROPER ( see S2 Table for full list of experimental results ) . We then assessed each of our 55 novel predicted target-replacer pairs for multicopy suppression , and found three of them to have replacer activity ( hisA , cysM , and metB were viable replacers for hisH , ilvA , and metC respectively; see S2 Table ) . Thus , in all we validated 10 out of 63 direct replacement predictions , which gives a success rate of 20% ( after eliminating 14 predictions for ptsI and glyA target genes , which were in practice untestable because the target knockout did not decrease growth from wt in our experiments; see S1 Text ) . Among the three validated novel target-replacer pairs we validated , the strength of the observed phenotype scaled with the strength of sequence similarity between the replacer gene and its match in the promiscuous gene tree ( i . e . , Panels 1–2 of Fig 1 ) . Namely , metB fully restored growth in ΔmetC cells , while the less homologous cysM and hisA ( in that order ) less efficiently supported growth following target knockout ( see S1 Text; also compare alignments in S4 , S5 and S6 Figs ) . Thus , these examples suggest that higher similarity to the function-assigning gene corresponds to higher activity or affinity for that enzyme’s substrates , although a more extensive study of multicopy suppression phenotypes would be needed to uphold this observation generally . Among the 98 indirect target-replacer pairs we predicted , two were for the target pdxB , which is by far the most ubiquitiously replaced target found in vitro across published multicopy suppression datasets ( in fact , pdxB is the sole focus of the Kim study: [8] ) . We confirmed our in silico predictions of these two target-replacer pairs using the manually curated E . coli model iAF1260 [22] . The target , pdxB , is a conditionally essential gene ( essential on M9 medium , but not on rich medium ) involved in the biosynthesis of pyridoxal 5’-phosphate ( P5P ) , the active form of Vitamin B6 . Vitamin biosynthesis reactions are good candidates for multicopy suppression , since vitamins are required in low amounts , so even a moderate flux through an over-expressed promiscuous enzyme could provide enough of the nutrient to enable growth [24] . P5P is an essential cofactor in all known living systems [25] . We tested our two predicted pdxB target-replacer pairs , and found one of them , thiG replacing pdxB , to be a true replacer [∆pdxB/thiG colonies were 1mm diameter , vs . 0 . 1–0 . 2mm diameter in ∆pdxB/empty , after 3 days incubation in replicate experiments on M9 medium; see S2 Table] . The observed phenotype was consistent with previously reported pdxB replacers ( 1mm colonies of a ∆pdxB/replacer strain after 1–2 days ( 3 cases ) or 3–5 days ( 4 cases ) at 37°C in equivalent growth conditions & temperature [8] ) . We explored the thiG-pdxB target-replacer pair in detail , as follows: Identifying promiscuous gene functions is a fundamental task in biology . Promiscuous functions have been causally associated with the evolution of new gene functions [24 , 31] , and may have contributed to the evolution of resistance to antimicrobials and other stresses [7] . Here , we present a new method termed PROPER that uses iterative PSI-BLAST-based phylogenetic trees for predicting potential promiscuous functions based on weaker similarities than are typically considered in assigning gene functions . Based on these predictions , we experimentally validate 4 novel target-replacer pairs , one of which ( thiG replacing pdxB ) , was found by coupling our promiscuity predictor with a GEM . Finally , we predict and then experimentally validate the promiscuous active site of thiG for this replacing activity , revealing a striking new promiscuous route for the production of an essential nutrient , p5p , in E . coli . While sequence similarity-based methods ( e . g . , BLAST ) are used ubiquitously to determine primary gene functions , their ability to call promiscuous functions is unclear . Our method gives a first estimate , as we were able to experimentally validate around 20% of direct target-replacer pairs we could test in the lab ( This is 100x higher than expected if guessing target-replacer pairs randomly ) . It is likely that the number of target-replacer pairs we validated is an underestimate of true promiscuity , as some true replacers might not show during multicopy suppression due to incorrect expression levels ( e . g . , due to non-optimal IPTG concentrations ) , insufficient effects of target knockout , or other confounding factors . Notably , many of our predictions might be correct even though they cannot be directly validated through multicopy suppression ( due to the necessity in multicopy suppression that the target gene is conditionally essential ) . Our comparison to the work of [17] emphasizes this point , as several of our predicted target-replacer pairings came up in that work , regardless of the fact that the target genes examined in that study are not conditionally essential . The promiscuous functions we predict thus might represent a large underground repertoire that could be activated under the right kinds of selective pressure . Our method for predicting promiscuous gene functions utilizes the RAST database and automatically constructed metabolic models from SEED , resources that include thousands of bacterial and archaeal species . While manually curated GEMS for E . coli are available ( e . g . , [32] ) , we chose to use the SEED GEM because it interfaces smoothly with the SEED and RAST databases , which was necessary for implementing GEM-PROPER without needing to reconcile thousands of metabolite and reaction names with an outside model . A second benefit of this choice is that although we focus here on predicting promiscuous gene functions in E . coli , our methods are generic and may easily be extended to any other organism in RAST . This could , for example , aid in determining resistance or adaptation mechanisms across major pathogenic strains , which could then be targeted in new cures . In all , this work constitutes the first ever genome-wide prediction of metabolic gene multicopy suppression , and the first integration of such predictions with a GEM to achieve network-reliant indirect predictions . We begin with an unsupervised , large-scale method to predict enzyme promiscuity , and finally dial in on and then experimentally verifying a key prediction of biological significance in E . coli . Without such a systems approach , it would have been extremely difficult to identify thiG as a candidate replacer for pdxB . Thus , this study serves as a successful example where systems biology can provide a roadmap for biological investigation , identifying the most promising targets to then follow up on with more costly and time-consuming experiments . PROPER proceeds in three steps , as outlined in Fig 1: ( 1 ) building gene similarity trees for all genes in E . coli; ( 2 ) using these trees to build a matrix of promiscuous gene functions; and ( 3 ) applying this matrix to determine potential promiscuous gene replacers for target genes of interest . In GEM-PROPER , step ( 3 ) is replaced by a GEM-based approach , in which we search for promiscuous functions ( from step 2 ) that can rescue in silico growth of a GEM model after knockout of the target gene . Here , we describe these steps in detail: Multicopy suppression is an established assay that has been described elsewhere ( see panel 4 in Fig 1 , and , e . g . , [8 , 19] ) . In our experiments , we tested specific target-replacer pairs in the manner used by Kim to validate target-replacer pairs they identified in their large-scale screen . Namely , we inserted a plasmid containing the replacer gene with an IPTG-induced promoter into a KEIO strain with the target gene knocked out , and assessed whether this target-replacer strain grew better than the target knockout strain without the replacer . Specifics of strain construction and the experiments follow: E . coli strains containing specific gene knockouts were obtained from the KEIO collection [36] , which contains single-gene knockout strains for the majority of genes in E . coli . Plasmids for IPTG-inducible expression of predicted replacer genes were obtained from the ASKA collection [37] , which contains plasmids that overexpress nearly every individual gene in E . coli . Both collections are obtainable from the National Bioresource Project at the National Institute of Genetics , Japan . KO target strains were made electro-competent ( 4x washes with ddw/10% glycerol ) and transformed with ASKA plasmids over-expressing the corresponding predicted replacer gene . When we were developing our assay , we found that knockout colonies usually grew after some amount of time on M9 medium regardless of whether the replacer gene had been added to the plasmid , despite the fact that each of the target strains we tested was previously reported to be non-viable on M9 medium . The previous study by Patrick [19] had not plated background target-deleted strains that had empty plasmids inserted . Due to the eventual growth in most strains , this was an important negative control for judging that a replacer was actually compensating for the loss of the target and improving growth ( and was done in Kim: [8] ) . Therefore , for each KO strain we additionally transformed with an empty ASKA plasmid as a negative control . We judged a strain to be a true replacer only if it came up consistently stronger and/or earlier than the strain carrying the empty plasmid . In the KEIO collection each knocked out gene is replaced by the kanamycin resistance gene ( KanR ) . Thus we verified correct location of the knockout in each target strain by amplifying and sequencing the area directly flanking the kanamycin insertion , using a kanamycin universal primer and a strain-specific reverse primer located downstream of the knocked out gene . Since the orientation of the kanamycin insertion depends on the orientation of the original knocked out gene , the KanR primer 5’-ATATTGCTGAAGAGCTTGG was used when the original gene had been forward coded and the KanR primer 5’-AATGAACTCCAGGACGAG was used if the original gene had been reverse coded . Correct sequence of the replacer gene was verified by amplifying and sequencing the ASKA plasmid insert ( forward: 5′-ATC ACC ATC ACC ATA CGG AT; reverse: 5′-CTG AGG TCA TTA CTG GAT CTA ) . Starter cultures of each target-replacer ( TR ) pair were grown overnight in LB+CAP , washed and serially diluted in saline ( 0 . 9% NaCl ) . Aliquots ( 100 ul ) of 10−6 dilution were plated on LB+chloramphenicol and on M9-glucose-kanamycin-chloramphenicol ( 1× M9 salts , 2 mM MgSO4 , 0 . 1mMCaCl , 0 . 4% glucose , 34 μg/mL chloramphenicol , 30μg/ml kanamycin ) containing IPTG . All TR pairs were plated on both M9 with 50μM and 125μM IPTG; TR pairs of particular interest ( hisA/∆hisH , cysM/∆ilvA and thiG/∆pdxB ) were plated on 250μM as well . In each TR plating experiment , the empty and frvX negative controls were plated alongside the TR pairs . Plates were wrapped in nylon to avoid dehydration , incubated at 37 and monitored during the next three weeks . Nearly all negative control plates showed some growth phenotype , ranging from very strong ( normal sized colonies within 3 days ) to very weak ( pinpoint colonies after 3 weeks; S3 Table ) . Thus , a TR pair was considered valid only if colonies consistently formed at a higher rate than in both negative control plates . Each plating experiment was repeated 2–3 times . Growth of TR pairs that showed exceptionally high growth on the negative control plates ( ΔglyA , ΔptsI and ΔpabB ) was also monitored in liquid culture . Cultures in LB+CAP were washed once in saline and re-suspended at a dilution of 1:100 into M9-glucose-kanamycin-chloramphenicol minimal media supplemented with IPTG ( 50 μM ) . Growth measurements were performed in a 96-well plate incubated for 19–24 h at 37°C in a temperature-controlled plate reader with continuous shaking ( ELX808IU-PC; Biotek ) , and OD595 was monitored every 15 min . Each TR pair was loaded into 2 duplicate wells . Growth of the negative controls ( empty and frvX plasmids ) for each target strain was likewise monitored during every run . It was reported in [8] that knockout strains of pdxB sometimes display a heterogeneous phenotype , with some growing on minimal medium and others not . To be sure that this was not a factor in our experiments verifying the thiG/∆pdxB replacer-target pair , we grew individual colonies of ∆pdxB and confirmed that there was no heterogeneity in their growth on M9 medium . Overnight cultures of each strain were grown in LB with 30 μg/mL kanamycin ( ΔpdxB cells ) and 25 μg/mL chloramphenicol ( ASKA plasmids ) selection . Cells were washed in 1x PBS and diluted 1:100 into M9-glucose-kanamycin-chloramphenicol for plate seeding; M9 was prepared without a nitrogen source . A checkerboard matrix was generated in 2mL deep-well , 96-well assay plates by serial dilution of NH4Cl ( 22 . 5% w/v maximum concentration ) across plate columns and IPTG ( 500 μM maximum concentration ) across plate rows . Wells were uniformly inoculated with cells , and each well contained a final volume of 1 . 2 mL . Plates were sealed with gas-permeable membranes and grown in a light-protected , microplate incubator shaker at 37°C and 700 RPM; 700 RPM was determined to be equivalent to 300 RPM in a standard incubator shaker . Samples for OD600 measurements were taken at designated timepoints using a SpectraMax M5 microplate multimode plate reader ( Molecular Devices ) , and the gas permeable membrane resealed after each timepoint . 100 μL samples were taken for OD600 measurements to minimize the total volume loss . For batch culture experiments , cells were prepared from overnight cultures as described above . Cells were diluted 1:100 in 30 mL of M9-glucose-kanamycin-chloramphenicol , containing ~3% NH4Cl ( w/v ) and 10 μM IPTG , in 250 Erlenmeyer flasks . Cultures were grown at 37°C and 300 RPM in an incubator shaker , with samples for OD600 measurements taken at designated timepoints . The X-ray structure of pdxS ( from B . bacillus ) in complex with pdxT , was downloaded from the pdb database , and the residues of the pdxS active site were identified from the publication ( active residues are: K149 , D102 , K81 and D24 ) [38] . Multiple pdxS units from the multimeric structure were overlaid and were found to be coincident ( as can be seen in Figs 2B and S8 ) . No structure of thiG was available from E . coli , so a homology-model was built using the SWISSMODEL pipeline ( http://swissmodel . expasy . org/; [39] ) based on the thiG template from Thermus thermophilus ( 51 . 98% seq identity; PDB ID 2htm [40] ) . The structures of pdxS were subsequently aligned with that of thiG using pyMol [41] . Low ( 50μM ) IPTG concentration was sufficient to induce all of the replacers . Interestingly , increasing the IPTG concentration affected growth sometimes positively and sometimes negatively , depending on the specific strain . Increasing IPTG concentration up to 250μM caused increases in colony size and number , and decreases in incubation times , for hisA/∆hisH and cysM/∆ilvA . Conversely , growth of purE/∆purK , one of the target-replacer pairs predicted by both Patrick and us , was almost entirely inhibited when IPTG concentration was increased to 125uM . These results illustrate that over-expression can also have deleterious effects [42] . The optimal level of expression ( and hence the optimal IPTG concentration ) depended on the specific target-replacer combination . In agreement with this , in several target strains , over-expression of the randomly chosen gene frvX caused a decrease of the background seen in the empty plasmid control ( see S3 Table ) .	Many enzymes can perform secondary functions at low affinities or rates , but such ‘promiscuous’ functions have never been predicted on a genome-wide scale . Here , we present the first genome-wide method to predict promiscuous functions of metabolic genes , which we apply to E . coli . Notably , we predict and validate several new cases where a ‘replacer’ gene can compensate for the loss of an essential ‘target’ gene through a promiscuous activity . Next , we couple our method with a genome-scale metabolic model , in order to search for ‘replacer’ genes that compensate for essential ‘target’ genes by metabolically bypassing them . We use this network-augmented approach to uncover a novel promiscuous pathway for the production of pyridoxal 5’-phosphate ( the active form of Vitamin B6 ) in E . coli . This study represents an important step in understanding promiscuous functions in bacteria , and is a prime example of a systems-level analysis leading to new biological insight .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "sequencing", "techniques", "medicine", "and", "health", "sciences", "pathology", "and", "laboratory", "medicine", "pathogens", "computational", "biology", "enzymology", "database", "searching", "bacillus", "microbiology", "gene", "function", "prokaryotic", "models", "model", "organisms", "phylogenetic", "analysis", "genome", "analysis", "enzyme", "metabolism", "protein", "structure", "molecular", "biology", "techniques", "bacteria", "enzyme", "chemistry", "bacterial", "pathogens", "research", "and", "analysis", "methods", "sequence", "analysis", "sequence", "alignment", "proteins", "medical", "microbiology", "microbial", "pathogens", "biological", "databases", "molecular", "biology", "molecular", "biology", "assays", "and", "analysis", "techniques", "protein", "structure", "comparison", "biochemistry", "database", "and", "informatics", "methods", "bacillus", "subtilis", "genetics", "biology", "and", "life", "sciences", "genomics", "genomic", "databases", "organisms", "macromolecular", "structure", "analysis" ]	2016	Systems-Wide Prediction of Enzyme Promiscuity Reveals a New Underground Alternative Route for Pyridoxal 5’-Phosphate Production in E. coli
Activation of pattern recognition receptors and proper regulation of downstream signaling are crucial for host innate immune response . Upon infection , the NF-κB and interferon regulatory factors ( IRF ) are often simultaneously activated to defeat invading pathogens . Mechanisms concerning differential activation of NF-κB and IRF are not well understood . Here we report that a MAVS variant inhibits interferon ( IFN ) induction , while enabling NF-κB activation . Employing herpesviral proteins that selectively activate NF-κB signaling , we discovered that a MAVS variant of ~50 kDa , thus designated MAVS50 , was produced from internal translation initiation . MAVS50 preferentially interacts with TRAF2 and TRAF6 , and activates NF-κB . By contrast , MAVS50 inhibits the IRF activation and suppresses IFN induction . Biochemical analysis showed that MAVS50 , exposing a degenerate TRAF-binding motif within its N-terminus , effectively competed with full-length MAVS for recruiting TRAF2 and TRAF6 . Ablation of the TRAF-binding motif of MAVS50 impaired its inhibitory effect on IRF activation and IFN induction . These results collectively identify a new means by which signaling events is differentially regulated via exposing key internally embedded interaction motifs , implying a more ubiquitous regulatory role of truncated proteins arose from internal translation and other related mechanisms . In response to pathogen infection , host cells initiate an immediate innate immune response to defeat pathogen propagation [1 , 2 , 3] . The retinoic acid-inducible gene I ( RIG-I ) and melanoma differentiation antigen 5 ( MDA5 ) are cytosolic receptors that sense infecting viruses via RNA with distinct structural features [4 , 5 , 6] . Upon RNA association , RIG-I and MDA5 dimerize with the mitochondrion antiviral signaling ( MAVS ) adaptor that , in turn , triggers the activation of IKK ( IKKα and β ) and TBK-1 or IKKε ( also known as IKKi ) kinase [7 , 8 , 9 , 10] . IKKα or β phosphorylates the inhibitor of NF-κB and induces its subsequent degradation , unleashing NF-κB to translocate into the nucleus and up-regulate gene expression [11 , 12] . TBK-1 and IKKε phosphorylate the interferon regulatory factors ( IRF ) to enable the expression and secretion of interferons ( IFN ) , e . g . , interferon β [13 , 14] . As such , these signaling events cumulate in establishing an effective antiviral state . To survive in the presence of active host immune defense response , pathogens have evolved a plethora of strategies to evade and exploit host immune signaling events [15 , 16] . RIG-I and MDA5 are key cytosolic sensors that detect intracellular RNA . Not surprisingly , RNA viruses deploy various mechanisms to disrupt signal transduction downstream of these receptors [16 , 17] . An elegant example that shared by multiple RNA viruses , including hepatitis C virus , picornavirus and enterovirus , is to cleave the MAVS adaptor off the mitochondrion membrane with a viral NS4/NS5 protease , thereby shutting down IFN induction in response to viral infection [9 , 18 , 19 , 20 , 21] . Our previous work with murine gamma herpesvirus 68 ( γHV68 ) , a model herpesvirus closely-related to Kaposi’s sarcoma-associated herpesvirus ( KSHV ) and Epstein-Barr virus ( EBV ) , showed that γHV68 hijacks MAVS and IKKβ to promote viral transcriptional activation and disable cytokine gene expression [22 , 23 , 24] . As such , loss of MAVS impaired lytic replication of γHV68 , in stark contrast to the increased viral replication of diverse RNA viruses [25 , 26] . These results suggest that gamma herpesviruses have evolved strategies to activate RIG-I and/or MDA5 , receptors upstream of MAVS . Indeed , we have recently reported a new mechanism of RIG-I activation that is enabled by a viral pseudo enzyme [27] . These studies collectively defined intricate viral immune evasion and exploitation strategies . The RIG-I and MDA5-dependent innate immune signaling bifurcates downstream of MAVS into the IKK-NF-κB and TBK-1/IKKε-IRF cascades . It is not well understood how these two signaling ramifications can be differentially regulated downstream of shared RIG-I and MAVS , and other common receptors and adaptors . We have recently discovered conserved gamma herpesvirus proteins , vGATs that preferentially activate NF-κB via deamidating RIG-I [27] . The mechanism of action of the herpesviral protein is unique in that: a ) viral proteins , in contrast to RNA , activates RIG-I; b ) these viral factors induce RIG-I activation via deamidation of key residues , a new means distinct from that induced by RNA; and c ) the viral factor selectively triggers NF-κB activation , but not IRF activation and IFN induction , via RIG-I and MAVS . The differential activation downstream of MAVS by vGAT prompted us to dissect the underpinning mechanism . In search for factors that contribute to the differential activation of RIG-I-dependent signaling , we discovered a MAVS variant of ~50 kDa ( designated MAVS50 ) that arises via internal translation from the second initiation codon . Notably , the MAVS50 variant was recently reported by Brubaker S . W . et al [28] . We discovered that , though lacking the N-terminal caspase recruitment domain ( CARD ) , MAVS50 exposes its TRAF-binding motifs within the very N-terminus to deregulate IRF activation and IFN induction . Interestingly , MAVS50 is sufficient to activate NF-κB , but not IFN induction . MAVS50 inhibits MAVS70-mediated innate immune signaling via competing with full-length MAVS for binding to TRAF2 and TRAF6 molecules . Sum of both leads to the specific inhibition of IRF activation and IFN induction . Consequently , MAVS50 expression increased VSV lytic replication . Mutations ablating TRAF-binding ability impaired MAVS50 to inhibit IFN-β induction in response to Sendai virus ( SeV ) infection and to increase VSV replication . These findings identify a delicate mechanism of selective innate immune activation via truncated proteins that expose their key protein-interacting domains . We have previously identified gamma herpesviral homologues of glutamine amidotransferase ( referred to as vGAT ) activate RIG-I via deamidation [27] . We noted a remarkable feature of this innate immune activation is the preferential activation of the NF-κB signaling cascade , but not that of IRF and IFN induction . In an experiment that aims to examine MAVS activation by Sendai virus ( SeV ) infection or expression of γHV68 vGAT , we observed that a smaller isoform of MAVS , of ~50 kD ( designated MAVS50 ) , did not migrate into the Triton X-100-insoluble fraction in cells infected with SeV or expressing γHV68 vGAT ( Fig 1A ) . In contrast , the full-length MAVS , designated as MAVS70 , accumulated in the Triton X-insoluble fraction , indicative of its activation [8] . This result suggests that MAVS50 likely possesses function distinct from its kin , MAVS70 . We then set out to determine the nature of the MAVS50 variant . Visual inspection of MAVS mRNA revealed an internal translation initiation site at codon 142 ( Fig 1B ) . We reasoned that MAVS50 , if produced from internal translation from the second initiation codon , lacks the N-terminal region including the entire CARD domain . We employed two antibodies , one was raised against a polypeptide of the first 135 amino acids and the other against an internal sequence encompassing amino acids 150 to 250 , to differentiate these two putative MAVS isoforms . When whole cell lysates were analyzed by immunoblotting , we found that only MAVS70 reacted with the antibody against the first 135 amino acids , whereas both MAVS70 and MAVS50 reacted with the antibody against the internal region ( aa 150–250 ) ( Fig 1C ) . This result indicates that MAVS50 lacks the amino-terminal region . We then engineered a MAVS construct that carries an amino-terminal Flag epitope and a carboxyl-terminal HA epitope to probe MAVS expression . Such a dually-tagged MAVS construct yielded a single MAVS species of 70 kDa reacting with anti-Flag antibody , and both species of ~70 kDa and 50 kDa reacting with anti-HA antibody . These two species are of similar sizes to endogenous MAVS ( Fig 1D ) . To determine whether MAVS50 was produced from the second initiation codon , we mutated the second ATG ( methionine ) into TGC ( cysteine ) within the cDNA of MAVS70 . To exclude that MAVS50 is a product of MAVS70 due to internal proteolytic cleavage , we deleted the first ATG of MAVS70 . We also generated a construct that contains the cDNA sequence encoding amino acids 142 to 540 of MAVS70 . As shown in Fig 1E , deletion of the first initiation codon abolished the expression of MAVS70 , while produced more MAVS50 than a construct containing wild-type MAVS . The increased MAVS50 expression likely stems from the lack of competition of translation initiation at the first AUG codon . We noted additional species larger than 50 kDa were produced from the construct missing the first initiation codon , suggesting that these proteins are produced from internal translation using non-AUG codons upstream of the second AUG ( 142 ) initiation codon . As expected , mutation of the second initiation codon abolished the expression of MAVS50 . Moreover , the MAVS construct containing the sequence encoding amino acids 142–540 yielded a MAVS protein migrating identically as MAVS50 . These results collectively support the conclusion that MAVS50 is produced from internal translation initiation using the second AUG codon . Next , we probed a number of human cell lines for the expression of MAVS50 and found that MAVS50 , similar to MAVS70 , was abundantly expressed from HEK 293T , Jurkat T cells , BJAB B cells , HUVEC endothelial cells , HCT116 colorectal cells , although both isoforms were detected at very low level in HeLa cervical cells ( Fig 1F ) . We further examined the expression of MAVS70 and MAVS50 expression in cells infected with SeV and herpes simplex virus type 1 ( HSV-1 ) , prototype RNA and DNA viruses , respectively . Both viruses modestly induced the expression of MAVS50 at late time points post-infection , specifically 24–48 ( Fig 1G ) and 12 ( Fig 1H ) hours post-infection for SeV and HSV-1 , respectively . MAVS serves as an adaptor to relay signaling from RIG-I and MDA5 receptor to downstream kinases that bifurcate to activate NF-κB and IRF transcription factors [7 , 8] . To examine the roles of MAVS50 in these signaling cascades , we over-expressed MAVS50 and examined signaling events leading to NF-κB and IRF activation . Using reporter assays , we found that MAVS50 expression activated NF-κB ( Fig 2A ) in a dose-dependent manner . By contrast , MAVS50 expression did not up-regulate the promoter of IFN-β , but MAVS wild-type and MAVS70 did ( Fig 2B ) . Consistent with the NF-κB activation , over-expressed MAVS50 also up-regulated the kinase activity of IKKβ by an in vitro kinase assay , in comparison to RIG-I-N and MAVS70 ( Fig 2C ) . TRAF molecules are important adaptors downstream MAVS and are implicated in specific activation of NF-κB and IRF transcription factors . Thus , we examined MAVS50 interactions with a panel of six TRAF molecules by co-IP assays in transfected 293T cells . While MAVS70 interacted with all TRAF molecules except TRAF4 , MAVS50 demonstrated preferential interaction with TRAF2 and TRAF6 ( S1A and S1B Fig ) . This result suggests that MAVS50 , in comparison to MAVS70 , is distinct in interacting with downstream TRAF adaptors . When signaling events of TBK-1 and IRF activation were examined , we found that MAVS50 expression had no detectable effect on the kinase activity of TBK-1 , nor the dimerization of IRF3 ( S1C and S1D Fig ) , supporting the conclusion that MAVS50 does not activate the IRF signaling cascades . As controls , RIG-I-N ( 2CARDs ) and MAVS70 potently activated TBK-1 by kinase assay and induced IRF3 dimerization by native gel electrophoresis . These results collectively show that MAVS50 preferentially activates the IKKβ-NF-κB signaling cascade . MAVS is characterized by a CARD-mediated oligomerization in provoking downstream signaling events [8 , 29] . MAVS50 lacks the CARD domain and we determined whether MAVS50 forms oligomers by size exclusion chromatography . When purified from 293T cells , MAVS70 was eluted in fractions corresponding to ~670 kDa , consistent with the notion that MAVS70 forms large oligomers ( Fig 2D ) . However , MAVS50 was eluted in fractions corresponding to ~120 kDa , suggesting that MAVS50 forms smaller oligomers , likely dimer or tetramer . This result is consistent with the critical roles of CARD domain in mediating large signaling-competent oligomers [30 , 31] . Given that MAVS50 lacks the CARD for interaction with RIG-I , we determined whether MAVS50 could relay signal transduction downstream of RIG-I . Thus , we knocked down the expression of endogenous MAVS ( both isoforms ) ( Fig 2E ) and “reconstituted” MAVS expression with MAVS wild-type , MAVS70 or MAVS50 ( Fig 2F ) . These “reconstituted” cells were then used to examine RIG-I-dependent signaling in response to SeV infection . Upon SeV infection , 293T cells “reconstituted” with wild-type MAVS and MAVS70 activated TBK-1 as determined by TBK-1 phosphorylation at ser172 , a marker for TBK-1 activation ( Fig 2G ) . However , MAVS50 expression failed to do so . When mitochondrion-enriched fractions were analyzed for phosphorylated TBK-1 ( pS172 ) , we found that phosphorylated TBK-1 was abundant in the mitochondrion-enriched fraction from cells “reconstituted” with wild-type MAVS and MAVS70 , but not that of cells expressing MAVS50 ( Fig 2H ) . Consistent with this , “reconstituted” expression of wild-type MAVS and MAVS70 restored robust expression of IFNb , CCL5 , ISG56 , IL-8 and Viperin , indicative of activation of IRF and NF-κB ( Fig 2I and S2A Fig ) . However , “reconstituted” expression of MAVS50 failed to up-regulate the transcription of these anti-viral cytokines . Enzyme-linked immunoassay ( ELISA ) further show that IFN and CCL5 were produced from knockdown cells expressing wild-type MAVS and MAVS70 ( Fig 2J ) . We observed marginal but detectable level of IFNb and significantly ISG56 in resting cells that were “reconstituted” for the expression of MAVS wild-type and MAVS70 . Similar to MAVS wild-type and MAVS70 , MAVS50 localizes to the mitochondrion and peroxisome when expressed in 293T cells ( S2B and S2C Fig ) , consistent with the notion that MAVS is targeted to the mitochondrion via a transmembrane tail [8] . Thus , MAVS50 does not relay signal transduction from RIG-I to downstream molecules , re-enforcing the critical role of CARD in RIG-I-mediated signaling . MAVS50 lacks the CARD domain and fails to relay signal transduction downstream of RIG-I . Considering that MAVS50 possesses most of the sequence of MAVS70 , we reasoned that MAVS50 likely regulates MAVS70-mediated signaling . To test this hypothesis , we first examined whether MAVS50 can interact with MAVS70 and itself . In transfected 293T cells , V5-tagged MAVS70 and MAVS50 were readily detected in protein complexes precipitated with anti-Flag antibody against Flag-MAVS50 ( Fig 3A ) , indicating that MAVS50 can interact with MAVS70 and MAVS50 . This result is consistent with our observation that MAVS50 eluted as ~120 kDa in gel filtration , which implies self-oligomerization of MAVS50 ( Fig 2D ) . Due to the largely overlapping sequence between MAVS70 and MAVS50 , it is technically challenging to probe the interaction between endogenous MAVS70 and MAVS50 . Thus , we established a stable 293 cell line that expresses Flag-tagged MAVS50 under the control of doxycycline in a dose-dependent manner ( S2D Fig ) . Precipitation of MAVS50 effectively pulled down MAVS70 , indicating that MAVS50 physically associates with endogenous MAVS70 ( Fig 3B ) . We further analyzed the interaction between MAVS70 and MAVS50 with gel filtration that is routinely used to assess protein complex formation . Purified MAVS50 was predominantly eluted in fractions corresponding to proteins of ~120 kDa . However , when MAVS50 and MAVS70 were co-expressed in 293T cells , purified MAVS50 was eluted in fractions corresponding to ~670 kDa oligomer ( Fig 3C ) . These results show that MAVS70 can convert MAVS50 into oligomers of larger sizes . Moreover , MAVS70 was also detected in fractions that were enriched with oligomerized MAVS50 , indicating that MAVS70 is integrated in the MAVS50 oligomers and vice versa ( Fig 3C ) . Finally , we assessed the elution pattern of MAVS70 and MAVS50 in lysates of three representative cell lines , including THP-1 monocyte , 293T fibroblast and HeLa cervical epithelial cells , by size exclusion chromatography . MAVS70 and MAVS50 co-eluted in fractions corresponding to ~220–440 kDa in THP-1 monocytes , 293T and HeLa cells ( Fig 3D ) . A notable difference in the elution patterns of MAVS70 and MAVS50 was observed , i . e . , MAVS70 were more evenly distributed in fractions 24 and 26 , whereas MAVS50 was predominantly eluted in fraction 24 . Furthermore , upon Sendai virus infection , MAVS50 increased in fraction 26 that MAVS70 peaked in elution ( S3 Fig ) . Similar result was observed for 293T cells infected with murine gamma herpesvirus 68 , a DNA virus ( S3 Fig ) . These observations suggest that MAVS 50 preferentially associate with the larger size of MAVS70 oligomers . Collectively , these results indicate that MAVS50 physically interacts with MAVS70 . To determine the effect of MAVS50 on MAVS70-mediated signaling , we assessed activation of NF-κB and IFN-β promoter by reporter assays . We found that MAVS50 inhibited MAVS70-induced transcription of the IFN-β promoter in a dose-dependent manner ( Fig 4A ) . By contrast , MAVS50 did not significantly impact the NF-κB activation by MAVS70 ( Fig 4B ) . We noted that MAVS50 was capable of activating NF-κB . We then expressed exogenous MAVS50 either by lentivirus transduction or transient transfection , and examined host cytokine gene expression and viral replication . When MAVS50 was expressed in 293T cells by lentivirus transduction ( Fig 4C ) , while ISG56 expression was not significantly impacted , the expression of IFNβ was reduced by 50% in response to SeV infection ( Fig 4D ) . Conversely , exogenously expressed MAVS50 enhanced the replication of vesicular stomatitis virus ( VSV ) , a prototype RNA virus , by fluorescence microscopy ( Fig 4E ) . Plaque assay further showed that MAVS50 expression increased VSV replication by more than 5-fold ( Fig 4F ) . Taken together , MAVS50 inhibits IFN-β induction in response to viral infection . The cytosolic sensor-mediated IFN induction pathway constitutes of key signaling molecules , including RIG-I/MDA5 , MAVS , TBK-1/IKKε , and IRF3 ( Fig 5A ) . Over-expression of these components is sufficient to activate downstream signaling events , cumulating in the up-regulation of IFN expression . To identify the point of inhibition by MAVS50 , we employed reporter assay that takes advantage of the IFN-β promoter as a surrogate and the over-expressed key components outlined in Fig 5A . While MAVS50 inhibited transcription of the IFN-β promoter induced by RIG-I-N and MAVS70 ( Figs 5B and 3D ) , MAVS50 had no inhibitory effect on the IFN-β promoter induced by TBK-1 , IKKε and the constitutively active IRF3-5D mutant ( Fig 5C and S4A and S4B Fig ) . These results suggest that MAVS50 targets a step between MAVS70 and TBK-1 , a link between the common adaptor and bifurcated downstream signaling events that trigger IRF activation and IFN induction . Considering that MAVS50 contains TRAF2- and TRAF6-binding motifs within its N-terminus and that TRAF molecules serve as link downstream of MAVS , we reasoned that MAVS50 likely targets TRAF molecules to modulate MAVS70-mediated signaling of the IRF branch . We then assessed TRAF6 interaction with MAVS70 and MAVS50 by co-immunoprecipitation . When TRAF6 was precipitated in transfected 293T cells , MAVS50 was readily detected and MAVS70 was detected at background level ( Fig 5D ) . We noted that MAVS70 expression consistently diminished the interaction between MAVS50 and TRAF6 . We further probed TRAF6 interaction with MAVS50 or MAVS70 in 293T cells that were “reconstituted” with exogenous MAVS70 and MAVS50 . Co-IP assays demonstrated that TRAF6 was readily detected when MAVS50 was precipitated in 293T cells ( Fig 5E ) . Although TRAF6 was detected at basal level when co-precipitated with MAVS70 , we believe this is likely due to the strong interaction between MAVS50 and TRAF6 . Interestingly , although MAVS50 can interact with TRAF3 , MAVS50 failed to precipitate with TRAF3 in the presence of MAVS70 ( S4C Fig ) . This result suggests that MAVS50 weakly interacts with TRAF3 . To determine whether MAVS50 can compete with MAVS70 for binding to TRAF6 , we took advantage of the MAVS knockdown cells that were “reconstituted” with exogenous V5-tagged MAVS70 to examine MAVS70 interaction with TRAF6 by Co-IP assay . We found that increasing amount of MAVS50 reduced TRAF6 precipitated with anti-V5 ( MAVS70 ) in a dose-dependent manner ( Fig 5F ) . Interestingly , MAVS50 expressed at a lower level increased the amount of TRAF6 precipitated with MAVS70 ( compare lane 2 and 3 ) , suggesting that MAVS50 integrates into MAVS70 to bridge an interaction between MAVS70 and TRAF6 . With higher levels of MAVS50 , MAVS50 effectively reduced TRAF6 precipitated by MAVS70 , indicative of competition between MAVS50 and MAVS70 for association with TRAF6 ( compare lanes 4 and 5 to lanes 2 and 3 ) . In SeV-infected cells , the effect of MAVS50 on interaction between MAVS70 and TRAF6 was reduced , suggesting that SeV infection partly inhibits MAVS50 action . We also recognized that SeV infection slightly reduced , rather than increased , MAVS70 interaction with TRAF6 . This is likely due to the “reconstituted” expression of MAVS70 that already activated downstream signaling in resting cells ( Fig 2I ) and further activation by SeV infection likely precipitated MAVS70 out from the Triton X-100-soluble fraction . Nevertheless , MAVS50 competes with MAVS70 for binding to TRAF6 . A major TRAF2-binding motif ( PVQE , [P/S/A/T]x[Q/E]E ) locates at the very amino terminus of MAVS50 and a TRAF6-binding motif ( PGENSE , PxExx[Ar/Ac]; Ar , aromatic; Ac , acetic ) [32 , 33] immediately follows the TRAF2-binding motif ( Fig 6A ) . To probe the contribution of these N-terminal TRAF-binding motifs , we mutated the critical residues of TRAF2- and TRAF6-binding motifs into alanines , thus named M2 and M6 of MAVS50 ( S5A Fig ) , and examined their interactions with TRAF2 or TRAF6 by co-IP assay . As expected , mutations within the TRAF2-binding motif abolished MAVS50 interaction with TRAF2 , while mutations within the TRAF6-binding motif had no effect on MAVS50 interaction with TRAF2 ( Fig 6B ) . MAVS50 mutant ablated both TRAF2- and TRAF6-binding motifs lost the interaction with TRAF2 . Surprisingly , mutating the TRAF2-binding motif nearly abolished the MAVS50 interaction with TRAF6 ( Fig 6C ) . Mutations within the immediate downstream TRAF6-binding motif , although reduced MAVS50 interaction with TRAF6 , had less effect than mutations within the TRAF2-binding motif ( Fig 6C ) . Simultaneously mutating the TRAF2- and TRAF6-binding motifs reduced MAVS50 association with TRAF6 to residual level . The residual level of TRAF6 interaction of MAVS50 M2 , 6 mutant is likely due to the transmembrane-proximal TRAF6-binding motif . Additionally , the MAVS50 M2 , 6 mutant also demonstrated reduced interactions with TRAF3 and TRAF5 in transfected 293T cells ( S5B and S5C Fig ) . Mutations within the TRAF2- and TRAF6-binding motifs had no significant effect on MAVS50 interaction with MAVS70 by co-IP assay ( S5D Fig ) , suggesting that these MAVS50 mutants are functionally competent . These results indicate that the very amino-terminal TRAF2-binding motif is critical for binding to both TRAF2 and TRAF6 , and likely other TRAFs , suggesting that the TRAF2-binding motif is a functionally degenerate interaction motif for more than one TRAF molecule . To determine the role of the N-terminal TRAF-binding motifs of MAVS50 , we examined the effect of MAVS50 mutant that harbors mutations in both TRAF2- and TRAF6-binding motifs , designated MAVS50-M2 , 6 . Compared to wild-type MAVS50 , MAVS50-M2 , 6 was significantly impaired to inhibit MAVS70-induced IFN-β expression by reporter assay ( Fig 6D ) . We then expressed exogenous MAVS50 or MAVS50M2 , 6 mutant by lentivirus ( S5E Fig ) and infected these cells with SeV . Real-time PCR analysis showed that the MAVS50-M2 , 6 mutant failed to inhibit IFNb gene expression in response to SeV infection , but wild-type MAVS50 did ( Fig 6E ) . Consistent with this , wild-type MAVS50 expression increased VSV replication , but MAVS50-M2 , 6 mutant completely lost the ability to promote VSV replication ( Fig 6F and 6G ) . These results collectively demonstrate the critical role of the TRAF-binding motifs of MAVS50 , locating within the very N-terminal region , in inhibiting IFN induction downstream of RIG-I . Upon sensing pathogen-associated molecular patterns , pattern recognition receptors initiate signaling events that bifurcate into NF-κB and IRF activation downstream of common adaptor molecules . While NF-κB activation and cytokine production are important for inflammatory response that attracts other immune cells to the site of infection , activated IRF and secreted IFN exert immediate antiviral effect within the site of infection . How NF-κB and IRF activation , triggered by shared upstream signaling molecules such as RIG-I and MAVS , are differentially regulated is not well understood . We report here that the MAVS50 variant , translated from an internal initiation codon , effectively competes with MAVS70 for binding to TRAF molecules . In doing so , MAVS50 inhibits MAVS70-mediated signal transduction , specifically IRF activation and interferon induction . Our work agrees well with a recent study reporting that MAVS50 suppressed RIG-I-dependent IFN induction [28] . Additionally , linear ubiquitination of NEMO , the scaffold protein of IKK and TBK-1 kinase complexes , was previously reported to dampen IFN induction while stimulating NF-κB activation [34] . This activity requires the LUBAC E3 ligase that catalyzes linear ubiquitin chain assembly on NEMO and disrupts MAVS interaction with TRAF3 , which relays signaling from MAVS to TBK-1 and IRF activation . A more recent study showed that cholera toxin induced RIG-I-dependent signaling events with a signature of NF-κB activation , although the molecular mechanism underpinning this preferential activation is not clear [35] . These studies , including our current work , highlight a recurring theme in differential regulation of the two signaling ramifications downstream of shared receptors that sense invading pathogens , pointing to a common shift from IFN induction to a NF-κB-dependent inflammatory response . Under conditions of pathogen infection , it is likely that IFN induction is the immediate robust response of the innate immune phase , whereas NF-κB activation and cytokine secretion constitute a modest but sustained inflammatory response . Given the pro-survival roles of NF-B activation , it is not surprising that viruses often usurp NF-κB activation or upstream signaling events to facilitate their infection , such as HIV and herpesviruses [15 , 36] . By contrast , IRF activation and IFN signaling promote cell death . In contrast to what was reported by Brubaker et al . , we did not observe cell death induced by MAVS70 and MAVS50 . This discrepancy may stem from the difference in our experimental conditions . Despite of missing the N-terminal CARD domain that mediates hetero-oligomerization with RIG-I and self-oligomerization of MAVS [30 , 31] , MAVS50 forms small oligomers that correspond to the size of a dimer or tetramer analyzed by gel filtration . Consistent with that , MAVS50 can homo-dimerize as determined by co-immunoprecipitation . This intriguing observation suggests the existence of unknown sequence that , in addition to CARD , mediates MAVS dimerization . To define a dimerization domain , we have applied serial truncations from the N-terminus of MAVS50 and performed co-IP assays . Unfortunately , we failed to pinpoint a key homo-dimerization sequence , implying that MAVS50 homo-dimerization requires a structural sequence , rather than the primary linear sequence . Alternatively , other cellular factors , including mitochondrial membrane , may scaffold the homo- and hetero-dimerization of MAVS70 and MAVS50 . Nevertheless , MAVS50 is prone to form oligomer and , when it is over-expressed , is sufficient to trigger NF-κB activation . On the other hand , “reconstituted” expression of MAVS50 in cells that endogenous MAVS isoforms , both MAVS70 and MAVS50 , were depleted by shRNA-mediated knockdown failed to trigger the expression of inflammatory cytokines and IFN-β in response to SeV infection . This result indicates that MAVS50 can not relay signal transduction from RIG-I to NF-κB and IRF transcription factors and requires MAVS70 to do so , re-enforcing the critical role of the CARD domain in assembling the RIG-I-MAVS signaling platform . Indeed , MAVS50 interacts with MAVS70 and MAVS70 expression converted MAVS50 from oligomers of ~120 to those of ~670 kDa . When purified MAVS50 was analyzed by gel filtration , MAVS70 was detected in fractions that were enriched for oligomerized MAVS50 . Thus , MAVS70 can incorporate into MAVS50 oligomers , and vice versa . These activities enable MAVS50 to serve as a modulator of the MAVS-dependent immune pathways via interaction with MAVS70 and TRAF molecules , key amplifiers at the crossroad in innate immune signaling . However , MAVS70 expression diminished the interaction between MAVS50 and TRAF molecules , including TRAF6 and TRAF3 . These results suggest that the innate immune signaling , activated by MAVS70 expression , is capable of inactivating MAVS50 and potentially releasing the MAVS50-mediated inhibition . This hypothesis remains to be examined in the near future . How does MAVS50 differentially alter MAVS70-dependent signaling , i . e . , inhibiting IRF activation and IFN induction while weakly stimulating NF-κB activation ? Based on our findings , we propose the following hypothetical model that summarizes the action of MAVS50 in specific inhibiting IFN induction ( Fig 7 ) . Upon stimulation such as activated RIG-I , MAVS70 forms large oligomers in the form of prion-like polymers or fibrils , resulting potent activation of both NF-κB and IRF transcription factors . By positioning TRAF-binding motifs at the very N-terminus , MAVS50 interacts more strongly with TRAF6 , and likely TRAF2 , than MAVS70 . In doing so , MAVS50 efficiently sequesters TRAF6 from the prion-like MAVS70 polymers , attenuating the MAVS70-mediated signaling . When activated MAVS70 undergoes oligomerization , MAVS50 is induced to oligomerize via interacting with MAVS70 . Given its high affinity for TRAF adaptor molecules , oligomerized MAVS50 is sufficient to induce NF-κB activation , but not IRF activation and IFN induction . Sum of both results is the specific inhibition of IFN induction and modest NF-κB activation by MAVS50 . Then , how does MAVS70 activate both NF-κB and IRF , while MAVS50 activates only NF-κB . In SeV-infected or vGAT-expressing cells , MAVS70 migrated into the Triton X-100-insoluble fraction , whereas MAVS50 remained in the soluble fraction . In transfected 293T cells , MAVS50 forms smaller oligomers than MAVS70 . These findings largely agree with the notion that higher order of MAVS70 oligomers form fibrils and precipitates out from Triton X-100-containing solution , whereas MAVS50 does not [8 , 31] . Together with TBK-1 that phosphorylates an IRF3-binding domain , MAVS70 fibrils provide a signaling platform that enables IRF3 activation and IFN induction [37] . Thus , it is conceivable that MAVS70 and MAVS50 form at least two types of oligomers that are of distinct sizes . The large oligomeric MAVS70 is capable of activating NF-κB and IRF , while the smaller MAVS50-containing oligomer only activates NF-κB . Perhaps , integration of MAVS50 into the MAVS70 larger oligomers shifts the signaling capacity of the complex from activating both NF-κB and IRF to that activating only NF-κB . This possibility remains to be formally tested in the future . Compared to MAVS70 , MAVS50 preserves all three TRAF-binding motifs , i . e . , TRAF2- and TRAF6-binding motifs . Indeed , MAVS50 demonstrated preferential interaction with TRAF2 and TRAF6 , while MAVS70 interacted with five TRAFs except TRAF4 . This result agrees with the notion that the N-terminally exposed TRAF-binding motifs are better accessed and endow MAVS50 better interactions with TRAF2 and TRAF6 . Intriguingly , mutations within the putative TRAF2-binding motif , but not those within the predicted TRAF6-binding motif , abolished MAVS50 interaction with both TRAF2 and TRAF6 . Similarly , the MAVS50 M2 , 6 mutant was impaired to interact with TRAF3 and TRAF5 . These results highlight the critical role of the very N-terminally exposed TRAF2-binding motif in interacting with more than one TRAF molecule , implying the degeneracy of the TRAF2-binding motifs in recruiting TRAF molecules and relaying signal transduction . It is important to note that , compared to MAVS50 , these MAVS50 mutants were expressed well and interacted with MAVS70 , suggesting that the defect in TRAF-binding is not due to overall protein misfolding . Alternatively , it is possible that TRAF2 recruited by the terminally positioned TRAF2-binding motif facilitates the interaction of MAVS50 with TRAF6 and other TRAFs [7 , 34] . This is supported by the observation that MAVS recruits multiple TRAFs with non-redundant roles in innate immune signaling [38] and TRAF molecules are prone to oligomerize [39] . The space between these two TRAF-binding motifs may not permit two TRAF molecules to dock on the N-terminus of MAVS50 , but multiple TRAFs decorating on oligomerized MAVS50 are possible [32 , 33] . Nevertheless , by positioning the TRAF2-binding motif at the very N-terminus , MAVS50 effectively competes with MAVS70 for recruiting these key signaling molecules , resulting in attenuation of the MAVS70-mediated signal transduction . The MAVS50 variant is generated from internal translation initiation , such that two TRAF2-binding motifs are exposed at the very N-terminus to facilitate protein interaction . Such a delicate mechanism , crafted by millions of years of evolution , symbolizes the importance of the tight regulation of MAVS-dependent innate immune signaling in response to viral infection . Notable regulatory action alike is exemplified by alternatively spliced variant of RIG-I [40] and TRIF [41] , and a LGP2 allele likely due to gene duplication [42 , 43] . While alternative splicing has been extensively studied , the contribution of internal translation initiation in regulating fundamental biological processes such as immune response is not well understood . On the other hand , internal translation initiation is one of the mechanisms that viruses frequently deploy to maximize the coding capacity of their limited genomes [44 , 45] and that critical cellular genes avoid shutdown under stressed conditions [46 , 47] . It is probable that internal translation initiation of MAVS50 is activated to enhance its expression in virus-infected cells that the cap-dependent translation is suppressed . Expression of MAVS50 will alleviate and prevent an overacting antiviral immune response , thereby promoting cell survival . The roles of MAVS50 under this stressed condition remain to be determined . Unless otherwise specified , all genes were cloned into pcDNA5/FRT/TO ( Invitrogen ) for transient expression , and pCDH-EF-puro-CMV-MCS ( System Bioscience ) for lentiviral expression . All cloned cDNAs were confirmed by DNA sequencing . HEK 293T ( ATCC ) , HEK 293T-Rex cells ( Invitrogen ) , BHK21 ( ATCC ) , THP-1 ( ATCC ) were maintained in Dulbecco’s modified Eagle’s medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) , penicillin ( 100 U/mL ) , and streptomycin ( 100 μg/mL ) . Human Jurkat ( ATCC ) T lymphoid cells were maintained in RPMI 1640 supplemented with 10% fetal bovine serum ( FBS ) , penicillin ( 100 U/mL ) , and streptomycin ( 100 μg/mL ) . All cells were cultured at 37°C in an atmosphere of 5% CO2 . VSV-GFP virus was amplified in BHK-21 cells . Viral titers were determined by a plaque assay using NIH3T3 monolayer . The following antibodies were used in this study: Anti-human MAVS1-135 ( Santa Cruz Biotechnology ) , anti-human MAVS150-250 ( Abcam ) , anti-human MAVS150-200 ( Bethyl group ) , anti-β-actin ( Abcam ) , anti-Flag ( Sigma ) , anti-V5 ( Bethyl Group ) , anti-HA ( Covance ) . Antibodies against IKKβ and IKKγ were kindly provided by Dr . Ebrahim Zandi ( University of Southern California ) . Luciferase reporter assays were performed as previously described [48] . Briefly , HEK293T cells ( 1 x 105 cells/well ) were seeded in 24-well plates 16 hours prior to transfection . Cells were transfected with NF-ĸB or IFN-β reporter plasmid cocktail ( including 50 ng of NF-ĸB or IFN-β promoter luciferase reporter plasmid and 100-ng of pGK-β-GAL plasmid ) and an expression plasmid , by calcium phosphate transfection method . At 30 hours post-transfection , cell lysates were used to measure the firefly luciferase activity and β-galactosidase activity . Immunoprecipitation and immunoblotting were carried out as previously described [22 , 23 , 49] . Briefly , cells were harvested , rinsed with ice-cold PBS , and lysed with NP40 buffer ( 50 mM Tris-HCL [pH 7 . 4] , 150 mM NaCl , 5 mM EDTA , 1% NP40 ) supplemented with protease inhibitor cocktail . Centrifuged cell lysates were then pre-cleared with Sepharose 4B beads , and subjected to precipitation with antibody-conjugated agarose ( Sigma ) at 4°C for 4–6 hours . Precipitated proteins were extensively washed with NP40 buffer and eluted with 1x SDS-PAGE loading buffer by boiling at 95°C for 5–10 minutes . For immunoblotting analysis , whole cell lysates ( WCL ) or precipitated proteins were resolved by SDS-PAGE , and transferred to nitrocellulose membrane . Immunoblotting analysis was performed with indicated primary antibodies and proteins were visualized with IRDye800- or IRDye680-conjugated secondary antibodies ( Licor ) using an Odyssey infrared imaging system ( Licor ) . Gel filtration was performed as previously described by Zandi et al [11] . Briefly , WCL or purified proteins were applied to superpose 6 or superdex 200 column ( GE Bioscience ) and subjected to gel filtration analysis with buffer B ( 1 mM EDTA , 0 . 5 mM EGTA , 150 mM NaCl , 20 mM Tris-HCl [pH 7 . 6] , 0 . 5% Triton X-100 , 20 mM NaF , 20 mM β-glycerolphosphate , 1 mM Na3VO4 , 5 mM benzamidine , 2 . 5 mM metabisulphite ) . Elutions were collected in 0 . 5 ml fractions and were analyzed by immunoblotting . qRT-PCR was performed as previously described [22 , 23] . Briefly , total RNA was extracted from HEK293T cells using TRIzol reagent ( Invitrogen ) . To remove genomic DNA , total RNA was digested with RNase-free DNase I ( New England Biolab ) . First-strand cDNA was synthesized from 1 μg total RNA , using reverse transcriptase ( Invitrogen ) . The abundance of cytokine mRNA was assessed by qRT-PCR , using SYBR Green Master Mix ( Applied Biosystems ) . Human β-Actin was used as an internal control . Commercial ELISA kits used in this study include: Human IFNβ ( Thermo Scientific ) and human CCL5 ( R&D Systems ) . The supernatants from cultured cells were collected at the indicated time points after Sendai Virus infection . Cytokine levels in the supernatants were assessed according to manufacturer’s instruction . HEK293T cells were transfected with the indicated plasmids . Cells were lysed and anti-IKKγ antibody was used to precipitate endogenous IKKα/β/γ complexes . The precipitated complexes were subjected to in vitro kinase assay as previously described [49] . The kinase reaction mixture consisted of GST-IĸBαNT as the substrate , [γ-32P]ATP , and precipitated kinase complex in 20 μl kinase buffer ( 25 nM HEPES [pH 7 . 5] , 50 mM KCL , 2 mM MgCl2 , 2 mM MnCl2 , 1 mM DTT , 10 mM NaF , 20 mM β-glycerolphosphate , and 1 mM sodium orthovandate ) . The reaction mixture was incubated at room temperature for 40 minutes . Phosphoryation of IĸBα was analyzed by autoradiography . Lentivirus production was performed as previously described [22 , 48] . Briefly , HEK293T cells were transfected with packaging plasmids ( DR8 . 9 and VSV-G ) and the pCDH lentiviral expression plasmids or shRNA plasmids . At 72 hours post-transfection , supernatant was harvested and , if necessary , concentrated by ultracentrifugation . HEK293T cells were then infected with lentivirus in the presence of polybrene ( 8 μg/ml ) . Cells were selected and maintained in complete media . Cells were lysed and homogenized using hypotonic buffer solution ( 20 mM Tris-HCl , pH 7 . 4 , 10 mM NaCl , 3mM MgCl2 ) . The homogenates were centrifuged at 500xg for 5 minutes . The supernatant ( S1 ) was centrifuged at 5000xg for 10 minutes to precipitate crude mitochondria . The crude mitochondria fraction ( P5 ) was then lysed and analyzed by immunoblotting . The statistical significance ( P-value ) was calculated using unpaired two-tailed Student’s t test . A P-value of <0 . 05 was considered statistically significant .	Host innate immune signaling plays critical roles in defeating pathogen infection . In response to viral infection , cellular signaling events cumulate in the activation of NF-κB and interferon regulatory factors . How these two signaling ramifications are differentially regulated remains an open question . Here we report an internally translated MAVS variant deregulates IRF activation via exposing N-terminal TRAF-binding motifs . As such , the short form of MAVS efficiently competes for binding to TRAF2 and TRAF6 against full-length MAVS , thereby sequestering key adaptors from the signaling cascades mediated by full-length MAVS . Our study uncovers a delicate regulatory mechanism of truncated proteins bearing key protein-interacting motifs that is enabled by internal translation initiation and potentially other relevant means .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[]	2015	An Internally Translated MAVS Variant Exposes Its Amino-terminal TRAF-Binding Motifs to Deregulate Interferon Induction
Gene expression is controlled by the combinatorial effects of regulatory factors from different biological subsystems such as general transcription factors ( TFs ) , cellular growth factors and microRNAs . A subsystem’s gene expression may be controlled by its internal regulatory factors , exclusively , or by external subsystems , or by both . It is thus useful to distinguish the degree to which a subsystem is regulated internally or externally–e . g . , how non-conserved , species-specific TFs affect the expression of conserved , cross-species genes during evolution . We developed a computational method ( DREISS , dreiss . gerteinlab . org ) for analyzing the Dynamics of gene expression driven by Regulatory networks , both External and Internal based on State Space models . Given a subsystem , the “state” and “control” in the model refer to its own ( internal ) and another subsystem’s ( external ) gene expression levels . The state at a given time is determined by the state and control at a previous time . Because typical time-series data do not have enough samples to fully estimate the model’s parameters , DREISS uses dimensionality reduction , and identifies canonical temporal expression trajectories ( e . g . , degradation , growth and oscillation ) representing the regulatory effects emanating from various subsystems . To demonstrate capabilities of DREISS , we study the regulatory effects of evolutionarily conserved vs . divergent TFs across distant species . In particular , we applied DREISS to the time-series gene expression datasets of C . elegans and D . melanogaster during their embryonic development . We analyzed the expression dynamics of the conserved , orthologous genes ( orthologs ) , seeing the degree to which these can be accounted for by orthologous ( internal ) versus species-specific ( external ) TFs . We found that between two species , the orthologs have matched , internally driven expression patterns but very different externally driven ones . This is particularly true for genes with evolutionarily ancient functions ( e . g . the ribosomal proteins ) , in contrast to those with more recently evolved functions ( e . g . , cell-cell communication ) . This suggests that despite striking morphological differences , some fundamental embryonic-developmental processes are still controlled by ancient regulatory systems . Gene regulatory networks systematically control the gene expression dynamics . These networks are highly modular , and consist of various sub-networks . Each sub-network contains a number of regulatory factors representing a subsystem that drives specific gene regulatory functions [1 , 2] . The subsystems interact with one another , and work together to carry out the entire gene regulatory function . For example , the gene expression in embryogenesis is controlled by the combinatorial effects of various regulatory subsystems composed of complex evolutionary regulatory networks [3] . These regulatory subsystems drive very diverse developmental programs , from the highly conserved ( e . g . DNA replication ) to the species-specific ( e . g . body segmentation ) . As such the orthologous genes that are evolutionary conserved genes across species can therefore be regulated by both orthologous and species-specific transcription factors ( TFs ) [4] . The orthologous TFs form an “internal” regulatory network , while the species-specific TFs form an “external” one . Unfortunately , existing experimental gene expression data cannot decouple the expression components that are driven by the different subsystems . Thus , computational methods are required to assess the contribution from each factor or subsystem from the gene expression data . In this study , we propose a novel computational method , DREISS—dynamics of gene expression driven by external and internal regulatory networks based on state space model . Using DREISS , we are able to identify temporal gene expression dynamic patterns for evolutionarily conserved genes during embryonic development , as driven by conserved and species-specific regulatory subsystems . These results advance our current understanding of gene regulatory networks during evolution , as well as the differentiation during development . Developmental gene regulatory networks control gene expression during the developmental processes . These particular regulatory networks have evolved , making it difficult to understand their regulatory mechanisms at the system level . Hence , one typically compares developmental gene expression across species to infer biological activities of developmental gene regulatory networks . For example , embryogenesis provides a platform to study the evolution of gene expression between different species . Recent work has showed that significant biological insight can be gained by cross-species comparisons of the expression profiles during embryogenesis for worms [5] , flies [6] , frogs [7] and several other vertebrates [8] . It was found that the orthologous genes have minimal temporal expression divergence during the phylotypic stage , a middle phase during the embryonic development across species within the same phylum . These patterns are often characterized as “hourglass” [9] . In addition , the conserved hourglass patterns were observed even within a single species while comparing the developmental gene expression data across distant species , such as worm and fly [10]; i . e . , the expression divergence among evolutionarily conserved genes become minimal during the phylotypic stage in both worm and fly . However , much less is known about how the orthologous genes in each species eventually contribute to their species-specific phenotypes due to the lack of appropriate computational approaches . Thus , we aim to use DREISS to discover the components of the orthologous gene expression during embryonic development driven by species-specific transcription factors . The state-space model has been widely used in engineering [11] , and also in biology for the analysis of gene expression dynamics [12–14] . It models the dynamical system output as a function of both the current internal system state and the external input signal . A well-known example in engineering is the vehicle cruise control system where the system state can be the vehicle’s speed . Based on the road conditions , the cruise control requires various fuel amounts in order to keep the desired speed level . In biology , we can look at the transcription factors and microRNAs as internal and respectively external regulatory factors of the protein-coding genes expression ( See more internal-external examples in S1 Table ) . Similarly , the state-space model can be applied for studying the expression of orthologous genes at different developmental stages using information regarding their expression ( internal ) and species-specific regulatory factors ( external ) at the current known developmental stage . Unlike earlier studies that calculate the expression correlation between individual genes , the state-space model predicts the temporal causal relationships at the system level; i . e . , the state at a time is determined by the state and external input at the previous time . The earlier work applied the state-space model to study the gene expression dynamics focusing on small-scale systems , and did not explore the analytic dynamic characteristics of the inferred state-space models . The complex and large-scale biological datasets , especially temporal gene expression data , are very noisy , and high dimensional ( i . e . , the number of genes is much greater than the number of time samples ) , thereby preventing an accurate estimation of the state-space model’s parameters . The dimensionality reduction techniques have thus been used to project high-dimensional genes to low-dimensional meta-genes ( i . e . , the selected features representing de-noised and systematic expression patterns [1 , 15 , 16] ) as well as the principal dynamic patterns for those meta-genes [17 , 18] . Using DREISS , we are able to apply the dimensionality reduction to the gene expression data , and develop an effective state-space model for their meta-genes , and then identify a group of canonical temporal expression trajectories representing the dynamic patterns driven by the effective conserved and species-specific meta-gene regulatory networks according to the model’s analytic characteristics . These dynamic patterns reveal temporal gene expression components that are controlled by conserved or species-specific GRNs . DREISS is a general-purpose tool and can be used to study the gene regulatory effects from any different subsystems for a given group of genes . As an illustration , we applied DREISS to the gene expression data during embryonic development for two model organisms , worm ( Caenorhabditis elegans ) and fly ( Drosophila melanogaster ) . In both species , we were able to identify the expression patterns of worm-fly orthologs driven by the conserved regulatory network consisting of the worm-fly orthologous TFs ( i . e . , the conserved regulatory subsystems between two species ) , as well as the worm/fly-specific regulatory network consisting of non-orthologous TFs ( i . e . , the species-specific regulatory subsystem ) . Our results reveal that , in addition to executing conserved developmental functions between worm and fly , their orthologous genes are also regulated by species-specific TFs to involve in species-specific developmental processes . In summary , DRIESS provides a framework to analyze both distantly and closely related species allowing for a better understanding of the gene regulatory mechanisms during development . A gene regulatory network is made up of various subsystems [1 , 2] . These subsystems work together to execute regulatory functions . Given a group of N1 genes in a subsystem , defined as the internal gene set , Ω , their gene expression levels are not only controlled by internal interactions among Ω , but also affected by the regulatory factors from other subsystems outside Ω . We define an external gene set , Ψ consisting of those external regulatory factors . For example , we consider the worm-fly orthologous genes as internal set Ω . The worm-fly orthologous TFs from internal set Ω are the internal regulatory factors , and non-orthologous TFs such as worm- or fly- specific TFs are the external regulatory factors . Both the internal and external regulatory factors control gene expressions in dynamic ways ( i . e . , their regulatory signals at the current time will affect gene expressions at subsequent times ) . Thus , the regulatory mechanisms for gene expressions form a control system . In this study , we used a state-space model ( defined by linear first-order difference equations , Fig 2A ) to formulate temporal gene expression dynamics for internal set Ω ( comprising N1 genes ) with external regulation from external set Ψ ( comprising N2 genes ) at time points 1 , 2 , … , T as follows: Xt+1=AXt+BUt ( 1 ) , where the vector Xt∈RN1×1 , the “state” , includes N1 gene expression levels at time t in Ω , and the vector Ut∈RN2×1 , the “input or control” , includes N2 gene expression levels at time t in Ψ . The system matrix A∈RN1×N1 captures internal causal interactions among genes in Ω ( i . e . , the ith , jth element of A , Aij describes the contribution from the jth gene expression at time t to the ith gene expression at the next time t+1 ) , which instantiates a gene regulatory network . The control matrix B∈RN1×N2 captures external causal regulations from the genes in Ψ to genes in Ω ( i . e . , the ith , jth element of B , Bij describes the contribution from the jth gene expression in Ψ at time t to the ith gene expression in Ω at the next time t+1 ) . R represents the real number domain . According to the state space model ( 1 ) , the gene expression dynamics in Ω is determined by the system matrix A and the control matrix B . In particular , based on Eq 1 , the state Xt can be expanded as follows: Xt=AXt−1+BUt−1=A ( AXt−2+BUt−2 ) +BUt−1=A2Xt−2+ABUt−2+BUt−1=A3Xt−3+A2BUt−3+ABUt−2+BUt−1=⋯=At−1X1+At−2BU1+At−3BU2+⋯+ABUt−2+BUt−1=At−1X1⏟XtINT+∑k=1t−2AkBUt−1−k⏟XtINTER+BUt−1⏟XtEXT ( 2 ) , where XtINT=At−1X1 is defined as the expression vector of the gene components driven only internally by genes in Ω . The rest terms ∑k=1t−2AkBUt−1−k+BUt−1 captures the expression expression vector of the gene components in Ω affected externally by the genes in Ψ . In particular , XtEXT=BUt−1 represents the expression vector of gene components in Ω driven purely by the genes in Ψ since it only involves B and U , and XtINTER=∑k=1t−2AkBUt−1−k captures the expression vector of gene components in Ω driven by the interactions between internal and external groups for involving A , B and U . In this paper , we mainly focus on the purely internal dynamics . As for the external-related terms , we should emphasize that any subdivision of the rest of the terms ∑k=1t−2AkBUt−1−k+BUt−1 is completely arbitrary . That is , although we subdivided it into a purely external term and an interaction term here , one could subdivide it multiple ways . That is , given a particular type of subdivision , each of the subdivided terms sums up a group of terms from AkBUt−1−k , k = 0 , 1 , 2 , … , t-2 . For example , one can look at ∑k=2t−2AkBUt−1−k+ ( ABUt−2+BUt−1 ) , where ABUt−2 + BUt−1 shows the contribution from the inputs up to two time steps back to Xt . The temporal gene expression experiments normally have limited time samples ( for example , there may only be a dozen time points ) , which are far less than the time samples needed to estimate the large matrices A and B when internal and external groups , Ω and Ψ are composed of hundreds or thousands of genes . One way to deal with lack of time samples is dimensionality reduction . Thus , we project high dimensional temporal gene expressions to much lower dimensional meta-gene expression levels using a dimensionality reduction technique ( Fig 2B ) . Those meta-gene expression levels should capture original gene expression patterns , such as the ones having the greatest degree of co-variation . We calculate the meta-gene expression levels as follows: X˜t=WXXt;U˜t=WUUt ( 3 ) , where X˜t∈RM1×1 , the “meta-gene state” at time t , includes M1 ( << N1 and <T ) meta-gene expression levels; i . e . , the first M1 elements of the tth row of the matrix whose columns are right-singular vectors of the matrix [X1 X2 ⋯ XT] in Ω by the singular value decomposition ( SVD ) [19]; the vector U˜t∈RM2×1 , the “meta-gene input or control” at time t , includes M2 ( << N2 and <T ) meta-gene expression levels; i . e . , the first M2 elements of the tth row of the matrix whose columns are right-singular vectors from SVD of the matrix [U1 U2 ⋯ UT] in Ψ; WX∈RN1×M1 is the linear projection matrix of SVD from M1 meta-gene expression space to N1 gene expression space in Ω , WU∈RN2×M2 is the linear projection matrix of SVD from M2 meta-gene expression space to N2 gene expression space in Ψ , and ( . ) * is a pseudo-inverse operation; i . e . , WW = I , where I is the identity matrix . Next , we obtain the effective state-space model for meta-genes using linear projections WX and WU between genes and meta-genes as follows ( Fig 2C ) . By replacing ( 1 ) using ( 3 ) , we obtain that WXX˜t+1=AWXX˜t+BWUU˜t . ( 4 ) After multiplying the pseudo-inverse of WX , WX∈RM1×N1 s . t . WXWX=I where I is an identity matrix , at both sides of ( 4 ) , we have that X˜t+1=WXAWX⏟A˜X˜t+WXBWU⏟B˜U˜t=A˜X˜t+B˜U˜t ( 5 ) , where the effective meta-gene system matrix A˜=WXAWX∈RM1×M1 captures internal causal interactions among meta-genes in Ω ( i . e . , an element of A˜ , A˜ij describes the contribution from the jth meta-gene expression at time t to ith meta-gene expression at time t+1 ) , and the effective control matrix B˜=WXBWU∈RM1×M2 captures external causal regulations from meta-genes of Ψ to meta-genes of Ω ( i . e . , the ith , jth element of B˜ , B˜ij describes the contribution from the jth meta-gene expression in Ψ at time t to ith meta-gene expression in Ω at time t+1 ) . Eq 5 describes the effective state space model for the meta-genes of Ω , whose expression dynamics is determined by A˜ and B˜ . Because the meta-gene dimension , M1 ( M2 ) is less than T , and much less than N1 ( N2 ) , we can estimate A˜ and B˜ as follows . We rewrite Eq 5 as a matrix product on the right side: X˜t+1=A˜X˜t+B˜U˜t=[A˜B˜][X˜tU˜t] . ( 6 ) By applying Eq 6 to time points , 2 , 3 , … , T , we then obtain that [X˜2X˜3⋯X˜T]⏟Ζ=[A˜B˜][X˜1X˜2⋯X˜T−1U˜1U˜2⋯U˜T−1]⏟Υ ( 7 ) , where Ζ∈RM1× ( T−1 ) and Υ∈R ( M1+M2 ) × ( T−1 ) . Because of dimension reduction , Υ has more columns than rows so that it has right pseudo-inverse . Thus , the effective internal system matrix A˜ and external control matrix B˜ can be estimated by: [A˜B˜]=ΖΥ ( 8 ) , where Υ∈R ( T−1 ) × ( M1+M2 ) is the right pseudo-inverse of Υ; i . e . , ΥΥ = I , with M1<N1 , M2<N2 , M1+M2<T , t = 1 , 2 , … , T . It is worth noting that if we do not reduce the dimensionality , and obtain Eq 7 from Eq 5 , then Υ will have much more rows than columns so that it doesn’t have right pseudo-inverse; i . e . , there doesn’t exist a matrix Υ* such that ΥΥ* is a full-rank identify matrix . In addition , the condition of M1+M2<T also makes ΥΥ* be a full-rank identify matrix . The analytic solution to a general first-order linear matrix difference equation [20] , Qt+1 = CQt is Qt = CtQ0 = ( HEH-1 ) tQ0 = HEtH-1Q0 = HEtS , where the columns of the matrix H are eigenvectors of C , the diagonal elements of the diagonal matrix E are eigenvalues of C such that CH = HE , and the vector S = H-1Q0 . Then , if we rewrite Qt by a linear combination of the time exponential of eigenvalues of C , we have that Qt=HEtS=∑i=1mcαitsiHi=∑i=1mcαitKi , where mc is the total number of eigenvalues of C , αi is the ith eigenvalue of C , si is the ith element of S , Hi is the ith eigenvector of C ( i . e . , the ith column of H ) , and Ki = siHi is the coefficient vector of Qt over the tth time exponential of αi . By Eq 5 , the matrix Ã determines the meta-gene states components whose expression dynamics are internally controlled by the meta-genes of Ω . As Eq 2 , we define the expression of the meta-gene components driven only internally by themselves in Ω at time t as X˜tINT , an M1-dimensional vector; i . e . , their expression at two adjacent time points have X˜t+1INT=A˜X˜tINT∈RM1×1 . According to the above analytic solution , it can be a linear combination of M1 dynamic patterns determined by the eigenvalues of the effective system matrix A˜ as follows: X˜tINT=∑p=1M1λptK˜p; i . e . , the internally driven component of ith meta-gene’s expression across all time points , [X˜1INT ( i ) X˜2INT ( i ) …X˜TINT ( i ) ]=∑p=1M1K˜p ( i ) [λp1λp2…λpT]⏟pthiPDP ( 9 ) , where λp and K˜p∈CM1×1 are the pth eigenvalue of A˜ and its coefficient vector from the analytic solution , which determines the pth dynamic pattern driven by effective internal regulations , defined as the pth internal principal dynamic pattern ( iPDP ) = [λp1λp2…λpT] , in which λpt represents the tth power of λp , and Ξ ( i ) represents ith element of the vector Ξ . C represents the complex number domain . If an eigenvalue λ is complex when A˜ is asymmetric , then its conjugate λ¯ is also an eigenvalue , so we sum its iPDP and its conjugate eigenvalue , λ¯’s iPDP , as a unified iPDP with real elements equal to [λp1+λ¯p1λp2+λ¯p2…λpT+λ¯pT] . The internal principal dynamic patterns ( iPDPs ) represent canonical temporal expression trajectories , which can be either increasing , or damped oscillation and so on depending on iPDP’s eigenvalues ( Fig 3 ) . The iPDPs can be ordered by sorting their eigenvalues . Also by Eqs 2 and 5 , the expression of the meta-gene states components driven purely by the external group Ψ at time t is defined as X˜tEXT , an M1-dimensional vector , and its expression dynamics is determined by the equation X˜t+1EXT=B˜U˜t∈RM1×1; i . e . , the externally driven components of meta-gene states at two adjacent time points . In particular , the externally driven component of ith internal meta-gene’s expression across time points: [X˜2EXT ( i ) X˜3EXT ( i ) …X˜TEXT ( i ) ]=∑q=1M2B˜i , q[U˜1 ( q ) U˜2 ( q ) …U˜T−1 ( q ) ]⏟qthePDP ( 10 ) , where X˜tEXT ( i ) and U˜t ( q ) are ith and qth elements of X˜tEXT and U˜t , respectively with t = 1 , 2 , … , T , the vector [U˜1 ( q ) U˜2 ( q ) …U˜T−1 ( q ) ]⏟ is defined as qth external principal dynamic pattern ( ePDP ) , and B˜i , q is the element of B˜ at ith row and qth column , which is also the coefficient of the externally driven component of ith internal meta-gene’s expression over qth ePDP . Based on Eq 2 , the expression of the meta-gene components driven by the interactions between internal and external meta-genes is given by X˜tINTER=∑k=1t−2A˜kB˜U˜t−1−k . In this paper , we focus on the purely driven internal patterns ( i . e . , iPDPs ) and compare them across different biological systems . Because genes and meta-genes have linear relationships in terms of their expression levels as described in Eq 2 , the components of gene expression levels in Ω driven by internal regulations , XtINT∈RN1×1 can be also expressed as linear combinations of M1 iPDPs: XtINT=WXX˜tINT=∑p=1M1λptWXK˜p⏟Cp=∑p=1M1λptCp;i . e . , the internally driven component of ith gene’s expression across all time points , [X1INT ( i ) X2INT ( i ) …XTINT ( i ) ]=∑p=1M1Cp ( i ) [λp1λp2…λpT]⏟pthiPDP ( 11 ) , where Cp=WXK˜p∈CM1×1 represents the gene coefficient vector for pth iPDP . Similarly , the gene expression components driven by external genes in Ψ , XtEXT∈RN1×1 can be also expressed as linear combinations of M2 ePDPs: XtEXT=WXX˜tEXT=WXB˜⏟DU˜t=DU˜t;i . e . , the externally driven component of ith gene’s expression across all time points , [X2EXT ( i ) X3EXT ( i ) …XTEXT ( i ) ]=∑q=1M2Di , q[U˜1 ( q ) U˜2 ( q ) …U˜T−1 ( q ) ]⏟qthePDP ( 12 ) , where XtEXT ( i ) is ith element of XtEXT with t = 1 , 2 , … , T , and Di , q is the element of D=WXB˜ at ith row and qth column , which is also the coefficient of the externally driven component of ith gene’s expression over qth ePDP . DREISS enables us to compare expression dynamic patterns between two or more temporal gene expression datasets even though they have different numbers of samples , as well as differences in the times at which those samples were collected . For example , we can apply DREISS to two different datasets of the same group of genes , and identify both the common ( similar ) and the specific ( different ) dynamic patterns driven by internal regulations captured by the eigenvalues of the effective system matrices between the two datasets . In this paper , we apply DREISS to 3 , 153 one-to-one orthologous genes between worm ( Caenorhabditis elegans ) and fly ( Drosophila melanogaster ) as internal group , Ω to study their expression dynamics during embryonic development [10] . We refer to species-specific TFs as external regulations; i . e . , external group Ψ . We found that worm-fly orthologs have similar internal dynamic patterns , which may be mainly driven by conserved TFs , but have very different external dynamic patterns driven by species-specific TFs between worm and fly embryonic developmental stages . The data is summarized as follows . We define internal group Ω as 3 , 153 one-to-one orthologous genes between worm and fly during embryonic development , and external group Ψ as all the species-specific TFs ( 509 worm-specific TFs , 442 fly-specific TFs ) [21 , 22] . We used their temporal gene expression levels ( as measured by the RPKM values in RNA-seq ) during embryonic development from the modENCODE project [10] . The worm embryonic development dataset includes T = 25 time stages at 0 , 0 . 5 , 1 , 1 . 5 , … , 12 hours , and the fly dataset includes T = 12 time stages at 0 , 2 , 4 , … , 22 hours , but t = 1 , 2 , . . , 25 for worm and t = 1 , 2 , … , 12 for fly are used in this paper , representing the relative time points for the entire embryonic development processes . Because M1+ M2<T in Eq 8 , we choose M1 = M2 = 5 meta-genes for fly ( T = 12 ) , and find that five meta-genes of Ω and five meta-genes of Ψ capture ~98% of the co-variation of orthologous gene expressions and fly-specific TF gene expressions , respectively . In order to compare worm and fly , we also choose M1 = M2 = 5 meta-genes for worm , which capture ~98% of the co-variation of orthologous gene expressions and worm-specific TF gene expressions . We find that the meta-gene canonical temporal expression trajectories driven by conserved regulatory networks ( i . e . , internal principal dynamic patterns , iPDPs ) include four major patterns in both the worm and fly embryonic developmental process by order of eigenvalues: 1 ) a late highly varied pattern; 2 ) an early fast decaying pattern; 3 ) a slowly increasing pattern; and 4 ) an oscillating pattern ( Fig 4A ) ; i . e . , the pattern of canonical trajectories ( VL , D , I , O ) in Fig 3 . In contrast to the observed iPDP similarities , we find that worm and fly have very different external principal dynamic patterns ( ePDPs ) ( Fig 4B ) ; i . e . , the expression dynamic patterns driven by species-specific TFs . The principal dynamic patterns driven by the worm-specific regulatory network; i . e . , worm ePDPs , include a varied pattern that decreases until the middle stage and then increases , an increasing pattern , a varied pattern with a peak entering middle stage , a pattern that varies early and then increases during the embryonic development , and a cosine-like oscillating pattern with roughly two periods during the embryonic development . The fly ePDPs , however , have a varied pattern with low expression at the early stage , a sine-like oscillating pattern with roughly one period during the embryonic development , an increasing pattern , another sine-lie oscillating pattern with roughly two periods during the embryonic development , and a varied pattern that is like damped oscillation . In addition , we checked the sensitivity of iPDPs to small perturbations to internal/external regulatory networks by the leave-one-out method; i . e . , we removed one gene in the internal/external group , ran DREISS , and obtained the ordered iPDP eigenvalues for the remaining genes . We repeated the leave-one-out method for all genes , and finally found the ranges in which iPDP eigenvalues vary shown as error bars in S1 Fig . We can see that the iPDP eigenvalues almost stay at the same values ( small error bars ) for both worm and fly , which implies that the principal dynamic patterns of worm-fly orthologous genes driven by their conserved regulatory network are robust to small changes . The above results suggest that the conserved regulatory networks from orthologous meta-genes between worm and fly have similar effects to orthologous meta-genes , given their similar iPDPs ( i . e . , both have four patterns , as described above ) . The species-specific regulatory networks from species-specific meta-genes ( i . e . , worm-specific or fly specific TFs ) have effects that differ from the orthologous meta-genes for their different ePDPs . In addition , the expression dynamic patterns driven by the interactions between internal orthologous genes and external species-specific TFs are also different between worm and fly ( S2 Fig ) . In both worm and fly , we observe the similar four types of internally driven canonical temporal expression trajectories; i . e . , four matched internal principal dynamic patterns ( iPDPs ) ( Fig 4A ) . Thus , we are interested in seeing how individual orthologous genes relate to those dynamic patterns . We find that the worm-fly orthologous genes have correlated coefficients over each of the four iPDPs . Based on Eq 10 , we can obtain the coefficients of orthologous genes for each iPDP . We find that their coefficients are significantly correlated between worm and fly iPDPs with a similar pattern ( Fig 5 ) : r = 0 . 33 ( p<2 . 2e-16 ) for the highly varied pattern at late embryonic development stages ( first iPDP ) , r = 0 . 66 ( p<2 . 2e-16 ) for the fast decaying pattern at early embryonic development stages ( second iPDP ) , r = 0 . 67 ( p<2 . 2e-16 ) for the slowly increasing pattern during embryonic development ( third iPDP ) , and r = 0 . 73 ( p<2 . 2e-16 ) for the oscillation pattern during embryonic development ( forth iPDP ) , where r represents Spearman correlation of iPDP coefficients of 3 , 153 orthologous genes between worm and fly . This implies that , not only do the orthologous meta-genes have similar internal ( conserved ) regulatory effects ( i . e . , similar iPDPs ) , but the worm-fly orthologous genes also have similar internally-driven expression dynamics as resulted from their significantly correlated coefficients for iPDPs . The ePDPs between worm and fly generally do not show a high degree of matching similarity , but the worm ePDP No . 2 , and the fly ePDPs No . 3 are roughly representing the growing patterns . We find that orthologous gene correlation coefficients between these ePDP patterns are very small ( Spearman correlation r = -0 . 22 of the orthologous gene coefficients of worm ePDP No . 2 and fly ePDP No . 3 ) . The ribosome produces proteins , which is an ancient process and conserved across worm and fly , organisms separated by almost a billion years of evolution . The ribosomal genes are highly expressed during embryogenesis , since intensive cell division and migration require a large amount of proteins to be synthesized . We collected 195 ribosome-related genes based on the GO annotations . We ranked the coefficients of orthologous genes for each iPDP and ePDP in ascending order , and compared the rank values of iPDP and ePDP coefficients of ribosomal genes . We found that their average ranks of iPDP coefficients are significantly larger than ePDP ones in both worm ( t-test p<2 . 2e-16 ) and fly ( t-test p<2 . 6e-13 ) as shown in Fig 6 . This means that the ribosomal gene expression is significantly more influenced by the conserved regulatory network than by the species-specific regulatory network , which is consistent with ribosomal genes having conserved functions during embryonic development . The orthologous genes related to signal transduction for cell-cell communication ( a significantly more recent evolutionary adaptation relative to the ribosome ) exhibit the opposite trend . We found that 320 signaling genes from GO annotations have significantly larger average rank values of ePDP coefficients than iPDP ones in both worm ( t-test p<5 . 6e-11 ) and fly ( t-test p<8 . 3e-4 ) , as shown in Fig 6 . This result implies that the signaling gene expression is significantly more driven by the species-specific regulatory network than by the conserved regulatory network , which is consistent with the signaling genes being commonly associated with species-specific functions , such as body plan establishment and cell differentiation . We next turn to the biological meaning of individual canonical temporal expression trajectory for iPDPs and ePDPs . For the fast-decaying pattern ( 2nd iPDP ) , we find that the DNA replication is significantly enriched in Top 300 ( ~10% ) orthologous genes that have the most negative coefficients for this pattern , in both worm ( p<1 . 6e-8 ) and fly ( p<4 . 5e-6 ) . The GO enrichment analysis was performed using DAVID [23] . The very negative coefficients for the fast decaying pattern mean high positive coefficients for a fast-growing pattern ( vertically flipped 2nd iPDPs of worm and fly represent a fast-growing pattern ) , showing a drastic increase at the beginning of embryogenesis , then remain flat during the late embryogenesis ( red curves in Fig 7 ) . Most of the cell division of embryogenesis in both worm and fly happens approximately within the first 300 minutes . Then , the cell elongation and migration start to dominate the development [24 , 25] . The mRNA abundance of the genes involved in DNA replication may change accordingly . This is well reflected by the second iPDP . Interestingly , the original expression patterns of those top orthologous genes actually do not have fast-growing patterns ( black curves in Fig 7 ) , probably because of the combined effects of both conserved and species-specific GRN . Maternal mRNAs , which are pre-loaded before fertilization , may also mask the fast growing pattern of DNA replication genes . This pattern could only be observed after we separated the effect of two types of TFs using DREISS . In addition , we did not find any enrichment of DNA replication in top genes of other iPDPs ( p>0 . 05 ) . Therefore , the fast-growing iPDP patterns identified by our method reveal conserved regulation on the elementary cellular process of both species ( i . e . DNA replication ) . Besides a fast growing pattern driven by conserved worm-fly orthologous TFs , we also identified a fast growing pattern driven by non-conserved TFs for the two species . The Top 300 orthologous genes ( ~10% ) with the fast-growing worm ePDP ( ePDP No . 2 ) ( i . e . , driven by species-specific regulatory networks ) are enriched in ‘proteasome’ ( p<9 . 8e-16 ) . Protein degradation is not only a key process in apoptosis , but also throughout the entire course of development [26 , 27] . For example , eliminating proteins that are no longer needed is a vital process during embryo development; e . g . , the maternal proteins need to be cleaned as the embryogenesis proceeds ) . Previous reports also showed that different species usually have different maternal mRNA in the oocyte , which indicates that species-specific strategies might be utilized to regulate the protein degradation process [28] . In this study , after separating the effect of conserved and non-conserved regulatory networks , we observed that the protein degradation is significantly enriched in the genes majorly driven by species-specific TFs in worms . In contrast , the Top 300 orthologous genes with fast growing fly ePDP3 are enriched in ‘mitotic cell cycle’ ( p<3 . 5e-29 ) , ‘translation’ ( p<1e-30 ) and ‘mitochondrion’ ( p<7 . 7e-20 ) . Those enriched function related to energy generation is probably indicative of the large energy requirement during fly embryogenesis [29] , which did not provide the evolutionary conservation of this energy-related gene regulation . Our result reveals that the fly genes associated with respiration are more up-regulated by fly-specific TFs relative to conserved TFs , and that this up-regulation evolved after the separation of worm and fly . Besides the fast-growing pattern driven by species-specific TFs , we also observed some other interesting patterns . For example , worm ePDP3 displays a dramatic peak about 5 hours after fertilization . Among the Top 300 worm orthologous genes of this pattern , genes involved in synaptic transmission ( p<5 . 6e-9 ) and cell-cell signaling ( p<1e-7 ) are over-represented , suggesting that they are transiently activated in this stage of embryogenesis by worm-specific TFs . This observation indicates the gene regulatory network for these genes have evolved after the speciation . We applied DREISS to another example ( also see supplement ) about cancer . We are also interested to identify the gene expression dynamic patterns driven by conserved and human-specific regulatory networks during breast cancer cell cycle . Thus , we applied DREISS to a time-series gene expression data for human estrogen-responsive breast cancer cell line ( ZR-75 . 1 ) before and after hormonal stimulation , which has 12 time points covering a complete mitotic cell cycle ( 0–32 hours ) of hormonal stimulated cells [30] . The internal group , Ω is defined as a set of cross-species conserved human genes ( i . e . , 1132 worm-fly-human orthologs including 150 orthologous TFs ) , and the external group , Ψ consists of 1870 human-specific TFs . As shown in S3 Fig , the internally driven principal dynamic patterns ( iPDPs ) of conserved human genes include an oscillation trajectory whose period is roughly equal to a full cell cycle ( iPDP No . 4 ) , but the externally driven patterns ( ePDPs No . 2–4 ) oscillates more frequently than internal one , which suggests that though the evolutionarily conserved TFs regulate the normal cell cycle , the human specific TFs potentially drive the abnormal cycling behaviors of conserved gene expression responding to the hormonal stimulation . In this paper , we presented a novel computational method , DREISS , which decomposes time-series expression data of a group of genes into the components driven by the regulatory network inside the group ( internal regulatory subsystem ) , and the components driven by the external regulatory network consisting of regulators outside the group ( external regulatory subsystem ) . DREISS is a general-purpose tool that can be used to study the gene regulatory effects of any interested biological subsystems such as protein-coding transcription factors , micro-RNAs , epigenetic factors and so on . As an illustration , we applied DREISS to the time-series gene expression datasets for worm and fly embryonic developments from the modENCODE project [10] , and compared the worm-fly orthologous gene expression dynamic patterns driven by the conserved regulatory network ( i . e . , regulation effects from orthologous TFs ) , with the patterns driven by the species-specific regulatory networks ( i . e . , regulation effects from worm or fly specific TFs ) . We found that the conserved TFs drive similar genomic functions , but non-conserved TFs drive species-specific functions of orthologous genes between worm and fly , implying that , in addition to having ancient conserved functions , orthologous genes have been regulated by evolutionarily younger GRNs to execute species-specific functions during the evolution . This work can be easily extended to study the regulatory effects from orthologous TFs and species-specific TFs to species-specific genes . For example , one can find the expression dynamic patterns of worm/fly specific genes driven by specific TFs , and identify the genes with strong patterns associated with worm/fly specific functions , such as body formations . To the best of our knowledge , DREISS is the first method to reveal how the evolution of GRNs affects gene expression during embryogenesis . We emphasize that DREISS is a general-purpose method ( a free downloadable R tool available from github . com/gersteinlab/dreiss ) . Users can define the internal group ( Ω ) and external group ( Ψ ) according to their interests . For example , if users want to identify the protein-coding expression patterns driven by miRNAs , they can define miRNAs as an external group and protein-coding genes as an internal group . Additionally , DREISS can be applied to more than two datasets , such as comparing worm , fly and human embryonic stem cell developmental data , and finding their conserved and specific developmental expression patterns . The expression patterns driven by human-specific regulatory factors will potentially help us understand human-specific developmental processes along with the associated human genes . Due to the limited time samples in gene expression datasets , DREISS uses the simple linear state space model ( i . e . the first order linear invariant difference equation ) to model the temporal gene expression dynamics , and identify principal temporal dynamic patterns . This model assumes that the gene regulatory networks controlling temporal gene expression dynamics does not change across the entire biological process such as ( A , B ) in Eq 1 . Thus , based on the analytic analysis , the principal dynamic patterns ( PDPs ) must follow a small set of canonical temporal trajectories ( Fig 3 ) . With the rapidly increasing gene expression data , we can extend DREISS to more advanced models such as switched and hybrid system models , non-linear models [31] , which will allow us to study the gene regulatory networks are time varying , and potentially find the more temporal gene expression patterns capturing the more complex gene regulatory activities .	The dynamics of a biological system can be controlled by its own internal mechanisms and external perturbations . To gain intuition on this , we may draw a comparison with a mass hanging from a spring . The mass will move naturally by itself but its dynamics is also affected by one’s pulling it . That is , the dynamics of the mass is governed by the effect of the external perturbations superimposed on the internal mechanism of the spring ( i . e . Hooke’s law ) . Similarly , given a group of genes , their temporal gene expression dynamics can be controlled by both transcription factors inside the group and external regulatory factors . Therefore , it is useful to identify the expression dynamics that are exclusively controlled by internal or external factors and compare them across various systems . While state-space models have been widely used to decouple the internal and external effects in physical systems , such as the mass and spring , typical biological systems do not have enough time samples to infer all the model’s parameters , and applications of state-space models were not very effective in these instances . Hence , we developed a general-purpose computational method by integrating state-space models and dimensionality reduction to identify temporal gene expression patterns driven by internal and external regulatory networks . We applied our method to the embryonic developmental datasets in the worm and fly ( and also in a human cancer context ) . We successfully identified the temporal expression dynamics of cross-species conserved genes that were driven by conserved and species-specific regulatory networks .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "genetic", "oscillators", "developmental", "biology", "mathematics", "linear", "algebra", "algebra", "gene", "regulatory", "networks", "embryogenesis", "gene", "expression", "genetics", "gene", "regulation", "biology", "and", "life", "sciences", "physical", "sciences", "computational", "biology", "evolutionary", "biology", "eigenvalues", "evolutionary", "genetics", "evolutionary", "developmental", "biology" ]	2016	DREISS: Using State-Space Models to Infer the Dynamics of Gene Expression Driven by External and Internal Regulatory Networks
A recent study reported neutralizing antibodies to West Nile virus ( WNV ) in horses from four ranches of southern Pantanal . To extend that study , a serosurvey for WNV and 11 Brazilian flaviviruses was conducted with 760 equines , 238 sheep and 61 caimans from 17 local cattle ranches . Among the tested equines , 32 were collected from a ranch where a neurologic disorder outbreak had been recently reported . The sera were initially screened by using a blocking ELISA and then titrated by 90% plaque-reduction neutralization test ( PRNT90 ) for 12 flaviviruses . Employing the criterion of 4-fold greater titer , 78 ( 10 . 3% ) equines were seropositive for Ilheus virus , 59 ( 7 . 8% ) for Saint Louis encephalitis virus , 24 ( 3 . 2% ) for WNV , two ( 0 . 3% ) for Cacipacore virus and one ( 0 . 1% ) for Rocio virus . No serological evidence was found linking the neurological disease that affected local equines to WNV . All caimans and sheep were negative by blocking ELISA for flaviviruses . There were no seropositive equines for Bussuquara , Iguape , Yellow fever and all four Dengue virus serotypes . The detection of WNV-seropositive equines in ten ranches and ILHV and SLEV-seropositive equines in fourteen ranches of two different sub-regions of Pantanal is strong evidence of widespread circulation of these flaviviruses in the region . Flaviviruses represent a group of mosquito-borne viruses in Brazil that are annually involved in a large number of human cases of dengue fever countrywide and sporadic local outbreaks of sylvatic yellow fever [1] , [2] . Outbreaks caused by other flaviviruses have also been reported in the country . In the 1970s , the largest Brazilian epidemic of arbovirus encephalitis was caused by Rocio virus ( ROCV ) in southeast Brazil [3] . More recently , an outbreak of hemorrhagic manifestations was linked to Saint Louis encephalitis virus ( SLEV ) [4] . Sporadic human cases caused by other sylvatic flaviviruses , including Bussuquara virus ( BSQV ) , Ilheus virus ( ILHV ) and Cacipacore virus ( CPCV ) have also been reported in Brazil [5] , [6] , [7] . Furthermore , yellow fever epizootics in howler monkeys were reported in 2008 and 2009 . Approximately 200 carcasses tested positive for Yellow fever virus ( YFV ) and about 2000 deaths were reported [8] . Thirteen flaviviruses have been reported in Brazil , listed here in chronological order of discovery: YFV , ILHV , BSQV , SLEV , ROCV , CPCV , Dengue virus 1 ( DENV-1 ) and Dengue virus 4 ( DENV-4 ) , Dengue virus 2 ( DENV-2 ) , Iguape virus ( IGUV ) , Naranjal-like virus ( NJLV ) , Dengue virus 3 ( DENV-3 ) and Culex flavivirus ( CXFV ) [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] , [18] , [19] , [20] . In 2009 , serological evidence of West Nile virus ( WNV ) infection in Brazilian horses , was collected for the first time , in the Pantanal wetland region of Mato Grosso do Sul state ( MS ) [21] . The Pantanal wetland is a subtropical region of great biodiversity with strong potential for maintenance and evolution of mosquito-borne viruses . Comprising approximately 140 , 000 km2 , the Pantanal is a vast sedimentary floodplain characterized by seasonal flooding which determines specific ecosystem processes , with the occurrence of plants and animals that are adapted to the annual shrinking and expansion of habitats due to the seasonal hydrological regime [22] . The region , which covers mainly Brazilian but also Paraguayan and Bolivian territories , is ecologically classified into 11 sub-regions according to vegetation , flooding and physiography . In Brazil , the Pantanal is located within the states of Mato Grosso ( MT ) and MS in the west-central region of the country [23] . Most of the flaviviruses isolated in Brazil are unknown or understudied in the Pantanal . A small number of investigations in the Nhecolândia Sub-region of the Pantanal , MS , have detected serological evidence for four flaviviruses , including ILHV , SLEV , WNV and CPCV [21] , [24] , [25] . Recently , three other serological studies reported the detection of neutralizing antibodies for WNV in equines from MT and again in MS , including from outside the boundaries of the Pantanal [26] , [27] , [28] . Complete surveys for flaviviruses in the Pantanal have not been reported . Therefore , we conducted a serosurvey for WNV and the 11 Brazilian flaviviruses of potential medical importance , utilizing equines , sheep and caimans as indicators . The collections for this study were approved by the Animal Ethics Committee of Fundação Oswaldo Cruz ( License CEUA-Fiocruz LW-1/12 , protocol P-74/10-5 ) in compliance with the requirements of Brazilian Law 11794/2008 , which rules on the scientific use of animals , including the principles of the Brazilian society of Science in laboratory animals . The collections were also approved by the Brazilian Institute of Environment and Natural Resources ( licenses IBAMA 18363-1/2009 and 18363-2/2010 ) . Caimans were captured in two ranches from sites where a high concentration of these animals was observed . Caiman blood samples were obtained by puncture of the internal jugular vein between the 1st and 2nd cervical vertebrae as described previously [29] . Information recorded about caimans included gender , weight and snout-vent length . Blood samples from equines , including horses , donkeys and mules , and from sheep were taken by puncture of the jugular vein in 14 and nine different ranches , respectively . Gender , age , tameness , breed , vaccination status , travel history outside of Pantanal and history of abnormalities , such as clinical signs involving the central or peripheral nervous system , were recorded for each equine sample . Information about gender , age , breed and travel history outside of Pantanal were also recorded for each sheep . All animals sampled appeared healthy except for one horse that presented with clinical signs suggestive of neurological disorder . Additionally , 36 horse samples collected in February 2010 from a ranch comprising roughly a 900-square-km area in the Nabileque Sub-region of Pantanal , also located in MS ( Figure 1 ) were tested . An outbreak of an undiagnosed neurological syndrome had caused the death of 16 equines at this location from December 2009 to January 2010 . All equines here were vaccinated for rabies and tested negative for Trypanosoma evansi , a common pathogen involved in incoordination and instability of hind limbs in Pantanal equines . These equines had not been vaccinated for equine encephalitis viruses at the time of the outbreak . After the first deaths , the herd was vaccinated for eastern and western equine encephalitis viruses . All samples were initially screened for flavivirus-reactive antibodies by blocking ELISA as described previously [30] . Briefly , the ability of the sera to block the binding of the flavivirus group-reactive monoclonal antibody 6B6C-1 to the cell lysate-derived antigen for WNV was compared to the blocking ability of negative control horse serum . Samples were considered seropositive when the inhibition values produced were ≥30% . Samples with inhibition values between 19 and 29% were repeated twice to confirm the negative results . Blocking ELISA-positive samples were then heat-inactivated and tested by PRNT90 , as previously described [31] , for WNV and 11 Brazilian flaviviruses . Briefly , serial two-fold dilutions that ranged from 1∶10–1∶320 or 1∶20–1∶640 of each blocking ELISA seropositive sample were tested for their ability to neutralize plaque formation by WNV , SLEV , ILHV , CPCV , BSQV , IGUV , YFV ( vaccine strain 17D ) , ROCV , DENV-1 , DENV-2 , DENV-3 ( recombinant ChimeriVax Dengue 3 virus ) and DENV-4 . Caiman and sheep samples that were negative by blocking ELISA were also tested by PRNT90 for WNV . All reference viruses used for PRNT were previously submitted to partial nucleotide sequencing of the N terminal region of NS5 gene to confirm viral identity . High identity scores were obtained with the following sequences deposited at GenBank: DENV-1 ( FJ562106 ) , DENV-2 ( GQ398257 ) , DENV-3 ( recombinant ChimeriVax Dengue 3 virus ) ( JN811143 ) , DENV-4 ( GQ199880 ) , ROCV ( AY632542 ) , SLEV ( EF158048 ) , ILHV ( AY632539 ) , YFV ( JN628279 ) , BSQV ( AY632536 ) , CPCV ( AF013367 ) , IGUV ( AY632538 ) , and WNV ( JN819325 ) . Serum was considered seropositive to a virus when it reduced at least 90% of the formation of plaques of this virus and its reciprocal neutralizing antibody titer was at least four-fold greater than what was observed for the other 11 tested flaviviruses . The PRNT is the most specific serological test for the differentiation of flavivirus infections in convalescent serum samples [32] . Type-specific antibodies against flaviviruses can be distinguished using the PRNT [33] . The conservative criterion of four-fold greater titer is based on the results of cross-neutralization tests with flaviviruses and their respective antisera used for studies of antigenic relationships [34] , [35] . Two or more flaviviruses are distinct from each other by quantitative serological criteria ( four-fold or greater differences between homologous and heterologous titers of both serum samples ) [32] . The Cochran-Armitage Trend Test was used to test for positive trends in seroprevalence among age classes ( StatXact 10 . 0 , Cytel Software Corporation , Cambridge , MA ) All caiman and sheep serum samples were negative for flaviviruses-specific antibodies using the 6B6C-1 monoclonal antibody by blocking ELISA . When tested by PRNT90 for WNV , all caiman samples confirmed negative . Sixty caiman samples had PRNT90 titer <10 and one <40 for WNV ( low sample volume of the latter required testing at a lowest dilution of 1∶40 ) . From 238 sheep samples tested , 235 had PRNT90 titers <10 , two showed low neutralizing antibodies titers of 20 and 10 and one <40 . From 760 equine samples initially screened for flavivirus-reactive antibodies by blocking ELISA , 396 ( 52 . 1% ) were positive . When this subset was tested using PRNT90 , 332 ( 43 . 7% ) had neutralizing reactivity ( PRNT90 titer ≥10 ) for ILHV , 251 ( 33% ) for SLEV , 172 ( 22 . 6% ) for WNV , 139 ( 18 . 3% ) for CPCV , 130 ( 17 . 1% ) for ROCV , 62 ( 8 . 2% ) for IGUV , 14 ( 1 . 8% ) for YFV , 12 ( 1 . 6% ) for BSQV , four ( 0 . 5% ) for DENV-1 , three ( 0 . 4% ) for DENV-2 , one ( 0 . 1% ) for DENV-4 and none for DENV-3 ( Table 1 ) . Employing the criterion of 4-fold greater PRNT90 titer , 77 ( 10 . 1% ) equines were seropositive for ILHV , 59 ( 7 . 8% ) for SLEV , 24 ( 3 . 2% ) for WNV ( see Table 2 ) , two ( 0 . 3% ) for CPCV and one ( 0 . 1% ) for ROCV . From the remaining 232 blocking ELISA positive samples , 227 ( 29 . 9% ) were considered seropositive for undifferentiated flavivirus due to cross-reactivity and five ( 0 . 6% ) were considered seronegative with neutralizing antibodies titers <20 or <10 for all of the flaviviruses that were assayed by PRNT90 . These five may represent infections by an unknown flavivirus . There were no seropositive equines using the four-fold antibody titer difference criterion for BSQV , YFV , IGUV , DENV-1 , DENV-2 , DENV-3 and DENV-4 . Monotypic neutralization reactions ( i . e . a sample that reacted with only one of the 12 flaviviruses in the panel ) were detected for ILHV ( n = 27 ) , SLEV ( n = 11 ) and WNV ( n = 6 ) . The only horse from Nhecolândia Sub-region that presented neurological signs was seropositive for undifferentiated flavivirus with the following PRNT90 titers: 80 for SLEV , 40 for ILHV , 20 for YFV , 10 for ROCV and <10 for all other flaviviruses tested . Neutralizing antibodies for WNV were detected in 172 equines from all 15 ranches that equines were tested . Considering the four-fold higher titer criterion as confirmation , the 24 confirmed WNV-seropositive equines were detected in ten ranches comprising an area of approximately 2 , 500 square-km in the Nhecolândia and Nabileque sub-regions of the Pantanal , located in MS ( Figure 2 ) . Besides the 24 WNV-seropositive equines , samples from 37 other equines presented PRNT90 titers for WNV ≥40 and among them 26 showed equal or higher titers for one of the other flaviviruses tested . For these 37 equines , the serology profile could not be interpreted to determine whether WNV reactivity was due to infection with WNV or cross-reactivity from other flavivirus infections . In fact , six of them were considered seropositive for one of the 11 other flavivirus tested ( Table 3 ) . Low level of neutralizing antibodies was detected for YFV , BSQV , IGUV , DENV-1 , DENV-2 and DENV-4 even though none of the equines were considered infected with these viruses . In some cases , the PRNT90 titers for YFV , BSQV and IGUV exceeded 20 . Both ILHV- and SLEV-seropositive equines were detected in 14 of the 15 ranches that equines were tested , comprising roughly a 3 , 100-square-km area . The only ranch that had no ILHV- and SLEV-seropositive equines was ranch PS ( Figure 2 ) . From 760 equines tested , 441 ( 58% ) had date of birth available . Stratifying the serology results by age demonstrated that risk of exposure to flaviviruses increased with age for flaviviruses in general and for ILHV ( 1-tailed p<0 . 001 for both ) , but not for SLEV ( p = 0 . 17 ) or WNV ( p = 0 . 46 ) ( Figure 3 ) . SLEV and WNV appeared to increase in prevalence with age for equines <9 years of age , but not for older animals . Equines seropositive for CPCV and ROCV were sampled in October 2009 . The two CPCV-seropositive horses were five and seven years of age and were from ranch PQ , which comprises an area of approximately 140 square-km . The only horse that was seropositive for ROCV had no record of birth date and was from ranch PM that encompasses a 360-square-km area . At the end of 2009 , an outbreak of neurologic disease among horses was reported from ranch PV located in the Nabileque Sub-region . From 36 equines sampled , 23 ( 63 . 8% ) were seropositive for flavivirus infection by blocking ELISA . When this subset was tested by PRNT90 , 11 ( 30 . 5% ) confirmed seropositive for SLEV , nine ( 25 . 0% ) for undifferentiated flavivirus ( es ) , two ( 5 . 5% ) for WNV , and one ( 2 . 8% ) for ILHV ( Table 4 ) . Of the nine undifferentiated flavivirus infections , five had SLEV titers >20 and five had higher SLEV titers than other flaviviruses in the panel , indicating that a majority of these could have had a history of SLEV infection . In addition to the relatively low seroprevalence for WNV of just 5 . 5% in this sample of 36 equines , four equines from this group that recovered from the recent neurologic syndrome were serologically negative for WNV ( Table 4 ) . In spite of recent vaccination against equine encephalitis viruses , most equines had low titers for equine encephalitis viruses . These results will be described separately . In the present study , serological evidence of ILHV , SLEV , WNV , CPCV and ROCV circulation was detected in equines of the Pantanal . Severe disease has been linked to these viruses in humans [7] , [13] , [43] , [58] , [59] and for WNV also in equines [60] . WNV and SLEV circulation in the Pantanal appears to be a recent phenomenon . In the last three decades , the Pantanal has been impacted by the conversion of natural vegetation into agricultural fields and cattle pasture , with alteration and loss of natural habitats and biodiversity . Moreover , major negative impacts occur in uplands , with drastic deforestation of savanna vegetation , where main rivers feeding the Pantanal have their springs [22] , [61] . These environmental changes in the Pantanal environs can directly impact the fluctuations of vector and vertebrate host populations , which could affect arbovirus transmission dynamics in the region . Loss of wetlands and other natural habitats typically leads to reduced biodiversity , and consequently increased risk of outbreaks of vector-borne diseases [62] . Preserving large wetland areas sustains biodiversity and may represent a valuable ecosystem-based approach for controlling WNV outbreaks [63] . Indeed , an outbreak of neurologic disease among equines reported herein occurred in a ranch that encompasses drier plateau ecosystem on the periphery of the Pantanal wetlands . Serology data suggest that an unknown flavivirus closely related to SLEV or a newly recognized equine-virulent strain of SLEV may be associated with this neurologic syndrome . Considering the highly conservative criteria used in the present study , the detection of seropositive equines for ILHV , SLEV , WNV , CPCV and ROCV is strong evidence for local circulation of these flaviviruses in the Pantanal of MS , Brazil . These flaviviruses should be included as differential diagnosis in future flavivirus serosurveys to be held in the region . For PRNT studies , not only these flaviviruses , but also YFV , BSQV and IGUV should be included . It appears that WNV circulated recently in the Pantanal . The detection of WNV-seropositive equines in at least ten ranches is strong evidence of recent widespread circulation , but not necessarily establishment of WNV in the region . Circulation of WNV in this area has not yet been confirmed by virus isolation . Because detection of antibodies is indirect evidence of flavivirus circulation and because novel flaviviruses may exist in the region , we encourage efforts to isolate viruses to confirm the circulation of these flaviviruses . In future investigations of equine epizootics , efforts should be concentrated on the collection of samples suitable for viral isolation .	West Nile virus is maintained in cycles between birds and mosquitoes and recently reemerged as a worldwide major public health and veterinarian concern as the cause of human and equine encephalitis outbreaks . Recent studies have reported serological evidence of West Nile virus circulation in Pantanal , west-central region of Brazil . However , considering the co-circulation of various cross-reactive flaviviruses in Brazil and that most of the flaviviruses isolated in Brazil are unknown or understudied in the Pantanal , serological results should be interpreted with caution . Therefore , we conducted a serosurvey for West Nile virus and the 11 Brazilian flaviviruses of potential medical importance , utilizing equines , sheep and caimans as indicators , including 32 equines collected from a ranch where a neurologic disorder had been recently reported among the equines . We found serological evidence of Ilheus , Saint Louis encephalitis , West Nile , Cacipacore and Rocio viruses in Pantanal equines . West Nile virus infection was not associated with the neurological disease of equines . The detection of WNV-seropositive equines in ten ranches and ILHV- and SLEV-seropositive equines in fourteen ranches of two different sub-regions of Pantanal is strong evidence of widespread circulation of these flaviviruses in the region .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "emerging", "viral", "diseases", "virology", "biology", "microbiology" ]	2014	Serological Evidence of Widespread Circulation of West Nile Virus and Other Flaviviruses in Equines of the Pantanal, Brazil
Cells respond to their environment by modulating protein levels through mRNA transcription and post-transcriptional control . Modest observed correlations between global steady-state mRNA and protein measurements have been interpreted as evidence that mRNA levels determine roughly 40% of the variation in protein levels , indicating dominant post-transcriptional effects . However , the techniques underlying these conclusions , such as correlation and regression , yield biased results when data are noisy , missing systematically , and collinear---properties of mRNA and protein measurements---which motivated us to revisit this subject . Noise-robust analyses of 24 studies of budding yeast reveal that mRNA levels explain more than 85% of the variation in steady-state protein levels . Protein levels are not proportional to mRNA levels , but rise much more rapidly . Regulation of translation suffices to explain this nonlinear effect , revealing post-transcriptional amplification of , rather than competition with , transcriptional signals . These results substantially revise widely credited models of protein-level regulation , and introduce multiple noise-aware approaches essential for proper analysis of many biological phenomena . Cellular protein levels reflect the balance of mRNA levels , protein production by translation initiation and completion , and protein removal by degradation , secretion , and dilution due to growth [1–3] ( Fig 1A ) . A standard quantitative model for protein-level regulation [4 , 5] is ∂ P i ∂ t = τ i M i - δ i P i ( 1 ) where Pi is the cellular protein level ( molecules per cell ) of gene i , Mi is the mRNA level , and τi and δi are the mRNA translation and net protein removal rates , respectively . According to this model , at steady-state , protein levels will be proportional to mRNA levels with proportionality constants of τi/δi: P i = τ i δ i M i ( 2 ) such that if rates of translation and removal did not vary by gene , and in the absence of experimental noise or other variation , steady-state mRNA and protein levels would correlate perfectly [1] . Consequently , the mRNA–protein correlation observed in global measurements of mRNA and protein levels has been intensely studied , and deviations from perfect correlation used to quantify the contribution of post-transcriptional processes to cellular protein levels [1 , 3 , 6–9] . The consensus across these studies holds that , in a wide array of organisms , transcriptional regulation explains 30–50% of the variation in steady-state protein levels , leaving half or more to be explained by post-transcriptional regulatory processes [3 , 7 , 9–16] . Higher correlations are observed , generally for subsets of less than half the genome [1 , 9 , 17] . Low observed mRNA–protein correlations have motivated the search for alternate forms of regulation capable of accounting for the majority of protein-level variability [3 , 9 , 13] . In one proposal , mRNA levels serve mainly as an on-off switch for protein expression , imposing coarse control over protein levels which is then tuned by post-transcriptional mechanisms [9] . Recent studies have indeed uncovered wide between-gene variation in post-transcriptional features such as inferred translation rates [18] and protein degradation rates [3] . However , as frequently noted [1 , 7 , 9 , 10 , 19–21] , noise in measurements can cause many of the observations attributed to post-transcriptional regulation . Here , noise encompasses variability due to cell-to-cell variation , growth conditions , sample preparation and other effects due to experimental design [22] , and measurement biases and error [10 , 21 , 23] . Uncorrelated noise between mRNA and protein measurements will reduce the observed mRNA–protein correlation relative to the true value [24] , while inflating the variation in measurements of translational efficiency and other post-transcriptional processes . Most studies , particularly of protein levels , cover only a subset of known genes , due to factors such as signal-to-noise limitations , method biases , and continual revision of the coding-sequence annotations used to design and analyze assays . Limited and variable transcriptome and proteome coverage complicate analyses further , making it difficult to compare studies and to synthesize a holistic view of regulatory contributions . Missing data tends to reduce the precision of estimates , if data are missing at random ( MAR ) . However , most quantification methods are biased toward detection of more abundant mRNAs and proteins [9] . Data which are not missing at random ( NMAR ) in this way have reduced variance or restricted range . Range restriction , in turn , tends to systematically attenuate ( reduce in absolute magnitude toward zero ) the observed correlations and regression coefficients relative to complete data [25 , 26] . That is , biased detection produces biased estimates of the mRNA–protein correlation , leading to incorrect conclusions about regulatory contributions [27] . In many comparisons of the roles of transcriptional and post-transcriptional regulation , protein levels are correlated with or regressed on various predictors ( mRNA level and half-life , codon usage , amino-acid usage , etc . ) to determine relative contributions to protein-level variation [1 , 3 , 4 , 14 , 18 , 21] . If mRNA levels are found to explain a certain percentage , say X , in protein levels , then the other predictors are asserted to explain no more than 100−X percent of the variance [3 , 9 , 21 , 28 , 29] . A basic assumption of such analyses is that transcriptional and post-transcriptional regulation vary independently between genes . Several of the same studies report that high-expression genes show signs of more efficient translation [3 , 4 , 18] ( reviewed in [1] ) , raising concerns about the validity of this assumption . A related assumption of these analyses , one encoded in the standard functional model above , is that mRNA and protein levels are proportionally or linearly related [1 , 5]; the slope of this line is the mean number of proteins per mRNA . More often , the data are plotted on a log-log scale , where linearity appears as a slope of 1 . Consistent with this , ordinary least-squares linear regression shows that the slope is quite close to 1 for E . coli ( 0 . 96 ) and budding yeast ( 1 . 08 ) [17] , and estimates of proteins per mRNA have been reported roughly constant across mRNA expression levels in a prominent study [30] . However , like correlations , slopes estimated by standard linear regression are biased downward by noise in mRNA level measurements , an effect called regression dilution bias[31] which affects any regression where the independent variable is measured with error . A frequently encountered case is that , given two measurements X and Y , the slope from regressing Y on X is not the inverse of the slope of regressing X on Y[32–34]; this is regression dilution bias at work . Consequently , linear regression cannot be used to estimate the functional relationship between mRNA and protein levels , raising the question of what the true functional relationship is . Use of nonparametric methods avoids assumptions of linearity [1] , at the cost of destroying genuine information about the dynamic range of gene expression and its determinants . Analytical solutions to many of these problems exist—notably , Spearman introduced a correction for noise-induced attenuation of correlation estimates more than a century ago [24]—yet have largely failed to find their way into the hands of groups carrying out gene-regulation experiments and analyses ( with a few exceptions [15] ) . Some problems remain almost entirely unaddressed , such as providing accurate estimates of the functional relationship between variables measured many times with correlated noise yielding variably and systematically missing values . Here , we develop and integrate approaches to address all of these challenges , with the aim of providing more comprehensive and rigorous estimates of the relationship between mRNA and protein levels than have previously been possible . To do so , we take advantage of the rapid , continual progress made in global measurement of mRNA and protein levels by multiple methods [6 , 17 , 30 , 35–41] . All of these methods were first employed at the genome scale in studies profiling gene expression during log-phase growth of budding yeast in rich medium , a de facto standard . These studies often compare results against previous studies , evaluating agreement , precision , coverage , and dynamic range while pointing out relative advantages of each approach ( e . g . [17 , 18 , 30 , 37 , 38 , 40] ) . Our efforts to synthesize these data into a coherent whole are grounded in the stance that all these works constitute measurements of the same underlying quantities—average mRNA and protein levels in a large cell population prepared under narrowly defined conditions—whether or not such measurements were the study goal . Systematic differences between approaches due to experimental choices will introduce variation which may not be distinguishable from simple inaccuracy in measurement . We treat this variation as experimental noise without prejudice . Distinctions between biological variability , measurement error , method bias , and other sources of noise are of course important , particularly in deciding how to control or manage noise . These distinctions may also depend on one’s perspective . For example , unintentional differences in growth conditions may lead two groups following the same protocol to make measurements on samples which inevitably are , in truth , biologically different , such that error-free measurement would reveal differences in mRNA and protein levels . In one sense , these differences reflect biological variability; in an equally valid sense , they represent experimental noise . Similarly , intentional protocol differences that are not meant to alter measurement accuracy ( such as use of new methods intended to make measurements more precise ) , yet carry known and unknown biases , may also introduce noise . Here , we take an empirical approach to noise which does not involve divining intent . Versions of this approach are taken , often implicitly , by the many previous analyses that integrate experiments from multiple groups [8 , 9 , 17 , 18 , 30] . Our results reveal that , once noise is accounted for , mRNA and protein levels correlate much more strongly under these experimental conditions than previously appreciated , with a correlation coefficient of r = 0 . 93 . We find that protein levels are not proportional to mRNA levels , but instead are more steeply related , an effect we show is consistent with measurements of translational activity . Transcriptional and post-transcriptional regulation act in a concerted , non-independent manner to set protein levels , inconsistent with common attempts to divvy up and assign protein-level variance to each mechanism . As a byproduct , we generate what by several measures is the most complete and accurate quantitative transcriptome and proteome available , in average molecules per haploid cell , for this widely studied organism under these well-studied conditions . Finally , we highlight and introduce methods for analyzing correlations and functional relationships between measured data which may be used broadly . We collected 38 measurements of mRNA levels and 20 measurements of protein levels from 13 and 11 separate studies respectively , each of haploid S . cerevisiae growing exponentially in shaken liquid rich medium with 2% glucose between 22°C and 30°C ( Table 1 ) . As described in the Introduction , we assume , for modeling purposes , that each replicate in each experiment constitutes a measurement of the true per-gene mean mRNA and protein levels under these narrowly defined conditions . These data cover varying amounts of the genome and display a wide range of correlations between studies ( Fig 1B , Pearson correlations on log-transformed values with zeros and missing values omitted ) . Although correlations of replicates within studies are quite high [9] , with median r = 0 . 97 for mRNA and 0 . 93 for protein levels , between-study correlations are far more modest , r = 0 . 62 for mRNA measurements and 0 . 57 for protein measurements ( Fig 1C ) . That is , data from a typical mRNA study explains 39% of the variance in another study ( r2 = 0 . 39 ) and a typical protein study’s results explain only 32% in another study’s variance , consistent with previous studies reporting wide variation between studies [16] . Strong outliers indicate high reproducibility for a two pairs of studies ( Fig 1C ) , but each such outlier is a correlation between separate studies done by the same research group , suggesting the presence of additional variability sources between groups . Coverage of the 5 , 887 verified protein-coding genes in yeast [42] also varies widely across pairs of studies ( Fig 1C ) . Coupled with high within-study reproducibility , the low between-study reproducibility indicates the presence of large systematic errors between studies . In a single study [38] , mRNA levels in a commercially prepared sample were measured using two methods , a commercial microarray and single-molecule RNA sequencing . These measurements correlate with r = 0 . 86 ( 73% of the variance explained in one measurement by the other ) , quite similar to the r = 0 . 84 correlation of the single-molecule measurement with an independent RNA-Seq dataset on RNA from a different study [43] . These data hint , coupled with similar observations in other biological systems [44] , that high within-study reproducibility is likely to reflect reproducible biases associated with use of a single measurement technique in addition to reproducible features of the biological sample . Correlations are modest even between studies using similar methods ( e . g . , r = 0 . 81 between two RNA-Seq datasets using Illumina instruments [18 , 43] ) . Comparing mRNA studies performed using similar or different methods on a shared set of 4 , 595 genes revealed a consistent bias toward higher median correlations between studies using similar methods , but these differences were not statistically distinguishable ( Fig 1D , no t-test P < 0 . 05 for differences in correlation when comparing studies employing shared methods versus independent methods after false discovery rate correction ) . Between-study correlations quantify the studies’ mean ratio of true variance to total variance , termed the reliability [15 , 45 , 46] ( see Methods ) . In turn , setting aside sampling error , the maximum observable correlation between any two datasets is equal to the geometric mean of their reliabilities . Because virtually all reported global mRNA–protein correlations involve mRNA and protein levels measured in separate studies , between-study reliabilities are the relevant quantity . The modest reliability values—setting aside those of the same group reporting two studies , which we exclude from this analysis—sharply limit the maximum observable mRNA–protein correlations . This limit has startling consequences: if steady-state mRNA and protein levels actually correlated perfectly ( true r = 1 . 0 ) , then given the median observed between-study correlations in Fig 1C , we would expect to observe mRNA–protein correlations of only r = 0 . 57 × 0 . 62 = 0 . 60 . The data reveal a wide range of modest mRNA–protein correlations with a median of r = 0 . 54 ( Fig 2A ) quantified either by the Pearson correlation between log-transformed measurements or the nonparametric Spearman rank correlation ( S1 Fig; both measures produce similar results and we employ the former throughout ) . The largest pair of datasets covers 4 , 367 genes and shows an mRNA–protein correlation of r = 0 . 618 ( r2 = 0 . 38 , 38% of protein-level variance explained by mRNA levels ) , close to consensus values [9] . The largest dataset containing replicated measurements of mRNA and protein in at least two studies yields similar correlation values; notably , averaging paired measurements together and correlating the averages increases the apparent correlation ( Fig 2B ) . This averaging effect has a simple explanation: if experimental noise drives down the mRNA–protein correlation , and noise is to some extent random between studies , then averaging together measurements from different studies will increase the correlation as random noise dilutes out and signal titrates in . However , exploiting averaging comes with hidden dangers when using these data . Averaging requires multiple measurements . Few protein datasets cover even half the genome , and incomplete data tend to be biased toward abundant proteins , as revealed by examining levels in a large dataset when restricted to proteins detected in smaller datasets ( Fig 2C ) ; it is plausible that higher-expression proteins correlate more strongly with mRNA levels . We therefore checked for an averaging effect using a subset of the data with a minimum level of reproducibility , at least eight mRNA and eight protein measurements , which includes 549 genes . This high-coverage gene subset does encode more highly abundant proteins relative to the rest of the genome as assessed by western blotting ( Fig 2D ) . As a benefit , however , changes in correlation due to averaging within this subset do not merely reflect underlying systematic changes in the expression levels of the analyzed genes . In this subset , the observed mRNA– protein correlation rises markedly as more measurements are averaged together ( Fig 2E ) , more than doubling in the apparent protein-level variance explained by mRNA level ( from 33% to 72% ) simply by averaging together more measurements of the same genes . These data strongly indicate that experimental noise substantially reduces the apparent correlation between mRNA and protein levels . The foregoing analyses involve estimates uncorrected for noise , which as described in the Introduction do not properly estimate the true correlation between the variables being measured . We will first incorporate noise-aware estimates of the true correlation , and then address the more challenging problem of accounting for missing data to arrive at a true genome-scale estimate of the mRNA–protein correlation . Reduction of correlations by noise can be corrected using information from repeated measurements , assuming the noise is uncorrelated across measurements [24 , 46] . Quantitative corrections for correlation attenuation were first introduced more than a century ago by Spearman [24] , are widely used in the social sciences [46–48] , and have found recent applications in biology [15 , 45 , 49–52] . Given two measurements each of variables X and Y , each with uncorrelated errors , the true correlation can be estimated using only correlations between the four measurements X1 , X2 , Y1 , Y2 ( see Methods ) : r ^ X Y true = r X 1 Y 1 r X 2 Y 2 r X 1 Y 2 r X 2 Y 1 4 r X 1 X 2 r Y 1 Y 2 ( 3 ) The correction reflects a simple intuition: the denominator quantifies the reliabilities of the measurements , which determine the maximum observable correlation , and the numerator quantifies the observed correlation using a geometric mean of four estimates and is divided by this maximum value to yield an estimate for the true value . The estimate is not itself a correlation coefficient , and may take values outside ( −1 , 1 ) due to sampling error [46] . Also note that there is no P-value associated with this estimate; statistical testing for significant association using uncorrected correlation measures remains valid . To demonstrate and test Spearman’s correction , we applied it to simulated data generated to mimic key features of mRNA and protein data , but with a known underlying correlation and known measurement reliability . We generated data for 5 , 000 simulated genes with a range of correlations and fixed reliability; a fixed correlation and a range of reliabilities; and a fixed correlation and reliability with a range of data missing at random , or non-randomly , with a detection bias against low-expression genes . We then measured the observed correlation , uncorrected for noise , and used Spearman’s correction to estimate the true correlation . At each set of parameters , we generated 50 transcriptome/proteome pairs to assess reproducibility . As shown in Fig 3A–3C , noise reduces correlations in a non-negligible way . Given an actual correlation of 0 . 9 , and a reliability of 0 . 7 , higher than the mean values for real data ( cf . Fig 1C ) , the observed correlation has a mean of 0 . 631±0 . 009 ( standard deviation ) , whereas Spearman’s correction yields a median value of 0 . 901±0 . 007 , closely matching the true value . Spearman’s correction performs well over a wide range of reliabilities ( Fig 3B ) and when data are missing at random ( Fig 3C ) , cases where observed correlations provide a wide range of estimates that are all systematically incorrect . Smaller datasets lead to increased variability of the Spearman estimate due to sampling error ( Fig 3C ) . When faced with data biased toward detection of high-abundance mRNAs and proteins , Spearman’s correction systematically underestimates the true correlation ( Fig 3D ) , as expected due to restriction of range effects . Using Spearman’s correction on real data , we estimated mRNA–protein correlations for pairs of mRNA- and protein-level studies , obtaining a median corrected correlation of 0 . 92 . Variability due to sampling error was large for small datasets as expected ( cf . Fig 3C and 3D ) , and decreased as dataset size increased , with estimates stabilizing for large datasets ( > 3000 genes ) at a mean of r = 0 . 88±0 . 02 ( Fig 2A ) . This value is echoed by consideration of the largest dataset with two mRNA [38 , 43] and two protein [30 , 40] measurements each ( Fig 2B ) . For these data , the four observed mRNA–protein correlations are r = 0 . 60 , 0 . 63 , 0 . 62 and 0 . 64 , and the correlation between mRNA and protein measurements are rmRNA = 0 . 86 and rprotein = 0 . 57 respectively , yielding the corrected estimate r ^ true = 0 . 60 × 0 . 63 × 0 . 62 × 0 . 64 4 0 . 85 × 0 . 57 = 0 . 89 . As demonstrated , Spearman’s correction , while useful , does not address biases due to data that are systematically missing . Spearman’s correction also assumes uncorrelated errors , and thus has no mechanism for handling correlated errors arising due to , for example , protocol similarities within a study or use of similar measurement techniques between studies . Actual datasets show evidence for all of these effects ( Fig 1 ) . Extending estimates to the full genome , accounting for structured noise and non-randomly missing data , requires a more sophisticated approach . Even seemingly simple approaches to reduce noise , such as averaging measurements normalized to the same scale , are unworkable as strategies for estimating genome-scale mRNA–protein relationships: only 16 proteins are detected by all 11 protein quantification studies , and these proteins are all highly abundant . Throwing out smaller datasets discards potentially valuable measurements , and it is unclear when to stop , since all datasets are incomplete to some degree . To address these challenges , we adapted structural equation modeling to admit nonrandomly missing data ( see Methods ) . We introduce a structured covariance model ( SCM ) , adapted with important modifications from recent work [27] , that explicitly accounts for structured noise arising from replicates and use of shared measurement techniques , explicitly estimates noise at multiple levels and the nonlinear scaling factors linking underlying variables , and allows inferences of latent covariance relationships with imputation of missing data ( Fig 4 ) . The SCM accurately estimates true correlations in simulated data when substantial data are missing nonrandomly , a case on which Spearman’s correction produces severely biased estimates ( Fig 3D ) . Fitting the SCM to real data yields estimates of whole-genome steady-state mRNA–protein correlation of r = 0 . 926±0 . 004 across all 5 , 854 genes for which an mRNA has been detected in at least one of the 38 mRNA quantitation experiments ( Fig 2A ) . That is , mRNA levels explain 86% of variation in protein levels at the whole-genome scale . We emphasize that the SCM does not involve any attempt to maximize the mRNA–protein correlation or any assumptions about the strength of the correlation . To examine the influence of low-coverage datasets on the correlation estimate , we re-fit the SCM on data restricted to studies with no more than 60% or 80% missing values ( cf . Table 1 ) , resulting in essentially unchanged correlation estimates of r = 0 . 919 and r = 0 . 933 , respectively . Including these smaller datasets does not alter these estimates significantly . The SCM integrates all data to produces mean and variability estimates of mRNA and protein levels , yielding a dataset in which mRNA levels have been quantified for 5 , 854 genes and protein levels have been quantified for 4 , 990 genes in at least one study . To evaluate the accuracy of these estimates , we linearly scaled them to molecules per haploid cell using high-quality published values for mRNA per cell and protein per cell . Estimates of the number of mRNA molecules per cell range from 15 , 000 to 60 , 000 molecules per cell [36 , 53] . A more recent study argued that the earlier , lower estimate resulted from misestimation of mRNA mass per cell and average mRNA length , with 36 , 000 molecules per cell as a revised estimate also supported by independent measurements [54] . The higher estimate resulted from rescaling the lower estimate to match expression of five genes measured by single-molecule fluorescence in situ hybridization ( FISH ) [53] . We adopted the 36 , 169 mRNA molecules per cell estimate [54] . Scaled to 4μg of protein in 1 . 5×106 cells ( 2 . 7pg protein per yeast cell in cells roughly 30 μm3 in size ) [55] , SCM protein levels sum to just over 35 million protein molecules per haploid cell , similar to the 50 million molecules per cell estimated previously [20] within the variation in total protein extraction from haploid yeast cells ( cf . [56] , which estimates 4 . 95pg per cell ) . Scaled SCM per-gene means provide the best point-estimates of molecules per cell ( Fig 5A ) , although the correlation between estimates of means is necessarily higher than the estimated true correlation , since each estimate contains error . For a more representative global view of mRNA and protein levels , we draw a sample from the SCM estimates according to each gene’s mean and variance in levels ( Fig 5B ) . Correlations between sampled mRNA and sampled protein levels ( r = 0 . 923 ) are consistent with the inferred underlying correlation . We then compared scaled SCM estimates to small-scale gold-standard , independent measurements of absolute mRNA and protein levels not used in our analysis . ( No genome-scale gold-standard measurements of mRNA or protein levels exist for yeast or any other organism . ) SCM estimates of absolute mRNA levels matched FISH measurements well [53] ( average difference of 1 . 2-fold between estimated and measured levels [Fig 5B] , with one outlier estimate overshooting the FISH value by 1 . 7-fold ) . Notably , these results demonstrate that the FISH estimates are compatible with roughly 36 , 000 mRNA molecules per cell during exponential growth as reported [54] , and do not require the almost two-fold higher number of cellular mRNAs extrapolated in the FISH study . Absolute protein levels for a set of 21 proteins differing up to 25 , 000-fold in cellular abundance have been measured using single-reaction monitoring ( SRM ) spiked with stable-isotope standards [57] . SCM estimates correlate better with these absolute levels ( r = 0 . 94 between log-transformed values ) than does any individual dataset . This includes the only study , using western blotting [30] , which reports levels for all 21 proteins ( r = 0 . 90 ) ( Fig 5C , average difference of 1 . 4-fold between SCM estimates and SRM measurements , compared to 1 . 8-fold using western blotting ) . Relative protein levels estimated by integrating multiple datasets using an alternative approach in which noise is not modeled [16] correlate with absolute levels less well ( r = 0 . 88 ) than do the SCM estimates . The structured covariance modeling approach thus estimates steady-state cellular mRNA and protein levels with an unmatched combination of completeness and accuracy . To evaluate imputation of missing data , we focused on the 864 genes with a detected mRNA but no protein detected in any of the 11 studies . Some of these genes encode well-studied proteins such as the proteasomal regulator Rpn4p and the cyclin Cln3p , indicating clear false negatives . For a systematic evaluation , we turned to ribosome profiling studies [18] , which quantify ribosome-protected mRNA fragments normalized for gene length ( ribosome density ) , providing an estimate of the mRNAs being actively translated in vivo . At least one of five studies under compatible experimental conditions detects ribosomes in the coding sequence of 637 of these 864 genes , suggesting active translation . Normalized ribosome density for this restricted set of genes correlates with the imputed protein levels ( Fig 5E , r = 0 . 55 ) , despite the attenuating effect of range restriction . Because the missing protein data correspond to genes at the detection limit of these ribosome-profiling studies , we predict that many of the remaining genes will be found to produce proteins at low levels during exponential growth . The SCM estimates serve as predictions for the levels of these as-yet undetected proteins . Our results indicate that the true correlation between steady-state mRNA and protein levels in exponentially growing budding yeast is far higher than previously recognized , explaining the vast majority of variation in protein levels on a log scale . In many previous analyses , this would be equivalent to demonstrating a minor role for other forms of regulation: if the variation in protein levels were a pie , and mRNA levels took a slice , other forms of variation would get only the leftovers . As we will show , such competition is largely illusory . Positive evidence exists for strong post-transcriptional contributions to protein levels . The dynamic range of protein abundance is wider than mRNA abundance , which must reflect dynamic-range amplification by post- transcriptional regulation [9] . Indeed , wide per-gene variation exists in measurements of translational efficiency [18 , 58 , 59] . The report that translational activity , estimated by ribosome profiling , explained more than twice the protein-level variation than did measured mRNA levels [18] prompted us to more closely examine these results . We reproduced these comparisons , and found that subsequent ribosome-profiling studies [58–61] confirmed the strong predictive power of ribosome density for the protein levels originally employed , which came from a single study [40] ( Fig 6A ) . We wondered whether these findings might reflect experimental noise that differed between the mRNA and ribosome-footprint measurements in the original study . Correlations using SCM-estimated protein levels are substantially higher for both SCM-estimated mRNA levels and ribosome density measured in all studies , consistent with reduction of noise in the SCM estimates ( Fig 6A ) . SCM-estimated mRNA levels predict protein levels better ( r = 0 . 926 ) than any of the individual ribosome profiling studies ( Fig 6A ) . This likely reflects remaining noise and systematic bias in the profiling studies , since using Spearman’s correction to estimate the true correlation between ribosome density and protein level yields correlations of r = 0 . 88 against SCM-estimated protein levels and r = 0 . 91 using the largest two largest protein-level datasets , measured by mass-spectrometry and western blotting . These results suggest that , contrary to previous reports , measures of translation and mRNA level have essentially equivalent and quite strong predictive power for protein levels . However , major contributions to protein levels from mechanisms other than mRNA level become obvious upon inspection of the data . The dynamic range of protein expression ( from fewer than 50 to more than 1 , 000 , 000 molecules per cell [30 , 57] ) is wider than that of mRNA levels ( e . g . from 0 . 1 to 89 molecules per cell in a landmark early study [36] ) . In the SCM estimates , the full range of mRNA expression is roughly 10 , 000-fold ( 0 . 02 to 253 molecules per cell on average ) , whereas the range of protein expression is more than 1 , 000 , 000-fold ( an average of 0 . 4 molecules to 1 . 3 million molecules per cell ) . Since both mRNA and protein are roughly lognormally distributed , the ratio of log-transformed ranges , 1 . 6 , yields a rough measure of relative variation . ( This relative variation is unchanged when attention is restricted to the central 95% of the mRNA and protein distributions to mute outlier effects . ) Individual mRNA and protein datasets vary but confirm similar differences in dynamic range ( S3 Fig ) . We address more representative estimates of relative dynamic range below . As previously noted [9] , this dynamic-range amplification must involve post-transcriptional variation . The standard use of a logarithmic scale raises some questions about the interpretation of dynamic range . What does a ten-fold difference mean , if it is between 0 . 01 to 0 . 1 molecules per cell rather than between 1 and 10 molecules per cell ? Are fractional numbers meaningful ? We proceed as though they are . Fractional molecules per cell in a population average may indicate mRNAs or proteins present in only a fraction of cells in the population , which can arise in many ways , from conditional expression ( e . g . during a segment of the cell cycle ) to incomplete repression ( leakiness ) . Here , estimates of levels reflect the measurements but confer no particular interpretation . We note that no obvious break or cutoff exists in the data or the SCM estimates to suggest a gene-expression threshold below which the biology changes qualitatively . A consequence of two facts—the higher dynamic range of protein levels than of mRNA levels , and the strong log-log linear correlation between the two—is that steady-state protein levels cannot be ( even noisily ) proportional to steady-state mRNA levels at the genome scale . In the standard model ( P i = τ i δ i M i with protein P and mRNA M for gene i , cf . Eq 2 ) , steady-state protein levels will be roughly proportional to steady-state mRNA levels on a log-log scale assuming translation rates and degradation rates are uncorrelated with mRNA levels . This is most easily seen considering the case of constant translation and degradation rates ( τi = τ and δi = δ , respectively ) across all genes , such that P i = τ δ M i 1 where we have made explicit the exponent of 1 . In this case , ln P i = 1 × ln M i + ln ( τ δ ) . Deviations from proportionality can be captured by deviations from a log-log slope of 1 . As described in the Introduction , several studies have estimated slopes very near 1 , but have not accounted for error-induced systematic underestimation of slopes due to regression-dilution bias [31] . We therefore used a noise-tolerant regression technique closely related to principal component analysis known as ranged major-axis ( RMA ) regression [33] , which yielded a range of slopes systematically higher than the ordinary least-squares regression slopes ( Fig 6B and 6C ) and have a median of 1 . 54 . Unlike OLS , RMA regression permits error in both variables and is symmetric , such that regression of Y on X produces the inverse slope to that obtained by regression of X on Y . Other techniques with the same symmetry property but different technical assumptions each yield slopes substantially larger than 1 and larger than OLS estimates ( S2 Fig ) . The estimated slopes for individual pairs of datasets span a wide range , even using RMA and limiting attention to large datasets ( Fig 6B ) , suggesting the existence of systematic biases , toward increased and decreased variance , separating these studies . The presence of such biases in protein-quantitation studies , though not their precise source , has been previously described [62] . The SCM approach , which accounts for both noise and missing data , yields an estimated slope of 1 . 69 , compatible with the range of estimates from noise-aware methods on individual pairs of datasets ( Fig 6B and 6C ) and also similar to the expectation ( 1 . 6 ) derived from examination of the relative dynamic ranges above . Steady-state protein levels therefore reflect a dramatic multiplication of the transcriptional signal: rather than competing with transcriptional regulation as often assumed , post-transcriptional regulation cooperates . If translational activity drives much of this cooperative amplification , higher-expressed mRNAs must tend to be more highly translated . Such an effect was noted in passing in the earliest ribosome-profiling study [18] . Several additional such studies satisfying our experimental criteria have been performed since [58–61] , which allows us to more thoroughly quantify the relationship between levels of translation and expression . The coverage of these datasets is excellent , so we focus on the 4 , 435 genes for which all five studies report ribosome density measurements . Using these data , we found a markedly supralinear relationship between relative translational activity ( estimated by median ribosome density ) , and SCM-estimated mRNA levels ( Fig 6D ) with a log-log slope of 1 . 68 . As this result implies , translational efficiency ( TE ) ( median ribosome density divided by median normalized mRNA levels within these same studies [18 , 58] ) increases with SCM-estimated mRNA level ( Fig 6E , Spearman rank correlation r = 0 . 65 ) , with some evidence for a ceiling or saturation effect at high expression levels . These results provide strong evidence that highly expressed genes generate highly translated mRNAs . RMA regression of ribosome density against SCM mRNA levels yielded a slope of 1 . 70 , compared to a slope of 1 . 72 of SCM protein levels against mRNA levels ( Fig 6F ) , suggesting that increases in translational activity accompanying elevated mRNA expression are sufficient to generate the broader dynamic range of protein levels relative to mRNA levels . A subtle possibility is that the SCM estimates have a compressed dynamic range relative to true values , which would inflate both the slope of the translational-activity–mRNA relationship and the correlation between TE and mRNA levels . To address this possibility , we exploited the fact that three of the accompanying mRNA-level measurements in the ribosome-profiling studies [58 , 60 , 61] were not used in our SCM estimates and therefore constitute an independent , modern , replicated mRNA dataset . The median of these recent measurements correlate well with our SCM estimates ( r = 0 . 90 , Pearson correlation on log-transformed values ) and the SCM and recent measurements have statistically indistinguishable distributions ( S4 Fig ) . High- and low-expression genes deviate slightly consistent with experimental error in RNA-seq at the low end and compression of the SCM estimates at the very high end . We regressed ribosome densities and protein levels against these recent mRNA levels . Slopes were lower but substantially above 1 . 0 ( 1 . 46±0 . 02 and 1 . 49±0 . 03 [95% confidence intervals] for translational activity and SCM protein levels versus recent mRNA measurements , respectively , Fig 6F ) . Importantly , calculation of the slope of translational activity versus recent mRNA level does not involve our SCM measurements at all , and thus provides independent evidence that translational activity levels have a wider dynamic range than mRNA levels . To provide an overall view of relative dynamic ranges , we plotted the distribution of estimated numbers of steady-state mRNAs and proteins per gene . We used ribosome density measurements to estimate the number of ribosomes engaged in translating each mRNA species in a typical haploid cell ( Fig 6G ) , assuming 200 , 000 ribosomes per cell [63] of which 85% are engaged in active translation ( see Methods ) . Both SCM and recent mRNA levels show a similar dynamic range for most mRNA species , and a narrower distribution than ribosome or protein levels . In summary , measured variation in translational activity correlates strongly with mRNA level and is sufficient to quantitatively account for the strong nonlinear relationship between mRNA levels and protein levels . The analysis above illustrates a fundamental asymmetry: although absence of post-transcriptional regulatory processes would produce a perfect mRNA–protein correlation [1] , a perfect mRNA–protein correlation need not indicate a negligible post-transcriptional contribution to relative protein levels . Contrary to the conclusions of many analyses , it is possible for mRNA levels and ( for example ) translation rates to each explain more than 50% of protein-level variation . Both processes could each contribute 100% of protein-level variation . All that is required is that their contributions not be independent . To see this , consider the following toy model for regulation of protein levels which does not involve assuming that translation rates are independent of mRNA levels: ∂ P i ∂ t = τ i M i − δ i P i standard model , cf . Eq . 1 with δi=δτi=α ( Miϵi ) γln ϵi~N ( 0 , σ ) constant degradation ratetranslation rate rises nonlinearly with mRNA levelnoisy evolved correlation between mRNA levels and translation rates Despite appearances , the functional relationship between translation rates and mRNA levels does not imply or depend on mechanistic properties of transcription and translation . All variance in this model ( as in all analyses in the present work ) derives from differences between genes , so the functional relationship merely describes an empirical correlation . As described in more depth in the Discussion , such a correlation can arise if genes have evolved differential translational efficiencies tuned to multiply transcriptional signals . In this toy model , with εi = 1 ( or more generally σ = 0 ) , translation rates and mRNA levels reinforce each other perfectly albeit nonlinearly . Under these conditions , steady-state mRNA levels explain 100% of the steady-state protein-level variation on a log scale . Translational regulation also explains 100% of the protein-level variation . P i = α M i γ δ M i = α δ M i 1 + γ steady-state protein levels ln P i = ln α δ + ( 1 + γ ) ln M i log protein levels are linearly related to log mRNA levels = - ln ( δα1/γ ) + ( 1 + 1 γ ) ln τ i log protein levels are linearly related to log translation rates Adding variation to translation rates ( σ > 0 ) and fixing other parameters allows close reproduction of the SCM estimates on several dimensions ( Fig 7A and 7B; source code including parameters presented in Methods ) . Both datasets have similar mRNA–protein correlations ( r = 0 . 926 for experimental data , r = 0 . 922 for toy model ) , similar log-log slopes ( 1 . 69 for both ) , and similar dynamic ranges for mRNA and protein levels . The critical difference between this model and the standard model for protein-level variation , Eq 1 , is the evolved strong positive correlation between mRNA levels and translational efficiency . This , too , is evident in experimental data when calculating translational efficiency ( Fig 7C , RMA slope = 0 . 71 , Spearman r = 0 . 62 ) . The correlation is mirrored by the toy model , where translation rate per mRNA and mRNA level can be directly compared ( Fig 7D; RMA slope = 0 . 74 , Spearman r = 0 . 74 ) . The experimental data are substantially missing at the low end , which will tend to attenuate the correlations . The toy model does not capture the apparent saturation of translational efficiency at high mRNA levels ( Fig 7C ) . Assuming this effect is real , other mechanisms , such as decreased rates of protein turnover , must be added to the toy model to even better reflect the data , which we leave for future detailed modeling . Our results demonstrate that the frequently reported result that steady-state mRNA levels explain less than half ( 30–50% ) of the variation in protein levels constitutes a significant underestimate . In exponentially growing budding yeast , the best-studied system and source of many of these claims , we find that the true value at the whole-genome scale , taking into account the reductions in correlation due to experimental noise and missing data , is closer to 85% . Many thoughtful studies have tackled this problem before , arriving at results that match ours on certain dimensions , but via quite different approaches . Previous work has employed versions of Spearman’s correction [15] , contended with differences in dynamic range by adopting nonparametric approaches [1 , 17] , and integrated multiple datasets [8 , 11 , 16 , 17] . All of these works have reached conclusions which differ from the portrait assembled here . Our analysis transcends these studies on several fronts . The present study incorporates more measurements than any previous work . We distinguish between correlations between measurements and estimates of underlying correlations accounting for between-study reliability , a critical difference that has largely eluded previous work . The structured covariance model natively handles nonrandomly missing data to provide more complete and accurate molecules-per-cell estimates than previous studies . Most importantly , we have not relied on the common but mistaken assumption that different modes of regulation act independently . A consistent approach in the literature has been to pit transcriptional and post-transcriptional variation against each other , both analytically and rhetorically ( e . g . , “transcriptional regulation is only half the story” [28] ) . As we have shown , the data do not fit this competitive paradigm , and even invalidate some of its analytical assumptions , such as independence and non-collinearity . The competitive versus cooperative aspects of post-transcriptional regulation come to the fore when considering the dynamic ranges of gene expression . A wider range of protein than mRNA levels is well-established in a range of organisms [3 , 15 , 64] , and our results further cement this observation . However , dynamic-range variation could be achieved in different ways , captured by two extremes . At one extreme , post-transcriptional regulatory variation is uncorrelated with transcriptional regulation , reducing the contribution of mRNA levels to protein levels . At the other extreme , post-transcriptional variation correlates strongly with transcriptional regulation , multiplying the transcriptional signal with little interference . In both cases , post-transcriptional regulation amplifies the dynamic range of gene expression , but only in the latter case does it also faithfully amplify the mRNA signal itself . Our data clearly and convergently indicate that the biology , at least for this organism under these conditions , lies toward the latter , cooperative extreme . Coordinated transcriptional and translational signal amplification may explain a range of other observations , particularly regarding proteins-per-mRNA ( PPM ) ratios , which are frequently used to isolate signs of post-transcriptional regulation . Because post-transcriptional amplification correlates strongly with mRNA levels , PPM will remain correlated with mRNA , and as a consequence , any sequence features correlated with mRNA will tend to correlate with PPM as well . As an example , amino-acid composition correlates with PPM in yeast [17] , with valine/alanine/glycine frequencies higher in high-PPM sequences and leucine/asparagine/serine frequencies lower in high-PPM sequences . These are precisely the same amino acids previously shown to vary most strongly in frequency , in the same directions , with increasing mRNA abundance [65] . Similarly , many other correlates of PPM are also correlates of mRNA levels ( codon bias , tRNA adaptation ) , including mRNA level itself [1 , 11] . For features such as codon bias , which arises in response to selection for translational efficiency [66] , association with increased PPM might seem an obvious causal link , but because codon bias strongly associates with mRNA level , the null expectation is that it will correlate with PPM even if codon bias had no effect on translational activity at all . Analyses of the determinants of protein levels must contend with the collinearity and non-independence of contributing processes . The strong correlation between steady-state mRNA and protein levels may seem to validate the use of mRNA levels as relatively faithful proxies of protein levels . We urge caution , as a tempting conclusion—that mRNA changes serve as faithful proxies for protein changes—does not follow . Attempts to infer the correlation between mRNA and protein changes from steady-state mRNA–protein correlations confuse two distinct and complex phenomena . The genome-scale relationship between mRNA levels and protein levels is an evolved property of the organism , reflecting tuning by natural selection of each gene’s transcriptional and post-transcriptional controls , rather than a mechanistic input-output relationship between mRNA and protein mediated by the translational apparatus . Two genes with steady-state mRNA levels differing by 10-fold may have 500-fold differences in protein levels due to evolved differences in their post-transcriptional regulation . These evolved steady-state differences do not predict how the protein levels for these genes will change if both mRNAs are induced 10-fold , because evolution does not occur on this timescale; the changes in protein levels are instead dictated by the cellular mechanisms of translation . An important intermediate case between the evolutionary and mechanistic cases is variation in mRNA and protein levels in individuals across a genetically diverse population . The potential for correlations between mRNA and protein relies upon substantial true variance in mRNA levels . In population-variation studies , one expects relatively few variants and resulting variation far lower than the orders of magnitude considered here . Correspondingly , in such studies mRNA-change–protein-change correlations may be low even given a strong underlying link between mRNA and protein levels . If the nonlinear multiplication of mRNA levels into protein levels is an evolved property , what mechanism ( s ) has evolution exploited ? The present work supports a particular class: the increased density of ribosomes on high-expression mRNAs , with variation sufficient to account for the nonlinearity , suggests increased rates of translation initiation as the major contributor . Correspondingly , recent work has shown that in yeast and a wide range of other organisms , the stability of mRNA structures in the 5’ region weakens as expression level increases , favoring more efficient translation initiation [67] , and wide variation in heterologous protein levels can be achieved by varying mRNA stability near the initiation site [68 , 69] . Several limitations still attend our approach . By assuming single multiplicative errors per experiment , we ignore variation in per-gene error which may be systematically different between low- and high-expression genes and/or systematically affect particular measurement techniques [62] . For example , limitations in the dynamic range of a measurement technique will tend to compress the resulting measurements , causing such systematic errors . Our model does not contend with distortions possibly imposed by alterations to 3’ regulatory signals ( e . g . tagging with affinity epitopes [30] or fluorescent proteins [39] to enable protein detection ) , or with variability in quantification due to propensities of particular mRNAs to be more efficiently sequenced or for their protein products to be unusually amenable to mass-spectrometric detection . The lack of any gold-standard genome-scale measurements hinders detection of such biases . Our results underscore the urgent need for such standard measurements of absolute mRNA and protein levels to enable identification and correction of systematic errors in established and emerging gene-expression measurement techniques . More sophisticated models for experimental error at many levels , which would be informed by but need not wait for such gold-standard measurements , also promise to provide higher-fidelity biological estimates from existing data . We infer a higher mRNA–protein correlation ( r = 0 . 93 ) here than when using an earlier , related model [27] ( r = 0 . 82 ) , a difference we attribute to two factors . First , the present analysis stratifies by measurement technology using all data , whereas the previous estimate did not , although in that study , stratifying by technology on a reduced dataset yielded r = 0 . 86 [27] . Here , using all data and treating technology-related experimental noise separately from other sources of noise , we are able to average out more systematic technology biases , likely producing superior estimates of the associated measurement variability and reducing noise-induced attenuation of the mRNA–protein correlation . Second , in the present analysis , population-averaged protein levels and mRNA levels are constrained to each have a single underlying variance , whereas in the earlier study each experimental replicate had a separate variance . Inference of artificial experiment-specific variances spread variability across experiments ( overfitting ) , where in the present analysis , we adopt the more biologically plausible stance that the true underlying mRNA and protein population-average distributions each have a characteristic variance which is measured by each experimental replicate . The present model , deprived of extra parameters , infers higher correlations . Our study considers a single well-studied growth condition for a single well- studied organism , raising questions about how to generalize this work . The principles of accounting for noise , but not precise results , can and should be extrapolated to regulatory contributions in other settings and other organisms . An influential study on mouse fibroblasts measured mRNA and protein levels and degradation rates for thousands of genes [3] , concluding that mRNA levels explained 41% of the variation in protein levels , with most variation instead explained by translational regulation . Our results indicate many ways in which the results of this study could be profitably revisited . Indeed , a recent follow-up study concluded that , once effects of error and missing data were accounted for , mRNA levels explain 75% or more of the protein-level variation in these data [21] . The protein regulatory environment of rapidly dividing cells differs from that of many other cellular states . The faster cells divide , the more rapidly protein molecules partition into daughter cells , adding an approximately constant amount to all protein removal rates and consequently reducing between-gene variation in these rates . This will tend to increase the dependence of protein levels on mRNA levels , and decrease the dependence on degradation rates , during proliferation . In addition to cellular state , regulatory contributions depend on timescale . Post-transcriptional processes must dominate protein-level changes within seconds to a few minutes of a stimulus or signal; transcriptional responses , particularly in eukaryotes , where transcription and translation are uncoupled , are all but powerless at this timescale . As such , the notion of general determinants of protein levels without regard to timescale has questionable utility . A final theme emerging from our study is that careful empirical studies , coupled with noise-aware analyses , are needed to determine regulatory contributions for any cellular condition of interest at any timescale . Let us assume we wish to measure latent variables ϕ and ψ but , due to noise , actually observe variables X = ϕ+εX and Y = ψ+εY where the random noise variables εX and εY are uncorrelated and mean zero . The reliability α X = Var ( Φ ) Var ( X ) = Var ( Φ ) Var ( Φ ) + Var ( ϵ X ) ( 4 ) quantifies the ratio of signal variance to total ( signal plus noise ) variance in X . Given two random variables X1 and X2 representing replicate measurements of ϕ , the latent ( true ) variance is Cov ( X1 , X2 ) = Cov ( ϕ+εX1 , ϕ+εX2 ) = Cov ( ϕ , ϕ ) = Var ( ϕ ) , where the error terms vanish because they are uncorrelated by assumption . Thus , the Pearson correlation between replicates is ρ X 1 , X 2 = Cov ( X 1 , X 2 ) Var ( X 1 ) Var ( X 2 ) = Cov ( Φ , Φ ) Var ( X 1 ) Var ( X 2 ) = Var ( Φ ) Var ( X 1 ) Var ( Φ ) Var ( X 2 ) = α X 1 α X 2 , ( 5 ) which is the geometric mean of the reliabilities of the two measurements . We wish to infer the Pearson correlation coefficient between latent variables ρ ϕ , ψ = Cov ( ϕ , ψ ) Var ( ϕ ) Var ( ψ ) but , due to noise , we observe random variables ρ X , Y = Cov ( X , Y ) Var ( X ) Var ( Y ) = Cov ( Φ , ψ ) ( Var ( Φ ) + Var ( ϵ X ) ) ( Var ( ψ ) + Var ( ϵ Y ) ) ≤ ρ Φ ψ . ( 6 ) with equality only when Var ( εX ) = Var ( εY ) = 0 ( i . e . there is no noise ) . Uncorrelated noise has no average effect on the numerator because errors cancel ( see above ) , but the error terms in the denominator do not cancel . This effect additively inflates the variances in the denominator , biasing the observed correlations downward relative to the truth . Given the reliabilities αX and αY , Spearman’s correction is given by ρ Φ ψ = ρ X Y α X α Y ( 7 ) To estimate ρϕψ , we need estimates of ρXY , αX and αY . A natural estimator replaces these population quantities with the sample correlation coefficients , rxy , α ^ x and α ^ y with α ^ x = r x 1 , x 2 α ^ y = r y 1 , y 2 where x1 , x2 are realizations of X and y1 , y2 are realizations of Y . These replicates are used to estimate reliabilities . The true correlation , ρϕ , ψ , can then be estimated using only correlations between measurements: r ^ Φ ψ = r x 1 y 1 r x 2 y 2 r x 1 x 2 r y 1 y 2 = r x 1 y 1 r x 2 y 2 α ^ x α ^ y We extend this estimate to r ^ Φ ψ = r x 1 y 1 r x 2 y 2 r x 1 y 2 r x 2 y 1 4 α ^ x α ^ y which has the further desirable properties of exploiting all pairwise correlations and being independent of the choice of indices . Taking this approach to its logical conclusion , given a set of N measurements of ϕ and M measurements of ψ , we propose the estimator r ^ Φ ψ = ( ∏ i , j N , M r x i , y j ) 1 N × M ( ∏ i < i ′ N r x i , x i ′ ) 1 N ( N - 1 ) ( ∏ j < j ′ M r y j , y j ′ ) 1 M ( M - 1 ) , where the numerator is the geometric means of all pairwise correlations , and the demoninator is the square root of the product of the geometric means of the pairwise reliability estimates ( correlations between measurements ) for each variable . We gathered 38 measurements from 13 studies measuring mRNA expression , and 20 measurements from 11 studies measuring protein concentrations , yielding a total of 58 high-throughput measurements of mRNA and protein levels from a maximum of 5 , 854 genes in budding yeast . The measurements were taken using different technologies including custom and commercial microarrays , competitive PCR , high-throughput RNA sequencing , flow cytometry , western blotting , scintillation counting of 35S-labeled protein on 2D gels , and liquid chromatography coupled to tandem mass spectrometry ( LC-MS/MS ) using a range of labeling and quantification techniques . All yeast cultures were haploid S . cerevisiae growing in shaken liquid rich medium with glucose between 22°C and 30°C and sampled during the exponential growth phase . Details of the datasets are summarized in Table 1 . For analytical purposes , we treat data from one study [38] which performed two independent measurements using different methods as two studies ( RNA-Seq and microarray ) , one per method . This study’s RNA-Seq employed a single-molecule sequencing method , smsDGE; we treat this as an RNA-Seq dataset . We downloaded ribosome-profiling data from the primary sources for five studies [18 , 58–61] . Within-study replicates were averaged; for one study of translational inhibitors [59] , the no-inhibitor and 1×-inhibitor replicates were averaged . Summary ribosome density and mRNA levels for these datasets were computed by log-transforming all instances of each type of measurement , subtracting the grand median value , and exponenentiating the per-gene median of the resulting values . To ensure the measurements were independent of SCM estimates , we excluded the mRNA levels from Ingolia and colleagues from the mRNA estimates , leaving three studies ( one ribosome-profiling study did not report mRNA levels [59] ) . To preserve the measured dynamic range in the data , no scaling of variance was performed . Translational efficiency was computed as the median normalized ribosome density ( five studies ) divided by the median normalized mRNA level ( three studies ) , ensuring these results are independent of the SCM estimates . To estimate the number of ribosomes translating each mRNA species , we multiplied median ribosome densities ( which are proportional to ribosomes per nucleotide ) by gene length , then normalized the resulting distribution to sum to 200 , 000 ribosomes per haploid cell [63] . We further assumed that approximately 85% of ribosomes are engaged in active translation during rapid growth [70] . Raw data ( with missing values ) , data normalized and imputed using the SCM , and merged molecules-per-cell estimates are archived in Dryad ( http://datadryad . org ) with DOI doi:10 . 5061/dryad . d644f . All analyses were carried out using R [71] using custom scripts which may be downloaded from GitHub ( http://github . com/dad/mrna-prot ) . Regression analyses using major-axis ( MA ) , scaled major-axis ( SMA ) , and ranged major-axis ( RMA ) regression were performed using the package lmodel2 . RMA was performed using interval ranges . The model has two components: an observation model p ( Ii , j∣Xi , j ) , which provides the probability of observing a value for mRNA/protein i in replicate j , given the underlying mRNA/protein level , and a hierarchical model p ( Xi , j∣… ) for the underlying mRNA/protein levels themselves . The full model is specified as Xi , j=Li , l[ j ]Gl[ j ]+Ti , t[ j ]+Ei , k[ j ]+Ri , j+νj ( 8 ) L i ∼ 𝓝 2 ( 0 , Ψ ) ( 9 ) T i , t ∼ 𝓝 N T ( 0 , τ t ) ( 10 ) E i , k ∼ 𝓝 ( 0 , ξ k ) ( 11 ) R i , j ∼ 𝓝 ( 0 , θ j ) ( 12 ) p ( I i , j = 0 \| X i , j = x ) = 1 1 + exp ( - η k [ j ] 0 - η k [ j ] 1 X i , j ) . ( 13 ) Random variables Li , l correspond to the true denoised protein ( l = 1 ) and mRNA ( l = 2 ) levels , for mRNAs and proteins i = 1 , … , N , and Li = [Li , 1 , Li , 2]′ . The random variables Ti , t and Ei , k capture common technological variation and batch effects , respectively , t = 1 , … , Nt , k = 1 , … , NE . Ri , j are experimental noise for replicate j = 1 , … , NR . Both technology effects and batch effects between experiments are assumed to be independent , Cov ( Ti1 , t1 , Ti2 , t2 ) = 0 if t1 ≠ t2 , and Cov ( Ei1 , k1 , Ei2 , k2 ) = 0 if k1 ≠ k2 . Measurement noise is independent between replicates , Cov ( Ri1 , j1 , Ri2 , j2 ) = 0 if j1 ≠ j2 . The parameters νj corresponds to the normalizing constants of the mRNAs/proteins within a replicate ( on the log-scale , normalizing constants become offsets ) . The coefficient Gl represents the log-variance of the denoised true mRNA or protein . The ratio A = G a b u n d G a b u n d represents the amount of post-transcriptional amplification of mRNA to protein . At steady state we expect P i ≈ M i A for protein Pi and mRNA Mi . This model falls into the class of models that were extensively studied in an earlier work [27] . The results are largely insensitive to deviations from parametric modeling assumptions and to several details of prior specifications . Below is R code to reproduce the toy model in Fig 7 . # Random number seed set . seed ( 115 ) # Number of genes n <- 5854 # Exponent of empirical ( evolved ) relationship between steady-state # mRNA levels and translation rates gamma <- 0 . 56 # Scaling factor , 1/time alpha <- 0 . 1 # Degradation rate , 1/time delta <- 0 . 001 # Standard deviation of mean-zero variation added to log mRNA levels to yield # unscaled log translation rates te . variation <- 1 . 1 # Steady-state mRNA levels in molecules/cell ( log-normal ) # Mean and variance are equal to those of the SCM mean estimates log . m <- rnorm ( n , mean = 1 . 09 , sd = 1 . 25 ) m <- exp ( log . m ) # Translation rate -- add log-normal variation to , and scale , mRNA levels tau <- alphaexp ( log . m + rnorm ( n , mean = 0 , sd = te . variation ) ) ^gamma # Steady-state protein levels in molecules/cell ( log-normal ) prot . variation <- 0 . 55 p <- ( tau/delta ) exp ( log . m + rnorm ( n , mean = 0 , sd = prot . variation ) ) # Plot protein vs . mRNA plot ( m , p , log = ‘xy’ , pch = 16 , las = 1 , xlab = ‘mRNA level ( mol . /cell ) ’ , ylab = ‘Protein level ( mol . /cell ) ’ ) # Plot translation rate vs . mRNA plot ( m , tau , log = ‘xy’ , pch = 16 , las = 1 , xlab = ‘mRNA level ( mol . /cell ) ’ , ylab = ‘Translation rate per mRNA ( proteins/sec ) ’ )	Cells respond to their environment by making proteins using transcription and translation of mRNA . Modest observed correlations between global steady-state mRNA and protein measurements have been interpreted as evidence that mRNA levels determine roughly 40% of the variation in protein levels , indicating dominant post-transcriptional effects . However , the techniques underlying these conclusions , such as correlation and regression , yield biased results when data are noisy and contain missing values . Here we show that when methods that account for noise are used to analyze much of the same data , mRNA levels explain more than 85% of the variation in steady-state protein levels . Protein levels are not proportional to mRNA levels as commonly assumed , but rise much more rapidly . Regulation of translation achieves amplification of , rather than competition with , transcriptional signals . Our results suggest that for this set of conditions , mRNA sets protein-level regulation , and introduce multiple noise-aware approaches essential for proper analysis of many biological phenomena .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[]	2015	Accounting for Experimental Noise Reveals That mRNA Levels, Amplified by Post-Transcriptional Processes, Largely Determine Steady-State Protein Levels in Yeast
Intrinsic antiviral resistance represents the first line of intracellular defence against virus infection . During herpes simplex virus type-1 ( HSV-1 ) infection this response can lead to the repression of viral gene expression but is counteracted by the viral ubiquitin ligase ICP0 . Here we address the mechanisms by which ICP0 overcomes this antiviral response . We report that ICP0 induces the widespread proteasome-dependent degradation of SUMO-conjugated proteins during infection and has properties related to those of cellular SUMO-targeted ubiquitin ligases ( STUbLs ) . Mutation of putative SUMO interaction motifs within ICP0 not only affects its ability to degrade SUMO conjugates , but also its capacity to stimulate HSV-1 lytic infection and reactivation from quiescence . We demonstrate that in the absence of this viral countermeasure the SUMO conjugation pathway plays an important role in mediating intrinsic antiviral resistance and the repression of HSV-1 infection . Using PML as a model substrate , we found that whilst ICP0 preferentially targets SUMO-modified isoforms of PML for degradation , it also induces the degradation of PML isoform I in a SUMO modification-independent manner . PML was degraded by ICP0 more rapidly than the bulk of SUMO-modified proteins in general , implying that the identity of a SUMO-modified protein , as well as the presence of SUMO modification , is involved in ICP0 targeting . We conclude that ICP0 has dual targeting mechanisms involving both SUMO- and substrate-dependent targeting specificities in order to counteract intrinsic antiviral resistance to HSV-1 infection . The ubiquitin pathway regulates many essential cellular processes including protein degradation , the cell cycle , transcription and DNA repair . It is not surprising that many viruses have therefore evolved strategies to take advantage of this pathway in order to enhance their replication ( for a recent review see [1] ) . During herpes simplex virus type-1 ( HSV-1 ) infection , one of the first viral proteins to be expressed is ICP0 ( infected cell protein 0 ) , an E3 ubiquitin ligase of the RING finger class that is required for the efficient initiation of lytic infection and productive reactivation of viral genomes from latency ( reviewed in [2] ) . Whilst the exact mechanisms by which ICP0 stimulates infection remain to be elucidated , it is clear that the ubiquitin ligase activity of ICP0 plays a fundamental role in regulating infection . Deletion or point mutations within the RING finger of ICP0 that inactivate its ubiquitin conjugation activity completely impair its ability to stimulate lytic infection and the reactivation of quiescent viral genomes [3]–[8] . During infection , ICP0 localizes to promyelocytic leukemia ( PML ) nuclear bodies ( PML-NBs , also known as ND10 or PODs ) where it induces the proteasome-dependent degradation of PML , its small ubiquitin-like modifier ( SUMO ) -modified isoforms , and SUMO-modified Sp100 [9]–[11] . Recent findings suggest that these ND10 proteins play a role in contributing to intrinsic antiviral defence , as depletion of these proteins increases the likelihood of an ICP0-null mutant virus entering productive infection [12] , [13] . The available evidence is consistent with the hypothesis that ICP0 targets specific cellular proteins for proteasome-dependent degradation in order to inhibit ( or relieve in the case of latency ) cellular mechanisms that would otherwise repress viral transcription [14] . However , the mechanism ( s ) by which ICP0 targets these cellular proteins for degradation remains unclear . One of the earliest detectable events during HSV-1 infection is a cellular response that leads to the accumulation of ND10 components at sites closely associated with viral genomes soon after they have entered the nucleus . This response is rapidly counteracted by the ubiquitin ligase activity of ICP0 , a phenotype that correlates well with its ability to stimulate lytic infection and reactivation from quiescence [8] , [15] , [16] . We have recently shown that this recruitment is dependent upon SUMO Interaction Motifs ( SIMs ) within these proteins and that mutation of these motifs inhibits their abilities to repress viral replication [17] . These data implicate a role for the SUMO conjugation pathway in mediating intrinsic resistance to HSV-1 infection . However , several important questions remain outstanding . For example , how does ICP0 target PML and its SUMO-modified isoforms for degradation , what other cellular factors are also targets for ICP0-mediated degradation , and how does ICP0 inhibit the recruitment of proteins other than PML to these repressive foci associated with incoming HSV-1 genomes ? Recent findings have highlighted a link between ubiquitin ligase targeting and the proteasome-dependent turnover of SUMO-modified proteins through the discovery of SUMO Targeted Ubiquitin Ligases ( STUbLs ) . These proteins represent a class of RING finger ubiquitin ligases that contain SIMs that were initially identified in yeast through the characterization of Slx5 and Slx8 in S . cerevisiae , Rfp1 and Rfp2 in Schizosaccharomyces pombe , and more recently RNF4 in mammalian cells [18]–[23] . These SIM-containing ubiquitin ligases provide a means of regulating SUMO-modified substrates via their ubiquitination and proteasome-dependent degradation . SIMs typically consist of a short core of hydrophobic amino acids , ( V/I/L ) -x- ( V/I/L ) - ( V/I/L ) or ( V/I/L ) - ( V/I/L ) -x- ( V/I/L ) , which form a β-strand that binds in a groove formed between the α-helix and a β-strand of SUMO [24]–[26] . SIMs are often followed by acidic or phospho-serine residues that enhance the SIM-SUMO interaction [27] , [28] . The best characterized mammalian STUbL is RNF4 , a SIM-containing RING finger ubiquitin ligase that promotes the degradation of SUMO-modified PML following arsenic trioxide treatment . Multiple SIMs within the N-terminus of RNF4 mediate its association with poly-SUMO chains anchored on PML , leading to the ubiquitination of both SUMO and PML [22] , [23] , [29] , [30] . Given the parallels between the STUbL activity of RNF4 and ICP0’s ability to degrade PML , and as ICP0 had previously been shown to induce the loss of SUMO conjugates following ectopic expression of tagged SUMO-1 [9] , we decided to investigate if ICP0 represented a viral STUbL . Here we report that ICP0 contains multiple SIM-like sequences ( SLSs ) , one of which shares homology to a SIM previously characterized within hDaxx and HCMV IE2 , and that mutation of specific SLSs inhibits its ability to interact with and promote the ubiquitination of SUMO-2 chains . We demonstrate that during infection ICP0 induces the global degradation of high molecular weight ( MW ) SUMO conjugates in a RING finger- and proteasome-dependent manner , although certain SUMO-conjugates are degraded more rapidly than others . Utilizing a panel of ICP0-expressing cell lines , we show that combined mutation of several SLSs has a clear detrimental effect upon ICP0 function and its ability to degrade SUMO-conjugates . Using PML as a model substrate , we demonstrate that ICP0 preferentially induces the degradation of all SUMO-modified PML isoforms , but can additionally target PML isoform I for degradation in a SUMO modification-independent manner . We also demonstrate that the SUMO conjugation pathway plays an important role in mediating the recruitment of ND10 components to sites associated with incoming HSV-1 genomes . Depletion of Ubc9 , the sole E2 SUMO conjugating enzyme , partially relieves this cell-mediated repression mechanism and increases the replication efficiency of an ICP0-null mutant virus . Taken together our data demonstrate that the SUMO conjugation pathway contributes to intrinsic antiviral resistance and that these activities are counteracted by both SUMO- and substrate-dependent targeting specificities of ICP0 . Previous studies have shown that several ND10 proteins , namely PML , Sp100 , hDaxx and ATRX , contribute to intrinsic resistance to HSV-1 infection , and that this cellular response is counteracted by the viral ubiquitin ligase ICP0 [12] , [13] , [31] . An important aspect of intrinsic antiviral resistance is the ability of these constitutively expressed proteins to relocate to sites associated with incoming viral genomes in order to mediate the transcriptional repression of viral gene expression [15] , [16] , [31] . We have recently shown that mutation of SIMs within PML and hDaxx inhibited the ability of these proteins to localize to HSV-1 genomes and thereby restrict the replication of an ICP0-null mutant virus [17] . Furthermore , in cells depleted of PML additional SUMO-2/3 conjugates were observed to localize to sites associated with incoming viral genomes [17] . These data imply a potential mechanistic role for SUMO conjugation in mediating intrinsic antiviral resistance . To test this hypothesis , human diploid fibroblasts ( HFs ) were transduced with lentiviruses expressing control ( shNeg ) or anti-Ubc9 shRNAs and analyzed in assays monitoring intrinsic resistance to HSV-1 infection , either in the presence or absence of ICP0 ( Figures 1 and S1 ) . Ubc9 is the sole E2 SUMO conjugating enzyme and is essential for SUMO conjugation . Cells stably expressing an shRNA to Ubc9 exhibited significant depletion in Ubc9 expression , as well as decreased abundance of SUMO-1 and SUMO-2/3 conjugates in comparison to control cells ( Figure 1A and data not shown ) . Ubc9-depleted cells also showed significant reductions in SUMO modification of both PML and Sp100 ( Figure 1A ) , dispersal of intra-nuclear SUMO-1 and SUMO-2/3 conjugates , and dramatic changes in the number and distribution of ND10 ( Figure 1B , bottom panels ) . These data demonstrate that depletion of Ubc9 efficiently restricts the SUMO conjugation pathway , affecting the overall abundance of SUMO-conjugates and ND10 integrity . To investigate the role of SUMO conjugation in intrinsic antiviral resistance we examined the relative plaque formation efficiencies ( PFE ) of both wild type ( wt ) and ICP0-null mutant ( ΔICP0 ) HSV-1 in control and Ubc9-depleted cells . Wt HSV-1 PFE was unaffected by depletion of Ubc9 ( Figure 1C ) . In contrast , ICP0-null mutant ( ΔICP0 ) HSV-1 infection exhibited a 10-fold increase in PFE in Ubc9-depleted cells compared to control cells ( Figures 1C and 1D ) . Although significant depletion of Ubc9 was achieved ( Figure 1A ) , the SUMO pathway is required for cell division [32]-[34] . In accordance , we noted that Ubc9 depletion could not be maintained long-term and that depleted cells had to be analyzed soon after isolation . Therefore , it is likely that some Ubc9 remains in order to allow limited cell division to occur while under selection . The observed increase in ICP0-null mutant PFE may therefore represent an underestimate in the repressive role that SUMO conjugation plays during HSV-1 infection in the absence of ICP0 expression . We note that depletion of Ubc9 itself does not impact upon the infection process per se , as wt virus was unaffected in plaque forming efficiency in comparison to control cells ( Figure 1C ) . We next investigated the recruitment of SUMO and PML to sites associated with incoming viral genomes in the absence of ICP0 expression ( Figure S1 ) . Viral genomes can be visualized by the appearance of punctate foci containing the major HSV-1 transcription activator ICP4 [15] , which binds strongly to viral DNA . Infection of control cells resulted in a significant localization of SUMO-1 and SUMO-2/3 conjugates , as well as PML , to sites closely associated with viral genomes ( Figure S1 , shNeg panels ) . In contrast , recruitment of PML and SUMO-conjugates was greatly reduced in cells depleted of Ubc9 ( Figure S1 , shUbc9 panels ) . We conclude that the SUMO conjugation pathway is required for the efficient recruitment of intrinsic antiviral factors to sites associated with incoming HSV-1 genomes and the efficient repression of ICP0-null mutant replication . ICP0 has been reported to induce the proteasome-dependent degradation of PML and its SUMO-modified isoforms [9]-[11] , as well as other SUMO-1 conjugated proteins following the ectopic expression of myc-tagged SUMO-1 [9] , [35] , [36] . We therefore decided to investigate the effect of ICP0 on the stability of endogenous SUMO-1 and SUMO-2/3 conjugated proteins . HSV-1 infection of HFs at a high multiplicity of infection ( MOI ) of 5 plaque forming units ( pfu ) per cell induced a general loss of high MW SUMO-conjugates in an ICP0- , RING finger- and proteasome-dependent manner ( Figure 2A , an independent experiment is shown in Figure S2B ) . Intriguingly , instead of a decrease , a significant increase in the levels of both SUMO-1 and SUMO-2/3 conjugates was detected at this MOI with ICP0-null ( ΔICP0 ) and ICP0 RING finger deletion ( ΔRING ) mutant viruses ( Figure 2A ) . The relative infection efficiencies of the wt and mutant viruses were compared by detection of the viral DNA polymerase accessory factor UL42 . While the mutant viruses exhibited some defect in viral gene expression at this MOI , this was not sufficient to explain the differences in SUMO-conjugate expression levels during infection ( Figure 2A ) . The C-terminus of ICP0 , encompassing its USP7 binding domain and ND10 localization sequences , was also required for efficient SUMO-conjugate degradation ( Figure S2C ) . This activity occurred in a number of other cell types , including primary keratinocytes ( HaCat , data not shown ) and HepaRG hepatocytes ( Figure S3A ) . As observed previously [19] , [37] , addition of the proteasome inhibitor MG132 not only inhibited ICP0-mediated degradation of SUMO-conjugates , but also led to a substantial increase in their abundance ( Figure 2A ) . Degradation of SUMO-conjugates in general occurred less rapidly than that of PML and its SUMO-modified isoforms in both HFs and HepaRG cell types ( Figure S3A ) . This suggests that the identity of a given substrate , as well as the fact that it is conjugated to SUMO species , influences the efficiency of ICP0-mediated degradation . These data are consistent with the initial localization of ICP0 to SUMO-conjugates localized at ND10 prior to bringing about their degradation ( Figures 2B and 2D , with further details in Figures S4A and S4C ) . Interestingly , even in PML-depleted cells , ICP0 colocalized with SUMO-2/3 conjugates , but not SUMO-1 conjugates , at the earliest stages of infection when only very low levels of ICP0 were present ( Figures 2C , 2E , S4B and S4D ) . Therefore , ICP0 localizes to sites that contain condensed SUMO-conjugates , either in the presence or absence of PML . ICP0-mediated degradation of SUMO-conjugates was also independent of PML expression ( Figure S3B ) . We conclude that ICP0 exhibits properties related to those of a STUbL , inducing the rapid loss of SUMO-modified PML followed by widespread proteasome-dependent degradation of SUMO-conjugates during infection in a RING finger-dependent manner . As ICP0 displayed STUbL-like properties , we next analyzed its polypeptide sequence for the presence of SIMs and found seven potential SIM-like sequences ( SLS ) , six that conform to the consensus ( with the possible exceptions of SLS-2 and -6 that contain proline residues ) , and one ( SLS-4 ) that has homology [IVISDS] to a SIM previously identified in hDaxx [38] and the HCMV regulatory protein IE2 [39] ( Figures 3A and 3B ) . These predicted SLSs of ICP0 are distributed throughout the entire ORF , with SLS-1 and -2 close to the RING finger , SLS-3 and -4 in the middle and SLS-5 to -7 in the C-terminal third of the protein . SLS-4 lies adjacent to a known phospho-serine region ( Figure 3B ) that is required for ND10 disruption in certain cell types [40] , [41] . Comparison of ICP0 sequences from HSV-1 and HSV-2 demonstrated that six of the predicted SIMs are conserved , the exception being SLS-3 ( Figure 3B ) . Consistent with the presence of potential SIMs , yeast two-hybrid ( Y2H ) assays demonstrated that ICP0 interacted with both SUMO-2 and -3 ( Figure 3C ) . ICP0 did not interact with SUMO-1 in this assay system , even though SUMO-1 was capable of interacting with hDaxx in a SIM-dependent manner ( Figure 3C and S5 , [38] ) . USP7 was used as a positive control for ICP0 interaction [42] . The ICP0-SUMO interaction was not mediated through covalent linkage , as SUMO-2/-3 mutants lacking the C-terminal di-glycine motif still interacted with ICP0 ( Figure 3C ) . Using a panel of C-terminal ICP0 deletion mutants we found that the SUMO-2/-3 interaction occurred in an ICP0 RING finger-independent manner and required residues 241-388 ( Figure 3D ) . These data suggest that either SLS-3 or SLS-4 could mediate the SUMO-2/-3 interaction . As SLS-3 is not conserved in HSV-2 ( Figure 3B ) , we investigated the requirement for SLS-4 and two other SIM-like sequences ( SLS-5 and -7 ) for interaction with SUMO in the context of full-length ICP0 . Mutation of SLS-4 abrogated the SUMO-2/-3 interaction in the Y2H system , but mutation of SLS-5 or -7 , either individually or combined , had no effect ( Figures 3E and 3F ) . We conclude that SLS-4 constitutes a genuine SIM that is specific for SUMO-2/3 . Given that the C-terminal third of ICP0 was shown to be required for the efficient degradation of SUMO-conjugates during infection ( Figure S2C ) , in vitro pull-down assays were performed using purified C-terminal fragments of ICP0 ( residues 594-775 linked to GST ) in order to determine if these sequences mediated any interaction with SUMO ( Figure 3G ) . In contrast to the Y2H analysis ( Figure 3F ) , the C-terminal fragment of ICP0 interacted with SUMO-1 , but not SUMO-2 , in this assay . Individual mutation of SLS-5 , but not SLS-7 , disrupted this interaction . Surprisingly , the C-terminal region of ICP0 mutated in both SLS-5 and SLS-7 interacted with SUMO-1 at wt levels . We conclude from these data that the C-terminal portion of ICP0 is able to interact with SUMO-1 and that SLS-5 probably represents an authentic SIM for SUMO-1 . However , further analysis will be required in order to define the basis of ICP0-SUMO-1 interaction within this region . The RING finger domain of ICP0 has E3 ubiquitin ligase activity in vitro in the presence of the E2 ubiquitin conjugating enzymes UbcH5a ( UBE2D1 ) and UbcH6 ( UBE2E1 ) [3] , [43] . Ubiquitination reactions carried in the presence of purified poly-SUMO-2 chains demonstrated that ICP0 could catalyze the formation of high MW poly-SUMO-2 ubiquitin conjugates in a RING finger-dependent manner , which required sequences encompassing SLS-4 ( compare ICP0 . 1-323 with ICP0 . 1-396 , Figure 4A ) . Mutation of SLS-4 within the context of ICP0 . 1-396 significantly reduced ICP0’s ability to mediate the poly-ubiquitination of SUMO-2 chains in vitro ( Figure 4A and B ) , even though this mutant had ubiquitin ligase activity equivalent to that of full-length ICP0 and ICP0 . 1-396 ( Figure 4A , right-hand panels ) . Taken together , these data indicate that ICP0 can directly interact with and ubiquitinate poly-SUMO chains in solution and that SLS-4 plays a role in this process . It is of interest to note that while ICP0 . 1-396 can mediate the ubiquitination of poly-SUMO2 chains in vitro , this activity was reduced in comparison to full-length ICP0 ( Figure 4A and B ) , suggesting that additional sequences and/or post-translational modifications may contribute to ICP0’s ability to ubiquitinate poly-SUMO2 chains in solution . To investigate the role of the SLSs within ICP0 in the degradation of high MW SUMO-conjugates , inducible cell lines expressing wt or SLS mutant forms of ICP0 were analyzed for SUMO-conjugate abundance following doxycycline induction ( Figure 4C ) . Expression of both wt and mSLS-5/7 ICP0 resulted in reduced levels of both SUMO-1 and SUMO-2/3 conjugates at 24 hours post-induction , whereas the mSLS-4 or mSLS-4/5/7 mutants did not . Analysis of PML stability demonstrated that wt , mSLS-4 and mSLS-5/7 forms of ICP0 could all induce the substantial degradation of PML , whereas this was much less marked in the case of the triple mutant ICP0-mSLS-4/5/7 ( Figure 4C , an independent time course experiment is shown in Figure S3C ) . In agreement with infection data ( Figure S3A ) , these results imply that ICP0 exhibits substrate selectivity with regard to SUMO-conjugated species , in that PML is targeted more efficiently than SUMO conjugates in general . Furthermore , since the defect of the triple mutant in PML degradation is substantially greater than either the single SLS-4 or the double SLS-5/7 mutations , it is possible that the SLSs are acting cooperatively . Mutation of SLS-5 and -7 in the C-terminal region of ICP0 greatly enhanced its accumulation after induction , and while the reasons for this remain unknown , it could result in an underestimation of any defects caused by these mutations . As an important control , we found that all SLS mutant forms of ICP0 were able to induce the formation of colocalizing conjugated ubiquitin ( Figure S6 ) , indicating that these mutations did not compromise the E3 ubiquitin ligase activity of ICP0 per se . Collectively , the data in Figures 3 and 4 indicate that ICP0 directly interacts with and ubiquitinates SUMO , and combined mutations within SLS-4 , -5 , and -7 reduce ICP0-dependent degradation of high MW SUMO conjugates . We conclude that ICP0 has STUbL-like properties . We next tested whether ICP0 induces the degradation of PML in a SUMO modification-dependent manner . PML is expressed as a complex family of related isoforms that contain a SIM and multiple SUMO modification sites [24] , [44] , [45] ( Figure 5A ) . Using a series of cell lines expressing individual EYFP-linked PML isoforms in normal or PML-depleted backgrounds [46] , we found that the SUMO-modified forms of all PML isoforms were degraded in an ICP0- and proteasome-dependent manner during infection , consistent with previous studies analysing ICP0’s ability to induce the degradation of endogenous PML during infection [9]–[11] . In contrast , the unmodified forms of EYFP-PML were relatively resistant to degradation , with the exception of PML isoform I ( Figure 5B ) . Individual EYFP-PML isoform degradation was not dependent on endogenous PML as similar results were obtained in PML-depleted cells ( data not shown ) . Using isoforms I and IV as example substrates , we investigated the requirement for SUMO modification for ICP0 induced degradation of PML . Lysine to arginine mutations at residues 160 and 490 ( K160/490R ) , the two major SUMO modification sites [47] , were expressed in control and PML-depleted cells and monitored for their respective stabilities during infection . In contrast to PML . IV . K160/490R , PML . I . K160/490R was readily degraded in an ICP0-dependent manner ( Figures 5C and 5D ) . Although the double K160/490R mutation substantially reduced SUMO modification , some modified bands remained , particularly in the presence of endogenous PML . Inclusion of the K65R mutation ( K65/160/490R ) , affecting the other published SUMO modification site [47] , did not alter this banding pattern [17] . We noted that lysine 616 also falls within a good SUMO modification consensus sequence ( LKID ) . Additional mutation of K616R ( K160/490/616R ) , either in the presence or absence of the K65R mutation , reduced modification to undetectable levels in both endogenous and PML-depleted backgrounds ( Figure 5E; [17] ) . This mutant form of PML . I remained equally sensitive to ICP0-mediated degradation ( Figure 5E ) , indicating that sequences specific to exon 9 within PML . I ( Figure 5A ) , either directly or indirectly , confer additional sensitivity to ICP0-mediated degradation independent of PML SUMO modification status . These data suggest that ICP0 utilizes dual targeting mechanisms to mediate the degradation of PML during infection , one being a SUMO-dependent mechanism that leads to the preferential degradation of all SUMO-modified PML isoforms , and the other a SUMO modification-independent mechanism that can target PML . I for degradation via sequences encoded by exon 9 . We next tested the hypothesis that SLSs within ICP0 contribute to its ability to stimulate HSV-1 lytic infection and reactivation of gene expression from quiescence . Cell lines that express various mutants of ICP0 in an inducible manner were tested for their ability to stimulate plaque formation of a HSV-1 ICP0-null mutant virus and to reactivate gene expression from quiescent HSV-1 genomes . The use of these assays to analyze other mutants of ICP0 has been described in detail elsewhere [8] . N-terminal fragments 1–240nls or 1–340nls had negligible complementation activity in ICP0-null mutant virus plaque assays , whereas fragments 1–374nls , 1–396nls and 1–594 , that include SLS-4 , exhibited detectible levels of complementation activity ( Figure 6B ) . Mutation of SLS-4 within constructs 1–396nls and 1–594 virtually eliminated complementation , indicating that SLS-4 contributes to ICP0 activity in the context of these shorter fragments . Individual mutation of SLS -4 , -5 or -7 in full-length ICP0 had varying but lesser effects on ICP0 complementation efficiency . However , similar to PML degradation ( Figure 4C and S3C ) , the triple mutant was significantly less active than wt ICP0 ( Figure 6B ) . In assays monitoring the reactivation/derepression of β-galactosidase gene expression from cells harbouring quiescent HSV-1 genomes , large reductions in activity resulted from mutation of SLS-4 in the 1-396nls and 1–594 backgrounds , again highlighting the importance of SLS-4 in the context of these shorter ICP0 fragments ( Figure 6C ) . Comparison of the effects of individual and combined SLS mutants in the full-length ICP0 again demonstrated varying levels of activity , with the triple SLS mutant being the most defective with over a 50% drop in reactivation efficiency ( Figure 6C ) . We have observed that mutant ICP0 proteins frequently retain greater relative activity in this reactivation assay than in the complementation assay [8] , [48] . We conclude that SIM-like sequences within ICP0 significantly contribute to its biological functions , both in its ability to complement ICP0-null mutant plaque formation and , to a lesser extent , the reactivation of quiescent HSV-1 gene expression . However , it is likely that additional sequences within the C-terminal third of ICP0 also contribute to its functionality in these assay systems . We have shown that related RING finger viral orthologues of ICP0 , including BICP0 ( from bovine herpes virus type 1 ) , EICP0 ( from equine herpes virus type 1 ) , and PICP0 ( from pseudorabies virus ) , have E3 ubiquitin ligase activity in vitro and can partially substitute the functional requirement for ICP0 during HSV-1 ICP0-null mutant infection [48] . Importantly , these viral orthologues were shown not only to inhibit the recruitment of PML to sites associated with incoming HSV-1 viral genomes , but also to modulate the SUMO conjugation profile of PML and/or Sp100 [48] . We therefore investigated whether STUbL-like properties were a conserved activity of the ICP0 family of RING finger proteins . All ICP0 viral orthologues , including VICP0 from varicella zoster virus , contain multiple putative SIM-like sequences , with at least one SLS conforming to the SIM consensus ( Figure 7A and 7B bold highlights , [49] ) . SLS-3 of BICP0 also shares homology ( Figure 7B , solid box ) with the phospho-serine region downstream of SLS-4 of ICP0 ( IVISDS , Figure 3B ) . We note that the other viral orthologues also contain similar acidic and/or serine amino acid sequences following some of their putative SIM-like sequences ( Figure 7B , dashed boxes ) . These negatively charged motifs are similar to other examples that have been shown to enhance SIM-SUMO interactions [27] , [28] , [50] . Utilizing inducible cell lines that express ICP0 or related viral proteins , we found that like ICP0 ( Figure 4C ) , BICP0 , EICP0 , and PICP0 were all capable of reducing the overall abundance of high MW SUMO conjugates following doxycycline induction ( Figure 7C and S7 ) . As found previously [48] , VICP0 was expressed at insufficient levels to be active in this experimental system ( Figure S7 ) . Ubc9 levels remained unaltered in cells expressing ICP0 or related viral orthologues ( Figure 7C ) , indicating that the degradation of SUMO conjugates could not be explained by indirect degradation of Ubc9 . We conclude that the induced loss of SUMO conjugates by ICP0 and its viral orthologues occurs independently of viral infection and that STUbL-like activity is a conserved property of these related viral RING finger ubiquitin ligases . In this report we demonstrate that the viral ubiquitin ligase ICP0 has STUbL-like properties that contribute to its ability to counteract host-cell intrinsic resistance to HSV-1 infection . Intrinsic resistance represents the first line of intracellular antiviral defence and , unlike innate or acquired antiviral immunity , is mediated by pre-existing cellular factors that attempt to restrict viral replication during the initial stages of infection ( for reviews see [14] , [51] , [52] ) . In the case of herpesviruses , intrinsic resistance leads to the repression of viral gene expression , which may reflect an important biological aspect of how these viruses attain a quiescent state of infection prior to the establishment of latency . Whilst the role of chromatin modification on viral gene transcription has been investigated in many laboratories , we have identified an additional aspect to intrinsic resistance ( not necessarily unconnected ) that involves the SUMO conjugation pathway and components of ND10 [12] , [13] , [15] , [16] , [31] . We have recently shown that recruitment of PML and hDaxx to foci associated with incoming viral genomes is dependent upon their SIMs , and that additional SUMO-2/3 conjugates , as well as the SUMO E3 ligase PIAS2β , are also recruited to viral genomes in a PML-independent manner [17] . Consistent with these observations , we show here that depletion of Ubc9 , the sole SUMO E2 conjugating enzyme , restricts the cell’s ability to repress ICP0-null mutant virus replication ( Figure 1C and 1D ) and inhibits the recruitment of SUMO conjugates and PML to foci associated with incoming viral genomes ( Figure S1 ) . Collectively these data demonstrate that the SUMO conjugation pathway plays a role in intrinsic resistance to HSV-1 infection . This phenotype is consistent with previously documented roles for SUMO conjugation in transcriptional repression ( reviewed in [53] ) , the assembly of ND10 [38] , [44] , and the recruitment of chromatin modifying enzymes into ND10 [54] . It is therefore likely that in the context of intrinsic antiviral defence this pathway is involved in the assembly of a network of repressive factors that associate with viral genomes following their entry into the cell nucleus in order to bring about repression of viral transcription ( summarized in Figure 8 ) . Analogous to viral counter measures against innate and acquired immunity , viruses have evolved mechanisms to disarm intrinsic antiviral defence . One of the first proteins to be expressed during HSV-1 infection is ICP0 , a viral RING finger ubiquitin ligase that localizes to and disrupts ND10 by mediating the degradation of PML , its SUMO-modified isoforms , and SUMO-modified Sp100 [9]–[11] . The degradation and dispersal of ND10 constituent proteins correlates well with ICP0’s ability to counteract intrinsic defence , thereby aiding the efficient initiation of viral replication . However , the precise mechanism ( s ) by which ICP0 targets these ND10 proteins for degradation has remained elusive . Given the parallels between ICP0 and the cellular STUbL RNF4 [22] , [23] , [29] , we decided to investigate if ICP0 possessed STUbL-like properties that contribute to its ability to counteract intrinsic defence . Whilst we show that ICP0 shares some phenotypic similarities to RNF4 , a number of important differences have also been highlighted . Like RNF4 , ICP0 preferentially induces the degradation of SUMO-modified forms of PML , but unlike RNF4 [23] , [30] , ICP0 is able to target PML without the need for additional PML SUMO modification ( Figures 5B and S3A ) . In addition , ICP0 also induces the widespread degradation of SUMO-1 and SUMO-2/3 conjugate proteins during infection ( Figure 2A and S2B ) . Importantly , this activity is not dependent upon the presence of PML ( Figure S3B ) , indicating that this phenotype is not an indirect consequence of ND10 disruption . These data demonstrate that ICP0 targets additional SUMO-modified proteins for degradation other than those constitutively modified at ND10 , consistent with its ability to localize to SUMO-2/3 conjugates in PML-depleted cells during the initial stages of infection ( Figure 2E ) . These differences , plus others described below , indicate that ICP0 does not represent a precise viral orthologue of RNF4 . Although ICP0 contains many SIM-like sequences , only one of these was strongly identified as a functional SIM ( SLS-4 , IVISDS ) . This motif shares homology to previously characterized SIMs within hDaxx and the HCMV IE2 protein ( Figure 3B; [38] , [39] ) . SLS-4 is required for ICP0’s ability to interact with SUMO-2 ( Figure 3F ) , ubiquitinate poly-SUMO-2 chains in vitro ( Figure 4A ) , and to reduce the level of SUMO conjugates when expressed by itself in an inducible cell line system ( Figure 4C ) . These data support the hypothesis that ICP0 has STUbL-like activity . It is of interest to note that serine residues adjacent to SLS-4 can be phosphorylated [41] and are required for ICP0’s ability to disperse PML in transfection based assays in certain cell types [40] . It is plausible therefore that this particular SIM is regulated by phosphorylation , similar to those previously identified in SUMO ligases [27] . Whilst the extent to which the other SIM-like sequences contribute to SUMO interaction remains unclear , our data indicate that the C-terminal third of ICP0 can additionally interact with SUMO-1 , potentially in a SLS-5 dependent manner ( Figure 3G ) . However , we note that this region influences many aspects of ICP0 function , including USP7 and CoREST binding [42] , [55] , ICP0 multimerization and ND10 localization [56] . Whilst ICP0’s STUbL-like activity is important for its biological functions ( Figures 6B and 6C ) , it is important to note that ICP0 also directly interacts with and ubiquitinates other cellular proteins such as RNF8 , USP7 and p53 in a SUMO modification-independent manner [43] , [57] , [58] . Here we found that PML . I can also be targeted for degradation in a SUMO modification-independent manner ( Figures 5B–E ) . Thus , ICP0 clearly has both SUMO-dependent and -independent targeting specificities that may have an overall accumulative effect on counteracting intrinsic antiviral resistance . Furthermore , since SUMO-modified PML is degraded more rapidly than the bulk of SUMO conjugates , it appears that the identity of a SUMO-modified protein also influences ICP0 substrate targeting . Indeed , it is possible that optimal ICP0 targeting involves both SUMO- and substrate-dependent interactions that synergize to define the most avid ICP0 targets . Another important consideration is the distinction between biochemical targeting and spatial localization . SUMO-modified PML may be degraded more rapidly than other SUMO conjugates because ICP0 interacts more avidly with the former than the latter , or because the high concentration of ICP0 within ND10 preferentially enhances degradation of SUMO conjugates within these structures . While these two factors may well be related , we note that ICP0 colocalizes with SUMO-2/3 conjugates even in the absence of PML ( Figure 2E ) . Similarly , the reduced activity of the C-terminal deletion mutant ICP0 . 1-594 on both PML [9] and SUMO conjugates ( Figure S2C ) could be due to less efficient biochemical targeting at a molecular level ( Figure 3G ) , or because this mutant is diffusely distributed in the nucleus and not spatially targeted to ND10 where intra-nuclear SUMO conjugates accumulate . Viruses have evolved numerous mechanisms to exploit ubiquitin conjugation in order to create cellular environments that favour viral replication . Our data identify a dual targeting ( SUMO- and substrate-dependent ) mechanism through which ICP0 manipulates the cellular environment in favour of HSV-1 replication . Given that the STUbL-like properties of ICP0 appear to be a conserved activity of this family of viral RING finger ubiquitin ligases ( Figure 7C ) , it would be interesting to determine if this is a common mechanism for substrate targeting utilized by other viral ubiquitin ligases . The observation that ICP0 targets SUMO-conjugated proteins in general for proteasome-mediated degradation provides a plausible explanation for its ability to inhibit multiple factors involved in intrinsic antiviral defence ( summarized in Figure 8 ) . This activity may also account for other proposed roles for ICP0 in regulating other cellular pathways , including innate interferon-mediated defence and DNA damage response pathways , as components of these pathways have been shown to be regulated by SUMO modification [59] , [60] , [61] , [62] . We stress , however , that the target specificity of ICP0 is regulated by factors in addition to the presence of conjugated SUMO , and that the biologically relevant targets will comprise only a minority of total SUMO-conjugated proteins . Nonetheless , our observations suggest a more general role for SUMO conjugation in resistance to pathogen infection . Evidence in support of this hypothesis includes: ( i ) that the chicken adenovirus protein Gam-1 inactivates the SUMO conjugation pathway by targeting the SUMO E1 activating enzyme complex in order to stimulate viral transcription [63]; ( ii ) that a pathogenic bacterium impairs the SUMO modification pathway to enhance infection [64]; and ( iii ) that SUMO modification of transcriptional regulatory proteins is frequently associated with transcriptional repression [53] . Since the recruitment of ND10 components to sites associated with viral genomes occurs extremely quickly and is independent of viral transcription [16] , we propose that the SUMO pathway may regulate a process that responds to the entry of foreign DNA in general into the cell nucleus . Further investigation into ICP0’s SUMO-targeted and SUMO-independent ubiquitin ligase activities will provide insight into the cellular processes that regulate this response to infectious pathogens . Human foetal foreskin diploid fibroblasts ( HFs ) were grown in Dulbeccos Modified Eagles Medium with 10% fetal calf serum ( FCS ) . HepaRG cells [65] were grown in Williams Medium E with 10% fetal bovine serum Gold ( PAA Laboratories Ltd ) , 2 mM glutamine , 5 µg/ml insulin and 0 . 5 µM hydrocortisone . All cell growth media contained 100 units/ml penicillin and 100 µg/ml streptomycin . PML-depleted and control HepaRG and HFs cells were described previously [12] , [13] . Control and PML-depleted HepaRG cells reconstituted with individual PML isoforms expressed at close to endogenous levels and derivatives expressing PML isoforms I and IV with lysine to arginine substitutions at the known SUMO modification sites and at lysine residue 616 have been described previously [46] . PML isoforms are named according to [45] . Tetracycline inducible HepRG cells expressing ICP0 or alpha herpes viral orthologues have also been described previously [48] . Ubc9-depleted cells were constructed by lentiviral transduction , as described in [13] expressing a shRNA based upon a 19-mer ( 5’ GAAGTTTGCGCCCTCATAA 3’ ) within the Ubc9 open reading frame . Wild type HSV-1 strain 17syn+ , its ICP0-null mutant derivative dl1403 [66] , RING finger deletion mutant FXE [67] , and C-terminal truncation mutant E52 ( expressing ICP0 . 1-594 [56] ) were grown and titrated as previously described [68] . Derivatives of wt ( in1863 ) and ICP0-null mutant ( dl1403/CMVlacZ ) HSV-1 that contain a β-galactosidase gene linked to the human cytomegalovirus immediate-early promoter/enhancer inserted into the tk locus were used for plaque assays as described [12] . Cells in 24-well dishes at 1×105 cells per well were washed with phosphate buffered saline before harvesting in SDS-PAGE loading buffer . Proteins were resolved by SDS-PAGE and transferred to nitrocellulose membranes for western blotting . Monoclonal antibodies utilized recognised the following proteins: actin ( AC-40 , Sigma-Aldrich ) , tubulin ( T4026 , Sigma-Aldrich ) , ubiquitin ( P4D1 , Santa Cruz ) , ICP0 ( 11060 , [69] , UL42 [70] , ICP4 ( 58S , [71] , myc ( 9E10 sc-40 , Santa Cruz ) and PML 5E10 [72] . Rabbit polyclonal antibodies were used to detect Sp100 ( SpGH [73] ) , PML ( sc-9863 , Santa Cruz ) , Ubc9 ( ab30505 , AbCam ) ; EGFP ( ab290 , Abcam ) , SUMO-1 ( ab32058 , Abcam ) , and SUMO-2/3 ( ab3742 , Abcam ) . Cells on 13 mm glass coverslips were fixed and permeabilized using 2 . 5% non-buffered formaldehyde and 0 . 5% Triton-X100 in 10 mM HEPES ( pH 7 . 0 ) , 100 mM NaCl , 300 mM sucrose , 3 mM MgCl2 , 5 mM EGTA . The secondary antibodies used were Alexa 488 , 594 , and 633 conjugated donkey anti-rabbit , - sheep , and -mouse IgG ( Invitrogen ) . A glycerol-based mounting medium was used ( Citifluor AF1 ) . The samples were examined using a Zeiss LSM 510 confocal microscope with 488 nm , 543 nm and 633 nm laser lines and a x63 Plan-Apochromat oil immersion lens , NA 1 . 40 . Exported images were processed using Adobe Photoshop with minimal adjustment and assembled for presentation using Adobe Illustrator . Y2H analysis was based upon the Matchmaker 3 system ( Clontech ) using AH109 and Y187 yeast strains . ICP0 cDNAs encoding wt ICP0 . 1–775 , ICP0 . 1–775ΔRING ( FXE ) , ICP0 . 1–388 and 1–241 , along with hDaxx and hDaxx . mSIM ( C-terminal SIM mutant aa 733–740 IIVLSDSD to IGAGSDSD , [17] ) were cloned into pGAD-T7 in frame with the GAL4 activation domain ( AD ) . cDNAs encoding SUMO isoforms , their inactive conjugation mutants ( -GG to -AA ) , pp71 and USP7 were cloned into pGBK-T7 in frame with the GAL4 DNA binding domain ( BD ) . Transformed yeast colonies were picked , mated overnight , and diploids serially diluted prior to plating out onto selective medium ( as highlighted ) following the manufacturer’s guidelines . Colonies were allowed to grow for 72 hours prior to image capture . Full-length poly-histidine tagged ICP0 and ICP0ΔRING were purified as previously described [3] . Poly-histidine tagged ICP0 . 1–323 , ICP0 . 1–396 , ICP0 . 1–396SLS-4 , SUMO-1 and SUMO-2 were purified from bacterial extracts utilizing Nickel agarose affinity chromatography and dialysed into 50 mM Tris ( pH 7 . 5 ) , 150 mM NaCl , 2 . 5% glycerol , 2 mM MgCl2 , 1 mM DTT . Poly-HisSUMO-2 chains were purchased from Boston Biochem . In vitro ubiquitination reactions were carried out in the presence of 20 ng poly-HisSUMO-2 chains as essentially described in [43] utilizing 10 ng E1 activating enzyme , 30 ng UbcH5a , 2 . 5 µg ubiquitin ( Sigma-Aldrich ) , and 100 ng ICP0 , ICP0ΔRING , ICP0 . 1–323 , or ICP0 . 1 . 396 in the presence of 50 mM Tris ( pH 7 . 5 ) , 50 mM NaCl , 1 mM MgCl2 and 5 mM ATP . Reactions were carried out at 37°C for 90 minutes and terminated by the addition of boiling mix containing 8 M urea and 100 mM DTT . Quantification of poly-ubiquitinated poly-SUMO2 chains was performed by densitometry analysis of western blots using Quantity One software ( Bio-Rad ) . Glutathione-S-Transferase ( GST ) pull-downs were carried out in buffer H ( 50 mM HEPES pH 7 . 0 , 150 mM NaCl , 5 mM β-mercaptoethanol and 0 . 1% NP-40 ) using beads bound to either GST alone or GST linked to the C-terminal 594-775 amino acids of ICP0 ( GST-E52; [56] ) or equivalent fragments with mutations in SLS-5 , -7 , or -5/7 in 1 . 5 ml of precleared bacterial supernatants containing either His-tagged SUMO-1 or SUMO-2 for 90 minutes end-over-end at 4°C . The beads were washed three times in 1 ml buffer H and soluble complexes were eluted in 60 µl of 1x SDS-PAGE loading buffer . HepaRG cells expressing wt and 1–594 C-terminal truncation mutant ICP0 proteins in an inducible manner have been described previously [8] . A series of C-terminal truncation mutants of ICP0 were constructed using existing restriction sites or previously described EcoRI linker insertion mutants [5] as follows: ICP0 . 396nls and ICP0 . 1–517 used the NotI or MluI sites in the ICP0 cDNA; ICP0 . 1–547 , 1–374nls and 1–340nls used linker insertions E1 , E51 and E15; ICP0 . 1–241nls used the ICP0 truncation fragment including the first 241 codons described previously [3] . ICP0 . 1–547 and ICP0 . 1–517 include the normal ICP0 nuclear localization signal and the constructs include a C-terminal linker containing stop codons . The shorter truncation mutants ( with nls in the name ) contain a C-terminal linker encoding the SV40 T-antigen nuclear localization signal followed by a stop codon . Substitution mutants in SLS-5 ( VVAL to GGAL ) and SLS-7 ( VVLV to GGLV ) were constructed by a PCR splicing approach . MluI-SalI fragments containing the desired mutations were transferred into the wt ICP0 lentivirus expression vector . Substitution mutants in SLS-4 ( PIVI to PGAG ) were constructed by oligonucleotide synthesis of the region between the SfiI and NotI sites in the ICP0 cDNA , maintaining the coding potential except for the desired mutations , while decreasing the GC content by introduction of silent mutations , then rebuilding the oligonucleotides into the wt cDNA . The double mutant mSLS-5/7 was prepared by serial mutagenesis . The mSLS-4 mutants in the 1-594 and 1-396nls truncations and the mSLS-5/7 double mutant were constructed using the NotI site on the 3’ side of the mutated SLS-4 motif . All mutants were confirmed by extensive DNA sequence analysis after insertion into lentiviral vectors . Cell lines expressing these proteins were isolated as described previously [8] . Assays of complementation of plaque formation by ICP0-null mutant HSV-1 and derepression of quiescent HSV-1 after induction of ICP0 expression were performed and quantified as described [8] .	Viruses must evade several antiviral defences in order to establish a productive infection . These include antibody- and cell-mediated acquired immunity and interferon-regulated innate immunity . Recently , a third arm of antiviral defence has been discovered , so called intrinsic immunity . This aspect of antiviral resistance represents the first line of intracellular defence against virus infection and is mediated by pre-existing cellular factors that attempt to repress viral replication during the initial stages of infection . Like acquired and innate immunity , viruses have evolved mechanisms that overcome intrinsic defence . Here we show that in response to herpes simplex virus type-1 ( HSV-1 ) infection an important aspect of intrinsic immunity is regulated by the small ubiquitin-like modifier ( SUMO ) conjugation pathway . In response to this defence , the virus induces rapid degradation of specific SUMO-conjugated proteins , followed by widespread loss of SUMO-conjugated species in general . Inactivation of the SUMO pathway inhibits the cell’s ability to efficiently repress viral replication in the absence of this viral countermeasure . Our data identifies an important regulatory pathway that mediates intrinsic resistance to HSV-1 infection and describes the biochemical mechanism that the virus utilizes in order to counteract this antiviral defence .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "virology", "biology", "microbiology" ]	2011	A Viral Ubiquitin Ligase Has Substrate Preferential SUMO Targeted Ubiquitin Ligase Activity that Counteracts Intrinsic Antiviral Defence
The frequent dispensability of duplicated genes in budding yeast is heralded as a hallmark of genetic robustness contributed by genetic redundancy . However , theoretical predictions suggest such backup by redundancy is evolutionarily unstable , and the extent of genetic robustness contributed from redundancy remains controversial . It is anticipated that , to achieve mutual buffering , the duplicated paralogs must at least share some functional overlap . However , counter-intuitively , several recent studies reported little functional redundancy between these buffering duplicates . The large yeast genetic interactions released recently allowed us to address these issues on a genome-wide scale . We herein characterized the synthetic genetic interactions for ∼500 pairs of yeast duplicated genes originated from either whole-genome duplication ( WGD ) or small-scale duplication ( SSD ) events . We established that functional redundancy between duplicates is a pre-requisite and thus is highly predictive of their backup capacity . This observation was particularly pronounced with the use of a newly introduced metric in scoring functional overlap between paralogs on the basis of gene ontology annotations . Even though mutual buffering was observed to be prevalent among duplicated genes , we showed that the observed backup capacity is largely an evolutionarily transient state . The loss of backup capacity generally follows a neutral mode , with the buffering strength decreasing in proportion to divergence time , and the vast majority of the paralogs have already lost their backup capacity . These observations validated previous theoretic predictions about instability of genetic redundancy . However , departing from the general neutral mode , intriguingly , our analysis revealed the presence of natural selection in stabilizing functional overlap between SSD pairs . These selected pairs , both WGD and SSD , tend to have decelerated functional evolution , have higher propensities of co-clustering into the same protein complexes , and share common interacting partners . Our study revealed the general principles for the long-term retention of genetic redundancy . Genetic robustness in yeast cells accounts for insignificant phenotypic consequences upon deletion of many genes [1] , [2] . It is thought that such resilient design of the genetic program is achieved in two different ways . In the first scenario , genes performing related functions are distributed on alternate pathways [3] , [4] mimicking the electric parallel circuits so its alternate paths can compensate that blockage of one pathway . The second strategy to achieve robustness is by gene duplication , i . e . null mutation on one gene can be buffered by its paralogous copy which shares overlapping function [5] . This notion is supported by recent investigations which showed that mutual compensation is prevalent among paralogs [6]–[8] , but contradicts population genetic theories predicting that genetic redundancy is evolutionarily unstable [9] . The instability can be understood when considering the evolutionary fate of duplicated genes [10] . Upon duplication , the paralogs usually go through a short-lived and transient state of complete redundancy , followed by a non-functionalization process that leads to massive loss of duplicates [10] . To persist , duplicate genes usually have to functionally diverge , either through subfunctionalization ( partition of ancestral functions ) or neofunctionalization ( independent gain of novel functions ) [10]–[13] . Regardless of how the paralogs had navigated an evolutionary trajectory from the transient complete redundancy to the long-time retention , the sister paralogs are anticipated to share fewer functions as time progresses . Therefore , the missive loss of duplicated genes and the highly divergent functions between the long-term retained pairs appear to be contradictory to the genetic redundancy provided by paralogs . More perplexingly , even for the duplicates that have backup capacity , several recent studies reported that little functional similarity is shared between them [8] , [14] , leading to the hypothesis of “backup without redundancy” [14] . As these previous observations were made on small datasets from double-gene deletion experiments in budding yeast , it is necessary to re-examine the relationship between the cellular robustness and gene redundancy using more recent and larger datasets , and more importantly , to include paralogs arising from different evolutionary origins . In this study , we based our analysis on the synthetic genetic interactions derived from a recent landmark study , in which ∼2 , 000 genes were queried against the rest of the genome for synthetic genetic interactions ( epistasis ) [15] . This data set , larger than any other previous yeast double-deletion experiments , provides us a unique opportunity to systematically examine the genetic buffering between ∼500 duplicate gene pairs on a genome-wide scale . Moreover , this data set includes duplicate pairs from both whole-genome duplication ( WGD ) and small-scale duplications ( SSD ) , allowing us to compare duplicates with different origins in an unbiased manner . Our analysis confirmed the previous reports , which were based on much smaller datasets , about the prevalent mutual compensation among paralogs , both from WGD and SSD . However , in contrast with “backup without redundancy” , our further examination suggests that functional overlap/redundancy between paralogs is a key determinant of backup capacity between duplicates , with which the buffering potential of any given pair can be accurately predicted . More interestingly , although mutual compensation among duplicate genes is prevalent , we found that the evolution of genetic robustness by gene duplication follows a neutral mode , i . e . the loss of backup capacity being proportional to background mutations accumulated in the divergence time since duplication . Under the neutral mode , although massive duplicates had lost their mutual compensation , we also found natural selection plays a role in maintaining long-term retention of the backup capacity between a few duplicates , which requires slowly evolved functions between paralogs . Among the duplicate pairs assayed , we found 39 . 5% ( 105/266 ) of the WGD paralogs had significant aggravating interactions , in comparison with 18 . 4% ( 42/228 ) for SSD paralogs ( Figure 1 ) . The percentage of backup pairs for WGD was comparable to what was previously reported ( ∼35% ) [8] , where random spore analysis ( RSA ) and growth curve analysis ( GCA ) rather than SGA were used to determine the compensatory effects between WGD paralogs . It was interesting that the percentage of backup pairs was much lower for SSD pairs than WGD pairs ( Figure 1 ) ; such a reduced dispensability of SSD duplicates was previously speculated from single-gene deletion experiments [20] . We further designed two control sets to determine the statistical significance of the observed compensation between duplicated genes . First we randomly chose gene pairs that have genetic interactions regardless of being duplicate or singleton , and found only 7% of the pairs have aggravating interactions ( see Figure 1A and Materials and Methods ) . Second , we took all the duplicated genes and randomly grouped them into pairs , and found that only 6 . 6% of these random pairs have aggravating interactions . This ruled out the possibility that the observed preferential buffering between duplicates was simply due to that duplicate genes might have more aggravating genetic interactions than singleton genes . Comparing the percentages for the control sets with the percentages of 39 . 5% for WGD paralogs and 18 . 4% for SSD paralogs ( Figure 1B ) , our analysis established that duplicates indeed have excessive backup capacity , which results from their intrinsically shared properties . As SGA provides quantitative measurements for the interaction strength between any gene pairs [15] , we next studied the backup strength between paralogs . Compared with two control sets , we found the interaction strength between duplicate pairs was much stronger with the average scores of −0 . 42 and −0 . 33 for WGD and SSD , respectively , in sharp contrast with −0 . 07 and −0 . 069 for the two random control sets , respectively ( see Figure 1B , P = 8 . 54×10−36 for WGD , P = 1 . 87×10−6 for SSD and P = 0 . 06 between WGD and SSD ) . We note that these findings are in agreement with what was previous reported from analysis on much smaller datasets [6]–[8] , [14] . Taken together , our analysis established that strong genetic buffering capacity is prevalent between WGD and SSD paralogs , which provides enhanced genetic robustness in yeast cell . Intuitively , genetic robustness by redundancy between gene duplicates should be attributed to their functional similarities . However , conflicting observations were reported in the recent literature [6]–[8] , [14] . It was suggested that functional redundancy between buffering duplicates is minimal [8] , [14] , which gave rise to the hypothesis of “backup without redudancy” [14] . In these earlier studies , functional similarity between paralogs was characterized based on their resemblance in gene expression profiles , protein interactions , or genetic interaction profiles . However , two genes may still buffer each other even though they only have limited functional overlap , which does not require them to have near identical profiles of gene expression or genetic interactions . Supporting this notion , Kafri et al . proposed a model of transcriptional reprogramming , which predicted that differentially expressed duplicates were more likely to buffer each other [21] . Therefore complementary to indirect metrics that score overall functional similarity between duplicate copies ( inferred from sequences or expression profiles ) , a new metric is required to specifically and directly quantify the extent of functional overlap between paralogs . In this study , we used a metric called GO-div to gauge functional overlap between paralogs directly from their respective GO annotations ( see Materials and Methods and also Figure S1 ) . GO-div previously was used to benchmark data obtained from high-throughput experiments [22] , and here we adopted this approach to quantify functional overlap between duplicate genes . Conceptually , GO-div measures the semantic dissimilarity between the sets of Gene Ontology annotations associated with a pair of genes [22] and is calculated on the basis of resemblance between the “best matched” GO terms between sister paralogs , most notably not affected by other diverged functions ( see Materials and Methods and also Figure S1 for a schematic illustration ) . Higher GO-div indicates less functional overlap between paralogs while lower GO-div indicates both paralogs at least share some very specific functions even though they have diverged in other functions . Although current gene annotations might be incomplete , given the extensive effort in characterizing yeast genes in the past several decades , GO-div calibrates functional overlap between two genes at least within the best of our current knowledge . Complementary to GO-div , we also calculated the non-synonymous substitution rate per site ( Ka ) between paralogs to represent overall divergence in protein coding sequence between paralogs [23] . Worthy of note , GO-div was moderately correlated with Ka with R = 0 . 2 and P<0 . 05 . The statistical significance indicated their intrinsic consistency in characterizing functional similarity between gene pairs , while the weak correlation suggested that only 4% ( R2 ) of the variation in GO-div could be explained by Ka , highlighting the non-redundancy of using the two metrics in studying functional divergence . Among all the duplicate pairs we examined , we found that substantial functional redundancy between paralogs ( for both WGD and SSD duplicates ) was a key determinant of their genetic backup capability . First , as revealed by Figure 2 , duplicate pairs , either WGD ( Figure 2A ) or SSD ( Figure 2B ) , are more likely to buffer each other if they have less diverged functions; this trend stands when functional divergence was estimated either by the direct measure ( GO-div ) or by ka . Secondly , for the buffering pairs from both WGD and SSD , we found the buffering strength between the paralogs was significantly correlated with their functional divergence ( see Figure 2C and 2D ) scored by GO-div , having Pearson's R = 0 . 34 , P = 3 . 1×10−4 for WGD pairs and R = 0 . 37 , P = 0 . 01 for SSD pairs . The correlation is also significant when using Ka to approximate functional divergence between paralogs in both WGD and SSD , with Pearson's R = 0 . 41 , P = 1 . 5×10−5 for WGD pairs and R = 0 . 33 , P = 0 . 03 for SSD pairs . In addition , we also found expression divergence between duplicates ( see Materials and Methods ) is significantly correlated with their buffering strength for SSD paralogs with R = 0 . 33 , P = 0 . 03 , but not for WGD pairs . This lessened significance of the correlation highlights the superiority of using a direct metric to quantify functional redundancies between duplicates . Taken together , such a tight coupling ( see Figure 2C and 2D ) between buffering strength and functional overlap between paralogs suggested that the observed prevalent mutual compensation between paralogs ( Figure 1 ) is indeed maintained by their functional similarity , and the less diverged pairs tend to have stronger buffering strength . It is also important to note that WGD and SSD paralogs have different origins and functional propensities [16] , [24] , therefore our consistent observation on these two classes of duplicates suggested our conclusion was not biased towards particular function categories ( as shown in Figure 1 and Figure 2 ) . Having established the role of genetic redundancy in cellular robustness , we ask whether backup capacity can be predicted for any unseen duplicate pairs . To test this , we pooled the WGD and SSD duplicates together , labeled the 147 pairs ( WGD+SSD ) that have backup capacity as positive samples and the remaining non-backup pairs as negative . We characterized each pair with a feature vector , each element being a direct or indirect metric measuring their functional divergence , including Ka , sequence identity , expression divergence and GO-div . A support vector machine ( SVM ) was subsequently implemented to classify these paralogs into either with backup capacity or without . A 3-fold cross-validation , as demonstrated in Figure 2E , suggested that the degree of functional overlap between any paralog pairs was sufficient to distinguish backup pairs from non-backup pairs , with AUC = 0 . 74±0 . 05 . Such a high predictive power further strengthened our argument that backup between paralogs stems from their functional redundancy . It is also important to note that GO-div , which scores the specificity of the best-matched functions between paralogs , is the strongest indicator among all the features to predict backup capacity , and using GO-div alone can achieve AUC = 0 . 7 , higher than using combination of any other features ( AUC = 0 . 67 ) . In the above we described the presence of prevalent mutual compensation between paralogs ( Figure 1 ) and established that such compensation is maintained by functional overlap ( Figure 2 ) . However , such functional redundancy should be understood in a dynamic and evolutionary context because functional similarity between duplicate pairs might be due to a lack of sufficient divergence time , or due to the long-term retention by natural selection . We next decided to delineate the evolutionary trajectory of genetic robustness resulting from gene duplication events . For this purpose , we only considered SSD pairs because they have continuously tractable divergence times , which provide us a dynamic view of genetic robustness in the course of evolution . WGD pairs , however , have all resulted from a single ancient genome duplication event ∼100 millions years ago [17] , [25] , and thus the observed backup capacity between WGDs have presumably been retained by selection . In the above analysis , we have identified 42 pairs with backup capacity among a total of 228 SSD pairs ( Figure 1 ) . We then calculated Ks ( the synonymous substitution rate per site in coding sequences ) between these buffering SSD paralogs to approximate their divergence time [10] , [23] . Strikingly , as revealed in Figure 3A , we found Ks values among these buffering pairs showed a bi-modal distribution . The broad peak on the right ( with Ks>2 , Figure 3A ) represents very ancient paralogs that still maintain their backup capacity . It is known that most paralogs have to functionally diverge to achieve long-term retention [11]; the maintained backup capacity between these ancient pairs should result from severe purifying selection stabilizing their functional redundancy ( note that the ribosomal proteins have been removed ) . The peak on the left , centered at Ks = 0 . 18 , represents very recent duplicates . These recent duplicates have not had sufficient time to functionally diverge , and these very recent paralogs among the buffering pairs may be merely due to an “evolutionary inertia” . In other words , these paralogs are in an evolutionary “transient state” since it is uncertain whether the paralogs will be eventually retained in the genome or whether they could still keep sufficient functional overlap in the course of evolution to maintain mutual backup capacity . We also examined the remaining 186 SSD pairs , whose mutual compensation had been completely lost; we found the vast majority ( 88% ) is ancient pairs with Ks>2 , confirming that the loss of backup capacity needs sufficient divergence time . However , 8 pairs among them showed unusually low Ks values ( Ks<1 ) , where 6 pairs are uncharacterized open reading frames or hypothetical proteins with unknown functions . Such a discrepancy might have resulted from rapid loss of functional overlap between these hypothetical proteins . Furthermore , the observation that the majority of the duplicate pairs ( 186 non-buffering pairs , in comparison with 42 buffering pairs ) had lost backup capacity also suggested that maintaining long-term mutual compensation between duplicates is evolutionarily difficult because most mutations affecting fitness are deleterious , and genetic redundancy would be eventually eroded by rampant mutations . Therefore in a neutral mode , it is expected that the loss of buffering strength between paralog pairs should be proportional to the amount of background mutations , scaled by divergence time . However , in the alternative model , which assumes the presence of natural selection , no correlation was expected between these two variables . By using Ks to approximate the amount of background mutations during the divergence time since duplication , we were able to consider 10 pairs with Ks≤2 among the 42 SSD pairs with backup capacity . We did not include gene pairs with Ks greater than 2 since the substitutions might have been saturated , which made it inaccurate to estimate the synonymous rates of substitutions . Interestingly , we found a tight correlation between Ks and the buffering strength between paralogs , with Pearson's correlation R = 0 . 85 , P = 1 . 8×10−3 ( Figure 3B ) , suggesting ∼72% ( R2 ) of the variation in backup strength between these duplicates could be explained by Ks . Furthermore , the proportionality of the two variables is characterized by the slope ( k = 0 . 41 ) of the regression line in Figure 3B , suggesting that a 0 . 41-fold decrease in buffering strength is accompanied with an increase of Ks by every unit . As Ks is highly correlated with Ka , which is an indicator of functional divergence in protein coding sequences , we sought to determine whether Ka was a confounding factor for the correlation between Ks and the buffering strength between the paralogs . We performed a partial correlation analysis; by controlling for the third variable Ka , we found the significant correlation between Ks and the buffering strength still remains ( R = 0 . 64 , P = 0 . 06 ) , while by controlling for Ks , Ka is no longer significantly correlated with the buffering strength ( R = 0 . 41 , P = 0 . 27 ) . This trend is also confirmed on another set of duplicate pairs with no restriction of best reciprocal BLAST matches to include more samples; again significant correlation between Ks and the buffering strength still stands when controlling for Ka ( R = 0 . 68 , P = 0 . 02 ) , while the correlation between Ka and the buffering strength is absent when controlling for Ks ( R = 0 . 32 , P = 0 . 33 ) . This analysis suggested that for duplicate pairs , their gradual loss of mutual buffering strength is scaled by the amount of background mutations ( approximated by Ks ) , not by the non-synonymous mutations ( Ka ) . To further support this neutral mode of evolution , the buffering WGD pairs serve as a negative control as backup capacity between WGD paralogs has been long-term stabilized by natural selection . In this scenario , the background mutations are expected not to correlate with the buffering strength between duplicate pairs . By considering 18 WGD pairs with mutual compensation and Ks≤2 , our partial correlation analysis ( after controlling for Ka ) consistently confirmed this prediction with R = 0 . 4 and P = 0 . 1 . Therefore , we concluded that , unless severe natural selection stabilizes genetic redundancy between paralogs for their backup capacity , mutual buffering is generally unstable between paralogs and will be eventually lost given sufficient amount of background mutations , which is proportional to divergence time between paralogs . For those ancient pairs ( Ks>2 ) that have still retained mutual buffering ( the right peak in Figure 3A ) , some of them exhibit very strong buffering capacity with buffering strength less than −0 . 7 ( Figure 3C ) . This highlighted the effects of selective pressure in stabilizing functional redundancy between these SSD paralogs . However , it is known that duplicate genes generally have to functionally diverge to achieve long-term retention in the genome [10]–[12]; therefore cells must have adopted some strategies to satisfy these conflicting requirements . One interesting example is an ancient pair STV1 and VPH1 with Ks>4 . Their coding sequences have significantly diverged ( Ka>0 . 4 ) but have maintained significant functional overlap with GO-div being much smaller than 0 . 01 , manifested by their common function in vacuolar acidification . The protein products of both genes ( Stv1p and Vph1p ) have a vacuolar-ATPase V0 domain for proton transportation across membranes; however , Stv1p is localized in Golgi and endosomes while Vph1p is localized in vacuole [26]–[29] . Therefore the observed backup capacity between these two paralogs in our study suggests that their function in normal conditions is likely to be specialized for different cellular compartments , but upon perturbations , they could be alternately used to buffer the loss of their respective paralogs . Supporting this scenario , previous experiments have shown that the moderate growth defects of Δvph1 mutant could be rescued by over-expression of Stv1p , which led to re-localization of some Stv1p to the vacuole where Vph1p is specifically localized [29] . Therefore this example represents a strategy allowing long-term retention of duplicates by diversifying their sub-cellular localization to retain the same functions , with which functional redundancy between duplicates could be maintained for their long-term mutual buffering . Lastly , we probed the general genetic properties of these ancient pairs that have maintained their long-term backup capacity . For this purpose we only considered the stabilized buffering paralogs and excluded those transient buffering pairs ( such as gene pairs around the left peak in Figure 3A ) . For WGD pairs , as their mutual compensation remains strong ( Figure 1B ) even after ∼100 million years of evolution [25] , mutual compensation between WGDs is most likely to have been stabilized by natural selection . Therefore we compared the 105 WGD duplicates with retained backup capacity against 161 WGD pairs that had lost their backup capacity . Similarly , for SSD paralogs , we only considered those paralog pairs with sufficient divergence time ( Ks>2 ) and excluded the “transient” buffering paralog pairs since their backup capacity might be eventually lost ( see Figure 2A and 2B ) . In the end , we were able to compare 32 ancient SSD backup pairs ( Ks>2 ) with the 163 non-backup pairs within the same age range ( Ks>2 ) . Compared with the non-buffering paralog pairs , the stabilized buffing paralogs have significantly overlapping functions for both WGD and SSD pairs ( 20–30% lower than non-backup pairs , Figure 4A and 4B ) , characterized by GO-div and Ka . Particularly for WGD , as they originated from a single duplication event ∼100 million years ago , the observed elevated functional redundancy between the buffering pairs indicates decelerated functional evolution between these duplicates . As the divergence in protein sequence ( nonsynonymous substitutions ) can also cause divergence in three-dimensional structures , we next examined the difference in secondary structures between these pairs . As expected , we confirmed that functional similarity between buffering pairs from WGD could be also reflected by their structural similarities , with backup pairs usually having similar secondary structural conformations ( Figure 4C ) . We also collected protein interactions from BioGrid ( see Materials and Methods ) [30] and found 62% of the backup WGD paralogs have at least one shared interacting protein while the percentage substantially decrease to 40% for non-backup WGD paralogs ( Figure 4D , P = 5 . 95×10−4 , chi-square test ) . We performed a similar analysis on the 32 SSD paralogs pairs , but did not find the excessive shared protein interactions in comparison with the matched control . It is likely due to insufficient sample size for SSD backup pairs . In addition , unlike WGD pairs , buffering between SSD pairs is typically weaker than WGD pairs ( Figure 1B ) . Therefore it is likely that the subtle buffering between SSD pairs might not be captured in our analysis of protein interactions . However , regardless of WGD and SSD , we did find the buffering pairs shared some common characteristics . With a total of 392 literature-curated protein complexes examined , we found both WGD and SSD buffering pairs were more likely to be co-clustered in the same protein complexes , with the percentage of ∼18% for the buffering pairs , compared with only ∼5–8% for the non-buffering pairs ( see Figure 4D ) . Worthy of note , previous work showed preferential co-clustering of WGD pairs in protein complexes [31]; thus the further elevated propensity of co-clustering for these buffering WGD pairs reveals a strategy of genetic buffering between duplicates: within the same complex , the backup subunit is always ready to take place of the malfunctioned ones . However , it is important to note that even in the same protein complex , the paralogs still have substantial divergence in sequences and expression profiles , which might indicate the underlying regulatory reprogramming to regulate such a backup strategy [21] . This notion can be best illustrated by one example of a buffering pair derived from SSD , Hos2p ( YGL194C ) and Rpd3p ( YNL330C ) : both proteins are involved in the histone deacetylase complex; however , Rpd3p is also a member of Rpd3L complex , Rpd3S complex , Sin3 complex and HDB complex . Therefore by differentiating their functions , the two paralogs could achieve long-term retention while their co-clustering in histone deacetylase complex enables their mutual buffering capacity . There are long-standing debates about the extent and mechanism of genetic robustness contributed by gene duplication . On one hand duplicated genes do show markedly elevated dispensability than singleton genes , which was speculated to result from mutual compensation between paralogs [5] . Alternatively , it was also proposed that such elevated dispensability of duplicates merely results from higher “duplicability” of less important ancestral genes [32] . Therefore , to determine the extent to which yeast paralogs could buffer each other , a systematic interrogation of double-knockouts of yeast paralogs is essential . In this work , we analyzed mutual buffering between yeast paralogs for ∼500 non-redundant WGD and SSD duplicate pairs , a set much larger than what was previously examined . With this largest dataset to this date , we established that merely relying on functional overlap , we are able to accurately predict buffering capacity between paralogs ( with AUC>0 . 74 ) . We further considered the functional redundancy in an evolutionary context , and found recent pairs usually maintain transient functional overlap , and the resulting mutual compensation should be mainly attributed to a lack of sufficient divergence time . However , we also uncovered an appreciable portion of duplicates with long-term retained backup capacity stabilized by selection , which is explained by their conserved functional overlap . Although both WGD and SSD paralogs could have buffering capacity , substantial difference existed between these two sets , As shown in Figure 1 , it is clear that WGD pairs are far more likely to buffer each than SSD pairs ( 39% vs 18% ) ; WGD pairs also have stronger buffering strength . We reasoned that this disparity might have resulted from differential evolutionary mode between WGD and SSD paralogs [24] . It is known that dosage balance plays an important role in WGD retention [24] , [33]; thus the retained WGD paralogs we observed here are expected to be under stronger functional constraints , which reduce the rate of functional divergence between WGD paralogs . Given these facts , their preferential mutual buffering is then anticipated . While the stoichiometric constraints on WGD pairs provide an explanation to the long-term retained redundancy for backup capacity , we for the first time presented convincing evidence that natural selection also acted on SSD pairs , which were presumably under less stoichiometric constraints compared with WGDs [16] , [24] . This finding is interesting as it revealed a different mode of evolution from that of WGD pairs . Without severe stoichiometric constraints , the duplicated copies could have the freedom to experience functional dispersal for their long-term retention , which may bring substantial genetic novelty into an existing system [13] . On the other hand , however , functional overlap between the duplicated copies and the progenitor copies was also selected and therefore stabilized during the course of evolution , which promotes cellular robustness . Though both sides are beneficial for a cell , they are conflicting in nature because most duplicates have to experience substantial functional divergence to achieve their long-term retention [10] , [11] , and genetic redundancy would be eventually eroded by rampant mutations in the process of functional divergence . However , in our study the observed natural selection on backup capacity between some ancient SSD pairs ( Figure 3A and 3C ) suggests that in some circumstances , genetic redundancy can still be long-term retained by natural selection , and that cells must have evolved effective strategies to balance the conflicting needs , promoting genetic novelty and systems robustness simultaneously . This point is best demonstrated by the example of STV1 and VPH1 , which retained their long-term backup capacity by specializing their overlapping function in different cellular organelles . Based the results described in this paper , we propose that genetic redundancy essentially comes from functional overlap between paralogs [7] . This notion is consistently supported in our study by employing GO-div to specifically quantify the degree of functional overlap between paralogs; this approach is not affected by other differentiated functions that are not shared by the paralogs . In contrast , as shown in previous work , when using expression profiles or genetic interaction profiles to estimate functional divergence on a global scale ( in comparison with localized function overlap ) , strong association between functional similarity and backup capacity is not always observed [14] , [31] . Despite the prevalence of mutual compensation between paralogs , our analysis revealed that a large number of duplicate pairs had lost their backup capacity . Indeed , we showed that the erosion of backup capacity between paralogs is essentially a neutral process , with the buffering strength correlated with the amount of background mutations , and proportional to divergence time . Consequently unless evolutionarily stabilized , mutual compensation between most paralogs is an evolutionarily transient state , and cannot substantially contribute to the cellular robustness on a large evolutionary scale . Therefore , beyond genetic redundancy between duplicates , future research is needed to explore other mechanisms contributing to the global robustness in a cell . We compiled yeast duplicates from Guan et al . [16] , where the authors used an improved algorithm to detect paralogs based on Kellis et al [17] , [18] . In our study , we studied 495 WGD and 667 SSD paralogs with sequence identity ≥20% as they represent the most confidently assigned paralogs . The SSD pairs were derived from the best reciprocal matches , with one gene being involved in only one pair . Among the WGD and SSD paralogs , we removed the pairs with at least one copy annotated to be ribosomal proteins , and the annotation was based on gene ontology ( GO , as of Jan 2009 ) . We mapped the paralogs onto the newly released yeast genetic interaction data generated by high-density synthetic genetic arrays ( SGA ) [15] , and retained a total of 328 pairs that have quantitative genetic interactions . We further complemented this list by quantitating additional 166 pairs with the same platform as Costanzo et al . [15] . In total , we studied 494 non-ribosomal paralogs in this study , in which 266 were WGD paralogs and 228 were SSD paralogs . The scoring scheme for genetic interactions is detailed in Costanzo et al . [15]; the significant negative genetic interactions ( interaction score is smaller than 0 and P-val is less than 0 . 05 ) between paralogs indicate their mutual backup capacity , implicating that double deletion of a pair induces much sicker growth defect that expected from single-deletions . Therefore the more negative the scores are , the stronger backup capacity is expected . To determine whether the duplicate pairs have an excess of backup capacity , we generated an ensemble of 1 , 000 randomized controls . For each control group , we randomly sampled 1 , 000 gene pairs hat have genetic interaction assayed in Costanzo et al . Then the percentage of random pairs having negative genetic interaction can be determined for each control group , and the distribution of the percentages can be estimated from the 1 , 000 randomized controls . To determine whether buffering between paralogs was stronger than random pairs , for each control , we calculated average scores for the pairs maintaining negative interactions , and the distribution of the average scores can be estimated from the 1 , 000 randomized controls . Sequence divergence between paralogs is estimated by synonymous ( Ks ) and non-synonymous ( Ka ) substitutions per site by aligning the coding sequences of two genes . We downloaded yeast gene sequences from SGD ( Saccharomyces Genome Database ) , and implemented PAML to calculate Ks and Ka [34] . Functional similarities between two genes can be measured by their semantic similarities in the Gene Ontology ( GO ) hierarchy [22] , [35]; this approach had been successfully used to benchmark data from high-throughput experiments [36] . In this study , we adopted this approach to quantify functional overlap between duplicate genes . We considered all the GO terms in the hierarchy of Biological Process ( BP ) , and these terms represent a corpus , with which each gene is annotated . We did not consider terms in the hierarchies of Cellular Component ( CC ) and Molecular Function ( MF ) because CC is not a direct indicator of functional similarity . MF depicts gene activities at the molecular level , and one or more assemblies of MF define a BP term [37] . Therefore considering BP terms implicitly covers MF annotations . In addition , recent work showed using BP reaches the best performance than using CC and MF terms [22] . Considering annotation quality , we excluded all the electronic annotations ( with the code of IEA ) . For a duplicate pair , as shown in Figure S1A , copy A is annotated with m terms and copy B is annotated with n terms , so GO-div between copy A and B is defined by: ( 1 ) where T ( i , j ) is the semantic similarity between term i and j . Calculation of the term-term semantic similarity T in a GO hierarchy is demonstrated in Figure S1B by following the protocol described in [22] and [36] . The rationale is that two terms are more similar if they share a very specific ancestral term , and the specificity of a term x is defined by the probability , p ( x ) , of randomly sampling the term x and all its ( recursive ) children terms from the BP term collection [35] . With this , the term-term similarity T ( m , n ) , using the term Am for gene copy A and the term Bn for gene copy B as an example ( as illustrated in Figure S1A ) , is defined by: ( 2 ) S ( m , n ) is the set of parent terms shared by m and n ( see Figure S1B ) , and the numerator of Eq . [2] essentially is to calculate the information content of the most specific parental term ( s ) shared by m and n ( see Figure S1B ) . The denominator is a normalization constant to scale the score between 0 and 1 . Thus for two terms , if both terms are specific ( deep in the GO tree ) while their common ancestor term is also very specific , then the two terms receive high score T , indicating great semantic similarity between the two terms . At an extreme , when two terms are only overlapped at the root term ( Biological Process ) , then p ( x ) = 1 , giving T = 0 . Collectively , GO-div computes all possible term-term similarity for a duplicate pair ( Figure S1A ) , and scores the best matched GO term pair ( s ) . In this regard , compared with other metrics for characterizing overall functional similarity between duplicates , GO-div is more suitable to quantify “functional overlap” between paralogs . We collected expression profiles for each yeast genes across 549 physiological conditions [38]–[40] . Expression divergence between paralogs is then defined as 1 minus correlation coefficients of expression profiles between sister paralogs . We trained a SVM using RBF kernel with Gaussian variance and penalty for soft margin . We predicted protein secondary structures using PSIPRED [41] , which achieves prediction accuracy >80% . For the predicted structures , we compared the structural resemblance using a newly introduced approach [42] , where structural characteristics are encoded in a feature vector comprised of transition probabilities among the basic structural building blocks including α helices , β strands and coils . Information discrepancy between two feature vectors of sister paralogs was calculated to quantify dissimilarity in secondary structures , D , and greater D indicates more dissimilar secondary structures . This approach has been shown to be robust and objective to classify protein structures [42] . We downloaded protein-protein interactions from BioGrid ( version 2 . 0 . 52 ) [30] , and retained protein interactions reported from two-hybrid assays and affinity capture-mass spectrometry . The derived protein interaction network covers 4 , 873 genes mediate 33 , 949 protein interactions . Protein complexes were curated by merging annotations from SGD ( Saccharomyces Genome Database ) , GO ( Gene Onotology ) and MIPS ( The Munich Information Center for Protein Sequences ) .	Eukaryotic cells show remarkable robustness against external perturbations , which has been thought to be attributed , at least in part , to the extensive gene duplication events in eukaryotic genomes . By duplication , genes are likely to gain redundant copies for backup purposes , however , this notion contradicts the population genetic theory that genetic redundancy is evolutionarily unstable . In this study , we used yeast as a model organism to delineate the evolutionary trajectory of genetic robustness by gene duplication , utilizing the comprehensively characterized synthetic genetic interaction data in the yeast genome . We showed that the evolution of genetic robustness by duplication follows a neutral mode , with the loss of backup capacity proportional to the divergence time . However , natural selection was also acting on a few pairs to maintain their long-term backup capacity; and these pairs are slowly evolving , are co-clustered in the same protein complexes , and tend to interact with the similar partners . This study unravels the general principles underlying the evolution of the cellular robustness arising from genetic redundancy .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "computational", "biology/molecular", "genetics", "computational", "biology/systems", "biology", "evolutionary", "biology/bioinformatics", "evolutionary", "biology/genomics" ]	2010	The Cellular Robustness by Genetic Redundancy in Budding Yeast
Neuroimaging has identified many correlates of emotion but has not yet yielded brain representations predictive of the intensity of emotional experiences in individuals . We used machine learning to identify a sensitive and specific signature of emotional responses to aversive images . This signature predicted the intensity of negative emotion in individual participants in cross validation ( n =121 ) and test ( n = 61 ) samples ( high–low emotion = 93 . 5% accuracy ) . It was unresponsive to physical pain ( emotion–pain = 92% discriminative accuracy ) , demonstrating that it is not a representation of generalized arousal or salience . The signature was comprised of mesoscale patterns spanning multiple cortical and subcortical systems , with no single system necessary or sufficient for predicting experience . Furthermore , it was not reducible to activity in traditional “emotion-related” regions ( e . g . , amygdala , insula ) or resting-state networks ( e . g . , “salience , ” “default mode” ) . Overall , this work identifies differentiable neural components of negative emotion and pain , providing a basis for new , brain-based taxonomies of affective processes . Emotions are a class of psychological states comprised of physiological responses , expressive behavior , and subjective experiences that are central to our daily lives and to multiple forms of psychopathology [1] and chronic medical diseases [2] . Emotional information organizes physiological , cognitive , and motor systems into adaptive [3] , organism-wide responses to events and situations relevant for survival and well-being [4–6] . These responses allow us to pursue resources and avoid harm [7] , translate cognitive goals into motivated behavior [8] , and navigate the social world [9 , 10] . Conversely , emotional dysregulation is at the heart of many brain- and body-related disorders ( e . g . , mood , anxiety , personality , cardiovascular , and substance use disorders ) and likely cuts across traditional diagnostic boundaries [11] . Thus , understanding the neurobiological mechanisms that generate and mitigate negative emotional experience is paramount to understanding both human flourishing and dysfunction . The importance of understanding the “emotional brain” has motivated hundreds of neuroimaging studies in healthy humans [12 , 13] and those suffering from psychopathology [14–16] . The promise of these studies for basic research is that they will permit a brain-based taxonomy of emotional processes , avoiding the sole reliance on psychological categories [17 , 18] , while the hope for clinical development is to provide transdiagnostic markers for psychopathology that can identify functional brain dysregulation [19] and physical health risk [2 , 20] , predict treatment response [21 , 22] , and guide new , brain-based treatments [23 , 24] . In spite of this promise , fundamental requirements must be met before neuroimaging findings can be considered brain representations of emotion that are useful for translational purposes [25] . Previous work has identified many brain correlates of emotional versus nonemotional stimuli [12] and physiological responses [26 , 27] but has yet to uncover brain signatures diagnostic of an individual’s emotional experience . For example , the amygdala , dorsal anterior cingulate ( dACC ) , anterior insula ( aINS ) , and other regions reliably respond to aversive stimuli [28] , and functional alterations in these regions are considered prominent features of anxiety disorders [14 , 29] . However , activation in these regions does not imply an emotional experience . Amygdala activation can occur in the absence of emotional experience [30] and does not appear to be involved in all aversive experiences [31] . In addition , the dACC and aINS are among the most frequently activated regions in the brain across all types of emotional and nonemotional states [28] and have recently been conceptualized as network “hubs” that may be integrating cognitive , emotional , and motivational information [32 , 33] . One factor that contributes to this limitation is that the vast majority of studies focus on comparing types of stimuli [12] , e . g . , “negative” versus “neutral” images , rather than finer grained differences in reported experience [34] . While these emotion-related comparisons are assumed to reflect “affective processing , ” confounds with attention , salience , and other processes may render many findings superfluous to emotional experience . Thus , there is a pressing need for neural signatures that are optimized to predict emotional experiences and functional outcomes . These indicators should: ( 1 ) specify a precise set of brain voxels that can be tested in new individuals and prospectively applied to new samples and ( 2 ) be sensitive and specific to a class of affective experiences ( e . g . , negative emotion and not other states such as attention or arousal ) [35] . Machine learning provides a new toolbox of algorithms suited for developing sensitive and specific signatures of psychological processes [36–39] , particularly when those signatures involve measures across multiple neural systems , as is likely to be the case with emotional experience [12 , 18 , 40] . Standard neuroimaging methods generally preclude estimation and optimization of the strength of the brain experience correspondence [28 , 41–43] , but cross validated machine learning analyses can identify whether brain effects are of sufficient magnitude ( e . g . , sensitive enough ) and specific enough to have translational utility . These techniques have recently shown great promise in identifying patterns that discriminate among types of affective experiences from brain [35 , 44–46] and physiology [47] , discriminating patient from control groups [19 , 48] , and predicting treatment response [49] . Here , we use machine learning in a large sample ( n = 183 ) to identify the brain systems that predict the intensity of negative affective experiences elicited by viewing images from the International Affective Picture System ( IAPS ) [50] , which is among the most robust methods of eliciting brief affective experiences ( d = 0 . 81 ) [51] . In spite of the widespread use of IAPS images in basic and clinical research ( e . g . , it is the primary affective task in the human connectome project [52] ) , the brain mechanisms that underlie the genesis of the negative experiences they evoke have not been clearly identified . In addition , it is unclear ( a ) whether it is possible to identify a pattern that strongly predicts emotional experience prospectively in out-of-sample individuals , ( b ) which brain systems are involved ( cortical , subcortical , or both ) , and ( c ) whether brain activity that tracks negative affect is specific for negative affect , or whether it codes for “salience , ” arousal , or more general features of stimulus processing . Answers to all of these questions are critical for continued progress in both basic affective and clinical sciences . We address each of these questions by developing a multivariate pattern that predicts negative emotion and assess its sensitivity and specificity relative to pain—another type of arousing , salient , negative experience . Finally , to examine the distributed versus localized nature of the signature , we examined the subsystems necessary and sufficient for accurately predicting negative emotional experience . We used Least Absolute Shrinkage and Selection Operator and Principle Components Regression ( LASSO-PCR ) [35 , 53] to identify a distributed Picture Induced Negative Emotion Signature ( PINES ) that monotonically increased with increasing affective ratings in leave-one-subject-out cross validated analyses ( n = 121 ) . To apply the model to data from individual test subjects in both cross validation ( n = 121 ) and separate hold-out test datasets ( n = 61 ) , we calculated the pattern response—the dot product of the PINES weight map and the test image—for individual subjects’ activation maps for each of 5 levels of reported negative emotion ( see Fig 1 ) . The resulting continuous values reflect the predicted intensity of negative emotion for a given activation map . We used these values to classify which of two conditions elicited a stronger negative emotion for an individual ( a “forced-choice” test ) [35] , providing accuracy estimates ( Fig 1E ) . We also used similar classification tests , described below , to evaluate the sensitivity and specificity of PINES responses to negative emotion versus pain . We focus primarily on results for the test sample , as it was completely independent of all model-training procedures and provides the strongest evidence for generalizability [54] . The PINES accurately predicted ratings of negative emotional experience in both cross validation and hold-out test datasets ( Fig 2 ) . For individual participants in the cross validation sample , the average root mean squared error ( RMSE ) was 1 . 23 ± 0 . 06 ( standard error; SE ) rating units , and the average within-subject correlation between predicted and actual ratings was r = 0 . 85 ± 0 . 02 ) . Accuracy was comparable in the test sample ( RMSE = 0 . 99 ± 0 . 07 , r = 0 . 92 ± 0 . 01 ) . The PINES accurately classified highly aversive ( rating 5 ) versus nonaversive ( rating 1 ) pictures with 100% forced-choice accuracy in both cross validation and test samples ( Fig 2B ) . Classification accuracy was also high in both the highly aversive range ( rating of 5 versus 3: forced-choice = 91%; test sample ) and the moderately aversive range ( rating of 3 versus 1: 100%; test sample ) ( See S1 Table ) . We also assessed single-interval classification based on a single image rather than a relative comparison ( Table 1 ) , which were only slightly less accurate ( Table 1 ) . Comparisons with Support Vector Regression ( SVR ) , another popular algorithm , indicate that these results appear to be robust to the choice of algorithm and , to a large extent , the amount of data used in the training procedure ( see S1 Methods ) . The PINES pattern included reliable predictive weights across a number of cortical and subcortical regions ( Fig 2A ) . Positive weights ( greater activity predicts more negative emotion ) were found in many regions typically associated with negative emotion [12 , 40] , including the amygdala , periaqueductal gray ( PAG ) , aINS , dorsomedial prefrontal cortex ( dmPFC ) , ventral occipital cortex , presupplementary motor area ( preSMA ) , ventromedial temporal lobe ( mTL ) , and posterior cingulate cortex ( PCC ) . Negative weights were found in the bilateral parahippocampal gyrus , right superior temporal gyrus , left temporal parietal junction ( TPJ ) , right caudate , and occipital and somatomotor cortices . These regions likely comprise multiple functional systems , as we describe in more detail below . Though the PINES comprises nonzero predictive weights across the brain ( see S1 Fig ) , supplementary analyses indicated that a sparse pattern thresholded at p < . 001 , as shown in Fig 2 ( 1 . 6% of in-brain voxels ) , was sufficient to predict emotional experience with comparable sensitivity to the full model ( see S1 Methods and S5 Fig ) . Affect systems may be organized by valence so that a brain signature for negative affect may be found across stimulus modalities and contexts , or in a modality-specific manner , such that there is not one “negative affect system” but many . Testing these hypotheses requires comparing multiple types of negative affect across modalities . Here , we assessed the generalizability and specificity of the PINES response across IAPS pictures and somatic pain , which is a negative , arousing experience arising from a different modality . We employed two types of analyses to examine the PINES specificity . First , we compared the spatial topography of the PINES to another pattern map , the Neurologic Pain Signature ( NPS ) , which shows high sensitivity and specificity to somatic pain across multiple studies [35] . The PINES and NPS maps were almost completely uncorrelated ( robust ranked spatial correlation , ρ^ = −0 . 01; Fig 4 ) . Several regions showed positive weights in both maps , including the anterior cingulate ( ACC ) , insula , and amygdala . As shown in Fig 5C , however , the weight patterns within these regions were also uncorrelated ( bilateral ACC , ρ^ = 0 . 04 , insula , ρ^ = −0 . 05 ) , though weights in the amygdala were modestly correlated ( ρ^ = 0 . 21 ) . Second , we assessed the specificity of the pattern responses in the test IAPS ( n = 61 ) and thermal pain ( n = 28 ) [56] datasets . The PINES accurately predicted negative affect in the IAPS dataset ( n = 61 ) but showed no response to increasing pain intensity in the pain dataset ( Fig 4 ) . Conversely , the NPS responded robustly to increasing pain but showed no response to increasing negative affect in the IAPS dataset . To further assess sensitivity and specificity , we examined how well responses in each pattern could discriminate ( a ) high pain versus high negative affect , ( b ) high versus low pain , and ( c ) high versus low negative affect ( Table 1 ) . Because this involves comparing responses from two separate , imbalanced test sets ( n = 61 versus n = 28 ) , the analyses described below employ single interval classification , in which individual images are tested for suprathreshold responses independently ( as compared to relative within-subject differences in forced-choice classification ) . The threshold was determined by finding the point that minimized signal detection response bias ( see Methods for details ) , and we report balanced emotion classification accuracy ( chance = 50% ) , sensitivity , and specificity ( See S2 Table for equivalent forced-choice analyses ) . Another question is whether the precise pattern of activity specified in the PINES uniquely captures negative affect , or whether regions and networks previously used in the literature are sufficient . In order to fully appreciate the sensitivity and specificity of the PINES , it is necessary to compare it to the standard univariate approach , which typically examines average activation within ROIs compared to baseline activity . In this analysis , we examined the average response to emotion and pain stimuli within anatomical ROIs and canonical networks defined in large-scale resting-state studies [57] . Defining a brain pattern sensitive and specific to a type of negative emotion is a critical first step towards developing meaningful models of brain representations of emotion . Here , the development of the PINES affords the opportunity to characterize the basis of this pattern representation within and across brain networks . Constructionist theories of emotion [12 , 18] predict that negative affect is created by interactions among discrete subnetworks that span multiple brain systems , whereas more traditional modular views predict that one system may be sufficient . We tested whether the PINES might be composed of multiple distinct subnetworks and whether responses in multiple subnetworks are necessary for predicting emotional responses . If so , the negative affect captured by the PINES might be considered a truly multisystem distributed process . For this analysis , we calculated pattern responses within each of the largest regions in the PINES ( p < . 001 , k = 10 voxels; see S1 Methods ) for every individual trial within each participant and used a robust clustering algorithm to group the PINES regions into separate networks based on similar patterns of trial-by-trial covariation ( see Methods ) . The best solution contained nine separate clusters , which provides a descriptive characterization of the subnetworks that comprise the PINES ( Fig 6 , S3 Table ) that is broadly consistent with constructionist accounts of emotion [12] and previous meta-analyses of emotion-related networks [17] . These subnetworks included ( a ) two networks encompassing different parts of the visual cortex ( e . g . , lateral occipital cortex [LOC] and occipital pole ) consistent with the visual modality of the stimuli , ( b ) a left amygdala-right aINS-right putamen network , which has been implicated in multiple forms of arousal and salience , ( c ) a network that includes bilateral posterior parahippocampi and the precuneus , which are broadly involved in memory and other forms of contextual processing , and ( d ) a network that includes parts of the dmPFC and PCC that are likely involved in social cognition but are distinct from more executive processes [64 , 65] . An additional network that includes the right somatosensory cortex and contralateral cerebellum may be involved in preparing for the rating action but may also play a more fundamental role in the emotion generation process [66] . These results have theoretical implications for the neurobiology of emotion in terms of both the diversity of processes underlying affective experiences and how they are represented in the brain . Emotions are often defined as a composite of multiple intrinsically inter-related processes ( e . g . , autonomic arousal , expressive behavior , action tendencies , interoception , and conscious experiences ) . Theories differ widely on how these processes combine to give rise to emotional experience [1] , but most major theories suggest that cognitive , sensory , motor , motivational , and interoceptive processes are critical ingredients of emotional experience . For example , appraisal theories view emotion as a dynamically unfolding process and emphasize the role of appraisals [8 , 68 , 69] , embodied affect theories emphasize interoceptive and somatomotor representations [70] , and constructionist theories view emotions as being constructed from all of these component processes [12 , 18] . In spite of this richness , since MacLean [71] , theories of the emotional brain have treated emotion as a singular faculty that is localizable to a specific system . Often , this view has translated into “structure-centric” theories of emotional experience; e . g . , the amygdala is critical for fear [72] , the ACC for pain affect [73] , and the insula for disgust [74] . In other cases , this view translates into circumscribed pathways or networks for “core affect” [17] and emotional awareness [75] . It remains unclear how far the structure-centric view can take us in understanding the brain bases of emotional experience . The regions most strongly identified with emotion are also intimately involved in a wide array of cognitive functions such as attention , error monitoring , associative learning , and executive control [33] . Recent connectivity [63] and functional diversity analyses [32] suggest that these regions are not solely processing affective signals but rather represent functional “hubs” for integrating many types of information . As the limitations of the structure-centric view are increasingly widely recognized [12 , 33] , researchers have moved towards the identification of intrinsically connected networks conserved both at rest and during active tasks [76] . These networks have been labeled with process-general names including the “salience network” [63] , “default mode” network [77] , and others , and a modern incarnation of the “emotional brain” theory suggests that the basis of emotional experience is encapsulated in one or a few of these networks such as the “limbic” network named after MacLean’s original formulation . Our results corroborate the view that structure-centric—and even network-centric—models of emotion are limited and provide an alternative model for the brain representation of emotional experience . In this study , we targeted the conscious experience component , which is the defining feature of subjective distress and suffering . None of the anatomical regions identified in previous literature ( e . g . , amygdala , ACC , insula ) predicted the intensity of emotional experience or discriminated emotion from pain in this study . This suggests that the effects identified in previous work using traditional statistical parametric mapping approaches are small and unlikely to serve as effective signatures of the type or magnitude of an emotional experience in an individual person . Furthermore , activity in predefined networks was insufficient to capture negative emotion ratings , demonstrating that the pattern we identified using targeted machine-learning analysis is not reducible to these more process-general networks . The fact that networks and regions defined a priori , even from very large resting-state samples [57] , were insufficient to capture emotional experience here has broad implications for the study of emotion and attempts to identify biomarkers for mental health disorders going forward [21 , 25 , 49] . Finally , our clustering analysis of the PINES map indicated that multiple , separable subnetworks distributed widely throughout the brain made independent contributions to predicting emotional experience . Importantly , no single subnetwork appeared to be necessary or sufficient in characterizing the emotional experience , as the accuracy in predicting the magnitude or type of experience did not significantly decrease when any given network was omitted . This pattern is consistent with both appraisal [68 , 69] and constructionist theories of emotion [12 , 78] , which posit that emotional experiences result from interactions between core affect , sensory , memory , motor , and cognitive systems [40] . Overall , these results provide an important step towards identifying emotion-related patterns that can serve as indicators for components of emotional experience . Such signatures can be used as intermediate phenotypes for genetic or risk-stratification studies , and they may provide objective neurobiological measures that can supplement self-report . The identification of intermediate brain-based phenotypes is critical , as self-reported emotion can be affected by many independent processes [8 , 18 , 68]—e . g . , core experience , self-reflection , decision-making heuristics , and communicative intentions—which have different implications for understanding what exactly treatments that modulate emotion are measuring and which processes are affected by interventions . We close with several key points and future directions . Importantly , the PINES is not necessarily a biomarker of negative emotion in general . We have demonstrated that it is a signature for the type of affect induced by aversive IAPS images , but its transferability to other emotional states ( e . g . , emotion induced by recall , rejection , positive emotion , or stress ) remains to be tested . Such tests are a long-term program of future research that must span many studies and papers . We still know very little about the underlying structure of affect and which types of emotional responses can be cross predicted by the same brain markers . It is possible that the PINES captures some types of negative emotion and not others , and findings to this effect will help us move beyond the categories proscribed in our language to develop a more nuanced , brain-based view of affective processes [7 , 17] . In addition , testing the specificity and transfer of the PINES across many different kinds of affect is a key to developing more robust and specific markers . The PINES can undoubtedly be improved . For example , with further development and testing , it may be differentiated into markers for more specific types of emotional experiences ( e . g . , emotion categories like fear , disgust , etc . or canonical affect-inducing appraisals ) . In addition to types of affect , the PINES can be tested for responses across patient groups ( e . g . , schizophrenia , depression , or anxiety ) and treatments thought to affect emotion ( e . g . , self-regulation , drug treatment , psychotherapy , etc . ) . This study provides a foundation and a benchmark for such future developments .	Emotions are an important aspect of human experience and behavior; yet , we do not have a clear understanding of how they are processed in the brain . We have identified a neural signature of negative emotion—a neural activation pattern distributed across the brain that accurately predicts how negative a person will feel after viewing an aversive image . This pattern encompasses multiple brain subnetworks in the cortex and subcortex . This neural activation pattern dramatically outperforms other brain indicators of emotion based on activation in individual regions ( e . g . , amygdala , insula , and anterior cingulate ) as well as networks of regions ( e . g . , limbic and “salience” networks ) . In addition , no single subnetwork is necessary or sufficient for accurately determining the intensity and type of affective response . Finally , this pattern appears to be specific to picture-induced negative affect , as it did not respond to at least one other aversive experience: painful heat . Together , these results provide a neurophysiological marker for feelings induced by a widely used probe of negative affect and suggest that brain imaging has the potential to accurately uncover how someone is feeling based purely on measures of brain activity .	[ "Abstract", "Introduction", "Results", "Discussion" ]	[]	2015	A Sensitive and Specific Neural Signature for Picture-Induced Negative Affect
Animal-vegetal ( AV ) polarity of most vertebrate eggs is established during early oogenesis through the formation and disassembly of the Balbiani Body ( Bb ) . The Bb is a structure conserved from insects to humans that appears as a large granule , similar to a mRNP granule composed of mRNA and proteins , that in addition contains mitochondria , ER and Golgi . The components of the Bb , which have amyloid-like properties , include germ cell and axis determinants of the embryo that are anchored to the vegetal cortex upon Bb disassembly . Our lab discovered in zebrafish the only gene known to function in Bb disassembly , microtubule-actin crosslinking factor 1a ( macf1a ) . Macf1 is a conserved , giant multi-domain cytoskeletal linker protein that can interact with microtubules ( MTs ) , actin filaments ( AF ) , and intermediate filaments ( IF ) . In macf1a mutant oocytes the Bb fails to dissociate , the nucleus is acentric , and AV polarity of the oocyte and egg fails to form . The cytoskeleton-dependent mechanism by which Macf1a regulates Bb mRNP granule dissociation was unknown . We found that disruption of AFs phenocopies the macf1a mutant phenotype , while MT disruption does not . We determined that cytokeratins ( CK ) , a type of IF , are enriched in the Bb . We found that Macf1a localizes to the Bb , indicating a direct function in regulating its dissociation . We thus tested if Macf1a functions via its actin binding domain ( ABD ) and plectin repeat domain ( PRD ) to integrate cortical actin and Bb CK , respectively , to mediate Bb dissociation at the oocyte cortex . We developed a CRISPR/Cas9 approach to delete the exons encoding these domains from the macf1a endogenous locus , while maintaining the open reading frame . Our analysis shows that Macf1a functions via its ABD to mediate Bb granule dissociation and nuclear positioning , while the PRD is dispensable . We propose that Macf1a does not function via its canonical mechanism of linking two cytoskeletal systems together in dissociating the Bb . Instead our results suggest that Macf1a functions by linking one cytoskeletal system , cortical actin , to another structure , the Bb , where Macf1a is localized . Through this novel linking process , it dissociates the Bb at the oocyte cortex , thus specifying the AV axis of the oocyte and future egg . To our knowledge , this is also the first study to use genome editing to unravel the module-dependent function of a cytoskeletal linker . Cellular polarity organizes the intracellular space into cytoplasmic domains that mediate cellular functions across diverse cell types . For instance , oocytes are polarized in many species with the formation of the Balbiani Body ( Bb ) ( also called the mitochondrial cloud in Xenopus ) adjacent to the nucleus . The Bb is a large granule conserved from insects to mammals that tightly aggregates RNAs , proteins , ER , and mitochondria . The Bb granule is a non-membrane bound compartment that isolates its content from the cytoplasm . There is evidence in zebrafish and Xenopus stage I oocytes that the Bb forms through the assembly of Bucky ball amyloid-like fibers that capture Bb components and give rise to the large Bb granule [1–3] . Later , by the end of stage I ( stage II in Xenopus ) , the Bb dissociates at the oocyte cortex and its components become docked at the now defined oocyte vegetal pole . This establishes the animal-vegetal ( AV ) axis of the oocyte and future egg , which in turn defines the anterior-posterior axis of the embryo [4] . Hence , elucidating the mechanism of Bb disassembly is relevant to understanding two conserved and linked processes; the establishment of cell polarity and the disassembly of an amyloid-like structure such as the large Bb granule . The two proteins known to be necessary for Bb function in zebrafish , Bucky ball ( Buc ) and Microtubule-actin crosslinking factor 1a ( Macf1a ) , were discovered via a zebrafish maternal-effect mutant screen in our lab [2 , 5 , 6] . In eggs from buc or macf1a mutant females , the cytoplasm fails to segregate to form the blastodisc at the animal pole , and instead is radially distributed around the yolk [6] . Lacking AV polarity , development aborts shortly thereafter [6] . During early stage I of zebrafish oogenesis , Bucky ball is required for Bb formation , and the Xenopus Buc ortholog , Xvelo , is the most abundant protein in the frog Bb [1 , 3] . buc mutant oocytes lack a Bb and RNAs normally carried within the Bb are dispersed throughout the cytoplasm and never localize to the vegetal pole [2 , 3 , 7] . Xvelo self-aggregates in vitro and in vivo forming a matrix of amyloid-like fibers that may entrap mitochondria to create the Bb [1 , 8] . These amyloid-like aggregates are very stable and difficult to disrupt [1] . However , the Bb naturally disassembles by the end of stage I of oogenesis . Macf1a is the only known functional player in this process . In macf1a mutant oocytes the Bb forms and accumulates RNA normally , however , the Bb becomes enlarged and never disassembles [5] . All macf1a mutant oocytes also develop an acentric nucleus phenotype beginning at mid-stage 1 of oogenesis [5] . It is unknown if this phenotype is functionally linked to AV polarity and Bb disassembly . Understanding the role of Macf1a in Bb dissociation could provide insight into the dissociation of similar amyloid-like aggregations in pathological conditions . Macf1 is a conserved , giant cytolinker that can interact with all cytoskeleton components: microtubules ( MTs ) , actin filaments ( AF ) and intermediate filaments ( IF ) . Macf1 is modular in that distinct domains can interact with each of these cytoskeleton components , and these distinct domains are expressed in multiple isoforms [9–11] . Macf1 functions in a variety of tissues and processes in different species , all integrating the cytoskeleton in cellular functions . In mice , Macf1 is essential for early development and mutant embryos die during gastrulation [12] . In mammalian cells , Macf1 acts as a plus ( + ) tip MT-binding protein and mediates MT and actin integration at the cell cortex [10 , 13] . Keratinocytes require Macf1 for cell migration , where Macf1 integrates MTs and actin cables to maintain focal adhesions [13 , 14] . Similarly , using the gene trap line Gt ( macf1a-citrine ) ct68a in zebrafish , Antonellis et al [14] showed that Macf1a-Citrine fusion protein likely functions in connecting MTs to actin in hair cells and participates in apical-basal polarity . In invertebrates , the macf1 orthologs , shot ( fly ) and VAB-10 ( worm ) , have diverse functions in axon targeting , nuclear migration , epidermal attachment , and germ cell maintenance [15–23] . How Macf1a interacts with the cytoskeleton to disassemble the Bb and establish oocyte polarity remains undetermined [8 , 24 , 25] . Technical constraints have restricted the study of macf1a , since it is a large gene spanning ~300 kb of the zebrafish genome , with the longest predicted ORF of ~25 kb ( NCBI: XP_001920094 . 1 ) . Using transgenes for such large transcripts is difficult and subject to variable expression due to heterogeneous insertion sites . Thus , to unambiguously determine how Macf1a acts in AV polarity establishment , we targeted the macf1a endogenous gene to address Macf1a domain function in its normal physiological context . Here we investigated the localization and function of Macf1a and the cytoskeleton in regulating the Bb and oocyte nucleus positioning . We found that Macf1a and cytokeratins ( a type of IF ) localize to the Bb , and that Macf1a associates with actin at the cortex upon Bb disassembly . Disruption of cortical actin in late stage I oocytes causes detachment of Bb components from the cortex , partially phenocopying the macf1a mutant . In contrast , disruption of MTs does not affect the Bb or nuclear positioning . Based on these results , we tested the hypothesis that Macf1a functions via its ABD and/or PRD ( IF binding domain ) to regulate Bb disassembly at the cortex . To test this , we used CRISPR/Cas9 genome editing technology to delete these domains by targeting the macf1a endogenous gene . This method harnesses the modular structure of the Macf1 cytoskeleton-binding domains to specifically interrogate single Macf1a domain functions in Bb disassembly and nucleus positioning . Our results reveal that the Macf1a ABD is essential for Bb disassembly and correct nuclear positioning . Surprisingly , we found that the Macf1a PRD domain is dispensable for both of these processes . To our knowledge , this is the first study to use genome editing to precisely target the module-dependent function of a cytoskeletal linker . The Bb progresses during stage IB of oogenesis from its initial location adjacent to the nucleus , to the oocyte cortex by late stage IB where it disassembles . To characterize this process in macf1a mutants , we followed protein and RNA markers of the Bb in mutant and wild type ( WT ) stage IB oocytes ranging from ~50 to 140 μm in diameter [26] . The Bb markers we selected were Buc , the only protein known to be required for Bb formation in vertebrates , and dazl , an mRNA component of the germ plasm ( Fig 1B–1D ) . To examine dazl RNA localization , we used the highly sensitive hybridization chain reaction ( HCR ) method that allows detection of low RNA concentrations with minimal background [27] . We observed that Buc is recruited to the Bb in macf1a mutant oocytes ( Fig 1D’ and 1D” ) , similar to dazl and other previously examined Bb components ( Fig 1B ) [5 , 28] . Unlike in WT , in macf1a mutants Buc and dazl fail to localize to the vegetal cortex and instead remain in a persistent and enlarged Bb ( Fig 1B–1D” ) . We used the Buc immunofluorescence localization pattern to quantify the progression of Bb disassembly during stage I of oogenesis in WT and macf1a mutant oocytes ( Fig 1C and 1D” ) . In late stage IB WT oocytes , as the Bb disassembles at the cortex , Buc dissociates from the Bb and localizes to the vegetal cortex ( Fig 1C and 1C” ) . To quantitatively evaluate Bb disassembly , we identified the Bb by Buc immunostaining , then compared the signal intensity of Buc within the Bb ( Fig 1E ) versus outside the Bb ( Fig 1E’ ) during disassembly ( see methods ) . We measured the total Buc immunofluorescence area versus Buc localized to the Bb to estimate a Bb disassembly ratio ( Bb Buc/Buc total ) , along with measuring the oocyte diameter throughout stage I ( Fig 1F ) . In 60 micron ( μm ) early stage IB oocytes when the mature , compact Bb has formed ( Fig 1C ) ( Elkouby et al , 2016 ) , the Buc disassembly ratio is ~1 . This ratio decreases to ~0 towards the end of stage I when the Bb disassembles and Buc is unloaded at the cortex ( Fig 1F ) . Using this method , we found that the Bb begins to dissociate at the vegetal pole in WT oocytes that were 95 to 110 μm in diameter , and reached the midway point ( 0 . 6 to 0 . 4 Buc disassembly ratio ) in oocytes 115 to 125 μm in diameter ( Fig 1F ) . In the largest WT stage IB oocytes ( 135 to 160 μm in diameter ) the disassembly ratio was ~0 . 15 . In contrast , the Bb disassembly ratio in macf1ap6cv and macf1sa12708 mutant oocytes of a similar diameter range did not appreciably decrease below 1 ( Fig 1F , red squares and green triangles , respectively ) . These results demonstrate that Macf1a is essential to dissociate the Bb granule and relocalize Buc , the essential Bb-forming protein , from the Bb to the cortex to establish AV oocyte polarity . Furthermore , this analysis indicates that the Bb does not dissociate during a short window of time , but rather disassembles over a period of oocyte growth that corresponds to an almost 2-fold increase in volume ( from <120 μm to 145 μm diameter ) . Mutant macf1a oocytes display both an asymmetrically positioned nucleus and a Bb dissociation defect . These defects may represent two independent functions of Macf1a , one in Bb disassembly and one in nuclear positioning , or one of the defects may be a secondary effect caused by the other defect . For example , the enlarged , persistent Bb of macf1a mutants could displace the nucleus , causing it to become acentric . To determine if the Bb defect causes the nucleus to become asymmetric , we generated double mutants of macf1a and the buc mutant , which never forms a Bb . We hypothesized that if Macf1a functions independently in Bb disassembly and nuclear positioning , then a buc p106re; macf1a p6cv double mutant should display an absence of the Bb and a nuclear positioning phenotype . However , if the Bb is absent and the nucleus is no longer acentric in the double mutant , then the nuclear defect can be considered a secondary effect of the Bb defect in macf1a mutants . We analyzed buc; macf1a double mutant oocytes and found that the Bb was absent , while the acentric nuclear phenotype remained ( Fig 2A–2H ) . These results strongly support Macf1a functioning independently in Bb disassembly and nuclear positioning . We next tested if Macf1a regulates Bb disassembly and nucleus positioning directly by associating with the Bb and nucleus or if it acts via an indirect mechanism , for example , by localizing to the oocyte cortex and regulating these processes . To investigate this question , we examined the intracellular localization of Macf1a protein in stage I WT oocytes using an antibody against mouse Macf1 [9] and took advantage of a zebrafish gene trap line , Gt ( macf1a-citrine ) ct68a , inserted in a macf1a intron between exons 57 and 58 [29] . Using the Macf1 antibody , we found that Macf1a localizes to the Bb and nuclearly ( Fig 3A ) . Importantly , no immunostaining was observed in the macf1asa12708 mutant allele , indicating that the antibody is specific to Macf1a ( Fig 3D ) . In later stage I oocytes , Macf1a localization recapitulated the dynamics of Bb disassembly at the vegetal cortex , progressively dissociating from the Bb and localizing to the oocyte cortex ( Fig 3A–3C ) . Immunostaining for the Citrine insertion in Macf1a also showed similar localization to the Bb and followed its dynamics ( Fig 3E–3G ) . Perinuclear localization was also observed in about half the oocytes ( 9/20 ) ( Fig 3H ) , though it did not fully recapitulate the nuclear distribution observed with the Macf1 antibody . We postulate that Macf1 levels are lower in the nucleus than in the Bb based on our ability to detect nuclear Macf1 only in conditions of high Macf1 antibody concentrations ( see methods ) , whereas Macf1 in the Bb is consistently detected using high and lower Macf1 antibody concentrations ( discussed further later related to testing the function of the Macf1-ABD ) . Importantly , these results support a model where Macf1a functions directly in the Bb to regulate its dissociation and anchoring to the oocyte cortex . Macf1a localization to the nucleus suggests a direct role for it in positioning the nucleus as well . Macf1a contains binding domains for several different cytoskeletal elements and acts to integrate cytoskeletal systems in other models , thus we expect that Macf1a interacts with the oocyte cytoskeleton to regulate Bb dissociation and nuclear positioning . To determine which cytoskeletal components interact with Macf1a , we examined the distribution of actin and MTs in stage I oocytes . With this purpose , we performed live imaging experiments using the transgenes Tg ( actb1:lifeact-GFP ) [30] and Tg ( ef1a:dclk-GFP ) [31] to visualize actin and MTs , respectively . As shown in live and fixed samples , actin appeared as a thick cortical layer and intranuclearly , but was not present in the Bb in stage IB oocytes ( Fig 4A , 4F and 4G ) [5] . Actin localized similarly in macf1p6cv mutant oocytes ( Fig 4C ) . Upon Bb disassembly , Macf1 spreads at the cortex and is closely associated with cortical actin ( Fig 4B ) . We then tested the function of cytoskeletal components in stage I oocytes using pharmacological inhibitors . To test actin function , we disrupted actin filaments with Latrunculin A ( LatA ) and evaluated its effect on the Bb and nuclear positioning . After 6 hours ( h ) of treatment , we found that actin filaments were moderately affected ( S1A and S1E Fig ) , and only after 10-12h of LatA treatment were actin filaments greatly reduced ( Fig 4F–4G’ , S1B and S1F Fig ) . We treated ovaries with LatA for 12h or 20h , then fixed and stained for Buc . After 12h of LatA treatment , Buc appeared detached from the oocyte cortex in a few oocytes ( 4/22 ) , three of which also displayed an acentric nucleus ( 3/22 oocytes ) ( S1G and S1H Fig ) . After 20h of LatA treatment these effects were stronger . In control conditions Buc remained in the Bb or at the cortex ( Fig 4H–4I ) , while in LatA treated oocytes the Bb was closer to the nucleus as in a pre-disassembly stage ( Fig 4J–4K ) . Additionally , the nucleus was acentric in many oocytes ( 15/40 ) ( Fig 4K ) . It is possible that the stronger effect of 20 versus 12h of LatA treatment is due to actin filaments that remain but are undetectable after 12h of treatment , which are sufficient to preserve Buc at the cortex , but are effectively disrupted after 20h of treatment . Thus , disruption of actin partially phenocopies the macf1a mutant phenotype ( Fig 1B ) , suggesting that Macf1a may interact with cortical actin to mediate Bb disassembly and nuclear positioning . In agreement with previous reports on fixed tissue in Xenopus [32] , MT networks in live oocytes were present throughout the cytoplasm and were enriched perinuclearly ( Fig 4L ) . Though we detected stable MTs ( acetylated ) in the Bb in some oocytes ( Fig 4R , 10/25 oocytes ) , they were not enriched there . We addressed MT function by live imaging of ovaries treated with nocodazole for 2 and 10 h , or by incubating them at 4°C to also depolymerize stable MTs . In both cases , we found that depolymerization of MTs did not affect the Bb or nuclear positioning ( Fig 4L–4S , S1–S4 Movies ) . This suggests that MTs do not play a role in regulating the Bb structure , nuclear positioning or in cortical attachment like we observed for actin . Finally , we examined the distribution of IF using a Pan-Cytokeratin ( CK ) type II antibody . We found that CK was distributed in a punctate pattern in the Bb and cortically in stage 1B WT oocytes ( Fig 5A ) . We quantified the CK distribution in the Bb in WT using a custom MATLAB program and found that CK puncta density is significantly enriched in the Bb ( 4-fold increase ) compared to the cytoplasm ( Fig 5C and 5D ) . Interestingly , in macf1a mutants CK puncta were still significantly enriched in the Bb , though cytoplasmic CK appears accumulated around the nucleus , similar to the detachment of other components from the cortex in macf1a mutants ( Fig 5B and 5D ) . Thus , Macf1a may function in Bb disassembly via integrating CK in the Bb to actin at the cortex , and CK localization to the Bb is likely Macf1a independent . The modular domain structure of Macf1a allows us to interrogate Macf1a by targeting specific cytoskeleton-binding functions . We postulated that Macf1a mediates Bb disassembly by binding cortical actin via its Actin binding domains ( ABDs ) and by binding CK within the Bb via its Plectin repeat domain ( PRD ) , integrating the cortical actin and CK and therefore disassembling the Bb at the oocyte cortex . Results supporting this hypothesis are that CK and Macf1a are both enriched in the Bb and actin depolymerization disrupts Bb cortical anchoring . In addition , because we observed Macf1a localized to the nucleus and disruption of actin causes acentric nuclear positioning , the Macf1a-ABDs may also be required to position the nucleus in the oocyte . To interrogate Macf1a-ABD and -PRD functions in oocyte polarity , we developed a CRISPR/Cas9 approach to make large deletions in the endogenous macf1a gene . We designed sgRNAs targeting the introns flanking the ABD and the PRD encoding exons to remove each domain ( Fig 6 , and see methods ) . Importantly , the deleted exons did not alter the reading frame of the macf1a ORF . Our strategy was to first inject each sgRNA singly with Cas9 protein into 1-cell stage F0 embryos to confirm high frequency cutting ( Table 1 ) . Cutting efficiency was assayed by PCR amplification of genomic DNA spanning the target site , followed by high resolution melt analysis ( HRMA ) of the PCR product [33] . We increased the amount of sgRNA injected until every F0 embryo showed CRISPR-induced HRMA shifts . Then we targeted specific macf1a domains by simultaneously injecting two sgRNAs with Cas9 into 1-cell stage embryos . We optimized the sgRNA concentration further o obtain high frequencies of deletions without embryo abnormalities . We detected the deletions by PCR analysis of genomic DNA , then sequenced the PCR products to confirm that the deletions were consistent with our predictions . We raised F0 injected fish and tested for germline transmission in F1 embryos . Since F0 fish display germline mosaicism for the induced mutations , we do not expect Mendelian ratios in the F1 ( Table 1 ) . All future generations examined displayed normal Mendelian ratios of homozygotes , heterozygotes and wild-types . The Macf1a-ABD is composed of two calponin-homology ( CH ) domains: CH1 and CH2 . A single CH domain can bind actin , but many actin binding proteins contain two CH domains . Macf1 and some Plakin isoforms in other species contain either the full CH1-CH2 ABD or the CH2 domain alone [34] . We designed CRISPR sgRNAs to generate deletions that would recapitulate the known Spectraplakin isoform lacking CH1 , as well as to test if the full ABD functions in Bb disassociation and nucleus positioning . Thus , we targeted the CH1 domain , as well as both CH1and CH2 domains . To delete macf1a-CH1 we targeted introns 3 and 5 , deleting exons 4 and 5 , which spans 70% of the macf1a-CH1 coding sequence ( Fig 6A and 6B ) . To delete macf1a-CH1-CH2 we targeted introns 3 and 8 , removing exons 4–8 , which encompass 85% of the entire macf1a-CH1-CH2 ( Fig 6B ) . We succeeded in deleting 10 . 3 kb and 17 . 8 kb of genomic DNA to remove the Macf1a-CH1 and the Macf1a-CH1-CH2 , respectively , through germ line transmitted mutations ( Fig 6A and 6B , S2A and S2C Fig ) . The Macf1a-PRD is entirely contained within exon 35 ( NCBI: XP_001920094 . 1 ) . We targeted the immediately preceding intron 34; however , due to the small size of introns 35–37 , and to preserve the transcript ORF , we chose intron 38 as the second CRISPR target for deleting the PRD ( Fig 6A and 6C ) . Importantly , the 174 amino acid region encoded by exons 36–38 that were targeted for deletion does not appear to include key components of Macf1a functional domains ( based on a SMART domain prediction ) . We succeeded in deleting 9 . 1 kb of genomic DNA to remove the Macf1a-PRD in a germ line transmitted allele ( Fig 6C , S2B and S2C Fig ) . From the macf1a p1CH1 and macf1a p2CH1CH2 mutants , we sequenced ovary cDNA spanning exons 3 to 8 , which encode the ABD , and confirmed the lack of the intended exons in macf1p1CH1 and macf1a p2CH1CH2 ( Fig 6B , S2A Fig ) . For the PRD , we detected in both WT and macf1ap3PRD mutant oocytes a small RT-PCR band ( ~750 bp ) ( S2C Fig ) that sequence analysis showed corresponds to an alternatively spliced transcript that does not include exons 35–39 , as discussed above ( macf1a cDNA results ) . In addition , we detected a larger band ( 970bp ) only present in the macf1ap3PRD mutant that sequence analysis showed is generated by the deletion of exons 35 ( PRD ) to 38 . We amplified in ovary cDNA an ~1 . 5 kb band from the flanking exon 34 into exon 35 in WT ( confirmed by sequence analysis ) , which was absent in macf1ap3PRD mutants ( S2C Fig ) . Together , these results confirmed that we deleted the PRD from the macf1a endogenous locus and , generated a transcript lacking the Macf1a-PRD . In summary , we deleted the exons encoding the Macf1a ABD and PRD from the macf1a gene using CRISPR/Cas9 technology . The transcript produced from the genome-edited mutants lack these specific domains but preserve the macf1a transcript ORF . To determine the role of the Macf1a ABD in Bb progression and nuclear positioning , we analyzed the ovaries of macf1a p1CH1 and macf1a p2CH1CH2 deletion mutants . First , we generated transheterozygotes of macf1a p1CH1/macf1asa12708 and macf1a p2CH1CH2 /macf1asa12708 ( see methods ) . We found that in both cases the oocytes displayed a fully penetrant macf1a null mutant phenotype: the nucleus was acentric and the Bb failed to disassemble in late stage I oocytes ( Fig 7A ) . Then we examined homozygous mutant ovaries of each allele . We found that macf1a p2CH1CH2 homozygous mutant oocytes showed a macf1a null phenotype . However , macf1a p1CH1 homozygous mutant females exhibited an incompletely penetrant macf1a null phenotype , with ovaries from one female showing either a WT or mutant phenotype in the full set of oocytes ( Fig 7A ) . Importantly , the Macf1a mutant protein was produced and localized to the Bb in both macf1a p1CH1 and macf1a p2CH1CH2 mutant oocytes as in WT , and it showed similar expression levels in immunostaining ( Fig 8A–8C ) . We did not clearly detect Macf1 in the nucleus , most likely as a result of using lower Macf1 antibody concentrations for these experiments ( see methods and discussion ) . Similar to the macf1asa12708 mutant phenotype , Buc remains localized in the persisting , enlarged Bb of macf1a p1CH1/macf1asa12708 , macf1a p1CH1 , macf1a p2CH1CH2 /macf1asa12708 and macf1a p2CH1CH2 oocytes ( Fig 7 ) . AV polarity is also affected in the eggs of these mutant females ( Fig 8D , Table 2 ) . The incomplete penetrance of the macf1a p1CH1 ovary mutant phenotype is also observed in the AV egg phenotype ( Fig 7B , Table 2 ) . These findings show that the Macf1a-ABD mediates Bb granule dissociation at the cortex , nuclear positioning , and is essential for defining the AV axis ( Fig 9 model ) . Furthermore , both the CH1 and CH2 ABDs provide the robust function needed for Bb disassembly and nuclear positioning , whereas the CH2 domain alone can suffice but often is insufficient . Unexpectedly , when we analyzed the ovaries of the macf1ap3PRD/macf1asa12708 and macf1ap3PRD deletion mutants , which lack the domain that can interact with IF , Bb disassembly was not affected ( Fig 7A ) . In macf1ap3PRD mutant oocytes , Buc localizes to the Bb normally and the nucleus is centrally located as in WT . Furthermore , eggs from macf1ap3PRD/macf1asa12708 and macf1ap3PRD homozygous mutant females display normal AV polarity and development ( Fig 8D , Table 2 ) . These results show that the Macf1a-PRD is not required to localize Macf1a to the Bb or to disassemble the Bb RNP granule . Thus the Macf1a-PRD is dispensable for Macf1a function in AV polarity and nuclear positioning . Macf1a is a cytolinker that integrates cytoskeleton components in different cellular contexts [9 , 10 , 13 , 20 , 35]; however , a function for Macf1a in dissociating a large granule like the Bb is unprecedented , and so it is unclear how it may integrate MTs , actin , or IFs in Bb disassembly . Here , we developed a novel approach to unravel Macf1a domain-dependent function by interrogating , for the first time , the very large endogenous macf1a gene through CRISPR-Cas9 mediated exon deletions ( Fig 6 ) . We found that the Macf1a ABD is essential for Bb disassembly , whereas the IF interacting domain of Macf1a , the PRD , is dispensable . Since we did not detect MT enrichment in the Bb , nor a defect when MTs were depolymerized ( Fig 4L–4S ) , it is possible that Macf1a does not function as a cytoskeletal linker in this context . Rather it may link Bb components to the oocyte cortex through its ability to bind cortical actin and localize to the Bb ( Fig 9 ) . Thus , it may function to link two structures together , but not through its canonical cytoskeletal specific cross-linking function . C . elegans P-granules have become an excellent model for studying the assembly/disassembly dynamics of RNP granules . P granules form via the assembly of MEG proteins , which are intrinsically disordered proteins ( IDPs ) acting like a scaffold for P granule condensation [36] . MEG is the target of kinases and phosphatases that regulate the disassembly and assembly of P granules , respectively , and are crucial in establishing the anterior-posterior axis of the embryo [36 , 37] . Similarly , Buc and its ortholog in Xenopus Xvelo are IDPs functioning in Bb granule formation . The N-terminal region of Xvelo contains a prion-like domain required for Xvelo assembly in an amyloid-like matrix that can enclose Bb components [1] . In zebrafish oocytes , Bb cortical disassembly defines the future oocyte and egg vegetal pole , and Macf1a is the only known factor to function in this process . Our data show that Buc remains localized in a persistent Bb in macf1a mutants , suggesting that Macf1a either directly triggers Bb disassembly at the cortex or is required to mediate Bb component interaction with cortical factors that trigger Bb disassembly . Our results show that deletion of the CH1-CH2 actin binding domains in an otherwise intact Macf1a protein that localizes to the Bb causes a failure in Bb granule dissociation . These defects are indistinguishable from those of the two macf1a nonsense alleles . A macf1a mutant lacking the CH1 binding domain , leaving intact the CH2 domain in macf1a p1CH1/macf1asa12708 transheterozygotes also exhibits a fully penetrant mutant phenotype; however , the phenotype shows incomplete penetrance in macf1a p1CH1 homozygous mutants . This suggests that macf1a p1CH1 is an hypomorphic allele , unable to restore function over the null allele ( transheterozygous ) but partially restoring function in a homozygous condition . Interestingly , Macf1 and other Spectraplakins generate isoforms lacking the CH1 ABD that bind actin with lower affinity [9 , 38] . Although differences in genetic background may be affecting the penetrance of the phenotype , these results are consistent with the Macf1a-CH2 ABD having a lower affinity to actin that is frequently insufficient to link the Bb to cortical actin , revealing the functional importance of the conserved CH1-CH2 domains in Spectraplakins . These results indicate that the Macf1a-ABD interacts with cortical actin to anchor the Bb to the oocyte cortex where it undergoes dissociation . Consistent with this , actin disruption phenocopied aspects of the macf1a mutant phenotype , causing the detachment of the Bb from the cortex . It is possible that Macf1a interaction with actin is sufficient to trigger Bb disassembly , or that other factors are required in addition to Macf1a . P granules , for instance , disassemble upon phosphorylation of structural components [36]; thus , Buc may be phosphorylated at the cortex in a Macf1a-ABD dependent manner . Our analysis shows that the Bb does not disassemble all at once , but instead dissociates progressively during a period of about two-fold growth in oocyte volume . During this period the Bb becomes progressively smaller , although remains largely spherical , with Macf1a , Buc and other Bb components relocating as puncta to the oocyte cortex . We hypothesize that the Bb dissociates progressively with peripheral regions dissociating prior to more internal regions . During this process , it remains unclear whether Bb components fully dissolve at the cortex upon disassembly and then reassemble in aggregates that are docked cortically in a two- step process , or if the Bb fragments into smaller granules , with each granule anchored in a single step to the oocyte cortex . Either way , the Macf1a-ABD is a key activity in the process , which could be accompanied with modifications to Buc and/or other Bb components that act in Bb dissociation at the cortex . Shot , the Macf1 ortholog in Drosophila , also plays a key role in establishing oocyte polarity . Shot is required for anchoring MTs to the anterior and lateral oocyte cortex by interacting with the MT minus-end binding protein Patronin , which together with Shot functions as a noncentrosomal MT organizing center [39] . The Shot-ABD is required for its localization to cortical actin . The MT network organized by Shot/Patronin is key to setting up anterior-posterior polarity of the Drosophila oocyte . Both Macf1a and Shot function through their ABDs and localize with actin at the oocyte cortex , however , the mechanisms by which they act are likely distinct . In Drosophila , the assembly of MTs is downstream of Shot polarized cortical localization , which is regulated by Par-1 , whereas Macf1a localizes at the oocyte cortex after Bb disassembly . Nevertheless , a role for Macf1a more broadly in linking MTs to the cortex is consistent with an absence of MTs at the cortex in zebrafish macf1a mutant oocytes [5] , where it could act with Patronin to organize the MT network at a stage when the centrosome is absent [40] . Our live imaging data suggest that disruption of MTs does not affect the Bb . In keratinocytes and fibroblasts Macf1a localizes to MTs , and can function as a plus tip binding protein , stabilizing them [9 , 10] . Also in zebrafish hair cells , the expression of the Macf1a-Citrine fusion in Gt ( macf1a–citrine ) ct68a line shows that Macf1a colocalizes with actin and microtubules [14] . However , MTs are not enriched in the Bb and disruption of the MT network with nocodazole and cold treatment does not affect Bb morphology , RNA localization or cortical attachment ( Fig 4L–4S ) [8 , 25 , 32 , 41 , 42] . Thus , we do not consider a role for MTs or the Macf1a-MTBD currently in Bb disassembly . Nevertheless , since Macf1a stabilizes MTs and connects them to cortical actin in other cell types [9 , 10 , 13] , we cannot rule out a contribution of MTs , until the conditions for culturing zebrafish oocytes allow in vivo visualization of Bb disassembly dynamics and the role of MTs in the process can be addressed . Future studies interrogating the Macf1a-MTBD will also clarify its contribution in oogenesis . We found here that Macf1a localizes to the nucleus in the oocyte , suggesting that Macf1a plays a direct role in positioning the nucleus centrally in the oocyte . We postulate that Macf1a nuclear concentration is lower than in the Bb , as evidenced by the absence of Macf1a nuclear staining in conditions of lower Macf1 antibody concentration ( Fig 8 versus Fig 3 , see methods ) . Our observation is supported by a recent study describing the nuclear proteome of frog oocytes , that quantified the nucleocytoplasmic partitioning of ~9 , 000 proteins [43] . Among those is Macf1 , showing a nucleocytoplasmic partition of ~0 . 08 , suggesting that Macf1 is ~12X higher in the cytoplasm than in the nucleus . Further evidence from the macf1a; buc double mutant shows that the nuclear localization defect is not caused by the Bb defect ( Fig 2 ) . Macf1a may interact with proteins residing in the nuclear envelope ( NE ) to connect the nucleus to the cytoskeleton . Although Nesprins residing in the outer NE interact with MTs to regulate nuclear positioning [44–50] , our data indicate that disruption of MTs does not affect nuclear positioning . Rather , we found that actin disruption often leads to acentric nuclear positioning ( Fig 4K ) , albeit we did not detect actin filaments around the nucleus and instead as an intranuclear mesh . Considering that the Macf1a ABD is required for nuclear positioning , it is possible that it interacts with cytoplasmic or perinuclear actin filaments that are undetectable in our conditions or , alternatively , that the Macf1a ABD interacts with a yet unknown partner in positioning the nucleus . Future studies will be required to deduce the mechanism . In summary , we developed a genome editing approach using CRISPR-Cas9 to interrogate the cytolinker function of Macf1a . With this approach , we identified the Macf1a ABD as essential for Bb RNP granule disassembly and , thus , for establishing AV polarity of the oocyte and egg . The Macf1a ABD is also required for centric nucleus positioning in the oocyte . Moreover , we determined that the Macf1a PRD is dispensable for Macf1a function in oogenesis indicating that Macf1a does not interact with CKs to regulate Bb dissociation or nucleus localization , though we cannot rule that other Macf1 domains may be interacting with CKs . This is the first study to address Macf1a functional domains by targeting its endogenous locus . Although the use of CRISPR technology is widespread nowadays , applications like the one presented here take a step further in developing more powerful and elegant genome editing approaches than solely generating null alleles , to reveal a deeper understanding of basic cellular mechanisms . We expect that applying similar strategies will be valuable to understand the function of other modular proteins like Macf1 , including the related genes dystonin [51] , dystrophin [52] and plectin [53] , which like macf1 [54] can lead to complex human diseases . All animal studies were approved by the University of Pennsylvania Institutional Animal Care and Use Committee ( Protocol number 804214 ) . Animal care and use adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals . Ovaries were collected from 3 to12 month old adult fish of TU wild type , macf1ap6cv [5] , macf1asa12708 ( Sanger Center Mutation Resource ) , bucp106re [2 , 3] , Tg ( ef1a:dclk-GFP ) [31] , Tg ( bactin2:HsENSCONSIN17- 282-3xEGFP ) [55] , Tg ( actb1:lifeact-GFP ) [30] and Gt ( macf1a–citrine ) ct68a [29] . For genotyping we used the following primers and PCR conditions: For macf1asa12708 genotyping , we used the KASParTM genotyping protocol of LGC Genomics [56] . KASPar sequence: CTGGTAGCCATGTCCTCCTCTGAGGATGAAGGCAGTCTGCGCTTTATTTA[T/A]GAGCTGCTGGGATGGGTWGAAGAAAMGCAAGATCTGCTGGAGCGAGCTGA . For Gt ( macf1a–citrine ) ct68a genotyping , we used For: ACGTAAACGGCCACAAGTTC , Rev: AAGTCGTGCTGCTTCATGTG . Denaturing: 94°C , 1:00 min . Annealing: 60°C , 45 sec Extension: 72°C , 30 sec X 30 cycles . The macf1ap6cv/+ [5] and macf1asa12708/+ heterozygotes exhibit a wild type phenotype and were used as controls alongside macf1ap6cv [5] and macf1asa12708 homozygous mutant alleles . In addition , both homozygous mutant alleles show indistinguishable mutant phenotypes , and we used them arbitrarily depending on fish availability . Ovaries were dissected from euthanized females and dissected carefully to preserve stage I-II oocytes and to remove later stage oocytes . Then the dissected ovaries were digested with 1 . 5 mg/ml collagenase I ( Sigma-Aldrich ) for 15 minutes in L-15 Medium ( Sigma-Aldrich ) . Ovaries were fixed according to the acid fixation method [57] and kept overnight in 4% formaldehyde . Following washes in PBS , ovaries were kept in cold methanol at -20°C for at least 6 hours before use . For immunostaining , ovaries were washed and rehydrated in decreasing methanol percentages ( 75% , 50% , 25% ) and washed finally in PBS 3 times ( x ) . Ovaries were incubated in blocking solution containing PBS ( 0 . 3% Triton X-100 , 1% BSA ) for 1 . 5 to 2h . Primary antibodies were diluted in blocking solution and incubated overnight at 4°C . Ovaries were washed in blocking solution 4x15min and incubated with secondary antibodies in blocking solution for 90 min . Ovaries were washed in PBT ( 0 . 1% Triton ) 4x15min and lastly incubated in PBT with DAPI ( 1:1000 ) and DiOC6 ( 1ug/ml ) ( Calbiochem ) for 1–2 h . Then they were washed in PBT 4x10min , transferred into vectashield ( Vector labs ) and mounted for imaging . For Macf1 staining in Fig 3A–3C , after collagenase treatment we collected oocytes using a cell strainer ( 100 μm ) ( FisherbrandTM ) , pipetted them into a 12-well plate , and performed the immunostaining in small volumes following the protocol described in [58] . This protocol for isolated oocytes may favor antibody penetration compared to the whole mount method used for Citrine staining and in Fig 8 . For CK detection , we used ovary tissue sections , where ovaries were first dissected and fixed in 4% formaldehyde overnight . After several washes in PBS , the ovaries were immersed in 30% sucrose for 48–72 hours to cryoprotect the samples , and stored in -20°C until cryosections were obtained . The immunostaining was performed on the mounted tissue sections . Images were acquired on a Zeiss LSM 710 confocal microscope using a 40X lens . The acquisition setting was set between samples and experiments to: XY resolution = 512x512 pixels , pinhole adjusted to 1 . 1μm of Z thickness , increments between stack images were 1μm , laser power and gain were set for each antibody . Acquired images were adjusted only in contrast/brightness . All figures were made using Adobe Photoshop and Illustrator CC 2014 . Cytokeratin puncta were quantified using a MATLAB ( Version 8 . 2 . 0 . 701 R2013b 64-bit , MathWorks ) script . All images were pre-processed to reduce noise and to separate each oocyte into a single image . Cytokeratin positive puncta were identified using a simple threshold . The oocytes were sectioned into three regions ( cytoplasm , nucleus , and Balbiani body ) according to the intensity of DiOC6 staining . Additionally , a region of just the perimeter of the cytoplasm was defined as any cytoplasm within one Balbiani-body-diameter of the cytoplasmic membrane . Puncta density was calculated as puncta per area for each region . Ovaries were dissected from adult fish ( 3 to 12 months ) in L-15 media ( 60% in Hanks solution with gentamycin ( 50μg/ml ) ( Gibco ) ) supplemented with FBS ( 10% ) and insulin ( 15μg/ml ) at 28°C . In the media , further dissection of the ovary was performed to isolate small pieces containing mainly stage I oocytes . The ovaries were placed in a glass bottom dish and embedded in low-melt agarose ( 0 . 5% ) prepared in the media solution . The dish was filled with media solution containing MitoTracker ( 500 nM , Molecular Probes ) for 2–3 hours , then the media was replaced once and ovaries kept in the media throughout the imaging . Nocodazole ( 50μM ) ( EMD Milipore ) and latrunculin A ( Sigma-Aldrich ) ( 20μg/ml ) were used to destabilize MTs and actin , respectively . When live imaging began , ovaries had been exposed to the drugs in the incubation media for no longer than 15–20 min . The control group was treated with DMSO in the same conditions . We used Mitotracker staining to monitor oocyte health and viability during the treatment . For analysis of LatA treatment in fixed samples , ovaries from the same fish were divided into Latrunculin A ( Sigma-Aldrich ) ( 20μg/ml ) and DMSO treated groups with two replicates for each condition . We tested incubation times of 6 , 12 and 20 hours . After the treatment , ovaries were fixed as above and stained for Buc . We measured the oocyte diameter and analyzed the effect of LatA treatment only in stage I oocytes . For cold treatment , ovaries were dissected as explained previously; ovaries were kept in culture media , and divided in tubes placed in ice in a cold room ( 4°C ) and controls kept at 28°C , both for 120 min . Then ovaries were fixed and processed for immunostaining . To generate macf1a cDNA we used the Superscript First-cDNA synthesis system ( Invitrogen ) . First , using the macf1a predicted ORF sequence ( ~25 kb ) ( NCBI: XP_001920094 . 1 ) , primers were designed to amplify a large piece of macf1a cDNA ( ~19 kb ) made from ovary RNA . Forward: CCACCGAAAAACAGGAGAACAC; Reverse: GCTCCACTTGAAACCTCTTCGC . Instead a ~6 . 5 kb product was obtained that was cloned into pCR-XL-TOPO ( Invitrogen ) ( kindly provided by Tripti Gupta ) . We sequenced the 6 . 5 kb cDNA and found that it contained three regions from the macf1a predicted cDNA sequence: bp 3767–4482 , 12314–15370 and 20453–23180 . The 4482–12314 gap corresponds to exons 35–39; exon 35 contains the Macf1a PRD ( ~7 kb ) domain . We amplified from ovary cDNA ~1 . 5 kb of exon 35 ( NCBI:XP_001920094 . 1 ) using primers from exon 35 and flanking exon 34 . We also amplified from ovary cDNA and sequenced exons 36–39 , from flanking exon 35 . In the ~25 kb predicted transcript ( NCBI: XP_001920094 . 1 ) , 29 spectrin repeats are predicted . We identified 2 spectrin repeats in exons 36–39 , 14 spectrin repeats in the 6 . 5 kb cDNA spanning sequences 12314–15370 , and another 13 were identified in cDNA by sequencing the gap in the 6 . 5 kb cDNA between 15370–20453 bp , corresponding to exons 57–75 , for a total of 29 spectrin repeats . To complete the macf1a cDNA , we performed 5’ RACE , amplifying overlapping fragments upstream of 3955 bp of the 25 kb predicted transcript ( NCBI:XP_001920094 . 1 ) to assemble a fragment containing the macf1a CH1-CH2 ( ABD ) and Plakin domains . Using 3’ RACE from the 23 , 180 bp position , we assembled a fragment containing the MTBD ( ~3 kb ) . See the list of primers in S1 Table . macf1a deletion mutants were created using CRIPSR-Cas9 mediated mutagenesis . The intron targets selected contained a PAM sequence for Cas9 targeting . The sgRNA constructs to target macf1a introns were purchased from the University of Utah Mutation Generation and Detection Core , which cloned them into plasmids containing an upstream T7 promoter and flanked by a DraI restriction site . To synthesize sgRNAs , we used the T7 MEGAshortscript kit ( Ambion ) and a clean-up step with MEGAclear kit ( Ambion ) , following the manufacturer’s protocol . For injections , we mixed sgRNAs ( Table 1 ) and Cas9 protein ( 180–200 pg ) ( purchased from the University of Utah Mutation Generation and Detection Core for Crispr reagents ) and injected 1 . 2–1 . 5 nl of the CRISPR mix into one-cell stage zebrafish embryos . At 24 to 48 hpf , a fraction ( ~25% ) of the injected embryos were euthanized and DNA was extracted . We performed HRMA analysis on single embryos using MeltDoctor HRM Master Mix ( Applied Biosystems ) to determine the mutation rate , and PCR analysis to detect genomic deletions . When a high ( ~80–100% ) mutation rate was obtained , the remaining injected embryos were raised to adulthood . We crossed F0 adults carrying macf1a p1CH1 and macf1a p2CH1CH2 deletions to wild type fish to produce F1 macf1a p1CH1 and macf1a p2CH1CH2 heterozygotes . We then crossed these heterozygotes to either a macf1asa12708/+ female or a macf1asa12708 homozygous male to obtain F2 transheterozygous ovaries of macf1a p1CH1 /macf1asa12708 or macf1a p2CH1CH2 /macf1asa12708 . In the next generation , we incrossed the F2 fish to obtain homozygous mutants for analysis . HRMA primers: For PCR deletion analysis , we combined three primers to amplify the wild type and mutant alleles . PCR conditions: All statistical analysis and plotting was performed using the GraphPad Prism 6 and Excel software . See S1 Methods .	The totipotent egg of most vertebrates is polarized in a so called animal-vegetal axis that is essential to early embryonic development . The animal-vegetal axis is established in the early oocyte by the dissociation of the Balbiani Body ( Bb ) . The Bb is a large RNA-protein granule , conserved from insects to mammals , that forms next to the oocyte nucleus and dissociates later at the oocyte cortex . Importantly , Bb dissociation at the oocyte cortex defines the future vegetal pole of the egg . Macf1a , a cytolinker , is the only factor known to regulate Bb dissociation . However , how the giant Macf1a protein with multiple functional domains can interact with the cytoskeleton to regulate Bb disassembly is unknown . Here , we unravel Macf1a function via interrogating , for the first time , individual macf1a-encoded domains of the gene in its normal chromosomal location for their requirement in Bb dissociation and ultimately in egg polarity establishment . The method presented here is applicable to other cytolinkers involved in human disease .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "medicine", "and", "health", "sciences", "reproductive", "system", "vertebrates", "animals", "xenopus", "germ", "cells", "animal", "models", "oocytes", "model", "organisms", "amphibians", "experimental", "organism", "systems", "forms", "of", "dna", "dna", "cellular", "structures", "and", "organelles", "cytoskeleton", "research", "and", "analysis", "methods", "contractile", "proteins", "genome", "complexity", "actins", "genomics", "animal", "cells", "proteins", "frogs", "complementary", "dna", "xenopus", "oocytes", "ovaries", "biochemistry", "cytoskeletal", "proteins", "cell", "biology", "ova", "anatomy", "nucleic", "acids", "phenotypes", "genetics", "biology", "and", "life", "sciences", "cellular", "types", "computational", "biology", "introns", "organisms" ]	2017	Microtubule-actin crosslinking factor 1 (Macf1) domain function in Balbiani body dissociation and nuclear positioning
Despite its role in homogenizing populations , hybridization has also been proposed as a means to generate new species . The conceptual basis for this idea is that hybridization can result in novel phenotypes through recombination between the parental genomes , allowing a hybrid population to occupy ecological niches unavailable to parental species . Here we present an alternative model of the evolution of reproductive isolation in hybrid populations that occurs as a simple consequence of selection against genetic incompatibilities . Unlike previous models of hybrid speciation , our model does not incorporate inbreeding , or assume that hybrids have an ecological or reproductive fitness advantage relative to parental populations . We show that reproductive isolation between hybrids and parental species can evolve frequently and rapidly under this model , even in the presence of substantial ongoing immigration from parental species and strong selection against hybrids . An interesting prediction of our model is that replicate hybrid populations formed from the same pair of parental species can evolve reproductive isolation from each other . This non-adaptive process can therefore generate patterns of species diversity and relatedness that resemble an adaptive radiation . Intriguingly , several known hybrid species exhibit patterns of reproductive isolation consistent with the predictions of our model . The evolutionary significance of hybridization has been a hotly debated topic for decades [1] . Homoploid hybrid speciation , speciation that occurs as a result of hybridization without a ploidy change [2 , 3] , is generally thought to be an exceptionally rare outcome of hybridization , and there are indeed only a handful of well-supported cases of this phenomenon [4] . Though it is not uncommon for species’ genomes to exhibit evidence of past hybridization , hybrids are often thought to be weakly isolated from parental species , though few studies have explicitly investigated this . Empirical research on homoploid hybrid speciation over the last decade has primarily focused on the role of hybrid phenotypes in establishing reproductive isolation between hybrids and parental species [5–9] . Hybrids can have recombinant or transgressive traits that differentiate them from parental species . In some cases , these traits can allow hybrids to occupy new niches . For example , in Rhagoletis fruit flies , hybrid lineages have novel host preferences , potentially contributing to reproductive isolation between hybrids from parental species [10 , 11] . Similarly , if hybrid lineages have novel mate preferences , this can isolate hybrids from parental species via assortative mating , a mechanism which has been implicated in hybrid speciation in Heliconius butterflies ( [8] , and see [12] for a model of this process ) . This work has lead to the idea that novel hybrid phenotypes are key to hybrid speciation [13] . Despite several well-documented examples [6 , 8] , it has been difficult to determine the evolutionary importance of hybrid speciation , in part because few theoretical models have been developed . The existing models of hybrid speciation simulate either positive selection on certain hybrid genotypes or inbreeding [9 , 12 , 14] . In one model [14 , 15] , novel combinations of underdominant parental inversions can fix in hybrid populations , particularly if the novel inversion combination is under positive selection or if rates of inbreeding ( selfing ) are high ( see Discussion ) . Though there is evidence that this process combined with ecological factors was involved in the formation of hybrid Helianthus sunflower species [5 , 6 , 16] , the basis for invoking positive selection on recombinant inversion genotypes is unclear . Later versions of this model incorporated ecological differentiation between hybrid and parental species and showed that hybrid speciation occurred frequently if hybrids had higher fitness than parental species in an unoccupied niche [9 , 17] . Though hybridization often generates novel traits [18–20] it is difficult to evaluate the likelihood that these traits will be more fit than parental types ( ecologically or intrinsically ) , making it difficult to predict the importance of hybridization in generating new species by positive selection on hybrid genotypes . The genetic incompatibility of hybrids constitutes a key component of reproductive isolation between many species , and is the basis for the biological species concept . While previous models of hybrid speciation incorporated inversions [21] , here we investigate the potential role of negative epistatic interactions , another important genetic mechanism of speciation . The first genetic model of speciation , described by Bateson , Dobzhansky and Muller ( the BDM incompatibility model , S1 Fig . , [22–24] ) , predicts that mutations at two genetic loci differentially accumulating along two lineages can negatively interact in their hybrids . Empirical research has shown that these types of negative epistatic interactions are remarkably common [25–29]; reviewed in [24 , 30 , 31] . Though the theory of genetic incompatibilities was originally formulated in the context of allopatrically diverging species , more recent research has investigated dynamics of these incompatibilities in the context of hybrid zones . Under the simplest BDM scenario , derived genotypes are presumed to be neutral , meaning that they have the same fitness as ancestral genotypes . When there is gene flow between species , neutral BDMs are predicted to fix for genotype combinations that are compatible with either parental species [32] , rendering them ineffective barriers to gene flow [33 , 34] . However , incompatibilities may also frequently arise due to adaptive evolution or coevolution of pairs of loci along lineages ( S1 and S2 Figs [24 , 32 , 35–37] ) . Such incompatibilities are more effective barriers to gene flow than neutral BDM incompatibilities ( [38] , see also [39] ) . In its initial description , the BDM model envisioned incompatibilities that cause complete hybrid inviability or sterility , but many negative epistatic interactions in interspecific crosses have more moderate effects on fitness ( e . g . [40–42] ) , allowing hybrid populations to persist . With few exceptions , previous work on genetic incompatibilities has focused on their role in maintaining reproductive isolation between parental species . As a result , hybrid populations have primarily been modeled as tension zones , but incompatibilities may also have interesting dynamics within isolated hybrid populations ( i . e . hybrid swarms ) . Here we present a new model in which reproductive isolation between hybrid and parental populations emerges as a consequence of selection against incompatibilities in a hybrid swarm . Selection on a single adaptive or coevolving incompatibility pair can result in the fixation of genotype combinations that contribute to isolation between the hybrid population and one or the other parental species . Here , we show that in the presence of multiple pairs of such incompatibilities , this process can result in the rapid evolution of reproductive isolation of hybrid populations from both parental species ( Fig . 1 ) . Two features of this model make it particularly plausible biologically . First , as noted above , negative epistatic interactions are common , providing ample raw material for our model . Second , hybrid populations in which hybrids are abundant are common in nature ( e . g . [43–45] ) . Though ecological and sexual selection are important factors in the few well-documented cases of hybrid speciation [6 , 8] , our results suggest that hybrids can evolve reproductive isolation as a result of selection against genetic incompatibilities alone . In the simplest model of a hybrid population , an equal mixture of individuals from both parental species form a new isolated population and mate randomly with respect to genotype ( Figs . 1 and S3 ) , such that the first mating event generates 50% F1 hybrids and 25% each parental species . Using theory of two-locus selection [46 , 47] , hereafter the “deterministic two-locus model” , one can model the effect of selection at two polymorphic loci on gamete frequencies of a diploid sexual population ( see Methods and S1 Text ) . This model describes the dynamics of two loci subject to any arbitrary fitness matrix . Here , we focus on fitness matrices for three types of incompatibilities that may commonly arise between species ( S1 and S2 Figs; [30 , 35] ) : 1 . BDM incompatibilities arising from neutral substitutions , 2 . BDM incompatibilities arising from adaptive substitutions , and 3 . BDM incompatibilities arising from coevolution between loci . Applying the two-locus selection model to these incompatibility types , one can see that the direction of fixation depends on the initial frequency of the parental alleles ( f , see S3 and S4 Figs ) and dominance at each locus ( h , S4 Fig . ; see also [48] ) . This purely deterministic model of selection on hybrid incompatibilities is unrealistic because even large populations experience some degree of genetic drift . We thus extended the model to include genetic drift , which can affect the speed and direction of fixation of incompatibility pairs ( S5 Fig . ) . For neutral BDM incompatibilities ( S1 Fig . ) , this model does not predict fixation of genotypes incompatible with either parental species ( S4 Fig . ) . In contrast , for coevolving or adaptive BDM incompatibilities ( S1 Text , S1 and S2 Figs ) , the two-locus finite population model predicts that at equal admixture proportions ( f = 0 . 5 ) , a single incompatibility pair has a 50% chance of fixing for either parental allele combination ( Fig . 2 , S2 Text , S1 Table ) . Interestingly , while genetic drift in small populations could accomplish the same thing ( 9 ) , the process described here occurs rapidly in large populations and is driven by deterministic selection ( Fig . 2 ) . Given these dynamics , it is clear that large hybrid populations with two or more of these types of hybrid incompatibilities could , in principle , fix for one parental genotype at one incompatibility pair and the other parental genotype at the other incompatibility pair ( Fig . 1 ) . This outcome would result in reproductive isolation of the hybrid population from both parental species . With two codominant incompatibility pairs and equal admixture proportions , the probability that a hybrid population will become isolated can be predicted by a simple binomial . However the binomial prediction breaks down when there is variation in dominance , admixture proportions , or linkage between incompatibilities , and thus we explore these further by simulation . To investigate the dynamics of multiple incompatibility pairs , we simulated a large , randomly-mating and spatially isolated hybrid population with two pairs of unlinked hybrid incompatibility loci ( see Methods; S3 Fig . , setting m1 = m2 = 0 ) . The fitness scheme used is that of a coevolutionary incompatibility model ( S2 Fig . ) , assuming that incompatibilities are codominant ( i . e . h = 0 . 5 ) , that fitness is symmetric with respect to the parental source of alleles ( i . e . wij = wji ) and that the cumulative fitness effects of multiple incompatibility pairs is multiplicative . If hybrid populations fixed for the parent species 1 genotype at one incompatibility pair and the parent species 2 genotype at the other , we considered the hybrid population as having evolved reproductive isolation from both parental species ( albeit weaker than between the two parental species ) . While selection against hybrids will sometimes be so extreme that few hybrids will survive ( or reproduce ) in the population ( see simulations below ) , selection against hybrids can also be more moderate , allowing hybrids to persist [41 , 45 , 49–53] . In simulations of this moderate selection scenario , reproductive isolation between hybrid and parental populations can evolve frequently and rapidly ( Fig . 3 ) . For example , for two incompatibility pairs with selection coefficients ( s ) of 0 . 1 , 47±2% of simulated hybrid populations became isolated from both parental species within an average of ~200 generations . Exploring a range of s ( 0 . 1–0 . 5 , S6 Fig . , S2 Table ) , initial admixture proportions ( f = 0 . 3–0 . 7 , S7 Fig . ) , and population sizes ( 100–10 , 000 diploids , S3 Table ) , we conclude that , unless fitness of hybrids is low ( i . e . F1 fitness <0 . 5 ) or ancestry of the founding population is substantially skewed ( >60% one parental species ) , reproductive isolation evolves rapidly and with surprisingly high probability ( 27±2% to 43±2% of the time; on average within 75 ± 16 to 258 ± 38 generations , see S3 Text ) . In the above simulations , we assume that selection on different hybrid incompatibility interactions is symmetrical ( s1 = s2 , S2 Fig . ) , but it is unlikely that selection is truly equal on different hybrid genotype combinations . When fitness is completely asymmetrical ( i . e . s1 = 0 in S1 Fig . , as for neutral BDM incompatibilities ) , only strong genetic drift can cause the fixation of genotype pairs that are incompatible with either parental species , as selection cannot do so ( see S4 , S8 , S9 Figs , S4 Text ) . This reliance on genetic drift implies that this process will be slow unless an extreme bottleneck is invoked . In contrast , the dynamics of BDM incompatibilities resulting from adaptation within parental lineages can be quite different ( S1 Fig . ) . Notably , while selection may also be highly asymmetric in such cases [38 , 54] , derived alleles have higher fitness than ancestral alleles , allowing for the fixation of genotype combinations that are incompatible with both parental species . We find that isolation evolves with similar frequency under asymmetric selection as long as selection is strong relative to drift ( S3 Text , S4 Table ) , because even weak selection will prevent the fixation of the ancestral genotype . Above we simulated codominant hybrid incompatibilities ( h = 0 . 5 ) , but the two-locus model ( S4 Fig . ) shows that patterns of fixation are different depending on the value of h . In particular , when h is zero or unity , fixation is not strongly dependent on admixture proportions ( S4 Fig . ) . When we simulate variation in dominance among incompatibility interactions ( see S3 Text , S5 Table ) , we find that reproductive isolation between hybrid populations and parental species evolves with comparable frequency ( 42–48±2% vs 47±2% under the codominant scenario ) . Recent empirical studies have suggested that most species are distinguished by multiple hybrid incompatibilities [30 , 41 , 55–59] . Theoretically , barring extinction of the hybrid population ( see simulations below ) , increasing the number of pairs of incompatibilities should increase the probability that a hybrid population will evolve isolation from both parental species . In order to illustrate this , we simulated 3–6 unlinked hybrid incompatibility pairs ( S5 Text ) . As expected , increasing the number of hybrid incompatibilities increases the probability that the hybrid population will be isolated from each parental species by at least one incompatibility pair ( >90% with 6 incompatibility pairs , Figs . 3 , S6 , S5 Text ) . We assume in most of our simulations that loci involved in hybrid incompatibilities are completely unlinked . As the number of incompatibilities increases , this becomes unlikely . Genetic linkage between loci involved in different epistatic interactions can reduce the frequency at which hybrid populations evolve isolation because alleles are more likely to fix for the same parental genotype ( S10 Fig . , S5 Text , S6 Table ) . Interestingly , when coevolving incompatibility loci are linked to a neutral BDM incompatibility , this does not significantly lower the frequency at which hybrid populations evolve reproductive isolation ( S5 Text ) . Furthermore , linkage between coevolving incompatibilities and neutral BDM incompatibilities can more frequently result in fixation of neutral BDM incompatibilities for a parental genotype ( 16±2% ) , resulting in stronger isolation between hybrid and parental populations ( S5 Text ) . The above simulations focus on simple models that show this process can occur in principle . To capture more biological realism in the number and types of incompatibilities , we simulated 20 incompatibility pairs with randomly determined genomic position and dominance , exponentially distributed selection coefficients ( mean s = 0 . 05 ) and variation in asymmetry of selection ( see above and S5 Text ) . In these simulations , 95% of populations developed isolation from both parental species . On average , the hybrid population first evolved isolation from both parental species after ~250 generations and was isolated from each by 7 incompatibility pairs within 1000 generations . Since incompatibility pairs with the largest fitness effects tend to fix first , hybrid populations developed considerable reproductive isolation from parental species even before all incompatibilities were fixed in the population ( Figs . 4 and S11 ) . Overall , our simulations suggest that rapid evolution of reproductive isolation of hybrid populations is likely when parental species are separated by several hybrid incompatibility pairs . Reproductive isolation between hybrids and parental species is less likely to evolve as the fitness of hybrids decreases . For example , if we repeat the simulations above ( i . e . the 20 incompatibility pairs with exponentially distributed selection coefficients ) , if s- = 0 . 1 , the average fitness of an F1 hybrid between the two parental species is 0 . 38 and isolation evolves in only 56±2% of simulations . When s- = 0 . 2 , the average fitness of hybrids is 0 . 1 , and only 1 . 4±0 . 5% of simulated populations develop isolation and parental genotypes dominated in these populations . Thus , this process is unlikely to occur between species in which post-zygotic isolation is nearly complete . Similarly , if parental individuals in the simulated hybrid population mate assortatively with conspecifics , reproductive isolation between hybrids and parental species is significantly less likely to evolve ( S6 Text ) . The reasons for this are two-fold: assortative mating prevents the formation of a large hybrid population , and parentals outcompete early generation hybrids that are still segregating for parental incompatibilities . We model a completely isolated hybrid swarm , but many hybrid populations experience gene flow with parental species . Ongoing migration may impede the evolution of reproductive isolation by preventing the fixation of genetic incompatibilities . To evaluate this , we simulated hybridization scenarios with ongoing migration ( S3 and S13 Figs , 4Nm = 8–20 ) . Even with substantial gene flow from parental populations , hybrid populations evolved reproductive isolation from them at high probability ( i . e . 38±2% of simulations with two incompatibility pairs , s = 0 . 1 and 4Nm = 8; S6 Text; S7 Table ) . In the above simulations , we assume that migration is symmetrical from both parental species , but asymmetrical migration may be common in hybrid zones ( e . g . [60–62] ) . To explore how asymmetrical migration could influence our results , we varied asymmetry in migration rates ( S6 Text ) . As expected , when migration rates were high and strongly asymmetrical ( S12 Fig . ) , hybrid reproductive isolation from both parental species evolved infrequently . However , in less extreme cases , hybrid reproductive isolation was still observed frequently ( e . g . >20% of simulations with 4Nm<20 , S12 Fig . ) . It is interesting to consider the fact that chance plays an important role in the direction that incompatibility pairs fix . As a result , one would expect that two or more independently formed hybrid populations from the same pair of parental species could evolve isolation from each other . To demonstrate this effect , we simulated two hybrid populations formed from the same pair of parental species ( S14 Fig . ) . In the absence of migration , the two hybrid populations evolved isolation from each other frequently ( 50±5% , as expected given two hybrid incompatibility pairs , see S6 Text; S8 Table ) . Remarkably , this outcome is still observed with relatively high gene flow between the two hybrid populations ( 24±4% with 4Nm = 8 and two hybrid incompatibility pairs , S6 Text; S8 Table ) . We describe a new model of the evolution of reproductive isolation of hybrid populations , a first step towards hybrid speciation . Unlike previous models of hybrid speciation , our model does not assume positive selection on hybrid genotypes or inbreeding , but rather deterministic selection against hybrid incompatibilities in randomly mating hybrid populations . With moderate selection ( i . e . s≤0 . 2 ) on two or more incompatibility pairs in an allopatric hybrid population , reproductive isolation from both parental species emerges with ~50% ( or higher ) probability . Hybrid reproductive isolation also evolves frequently with substantial levels of ongoing migration between hybrids and parental species ( 4Nm < 20 each parent ) . Another striking result of our simulations is the speed with which reproductive isolation evolves between hybrids and parental species . Depending on parameters , reproductive isolation can emerge in fewer than 100 generations with moderate selection ( S3 Text ) . The idea that hybrid speciation can occur rapidly has been supported by experimental results [14 , 63 , 64] and to some extent by previous models of hybrid speciation [9 , 14] . Our model suggests that simple selection on incompatibilities in hybrid populations could also lead to rapid reproductive isolation on timescales much faster than expected for allopatric speciation due to the accumulation of neutral BDM incompatibilities . Given that epistatic incompatibilities are common , our results on the probability and speed of isolation suggest that this process may frequently occur in hybrid populations . Previous empirical work has emphasized the importance of ecological differentiation between hybrid and parental populations or positive selection on hybrid genotypes as a route to hybrid reproductive isolation [6 , 8–10 , 12 , 63 , 65] . The novel finding of our simulations is that reproductive isolation evolves readily in hybrid populations without positive selection on hybrids . However , the two are not mutually exclusive and ecological factors , which have been shown to underlie several cases of hybrid speciation [6 , 8 , 63] , may complement selection on genetic incompatibilities to further strengthen reproductive isolation . For example , in Helianthus , a combination of chromosomal rearrangements and novel hybrid phenotypes are important in distinguishing hybrid and parental species [6 , 66] . Like other models ( [9 , 14] ) , our model predicts that isolation between hybrids and parental species is inherently weaker than isolation between the two parental species . We propose that fixation of incompatibilities could be a crucial step in initially limiting gene flow between hybrids and parental species , allowing for the development of other isolating mechanisms . For example , theoretical work predicts that reinforcement can develop even when selection against gene flow is moderate [67–70] . Previous models of hybrid speciation have incorporated species-specific inversions that are assumed to be underdominant . Under this “underdominant inversion” model , hybrid populations can fix for novel inversion combinations , resulting in isolation between hybrid and parental species [15] . Simulation results under this model have suggested that inbreeding [14] or positive selection on hybrid genotypes [9 , 14] is important for the evolution of hybrid reproductive isolation . However , past simulation efforts focused on hybrids in a tension zone , either with no spatial isolation from parental species [14] or with high migration rates from parental species [17] . To investigate the dynamics of the underdominant inversion model in situations where migration is more restricted , we simulate the underdominant inversion model in an isolated hybrid swarm scenario that is similar to our epistatic incompatibility model ( S7 Text ) . Interestingly , we find that isolation evolves frequently under this model even without positive selection ( ~40% of simulations , see S7 Text ) . These results show that , in hybrid-dominated populations , the inversion model has similar behavior to our model of selection against negative epistatic interactions ( S7 Text ) . Which mechanism of isolation is more prevalent in hybrid populations will depend on the frequency of hybrid incompatibilities of each type . Empirical evidence suggests that while underdominance can be a common isolating mechanism in plants ( reviewed in [21] ) , negative epistatic interactions may be a more common mechanism of reduced hybrid fitness in animals [24] . It is important to note several factors that may influence how common our epistatic interactions model of hybrid speciation will be in natural populations . First , our model assumes that hybrids are abundant in a population and , while this appears to be reasonably common ( see S6 Text; S9 Table ) , this is clearly not a feature of all hybrid zones . We also note that our model only represents fitness in terms of genetic incompatibilities and that hybrid populations can have lower fitness as a result of ecological or sexual selection . For example , in our simulations , we assumed random mating between hybrids and parentals . But when parental species exert negative sexual selection against hybrids , hybrid populations are significantly more likely to be outcompeted by parentals ( S10 Table ) . There is substantial variation in the mating preferences of parentals for hybrids [71] . In two species of cyprinidontiform fishes , male and female parentals mate readily with hybrids [45 , 72 , 73] , while mice discriminate against them [74] . This suggests that the likelihood of this process will depend in part on the biology of the hybridizing species . An additional consideration is that hybrid reproductive isolation is most likely to evolve during a particular window of divergence between parental species . When the fitness of hybrid populations is low ( i . e . corresponding to high levels of divergence between parental species ) , they are more prone to extinction or displacement by parentals ( S6 Fig . , S5 Text ) . This suggests that the evolution of hybrid reproductive isolation through this mechanism is most likely to occur in a period of evolutionary divergence during which species have accumulated some hybrid incompatibilities but have not diverged to the point at which hybrids are largely inviable . The most detailed work characterizing genetic incompatibilities has been between Drosophila species , where hybrids generally have substantially reduced fitness compared to parents [56 , 57 , 75] . Hybrids between several other species studied to date , however , are affected by fewer incompatibilities or incompatibilities of weaker effects [26 , 55 , 59 , 76–79] . Such groups may be more likely to form hybrid populations , and should be the focus of future empirical research . In addition , even species that currently have strong isolation may have historically produced hybrid populations , though investigating ancient hybrid speciation by the mechanism we describe would be challenging . This is because if parental and hybrid lineages have diverged substantially since the time of initial hybridization it may not be possible to determine whether or not incompatibilities were initially derived from parental genomes . It is interesting to note that reduced frequency of reproductive isolation with increasing selection on hybrids can be mitigated to some extent by an increase in the total number of hybrid incompatibility pairs . In our simulations , we see a positive relationship between the number of interactions and the probability of developing reproductive isolation , and a negative relationship between the total strength of selection on hybrids and the probability of developing reproductive isolation ( Figs . 3 and S6 ) . This tradeoff suggests that reproductive isolation can evolve between hybrid and parental populations even when the fitness of hybrids is low ( as in Figs . 3 , 4 , and S6 , keeping in mind that extinction occurs frequently when hybrid fitness is nearly zero ) . Similarly , our model is sensitive to skewed initial admixture proportions , but increasing the number of hybrid incompatibility pairs increases the probability that skewed hybrid populations will be isolated from both parental species by at least one incompatibility ( S7 Fig . ) . For example , with two incompatibility pairs , the probability of isolation from both parental species in an ancestry-skewed population ( 65% parent 1 ) was 7% while with four incompatibility pairs the probability rose to 15% . In addition , because discrete populations in a cline often span a range of admixture proportions ( e . g . [80–82] ) , it is likely that some hybrid populations will fall in the range where we predict that isolation can evolve . On the other hand , our results show that high levels of migration ( as might be observed in continuous clines ) can prevent isolation; future research should investigate the dynamics of this process in a range of hybrid zone structures . Finally , our model assumes that coevolving incompatibilities or BDM incompatibilities arising from adaptive evolution frequently occur between species . Accumulating evidence suggests that incompatibilities arising from coevolution may be common [30 , 36 , 83–86] . For example , in marine copepods , coevolution between cytochrome c and cytochrome c oxidase results in a reciprocal breakdown of protein function in hybrids [86] . In addition , the fact that many known incompatibility genes involve sexual conflict , selfish genetic elements , or pathogen defense suggests an important role for coevolution in the origin of incompatibilities [36 , 83 , 87 , 88] . Our model also applies to BDM incompatibilities that arise due to within-lineage adaptation , assuming that the fitness advantage of the derived alleles is not dependent on the parental environment . It is currently unknown whether incompatibilities are more likely to be neutral or adaptive . Though there is evidence for asymmetric selection on many hybrid incompatibilities [28 , 29 , 89] , neutrality has not been established in these cases . Anecdotal evidence supports the idea that adaptive incompatibilities are common , since many of the genes underlying hybrid incompatibilities identified so far show evidence of positive selection within lineages [90] , but the relative frequency of adaptive and neutral BDM incompatibilities awaits answers from further empirical research . Intriguingly , theoretical work also suggests that neutral BDM incompatibilities are unlikely to persist if there is gene flow between species [32] . The patterns predicted by our model are testable with empirical approaches . A large number of studies have successfully mapped genetic incompatibilities distinguishing species [25 , 26 , 41 , 56 , 57 , 79 , 91] . Ancestry at these sites can be determined in putative hybrid species , and the relative contribution of parental-derived incompatibilities to reproductive isolation can be determined experimentally . For some species , it may be possible to evaluate the dynamics of incompatibilities relative to the genetic background in experimentally generated hybrid swarms [92] . We predict that many hybrid populations exhibiting postzygotic isolation from parental species will have fixed incompatibility pairs for each parental species . Several cases of hybrid speciation report reduced fitness of offspring between parental and hybrid species consistent with the mechanism described here [6 , 16 , 53 , 93] and are promising cases for further empirical research . Strikingly , a recent study on Italian sparrows concludes that reproductive isolation between parental and hybrid species is partly due to the fixation of parental-derived incompatibilities [94] . An intriguing implication of our model is that independently formed hybrid populations between the same parental species can develop reproductive isolation from each other . The likelihood of this outcome increases with the number of incompatibility pairs . In sunflowers , empirical studies of ecologically-mediated hybrid speciation have identified multiple hybrid species derived from the same parental species [95] . It is interesting to note that selection against hybrid incompatibilities could generate the same pattern in replicate hybrid populations . In fact , this mechanism could generate a species phylogeny pattern similar to that expected from an adaptive radiation , with multiple closely related species arising in a relatively short evolutionary window . This finding is striking because our model does not invoke adaptation and suggests that non-adaptive processes ( i . e . selection against incompatibilities ) could also explain clusters of rapidly arising , closely-related species . To characterize evolution at hybrid incompatibility loci in hybrid populations without drift , we used the equations described by Karlin and others [46 , 47] to calculate changes in allele frequency as a result of two-locus selection . The frequency of gamete i at generation t is given by xi ( t ) =xi ( t-1 ) wiw-+ϵirDw14w- , ( 1 ) where ε1 = ε4 = -ε2 = -ε3 = -1 , the marginal fitness of allele i , wi=∑j=14xjwij , ( 2 ) the mean fitness of the population , w-=∑i=14∑j=14xixjwij , ( 3 ) w14is the fitness of a double heterozygote , r is the recombination rate and D is linkage disequilibrium between the two loci . These equations assume random mating , non-overlapping generations and that fitness depends only on two-locus genotype and not on whether the chromosome was maternally or paternally inherited ( i . e . wij = wji ) . To model changes in allele frequencies over time , we developed a custom R script ( available from github: https://github . com/melop/twolocusmodel ) . Iterating through the change in allele frequencies each generation as a result of selection gives the expected patterns of fixation at incompatibility loci without genetic drift ( S4 Fig . ; see also [48] ) . The deterministic two-locus model of fixation of hybrid incompatibilities does not realistically predict expected patterns in natural populations because even large populations will have some level of genetic drift . To model drift , we added multinomial sampling of N diploid individuals and recalculated allele frequencies each generation ( available from github: https://github . com/melop/twolocusmodel ) . Patterns of fixation incorporating genetic drift through multinomial sampling show similar dynamics to the model lacking genetic drift , with the exception of several equilibrium states specific to the latter ( see S5 Fig . , S2 Text ) . Exact results for more than two loci have proven difficult to obtain [96–99] . As a result , we developed a custom c++ program , called admix’em ( github: https://github . com/melop/admixem ) , to simulate more complex scenarios . The code allows one to specify the number and length of chromosomes and the genomic locations of hybrid incompatibilities and neutral markers . The current implementation assumes non-overlapping generations and diploid sexual individuals . When modeling linkage , we assume a uniform recombination rate and one recombination event per chromosome per meiosis . Unless otherwise specified , we model all pairs of hybrid incompatibility loci as unlinked . As we are interested in short-term dynamics , we do not implement mutation . Selection coefficients are assigned to particular allelic combinations according to a hybrid fitness matrix ( see S1 and S2 Figs ) . Based on each individual’s genotype at the hybrid incompatibility loci , we calculate total individual fitness w , defined as the probability of survival of that individual . Total fitness across multiple incompatibility pairs is assumed to be multiplicative . Each female mates with one randomly selected male ( but we also accommodate assortative mating , see S6 Text ) , and produces a Poisson distributed number of offspring with a mean = 2 . After selection , if the carrying capacity ( N ) is not reached , additional offspring from the same mating events are drawn from a Poisson distribution with a new mean = ( carrying capacity—current population size ) /number of females . This process is repeated until carrying capacity is reached or females have no available gametes ( set to a maximum of 10 ) . A potential concern with this approach for maintaining a constant population size is that it could artificially preserve a hybrid population that would otherwise be ephemeral by continuing to sample offspring ( up to 10 per female in our simulations ) . However , because parentals are present in each population ( see below , at 50% frequency each parental species in the initial population ) , this allows for out-competition of hybrids by parentals when hybrid fitness is low . All reported results are based on 500 replicate simulations , which were conducted for 2000 generations . In the majority of simulations ( except S3 and S4 Texts ) the hybrid population is initially colonized by 500 individuals of each parental species . Hybrid and parental populations were modeled as spatially distinct with migration parameters between them; most simulations specified one hybrid population formed between two parental populations ( S3 Fig . ) but we also simulated a stepping-stone model and a model with multiple independently formed hybrid populations ( S6 Text , S13–S14 Figs ) . In simulations with migration , the number of migrating individuals each generation was determined by drawing from a binomial distribution with a mean equal to the number of migrating individuals . Details on individual simulations and results can be found in the supporting text . Hybrid populations are considered to have evolved reproductive barriers from both parental species if they fix at least one incompatibility from each parental type; the strength of reproductive isolation between hybrids and parental species will depend on the selection coefficient and number of incompatibilities .	Understanding the origin of species is one of the central challenges in evolutionary biology . It has been suggested that hybridization could generate new species because hybrids can display novel combinations of traits that induce reproductive isolation from their parental species ( called “hybrid speciation” ) . Existing models predict that this should only occur in special cases , and indeed there have been only few well-supported examples . We describe a new model of hybrid reproductive isolation that results from selection against genetic incompatibilities in hybrids , which are predicted to be common . Simulations reveal that hybrid populations rapidly and frequently become isolated from parental species by fixing combinations of genes that hinder successful reproduction with parental species . We propose that this process could be an important mechanism for the formation of new species .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[]	2015	Reproductive Isolation of Hybrid Populations Driven by Genetic Incompatibilities
In all eukaryotes , histone variants are incorporated into a subset of nucleosomes to create functionally specialized regions of chromatin . One such variant , H2A . Z , replaces histone H2A and is required for development and viability in all animals tested to date . However , the function of H2A . Z in development remains unclear . Here , we use ChIP-chip , genetic mutation , RNAi , and immunofluorescence microscopy to interrogate the function of H2A . Z ( HTZ-1 ) during embryogenesis in Caenorhabditis elegans , a key model of metazoan development . We find that HTZ-1 is expressed in every cell of the developing embryo and is essential for normal development . The sites of HTZ-1 incorporation during embryogenesis reveal a genome wrought by developmental processes . HTZ-1 is incorporated upstream of 23% of C . elegans genes . While these genes tend to be required for development and occupied by RNA polymerase II , HTZ-1 incorporation does not specify a stereotypic transcription program . The data also provide evidence for unexpectedly widespread independent regulation of genes within operons during development; in 37% of operons , HTZ-1 is incorporated upstream of internally encoded genes . Fewer sites of HTZ-1 incorporation occur on the X chromosome relative to autosomes , which our data suggest is due to a paucity of developmentally important genes on X , rather than a direct function for HTZ-1 in dosage compensation . Our experiments indicate that HTZ-1 functions in establishing or maintaining an essential chromatin state at promoters regulated dynamically during C . elegans embryogenesis . In genomes ranging from protozoa to humans , specialized regions of chromatin are created by the local incorporation of variant histones into nucleosomes . The histone H2A variant H2A . Z is one such highly conserved variant , though the biophysical and biological function of H2A . Z incorporation into chromatin remains unresolved . Evidence from Tetrahymena suggests a function for H2A . Z in transcriptional activation due to its localization to the transcriptionally active macronucleus [1]–[3] . This function is consistent with genome-wide studies of Htz1 occupancy in S . cerevisiae ( hereafter “yeast” ) , which revealed Htz1 incorporation flanking a nucleosome-free region upstream of most genes . It has been hypothesized that H2A . Z-containing nucleosomes may contribute to transcriptional activation by being less stable than H2A-containing nucleosomes [4]–[6] . However , others have reported that H2A . Z-containing nucleosomes are in fact slightly more stable than canonical nucleosomes [7]–[9] . This seeming contradiction may have been resolved by studies examining H2A . Z in combination with the histone H3 variant H3 . 3 . In combination with histone H3 , H2A . Z nucleosomes were at least as stable as H2A nucleosomes , but the combination of H2A . Z and H3 . 3 results in highly unstable nucleosomes [4] . This instability in conjunction with H3 . 3 could facilitate timely and efficient gene activation . Indeed , in yeast cells lacking H2A . Z , the activation of genes in response to heat shock or galactose is delayed , and recruitment of RNA polymerase II and TATA-binding protein to responsive promoters is diminished [10] , [11] . H2A . Z is also required for a form of “transcriptional memory” in yeast , in which recently transcribed chromatin is retained at the nuclear membrane to allow rapid re-activation of the gene [12] . Recent high-resolution mapping of H2A . Z in human cells also revealed a positive correlation between H2A . Z occupancy and transcription , providing additional support for an H2A . Z function in transcriptional activation [13] . Despite the wealth of evidence for a positive association between H2A . Z and transcription , other genetic and cytological evidence suggests that H2A . Z also functions in gene silencing . The functional homolog of H2A . Z in Drosophila , H2Avd , is localized to both euchromatin and heterochromatin on polytene chromosomes , including the heterochromatic chromocenter [14] , [15] . By genetic criteria , H2Avd is considered to have a repressive function . H2Avd mutations are enhancers of Polycomb mutant phenotypes , suppressors of Trithorax group mutant phenotypes , and suppressors of position-effect variegation [16] . Further evidence for a repressive function is found in mice , where H2A . Z promotes heterochromatin protein HP1α binding and co-localizes with HP1 at pericentric heterochromatin [17] , [18] . In mammalian cells , mono-ubiquitylation of the H2A . Z C-terminus may distinguish “repressive H2A . Z” from “activating H2A . Z” , particularly on the silent X chromosome [19] . Even within the yeast literature , there are conflicting conclusions regarding correlation with transcriptional activity and RNA Polymerase II . One study found no correlation between Htz1 occupancy and transcription rate of the downstream gene [11] , while others reported an inverse correlation with transcription rate [20]–[22] . The resolution of these apparently contradictory activating and silencing functions could be explained by a requirement for H2A . Z in regulating the precise timing and kinetics transcription , rather than simply promoting an “on” or “off” transcriptional state . This potentiation of transcription would be especially critical during periods of dynamic transcriptional regulation , such as occurs in development and environmental responses . There is growing evidence for this hypothesis . In C . elegans , knockdown of htz-1 by RNAi caused expression of genes dependent on the FoxA transcription factor PHA-4 to be delayed [23] . Furthermore , HTZ-1 and components of the C . elegans Swr1 complex ( SSL-1 ) required for HTZ-1 deposition have been identified in genetic screens for suppressors of vulval induction , a process highly dependent on precise timing of transcriptional cascades and tightly coordinated with cell divisions [24] , [25] . Another clue to the function of H2A . Z may lie in the fact that it is required for viability in all metazoans tested [26]–[29] , but is not required for viability in single-celled yeast . A lack of H2A . Z during metazoan development typically causes defects that lead to late embryonic lethality [23] , [26] , [27] , [29] , [30] . This is consistent with expression of H2A . Z in mice , where the undifferentiated cells of the inner cell mass have low H2A . Z protein levels , with H2A . Z protein levels increasing as the cells differentiate into extraembryonic endoderm [17] . Whether H2A . Z has been associated with gene activation or repression in one study versus another may not represent a universal regulatory function for H2A . Z , but may instead be a reflection of the specific biological conditions under which the function of H2A . Z was observed in a given experiment , and the temporal resolution of the particular assays employed . In this light and with a focus on development , we used Chromatin ImmunoPrecipitation on DNA microarrays ( ChIP-chip ) , genetic mutation , and RNAi to interrogate the function of HTZ-1 during embryogenesis in C . elegans . HTZ-1 knockdown by RNAi has been previously shown to cause embryonic lethality [23] . To further characterize the function of HTZ-1 ( R08C7 . 3 ) in C . elegans development , we analyzed animals harboring a deletion in the C . elegans htz-1 gene . The mutant htz-1 ( tm2469 ) contains a deletion of 345 bp of the htz-1 gene , thereby eliminating 97 of the 140 predicted amino acids and making it a likely genetic null . The majority of homozygous htz-1 ( tm2469 ) offspring from htz-1 ( tm2469 ) /+ heterozygotes ( denoted as maternal +; zygotic − , or M+Z− ) animals are rescued from embryonic lethality by a maternal contribution of HTZ-1 . These rescued animals develop into worms exhibiting grossly normal morphology and germ cell proliferation until late adulthood ( Figure 1A–B ) . Of the M+Z− animals that reach adulthood , 80% are sterile and do not generate any embryos , instead producing unfertilized oocytes that eventually fill the uterus ( Figure 1B ) . In 20% of the rescued animals , M−Z− embryos are observed in the uterus ( Figure 1C ) . None of the embryos produced by M+Z− mothers were expelled from the uterus onto plates , indicating that the M+Z− mothers have an egg-laying defect ( Egl ) . Somewhat unexpectedly , 28% of the M−Z− embryos ( n = 32 ) progressed through embryogenesis to produce a few hatched larvae . All of these M−Z− escapers arrest at the first larval stage ( Figure 1D–E ) . The M−Z− embryos that hatched tended to arise from the first few eggs produced by M+Z− mothers , suggesting that in these animals HTZ-1 were still maternally provided at very low levels , but subsequent divisions of the germ cell precursors diluted HTZ-1 such that later embryos received a level below that required for viability . The viability and semi-fertility of the htz-1 ( tm2469 ) M+Z− offspring suggested that the maternal load of HTZ-1 received by an embryo is sufficient to allow it to reach adulthood with defects limited to germ cells and specification of cells in post-embryonic lineages , for example vulval development . To test this , we targeted the maternal complement of htz-1 mRNA using RNAi . Direct injection of dsRNA into the gonad of adult wild-type animals produced a more severe phenotype than was observed in M+Z− offspring . Instead , the RNAi phenotypes are consistent with those observed in htz-1 ( tm2469 ) M−Z− embryos . Specifically , embryonic lethality was observed for 70% of the embryos , with the remaining animals dying as larvae ( Figure 1F–G; Text S1 ) . We verified that the htz-1 dsRNA injections did not cross-react with H2A mRNA by showing that expression of a GFP-tagged version of H2A was not affected ( Figure 2A–C ) . We interpret the progression of phenotypes resulting from either RNAi treatment or genetic mutation to indicate that HTZ-1 is required for both embryogenesis and for post-embryonic development . We propose that the occasional escape from lethality occurs due to perdurance of maternal HTZ-1 protein or RNA for as long as two generations , or in the case of RNAi , a failure to completely eliminate HTZ-1 protein or message in the offspring of injected mothers ( Discussion ) . HTZ-1 RNA is abundant in the form of a maternal contribution , and remains abundant throughout the majority of embryogenesis , suggesting that the function of HTZ-1 in development is widespread [31] . To investigate the distribution of HTZ-1 protein , we generated polyclonal antisera specific to a unique peptide sequence in the C-terminal region of HTZ-1 ( Methods ) . The antibody recognized a single band of 15 kD on western blots of C . elegans protein extract , corresponding to the predicted molecular weight of HTZ-1 ( Figure S1 ) . Using these antibodies , we stained whole embryos and adults and found that HTZ-1 protein is present in all cell types throughout all stages of development . HTZ-1 protein levels are low in early embryos ( 1–12 cell ) , but increase as development progresses ( Figure 2D–F ) . HTZ-1 protein becomes detectably incorporated into chromosomes by the four-cell stage , coincident with the onset of zygotic transcription . This occurs in both wild-type and M+Z− embryos , demonstrating that zygotic transcription of htz-1 itself is not required for incorporation of HTZ-1 protein into chromatin . In wild-type adults , HTZ-1 protein is observed in both somatic and germline precursor cells ( data not shown ) . No HTZ-1 protein was observed by immunofluorescence in M+Z− adult gonads or their M−Z− embryos ( Figures G–O ) . In addition , no protein staining was observed in the offspring of animals injected with HTZ-1 RNAi ( Figure 2P–R ) . The low levels of HTZ-1 protein in young embryos , despite abundant htz-1 mRNA , suggests that much of the maternal contribution is RNA-based , with HTZ-1 protein levels controlled post-transcriptionally ( Figure 2D–F ) . Another case in which HTZ-1 protein levels do not depend on zygotic transcription can be inferred from the presence of HTZ-1 protein in the germline precursors ( P lineage ) . In these cells , HTZ-1 protein is present in chromatin at levels comparable to the surrounding somatic blastomeres , despite the repression of zygotic mRNA production in the P lineage ( Figure 2D–F ) [32] . HTZ-1 protein is also observed in the chromatin of the primordial germ cells Z2 and Z3 ( data not shown ) , which undergoes a dramatic erasure of histone H3 modifications during development [33] , [34] . To determine the genomic locations at which HTZ-1 functions , we performed ChIP-chip of HTZ-1 from extracts of wildtype N2 C . elegans embryos ( Methods ) . For detection of ChIP-enriched loci , we used DNA microarrays consisting of 50-bp oligonucleotide probes that tile across the entire genome with 86-bp start-to-start spacing ( Methods ) . Peaks of HTZ-1 binding were identified using ChIPOTle [35] . Throughout the genome , 5163 sites of HTZ-1 incorporation were found , with 85% of the peaks occurring within intergenic regions . Intergenic regions are defined as those that occur outside the boundaries defined by the translation start and stop sites of annotated transcripts or predicted genes . Under this definition , intergenic regions comprise 58% of the bases in the genome . Of the peaks within an intergenic region , 71% were within the 2-kb upstream of an annotated translation start site , 25% were within 2-kb of the translation stop , and only 4% were greater than 2-kb upstream of a translation start site . Among the 15% of peaks found to occur within an annotated transcription unit , most occurred near the 5′ end ( median +545 bp downstream of the annotated translation start site ) . Therefore , like yeast Htz1 , C . elegans HTZ-1 is preferentially incorporated into intergenic regions , specifically at promoters ( Figure 3A ) . We next investigated whether HTZ-1 was incorporated specifically at sites of transcriptional initiation . The majority of transcription initiation sites are not well-annotated in C . elegans , due in part to the prevalence of trans-splicing [36] . Therefore as a proxy for transcription initiation sites , we plotted HTZ-1 binding relative to annotated translation start codons . On average , the peak of HTZ-1 incorporation occurs just upstream of the translation start codon ( Figure 3B ) , which we interpreted to indicate incorporation at or near sites of transcription initiation . To further test whether the observed signal represents sites of transcription initiation , we took advantage of a unique feature of the C . elegans genome . Approximately 15% of C . elegans genes are predicted to reside in operons that are transcribed as a large polycistronic pre-mRNA , which is then trans-spliced into mRNAs for the individual genes [37] . We plotted HTZ-1 incorporation relative to the first gene in operons , where transcription is expected to initiate , and also plotted incorporation relative to internal genes , where transcription is not expected to initiate . Indeed , HTZ-1 incorporation is generally observed upstream of the first gene in an operon , and does not generally occur upstream of internal genes ( Figure 3C ) , indicating that C . elegans HTZ-1 is incorporated primarily at or near sites of transcription initiation . We also observed some important exceptions to this general rule , which are discussed below . Currently most C . elegans operons are identified primarily by two criteria: the appearance of two or more genes in close proximity that are transcribed on the same strand , and the isolation of a downstream RNA transcript with an SL2 trans-spliced leader [38] , [39] . In this way , a total of 1118 putative operons have been identified ( genome release ws170 ) . However , these criteria are imperfect , and do not provide information about genes that may be regulated both as part of an operon and by their own independent promoter . Independent transcription events within operons have been difficult to detect because the 5′ ends of mRNAs produced by either trans-splicing of a poly-cistronic mRNA or an independent transcription event are not readily distinguishable . To identify genes that are likely to be regulated both as part of an operon and individually , we examined incorporation of HTZ-1 at internally encoded genes of annotated operons . Overall , 75% of operons contained at least one HTZ-1 peak . A gene within an operon was more than twice as likely as a non-operon gene to have an HTZ-1 peak at its promoter ( Figure 4A–B ) . Of operons containing at least one site of HTZ-1 incorporation , 85% contained a peak upstream of the first gene , as one might expect . However , 49% of operons with HTZ-1 incorporation at the first gene also exhibited an internal peak of HTZ-1 incorporation . This strongly suggests internal transcription start sites at 416 ( 37% ) of the currently annotated operons ( Figure 4B , Table S2 ) . Because some operons contain multiple internal HTZ-1 peaks , this represents a total of 455 putative independently regulated genes within annotated operons . This is likely to be an underestimate , since the HTZ-1 localization data is derived only from embryonic extracts , meaning that genes and operons regulated specifically in adults or germ cells are not represented . The unexpectedly high number of individually regulated genes within operons may to some extent reflect a mis-annotation of operons based on traditional criteria . To show that internal HTZ-1 incorporation can occur at verified operons , we examined CEOP1456 , one of the first characterized operons , supported by cistronic RNA evidence [37] , [40] . In this well-characterized operon , both HTZ-1 and RNA Polymerase II occupy the chromatin immediately upstream of the internal kin-10 gene , strongly suggesting independent regulation ( Figure 4C ) . Recently , differential regulation of genes driven by internal operon promoters was demonstrated using a GFP reporter assay [41] . We find that one-third of these internal promoters are occupied by HTZ-1 in embryos ( Table S2 ) . A time-course of the early embryonic transcription [31] provides evidence that genes within operons that contain multiple sites of HTZ-1 incorporation exhibit differential expression ( Figure S2 ) . In contrast to yeast , in which Htz1 is incorporated into nearly every promoter [42] , our ChIP-chip data indicate that HTZ-1 is incorporated into the promoters of only 23% of C . elegans genes ( Methods ) . To determine what might be held in common among the particular subset of genes that were occupied by HTZ-1 , peaks were annotated to gene promoters , assigned Gene Ontology ( GO ) terms according to the nearest downstream gene , and evaluated with GO::TermFinder [43] . To avoid ambiguous gene assignments , only peaks annotated to unidirectional promoters or within coding regions were used in the input set . We found that GO terms associated with metazoan development and positive regulation of growth were strongly over-represented among HTZ-1 bound genes , while no overrepresented GO term was associated with the non-HTZ-1 bound genes ( Table 1 , Table S1 ) . This finding suggests that HTZ-1 functions preferentially at the promoters of genes essential for growth and development . We next sought to examine the relationship between HTZ-1 occupancy at promoters and transcriptional activity during embryogenesis . We found that , in general , transcript levels [31] were positively correlated with HTZ-1 promoter occupancy ( Figure 5A; Spearman rank-order correlation = 0 . 35 ) . A positive correlation was also observed between RNA levels reported by a completely independent study [44] and HTZ-1 occupancy ( Figure S3 ) . Despite the positive overall correlation between occupancy and transcript levels , the relationship becomes negative at promoters of genes with very high transcript abundance ( Figure 5A ) . This observation is consistent with a general loss of nucleosomes upstream of highly transcribed genes [45] , [46] . We sought to establish a more direct link between HTZ-1 occupancy and transcription , so we determined the genome-wide occupancy of RNA polymerase II by ChIP-chip using an antibody specific to the C-terminal domain heptapeptide ( 8WG16 , Methods ) . At gene promoters , HTZ-1 occupancy was strongly correlated with RNA Polymerase II occupancy ( Figure 5B ) . In fact , the correlation was stronger than that observed between HTZ-1 occupancy and transcript levels ( Spearman rank-order correlation = 0 . 57 ) . Consistent with the correlation with transcript levels , at the promoters most highly occupied by RNA Polymerase II , the correlation with HTZ-1 occupancy was negative . Again , this observation is likely due to general nucleosome loss at the promoters of highly transcribed genes , for example those that encode the histone and ribosomal proteins [45] , [46] . Temporal regulation gene expression during embryogenesis may also affect this correlation and is considered in the Discussion . To further illustrate the relationship between HTZ-1 localization and polymerase occupancy , the 4650 genes with HTZ-1 incorporated into their promoters were aligned according to their translation start site , and average RNA polymerase II occupancy relative to the start site was plotted ( Figure 5C ) . HTZ-1-occupied promoters were on average occupied by RNA Polymerase II , whereas genes lacking HTZ-1 were not ( Figure 5D ) . At promoters occupied by HTZ-1 , the average peak of HTZ-1 occupancy was at negative 12 bp relative to the translation start , while the average peak of RNA Polymerase II occupancy was slightly upstream at negative 98 bp ( Discussion ) . An important consideration in interpreting these relationships is that our experiments were performed using extract derived from a mixed population of embryos composed of many cell types . Therefore , our results are a projection of HTZ-1 occupancy throughout embryogenesis and represent a temporal and spatial average of the relationship between HTZ-1 , RNA Polymerase II , and transcription ( Discussion and Text S1 ) . To examine if HTZ-1 occupied promoters direct a stereotypic pattern of gene expression , we compared HTZ-1 occupancy , RNA Polymerase II occupancy , and transcription at each gene using a published time-course of transcript abundance during embryonic development [31] . Promoters occupied by HTZ-1 were clustered according to our RNA Polymerase II promoter occupancy data and the change in transcript abundance relative to the onset of zygotic transcription . To avoid ambiguity , transcripts that were highly maternally loaded ( >100 parts per million ( ppm ) ) were removed from analysis . Consistent with the aggregate analysis , RNA Polymerase II is abundant at most HTZ-1 occupied genes ( Figure 6A ) , while promoters at which HTZ-1 is not incorporated generally lack RNA Polymerase II ( Figure 6B ) . However , a large proportion of genes downstream of promoters occupied by both HTZ-1 and RNA polymerase II produce low transcript levels ( Figure 6A ) , and conversely some genes produce high transcript levels despite low levels of HTZ-1 and RNA polymerase II at their promoters ( Figure 6B; Discussion ) . Therefore , while HTZ-1 is strongly linked to RNA Polymerase II occupancy in aggregate , HTZ-1 bound promoters do not specify a stereotypic pattern of transcriptional regulation during development , suggesting that RNA polymerase occupancy and transcript levels are decoupled at some promoters . The sex chromosomes are often sites of specialized chromatin , harboring unique histone variants and chromatin modifications . To determine whether HTZ-1 was differentially localized to X , we co-stained embryos with anti-HTZ-1 in combination with either anti-DPY-27 , which marks the X chromosomes in embryos of greater than about 30 cells ( Figure 7A–D ) , or anti-MES-4 , which marks the autosomes but not X chromosomes in early embryos ( Figure 7E–H ) . In embryos that had initiated somatic dosage compensation , HTZ-1 incorporation was noticeably reduced on the X chromosomes , which was marked by DPY-27 staining ( Figure 7A–D ) . However , co-staining with MES-4 revealed HTZ-1 under-representation on X even before the onset of somatic dosage compensation ( Figure 7E–H ) . These results indicate that in both early-stage embryos before the onset of dosage compensation and late-stage C . elegans embryos after dosage compensation is established , there is significantly less HTZ-1 associated with the X than with autosomes . We next aimed to ensure that reduction of HTZ-1 we observed on the X chromosome by immunofluorescence was not due to epitope exclusion . This concern was prompted by reports that mammalian H2A . Z on the inactive X chromosome is ubiquitylated , and that this modification can interfere with recognition by antibodies raised against a C-terminal peptide epitope [19] . Our antisera were also raised against a C-terminal peptide . To address this concern , we co-stained embryos expressing a HTZ-1:YFP transgene with anti-DPY-27 and anti-YFP antibodies . We observed a similarly reduced YFP signal coincident with regions of DPY-27 signal . This serves as independent evidence that within in the same nucleus , X chromatin has less HTZ-1 incorporation than autosomes ( Figure S4 ) . Two explanations for the under-incorporation of HTZ-1 on X immediately come to mind . One is that less HTZ-1 is incorporated on X as part of the C . elegans dosage compensation mechanism . A second explanation , which we favor for the reasons presented below , is that genes important for development , whose promoters tend to be occupied by HTZ-1 , are under-represented on the X chromosome [47]–[50] . To distinguish these possibilities , we examined at high resolution the sites of HTZ-1 incorporation on the X chromosomes relative to the autosomes ( Figure 7I ) . One possible variation of the “dosage compensation” hypothesis predicts that sites of HTZ-1 incorporation are excluded or diminished on X as a consequence of the transcriptional repression imposed by the DCC . In this case , one would expect HTZ-1 occupancy on X to be excluded from sites occupied by the dosage compensation machinery [51] . Contrary to this prediction , we found strong co-localization of HTZ-1 incorporation and DCC binding , such that over 62% of HTZ-1 peaks are coincident with a DPY-27 peak ( Figure 7J–K , Figure S7 ) . The highly concordant binding pattern of HTZ-1 and the DCC on X would appear to rule out a function for HTZ-1 as a direct negative regulator of DCC binding to autosomes ( Discussion ) . We then considered the possibility that HTZ-1 incorporation is in fact a requirement for the loading of the DCC onto X . However , there are far more sites of DCC localization on X than HTZ-1 incorporation , meaning that most DCC-bound loci are not sites of HTZ-1 localization . For example , while both HTZ-1 and DPY-27 are incorporated at the X-linked dpy-23 promoter , HTZ-1 is not incorporated at the well-characterized DCC recruitment site rex-1 , which occurs just 5 kb downstream of dpy-23 ( Figure 7L ) [51] , [52] . We also examined in more detail apl-1 and lin-15 , two of the few genes known with some certainty to be dosage compensated [53] , [54] . Although the DCC and RNA Polymerase II are present at both loci , HTZ-1 is present at lin-15 , but not at apl-1 ( Figure S5 ) , reinforcing the interpretation that HTZ-1 is not required for dosage compensation . Conversely , the under-representation of HTZ-1 on X is not dependent on the dosage compensation process , because it is evident in the germline and before the onset of somatic dosage compensation ( Figure 7H ) . The alternative “developmental gene” hypothesis for the under-incorporation of HTZ-1 on X is based on the observation that only about half as many essential genes occur on X as would be expected to occur on an autosome of the same size ( 201 vs . 562 expected , wormbase release ws170 ) [47]–[50] . This hypothesis predicts that there would be fewer sites of HTZ-1 incorporation on X , but that those that do occur on X behave like those on autosomes . The X harbored 495 HTZ-1 peaks , about half the number expected from a hypothetical autosome with the size and gene density of X ( p-value = 2 . 05×10−43 and 8 . 09×10−93 respectively , Figure 7I ) . There was no significant difference between the median height and width of HTZ-1 peaks on X ( z-score = 2 . 28 and 774 bp , respectively ) as compared to the median height and width of HTZ-1 peaks on autosomes ( z-score = 2 . 20 and 860 bp , respectively ) ( Figure S8 ) . This indicates that while HTZ-1 incorporation occurs at fewer loci on X , where it does occur the degree of incorporation is the same as the autosomes . The most parsimonious explanation for the under-representation of HTZ-1 on the X is that the types of genes that require HTZ-1 for proper regulation are themselves under-represented on the X chromosome . Nonetheless , HTZ-1 is likely to have an indirect function in the dosage compensation process by affecting the regulation of genes required for dosage compensation . Strong HTZ-1 incorporation is observed at the promoters of sdc-1 , sdc-2 , sdc-3 , dpy-27 , mix-1 , and dpy-30 , all of which are required for dosage compensation . Although any number of complex scenarios involving a direct relationship between HTZ-1 and the canonical dosage compensation process remain possible , we interpret the under-representation of sites of HTZ-1 localization on X to be a simple consequence of the under-representation of germline and developmentally important genes on the X chromosome ( Discussion ) . The C . elegans genome has been shaped by the developmental programs it must coordinately execute . The general requirement of H2A . Z for development in metazoans suggests a function for H2A . Z in establishing or maintaining a specialized chromatin state at developmentally regulated promoters [27]–[30] , [55] . In this study , we have established that HTZ-1 is incorporated upstream of genes critical for development , and that maternally provided HTZ-1 is sufficient for C . elegans embryogenesis . We infer by the progressively deteriorating phenotype suffered by offspring lacking HTZ-1 that HTZ-1 is required for both embryogenesis and post-embryonic development . The function of HTZ-1 in pharyngeal organogenesis may provide a model for the mechanism by which HTZ-1 is generally required for C . elegans development . The development of the pharynx relies on precise temporal regulation of transcription activation , mediated in part by PHA-4 , a FoxA transcription factor [56] , [57] . HTZ-1 depletion enhances defects in pharyngeal organogenesis associated with loss of PHA-4 , and activation of PHA-4-dependent promoters is delayed in the absence of HTZ-1 [23] . This is reminiscent of the delay of yeast GAL gene activation in the absence of Htz1 [10] , and indicates a conserved role for H2A . Z in facilitating timely gene expression . Previous genome-wide studies in yeast and other organisms have reached differing conclusions regarding the relationship between H2A . Z , RNA Polymerase II , and transcription [11] , [12] , [15] , [16] , [20]–[22] , [42] , [58] , [59] . Functional divergence between yeast Htz1 and metazoan homologs are a possible source of the discrepancy . Consistent with this , C . elegans HTZ-1 is only 61% identical to yeast Htz1 , but 77% identical to Drosophila H2Avd , and 83% identical to mouse or human H2A . Z ( Figure S6 ) . In C . elegans , we found that HTZ-1 is incorporated specifically at promoters , where its occupancy is strongly correlated with RNA polymerase II occupancy , and to a lesser degree with transcript levels ( see Text S1 ) . This suggests that RNA polymerase II is present at some HTZ-1 occupied promoters without being linked to a corresponding increase in transcripts . One possible explanation is pausing of RNA polymerase II near initiation sites . This phenomenon is common in human and Drosophila cells [15] , [60]–[62] but has not yet been established to occur in C . elegans . The 8WG16 RNA polymerase II antibody we used is probably not the appropriate choice for making conclusions about RNA Pol II pausing , because the antibody recognizes primarily the unphosphorylated RNA Pol II CTD , but it is known to have some cross-reactivity with both CTD-Ser5P and CTD-Ser2P . RNA Polymerase II pausing would be more appropriately examined with an independent , non-C-terminal domain RNA Pol II antibody or a CTD-Ser5P specific antibody . Nonetheless , using the 8WG16 antibody , we observed a small number of genes ( about 300 , or ∼1 . 5% ) with promoter-restricted RNA Polymerase II . A recent genome-wide study of the Drosophila H2A . Z homolog at mononucleosome resolution reported that an H2A . Z-containing nucleosome was often positioned just downstream of a paused RNA polymerase II [15] . Although we did not observe any relationship , positive or negative , between HTZ-1 occupancy and this putative paused state , peak HTZ-1 occupancy occurs about 80 bp downstream of peak RNA Pol II occupancy at promoters ( Figure 5C ) . Thus , the putative poised state may in some cases be facilitated by HTZ-1 , and could contribute to the efficient and timely activation of developmental promoters . Indeed , our data does not formally exclude the possibility that H2A . Z functions to dampen transcription [63] . In Drosophila and mammalian cells , H2A . Z plays a role in gene silencing by participating in the assembly of heterochromatin [64] , [65] . While another study of C . elegans HTZ-1 argues against a repressive role [23] , and we observe high levels of expression from many genes that contain HTZ-1 at their promoters , we cannot exclude the possibility that transcription at these loci would be even higher in the absence of HTZ-1 . One key question for future studies concerns how H2A . Z is directed to developmental promoters . Sequence-specific transcription factor binding at promoters is likely an important driver of Swr1-mediated H2A . Z incorporation [23] , [66] . At the human p21 promoter , sites of p53 binding are occupied by H2A . Z and p400 ( a human Swr1 homolog ) and this enrichment is dependent on p53 binding [13] . In C . elegans , association of HTZ-1 with pharyngeal promoters is dependent upon the presence of promoter PHA-4 motifs [23] . This requirement of PHA-4 for HTZ-1 association may be one specific example of the general mechanism underlying the specificity of HTZ-1 for developmental promoters . Studies in yeast implicate histone tail acetylation as another important factor . Histone H4K16 acetylation is a prerequisite for Htz1 association near yeast telomeres [16] , [67] . Yeast Htz1 recruitment is reduced in the absence of Bdf1 , a bromodomain containing protein that binds acetylated histone tails , and GCN5 , a histone acetyltransferase that acetylates Histone H3 tails [11] . The NuA4 histone acetyltransferase complex , which interacts with Bdf1 and is responsible for bulk H4 acetylation and acetylation of Htz1 itself , shares multiple non-catalytic components with the Swr1 complex [11] , [68]–[70] . Nucleosome free regions ( NFRs ) at promoters may also play a role . Htz1 was deposited at sites flanking NFRs , which often harbor 22-nt motif that contained a Reb1 transcription factor binding site [20] . Insertion of this motif at an ectopic location was sufficient for NFR formation and flanking Htz1 incorporation . The incorporation of HTZ-1 at sites of transcription initiation suggests that HTZ-1 may be useful for identifying previously unannotated promoters . Our observation of HTZ-1 incorporation upstream of subsequent genes within operons implies the existence of independently regulated internal promoters in at least one-third of all currently annotated operons . Alternatively , operons may be less prevalent than the current genome annotation indicates . Indeed , a recent publication found evidence for functionally distinct internal promoters at 66 out of 238 ( 27% ) downstream operon genes tested [71] , a proportion of operons similar to which we found internal HTZ-1 incorporation . Additionally , transcript evidence from a published time-course during early development [31] provides evidence for independent regulation of some internal operon genes ( Figure S2 ) . Immunofluorescence and ChIP-chip experiments reveal a significant under-incorporation of HTZ-1 on the X chromosome relative to the autosomes . We explored three explanations for this under-representation: differential detection of the HTZ-1 protein specifically on the X; a function for HTZ-1 in dosage compensation; or an under-representation of developmentally important genes , which tend to be HTZ-1 targets , on X [47]–[50] . The first possibility is reasonable because mammalian H2A . Z can be ubiquitylated on its C-terminus , and this mark distinguishes H2A . Z incorporated on the heterochromatin and silent X chromosome [19] . The C-terminal residues are conserved in C . elegans HTZ-1 , and include the antigen to which the antibody was raised . However , analysis of HTZ-1::YFP localization , in which detection would not be affected by modification state , indicates that under-representation on X is not due to epitope occlusion by ubiquitylation or any other cause ( Figure S4 ) . The second possibility concerns C . elegans dosage compensation , during which the two hermaphrodite X chromosomes undergo chromosome-wide reduction in expression to match the output of the single male X chromosome [54] . While we observed a high degree of overlap between sites of HTZ-1 incorporation and sites of DPY-27 binding , the converse did not hold true . Many sites of DCC binding occur in areas of no HTZ-1 incorporation , including known recruitment sites such as the rex-1 locus . This suggests that Dosage Compensation Complex binding does not require HTZ-1 . We interpret the extensive co-localization on X to indicate independent functions of HTZ-1 and the DCC , both of which act upstream of genes active during embryogenesis . This interpretation is further supported by very few instances of overlap between HTZ-1 and DPY-27 at sites away from promoters . For example , only nine HTZ-1 peaks on X overlap with the 219 DPY-27 peaks found in the region downstream of genes . Of the nine overlaps , all may be explained as having promoter function: six occurred in regions near another gene promoter , and the remaining three were bound by RNA Polymerase II despite the absence of a gene annotation . Finally , HTZ-1 is under-incorporated on X both prior to and subsequent to the activation of somatic dosage compensation during embryogenesis . The simplest explanation for the under-representation of HTZ-1 on X is that during embryogenesis HTZ-1 and the DCC both tend to bind at the transcription initiation sites of active genes important for development ( Figure 7K ) . Genes essential for development are approximately 2-fold under-represented on the X [47] , [48] , [50] , which is consistent with the approximately 2-fold under-incorporation of HTZ-1 on X . Although there are fewer sites of HTZ-1 incorporation on X , the individual sites of HTZ-1 incorporation on X do not differ in any quantifiable way from sites of incorporation on autosomes ( Figure S8 ) . A ChIP-chip or ChIP-seq experiment revealing similar under-representation of HTZ-1 on the male X would provide further evidence against the direct involvement of HTZ-1 in dosage compensation . As stated in the results , HTZ-1 is likely to have an indirect function in dosage compensation and many other developmental processes by virtue of its incorporation at a wide spectrum of developmentally important genes . Nonetheless , our results do not completely rule out a direct positive or negative role for HTZ-1 in dosage compensation complex targeting or function . For example , the HTZ-1 on X could be post-translationally modified differently from HTZ-1 on autosomes [19] , [72] , [73] , thereby conferring a distinct function for HTZ-1 depending on where in the genome it is incorporated . One challenge in understanding H2A . Z function is integrating very diverse types of data , each of which lends clues to H2A . Z function but has its own limitations . For example , our experiments were performed in an unsynchronized population of embryos composed of multiple cell types . Therefore , the results presented here are a static projection of the dynamic activation and repression events that are occurring at gene promoters . How this might be manifested in our dataset can be illustrated by considering a previous time-course study of HTZ-1 at the myo-2 promoter [23] . HTZ-1 was not incorporated into the myo-2 promoter when it was repressed , but was transiently incorporated at the onset of transcription . HTZ-1 was then lost as myo-2 became highly expressed later in development . In a temporal projection of these results , as occurs in our dataset , it appears as if HTZ-1 was very weakly incorporated ( if at all ) into the myo-2 promoter ( Text S1 ) . Our genome-wide data indicates a preferential incorporation of HTZ-1 at developmentally dynamic promoters , and a loss of HTZ-1 at very highly transcribed genes . We infer that the general conclusions implied by the previous time-course study conducted at the myo-2 locus are now extended to the entire genome by our data [23] . The biophysical properties of nucleosomes containing H2A . Z provide clues about how H2A . Z could facilitate precise , coordinated developmental transcriptional programs . Incorporation of H2A . Z into an otherwise canonical nucleosome appears to have slight stabilizing effects , but incorporation of H2A . Z into nucleosomes containing H3 . 3 , a mark of active transcription , is reported to cause instability [4] , [5] . Furthermore , H2A . Z incorporation may alter associations between the histone proteins within the nucleosome , regulating the formation of higher-order chromatin fibers [18] , [74] . The H3 . 3 mediated instability could allow H2A . Z to facilitate or maintain a nucleosome-free region at promoters upon activation , while the effect of H2A . Z on higher-order structures could promote the maintenance of transcription by promoting more precise nucleosome positioning at promoters or by promoting the assembly of a specialized higher-order chromatin state [4] , [15] , [18] , [21] , [42] , [75] . In this way , HTZ-1 may aid the C . elegans embryonic genome in executing rapid transitions between quiescence and activity as developmental programs are executed . The RNA polymerase II monoclonal antibody 8WG16 was obtained from Covance . A mouse ascites polyclonal anti-HTZ-1 antibody was made to the HTZ-1 specific C-terminal peptide ( N-PGKPGAPGQGPQ-C ) by Invitrogen-Zymed . Rabbit DPY-27 polyclonal antibodies used for immunofluorescence were generously provided by Dr . B . J . Meyer ( UC Berkeley ) . Rabbit polyclonal MES-4 antibodies were generously provided by Dr . S . Strome ( UC Santa Cruz ) . AlexaFluor donkey anti-mouse 488 IgG ( Invitrogen A-21202 ) , AlexaFluor donkey anti-rabbit 594 IgG ( Invitrogen A-21207 ) were used as secondary antibodies for immunofluorescence . Anti-mouse HRP and ECL Plus ( Amersham ) were used for western blot visualization . ChIP-chip analysis was performed in the N2 Bristol strain . htz-1 ( tm2469 ) was obtained from Shohei Mitani and balanced over nT1 ( qIs51 ) IV , V . KW1665 ( htz-1 ( tm2469 ) IV/nT1 ( qIs51 ) IV , V ) is maintained by selecting GFP-positive heterozygotes . All strains were cultured under standard conditions [76] at 20°C using E . coli strain OP50 or HB101 as a food source . Adult hermaphrodites gravid with embryos were dissected in 1× PBS ( 137 mM NaCl , 2 . 7 mM KCl , 8 mM Na2HPO4 and 2 mM KH2PO4 ) on a slide . Paraformaldehyde was then added to 5% . The slide was incubated at room temperature for 2 minutes with a cover slip in place , and placed on dry ice for approximately 20 min . The cover slip was removed rapidly with a razor , and the slide was then placed into 95% ethanol for 2 minutes , followed by incubation in PBST ( 1× PBS+0 . 1% Tween-20 ) for 30 minutes . Slides were incubated with primary antibody at 1∶500 ( HTZ-1 ) or 1∶100 ( DPY-27 and MES-4 ) dilution overnight and with secondary antibody ( 1∶500 ) for approximately 3 hours . 4′ , 6-diamidino-2-phenylindole ( DAPI ) was used to stain DNA . Slides were mounted using ProLong® Gold antifade reagent ( Invitrogen P36934 ) . Staining was visualized using a Leica DMRXA microscope outfitted with a Cooke Sensicam . Capture and analysis of immunofluorescence images was performed using either Volume Scan ( Vaytek ) and Image-Pro Plus ( Media Cybernetics ) or SimplePCI . 2 ( Hamamatsu Corporation ) imaging software . Nomarski DIC microscopy imaging was performed with a Leica DMRXA microscope and SimplePCI . 2 software . dsRNA was generated by in vitro transcription reaction using a Promega RiboMAX Large Scale RNA Production Systems T7 kit ( Promega #P1300 ) . Direct injection of concentrated dsRNA into adult gonads was required to obtain significant levels of embryonic lethality and larval arrest , as standard feeding and soaking methods did not result in sufficient depletion of the maternal HTZ-1 . 1 . 2 µg/µL htz-1 dsRNA was injected into young adult eri-1 ( mg366 ) or N2 animals . The injected animals were allowed to recover and lay embryos overnight , then transferred to a new plate for collection and phenotypic scoring of affected embryos for 9–12 hrs . Following the 9–12 hour period , the adults and the embryos in utero were dissected and processed for immunofluorescence . The phenotypes of hatched larvae were observed and analyzed by DIC light microscopy 2–3 days after hatching . N2 animals were used for phenotype counts and staining experiments; eri-1 ( mg366 ) animals were used for DIC RNAi phenotype experiments . Embryos were prepared by bleaching from gravid N2 adults grown in S-basal media liquid culture . Live embryos were cross-linked using 2% formaldehyde for 30 minutes at room temperature followed by quenching with 125 mM glycine for 15 minutes . Embryos were then washed twice with M9 Buffer , once by ChIP buffer , and frozen at −80°C . Extracts were prepared by resuspending embryo pellets in 1 volume ChIP Buffer ( 50 mM HEPES-KOH pH 7 . 5 , 300 mM NaCl , 1 mM EDTA pH 8 . 0 , 1% TritonX-100 , 0 . 1% sodium deoxycholate , 10% glycerol , protease inhibitors ( Calbiochem ) ) , followed by dounce homogenization ( 50× ) and sonication ( 4× , 1 s on , 0 . 5 s off , at 20% amplitude on ice ) using a Branson Digital Sonifier 450 . In a volume of 500 µL , 2 mg extract was used for each ChIP . 100 mg ( 5% ) of the extract was set aside as “Input” and 400 µL elution buffer ( 0 . 1 M NaHCO3 , 1% SDS ) was added . Two ( anti-RNA Pol II ) or six ( anti-HTZ-1 ) µg of antibody was added to each IP sample and incubated overnight at 4°C . Immune complexes were purified with 10 µL protein-A sepharose ( Amersham ) and washed 5 minutes with 1 . 5 mL of each of the following solutions: ChIP Buffer , ChIP Buffer with 500 mM NaCl , ChIP Buffer with 1 M NaCl ( HTZ-1 IPs only ) , LiCl solution ( 10 mM Tris-HCl pH 8 . 0 , 250 mM LiCl , 0 . 5% NP-40 , 0 . 5% sodium deoxycholate , 1 mM EDTA ) , and TE ( 10 mM Tris-HCl pH 8 . 0 , 1 mM EDTA ) . Samples were treated with 20 µg RNAse for 30 minutes at 37°C . IP samples were eluted twice with 200 µL elution buffer . NaCl was added to 200 mM and crosslinks were reversed by incubation overnight at 65°C . DNA was purified using Zymo DNA purification columns and amplified using LM-PCR [51] . Microarrays used were previously described ( GEO GPL4614 and GPL4619; [51] ) . Four independent HTZ-1 ChIP biological replicates were performed , one of which was a dye-swap ( ChIP 4 ) . RNA Polymerase II ChIPs were performed from extracts used for HTZ-1 ChIPs 1 and 2 . DPY-27 and “no antibody” datasets were published previously ( GEO GSE6739; [51] ) . HTZ-1 , RNA Polymerase II , and no antibody raw intensities were normalized by median centering log2 ratios ( IP/input ) . Normalized log2 ratios from each experiment were converted to standardized z-scores and combined by taking the median of experiments . Raw data for HTZ-1 and RNA Polymerase II can be found at NCBI GEO accession number GSE10201 . Peaks were derived using a Perl implementation of ChIPOTle ( https://sourceforge . net/projects/chipotle-perl/ ) [35] using a window size of 500 bp , step size 86 bp , at a Bonferroni corrected p-value of 1×10−9 . Any HTZ-1 peaks overlapping a peak found in the mock “No antibody” IP were removed from analysis . Peaks were annotated using Wormbase genome release 120 ( Table S3 ) . Maximum probe centers of peaks were annotated either to an intergenic or coding region , exclusively . Annotation distribution statistics were calculated using an unpublished C . elegans implementation of Cis-Element Annotation Software ( CEAS ) ( [77] , X . Shirley Liu , unpublished ) . Genome browser views were generated using the UCSC genome browser ( http://genome . ucsc . edu ) , using the ce2 ( March 2004 ) /ws120 genome build . Analysis for overrepresentation of Gene Ontology terms was done with GOTermfinder [43] , accessed November 11 , 2007 at http://go . princeton . edu/ . The figure of 23% of C . elegans genes being incorporated with HTZ-1 is based on annotating HTZ-1 peaks to the closest promoters and coding regions of 4650 genes ( 23 . 4% ) . In the case of bidirectional promoters ( 1800 ) , both genes downstream were counted . Therefore , this number may be an overestimate if HTZ-1 functions only at one of the two genes in these cases . HTZ-1 or RNA Polymerase II occupancy at each gene promoter was scored by averaging all probes within a 1-kb window centered on translation start sites . Transcript abundance data was obtained from a published study [31] and compared by averaging the last 3 timepoints from this study . To avoid ambiguity from maternally loaded RNAs , genes with high maternal transcript abundance ( >100 parts per million ) were removed from the clustering analysis . Each time point was divided by the 0 minute time point ( the onset of zygotic transcription ) and log2 transformed . K-means clustering ( k = 6 , 1000 iterations , similarity metric = spearman rank correlation ) was performed using Cluster 3 . 0 [78] and visualized using Treeview [79] .	To fit within a cell's nucleus , DNA is wrapped around protein spools composed of the histones H3 , H4 , H2A , and H2B . One spool and the DNA wrapped around it are called a nucleosome , and all of the packaged DNA in a cell's nucleus is collectively called “chromatin . ” Chromatin is important because it modulates access to information encoded in the underlying DNA . Spools with specialized functions can be created by replacing a typical histone component with a variant version of the histone protein . Here , we examine the distribution and function of the C . elegans histone H2A variant H2A . Z ( called HTZ-1 ) during development . We demonstrate that HTZ-1 is required for proper development , and that embryos are dependent on a contribution of HTZ-1 from their mothers for survival . We mapped the location of HTZ-1 incorporation genome-wide and found that HTZ-1 binds upstream of 23% of genes , which tend to be genes that are essential for development and occupied by RNA polymerase . Fewer sites of HTZ-1 incorporation were found on the X chromosome , probably due to an under-representation of essential genes on X rather than a direct role for HTZ-1 in X-chromosome dosage compensation . Our study reveals how the genome is remodeled by HTZ-1 to allow the proper regulation of genes critical for development .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "genetics", "and", "genomics/genomics", "molecular", "biology/histone", "modification", "developmental", "biology", "molecular", "biology/transcription", "initiation", "and", "activation", "genetics", "and", "genomics/gene", "expression", "genetics", "and", "genomics/epigenetics", "molecular", "biology", "molecular", "biology/chromatin", "structure" ]	2008	The Genomic Distribution and Function of Histone Variant HTZ-1 during C. elegans Embryogenesis
Proper functioning of working memory involves the expression of stimulus-selective persistent activity in pyramidal neurons of the prefrontal cortex ( PFC ) , which refers to neural activity that persists for seconds beyond the end of the stimulus . The mechanisms which PFC pyramidal neurons use to discriminate between preferred vs . neutral inputs at the cellular level are largely unknown . Moreover , the presence of pyramidal cell subtypes with different firing patterns , such as regular spiking and intrinsic bursting , raises the question as to what their distinct role might be in persistent firing in the PFC . Here , we use a compartmental modeling approach to search for discriminatory features in the properties of incoming stimuli to a PFC pyramidal neuron and/or its response that signal which of these stimuli will result in persistent activity emergence . Furthermore , we use our modeling approach to study cell-type specific differences in persistent activity properties , via implementing a regular spiking ( RS ) and an intrinsic bursting ( IB ) model neuron . We identify synaptic location within the basal dendrites as a feature of stimulus selectivity . Specifically , persistent activity-inducing stimuli consist of activated synapses that are located more distally from the soma compared to non-inducing stimuli , in both model cells . In addition , the action potential ( AP ) latency and the first few inter-spike-intervals of the neuronal response can be used to reliably detect inducing vs . non-inducing inputs , suggesting a potential mechanism by which downstream neurons can rapidly decode the upcoming emergence of persistent activity . While the two model neurons did not differ in the coding features of persistent activity emergence , the properties of persistent activity , such as the firing pattern and the duration of temporally-restricted persistent activity were distinct . Collectively , our results pinpoint to specific features of the neuronal response to a given stimulus that code for its ability to induce persistent activity and predict differential roles of RS and IB neurons in persistent activity expression . Working memory reflects the temporary storage of information that is necessary for immediate decisions/actions . Delay-period activity , which corresponds to neural activity that persists after the end of the initiating stimulus , represents the cellular correlate of working memory [1] , [2] . This activity , referred from now on as persistent activity , is stimulus-selective: a specific pyramidal neuron will only exhibit persistent activity if a stimulus is presented in specific locations of the visual field , in the spatial working memory tasks for example , which represents the neuron's memory field . [3] , [4] . ‘A large body of work has been devoted to understanding the biophysical mechanisms underlying induction and maintenance of persistent activity , which have emphasized the importance of a delicate balance between excitatory and inhibitory recurrent network connections [5] , [6] , [7] , [8] , [9] , as well as the contribution of intrinsic cellular conductances [10] , [11] , [12] , [13] . However , very little is known regarding the cellular mechanisms that enable stimulus selectivity in the PFC . How does a neuron ‘recognize’ the relevant stimulus and therefore , enters a persistent activity state ? Previous studies have suggested that formation of these memory fields entails proper inhibitory transmission [14] , as well as fine interactions between pyramidal neurons and interneurons [15] , similar to the mechanisms underlying the formation of orientation columns [16] . However , additional cell-specific features , such as the latency to the first action potential or the sequence of inter-spike intervals ( ISIs ) , could also be involved in the formation of memory fields , as shown in the visual cortex [17] . In the prefrontal cortex ( PFC ) , the brain area heavily involved in mediating working memory functions and expression of persistent activity , layer V pyramidal neurons come in at least two flavors with respect to their firing patterns: intrinsic bursting ( IB ) neurons , characterized by an initial burst of action potentials ( APs ) followed by single APs or regular spiking ( RS ) neurons , characterized by a sequence of single APs [18] , [19] . These neurons can also be categorized as adapting , whose firing frequency in response to a constant current step decreases during the stimulation and non-adapting [20] , [21] . These different pyramidal neuron subtypes based either on their morphology or firing pattern , can form distinct sub-networks [22] , [23] , [24] that project to different subcortical areas , such as the pons or the striatum , suggesting that they might serve distinctive functional roles . This is further supported by recent data showing that cortico-pontine pyramidal neurons , compared to cortico-cortical neurons , have increased levels of the hyperpolarization activated cation current ( H-current ) , contributing to increased temporal summation and increased amplitude of the slow afterdepolarization ( dADP ) , which in turn facilitates the probability of persistent activity induction [25] . The present study uses detailed compartmental models of IB and RS neuron sub-types to identify ( a ) features of incoming signals that determine persistent activity induction ( stimulus selectivity ) and ( b ) characteristics of the neuronal response to these signals that may be used by downstream neurons to decode information about the probability of persistent activity emergence ( encoding of preferred stimuli ) . Our results predict that stimulus-selectivity is tightly linked to the spatial location of activated synapses . Moreover , while the properties of persistent activity differ between the two subtype models , in both neurons the latency to the first action potential and the initial inter-spike-intervals of the stimulus-induced response contain predictive information regarding the emergence of persistent activity , providing a mechanism for encoding and propagating the occurrence of preferred signals . Two biophysical mechanisms have thus far been implicated in the generation of persistent activity: the NMDA [7] , [9] , [29] , [30] and the CAN conductance [7] , [10] , [11] , [29] . The NMDA current was validated with respect to the AMPA current based on experimental data from connected layer V PFC pyramidal neurons showing that the NMDA-to-AMPA ratio is 1 . 2 , and that NMDA currents have relatively slow kinetics of inactivation ( Fig . 2A , B1 , B2 ) [31] . Furthermore , it has been shown that basal dendrites of layer V pyramidal neurons exhibit NMDA spikes at their basal dendrites ( Fig . 2C ) [32] , [33] . These NMDA spikes are generated in an all-or-none manner , and once generated , stronger stimuli affect mostly the duration of the NMDA spike , while a slight increase in the amplitude may also be seen [32] . We tested whether NMDA spikes could be evoked at the basal dendrites of the neuron models . A dendritic branch located about 100 µm from the soma was stimulated with an increasing number of excitatory synapses . When using at least 40 excitatory synapses to induce the necessary depolarization , dendritic NMDA spikes could be evoked in both model neurons . Further increase in the number of synapses resulted in an increase of the NMDA spike duration along with a slight increase in the spike amplitude , in accordance with the experimental data ( Fig . 2D1 , D2 ) . Layer V PFC pyramidal neurons have been shown to exhibit a delayed afterdepolarization ( dADP ) following stimulation of Gq-coupled receptors [34] , [35] . This dADP is induced following a burst of action potentials and has small amplitude ( average ∼3 mV ) and very slow kinetics ( decay τ = 3 sec ) [35] , rendering it a possible mechanism for induction and maintenance of persistent activity [11] . The dADP has been shown to be primarily generated by the CAN current [36] and possibly results from the activation of TRPC4/5 channels which are found in layer V PFC pyramidal neurons [35] . We simulated the dADP by including an additional ionic mechanism , which is mainly dependent on two variables: a ) the half point of calcium-induced activation and b ) the rate of inactivation . These two variables were adjusted so that the dADP was induced following a burst of 5 spikes , but was much smaller following just 2 spikes ( Fig . 3A , B ) , in accordance with experimental findings [11] . The amplitude of the dADP could be modified by changing the conductance of the CAN mechanism ( Fig . 3C ) and was kept within the physiological range ( 2–8 mV ) , based on existing experimental data ( [11] and Fig . 3D ) . Cortico-cortical connections that are thought to underlie the emergence and maintenance of persistent activity [7] , [37] form synapses onto the basal dendrites of pyramidal neurons [38] . Therefore , the basal dendrites of the model neurons were stimulated with a total of 200 excitatory synapses ( containing both AMPA and NMDA receptors ) , evenly distributed within a few basal dendrites ( see Methods ) , 10 times at 20 Hz ( synchronously ) , while the soma was stimulated with 5 inhibitory synapses ( both GABAA and GABAB ) at 50 Hz ( also synchronously ) [39] . This stimulation protocol was repeated 50 times and the location ( set of dendritic branches ) , but not the stimulation time , of activated synapses varied between trials ( see Methods ) . Persistent activity was induced in a probabilistic manner , in a percentage of these trials . Synaptic stimulation alone ( in the absence of the CAN mechanism ) did not lead to persistent activity in any single cell model , even when the number of stimulated synapses was gradually increased up to 400 . However , since neuromodulators , such as dopamine and serotonin , are known to increase NMDA currents in layer V PFC pyramidal neurons [40] , [41] , [42] and the NMDA conductance is required in large-scale networks for stabilizing persistent activity [7] , we next tested whether increasing the NMDA current by 25% could induce persistent activity in the single neuron models . Increasing the NMDA-to-AMPA ratio ( abbreviated “N” , with equal to the ratio ) from 1 . 2 to 1 . 5 did not induce persistent activity in any model neuron ( data not shown ) although it resulted in decreased inter-spike-intervals ( ISIs ) of the neuronal response during the stimulus ( Supplemental Table S1 ) . The latter concurs with experimental data showing modulation of neuronal excitability by NMDA in vitro [43] . Activation of the dADP mechanism on the other hand , resulted in induction of persistent activity , that is , neuronal activity that lasted more than 3 seconds following the end of the stimulus ( Fig . 4A2 , B2 ) in both neuron models . Increasing the magnitude of the CAN conductance ( i . e . , increasing the amplitude of the resulting dADP , tested with five somatic step pulses , within the physiological range ) increased the probability of inducing persistent activity . We characterized the magnitude of the CAN current that would induce persistent activity with at least 50% probability in the 50 experimental trials in which the spatial arrangement of the synapses on basal dendrites was varied ( i . e . , at least 25/50 trials exhibited persistent activity ) . The dADP required for induction of at least 50% persistent activity for the RS and IB neuron models was 3 . 2 and 3 . 9 mV , respectively ( Fig . 4C , white bars ) and dropped by 1 . 3 mV in both models when the NMDA-to-AMPA ratio increased to 1 . 5 ( Fig . 4C , black bars ) . Note , however , that the slightly larger dADP in the IB model cell corresponds to a smaller CAN conductance compared to the RS model cell ( Fig . 3C ) . This can be explained by the enhanced R-type calcium and persistent sodium currents in the IB model cell which may contribute to the long-lasting depolarization produced by the CAN mechanism , thus partially substituting the CAN conductance . Taken together , these findings show that induction of persistent activity requires a larger dADP ( although a smaller CAN conductance ) in the IB than the RS model cell . In the previous analysis , we classified persistent activity as the neuronal activity that continues past the end of the initiating stimulus and lasts at least 3 seconds . However , the neuron models could exhibit self-terminated persistent activity ( 500–2000 ms ) ( Fig . 4D , white bars ) , even in the experimental trials classified as not having persistent activity ( ‘no persistent’ trials ) . We notice that the RS neuron model exhibits significantly less temporally-restricted persistent activity compared to the IB neuron model ( Fig . 4D , white bars , p<0 . 001 ) . Increasing the NMDA-to-AMPA ratio however facilitates short-lasting persistent activity to a much larger extent in the RS than the IB model ( Fig . 4D , black bars , p<0 . 001 ) . A possible explanation could lie in the fact that the IB model cell has larger R-type calcium and persistent sodium currents , which together with the CAN mechanism contribute to the prolonged depolarization needed for persistent activity . Thus , an additional slow increase in Ca++ influx due to enhanced NMDARs would have a greater impact on the RS model , where the primary conductance responsible for the dADP is the CAN conductance , than the IB model , where several mechanisms -with different kinetics- already contribute to this depolarization . Furthermore , this short-lasting persistent activity could be significant in an in vivo situation where network mechanisms could maintain it for longer periods of time . These findings show that , while persistent activity in the single neuron models is primarily dependent on the CAN current , altering the NMDA-to-AMPA ratio modulates the duration of persistent activity that lasts less than 3 seconds . Having characterized the conditions leading to persistent activity emergence in both model cells , our next goal was to search for features of the input and/or the models' response that would be associated with stimulus-selectivity . The presence of ‘memory fields’ has been shown in individual PFC neurons with respect to delay-period activity [44] . That is , a specific neuron exhibits robust delay-period activity ( i . e . , an increase in firing rate during the delay compared to the stimulus period ) only for a specific set of locations in the visual field [3] . The way a PFC pyramidal neuron , however , identifies its memory field remains an open question . It is possible that different incoming stimuli , such as stimuli located in different parts of the visual field , activate synapses in different dendritic branches on PFC pyramidal neurons and this spatial specificity of inputs is in turn used to discriminate between preferred ( i . e . , those leading to persistent activity ) and non-preferred stimuli . In our models , we used 50 simulation trials , in which the set of dendritic branches containing synaptic mechanisms varies with each trial ( see Methods ) . This variability in the location of incoming contacts could be assumed to represent different incoming stimuli [45] , hence , we conjecture that the spatial location of activated synapses may play a role in persistent activity induction . To test this hypothesis , we measured the distance from the soma and the center of each dendritic branch that contained stimulated synapses , and averaged the values of each of these features for all dendritic branches in ‘persistent’ versus ‘no persistent’ trials . We found that in both neuron models , synaptic mechanisms were on average located further away from the soma for the ‘persistent’ trials , compared to the ‘no persistent’ trails and this difference was statistically significant ( p<0 . 001 ) ( Fig . 5B , RS model and Supplemental Fig . S3 , IB model ) . The distributions of all activated dendritic segments in ‘persistent’ and ‘no persistent’ trials ( Fig . 5C , RS model and Supplemental Fig . S3 , IB model ) show that this difference stems from a rightward shift as well as a change in the shape of the ‘persistent’ trial distribution due to the activation of dendritic segments located further away from the soma . A possible mechanistic explanation as to why inputs that are further away from the soma lead to persistent firing may be linked to the generation of NMDA spikes . As shown in Fig . 5D , the magnitude of NMDA spikes is much larger when synapses are stimulated in distal compared to proximal locations within the basal dendrites of both model neurons . It is thus possible that distal inputs lead to persistent activity emergence via the facilitation of NMDA spikes which in turn promote the supralinear integration of synaptic inputs [46] and provide much larger and longer lasting somatic depolarizations . These findings are supported by recent experimental data showing that inputs to proximal basal dendrites of cortical pyramidal neurons sum linearly and require precise temporal coincidence for effective summation , whereas distal inputs are combined supralinearly over broader time windows in an NMDAR-dependent manner [47] . Finally , these findings suggest that the relative distance of incoming signals from the cell body may code for the neuron's memory field and therefore , their ability to induce persistent activity . Since the spatial location of incoming contacts is significantly different between ‘persistent’ and ‘no persistent’ trials , it is likely that these differences are reflected in the neuronal response to these stimuli and , if so , this information can be used by downstream neurons to decode the upcoming emergence of persistent activity before it occurs [48] . To test this hypothesis , we first examined whether features of the neuronal response to the stimulus , such as the average firing frequency or the AP latency differed between preferred and non preferred inputs . We found that the average ISIs of the neuronal response during the stimulus was not different between ‘persistent’ and ‘no persistent’ trials in either the RS or the IB neuron model ( see Supplemental Table S2 ) . However , the first AP latency of the models' response was clearly different in the ‘persistent’ trials when compared to the ‘no persistent’ trials ( see Fig . 6A–D and Supplemental Table S3 ) . Specifically , ‘persistent’ trials in the RS model had AP latencies that were significantly larger than the ‘no persistent’ trails for both NMDA-to-AMPA ratios tested ( p<0 . 001 ( N1 . 2 ) and p<0 . 001 ( N1 . 5 ) , non-overlapping boxes in Fig . 6C , D ) . For the IB model neuron , differences in the AP latencies were highly significant only when the NMDA-to-AMPA ratio was increased to N1 . 5 ( p = 0 . 0065 ( N = 1 . 2 ) , overlapping boxes in Fig . 6C and p<0 . 001 ( N = 1 . 5 ) , non-overlapping boxes in Fig . 6D ) . In both models , persistent activity emergence was associated with a slightly slower onset of the neuronal response , which could be explained by the more distal location of activated synapses , compared to the ‘no persistent’ trials ( Fig . 5B–C and Supplemental ) . Although the differences in the AP latencies between ‘persistent’ and ‘no persistent’ trials were small ( 200–300 µs ) , recent studies have shown that even submillisecond differences in AP emergence or width could represent meaningful coding parameters for neurons [49] , [50] , [51] , suggesting that the magnitude of the AP latency maybe used to code for the occurrence of a preferred stimulus . To test whether differences in the AP latency can discriminate between ‘persistent’ and ‘no persistent’ trials in a more systematic manner , we assessed the ability of the AP latency values to predict the emergence of persistent activity using Linear Discriminant Analysis ( LDA ) . For this , we used a training set ( consisting of the AP latencies for 20 ‘persistent’ and 10 ‘no persistent’ trials ) in order to determine the optimal cut-off that separates the two distributions based solely on the value of the AP latency . The method was validated using leave-five-out cross validation ( LFOCV ) and subsequently tested on a previously unseen set of another 30 trials ( see Methods ) , to assess how well can the AP latency of the response to a new input determine whether this input will induce persistent activity or not . Each observation ( i . e . trial ) of the test set was passed through the 6 ‘trained’ LDA models produced by the LFOCV and a class label was assigned by each model ( 0 for ‘no persistent’ and 1 for ‘persistent’ ) . All model outputs were then averaged and if the average was 0 . 5 or higher , then that specific observation was classified as a ‘persistent’ trial , otherwise it was classified as a ‘no persistent’ trial . Using the percentage of correctly predicted ‘persistent’ ( sensitivity ) and ‘no persistent’ ( specificity ) trials to assess the method's performance accuracy , we found that discrimination was more successful in the RS than the IB model neuron . Specifically , for the RS model and an NMDA-to-AMPA ratio of 1 . 2 , the sensitivity of the method was very high ( 100% or 1 ) , while the specificity was a bit lower ( 0 . 7 ) , resulting in total accuracy of 0 . 85 ( Fig . 6E–F , black squares ) . This means that out of the 30 trials tested , all 20 ‘persistent’ trials and 7/10 of the ‘no persistent’ trials are correctly identified by their respective AP values . For the IB model , the sensitivity value was 0 . 8 , the specificity was only 0 . 4 , and the total accuracy was 0 . 6 , considerably decreased compared to the RS model ( Fig . 6E–F , ‘x’ marks ) . However , for a larger NMDA-to-AMPA ratio ( N 1 . 5 ) , the performance accuracy was high for both models , with the RS cell reaching 100% and the IB cell reaching 90% ( Fig . 6G ) . Taken together , these findings suggest that the AP latency may be used as a discriminatory feature for signaling whether a given stimulus will or will not lead to persistent firing , that the accuracy of this prediction is higher in the RS than the IB model neuron and is strongly dependent on the NMDA contribution . We next investigated whether some other characteristic of the stimulus-induced response can better predict persistent activity emergence even for a lower NMDA-to-AMPA ratio . Towards this goal , we used the ISIs during the stimulus-induced response as input to a linear perceptron ( see Methods and Fig . 7A for a graphical illustration ) and tested whether ‘persistent’ trials could be discriminated from ‘no persistent’ trials based on these features . The perceptron was trained and validated with 30 trials ( as in LDA ) , using the leave-one-out cross validation ( LOOCV ) method and subsequently tested on a previously unseen set of another 30 trials ( see Methods and LDA analysis above ) . Sensitivity and specificity measures were again used to assess the method's performance accuracy . The sensitivity of the perceptron for both the RS and IB neuron models was 100% when the first 2 , first 5 , or all ISIs of the stimulus-induced response were used as input ( Fig . 7B ) . Similarly , specificity of the perceptron for both models was above 80% , with the IB slightly better than the RS model ( 0 . 9 or greater , Fig . 7C ) for all ISI sequences tested . In particular , the specificity for the RS model was 80% , 90% and 90% when the first 2 , 5 or all ISIs were used as input features ( Fig . 7C , black squares ) whereas the specificity of the IB model was 90% , 100% and 90% , respectively ( Fig . 7C , ‘x’ marks ) . The perceptron's performance was also assessed on shuffled datasets ( in which ‘persistent’ and ‘no persistent’ trials were randomly labeled ) and the performance was severely degraded: the sensitivity dropped to 0% and the specificity to 60% for both RS and IB models . These results show that the initial ISIs of the stimulus-induced response contain highly accurate predictive information regarding the emergence of persistent activity in both the IB and the RS model neurons . Overall , our findings suggest that temporal characteristics of the stimulus-induced response , such as the first spike latency and the first 2 ISIs , contain significant predictive information about the emergence of persistent activity beyond the end of preferred stimuli while average characteristics such as the firing frequency don't capture such information , in accordance with data from other brain regions [48] , [52] , [53] . These findings are particularly important as they pinpoint specific features of the neuronal response , at the single neuron level , which are common across two major sub-types of pyramidal neurons and which encode stimulus preference with respect to persistent activity emergence . If experimentally validated , these findings suggest a potential mechanism by which stimulus-selectivity that initiates in primary cortices may be decoded by downstream PFC pyramidal neurons within less than 100 miliseconds from the stimulus presentation , and this rapid decoding may have serious implications for the expression of goal-directed behaviors that have been documented in the PFC [54] . Since both model neurons seem to use similar codes for stimulus-selective persistent activity induction , we wondered whether their different firing patterns influenced persistent activity at a different level . We thus contrasted the properties of persistent activity , such as its induction threshold , firing frequency and firing pattern and their dependence on CAN and NMDA in the two model cells . The ability of PFC pyramidal neurons to display neuronal activity that persists after the end of stimulation was first recorded in vivo in monkeys [57] . This persistent activity has been considered as a network property and particularly a property of recurrent networks due to reverberating excitation [5] . Thus , single neurons have to be part of a recurrent network in order to exhibit persistent activity firing . This hypothesis was further corroborated by the fact that persistent activity could not be induced in neurons recorded in PFC slices where many of the recurrent connections could be severed . Following a modification of the artificial cerebrospinal fluid used , persistent activity lasting for about 1–2 seconds could be recorded from single PFC pyramidal neurons in the slice preparation . This persistent activity , or rather the ‘UP’ state , which occurs both spontaneously and following a stimulus [8] , [9] , [30] , is mediated by AMPA and NMDA [9] and is modulated by GABAB currents [58] and dopamine [59] , [60] . Single neurons have been shown to exhibit persistent activity following activation of metabotropic receptors , such as the muscarinic acetylcholine receptor ( mAchR ) and the metabotropic glutamate receptors ( mGluR ) [10] , [11] , [12] , due to an underlying depolarizing envelope ( i . e . , dADP ) activated by these receptors [34] , [35] , [61] , [62] . The average dADP in PFC pyramidal neurons following a short 20 Hz stimulus ranges between 2 and 8 mV , not large enough to induce persistent activity by itself , as it has been suggested for enthorhinal cortical neurons [63] . Our computational study shows that this small depolarization when coupled with synaptic activation can induce persistent activity in single neurons . The common characteristic of both dADP and NMDA mechanisms is their slow inactivation kinetics , previously suggested to be required for the persistent activity to maintain ‘physiological’ firing rates [31] . Our study examined the role of these mechanisms in persistent activity . We showed that while an increased NMDA modulates the neuronal firing rate during the stimulus , increasing the CAN current specifically increases the firing frequency of persistent activity . Furthermore , while increasing the NMDA current increases the variability of firing during persistent activity , increasing the CAN current decreases this variability . Based on analysis from in vivo delay-period activity , stimulus-selective -and thus more informative ( or significant for mediating behavior ) - persistent activity has high firing frequency rates ( increased compared to cue-response ) as well as increased variability [55] . Our results suggest a dual role for both NMDA and CAN current mechanisms: CAN current acts to enhance persistent activity firing but makes it more regular , while NMDA acts to decrease persistent activity firing but increases its irregularity . Thus , a delicate balance between these two mechanisms in vivo is likely to be critical for proper persistent activity firing . Persistent activity in PFC is stimulus-selective , that is , a neuron will only exhibit persistent firing to specific stimuli , for example stimuli that appear on a specific location of the visual field [44] . The selection of stimuli that a neuron responds to is called a ‘memory field’ , in analogy to the receptive fields in the visual cortex [64] , or the place fields in the hippocampus [65] . Inhibitory mechanisms play a significant role in shaping the memory fields in PFC , since blockade of GABAA receptors disrupts the emergence of stimulus-selective persistent activity [14] . In our study , we made the assumption that different environmental stimuli could be mapped as different spatial arrangements of synaptic inputs on the basal dendrites . Dendritic activation has been shown to map direction-selective responses in the fly [45] as well as place cells in the hippocampus [66] , hence , it is possible that different spatial locations in the neuron's receptive field correspond to the activation of spatially distinct synaptic patterns . The fact that persistent activity emergence in the model neurons is associated with activation of synapses that are located further away from the soma suggests that perhaps in vivo circuits are refined so that stimuli within a memory field project to more distal basal dendrites compared to stimuli outside the neuron's memory field . Since persistent activity is characterized by slow kinetics , it is likely that inputs to distal dendrites , which are generally characterized by slower integration , are more suitable for carrying signals related to persistent activity induction . Finally , stimulation of distal dendrites can generate larger NMDA spikes ( Fig . 5 and [67] ) which will in turn prolong the window for temporal summation of incoming signals thus resulting in larger and longer-lasting somatic depolarization . Therefore , distal inputs may facilitate persistent activity emergence via the enhancement of NMDA spikes [47] . In addition to a strong link between the spatial arrangement of preferred stimuli and the emergence of persistent activity , our data showed that features of the neuronal response during the stimulus such as the AP latency [48] and the first few inter-spike-intervals , can code for the emergence of persistent activity . Specifically , we found that the AP latency in ‘persistent trials’ is on average significantly longer compared to ‘no persistent’ trials in both the RS and IB model neurons . This may be due to the fact that ‘persistent trials’ corresponded to synaptic arrangements in which activated synapses were located significantly further away from the soma than in ‘no persistent’ trials . Although the differences in the AP latencies between ‘persistent’ and ‘no persistent’ trials were submillisecond , several studies suggest that they could still be decoded by downstream neurons [49] , [50] , [51] . This finding adds to the coding capabilities of the AP latency which has also been found to code for differences in spatiotemporal characteristics of the input in CA1 model neurons [48] as well as the location of sound in secondary auditory neurons [52] . While AP latency was not as powerful predictor at lower NMDA-to-AMPA ratio , particularly in the IB model cell , stimulus selectivity was encoded in the first few inter-spike-intervals of the stimulus-induced response . For both the RS and IB model neurons the emergence ( or not ) of persistent activity could be predicted with high accuracy when utilizing the first few ( 2 or 5 ) ISIs . These findings are particularly important as they suggest the potential decodability of preferred stimuli by neuronal circuits downstream the L5 PFC pyramids , as early as a few hundreds of milliseconds following the stimulus presentation and long before the emergence of persistent activity . In support of this conjecture , recent data suggest that PFC neurons can categorize input signals as early as the stimulus presentation time [68] , [69] . This information could in turn be used by downstream striatal [70] and pontine neurons [71] , [72] to prepare for the execution of a specific movement and may provide a neuronal basis for goal-directed behavior . Overall , our findings regarding the coding of information in ISIs are in agreement with studies from other brain regions where ISI sequences were shown to contain more information about receptive fields in the visual cortex than the average firing frequency [17] , and could be used to filter and modulate receptive fields in retinal ganglion cells [73] . Recently , in vivo patch-clamp techniques uncovered the importance of intrinsic cellular features in active place cell in the hippocampus [74] . Thus , it is now possible to use patch-clamp recordings in PFC pyramidal neurons during virtual working memory tasks to test the prediction that cellular features such as the AP latency or the ISIs can be used to code for the occurrence of preferred stimuli and the emergence of persistent activity . While both IB and RS pyramidal neurons have been documented in the prefrontal cortex [19] , [20] , their functional role remains unclear . According to recent studies , there could be a link between neuronal sub-types and their preferred target areas . For example , both RS and IB cortical neurons project to the pons ( cortico-pontine ) or the striatum but no IB neurons project to the contralateral cortex ( cortico-cortical ) [23] . Similarly , IB neurons in the distal parts of the subiculum project primarily to the medial enthorhinal cortex but not the amygdala [75] . This segregation is likely to be associated with some form of functional specialization of RS and IB neurons . Furthermore , corticopontine neurons , which consist of both RS and IB neurons , in PFC seem more likely to express persistent activity in response to acetylcholine modulation compared to cortico-cortical neurons , in which no IB neurons are found [25] . Our findings are in line with this hypothesis as they support a differential role of RS and IB pyramidal neurons in persistent activity emergence . Moreover , since different neuronal properties have been suggested to provide a recurrent network with different persistent activity characteristics [76] , our data suggest that RS and IB neurons may form distinct subnetworks when connected in a recurrent network . Future modeling and experimental work is needed to further investigate this hypothesis . Dopamine , acting through D1/5 receptors , increases the NMDA current [42] while it decreases the dADP [11] . Our computational study , as well as a previous one [77] , suggested that increasing the NMDA component of synaptic stimulation decreases the CAN current required for induction of persistent activity . Thus , D1 signaling seems to modulate both of these mechanisms in order to maintain stability of neuronal excitability . In the case where only NMDA currents are increased while the dADP amplitude in response to metabotropic receptors remains the same , persistent activity will be elicited even in response to non-relevant stimuli . On the other hand , if dADP alone was reduced without any change in the NMDA currents , then no stimulus would be able to induce persistent activity . Our modeling work showed that when increasing the NMDA current in a similar amount that DA does , the ISI variability increases in both model neurons but it remains elevated for the entire 3 second recoding period of persistent activity only in the IB model neuron ( Fig . 9F ) . Furthermore , increasing the NMDA current also changes the persistent activity properties of the RS model neuron to resemble those of the IB model neuron , with a bursting firing pattern during persistent activity and the emergence of time-limited persistent activity ( Fig . 9B and 4D ) . Our results are in agreement with the well established idea that an increase in DA is necessary for proper expression of persistent activity since the properties of persistent activity in our model neurons are closer to the ones observed in vivo , when the NMDA current contribution is increased . Furthermore , increasing the NMDA current contribution also improves the ‘persistent’ vs . ‘no persistent’ trials discrimination particularly in the IB model neuron , enhancing the coding capabilities of this neuronal subtype , since both the AP latency and/or the first few ISIs can be used to predict the emergence of persistent activity . DA also modulates other biophysical mechanisms in pyramidal neurons of the prefrontal cortex , such as the L-type calcium channels [78] , [79] , sodium currents [27] , [80] and potassium currents [81] , [82] . Modulation of all these mechanisms is likely to affect properties of persistent activity; however , such an analysis is beyond the scope of this work . Our detailed model reproduces closely the electrophysiological activity of PFC pyramidal neurons . Nonetheless , sources of inaccuracy may have been introduced since the experimental data used to constrain the model are products of in vitro preparations . In that sense however , model limitations do not significantly differ from those of the in vitro preparations whose findings are readily replicated by the model . Simplifications that have been adopted in this work include: ( i ) a strictly phenomenogical model of the CAN current , ( ii ) absence of background synaptic activity which is known to occur in vivo ( although we do include membrane noise ) , ( iii ) stimulation delivered only to the basal dendrites of the model neurons ( which are known to receive the majority of inputs from other cortical areas ) , while more spatially distributed stimulation in combination with the experimentally observed accumulation of extracellular potassium [83] could reduce the threshold for persistent activity induction and require more complex spatial coding features , ( iv ) same biophysical mechanisms in both model cells , yet different conductance values for the R-type calcium and sodium currents , while in nature , there is probably some variability , and ( v ) no modeling of plasticity or neuromodulator effects . In spite these simplifications , our model findings are important as they have identified neuronal features that could code for the emergence of persistent activity in both neuronal subytpes found in the cortex , as well as differential properties of persistent activity between RS and IB neurons . In summary , our modeling results allow the formulation of several predictions , which when tested experimentally could further the current knowledge on persistent activity , its underlying mechanisms and its contribution to working memory . First , we predict that the location of activated synapses is critical for the emergence of persistent firing: stimuli that lead to persistent activity consist of inputs that arrive in distal parts of the basal tree , far away from the soma . This prediction could be easily tested experimentally in slice preparations that generate Up and Down states . Specifically , synaptic stimulation of proximal versus distal basal dendrites should have a different impact on the probability of generating Up states and/or modulating the firing frequency or duration of the Up state . Second , we have identified the AP latency and first few ISIs of the stimulus-induced response as features that could discriminate between stimuli that result in persistent activity or not . These findings could be tested experimentally in slice preparations and/or in vivo when a stimulus is used to induce persistent activity . Finally , our models predict that the dADP threshold for persistent activity induction is lower in RS than IB neurons , suggesting that RS neurons should comprise the majority of layer 5 PFC pyramidal neurons that exhibit persistent firing . This can also be tested experimentally in slice preparations that generate Up and Down states , where the effect of synaptic stimulation on the UP state is contrasted between RS and IB neurons . Overall , our modeling work identifies key features of the neuronal response that could predict the emergence of persistent activity and pinpoints a differential role of RS and IB model neurons in persistent activity properties . Both the RS and IB models were validated with respect to passive and active membrane properties as well as apical and basal dendritic responses ( see Supplemental Fig . S1 ) . Dendritic and somatic voltage traces in response to glutamate release were validated based on experimental data [32] ( see Supplemental Fig . S2 ) . Dendrites were stimulated with a total of 200 excitatory synapses ( containing both AMPA and NMDA receptors ) , equally distributed in 10 different dendritic branches ( 20 synapses on each branch ) which were activated 10 times at 20 Hz . The 10 dendritic branches were selected randomly from the pool of all basal dendrites . Synapses were distributed at random locations within each branch , according to a uniform distribution . The soma of the neuron model was stimulated with 5 inhibitory synapses [39] at 50 Hz . Both excitatory and inhibitory synapses were activated synchronously , without any temporal variability between different trials . Since we were interested in studying the suprathreshold response of neurons to a specific stimulus , we used the above number of synapses for which the neuron model responded with at least 5 APs in the 10 event stimulus . In addition , for best simulation of membrane potential fluctuations as observed in vitro due to the stochastic ion channel noise [88] , [89] , an artificial current with Poisson characteristics was injected in the soma of both RS and IB neuron models . This simulation of channel noise is simple compared to ones recently reported [89] , yet sufficient for the complexity of the model used and the purpose of the current study . We define persistent activity as the prolongation of neuronal activity following the end of the stimulus for at least 3 seconds . All simulations were recorded for 5 seconds , and if neuronal activity persisted past the 3 seconds following the stimulus , it did not stop before the 5 sec recording . Simulations included 50 repetition trials for each condition ( i . e . specific set of NMDA and CAN conductances ) , where the spatial distribution of activated synaptic mechanisms at the different basal dendritic branches changed in each trial . For the data analysis , only conditions in which persistent activity emerged in at least 50% of the runs were used , unless otherwise noted . Estimation of inter-spike-intervals ( ISIs ) of the simulated neuronal responses , as well as generation of the ISI histograms was performed with custom-made macros using IgorPro software ( Wavemetrics , Inc ) and Matlab ( Mathworks , Inc ) . Prediction of persistent activity emergence based on the ISI values was done using a custom-made Artificial Neural Network written in Java . The network used was a simple perceptron , which was initially trained with 30 randomly selected trials , validated ( using leave-one-out cross validation ) and subsequently tested on another 30 trials ( both training and test sets comprised of 20 persistent and 10 no persistent trials ) . Prediction of persistent activity emergence based on the AP latency values was done using Linear Discriminant analysis ( LDA ) in Matlab ( Mathworks , Inc . ) , with code downloaded from the file exchange site ( http://www . mathworks . com/matlabcentral/fileexchange/29673-lda-linear-discriminant-analysis ) . The method was initially trained with 30 randomly selected trials , validated ( using leave-five-out cross validation ) , and then tested on another set of unseen 30 trials , similarly to the Perceptron analysis . The following conditions were used for the prediction analysis: RS neuron model −N1 . 2/−N1 . 5 , 40 persistent trials , 20 no persistent trials; IB neuron model −N1 . 2/−N1 . 5 , 40 persistent trials , 20 no persistent trials ( as shown below in the text: N = NMDA-to-AMPA ratio , dADP = delayed afterdepolarization ) . In each trial , synaptic mechanisms were placed according to a uniform distribution within 10 dendritic branches which were selected at random . For each of the selected dendritic branches , the path distance from the center of the branch to the soma was calculated and used to estimate the average dendritic distance for any given trial . All experimental results shown here are parts of the neuronal database recorded by Kyriaki Sidiropoulou while she was at Dr . Francis White laboratory and have been reported in previous publications [11] , [28] . The model is available for download from ModelDB ( Link: https://senselab . med . yale . edu/modeldb/ShowModel . asp ? model=144089 , accession number: 144089 ) .	Memory , referred to as the ability to retain , store and recall information , represents one of the most fundamental cognitive functions in daily life . A significant feature of memory processes is selectivity to particular events or items that are important to our survival and relevant to specific situations . For long-term memory , the selectivity to a specific stimulus is seen both at the behavioral as well as the cellular level . For working memory , a type of short-term memory involved in decision making and attention processes , stimulus selectivity has been observed in vivo using spatial working memory tasks . In addition , persistent activity , which is the cellular correlate of working memory , is also selective to specific stimuli for each neuron , suggesting that each neuron has a ‘memory field’ . Our study proposes that both the location of incoming inputs onto the neuronal dendritic tree and specific temporal features of the neuronal response can be used to predict the emergence of persistent activity in two neuron models with different firing patterns , revealing possible mechanisms for generating and propagating stimulus-selectivity in working memory processes . The study also reveals that neurons with different firing patterns may have different roles in persistent activity expression .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "working", "memory", "cognitive", "neuroscience", "computational", "neuroscience", "single", "neuron", "function", "biology", "neuroscience", "coding", "mechanisms" ]	2012	Predictive Features of Persistent Activity Emergence in Regular Spiking and Intrinsic Bursting Model Neurons
The point of attachment of spindle microtubules to metaphase chromosomes is known as the centromere . Plant and animal centromeres are epigenetically specified by a centromere-specific variant of Histone H3 , CENH3 ( a . k . a . CENP-A ) . Unlike canonical histones that are invariant , CENH3 proteins are accumulating substitutions at an accelerated rate . This diversification of CENH3 is a conundrum since its role as the key determinant of centromere identity remains a constant across species . Here , we ask whether naturally occurring divergence in CENH3 has functional consequences . We performed functional complementation assays on cenh3-1 , a null mutation in Arabidopsis thaliana , using untagged CENH3s from increasingly distant relatives . Contrary to previous results using GFP-tagged CENH3 , we find that the essential functions of CENH3 are conserved across a broad evolutionary landscape . CENH3 from a species as distant as the monocot Zea mays can functionally replace A . thaliana CENH3 . Plants expressing variant CENH3s that are fertile when selfed show dramatic segregation errors when crossed to a wild-type individual . The progeny of this cross include hybrid diploids , aneuploids with novel genetic rearrangements and haploids that inherit only the genome of the wild-type parent . Importantly , it is always chromosomes from the plant expressing the divergent CENH3 that missegregate . Using chimeras , we show that it is divergence in the fast-evolving N-terminal tail of CENH3 that is causing segregation errors and genome elimination . Furthermore , we analyzed N-terminal tail sequences from plant CENH3s and discovered a modular pattern of sequence conservation . From this we hypothesize that while the essential functions of CENH3 are largely conserved , the N-terminal tail is evolving to adapt to lineage-specific centromeric constraints . Our results demonstrate that this lineage-specific evolution of CENH3 causes inviability and sterility of progeny in crosses , at the same time producing karyotypic variation . Thus , CENH3 evolution can contribute to postzygotic reproductive barriers . Centromeres are the site where spindle microtubules attach to chromosomes during cell division . This attachment is mediated via a multi-protein complex called the kinetochore , a structure essential for the stable inheritance of genetic information . Contrary to expectation , the centromere is not a genetic locus in the traditional sense of being defined by its DNA sequence [1 , 2] . The DNA sequence underlying the centromere is not evolutionarily conserved and in most species , is composed of megabases of rapidly evolving tandem repeats [3] . However , these repeats are not essential to centromere formation since neocentromeres or the gain of new centromeric activity has been observed over unique DNA sequences as well [4–6] . The common denominator to all centromeres , old and new , is the presence of a centromere specific histone variant of H3 called CENH3 ( or CENP-A ) [7] . This and other evidence [8–10] indicate that in both plants and animals , the location of centromeres is specified epigenetically by the presence of CENH3 . Despite this ancient and conserved role of CENH3 in maintaining genetic integrity , the CENH3 protein sequence is not evolving under purifying selection . In contrast to the nearly invariant histone H3 , CENH3 homologs are highly divergent . For example , CENH3 from Arabidopsis thaliana and Arabidopsis arenosa , sister species that shared a common ancestor approximately 5 MYA , differ at 23 of 178 amino acid positions while canonical Histone H3 has accumulated only 4 substitutions out of 136 amino acid positions since the divergence of plants and animals . In the Brassicaceae and in Drosophila , the diversification of CENH3 at both the Histone Fold Domain ( HFD ) and the N-terminal tail appears to be driven by adaptive evolution under natural selection [11 , 12] . This accelerated evolution is especially pronounced at the N-terminal tail of CENH3 , which is hyper-variable both in its length and sequence . Why a structure essential for stable inheritance of genetic material is composed of genetically unstable units is a fundamental unsolved question in the field of chromosome biology . The “centromere drive” hypothesis proposed by Henikoff and Malik puts forward genetic conflict as the source of this striking diversification [13] . This model supposes that DNA sequence can influence centromere function . Female meiosis in animals and plants is asymmetric , in that only one product survives to become the egg cell . If a sequence variant evolves that can preferentially segregate into the surviving egg cell , it will rapidly sweep through the population [14 , 15] . However , such driving chromosomes would be associated with fitness costs including fixation of linked deleterious mutations , sterility due to non-disjunction and in the case of sex chromosomes , skewed sex ratios . This in turn is expected to set off the evolution of centromere-associated proteins to suppress the selfish transmission of this centromere . Cycles of centromere drive and suppression could result in the rapid diversification of centromeres and associated factors . One outcome of divergence in centromere components , DNA and/or proteins , could be the evolution of incompatibilities in the segregation machinery , leading to the reproductive isolation of populations . While there is strong evidence attributing expansion of centromeric repeats to meiotic drive [16] , whether CENH3 or other centromeric proteins are co-evolving with DNA sequences to suppress instances of drive remains speculative . The functional consequences of CENH3 divergence are difficult to address because CENH3 is an essential gene and most model systems cannot tolerate the segregation errors caused by mutations or modification to its function . In D . melanogaster and mammalian cells , RNAi has been used to down-regulate CENH3 levels [17 , 18] . However , the interpretation of any loss-of-function phenotypes is confounded by the persistence of CENH3 through multiple rounds of cell division . In contrast , a cenh3 null mutant in A . thaliana allows us to completely replace the endogenous protein with transgenic variants . In addition , A . thaliana has high-copy centromeric repeats similar in organization to most plants and animals [19] , making it an attractive system for testing general principles of centromere function . Also unique to A . thaliana is the CENH3-mediated genome elimination system [20] , which we have leveraged as a sensitive genetic assay for centromere function in this study . This genetic assay is based on the discovery that when a cenh3 null mutant expressing a GFP-tagged chimeric CENH3 ( GFP-tailswap ) is crossed to a wild type , missegregation of chromosomes from the GFP-tailswap parent is observed [20] . Since A . thaliana has a high tolerance to aneuploidy , the F1 progeny capture a wide range of segregation errors . In the most extreme cases , all the chromosomes from the GFP-tailswap parent are lost ( genome elimination ) yielding haploid offspring that inherit chromosomes only from the wild-type parent . Importantly , segregation errors are only observed in crosses to wild type and not during normal vegetative growth or when GFP-tailswap plants are selfed . This implies that chromosome missegregation in the F1 zygote is the result of competition between wild-type centromeres and defective centromeres built on the artificial chimeric CENH3 . Thus , the frequency of segregation errors and genome elimination can be used as a sensitive assay for centromere function . We were interested in asking what would happen if instead of using an artificial chimeric construct we simply replaced the endogenous CENH3 with natural variants from related species . Previous studies using GFP-tagged versions of CENH3 orthologs had found a very narrow evolutionary window of functional complementation [21] . This leads to the conclusion that plant CENH3s are evolving under unique and highly dissimilar lineage-specific functional constraints [21] . Here , using untagged natural variants of CENH3 we observed the following: 1 ) Despite extensive sequence divergence , the essential functions of CENH3 are conserved across a much broader evolutionary time-scale than previously thought; 2 ) Naturally evolved divergence in CENH3 can contribute to genetic instability by causing chromosome missegregation , generating not only aneuploids and haploids , but also novel genetic rearrangements; 3 ) It is the divergence in the fast evolving N-terminal tail domain that is responsible for segregation defects and 4 ) The N-terminal tail appears to be evolving in a modular fashion . With these results , we argue that the core functions of CENH3 have remained unchanged over long evolutionary periods while the N-terminal tail of CENH3 is evolving as a species-specific optimized platform for centromere organization . Finally , our study presents the first direct evidence for the role of CENH3 divergence in speciation . A . thaliana is a member of the mustard family ( Brassicaceae ) , known for its agriculturally important Brassica crops . Analysis of CENH3 homologs from several species within the mustard family revealed that it is adaptively evolving , both at the Histone Fold Domain ( HFD ) and the N-terminal tail ( NTT ) [11] . Ravi et al . ( 2010 ) [21] had assayed CENH3s from species within the Brassicaceae and beyond for functional complementation of cenh3–1 , a CENH3 null mutation in A . thaliana . They found that GFP-tagged CENH3 from Brassica rapa and Zea mays localized at A . thaliana centromeres , but only GFP-tagged CENH3 from the closely related species A . arenosa rescued embryo lethality of the cenh3–1 . A caveat to these experiments was the presence of the GFP-tag . GFP-tagged A . thaliana CENH3 largely complemented the functions of the A . thaliana cenh3–1 mutation , but when crossed to wild type segregation errors were observed at a low frequency . This hinted that the GFP-tag is not entirely neutral . Thus , to assay only the effects of naturally evolved variation on CENH3 function , we decided to test complementation of the cenh3 null mutant using native untagged proteins . We chose CENH3 from B . rapa and Lepidium oleraceum , two species nested within the Brassicaceae family . L . oleraceum is more closely related to A . thaliana than B . rapa , but more distantly than A . arenosa [22] . To test for complementation , we transformed cenh3–1/CENH3 heterozygotes with constructs expressing genomic sequence encoding L . oleraceum CENH3 ( LoCENH3 ) and B . rapa CENH3 ( BrCENH3 ) under the endogenous A . thaliana CENH3 promoter . We recovered transformants that were homozygous for the cenh3–1 mutation for both variants in the T1 generation . This result is revealing in two ways: firstly it shows that the GFP-tag interferes with CENH3 function and secondly , it indicates that the previously defined boundary of functional complementation is incorrect . We further characterized the extent of mitotic and meiotic complementation in the T2 generation . A . thaliana plants homozygous for cenh3 null mutation expressing transgenic L . oleraceum CENH3 or B . rapa CENH3 were phenotypically indistinguishable from wild type ( Fig . 1A ) . We therefore conclude that B . rapa and L . oleraceum CENH3 can fully complement A . thaliana CENH3 mitotic functions required for vegetative growth . Transgenic lines for both CENH3 variants in a cenh3–1 homozygous background were also self-fertile . To assay meiotic complementation , we wanted to identify plants that were homozygous for both the cenh3 null mutation and variant CENH3 transgene . Following segregation ratios of the transgene is not informative in a cenh3–1 homozygous mutant background , since individuals without transgenic CENH3 cannot survive . Therefore , we decided to use frequency of seed death in selfed siliques of T2 plants to infer the zygosity of the CENH3 transgene . Individuals that are cenh3 -/- and heterozygous for the transgene are expected to produce 25% seed death upon selfing . Assuming that the transgene is inserted at a single locus , individuals homozygous for the transgene are expected to produce 0 to less than 25% seed death if fully or partially complementing the meiotic functions of the endogenous A . thaliana CENH3 . Using this criterion to infer the zygosity of the transgene , we measured fertility of A . thaliana plants in which the endogenous CENH3 is replaced by L . oleraceum CENH3 or B . rapa CENH3 . We measured seed set and frequency of abnormal seeds in selfed siliques from three independent transformation events for each construct ( Fig . 1C ) . The complemented lines were comparable to wild type for both measures of fertility . Furthermore , viability-stained anthers from the same complemented lines showed live pollen numbers and appearance indistinguishable from wild type ( Fig . 1B ) . For L . oleraceum CENH3 complemented lines we further analysed meiosis cytologically with DAPI stained chromosome spreads from pollen mother cells ( PMCs ) in two T1 families , 2 and 19 . Prophase I of meiosis in both lines was indistinguishable from wild type ( S1A Fig . ) . Chromosome segregation in PMCs at both meiotic divisions was checked for segregation errors . In the T1 = 19 family , metaphase I ( n = 26 ) , anaphase I ( n = 7 ) , metaphase II ( n = 40 ) , anaphase II ( n = 5 ) and telophase II ( n = 21 ) PMCs were scored , none of which displayed segregation errors ( Fig . 2 ) . Careful inspection of all post-prophase I PMCs sampled revealed some limited chromosome fragmentation in one anaphase II cell ( S1B Fig . ) . Although , the origin of this cannot be ascertained at present , its low frequency is unlikely to compromise fertility . Thus , we conclude that CENH3 orthologs can complement the essential mitotic and meiotic functions of A . thaliana CENH3 under standard growing conditions . Next , we wanted to test how A . thaliana centromeres built on CENH3 variants functioned in comparison to those built on the native A . thaliana CENH3 . To do so , we crossed them as females with pollen from wild-type ( CENH3 +/+ ) Landsberg erecta ( Ler ) homozygous for the gl1–1 glabrous mutation , which confers a trichomeless phenotype . We chose Ler as the CENH3 wild-type parent because the complemented lines were generated in the Col-0 accession . This allows us to use polymorphisms between Col-0 and Ler to determine the parent of origin for all the chromosomes in the F1 . In a standard cross we expect only F1 diploid hybrids with trichomes . However , if replacing the endogenous CENH3 with natural variants creates weak centromeres , then we can expect mitotic missegregation in the F1 zygote . The first indication of abnormal segregation in these crosses was the observation that 14–47% seeds aborted during development ( Table 1 ) . In contrast to the uniformly tan-colored plump seeds generated when the complemented lines are selfed , dark nearly black shriveled seeds were seen in crosses to wild type . Upon germination of F1 seeds from L . oleraceum CENH3 and B . rapa CENH3 crosses , we recovered diploid , aneuploid and haploid progeny . All haploids were sterile and paternal on the basis of having a trichomeless appearance , an expression of the recessive gl1–1 mutation . We confirmed the haploid genome content of 11 phenotypically selected haploids by flow cytometry ( S2 Fig . ) . Crosses were between cenh3–1/cenh3–1 + CENH3 transgene females and pollen from wild type Landsberg CENH3 +/+ strain homozygous for the gl1–1 glabrous mutation . Sterile offspring expressing the recessive gl1–1 trichomeless phenotype were scored as paternal haploid . Offspring with developmental defects were scored as aneuploid . Fertile wild-type offspring were scored as diploid . For each CENH3 construct we tested two individuals from each of the three independent transformation events ( T1 families ) in crosses to wild-type Ler gl1–1 . Substantial variation in the frequency of haploids was observed between the different T1 families ( Table 1 ) . While the source of this variability is unclear , it is consistent with the variable haploid induction rates observed when GFP-tailswap is crossed to wild type . In cases where cenh3–1 is complemented with L . oleraceum CENH3 , the frequency of genome elimination ranged from 2 to 11% . For B . rapa CENH3 complemented cenh3–1 , the range was 1 to 2% . Although , L . oleraceum is more closely related to A . thaliana than B . rapa , substituting endogenous A . thaliana CENH3 with the L . oleraceum ortholog in an A . thaliana plant appeared to have a greater destabilizing effect on A . thaliana centromere as inferred from the larger frequency of genome elimination on average ( 6 ± 2 . 4% vs . 1 ± 0 . 2% ) . We have not observed any instances of aneuploidy and haploidy in the selfed progeny of the complemented lines ( S3 Fig . ) . In addition meiosis in L . oleraceum T1 families 2 and 19 , which generated the highest frequency of haploids and aneuploids , is wild type in appearance ( Fig . 2 ) . From the absence of meiotic defects during selfing , we infer that the segregation errors and genome elimination observed in the crosses to wild type ( CENH3 +/+ ) are not the byproduct of meiotic dysfunction in the inducer parent , but rather the consequences of postzygotic interactions in the hybrid embryo . From this we conclude that natural variation in CENH3 can cause centromere-mediated genome elimination and contribute to genetic instability through changes in ploidy . One of the hallmarks of centromere-mediated genome elimination is the generation of aneuploid progeny at a relatively high frequency ( ~30% for GFP-tailswap ) [20] . Aneuploids have imbalanced karyotypes that perturb gene dosage , with large and variable phenotypic consequences . A . thaliana aneuploids exhibit morphological phenotypes in a wide variety of traits including abnormal leaf morphology , irregular branching patterns and infertility [23] . Using these criteria , we estimated that in crosses of Ler gl1–1 ( as the wild-type pollen parent ) to L . oleraceum CENH3 and B . rapa CENH3 complemented lines , the incidence of aneuploidy is 11 . 3% and 8 . 3% respectively ( Table 1 ) . We selected 48 phenotypically aneuploid progeny from each cross for whole genome sequencing to determine the relative dosage of each chromosome using a bioinformatics approach . Chromosomes and subchromosomal regions that vary from the expected number of 2 can be readily identified by increased or decreased read count relative to the rest of the genome [23] . We identified chromosomal imbalances in 73 of the 96 individuals selected for sequencing ( S4 Fig . , S5 Fig . and S1 Table ) . In this dataset we found three classes of aneuploid chromosome types and an example of each is shown in Fig . 3 ( B–D ) . As a comparison diploid Col/Ler individual with 2 copies of each of the five A . thaliana chromosome is shown in Fig . 3A . The first class contains numerical aneuploids where whole chromosomes are duplicated , as exemplified by an individual trisomic for Chr3 ( Fig . 3B ) . The second class contains aneuploids with truncated chromosomes , such as , for example , an extra copy of Chr5 with a truncated left arm ( Fig . 3C ) . Lastly , the third class displays dosage variation consistent with chromosomes that shattered and have gained or lost DNA segments multiple times across the entire length of the chromosome . An example for a shattered Chr2 is shown in Fig . 3D . Based on our low pass sequencing analysis we cannot infer the chromosomal organization of these dosage variants presented here . Using SNPs between the parental lines , we were able to infer the origins of the copy variant regions ( SNP plots in Fig . 3A–D ) . In all three classes of dosage variants , the DNA contributing to the increased copy number originated from the transgenic Col-0 parent , in which the endogenous CENH3 had been replaced by an evolutionary variant . We even observed the loss of heterozygosity in the shattered Chr2 ( Fig . 3D ) , as a result of the complete loss of the Col-0 chromosomal regions . The largest fractions of aneuploids from these crosses were products of whole chromosome missegregation events ( Fig . 3E and F ) . However , there were also a considerable number of aneuploids with sub-chromosomal changes in copy number . This variation in dosage implies the creation of novel genetic karyotypes . In summary , centromeres built on CENH3 variants appear to missegregate in crosses to wild type . One consequence of which is aneuploidy and segmental dosage variants and with that the introduction of a broad range of phenotypic diversity [24] . Since our results negated the previously identified limits of CENH3 functional complementation , we decided to sample a larger evolutionary space . Flowering plants are divided into two major groups: monocots and dicots that diverged from each other 146–161 MYA . Rosids are the largest clade within the dicots , comprising of around 70 , 000 species including the model plant A . thaliana [25] . To better understand the extent of variation in CENH3 across the plant kingdom , we collated 67 CENH3 sequences from public databases that included homologs from green algae , mosses , monocots and dicots ( S2 Table ) . Using protein sequence from the HFD we generated a multiple sequence alignment and constructed a phylogeny of CENH3 in the plant kingdom ( Fig . 4A ) . This CENH3-HFD based gene tree was largely congruent with the accepted evolutionary relationships between these species ( Fig . 4A ) . The most striking feature of the tree is the size of its branches and the variation in their lengths , illustrating the rapid and variable rates of CENH3 evolution . We chose to test CENH3 from two additional species at increasing degrees of evolutionary distance from A . thaliana: grapevine ( Vitis vinifera ) , one of the earliest diverging rosid species considered a basal rosid , and corn ( Zea mays ) , a monocot . To test the functional complementation of these distant species , we made constructs expressing V . vinifera CENH3 and Z . mays CENH3 cDNA under control of the endogenous A . thaliana CENH3 promoter . These transgenes were transformed into cenh3–1/CENH3 heterozygotes . We recovered both V . vinifera CENH3 and Z . mays CENH3 transformants in a cenh3–1 homozygous background in the T1 generation ( Fig . 4B and S6 Fig . ) . V . vinifera and Z . mays CENH3 have 21 and 38 amino acid substitutions respectively , relative to the 97 amino acid positions in the HFD of A . thaliana CENH3 ( S7 Fig . ) . Hence , it was surprising that both V . vinifera CENH3 and Z . mays CENH3 were able to complement the embryo lethality of the cenh3–1 resulting in plants undistinguishable from the wild type . To the extent that the complemented lines were self-fertile , we can say that both variants also fulfilled the essential meiotic functions of A . thaliana CENH3 ( Fig . 4B and S6 Fig . ) . The L . oleraceum CENH3 gene has 12 amino acid substitutions in its HFD relative to A . thaliana and 31 in its N-terminal tail . We generated chimeric proteins in which the N-terminal tail of L . oleraceum CENH3 was fused to the HFD of A . thaliana CENH3 , and vice versa ( Fig . 1A ) . We assayed complementation of cenh3–1 and found that both chimeras complemented the embryo lethality of the cenh3–1 mutation in the T1 generation . The chimeric CENH3s were also similar to wild type with respect to pollen viability as determined by viability staining and in number and appearance of developing seeds within siliques ( Fig . 1B and 1C ) . We then tested the functionality of centromeres built on these chimeric CENH3 transgenes by making crosses to wild type . It was immediately apparent by visual inspection of the resulting F1 seeds that the two chimeras had entirely different effects . The F1 seeds from the chimera with A . thaliana N-terminal tail fused to L . oleraceum HFD ( AtNTT-LoHFD ) crossed to wild type appeared largely normal while most of the F1 seeds from the L . oleraceum N-terminal tail fused to A . thaliana HFD ( LoNTT-AtHFD ) were abnormal in appearance ( Table 1 ) . We failed to obtain F1 seed germination from crosses of LoNTT-AtHFD to the wild type except from a single T1 family . In this respect , the function of the chimera , LoNTT-AtHFD , is reduced compared to the full-length L . oleraceum CENH3 . We only recovered 124 F1 progeny from the LoNTT-AtHFD cross , of which 2 were haploids and 23 were phenotypically aneuploid . In contrast , we recovered a large number of F1 progeny from the crosses with AtNTT-LoHFD . However , out of a total of 554 F1’s none were haploids . This indicates that restoring the N-terminal tail to the endogenous sequence is sufficient to restore activity to a level similar to wild-type . Since our genetic assays highlight a critical role of the N-terminal tail sequence in segregation and genome elimination , we were interested in identifying patterns in its sequence evolution . N-terminal tails of CENH3 proteins are hyper-variable both in their amino acid sequence and length , ranging from 23 amino acids ( Pisum sativum ) to 194 amino acids ( Brachypodium distachyon ) . Thus , reconstructing the evolutionary history of N-terminal tails from alignments of distant CENH3 lineages is not possible . Instead , we decided to use an alignment free approach and used the motif search program MEME to identify short conserved blocks of sequence homology in the otherwise unstructured N-terminal tail . A similar approach investigating N-terminal tail evolution in Drosophila species identified three conserved blocks of homology shared by all CENH3 alleles in that clade [26] . Our analysis of N-terminal tails includes variation from a significantly broader evolutionary timescale , with CENH3 sequences ranging from green algae to flowering plants . We identified seven stretches of conserved protein sequences , which we have termed Blocks 1–7 ( Fig . 5A , S3 Table ) . The over-representation of Brassicaceae-clade specific motifs ( 4 of 7 Blocks ) is a reflection of our sampling bias , in which 22 of the 67 N-terminal tail sequences were from species within the Brassicaceae . Several interesting patterns were immediately apparent: First , Block 1 and Block 2 were identified in nearly all plant CENH3s and in canonical Histone H3 ( Fig . 5A ) . It appears that while the intervening sequence is highly variable in both length and content , the N- and C-terminus of N-terminal tails are evolving under strict constraint . These Blocks were not identified in H . sapiens CENH3 . Second , in several instances where a species’ genome carries two copies of CENH3 , there was differential retention of Blocks between the two copies , a situation analogous to sub-functionalization post gene duplication . For example , copy A of CENH3 in Arabidopsis lyrata is missing Block 6 but retained Block 3 , while copy B is missing Block 3 but has retained Block 6 . In Hordeum vulgare , the monocot-specific Block 7 is retained in copy A , but lost in copy B . Third , isolated Blocks were identified across long evolutionary distances ( Fig . 5B ) . For example , Brassicaceae-specific Block 4 was absent in all other lineages but present in V . vinifera , a basal rosid . Similarly , Block 6 that is present in most , but not all , Brassicaceae species , was also identified in two distant rosid species , Phaseolus vulgaris and Glycine max . The most parsimonious explanation for this pattern is that sequences homologous to Block 4 and Block 6 were present in the N-terminal tail of the ancestral CENH3 and were selectively retained or lost in the different rosid species . These observations suggest a modular evolutionary pattern where the constraints on individual Blocks are independent of one another . An outcome of this might be that the N-terminal tails acquire lineage-specific configuration of Blocks , thereby generating combinatorial sequence diversity . The results obtained in this study provide new and dramatically different information about CENH3 function and evolution from that previously available [21] . We observed wide complementation of a CENH3 loss-of-function mutation , while previous studies failed to obtain complementation except in the case of CENH3 from a very close relative . The difference lies quite simply in the use of untagged versus GFP-tagged CENH3 proteins in functional complementation assays . Furthermore , a recent study of CENH3CSE4 dynamics in yeast found that fusion of the GFP-tag to the CENH3CSE4 protein altered its function [27] . Taken together , it is apparent that presence of the GFP-tag significantly interferes with centromere function and protein modified with this fusion has limited use as a proxy for wild-type CENH3 activity . The role of CENH3 in centromere determination is thought to have originated in an early eukaryotic ancestor [28] . Functional homologs of CENH3 have been identified in plants , animals , fungi and protists [29 , 30] . This essential gene exists as a single copy in nearly all species . Given the absence of gene duplicates and opportunities for sub-functionalization , this diversity in CENH3 protein sequences is puzzling and begs the question: how conserved are the functional requirements for making a centromere ? This question has been asked in at least four different model organisms using primarily two assays: localization of evolutionarily distant CENH3s to the endogenous centromere and functional complementation of the endogenous CENH3 with evolutionary variants [18 , 21 , 31–33] . Two contrasting patterns of CENH3 functional conservation are apparent from the literature and this study . The first pattern is one of shared constraint over long evolutionary distances and the second is that of extreme lineage-specificity . In mammalian cells , GFP-tagged CENH3s from C . elegans and S . cerevisiae localized to centromeres . In addition , S . cerevisiae CENH3 rescued mammalian cells from mitotic arrest induced by depletion of the endogenous CENH3 [18] . In Arabidopsis , centromeric localization of complementing CENH3 does not extend as far as yeast [21] but CENH3 from Z . mays , a distant monocot species , can functionally substitute for the endogenous CENH3 ( Fig . 4B and S6 ) . In contrast , in D . melanogaster , GFP-tagged CENH3 from a species within the same genus failed to localize to centromeres [31] . In budding yeast , functional complementation of CENH3 is limited to the closely related hemiascomycetes [33] . Hemiascomycetes are unique in having ‘point centromeres’ that are genetically defined by a 125-bp sequence . Point centromeres are a derived evolutionary characteristic [28 , 34] and a plausible argument is that this specialized centromeric structure places severe lineage-specific constraints on CENH3 function , thereby restricting the limits of functional complementation . The results presented here argue that functional conservation despite sequence divergence is the norm , while stringent functional constraints might be symptomatic of a derived idiosyncratic centromere . In this study we have asked not only whether a divergent CENH3 can functionally complement the endogenous A . thaliana allele , but also how well it complements those functions by providing a quantitative measure of the effect of CENH3 divergence on segregation fidelity . This measure has been possible because A . thaliana , like most plants , has a high tolerance for genomic dosage imbalance [35–37] , thereby allowing recovery of the products of missegregation . Strikingly , complemented lines that had no fertility issues when fertilized by pollen of the same genotype , displayed large-scale segregation errors when crossed to wild type . Significant fractions of the recovered F1 progeny were either aneuploid or haploid ( Table 1 ) . In all cases the missegregated chromosomes originated from the parent expressing the divergent CENH3 ( Fig . 3 , S4 and S5 ) . This clearly implies that centromeres built on the divergent CENH3 , while able to complement essential functions , are deficient in comparison to the endogenous A . thaliana CENH3 . What is the molecular basis of this functional deficiency ? Answering this question constitutes an exciting next challenge since it will uncover species-specific adaptations to centromere function and shed light on what is driving the rapid evolution of this ancient biological structure . Genome elimination as a barrier to interspecies hybridization has been observed in several taxa [38] . It had been previously shown that engineering modifications to CENH3 , namely fusing an N-terminal GFP-tag and swapping the N-terminal domain with one from Histone H3 . 3 ( GFP-tailswap ) , causes segregation errors and genome elimination . Our results now show that naturally occurring divergence in CENH3 has the same effect . The most parsimonious explanation is that the underlying mechanistic basis of genome elimination in these different systems is shared while differing quantitatively in its outcome . In contrast to the male-sterile GFP-tailswap construct , CENH3 evolutionary variants are perfectly fertile when selfed , imposing no obvious fitness cost per se ( Fig . 1 and 2 ) . This highlights the fact that unlike the artificial GFP-tailswap construct , the naturally occurring mutations in CENH3 have evolved under functional constraint and can fulfill the conserved , essential functions even in the context of a non-native centromere , at least under standard growth conditions . However , in crosses to gametes with wild-type centromeres , the difference in parental CENH3s produces inviable ( aborted seeds ) and sterile ( aneuploid and haploid ) F1 progeny . In addition to these fitness penalties , the cross creates genetic novelty including instances of chromosomal breakage and shuffling of the resulting segments ( Fig . 3B-D ) . Aneuploidy and elimination of the haploid inducer genome are likely a linked phenomenon . Interestingly , fragmented chromosomes have been observed in other systems where genome elimination follows from an interspecific hybridization event [39 , 40] . In the natural barley wide crosses and in wheat and pearl millet hybrids , micronuclei formation is observed during the process of genome elimination [39 , 41] . Chromosomes within micronuclei could be targeted for elimination or be rescued by the cell , resulting in potential aneuploid progeny . While most aneuploid karyotypes have a deleterious fitness effect , recent studies have shown that aneuploidy is able to confer adaptive phenotypes under various stress conditions [42 , 43] . In summary , our data strongly supports a role for CENH3 divergence in speciation , not only as a means for creating a postzygotic reproductive barrier but also as a driver of genetic novelty . A major finding from our work is that it is divergence in the L . oleraceum N-terminal tail that is critical for the missegregation phenotype . Fusing A . thaliana N-terminal tail to a divergent HFD improved its function , while fusing a divergent N-terminal tail to the A . thaliana HFD corrupts its function . In fact this second chimera showed a more severe missegregation phenotype than the full-length divergent CENH3 ( Table 1 ) . This suggests that the two domains of CENH3 might be co-evolving with one another , thus in some cases a chimera between two non-adapted domains could create an allele that is worse than the sum of its individual parts . Nevertheless , our results show that , despite sequence divergence , the HFD of CENH3 from a distant species can be functionally interchanged . Domain-swap experiments have revealed that regions within the HFD are required for centromere localization [31 , 44] . A plausible hypothesis is that the structural and functional constraints on the HFD are essentially unchanging , while the N-terminal tail is evolving to accommodate lineage-specific differences in centromeric environment . Our examination of N-terminal tail sequences across the plant kingdom suggests a pattern where blocks of sequence homology are being lost and gained in a lineage-specific manner ( Fig . 5A ) . A tempting conjecture is that these blocks of homology represent functional modules , such as interactions with other centromere-associated proteins . If this was the case we could expect lineage-specific diversity in centromeric machinery , with the integration ( or subtraction ) of lineage-specific interactions into the ancestral centromere network . Consistent with this expectation , a recent study recently delineated the evolutionary trajectory of Umbrea , a neogene that has gained essential centromeric functions in specific Drosophila lineages [45] . While this is in no way conclusive , we propose that the idiosyncratic rewiring of centromeric chromatin constitutes a potential driving force for the evolution of the N-terminal tail of CENH3 . In summary , our results argue that while CENH3 from all species perform conserved functions , each CENH3 is adapted to its own unique cellular , most likely centromeric , environment . Why there should exist so many diverse solutions to the problem of packaging centromeric chromatin remains enigmatic . However , we demonstrate that this lineage-specific diversification of CENH3 has the potential to contribute to the genetic diversification and reproductive isolation of populations . Plants were transformed by the Agrobacterium floral dip method using standard protocols . Plants were grown under 16 hr of light/8 hr of dark at 20°C . For each cross , at least five flowers from an early inflorescence were emasculated and pollinated one day later with wild type pollen . F1 seeds were first sown in 0 . 5X MS plates containing 1% sucrose to maximize germination efficiency and then transplanted to soil . The L . oleracem CENH3 coding region including introns was PCR amplified from genomic DNA with the addition of SalI and XbaI sites at the ends . This PCR product was then cloned using standard restriction enzyme cloning into CP225 , a cassette vector generated by Ravi et al . ( 2010 ) [21] . This vector is based on pCAMBIA1300 and carries the endogenous A . thaliana CENH3 promoter region i . e . 1489 bp upstream of the ATG , followed by a small linker region containing SalI and XbaI sites and finally the CENH3 transcriptional terminator i . e . 585 bp downstream of the STOP codon . All other constructs were cloned into a new Gateway-compatible cassette vector SM2 that was derived from the above CP225 . To construct this vector , we used three-fragment multi-site gateway technology ( Life technologies , cat# 12537–023 ) that allows simultaneous assembly of three DNA fragments in a defined order into a destination vector . The first and third fragments are the endogenous A . thaliana CENH3 promoter and terminator respectively , while the second fragment can be any CENH3 variant being tested . We PCR amplified the promoter and terminator sequences from CP225 flanked by the appropriate attB sites and recombined them via the BP reaction into pDONR 221 P1-P4 and pDONR 221 P3-P2 respectively , generating the following entry clones: pENTR L1-promoter-L4 and pENTR L3-terminator-L2 . Next , we integrated these two along with pENTR R4-pLac-Spec-R3 , the control entry clone for the second fragment , into the destination vector through a single LR reaction . The destination vector was a generous gift from the Pikaard Lab and was a modified pEARLEYGATE302 binary vector that has an additional ampicillin resistance gene for bacterial selection . We then did a reverse BP reaction with this intermediate expression plasmid and pDONR 221 P4r-P3r to replace the placeholder in the second fragment with the Gateway negative selection cassette [CmR-ccdB] generating the final cassette vector , SM2 = CENH3 promoter-attL4-CmR-ccdB-attL3-terminator in pEARLEYGATE302 . The B . rapa CENH3 genomic sequence was PCR amplified from the GFP-tagged B . rapa CENH3 plasmid generated in Ravi et al ( 2010 ) [21] . A chimeric transgene combining the A . thaliana N-terminal tail domain with L . oleraceum HFD was constructed by overlapping PCR . The N-terminal domain included genomic sequence coding for CENH3 starting from the “ATG” up to but not including the “PGTVAL” motif and the HFD extended from the “PGTVAL” motif to the STOP codon . The reciprocal construct with L . oleraceum N-terminal tail domain and A . thaliana HFD was similarly constructed . Transgenic variants outside the Brassicaceae were generated using CENH3 cDNA . Z . mays CENH3 was PCR amplified from plasmid generated in Ravi et al ( 2010 ) [21] . CENH3 cDNA from V . vinifera was synthesized by GenScript USA Inc . Piscataway , NJ based on the Genbank sequence , 225454488 . Genomic DNA preparation and PCR genotyping were performed using standard methods . cenh3–1 was genotyped with dCAPS primers . To genotype the cenh3–1 mutation in lines with the construct A . thaliana N-terminal tail domain fused to L . oleraceum HFD , we first performed a PCR reaction with one primer outside the CENH3 promoter genomic DNA fragment present in the transgene . This PCR product was then used as the template in the dCAPS genotyping reaction . For each construct transgene-specific PCR primers were designed and used to confirm the genotype of each transgenic line . Primer sequences are available on request . Representative images of rosettes were taken 25 to 30 days after germination . The percentage of normal seeds was determined by visual inspection using a dissecting microscope . On average , seeds from five individual siliques were pooled and counted for one individual from each T1 family identified as CENH3 transgene +/+ cenh3 -/- . Alexander staining of anthers was done according to published protocols [46] . DAPI stained male meiotic chromosome spreads were prepared as described in Ross et al . [47] , and imaged using an Olympus BX61 epifluorescence microscope and Digital Scientific SmartCapture 3 software Flow cytometric determination of genome content was performed on floral buds using published protocols [48] . 0 . 1g leaf tissue from aneuploid plants were collected and purified using DNA Phytopure Kit ( GE ) . Genomic DNA libraries were prepared using the standard NEB Next DNA Library Prep with NEXTFlex-96 Adapters from BIOO Scientific , pooled and sequenced on Illumina HiSeq 2000 for 50bp single reads . The resulting reads were mapped to TAIR10 using BWA followed by chromosome dosage analysis using the protocol described in Henry et al ( 2010 ) [23] . All the individuals that were sequenced and analyzed are identified with a unique FRAG identifier and are described in S2 Table . Reference IDs for all sequences used in this study are available in S1 Table . Multiple alignments of protein sequences encoding the histone fold domain of CENH3s was generated using MUSCLE and refined manually [49] . Evolutionary analyses were conducted in MEGA6 [50] . Phylogenetic history was inferred using the Maximum Likelihood method . The analysis involved 71 protein sequences . All positions containing gaps and missing data were eliminated . There were a total of 85 positions in the final dataset . MEME [51] with default parameters was used to identify statistically significant blocks of sequence homology in N-terminal tails extracted from 67 plant CENH3 sequences available from public databases .	As populations evolve into new species they acquire mutations that are compatible with their own genetic background , but often lead to defects when crossed to others . Here , we show that naturally evolved differences in the centromere-specific histone H3 ( CENH3 ) can contribute to this process . Unlike canonical histones , CENH3 differentiates rapidly even between closely related species . To better understand the functional role of natural CENH3 variation , we complemented a null allele of Arabidopsis with progressively more distant orthologs . Contrary to previous findings , we discovered that all tested variants , even the highly diverged maize CENH3 , could restore normal growth and reproduction in selfing individuals . However , when crossed to the wild type , hybrid progeny suffered from extensive mis-segregation . Genotypes include simple aneuploids , novel genetic rearrangements , and in extreme cases haploids where all the chromosomes from one parent are lost . This indicates that while wide variation in CENH3 is compatible with its essential function , epigenetically different centromeres do not function well when brought together in a hybrid embryo . A better understanding of haploid generation would have profound effects on plant breeding and our results suggest that the natural variation of CENH3 could offer a cache of testable variation .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[]	2015	Naturally Occurring Differences in CENH3 Affect Chromosome Segregation in Zygotic Mitosis of Hybrids
The omnipresence of allosteric regulation together with the fundamental role of structural dynamics in this phenomenon have initiated a great interest to the detection of regulatory exosites and design of corresponding effectors . However , despite a general consensus on the key role of dynamics most of the earlier efforts on the prediction of allosteric sites are heavily crippled by the static nature of the underlying methods , which are either structure-based approaches seeking for deep surface pockets typical for “traditional” orthosteric drugs or sequence-based techniques exploiting the conservation of protein sequences . Because of the critical role of global protein dynamics in allosteric signaling , we investigate the hypothesis of reversibility in allosteric communication , according to which allosteric sites can be detected via the perturbation of the functional sites . The reversibility is tested here using our structure-based perturbation model of allostery , which allows one to analyze the causality and energetics of allosteric communication . We validate the “reverse perturbation” hypothesis and its predictive power on a set of classical allosteric proteins , then , on the independent extended benchmark set . We also show that , in addition to known allosteric sites , the perturbation of the functional sites unravels rather extended protein regions , which can host latent regulatory exosites . These protein parts that are dynamically coupled with functional sites can also be used for inducing and tuning allosteric communication , and an exhaustive exploration of the per-residue contributions to allosteric effects can eventually lead to the optimal modulation of protein activity . The site-effector interactions necessary for a specific mode and level of allosteric communication can be fine-tuned by adjusting the site’s structure to an available effector molecule and by the design or selection of an appropriate ligand . The traditional emphasis on complementarity between the drug and the catalytic site has inarguably formed a foundation in the current drug discovery approaches . However , many important drug targets share a conserved substrate binding site [1–5] , rendering drug toxicity as a result of the off-target binding . The allosteric regulation of protein activity via effector binding has been increasingly favoured in the drug discovery [1 , 2 , 6] . It is well recognized that potentially druggable allosteric sites are ubiquitous in most if not all dynamic proteins [7] , turning the key advantage of targeting allosteric sites [5]—non-competitive fine-tuning of protein activity at a distance—into a new paradigm in the drug design . For example , it was shown that allosteric drugs provide a way to modulating the activity of kinases that underlie a multitude of human diseases , bypassing the problem of low specificity with the conserved ATP binding pocket [8 , 9] . The allosteric drugs for GPCRs provide greater subtype selectivity among GPCR receptor families , while avoiding receptor desensitization typical for the orthosteric ones [10–12] . Currently , among the notable marketed drugs targeting GPCRs are cinacalcet [13] and maraviroc [14] , an allosteric agonist for calcium-sensing receptor and an antiretroviral allosteric antagonist for the CCR5 receptor , respectively . One of the major hurdles in the development of allosteric drugs lies in the finding of allosteric sites [5 , 15–17] , for which a repertoire of experimental and computational methods is being developed . High-throughput fragment-based screening using a large chemical library formed the main thrust in the identification of potential allosteric sites and lead compounds in pharmaceutical research [18–20] . A number of site-directed approaches have been employed for detection and probing of allosteric sites and their modulatory effects , including disulfide trapping [21] , alanine scanning [22] , hydrogen-deuterium exchange mass spectrometry [23 , 24] and photoaffinity [25 , 26] . However , above experimental approaches while powerful , are relatively costly and time-consuming compared to any extensive analysis performed in silico . Computational approaches for finding the allosteric sites can be broadly classified as sequence-based and structure-based methods [3 , 4 , 27] . Sequence-based techniques utilize sequence homology inferred from the multiple sequence alignment to identify the co-evolving amino acids that constitute catalytic and allosteric sites [28] . However , complexity of the site-ligand interactions and energetics result in strong limitations on the predictive power of the sequence-based approaches [5 , 28 , 29] . Structure-based methods analyse binding pockets based on their topological and physicochemical features [5 , 30] . These approaches are strongly biased towards binding pockets that exhibit detectable curvature in the static 3D structure , in which case latent allosteric sites that can exist in a subset of a protein conformational ensemble may be left undetected . We have recently introduced a structure-based perturbation model of allostery [31] , which quantitatively describes the causality and energetics of allosteric communication , by simulating ligand binding as a local alteration in the protein inter-residue network of interactions . Because of the observation that perturbation at allosteric sites can affect distant functional site via modifying the energetics of the whole protein and assuming the reversibility of allosteric signalling , we hypothesized here that allosteric sites could be detected by perturbing the functional ones [5] . In order to test this hypothesis and to explore its predictive power , the reverse perturbation approach was developed here . Using a heterogeneous set of 13 classical allosteric proteins from previous studies [31–33] , dubbed here the “classical set” , we found that perturbation at the functional sites allows one to identify known allosteric ones . In order to estimate the predictive power of reverse perturbation method , it was necessary to introduce an operational definition of the allosteric site in the framework of elastic network model of protein . Specifically , assuming that allosteric signalling occurs between non-overlapping distant sites , a distance condition was set to ensure communication and not direct physical interaction between residues of the functional and allosteric sites . Using the classical set and a new collection of 41 allosteric proteins from the benchmarking set [34] , predictive power of the reverse perturbation method is shown . Furthermore , we argue that in addition to the widely addressed case of predicting latent allosteric sites [35] , the task of inducing allosteric signalling with a desired and tunable level of agonistic/antagonistic activity can be naturally formulated . We show that the reverse perturbation approach opens the way for achieving above goal , allowing one to find targets for allosteric effectors and to optimize structures/compositions and interactions of corresponding regulatory site-effector pairs that will provide a desired allosteric response at the functional site . In the structure-based statistical mechanical model of allostery [31] , ligand binding is modelled as a perturbation of the harmonic network associated with the protein . The perturbation is defined as a stiffening harmonic restraint applied to the residue pairs that compose the binding site of interest . As a result of the perturbation , residues in the binding site experience an increase of rigidity in comparison with the residues in the unperturbed binding site . We have shown that in the event of allosteric communication , the perturbation of the allosteric sites induces a response at the functional ones by altering their energetics and fluctuation dynamics [31] . Specifically , as a result of the perturbation in a binding site , a per-residue free energy change Δgi is obtained for each residue of the protein , which is the signature of the change in the amount of work exerted in the environment of residue i . Here we investigate the hypothesis of reversibility of allosteric communication , according to which one can detect a change in the free energy on the residues of allosteric sites when a perturbation–simulated binding–is applied in the functional sites ( Fig 1 ) . To begin with , we analysed a set of classical allosteric proteins , dubbed here “classical set” [32 , 33] , to directly test the hypothesis of the reverse perturbation approach in identifying allosteric sites in proteins of different sizes , oligomerization states and functions . Table 1 contains the average free energy changes ( averaged over all residues in corresponding sites in case of oligomeric proteins ) observed as a result of the allosteric signalling in known allosteric sites ( ΔgA ) and in the restrained functional sites ( ΔgF ) in 13 proteins of the classical set . First column: Protein name , oligomerization state , and total number of residues . Second column: PDB ID of the protein . The third and fourth columns represent the perturbation applied to functional sites and the resulted ΔgF values , respectively ( see Eq 7 in Materials and Methods ) . The fifth and sixth columns show the known allosteric sites ( A—allosteric activator , I—allosteric inhibitor ) and the ΔgA values in the allosteric sites in response to the reverse perturbation , respectively . The seventh column provides the average free energy differences ΔgU over all the residues in the protein , giving an estimate on the protein stability changes caused by the applied perturbation . The last column gives the proximity between the corresponding functional and allosteric sites in a protein subunit ( see Eq 8 in Materials and Methods ) . The Anthranilate synthase ( AnthS ) from Serratia marcescens is a heterotetramer consisting of a dimer of TrpE and TrpG subunits . Upon perturbation of the glutamine substrate binding sites , the tryptophan ( inhibitor ) binding sites in the larger TrpE subunits show a positive free energy difference response ( ΔgTRP ( 2×GLU ) = 0 . 42 kcal/mol ) , compared to the overall decrease in the free energy of the structure and stabilization of the entire TrpG subunits upon perturbation ( ΔgAnthS ( 2×GLU ) = -0 . 29 kcal/mol , Fig 2A and Table 1 ) . The aspartate carbamoyltransferase ( ATCase , Fig 2B ) from Escherichia coli is a heterododecameric enzyme composed of two trimers of catalytic subunits in the centre of the oligomer and three dimers of the peripheral regulatory subunits . Simulated binding in the catalytic sites of ATCase ( PAL sites ) increases the configurational work exerted at the allosteric domain in the regulatory subunits that contain allosteric activator ATP and inhibitor CTP ( ΔgATP−CTP ( 6×PAL ) = 1 . 75 kcal/mol , Table 1 ) , but not the zinc-binding domain which plays a structural role in the complex assembly . Allosterically activated by cyclic AMP , the catabolite activator protein ( CAP ) from E . coli is a classical model system of transcriptional activation . The cAMP binding causes large rotation in the DNA-binding domain of CAP , which is a prerequisite for interactions with DNA [36] . Negative cooperativity observed for the binding of second cAMP molecule was discussed earlier [31] , and it was shown that mutations in the cAMP-binding pocket decreasing the affinity between CAP and cAMP enhance negative cooperativity [37] . Using the DNA-bound conformation of CAP , we simulated binding in the DNA interaction sites and observed that allosteric site in the N-terminal cAMP-binding domains exhibits a large positive free energy change ( ΔgcAMP ( 2×DNA ) = 1 . 73 kcal/mol ) , higher than the average free energy change ( ΔgCAP ( 2×DNA ) = 0 . 71 kcal/mol ) of the homodimer ( Fig 2C and Table 1 ) . For the homotetrameric 3-deoxy-D-arabino-heptulosonate-7-phosphate synthase ( DAHPS ) from E . coli , binding of the phosphoenolpyruvate ( PEP ) substrate to the TIM barrel domain was simulated . We observed a positive free energy change ( ΔgPHE ( 4×PEP ) = 0 . 80 kcal/mol ) in the center of the quaternary complex , where the inter-subunit pockets for allosteric inhibitor phenylalanine ( PHE ) are located , while the whole structure yields rather a modest positive free energy change ( ΔgDAHPS ( 4×PHE ) = 0 . 15 kcal/mol , Fig 2D and Table 1 ) . The dimeric arginine kinase ( DAK ) from Apostichopus japonicus is known to exhibit negative cooperativity , in which binding of substrates ( ARG and ATP ) to one subunit causes a conformational change in the free subunit . Specifically , “large outward reorganization of the other subunit ( open state ) which result in the release of products” , that precludes substrate binding was experimentally observed [38] . We found that perturbing the catalytic sites in one subunit destabilizes the unperturbed one ( ΔgARG ( 1×ARG/ATP ) = 1 . 01 kcal/mol and ΔgATP ( 1×ARG/ATP ) = 0 . 48 kcal/mol , Fig 3A and Table 1 ) , presumably reducing its affinity to substrates . The NAD binding sites in the Rossmann fold were perturbed in the human NAD-dependent malic enzyme ( NADME ) . Fig 3B shows that configurational work propagates from the outer domains of the homotetramer to the dimerization interface where the allosteric activator fumarate ( FUM ) binds , increasing the amount of work exerted in these sites ( ΔgFUM ( 4×NAD ) = 2 . 90 kcal/mol ) to a much larger extent than the overall free energy change ( ΔgNADME ( 4×NAD ) = 0 . 67 kcal/mol , Fig 3B and Table 1 ) . Noteworthy , latent allosteric sites could also exist in the extended area near the core of this quaternary complex where large positive configurational work is observed as a result of perturbation at the functional sites . A classical allosteric enzyme with a long history of studies , the phosphofructokinase ( PFK ) from Bacillus stearothermophilus , is a homotetramer . When the substrate and cofactor binding sites for fructose-6-phosphate ( F6P ) and ATP were perturbed , the allosteric response at the regulatory exosites ( that bind activator ADP and inhibitor phosphoenolpyruvate ( PEP ) ) located in the dimerization interface is manifested in the increase of the free energy ( ΔgADP ( 4×F6P/ATP ) = 1 . 35 kcal/mol and ΔgPEP ( 4×F6P/ATP ) = 1 . 47 kcal/mol ) detected in these sites , respectively ( Fig 3C and Table 1 ) . The D-3-phosphoglycerate dehydrogenase ( PGDH ) of E . coli is a homotetramer with a ring-shaped quaternary structure . Simulated binding at the substrate AKG and cofactor NAD sites , which are located in the interface between corresponding substrate and cofactor domains , allows one to identify the binding sites for allosteric inhibitor serine ( SER ) in the distant peripheral regulatory domains . The serine binding sites show a large positive free energy change ( ΔgSER ( 4×AKG/NAD ) = 2 . 62 kcal/mol ) in comparison with the negligible background free energy increase ( Fig 3D and Table 1 ) . The overall change in the free energy of the small monomeric human protein tyrosine phosphatase 1B ( PTP1B ) is negative upon restraining the catalytic site BPM ( ΔgPTP1B ( 1×BPM ) = -2 . 12 kcal/mol ) , pointing to the strong stabilizing role of this perturbation . We observed a slight increase of the free energy at the known allosteric site 892 ( Δg892 ( 1×BPM ) = 0 . 42 kcal/mol , Fig 4A and Table 1 ) . Additionally , the beta sheet distant from the catalytic site exhibited large increase in the free energy change , suggesting the presence of a potential latent allosteric site ( Fig 4A ) . The uracil phosphoribosyltransferase from Sulfolobus solfataricus ( SSUPRT ) is a homotetramer that catalyzes the production of uridine 5′-monophosphate ( UMP ) . As a result of perturbing the catalytic sites UMP in the outer regions ( Fig 4B ) , residues in the binding site for the allosteric inhibitor CTP at the central interface show the positive free energy change ( ΔgCTP ( 4×UMP ) = 2 . 05 kcal/mol , Table 1 ) compared to the overall small average free energy change ( ΔgSSUPRT ( 4×UMP ) = -0 . 05 kcal/mol ) . The homodimer of threonine synthase ( ThrS ) from Arabidopsis thaliana features an extensive interface between the subunits , in which the allosteric activator S-adenosylmethionine ( SAM ) binds . Similar to SSUPRT , when the protein’s catalytic sites pyridoxal-L-phosphate PLP are stabilized , the configurational work exerted on the binding sites for SAM increases ( ΔgSAM ( 2×PLP ) = 2 . 27 kcal/mol , Table 1 ) in contrast to the small average free energy change ( ΔgThrS ( 2×PLP ) = 0 . 05 kcal/mol , Fig 4C ) . The bovine glutamate dehydrogenase ( BGDH ) and the glucosamine-6-phosphate deaminase ( G6PD ) from E . coli , are large homohexamers consisting of a dimer of trimers , and a trimer of dimers , respectively . The well-studied allosteric regulation of BGDH features a repertoire of metabolites as well as several characterized allosteric sites [39 , 40] . It is known that binding of glutamate ( GLU ) substrate and NADP+ ( NDP ) cofactor to the catalytic cleft elicits large-scale conformational changes in this complex molecular machine . For example , the antenna region ( Fig 5A ) , which consists of intertwined alpha helices and is evolutionarily conserved in the animal kingdom , is essential to the allostery of BGDH . The antenna serves as the inter-subunit relay of allostery , and its motion is regulated by the allosteric activator ADP and inhibitor GTP . Remarkably , we observed the large increase in the free energy of allosteric response ( 5 kcal/mol ) in the antenna region upon perturbation of the catalytic sites , in comparison to the small background free energy change ( ΔgBGDH ( 6×GLU/NDP ) = 0 . 31 kcal/mol , Fig 5A and Table 1 ) . On the other hand , a weak response is detected in the ADP and GTP binding sites ( ΔgADP ( 6×GLU/NDP ) = 0 . 18 kcal/mol and ΔgGTP ( 6×GLU/NDP ) = -0 . 18 kcal/mol , Table 1 ) . The configurational work exerted in the antenna region is consistent with experimental data , which shows rotation of the helices as the bound catalytic cleft closes [39 , 40] . We found that , in agreement with experimental data in which the presence of antenna region was shown to be required for allosteric signalling , this region plays a role of the allosteric modulator , suggesting , in turn , the presence of latent allosteric sites in this region ( Fig 5A ) . In the case of homohexameric G6PD ( Fig 5B ) , the allosteric response results in the positive free energy change globally distributed in the complex upon perturbation in all catalytic sites ( alpha-D-glucosamine 6-phosphate ( AGP ) , ΔgG6PD ( 6×AGP ) = 0 . 66 kcal/mol ) , including the inter-subunit interface where the allosteric activator N-acetylglucosamine 6-phosphate ( 16G ) binds ( Δg16G ( 6×AGP ) = 0 . 78 kcal/mol , Table 1 ) . In particular , residues at the core appeared to be highly affected by the perturbation , especially Cys219 , which forms disulphide bridges between subunits . Perturbing the catalytic site of one subunit essentially freezes the entire subunit including the proximal allosteric site for the activator 16G , increasing the free energy in the remaining unperturbed subunits ( S1 Fig ) . It was shown in X-ray crystallography [41] and fluorescence spectroscopy [42] experiments that the substrate binding to the catalytic site of G6PD induces structural and dynamic changes in all subunits of the protein , consistent with our observations . The classical set of proteins analysed here contains an equal proportion of protein structures in the apo and bound forms . Using crystal structures of PFK and PGDH with or without bound ligands , we show that the free energy profiles are largely similar ( S2 and S3 Figs ) , allowing to work with only one available structure . In all proteins of the classical set , a positive free energy difference was detected in distant allosteric sites upon perturbation of the substrate and/or cofactor binding , as well as an anticipated negative one in the regions where the perturbation is applied . Despite the different modes of allosteric regulation , binding sites for both allosteric activator and inhibitor exhibit an increase of configurational work exerted upon perturbation of the functional ones . For example , in enzymes with overlapping binding sites for activator and inhibitor , such as the ATCase ( Fig 2B ) and the PFK ( Fig 3C ) , both sites show positive Δgi values . Therefore , the positive free energy difference as a result of the functional site perturbation serves in our approach as the standard quantitative indicator for the allosteric sites . The profiles of the free energy changes due to the reverse allosteric signalling upon perturbation at the functional sites typically show rather extensive areas of the positive free energy difference , which encompass the binding sites of known allosteric effectors . Therefore , the locations that yield a large free energy change can , actually , be allosterically coupled to corresponding catalytic site , and , therefore , can potentially contain unknown/latent allosteric sites . For example , the reverse perturbation analysis suggests the presence of latent allosteric sites in the large TrpE subunit of AnthS ( Fig 2A ) , in the cAMP-binding domain of CAP ( Fig 2C ) , in PTP1B ( Fig 4A ) , as well as in the dimerization interface in NADME ( Fig 3B ) , PFK ( Fig 3C ) and ThrS ( Fig 4C ) . Furthermore , the antenna region in BGDH ( Fig 5A ) and multiple locations in G6PD ( Fig 5B ) can also contain latent allosteric sites that provide complex allosteric regulation of these large protein complexes . Fig 5 shows two cases , BGDH and G6PD , in which it was not possible to identify known regulatory sites proximal to the perturbed catalytic ones . In these proteins , the perturbation of the catalytic site stabilizes the region proximal to the binding sites due to the direct interaction of residues in these adjacent sites . These cases show that the separation of allosteric and the regulated functional sites is instrumental in allosteric communication , as postulated in the seminal Monod-Changeux-Jacob paper [43] that allosteric proteins should “…possess two , or at least two , stereospecifically different , non-overlapping receptor sites . ” Therefore , for any high-throughput analysis and prediction of regulatory exosites , it is crucial to introduce a quantitative measure for spatial separation between catalytic and allosteric sites . Here , we have devised an operational definition of the allosteric site in the framework of the elastic network model of protein . To obtain a quantitative criterion for defining the allosteric site we introduce a notion of proximity , which is the fraction of interacting residues among all possible pairs that can be formed between residues of two sites–functional and the candidate allosteric ones . The distance cutoff for defining the interaction between residues is chosen according to the distance cutoff ( 11 Å ) used in the microscopic allosteric potential of the original model [31] ( see also Materials and Methods ) . Taking the most conservative approach , we have chosen the upper limit cutoff 11Å , which allows one to predict most of the allosteric sites . A site is considered allosteric if the corresponding proximity with the functional site does not exceed the threshold value ( selection of the threshold value is explained below ) . We illustrate this definition by two homologous homotetramers , fructose 1 , 6-bisphosphatase 1 ( FBPase 1 ) from E . coli ( PDB ID: 2q8m , Fig 6A ) and from Sus scrofa ( PDB ID: 1kz8 , Fig 6B ) , which control the gluconeogenesis pathway [44] . Adenosine monophosphate ( AMP ) and glucose-6-phosphate ( BG6 ) bind to two distinct allosteric sites of FBPase 1 from E . coli , inhibiting hydrolysis of fructose 1 , 6-bisphosphate ( FBP ) [45] . The inhibitor binding at BG6 site would perturb the adjacent active FBP site via direct interaction between residues in these overlapping sites , whereas the distant AMP site regulates the FBP site by long-range allosteric signalling . Therefore , using the reverse perturbation approach , the distant AMP site can be readily identified based on the increase in free energy of allosteric response ( ΔgAMP ( 4×FBP ) = 1 . 67 kcal/mol ) . However , simulated binding in the FBP site ( ΔgFBP ( 4×FBP ) = -0 . 77 kcal/mol ) decreases the free energy of the adjacent BG6 site ( ΔgBG6 ( 4×FBP ) = -1 . 62 kcal/mol ) , which is not a true allosteric site because of the 10% proximity to the FBP site ( Fig 6A ) . Non-allosteric nature of interactions between the BG6 and FBP sites can be further confirmed by the weak decrease of the free energy observed upon direct perturbation via simulated binding to BG6 ( ΔgFBP ( 2×BG6 ) = -0 . 1 kcal/mol ) . On the contrary , the direct perturbation at the distant AMP ( inhibitor ) binding site increases the configurational work exerted at the FBP site ( ΔgFBP ( 4×AMP ) = 0 . 44 kcal/mol ) , yielding the allosteric nature of communication between these sites . The drug screening against FBPase 1 from Sus scrofa revealed that one of the anilinoquinazoline compounds , PFE , inhibits gluconeogenesis by binding to an allosteric site at the subunit interface [46] . Upon perturbing the catalytic FBP site , both AMP and PFE sites ( with 0 and 2% proximity to the FBP site , respectively ) , displayed a similar increase in the free energy ( ΔgAMP ( 4×FBP ) = 1 . 36 kcal/mol and ΔgPFE ( 4×FBP ) = 1 . 45 kcal/mol , Fig 6B ) . The configurational work exerted at the PFE site shows that the low proximity between functional and allosteric sites is a necessary condition for the correct definition of the latter . Additional analysis of several proteins that result in both successes and failures in detection of allosteric sites have led to the following operational definition: a site is considered allosterically coupled to the regulated catalytic one if the proximity between them is no more than 2% . The sites’ proximities obtained for each protein in the classical set ( Table 1 ) show that all proteins have non-overlapping functional and allosteric sites except the BGDH and G6PD , where the proximities are 7 and 10% , respectively , hence failed to be detected as allosteric ones . This operational definition is important in order to correctly estimate the predictive power of reverse perturbation approach , allowing us to operate with proteins with different architectures and mutual locations of the functional and allosteric sites . With the set of allosteric sites obtained on the basis of above operational definition , one can analyse the predictive power of the reverse perturbation approach . A successful prediction of allosteric sites is possible if their residues will exhibit a large free energy change upon perturbation of the functional site . The receiver operating characteristic ( ROC ) curve is used here to quantify the proportion of true positive ( those that belong to a known allosteric site ) and false positive ( those not belonging to a known allosteric site ) among residues with a large free energy change , upon perturbation at the corresponding functional sites ( see Materials and Methods for details ) . Using the classical set of allosteric proteins , we show that the true positive rate increases more rapidly than the false positive rate for most allosteric sites , indicating that the majority of residues of known allosteric sites ( true positives ) are indeed located near the positive tail of the Δgi distribution ( Fig 7A ) . The known allosteric sites in the NADME ( Fig 3B ) , PGDH ( Fig 3D ) and SSUPRT ( Fig 4B ) can be precisely determined based on the large increase in the free energy . It is important to emphasize , however , that presence of latent allosteric sites challenges the task of validating the prediction of the known ones , as some of the residues belonging to these sites can be erroneously scored as false positives . For example , the residues that show a large increase in Δgi values in potential allosteric sites , such as the antenna region in the BGDH , turn out to be false positives . Good predictive power can be achieved for known allosteric sites in the ATCase ( Fig 2B ) , DAHPS ( Fig 2D ) , DAK ( Fig 3A ) , PFK ( Fig 3C ) and ThrS ( Fig 4C ) . However , for the AnthS ( Fig 2A ) , CAP ( Fig 2C ) and PTP1B ( Fig 4A ) , the ROC curves are close to the diagonal line , indicating low predictive power for the known allosteric sites in these proteins ( Fig 7A ) . This is likely due to the pronounced increase in free energy over the protein domains caused by catalytic site perturbation , suggesting the presence of latent allosteric sites that contribute to the false positives . For DAK ( Fig 3A ) , the ROC curves for ARG ( Fig 7A ) and ATP binding sites in the free subunit ( Fig 7B ) vary greatly as the former site exhibits a higher increase in free energy , allowing its detection as the allosteric site . In the BGDH ( Fig 5A ) and G6PD ( Fig 5B ) , the regulatory exosites are too close to the restrained catalytic site , hence , not satisfying the aforementioned operational definition ( Fig 6B ) . To complement the classical set of 13 allosteric proteins , we have used the operational definition of allosteric sites to obtain 41 proteins with 48 unique experimentally-determined allosteric sites from the benchmarking set of allosteric proteins [34] ( S1 Table ) . We calculated the free energy profile for each of these proteins upon simulated ligand binding at the functional site ( S4 Fig ) . Similar to the classical set , the predictive power of the reverse perturbation approach was estimated for each protein in this heterogeneous additional set . The area under the ROC curves ( AUC ) indicates good predictive power of the method ( Fig 7C ) , showing that reverse perturbation approach allows one to successfully detect the majority of known allosteric sites in the additional set . In addition to the detection of known allosteric sites , the reverse perturbation approach delineates extended protein regions which are also characterized by the positive free energy change . This observation along with multiple indications of the presence of latent allosteric sites [5] in different proteins show that allosteric response can also be artificially induced by the interactions with rationally selected sets of residues belonging to latent or de novo designated allosteric sites . A classical allosteric enzyme , PFK ( Fig 3C ) , which is regulated by the activator ADP and inhibitor PEP binding to the overlapping binding sites , serves as an excellent illustration that large difference in allosteric response can be caused by minor changes in the composition of an allosteric site ( Fig 8 ) . Specifically , ligand binding to the PEP site , which is a part of the larger ADP binding site results in a mild increase of free energy in the functional F6P site ( ΔgF6P ( 4×PEP ) = 0 . 30 kcal/mol ) compared to the global free energy change of the homotetramer ( ΔgPFK ( 4×PEP ) = 0 . 10 kcal/mol ) . At the same time , a large increase in free energy is observed at F6P site upon simulated binding to ADP site ( ΔgF6P ( 4×ADP ) = 0 . 70 kcal/mol , Fig 8 ) . The modes of regulation in PFK , along with many other allosteric enzymes with overlapped activator and inhibitor binding sites in the literature [5] ( see also , for example , ATCase , Fig 2B ) , highlights the importance of tuning allosteric response by varying the binding site composition . We consider here as case studies , the subunit interfaces of the NADME ( Fig 3B ) , ThrS ( Fig 4C ) and FBPase 1 ( Fig 6A ) , which can apparently be used to induce a required allosteric response . In our model , an allosteric site consists of a set of residues whose perturbation results in the large free energy change in the corresponding functional site . In a given protein , to obtain a magnitude of regulation comparable with that of the native effectors , the allosteric response at the functional sites originating from the newly designated regulatory sites should be comparable with that obtained from known allosteric sites . The homotetrameric NADME is allosterically activated by the binding of fumarate ( FUM ) at the dimerization interface , causing a slight decrease in free energy at the catalytic site ( ΔgNAD ( 4×FUM ) = -0 . 18 kcal/mol ) . Based on the reverse perturbation analysis in NADME ( Fig 3B ) and in order to illustrate the possibility to induce an allosteric response , four sites were putatively defined in the protein region corresponding to the increase of the free energy: site 1 ( red ) and site 2 ( green ) are located at the dimerization interface where fumarate binds , site 3 ( yellow ) is situated at the tetramerization interface where ATP binds to stabilize the functional tetrameric form of the enzyme [47] , site 4 ( cyan ) is close to the tetramer’s core , and site 5 ( magenta ) , which is located in the protein region with negligible free energy change is used as a negative control ( Fig 9 ) . We show that a range of allosteric responses can be induced upon simulated binding to sites 1–4 for different modulation of the catalytic activity . Perturbation of site 1 induces a stronger decrease of the free energy in the catalytic site ( ΔgNAD ( 4×site1 ) = -0 . 80 kcal/mol ) than that of the native FUM allosteric site ( -0 . 18 kcal/mol ) . Similar to the effect of the FUM–site binding , perturbation of site 2 ( ΔgNAD ( 4×site2 ) = -0 . 17 kcal/mol ) and the site 3 ( ΔgNAD ( 4×site3 ) = -0 . 12 kcal/mol ) causes a slight decrease in the free energy in the catalytic site . On the other hand , simulated binding to site 4 strongly increases the free energy in the functional site ( ΔgNAD ( 4×site4 ) = 0 . 86 kcal/mol ) . The reverse perturbation analysis has also detected areas that are not allosterically coupled to the catalytic site , indeed simulated binding in site 5 ( chosen as a negative control ) induces only weak response in the catalytic site ( ΔgNAD ( 4×site5 ) = 0 . 05 kcal/mol ) . The generic nature of inducing allosteric response on the basis of the reverse perturbation approach is further illustrated with examples of allosteric signalling of different modes and magnitudes obtained in the cases of FBPase 1 ( S5 Fig ) and ThrS ( S6 Fig ) . Above analysis shows that in addition to the quest for latent allosteric sites , a more general question about the inducing of desired allosteric response can be formulated . Using site 1 of NAD-dependent malic enzyme as a test case , we show that the response can be further fine-tuned to achieve the necessary modulation in the catalytic site by varying the site’s composition ( Fig 10 ) . Restricting perturbation of site 1 ( blue ) to a subset of residues ( Leu118 , Ala119 , Gln122 ) substantially weakens the allosteric response ( from -0 . 80 kcal/mol to -0 . 14 kcal/mol , Fig 10 ) . Replacing Leu118 with the closest neighbours Gly117 and Ser121 recovers stronger allosteric signal ( -0 . 67 kcal/mol and -1 . 06 kcal/mol , respectively , S7 Fig ) . Alternatively , considering the subset of residues Gln122 , His125 and Ile126 ( red ) which produces an allosteric effect , which is comparable ( -0 . 82 kcal/mol ) to that induced by perturbing the whole site 1 ( Fig 10 ) . Allosteric signalling can also be fine-tuned to be weaker by replacing Gln122 with Ala119 , Cys120 or Ser121 , resulting in the allosteric responses -0 . 21 kcal/mol , -0 . 21 kcal/mol and -0 . 29 kcal/mol , respectively , at the regulated functional site ( S7 Fig ) . Sites 1–5 are well separated from the distant NAD site with less than 2% proximity , which ensures the allosteric nature of signalling from these putatively designated sites . Inducing and fine-tuning the allosteric response should be further complemented by the rational design or selection of allosteric effectors . For example , minimal sets of residues as the basis for targeted allosteric response ( e . g . blue and red sets , Fig 10 ( left , bottom ) and S7 Fig ) can be incrementally changed residue-by-residue ( Fig 10 ) in order to achieve required strength of the allosteric signal . Alternatively , one can explore the repertoire of all possible binding sites , adjusting them for a given lead ligand ( combined red-blue sets , Fig 10 , right ) . Allostery is a universal phenomenon where ligand-binding at a regulatory site causes a change in the ligand-affinity and/or activity at the coupled functional site [43] . Serious attempts to utilize the advantages of targeting allosteric sites instead of the orthosteric ones have only started in recent years , and the concept of allosteric drugs has since formed an important part in drug discovery [1 , 2 , 5 , 6] . Prediction of allosteric sites that can remotely regulate the dynamics at the functional site of interest has been shown to be a challenging task [5 , 15 , 16] . In this paper we test the hypothesis of the reversibility of allosteric communication , according to which the perturbation at the functional site results in a signal that propagates towards allosterically active protein regions . We show that in most of the cases reverse perturbation at functional sites causes an increase of the free energy in protein regions that are dynamically coupled to them , which were subsequently used for the detection of allosteric sites . In general , the topology of the protein plays a non-trivial role in the propagation of the allosteric signal , especially when protein activity is regulated by more than one effectors . Using the protein set consisting of 13 classical allosteric proteins and an independent benchmark set of 41 proteins , we show that known allosteric sites can indeed be identified by the reverse perturbation method . Good predictive power of the method was obtained in all proteins of both classical and benchmark sets when the allosteric sites are spatially separated from the functional ones according to the operational definition of allosteric site , which , in turn , is based on the original Monod-Changeux-Jacob’s formulation [43] of allostery and relationship between the functional and the regulatory exosites . After showing that known allosteric sites can be detected via the reverse perturbation method , we addressed the question of the allosteric sites identification . Using the NAD-dependent malic enzyme as an example , we show that simulated binding to two putatively defined sites in the protein region obtained from the reverse perturbation method ( sites 2 and 3 ) results in the allosteric response at the functional site similar to that caused by the native effector fumarate . Moreover , the prediction of allosteric sites from the protein regions obtained via the reverse perturbation method may result in multiple solutions . A repertoire of overlapping and non-overlapping sites can induce comparable allosteric signals upon binding to these sites . This redundancy , suggests that the practical task on the detection of known and prediction of latent allosteric sites can be turned into the general problem of how to induce and fine-tune an allosteric response . The case studies of NAD-dependent malic enzyme , threonine synthase , and fructose-1 , 6-bisphosphatase were used here to show that a range of allosteric responses can be induced upon simulated binding in sites putatively defined in the regions with increased free energy obtained via the reverse perturbation approach . For example , while simulated binding in the fumarate and putatively designated sites 1–3 of NAD-dependent malic enzyme result in a free energy decrease at the catalytic site , perturbation in site 4 strongly increases the free energy in the catalytic site . Two additional examples , FBPase 1 and ThrS , further illustrate the possibility to adopt the reverse perturbation method for inducing the allosteric response . Further , a presence of the overlapping sites of activators and inhibitors , such as ADP and PEP in PFK , ATP and CTP in ATCase and many others , calls for the fine tuning of the induced allosteric signalling via the rational design of the interactions between ligand and binding site . We exemplify the fine-tuning of the allosteric effect by varying the composition of a designated binding site 1 of NAD-dependent malic enzyme , which results in different levels of the catalytic activity modulation . In general , knowing the allosteric response in the functional sites upon binding of the native allosteric ligand , it is possible to select new allosteric sites and/or ligands that cause the same effect as the natural ones and , therefore , can be considered as new regulatory exosites . Further , an exhaustive calculation of the allosteric effect caused by every residue would be instrumental for rationally defining the candidate/potential allosteric sites in the absence of preliminary experimental data . The latter can be obtained in mutation experiments that measure the allosteric effects of residue-by-residue substitutions on the protein activity [48] , providing thus a foundation for the experimental verification and further improvement of the computational model . Despite two major limitations of the current structure-based perturbation model , the lack of sequence information , and the coarse-grained description of proteins on the basis of structure-based Cα model , the reverse perturbation method should be regarded as a general strategy in finding and exploring allosteric sites . For example , a reverse perturbation-like strategy was used in a recent experimental work where hydrogen exchange mass spectroscopy method was used for characterizing the allosterically active regions of protein Hsp90 induced by the orthosteric binding [49] . It is important to note , that modularity of the structure-based perturbation model is instrumental for improving the accuracy of calculations , as it was done , for example , in Gehrig , S . et al . [50] , where instead of using normal modes the slowest principal components ( PCA ) calculated from the MD trajectories were used for the calculations of the allosteric potential and the corresponding free energy derived in our model . However , increase of the accuracy by implementing the MD simulations will come at a price of the calculation speed . An alternative way of the model improvement would be via including the sequence dependence in the energy function , which preserves , at the same time , the model’s efficiency in terms of the high speed of calculations . To conclude , the task of inducing and fine-tuning of allosteric response can be generalized and formulated in the following sequence of steps: ( i ) finding the potential regulatory exosites via reverse perturbation approach; ( ii ) optimizing the compositions/structures of the binding exosites that can induce a required allosteric signalling upon binding to them; ( iii ) selection of the appropriate ligands that interact with the chosen allosteric site with sufficient binding affinity; ( iv ) allocation of the regulatory exosites that provide required allosteric effect at the corresponding functional site in the case of pre-existing library of ligands . By exploring the possibility to detect known , finding latent , and designing new regulatory exosites and corresponding allosteric effectors , the reverse perturbation method introduced in this work provides a conceptual framework aimed at the optimization of the allosteric regulation of protein activity . We used here the set of 13 allosteric enzymes included in previous studies [31–33] , which we refer to as the “classical set” . An additional set of 41 allosteric proteins with 48 experimentally-determined allosteric sites is obtained from the benchmarking collection of allosteric proteins ASBench [34] . In collecting above additional set , we applied the following requirements: ( i ) structures lacking information on functional sites , on parts of the structures , proteins that change the oligomerization state and structures in which regulation involves protein-protein interactions were omitted together with other cases of missing annotation; ( ii ) based on the operational definition of the allosteric site and applying the “proximity threshold” ( see below ) of no more than 2% , we obtained the final list of 48 sites in 41 proteins ( S1 Table ) . Interacting residues were extracted from the structures , based on the distance cutoff of 4 . 5 Å between the heavy atoms of protein residues and those of the ligand . The quaternary structure assemblies were obtained from the PDBePISA [51 , 52] with all water molecules , ions and ligands removed . The apo form of protein complexes was used whenever available , except for CAP due to the large structural difference between its apo and DNA-bound forms . The structure-based statistical mechanical model of allostery used in this work consists of three parts , which are described in detail in a previous work [31] . First , Cα harmonic models are built for the ligand-free ( unperturbed ) and ligand-bound ( perturbed ) systems from a single crystal structure . The presence of a bound ligand , i . e . perturbation , is modeled via the harmonic restrain of all residue pairs in the binding site . For the unperturbed system , the harmonic potential for all pairs of Cα atoms i and j is given by E ( 0 ) ( r−r0 ) =∑pairsi , jki , j ( di , j−di , j0 ) 2 ( 1 ) where r is the 3N-dimensional vector of coordinates of Cα atoms and r0 is a vector of Cα atoms from the reference structure . The di , j is the distance between any pair of Cα atoms i and j , the corresponding distance in the reference structure is di , j0 . The distance-dependent force constant ki , j decays as ( 1/di , j0 ) 6 with a global cutoff of 25 Å [53] . The potential associated with the ligand-bound state ( B ) with n bound sites S = {s1 , s2 , … , sn} is given by E ( B ) ( r−r0 ) =E ( 0 ) ( r−r0 ) +α∑n∑pairsi , j∈snki , j ( di , j−di , j0 ) 2 ( 2 ) where α = 100 is a stiffening factor of the perturbed site . Protein’s configurational ensembles are characterized by the first 10 slowest normal modes eμ ( 0 ) and eμ ( B ) for the free and ligand-bound systems , respectively . Previous studies have shown that the conformational transitions in allosteric communication are well described by the first 10 low frequency normal modes [32 , 33] . Second , we define a microscopic allosteric potential associated with a residue i Ui ( σ ) =12∑μεμ , iσμ2 , ( 3 ) which measures the total elastic work acting on a residue i as a result of the change in displacements of all its neighbours caused by the normal mode μ . The generic configuration of residue i is obtained from the reference configuration ri0 by superimposing the vectors eμ , i such as ri ( σ ) =ri0+∑μσμeμ , i , where σ = ( σ1 , … , σμ , … ) is a vector of Gaussian amplitudes . Thus , the residue configuration ri ( σ ) is uniquely identified by the state vector ( σ1 , … , σμ , … ) . The parameters εμ , i measures the elastic stress on the residue i and its neighbours j as result of the motion associated with the mode eμ εμ , i=∑j:di , j0<dcc\|eμ , i−eμ , j\|2 , ( 4 ) where c = 1 kcal/mol/Å2 and a distance cutoff dc = 11 Å . The allosteric potential in Eq 3 is evaluated for both protein states , ligand free ( 0 ) and ligand bound ( B ) ones , respectively . Finally , in order to obtain a per-residue free energy , the allosteric potential is integrated over all possible configurations σ , resulting in the partition function zi=Πμ ( 2πkBTεμ , i ) 1/2 , and , consecutively in the free energy gi = −kBT ln zi . The per-residue free energy difference between the unperturbed ( 0 ) and perturbed ( B ) protein states is Δgi=12kBT∑μlnεμ , i ( B ) εμ , i ( 0 ) . ( 5 ) To compare the relative strength of the free energy change Δgi for one residue to the effects on the corresponding subunit , the following average is considered ΔgU=1nU∑i∈UΔgi , ( 6 ) where nU is the total number of residues in the subunit U . In the reverse perturbation approach , the functional sites are perturbed , and , as a result , the change in the free energy at the levels of residues and sites is evaluated throughout the protein . To analyse the Δgi values of every residue in the oligomeric enzyme and to obtain the corresponding Δgi profile , the Δgi of corresponding residues from different subunits are averaged . Both functional and allosteric sites are indicated on the Δgi profile ( Figs 2–6 and S1–S4 Figs ) . The change of the free energy in the site s is also estimated as the average of the free energy changes among the residues in this site ΔgS=1ns∑i∈sΔgi , ( 7 ) where ns is the total number of residues in the site s . Eq 7 is used to obtain the free energy changes for every functional and allosteric site . The operational definition of the allosteric site is based on the restriction on a spatial proximity of communicating functional and regulatory sites . For every pair of residues i and j , the number of physically interacting pairs on the basis of the distance cutoff dc = 11 Å is obtained . The proximity is defined as the fraction of interacting pairs over the total number of pairs between the residues of functional and candidate allosteric sites under consideration: PX=ndi , j<dcni , j×100% ( 8 ) A ligand-binding site is defined as allosteric for the corresponding functional site within the same subunit if the proximity PX is no more than 2% . A distribution of the free energy changes Δgi is obtained upon perturbation of the catalytic sites for each protein in the classical and benchmark sets . For most of the allosteric proteins , all residues in the known allosteric sites exhibit an increase in the free energy , hence only the positive range of the Δgi distribution is used . In the PTP1B and BGDH , the entire Δgi distributions are used as the known allosteric sites contain equal proportion of residues with gain or loss of the free energy . For plotting the ROC , the first bin corresponding to the residues within the top 5% of the Δgi distribution is first used . A sequence of bins with decreasing thresholds with a 5% step is defined to obtain a series of true and false positive rates of the ROC curve . For residues with Δgi above the threshold , a true positive is scored if the residue in the crystal structure is located within 4 . 5 Å from the allosteric effector , whereas a false positive indicates that the residue does not belong to a known allosteric site . The harmonic models of proteins and the normal modes analysis are performed using the MMTK package [54] . Fourier approximation is used in the calculation of normal modes [55] . UCSF Chimera [56] is used to generate the illustrations .	Recent advances in the development of allosteric drugs allow one to fully appreciate the sheer power of allosteric effectors in the avoiding toxicity , receptor desensitization and modulatory rather than on/off mode of action , compared to the traditional orthosteric compounds . The detection of allosteric sites is one of the major challenges in the quest for allosteric drugs . This work proposes a “reverse perturbation” approach for identifying allosteric sites as a result of a perturbation applied to the functional ones . We show that according to the traditional Monod-Changeux-Jacob’s definition of allostery , considering non-overlapping regulatory and functional sites is a critical prerequisite for the successful detection of allosteric sites . Using the reverse perturbation method , it is possible to determine wide protein regions with a potential to induce an allosteric response and to adjust its strength . Further studies on inducing and fine-tuning of allosteric signalling seem to be of a great importance for efficient design of non-orthosteric ligands in the development of novel drugs .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "allosteric", "regulation", "cell", "physiology", "medicine", "and", "health", "sciences", "crystal", "structure", "engineering", "and", "technology", "condensed", "matter", "physics", "enzymology", "protein", "structure", "pharmacology", "crystallography", "thermodynamics", "enzyme", "chemistry", "solid", "state", "physics", "proteins", "cell", "communication", "enzyme", "regulation", "molecular", "biology", "communication", "equipment", "drug", "discovery", "free", "energy", "physics", "biochemistry", "drug", "research", "and", "development", "cell", "biology", "equipment", "biology", "and", "life", "sciences", "antennas", "physical", "sciences", "macromolecular", "structure", "analysis" ]	2018	Reversing allosteric communication: From detecting allosteric sites to inducing and tuning targeted allosteric response
During plant-pathogen interactions , the plant may mount several types of defense responses to either block the pathogen completely or ameliorate the amount of disease . Such responses include release of reactive oxygen species ( ROS ) to attack the pathogen , as well as formation of cell wall appositions ( CWAs ) to physically block pathogen penetration . A successful pathogen will likely have its own ROS detoxification mechanisms to cope with this inhospitable environment . Here , we report one such candidate mechanism in the rice blast fungus , Magnaporthe oryzae , governed by a gene we refer to as MoHYR1 . This gene ( MGG_07460 ) encodes a glutathione peroxidase ( GSHPx ) domain , and its homologue in yeast was reported to specifically detoxify phospholipid peroxides . To characterize this gene in M . oryzae , we generated a deletion mutantΔhyr1 which showed growth inhibition with increased amounts of hydrogen peroxide ( H2O2 ) . Moreover , we observed that the fungal mutants had a decreased ability to tolerate ROS generated by a susceptible plant , including ROS found associated with CWAs . Ultimately , this resulted in significantly smaller lesion sizes on both barley and rice . In order to determine how this gene interacts with other ( ROS ) scavenging-related genes in M . oryzae , we compared expression levels of ten genes in mutant versus wild type with and without H2O2 . Our results indicated that the HYR1 gene was important for allowing the fungus to tolerate H2O2 in vitro and in planta and that this ability was directly related to fungal virulence . Molecular oxygen , itself relatively nontoxic , is important to most living organisms on this planet . However , its derivatives , reactive oxygen species ( ROS ) , can lead to oxidative destruction of cells [1] . For example , in mammals , ROS can accelerate aging by making holes in membranes , or by stealing electrons from DNA , which may result in cancer and other severe diseases [2] . However , animals , plants and fungi have all adapted to use ROS as key signaling molecules [3] . In plants , ROS play a more positive role as a defense mechanism against attacking pathogens , and are often produced as a first line of defense [4] . In the plant-pathogenic fungus , Magnaporthe oryzae , ROS regulation plays important roles in both development and virulence . ROS itself has been shown to accumulate in the developing and mature appressorium , or fungal penetration structure , while the two NADPH oxidases in M . oryzae , NOX1 and NOX2 are required for proper development of appressoria , as well as full virulence [5] . The catalase gene family member , encoded by CATB , was shown to also be involved in cell wall integrity as well as virulence , as deletion mutants were altered in hyphal , spore and appressorial morphology [6] . Organisms , therefore , must carefully balance the toxic effects of ROS and the need for ROS in cellular signaling . There are five major types of ROS in plants: superoxide ( O2− ) , hydrogen peroxide ( H2O2 ) , hydroxyl radical ( OH ) , nitric oxide ( NO ) , and singlet oxygen ( 1O2 ) . In plant cells , organelles with an intense rate of electron flow or high oxidizing metabolic activity are major sources of ROS generation [7] . These organelles include mitochondria , chloroplasts and peroxisomes . ROS are also generated via enzymatic sources , such as membrane-associated NADPH oxidases , cell wall peroxidases and oxalate oxidases [8] . ROS play a crucial role during plant defense responses . Oxidative bursts have been detected when plant cells are inoculated with biotrophic pathogens [9] , hemi-biotrophic pathogens [10] , necrotrophic pathogens [11] , and pathogen elicitors [12] . More recent studies with M . oryzae that causes rice blast disease , demonstrated that rice produces H2O2 shortly after inoculation with a virulent strain of the fungus [13] , [14] . The toxic effects of ROS can directly kill pathogens , and as a result , pathogens have developed counter measures [5] . The coexistence of hosts and pathogens side-by-side determines that the increase of resistance in a host will be balanced by the change of virulence in a pathogen , and vice versa . A metabolite fingerprint study of three rice cultivars infected by M . oryzae provided evidence for suppression of plant-associated ROS generation during compatible interactions [9] . Fungal-produced catalase was secreted during infection , and appeared to play a role in breaking down the plant-produced H2O2 , allowing the disease cycle to occur; in the absence of catalase , infection was largely blocked by the plant's ROS [15] . ROS production and mitigation is a multifaceted process , incorporating many genes and pathways [1] . One mechanism of sensing and ultimate detoxification of ROS in yeast is via the Hyr1 gene , formerly termed Gpx3/Orp1; this gene , upon ROS induction , activates its partner protein yAP1 , which is a bZip transcription factor involved in activating cellular thiol-redox pathways , and arguably one of the most studied ROS-sensing proteins in yeast [16] . This AP1-like ( activator protein ) transcription factor regulates H2O2 homeostasis in Saccharomyces cerevisiae ( S . cerevisiae ) , which in turn governs the synthesis of glutathione [17] . Hyr1p plays a key role during the oxidative response in S . cerevisiae [18]; after being directly oxidized by H2O2 , it forms an intermolecular disulfide bond with yAP1 [19] . A conserved cysteine residue at position 598 in Yap1p becomes active by forming an inter-molecular disulfide bond with the Cys36 of Hyr1p . This transient inter-molecular linkage is then resolved to a Yap1p intra-molecular disulfide bond between the cysteines at positions C303-S-S-C598 . During this process , the Yap1 protein is released by Hyr1p in its active form , which is then transported to the nucleus [20] . This conformational change shields its nuclear export signal from the interacting protein Crm1p , allowing it to remain in the nucleus and control a suite of antioxidant genes [21] , [22] . Although YAP1 gene homologs have been analyzed in several plant pathogenic fungi such as Aspergillus fumigatus , Alternaria alternata , Cocholiobolus heterostrophus , Botrytis cinerea and Ustilago maydis [16] , [20] , [23] , [24] , [25] , [26] , HYR1 has yet to be studied in filamentous fungi . In this study , we closely examined the HYR1 homolog in M . oryzae as a candidate mechanism for coping with a ROS-intensive host environment . We demonstrated that HYR1 was indeed involved in detoxifying or preventing plant basal immune responses including plant-generated ROS and callose deposits during initial stages of infection , which was correlated with its role as a virulence factor . As one of the key members during the oxidative stress response , the yeast Saccharomyces cerevisiae Hyr1/YIR037W ( formerly termed Gpx3 ) was reported to be a glutathione-dependent phospholipid peroxidase ( PhGpx ) that specifically detoxifies phospholipid peroxides [19] . In order to identify the corresponding gene in M . oryzae , we performed a BlastP analysis against the fully sequenced genomic database of M . oryzae housed at the Broad Institute . Using an E-value of 1e-3 returned a single hit located on Supercontig 20 , with an accession number of MGG_07460 . 6 . It is 1315 bp long including two introns , with an open reading frame of 783 bp , which encodes a 172-amino acid protein . A sequence analysis was performed using Prosite on the ExPASy Proteomics Server ( http://ca . expasy . org/prosite/ ) . Hits revealed a glutathione peroxidase active site at amino acid positions 27–42 , and a glutathione peroxidase signature at amino acid positions 66–73 ( Figure 1A ) . When a BlastP search was performed against GenBank at NCBI , numerous hits were returned with high similarity scores , from many organisms including fungi and bacteria . An alignment indicates that the putative GSHPx domains of Hyr1 are highly conserved across different organisms ( Figure 1B ) . The MoHyr1 protein shares the highest amino acid conservation with the model , non-pathogenic fungus , Neurospora crassa ( 93% similarity and 73% identity ) , but shares between 81 and 90% similarity with eight other plant pathogenic filamentous fungi examined ( Table S1 and Figure 1C ) . Secondary structure of the HYR1 protein was determined by PSIPRED [27] , and consists of eight β-sheets ( or strands ) and four α-helices ( Figure 2 ) . As described in Zhang et al [18] , the ScHyr1p showed a typical ‘thioredoxin fold’ , also consisting of four β-sheets surrounded by three α-helices [28] . We compared the crystal structure of ScHyr1p with the predicted tertiary structure of MoHyr1 protein , generated with PyMOL ( http://www . pymol . org/ ) . The MoHyr1 predicted structure appears similar to a canonical thioredoxin fold , showing four β-sheets , with β1 and β2 running parallel and β3 and β4 running anti-parallel , surrounded by three α-helices ( Figure 2 ) . We located three positionally conserved cysteines in our HYR1 protein model compared to yeast , and these are marked in Figures 1B and 2 . Two important cysteines , Cys39 and Cys88 , likely correspond with two active sites found in the yeast Hyr1p , Cys36 and Cys82 . Together , our in silico data suggest that we have identified the structural homolog of the ScHyr1 from yeast , and that this gene is highly conserved across filamentous fungi . In order to functionally characterize the MoHYR1 gene , we obtained the ATCC S . cerevisiae Δhyr1 mutant and its wild type parent for complementation tests . Our hypothesis was that based on its sequence and predicted tertiary structure , the MoHYR1 gene would rescue the yeast mutant when grown on non-permissive concentrations of hydrogen peroxide . As shown in Figure 3 , the yeast mutant and the wild type strain both grow well on 0 and 2 mM H2O2 . However , growth of yeast Δhyr1 was significantly hindered in 4 mM H2O2 . The wild type MoHYR1 gene was transformed into the yeast mutant , which restored partial growth on this higher concentration . To further support our hypothesis , we constructed mutations in the two conserved cysteine residues at positions 39 and 88 . Neither of the mutations rescued the yeast phenotype on hydrogen peroxide ( Figure 3 ) . To explore the biological role of the MoHyr1 protein in the development and pathogenicity processes of M . oryzae , the deletion mutant Δhyr1 was generated through homologous recombination of the MoHYR1 open reading frame with a gene conferring hygromycin resistance ( hygromycin phosphotransferase; HPH ) ( Figure S1A ) . A gene deletion fragment was generated by nested PCR amplification of the 5′ flanking region of MoHYR1 , the HPH gene , and 3′ flanking region of MoHYR1 , using adapters to link the three pieces together . This gene deletion fragment , which contained flanking regions homologous to the MoHYR1 gene , was introduced into protoplasts of M . oryzae via PEG-mediated fungal transformation . After PCR screening of successful knockouts and ectopics using primer pairs outside the flanking regions and inside the HPH gene , two Δhyr1 knockout mutants ( B25 , B33 ) and two ectopic mutants ( B40 , B60 ) were identified ( Figure S1B ) and confirmed with Southerns ( Figure S1C ) . Real-time qRT-PCR was also employed to confirm full deletion of the MoHYR1 gene and no transcripts were detected . Deletion mutant Δhyr1 ( B33 ) was complemented with a full-length copy of the MoHYR1 gene linked to the cerulean fluorescent protein ( Figure S1D , see Materials and Methods ) . HYR1p in yeast was reported to not only be a sensor of ROS , but to have scavenging properties as well [19] . To investigate the role of MoHYR1 in scavenging H2O2 during vegetative hyphal growth , we inoculated the same amount of initial mycelia into complete media ( CM ) containing 0 , 5 and 10 mM H2O2 . No significant differences were detected among wild type , the Δhyr1 knockout mutants and the ectopics when growing in 0 mM H2O2 . However , the mycelial growth of the Δhyr1 knockout mutants was severely and significantly affected at 10 mM H2O2 ( Figure S2A and B ) . By contrast , the wild type and ectopics did not display much difference in mycelial growth at any concentration . The complemented mutant line grew slightly better than wild type in all concentrations of H2O2 , and upon Southern analysis , we found that four copies had inserted into the genome ( Figure S1E ) . Together , these data indicated that MoHYR1 was responsible for the H2O2 growth tolerance phenotype . To determine the role of MoHYR1 in virulence , we drop-inoculated detached leaves of three week-old blast-susceptible barley cultivars with conidia from two independently generated Δhyr1 mutants , B25 and B33 ( Figure 4A ) . The mutants were still able to cause disease lesions , but there was a measurable and significant reduction in lesion size compared to those produced by wild type , ectopics , and the complemented line ( Figure 4B ) . The complemented line , hyr1- C , restored full virulence to the Δhyr1 mutant , B33 . All pathogenicity assays were repeated on the susceptible rice cultivar Maratelli , with similar results ( Figure 4C ) using the spray-inoculation technique . Disease was also quantified on rice using a “lesion type” scoring assay [29] and error bars show that while lesion types 1–3 do not differ between the mutants , ectopics and wild type , lesion types 4 and 5 ( severe , coalescing ) did not form on mutant-inoculated plants ( Figure 4D ) Interestingly , no other developmental phenotype examined was compromised in the Δhyr1 mutant , including growth rate , conidia production and shape , germ tube and appressorial formation ( Table 1 ) . A fundamental question we wanted to assess was whether MoHYR1 was required for infection-related activities in planta . The M . oryzae's disease cycle is initiated when the conidium contacts a hydrophobic surface , inducing it to germinate . The germinated conidium forms a germ tube and appressorium that penetrates the plant surface via turgor pressure and forms a thin penetration peg into the first plant cell [30] . Thus , we first examined whether ROS was present during any of these processes , and if so whether MoHYR1 was involved in coping with it . We inoculated susceptible rice and barley cultivars with the Δhyr1 mutants , ectopics and wild type . ROS was detected using the indicator 2′ , 7′-dichlorodihydrofluorescein diacetate ( H2DCFDA ) [31] . Conidia of wild type , ectopics and the Δhyr1 mutant all elicited some degree of ROS when inoculated onto barley leaves ( Figure 5A–C ) , whereas ROS was undetectable under the same imaging conditions when non-inoculated leaves were stained ( data not shown ) . The Δhyr1 mutants showed the strongest ROS signal 24 hours post inoculation ( hpi ) compared to the others . The signal continued in this manner through 48 hours ( data not shown ) . These experiments were repeated six times and the results were consistent across the two independent Δhyr1 mutant lines . ROS signals were quantified via counting the number of ‘ROS haloes’ found around appressoria and expressing this as a percentage of appressoria counted per sample; a significant difference in signals was observed between the mutants , wild type , and ectopics ( Figure 5D ) . These results indicate that in the absence of the MoHYR1 gene , the fungus can no longer manage the ROS that is generated during initial infection events , or loses the ability to effectively cope with it . To better understand the reason for reduced virulence in the Δhyr1 mutant , we wished to determine whether internal fungal levels of ROS were altered in the absence of the gene . The deletion mutant and wild type were grown on complete media and stained with nitroblue tetrazolium ( NBT ) for production of superoxide anions ( Figure S3 ) . Results showed little differences between mutant and wild type when examining the entire colony ( Figure S3E and F ) or aerial hyphae ( Figure S3A–D ) . Figure 5C suggested that reactive oxygen species localized mainly around the appressoria . Upon closer inspection , we observed that the ROS “haloes” around the appressoria usually localized directly underneath the appressoria ( Figure 6 ) . Previous studies had demonstrated that the rice blast fungus also generates internal ROS during infection-related development , particularly during appressorial maturation and furthermore , that ROS can be secreted from the fungus itself [5] . In order to identify the source of the reactive oxygen species detected in our experiment , we inoculated M . oryzae onto the hydrophobic side of gel-bond , which can mimic the plant surface and induce ROS production in vitro [32] . The result shown in Figure 7 indicated that first , M . oryzae does generate ROS during germ tube and appressorial formation; second , the reactive oxygen species generated by M . oryzae were mostly intracellular and did not appear to be secreted or defused; and finally , that ROS were relatively weak in the fungal structures by 24 hpi . These observations occurred in the wild type , ectopic and mutant lines , indicating little difference in internal ROS levels regardless of the presence of HYR1 . Altogether , these results were different from what we observed in planta , which was a strong ROS signal from 24–48 hpi . In order to identify the source of the ROS detected during susceptible interactions , we used diamino-benzidine ( DAB ) to study the ROS distribution pattern . Barley leaves were inoculated with Δhyr1 mutant then stained with DAB and imaged using confocal reflected light signal to visualize the DAB deposits from a top view of an interaction site ( Figure 8A ) . The leaf samples from this same interaction site was processed further and embedded in epoxy resin to obtain a cross-section using a correlative microcopy approach . The confocal images suggested that the dark region ( DAB ) was localized immediately adjacent and inside the plant cell wall ( Figure 8B ) centered around the penetration peg ( arrowhead - Figure 8B ) . The second piece of evidence resulted from scavenging for ROS with ascorbic acid , an antioxidant that detoxifies hydrogen peroxide [33] . When 0 . 5 mM ascorbic acid was mixed with Δhyr1 mutant conidia , inoculated onto plants and stained with H2DCFDA , ROS haloes were clearly observed ( Figure 9A ) . However , when barley leaves were pre-treated with ascorbic acid , then inoculated and stained with H2DCFDA , almost no ROS haloes were detected ( Figure 9B ) . This experiment was repeated with another ROS-inhibitor called DPI ( diphehyleneiodonium chloride ) , with similar results ( data not shown ) . Ascorbic acid-treated leaves were also inoculated with mutant conidia and allowed to incubate in the growth chamber for six days , after which time we observed wild type lesions ( Figure 9C ) . This suggested that the ROS haloes observed in this experiment are likely originated from the plant . Futhermore , we analyzed previously characterized nox1 and nox2 mutants for ROS haloes; in M . oryzae , NOX1 and NOX2 code for NAPDH oxidases , and are largely responsible for producing internal ROS [5] . We hypothesized that if ROS was emanating from the plant , than the loss of the NOX genes should have no effect on haloes . Overall , haloes can still be produced when either of the nox mutants , or its parental strain , Guy11 was inoculated onto barley leaves ( Figure S4A–F ) . While there was a slight significant difference among the number of haloes observed when looking at the individual mutants ( nox1 made slightly more than nox2 ) , there was no significant difference between mutants and wild type ( 20–30 appressoria were counted per strain , and the percentage of those with haloes , reported; Figure S4G ) . Since our data strongly suggested that Δhyr1 mutants had a lower capacity to eradicate plant-generated ROS during early stages of infection . Our next goal was to determine whether this gene played a role in fungal tolerance to ROS generated immediately following inoculation . In order to carry out this experiment , we inoculated susceptible barley leaves with either the Δhyr1 mutants or the wild type conidia , and imaged them 1 hpi . The ROS dye H2DCFDA was injected directly into the leaves , so the result only showed the redox status inside the leaves , and not inside the fungus , which might have skewed the results . Our data revealed that ROS was detected 1 hpi , which indicated that the plant detected and responded to the pathogen at an early time point ( indicated by ROS fluorescence in the mesophyll cells; Figure S5A ) . A quantitative analysis of the signal intensities by ImageJ ( available at http://rsb . info . nih . gov/ij; developed by Wayne Rasband , National Institutes of Health , Bethesda , MD ) revealed no significant differences when inoculated with the Δhyr1 mutants or with the wild type conidia ( Figure S5B ) . We thus concluded that the MoHYR1 gene does not play a role in ameliorating an early , or immediate , plant defense response . To test whether MoHYR1 had any impact on plant-produced ROS that may occur later during infection , we inoculated Δhyr1 mutant conidia or wild type conidia onto barley leaves and stained with DAB at 24 hpi ( Figure 10 ) . Results indicated that the Δhyr1 mutant was unable to block ROS produced at 24 hpi , where the ROS was both detected in an entire plant epidermal cell , as well as in plant cells that were not in direct contact with the pathogen ( Figure 10 ) . It has been documented that the presence of reactive oxygen species around CWAs is a biochemical marker for non-penetrated cells during the interaction between barley and barley powdery mildew , Blumeria graminis [34] . To determine whether the ROS observed during a susceptible barley-M . oryzae was related to CWAs , we performed aniline blue staining on inoculated leaves . At 24 hpi , we found callose deposits specifically localized around the appressoria and penetration sites ( Figure 11 ) . Sequential correlative staining with H2DCFDA for ROS followed by analine blue for callose , showed a strong positional correlation between the two host responses when overlaid ( Figure 11C ) . CWAs are believed to physically block pathogen penetration [34] . To further characterize the CWAs formed during the barley- M . oryzae interaction , we examined leaves that had been inoculated with M . oryzae 24 and 40 hpi with either mutant or wild type conidia . The result showed that classical CWAs were formed within 24 hpi in both strains and no other differences in CWA morphology could be detected ( Figure 12 ) . Given the fact that increased ROS accumulation occurs in the absence of MoHYR1 , we next tried to determine whether the ROS scavenging system was impaired in the Δhyr1 mutants . We used real-time quantitative real time reverse transcription PCR ( real-time qRT-PCR ) to compare the expression of general antioxidant and redox control gene orthologs in both M . oryzae wild type and Δhyr1 strains ( Figure 13 ) . Primer pairs for the following genes were employed to examine gene expression: YAP1 ( MGG_12814 . 6 ) , GSH1 ( γ-glutamylcysteine synthetase; MGG_07317 . 6 ) , GSH2 ( glutathione synthetase; MGG_06454 . 6 ) , GLR1 ( glutathione reductase; MGG_12749 . 6 ) , GTT1 ( glutathione transferase 1; MGG_05677 . 6 ) , SOD1 ( Cu/Zn superoxide dismutase; MGG_03350 . 6 ) , CAT1 ( catalase 1; MGG_10061 . 6 ) , GTO1 ( omega class glutathione transferase 1; MGG_05367 . 6 ) , and cyt c per ( cytochrome c peroxidase; MGG_10368 . 6 ) . The housekeeping gene encoding Ubc ( ubiquitin conjugating enzyme; MGG_04081 . 6 ) was used as an internal control . We also included the gene MoHYR1 ( MGG_07460 . 6 ) in this experiment to confirm its deletion in the mutant lines . The expression patterns of these ten genes were placed into two categories . The first category ( Figure 13A ) is comprised of four genes that show increased expression in the wild type strain after induction with hydrogen peroxide , while expression in the mutant line is low and unchanging . GTT1 , GR and GSH1 belong to this category , along with the HYR1 partner protein YAP1; YAP1 also shows slight but significant differences in expression in the Δhyr1 mutant line with and without H2O2 , and has a higher expression level compared to the wild type strain without ROS . The second category contains genes whose expression does not significantly change , both in response to H2O2 , as well as in the presence of the MoHYR1 gene . This category includes six genes: cyt c per , CAT I , Cu/Zn SOD , GTT I , GSHII and MoHYR1 ( Figure 13B ) . HYR1 shows no expression at all in the mutant line , which was to be expected . We evaluated the sub-cellular localization pattern of the MoHYR1 protein during infection , conidia of a M . oryzae deletion line ( Δhyr1 B33 ) transformed with cerulean-MoHYR1 N-terminal fusion ( the same construct that was used for complementation ) , was inoculated onto barley leaves and observed during the following time points: 1 hpi , 6 hpi , 12 hpi , 24 hpi and 72hpi . At 1 hpi , MoHYR1 was mainly localized in the conidial vacuoles and with low levels in the cytoplasm . When the germ tube formed , the protein was present throughout the germ tube ( Figure 14A ) . At 6 hpi , the MoHYR1 protein showed increased cytoplasmic localization in the appressorium and conidium and at 12 hours , a concentration of HYR1 in the appressorial cytoplasm ( Figures 14B and C ) . At the later time point , 24 hpi , the protein appeared to be localized in the vacuoles with reduced levels in the cytoplasm ( Figure 14D ) , and a later , invasive stage time point suggests the protein was again cytoplasmically localized ( Figure 14E ) . In M . oryzae , MoHYR1 is the only sequence homolog of the yeast glutathione-dependent peroxidase , HYR1p , formerly termed Gpx3 [35] . In yeast , HYR1p senses H2O2 through two highly conserved cysteines that are redox sensitive . Mutations in either of these two cysteines leads to a non-functional HYR1 [18] . Indeed , we found that the wild type MoHYR1 , but not the MoHYR1 cysteine mutants , was able to partially rescue the yeast HYR1p mutant on non-permissive levels of H2O2 . This result is similar to Δyap1 yeast mutants complemented with homologs from two pathogenic filamentous fungi , Cochliobolus heterostrophus and Ustilago maydis , as both homologs partially complemented the yeast mutation [20] , [23] . These data suggested that MoHYR1 may function similarly during redox sensing and the subsequent signaling that leads to ROS detoxification . This model was further supported by the presence of ROS haloes located underneath appressoria during infection with a much greater frequency in the Δhyr1 mutant compared to the wild type strain . The increase in ROS haloes in Δhyr1 mutants correlated with significantly smaller lesions sizes when inoculated on susceptible rice and barley plants , suggesting that ROS scavenging regulated by MoHYR1 was required for full virulence . This was supported by a rescuing of the Δhyr1 mutant phenotype to wild type lesions by scavenging plant-derived ROS with ascorbic acid or disrupting plant-derived ROS generation with DPI . These results were similar to a gene recently reported on in the rice blast fungus called DES1 for Defense Suppressor 1 [14] . DES1 was also involved in virulence and triggers a stronger plant response upon infection , manifested by both an increase of the oxidative burst , as well as expression of two plant defense genes . Intriguingly , DES1 has no known functional domains and from sequence analysis , its function cannot be predicted , although it is well-conserved throughout fungi . It is also noteworthy that expression of MoHYR1 was tested in the Δdes1 mutant , and found to be slightly down-regulated . This could suggest that HYR1 and DES1 represent two semi-redundant , semi-dependent mechanisms evolved to cope with the plant defense response . Equally interesting is a gene recently identified in the plant and human fungal pathogens , Alternaria brassicicola and Aspergillus fumigatus , respectively , called tmpL [16] . This membrane-localized gene contains a FAD/NADP-binding domain and had not yet been studied in fungi . A deletion of tmpL resulted in a severely reduced virulence defect and hypersensitivity of exogenous oxidative stresses , however when the YAP1 gene was over-expressed in the deletion line , it rescued these and other mutant phenotypes , suggesting tmpL , YAP1 and presumably HYR1 may act in a concerted pathway to sense and trigger ROS scavenging pathways . A successful pathogen , which has the ability to detoxify ROS , will subsequently have fewer barriers to overcome before reaching its ultimate goal , which are the cell contents . Our results with the MoHYR1 gene suggest that while there might be no effect of MoHYR1 on ROS that's produced immediately by the plant ( Figure S3 ) , there is subsequent ROS production which MoHYR1 clearly helps the fungus overcome ( Figure 10 ) . Metabolic profiling performed by Talbot and colleagues ( 2008 ) provides support for this concept , revealing a M . oryzae-induced host metabolism re-programming that suppressed or delayed plant-produced ROS during susceptible interactions . Although supporting evidence has shown that M . oryzae can produce ROS during infection related development [5] , through scavenging experiments , the ROS observed in our studies appear to be largely plant-generated . Internal fungal ROS was unaffected by the absence of the MoHYR1 gene in vitro . Furthermore , ROS haloes were not disrupted by the ROS scavenger , ascorbic acid , when applied only to conidia , but were disrupted when ascorbic acid was specifically applied to leaves . Several pathways for plant-generated ROS include cell wall-bound peroxidases [1] . Plants defend themselves against pathogens by a battery of cell wall-associated defense reactions , including generation of ROS and cross-linking of lignin compounds [34] . During the interaction between a French bean ( Phaseolus vulgaris ) and a cell wall elicitor from Colletotrichum lindemuthianum , ROS appears to originate from cell wall peroxidases [36] . Apoplastic alkalization has been shown to be important in this process [34] . ROS generated from cell wall peroxidases also serve as key molecules required for lignification and cross-linking of cell walls [34] . In a study carried out between barley and the powdery mildew fungus , barley cell wall localized peroxidase HvRBOHA is responsible for generating H2O2 , which was only present in non-penetrated cells [37] . Our results , particularly in Figure 8B , suggest ROS localized up against the plant cell wall . Further investigations into M . oryzae-host interactions will include analyzing plant defense-related genes , including the barley cell wall peroxidase . Callose and ROS are two plant defensive compounds known to be involved in cell wall appositions , which are deposited during both compatible and incompatible interactions [34] . H2O2 played an important role in this process and enzymatic removal of H2O2 by catalase significantly reduces the frequency of phenolic deposition [34] . Several components were reported to be essential for this oxidative burst: peroxidases , a calcium influx and K+ Cl− efflux , extracellular alkalization , and post-Golgi vesicles [38] . ROS around the CWA areas might function as signal compounds to gather the vesicles and components needed for mature CWAs . We observed that ROS and callose deposits were positionally related during attempted penetration by both wild type and Δhyr1 mutants , immediately below the appressorium . From this result , we hypothesize that ROS generated by plant defenses activates CWA formation in a susceptible host and experiments to determine the timing of deposition of ROS versus callose are currently underway . A hypothesis that follows from these data is that when the MoHYR1 gene is deleted , the plant responds as though it's being challenged with an avirulent pathogen . As early as 12 hours post inoculation , we observed that barley leaves inoculated with Δhyr1 mutants showed higher ROS signals compared with leaves inoculated with wild type . These data were consistent using two staining methods , H2DCFDA and DAB . In leaves inoculated with wild type , ROS was detected around appressoria but was mostly observed inside fungal structures . However , ROS was seen both around appressoria and adjacent cells when inoculated with the Δhyr1 mutants . Whole cells filled with ROS were also observed when inoculated with Δhyr1 mutants , which was related with HR-type cell death . All these data indicated that HYR1 might function to suppress later plant-generated ROS , either by detoxifying it directly , or manipulating plant ROS secretion-related gene expression . While our data showed that HYR1 likely played an important role in ROS-detoxification processes , our experiments did not preclude other ROS tolerance mechanisms in the fungus , particularly since mutants were reduced in virulence , but not completely non-pathogenic . Such mechanisms might involve the aforementioned DES1 and tmpL genes . Currently , we are characterizing the MoYAP1 homolog in M . oryzae; our initial Δyap1 mutant data suggested this gene was dispensable for pathogenicity , much like what has been found in Botrytis cinerea , Aspergillus fumigatus and Cochliobolus heterostrophus [23] , [25] , [26] . Intriguingly , YAP1 did appear to be essential for virulence in Ustilago maydis and Alternaria alternata [20] , [25] , suggesting that fungal lifestyle ( i . e . necrotrophic vs . biotroph ) had little to do with this particular oxidative stress pathway , and further supporting redundant pathways . Our real-time qRT-PCR data showed that YAP1 increases in expression when wild type was challenged with H2O2 and we also noted a decrease in YAP1 gene expression in the Δhyr1 mutant background . One interpretation of this result was that the fungal cell might be compensating for the absence of HYR1 , by boosting expression of its partner gene . The glutathione pathway-related genes GLR1 , GTO1 and GSH1 , all increased during H2O2 challenge in the wild type however had extremely decreased expression in the mutant line , regardless of ROS . This suggested that these genes were reliant upon HYR1 , which was not unexpected , since the glutathione pathway was shown to be regulated YAP1 , which occurs after interacting with HYR1 [17] . Our results were also in keeping with the C . heterostropus Yap1 homolog mutant Δchap1 , which showed extremely low levels of both GLR1 and GSH1 [23] . Interestingly , we did not observe any of the other genes increasing in expression in the mutant background; this suggested that at least for the genes that we chose such as CAT1 and SOD1 , they did not provide compensatory mechanisms for a loss of HYR1 . While this is one hypothesis , it is also possible that these genes are regulated at the protein level , as was found in the A . fumigatus mutant , ΔAfyap1; both CAT1 and SOD1 were among the proteins down-regulated in the mutant [39] , and this could also hold true for the Δhyr1 mutant . Likewise , catalase , SOD and peroxidase activities were measured in the A . alternata mutant ΔAaAp1 [25] . A transcriptomic study on the Δhyr1 deletion mutant would answer many of these questions; further , such a study would uncover redundant pathways of ROS detoxification masked by the presence of MoHYR1 . While numerous studies have examined localization of the Yap1p , we were unable to find any studies on the localization of HYR1 either in yeast or filamentous fungi . Our data revealed that the HYR1 protein mostly localized either to the cytosol or to vacuoles , during early stage infection events on barley ( germ tube , early appressorial formation , appressorial maturation and penetration ) . At one hpi , MoHYR1 was mainly moving through the germ tube , although it was difficult to definitively ascertain which organelle it might be associated with . At twelve hpi , the MoHYR1 protein shows cytoplasmic localization , mainly expressed in the cytosol of the appressorium . We suspect that by twenty-four hours , the fungus had penetrated and gained ingress to the first epidermal cell; indeed cell biology studies on events following initial penetration suggested that M . oryzae bulbous hyphae fill an entire rice leaf sheath cell and were in the process of moving onto the next one by twenty-seven hours post-inoculation [40] . Its vacuolar localization at this time-point could reflect that fact that it was no longer needed by the fungus , which had circumvented the plant's oxidative burst and at that point growing in the first epidermal cell . We examined a later time-point at 72 hpi and found the HYR1 gene to be once again cytoplasmically localized , perhaps indicating a requirement for this pathway at the invasive growth stage . In conclusion , we identified and characterized the MoHYR1 gene , a functional homolog of the yeast Hyr1 ( or Gpx3 ) gene . Although MoHYR1 does not cause dramatic effects in the disease phenotype , it nevertheless played an important role in virulence . This effect appeared to be related to the deletion mutant's inability to tolerate plant-generated ROS , or at least to do so in a timely and effective manner to cause wild type levels of disease . Together , our results help to define a mechanism that , while well-studied in yeast , has not yet been examined in filamentous fungi; furthermore , our studies pose additional questions to be answered regarding the role of the glutathione pathway in scavenging ROS in filamentous fungi , how this aids in pathogenicity and what other underlying redundant scavenging pathways exist . Rice-infecting M . oryzae , strain 70–15 ( Fungal Genetics Stock Center 8958 ) was used as the wild type strain throughout this project , and the strain from which mutants and transgenics were derived . All strains were maintained at 25°C under constant fluorescent light on complete medium ( CM 1 liter: 10 g sucrose , 6 g yeast extract , 6 g casamino acid , 1 ml trace element ) . Oatmeal agar medium ( OAM 1 liter: 50 g oatmeal and 15 g agar ) was used for sporulation . Conidia were harvested 10–12 days after plating . Yeast strains BY4741 ( wild type ) and BY4741 YIR037W ( Δhyr1 mutant ) were ordered from the American Type Culture Collection , grown out and maintained on YPD medium . Constructs for transformation were built using standard PCR reaction conditions and programs; briefly , pJS371 used overlapping primers to make an intron-free version of the MoHYR1 gene in pJS318 . Using the intron-free plasmid , overlapping primers were used to make Cys39Ala and Cys88Ala mutant versions of the coding sequence . These were cloned into pCRScript ( pJS372 & pJS373 , respectively ) . The yeast HYR1 gene ( ScHYR1 ) was then amplified from Sc46 and cloned into pRS423 , the His3 episomal plasmid , pJS374 . These plasmids then form the basis of the genes to be tested: MoHYR1 wild type , the 2 cysteine mutants of MoHYR1 and the ScHYR1 gene . These four genes are under the same promoter and terminator . Therefore ScHYR1 was engineered to have an NcoI site at the ATG and a BamHI site at the beginning of the terminator ( pJS375 ) . Since the Magnaporthe gene has a natural NcoI site at the ATG , the 3 genes of the MoHYR1 are cloned into pJS379 as NcoI/BamHI fragments ( pJS381 , pJS382 , pJS383 ) . For the complementation assays , five-microliter drops from serial dilutions from cultures with anOD600 of 0 . 5 were spotted on plates with and without 0 , 2 and 4 mM H2O2 and grown for 2 days at 30°C . This experiment was repeated 10 times . In total , the following plasmids were used in this part of the study: pSM387 ( = pRS423 ) HIS3 yeast episomal plasmid; pJS374 pSM387 + ScHYR1; pJS381 ScHYR1-Pro::MoHYR1::ScHYR1Term; pJS382 ScHYR1-Pro::MoHYR1_Cys36Ala::ScHYR1Term; pJS383 ScHYR1-Pro::MoHYR1_Cys82Ala::ScHYR1Term . Rice cultivar Maratelli ( a gift from the Dean Lab; Raleigh , NC ) and barley cultivar Lacey ( Johnny's Selected Seeds; Winslow , ME ) were used throughout this study , as both are susceptible to M . oryzae strain 70–15 . Rice was grown in a growth chamber at 80% humidity , and 12 h:12 h day:night cycles , at 28°C . Barley was grown in a growth chamber at 60% humidity , and 12 h:12 h day:night cycles , at 24°C ( day ) and 22°C ( night ) . The targeted gene deletion was accomplished using the homologous recombination method . We amplified 5′ and 3′ flanking regions of Hyr1 using primer pairs #1 and 2 ( Table S2 ) . Flanking regions were then linked via adaptor-mediated PCR to a 1 . 3 kb HPH coding sequence , providing resistance to the antibiotic hygromycin ( Alexis Biochemicals , San Diego , CA ) . The entire length of the deletion fragment was 3 . 7 kb . Fungal protoplasts of the wild type 70-15 were directly transformed with the nested PCR product ( primers used were forward primer of primer pair #1 and reverse primer of primer pair #2 ) . Protoplast generation and subsequent transformation were conducted by following established procedures [41] . To confirm the knockout mutant , the genomic DNA of candidate strains was extracted and amplified with primer pairs #3 , 4 and 5 ( Table S2 ) . Equal-sized pieces of mycelia were cut with #3 cork-borer tool ( 0 . 7 cm in diameter ) , and immersed in 10 ml of liquid CM at 25°C in darkness . Colonies were grown in CM containing H2O2 at concentrations of 0 mM , 5 mM and 10 mM . Colonies were removed from each well , vacuum filtered to dryness , and measured on a scale one week post-immersion . For point or drop inoculations , conidia were harvested from 12-day-old cultures grown on OMA in 20 µl of a 0 . 2% gelatin ( Acros organics , New Jersey ) suspension , for a final concentration of 1–5×105 conidia/ml . Point two percent gelatin was used as a non-inoculated control for pathogenicity assays . For drop inoculations , three week old leaves of Maratelli or Lacey were detached and laid flat in a humid chamber ( 90 mm Petri dish with moist filter paper ) . Twenty microliters of conidial suspensions , or gelatin alone , were dropped onto each leaf and kept in darkness overnight at ∼25°C . The next day , remaining water drops were wicked off and moved to a growth chamber under constant fluorescent light . For spray inoculations , conidial suspensions ( 10 ml; concentration as above ) in 0 . 2% gelatin were sprayed onto three week old Maratelli or Lacey seedlings . Inoculated plants were placed in a dew chamber at 25°C for 24 hours in the dark , and then transferred into the growth chamber with a photoperiod of 16 h:8 h light:dark cycles . Disease severity was assessed seven days after inoculation . Quantitative real time reverse transcription PCR ( real-time qRT-PCR ) was carried out using primer pairs for the following genes: YAP1 ( MGG_12814 . 6 ) , GSH1 ( MGG_07317 . 6 ) , GSH2 ( MGG_06454 . 6 ) , GLR1 ( MGG_12749 . 6 ) , GTT ( MGG_06747 . 6 ) , GTO1 ( MGG_05677 . 6 ) , GTT1 ( MGG_09138 . 6 ) , SOD1 ( MGG_03350 . 6 ) , CAT1 ( MGG_10061 . 6 ) and cytochrome c peroxidase ( MGG_10368 . 6 ) . The housekeeping gene encoding ubiquitin conjugating enzyme ( MGG_00604 . 6 ) was used as an internal control . We also included the gene MoHYR1 ( MGG_07460 . 6 ) to confirm its deletion in the mutant lines . Primer pairs are listed in Table S3 . Seventy-five nanograms of cDNA generated from mycelium grown as per the H2O2 experiments described above ( generated from the 0 mM and 5 mM H2O2 samples ) , was used as templates for each reaction . The mycelia were fragmented in a blender as per the protocol by Mosquera et al [42] , before being inoculated into liquid complete medium . After 2–3 days , the mycelia were blended again to ensure the largest amount of actively growing fungal tips . The H2O2 experiment was performed 24 hours after the 2nd blending , and RNA was extracted . PCR reaction conditions were as follows for a 25 µl reaction: 13 µl H2O , 10 µl 5 Prime SYBR Green Master Mix ( Fisher Scientific , Waltham , MA ) , 0 . 5 µl Forward Primer ( for a final concentration of 2 µM; Integrated DNA Technologies , Coralville , IA ) , 0 . 5 µl Reverse Primer ( for a final concentration of 2 µM ) and 1 µl template DNA . Conditions for real-time quantitative RT-PCR conditions were as follows: 95°C for 2 min; 95°C for 15 sec , 58°C for 15 sec , 68°C for 20 sec ( cycle 40 times ) ; 95°C for 15 sec; 60°C for 15 sec ( melting curve ) ; 60°C –95°C for 20 min; 95°C for 15 sec; lid temperature constant at 105°C . The 2−ΔΔCt method was used for generating the data . ΔΔCt is defined as ΔCt treatment - ΔCt calibrator . cDNA from the strain 70-15 in 0 mM H2O2 was used as the calibrator for comparison of gene expression in 5 mM H2O2 in both the Δhyr1 deletion lines as well as the wild type For both the ΔCt treatment and ΔCt calibrator , ΔCt is defined as Ct gene - Ct housekeeping-gene . For the calibrator , which is 0 µM H2O2 , this value would be 2−0 or 1 . These experiments were repeated twice with similar results . A HYR1 N-terminal cerulean fusion construct was generated by fusion PCR . Briefly , using M . oryzae genomic DNA as a template , a 1 kb promoter region of HYR1 was amplified with primers 6 and 7 ( Table S2 ) . Another set of primers , 8 and 9 , were used to amplify the 2 . 4 kb HYR1 open reading frame . Three resulting fragments , the 1 kb promoter fragment , the 1328 bp ORF ( including 709 bp of terminator sequence ) and 740 bp cerulean fluorescent protein coding sequence [43] , were mixed and subjected to a second fusion PCR with primers 7 and 8 . The resulting 3 . 1 kb PCR product was generated with BamHI and NotI restriction enzymes ( New England Biolabs , Beverly , MA ) and cloned into pBlueScript II SK+ . The construct was fully sequenced and found to be correct , hence was co-transformed into the M . oryzae Δhyr1 knockout mutant protoplasts to make Cerulean-HYR1 fusion transformants . Transformants with expected genetic integration events were identified by PCR using primers pairs 6 and 10 ( Table S2 ) . Properly transformed Δhyr1 mutants were also used as the complemented lines , in Figures 3 and 4 , designated as “hyr1-C” . Ten-fourteen day old rice and eight day old barley plants were used and collected 24 hours after being inoculated with 10–12 day old conidia ( methods as described above ) . All staining procedures were performed with both rice and barley , however barley was best-suited for microscopy , hence all micrographs shown in this study are of barley . For experiments with 29 , 79-dichlorofluorescin diacetate ( H2DCFDA ) ( Invitrogen , Carlsbad , CA ) , inoculated tissue were collected and incubated for 60 min at room temperature in 5–20 mM H2DCFDA dissolved in DMSO ( less than 0 . 005% final concentration ) , then washed with 0 . 1 mM KCl , 0 . 1 mM CaCl2 ( pH 6 . 0 ) and left for 60 min at 22°C before experimentation . Dye excitation was at 488 nm; emitted light was detected with a 500–550 band pass emission filter . DAB staining was carried out using the protocol developed by Thordal Christensen et al [44] . Briefly , leaves were cut at the base with a razor blade and placed in a 1 mg/mL solution of DAB for 8 h under darkness at room temperature . Leaves were decolorized by immersion in ethanol ( 96% ) for 4 h followed by 2 hours in PBS buffer before imaging . A third method of ROS detection was employed for examining ROS internal to , or secreted from , the fungus . Nitroblue tetrazolium ( Sigma-Aldrich , St . Louis ) was used at 4 mg/mL ( in deionized water ) and the staining performed for 5 min∼30 min at room temperature prior to observation . In order to eliminate the ROS generated by fungus , conidia of Δhyr1 ( B25 ) and wild type ( 70-15 ) were mixed with 0 . 5 mM ascorbic acid ( AsA ) and inoculated onto the leaf surface . Leaves were stained for ROS at 24 hpi . In order to eliminate ROS generated from the plant , leaves were first treated with 0 . 5 mM ascorbic acid for 1 hour . To remove excess AsA , leaves were then washed with 0 . 1 mM KCl , 0 . 1 mM CaCl2 ( pH 6 . 0 ) buffer three times for 5 minutes each . Finally , leaves were inoculated with conidia 1hpi and stained for ROS 24 hpi . Additionally , barley leaves were injected with 5 µM DPI ( diphenyleneiodonium; Sigma , St Louis ) , then washed and inoculated , as above . Calcofluor White M2R ( Fluorescent brightener 28 , F-6258 , Sigma , St Louis ) was used for detection of the fungal cell wall . We made 10 , 000-fold dilutions from a saturated Calcofluor White stock solution . For experiments involving conidia on gel-bond ( VWR , Arlington Heights , IL ) , Calcofluor White was applied 1 , 4 , 8 , 12 , and 24 hours post inoculation , incubated for 15 minutes , then gently rinsed off with 1X PBS buffer . For experiments involving inoculated plants , inoculated or non-inoculated ( control ) leaf tissue was collected and immersed in working solution for 15 minutes , then gently rinsed with 0 . 1 mM KCl , 0 . 1 mM CaCl2 ( pH 6 . 0 ) . For CWAs staining , we cleared inoculated or non-inoculated ( control ) leaves in ethanol:acetic acid ( 6:1 v/v ) overnight and washed them with water . Subsequently , cleared leaves were incubated in 0 . 05% aniline blue ( w/v ) in 0 . 067 M K2HP04 buffer at pH 9 . 2 overnight and rinsed gently in sterilized deionized water for microscopy . Inoculated barley leaves were stained using DAB and rinsed several times in PBS . Thereafter , samples were fixed in 2% paraformaldehyde ( Electron Microscopy Sciences , Hatfield , PA ) and 2% glutaraldehyde ( Electron Microscopy Sciences , Hatfield , PA ) in sodium cacodylate ( Electron Microscopy Sciences , Hatfield , PA ) buffer for 1 hour overnight . Samples were then rinsed three times , 15 min each , in sodium cacodylate and post-fixed with 2% OsO4 in sodium cacodylate for 3–5 hours on a rotator . Again , samples were then rinsed three times for 15 min each , with water on a rotator . Samples then underwent an ethanol dehydration series ( 25% , 50% , 80% ETOH; 20 min each ) on a rotator . Samples were primed with 1% gamma-glycidoxylpropyl trimethoxysilane in 80% ETOH overnight at room temperature and then washed three times for 15 min each in 100% ETOH on a rotator . Samples then underwent a series of infiltrations on a rotator as follows: 100% ETOH/n-BGE ( Electron Microscopy Sciences , Hatfield , PA ) ( 1:1 ) for 30 min , 100% n-BGE for 30 min , n-BGE/Quetol-651 ( Electron Microscopy Sciences , Hatfield , PA ) ( 1:3 ) for 1 hour , n-BGE/Quetol-651 ( 1:1 ) for 1 hour , n-BGE/Quetol-651 ( 3:1 ) for 1 hour , 100% Quetol-651 for 1 hour , 100% Quetol-651 for 1 hour , 100% Quetol-651 overnight and 100% Quetol-651 for 1 hour . Finally , samples were embedded and polymerized in an oven at 60°C for about 24 hours . BlastP analysis was done against the fully sequenced genomic database of M . oryzae housed at the Broad Institute , using an e-value of 1e-3 . ClustalW ( X2 ) was used to perform the full alignment and generate the phylogenetic tree . The final tree image was generated with Tree Viewer . The HYR1 protein secondary structure was predicted using the PSIPRED protein structure prediction server . The structural image of the HYR1 protein was created using the PyMOL molecular viewer . All student t-tests were performed using JMP8 ( SAS Institute Inc . 2007 . <Title> . Cary , NC: SAS Institute Inc . ) . Confocal images were taken with Zeiss LSM510 or Zeiss LSM5 DUO using a C-Apochromat 40X ( NA = 1 . 2 ) water immersion objective lens . H2DCFDA ester was excited at 488 nm and fluorescence was detected using a 505–550 nm band pass filter . Calcofluor white was excited at 405 nm and detected using 420–470 nm band pass filter . Cerulean was excited at 458 nm and detected using a 475 long pass filter . We also used transmitted light and reflected light for some confocal experiments .	Reactive oxygen species ( ROS ) are antimicrobial compounds and also serve as stimulators and products of plant defense reactions . ROS appear to be active in the critical zone where pathogens and plants come in contact . Therefore , understanding the source , the role , and the destination of ROS in each interacting partner will be crucial for understanding the pathogen-host molecular battle . In this study , we focused on one potential fungal mechanism for ameliorating effects of plant-produced ROS during the early stages of infection . Characterizing the MoHYR1 gene from the rice blast fungus Magnaporthe oryzae , suggested that MoHYR1 was involved in overcoming plant defense-generated ROS . The deletion of this gene caused a virulence defect in M . oryzae , which we believe was associated with the mutant's inability to detoxify plant-generated ROS . Together , our data suggested that HYR1 is a virulence factor in the rice blast pathogen , and its role in virulence was directly related to sensing and managing plant-generated ROS during early infection events . HYR1 is part of a ROS scavenging and sensing pathway that is well characterized in yeast , and our study is the first to examine this important gene in filamentous fungi .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "microbiology/plant-biotic", "interactions", "plant", "biology/plant-biotic", "interactions", "cell", "biology/cellular", "death", "and", "stress", "responses", "cell", "biology/microbial", "growth", "and", "development", "pathology/molecular", "pathology", "cell", "biology/plant", "cell", "biology", "molecular", "biology/translational", "regulation", "cell", "biology/gene", "expression" ]	2011	HYR1-Mediated Detoxification of Reactive Oxygen Species Is Required for Full Virulence in the Rice Blast Fungus
Calcium ( Ca ) sparks are elementary events of biological Ca signaling . A normal Ca spark has a brief duration in the range of 10 to 100 ms , but long-lasting sparks with durations of several hundred milliseconds to seconds are also widely observed . Experiments have shown that the transition from normal to long-lasting sparks can occur when ryanodine receptor ( RyR ) open probability is either increased or decreased . Here , we demonstrate theoretically and computationally that long-lasting sparks emerge as a collective dynamical behavior of the network of diffusively coupled Ca release units ( CRUs ) . We show that normal sparks occur when the CRU network is monostable and excitable , while long-lasting sparks occur when the network dynamics possesses multiple metastable attractors , each attractor corresponding to a different spatial firing pattern of sparks . We further highlight the mechanisms and conditions that produce long-lasting sparks , demonstrating the existence of an optimal range of RyR open probability favoring long-lasting sparks . We find that when CRU firings are sparse and sarcoplasmic reticulum ( SR ) Ca load is high , increasing RyR open probability promotes long-lasting sparks by potentiating Ca-induced Ca release ( CICR ) . In contrast , when CICR is already strong enough to produce frequent firings , decreasing RyR open probability counter-intuitively promotes long-lasting sparks by decreasing spark frequency . The decrease in spark frequency promotes intra-SR Ca diffusion from neighboring non-firing CRUs to the firing CRUs , which helps to maintain the local SR Ca concentration of the firing CRUs above a critical level to sustain firing . In this setting , decreasing RyR open probability further suppresses long-lasting sparks by weakening CICR . Since a long-lasting spark terminates via the Kramers’ escape process over a potential barrier , its duration exhibits an exponential distribution determined by the barrier height and noise strength , which is modulated differently by different ways of altering the Ca release flux strength . Calcium ( Ca ) is a ubiquitous signaling ion in biology , regulating both normal biological pathways as well as disease processes [1–3] . Besides biological signal transduction , Ca is required for muscle contraction and plays a key role in generating both normal and abnormal cardiac rhythms [4 , 5] . Intracellular Ca is stored in endoplasmic reticulum ( ER ) or sarcoplasmic reticulum ( SR ) membrane networks exhibiting complex structures within the cell . Ca is released from these internal stores via Ca release channels , called IP3 receptors ( IP3Rs ) or ryanodine receptors ( RyRs ) . IP3Rs or RyRs are clustered on the membrane of the ER/SR , forming discrete Ca release units ( CRUs ) . Opening of IP3Rs or RyRs is sensitized by cytoplasmic Ca , forming a positive feedback loop called Ca-induced Ca release ( CICR ) . CICR causes the IP3Rs or RyRs to open and close collectively in a cluster , resulting in random and discrete Ca release events , called Ca sparks [6] . A spark can be elicited by Ca entry from sarcolemmal Ca channels or occur spontaneously via CICR . Ca sparks have been considered as the dynamical elements which interact to generate sub-cellular and cellular dynamics for Ca signaling and muscle contraction , such as Ca waves and oscillations [7–12] and more complex nonlinear dynamics in the heart [5 , 13–18] . A Ca spark normally lasts for 10 to 100 ms , but long-lasting sparks with duration of several hundred milliseconds to seconds have been widely observed in different types of cells under various conditions [19–28] . Paradoxically , the transition from normal sparks to long-lasting sparks can be induced both by agents that increase RyR open probability [19–24] ( e . g . , Fig 1A ) and agents that decrease RyR open probability [25–27] ( e . g . , Fig 1B ) . Long-lasting sparks have also been observed near the nucleus in ventricular myocytes without altered RyR open probability [28] . Xiao et al [20] proposed a single channel mechanism for long-lasting sparks induced by FK506 or rapamycin by showing that FK506 resulted in prolonged subconductance open states of RyRs . However , it has also been shown that FK506 or rapamycin disrupt the coupled gating of RyRs , substantially shortening their open time [29] , contrary to prolongation of spark duration by FK506 . Moreover , reducing RyR open probability by tetracaine or Mg2+ does not increase the open time , but still can result in long-lasting sparks [25 , 27] . In a simulation study using a single CRU model with network SR ( NSR ) Ca concentration held constant , Sobie et al [30] showed that decreasing the cooperative effect of coupled gating ( equivalently increasing RyR open probability ) promoted long-lasting sparks , which was further verified theoretically by Hinch [31] . Hinch showed that the transition from normal sparks to long-lasting sparks occurs when the deterministic dynamical system governing CICR at the single CRU level changed from monostable to bistable . Using a similar single CRU model , Stern et al [32] showed that long-lasting sparks could be induced by increasing the intra-SR Ca diffusion rate or increasing RyR open probability . However , none of the single CRU studies have explained the experimental observations that reducing RyR open probability can also promote long-lasting sparks . Although a spark is a collective behavior of the IP3Rs/RyRs clustered in a CRU , the CRUs in a cell are not isolated from each other , but are coupled via intra-SR and cytosolic Ca diffusion . Therefore , the spark behavior of a CRU also depends on the behaviors of the neighboring CRUs , which can only be understood in the context of the CRU network of the cell , rather than a single CRU . In this study , we propose a theory for the transition from normal brief sparks to long-lasting sparks based on a coupled CRU network , which unifies the seemingly contradictory experimental observations described above . We show theoretically and in computer simulations that when the Ca release flux of a CRU is low , CICR cannot be maintained at the single CRU scale , resulting in normal short sparks . When the Ca release flux is high , CRUs fire frequently , and the firing competition between neighboring CRUs prevents the CRUs from sustaining the CICR state , also resulting in normal short sparks . However , when the Ca release flux is in an intermediate range , Ca release is strong enough to maintain CICR but low enough to have a low spark frequency . Consequently , the quiescent neighboring CRUs provide the additional source of Ca to prevent SR depletion to a level causing CICR termination , thereby resulting in long-lasting sparks . From a fundamental nonlinear dynamics perspective , our findings reveal that different sparse patterns of long-lasting sparks correspond to different dynamical attractors of the CRU network . Those attractors are stable in the deterministic limit , where the spark duration is infinite , and metastable in the presence of RyR channel stochasticity , where long-lasting sparks terminate via Kramers-like escape across the barrier between the metastable firing state and the non-firing state . As a result of this escape process , the spark duration exhibits an exponential distribution determined by the height of the barrier and the noise strength . We first carried out simulations using a ventricular myocyte model developed by Restrepo et al [14] with modifications ( see Methods ) to recapitulate the experimental observations of long-lasting sparks induced by ryanodine [21] or FK506 shown in Fig 1 [20] . Fig 2 shows a transition from short sparks to long-lasting sparks in the model when the RyR open probability was increased by prolonging the open time of the RyRs to simulate FK506 in Fig 1A . The SR Ca load ( cj in Fig 2A ) under control conditions was ~2 . 0 mM and depleted to ~1 . 0 mM during a spark , while the SR load after simulated FK506 was reduced to ~1 . 4 mM and depleted to ~0 . 7 mM during a spark . The spark frequency ( i . e . , spark probability ) became higher after RyR open probability was increased ( Fig 2B ) , agreeing with the experimental observations [21] . The left panel and right panel in Fig 2C show the spark duration distributions before and after FK506 , respectively . Under the control condition , the spark duration exhibited a bell-shaped ( close to Gaussian ) distribution . After FK506 , the spark duration exhibited an exponential distribution . These properties also agree well with the experimental observations [20 , 21] . Fig 3 shows a transition from short sparks to long-lasting sparks when the RyR open probability was decreased by increasing the closed time of the RyRs to simulate the effects of tetracaine in Fig 1B [27] . In this case , the SR Ca load ( cj in Fig 3A ) under control conditions was ~0 . 9 mM and depleted to ~0 . 5 mM during a spark , while the SR load after simulated tetracaine was increased to ~1 . 2 mM and depleted to ~0 . 6 mM during a spark . The spark frequency became lower after the RyR open probability was reduced ( Fig 3B ) , also agreeing with experimental observations [33 , 34] . The spark duration distribution exhibited a bell-shaped distribution under control and became exponential after tetracaine ( Fig 3C ) , which also agrees with the experimental observations [27] . We then systematically investigated the effects of altering the Ca release strength of the CRUs on spark dynamics ( Fig 4 ) . We used four ways to alter the Ca release strength: the RyR closed-to-open rate constant ( scaled by α , Fig 4A ) , the RyR open-to-closed rate constant ( scaled by β , Fig 4B ) , the single channel conductance of RyR ( scaled by γ , Fig 4C ) , and the number of RyRs in a CRU ( RyR cluster size N , Fig 4D ) . We plotted the SR Ca load and spark duration using box plots for different release strength ( Note: the x-axis is not a linear scale ) . Decreasing α , which decreases the RyR open probability by increasing the closed time of the RyRs ( simulating tetracaine [27 , 33] ) , first increased and then decreased the spark duration . SR Ca load increased monotonically as α decreased ( Fig 4A ) . Note that the spark duration was normal ( short ) at both low and high RyR open probability , while long-lasting sparks occurred at the intermediate range . Increasing β , which decreases the RyR open probability by shortening the open time of RyRs ( simulating flecainide [33] ) , had the same effect as decreasing α , but the spark duration was shorter for the same RyR open probability ( note: the steady-state RyR open probability is the same for the same α/β ratio , see Methods ) . We observed the same non-monotonic behavior of spark duration for altering the single channel conductance of RyR ( Fig 4C ) and the RyR cluster size ( Fig 4D ) . In these two cases , despite the large change in spark duration , the SR Ca load only changed slightly . In all four cases , the spark frequency decreased as the Ca release flux strength was reduced . The results in Fig 4 show that for some conditions , long-lasting sparks occur with an increase in RyR open probability , whereas for other conditions , they occur with a decrease . Indeed , the conditions before the RyR open probability was altered were different in Figs 2 and 3 where long-lasting sparks were induced by increasing and decreasing RyR open probability , respectively . Although the conditions were different as indicated by the baseline spark frequencies ( Fig 2B and Fig 3B ) , the spark duration distributions under two conditions were similar ( Figs 2C and 3C ) . When long-lasting sparks were induced by increasing the RyR open probability , the spark frequency increased ( Fig 2B ) , while when long-lasting sparks were induced by decreasing RyR open probability , the spark frequency decreased ( Fig 3B ) . These behaviors agree well with experimental results [21 , 26 , 27] , supporting the modeling prediction that long-lasting sparks can be induced by different alterations of RyR properties starting from different initial states . To elucidate the mechanism of long-lasting sparks , we developed a theory using coupled CRUs and carried out simulations to verify the theory . Theory—According to previous studies [5 , 31 , 32] , normal brief sparks occur when the corresponding deterministic limit of the CRU model is monostable , and long-lasting sparks occur when it is bistable . Here we use deterministic models of uncoupled single and coupled CRU systems to perform theoretical analyses by investigating their steady-state solutions . Consider a single CRU model with three compartments indicated in Fig 5A . The differential equations describing the Ca dynamics are: dcidt=Jrel−Jup=γNp ( cj−ci ) −μciδ+ci ( 1 ) dcjdt=−Jrel+JNj=−γNp ( cj−ci ) +DNj ( cN−cj ) ( 2 ) dpdt=αci2 ( 1−p ) −βp ( 3 ) where ci , cj and cN are Ca concentrations in cytosolic , junctional SR ( JSR ) and network SR ( NSR ) , and p is the open probability of a single RyR . N is the number of RyRs in a CRU , γ is the single channel conductance of the RyRs , μ is the SERCA pump strength , δ is its half-saturation value , and DNj is the Ca diffusion constant from NSR to JSR . The RyR model ( Eq 3 ) is described by a 2-state model with an open state and a closed state . The rate constant for closed-to-open is αci2 and the rate constant for open-to-closed is β . Instead of fixing cN to a constant as in previous models [30–32] , we assume that the total Ca of the CRU is constant , i . e . , c0 = cN + ci + cj . We further assume that RyRs reach their steady-state rapidly , which implies that the ci-nullcline is the solution of Eq 1 for dci/dt = 0 with ps=αci2αci2+β ( the thick black lines in Fig 5A ) , which yields: ci=0 ( 4 ) and cj=ci+μci ( δ+ci ) γNps . ( 5 ) The cj-nullcline in turn is the solution of Eq 2 when dcj/dt = 0 ( thin black line in Fig 5A ) , yielding: cj=γNpsci+DNj ( c0−ci ) 2DNj+γNps ( 6 ) The steady-state solutions ( fixed points of the system ) correspond to the intersections of those two nullclines . The system always has a steady-state solution according to Eqs 4 and 6: ci = 0 , ps = 0 , and cj = c0/2 . Additional steady-state solutions can be obtained using Eqs 5 and 6 . Subtracting Eq 6 from Eq 5 , one has Δcj=μci ( δ+ci ) γNps−DNj ( c0−3ci ) 2DNj+γNps . ( 7 ) According to Eq 7 , when Δcj > 0 holds for all ci>0 , no additional intersections of the two nullclines and thus no additional steady states exist , so that the system is monostable . When this condition fails , new steady-states occur , leading to bistability in which CICR is maintained in the deterministic limit . It can be easily shown from Eq 7 that the condition Δcj > 0 for all ci>0 tends to fail by increasing c0 , DNj , or the Ca release flux strength ( i . e . , γNps ) . As shown in Fig 5A , increasing RyR open probability moves the ci-nullcline downward but has a small effect on cj-nullcline ( from black to cyan in Fig 5A ) , promoting bistability . Thus increasing RyR open probability or Ca diffusion from NSR to JSR tends to result in bistability for long-lasting sparks , agreeing with the previous studies [30–32] . However , if the CRU under control conditions is monostable , i . e . , Δcj > 0 holds for all ci>0 then this will always hold for reduced γNps . Therefore , reducing RyR open probability suppresses bistability and thus long-lasting sparks , which cannot explain the experimental observations and simulation results that reducing RyR open probability also promotes long-lasting sparks . Next , we consider networks of coupled CRUs . First consider two coupled CRUs ( Fig 5B ) . To facilitate analytical treatment , we assume no cytosolic Ca diffusion between the two CRUs , which are then only coupled via Ca diffusion in the NSR . The equations describing the variables of the second CRU are the same as Eqs 1–3 and an additional equation is needed for the coupled system to describe the Ca concentration in the NSR , i . e . , dCNdt=DNN ( cN−CN ) +DNj ( Cj−CN ) +μCiδ+Ci ( 8 ) where we denote the variables of the second CRU with capital letters and DNN is the diffusion constant of Ca diffusion between the NSR of the two CRUs . The total Ca of the two coupled CRUs remains constant , satisfying: 2c0 = cN + ci + cj + CN + Ci + Cj . Since the two CRUs are identical , the coupled system can always have a uniform steady state solution which is the same steady-state solution as the one of an isolated single CRU . However , it is possible that spatially non-uniform solutions exist . Assuming that the second CRU does not fire , staying at its steady state Ci = 0 ( Cj = CN ) , then the ci-nullcline of the first CRU is unchanged , but the cj-nullcline of the first CRU becomes cj=γNpsci+DNj ( 2c0−ci ) /34DNj/3+γNps ( 9 ) As shown in Fig 5B , the cj-nullcline ( Eq 9 ) is elevated from that of a single CRU ( Eq 6 ) . This implies that when two monostable CRUs are coupled together , and only one of them fires , the system can become bistable , i . e . , the unfired CRU provides an additional source of Ca via SR diffusion so that the SR of the firing CRU is not depleted to the level that terminates CICR . This bistability can be enhanced when the number of diffusively-coupled CRUs in the network is further increased ( Fig 5B ) . Note that the number of CRUs needed for bistability still relies on RyR open probability or the Ca release flux strength as in the single CRU . For example , reducing RyR open probability elevates the ci-nullcline but has little effect on the cj-nullclines ( red curves in Fig 5B ) . One can infer from the preceding theoretical analysis that in a network of coupled CRUs , different non-uniform steady-state solutions corresponding to different spatial firing patterns of sparks can exist in addition to the homogeneous monostable steady-state solutions , so that the network dynamics exhibits multi-stability . The specific pattern that the system selects depends on initial conditions . To test this prediction , we performed simulations in a one-dimensional cable of coupled CRUs with the single CRU described by Eqs 1–3 . The parameters were kept the same as for the control ( black curves ) in Fig 5B . Fig 5C shows a spatially period-2 and period-5 solutions , while Fig 5D shows two different random spatial patterns in the cable . For the sake of theoretical treatment , we did not allow Ca diffusion in the cytosolic space in the theoretical analysis . If Ca diffusion is allowed in the cytosolic space , it becomes more difficult for the non-uniform steady-state solutions to form . This is because cytosolic Ca diffusion tends to synchronize spark firing , thereby favoring the spatially uniform state . For example , after we added Ca diffusion in the cytosolic space with a small diffusion constant , the spatially period-2 solution was no longer observed , but higher periods as well as random patterns with longer spatial scales could still exist ( Fig 5E ) . Importantly , contrary to cytosolic Ca diffusion , enhancing intra-SR Ca diffusion promotes multi-stability ( see results in Fig 6 below ) . Based on the analysis above , we can make the following theoretical predictions for Ca sparks in coupled CRU networks . Differing from its deterministic limit , in which the dynamically selected spark patterns is dictated by the initial conditions , the CRU network system ( in a real or model cell ) exhibits spontaneous sparks which occur randomly in space and time . Therefore , the selection of a spark pattern depends on the spark firing rate . When the spark probability is high , many CRUs fire , and there are not enough unfired CRUs to allow Ca to diffuse via the NSR from the unfired to the fired CRUs to maintain CICR ( corresponding to the black nullclines in Fig 5B ) . Thus the chance of forming a long-lasing spark is very low if not zero . As RyR open probability is reduced , the ci-nullcline moves upwards . However , since the spark probability also decreases , the cj-nullcline also moves upwards ( this is different from a single CRU in which the cj-nullcline hardly moves ) since the unfired CRUs provide additional Ca . As long as the cj-nullcline moves faster than the ci-nullcline does , the two can intersect to result in long-lasting sparks . Once a CRU fires as a long-lasting spark , the SR Ca in its neighboring CRUs is decreased . This suppresses the probability that neighboring CRUs will fire since the spark probability increases exponentially with SR Ca [30] , thereby further stabilizing the long-lasting sparks . Stabilization is even more pronounced for faster SR Ca diffusion . As the RyR open probability decreases further , CRUs fire even less frequently . However , as a response to this decrease , the cj-nullcline will no longer move upwards as much as the ci-nullcline since adding more distal unfired CRUs does not help to elevate the cj-nullcline . Eventually , the two nullclines lose their intersections required for long-lasting sparks , and only the brief sparks occur . This theoretical framework allows us to interpret the computer simulation results shown in Fig 4 . A reduction of the Ca release flux strength first promoted the transition from normal short sparks to long-lasting sparks but then further reduction of release flux caused a transition back to short sparks . Therefore , long-lasting sparks result from the balance between RyR open probability , which controls the spark frequency and hence the average ratio of firing and non-firing CRUs in the network , and intra-SR Ca diffusion that promotes long-lasting sparks in regions of the cell where this ratio is sufficiently low for a firing CRU to be surrounded by several non-firing CRUs . The multiplicity of firing patterns reflects different network configurations of firing and non-firing CRUs . Computer simulations—To further confirm the theoretical predictions that CRU coupling and spark rate are two of the key factors in promoting long-lasting sparks , we carried out additional simulations using the same detailed stochastic CRU model as in Figs 2–4 . First , we gradually coupled more and more CRUs together in a 3D geometry to demonstrate the coupling effect ( Fig 6 ) . When the parameters were set using the control conditions in Fig 3 , adding more CRUs only slightly increased spark duration ( Fig 6A ) , which all followed bell-shaped distributions . When the parameters were set using the conditions simulating tetracaine , spark duration was brief for small CRU networks , but increased dramatically with network size ( Fig 6B ) . The spark duration distributions changed from bell-shaped distributions to exponential distributions . When the intra-SR Ca diffusion constant was reduced , the spark duration became shorter ( Fig 6C ) . In other words , enhancing intra-SR Ca diffusion promotes long-lasting sparks . We then carried out simulations to demonstrate the effect of spark rate on Ca spark dynamics . In the case of tetracaine , the RyR open probability was reduced , and the spark probability was low . To increase the spark rate , we applied stimuli periodically in space ( bars in Fig 7A ) to fire sparks at a certain time point after the system reached equilibrium ( arrow in Fig 7A ) . When the stimuli were applied to all CRUs , the spark durations were brief as under normal control conditions . As the spatial distance between the stimulated sites increased , the averaged spark duration prolonged and saturated as the distance increased further . In another type of simulation , we used the control condition as in Fig 3 in which the spark rate was high and the spark duration was brief . To reduce the spark rate of the CRU network , we shut off the RyRs in a portion of the CRUs in the network , which caused long-lasting sparks to occur ( Fig 7B ) . Finally , we demonstrated the effect of changing the nullclines by reducing the RyR conductance . The simulations were the same as in Fig 7A , but we kept the distance between the stimulated sites at 8 CRUs . As shown theoretically , reducing γ makes the two nullclines move further apart and eventually lose their intersection required for bistability . As shown in Fig 7C , the spark duration first increased and then decreased , becoming brief when γ became small enough . Based on the dynamical analysis , when the deterministic limit has one stable solution ( Fig 5A ) , normal sparks occur . Random opening of RyRs can trigger CICR , causing most of the RyRs to open ( arrow 1 in Fig 8A ) , which then depletes the SR ( arrow 2 in Fig 8A ) . When the SR Ca is too low , CICR cannot be maintained either due to the release flux becoming too small or because of luminal Ca-dependent regulation of RyRs . The release stops and SR is refilled by the SERCA pump to the steady state ( arrow 3 in Fig 8A ) . This is a typical excitable transient process in which the duration of the spark is determined by the duration of the excitable transient corresponding to the time it takes to deplete the JSR Ca below the critical level to sustain CICR . This explains why the duration of normal sparks exhibits a bell-shaped distribution . For long-lasting sparks , the deterministic limit has two stable states ( Fig 5B ) . CICR cannot automatically shut off but rather is maintained at the high stable state . Termination of the release is a stochastic transition across a potential barrier ( arrow 3 in Fig 8B ) , i . e . , the classical Kramers’ escape process [35] and the spark duration is then the first-passage time across the barrier . To understand what determines the duration of the long-lasting sparks , we use the Langevin equation to describe the stochastic opening of the RyRs in a CRU as [36]: dpdt=ko ( 1−p ) −kcp+ko ( 1−p ) +kcpNξ ( t ) ( 10 ) where p is the open probability of RyR and N is the number of RyRs in a CRU . ξ ( t ) is a Gaussian white noise with < ξ ( t ) > = 0 and < ξ ( t ) ξ ( t' ) > = δ ( t−t' ) . ko and kc are the transition rate constants with ko being a function of Ca concentrations in the dyadic space ( cp ) and JSR ( cj ) and kc = β ( see Eqs 26 and 27 in Methods ) . Since cp changes quickly due to the small volume of the dyadic space , one can represent it by a function of p using a quasi-steady state approximation [14 , 37 , 38] . Eq 10 can then be rewritten as dpdt=f ( p , cj ) +ε ( p , cj ) ξ ( t ) ( 11 ) where f ( p , cj ) =ko ( 1−p ) −kcp=αh ( cj ) ( vpcs/τp+γNpcjvp/τp+γNp ) 2 ( 1−p ) −βp ( 12 ) and h ( cj ) is a Hill function describing luminal Ca-dependent regulation ( see Methods ) , vp is the dyadic space volume , τp is the diffusion time constant from the dyadic space to sub-membrane space , and cs is the Ca concentration in the sub-membrane space ( see Restrepo et al [14] ) . We further assume that the noise strength can be approximated using the steady-state values , i . e . , ε=2ps ( 1−ps ) Nτ ( 13 ) where ps is the steady-state open probability: ps=koko+kc=11+kc/ko ( 14 ) and τ is the relaxation time of the RyRs: τ=1ko+kc ( 15 ) With a constant noise strength approximation , the transition rate across the potential barrier is [39]: rk= ( 2π ) −1U" ( pa , cj ) \|U" ( pb , cj ) \|exp[−U ( pb , cj ) −U ( pa , cj ) ε] ( 16 ) where U ( p , cj ) = −∫f ( p , cj ) dp is the potential function , U ( pa , cj ) the potential valley , U ( pb , cj ) the potential barrier ( see schematic plot in Fig 8C ) , and U" ( p , cj ) =∂2U ( p , cj ) ∂p2 . The spark duration , T , is approximately the first-passage time of the system across the barrier , whose distribution can be obtained as follows . At time T , the total probability , q ( T ) , that the system still stays in the potential valley obeys dq ( T ) / dT = −rkq ( T ) , and the total probability that the system has crossed the barrier is then P ( T ) = 1 − q ( T ) . The escape rate at time T is ρ ( T ) = dP ( T ) / dT , which gives rise to: ρ ( T ) =rkexp ( −rkT ) ( 17 ) and the average is <T>=∫0∞Tρ ( T ) dT=1/rk ( 18 ) The standard deviation of exponential distribution is: σ = < T > . Therefore , the duration of long-lasting sparks exhibits an exponential distribution . Our simulation results and the experimental observations agree with this theoretical prediction . Based on the Kramers’ theory ( Eqs 16–18 ) , the spark duration and its variance are determined by the noise strength and the height of the potential barrier , which depend on the SR Ca content , the RyR open probability , the RyR cluster size , as well as the relaxation time of the RyRs . The theory can explain qualitatively the results shown in Fig 4 that reducing the Ca release flux by different ways can promote long-lasting sparks but results in different average spark durations for the same change of release flux strength . For example , simulations showed that reducing the single channel conductance of RyR ( γ ) yields longer averaged spark duration than reducing the number ( N ) of RyRs in a CRU ( Fig 8D , and compare the results in Fig 4C and4D ) , even though those two interventions reduce the Ca release flux by the same amount . This is because reducing N also increased the noise strength , which resulted in shorter spark durations for the same reduction in Ca release flux . Based on the theory above , if one changes N and γ while keeping γN = constant , there will be no change in the driving force according to Eq 12 and thus no change in the potential barrier . According to Eq 13 , the noise strength will be changed , inversely proportional to N . Therefore , using Eqs 16 and 18 , one obtains that the average spark duration increases exponentially with N , i . e . , <T>∝exp ( μN ) ( 19 ) Therefore , increasing the number of RyRs in a CRU while keeping the release flux unchanged , although reducing the noise strength , increases the spark duration and its variance . To confirm this theoretical prediction ( Eq 19 ) , we carried out simulations by changing N and γ while keeping γN = constant in the ventricular myocyte model , and the average spark durations from the simulations indeed increased with N over a 5-fold change in N ( Fig 8E ) . The standard deviation is slightly greater than <T> ( Fig 8F ) , which may be due to the fact that the spark distributions are not precisely exponential in our simulations . In Fig 4A and 4B , RyR open probability was reduced by either reducing α ( tetracaine ) or increasing β ( flecainide ) to cause long-lasting sparks . For the same α/β , the RyR open probability was the same , but the spark duration was shorter in the case of increasing β than the case of decreasing α ( one example shown in Fig 8G ) . If one varies α and β while maintaining the same α/β ratio , then both the driving force in Eq 12 ( and thus the potential U ) and the noise strength in Eq 13 scale with α . Using Eqs 16 and 18 , one has <T>∝1/α ( 20 ) Using the ventricular myocyte model , Fig 8H shows that changing α and β while keeping α/β = 1 produces a linear relation between <T> and 1/α , which agrees with the theoretical prediction . In these simulation results , the standard deviation is slightly greater than <T> ( Fig 8I ) . It has been hypothesized that agents ( e . g . , FK506 , rapamycin , and ryanodine ) that increase RyR open probability promote long-lasting sparks by inducing long subconductance open states of RyRs [20 , 21] . However , long-lasing sparks have also been observed after tetracaine , Mg2+ , or ruthenium red which reduce RyR open probability but do not increase RyR open time [25–27] . In our simulation , the average RyR open time ( = 1/β ) is ~ 1 ms , while the average spark duration can be several hundred milliseconds or seconds . Therefore , long-lasing sparks cannot be explained solely by the single channel properties of individual CRUs . The mechanisms of long-lasting sparks have been investigated both theoretically and computationally using single CRU models . Sobie et al [30] were the first to simulate long-lasting sparks using a single CRU model in which they showed that decreasing the coupled gating of RyRs ( which equivalently increased the RyR open probability ) promoted long-lasting sparks . Hinch [31] used theoretical analysis to show that normal brief sparks are stochastic events of a system whose deterministic limit is a monostable excitable system , while long-lasting sparks are stochastic events of a system whose deterministic limit is a bistable system . He also derived analytical formulism of spark duration distributions for both normal and long-lasting sparks . In a more recent study [32] , Stern et al revisited the mechanisms of Ca spark termination , in particular the effects of Ca diffusion between NSR and JSR , and showed that long-lasting sparks are metastable solutions of a CRU which is potentiated by increased RyR Ca sensitivity , increased RyR open probability , or increased Ca diffusion from NSR to JSR . All these previous studies and the theoretical analysis in the present study show that , when using a single CRU model , the transition from normal brief sparks to long-lasting sparks is promoted by increasing RyR open probability , which cannot explain the experimental observation that reducing RyR open probability also promotes long-lasting sparks . Although Ca sparks are firings of individual CRUs , they are observed experimentally not in isolated CRUs but in a coupled network of CRUs . In a CRU network , CRUs are coupled via Ca diffusion in the NSR and cytosol . Thus CRU firings are affected by and also affect the neighboring CRUs . In a simulation study using a one-dimensional chain of coupled CRUs by Gaur and Rudy [40] , long-lasting openings was observed with reduced Ca flux due to impaired luminal Ca sensor and buffering , agreeing with the observation that reducing RyR open probability induces long-lasting sparks , but the underlying mechanism was not investigated . In the present study , by focusing on networks of diffusively coupled CRUs , our analysis sheds new light on the mechanisms of long-lasting sparks , thereby helping to unify seemingly contradictory experimental observations . Specifically , in the deterministic limit of vanishing channel stochasticity , multiple stable solutions can co-exist in a coupled CRU network , exhibiting periodic and random spatial patterns of continuously firing CRUs surrounded by quiescent non-firing ones . In the real or model cells , sparks fire randomly , and the occurrence of long-lasting sparks is a random pattern selection process among the multiple solutions of the CRU network that become metastable in the presence of noise . When the spark probability is high , the system selects the uniform solution , resulting in brief sparks . When the spark probability becomes lower , the system selects non-uniform solutions , resulting in long-lasting sparks maintained by intra-SR diffusion of Ca from unfired neighboring CRUs . When the spark probability becomes very low , no multi-metastability occurs in the CRU network , again resulting in brief sparks only . Therefore , long-lasting sparks can be generally understood to result from two competing processes: spark frequency controlled by RyR open probability and the formation of multi-metastable firing patterns of the CRU network promoted by increased diffusive coupling . Due to this competition , long-lasting sparks occur in the intermediate range of RyR open probability , explaining why reducing or increasing RyR open probability can induce long-lasting sparks in different experiments . Since reducing RyR open probability or the Ca release flux may cause an increase in SR Ca load ( Fig 4 ) , the question arises as to whether the increase of SR Ca load is responsible for the transition from normal sparks to long-lasting sparks . Based on the theoretical analysis ( e . g . , the nullcline analysis in Fig 5 or Eq 7 ) , increasing the SR Ca load ( or increasing the total Ca ) will potentiate bistability by moving the cj-nullcline upwards and thus promote long-lasting sparks . However , reducing the Ca release flux will also weaken CICR and move the ci-nullcline upwards , making it difficult to sort out whether increased SR Ca load or weaker CICR plays a dominant role in sustaining long-lasting sparks using the single CRU analysis . At the transition from short sparks to long-lasting sparks obtained by decreasing the Ca release flux in the simulations ( cf . Fig 4 ) , the SR Ca load only increased slightly ( see , in particular , the cases shown in Fig 4C and 4D ) . However , in those simulations , the spark frequency decreased quickly , indicating that there were many unfired CRUs that could provide Ca to avoid the Ca in the JSR of the firing CRUs to be depleted below a critical value for persistent firings . In fact , Zima et al [27] showed that increasing SR load alone ( by 36% ) without blocking RyR open probability would not produce long-lasting sparks . If the increase in SR load alone were responsible for the induction of long-lasting sparks , all CRUs would fire long-lasting sparks . However , only a very few of the CRUs exhibited long-lasting sparks in our simulations as well as in experiments [27] . We conclude that in both our simulations and experiments , increase in SR load might contribute to the induction of long-lasting sparks , but that the reduction in spark frequency is likely to be the major contributor by allowing Ca to diffuse from unfired to fired CRUs so as to prevent Ca depletion below the critical level for termination of CICR . In real ventricular myocytes , the RyR clusters are heterogeneous [41] , which may also contribute to the induction of long-lasting sparks . Since the number of RyRs and their spatial distribution within a CRU may vary , different CRUs may exhibit different CICR properties . As a result , some CRUs may have a higher likelihood to fire and hence preferentially drain Ca from unfired CRUs , making those unfired CRUs even less likely to fire . The occurrence of long-lasting sparks may also be affected by the heterogeneity of the NSR network in which a given CRU may functionally link different numbers of CRUs in different regions . As shown by Zima et al [26] , CRUs with fewer linked neighbors tended to fire short sparks while CRUs with more neighbors tended to fire long-lasting sparks , indicating that the supply of Ca from the neighboring unfired CRUs is important for maintaining the long firings . Due to the CRU heterogeneity in the real system , both brief and long-lasting sparks can co-exist in the same cell , which may account for the modularity of spark duration distributions seen in experiments [27] . However , while although heterogeneity alone can explain why some CRUs preferentially exhibit long-lasting sparks , it cannot explain why reducing RyR open probability promotes long-lasting sparks . We need to point out that whether increasing RyR open probability or decreasing RyR open probability promotes long-lasting sparks depends on the conditions of the system , such as temperature , intracellular Na level , as well as the species . As shown in Figs 2 and 3 , when the cell had a high SR load and low spark frequency , increasing RyR open probability promoted long-lasting sparks , but when the cell had a low SR load and high spark frequency , reducing RyR open probability promoted long-lasting sparks . Since the RyR properties and intracellular Na level that affect Ca load [42] may vary with species and diseases , such that in some cases , decreasing RyR open probability may promote long-lasting sparks , whereas in others , increasing RyR open probability may be required . The mechanism of Ca spark termination has been debated for decades and different biological causes and mechanisms have been proposed [43–47] , including stochastic attrition [43] , cytosolic Ca-dependent inactivation [43] , allosteric coupling [30] , luminal Ca-dependent inactivation [48 , 49] , SR Ca depletion [30] , and induction delay [50] . Theoretically , any of these mechanisms or a combination of several of them can terminate a Ca spark , but which of them exists in the real cell has been the subject of debate for decades . Although the exact biological causes remain unclear , two qualitatively distinct dynamical mechanisms can be distinguished from the basic nonlinear dynamics perspective of the present article . As illustrated in Fig 8A , a normal spark is an excitable transient of a stochastically excitable system . Randomly opening one or more RyRs or LCCs results in an increase of Ca in the dyadic space , which can elicit CICR to cause RyRs to open . The CICR keeps the RyRs open which depletes the JSR . When the JSR Ca is depleted below a critical level , the CICR stops because: 1 ) the Ca in the JSR is so low so that the Ca flux through the RyRs cannot maintain the Ca level in the dyadic space to sustain the CICR; and 2 ) Reduced SR Ca causes inactivation of the RyRs so that the number of open RyRs decreases , further reducing the Ca flux required to sustain CICR . As CICR declines , all RyRs close and the spark terminates . However , when the JSR Ca cannot be depleted to the critical level , CICR is sustained ( Fig 8B ) . Termination of CICR is due to stochastic fluctuations of RyR openings that cause the system to cross the potential barrier ( Fig 8C ) , i . e . , due to stochastic fluctuations ( or stochastic attrition [32] ) , the number of open RyRs at a certain time is below a critical number so that the Ca flux is not strong enough to sustain CICR . Since termination of a normal spark is via JSR Ca depletion , its duration is determined by the Ca release flux strength . The stochastic channel noise causes the spark duration to fluctuate around its mean , which is not highly variable . In contrast , a long-lasting spark ( Fig 8B ) is a stochastic bistable switch whose dynamics follow Kramers’ escape theory . The spark duration in this case exhibits an exponential distribution with a large variation . As shown in our theoretical analysis ( Eqs 13–18 ) , the average spark duration strongly depends on the relaxation time ( τ ) , open probability of the RyRs , and the size of RyR clusters and thus on the specific ways of changing the Ca release flux strength . The theoretical predictions agree well with the behaviors observed in simulations of the detailed ventricular myocyte model . Therefore , the two types of sparks are caused by two distinct dynamical mechanisms , and the transition from brief to long-lasting sparks is a transition from monostability to bistability or multistability . Several limitations of the present study need to be pointed out . The CRU network is a homogenous model , while the real CRU network is highly heterogeneous [41] . We used a very simple model of the spatial structure of a CRU , while spatially detailed models of CRUs have been developed recently [50–53] , and a new mechanism of spark termination was shown in such detailed models [45 , 50] . Adding the detailed structural CRU information may reveal additional mechanistic insights into long-lasting sparks . Finally , the detailed spark dynamics may also depend on the specific RyR model used . We used a 2-state RyR model that incorporated the luminal Ca regulation of RyR , simplified from the original 4-state model . As we showed in Fig 9 , the two models exhibited similar spark dynamics despite small quantitative difference . Nevertheless , the dynamical mechanisms of the transition from brief to long-lasting sparks revealed in the present are likely generic ones that are applicable to Ca spark dynamics in the real systems . We used the ventricular myocyte model developed by Restrepo et al [14] , which contains a three-dimensional network of 19 , 305 ( 65x27x11 ) CRUs ( Fig 9A ) with CRU spacing being 1 . 84 μm in the longitudinal direction and 0 . 9 μm in the transverse direction , corresponding to a cell of dimension ~120×25×10 μm3 . The CRUs are coupled via Ca diffusion in the cytosolic space and SR . Each CRU contains five sub-volumes ( lower panel in Fig 9A ) : NSR , JSR , dyadic space or proximal space , sub-membrane space , and cytosolic space . Each CRU has a cluster of 100 RyR channels simulated using random Markov transitions . The Ca concentrations in the five compartments for an arbitrary CRU are described by the following differential equations: dcidt=βi ( ci ) ( Idsivsvi−Iup+Ileak−ITCi+Ici+Ib ) ( 21 ) dcsdt=βs ( cs ) ( Idpsvpvs+INCX−Idsi−ITCs ) ( 22 ) dcpdt=βp ( cp ) ( Irel+ICa , L−Idps ) ( 23 ) dcnsrdt= ( Iup−Ileak ) vivnsr−Itrvjsrvnsr+Icnsr ( 24 ) dcjsrdt=βjsr ( cjsr ) ( Itr−Irvpvjsr ) ( 25 ) where ci is the free Ca concentration in the cytosolic space , cs is the free Ca concentration in the sub-membrane space , cp is the free Ca concentration in the proximal space ( dyadic space ) , cjsr is the free Ca concentration in the junctional SR , and cnsr is the free Ca concentration in the network SR . The β terms account for instantaneous buffers in corresponding compartments using the rapid buffering approximation [54] . INCX is Na-Ca exchange flux and ICa , L is the L-type Ca flux ( Note: the same symbols as the membrane ionic currents were used but they are ion fluxes ) . Iup is the SERCA uptake current representing total flux into the NSR , Ileak is the leak current from NSR to cytosol , and Irel is the total Ca efflux from the JSR . Idsi , Idps and Itr are the diffusion currents from adjacent compartments , ITCi and ITCs are the troponin C dynamic buffering currents in cytosol and sub-membrane spaces , and Ici and Icnsr are the diffusive currents between neighboring CRUs in the corresponding compartments . vs , vi , vp , vnsr , and vjsr are the volumes of the sub-membrane space , cytosolic space , dyadic space , NSR , and JSR , respectively . We made the following changes ( and the small corrections stated in Restrepo and Karma [55] ) from the original model [14]: The changed parameters from the original model [14] are listed in Table 1 . The RyRs were simulated using a stochastic simulation method described previously [38] . The time step for integration of the differential equation is 0 . 01 ms . The total simulation time is at least 100 s and data were collected after 10 s to ensure that the system reached its steady-state condition . Simulations were done using Graphic Processor Units ( Nvidia Tesla K20 ) with CUDA and C language . The source code of the ventricular myocyte model is available at the following link , https://bitbucket . org/quslab/long-lasting-spark-model/src .	Calcium ( Ca ) sparks , resulting from Ca-induced Ca release , are elementary events of biological Ca signaling . Sparks are normally brief , but long-lasting sparks have been widely observed experimentally under various conditions . The underlying mechanisms of spark duration or termination and the corresponding determinants remain a topic of debate . In this study , we demonstrate theoretically and computationally that normal brief sparks are excitable transients , while long-lasting sparks are multiple metastable states emerging in the diffusively coupled Ca release unit network , as a result of cooperativity and release competition among the Ca release units . Termination of a long-lasting spark is a Kramers’ escape process over a potential barrier , and the spark duration is the first-passage time , exhibiting an exponential distribution .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[]	2016	Long-Lasting Sparks: Multi-Metastability and Release Competition in the Calcium Release Unit Network
This work examines the computational architecture used by the brain during the analysis of the spectral envelope of sounds , an important acoustic feature for defining auditory objects . Dynamic causal modelling and Bayesian model selection were used to evaluate a family of 16 network models explaining functional magnetic resonance imaging responses in the right temporal lobe during spectral envelope analysis . The models encode different hypotheses about the effective connectivity between Heschl's Gyrus ( HG ) , containing the primary auditory cortex , planum temporale ( PT ) , and superior temporal sulcus ( STS ) , and the modulation of that coupling during spectral envelope analysis . In particular , we aimed to determine whether information processing during spectral envelope analysis takes place in a serial or parallel fashion . The analysis provides strong support for a serial architecture with connections from HG to PT and from PT to STS and an increase of the HG to PT connection during spectral envelope analysis . The work supports a computational model of auditory object processing , based on the abstraction of spectro-temporal “templates” in the PT before further analysis of the abstracted form in anterior temporal lobe areas . The concept of an auditory object is controversial [1] . The term can be applied to a sound source like a voice , or an acoustic event generated by a source such as a vowel sound . In both cases , there are features of the object that are independent of the detailed structure of the sound: we can recognise the same vowel , or voice , regardless of the pitch . In these examples , the spectral envelope of the sound determines the particular vowel sound produced , and is , in general , one of the important acoustic features that determine its perceived timbre ( Figure 1; spectrogram of the same vowel at different pitch ) . In this experiment we consider the “abstraction” of the spectral envelope a critical aspect of auditory cognition that defines auditory objects before semantic processing . Such analysis allows generalisation between different exemplars ( e . g . , the same vowel at a different pitch ) in an analogous manner to the generalisation between visual objects that are seen from different perspectives . The analysis of spectrum in the central auditory system begins in the cochlear nucleus [2] , in which models specify sharpening of spectral representation by lateral inhibition [3] . Although relatively less is known about the representation of spectrum in the inferior colliculus and auditory thalamus , a general understanding is that the sharpening of spectrum representation continues in these centres [4] . At the level of the primary cortex , however , animal studies show that representation of spectrum is more complex . Specifically , a given spectrum at the cortex is represented at multiple scales in which a given spectrum has multiple representations at different levels of spectral resolution [5] . Mathematically , this representation has been called ripple analysis , where a given spectrum is decomposed into a sum of ripples of different ripple densities and velocities [6] . Neurons in the primary auditory cortex allow spectral analysis by selectively responding to a fixed ripple density and ripple velocity . The complex spectral analysis in the cortex [7] has not been demonstrated in subcortical areas to the same extent [8] . In this study , we examine the human cortical representation of the spectral envelope independently of the fine structure of the spectrum , a process for which there are a priori grounds for specifying cortical models that are based on initial complex representations in the primary auditory cortex . We have previously demonstrated bilateral activation of the planum temporale ( PT ) and right-lateralised activation of the superior temporal sulcus ( STS ) during the analysis of the spectral envelope in a conventional analysis of functional magnetic resonance imaging ( fMRI ) data [9] . Such analyses identify the regional nodes of a network that are active during the task without demonstrating the pattern of connections that determines the dynamics of the system: there are multiple mechanisms by which the measured task-induced regional responses could be explained . In the current study , we go beyond classical structure–function correlations and characterise formally the functional interactions between auditory areas involved in spectral envelope analysis . This system identification approach rests on the mathematical characterization of the causal and context-dependent influences that system elements exert upon each other ( i . e . , effective connectivity [10–13] ) . We use dynamic causal modelling ( DCM ) and Bayesian model selection [14] to address two fundamental questions about the biological computations that attend auditory processing . First , we assess the general structure of the HG–PT–STS network for auditory object processing . In particular , we address the critical question of whether analysis in PT and STS occurs in a serial ( hierarchical ) fashion , based on connections from HG to PT and from PT to STS , or whether the analysis is based on parallel processing that is mediated by connections from HG to both PT and STS . Second , we address how connection strengths between elements of this cortical network are modulated or enabled during the spectral envelope processing . The approach allows a direct test of a computational mechanism we have suggested previously [15] . This scheme is based on an initial stage of abstraction of the properties of the stimulus that occurs at the PT , before further processing of the abstracted “template” in areas that are concerned with categorical and semantic processing of auditory stimuli . The demonstration of a serial mechanism based on the PT as an intermediate stage would be consistent with such a scheme . In brief , our results provide strong support for a serial model with increase of the connection strength of the first stage from the HG to PT during spectral envelope analysis . The results suggest a single “stream” for auditory object analysis , and are congruent with macaque models based on a predominant pattern of connectivity from core to belt to parabelt areas . We assessed the ability of different network models to explain the variation in measured fMRI blood oxygenation level–dependent ( BOLD ) responses in the HG , PT , and STS in the right hemisphere during the extraction of the spectral envelope of generic sounds without any semantic association . Two broad classes of models , serial and parallel , were defined as shown in Figure 2 . The serial models contain connections from HG → PT and thence from PT → STS . In contrast , parallel models postulate connections from the HG to both the PT and STS . The models within each family differ with respect to the back connections specified , and with respect to the specific site of the modulatory effect of spectral envelope analysis . The models were compared using Bayesian model selection implemented within SPM ( http://www . fil . ion . ucl . ac . uk/spm/software/spm5 ) . The selection procedure estimates the probability of each model given the data using Akaike information criterion ( AIC ) and Bayesian information criterion ( BIC ) approximations to each model's log-evidence or marginal likelihood . Figure 3 shows the evidence for the models , determined separately using AIC and BIC , in eight participants . In this figure , we have assumed all models were equally likely a priori . This allows us to treat the normalised marginal likelihood as the conditional probability of each model . Model 1 is the optimal model over all participants , with the exception of participant 7 . The parameters for this model specify a serial model with connectivity ( HG → PT → STS ) and modulation of connection from HG → PT during the analysis of the spectral envelope . In addition to the individual inference , Table 1 shows the group Bayes factor ( GBF ) for model 1 with respect to the other 15 models . Given candidate hypotheses ( models ) i and j , a Bayes factor of 150 corresponds to a belief of 99% in the statement that “hypothesis i is true” . Following the usual conventions in Bayesian statistics [14 , 16] , this corresponds to “strong” evidence in favor of model i ( compare Table 2 ) . All the values of the GBF for model 1 with respect to all other models is greater than 150 , corresponding to very strong evidence for the serial model number 1 . Plots of measured and predicted BOLD time series for a single participant are shown in Figure S3 . This figure shows that the BOLD response in all three areas , particularly in STS , is fitted well by the optimal model . This demonstrates that ( 1 ) activity in the PT can be explained as a function of the input from the HG and its modulation during spectral envelope processing , and that ( 2 ) STS activity can be explained as a function of the input from the PT ( compare the structure of model 1 as shown in Figure 2 ) . Estimates of the interregional connection strengths and their modulation for each participant and probabilities that the coupling estimates are greater than zero are shown in Tables 3 and 4 , respectively . The probabilities that the connection strengths are greater than zero are all ~1 . 00 , with the exception of the PT → STS connection in participant 5 . Furthermore , the probability that the modulation of the strength of the connection from HG → PT is greater than zero is ~1 . 00 in all participants except 1 and 7 , where the probability is greater than 0 . 9 . A further t-test was carried out on intrinsic and modulatory connection strengths to assess the group level connection strengths . The mean values of HG → PT and PT → STS intrinsic connection strengths are 0 . 37 ( p < 0 . 01 ) and 0 . 48 ( p < 0 . 01 ) , respectively . The mean value of modulatory HG → PT ( measured in percentage increase ) in connection strength is 109 . 29 ( p < 0 . 01 ) Theoretically , a large number of models other than those considered above are possible . The choice of these models was motivated by preliminary analysis of the data . This analysis showed that ( 1 ) inclusion of modulation of the HG → PT pathway is critical to model performance ( as evaluated using AIC and BIC ) , and ( 2 ) addition of a feedback path ( from PT → HG ) led to poorer model performance . Since AIC and BIC strike a tradeoff between predictability and cost ( measured in terms of number of parameters ) of the model , this implies that the feedback path does not significantly increase the predictability but adds to the cost . We have estimated 54 further models that include back connections to HG from PT and STS ( in both serial and parallel models ) and HG → STS → PT models . These models , which are anatomically and functionally plausible , are schematically represented in Figure S1 . A plot of posterior probabilities of all the 70 models , with the first 16 as shown in Figure 1 and the next 54 as shown in Figure S1 , is shown in Figure S2 . On evaluating all the 70 estimated models ( Figure S2 ) , participants 1 to 6 continued to show very strong evidence in favour of model 1 . In participants 7 and 8 , however , the model selection procedure failed to identify an optimal model , although for different reasons . In participant 7 , there was no decisive evidence in favour of any model: the Bayes factor for comparing model 10 to model 1 ( the latter being the optimum model in the first six participants ) was only 1 . 4 . This designates very little evidence in favour of model 10 and is substantially below the threshold ( i . e . , 3 ) that is commonly used in Bayesian statistics to decide between two models [16] . In contrast , in participant 8 , the two approximations to the model evidence ( AIC and BIC ) favoured different models ( 21 and 19 , respectively ) . Similarly , model 1 was superior to model 21 according to the BIC criterion , but inferior according to the AIC criterion . These contradictory constellations represent a limitation of the model selection procedure adopted here , which , in cases like this particular participant , prevents one from drawing a firm conclusion about which model is optimal [14] . Overall , therefore , six out of eight participants showed strong evidence in favour of model 1 , and the remaining two participants failed to show consistent evidence in favour of any one model . A major challenge in auditory cognition is to relate cognitive processes to dynamic interactions among cortical regions . DCM was designed specifically for functional imaging data to model and draw inferences about effective connectivity between different regions . The present study aimed to understand the systems-level organisation of the computational mechanisms in the HG , PT , and STS invoked for the analysis of the spectral envelope of sounds . Two broad categories of models , serial and parallel , were specified a priori . The data provide very strong evidence for a serial model in which analysis of the spectral envelope specifically enhances the connection from the HG to the PT . In contrast to the visual system [17–22] , the effective connectivity between auditory areas has not been studied extensively . The few previous studies used structural equation modelling ( SEM; [23–25] ) . The present study is the first to use DCM to examine the auditory system . Effective connectivity between the HG and the PT was suggested by a previous study [23] using SEM . However , the results were inconsistent across participants and also between the group analysis and individual participant analyses . Another study using SEM [25] considered the connection between the HG and PT and frontal areas , but did not examine local connection with the STS . In the present study , we have provided evidence for a consistent model across participants based on serial analysis in the HG , PT , and STS . The group analysis also concurs with the individual participant analysis . We now consider the limitations of the model supported by our data , and its biological significance . DCM models the causal influence of the neural activity in one area on another , where those areas have a direct or indirect anatomical connection . The connections that we have specified a priori are plausible , given available macaque data [26] , but data on the interconnections between human cortical areas are limited to a small number of postmortem dye-tracing studies [27] and “opportunistic” studies of neurosurgical patients [28] . The model supported here is both consistent with the existence of direct anatomical projections between the HG and the PT and between the PT and the STS , and also provides evidence for the functional expression of these projections . The connection between the HG and the PT is supported by the neurophysiological evidence and tracing studies above , but further evidence for the anatomical connection between the PT and the STS , particularly , is required . Such evidence might accrue from further postmortem work or the application of in vivo techniques such as diffusion tensor imaging . In this study , we have tested the simplest possible model to describe the data in individual participants . In particular , the HG volume we have used is likely to contain primary and secondary areas: we have previously argued [29] that there are three functional areas in the HG that correspond to the three macaque “core” areas A1 , R , and RT . Kaas and Hacket [26] have described a macaque scheme based on a pattern of connectivity that extends from core to belt to parabelt areas . The connectivity structure of the serial model that was selected as optimal in this study is consistent with such a scheme , if PT contains the homologues of belt areas . However , the detailed pattern of interconnections within the HG could not be assessed in this dataset , as three distinct functional areas in the HG were not demonstrated in all the individual participants . The extent to which the analysis may involve several different functional areas within the HG before analysis occurring in the PT therefore cannot be determined . Like the HG , the PT is a large anatomical area , corresponding to the cytoarchitectonic area Te 3 . 0 [30] , within which there may be a number of functional subdivisions . Homology with the macaque becomes even more difficult than in the case of core areas . One possibility is that there may be “belt homologue” areas in the PT adjacent to the three “core” areas suggested in the HG . The connectivity between the HG and the PT identified in this analysis would then be broadly congruent with the core-to-belt projections that have been identified in the macaque [26] . Recording work in the macaque suggests that more anterior belt areas are critical for auditory object analysis , although the distinction between anterior and posterior auditory areas is not as marked as in the case of spatial analysis [31] . Connections to the STS are much more difficult to characterise in terms of homology , especially in view of the existence of three temporal gyri in the human and two in the macaque . Human functional data for the STS demonstrate complex cognitive analysis , including voice processing [32] and the integration of auditory and visual object information [33] . Whether or not the macaque homology holds , however , the other human studies suggest a role for the STS in associative analysis , whereas the serial analysis we have demonstrated is also hierarchal: perceptual analysis in the earlier areas ( HG and PT ) precedes more complex associative analysis in the later area ( STS ) . The model identified as optimal by our Bayesian selection procedure is characterised by a serial architecture in which high activity in higher auditory areas during spectral envelope extraction is explained by a modulation of the HG → PT connection . In neurophysiological terms , this means that only the HG → PT connection is dependent on the spectral envelope modulation , and that the induced context-dependent response in the PT is simply relayed on to the STS . In functional terms , this means that spectral envelope analysis is likely to be completed at the stage of the PT , and that the differential responses in the STS are a downstream reflection of this process . In contrast , if we imagine that model 6 ( compare Figure 2 ) had been selected as optimal , the interpretation would have been that context-dependent connectivity was restricted to the STS → PT connection and that functionally , spectral envelope analysis is likely to be performed at the level of the STS and the results fed back to the PT via the STS → PT connection . Completion of spectral envelope analysis at the PT in the absence of a task is consistent with the “obligatory” abstraction of templates before the PT that does not depend on the existence of a task . It will be of interest in future studies to see if the presence of an active task produces modulation of the second serial stage between the PT and the STS . This is also interesting in terms of the idea that the PT may be a critical computational “hub” where spectro-temporal “templates” are extracted before analysis in higher centres that assess the significance of a particular template ( such as its relevance to position in space or semantic category ) [15] . This abstraction is homologous to feature extraction or selection in machine learning: feature selection is a process commonly used in machine learning , in which features available from the data are selected for subsequent inference and learning . The spectral envelope corresponds to a type of template that is independent of the spectral fine structure of the sound , and is important for source identification independent of the pitch of the source or whether it was producing a harmonic sound or noise . There are certain limitations of the models that have been tested in the present work . First , only cortical connections have been considered . The thalamic connections were not included in the models , because of ( 1 ) the absence of activation in the auditory thalamus due to our experimental manipulation and ( 2 ) the evidence from animal work that complex spectral analysis first occurs in the primary auditory cortex [8] . Also , only models of the right hemisphere have been considered here , and hemispheric interactions have been ignored . This is because the conventional fMRI analysis of spectral envelope processing has consistently demonstrated a dominant role of the right hemisphere , with substantially less involvement of the left hemisphere . Sequences of harmonic or noise stimuli were synthesised digitally at a sampling frequency of 44 . 1 kHz and 16-bit resolution . The harmonic stimuli were harmonic series , whereas the noise stimuli were random-phase noise . The stimuli were synthesised in the frequency domain , allowing the same spectral envelope to be applied to either harmonic or noise sounds . The duration of stimuli was 500 ms ( with a 20-ms gating window ) . Synthesized sounds were used to form two sets of sequences . The first set , called “all-harmonic” , consisted of harmonic sounds only; the second set , known as “alternating” , consisted of alternating harmonic and noise sounds . The “all-harmonic” set has three experimental conditions: ( 1 ) the spectral envelope and pitch ( fundamental frequency ) of the sounds in the sequence are fixed; ( 2 ) the spectral envelope is fixed , but the fundamental frequency of sounds in the sequence is changing; and ( 3 ) the spectral envelope is changing , but fundamental frequency is fixed . The fundamental frequencies of the sounds in this set are 120 , 144 , 168 , or 192 Hz , either fixed or varied between successive sounds in the sequence . The “alternating” set has two conditions: ( 1 ) harmonic and noise sounds alternating with fixed envelope and ( 2 ) harmonic and noise sounds alternating with changing spectral envelope . In total , the experiment has six conditions , five as described above , and the silence condition . The conditions are schematically shown in Figure 4 . Change in fundamental frequency f0 is perceived as change in pitch , whereas change in the spectral envelope is perceived as a change in the identity of the source . The critical contrast in the group analysis to assess the “extraction” of the spectral envelope is the contrast between the two alternating conditions with changing and fixed spectral envelopes when the fine spectral structure of the stimuli is continually changing . The difference between these two conditions corresponds to an alteration in the perceived source over and above the low-level analysis of the fine spectral structure . That contrast was used to define spectral envelope extraction at the group level . The total duration of each sequence was 7 . 5 s or 8 s . Before carrying out the fMRI experiment , the participant's ability to perceive the change in the spectral envelope was assessed in a separate psychophysical experiment . The same elements of the sequences used in the fMRI experiment were presented to the participants in a two-interval–two-alternative forced-choice paradigm . The task was to detect change in pitch ( in all harmonic sequences ) or change in the spectral envelope ( all-harmonic or alternate conditions ) . Participants were able to detect harmonic sequences with pitch change or spectral shape change with 100% accuracy . Participants were also able to detect change in the spectral envelope in ( alternating condition ) with 100% accuracy . The changes in spectral shape therefore could be reliably detected independent of fine spectro-temporal changes . Data from eight healthy volunteers were used for DCM . All participants gave their informed consent , and the experiment was carried out with approval of the local ethics committee . fMRI data were acquired from a 1 . 5-T Siemens SONATA system ( http://www . siemens . com ) using gradient echo planar imaging ( echo time = 50 ms; flip angle = 90 degrees ) in a sparse image acquisition protocol [34] . Stimuli were presented diotically at a fixed sound-pressure level of 80 dB during the silent phase of the protocol . A whole-brain volume of 48 slices ( 2-mm thickness , in plane resolution 3 × 3 mm2 ) was acquired every 12 . 5 s with a time for acquisition of 4 . 32 s . Participants were instructed to attend to the stimuli with their eyes closed . In a typical trial of image acquisition , stimulus is first presented for about 8 s , followed by image acquisition that lasts 4 . 32 s . There was no active auditory discrimination task , however; to maintain attention , participants were asked to signal the end of each sequence by pressing a button box under the right hand . The experiment was divided in two runs , with 16 scans acquired for each condition in each run . The order of conditions was fully randomised . Images were realigned , normalized to a standard EPI template , and smoothed with a 3-D Gaussian kernel with full-width half-maximum of 8 mm . Regressors for the design matrix were created by convolving boxcar stimulus functions ( representing stimulus events ) with a canonical hemodynamic response function . Linear contrasts of parameter estimates were created for each participant . Finally , a random-effects group analysis was performed by comparing the participant-specific contrast images with the appropriate t-tests to produce a statistical parametric map . The general goal of DCM is to provide mechanistic explanations , in terms of connectivity and its modulation , for local effects observed in a conventional univariate analysis . The SPM results of the present data demonstrated a neural system in which the lowest level ( i . e . , the primary auditory cortex in the HG ) does not show any significant activity differences between experimental conditions , but is uniformly driven by auditory stimulation . In contrast , higher auditory areas ( PT and STS ) show higher activity when spectral envelope extraction is required . These two observations could be potentially explained by a network model ( i . e . , DCM ) in which a “neutral” input area , perturbed by auditory stimuli per se , drives two higher auditory areas differentially ( i . e . , some or all of the efferent connections of the input area are modulated by spectral envelope extraction ) . Our DCM included three areas ( HG , PT , and STS ) in the right hemisphere . These areas were identified for each participant based on the coordinates of the peak activation obtained in the group analysis . For the HG , the contrast ( condition 4 + condition 5 ) versus silence was used to define the centre of the volume from which the time series was extracted . For the PT and the STS , the contrast between the alternating sequences with variable and fixed spectral envelopes ( condition 5 versus condition 4; see Figure 4 ) was used to define the centres of the volumes . The centre of each volume ( defined as a sphere of 4-mm radius ) was located at the local maximum that was nearest to the peak coordinates in the group analysis . The selected local maximum was constrained to lie within 16 mm ( twice the width of the Gaussian smoothing kernel ) of the group peak coordinates and within the same anatomical gyrus/sulcus as the group activation . The coordinates of peak activation for the three volumes in each participant are given in Table 5 . A summary time series from each of the three regions was furnished by the principal eigenvariate of measurements recorded from all significant voxels located within the volume . From a system theory point of view , the brain can be treated as a nonlinear input–output dynamic system that can be excited by controlled stimuli and which response ( hemodynamic response here ) can be measured . The central idea behind DCM is to estimate and draw inferences about the causal interaction between different regions of the brain by identifying a model for the system using input–output measurements . In DCM , three different sets of parameters are used . The first set of parameters , known as intrinsic parameters , models the anatomical or hardwired connection strengths between the regions . These parameters represent the influence that one region has over the other in the absence of any external excitation of the system . The second set of parameters , known as modulatory parameters , models the change in intrinsic connection strength that is induced by the external experimental input . These parameters are therefore input-specific and are also referred to as “bilinear terms or parameters . ” The third set of parameters models the direct influence of an external stimulus on a given region . The conventional general linear model analysis is based on the assumption that any external stimulus has a direct influence on a region; therefore , it is the third set of parameters on which a general linear model analysis is based exclusively . DCM , therefore , can also be regarded as more general , with the general linear model analysis being a specific situation in which the interaction parameters ( first and second sets ) are assumed to be zero . DCM has several advantages over other models of effective connectivity ( e . g . , SEM [20] , multivariate autoregression [35] , or Granger causality [36]; see [13 , 37] for details ) . For example , DCM takes temporal order ( and autocorrelation of the fMRI time series ) into account . It further allows one to model the effects of experimentally controlled manipulations as either affecting regional activity directly ( e . g . , sensory inputs ) or modulating the strengths of connections , and does not need to assume that the system is driven by stochastic innovations . Most important , however , DCM is currently the only model of effective connectivity that combines a neural population model with a biophysical hemodynamic forward model , and is thus able to model how system dynamics at the ( hidden ) neuronal level translates into measured BOLD signals . In brief , DCM is based on a bilinear model of neural population dynamics that is combined with a hemodynamic model [38 , 39] , describing the transformation of neural activity into predicted BOLD responses . The neural dynamics are modelled by the following bilinear differential equation: where z is the state vector ( with one state variable per region ) , t is continuous time , and uj is the j-th experimental input to the modelled system ( i . e . , some experimentally controlled manipulation ) . This state equation represents the strength of connections between the modelled regions ( the A matrix ) , the modulation of these connections as a function of experimental manipulations ( e . g . , changes in task; the B ( 1 ) . . . B ( m ) matrices ) , and the strengths of direct inputs to the modelled system ( e . g . , sensory stimuli; the C matrix ) . These parameters correspond to the rate constants of the modelled neurophysiological processes . Combining the neural and hemodynamic model creates a joint forward model , which is inverted using conventional techniques ( expectation maximisation ) to give the posterior density of the parameters . Under Gaussian assumptions , this density can be characterised in terms of its maximum a posteriori estimate and its posterior covariance . This density obtains by optimising a free-energy bound on the models log-evidence or marginal likelihood . DCM is a hypothesis-driven technique in which model space is specified a priori . The first objective of the present study was to test if the coupling between the HG , PT , and STS is serial or parallel . To address this , two broad categories of models , serial and parallel , were specified ( Figure 2 ) . In the serial models , auditory inputs entering the HG reach the STS via the PT , and , thus , processing in the STS depends on inputs from the PT . In contrast , in the parallel models , the HG connects to both the PT and the STS , thus enabling a parallel processing in the PT and STS . The second objective was to determine where , in the best model , task requirements ( i . e . , spectral envelope analysis ) led to changes ( i . e . , modulation ) in the connection strengths . The modulatory input is defined as condition 5 of the experiment . In total , 16 models ( nine serial , seven parallel ) were inverted and compared using their log-evidence . These models are shown in Figure 2 . To rule out the possibility of other theoretically possible models , 54 additional models shown in Figure S2 were also estimated , and their log-evidence was computed . A general problem that arises in any modelling exercise is to decide , given some data , which of several competing models is optimal . A number of criteria have been proposed in the modelling literature [40] . From a Bayesian perspective , an optimal criterion is the model evidence ( i . e . , the probability p ( y \| m ) of obtaining the data y given a particular model m [16] ) . Critically , the model evidence not only takes into account the relative fit of competing models , but also their relative complexity ( i . e . , the number of free parameters ) . This is important because there is a tradeoff between the fit of a model and its generalizability ( i . e . , how well it explains different datasets generated from the same underlying process ) . As the number of free parameters is increased , model fit increases monotonically , whereas beyond a certain point , model generalizability decreases . The reason for this is “overfitting”: an increasingly complex model will , at some point , start to fit noise that is specific to one dataset and thus become less generalisable across multiple realizations of the same underlying generative process . As the model evidence cannot always be derived analytically , two commonly used approximations are the AIC and the BIC [14] . These approximations , however , do not necessarily give identical results because the BIC favours simpler models , whereas the AIC is biased toward more complex models . Here , we have adopted the usual conventions ( compare [14] ) ( 1 ) that a conclusion can only be drawn if these two criteria agree , and ( 2 ) that the more conservative of the two estimates is chosen . Finally , the relative evidence of one model as compared with another is expressed by the so-called “Bayes factor”: where BF12 is the Bayes factor of model 1 with respect to model 2 . Following the selection of a best model for each participant , the optimal model for a group of participants can be determined by the GBF , which is the product of the Bayes factors for each individual participant [41] .	The past decade has seen a phenomenal rise in applications of functional magnetic resonance imaging for both research and clinical applications . Most of the applications , however , concentrate on finding the regions of the brain that mediate the processing of a cognitive/motor task without determining the interaction between the identified regions . It is , however , the interactions between the different regions that accomplish a given task . In this study , we have examined the interactions between three regions—Heshl's gyrus ( HG ) , planum temporale ( PT ) , and superior temporal sulcus ( STS ) —that have been implicated in processing the spectral envelope of sounds . The spectral envelope is one of the dimensions of timbre that determine the identity of two sounds that have the same pitch , duration , and intensity . The interaction between the regions is examined using a system-based mathematical modelling technique called dynamic causal modelling ( DCM ) . It is found that flow of information is serial , with HG sending information to PT and then to STS with the connectivity between HG to PT being effectively increased by the extraction of spectral envelope . The study provides evidence for an earlier hypothesis that PT is a computational hub .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "neuroscience", "homo", "(human)" ]	2007	Hierarchical Processing of Auditory Objects in Humans
Human immunodeficiency virus ( HIV ) persistence in latently infected resting memory CD4+ T-cells is the major barrier to HIV cure . Cellular histone deacetylases ( HDACs ) are important in maintaining HIV latency and histone deacetylase inhibitors ( HDACi ) may reverse latency by activating HIV transcription from latently infected CD4+ T-cells . We performed a single arm , open label , proof-of-concept study in which vorinostat , a pan-HDACi , was administered 400 mg orally once daily for 14 days to 20 HIV-infected individuals on suppressive antiretroviral therapy ( ART ) . The primary endpoint was change in cell associated unspliced ( CA-US ) HIV RNA in total CD4+ T-cells from blood at day 14 . The study is registered at ClinicalTrials . gov ( NCT01365065 ) . Vorinostat was safe and well tolerated and there were no dose modifications or study drug discontinuations . CA-US HIV RNA in blood increased significantly in 18/20 patients ( 90% ) with a median fold change from baseline to peak value of 7 . 4 ( IQR 3 . 4 , 9 . 1 ) . CA-US RNA was significantly elevated 8 hours post drug and remained elevated 70 days after last dose . Significant early changes in expression of genes associated with chromatin remodeling and activation of HIV transcription correlated with the magnitude of increased CA-US HIV RNA . There were no statistically significant changes in plasma HIV RNA , concentration of HIV DNA , integrated DNA , inducible virus in CD4+ T-cells or markers of T-cell activation . Vorinostat induced a significant and sustained increase in HIV transcription from latency in the majority of HIV-infected patients . However , additional interventions will be needed to efficiently induce virus production and ultimately eliminate latently infected cells . ClinicalTrials . gov NCT01365065 One of the major barriers to a cure for human immunodeficiency virus ( HIV ) infection are long lived latently infected memory CD4+ T-cells that persist in patients on suppressive antiretroviral therapy ( ART ) [1] , [2] . One approach currently being investigated to eliminate latently infected cells is to induce production of virus from latently infected cells making the recently activated latently infected cell susceptible to death from virus-induced cytolysis or induction of HIV-specific T-cells [3] . Histone deacetylase inhibitors ( HDACi ) can activate HIV production efficiently in nearly all latently infected cell lines [4]–[9] and in many but not all primary CD4+ T-cell models of latency [10] . Using resting CD4+ T-cells from HIV-infected patients on cART ex vivo , HDACi induce both virus transcription and production of free virus [11] , although the amount of virus produced from resting CD4+ T-cells is significantly less than that induced by a T-cell mitogen [12] . The pan HDACi vorinostat , the first HDACi to be licensed for the treatment of cutaneous T-cell lymphoma [13] is a less potent activator of latent HIV than other HDACi such as romidepsin [11] , [12] , [14] , but has been clearly shown to induce virus production from resting CD4+ T-cells from HIV-infected patients on cART ex vivo by some groups in both the absence [15] or presence [8] , [16] of activated feeder cells while other groups have shown minimal virus production [12] , [17] . There has been recent data suggesting that HDACi may only activate HIV transcription through stimulation of a host gene promoter leading to the production of chimeric host-HIV transcripts , or read-through transcripts , and not true CA-US HIV RNA raising a concern that HDACi are unable to induce virion production [12] . However , studies using other models of ex vivo stimulation of resting CD4+ T-cells from HIV-infected patients on ART have not supported these findings [11] . Furthermore , recent data from a clinical trial of the HDACi romidepsin clearly demonstrated that virus could be produced in vivo following intravenous administration of this HDACi to HIV-infected patients on ART [18] . Recently , vorinostat was demonstrated to activate HIV transcription in vivo in resting memory CD4+ T-cells in HIV-infected subjects on ART who had been selected based upon ex vivo increase in HIV transcription by vorinostat [16] , [19] . We hypothesized that a multi-dose course of vorinostat would increase HIV transcription in CD4+ T-cells in blood from unselected HIV-infected patients on suppressive ART . We aimed to determine the safety and tolerability of short course vorinostat in HIV-infected patients on ART and to determine the effect on cell associated unspliced ( CA-US ) HIV RNA in CD4+ T-cells and the number of latently infected cells in blood and rectal tissue . Participants' median baseline CD4+ T-cell count was 721 ( IQR 621 , 907 ) cells/µl and duration of virus suppression was 5 . 0 ( IQR 3 . 9 , 7 . 5 ) years ( Table 1 ) . All enrolled subjects completed the study as planned . Adverse events were mild or moderate in severity ( Tables S1 and S2 ) and there were no significant interactions with ART ( Table S3 ) . Changes in histone acetylation were measured by flow cytometry ( for acetylated ( Ac ) H3 , Ac lysine and AcH4 ) or western blot ( Ac H3 ) ( Figure 1A–C ) and using flow cytometry we observed a statistically significant increase in Histone 3 and total lysine ( K ) acetylation during administration of vorinostat which returned to baseline following cessation of drug ( Figure 1D ) . The time to peak acetylation for each patient was variable with 9 of 20 participants only achieving peak acetylation of H3 or lysine , after 7 days of drug . There was a significant increase in CA-US HIV RNA in CD4+ T-cells from blood between baseline and the primary endpoint at day 14 ( p<0 . 001 ) . Intra-individual change in CA-US HIV RNA from baseline was significant in 90% ( 18/20 ) of participants on at least one time point during vorinostat dosing . The median fold change in CA-US HIV RNA from baseline to peak in CD4+ T-cells from blood was 7 . 4 ( IQR 3 . 4 , 9 . 1 ) and in rectal tissue from baseline to day 14 was 1 . 4 ( IQR 0 . 8 , 2 . 8; Figure 2A; individual fold changes shown in Figure S2 ) . The time to peak change in CA-US HIV RNA in blood varied from 8 hours to 84 days ( Figure 2B ) . CA-US HIV RNA at peak and day 84 correlated with baseline values ( p<0 . 0001 for both; ρ = 0 . 23 and 0 . 39 respectively; Figure 2C ) . The increase in CA-US RNA was statistically significant by 8 hours after first dose and remained elevated throughout follow-up , including throughout the 70 day period after vorinostat dosing ( p<0 . 001 for all time points for both comparison of raw data in copies per million 18s or when measured by fold-change , except day 28; Figure 3A ) . Using a generalised estimating equation ( GEE ) analysis , the mean fold change in CA-US HIV RNA relative to baseline at time points during vorinostat was 2 . 65 ( 95% CI 1 . 76 , 3 . 52 , p = 0 . 023 ) and at time points after vorinostat ( study days 21 , 28 and 84 ) was 3 . 00 ( 95% CI 2 . 16 , 3 . 84 , p = 0 . 018 ) . There was a trend toward an increase between baseline and day 14 in CA-US HIV RNA in CD3+ T-cells from rectal tissue ( p = 0 . 08; Figure 3A ) . There was no statistically significant correlation between changes in CA-US HIV RNA and changes in acetylation of H3 , lysine and H4 ( p>0 . 05 for all comparisons ) . The detection of an increase in HIV RNA in plasma following vorinostat was measured in real time using a commercial assay ( Roche ) with a lower limit of detection ( LLOD ) of 20 copies/ml and on batched frozen plasma using a more sensitive assay that that had a LLOD of 0 . 3 copies/ml ( single copy assay , SCA ) [20] . Despite the significant increase in CA-US HIV RNA , we found no significant increase in plasma HIV RNA using the SCA ( Figure 3B ) or the commercial HIV RNA assay . One participant had a qualitative increase ( from <LLOD to >LLOD ) in plasma HIV RNA at more than one time point during the study ( peak 160 copies/ml at day 7 ) with a significant increase in CA-US HIV RNA ( peak at day 28 ) and marked increase in PD-1 expression on CD8+ T-cells ( Figure 4 ) . This patient had had evidence of long term durable control of HIV RNA on ART with plasma HIV RNA <50 copies/ml , measured on 13 occasions over 5 years prior to enrolment in this study . Consistent with the absence of production of HIV RNA in plasma , we found no change in HIV DNA ( Figure 3C ) , no change in integrated DNA ( n = 11; Figure S1A ) nor inducible virus using a novel limiting dilution quantitative assay that detects multiply spliced ( MS ) tat and rev HIV RNA following activation with phorbol myristate acetate ( PMA ) and ionomycin ( n = 6; Figure S1B ) . Finally , there was no change in HIV DNA in rectal tissue prior to and following vorinostat ( Figure 3C ) Given there are no sensitive markers to determine HIV protein expression in vivo , we measured gag-specific T-cells as a strategy to detect any potential change in protein expression and/or the adaptive immune response following vorinostat . Gag-specific CD4+ and CD8+ T-cells did not increase following vorinostat , despite an increase in staphylococcal enterotoxin B ( SEB ) -specific interferon gamma ( IFN-γ ) producing CD8+ T-cells over the study duration measured by either fold change compared to baseline ( p = 0 . 04; n = 11; Figure 5A ) or the absolute percentage of CD8+ T-cells expressing IFN-γ ( p = 0 . 04; ) . There was an increase in regulatory T-cells observed that returned to baseline following cessation of drug ( Figure 5B ) . There were no changes in markers of immune activation and differentiation in blood and rectal tissue ( Figure S3 ) . We next used Illumina microarrays to characterize the kinetics and nature of host gene expression changes in whole blood following vorinostat administration and to determine whether there was a distinct transcriptional profile associated with changes in CA-US RNA . We observed highly significant changes in expression of multiple host genes compared to baseline at all time points including at two hours post vorinostat ( Figure S4 ) . HDACi compounds mediate dynamic changes in chromatin states by binding to HDACs , promoting open chromatin structure , DNA hypomethylation , and histone and non-histone protein acetylation which can lead to enhanced accessibility for the basal transcription machinery . This leads to a reduction in replication fork velocity and an increase in DNA replicative stress culminating in DNA damage and double strand DNA breaks [21] , [22] . DNA double-strand break repair is essential for maintenance of genome stability following vorinostat administration . We were able to track temporal changes in gene expression associated with enhanced transcriptional activity and the DNA damage response ( DDR ) over the course of 24 h after the 1st dose of vorinostat ( Figures 6A and 6B ) and subsequent time-points up to 70 days after the last dose . At two hours following the first dose of vorinostat , we saw a burst in transcriptional activation including upregulation of sequence-specific DNA binding transcription factors YY1 , Serum Response Factor ( SRF ) and SERTA Domain-Containing Protein 3 ( SERTAD3 ) , a potent co-activator of E2F responsive promoters [23] ( Figure 6A ) . We also found upregulated expression of several Plant Homeo Domain ( PHD ) zinc finger proteins including Bromodomain and PHD Finger-containing protein 1 BRPF1 , PHF13 , PHF23 , and Inhibitor of Growth family member 2 ( ING2 ) [24] , [25] . These proteins function as histone readers recognizingH3K4 methylation and recruit histone acetyl-transferase complexes involved p53/TP53cell cycle arrest ( GADD45ACDKN2D ) and DNA repair ( RAD9A , MSH6 , BRCA1 associated RING domain 1 BARD1 , REPIN1 , and POLE3 ) ( Figure 6B ) [22] . Heat shock proteins ( HSP ) , HSP70B , HSP60 , HSP40A3 , and HSP90B1 associated with the unfolded protein response were also significantly up-regulated at this time point ( Figure 6B ) , in parallel with HDAC1 and HDAC2 , the multi-subunit HDAC-mSin3a complex ( SIN3A , SAP30 , ING2 , ARID4A , ARID5A , KDM5B ) and several chromodomain proteins , which are part of the Polycomb Repressive Complex 1 ( PRC1 ) ( orange arrowheads , Figure 6B ) . Importantly , components of the splicesome and nuclear export proteins were upregulated at two hours indicating functional splicing machinery ( black arrowheads Figure 6B ) . Since previous studies found that a switch in the subunit composition of the switch/sucrose-non fermenting ( SWI/SNF ) ATP-dependent chromatin remodeling complexes between BRG1-Associated Factor ( BAF ) and PBAF was required for nucleosomal repositioning and initiation of transcription at the HIV LTR , we searched for differential expression of variant subunits exclusive to PBAF complexes [26]–[28] . SWI/SNF complexes control the nucleosome landscape at active promoters and are required for HIV Tat transactivation [29] . Interestingly , we found that two core components of the SWI/SNF ATP-dependent BAF chromatin remodeling complex , SWI/SNF related , matrix associated , Actin dependent Regulator of Chromatin , subfamily D1 ( SMARCD1 ) , member 1 BAF60 , and SMARCB1 ( SNF5/INI1 ) were down-regulated two hours following vorinostat ( red arrowhead Figure 6B ) . SNF 5/INI1 is an essential HIV host cell factor involved in the integration of viral cDNA into active genes [30] . Its expression interferes with viral replication and knockdown has been shown to increase expression of HIV 2LTR circles , a dead end product of HIV replication [30] . In addition , recent studies suggest that loss of SNF5 activity results in the disruption of nucleosome occupancy at transcriptional start sites ( TSS ) of gene promoters resulting in the upregulation of E2F target genes associated with cell cycle and proliferation [28] . Vorinostat induced transient expression of both the serine/threonine kinase CDK9 ( red arrowhead Figure 6B ) and CCNT2 , core subunits of the positive transcription elongation factor b ( P-TEFb ) , and AFF4 and ELL , components of the super elongation complex ( SEC ) ( Table S4 ) [31] , [32] . We also found vorinostat induced expression of BRD2 , a member of the bromodomain and extraterminal ( BET ) protein family that can compete with TAT for binding to P-TEFb and suppress HIV transcription in latently infected cells ( Table S4 ) . We did not see differential expression of CycT1 that specifically interacts with Tat as part of the SEC . The rapid up-regulation of HDACs and chromatin co-repressor complexes two hours after the initial dose of vorinostat result in restoration of the normally repressive chromatin architecture leading to a shutdown of the transcription of early response genes by 8 hours that was maintained at 24 hours after vorinostat administration ( Figure 6A ) [21] . Most importantly , TNF-α and NF- κB which can promote HIV replication [10] were strongly upregulated at two hours after vorinostat and turned off by 8 hours . This is supported by the downregulation of CCAAT/enhancer binding protein ( C/EBP ) , an important transcriptional amplifier of inflammatory response genes [33] , and consistent with the anti-inflammatory effects of HDACis . Using linear regression , we observed a strong correlation between changes in CA-US HIV RNA with the transcriptional profile and pathways observed at two hours with 2201 genes differentially expressed with a p<0 . 05 ( Figure 7A ) . Pathway enrichment analysis showed upregulation of MAPKs that activate both the stress-activated protein kinase ( SAPK ) , JNK kinase pathways and extracellular signal-regulated protein kinase ( ERK2 ) and NF-κB pathways , ( black arrows Figure 7B ) . These pathways play a key role in upregulating AP-1 activity and T cell activation and differentiation and correlated with an increase in CA-US HIV RNA ( Figure 7B ) . The expression of cyclin dependent-kinase subunit 2 ( CKS2 ) , CDK2 , and CCND2 and CCND3 regulating G1-S phase transition at two hours post vorinostat was counterbalanced by the upregulation of inhibitors of cyclins CDK2 and CDK4 holoenzymes , CDKN1A , and CDKN2D p19ink4D respectively , and GADD45A associated with p53-dependent cell cycle G1 phase arrest ( green arrows ) ( Figures 7B ) . In addition , gene expression associated with the ER stress response and p53-mediated apoptosis also correlated with the increase in CA-US HIV RNA , ( red arrows ) . Given it remains unclear whether repeated doses of vorinostat induce similar upregulation of host gene , or HIV RNA expression [19] , we compared changes in the gene expression two hours following the first and two hours following the seventh dose of vorinostat . Similar responses in gene expression and pathway activity were seen in four of the five subjects ( Figure 8 and Figure S5 ) . There were some differences in genes associated with cell cycle arrest and survival ( e . g . CDKN1A ( p21 ) , BCL2L and FOXO3 ) suggesting a shift from the pro-apoptotic DDR seen after the first dose to an oxidative stress response and cell survival after the seventh dose ( Figures 8B and 8C ) . We also compared changes after the first and seventh doses using gene set enrichment specific to different peripheral blood mononuclear cells ( PBMC ) subsets . We applied a method for reconstruction of subset-dependent gene expression modules using gene set enrichment and subset-specific network analysis . These cell subset-dependent modules are referred to as Nakaya modules , where cell subset specific gene activity is visualized in the radial plot shown in Figure S5C . The radial plot shows a wedge of color that points outward ( increased expression ) or inward ( decreased expression ) . We found similar enrichment in gene expression for both the first and seventh doses in T cells , B cells , monocyte and plasmacytoid dendritic cells ( pDCs ) and significant downregulation of natural killer ( NK ) cell activity . Notably , the myeloid dendritic cell ( mDC ) population was significantly enriched in downregulated genes after the seventh dose only . Given that we observed persistent changes in CA-US HIV RNA out to 84 days ( Figure 2 ) , we asked whether significant changes in host gene expression also persisted and indeed found distinct changes in host gene expression through day 84 ( Figure 9A , S4 ) . Comparing differences in gene expression between day 1 , day 14 and day 84 normalized to baseline ( F test; Figure 9A ) , we saw a subset of differentially expressed genes after 14 days of vorinostat administration that reflected cellular detoxification processes whereas long term changes seen at day 84 ( 70 days after discontinuation of vorinostat ) were significantly associated with protein ubiquitination and upregulation of MHC Class I antigen presentation ( Figure 9B ) . Genes upregulated early ( 2 hours to one day ) after vorinostat administration occurred mostly in T cells while at day 84 , monocytes and mDCs showed the bulk of upregulated genes ( Figures 9C and 9D ) . In this study of 14 days of continuous daily dosing , vorinostat was safe , relatively well tolerated and induced a sustained increase in CA-US HIV RNA in CD4+ T-cells from blood in the majority of participants , suggesting that the latent HIV provirus in most virally suppressed adults is susceptible to HDAC inhibition . Despite the significant increase in CA-US HIV RNA following vorinostat , there was no significant change in plasma HIV RNA in all but one subject . This may be because vorinostat induced an increase in HIV transcription , as measured by CA-US HIV RNA , but subsequent blocks in virus production were not reversed , including synthesis or export of multiply spliced HIV RNA from the nucleus to the cytoplasm [34] and/or inhibition of translation by cellular microRNA [35] . Alternatively , this may be due to induction of HIV transcription in only a small subset of latently infected cells [9] , [36] or the production of incomplete short or read through transcripts , as described in ex vivo models [12] , [37] . Recent studies using resting CD4+ T-cells isolated from HIV-infected patients on ART stimulated with vorinostat and other HDACi , have also demonstrated that production of free virus from latently infected cells was rarely observed , even when an increase in CA-US HIV RNA was clearly detected [9] , [11] , [12] . Our in vivo findings are consistent with these in vitro observations although it is important to note that we did see evidence for the upregulation of genes associated with splicesome assembly and nuclear export of mRNA at 2 h suggesting functional splicing machinery ( Figure 6B ) . HDACi have a broad range of effects on both the adaptive and innate immune responses , and have been shown to induce both the number and function of regulatory T-cells [38] , as we show in this study . Vorinostat has also been shown to reduce anti-CD3 mediated T-cell proliferative responses in a dose dependent fashion [39] , although interestingly , in this study we observed a significant increase in mitogen-specific IFN-γ CD8+ T-cell responses over the duration of follow-up . However , despite these diverse changes in global T-cell function , we still saw no change in gag-specific T-cells in this study most likely because translation and HIV protein expression were insufficient to induce priming or recall of HIV-specific T-cells . The timing and magnitude of increase in both CA-US HIV RNA and histone acetylation were highly variable , consistent with previous reports of HDACi in patients with malignancy [40] . The lack of a statistically significant relationship between changes in HIV CA-US RNA and any of the markers of acetylation supports recent data suggesting that the activity of HDACi in stimulating HIV transcription may not occur by the direct effects on histone acetylation , but via effects on other proteins such as the release of free P-TEFb from the inhibitory complex 7SK snRNP [31] . HDACi have a wide ranging effect on acetylation of both histone and non-histone proteins [41] and therefore other , yet unidentified , proteins may also have played a role in activating HIV transcription . Using gene array we showed a strikingly consistent change in expression in multiple genes in all subjects as quickly as two hours following the initial dose . Although it is difficult to ascribe direct effects of vorinostat on specific pathways associated with chromatin remodeling , transcriptional activation , spliceosome activity or mRNA export with expression of CA-US HIV RNA , our analysis of host gene expression showed that vorinostat can effectively promote or prime a favorable epigenetic environment for HIV transcription . The coordinated and highly consistent changes in host gene expression that favour HIV transcription , also argues against random , non-specific read-through [12] . However , we acknowledge that distinguishing between read-through transcripts and viral CA US-RNA would need to be measured directly on these samples to confidently exclude the possibility of only read-through transcription occurring . Hence , the mechanisms by which vorinostat and perhaps other HDAC inhibitors affect HIV reactivation might be multi-factorial , with the drug directly stimulating HIV transcription and indirectly altering expression of other host genes that affect this process . Induction of HIV transcription is a critical first step in activating virus production from latency . Therefore , HDACi such as vorinostat are likely to play an important role in combination activation strategies given the significant synergism demonstrated when combining HDACi with other agents , such as protein kinase C activators and/or BET inhibitors , leading to the induction of virus production from latently infected cells from HIV-infected patients on ART ex vivo [12] , [42] , [43] . It remains unclear what the optimal dosing of vorinostat should be in activating HIV transcription . A recent study using an unconventional dosing schedule of 3 days per week of vorinostat demonstrated an increase in CA-US HIV RNA in 3 of 5 participants following multiple but intermittent doses of drug [19] . The authors noted a reduced response in production of CA-US HIV RNA after the 11th and 22nd doses , compared to the first dose , although there were too few patients and time points studied to determine whether there were any statistically significant differences in the magnitude of the changes observed [19] . In our study , multiple doses of vorinostat were clearly well tolerated and there were similar robust changes in host gene expression following the first and seventh doses . Although an increase in CA-US HIV RNA was observed in all participants who responded within 8 hours , in some participants the maximum fold increase in CA-US HIV RNA was only observed after 7 or 14 doses , arguing that multiple doses of vorinostat may be needed to maximize any changes in HIV transcription . Whether this is the case for other HDACi remains unknown . There are now more potent HDACi than vorinostat in clinical trials for malignancy and activation of latent HIV [44] , that have significantly enhanced potency in activating latent HIV infection in vitro , including panobinostat , entinostat and romidepsin [11] , [12] , [44] , [45] , and these more potent agents will hopefully show greater potency than vorinostat in vivo [44] . An unexpected finding in our study was the unique changes in host gene expression detected out to day 84–70 days off vorinostat when compared to baseline . This was despite a return to baseline levels of histone acetylation upon cessation of drug following the sustained increase whilst on vorinostat . It will be important to determine if the expression of IFN inducible genes ( ISGs ) associated with antigen processing and Class I antigen presentation at day 84 reflects long-term epigenetic changes at the viral LTR that might facilitate virus reactivation and/or that promote innate immune responses . These changes may explain the observed increased CA-US HIV RNA expression to day 84 in this study . The correlation between baseline and both peak and day 84 levels of CA-US HIV RNA suggests activation of transcription may be more efficient in patients with higher levels of basal HIV transcription . Latently infected resting memory CD4+ T-cells represent the most significant barrier to HIV eradication . In this study we demonstrated that although short-course vorinostat clearly induced CA-US HIV RNA , there was no evidence of production of free virus or elimination of latently infected cells . Vorinostat may be a feasible component of larger HIV eradication studies given its tolerability and the induction of HIV transcription in the majority of participants without the need for ex vivo screening , although the prolonged changes in host gene expression require careful long term follow up . The lack of change in the number of infected CD4+ T-cells suggests that vorinostat did not impact the size of the HIV reservoir despite a clear effect on HIV transcription . Additional interventions will be needed to efficiently induce virus production and ultimately to eliminate latently infected cells . We recruited twenty adults aged 18–60 years receiving at least three antiretroviral agents , with a plasma HIV RNA <50 copies per mL for at least three years ( excluding single viral ‘blips’ ) , a CD4+ T-cell count >500 cells/µL and documented subtype B HIV-1 infection . We excluded individuals with significant acute illness , hepatic or cardiac disease , diabetes , malignancy , transplantation or recent use of immunomodulatory agents . We initially excluded patients receiving protease inhibitor regimens but this was modified once further data had become available and subsequently there were no exclusions on the basis of antiretroviral regimen . Participants provided informed consent and The Alfred Human Research Ethics Committee approved the study . The study is registered at ClinicalTrials . gov ( NCT01365065 ) . Participants received vorinostat 400 mg orally once daily for 14 days . Blood was collected at 0 , 2 , 8 and 24 hours , and on days 7 , 14 , 21 , 28 and 84 . Rectal biopsies were performed at baseline and on day 14 . Participants were monitored for clinical and laboratory adverse events , graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events ( Version 4 . 0 ) . A Data Safety Monitoring Board reviewed safety and tolerability data after the first and tenth participants had completed dosing . The primary study objective was to evaluate the effect of vorinostat on HIV transcription . We measured CA-US HIV RNA because this is the first product of HIV transcription and is required to ultimately synthesise MS-HIV RNA , viral proteins and single stranded viral RNA needed for new virion production [46] , [47] . Secondary efficacy endpoints were a sensitive measure of plasma HIV RNA with a lower limit of detection ( LLOD ) of 0 . 3 copy per mL [20] cell associated and integrated HIV DNA ( a measure of the total number of infected CD4+ T-cells ) ; and histone ( H3 , H4 and lysine ) acetylation ( a pharmacodynamic marker of vorinostat activity ) . Safety endpoints were plasma HIV RNA measured using a commercial assay with a LLOD of 20 copies per ml ( TAQMAN v2 , Roche ) , adverse events , serious adverse events , dose limiting toxicity , CD4+ T-cell count and plasma trough concentrations of antiretroviral agents . Because HDACi can activate DNA viruses we quantified cytomegalovirus ( CMV ) and Epstein-Barr virus ( EBV ) DNA at baseline and day 28 . Trough concentrations of non-nucleoside reverse transcriptase inhibitors ( NNRTI ) or protease inhibitors ( PI ) in blood were performed at baseline and day 14 using a validated high performance liquid chromatography ( HPLC ) assay . Thawed PBMC were permeabilised , fixed with 90% methanol , then stained using antibodies to acetylated ( Ac ) histone ( H ) 3 , and Ac lysine ( Millipore , Billerica , MA ) and Ac H4 ( kind gift from Dr Jeff Lifson , National Cancer Institute Frederick , Frederick , MD ) and associated isotype controls with secondary staining with either goat-anti-rat PE or goat anti-mouse-FITC ( Invitrogen ) . Lymphocytes were gated by size and data expressed as MFI above isotype control . Fold changes were determined by comparison of MFI at each time point above baseline MFI . CD4+ T-cells were isolated from stored peripheral blood mononuclear cells ( PBMC ) using a CD4+ T-cell isolation kit and magnetic-activated cell sorting ( MACS ) columns ( Miltenyi Biotec , Teterow , Germany; purity >95% ) and RNA and DNA extracted ( Allprep isolation kit , Qiagen ) . For quantification of CA-US RNA , a semi-nested real time quantitative ( q ) PCR was used with a first round amplification of 15 cycles to ensure that following second round amplification the assay was in the linear range between 1 to 46 , 000 input copies , as previously described by Pasternak et al [48] . The second round used primers to gag [46] . HIV RNA copy numbers were standardised to cellular equivalents using an 18s RNA real time LUX PCR primer set ( Invitrogen ) . The LLOD for CA-US HIV RNA was 1 copy per well . PCR amplification of cDNA for CA-US HIV RNA was performed in quadruplicate with an intra-assay coefficient of variation ( CV ) of 32% . In all assays , a no reverse transcriptase ( RT ) control was used . If there was any amplification from the no RT control , ie . evidence of DNA contamination , a second stored sample was re-extracted . If contaminating DNA persisted , the reading was excluded . Repeat extraction was required for only 2 of a total of 200 samples analysed for this study . HIV DNA was quantified as previously described [49] . PCR for HIV DNA was performed in triplicate for all samples with an intra-assay CV of 21% . Integrated DNA was measured in total CD4+ T-cells as previously described [50] . Single cell mononuclear cell suspensions were obtained from rectal biopsies [51] . Cells were stained using a cocktail of antibodies to CD3 , CD8 , CD45 and CD4 ( Multitest , BD Biosciences , Franklin Lakes , NJ ) and sorted for CD45+CD3+ cells using high speed flow cytometry ( FACSAria , BD Biosciences ) . CA-US HIV RNA and HIV DNA were quantified as above . CD4+ T cells were isolated from PBMCs from study participants by negative magnetic selection ( StemCell ) , and stimulated with phorbol myristate acetate ( PMA; 100 ng/mL ) and ionomycin ( 1 µg/mL ) for 12 h . Serial dilutions of the stimulated cells were placed in a 96 well plate directly in RT-PCR buffer using 1 in 10 dilutions ( 4 times ) and with 24 replicates at each dilution . MS HIV RNA was quantified by semi nested real time PCR with primers in tat and rev as previously described [52] with some minor modifications . The frequency of positive cells was calculated using the maximum likelihood method [53] and this number was then expressed as a frequency of cells with inducible MS HIV RNA per million CD4+ T-cells . Immune activation and differentiation were quantified as previously described [54] . In brief , one million thawed PBMC were stained with either an activation or differentiation panel for 15 minutes at 37°C prior to fixation in formaldehyde . Both panels included CD3 V450 ( Becton Dickinson ) ; CD4 PE-Texas Red ( Invitrogen ) ; CD8 Qdot605 ( Invitrogen ) . Activation panel included HLA-DR FITC; PD-1 AF647; CD38 PE; CCR5 PE-Cy5; 45RA PE-Cy7 ( all Becton Dickinson ) ; CCR7 APC eFluor-780 ( eBioscience ) . Differentiation panel included CD45RA PE; CD28 PE-Cy5; CCR7 PE-Cy7; CD31 FITC ( all Becton Dickinson ) ; CD57 AF647 ( Biolegend ) ; CD27 AF780 ( eBioscience ) . For Tregs , PBMC were surface stained with CD4 PerCP , CD127 PE and CD25 FITC ( all Becton Dickinson ) followed by intracellular staining using eBioscience FoxP3 staining kit and FoxP3 APC as per manufacturer's instructions . Data was acquired on a BD LSR-Fortessa and analysed using FlowJo version 10 . Thawed PBMC were rested for 12 hours prior to stimulation of 1 . 5 million cells each with Brefeldin A ( Sigma Aldrich ) and either gag peptides ( 1 ug/peptide/mL; NIH AIDS reagent program ) ; Staphylococcal enterotoxin B ( SEB; 1 ng/mL ) or Dimethyl sufoxide ( DMSO ) for 6 hours . Cells were then surface stained with CD3 AlexaFluor700 , CD8 Pacific Blue , CCR7 PE-CF594 , PD-1 PE-Cy7 ( all BD Biosciences ) , CD4 Qdot 605 ( Invitrogen ) , CD45RA Brilliant Violet 650 , CD19 Brilliant Violet 510 ( Biolegend ) , CD27 APCe780 , and aqua fluorescent reactive dye ( Invitrogen ) , permeabilised with Saponin and stained intracellularly with IL-2 PerCP-Cy5 . 5 , IFNγ APC and TNFα Alexa Fluor 488 ( all BD Biosciences ) prior to fixation . Cells were acquired within 24 hrs using a BD LSR-II and analysed using FlowJo version 9 and 10 . Blood was collected directly into Paxgene tubes and cells lysed for RNA extraction as per manufacturer's instructions ( Qiagen , Valencia , CA ) . Reverse transcription reactions were performed to obtain cDNAs which were hybridized to the Illumina Human HT-12 version 4 Expression BeadChip according to the manufacturer's instructions and quantified using an Illumina iScan System . The data were collected with Illumina GenomeStudio software . Analysis of the genome array output data was conducted using the R statistical language [55] and the LIMMA statistical package [56] from Bioconductor [57] . First , arrays displaying unusually low median intensity , low variability , or low correlation relative to the bulk of the arrays were tagged as outliers and were discarded from the rest of the analysis . Quantile normalization followed by a log2 transformation using the Bioconductor package LIMMA was applied to process microarrays . The LIMMA package was used to fit a linear model to each probe and to perform a ( moderated ) Student's t test on various differences of interest . For data mining and functional analyses , genes that satisfied a p-value ( 0 . 05 ) were selected . Probes that did not map to annotated RefSeq genes and control probes were removed . When indicated , the expected proportion of false positives , the false discovery rate ( FDR ) , was estimated from the unadjusted p-value using the Benjamini and Hochberg method [58] . The full dataset was composed of 9 patients with 8 time points each . Samples were stratified into the following groups: Day 0 ( baseline ) , 2 hours , 8 hours , Day 1 , Day 14 ( all on vorinostat ) ; and Day 84 ( off vorinostat ) . One set of analyses compared each time point to baseline ( to show the persistent effect of vorinostat over time ) . The other set compared baseline gene expression at 2 hours , 8 hours and 1 day ( to isolate the early effects of vorinostat following the initial dose ) . In 5 individuals , the additional time points including Day 7 and Day 7+ 2 hours were collected to determine if changes seen 2 hours after the first dose were the same as those seen at 2 hours following the 7th dose . Heatmaps of genes differentially expressed between different groups and baseline were produced ( Figure S3 ) . Two ANOVA ( F-test ) heatmaps comparing the groups of interest: 2 hours , 8 hours , Day 1 ( Figure 6A ) and Day 0+2 hours , Day 7 , Day 7+2 hours ( Figure 8A ) were produced . The top 50 statistically significant genes are shown as symbols and plotted on the row names of the heatmaps . Gene expression within each heatmap is represented as gene-wise standardized expression ( Z-score ) , with p-value<0 . 05 chosen as the significant level . Gene Set Enrichment Analysis ( GSEA ) [59] was performed on the various contrasts of interest . GSEA is a statistical method to determine whether members of a particular gene set preferentially occur toward the top or bottom of a ranked-ordered gene list where genes are ranked by the strength of their association with the outcome of interest . More specifically , GSEA calculates a net enrichment score ( NES ) that reflects the degree to which a set of genes is over-represented among genes that are differently expressed . We apply a nominal p-value cutoff of 0 . 05 when plotting the top enriched pathways on a checkerboard figure . The NES and p-value rankings usually go hand in hand i . e . : the top 10 NES pathway scores equate with the top 10 significant p-values ) . Since gene expression is tightly regulated , we do not apply statistical cutoffs on the actual FC's of genes when performing pathway enrichment . We try to include as many genes as possible to capture enriched and coregulated transcripts within a pathway that are an indication of relative pathway activity . The significance of an observed NES is obtained by permutation testing: resorting the gene list to determine how often an observed NES occurs by chance . Leading Edge analysis was performed to examine the particular genes of a gene set contributing the most to the enrichment . Two different databases were used: Ingenuity Pathway Analysis software ( Ingenuity H Systems , www . ingenuity . com ) was used to mine canonical pathways while MSigDB ( www . broadinstitute . org/msigdb; Broad Institute , Cambridge , MA ) was used to mine chromatin and splicing pathways . A list of significant pathways ranked by p-value and NES is provided ( Dataset S1 ) . Linear regression analysis was performed between CA-US HIV RNA at 2 hours and the gene expression 2 hours after the first dose of vorinostat versus baseline ( n = 9 ) . CA-US HIV RNA was plotted as a continuous variable and correlated with distinct gene expression profiles at low and high levels of CA-US HIV RNA ( Figure 7A ) . About 2000 features passed the p-value cut off of <0 . 05 . Pathway analysis was performed on the regression features and a checkerboard figure with some of the top resulting pathways was produced ( Figure 6B ) . A radial plot ( Figure S4C ) illustrating the different enrichment scores in PBMC cell specific subsets [60] between samples collected at day 0+2 hours and day 7+2 hours is shown . Checkerboard figures were used as a representation of the pathway analysis results representing the top genes and the top pathways for a specific contrast . Checkerboard plots show the top 10 enriched pathways on one axis and leading edge analysis ( genes contributing to that enrichment ) on the corresponding axis . This approach allows quick visualization of what genes are up regulated ( red ) or down regulated ( blue ) in the respective pathway at the specified contrast . Checkerboard analysis was also performed on a cell subset level [60] and the same plots generated together with a subset enrichment heat map displaying the contrasts on the x-axis and the subset on the y-axis ( Figure 9C ) . A sample size of 20 patients gave 80% power to detect an increase in CA-US HIV RNA of 40 copies/million CD4 T cells ( primary endpoint ) and an increase in plasma HIV RNA using the single copy assay of 0 . 4 log ( secondary efficacy endpoint ) at a p<0 . 05 level of significance . Categorical variables were summarised using frequency and percentage whilst continuous variables were summarised using mean and standard deviation ( SD ) or median and inter-quartile range ( IQR ) as appropriate . Spearman rank correlation coefficients were calculated between virologic and immunologic measurements . Intra-individual comparisons of CA-US RNA and HIV DNA between baseline and post-baseline time points were performed using parametric summary measures of replicate PCR data and a parametric paired t-test as PCR replicate data were derived using a standard curve and thus approximated a normal distribution . Bonferroni adjustment was made for multiple comparisons . Whilst we were prepared to presume normality at the level of the individual replicate data directly derived from the standard curve , we were less prepared to extend this assumption of approximate normality to the more sparse , more severely skewed summary data items not directly derived from the standard curves , opting for the more conservative non-parametric approach . As such , comparisons of fold change MFI for acetylation , CA-US RNA , fold change in CA-US RNA , HIV DNA , SCA , activation and differentiation markers , integrated DNA , TILDA and ICS between baseline and subsequent time points across all patients used a non-parametric Wilcoxon signed rank test . For each statistical test , a sensitivity analysis was run consisting of parallel non-parametric testing where parametric analysis was chosen , and conversely parallel parametric testing where non-parametric methodologies were used . In each analysis there was no difference in the pattern of significance or , with regards to the modeling , the direction of the coefficients . Comparisons in CA-US HIV RNA , HIV DNA , SCA , integrated DNA and ICS between pre-vorinostat , on vorinsotat and off vorinostat time periods were also performed using a Generalised Estimating Equations ( GEE ) , using a Gaussian family structure , a link identity function and an exchangeable within-group correlation structure . A robust variance estimator was used secondary to the small sample size and the deviations from normality exhibited in the summary measure data . We further extended the GEE modeling to estimate fixed effects for assay to correct for intra-assay variability by using approximations proposed by Sutradhar and Rao [61] for GEE and further developed by Feddag et al [62] . Outcome variables were log10 transformed . All reported p values were two-tailed . A Bonferroni deflation of significance was applied for multiple comparisons , otherwise p<0 . 05 was considered significant . All analyses were conducted in Stata version 12 ( StataCorp , College Station , Texas ) .	The major barrier to curing HIV is the long term persistence of latently infected resting memory T-cells in HIV-infected patients on antiretroviral therapy ( ART ) . One strategy being pursued to eliminate latently infected cells is to activate HIV production from latently infected cells with the aim of killing latently infected cells via virus induced cell death or stimulation of an HIV-specific immune response . Histone deacetylases ( HDACs ) are important in maintaining HIV latency . Vorinostat , an inhibitor of HDACs ( HDACi ) licensed for the treatment of some malignancies , has been shown in laboratory studies and a clinical study of selected individuals to disrupt HIV latency . We examined the ability of standard dose vorinostat given daily for 14 days to activate latent HIV infection in unselected HIV-infected individuals on ART . The study showed evidence of activation of latent HIV infection in 18/20 ( 90% ) of individuals and was safe and generally well tolerated . There were significant early changes in host gene expression , which persisted during and after the period of vorinostat . No changes were seen in immune activation or number of latently infected cells . Vorinostat was able to activate latent HIV infection in most individuals . Additional interventions will be needed to eliminate latent HIV infection .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "viruses", "immunodeficiency", "viruses", "infectious", "diseases", "hiv", "infections", "medicine", "and", "health", "sciences", "medical", "microbiology", "hiv", "viral", "pathogens", "microbial", "pathogens", "biology", "and", "life", "sciences", "microbiology", "viral", "diseases", "hiv-1", "hiv-2", "organisms", "retroviruses" ]	2014	Activation of HIV Transcription with Short-Course Vorinostat in HIV-Infected Patients on Suppressive Antiretroviral Therapy
Recognizing that certain biological functions can be associated with specific DNA sequences has led various fields of biology to adopt the notion of the genetic part . This concept provides a finer level of granularity than the traditional notion of the gene . However , a method of formally relating how a set of parts relates to a function has not yet emerged . Synthetic biology both demands such a formalism and provides an ideal setting for testing hypotheses about relationships between DNA sequences and phenotypes beyond the gene-centric methods used in genetics . Attribute grammars are used in computer science to translate the text of a program source code into the computational operations it represents . By associating attributes with parts , modifying the value of these attributes using rules that describe the structure of DNA sequences , and using a multi-pass compilation process , it is possible to translate DNA sequences into molecular interaction network models . These capabilities are illustrated by simple example grammars expressing how gene expression rates are dependent upon single or multiple parts . The translation process is validated by systematically generating , translating , and simulating the phenotype of all the sequences in the design space generated by a small library of genetic parts . Attribute grammars represent a flexible framework connecting parts with models of biological function . They will be instrumental for building mathematical models of libraries of genetic constructs synthesized to characterize the function of genetic parts . This formalism is also expected to provide a solid foundation for the development of computer assisted design applications for synthetic biology . “How much can a bear bear ? ” This riddle uses two homonyms of the word “bear” . The first instance of the word is a noun referring to an animal , and the second is a verb meaning “endure” . Although the word “bear” has over 50 different meanings in English , its meaning in any given sentence is rarely ambiguous . In a simple case like this riddle , the meaning of each word can be deciphered by looking at other words in the same sentence . In other cases , it is necessary to take into account a broader context to properly interpret the word . For instance , it may be necessary to read several sentences to decide if “bear claw” refers to a body part or a pastry . A reader will progressively derive the meaning of a text by recognizing structures consistent with the language grammar . It is often difficult to understand the meaning of a text by relying exclusively on a dictionary . It is interesting to compare this bottom-up emergence of meaning with the top-down approach that made genetics so successful . The discipline was built upon a quest to define hereditary units that could be associated with observable traits well before the physical support of heredity was discovered [1] , [2] . The one-to-one relationship between genes and traits was later refined by Beadle and Tatum's hypothesis that the gene action was mediated by enzymes [3] , [4] . Cracking the genetic code has been one of the major milestones in understanding the information content of nucleic acids sequences . By demonstrating the colinearity of DNA , RNA , and protein sequences , the genetic code was instrumental in the identification of specific DNA sequences as genes . The influence of this legacy on contemporary biology cannot be underestimated . Models used in quantitative genetics predict phenotypes from unstructured lists of alleles at different loci [5] , [6] . Similarly , genome annotations remain very gene-centric . Most bioinformatics databases have been designed to collect information relative to coding regions or candidate genes . Few , if any , annotations of non-coding regions or higher order structures are being systematically recorded even for model organisms like yeast [7] , [8] . Yet , despite its success , the notion of gene appears insufficient to express the complexity of the relation between an organism genome and its phenotype [1] , [9] The elucidation of the molecular mechanisms controlling gene expression has revealed a web of molecular interactions that have been modeled mathematically to show that important phenotypic traits are the emerging properties of a complex system [10]–[15] . The development of this more integrated understanding of the cell physiology leads to a progressive adoption of the more neutral notion of genetic part as a replacement for the notion of genes associated with specific traits . Making sense of the list of parts generated in genomics , proteomics , and metabolomics has been a major challenge for the systems biology community [16]–[21] . It is becoming apparent that the genetic code captures only a small fraction of the information content of DNA molecules [22] , [23] . Yet , if there is a general agreement that the cell dynamics is somehow coded in genetic sequences , no formal relationship between DNA sequences and dynamical models of gene expression has been proposed so far . In particular , the formalization of the biological functions of genetic parts has remained elusive . As a result , building models of gene networks encoded in DNA sequences remains a labor-intensive process . This limitation has hampered the development of large families of models needed to analyze phenotypic data generated by libraries of related genetic constructs [24]–[28] . Synthetic biology is likely to be instrumental in refining our understanding of the design of natural biological systems [29] . Just like the genetic code was partly elucidated through the de novo chemical synthesis of DNA molecules [30] , [31] , the redesign of genomic sequences will shed a new light on the relations between structure and function in genetic sequences [32]–[34] . By considering biological parts as the building blocks of artificial DNA sequences [35] , designing new parts that do not exist in nature [26]–[28] , and making parts physically available to the community [36] , synthetic biology calls for a systematic functional characterization of genetic parts [37] . These efforts are still limited by the difficulty in expressing how the function of biological parts may be influenced by the structure of the DNA sequence in which they are used . It has been shown that a partial redesign of the genomic sequences of two viruses had a significant effect on the virus fitness even though the redesigns preserved the protein sequences [33] , [38] . Just as the context of the expression “bear claw” helps understand its meaning , it is necessary to consider the entire structure of the DNA molecule coding for particular genes to appreciate how those genes contribute to the phenotype . One possible approach to this problem is to extend the linguistic metaphor used to formulate the central dogma . The notions of genetic code , transcription , and translation are derived from a linguistic representation of biological sequences . Several authors have modeled the structure of various types of biological sequences using syntactic models [39]–[46] . However , these structural models have not yet been complemented by formal semantic models expressing the sequence function . An interesting attempt to use grammars to model the dynamics of gene expression did not rely on a description of the DNA sequence structure . Instead , this grammar described how various inducible or repressible promoters can transition between different states under the control of environmental parameters [47] . The simple semantic model stored in a knowledge base established a correspondence between the strings generated by the syntax and the physiological state of the cell . The Sequence Ontology [48] and the Gene Regulation Ontology [49] represent other attempts to associate semantic values with biological sequences . Their controlled vocabularies can be used by software applications to manage knowledge . However , the semantics derived from these ontologies is a semantics of the sequence annotation , not of the sequences themselves . We recently described a fairly simple syntactic model of synthetic DNA sequences [50] capable of generating a large number of previously published synthetic genetic constructs [24] , [25] , [51] . We have now enhanced this initial syntactic model with a formal semantic model capable of expressing the dynamics of the molecular mechanisms coded by the DNA sequences . Specialized terms like syntax , semantics , and others are defined in Table 1 . Our approach uses attribute grammars [52] , a theoretical framework developed in the 60s to establish a formal correspondence between the text of a computer program and the series of microprocessor operations it codes for [53] , [54] . Even though other types of semantic models have been developed since then [55] , [56] , attribute grammars still represent a good compromise between simplicity and expressivity , an important characteristic to ensure that the framework can be used by non-computer scientists . Attribute grammars make it possible to use well characterized compilation algorithms to translate a DNA sequence into a mathematical model of the molecular interactions it codes for . As the static source code of a program directs the dynamic series of operations carried out by the microprocessor based on user inputs , the compilation process translates the static information of cells coded by DNA sequences into a dynamical model of the development of a phenotype in response to environmental influences [57] . The translation of a gene network model from a genetic sequence is very similar to the compilation of the source code of a computer program into an object code that can be executed by a microprocessor ( Figure 1 ) . The first step consists in breaking down the DNA sequence into a series of genetic parts by a program called the lexer or scanner . Since the sequence of a part may be contained in the sequence of another part , the lexer is capable of backtracking to generate all the possible interpretations of the input DNA sequences as a series of parts . All possible combinations of parts generated by the lexer are sent to a second program called the parser to analyze if they are structurally consistent with the language syntax . The structure of a valid series of parts is represented by a parse tree [50] ( Figure 2 ) . The semantic evaluation takes advantage of the parse tree to translate the DNA sequence into a different representation such as a chemical reaction network . The translation process requires attributes and semantic actions . Attributes are properties of individual genetic parts or combinations of parts . Semantic actions are associated with the grammar production rules . They specify how attributes are computed . Specifically , the translation process relies on the semantic actions associated with parse tree nodes to synthesize the attributes of the construct from the attributes of its child nodes , or to inherit the attributes from its parental node . In our implementation , the product of the translation is a mass action model of the network of molecular interactions encoded in the DNA sequence . By using the standardized format of Systems Biology Markup Language ( SBML ) , the model can be analyzed using existing simulation engines [58]–[60] . We have developed a simple grammar compact enough to be presented extensively , yet sufficiently complex to represent basic epistatic interactions . The grammar generates constructs composed of one or more gene expression cassettes . The gene expression cassettes are themselves composed of a promoter , cistron , and transcription terminator . Finally , a cistron is composed of a Ribosome Binding Site ( RBS ) and a coding sequence ( gene ) . The syntax is composed of 12 production rules ( P1 to P12 ) displayed in bold characters in Figure 3 where each entry is composed of a rewriting rule ( bold ) , and semantic actions ( curly brackets ) . The symbol ε refers to an empty string , [ , ] to a list , [] to an empty list , and the ‘+’ sign indicates the concatenation operation on two lists . This syntax is comparable to the one described previously [50] except that we introduced the extra non-terminal restConstructs to allow the generation of constructs with multiple cassettes without introducing parsing problems due to direct left recursions [61] . The attributes of a part include the kinetic rates related to this part and the interaction information . For example , the attributes of a promoter include a transcription rate along with a list of proteins repressing it and the kinetic parameters of the protein-DNA interactions . For non-terminal variables corresponding to combinations of parts such as cistrons , the attributes include a list of proteins , a list of promoters , and a list of chemical equations . The equation list is used to store the model of the system behavior , while the lists of promoters and proteins are recorded for computing the molecular interactions resulting from the DNA sequence . The complete set of attributes used in this simple grammar is listed in Table 2 . If many attributes can be computed locally by only considering a small fragment of the DNA sequence , other attributes are global properties of the system . For instance , the computation of protein-DNA interactions requires access to a global list of proteins expressed by the constructs . However , this list is not available until all of the different cassettes have been parsed . The problem is overcome by using a multiple-pass compilation method . In the first pass , the compiler does not do any structural validation but builds the list of proteins in the system and passes the list as an inherited attribute to the second pass . In the second pass , the promoter-protein interactions can be calculated locally at the level of each cassette . Rules P1 to P5 define the structure of a design , while rules P6 to P12 cover the selection of a specific part for each category . In the semantic action , the relation between an attribute and its variable is indicated by a dot and constants are enclosed by brackets . For instance , gene . mRNA_degration_rate = [k6] indicates that the value of the attribute mRNA_degration_rate of a gene is a constant k6 . The attribute repressor_list used in P6 and P7 includes the name of the repressor , the stoichiometry , and the kinetic constants of the forward and reverse reactions of the protein-DNA interaction . Table S1 details the parsing steps and computational dependence of each step . Finally , the equation writing operations are handled by functions typed in italics in Figure 3 and defined in Figure 4 . The translation of the DNA sequence into a mathematical model is available as the equation_list attribute of constructs . The model outputs are generated by equations generators , which are purposely decoupled from the semantic actions . The decoupling enables the flexibility of using different equation formats to describe a biological process . The translation of the construct composed of the parts pro_u rbsA gene_v t1 pro_v rbsB gene_u t1 generates the equations displayed in the [Reactions] section of Figure 5 . Each line is composed of a reaction index ( R1 to R12 ) , the chemical equation itself , and one or two reaction parameters depending on the reaction reversibility . The initial values have been computed by assigning 1 to variables representing DNA sequences and prompting the user to set the initial condition of proteins . The scripts and data used in this report are available in Dataset S1 . The semantic model presented in the previous section is completely modular since the parameters of the model describing the construct behavior are attributes of individual parts , not of higher order structures . For instance , in the previous model ( Figures 3 and 4 ) , translational efficiency is primarily determined by the RBS sequence [62] , [63] . This association between RBS and translation rate was successfully used to design one of the first artificial gene networks [24] and is still used by many synthetic biology software applications [64]–[67] . Yet , it is also well known that translation initiation can be attenuated by stable mRNA secondary structures [68]–[70] . This leads to a situation where a translational rate can no longer be considered the attribute of an individual part but needs to be considered as the attribute of a specific combination of parts . This type of context-dependency can naturally be expressed using attribute grammars since the translation reaction is computed at the cistron level , not at the level of individual parts . Rule P5 of Figure 3 can be modified by introducing a new function to retrieve the translation rate for specific combination of gene and RBS . The get_translation_rate function checks for specific cases of interactions between an RBS and coding sequence first . If none is found , then the default RBS translation rate is used . This approach is illustrated in Table 3 using previously published data demonstrating the interference between the RBS and coding sequence [68] . Specifically , this report provides the relation expression observed in 23 different constructs generated by combining different variants of the RBS and MS2 coat protein gene . This data set has been reorganized in Table 3 by sorting the constructs according to the RBS and gene variants they used . Three of the constructs using the WT RBS sequence resulted in a maximum level of expression while the expression of the gene variants ORF4 , ORF5 , and ORF6 were expressed at a much lower level due to the greater stability of the mRNA secondary structure . A similar pattern is observed for other RBS variants ( RBS1 , RBS2 , RBS3 , RBS7 ) . For all of these RBS variants , it is possible to define the translation_rate function by associating the default translation rate with the maximum expression rate . Specific translation rates associated with particular pairs of RBS and gene variants are recorded separately . The semantic model in Figures 3 and 4 is a compact proof of concept example , but it does not capture a number of features commonly found in actual genetic constructs . In order to demonstrate that our approach is capable of modeling more realistic DNA sequences , we have extended this semantic model ( Supplementary Materials ) to translate the DNA sequences of previously published DNA plasmids that include polycistronic cassettes in different orientations [24] . This plasmid library was generated by 32 different genetic parts ( three promoters: pLtetO-1 , pLs1con , ptrc-2; eight RBS: rbsA to rbsH; and four genes: tetR , cIts , lacI , and gfp and one terminator , all in both orientations ) . The syntax generates 72 different single gene expression constructs in each orientation . By combining two genes repressing each other in a construct , it is possible to make bistable artificial gene networks that are represented in Figure 6 . These bistable networks can be used as a genetic switch . To demonstrate the potential use of a semantic model to search for a desirable behavior in a large genetic design space , we have generated the DNA sequences of all 41 , 472 possible sequences ( 722×8 RBS for the reporter gene ) having the same structure as previously described switches . All sequences were translated into separate model files and a script was developed to perform a bistability analysis of each model . Parameters of the semantic model were obtained by qualitatively matching the experimental results of the six previously published switches [24] and are summarized in Table S2 . Most of the automatically generated sequences led to inherently non-bistable networks because the necessary repressor/promoter pairs did not match . Since this specific example is particularly well understood , we could have generated a limited number of targeted constructs . Yet , we chose to generate all possible sequences to demonstrate the generality of our approach . In particular , it was important to evaluate the computational cost of generating and translating DNA sequences to ensure that it would not prevent a systematic exploration of more complex design spaces . It takes only minutes to generate 41 , 472 sequences and translate them into SBML files . Hence , the computational cost of this step is negligible compared to the time required by the simulation of the SBML files . Bistability was tested numerically by integrating the differential equations until they converged to a steady state starting from two different initial conditions . The two initial conditions started with one protein level very high and the other very low and vice versa . We characterized the bistability by computing the ratio of reporter concentration for the two steady state values . In order to globally verify the behavior of this large population of models , we focused on the 3 , 072 constructs potentially capable of bistability , 1 , 408 of which were found to be bistable . We further reduced the number of constructs used to verify the translation process from 3 , 072 to 384 by assuming that two constructs differing only in the RBS in 5′ of the reporter gene would produce the same ratio of steady state values . Figure 6 visualizes the behavior of these 384 constructs . Constructs that are not bistable have a ratio of 1 . This ratio gives insight into how the construct is expected to be experimentally detectable . Since most experimental methods cannot give an exact value of protein concentration , a high ratio is desired to rise above experimental noise . Each of the 6 windows is analogous to the previously described two-parameter bifurcation diagram for that pair of repressors [24] . This gives confidence that both the semantic model of DNA sequences and the compiler used to translate automatically generated DNA sequences give results consistent with manually developed models of this family of gene networks . In the long term , the advantage to our approach over a traditional two-parameter bifurcation is the association of discrete parameter values with specific parts . This will prove particularly valuable when the context-dependencies of parameter values are better documented experimentally . This example demonstrates the benefit of building a semantic model of synthetic DNA sequences . Even a small library of genetic parts can generate large numbers of artificial gene networks having no more than a few interacting genes . A syntactic model describing how parts can be combined into constructs is a compact representation of the genetic design space generated from the parts library . While it is possible to manually build mathematical models capturing the dynamics of some of these artificial gene networks individually , it becomes desirable to automate the process to ensure the model consistency when building large families of related models derived from the same parts library . By considering genetic parts as the terminal symbols of an attribute grammar , it becomes possible to automatically generate models of numerous artificial gene networks derived from this parts library and quickly identify the optimal designs [71] . The parameter values used in the previous example were selected to match an extremely small set of six experimental data points . Although the under-determination of the model does not make it possible to precisely estimate the value of these parameters , the example illustrates how the framework could provide valuable guidance in selecting specific parts for a design . Considering that the exact value of parameters for parts is still a far off perspective , the automatic exploration of the design space presented here will provide useful guidance in construct design . For example , robust constructs from the cusp interior of the tetR/cI and lacI/cI pairings could be built and tested while less robust switches based on the lacI/tetR pairing would be avoided . As more is learned about these parts including the specific rates in different genetic contexts , the predictive ability of such maps will increase . Other motifs could be explored in a similar manner . For example , oscillators [11] could be explored by permuting parts and calculating the model-predicted existence of oscillations as well as their period or amplitude . The approach presented in this report will be implemented into GenoCAD [72] , the web-based tool we have developed to give biologists access to our syntactic design framework . Through GenoCAD , users will benefit from the syntactic and semantic models of various parts sources ( GenoCAD provided library , MIT Registry of Standard Biological Parts , or user created parts library ) . Initially , users will be able to translate their designs into SBML files that could be imported in SBML-compliant simulation tools ( www . sbml . org/SBML_Software_Guide ) for further analysis . At a later stage , simulation results and more advanced numerical analyses will be seamlessly integrated in GenoCAD's workflow . One of the major obstacles toward the implementation of such semantic models in GenoCAD is the development of a data model allowing users to understand and possibly edit the functional model of the parts they use . A function description language called Genetic Engineering of living Cells ( GEC ) was recently introduced to specify the properties of a design [67] . GEC is capable of finding a DNA sequence that implements the desirable phenotypic functions . Several other software applications have been recently released to design biological systems from standardized genetic parts . ASMPART [65] , SynBioSS [66] , a specialized ProMot package [64] and TinkerCell ( www . tinkercell . com ) illustrate this trend . These tools are still exploratory . One of their limitations is the requirement to define parts in a specialized format , such as SBML or Modeling Description Language ( MDL ) . Furthermore , instead of defining parts interactions in the underlying parts data models , these tools rely on the user to manually define them textually [66] or graphically [64] . As a result of this specific limitation , several of these tools do not appear suitable for the automatic exploration of a design space . Moreover , they tend to rely on a loosely defined relationship between the structure of the genetic constructs and their behavior . They allow parts to be assembled in any order without regard for biological viability . Still , the scripts developed to generate our results are of lesser importance than the application of the theory of semantics-based translation using attribute grammars to the translation of DNA sequences into dynamical models representing the molecular interactions they encode . Since this approach is used to develop the compilers of many computer languages [56] , [73] , a wealth of existing theoretical results and software tools can find new applications in the life sciences . For instance , we have implemented semantic models of DNA sequences into two widely used but very different programming environments , Prolog [74] and ANTLR [75] . Future research efforts will need to investigate the pros and cons of different compiler generators and different parsing algorithms for analyzing even genome-scale DNA sequences and how they impact the ability of grammars to express various features of DNA sequences . Also , the type of attributes associated with parts is flexible . Here we primarily use mass action kinetic rates as attributes , but we could just as easily have used an emerging synthetic biology unit like polymerase per second ( PoPS ) [37] , [76] . Ultimately , tools capable of automatically generating models of the behavior of synthetic DNA sequences will be important for the advancement of synthetic biology [71] . However , these tools will need to be able to express that the contribution of a genetic part to the phenotype of an organism depends largely on the local and global context in which it is placed . The interference between RBS and coding sequence is just one example of the biological complexity that computer assisted design applications will have to properly consider . Before it will be used to build synthetic genetic systems meeting user-defined specifications , the semantic model of DNA sequences presented in this report will be instrumental in the quantitative characterization of structure-function relationships in synthetic DNA sequences . The vision of applying quantitative engineering methods to biological problems has been recognized as a promising avenue to biological discovery [29] . The critical role of artificial gene networks in the characterization of molecular noise affecting the dynamics of gene networks [77] illustrates the potential of synthetic biology as a route to refine the understanding of basic biological processes . Ongoing efforts aim to carefully define how parts should fit together syntactically and what attributes are needed to characterize their function . For example , the sequence between the RBS and the start codon has been shown to play an important role in translation rate [63] . The question arises whether the RBS should be defined to include the spacing , or if there should be a separate parts category for the spacer . The rapid development of gene synthesis techniques [78] will make it possible to investigate these questions with a base-level resolution . Beyond libraries of parts for designing expression vectors , similar curation efforts could lead to the identification of parts in genomic sequences , whereby the hypothetical function of these parts as they are expressed in attribute grammars could be tested by genome refactoring [33] .	Deciphering the genetic code has been one of the major milestones in our understanding of how genetic information is stored in DNA sequences . However , only part of the genetic information is captured by the simple rules describing the correspondence between gene and proteins . The molecular mechanisms of gene expression are now understood well enough to recognize that DNA sequences are rich in functional blocks that do not code for proteins . It has proved difficult to express the function of these genetic parts in a computer readable format that could be used to predict the emerging behavior of DNA sequences combining multiple interacting parts . We are showing that methods used by computer scientists to develop programming languages can be applied to DNA sequences . They provide a framework to: 1 ) express the biological functions of genetic parts , 2 ) how these functions depend on the context in which the parts are placed , and 3 ) translate DNA sequences composed of multiple parts into a model predicting how the DNA sequence will behave in vivo . Our approach provides a formal representation of how the biological function of genetic parts can be used to assist in the engineering of synthetic DNA sequences by automatically generating models of the design for analysis .	[ "Abstract", "Introduction", "Model", "Results", "Discussion" ]	[ "computational", "biology/synthetic", "biology", "genetics", "and", "genomics/complex", "traits", "computer", "science/numerical", "analysis", "and", "theoretical", "computing", "genetics", "and", "genomics/gene", "function", "computational", "biology/systems", "biology" ]	2009	Modeling Structure-Function Relationships in Synthetic DNA Sequences using Attribute Grammars
RNA-protein binding is critical to gene regulation , controlling fundamental processes including splicing , translation , localization and stability , and aberrant RNA-protein interactions are known to play a role in a wide variety of diseases . However , molecular understanding of RNA-protein interactions remains limited; in particular , identification of RNA motifs that bind proteins has long been challenging , especially when such motifs depend on both sequence and structure . Moreover , although RNA binding proteins ( RBPs ) often contain more than one binding domain , algorithms capable of identifying more than one binding motif simultaneously have not been developed . In this paper we present a novel pipeline to determine binding peaks in crosslinking immunoprecipitation ( CLIP ) data , to discover multiple possible RNA sequence/structure motifs among them , and to experimentally validate such motifs . At the core is a new semi-automatic algorithm SARNAclust , the first unsupervised method to identify and deconvolve multiple sequence/structure motifs simultaneously . SARNAclust computes similarity between sequence/structure objects using a graph kernel , providing the ability to isolate the impact of specific features through the bulge graph formalism . Application of SARNAclust to synthetic data shows its capability of clustering 5 motifs at once with a V-measure value of over 0 . 95 , while GraphClust achieves only a V-measure of 0 . 083 and RNAcontext cannot detect any of the motifs . When applied to existing eCLIP sets , SARNAclust finds known motifs for SLBP and HNRNPC and novel motifs for several other RBPs such as AGGF1 , AKAP8L and ILF3 . We demonstrate an experimental validation protocol , a targeted Bind-n-Seq-like high-throughput sequencing approach that relies on RNA inverse folding for oligo pool design , that can validate the components within the SLBP motif . Finally , we use this protocol to experimentally interrogate the SARNAclust motif predictions for protein ILF3 . Our results support a newly identified partially double-stranded UUUUUGAGA motif similar to that known for the splicing factor HNRNPC . RNA-protein binding is a fundamental biological interaction vital to the diverse functions of RNA , including key roles in RNA splicing , translation , localization and stability [1–4] . However , the sequence features that determine affinity to RNA-binding proteins ( RBPs ) are unknown for most RBPs , including the vast majority of the hundreds of RBPs in the human proteome . Moreover , even for RBPs with known binding motifs , existing sequence motifs are only weakly predictive of which RNA regions will be bound . Deciphering these RNA binding features is crucial for mechanistic understanding of RNA-protein binding and understanding how RNA regulation impacts human health . RNA-protein interactions are known to play a role in a wide variety of diseases including muscular dystrophy , fragile X syndrome , mental retardation , Prader-Willi syndrome , retinitis pigmentosa , spinal muscular atrophy , and cancer [1–5] . Short single motifs are usually used to describe RNA-protein binding elements , e . g . as compiled in the RBPDB experimental database [6] , but such motifs have often had poor predictive power . As an example , Hogan et al identified transcripts bound to 40 yeast RBPs and then searched their UTR regions for overrepresented sequences [7] . They were able to find statistically significant motifs for only 21 RBPs , and in many cases previously known motifs could not be found . This issue of poor predictive power for single motifs has continued even with finer resolution assays such as CLIP-seq , which can localize binding sites to within a few nucleotides [8 , 9] . For example , in CLIP-seq for LIN28–RNA interaction sites in human somatic and embryonic stem cells [10] , the most overrepresented sequence motif ( GGAGA ) was found in less than 13% of the sites . A possible explanation for this problem is that proteins have the potential to interact with multiple sequence motifs . For instance , it is known that Gemin5 , a peripheral protein of the survival of motor neuron ( SMN ) complex in metazoan organisms [11–13] , is responsible for recognition of the Sm site of snRNA [14 , 15] . This recognition is mediated by a WD40 repeat domain located in the N-terminus [16–18] , yet there is also a bi-partite non-canonical RNA-binding domain at the C-terminus which modulates IRES-dependent translation [19 , 20] . However , computational methods to distinguish multiple motifs simultaneously have not been developed . Existing computational approaches for RNA motif detection , which have been geared toward single motif discrimination , have had moderate success . RNA motif analysis has often been carried out by repurposing DNA motif finder tools such as MEME [21] , PhyloGibbs [22] or cERMIT [23] , but these methods cannot take into account RNA secondary structure . Many known RBPs do bind to single stranded RNA ( ssRNA ) , but it remains unclear how much secondary structure impacts binding . Some methods have incorporated aspects of RNA structure , e . g . by biasing for single stranded regions [24 , 25] or searching over a limited set of structural contexts ( paired , loop , unstructured , miscellaneous ) [26–28] . However , the predictive power of these methods remains low , likely because of the limited number of considered contexts compared to the diversity of possible RNA structures . For example , Kazan et al . tested their algorithm on 9 RBP-interaction sets and found an average AUC value of only 0 . 64 [28] . Approaches that consider structural contexts using machine learning algorithms such as Support Vector Machines [29] , Hidden Markov Models [30 , 31] or Deep Learning [32–34] have been developed . Some have improved cross-validation AUC values to 0 . 8 to 0 . 9 [34–36] , though common caveats to current approaches are that they rely on immunoprecipitation training sets with uncertain specificity , that they have not been developed to handle multiple motifs , or that they have abstracted structural constraints rather than considering exact RNA structures . Recently , Maticzka and colleagues developed the graph kernel-based GraphProt to handle sequence and structure together and applied it to learn motifs from CLIP-seq data [37] , finding motifs that were predictive of binding for the protein PTB . However , this approach has not been tested for RBPs that bind to double stranded RNAs , and it is unknown whether the effectiveness would depend on the types of structures to which individual proteins bind . Moreover , GraphProt reports at most one motif and classifies the remaining data as noise . A more general approach would be to use clustering to allow for multiple possible motifs . A related method is GraphClust [38] , which uses a sequence/structure graph kernel to cluster RNAs , and a recent extension called RNAscClust [39] incorporates orthologous sequence conservation to improve the RNA folding estimates into the clustering process . However , these methods are tailored to cluster non-coding RNAs , and it is unknown if they would be effective for the clustering of CLIP-seq sites . Here , we propose a method , SARNAclust ( Semi-Automatic RNA clustering ) , to cluster , as opposed to classify , RNA motifs that bind to a given RBP from CLIP-seq data . To our knowledge , this is the first approach to attempt to cluster CLIP-seq peaks in order to discover potentially multiple RNA motifs that bind to a given RBP . The most related approach we know of is AptaTrace [40] , which uses clustering to identify multiple possible RNA motifs from HT-SELEX experiments . However , AptaTrace is not optimized for CLIP-seq data since it relies on k-mer context information during evolution of a sequence pool over multiple SELEX rounds , while CLIP-seq provides a static snapshot . Another recent method , RNAcompete-S [41] clusters multiple components that contribute to a single binding motif , but is not designed to handle distinct motifs arising from separate binding domains . Here we present a novel pipeline to address these problems in RNA-protein motif identification . We first describe our pipeline , which consists of 3 steps ( peak discovery , motif discovery and motif experimental validation ) , with particular attention to the novel computational motif discovery algorithm SARNAclust . Next we benchmark SARNAclust on synthetic data and validate our experimental protocol on a known double-stranded RNA motif . We then show SARNAclust motif predictions for a set of RBPs with eCLIP data and experimentally validate the motif of one such RBP . Finally , we discuss the results and implications of this pipeline in the Discussion section . We present a mixed computational/experimental pipeline to derive RNA motifs that bind to a given RBP based on immunoprecipitation data . The motivation for our pipeline is two-fold: first , to discover motifs where both sequence and structure are necessary; and second , to enable identification of more than one motif per RBP through optimized clustering over the CLIP peaks . Our complete software pipeline includes source code to process data files from a CLIP experiment ( see Methods ) , to calculate secondary structure of the peaks using RNAfold , to cluster peaks according to sequence only , and to cluster peaks according to sequence/structure using SARNAclust . In addition , we provide a protocol for experimental validation of candidate motifs , including in silico design of instances of the motif using RNAiFold [42 , 43] . Fig 1 shows the flowchart of our pipeline . S1 Fig shows the flowchart of the peak analysis . A key element of SARNAclust is the graph transformation that allows for the calculation of a similarity value between pairs of sequence/structures . These similarity values provide the input for the clustering of CLIP peaks . Flexible parameters in SARNAclust allow it to be used as a guidance system to identify well-supported motifs and test their key features . To test the effectiveness of SARNAclust , we generated 100 sequences for each of the 5 synthetic motifs in Table 1 . We then combined these 500 sequences with 1000 random sequences to act as noise and then tested the ability of SARNAclust to sort these into separate clusters . Each synthetic motif corresponds to a hypothetical RNA motif that would bind a protein binding domain . The 5 motifs represent: a special structure with no sequence conservation ( special_structure ) or a conserved sequence within a certain structural context in a hairpin loop ( GAGA_in_Hairpin ) , in a bulge ( AUG_in_Bulge ) , in an external loop ( pyrimidine_tract ) or in a double stranded region ( GGUCG_in_left_stem ) . Sequences for each motif were generated using RNAdualPF [48] , which samples from the low energy ensemble of sequences compatible with the given structure and with the corresponding sequence constraints ( see Table 1 ) . The 1000 random sequences were generated uniformly randomly ( i . e . sampling each nucleotide with 0 . 25 probability ) with lengths distributed the same as the lengths of the synthetic motifs . All motif and random sequences can be found in S1 Data . Clustering of these 1500 sequences indicated that SARNAclust was able to distinguish multiple clusters corresponding to the original motif groups . As a clustering method we used DBSCAN from the sklearn package , surveying over possible values for the threshold parameter that specifies the minimal similarity for two data points to be in the same cluster , and with graph kernel options R = 2 and D = 2 . This threshold is a dissimilarity threshold—at a threshold of x , 2 sequence/structures cannot be in the same cluster if their similarity measure is less than 1-x . Fig 3 shows the different V-measure ( a measure of clustering quality , see Methods ) values for each graph transformation and each threshold value . S2 Data shows other measures of quality of clustering ( see Methods ) assessed by comparing the true cluster label for each sequence versus the one yielded by the clustering algorithm . We observed that SARNAclust was able to recover each category of motif , though optimization of the choice of graph transformation enhanced detection of each motif class . As can be seen , use of each graph transformation yielded the corresponding motif at low to mid threshold values but false positives increased as the threshold parameter increased . For instance , option 4 finds the GAGA_in_hairpin motif easily at low thresholds . Although option 6 performs well , it benefits from the fact that most motifs do not have external loops and may not be as general as other options . For most motif instances option 6 is only able to use the bulge graph features to discriminate motif instances from one another . The GraphProt-like options ( 1 and 2 ) perform well at high threshold values , but cannot successfully cluster at low thresholds . This is due to the excess number of features specified in this graph transformation , making it difficult to cluster instances unless they are nearly identical . Option 2 contains fewer features than option 1 and thus performs better . Options 10 and 11 are simplified GraphProt-like versions and achieve the best results with v-measure values of over 0 . 95 . Option 9 achieves high quality values as well , and for a large range of threshold , especially with respect to FMS . For comparison , we also applied GraphClust to the same set of synthetic motifs . We note that we did not choose the GraphClust extension RNAscClust here because the folded structures are pre-determined for these synthetic designed sequences , and therefore the folding improvements of RNAscClust do not offer any advantages . We used the default GraphClust parameters except graph kernel R and D , which were set to the values we used in SARNAclust , and the minimum length of sequence was set so that all the 1500 sequences would be considered . GraphClust returns by default 5 clusters , which we would expect to correspond to the 5 synthetic motifs . GraphClust returns the seed and extended sequences for each instance , and we calculated several clustering quality measures for each ( Adjusted Rand Index ( ARI ) , Adjusted Mutual Information ( AMI ) , Homogeneity Score ( HS ) , Completeness Score ( CS ) , V-measure score ( VMS ) and Fowlkes-Mallows score ( FMS ) ) . SARNAclust outperformed GraphClust as shown in Table 2 . Note that the GraphClust v-measure values in both scenarios are under 0 . 1 , while SARNAclust with option 11 achieves better results for almost all thresholds , including >0 . 95 at threshold 0 . 5 . This indicates that the difference between clustering CLIP peaks and RNAs are substantial enough that the SARNAclust provides superior performance over GraphClust . This is likely due to the large combinatorial complexity of GraphClust’s parameter space . We also compared whether a classification-based approach to motif detection could identify the synthetic motifs as well as SARNAclust . To handle multiple motifs , we used classification to identify the best motif iteratively , at each step removing the sequences containing the prior best motif . For this comparison we chose RNAcontext [28] , which uses classification to identify one motif at a time based on sequence and structure . Remarkably , the first iteration of RNAcontext could not find any of the synthetic motifs ( S3 Fig ) . In fact , the sequence motif ( of length 11 ) returned has a very low information content of 5 bits ( when a completely fixed sequence of length 11 would have an information content of 22 bits ) . This is likely because the RNAcontext approach is not well-suited to the benchmark set as it contains multiple signals from distinct overrepresented motifs as well as noise . Interference among the motifs apparently causes RNAcontext to be unable to report any single motif with high confidence . As part of our pipeline for identifying binding motifs , we developed a targeted RNA Bind-N-Seq ( RBNS ) protocol [49] to experimentally test motif predictions . We first tested this protocol on Stem Loop Binding Domain Protein ( SLBP ) , which binds a known motif [50] found in the 3’UTR of histone mRNAs . In the RNA Bind-N-Seq protocol , randomly generated 40-mers are tested for their efficiency in binding a protein . Because any RNA molecule of length 40 can form over 200 trillion secondary structures , it is not possible to fully sample this space . Therefore , we performed RBNS measurements of SLBP-RNA binding with several thousand designed sequences to ascertain the validity of our experimental validation approach . To do this , we first used RNAiFold [43] ( See Methods ) to design four different types of sequences as illustrated in S1 Table ( 153 , 4106 , 4107 and 4106 sequences of each respectively ) . These sequences were chosen to test whether binding requires sequence conservation in the loop region or the stem region of the motif , and also whether relocation of the loop sequence to a different structural context ( i . e . a bulge ) could still lead to protein-RNA binding . We then performed RBNS in duplicate using purified GST-SBP-SLBP ( Glutathione-S-transferase Streptavidin-Binding Peptide SLBP ) to pull down the designed RNA sequences [49] . As a nonspecific binding control , we also performed RBNS with the same RNA against purified GST-SBP . Each protein was expressed in E . coli and affinity purified ( S4 Fig ) , and pulled down RNA was reverse transcribed with a primer containing a 10 nt random sequence to enable collapsing of PCR duplicates during data analysis . The resulting cDNA was then PCR amplified to attach Illumina sequencing primers and indices . Only the consensus motif [50] exhibited a clear shift from the control , indicating that the motif definition is specific and that all the variant versions of the motif have decreased binding . Fig 4A shows the difference between GST-SBP RBNS and GST-SBP-SLBP RBNS , quantified by the shift in percentage of reads of each type . Only the consensus motif has a significant enrichment with respect to the control ( t-test p-val = 0 . 00147 ) . To assess p-values of individual sequences , we used DEseq [51] to compare the read counts ( S3 Data ) of sequences in the pool to the controls . This analysis showed that only sequences from the consensus motif bind to SLBP significantly . Moreover , all but 7 of these consensus sequences are significantly overrepresented in the SLBP bound pool ( Adjusted p-val > 0 . 01 ) . Furthermore , S5 Fig shows the sequence logos for all the consensus sequences that bind or do not bind significantly , respectively . The logos indicate that long stretches of U’s near the apical region of the hairpin loop compromise binding affinity , which is to be expected since they are energetically unfavorable and therefore prone to render the hairpin unstable . To further validate these results and the validity of our RBNS-like experimental protocol , we performed several gel shift experiments ( Fig 4B ) . We incubated 6 RNA probes selected from the RBNS data with purified GST-SBP-SLBP . S3 Data shows the 6 selected sequences highlighted in red . These include 2 from the consensus binding group , one with strong binding affinity in the RBNS assay ( consensus A ) and one with no significant binding affinity ( consensus B ) . There are also 4 extra sequences from the remaining types where the RBNS binding signal was not significant . As expected , only the consensus A sequence shows binding to SLBP , confirming our conclusions from the p-value analysis and validating the RBNS protocol . Given these validations of the computational and experimental pipeline , we then applied SARNAclust to predict motifs from real immunoprecipitation data . First , we verified that SARNAclust could find the motif for SLBP . In order to do so , we downloaded SLBP eCLIP [52] data from the ENCODE project ( www . encodeproject . org ) . After applying our peak discovery pipeline we were left with only 49 peaks , most of them ( i . e . 35 ) found indeed in histone genes . After calculating the secondary structure of each peak using RNAfold , we ran SARNAclust . Fig 5 shows the motif found for options 1 , 2 and 10 at threshold 0 . 6 , as well as option 7 at threshold 0 . 5 . These are the most suited options since they account for sequence in double stranded regions , which is important for SLBP . Options 7 and 10 yield less specific clustering , meaning that they need higher thresholds ( i . e . the clusters contain more sequences ) to find the motif , consistent with their being coarser representations of the sequence/structure . As can be seen , the motif found is very similar to the canonical motif [50] . No other clusters were found , showing SARNAclust is effective even if only one motif exists . We then used SARNAclust to predict motifs for several RBPs from the ENCODE project , which is generating RNA crosslinking immunoprecipitation assays that are expected to eventually cover >200 known human RNA Binding Proteins using eCLIP [52] . We downloaded a set of 20 RBPs from ENCODE eCLIP experiments at www . encodeproject . org , each with 2 replicates and a control . We selected these RBPs due to the fact that they contain either double stranded RNA binding domains or unknown RNA binding domains . We discarded all helicases since they are known to promiscuously bind to double-stranded RNAs with no clear motif . We identified novel motifs with SARNAclust using the same graph kernel parameters as in the synthetic data section ( R = 2 , D = 2 ) and the same DBSCAN algorithm for clustering . We used options 9 , 10 , 11 for their overall performance on the synthetic data , along with option 2 for its similarity to GraphClust , at the best performing thresholds ( 0 . 3–0 . 55 ) . Table 3 shows the list of RBPs chosen for this study , along with their RNA binding domains and the number of peaks found by our peak discovery pipeline . Note that none of these RBPs had previously known motifs in the two most relevant motif databases: RBPDB [6] and ATtRACT [53]; and only one RBP ( EIF4G ) has a motif described in the two recent publications on ENCODE eCLIP [54] and RBNS [55] . We found motifs for several proteins , with results dependent on the choice of options ( see S2 Table ) . However , analysis of the data under the GraphProt algorithm ( equivalent to SARNAclust with option 2 ) was unable to find any clusters for all but 2 RBPs . Similarly , RNAcontext yielded motif predictions with low Area Under the Receiver Operating Curve values , ranging from 0 . 111 ( NKRF ) to 0 . 546 ( AKAP8L ) , and most motifs had low sequence complexity and information content . In contrast , SARNAclust with options 9 , 10 and 11 was able to find clusters almost for all RBPs ( all of them for option 9 , all but 1 for option 11 and all but 5 for option 10 ) . We focused on clusters with sequence conservation signal in addition to structural information , as those without sequence information would be more difficult to interpret and experimentally validate . This is also why we removed helicases , as we would expect those to lack sequence conservation . Nevertheless , we were able to find several interesting cases , as described below . SARNAclust results were consistent with k-mer analysis but provided additional structural context . K-mer overrepresentation results [3] for k = 4 , 5 , 6 , 7 , 8 , 9 on these RBPs are shown in S3 Table . Most of the k-mers found were either GU repeats or a stretch of guanines GGGG . SARNAclust was able to find these GU repeat motifs as well simultaneously with structure . However , SARNAclust did not reveal other candidate motifs for those proteins , suggesting that those RBPs with repetitive motifs bind to double stranded structures indiscriminately . Because of the strong predictive motif for ILF3 from SARNAclust , as evidenced by its clusters and signal for double-strandedness , we next used the RBNS approach to validate these motifs . ILF3 is known to be involved in many processes such as transcription , translation , regulation of cell cycle or viral replication [57] . However splicing has only recently been reported as a possible function [58] , and confirmation of this HNRNPC-like motif would shed light on a potentially novel function . We therefore used RBNS to test the binding of the predicted motifs with ILF3 and whether it requires a specific RNA structure . Using RNAiFold we designed sequences for 4 different perturbations of the motif as shown in S4 Table ( 19 , 2680 , 2680 , and 2680 sequences of each motif class respectively ) . For each motif class , we attempted to generate a few thousand sequences . However , only 19 designed sequences were obtained for the UUUUUGAGA-unpaired motif class due to the fact that unpaired structures tend to have higher free energies than paired structures ( see Methods ) . Similarly as for SLBP , we performed RBNS with purified GST-SBP-ILF3 using an RNA pool based on the motifs in S4 Table . Fig 7 shows the shift in percentage of reads of each type and its difference between GST-SBP-1 , GST-SBP-2 non-specific binding controls and ILF3-1 , ILF3-2 samples . Only the motif UUUUUGAGA-paired exhibited a significant positive shift from the control ( p-val<0 . 005 using the t-test ) , supporting the novel motif . We also used DEseq to analyze differential representation of sequences in the ILF3 bound and unbound pools ( S4 Data ) . Only sequences from the UUUUUGAGA-paired motif showed significant binding to ILF3 , confirming and specifying the computationally discovered motif . 1551 out of 2680 sequences showed increased binding ( T-test , p-adj<0 . 05 ) , while only 6 sequences showed decreased binding . These targeted RBNS results suggest that among the motifs tested , that the UUUUUGAGA-paired shows the strongest binding . In comparison , we observed no significant enrichment or depletion in binding from any of the 19 sequences representing the UUUUUGAGA-unpaired motif . The paired and unpaired sets of sequences each spanned a range of base compositions . The only systematic difference in the two sets was that all of the unpaired motif sequences contained a poly-A sequence that bound to the UUUUU region while preventing pairing of the GAGA region . This is likely a structural constraint , though we cannot rule out that the polyA sequence could also have a sequence-dependent effect on binding . Because the sequences from all motif groups were incubated with the protein at the same time , we note that the lack of enrichment of other motifs is a comparative effect impacted by the stronger binding of UUUUUGAGA-paired motif sequences . Another factor is that RNA Bind-n-Seq does not capture indirect binding interactions mediated through multiprotein complexes , which may be relevant for some of the other motifs . A special case is the GU repeat motif predicted by SARNAclust for ILF3 ( GU-repeats motif in S2 Table ) , which did not show enriched binding . GU-rich motifs were predicted for many other ENCODE RBPs as well ( [54]and S3 Table ) , and we speculate that the presence of such sequences in CLIP data may be due to experimental noise . These results indicate that ILF3 binds to a UUUUUGAGA motif with most nucleotides in double-stranded regions . They also suggest a relationship between ILF3 and HNRNPC , which has been reported to have a similar motif , though that motif was reported to be single stranded . To explore this further , we analyzed the overlap of ILF3 and HNRNPC peaks in ENCODE eCLIP data . Overall , ILF3 has 822 peaks that fall in anti-sense Alus . 322 of these peaks are shared with HNRNPC peaks , and 146 of those ( 45% ) contain the UUUUUGAGA sequence . Conversely , ILF3 has 285 peaks in anti-sense Alus covering a UUUUUGAGA sequence , and 146 of such peaks ( 51% ) are shared with HNRNPC . Thus from both perspectives a substantial fraction of sites overlap between ILF3 peaks , HNRNPC peaks , and UUUUUGAGA sequences within anti-sense Alus . Our motif analysis predicts 67 of the 146 common sites ( 46% ) to have the paired version of the UUUUUGAGA motif , suggesting there may be some flexibility in RNA structure at the overlapping sites . SARNAclust is a novel computational method that can effectively process and analyze data from CLIP experiments in order to predict RNA motifs likely to bind individual proteins . A key novelty of SARNAclust is that it can assess RNA binding motifs at the level of the complete RNA structure , rather than only taking into account abstractions of structural context . The SARNAclust approach of clustering rather than classifying distinguishes it from prior methods , allowing it to identify motifs even without training data . This is an important aspect for CLIP-seq , for which the specificity of experimental measurements is not well understood due to diverse effects such as multiple binding modalities and sources of noise . Application of SARNAclust and the new RBNS validation approach allowed us to experimentally verify ILF3 binding to a newly predicted UUUUUGAGA motif . The signal for this was distinct from repetitive GU or CU motifs , supporting the idea that those repetitive sequences are not true binding sequences . More broadly , SARNAclust allowed us to investigate the relative importance of structure , which has been challenging for RNA-protein interactions , and we found that structure significantly affected the RBNS results for both SLBP and ILF3 . Structural changes to each of several components of the SLBP motif reduced binding , and the new motif for ILF3 exhibited a bias for double-strandedness . Although identification of RBPs that bind to multiple motifs will require further investigation , the multi-domain structure of many RBPs suggest this is a likely possibility . The combination of SARNAclust and our target RBNS validation already allows us to separate multiple distinct signals from noise , making it suited to this ongoing challenge . In contrast , other methods have more difficulty in resolving multiple signals simultaneously . In addition to our results on synthetic motifs , we found that when we selected 1000 ILF3 peaks at random and inputted them to RNAcontext , we found no similar motif to the ones output by SARNAclust ( S6 Fig ) . The similarity of the new ILF3 motif to that for HNRNPC is intriguing , as it was shown in [56] that HNRNPC competes with another protein U2AF2 for binding of 3’ splice sites to regulate the inclusion/exclusion of exons . They concluded that HNRNPC prevents inclusion of cryptic exons while U2AF2 promotes it , with RBP binding often occurring in antisense Alu elements . Based on this competition , we would expect U2AF2 to have a similar binding site to HNRNPC . However , the predicted motif for HNRNPC is much more similar to that for ILF3 than it is to the predicted U2AF2 motif ( Table 3 ) . We speculate that ILF3 might compete with either HNRNPC or U2AF2 for binding of similar regions . In this paper we have introduced a new pipeline with a powerful clustering algorithm SARNAclust for analyzing CLIP data in order to cluster CLIP peaks into different binding motifs . We have verified the effectiveness of SARNAclust on synthetic data and used RNA Bind-n-Seq to experimentally validate predictions for new and known motif predictions from ENCODE data . These studies included surveying over different biophysical models and clustering thresholds to identify those likely to work best for real datasets ( i . e . options 9 , 10 , 11 at clustering thresholds 0 . 3–0 . 55 ) . We have also shown the utility of our RBNS approach by validating its results using gel shift experiments . Still we are we are cognizant of the fact that different RBPs will vary in binding affinity and modality , particularly those with different types of RNA recognition motifs , and further studies will be needed to confirm the generality of these methods for all RBPs . In the future and as more eCLIP data sets for double-stranded binding RBPs become available , we expect SARNAclust will be a valuable tool to discover new motifs , to probe the combinatorial interactions of RNA-binding proteins , and to elucidate their functional importance . ENCODE data ( www . encodeproject . org ) correspond to a set of CLIP experiments described as enhanced CLIP ( eCLIP ) , which modifies the iCLIP method to include improvements in library preparation of RNA fragments . See [52] for details . All data were downloaded through the ENCODE Project website . For proteins with more than one experimental cell type , we used the data from the K562 female cell line . Probes were in vitro transcribed and biotinylated using the Pierce RNA 3’ end biotinylation kit . 1 nM of biotinylated probe was incubated with or without 320 nM GST-SBP-SLBP in binding buffer consisting of 10 mM HEPES ( pH 7 . 3 ) , 20 mM KCl , 1 mM MgCl2 , 20 mM DTT , 5% glycerol . The incubation period was 30 minutes , followed by gel electrophoresis on a native TBE 4% polyacrylamide gel and transfer to a nylon membrane , all at 4°C . Membranes were processed using the ThermoFisher Scientific Chemiluminescent Nucleic Acid Detection Module Kit . Images were captured on a Kodak ImageStation 4000MM Pro .	RNA-protein binding is critical to gene regulation , and aberrant RNA-protein interactions play a role in a wide variety of diseases . However , molecular understanding of these interactions remains limited because of the difficulty of ascertaining the motifs that bind each protein . To address this challenge , we have developed a novel algorithm , SARNAclust , to computationally identify combined structure/sequence motifs from immunoprecipitation data . SARNAclust can deconvolve multiple motifs simultaneously and determine the importance of specific features through a graph kernel and bulge graph formalism . We have verified SARNAclust to be effective on synthetic motif data and also tested it on ENCODE eCLIP datasets , identifying known motifs and novel predictions . We have experimentally validated SARNAclust for two proteins , SLBP and ILF3 , using RNA Bind-n-Seq measurements . Applying SARNAclust to ENCODE data provides new evidence for previously unknown regulatory interactions , notably splicing co-regulation by ILF3 and the splicing factor hnRNPC .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "rna-binding", "proteins", "applied", "mathematics", "simulation", "and", "modeling", "algorithms", "mathematics", "clustering", "algorithms", "immunoprecipitation", "sequence", "motif", "analysis", "research", "and", "analysis", "methods", "sequence", "analysis", "rna", "structure", "sequence", "alignment", "bioinformatics", "proteins", "molecular", "biology", "precipitation", "techniques", "biochemistry", "rna", "rna", "folding", "double", "stranded", "rna", "nucleic", "acids", "database", "and", "informatics", "methods", "biology", "and", "life", "sciences", "physical", "sciences", "macromolecular", "structure", "analysis" ]	2018	SARNAclust: Semi-automatic detection of RNA protein binding motifs from immunoprecipitation data
Fatty acid ( FA ) binding proteins ( FABPs ) of helminths are implicated in acquisition and utilization of host-derived hydrophobic substances , as well as in signaling and cellular interactions . We previously demonstrated that secretory hydrophobic ligand binding proteins ( HLBPs ) of Taenia solium metacestode ( TsM ) , a causative agent of neurocysticercosis ( NC ) , shuttle FAs in the surrounding host tissues and inwardly transport the FAs across the parasite syncytial membrane . However , the protein molecules responsible for the intracellular trafficking and assimilation of FAs have remained elusive . We isolated two novel TsMFABP genes ( TsMFABP1 and TsMFABP2 ) , which encoded 133- and 136-amino acid polypeptides with predicted molecular masses of 14 . 3 and 14 . 8 kDa , respectively . They shared 45% sequence identity with each other and 15–95% with other related-members . Homology modeling demonstrated a characteristic β-barrel composed of 10 anti-parallel β-strands and two α-helices . TsMFABP2 harbored two additional loops between β-strands two and three , and β-strands six and seven , respectively . TsMFABP1 was secreted into cyst fluid and surrounding environments , whereas TsMFABP2 was intracellularly confined . Partially purified native proteins migrated to 15 kDa with different isoelectric points of 9 . 2 ( TsMFABP1 ) and 8 . 4 ( TsMFABP2 ) . Both native and recombinant proteins bound to 11- ( [5-dimethylaminonaphthalene-1-sulfonyl]amino ) undecannoic acid , dansyl-DL-α-amino-caprylic acid , cis-parinaric acid and retinol , which were competitively inhibited by oleic acid . TsMFABP1 exhibited high affinity toward FA analogs . TsMFABPs showed weak binding activity to retinol , but TsMFABP2 showed relatively high affinity . Isolation of two distinct genes from an individual genome strongly suggested their paralogous nature . Abundant expression of TsMFABP1 and TsMFABP2 in the canal region of worm matched well with the histological distributions of lipids and retinol . The divergent biochemical properties , physiological roles and cellular distributions of the TsMFABPs might be one of the critical mechanisms compensating for inadequate de novo FA synthesis . These proteins might exert harmonized or independent roles on lipid assimilation and intracellular signaling . The specialized distribution of retinol in the canal region further implies that cells in this region might differentiate into diverse cell types during metamorphosis into an adult worm . Identification of bioactive systems pertinent to parasitic homeostasis may provide a valuable target for function-related drug design . Neurocysticercosis ( NC ) , caused by infection of the central nervous system ( CNS ) with Taenia solium metacestode ( TsM ) , represents one of the most common CNS helminthic diseases and invokes formidable public health problems . NC is associated with several neurological manifestations including seizure , headache and focal neurologic deficits , which may vary according to the location , number and viability of the parasites within the brain [1] . NC is endemic worldwide , but is more prevalent in Latin America , the Indian subcontinent , Sub-Saharan regions and Southeast Asian countries , where approximately 50 million people are at risk of infection . NC has been increasingly detected in developed countries due mainly to immigrants from endemic areas [2] , [3] . The clinical aspects , neuroimaging and serodiagnosis of NC have been relatively well characterized [4 and references therein] . However , the functional aspects of the pathogen including cellular biochemical and molecular mechanisms inherent to the maintenance of cellular homeostasis have largely remained elusive . Parasitic helminths exploit limited lipid metabolism due to low levels or an absence of enzymes involved in the oxygen-dependent pathway . They depend mostly on essential lipids imported from their host and have evolved special hydrophobic ligand binding systems to ensure their long-survival in the harsh , low-oxygen tension host environments [5] . A series of lipid binding proteins have been characterized from the platyhelminths . The hydrophobic ligand binding proteins ( HLBPs ) are small α-helix rich 7–10 kDa molecules with extremely hydrophobic binding site ( s ) . Their functions included uptake and storage of the hydrophobic molecules , and cellular protection by lowering free fatty acid ( FA ) concentrations below toxic levels [6] , [7] . Some of these molecules , especially those of TsM , are reliable serodiagnostic biomarkers for NC [4] , [8] , [9] . No orthologous protein has been identified in other organisms . The molecules form a novel cestode-specific HLBP family [10] showing unique properties including oligomer/multimer formation in normal physiological conditions [11] . FA binding proteins ( FABPs ) are cytosolic proteins of approximately 15 kDa . They have been implicated in intracellular uptake , transport and storage of hydrophobic ligands , regulation of lipid metabolism and sequestration of excess toxic FAs [12] , as well as in signaling and regulation of gene expression [13] , [14] . The proteins bind non-covalently to hydrophobic ligands , especially to FAs and retinol . They belong to the intracellular lipid binding protein ( iLBP ) , which comprises the calycin superfamily , together with the avidin and lipocalin families . The family members show varying degrees of sequence identity among the members ( approximately 15–70% ) , but conserve a characteristic β-barrel structure , which consists of 10 anti-parallel β-strands and two α-helices [15] . During the course of chordate evolution , an ancestral iLBP gene has diverged into three subfamilies of FABP , cellular retinoic acid binding protein ( CRABP ) and cellular retinol binding protein ( CRBP ) , after gene duplication [16] . Each of the protein subfamilies is subdivided into several isoforms with tissue-specific distribution and function in vertebrates [17] . Several FABPs from platyhelminths including Echinococcus granulosus , Mesocestoides vogae ( syn . M . corti ) , Schistosoma spp . and Fasciola spp . have been characterized [18]–[25] . These proteins display structural and biochemical properties similar to vertebrate orthologs , especially the human heart-type FABP [26] . They are not only involved in trafficking , storage/utilization of intracellular FA and protection of several intracellular enzymes from the detergent effects of FAs , but also in inducing antibody responses and protective immunity in the hosts [25] , [26] . Although the crystal structure of E . granulosus FABP1 has been elucidated [27] , highly limited information is available regarding structural/functional diversification and tissue specificity of FABPs in platyhelminths . Our previous ex vivo experiments with viable TsMs demonstrated that secretory TsM HLBP shuttles FAs in the surrounding host tissues and conveys them into the parasite across the biological barrier [11] , while downstream molecules responsible for the intracellular trafficking and assimilation of the transported FAs have not been elucidated . In this study , we isolated two novel genes encoding FABPs and investigated their biochemical and functional properties , which might act as the intracellular counterparts of the HLBP by mediating intracellular transportation of hydrophobic molecules . All animals used in this study were housed in accordance with guidelines from the Association for the Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) . All protocols were approved by the Institutional Review Board and conducted in the Laboratory Animal Research Center of Sungkyunkwan University ( protocol 2006-02-048 ) and Universidad Autónoma de Sinaloa , Mexico ( 2008 ) . TsMs were collected from naturally infected pig in Sinaloa state , Mexico . Intact worms were individually collected and washed with physiological saline >10 times . Cyst fluid ( CF ) was collected as previously described [11] . The whole worm , scolex and neck , and bladder wall were separately homogenized with a Teflon-pestle homogenizer in phosphate buffered saline ( PBS; 100 mM , pH 7 . 2 ) supplemented with protease inhibitor cocktail ( 1 tablet/25 ml; Complete; Roche ) . The CF and homogenates were centrifuged for 1 h at 20000 g . The supernatants were employed as crude CF and the respective extracts . Thirty fresh worms were incubated in 25 ml RPMI 1640 ( Gibco ) supplemented with the protease inhibitor cocktail for 1 h at 37°C . Addition of protease inhibitor cocktail into culture medium did not induce harmful effects on the excretory-secretory products ( ESP ) [28] . The incubation medium was harvested and centrifuged at 500 g for 10 min followed by 20000 g for 1 h . The supernatants were used as ESP . All procedures were done at 4°C unless otherwise specified . Samples were stored at −80°C until use . We previously constructed a TsM cDNA library using the lambda Uni-ZAP system and determined the nucleotide sequences of the randomly picked clones from 5′-regions with the universal T3 promoter primer [29] . We selected two clones , designated TsMFABP1 and 2 , which showed significant degrees of sequence identity with numerous FABPs during BLAST analysis of the GenBank databases at the NCBI ( http://www . ncbi . nlm . nih . gov/ ) . The TsM cDNA library was screened by polymerase chain reaction ( PCR ) using vector ( T3 and T7 promoter primers ) and TsMFABP-specific primers ( sense , 5′-GGCACGAGGATCAGATCGGGTGGTC-3′ and antisense , 5′-AGAGGGCGCTTTTGTATTTCACGTC-3′ for TsMFABP1; and sense , 5′-TAATTAACCCTCACTAAAGGGAAC-3′ and antisense , 5′-AAAAGGTGTCAAAGTGGGCTTGTTG-3′ for TsMFABP2 ) . T3 promoter primer and the antisense primers were used to amplify the 5′-regions of the respective genes . The sense primer and T7 promoter primer were employed to amplify their 3′-regions . The thermal cycler profile included preheating at 94°C ( 2 min ) , 35 cycles at 94°C ( 40 sec ) , 60°C ( 30 sec ) and 72°C ( 1 min ) with a final extension at 72°C ( 10 min ) . Amplicons were ligated into the pGEM-T Easy vector ( Promega ) and sequenced using the ABI Prism Dye Terminator Cycle Sequencing Core Kit ( Perkin Elmer ) and a Bioapply 3730 XL automated DNA sequencer ( Perkin Elmer ) . In order to increase the accuracy of nucleotide sequences , we used the high fidelity Pfu DNA polymerase ( Clontech ) during the PCR amplification and determined them from both strands of five clones . Contig cDNAs were obtained by overlapping the 5′- and 3′-region sequences . Their integrity was further confirmed by PCR using primers matched to each terminus of the contig sequences . The genomic structures were determined by amplifying each of the homologous DNAs from genomic DNAs extracted from a single worm . The genomic sequences were aligned with their corresponding mRNA sequences by considering the exon-intron boundary sequences , after which their chromosomal structures were determined . The coding profiles and homology patterns were analyzed with the ORF Finder and BLAST programs ( NCBI ) . A search for the functionally and structurally conserved protein domains was conducted using ProfileScan ( http://myhits . isb-sib . ch/cgi-in/motif_scan ) . The secondary structures were predicted by PDH software . The tertiary structures were predicted by comparative modeling method by ESyPred3D ( http://www . fundp . ac . be/sciences/biologie/urbm/bioinfo/esypred/ ) using the E . granulosus FABP1 ( Protein Data Bank id . 1O8V; 95% identity ) as a template and visualized with PyMol [30] as a template . The quality of predicted tertiary models was further evaluated by calculating template modeling score ( TM-score ) and root mean square deviation ( RMSD ) between TsMFABPs and other related proteins with the I-TASSER program ( http://zhanglab . ccmb . med . umich . edu/I-TASSER/ ) , which combined the methods of threading , ab initio modeling and structural refinement [31] . In order to retrieve the closely matched sequences from a variety of GenBank genomic databases , the deduced amino acid ( aa ) sequences of TsMFABPs were used as queries in the BLAST searches . A total of 168 sequences were selected by considering both the homology values and taxonomical distributions . Human proteins representing distinct subfamilies of iLBP were additionally retrieved from the databases . The aa sequences of two data sets were separately aligned with ClustalX and optimized using GeneDoc . The alignments were used as inputs to analyze the phylogenetic relationships among the members with MEGA program ( ver4 . 0 ) . The sequence divergences were calculated with the Jones-Taylor-Thornton ( JTT ) substitution model and indels between pairs of sequences were regarded as missing data . The phylogenetic trees were constructed by the neighbor-joining algorithm . The statistical significance of each branching node was evaluated employing 1000 random samplings of the input alignments by the SEQBOOT program . The cDNAs corresponding to the predicted ORF region of TsMFABPs were PCR-amplified with specific primers containing cleavage sites for restriction enzymes ( underlined ) of BamHI and XhoI ( TsMFABP1 , 5′-CGCGGATCCATGGAGCCATTCATCG-3′ and 5′-TGACTCGAGTTACGCTGCCTTAAC-3′; TsMFABP2 , 5′-GCGGATCCATGACCTCAAGTGAG-3′ and 5′-CACTCGAGTCAGCTCTTCTGCCG-3′ ) and directionally cloned into the pET-28a expression vector ( Novagen ) following enzyme digestion . The plasmids were transformed into Escherichia coli BL21 ( DE3 ) ( Novagen ) . Each of single colonies containing the insert was used to initiate a liquid culture and expression of recombinant protein was induced by 0 . 5 mM isopropyl β-D-1-thiogalactopyranoside ( IPTG ) . The recombinant proteins were purified from bacterial lysates by Ni-NTA affinity chromatography using a HiTrap chelating Sepharose column ( Amersham Biosciences ) . The recombinant proteins were monitored by 15% reducing SDS-PAGE with Coomassie Brilliant Blue ( CBB ) G-250 staining . Polyclonal antisera against each recombinant protein were raised in specific pathogen-free , 6-week-old female BALB/c mice by consecutive subcutaneous inoculation of the respective proteins ( 30 µg ) in Freund's adjuvant at 2-week intervals . A final booster was done by intravenous injection of 10 µg/100 µl PBS without adjuvant through tail vein . One week later , blood was collected by cardiac puncture . The immune sera were obtained by centrifugation at 3000 g for 10 min . IgG fractions were isolated using a Protein G affinity chromatography column ( Amersham Biosciences ) and stored at −80°C until required . TsM whole worm extracts were fractionated by a Superdex 75 prep grade ( HiLoad , 16×60 cm-long ) molecular sieve fast protein liquid chromatography ( FPLC ) system ( AKTA; Amersham Biosciences ) , which was equilibrated with Tris-HCl ( 20 mM , pH 8 . 0 ) containing 150 mM NaCl . The extracts ( 10 mg proteins/3 ml ) were applied to the column ( flow rate; 0 . 5 ml/min ) and 85 fractions ( each 1 . 5 ml aliquot ) were allocated according to their absorbance at 280 nm monitored by UNICORN ( ver3 . 0 ) . Fractions showing high lipid-binding activity with concomitant positive reactions against anti-rTsMFABPs were pooled , dialyzed against Tris-HCl ( 20 mM , pH 9 . 0 ) and concentrated . Ion-exchange chromatography was further conducted on a 2×10 cm-long DEAE-Sepharose column ( Amersham Biosciences ) equilibrated with Tris-HCl ( 20 mM , pH 9 . 0 ) . Elution was done with a step-wise gradient NaCl concentrations ( 0 , 20 , 40 , 60 , 80 and 100 mM ) with the same buffer . Active fractions identified as above were dialyzed against 20 mM Tris-HCl ( pH 8 . 0 ) containing 150 mM NaCl , concentrated and stored at −80°C until use . Respective TsM extracts ( 10 µg ) and rTsMFABPs ( 100 ng ) were resolved by 15% SDS-PAGE under reducing conditions . For 2-DE , partially purified TsMFABPs ( 10 µg ) were mixed with rehydration buffer ( 6 M urea , 2 M thiourea , 2% CHAPS , 0 . 4% dithiothreitol [DTT] , 0 . 5% IPG buffer and 0 . 002% bromophenol blue [BPB] ) , loaded on IPG strips ( pH 6–11 ) with a cup-loading instrument ( IPGphor; Amersham Biosciences ) and focused for a total of 35 kVh . Second-dimension SDS-PAGE was done by 15% gels ( 160×160×1 mm ) . The separated proteins were visualized with CBB G-250 or transferred to nitrocellulose ( NC ) membranes ( Schleicher & Schuell ) . The membranes were blocked for 1 h with Tris buffered saline ( 100 mM , pH 8 . 0 ) containing 0 . 05% Tween 20 and 5% skim milk ( blocking buffer ) , after which they were incubated overnight with specific mouse antibodies ( 1∶2000 dilutions ) in blocking buffer . The membranes were incubated with a 1∶4000 dilution of horseradish peroxidase ( HRP ) -conjugated goat anti-mouse IgG ( Cappel ) for an additional 1 h . The reactions were developed with an enhanced chemiluminescence ( ECL ) kit ( Pierce ) . For quantitative analysis , all the immunoblot images were developed after 1 min exposure . All the proteins were delipidated for 2 h using Sephadex-LH ( Sigma-Aldrich ) prior to assay . The ligand binding profile of the native and rTsMFABPs were detected spectrofluorometrically using fluorescent FA analogs , including 11- ( [5-dimethylaminonaphthalene-1-sulfonyl]amino ) undecannoic acid ( DAUDA ) , dansyl-DL-α-aminocaprylic acid ( DACA ) ( Molecular Probes ) , retinol ( Sigma-Aldrich ) and naturally fluorescent cis-parinaric acid ( cPnA; Molecular Probes ) . Fluorescence emission spectra were recorded at 25°C with a total volume of 200 µl per well using black 96-well Microfluor 1 plates and an Infinite M-200 automated multi-detector ( Tecan ) . The emission and excitation wavelengths for DAUDA , DACA , retinol and cPnA were 519 , 519 , 325 and 420 nm , and 345 , 350 , 350 and 315 nm , respectively . We included the TsM 120-kDa protein ( 10 µM ) and recombinant 18 kDa ( 5 µM; a subunit of the TsM 120-kDa protein ) , which were proven not to have FA-binding activity [32] , as negative controls during the measurements . All fluorescent stock compounds ( 10 mM dissolved in ethanol ) were stored at −20°C in a dark room and were freshly diluted in ethanol prior to use . The equilibrium dissociation constants ( Kd ) of the proteins bound to DAUDA , retinol and cPnA were estimated by adding increasing concentrations of respective ligands ( 0 . 1–10 µM for FA analogs and 0 . 1–10 mM for retinol ) in a micro-quartz plate . Fluorescence intensities were normalized to the peak fluorescence intensity and corrected for background fluorescence of the ligand alone at each concentration . Corrected data were analyzed using the one-site saturation model and best fit algorithm contained in SigmaPlot9 software ( y = VmaxX/Km+X , where y is relative fluorescence and X is concentrations of lipid ligand . Vmax can be substituted as Fmax [maximum fluorescence] ) . Competition assays were carried out by monitoring the change in fluorescence intensity at the peak transmission wavelength measured for either rTsMFABPs:DAUDA , rTsMFABPs:retinol or rTsMFABPs:cPnA complex in the presence of 10-fold excess oleic acid . Fresh TsMs were evaginated in the presence of 1% bile salts ( Sigma-Aldrich ) in RPMI 1640 ( pH 7 . 2 ) at 37°C overnight . The worms were fixed in 4% paraformaldehyde in PBS ( 50 mM , pH 7 . 4 ) at 4°C , dehydrated with a graded alcohol and embedded in paraffin . Sections 4 µm in thickness were cut , deparaffinized and rehydrated . A stock solution was prepared by dissolving Nile red ( 9-diethylamino-5H-benzo[α]phenoxazine-5-one , 100 µg/ml; Sigma-Aldrich ) in acetone and stored at −20°C in the dark until use . The stock solution ( 10 µl ) was diluted in 70% glycerol ( 10 ml ) just prior to use . A drop of diluted Nile red solution was placed on the fresh TsM sections for 1 h at 4°C . The slides were mounted on Paramount Aqueous mounting medium ( DAKOCytomation ) and observed using a LSM510 Meta DuoScan confocal microscope ( Carl Zeiss ) . The locality of retinol ( vitamin A ) was observed on the 10 µm-thick cryosectioned TsM sections under an Axioplot light/fluorescent microscope ( excitation filter BP365/12 , barrier filter BP495/40; Carl Zeiss ) [33] . Since treatment of worm sections with organic solvent removed retinol and the biochemical was quickly oxidized when exposed to the air , unfixed and unstained frozen sections were observed immediately after mounting . The tissue distribution of TsMFABPs was determined on evaginated worm sections using the respective antibodies . Worm sections ( 4 µm-thick ) were treated with 3% hydrogen peroxide for 5 min and blocked with PBS supplemented with 3% bovine serum albumin ( BSA ) and 0 . 05% Tween 20 ( PBS/T-BSA ) for 1 h . The sections were incubated with the respective antibodies ( 1∶200 dilutions in PBS/T-BSA ) overnight at 4°C . For fluorescent staining , rhodamine-conjugated goat anti-mouse IgG antibody ( Jackson ) was incubated for 1 h at 4°C . The slides were counterstained with 4′ , 6-diamidino-2-phenolindole ( DAPI , 10 µg/ml; Invitrogen ) for 5 min at 4°C in dark and observed under an Axioplot light/fluorescent microscope ( Carl Zeiss ) . Pre-immune mouse serum diluted to the same ratio was employed as a control . In situ hybridization was conducted using fluorescent Cy5-labeled probes ( TsMFABP1 anti-sense , 5′-CGCTGCCTTAACGTAGGTTCGCACGC-3′ and sense , 5′-GCGTGCGAACCTACGTTAAGGCAGCG-3′; TsMFABP2 antisense 5′-GCTCTTCTGCCGACGGTACATGTGCAC-3′and sense 5′-GTGCACATGTACCGTCGGCAGAAGAGC-3′ ) . The worm cryosections mounted on superfrost PLUS slides ( Sigma-Aldrich ) were rehydrated in 10% formamide and 2× SSC for 5 min , followed by treatment with proteinase K . The hybridization reactions were performed in hybridization solution ( 100 µl ) for 16 h at 55°C . The slides were then washed with washing buffer ( 20% formamide in 2× SSC ) 2 times for 30 min at 30°C . Nuclear staining was done by adding DAPI ( Invitrogen ) to the wash solution during the second wash . The slides were mounted with freshly prepared oxygen depleted mounting media . The signals were observed using a LSM510 Meta DuoScan confocal fluorescence microscope ( Carl Zeiss ) . Similarity analyses of TsM expressed sequence tag clones against the GenBank database and following cDNA library screening led to the identification of two full-length cDNAs , which displayed high structural similarity with the other known FABPs . The TsM genes , designated TsMFABP1 and TsMFABP2 , encoded an ORF for 133- and 136-aa polypeptide with predicted molecular masses of 14 . 3 and 14 . 8 kDa and isoelectric point ( pI ) values of 8 . 6 and 8 . 4 , respectively . The coding regions shared 48% and 45% identity with each other at the nucleotide and aa levels , respectively . The initial BLASTX searches with the TsMFABP sequences at the NCBI retrieved several hundred FABPs isolated from diverse organisms . They showed the highest matches to those of cestode parasites including E . granulosus and M . vogae ( identity >53% and E-value <3e-28 for TsMFABP1 , identity >42% and E-value <9e-18 for TsMFABP2 ) . Homology searches by the Hidden Markov models revealed the results similar to those with BLAST algorithms ( data not shown ) . The primary structures of TsMFABPs were compared with those of some cestode and human orthologs . As shown in Figure 1 , these molecules revealed variable degrees of sequence identity from 44%–95% , but tightly conserved several signatures and motifs representative of the FABP family . Motifs 1 , 2 , and 3 , spanned the βA-α1 ( 23 aa ) , βE ( 17 aa ) and βI–βJ ( 22 aa ) domains ( blue boxes ) . Nuclear localization signal with three basic aa residues was positioned at K18/R9 , R30/21 and K31/22 , and its regulation site was found at F58/62 , respectively ( red and dotted red boxes ) . Nuclear export signal was observed at L60/62 , V82/L82 and M92/L92 ( green boxes ) . Hormone-sensitive lipase binding site was recognized at K18/R9 ( blue arrow ) ( positions of respective aa residue denote each for TsMFABP1 and 2 ) . The GXW triplet , which is shared by the members of calycin superfamily , was found in the motif 1 ( orange box ) , but the TDY triplet found in the lipocalin family was not detected in the motif 2 of TsMFABPs and related proteins . Interestingly , TsMFABP2 contained two aa insertions between βB and βC ( 4 aa , BC loop ) , and between βF and βG ( 6 aa , FG loop ) . In addition , TsMFABP1 conserved a single site for protein kinase C and casein kinase II phosphorylation , while TsMFABP2 harbored three sites targeted for the casein kinase II phosphoylation ( purple boxes ) . The tertiary structures of TsMFABPs were readily simulated using the E . granulosus FABP1 ( Protein Data Bank id . 1o8vA ) as a template during homology-based modeling . The models were highlighted by the basic β-barrel composed of 10 anti-parallel β-strands ( βA–βJ ) and N-terminal helix-turn-helix motif ( α1 and α2 ) ( Figure S1 ) . The extra loops detected in TsMFABP2 were placed near the bottom of the barrel ( pinkish boxes ) . A similar structure for TsMFABP1 protein was predicted by different threading templates such as 1o8vA , 3rswA and 1hmsA by I-TASSER program ( confidence score 1 . 43 , TM-score 0 . 91±0 . 06 , RMSD 1 . 8±1 . 5 Å ) . The I-TASSER result with TsMFABP2 sequence was similar to that of TsMFABP1 , while the quality of predicted model seemed to be less significant , due probably to the extra BC and FG loops ( confidence score 0 . 50 , TM-score 0 . 78±0 . 10 , RMSD 3 . 6±2 . 5 Å ) . We deposited nucleotide sequence data under the accession numbers HQ259679 ( TsMFABP1 ) and HQ259680 ( TsMFABP2 ) in the GenBank database . A phylogenetic tree constructed with the aa sequences of 168 TsMFABP-related proteins demonstrated different clustering patterns between the protostomian and deuterostomian FABPs ( Figure S2 ) . The proteins isolated from the invertebrates were closely allocated to one another according to the taxonomical positions of their donor organisms , whereas those from higher animals appeared to be split into several monophyletic sub-clades containing each of the iLBP families , regardless of their donor sources . The relative phylogenetic positions of TsMFABPs were further examined against diverse human iLBP members ( Figure 1B ) . A neighbor-joining tree placed these platyhelminth proteins between the human myelin-adipocyte-heart FABP and the CRBP/CRABP subfamilies , suggesting that the platyhelminth proteins have not yet been differentiated into each of the subfamily lineages . The TsMFABP1 was interconnected to other cestode proteins by an internal node ( red arrow in Figure 1B ) , while TsMFABP2 comprised a single external node . The trematode proteins formed a clade separated from that of cestode homologs . The trees constructed using the maximum-likelihood ( TREE_PUZZLE ) and maximum-parsimony ( PHYLIP ) algorithms also showed a tree topology similar to that of neighbor-joining method ( data not shown ) . The genomic structure of TsMFABP genes was determined employing the genomic DNA extracted from a single worm . The genomic sequences of TsMFABPs contained a single intron of 84-bp ( TsMFABP1 ) or 3010-bp ( TsMFABP2 ) near the 3′-end of the respective ORFs . The intron was located prior to the first nucleotide of a codon ( phase 0 ) within both TsM genes ( Figure 1C ) . The intron appeared to be orthologous among the related genes used in the phylogenetic analysis , except for the M . vogae ( MvFABPs ) and out-group gene ( HsRBP ) , despite the great length polymorphism ( red vertical line with a red arrow , Figure 1A ) . This result suggested that the paralogous TsMFABP1 and 2 genes have arisen by duplication of an ancestral gene at least before divergence of cestode species . The bacterially expressed recombinant proteins were purified by Ni-NTA affinity chromatography . The rTsMFABPs migrated to approximately 18 kDa , which were slightly larger ( 3 kDa ) than that predicted by the aa sequences , due to the additional N-terminal tag ( Figure S3A ) . We also partially purified the native TsMFABPs through gel filtration followed by DEAE anion-exchange chromatography . TsMFABP1 and 2 were eluted at flow-through and 20 mM fractions , respectively . When these proteins were analyzed by 2-DE and subsequent immunoblotting probed with each of the specific antibodies , a single immunoreactive signal at 15 kDa and a pI value of ca . 9 . 2 ( TsMFABP1 ) or 8 . 4 ( TsMFABP2 ) was detected ( Figure S3B ) . The partially purified native and recombinant proteins were subjected to delipidation . Each of the proteins ( 1 µM ) was used in a hydrophobic ligand binding assay against the polarity-sensitive fluorophore-tagged FA analogs ( 0 . 1 µM ) and retinol ( 5 mM ) . The fluorescence emission of DAUDA was significantly increased with a blueshift from 550 nm to 500 nm , when mixed with the native or rTsMFABP1 ( Figures 2A and 2B ) indicating the engagement of fluorophore into a highly non-polar DAUDA binding site . The interactive binding was competitively inhibited by oleic acid in a dose-dependent manner ( Figure 2B , part of data not shown ) . The TsMFABP2 also bound to DAUDA , although its specific activity was lower than that of the TsMFABP1 . Both of the TsM proteins exhibited binding affinity toward retinol . Interestingly , the relative activities were reversed when retinol was provided as the hydrophobic ligand ( Figures 2C and 2D ) . The retinol-binding activity of rTsMFABP2 appeared to be higher than that of rTsMFABP1 . Other fluorescent FA analogs such as DACA and cPnA showed interaction modes comparable to those with DAUDA against the rTsMFABPs ( data not shown ) . No binding activity was detected in the reactions with the TsM 120-kDa and recombinant 18-kDa proteins , which were used as negative controls . The steady-state kinetics of binding reactions assayed using rTsMFABPs demonstrated saturation behavior in accordance with the increasing concentrations of DAUDA and cPnA ( 0 . 1–10 µM ) , and retinol ( 0 . 1–10 mM ) . The dissociation constants ( Kd ) of rTsMFABP1 were calculated to be 2 . 15 µM , 0 . 28 µM and 1 . 78 mM for DAUDA , cPnA and retinol , respectively , whereas the equivalent values for rTsMFABP2 were determined to be 9 . 40 µM , 0 . 64 µM and 0 . 98 mM , respectively . The binding rate constants ( Vmax/Kd ) against each of the hydrophobic ligands were also highly distinguishable between the rTsMFABP1 and rTsMFABP2: 139 . 3 versus 38 . 7 for DAUDA , 2172 . 9 versus 483 . 0 for cPnA and 172 . 1 versus 211 . 3 for retinol ( Table 1 ) . We examined tissue expression pattern of TsMFABPs employing soluble TsM proteins extracted from different anatomical compartments . TsMFABPs were expressed in the TsM parenchyme including scolex and neck , and bladder wall , although that of TsMFABP2 appeared relatively low in the bladder wall . Interestingly , the CF and ESP proteins , where the excretory-secretory proteins accumulate , reacted with the antiserum specific to rTsMFABP1 , but not with that against rTsMFABP2 . The same blot probed with preimmune mouse serum did not exhibit any response ( Figure 3A ) . The histological distribution of TsMFABPs was further examined on the TsM sections by immunohistochemical staining . Figure 3B ( panel a ) presents an evaginated worm section stained with hematoxylin-eosin , in which characteristic tissues/organs of TsM including scolex , neck , spiral canal , loose tissue and bladder wall were observed . These two proteins exhibited principally similar anatomical distribution in the worm section , but some variable pattern was also recognized . Anti-rTsMFABP1 antibody mainly reacted with protein ( s ) scattered in the bladder wall and spiral canal . The signal appeared to be prominent in the subtegumental nuclear layer and the spherical cell body-like compartments scattered through the fibrillar stroma of the bladder wall ( panel c ) . The antibody revealed a similar reaction pattern in the spiral canal . In the neck , the reaction intensity was relatively weak and was largely restricted in the nuclear layer zone ( panel d ) . In contrast , protein ( s ) in CF showed fairly weak positive reactions , which suggested that small amount of TsMFABP1 are secreted into surrounding environments . The scolex did not exhibit any detectable reaction ( panel e ) . The TsMFABP2 was intensely localized in the subtegumental regions and relatively less in the stroma beneath the subtegumental nuclear layer of the neck and spiral canal ( panel h ) . The bladder wall revealed weak positive reactions ( panel g ) , while CF and scolex did not show any detectable signal ( panels f and i ) . The expression patterns observed at the protein levels matched well with the results obtained by in situ hybridization , in which each transcript was stained with Cy5-labeled , gene-specific antisense probes on the TsM cryosections ( Figure 4 ) . Both transcripts created high signals at spiral canal region . Lipid molecules are shuttled by HLBPs/FABPs during the intracellular trafficking . We analyzed the tissue distribution of lipid droplets and retinol in the worm sections . As shown in Figure 5 ( panels a–d ) , the lipid droplets stained with Nile red were primarily distributed within the bladder wall and spiral canal in a scattered fashion , and less in the subtegumental regions of the neck and scolex ( yellow arrows ) . The distribution density of the droplets was found to be irregular across the bladder wall . Strong signals were detected in the outer surface regions of the bladder wall membrane ( white arrows ) , whereas the inner regions were faintly stained with the hydrophobic dye . Nile red was also stained with some hydrophobic droplets/molecules in the region filled with CF , which has been suggested to act as a reservoir for lipid molecules taken from host environments [11] . Retinol , when exposed and excited with ultraviolet light , emits a natural , green fluorescence and faded away within 20 sec [34] . Retinol was largely restricted in the outer membranous region of the spiral canal and bladder wall compared to those of Nile red-positive molecules ( Figure 5 , panels f and g ) . Retinol was also detected in CF ( panel e ) , although no significant signal could be observed by the anti-rTsMFABP2 antibody ( Figure 5 , panel f ) . Hooklets showed non-specific epifluorescence ( panel h ) HQ259679 ( TsMFABP1 ) and HQ259680 ( TsMFABP2 ) .	Neurocysticercosis ( NC ) , an infection of the central nervous system with Taenia solium metacestode ( TsM ) , constitutes a leading cause of adult-onset seizures in endemic areas . Like other helminths , TsM is incapable of synthesizing lipid molecules . It should be equipped with a specialized system for lipid transportation from the host to ensure its long-survival . Such a transport system may be a target for function-associated drug design . We characterized two novel fatty-acid ( FA ) -binding TsM proteins ( TsMFABP1 and TsMFABP2 ) . Native and recombinant proteins bound to several FA analogs and retinol at micromolar and millimolar concentrations . Their binding was specifically inhibited by oleic acid . TsMFABP1exhibited high affinity toward FA analogs , while TsMFABP2 showed preferential affinity to retinol . Both TsMFABPs were predominantly expressed in the canal region of the worm , where lipids and retinol were abundantly distributed . The two paralogous TsMFABPs have undergone ( or are still undergoing ) structural diversification and following functional divergence to act as FABP or retinol binding protein , similar to the intracellular lipid binding proteins of deuterostomian animals . The canal region specific distribution of lipids , retinol and FABPs further suggested that cells in this area might differentiate into diverse cells to compose huge numbers of the proglottids , thereby playing vital roles in the parasite growth and development .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "medicine" ]	2012	Structural and Binding Properties of Two Paralogous Fatty Acid Binding Proteins of Taenia solium Metacestode
In 2010 , WHO recommended the use of new short-course treatment regimens in kala-azar elimination efforts for the Indian subcontinent . Although phase 3 studies have shown excellent results , there remains a lack of evidence on a wider treatment population and the safety and effectiveness of these regimens under field conditions . This was an open label , prospective , non-randomized , non-comparative , multi-centric trial conducted within public health facilities in two highly endemic districts and a specialist referral centre in Bihar , India . Three treatment regimens were tested: single dose AmBisome ( SDA ) , concomitant miltefosine and paromomycin ( Milt+PM ) , and concomitant AmBisome and miltefosine ( AmB+Milt ) . Patients with complicated disease or significant co-morbidities were treated in the SDA arm . Sample sizes were set at a minimum of 300 per arm , taking into account inter-site variation and an estimated failure risk of 5% with 5% precision . Outcomes of drug effectiveness and safety were measured at 6 months . The trial was prospectively registered with the Clinical Trials Registry India: CTRI/2012/08/002891 . Out of 1 , 761 patients recruited , 50 . 6% ( n = 891 ) received SDA , 20 . 3% ( n = 358 ) AmB+Milt and 29 . 1% ( n = 512 ) Milt+PM . In the ITT analysis , the final cure rates were SDA 91 . 4% ( 95% CI 89 . 3–93 . 1 ) , AmB+Milt 88 . 8% ( 95% CI 85 . 1–91 . 9 ) and Milt+PM 96 . 9% ( 95% CI 95 . 0–98 . 2 ) . In the complete case analysis , cure rates were SDA 95 . 5% ( 95% CI 93 . 9–96 . 8 ) , AmB+Milt 95 . 5% ( 95% CI 92 . 7–97 . 5 ) and Milt+PM 99 . 6% ( 95% CI 98 . 6–99 . 9 ) . All three regimens were safe , with 5 severe adverse events in the SDA arm , two of which were considered to be drug related . All regimens showed acceptable outcomes and safety profiles in a range of patients under field conditions . Phase IV field-based studies , although extremely rare for neglected tropical diseases , are good practice and an important step in validating the results of more restrictive hospital-based studies before widespread implementation , and in this case contributed to national level policy change in India . Clinical trial is registered at Clinical trial registry of India ( CTRI/2012/08/002891 , Registered on 16/08/2012 , Trial Registered Prospectively ) . Visceral leishmaniasis ( VL , also known as kala-azar ) is an ultimately fatal disease with 10 , 311 reported cases in the Indian subcontinent in 2014 [1] , although under-reporting means that the real number is likely to be higher [2] . The number of reported cases in India has progressively declined in recent years from 33 , 187 in 2011 to 6245 in 2016 , an approximate annual reduction of 30–35% [3]; this may be due to a number of factors , including the VL elimination initiative in South-East Asia , the natural incidence cycles of the disease , and improvements in social conditions . Early and effective treatment is one of the pillars of the VL elimination strategy . Historically , a number of drugs have been used in India in monotherapy , including pentavalent antimonials , amphotericin B deoxycholate , miltefosine , paromomycin , and liposomal or lipid formulations of amphotericin B [4 , 5] . Pentavalent antimonials , the only available treatment for VL for decades , are no longer recommended in the most endemic state of Bihar due to development of resistance , with treatment failure reaching more than 60% in some villages [6] . Miltefosine was introduced into the national program as an orally administered 28-day monotherapy in 2005 , with very satisfactory cure rates . However , its efficacy decreased from 96% to 90% within a decade of use in India [7 , 8] , with higher reported failure rates in children , likely to be related to the inappropriate linear dosage which was used [9] . In Nepal , a 10% failure rate for miltefosine at 6 months doubled to 20% at 12 months follow-up . With limited drugs available , there was a need to preserve the existing drugs and to develop shorter and safer treatment regimens [10] . Amphotericin B deoxycholate is a highly efficacious drug with a cure rate of 97% , but requires in-patient treatment for up to a month , which , coupled with infusion and drug-related adverse effects , has limited its utility [11] . AmBisome ( Gilead Pharmaceuticals , Foster City , CA , USA ) is a brand name for liposomal amphotericin B ( AmB ) . It has been studied extensively at a range of doses and shows excellent safety and efficacy . In a study carried out by Sundar et al . , a single 10 mg/kg dose of AmB had 95 . 7% efficacy and was safer than conventional amphotericin B deoxycholate [11] . An earlier phase III non-inferiority clinical trial in India comparing conventional amphotericin B deoxycholate with three different low-dose combinations ( AmBisome 5 mg/kg plus 7 days of miltefosine; AmBisome 5 mg/kg plus 10 days of paromomycin; miltefosine plus paromomycin both for 10 days ) found all three to be non-inferior with final cure rates of ≥97% at 6 months [12] . In 2010 , the WHO recommended these combination regimens along with a single dose of 10mg/kg AmBisome ( known as Single Dose AmBisome/SDA ) as first line treatments in South Asia [13] based on economic , safety , and efficacy considerations [14–16] . However , these hospital-based studies were restricted in sample size , conducted under very controlled conditions , and mostly excluded unwell or patients from more vulnerable groups ( e . g . pregnant women or the very young/old ) . As such , the Drugs for Neglected Diseases initiative ( DNDi ) , in collaboration with Rajendra Memorial Research Institute of Medical Science ( RMRIMS ) , State Health Society Bihar , and Médecins Sans Frontières ( MSF ) , conducted this field effectiveness study to better determine the safety and feasibility of these treatment regimens under field conditions within public healthcare facilities in Bihar , India . The protocol was approved by the Institutional Ethics Committee of RMRI Patna , Ethics Review Board of Mèdecins Sans Frontières , London School of Hygiene & Tropical Medicine ( Ref 6046 ) , Indian Council of Medical Research , Drug Controller General of India and National Vector Borne Disease Control Programme . Written informed consent was obtained by a treating physician . For children , consent of parents or of a legal representative was obtained . This study was an open label , prospective , non-randomized , non-comparative multicenter phase IV clinical trial conducted through government hospitals and primary health clinics ( PHCs ) in Bihar state , India . The study was conducted from August 2012 to September 2015 in two districts ( Vaishali and Saran ) and at the Rajendra Memorial Research Institute of Medical Sciences ( RMRIMS ) , a government research institute specializing in VL located in Patna . All patients meeting a case definition of VL defined as fever for more than 2 weeks , splenomegaly , and confirmed with a positive rK-39 rapid diagnostic test ( InBios , USA ) were included in the study . Relapse cases with a confirmatory parasitological diagnosis were also eligible . Patients with concurrent PKDL , HIV and those reporting a history of hypersensitivity to the investigational drugs were excluded . Upon confirmation of VL , written informed consent was obtained by a treating physician . For children , consent of parents or of a legal representative was obtained . Prior to treatment , blood was taken for haemoglobin , alanine aminotransferase ( ALT ) , aspartate aminostransferase ( AST ) , and serum creatinine . Other tests were performed when medically indicated . For women aged 12–55 years , a urinary pregnancy test was also conducted , with all pregnant women being referred for SDA treatment . Due to the teratogenicity of miltefosine , women with child-bearing potential unwilling to use long-acting injectable contraception during and for three months after treatment were also referred for SDA treatment . Height and weight were measured for all patients at admission . Anthropometric indicators appropriate for patient age were calculated using the latest World Health Organization ( WHO ) Multicentre Growth Reference [17] . Severe wasting was defined based on WHO criteria ( weight for height Z-score <-3 for children <5 years; BMI-for-age Z-score <-3 for those 5–19 years; and BMI <16 . 0 for adults ) . Severe anaemia was defined as haemoglobin <7 g/dL for children <5 years; <8 g/dL for 5 years and older; moderate anaemia defined as <11 g/dL but above the cut-off for severe anaemia [18] . Patients with haemoglobin <4 g/dl , serious concomitant infection ( e . g . severe pneumonia ) , complicated severe malnutrition , TB/VL co-infection , or children <2 years of age were referred to the MSF VL treatment unit within Hajipur district hospital or RMRIMS for further specialist management . These patients were treated with SDA as per physician decision and included in the study . The three regimens evaluated were: a 10 mg/kg single intravenous dose of AmBisome ( SDA ) ; a 5 mg/kg single intravenous dose of AmBisome plus 7 days of linear dosage oral miltefosine ( AmB+Milt ) ; and 11 mg/kg intramuscular base paromomycin plus linear dosage oral miltefosine for 10 days ( Milt+PM ) . Linear dosage of miltefosine was 2 doses of 50 mg ( morning and evening ) for patients ≥12 years weighing more than 25 kg , or a single morning dose of 50 mg for those weighing less than 25 kg . Children of 2–11 years were given miltefosine at a dose of 2 . 5 mg/kg/day orally divided into two daily doses . The SDA regimen was administered in 5% dextrose over approximately 2 hours after completion of a test dose of 1 mg to check for hypersensitivity over 30 minutes; patients were discharged the following day from the district hospital where clinical conditions allowed a safe return home . The AmB+Milt regimen consisted of AmBisome 5 mg/kg , administered as above on day 1 , with oral miltefosine on days 2 to 8 to be taken at home with advice to return in case of any adverse event . The Milt+PM regimen consisted of the oral miltefosine dose plus intramuscular paromomycin ( 11 mg/kg/day in a single daily dose ) given concomitantly daily for 10 days . Patients treated at the district hospital were admitted to a VL ward for the 10 days of treatment , whereas patients enrolled at PHC level were managed on an outpatient basis , returning each day for the injection . Patients who failed to return for ambulatory treatment were actively traced by telephone and , if necessary , in person to ensure maximum compliance . Following national regulatory recommendation as part of the study approval process , children were only treated at the district hospitals under the supervision of a paediatrician . At the specialist RMRI facility , all three modalities were used , based on clinician decision . Patients that relapsed with any of the three treatment regimens were given rescue treatment as per physician decision . Patients were asked to return for two post-treatment follow-up visits . The first was scheduled 7–20 days after treatment onset to assess initial cure . A second follow-up visit was planned at 6 months ( with a 5–10 month window period ) after treatment onset , to assess final cure . Patients were actively traced if they did not attend follow-up visits . Treatment stopped was defined as treatment stopped early by the attending clinician for any reason . Default was defined as failure to finish treatment against medical advice . Relapse was defined as recurrence of clinical symptoms and visualization of parasites in spleen or bone marrow aspirate before the 6 month follow up period . Death was reported if it occurred from any cause up to 6-months post-treatment . Lost to follow-up was defined as a patient who was unable to be traced at the 6 months follow up window . For effectiveness analyses , the primary outcome was final cure defined as a negative test of cure at the end of treatment , absence of clinical signs and symptoms of VL and no relapse up to 6 months follow-up . In the ITT analysis , the final cure rate for SDA was 91 . 4% ( 95% CI 89 . 3–93 . 1 ) , AmB+Milt 88 . 8% ( 95% CI 85 . 1–91 . 9 ) , and Milt+PM 96 . 9% ( 95% CI 95 . 0–98 . 2 ) . In the complete case analysis , cure rates were SDA 95 . 5% ( 95% CI 93 . 9–96 . 8 ) , AmB+Milt 95 . 5% ( 95% CI 92 . 7–97 . 5 ) and Milt+PM 99 . 6% ( 95% CI 98 . 6–99 . 9 ) ( Table 2 ) . Relapse rates varied by drug regimen and were higher for children than for those older than 12 years ( Tables 3 and 4 ) . Those with illness that had lasted 8 weeks or less were also more likely to have relapse at 6 months . Serious adverse events were infrequent in all study arms . There were 5 serious adverse events ( SAE ) in the SDA arm . Anaphylactic reaction occurred during treatment in one patient and was related to the study drug AmBisome . There were four SAEs after completion of treatment and discharge from hospital: one asymptomatic atrial ectopic possibly related to AmBisome and three other SAEs that were unrelated to the study drugs: TB empyema , hospitalization due to dehydration and elevated creatinine , and lower respiratory tract infection . All SAEs resolved completely with no sequela . The most common adverse events were gastrointestinal ( nausea , vomiting , diarrhoea , abdominal pain ) and back pain ( Table 5 ) . Adverse events leading to treatment interruption were rare ( <1% for all three drug regimens ) . Previous phase-3 randomized controlled trials have shown these regimens were non-inferior to treatment with standard amphotericin B deoxycholate with ITT final cure of 93·0% for SDA ( 95% CI 87 . 5–96 . 3 ) , AmB+Milt 97·5% ( 95% CI 93 . 3–99 . 2 ) , and Milt+PM 98·7% ( 95% CI 95 . 1–99 . 8 ) [12] . Cure rates by ITT in this study , while not quite as high , still achieved acceptable levels with the differences largely due to loss to follow-up . Earlier DNDi conducted a phase-3 clinical trial in Bangladesh to assess safety and efficacy of short course combination regimens in field conditions at Upazila level that provided excellent efficacy outcome ( ≥95% ) and very good safety profile [19] . The demographic characteristics of the population enrolled in the study correspond to those defined for this area , roughly 70% patients older than 12 years , 30% of them women of child-bearing age . The effectiveness in complete case analysis in this study was slightly higher in the Milt+PM arm than in the other two arms , which may partly be due to the higher number of clinically unwell patients being allocated to the SDA treatment arm . Despite a range of clinical severity , presentations , and patient demographics , all of the treatments showed excellent safety profiles . This study was non-comparative , both SDA and combination of Milt+PM had satisfactory effectiveness of >90% which corroborates the decision by the Indian control program to use these treatments in the elimination program . No complications were seen in pregnant ( treated with SDA ) or extremely young patients . Generally , the treatments were easily prepared and administered by health care providers , and appeared to be well accepted by patients . When elimination was first envisioned , oral miltefosine was proposed to be used primarily in the attack phase due to its acceptability . In parallel to the provisional results of this study , the WHO included India in the AmBisome donation programme , resulting in India adopting these new treatment modalities within the national elimination programme , replacing miltefosine monotherapy . To date , this has proved to be a very effective strategy , with over 12 , 000 patients having been treated in the attack phase with SDA within the public health sector across the Indian subcontinent with excellent safety and efficacy [14] . Although widely implemented , SDA is not without its limitations–complex storage and preparatory requirements mean that its safe use is contingent on logistical support that is not required for Milt+PM , for example . The unintended consequence of this has been the neglect of alternative drug combinations , which has resulted in a lack of stock and awareness of these regimens in the national programme . Considering the limited number of therapeutic options available , it is critical to ensure that procurement and availability of all three drugs is ensured within the elimination framework . Currently , all three WHO supported formulations of these drugs are produced by single source manufacturers AmBisome ( Gilead Sci . , USA ) , miltefosine ( Knight Therapeutic Inc . , Canada ) , and paromomycin sulphate ( Gland Pharma , India ) [16] , making the supply chain sensitive to factory and quality issues should they arise . This reflects the urgent need for investment in bio-equivalence studies , technology transfer , and alternative production , which may potentially need to be centralized and pooled to ensure adequate market conditions . Moreover , all these limitations justify strengthening the development of new chemical entities ( NCEs ) that are needed in the form of short-course oral combinations , to replace the existing drugs in the Indian subcontinent and worldwide [20] . Although resistance to amphotericin B has yet to be demonstrated in vivo despite decades of use , prolonged use of monotherapies such as miltefosine and paromomycin have resulted in reduced drug susceptibility , and potential mechanisms of amphotericin B resistance have been described [21] . Reduced drug susceptibility for SSG and Milt were only determined well after they had progressed to unacceptable levels; as such it is critical that the national programme develops sentinel surveillance for drug susceptibility monitoring of VL drugs so that early signals can be generated that can guide more rational use of existing therapeutic options . Such initiatives are underway in India [22] but are yet to be developed in Bangladesh or Nepal . There are also a number of challenges that need to be considered for the Milt+PM and AmB+Milt regimens . Although compliance was very high in this study ( >99% ) , this was based on patients being actively traced to complete treatment and a large proportion being managed as inpatients for the duration of treatment . The PHC system in Bihar remains weak and overburdened with long waiting times and irregular timings–thus returning daily for treatment for a period of 10 days becomes an additional economic burden for patients and caregivers and is likely to result in reduced treatment compliance . Additionally , for Milt containing regimens , there is a requirement for women of reproductive age to take a pregnancy test , and , if negative , to comply with contraceptive cover during treatment and for 3 months afterwards , something that has generally been poorly followed under programmatic conditions . Given that the most common adverse event related to Milt is vomiting , contraceptive injections remain the most suitable option , recently been made available in India within the public health sector [23] . As such , clear coordination and preparation on safety messaging for all treatments evaluated in this study is required . There are a number of limitations to this study . Although it was originally planned that each site would use a particular regimen , there was a degree of mixing of treatments between sites . Additionally , children were under-represented due to regulatory demands , while the majority of patients receiving the Milt+PM arm received treatment as in-patients , reducing the validity of the feasibility interpretation of this arm in normative settings . Finally , the majority of patients in two of the treatment arms were treated by MSF doctors , supporting activities at Hajipur hospital . This is the largest prospective study conducted using the revised WHO recommended VL treatment regimens for the Indian subcontinent , and to the authors’ knowledge , the first NTD based phase 4 study within the Indian subcontinent . The results were used by the Indian national programme to support policy change , introducing SDA and the different combinations as treatment options in the elimination strategy .	Treatment is one of key strategies for visceral leishmaniasis control and elimination . Historically a number of monotherapy drugs for VL treatment were used in India including pentavalent antimonials , amphotericin B deoxycholate ( AmB ) , and miltefosine ( MF ) . With the limited number of drugs available there was a need to preserve existing drugs and to develop shorter and safer treatment regimens . Three short-course combination regimen including AmBisome , miltefosine and paromomycin have been evaluated in a phase III clinical trial conducted in India ( 2008–2010 ) . All showed an excellent safety profile and an efficacy of at least 97% in controlled conditions . In 2010 , WHO recommended the use of new short-course treatment regimens in kala-azar elimination efforts for the Indian subcontinent . Although phase 3 studies have shown excellent results , there remains a lack of evidence on a wider treatment population and the safety and effectiveness of these regimens under field conditions within national program settings . This study was implemented in field conditions with treatment provided by government doctors , providing further evidence for scaling up new regimens in national program contexts within the public health sector and contributing to national policy change in India .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "antimicrobials", "kala-azar", "medicine", "and", "health", "sciences", "clinical", "research", "design", "drugs", "geographical", "locations", "microbiology", "india", "tropical", "diseases", "antifungals", "parasitic", "diseases", "anemia", "research", "design", "pharmaceutics", "neglected", "tropical", "diseases", "pharmacology", "research", "and", "analysis", "methods", "infectious", "diseases", "mycology", "zoonoses", "amphotericin", "adverse", "events", "protozoan", "infections", "hematology", "people", "and", "places", "asia", "drug", "research", "and", "development", "leishmaniasis", "microbial", "control", "biology", "and", "life", "sciences", "drug", "therapy" ]	2018	Field safety and effectiveness of new visceral leishmaniasis treatment regimens within public health facilities in Bihar, India
The majority of recently emerging infectious diseases in humans is due to cross-species pathogen transmissions from animals . To establish a productive infection in new host species , viruses must overcome barriers to replication mediated by diverse and rapidly evolving host restriction factors such as protein kinase R ( PKR ) . Many viral antagonists of these restriction factors are species specific . For example , the rhesus cytomegalovirus PKR antagonist , RhTRS1 , inhibits PKR in some African green monkey ( AGM ) cells , but does not inhibit human or rhesus macaque PKR . To model the evolutionary changes necessary for cross-species transmission , we generated a recombinant vaccinia virus that expresses RhTRS1 in a strain that lacks PKR inhibitors E3L and K3L ( VVΔEΔK+RhTRS1 ) . Serially passaging VVΔEΔK+RhTRS1 in minimally-permissive AGM cells increased viral replication 10- to 100-fold . Notably , adaptation in these AGM cells also improved virus replication 1000- to 10 , 000-fold in human and rhesus cells . Genetic analyses including deep sequencing revealed amplification of the rhtrs1 locus in the adapted viruses . Supplying additional rhtrs1 in trans confirmed that amplification alone was sufficient to improve VVΔEΔK+RhTRS1 replication . Viruses with amplified rhtrs1 completely blocked AGM PKR , but only partially blocked human PKR , consistent with the replication properties of these viruses in AGM and human cells . Finally , in contrast to AGM-adapted viruses , which could be serially propagated in human cells , VVΔEΔK+RhTRS1 yielded no progeny virus after only three passages in human cells . Thus , rhtrs1 amplification in a minimally permissive intermediate host was a necessary step , enabling expansion of the virus range to previously nonpermissive hosts . These data support the hypothesis that amplification of a weak viral antagonist may be a general evolutionary mechanism to permit replication in otherwise resistant host species , providing a molecular foothold that could enable further adaptations necessary for efficient replication in the new host . There are at least 868 described zoonotic microbial pathogens , 33% of which are capable of human to human transmission [1] . Recent viral zoonoses have led to some of the most devastating and medically relevant outbreaks in modern history , including SARS coronavirus , pandemic influenza , and HIV/AIDS , highlighting the urgent need to understand how viruses adapt to infect new species . At a population level , factors influencing the transmission of zoonotic pathogens to humans include increasing population density , greater contact with wildlife , increased travel , and poor public health infrastructure [2] , [3] . However , these factors only allow the microbe increased access to new hosts; they do not directly enable it to adapt to and replicate in the new species . Intermediate hosts , animals that are not the natural host of a virus but are still permissive or semi-permissive for viral replication , play a critical role in cross-species transmission . These hosts can facilitate increased contact between a virus and a new host , and drive adaptive changes that may improve virus replication ( Reviewed in [4] ) . For example , spill-over of Nipah virus from fruit bats into pigs , the intermediate host , increased human exposure to the virus and resulted in eventual human outbreaks in Malaysia [5] , [6] . In another example , lentiviral adaptation through intermediate chimpanzee hosts led to both increased contact with humans , and adaptive genetic changes permitting the virus to inhibit the human versions of several host restriction factors ( Reviewed in [7] ) . At a molecular level , the initial success of a virus after entry into a new host cell depends on its ability to overcome cellular host restriction factors . A subset of these proteins inhibits specific virus families , such as the restriction of retroviruses mediated by TRIM5α [8] . However , other restriction factors , including protein kinase R ( PKR ) , block the replication of multiple virus families . PKR is activated by binding to double-stranded RNA ( dsRNA ) , a common byproduct of both RNA and DNA virus replication , followed by dimerization and autophosphorylation . Activated PKR then phosphorylates the α-subunit of eukaryotic initiation factor 2 ( eIF2α ) , ultimately arresting translation initiation [9] . In response to the broad and potent barrier to viral infection imposed by PKR , most virus families have evolved at least one mechanism to inhibit the PKR pathway [10] . This conflict between host restriction factors and their viral antagonists results in an “arms race” leading to rapid evolution of both sets of genes [11] . The extraordinarily high rate of positive selection ( dN/dS ) among primate PKR genes reflects the evolutionary pressure on PKR to evade virus antagonists [12] , [13] . To productively infect new host species , virus antagonists must rapidly adapt to escape these differences in PKR . In the current study , we modeled the process of viral adaptation through an intermediate , minimally permissive host . We experimentally evolved a recombinant vaccinia virus expressing the rhesus cytomegalovirus PKR antagonist rhtrs1 in African green monkey cells expressing RhTRS1-resistant PKR . We demonstrate that amplification of the exogenous rhtrs1 locus was an early adaption that is sufficient to rescue virus replication in minimally-permissive AGM cells . This amplification of the rhtrs1 locus was also sufficient to expand the species tropism of these viruses , enabling them to infect both human and rhesus cells substantially better than the initial virus . Importantly , rhtrs1 amplification did not occur when the initial virus was passaged through human fibroblasts , suggesting that amplification in AGM cells was a necessary intermediate step to expand the virus host range . Gene amplification is a universal mechanism of rapid adaptation occurring in eukaryotes [14] , [15] , prokaryotes [16] , and viruses [17] , [18] , enabling diverse adaptations including evasion of host restriction factors ( Reviewed in [19] ) . Our results suggest that gene amplification in an intermediate host may be a risk factor for broad cross-species transmission independent of other adaptive events . The rhesus cytomegalovirus ( RhCMV ) PKR inhibitor TRS1 ( RhTRS1 ) can block PKR activation and rescue replication of a vaccinia virus mutant lacking the PKR inhibitor E3L ( VVΔE3L+RhTRS1 ) in several African green monkey ( Chlorocebus aethiops , AGM ) cell lines [20] . However , we discovered that a recombinant vaccinia virus expressing RhTRS1 and lacking both of the known vaccinia PKR antagonists , E3L and K3L , ( VVΔEΔK+RhTRS1 ) produced 100 to 1000-fold less virus in AGM-derived PRO1190 cells relative to VV-βg ( which contains both E3L and K3L ) although it replicated almost as efficiently as VV-βg in AGM-derived BSC40 cells ( Fig . 1A ) . Thus , RhTRS1 varies in its ability to support VVΔEΔK replication in different AGM cells . This VVΔEΔK+RhTRS1 replication defect in PRO1190 cells may be due to incomplete inhibition of PKR in these cells by RhTRS1 . To test this hypothesis , we generated PRO1190 cells stably expressing either a PKR-specific shRNA ( PRO1190-PKR kd ) , which resulted in a 56% reduction of PKR expression , or a scrambled control shRNA ( PRO1190-ctrl kd ) ( Fig . 1B ) . Similarly , PKR-specific RT-qPCR demonstrated a 60% reduction in PKR mRNA from PRO1190-PKR kd cells relative to PRO1190-ctrl kd cells ( 2578 copies or 6503 copies PKR/ng total RNA respectively ) but little difference between PRO1190 and PRO1190-ctrl kd cell PKR levels ( Fig . S1 ) . In PRO1190-PKR kd cells , VVΔEΔK+RhTRS1 replication was almost completely rescued to VV-βg levels ( Fig . 1C ) . We detected a similar increase in VVΔEΔK+RhTRS1 replication after transiently transfecting PRO1190 cells with a siRNA specific for PKR , but not a control siRNA ( data not shown ) . Sequence analysis of PRO1190 PKR identified three non-synonymous mutations relative to a previously reported AGM PKR ( GenBank # EU733254 ) that is sensitive to RhTRS1 ( Fig . S2 , [20] ) . Interestingly , one of these single nucleotide variants is heterozygous in PRO1190 cells , and changes a residue ( T577M ) that is evolving under positive selection in primates [12] . Although additional studies will be needed to determine whether one or more of these PRO1190 PKR polymorphisms is responsible for increased resistance to RhTRS1 , the results shown in Fig . 1 demonstrate that the block to VVΔEΔK+RhTRS1 replication in PRO1190 cells is mediated by PKR . To determine whether VVΔEΔK+RhTRS1 could adapt to overcome the PKR-mediated block to replication in PRO1190 cells , we utilized a system of experimental evolution . We infected PRO1190 cells with VVΔEΔK+RhTRS1 at a low multiplicity of infection ( MOI = 0 . 1 ) . 48 hours post infection ( hpi ) we lysed the infected cells , titered the resulting virus , and infected new PRO1190 cells , repeating this cycle multiple times . After four passages we observed a 10- to 100-fold increase in viral replication that remained stable for at least three subsequent passages in each of three independent lineages ( Fig . 2 ) . We next performed a competition assay to assess the relative fitness of the passaged virus in comparison to the initial VVΔEΔK+RhTRS1 virus . We co-infected PRO1190 cells with either VVΔEΔK+RhTRS1 or VV-A , both of which express eGFP , and the same competitor , VVΔE3L+RhTRS1 , which expresses β-gal ( MOI = 0 . 1 for each virus ) . Virus produced 48 hpi was titered on permissive BSC40 cells and the specific progeny viruses were enumerated by detecting β-gal ( VVΔE3L+RhTRS1 ) and eGFP ( VVΔEΔK+RhTRS1 and VV-A ) expression in the plaques ( Fig . S3 ) . VVΔEΔK+RhTRS1 replicated ∼3-fold better than VVΔE3L+RhTRS1 , whereas VV-A replicated 290-fold better than VVΔE3L+RhTRS1 , confirming that serial passage through PRO1190 cells increased the fitness of passaged viruses approximately 100-fold relative to the initial VVΔEΔK+RhTRS1 . Since the passaged viruses replicated more efficiently in a minimally permissive cell line , we investigated the ability of the passaged viruses to replicate in primary cells from more divergent primates . We have shown that RhTRS1 does not inhibit human or rhesus PKR in the context of VVΔE3L [20] . To determine whether the adaptations that evolved during serial passage in PRO1190 cells affected the virus species tropism , we infected primary human foreskin fibroblasts ( HFF ) or primary rhesus fibroblasts ( RF ) with VVΔEΔK+RhTRS1 , each of the three passaged virus pools or VV-βg at an MOI of 0 . 1 ( Fig . 3 ) . As expected , VVΔEΔK+RhTRS1 replicated poorly and VV-βg replicated efficiently in each cell type . Remarkably , all three passaged pools replicated between 1000- to 10 , 000-fold better than VVΔEΔK+RhTRS1 in both HFF ( Fig . 3 , center ) and RF ( Fig . 3 , right ) . For each pfu of VV-A , VV-B and VV-C used to infect the cells , 5 . 2 , 1 . 9 , and 6 . 3 pfu of progeny emerged from HFF and 40 . 8 , 7 . 3 , and 3 . 7 emerged from RF , respectively , suggesting that these viruses were sufficiently well adapted to enable continuous propagation in these cells ( see below ) . However , these virus pools still replicated 10- to 100-fold better in PRO1190 than in either human or rhesus cells ( Fig . 3 , left ) . Thus , adaptation of VVΔEΔK+RhTRS1 in minimally permissive AGM fibroblasts also provides a substantial replication benefit in human and rhesus cells expressing distantly related PKR proteins . To elucidate the underlying mechanism for this gain of fitness , we harvested DNA from passage 7 viruses for genetic analyses and then passaged the viruses once more in PRO1190 to generate viral stocks for biochemical and infectivity analyses . Gene amplification as a mechanism of rapid adaptation in vaccinia virus has been well documented [17] , [18] , [21] . To determine whether gene amplification could account for the broadly improved replication of passaged VVΔEΔK+RhTRS1 we performed paired-end Illumina based deep sequencing ( Short Read Archive #SRP033208 ) . Based on read depth , we detected duplication of the rhtrs1 locus in all three passage 7 virus pools but not in VVΔEΔK+RhTRS1 ( Fig . 4A ) . Each of the passaged pools contained between 1 . 4 and 1 . 9 copies of rhtrs1 per genome , although these numbers reflect averages of a heterogeneous population of viral genomes . Confirming this estimate of rhtrs1 copy number , the frequency of reads in which we detected a recombination site near the rhtrs1 locus increased as a percentage of total reads in viruses predicted to have more copies of rhtrs1 ( Fig . 4B ) . We used PCR to confirm that the rhtrs1 locus was amplified , using externally directed primers specific to rhtrs1 that only amplify a product if there is a tandem duplication of the gene ( Fig . S4A ) . We detected 3 kb products in all three virus pools , and 2 . 3 kb and 1 . 8 kb products only in the VV-A virus pool ( Fig . S4B ) . We were unable to obtain enough of the 2 . 3 kb band for further analysis , but we did characterize the 3 kb and 1 . 8 kb products by Sanger sequencing . The larger product was identical in all three passaged virus pools , and represents a recombination between the vaccinia virus gene L5R upstream of rhtrs1 with J2R downstream of rhtrs1 . In the smaller product J2R recombined with the neoR gene , which was introduced as a selection marker during construction of VVΔEΔK+RhTRS1 ( Fig . S3C ) . These two sites represented the predominant recombination sites ( 85 . 5% and 2 . 8% respectively ) identified by Illumina deep sequencing . However , we found additional minor recombination sites by Illumina deep sequencing , including a 15 kb duplication in VV-B . The presence of an identical recombination site in all three passaged virus pools suggests that duplication may have been present at a very low frequency in the initial virus population even though we did not detect it in the Illumina sequencing data . Regardless , taken together these data demonstrate that the copy number of rhtrs1 in the viral genome was substantially increased by passage through PRO1190 cells . Unlike a previous study which identified adaptive point mutations arising after locus expansion [17] , we did not detect any point or indel mutations in rhtrs1 in any of the passaged virus pools . However , we identified 13 vaccinia virus gene mutations present at greater than 5% frequency in at least one of the pools ( Table 1 ) . All three passaged pools had one or more of four different single nucleotide deletions within the A35R gene at frequencies ranging from 12 to 42% . Transition mutations affecting the A24R and A37R genes were present at >50% frequencies in VV-A and VV-B respectively , but were rare or absent in the other viral pools . None of the other mutations were detected in all three pools or occurred at >50% frequency in any pool , so are unlikely to account for the improved replication of the passaged viruses . However , the presence of these VV gene mutations raised the question of whether the expanded species tropism we observed was due to the VV gene mutations or to rhtrs1 amplification . If rhtrs1 amplification alone is sufficient for the observed increase in fitness , we reasoned that overexpression of rhtrs1 in trans might rescue VVΔEΔK+RhTRS1 replication . To investigate this possibility , we stably transduced rhtrs1 into HFF ( HFF+RhTRS1 ) , and confirmed RhTRS1 expression by immunoblot ( data not shown ) . We also prepared a control cell line ( HFF-LHCX ) by transducing the empty vector , LHCX , into HFF . In the control cells , VVΔEΔK+RhTRS1 replicated approximately 1000-fold less efficiently than VV-βg ( Fig . 5A ) . In HFF+RhTRS1 , VVΔEΔK+RhTRS1 replication increased more than 100-fold . Thus , combined expression of RhTRS1 from genes in both the cell and the infecting virus potentiated VVΔEΔK+RhTRS1 replication , supporting the hypothesis that rhtrs1 amplification alone is sufficient to expand the species tropism of VVΔEΔK+RhTRS1 . Although rhtrs1 amplification provided a substantial growth benefit in HFF and RF , the passaged viruses still replicated 100- to 1000-fold less efficiently than VV-βg ( Fig . 3 ) . To determine whether this incomplete rescue in HFF was due to incomplete PKR inhibition or represented a second block to replication , we infected HFF stably transduced with either a PKR specific shRNA ( HFF-PKR kd ) that reduces PKR expression >95% , or a non-specific shRNA ( HFF-ctrl kd ) [22] . Knocking down PKR increased VVΔEΔK+RhTRS1 replication ∼1000-fold , indicating that PKR is a major barrier to replication in these cells ( Fig . 5B ) . All three passaged virus pools replicated ∼10-fold better in HFF-PKR kd cells than in HFF-ctrl kd cells , suggesting that rhtrs1 amplification , which fully inhibits PRO1190 PKR , only partially inhibits human PKR . However , these viruses all replicated ∼10-fold less well than VV-βg in the HFF-PKR kd cells . This remaining replication defect may be due to incomplete PKR knockdown in these cells [20] , although it is also possible that an additional host factor inhibits VVΔEΔK+RhTRS1 replication in HFF . Unlike many known PKR inhibitors that block PKR phosphorylation [10] , [23] , RhTRS1 inhibits the PKR pathway after PKR phosphorylation but prior to eIF2α phosphorylation [20] . However , it is possible that RhTRS1 amplification inhibits PKR through an alternative mechanism , such as dsRNA sequestration . To determine whether rhtrs1 amplification altered the mechanism of PKR inhibition , we infected PRO1190 with VV-βg , VVΔE3L , VVΔEΔK+RhTRS1 , and VV-A at an MOI of 3 , and collected cell lysates 24 hpi . 35S metabolic labeling demonstrated that VVΔEΔK+RhTRS1 expressed vaccinia virus proteins in PRO1190 cells , though in much lower quantities than VV-βg , consistent with the former virus being unable to inhibit PKR completely ( Fig . 6 , top left panel ) . In contrast , VV-A produced abundant vaccinia virus proteins , similar to VV-βg , confirming that this virus efficiently inhibits the PKR pathway . We used immunoblot analyses to determine the stage of the PKR pathway inhibited by each virus ( Fig . 6 , lower panels ) . RhTRS1 expression was noticeably higher in the VV-A infected cells compared to those infected with VVΔEΔK+RhTRS1 , consistent with the better replication of VV-A in PRO1190 cells ( Fig . 6 , lanes 4 and 5 ) . PKR phosphorylation was elevated in all virus infected cells except VV-βg , consistent with our previous report [20] that even when RhTRS1 blocks the PKR pathway , it does not block PKR autophosphorylation ( Fig . 6 , lanes 3–5 ) . Phospho-eIF2α levels in VVΔEΔK+RhTRS1 infected cells were intermediate between mock and VVΔE3L infected PRO1190 cells , suggesting that a single RhTRS1 gene weakly inhibits the PKR pathway ( Fig . 6 , lane 4 ) . Infection with VV-A resulted in low levels of eIF2α phosphorylation , similar to that detected in VV-βg infected PRO1190 cells ( Fig . 6 , compare lanes 2 and 5 ) , indicating that RhTRS1 amplification is sufficient to completely inhibit PKR-mediated translational shutdown at a stage after PKR phosphorylation . Together , these data suggest that PKR-mediated pressure in PRO1190 cells selected for rapid amplification of the rhtrs1 locus , and that this amplification was sufficient to enable the virus to block PKR-mediated defenses in PRO1190 cells without altering the mechanism of RhTRS1-mediated PKR inhibition . In previous studies , RhTRS1 alone was insufficient to inhibit human PKR [20]; however , the passaged viruses reported here replicate substantially better in HFF than VVΔEΔK+RhTRS1 , suggesting that amplification of rhtrs1 is able to inhibit PKR at least partially in these cells . To test this hypothesis and determine whether the mechanism of PKR inhibition was the same in HFF as it is in PRO1190 , we infected HFF ( MOI = 3 ) and prepared cell lysates 24 hpi . In contrast to infection of PRO1190 cells , infection of HFF with VVΔEΔK+RhTRS1 resulted in nearly complete shut off of protein synthesis by 24 hpi ( Fig . 6 , top right panel ) and produced only trace amounts of RhTRS1 . Compared to VVΔEΔK+RhTRS1 , VV-A infection of HFF resulted in detectable , though still low overall levels of 35S labeled proteins , and much more RhTRS1 ( Fig . 6 , lanes 9 and 10 ) . PKR phosphorylation and eIF2α phosphorylation were elevated after infection with both VVΔEΔK+RhTRS1 and VV-A compared to mock or VV-βg controls ( Fig . 6 , compare lane 9 to lanes 6 and 7 ) . These data suggest that a single copy of rhtrs1 was insufficient to inhibit the translational shutoff mediated by human PKR , but amplification of this weak antagonist resulted in partial inhibition of human PKR allowing enough protein synthesis to support a modest level of virus replication . Because selection in AGM cells resulted in a broad expansion of viral species tropism , we investigated whether passage of VVΔEΔK+RhTRS1 directly in HFF would similarly select for mutants , such as rhtrs1 amplification , that improved replication in HFF . We therefore serially infected PRO1190 cells and HFF with VVΔEΔK+RhTRS1 in parallel ( Fig . 7 ) . In PRO1190 cells , viral fitness again increased after four passages . In contrast , virus replication was strongly inhibited in HFF , and we were unable to detect any viral replication after three rounds of infection in all three pools . These data suggest that , under these experimental conditions , adaptation in PRO1190 cells was a necessary intermediate step for improved replication in HFF . Finally , we evaluated whether the improved replication of the viruses that had been passaged in PRO1190 cells was sufficient to enable stable propagation in HFF . We therefore serially infected HFF with VV-A , VV-B and VV-C at low multiplicity of infection ( MOI = 0 . 1 at each passage ) . We were able to propagate these viruses in HFF for at least four passages . Moreover , replication increased between 5- to 14-fold after only two passages , suggesting further adaptation occurred in HFF cells ( Fig . 7B ) . To define mutations that may have evolved during serial passage in HFF , we performed paired-end Illumina based deep sequencing on viral DNA isolated after the fourth round of passage in HFF . Again , we did not find any mutations in rhtrs1; however , we detected an average rhtrs1 copy number of 1 . 7 , 2 . 8 , and 2 . 7 for the virus pools derived from VV-A , VV-B and VV-C , respectively , representing an average expansion of the rhtrs1 locus by approximately one additional copy relative to the PRO1190-adapted viruses . In addition , two VV gene mutations ( in F7L [indel] and J6R ) that arose during PRO1190 adaptation were lost after HFF adaptation , and two new mutations ( in F7L [missense] and H4L ) of uncertain significance appeared during HFF adaptation ( Table S3 ) . Taken together , our study suggests the hypothesis that gene amplification acts broadly to increase replication in a variety of hosts , and may provide a molecular foothold that allows for continued species-specific adaptation of the virus in more resistant hosts . Cross-species pathogen transmissions have been responsible for more than 60% of all emerging infectious diseases in humans during the past 70 years [24] . Human immunodeficiency viruses , avian influenza viruses , and the recently described Middle East respiratory syndrome coronavirus exemplify the ongoing threat and potential of animal viruses to spread to and among humans , highlighting the urgent need to understand the mechanisms underlying cross-species transmission and adaptation to new hosts . One such mechanism , genetic locus amplification in response to selective pressure , has been observed in both viruses and bacteria [16]–[18] , [21] . Here we demonstrate that amplification of the exogenous gene rhtrs1 is sufficient to block potent PKR-mediated inhibition and improve VVΔEΔK+RhTRS1 replication in AGM-derived PRO1190 cells . This adaptation also expanded the species tropism of the virus , enabling markedly improved replication in otherwise resistant human and rhesus monkey cells . Importantly , VVΔEΔK+RhTRS1 failed to replicate in HFF to a level sufficient to sustain transmission upon serial passage , demonstrating that adaptation in PRO1190 was a critical intermediate step to expand the viral species tropism . Thus , the process of adaptation in one host may increase the likelihood of virus transmission to a variety of divergent species . Under PKR-mediated selective pressure , the rhtrs1 locus amplified during serial passage of VVΔEΔK+RhTRS1 in minimally permissive PRO1190 cells ( Fig . 4 ) . It is not clear whether the initial duplication ( s ) occurred during preparation of the VVΔEΔK+RhTRS1 stock in BSC40 cells or during the first few passages in the PRO1190 cells . The observation of a faint PCR product from the starting virus , VVΔEΔK+RhTRS1 using outward directed rhtrs1 primers in one of three experiments ( Fig . S4 ) , and the detection of an identical recombination break point in all 3 independently passaged virus pools suggest the amplification may have been present at a very low level in the starting virus . However , amplification of the locus in VVΔEΔK+RhTRS1 occurred below the level of Illumina deep sequencing detection ( Fig . 4 ) , supporting the idea that if rhtrs1 duplications are present they are rare in the initial VVΔEΔK+RhTRS1 stock . Additionally , we detected recombination between J2R and neoR ( Fig . S4 ) only in the VV-A pool , suggesting that recombination events did arise during serial passage . Regardless of when they arose , amplifications of the rhtrs1 locus were substantially enriched during virus passage under selective pressure ( Figs . 2 and 4 ) . A previous study demonstrated amplification of K3L as an adaptive mechanism against human PKR with strikingly similar kinetics to the current study [17] . Together , these two studies support the hypothesis that preexisting or frequently arising de novo gene duplications enable vaccinia virus to adapt rapidly to selective conditions imposed by relatively resistant host restriction factors . The “accordion hypothesis” of rapid evolution posits that gene amplification provides a replication benefit to the virus and those extra copies of a weak viral antagonist of host defenses provide additional templates to acquire potentially adaptive mutations . Indeed , Elde , et al . detected such an adaptive mutation in K3L ( H47R ) , apparently arising after amplification of the locus [17] . Therefore , we were surprised that no mutations arose in rhtrs1 , although it may be that additional rounds of replication would reveal such mutations . We did , however , detect 13 mutations in endogenous vaccinia virus genes that occurred at >5% frequency ( Fig . 4 ) . While our data suggest that none of these mutations are necessary to expand the species tropism of VVΔEΔK+RhTRS1 ( Fig . 5 ) , we have not ruled out the possibility that they may provide some replication benefit . The presence of mutations in A24R , A35R , and A37R are the most intriguing , as they were either present at >50% frequency in one pool ( A24R and A37R ) or detectable in all three pools ( A35R ) . None of these genes has been previously implicated as a PKR antagonist . A37R is conserved across multiple poxvirus families ( http://www . poxvirus . org ) , but its function is unknown . A24R is a subunit of RNA polymerase . If the mutation we identified acts like some other reported RNA polymerase mutations to decrease transcription elongation [25] , [26] , this mutation might result in less dsRNA production and therefore less PKR activation . A35R is a gene of unknown biochemical function that may be involved in evasion of the adaptive immune response [27] , [28] . All three passaged virus pools contained nucleotide deletions in A35R at greater than 10% frequency . A35R orthologs are conserved across most poxviruses but , intriguingly , variola virus contains a truncation in its A35R gene , demonstrating that similar truncations have evolved in the past . Further studies are underway to evaluate the potential contributions of these mutations to viral replication . Consistent with previous studies of RhTRS1 in other AGM cell types [20] , PKR , but not eIF2α , was phosphorylated in PRO1190 cells infected with VV-A . Thus , the improved replication of the passaged virus pools in PRO1190 cells was likely due to enhancement of the basic RhTRS1-mediated inhibition of PKR , and not due to another mechanism , such as dsRNA sequestration or reduction in the abundance of dsRNA . VVΔEΔK+RhTRS1 infected PRO1190 cells had phospho-eIF2α levels lower than that found in VVΔE3L infected cells ( Fig . 6 , lanes 3 and 4 ) , suggesting that even just a single copy of rhtrs1 is able to inhibit PKR function to a small degree , but amplification of the locus appears to be needed to express enough RhTRS1 to inhibit eIF2α phosphorylation potently and enable efficient viral replication . It is also possible that elevated expression of RhTRS1 is necessary to block another activity of PKR , such as autophagy or inflammasome responses [29]–[31] that might aid in replication . Adaptation of VVΔEΔK+RhTRS1 to PRO1190 cells provided a substantial replication advantage in both human and rhesus monkey fibroblasts . Although VV-A replicated to much higher titers and expressed more RhTRS1 than VVΔEΔK+RhTRS1 in HFF , 35S metabolic labeling revealed relatively low protein synthesis rates and eIF2α phosphorylation was still elevated after VV-A infection compared to VV-βg . A substantial proportion of the block to VV-A replication in HFF is still mediated by PKR despite RhTRS1 overexpression ( Fig . 5B ) , suggesting that further adaptation in HFF may be necessary to block the PKR pathway in HFF completely . These results indicate that RhTRS1 overexpression blocks human PKR incompletely . Nonetheless , viruses that had adapted by initial passage in PRO1190s replicated in HFF at a level sufficient to enable sustained passage in HFF ( Fig . 7B ) . Furthermore , these PRO1190-adapted viruses acquired additional changes as a result of serial passage in HFF , although the biological relevance of these changes is currently unclear . Combined with the observation that serial passage of VVΔEΔK+RhTRS1 in HFF failed to generate any adapted viruses ( Fig . 7A ) , these data suggest that adaptation of rhtrs1 in minimally permissive PRO1190 cells was a critical intermediate step in the generation of a virus with broadened species tropism . Cross-species pathogen transmission is an important source of emerging infections worldwide . Our study illustrates that gene amplification of a weak viral antagonist of PKR can broaden the host range of vaccinia virus . The presence of gene families in other large DNA viruses , e . g . the cytomegalovirus US22 family , of which rhtrs1 is a member , provides indirect evidence that episodes of locus amplification have also occurred in other viruses . Adaptation of a viral antagonist through non-synonymous mutations has the potential to confer a species-specific advantage for the virus in a specific host . In contrast , gene amplification and subsequent over-expression of the antagonist is more likely to increase protein activity through mass action effects in a variety of hosts . Thus , gene amplification may be a common evolutionary strategy employed by large DNA viruses , permitting modest replication in otherwise resistant host species and providing a molecular foothold that enables further adaptations necessary for more efficient replication and spread in the new host . AGM fibroblasts ( PRO1190 , Coriell Institute for Medical Research ) human foreskin fibroblasts ( HFF ) , rhesus fibroblasts ( RF ) , BSC40 and HeLa cells , and derivative cell lines were maintained in Dulbecco's modified Eagle's medium supplemented with 10% NuSerum ( BD Biosciences ) as previously described [32] . At times , PRO1190 cell lines were also propagated in Minimal Essential Medium with 20% fetal calf serum and anitbiotics to enable more rapid growth . PRO1190 cells were transduced with a nonsilencing control or PKR-targeting shRNA lentiviral vectors ( Open Biosystems , catalogue numbers RHS4430-98819555 and RHS4346 , respectively ) and selected in puromycin ( 5 µg/ml ) to generate PRO1190-ctrl k/d and PRO1190-PKR k/d cells . pEQ1364 was constructed by moving the RhTRS1 gene , with a C-terminal biotinylation signal and 6x-His tag , as a HindIII/PmeI fragment from pEQ1215 [20] into the HindIII/HpaI sites of pLHCX ( Clontech Laboratories , Inc ) . HFF-LHCX and HFF+RhTRS1 were produced by transducing HFF with retroviral vectors made using LHCX and pEQ1364 , respectively and selecting with hygromycin B ( 100 µg/ml ) . Vaccinia virus ( VV ) Copenhagen strain ( VC2 ) [33] and VVΔE3L [34] , both obtained from Bertram Jacobs ( Arizona State University ) , and VV-βg ( VC2-LacZ in [20] were propagated and titered in BSC40 cells . VC-R2 ( VVΔE3LΔK3L ) was constructed by replacing the E3L gene in the K3L-deleted VACV vP872 strain ( ΔK3L in VC2 background , provided by Bertram Jacobs ) [35] by homologous recombination . The 518 bp 5′ arm was created by PCR amplification of VC2 DNA with primers C15 ( 5′-GATTAAGGGTACTAGCGGCACCG′3′ ) ×C16 ( 5′-TTTTAGAGAGAACTAACACAACCAGC-3′ ) . The 512 bp 3′ arm was created by PCR amplification of VC2 DNA with primers C19 ( 5′-GTGTAGTAAGCTAGCGAGCTCGGTACCTTCTAGTTATCAATAACAGTTAGTAGTTTAG-3′ ) ×C20 ( 5′-CCAACAAACTGTTCTCTTATGAATCG-3′ ) . The reading frame of EGFP including the PEST sequence was amplified with primers C17 ( 5′-GCTGGTTGTGTTAGTTCTCTCTAAAACCCGGGATCCACCGGTCGCC-3′ ) ×C18 ( 5′-GGTACCGAGCTCGCTAGCTTACTACACATTGATCCTAGCAGAAGC-3′ ) using pD2EGFP-N1 ( Clonetech ) as the template . PCR products were gel-purified and mixed together as template for fusion PCR using C15×C20 . PfuUltra polymerase ( Agilent Technologies ) was used for these PCR reactions . PCR products were cloned into the pCR2 . 1 TOPO vector to generate plasmid S96 . S96 was used as template for PCR amplification of marker +5′ and 3′ arms using C15×C20 followed by gel-purification . BS-C-1 cells grown on 12 well plates were infected with vP872 at MOI = 2 and transfected 2 hours after infection with 1 µg of the purified PCR product . Cell lysates were collected after 18 hours and plated in a dilutions series on RK13+E3L+K3L cells [36] . Green plaques were picked after 48 hours at the highest dilution possible and plaque purified an additional three times on RK13+E3L+K3L cells . EGFP expression in VC-R2 is under the control of the endogenous E3L promoter . VVΔEΔK was propagated and titered using HFF+TRS1 cells ( HF-TRS1 in [37] ) . VVΔEΔK+RhTRS1 was constructed by homologous recombination of plasmid pEQ1233 [20] into the thymidine kinase ( TK ) locus of VVΔEΔK . Recombinant virus was plaque purified three times in BSC40 under G418 selection and subsequently propagated and titered on BSC40 cells . Total RNA from PRO1190 cells was amplified using previously reported PKR specific primers [12] . The amplification product was gel purified and cloned using the StrataClone PCR cloning kit ( Agilent ) . Multiple plasmids were submitted for Sanger sequencing using PKR specific sequencing primers ( #859: 5′-ATGGCTGGTGATCTTGCAC; #860: 5′-GTGAACAACTCACTTGCTTC; #861: 5′-GAAACTAGACAAAGTTTTGGC; #862: 5′-CTAACATGTATGTCGTTCCT; #863: 5′-AAGGCACTTAGTCTTTGATC; #864: 5′-TCTGATATCTCAAGCAATGC ) . Contigs were assembled and curated in Geneious Pro v4 . 8 . 5 ( GenBank #KF728076-7 ) . The predicted amino acid sequence was aligned to the predicted amino acid sequences of previously reported AGM ( GenBank # EU733254 ) , rhesus macaque ( GenBank# EU733261 ) , and human PKR ( GenBank # NM001135651 ) using ClustalW2 ( http://www . ch . embnet . org/software/ClustalW . html ) [38] . For every round of infection , triplicate confluent 10 cm dishes ( Figure 2 ) or 6-well plates ( Figure 7 ) of PRO1190 or HFF were infected with VVΔEΔK+RhTRS1 ( MOI = 0 . 1 ) . Two days post-infection , cells were collected , pelleted and resuspended in 1 mL DMEM+10% NuSerum . After three freeze/thaw cycles , virus titers were determined on BSC40 by plaque assays and used for the next round of infection . If insufficient virus was produced to infect cells at MOI = 0 . 1 , the entire volume of virus lysate was used in the subsequent round of infection . Vaccinia virus DNA from passage 7 in PRO1190 ( Figure 2 ) or passage 4 in HFF ( Figure 7 right ) viruses was purified from infected cell cytoplasmic extracts for genetic analyses described below [39] . The passage 7 pools were further expanded ( passage 8 ) in PRO1190 for use in virologic assays . PRO1190 cells were co-infected with 0 . 1 MOI of either VVΔEΔK+RhTRS1 or VV-A , and 0 . 1 MOI of VVΔE3L+RhTRS1 as a common competitor . Two days post-infection cells were collected , pelleted and resuspended in 1 mL DMEM+10% NuSerum . After three freeze/thaw cycles , virus titers were determined on BSC40 by plaque assays . VVΔE3L+RhTRS1 plaques were detected with the β-gal substrate ImaGene Red C12RG following the manufacturer's directions ( Life Technologies ) . Plaques were imaged on a Typhoon Trio imager ( eGFP - 488 nm excitation , 520 BP 40 filter; ImaGene Red - 532 nm excitation , 580 BP 30 filter ) at 50 µm/pixel resolution , and classified as GFP+ , ImaGene Red+ , or double positive using ImageJ software ( http://rsb . info . nih . gov/ij/ ) . Cells were mock infected or infected with vaccinia viruses ( MOI = 3 ) . One day postinfection , the cells were lysed in 2% sodium dodecyl sulfate ( SDS ) . Equivalent amounts of the lysates were separated on 10% SDS-polyacrylamide gels , transferred to polyvinylidene difluoride ( PVDF ) membranes , and probed with one of the following antibodies: PKR ( sc-6282; Santa Cruz Biotechnology , Inc . ) , phospho-PKR ( T446; 1120-1; Epitomics ) , eIF2α or phospho-eIF2α ( Ser51 ) antibody ( both from Cell Signaling Technology , catalog numbers 9722 and 9721 , respectively ) , TRS1 α999 [37] , or actin ( A2066; Sigma ) . All purchased antibodies were used according to the manufacturer's recommendations . Proteins were detected using the Western Star chemiluminescent detection system ( Applied Biosystems ) according to the manufacturer's recommendations . Densitometry measurements were performed using ImageJ . Total RNA was extracted from PRO1190 , PRO1190-ctrl kd , and PRO1190-PKR kd cells with TRIzol reagent following the manufacturer's protocol ( Invitrogen ) , and 1 ng of total RNA was assayed per reaction . The standard curve was generated from 10-fold serial dilutions of a PRO1190 PKR containing plasmid ( described above in PRO1190 PKR sequence analysis , pEQ1334 ) diluted in 200 pg/µL salmon sperm DNA . All samples were amplified in triplicate using the GoTaq 1-step RT-qPCR kit following the manufacturer's protocol ( Promega ) using PKR specific primers ( #1063: 5′-CACAGAATTGACGGAAAGAC; #1064: 5′-ATCCCAACAGCCATTGTAGT ) . RT-qPCR was performed on a Rotor-Gene Q thermocycler ( Qiagen ) with temperature holds at 37°C×15 min and 95°C×10 min followed by 40 cycles of 95°C×10 s , 60°C×30 s . Raw data was analyzed using the included Rotor-Gene Q series software using the automatic cycle threshold ( Ct ) setting for assigning baseline and threshold for Ct determination . Nonlinear regression analysis to determine PKR copy number in the experimental samples was performed in GraphPad Prism 6 . PRO1190 and HFF cells were mock infected or infected with the indicated viruses ( MOI = 3 ) . At 24 hpi , the cells were labeled for 1 h with 100 µCi/ml L-[35S]methionine/L-[35S]cysteine ( EasyTag express protein labeling mix; PerkinElmer ) in medium lacking methionine and cysteine . The cells were then lysed in 2% SDS . Equivalent amounts of protein from each sample were separated on 10% SDS-polyacrylamide gels , dried , and visualized by autoradiography .	The spread of microbes from animals to humans has been responsible for most recently emerging human infectious diseases , including AIDS , bird flu , and SARS . Therefore , understanding the evolutionary and molecular mechanisms underlying cross-species transmission is of critical importance for public health . After entering a new host cell , the success of a virus depends on its ability to overcome antiviral factors in the cell , such as protein kinase R ( PKR ) . To investigate the process of virus transmission between species , we employed a recombinant vaccinia virus ( VVΔEΔK+RhTRS1 ) expressing the rhesus cytomegalovirus PKR antagonist RhTRS1 . This protein inhibits some African green monkey ( AGM ) PKRs; however , it does not inhibit human or rhesus variants of PKR . Serial passaging VVΔEΔK+RhTRS1 in RhTRS1-resistant AGM cells resulted in rhtrs1 duplication in the viral genome , which improved VVΔEΔK+RhTRS1 replication in AGM cells . Remarkably , rhtrs1 duplication also enhanced virus replication in human and rhesus cells . In contrast , passage of VVΔEΔK+RhTRS1 in human cells , without prior adaptation in AGM cells , did not improve VVΔEΔK+RhTRS1 replication . These results support the hypothesis that amplification of a weak viral antagonist of a host defense protein in one species may enable cross-species transmission into new hosts that are nonpermissive to the initial virus .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "medicine", "infectious", "diseases", "mechanisms", "of", "resistance", "and", "susceptibility", "viral", "immune", "evasion", "virology", "microbial", "evolution", "host-pathogen", "interaction", "biology", "microbiology", "evolutionary", "biology", "viral", "diseases", "viral", "evolution" ]	2014	Adaptive Gene Amplification As an Intermediate Step in the Expansion of Virus Host Range
Genetic diversity is maintained by continuing generation and removal of variants . While examining over 800 , 000 DNA variants in wild isolates of Caenorhabditis elegans , we made a discovery that the proportions of variant types are not constant across the C . elegans genome . The variant proportion is defined as the fraction of a specific variant type ( e . g . single nucleotide polymorphism ( SNP ) or indel ) within a broader set of variants ( e . g . all variants or all non-SNPs ) . The proportions of most variant types show a correlation with the recombination rate . These correlations can be explained as a result of a concerted action of two mutation mechanisms , which we named Morgan and Sanger mechanisms . The two proposed mechanisms act according to the distinct components of the recombination rate , specifically the genetic and physical distance . Regression analysis was used to explore the characteristics and contributions of the two mutation mechanisms . According to our model , ~20–40% of all mutations in C . elegans wild populations are derived from programmed meiotic double strand breaks , which precede chromosomal crossovers and thus may be the point of origin for the Morgan mechanism . A substantial part of the known correlation between the recombination rate and variant distribution appears to be caused by the mutations generated by the Morgan mechanism . Mathematically integrating the mutation model with background selection model gives a more complete depiction of how the variant landscape is shaped in C . elegans . Similar analysis should be possible in other species by examining the correlation between the recombination rate and variant landscape within the context of our mutation model . Genetic diversity is maintained by interplay between the continuing generation and removal of variants; variants are produced by mutation and removed by genetic drift or natural selection during evolution . The modes of generating mutation are biologically diverse , and variants accumulate throughout the life history of a species [1] . Natural selection can drive non-neutral variants to extinction or fixation . Neutral variants can be removed either by genetic drift [2] , which is driven by simple chance , or by natural selection because of linkage or proximity to a non-neutral variant [3 , 4] . The variants that are closely linked to a beneficial mutation may become concurrently fixed as a result of selective sweeps [3] , and the variants that are closely linked to a deleterious mutation can become concurrently extinct as a result of background selection [4] . Thus genetic diversity is shaped by both mutation and natural selection . In many species including Caenorhabditis elegans , higher frequency of variants is observed in the regions of high recombination [5–7] . Natural selection , notably selective sweep and background selection , is the favored explanation for the existence of this pattern [8] , but other proposed explanations include elevated rates of mutation associated with higher recombination , sequestration of important DNA or genes in the regions of low recombination , and sequencing analysis error [9 , 10] . The recombination-based mutation mechanism was arguably ruled out in Drosophila melanogaster early on [5] but was raised as a possible explanation for the variant distribution in humans [11 , 12] . In C . elegans , background selection in combination with selective sweep were shown to shape the variant distribution [7 , 13 , 14] , but a study of codon usage suggested that elevated rates of mutation exist in regions of high recombination [15] . The analysis of C . elegans mutation accumulation ( MA ) strains does not show a correlation between the recombination rate and the accumulation of mutations and thus strongly argues against a substantial role of mutation [16 , 17] , but it is possible that culturing condition in the laboratory leads to mutation rates that do not reflect the mutation rates in the wild environment . Thus in shaping the variant distribution , natural selection is generally agreed as an important factor while mutation is thought to play a lesser role in C . elegans [7 , 13] and perhaps an insignificant role in many species [18–20] . In the present study , we performed a more complete examination of genetic diversity by a previously untried analysis of the composition of variants ( e . g . the proportion of specific variant types ) , which complements the standard analysis of the distribution of variants ( i . e . variant frequency and density ) . Here , we use the term proportion for a single specific variant type and the term composition for proportions of all variant types . Over 800 , 000 homozygous variants present in 40 wild isolates of C . elegans , which had been whole-genome sequenced concurrently [10] , form the basis of our analysis . First , we demonstrated a strong correlation between the recombination rate and the proportion of many specific variant types . This new correlation is as good as the known correlation between the recombination rate and variant distribution . This new correlation demands an explanation and provides a new avenue for studying genetic diversity . Most other genomic features , such as GC content , expression level , and essential gene exon density , show weaker correlation than the recombination rate whereas exon and repetitive sequence densities show a similar level of correlation as the recombination rate . The correlation between the recombination rate and variant composition remain strong using only the variants that do not affect exons or only the variants that affect the DNA outside repetitive sequences . We discuss natural selection and mutation as possible actors responsible for the correlation . To explain the correlation , we present a mutation model and associated mathematical equations , which are based on the premise of a combined action of mutation mechanisms with distinct basis of mutation generation probabilities . Using regression analysis , we estimated the contributions of different mutation mechanisms and their properties . The mutation model also can be used to re-examine the correlation between the recombination rate and variant distribution . Furthermore , the mutation model can be integrated with background selection model to depict a more complete history of the landscape of variants in C . elegans . In addition to explaining the correlation between the recombination rate and variant composition , our analysis suggests a greater importance of mutation in shaping the variant distribution in C . elegans than previously thought . We discovered a striking pattern while examining the composition of variants such as indels of size between 40 and 699 base pairs ( i40-699 ) . The proportion of i40-699 out of all variants is higher near autosomal ends and lower in autosomal centers ( Fig 1A ) . For every genomic interval , the proportion of i40-699 out of all variants is calculated by dividing the number of i40-699 by the number of all variants . All variants include all SNPs , all indels , and all other complex variants . For example , a genomic interval with 70 SNPs , 30 indels including a single i40-699 , and no other complex variants has 1% i40-699/variants by this metric of variant type proportions . By polymerase chain reaction ( PCR ) assay , we have positively verified at least 121 out of 124 ( 97 . 5% ) i40-699 in CB4856 , which suggests a high quality of variant calling for the indels of this size range . The pattern of higher proportion of i40-699 out of all variants in autosomal ends is reminiscent of the pattern of variant distribution reported in prior publications [7 , 10 , 21 , 22] . The underlying reason for the pattern of the variant distribution was attributed to the associated recombination rate , which is higher in autosomal ends , and thus we tested the correlation between the recombination rate and the proportion of i40-699 out of all variants . The strength of correlation can be examined by a number of correlation tests including the methods of Pearson [23] , Spearman [24] , and Kendall [25] as well as by simple linear regression . The possible range of values for Pearson's r , Spearman's rho , Kendall's tau is between 1 and -1 with a larger absolute value indicating a stronger correlation . Using linear regression , useful indicators include the t value associated with the slope and the R2 value . Correlation tests show a strong positive correlation between the recombination rate and the proportion of i40-699 out of all variants ( Fig 1C , Pearson's r = 0 . 61 , Spearman's rho = 0 . 57 , Kendall's tau = 0 . 4 ) . By linear regression , the t value of 7 . 6 associated with the slope indicates that the slope value is 7 . 6-fold greater than the standard error ( p < 0 . 001 , R2 = 0 . 36 ) . A positive correlation means that the proportions of i40-699 out of all variants are higher in the regions of high recombination . A strong correlation raises the possibility of recombination rate being a causative factor in establishing different proportions of i40-699 out of all variants across the genome . For a more thorough analysis of variant composition , we examined the proportions of 30 variant types out of all variants , the proportions of 26 non-SNP subtypes out of all non-SNPs , and the proportion of transitions ( Ts ) and transversions ( Tv ) out of all SNPs . Indel variant types range from indels of single base pair change ( i1 ) to indels of 100 to 4999 base pair change ( i100-4999 ) . We define non-SNP as variant types encompassing indels and substitutions of two or more adjacent base pairs as well as combination of these mutation types , which include variants without any change in the number of base pairs ( i0 ) but is not a SNP . The proportions of SNPs out of all variants and transitions ( Ts ) out of SNPs showed strong negative correlation with the recombination rate with lower proportion in the regions of high recombination ( Fig 1D , S1 Table ) . A negative correlation for SNPs out of all variants means that SNPs constitute a smaller proportion of all variants in the regions of high recombination , which coincide with autosomal centers ( Fig 1B ) . A negative correlation does not mean that there are fewer SNPs in the regions of high recombination , and in fact there are more SNPs in the regions of high recombination . Most variant type proportions show a strong positive correlation with the recombination rate ( S1 Table ) . The range of Pearson's r using the absolute value is from 0 . 015 to 0 . 67 with 25% quartile of 0 . 33 , median of 0 . 50 , 75% quartile of 0 . 59 , and mean of 0 . 45 ( Fig 2A ) . The weakest correlations are associated with i0 , i1 , i4 and i5 as indicated by the weak Pearson's r values for i0/non-SNPs ( -0 . 015 ) , i1/all variants ( 0 . 17 ) , i4/non-SNPs ( -0 . 16 ) , and i5-9/non-SNPs ( -0 . 07 ) . Similar conclusions can be made using Spearman's rho and Kendall's tau ( S1 Table ) . Thus , strong correlation with the recombination rate exists for the proportion of most but not all variant types . As a comparison , we examined the well-known correlation between the recombination rate and the distribution of variants using the same analysis . Correlation tests show strong positive correlation for all variants ( Pearson's r = 0 . 53 , Spearman's rho = 0 . 65 , Kendall's tau = 0 . 46 ) , and linear regression analysis shows a t of slope of 6 . 3 and R2 of 0 . 28 ( Fig 1E ) . The correlation is positive for all variant types , but the correlation is somewhat weaker for SNP ( Pearson's r = 0 . 46 ) , Ts ( Pearson's r = 0 . 42 ) , Tv ( Pearson's r = 0 . 5 ) , and especially i0 ( Pearson's r = 0 . 25 ) . All other indels including i40-699 show a strong positive correlation ( Pearson's r = 0 . 71 for i40-699 , S2 Table ) . Using Spearman's rho and Kendall's tau , the trend is similar but with smaller differences in the strength of correlation among different variant types ( S2 Table ) . Together , the strength of the newly observed correlation between the recombination rate and variant composition is similar to that between the recombination rate and variant distribution . The correlation between the recombination rate and variant composition is similar among autosomes , but the correlation is often weaker for the X chromosome ( S2 Fig ) . Interestingly , the established correlation between the variant distribution and the recombination rate is also weaker for the X chromosome ( S2E Fig ) . This suggests that if recombination rate is indeed a contributing factor in the distribution and composition of variants , the effect is smaller in the X chromosome . To further test the relevance of the correlation between the recombination rate and variant composition , we also examined the correlation between variant composition and several other genomic features . Other genomic features examined are exon density , coding DNA sequence ( CDS ) density , essential gene density using either the gene length or exon length , repetitive sequence density using inverted and tandem repeats , GC content , expression level , and five different measures of chromatin state . Some of these factors , including exon density and repeat sequence density , are correlated with the recombination rate in C . elegans [26 , 27] . Chromatin states in C . elegans have been placed into five groups [28] , and some of these chromatin states ( e . g . methylation state of histone H3K4me1 as measured by ChIP-seq ) are known to be correlated with other genomic features , such as expression level or repeat sequence density [28] . Details of the results including all p values are shown in S3 and S4 Tables , and highlights are summarized in the next two paragraphs . Using 58 definitions of the proportion of variant types , the correlations with the recombination rate show Pearson's r median absolute value of 0 . 5 ( Fig 2A , S1 Table ) . Examining the correlations between the proportion of variant types with other genomic features instead , the correlations are essentially of same strength with the repetitive sequence density ( Pearson's r median absolute value = 0 . 52 ) , with the exon density ( Pearson's r median absolute value = 0 . 54 ) , and with CDS density ( Pearson's r median absolute value = 0 . 48 ) ( S3 Table ) . The correlations are weaker with all other factors . The Pearson's r median absolute values between 0 . 3 and 0 . 4 are observed for H3K36me1 , H3K9me1 , and H3K27me1 histone methylation states . The Pearson's r median absolute value is less than 0 . 3 for all other factors including essential gene exon density . Comparable results were obtained using Kendall's tau , and Spearman's rho suggests a somewhat stronger correlation with exon and repetitive sequence densities ( S1 and S3 Tables ) . In summary , the correlation between the recombination rate and variant composition is among the strongest with the correlations between variant composition and exon or repetitive sequence densities showing a similar level of correlation . The distribution of variants was also examined as a comparison . Using the distribution of 31 different types of variants , the correlations with the recombination rate show a Pearson's r median value of 0 . 70 ( S2 Table ) . Examining the correlations between the distribution of variants and other genomic features instead , the correlations are slightly stronger with the repetitive sequence density ( Pearson's r median = 0 . 75 ) and weaker with the exon density ( Pearson's r median = -0 . 57 ) . By Pearson's r , the correlations are weaker with all other factors ( S4 Table ) . Comparable results were obtained using Spearman's rho ( S2 and S4 Tables ) . Using Kendall's tau , the strongest correlation was observed with the recombination rate ( tau median = 0 . 72 ) followed by repetitive sequence density ( tau median = 0 . 57 ) . Taken together , the relative strength of the correlation between the recombination rate and variant composition compared to other genomic features is arguably as strong as the relative strength of the correlation between the recombination rate and variant distribution . We asked whether exon density or repetitive element density , which are correlated with the recombination rate , might be responsible for the existing landscape of variant composition rather than the recombination rate . First , we examined the correlation using only the variants that affect exons as opposed to using the variants that do not affect exons . By Pearson's r , similarly strong correlations are observed using all variants ( median absolute value = 0 . 5 ) and using only the variants that do not affect exons ( median absolute value = 0 . 53 ) ( S5 Table ) . On the other hand , considerably weaker correlations are observed using only the variants that affect exons ( Pearson's r median absolute value = 0 . 12 ) ( S6 Table ) . Similar conclusions can be made using Spearman's rho and Kendall's tau values . Thus , the variants that affect exons are a minor actor in shaping the correlation between the recombination rate and variant composition . As for repetitive sequence density , we examined the correlation using only the variants that affect DNA outside repetitive sequences . This was done because examining the correlation by separating variants into those that affect and those that do not affect repetitive sequences was less satisfactory ( S7 and S8 Tables ) . Using the variants that affect DNA outside repetitive sequences , the correlations are strong ( Pearson's r median absolute value = 0 . 49 ) ( S9 Table ) unlike when using only the variants that are located inside repetitive sequences ( Pearson's r median absolute value = 0 . 11 ) ( S10 Table ) . Similar conclusions can be made using Spearman's rho and Kendall's tau values . For highly repetitive sequences , sequencing analysis and variant calling may be problematic , and perhaps different variant types are affected to a different level depending on the degree of repetition . However , we cannot discount the possibility of the real absence of correlation between the recombination rate and variant composition in repetitive sequences . In any case , repetitive sequences have only a minor effect in the observed correlation between the recombination rate and variant composition . Here , we discuss forces that can affect variant composition , and we will consider natural selection first . Natural selection , which acts by the fitness of phenotype , can be divided into the direct natural selection of non-neutral variants and the collateral removal of neutral variants , which depends on the linkage or the recombination rate between the selected non-neutral variant and nearby neutral variants . In the latter case , linkage-based selection models , such as selective sweep [3] and background selection [4] , has been used to explain the correlation between the recombination rate and variant distribution in light of their effect in removing more neutral variants in the regions of low recombination . These linkage-based models predict an aspect of the Hill-Robertson effect [29 , 30] . However , the Hill-Robertson effect , selective sweep , and background selection do not affect variant composition because linked neutral variants are removed regardless of their type ( e . g . SNP and indel ) . In other words , linkage-based selection models cannot explain the correlation between the recombination rate and variant composition . On the other hand , direct natural selection of non-neutral variants should affect variant composition . Here , exon density is a likely modifier . Notably , indels are more likely to be deleterious than SNPs in protein coding sequence [31] , and transversions are more likely to cause a non-conservative amino acid change because of the nature of the triplet code and wobble position . Indeed , the proportion of SNPs out of all variants is quite different among the variants that affect exons ( 84% ) as opposed to the variants that do not affect exons ( 72% ) ( Fig 2B and 2C , S11 Table ) . Likewise , transitions ( Ts ) are 64% of SNPs that affect exons and 54% of SNPs that do not affect exons ( Fig 2D and 2E ) . Using 58 physical-distance forward ( pf ) intervals , there is a clear difference ( two sample t-test , p < 0 . 001 ) in the proportions between these variant populations . The difference is even greater if the variants that affect CDS are examined instead ( SNPs = 89% of all variants , Ts = 66% of SNPs ) . In comparison , the variant composition is similar for the variants that affect introns but not exons and for the variants that affect neither coding sequences nor introns ( S3 Fig ) . Thus , direct natural selection appears to have a stronger effect on the variant composition in exons , for example , by preferentially removing indels over SNPs and Tv over Ts . We asked whether direct selection or exon density , which happens to be correlated to the recombination rate , is responsible for the correlation between the recombination rate and variant composition . A simple model would predict a uniform variant composition in exons and a different uniform variant composition outside exons . Such a model by itself does not predict a correlation between the recombination rate and variant composition for the population of variants that affect exons as well as for the population of variants that do not affect exons . However , strong correlation exists between the recombination rate and variant composition for the variants that do not affect exons ( S5 Table ) . Thus the action of direct natural selection simply according to the exon density cannot explain the correlation between the recombination rate and variant composition . Even in exons and CDS , where natural selection is expected to be a stronger force , mutation and other factors likely act in shaping the variant composition . Larger indels are fewer in number than smaller indels . For example , the number of indels of 2 base pairs or i2 is ~2 . 7-fold that of i3 , and the number of i3 is ~1 . 6-fold that of i4 ( S11 Table ) . However , an i3 in many cases should be less deleterious than an i1 , i2 , or i4 given the structure of the triplet code in RNA translation . Examining only the variants that affect CDS , the number of i3 ( n = 1072 ) is greater than the number of i2 ( n = 508 ) but smaller than the number of i1 ( n = 1583 ) ( S11 Table ) . Thus , natural selection is not the sole factor responsible for the presence of fewer larger indels . For example , mutation in a cell may be more prone to generate a SNP rather than i1 and more prone to generate an i1 than i2 or larger indels . Coordinated defense of a cell or an organism against mutation , such as DNA repair and apoptosis , may also contribute to the relative rarity of larger indels . Direct natural selection is likely a smaller factor in determining variant composition outside exons , and conversely mutation should be of greater importance outside exons . For direct natural selection to be responsible for the correlation between the recombination rate and variant composition , selection must act either according to the recombination rate itself or according to another genomic feature , which happens to be correlated to the recombination rate . Repetitive sequences are correlated with the recombination rate , but repetitive sequence density is unlikely to be the relevant factor because strong correlation exists between the recombination rate and variant composition for the variants that affect DNA outside repetitive sequences ( S9 Table ) . Genomic features such as CDS , exons of essential genes , highly conserved domains , and regulatory elements all may be affected differently by direct selection . However , we think that all of these genomic features are unlikely to explain the correlation between the recombination rate and variant composition . Alone by natural selection , we are unable to conceptualize how disparity in the variation composition can be generated in a genome according to the recombination rate , although we cannot definitively eliminate the existence of such a mechanism . Later in this manuscript , we will discuss how natural selection may modify a pre-existing correlation between the variant composition and recombination rate . In brief , natural selection affects variant composition except probably not de novo by the recombination rate . Mutation is a good candidate for the causative factor that is responsible for the correlation between the recombination rate and variant composition . However , we need a more comprehensive model than simply stating that high recombination rate causes high mutation rate . For a more comprehensive model , we classified all mutation mechanisms by their relationship to the recombination rate . The recombination rate of a genomic interval is determined by the genetic distance and physical distance of the interval . We suggest that all mutation mechanisms can be classified into two groups , one dependent on the relative genetic distance and the other dependent on the relative physical distance . Two general assumptions are that relative contributions of the mutation mechanisms are constant and that there are fixed probabilities for the generation of different mutation types . This mutation model is discussed in depth below . First , consider the mutation mechanisms that act according to the physical distance . Such a mutation mechanisms would generate three times as many variants in an interval of 3 megabase ( Mb ) compared to an interval of 1 Mb . A key underlying assumption is that the probability of mutation being generated is equal from one nucleotide to the next nucleotide . For example as a chromosome is replicated by a DNA polymerase , one nucleotide is as likely to become a site of mutation as the next nucleotide . The probability of mutation being generated at each base pair position during replication is assumed to be a constant . We named this group of mutation mechanisms as the Sanger mechanism for the critical role that the Sanger sequencing method played in determining the physical distance of DNA . We note that some chemical mutagens and irradiation may affect some nucleotide more than others , and homopolymers may be more liable to slippage resulting in a loss or a gain of base pairs . Such caveats aside , the number of variants generated by the Sanger mechanism for a genomic interval of a given physical length is defined as follows: #variantsbySangermechanism=Sangercoefficient ( S ) physicaldistance ( n ) Next , consider the mutation mechanisms that act according to the genetic distance . Such mechanisms would generate twice as many variants in an interval of 2 centimorgans ( cM ) compared to an interval of 1 cM . Using the strictest definition , the probability of mutation generation here depends on the probability of a base pair position becoming a site of meiotic chromosomal crossover . We call this group of mutation mechanisms as the Morgan mechanism . Unequal crossover of chromosomes leading to mutation has been known from the early days of modern biology [32] , and direct evidence for mutagenic effect of recombination was shown recently [33] . We hypothesize that the Morgan mechanism is not limited to mutations arising directly from errors in chromosomal crossovers . Specifically , chromosomal crossover is known to be preceded by programmed DNA double-strand breaks , which is catalyzed by Spo11 topoisomerase-like protein [34] . Meiotic double-strand breaks can be repaired through many different pathways with or without a chromosomal crossover [35] . Hotspots for double-strand breaks and the frequency of these hotspots may determine the recombination rate . Thus , a suitable definition of the Morgan mechanism may encompass mutations that are generated as a consequence of programmed DNA double-strand break that may lead to recombination . By this interpretation , the probability of meiotic DNA double-strand break is the primary contributor to a constant that we call the Morgan coefficient . The Morgan coefficient and the genetic distance of a given interval have the following relationship with the number of mutations generated by the Morgan mechanism: #variantsbyMorganmechanism=Morgancoefficient ( M ) geneticdistance ( g ) To formulate a mathematical model that combines the effect of the Morgan and Sanger mechanisms , the contribution of the two mechanisms needs to be normalized to the same time scale . Therefore , both Morgan and Sanger coefficients should reflect the number of potential mutation generation events in a single generation . For example , the probability of mutation generation by the Morgan mechanism should be modified by ten if there are ten programmed meiotic DNA double-strand breaks in the life cycle of an organism starting from a fertilized egg to the next fertilized egg . Replication of a 100 Mb C . elegans genome would involve 100 , 000 , 000 , 000 base pairs with a probability of mutation at each base pair , and the sum of the probabilities should be multiplied by the average number of cell cycles involved going from one generation to the next . The number of variants generated by Morgan and Sanger mechanism in a single generation is defined as follows: #ofvariantsgeneratedin1generation= ( Mg+Sn ) 1generation There is a fixed ratio that describes the relative relationship between the Morgan and Sanger coefficients . We call this the R coefficient: Rcoefficient=S/M ( fixedratio ) Relative contributions of the Morgan and Sanger mechanisms can be written as shown below using the R coefficient and without the S coefficient: #ofvariantsvo= ( r+R ) nMd See S1 Text for the algebra used to derive the equation above . Here , r is the recombination rate , which is the genetic distance ( g ) divided by the physical distance ( n ) . The R coefficient has the same units as the recombination rate . The factor d for divergence is the number of generations . The contribution of variant generation by the Morgan mechanism is r * n * M * d , and the contribution by the Sanger mechanism is R * n * M * d . We used uppercase letters for constants and lowercase letters for variables . We expect that a well-known population genetics term of mutation rate μ corresponds to the sum of ( r + R ) * n * M where r and n include all r and n values for all genomic intervals . A smaller R coefficient means that more contribution by Morgan mechanism and consequentially a higher density of variants in the intervals of high recombination rate . In brief , this equation can be used to examine the correlation between the recombination rate and variant distribution . To explain the correlation between the recombination rate and variant composition , we need to discuss the probability of generating a specific variant type . The key assumption here is that there is a fixed probability of a mutation generated being a specific mutation type for each mutation mechanism . Thus , we expect that Morgan and Sanger mechanisms would have different probabilities of generating a specific mutation type . The coefficient FM defines the probability of generating a specific variant type by a Morgan mechanism mutation generation . Likewise , FS defines the probability of generating a specific variant type from a Sanger mechanism mutation generation . An FM of 1 or 0 for SNP means that the Morgan mechanism generates SNP with a probability of 100% or 0% , respectively . The proportion of a specific variant type in an interval is: proportionoutofallvariantsfo= ( rFM+RFS ) / ( r+R ) . For each interval , the numerator ( r * FM + R * FS ) may correspond to the number of a specific variant type , and the denominator ( r + R ) may correspond to the number of all variants in the interval . The same equation can be used with denominator being something other than all variants ( e . g . SNPs or non-SNPs , see S1 Text ) . For the proportion of a specific variant type , a larger FM value than the FS value should lead to a positive correlation between the recombination rate and the proportion . On the other hand , a smaller FM value than the FS value should lead to a negative correlation , and equal values of FM and FS should lead to a zero correlation . While we started the mutation model without considering non-mutation factors , it may be worth discussing what else may affect the values of R and F coefficients . In our opinion , the values of both R and F coefficients can be affected by DNA repair and apoptosis , which may affect Morgan and Sanger mechanisms differently . Germline apoptosis in C . elegans is stochastic and is differently activated than somatic apoptosis [36] , and DNA damage-induced apoptosis involves a conserved genomic integrity checkpoint pathway [37] . It is possible that generation of a larger indel may be more likely to trigger apoptosis in a germ cell even if there is no selective advantage for a specific small indel over a specific large indel . It is also conceivable that the rate of triggering apoptosis in response to DNA damage is different for meiosis and mitosis . Given these scenarios , incorporating their effect on the R and F coefficients into the mutation model may be useful for DNA repair and apoptosis . Furthermore , natural selection can act in a similar manner to change the variant landscape , which we will discuss in more detail later ( see the last section of Discussion ) . In brief , our model invoking mutation and biological mechanisms that directly respond to the mutation can explain the correlation between the recombination rate and variant distribution as well as the correlation between the recombination rate and variant composition . Using the equation for the proportion of specific variant type , it should be possible to estimate the FM , FS and R coefficients given the recombination rate of each genomic interval and the corresponding proportion of a specific variant type in the interval . The values that best fit the observation is determined by using least squares ( LS ) regression analysis [38 , 39] . A simple analysis would use the recombination rate of each interval as defined , for example by WormBase [40] , and the corresponding proportion of each specific variant type using all variants . However , such a simple analysis ignores the stronger effect direct natural selection has on exons . To avoid a potentially large complicating effect of direct natural selection , the main analysis was done using only the variants that do not affect exons . A potentially complicating factor in analyzing only the variants that do not affect exons is that a substantial part of the physical length of the genome is occupied by exons . Notably , 17 . 2% of all variants affect CDS , and 24 . 1% of all variants affect exons ( S11 Table ) whereas CDS and exons occupy 25 . 1% and 31 . 4% , respectively , of C . elegans chromosomes ( S12 Table ) . Given that a large part of each genomic interval is being removed from the analysis , it may be worth asking whether the recombination rate of each genomic interval should be recalculated . Such a recalculation is possible only to a limited degree because genetic positions have been determined at the level of genes only rather than at the level of exons . For simplicity , we chose not to recalculate the recombination rate of each interval using a simplistic ( although not exactly correct ) assumption that exon density is even within each interval . With an even distribution of exons , we can assume that the genetic distance and the physical distance is removed in an equal and proportionate manner when exons are excluded from the analysis . To obtain meaningful estimates using the standard pf interval boundary definition alone , we used aggregated data from 37 wild isolates for our main analysis . Only 37 of the 40 individual isolates were used because a trio ( MY2 , MY14 and JU1171 ) and a pair of the isolates ( ED3057 and ED3072 ) are nearly identical throughout the entire genome , as also noticed by others [41]; ED3072 , JU1171 , and MY14 were removed from the analysis . Only genomic intervals with minimum of 300 variants were used . First , we show the analysis of a small set of the proportion of variant types with easily interpretable results . Using the proportions of SNPs out of all variants for the aggregation of individual isolates , FM and FS are 0 . 67±0 . 02 ( standard error of the estimate ) and 0 . 72±0 . 00 , respectively ( Fig 3B ) . This means that 67% of the mutations generated by the Morgan mechanism are SNPs whereas 72% of the mutations generated by the Sanger mechanism are SNPs . Thus , both mutation mechanisms produce primarily SNPs with a somewhat higher probability for the Sanger mechanism . Examination of yeast RAD52 recombinational repair mutants revealed a higher frequency of point mutations than other mutation types [42] , and thus the estimation that Morgan mechanism generates SNPs predominantly is not surprising . Using the proportions of i40-699 out of all variants , the estimated probabilities of generating i40-699 are 2 . 2±0 . 3% and 0 . 5±0 . 0% for the Morgan and Sanger mechanisms , respectively ( Fig 3A ) , which suggest that the Morgan mechanism is >4-fold more likely to generate an i40-699 than the Sanger mechanism . Meanwhile , the estimates of the R coefficient based on i40-699 and SNPs out of all variants are 9±3 and 8±5 , respectively . Using i40-699 out of non-SNPs and Ts out of SNPs , the R coefficient estimates are 8±2 and 8±5 , respectively ( Fig 3C and 3D ) . An R coefficient of 8 suggests that 26% of all the variants in C . elegans are generated by the Morgan mechanism in the wild environment . For a more complete analysis , we examined the correlation with the recombination rate using 58 definitions of the proportion of variant types . A wide range of the estimates of the R coefficient was obtained with a wide range of accompanying t and p values in our main analysis ( S13 Table ) . We also present the estimates derived from LS regression using only the variants that do not affect CDS as well as using all variants ( S14 and S15 Tables ) . We also present the estimates derived using a highly processed variant set , which involved combining overlapping variants and censoring of overshadowed variants ( S16 Table , see Methods ) . In addition , we present the estimates derived from the whole variant data , which produce estimates with poorer t and p values , rather than aggregated data from 37 wild isolates ( S17 Table ) . Here , better estimates need to be sorted from questionable estimates . The quality of the R and F coefficient estimates can be assessed by examining the associated t and p values . For example , an R coefficient value below zero is biologically irrelevant , and thus any estimate that suggests the possibility of the true R coefficient value below zero can be considered weak . Such weak estimates of the R coefficient can be censored by using the t value associated with the estimate of R . Using a cutoff of t > 1 , the R coefficient estimates range from 0 . 7 to 55 . 3 with a median value of 9 . 6 ( Fig 4A ) . Thirty-four of 56 ( 61% ) definitions of the proportion of variant types pass the t > 1 cutoff . We removed non-SNPs out of all variants and Tv out of SNPs from the consideration because these definitions gave identical R coefficient estimates as SNPs out of all variants and Ts out of SNPs , respectively . The p value , which has an inverse relationship with the t value , of the estimates of the R coefficients also can be used as a cutoff . Using p-value based cutoffs , the median R coefficient value is 9 . 8 with p < 0 . 05 ( n = 23 ) and 9 . 2 with p < 0 . 01 ( n = 17 ) ( Fig 4A , S13 Table ) . Using Benjamini-Hochberg method [43] , p ~ 0 . 08 for the R coefficient was considered significant using false discovery rate of 0 . 25 whereas p ~ 0 . 015 was considered significant using false discovery rate of 0 . 05 . Using the variants that do not affect CDS or using all variants instead , the median estimates of the R coefficient change only slightly ( S14 and S15 Tables , median R coefficient is 10 . 8 and 10 . 6 , respectively , using p < 0 . 05 cutoff ) . Given these estimates and their accompanying supporting statistic values , it seems reasonable to propose an approximate R coefficient value of 10 for C . elegans . The estimates of FM and FS coefficients presuming a uniform R coefficient value ( e . g . R = 10 ) are presented in S18 Table . An R coefficient of 10 suggests that 22% of all the variants in C . elegans are generated by the Morgan mechanism in the wild environment . It is possible that the proposed R coefficient value of 10 is too big . With additional processing of overlapping variants and censoring of overshadowed variants ( see Methods ) , the median R coefficient estimate is 5 . 5 using the variants that do not affect exons ( p < 0 . 05 cutoff , S16 Table ) . We reasoned that a range of acceptable R coefficient value also could be obtained by examining the quality of FM and FS coefficient estimates from LS regression analysis using a presumed the value of the R coefficient . By definition , the FM and FS coefficients must be in a range between 0 and 1 . Thus , we declared that a presumed R coefficient value that suggests an FM or FS coefficient value outside this range is incorrect . We examined the estimates of the FM and FS coefficients using presumed R coefficient values between 0 . 01 and 9900 . Here , an upper limit of 57 for the R coefficient is suggested by the analysis of indels of 1 base change ( i4 ) out of non-SNPs whereas a lower limit of ~0 . 9 is suggested by i40-699 out of all variants ( S4 Fig , S13 Table ) . By this criterion , 0 . 9 and 57 can be considered the lower and upper limits , respectively , of the R coefficient in C . elegans . This range is compatible with the proposed R coefficient value of 10 as well as the alternate R coefficient value of 5 . 5 , which suggests a 34% contribution by the Morgan mechanism . Variant distribution can be analyzed using the mutation model , and LS regression analysis should allow a different method of estimation of the R coefficient as well as estimation of the product of the Morgan coefficient and the divergence factor ( M * d ) . Again , our standard analysis uses the variants that do not affect exons . Here , the estimate of the R coefficient is 3 . 5±0 . 9 ( Fig 5A , S19 Table ) , and the estimate of M * d is 1500±200 . This R coefficient estimate of 3 . 5 is accompanied by a large t value ( t = 3 . 9 ) and a small p value ( p < 0 . 0002 ) , which indicate a higher degree of confidence than the R coefficient estimates derived from the analysis of variant composition . Using all variants and the variants that do not affect CDS instead , R coefficient estimates are 3 . 3±1 and 3 . 5±0 . 9 ( S19 Table ) . The DNA length of the interval was adjusted when using a subset of variants by subtracting the length of DNA occupied by exons or CDS . Processing of overlapping variants and censoring of overshadowed variants were not performed for the analysis of variant distribution . Notably , this R coefficient estimate of 3 . 5 is considerably smaller than the proposed R coefficient value of 10 derived from the analysis of variant composition . R coefficients of 3 . 5 and 10 predict up to a 3 . 8-fold and 2-fold difference , respectively , in the variant distribution between the regions of high and low recombination ( Fig 4B ) . The discrepancy in the R coefficient estimates between the analysis of variant distribution and the analysis of variant composition can be reconciled . Earlier , we mentioned that linkage-based selection models could also explain the skewing of distribution of variants according to the recombination rate . This means that a proper analysis of the correlation between the recombination rate and variant distribution must account for both mutation and the Hill-Robertson effect in general ( e . g . through background selection or selective sweep ) . On the other hand , the Hill-Robertson effect can be ignored in the analysis of variant composition . This means that an R coefficient of 10 from the analysis of variant composition reflects the action of mutation only whereas an R coefficient estimate of 3 . 5 from the analysis of variant distribution reflects the combined action of mutation and selection . A more complete depiction of how the variant landscape is shaped is possible by combining our mutation model with a linkage-based selection model that also affects variant distribution according to the recombination rate . We chose background selection for a demonstration here because background selection is generally assumed to have a greater effect than selective sweep in C . elegans . A combined model containing both the background selection and our mutation model can be made by a coupling of the equations from the two models as follows: vb= ( r+R ) dnMexp ( -U/ ( sd+r* ( 1-F ) ) ) / ( 1+F ) . Here , mutation model equation vo = ( r + R ) * d * n * M was used to substitute the number of variants without background selection ( vo ) in the background selection equation . Here , vb is the number of variants after background selection , U is the deleterious mutation rate , sd is the average selection coefficient , and F is the inbreeding coefficient , with a relationship to the outcrossing rate , c: F = ( 1-c ) / ( 1+c ) [4 , 7] . Both sd and F coefficients should have values between 0 and 1 . Because background selection does not affect variant composition , the R coefficient estimate from the analysis of variant composition should be used here . Others have determined some of the values used in the background selection model . The deleterious mutation rate ( U ) in C . elegans has been measured by many approaches with a wide range from 0 . 003 to 0 . 48 , and 0 . 48 seems to be the current accepted value [44–49] . The rate of outcrossing for wild C . elegans has been estimated in a range between ~1% and 1 . 7% previously [50–52] , except for an outlier estimate of 22% [53] . Thus , we performed LS regression analysis using presumed values of 0 . 48 and 1 . 7% for the U and outcrossing rate , respectively . The estimates of the sd selection coefficient vary depending on the assumed values of the R coefficient and M * d ( Fig 5C ) . Here , our proposed R coefficient value of 10 was used alongside alternate values of 3 . 5 and 60 . The estimates of sd change depending on the assumed value of M * d , and the range of acceptable values of M * d also changes depending on the presumed R coefficient value . The estimates of sd with the best t and p values were between 0 . 2 and 0 . 29 using presumed R coefficient values of 3 . 5 , 10 and 60 . These best sd and M * d values were used along with the presumed R coefficient values to generate the lines of best fit between the recombination rate and the distribution of variants ( Fig 5B ) . We suspect that bigger M * d values associated with larger presumed R coefficients simply reflect a bigger value of the M coefficient rather than a bigger divergence d value . Examining the sum of squares , the pseudo-R2 values are similar between the different presumed values of the R coefficient ( R2 = 0 . 33 for R of 3 . 5 , R2 = 0 . 43 for R of 10 , R2 = 0 . 46 for R of 60 ) . In comparison , R2 is 0 . 48 using the mutation model alone with the R coefficient of 3 . 5 . Our opinion is that the differences in the R2 values are too small to be used to pick the value of the R coefficient . Nevertheless , these analyses using the combined model allow a fuller exploration of how variant landscape have been generated than by using a single model . If mutation can generate a disparity in the landscape of variant composition according to the recombination rate , then natural selection theoretically can amplify or reduce the disparity by a simple mechanism . Take a scenario where two genomic regions accumulate different numbers of variants in the absence of selection ( e . g . region A with 500 indels and 500 SNPs and region B with 20 indels and 80 SNPs ) . For simplicity , all variants are outside exons . Now , extend the scenario so that selection removes a higher percentage ( e . g . 50% ) of the indels and a lower percentage ( e . g . 0% ) of the SNPs . Here , selection has changed the variant composition ( and distribution ) to a different degree for the two regions , and the recombination rate is one of the underlying factors . In a simulation using the pf boundary definition , the real R coefficient value of 10 , FM for SNP of 0 . 8 , and FS for SNP of 0 . 7 without selection , introducing the action of selection that removes 50% of the indels and none of the SNPs would change the variant composition so that LS regression analysis would generate an apparent R coefficient estimate of 10 . 6 instead of 10 . Selection is likely to remove indels more than SNPs and larger indels more than smaller indels , and FM values are larger than FS values for most indel types for C . elegans . Thus , selection in most cases would reduce rather than amplify the difference in the variant composition between the regions of low and high recombination . This means that our R coefficient estimates probably underestimate the real contribution of the Morgan mutation mechanism . We suspect that selection would discriminate between indels and SNPs to a lesser degree than in the example given here ( i . e . 50% versus 0% ) , and thus the actual effect on the estimation of the R coefficient is probably smaller . As discussed earlier , DNA repair and apoptosis in direct response to mutation can also affect variant composition in a similar manner . Further improvements in the estimation of the R and other coefficients may be possible through data aggregation using multiple interval boundary definitions ( S12 Table ) . Here in addition to the standard pf boundary definition , we used three other definitions ( pr for physical reverse , gf and gr for genetic forward and reverse , which are 5 cM-based intervals ) . The estimates of R coefficient from variant composition analysis fluctuate considerably with different interval boundary definitions ( S13–S17 Tables ) . This effect of the genomic interval boundary choice is considerably larger than the effect of removing variants affecting exons or CDS from the analysis . For example using the variants that do not affect exons in the analysis of variant composition , the median R coefficient estimates are smaller using the other three interval boundary definitions ( S13 Table , median R = 9 . 8 , 9 . 1 , 5 . 2 , and 3 . 1 with a cutoff of p < 0 . 05 using pf , pr , gf , and gr interval boundary definitions , respectively ) . After aggregating the data using all four boundary definitions , the median R coefficient estimate is 7 . 2 , which suggests a 32% contribution by the Morgan mechanism . It should be noted that the associated t and p values obtained using aggregated boundary definitions should be treated differently than when single boundary definition is used . Notably in the analysis of variant distribution , the R coefficient estimate also drops to 2 . 5 with the aggregated data using all four boundaries ( S19 Table ) . Thus , a better R coefficient value may be 7 rather than 10 , and a better estimate may be possible by more aggregation . Errors in sequencing analysis , mutation detection , and mutation annotation are a potential problem in the analysis . More sequencing data , which is exemplified by a higher-quality sequencing data in the recently published genome of CB4856 [41] , may help , but more important probably is a refinement in variant calling and better reconciliation of conflicting variant calls . Sequencing analysis can be problematic for highly repetitive sequences , which are enriched in the regions of high recombination . Homopolymer runs pose a different problem , and it is worth noting that Million Mutation Project ( MMP ) observed a more even distribution of mutations across the genome in artificially mutagenized C . elegans when homopolymer runs were excluded from the analysis [10] . Homopolymers presumably disproportionately affect variant calling of very small indels , and better estimates from LS regression analysis may be possible after accounting for homopolymers . Perhaps the most important factor is a large number of apparent conflicts in the variant annotation , which can be described as overlapping and overshadowed variants ( see Methods ) . These conflicts appear to be a consequence of not reconciling the results of multiple independent mutation detection methods . From the analysis of variant composition using all four boundary definitions after accounting for these conflicts , we obtained a median R coefficient value of 4 . 8 ( S16 Table ) , which suggests a 42% contribution by the Morgan mechanism . Other improvements in the analysis , such as a better calculation of the recombination rate of the interval when exons are removed from the analysis , may also be useful . Our mutation model suggests that there are more variants in autosomal ends and in regions of high recombination in C . elegans in part because Morgan mechanism is a significant source of mutation generation . Perhaps Morgan mechanism plays a big part in mutation generation because double-strand breaks are inherently dangerous and liable to cause mutations . The dangers of breaking DNA , even when done deliberately , may also explain the higher proportion of larger indels associated with the Morgan mechanism . Alternatively , the fidelity of DNA replication may be more important for somatic cells than germ cells , and generating mutations in germ cells may facilitate evolution with minimal disruption in the development and physiology of individual organism . Similarly , larger indels may constitute a higher proportion of variants in the regions of high recombination in part because larger indels are more tolerated by DNA repair and apoptosis during meiosis . A potential argument against our proposed R coefficient values between 4 . 8 and 10 comes from the analysis of mutation-accumulating ( MA ) strains [16 , 17] , which show a distribution of variants that cannot be reconciled with >2-fold difference between the regions of high and low recombination ( Fig 4B ) . When the variant distribution in the MA strains is examined , the R coefficient estimate is 64±101 ( p = 0 . 53 ) using the standard pf boundary definition and 35±18 ( p = 0 . 05 ) using all four boundary definitions together . These R coefficient values suggest a 4% or 9% contribution by the Morgan mechanism , respectively . Meaningful analysis is not possible with variant composition in MA strains because the total number of MA variants is very small ( n = 391 ) . The authors of the MA strains suggested that the unequal distribution of variants in the wild population is likely a result of natural selection and not mutation . However , the generation of mutations in MA strains perhaps does not accurately reflect the generation of mutations in the wild environment , which is a possibility also suggested by the authors of the MA strain studies [16 , 17] . Perhaps laboratory growth condition dramatically increases Sanger mechanism or decreases Morgan mechanism in C . elegans . If this is true , mutation-accumulating strains grown in laboratory may be irrelevant for studying the relative importance of Morgan and Sanger mutation mechanisms in the wild environment . In this study , we showed that examining the composition of variants has a potential to reveal many interesting facets of molecular evolution . The proportions of many variant types in different genomic intervals show strong correlation with the recombination rate . By separating mutation mechanisms conceptually according to those dependent on genetic and physical distances , our mutation model describes how the composition of variant as well as the distribution of variants in the genome may become uneven throughout the C . elegans genome . We used this mutation model to systemically and holistically estimate the probabilities of generating specific mutation types by the putative Morgan and Sanger mechanisms in C . elegans . Since this mutation model is compatible with existing natural selection models , such as background selection , a more comprehensive analysis of genetic diversity is now possible . We think that this new metric of the proportion of variant types together with our mutation model have a potential to be generally useful in the analysis of genetic diversity in other species . It would be especially interesting to see if a substantial contribution of Morgan mechanisms in mutation generation is the rule rather than an exception when variant composition is examined in other species including in humans . For computational processing and analysis , custom scripts were written and executed using R [54] and R Studio . The R package minpack . lm by Andrej-Nikolai Spiess and Katherine M . Mullen was used to perform LS regression analysis using Levenberg-Marquardt algorithm . The R package seqinr maintained by Simon Penel was used to determine GC content . The R scripts along with source and output files , which are needed to execute the scripts , are available at Dryad digital repository [55] . The C . elegans variant data and genomic features were obtained using the newest version WS256 of WormBase ( http://www . wormbase . org ) [40] . Most of the data were obtained using the WormBase Genome Browser GBrowse ( http://www . wormbase . org/tools/genome/gbrowse/c_elegans_PRJNA13758/ ) . Specifically , the variants were obtained by using Polymorphisms track of GBrowse . Other GBrowse tracks used were: Curated Genes , Curated Genes ( Protein-coding ) , Curated Genes ( noncoding ) , RNAseq for expression level , Tandem and Inverted Repeats for repetitive sequences , and various Histone Modifications ChIP-Seq ( H3K4me1 , H3K36me1 , H3K9me1 , H3K27me1 , and H3K27me3 ) for chromatin state . GFF3 files were obtained using GBrowse in all cases except WIG ( signal ) files were obtained for chromatin state . Curated Genes contain the details on genetic and physical location as well as exons and introns , and Curated Genes ( protein-coding ) and Curated Genes ( noncoding ) contain the information specific to coding DNA sequences and non-coding RNA , respectively . ChIP-seq and RNAseq data were generated by the modENCODE project [56] . The WS256 version PRJNA13758 of the C . elegans genomic DNA sequence [27] was obtained using the WormBase FTP site ( ftp://ftp . wormbase . org/pub/wormbase/species/c_elegans/sequence/genomic/ ) . To obtain a list of essential genes , we used the queries lethal , sterile , larval lethal , larval arrest , embryonic lethal , and zygotic lethal as mutant phenotypes using WormMine version WS253 ( http://intermine . wormbase . org/tools/wormmine/begin . do ) . We use four sets of interval boundary definitions ( S12 Table ) . Two sets are organized by one megabase ( Mb ) , and two other sets are organized by five centimorgans ( cM ) . The boundaries are set starting from either the left or the right telomere of the chromosome . The final intervals at the chromosomal ends are less than 1 Mb or 5 cM of DNA , except any interval of smaller than 0 . 2 Mb is joined with a neighboring interval to minimize noise . The intervals from the different boundary definitions do not share common end points except for the telomeres . Total of 102 and 58 intervals exist according to the boundary definitions using 1 Mb and 5 cM , respectively . The total exon space was calculated after accounting for duplicates and overlaps . Specifically , duplicate exons that share the same 5' and 3' ends were removed . Overlapping exons were combined into one exon . Any exon that is located completely within another exon was removed . The same process was used to calculate the genomic space occupied by essential genes , the genomic space occupied by the exons of essential genes , and the genomic space occupied by repetitive sequences . For introns , an additional process of subtracting the space occupied by exons within introns was performed . The density of these elements is defined as the physical length of space occupied by the element ( e . g . exons ) divided by the physical length of the genomic interval . RNAseq GFF3 file was used to calculate expression level , and WIG files were used to measure chromatin state . Briefly , we multiplied the interpreted depth of coverage by RNAseq by the length of each RNAseq reads . To obtain normalized expression level for each genomic interval , we then divided the sum of these products by the total length of the genomic interval . Similarly , the sum of the levels of methylation state for 10 base pair stretches in WIG files were obtained for each genomic interval . To obtain normalized chromatin state level for each genomic interval , we divided this sum by the number of the 10-base pair stretches corresponding to each genomic interval . The variant data obtained from WormBase using the Polymorphisms track of GBrowse contain a great deal of information . The information includes the start and end positions of the variants relative to the N2 reference genome . Other information includes a list of wild isolates with the variant and the consequence of the variant on the protein coding sequence . We focused solely on 40 wild isolates out of hundreds of C . elegans wild isolates with variant data . The reason for choosing these 40 isolates was that these 40 have been sequenced via whole-genome sequencing by the Million Mutation Project ( MMP ) and thus have the most complete data . The 869 , 019 variants in the 40 wild isolates account for over 97% of all natural variant records in the WS256 WormBase polymorphism data . Four different methods were used by the MMP to identify variants [10] , specifically a process using the standard suite of SAMtools software ( mpileup , bcftools , vcfutils . pl ) , a two-step process involving scanning of mpileup for "gapped" reads to identify small indels ( <200 bp ) , a process involving examination of "split" reads generated by phaster to identify indels of 100–5000 bp , and a process of examination of variations or changes in read coverage to identify larger copy number variants . The MMP-specific variant dataset is available separately ( http://genome . sfu . ca/mmp/ ) , but the physical position of the variants in the MMP curation often does not perfectly match the reference genome position provided by WormBase . Many of the variants annotated by WormBase were identified over a long period of time for some of the 40 wild isolates . Other methods of variant identification included light shotgun sequencing [21 , 57] and oligo array comparative genome hybridization [22 , 58 , 59] . The WormBase GFF3 annotation includes the laboratory source of the variant data , which is only useful for distinguishing the variants from light shotgun sequencing from all other methods because the method of variant identification is not specified for each variant . We found many inconsistencies of annotation in the WormBase curated variants , including adjacent , overlapping , and overshadowed variants . For example , substitutions of two or more adjacent base pairs are often not curated as a single variant even in cases of the adjacent variants being present in an identical set of wild isolates . Combining such adjacent variants seems absolutely warranted . More complicated are a small number of variants occupying an overlapping space in the genome . Arguably the most consequential are a large number of variants that are completely enveloped , or overshadowed , by a large deletion . Overlapping and overshadowing of variants may stem from complex genome rearrangement , such as a combination of deletion ( s ) and duplication ( s ) . Another possible source of inconsistency is the differences in sample sources , but we found many similar inconsistencies using only the MMP data . Therefore , the main problem appears to be different results derived from multiple independent variant calling methods and a lack of a thorough and holistic reconciliation of the conflicting results . After combining all adjacent variants that are present in identical set of wild isolates , the total number of variants was reduced from 869 , 019 to 853 , 815 . Our main analysis was performed using these 853 , 815 processed variants either as a whole or as an aggregation in 37 of the 40 wild isolates . For further processing involving overlapping and overshadowed variants , we did not require that the variants of interest are present in identical set of wild isolates . Overlapping and overshadowing variants were processed at the level of individual wild isolates only , and these more processed variants were used for supplementary analysis . Overlapping variants were combined into a single variant , and overshadowed variants were censored . In part because of the presence of some very large indels including 144 variants of >100 , 000 base pair deletions ( including a deletion of >1 . 4 Mb ) , this process reduced the number of variants by 11% to 28% depending on the wild isolate ( median = 19% ) . The location of the variants was annotated relative to exons , introns , CDS , ncRNA , and repetitive sequences . The results of this annotation at the whole-genome level are summarized in S11 Table . Variants are counted as affecting a genetic feature ( e . g . exon ) if at least one base pair of the genetic feature was changed . Variants were counted as being inside a genetic feature only if no DNA outside the genetic feature was affected . Similarly , variants were counted to be within an interval only if both the starting position and the end position of the variant lie within the defined interval . By our definition , the variants that do not affect exons are the variants that do not affect any of the following genetic features: coding DNA sequence ( CDS ) , 5' UTR , 3' UTR , pseudogene , and non-coding RNA ( i . e . miRNA , tRNA , rRNA , snRNA , snoRNA , piRNA , lincRNA , scRNA ) . CDS is the sole consideration for annotation of the variants that do not affect CDS . WormBase annotations were used to check our annotation results that were examined relative to CDS ( n = 146 , 746 ) . All but 70 variants that we annotated as affecting CDS had WormBase annotation term = Nonsense , = Frameshift , = Silent , = Coding_exon , or = Readthrough . In the case of these 70 exceptions , WormBase used the annotation term = Splice_site instead . The conflict is there because WormBase used a shorter exon rather than a larger exon in these 70 cases . Conversely , our annotation of the variants that affect CDS missed 26 variants that WormBase annotated with the terms = Frameshift or = Coding_exon . Of these 26 , WormBase correctly annotated 14 insertions affecting splice site as = Frameshift . WormBase annotation looked incorrect to us in the other 12 cases .	DNA variants in the world population of a species reflect the genetic diversity of the species . While examining variants in whole-genome sequenced wild isolates of the nematode worm Caenorhabditis elegans , we discovered apparent correlations between the recombination rate and the proportion of many variant types . To explain this correlation , we present a model of a concerted action of two groups of mutation mechanisms , which act according to different components of the recombination rate . This model can also explain how mutation mechanisms as a whole can affect the genomic landscape of the variant distribution . Using this mutation model , we systemically and holistically estimate the probabilities of generating specific mutation types by distinct groups of mutation mechanisms in C . elegans . Since this mutation model can be mathematically combined with existing natural selection models , such as background selection , a more comprehensive analysis of genetic diversity is now possible . We expect that the mathematical equations we present here can be used for refining computer simulations of evolution and coalescent modeling . Within the context of this mutation model , the correlation between the recombination rate and the proportion of variant types may serve as a useful new metric for analysis of genetic diversity in other species .	[ "Abstract", "Introduction", "Results", "and", "Discussion", "Methods" ]	[ "invertebrates", "caenorhabditis", "invertebrate", "genomics", "animals", "animal", "models", "regression", "analysis", "mutation", "caenorhabditis", "elegans", "model", "organisms", "mathematics", "statistics", "(mathematics)", "experimental", "organism", "systems", "dna", "recombination", "dna", "research", "and", "analysis", "methods", "sequence", "analysis", "bioinformatics", "mathematical", "and", "statistical", "techniques", "statistical", "methods", "animal", "genomics", "biochemistry", "dna", "sequence", "analysis", "nucleic", "acids", "natural", "selection", "linear", "regression", "analysis", "database", "and", "informatics", "methods", "genetics", "nematoda", "biology", "and", "life", "sciences", "physical", "sciences", "genomics", "evolutionary", "biology", "evolutionary", "processes", "organisms" ]	2017	Effect of mutation mechanisms on variant composition and distribution in Caenorhabditis elegans
Nitric oxide ( NO ) regulates neuronal function and thus is critical for tuning neuronal communication . Mechanisms by which NO modulates protein function and interaction include posttranslational modifications ( PTMs ) such as S-nitrosylation . Importantly , cross signaling between S-nitrosylation and prenylation can have major regulatory potential . However , the exact protein targets and resulting changes in function remain elusive . Here , we interrogated the role of NO-dependent PTMs and farnesylation in synaptic transmission . We found that NO compromises synaptic function at the Drosophila neuromuscular junction ( NMJ ) in a cGMP-independent manner . NO suppressed release and reduced the size of available vesicle pools , which was reversed by glutathione ( GSH ) and occluded by genetic up-regulation of GSH-generating and de-nitrosylating glutamate-cysteine-ligase and S-nitroso-glutathione reductase activities . Enhanced nitrergic activity led to S-nitrosylation of the fusion-clamp protein complexin ( cpx ) and altered its membrane association and interactions with active zone ( AZ ) and soluble N-ethyl-maleimide-sensitive fusion protein Attachment Protein Receptor ( SNARE ) proteins . Furthermore , genetic and pharmacological suppression of farnesylation and a nitrosylation mimetic mutant of cpx induced identical physiological and localization phenotypes as caused by NO . Together , our data provide evidence for a novel physiological nitrergic molecular switch involving S-nitrosylation , which reversibly suppresses farnesylation and thereby enhances the net-clamping function of cpx . These data illustrate a new mechanistic signaling pathway by which regulation of farnesylation can fine-tune synaptic release . Throughout the central nervous system ( CNS ) , the volume transmitter nitric oxide ( NO ) has been implicated in controlling synaptic function by multiple mechanisms , including modulation of transmitter release , plasticity , or neuronal excitability [1–3] . NO-mediated posttranslational modifications ( PTMs ) in particular have become increasingly recognized as regulators of specific target proteins [4] . S-nitrosylation is a nonenzymatic and reversible PTM resulting in the addition of a NO group to a cysteine ( Cys ) thiol/sulfhydryl group , leading to the generation of S-nitrosothiols ( SNOs ) . In spite of the large number of SNO-proteins thus far identified , the functional outcomes and mechanisms of the underlying specificity of S-nitrosylation in terms of target proteins and Cys residues within these proteins are not clear . Synaptic transmitter release is controlled by multiple signaling proteins and involves a cascade of signaling steps [5] . This process requires the assembly of the soluble N-ethyl-maleimide-sensitive fusion protein Attachment Protein Receptor ( SNARE ) complex and associated proteins , the majority of which can be regulated to modulate synapse function . Regulatory mechanisms include phosphorylation of SNARE proteins [6] as well as SNARE-binding proteins such as complexin ( cpx ) , which have been reported at different synapses such as the Drosophila neuromuscular junction ( NMJ ) [7] or in the rat CNS [8] . Several contrasting effects on transmitter release are induced by NO-mediated PTMs [9] . Other forms of protein modification to modulate cellular signaling include prenylation , an attachment of a farnesyl or geranyl-geranyl moiety to a Cys residue in proteins harboring a C-terminal CAAX prenylation motif . This process renders proteins attached to endomembrane/endoplasmic reticulum ( ER ) and Golgi structures until further processing , as shown for Rab GTPases [10–12] . Farnesylation also regulates mouse cpx 3/4 [13] and Drosophila cpx function [14–17] . The Cys within CAAX motifs can also undergo S-nitrosylation , which interferes with the farnesylation signaling [18]; however , direct evidence in a physiological environment is lacking . Cpx function has been studied in many different systems and there is controversy regarding its fusion-clamp activity . Cpx supports Ca2+-triggered exocytosis but also exhibits a clamping function [19–24] . Analysis of mouse cpx double-knockout neurons lacking cpx 1 and 2 found only a facilitating function for cpx on release , and different D . melanogaster and Caenorhabditis elegans cpx mutant lines exhibit altered phenotypes in clamping or priming/fusion function [14–17 , 24–27] , illustrating the controversial actions of cpx . Here , we investigated the effects of NO on synaptic transmission and found that NO reduces Ca2+-triggered release as well as the size of the functional vesicle pool , which was reversed by glutathione ( GSH ) signaling . At the same time , spontaneous release rates were negatively affected by NO . We confirmed that cpx is S-nitrosylated and that NO changes the synaptic localization of cpx , as also seen following genetic and pharmacological inhibition of farnesylation . Thus , we propose that the function of cpx is regulated by S-nitrosylation of Cys within the CAAX motif to prevent farnesylation . This increases cpx-SNARE-protein interactions , thereby rendering cpx with a dominant clamping function , which suppresses both spontaneous and evoked release . Previously , we found that enhancing endogenous nitric oxide synthase ( NOS ) activity induced by overexpression of D . melanogaster NOS ( DmNOS ) caused a reduction in synaptic strength at the Drosophila NMJ synapse [28] . To examine the effects of NO on glutamatergic transmission in more detail , we exposed wild-type ( WT ) w1118 control ( Ctrl ) larvae to NO donors , which provide an estimated NO concentration of about 200 nM [29] . When recording evoked excitatory junction currents ( eEJCs ) up to 70 min during NO incubation , the amplitudes started to decline significantly after 35 min ( Fig 1A and S1 Data , p < 0 . 05; n = 3 each ) . Mean eEJC amplitudes and quantal content ( QC ) at 50 min for Ctrl ( 122 ± 7 nA , QC: 200 ± 15 , n = 20–22 ) and NO treatment ( 59 ± 7 nA , QC: 93 ± 10 , n = 14 ) are shown in Fig 1B . As the canonical NO-cGMP pathway is active in Drosophila [30] and potentially responsible for this observation , we blocked the soluble guanylyl cyclase ( sGC ) with 1H-[1 , 2 , 4]oxadiazolo[4 , 3-a]quinoxalin-1-one ( ODQ , 50 μM ) . Interestingly , ODQ did not prevent the effects of NO , suggesting a cGMP-independent mechanism ( amplitudes: Ctrl + ODQ: 127 ± 5 nA , NO + ODQ: 70 ± 7 nA , QC: Ctrl + ODQ: 200 ± 22 , NO + ODQ: 130 ± 11 , Fig 1B , n = 10–16 ) . As Drosophila has endogenous NO signaling and produces neuronal NO in a Ca2+/calmodulin-dependent manner [31 , 32] , we used NOS knockout-like ( NOS “null” ) larvae to assess endogenous NO modulation of release . We used two different lines with strongly reduced DmNOS showing NOS “null” activity ( NOSC and NOSΔ15 [33 , 34] ) and we would expect that lack of endogenous NO generation has the opposite effects on release . When recording eEJCs , both genotypes exhibited a tendency towards larger eEJC amplitudes and QC ( Fig 1C ) and , in addition , we detected an increased presynaptic release probability ( pvr ) in NOSC NMJs , as indicated by the reduced paired pulse ratio ( PPR ) at 20 ms ISI ( 0 . 80 ± 0 . 03 [n = 11] , p = 0 . 002 , Student t test ) compared to WT Ctrls ( 0 . 93 ± 0 . 03 [n = 17] ) , indicating endogenous nitrergic effects on release probabilities . To further understand the effects of NO on release , we analyzed miniature EJCs ( mEJCs ) under the same conditions . NO had no effect on mEJC amplitudes or decay kinetics; however , the frequency was reduced following NO and NO+ODQ incubation ( Ctrl: 2 . 0 ± 0 . 2 nA [n = 25] , NO: 1 . 1 ± 0 . 1 nA [n = 16] , NO+ODQ: 1 . 0 ± 0 . 2 nA [n = 8] , ODQ: 1 . 7 ± 0 . 2 nA [n = 11] , Ctrl versus NO: p < 0 . 01 , Ctrl versus NO+ODQ: p < 0 . 05 , Fig 1D ) . This suggests that NO is unlikely to affect synaptic vesicle filling or composition/activity and density of postsynaptic D . melanogaster glutamate receptors ( DmGluR ) [35] . We tested miniature events in the NOS “null” mutants and confirmed a further inhibitory role of NO signaling on release , with mEJC frequencies being significantly enhanced in NOSΔ15 ( 3 . 5 ± 0 . 5 s−1 [n = 4] , p = 0 . 001 ) and NOSC ( 3 . 5 ± 0 . 4 s−1 [n = 16] , p = 0 . 04 ) larvae compared to Ctrl ( Fig 1E ) , without affecting mEJC amplitudes ( NOSΔ15: 0 . 8 ± 0 . 1 nA [n = 13] , NOSC: 1 . 1 ± 0 . 3 nA [n = 3] Fig 1E ) or decay kinetics ( NOSΔ15: 8 . 9 ± 0 . 6 ms [n = 12] , NOSC: 9 . 4 ± 0 . 3 ms [n = 4] , p > 0 . 05 versus Ctrl ) . Thus , reduction of endogenous NOS activity shows opposite effects to elevation of NO levels , confirming the inhibitory action of NO on evoked and spontaneous vesicle release . As the data imply cGMP-independent signaling , we wanted to confirm that cGMP levels are not altered following NO stimulation . Thus , we measured cGMP directly in isolated larval brains . NO application did not raise cGMP levels ( at 50 min: Ctrl: 2 . 4 ± 0 . 5 pmol/mg , NO: 3 . 0 ± 0 . 6 pmol/mg , p > 0 . 05 [n = 30 each] , Fig 1F ) . Cyclase inhibition in the presence of NO did not significantly reduce cGMP levels , confirming lack of NO-induced neuronal cGMP accumulation . We found that any generated cGMP was broken down by phosphodiesterase DmPDE5/6 [36] , as cGMP increased following NO stimulation only with PDE inhibition ( 20 μM zaprinast [Zap]; NO+Zap: 50 . 2 ± 8 . 3 pmol/mg , p < 0 . 0001 ) , while Zap alone had no effect ( Zap: 4 . 6 ± 2 . 0 pmol/mg , p > 0 . 05 ) . To assess whether NO is produced endogenously to induce modulation of synaptic function as observed above , we expressed FlincG3 presynaptically and stimulated NMJs at 20 Hz ( for 10 s every minute for 20 min ) . As shown in Fig 1G , 20 Hz stimulation induced a significant increase in fluorescence , confirming endogenous presynaptic generation of NO ( Ctrl: 62 ± 4 arbitrary units [a . u . ’s] , Stim: 96 ± 8 a . u . ’s , Fig 1H [n = 13–15 boutons] , p < 0 . 01 ) . Importantly , addition of the NO donor did not further increase the fluorescence , indicating that activity-induced synaptic NO concentrations reach similar levels ( NO: 93 ± 7 a . u . ’s ) . A potential target of NO signaling is mitochondria [37] , which are required for the energy to maintain vesicle recycling and synaptic transmission [38] . Thus , we measured mitochondrial activity in third instar larvae under the same conditions ( 50 min NO incubation ) and found that mitochondrial activity was unaffected by NO ( S1 Fig and S9 Data ) , suggesting that the effects of NO on synaptic transmission are not due to ATP depletion . Together , these data suggest that NO has a presynaptic effect on transmitter release , which is independent of cGMP signaling . Several mechanisms contribute to the regulation of synaptic strength [39] , including altered pvr , alterations in the number of readily releasable vesicles and release sites ( N ) or quantal size ( q ) . Alterations in q are likely not involved in the NO-induced effects observed based on our mEJC data above ( Fig 1 ) . We next assessed additional release parameters , including pvr , N , vesicle pool size , and Ca2+ dependency of release in NOS “null” and WT NMJs following nitrergic signaling . We determined pool size via a method successfully applied at the Drosophila NMJ , by analyzing the cumulative QC of trains of higher frequency stimulation [40] . Stimulation at 50 Hz for 500 ms in 1 . 5 mM extracellular calcium concentration ( [Ca2+]e ) retrieves vesicles from the readily releasable pool ( RRP ) [41] . This stimulation pattern induced mild depression in Ctrls and strong initial facilitation of trains under NO conditions ( Fig 2A and S2 Data ) . Cumulative QC analysis revealed a pool size of 453 ± 37 ( n = 17 ) in Ctrl and 185 ± 18 in NO-exposed NMJs ( n = 16 , p < 0 . 01 ) , suggesting a strong reduction in ready-releasable/recycling vesicles ( Fig 2A–2D ) . Supporting the above data , pool size estimation in the presence of ODQ confirmed cGMP independence ( NO+ODQ: 310 ± 33 [n = 9] , p < 0 . 05 versus Ctrl; ODQ alone: 501 ± 34 [n = 9] , p > 0 . 05 versus Ctrl , Fig 2A–2D ) . And importantly , analysis of the vesicle pool sizes in NOS “null” mutants revealed a strong 2-fold increase compared to w1118 Ctrl and an over 5-fold increase compared to NO application ( NOSC: 975 ± 161 [n = 11]; NOSΔ15: 958 ± 139 [n = 4] , p < 0 . 001 versus Ctrl , Fig 2A–2D ) . To exclude any potential developmental effects caused by NOS deficiency that could account for these strong increases in release , we assessed NMJ morphology and ultrastructure . We analyzed the total volume of NMJs ( horseradish peroxidase [HRP] signal ) and the number of Bruchpilot ( Brp ) puncta/NMJ volume of z-stack confocal images ( S2A and S2B Fig and S9 Data ) and measured the number of AZs , T-bars per Ib bouton , and vesicles within a 250-nm semicircle around the AZ ( S2C and S2D Fig ) . These data indicated that reduced NOS activity has no developmental impact on the structure of NMJs and synaptic boutons and can therefore not explain the physiological differences observed above . In addition to changes in release , NO could also exert its effects indirectly via modulating transmitter uptake and pool recovery . To exclude this possibility that altered recovery from depression affected the above pool estimations , we examined eEJC recovery . Following depletion of vesicle pools during a 50-Hz train ( 1 s ) , we measured the time course of recovery over the following 60 s . NO did not show any effects on the time constant of recovery ( S3 Fig and S9 Data ) . In order to test whether NO acts specifically on RRP or also affects the availability of other pools , we stimulated the NMJ for longer periods ( 8 s ) at 50 Hz . This prolonged stimulation leads to recruitment of vesicles from the reserve pool ( RP ) [42 , 43] . Analysis revealed that NO also caused a strong reduction of release from the RP ( Fig 2E–2H , Ctrl: 11 , 160 ± 1 , 645 [n = 6]; NO: 5 , 286 ± 798 [n = 7] , p = 0 . 0062 ) . One important protein that regulates vesicle clustering and release of neurotransmitter is the phosphoprotein synapsin ( syn ) , which regulates recycling of RP vesicles in Drosophila NMJs [43] . We tested whether modulation of syn could be responsible by employing larvae deficient in this protein from the Syn97-null mutation [44] . These larvae did not exhibit any reduction in single-stimulus QC compared to Ctrls , but prolonged recruitment ( 500 ms at 50 Hz ) showed reduced vesicle availabilities . Importantly , incubation of Syn97 larvae with NO led to further reduction of both parameters ( S4 Fig and S9 Data ) , suggesting that NO effects are via a different signaling route . Based on these data , we suggest that NO decreases release of vesicles from the RRP and RP but does not affect the rate of vesicle pool recovery from depletion . The NO-mediated effects appear to be independent of syn , suggesting an event downstream of vesicle recruitment per se . We next applied an independent approach to estimate the synaptic parameters: fluctuation analysis [45] to estimate the number of functional release sites N . eEJCs were elicited at varying calcium concentrations ( [Ca2+]e: 0 . 5–3 mM , 0 . 2 Hz ) and amplitudes were plotted over [Ca2+]e ( Fig 3A and 3B and S3 Data ) . NO exposure led to reduced release across different Ca2+ concentrations ( 0 . 75–3 mM ) . N was estimated from parabolic fits to the variance-mean plots for each NMJ ( Fig 3C ) . This analysis revealed a strong reduction in N following NO exposure ( Fig 3D , NCtrl: 630 ± 104 [n = 5] , NNO: 117 ± 32 [n = 6] , p = 0 . 0006 ) . The estimation of N from the fluctuation analysis ( about 600 ) in Ctrl is in accordance with previously reported electron microscopy ( EM ) data showing a number of about 500 vesicles per NMJ [46] . These data confirm that NO most likely reduces the number of releasable vesicles by preventing vesicle fusion at individual release sites . The reduced QC seen following NO exposure can also be attributable to a change in the Ca2+ dependency of release , so we determined whether the reduced transmitter release is due to altered Ca2+ cooperativity of release [47] . The Hill slope was strongly reduced by NO ( Ctrl: 3 . 2 ± 0 . 4 [n = 6] , NO: 1 . 8 ± 0 . 7 [n = 5] , p = 0 . 0024 , Fig 3E ) ; however , the half maximal effective Ca2+ concentration ( EC50 ) was unaltered ( Ctrl: 1 . 0 ± 0 . 03 , NO: 1 . 0 ± 0 . 09 , p > 0 . 05 , Fig 3E ) , indicating that sensitivity to Ca2+ was not affected by NO . To further assess nitrergic effects on pvr , we used the PPR approach by delivering two pulses with interspike intervals ( ISIs ) between 10 and 200 ms at two different [Ca2+]e ( 1 and 1 . 5 mM , Fig 3F and 3G ) in Ctrl and NO-treated NMJs . Analysis showed that Ctrl NMJs only exhibit slight potentiation at low Ca2+ and high ISI , indicative of low pvr . In contrast , pvr in the presence of NO was decreased , as shown by an increased PPR ( potentiation at all ISI at 1 mM Ca2+ and 20 and 40 ms ISI at 1 . 5 mM Ca2+ , p < 0 . 05 , Ctrl versus NO at each ISI ) , which is also in agreement with elevated pvr in NOS “null” larvae . With about 500 release sites per NMJ and a QC of 200 ( Ctrl ) and 90 ( NO ) , our data present estimated pvr values of 0 . 33 ( Ctrl ) and 0 . 16 ( NO ) , with Ctrl values similar to estimates made previously in WT larvae [40] . Previously , we have shown that NO signaling can suppress mammalian P/Q and N-type Ca2+ channels [48] . In order to test whether altered Ca2+ influx could cause the observed effects on evoked release at the NMJ , we tested whether NO application for 60 min changed presynaptic Ca2+ levels during a train of synaptic stimulation . GCaMP5 was expressed presynaptically and activity-evoked Ca2+ influx in type 1b NMJ boutons was imaged at different extracellular Ca2+ concentrations ( 0 . 25–3 mM ) . Our data showed that NO had no effect on stimulated Ca2+ levels at any concentration tested ( ΔF/F0 , myrGCamP5: 3 mM Ca2+: Ctrl: 0 . 70 ± 0 . 09 , NO: 0 . 78 ± 0 . 12 [n = 13–18 boutons from 4–6 NMJs each] , p > 0 . 05; Fig 3H and 3I; GCaMP5: 0 . 25 mM Ca2+: 0 . 24 ± 0 . 03 , NO: 0 . 14 ± 0 . 03 , 0 . 5 mM Ca2+: Ctrl: 0 . 42 ± 0 . 08 , NO: 0 . 50 ± 0 . 07 , 1 . 5 mM Ca2+: Ctrl: 1 . 13 ± 0 . 14 , NO: 1 . 18 ± 0 . 21 [n = 28–46 boutons from 7–11 NMJs each] , p > 0 . 05; S5 Fig and S9 Data ) . Together , the data suggest that NO reduced evoked release and the frequency of spontaneous release , likely due to reduced release probability and Ca2+ cooperativity , which manifests itself in reduced vesicle fusion . We showed that the Ca2+ dependence of release , but not Ca2+ entry per se , was reduced by NO , which indicates a possible modulation of SNARE ( -associated ) protein interactions via NO-mediated PTMs . S-nitrosylation is a reversible non-enzymatic protein modification , the levels of which can be regulated via S-nitrosoglutathione reductase ( GSNOR ) , the sole alcohol dehydrogenase 5 ( ADH-5 ) isozyme in vertebrate brains [49] , which has a homologue in Drosophila ( encoded by the formaldehyde dehydrogenase [fdh] gene ) . This de-nitrosylation process requires GSH . GSH is produced from L-glutamate and Cys via the enzyme glutamate-cysteine ligase ( GCL ) , the rate-limiting step in GSH synthesis in fly [50] . The Drosophila GCL holoenzyme is heterodimeric , consisting of a catalytic ( DmGCLc ) and a modifier ( DmGCLm ) subunit , each encoded by a unique gene , and overexpression of either subunit increases cellular GSH levels [50] . In order to assess the contributions of SNO formation to the physiology at the NMJ , we investigated the effects of altering neuronal GSH levels . If NO mediates its observed actions via SNO formation , we should be able to prevent/reduce the effects on transmitter release by providing elevated GSH levels by ( i ) GSH supplementation , ( ii ) overexpression of GSNOR ( fdh ) , or ( iii ) overexpression of GCL ( DmGCLm/c ) and , inversely , enhance NO effects by using RNA interference ( RNAi ) expression of the above proteins . We tested first the recovery of NO-mediated reduction of eEJC amplitudes following NO exposure for 50 min by washing out NO . eEJC amplitudes recovered slightly ( Fig 4A , green and S4 Data ) ; however , when washing in GSH ( 150 μM ) , the amplitudes recovered to control levels after 15 min ( GSH [blue] versus NO at 50 min [red] , p < 0 . 05 ) , indicating a GSH-mediated reversal . To characterize effects of endogenous GSH formation , we used elav-Gal4-driven UAS-fdh31 , UAS-DmGCLm , and UAS-DmGCLc overexpression . It has been shown that overexpression of either DmGCLc or DmGCLm results in enhanced enzyme activity and elevated GSH levels [50] , GSNOR overexpression ( elav > UAS-fdh31 ) reduces global S-nitrosylation in fly , and conversely , GSNOR-RNAi expression ( elav > UAS-fdhri34 ) elevates SNO protein levels [51] . Overexpression of GSNOR and GCLm/c ( Fig 4B–4D ) prevented NO effects on QC ( GSNOR: 238 ± 20 [n = 11] , DmGCLm: 197 ± 32 [n = 7] , DmGCLc: 215 ± 39 [n = 6] , GSNOR+NO: 223 ± 24 [n = 8] , DmGCLm+NO: 177 ± 10 [n = 7] , DmGCLc+NO: 329 ± 26 [n = 3] , p > 0 . 05 ) and vesicle pool sizes ( GSNOR: 438 ± 51 [n = 10] , DmGCLm: 400 ± 99 [n = 7] , DmGCLc: 496 ± 93 [n = 6] , GSNOR+NO: 472 ± 34 [n = 8 ) , DmGCLm+NO: 360 ± 40 [n = 7] , DmGCLc+NO: 685 ± 148 [n = 3] , p > 0 . 05 , Fig 4B–4D ) . These data confirm that by enhancing GSNOR and GCL activities , thereby elevating intracellular GSH levels , the effects of NO on pool size and pvr ( PPR at 20 ms ISI; w1118 Ctrl [0 . 93 ± 0 . 03] versus NO [1 . 2 ± 0 . 07] , p < 0 . 0001 , GSNOR overexpression [0 . 88 ± 0 . 02] , +NO [0 . 84 ± 0 . 05]/GCLm overexpression [0 . 87 ± 0 . 04] , +NO [0 . 92 ± 0 . 03]/GCLc overexpression [0 . 99 ± 0 . 07] , +NO [0 . 96 ± 0 . 02] , p > 0 . 05 , Fig 4E ) were precluded , suggesting that this was due to reduced SNO formation . Furthermore , overexpression of GSNOR , DmGCLm , and DmGCLc prevented the reduction in mEJC frequency following NO exposure ( fGSNOR: 2 . 4 ± 0 . 3 s−1 [n = 13]; fDmGCLm: 3 . 0 ± 0 . 3 s−1 [n = 13]; fDmGCLc: 1 . 5 ± 0 . 42 s−1 [n = 5]; fGSNOR+NO: 1 . 7 ± 0 . 2 s−1 [n = 7]; fDmGCLm+NO: 2 . 9 ± 0 . 4 s−1 [n = 13]; fDmGCLc+NO: 0 . 4 ± 0 . 1 s−1 [n = 3] , p > 0 . 05 versus w1118 Ctrl and versus each Ctrl , Fig 4F ) without affecting mEJC amplitudes ( GSNOR: −0 . 6 ± 0 . 07 nA [n = 13]; DmGCLm: −0 . 7 ± 0 . 07 nA [n = 13]; DmGCLc: −0 . 5 ± 0 . 07 nA [n = 5]; GSNOR+NO: −0 . 6 ± 0 . 07 nA [n = 7]; DmGCLm+NO: −0 . 6 ± 0 . 08 nA [n = 13]; DmGCLc+NO: −0 . 6 ± 0 . 07 nA [n = 3] , p > 0 . 05 versus w1118 Ctrl and versus each Ctrl , Fig 4F ) or decays ( GSNOR: 7 . 5 ± 0 . 2 ms [n = 13]; DmGCLm: 9 . 7 ± 0 . 4 ms [n = 13] , DmGCLc: 6 . 7 ± 0 . 3 ms [n = 5] , GSNOR+NO: 6 . 2 ± 0 . 2 ms [n = 7] , DmGCLm+NO: 7 . 9 ± 0 . 5 ms [n = 13] , DmGCLc+NO: 6 . 2 ± 0 . 4 ms [n = 3] , p > 0 . 05 versus w1118 Ctrl and versus each Ctrl , Fig 4F ) . Furthermore , the reduction of endogenous GSNOR and DmGCLm activities ( elav > UAS-RNAi ) caused partial electrophysiological phenotypes , such as a decrease in eEJC amplitudes , QC , or vesicle pool size compared to w1118 Ctrl , with NO having no further major negative effects ( S6 Fig and S9 Data ) . We next asked which signaling routes and PTMs are involved in NO modulation of release . The SNARE-binding and fusion-clamp protein cpx regulates not only the Ca2+ cooperativity of evoked release but also spontaneous release [14] as well as release probabilities [52] , thereby presenting a strong candidate for mediating the observed NO-induced changes . Cpx acts by binding to the SNARE complex , thereby promoting the clamping of release , and only when replaced by synaptotagmin 1 in response to Ca2+ influx will vesicle fusion be initiated [14 , 20] . Dmcpx function can be regulated by protein kinase A ( PKA ) phosphorylation of serine126 ( Ser126 ) [7] or by prenylation at the C-terminus [15 , 16] . In order to test whether cpx is required to exert NO effects , we first used cpx null mutants ( cpxSH1 , cpx-/- ) [14] . In these animals , we detected a strong reduction in evoked release and QC ( 22 . 6 ± 3 . 2 [n = 11] , p < 0 . 0001 versus Ctrl ) , which was unaffected by NO ( 13 . 8 ± 2 . 0 [n = 4] , p > 0 . 05 versus cpx-/- , p < 0 . 0001 versus Ctrl , Fig 5A–5D and S5 Data ) . Similarly , when comparing the vesicle pool size , cpx-/- NMJs showed a strong reduction ( 26 ± 6 [n = 11] , p < 0 . 0001 versus Ctrl ) , which again was unaffected by NO ( 22 ± 5 [n = 4] , p > 0 . 05 versus cpx-/- , p < 0 . 0001 versus Ctrl , Fig 5A–5D ) . These data confirm that cpx is required for NO to induce suppression of evoked release and available vesicle pool size and suggest that NO might enhance the clamping function of cpx in WT larvae . We next tested the impact of NO on the clamping ability of cpx by characterizing spontaneous release . Interestingly , the frequency of spontaneous events inversely correlates with endogenous cpx levels [14] . We analyzed mEJCs in cpx-/- muscle 6 ( m6 ) , which exhibited an extremely high frequency [14] ( >40 × w1118 , Fig 5E ) . NO did not reduce the mEJC frequency in those preparations , although a precise analysis is difficult due to strong overlap of single mEJCs [14] . In order to allow more accurate frequency measurements in cpx-/- animals , we used neighboring muscle 5 ( m5 ) , posessing a synapse with approximately 4-fold fewer release sites compared to m6 . Similar to m6 , cpx-/- increased mEJC frequencies >10-fold compared to Ctrl ( m5: w1118: 0 . 8 ± 0 . 2 s−1 [n = 3] , cpx-/-: 11 . 6 ± 0 . 8 s−1 [n = 6] , p < 0 . 0001 ) ; however , following NO exposure , this preparation did not show any change in mEJC frequency ( m5 cpx-/- + NO: 9 . 7 ± 0 . 9 s−1 [n = 5] , p > 0 . 05 versus m5 cpx-/- , Fig 5E and 5F ) , suggesting the requirement of cpx for the observed nitrergic effects . Nevertheless , we recorded from m6 of heterozygous animals , which exhibit higher frequencies than w1118 but are still accurately quantifiable ( m6 cpx+/-: 4 . 6 ± 0 . 8 s−1 [n = 5] ) . Here , NO induced a strong reduction in the frequency ( m6 cpx+/- + NO: 0 . 6 ± 0 . 2 s−1 [n = 5] #p < 0 . 05 versus m6 cpx+/- Ctrl , Fig 5E and 5F ) , similar to that seen in w1118 . These data confirm that NO only modulates spontaneous release frequencies in the presence of cpx . Together , these data show that in the absence of cpx , NO causes no electrophysiological phenotypes . The NO-mediated reduction of eEJC amplitudes , QC , pool size , and mEJC frequency all require the presence of cpx , suggesting that its modulation might be responsible for the observed nitrergic effects , which could be explained by a gain-of-clamping function [53] . This potential effect was further investigated by using the established paradigm of activity-induced enhancement of spontaneous release at the Drosophila NMJ [7] . We assessed whether NO modulation of release also affects this activity-dependent signaling , which would strengthen the role of cpx as a target for nitrergic regulation and a general regulatory mechanism . PKA has been reported to modulate mEJC frequency potentiation in a cpx overexpression model ( Dmcpx 7B , [7] ) . We confirmed that high frequency stimulation ( 50 Hz for 3 s ) led to an enhanced mEJC frequency in w1118 NMJs relative to baseline ( Ctrl: 1 . 9 ± 0 . 2-fold [n = 13] , Fig 5G and 5H ) . Interestingly , repeating this protocol in larvae exposed to NO showed a lack of frequency potentiation ( NO: 0 . 8 ± 0 . 1-fold [n = 14] , p < 0 . 05 versus Ctrl ) , which was also ODQ independent ( NO + ODQ: 1 . 0 ± 0 . 1-fold [n = 7] , p > 0 . 05 versus NO , Fig 5G and 5H ) . To test whether the manipulation of PTMs also affects nitrergic suppression of frequency potentiation , we used larvae overexpressing GCLm and GSNOR and NOS “null” larvae . We found that GCLm and GSNOR overexpression occluded nitrergic effects on suppression of mEJC frequency potentiation , whereas the lack of NO signaling led to enhanced potentiation ( GCLm + NO: 2 . 3 ± 0 . 4-fold [n = 7] , GSNOR + NO: 2 . 5 ± 0 . 5-fold [n = 7] , NOS “null” [comprised of n = 5 NOSC and n = 3 NOSΔ15]: 3 . 8 ± 0 . 3-fold , *p < 0 . 01 versus Ctrl , ##p < 0 . 01 versus NO , ####p < 0 . 001 versus NO , Fig 5G and 5H ) . These data show that NO suppresses the activity-mediated increase in mEJC frequency and suggest that , similar to phospho-incompetent cpx mutants [7] , nitrergic modulation of WT cpx produces an inhibitory action on spontaneous release . The lack of PTM signaling leads to an enhanced frequency potentiation , strengthening the notion that NO-mediated effects are responsible for suppression of synaptic release and our data point towards modulation of cpx as a key signaling mechanism . Having shown that cpx signaling is involved in NO-mediated effects on spontaneous and evoked release , we next considered if S-nitrosylation of the Cys residue within the C-terminus of cpx possessing the CAAX motif could explain the observed results . Importantly , prenylation has been studied in several genetically modified cpx proteins in which the CAAX motif was eliminated [15 , 16] . These studies suggest that deletion of final parts of the C-terminus/final amino acid affects cpx localization , interactions with SNARE-proteins , and , subsequently , its function . To explore the effects of cpx farnesylation more in detail , we made use of Drosophila lines expressing green fluorescent protein ( GFP ) -tagged WT and mutant cpx ( cpx1257 , lacking the final amino acid [16] ) , referred to as CpxΔX . This mutant has been shown to exhibit altered co-localization with syntaxin at the dorsolongitudinal flight muscle ( DLM ) neuromuscular synapse . We assessed localizations of WT and mutant cpx at the NMJ ( elav > UAS-cpx-GFP , elav > UAS-cpx1257-GFP ) with respect to their interaction with the AZ protein , Brp . WT cpx exhibits diffuse localization within boutons ( as previously reported [15] ) with little co-localization with Brp ( Fig 6A and 6B and S6 Data ) . In contrast , the mutant form , lacking farnesylation , is highly co-localized with Brp , as indicated by the increase in Pearson’s coefficient ( Fig 6A and 6B; WT cpx: 0 . 35 ± 0 . 30 [n = 9] , CpxΔX: 0 . 65 ± 0 . 03 [n = 9] , p < 0 . 0001 ) . These data confirm that preventing cpx farnesylation results in enhanced co-localization with AZ . To further support these data , we conducted high-resolution stimulated emission depletion ( STED ) microscopy [54] and analyzed the Pearson’s coefficient for the co-localization of Brp with cpx . This experiment verified the confocal data showing enhanced co-localization of CpxΔX with Brp versus WT cpx ( WT cpx: 0 . 13 ± 0 . 02 [n = 25] , CpxΔX: 0 . 27 ± 0 . 02 [n = 23] , p < 0 . 0001 , Fig 6C and 6D ) . As Dmcpx possesses a predominant clamping function [23] , we propose that NO could lead to a reduction in farnesylation , a consequent stronger interaction with the SNARE complex at the AZ , and thereby enhance its clamping function upon transmitter release . To specifically confirm co-localizations , we used the high-resolution proximity ligation assay ( PLA ) , with which we imaged interactions of Brp with cpx . We used both lines , WT cpx-GFP and CpxΔX-GFP expressing larvae , and found that PLA signals are strongly enhanced at NMJs expressing the mutant cpx ( Fig 6E and 6F; WT cpx: 0 . 04 ± 0 . 004 [n = 9] , CpxΔX: 0 . 12 ± 0 . 02 [n = 9] , p = 0 . 009 ) . As the co-localization data may depend upon expression of GFP-tagged cpx , we confirmed equal GFP expression levels in both lines by immunoblotting ( S9A Fig ) . These co-localization and PLA experiments confirm an enhanced association of a mutated farnesylation-incompetent cpx with Brp and suggest that lack of farnesylation renders cpx in close proximity to release sites of AZs . In order to assess this possibility further , we used pharmacological and genetic tools to modulate cpx farnesylation and compared protein localization and synaptic release following farnesyl transferase ( FTase ) inhibition and NO exposure . Reduced expression of the Drosophila ortholog of FTase or inhibition of FTase by L-744 , 832 and GGTI-298 have strong effects on fly lethality [55] , implicating a crucial function of this signaling in fly . First , we tested how FTase inhibition ( 20 μM L-744 , 832 + 10 μM GGTI-298 ) and NO exposure affect cpx co-localization with the SNARE complex proteins syntaxin and synaptotagmin or Brp , using the PLA . We measured total PLA volume of NMJ z-stacks and normalized PLA signals to NMJ volume . We found that both treatments ( depicted as “farnesyl inh” and “NO , ” Fig 7A and 7B and S7 Data ) led to enhanced co-localization of cpx with syntaxin and Brp ( syntaxin-cpx: Ctrl: 0 . 04 ± 0 . 007 , NO: 0 . 12 ± 0 . 02 , farnesyl inh: 0 . 11 ± 0 . 02 , Brp-cpx: 0 . 02 ± 0 . 007 , NO: 0 . 08 ± 0 . 03 , farnesyl inh: 0 . 09 ± 0 . 05 , Fig 7A and 7B; p < 0 . 01 , p < 0 . 001 versus Ctrl ) , suggesting that NO PTMs and farnesylation inhibition enrich cpx at the AZ . When analyzing the interactions between the Ca2+ sensor synaptotagmin and cpx , we found that this interaction was completely suppressed following treatments ( Ctrl: 0 . 2 ± 0 . 06 , NO: 0 . 03 ± 0 . 006 , farnesyl inh: 0 . 04 ± 0 . 007 , Fig 7A and 7B; p < 0 . 01 versus Ctrl ) . The PLA data were further supported by STED imaging studies showing identical changes in protein co-localization , as determined by Pearson’s coefficient analysis ( S7 Fig and S9 Data ) . One possibility to allow for greater amounts of cpx to be available for binding to SNAREs is by enhancing free and soluble cytosolic levels as a consequence of reduced farnesylation . Farnesylation of cpx results in its membrane tethering , and thus protein fractions , which are membrane bound , are less mobile than soluble cytosolic proteins . To assess the mobility of potentially farnesylated versus soluble ( non-farnesylated ) cpx and thus distinguish between these two pools of cpx , we performed fluorescence recovery after photobleaching ( FRAP ) analysis of GFP-tagged WT and farnesylation-incompetent cpx ( CpxΔX ) . Although a previous study did not detect differences between farnesylated versus non-farnesylated cpx isoform using this method with a photo-bleaching area of half a bouton [15] , we found that accurate FRAP analysis of cpx-GFP mobility can only be performed by using substantially smaller bleaching areas , as reported previously [56] ( S8 Fig and S9 Data ) . Using this approach , we found that bleaching an area of 2 . 5 μm2 ( instead of >10 μm2 ) generally leads to faster recovery rates ( S8 Fig and S9 Data ) . Our data confirmed that lack of farnesylation ( CpxΔX ) allows for greater movement of cpx and faster recovery ( tau: WT cpx: 18 . 1 ± 1 . 7 ms , CpxΔX: 11 . 9 ± 1 . 2 ms [p < 0 . 05] , WT Cpx + NO: 8 . 8 ± 0 . 8 ms [p < 0 . 0001] , n = 18–20 , Fig 7C ) , as expected for a soluble protein . Our data further show that NO treatment caused the same increase in recovery rates ( Fig 7C ) , suggesting that NO also prevented farnesylation . These data suggest that due to enriched local levels , cpx outcompetes synaptotagmin for SNARE binding at the AZ , thereby displacing synaptotagmin , as reported previously in biochemical studies [53] . Our data show that pharmacological and genetic inhibition of farnesylation promotes cpx co-localization with the AZ and supports the notion that this negatively impacts on synaptotagmin-SNARE complex binding , subsequently reducing release . The specificity of the PLA was corroborated by lack of Brp-cpx PLA signals in cpx-/- larvae ( S9B–S9D Fig ) . Next , we explored the possibility of whether specific inhibition of FTase activity by L-744 , 832 and GGTI-298 and FTase RNAi mimics the effects of NO on synaptic transmission . We found that , in both conditions , the frequency of mEJCs was reduced to similar values seen following NO exposure ( fmEJC: L-744 , 832 + GGTI-298: 0 . 7 ± 0 . 1 s−1 [n = 8] , p = 0 . 0051 versus Ctrl , FTase RNAi: 0 . 9 ± 0 . 2 s−1 [n = 9] , p = 0 . 0136 versus Ctrl , Student t test , Fig 7D ) . Importantly , both L-744 , 832 + GGTI-298 and FTase RNAi expression reduced evoked transmission and available vesicle pool size to levels similar to those following NO incubation ( L-744 , 832 + GGTI-298: eEJC: 56 ± 5 nA , QC: 80 ± 13 [n = 9] , pool size: 180 ± 27 [n = 9] , p < 0 . 0001 versus each w1118 Ctrl; FTase RNAi: eEJC: 75 ± 5 nA , QC: 82 ± 6 [n = 9] , pool size: 120 ± 17 [n = 9] , p < 0 . 0001 versus each w1118 Ctrl , Student t test , Fig 7E and 7F ) . These data suggest that the farnesylation status of cpx mediates nitrergic effects , resulting in changed SNARE protein interactions , which determines the physiological outcome of cpx . To further investigate the effects of NO directly on the prenylation process , we employed the well-characterized GFP-CAAX transfection model [57] . Here , human embryonic kidney ( HEK ) cells were transfected with GFP-CAAX ( K-Ras motif ) and the membrane association was assessed in response to prenylation inhibition and NO treatment . In control conditions , GFP exhibited a strong fluorescence signal at the membrane , which disappeared and redistributed into the cytosol following pharmacological inhibition of prenylation ( L-744 , 832 + GGTI-298 , p < 0 . 0001 ) , confirming the prenylation-mediated localization of GFP-CAAX to the membrane ( Fig 8A and S8 Data ) . Importantly , we showed that NO treatment ( propylamine propylamine NONOate [PAPA-NONOate] , p < 0 . 0001 ) induced a similar phenotype , with GFP being localized predominantly in a cytosolic manner—suggesting that NO prevents farnesylation through the same pathway ( Fig 8A ) . To confirm that the Cys within the CAAX motif can undergo S-nitrosylation , we performed the Biotin Switch Assay on cpx-3 from isolated mouse retinas . NO donor incubation induced a >2-fold increase in SNO-cpx ( Fig 8B ) , confirming this PTM on cpx and suggesting that this PTM is responsible for NO-induced changes in localization and function of cpx . To specifically confirm the effects of S-nitrosylation and SNO interaction with farnesylation of cpx in Drosophila , we generated and expressed a nitroso-mimetic cpx mutant ( Dmcpx 7AC140W ) in a cpx null background ( cpxSH1 ) and assessed synaptic responses . The Cys140 of Dmcpx is located within a hydrophobic region , as predicted in the Kyle Doolittle plot , which favors S-nitrosylation [58] . This mutant exhibits reduced evoked responses , QC , and vesicle pool sizes ( eEJC: 70 ± 7 nA , QC: 106 ± 8 , pool size: 204 ± 23 [n = 15 each] , p < 0 . 0001 versus each w1118 Ctrl , Fig 8C and 8D ) , indicating that the mimicking of S-nitrosylation and simultaneous lack of farnesylation of cpx caused the observed changes . Importantly , this mutation also induced a reduction in spontaneous activity ( fmEJC: 1 . 3 ± 0 . 2 s−1 [n = 15] , p < 0 . 05 versus w1118 Ctrl , Fig 8C and 8D ) , reinforcing the argument of enhanced clamping function due to SNO formation and lack of farnesylation . The expression of WT cpx in the null background did not affect QC , pool size , or mEJC frequency ( QC: 167 ± 17 [n = 5]; pool size: 381 ± 76 [n = 5]; fmEJC: 2 . 4 ± 0 . 4 s−1 [n = 10 each] , p > 0 . 05 versus each w1118 Ctrl ) . To confirm changes in localization of Dmcpx 7AC140W , we analyzed PLA signals and found that Dmcpx 7AC140W highly co-localizes with Brp , in strong contrast to WT cpx ( WT: 0 . 025 ± 0 . 013 , Dmcpx 7AC140W: 0 . 17 ± 0 . 03 [n = 6–7] , p < 0 . 0001 , both expressed in cpx-/- background , Fig 8E and 8F ) . The data from the PLA experiments were confirmed by STED confocal microscopy , showing significantly higher Pearson’s coefficients for the co-localization of the cpx mutant C140W with Brp relative to the interaction of WT cpx with Brp ( WT cpx: 0 . 13 ± 0 . 03 , Dmcpx 7AC140W: 0 . 34 ± 0 . 02 [n = 20–24] , p < 0 . 0001; Fig 8E and 8G ) . These data demonstrate that independent approaches to block farnesylation ( and mimic of cpx-SNO ) recapitulate nitrergic modulation of release and protein localization and therefore link for the first time NO-induced PTM and farnesylation signaling of cpx . We propose that S-nitrosylation acts as a novel endogenous pathway to alter cpx farnesylation signaling and protein–protein interactions and thereby allows a fine-tuning of synaptic function . NO regulates a multitude of physiological and pathological pathways in neuronal function via generation of cGMP , thiol-nitrosylation , and 3-nitrotyrosination in health and disease [59] . Here , we show by employing biochemical and genetic tools in Drosophila , mouse , and HEK cells that NO can S-nitrosylate cpx and modulate—in a cGMP-independent manner—neurotransmitter release at the NMJ by interfering with its prenylation status , thereby affecting the localization and function of this fusion-clamp protein . We found that these nitrergic effects are reversed by GSH application or overexpression of GSH-liberating and de-nitrosylating enzymes ( GCLm/c , GSNOR ) . GSH is the major endogenous scavenger for the NO moiety by the formation of S-nitrosoglutathione ( GSNO ) and consequently reduces protein-SNO levels via trans- and de-nitrosylation . The suppression of NOS activity facilitates synaptic function and the data support the notion that endogenous or exogenous NO enhances S-nitrosylation , reduces cpx farnesylation , and diminishes release . Of the numerous synaptic molecules involved in release , cpx in particular has been implicated in the regulation of both evoked and spontaneous release due to its fusion-clamp activity . Despite the seemingly simple structure of cpx , its physiological function is highly controversial , as this small SNARE-complex binding protein can both facilitate but also diminish fast Ca2+-dependent and spontaneous release , depending on the system studied [22 , 25 , 53 , 60] . In addition , there are different mammalian isoforms of cpx ( 1–4 ) , which differ in their C-terminal region , with only cpx 3/4 containing the CAAX prenylation motif . Farnesylation in general determines protein membrane association and protein–protein interactions [61] , and some cpx isoforms , such as muscpx 3/4 and Dmcpx 7A , are regulated in this manner [13 , 23 , 62] . However , muscpx 1/2 does not possess a CAAX motif , suggesting differential regulatory pathways to modulate cpx function . In Drosophila , there are alternative splice variants resulting from a single cpx gene , but the predominant isoform contains the CAAX motif ( Dmcpx 7A ) , implicating the importance of this signaling molecule [15 , 16] . The other splice isoform ( Dmcpx 7B ) lacks the CAAX motif and is expressed at about 1 , 000-fold lower levels at the larval stage [15] , thus making Dmcpx 7A the dominant isoform to be regulated by farnesylation . However , the lack of Dmcpx 7B phosphorylation by PKA induces similar phenotypes as seen in our experiments when assessed following an induction of activity-dependent potentiation of mEJC frequency [7] , which also may involve cpx–synaptotagmin 1 interactions . Interestingly , both depletion and excessive levels of cpx suppress Ca2+-dependent and -independent exocytosis [63] . Cpx may promote SNARE complex assembly and simultaneously block completion of fusion by retaining it in a highly fusogenic state . Ca2+-dependent fusion is promoted below a concentration of 100 nM of cpx , whereas above 200 nM , it exhibits a clamping function resulting in a bell-shaped response curve [64] . Previous work suggests that synaptotagmin 1 , once bound to Ca2+ , relieves the cpx block and allows fusion . Another study reported that selective competition between cpx and synaptotagmin 1 for SNARE binding allows regulation of release [53] . Our data are in agreement with the latter findings , as we observed reduced synaptotagmin 1–cpx interactions following the block of farnesylation ( Fig 7 ) , indicating fewer synaptotagmin molecules binding to the SNARE complex to displace cpx . This limited replacement of cpx by synaptotagmin has been implicated in biochemical studies showing that local excess of cpx inhibits release , presumably by outcompeting synaptotagmin binding [53 , 60] . Thus , synaptotagmin-SNARE binding is strongly dependent upon the local concentration of cpx [53] . Alternatively , and we cannot exclude this possibility , the modulation of cpx may simply alter its binding to the SNARE complex without directly displacing synaptotagmin , but interpretation of the data from our assays ( PLA , co-localization ) would not allow us to distinguish between these possibilities . Our data are compatible with the idea that cpx binds to the SNARE complex , facilitates assembly , and then exerts its clamping function by preventing full fusion due to SNARE complex stabilization and subsequent increased energy barrier to allow fusion . Our model could provide an explanation of how cpx can be regulated to signal downstream to modulate transmitter release . So far , there are no data available , apart from mutation studies , as to how cpx function can be altered . We provide data indicating a physiologically relevant mechanism to adjust cpx function , possibly to the requirements of the neuron to adjust synaptic transmission . This likely occurs due to Cys S-nitrosylation and suppression of farnesylation , allowing greater amounts of hydrophilic cpx , not bound to endomembranes , to be available for binding with the SNARE complex in an altered configuration . This cross signaling between nitrosylation/farnesylation has been proposed to act as a molecular switch to modulate Ras activity [65] . Our data show that enhanced nitrergic activity and blocking farnesylation , either genetically ( CpxΔX ) or pharmacologically , alters the localization of cpx at the Drosophila NMJ and that of GFP-CAAX in HEK cells ( Figs 6–8 ) . Furthermore , by using a nitroso-mimetic cpx mutant , we found enhanced co-localization of cpx with the AZ protein Brp , implying a localization-function relationship ( Fig 8 ) . This consequently increases the net-clamping function because of elevated local concentrations of cpx . Dmcpx specifically exhibits a strong clamping function , as shown following overexpression in hippocampal neurons , which causes suppression of evoked and spontaneous release accompanied by a reduction of the release probability [23] or reduced vesicle fusion efficiency in in vitro assays [64] . Two independent studies eliminating the CAAX motif in Dmcpx ( cpx572 and cpx1257 ) investigated localization-function interactions and showed disagreeing effects on both release and cpx localization [15 , 16] . In particular , it has also been shown that the truncated cpx ( cpx572 , lacking the last 25 amino acids ) does not co-localize with Brp [15] . Interestingly , this mutant causes a strong decrease in C-terminal hydrophobicity and a modest physiological response ( increased mini frequency , decreased evoked amplitudes equivalent to a loss of clamping and loss of fusion function ) relative to the total knock-out ( KO ) . In contrast , the cpx mutant with single amino acid deletion ( cpx1275 ) causes no effect on evoked but identical effects on the frequency of spontaneous release , suggesting a lack of clamping but no lack of fusion function . In addition , this mutant now co-localizes with the AZ at the NMJ [16] . These two studies indicate that the different mutations cause contrasting electrophysiological and morphological phenotypes , indicating that it is due to the nature of the mutation ( lack of the last 25 amino acids versus 1 amino acid ) , which highlights the importance of a functional C-terminus . More recent studies have shown that deletions of the final amino acids ( 6 or 12 residues ) completely abolished the membrane binding of cpx-1 , impairing its inhibitory function and confirming the requirement of an intact C-terminus for inhibition of release [66 , 67] . Here , we use an endogenous cpx with intact hydrophobic C-terminus , allowing physiological membrane binding . This is essential for inhibitory function , as the C-terminus is required for selective binding to highly curved membranes , such as those of vesicles [68] . Thus , as we used different approaches to alter farnesylation and generated a single amino acid mutant cpx ( Dmcpx 7AC140W ) , leaving the C-terminus intact , our studies were performed under conditions of endogenous regulation of cpx function and thus provide new functional data on cpx signaling . Importantly , our data show that this regulation alters cpx function , and this is the first study to provide an explanation for the differential effects observed using cpx mutants or even cpx protein fragments in mammals , worm , and fly in various cross-species rescue experiments [20 , 23] . Our data are in agreement with a model that non-farnesylated hydrophilic and soluble cytosolic cpx binds to the vesicular membrane via its C-terminal interactions , thereby exerting its inhibitory effect . When proteins are farnesylated , they are likely tethered to endomembranes , other than vesicle membranes [12] . It has to be distinguished between cpx interaction with the vesicle membrane as a result of the hydrophobic C-terminus , allowing cpx to become in close proximity to the AZ , and cpx endomembrane binding following farnesylation , which prevents cpx interactions with the AZ . However , in our case , SNO modification may enhance the binding to other proteins ( e . g . , SNAREs ) , thereby augmenting the effects . These additional interactions with unknown binding partners may affect proper cpx function and explain some of the discrepancies seen in studies using other genetically altered cpxs . In summary , our study provides new data to illustrate a potential mechanism to regulate cpx function in a physiological environment , and we showed that NO acts as an endogenous signaling molecule that regulates synaptic membrane targeting of cpx , a pathway that may reconcile some of the controversial findings regarding cpx function . We suggest that increased S-nitrosylation and consequent lack of farnesylation leads to enhanced cytosolic levels of a soluble hydrophilic cpx and less endomembrane-bound fractions ( Fig 9 ) , because farnesylation-incompetent proteins remain in the cytosol [12] . These novel observations advance our understanding of similar nitrergic regulation of farnesylation that may be relevant for mammalian cpx-dependent synaptic transmission at the retina ribbon synapse and other brain regions [13] . Finally , this work has broader implications for physiological or pathological regulation of the prenylation pathway not only during neurodegeneration and aging , when enhanced S-nitrosylation might contribute to abnormal farnesylation signaling [69 , 70] , but also in other biological systems in which nitrergic activity and prenylation have important regulatory functions such as in cardio-vasculature or cancer signaling [71] . Flies were raised on standard maize media at 25 °C at a 12-h LD cycle . The elav-Gal4 [C155] driver was obtained from the Bloomington Stock Center ( Indiana , US ) . The UAS-RNAi lines ( GCLm [CG4919] , GCLc [CG2259] , and Fnta [CG2976] ) were purchased form the Vienna Drosophila Resource Centre ( VDRC ) . The use of the UAS-Gal4 bipartite expression system to drive pan-neuronal expression excludes potential postsynaptic effects . The elav-Gal4 driver ( female flies ) and the UAS responder lines ( male flies ) were crossed to obtain offspring expressing the genes of interest and w1118 were used as Ctrls . The fluorescent Ca2+ sensor GCaMP5 was tethered to the plasma membrane with an N-terminal myristoylation ( myr ) sequence as described previously [72] . The UAS-myrGCaMP5 and cpxSH1 null mutant lines were provided by Troy Littleton ( MIT , Cambridge , MA ) [73] . GCaMP5 was expressed in glutamatergic neurons ( OK371-Gal4; UAS-GCaMP5 ) . cpx expression levels are shown for w1118 and cpxSH1 larvae in S9B and S9C Fig . UAS-fdh31 ( expression of fdh homologue of mammalian GSNOR/ADH-5 ) and UAS-fdhri34/25 ( expression of fdh RNAi ) mutant transgenic lines were kindly provided by Li Liu Institute of Biophysics , Chinese Academy of Sciences , Beijing , China ) [51] . NOSΔ15/NOSC lines were provided by Patrick O’Farrell ( UCSF , San Francisco , CA ) . NOSΔ15 deletion removes sequences encoding residues 1–757 , encompassing the entire oxygenase domain and including regions that bind the catalytic heme and the substrate rendering the lines NOS “null” [33 , 34] . The syn-null mutant transgenic line ( Syn97 ) was generously provided by Erich Buchner ( Universitätsklinikum Würzburg , Germany ) [44] . UAS-EGFP-cpx and UAS-EGFP-cpx1257 transgenic lines were kindly provided by Fumiko Kawasaki ( Penn State University , PA ) [16] . UAS-GCLm and UAS-GCLc transgenic lines were provided by William C . Orr ( Southern Methodist University Dallas , TX ) . cDNAs encoding for cpx 7A was a gift from Troy Littleton and used as a template for downstream PCRs . Cys 140 of cpx7A isoform was mutated to tryptophan to generate S-nitrosylation mimic mutant . PCR products , which include XhoI and XbaI restriction sites , were cloned into the pJFRC2 vector [74]—a gift from Gerald Rubin ( Addgene plasmid no . 26214 ) —by standard methods . The resulting constructs were injected into attP40 Drosophila strains . The resulting transgenic lines ( Dmcpx7AC140W and WT Dmcpx ) were crossed into a cpxSH1 background [7] using standard balancing techniques . The FlincG3 ORF was amplified from pTriEx4-H6-FGAm ( FlincG3 ) ( Addgene plasmid no . 49202 ) and the resultant PCR product cloned into pUASTattB by the Protein Expression Laboratory ( PROTEX ) , University of Leicester . Microinjection of the pUASTattB plasmid was performed by the University of Cambridge , Department of Genetics Fly Facility . TEVC recordings were performed as described previously [75] . Sharp-electrode recordings were made from ventral longitudinal m6 in abdominal segments 2 and 3 of third instar larvae using pClamp 10 , an Axoclamp 900A amplifier and Digidata 1440A ( Molecular Devices , US ) in hemolymph-like solution 3 ( HL-3 ) [76] . Recording electrodes ( 20–50 MΩ ) were filled with 3 M KCl . mEJCs were recorded in the presence of 0 . 5 μM tetrodotoxin ( Tocris , UK ) . All synaptic responses were recorded from muscles with input resistances ≥4 MΩ , holding currents <4 nA at −60 mV and resting potentials more negative than −60 mV at 25 °C , as differences in recording temperature cause changes in glutamate receptor kinetics and amplitudes [77] . Holding potentials were −60 mV . The extracellular HL-3 contained ( in mM ) : 70 NaCl , 5 KCl , 20 MgCl2 , 10 NaHCO3 , 115 sucrose , 5 trehalose , 5 HEPES , and 1 . 5 CaCl2 ( 0 . 5–3 . 0 mM in Fig 3 and S3 Data , as specified ) . Average single eEJC amplitudes ( stimulus: 0 . 1 ms , 1–5 V ) are based on the mean peak eEJC amplitude in response to 10 presynaptic stimuli ( recorded at 0 . 2 Hz ) . Nerve stimulation was performed with an isolated stimulator ( DS2A , Digitimer ) . Paired-pulse experiments were performed by applying 5 repetitive stimuli ( 0 . 2 Hz ) at different intervals ( 20 , 40 , 100 , 200 ms ) for each cell at each ISI . All data were digitized at 10 kHz and for miniature recordings , 200-s recordings , we analyzed to obtain mean mEJC amplitudes , decay , and frequency ( f ) values . QC was estimated for each recording by calculating the ratio of eEJC amplitude/average mEJC amplitude , followed by averaging recordings across all NMJs for a given genotype . mEJC and eEJC recordings were off-line low-pass filtered at 500 Hz and 1 kHz , respectively . Materials were purchased from Sigma-Aldrich ( UK ) unless otherwise stated . Approximately 40 eEJCs were elicited at different [Ca2+]e , ranging from 0 . 5 to 3 mM to give mean eEJC amplitudes ( I ) . The mean eEJC is given by I = Npvrq [45] , with N being the number of independent release-ready vesicles , pvr the vesicular release probability , and q the quantal size at each given [Ca2+]e . The eEJC variance was calculated as previously described [45] . The plots of the variance-mean were obtained for each cell and fitted with the parabolic function Var ( I ) = I2/N + qI that was a constraint to pass through the origin . Upon fitting the parabola , pvr and q were calculated using the equations: q = A/ ( 1+CV2 ) and pvr = I ( B/A ) ( 1+CV2 ) where CV2 is the coefficient of variation of the eEJC amplitudes at a given [Ca2+]e concentration calculated as CV2 = ( eEJCs standard deviation/mean amplitude ) 2; A and B were obtained from the fitting parameters . Estimated values were not corrected for variability in mEJC amplitude distributions or latency fluctuations . Ca2+ cooperativity was assessed by plotting eEJC amplitudes over [Ca2+]e and fitted with the Hill equation ( mean eEJC amplitude plotted versus different [Ca2+]e: eEJC ( [Ca2+] ) = eEJCmax[1+ ( EC50/[Ca2+] ) slope]−1 ) , yielding the Hill slope as a measure of Ca2+ cooperativity . The apparent size of the RRP was probed by the method of cumulative eEJC amplitudes [78] . Muscles were clamped to −60 mV and eEJC amplitudes during a stimulus train ( 50 Hz , 500 ms [of a 1-s train] ) were calculated as the difference between peak and baseline before stimulus onset of a given eEJC . Receptor desensitization was not blocked as it did not affect eEJC amplitudes , because a comparison of the decay of the first and the last eEJC within a train did not reveal any significant difference in decay kinetics . The number of release-ready vesicles ( N ) was obtained by back extrapolating a line fit to the linear phase of the 500-ms cumulative eEJC plot ( the last 200 ms of the train ) to time zero . N was then obtained by dividing the cumulative eEJC amplitude at time zero by the mean mEJC amplitude recorded in the same cell . To calculate the QC in the train , we used mean mEJC amplitudes measured before the train . Third instar larvae were dissected in ice-cold PBS then fixed in 4% paraformaldehyde . After permeabilization with PBS-0 . 1% Triton ( PBS-T ) and blocking with PBS-T containing 0 . 2% bovine serum albumin ( BSA ) and 2% normal goat serum , larval fillets were incubated at 4 °C overnight in solutions of primary antibody . The following antibody dilutions were used: NC82 ( supernatant ) anti-Brp ( Bruchpilot ) 1:200 , cpx ( 1:500 ) , syntaxin ( 1:200 ) , synaptotagmin ( 1:200 ) , and GFP ( 1:200 ) . After 3 × 10 min washes in PBS-T , larvae were incubated with AlexaFluor 488 goat anti-HRP ( Jackson Immuno Research ) and AlexaFluor 546 goat anti-mouse 1:500 dilution for 90 min at room temperature . Larvae were mounted using Vectashield mounting medium ( Vector Labs ) and NMJ 6/7 ( segments A2 and A3 ) images were acquired with a Zeiss laser-scanning confocal microscope ( LSM 510 , Zeiss ) . Image analysis was performed with ZEN ( Zeiss ) and Volocity 6 . 3 software . Images were acquired on a Leica TCS SP8 system attached to a Leica DMi8 inverted microscope ( Leica Microsystems ) . Excitation light ( 488 nm for AlexaFluor488 or 561 nm for AlexaFluor568 ) was provided by a white light laser with a repetition rate of 80 MHz . Images were acquired using a 100× 1 . 4 NA oil immersion objective and fluorescence was detected through a bandpass of 495–550 nm ( AlexaFluor488 detection ) or 570–650 nm ( AlexaFluor 561 detection ) . Gated STED imaging of samples was achieved through use of 592-nm and 660-nm depletion lasers with a time gate set to 1 . 8–8 ns using the Leica STED 3X system . All images were acquired with 32-line averages and 22 × 22 nm pixel size . Images were taken using an LSM 510 confocal microscope ( Zeiss ) . The size of the bleaching area was optimized as shown previously [56] . Bleaching areas were selected within each bouton ( about 2 . 5 μm2 ) and images acquired every 10 s . Data were fitted with a single exponential to reveal tau values of fluorescence recoveries . The assay was performed as described [79] . Briefly , dissected third instar larvae were fixed in Bouin’s solution for 15 mins on ice , washed in PBT ( PBS with 0 . 1% Triton ) 3 times for 10 min each and blocked in PBT/1% BSA for 1 h . Larvae were incubated overnight at 4 °C in mouse and rabbit antibodies against the 2 proteins of interest , diluted in PBT/1% BSA . Primary antibodies used were anti-rabbit cpx ( Littleton ) , anti-rabbit GFP ( Abcam ) , anti-mouse Brp ( Developmental Studies Hybridoma Bank [DSHB] ) , anti-mouse syntaxin ( DSHB ) , and anti-mouse Synaptotagmin ( DSHB ) . All antibodies were used at 1:200 dilution . The next day , PLA probe binding , ligation , and amplification steps were performed as described [79] . Before mounting , larvae were counterstained with AlexaFluor 488 goat anti-HRP ( Jackson Immuno Research ) at 1:500 dilution for 40 mins . PLA signals were only measured within the HRP signals . PLA signal and NMJ volumes of z-stack images were analyzed in Volocity 6 . 3 . PLA signals were only measured within the HRP signals . All PLA signals were expressed relative to total NMJ volume ( S10 Fig and S9 Data ) . A plasma membrane targeted eYFP CAAX protein was constructed by fusing the last 15 amino acids of Human K-Ras isoform b with the C-terminus of eYFP . A short linker sequence GTMASNNTASG was inserted between the last amino acid of eYFP and the membrane targeting CAAX sequence . The resulting construct was subcloned into expression vector pcDNA5 frt and verified by DNA sequencing . HEK293 FT cells were plated on poly-d-lysine coated glass coverslips in 6 well plates and transfected with 0 . 5 g eYFP CAAX per well using polyethylenimine ( PEI ) at a ratio of 1 g DNA to 6 g PEI . Prior to imaging , cells were treated for 12 h with the NO donor DETA-NONOate or a combination of the farnesyl transferase inhibitor L-744 , 832 ( 20 μM ) and the geranylgeranyltransferase I inhibitor GGTI-298 ( 20 μM ) . Cells were then washed 3 times with PBS and fixed for 15 min with 4% paraformaldehyde . Coverslips were mounted on glass microscope slides with VectaShield H1500 and observed using a Zeiss laser scanning confocal microscope . Animals were kept in the dark 3 h before removing the retinas in order to decrease basal levels of protein nitrosylation . Retinas were kept in DMEM ( Gibco 31053–028 ) with protease inhibitors ( Complete ) and treated with NO donors ( GSNO and PAPA-NONOate , 20 μM ) for 40 min at room temperature and protected from light . The biotin-switch assay was performed with the S-nitrosylated Protein Detection kit ( Cayman Chemical , 10006518 ) in the dark . Bradford assay was performed and equal amount of proteins were incubated with Streptavidin beads ( Sigma ) overnight . Western blot was performed with cpx 3 antibody ( Synaptic Systems ) , 1:1 , 000 . Third instar larvae were “filleted” in phosphate-buffered saline at room temperature and then fixed in 2% ( wt/vol ) glutaraldehyde in 0 . 1 M sodium cacodylate buffer ( pH 7 . 4 ) at 4 °C overnight . They were postfixed with 1% ( wt/vol ) osmium tetroxide/1% ( wt/vol ) potassium ferrocyanide for 1 h at room temperature and then stained en bloc , overnight , with 5% ( wt/vol ) aqueous uranyl acetate at 4 °C , dehydrated and embedded in Taab epoxy resin ( Taab Laboratories Equipment Ltd , Aldermaston , UK ) . Semi-thin sections , stained with toluidine blue , were used to identify areas containing synaptic regions ( m6/7 in regions A2/A3 ) . Ultra-thin sections were cut from these areas , counterstained with lead citrate , and examined in an FEI Talos transmission electron microscope ( FEI Company [Thermo Fisher Scientific Inc . ] , Hillsboro , OR ) . Images were recorded using an FEI Ceta-16M CCD camera with 4k × 4k pixels . SV measurements were made using ImageJ software . A total of about 500–900 SVs were measured in 5–10 boutons from 3 animals per genotype . Wandering third instar larvae expressing presynaptic UAS-myrGCaMP5 or UAS-GCaMP5 using the pan-neuronal C155 or glutamatergic neuronal OK371 driver , respectively , were dissected in low Ca2+ HL-3 saline ( 0 . 2 mM CaCl2 ) at room temperature . The motor nerves were carefully snipped below the ventral nerve cord , and the CNS was removed . The preparation was washed several times with HL-3 containing 1 . 5 mM Ca2+ . Nerve stimulation was performed with an isolated stimulator ( DS2A , Digitimer ) and images were recorded before , during ( 2–6 s in a train at 60 Hz ) and after the stimulation period ( 8 s ) in HL-3 containing 3 mM Ca2+ or during 15 s in a 20-s train at 20 Hz at indicated Ca2+ levels in the presence of 5 mM L-glutamic acid . We acquired images at a rate of 1 image per 4 s using a Zeiss laser-scanning confocal microscope ( LSM 510 Meta; Zeiss ) with a 63× 1 . 0 NA water immersion objective ( Zeiss ) . Excitation was set at 488 nm ( Argon laser ) using a dichroic mirror 490 nm and a bandpass filter 500–550 nm . Low sampling rates were sufficient to investigate Ca2+ plateau levels during the 8-s stimulation periods [80] . A single confocal plane of muscle pair 6/7 NMJ in segments A2 or A3 was imaged to establish a baseline . Small z-drifts were manually corrected during the imaging session . Imaging sessions in which significant movement of the muscle occurred were discarded . Images were analyzed using Volocity 6 . 3 Image Analysis software ( PerkinElmer ) . Single bouton fluorescence intensities were measured ( average within a bouton ) and bouton ΔF/F0 values were averaged for each NMJ . NMJs of larvae expressing UAS-FlincG3 presynaptically were imaged as described above to measure GCaMP fluorescence . To prevent cGMP breakdown by PDE activity , preparations were incubated with 10μM Zap prior to imaging . High resolution respirometry was performed with an Oroboros O2K oxygraph ( Oroboros Instruments Ltd . ) . For each measurement , 3 third instar larvae were homogenized in 100 μL of respiration buffer MiR05 [81] . Leak state respiration was measured after adding 5 mM of pyruvate , 2 mM of malate , and 10 mM of glutamate . Oxphos capacity supported by Complex I was measured after addition of 1 . 25 mM ADP . After addition of 10 mM succinate , Oxphos capacity supported by both Complex I and Complex II were measured . Free Oxphos capacity was calculated as the difference Oxphos–Leak . Respiratory Ctrl ratios ( RCRs ) were calculated as the ratio Oxphos/Leak . Larval brains ( 30 per condition ) were isolated and assessed for cGMP production . Briefly , brain extracts were diluted 5-fold in 100 mM sodium acetate , pH 6 . 2 , and acetylated by consecutive addition of triethylamine ( 10 μL ) and acetic anhydride ( 5 μL ) and used in the radioimmunoassay [82] within 60 min . Cyclic GMP standards ( 100 μL; 0–4 nM ) were treated identically . Acetylated samples ( 100 μL ) were mixed with 2′-O-succinyl 3-[125I]-iodotyrosine methyl ester cyclic GMP ( GE Healthcare , IM107 ) ( 50 μL , about 3 , 000 d . p . m . made up in 50 mM sodium acetate , 0 . 2% BSA , pH 6 . 2 ) , and 100 μL of anti-cyclic GMP antibody ( GE Healthcare , TRK500; diluted in 50 mM sodium acetate , 0 . 2% BSA , pH 6 . 2 ) . Samples were intermittently vortex mixed during a 4-h incubation at 4 °C . Free and bound cyclic GMP was separated by charcoal precipitation with 500 μL of a charcoal suspension ( 1% [w/v] activated charcoal in 100 mM potassium phosphate , 0 . 2% BSA , pH 6 . 2 ) . After vortex mixing for 5 min , samples were centrifuged ( 13 , 000 × g , 4 min , 4 °C ) and radioactivity determined in an aliquot of supernatant ( 600 μL ) . Unknown values were determined from the cyclic GMP standard curve using GraphPad Prism 7 ( GraphPad Software Inc . , San Diego , CA ) . Data points represent 2 measurements of 30 brains for each condition . NO donor solutions were made freshly from stock solutions on the day and working solutions ( 200 μM sodium nitroprusside [SNP] and 5 μM PAPA-NONOate , each releasing about 200 nM NO ) [29] were kept on ice for up to 6 h . All experiments to assess NO signaling were made between 40 and 60 min of NO exposure ( NO: 200 μM SNP , 20 μM PAPA-NONOate; presented data comprise responses following incubation with either donor as they are not different from each other [Student t test , p > 0 . 05]; 500 μM SNP was used in Figs 7A , 7B , 8E and 8F [S7 and S8 Data] ) . Incubations with drugs: Zap ( PDE inhibitor ) , ODQ ( sGC inhibitor ) , L-744 , 832 , and GGTI-298 ( FTase inhibitors ) incubation for 1 h; both block FTase with an IC50: 1 . 8 nM and IC50: 203 nM , respectively [83] . Drugs were purchased from Tocris or Sigma . Statistical analysis was performed with Prism 6 . 3 and 7 and InStat 3 ( Graphpad Software Inc . , San Diego , CA ) . Statistical tests were carried out using an ANOVA test when applicable with a posteriori test ( 1-way ANOVA with Tukey’s multiple comparisons test ) or unpaired Student t test , as indicated . Data are expressed as mean ± SEM where n is the number of boutons , NMJs , or larvae as indicated and significance is shown as p < 0 . 05 , p < 0 . 01 , p < 0 . 001 , and ***p < 0 . 0001 .	One way neurons communicate with each other and with other tissues , such as muscle , is by releasing chemical compounds known as neurotransmitters at sites of interaction known as synapses . This synaptic transmission can be finely regulated by both the releasing neuron and the receiving neuron or muscle cell . Many signaling molecules and pathways are involved in neurotransmitter release . In this study , we have investigated one of such pathways and its role in modulating neurotransmitter release at the neuromuscular synapse of the larva of the fruit fly Drosophila melanogaster . This regulation involves nitric oxide , a freely diffusible reactive molecule that can be generated in response to activity in the motor neuron . Several neuronal proteins can be modified by nitric oxide , and our study identified a specific target molecule that regulates neurotransmitter release . This protein , called complexin , undergoes a posttranslational modification in response to increased levels of nitric oxide , changing its localization and function at the synapse and modulating neurotransmission . Our findings can explain how neurons may modulate communication in an activity-dependent manner utilizing nitric oxide signaling .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "invertebrates", "medicine", "and", "health", "sciences", "neurochemistry", "neuromuscular", "junctions", "vesicles", "nervous", "system", "green", "fluorescent", "protein", "electrophysiology", "neuroscience", "prenylation", "animals", "animal", "models", "farnesylation", "developmental", "biology", "drosophila", "melanogaster", "model", "organisms", "luminescent", "proteins", "experimental", "organism", "systems", "cellular", "structures", "and", "organelles", "drosophila", "research", "and", "analysis", "methods", "s-nitrosylation", "proteins", "neurochemicals", "nitric", "oxide", "life", "cycles", "chemistry", "insects", "arthropoda", "biochemistry", "larvae", "eukaryota", "anatomy", "synapses", "cell", "biology", "post-translational", "modification", "physiology", "biology", "and", "life", "sciences", "chemical", "reactions", "physical", "sciences", "neurophysiology", "organisms" ]	2018	Nitric oxide-mediated posttranslational modifications control neurotransmitter release by modulating complexin farnesylation and enhancing its clamping ability
Most identified Drosophila appendage-patterning genes encode DNA-binding proteins , whose cross-regulatory interactions remain to be better characterized at the molecular level , notably by studying their direct binding to tissue-specific transcriptional enhancers . A fine-tuned spatio-temporal expression of bric-a-brac2 ( bab2 ) along concentric rings is essential for proper proximo-distal ( P-D ) differentiation of legs and antennae . However , within the genetic interaction landscape governing limb development , no transcription factor directly controlling bab2 expression has been identified to date . Using site-targeted GFP reporter assay and BAC recombineering , we show here that restricted bab2 expression in leg and antennal imaginal discs relies on a single 567-bp-long cis-regulatory module ( CRM ) , termed LAE ( for leg and antennal enhancer ) . We show that this CRM ( i ) is necessary and sufficient to ensure normal bab2 activity in developing leg and antenna , and ( ii ) is structurally and functionally conserved among Drosophilidae . Through deletion and site-directed mutagenesis approaches , we identified within the LAE essential sequence motifs required in both leg and antennal tissues . Using genetic and biochemical tests , we establish that in the LAE ( i ) a key TAAT-rich activator motif interacts with the homeodomain P-D protein Distal-less ( Dll ) and ( ii ) a single T-rich activator motif binds the C2H2 zinc-finger P-D protein Rotund ( Rn ) , leading to bab2 up-regulation respectively in all or specifically in the proximal-most ring ( s ) , both in leg and antenna . Joint ectopic expression of Dll and Rn is sufficient to cell-autonomously activate endogenous bab2 and LAE-driven reporter expression in wing and haltere cells . Our findings indicate that accuracy , reliability and robustness of developmental gene expression do not necessarily require cis-regulatory information redundancy . Fine-tuned spatial and temporal transcriptional regulation is essential to ensure proper development [1] , [2] . Reliability of developmental gene regulation is governed by tissue-specific cis-regulatory modules ( CRM ) or “enhancers” , often situated far away from gene promoters , whereas robustness and accuracy of gene expression could be ascribed to partially-redundant “shadow” enhancers [3] . Morphogenesis of the Drosophila adult leg along the proximo-distal ( P-D ) axis offers a good model to decipher how patterning genes are tightly controlled at the transcriptional level and integrated within the well characterized limb-specific genetic cascades [4]–[9] . Nevertheless , owing to the large number of leg patterning genes encoding DNA-binding proteins and the complexity of their cross interactions , it is a challenge to determine which transcription factors ( TF ) are directly implicated in the regulation of a given P-D gene and how they functionally interact upon binding to discrete DNA sites . Toward a full understanding at the molecular level of the genetic interaction landscape governing limb development , it is thus crucial to gain better knowledge of the CRM ( s ) controlling the fine-tuned expression of each member of the underlying gene regulatory network . The Drosophila leg is composed of ten segments which are articulated to each other by characteristic joints [8] . The distal portion of the adult leg , the tarsus , is divided into five segments ( ts1–5 ) , and P-D tarsal patterning occurs by successive intercalations of new positional fates within the growing leg imaginal disc [5] , [9] . Early on during this process , wingless and decapentaplegic signalling pathways regulate the expression of transcription factors encoded by Distal-less ( Dll ) , dachshund ( dac ) and homothorax ( hth ) P-D genes [10]–[13] . Dll , dac and hth are activated in concentric domains that define the distal , medial and proximal parts of the adult appendages , respectively [14] , [15] . Together with EGFR signalling emanating from the distal-most cells , these so-called leg “gap” genes in turn regulate downstream TF-encoding genes , that include spineless , rotund , bric-a-brac1/2 , BarH1/2 and apterous , all expressed dynamically in the tarsal domain [5] , [9] . Unlike the leg , in the antennal imaginal disc Dll and hth expression domains overlap , leading to the specific activation of spalt and cut [16] . Except for the regulation of the early-acting Dll and dac gap genes [14] , [15] , few direct interactions have been established within the leg and antennal P-D regulatory cascades . Here , we choose the bric-a-brac ( bab ) locus as a model to study the integrated regulation of P-D patterning genes implicated in distal leg and antennal segmentation . The bab locus consists of two paralogous genes , bab1 and bab2 , encoding BTB transcription factors [17] , [18] . Although both genes are partially redundant in other developmental processes , only bab2 is indispensible for distal leg and antennal segmentation [17] , [18] . However , both bab1 and bab2 display dynamic expression in similar restricted P-D sub-domains with distinct expression patterns between leg and antenna [17] , [18] . Initially expressed homogeneously within the Dll-expressing distal domain in early-mid third-instar larvae ( L3 ) , the bab1/bab2 expression pattern in late L3 resolves to four concentric rings in the leg or two concentric rings in the antennal imaginal discs [17] . Later on at pupal and adult stages a P-D expression gradient within each ring is observed which is essential for ts2–4 and antennal a3–5 segment joint formation [17]–[19] . In both developing leg and antenna , all aspects of the bab2 expression pattern require the activity of the homeodomain Dll protein [19]–[21] . In addition to Dll , bab2 expression is dependent on spineless ( ss ) , at the exclusion of the distal-most ring [19] . Indeed , spineless encodes a bHLH-PAS family TF that is transiently expressed in ts1–3 from early to mid L3 stage [19] , [21] , [22] . Of note , unlike leg , ss expression is maintained in developing antenna [22] , notably under the direct control of Dll [23] . In addition to positive inputs from Dll and ss , bab2 expression is restricted proximally by dac activity [19] and distally by a gradient of EGFR signalling [21] , [24] . Lastly , graded bab2 expression , both distally and proximally , has been linked to Notch signalling via a repressive effect of bowl activity [25] , [26] . Although several bab2 regulators have been identified , to date none has been shown to be direct and no limb-specific CRM has been identified . Starting from a previous systematic identification of tissue-specific enhancers within the 150-kilobase ( kb ) -long bab locus performed by Williams et al . [27] , we characterized here an evolutionarily-conserved 567 base pair ( bp ) CRM , which reproduces expression of the bab2 endogenous gene , both in leg and antennal tissues . This CRM ( termed LAE , for leg and antennal enhancer ) is physically and functionally conserved between D . melanogaster and D . virilis . We find that the LAE is both necessary and sufficient in-vivo to ensure proper bab2 expression in leg and antennal imaginal discs , and for normal segmentation of the mature appendices . Using targeted deletions and site-directed mutagenesis , we show that leg and antennal cis-regulatory elements are closely associated . Furthermore , activation of bab2 expression in proximal- and distal-most rings is dependent on separate DNA elements . Moreover , we show that discrete essential LAE sites interact with Distal-less and Rotund transcription factors leading to bab2 activation in all or specifically in the proximal-most expressing cells , respectively . Finally , ectopic co-expression of Dll and Rn is sufficient to instruct wing and haltere cells to up-regulate bab2 . Taken together , our work indicates that a single enhancer , under the direct control of the P-D proteins Dll and Rn , is necessary and sufficient to reliably govern Drosophila bab2 expression in distinct limb morphogenetic fields . A systematic study of the 150-kb bab locus identified leg-specific cis-regulatory elements within a 11 kb region encompassing two overlapping genomic fragments ( termed BP42 and BP47 ) localized between the bab1 and bab2 transcription units ( Figure 1A ) [27] . Both BP42 and BP47 fragments also reproduce the antennal bab2 expression ( Supplementary Figure S1 ) . To identify limb-specific bab CRMs , we further dissected the relevant 11-kb region , using a sequence-directed GFP reporter assay ( see Materials and methods ) [28] . Six overlapping genomic fragments ( #1 to 6 ) ( Figure 1A ) were examined for GFP expression in both developing leg and antenna . Only fragments #3 and #4 drove strong GFP expression in leg as well as antennal tissues ( Figure S1 ) indicating that the relevant cis-regulatory information is located within the 1 . 5 kb sequence shared by BP42 and BP47 . In confirmation of this , a fragment ( #7 ) containing only this 1 . 5 kb region was sufficient to reliably reproduce bab2 expression in leg and antennal tissues ( Figure S1 ) . Before dissecting further the 1 . 5 kb fragment , we examined its evolutionary conservation among 12 Drosophila species whose genome sequences were available [29] . LAE sequences were identified and aligned . Only three >20 bp motifs ( termed CR1-3 ) are highly conserved among Drosophilidae ( Figures 1B and S2 ) . We therefore tested the activity of a 567-bp fragment ( #8 ) encompassing the CR1-3 motifs , and found that its activity faithfully recapitulated spatial and temporal bab2 expression in both developing leg and antenna ( Figure 1C–D ) . From two Drosophilidae species having diverged 40–60 millions years ago , we then tested the equivalent 0 . 7-kb D . virilis region and found that its regulatory activity was similar to that of the D . melanogaster LAE ( Figure 1D , compare G–G′ and H–H′ to E–E′ and F–F′ , respectively ) , supporting the functional importance of the conserved CR1-3 motifs . Taken together these data indicate that the evolutionarily-conserved 567-bp region contains regulatory information sufficient to recapitulate limb-specific bab2 expression , both in developing leg and antenna . We therefore termed this region LAE , for leg and antennal enhancer . To define whether the LAE is also necessary to ensure normal bab expression in-vivo , we used a P[acman] BAC construct ( 26B15 ) [30] , including the bab2 transcription unit and the LAE-containing intergenic region ( see Figure 1A ) . Adults homozygous for the babAR07 null allele display shortened tarsi with segmental joint fusions , particularly the fully penetrant fusion of ts4–5 ( Figure 2A–B ) [18] . A single copy of the intact 26B15 BAC construct was sufficient to restore normal Bab2 protein expression in both developing leg and antenna ( Figure 2A , J and O , respectively ) , as well as to rescue bab mutant phenotypes ( Figure 2A–B , arrowheads in E ) , suggesting the lack of a remote shadow enhancer located within bab1 or elsewhere in the vicinity of the bab locus . In contrast , no appendage-specific Bab2 expression could be detected for a LAE-deleted version of the 26B15 BAC construct ( Figure 2A–B , K and P ) . Further , phenotypic rescue could neither be observed ( Figure 2A–B , see arrows in F ) . We conclude that in our experimental conditions the LAE cis-regulatory module appears to be strictly required for bab2 expression and function in developing limbs . Of note , the wild-type 26B15 BAC and its LAE-minus version were both capable of partially restoring dominant abdominal pigmentation defects of babAR07 heterozygous females ( Figure S3 ) ( see discussion ) . To ask whether the LAE is also sufficient to ensure normal Bab2 expression in developing limbs , we then tested the ability of an LAE-driven bab2 cDNA construct ( LAE-bab2cDNA ) to rescue babAR07 phenotypes . Normal leg segmentation and limb-specific Bab2 expression were restored by a single copy of the LAE-bab2cDNA construct ( Figure 2A–B , G , L and Q ) , demonstrating that the LAE is both essential and sufficient for limb-specific bab2 expression and function . To functionally dissect the LAE , we tested serially truncated constructs removing either 3′ or 5′ sequences ( Figure 3 ) . Deletion of 89 bp at the 3′-end ( F3 construct ) , that removed the CR3 motif , led to a modest decrease in GFP expression ( 72 and 77% of signal strength , compared to the entire LAE , in leg and antennal tissues , respectively ) without affecting expression in the right cells ( Figure 3 , B–B′ and I–I′ ) . Deletion of 248 additional bp ( S5 construct ) further reduced the level of expression ( 35 and 32% in leg vs . antenna ) ( Figure 3 , C–C′ and J–J′ , respectively ) . Deletion of 93 additional bp at the 3′-end ( S1 construct ) , removing the CR2 sequence , led to a drastic reduction of the signal strength ( 10 and 3% ) , with a nearly-complete loss of GFP expression in the distal-most bab2-expressing cells , in either leg or antennal tissues ( Figure 3 , arrows in D–D′ and K–K′ , respectively ) . Of note , a deletion of 37 additional bp , non-strictly-conserved across all Drosophilidae , led to nearly-complete loss of all rings ( not shown ) , indicating their critical role . These 3′-deletion data support the following conclusions: ( i ) 327 bp from the 3′ half , including CR3 , are required for signal intensity; ( ii ) whereas the remaining 230 bp from the 5′-half , including CR1-2 , are sufficient for limb-specific expression , among which ( iii ) 93 bp encompassing CR2 are required for making the distal bab2-expressing ring . Conversely , a 100-bp 5′ deletion removing the CR1 sequence ( F1 construct ) led to complete loss of limb-specific GFP expression ( Figure 3 , F–F′ and M–M′ ) . This indicates a key regulatory function for CR1 in all bab2-expressing leg and antennal cells . Nevertheless , the 68-bp CR1 sequence alone ( B1 construct ) did not drive either leg- or antennal expression ( Figure 3 , G–G′ and N–N′ , respectively ) . These data indicate that CR1 is critical but not sufficient for GFP-reporter activity in both developing appendages . As the CR2-containing F1 construct deleted for CR1 did not allow GFP expression ( above ) , we conclude that CR2 is also not sufficient for LAE activity in the distal-most bab2-expressing ring . Given that the CR1 sequence is not sufficient for LAE activity , we then examined the functional significance of the poorly-conserved 32-bp-long 5′-flanking region ( see Figure S2 ) . Its deletion ( S3 construct ) led to reduced expression levels in the two proximal-most leg rings ( spanning tarsi ts1–2 and ts2–3 ) and the proximal antennal ring ( spanning a3–4 segments ) ( Figure 3 , brackets in E–E′ and L–L′ , respectively ) , but specificity retained ( i . e . , expression in the right cells at the late L3 stage ) . Taken together these data indicate that ( i ) the entire LAE is required to recapitulate normal spatial , temporal and quantitative bab2 expression patterns; ( ii ) leg and antennal expression employ shared regulatory information; ( iii ) the 337 bp 3′-part is only required for signal strength; and finally ( iv ) the 230 bp 5′-part is critical for signal specificity , with a key role of the CR1 sequence , while the 32 bp 5′-end and CR2 sequences are required quantitatively for normal expression levels in the proximal- and distal-most rings , respectively . Its central role in limb-specific expression led us to dissect the 68-bp-long CR1 sequence by substituting each 8 bp by a linker sequence ( see Materials and methods ) ( Figure 4A ) . Though all eight CR1-mutated LAE constructs ( LS1-8 ) detectably affected GFP expression qualitatively and/or quantitatively ( Figures 4B and S4 ) , four ( LS1 , 2 , 5 and 8 ) displayed strong defects . LS2 showed almost complete loss of GFP expression in both leg and antennal tissues ( Figure 4B , E–E′ and J–J′ , respectively ) , indicating it affects a crucial positive input . LS1 led to strong up-regulation in inter-ring cells in the developing leg ( Figure 4B , arrows in D–D′ ) , suggesting a distinct role in inter-ring suppression . LS5 showed globally decreased GFP expression ( 30% ) , but more pronounced in the proximal-most leg and antennal ring ( Figure 4B , brackets in F–F′ and K–K′ , respectively ) . Finally , LS8 displayed decreased GFP expression specifically in the proximal-most bab2-expressing ring , again in both leg and antennal tissues ( Figure 4B , brackets in G–G′ and L–L′ , respectively ) . As to milder effects of other mutants , ( i ) the LS4 and LS6 constructs displayed a partial GFP intensity decrease in the proximal-most ring , particularly marked in leg tissues; ( ii ) LS7 showed a partial inter-ring de-repression in leg tissues; ( ii ) whereas LS3 and LS6 exhibited a slight GFP intensity decrease specifically in antennal tissues ( 50% each ) ( Figure S4 ) . Taken together these site-directed mutagenesis data indicate ( i ) tightly-associated leg and antennal CR1 regulatory information , and ( ii ) that the LS2 sub-sequence is indispensible for LAE activity . Having identified key cis-regulatory DNA elements , we next sought positively acting factors that directly bind there . We noticed that the critical LS2 motif is embedded within an A/T-rich sequence ( AAAATTAATGGTAATAA ) , including three potential homeodomain ( HD ) -binding sites ( TAAT or ATTA motifs ) of which two were disrupted in the LS2 mutant . Given that HD-containing Dll protein is cell-autonomously required for bab2 expression [21] , we therefore examined whether the LAE-GFP reporter also requires Dll activity , using clonal analysis . As for endogenous bab2 , GFP expression was abolished in mutant leg and antennal clones for DllSA1 , a protein-null allele ( Figure 5A , D–D″′ and E–E″′ , respectively ) . This indicates that Dll is cell-autonomously required for LAE regulatory activity , and therefore suggest a direct binding of the HD-containing Dll transcription factor to the LAE , potentially through the key activating LS2 motif within CR1 . In addition to the 3 present in CR1 , the entire LAE comprises 8 additional putative HD-binding sites of which 7 are clustered within the CR2 and CR3 sequences ( Figures 5B and S2 ) . To test whether Dll is able to bind in-vitro to the TAAT/ATTA-containing LAE sequences , we used an electrophoretic mobility shift assay ( EMSA ) . All LAE fragments including one to six of the 11 TAAT/ATTA motifs bound in-vitro translated Dll , albeit with distinct affinities ( Figure 5B ) . Unexpectedly , given the key role in-vivo of the LS2 region , each TAAT or ATTA site in CR1 showed rather low in-vitro affinity ( DNA fragments #1–2 ) , when tested individually . We therefore examined whether Dll binds with a higher efficiently to a larger DNA fragment including the three TAAT/ATTA motifs present in CR1 ( i . e . , AAAATTAATGGTAATAA ) . As a matter of fact , Dll strongly bound this extended fragment ( #6 ) ( Figure 5B ) . Furthermore , stable interaction was strictly dependent upon the three TAAT/ATTA motifs . Whereas singly-mutated ( TAAT vs . CCCC ) fragments bound Dll with lower affinity , the mutation of the double overlapping TAAT/ATTA sites ( ATTAAT vs . CCCCAT; probe #9 ) showed a stronger effect , while Dll binding was abolished when all three TAAT/ATTA motifs were mutated ( probe #7 ) . To investigate whether the extended AT-rich sequence mediates in-vivo regulation by Dll , we then introduced the same TAAT-mutated sequence ( i . e . , CAAACCCCATGGCCCCAAGCA ) into the LAE-GFP reporter ( H4 construct ) . The H4 construct no longer expressed GFP either in leg or antennal tissues ( Figure 5C , F–F′ and G–G′ , respectively ) , indicating that these three HD-binding sites are crucial for LAE activity in-vivo . Surprisingly , mutation of the two overlapping TAAT/ATTA motifs was silent in-vivo ( LS3 mutant; Figure 4B ) . In fact , a new HD-binding site capable of stably interacting with Dll in-vitro ( not shown ) has been fortuitously created in LS3 ( see Figure 4A ) , providing a likely rationale for this discrepancy . Taken together , in-vitro and in-vivo data establish that Dll-dependent activation of bab2 expression in developing leg or antenna , involves direct binding to the conserved HD-binding sites of the CR1 regulatory sequence . Having shown above that normal LAE activity in the proximal-most bab2-expressing cells is mediated by its 32-bp 5′-end region ( R32 ) ( Figure 3 , E–E′ and L–L′ ) , we then sought for candidate transcription factors encoded by known limb P-D patterning genes . One , spineless , has been previously shown to regulate bab2 expression in the leg proximal-most rings [19] . However , we found no evidence for any putative binding site for the Ss/Tango bHLH-PAS heterodimer ( GCGTG ) [31] , [32] in R32 , and even in the entire LAE . A second candidate was rotund ( rn ) , a spineless target gene [9] , [33] encoding a C2H2 zinc-finger protein and whose expression corresponds to proximal tarsal segments [9] . First , we compared expression patterns of both genes , using rn-Gal4 and UAS-GFP constructs . GFP expression takes place in the right cells and at the right time [34] , but is still detected beyond the mid L3 stage due to perdurance of Gal4 and GFP proteins . At late L3 stage , perduring rn-Gal4 expression is only detected in the proximal-most Bab2-expressing tissues , and extends more proximally than bab2 , particularly in antennal tissues ( Figure 6 , A–A″′ and B–B″′; see yellow and white brackets , for distal versus proximal bab2-only and rn-only expressing cells , respectively ) . To analyse the role of rn on bab2 and LAE-driven expression , we generated mitotic clones ( GFP deficient ) of cells homozygous for the rn16 null allele . Strong cell-autonomous reduction of endogenous bab2 ( in blue ) and LAE-RFP reporter ( in red ) expression was then observed for leg and antennal clones overlapping the proximal-most bab2-expressing rings ( Figure 6 , C–C″′ and D–D″′ , respectively ) . This indicates that rn activity on bab2 is spatially restricted and contributes to bab2 expression only in ts1–2 leg and a3–4 antennal tissues . Next , we examined whether the Rn protein activates bab2 expression via the R32 sequence at the LAE 5′end , specifically required for GFP reporter expression in the proximal-most tarsal and antennal bab2-expressing ring ( s ) . To further confirm its functional requirement , we deleted the R32 sequence in the context of the 230-bp S5-GFP construct ( Figure 3 ) , recapitulating leg and antennal bab2 expression , although with lower level than the full size LAE ( Figure 7A , D–D′ and G–G′ , respectively ) . The R32-deleted S5 fragment ( S10 ) drove relatively strong GFP expression in the distal-most ring but only very weakly in the proximal-most ring ( s ) , particularly in antennal tissues ( Figure 7A , brackets in E–E′ and H–H′ ) . These data confirm that the R32 sequence is required for full LAE-driven expression in proximal bab2-expressing cells in both tarsal and antennal tissues . The R32 sequence includes a 6-bp-long oligo-T track ( T6 ) embedded within a 13-bp-long T-rich ( T13 ) sequence ( TTCGTTTTTTGTT ) , that resembles binding sites for the vertebrate Rn homolog [35] . To test whether the Rn protein is able to bind effectively to T13 in LAE , EMSA experiments were performed with DNA probes covering either the T13 sequence ( probe #1 ) alone or the complete R32 sequence ( #2 ) ( Figure 7B ) . Though in-vitro translated Rn bound both probes , the R32 fragment was bound about 5-fold more than T13 ( Figure 7B , compare lanes 1 and 2 ) . R32 includes a sequence matching the consensus binding site for Drosophila Rn , as determined by recent bacterial one-hybrid binding site data [32] ( see Figure 7B ) . The T6 track appeared critical for specific binding , as R32 and T13 fragments mutated in the T6 track ( probes #3–4 ) were not stably bound by Rn . The EMSA experiments thus indicate that the LAE T13 region constitutes a strong binding site for Rn . To examine the functional importance of the Rn binding site in-vivo , we added the T13 sequence to the truncated S10 construct to yield the H3 construct . We found that the H3-GFP reporter activity in leg and antennal tissues was similar to that of S5-GFP ( Figure 7A , brackets in F–F′ and I–I′ , respectively ) . Even though the GFP expression level remained somewhat low in the developing antenna , we conclude that the T13 sequence is sufficient for GFP-reporter activation in the proximal-most bab2-expressing rings in both the leg and antennal imaginal discs . To further confirm that T13 is essential to mediate direct up-regulation of bab2 by the Rn activator , we performed rotund gain-of-function ( GOF ) experiments , using the S5 and S10 reporter constructs . Ectopically-expressed Rn ( Dll>Rn ) activated endogenous bab2 and S5-GFP expression throughout the Dll-expressing leg and antennal cells ( Figure 7C , J–J′ and L–L′ , respectively ) , whereas the T13-deficient S10 construct displayed GFP expression neither in leg nor in antennal tissues ( Figure 7C , K–K′ and M–M′ , respectively ) . Surprisingly , the normal GFP-expressing rings of cells were also no longer detected , suggesting that Rn protein behaves as an indirect repressor , in addition to being a direct bab2 activator through the T13 sequence . Taken together with the clonal analyses , we deduce that ( i ) Rn activity is necessary ( in proximal but not in distal rings ) and sufficient ( when ectopically expressed throughout the Dll domain ) for bab2 and LAE-driven activation and ( ii ) this regulation is direct , via Rn binding to the LAE T13 sequence . The above experiments indicated that both Dll and Rn are required together for full bab2 and LAE-dependent reporter activation . This raised the possibility that Dll and Rn together can instruct cells to up-regulate the LAE . To investigate their joint instructive properties , we mis-expressed Dll , Rn or Dll+Rn in wing , haltere and eye tissues using flip-out GOF experiments [36] . bab2 is weakly expressed in restricted domains in developing dorsal appendages ( around the wing pouch and in the haltere pouch [19] ) but silent in eye tissues . Of note , LAE-driven reporter expression could not be detected either in developing wing , haltere or eye ( not shown ) . Sustained ectopic expression of Dll protein in developing wing disc induces the entire leg P-D differentiation program including bab2 expression [37] . We therefore used an hsp70-Flp construct to generate small or even single cell clones through heat induction in second- or third-instar larval tissues . In such conditions , mis-expressed Dll appeared inefficient in ectopically activating bab2 as well as LAE-RFP expression in eye , wing and haltere discs ( 0/50 , 9/180 and 0/150 examined clones , respectively ) ( Figure 8 , A–A″′; not shown ) . Significantly , the few RFP positive clones in the wing disc all corresponded to cells that normally express bab2 ( not shown ) . In equivalent analyses with mis-expressed Rn , eye , wing and haltere clones induced neither endogenous bab2 nor LAE-RFP expression ( n>50 for each tissue ) ( Figure 8 , B–B″′; not shown ) , even in wing and haltere cells normally expressing bab2 . In striking contrast , on co-expressing Dll+Rn proteins , a large proportion of examined wing and haltere clones ( 120/180 and 130/180 , respectively ) cell-autonomously activated both bab2 and LAE-RFP expression ( Figure 8 , C–C″′; not shown ) . Further , in most of the LAE-RFP non-expressing wing and haltere clones , Dll protein was not detectably accumulated ( not shown ) . Significantly , eye clones co-expressing Dll+Rn failed to activate endogenous bab2 and LAE-driven reporter genes ( not shown ) , indicating tissue specificity for their joint instructive properties . These data establish that ectopic co-expression of Dll and Rn is sufficient in instructing dorsal appendage cells to activate endogenous bab2 and LAE-driven reporter expression , presumably adopting a “proximal-ring transcriptional mode” , supporting ( i ) their direct binding to the LAE and ( ii ) a tissue-specific functional synergy involving these two transcription factors . The Drosophila leg and antenna are thought to be homologous structures evolved from a common ancestral appendage , as shown by leg-to-antenna or antenna-to-leg transformations caused by mis-expression of P-D patterning or homeotic genes [16] , [22] , [39]–[42] . To our knowledge , bab2 is the first example of a developmental gene for which a single transcriptional enhancer is shown to be both necessary and sufficient to accurately ensure a complex gene expression pattern in distinct limb morphogenetic fields . As none of our mutated LAE constructs specifically affected antennal or leg expression , our findings thus support the idea that an ancestral P-D genetic cascade emerged before limb diversification in insects . bab2 expression in developing leg and antennal discs is dynamic and complex , going from broad regional expression at early L3 stage to precisely positioned rings , and later on , to graded expression at pupal and adult stages [17]–[19] . Limb-specific regulation occurs through the 230-bp-long 5′ half of the LAE , which integrates positive inputs from both Dll and Rn transcription factors , whereas the 327-bp-long 3′ half appears to be required for signal amplification to confer robust expression , but is unable alone to drive any GFP-reporter expression . The presence of this signal intensity “booster” could explain the apparent lack of shadow enhancer [38] , as indicated by our phenotypic rescue experiments ( Figure 2 ) . Conversely , the 26B15 BAC construct partially rescues the bab mutant abdominal pigmentation defects ( Figure S3 ) , suggesting it may contain a shadow enhancer to assist the abdominal cis-regulatory elements identified previously within the bab1 transcription unit ( i . e . , excluded from the 26B15 BAC; see Figure S3 ) [27] . Even though we cannot formally exclude that the bab locus includes a remote secondary enhancer assisting the LAE in other environmental or genetic backgrounds , our data suggest that a single locus can harbor both partially-redundant ( abdominal ) and “master” ( limb ) tissue-specific transcriptional enhancers with distinct functional constraints . In this study , we have established that bab2 is a direct target of the Dll homeodomain-containing transcription factor , acting through at least the critical AAATTAATGGTAAT composite binding site present in the CR1 sequence ( Figure 9A ) . Interestingly , similar A/T-rich Dll binding sites are also present in enhancers of ss and dac ( i . e . , AATTTAATGGTAAA and AAATTATATTTAAT , respectively ) , two other direct Dll target genes [15] , [23] , suggesting a conserved CRM grammar for Dll-regulated genes . Although Dll protein is expressed throughout the larval , pupal and adult stages , onset of bab2 expression starts only at the early-mid L3 stage , in the form of a circular domain within the Dll-expressing distal territory ( Figure 9B ) [19] . Consistently , we have shown that the Dll transcription factor is required but not sufficient for cell-autonomous bab2 expression ( Figure 8A ) . In addition to the critical CR1 TAAT-rich sequence , we have shown that Dll protein also binds strongly in-vitro to the other HD-binding sites ( Figure 5B ) . It is thus formally possible that the Dll transcriptional activator may contribute to the signal intensity “booster” effect of the LAE 3′-half ( Figure 9A ) . However , signal boosting sequences situated in the middle of the LAE ( i . e . , between positions 231–478; see Figure 3 ) do not contain TAAT motifs , indicating that at least one other activating transcription factor is involved . In conclusion , in addition to Dll , characterization of new TF ( s ) interacting with LAE 3′-half sequences may help to better understand CRM activity , both in terms of tissue specificity and expression enhancement . The specific requirement for rotund activity in bab2 proximal regulation contrasts with data reported in St-Pierre et al . [34] , showing apparently normal bab2 expression in rn mutant leg discs . However , although a large bab2-expressing ring does indeed appear in mid L3 rn mutant larvae , the mature 4-ring pattern never emerges later on , and instead two presumably-distal rings are detected at late L3 stage ( not shown ) . For the first time , we report that ectopically-expressed Rn protein is sufficient to activate and maintain bab2 throughout the Dll expression domain , in both developing leg and antenna ( Figure 7C ) . In light of the dependence of rn expression in leg ts1–3 tissues on ss activity [9] , our data may provide a rationale for why ss activity is required in proximal bab2 regulation [19] . Moreover , Dll and Rn transcriptional activators , that are both necessary for full bab2 expression in leg and antennal proximal-most ring cells , are also jointly sufficient to ectopically activate bab2 ( and LAE-RFP reporter ) expression in wing and haltere but not in eye tissues , thus forming a context-specific instructive couple . The molecular basis of their functional synergy remains to be deciphered . Furthermore , in the proximal bab2-expressing domain , Dll and Rn are likely to function together with additional , still-unknown activators binding to LS4–5 and LS8 sequences ( Figures 4B and S4B ) , and whose identification will certainly provide insights into the tissue specificity of the LAE . Rotund transcription factor is required for proximal bab2 expression in both legs and antennae . However , unlike antenna , rn ( and spineless ) expression is transient in leg tissues . In addition to their previously described roles in ts1–3 growth [9] , [33] , we propose that transiently-expressed ss and rn counteract repressive activities of dac and/or bowl , both of which are dynamically expressed during the critical L3 stage [9] , [25] . In fact , de Celis and Bray [25] anticipated that a transiently-expressed bab2 activator should be present to relieve transient repression by bowl . Consistent with this view , ss activity represses bowl ( and dac ) expression [25] , [33] , in addition to its role in rn activation . As bowl ( and dac ) expression has decayed in the tarsal cells which earlier transiently expressed ss and rn , maintenance of bab2 expression would no longer require Rn activity . Maintenance of antennal Rn ( and Ss ) expression may in fact counteract additional bab2 repressors whose expression persists throughout development , such as hth and spalt ( Figure 9B ) [19] . Identification of repressive elements within the LAE will help address these issues . None of 16 different LAE reporter constructs displayed detectable de-repression in the proximal- and distal-most territories , suggesting thus the existence of functionally-redundant repressive DNA elements or alternatively of a competition between transcriptional repressors and activators for binding to the same sites . Of note , site-directed mutagenesis identified a leg inter-ring repression element ( LS1 motif ) at the CR1 3′-end . Inter-ring repression has been linked to Notch signalling [25] . However , the absence of putative Su ( H ) binding sites from the LAE sequence suggests that Notch-mediated repression is likely to be indirect . As the LS1 repressive motif is precisely located in the immediate vicinity of the critical composite HD-binding site ( Figure 9A ) , competitive binding processes may well operate between Dll and LS1-bound repressor ( s ) . Whatever the nature of the latter , a functional link with Dac , a non-specific DNA binding protein known to function as part of a multi-protein complex [43] , certainly will deserve to be investigated . Our findings , coupled with results of previous studies [19] , allow us to propose a model for distal limb-specific bab2 regulation ( Figure 9 ) . bab2 expression starts during the early-mid L3 stage at about 84 hours ( hr ) after egg laying ( AEL ) as a circular domain nested within the earlier-initiated Dll expression domain [44]–[46] . Our data indicate that Dll is required but not sufficient to activate bab2 . We suppose that a hypothetical distal activator ( X ) binding to CR2 ( Figure 9A ) may contribute to the onset of bab2 expression . In response to EGFR signalling , bab2 down-regulation in the distal-most leg territory occurs in mid L3 tissues ( at ∼90–96 hr AEL ) , giving rise to a single large ring ( Figure 9B , depicted in dark green ) . Note that EGFR-mediated bab2 repression has not been yet described in the distal antenna . From about the same stage , as rotund expression starts ( ∼84–96 hr AEL ) , a second ring emerges proximally , both in the antenna and the leg tissues ( Figure 9B , light green ) . By late L3 stage ( 120 hr AEL ) , bab2 expression consists of two well-separated rings in the developing antenna and of four concentric rings of distinct intensities in the developing leg . The two proximal-most rings ( light green ) , depending on transient Rn activity , and the two distal-most rings ( dark green ) emerge both by tarsal growth and inter-ring down-regulation ( through at least the LS1-binding repressor ) . Of note , rotund activity is indirectly required for the appearance of the two distal-most rings , due to its role in ts3 growth [9] . Lastly , in the developing leg , precise ring positioning depends on repression by EGFR and Notch signalling [21] , [24] , the molecular bases for their action remaining to be deciphered . The LAE identified in the present work is systematically located between bab1 and bab2 transcription units of 22 Drosophilidae species for which the entire bab locus sequence is available ( not shown ) . This suggests strong topological constraints during evolution . As ( i ) the duplicated bab genes are merely co-expressed during leg and antennal development [17] , ( ii ) we have shown a critical role of the LAE in ensuring bab2 expression ( this study ) and ( iii ) no other limb-specific cis-regulatory regions could be identified within the 150-kb bab locus [27] , we thus infer that the LAE is likely to reliably govern bab1 expression as well . Consistent with our assumption that strong functional constraints have operated during evolution , a LAE-like sequence with partially conserved CR1-2 sequences is even present between paralogous bab1 and bab2 genes of the tsetse fly Glossina morsitans ( Figure S5A and not shown ) , which diverged from Drosophilidae about 260 million years ago [47] . Furthermore , we have shown that the LAE from D . virilis ensures normal regulatory functions in D . melanogaster . Although strongly conserved among Sophophora subgenus species ( i . e . D . melanogaster group ) , the T13 Rn-binding site is poorly conserved in D . virilis and related Drosophila subgenus species ( Figure S5B ) . It is formally possible that Rn has been co-opted recently in the Sophophora subgenus . Alternatively , D . virilis Rn could act through subgenus-specific LAE sequences . Consistent with this hypothesis , a T13-related sequence , only conserved among Drosophila subgenus species , is located between CR1-2 ( Figure S5B ) . Investigating the molecular basis of Rn action in the positive regulation of bab2 may provide an entry point to tackle these evolutionary issues . Drosophila lines were grown on standard yeast extract-sucrose medium . The vasa-PhiC31 ZH2A attP stock was kindly provided by F . Karch and was used to generate most of the transgenic GFP and RFP reporters and the two P[acMan] BAC constructs . rn or Dll mutant clones were generated by 30 minute ( mn ) heat shocks at 38°C , in early first- to late second-instar larvae of genotypes: ( i ) y w hsFlp; FRT82B Ub-GFP/FRT82B rn16 and ( ii ) y w hsFlp; arm-lacZ FRT42D/DllSA1 FRT42D , respectively . Flip-out clones over-expressing either Dll , Rn or both Dll plus Rn were generated by 40mn heat shocks at 38°C , in second- or third-instar larvae of genotypes: ( i ) y w LAE-RFP hsFlp; Pact>y+>Gal4 , UAS-GFP/UAS-Dll , ( ii ) y w LAE-RFP hsFlp; Pact>y+>Gal4 , UAS-GFP/UAS-Rn1 and ( iii ) y w LAE-RFP hsFlp; Pact>y+>Gal4 , UAS-GFP/UAS-Dll UAS-Rn1 , respectively . UAS-GFP . [S65T] , UAS-Rn1 and rn-Gal4 lines were obtained from the Bloomington stock center . UAS-Dll and DllEM212-Gal4 line were provided by S . Cohen and M . Suzanne , respectively . Genomic DNA fragments from the D . melanogaster or D . virilis bab locus were amplified by standard PCR ( using the following primer pairs: cccgaattcGCGCCTAACTAGCCAACAAT/cccggatCCTTTGACTCCGCTTTCGTCTTC and cccgaattcGAAACATCACGTTATCTAGCCACA/cccggatccAGAGTTGCTTGCACACACTCAC , respectively; BamHI and EcoRI restriction sites are underlined ) , cloned into pBP-S3aG , a home-made derivative of the attB-containing pS3aG plasmid obtained from T . M . Williams and S . Carroll [27] . Of note , the pBP-S3aG vector includes the TATA-less bab2 minimal promoter . The pLAE-RFP and pLAE-Bab2 plasmid constructs were made by substituting the GFP insert of the pLAE-GFP construct by a pH2B-RFP insert ( obtained from A . Vincent ) or a bab2 full-length cDNA ( from pNB-bab2 ) , respectively . Site-directed mutagenesis ( including linker scanning ) was performed by PCR , using the overlap extension method [48] . All constructs were sequence-verified . BAC recombineering and PhiC31-mediated germline transformation were performed as described [49] . The LAE was validated through insertions within several attP genomic sites , including the 2A platform on the X chromosome that was used for all constructs reported in this study . The genomic sequences homologous to the D . melanogaster LAE were recovered by Blat analysis at the UCSC Genome Browser website ( http://genome . ucsc . edu/cgi-bin/hgBlat ? command=start ) and aligned with MAFFT ( http://mafft . cbrc . jp/alignment/server/ ) . The multiple alignment was then shaded with Boxshade ( http://www . ch . embnet . org/software/BOX_form . html ) . Third-instar larval imaginal discs were prepared and stained using standard procedures . Confocal analyses were done with a LEICA TCS SP5 microscope . Rat anti-Bab2 [17] and rabbit anti-Dll [50] antibodies were used at 1/2000 and 1/200 , respectively . For each reporter construct , GFP fluorescence quantification was obtained from 10 distinct T2 leg and eye-antennal discs ( i ) dissected from late third-instar larvae grown in the same environmental conditions ( temperature and larval density ) , and then ( ii ) fixed in the same conditions . Confocal images were analyzed with ImageJ software , using the same area ( centered on the distalmost ring of cells ) , laser excitation settings ( 75% maximal detection ) and brightness/contrast image acquisition . Dll and Rn proteins were synthesized by coupled in-vitro transcription/translation with T7 RNA polymerase and rabbit reticulocyte lysate ( TNT assay , Promega ) . pET-Dll and pCS2-MycRn plasmid constructs were obtained from S . Cohen and P . Couso , respectively . EMSA experiments were performed mainly as described [51] , using Novex 6% DNA retardation gels ( InvitroGene ) . Probes were assembled from synthetic oligonucleotides including 4 additional G bases at their 5′-ends , labeled with the Klenow fragment of E . coli DNA polymerase I in presence of [α-32P]dCTP , and purified on mini Quick spin columns ( Roche ) . Specific activities were roughly similar for all tested probes . Free and shifted probes were revealed with a PhosphorImager .	In insects , leg and antenna are homologous limbs , though derive from a single ancestral appendage . In Drosophila , leg and antennal development along the proximo-distal ( P-D ) axis relies on relatively-well known genetic cascades , in which most appendage-patterning genes encode transcription factors ( TF ) . However , their cross-regulatory interactions remain to be better characterized at the molecular level . A fine-tuned expression of the bric-a-brac2 ( bab2 ) gene is essential for normal leg and antennal segmentation . However , within the genetic cascades governing P-D limb development , no TF directly controlling bab2 expression has been identified to date . We show here that restricted bab2 expression in developing leg and antenna is governed by a single enhancer , termed LAE , which is necessary and sufficient in-vivo to ensure bab2 functions there . We show that leg and antennal cis-regulatory elements are closely associated and that essential LAE sites interact with Distal-less ( Dll ) and Rotund ( Rn ) TFs , leading to bab2 activation in all or specifically in the proximal-most expressing cells , respectively . Finally , joint ectopic expression of Dll and Rn is sufficient to instruct wing and haltere cells to up-regulate bab2 . Taken together , our work indicates that a single enhancer is necessary and sufficient to reliably govern bab2 expression in distinct morphogenetic fields .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "developmental", "biology", "model", "organisms", "molecular", "cell", "biology", "genetics", "biology", "genomics", "evolutionary", "biology" ]	2013	Drosophila Distal-less and Rotund Bind a Single Enhancer Ensuring Reliable and Robust bric-a-brac2 Expression in Distinct Limb Morphogenetic Fields
Toxoplasma gondii has evolved a number of strategies to evade immune responses in its many hosts . Previous genetic mapping of crosses between clonal type 1 , 2 , and 3 strains of T . gondii , which are prevalent in Europe and North America , identified two rhoptry proteins , ROP5 and ROP18 , that function together to block innate immune mechanisms activated by interferon gamma ( IFNg ) in murine hosts . However , the contribution of these and other virulence factors in more genetically divergent South American strains is unknown . Here we utilized a cross between the intermediately virulent North American type 2 ME49 strain and the highly virulent South American type 10 VAND strain to map the genetic basis for differences in virulence in the mouse . Quantitative trait locus ( QTL ) analysis of this new cross identified one peak that spanned the ROP5 locus on chromosome XII . CRISPR-Cas9 mediated deletion of all copies of ROP5 in the VAND strain rendered it avirulent and complementation confirmed that ROP5 is the major virulence factor accounting for differences between type 2 and type 10 strains . To extend these observations to other virulent South American strains representing distinct genetic populations , we knocked out ROP5 in type 8 TgCtBr5 and type 4 TgCtBr18 strains , resulting in complete loss of virulence in both backgrounds . Consistent with this , polymorphisms that show strong signatures of positive selection in ROP5 were shown to correspond to regions known to interface with host immunity factors . Because ROP5 and ROP18 function together to resist innate immune mechanisms , and a significant interaction between them was identified in a two-locus scan , we also assessed the role of ROP18 in the virulence of South American strains . Deletion of ROP18 in South American type 4 , 8 , and 10 strains resulted in complete attenuation in contrast to a partial loss of virulence seen for ROP18 knockouts in previously described type 1 parasites . These data show that ROP5 and ROP18 are conserved virulence factors in genetically diverse strains from North and South America , suggesting they evolved to resist innate immune defenses in ancestral T . gondii strains , and they have subsequently diversified under positive selection . Intracellular parasites have to contend with immune responses mounted by the hosts they infect if they are to survive long enough to effectively transmit . The parasite Toxoplasma gondii is one of the more successful parasites in terms of transmission as it can infect all mammals and many birds [1] , both of which serve as intermediate hosts where the parasite propagates asexually . The parasite undergoes sexual recombination only within the intestinal tract of members of the Felidae family when these predators ingest chronically infected intermediate hosts [2] . Transmission by this predator-prey life cycle has shaped the co-evolution of parasite virulence factors and host responses . T . gondii readily infects humans but mainly causes disease in situations where the immune system has become compromised , although associations between parasite genotype and disease severity may also occur in healthy individuals [3] . The association of virulence with parasite genotypes has been more fully investigated in laboratory mice , where this trait can be more easily assessed and where the parasite has likely had more opportunities to evolve within a host that is common prey of cats . Initially it was shown that parasite isolates from Europe and North America differed in their ability to cause disease in laboratory mice . Type 1 strains lead to lethal infection in all conventional strains of laboratory mice including outbred animals ( LD100 = 1 ) during the first two weeks of infection ( i . e . the acute phase ) , and any survivors observed at low inoculum are invariably not infected ( i . e . they remain serologically negative ) [4 , 5 , 6] . In contrast , type 2 strains show intermediate virulence ( LD50 varies with mouse strain , although these strain generally do not cause mortality in outbred lines ) , and type 3 strains are avirulent ( they are not lethal in inbred mice at any dose ) , respectively [4 , 5 , 6] . These phenotypic differences prompted the generation of genetic crosses between these three strain types , resulting in the identification of the rhoptry kinase ROP18 as a virulence factor in genetic crosses between type 1 and type 3 or between type 2 and type 3 strains [7 , 8] , and the rhoptry pseudokinase ROP5 as a virulence factor in genetic crosses between type 2 and 3 or between type 1 and 2 strains [9 , 10] . Together these proteins are involved in resisting innate immune mechanisms that are induced by IFNg stimulation [11] . The parasite secretes ROP5 and ROP18 into the host cell during invasion where they co-localize to the surface of the parasitophorous vacuole [12] . ROP18 is tethered to the vacuole membrane by an N-terminal low complexity region [13] and this association is required for its virulence enhancing properties [14] . At the vacuole surface , ROP18 phosphorylates host immune-related GTPase ( IRGs ) and prevents them from accumulating on the membrane and destroying the parasite vacuole [12 , 15] . The association of ROP18 with ROP5 increases the phosphorylation activity of ROP18 [16] , and may also make substrates available to the kinase by binding to members of the IRG family [17 , 18 , 19] . Biochemical studies subsequently showed that ROP5 also associates with ROP17 , and that this kinase works synergistically with ROP18 to phosphorylate and block IRGs [20] . Deletion of either ROP17 or ROP18 alone leads to a modest attenuation in the type I RH strain , evident as a delay in time until death; however , mice still succumb to low challenge doses ( i . e . 100 parasites ) . In contrast , deletion of both ROP17 and ROP18 leads a marked attenuation of virulence where mice survive high challenge does ( ≥ 105 parasites ) [20] , similar to the deletion of ROP5 ( ≥106 parasites ) [9 , 10] . South American stains are much more genetically diverse than those in the North and our current estimate of the population structure consists of 6 major clades that contain a total of 16 distinct haplogroups [21 , 22] . Notably , most common haplotypes from South America are also virulent in mice [23] , similar to the type I lineage of North America where this phenotype is otherwise rare . Importantly , South American strains also exhibit the trait that a single infectious organism is lethal in all conventional laboratory mice ( i . e . LD100 = 1 ) and they do not give rise to serologically positive surviving animals [23] . Previous studies have shown that ROP5 alleles in North and South American strains are correlated with resistance to IRG coating of the parasitophorous vacuole [18] and that expression of ROP18 from the South American strain RUB ( type 5 ) confers virulence to an avirulent type 3 parasite [24] . However , the precise contributions of ROP5 , ROP18 , or other genetic factors , to acute virulence in these strains are unknown . For that reason , we mapped the genetic basis for differences in acute virulence between the intermediate virulence type 2 ME49 strain and the highly virulent type 10 VAND strain using a recently conducted genetic cross [25] . The type 2 ME49 parental strain was isolated from a sheep in the USA [26] , whereas the type 10 VAND parental strain was originally isolated from an immunocompetent human adult in French Guiana [27] . Using quantitative trait locus ( QTL ) mapping and CRISPR-Cas9 generated knockout ( KO ) lines we were able to identify ROP5 as the single major virulence factor underlying the virulence differences in laboratory mice between type 2 and type 10 parasites . Additionally assessment of the contribution of ROP18 to virulence in several South American strains revealed a stronger phenotype than has been previously described for North American type 1 strains . These findings demonstrate that the genetic basis of mouse virulence is conserved across diverse lineages of T . gondii . We utilized a previously described genetic cross between type 2 ME49-FUDRr and type 10 VAND-SNFr strains [25 , 28] . Whole-genome sequencing of 24 progeny and mapping based on genome-wide SNP analysis was previously used to identify the molecular basis of sinefungin resistance [29] . Here we used the existing genetic map from this cross to analyze virulence of 24 recombinant progeny in CD-1 outbred mice . The parental ME49-FUDRr strain generally does not cause lethal infection in CD-1 mice and 7 of the progeny inherited this trait , where most or all of the mice infected with these progeny survived the 30 day course of the experiment ( Fig 1A ) . There were 14 progeny that acquired the virulent trait from the VAND-SNFr parent , exhibiting ≥ 75% mortality in CD-1 mice ( Fig 1A ) . Only three progeny showed an intermediate phenotype , with 50% of the mice succumbing to infection when infected with these progeny ( Fig 1A ) . A genome-wide primary QTL scan of virulence ( % mortality ) run as a continuous phenotype generated one significant peak between the physical locations of 0 . 48 MB and 2 . 94 MB on chromosome XII with a log10 of odds ( LOD ) score of 6 . 95 ( Fig 1B ) . A second QTL was detected in the primary scan with a LOD score of 2 . 54 on chromosome Ia; however , this peak did not reach significance ( Fig 1B ) . The single peak on chromosome XII accounted for 71% of the effect-size , or variance in the virulence phenotype . A secondary scan with the primary QTL on XII run as an additive covariate failed to produce additional significant peaks ( Fig 1C ) . Although no secondary peaks were identified , analysis of the virulence phenotype using a two-locus model identified a significant interaction between the primary peak on XII and chromosome VIIa , with a LOD score of 7 . 68 . This finding is of interest as the known virulence factors ROP5 and ROP18 are located on chromosomes XII and VIIa , respectively . Because the peak on XII includes the tandem repeat of the ROP5 gene that has been shown to be a virulence factor in other strains , we chose to investigate ROP5 further . To delete the large locus spanning all copies of ROP5 on chromosome XII , we modified the T . gondii CRISPR-Cas9 plasmid [30] to express two separate single-guide RNAs ( sgRNAs ) . The two sgRNAs target unique sites upstream and downstream of the ROP5 locus ( Fig 2A ) . When expressed in T . gondii , the sgRNAs should generate two double-strand breaks in the genome and allow for replacement of the intervening region with a selection cassette containing homologous regions outside the sgRNA cut sites . This strategy is expected to generate a knockout of all ROP5 alleles including the expanded copy number ROP5 genes represented by TGVAND_308090 , and two adjacent single copy genes that encode predicted pseudokinases that are paralogs of ROP5: TGVAND_308093 and TGVAND_308096 ( Fig 2A ) . We co-electroporated the double-CRISPR plasmid with a loxP-DHFR-mCherry-loxP selection cassette into VAND and acquired stable parasites after selection with pyrimethamine . A PCR screen for integration of the loxP-DHFR-mCherry-loxP selection cassette at the ROP5 locus identified parasite clones that were positive for homologous integration ( Fig 2B ) and lacked all ROP5 coding regions ( S1A Fig ) . Western blot analysis with rabbit α-ROP5 confirmed that no protein expression was detectable ( Fig 2C ) . Wild type VAND and two different clones of VANDΔrop5 parasites were injected i . p . into CD-1 mice to test virulence . Only mice infected with wild type parasites succumbed to infection . Mice infected with VANDΔrop5 survived the course of the experiment , even at a high dose of 105 parasites ( Fig 2D ) . Surviving mice from these experiments ( Fig 2D ) were tested for the generation of T . gondii antibodies via ELISA to ensure they were initially infected ( S2 Fig ) . To confirm that ROP5 is sufficient for virulence we complemented the VANDΔrop5 parasite with the TOXOM52 cosmid that spans the _308090 tandem ROP5 locus , but does not include the adjacent _308093 or _308096 genes . TOXOM52 transgenic parasites were positive by PCR for the ROP5 coding sequence ( S1A Fig ) and showed wild type levels of ROP5 protein expression ( Fig 2C ) . Mice infected with VANDΔrop5::TOXOM52 succumbed to infection in a similar timeframe as those infected with wild type parasites , demonstrating that the tandemly repeated _308090 ROP5 alleles are sufficient to restore virulence to wild type levels ( Fig 2D ) . These results confirm that ROP5 is the major locus responsible for virulence differences between type 2 ME49 and type 10 VAND . Genetic crosses have now implicated ROP5 as a major virulence factor using type 1 GT1 [9] ( Clade A ) , type 2 ME49 ( Clade D ) , type 3 VEG ( Clade C ) [7 , 10] and type 10 VAND ( Clade F ) parasites as parental strains . These strain types represent a broad , yet incomplete , spectrum of the global T . gondii genetic diversity ( Fig 3A ) [21 , 22] . Given this , we were interested to test the contribution of ROP5 to virulence in strains from Clade B , a genetically distinct group not represented by the currently available genetic crosses . Using the double-CRISPR strategy applied to generate the VANDΔrop5 strain , we created Δrop5 parasites for virulent South American type 8 TgCtBr5 and type 4 TgCtBr18 strains . Knockouts were confirmed by PCR screening of clones for integration of the selection cassette ( S3A and S3B Fig ) , loss of the ROP5 coding region ( S1B and S1C Fig ) , and loss of ROP5 protein expression by Western blotting ( Fig 3B and 3D ) . Mice succumbed to infection with wild type strains of TgCtBr5 or TgCtBr18 , yet when mice were infected with either TgCtBr5Δrop5 or TgCtBr18Δrop5 they survived the course of the experiment ( Fig 3C and 3E ) . These data demonstrate that ROP5 is also a major contributor to virulence in Clade B strains of T . gondii . ROP5 and ROP18 have previously been shown to function together to resist host IRG-mediated parasite killing [16 , 18] . Although ROP5 plays a dominant role in many strains , the effect of deleting ROP18 varies with different backgrounds . For example , deletion of ROP18 results in a delay in the time until death in the RHΔrop18 strain [12] and a modest shift in the LD50 in GT1Δrop18 strain [30] parasites . Consequently , we assessed the role of ROP18 in the virulence of South American VAND strains . Using the sgRNA-ROP18 CRISPR disruption strategy described previously [30] , we deleted ROP18 from the VAND-SNFr parental strain . PCR screening for integration of the DHFR* selection cassette at the ROP18 locus was confirmed in stable clones ( S4 Fig ) , resulting in the loss of ROP18 protein expression , as determined by Western blotting ( Fig 4A ) . Unlike the delayed death phenotype seen previously for type 1 RHΔrop18 parasites , we observed 100% survival rates for CD-1 mice infected with two different clones of VANDΔrop18 parasites ( Fig 4B ) , even at a dose of 105 parasites . Virulence was restored in VANDΔrop18 ( Fig 4B ) when it was complemented with a ROP18-Ty tagged version of the protein ( Fig 4A ) . To determine how the loss of ROP18 would affect virulence in Clade B strains , we knocked out ROP18 in TgCtBr5 and TgCtBr18 using the same strategy used in VAND . After pyrimethamine selection we obtained parasite clones that integrated the DHFR* selection cassette at the ROP18 locus ( S4B and S4C Fig ) , resulting in the loss of expression of ROP18 as determined by Western blotting ( Fig 4C and 4E ) . Similar to type 10 VAND parasites lacking ROP18 , TgCtBr5Δrop18 and TgCtBr18Δrop18 parasites were completely avirulent in CD-1 mice , even at a dose of 105 ( Fig 4D and 4F ) The locus encoding ROP5 shows evidence of expansion of tandem copies of the gene in different strains of T . gondii [10 , 16] . Alignments of genome-wide sequence to the assembled ME49 chromosomes revealed that the copy number for the ROP5 gene ( _308090 ) varies between strains , whereas the more divergent _308093 and _308096 ROP5 paralogs that lie adjacent in the genome are each single copy ( Fig 5A ) [31] . The number of copies of ROP5 ( _308090 ) ( i . e . copy number variation ( CNV ) ) for each of the strains was estimated by comparing the trace reads to the CDS , yielding the following estimates: GT1 ( 5 ) , ME49 ( 10 ) , VEG ( 4 ) , TgCtBr18 ( 6 ) , TgCtBr5 ( 7 ) , and VAND ( 8 ) ( Fig 5A and 5B ) . Previous studies have shown that each strain contains a dominant allele that is expressed at higher levels based on CNV . Dominant alleles have been referred to as “C” , while less common alleles were labelled as “A” or “B” , in work reported by Resse et al . , [10] , while these two categories were called M for major ( C alleles ) and m for minor alleles ( B and A alleles ) by Behnke et al . , [9] . ROP5 alleles in VAND and TgCtBr5 were also previously named using A and B alleles , based on how they grouped phylogenetically with a collection of various strains including the type 2 strain ME49 [18] . We were able to estimate the CNV for the individual ROP5 alleles from these alignments for the strains VAND and TgCtBr5 ( Fig 5C ) . These genes can also be defined as major and minor alleles based on copy number: TgCtBr5-B1 corresponds to the major allele ( M ) , while TgCtBr5-B2 is a minor allele ( m1 ) ; VAND-B2 is the major allele ( M ) ; VAND-B1 , VAND-B3 , and VAND-A are minor alleles ( m1-3 , respectively ) . Although the number of ROP5 copies found in a strain doesn’t correlate with virulence , amplification of specific alleles types is associated with virulence in the murine model [10 , 16] . For example , the major alleles shared by type 1 and 3 strains ( GT1-M , VEG-M corresponding to ROP5CI/III ) are associated with virulence , while those from type 2 ( ME49-M or ROP5CII ) are not [9 , 10] . To expand on this relationship , we conducted phylogenetic analysis using the previous sequences for ME49 , GT1 , VEG [16] and VAND , TgCtBr5 [18] ROP5 alleles that are available from NCBI . Phylogenetic analysis of the ROP5 protein sequences revealed a lack of monophyly among the three strain types ( Fig 5B ) . However , most alleles from the virulent strains group together , with VAND-B1 ( m1 ) , VAND-B2 ( M ) , TgCtBr5-B1 ( M ) , and TgCtBr5-B2 ( m1 ) being closer to GT1-M . This cluster also included the major type 3 allele in VEG ( VEG-M ) , which is functionally similar to the major allele in type 1 strains ( the avirulence of VEG is due to hypo-expression of ROP18 not a defect in ROP5 ) . In contrast , GT1-m2 , VEG-m2 , and VAND-A from virulent parasites group with the avirulent ME49-M and ME49-m1 alleles ( Fig 5B ) . VAND-B3 appeared as a more divergent allele occurring on a long branch at the base of the tree ( Fig 5B ) . The function of ROP5 alleles in virulence may be attributed to differences the polymorphic surface of ROP5 , which interacts with host IRGs and likely contributes to IRG specificity [19] . Indeed , when we analyzed the ratio of non-synonymous to synonymous mutations we detected 23 codon sites evolving under significant positive diversifying selection , with the majority of codon sites ( 74% of positive codons ) localized within the regions encoding the polymorphic surface of ROP5 ( S1 Table and Fig 5C ) . Interestingly , one codon site that evolves under positive diversifying selection was found within the DFG motif that is normally part of the activation loop in active kinases ( S1 Table and Fig 5C ) . To provide greater insight into the region of ROP5 that is under diversifying selection , we mapped these residues on the recently described binding site for Irga6 as revealed by a X-ray co-crystal structure [19] . A majority of the residues in ROP5 that are under positive selection cluster tightly to a region that interacts directly with Irga6 ( Fig 6A ) . A total of 13 residues ( 56 . 5% ) under positive selection are directly at the interface between ROP5 and Irga6 ( purple in Fig 6B , S1 Table ) , as identified by PDBePISA [32 , 33] . An additional 3 residues ( 13% ) under positive selection are close to the interface and may play a supporting role in stabilizing the primary interface residues ( S1 Table and Fig 6 ) . Our studies extend the application of forward genetic crosses and reverse genetic engineering using CRISPR-Cas9 to South American strains to provide an expanded framework for examining the basis of acute virulence in the mouse model . These studies reveal a conserved role for ROP5 and ROP18 in acute virulence in the mouse , likely due to their previously demonstrated role in protecting the parasite from host innate immune responses [12 , 15] . The identification of only a single significant QTL in analyzing phenotypic differences among progeny from the genetic cross between type 2 ME49 and type 10 VAND led to the identification of ROP5 as the major locus in controlling acute virulence in the mouse . Although ROP18 was not detected in a primary scan , a two-locus interaction occurred between regions containing ROP5 ( chromosome XII ) and ROP18 ( chromosome VIIa ) , suggesting that these two genes collectively explain the major virulence differences among diverse strains of T . gondii in the mouse . This prediction was borne out by reverse genetic engineering of ROP5 and ROP18 disruptant mutants in other South American strains , the latter of which showed more severe defects in acute virulence than previously observed in the type 1 background . The widespread importance of these two virulence factors suggests that they evolved before the global spread and divergence of the parasite , consistent with recent reports that functional alleles of ROP5 and ROP18 also exist in the close relative Hammondia hammondi [34] , even though this parasite is not virulent in mice . Collectively , these data suggest that ROP5 and ROP18 have been important in the evolution of these parasites within their rodent hosts , likely due to their interaction with the IRG family . Despite extensive polymorphism between alleles , the major difference in the contribution of ROP18 to mouse virulence is due to different expression levels , being high in types 1 and 2 , and very low in type 3 [7 , 8] . Complementation of a type 3 strain by over-expression of either ROP18I [8] or ROP18II [7] was sufficient to restore virulence . Additionally previous studies have shown that complementation of an avirulent type 3 strain with type 1-like alleles of ROP18 from South American strains results in a similar gain of virulence [24] . Combined with the fact that ROP18 was not seen as a major QTL in the cross between type 1 and 2 , where ROP5 was mapped [9] , these data suggest that different alleles of ROP18 function similarly , and their differential contribution to virulence is primary due to expression level . Given these previous findings , it is perhaps not surprising that ROP18 was not identified as a major QTL in the present cross . However , it was somewhat surprising that ROP5 was identified as the gene responsible for the majority of difference in acute virulence between type 2 ME49 and type 10 VAND , given the extensive genetic differences between these strains [21 , 22] . The mechanism by which ROP18 and ROP5 thwart innate immunity has been well established in North American strain types , where the combination of alleles present in type 1 strains leads to resistance to IRG loading relative to type 2 or 3 strains [11] . Although we did not test the recruitment of IRGs directly in this study , the role of ROP5 and ROP18 in the virulence of South American strains is likely due to their function in resisting host IRG recruitment to the parasitophorous vacuole . Consistent with this prediction , previous studies have shown that in IFNg-activated host cells , vacuoles containing virulent VAND ( type 10 ) and TgCtBr5 ( type 8 ) parasites recruit less Irgb6 than susceptible type 2 and 3 parasites [18] . The only other single peak observed on chromosome Ia was below the threshold of significance . In a previous genetic cross between type 1 and 3 , a minor QTL was detected on chromosome Ia from markers M48 ( ~330 kbp ) –AK4 ( ~485 kbp ) with a LOD score of 2 . 38 [8] . However , in the present study , the QTL peak on chromosome Ia lies at marker MV24 near the right end of the chromosome ( ~1 , 500 kbp ) with a LOD score of 2 . 54 , and a marginal level of significance . There is also a second non-significant peak in the center of the chromosome from MV2 ( ~570 kbp ) to MV9 ( ~830 kbp ) . These findings suggest that genes on this chromosome play a minor role in acute virulence in the mouse , although the resolution of the present data is insufficient to precisely map these or to identify candidate genes . Although forward genetic mapping allowed us to identify a locus containing ROP5 as important in the virulence of South American strains , the newly adapted CRISPR-Cas9 technology was critical for testing its function by reverse genetics . The default DNA damage repair pathway in T . gondii is nonhomologous DNA end joining , and for this reason the use of a Δku80 background is important to efficiently generate knockout parasites [35 , 36] . Several attempts were made to generate a VANDΔku80 strain for use in this study but all were unsuccessful , perhaps because VAND is a low passage isolate that has not been adapted to in vitro culture conditions for long periods of time . However , gene disruption using CRISPR-Cas9 is highly efficient even in wild type KU80 proficient parasites , as shown previously [30] , and this technique was also successfully applied here to diverse strains . We also extended this strategy by creating a CRISPR-Cas9 plasmid that expressed two sgRNAs targeting sites surrounding a large genomic locus . This strategy was efficient at deleting the ~30 kb ROP5 locus in three different low-passage South American strains . The acute virulence phenotype of VANDΔrop5 mutant was restored using a cosmid that only contains the ROP5 cluster , but not the adjacent pseudokinase paralogs , indicating that these adjacent genes contribute little to the phenotype and that the ROP5 alleles are the primary determinants of acute virulence . Phylogenetic analysis revealed that VAND and TgCtBr5 contain ROP5 alleles that are similar to the major alleles in the virulent type 1 strain GT1 , with only minor alleles being similar to the type 2 strain ME49 . Based on this similarity , we decided to complement the VAND ROP5 locus knockout with a cosmid containing the cluster from the type I RH strain . We have not tested the function of individual alleles in restoring this phenotype , nor complemented directly with individual VAND alleles , and further testing would be needed to establish if they have conserved functions . Interestingly , all of the ROP5 alleles contain a functional change in HRD motif , that forms part of the catalytic triad , where major alleles in the virulent strains ( those related to GT1-M corresponding to ROP5CI ) and some minor alleles such as GT1-m1 ( ROP5BI ) contain “H” , while the major allele in the type 2 strain ME49 ( ME49-M , ROP5CII ) , and minor alleles such as VAND-A , GT1-m2 ( ROP5AI ) , and VEG-m2 contain a “R” . In contrast , VAND-B1 and VAND-B2 alleles show a new residue in the catalytic triad of “L” , which again is predicted to lack kinase activity . None of these residues is predicted to be active as this residue is normally a D in active kinases [37] . The functional significance of these changes in the catalytic triad of ROP5 is unclear as the interface that interacts with IRGs is on the opposite face of the molecule [19] . However these different alleles also correlate with differences in function in laboratory mice: the former group , including GT1-M , GT1-m1 , and related alleles , inhibit recruitment of IRGs to the parasitophorous vacuole in vivo , while the later group , including ME49-M and related alleles , do not [16 , 17] . The diversity of alleles found in ROP5 suggests that alleles that lack activity in the mouse may be active in others hosts , for example other rodents or birds . Previous studies have emphasized that this binding interface is polymorphic in ROP5 [19] , and our analysis of residues that are under strong positive selection in ROP5 further defines this interface as a cluster of residues that directly interacts with Irga6 . Additional nearby polymorphic residues in ROP5 may act to influence the Irga6 interaction through allosteric interactions , or they may be important in binding to other polymorphic alleles of Irga6 . Importantly , IRGs also show polymorphism in this region of the molecule [19] , suggesting that ROP5 is under selective pressure for enhanced binding while IRGs have evolved to avoid recognition . In support of this model , binding of ROP5 to Irga6 has been shown to directly limit polymerization in vitro [19] , as well as to enhance phosphorylation of IRGs by ROP18 in vitro and in vivo [16] . Collectively , these findings provide evidence for the hypothesis that the polymorphic regions of ROP5 are evolving under positive diversifying selection driven by interaction with host innate immunity factors . We were also able to efficiently delete ROP18 in different genetic backgrounds , demonstrating the potential of CRISPR-Cas9 for genetic manipulation of diverse strains of T . gondii . Unlike ROP5 knockouts where all Δrop5 strains tested are equally avirulent despite different genetic backgrounds , there are large differences in the phenotype of Δrop18 disruptants between strains . For example , previous studies have reported that deletion of ROP18 in the type 1 RH strain led to delayed time until death , but not a change in the lethal dose [12] vs . a several log increase in the lethal dose using the type 1 GT1 strain [30] . In contrast , deletion of ROP18 in type 4 , 8 , and 10 strains led to a complete loss of virulence with no animals dying at a challenge dose of up to 105 ( present study ) . The reasons for these differences are uncertain but are unlikely to be due to experimental variation in the mouse model since they were obtained in the same laboratory using a standard protocol that is highly reproducible . Type 1 strains normally result in lethal infection in all strains of conventional laboratory mice with an inoculum of a single parasite [4 , 5 , 6] . In contrast animals infected here with Δrop18 mutants in type 4 , 8 , and 10 did not succumb to infection , yet had very high antibody titers , reflecting an active infection . Previous studies have stressed that the major contribution of ROP18 in acute virulence appears to be due to expression level differences , and yet all the strains tested here show high levels of expression [24] . Hence , it is not immediately clear why the contribution of ROP18 to acute virulence would vary so extensively among the strains tested here . Once possibility is that the small number of polymorphisms in ROP18 alleles found between type 1 strains and the isolates studies here may account for these differences . Consistent with this , previous studies have also shown that virulent alleles of ROP18 are also under diversifying selection [24] . The marked differences in attenuation with the deletion of ROP18 might also be due to the relative importance of other genes that contribute to acute virulence . For example , the active rhoptry kinase ROP17 [20] or the secreted dense granule GRA7 [38] , which have been shown to be important in the type I RH strain , may partially compensate for loss of ROP18 in some backgrounds . The fact that ROP18 plays a less important role in the RH strain , may also be due to its passage history , which has been propagated extensively in mice and in vitro since its original isolation more than 75 years ago [39] , resulting in numerous changes that favor growth . Further studies of the precise function of various ROP18 alleles in different genetic backgrounds would be necessary to decipher among these alternatives . Despite their conserved functions in the murine model , ROP5 and ROP18 do not appear to mediate resistance to IFNg-activated control of replication in human cells [18] . Because humans lack the expanded IRG family [40] , other mechanisms of control likely contribute to cell autonomous control of replication . However , ROP18 has also been shown to phosphorylate host cellular protein ATF6-β targeting it for degradation , thus hindering antigen presentation [41] . This pathway may contribute to disruption of adaptive immunity in hosts other than murine , including humans . Consistent with this possibility , among human patients with ocular toxoplasmosis in Columbia , mouse-virulent alleles of ROP18 were correlated with more severe disease [42] . Additionally , certain South American strains appear to be more virulent in humans in that they are associated with higher rates of ocular disease in the case of Clade B [43] , and they are associated with severe disease in healthy individuals in the case of Clade F [44 , 45 , 46] . The parasite factors responsible for increased acute disease severity in humans are unknown , but these traits could be suitable for genetic mapping if appropriate screens were available for assessing these phenotypes in vitro . Although it is unlikely that resistance in human cells is a driving factor for evolution of T . gondii virulence , given the fact that humans are rarely involved in transmission , such traits could represent conserved functions that are important in multiple vertebrate hosts . Using forward genetic mapping and reverse genetic engineering with CRISPR-Cas9 , we demonstrate the conserved nature of the ROP5 and ROP18 virulence factors in South American strains of T . gondii . We also found a more dramatic role for ROP18 in the virulence of these strains than had been appreciated using North American isolates of the parasite . Recent genetic studies in the house mouse reveal novel alleles of IRGs that are responsible for their enhanced resistance to otherwise virulent type 1 lineages of T . gondii [47] . The predator-prey relationship between the house mouse and domestic cat has likely existed for at least 10 , 000 years in association with human settlement [48] . In contrast , strains harboring similar ROP18 and ROP5 alleles have been estimated to have diverged several million years ago [22] . Collectively these studies suggest that parasite virulence determinants are co-evolving against the pressure of murine immunity factors in a backdrop of more ancient history that is conserved among diverse strains of T . gondii . All animal experiments were conducted according to the U . S . A . Public Health Service Policy on Humane Care and Use of Laboratory Animals . Animals were maintained in an Association for Assessment and Accreditation of Laboratory Animal Care International-approved facility . All protocols were approved by the Institutional Care Committee at the School of Medicine , Washington University in St . Louis . The parents and progeny of the genetic cross between type 2 ME49-FUDRr and type 10 VAND-SNFr strains [25] , and parasite strains TgCtBr5-SNFr and TgCtBr18-SNFr were grown as tachyzoites in human foreskin fibroblast ( HFF ) monolayers . The TgCtBr5 and TgCtBr18 strains were made SNF resistant using N-ethyl-N-nitrosourea as described previously [49] . Cultures were maintained in D10 media composed of Dulbecoc’s modified Eagle medium ( DMEM ) , 10% fetal bovine serum ( FBS ) , gentamycin , and glutamine at 37°C with 5% CO2 . Infected monolayers were scrapped and force passaged through 21 gauge needles to harvest parasites . CD-1 mice were acquired from commercial vendors ( Jackson Labs ) and housed under SPF conditions at Washington University School of Medicine . Acute virulence was determined by i . p . injection of indicated number of tachyzoites into groups of five 8–12 week old female outbred CD-1 mice per experiment . Survival was monitored for 30 days , after which surviving mice were bled to collect serum for serological testing by enzyme-linked immunosorbent assay ( ELISA ) , as described previously [50] . The percentage of surviving animals was determined by the number of animals that succumbed / total number of animals that were infected x 100 . Survival values were corrected for animals that failed to become infected ( i . e . those that remained serologically negative ) . The virulence phenotype was analyzed by QTL mapping using J/qtl [51] to establish the genome-wide association of phenotypes with genotypes of progeny from the genetic cross between type 2 ME49-FUDRr and type 10 VAND-SNFr strains described previously [25] . Virulence QTLs were confirmed by interval mapping based on a one-dimensional genome scan with 1 , 000 permutation to generate a log likelihood plot ( P < 0 . 001 ) . Covariant analysis and a two-dimensional genome scan were conducted using J/qtl to identify minor QTLs and epistatic interactions , respectively using criteria above . In order to make the double-CRISPR plasmid used to knockout the ROP5 locus , two single sgRNA CRISPR plasmids were constructed . The pSAG1::CAS9-U6::sgUPRT plasmid ( Addgene #54467 ) was used as a template to amplify with Set1-1 or Set1-2 primers ( S2 Table ) and PCR products were processed using the Q5 Site-Directed Mutagenesis Kit ( NEB ) , as described previously [30] to create the pSAG1::CAS9-U6::sg5’ROP5 and pSAG1::CAS9-U6::sg3’ROP5 plasmids . The sgRNA cassette ( containing the U6 promoter , sgRNA , and scaffold ) was PCR amplified with Q5 High-Fidelity DNA polymerase ( NEB ) from the pSAG1::CAS9-U6::sg3’ROP5 plasmid using Set2-1 primers ( S2 Table ) that contain terminal KpnI and XhoI restriction enzyme sites . The sg3’ROP5 cassette PCR product and the pSAG1::CAS9-U6::sg5’ROP5 plasmid were cut with KpnI/XhoI ( NEB ) , buffer 4 at 37°C overnight , gel purified on a 1% agarose gel , and purified with the Qiagen Gel Purification Kit ( Qiagen ) . The digested and purified PCR products and plasmid were ligated using T4 ligase ( NEB ) at 16°C over-night , ligations were transformed into OmniMAX chemically competent cells ( Invitrogen ) , and a correct clone was identified with KpnI/XhoI restriction fragment mapping . The plasmid named pSAG1::CAS9-U6::sg5’ROP5-sg3’ROP5 plasmid contains two sgRNAs targeting the 5’ and 3’ side of the ROP5 locus . To select for the excision of the ROP5 locus in pSAG1::CAS9-U6::sg5’ROP5-sg3’ROP5 transfected parasites , we constructed a selection cassette flanked by homologous regions ( HR ) adjacent to the 5’ and 3’ sgROP5 CRISPR cut sites using Gibson Assembly ( NEB ) . Four PCR fragments were amplified using Q5 High-Fidelity DNA polymerase ( NEB ) from the following template/primer pairs: ( 1 ) the p-loxP-DHFR-mCherry-loxP plasmid with Set3-1 primers , ( 2 ) VAND genomic lysate with Set 3-2primers , ( 3 ) VAND genomic lysate with Set3-3 primers , ( 4 ) the pUC19 plasmid with Set3-4 primers ( S2 Table ) . PCR products were gel purified as above and combined in a Gibson Assembly reaction ( NEB ) using manufactures protocol . The reaction was transformed into NEB 5-alpha cells and a correct clone was identified using three different restriction fragment mappings ( BamHI/NotI , BamHI/ScaI , and ScaI/NotI ) and named pUC19-ROP5HRs-loxP-DHFR-mCherry . This plasmid was used as a template in a Q5 High-Fidelity DNA polymerase PCR to amplify the 5’ROP5HR-loxP-DHFRmCherry-loxP-3’ROP5HR cassette using the Set4-1 primers ( S2 Table ) . This amplified cassette was gel purified as above for use as a selection cassette in creating ROP5 KO parasites . The VAND-SNFr , TgCtBr5-SNFr , and TgCtBr18-SNFr strains were individually electroporated with the double-CRISPR pSAG1::CAS9-U6::sg5’ROP5-sg3’ROP5 plasmid and a PCR amplified 5’ROP5HR-loxP-DHFRmCherry-loxP-3’ROP5HR selection cassette . Parasites were selected in 2 μM pyrimethamine ( Sigma-Aldrich ) , cloned , and screened by PCR for integration of the DHFR-mCherry cassette at the ROP5 locus using Set5-1 primers for 5’ integration , Set5-2 primers for 3’ integration , Set5-3 for the ROP5 coding sequence and Set5-4 primers to the GRA1 promoter as a positive control for the PCR ( S2 Table ) . Knockout parasites were identified , protein harvested , and Western blotted to confirm the loss of ROP5 protein . In order to complement with a cosmid containing the DHFR pyrimethamine selection cassette , we excised loxP-DHFR-mCherry-loxP from a VAND ROP5 knockout clone using Cre recombinase . The VANDΔrop5 #10 strain was electroporated with the pmin-Cre-eGFP plasmid [52] , cloned without selection , and mCherry negative parasites were screened by immune-fluorescence assay ( IFA ) using primary rat aντι-mCherry mAb ( Life Technologies ) and secondary goat aντι-rat Alexa Fluor 594 ( Life Technologies ) , to identify a VANDΔrop5ΔdhfrmCherry parasite . Sensitivity to 2 μM pyrimethamine was confirmed for VANDΔrop5ΔdhfrmCherry . To generate the complement , the VANDΔrop5ΔdhfrmCherry parasite was electroporated with the TOXOM52 cosmid , derived from the type I RH strain ( http://toxomap . wustl . edu/cosmid . html ) [53] . The parasites were selected with 2 μM pyrimethamine ( Sigma-Aldrich ) , and clones were screened for expression of ROP5 by IFA using primary rabbit anti-ROP5 and secondary goat anti-rabbit Alexa-Fluor 594 ( Life Technologies ) . ROP5 expressing clones were identified , protein harvested , and Western blotted to confirm wild type levels of expression in the complement , creating VANDΔrop5::ROP5 . The ROP18 KOs and complement were created as described previously [30] . Briefly , the VAND-SNFr , TgCtBr5-SNFr , and TgCtBr18-SNFr strains were individually electroporated with the pSAG1::CAS9-U6::sgROP18 CRISPR plasmid and a PCR amplified selection cassette containing ROP18 homologous regions flanking a DHFR* pyrimethamine resistance marker . Parasites were selected with 2 μM pyrimethamine ( Sigma-Aldrich ) , cloned by limiting dilution , and PCR screened for integration of the selection cassette at the ROP18 CRISPR cut site . Knockout clones were identified , proteins harvested , and Western blots were run to confirm the loss of ROP18 protein . To create a complemented strain , the VANDΔrop18 #1 clone was electroporated with the pSAG1::CAS9-U6::sgUPRT plasmid and a PCR amplified selection cassette containing UPRT homologous regions flanking an IMC1p-ROP18Ty expression construct . Parasites were selected with 10 μM FUDR ( Sigma-Aldrich ) , cloned and screened for expression of Ty by IFA: using primary α-Ty-BB2 mAb and secondary goat-α-mouse Alexa Fluor 488 ( Life Technologies ) . ROP18-Ty expressing clones were identified and protein harvested , and Western blots were run to confirm wild type expression levels of ROP18 in the complement , VANDΔrop18::ROP18 . Genomic lysates were harvested from various parasite strains for use as templates in PCR screens . Parasites were grown in HFF monolayers , harvest by scrape/needle pass , spun at 400 rcf and supernatant was aspirated from the parasite pellet . The pellet was suspended in genomic lysate buffer ( 1X PBS with 10% Taq polymerase buffer ( Invitrogen ) and 400 ng/μl proteinase K ( Life Technologies ) ) , the sample was incubated at 37°C for 1hr , 50°C for 30 min , and 95°C for 5–10 min , and spun to pellet debris . Genomic lysates were used in PCR screens to detect the integration or removal of selection cassettes . For the ROP5 knockouts and complementation the primer sets outlined in S2 Table were used , and for the ROP18 disruptants and complementation the primer sets described previously [30] were used in a Taq polymerase ( NEB ) PCR . PCR products were run on a 1% agarose gel and imaged on a UV light box . Parasites were grown and harvested as above , counted on a hemocytometer , the pellet was resuspended in protein lysate buffer ( 1X PBS , 1% Triton-X , 1% DNase , 1% protease inhibitors ( Sigma-Aldrich ) ) to 1 X 106 parasites/μl , vortexed , and spun to pellet debris . Each sample contained protein from 5 X 106 parasites in 1X sample buffer with 10% β-Me ( Sigma-Aldrich ) and was incubated for 5–10 min at 95°C . Samples were resolved on a 10% SDS-polyacrylamide ( Bio-Rad ) gel at 100V for 1 hr . Gels were transferred to nitrocellulose , blocked in 5% milk , and probed with ( all diluted to 1:10 , 000 ) primary—rabbit α-ROP5 or rabbit α-ROP18 and mouse mAb DG52 α-SAG1 antibodies , and secondary antibodies consisting of goat α-rabbit IgG IRDye 680 ( LI-COR Biosciences ) and goat α-mouse IgG IRDye 800 ( LI-COR Biosciences ) . All Blots were imaged on a LI-COR Odyessy CLx infrared imaging system ( LI-COR Biosciences ) . Network analysis using genome wide SNP data from the 16 Toxoplasma gondii reference genomes and the TgCtBr18 genome ( obtained from ToxoDB ) was conducted using SplitsTree v4 . 4 [54] to generate unrooted phylogenetic networks using a neighbor-net method and 1 , 000 bootstrap replicates , as previously described [31] . ROP5 alleles used in this study were defined in previous studies [9 , 18] and were obtained from NCBI accession records for the following strains: GT1 , VEG , and ME49 alleles ( BK008043 . 1-BK008057 . 1 ) , VAND alleles ( JQ743716 . 1 , JQ743730 . 1 , JQ743760 . 1 , JQ743772 . 1 , JQ743778 . 1 , JQ743783 . 1 ) , TgCtBr5 alleles ( JQ743734 . 1 , JQ743742 . 1 , JQ743743 . 1 ) . Briefly , ROP5 alleles for GT1 , VEG and ME49 were previously reconstructed in Behnke et al . [9] by aligning Sanger sequences of each strain to a reference ROP5 CDS allowing for the reconstruction of SNPs patterns . In this case , the Sanger sequences are long enough to combine overlapping reads containing the same SNP patterns together to define novel alleles , as detailed in Figure S6 of Behnke et al . , [9] . Alleles for VAND and TgCtBr5 were previously defined by allele-specific amplification and direct sequencing as described in Niedelman et al . [18] . Copy-number variation ( CNV ) for the ROP5 locus was determined as previously described [31] . Briefly , NextGen sequence reads of the respective genomes were aligned to the ME49 assembled chromosomes using bowtie2 –end-to-end [55] the average 1X read bases for each genome and the average read bases for TGME49_308090 , TGME49_308093 , and TGME49_308096 were determined using samtools mpileup exported data [56] . The reads surrounding the ROP5 locus for each genome were exported with samtools mpileup and data were graphed in R . To determine CNV for individual ROP5 alleles , the reads aligning to the region surrounding TGME49_308090 were exported from the bowtie2 alignment using mpileup and realigned to the ME49 assembled chromosomes using CLC Genomics Workbench . Reads that corresponded to a particular allele were identified based on SNP pattern and the allele CNV was estimating by calculating the ratio of allele reads as compared to the total CNV for the ROP5 locus [9] . DNA sequences were translated and aligned using MUSCLE with Geneious v7 . 0 . 4 . Phylogenetic analysis of the protein sequences was implemented in RAxML [57] using maximum likelihood and the JTT+I+G+F model . For maximum likelihood , we used the ModelGenerator v85 [58] to determine the most appropriate amino acid substitution model based on the Akaike Information Criterion . Positively selected codons were estimated by the Fast Unconstrained Bayesian AppRoximation method using HyPhy package [59 , 60] . Posterior probability values above 0 . 95 were chosen for codon sites where the distribution of non-synonymous substitution rates ( β ) are significantly greater than the synonymous ( α ) substitution rate . Amino acid residues at the ROP5/IRGa6 interface were identified by PDBePISA [32 , 33] based on the crystal structure of ROP5BI ( corresponding to ROP5 GT1-m1 in the present study ) bound to murine IRGa6 ( PDB code 4LV5 ) [19] . Residues close to the interface were identified by visual inspection . Figures were generated in PyMOL .	Parasites and the hosts they infect are in constant struggle with each other for survival . On the one hand , the host needs to control parasite growth , while the parasite needs to evade the host response long enough to allow for efficient transmission . The parasite Toxoplasma gondii has evolved virulence factors ROP5 and ROP18 to evade innate immune mechanisms of its natural intermediate host , small rodents . These genes were initially identified in clonal parasite types isolated from Europe and North America , but the factors that contribute to virulence in genetically divergent South American strains have not been tested . Here we used forward and reverse genetic analyses to show that ROP5 and ROP18 are also major virulence factors in genetically distinct virulent South American strains . Given that ROP5 and ROP18 function as virulence factors in strains from North America , Europe , and South America they likely acquired their functions before Toxoplasma gondii radiated into its present global population structure .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[]	2015	Rhoptry Proteins ROP5 and ROP18 Are Major Murine Virulence Factors in Genetically Divergent South American Strains of Toxoplasma gondii
The widespread use of antibiotics is selecting for a variety of resistance mechanisms that seriously challenge our ability to treat bacterial infections . Resistant bacteria can be selected at the high concentrations of antibiotics used therapeutically , but what role the much lower antibiotic concentrations present in many environments plays in selection remains largely unclear . Here we show using highly sensitive competition experiments that selection of resistant bacteria occurs at extremely low antibiotic concentrations . Thus , for three clinically important antibiotics , drug concentrations up to several hundred-fold below the minimal inhibitory concentration of susceptible bacteria could enrich for resistant bacteria , even when present at a very low initial fraction . We also show that de novo mutants can be selected at sub-MIC concentrations of antibiotics , and we provide a mathematical model predicting how rapidly such mutants would take over in a susceptible population . These results add another dimension to the evolution of resistance and suggest that the low antibiotic concentrations found in many natural environments are important for enrichment and maintenance of resistance in bacterial populations . Antibiotics represent one of mankind's most important medical inventions but during the last decades the continuing rapid development of antibiotic resistance has emerged as one of the most serious health care problems , both in community and hospital settings [1] , [2] , [3] . Whereas some resistance-conferring genes were most likely originally selected to serve metabolic functions and/or for signal trafficking or protection against competing antibiotic-producing bacteria [4] , the recent worldwide enrichment and spread of highly resistant pathogenic bacteria in the micro-biosphere has largely been driven by human activities , including the extensive use and misuse of antibiotics in human and veterinary medicine and in agriculture [2] , [3] , [5] , [6] , [7] . While it is evident that the high concentrations of antibiotics used therapeutically can select for resistant mutants , it still remains unclear how important the low antibiotic concentrations that due to anthropogenic input pollute natural ( e . g . aquatic or soil ) environments [8] , [9] , [10] , that are produced naturally by antibiotic-producing micro-organisms or that are present in certain human/animal body compartments during therapeutic or growth promotion use , are for the selection and enrichment of resistant mutants . In pharmacodynamic models it is generally assumed that selection of resistant bacteria only occurs at concentrations between the minimal inhibitory concentration ( MIC ) of the susceptible wild type population ( MICsusc ) and that of the resistant bacteria ( MICres ) [11] , [12] ( mutant selective window hypothesis , see Fig . 1A ) and that concentrations below the MICsusc will not inhibit growth of the susceptible bacteria and therefore not be selective . Earlier studies on selection with small differences in bacterial susceptibility to antibiotics show that selection can efficiently act on minute differences to select for resistance [13] , [14] , [15] . Furthermore , using an elegant color-based assay a recent study has shown qualitatively that levels of antibiotics below the MIC can enrich for resistant bacteria [16] . Here , we further explore the mutant selective window assumption and as outlined schematically in Fig . 1A , we examine for two bacterial species and three antibiotics how far below the MICsusc pre-existing and de novo generated resistant mutants can be selectively enriched because of minute reductions in the growth rate of their susceptible counterparts . To determine if exposure to very low antibiotic concentrations ( <<MICsusc ) can result in enrichment for resistant mutants , we used several well-defined mutants of Escherichia coli and Salmonella enterica ( Var . Typhimurium LT2 ) ( Table S1 in Text S1 ) and three different classes of antibiotics with high importance to human and veterinary medicine ( tetracyclines , fluoroquinolones and aminoglycosides ) . The resistance markers used were Tn10dTet ( confers tetracycline resistance ) , gyrA ( S83L and D87N ) , ΔmarR , and ΔacrR mutations ( confer ciprofloxacin resistance ) and rpsL ( K42R ) ( confers streptomycin resistance ) , all of which are found in clinical isolates of several different bacterial species . Using highly sensitive competition experiments between isogenic pairs of susceptible and resistant strains , we show that selection of resistant bacteria can occur at antibiotic concentrations far below the minimal inhibitory concentration . Finally , we present a mathematical model , showing how resistant mutants are expected to arise de novo and spread in bacterial populations at these sub-MIC levels of antibiotics . An initial examination of the effect of low antibiotic concentrations was performed in single cultures where a susceptible wild-type and a resistant mutant carrying a Tn10dTet were grown separately in the presence of different concentrations of tetracycline . As shown in Fig . 1B and Table S2 in Text S1 , concentrations far below MICsusc reduced the exponential growth rate of the susceptible strain without any apparent effect on the resistant strain . For example , at a concentration 1/30 of the MICsusc , the susceptible strain grew about 15% slower than without antibiotic whereas the resistant mutant seemed unaffected , suggesting that resistant strains are strongly selected at these low concentrations . To increase the sensitivity of these assays and allow detection of extremely small differences in growth rates , we performed competition experiments between pairs of susceptible and resistant strains . The MICs for the susceptible and resistant mutants were: S . typhimurium wild type ( streptomycin = 4 ug/ml , tetracycline = 1 . 5 ug/ml ) , rpsL K42R >1024 ug/ml and Tn10dTet strain = 128 ug/ml; E . coli wild type ( ciprofloxacin = 0 . 023 ug/ml ) , gyrA S83L ( ciprofloxacin = 0 . 38 ug/ml ) , gyrA D87N ( ciprofloxacin = 0 . 25 ug/ml ) , ΔacrR ( ciprofloxacin = 0 . 047 ug/ml ) and ΔmarR ( ciprofloxacin = 0 . 047 ug/ml ) . The strains were genetically tagged with variants of the green fluorescent protein gene ( yfp and cfp , encoding yellow- and cyan-fluorescent proteins , respectively ) to allow counting of large populations of competing cells by fluorescence activated cell sorting ( FACS ) , thereby significantly reducing any experimental errors associated with counting of small populations . The competing strains were isogenic except for the resistance determinant and the yfp and cfp genes producing the respective fluorescent proteins . Control experiments showed that the difference in fitness cost between the cfp and yfp markers had a negligible impact on growth rates ( Fig . S1 ) . The strains were competed for up to 80 generations by serial passage in batch cultures in the presence of different concentrations of either one of the antibiotics tetracycline , ciprofloxacin ( a fluoroquinolone ) and streptomycin ( an aminoglycoside ) as well as in the absence of drug ( Fig . 2A–D , Fig . 3A–H , Fig . 4A–E ) . As shown by our previous studies [17] , this experimental set-up allows detection of growth rate differences at least as small as 0 . 3% , which approaches the limit of sensitivity set by the interference of periodic selection events . Whereas the growth rate measurement shown in Fig . 1B only measured the exponential phase of growth , the competition experiments represent a composite of growth and survival in lag phase , exponential phase and stationary phase that allows examination of the whole growth cycle . The data presented in Fig . 2A , C , Fig . 3A , C , E , G and Fig . 4A–D shows how the ratio of resistant:susceptible strains changes as a function of the number of generations of growth at different concentrations of antibiotic . Each line represents one competition experiment and the slope is a measure of the selection coefficient ( s-value ) . Thus , the negative slope obtained in the absence of antibiotic is a measure of the fitness cost of the antibiotic resistance mechanism and a positive slope indicates that the resistant mutant is enriched . When the s-values obtained from these experiments are plotted as a function of antibiotic concentration the intercept , s = 0 , represents what we specify as the minimal selective concentration ( MSC ) where the fitness cost of the resistance is balanced by the antibiotic-conferred selection for the resistant mutant ( Fig . 2B , D , Fig . 3B , D , F , H and Fig . 4E ) . Depending on the antibiotic and the type of resistance mutation examined the MSC varied between 1/4 and 1/230 of the MICsusc . For streptomycin the MSC value was 1/4 of the MIC value of the susceptible strain ( Fig . 2B ) , for tetracycline 1/100 ( Fig . 2D ) and for ciprofloxacin it varied between 1/10 ( Fig . 3B ) and 1/230 ( Fig . 3D ) of the MICsusc depending on the particular resistance mutation . These values correspond to absolute antibiotic concentrations of 1 µg/ml ( streptomycin ) , 15 ng/ml ( tetracycline ) , and 2 . 5 ng/ml to 100 pg/ml ( ciprofloxacin ) . The competitions performed with a small initial fraction of resistant mutants also showed that the selection coefficients are independent of the initial frequency of resistant mutants . Even at initial frequencies as low as 10−4 , the same enrichment ( i . e . same selection coefficient ) of the resistant mutants could be observed as at a 1∶1 ratio ( compare Figures 2D and 4E ) . Since the resistant mutants could be enriched from very low initial fractions ( 10−4 ) we also tested whether resistant mutants could be selected de novo from a susceptible population . To this end we grew 20 independent lineages of a susceptible wild type Salmonella typhimurium LT2 strain for 700 generations at 1/4 of the MIC of streptomycin and continuously screened for resistant cells by plating on different concentrations of streptomycin . At this low level of antibiotic we could observe rapid enrichment of de novo resistant mutants ( Fig . 5 ) . Thus , within 200 to 400 generations , a considerable enrichment of mutants with resistances between 2 and 16 times the MIC of the starting strain ( 8–64 µg/ml ) could be seen , and after 500 to 600 generations also high-level resistant mutants ( 24–32 times MIC of the wild type = 96–128 µg/ml ) appeared . After 400 generations , all 20 lineages contained subpopulations with a MIC higher than 32 µg/ml ( 8 times MIC ) , and after 600 generations 14 of the lineages had subpopulations with a MIC higher than 64 µg/ml ( 16 times MIC ) . Using the method described above , 20 lineages of wild type E . coli were grown for 600 generations in sub-MIC levels of ciprofloxacin . After 500 generations of growth at 1/10 of the MIC , five of the lineages had subpopulations ( >1% of the population ) with low level resistance ( 2-fold higher MIC than the susceptible parental strain ) to ciprofloxacin , and after 600 generations , one out of twenty lineages had a subpopulation of cells with an MIC 8-fold higher than the susceptible parental strain ( see Fig . S2 ) . We also calculated ( Appendix in Text S1 ) how rapidly de novo generated resistant mutants would take over in a susceptible population at low antibiotic concentration , as determined by mutation rates ( u ) , population sizes ( N ) , and the fitness advantage ( s ) in the presence of antibiotics . s depends on the antibiotic concentration above the MSC as shown in Fig . 2 and 3 . When no resistance mutants are present initially , the time to fixation can be expressed as The first term is the stochastic waiting time for the first surviving mutant to appear and the second term is from the subsequent growth to 50% presence . For small values of uN <0 . 1 , the first term dominates and fixation may be slow . For large values , uN >1 , the second term dominates and fixation can be fast , ca . 100–1000 generations for s between 0 . 1–0 . 01 ( Fig . 6 ) . In this limit , resistance mutants appear so frequently that it makes little difference to the fixation time if they are present initially or not . In this context it is worth noting that sub-MIC levels of several antibiotics , most pronounced for fluoroquinolones , have been shown to increase bacterial mutation rates which potentially could reduce the waiting time and thereby increase the rate of mutant take-over [18] . Antibiotic concentrations in natural environments can vary extensively depending on the particular environment . For example , in connection with polluting pharmaceutical industries or at sewage outlets from hospitals the concentrations can reach very high levels ( mg/ml ) , with fluoroquinolones frequently reaching the highest levels [19] , [20] , [21] whereas in aquatic environments or in soil levels are typically much lower [8] . The presented data suggests that even in those environments with very low antibiotic concentrations , maintenance and selection of resistant bacteria can occur . For example , the MSC for ciprofloxacin and tetracycline obtained from our experiments correspond to 100 pg/ml and 15 ng/ml , respectively , similar to concentrations that can be found in some aquatic and soil environments [8] . Thus , the surprisingly high frequencies of antibiotic-resistant bacteria found in animals from relatively pristine environments [22] , [23] , [24] could conceivably be partly explained by enrichment due to sub-MIC selective effects . These findings are also highly relevant with regard to the question of reversibility of resistance . Since most antibiotic resistance mechanisms are associated with a fitness cost it has been proposed that the fitness costs of resistance will allow susceptible bacteria to out-compete resistant bacteria if the antibiotic selective pressure is reduced . However , most available data suggests that the rate of reversibility will be slow or absent at the community level [25] . Several factors could contribute to this irreversibility , including the absence of a fitness cost , reduction of the fitness cost through compensating mutations and genetic co-selection between the resistance-conferring gene and another gene under selection . In addition , the sub-MIC selection observed here could be a significant contributor to this long-term persistence of resistance where very low antibiotic concentrations in the environment are sufficient to maintain the existing resistant bacteria in the population by further balancing the fitness cost of the resistance . This can be particularly important for bacterial pathogens whose normal life cycle involves growth in soil environments ( e . g . P . aeruginosa ) or periodic growth in aquatic environments ( e . g . E . coli ) . From the slope of the graphs in Fig . 2B , D and Fig . 3B , D , F and H we can infer that the fitness cost of the resistance mutation has a major influence on the value of the MSC . This cost must first be overcome by a negative effect of antibiotics on the susceptible bacteria before resistant bacteria will be selected , shifting the MSC towards higher concentrations . Reducing this cost will shift the curve upwards and lower the MSC . It is also evident that the increased resistance of the mutants ( difference between MICsusc . and MICres . ) or the mode of resistance ( point mutation or efflux pumps ) has little effect on MSC relative to the fitness cost . Since our data was obtained in defined genetic backgrounds with single point mutations or deletions commonly found for the antibiotics tested , the fitness cost represents the cost of a de novo resistance mutation . However , in most resistant strains found clinically the fitness cost of resistance is frequently compensated for by secondary mutations without a loss of resistance [25] . Such fitness compensation has been described for resistance to many different antibiotics including fluoroquinolones and streptomycin [26] , [27] , [28] . This implies that the antibiotic concentrations at which such compensated resistant strains will be selected can be even lower than what we have measured here . Another significant implication from our and the findings of others is that the widely used concept of the mutant selective window needs modification . Thus , in pharmacodynamics it is generally assumed that antibiotic concentrations below the MIC do not confer selection and that the mutant selective window—the concentration range in which the resistant mutant is enriched—extends between the MIC of the susceptible wild type and the MIC of the resistant mutants [11] , [12] . However , our results imply that the biologically relevant sub-MIC selective window is much wider and needs to include antibiotic concentrations several hundred-fold below MICsusc ( Fig . 1A ) . Furthermore , the methodology described here could be used to probe the biologically active antibiotic concentrations in different environments , including for example animal models . Thus , by performing competitions between genetically tagged susceptible and resistant strains in animals treated with different antibiotic concentrations one can from the enrichment rate of resistant bacteria infer the biologically active concentration of antibiotic at the site of bacterial growth . At selection above the MIC of a strain , the main driving force of the selection is antibiotic resistance , while the fitness cost of the mutation is less critical . Even mutations with a very high cost will be selected , since competitors in the form of susceptible bacteria will be eliminated . At sub-MIC levels , however , the situation is different since the susceptible bacteria will not die , they will only grow slower . Because of this , resistance mutations conferring high fitness costs will not be enriched; only mutations where the fitness cost is lower than the growth reduction caused by the antibiotic in the susceptible bacteria will be competitive . This suggests that a new spectrum of low-cost or no-cost resistance mutations might be enriched during such conditions . The data in Fig . 5 , show that these sub-MIC levels of antibiotics do not only enrich for pre-existing resistant mutants , but they can also select for resistant mutants de novo from a susceptible population . It is interesting to note that despite the low antibiotic concentrations used , mutants with high resistance levels were enriched . Since the streptomycin concentration chosen for the de novo mutant selection experiment is the same as the MSC determined in the competitions between wild type and an rpsL K42R mutant , the enriched resistant bacteria are likely to carry resistance mutations with a fitness cost that is significantly lower than the previously studied rpsL mutation . In the presented experiments pre-existing mutants were rapidly enriched in competitions with susceptible strains . From the mathematical model we can infer a similar situation for de novo resistant mutants , especially in large populations where uN >1 and at antibiotic concentrations where 0 . 01< s <1 . 0 . In those situations resistant mutants rapidly appear and within 100–1000 generations of growth they will take over the population . The model is supported by the experiments shown in Fig . 5 and Fig . S2 , where de novo mutants continuously increased in frequency during 600–700 generations of growth in the presence of sub-MIC levels of antibiotics . In conclusion , the presented data suggests that the very low antibiotic levels which are present in many natural environments or generated in certain body compartments during treatment are relevant for the enrichment and maintenance of pre-existing resistant mutants as well as for the de novo selection of new mutants . These results emphasize the importance of introducing measures that reduce antibiotic levels in the environment and use of treatment dosing regimens that preclude prolonged time periods of sub-MIC levels of antibiotics . Strains used in this study were derived from Escherichia coli MG1655 and Salmonella enterica serovar Typhimurium LT2 ( designated S . typhimurium in the text ) and are listed in Table S1 in Text S1 . The resistant strains were constructed by P22 transduction ( S typhimurium ) or P1 transduction ( E coli ) of the resistance genes into the parental strains . The liquid and solid media used for bacterial growth were Mueller–Hinton broth ( Becton Dickinson , MD , USA ) , Mueller–Hinton agar ( Mueller–Hinton broth supplemented with 1 . 5% agar ) and Luria–Bertani ( LB ) agar ( Sigma-Aldrich , MO , USA ) . Strains were grown at 37°C , and liquid cultures were aerated by shaking . Growth rates were measured at 37°C in Mueller-Hinton broth , with or without tetracycline present , using a Bioscreen C Analyzer ( Oy Growth Curves Ab Ltd , Helsinki , Finland ) . Each well was inoculated with a 1000-fold dilution of an overnight culture and measurements at each antibiotic concentration were made in quadruplicate . The cultures were grown for 24 hours with continuous shaking , and OD600 measurements were taken every 4 min . The calculations were based on OD600 values between 0 . 02 and 0 . 1 , where growth was observed to be exponential . The sensitive strain ( DA6192 ) and the resistant strain ( DA17822 ) were grown in separate experiments , and the relative growth rates were calculated as the derived growth rates divided by the growth rate of the same strain grown without antibiotics . MIC assays of tetracycline and ciprofloxacin were performed by broth macrodilution in 10 mL tubes . Tubes containing Mueller-Hinton broth ( 1 mL ) supplemented with different concentrations of antibiotics were inoculated with 1 µL of an overnight bacterial culture grown at 37°C . The tubes were incubated at 37°C with shaking for 16 to 18 hours , the tetracycline cultures protected from light to avoid degradation of the antibiotic . The MIC was set to the lowest concentration of antibiotic yielding no visible growth . The MIC of streptomycin was determined by Etest according to the instructions of the manufacturer ( AB bioMerieux , Solna , Sweden ) . Etests were performed on Mueller-Hinton agar plates incubated for 16–18 h at 37°C . Limited sampling of competitors ( <103 cells ) commonly introduces statistical uncertainties in competition experiments and more accurate measurements of resistant mutant to wild type cell ratios can be obtained with the aid of chromosomal copies of either the cyan ( cfp ) or yellow ( yfp ) variants of green fluorescent protein gene ( gfp ) . These allow tracking of large numbers of single cells ( 105 cells ) using a fluorescence activated cell sorter ( FACS ) . The cfp/yfp genes were inserted into galK using the λ Red system as previously described [29] and moved by phage P22 transduction or phage P1 transduction into the various strains . Overnight cultures grown in Mueller-Hinton medium of the susceptible wild type strains with either cfp or yfp , were mixed 1∶1 , 10∶1 , 102∶1 , 103∶1 and 104∶1 with the isogenic resistant mutant carrying the other marker and maintained by 1000-fold serial dilution ( resulting in 10 generations of growth per serial passage ) every 24 hours for up to 4 to 6 serial passages . The ratio of resistant to susceptible cells in the population was determined at each serial passage by counting 105 cells using a fluorescence-activated cell sorter ( BD FacsAria ) . The selection coefficients were determined using the regression model s = [ln ( R ( t ) /R ( 0 ) ) ]/[t] , as previously described [30] where R is the ratio of resistant to susceptible . This protocol allowed reproducible determinations of fitness differences as small as s = 0 . 003 [17] . Two independently constructed sets of each wild type strain , marked with either cfp or yfp , were also included to measure the relative impact on growth rates of having a cfp marker compared to yfp . These control experiments showed that over 40 generations of competition , the difference in cost between the markers had a negligible impact on growth rates . ( Fig . S1 ) . The competition experiments performed with a low initial fraction of resistant mutants were done with tetracycline due to the long time required for the appearance of de novo tetracycline resistant mutants that might disturb the competition experiments . To investigate whether sub-inhibitory antibiotic concentration could also select for de novo generated resistant mutants , susceptible bacteria was serially passaged at 1/4 of the MIC of streptomycin and at 1/10 of the MIC of ciprofloxacin . A total of 20 independent lineages of S . typhimurium LT2 was serially passaged by 1000-fold dilution in 1 ml batch cultures every 24 hours for 700 generations ( 10 generations of growth per serial passage ) in Mueller-Hinton medium containing 1 µg/ml streptomycin , and 20 independent lineages of E . coli MG1655 were serially passaged by 1000-fold dilution in 1 ml batch cultures every 24 hours for 600 generations in Mueller-Hinton medium containing 2 . 3 ng/ml ciprofloxacin . The lineages were started from overnight cultures from independent colonies , using an initial bottleneck of approximately 104 cells to minimize the number of preexisting resistant mutants . The percentage of resistant cells in each culture was monitored by plating approximately 105 cells onto LB agar containing different concentrations of antibiotics every 100 generations and counting the number of colonies . A subset of these cells were restreaked on the same antibiotic concentration to confirm that they were resistant . Tn10 ( Transposon Tn10 tetracycline resistance and repressor genes tetA and tetR ) GenBank: J01830 . 1 rpsL ( 30S ribosomal protein S12 ) GenBank: AAL22311 . 1 Swiss-Prot: P0A7S6 gyrA ( DNA gyrase subunit A ) GenBank: AAC75291 . 1 Swiss-Prot: P0AES4 acrR ( HTH-type transcriptional regulator AcrR ) GenBank: AAC73566 . 1 Swiss-Prot: P0ACS9 marR ( Multiple antibiotic resistance protein MarR ) GenBank: AAC74603 . 2 Swiss-Prot: P27245 gfp ( Green fluorescent protein ) GenBank: AAA27722 . 1 Swiss-Prot: P42212	Antibiotic resistance has emerged as a very significant health care problem due to the extensive use and misuse of antibiotics in human and veterinary medicine and in agriculture . It remains unclear where most of the resistant bacteria have been selected , and in particular if the low antibiotic concentrations that are present in natural environments or in human/animal body compartments during therapeutic or growth promotion use , are important for the selection and enrichment of resistant mutants . The presented data shows that for several clinically used antibiotics extremely low concentrations , similar to the concentrations found in natural environments , can select for resistant bacteria . These results suggest that antibiotic release into the environment might be a significant contributor to the emergence and maintenance of resistance and emphasize the importance of introducing measures to reduce antibiotic pollution .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "biology", "microbiology", "evolutionary", "biology", "population", "biology" ]	2011	Selection of Resistant Bacteria at Very Low Antibiotic Concentrations
The steroid hormone ecdysone coordinates insect growth and development , directing the major postembryonic transition of forms , metamorphosis . The steroid-deficient ecdysoneless1 ( ecd1 ) strain of Drosophila melanogaster has long served to assess the impact of ecdysone on gene regulation , morphogenesis , or reproduction . However , ecd also exerts cell-autonomous effects independently of the hormone , and mammalian Ecd homologs have been implicated in cell cycle regulation and cancer . Why the Drosophila ecd1 mutants lack ecdysone has not been resolved . Here , we show that in Drosophila cells , Ecd directly interacts with core components of the U5 snRNP spliceosomal complex , including the conserved Prp8 protein . In accord with a function in pre-mRNA splicing , Ecd and Prp8 are cell-autonomously required for survival of proliferating cells within the larval imaginal discs . In the steroidogenic prothoracic gland , loss of Ecd or Prp8 prevents splicing of a large intron from CYP307A2/spookier ( spok ) pre-mRNA , thus eliminating this essential ecdysone-biosynthetic enzyme and blocking the entry to metamorphosis . Human Ecd ( hEcd ) can substitute for its missing fly ortholog . When expressed in the Ecd-deficient prothoracic gland , hEcd re-establishes spok pre-mRNA splicing and protein expression , restoring ecdysone synthesis and normal development . Our work identifies Ecd as a novel pre-mRNA splicing factor whose function has been conserved in its human counterpart . Whether the role of mammalian Ecd in cancer involves pre-mRNA splicing remains to be discovered . The insect steroid hormones , ecdysteroids , regulate growth , stimulate molting , and orchestrate tissues to undergo complex morphogenetic changes during metamorphosis [1]–[3] . Temporal control of ecdysteroid synthesis by the larval prothoracic gland ( PG ) is therefore critical ( for recent reviews see [4] , [5] ) . The biosynthetic pathway commences by converting cholesterol to 7-dehydrocholesterol by a Rieske oxygenase Neverland ( Nvd ) [6] . A short-chain dehydrogenase/reductase encoded by shroud ( sro ) [7] and cytochrome P450 ( CYP ) enzymes encoded by spook ( spo ) /spookier ( spok ) , phantom ( phm ) , disembodied ( dib ) , and shadow ( sad ) then catalyze the subsequent steps to produce ecdysone ( E ) [8] , [9] . Once released from the PG , E becomes hydroxylated in peripheral tissues by another CYP , Shade ( Shd ) , to yield the main active hormone , 20-hydroxyecdysone ( 20E ) [10] . Reflecting the necessity of 20E for early cuticle formation , Drosophila melanogaster loss-of-function mutants that are available for sro [7] , spo , phm , dib , sad , and shd [8] die as embryos . Understandably , Drosophila mutants that display reduced steroid titers during larval development provide invaluable experimental tools . Among these , the ecdysoneless1 ( ecd1 ) mutants are homozygous viable at 22°C , but exposure to 29°C reduces their ecdysteroid titer and causes a developmental arrest [11] , [12] . The ecd1 allele has been widely used since its discovery in 1977 [11] to test effects of ecdysteroid signaling on a number of processes from morphogenesis to reproduction to behavior . Yet why these mutants lack the hormone has not been determined . Our original identification of the ecd gene [13] has revealed homology from fission yeast to humans but none that would illuminate the mode of Ecd action . Mammalian Ecd ( also known as SGT1 and hEcd in humans ) has been shown to stimulate cell proliferation by interacting with the Retinoblastoma ( Rb ) proteins [14] . Conditional deletion of the Ecd gene from mouse embryonic fibroblasts stalls these cells at the G1-S phase transition , suggesting that Ecd normally lifts the inhibitory effect of Rb on E2F-dependent cell cycle progression [14] . High hEcd expression has been correlated with malignancy of human breast [15] and pancreatic [16] tumors . In a mouse model , Ecd has been suggested to promote tumorigenesis via enhancing glucose import and glycolysis in the pancreatic tumor cells [16] . These observations illustrate the emerging importance of Ecd , but an underlying mechanism for Ecd action is still lacking . A systematic mapping of Drosophila protein-protein interactions [17] has uncovered contacts between Ecd and proteins responsible for pre-mRNA splicing . Among these are members of the U5 small nuclear ribonucleoprotein particle ( snRNP ) complex , including orthologs of the RNA helicase Brr2 , the GTPase Snu114 , the Aar2 protein , and the highly conserved spliceosome core component , Prp8 [18] . Another global proteomic study [19] has detected a corresponding interaction between human Prp8 and hEcd . The spliceosome is a most elaborate machinery and pre-mRNA splicing is extensively coupled with transcription as well as with posttranscriptional mRNA surveillance and degradation of incorrectly spliced transcripts [20] . This network involves hundreds of proteins whose individual functions are often inferred from work on yeast , or still remain unknown [21]–[23] . Genetic studies in Drosophila have shown essential roles of several spliceosome components such as Prp8 , Prp38 , Prp31 , or BCAS2 for tissue growth , cell proliferation or cell viability [24]–[27] . Proteomic dissection of spliceosomal complexes has revealed a large overlap in protein composition between Drosophila and humans while suggesting novel and/or fly-specific components [28] . With the hypothesis that Ecd might be a new player in pre-mRNA splicing , we have verified that Ecd interacts with a complex containing Drosophila Prp8 , Aar2 , Brr2 , and Snu114 orthologs . Consistently with a vital role in pre-mRNA splicing , we show that Ecd is required by dividing imaginal disc cells for survival , even when their apoptosis is blocked or cell growth enhanced . In the larval PG , loss of Ecd compromises pre-mRNA splicing and abolishes protein expression of the essential steroidogenic enzyme Spok ( CYP307A2 ) , thus accounting for the systemic steroid deficiency in ecd mutants . Remarkably , human Ecd can functionally substitute for its fly ortholog . The Drosophila Ecd protein has been recently associated with the U5 snRNP complex [17] . In order to verify the proteome-wide data , we expressed Myc epitope-tagged Ecd in Drosophila S2 cells and using mass spectrometry we examined material that co-precipitated with Myc::Ecd ( Figure 1A ) . This analysis identified three proteins of the U5 snRNP complex that were previously shown to associate with Ecd [17] , namely orthologs of the budding yeast ( Saccharomyces cerevisiae ) Prp8p , Snu114p , and Brr2p that are encoded by D . melanogaster genes CG8877/prp8 , CG4849/eftud2 , and CG5931/l ( 3 ) 72Ab , respectively . Our mass spectrometry did not detect another Ecd interactor , identified as a product of the CG12320 gene [17] that is homologous to S . cerevisiae Aar2p . For clarity , we will refer to the Drosophila proteins as Prp8 , Snu114 , Brr2 , and Aar2 . To test the individual protein-protein interactions , we cloned the above Drosophila genes and performed co-immunoprecipitation with pairs of their epitope-tagged products expressed in S2 cells . First , we verified interactions between the fly counterparts of the known spliceosomal components . As expected , binding occurred between Prp8 and Snu114 , Prp8 and Brr2 , and Prp8 and Aar2 ( Figure 1B ) . Of these four proteins , Prp8 and Aar2 formed complexes with Ecd , whereas binding between Ecd and either Brr2 or Snu114 was not detected ( Figure 1C ) . It was of interest to examine interactions of mutated Ecd versions that occur in Drosophila . We chose the temperature-sensitive ecd1 allele that carries a single Pro-656 to serine substitution and the non-conditional lethal , ecdl ( 3 ) 23 , where the Ecd polypeptide is prematurely terminated after Ser-649 , thus lacking the C-terminal 35 amino acids [13] . Both the Ecd1 ( tested at a non-permissive temperature of 30°C ) and Ecdl ( 3 ) 23 mutant proteins retained the capacity to bind Prp8 and Aar2 ( Figure 1C ) . This indicates that the steroid-deficiency and lethality phenotypes in Drosophila result from a failure of the mutant Ecd protein to perform functions other than binding Prp8 and Aar2 . The above results suggest that Ecd interacts with at least two well-established members of the U5 snRNP complex , of which Prp8 is evolutionarily the best-conserved core protein of the spliceosome [18] . In S . cerevisiae , a part of the U5 snRNP complex that contains Prp8p , Snu114p , and Aar2p assembles in the cytoplasm before being imported to the nucleus , where Brr2p replaces Aar2p [29] , [30] . Localization of Drosophila Prp8 , Snu114 , Aar2 or Brr2 has not been previously reported . Using the epitope-tagged proteins in transfected S2 cells , we detected Prp8 , Snu114 , and Aar2 predominantly in the cytoplasm , whereas Brr2 occurred in the nucleus ( Figure 2A ) . Expression of Flag::Prp8 in transgenic Drosophila larvae revealed that Prp8 also resided in the cytoplasm of the PG ( Figure S1A ) . The Ecd protein was previously detected in the cytoplasm of Drosophila PG cells [13] . We confirmed this result with a new antibody raised against the N-terminal part of Ecd for the endogenous protein ( Figure S1B , S1C ) and for Ecd overexpressed in the PG using the phm-Gal4 driver ( Figure S1D ) . This antibody also detected a clear cytoplasmic signal in the wing imaginal discs ( Figure S2 ) . The cytoplasmic localization of hEcd in human cells was shown to depend on leptomycin B-sensitive nuclear export [31] . In transfected S2 cells , GFP-tagged Ecd co-localized in the cytoplasm with Prp8 ( Figure 2A ) . Only when the cells were treated with leptomycin B , some of the GFP::Ecd fusion protein was retained in the nucleus ( Figure 2B ) . Therefore , together with the spliceosomal proteins with which Ecd interacts , Ecd primarily resides in the cytoplasm . Like its human counterpart , Ecd is subject to active nuclear export . If Ecd plays a role in pre-mRNA splicing as suggested by its protein interactions , its loss would be expected to disturb vital cellular functions as has been found for several Drosophila spliceosomal proteins including Prp8 [24]–[27] . To track the fate of Ecd-deficient imaginal cells in vivo , we employed the MARCM technique [32] using the non-conditional ecdl ( 3 ) 23 allele that behaves as a genetic null [13] . In contrast to control , no mitotic clones homozygous for this mutation were found in eye/antennal or wing imaginal discs of third-instar larvae ( Figure 3A , 3B , 3E , 3F ) . Rare and extremely small ecdl ( 3 ) 23 clones could be detected when cell death was prevented in these clones with the anti-apoptotic protein p35 , or when their cellular growth was enhanced with activated RasV12 ( Figure 3C , 3D ) . To achieve a milder loss-of-function effect , we generated transgenic flies for RNAi-mediated silencing of ecd . Observation of wing discs two days after heat-dependent clonal induction of ecd dsRNA revealed frequent ecdRNAi clones . These gradually disappeared over the next 24 hours ( Figure 3G , 3H ) , indicating that the cells autonomously required Ecd to survive . To see the effect of Ecd knockdown on morphogenesis , we induced ecd RNAi in a restricted region along the anterior-posterior boundary of the wing imaginal disc using the dpp-Gal4 driver . Figure S2 shows depletion of the Ecd protein from the dpp-expressing cells . The loss of Ecd disrupted the regular pattern of the dpp expression domain , and the affected cells underwent apoptosis as assessed by active Caspase 3 staining ( Figure 4A , 4B ) . Consequently , the adult wings of dpp>ecdRNAi flies displayed reduced size of specific intervein regions and loss of the anterior crossvein ( Figure 4E , 4F ) . The same anomalies were induced by RNAi against the components of the U5 snRNP complex , Prp8 ( Figure 4C , 4G ) and Brr2 ( Figure S3A , S3B ) . Overexpression of Prp8 in the Ecd-deficient cells did not suppress the ecd RNAi phenotype ( Figure S3C , S3D ) , suggesting that Ecd plays a unique role that cannot be substituted by surplus of its partner protein . Strikingly , apoptosis of imaginal disc cells expressing ecd dsRNA as well as the adult wing defect could be averted by supplementing the human ortholog , hEcd ( Figure 4D , 4H ) . To address the putative role of Ecd in ecdysone biosynthesis , we removed the Ecd protein specifically from the PG by triggering ecd RNAi with the phm-Gal4 driver [33] ( Figure S1C ) . Affected larvae reached the third instar but were unable to pupate . Unlike in the imaginal discs , depletion of Ecd was not cell-lethal in the polyploid PG , although it reduced the size of PG cells and nuclei ( Figure 5A , 5B ) . Using available antibodies , we could therefore examine expression of steroidogenic enzymes , namely CYP307A2/Spok and CYP306A1/Phm , in late-third instar phm>ecdRNAi larvae . The strong Spok staining in control PGs was completely lost in the Ecd-deficient gland , whereas the Phm signal appeared only partly reduced ( Figure 5A , 5B ) . While both enzymes are indispensable for E biosynthesis , Spok acts upstream of Phm in the larval PG to mediate the essential conversion of 7-dehydrocholesterol to ketodiol [33] , [34] . Thus , PGs deprived of Spok cannot synthesize ecdysone even in the presence of Phm . As was the case in the wing imaginal discs , expressing the human Ecd ortholog in the PG compensated for the depletion of the endogenous Ecd protein . hEcd improved morphology of the gland ( Figure 5C ) and restored expression of Spok ( Figure 5C′ ) . Similarly to ecd RNAi , knockdown of the spliceosomal proteins Prp8 ( Figure 5D ) and Brr2 ( Figure S4A , S4B ) also reduced size of the PG and eliminated the Spok protein while only partially reducing Phm levels . These results suggested that expression of Spok , but not of Phm , was highly sensitive to perturbation of pre-mRNA splicing . Interestingly , the phm gene contains two small introns , whereas spok lies within the heterochromatin and its single intron separates the two coding exons by nearly 30 kilobases ( Flybase , http://flybase . org/ ) ( Figure 6A ) . To examine whether the absence of Ecd affects spok pre-mRNA splicing , we determined the pre-mRNA:mRNA ratio , a standard measure of intron retention or splicing defect ( e . g . , [35]–[37] ) , in phm>ecdRNAi larvae and in larvae homozygous for the temperature-sensitive ecd1 allele . mRNA levels were measured using quantitative reverse-transcription ( qRT ) PCR with primers positioned in two separate exons , and pre-mRNA was amplified with primer pairs spanning exon-intron boundaries or primers within the intron ( Figure 6A ) . All RNA samples were free of residual genomic DNA and qRT-PCR data were normalized to rp49 mRNA levels that did not change appreciably between control and ecd-deficient larvae ( Figure S5 ) . Because spok and phm are expressed specifically in the PG during larval development , we were able to use RNA from entire larvae . As assessed with all three pre-mRNA-specific primer sets , spok pre-mRNA:mRNA ratio strongly increased in both phm>ecdRNAi larvae and in ecd1 mutants upshifted to 29°C ( Figure 6B ) . The relative enrichment of unspliced to spliced spok transcript was markedly stronger upon PG-specific knockdown of ecd than in the hypomorph mutants . Unexpectedly , loss of ecd also reduced levels of spok pre-mRNA , again more substantially in phm>ecdRNAi larvae ( Figure 6C ) , suggesting that in addition to splicing , either spok transcription or stability of its pre-mRNA might have been affected . Consistently with the absence of the Spok protein ( Figure 5B′ ) , ecd RNAi diminished spok mRNA ( Figure 6D ) . Levels of phm pre-mRNA and mRNA were also lowered in PGs of ecd-deficient animals relative to controls ( Figure 6C , 6D ) . However , in contrast to spok there was no significant increase of the pre-mRNA:mRNA ratio and therefore no appreciable intron retention in the case of phm ( Figure 6B ) . These data suggested that Ecd was required for splicing of the large intron from the spok pre-mRNA . For comparison , we examined the effect of Prp8 knockdown . As expected , spok mRNA disappeared in phm>prp8RNAi larvae due to compromised splicing ( Figure 6E ) . However , because the absolute amount of unspliced spok transcript accumulated upon prp8 RNAi , the pre-mRNA:mRNA ratio rose dramatically more than in the case of ecd RNAi ( Figure 6E ) . Finally , we tested whether human Ecd restored Spok expression at the level of pre-mRNA splicing . In a separate set of experiments , we confirmed that the loss of spok mRNA was accompanied by intron retention ( increased pre-mRNA:mRNA ratio , Figure 6F ) . Importantly , expression of hEcd in the PG of phm>ecdRNAi larvae restored levels of spok mRNA , pre-mRNA , and of their normal proportion ( Figure 6F ) , suggesting that the human protein could substitute for Ecd in splicing of spok pre-mRNA . Because the PG produces a circulating hormone , knockdown of Ecd would be expected to affect the entire organism even when restricted to the gland . In agreement with the absence of the Spok/CYP307A2 protein ( Figure 5B′ ) , third-instar phm>ecdRNAi larvae on day 6 after egg laying ( AEL ) showed reduced ecdysteroid titer ( Figure S6A ) . Compared to controls , they were retarded in growth ( Figure 7A–C ) . Without interruption , they continued to feed for another 10 or more days , producing large “permanent” larvae ( Figure 7D ) . This phenotype corresponded with effects of lacking ecdysteroid surge and inadequate expression of steroidogenic genes including spok and phm [38] , [39] . Although feeding phm>ecdRNAi larvae on day 5 AEL with 20E was not sufficient for pupation , it induced wandering behavior ( Figure S6B ) . Importantly , substituting hEcd for the depleted fly protein in the PG increased ecdysteroid titer ( Figure S6A ) and restored normal growth and development of phm>ecdRNAi animals to adults ( Figure 7A , 7E , 7F ) . The PG-specific knockdown of Ecd was accompanied by lowered mRNA levels of at least two 20E-response genes , the Ecdysone receptor ( EcR ) and E74 , as assessed in whole phm>ecdRNAi larvae ( Figure 7G ) . A similar reduction in EcR and E74 mRNAs occurred in ecd1 mutant larvae at 29°C ( Figure 7G ) . However , ecd1 mutants could have lost these transcripts for two different , mutually non-exclusive reasons: ( 1 ) due to the lack of circulating ecdysteroids , and/or ( 2 ) because Ecd may be required for splicing of the EcR and E74 pre-mRNAs , both of which contain multiple large introns . Indeed , we found a major rise in the pre-mRNA:mRNA ratio for EcR ( but not for E74; data not shown ) in ecd1 mutants , and a much less pronounced increase of this ratio in phm>ecdRNAi larvae ( Figure 7H ) . These data suggest that EcR pre-mRNA splicing was sensitive to loss of ecd function in tissues throughout the body . In support of this notion , we observed that dietary 20E did not induce EcR mRNA expression in ecd1 mutants at 29°C nearly as efficiently as under the permissive temperature or in phm>ecdRNAi larvae ( Figure S6C ) , where ecd function in the peripheral tissues was unaffected . A proteome-wide study [17] has uncovered interactions of Drosophila Ecd with multiple U5 snRNP-associated spliceosomal proteins . We likewise found a complex of Ecd with Prp8 , Snu114 , and Brr2 using mass spectrometry , and interactions of Ecd with Prp8 and Aar2 by immunoprecipitation . We detected Brr2 in cell nuclei , whereas Ecd co-localized with Prp8 , Snu114 , and Aar2 in the cytoplasm of Drosophila cells . In contrast , S . cerevisiae Prp8p resides in the yeast nuclei owing to its nuclear localization signal [29] . Although Prp8 is remarkably well conserved [18] , this particular sequence shows poor homology with the fly protein . Nuclear import of Drosophila Prp8 might therefore rely on another mechanism . Nonetheless , our results conform to a current model from S . cerevisiae , where Prp8p , Snu114p , and Aar2p preassemble in the cytoplasm , then upon nuclear import Aar2p is replaced by Brr2p as the U5 snRNP complex matures [29] , [30] . Nucleo-cytoplasmic shuttling of Ecd , which also occurs in human cells [31] , corresponds with such translocation . Phenotypes caused by disrupted pre-mRNA splicing match those inflicted by loss of Ecd . Cell death of Drosophila imaginal disc clones lacking spliceosomal proteins Prp38 , MFAP1 , and BCAS2 has been reported [24] , [25] . Our data show that knockdown of the Prp8 and Brr2 proteins caused apoptosis of imaginal disc cells and recapitulated a specific ecd RNAi wing phenotype . Removal of Prp8 or Brr2 from the PG eliminated the Spok protein , confirming that its expression was sensitive to deficiency in pre-mRNA splicing factors . Similar to Prp8 knockdown , spok pre-mRNA:mRNA ratio strongly increased in ecd1 or phm>ecdRNAi larvae , thus evidencing a role of Ecd in spok pre-mRNA splicing . However , removal of Ecd also affected transcription as judged from reduced levels of spok or phm pre-mRNAs . It is important to note here that pre-mRNA splicing is not an isolated event but that it is intimately coupled with transcriptional elongation [44] , [45] as well as with 3′-end processing , quality control , and nuclear export or degradation of the RNA product ( reviewed in [20] ) . Indeed , there are indications that hEcd may be involved in transcriptional regulation in mammalian cells [31] . An intriguing question is why would dysfunction of Ecd impact splicing of spok but not phm pre-mRNA in the steroid-producing gland ? Surprisingly , even the relatively few and simple intron-containing genes in S . cerevisiae have been shown to respond differently to mutations in individual spliceosome core components [36] . In multicellular organisms , spliceosomal proteins such as Prp8 are not ubiquitously expressed and not equally required to splice every pre-mRNA [46] . Instead , many perform exquisitely tissue-specific , gene-specific , and even exon-specific functions . An RNAi screen in Drosophila cells has uncovered the necessity of several core spliceosomal proteins including Brr2 for inclusion of particular exons in different , alternatively spliced transcripts [47] . A striking example of how remarkably specific physiological functions may be affected by faulty pre-mRNA splicing is the human disease retinitis pigmentosa ( RP ) , where degeneration of photoreceptor cells leads to blindness . Autosomal-dominant forms of RP are linked to mutations in Prp8 , Prp31 , Prp3 , and PAP1 homologs that contribute to the U5•U4/U6 tri-snRNP complex [46] , [48] . One plausible explanation for the sensitivity of the photoreceptor neurons to defective splicing is insufficient production of mRNAs that are highly expressed in these cells [48] . It has been argued that ubiquitously expressed pre-mRNAs carry universal splicing signals in order to be correctly spliced in any cell type , whereas transcripts with restricted expression patterns might rely on less robust , tissue-specific splicing signals that are more prone to fail in the absence of individual splicing factors [46] . Although the mechanism of Ecd action remains to be determined , we presume that spok pre-mRNA might be particularly sensitive to the absence of Ecd , Prp8 or Brr2 for similar reasons . In addition to abolishing the expression of Spok , loss of ecd function partially reduced mRNA and pre-mRNA levels of phm ( Figure 6C , 6D ) and of two other genes ( nvd and dib; data not shown ) that are likewise required for E biosynthesis . However , the lower expression of phm , nvd and dib was not accompanied by increased pre-mRNA:mRNA ratios , indicating that pre-mRNA splicing of these genes did not depend on Ecd . These results might suggest that Ecd exerts some effect on transcription or transcript stability . Alternatively , Ecd might be required for pre-mRNA splicing of a factor acting upstream of the steroidogenic gene expression . One such candidate is EcR , which has recently been shown to mediate a positive-feedback of 20E on the expression of phm , dib , sro and sad , but not of spok [49] . Elevated pre-mRNA:mRNA ratio indicated that splicing of EcR pre-mRNA was compromised in ecd1 mutants ( Figure 7H ) , and the EcR protein was markedly reduced in the PGs of phm>ecdRNAi larvae ( Figure S6D ) . Therefore , whereas the absence of the Spok protein primarily resulted from disrupted splicing , the partial reduction of nvd , phm and dib expression might be attributable to the lack of EcR in the PG of phm>ecdRNAi larvae . A failure of the PG to produce E inevitably provokes systemic developmental defects in peripheral organs . However , an issue arises with the ecd1 mutants as to whether any observed tissue-specific phenotype may be ascribed to the lacking hormone . Because Ecd is required cell-autonomously , effects of ecdysteroid deficiency and those caused by loss of any cell-autonomous Ecd function cannot be discriminated in ecd1 mutant background . For example , low expression of 20E-response genes such as EcR in ecd1 larvae likely results from a combined impact of disrupted E synthesis in the PG and compromised EcR pre-mRNA splicing in the peripheral tissues . The problem of 20E-dependent and 20E-independent effects of the ecd1 mutation may be reflected by a recent transcriptome analysis , revealing that of about a thousand genes affected in ecd1 background , only a minority were regulated by the 20E receptor , EcR [43] . Based on the elimination of ecdl ( 3 ) 23 mutant clones from the imaginal discs , we conclude that cells lacking Ecd cannot be sustained within proliferating tissue context . This was the case even when the mutant clones were protected from apoptosis with p35 , or enhanced for growth through RasV12 . Small ecdl ( 3 ) 23 or ecdRNAi clones were replaced by surrounding cells without phenotypic consequences for the adult . In contrast , ecd RNAi delivered to larger areas was lethal or caused visible defects , such as the aberrant wings in dpp>ecdRNAi flies . Rarely emerging MS1096>ecdRNAi flies had vestigial wings ( Figure S7A ) similar to those induced by depletion of the BCAS2 spliceosomal protein under the same Gal4 driver [25] . Interestingly , imaginal disc cells were not affected by triggering ecd RNAi after their proliferative phase was completed , as in the eyes of GMR>ecdRNAi flies ( Figure S7B ) . GMR-Gal4 is a strong but late-acting driver expressed predominantly in the post-mitotic cells posterior to the eye morphogenetic furrow [50] . Intriguingly , deletion of Ecd in mouse embryonic fibroblasts has been shown to cause a proliferative block and to reduce expression of several cell-cycle regulators such as CyclinE1 ( CycE1 ) downstream of the transcription factor E2F [14] . However , we were unable to advance proliferation of ecdRNAi clones in imaginal discs by expressing Drosophila CycE [51] or combinations of either E2F with its partner DP [52] or CycD with Cdk4 [53] ( data not shown ) . Therefore , the lack of Ecd in proliferating tissues cannot be compensated by gain of individual factors that stimulate cellular growth or cell cycle progression . These results suggest that proliferating cells are particularly sensitive to loss of Ecd . Such sensitivity corresponds to the fact that splicing factors prevailed among genes positively screened as being required for cell division in a human cell line [54] . Unlike some of the core spliceosomal factors , no Ecd-like protein has been found in the budding yeast , S . cerevisiae , in which only 5% of the genes contain introns . This contrasts with 43% of intron-containing genes in the fission yeast , Schizosaccharomyces pombe [55] , where the spliceosomal protein composition is more akin to humans than to the budding yeast [56] . Interestingly , S . pombe has an Ecd ortholog , Sgt1p , which resides in the cell nuclei and is essential for the yeast growth [57] . Conditional Sgt1 mutation alters expression of genes involved in virtually all cellular processes including metabolism , and impairs growth on glucose media . Sgt1p is thought to regulate transcription [57] , although its mode of action remains unknown . Expression of Sgt1p failed to substitute for Drosophila Ecd ( data not shown ) , likely reflecting a remote homology between the two proteins ( 21% overall amino acid identity ) . In contrast , human Ecd that is moderately homologous to fly Ecd ( 31% ) rectified pre-mRNA splicing and expression of spok in the PG , permitting development of phm>ecdRNAi adults . In mammals , Ecd has been implicated as a positive regulator of cell-cycle promoting genes , of cell cycle progression itself , and of cancer development [14]–[16] . Since hEcd has been independently detected in a complex containing human Prp8 [19] , it will be of interest to know if any of those effects involve splicing of particular pre-mRNAs . Considering the evidence from S . pombe , Drosophila and mammals , we suspect that the role of Ecd in pre-mRNA splicing may correlate with the evolutionarily growing importance of splicing in complex multicellular organisms . The following Drosophila strains were used: w1118 , ecd1 [11] , ecdl ( 3 ) 23 [13]; a gift of Dr I . Zhimulev ) , FRT2A ( BL1997 ) , ecdl ( 3 ) 23 FRT2A/TM6B , y w hsFLP; ecdl ( 3 ) 23 FRT2A/TM6B , UAS-rasV12/CyO; ecdl ( 3 ) 23 FRT2A/TM6B , UAS-p35/CyO; ecdl ( 3 ) 23 FRT2A/TM6B , UAS-prp8RNAi ( VDRC , 18565 ) , UAS-brr2RNAi ( VDRC , 110666 ) , phm-Gal4 , UAS-CD8::GFP [33] , and dpp-Gal4 , UAS-GFP [58] . New transgenic lines carrying UAS-ecd , UAS-ecdRNAi , UAS-Flag::hEcd , UAS-Flag::prp8 , and UAS-Flag::Sgt1 constructs were established with standard P-element germline transformation . If not specified otherwise , crosses were carried out at 25°C . Crossing all Gal4 driver lines to w1118 background provided controls for each experiment . Mutant clones within eye/antennal imaginal discs were generated using the MARCM ( mosaic analysis with a repressible cell marker ) method [32] with eyFLP , act>y+>Gal4 , UAS-GFP/CyO; FRT2A tubGal80/TM6B flies as described [59] . To induce ecdRNAi “flip-out” clones , hsFLP; act>y+>Gal4 , UAS-GFP/CyO [60] females were crossed to UAS-ecdRNAi males at 20°C . Recombination was induced by exposing progeny to heat shock at 37°C for 30 min , followed by incubation for 48 and 72 h at 25°C prior to dissection . Temperature-sensitive ecd1 and control w1118 larvae were grown at 22°C until day 4 AEL , then placed to 37°C for 45 min and to 29°C until dissection on day 6 AEL . Coding Drosophila melanogaster DNA sequences of prp8 ( CG8877 ) , snu114 ( CG4849 ) , aar2 ( CG12320 ) , brr2 ( CG5931 ) , ecd ( CG5714 ) , ecdl ( 3 ) 23 ( amino acids 1–650 of Ecd ) , and human Ecd ( hsgt1 , NP_009196 . 1 ) were amplified from respective cDNAs using the Phusion polymerase ( New England Biolabs ) . Schizosaccharomyces pombe Sgt1 ( SPAC1002 . 10c ) was amplified from genomic DNA . See Table S1 for all PCR primers . The fragments were cloned into pTFW , pTMW or pTGW vectors enabling expression of proteins with N-terminal Flag , Myc , or GFP tags , respectively ( T . Murphy , Carnegie Institution of Washington ) , using the Gateway cloning system ( Invitrogen ) . The ecd1 P656S mutant was recreated using site-directed mutagenesis by CCT to TCT codon transition . To generate UAS-ecdRNAi , a 497-bp cDNA fragment from the ecd gene was amplified and cloned as inverted repeat into the pWIZ vector [61] . A polyclonal rat anti-EcdNterm antibody was raised ( Eurogentec ) against bacterially expressed Drosophila Ecd polypeptide ( amino acids 204–458 ) . The corresponding ecd cDNA fragment was cloned into the pET28b plasmid ( Novagen ) , and the antigen was purified using a hexahistidine tag . Drosophila S2 cells were cultured at 25°C in Shields and Sang M3 insect medium ( Sigma Aldrich ) containing 8% fetal bovine serum and antibiotics ( Pen/Strep , Gibco ) . Cells were transfected in serum free medium using X-TremeGENE ( Roche Applied Science ) according to manufacturer's instructions . Expression of UAS-driven genes was induced by co-transfection with a pWA-GAL4 plasmid expressing Gal4 under an actin5C promoter ( a gift from Y . Hiromi ) . Nuclear export was inhibited by incubating cells with 5 ng/ml leptomycin B ( Biomol ) for 4 h . Cells expressing the temperature-sensitive Ecd1 protein variant were upshifted to non-permissive temperature of 30°C for 45 min prior to processing . For immunoprecipitation , transfected S2 cells were lysed in 50 mM Tris-HCl ( pH 7 . 8 ) , 150 mM NaCl , 1 mM EDTA ( pH 8 . 0 ) , 1% Triton X-100 , 0 . 01% Igepal , and protease inhibitors ( Roche Applied Science ) . The lysate ( 300–500 µg of total protein ) was incubated overnight with 15 µl of anti-Flag ( Invitrogen ) or anti-Myc ( Medical and Biological Laboratories ) magnetic beads at 4°C . After five washes in lysis buffer , proteins were recovered in two consecutive elution steps , each with 50 µl of 0 . 1 M glycine-HCl ( pH 3 . 0 ) for 5 min , and neutralized with 10 µl of 0 . 5 M Tris-HCl ( pH 7 . 8 ) and 1 . 5 M NaCl . Upon SDS-PAGE , proteins were detected by immunoblotting with mouse anti-Flag M2 ( 1∶1000 , Sigma Aldrich ) , rabbit anti-c-Myc ( 1∶1000 , sc-789 , Santa Cruz ) or rabbit anti-GFP ( 1∶2000 , Acris ) antibodies , followed by incubation with corresponding HRP-conjugated secondary antibodies . Chemiluminescent signal was captured using ImageQuant LAS4000 reader ( GE Healthcare ) . Protein extracts from S2 cells containing Myc::Ecd or the empty pTMW vector ( for control ) were resolved on SDS-PAGE . Upon silver staining ( SilverQuestTM , Invitrogen ) , bands of interest were cut out . In-gel tryptic digestion with 12 . 5 ng/µl porcine trypsin ( Promega ) in 10 mM NH4HCO3 and further extractions were performed as described [62] . Collected extracts were concentrated by vacuum centrifugation and desalted using STAGE Tip C18 spin columns ( Proxeon , Thermo Scientific ) [63] . Eluted peptides were vacuum-concentrated and resuspended to a final volume of 20 µl in 0 . 5% acetic acid , of which 10 µl were used for analysis . Reversed-phase liquid chromatography ( LC ) coupled to nano-flow electrospray tandem mass spectrometry ( MS ) were carried out using an EASY nLC II nano-LC ( Proxeon , Thermo Scientific ) with a C18 column ( internal diameter 75 µm ) coupled to a LTQ Orbitrap mass spectrometer ( Thermo Scientific ) . Peptide separation was performed at a flow rate of 250 nl/min over 60 min ( 10–40% acetonitrile; buffer A: 0 . 1% formic acid in water; buffer B: 0 . 1% formic acid in acetonitrile ) . Survey full scan MS spectra ( m/z 350 to 2000 ) of intact peptides were acquired in the Orbitrap at a resolution of 30 , 000 using m/z 445 . 12003 as a lock mass . The mass spectrometer acquired spectra in data dependent mode and automatically switched between MS and MS/MS acquisition . Signals with unknown charge state and +1 were excluded from fragmentation . The ten most intense peaks ( threshold 500 ) were isolated and automatically fragmented in the linear ion trap using collision induced dissociation ( CID ) . The search algorithm Mascot [64] , implemented in the Proteinscape software ( Bruker ) , was used for peptide and protein identification . MS/MS data were searched using the canonical and isoform sequence database of the Drosophila melanogaster complete proteome , provided by the UniProt Consortium . Oxidation of methionine residues was used as a variable modification and carbamidomethylation of cysteine residues as a fixed modification . For Orbitrap data , 10 ppm mass tolerance was allowed for intact peptide masses and 0 . 8 Da for CID fragment ions detected in the linear ion trap . Peptides were filtered for Mascot score ≤20 . Protein identifications were based on at least two peptides . Drosophila S2 cells grown on cover slips , dissected ring glands and imaginal discs were processed as described [65] and stained overnight at 4°C ( tissues ) or 2 h at room temperature ( S2 cells ) with the following antibodies: rat anti-EcdNterm ( 1∶500 , this study ) , guinea pig anti-Spok ( 1∶1000 ) and rabbit anti-Phm ( 1∶300 ) [33] , rabbit anti-cleaved Caspase-3 ( 1∶500 , ASP175 , Cell Signaling , #9661 ) , rabbit anti-pH3 ( 1∶100 , Cell Signaling , #9701 ) , mouse anti-Flag M2 ( 1∶500 , Sigma Aldrich ) , rabbit anti-c-Myc ( 1∶500 , sc-789 , Santa Cruz ) , and mouse anti-Lamin ( 1∶500 , ADL67 . 10 ) , mouse anti-EcR ( 1∶200 , DDA2 . 7 ) , and mouse anti-Fasciclin III ( 1∶300 ) ; the latter three from the Developmental Studies Hybridoma Bank ( DSHB , Iowa ) . After washing , samples were incubated with corresponding secondary antibodies coupled to Cy3 or Cy5 ( Jackson ImmunoResearch ) , counterstained with DAPI ( 0 . 5 µg/ml , Invitrogen ) to visualize nuclei , and mounted in Dabco:Mowiol ( Sigma-Aldrich ) . Confocal single sections and stacks were acquired at room temperature with Olympus FV1000 confocal microscope . Maximum projections were generated using Fluoview 2 . 1c Software ( Olympus ) and ImageJ [66] . Final image processing including panel assembly , brightness and contrast adjustment were done in Photoshop CS5 . 1 ( Adobe Systems , Inc . ) . To allow comparison among genotypes , images were taken and processed with the same settings . Adult wings were mounted in Hoyer's medium . Images of larvae , flies , adult wings , and Z-stacks of adult eyes were taken using Leica M165 FC fluorescent stereomicroscope equipped with a DFC490 CCD camera . Images were processed using a Multifocus module of the LAS 3 . 7 . 0 software ( Leica ) . Total RNA was isolated from eight third-instar larvae with Isol-RNA Lysis Reagent ( 5 Prime ) . cDNA was synthesized from 2 µg of RNA treated with DNase I ( Ambion ) using random primers and Superscript III reverse transcriptase ( Invitrogen , Carlsbad , CA ) . PCR was performed in triplicates with the SYBR green mix ( Bio-Rad , Hercules , CA ) using the CFX96 ( Bio-Rad , Hercules , CA ) or the 7900HT ( Applied Biosystems ) real-time PCR systems . qRT-PCR primers ( Table S1 ) were designed to anneal at 62°C . All primers were initially tested by qPCR on serially diluted DNA templates and those deviating from typical standard curves were redesigned . Data were normalized to rp49 transcript levels and fold changes were calculated using the ΔΔCT method [67] or the Relative standard curve method [68] . At least four biological replicates were analyzed per each experiment . For hormone feeding , 40 larvae aged 5 days AEL were transferred per vial on food containing 100 µg/ml of 20E . Control food contained solvent only ( 6% w/w ethanol ) . Total RNA was isolated after 4 h ( for ecd1 ) or one day of feeding ( 6 days AEL for phm-Gal4 experiments ) ; wandering behavior was photographed on day 7 AEL . To measure ecdysteroid titer , larvae 6 days AEL ( 15 animals per assay ) were homogenized in 500 µl of methanol and centrifuged at 20 , 000 g . The pellets were re-extracted with 300 µl methanol and supernatants pooled and vacuum dried . The dried extracts were processed using the EIA immunoassay system ( Cayman Chemical ) as described [69] . Ecdysteroid concentration was calculated from an eight-point standard curve of serially diluted 20E . An unpaired two-tailed Student's t-test with unequal variation and one-way analysis of variance ( ANOVA ) with a post hoc Newman-Keuls Multiple Comparison Test were used to determine statistical significance for changes in gene expression in all qRT-PCR experiments , larval size , and ecdysteroid measurements .	Steroid hormones perform pivotal roles in animal development , sexual maturation , reproduction , and physiology . Also insects possess a hormonal steroid , commonly known as ecdysone , that was originally found to promote ecdyses in growing larvae and their metamorphosis to adults . Since the discovery of ecdysone-inducible puffs on the polytene chromosomes in the 1960's , genetics of the Drosophila fruit flies has substantially advanced our understanding of steroid hormone impact on gene regulation during development . In the present study , we have solved an old puzzle of the “ecdysoneless” mutant that has traditionally provided investigators with a steroid-deficient animal model . Unexpectedly , we find that the Ecdysoneless protein ( Ecd ) does not primarily regulate ecdysone biosynthesis but that a critical steroidogenic enzyme requires Ecd for splicing of its precursor mRNA ( pre-mRNA ) . Ecd physically contacts the immensely complex pre-mRNA splicing machinery . Outside the ecdysone-producing gland , Ecd is essential for survival of dividing cells within developing tissues . Despite vast evolutionary distance , a human homolog of Ecd can functionally substitute for its counterpart in the fly . A conserved role of Ecd in pre-mRNA splicing might underlie a recently described involvement of mammalian Ecd in cell cycle progression and its contribution to malignancy of certain tumor types .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "biotechnology", "endocrine", "signaling", "animal", "genetics", "gene", "function", "genetics", "of", "disease", "developmental", "biology", "organism", "development", "molecular", "development", "genetic", "engineering", "gene", "expression", "nuclear", "receptor", "signaling", "biochemistry", "rna", "signal", "transduction", "rna", "processing", "cell", "biology", "genetics", "biology", "and", "life", "sciences", "metamorphosis", "molecular", "cell", "biology", "cell", "signaling" ]	2014	Unexpected Role of the Steroid-Deficiency Protein Ecdysoneless in Pre-mRNA Splicing
Behaving in the real world requires flexibly combining and maintaining information about both continuous and discrete variables . In the visual domain , several lines of evidence show that neurons in some cortical networks can simultaneously represent information about the position and identity of objects , and maintain this combined representation when the object is no longer present . The underlying network mechanism for this combined representation is , however , unknown . In this paper , we approach this issue through a theoretical analysis of recurrent networks . We present a model of a cortical network that can retrieve information about the identity of objects from incomplete transient cues , while simultaneously representing their spatial position . Our results show that two factors are important in making this possible: A ) a metric organisation of the recurrent connections , and B ) a spatially localised change in the linear gain of neurons . Metric connectivity enables a localised retrieval of information about object identity , while gain modulation ensures localisation in the correct position . Importantly , we find that the amount of information that the network can retrieve and retain about identity is strongly affected by the amount of information it maintains about position . This balance can be controlled by global signals that change the neuronal gain . These results show that anatomical and physiological properties , which have long been known to characterise cortical networks , naturally endow them with the ability to maintain a conjunctive representation of the identity and location of objects . In what follows , we show an example of a retrieval bump in a 2D recurrent network with metrically organised connectivity . We consider a two dimensional network containing N = 4900 neurons in total . The neurons are arranged on a regular lattice with 70 neurons on each side and distance l between neighbouring sites . The connections between neurons have a metric structure: a neuron in position ri is connected to a neuron in position rj with probability ( 13 ) In the simulations reported here the width of the connectivity , σ , is set to 7 . 5l . Since l is the distance between two adjacent neurons , this means that the probability that two adjacent neurons are connected to each other is ∼0 . 7 . Experimental data estimate this probability to be 0 . 5–0 . 8 [30] . The gain of all neurons in the simulations reported in this section is set to a background level g = 0 . 5 . At the beginning of the simulation a 15×15 square centred on the neuron in position ( 58 , 58 ) is chosen . The activity of neurons inside this square are initialised to their activity in the first stored pattern while the activity of other neurons are set to zero , that is in the beginning of simulation if ri is in the square and νi = 0 if ri is outside it . In this way at the beginning of the simulation the dot product overlap with the first pattern and the others have the following values Fig . 1A shows the local overlap with the cued pattern ( μ = 1 ) at the beginning of a simulation . The local overlap ( Eq . ( 8 ) ) with the cued pattern after 200 synchronous updates is shown in Fig . 1B and the distribution of activity {νi} , also after 200 time steps , is shown in Fig . 1C . We see that the activity of the network is concentrated on a part of the 2D network , and so is the local overlap . The important point is that this final pattern of activity has a high dot product overlap with the cued pattern but not with other stored patterns , i . e . Thus by calculating these dot products , or equivalently calculating the sum of the local overlaps miμ over i , in the end of the simulation we can say which pattern was presented , i . e . in this example the first pattern . The spatial distribution of activity would have been different ( Fig . 1D ) , if instead of the probability distribution in Eq . ( 13 ) , we had used a uniform distribution In this case , too , by cueing one of the patterns , as we did for the metrically organised network , after 200 time steps , we have m1 ( t = 200 ) ≈0 . 8 and mμ≠1 ( t = 200 ) ≈0 , thus indicating retrieval of the pattern . The difference between the two connectivity models emerges , however , in the final distribution of activity . Whereas in Fig . 1D the activity is distributed uniformly across the network ( at a gross spatial scale , since at a fine scale individual units are activated in relation to their selectivity for the cued pattern ) , in Fig . 1C the activity is localised over a portion of the 2D network . So , metric recurrent connections , as predicted by the mathematical analyses of attractor states and as confirmed by many other simulations , allow activity to stabilise in spatially modulated distributions . Even though Fig . 1 shows the possibility of localised retrieval in the network with the Gaussian connectivity in Eq . ( 13 ) , a critical observation is that in Fig . 1B the final local overlap is in a different position than the initial cue ( Fig . 1A ) . The trajectory that the peak of the local overlap follows during the retrieval process is shown in Fig . 2 . The green square shows the peak at the beginning of the simulation , before any updates take place ( Fig . 1A ) , and the red circle shows the peak after 200 time steps ( Fig . 1B ) . It is clear that , during retrieval , the “bump” of activity drifts away from its initial position . This raises the question addressed in this paper , of whether where information in the cue can be preserved by spatially modulated attractor states . Can we code the position of an object by the position of the peak of the retrieval bump ? The answer to this question depends on whether the retrieval process can end with the peak of the bump on the intended position . We first examine whether the position of the cue ( which can be thought of as the position of an object in the retina ) determines the positions of the retrieval bump . If the retrieval bump appears at the same position as ( or is uniquely determined by ) the centre of the cue , it is possible to read the activity of the network and simply decode both what information , that is , which cue has been presented ( the pattern with the highest overlap with final activity ) , and , in addition , where it has been presented: object position is simply coded by the position of the centre of the bump . To examine the relation between the position of the initial cue and the final position of the retrieval bump , we ran simulations in which the position of the initial cue was systematically changed across the network and the distance between the position of the retrieval bump and the position of the initial cue was measured . In Fig . 3 , we summarise the results from simulating a network of 70×70 neurons with the Gaussian connectivity pattern Eq . ( 13 ) with σ = 7 . 5l , as used in Fig . 1 . At the beginning of each trial , the first pattern was cued by initialising the activity of neurons in the following way: , if neuron i was within a 15×15 square , whose centre was varied , across trials , over the entire network; while νi = 0 , if neuron i was outside the square . The activity of all neurons was then synchronously updated for 200 time steps and the local overlap with each pattern was monitored . Fig . 3A shows that the position of the bump at the end of each trial ( red circles ) does not match the peak of the local overlap with pattern 1 at the beginning of the trial ( green squares ) . The bump drifts away from its initial positions , and stabilises on one of , in this particular case , 4 final positions . This small number of final stable positions indicates that one cannot decode from the final position of the retrieval bump where the cue was located , at least not with high accuracy . In fact , by looking at the final position of the bump , one might say whether the initial position of the cue was among the 23 initial positions that converge to the upper left red circle or among the 10 initial positions that converge to the lower right red circle , but nothing more . The small number of final stable positions of the bump resembles what has been noticed before in recurrent networks with distance dependent weights between neurons but without stored memory patterns . In such models the synaptic weight between two neurons is generally taken to be excitatory at short distances while inhibitory at long distances [51]–[58] . The distance between two neurons in these models can be anatomical distance , or distance , in the feature space , between the features that the neurons are selective for . Models of this type have been used to conceptualise how local networks of orientation selective neurons in visual cortex [52] , head-direction neurons [53] , location selective neurons in prefrontal cortex [54] and hippocampal neurons [57] , [58] can maintain selectivity after the external input has been removed . Studies on rate based models [51]–[53] as well as networks with spiking neurons [54]–[56] show that , under very mild conditions , the stable activity profile of these networks is of the form of a localised “bump” . If the network is strictly homogeneous , the bump can potentially exist anywhere on the network , and it can be smoothly moved from one position to the other . Any small inhomogeneity in the network , however , fractures the continuum of solutions , which therefore represents an ideal limit case , and stable bumps are allowed only at a number of discrete positions [53] , [57] , [59]–[61] . Coming back to the associative memory network with metric connectivity , it is clear that inhomogeneity is an unavoidable part of its overlaid memory structure . Synaptic weights are required to be different from each other in such a network , to support the retrieval of memory patterns , a situation where a neuron can be active while its nearest neighbour is inactive . As a result , a retrieval bump in our model cannot be maintained at any arbitrary position on the network . Even though the final position of the bump cannot accurately tell where the cue was initiated , it may still be able to code for a large number of positions in a network with realistic size . This happens if the number of final stable positions increases with the size of the network . To examine this relation , we scaled up the simulated network . The result of such scaling analysis is reported in Fig . 4 , which shows the number of final positions resulting with different network sizes , while keeping the number of connections and the width of connectivity constant . One sees a roughly linear increase in the number of stable bump positions . The approximately linear scaling of the number of final positions with network size indicates that a large number of positions can be represented in realistically large networks , but not any arbitrary position: with our regular 2D lattices and our parameters , the number of stable bump positions is about one thousand times smaller than the number of lattice nodes . Furthermore , the few stable positions of the retrieval bump are different for different patterns ( data not shown ) . This makes the representation of position dependent on object identity and thus hard to decode . We ask , therefore , whether it is indeed possible to stabilise bumps at any arbitrary position . This is discussed in the following sections . In this section we show that the bump of activity can be stabilised on an arbitrary position on the network if neurons around that position have a slightly higher linear gain than the rest of the neurons . This increase in the linear gain applies to all neurons in that area in the same manner , whether they are selective for the cued pattern or not; that is , it is not pattern selective and solely reflects object position . This local gain modulation can be triggered by an attentional mechanism that modulates the responsiveness of neurons in the part of the network which corresponds to the position of the object . It could also be produced by the pattern itself: when the cue to initialise retrieval is given to the network , the mean activity of the part of the network that receives the cue would be higher than the rest of network . This could trigger changes in the gain of the neurons that may last for several seconds [62]–[64] . We leave discussing the sources of the gain modulation to the Discussion section and first answer the following questions . Can such localised gain modulation stabilise the bump at any desired position and , if so , how strong should it be ? How does localised gain modulation affect pattern retrieval ? Suppose that a non-pattern-selective signal changes the gain of those neurons which correspond to the position of the object in the visual scene . The effect of such gain modulation is shown in Fig . 5 . In the simulations of Fig . 5 , the activity of ≈4 . 6% of the neurons , randomly distributed across the network , are initially set to their activity in the first pattern , while the rest are silent ( note that the quality of the cue is then the same as what we used in the simulations of Fig . 3 , but now the cue is not localised ) . The localised gain modulation is incorporated into the simulations by first choosing , at each trial , a square box at a different position over the network . The linear gain of neurons inside the square is then increased by a factor of β relative to that of the other neurons in the network . The position of the centre of the high gain square box is in fact chosen in exactly the same way as we chose the centre of the cue in Fig . 3 , i . e . at the nodes of a regular lattice , shown as green squares in Fig . 5A and Fig . 5C . The result of such change in the spatial distribution of the gain is evident for β = 1 . 5 ( Fig . 5A , 5B ) and even more for β = 3 ( Fig . 5C , 5D ) . Even though the pattern-selective cue does not contain spatial information , a spatially selective increase in the linear gain of the neurons in a restricted region of the network helps localising the bump in that region . Notably , as shown in Fig . 5D , the distance that the peak in the local overlap drifts from the initial focus is minimal , particularly for successful trials ( red circles ) ( d ) , whereas averaging across unsuccessful runs ( black circles ) ( d* ) substantially increases the drift , as if jumping to the wrong basin of attraction in the space of patterns facilitates similar jumps in physical space . It should be noted that while in Fig . 3 the localised cue had been removed after initialising the activity , in the results shown in Fig . 5 the change in gain is maintained throughout the simulation . It is true that keeping the localised cue would have helped localising the bump at the right position , without gain modulation , but the essential difference between the two mechanisms should be appreciated: the change in gain is independent of the memory pattern to be retrieved and could thus be produced by a mere spatial signal , with little specific information content besides spatial position itself . The pattern-selective cue , instead , can be thought to commit the informational resources ( e . g . , the channel capacity [65] ) of the ventral visual form processing stream , and it makes sense to hypothesise that it should be removed as soon as possible , to make room for the analysis of other objects by the same pathway . Even though increasing the gain in a spatially restricted part of the network stabilised the final bump there , there is a disadvantage with this strategy: by using such non-uniform gain , the number of successful runs decreases . Remember that the quality of the cue used in Fig . 5 is the same as the one in Fig . 3 , however , there were no unsuccessful runs in Fig . 3 and Fig . 5A , whereas there are 12 unsuccessful runs in Fig . 5C ( shown by black circles ) : better preservation of spatial information ( higher gain modulation ) is accompanied by , in this example , a higher number of unsuccessful runs . This suggests that preservation of spatial information through gain modulation affects the retrieval of the pattern . In Fig . 5 the effect is negative , an interference , but as we show below it can also be a positive effect . In the following sections , we quantify this interaction using information theory and demonstrate efficient ways to minimise the negative interaction . In order to quantify the interaction between what and where information , we use Shannon information theory . We estimate the amount of information that the activity of the network , after retrieval , represents about what and where . We do this for different degrees of gain modulation , levels of the average gain , number of stored patterns and also different ways of presenting the cue . This provides us with a quantitative picture of the relation between what and where information . We denote by Iwhat and Iwhere , the amount of information about what and where , respectively . To compute Iwhat , we look at network activity after 200 times steps and compute its overlap with all stored pattens ( Eq . ( 9 ) ) . The pattern with the highest overlap is considered as retrieved and Iwhat measures how much information knowing this retrieved pattern gives us about which pattern was presented . Iwhere , on the other hand , is the information between the position of the bump of activity after retrieval and the centre of the gain modulated area ( we put Iwhere = 0 when there is no gain modulation; see section “Continuous attractors are fragmented by superimposed memories” ) . For exact definitions and details about how we compute Iwhat and Iwhere from the simulations see section “Mutual information measures” in the Materials and Methods . To start with , we consider a network ( with the architecture used before ) that has stored p patterns and assume that in the beginning of the simulations a cue similar to one of the patterns is presented ( the exact cue presentation is described in the three Conditions below ) . All neurons have a background gain of g . During recall , either the gain of all neurons is kept equal to g , which is the case of uniform gain , or the localised gain modulation mechanism is turned on . In the latter case the gain of the neurons inside a 15×15 square whose centre is on one of 49 preassigned positions on the network is boosted to βg . Different values of β are chosen in different simulations . In each run , one of the patterns is chosen as a cue and one of the 49 positions is chosen as the centre of the high gain region . As in the previous sections , the centre of the squares surrounding the high gain region is chosen from one of the 49 nodes of a 7×7 regular lattice covering the entire 2D network . Each pattern and each of the 49 positions for the high gain region is used exactly once . We first calculate Iwhat and Iwhere for a network with the global gain chosen to be g = 0 . 5 . We do this for the case of uniform gain ( all neurons have the same gain , thus equal to the background gain g ) , three degrees of gain modulation , with β = 1 . 5 , 2 and 3 , and three values of p = 5 , 10 and 15 . We consider three alternative ways in which the cue can be presented to the network . These cueing conditions and the resulting Iwhat−Iwhere relation are described below . In the previous section , the background gain was g = 0 . 5 . Without gain modulation , the network could reach high Iwhat values , sometimes retrieving all stored patterns , even from a very small initial cue . With gain modulation , Iwhere increased but Iwhat decreased . Here , we show that when the background gain is low , the interaction can be reversed , that is , gain modulation can actually increase both Iwhat and Iwhere . We set the background gain to g = 0 . 25 . As shown in Fig . 7 , for the case of complete cue ( as in Condition 1 above ) even without gain modulation Iwhat is very small . When incorporating a gain modulation mechanism , however , the amount of what information maintained by the network increases , together with the amount of where information . In section “Low gain regime versus high gain regime” ( see Materials and Methods ) , we discuss why the relation between Iwhat and Iwhere is different in the low gain and high gain regimes . Intuitively , the reason is as follows . Successful retrieval occurs only when the gain of the neurons that support the retrieved pattern is between a minimum gmin and a maximum gmax . In the low gain regime , the level of background gain is below gmin and the network cannot retrieve the patterns . When the gain is increased in part of the network , then it may enter the range [gmin , gmax] , allowing for retrieval to occur . At the same time , since that region has a higher gain , the retrieval bump does not drift away . When the background gain is high , instead , gain modulation stabilises the bump in the gain modulated area . This is accompanied , however , by a decrease in the size of the bump . The reason is that the higher neuronal gain increases the firing rate of neurons inside the bump ( the peak of the bump is higher ) and , to comply with the constant mean activity condition ( Eq . ( 3 ) ) , this increase in the peak activity is accompanied by a decrease in the spatial extent of the bump . Therefore , fewer connections are involved in retrieving the pattern and Iwhat decreases . As expected from this argument , increasing β too much even in the low gain regime should decrease Iwhat . This can be seen in Fig . 7 for β = 5 and p = 10 . When a retrieval bump is localised on a particular position , one can in principle use the rest of the network to retrieve other patterns , in the form of additional bumps of activity . If they can coexist with the first bump , the network would then be able to represent the position and identity of multiple objects simultaneously , without encountering the problem of binding . In random networks with no metric connectivity nor localised retrieval , retrieving multiple patterns is indeed possible , at very low storage loads [1] , [45] , [66]; in these networks , however , it is not possible to represent the position of the objects , which has to be represented elsewhere . If the what and where of multiple objects are represented in different networks , a binding problem arises . The localised retrieval process described here does not suffer from this problem . It is then important to assess the conditions which make it possible to stabilise ( at least ) two retrieval bumps simultaneously . Assume that a pattern is retrieved and , using localised gain modulation , the bump of activity is stabilised on a desired position . A second cue may then be presented to the network at another position . Even though most of the connections to each neuron in the network come from nearby neurons , the second pattern would still affect the first retrieval bump , because of the global inhibition in the simplest version of our model , as inhibition is taken to regulate a common threshold , such that the mean activity of the network is constant ( Eq . ( 3 ) ) . This introduces interactions between distal neurons , which are not directly connected by excitatory synapses , and such interactions are generally disruptive . A simple way to reduce such interaction is to assume that when the local mean activity in part of the network exceeds some limit value , the threshold is raised but only locally , regardless of the activity of neurons outside that region . The local threshold may also be regulated downward , to facilitate the emergence of a retrieved pattern in a region which would otherwise be kept at too low a mean activity level . With such additional provisions , multiple bumps can be formed and stabilised , as shown in the example in Fig . 8 . In our model , independent attractors are set up in a local cortical network only for object identity , as position invariant representations; but they can be accessed in a spatially focused mode , leading to position dependent activity . Associating a single representation to an object , which is then modulated by position , is a particular case of what in cognitive neuroscience parlance is sometimes referred to as type ( e . g . table ) and token ( particular instance of a type: e . g . a table in a particular position ) [76] , [77] . In the language of our model , the type is the original pattern of activity associated to an object and the token is the bumpy pattern that is localised in a particular position . An alternative mechanism is to store attractors associated to object-position pairs , that is storing a neural activity pattern for each token [78] . In this way , when a particular object is presented in a particular position , the attractor corresponding to the object-position pair would be activated , and could remain activated even after the object has been removed from the scene . The problem is that models which hypothesise distinct , discrete attractors for each possible object-position combination would certainly violate any conceivable storage capacity limit , because of the infinitely large number of possible positions of an object [79] . Furthermore , there is a major difference between the nature of what and where information , which makes attractors associated to object-position pairs unlikely: as opposed to what information , to which the brain can contribute from the information that it has previously stored , the brain does not usually retrieve positional information from memory , but rather has to maintain it as well as it can . Thus , it would seem rather implausible that the brain uses its storage capacity , arguably its most precious resource [29] , to store something that it does not have to retrieve . The difference between these two mechanisms is directly reflected in the storage capacity required for object-position attractors , in order to represent the same amount of information as the mechanisms studied here does through spatial modulation . Representing 6 bits of Iwhere and 2 bits of Iwhat ( corresponding to the black diamonds in Fig . 7 ) would require the storage of 2 ( 6+2 ) = 256 distinct object-position attractors . This is ca . 2 . 5 times beyond the number of attractors that a randomly connected network , with the same number of connections per neuron and the same mean activity level as what we used , could store [32] . This exorbitant requirement is due to effectively committing storage space separately to each pair , instead of using the physical arrangement of neurons in the tissue to represent Iwhere . Analytical results valid in the limit of large networks and optimal storage further support this conclusion , as we show in section “Comparision with other models” in the Materials and Methods . There , we also show that the difference in the efficiency of the two models will be even more pronounced for larger networks . There is , of course , a price to pay: the addition of a gain modulation mechanism to stabilise the position of the bump . In what follows , we discuss the possible physiological substrates of this gain modulation . In our model , localised gain modulation is crucial for maintaining where information as what information is being retrieved , and for maintaining both what and where information after the retrieval process is completed . When an object is presented as a stimulus , a signal should trigger an increase in the gain of neurons in an appropriate part of the network . Such higher gain should then be maintained by the same or a distinct mechanism during retrieval and thereafter , when the object is not present anymore but information about it has to be used ( e . g . during the delay period of a delay-match-to-sample task ) . What mechanisms can trigger the neuronal gain ? In vivo studies show that increasing the activity of a local cortical network increases the gain of its neurons [80]–[82] . Therefore , any mechanism that increases the mean activity of a part of the network could be used for triggering the gain modulation . One such source of increase in the activity is the cue itself . This requires that the pattern selective cue retains some spatial information; a scenario which we have shown to be particularly effective in minimising the trade-off between what and where information ( see Fig . 6C ) . Although this mechanism would be effective in this sense , it is doubtful whether it could be the only source of gain modulation in high level visual cortices . This is because experimental studies show that the position of the peak of the activity in visual cortical areas during visual stimulation is strongly correlated with the categorical properties of the stimulus and exhibits a weaker level of retinotopy [83]–[85] ( see also the following section “Storing patterns with spatial prefrence” ) . The situation may be different in more advanced cortical areas , such as PFC , in which such categorical maps have not been reported . Another possible source for increasing the gain is attentional signals . In this case the increase in the activity level required for gain modulation is induced by the attentional signal and the position of the bump corresponds to the position of the attentional spotlight . There are several reasons that make attention a likely source of activity localisation through gain modulation . fMRI studies on human subjects show that the retinotopic representation of the position of an attended object in visual cortices show increased activity [86]–[94] . Evidence from monkey neurophysiology also supports the idea that the attentional spotlight increases the gain of neurons inside the spotlight [95]–[100] . Furthermore , many studies in cognitive neuropsychology suggest that spatial , focal attention is critical to allow the binding of what and where information [77] , [101] , referred to as type and token information respectively [76] . Finally , a recent neuroimaging study shows that attention strongly enhances retinotopic representation in object selective visual areas , thus supporting the idea that attentional gain modulation is important for combined representation of what and where [102] . Although , these studies point to attentional signals as a strong candidate for initiating the gain modulation , a contribution may still be given by a weakly retinotopic initial cue . Further experimental work is required to disentangle the relative effect of the initial cue and attention on triggering the gain modulation . Once the increase in the gain of neurons in the right part of the network is triggered , it should be maintained during retrieval . Although the same mechanisms that initiated gain modulation can keep the gain high during retrieval , a promising mechanism for maintaining high level of gain , particularly after the stimulus is removed , is single neuron memory . Several studies show that the recent history of spiking increases the responsiveness of neurons , and that this increase can last for several seconds , thus exhibiting a form of single neuron short-term memory [62]–[64] . Assuming that such single neuron short-term memory mechanisms are responsible for the higher gain of neurons inside the bump , global signals that turn them on or off can strongly affect the level of what and where information that the network represents in its activity . As discussed above , the attentional signal may trigger the increase in neuronal gain and maintain it elevated for some time . After the attentional signal is removed , the increase in neuronal gain can be maintained by single neuron short term memory mechanisms . Attention can then be directed to another object , while what and where information about the first object is still decodable from neuronal activity . How long this information survives depends on how long the short term increase of the gain can be maintained by single neuron mechanisms . Understanding such mechanisms and comparing their time scale with behavioural times for maintaining combined what and where information , as well as pharmacologically interfering with them , one can test whether our model is relevant to real visual perception . One of the roles of attention is to bias the competition for limited processing resources in favour of the object that it is acting on [103] , [104] . Therefore , if the localised gain modulation that is needed in our model for combining what and where is induced by attention , it should be able to do the same . This is verified by computer simulations as shown in Fig . 9 . Two localised partial cues , corresponding to two different objects , are simultaneously given to a network . When the neuronal gain is uniform , the object with the larger cue will be retrieved , while the other one will be suppressed . However , if the neuronal gain in the area that receives the smaller cue is sufficiently large , the competition will be biased in favour of it . Interestingly , the level of gain modulation that is required to bias the competition towards the object with the small cue depends on the width of the connectivity , σ . Increasing the width of the neuronal connectivity increases the minimum level of gain modulation that is required for biasing the competition . This emphasises the role of local connectivity . In the model presented here , the units are taken to be arranged on a retinotopic patch of cortex , corresponding to at least a portion of visual field , but we assumed patterns of activity to be generated from a spatially uniform distribution ( see Eq . ( 4 ) ) . A more realistic model , however , should allow for the storage of spatially organised patterns [105] . This is important since , in the case of high level visual cortical areas , the overall position of intense neural activity during visual stimulation is strongly correlated with object identity or category . Regions in the visual cortex have been located that are preferentially selective for faces [106]–[109] , pictures of scenes [110] , [111] and buildings [112] , and complex object features [113] . This strong categorical map may coexist with a retinotopic map . The details of this combined organisation are far from clear , however , particularly insofar as it is expressed in the putative attractor states , after the stimulus is removed ( e . g . during delay periods ) , which is the situation relevant to our study . During visual stimulation , and when attention is not a main factor , some studies suggest that there is a weak retinotopy , with only a peripheral versus central bias and no angular representation [83]–[85] . Others , on the other hand , report the existence of multiple precise retinotopic maps in the same regions [114]–[116] , although still much weaker than the level of retinotopy in primary visual areas [117] . As mentioned in the previous section , such retinotopic maps could be enhanced by attention [102] . To include the coexistence of categorical and retinotopic maps in the model presented here , one might consider two limit cases , which roughly correspond to these two views . In the first case , category specificity and weak retinotopy coexist at the same spatial scale; one should then assume , in a refined model , that patterns are generated from multiple distributions , each of them corresponding to one category of objects , and patterns drawn from each have higher activity at a preferred position on the network . In this case , when there is no gain modulation the peaks of the retrieved patterns cluster depending on which distribution they came from . The peaks will also be more weakly correlated with the position of the cue compared to the case of spatially uniform patterns that we have discussed . With attentional gain modulation , one expects to see a clearer retinotopic map . This is in fact consistent with the abovementioned finding that attentional gain modulation enhances the retinotopic representation in advanced visual areas [102] . In the second limit case , retinotopy is expressed in object selective visual areas at a finer scale than category specificity , in which case one should allow for the present model to be simply multiplexed , to include one array on a distinct cortical patch for each object category . Further work is required , especially in view of many intermediate possibilities , to assess , for example , how much more gain modulation would be needed in order to stabilise a bump of activity away from its preferred position , and how this would affect retrieval . The ability to represent what and where information in the same network has also been proposed to be crucial to understand the functional significance of the differentiation among cortical layers [118] . Whereas most network models used to study attractor dynamics in associative memory do not consider cortical lamination , the core hypothesis of the proposal is that layer IV units , by virtue of their distinct connectivity , may privilege the representation of position information . Furthermore , through less adaptive spiking activity they may influence the dynamics of pyramidal units in the superficial layers only after these have engaged the attractor basin that leads to retrieve object identity . The differentiation was shown to be advantageous , in the model , through computer simulations , conducted with external inputs maintained active . In this regime no assessment was possible of whether genuine dynamical attractors had indeed been formed during memory storage , that will drive network dynamics in the absence of the cue . While the present work clarifies the conditions allowing a single layer network to represent what and where information , how they could be realized in a network with differentiated cortical layers remains to be explored . In discussing what and where information , we have made explicit reference , here , to object identity and position in the visual field . Where information could however be any feature that is mapped in the gross topography of the cortical sheet , such as frequency in the auditory system [119] , and in relation to which there is no meaning to using attractor dynamics in order to refine the afferent signal with what is stored in memory . In fact , this mapping need not even be topographically organised: the crucial factor is the existence of a map ( topographic or not ) [120] , that is produced as a result of the dependence of in Eq . ( 5 ) on i and j , and that is independent of the stored patterns . Where information would ideally be expressed by a continuous attractor and thus maintained e . g . as delay activity , except that continuity at a fine scale is disrupted by the storage of what memories . What information could instead be any feature that could benefit from attractor dynamics , because of its uneven statistical distribution , which makes some interpretation of the afferent signal more likely than others . If synaptic weights are produced by Eq . ( 5 ) , the weights of the connections that originate from a given neuron can be both negative and positive . This is against Dale's law and against our assertion that all neurons in the model network are excitatory . In this section , we show how the model described in sections “Firing rate description of the network” and “Stored memory patterns and synaptic weights” ( see Model ) can be conceptually derived from a more realistic formulation , in which all synaptic weights are positive . Let us first consider a network in which the firing rate of neuron i at time t+1 is determined by ( 14 ) in which Thi is the threshold of neuron i , Ii is its inhibitory input , and ( 15 ) The synaptic weights , Wij , in this network take the following form ( 16 ) where Jback is the background weight , ϖij = 1 if there is a connection from neuron j to neuron i and ϖij = 0 otherwise , and C is the average number of connections per neurons . For sufficiently large Jback , the resulting synaptic weights in Eq . ( 16 ) will be all positive . We can now show that a network with uniform threshold , as assumed in Eq . ( 2 ) , and synaptic weights of the form Eq . ( 5 ) , has equivalent dynamics as described by Eqs . ( 14 ) and ( 16 ) , when an additional condition is satisfied . Combining Eq . ( 14 ) with Eqs . ( 15 ) and ( 16 ) , the firing rate of neuron i can be written in terms of the firing rate of the other neurons as ( 17 ) in which Jij is the weight of the connection from neuron j to neuron i according to the prescription Eq . ( 5 ) . The assumption we now make is that the inhibitory feedback reacts in such a way that for each neuron , the last three terms in the parenthesis in Eq . ( 17 ) together become equal to a uniform effective threshold , Th . This effective threshold is simply chosen such that Eq . ( 3 ) holds . In this way , Eq . ( 17 ) reduces to ( 18 ) which is the same as Eq . ( 2 ) . In this section we briefly describe how the self-consistent equation for the local overlap with the retrieved pattern ( Eq . ( 10 ) ) can be derived . We refer the reader to [32] , [33] for more details . To start with , we assume , without loss of generality , that the first pattern ( μ = 1 ) is retrieved and therefore for . Using Eqs . ( 1 ) , ( 5 ) and ( 8 ) , we then write the input to neuron i as ( 19 ) Denfining zi as ( 20 ) and combining Eq . ( 19 ) and Eq . ( 2 ) , the activity of neuron i can be written as ( 21 ) Inserting vi from Eq . ( 21 ) into Eq . ( 8 ) we arrive at the following self-consistent equation for ( 22 ) Averaging the right hand side of Eq . ( 22 ) over the distribution of zj , η , and the connectivity pattern , yields the following equation ( which is the same as Eq . ( 10 ) ) ( 23 ) where 〈〉η , stands for averaging over the distribution of η , is the probability of connection ( Eq . ( 7 ) ) , and F̅j is the gain function , F , averaged over the distribution of zj ( 24 ) We now find the distribution of zj , which we denote by zj Pri ( zi ) . To do this we note that if the first pattern is retrieved , vjs , on the right hand side of Eq . ( 20 ) will be independent from each other and from ημ for μ≠1 . The assumption of independence is strictly correct when the network is highly diluted , that is when the number of presynaptic neurons shared by any two postsynaptic neurons is small [121] , [122] . When the network is not highly diluted , the calculation will be more involved , but yields qualitatively the same results [32] , [123] . Thus , for the sake of simplicity , we assume that the assumption of independence holds; for a complete derivation we refer the reader to the aforementioned references . With this independence assumption , the right hand side of Eq . ( 20 ) will be a sum of independent random variables , and therefore , Pri ( zi ) will be a Gaussian distribution . In the following we show that the mean of this Gaussian distribution is zero and also find a self-consistent equation for its variance . Noting that ( 25 ) ( 26 ) and using Eq . ( 21 ) , we get the following equations ( 27 ) ( 28 ) where indicates averaging over the distribution of η and ϖij , and 〈〉η indicates averaging over η . From Eq . ( 27 ) , we see that the mean of Pri ( zi ) is zero . In order to find the variance of Pri ( zi ) , we should average both sides of Eq . ( 28 ) over the distribution of zj . This is because , in the limit of large N and large C , this variance is expected not to depend on the exact realisation of any zj in the right hand side of Eq . ( 28 ) , but only on its statistical distribution . Performing this average yields the following equation for the variance that we denote by ρi2 ( 29 ) Equations ( 23 ) and ( 29 ) form a closed set of equations whose solutions determine the steady states of the system . Finding mi and ρi that satisfy these equations , we can find the activity of neurons in the steady states by plugging them in Eq . ( 21 ) . In the case of a randomly connected network , that is when is independent of i and j , and gi are also the same for all neurons , the solution of Eqs . ( 23 ) and ( 29 ) will be of the form mi = m and ρi = ρ . In this case the only spatial dependence of the steady state activities , Eq . ( 21 ) , will come from the dependence of ηi1 on i and since they are generated identically for each i , the probability that a neuron is active in the steady state will be uniform over the network . Spatially localised retrieval can be observed when depends on the distance between i and j . In this section we show how we compute what and where information , Iwhat and Iwhere , from simulations . We estimate the amount of what information , Iwhat , from the frequency of successful retrieval runs . To see how , let us assume that we cue pattern μc . Then after some time we look at the pattern of activity of the network , compute its dot product overlap with all stored patterns ( Eq . ( 9 ) ) and find that pattern μr , say , has been “retrieved” in that particular run , i . e . , it has the highest overlap with the activity of the network . We denote the probability of retrieving pattern μr given that we have cued pattern μc by Pr ( μr\|μc ) . Estimating this probability from the simulations , we can compute the information that the pattern of activity gives us about which pattern was presented as ( 30 ) where Pr ( μc ) is the probability of cueing pattern μc and ( 31 ) In the simulations all patterns are presented an equal number of times , therefore , ( 32 ) We denote the fraction of successful runs ( when μc = μr ) that we measure from the simulations by f , that is ( 33 ) Since in unsuccessful runs ( when μc≠μr ) , all patterns , except for μc are a priori equally likely to be retrieved , we have ( 34 ) Using Eqs . ( 32 ) – ( 34 ) in Eq . ( 30 ) , we can thus write for fixed degree of gain modulation , fixed background gain , and fixed number of patterns , ( 35 ) Note that the above is , strictly speaking , only a measure of the information implicit in the selection among the p patterns operated by attractor dynamics; under certain conditions , however , it can also serve as an indicator of the total information available in the firing pattern itself [124] . Iwhere is the mutual information between the peak of the local overlap after 200 time steps and the centre of the gain modulated area ( or the centre of the cue when there is no gain modulation ) . To estimate where information , Iwhere , we first measure the distance between the peak of the final overlap of the successful runs and the centre of the gain modulation , for each cued pattern . Then we make a histogram of these distances and calculate the fraction of runs which fall in any of the 10 distance bins chosen to be b1 = [0 , 5] , b2 = [5] , [10] , … , b10 = [45] , [50] . In this way we have the conditional probability , Pr ( k \| x ) , of having the peak of the activity in the kth distance bin , given that the peak was initially at position x on the lattice . With N neurons on each side of the lattice , we have Pr ( x ) = 1/N2 , and we can write Iwhere as ( 36 ) in which we have used the fact that Pr ( k \| x ) does not explicitely depend on x and we can simply denote it by Prk . Similarly to what we do for Iwhat , we have also assumed that for any such ring between the circles of radius 5k and 5 ( k−1 ) , centred on the gain modulation square , the final bump can be anywhere , with equal probability , on the ring . In this expression the factor 2k−1 accounts for the fact that the area covered by the kth bin is 2k−1 times the area of the first bin , and hence its a priori probability is 2k−1 times higher . The first term in Eq . ( 36 ) , is the maximum information value , Iwhere≃6 bits , in this approximation , i . e . , the logarithm ( in base 2 ) of the ratio between the “area” of the network ( 4900 ) and that of the smallest bin ( 5×5×π ) , and is achieved when all successful runs end up with a bump at d≤5 from its intended position . In this section we discuss why in the low gain regime , gain modulation aids retrieval of the patterns whereas in the high gain regime it has a negative effect . We start from the self-consistent equations , Eqs . ( 23 ) and ( 29 ) . Assume that the steady state of the network is a bump of activity over a part of the network with single neuron gain βg , whereas the rest of the network is silent with gain g . Furthermore , assume that mi and ρi that satisfy Eqs . ( 23 ) and ( 29 ) are nonzero inside the bump and zero elsewhere . Consider that inside the bump mi = m and ρi = ρ , where m and ρ can be regarded , just for simplicity , to be roughly constant . Then from Eqs . ( 23 ) and ( 29 ) we have: ( 37a ) ( 37b ) where α = p/C is the storage load and ( 38 ) Eqs . ( 37 ) are of the form of mean-field equations of a recurrent network with non-metric connections [79] , [125] ( assuming uniform values for mi and ρi inside the bump and zero outside is equivalent to assuming that the part of the network , over which the bump is formed , is behaving as an independent network ) . For each value of α , Eqs . ( 37 ) have non-zero solution for m , and thus the network can retrieve the stored patterns , if and only if gmin ( α ) <βg<gmax ( α ) , where gmax ( α ) and gmin ( α ) are functions of α . The effect of background gain g can now be readily seen . When g<gmin ( α ) retrieval does not happen without gain modulation . With gain modulation , however , the neuronal gain of the part of the network that is gain modulated will be boosted by a factor of β and for large enough β , the neuronal gain will be in the regime that supports retrieval i . e . gmin ( α ) <βg<gmax ( α ) . When the background gain g is high , βg can exceed gmax ( α ) , thus retrieval will not be successful . In this section , we discuss why it is more efficient to spatially modulate attractor states associated to objects , than to store distinct attractors for different positions of each object . Under optimal conditions , the number of attractors that an associative memory with C connections per neuron , but without metric connectivity , can retrieve is ( 39 ) where k is a constant that is primarily determined by the sparsity of the stored patterns [79] . Metric connectivity , which enables localised retrieval , decreases k by a moderate factor γ1≃3−4 [32] . Localised gain modulation , that stabilises the bump at an arbitrary position , decreases k again by another factor , γ2 , that for the parameters and network size we used turns out to be γ2≃4 . This is actually an overestimation of the decrease in storage capacity due to localised gain modulation , for realistic size networks . This is because when we calculate the mean of the right hand side of Eq . ( 22 ) over the distribution of connectivity patterns and η to get Eq . ( 23 ) , we ignore the fluctuations around this mean , that behave as . These fluctuations are what break the translational symmetry of the self-consistent equation , Eq . ( 23 ) , and make the bump favour a few positions over the others , and are compensated for by the localised gain modulation . As a result , less gain modulation is required for stabilising the bump when there are more connections per neuron . However , even with this estimate for γ2 , the process described here results in a moderate reduction in storage capacity ( 40 ) The spatial modulation described here can represent positional information with a resolution , where l is the lattice spacing and Np is the number of distinct position that can be resolved-in a large network , Np∼O ( N ) ( see Fig . 5 ) . On the other hand , the naive storage of distinct , unrelated attractors for each object position pair decreases the number of objects , whose identity could be retrieved , to ( 41 ) illustrating the wasteful use of memory resources for positional information , which in itself requires no memory . An alternative arrangement might be to associate attractors to objects , but allow each attractor to be a continuous 2D manifold , different for each object , so that position can be represented by the position of a bump of activity on such attractor manifold , unrelated to the position of the active neurons in the tissue . This arrangement corresponds to the multiple spatial charts model of Samsonovich and McNaughton [58] , introduced to account for the ability of rodents to track their own position in multiple spatial environments , by coding it as a group of coactive hippocampal place cells , which comprise a bump on a chart corresponding to each environment . Instead of assigning distinct charts to distinct spatial contexts , such as a square recording box rather than a circular one , one could well assign distinct charts to distinct objects , each of which would then have its own “private” continuous or quasi-continuous attractor , unrelated to the 2D arrangement of neurons in the tissue . The mathematical analysis of the multiple charts model [126] reveals that a network can store a number of charts equivalent to the number of attractors in a standard associative network of the same connectivity , reduced by a factor Nb , which is the number of place cell ensembles , uncorrelated with each other , required to “tile” a chart . In the simplest version of the model , each neuron shows a single place field in each environment ( at a different spatial position in each chart ) covering a fraction a of the total area of the environment . Then Nb≈ ( 1/a ) and , although the number of positions that can be represented accurately can be larger than Nb , still a≪1 for the network to be able to resolve position in space . Therefore , adapting the hippocampal model would also yield a lower capacity ( 42 ) because of the cost of creating a separate “virtual” space for each object . Simply utilising the position of neurons in the tissue to represent physical position for all objects , and reserving memory resources for object identity , provides the most efficient solution to combine what and where information . Note instead that in the hippocampus , to the extent that it utilises coactivity patterns to discriminate between different spatial contexts [127] , the position of neurons in the tissue cannot be used to code for position in real space , and in fact place field position in the chart is found to be unrelated to cell position in the tissue [128] . It is also worth mentioning that the same problem that we encountered for stabilising the bump at an arbitrary position will also appear in models that associate a distinct chart to each object [57] . Therefore , an extra mechanism will be required in this case , too , and the real pmax will be smaller than pmax in Eq . ( 42 ) by a factor similar to γ2 in our model .	Forming a coherent picture of our surrounding environment requires combining visual information about the position of objects ( where information ) with information about their identity ( what information ) . It also requires the ability to maintain this combined information for short periods of time after the stimulus is removed . Here , we propose a theoretical model of how this is accomplished in the brain , particularly when sensory input is incomplete , and missing what information should be supplied from what is stored in memory . The main idea is that local connectivity in cortical networks can allow the formation of localised states of activity . Where information can then be represented by the position of such “bumps” , and what information by the fine structure of the neuronal activity within them . We show that there is a difficulty with implementing this idea: noise and heterogeneity in connectivity cause bumps to drift , thereby losing where information . This problem can be solved by incorporating a localised increase in neuronal gain; this , however , interferes with retrieving what information and maintaining it in working memory . We quantify this interference via theoretical analysis of the model and show that , despite the interference , the proposed mechanism is an efficient one in retrieving what information while representing where information .	[ "Abstract", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "neuroscience/cognitive", "neuroscience", "neuroscience/theoretical", "neuroscience", "computational", "biology/computational", "neuroscience" ]	2008	Representing Where along with What Information in a Model of a Cortical Patch
Clathrin and the multi-subunit adaptor protein complex AP2 are central players in clathrin-mediated endocytosis by which the cell selectively internalizes surface materials . Here , we report the essential role of clathrin and AP2 in phagocytosis of apoptotic cells . In Caenorhabditis elegans , depletion of the clathrin heavy chain CHC-1 and individual components of AP2 led to a significant accumulation of germ cell corpses , which resulted from defects in both cell corpse engulfment and phagosome maturation required for corpse removal . CHC-1 and AP2 components associate with phagosomes in an inter-dependent manner . Importantly , we found that the phagocytic receptor CED-1 interacts with the α subunit of AP2 , while the CED-6/Gulp adaptor forms a complex with both CHC-1 and the AP2 complex , which likely mediates the rearrangement of the actin cytoskeleton required for cell corpse engulfment triggered by the CED-1 signaling pathway . In addition , CHC-1 and AP2 promote the phagosomal association of LST-4/Snx9/18/33 and DYN-1/dynamin by forming a complex with them , thereby facilitating the maturation of phagosomes necessary for corpse degradation . These findings reveal a non-classical role of clathrin and AP2 and establish them as indispensable regulators in phagocytic receptor-mediated apoptotic cell clearance . Phagocytosis of apoptotic cells is critical to tissue remodeling , suppression of inflammation and control of immune responses [1] , [2] . During phagocytosis , cell corpses are firstly engulfed and subsequently degraded by phagocytes , both phases being controlled by evolutionarily conserved regulators . In the lifetime of a C . elegans hermaphrodite , 131 somatic cells and about half the germ cells undergo apoptosis and the resulting cell corpses are quickly removed by neighboring cells in the soma or by sheath cells encasing the germ line . The engulfment of cell corpses is essentially controlled by two partially redundant signaling pathways that induce the cytoskeletal reorganization of engulfing cells [3] . In one pathway , the intracellular molecules CED-2/CrKII , CED-5/DOCK180 , and CED-12/ELMO act through a protein interaction cascade to induce the activation of the small GTPase CED-10/Rac1 , leading to the cytoskeleton reorganization necessary for engulfment [4]–[7] . In addition , the phosphatidylserine ( Ptdser ) receptor PSR-1 likely binds Ptdser , an “eat me” signal , and acts upstream of CED-2 , -5 , and -12 to regulate engulfment [4] . Two other signaling modules , INA-1/integrin-SRC-1/Src and UNC-73/TRIO-MIG-2/RhoG , were also found to function through the CED-5-CED-12 motility-promoting complex to facilitate CED-10 activation for corpse engulfment [8] , [9] . In addition , a non-canonical Wnt pathway consisting of the MOM-5 receptor , GSK-3 kinase and APC/APR-1 may act through CED-2 to regulate CED-10 activity for cell corpse engulfment during early embryo development [10] . In the other pathway , the phagocytic receptor CED-1 , which shares homology with the human scavenger receptor SREC , LRP/CD91 and MEGF10 , and Drosophila Draper and Six-microns-under ( SIMU ) [11]–[15] , recognizes apoptotic cells by interacting with TTR-52 , a PtdSer-binding protein secreted from engulfing cells [16] . The adaptor protein CED-6/Gulp likely acts downstream of CED-1 to transduce engulfing signals to other effectors including the large GTPase DYN-1/dynamin , resulting in cell corpse engulfment and formation of phagosomes [14] , [17] , [18] . In addition , the ABC transporter CED-7 is also required for cell corpse recognition by CED-1 in embryos [11] , [19] . Recent studies suggest that CED-7 acts with TTR-52 and NRF-5 , another secreted PtdSer-binding protein , to mediate PtdSer transfer from cell corpses to phagocytes , thus promoting the recognition of cell corpses by CED-1 [20] , [21] . Subsequent to corpse internalization , CED-1 is recycled from the phagosome back to the plasma membrane by the retromer complex [22] . Phagosomes enclosing cell corpses then undergo a maturation process by dynamically fusing with endocytic organelles including early and late endosomes as well as lysosomes , leading to formation of phagolysosomes in which cell corpses are ultimately digested . It has been found that several molecules required for endocytic trafficking , such as DYN-1/Dynamin , the phosphatidylinositol-3 kinase ( PI3K ) VPS-34 , small GTPases and their regulators or effectors including RAB-2 , RAB-5 , TBC-2 , RAB-7 , RAB-14 , and the HOPS complex , act in an ordered manner to regulate phagosome maturation [23]–[29] . As the phagolysosome forms , it is progressively acidified in order to activate lysosomal enzymes needed for cell corpse digestion [30] . The phagocytic receptor CED-1 plays a leading role in apoptotic cell clearance by recognizing cell corpses and transducing signals for engulfment and phagosome maturation . Nevertheless , it remains largely unknown how the CED-1-mediated signaling pathway triggers the cytoskeletal reorganization required for corpse internalization . In addition , the mechanisms governing the transition from corpse internalization to phagosome maturation are poorly understood . Interestingly , CED-1-mediated phagocytosis of cell corpses appears to resemble clathrin-mediated endocytosis ( CME ) of cell surface molecules in that both events cause receptor-dependent internalization of extracellular cargoes differing only in size [31] . In CME , recognition of the cytoplasmic domains of plasma membrane receptors by adaptor proteins triggers the formation of clathrin-coated vesicles ( CCVs ) with diameters ranging from 10–200 nm [32] , [33] . The formation of cargo-containing CCVs requires several protein module-mediated events , including FCH domain-only ( FCHO ) complex-mediated initiation , adaptor protein 2 ( AP2 ) -dependent cargo selection and coat building , dynamin-mediated scission , and auxilin- and heat shock cognate 70 ( HSC70 ) -dependent uncoating [32] . Recent studies revealed that some of the molecules required for CME are involved in phagocytosis of pathogens or the maturation of phagosomes containing apoptotic cells . For example , clathrin and the adaptor protein Dab2 were found to be important for phagocytosis of pathogenic bacteria by mammalian cells [34] . In C . elegans , DYN-1/dynamin regulates the initiation of phagosome maturation [23] . However , whether other components of the CME pathway play a role in apoptotic cell clearance remains unknown . In this study , we investigated the mechanisms underlying CED-1-mediated cytoskeleton reorganization for phagocytosis and uncovered the role of CME regulators in apoptotic cell clearance . Our findings revealed that clathrin and the AP2 complex , the central players in CME , are critical to apoptotic cell removal in C . elegans . We found that clathrin and AP2 act downstream of CED-1 and CED-6 by forming a complex with them , which mediates the rearrangement of the actin cytoskeleton required for cell corpse engulfment . In addition , we demonstrated that LST-4 , the C . elegans homolog of Snx9/18/33 , functions at an early step of phagosome maturation by promoting phagosomal association of DYN-1/dynamin . Furthermore , we demonstrated that clathrin and AP2 also interact with LST-4 and DYN-1 to regulate the initiation of phagosome maturation required for cell corpse degradation . These findings suggest that clathrin and AP2 play essential roles in the phagocytic receptor CED-1-mediated apoptotic cell clearance pathway by regulating cytoskeletal reorganization and facilitating phagosome maturation . To explore how the phagocytic receptor CED-1 and its downstream adaptor CED-6 function to induce cytoskeletal reorganization for cell corpse engulfment , we firstly sought to identify proteins that are in complex with endogenous CED-1 and/or CED-6 . Using antibodies specific for the C-terminus of CED-1 ( CED-1C ) [22] and CED-6 , we performed immunoprecipitations in whole cell lysates of wild type ( N2 ) , ced-1 ( e1735 ) and ced-6 ( n1813 ) strong loss-of-function mutants , and analyzed proteins in the precipitates using liquid chromatography-coupled tandem mass spectrometry ( LC-MS/MS ) . Interestingly , multiple peptides of the heavy chain of clathrin ( CHC-1 ) were identified from proteins co-precipitated with CED-1 and CED-6 in the N2 lysate but not in lysates of ced-1 ( e1735 ) or ced-6 ( n1813 ) mutants ( Figure S1A and S1B ) . We therefore used RNA interference ( RNAi ) to deplete chc-1 and examined the persistence of cell corpses in C . elegans germ lines . We found that germ cell corpses accumulated significantly in an age-dependent manner in chc-1 ( RNAi ) animals . A similar increase was observed at 25°C in a chc-1 temperature-sensitive mutant , b1025ts [35] , though to a lesser extent than in chc-1 ( RNAi ) animals ( Figure 1A ) . These results indicate that loss of clathrin function caused accumulation of apoptotic cells in C . elegans germ lines . Previously it was also reported that clathrin RNAi induced an elevation in the number of germ cell corpses [23] , but how clathrin functions in this process remains unclear . Given the central role of clathrin in CME , we went on to investigate whether inactivation of other regulators required for CME could also result in accumulation of apoptotic cells . We used RNAi to deplete C . elegans homologs of individual mammalian proteins involved in CME and examined the persistence of germ cell corpses . Compared to animals with control RNAi , a significant increase in germ cell corpses was observed in animals treated with RNAi of apa-2 , apb-1 and dpy-23 , which encode the α , β2 and μ2 subunits of the AP2 complex , respectively ( Table S1 ) . A time-course analysis confirmed that germ cell corpses increased significantly in an age-dependent manner in animals with RNAi of apa-2 , apb-1 , and dpy-23 , but not aps-2 , which encodes the σ2 subunit of the AP2 complex ( Figure 1B ) . In addition , RNAi of lst-4 and dyn-1 , which encode C . elegans homologs of mammalian sorting nexins 9/18/33 and dynamin , respectively , also led to a strong accumulation of germ cell corpses ( Table S1 ) [18] , [22] , [36] . Clathrin and the AP2 complex are essential for formation of CCVs while sorting nexin 9 and dynamin are required for scission of CCVs from the plasma membrane during endocytosis [32] , [37] . To distinguish whether the increase in germ cell corpses in above RNAi animals resulted from excessive apoptosis or defective cell corpse clearance , we measured the duration of cell corpses using time-lapse recording . In animals with control RNAi , the average duration of germ cell corpses was 29 . 3±2 . 5 min ( mean±SEM , standard error of the mean ) . In contrast , the majority of germ cell corpses in chc-1 ( RNAi ) and apb-1 ( RNAi ) animals persisted longer than 120 min , and no cell corpses existed less than 60 min ( Figure 1C ) , suggesting that inactivation of clathrin or AP2 likely caused defective cell corpse clearance . To prove this , we performed transmission electronic microscopy ( TEM ) analysis to examine the engulfment of germ cell corpses . In chc-1 ( RNAi ) animals , 28 out of 50 germ cell corpses examined ( 56% ) from 5 gonad arms appeared not to be fully engulfed by sheath cells ( Figure 1D ) . Similarly , 12 of 25 corpses ( 48% ) from 7 gonad arms of apb-1 ( RNAi ) animals were found not to be internalized ( Figure 1D ) . In contrast , in animals treated with gla-3 RNAi , which induces excessive germ cell apoptosis without affecting cell corpse clearance , 100% of cell corpses were fully internalized by gonad sheath cells ( Figure 1D ) [22] , [38] . These findings indicate that the engulfment of cell corpses was impaired when clathrin and the AP2 complex were down-regulated . In C . elegans , the ced-1/6/7 and ced-2/5/12/10 signaling pathways act redundantly to mediate cell corpse engulfment [3] . As our findings revealed that CHC-1 and AP2 are involved in cell corpse engulfment , we tested whether depletion of individual AP2 components and chc-1 could exert an additive effect on defective corpse engulfment in mutants deficient in either engulfment pathway . In ced-1 ( e1735 ) and ced-6 ( n2095 ) strong loss-of-function mutants in the cell corpse recognition pathway , RNAi of apb-1 , dpy-23 , or chc-1 did not obviously change the numbers of germ cell corpses at all adult ages examined ( Figure 2A ) . In contrast , RNAi depletion of these three genes significantly enhanced the numbers of germ cell corpses in ced-2 ( n1994 ) and ced-5 ( n1812 ) strong loss-of-function mutants affecting the cytoskeletal reorganization pathway ( Figure 2B ) . These results suggest that chc-1 and genes of the AP2 complex likely act within the same genetic pathway as ced-1 and ced-6 to regulate cell corpse clearance . To determine whether clathrin and AP2 are directly involved in formation of cell corpse-containing phagosomes , we generated transgenes expressing green fluorescence protein ( GFP ) -tagged CHC-1 , APA-2 and DPY-23 under the control of their own promoters . We found that these GFP-tagged proteins associated with phagosomes in the germ line ( Figure S2A ) . Similarly , GFP-tagged APA-2 ( APA-2::GFP ) and mCherry-fused CHC-1 ( mCherry::CHC-1 ) driven by the engulfing cell-specific ced-1 promoter also appeared on phagosomes ( Figure 2C and 2D ) . The fusions of CHC-1 with GFP and mCherry were functional in that both GFP::CHC-1 driven by the chc-1 promoter ( yqEx480 ) and mCherry::CHC-1 driven by the ced-1 promoter ( yqIs98 ) rescued the cell corpse phenotype in chc-1 ( b1025ts ) mutants ( Figure S2B ) . Likewise , APA-2::GFP driven by the apa-2 promoter ( yqEx481 ) or the ced-1 promoter ( yqIs99 ) rescued the accumulation of germ cell corpses in animals treated with RNAi complementary to the 3′ untranslated region ( 3′ UTR ) of apa-2 ( Figure S2C ) . Interestingly , the labeling of cell corpses by APA-2::GFP was strongly inhibited by chc-1 RNAi , and RNAi of apa-2 and apb-1 conversely suppressed corpse labeling by mCherry::CHC-1 ( Figure 2C and 2D; Figure S7G; and data not shown ) . Thus , AP2 and CHC-1 likely form a complex and are recruited to phagosomes in a mutually dependent manner . Alternatively , CHC-1 is required for the stabilization of AP2 at the phagosome . Furthermore , in ced-1 ( e1735 ) , ced-1 ( RNAi ) and ced-6 ( RNAi ) animals , few germ cell corpses were labeled by APA-2::GFP or mCherry::CHC-1 , suggesting that CED-1 and CED-6 are required for phagosomal association of AP2 and clathrin ( Figure 2C and 2D ) . On the other hand , we also examined the effect of loss of chc-1 and apb-1 on cell corpse recognition by CED-1::GFP and phagosomal recruitment of GFP::CED-6 , both of which functionally rescued the defective cell corpse clearance in ced-1 ( e1735 ) and ced-6 ( n2095 ) mutants ( Figure S2D and S2E ) . The surrounding of germ cell corpses by CED-1::GFP or GFP::CED-6 in apb-1 ( RNAi ) and chc-1 ( RNAi ) animals was similar to that in wild type and in lst-4 ( tm2423 ) mutants affecting phagosome maturation ( see below ) ( Figure S3 ) , indicating that loss of apb-1 and chc-1 did not affect cell corpse recognition by CED-1 and phagosomal recruitment of CED-6 . Collectively , these data suggest that phagosomal recruitment of AP2 and clathrin occurs downstream of CED-1 and CED-6 . Next we investigated whether loss of clathrin or AP2 function affects the rearrangement of the actin cytoskeleton , which is required for internalization of cell corpses . For this purpose , we generated transgenes expressing GFP-fused ACT-1 , an actin isoform that controls cytoplasmic microfilament function , and GFP-tagged Drosophila Moesin ( GFP::Moesin ) [39] , a filamentous actin ( F-actin ) -specific-binding protein , both of which were driven by the engulfing cell-specific ced-1 promoter . In wild-type animals , about 50% of germ cell corpses were surrounded by GFP::ACT-1 . In ced-1 ( e1735 ) or ced-6 ( RNAi ) animals , however , the labeling of germ cell corpses by GFP::ACT-1 was strongly reduced ( Figure 3A and 3B ) . Similar reduction in labeling of germ cell corpses by GFP::ACT-1 was observed in chc-1 ( RNAi ) and apb-1 ( RNAi ) animals ( Figure 3A and 3B ) . Similarly , whereas about 60% of germ cell corpses were positive for GFP::Moesin in the wild type , only 10–20% of germ cell corpses were encircled by GFP::Moesin in animals treated with RNAi of ced-1 , chc-1 or apb-1 ( Figure S4 ) . Thus , loss of chc-1 and apb-1 resulted in a failure in actin cytoskeleton rearrangement required for cell corpse engulfment , like that caused by loss of ced-1 or ced-6 . To prove this , we examined the recruitment of mCherry-tagged ACT-1 by phagosomes positive for CED-1::GFP or GFP::CED-6 in chc-1 ( RNAi ) and apb-1 ( RNAi ) animals . RNAi of chc-1 or apb-1 caused a strong decrease in labeling of CED-1::GFP-positive phagosomes by mCherry::ACT-1 from 77% to 35–40% ( Figure 3C and 3E ) . Similarly , only 37–41% of GFP-CED-6-positive phagosomes were labeled by mCherry::ACT-1 in animals with RNAi of chc-1 or apb-1 , compared to 85% in animals with control RNAi ( Figure 3D and 3F ) . Taken together , these findings suggest that CHC-1 and AP2 act downstream of CED-1 and CED-6 to mediate the rearrangement of the actin cytoskeleton required for corpse engulfment . To understand how clathrin and AP2 may cooperate with CED-1 and CED-6 to control cytoskeletal rearrangement in the CED-1 cell corpse engulfment pathway , we first tested whether AP2 components and CHC-1 can physically interact with CED-1 using in vitro GST pull-down assays . The GST-fused intracellular region of CED-1 ( CED-1C , amino acids 933–1111 ) , but not GST , directly interacted with 35S-labeled APA-2 synthesized by in vitro translation ( Figure 4A ) . However , other AP2 subunits labeled by 35S and the purified recombinant C-terminal region of CHC-1 ( CHC-1C , amino acids 825–1628 ) did not show a detectable interaction with CED-1C , suggesting that CED-1 likely interacts with the AP2 complex through its α subunit ( Figure 4A ) . To determine whether the intracellular region of CED-1 is important for phagosomal recruitment of CHC-1 and/or AP2 , we examined if expression of a GFP-fused CED-1 with the C-terminal region deleted ( CED-1ΔC::GFP , smIs110 ) could rescue the defective phagosomal recruitment of mCherry::CHC-1 in ced-1 ( e1735 ) mutants . CED-1ΔC::GFP failed to rescue the cell corpse phenotype ( Figure S2D ) but labeled cell corpses in ced-1 ( e1735 ) mutants; however , barely any mCherry::CHC-1 was found to co-localize with CED-1ΔC::GFP ( Figure 4B ) . In contrast , the full-length CED-1::GFP ( smIs34 ) fully rescued the cell corpse phenotype in ced-1 ( e1735 ) animals ( Figure S2D ) and about 60% of CED-1::GFP-positive cell corpses were labeled by mCherry::CHC-1 . Thus the intracellular region of CED-1 is required for phagosomal recruitment of CHC-1 . We also tested the interaction of GST-fused CED-6 with individual AP2 components and CHC-1 , and found that GST-CED-6 directly interacted with 35S-labeled APA-2 , APB-1 and DPY-23 , and His6-tagged CHC-1C ( Figure 4A ) . These findings suggest the possibility that CED-1 and CED-6 form a complex with the AP2 complex and CHC-1 . To prove this , we examined the interaction of CHC-1 and individual components of the AP2 complex with endogenous CED-1 and CED-6 by performing immunoprecipitations with CED-1C- or CED-6-specific antibodies . We found that endogenous CED-6 associated with CED-1 immunoprecipitated by the CED-1C antibody , providing direct evidence that CED-1 and CED-6 form a complex in C . elegans ( Figure 4C ) . Importantly , APA-2::GFP was co-immunoprecipitated with endogenous CED-1 and CED-6 ( Figure 4D and 4E ) , and similar co-immunoprecipitation of DPY-23::GFP and GFP::CHC-1 with endogenous CED-6 was observed ( Figure 4F and 4G ) . The specificity of these in vivo protein interactions was supported by the absence of any interaction , using the same immunoprecipitation assay , between CED-6 and GFP-fused LST-4 and DYN-1 , two factors required for phagosome maturation ( see below ) ( Figure 4H and 4I ) . Thus the in vivo interactions of CHC-1 and individual AP2 components with CED-6 or CED-1 are consistent with their direct interactions in vitro , suggesting that the AP2 complex and CHC-1 likely fulfill their functions in cell corpse engulfment by forming a complex with CED-1 and CED-6 . As EM analysis indicated that a significant proportion of germ cell corpses were still internalized by sheath cells in chc-1 ( RNAi ) and apb-1 ( RNAi ) animals , we wondered whether maturation of phagosomes containing cell corpses was affected in these animals . To assess this , we examined the acidification of phagosomes with LysoSensor Green DND-189 , an indicator of normal progression of phagosome maturation . We found that germ cell corpses were mostly negative for LysoSensor Green DND-189 staining in chc-1 ( RNAi ) and apb-1 ( RNAi ) animals compared to animals with gla-3 RNAi that induces an elevation in apoptosis without affecting cell corpse clearance [38] , suggesting that the maturation of phagosomes was inhibited ( Figure 5A ) . To corroborate this conclusion , we examined phagosomal recruitment of several effectors essential for phagosome maturation in apb-1 ( RNAi ) and chc-1 ( RNAi ) germ lines , including GFP-fused RAB-5 ( GFP::RAB-5 ) and mCherry-fused RAB-14 ( mCherry::RAB-14 ) , two small GTPases required for phagosomal progression from early to late stages , and mCherry-fused NUC-1 ( NUC-1::mCherry ) , a lysosomal DNase that indicates the formation of phagolysosomes [27] . We found that the labeling of cell corpses by these phagosomal markers in apb-1 ( RNAi ) and chc-1 ( RNAi ) animals was greatly reduced compared to that in wild type ( Figure 5B–5E ) , indicating that phagosomes in these animals arrested at an early stage of maturation . Taken together , these data indicate that clathrin and AP2 act at an early stage of phagosome maturation , impairment of which inhibited the progression of phagosomes from early to late stages . To elucidate how AP2 and CHC-1 function in phagosome maturation in addition to their role in cell corpse engulfment , we sought to determine their functional interactions with two other regulators identified in our screen , LST-4 and DYN-1 , the C . elegans homologs of mammalian Snx9/18/33 and dynamin , respectively . Snx9 and dynamin act together with AP2 and clathrin to regulate the formation of CCVs in CME [32] , [40] , [41] . DYN-1 was previously shown to act at an early stage of phagosome maturation by forming a complex with VPS-34 and RAB-5 whereas LST-4 likely affects cell corpse degradation at a similar stage to DYN-1 [22] , [23] , [36] , [42] . As the first step towards our goal , we set out to clarify the role of LST-4 in phagosome maturation by comparing the cell corpse phenotype of two deletion mutants , tm2423 and qx159 . We found that these mutants accumulate germ cell corpses to similar levels in an age-dependent manner ( Figure S5A and S5B ) . In addition , around 70% of lst-4 ( tm2423 ) germ cell corpses were found to be encircled by GFP::Moesin , compared with 60% in wild type , indicating that loss of lst-4 did not affect cell corpse internalization ( Figure S5C and S5H ) . However , germ cell corpses labeled by the early phagosome markers YFP::2xFYVE , GFP::RAB-5 , and mCherry::RAB-14 , and the late phagosome marker GFP::RAB-7 , were greatly reduced in lst-4 ( tm2423 ) animals , indicating that loss of lst-4 inhibited the recruitment of factors required for phagosome maturation ( Figure S5D–S5H ) . Moreover , loss of lst-4 also blocked phagosome acidification as the majority of germ cell corpses were negative for LysoSensor Green DND-189 staining in either lst-4 ( tm2423 ) single mutants or double mutants of lst-4 ( tm2423 ) with vps-18 ( tm1125 ) that was previously shown to cause defective phagolysosome formation but not phagosome acidification [29] ( Figure S6A and S6B ) . This contrasts to the high proportion of corpses stained by the same dye in gla-3 ( RNAi ) animals , in which cell corpses are normally removed , and in vps-18 ( tm1125 ) single mutants ( Figure S6A and S6B ) . Importantly , we found that LST-4 was recruited to phagosomes using LST-4::GFP or LST-4::mCherry fusions that fully rescued the cell corpse phenotype in lst-4 ( tm2423 ) mutants , even though they appeared cytoplasmic in several tissues ( Figure S6C–S6E; Figure S7A ) . The phagosomal association of LST-4 was blocked by loss of ced-1 and ced-6 but not RNAi depletion of dyn-1 , rab-5 and rab-7 , three genes required for phagosome maturation but not corpse engulfment ( Figure S6E and S6F ) . Together , these findings , in agreement with the results obtained by Almendinger et al . [36] , establish that LST-4 acts at an early stage of phagosome maturation . We next characterized the functional interaction between LST-4 and DYN-1 . In animals co-expressing LST-4::mCherry and DYN-1::GFP , which is able to rescue the defective cell removal in dyn-1 ( ky51 ) mutants , both proteins were found to colocalize on phagosomes ( Figure S7A and S7B ) . Time-lapse analysis revealed that both proteins were simultaneously recruited to the phagosome and quickly formed a crescent-like structure , before dissociating from the phagosome at the same time ( Figure 6A ) . Using immunoprecipitation we further found that these two proteins associated with one another in C . elegans ( Figure S7C , top panel ) whereas they did not show detectable in vivo interaction with CED-6 ( Figure 4H and 4I ) . Consistent with this , His6-tagged recombinant LST-4 directly interacted with GST-fused DYN-1 , which confirmed the in vitro interaction of these two proteins reported previously [42] . Nevertheless , no interaction of LST-4His6 or DYN-1His6 with CED-1C or CED-6 was detected in the same GST pull-down assay ( Figure S7C , bottom panel ) . Together these results indicate that LST-4 and DYN-1 form a complex to regulate phagosome maturation but do not act in complex with CED-1 or CED-6 . To further determine the effect of LST-4-DYN-1 interaction on their association with phagosomes , we monitored the dynamic association of DYN-1::GFP with phagosomes in germ lines of wild-type and lst-4 ( tm2423 ) animals , and phagosomal association of LST-4::GFP in wild-type and dyn-1 ( RNAi ) germ lines . In the wild type , DYN-1 was initially localized to the periphery of the phagosome and then quickly became enriched to form a large patch-like structure ( Figure 6B , 0–14 min ) . DYN-1 then became more evenly distributed on the phagosome before forming punctate structures ( Figure 6B , 14–56 min ) , which likely represent the dissociation of DYN-1 from the phagosome . Unlike in wild-type , DYN-1::GFP neither became sharply enriched nor formed an obvious patch on phagosomes in lst-4 ( tm2423 ) mutant germ lines ( Figure 6B ) , suggesting that loss of lst-4 likely affected the enrichment or stabilization of DYN-1 on phagosomes . These findings are in agreement with the observations made previously by Lu et al . that loss of lst-4 impaired the phagosomal association of DYN-1 during embryonic cell corpse removal [42] . In addition , we noticed that DYN-1::GFP was more enriched on phagosomes when co-expressed with LST-4::mCherry ( compare Figure 6A and 6B ) . On the other hand , RNAi depletion of dyn-1 seemed not to affect the association of LST-4 with phagosomes , because no obvious difference in the dynamic association of LST-4::GFP with phagosomes was observed between dyn-1 ( RNAi ) and control RNAi animals ( Figure 6C ) . These results , together with the findings made by Lu et al . and Almendinger et al . [36] , [42] , establish that LST-4 promotes phagosomal activity of DYN-1 . Importantly , we further found that lst-4 ( tm2423 ) mutants expressing DYN-1::GFP ( qxIs139 ) displayed an obvious reduction in germ cell corpses compared with the same mutants without DYN-1::GFP expression ( Figure 6D ) . For example , lst-4 ( tm2423 ) animals expressing DYN-1::GFP ( qxIs139 ) contained 6 . 3±0 . 6 ( mean±SEM ) and 7 . 6±0 . 6 corpses per gonad arm at adult ages of 36 and 48 h post L4 , respectively , compared with 29 . 9±0 . 6 and 55 . 2±1 . 1 in lst-4 ( tm2423 ) mutants ( Figure 6D ) . In contrast , dyn-1 RNAi caused similar levels of germ cell corpse accumulation in both wild type and animals expressing LST-4::GFP ( yqIs114 ) ( Figure 6E ) . Taken together , these findings provide strong evidence that LST-4 forms a complex with DYN-1 and acts through the latter to promote the initiation of phagosome maturation . Having demonstrated that both CHC-1-AP2 and LST-4-DYN-1 complexes act at a very early stage of phagosome maturation , we asked how CHC-1 and AP2 might affect LST-4 and DYN-1 . Firstly , we tested if depletion of chc-1 and AP2 had an additive role in cell corpse accumulation in lst-4 ( tm2423 ) animals , and found that RNAi of chc-1 , apb-1 , and dpy-23 did not affect the numbers of germ cell corpses in lst-4 ( tm2423 ) mutants ( Figure 7A ) . Secondly , we investigated whether phagosomal association of LST-4 and DYN-1 were affected by inactivating chc-1 and the AP2 complex . In chc-1 ( RNAi ) and apb-1 ( RNAi ) germ lines , the labeling of germ cell corpses by LST-4::GFP and DYN-1::GFP was strongly reduced compared to that in wild type ( Figure 7B and 7C; Figure S7D ) , indicating that CHC-1 and AP2 are important for phagosomal association of LST-4 and DYN-1 . In contrast , phagosomal association of APA-2::GFP and mCherry::CHC-1 in lst-4 ( tm2423 ) , dyn-1 ( ky51 ) , or dyn-1 ( RNAi ) animals were similar to that in wild type ( Figure S7E–S7G ) . These results suggest that CHC-1 and the AP2 complex function upstream of LST-4 and DYN-1 in phagosome maturation . Finally , we tested whether CHC-1 and individual AP2 subunits could directly interact with LST-4 and/or DYN-1 . As shown in Figure 7D , 35S-labeled APA-2 , APB-1 , and DPY-23 interacted with GST-fused DYN-1 and LST-4 , but not GST or GST-fused CED-9 , an anti-apoptotic protein acting at the cell-killing stage [43] . Similarly , purified recombinant CHC-1C interacted with GST-DYN-1 and GST-LST-4 but not GST or GST-fused CED-9 . Thus clathrin and AP2 likely form a complex with LST-4 and DYN-1 . To prove this , we examined whether DYN-1 and LST-4 are indeed in complex with CHC-1 and AP2 in C . elegans . Using immunoprecipitation we found that mCherry::CHC-1 associated with DYN-1::GFP in animals co-expressing these two proteins ( Figure 7E ) . Similarly , mCherry::CHC-1 and LST-4::GFP associated with one another as revealed by co-immunoprecipitation ( Figure 7F ) . In addition , we found that mCherry::LST-4 co-immunoprecipitated with APA-2::GFP and DPY-23::GFP ( Figure 7G and 7H ) , indicating that LST-4 interacts with AP2 components in C . elegans . Taken together , the in vitro and in vivo protein interactions among LST-4 , DYN-1 , CHC-1 and AP2 components ( Figure 7D–7H , Figure S7C ) strongly suggest that clathrin and AP2 form a complex with LST-4 and DYN-1 , thereby promoting phagosome maturation during cell corpse clearance . During phagocytosis , the phagocytic receptor CED-1 recognizes cell corpses and transduces engulfment signals to the CED-6 adaptor . DYN-1/dynamin was also reported to participate in the ced-1 pathway for corpse engulfment , and likely acts downstream of CED-1 and CED-6 [18] . Nevertheless , it is not clear how these factors coordinate to induce the rearrangement of the actin cytoskeleton , a key event required for cell corpse internalization . Although it was previously proposed that the two engulfment pathways for cytoskeletal reorganization converged on the CED-10 GTPase , the molecular link between the phagocytic receptor CED-1 and CED-10 remains to be identified [44] . In this study , we explored the role of major regulators of CME , a process that internalizes cell surface materials by use of clathrin-coated vesicles , in phagocytosis of apoptotic cells . Our findings revealed that clathrin and the AP2 complex are essential players in the process of cell corpse engulfment . Inactivation of the clathrin heavy chain CHC-1 or individual components of AP2 resulted in accumulation of cell corpses in the C . elegans germ line . Moreover , RNAi of chc-1 or AP2 components significantly enhanced the engulfment defects in ced-2 and ced-5 strong loss-of-function mutants but not mutants deficient in ced-1 and ced-6 , suggesting that the chc-1 and AP2 genes likely act within the same genetic pathway as ced-1 and ced-6 . Our results demonstrated that CHC-1 and the AP2 complex associate with phagosomes containing cell corpses in an inter-dependent manner and their phagosomal recruitment requires CED-1 and CED-6 . Importantly , loss of clathrin or AP2 function severely impaired the rearrangement of the actin cytoskeleton required for corpse engulfment . Altogether these findings provide strong evidence that clathrin and AP2 function downstream of CED-1 and CED-6 and likely mediate the cytoskeletal reorganization required for cell corpse internalization ( Figure 8 ) . Our findings suggest that clathrin and the AP2 complex serve a dual role in the process of apoptotic cell removal ( Figure 8 ) . On one hand , clathrin and AP2 are important for cell corpse engulfment by acting downstream of CED-1 and CED-6 to mediate cytoskeletal rearrangement in engulfing cells ( Figure 8 ) . This function is likely achieved by forming a protein complex with CED-1 and CED-6 . In support of this conclusion , we found that CED-1 indeed forms a complex with CED-6 in vivo . Remarkably , our immunoprecipitation and in vitro GST pull-down results revealed that the CED-6 adaptor protein directly interacts with CHC-1 and individual components of the AP2 complex in C . elegans . Interestingly , we found that the CED-1 receptor likely interacts with AP2 via the α subunit of the latter . Because clathrin can function as an actin organizer at large membrane interfaces that far exceed the size of conventional CCVs [45] and loss of CHC-1 and AP2 function caused defective recruitment of actin around cell corpses , we propose that the formation of a protein complex by CED-1 , CED-6 , AP2 and CHC-1 provides a hub for recruitment and assembly of actin for cell corpse engulfment . On the other hand , clathrin and the AP2 complex are essential for phagosome maturation following corpse internalization . Loss of CHC-1 and AP2 function abrogated the acidification of phagosomes and inhibited phagosomal recruitment of downstream effectors required for phagosome maturation . In addition , our data demonstrated that LST-4 interacts with DYN-1 to promote its association with phagosomes . Clathrin and AP2 facilitate phagosomal association of the LST-4-DYN-1 complex by interacting with them , thereby promoting the initiation of phagosome maturation ( Figure 8 ) . Notably , whereas AP2 and CHC-1 were found to form complexes with either CED-1-CED-6 or LST-4-DYN-1 , no protein interaction of CED-1 or CED-6 with LST-4 or DYN-1 was detected by either co-immunoprecipitation or GST pull-down assays . Thus clathrin and AP2 likely form two types of complex with factors required for engulfment and phagosome maturation , establishing them as a molecular link between engulfment and phagosome maturation in apoptotic cell clearance mediated by the phagocytic receptor CED-1 ( Figure 8 ) . The recruitment of actin around germ cell corpses mediated by the complex of CED-1 , CED-6 , AP2 and clathrin may resemble the pathway used by mammalian cells to phagocytose pathogens . In mammalian cells , clathrin and some other regulators of CME are found to be essential for invasion of pathogenic bacteria , fungi and large viruses [34] , [46]–[51] . For example , clathrin and dynamin were found to localize to bacterial entry foci during the invasion of Listeria monocytogenes and inactivation of major regulators of CME , such as Grb2 , EPS15 , CIN85 and CD2AP , severely inhibited bacterial internalization [34] . Further studies revealed that upon bacterial infection , the clathrin heavy chain CHC undergoes Src-dependent phosphorylation , which in turn initiates the accumulation of clathrin coats at bacterial adhesion sites . Through interaction of the clathrin light chain CLC with the actin-interacting protein Hip1R , actin is recruited and assembled at bacteria-host adhesion sites , leading to bacterial internalization . Thus the clathrin-coated pits that accumulate at bacterial entry sites serve as platforms for the actin polymerization needed for phagocytosis [51] . Intriguingly , the clathrin adaptor Dab2 , but not AP2 , is critical for clathrin recruitment to L . monocytogenes entry sites [34] . In C . elegans , our findings indicate that clathrin is similarly required for the actin rearrangement needed for phagocytosis of apoptotic cells . Nevertheless , RNAi depletion of hipr-1 and clic-1 , which encode C . elegans homologs of mammalian HipR1 and clathrin light chain , respectively , did not induce a similar level of corpse accumulation to that caused by chc-1 RNAi ( Table S1 ) . We also performed RNAi to deplete several other putative actin-binding proteins predicted by the STRING protein interaction prediction program ( http://string-db . org/ ) but failed to detect an obvious accumulation of germ cell corpses ( data not shown ) . Thus it is possible that multiple factors may function redundantly to mediate the recruitment of actin by the CED-1-CED-6-AP2-clathrin complex . Further studies will be necessary to unveil the underlying mechanism . In addition , unlike clathrin recruitment during bacterial phagocytosis by mammalian cells , phagosomal recruitment of clathrin requires the AP2 complex in C . elegans . The requirement for different adaptors may be attributed to the use of different receptors for engulfment of apoptotic cells and bacteria . Besides , as the sizes of cell corpses in C . elegans are normally ≥1 µm , which is much larger than endocytic CCVs ( <200 nm ) , it also remains to be determined how clathrin is assembled ( i . e , clathrin per se , clathrin-coated vesicles , or clathrin-coated pits ) when it associates with phagosomes . Moreover , the engulfment of cell corpses in C . elegans appears to involve fewer CME regulators compared with mammalian phagocytosis of pathogens . In our unbiased RNAi screen of C . elegans CME regulators , we found that only CHC-1 and AP2 components obviously affected cell corpse engulfment and degradation while LST-4/Snx9/18/33 and DYN-1/dynamin were essential for phagosome maturation; in contrast , RNAi inactivation of several major CME regulators has been shown to inhibit bacterial infection of mammalian cells . Thus , whereas both cell corpse engulfment in C . elegans and pathogen invasion in mammals make use of clathrin for actin rearrangement , other factors may differ owing to the requirement of distinct signaling mechanisms . Remarkably , MEGF10 , the mammalian ortholog of CED-1 , was reported to interact with the μ2 subunit of AP2 in a yeast 2-hybrid screen , and the existence of a protein complex containing MEGF10 and AP2 subunits was further confirmed by a protein purification assay [52] , [53] . More recently , Drosophila Ced-6 was identified as a clathrin-associated sorting protein ( CLASP ) as it binds to clathrin and AP2 via the C-terminal region [54] . Furthermore , the phosphotyrosine-binding domain ( PTB domain ) of Drosophila Ced-6 specifically recognizes a noncanonical sorting signal in the vitellogenin receptor Yolkless . Thus Ced-6 participates in clathrin-mediated yolk uptake in Drosophila egg chambers [54] . In addition , the mammalian homolog of CED-6 , Gulp , can also interact with both clathrin and AP2 [54] , [55] . In our study we found that clathrin and AP2 act in phagocytic receptor-mediated cell corpse removal by forming a protein interaction cascade with CED-1 and CED-6 to regulate the actin rearrangement required for engulfment and with LST-4 and DYN-1 to promote phagosome maturation needed for corpse degradation . Given that the major factors for apoptotic cell engulfment are evolutionarily conserved and the interactions of clathrin and AP2 with CED-6 and/or CED-1 similarly exist in C . elegans , Drosophila and mammals , our discovery that clathrin and AP2 play an essential role in removal of apoptotic cells suggests that the non-classical function of clathrin and its adaptor proteins in phagocytosis is likely conserved across diverse species . The Bristol strain N2 was used as wild type . lst-4 ( tm2423 ) deletion mutants were provided by Dr . Shohei Mitani ( Tokyo Women's Medical University , Tokyo , Japan ) . lst-4 ( qx159 ) mutants were isolated in Dr . Xiaochen Wang's lab ( National Institute of Biological Sciences , Beijing ) . The lst-4 ( qx159 ) mutation is a deletion of 4573 bp including 1962 bp of the lst-4 gene ( from exon 4 to the stop codon ) and 2611 bp downstream of the lst-4 open reading frame ( ORF ) , which also affects the gene Y37A1B . 4 . The flanking sequences of the deletion region are 5′-TGCCCAGAAATTTTATTTTT-3′ and 5′-ATGTTCTTGTTGACCTTATT-3′ . Other mutant alleles used in this study are listed by linkage groups: LG I: ced-1 ( e1735 ) , ced-12 ( n3261 ) . LG III: ced-6 ( n1813 ) , ced-6 ( n2095 ) , chc-1 ( b1025ts ) . LG IV: ced-2 ( n1994 ) , ced-5 ( n1812 ) . LG V: unc-76 ( e911 ) . LG X: dyn-1 ( ky51 ) . The integrated arrays qxIs405 ( Pced-1gfp::act-1 ) , qxIs105 ( Prab-14 mcherry::rab-14 ) , qxIs139 ( derived from qxEx957 ) ( Pced-1dyn-1::gfp ) , qxIs408 ( Pced-1gfp::rab-5 ) , qxIs66 ( Pced-1gfp::rab-7 ) , qxIs257 ( Pced-1 nuc-1::mcherry ) were provided by Dr . Xiaochen Wang . The integrated arrays smIs34 ( Pced-1 ced-1::gfp ) and smIs110 ( Pced-1 ced-1ΔC::gfp ) were provided by Dr . Ding Xue ( University of Colorado , Boulder ) . The integrated array opIs334 ( Pced-1yfp::2×fyve ) was provided by Dr . K . S . Ravichandran ( University of Virginia , Charlottesville , VA ) and Dr . M . O . Hengartner ( University of Zurich , Zurich , Switzerland ) . The integrated array yqIs120 ( Pdpy-23dpy-23::gfp ) was generated by integrating an extrachromosomal transgene harboring the pMG4 ( Pdpy-23dpy-23::gfp ) plasmid kindly provided by Dr . Erik M . Jorgensen ( University of Utah , Salt Lake City , UT ) and the unc-76 rescuing plasmid in an unc-76 ( e911 ) background . Other strains used in this study carrying integrated or extrachromosomal arrays are as follows: yqIs98 ( Pced-1mCherry::chc-1 ) , yqIs99 ( Pced-1apa-2::gfp ) , yqIs100 ( Pced-1mCherry::act-1 ) , yqIs101 ( Pced-1gfp::ced-6 ) , yqIs112 ( Pced-1gfp::chc-1 ) , yqIs114 ( Plst-4lst-4 ( cDNA ) ::gfp ) , yqIs119 ( Plst-4lst-4 ( cDNA ) ::mCherry ) , yqIs121 ( Pced-1gfp::Moesin ) , yqEx368 ( Plst-4::lst-4 ( cDNA ) ::gfp ) , yqEx376 ( Plst-4::lst-4 ( gDNA ) ::gfp ) , yqEx480 ( Pchc-1chc-1::gfp ) , yqEx481 ( Papa-2apa-2::gfp ) . Animals carrying the stably integrated array were outcrossed with the N2 strain 4 times . C . elegans cultures and genetic crosses were performed essentially according to standard procedures [56] . Deletion strains were outcrossed with the N2 strain at least 4 times . C . elegans transformation was carried out essentially as described before [57] . The Pchc-1chc-1::gfp construct was generated by cloning a genomic DNA fragment containing a promoter region of 3 kb and the open reading frame ( ORF ) of the chc-1 gene in frame with GFP into the pPD95 . 77 vector . The Papa-2apa-2::gfp construct was similarly generated by cloning a genomic fragment containing a promoter region of 2 kb and the ORF of apa-2 . Genomic DNA containing the ORF of chc-1 was amplified and inserted into Pced-1mCherry1 or Pced-1gfp1 via the KpnI site to generate the Pced-1mCherry::chc-1 and Pced-1gfp::chc-1 constructs . To generate Pced-1apa-2::gfp , a genomic fragment containing the apa-2 ORF was amplified and inserted into Pced-1gfp3 via the KpnI site . To construct Pced-1gfp::Moesin , the C-terminal of Moesin was amplified from the plasmid pJWZ6 [58] ( provided by Dr . David R . Sherwood , Duke University ) and inserted into Pced-1gfp1 via the KpnI site . To generate Plst-4lst-4 ( cDNA ) ::gfp , a DNA fragment containing a 2 kb promoter region and the first intron followed by the remaining cDNA sequence of the lst-4 isoform c was inserted between the HindIII and KpnI sites of the vector pPD95 . 77 . Plst-4lst-4 ( cDNA ) ::mCherry was derived from Plst-4lst-4 ( cDNA ) ::gfp by replacing gfp with mCherry . To generate Plst-4lst-4 ( gDNA ) ::gfp , a genomic fragment containing a 2 kb promoter region and the lst-4 genomic ORF were amplified and inserted between the XbaI and XmaI sites of the vector pPD95 . 77 . RNAi experiments were performed by using bacterial feeding assays as described previously [59] . In most cases , L4-stage animals were transferred to plates seeded with bacteria expressing either control double-stranded RNA ( dsRNA ) ( L4440 empty vector ) ( Control RNAi ) or dsRNA corresponding to the open reading frames of genes of interest . RNAi of apa-2 with its 3′ UTR was performed by feeding animals with bacteria expressing dsRNA corresponding to the 3′UTR of 516 bp . Germ cell corpses and other phenotypes were observed in adults of the progeny . For RNAi of chc-1 , apb-1 and dyn-1 , which may cause embryonic lethality in the progeny , L3- to L4-stage animals were transferred to plates seeded with bacteria expressing dsRNA of individual genes and phenotypes were observed in adults of the same generation . Cell corpses in synchronized animals were scored under Nomarski optics . To quantify germ cell corpses , cell corpses in the germline meiotic region of one gonad arm in each of at least 15 animals were scored at various adult ages ( 12 , 24 , 36 , 48 and 60 h after the L4 stage ) . The average numbers of germ cell corpses from one gonad arm were calculated for each adult age . Data derived from different genetic backgrounds were compared using unpaired t-tests . For cell corpse analysis of chc-1 ( b1025ts ) mutants , animals were grown to L4 at 20°C and then shifted to 25°C . Germ cell corpses were scored at 12 , 24 , 36 and 48 h after the shift . To quantify the percentage of germ cell corpses labeled by various phagosomal markers , adult animals at 36 h after the L4 molt were mounted on agar pads in M9 buffer ( 1 litre contains: 3 g KH2PO4 , 6 g Na2HPO4 , 5 g NaCl , 1 mM MgSO4 ) with 2 mM levamisole and then examined by fluorescence microscopy . To analyze the labeling of germ cell corpse by phagosomal markers in dyn-1 ( ky51ts ) mutants , animals were grown to L4 stage at 20°C and then shifted to 25°C; cell corpses were analyzed 24 h after the shift . To view germ cell corpses in dyn-1 RNAi-treated animals , L4 larvae were cultured on RNAi plates and germ cell corpses were analyzed 24 h after the L4 molt . Adult animals ( 36 h after the L4 molt ) were dissected in gonad dissection buffer ( 60 mM NaCl , 32 mM KCl , 3 mM Na2HPO4 , 2 mM MgCl2 , 20 mM Hepes , 50 µg/ml penicillin , 50 µg/ml streptomycin , 100 µg/ml neomycin , 10 mM glucose , 33% FCS , and 2 mM CaCl2 ) containing 1 µM LysoSensor Green DND-189 ( Invitrogen ) and examined by fluorescence microscopy . To measure the duration of germ cell corpses , animals were mounted in M9 buffer containing 2 mM levamisole , sealed with beeswax and Vaseline ( 1∶1 ) , and recorded under Nomarski optics at 20°C . The gonadal region was recorded every 1 min at 1 µm/section for 20 Z-sections . Images were captured using a Zeiss Axioimager M1 coupled with an AxioCam monochrome digital camera and Axiovision rel . 4 . 7 software . Animals were constantly examined for viability during recording . L3- or L4-stage animals were fed with bacteria expressing dsRNA of chc-1 or apb-1 . 30 h later , animals were collected for fixation , embedding and sectioning following a procedure essentially as described by Gumienny et al . [60] . Cell corpse photographs were taken with a JEM-1400 Transmission Electron Microscope . Germ cell corpses and the neighboring gonadal sheath cells were analyzed to determine whether individual cell corpses were engulfed . Recombinant GST-CED-1C , GST-CED-6 , GST-DYN-1 , GST-LST-4 , GST-CED-9 proteins were expressed in bacterial BL21 ( DE3 ) cells and purified with glutathione-Sepharose beads ( Amersham ) according to the instructions provided by the supplier . His6-tagged CHC-1C ( amino acids 825–1682 ) , LST-4 , DYN-1 and mCherry were purified with Ni-NTA beads . 35S-labeled APA-2 , APB-1 and DPY-23 were prepared by in vitro translation . Purified GST , GST-CED-9 , GST-CED-1C or GST-CED-6 proteins ( 3 µg of each ) immobilized on glutathione-Sepharose beads was incubated with 35S-labeled APA-2 , APB-1 , DPY-23 or CHC-1CHis6 , LST-4His6 and DYN-1His6 at 4°C for ≥4 h and washed extensively . Bound proteins were resolved on sodium dodecyl sulfate ( SDS ) polyacrylamide gels ( SDS-PAGE ) and visualized by autoradiography or immunoblotting with anti-His6 antibody . CED-1C antibody was generated previously [22] . CED-6 and mCherry antibodies were generated in guinea pigs or rabbits by injecting recombinant proteins . GFP polyclonal antibody ( Catalog # E022200-02 , rabbit ) was purchased from EarthOx , LLC . ( San Francisco , CA , USA ) . GFP monoclonal antibody ( GFP ( B-2 ) :sc-9996 , mouse ) was purchased from Santa Cruz Biotechnology , Inc . Whole cell lysates were prepared from indicated strains and immunoprecipitations were performed essentially as described before [22] using individual antibodies . Precipitated proteins were resolved by SDS-PAGE and subjected to LC-MS analysis or detected with antibodies . Adult animals ( 24 h post L4 molt ) were anesthetized with 0 . 1 mM levamisole in M9 buffer , mounted on 2% agar pads , and maintained at 22°C . Time-lapse images of DYN-1::GFP , LST-4::mCherry and LST-4::GFP were captured every 2 min by using an imaging system consisting of an Axiovert 200M microscope ( Carl Zeiss MicroImaging , Inc . ) equipped with a 100× , 1 . 45 N . A . objective , an EM CCD camera ( Hamamatsu model , C9100-13 ) , and the 488 nm and 561 nm lines of an Argon and Krypton laser attached to a spinning disk confocal scan head ( Yokogawa CSU10 obtained from Solamere Inc . ) .	Phagocytosis of apoptotic cells is an indispensable part of the cell death program . During phagocytosis , the evolutionarily conserved CED-1 family phagocytic receptors recognize cell corpses and transduce engulfment signals to induce the formation and maturation of phagosomes . However , it remains largely unknown how the CED-1 signaling pathway induces the cytoskeletal reorganization required for corpse internalization and initiates phagosome maturation . Interestingly , cell corpse phagocytosis appears to resemble clathrin-mediated endocytosis ( CME ) of cell surface molecules in that both events cause receptor-dependent internalization of extracellular cargoes differing only in size . In CME , the recognition of plasma membrane receptors by adaptor proteins such as the AP2 complex triggers the formation of clathrin-coated vesicles ( CCVs ) with diameters ranging from 10–200 nm . Nevertheless , it is not known whether clathrin and AP2 also play a role in phagocytosis of apoptotic cells that are much larger than CCVs . Here we provide genetic , cell biological , and biochemical experimental findings to demonstrate that clathrin and AP2 act downstream of CED-1 to regulate the actin cytoskeleton rearrangement required for cell corpse internalization and cell corpse degradation . Clathrin and AP2 form two types of complex with factors required for engulfment and phagosome maturation . These findings establish clathrin and AP2 as essential players in phagocytic receptor-mediated apoptotic cell clearance .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "molecular", "cell", "biology", "genetics", "biology", "model", "organisms" ]	2013	Clathrin and AP2 Are Required for Phagocytic Receptor-Mediated Apoptotic Cell Clearance in Caenorhabditis elegans
The Americas have suffered a dramatic epidemic of Zika since May in 2015 , when Zika virus ( ZIKV ) was first detected in Brazil . Mosquitoes belonging to subgenus Stegomyia of Aedes , particularly Aedes aegypti , are considered the primary vectors of ZIKV . However , the rapid spread of the virus across the continent raised several concerns about the transmission dynamics , especially about potential mosquito vectors . The purpose of this work was to assess the vector competence of the house mosquito Culex quinquefasciatus from an epidemic Zika area , Rio de Janeiro , Brazil , for local circulating ZIKV isolates . Culex quinquefasciatus and Ae . aegypti ( positive control of ZIKV infection ) from Rio de Janeiro were orally exposed to two ZIKV strains isolated from human cases from Rio de Janeiro ( Rio-U1 and Rio-S1 ) . Fully engorged mosquitoes were held in incubators at 26 ± 1°C , 12 h:12 h light:dark cycle and 70 ± 10% humidity . For each combination mosquito population—ZIKV strain , 30 specimens were examined for infection , dissemination and transmission rates , at 7 , 14 and 21 days after virus exposure by analyzing body ( thorax plus abdomen ) , head and saliva respectively . Infection rates were minimal to completely absent in all Cx . quinquefasciatus-virus combinations and were significantly high for Ae . aegypti . Moreover , dissemination and transmission were not detected in any Cx . quinquefasciatus mosquitoes whatever the incubation period and the ZIKV isolate . In contrast , Ae . aegypti ensured high viral dissemination and moderate to very high transmission . The southern house mosquito Cx . quinquefasciatus from Rio de Janeiro was not competent to transmit local strains of ZIKV . Thus , there is no experimental evidence that Cx . quinquefasciatus likely plays a role in the ZIKV transmission . Consequently , at least in Rio , mosquito control to reduce ZIKV transmission should remain focused on Ae . aegypti . A Zika virus ( ZIKV ) epidemic has rapidly spread throughout tropical and subtropical zones of the American continent since early 2015 [1] . Brazil was likely the starting point of the Zika pandemic in the Americas [2 , 3] . The Zika virus pandemic has spread to North America too . By July 2016 , 45 American countries or territories have already reported active ZIKV transmission ( http://www . cdc . gov/zika/geo/active-countries . html ) . ZIKV is a positive-sense , single-stranded RNA mosquito-borne-virus of 10 , 807 nucleotides belonging to family Flaviviridae , genus Flavivirus . It is composed of three major lineages: East African , West African , and Asian [4] . ZIKV was first isolated from a sentinel rhesus monkey in the Zika forest in Uganda in 1947 [5] . The second ZIKV isolations were obtained from 20 pools of the forest canopy feeder mosquito Aedes ( Stegomyia ) africanus captured in the same area [6] . Almost 70 years have passed and little is known about natural ZIKV vectors . Aedes mosquitoes are considered the primary vectors of ZIKV in Africa with reported viral isolations from several species , especially from Ae . africanus [1 , 7–10] . ZIKV was also isolated from several other mosquito species belonging to genus Aedes ( subgenera Stegomyia and Diceromyia ) , Mansonia and Culex , and horse flies from the wild in Uganda [8] . More recently , natural infections screened by molecular methods in sylvatic African mosquitoes were again predominantly found in Aedes belonging to subgenus Stegomyia , but also in other species of Aedes , Mansonia , Culex , Anopheles [9 , 10] . Nevertheless , ZIKV transmission in the wild has remained poorly understood . Only two sylvatic species ( Ae . vittatus and Ae . luteocephalus ) proved to be able to transmit ZIKV in laboratory assays [11] . The domestic mosquito Ae . ( Stegomyia ) aegypti was early shown to be competent to experimentally transmit ZIKV [12] . Due to its high anthropophilic and domestic behaviors and virus detection in field caught specimens [13 , 14] , this mosquito has been incriminated as the urban and periurban vector in Africa and Asia [1 , 15] . ZIKV has only recently emerged outside of its natural distribution in Africa and Asia , and has caused a series of epidemics in urban and periurban sites on Pacific islands [16–20] before reaching the Americas , probably in 2013 [21] . The spreading virus belonged to the Asian genotype [21] . Despite multiple efforts , mosquito vectors involved in the ZIKV outbreaks across the Pacific Ocean in 2007–2015 were not identified . Ae . aegypti and other local members of subgenus Stegomyia ( Ae . hensilli and Ae . polynesiensis ) were thought to be potential vectors [16 , 22 , 23] . Ae . ( Stegomyia ) albopictus was found naturally infected with ZIKV in urban sites in Gabon in 2007 [24] and Mexico ( http://www . paho . org/hq/index . php ? option=com_docman&task=doc_view&Itemid=270&gid=34243&lang=en ) . Additionally , Ae . aegypti from Singapore were competent to transmit the African ZIKV genotype in the laboratory [25] . Thereafter , Ae . albopictus has been considered a potential vector of ZIKV throughout its geographical range , concomitantly or not with Ae . aegypti [1 , 24 , 26 , 27] . With the arrival of the ZIKV Asian genotype in the Americas , the global number of suspected and confirmed ZIKV cases reached levels never seen previously [28 , 29] . Besides , the rapid geographical spread , the increased incidence of severe congenital troubles , such as microcephaly , and Guillain-Barré syndrome associated with ZIKV in Brazil led the World Health Organization to declare the ZIKV epidemic a Public Health Emergency of International Concern [1 , 30] . ZIKV proved to have a high potential for geographic expansion in regions wherever Ae . aegypti mosquitoes are present , concomitantly with Dengue viruses 1–4 and Chikungunya virus prone areas of transmission , as it has occurred in Brazil and other American tropical and subtropical countries [29 , http://www . cdc . gov/zika/geo/active-countries . html] . American Ae . aegypti and Ae . albopictus populations showed to be competent to transmit the ZIKV belonging to the circulating genotype , but displayed heterogeneous infection , dissemination and transmission rates in laboratory assays [26] . However , Ae . aegypti and Ae . albopictus populations from Brazil and USA exhibited low transmission efficiency to ZIKV [26] , which appeared inconsistent with the rapid Zika spread throughout the Americas . Two main hypotheses might explain this scenario: ( 1 ) The large number of humans susceptible to ZIKV combined with high densities of anthropophilic Aedes mosquitoes compensate their relatively low vector competence to ZIKV [26] . ( 2 ) Although the recent ZIKV pandemic has occurred only in Stegomyia-infested zones and Ae . aegypti has been suggested to be the main vector , other anthropophilic , domestic and usually abundant mosquitoes such as Culex species could contribute to ZIKV transmission [1 , 31] . For example , Culex species belonging to the Pipiens Assemblage [32] , such as Cx . quinquefasciatus , were likely candidate due their high human-biting frequency and distribution in urban epidemic centers ( http://www . reuters . com/article/us-health-zika-brazil-idUSKCN0W52AW ) . There is no information whether Cx . quinquefasciatus can transmit the virus or the potential role of this mosquito in the ZIKV transmission in nature . We herein comparatively assess the vector competence of Cx . quinquefasciatus and Ae . aegypti populations from Rio de Janeiro for two local ZIKV isolates . Cx . quinquefasciatus populations tested in this study were collected from four districts of Rio de Janeiro: Manguinhos ( MAN , 22°52’20”S 43°14’46”W ) , Triagem ( TRI , 22°53’56”S 43°14’44”W ) Copacabana ( COP , 22°58’8 . 3”S 43°11’21”W ) and Jacarepaguá ( JAC , 22°57’42”S 43°24’11”W ) . For comparison , we used two populations of Ae . aegypti from Rio de Janeiro , Brazil: Urca ( URC , 22°56’45”S 43°09’43”W ) and Paquetá ( PAQ , 22°45’44”S 43°06’26” ) . The mosquitoes were concurrently collected as larvae or eggs using ovitraps from January to March 2016 to initiate laboratory colonies . Each colony was started with at least 200 field-collected individuals from more than five distinct collecting sites and traps . Field collected larvae and eggs were hatched and reared in insectaries ( 26 ± 1°C; 70 ± 10% RH; 12 h:12 h light:dark cycle ) . Larvae were reared in pans ( ~100 larvae/pan measuring 30 x 21 x 6 cm ) containing 1 liter of dechlorinated tap water supplemented with yeast tablets . Adults were kept under the same insectary controlled conditions described above , and supplied with a 10% sucrose solution . All experimental oral infections were performed with mosquitoes of the F1 generation , except for TRI ( laboratory colony ) and PAQ ( F2 ) . Mosquitoes were challenged with two ZIKV strains of the Asian genotype , named Rio-U1 and Rio-S1 , respectively isolated from urine and saliva of two patients in January 2016 , living in distinct districts in Rio de Janeiro [33] . The viral samples were isolated , kept anonymized and provided by Bonaldo et al . [33] , whose the institutional review board at Fundação Oswaldo Cruz has previously approved their study protocol . Viral stocks were obtained after two passages of the isolates onto Vero cells maintained with Earle’s 199 medium supplemented with 5% fetal bovine serum ( FBS ) , under an atmosphere containing 5% CO2 , and incubated at 37°C . Viral titer in supernatants were estimated by plaque-forming unit ( PFU ) assays of serial dilutions on Vero cells maintained at 37°C for 7 days and expressed in PFU/mL . Samples were kept at -80°C until use . The comparison of genomic sequences of ZIKV strains Rio-U1 ( KU926309 ) with Rio-S1 ( KU92630 ) yielded 99 . 6% identity , displaying six amino acid variations in the viral proteins . Phylogenetic analysis showed 99 . 7% identity of Rio-U1 and Rio-S1 strains with ZIKV isolates from Guatemala and other Brazil regions , including a Zika-associated microcephaly case . They all cluster ( bootstrap score = 97% ) within the Asian genotype and share a common ancestor with the ZIKV strain that circulated in French Polynesia in November 2013 [33] . Five to seven day-old females were isolated in feeding boxes and starved for 24 h and 48 h for Aedes and Culex mosquitoes , respectively . All mosquitoes were exposed to the infectious blood-meal containing a final viral titer of 106 PFU/mL which consists of a mixture of two parts of washed rabbit erythrocytes and one part of the viral suspension added with a phagostimulant ( 0 . 5 mM ATP ) . Females were fed through a pig-gut membrane covering the base of glass feeders containing the infectious blood-meal maintained at 37°C . Mosquito feeding was limited to 60 min . Only fully engorged females were incubated at 26°C constant temperature , 70 ± 10% RH and 12 h:12 h light:dark cycle , with daily access to 10% sucrose solution . When available , samples of 30 mosquitoes of each population were examined at 7 , 14 and 21 days after virus exposure , hereinafter abbreviated as “dpi” . Mosquitoes were individually processed as follows: abdomen and thorax ( herein after referred to as body ) were examined to estimate viral infection rate , head for dissemination and saliva for transmission . Each female was handled at a time , by using disposable and disinfected supplies to avoid contamination between individuals and between tissues of the same mosquito as previously described [34] . For the determination of viral infection and dissemination rates , each mosquito body and head were respectively ground in 500 μL and 250 μL of medium supplemented with 4% FBS , and centrifuged at 10 , 000 x g for 5 min at +4°C before titration . Body and head homogenates were serially diluted and inoculated onto monolayers of Vero cells in 96-well plates . After 1 h incubation of homogenates at 37° C , 150 μL of 2 . 4% CMC ( carboxymethyl cellulose ) in Earle’s 199 medium was added per well . After 7 days incubation at 37° C , cells were fixed with 10% formaldehyde , washed , and stained with 0 . 4% crystal violet . Presence of viral particles was assessed by detection of viral plaques . Additionally , body and head homogenates were individually submitted to specific ZIKV RNA detection and quantification through RT-qPCR , using the SuperScript III Platinum one-step RT-qPCR ( Invitrogen ) in QuantStudio 6 Flex Real-Time PCR System ( Applied Biosystems ) . For each reaction , we used 600 nM forward primer ( 5’-CTTGGAGTGCTTGTGATT-3’ , genome position 3451–3468 ) , 600 nM reverse primer ( 5’-CTCCTCCAGTGTTCATTT-3’ , genome position 3637–3620 ) and 800 nM probe ( 5’FAM- AGAAGAGAATGACCACAAAGATCA-3’TAMRA , genome position 3494–3517 ) . The sequences of this primer set were provided by Isabelle Lepark-Goffart ( French National Reference Centre for Arboviruses , IRBA , Marseille , France ) . The reverse transcription was performed at 45° C for 15 min . The qPCR conditions were 95° C for 2 minutes , followed by 40 amplification cycles of 95° C for 15 sec , 58° C for 5 sec and 60° C for 30 sec . For each run , numbers of ZIKV RNA copies were calculated by absolute quantitation using a standard curve , whose construction details are described elsewhere [33] . In order to assess the transmission rate ( TR ) and transmission efficiency ( TE ) , mosquito saliva was collected in individual pipette tips containing 5 μL FBS and processed by PFU assays , as previously described [26] . Accordingly , mosquito saliva was inoculated onto Vero Cell monolayer in 6-well plates incubated 7 days at 37° C , under 3 mL with 2 . 4% CMC in Earle’s 199 medium overlay , and stained as described above . Viral titers of saliva were expressed as PFU/saliva . Infection rate ( IR ) refers to the proportion of mosquitoes with infected body ( abdomen and thorax ) among tested mosquitoes . Disseminated infection rate ( DIR ) corresponds to the proportion of mosquitoes with infected head among tested mosquitoes ( i . e . ; abdomen/thorax positive ) . Transmission efficiency ( TE ) represents the proportion of mosquitoes with infectious saliva among the initial number of mosquitoes tested . Transmission rate ( TR ) represents the proportion of mosquitoes with infectious saliva among mosquitoes with disseminated infection . To compare the viral load , the Wilcoxon signed rank test was adopted to analyze pairwise comparison at 7 , 14 and 21 dpi for each mosquito population and tested virus strain . Significant difference was established when p-values were lower than 0 . 05 . Data analyses were conducted with PRISM 5 . 0 software ( GraphPad Software , San Diego-CA , USA , 2007 ) . This study was approved by the Institutional Ethics Committee on Animal Use ( CEUA-IOC license LW-34/14 ) at the Instituto Oswaldo Cruz . No specific permits were required for performing mosquito collection in the districts in Rio de Janeiro . We comparatively evaluated the susceptibility to infection of Cx . quinquefasciatus and Ae . aegypti from Rio de Janeiro to two ZIKV strains locally isolated . Infection rates ( IR ) were negligible to null in Cx . quinquefasciatus , whereas they remained very high for Ae . aegypti , ( Fig 1A ) . With few exceptions , the IRs were of 100% in the two tested Ae . aegypti populations ( URC and PAQ ) at 14 and 21 dpi , for both virus isolates . In addition , when examining Ae . aegypti from URC , 80% have already been infected by 7 dpi ( Fig 1A ) . In contrast , none of the four Cx . quinquefasciatus populations was likely to become infected except for 1 of 30 TRI Cx . quinquefasciatus challenged with ZIKV Rio-U1 , at 14 dpi ( viral load: 1 , 814 RNA copies/ml; 7 . 0 PFU/ml ) ( Fig 1A ) . ZIKV RNA copies ( 1 , 453 RNA copies/ml ) were detected in 1 of 16 MAN Cx . quinquefasciatus at 14 dpi challenged with the same ZIKV strain . However infective viral particles were not detected in the homogenate of this specimen in repeated PFU assays . Viral load estimated in bodies of Ae . aegypti tended to increase with incubation time ( Fig 2 ) , and the lowest values being detected at 7 dpi ( median: 1 . 1 x 106 RNA copies/ml , mean ± SE: 2 . 3 x 106 ± 2 . 4 x 106 RNA copies/ml ) and the highest at 21 dpi ( median: 1 . 5 x 109 RNA copies/ml , mean ± SE: 1 . 3 x 109 ± 8 . 3 x 108 RNA copies/ml ) . Accordingly , viral load was significantly higher at 21 dpi than at 7 ( p = 0 . 0098 ) and 14 dpi ( p = 0 . 009 ) . Viral loads at 14 dpi in bodies of Ae . aegypti from PAQ [IR: 100% , Fig 1; viral load: 1 . 6 x 108 RNA copies/mL ( median ) ; 2 . 6 x 108 ± 2 . 8 x 108 RNA copies/mL ( mean ± SE ) , Fig 2] were significantly higher than for URC [IR: 90 . 9% , Fig 1 , viral load: 2 . 1 x 107 RNA copies/mL ( median ) ; 2 . 6 x 108 ± 4 . 3 x 108 RNA copies/mL ( mean ± SE ) , Fig 2] when challenged with the same ZIKV isolate ( Rio-U1 ) . Cx . quinquefasciatus did not showed viral dissemination regardless of the incubation period whereas dissemination infection rates ( DIR ) were consistently high ( ~85–97% ) in Ae . aegypti at 14 and 21 dpi irrespective the ZIKV strain ( Fig 1B ) . Accordingly , transmission determined by detecting infective viral particles in mosquito saliva was not observed in any pair of Cx . quinquefasciatus population-ZIKV strain regardless the time point of examination ( Fig 1C ) . In contrast , significantly high transmission rates ( TR: 71 . 6–96 . 5% ) and transmission efficiency ( TE: 60 . 6–93 . 3% ) were observed in local Ae . aegypti ( PAC and URC ) at 14 dpi ( Fig 1C and 1D ) . At 14 dpi , viral load in the head of Ae . aegypti from URC infected with ZIKV Rio-S1 ( Fig 2 ) were significantly higher ( median: 1 . 2 x 107 RNA copies/mL; mean ± SE: 1 . 4 x 107 ± 9 . 5 x 106 RNA copies/mL ) compared to ZIKV Rio U1 ( median: 3 . 6 x 106 RNA copies/mL mean ± SE: 6 . 3 x 106 ± 7 . 8 x 106 RNA copies/mL , Fig 2 ) ( p = 0 . 0003 ) . When challenged with the same ZIKV isolate ( Rio-U1 ) , viral load in heads at 14 dpi was significantly higher in Ae . aegypti from PAQ ( median: 1 . 8 x 107 RNA copies/mL , mean ± SE: 3 . 7 x 107 ± 5 . 0 x 107 RNA copies/mL , Fig 2 ) than URC ( p = 0 . 0018 ) . As expected , DIR was lower ( DIR = 40% ) in Ae . aegypti ( URC ) at 7 dpi , and no transmission was observed at this time point ( Fig 1B–1D ) . TRs and TEs at 14 dpi were higher for PAQ compared to URC Ae . aegypti challenged with the same ZIKV isolate ( Rio-U1 ) ( Fig 1C and 1D ) , although viral load did not differ ( p = 0 . 4203 ) between mosquito populations ( Fig 3 ) . Also , comparisons of viral loads in saliva of URC Ae . aegypti challenged with different ZIKV isolates did not show any difference ( 40 . 3 ± 64 . 5 PFU Rio-S1/saliva versus 34 . 2 ± 69 . 0 PFU Rio-U1/saliva; p = 0 . 3388 ) ( Fig 3 ) . No significant difference was apparent ( p = 0 . 2212 ) in viral load in saliva between 14 and 21 dpi ( Fig 3 ) . The Zika epidemics has affected nearly all American countries with ca . 445 , 000cumulative suspected cases , with 91 , 962 confirmed infections and 9 deaths due to ZIKV as of August 5 , 2016 ( http://ais . paho . org/phip/viz/ed_zika_cases . asp ) . South American countries had nearly 74% of the continental Zika suspected cases , with ca . 5% ( 165 , 932 suspected cases ) from Brazil . The incidence rate in Brazil is 81 . 2/100 , 000 inhabitants Zika suspected cases , with 1 , 749 cases of microcephaly associated to ZIVK infection diagnosed by clinical , epidemiological and/or laboratory criteria as of May 2016 ( http://www . paho . org/hq/index . php ? option=com_content&view=article&id=11599&Itemid=41691 ) . Rio de Janeiro is one of the highest risk areas in Brazil , with an incidence of 278 . 1/100 , 000 suspected Zika cases as of July 2016 ( http://portalsaude . saude . gov . br/images/pdf/2016/julho/15/2016-boletim-epi-n28-dengue-chik-zika-se23 . pdf ) . To face such a severe health crisis , efficient and effective mosquito control strategies are essential . However , it depends on the definition of primary and/or potential local mosquito vectors . Other ZIKV transmission mechanisms besides Ae . aegypti have been observed . For instance , sexual ZIKV transmission between humans has been observed [35] . Natural ZIKV infections detected in several mosquito genera and even in horse flies would suggest that ZIKV could potentially infect a large range of mosquito species and even other hematophagous flies [31 , 33 , 36] . However , there is no evidence regarding the role of other mosquitoes or flies besides Aedes ( Stegomyia ) species in the ZIKV transmission in nature in the Americas . Indeed , there are no data whether other anthropophilic and domestic mosquitoes besides Ae . albopictus , and notably Ae . aegypti can transmit ZIKV . In this work , we demonstrate for the first time , under laboratory conditions , that Cx . quinquefasciatus are not competent to transmit two ZIKV strains circulating in Brazil . Four tested populations were minimally infected with ZIKV and were unable to transmit this virus . In contrast , two Ae . aegypti populations were highly susceptible to ZIKV infection and dissemination , and competent to transmit the same virus strains . This is consistent with Ae . aegypti being more likely to sustain the current ZIKV outbreak in Rio de Janeiro and probably in other tropical American zones . The Zika control program in Brazil , as well as in all epidemic American countries , consists essentially in intensifying and reinforcing the current strategies to control dengue for decades , which focuses in reducing Ae . aegypti density and longevity through eliminating or treating potential larval habitats and insecticide spraying ( http://www . who . int/tdr/publications/documents/dengue-diagnosis . pdf ) . However , the traditional vector control strategies have usually failed to efficiently reduce dengue transmission and spread , even when properly adopted [38] . Several reasons have been identified to explain these failures , among which are insufficient community engagement and management and high insecticide resistance in the target species , the mosquito Ae . aegypti [39–41] . Intensifying Ae . aegypti control activities has also been unsuccessful in stemming the rapid spread of ZIKV [1] . Therefore , new technologies are urgently needed to adequately and better mitigate ZIKV transmission , likely requiring combinations of several approaches . For instance , it has been recently demonstrated that Wolbachia-infected Ae . aegypti from Brazil blocks ZIKV transmission [42] . In addition , local control programs should design specific control strategies against the potential vector Ae . albopictus , since it has been shown to transmit ZIKV in laboratory [25 , 26 , 37] , with ZIKV detections in field-collected specimens [24 , http://www . paho . org/hq/index . php ? option=com_docman&task=doc_view&Itemid=270&gid=34243&lang=en] . The first determination of vector competence to ZIKV in American Ae . aegypti populations was conducted with a viral isolate from New Caledonia , as at the time of that evaluation , no local ZIKV strain were available . Nonetheless , the sequence of NS5 gene of ZIKV from New Caledonia displayed 99 . 4% identity with ZIKV from Brazil [26] . One Brazilian Ae . aegypti population , from Tubiacanga , Rio de Janeiro were challenged with the ZIKV New Caledonia . High susceptibility to infection and moderate dissemination rate , but with low transmission were found , suggesting unexpectedly low competence of local Ae . aegypti for ZIKV [26] . Our newly data with two Ae . aegypti populations from Rio de Janeiro ( URC and PAC ) orally challenged with two locally circulating ZIKV isolates ( Rio-U1 and Rio-S1 ) revealed very high dissemination and moderate to high transmission . Similar results were found when testing the URC mosquito population with two ZIKV strains isolated in 2015 from other Brazilian cities [42] . These differences in vector competence may be explained by the concept that the outcome of transmission depends on the specific pairing of vector and virus genotypes [43] . Similar to other ZIKV strains isolated during the epidemic in Brazil , sequences of virus strains used in the present study clustered with Asian clade , including sequences from New World , Malaysia , Micronesia and Pacific . Thus , the New Caledonian [26] and Brazilian strains are genetically nearly identical . Phylogenetic and molecular clock analyses are consistent with a single introduction of ZIKV from the Pacific area into the Americas , probably more than 12 months before the detection of ZIKV in Brazil [21] . It is possible that some genome evolution not yet identified has rapidly shaped ZIKV to New World Ae . aegypti populations , highlighting the genetic specificity and potential for local adaptation between arboviruses and mosquito vectors previously described for dengue [44] . To evaluate the potential role of a mosquito species to transmit an arbovirus like ZIKV requires examination of multiple components governing vectorial capacity , of which vector competence is simply one . Ecological , epidemiological , environmental and climatic factors influence both vector competence and vectorial capacity . Thus , distinct geographical populations of a mosquito species can greatly diverge in their vector competence when exposed to different virus strains , since the outcome of infection depends on the specific combination of mosquito and virus genotypes [45 , 46] . Thus , our demonstration that Cx . quinquefasciatus from Rio are not able to transmit ZIKV does not completely rule out the possibility that domestic Culex mosquitoes from other origins may exhibit different vector competence . Nevertheless , to now at least , there is no evidence that the southern house mosquito Cx . quinquefasciatus is a potential ZIKV vector . Our study with four Cx . quinquefasciatus populations from Rio challenged with two recently isolated virus strains from the same location where mosquitoes were collected showed that this species is not competent to transmit ZIKV . Similar result was obtained when the closely related species Cx . pipiens from USA was challenged with a ZIKV isolated from Puerto Rico [47] . Moreover , besides being incompetent to transmit ZIKV in the laboratory , neither Cx . quinquefasciatus nor any other species of the Pipiens Assemblage has been found naturally infected in the American ZIKV transmission area [48 , 49] or during the 2007 Zika outbreaks in the South Pacific island of Yap ( Micronesia ) [1 , 16] and in Gabon [24] where thousands of Cx . quinquefasciatus have been screened . Therefore , there is no reason to think that mosquito control efforts against Cx . quinquefasciatus to reduce Zika transmission , at least in Rio de Janeiro , Brazil . Mosquito measures to mitigate ZIKV transmission should remain focused on Ae . aegypti .	The pandemic Zika epidemic has affected nearly all American countries . The etiological agent is a mosquito-borne-virus originated from Africa that spread to Asia and more recently , to the Pacific region and the Americas . We experimentally demonstrated that the common house nightly biting mosquito Culex quinquefasciatus from Rio de Janeiro was not susceptible to locally circulating Zika virus ( ZIKV ) strains . Dissemination was not observed in Cx . quinquefasciatus regardless of the ZIKV isolate used and the incubation period after the ingestion of an infected blood meal . No infectious ZIKV particle was detected in the saliva of the four Cx . quinquefasciatus populations examined until 3 weeks after virus exposure . In contrast , we confirmed that local Aedes aegypti mosquitoes can be infected , disseminate ZIKV at significantly high rates , and assured moderate to very high viral transmission after day 14 of virus exposure . We concluded that Cx . quinquefasciatus is not competent to transmit local ZIKV . Our results support that mosquito control should focus on Ae . aegypti to reduce Zika transmission .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "invertebrates", "medicine", "and", "health", "sciences", "body", "fluids", "pathology", "and", "laboratory", "medicine", "viral", "transmission", "and", "infection", "pathogens", "geographical", "locations", "microbiology", "saliva", "animals", "viral", "vectors", "viruses", "rna", "viruses", "infectious", "disease", "control", "insect", "vectors", "viral", "load", "infectious", "diseases", "south", "america", "medical", "microbiology", "epidemiology", "microbial", "pathogens", "disease", "vectors", "insects", "brazil", "arthropoda", "people", "and", "places", "mosquitoes", "anatomy", "flaviviruses", "virology", "viral", "pathogens", "physiology", "biology", "and", "life", "sciences", "organisms", "zika", "virus" ]	2016	Culex quinquefasciatus from Rio de Janeiro Is Not Competent to Transmit the Local Zika Virus
Paracoccidioidomycosis ( PCM ) is the most prevalent deep mycosis in Latin America and is caused by fungi from the Paracoccidioides genus . Virulence factors are important fungal characteristics that support the development of disease . Aspartyl proteases ( Saps ) are virulence factors in many human fungal pathogens that play an important role in the host invasion process . We report here that immunization with recombinant Sap from Paracoccidioides brasiliensis ( rPbSap ) imparted a protective effect in an experimental PCM model . The rPbSap-immunized mice had decreased fungal loads , and their lung parenchyma were notably preserved . An aspartyl protease inhibitor ( pepstatin A ) significantly decreased pulmonary injury and reduced fungal loads in the lung . Additionally , we observed that pepstatin A enhanced the fungicidal and phagocytic profile of macrophages against P . brasiliensis . Furthermore , PbSAP expression was highly altered by environmental conditions , including thermal stress , dimorphism switching and low pH . Hence , our data suggest that PbSap is an important virulence regulator in P . brasiliensis . The pathogenic fungi Paracoccidioides brasiliensis and Paracoccidioides lutzii are the etiological agents responsible for paracoccidioidomycosis ( PCM ) . PCM is geographically limited to Latin America; Brazil has the highest number of reported cases [1 , 2] . Paracoccidioides is a thermally dimorphic fungus that exists in a mycelial phase at 25°C or ambient temperature and , when grown at 37°C , appears as yeast . P . brasiliensis infection occurs after the inhalation of conidia produced during the mycelial phase , which triggers differentiation into pathogenic yeast cells in the lungs . The development of disease depends both on factors associated with the host immune response and on the characteristics of the infectious agent , especially its virulence . However , few molecules have been effectively characterized as virulence factors in P . brasiliensis [3] . The extracellular proteases of pathogenic fungi perform important functions during infection; for example , some hydrolytic enzymes promote adhesion and tissue invasion by hydrolyzing proteins in host cell membranes [4] . The activity of proteinases and phospholipases is directly related to the establishment of infection [4–6] . Hydrolytic enzymes produced by Candida albicans represent the greatest factors that have been associated with virulence [7] . Aspartyl proteases are a family of proteolytic enzymes that play an important role in host invasion in many pathogenic fungi . C . albicans exploits many virulence factors to infect the host . The most important is a family of ten secreted aspartic proteases ( Saps ) that cleave numerous peptides and proteins , often deregulating the host's biochemical homeostasis [4 , 6] . Saps are also known to be immunogenic and to induce protective host defense in animal models [8 , 9] . The aspartyl protease of Aspergillus fumigatus , aspergilopepsin , is secreted in large quantities during lung infection in an animal model [10] . An aspartyl protease associated with the cell wall was detected in Coccidioides posadasii , and the recombinant protein was reported to be a putative candidate for a new vaccine [11] . In Paracoccidioides lutzii , a secreted aspartyl protease ( PbSap ) with highest identity to the aspartyl protease of C . posadasii ( 88% ) , followed by A . clavatus ( 87% ) and A . terreus ( 87% ) , was identified and characterized . The similarity of PbSap to the Saps of C . albicans ranges from 40–47% [12] . In addition , analysis of the transcriptome of P . lutzii during its transition from mycelium ( infecting form ) to yeast ( pathogenic form ) revealed that the PbSap transcript is up-regulated in the yeast phase of the fungus [13] . Recently , analysis using real-time qPCR during biofilm formation by P . brasiliensis revealed an increase in the expression of aspartyl proteinase genes; these profiles are potentially associated with virulence [14] . Quantitative proteomic analyses using the same P . brasiliensis isolate ( Pb18 ) with different degrees of virulence showed significant differences in the protein content between isolates , and the proteins found to be differentially expressed in the virulent isolates were presented as potential virulence factors [15] . Among the proteins with increased expression in the virulent isolate was the vacuolar A protein , also known as aspartyl protease ( PbSap ) . The expression of this protein was 8-fold higher in virulent Pb18 ( vPb18 ) than in the attenuated Pb18 ( aPb18 ) isolate . This result was validated by real-time PCR . In addition , the expression of PbSap was increased in an aPb18 isolate after two consecutive animal passages [15] . In the present study , we produced and purified recombinant PbSap ( rPbSap ) , which was recognized using serum from PCM patients . Immunization with rPbSap led to a reduction in fungal load in an experimental PCM model . In addition , the expression of PbSap can be modulated during dimorphism or with pH changes . Finally , the effect of pepstatin A , an aspartyl protease inhibitor in C . albicans , was investigated in the PCM model . After exposure to the proteinase inhibitor , fungal loads in the lungs were significantly reduced . These inhibitory effects caused by pepstatin A application reduced the virulence phenotype of experimental PCM , indicating the usefulness of the protease inhibitor as a potential antifungal agent in PCM . P . brasiliensis isolate Pb18 was used in all experiments . Yeast extract-peptone-dextrose modified medium ( mYPD ) [0 . 5% yeast extract , 1% casein peptone , and 0 . 5% glucose , pH 6 . 5] was used to cultivate yeast or mycelium cells , which were cultured at 37°C or 25°C , respectively . To induce protease secretion , P . brasiliensis yeast cells were also cultured in mYPD liquid supplemented with 1% bovine serum albumin ( BSA ) . Male BALB/c mice ( 6 to 8-weeks-old ) were maintained under specific pathogen-free conditions at a temperature of 23–24°C with a light/dark cycle of 12 h and were provided with food and water ad libitum . Animal experimentation was approved by the Ethics Committee on the use of animals at the Federal University of São Paulo ( CEP 1631220814 ) . Animals were handled according to the Brazilian National Council for Animal Experimentation Control ( CONCEA ) guidelines . PbSAP transcripts from P . brasiliensis grown under different experimental conditions were quantified by RT-qPCR . Total RNA was obtained from Pb18 yeast cells using the TRIzol reagent ( Thermo Fisher Scientific , Bremen , GA , USA ) after oxidative , osmotic and thermal stress conditions and during M-Y and Y-M transitions , as described previously [16–18] . Next , cDNA was synthesized with RevertAid Premium Reverse Transcriptase ( Thermo Fisher Scientific , Bremen , GA , USA ) according to the manufacturer's instructions . The cDNA preparations were used in qRT-PCR reactions to measure the expression levels of PbSAP . Reactions were performed using the SYBR Green/ROX qPCR Master Mix 2X ( Thermo Fisher Scientific , Bremen , GA , USA ) according to the manufacturer's instructions . Briefly , a 10-μL total volume was used for each PCR reaction , which consisted of 1× SYBR GreenPCR Master Mix , 250 nmol of the reverse primer , 250 nmol of the forward primer and 2 μL of cDNA . The cycling parameters were 50°C for 10 min , 95°C for 5 min and 40 cycles of 95°C for 30 s and 60°C for 1 min . A non-template control was used to detect any contamination . The quality of the reactions was determined from the dissociation curves . The results obtained were analyzed for baseline and threshold cycle values ( Ct ) using Step OnePlus software ( Thermo Fisher Scientific , Bremen , GA , USA ) . The relative expression ratio ( experimental/control ) was determined based on the 2-ΔΔCT method [19] after normalization to the transcript levels of α-tubulin ( α-TUB ) and 18S ribosomal RNA ( 18S ) . DNA contamination was evaluated via PCR amplification of the GP43 gene ( Accession No . U26160 ) . Negative controls contained neither DNA nor RNA . The nucleotide sequences of the forward and reverse primers were as follows: GP43 5’-TGTCACCCTTTTGCCAGT TG-3’ and 5’-TTCCCAAAACGGCTTCGA-3’; PbSAP , 5’-GATGACTCTGAGGCTACCTTTG-3´ and 5´-ATCGAGATCAACCTCCCAGTA-3’; α-TUB , 5’-CGGCATATGGAAATACATGGC-3’ and 5’ GTCTTGGCCTTGAGAGATGCA 3’; and 18S , 5'-CGGAGAGAGGGAGCCTGAGAA-3' and 5’-GGGATTGGGTAATTTGCGC-3’ . Experiments were performed in biological triplicate . Western blot analysis was performed with sera from hyperimmune mice or patients with PCM . Purified rPbSap was electrophoresed on 12% SDS-PAGE gels and electroblotted to nylon membranes that were blocked with 5% ( w/v ) non-fat dried milk in 1x TBS-T for 1 h at room temperature with shaking . Aspartyl protease was detected using sera from mice containing anti-rPbSap polyclonal antibodies ( produced in this work ) or sera from PCM patients . After incubation with the appropriate HRP-conjugated secondary antibodies anti-mouse IgG ( whole molecule ) ( at 1:2 , 000 dilutions , KPL ) , the blots were developed using the Super Signal system ( Thermo-Pierce–Rockford , IL ) . Image acquisition and densitometry were performed using a chemiluminescence documentation system ( UVITEC , Cambridge , UK ) . Negative controls included preimmune sera from mice or sera from healthy patients . Total protein extracts from yeast were obtained as described [22] and subjected to western blot analysis using the anti-rPbSap polyclonal antibody to detect PbSap in total protein extracts from Pb18 yeast cells . P . brasiliensis yeast cells were cultured in mYPD liquid ranging from pH 3–6 . 5 using BSA as a substrate at 37°C for 7 days with shaking . Total protein extracts were obtained as described previously and electrophoresed on 12% SDS-PAGE gels that were then stained with Coomassie brilliant blue . Aspartyl protease was detected using the anti-rPbSap polyclonal antibody . P . brasiliensis yeast cells were cultured in mYPD liquid using BSA as a substrate at 37°C for 7 days in low pH ( pH 4 . 0 ) . Total protein extracts were obtained and electrophoresed on 12% SDS-PAGE . The 66 kDa and 44 KDa lines were cut into 5 slices and slices from identical gel areas were combined . The in-gel digestion was carried out according to Stenballe and Jensen [23] . Protein digestion was identified by mass spectrometry on a Bruker MALDI-TOF instrument , according to Gyndry et al . , [24] . P . brasiliensis yeast cells were cultured in mYPD until they reached exponential growth , and 1x106 yeast cells were fixed for 30 min at room temperature in ice-cold methanol ( Merck ) and blocked overnight at 4°C in 1x PBS solution containing 3% BSA . The yeast cells were incubated with the anti-rPbSap polyclonal antibody at a 1:100 dilution for 4 h at room temperature , followed by incubation with the conjugate antibody ( Anti-Mouse IgG ( whole molecule ) -FITC antibody , Sigma Aldrich , St . Louis , MO , USA ) at a 1:100 dilution for 2 h at room temperature . The cells were co-stained with 5 mM Calcofluor white to visualize their cellular morphology . Microscopy slides were mounted with a small aliquot of the preparations in antifading Vectashield ( Vector Laboratories , Burlingame , CA , USA ) and sealed . The double labeling was analyzed by confocal microscopy ( Carl Zeiss LSM-510 NLO , Germany ) , and multiple images ( at least 5–6 slices of Z = 0 . 45 μm ) were captured . Image acquisition and comparative analyses of intensity were performed under similar instrument adjustment parameters . BALB/c mice were immunized with 100 μg of purified rPbSap by intraperitoneal injection as previously described . Imject Alum adjuvant ( Thermo Fisher ) was used as an adjuvant . Three groups of mice ( n = 6 animals each ) were used . Two controls were included: non-immunized mice and mice injected with adjuvant without the recombinant protein . After immunization , the mice were challenged i . t . with 1x105 mYPD-grown virulent yeast cells ( vPb18 ) [15] in 40 μL of saline solution . Briefly , the mice were anesthetized , their trachea was exposed and injected with 1x105 viable vPb18 yeast cells . The incisions were sutured with 4–0 silk . After 30 days of infection , the lungs were excised , and the numbers of viable microorganisms in the lungs were determined by enumerating CFUs [25] . Fragments of lungs were fixed in 10% buffered formalin ( Merck , Germany ) for 24 h and subjected to histopathological analysis via hematoxylin and eosin ( H&E ) staining . IFN-γ production was analyzed in lung tissue using quantitative enzyme-linked immunosorbent assay ( ELISA ) . These experiments were performed four times . Male BALB/c mice were infected i . t . with 1x105 viable vPb18 yeast cells as previously described , and after 30 days of infection , the animals were treated 4 times , once a week , with 0 . 06 mg/kg of pepstatin A ( Sigma Aldrich , St . Louis , MO , USA ) in 200 μL of saline solution or with 10 mg/kg of itraconazole ( ITR ) ( Sigma Aldrich , St . Louis , MO , USA ) [26] via intraperitoneal injection . BALB/c mice were divided into three groups ( n = 6 animals per group ) as follows: 1 ) Control ( untreated infected mice ) ; 2 ) Infected mice treated with pepstatin A and 3 ) Infected mice treated with ITR . After 30 days of infection , all animals were euthanized , CFUs in the lungs were determined and histopathological analysis ( H&E staining ) was also performed . In vitro phagocytosis experiments were performed with the THP1 cell line . Phagocytic tests were performed according to Parente-Rocha et al . [27] with minor modification . THP1 cells ( 106 ) were plated in 24-well tissue culture plates ( TPP , Switzerland ) with one coverglass per well in RPMI medium ( Gibco , Gaithersburg , MD , USA ) with 10% fetal bovine serum and then incubated for 24 h at 37°C and 5% CO2 for adherence . After 24 h , adherent THP1 cells were differentiated into macrophages using 50 ng/mL phorbol 12-myristate 13-acetate ( PMA ) ( Sigma Aldrich , St . Louis , MO , USA ) in RPMI medium for 24 h at 37°C and 5% CO2 . After 24 h , the macrophages were stimulated with 2 ng/mL recombinant murine IFN-γ ( BD Biosciences , San Jose , CA , USA ) and reincubated at 37°C and 5% CO2 overnight . Then , 106 yeast cells of P . brasiliensis were treated with pepstatin A at 15 or 30 μM for 1 h at 37°C in RPMI medium . Cells were recovered by centrifugation at 10 , 000 g for 5 min and added to the macrophages at a yeast:macrophage cell ratio of 10:1 , followed by incubation at 37°C and 5% CO2 for 48 h in fresh RPMI medium containing 20 ng/mL IFN-γ . The coverglasses were stained with a 1/20 solution of Giemsa ( Sigma Aldrich , St . Louis , MO , USA ) , and the e phagocytic index was determined according to Popi et al . [28] . To assess the effect of pepstatin A on the antifungal activity of macrophages , P . brasiliensis yeast cells were treated with pepstatin A at 15 or 30 μM for 1 h at 37°C . Cells were recovered by centrifugation at 10 , 000 g for 5 min and added to the macrophages at a yeast:macrophage cell ratio of 10:1 , followed by incubation at 37°C and 5% CO2 for 48 h in fresh RPMI medium containing 20 ng/mL IFN-γ . The numbers of P . brasiliensis yeast cells were determined by enumerating CFUs . After interacting , the macrophages were lysed by the addition of cold sterile water and plated on BHI agar supplemented with 4% fetal bovine serum and 5% spent culture medium of Pb192 . CFUs were enumerated after 7 days of incubation at 37°C . Data are expressed as the means ± SD . Significance was assessed by one-way analysis of variance , with Student’s t-tests used for comparisons . Results with P<0 . 05 were considered statistically significant . The PbSAP coding sequence from the Pb18 strain was 1 , 460 bp long , consistent with the predicted sequence ( UniProt accession number PADG_00634 ) , and it contained an open reading frame of 1 , 200 bp . Tacco and co-workers [12] identified and characterized an aspartyl protease in the Pb01 strain of P . brasiliensis ( currently called Paracoccidioides lutzii ) . This protein has 92% identity to PbSap from the Pb18 strain , and both aspartyl proteases share important attributes that are essential for SAP function . The similarity of PbSAP to the C . immitis and C . posadassi SAPs was 76% , followed by a similarity of 73% to the A . clavatus , A . terreus and Histoplasma capsulatum SAPs . PbSAP showed identities to the C . albicans SAPs 1–10 ranging from 24–30% , with high similarity to SAP8 . Gene expression of PbSAP was assessed during dimorphic switching and oxidative , osmotic and thermal stress . Temperature controls the dimorphic transition of P . brasiliensis , and pathogenicity is intimately linked to dimorphic switching [16] . In the mycelium-to-yeast transition ( M-Y transition ) , the transcript levels of PbSAP were up-regulated during the early stage ( 10 h—31 . 6-fold ) , and they remained elevated after the complete transition ( 120 h– 9 . 5-fold ) ( Fig 1A ) . Conversely , during the yeast-to-mycelium transition ( Y-M transition ) , we noticed down-regulated transcript levels of PbSAP ( Fig 1B ) . The most potent virulence factors are up-regulated during the M-Y transition [29] , thus providing early evidence that PbSAP is a possible virulence regulator . In addition , the PbSAP gene showed an mRNA expression pattern similar to those of the HSP70 , HSP82 , and HSP104 genes in both transitions , as observed by Goldman and co-workers ( 2003 ) [30] . Therefore , we investigated the 5' UTR region of PbSAP for the presence of putative heat-shock elements ( HSE ) . The target genes of the heat-shock transcription factor ( HSF ) contain a cis-acting sequence , the HSE , which consists of multiple inverted repeats of the sequence 5´-nGAAn-3´ . We did not detect obvious HSE ( classical HSE , cHSE ) in the PbSAP promoter region . However , the promoter region of PbSAP contained continuous inverted repeats of pentanucleotide units with sequences that diverged from cHSE ( Fig 1C ) . We detected five putative non-classical HSE ( ncHSE ) motifs ( Fig 1C ) . These putative ncHSE motifs contain different gap lengths between the units ( nGAAn ) that range from 3 to 15 bp . Thus , we investigated whether the PbSAP gene is heat-inducible . The expression of PbSAP was 40-fold greater in response to thermal stress at 42°C for 1 h than in the control yeast at 37°C ( Fig 1D ) . These data suggest that the PbSap protein is heat-inducible , despite its lack of cHSE . We also analyzed PbSAP expression after oxidative and osmotic stress . The transcript levels of PbSAP increased in the early stages ( 2 h , 3-fold ) of oxidative stress ( Fig 1E ) and osmotic ( 3-fold ) stress ( Fig 1F ) . In E . coli , His-tagged PbSap ( 400 amino acids ) was expressed as a major insoluble cytoplasmic protein . A molecular mass of approximately 44 . 65 kDa was calculated for rPbSap , including its vector sequence , and its observed SDS-PAGE mobility was compatible with this value ( Fig 2A ) . Lysis of the bacterial cells was followed by purification of rPbSap using a 6x Histidine-tag/Ni-NTA system ( Fig 2B ) . rPbSap protein was refolding and used to immunize mice . Anti-rPbSap mouse immune sera recognized rPbSap at titers up to 1/2 , 000 when tested with 200 ng of recombinant protein ( Fig 2C ) . No reactions were detected with mice preimmune sera . Using anti-rPbSap antibodies with total protein extracts from Pb18 , we noticed only a single band with a molecular mass of 66 kDa , which corresponds to the glycosylated form of PbSap ( Fig 2D ) . In P . brasiliensis , the PbSap deduced amino acid sequence revealed two N-glycosylation sites predicted at positions 139–142 and 339–342 , the same N-glycosylation sites predicted to aspartyl protease of P . lutzii . We performed an assay using endoglycosidase H to confirm that PbSap of P . brasiliensis is also glycosylated . After treatment of the total protein extract from P . brasiliensis yeast cells with endoglycosidase H a protein with molecular weight of 44 kDa was recognized by polyclonal antibodies anti-rPbSap ( S1 Fig ) . These data support the important information that 66 kDa protein is the glycosylated form of PbSap in P . brasiliensis . According to the Clustal Omega algorithm ( http://www . ebi . ac . uk/Tools/msa/clustalo/ ) , the PbSap sequence has a 13-amino-acid region ( LLAATTTLLGTSSA ) that contains T cell epitope characteristics . In addition , we also evaluated the immunogenicity of PbSap using western blot analysis with PCM patients’ sera . When rPbSap was used in a western blot assay to detect specific antibodies in sera from ten PCM patients , we observed that all the sera recognized rPbSap ( Fig 2E ) . No reactions were detected with sera from healthy individuals ( Fig 2E ) . The recombinant protein rPbSap was not recognized by sera from patients with candidiasis , histoplasmosis and aspergillosis . These data suggest that PbSap can stimulate a specific immune response and act as an antigen of P . brasiliensis . Aspartyl proteases generally act at low pH [4] . To assess protease production , P . brasiliensis yeast cells were grown over a pH range using BSA to induce protease secretion . PbSap expression increased considerably at low pH ( 4 . 0 ) ( Fig 3A ) . Western blot analysis using the polyclonal anti-rPbSap antibody showed that intracellular and secreted PbSap expression was substantially higher at low pH ( Fig 3B ) . Interestingly , under these growth conditions , the Pb18 exhibited slow growth in acidic pH . In Fig 3B , the polyclonal antibodies recognized two bands with distinct size , 66 kDa and 44 kDa . We supposed that polyclonal antibody recognized PbSap in the glycosilation form ( 66 kDa ) and no-glycosilaton form ( 44 kDa ) . Mass spectrometry analysis was performed to confirm this hypothesis . Initially , the bands corresponding to 66 kDa and 44 kDa were excised from the Comassie-stained 1D gel and digested with trypsin . Protein digestion were identified by mass spectrometry by MALDI-TOF-MS/MS . Peptides mass fingerprints obtained in both bands were similar to the prediction fragmentation profile when PbSap was digested using trypsin . An immunofluorescence assay was used to evaluate the immunolocalization of PbSap using P . brasiliensis yeast cells cultured at pH 6 . 5 or 4 . 0 using BSA . We used anti-rPbSap mouse serum in the confocal microscopy experiments and observed its labeling pattern in yeast cells of isolate Pb18 growing at pH 6 . 5 or 4 . 0 . Fig 4 shows that the yeast cells grown at pH 4 . 0 ( upper panel ) and pH 6 . 5 ( middle panel ) had similar labeling profiles , with surface fluorescence and intense intracellular labeling , especially for the cells grown at pH 4 . 0 ( Fig 4 ) . Control images ( lower panel ) suggested that the surface labeling was specific for anti-PbSap because the preimmune antibody did not label the yeast cells ( Fig 4 ) . Collectively , these data suggest that PbSap expression increases at low pH and that acidic pH conditions enhance PbSap activity . Groups of BALB/c mice were intraperitoneally ( i . p . ) immunized with recombinant protein or with adjuvant in three rounds of immunization . The production of a specific antibody response to rPbSap in the immunized mice was confirmed by western blot analysis ( S2 Fig ) . After 3 immunizations over 30 days , the BALB/c mice were inoculated intratracheally ( i . t . ) with yeast cells . The fungal loads in the lungs were evaluated by enumerating the colony forming units ( CFUs ) after 30 days of infection . The lungs of mice that were immunized with rPbSap had significantly fewer CFUs ( 1 , 431 ± 511 CFUs/g of tissue ) than the controls ( Fig 7A ) : the fungal burdens in the lungs of the mice in the adjuvant and unimmunized groups were 6 , 636 ± 228 and 6 , 545 ± 172 CFU/g of tissue , respectively ( Fig 7A ) . Lung tissue from mice immunized with rPbSap were stained with H&E and compared to unimmunized infected tissues ( Fig 7B ) . As expected , the unimmunized infected lungs showed dense cell infiltrates with high numbers of fungal cells disseminated throughout the lung parenchymas . Similar results were observed in mice treated only with adjuvant . For mice immunized with rPbSap , we observed lung parenchymas that were significantly preserved and lacked fungal cells ( Fig 7B ) . IFN-γ production from lung tissue was assessed from controls and BALB/c mice immunized with rPbSap . Our results showed significant increase in IFN-γ production in the lung cells of immunized group compared with controls ( Fig 7C ) . Taken together , our results showed that immunization with rPbSap prevent the aggravation of P . brasiliensis infection . Given the increasing worldwide incidence of fungal infections , characterizing new virulence factors is very important for understanding the pathogenicity of fungi . In Candida spp . , Sap is the most important virulence factor [4] . This protein family is encoded by at least 10 SAP genes ( SAP1–SAP10 ) [4 , 33] , which are differentially expressed during distinct patterns of infection [33 , 34] . In a previous report , we compared the proteomes of virulent and attenuated P . brasiliensis; several proteins that have been described as virulence regulators in other fungi were up-regulated in the virulent Pb18 strain [15] . Vacuolar protease A ( PADG_00634 ) , also named aspartyl protease , was one of the up-regulated proteins in virulent P . brasiliensis . Transcriptome analysis of the M-Y transition in the Pb01 strain revealed that aspartyl protease transcript levels were up-regulated [13] . The same profile was observed in Pb18; transcript levels of PbSAP increased during an early stage of the M-Y transition and remained elevated until it completed . In contrast , during the Y-M transition , we observed a decrease in the transcript levels of PbSAP . Temperature changes represent a stressor , and in Pb18 , the yeast cells respond to this stress by increasing their transcript levels of PbSAP . Dimorphic switching appears to be intimately linked to pathogenicity . Previous studies have reported the up-regulation of genes involved in diverse cellular pathways during the M-Y transition in P . brasiliensis , including genes encoding putative virulence factors as heat shock proteins HSP70 , HSP82 and HSP104 [3 , 30] . These heat shock have similar patterns of expression when compared with the transcript levels of PbSAP in the dimorphic transition . In addition , we observed several putative ncHSE motifs in the PbSAP promoter . Conventional HSE motifs are typically composed of three to six continuous and/or inverted pentameric units of nGAAn [35]; in ncHSE , the gaps between the repetitive units can vary [36] . Discontinuous HSEs , which are ncHSEs , are found in approximately half of the yeast heat-shock factor ( yHSF ) target genes [18 , 36–38] . Thus , these data suggest that PbSap could participate in the temperature change response of P . brasiliensis and consequently contribute to its adaptation to the host environment . To assess the PbSap antigenicity profile , we produced it as a 44-kDa recombinant protein and incubated it with sera from patients diagnosed with PCM , which efficiently recognized rPbSap in an immunoblotting assay . On the other hand , the rPbSap was not recognized by sera from patients with candidiasis , histoplasmosis and aspergillosis . Serological analysis is the main diagnostic indicator of PCM [39] , and the characterization of new antigens can improve the serodiagnosis of PCM . During infection , P . brasiliensis antigens activate B lymphocyte cells , which produce immunoglobulins ( Igs ) that play a role in host defense against a variety of pathogens . High concentrations of Igs have been found in the bodily fluids of PCM patients [40] . Anti-Sap antibodies were observed in sera from patients with candidemia , indicating the presence of Sap antigens during systemic infection [41] . However , further studies are necessary to confirm the potential of PbSap for PCM diagnosis . In this study , we report the efficacy of rPbSap immunization . The data obtained showed that rPbSap immunization decreased fungal burden in an experimental PCM model . Saps from C . albicans are known to be immunogenic and to strongly induce a host protective response in animal models [8] . The identification of P . brasiliensis antigens and the evaluation of their role in parasite-host interactions are fundamental for the study of immunization procedures , which may lead to the prevention of infection . Some immunization experiments using P . brasiliensis recombinant proteins have already been performed . Morais and co-workers [42] conducted immunization procedures using rPb27 and showed that the rPb27-immunized group had lower levels of lung fibrosis , demonstrating the protective effect of rPb27 . Assis-Marques and co-workers [43] conducted similar studies using rGp43 and observed that immunized animals had less frequent , more compact granulomas with fewer fungal cells as well as fewer fungal cells in the lungs and spleen than non-infected animals . In addition , the organs of immunized animals showed high levels of interleukin ( IL ) -12 and interferon ( IFN ) -γ , suggesting a protective T helper 1 ( Th1 ) response . Muñoz and co-workers [44] demonstrated that immunization with P10 , a peptide from the Gp43 antigen , promotes a specific immune response even in immunocompromised BALB/c mice that is able to confer protection against P . brasiliensis . The protection conferred by immunization with rPbSap suggests that the antigen appears to be a promising candidate against P . brasiliensis infection . However , additional characterization can contribute to better evaluate and to compare PbSap protection with other antigens such as , rPb27 , rGp43 and P10 . Antisense RNA technology and Agrobacterium tumefaciens-mediated transformation have confirmed the importance of rPb27 and rGp43 in P . brasiliensis virulence [45 , 46] . To understand the role of PbSap in experimental PCM we inhibited enzymatic activity using a pharmacological inhibitor ( pepstatin A , an inhibitor of aspartyl protease ) . Our results suggest that PbSap is a potential virulence regulator , considering that inhibition of PbSap using pepstatin A significantly reduced experimental PCM infection . Pepstatin A has also shown to be a potent prophylactic agent for the prevention of lethal C . albicans infections [6] . HIV protease inhibitors that are used in the treatment of HIV disease efficiently reduce candidiasis by inhibiting Sap activity [47–50] . Indinavir affects the virulence of Cryptococcus neoformans [51] by reducing its protease activity and capsule production [52] . Saquinavir or ritonavir synergistically interacts with itraconazole against H . capsulatum [53] . The inhibition of PbSap represents a promising strategy for the design of new drugs , and a combined therapy using anti-fungal drugs and Sap inhibitors could increase the efficiency of treatment for P . brasiliensis infection . PCM is considered a classic granulomatous disease and macrophages play an important role in defense against the disease [3] . P . brasiliensis is a facultative intracellular pathogen that can survive inside macrophages , and this behavior is likely an important factor in its pathogenicity [27 , 54] . Using a proteomics approach , several proteins that were up-regulated in P . brasiliensis yeast cells during their interaction with macrophages have been described as virulence factors [27] . The result of this interaction is the phagocytosis of P . brasiliensis by macrophages [3] , which is followed by fusion of the microorganism-containing phagosomes with cellular lysosomes to form phagolysosomes . The low pH ( 4 . 7–4 . 8 ) inside the phagolysosomes supports the activity of the host’s acid lysosomal enzymes and is an essential host strategy for killing most pathogens [54 , 55] . However , the low pH is also optimal for the enzymatic activity of aspartyl proteinase [4 , 31] . After their ingestion by phagocytic cells , C . albicans and C . tropicalis were shown to express Sap antigens [56] , suggesting a role for Sap in fungal adaptation to the intracellular environment . Our data demonstrated that at low pH , the expression of PbSap increased , while only low levels of expression were observed above pH 6 . 5 . In mammalian hosts , few environments have a low pH . The expression of PbSap is possibly enhanced by phagosome–lysosome fusion due to the pH change that occurs . As reported previously , the expression of Sap4 , Sap5 and Sap6 increased in C . albicans during its interaction with macrophages [57] . Phagosome-lysosome fusion intensifies the activity of potential pathogenic factors , which benefits microbial intracellular survival , as observed for Trypanosoma cruzi and Mycobacterium tuberculosis [58] . In C . neoformans the major aspartyl peptidase 1 ( May1 ) showed relevant importance for survival in acidic host environments such as macrophage phagolysosomes . In a mouse model may1Δ strains displayed attenuated profile [59] . Collectively , our results suggest that PbSap plays a role in the adaptation of P . brasiliensis to the defense systems of its hosts . While it is not requisite for fungal survival , PbSap is an essential element for fungal infection , acting as a potential virulence regulator . In addition , the inhibition of PbSap ameliorates P . brasiliensis infection and its progression , demonstrating that this protein can be used as a molecular target for the development of new strategies and therapies for PCM and other fungi . Additional studies to evaluate the immune response and long-term infection in immunized mice can contribute to better understand the role of PbSap in the immune response and its contribution to fungal-host interaction . In addition , studies using rPbSap for the serodiagnosis of PCM could also help to improve and standardize immunological diagnoses as well as patient follow-up . This approach is under investigation in our laboratory .	Paracoccidioidomycosis is a systemic fungal disease occurring in Latin America and more prevalent in South America . The disease is caused by the dimorphic fungus Paracoccidioides spp . , its pathogenesis is multifactorial and few virulence factors have been recognized in this fungus . Some virulence factors , like the secreted aspartic proteases ( Saps ) , have been described playing an important role in adhesion , invasion and tissue damage in many pathogenic microorganisms . Thus , the inhibition of one of the many infection stages may contribute to the containment of the pathogen and thus should help in the treatment of disease . We have here produced an aspartyl protease recombinant of Paracoccidioides ( rPbSap ) and shown that rPbSap-immunized mice decreased disease progression . Besides that , when infected mice were treated with an aspartyl protease inhibitor we also observed a significant reduction of fungal infection . In addition , PbSap expression was modulated by different stress conditions , especially in low pH environment . These findings provide insights for the design of new antifungal therapies for this important fungal disease .	[ "Abstract", "Introduction", "Materials", "and", "methods", "Results", "Discussion" ]	[ "blood", "cells", "medicine", "and", "health", "sciences", "yeast", "infections", "immune", "cells", "pathology", "and", "laboratory", "medicine", "enzymes", "pathogens", "immunology", "microbiology", "enzymology", "fungi", "experimental", "organism", "systems", "paracoccidioides", "brasiliensis", "fungal", "diseases", "fungal", "pathogens", "research", "and", "analysis", "methods", "infectious", "diseases", "mycology", "white", "blood", "cells", "animal", "studies", "proteins", "medical", "microbiology", "animal", "cells", "microbial", "pathogens", "yeast", "biochemistry", "candida", "macrophages", "eukaryota", "cell", "biology", "virulence", "factors", "biology", "and", "life", "sciences", "proteases", "yeast", "and", "fungal", "models", "cellular", "types", "organisms", "candida", "albicans" ]	2018	Secreted aspartyl proteinase (PbSap) contributes to the virulence of Paracoccidioides brasiliensis infection
The risk of severe adverse events following treatment of onchocerciasis with ivermectin in areas co-endemic with loiasis currently compromises the development of control programmes and the treatment of co-infected individuals . We therefore assessed whether doxycycline treatment could be used without subsequent ivermectin administration to effectively deliver sustained effects on Onchocerca volvulus microfilaridermia and adult viability . Furthermore we assessed the safety of doxycycline treatment prior to ivermectin administration in a subset of onchocerciasis individuals co-infected with low to moderate intensities of Loa loa microfilaraemia . A double-blind , randomized , field trial was conducted of 6 weeks of doxycycline ( 200 mg/day ) alone , doxycycline in combination with ivermectin ( 150 µg/kg ) at +4 months or placebo matching doxycycline + ivermectin at +4 months in 150 individuals infected with Onchocerca volvulus . A further 22 individuals infected with O . volvulus and low to moderate intensities of Loa loa infection were administered with a course of 6 weeks doxycycline with ivermectin at +4 months . Treatment efficacy was determined at 4 , 12 and 21 months after the start of doxycycline treatment together with the frequency and severity of adverse events . One hundred and four ( 60 . 5% ) participants completed all treatment allocations and follow up assessments over the 21-month trial period . At 12 months , doxycycline/ivermectin treated individuals had lower levels of microfilaridermia and higher frequency of amicrofilaridermia compared with ivermectin or doxycycline only groups . At 21 months , microfilaridermia in doxycycline/ivermectin and doxycycline only groups was significantly reduced compared to the ivermectin only group . 89% of the doxycycline/ivermectin group and 67% of the doxycycline only group were amicrofilaridermic , compared with 21% in the ivermectin only group . O . volvulus from doxycycline groups were depleted of Wolbachia and all embryonic stages in utero . Notably , the viability of female adult worms was significantly reduced in doxycycline treated groups and the macrofilaricidal and sterilising activity was unaffected by the addition of ivermectin . Treatment with doxycycline was well tolerated and the incidence of adverse event to doxycycline or ivermectin did not significantly deviate between treatment groups . A six-week course of doxycycline delivers macrofilaricidal and sterilizing activities , which is not dependent upon co-administration of ivermectin . Doxycycline is well tolerated in patients co-infected with moderate intensities of L . loa microfilariae . Therefore , further trials are warranted to assess the safety and efficacy of doxycycline-based interventions to treat onchocerciasis in individuals at risk of serious adverse reactions to standard treatments due to the co-occurrence of high intensities of L . loa parasitaemias . The development of an anti-wolbachial treatment regime compatible with MDA control programmes could offer an alternative to the control of onchocerciasis in areas of co-endemicity with loiasis and at risk of severe adverse reactions to ivermectin . Controlled-Trials . com ISRCTN48118452 Onchocerciasis ( also known as River Blindness ) is a chronic disease induced by the filarial nematode Onchocerca volvulus . An estimated 37 million individuals are infected worldwide with 90 million at risk of infection , mainly in Sub-Saharan Africa . Adult worm infections establish within subcutaneous nodules ( onchocercomas ) and produce microfilariae ( mf ) , which parasitize skin and eye tissues . Mf are the transmissive stage for black fly vectors and are also responsible for the major disease pathologies of onchocerciasis , including intense troublesome itching , dermatitis , atrophy , visual impairment and blindness . Currently , the only drug available to treat onchocerciasis is ivermectin ( MectizanTM , Merck ) . Ivermectin is generally a safe and effective microfilaricide and has been used successfully in community-directed treatment programs aimed at both reducing the burden of disease and controlling transmission since 1987 [1] , [2] . Ivermectin has some macrofilaricidal activity against female adult worms after 6 years of exposure [3] , or when given repeatedly at three-monthly intervals [4] , [5] . Higher doses of ivermectin do not improve on this activity and such regimens are contraindicated due to the occurrence of visual problems [6] . Another anti-filarial drug , diethylcarbamazine ( DEC ) , is also contraindicated due to the incidence of treatment-associated blindness and the frequent development of potentially life threatening adverse reactions , known as Mazzotti Reactions [7] , [8] . There are three major limitations of a sole reliance on ivermectin for onchocerciasis control . Firstly , its use in areas co-endemic with Loa loa , a tissue dwelling filariae that gives rise to blood circulating mf and found principally in forested regions in Africa . Reports of severe adverse reactions ( SAE ) , including encephalopathy , coma and death , in the Central Africa region following mass distribution of ivermectin have introduced serious concerns and disruptions to onchocerciasis control programs [8] . Although the mechanism of ivermectin-associated SAE has not been fully elucidated , L . loa mf have been detected in the cerebral spinal fluid of patients suffering severe adverse reactions , indicating that mf can cross the blood brain barrier . The intensity of L . loa mf in the blood has been determined to be a major risk factor in the development of SAE [8] . Secondly , because ivermectin principally targets the mf stage , continuous delivery of annual treatment is required for at least 15–17 years to interrupt transmission as demonstrated in some endemic areas of Africa [9] . In other endemic areas of Africa this strategy is unlikely to lead to the interruption of transmission due in part to civil strife and conflict , insufficient health infrastructure and political commitment to funding for sustained control programmes , which together compromise the eradicability of onchocerciasis in Africa [10] . The third limitation is that such a long term , community-based strategy based on a single drug intervention is potentially vulnerable to the development of drug resistance . Recent reports from Ghana show evidence of sub-optimal efficacy of ivermectin in communities receiving 6–18 rounds of treatment [11] , [12] , [13] . Parasites from these communities show genetic changes associated with resistance to ivermectin in other nematodes and increase the concern of resistance to ivermectin developing in onchocerciasis [14] , [15] . Considering the absence of any safe alternative to ivermectin , there is an urgent need to identify novel anti-filarial drugs . An ideal alternative would exhibit curative ( macrofilaricidal ) or permanent sterility and have minimal treatment-associated side effects and be safe to use in patients co-infected with L . loa . A promising approach is to use antibiotics such as doxycycline to target not the filariae itself , but the Wolbachia endosymbiotic bacterium that is found in all life stages of O . volvulus . Pilot , open-labelled trials in onchocerciasis have demonstrated that 6-week courses of 100 mg/day oral doxycycline cause >90% reductions in Wolbachia levels from filarial tissues followed by an almost complete and sustained absence ( 12–18 months ) of mf from the skin [16] , [17] , [18] . Deleterious effects on embryogenesis were determined by histological assessment of extirpated nodules . However , a clear adulticidal effect of doxycycline could not be determined in onchocerciasis patients after 18 months [18] . More recent placebo controlled trials with extended follow-up analysis have detected significant macrofilaricidal activity 21–27 months after receiving 4 to 6 week courses of 200 mg doxycycline [19] or a 5-week course of 100 mg doxycycline [20] . As L . loa is free of Wolbachia symbionts [21] , [22] antibiotic therapy is not an option for their treatment . This , however , could be an advantage for the treatment of concomitant onchocerciasis or lymphatic filariasis with antibiotics in individuals co-infected with L . loa without the risk of microfilaricidal induced SAE . We therefore carried out a randomized , double-blind , placebo controlled field trial to assess the efficacy and safety of a six-week course of 200 mg/day oral doxycycline with or without ivermectin for the treatment of onchocerciasis alone and in patients co-infected with L . loa . A proportion of the onchocerciasis patients were also co-infected with Mansonella perstans . Our primary objectives were to measure changes in a ) O . volvulus mf levels in the skin , b ) Wolbachia levels in adult O . volvulus tissues , c ) embryogenesis within female O . volvulus uteri and d ) adult motility and viability . Secondary objectives were to measure the incidence and severity of adverse events and changes in L . loa and M . perstans microfilaraemia . The experimental protocol for this study was designed in accordance with the general ethical principles outlined in the Declaration of Helsinki . The trial was approved by ethics committees of the Tropical Medicine Research Station , Kumba and the Research Ethics Committee of The Liverpool School of Tropical Medicine . Written informed consent was obtained from all participants , with the exception of those who were illiterate , where a literate witness signed on behalf of the participant and the participant added a thumbprint . The trial is registered with the current controlled trials registry , no: ISRCTN48118452 . The trial was community based and was undertaken in 6 satellite villages ( Bifang , Ebendi , Eka , Ngalla , Dinku and Olurunti ) of the market town of Widikum , in the North West Province of Cameroon ( between latitude 5° N 43–5° N 54 and between longitude 9° E 41–9° E 44 ) starting on 1st July 2003 and finishing on 31st March 2005 . The area is hyperendemic for onchocerciasis with a community prevalence of Loa loa ranging from 3 . 36%–14 . 29% [23] . Nodulectomy surgery was performed at Batibo Hospital under the direction of The District Health Officer . Individuals eligible for participation were adults of both sexes aged 15–60 , with a minimum body weight of >/ = 40 Kg , in good health without any clinical condition requiring chronic medication . Mf counts were assessed microscopically following skin biopsy using a Walser skin punch . Hepatic and renal function and pregnancy were assessed by dipstick chemistry . Exclusion criteria encompassed an O . volvulus microfilarial load <10 mf/mg , a L . loa microfilarial load >8000 mf/ml , hepatic and renal enzymes outside of normal ranges ( AST [0–40 IU/l , ALT [0–45 IU/l] and creatinine [3–126 µmol/l] ) pregnancy , lactation , intolerance to ivermectin , alcohol or drug abuse or anti-filarial therapy in the last 12 months . Participants received 2×100 mg capsules of doxycycline ( VibramycinTM , Pfizer ) or matching placebo supplied by the manufacturer , daily , for a total of 42 days following a breakfast meal . Four months after the start of treatment , participants received an oral dose of 150 µg/kg ivermectin ( MectizanTM , Merck & Co . Inc . ) or dummy pill ( non-matching lactose tablet ) . Treatment was delivered by trained community distributors who gave the drug/placebo to the participants and witnessed them swallowing the tablets . Outcome measurements encompassed: a ) the number of mf present in skin snip biopsies taken at baseline , 4 , 12 and 21 months after the start of treatment , b ) the quantity of a Wolbachia single copy gene ( Wolbachia surface protein; wsp ) within extirpated nodule tissue 21 months after the start of treatment and the immunohistochemical staining of Wolbachia within adult O . volvulus tissues 21 months after the start of treatment , c ) the histological assessment of the frequency of embryonic stages present within female O . volvulus uteri 21 months after the start of treatment , d ) the detection of adult O . volvulus motility within onchocercomas using ultrasonography e ) histological assessment of parasite viability , f ) the clinical monitoring and assessment of adverse reactions during primary drug allocation ( doxycycline ) or secondary drug allocation ( ivermectin ) in patients singly infected with O . volvulus or co-infected with L . loa and g ) the number of L . loa and M . perstans mf present in 50 µl thick blood smears taken at baseline , 4 , 12 and 21 months after the start of treatment . Two skin snip samples of approximately 1 mg were taken from the rear of the leg using a Walser skin punch . Skin snips were placed in 200 µl saline containing 2 mM EDTA and incubated overnight at room temperature . The following day the saline samples were mounted on glass slides , total numbers of released mf were counted using a compound microscope and the mean number of mf/snip was derived . Finger prick blood samples ( 50 µl ) were taken at baseline and 4 , 12 and 21 month after the start of treatment . Thick blood smears were made to count numbers of L . loa and M . perstans mf by microscopy . Onchocercomas were surgically removed under local anaesthesia from operable sites . Onchocercomas were halved and fixed in either 80% ethanol for histology or in stabilisation buffer ( RNAlater , Qiagen ) for DNA analysis . Genomic DNA was extracted and the quantity of Wolbachia wsp was determined by quantitative PCR as previously described [24] . Ethanol fixed tissue was embedded in paraffin wax blocks and several sections stained with haematoxylin and eosin . The embryonic status of the adult females was determined by counting the number of adult female cross sections which contained ova , morulae , curled microfilariae and straight microfilariae . Immunohistochemistry for the detection of Wolbachia used a polyclonal rabbit antisera raised to B . malayi Wolbachia surface protein ( WSP ) at a dilution of 1∶2000 [24] and for viability with a rabbit antisera to lysosomal aspartic protease of O . volvulus ( APR ) at a dilution of 1∶1000 [25] . The viability of parasites was assessed using criteria as previously described [19] . In brief , the criteria for dead worms included evidence of calcification without cuticle or nearly complete adsorbed , loss of body wall integrity , loss of nuclei and absence of APR staining . Ultrasonography ( USG ) was used to examine palpable onchocercomas as previously described [26] . Ultrasound examinations were performed 21 months after treatment start using a portable ultrasound system ( Sonosite 180 Plus® , Sonosite Washington , USA ) and a linear transducer ( L38mm ) with frequencies of 7 . 5–10 MHz . Patients were examined in a supine position in order to avoid artefacts due to movements . Each onchocercoma was scanned in longitudinal and transverse sections to detect motile adult filariae . The transducer was positioned at the largest diameter or at the largest echo-free area in case of cystic nodules . Imaging was carried out in panorama mode to provide optimal information . The detection of all onchocercomata was recorded with a camcorder ( SONY® PAL handycam , SONY Corp , Japan ) on video tapes . Onchocercomata were identified by a capsule of connective tissue , lateral shadowing , partly echo-free areas as sign for necrotic proceedings and acoustic shadowing , reflecting moving and static fragments of the adult worms [26] . Clinical monitoring of adverse reactions was undertaken by community health officers throughout the 6-week period of doxycycline treatment . Patients were asked by questionnaire for any side effects of the drugs as per protocol . Adverse events were assigned scores; 0 = no abnormality , 1 = mild , 2 = moderate and 3 = severe . Individuals were asked to report any signs and symptoms that were not experienced prior to drug administration . All symptoms were documented in patients' treatment cards and medication or hospitalisation was provided where necessary . For the assessment of adverse reactions to ivermectin , a scoring system previously described was utilized [27] . Incidence and severity of clinical symptoms consistent with ivermectin-associated adverse reactions ( such as increased body temperature and type and extent of skin rash ) were recorded immediately preceding ivermectin treatment and forty-eight hours following administration . Based on data from a previous study [17] , a reduction in the ( geometric ) mean mf load/mg skin of 50% at 6 months was considered to be clinically significant . Assuming a baseline ( geometric ) mean mf load of 9 . 40 and a standard deviation similar to that in the previous study , 25 patients would be sufficient to detect such a reduction with 80% power . To allow for up to a 15% drop-out rate over the study period , 30 patients were recruited into each treatment group . Randomization for onchocerciasis was block stratified based on baseline microfilaridermia . All L . loa co-infected patients were assigned to doxycycline + ivermectin treatment . Treatment allocation was assigned by randomized ID code ( by JDT and MJT ) and the course of treatment sealed in an envelope for allocation by the field team and district field officers . All study personnel and participants were blinded to the doxycycline and placebo treatment assignment for the duration of the study . Placebo tablets for assessment of ivermectin adverse events were not supplied in time for treatment allocation and so unmarked lactose tablets of similar size , shape and colour were used as an alternative . The ivermectin and dummy pills were assigned to individuals in sealed unmarked envelopes before being handed over to district health officers for drug delivery and these together with individuals responsible for the clinical assessment of adverse events were not involved in any subsequent analysis . Due to the lack of significant differences between groups in the severity or incidence of adverse reaction , cytokine analysis was not performed . The drop-out rates in the treatment groups were compared using Kaplan-Meier ( survival curve ) analyses . The age distributions and sex ratios of the groups were compared using one-way ANOVA and the Fisher exact test respectively . O . volvulus microfilaridermia and wsp copy number measurements were significantly positively skewed ( even after log transformation ) as assessed by the Kolmogorov–Smirnov test of Normality , so were evaluated using non-parametric analyses; changes in mf counts with time from baseline were assessed using Wilcoxon signed rank tests and differences in mf counts between the treatment groups were analysed using Mann-Whitney U tests . As group sizes were small , frequency of amicrofilaridermia , mf in onchocercomatous tissue , embryonic stages within uteri and adult worm movement were compared across the treatment groups using Fisher exact tests . All analyses were performed using the SPSS v11 , Stata8 and GraphPad Prism software packages . Figure 1 illustrates the trial profile . One hundred and fifty onchocerciasis patients fulfilling all entry criteria , were enrolled into the trial and randomized into one of the three treatment arms . Twenty-two additional onchocerciasis patients fulfilling all entry criteria were identified as positive for L . loa infection ( below the safety threshold of 8000 mf/ml ) and assigned into doxycycline + ivermectin regimen . 102/112 ( 91% ) of the individuals who started treatment completed the full course of doxycycline and 59/60 ( 98% ) placebo . For ivermectin 107/123 ( 87% ) and dummy pill 25/27 ( 93% ) individuals completed treatment . Additional drop-outs at the 4 , 12 and 21 month follow-up assessments were 11 , 21 and 7 , respectively . Thus , in total , 104 out of 172 enrolled patients ( 62% ) completed all treatment allocations and all subsequent follow-ups and were included for outcome analysis . There were no significant differences in the drop-out patterns over the follow up period between the three treatment groups ( log-rank test of survival p = 0 . 718 ) . The baseline characteristics of these patients are reported in Table 1 . Age and gender did not significantly differ between the three treatment groups . For the assessment of adverse events , patients co-infected with L . loa ( treated with doxycycline and ivermectin ) have been analysed as a distinct group in order to evaluate whether co-infection is associated with the occurrence of such events . For all other outcome analyses , these patients have been combined with O . volvulus single infected patients receiving doxycycline and ivermectin . Table 2 summarizes the recorded adverse events during the 6-week primary drug allocation of doxycycline or matching placebo and 48 hours following the secondary drug allocation of ivermectin or dummy pill allocation . Adverse events were recorded in 17 patients during the 6-week period of doxycycline or matching placebo allocation . The incidence of adverse event did not significantly deviate between doxycycline- or placebo-assigned patients or between O . volvulus single infected and O . volvulus + L . loa co-infected patients assigned doxycycline . Symptoms were mild and included itching , fever , headache , body pains and vertigo . One patient administered doxycycline developed fever and headache , which led to a temporary interruption of the treatment for 5 days . Anti-malarial drugs were given for three days and doxycycline treatment resumed after recovery . There was no evidence or complaint of symptoms consistent with doxycycline-associated photosensitivity . It was not necessary to discontinue primary drug allocation in any instance . Forty-eight hours following ivermectin or dummy pill allocation , symptoms consistent with adverse reactions were observed in 16 patients ( 12 . 1% of all patients present ) . The majority of the adverse reactions were graded as mild ( 14 patients ) with two patients in the ivermectin only group experiencing moderate adverse reactions . The frequency of either mild or moderate adverse reaction did not significantly differ between patients treated with ivermectin following doxycycline intervention and patients treated with ivermectin following placebo , matching doxycycline . Furthermore , the frequency of adverse reaction did not differ between patients treated with ivermectin and patients treated with a dummy pill . Figure 2 illustrates the changes in O . volvulus mf levels in skin from baseline . The analysis of these changes is summarized in Table 3 . Baseline , O . volvulus microfilaridermia did not significantly differ between treatment groups in the participants who completed the trial . At 4 months post-doxycycline allocation and immediately preceding ivermectin allocation , reductions in microfilaridermia had occurred in all treatment groups compared with baseline; these were statistically significant both for the doxycycline and placebo groups assigned for ivermectin allocation but not for the group assigned to doxycycline alone . No significant inter-treatment group differences in microfilaridermia were observable at 4 months . At 12 months post-doxycycline intervention ( 8 months post-ivermectin intervention ) , microfilaridermia was significantly reduced in all treatment groups . However , inter-treatment group differences were also apparent at this follow up . Doxycycline + ivermectin treated individuals had lower levels of microfilaridermia compared with both the ivermectin only and doxycycline only groups . Also , the incidence of amicrofilaridermia ( an absence of detectable mf in the skin ) was significantly higher in doxycycline + ivermectin groups ( 76 . 1% ) compared with both the ivermectin only ( 21 . 6% ) and doxycycline only ( 38 . 1% ) groups . At 21 months post-doxycycline intervention ( 17 months post-ivermectin intervention ) , significant microfilaridermia reductions from baseline persisted in all treatment groups . Inter-treatment differences between doxycycline + ivermectin and ivermectin groups were also preserved . Furthermore , the majority ( 89 . 1% ) of the doxycycline + ivermectin treatment group were amicrofilaridermic at 21 months , compared with 21 . 6% in the ivermectin only treatment group . At 21 months the doxycycline only group also showed greater reductions in microfilaridermia and increased frequency of amicrofilaridermia ( 66 . 7% ) compared with ivermectin only . When comparing between doxycycline + ivermectin and doxycycline only at 21 months , the combined treatment group showed significantly increased frequency of amicrofilaridermia . Table 4 summarizes the measurements of Wolbachia within adult O . volvulus derived from extirpated nodules at 21 months . In total 30 individuals who completed the trial were selected for nodulectomy ( selection based on the presence of palpable nodules in suitable sites for operation ) . Wolbachia wsp copy number , determined from genomic DNA extracted from one half of each extirpated nodule , was significantly lower in both doxycycline + ivermectin and doxycycline only treatment groups , compared with ivermectin only treatment . The depletion of Wolbachia in doxycycline treatment groups determined by PCR was corroborated by immunohistochemical staining of Wolbachia WSP within sections of adult O . volvulus tissues ( Table 4 , Figure 3 ) . In both doxycycline only and doxycycline + ivermectin treated individuals , WSP positive staining was not detected in any of the nodules examined . These frequencies were significantly lower when compared with the frequency of positive staining of O . volvulus tissues derived from patients treated with ivermectin only ( 67% ) . Histological observations of different embryonic stages within uteri and released mf in nodule tissue are summarized in Table 5 . The frequency of ova , morulae , curled stage and straight stage embryos within uteri were reduced in female worms derived from doxycycline + ivermectin treated individuals compared with ivermectin only treated individuals after 21 months . Similar reductions in frequency of embryonic developmental stages were observed in doxycycline only treated individuals compared with ivermectin only treated individuals . Presence of mf within nodule tissue sections were also reduced in both doxycycline + ivermectin and doxycycline only treated individuals compared with ivermectin only treated individuals at 21 months . Table 6 documents the results of USG examination . At the 21-month follow up , 50 patients with palpable nodules undertook ultrasound examination . Distinct adult worm movement could be observed within nodules in 44 . 0% of ivermectin treated patients . In comparison , the frequencies of detectable worm movement in doxycycline + ivermectin ( 7 . 1% ) or doxycycline only ( 11 . 1% ) treated individuals was significantly lower only for the doxycycline + ivermectin group . At 21 months the proportion of dead adult female worms from patients in the doxycycline + ivermectin ( 47% ) and doxycycline only groups ( 65% ) was significantly increased compared to the ivermectin only group ( 17% ) . There was also a significant reduction in the number of living adult female and male worms per patient ( Table 7 ) . Fluctuations in L . loa microfilaraemia from baseline are illustrated in Figure 4 and statistical differences summarized in Table 8 . The study design precluded a treatment comparison ( all L . loa patients were assigned doxycycline + ivermectin ) and therefore only longitudinal analyses were undertaken . No significant changes in L . loa microfilaraemia occurred at 4 months after doxycycline treatment ( immediately preceding ivermectin intervention ) . L . loa mf loads were significantly reduced at 12 months and 21 months after the start of doxycycline treatment ( 8 and 17 months after ivermectin treatment ) . M . perstans microfilaraemia had significantly increased from baseline at 4 months after doxycycline allocation , immediately preceding ivermectin intervention , in both doxycycline and placebo treatment groups ( Table 9 , Figure 5 ) . No inter-group differences in M . perstans mf levels at 4 months were observable . At 12 months after the start of doxycycline intervention ( 8 months following ivermectin intervention ) significant reductions in circulating mf compared with baseline levels had occurred in doxycycline + ivermectin treated individuals but not ivermectin only treated individuals . There was also a strong inter-group difference at 12 months in both M . perstans microfilaraemia and frequency of amicrofilaraemia ( 94 . 7% in doxycycline + ivermectin compared with 7 . 1% in ivermectin only treated individuals ) . By the 21-month follow up ( 17 months following ivermectin treatment ) , both treatment groups showed reductions in M . perstans microfilaraemia compared with baseline . There were no inter-group differences apparent at this stage with amicrofilaraemia occurring in the majority of both doxycycline + ivermectin ( 84 . 2% ) and ivermectin only ( 78 . 6% ) treated individuals . The major outcome of this trial is that a 6-week course of doxycycline alone is a highly effective treatment against onchocerciasis , leading to long term and profound suppression of microfilaridermia , embryogenesis and significant macrofilaricidal activity independently of ivermectin administration . It also shows that an additional treatment with ivermectin does not lead to any improvement in the macrofilaricidal or sterilising activity of doxycycline . Furthermore , our trial indicates that doxycycline treatment is well-tolerated in O . volvulus patients co-infected with low to moderate levels of L . loa parasitaemias . These findings promote the use of a doxycycline only regimen to treat onchocerciasis patients co-infected with L . loa . However , further trials are warranted to test safety and efficacy of doxycycline treatment in co-infected individuals at risk of developing serious adverse reactions to ivermectin . Should the results of such trials support our safety and efficacy findings reported here , a potential solution will be available to MDA programs currently disrupted by the threat of L . loa SAE . Previous trials on lymphatic filariasis that also demonstrated macrofilaricidal activity with doxycycline alone [28] suggest that this approach could be extended to co-infections of Wuchereria bancrofti and loiasis . The results of this trial confirms previous findings in onchocerciasis and lymphatic filariasis that a course of doxycycline sufficient to deplete Wolbachia by >90% results in the death of adult worms ( reviewed in [29] ) . Prior to the removal of the nodules for histochemical and PCR analysis we used ultrasonography ( USG ) to detect the in vivo motility of parasites . The USG data showed reduced parasite motility in doxycycline + ivermectin treated individuals compared with the ivermectin only group and suggests that USG maybe used as a non-invasive tool to assess potential macrofilaricidal activity prior to histological analysis of parasite viability . Histological and PCR analysis confirmed that doxycycline treatment resulted in loss of Wolbachia from the adult parasites , with an extensive loss of uterine contents reflecting a blockage of embryogenesis as previously observed [18] , [19] . Treatment with doxycycline + ivermectin or doxycycline alone is superior to ivermectin in achieving sustained reductions in skin mf . The kinetics of mf decline are in line with the different modes of actions of the two drugs . The slow decline in mf skin levels following doxycycline treatment is most likely a consequence of the block in embryogenesis preventing the release of mf into the skin . These kinetics are beneficial in avoiding the rapid death of mf , which in individuals with high parasite burden leads to inflammatory Mazzotti adverse events following anti-filarial drug treatment and are associated with the release of Wolbachia into the blood and tissues [30] , [31] . The low incidence of adverse events following ivermectin treatment in this trial probably reflect the relatively low O . volvulus microfilarial burden , which was further reduced by a reduction in O . volvulus microfilaridermia levels across all groups from baseline to the 4 month follow-up time point ( prior to ivermectin administration ) . The absence of SAE in L . loa co-infected individuals was as expected due to exclusion of patients with >8000 mf/ml parasitaemias . The tolerability of treatment with doxycycline is consistent with our previous experience in more than 1000 treated field trial volunteers . No experience of severe adverse event or evidence of photosensitivity has been recorded . For ethical and safety issues , our study design precluded enrolment of individuals co-infected with L . loa above the safety threshold for standard ivermectin treatment ( >8000 mf/ml ) . However , we noted no additional safety issues in L . loa co-infected patients during doxycycline allocation . The dropout rate for the trial did not differ between L . loa positive and negative treatment groups . Given that our trial demonstrates comparable efficacies of doxycycline with or without ivermectin , we believe further trials are warranted to determine the safety of doxycycline treatments in L . loa patients with >8000 mf/ml with O . volvulus co-infection . In this regard a trial of community directed delivery of doxycycline in an area of onchocerciasis and loiasis co-endemicity has been completed and demonstrates the feasibility of the use of large-scale doxycycline therapy for the control of onchocerciasis in such communities [38] . Some of the communities covered by this trial included those with a prevalence of loiasis in excess of 40% , where it can be estimated that more than 5% of this population would have microfilaraemia levels above the threshold of 8000 mf/ml . The lack of any SAE to doxycycline therapy in the 12 , 612 people that completed the course of treatment provides indirect evidence that doxycycline therapy appears safe in individuals with higher L . loa microfilaraemia . Although the results of this trial show no additional benefit of ivermectin to the macrofilaricidal and sterilising activity of doxycycline , the combination of both drugs improves reductions in O . volvulus mf intensity and frequency of amicrofilaridermia . The timing of the ivermectin treatment may be sub-optimal and may be different in populations exposed to ivermectin control rather than the ivermectin-naïve population treated here . The changes in L . loa microfilaraemia observed showed no change from baseline at the 4-month follow-up point . After ivermectin treatment mf loads were significantly reduced at both 12 and 21 month after the start of treatment as anticipated . The changes observed in M . perstans microfilaraemia showed a different pattern . At the 4-month follow-up mf levels had increased in both doxycycline and placebo groups and no differences between groups was observed . 8 months after ivermectin treatment there was a striking reduction in mf loads in the doxycycline group with no change from baseline levels observed in the ivermectin only group . At the 21-month follow up both groups showed marked reductions from baseline and high frequency of amicrofilaraemia . The difference between doxycycline and ivermectin only groups at the 12-month follow up would be consistent with a recent report that M . perstans is host to Wolbachia endosymbionts and a 6-week course of doxycycline leads to depletion of Wolbachia and microfilaraemia [32] . Although we were unable to confirm the presence of Wolbachia in M . perstans due to technical reasons , the observations of the changes to microfilaraemia at the 12-month follow up would be consistent with the presence and dependency of Wolbachia in M . perstans . If the elimination of onchocerciasis and lymphatic filariasis ( W . bancrofti ) as a public health problem is to be achieved in Africa , a solution to the problem of L . loa co-endemicity in Central Africa has to be found . Attempts to use regimes of low dose ivermectin have failed to provide sufficient reductions in mf suitable for MDP and are probably inadequate to prevent the occurrence of post-treatment neurological SAEs [33] . Further trials are currently underway with albendazole , which can reduce L . loa microfilaraemia following a twice-daily 21-day course [34] . Shorter 3-day treatments with albendazole failed to lead to sufficient reductions safe enough for ivermectin treatment [35] , [36] and so alternative regimes are required . One such regime , which we have demonstrated in this trial , is the targeting of onchocerciasis with anti-wolbachial therapy . Current regimes with doxycycline are restricted for widespread MDA due to contraindications in children under 8 years old and pregnancy and the logistics of 4–6 week courses of treatment . Trials to evaluate the minimum effective course of treatment with combinations of doxycycline and rifampicin in onchocerciasis is currently underway as part of the A-WOL drug discovery and development programme , which aims to optimise current anti-wolbachial drugs and discover new drugs with a more rapid efficacy and without the contra-indications of doxycycline [37] , ( A-WOL . com ) . In addition , community directed intervention ( CDI ) trials using a 6-week course of doxycycline have been completed and challenge the notion that prolonged courses of treatment cannot be effectively delivered through CDI [38] . It may be possible to use RAPLOA , a rapid diagnostic tool , to map areas of high risk of encephalopathy to define restricted areas where these regimes could be deployed . Such regimes can already be considered as a suitable treatment for individual cases of onchocerciasis or LF in patients co-infected with L . loa and the further development of anti-wolbachial regimes compatible with MDA could offer an alternative tool for the control on onchocerciasis and LF in Africa .	The control of onchocerciasis in Africa relies on the sustained delivery of ivermectin . In certain areas , annual treatments delivered with high population coverage for at least 15–17 years can break transmission . In other endemic settings this strategy alone is thought to be insufficient to eradicate the disease . One of the major limitations occurs in areas that are co-endemic with another filarial infection caused by Loa loa , due to the risk of a rare severe adverse event associated with the rapid killing of L . loa microfilariae in heavily parasitized individuals . There are also concerns over recent evidence of reduced efficacy of ivermectin and the possible development of resistance . An alternative approach is to target the Wolbachia bacterial endosymbionts of Onchocerca volvulus with the antibiotic , doxycycline . In an area of Cameroon co-endemic for onchocerciasis and loiasis we conducted a trial comparing doxycycline with or without ivermectin treatment to ivermectin treatment alone . A six-week course of doxycycline delivers macrofilaricidal and sterilizing activities , which is not dependent upon co-administration of ivermectin . Doxycycline is well tolerated in patients co-infected with moderate intensities of L . loa microfilariae . The trial indicates that anti-wolbachial therapy is a feasible alternative to ivermectin in communities co-endemic for onchocerciasis and loiasis .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "infectious", "diseases/helminth", "infections", "microbiology/parasitology", "infectious", "diseases/neglected", "tropical", "diseases" ]	2010	Macrofilaricidal Activity after Doxycycline Only Treatment of Onchocerca volvulus in an Area of Loa loa Co-Endemicity: A Randomized Controlled Trial
The transition from mitotic to meiotic cell cycles is essential for haploid gamete formation and fertility . Stimulated by retinoic acid gene 8 ( Stra8 ) is an essential gatekeeper of meiotic initiation in vertebrates; yet , the molecular role of STRA8 remains principally unknown . Here we demonstrate that STRA8 functions as a suppressor of autophagy during spermatogenesis in mice . Stra8-deficient germ cells fail to enter meiosis and present aberrant upregulation of autophagy-lysosome genes , commensurate with autophagy activation . Biochemical assays show that ectopic expression of STRA8 alone is sufficient to inhibit both autophagy induction and maturation . Studies also revealed that , Nr1d1 , a nuclear hormone receptor gene , is upregulated in Stra8-deficient testes and that STRA8 binds to the Nr1d1 promoter , indicating that Nr1d1 is a direct target of STRA8 transcriptional repression . In addition , it was found that NR1D1 binds to the promoter of Ulk1 , a gene essential for autophagy initiation , and that Nr1d1 is required for the upregulated Ulk1 expression in Stra8-deficient testes . Furthermore , both genetic deletion of Nr1d1 and pharmacologic inhibition of NR1D1 by its synthetic antagonist SR8278 exhibit rescuing effects on the meiotic initiation defects observed in Stra8-deficient male germ cells . Together , the data suggest a novel link between STRA8-mediated autophagy suppression and meiotic initiation . Meiosis is a fundamental process in sexual reproduction during which diploid cells halve their chromosome number by two rounds of cell divisions to generate haploid cells or gametes . In mammals , temporal regulation of meiosis in germ cells is sex-specific: meiosis in females begins during embryogenesis , whereas meiosis in males starts at puberty and persists throughout adulthood [1] . To enter meiosis , diploid cells must cease mitosis and then undergo one round of DNA replication , followed by the formation of DNA DSBs , meiotic chromosome pairing , cohesion , synapsis , and recombination . Although meiotic initiation has been extensively studied in some model organisms , such as yeast , flies , and worms , the mechanisms governing meiotic initiation in mammals remain elusive , largely because the molecular machinery controlling this process differs among species [2] . To date , the best characterized gatekeeper of meiotic initiation in vertebrates is stimulated by retinoic acid gene 8 ( Stra8 ) [3] . STRA8 is thought to act as a basic helix-loop-helix ( bHLH ) transcription factor , based on its DNA binding and transcriptional activity [4] . Interestingly , Stra8 is expressed in a precise tissue-specific and developmental manner , whereby it is transitorily expressed only in premeiotic germ cells , of both sexes , shortly before their entry into meiosis [5 , 6] . Functionally , Stra8 likely governs both meiotic initiation and early meiotic progression . In one study , Stra8-deficient germ cells ( exons 2–7 deleted ) of both sexes do not display molecular hallmarks of meiotic initiation ( e . g . , DSBs ) and fail to enter meiosis in juvenile mouse testes and in developing ovaries [7 , 8] . Another study that used a different Stra8-deficient mouse line , in which exons 2–4 were deleted on an F1 hybrid ( C57BL/6 x 129 ) background showed that Stra8 functions instead in early meiotic prophase in spermatogenesis [9] . Nevertheless , Stra8-deficient germ cells of both sexes undergo meiotic arrest [7–9] . Thus , STRA8 is an essential regulator of meiosis that likely acts as a transcription factor , but currently there is little information on its molecular role and functional targets implicated in meiosis [10] . In this study , we report the unexpected finding that STRA8 acts as a suppressor of autophagy . Autophagy is a catabolic process involving self-digestion of protein and cellular organelles through lysosomes [11] and a major stress response pathway that promotes cellular survival by supplying nutrients and energy . In addition , autophagy plays a critical role in maintaining protein and cellular organelle quality control [12] . In recent decades , studies have identified genes that encode essential molecular factors of the autophagy pathway [11] , including regulators of autophagy initiation ( VPS34/PIK3C3 , ULK1 ) , nucleation ( BECN1 ) , elongation ( ATG5-ATG12 , ATG7 , ATG16 , and LC3B ) , and maturation through lysosome biogenesis ( VPS18 and LAMP2 ) . Moreover , TFEB has been identified as a master regulator of autophagy by inducing a broad spectrum of autophagy-lysosome genes [13] . Interestingly , lack of autophagy has been shown to instigate DNA damage events , including DNA DSBs , in somatic cells [14–16] ( reviewed in ref . [17 , 18] ) . The reported findings support a mechanism whereby STRA8 suppresses germ cell autophagy via direct transcriptional inhibition of a second transcription factor , NR1D1 , which is needed for expression of the essential autophagy initiator ULK1 . The data show that loss of Nr1d1 expression or inhibition of NR1D1 function by its synthetic antagonist SR8278 exhibited rescuing effects on the meiotic initiation block observed in Stra8-deficient male germ cells . Together , our results suggest that STRA8-mediated autophagy suppression is a mechanistic feature of its role in meiotic initiation . To determine the molecular mechanism of STRA8-driven meiotic initiation , we used Stra8-deficient mice on a highly inbred C57BL/6 background [8] . We examined Stra8-deficient testes at 15–21 days postpartum ( d . p . p . ) by transmission electron microscopy . At this age , wild-type testes exhibited a normal presence of spermatogonia and meiotic spermatocytes ( Fig 1A , panels a and b ) , whereas Stra8-deficient ( Stra8-/- ) testes exhibited a complete lack of meiotic spermatocytes despite the presence of spermatogonia ( Fig 1B , panels a-c ) . Importantly , transmission electron microscopy revealed two novel characteristics in Stra8-deficient testes . First , the cellular integrity in the adluminal compartment of the seminiferous tubules appeared highly disrupted ( Fig 1B , panel b ) , with the degenerative cells in this region having lost membrane integrity and erupted their cellular contents into the lumen ( Fig 1B , panel d ) . Second , we frequently observed autophagosome structures ( 38 autophagosome structures observed in 200 germ cells from 2 Stra8-deficient testes ) , whereas in wild-type testes comparable autophagosome structures were not observed ( 0 autophagosome structures in 378 germ cells examined from 2 wild-type testes ) ( Fig 1A , panel b and c ) . Comparable autophagosome structures were not observed in Sertoli cells from both wild-type ( Fig 1A , panel e ) or Stra8-deficient testes ( Fig 1B , panel c and e ) . This is consistent with a recent report that autophagy was not detected in spermatogonia , early spermatocytes and Sertoli cells in rat testes [19] . The autophagosome structures observed in Stra8-deficient germ cells were located in the cytoplasmic region of both non-degenerative ( cells with intact membrane ) ( Fig 1B , panels f-i ) and degenerative cells ( Fig 1B , panel j ) . These double-membraned autophagosomes enclosed cellular organelles , including mitochondria ( Fig 1B , panels f—i ) and endoplasmic reticulum ( Fig 1B , panel j ) . Together , our transmission electron microscopy study revealed aberrant autophagosome formation in Stra8-deficient germ cells . During autophagy , autophagosomes serve as intermediate transport vesicles to target cellular components for degradation before their conversion into autolysosomes . Thus , aberrant appearance of autophagosomes in Stra8-deficient testes could result from either elevated autophagosome formation or impaired autophagosome turnover through fusion with lysosomes [20] . To distinguish between these possibilities , we examined autophagy activity in Stra8-deficient testes by using tandem fluorescent-tagged LC3 ( RFP-GFP-LC3 ) by breeding RFP-GFP-LC3 transgenic allele into wild-type and Stra8-deficient backgrounds [21] . LC3 is a soluble protein and is distributed ubiquitously in cells . Upon autophagy activation , the cytosolic form of LC3 ( LC3-I ) is conjugated to phosphatidylethanolamine ( PE ) to form LC3-II , which is recruited to autophagosomal membranes , thereby serving as a well-characterized marker for autophagosomes [22] . In this reporter system , autophagosome vesicles ( GFP-positive and RFP-positive puncta ) can be distinguished from acidified autolysosome vesicles ( GFP-negative and RFP-positive puncta ) due to acidic quenching of the GFP signal , but not the RFP signal , after fusion with lysosomes . Whole-mount immunofluorescence imaging of seminiferous tubules from wild-type testes showed occasional merged GFP-positive and RFP-positive puncta , indicative of autophagosomes ( Fig 2A ) . In contrast , seminiferous tubules from Stra8-deficient testes exhibited a significant increase in the number of both total vesicles ( RFP-positive puncta ) and autolysosome vesicles ( GFP-negative and RFP-positive puncta ) ( Fig 2A ) . Direct fluorescence imaging of testicular cross sections confirmed a profound increase of autophagy in 98% of Stra8-deficient seminiferous tubules ( Fig 2B ) . In contrast , 100% of the seminiferous tubules in wild-type testes exhibited diffuse GFP signal , suggesting cytoplasmic soluble LC3 and a lack of autophagosome formation ( Fig 2B , lower left panels ) . To further confirm that the lack of autophagosomes in wild-type seminiferous tubules did not result from rapid turnover , wild-type and Stra8-deficient juvenile mice were treated with chloroquine , a weakly basic lysosomotropic agent that can block autophagosome fusion with lysosomes . Whereas chloroquine treatment resulted in an accumulation of autophagosome vesicles ( GFP-positive and RFP-positive puncta ) in Stra8-deficient testes , we did not observe an appreciable effect of autophagosome vesicle accumulation in wild-type testes , suggesting low autophagy activity ( S1 Fig ) . Autophagy is an essential intracellular degradation process . To evaluate autophagic degradation ( flux ) in wild-type and Stra8-deficient testes , we examined the protein level of p62 ( or sequestosome 1 , SQSTM1 ) , a highly selective substrate for autophagic degradation [23] . The amount of p62 protein inversely correlate with autophagic flux activity: high levels of autophagy results in low p62 protein levels due to its degradation , while low levels of autophagy results in high p62 protein levels due to its accumulation . Wild-type testes contained seminiferous tubules with robust p62 protein accumulation at 21 d . p . p . ( Fig 3A ) . In contrast , p62 protein level was almost completely lost in age-matched Stra8-deficient testes ( Fig 3A ) . To determine if low p62 protein levels in Stra8-deficient testes were due to elevated autophagic degradation , possible changes due to Sqstm1 gene ( encoding p62 ) expression and autophagosome degradation ( by chloroquine treatment ) were evaluated . Quantification of Sqstem1 mRNA showed comparable levels in age-matched wild-type and Stra8-deficient testes ( Fig 3B ) and whereas chloroquine showed no appreciable effect on p62 protein levels in wild-type testes ( S2 Fig ) , chloroquine induced cytoplasmic p62 accumulation in germ cells of Stra8-deficient testes ( Fig 3C ) . Together , these data suggest there is a rapid autophagic degradation of p62 in Stra8-deficient germ cells , reflective of high autophagic flux . To help uncover the mechanism by which STRA8 influences autophagy , expression levels of 14 essential autophagy-lysosome genes were evaluated by quantitative RT-PCR ( qRT-PCR ) . For these studies , juvenile testes at 10 d . p . p . were used to assure that the germ cell content is comparable between wild-type and Stra8-deficient testes and , thus , observed differences between mRNA levels are not due to differences in germ cell numbers ( S3 Fig ) . Among these 14 genes , 6 genes , namely , Ulk1 , Atg5 , Map1lc3b , Vps18 , Lamp2 , and Tfeb , were significantly upregulated in Stra8-deficient testes ( Fig 4 ) . These genes encode essential factors for autophagosome formation ( ULK1 , ATG5 , Map1lc3b ) , lysosome function ( VPS18 , LAMP2 ) , as well as a master regulator of autophagy-lysosome genes ( TFEB ) . Together , these data suggest that induction of autophagy in Stra8-deficient testes results from upregulation of specific autophagy-lysosome gene expression . Our data in Stra8-deficient testes suggests that STRA8 may suppress autophagy by inhibiting autophagy-lysosome gene expressions . Currently , there are no available STRA8-expressing germ cell lines and , because Stra8 is transiently expressed on the verge of mitosis to meiosis transition , primary isolation and culture of Stra8-expressing cells could be challenging . Hence , to assess the role of STRA8 in autophagy suppression , STRA8 was ectopically and stably expressed in F9 embryonic carcinoma cells , a cell line regularly used for autophagy research [24] . Of note , autophagy machinery is present in every cell type and autophagy is an ongoing process in all cells; therefore , analysis of autophagy is often performed in cell lines [20] . To facilitate the identification of STRA8-expressing cells , STRA8 was tagged with GFP at its carboxyl terminus ( S4 Fig ) . First , to test whether STRA8 suppresses autophagy induction , we used three commonly used inducers of autophagy , namely , amino acid starvation , rapamycin ( mTOR inhibitor ) , and metformin ( AMPK inducer ) [25] . Autophagosome formation was detected by immunoblotting for LC3-II , which is a marker for autophagosomes [26] . It was found that , while LC3-II levels were significantly increased in control cells in all three conditions , there was no significant further increase of LC3-II levels in STRA8-expressing cells ( Fig 5 ) . These data suggest that STRA8 suppresses de novo autophagosome formation upon autophagy induction . Although de novo autophagosome formation is impaired by STRA8 upon autophagy induction ( Fig 5 ) , we noted that there was a significant increase of LC3-II under basal condition ( no autophagy induction ) in STRA8-expressing cells , suggesting that STRA8 also inhibits autophagosome maturation , which results in autophagosome accumulation ( upregulation of LC3-II ) ( Fig 6A ) . This result was confirmed at the cellular level by a significant increase of LC3 puncta ( Fig 6B ) . Inhibition of autophagy flux frequently leads to autophagosome accumulation . Indeed , in our in vitro RFP-GFP-LC3 assay to monitor autophagy flux , STRA8 expression induced a significant accumulation of autophagosome vesicles ( GFP-positive and RFP-positive puncta ) that failed to mature into autolysosome vesicles ( GFP-negative and RFP-positive puncta ) ( Fig 6C ) . LC3-II and p62 are selectively degraded by autophagy . During chloroquine treatment , LC3-II and p62 accumulate due to inhibited autophagic flux , thereby serving as an indicator of autophagy flux activity . We show that the increment of p62 accumulation is significantly smaller in STRA8-expressing cells , suggesting that autophagy flux is being inhibited by STRA8 . Consistent with the findings using in vitro RFP-GFP-LC3 assay , inhibition of autophagosome maturation by chloroquine resulted in a significant increase of LC3-II and p62 levels in control cells , but no significant further increase of LC3-II and p62 was observed in STRA8-expressing cells after chloroquine treatment ( Fig 6B , 6D and 6E ) . Collectively , these data suggest that STRA8 also blocks autophagosome maturation under basal conditions . To evaluate whether STRA8 influences autophagy-lysosome gene expression , the expression levels of the autophagy-lysosome genes that were upregulated in Stra8-deficient testes ( Fig 4 ) were examined in these cells . We found that STRA8 expression alone was sufficient to cause a significant decrease in their expression levels , including Ulk1 , Pik3C3 , LC3B , Vps18 , Lamp2 , and Tfeb ( Fig 6F ) . Together , these data suggest that STRA8 functions as a suppressor of autophagy by inhibiting autophagy-lysosome gene expression . Thus , loss of STRA8 function leads to the aberrant autophagy activation and upregulation of autophagy-lysosome gene expression in Stra8-deficient testes . STRA8 contains a highly conserved bHLH domain that exhibits DNA binding activity [4] . To gain mechanistic insight into STRA8-mediated autophagy suppression , two STRA8 mutants in the bHLH domain were generated ( S5A Fig ) : in the first mutant , point mutations were introduced in the first helix domain , which are known to disrupt the DNA binding activity of bHLH family transcription factors ( mHelix ) [27]; in the second mutant , point mutations were introduced in the basic domain , which disrupts the nuclear localization of STRA8 ( mNLS ) [4] . Both STRA8 mutants exhibited impaired nuclear localization ( S5B Fig ) and lost their ability to suppress autophagy activation ( S5C Fig ) as well as maturation ( S5D Fig ) . These data suggest that the bHLH domain of STRA8 is critical for its autophagy suppression function . To identify putative target gene ( s ) of STRA8 that could mediate its autophagy suppression function , we have performed an RNA-sequencing analysis in cells with transient ectopic expression of STRA8 under normal conditions . STRA8 upregulated 7 genes and downregulated 15 genes ( ≥ 2-fold change ) ( Fig 7A ) . Interestingly , none of the autophagy and lysosome genes upregulated in Stra8-deficient testes was detected under this condition , suggesting that STRA8 regulates autophagy through other target ( s ) . Among the regulated genes by STRA8 , we found that Nr1d1 , a gene downregulated by STRA8 , encodes a nuclear hormone receptor also known as Rev-erb-α . Nr1d1 is a critical circadian rhythm gene [28] . Recently , several studies have shown that NR1D1 acts as either an activator or an inhibitor of autophagy by regulating autophagy-lysosome gene expression , depending upon tissue context [29–31] . We therefore hypothesized that Nr1d1 could be a functional target of STRA8 . We confirmed that ectopic expression of STRA8 significantly reduced Nr1d1 levels in F9 cells and other cell types examined ( S6A and S6B Fig ) , while the bHLH mutants of STRA8 failed to inhibit Nr1d1 expression ( S6A Fig ) . Reciprocally , Stra8-deficiency induced a significant upregulation of Nr1d1 expression at mRNA level as detected by qRT-PCR analysis in testes ( S6C Fig ) and by in-situ hybridization ( Fig 7B ) . To further evaluate if the induction of Nr1d1 mRNA is intrinsic to germ cells , we isolated the c-Kit-positive integrin α6-low differentiating spermatogonia , in which Stra8 is predominantly expressed , as well as the c-Kit-negative and integrin α6-high undifferentiated spermatogonia population , in which Stra8 is yet to be fully activated ( S7 Fig ) [32] . We found that Nr1d1 is more significantly upregulated in differentiating spermatogonia isolated from Stra8-deficient testes ( Fig 7C ) . Moreover , we show that Stra8-deficient germ cells exhibited higher levels of NR1D1 expression at protein levels using immunofluorescence ( Fig 7D ) . To evaluate whether Nr1d1 is a direct genomic target of STRA8 , the proximal region of the Nr1d1 promoter was examined for possible STRA8 binding sites . This identified a highly conserved canonical E-box ( CAGCTG ) , the binding motif for members of the vertebrates bHLH protein family ( Fig 7E ) . Chromatin immunoprecipitation ( ChIP ) detected robust STRA8 binding to this region of the NR1D1 promoter ( Fig 7F ) , suggesting that STRA8 represses Nr1d1 transcription through E-box binding . Both bHLH mutants of STRA8 do not associate with the NR1D1 promoter at this E-box . Together , these data suggest that Nr1d1 is under direct transcriptional repression by STRA8 . To characterize whether STRA8 suppresses autophagy through NR1D1 , we noted that past studies have shown that NR1D1 regulates autophagy through modulating the expression of Ulk1 gene [29–31] , which encodes an essential autophagy initiator [33 , 34] . Ulk1 mRNA is upregulated in Stra8-deficient testes , which we further confirmed using in-situ hybridization ( Fig 8A ) . Moreover , we show that Ulk1 expression was significantly increased in c-Kit-positive integrin α6-low differentiating spermatogonia isolated from Stra8-deficient testes ( Fig 8B ) , concomitant with upregulation of NR1D1 expression ( Fig 7B–7D ) . Moreover , we examined a 2-kb region upstream of the transcription start site of mouse Ulk1 promoter and identified 3 RAR-related Orphan Receptor ( ROR ) DNA elements ( ROREs ) , which consists of ( A/G ) GGTCA and could be putative NR1D1 binding sties [35 , 36] . We therefore tested whether NR1D1 could directly regulate Ulk1 expression in mouse testis by a ChIP assay . Importantly , we found that NR1D1 binding affinity declined progressively from the distal to the proximal ROREs of the Ulk1 promoter ( Fig 8C ) , suggesting that NR1D1 activates Ulk1 expression by engaging directly on the distal ROREs . Transcriptional activation of Ulk1 is known to increase autophagy activity [37] . To test whether NR1D1 upregulation mediates the aberrant activation of Ulk1 transcription in Stra8-deficient testes , Ulk1 expression was evaluated in Stra8-/-;Nr1d1-/- double knockout mice . Testicular Ulk1 expression in double knockout mice was similar to that in wild-type , indicating that Nr1d1 is required for Ulk1 induction in Stra8-deficient testes ( Fig 8D ) . Together , the results suggest that STRA8 suppresses autophagy by transcriptionally repressing Nr1d1 expression and , consequently , inhibiting the expression of essential autophagy initiation gene Ulk1 . To test whether repression of Nr1d1 activation by STRA8 is required to initiate meiosis , we evaluated whether NR1D1 inhibition could rescue Stra8-deficient testicular germ cells from meiotic initiation arrest . Stra8-deficient germ cells fail to enter meiosis during the first round of meiotic initiation in juvenile mouse testes [8] . Thus , we evaluated meiotic initiation of testicular germ cells into leptotene spermatocytes at 10 d . p . p . . In agreement with the previous report [8] , we found that while spermatocytes at early meiotic prophase ( leptotene ) have appeared as a result of the first round of meiotic initiation in wild-type testes , these meiotic spermatocytes were absent in Stra8-deficient testes at this age ( Fig 9A and 9D ) . Consistently , germ cells exhibiting nuclear distribution of SYCP3 ( a synaptomeal complex protein ) [38] together with foci of γ-H2AX ( a hallmark of DNA DSBs ) [39] , two molecular characteristics of leptotene spermatocytes , were absent in Stra8-deficient testes ( Fig 9B and 9E ) . Furthermore , gene expression analysis showed that testicular levels of Spo11 , which encodes a topoisomerase essential for meiotic DSB formation [40 , 41] , Dmc1 , which encodes a recombinase functioning in meiotic DSB repair [42 , 43] , and Sycp3 , were significantly downregulated in Stra8-deficient mice ( S8 Fig ) . Therefore , these missing hallmarks of meiotic initiation in Stra8-deficient testes at 10 d . p . p . provide a platform to evaluate the potential rescuing effects of NR1D1 inhibition on meiotic initiation . Notably , juvenile testes from Stra8-/-;Nr1d1-/- mice at 10 d . p . p . contained germ cells with nuclear morphology resembling that of leptotene spermatocytes as observed in wild-type testes ( Fig 9A; S9A Fig ) . Consistently , immunostaining revealed germ cells that exhibit nuclear distribution of SYCP3 together with foci of γ-H2AX staining in these testes ( Fig 9B ) . Testicular levels of Spo11 , Dmc1 , and Sycp3 were significantly upregulated in Stra8-/-;Nr1d1-/- mice when compared to Stra8-/-;Nr1d1+/+ mice ( Fig 9C ) . Moreover , Nr1d1 knockout alone showed no appreciable effect on meiotic initiation ( Fig 9A; S9B and S9C Fig ) . Taken together , these results suggest that genetic loss of Nr1d1 exhibited rescuing effects on the meiotic initiation arrest in Stra8-deficient testicular germ cells . To further examine the effect of NR1D1 inhibition on meiotic initiation arrest in Stra8-deficient testicular germ cells , we treated Stra8-deficient mice with SR8278 , a synthetic NR1D1 antagonist [44] . Consistent with the results of genetic NR1D1 inhibition , we found that pharmacological inhibition recovered the appearance of germ cells with nuclear morphology as well as molecular hallmarks of leptotene spermatocytes in Stra8-deficient testes ( Fig 9D and 9E ) . In addition , SR8278 treatment significantly stimulated testicular levels of Spo11 , Dmc1 , and Sycp3 in Stra8-deficient testes ( Fig 9F ) . SR8278 treatment showed no appreciable effects on meiosis in wild-type testes ( Fig 9D; S9D and S9E Fig ) . Taken together , the results from both genetic and pharmacological inhibition of NR1D1 suggest that aberrant upregulation of NR1D1 contributes to the meiotic initiation arrest of Stra8-deficient germ cells and that inhibition of Nr1d1 expression is an important feature of STRA8-directed meiotic initiation . Despite dramatic sexual dimorphism in mammalian meiosis [45 , 46] , STRA8 appears to exhibit a comparable role in inducing both male and female meiosis [7 , 8] . Thus , to examine whether STRA8 adopts similar mechanism of autophagy suppression in inducing meiosis in females , we investigated female meiosis , which occurs during embryonic day 13 . 5 ( E13 . 5 ) to E16 . 5 . Consistent with the observations in postnatal Stra8-deficient testes , autophagosome structures were frequently identified in germ cells of Stra8-deficient E14 . 5 ovaries ( 17 autophagosomes observed in 119 germ cells from 2 ovaries ) . In contrast , similar structures were not observed in age-matched wild-type fetal ovaries ( 158 germ cells from 2 wild-type fetal ovaries examined ) ( Fig 10A and 10B ) . Moreover , autophagy-lysosome genes as well as the STRA8 target gene , Nr1d1 , were significantly upregulated in Stra8-deficient ovaries when compared to wild-type ovaries at the same developmental stage ( Fig 10C ) . However , similar changes were not observed in embryonic testes at this age , because at this age meiosis is still inactive and STRA8 is absent until after birth ( Fig 10D ) . Taken together , the results suggest that , similar to its role in male germ cell meiosis , STRA8 functions as a suppressor of autophagy in female meiosis . Stra8 was first reported as an essential gatekeeper of meiotic initiation in 2006 [7] . To our knowledge , the first study attempted to characterize the molecular function of STRA8 reports that STRA8 shuttles between nucleus and cytoplasm , but is mostly nuclear in freshly isolated germ cells [4] . This agrees with a study using immunostaining in PFA-fixed testicular sections , which localized STRA8 predominantly to the nucleus of meiosis-entering preleptotene spermatocytes [6] . In addition , protein-DNA cross-link studies showed STRA8 had DNA binding activity [4] . Moreover , STRA8 displays transcriptional activity when fused to a GAL4-DNA binding domain , [4] . Although these in vitro assays suggest STRA8 as a transcription factor , the targets of STRA8 and their molecular consequences have yet to be identified . To characterize the meiotic gene program regulated by STRA8 in vivo , Soh and colleagues conducted RNA-sequencing analysis in wild-type and Stra8-deficient fetal ovaries at E14 . 5 , when ovarian germ cells enter meiosis [47] . However , only meiotic genes were not selected for investigation by Soh and colleagues . Interestingly , examination of this earlier RNA-sequencing result revealed notable similarities to the data reported herein , namely , the autophagy-lysosome genes that exhibit significant upregulation in both Stra8-deficient testes at 10 d . p . p . and Stra8-deficient fetal ovaries at E14 . 5 in our study ( Fig 4 and Fig 10C , respectively ) , including Ulk1 , Atg5 , Map1lc3b , Lamp2 , Vps18 , and Tfeb , also exhibited tendency of being upregulated in Stra8-deficient fetal ovaries at E14 . 5 in the RNA-sequencing result of Soh and colleagues . Moreover , except for Map1lc3a , the autophagy-lysosome genes that were not upregulated in Stra8-deficient testes at 10 d . p . p . in our study ( Fig 4 ) , i . e . , Pik3c3 , Atg7 , Vps11 , Uvrag , Lamp1 , and Becn1 , were also not upregulated in Stra8-deficient fetal ovaries at E14 . 5 in the RNA-sequencing result of Soh and colleagues . Thus , changes in autophagy-lysosome gene expression in Stra8-deficient fetal ovaries at E14 . 5 detected by Soh and colleagues using RNA-sequencing align with the data presented in the current study . These data together suggest that STRA8 inhibits the expression of a selective autophagy-lysosome gene program , thereby suggesting STRA8 functions as s suppressor of autophagy . A more recent study by Shen and colleagues is focused on characterizing a potential role for STRA8 in preventing apoptosis of germ cells [48] . Generally speaking , autophagy is associated with a response to stress to prevent cells death; however , excessive autophagy leads to apoptosis [49] . Stra8-deficient germ cells exhibited profound activation of autophagy ( Figs 1–4 ) which may represent a condition of uncontrolled autophagy activation that leads to their germ cell apoptosis [8] . To date , the roles of autophagy regulation in meiosis remains unclear . In budding yeast ( S . cerevisiae ) , autophagy participates in the early phase of meiosis and is switched off upon meiotic division [50] . In fission yeast ( S . pombe ) , however , autophagy is thought to be required for chromosome segregation during meiotic division [51] . Thus , these observations underscore the concept that mechanisms of meiotic initiation could differ in model organisms [2] . Our observation that autophagy is being actively suppressed by STRA8 during meiotic initiation in mouse spermatogenesis is in accordance with a recent study of autophagy in rat spermatogenesis , in which autophagy was found only to be activated during late meiotic spermatocytes but not in spermatogonia and early spermatocytes [19] . This is in line with the finding in yeast , in which autophagy is required for proper meiotic chromosome segregation [51] . Together , our findings suggest that activation of autophagy is specifically prevented by STRA8 during meiotic initiation in mammalian germ cell development . It should be noted that Stra8-deficient germ cells , when rescued by inhibiting NR1D1 antagonist SR8278 , do not progress beyond leptotene stage ( S10 Fig ) , suggesting that STRA8 has additional role ( s ) beyond the initiation of meiosis through NR1D1-independent mechanisms . For instance , based on our RNA-seq analysis , 15 out of 22 STRA8-regulated genes of are noncoding RNAs ( Fig 7A ) . Given that non-coding RNAs play an essential role in meiosis prophase and homologous recombination [35] , it is possible that STRA8 regulates this stage of meiosis through noncoding RNAs . We observed that STRA8-deficiency resulted in a significantly autophagy-lysosome gene expression in testes; reciprocally , these autophagy-lysosome genes were downregulated in cultured cells with stable STRA8 expression . Based on these data and a past study that indicated STRA8 could bind to DNA and display transcriptional activity [4] , we had expected that STRA8 , like some other transcriptional regulators of autophagy , such as TFEB [13] or FXR [52] , could potentially regulate a wide spectrum of autophagy-lysosome genes directly . However , no autophagy-lysosome genes were identified as direct targets of STRA8 by transient ectopic expression of STRA8 in our in vitro RNA-sequencing analysis under basal condition . Instead , Nr1d1 was identified as a target repressed by STRA8 transcriptionally . To date , three studies have reported a role of NR1D1 in either inducing or inhibiting autophagy , all involving transcriptional regulation of Ulk1 [29–31] . While Woldt and colleagues reported that NR1D1-deficiency in murine muscle leads to upregulation of autophagy [30] , Chandra and colleagues reported that Nr1d1-knock down leads to reduced autophagy and downregulation of Ulk1 gene expression in human macrophage [29] . Interestingly , whether NR1D1 is an autophagy inducer or inhibitor remains unclear in the study reported by Huang and colleagues [31] . Thus , these studies point out a critical regulatory role of NR1D1 in autophagy through modulating autophagy-lysosome gene expression . Herein , the data show that genetic loss of NR1D1 prevented the upregulated Ulk1 expression normally observed in Stra8-knockout testes ( Fig 8D ) , supporting a role for NR1D1 as an inducer of autophagy by activating autophagy-lysosome gene expression during germ cell development . It is currently unclear why there is a robust activation of autophagy that is precisely counteracted by STRA8-mediated suppression of autophagy during the transition to meiosis . One possibility is that the activation of autophagy is simply a response to cellular stress . A second , and more intriguing , possibility is that autophagy induction also plays an essential role in meiotic initiation and is an area that warrants further investigation . Based primarily on models of somatic tumorigenesis , genetic disruption of critical autophagy factors can induce the formation of DNA DSBs upon metabolic stress [14 , 15] . In germ cells , DSBs are required to initiate meiosis and permit the exchange of genetic information between maternal and paternal chromosomes through homologous recombination [53] . Stra8 activation is required for meiotic DSB formation [7 , 8] . Thus , by characterizing STRA8 as a suppressor of autophagy , our work suggests autophagy suppression as a possible mechanism adopted by germ cells to form meiotic DSBs . The mechanisms underlying autophagy inhibition-induced DSB formation is currently unclear . Several studies show that autophagy-deficiency causes DNA damage through accumulation of autophagic substrate , p62 [54–56] . However , p62-deficient mice are reportedly to be fertile [57] , suggesting that p62 accumulation alone is dispensable for meiosis . In addition , loss of autophagy has been shown to affect DNA damage repair machineries , such as HIPα ( a molecule essential for chromosome condensation ) [16] and Chk1 ( a molecular regulator of DNA damage repair by homologous recombination ) [58] . Thus , a role for autophagy suppression in DSB formation during meiotic initiation remains to be determined . Mounting efforts have been directed to derive functional haploid gametes ( sperm or oocytes ) from spermatogonial stem cells , embryonic stem cells or induced pluripotent stem cells in cultures [59] . However , how to properly induce and sustain meiosis remains to be a major challenge in this field . Our study revealed that germ cells entering meiosis is exposed to an antagonistic pressure of simultaneous autophagy induction and STRA8-mediated autophagy suppression . Thus , manipulating autophagy pathway to mimic this in vivo condition may facilitate the induction of meiosis in cultures , thereby advancing the technology of in vitro haploid gamete production that may ultimately afford clinical utility in assisted reproduction technology . All genetically modified mice were obtained from Jackson Laboratory: Stra8-deficient mice ( Stock number: 023805 ) , RFP-GFP-LC3 transgenic mice ( Stock number: 027139 ) , and Nr1d1-deficient mice ( Stock number: 018447 ) . For chloroquine treatment , mice were injected intraperitoneally with chloroquine dissolved in PBS at 100 mg/kg body weight daily . For SR8278 treatment , mice at 7 days of age were injected intraperitoneally with vehicle ( 5% DMSO , 10% Cremephol , and 85% PBS ) or SR8278 at 100 mg/kg body weight daily for 3 days . Postnatal testes were dissected from male mice after euthanasia . For timed pregnancy , females at estrus stage were caged with males overnight , and the presence of vaginal plug was examined the following morning and the midday was defined as E0 . 5 . Pregnant females were euthanized when embryos reached E14 . 5 . Embryos were collected from uterine horns before gonad isolation under a binocular dissecting microscope . All procedures and care of animals were carried out according to the Massachusetts General Hospital ( MGH ) and the University of Kansas Medical Center ( KUMC ) Institutional Animal Care and Use Committee ( IACUC ) under IACUC protocol number 2012N000015 and 2018–2461 , respectively . Euthanasia was performed by CO2 inhalation followed by cervical dislocation ( MGH ) and decapitation ( KUMC ) . Tissue specimens were fixed in 2 . 5% glutaraldehyde in 0 . 1M sodium phosphate buffer ( pH 7 . 4 , Electron Microscopy Sciences , Hatfield , PA ) , then rinsed several times in 0 . 1M sodium cacodylate buffer . Specimens were post-fixed in 1 . 0% osmium tetroxide in cacodylate buffer for 1 hour at room temperature and rinsed several times in cacodylate buffer . Samples were then dehydrated through a graded series of ethanols to 100% and dehydrated briefly in 100% propylene oxide . Samples were pre-infiltrated 2 hours in a 2:1 mix of propylene oxide and Eponate resin ( Ted Pella , Redding , CA ) , then transferred into a 1:1 mix of propylene oxide and Eponate resin and allowed to infiltrate overnight on a gentle rotator . The following day , specimens were infiltrated with fresh 100% Eponate resin for several hours , embedded in flat molds with fresh 100% Eponate and allowed to polymerize 24–48 hours at 60°C . Thin ( 70nm ) sections were cut using a Leica EM UC7 ultramicrotome , collected onto formvar-coated grids , stained with uranyl acetate and Reynold's lead citrate and examined in a JEOL JEM 1011 at MGH or The JEOL JEM-1400 at KUMC transmission electron microscope at 80 kV . Images were collected using an AMT digital imaging system ( Advanced Microscopy Techniques , Danvers , MA ) . Human STRA8 expression plasmid was obtained from Origene ( RC213536 ) . To tag STRA8 with GFP , STRA8 cDNA was inserted into the pEGFP-N1 vector ( Clontech ) . Mutant STRA8 plasmids were generated using Q5 mutagenesis kit ( New England Biolabs ) . All plasmids were sequenced to confirm fidelity prior to use . F9 embryonic carcinoma cells ( ATCC CRL-1720; American Type Culture Collection ) and 293T were cultured at 37°C in a humidified atmosphere of 5% CO2-95% air in DMEM containing 4 . 5 g/l glucose and supplemented with 10% FBS ( Invitrogen ) , 2 mM L-glutamine , 50 μg/ml penicillin and 50 μg/ml streptomycin . F9 cells were cultured on 0 . 1% gelatin-coated tissue culture plates . Cells were transfected using Lipofectamine 2000 ( Invitrogen ) and then stably selected by G418 ( Invitrogen ) for 2 to 3 weeks , before STRA8-expressing cells ( GFP-positive ) were identified and isolated by flow cytometry . F9 cells expressing GFP alone were used as control . For LC3 staining , cells were fixed in ice-cold methanol for 2 minutes before immunostaining with antibody against LC3 ( 12741 , Cell Signaling Technology ) . Images were captured by using a ZEISS LSM 800 confocal microscopy with Airyscan . To induce autophagy , control or Stra8-expressing cells were treated with rapamycin ( 0 . 1 mM ) , metformin ( 2 mM ) , or Earle's balanced salt solution ( EBSS ) for 2 hours before cell lysates were collected . To monitor the effects of blocking autophagosome maturation , control or STRA8-expressing cells were treated with chloroquine ( 20 μM ) for 2 hours before cell lysates were collected . Hela cells carrying mRFP-GFP-LC3 were transfected by plasmid expressing empty pCMV6 vector or STRA8 ( not tagged with GFP ) . Then , cells were fixed by 4% paraformaldehyde for 10 minutes at room temperature before observation under fluorescence microscope . Testes were fixed in 4% paraformaldehyde , embedded in paraffin , and sectioned for analysis . Antibodies used include: p62 ( ab56416 , abcam; 1:2000 dilution for immunofluorescence and 1:10 , 000 dilution for immunohistochemistry ) , γ-H2AX 05–636 , Millipore; 1:1 , 000 dilution ) , NR1D1 ( sc-100910 , Santa Cruz Biotechnology; 1:500 dilution ) , SYCP3 ( sc-74569 , Santa Cruz Biotechnology; 1:500 dilution ) . For immunofluorescence , detection was performed using Alexa fluor 546-conjugated goat anti-rabbit antibody and Alexa fluor 488-conjugated goat anti-mouse secondary antibodies . For immunohistochemistry , detection was performed using goat anti-mouse or goat anti-rabbit as secondary antibody for horseradish peroxidase-based DAB detection ( DAKO ) . Images were captured using a Nikon ECLIPSE TE2000-S microscope and were analyzed by Image J software ( National Institutes of Health ) . Dissected seminiferous tubules from wild-type and Stra8-deficient testes carrying the RFP-GFP-LC3 reporter were fixed in 4% paraformaldehyde on ice for an hour . Tissue is washed with PBS to remove PFA . For whole-mount imaging , tissues of seminiferous tubules were mounted directly on glass slides in PBS for confocal microscopy by using a Nikon A1R microscope . For imaging on testicular cross sections , testes were cryosectioned . Then sections were washed with warm PBS to remove gelatin and were imaged under Nikon ECLIPSE TE2000-S microscope . No staining was performed under both conditions . Images from wild-type and Stra8-deficient samples mounted on the same slide were captured and processed in parallel using identical settings . Total protein was isolated in RIPA buffer supplemented with 1 mM PMSF ( Sigma ) and protease inhibitor cocktail ( Sigma P8340 ) . Lysates were cleared by centrifugation at 14 , 000 X g for 10 min at 4°C , and protein concentrations in supernatants were determined ( DC protein assay; BioRad ) . Equal amount of protein from each sample was mixed with LDS sample buffer ( Invitrogen ) plus sample reducing agent ( Invitrogen ) , and denatured for 10 min at 70°C . Proteins were resolved in Bis-Tris gels ( Thermo Fisher ) , and transferred to PDVF membranes . Blots were probed with antibodies against LC3A/B ( 12741 , Cell Signaling Technology; 1:1 , 000 dilution ) , p62 ( ab56416 , Abcam; 1:20 , 000 dilution ) , GFP ( sc-9996 , Santa Cruz Biotechnology; 1:1 , 000 dilution ) , or pan-actin ( MS-1295 , Thermo Fischer; 1:1 , 000 dilution ) , washed and reacted with horseradish peroxidase-conjugated goat anti-rabbit or anti-mouse IgG ( BioRad ) . Detection was performed with the Clarity ECL Western Blotting Substrate ( BioRad ) . 293T cells were transfected with plasmid expressing GFP-tagged wild-type and mutant STRA8 . 24 hours later , cell lysates were processed using the EZ-ChIP kit ( Millipore , Temecula , CA ) along with a rabbit polyclonal anti-GFP antibody ( ab290; Abcam ) for immunoprecipitation . Normal rabbit IgG was used as a negative control . Precipitated soluble chromatin was analyzed by PCR using primer sets to target E-box: forward , 5’-CCC TCC CCG GCT TCT CTC TCT CC-3’ , reverse , 5’-GCA AAC CTT GCA AAC GTG AGG GC-3’; and to target exon 8: forward , 5’-CCG GAC CTG CGG ACC CTG AAC AA-3’ , reverse , 5’-TCT GTA CAA GGG GGC AGC GGC AGA-3’ . To characterize NR1D1 binding to the Ulk1 promoter , testicular lysates were processed using the EZ-ChIP kit ( Millipore , Temecula , CA ) along with a rabbit polyclonal anti-NR1D1 antibody ( #13418; Cell Signaling Technology ) for immunoprecipitation . Normal rabbit IgG was used as a negative control . Precipitated soluble chromatin was analyzed by PCR using primer sets to target ROREs: primer set 1 , forward , 5’-AAT GGG TAT GTG CGA CAA CA-3’ , reverse , 5’-TGT CAT TTG GGG AGG GGT AT-3’; primer set 2 , forward , 5’-TGC CAA GTT TGA CAA CCT GA-3’ , reverse , 5’-CTG TAT GTG GGG ACG GAG AC-3’; primer set 3 , forward , 5’-GCA CCT GCC TTT AAT TCC AA-3’ , reverse , 5’-CGA CTG GTC TCG AAC TTG CT-3’ . MCF-7 cells were transiently transfected with control ( pCMV6 ) or STRA8 plasmid . Cells were collected after 24 hours . Total RNA was isolated using the RNeasy Mini Kit ( Qiagen ) . Following rRNA depletion using RiboZero kit ( Epicentre/Illumina ) , RNA-Seq libraries were constructed using NEBNext Ultra Directional RNA library prep kit for Illumina ( New England Biolabs ) and sequenced on Illumina HiSeq2500 instrument , resulting in approximately 25 million reads per sample on average . STAR aligner was used to map sequencing reads to transcripts in hg19 reference genome . Read counts for individual transcripts were produced with HTSeq-count , followed by the estimation of expression values and detection of differentially expressed transcripts using EdgeR . Principle component analysis ( PCA ) was performed on the union of differentially expressed transcripts in all samples . In situ hybridization ( ISH ) has been described previously [60] . Antisense Nr1d1 probe ( 825 bp ) and Ulk1 probe ( 843 bp ) were amplified from cDNA prepared from juvenile testes , and labeled with digoxigenin ( Roche Diagnostics ) . Testes were collected from age-matched wild-type and Stra8-deficient mice . Testicular cells were dissociated by two-step enzymatic digestion , followed by staining with PE-conjugated rat anti-human integrin α6 antibody ( BD Pharmingen , clone GoH3 ) and APC-conjugated rat anti-mouse c-Kit antibody ( BD Pharmingen , clone 2B8 ) as previously described [32] . 40 , 000–50 , 000 cells from each population were sorted directly into lysis buffer in NucleoSpin RNA XS kit ( Takara ) for subsequent RNA isolation . Equal amount of total RNA from each sample was reverse transcribed by using SuperScript III from Invitrogen . quantitative RT-PCR was conducted by using SsoAdvanced Universal SYBR Green Supermix ( BioRad ) . Primer sequences are listed below: β-actin forward , 5’-CTG CCG CAT CCT CTT CCT C-3’ reverse , 5’-GCC ACA GGA TTC CAT ACC CA-3’ Mvh forward , 5’-GCT TCA TCA GAT ATT GGC GAG T-3’ reverse , 5’-GCT TGG AAA ACC CTC TGC TT-3’ Maplc3A forward , 5’-CAC ATC CTG GGT AGG TCC TG-3’ reverse , 5’-AAT GAC AAA CCC CAC AGA GC-3’ Maplc3B forward , 5’-CGG CTT CCT GTA CAT GGT TT-3’ reverse , 5’-AAC CAT TGG CTT TGT TGG AG-3’ Atg12 forward , 5’-TCC CCG GAA CGA GGA ACT C-3’ reverse , 5’-TTC GCT CCA CAG CCC ATT TC-3’ Atg5 forward , 5’-TGT GCT TCG AGA TGT GTG GTT-3’ reverse , 5’-GTC AAA TAG CTG ACT CTT GGC AA-3’ Atg7 forward , 5’-GTT CGC CCC CTT TAA TAG TGC-3’ reverse , 5’-TGA ACT CCA ACG TCA AGC GG-3’ Pik3c3 forward , 5’-CCT GGA CAT CAA CGT GCA G-3’ reverse , 5’-TGT CTC TTG GTA TAG CCC AGA AA-3’ Tfeb forward , 5’-CCA CCC CAG CCA TCA ACA C-3’ reverse , 5’-CAG ACA GAT ACT CCC GAA CCT T-3’ Ulk1 forward , 5’-AAG TTC GAG TTC TCT CGC AAG-3’ reverse , 5’-CGA TGT TTT CGT GCT TTA GTT CC-3’ UVRAG forward , 5’-ACA TCG CTG CTC GGA ACA TT-3’ reverse , 5’-CTC CAC GTC GGA TTC AAG GAA-3’ Vps11 forward , 5’-AAA AGA GAG ACG GTG GCA ATC-3’ reverse , 5’-AGC CCA GTA ACG GGA TAG TTG-3’ Vps18 forward , 5’-ACG AGG ACT CAT TGT CCC G-3’ reverse , 5’-CAT ACC CAG AAT GGG GGA TGC-3’ Lamp1 forward , 5’-CAG CAC TCT TTG AGG TGA AAA AC-3’ reverse , 5’-ACG ATC TGA GAA CCA TTC GCA-3’ Lamp2 forward , 5’-TGT ATT TGG CTA ATG GCT CAG C-3’ reverse , 5’-TAT GGG CAC AAG GAA GTT GTC-3’ Sqstem1 forward , 5’-AGG ATG GGG ACT TGG TTG C-3’ reverse , 5’-TCA CAG ATC ACA TTG GGG TGC-3’ Beclin1 forward , 5’-ATG GAG GGG TCT AAG GCG TC-3’ reverse , 5’-TCC TCT CCT GAG TTA GCC TCT-3’ Nr1d1 forward , 5’-ATG CCC ATG ACA AGT TAG GC-3’ reverse , 5’-CGG TGT GGA GTT GTA GCT GA-3’ All experiments were replicated at least three times independently . Different mice , tissues or cells were used during each experimental replicate . Animal assignment to each experimental group was made randomly . Quantitative data from the experimental replicates were pooled and are presented as the mean ± SEM or mean ± SD as indicated in the figure legend . Compiled data were analyzed by Student’s t-test .	Meiotic initiation is a key feature of sexual reproduction that launches an intricate chromosomal program involving DNA double strand breaks ( DSBs ) , homolog pairing , cohesion , synapsis , and recombination . Vertebrate gene Stra8 is an essential gatekeeper of meiotic initiation . However , the molecular role of STRA8 and its target genes remain elusive . Using mouse spermatogenesis as a model , we report that STRA8 suppresses autophagy by repressing the transcription of a nuclear hormone receptor gene Nr1d1 , and in turn , silencing the expression of Ulk1 , a gene essential for autophagy initiation . Given that autophagy is critical for protein and cellular organelle recycling and for preventing genomic instability , our study suggests that this newly demonstrated function of STRA8 , as a suppressor of autophagy , may be an important mechanistic feature of its role in meiotic initiation .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "cell", "death", "meiosis", "medicine", "and", "health", "sciences", "autophagic", "cell", "death", "reproductive", "system", "spermatocytes", "vesicles", "cell", "cycle", "and", "cell", "division", "drugs", "cell", "processes", "antimalarials", "germ", "cells", "pharmacology", "cellular", "structures", "and", "organelles", "sperm", "animal", "cells", "chromosome", "biology", "gene", "expression", "chloroquine", "testes", "anatomy", "cell", "biology", "genetics", "biology", "and", "life", "sciences", "cellular", "types", "genital", "anatomy" ]	2019	Meiotic gatekeeper STRA8 suppresses autophagy by repressing Nr1d1 expression during spermatogenesis in mice
Meiotic crossovers ( COs ) shape genetic diversity by mixing homologous chromosomes at each generation . CO distribution is a highly regulated process . CO assurance forces the occurrence of at least one obligatory CO per chromosome pair , CO homeostasis smoothes out the number of COs when faced with variation in precursor number and CO interference keeps multiple COs away from each other along a chromosome . In several organisms , it has been shown that cytoskeleton forces are transduced to the meiotic nucleus via KASH- and SUN-domain proteins , to promote chromosome synapsis and recombination . Here we show that the Arabidopsis kinesin AtPSS1 plays a major role in chromosome synapsis and regulation of CO distribution . In Atpss1 meiotic cells , chromosome axes and DNA double strand breaks ( DSBs ) appear to form normally but only a variable portion of the genome synapses and is competent for CO formation . Some chromosomes fail to form the obligatory CO , while there is an increased CO density in competent regions . However , the total number of COs per cell is unaffected . We further show that the kinesin motor domain of AtPSS1 is required for its meiotic function , and that AtPSS1 interacts directly with WIP1 and WIP2 , two KASH-domain proteins . Finally , meiocytes missing AtPSS1 and/or SUN proteins show similar meiotic defects suggesting that AtPSS1 and SUNs act in the same pathway . This suggests that forces produced by the AtPSS1 kinesin and transduced by WIPs/SUNs , are required to authorize complete synapsis and regulate maturation of recombination intermediates into COs . We suggest that a form of homeostasis applies , which maintains the total number of COs per cell even if only a part of the genome is competent for CO formation . During meiosis , chromosomes inherited from the mother and father are mixed in a process termed homologous recombination , to generate unique chromosomes that will be transmitted to the next generation . This genetic mixing has sustained the evolution of eukaryotes . There are typically one to four exchange points –crossovers ( COs ) - between homologous chromosomes at each meiosis . The distribution of these COs is under a series of constraints [1] , [2] . First , there is at least one CO per chromosome pair ( obligatory CO or CO assurance ) . Indeed , beyond their genetic consequences , COs are also essential for holding homologous chromosomes together during meiosis I , ensuring their balanced distribution in daughter cells . Notably , a lack of or improper positioning of this obligatory CO causes aneuploidy in human oocytes [3] . Second , COs are subject to interference . This prevents the occurrence of COs next to each other , shaping an even distribution and limiting their number [4] . COs are also under homeostasis , meaning that their number tends to be stable even when faced with variation in precursor number [5]–[7] . Finally , looking at frequencies , COs are not homogenously distributed along the genome; hot and cold regions have been defined at the chromosome scale , and hotspots with a very high CO frequency have been observed at the kb scale [8] , [9] . COs are produced during meiotic prophase I concomitantly with and functionally connected to chromosome pairing and synapsis , which is the intimate association of homologous chromosomes lengthways with a protein structure , the synaptonemal complex ( SC ) . Recombination is initiated at early prophase I by the formation of DNA double-strand breaks ( DSBs ) which largely outnumber the eventual CO number [10] . DSBs are subsequently resected to yield 3′ overhangs that invade the homologous chromosome , a step in which the recombinase DMC1 plays a prominent role [11] . In plants , as in mammals and budding yeast , these early steps of recombination also promote homologous chromosome synapsis . Indeed mutants affected in DSB formation or homologous template invasion ( including Atdmc1 ) fail in both synapsis and CO formation [12]–[14] . DSB repair events form intermediates that are eventually resolved as either COs or non-crossovers ( NCO ) [15] , [16] in the context of the SC . DSB repair can also occur using the sister as a template , a process that does not lead to inter-homologue COs , as observed in the Atdmc1 mutant [12] , [13] or during haploid meiosis [17] where the ubiquitous recombinase RAD51 catalyzes sister repair . However the prevalence of such sister-mediated repair in wild-type Arabidopsis is unclear [15] . Homologous chromosome invasion events can mature into COs through at least two independent pathways . These two pathways coexist in budding yeast , mammals and Arabidopsis [1] , [18]–[21] . Class I COs , the most prevalent class , are subject to interference and their production is dependent on the ZMM proteins ( in Arabidopsis: SHOC1 ( AtZIP2 ) , PTD1 , AtHEI10 , AtZIP4 , AtMSH4 , AtMSH5 , AtMER3 [18] ) . Interestingly , in Arabidopsis zmm mutants most chromosomes do not form COs , but synapsis occurs normally . This shows that recombination intermediates which promote synapsis are produced in zmm mutants , even if they are not eventually converted into COs . Formation of the minor Class II COs ( ∼15% of CO ) , that do not display interference , involves MUS81 [1] , [18]–[21] and is down regulated by AtFANCM [22] , [23] . All these molecular events must be coordinated at the chromosome and cellular levels to shape CO distribution . Interestingly , it has been shown in several species that chromosome movement is particularly prominent at meiotic prophase and plays a significant role in chromosome pairing/synapsis and CO formation [24] , [25] . Telomeres in mammals , fungi and plants , or specific chromosome sites called pairing centers in the case of C . elegans , bind to the nuclear envelope where they are subject to cytoskeleton originated forces [26]–[31] . Telomere chromosome movements are also illustrated by a transient prophase I configuration called the bouquet , where telomeres cluster together on the nuclear envelope . This highly polarized nucleus stage has been described since the early 1900s [32] . Cytoplasmic forces are transduced to the chromosomes inside the nucleus , through the nuclear envelop which is intact during meiotic prophase I , via a chain of proteins ( reviewed in [24] , [25] ) . Central in this chain are the KASH ( Klarsicht/ANC-1/Syne-1 homology ) -domain and SUN ( Sad-1/UNC-84 ) -domain proteins . In yeasts , worms and mammals KASH-domain and SUN-domain proteins play a crucial role in meiotic chromosome movement and homologue pairing [25] , [27] , [29] , [33] , [34] . KASH-domain proteins localize to the outer nuclear membrane and interact with SUN-domain proteins which are inserted in the inner membrane . However , the connection to the cytoskeleton , at one end , and to the chromosome ( telomeres in most species and pairing centers in C . elegans ) associated protein at the other end , appears to rely on evolutionary divergent proteins ( reviewed in [24] , [25] ) . In S . cerevisiae KASH-domain proteins interact with actin [35] while S . pombe , C . elegans and mouse KASH-domain proteins interact with dynein [25] , a microtubule motor protein . In plants , the movement of chromosomes at meiotic prophase has been directly observed in maize and the application of specific depolymerizing drugs suggests that it depends on both tubulin and actin [31] . In Arabidopsis , where live imaging is not yet available , telomeres appear to be associated around the nucleolus in early meiotic prophase and are moved to the nuclear membrane preceding synapsis where they transiently cluster together ( but not in a tight manner as in the classical bouquet configuration observed in many species [36] ) . On completion of synapsis the paired telomeres are dispersed but remain attached to the nuclear membrane until diplotene when they dissociate from the nuclear membrane [37] . Thus , it is likely that telomere-mediated chromosome movement is also important for meiotic prophase I in Arabidopsis . Strongly supporting this hypothesis , the two Arabidopsis SUN-domain proteins were recently shown to be essential for completion of synapsis and normal CO formation ( S . J . A , unpublished data ) . The rice Kinesin1-like protein PSS1 has been shown to be essential for fertility and normal chromosome segregation at meiosis [38] , but its potential function in synapsis and recombination was not investigated . Here we identified the AtPSS1 Kinesin1-like protein as a major actor in meiosis , promoting synapsis and regulating CO formation in Arabidopsis . Our data suggest that the movement of AtPSS1 along microtubules generates cytoplasmic forces which could be transmitted to the chromosomes via a KASH-SUN module and coordinate synapsis and CO distribution . A previous report showed that mutation of the rice class I kinesin I ( named OsPSS1 ) leads to meiotic defects [38] . Reciprocal BLAST analysis and comprehensive sequence analysis of plant kinesins [39] unambiguously identified the product encoded by the Arabidopsis At3g63480 gene as the only putative orthologue of OsPSS1 . The two proteins share high amino acid sequence identity ( 59% ) . We identified three T-DNA insertion lines from the public collections: Atpss1-1 , Atpss1-2 and Atpss1-3 . Insertion of the TDNA in these loci was confirmed by sequencing the flanking sequences ( Figure 1 ) . Homozygous plants for all three lines have the same phenotype: normal vegetative growth but decreased fertility , as shown by reduced seed set ( 55±6 seeds per silique for wild type versus 27±5 for Atpss1-1 ) and reduced pollen viability ( Alexander staining , Figure S1 ) . Heterozygote plants for two Atpss1 mutations had the same phenotype showing that the three mutants are allelic . Transformation of the Atpss1-1 mutant with a 5 kb genomic region containing the AtPSS1 coding and regulatory sequences restored pollen viability ( 7 independent transformants , Figure S1 ) , confirming that the observed defects are due to disruption of the AtPSS1 gene . We used chromosome spreads to investigate male meiosis defects in the Atpss1-1 mutant . Wild-type Arabidopsis meiosis was described in detail in [40] , and the major stages are summarized in figure 2 . At leptotene chromosomes appear as thin threads ( Figure 2A ) , synapsis ( the close association of two chromosomes via an SC ) begins at zygotene and is complete by pachytene ( Figure 2B ) . The SC is then depolymerized at diplotene and chromosomes condense so that the five bivalents are visible ( pairs of homologues connected by COs ) ( Figure 2C ) . The bivalents align at metaphase I ( Figure 2D ) , and chromosomes separate from their homologue at anaphase I leading to the formation of two pools of five chromosomes and two nuclei ( Figure 2E ) . At the second meiotic division , the pairs of sister chromatids align on the two metaphase plates , and separate at anaphase II to generate four pools of five chromosomes , which gives rise to tetrads of four microspores ( Figure 2F ) . In Atpss1 mutants , leptotene and zygotene appeared similar to those in wild type ( Compare figure 2A and 2G ) . Accordingly immunolocalization of two axial element proteins , ASY3 [41] and the hormad domain containing protein ASY1 [42] did not reveal any difference between Atpss1 and wild type ( Figure S2 ) . However we were unable to find a typical pachytene stage among chromosome spreads of the Atpss1 mutant ( n>300 ) , as only partial synapsis was observed ( Figure 2H ) . Synapsis was further examined by immunolocalization of REC8 and ZYP1 [43] , which are chromosome axis and SC central element proteins , respectively ( Figure 3 ) . In flower buds whose size corresponds to late pachytene/diplotene stages , most wild-type cells showed almost complete synapsis , with the ZYP1 signal covering completely the REC8 signal ( Figure 3A ) . In contrast , we were unable to find any meiocytes in which SC had undergone complete polymerization ( n = 102 ) in Atpss1 . Atpss1 cells showed various levels of incomplete synapsis ( Figure 3B ) , ranging from 4 to 91% , with less than half of the REC8 axis being covered with ZYP1 signal in most cells ( distribution shown on Figure 3C ) . The observed partial ZYP1 loading could be the result of either delayed synapsis or failure in completing synapsis . However , the observation of diplotene stages on the same slides favors the hypothesis of incomplete synapsis ( see also below ) . At diakinesis and metaphase I , a mixture of univalents and bivalents ( on average 3 . 1±1 . 2 bivalents and 1 . 9±1 . 2 univalent pairs ) was observed in each Atpss1 allele ( Figure 4 ) , contrasting with wild type which always has five bivalents ( Figure 2D , J and 5 ) . FISH experiments using probes directed against 45S , 5S rDNA and the F8J2 BAC that allow the identification of the five Arabidopsis Col-0 chromosomes as described in [44] , suggested that each chromosome is affected in bivalent formation ( The univalent frequency for chromosomes 1 to 5 were respectively 28% , 37% , 42% , 42% and 26% . N = 43 Atpss1-1 cells ) . The presence of univalents resulted in missegregation of chromosomes in anaphase I and a subsequent aberrant number of daughter cells and/or unbalanced chromosome distribution ( Figure 2K , L ) . Overall , our results showed that AtPSS1 is required for full synapsis and normal levels of bivalent formation at male meiosis . Observation of pistils [45] , showed that 52% of the Atpss1 female gametophytes were defectives ( n = 150 ) . Further , univalents were detected at metaphase I of female meiosis ( Figure S3 ) , showing that AtPSS1 is essential for normal levels of bivalent formation in both male and female meiocytes . The presence of bivalents in Atpss1-1 implies that CO formation is not completely impaired in this mutant . The nature of the COs produced in the absence of AtPSS1 was investigated by epistasis tests with zmm and mus81 mutants , which are defective in class I and class II CO formation , respectively . Mutation of a ZMM in Atpss1 reduced bivalent formation from 3 . 1±1 . 2 to 0 . 3±0 . 4 , showing that most of the COs produced in the Atpss1 mutant are ZMM dependent . We then used MLH1 immunolocalization , a marker of class I COs , to explore CO distribution in Atpss1 . The total number of MLH1 foci per cell during diplotene and diakinesis was similar in Atpss1 ( 11 . 9±2 . 7 and 10 . 2±2 . 3 ) and wild type ( 11 . 1±1 . 7and 10 . 5±1 . 5 ) ( Figure 5 ) . However , we found that the distribution of MLH1 foci among chromosomes was significantly affected in the Atpss1 mutant , as shown in figure 5 . In wild type , 62% of the bivalents had exactly two MLH1 foci , 20% had three , 15% had one and less than 3% had four foci . In contrast , the number of MLH1 foci per chromosome was much more variable in Atpss1 , with the appearance of classes not observed in wild type ( Figure 5E ) . One quarter of the chromosome pairs appeared as univalents without MLH1 foci , fitting with the observed frequency of univalents at metaphase I , while bivalents with more than three foci were more frequent than in wild type ( 19 . 4% vs 2 . 7% ) . This suggests that CO distribution but not frequency is affected in Atpss1 . Measurements of recombination rates in six genetic intervals using pollen tetrad analysis [46] showed that CO frequency is not reduced but even slightly higher in Atpss1 ( Figure 6 , Table S1a and Table S1b ) . CO interference , measured genetically , was significantly reduced compared to wild type , to a level no longer detected ( Table S1b ) . While we cannot formally exclude that a low level of interference exists , this clearly establish that CO interference measured genetically is decreased in Atpss1 . This further suggests that relative CO distribution is disturbed in Atpss1 . Overall , the above data showed that synapsis is incomplete and CO distribution among chromosomes is affected in Atpss1 mutants . As both synapsis and COs are promoted by DSB formation and repair , we carried out immunolocalization studies with DMC1 , a protein which marks DSBs undergoing repair . In Atpss1 , DMC1 foci decorated all chromosome axes and their total number was higher compared to wild type ( +37% . 204±6 vs 279±8 . T-test p = 3 . 5 . 10−10 ) , suggesting that in the mutant DSB formation is enhanced or that DMC1 foci accumulate due to slower turnover ( Figure 7 ) . Thus in the mutant DSBs appear to occur on all chromosomes . We then examined whether the chromosome regions where COs occurred and that synapsed were the same . Because synapsis disappears before MLH1 foci numbers peak in Arabidopsis [47] , we used HEI10/ZYP1 co-immunolocalization to explore this question ( Figure 8 and S4 ) . Indeed , HEI10 marks recombination progression from numerous faint foci at leptotene ( Figure 8A ) to about ten large foci labeling class I CO sites from late pachytene ( Figure 8C ) to diakinesis ( Figure S4 ) [48] . At leptotene , Atpss1 and wild-type cells were indistinguishable with numerous small HEI10 foci ( Figure 8A and 8D ) , further suggesting that early recombination events are unaffected in the mutant . At early wild type pachytene , numerous foci of variable size are dispersed on the SC ( Figure 8B ) . At the same stage in Atpss1 , the synapsed regions were also decorated with numerous HEI10 foci , but the regions that failed to synapse were foci-free . At late pachytene , a small number of bright and homogeneous foci were observed in both wild type and the mutant ( Figure 8C , F and G ) . Remarkably , while the total length of the SC in Atpss1 pachytene cells was on average one third that of wild type , confirming partial synapsis , the average number of HEI10 foci per cell was unaffected ( Wild type: 10 . 3±1 . 9 , Atpss1: 11 . 2±1 . 2 , p = 0 . 19 ) ( Figure 8H ) . Accordingly , the number of HEI10 foci per 100 µm of SC was on average 3 . 1±0 . 7 for wild type and 10 . 9±4 . 8 for Atpss1 ( these measurements were made on a cell per cell basis , because the entanglement of Arabidopsis pachytene chromosomes makes it difficult to unambiguously follow individual SCs ) . While the density of HEI10 foci was relatively stable in wild type ( from 2 to 4 . 3 per 100 µm ) , it varied greatly in Atpss1 ( from 4 . 3 to 23 . 6 per 100 µm ) ( Figure 8H ) . This is strikingly illustrated by the extreme case shown in figure 8G , where seven HEI10 foci can be seen on a single 30 µm SC stretch . At diplotene and diakinesis , the number of HEI10 foci per cell was similar and stable in the wild type and mutant . However , consistent with the MLH1 data , the distribution of HEI10 foci among chromosomes was significantly modified in Atpss1 ( Figure S4 ) , confirming that CO distribution but not number is affected . In summary , in Atpss1 , COs and synapsis are jointly restricted to the same limited portion of the genome . Partial synapsis is accompanied by an increase in CO density per SC unit , resulting in –or caused by ( see discussion ) - an unaffected number of COs per cell . The MUS81 pathway ( Class II pathway ) is minor in Arabidopsis wild type . Its disruption reduces CO frequency by ∼10% , but does not affect bivalent formation [49] , [50] ( Figure 4 ) . Mutation of MUS81 in the Atpss1 background did not further reduce bivalent frequency ( Figure 4 ) , which is consistent with the conclusion above that most COs are ZMM-dependent in Atpss1 . At FANCM was previously shown to limit MUS81-dependant CO formation and bivalent formation is fully restored in zmm/Atfancm mutants due to a massive increase in class II COs [22] . Mutation of AtFANCM in Atpss1 did not increase the number of bivalents , suggesting that it did not restore CO formation in regions that are CO incompetent in the single Atpss1 mutant ( but this does not exclude that there is an increase in CO frequency in regions that are CO competent ) ( Figure 4 ) . However while bivalent formation was very low in Atpss1 Atzip4 , bivalent formation was restored in the Atpss1 Atfancm Atzip4 triple mutant back to the level observed in the single Atpss1 mutant ( Figure 4 ) . Altogether , these results suggest that , in Atpss1 , class II COs occur at a low frequency , and can be promoted by mutating AtFANCM but exclusively in regions that are also already competent for class I CO formation . AtPSS1 , which belongs to the kinesin family , appears to play a crucial role in meiosis . Kinesin proteins are characterized by their ability to walk on microtubules via a motor domain that uses ATP to promote repetitive conformation changes [51] . We thus tested if the motor function of AtPSS1 is important for its function in meiosis . For this , we expressed an AtPSS1 protein modified in the conserved arginine ( Arg-293>His ) that was previously shown to abolish the microtubule-stimulated ATPase activity [38] in the Atpss1-1 mutant . When Atpss1 plants were transformed with the control wild-type AtPSS1 gene , pollen viability and bivalent formation at metaphase I were fully restored ( 7 independent transformants ) . In contrast , transformation with AtPSS1-R293H , expressed behind the native AtPSS1 promoter , did not restore pollen viability and normal meiosis ( 4 independent transformants , see methods; Bivalent frequency: 4 . 2±1 ( n cells = 35 ) , 4 . 3±0 . 5 ( n = 6 ) , 3 . 5±0 . 6 ( n = 4 ) , 3 . 5±1 . 2 ( n = 15 ) , respectively ) , showing that the kinesin function of AtPSS1 is critical for its role in meiosis . In several model species , cytoskeleton-based forces were previously shown to be important for meiosis and to be transduced to the nucleus by KASH- and SUN-domain containing proteins [24] , [25] . In Arabidopsis , two SUN proteins were recently shown to be redundant and important for meiosis ( S . J . A . under review ) . As in Atpss1 , a mixture of bivalents and univalents are observed in Atsun1 Atsun2 double mutants . This defect is quantitatively identical in the Atpss1 , Atsun1 Atsun2 and the Atsun1 Atsun2 Atpss1 triple mutants ( Figure 4 ) , suggesting that SUN proteins and AtPSS1 may act in the same pathway . WIP1-3 proteins were also recently identified as KASH containing proteins in Arabidopsis , and shown to interact with SUNs [52] . This raised the possibility that AtPSS1 could be involved in transmitting forces to the meiotic nucleus via a WIP-SUN module . Yeast two-hybrid experiments showed that AtPSS1 interacts directly with WIP1 and WIP2 . The AtPSS1-WIP1 but not the AtPSS1-WIP2 interaction was confirmed by BiFC assays ( Figure S5 ) . The yeast two-hybrid also confirmed that WIPs interact with SUNs , as previously shown [52] ( Figure 9 ) . During meiotic prophase I , chromosome movements within the intact nucleus are prominent and have been shown to be involved in chromosome pairing , synapsis and recombination in a variety of species . Here we showed that AtPSS1 , the Arabidopsis kinesin-1 like protein [39] , is essential for full synapsis and is required for proper CO distribution . Furthermore , the bivalent shortage is identical when AtPSS1 , SUNs or both , are knocked out suggesting that SUNs and AtPSS1 act in the same pathway to regulate CO formation . In addition , AtPSS1 interacts with the KASH-domain proteins WIP1 and WIP2 which themselves interact with SUN proteins [52] . Finally , we showed that the kinesin motor domain of AtPSS1 is required for its meiotic function . Kinesin is a motor protein which walks along microtubules with high processivity and for long distances ( reviewed in [51] ) . We thus speculate that AtPSS1 moves along microtubules and generates forces that are transduced via a SUN-WIP module through the nuclear membrane to the chromosomes , promoting synapsis and regulating CO distribution ( see below ) . The proteins that would connect SUNs to the chromosome telomeres remain to be identified . These results add to a growing amount of evidence showing that the transduction of cytoplasmic forces through the nuclear membrane is an important and conserved promoter of meiotic recombination . It should be noted here that the function we propose for AtPSS1 appears to be fulfilled by dynein in many organisms , and that dynein is absent from flowering plant genomes [53] . The rice PSS1 is also essential for normal meiosis [38] . Even though recombination and synapsis have not been extensively analyzed in the rice Ospss1 mutants , univalent were observed at metaphase I , suggesting that the primarily defects may be similar to Atpss1 . This suggests that the meiotic function of AtPSS1 is conserved among flowering plants . We showed that AtPSS1 is required for full synapsis and normal CO formation . In most species , the search for homologous sequences by recombinase-coated 3′-ssDNA promotes both CO formation and homologous synapsis . Indeed , in Arabidopsis both COs and synapsis are absent in mutants affecting DSB formation , but also homologous sequence invasion ( RAD51 , DMC1 and their co-factors ) [12] , [14] , [18] . This appears to be a cooperative process as multiple repair events are required for initiation and progression of synapsis [54] , [55] . Atpss1 mutants have a novel defect: in each cell , COs and synapsis take place on only a subset of the genome ( which varies from 10 to 90% ) . Initial DSB formation and processing do not appear to be involved in these defects , as DMC1 foci and early HEI10 are present on all chromosomes in the mutant . The number of DMC1 foci was higher in the mutant than wild type , possibly reflecting a delay in recombination progression . The increased number of DMC1 foci may also reflect an increase of the number of DSBs in response to the downstream defects [56] . However , we suggest that only a subset of these DSBs is efficiently matured into potential CO precursors and promoters of synapsis . This is supported by the observation that the segments of chromosomes which were seen to synapse were also the places where early HEI10 foci progressively matured into intermediate and then late CO-marking-foci . This model implies that chromosome movement involving AtPSS1 is required to efficiently mature DMC1-coated-DSBs into CO/synapsis precursors . This movement could be simply required for the homology searching DNA “tentacle” [57] to reach the homologous chromosome which can be at some distance in the nucleus [58] . Alternatively , the movement may be required to resolve the entanglement/clutter/interlocking which likely arises from multiple chromosome pairing attempts in the limited space of the nucleus [55] . The DSBs present on the portions of chromosomes which failed to reach homologues are likely repaired using the sister chromatids as template , thus failing to promote synapsis and homologous CO . Such sister-mediated repair occurs genome-wide in haploid Arabidopsis , where DMC1-coated resected DSBs are repaired on the sister , or in diploid mutants where DMC1 or one of its partners is absent [12]–[14] , [59] . One intriguing feature of the Atpss1 mutant is that CO frequency per cell is not reduced , but instead the subparts of the chromosomes that do synapse and recombine make a similar total number of COs per cell as in wild type . This is strikingly shown in figure 8G , where a single SC stretch was formed in a cell on which seven class I COs occurred , while CO number rarely exceeds four on an entire wild-type chromosome . The smaller size of the competent regions appears to be compensated by an increased CO density , which implies that interference is no longer acting or that the distance at which interference spreads is reduced . Unfortunately , the difficulty in following individual SCs prevented us from cytologically measuring CO interference . The stable number of COs per cell in Atpss1 could reflect a form of CO homeostasis , which is defined as the tendency to preserve CO number despite a variation in DSB number through a modulation of the probability for DSBs to become COs [5] . We suggest that such homoeostasis applies in the Atpss1 mutant , and that the decrease in the number of CO-competent DSB is compensated for by an increased probability of the eligible DSB becoming a CO . It is possible that the total number of COs per cell is defined , and then ∼10 COs per cell occurs on licensed regions . However , the mechanism that would count the number of COs per cell remains elusive . Alternatively , we suggest that a feed-back loop could sense some unachieved event ( e . g . the presence of chromosomes lacking COs , or incomplete synapsis ) , and then increase the propensity of precursors to be designated for CO . This feed-back loop would therefore modulate the parameters of interference ( possibly by a progressive increase in CO-promoting mechanical stress or progressive increase in the sensitivity of precursors to this stress [60] , [61] ) . Finally , AtPSS1 could have a dual function , on one hand promoting synapsis and recombination intermediate maturation , and on the other preventing an excess of COs on selected regions , both via chromosome movement [24] . Col-0 lines were obtained from the collection of T-DNA mutants from the Salk Institute Genomic Analysis Laboratory ( Columbia accession ) ( SIGnAL , http://signal . salk . edu/cgi-bin/tdnaexpress ) and provided by NASC ( http://nasc . nott . ac . uk/ ) . Mutant alleles used in this study were: Atmsh5-2 ( SALK_026553 ) [62]; Atzip4-2 ( SALK_068052 ) [63]; Atmus81-2 ( SALK_107515 ) [49] , [50] , Atfancm-1 [22]; Atsun1 ( SAIL_84_G10 ) ; Atsun2 ( FLAG_026E12 ) . Details for all genotypes , primers used and PCR amplification conditions are shown in table S2 . Plants were cultivated in a greenhouse or growth chamber under the following conditions: photoperiod 16 h/day and 8 h/night; temperature 20°C day and night; humidity 70% . The six intervals tested in this study correspond to intervals I1b and I1c ( both located at the top of chromosome 1 ) , I2a and I2b ( both located at the bottom of chromosome 2 ) , and I5c and I5d ( both located at the top of chromosome 5 ) described in [64] . Tetrad analyses were carried out as described in [46] . The resulting tetrad data ( Table S1a ) were analyzed using the Perkins mapping equation . All double mutants were obtained by crossing plants , which were heterozygous for each mutation . The resulting hybrids were self-pollinated . PCR screening was then used to identify plants in the F2 progeny that were homozygous for both mutations . The anti-ASY1 polyclonal antibody was described by [42] . The anti-ZYP1 polyclonal antibody was described by [43] . The anti-DMC1 antibody was described in [63] , the anti-MLH1 antibody in [47] , and the anti-HEI10 in [48] . The anti-REC8 polyclonal antibody was described in [65] . Chromosome spreads of male meiocytes were prepared and stained with DAPI as described in [40] . Chromosome spreads for immunocytology was performed according to [42] . Observations were made using a Leica ( http://www . leica . com ) DM RXA2 microscope or a Zeiss ( http://www . zeiss . fr ) Axio Imager 2 microscope; photographs were taken using a CoolSNAP HQ ( Roper , http://www . roperscientific . com ) camera driven by OpenLAB 4 . 0 . 4 software or a Zeiss camera AxioCam MR driven by Axiovision 4 . 7 . All images were further processed with OpenLAB 4 . 0 . 4 , Axiovision 4 . 7 , or AdobePhotoshop 7 . 0 ( http://www . adobe . com ) . The AtPSS1 , AtWIP1 , AtWIP2 , AtWIP3 , AtSUN1 and AtSUN2 open reading frames were amplified from Arabidopsis cDNA clones ( Columbia ecotype ) using specific primers flanked by the AttB1 and AttB2 sites ( Table S2 ) , cloned into Gateway vector pDONR207 using BP recombination ( Invitrogen ) , and sequenced . Expression vectors were obtained after LR recombination ( Invitrogen ) between these entry vectors and destination vectors ( pGADT7-GW and pGBKT7-GW for Y2H , and pBiFP vectors for BIFC ) . Yeast two-hybrid interactions were tested using AtPSS1 , AtWIP1 , ATWIP2 , AtWIP3 , AtSUN1 and AtSUN2 as bait ( pGADT7-GW ) or as prey ( pGBKT7-GW ) by mating with the AH109 and Y187 yeast strains . For fluorescence complementation tests , transient expression of all eight compatible combinations between protein pairs ( i . e . , providing both parts of the YFP ) was assayed . Each expression vector was introduced into Agrobacterium tumefaciens strain C58C1 ( pMP90 ) by electroporation . Agrobacterium bacterial cultures were incubated overnight at 28°C with agitation . Each culture was pelleted , washed , and resuspended in infiltration buffer ( 13 g/L bouturage N°2 medium [Duchefa Biochemie] and 40 g/L sucrose , pH 5 . 7 ) to an OD600 of 0 . 5 . The inoculum was delivered to the lamina tissue of N . benthamiana leaves by gentle pressure infiltration through the lower epidermis . To enhance transient expression of BiFC fusion proteins , the P19 viral suppressor of gene silencing was coexpressed [66] . YFP fluorescence was detected three days after infiltration . Tissue was mounted in low-melting-point agarose ( 0 . 4% in water ) and viewed directly using an inverted Zeiss Observer Z1 spectral confocal laser microscope LSM 710 using a C-Apochromat ×63/1 . 20 W Corr objective ( Carl Zeiss ) . Fluorescence was recorded after an excitation at 514 nm ( Argon laser ) and using a selective band of 514 to 568 nm . A 5 kb AtPSS1 genomic fragment containing 1 . 5 kb of promoter region and the complete AtPSS1 gene was amplified using specific primers flanked by AttB1 and AttB2 sites ( Table S2 ) , cloned into Gateway vector pDONR207 using BP recombination ( Invitrogen ) , and sequenced . Directed mutagenesis was performed using the Quickchange Site-Directed Mutagenesis Kit ( Stratagene ) . The mutagenic primers used to generate the AtPSS1-R293H ( Arg codon cgc→His codon cac ) are shown in Table S2 . A LR reaction between the resulting vectors and the pGWB1 destination binary vector was performed .	In species that reproduce sexually , diploid individuals have two copies of each chromosome , inherited from their father and mother . During a special cell division called meiosis , these two sets of chromosomes are mixed by homologous recombination to give genetically unique chromosomes that will be transmitted to the next generation . Homologous recombination processes are highly controlled in terms of number and localization of events within and among chromosomes . Disruption of this control ( a lack of or improper positioning of homologous recombination events ) causes deleterious chromosome associations in the offspring . Using the model plant Arabidopsis thaliana we reveal here that the AtPSS1 gene is required for proper localization of these homologous recombination events along the genome . We also show that AtPSS1 , which belongs to a family of proteins able to move along the cytoskeleton , is likely part of a module that allows cytoplasmic forces to be transmitted through the nucleus envelope to promote chromosome movements during homologous recombination progression .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "biochemistry", "cytoskeletal", "proteins", "meiosis", "molecular", "motors", "kinesins", "cell", "biology", "synapsis", "proteins", "microtubule", "motors", "cell", "cycle", "and", "cell", "division", "genetics", "biology", "and", "life", "sciences", "dna", "cell", "processes", "dna", "recombination", "meiotic", "prophase", "homologous", "recombination" ]	2014	The Kinesin AtPSS1 Promotes Synapsis and is Required for Proper Crossover Distribution in Meiosis
Microbial translocation ( MT ) is the process by which microbes or microbial products translocate from the intestine to the systemic circulation . MT is a common cause of systemic immune activation in HIV infection and is associated with reduced frequencies of CD4+ T cells; no data exist , however , on the role of MT in intestinal helminth infections . We measured the plasma levels of MT markers , acute-phase proteins , and pro- and anti - inflammatory cytokines in individuals with or without hookworm infections . We also estimated the absolute counts of CD4+ and CD8+ T cells as well as the frequencies of memory T cell and dendritic cell subsets . Finally , we also measured the levels of all of these parameters in a subset of individuals following treatment of hookworm infection . Our data suggest that hookworm infection is characterized by increased levels of markers associated with MT but not acute-phase proteins nor pro-inflammatory cytokines . Hookworm infections were also associated with increased levels of the anti – inflammatory cytokine – IL-10 , which was positively correlated with levels of lipopolysaccharide ( LPS ) . In addition , MT was associated with decreased numbers of CD8+ T cells and diminished frequencies of particular dendritic cell subsets . Antihelmintic treatment of hookworm infection resulted in reversal of some of the hematologic and microbiologic alterations . Our data provide compelling evidence for MT in a human intestinal helminth infection and its association with perturbations in the T cell and antigen-presenting cell compartments of the immune system . Our data also reveal that at least one dominant counter-regulatory mechanism i . e . increased IL-10 production might potentially protect against systemic immune activation in hookworm infections . Microbial translocation ( MT ) is the process by which microbes or microbial products—such as lipopolysaccharide ( LPS ) and bacterial DNA—translocate from the intestinal lumen to the systemic circulation in the absence of overt bacteremia [1] . Activation of Toll-like receptors by LPS is then thought to lead to systemic immune activation [1] . LPS and 16 s ribosomal RNA ( common to most bacteria ) are often used as indicators of MT , while soluble CD14 ( sCD14 ) and LPS-binding protein ( LBP ) are used to establish evidence of direct LPS stimulation [1] , [2] . Presence of anti-LPS core antibodies ( Endo core LPS antibody , or EndoCAb ) is also used as a surrogate measure of circulating LPS [1] , [2] . MT is commonly observed in conditions associated with disruption of the gastrointestinal ( GI ) epithelial barrier such as inflammatory bowel disease , graft-versus-host disease , and chronic viral infections including human-immunodeficiency virus ( HIV ) and hepatitis C virus [1] , [2] . Although MT is known to occur in infections affecting the integrity of the gut epithelium [3] , [4] , very few studies have examined the occurrence of this phenomenon in intestinal helminth infections . Hookworm infections are common intestinal helminth infections ( affecting 740 million people worldwide ) known to cause intestinal injury and blood loss [5] . Hookworm infection in humans is caused by the helminth parasites Necator americanus and Ancylostoma duodenale . Infection is acquired by entry of infective-stage larvae through the skin during contact with contaminated soil , followed by larval migration through the heart and lungs and subsequent development into adults in the GI tract . The adults then produce eggs , which are deposited in the feces and develop into infective larvae , completing the life cycle . The host must therefore mount an immune response against a number of different life-cycle stages during a hookworm infection [5] , [6] . In addition , due the chronic nature of this infection , the parasite has been postulated to manipulate the host immune system to establish long-standing infection [6] . MT is also commonly associated with acute and chronic systemic immune activation and perturbations in T cell subset numbers [1] , [2] . Thus , in HIV infection , circulating LPS is associated with increased secretion of proinflammatory cytokines and decreased frequencies of CD4+ T cells as well as the selective loss of Th17 ( CD4+IL-17+ ) T cells [7] , [8] . To explore the relationship of MT , innate and adaptive immune homeostasis , and immune activation with hookworm infection , we measured markers of MT and acute-phase proteins , pro- and anti- inflammatory cytokines along with CD4+ , CD8+ T cell , NK cell , and B cell numbers as well as frequencies of CD4+ and CD8+ T cell and dendritic cell ( DC ) subsets in hookworm-infected ( INF ) and uninfected ( UN ) individuals . Our study provides evidence for the occurrence of MT in hookworm infection associated with perturbations in immune cell compartments . Our study also reveals that MT in hookworm infections ( unlike HIV ) does not directly translate to systemic immune activation , perhaps due to counter-regulatory measures , such as increased IL-10 production , induced by the parasite . We prospectively studied a group of 46 INF and 45 UN individuals in an area endemic for hookworm infections in Tamil Nadu , South India ( table 1 ) . This was a community-based study , and all individuals were recruited from the same village in Tamil Nadu and were of similar socio-economic status . Blood ( total volume of 10 ml with or without anti-coagulants ) and stool samples were collected from all recruited individuals within the same time period . All INF individuals were treated with a single dose of albendazole ( 400 mg ) . Follow-up blood and stool samples were obtained from 30 of the treated individuals 3 months following treatment . All individuals were examined as part of a clinical protocol approved by Institutional Review Boards of both the National Institute of Allergy and Infectious Diseases and the National Institute for Research in Tuberculosis ( NCT00375583 ) , and informed written consent was obtained from all participants . Single stool samples were collected , transported to the laboratory at ambient temperatures , and examined by direct microscopy and by formal-gasoline concentration techniques , as described previously [9] . Stool microscopy was used to exclude the presence of other intestinal helminths including Ascaris , Strongyloides , Trichuris , Enterobius , Taenia and Hymenolepis . Concomitant filarial infection was excluded by the TropBio Og4C3 enzyme-linked immunosorbent assay ( ELISA ) ( Trop Bio Pty . Ltd , Townsville , Queensland , Australia ) . Hematology was performed on all patients using the Act-5 Diff hematology analyzer ( Beckman Coulter , Brea , CA , USA ) . To inactivate plasma proteins , plasma samples were heated to 75°C for 5 min . LPS levels were measured using a limulus amebocyte lysate assay ( Cell Sciences Hycult Biotech , Canton , MA , USA ) according to the manufacturer's protocol . Commercially available enzyme-linked immunosorbent assay ( ELISA ) kits were used to measure plasma levels of lipid-binding protein ( LBP ) , endotoxin core antibodies IgG ( EndoCAb ) , intestinal fatty acid binding protein ( IFABP ) , ( all Cell Sciences Hycult Biotech ) , and sCD14 ( R&D Systems , Minneapolis , MN , USA ) . Plasma levels of C-reactive protein ( CRP ) , haptoglobin , serum amyloid A ( SAA ) , and α-2 macroglobulin ( α-2M ) were measured using the Bioplex multiplex ELISA system ( Bio-Rad , Hercules , CA , USA ) according to the manufacturer's instructions . Plasma levels of cytokines , TNF-α , IFN-γ , IL-12 , IL-17 and IL-10 ( Bio-Rad ) were measured using the Bioplex multiplex ELISA system in a subset of samples . Flow cytometry acquisition was done on BD FACS Canto II ( BD Biosciences , San José , CA , USA ) . Analysis was done using FlowJo software v9 . 4 . 10 ( TreeStar Inc . , Ashland , OR , USA ) . T , B and NK cells were enumerated in whole blood using BD Multiset 6-Color TBNK cocktail ( BD Biosciences ) . Naïve and memory T cell phenotyping ( Figure S1 ) as well as dendritic cell subset phenotyping ( Figure S2 ) were performed using lineage specific antibodies ( BD Pharmingen and eBioscience ) . Data analyses were performed using GraphPad PRISM ( GraphPad Software , Inc . , San Diego , CA , USA ) . All samples were tested in duplicate by ELISA . Geometric means ( GM ) were used for measurements of central tendency . Statistically significant differences were analyzed using the nonparametric Mann-Whitney U test and Wilcoxon matched pair test . Multiple comparisons were corrected using the Holm's correction for each set of analysis . Correlations were calculated by the Spearman rank correlation test . As shown in table 1 , INF individuals differed from UN individuals in exhibiting significantly lower hemoglobin levels ( P<0 . 0001 ) , hematocrit ( P = 0 . 0045 ) and red blood cell counts ( P = 0 . 0016 ) . In contrast , INF individuals had significantly increased numbers of eosinophils ( P<0 . 0001 ) and total white blood cell counts ( P = 0 . 0086 ) . To determine the association of MT and related markers with hookworm infection , we measured the plasma levels of LPS , LPB , sCD14 , EndoCAb , and IFABP in INF and UN individuals . As shown in figure 1 , INF had significantly higher levels of LPS ( GM of 271 EU/ml in INF vs . 122 . 8 in UN; P = 0 . 0156 ) , sCD14 ( GM of 12 . 8 ng/ml in INF vs . 8 . 8 in UN; P = 0 . 0045 ) , EndoCAb ( GM of 1132 GMU/ml in INF vs . 257 . 5 in UN; P<0 . 0001 ) , and IFABP ( GM of 141 . 8 pg/ml in INF vs . 58 . 5 in UN; P = 0 . 0054 ) in comparison to UN . Thus , hookworm infection is associated with elevated circulating levels of molecules often associated with MT . To determine the association of acute-phase proteins with hookworm infection , we measured the plasma levels of α-2M , CRP , haptoglobin , and SAA in INF and UN individuals . As shown in figure 2A , INF had significantly lower levels of CRP ( GM of 0 . 95 ng/ml in INF vs . 1 . 7 in UN; P = 0 . 0224 ) and haptoglobin ( GM of 8 . 1 ng/ml in INF vs . 13 . 5 in UN; P = 0 . 0113 ) but not α-2M and SAA in comparison to UN . Similarly , to determine the association of pro-inflammatory cytokines with hookworm infection , we measured plasma levels of TNF-α , IFN-γ , IL-12 , and IL-17 . As shown in figure 2B—with the exception of IL-17 , which was significantly decreased in INF ( GM of 390 pg/ml in INF vs . 682 . 8 in UN; P<0 . 0001 ) —no significant alterations in plasma levels of pro-inflammatory cytokines were observed in hookworm infection . Conversely , INF had significantly higher levels of IL-10 compared to UN ( GM of 689 pg/ml in INF vs . 404 in UN; P<0 . 0001[figure 2C] ) . Moreover , there was a significant correlation between plasma levels of LPS and those of IL-10 ( r = 0 . 520; P = 0 . 0002 ) indicating that individuals with increased LPS levels also had increased levels of IL-10 . Thus , hookworm infection was not associated with either acute-phase protein elevation or elevation of pro-inflammatory cytokines; there was however concomitant increases in IL-10 , which could mitigate the systemic immune activation often associated with MT . Because MT is associated with decreased CD4+ T cell counts in HIV infection [7] , we sought to determine the relationship between hookworm infection and/or MT with the numbers of CD4+ and CD8+ T cells , NK cells , and B cells . As shown in figure 3A , INF had significantly lower numbers of total CD8+ T cells ( GM of 624 cells/µl in INF vs . 743 in UN; P = 0 . 0052 ) but not total CD4+ T cells , NK cells , or B cells in comparison to UN . To determine the association of MT with alterations in T cell subsets , we measured the frequencies of different CD4+ and CD8+ T cell subsets in hookworm infection . As shown in figure 3B and C , we observed a significant decrease in percentages of effector memory CD4+ ( CD45RA−CCR7−; GM of 28 . 3% in INF vs . 42 . 8 in UN; P = 0 . 0121 ) and CD8+ ( CD45RA−CCR7−; GM of 11% in INF vs . 22 . 5 in UN; P = 0 . 0051 ) T cells but not central memory ( CD45RA−CCR7+ ) or naïve T cells ( CD45RA+CCR7+ ) or nTregs ( CD4+CD25+Foxp3+CD127dim ) . We also sought to determine whether hookworm infection is also associated with perturbations in the antigen-presenting cell compartment and hence determined the frequency of the DC subsets . As shown in figure 3D , INF had significantly lower frequency of pDCs ( Lin−HLA-DR+ CD123+; GM of 0 . 56% in INF vs . 1 . 4 in UN; P<0 . 0001 ) as well as mDCs ( Lin−HLA-DR+ CD11c+; GM of 3 . 3% in INF vs . 5 . 8 in UN; P = 0 . 0035 ) in comparison with UN . Thus , hookworm infection is associated with alterations in homeostatic levels of total CD8+ T cells , the relative frequencies of both CD4+ and CD8+ effector memory T cells as well as baseline frequency of DC subsets . The relationships between circulating levels of LPS and EndoCAb and CD8+ T cells and DC subsets were next assessed in INF individuals ( figure 4 ) . As shown in figure 4A , levels of LPS exhibited a significant negative correlation with baseline CD8+ T cell counts ( r = −0 . 446; P = 0 . 0030 ) as well as with mDC ( r = −0 . 527; P = 0 . 0003 ) and pDC ( r = −0 . 450; P = 0 . 0028 ) frequency in INF . Similarly , EndoCAb levels were significantly negatively correlated with the baseline frequency of mDC ( r = −0 . 452; P = 0 . 0016 ) and pDC ( r = −0 . 429; P = 0 . 0036 ) ( figure 4B ) . No significant correlation was observed between MT markers and T cell subsets in INF individuals . In addition , no significant correlation was observed between these parameters in UN individuals ( data not shown ) . Thus , the process by which MT occurs appears to exhibit a significant negative association with CD8+ T cell numbers and mDC and pDC percentages in hookworm infection . Of the 46 INF individuals treated , 30 were able to be assessed at three months following treatment . 22/30 were found to be hookworm free at 3 months with the remaining 8 either still harboring infection or reinfected . As shown in figure 5A , treatment ( and resultant cure ) of hookworm infections caused a significant decrease in the circulating levels of LPS ( GM of 93 . 6 EU/ml following treatment vs . at 172 baseline; P = 0 . 0039 ) and sCD14 ( GM of 8 . 8 ng/ml vs . 9 . 7; P = 0 . 0348 ) compared to pre-treatment levels . In contrast , in those who failed to cure their hookworm infection there were no significant alterations in the levels of LPS , LBP or sCD14 between those found pre- and 3 months following albendazole therapy ( figure 5B ) . However , it is possible that the lack of statistical power in this group could lead to a lack of change in the markers investigated . Nevertheless , successful treatment of hookworm infection was associated with significant increases in pDCs ( GM of 0 . 79% following treatment vs . 0 . 42 at baseline; P = 0 . 0214 ) and mDCs ( GM of 5 . 4% vs . 2 . 4; P = 0 . 0021 ) at 3 months following anthelmintic therapy ( figure 5C ) . MT is characterized by translocation of microbial products from the intestinal lumen into the circulation and has been shown to occur as a consequence of disruption to the barrier function of the intestinal epithelium [1] . Studies in experimental animal models suggest that intestinal injury and systemic endotoxemia are two major factors leading to morbidity in helminth infections . Infection with the enteric nematodes Trichinella spiralis and Strongyloides venezuelensis has been shown to be associated with enhanced leakiness of the intestinal epithelium and translocation of LPS into the circulation in experimental animal models [3] , [4] . This leakiness—mediated in part by activated mast cells—can lead to movement of bacterial LPS into the portal circulation [3] , [4] . Even in non-intestinal helminth infections , such as Schistosoma mansoni , in which adult parasites reside in the mesenteric veins , damage caused by worm eggs traversing the GI epithelium can result in systemic translocation of bacteria [10] , [11] , [12]; however , no study has examined the role of MT in a human intestinal helminth infection . Because morbidity from hookworm infections is directly related to intestinal injury and blood loss caused by attachment of worms to the intestinal mucosa and submucosa [5] , hookworms are likely helminth candidates to induce MT . Therefore , the present study sought to elucidate the systemic effects of MT in hookworm infection . We examined five important circulating microbial or related products in our study . LPS ( a key indicator of MT ) was found to be significantly elevated in INF individuals . This was accompanied by a significant increase in levels of sCD14 , EndoCAb , and IFABP . Although increased levels of LBP are a common feature of MT in other infections [7] , [13] , [14] , we did not observe any significant alteration in LBP levels in hookworm infections . LPS is commonly measured to assess quantitatively the degree to which MT and plasma LPS levels are directly associated with the degree of intestinal permeability following invasive GI surgery [15] . Similarly , sCD14 , a soluble receptor for LPS produced by monocytes and macrophages , is often increased following MT and is felt to be an indicator of LPS stimulation in vivo [16] . Naturally occurring IgG antibodies to the LPS core oligosaccharide are potent neutralizers of LPS activity and are commonly elevated following systemic endotoxemia [17] , [18] . Finally , IFABP is an intracellular epithelial protein in the stomach and small and large intestinal mucosa and appears in the circulation after epithelial damage . Hence , plasma IFABP levels are considered useful markers for early diagnosis of intestinal ischemia [19] . Our data suggest that IFABP may also be a potentially reliable marker of the breach in epithelial integrity associated with chronic intestinal infections . We have not performed a quantitative assessment of the parasite burden in the stool of these individuals and some of the differences observed within the infected group could potentially reflect infection intensity , while some of the UN individuals may have had light infections . Nevertheless , our study groups did not differ significantly in age , gender , socio-economic status and were from the same geographical area , indicating that these confounding factors may not play a significant role . Circulating microbial products are well known inducers of acute-phase proteins , with SAA and haptoglobin known to be markedly elevated following challenge with LPS [20] . Acute-phase proteins derive primarily from the liver , and plasma concentrations are felt to be a reflection of the response to pro-inflammatory cytokines [21] . Because MT is commonly associated with systemic immune activation , we measured plasma levels of acute-phase proteins α-2M , CRP , haptoglobin , and SAA . Our data show that hookworm infections are not associated with significant elevations in acute-phase proteins . Indeed , CRP and haptoglobin levels were actually significantly lower in INF individuals . To confirm that MT in hookworm infections is not associated with systemic inflammatory responses , we also measured plasma levels of pro-inflammatory cytokines . Our examination of cytokine expression levels revealed that pro-inflammatory cytokines are not significantly elevated in INF individuals , again confirming the lack of a systemic inflammatory milieu associated with MT in hookworm infection . This is in agreement with the cytokine profiles observed in previous studies of experimental human infection [22] , [23] , [24] . Although studies in HIV infection reveal a direct association between levels of MT markers such as LPS and pro-inflammatory cytokines [7] , [13] , the lack of systemic pro-inflammatory cytokines may be reflective of counterbalancing by regulatory cytokines or Treg populations commonly seen in chronic helminth infection [25] , [26] . While increased levels of IL-10 production in hookworm infected individuals has been observed before [23] , [27] , our data on the increased levels of IL-10 in INF suggest that at least one potential mechanism by which systemic immune activation fails to occur in hookworm infections might be due to the heightened levels of IL-10 . Indeed , IL-10 is known modulator of LPS induced systemic immune responses and protects against lethal damage and septic shock [28] , [29] . Our data also suggest that natural regulatory T cells may not play a significant role in dampening inflammation in hookworm infections ( see Fig . 3B ) . Again , this is in contrast to a previous report that suggested that regulatory T cells are present at increased frequencies in hookworm infections [30] . In addition , other parasite dervived products , such as the production of proteins , termed “helminth defense molecules , ” that have been shown to actively inhibit LPS-induced inflammation [31] , might also contribute to the absence of overt systemic inflammation . MT is associated with perturbed CD4+ T cell homeostasis in HIV infection and idiopathic lymphocytopenia [7] , [32] . Although the exact mechanism by which CD4+ T cell depletion is associated with MT is still unclear , it is widely believed that depletion of the intraepithelial and lamina propria CD4+ T cells could disrupt the integrity of the intestinal epithelium [1] . Interestingly , in hookworm infections , we observed significant perturbations in not only the CD8+ T cell compartment but also in frequencies of specific DC subsets . Our data on the lack of difference in the circulating levels of CD4+ T cells is in contrast to previous studies that had shown lower levels of CD4+ T cells in INF individuals [33] , [34] . Our data therefore suggest an important association between MT and homeostatic levels of CD8+ T cells and DCs , as we demonstrated a significantly negative relationship between the levels of LPS/EndoCAb and the frequencies of CD8+ T cells and those of pDC and mDC . Moreover , we also identified the presence of significantly lower frequencies of both CD4+ and CD8+ effector memory T cells in hookworm infections , although an association between MT with altered T cell memory needs to be demonstrated . Th17 cells have been shown to play an important role in mucosal defense against bacterial and fungal pathogens and maintenance of the integrity of the mucosal barrier [35] . Selective depletion of Th17 cells in HIV-1 disease has been attributed to MT [8] . While we have not examined Th17 cell frequency directly , we report that hookworm infection is associated with significantly reduced levels of the prototypical Th17 cytokine IL-17 . Finally , the lower frequencies of mDCs and pDCs , cells known to normally induce proinflammatory cytokines and chemokines [36] , [37] following TLR ligation , may also provide an explanation for the failure of the hookworm-induced MT to induce systemic immune activation . Having demonstrated that hookworm infection is associated with MT that , in turn , affects certain cell populations , we wanted to examine how treatment/cure of hookworm infection might influence these parameters . Our data on individuals treated for hookworm ( and proven cured by absence of eggs in the stool ) demonstrate a partial reversal of MT events and a partial reconstitution of the DC compartment . Thus , treatment with albendazole and resultant cure led to diminution of the plasma levels of LPS and sCD14 and to an increase in frequency of pDC and mDC . Although we did not observe a significant reversal in CD8+ T cell counts and plasma levels of EndoCAb or pro-inflammatory cytokines ( data not shown ) , our treatment data considerably strengthen the association of hookworm infection with increased gut permeability and altered immune cell homeostasis and argue for a causal relationship . In summary , our study describes a novel relationship between an intestinal helminth infection and the process of MT—a process that might contribute to immune-mediated pathogenesis . It also suggests new targets for modulation of the pathology associated with hookworm infection .	Hookworm infections affect more than half a billion people worldwide and cause morbidity in the form of intestinal injury and blood loss . Host immunologic factors that influence the pathogenesis of disease in these individuals are not completely understood . Circulating microbial products such as LPS and markers associated with microbial translocation ( transfer of microbes or microbial products from the intestine to the circulation ) have been shown to play an important role in disease pathogenesis of certain infections like HIV . We have attempted to elucidate the role of the above mentioned factors in disease pathogenesis by comparing the plasma levels of the various markers in a group of hookworm infected and uninfected individuals . We show that circulating levels of microbial translocation markers are elevated in hookworm infected individuals , a potential cause of morbidity in these infections . This is associated with changes in the host immune system , especially in terms of lymphocyte and dendritic cells subsets . However , microbial translocation is not accompanied by increased levels of acute phase proteins or pro-inflammatory cytokines indicating that the parasite has evolved mechanisms to dampen LPS induced inflammation . Thus , our study highlights a novel pathway of pathogenesis in an intestinal helminth infection and improves our understanding of the various factors involved in the complex host-parasite interaction .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "medicine", "infectious", "diseases", "clinical", "immunology", "gastroenterology", "and", "hepatology", "immunology", "biology" ]	2012	Evidence of Microbial Translocation Associated with Perturbations in T Cell and Antigen-Presenting Cell Homeostasis in Hookworm Infections
Conformational changes in allosteric regulation can to a large extent be described as motion along one or a few coherent degrees of freedom . The states involved are inherent to the protein , in the sense that they are visited by the protein also in the absence of effector ligands . Previously , we developed the measure binding leverage to find sites where ligand binding can shift the conformational equilibrium of a protein . Binding leverage is calculated for a set of motion vectors representing independent conformational degrees of freedom . In this paper , to analyze allosteric communication between binding sites , we introduce the concept of leverage coupling , based on the assumption that only pairs of sites that couple to the same conformational degrees of freedom can be allosterically connected . We demonstrate how leverage coupling can be used to analyze allosteric communication in a range of enzymes ( regulated by both ligand binding and post-translational modifications ) and huge molecular machines such as chaperones . Leverage coupling can be calculated for any protein structure to analyze both biological and latent catalytic and regulatory sites . The concept of allostery was originally formulated to describe cooperative ligand binding in oligomeric proteins . The first model of positive cooperativity in binding of oxygen to hemoglobin was proposed by Linus Pauling in 1935 [1] , but the term allostery was coined in connection with the phenomenological MWC ( Monod-Wyman-Changeux ) and KNF ( Koshland-Némethy-Filmer ) models , developed in the 1960s [2] , [3] , [4] . Since then , there have been numerous studies of the mechanisms of allosteric regulation [5] , [6] , applying different experimental [7] , [8] and computational approaches [9] to proteins as different as small single-domain enzymes , motor proteins [10] and chaperones [11] , [12] . Although much progress has been made , the dichotomy between the original MWC and KNF models , or their modern counter parts , conformational selection and induced fit , dominates the discussion of allostery to this day [6] . The two models do however not describe mutually exclusive scenarios [13] , [14] , [15]: in both cases there is a shift in the population of different functional states upon effector binding . The main difference between the two is whether binding precedes conformational change or not [14] . Transition pathway analysis is primarily a matter of kinetics , whereas the shift in conformational equilibrium is one of thermodynamics: the conformational states involved determine which binding sites are allosterically connected , and their relative stability before and after binding determines the effect of regulation [6] . The major task therefore is to use this understanding to find structural determinants and molecular mechanisms of allosteric communication between distant binding sites [16] . Recently we developed the concept of binding leverage to measure the ability of a generic ligand , binding at different sites , to couple to conformational transitions , and thus its potential to have an allosteric effect [15] . We showed that in the majority of the studied cases , known allosteric and active sites had high binding leverage . We treated each site individually under the assumption that a site that has high binding leverage is connected to the global dynamics of the protein , without any specification of what other sites could be connected . Here we move on to investigate how allosteric communication takes place between specific pairs of sites . We introduce the concept of leverage coupling , which provides a quantitative characteristic of allosteric communication . We will also demonstrate how binding leverage and leverage coupling can be used to analyze allosteric communication mediated by metal binding and phosphorylation , as well as the function of three chaperones ( GroEL-GroES , CCT and thermosome ) . To study site-site communication , we make the following assumption: sites that have high binding leverage for the same motion are more likely to be allosterically coupled than sites that only have high binding leverage for motion along independent degrees of freedom . To represent a set of independent degrees of freedom we will use low frequency normal modes , which describe coherent motion involving the whole protein , and thus allow communication across large distances . We do not propose that protein dynamics is best described by global harmonic motion , but recognize the fact that the modes have repeatedly been shown to describe functional conformational change for proteins [17] , [18] . We therefore use them as a set of basis vectors describing the allowed directions of motion around the folded state of a protein , and explore the possibility that movement along a given mode can have an independent functional relevance . We have illustrated the role of independent degrees of freedom in allostery for a toy protein in Figure 1 . This protein has four binding sites W , X , Y and Z , and we have included two normal modes in the illustration , indicated by red and green arrows . The green mode causes closing of site Z and opening of site X , and only slight deformations of the other two sites . The red mode causes opening of site X and closing of site Y . Small red and green arrows indicate the deformation at each site for either mode . Site X and Y both have high binding leverage under the red mode and sites X and Z have high binding leverage under the green mode . This means that the pairs X and Y and Z and X are allosterically coupled , whereas the other pairs of sites are only weakly coupled ( indicated by the thickness of the lines crossing the protein , connecting the corresponding sites ) . In practice , X could be a catalytic site , Z an activator site and Y an inhibitor site . There is only indirect competition between the effects of Z and Y , i . e . if an activator is present at Z the effect of an inhibitor at Y might be weaker , and vice versa . With other patterns of communication , there can of course also be cases where activator and inhibitor binding are mutually exclusive . Alternatively , if this protein was an oligomer , X , Z and Y could be identical sites with positive or negative binding cooperativity . To quantify the strength of communication between two sites P and Q , as described in the previous paragraph , we introduce the leverage coupling DPQ . In the following , lower case roman indices ( i , j ) will number residues , lower case greek indices ( μ , ν ) normal modes , and upper case roman indices ( P , Q ) sets of residues , such as probe locations ( see Methods ) or biological binding sites . We denote the binding leverage of probe location P due to normal mode μ as LPμ ( see Methods ) . The symbol ΔiP is 1 if residue , and 0 otherwise . The leverage λiμ for a given residue and normal mode is then This calculation is done because our simulations generate a highly redundant set of probe locations , i . e . the denominator above can be large . Similarly , for an arbitrary set of residues P , we writewhere the norm of P is the number of elements in the set . Next , we introduce the vector , where n is the number of modes considered . The scalar productis large only if the sets P and Q have high leverage for the same normal modes . We will call the quantity DPQ the leverage coupling between the two sites . For example , for the two normal modes in Figure 1 , DXY and DXZ are large , and DXW , DZY , DZW and DYW are small . Similarly , the matrix measures the normalized leverage coupling and has the range 0≤CPQ≤1 . Since DPQ is based on normal mode vectors that represent infinitesimal motion , and depends on the size of the probe used in the calculation of LPμ , the scale of leverage coupling values is arbitrary and unique to each protein . We therefore always compare the leverage coupling of specific sites to the average coupling between the residues not belonging to any sites , i . e . the background leverage coupling for a given structure . The leverage coupling DPQ gives a measure of the strength of site-site coupling , but depends directly on the magnitude of conformational change at the different sites . In molecular machines like the chaperones , the conformational change at binding sites is small compared to the large-scale functional motions . Here , the measure CPQ can be used instead of DPQ to see how binding sites are correlated with different modes of functional motion . In this case we are interested in comparing the values between different sites and look for the most correlated pairs of sites for a given protein . The range of color bars in all figures containing CPQ matrices is from 0 to 1 , which reflects the span of CPQ values . Finally , the special case where one of the sets only has one residue can be used to see how one site couples to the rest of the protein . We will denote this variant DPi , where P is the studied site and the index i runs over all residues . We study 15 enzymes regulated by ligand binding , 14 of which were studied in our papers on binding leverage [15] and local closeness [19] . The addition is the 20-meric enzyme GTP cyclohydrolase I ( GTPCHI ) , which is both activated and inhibited allosterically by different substances [20] , [21] . These 15 enzymes are supplemented by 5 additional proteins , to generalize the analysis to other types of regulation and non-enzymes: Glycogen phosphorylase ( GP ) is allosterically regulated by both phosphorylation and ligand-binding [22] . The serine-protease thrombin is allosterically regulated by sodium binding [23] . The type I ( GroEL-GroES ) and type II ( CCT , thermosome ) chaperones are molecular machines regulated by ATP binding and hydrolysis [24] . The simulation parameters for the proteins discussed in the main text are summarized in Table 1 . The binding leverage was calculated using the ten lowest frequency normal modes [15] . The analysis of all the other proteins in this paper is based on the calculations described in the above paper . To begin with , we will briefly try to give the reader some intuition of what the leverage profiles can look like and how they relate to each other . The leverage profile similarity ( defined in Methods ) for the 10 lowest frequency normal modes , excluding the trivial first six modes , is plotted in Figure 2A for four different proteins . A value of 1 indicates that the two corresponding modes affect the exact same sites , and 0 that there is no overlap . Also included in the same panel is the importance of each of these normal modes , Λμμ ( see methods ) . Like for leverage coupling , the scale of leverage profiles is arbitrary and only relative values are relevant . For adenylate kinase ( AdK ) , the most significant leverage profiles correspond to modes 1 , 2 and 3 . Of these profiles , Λ1 and Λ2 are very similar . Figure 2B shows that these two leverage profiles peak at the same position , whereas the third is spread over more residues . That the leverage profiles are similar means that binding leverage is high for the same sites under the corresponding normal modes , even though these modes are orthogonal . Also included in the figure is the total binding leverage along the sequence , which is the sum of λiμ over all modes μ . Almost all active site residues ( involved in ATP and AMP binding ) are located at peaks in the total binding leverage . Having verified that different sites have their highest binding leverage for different normal modes , we move on to the analysis of leverage coupling . Supplementary Figure S1 contains plots of the leverage coupling matrix DPQ for the proteins not discussed in detail in the main text . The figure illustrates that , with the exceptions of ATCase and PTP1B , which we showed were difficult to analyze with binding leverage [15] , there is generally a stronger coupling between at least some of the allosteric and active sites ( including homotropic communication ) than between these sites and the rest of protein . One can also see that some sites are more strongly coupled than others are . We will however not analyze these proteins in detail; instead , we will focus on a couple of noteworthy cases . The tetrameric enzyme phosphofructokinase ( PFK ) in Bacillus stearothermophilus has one regulatory site where it is activated by ADP binding and inhibited by phosphoenolpyruvate ( PEP ) binding . The individual low frequency normal modes for this protein are less similar to each other than for AdK and there are also more modes that contribute significantly to binding leverage ( Figure 2A ) . In Figure 3 we display the leverage coupling DPQ for the four effector sites ( P = 1–4 , ADP/PEP ) , the four active sites ( P = 5–8 , F6P ) and the remaining residues of the four chains ( P = 9–12 , BG ) . As indicated by the color bars , the figure displays values from 0 to the maximal value of leverage coupling measured , in each matrix . Interactions between the effector sites dominate the matrix , and interactions between effectors and active sites are also strong , whereas interactions between the four active sites are weak . The latter indicates that there could be cooperative binding of effector but not of substrate . Experiments have shown that substrate binding is only cooperative in the presence of PEP [25] . The normalized leverage coupling CPQ is high if the sites P and Q have their peaks in binding leverage for the same modes . The CPQ matrix in Figure 3 for PFK indicates that different sets of modes affect the effector and active sites – the correlations are strong within the two groups of sites , but weaker between them . To demonstrate the validity of this interpretation we also included the DPQ-matrices for four of the individual modes . The modes were chosen from the dominating ones in Figure 2A . Modes 1 and 2 primarily affect the effector sites . Mode 1 also involves some connections between effectors and substrate . Mode 4 essentially only affects the active sites , and is probably responsible for any ( weak ) substrate binding cooperativity . Mode 10 provides relatively strong connections between the active site and the allosteric site , and Figure 2A shows that this is the second most important mode . To illustrate the communication between sites we color the surface of the protein by the leverage coupling between one site and each residue of the protein , DPi ( see Methods ) in Figure 4 , the raw data can be found in Figure S2 . The coloring in this figure , and in similar ones below , uses cyan for DPi = 0 , and magenta for the maximal value of DPi over all residues i for a given site P , i . e . the coloring gives the pattern of communication for a given site , but no indication of coupling strength compared to other sites P . The studied effector site in PFK communicates most strongly with the other effector sites ( Figure 4B ) , whereas the active site is connected with the other active sites , as well as the allosteric site ( Figure 4C ) . This apparent asymmetry comes from the fact that the interaction between effector sites is stronger than between anything else , but the connection between the active site and the effector site has approximately the same strength as the connections between active sites . Noteworthy is also the fact that neither site has any strong connections to sites other than the functional ones . GTPCHI catalyzes the first step in the production of tetrahydrobiopterin ( BH4 ) from GTP . It has positive cooperativity with respect to GTP binding . Allosteric regulation depends on the presence of the GTPCHI feedback regulatory protein ( GFRP ) . In combination with phenylalanine , GFRP reduces the cooperativity of GTP binding , increasing the activity at low GTP concentrations [20] . The GFRP-GTPCHI complex can also be inhibited by BH4 [21] . Both BH4 and phenylalanine bind at similar locations at the GTPCHI-GFRP interface . The architecture of the GFRP-GPTCHI complex is illustrated in Figure 5B . GTPCHI is a homodecamer arranged in two pentameric rings , and the regulatory GFRP pentamers bind one to each ring . We analyze three sites in the GFRP-GTPCHI complex , the BH4 site ( BH2 in the crystal structure ) , the phenylalanine site ( PHE ) and the catalytic site ( CAT ) . We define the catalytic site as all residues interacting with the catalytic Zn , and also His-134 and His-201 as defined in the catalytic site atlas [26] . The two allosteric sites have overlapping locations at the GFRP-GTPCHI interface and therefore have large mutual leverage coupling , as can been seen in Figure 5A , but both also couple strongly to the active site . The coupling between catalytic sites is not very strong in this complex , which is consistent with the fact that GFRP and Phe reduce cooperativity . To test the role of GFRP in modifying cooperativity in terms of binding leverage we removed GFRP from the structure and redid the calculations . The bottom two panels of Figure 5A show the coupling between the 10 different catalytic sites with and without GFRP . The effect is not very strong , but it is clear that the GTPCHI catalytic sites in the structure without GFRP are more strongly coupled compared to the background , than in the structure with GFRP . The connections DPi between one of the allosteric BH4-sites and the rest of the protein are illustrated in Figure 5C ( raw data in Figure S3 ) . Similarly , the coupling to one of the active sites , with and without GFRP present , is shown in Figure 5D and E . In the GFRP-GTPCHI complex the regulatory sites and their surrounding residues have the strongest leverage coupling , as was also seen for the site-site coupling matrix DPQ . This figure however clearly illustrates that communication with the “background” only involves the surroundings of the effector binding sites , and does not involve any other distinct sites . The concepts of binding leverage and leverage coupling can be generalized to study other forms of allosteric communication . Therefore , we consider cases of regulation involving metal binding and phosphorylation . We study glycogen phosphorylase ( GP ) as a case of allosteric regulation via covalent modification . Glycogen phosphorylase has two main conformations: the inactive dimeric T state and the active tetrameric R state [22] , [27] . In addition , it has two forms , GPa and GPb , where the former is phosphorylated at Ser14 . Crystal structures are available for both R and T state forms of GPa and GPb , but the R state is favored for unliganded GPa , and the T state for unliganded GPb . Both GPa and GPb are heterotropically activated by AMP , and inhibited by ATP and other metabolites . Upon phosphorylation , residues 1–20 become more ordered and move to a new position , 30 Å or so away , as can be seen in Figure 6B . In our calculations we use crystal structures of rabbit muscle GP . PDB entry 1gpa , representing unliganded GPa , is used for normal mode calculations; in addition we use T state GPb ( 1a8i ) and AMP-activated R state GPb ( 3e3n ) to define different binding sites . To be able to analyze phosphorylation using binding leverage , we treat residues 10–20 as a peptide ligand binding at two different sites , P1 ( T state GPb ) and P2 ( R state GPa ) , and calculate the normal modes without the 20 first residues . Figure 6A shows DPQ for P1 and P2 , and also the active ( PLP ) and allosteric sites ( AMP ) . It is clear that the connections are strongest between P1 and the PLP site . There is an unexpectedly weak interaction between the AMP and PLP site . Since P1 seems more important than P2 we hypothesize that release of residues 1–20 upon phosphorylation from P1 is more important for allostery than binding to P2 . The role of P1 is however somewhat uncertain given that residues 1–20 are relatively disordered in GPb . The connections are more or less symmetric between chains indicating that phosphorylation of one chain can trigger a global conformational change . To illustrate the connections between the active site and the rest of the protein we have drawn DPi for the active site in Figure 6C and D ( raw data in Figure S4 ) . This figure clearly shows strong connections between the active sites themselves and with P1 , but also towards one side of the dimer interface , opposite to P2 , which could contain latent allosteric sites . We also analyzed yeast glycogen phosphorylase ( yGP ) , which is structurally very similar to rabbit muscle GP , but differently regulated . The N-terminal strand in yGP is 40 residues longer than in rabbit muscle GP . In the GPb form the strand binds to the active site instead of P1 , and in the GPa form it folds at the dimer interface , at a position similar to P2 above [28] . The differences in regulatory mechanism between these two proteins are thus primarily due to the differing length of the N-terminal strand . This strand is excluded in our calculations and we therefore do not expect any qualitative differences between the two variants . We analyzed yGP using the same parameters as above , based on PDB entry 1ygp , having removed all residues before position 22 ( using the 1ygp numbering ) . We found that the leverage coupling between the active site and the rest of the protein is essentially identical to that of rabbit muscle GP , indicating that P1 is a latent allosteric site in yGP ( data now shown ) . As an example of metal binding-induced allostery we study the serine protease thrombin which is allosterically regulated by sodium binding [23] . It is also controlled by two other allosteric sites: exosite I ( EX1 ) interacts with several different protein partners , and exosite II ( EX2 ) interacts with several polyanionic substrates [23] . We divide the active site into three groups , the catalytic triad ( CAT ) and two of the substrate recognition pockets P2 and P4 . The leverage coupling of this protein is shown in Supplementary Figure S5 . The binding leverage of the sodium site is very low , and coupling to other sites weak . The sodium-induced conformational change primarily involves side-chain rearrangements , which are not modeled by our procedure . The concept of binding leverage could be expanded to include side-chains at a significant computational cost . Single side-chain rearrangements are however not expected to be modeled by low frequency normal modes , which means that a more refined description of motions would probably also be required to model the sodium regulation . Above , we analyzed a set of enzymes , some of them very large with up to 3 000 residues ( GTPCHI-GFRP , ATCase and GDH ) , and found that leverage coupling gives an understanding of allosteric communication in these enzymes . To push the envelope even further we will now move to the chaperonins , molecular machines with about 8 000 residues . These large molecules are quite challenging to study , the main bottleneck in our analysis being the time required to generate the very large number of probe locations needed , and the calculations took roughly 30–40 CPU hours for each chaperone on a modern desktop PC . Chaperonins represent a different type of allostery compared to the homo- and heterotropic regulation seen in enzymes . These molecular machines cycle through a set of conformations to provide a protected chamber for protein folding . ATP binding and hydrolysis cause large conformational changes to facilitate substrate capture , folding and release [24] . We will analyze and compare the bacterial group I chaperonin ( GroEL-GroES ) and eukaryotic and archaeal group II chaperonins ( CCT , Thermosome ) to investigate differences in regulatory mechanisms . The concepts developed in this paper were designed to analyze coupling between distinct ligand binding sites in enzymes , but , given a regulatory site , we can detect which parts of the protein are likely to have conformational change coupled to binding at that site . When a domain is deformed , the domain itself does not have high binding leverage , but many of the domain's hypothetical binding sites do . In this context binding leverage is therefore rather a measure of the degree of deformation of a section of the protein . By computing the leverage coupling DPQ for a site P and a domain Q , we can see how binding at the site P couples to conformational change in domain Q , making it possible to analyze allosteric communication in molecular machines such as chaperones . The GroEL-GroES chaperone consists of two heptameric rings ( GroEL ) and a heptameric lid ( GroES ) attached to one of the GroEL rings ( see Figure 7 ) . The ring closest to GroES is called the cis-ring and the other the trans-ring . Each GroEL ring provides a folding chamber . The functional cycle roughly goes through the following steps [24] , [29]: After substrate has bound to one of the open GroEL rings , ATP binds cooperatively to the GroEL ring [30] and increases affinity for GroES [31] . GroES binding causes a large conformational change increasing the volume of the cis folding chamber and changing it from hydrophobic to hydrophilic [24] , allowing folding to take place [32] . ATP hydrolysis weakens the affinity for GroES and when substrate and ATP have bound to the trans ring GroES and substrate are released from the cis subunit . In addition to intra-ring communication , there is also inter-ring signaling , which ( i ) adjusts the trans ring to accept substrate after cis ATP hydrolysis; ( ii ) leads to the ejection of cis substrate as a result of trans ATP binding [33]; ( iii ) accelerates the ejection of cis substrate by simultaneous binding of ATP and polypeptide to the open trans ring [34] . According to cryo-EM analysis , the equatorial domains play a key role in the inter-ring signaling [35] . Here , we will study the allosteric communication between the cis ATP sites and the rest of the protein . Conformational changes in GroEL involve the equatorial , intermediate and apical subdomains ( see Figure 7 ) . ATP binds to the equatorial domain and GroES to the apical domain . ATP binding controls the expansion of the folding chamber which takes place when the intermediate domain swings away from the equatorial domain . The apical domain follows the intermediate domain in this motion , largely as a rigid body . ATP hydrolysis mainly induces an increased flexibility of the intermediate and apical domains [29] , which probably explains the looser attachment of GroES to GroEL-ADP7 than to GroEL-ATP7 . ATP binding and hydrolysis is positively cooperative within each ring and negatively cooperative between the rings , providing tight ATP binding to only one ring at a time [36] . Figure 8A shows the leverage coupling DPQ and the normalized CPQ , for the ATP sites , the three subdomains of the cis ring , the trans ring and GroES . The strongest connections are between the chains of GroES . Second in strength are the connections between the apical and intermediate domains and GroES , and between the apical and intermediate domains themselves . The ATP site is also only weakly connected to the protein , a result of the fact that the equatorial domain and the ATP site undergo much smaller conformational change than the other two domains . The normalized leverage coupling CPQ however shows that the ATP site is more correlated with the apical and intermediate domains than with the equatorial domain to which it belongs . Correspondingly , there are strong correlations within the trans ring , where the magnitude of leverage coupling is much lower . The high degree of symmetry of the subsquare of the CPQ matrix describing interactions between the ATP site and the intermediate domain , and partly also the apical domain , is consistent with the positive cooperativity observed for ATP binding within one ring . Finally , there is a weaker correlation between the trans equatorial and intermediate domains , and the cis ring , particularly between the equatorial domains of either ring . These connections could be involved in the negative cooperativity between the two rings . We also analyzed the coupling DPi for one of the ATP sites , two views of this measure are provided in Figure 7A ( raw data in Figure S7 ) . The coloring indicates that the inside of the cis cavity , the GroEL-GroES interface and the interface between apical and intermediate domains are most strongly communicating with the cis ATP site . There is hardly any connection to the trans ring . These findings should be related to the fact that the main function of ATP is to regulate the cis cavity and the interactions with GroES , and also to the positive cooperativity of ATP binding . The human chaperone CCT has a similar function to GroEL , but does not utilize an analog to GroES . It consists of octameric rings , with similar but non-identical chains , instead of heptameric ones . It is also regulated by ATP binding and hydrolysis with steps similar to those of GroEL [24] . ATP binding is not cooperative , regulation has been described as sequential rather than concerted [37] . This is also reflected in the fact that only a fraction of the 16 ATP pockets were populated in crystal structures ( 13 in the one we use ) . The leverage-coupling matrix in Figure 8B shows that some of the apical domains are strongly coupled to each other , but coupling between intermediate domains is weaker . The normalized leverage coupling matrix in the same figure , CPQ , indicates that ATP has a weaker correlation with the apical and intermediate domains in CCT than it does in GroEL . In this plot the chains are ordered alphabetically , i . e . the first eight elements along either axis for each domain ( apical , intermediate , equatorial ) belong to the same ring , and the last eight to the other . This means that for CCT , interactions between the rings are as strong as within them , which is clearly different from what we saw for GroEL where the trans ring was only weakly connected to the rest of the protein . On the other hand , in CCT there is a greater asymmetry in the allosteric connections within one ring than in GroEL-GroES , in particular between the ATP site and the intermediate domain . This asymmetry is seen from the anisotropy of the different subsquares of the CPQ matrix , and is consistent with the sequential regulation of this chaperone [37] . Figure 7B shows the leverage coupling DPi for one of the ATP sites of CCT ( raw data in Figure S7 ) . As for GroEL-GroES ( Figure 7A ) , the ATP site is more strongly connected to the inside of the cavity than the outside , but in this case the pattern is relatively symmetric between the rings . The strongest deviation from symmetry , and also the strongest visible leverage coupling , is to a nearby interface between intermediate and apical domains ( magenta area in the middle panel of Figure 7B ) . The archaeal thermosome is homologous to CCT , but has a higher degree of symmetry than CCT [38] . The results of the analysis of this protein can be found in Figures S6 and S7 . The leverage coupling DPQ and the normalized CPQ in Figure S6A shows a pattern similar to CCT; the communication between apical domains is strong , and ATP is more strongly connected to the intermediate domain than the equatorial domain . The thermosome however displays a higher degree of symmetry ( as indicated by the uniformity of subsquares in the matrices ) . The DPi surfaces for one of the ATP sites in Figure S6B also shows a higher degree of symmetry than for CCT; in particular , the coupling to the neighborhood of the studied site is not stronger than to the rest of the protein . The difference in the symmetry of DPi is especially clear when comparing the two corresponding curves in Figure S7 . Symmetry is usually associated with positive cooperativity: the difference in symmetry between CCT and the thermosome might therefore reflect a difference in cooperativity , within the rings . Comparing to previous computational works [11] , [12] , [29] , [39] , [40] , we analyze allosteric communication between subunits in complete structures of both group I and group II chaperonins . It allows us to detect symmetry in the interactions between subunits of the cis ring of GroEL-GroES and its absence in CCT . We show that leverage coupling helps to understand positive cooperativity in the cis ring and negative cooperativity in the inter-ring communication in GroEL-Gro-ES , non-cooperative mechanism in human CCT , as well as positive intra-ring cooperativity in archaeal thermosome . Despite the almost half-century long studies of allostery , the majority of the works represents analysis of individual proteins ( or groups of homologs ) and mechanisms of allostery characteristic for individual structures . In this work , we sought a structural characteristic that can be used to understand allosteric communication in proteins of different types and sizes , from small single-domain proteins to large multi-chain oligomers and chaperones . We resort here to the thermodynamic aspect of allosteric regulation , where the conformational equilibrium between different structural states and their relative stability determine allosteric communication between sites and effect of regulation . We developed the concept of leverage coupling based on the idea that long-range communication between allosteric sites can be mediated by coherent motion along independent conformational degrees of freedom . We have studied the allosteric regulation of a number of proteins controlled by ligand binding , phosphorylation , or metal binding . The analysis has provided new insight into the allosteric mechanisms involved . Two approaches to the problem have been applied , first an analysis of known biological sites , to see how they are connected to each other , and how coupling between them compares to the background . Second we have selected specific sites and analyzed how these are coupled to the rest of the protein , thus being able to identify important functional regions of the protein , that are communicating with these specific sites , and in some cases see how different sites are coupled to different parts of the protein . We began our analysis by showing that leverage coupling largely captures the important connections in a number of enzymes , and exemplified this for phosphofructokinase ( PFK ) and GTP cyclohydrolase I ( GTPCHI ) . We also showed that the role of GFRP in regulating homotropic cooperativity in GTPCHI was described well by leverage coupling . In the case study of allostery by phosphorylation in glycogen phosphorylase , we found indications that the active sites had high leverage coupling with the site where the unphosphorylated N-terminal segment binds ( in a low temperature crystal structure ) , and hypothesize that the release of this segment upon phosphorylation causes the functional regulation . Allosteric regulation by metal binding in thrombin can however not be explained by leverage coupling , at least not in the coarse-grained version employed here . Finally , we have demonstrated that leverage coupling can be used to analyze allosteric communication in three different chaperones , and captures the differences in cooperativity between CCT and GroEL-GroES . We were able to describe allosteric communication between structural subunits providing positive cooperativity within each ring and negative cooperativity between the rings via inter-ring communication . The concept of allosteric communication mediated by collective degrees of freedom , as presented here , is based on our understanding of the physical principles determining protein dynamics . Using normal modes and coarse-grained docking simulations is a crude approximation of these principles – a complete description of the processes involved requires a statistical mechanics analysis based on a reliable energy function and proper conformational sampling . However , our analysis is successful in identifying communicating pairs of sites in the majority of the studied proteins , supporting our assumption that allosteric regulation relies on coherent conformational changes of oligomeric proteins and their domains . We have furthermore demonstrated that different regulatory sites have different patterns of communication ( see for instance the difference between active and allosteric sites in PFK ) , which are determined by motion along independent structural degrees of freedom , in our case different normal modes . This finding gives strong support to the idea that the ability of particular sites to couple to certain modes of motion , and not others , as illustrated in Figure 1 , can provide directed and differential allosteric communication and regulation . We have thus moved beyond the framework defined by the classical KNF and MWC models , both in that we propose a molecular mechanism for connecting different sites , and in that we are able to predict and identify many functional sites . Using normal modes to represent independent conformational degrees of freedom , we find that these motions can be used not only to describe the allosteric transition geometrically – as many have done before – but also to explain allosteric connections between different binding sites and to identify latent allosteric sites . Novel allosteric connections predicted by leverage coupling can be used as targets in experimental inhibitor/activator design . The calculation of binding leverage involves two main steps , generation of possible ligand conformations through coarse-grained Monte Carlo simulations , and analysis of the generated binding sites with respect to motions deduced from one or more crystal structures [15] . Probe conformations in which the probe is highly stressed , under a given protein motion , have high binding leverage . Binding leverage models allostery based on the assumption that binding to sites where ligand-protein interactions are connected to important degrees of freedom can affect the conformational equilibrium . We used binding leverage to rank probe locations ( defined below ) and found that high-ranking probe locations matched active and allosteric sites in a wide range of proteins . Here , we will give a brief overview of the procedure , which was described in detail previously [15] . Ligand binding is simulated with a completely fixed Cα-representation of the protein chain and a freely moving probe ligand in the form of a peptide with one or more Cα-atoms . The probe and protein interact via a square well potential which is attractive for Cα-Cα distances between 5 . 5 and 8 Å . Distances shorter than 4 . 5 Å are forbidden . Potential binding sites , called probe locations , are generated by running a number of short docking simulations . A probe location is defined as the residues interacting with the probe at the end of a given simulation . Binding leverage measures the ability of a probe ligand to resist a given motion , for example that of a normal mode . A spring is placed between all residue pairs in a probe location whose interconnecting lines pass through the ligand . The binding leverage of a probe location is then calculated as the total change in spring potential energy U due to a given motion , i . e . where summation is over all springs , and the additional index μ numbers the motion vectors used , i . e . one leverage is calculated for each vector . If more than one motion is considered the binding leverage can be summed to a total binding leverage for the probe location . Cα normal modes were calculated using MMTK with default parameters for all cases [41] . For the large proteins GTPCHI , GroEL-GroES , CCT and the thermosome we used the Fourier-basis approximation [42] , in all other cases vibrational modes are used . The binding leverage of residues i under mode μ ( defined in the main text ) can be grouped into leverage profiles , where m is the number of residues . We write the scalar product between two profiles as The magnitude Λμμ of a leverage profile indicates the importance of the corresponding normal mode in the total binding leverage , and the normalized scalar productis close to one when the corresponding modes involve the same binding sites , and close to zero when the overlap is small .	What are the molecular mechanisms of allosteric communication in proteins ? We base our analysis on the hypothesis that a folded protein has a number of conformational degrees of freedom , which describe fluctuations around the native conformation and switching from/to functional states . Transitions between the protein states involved in function and its regulation are based on coherent conformational degrees of freedom . Motion of one part of a protein along such a degree of freedom , implies a correlated motion in other parts of the protein . By determining which binding sites are simultaneously affected by the same motion we find sites that are allosterically coupled , i . e . where binding at one site can cause a change in ligand-affinity at another . Leverage coupling , the quantity introduced to measure this type of connection , reflects allosteric communication between different binding sites . We show how it can be used to understand allostery in enzymes of different sizes as well as in large protein complexes such as chaperones . Analysis of leverage coupling provides guidance in targeting native and latent regulatory sites .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "physics", "protein", "chemistry", "theoretical", "biology", "biological", "data", "management", "biophysic", "al", "simulations", "biology", "computational", "biology", "biophysics" ]	2011	Coherent Conformational Degrees of Freedom as a Structural Basis for Allosteric Communication
In many species , oocyte meiosis is carried out in the absence of centrioles . As a result , microtubule organization , spindle assembly , and chromosome segregation proceed by unique mechanisms . Here , we report insights into the principles underlying this specialized form of cell division , through studies of C . elegans KLP-15 and KLP-16 , two highly homologous members of the kinesin-14 family of minus-end-directed kinesins . These proteins localize to the acentriolar oocyte spindle and promote microtubule bundling during spindle assembly; following KLP-15/16 depletion , microtubule bundles form but then collapse into a disorganized array . Surprisingly , despite this defect we found that during anaphase , microtubules are able to reorganize into a bundled array that facilitates chromosome segregation . This phenotype therefore enabled us to identify factors promoting microtubule organization during anaphase , whose contributions are normally undetectable in wild-type worms; we found that SPD-1 ( PRC1 ) bundles microtubules and KLP-18 ( kinesin-12 ) likely sorts those bundles into a functional orientation capable of mediating chromosome segregation . Therefore , our studies have revealed an interplay between distinct mechanisms that together promote spindle formation and chromosome segregation in the absence of structural cues such as centrioles . During mitosis , centriole-containing centrosomes duplicate and then move to opposite ends of the cell where they nucleate microtubules and form the spindle poles . However , oocytes of many species lack centrioles , and as a result , spindles in these cells assemble using a different pathway [1] . We are interested in understanding the molecular mechanisms underlying this unique , acentriolar pathway of spindle assembly . Using C . elegans oocyte meiosis as a model , we recently found that acentriolar spindle assembly in this system proceeds by: 1 ) formation of a cage-like structure comprised of prominent bundles of microtubules that are constrained by the disassembling nuclear envelope , 2 ) reorganization of this structure such that the microtubule minus-ends are sorted away from the chromosomes to the periphery of the array where they are focused into multiple nascent poles , and then 3 ) coalescence of these poles until bipolarity is achieved [2] . During this process , the microtubule bundles project into the space near the homologous chromosome pairs ( bivalents ) and then begin to form lateral associations with them , an interaction that is maintained through anaphase . These lateral associations contribute to the alignment of bivalents at metaphase [3] . Subsequently , during anaphase , spindle morphology changes: the spindle shrinks and rotates 90 degrees such that it is perpendicular to the cell cortex , the spindle poles broaden , and the microtubule bundles reorganize into a parallel array , creating open channels [4–6] . Anaphase then proceeds through two phases , with chromosome-to-pole movement through the open channels in Anaphase A , and spindle elongation driving chromosomes further apart in Anaphase B [7] . This unique mode of chromosome segregation is kinetochore-independent [8] , and instead relies on a complex of proteins containing AIR-2 ( Aurora B kinase ) that concentrates at the center of each bivalent [9 , 10] , forming a ring-like structure ( the “midbivalent ring” ) . These rings localize to chromosomes during spindle formation [3] and then are removed from chromosomes in anaphase , remaining in the channels in the center of the spindle [8] . In this system , numerous factors have been shown to contribute to different aspects of acentriolar spindle assembly ( e . g . , microtubule length regulation and spindle pole formation ) , including MEI-1/2 ( katanin ) , KLP-7 ( MCAK ) , ASPM-1 , dynein , and others ( reviewed in [11] ) . Moreover , the kinesin-12 family member KLP-18 promotes spindle bipolarity [3 , 12 , 13] , by sorting microtubule bundles and forcing the minus-ends outward where they can be organized into the spindle poles [2] . However , the factors that are required for bundling microtubules and stabilizing these bundles in the absence of centrioles are unknown . Furthermore , little is known about how the acentriolar anaphase spindle is organized and stabilized . During mitosis in C . elegans , the centralspindlin complex of CYK-4 ( RhoGAP ) and the kinesin-6 family member ZEN-4 ( MKLP1 ) binds to and bundles antiparallel microtubules in the midzone of the anaphase spindle , providing stability to the structure [14] . The centralspindlin complex also localizes to the meiotic anaphase spindle in oocytes , and although this complex is required for the completion of cytokinesis and polar body formation [15] , depletion has no effect on anaphase spindle morphology [8] . Another component important for anaphase spindle organization during C . elegans mitosis is the microtubule bundling protein SPD-1 ( PRC1 ) , which is required for proper central spindle structure [16–18] , and localizes to the midzone in mitosis [16] and meiosis [19 , 20] . However , depletion of SPD-1 from C . elegans oocytes does not produce an obvious phenotype [8] , making it unclear if this protein functions during oocyte meiosis . Now , we have identified KLP-15 and KLP-16 , members of the conserved kinesin-14 family of minus-end-directed kinesins [21] , as factors required for microtubule bundling and organization during acentriolar spindle assembly in C . elegans oocytes; in the absence of these proteins , spindles are unable to maintain stable microtubule bundles and as a result are severely aberrant at metaphase and early anaphase . However , despite these defects , microtubules are then able to reorganize into a spindle capable of mediating chromosome segregation during anaphase . Importantly , this unexpected spindle reorganization phenotype enabled us to gain new insights into the mechanisms underlying anaphase spindle organization and chromosome segregation during acentriolar meiosis , uncovering previously unidentified roles for SPD-1 and KLP-18 in anaphase . These studies have therefore revealed a role for minus-end kinesins in acentriolar spindle assembly in C . elegans oocytes and highlight how the interplay of multiple mechanisms functions to promote the formation of a bipolar spindle that is capable of faithfully segregating chromosomes in this specialized type of cell division . To identify proteins required for acentriolar spindle assembly in C . elegans oocytes , we performed a targeted RNAi screen of genes previously reported to be embryonic lethal , visually screening for spindle defects in a strain expressing GFP::tubulin and GFP::histone [3] . This screen identified KLP-15 and KLP-16 , two highly homologous minus-end-directed kinesins ( 91 . 1% identical in amino acid sequence and 93% identical in mRNA sequence ) of the kinesin-14 family [22] . We observed the same spindle defects when we used the RNAi library clone annotated as targeting klp-15 as we did when we used the klp-16 clone , likely because both RNAi constructs target both transcripts due to the high sequence similarity between them ( Figs 1A and S1 ) . Consistent with this interpretation , our RNAi conditions caused high embryonic lethality ( 95 . 6%; S2A Fig ) , whereas single deletion mutants of either motor were largely viable; klp-15 ( ok1958 ) had 14% embryonic lethality and a new deletion mutant we generated , klp-16 ( wig1 ) , had 2 . 6% embryonic lethality . Moreover , treatment of klp-15 ( ok1958 ) with the RNAi clone annotated as targeting klp-15 and treatment of klp-16 ( wig1 ) with the clone annotated as targeting klp-16 caused high embryonic lethality ( 89 . 8% and 94 . 6% , respectively; S2A Fig ) , consistent with the interpretation that both proteins are expressed , and that each RNAi library clone can target both proteins . These results suggest that KLP-15 and KLP-16 are redundant , and therefore , we refer to these proteins collectively as KLP-15/16 ( in describing assays and results where we cannot distinguish between them ) . Previous work from other groups had suggested a role for KLP-15/16 in the segregation of meiotic chromosomes , because in addition to embryonic lethality , inhibition of these proteins resulted in phenotypes such as polar body defects , a high incidence of male progeny ( which results from non-disjunction of the X chromosome in the oocyte ) and multiple female pronuclei in the one-cell stage embryo [21 , 23–28] . However , a careful analysis to determine the causes of these segregation errors had not been reported . Therefore , we performed detailed live and fixed imaging of oocytes following klp-15/16 ( RNAi ) . After nuclear envelope breakdown ( NEBD ) is initiated in control oocytes , microtubules form prominent bundles that organize into a cage-like structure ( Fig 1B , Fig 1C , arrows; S1 Movie ) . The microtubules are then sorted such that the minus-ends of the microtubule bundles , visualized by ASPM-1 [29–31] , are on the periphery of the array , where they are organized into multiple nascent poles that coalesce until bipolarity is achieved ( Figs 1B , 1C , 1D and S2C; S1 and S3 Movies [2] ) . In klp-15/16 ( RNAi ) oocytes , although the microtubule cage appears to initially form normally ( Fig 1C bottom; arrows ) , the microtubule bundles are not maintained and begin to fall apart , resulting in a disorganized array that lacks focused nascent poles . Then , the array collapses into a “microtubule ball” comprised of short microtubules surrounding the chromosomes ( Figs 1B , 1C and S2C; S1 and S2 Movies ) . ASPM-1 localization appears largely diffuse at both the microtubule “array” and “ball” stages ( Figs 1C , 1D and S3A; S4 Movie ) , suggesting that microtubule minus-ends are distributed throughout these structures . However , since there are examples of spindles where ASPM-1 does have areas of slight concentration within the microtubule ball structures ( S3A Fig , arrowheads ) , it is possible that these spindles have some small degree of microtubule organization . Our analysis therefore demonstrates that depletion of KLP-15/16 affects the early stages of spindle assembly , resulting in structures that lack prominent microtubule bundles past the cage stage . These two kinesins likely function redundantly at this stage , since spindles appeared normal in the klp-15 ( ok1958 ) and the klp-16 ( wig1 ) single mutants ( Figs 2C and S3C ) . Interestingly , despite the severe spindle defects in oocytes following klp-15/16 ( RNAi ) , we observed that mitotic spindles in the one-cell stage embryo formed normally ( Fig 1E; S5 Movie ) , suggesting that these proteins are not essential when centriole-containing centrosomes are present . Given the strong phenotype observed in oocytes upon KLP-15/16 depletion , we next assessed the localization of these proteins . To this end , we generated a peptide antibody against the N-terminal 20 amino acids of KLP-16; because there is only one amino acid different between KLP-15 and KLP-16 in this region , this antibody likely recognizes both proteins ( S1 Fig ) . Indeed , this antibody recognizes a band corresponding to the size of both proteins in a Western blot of control worms , and this band was greatly reduced when RNAi was performed using a clone from the RNAi library that had been annotated as targeting klp-16 ( Fig 2A ) . Furthermore , this band was also reduced when klp-15 ( ok1958 ) worms were treated with the clone annotated as targeting klp-15 , and when klp-16 ( wig1 ) worms were treated with the klp-16 clone ( Fig 2A ) , further demonstrating the specificity of the antibody and confirming that RNAi treatment using either of the RNAi library clones for klp-15 or klp-16 targets both proteins . Using this antibody , we found that KLP-15/16 localize in the cytoplasm prior to NEBD and then begin to accumulate on microtubules during the multipolar stage , becoming uniform on the spindle throughout metaphase and anaphase ( Fig 2B ) . This localization is specific and likely represents both proteins , as it is abolished following RNAi depletion of KLP-15/16 , but it is the same as wild-type in the klp-15 ( ok1958 ) mutant and the klp-16 ( wig1 ) mutant ( Fig 2C ) . We observed a similar localization pattern in a worm strain expressing KLP-16::GFP from the endogenous locus , but this strain also revealed clear localization of KLP-16 to microtubule bundles at the cage stage ( Figs 2D and S4B ) and to the centrosomes and mitotic spindle microtubules in one-cell stage embryos ( Figs 2E and S4C ) , patterns that were not apparent with the KLP-15/16 antibody ( quantification in Materials and Methods ) . This discrepancy is likely due to variability with the fixed imaging , because even for the stages where we could observe robust staining of spindle microtubules using the KLP-15/16 antibody , not all spindles were stained . Furthermore , when we stained oocytes from the KLP-16::GFP strain with a GFP antibody , we saw similar variability ( quantification in Materials and Methods ) ( S4D and S4E Fig ) , despite the fact that when this strain was viewed live , every oocyte/embryo imaged had bright KLP-16 fluorescence that appeared to mark spindle structures ( S4A Fig ) . Therefore , we conclude that KLP-15/16 localize to spindle structures through all stages of oocyte spindle assembly , and also to microtubules in mitotic one-cell stage embryos . Although KLP-15/16 localize to microtubule bundles at the cage stage ( Figs 2D and S4B ) , these motors are not necessary for the formation of these bundles ( Fig 1C ) , suggesting that they act redundantly with other microtubule associated factors at this initial stage of spindle assembly . Similarly , KLP-15/16 localize to spindle microtubules in mitotic embryos ( Figs 2E , S4C and S4E ) , but they are not necessary for the assembly of these spindles ( Fig 1E ) , potentially because centrosomes provide the primary source of microtubule organization in these cells . Taken together , the phenotype of klp-15/16 ( RNAi ) and the localization pattern of these proteins support a role for KLP-15/16 in acentriolar meiotic spindle assembly where they likely stabilize the microtubule bundles formed during the cage stage . These stabilized microtubule bundles can then be sorted by other molecular motors such as KLP-18 to achieve bipolarity [2] . While filming klp-15/16 ( RNAi ) oocytes , we made the surprising observation that although spindle assembly was severely aberrant , microtubules were often able to reorganize into a bundled structure capable of segregating chromosomes , suggesting the presence of a second , KLP-15/16-independent mechanism for bundling microtubules that operates during anaphase ( S6 Movie ) . Therefore , we used markers of anaphase progression to carefully examine anaphase in klp-15/16 ( RNAi ) oocytes , to better understand this mechanism . During meiosis in wild-type oocytes , separase ( SEP-1 ) relocates from the kinetochore to the midbivalent ring complex at anaphase onset and then disappears from the rings by late anaphase ( Fig 3A ) [4] . Aurora B ( AIR-2 ) , a component of the midbivalent ring complex , is removed from the rings at anaphase onset and relocalizes to the microtubules by mid anaphase ( Fig 3A ) [32 , 33] . Therefore , we used these markers to stage oocytes following klp-15/16 ( RNAi ) , allowing us to distinguish pre-anaphase ( AIR-2 on the ring structures , SEP-1 on kinetochore ) , early anaphase ( both proteins in rings ) , and mid/late anaphase ( AIR-2 on microtubules , SEP-1 gone ) . Using these markers to stage klp-15/16 ( RNAi ) spindles , we found that the microtubule ball configuration observed in our imaging ( Fig 1B , 1C and 1D ) represents a mixture of metaphase and early anaphase ( Fig 3A ) , although the structures in early anaphase tended to be smaller ( S3B Fig ) . ( Note that the spindles that we used for our linescan analysis in Fig 1D all were within the range of volumes observed for metaphase spindles ( S3B Fig ) ) . This analysis suggests that the metaphase disorganized microtubule array begins to shrink in preparation for anaphase , similar to what happens in wild-type spindles [6] . Following this stage , when AIR-2 has relocalized to the microtubules and SEP-1 is gone in mid/late anaphase , microtubules reform into a bundled structure and chromosomes are able to segregate into distinct masses ( Fig 3A ) . Despite this anaphase spindle reorganization , we observed segregation errors such as lagging chromosomes and segregation of chromosomes along different axes ( Fig 3B and 3C; S6 Movie ) that resulted in high levels of aneuploidy in MII oocytes ( Fig 3D ) , likely due to the severely aberrant metaphase spindles that were unable to align chromosomes ( Fig 1B and 1C ) . However , the high rates of aneuploidy also raised the possibility that the microtubules in the klp-15/16 ( RNAi ) anaphase spindles may not be organized like in wild-type spindles , where a high concentration of microtubule minus-ends are found at the spindle poles . To test this hypothesis , we assessed the localization of ASPM-1 and KLP-18 ( a kinesin that is enriched at the poles of wild-type oocyte spindles [12] ) , and found that the microtubules of anaphase spindles in klp-15/16 ( RNAi ) oocytes , although bundled , are likely not organized properly , since ASPM-1 and KLP-18 are distributed throughout the entire spindle instead of being enriched at the poles ( Fig 3B and 3C ) . Although it is possible that microtubules within the bundles are properly organized and that the signals to localize ASPM-1 and KLP-18 are defective , we favor the interpretation that the secondary mechanism we identified bundles microtubules without first sorting them , resulting in bundles comprised of microtubules of mixed polarity . Next , we wanted to uncover factors that are responsible for bundling anaphase microtubules in klp-15/16 ( RNAi ) oocytes . Two possible candidates are the centralspindlin complex ( comprised of ZEN-4 and CYK-4 ) and SPD-1 , since these proteins have clearly defined roles during anaphase in C . elegans mitosis [14 , 16 , 18 , 34] and have been shown to concentrate at the midzone of the anaphase spindle in C . elegans oocytes [8 , 19 , 20] . Therefore , we assessed the localization of SPD-1 and ZEN-4 at high resolution on C . elegans oocyte spindles . As expected from previous studies , we found that neither centralspindlin ( ZEN-4 ) nor SPD-1 localize to metaphase spindles in control oocytes ( Fig 4A ) . However , during anaphase , ZEN-4 and SPD-1 both become enriched in a short region at the center of the spindle ( Fig 4A ) , with similar though non-identical localization ( Fig 4C ) . Following klp-15/16 ( RNAi ) , we observed a similar pattern , with neither ZEN-4 nor SPD-1 present on the disorganized spindle structures prior to anaphase , but then prominent localization on the bundled microtubules between the sets of segregating chromosomes during anaphase ( Fig 4B ) . This localization was clear even in spindles with multiple sets of segregating chromosomes , where the bundles were not all oriented along the same axis ( Fig 4B , bottom zoom ) . Therefore , because centralspindlin and SPD-1 both localize to microtubule bundles following klp-15/16 ( RNAi ) , these factors are in a location where they could potentially contribute to the anaphase-bundling mechanism we identified . To test this hypothesis , we assessed a potential functional role for these proteins in anaphase microtubule bundling . In previous work , co-depletion of ZEN-4 and SPD-1 did not affect anaphase spindle structure [8] , suggesting that these proteins may not play a role in oocytes . However , our studies have revealed a mechanism that operates in parallel with KLP-15/16 ( since KLP-15/16 are normally present on anaphase microtubules , Fig 2B and 2D ) . Thus , we expect that single depletion of this putative anaphase bundling factor may have only a mild ( or no ) anaphase phenotype , but that depletion of KLP-15/16 in combination with the secondary factor would completely abolish anaphase bundling . Therefore , we tested each candidate by single depletion and also by co-depletion/inhibition with KLP-15/16 ( Fig 5A ) , and then scored microtubule bundling in mid/late anaphase ( using SEP-1 and AIR-2 as markers to stage anaphase , as before; Fig 5B ) . In addition to microtubule bundling , we also assessed chromosome segregation as a functional readout for anaphase spindle organization , by scoring whether chromosomes were able to segregate into distinct masses ( Fig 5B ) . Using these assays , we found that both single and double inhibition/depletion of ZEN-4 and SPD-1 had little effect on anaphase microtubule bundling and chromosome segregation ( Fig 5A and 5B ) , consistent with a previous study [8] . Moreover , oocytes where both ZEN-4 and KLP-15/16 were inhibited/depleted appeared similar to klp-15/16 ( RNAi ) alone , with most spindles containing bundled microtubules that were able to segregate chromosomes . However , we found that co-depletion of KLP-15/16 and SPD-1 largely abolished anaphase microtubule bundling and chromosome segregation ( Fig 5A and 5B ) and resulted in spindles with shorter microtubule lengths ( Fig 5C ) . Furthermore , we observed an increase in the percentage of embryos with a single large polar body and no maternal pronucleus under these conditions , suggesting that the meiotic divisions lacked a functional spindle on which DNA could segregate ( Fig 6A , 6B and 6C ) . Interestingly , our estimations of spindle microtubule lengths revealed that the microtubules in the spd-1 ( RNAi ) condition were somewhat longer than microtubules in the control ( Fig 5C ) , suggesting that SPD-1 may perform a subtle role in regulating spindle length in anaphase . This observation is reminiscent of studies of mitotic anaphase in C . elegans , where an SPD-1 mutant displays larger distances between segregating chromosomes than wild-type embryos , suggesting that the microtubule crosslinking activity of SPD-1 can act to slow the rate of spindle midzone elongation [35] . Taken together , these data highlight a previously unknown role for SPD-1 on acentriolar anaphase spindles . Given this finding , we more carefully assessed the dynamics of SPD-1 loading onto the spindle during anaphase . Live imaging of control oocytes expressing SPD-1::GFP and mCherry::histone revealed that SPD-1 begins to load onto the spindle between segregating chromosomes shortly after spindle rotation and then continues to accumulate as anaphase progresses ( Fig 4D; S7 Movie ) , consistent with previous studies [19 , 20] . Similar to control oocytes , following depletion of KLP-15/16 , SPD-1 loads onto microtubules in early anaphase , at the microtubule ball stage ( Fig 4D; S7 Movie ) . Subsequently , as SPD-1 accumulates on the spindle , prominent bundles begin to form ( Fig 4D; S7 Movie ) . This localization pattern , in combination with our functional analysis , is consistent with the interpretation that loading of SPD-1 in early anaphase provides a secondary bundling activity that provides spindle stability and allows for chromosome segregation . Our SPD-1::GFP imaging also revealed that when SPD-1 loads onto microtubules in klp-15/16 ( RNAi ) oocytes , the forming bundles start out randomly oriented but are then restructured into a largely parallel array ( Fig 4D; S7 Movie ) . Therefore , in addition to SPD-1 bundling microtubules , there is another mechanism working to reorganize these newly formed microtubule bundles into a functional orientation along which chromosomes are able to segregate . One candidate factor that could provide this function is KLP-18 , since this motor sorts microtubule bundles during spindle assembly [2] , and is present on anaphase spindles following klp-15/16 depletion ( Fig 3C ) . It is currently unknown whether KLP-18 also functions during anaphase , because the requirement for this protein earlier during spindle assembly has made it difficult to assess an anaphase-specific role; in klp-18 mutants or RNAi , chromosomes do not segregate into distinct groups in anaphase ( Fig 5D ) because the spindles are monopolar prior to anaphase onset [4 , 12] . However , the KLP-15/16 depletion phenotype offers a unique opportunity to address this question , since this condition has revealed a sorting activity that operates specifically during anaphase to generate parallel arrays of microtubule bundles . Notably , this activity does not require that the microtubules have been sorted previously; following klp-15/16 ( RNAi ) , the microtubule bundles start out randomly oriented ( Fig 4D ) yet they can still be organized into a parallel array . Therefore , this feature allowed us to explore a potential role for KLP-18 during anaphase by co-depleting it with KLP-15/16 . To determine if KLP-18 could be required for this anaphase reorganization activity , we depleted KLP-15/16 in a KLP-18 mutant , klp-18 ( tm2841 ) , which results in a predicted early stop that is thought to eliminate KLP-18 function [2] . We then stained the spindles for SEP-1 and AIR-2 to stage them as before , to determine if microtubules were able to reorganize into spindles capable of mediating chromosome segregation in late anaphase , as they do following klp-15/16 depletion in the wild type strain ( Figs 3A and 5A ) . Notably , we found that depletion of KLP-15/16 in the klp-18 ( tm2841 ) mutant results in a complete failure of microtubule reorganization and chromosome segregation in late anaphase ( Fig 5D ) , despite the fact that SPD-1 was still able to target to the microtubules ( Fig 5E ) and that the early anaphase configuration appeared similar to KLP-15/16 depletion in the wild-type strain ( Figs 3A and S5 ) . These results therefore suggest that KLP-18 could provide the anaphase spindle reorganization activity that we observe in the klp-15/16 ( RNAi ) condition . Although we cannot completely rule out the possibility that the metaphase defect in klp-18 mutant oocytes prevents the microtubule reorganization that normally occurs following KLP-15/16 depletion , we think that our data are at least suggestive that KLP-18 provides this activity during anaphase and that it may therefore have an anaphase role in wild-type spindles . Taken together , we therefore propose that two complementary activities facilitate the reorganization of anaphase spindle microtubules following KLP-15/16 depletion: 1 ) SPD-1 loads in early anaphase to generate prominent microtubule bundles of mixed polarity , and 2 ) KLP-18 acts to orient these bundles into a parallel array that is capable of segregating chromosomes . Finally , we wanted to further investigate the mechanism of chromosome segregation during KLP-15/16-independent anaphase . During wild-type meiosis , microtubule bundles run along the sides of chromosomes prior to anaphase onset . These lateral associations remain in place during anaphase , creating channels that the chromosomes reside in as they move towards spindle poles [4] and then spindle elongation drives chromosomes further apart [7 , 8] . Given that microtubules are completely disorganized prior to anaphase onset following KLP-15/16 depletion and , unlike wild-type spindles , have no discernable lateral associations with chromosomes , we wanted to determine what types of microtubule-chromosome contacts were established during anaphase to facilitate segregation . First , we asked if anaphase spindles in klp-15/16 ( RNAi ) oocytes are able to form any channels that are analogous to those observed in wild-type oocytes . To this end , we stained spindles for SUMO , to mark the ring structures [36] , and SPD-1 , to mark anaphase microtubule bundles . In control spindles , each channel is comprised of a pair of separating chromosomes with a ring in between , and SPD-1 marks the microtubule bundles adjacent to the ring . Therefore , line scans of these components in control spindles show an alternating pattern of SUMO/SPD-1 and SUMO/microtubules across the channels ( Fig 7A ) . In klp-15/16 ( RNAi ) oocytes , we found similar alternating patterns of these markers in a significant number of spindles ( 12/18 klp-15/16 ( RNAi ) spindles examined; Fig 7A ) ; showing that the spindles are capable of forming microtubule channels during anaphase . Importantly , we also observed microtubules associating laterally with the segregating chromosomes ( Fig 7B arrows and Fig 7C; these associations were seen in 22/31 klp-15/16 ( RNAi ) spindles ) suggesting that this type of association can be established in anaphase , even if these associations are not in place at anaphase onset . Despite the fact that channels can form during anaphase in klp-15/16 ( RNAi ) oocytes , we also found that some rings appeared to be on the periphery of the spindle , demonstrating that not all separating chromosomes end up in a channel with a ring in the center ( Fig 7A , arrowheads ) . To gain insight into this variability , we looked earlier in anaphase before the microtubules were reorganized into bundles . In early anaphase spindles , at the “microtubule ball” stage when homologous chromosomes first begin to come apart , we observed some rings embedded in the microtubule ball close to the separating chromosomes , but also some rings towards the periphery of the structure ( Fig 7B ) . Subsequently , when the microtubules are bundled and aligned into parallel arrays , rings can be seen both in channels between segregating chromosomes and also completely outside of the reorganized spindle ( Fig 7B ) . This is likely due to the fact that microtubule bundling and reorganization are occurring as chromosomes begin to come apart . This results in some chromosomes and rings becoming organized within channels while others are not . This behavior may also contribute to the presence of lagging chromosomes in these spindles ( Figs 3B , 5A and 7A ) . Therefore , while it is unlikely that the complete formation of a ring-containing channel is essential for chromosome segregation , the fact that lateral associations are established suggests that they could contribute to segregation in this context . Taken together , our data have revealed two distinct mechanisms that act to bundle microtubules in acentriolar spindles . Prior to spindle assembly , KLP-15/16 localize diffusely throughout the cytoplasm , which is in contrast to kinesin-14s from other organisms that have been shown to be sequestered in the nucleus [37 , 38] . Then , as acentriolar spindles begin to form , KLP-15/16 load onto microtubule bundles during the cage stage , stabilizing them to facilitate spindle assembly ( Fig 8 ) . This proposed function is consistent with previous studies of kinesin-14s in other organisms , which demonstrated that this family of kinesins is required for acentriolar spindle formation and localize uniformly to acentriolar spindle microtubules [30 , 39–41] . However , while depletion of kinesin-14s in other organisms predominantly results in defects such as unfocused poles and splayed microtubules , depletion of KLP-15/16 in C . elegans oocytes completely prevents bipolar spindle formation and abolishes microtubule bundling prior to anaphase , implicating these proteins in the stabilization of microtubule bundles comprising the acentriolar meiotic spindle . Furthermore , unlike most other organisms where kinesin-14s perform essential roles in mitosis [40 , 42–44] , KLP-15/16 are not required for mitotic spindle formation in C . elegans , suggesting a unique role for KLP-15/16 that is specific to acentriolar spindle assembly; this finding is also reminiscent of studies in Drosophila , where inhibition of the kinesin-14 Ncd has a much more severe phenotype in oocytes than it does in mitosis [38 , 42 , 45] . The presence of the “microtubule ball” comprised of short microtubules that ultimately forms prior to anaphase following klp-15/16 ( RNAi ) raises the intriguing possibility that the function of KLP-15/16 could be to stitch together short microtubules into longer bundles that can then be sorted and organized into a bipolar spindle . Since kinesin-14s contain a motor domain in the C-terminus and a microtubule binding domain in the N-terminus [37 , 43 , 46] , and it has been reported that this class of kinesins can stabilize and cross-link microtubules in a parallel configuration [47] , it is possible that these motors could contribute to this stitching activity . This interpretation is consistent with a previous electron microscopy study , which reported the presence of many short microtubules in a partial reconstruction of a C . elegans oocyte spindle [48] , and also with a study in Xenopus egg extracts that demonstrated that meiotic spindles are comprised of tiled arrays of short microtubules [49]; therefore a microtubule stitching activity might be something that is especially important in the context of acentriolar meiosis . Our studies suggest that KLP-15/16 could be factors that organize these short microtubules into longer bundles capable of mediating chromosome congression and segregation . In addition to revealing an important function for KLP-15/16 , our studies have yielded insights into previously unknown mechanisms promoting accurate chromosome segregation during acentriolar meiosis . First , we found that the PRC1-family protein SPD-1 provides a secondary activity that stabilizes microtubule bundles during anaphase ( Fig 5A , 5B and 5C ) . This activity was previously unidentified , as prior depletion of SPD-1 [8] , confirmed by our own studies ( Fig 5A ) , failed to reveal an obvious anaphase defect . However , this is likely because KLP-15/16 are present on the anaphase spindle stabilizing the microtubule bundles ( Fig 2B ) , making SPD-1 non-essential until these proteins are depleted . This discovery is similar to findings in fission yeast , where the SPD-1 homolog Ase1 is not essential on its own but provides a backup mechanism for bipolar spindle assembly under conditions where the kinesin-5 motor Cut7 and the kinesin-14 motor Pkl1 are deleted [50] . During mitosis in other organisms , homologs of SPD-1 are known to crosslink antiparallel microtubules [51–53] , and our data are consistent with SPD-1 performing a similar function in C . elegans oocytes . During wild-type meiotic anaphase , this protein loads onto the central region of the spindle ( Fig 4D ) , where this putative crosslinking activity could reinforce anaphase spindle structure . Under KLP-15/16 depletion conditions , SPD-1 loads at the microtubule ball stage ( Fig 4D ) , which contains many short microtubules that are likely randomly oriented ( Fig 1C ) ; given this configuration , the ability to crosslink antiparallel microtubules would enable SPD-1 to bundle microtubules . These SPD-1-stabilized microtubule bundles could then be sorted and aligned into a parallel array by the action of KLP-18 . Therefore , our studies have uncovered a new function for SPD-1 on C . elegans acentriolar spindles , and also represent the first demonstration that PRC1-family proteins play a role during oocyte meiosis . Furthermore , our work also suggests that KLP-18 may be functional during anaphase in these cells , since it appears to organize the microtubule bundles generated by SPD-1 ( Fig 5D and 5E ) , suggesting that this motor not only plays roles during bipolar spindle formation , but may also be required to maintain spindle organization as chromosomes segregate . Finally , our work has also shed light on the mechanisms driving chromosome segregation in this unique form of anaphase . We found that anaphase spindles in KLP-15/16-depleted oocytes are sometimes able to form channels with lateral microtubule-chromosome associations ( Fig 7 ) , despite the lack of microtubule bundles prior to anaphase onset . These data suggest that this form of microtubule-chromosome contact is preferred during anaphase and points to a role for the chromosomes providing significant structural cues for spindle organization as these associations can be established de novo following KLP-15/16 depletion . However , given that these laterally-associated bundles may be comprised of microtubules of mixed polarity ( Fig 3B and 3C ) , it is improbable that they would be able to efficiently facilitate directional chromosome movement ( a mechanism we proposed to drive normal Anaphase A [4] ) . Therefore , we suggest that the primary force driving segregation in the absence of KLP-15/16 is the elongation of these lateral bundles in an Anaphase-B type mechanism . The discovery that lateral associations are established during anaphase is also interesting since two other types of chromosome-spindle contacts have been proposed to facilitate segregation during wild-type anaphase: 1 ) elongating microtubules contacting the inside surfaces of separating chromosomes to provide a pushing force [8] and 2 ) chromosomes contacting the spindle poles , so that outward pole separation can drive segregation in Anaphase B [7] . It is possible that the first type of association contributes to segregation during KLP-15/16-independent anaphase; since not every chromosome ends up in a normal microtubule channel ( Fig 7 ) , some microtubules might randomly make contacts with chromosome surfaces and provide a pushing force , contributing to segregation alongside the bundles that are laterally-associated . Indeed , our data are consistent with this idea since we observe non-laterally-associated microtubules in the reorganized klp-15/16 ( RNAi ) anaphase spindles ( Fig 7 ) . In contrast , the second model proposed that spindle shrinkage enables the chromosomes to establish a physical tether to a cross-linked network of microtubules and pole proteins; outward sliding of interpolar microtubules would then drive the poles and the tethered chromosomes apart [7] . Our observation that pole proteins KLP-18 and ASPM-1 are distributed throughout klp-15/16 ( RNAi ) spindles both prior to and during anaphase ( Figs 1C , 1D , 3B and 3C ) makes it difficult to imagine how such a tether would efficiently form in this context , and we therefore speculate that pre-established spindle poles may not be absolutely required to segregate chromosomes in C . elegans oocytes ( although we cannot rule out the possibility that other pole proteins exhibit a higher level of organization in these mutant spindles ) . Moreover , our data also raise the possibility that spindle elongation could be capable of driving segregation even when the polarity of microtubules within the spindle is disrupted , potentially revealing an unusual mode of chromosome segregation that operates in this mutant context . In summary , our studies have uncovered a crucial role for KLP-15 and KLP-16 in C . elegans acentriolar spindle assembly , revealed a second , anaphase-specific mechanism dependent on SPD-1 operating in parallel to these kinesins , and provided new insights into anaphase spindle organization and chromosome segregation mechanisms during acentriolar meiosis . In this study , ‘wild-type’ refers to N2 ( Bristol ) or EU1067 worms grown on NGM/OP50 plates , and ‘control’ refers to the RNAi vector control ( L4440 ) . N2 ( Bristol ) ANA065: adeIs1[pMD191 , mex-5::spd-1::GFP] II ( gift from Marie Delattre ) ANA072: adeIs1[pMD191 , mex-5::spd-1::GFP] II; ltIs37[pAA64; pie-1::mCherry::his-58; unc-119 ( + ) ] IV ( gift from Marie Delattre ) EU716: zen-4 ( or153 ) IV ( from the CGC ) . For experiments using zen-4 ( or153 ) , plates were shifted to 25°C 16–18 hours before dissection and fixation . EU1067: unc-119 ( ed3 ) ruIs32[unc-119 ( + ) pie-1::GFP::H2B] III; ruIs57[unc-119 ( + ) pie-1::GFP::tubulin] ( gift from Bruce Bowerman ) OD57: unc-119 ( ed3 ) III; ltIs37[pAA64; pie-1::mCherry::his-58; unc-119 ( + ) ] IV; ltIs25[pAZ132; pie-1::GFP::tba-2; unc-119 ( + ) ] IV ( gift from Arshad Desai ) RB1593: klp-15 ( ok1958 ) I . ok1958 is a deletion allele of the last 391 amino acids of KLP-15 ( from the CGC ) SMW13: klp-18 ( tm2841 ) IV/nT1[qIs51]; unc-119 ( ed3 ) ruIs32[unc-119 ( + ) pie-1::GFP::H2B] III; ruIs57[unc-119 ( + ) pie-1::GFP::tubulin] ( Wolff et . al . , 2016 ) SMW15: klp-16 ( wig1 ) I . This strain was generated using a CRISPR approach detailed below . SMW16: Pklp-16::klp-16::GFP ( C1971>A–PAM site mutation ) I . This strain was generated using a CRISPR approach detailed below . SMW18: ( SMW16 x OD56 ) Pklp-16::klp-16::GFP ( C1971>A–PAM site mutation ) I; ltIs37 [ ( pAA64 ) pie-1::mCherry::his-58 + unc-119 ( + ) ] IV A CRISPR-based approach [54 , 55] was used to generate an endogenously tagged KLP-16::GFP strain ( SMW16 ) . Briefly , 27μM recombinant Cas9 protein ( IDT ) was co-injected with 13 . 6μM tracrRNA ( IDT ) , 4μM dpy-10 crRNA ( 5’—GCUACCAUAGGCACCACGAG- 3’ ) ( IDT ) , 1 . 34μM dpy-10 repair oligo ( Ultramer from IDT; 5’ -CACTTGAACTTCAATACGGCAAGATGAGAATGACTGGAAACCGTACCGCATGCGGTGCCTATGGTAGCGGAGCTTCACATGGCTTCAGACCAACAGCCTAT- 3’ ) , 9 . 6μM klp-16 crRNA ( 5’—UGUCUAGUUCAUAGACAUCU- 3’ ) ( IDT ) ; and 136ng/μL ssDNA klp-16 repair template into N2 worms , that were then allowed to produce progeny . Worms from plates containing rollers and dumpys were screened for GFP expression , and homozygous KLP-16::GFP worms were identified by PCR screening . To make the klp-16 repair template , we generated a C-terminal LAP tag using a GBlock ( IDT ) and Gibson Assembly to create an S-TEV-GFP construct . The tag was then amplified using PCR with primers that contained homology to the klp-16 gene with the final product containing 68 bp of homology upstream of the klp-16 stop codon and 100 bp of homology downstream of the stop codon . ssDNA was generated by asymmetric PCR . SMW16 ( KLP-16::GFP ) was also crossed with OD56 ( mCherry::histone ) to generate SMW18: Pklp-16::klp-16::GFP ( C1971>A—PAM site mutation ) I; ltIs37 [ ( pAA64 ) pie-1::mCherry::his-58 + unc-119 ( + ) ] IV . A CRISPR-based approach similar to the one above was used to generate a worm strain with a ~600 bp deletion in the klp-16 locus beginning ~100 bp upstream of the start codon . Essentially the same approach was used as above; the differences being two crRNAs ( 4 . 8μM ) ( 5’- AGGCGGAGUUUAAGUUUGAG-3’ and 5’- CUCCUCAAGAAGCGUCACUU-3’ ) ( IDT ) ( one upstream and one downstream of the klp-16 start codon , respectively ) , and a ssDNA oligo ( 4μM ) ( Ultramer from IDT; 5’-CAGCCATCTCACGCTCCAATTGCGCATTTCTCTCCTCAAGAAGCGTCACTTCTCAAACTTAAACTCCGCCTCTGAAAATTCCCGCCAAATCGGATGGATTAC-3’ ) were used in the injection mix . The repair Ultramer sequence is homologous to the sequence just upstream and downstream to the two CRISPR cut sites thereby deleting the ~600 base pairs . Worms from plates containing rollers and dumpys were screened by PCR and homozygous mutants were isolated . Protein domains were determined using PsiPred [56] and Paircoil2 [57] . Protein sequences were analyzed using Clustal Omega [58] . Proline-rich regions of proteins have been shown to bind microtubules [46] . The proline content of amino acids 1–149 is 14% for KLP-15 and 13% for KLP-16 . From a feeding library [26 , 59] , individual RNAi clones were picked and grown overnight at 37°C in LB with 100μg/ml ampicillin . Overnight cultures were spun down and plated on NGM ( nematode growth media ) plates containing 100μg/ml ampicillin and 1mM IPTG . Plates were dried overnight . Worm strains were synchronized by bleaching gravid adults and letting the eggs hatch overnight without food . L1s were then plated on RNAi plates and grown to adulthood at 15° for 5–6 days . Young adult worms grown on control plates or plates containing RNAi-expressing bacteria were transferred to new plates containing either control or RNAi-expressing bacteria and allowed to lay eggs for 24 hours at 15°C before being moved to another fresh plate of either control or RNAi-expressing bacteria . The eggs were allowed to hatch for 24 hours and then the progeny ( eggs and hatched worms ) were counted . For each parent worm this process was repeated twice , resulting in three days of progeny being counted . For each condition , the progeny of at least 15 worms were scored . Immunofluorescence was performed by freeze cracking embryos and plunging into -20°C methanol as described [60] . Embryos were fixed for 35–45 minutes , rehydrated in PBS , and blocked in AbDil ( PBS plus 4% BSA , 0 . 1% Triton X-100 , 0 . 02% Na-Azide ) for 30 minutes . Primary antibodies were incubated overnight at 4°C . The next day , embryos were washed 3x with PBST ( PBS plus 0 . 1% Triton X-100 ) , incubated in secondary antibody for 1 hour and 15 minutes , washed again as before , incubated in mouse anti-α-tubulin-FITC for 1 . 5 hours , washed again , and incubated in Hoechst ( 1:1000 in PBST ) for 15 minutes . Embryos were then washed 2x with PBST , mounted in 0 . 5% p-phenylenediamine , 20mM Tris-Cl , pH 8 . 8 , 90% glycerol or ProLong Gold antifade Mountant ( Molecular Probes ) , and sealed with nail polish; except for the overnight primary , the entire procedure was performed at room temperature . For experiments using the rabbit anti-KLP-16 antibody and staining of SPD-1::GFP with mouse anti-GFP , embryos were blocked in AbDil overnight at 4°C and incubated in primary antibody for 2 hours at room temperature . Primary antibodies used in this study: rabbit anti-ASPM-1 ( 1:5000 , gift from Arshad Desai ) , rabbit anti-SEP-1 ( 1:400; gift from Andy Golden ) , rabbit anti-KLP-18 ( 1:10 , 000 , gift from O . Bossinger ) , rabbit anti-ZEN-4 ( 1:500; gift from Michael Glotzer ) , mouse anti-SUMO ( 1:500; gift from Federico Pelisch ) , mouse anti-GFP ( 1:200; Invitrogen ) . Rat anti-AIR-2 was generated by Covance using the C-terminal peptide sequence KIRAEKQQKIEKEASLRNH ( synthesized by the Peptide Synthesis Core Facility at Northwestern University ) , then affinity purified and used at 1:1000 . Rabbit anti-KLP-16 was generated by Covance using the N-terminal peptide sequence CMNVARRRSGLFRSTIGAPPK ( synthesized by the Peptide Synthesis Core Facility at Northwestern University ) , then affinity purified and used at 1:2000 . Rabbit anti-SPD-1 was generated by Proteintech using the C-terminal peptide sequence CIASSTPSSAKKVLTRRNQFL , then affinity purified and used at 1:1000 . Directly conjugated mouse anti-α-tubulin-FITC ( DM1α , Sigma ) and Alexa-fluor directly conjugated secondary antibodies ( Invitrogen ) were used at 1:500 . All antibodies were diluted in AbDil . All fixed imaging and high resolution imaging of KLP-16::GFP and KLP-16::GFP;mCherry::histone was performed on a DeltaVision Core deconvolution microscope with a 100x objective ( NA = 1 . 4 ) ( Applied Precision ) . This microscope is housed in the Northwestern University Biological Imaging Facility supported by the NU Office for Research . Image stacks were obtained at 0 . 2μm z-steps and deconvolved using SoftWoRx ( Applied Precision ) . All images in this study were deconvolved and displayed as full maximum intensity projections of data stacks encompassing the entire spindle structure , unless stated otherwise . For KLP-16::GFP and KLP-16::GFP;mCherry::histone imaging , live worms were mounted in anesthetic ( 0 . 2% tricaine , 0 . 02% levamisole in M9 ) . Two-color live imaging was performed using a spinning disk confocal microscope with a 63x HC PL APO 1 . 40 NA objective lens . A spinning disk confocal unit ( CSU-X1; Yokogawa Electric Corporation ) attached to an inverted microscope ( Leica DMI6000 SD ) and a Spectral Applied Imaging laser merge ILE3030 and a back-thinned electron-multiplying charge-coupled device ( EMCCD ) camera ( Photometrics Evolve 521 Delta ) were used for image acquisition . The microscope and attached devices were controlled using Metamorph Image Series Environment software ( Molecular Devices ) . Typically , ten to twelve z-stacks at 1μm increments were taken every 20–45 seconds at room temperature . Image deconvolution was done using AutoQuant X3 ( Media Cybernetics Inc . ) . Images were processed using ImageJ . Images are shown as maximum intensity projections of entire spindle structure . Live , intact worms were mounted on 5% agarose , M9 pads in 50% live imaging solution ( modified S-basal [50mM KH2PO4 , 10mM K-citrate , 0 . 1M NaCl , 0 . 025mg/ml cholesterol , 3mM MgSO4 , 3mM CaCl2 , 40mM serotonin creatinine sulfate monohydrate] ) , 50% 0 . 1 micron polystyrene Microspheres ( Polysciences Inc . ) , and covered with a coverslip . The spinning disk microscope is housed in the Northwestern University Biological Imaging Facility supported by the NU Office for Research . For S3 Movie , EU1067 worms were picked into a solution of tricaine ( 2% ) and tetramisole ( 0 . 4% ) , and incubated for ~30 min . Worms were then pipetted onto a 3% agarose pad , covered with a coverslip , and imaged immediately on a DeltaVision Core deconvolution microscope ( same as above ) . Image stacks were obtained at 1μm z-steps at 10 second intervals using 2 × 2 binning , and then deconvolved . Video images are full projections of data stacks . Fig 1D: Slides made on the same day were imaged within an 8 hour window on a DeltaVision Core deconvolution microscope ( see Microscopy section ) using the same exposure conditions and times for all slides . In ImageJ , linescans of 154 x 75 pixels ( L x W ) were performed on 6 z-slice sum projections of representative spindles from control ( n = 8 ) and klp-15/16 ( RNAi ) ( n = 9 ) embryos . In control spindles , the linescans were done along the pole-to-pole axis . In spindles from klp-15/16 ( RNAi ) embryos , linescans were done straight along the x-axis of the image , since these spindles lack a discernable orientation . The average fluorescence intensity for each channel was graphed ( solid line ) along with the SEM ( standard error of the mean ) ( shaded area ) using the ggplot package in R Studio . The y-axes of the graphs are the same between control and experiment for a given channel . Fig 2B: α-KLP-16 staining for each stage of spindle assembly in wild-type oocytes/embryos . Oocytes in prophase with the nuclear envelope intact were scored as “localized” if the KLP-15/16 signal was primarily cytoplasmic . During all other stages , oocytes were scored as “localized” if the KLP-15/16 signal was colocalized with spindle microtubules . The quantification is as follows: diakinesis 81 . 8% ( n = 22 ) , cage 0% ( n = 13 ) , multipolar 31 . 3% ( n = 67 ) , bipolar 64% ( n = 114 ) , anaphase 51 . 9% ( n = 79 ) , mitotic spindles 3 . 7% ( n = 27 ) . Although not every spindle is stained , we think that this represents variability with the immunofluorescence procedure and with the antibody ( since we see 100% localization of KLP-16::GFP to oocyte spindle microtubules and to mitotic spindles; see S4A Fig ) . Fig 2C: α-KLP-16 staining was scored in klp-15/16 ( RNAi ) oocytes . The number of oocytes in which we could discern any spindle staining is as follows: microtubule ball stage 8% ( n = 75 ) , anaphase 2 . 9% ( n = 35 ) . α-KLP-16 staining was scored in klp-15 ( ok1958 ) oocytes . Oocytes in prophase with the nuclear envelope intact were scored as “localized” if the KLP-15/16 signal was primarily cytoplasmic . During all other stages , oocytes were scored as “localized” if the KLP-15/16 signal was colocalized with spindle microtubules . The quantification is as follows: diakinesis 100% ( n = 3 ) , cage 14% ( n = 7 ) , multipolar 44% ( n = 25 ) , bipolar 72 . 2% ( n = 18 ) , anaphase 38% ( n = 21 ) . α-KLP-16 staining was scored in klp-16 ( wig1 ) oocytes as above . The quantification is as follows: diakinesis 100% ( n = 10 ) , cage 42 . 9% ( n = 7 ) , multipolar 89% ( n = 19 ) , bipolar 81 . 8% ( n = 11 ) , anaphase 79 . 2% ( n = 24 ) . As with the control strain ( see Fig 2B quantification above ) , we think that the incomplete staining we observe is due to variability with the procedure and antibody . Fig 3B: Linescans of control anaphase spindles and anaphase spindles from klp-15/16 ( RNAi ) oocytes stained for ASPM-1 were performed using the arbitrary profile tool in SoftWorx ( Applied Precision ) . A spindle was scored as having staining at poles if the ASPM-1 signal was enriched at two ends of the spindle near the segregating chromosomes . ASPM-1 was enriched at the poles of 21/24 control spindles , but was largely diffuse along spindle microtubules in anaphase of klp-15/16 ( RNAi ) oocytes ( only 9/26 spindles could be classified as having any type of ASPM-1 enrichment , and this enrichment was not as strong as in the control spindles ) . Fig 3C: Linescans of control anaphase spindles and anaphase spindles from klp-15/16 ( RNAi ) oocytes stained for KLP-18 were performed using the arbitrary profile tool in SoftWorx ( Applied Precision ) . A spindle was scored as having staining at poles if the KLP-18 signal was enriched at two ends of the spindle near the segregating chromosomes . KLP-18 was enriched at poles in 11/12 control spindles but was diffuse in klp-15/16 ( RNAi ) spindles ( 0/6 had KLP-18 concentrated into poles ) . Fig 3D: Aneuploidy in MII embryos was quantified by counting the number of chromosomes in MII in immunofluorescence images of control and klp-15/16 ( RNAi ) embryos . An embryo was scored as ‘aneuploid’ if the number of chromosomes was not 6 . Fig 4A and 4B: ZEN-4 and SPD-1 staining was scored in control spindles and spindles from klp-15/16 ( RNAi ) oocytes . Staining of metaphase and anaphase spindles ( staged by AIR-2 localization ) was scored for each condition . ZEN-4: Control metaphase 1 . 6% ( n = 63 ) , Control anaphase 90% ( n = 20 ) ; klp-15/16 ( RNAi ) metaphase 2 . 9% ( n = 134 ) , klp-15/16 ( RNAi ) anaphase 80 . 3% ( n = 66 ) . SPD-1: Control metaphase 9 . 5% ( n = 42 ) , Control anaphase 97% ( n = 66 ) ; klp-15/16 ( RNAi ) metaphase 1% ( n = 100 ) , klp-15/16 ( RNAi ) anaphase 88 . 2% ( n = 85 ) . Fig 5A and 5B: Quantification of microtubule bundling and chromosome segregation was done using immunofluorescence images of anaphase spindles with SEP-1 gone and AIR-2 relocalized to the microtubules ( mid/late anaphase ) for the conditions shown . We scored microtubule bundling by eye , looking through the entire z-stack in SoftWorx ( Applied Precision ) . An anaphase spindle was scored as “bundled” if one or more microtubule bundles were discernable . Chromosomes were scored as “segregated” if two or more distinct masses of chromosomes were observed . The simple matching coefficient ( SMC ) for microtubule bundling and chromosome segregation = 0 . 82 ( n = 251 ) ; in other words , 82% of the spindles were scored as microtubules bundled and chromosomes segregated or as no microtubule bundles and no chromosome segregation . Fig 5C: To approximate anaphase microtubule lengths , we used the measure distances tool in SoftWorx ( Applied Precision ) . Using this tool , a line was manually drawn ( point by point ) along the most prominent spindle microtubule bundle through the 3D stack of an image to measure its full length . Fig 6A and 6B: Polar body number and maternal pronuclei number were quantified by scoring live EU1067 worms mounted in anesthetic ( 0 . 2% tricaine , 0 . 02% levamisole in M9 ) on a Leica DM5500B fluorescent microscope . Fig 7A: Linescans of control anaphase spindles and anaphase spindles from klp-15/16 ( RNAi ) oocytes stained for tubulin , SUMO , and SPD-1 were performed using the arbitrary profile tool in SoftWorx ( Applied Precision ) . A spindle was scored as having oscillations if one or more instances of alternating MTs/SUMO and SUMO/SPD-1 signal was observed . This analysis was done by examining both single z-slices and max projections of spindles . Oscillations were observed in 9/11 control anaphase spindles and 12/18 klp-15/16 ( RNAi ) anaphase spindles . Fig 7C: Lateral microtubule associations to chromosomes were scored in control anaphase spindles and anaphase spindles from klp-15/16 ( RNAi ) oocytes . A spindle was scored as having lateral microtubule/chromosome associations if a microtubule appeared to contact and run along the side of a chromosome . This analysis was done by examining both single z-slices and max projections of spindles . We observed clear lateral associations in 31/35 control anaphase spindles and in 22/31 klp-15/16 ( RNAi ) anaphase spindles . S2C Fig: Live , intact worms expressing GFP::tubulin , GFP::histone ( EU1067 ) fed either control or klp-16 ( RNAi ) -expressing bacteria were anesthetized in 0 . 2% tricaine , 0 . 02% levamisole in M9 and viewed on a Leica DM5500B widefield fluorescence microscope . Spindles in embryos in the -1 , spermatheca , and +1 positions within the gonad were scored for microtubule organization by eye . A spindle was scored as “multipolar” if it had prominent microtubule bundles that formed more than two organized poles . A spindle was scored as an “array” if the microtubule structure lacked prominent bundles and organized poles . A spindle was scored as “MT ball” if the microtubule structure had collapsed around the chromosomes and lacked prominent bundles and organized poles . S3B Fig: Spindle volumes were measured using the surfaces tool in Imaris ( Bitplane ) . Using the full 3D image stack , this tool renders a 3D surface based on fluorescence signal ( for our analysis , we used the tubulin signal ) . The volume of this 3D surface is then measured . The volumes of metaphase and early anaphase spindles ( staged by SEP-1/AIR-2 localization ) from klp-15/16 ( RNAi ) oocytes were measured and compared to the volumes of the spindles used for the linescan measurements in Fig 1D . This analysis allowed us to conclude that the spindles used in our linescans for Fig 1D are within the range of metaphase spindles based on spindle volume . S4A Fig: Live , intact worms expressing KLP-16::GFP , mCherry::histone ( SMW18 ) were anesthetized in 0 . 2% tricaine , 0 . 02% levamisole in M9 and viewed on a Leica DM5500B widefield fluorescence microscope . The localization of KLP-16::GFP was scored in oocytes/embryos in the -1 , spermatheca , and +1 positions within the gonad . KLP-16::GFP signal was scored as cytoplasmic if it was absent/reduced in the nucleus . Because the localization of KLP-16::GFP on the spindle looks identical to the organization of GFP::tubulin , we scored localization for the following categories: cage , multipolar , bipolar , and anaphase . The organization of the chromosomes visualized by mCherry::histone was used to identify and better stage the spindles . S4C Fig: α-GFP staining for each stage of spindle assembly in SMW16 ( KLP-16::GFP ) . Oocytes in prophase with the nuclear envelope intact were scored as “localized” if the α-GFP signal was primarily cytoplasmic . During all other stages , oocytes were scored as “localized” if the α-GFP signal was colocalized with spindle microtubules . The quantification is as follows: diakinesis 100% ( n = 2 ) , cage 0% ( n = 4 ) , multipolar 76 . 5% ( n = 17 ) , bipolar 96 . 2% ( n = 26 ) , anaphase 83 . 3% ( n = 24 ) , mitotic spindles 100% ( n = 5 ) . As with the KLP-15/16 antibody , we think that the lack of staining in all embryos represents variability with the immunofluorescence procedure , since we see robust localization of KLP-16::GFP to microtubules when we visualize this strain live . Seventy-five EU1067 , RB1593 , or SMW15 worms were picked off of control , klp-16 ( RNAi ) ( EU1067 and SMW15 ) , or klp-15 ( RNAi ) ( RB1593 ) plates onto new , empty ( no bacteria ) plates . The worms were washed off the plates with cold M9 and transferred to a 1 . 5ml microcentrifuge tube . Worms were pelleted by spinning at 2000 rpm for 1 minute , and the tube was put on ice for ~2 minutes to allow worms to slow down and form a tight pellet . The M9 was removed and the tube was filled with fresh , cold M9 and mixed . The worms were washed a total of 3 times . After the final wash , as much M9 was removed as possible and 2X SDS sample buffer was added to the remaining worm/M9 mixture and boiled for 10 minutes . Samples were run on a 10% SDS-PAGE gel and blotted . For western analysis , rabbit anti-KLP-16 antibody ( 1:10 , 000 ) and mouse anti-tubulin ( 1:5000 ) ( Sigma , DM1α ) were used .	When cells divide , they must assemble a microtubule-based structure called a spindle on which the chromosomes are segregated . While in most cell types the microtubules that comprise the spindle are nucleated and organized by centriole-containing centrosomes , female reproductive cells ( oocytes ) of many species lack centrioles and therefore spindles in these cells assemble using unique mechanisms . Using C . elegans as a model system , we set out to identify factors required for acentriolar spindle assembly in oocytes and found two microtubule motor proteins necessary for this process . When these motors are depleted in oocytes , microtubules fail to form stable bundles during spindle assembly , resulting in severely aberrant spindles . However , we were surprised to find that these disorganized microtubules were then able to reorganize into a spindle capable of segregating chromosomes during anaphase , revealing a second mechanism that can act to bundle and organize spindle microtubules . Studies of this unique anaphase reorganization phenotype then enabled us to uncover new proteins contributing to spindle organization , and to gain insights into the mechanisms driving chromosome segregation in this important cell type . The work presented here therefore deepens our understanding of the molecular mechanisms of acentriolar spindle assembly and chromosome segregation in oocytes .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "invertebrates", "rna", "interference", "microtubules", "anaphase", "chromosome", "structure", "and", "function", "caenorhabditis", "metaphase", "cell", "cycle", "and", "cell", "division", "cell", "processes", "animals", "germ", "cells", "animal", "models", "developmental", "biology", "oocytes", "caenorhabditis", "elegans", "model", "organisms", "experimental", "organism", "systems", "epigenetics", "embryos", "cellular", "structures", "and", "organelles", "cytoskeleton", "research", "and", "analysis", "methods", "embryology", "animal", "cells", "chromosome", "biology", "genetic", "interference", "gene", "expression", "biochemistry", "rna", "eukaryota", "cell", "biology", "ova", "nucleic", "acids", "genetics", "nematoda", "biology", "and", "life", "sciences", "cellular", "types", "organisms", "chromosomes" ]	2017	Interplay between microtubule bundling and sorting factors ensures acentriolar spindle stability during C. elegans oocyte meiosis
How nonenveloped viruses such as simian virus 40 ( SV40 ) trigger the lytic release of their progeny is poorly understood . Here , we demonstrate that SV40 expresses a novel later protein termed VP4 that triggers the timely lytic release of its progeny . Like VP3 , VP4 synthesis initiates from a downstream AUG start codon within the VP2 transcript and localizes to the nucleus . However , VP4 expression occurs ∼24 h later at a time that coincides with cell lysis , and it is not incorporated into mature virions . Mutation of the VP4 initiation codon from the SV40 genome delayed lysis by 2 d and reduced infectious particle release . Furthermore , the co-expression of VP4 and VP3 , but not their individual expression , recapitulated cell lysis in bacteria . Thus , SV40 regulates its life cycle by the later temporal expression of VP4 , which results in cell lysis and enables the 50-nm virus to exit the cell . This study also demonstrates how viruses can generate multiple proteins with diverse functions and localizations from a single reading frame . Eukaryotic cells use a range of mechanisms and pathways to shuttle molecules across their complex endomembrane network [1] . Viruses exploit these pathways to gain entry into the cell and navigate through the various membrane barriers [2 , 3] . However , no known cellular pathways exist for exporting nonenveloped DNA viral progeny from their intracellular site of assembly to the extracellular milieu . Therefore , nonenveloped viruses , including simian virus 40 ( SV40 ) , initiate their release from the host cell in a lytic manner . How these cytolytic viruses kill the host cell and rupture its membrane system to facilitate release remains largely unknown [4] . For many years , viral-induced cell lysis was thought to be a nonspecific consequence of viral protein overexpression or the production of large amounts of viral progeny causing changes in cell membrane permeability . However , the necessity for the lytic event to occur in a timely fashion after the completion of viral assembly suggests that cytolytic viruses have evolved a method to ensure that the induction of cell death takes place at the proper time . The enterovirus 2B protein , for example , is proposed to aid in the lytic release process due to its ability to induce calcium leakage from the endoplasmic reticulum ( ER ) and alter cell permeability [5] . The adenovirus death protein has also been implicated in viral release because its removal significantly prolongs the adenoviral life cycle [6] . In SV40 , the core structural proteins VP2 and VP3 are capable of permeabilizing bacterial membranes and post-translationally integrating into ER membranes in a VP1-regulated manner [7 , 8] . While these data are supportive of VP2 and VP3 potentially playing a role in the lytic release of SV40 , it remains to be determined if other viral or host cellular proteins are involved in this process . SV40 encodes two early gene products ( small and large T antigens ) , four late gene products ( VP1 , VP2 , VP3 , and agno ) , and microRNAs that help regulate the temporal expression of these proteins [9–11] . In this manuscript , we have identified a new SV40 gene product called VP4 that is expressed ∼24 h after the late structural proteins VP1 , VP2 , and VP3 . VP4 is essential for the timely lytic release of the viral progeny that enables the efficient spreading of SV40 in culture . VP4 oligomerizes with VP3 , and when expressed together , these two proteins possess a lethal lytic property that is conserved in bacteria . These data demonstrate that late in the replication process , following viral assembly , SV40 expresses VP4 to initiate the death of the host cell and the efficient release of its progeny . The in vitro translations of mRNAs encoding VP2 and VP3 from SV40 unexpectedly produced a protein which migrated at ∼15 kD , in addition to VP2 and VP3 ( Figure 1B , lanes 1 and 3 , asterisks ) . The coding sequence for VP2 contains four highly conserved in-frame AUG codons ( Figures 1A and S1 , Met residues ) . To investigate which of these potential AUG initiation sites was responsible for the synthesis of the smaller protein , an algorithm that predicts AUG initiation sites ( NetStart1 . 0 ) was employed [12] . Surprisingly , the highest probability score for an AUG translation initiation site corresponded to Met228 in VP2 , which , if utilized , would yield a protein with a molecular mass of 13 . 9 kD ( Figure 1A , scores in parentheses ) . Mutation of the ATG codon for Met228 in VP2 to ATA ( Ile ) in both VP2 and VP3 prevented the in vitro synthesis of the ∼15-kD protein , which we have termed VP4 ( Figure 1B ) . Note that in later figures , the VP2-Met228Ile mutant is designated as ΔVP4 . To investigate whether VP4 was expressed during infection , a series of mutant SV40 genomes were generated in the bacterial replication competent plasmid pSV40 [8 , 13 , 14] . In addition , the VP2-M295I genome had the ATG codon corresponding to Met295 ( VP2-M295I ) in VP2 mutated to ATA ( Ile ) . The ΔVP2/3 genome had both ATG initiation codons for VP2 ( Met1 ) and VP3 ( Met119 ) altered to ATA ( Ile ) , whereas the ΔVP2/3/4 genome had the initiation codons for VP2 , VP3 , and VP4 altered . The wild-type ( WT ) and mutant genomes were transfected into permissive BS-C-1 cells . The cells were collected 5 d post-transfection , after VP1 , VP2 , and VP3 expression , and the post-nuclear and nuclear lysates were analyzed by immunoblotting , because all of these proteins are synthesized in the cytoplasm and imported into the nucleus for viral assembly . All of the genomes correctly expressed and localized both the early viral protein large T antigen ( LT ) and the major capsid protein VP1 ( Figure 1C ) . Late viral protein synthesis from the VP2 reading frame was examined by immunoblotting with polyclonal antisera against VP3 ( αVP2/3 ) , which also recognizes VP2 . Cells transfected with WT SV40 or the VP2-M295I genome expressed three proteins that reacted with VP3 antisera: VP2 , VP3 , and the ∼15 kD VP4 ( Figure 1C , lanes 1 and 3 , αVP2/3 ) . In contrast , the ∼15-kD VP4 protein was not observed following transfection with the ΔVP4 and the ΔVP2/3/4 genomes ( Figure 1C , lanes 2 and 5 , αVP2/3 ) . Importantly , VP4 was synthesized in cells transfected with the ΔVP2/3 genome , which indicates that it was not produced by the proteolytic cleavage of VP2 or VP3 . In addition , VP4 predominantly localized to the nucleus upon cellular fractionation , whereas VP2 and VP3 were present in both the post-nuclear and nuclear samples ( Figures 1C and S2 ) . Significantly , this indicated that during infection , SV40 expresses the ∼15-kD VP4 protein by initiating its synthesis from Met228 in the VP2 reading frame , resulting in a polypeptide that corresponds to the C-terminal 125 amino acids of VP2 and VP3 ( Figure 1A ) . To determine the effects of the various mutations on SV40 propagation , the number of BS-C-1 cells expressing LT over the course of infection was quantified by immunofluorescent microscopy . Contrary to plaque assays , which indirectly measure viral infectivity based on cell death , this method directly monitors viral spreading by determining the number of cells expressing the viral LT [7 , 8] . Following transfection , equivalent numbers of primary infections were established from the various genomes as ∼3% of the cells expressed LT after 2 d ( Figure 2A and 2B ) . As expected , the particles generated in cells transfected with the ΔVP2/3/4 genome were incapable of establishing secondary infections . The viral particles produced from the VP2-M295I genome propagated as efficiently as WT , infecting ∼65% of the cells by 7 d , indicating that synthesis of a protein from the AUG codon corresponding to Met295 in VP2 was not required for SV40 infection . In contrast to WT and VP2-M295I , ΔVP4 only infected ∼10% of the cells by 7 d . This demonstrated that VP4 played a role in viral propagation , as limited cell to cell spreading was observed in its absence . In addition , co-transfection of ΔVP2/3 with ΔVP4 partially complemented the ΔVP4 mutation , as ∼20% of the cells were positive for LT by day 7 ( Figure 2A and 2B ) . Upon cell fractionation , VP4 was predominantly found in the nuclear pellet of cells infected or transfected with the SV40 genome ( Figure S2 and unpublished data ) . To more thoroughly explore the cellular localization of VP4 , immunofluorescence microscopy was employed . Because the polyclonal antibody raised to VP3 recognizes VP2 , VP3 , and VP4 , the localization of VP4 was analyzed after transfection with the ΔVP2/3 genome . This enabled any signal from the VP3 antisera to be attributed to VP4 and allowed the localization to be examined in a context similar to that of a viral infection , as LT and VP1 are still expressed . At 3 d post-transfection , VP4 was found to localize to the nucleus , where it was enriched at the inner nuclear membrane periphery and in punctate structures ( Figure 2C ) . In addition , VP4 appeared to support a significant alteration in the nuclear morphology . The nuclei , as defined by LT , showed an obvious circular shape and size increase in cells transfected with the ΔVP2/3 genome , which expresses VP4 . This was a drastic change from the cells transfected with the ΔVP2/3/4 genome , where VP4-like immunostaining was not observed and the cell nuclei were a smaller kidney shape ( Figure 2C ) . Thus , VP4 appears to contribute to the expansion of the nucleus that occurs prior to the lytic release of SV40 . To elucidate the role of VP4 in the viral life cycle , we initially determined whether VP4 was incorporated into the mature virion , as it contains the proposed VP1 pentamer–binding domains found in VP2 and VP3 [15 , 16] . Immunoblot analysis with VP2/3 antisera of SV40 virions isolated by filtration and centrifugation indicated that VP4 was not found in SV40 particles , but significant quantities were present in cellular lysates ( Figure 3A ) . Thus , VP4 is a nonstructural viral protein that must perform an integral function in the host cell to promote viral propagation . To test if the absence of VP4 from the SV40 particles was due to its inability to associate with VP1 , binding to VP1 pentamers was examined . VP1 pentamers were produced from bacteria by expressing a construct lacking the VP1 C-terminal arms ( VP1Δarm-His ) , which are necessary for interpentameric association and capsid assembly but are not required for VP2 and VP3 binding [8 , 17] . In vitro translated radiolabeled VP2 , VP3 , and VP4 were synthesized in the presence of VP1 pentamers ( co-translationally ) , or incubated with the VP1 pentamers after synthesis was completed ( post-translationally ) . Following isolation of the VP1 pentamers , a significant fraction of the in vitro translated VP2 and VP3 ( ∼30% above background ) was found to associate with the VP1 pentamers both co- and post-translationally ( Figure 3B ) . In contrast , only slight levels ( ∼3%–7% above background ) of VP4 associated with the VP1 pentamers using either mechanism ( Figure 3B ) . The absence of VP2 or VP4 from the translation reaction had little effect on the association of VP3 with the VP1 pentamers ( Figure S3A ) . Thus , the lack of VP4 binding to VP1 possibly explains why it is not found in the viral particle . To gain further insight into what role VP4 performs in the host cell that contributes to the viral life cycle , its temporal expression was examined in relation to the late proteins . VP2 and VP3 were first observed at 48 h post-transfection ( Figure 3C ) . Strikingly , VP4 initially accumulated at 72 h post-transfection , 12 to 24 h later than the start of VP1 , VP2 , and VP3 expression ( Figure 3C and [7 , 8] ) . The temporal separation of VP4 synthesis from viral assembly provided an additional explanation as to why VP4 was excluded from incorporation into SV40 particles . Furthermore , viral-induced cell lysis also initiated at 72 h post-transfection as monitored by the trypan blue staining of cells ( Figure 3C , bottom ) . The requirement of VP4 for efficient SV40 propagation and the finding that it is a nonstructural protein synthesized at later times coinciding with host cell lysis support a function for VP4 in viral-induced lysis of the host cell . VP3 expression has been previously shown to induce bacterial lysis by possessing an inherent lytic property , which suggests that VP3 may act as a viroporin [7] . The apparent involvement of VP4 in the lytic release of SV40 led us to investigate whether VP4 expression from the VP3 construct resulted in the previously observed bacterial lysis caused by VP3 . To monitor bacterial lysis , the expression of C-terminal His-tagged VP3 ( VP3-His ) , VP3-His with the VP4 initiation codon mutated to ATA ( VP3-M228I-His ) , VP4-His , and the negative control VP1-His was induced with isopropyl-β-D-thiogalactopyranoside ( IPTG ) , and the optical density ( OD ) at 600 nm of the Escherichia coli suspension was measured . As previously observed , the OD of the E . coli suspension significantly decreased following the induction of VP3-His expression ( Figure 4A ) [7] . Strikingly , E . coli expressing the VP3-M228I-His construct ( without the VP4 initiation codon ) , or VP4-His alone , remained viable and entered stasis as the OD reached a plateau similar to the OD of VP1-His ( Figure 4A ) . These findings indicate that the co-expression of both VP4 and VP3 is required to induce bacterial lysis . Next , an assay was utilized to characterize the viroporin activity of the late proteins and determine whether they compromise the bacterial membrane by forming lethal or nonlethal membrane perturbations . E . coli membrane permeability was examined by monitoring the sensitivity of the bacteria to the membrane-impermeable protein synthesis inhibitor hygromycin B following expression of VP3-His , VP3-M228I-His , VP4-His , and the negative control VP1-His ( for experimental design , see Figure S4 and [7] ) . Prior to induction with IPTG , hygromycin B had little effect on protein synthesis , as the E . coli proteins were readily labeled with 35S-Met/Cys ( Figure 4B , lanes 1 and 2 ) . At 40 and 70 min post-induction with IPTG , detectable levels of VP3-His were not produced in the absence or presence of hygromycin B . The lack of 35S-labeled VP3-His being observed was likely due to the lysis of the E . coli expressing both VP3-His and VP4-His before reaching synthesis levels that were sufficient for visualization ( Figure 4A and 4B ) . In contrast , all of the nonlethal constructs ( VP1-His , VP3-M228I-His , and VP4-His ) were expressed at detectable quantities in the absence of hygromycin B following IPTG induction ( Figure 4B , asterisks ) . Of the nonlethal constructs ( determined from Figure 4A ) , only VP3M228I-His rendered the bacterial membranes permeable to hygromycin B , as the inhibitor prevented the synthesis of 35S-labeled VP3-M228I-His . Together , these findings demonstrate that in the absence of VP4 , VP3 is capable of permeabilizing membranes in a nonlethal fashion . However , in the presence of VP4 , VP3 has a lethal property that results in membrane lysis . Small pathogenic proteins that form pores and induce cell lysis generally insert into cellular membranes in an oligomeric state [18] . The observed bacterial lysis following the co-expression of VP4 with VP3 , but not the individual expression of either construct , suggested that the lytic property was a result of VP4 oligomerization with VP3 . To investigate whether VP4 associates with VP3 , VP4 synthesized from VP2 and VP3 mRNAs was monitored for its ability to bind GST-VP3 purified from bacteria . VP4 ( ∼45% bound ) showed a substantially higher affinity for sepharose-bound GST-VP3 than VP2 and VP3 ( ∼25%–30% bound ) , the positive control VP1 ( ∼12% bound ) , and the negative control luciferase ( Figures 4C and S3B ) . These data imply that the resultant lethality observed in bacteria co-expressing VP3 and VP4 may be due to a change in the physical properties of VP3 as a consequence of its association with VP4 . To determine how the absence of VP4 affected the normal course of infection , SV40 particles were generated from BS-C-1 cells transfected with either the WT SV40 or ΔVP4 genomes . In BS-C-1 cells , both the viral-mediated and transfection-mediated life cycles of SV40 have been well characterized with lengths of 2 . 5 to 4 d and 3 to 5 d , respectively [7 , 8] . After transfection , ∼2% of the BS-C-1 cells were infected with each genome as determined by LT expression at 2 d . Trypan blue staining revealed that these low level transfection-initiated ΔVP4 infections required ∼24 d , or the length of six WT life cycles , to lyse the majority of the cell population . This was significantly longer than the similar low level transfection-initiated WT SV40 infection , which required ∼12 d , or three life cycles , to completely lyse a cell population of equivalent size . The viral-containing supernatant was isolated from the ΔVP4-transfected cells at 24 d and the WT-transfected cells at 12 d . The number of LT-expressing BS-C-1 cells was then calculated at 2 d post-infection with equivalent fractions of the ΔVP4 and the WT virus-containing supernatants isolated at 24 and 12 d post-transfection , respectively . Strikingly , ∼95% fewer cells were expressing LT at 2 d post-infection with the ΔVP4 virus-containing supernatant compared to those with the WT virus-containing supernatant . The aforementioned deficiencies suggested that the ΔVP4 strain had a prolonged life cycle due to a defect in the lytic release of its viral progeny . To investigate this possibility , BS-C-1 cells infected with equivalent amounts of infectious WT and ΔVP4 particles ( Figure 5A , ∼6% LT-positive cells at 2 d ) were monitored for viral-induced cell death by trypan blue staining . The primary WT-infected BS-C-1 cells began to die at 3 d post-infection , reaching a plateau at 4 d until the secondary infected cells started to lyse at 7 d ( Figure 5B ) . The primary ΔVP4-infected cells showed a noticeable 2-d lag in lysis , which initiated at ∼5 d and completed by 6–7 d ( Figure 5B ) . Together , these results demonstrate that SV40 expresses a novel later protein ( VP4 ) that initiates the efficient lytic release of its progeny from the host cell . SV40 has been extensively studied for the past 50 years , yet how it and other noneveloped cytolytic viruses cause cell lysis to facilitate progeny release is relatively unknown . However , new discoveries into the regulation of its temporal viral protein expression have recently been made [11] . Here , we demonstrate that SV40 expresses a very late nonstructural viral protein that we have termed VP4 . VP4 is encoded within the VP2 and VP3 transcript from a downstream in-frame AUG start codon , and its expression triggers membrane lysis to aid in viral release for the efficient propagation of SV40 . In support of the hypothesis that VP4 has a role in the lytic process , we observed that VP4 was expressed concurrently with viral-induced lysis ∼24 h after the known late structural proteins , and that its deletion significantly reduced SV40 spreading by prolonging the viral life cycle . Furthermore , SV40-induced cell lysis could be recapitulated in E . coli by co-expressing VP4 and VP3 , but not by their individual expression , indicating that SV40-induced lysis likely occurs independently of eukaryotic host factors . Altogether , this suggests that VP4 expression initiates the timely execution of the lytic process through a mechanism that relies on its association with VP3 . There are several reasons that explain why VP4 has gone undetected . First , the small molecular mass of VP4 causes it to migrate off the bottom of most gel separation systems used to visualize the larger structural proteins . Second , VP4 is not incorporated into the viral particle , but is predominantly found in the frequently discarded nuclear fraction . Next , VP4 is expressed very late , immediately before cell lysis . Finally , VP4 is found in lysed cells and is susceptible to proteolysis by trypsin ( Figure S2 ) , which is a method commonly used to harvest adherent cells . These properties explain why the identification of VP4 has been elusive and provide valuable insight into its function . It is unlikely that the observed defect in ΔVP4 propagation was caused by the mutations created in VP2 ( Met228 to Ile ) and VP3 ( Met109 to Ile ) . The ΔVP2/3 genome , which expresses VP4 and not VP2 and VP3 , was able to complement the ΔVP4 genome that contains VP2 and VP3 with the aforementioned Met to Ile mutations . The observed 2-fold rescue in viral propagation following the co-transfection of ΔVP2/3 with ΔVP4 was considered significant due to the fact that ΔVP2/3 has no cell to cell spreading capabilities by itself [7 , 8] . This further supports the conclusion that VP4 was required for the efficient propagation of SV40 . Cytolytic viruses such as polyomaviruses ( SV40 , JCV , BKV ) and picornaviruses ( poliovirus , rhinovirus , and coxsackievirus ) induce host cell lysis to release their progeny and prevent their encapsulation in cellular membranes . However , these viruses must regulate the induction of lysis to ensure that it only occurs after the viral assembly stage . SV40 is known to utilize a series of tightly controlled timing mechanisms to direct its replication . At the onset of infection , the early genes trigger viral genome replication before the synthesis of the structural proteins , which assemble around the genome . During assembly , VP1 acts as a second timer . The expression of VP1 prior to VP2 and VP3 ( Figure 6 , step A ) [8] pre-positions the VP1 pentamers to bind to the newly synthesized VP2 and VP3 , preventing their insertion into membranes ( steps B and C ) . These VP1–VP2 and VP1–VP3 complexes are then imported into the nucleus for virion assembly , and the amount of VP2 and VP3 starts to exceed the number of available VP1 pentamers ( steps D and E ) . At this stage , VP4 begins to accumulate , forming hetero-oligomers with VP3 , and possibly VP2 , which insert into the host cell membranes ( Figure S5 ) to initiate cell lysis and viral release ( Figure 6 , steps F and G ) . This model demonstrates how the later expression of VP4 triggers cell lysis . Viruses are known to evade immune detection and prolong the survival of the infected cell by inhibiting the apoptotic response [19] . However , once the replication process has completed , cytolytic viruses must override both the viral anti-apoptotic and the host apoptotic responses in order to induce the lytic death of the host cell . In SV40-infected cells , LT acts to prevent the apoptotic response by sequestering p53 [20 , 21] . To counteract the protective activity of LT and induce lysis at the appropriate time , SV40 expresses VP4 very late in assembly . Thus lysis , and hence the length of the SV40 life cycle , is regulated by the temporal expression of VP4 , explaining why its removal significantly delayed the lytic death of infected cells . Upon sequence analysis of the 12 known polyomaviruses from mammalian and avian species , 11 were found to possess a potential VP4 initiation codon at a similar position in their VP2 transcripts . This suggests that VP4 performs a conserved function , but a detailed investigation to determine whether VP4 regulates lysis in other polyomaviruses is necessary . For the lone exception ( goose hemorrhagic polyomavirus ) , the absence of a VP4 initiation codon could explain its lengthy life cycle ( 7 d until the first observed cytopathic effects in primary goose kidney cells ) and the inability to effectively produce the virus in culture [22] . A question of future concern now becomes , how is the later expression profile for VP4 controlled ? The requirement of VP4 to be co-expressed with VP3 in order to recapitulate cell lysis in bacteria indicated that the association of VP4 with VP3 dramatically alters the properties of these two proteins . VP3 is a viral structural protein that is thought to be involved in the penetration process [8] . It is expressed late , binds to VP1 pentamers , forms oligomers , and is able to post-translationally insert into ER membranes . In the case of VP4 , the structure appears to be altered in such a way that it hides the VP1-binding domain , explaining why VP4 and VP4-like constructs with small tags that contain the majority of VP4 are not able to bind VP1 [23] . In previous studies where VP1 binding to VP4-like and VP4-truncated constructs containing N-terminal GST-tags was observed , the larger GST-tags may have prevented this change in the conformation of VP4 [15 , 24] . Comparative analysis of VP2 , VP3 , and VP4 revealed that these proteins possess distinct hydrophobic properties that likely contribute to their functional and conformational variations ( Figure S6 and [8] ) . The evolutionary pressure placed on viruses to perform multiple tasks with small genomes has caused viruses to diversify their genomes by encoding a number of proteins within single transcripts . They accomplish this task by organizing the genes as large polyproteins or in a polycistronic manner in which different genes are contained within alternative reading frames or at downstream AUG initiation sites in a single reading frame . Within SV40 , VP3 and VP4 are expressed from successive downstream AUG codons in the same reading frame as VP2 . This strategy for the creation of N-terminally truncated proteins could also be useful for modifying the role of cellular proteins by removing regulatory or functional binding domains , or by altering their localization by deleting targeting sequences . Through these simple mechanisms , cells and their pathogens can significantly expand and diversify the protein population expressed from their genomes . The bacterial replication competent plasmid pSV40 , encoding the entire WT SV40 genome ( strain 776 ) , was obtained from H . Kasamatsu ( Los Angeles , California , United States ) . All of the SV40 mutant genomes were created by site-directed mutagenesis ( Stratagene , http://www . stratagene . com/ ) of the ATG codons corresponding to the indicated VP2 Met residues in pSV40 to ATA ( Ile ) codons . The ΔVP4 and VP2-M295I genomes had Met228 ( ΔVP4 ) or Met295 ( VP2-M295I ) within the VP2 reading frame altered to ATA ( Ile ) . The ΔVP2/3 genome had the ATG initiation codons for VP2 ( Met1 ) and VP3 ( Met119 ) in VP2 mutated to ATA ( Ile ) . The ΔVP2/3/4 genome contained the additional mutation of the ATG initiation codon for VP4 ( Met 228 ) altered to ATA ( Ile ) . Infectious genome production and SV40 particle isolation from African green monkey kidney cells ( BS-C-1 ) ( ATCC , http://www . atcc . org/ ) have been previously reported [8] . The VP1 and VP2/3 polyclonal antibodies were generous gifts from A . Oppenheim ( Jerusalem , Israel ) . Antibodies against LT were purchased from Calbiochem ( http://www . emdbiosciences . com/html/CBC/home . html ) , Alexa-488-conjugated goat anti-mouse from Molecular Probes ( http://probes . invitrogen . com/ ) , and HRP-conjugated goat anti-mouse and anti-rabbit from Amersham ( http://www . gelifesciences . com/ ) . The cells were lysed in the dish on ice with 1% NP-40 HBS , 2 . 4 mM NEM , 50 μM LLnL , 0 . 4 μM PMSF , and 20 μM leupeptin and collected by scraping . The NP-40 insoluble nuclear fraction was sedimented at 15 , 000g for 5 min and the post-nuclear supernatant was collected . The nuclear fraction was solubilized in 2% SDS , 50 mM Tris ( pH 7 . 5 ) , 1 mM EDTA , 150 mM NaCl , 50 μM LLnL , 0 . 4 μM PMSF , and 20 μM leupeptin . The lysates were resolved by SDS-PAGE , transferred to PVDF , and subjected to standard immunoblotting . Trypan blue analysis of SV40-infected cells was previously reported [7] . BS-C-1 cells were maintained in DMEM supplemented with penicillin-streptomycin and 5% FBS ( Invitrogen , http://www . invitrogen . com/ ) in a humidified 5% CO2 incubator at 37 °C . Glass coverslips containing cells were fixed , stained with LT antisera , and analyzed by immunofluorescence microscopy as previously described [7] . VP1 , VP2 , VP3 , and VP4 were PCR cloned into the pSP72 plasmid ( Promega , http://www . promega . com/ ) for in vitro translations and the pET21d vector ( Novagen , http://www . emdbiosciences . com/html/NVG/home . html ) for bacterial expression using standard techniques . Site-directed mutagenesis of Met228 within the VP2 reading frame to an Ile was used to create the VP3-M228I and VP3-M228I-His constructs . The expression and purification of VP1Δarm-His using Ni-NTA sepharose and GST-VP3 using GSH-sepharose from the BL21 E . coli Rosetta strain have been previously described [8] . All pSP72 plasmids containing VP1 , VP2 , VP3 , and VP4 were linearized with the restriction enzyme Nde I , and the pET21d plasmids were linearized with AlwN I . The cDNAs were transcribed with the T7 expression system from Ambion ( http://www . ambion . com/ ) and translated in 10-μl reactions as described previously [25] with 6 mM β-mercaptoethanol used instead of DTT for experiments involving bacterially produced VP1Δarm . For co-translational association with VP1Δarm , the protein was present during synthesis , while post-translational associations involved the addition of the protein after synthesis inhibition . To monitor oligomerization with GST-VP3 , the translation reactions were added post-translation to freshly purified GST-VP3 and GST bound to GSH-sepharose in 0 . 5 ml of PBS ( pH 7 . 3 ) , 0 . 5% TX-100 , and 5 mM DTT . The methods for isolating the complexes of VP1Δarm-His with VP2 , VP3 , and VP4 by Ni-NTA sepharose have been previously described along with those for isolating bacterial purified GST–VP3 complexes by GSH-sepharose [8] . The GenBank ( http://www . ncbi . nlm . nih . gov/Genbank/ ) protein sequences from the various polyomaviruses used in this paper are the following: VP2: African green monkey ( NP_848005 ) ; baboon ( YP_406552 ) ; BKV ( BAF03118 ) ; bovine ( NP_040785 ) ; budgerigar ( NP_848011 ) ; crow ( YP_529825 ) ; finch ( ABB04270 ) ; goose hemorrhagic ( NP_849167 ) ; hamster ( NP_056734 ) ; JCV ( AAM89343 ) ; murine ( NP_041268 ) ; murine pneumotropic ( NP_041235 ) ; and SV40 ( AAC59345 ) . The VP4 nucleotide and protein sequence data reported are available in the Third Party Annotation Section of the DDBJ/EMBL/GenBank databases under the accession number BK006135 .	The release of viral particles from an infected host cell is essential for a virus to spread within the host organism . Cytolytic viruses such as the common cold , poliovirus , and simian virus 40 ( SV40 ) release their progeny by inducing lysis or death of the host cell . For efficient viral spreading , it is critical that optimal numbers of the virus are assembled before cell lysis and release occurs . Therefore , the timing of cell lysis is an integral and well-controlled step in the viral life cycle . For many years , lysis has been thought to be a nonspecific consequence of viral protein overexpression and the massive production of viral progeny . As SV40 was the first mammalian virus sequenced almost 30 years ago , it is an excellent model virus for investigating the poorly understood mechanism of viral release . In this study , we have identified a novel SV40 protein named VP4 that is required for the timely lytic death of the host cell , and hence regulates the spread of SV40 . The late expression of VP4 offers a sufficient period for virion assembly to occur before it initiates the lytic release of the newly assembled viral progeny .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods", "Supporting", "Information" ]	[ "nonenveloped", "viral", "release", "viroporin", "cell", "biology", "simian", "virus", "40", "virology", "membrane", "lysis", "cell", "lysis" ]	2007	A Very Late Viral Protein Triggers the Lytic Release of SV40
Experiments with drug-induced epilepsy in rat brains and epileptic human brain region reveal that focal cooling can suppress epileptic discharges without affecting the brain’s normal neurological function . Findings suggest a viable treatment for intractable epilepsy cases via an implantable cooling device . However , precise mechanisms by which cooling suppresses epileptic discharges are still not clearly understood . Cooling experiments in vitro presented evidence of reduction in neurotransmitter release from presynaptic terminals and loss of dendritic spines at post-synaptic terminals offering a possible synaptic mechanism . We show that termination of epileptic discharges is possible by introducing a homogeneous temperature factor in a neural mass model which attenuates the post-synaptic impulse responses of the neuronal populations . This result however may be expected since such attenuation leads to reduced post-synaptic potential and when the effect on inhibitory interneurons is less than on excitatory interneurons , frequency of firing of pyramidal cells is consequently reduced . While this is observed in cooling experiments in vitro , experiments in vivo exhibit persistent discharges during cooling but suppressed in magnitude . This leads us to conjecture that reduction in the frequency of discharges may be compensated through intrinsic excitability mechanisms . Such compensatory mechanism is modelled using a reciprocal temperature factor in the firing response function in the neural mass model . We demonstrate that the complete model can reproduce attenuation of both magnitude and frequency of epileptic discharges during cooling . The compensatory mechanism suggests that cooling lowers the average and the variance of the distribution of threshold potential of firing across the population . Bifurcation study with respect to the temperature parameters of the model reveals how heterogeneous response of epileptic discharges to cooling ( termination or suppression only ) is exhibited . Possibility of differential temperature effects on post-synaptic potential generation of different populations is also explored . The World Health Organization identifies epilepsy as one of the most common neurological diseases affecting approximately 50 million people across all ages across the world [1] . According to the International League Against Epilepsy , a patient has epilepsy if he has had a seizure and his brain activity demonstrates a pathologic and enduring predisposition to have recurrent seizures [2] . Because of the risks involved with unanticipated seizures , treatment of the disease is required to improve long-term quality-of-life of the patients . Antiepileptic drugs such as anticonvulsants are usually given as first line treatment after being diagnosed with epilepsy . Pharmaceutical researches continually seek antiepileptic drugs that are more effective and have less side effects [3 , 4] . However , 20%-40% of patients diagnosed with epilepsy are found refractory to antiepileptic drug treatment [5 , 6] . Thus , alternative treatments are still being sought after [7 , 8] . Surgical treatment is done by performing a resection of the epileptic foci of the brain . Absolute remission however is not guaranteed , let alone possibilities of unintended outcomes since the method is largely invasive [9 , 10] . Although the success rate of surgical treatment is high , limitation in indication and cost significantly hinder intractable epilepsy patients in acquiring it . Another increasingly attractive treatment option involves electrical neurostimulation of specific neural region such as vagus nerve stimulation and deep brain stimulation [11 , 12] . In the previous decade , focal cooling of the epileptic brain area has been pursued as an alternative therapeutic treatment for epilepsy and other seizure-inducing brain injuries [13–15] . Studies in animals have shown that reversible cooling to a temperature as low as 15°C using an implantable cooling device is able to terminate epileptic discharges without affecting the normal brain tissue [16–19] . Earlier experiments even noted that focal cooling of the cortex for one hour above 0°C did not induce any irreversible histological change or motor dysfunction [20] . Focal cooling at 25°C was also demonstrated to suppress epileptic discharges in a human brain [21] . Epileptic seizures arising from post-traumatic brain injuries were also shown to be suppressed and can be further prevented by moderately cooling the brain down by a temperature reduction of 2°C [22] . In other studies , focal brain cooling has found potential use for treatment of other brain diseases such as ischaemia , stroke , and neonatal encephalopathy [23–26] . The ultimate goal especially for epilepsy studies is to develop a technique for epileptic seizure suppression by a temperature control , when detected , via an implantable cooling device as a solution for thermal neuromodulation . This is feasible if we have precise knowledge of how temperature can suppress or terminate seizures . While temperature effects on physiological properties of animal neurons have been well-studied in vitro [27–31] , mechanisms of how cooling suppresses epileptic discharges especially in vivo are still not clearly understood . In this study , we aim to identify prospective mechanisms and investigate them using a computational modelling approach . Neural mass models have been widely utilized to study brain activities [32–37] and gain relevant physiological insights from them . The model introduced by Wendling et al . [34] in particular was shown to produce different types of brain activities similar to intracranial EEG recordings . We explore prospective mechanisms of cooling on epileptic discharges by introducing temperature dependence in the neural mass model of Wendling et al . in light of findings observed in in vitro and in vivo experiments published in literature . In particular , changes in synaptic dynamics were reported from in vitro cooling experiments such as reduction in the efficacy of neurotransmitter vesicle release [38] , loss of dendritic spines [39] and reduced glutamate concentrations [40 , 41] , suggesting a possible synaptic mechanism . A recent study with patients with intractable epilepsy also reports reduced extracellular glutamate and GABA concentrations during focal brain cooling [42] . We then formulated temperature dependence in our chosen neural mass model by introducing a temperature factor in the post-synaptic impulse response function . Parameter estimation of the model is performed using EEG recordings from in vivo cooling experiments on an animal model of epilepsy . Although the model is able to reproduce termination of epileptic discharges reported in in vitro studies [43 , 44] , the results of modeling our experimental data ( in vivo ) reveal that this synaptic mechanism is not sufficient to explain epileptic discharges that are persistent during cooling although suppressed in magnitude . We propose that another mechanism is required to compensate the effect of this synaptic mechanism to be able to reproduce observed suppression of epileptic discharges during cooling in terms of reduction in both frequency and magnitude of discharges . We suggest some biological plausibility of this compensatory mechanism based from published results from cooling experiments . The temperature dependence is in the form of a temperature coefficient ( Q10 ) which represents the factor by which the rate of a process increases for every ten-degree rise in the temperature at which it takes place [45] . In this study , the Q10 values determine whether suppression or termination of epileptic discharges can be achieved . Such heterogeneous response of epileptic discharge activity to cooling is revealed by bifurcation patterns with respect to the temperature parameters of the model . All experiments were performed according to the Guidelines for Animal Experimentation of Yamaguchi University School of Medicine . The animals were anesthetized with urethane ( 1 . 25 g/kg , i . p . ) . Lidocaine , a local anesthetic , was applied at pressure points and around the area of surgery . Focal brain cooling experiments were performed at Yamaguchi University School of Medicine . In this study , we utilized their data for parameter estimation of our model . Details of the experiments can be found in [46] . Briefly , anaesthetized male Sprague-Dawley rats were induced with epilepsy using Penicillin G potassium . Continuous EEG recordings of the epilepsy-induced region of the brain were made before and during cooling . An Ag/AgCl electrode for recording EEGs ( Unique Medical Co . , Fukuoka , Japan ) was positioned stereotactically 2 mm below the cortical surface at the left sensorimotor cortex just beneath the cooling device . Five different rat experiments each were done at cooling temperatures 25°C , 20°C , and 15°C . To remove high frequency components and also match the represented frequencies in the model , the raw recordings underwent a 40-Hz low-pass filter using a fifth order Butterworth filter in Matlab . One-minute steady-state intervals before and during cooling were identified by an expert and were taken from the filtered data for the study . For the model estimation procedure , first , the data is further downsampled to 2kHz corresponding to a step size of 0 . 5 ms in the simulation . Next , both the downsampled data and the simulated EEG are normalized by dividing by their respective standard deviations of activity before cooling , thus , they are reported in arbitrary units ( au ) unless otherwise stated . Fig 1 shows a summary of the preprocessed data in which we concatenated one-minute steady state activities before and during cooling . Suppression of epileptic discharges during cooling was observed especially with 15°C cooling temperature ( Fig 1 ) . Epileptic discharges were suppressed in terms of magnitude ( lower magnitude during cooling ) in all cases . In most cases , frequency of epileptic discharges is lower during cooling although slightly higher in some cases . The average magnitude and frequency of epileptic discharges before and during cooling are summarized in Fig 2 with error bars indicating minimum and maximum values from five rats . In general , we can say that epileptic discharges are suppressed during focal cooling at all three cooling temperatures . Surprisingly , significant termination of epileptic discharges was observed only in two out of five rats with 15°C cooling temperature compared to most in vitro recordings reported in literature; epileptic discharges were generally persistent during cooling from these in vivo recordings . Different intracranial EEG activities such as spike-wave discharges and low-voltage high-frequency activity , have been widely explained using neural mass models—a class of models based on a mean-field approximation of the activity of a population of neurons . Neural mass models involve two major processes described by two functions: a firing response function and a post-synaptic impulse response function . The firing response function approximates the average firing rate of a population in response to an average input potential ( the average membrane potential of the population ) . Assuming a unimodal distribution of threshold potentials , the firing response function of a population of neurons can be described by a sigmoid function [47] given by S ( v ) = 2 e 0 1 + e ( v t h - v σ t h ) , ( 1 ) where vth is the average threshold potential at which the population fires at half the maximum firing rate e0 . The steepness of the sigmoid curve 1/σth is inversely related to the variability in thresholds of excitation of neurons in the population [47] . On the other hand , the average post-synaptic potential ( PSP ) input of a neuronal population to other populations to which it provides excitation or inhibition is given by the convolution of the post-synaptic impulse response function h ( t ) of the population and its average firing rate u ( t ) . Originally , the post-synaptic impulse response function is modelled using a sum of two exponentials [32] as compared from experimental data but was later simplified to h X ( t ) = G X g X t e - g X t ; t ≥ 0 , ( 2 ) where GX is the average post-synaptic gain and gX is the reciprocal of the average synaptic time constant of population X . Finally , the convolution vX ( t ) = hX ( t ) * u ( t ) is equivalent to the solution of the following second-order differential equation using Green’s Formula [48]: v X ′ ′ + 2 g X v X ′ + g 2 v = G X g X u . ( 3 ) The primary cell population also receives additional noisy input from subcortical afferents or other neural masses which makes the differential equation stochastic . Such can be solved numerically using stochastic methods such as Euler-Maruyama scheme . Finally , the average membrane potential of a population , which is the input to Eq ( 1 ) , is taken as the weighted summation of the average post-synaptic potentials of the afferent populations ( inhibitory populations have negative contribution ) . The weights are determined by the number of synaptic connections . The average membrane potential of the primary cell population is taken as representative of cortical EEG activity [32] . Different neural mass models vary in terms of the types of neurons that comprise a population and the interconnections among the populations ( feedback loops ) . Da Silva et al . [32] tried to explain alpha rhythm of brain activity by considering two populations: excitatory thalamocortical neurons as primary cell population and and a population of inhibitory interneurons . Jansen and Rit [33] extended this model using pyramidal cells as the primary excitatory neurons and two types of interneurons—excitatory and inhibitory . They also estimated the relations among the number of synaptic interconnections among the neuronal populations using animal records of cortical synapses found in literature . Wendling et al . [34] further differentiated slow and fast inhibitory interneurons based on the studies of [49 , 50] based from hippocampal connections . In their model , slow inhibitory interneurons project to the dendrites while fast inhibitory interneurons project to the soma or near the soma of pyramidal cells . Moreover , slow inhibitory interneurons provide inhibition to fast inhibitory interneurons . Although the model was patterned after neuronal connections in hippocampus , similar architecture has been seen in the neocortex ( see [51] for an extensive review ) . The block diagram of the model is shown in Fig 3 . The parameters of the model are summarized in Table 1 together with the standard values adopted in this study . Wendling et al . showed that their model is able to capture different brain activities observed in intracranial EEG recordings . By fixing the value of average excitatory synaptic gain , an activity map ( Figure 4 of [34] ) shows regions of different brain activities by varying the average synaptic gains of slow and fast inhibitory neuronal populations . They used their model to explain that fast epileptic activity can arise due to impaired GABAergic inhibition by slow inhibitory interneurons . They demonstrated this by estimating average synaptic gains in the model from intracranial EEG recordings of temporal lobe epilepsy ( Figure 5 and 6 of [34] ) . In this study , we used the same model and show that it strongly captures the discharge activity of the animal model of epilepsy used in the experiments . In this study , we try to explain how cooling works in suppressing epileptic discharges by introducing temperature dependence in the neural mass model of Wendling et al . particularly for epileptic discharges . Our formulation starts with reduction in concentration of neurotransmitters as reported in in vitro studies . We model this effect as an attenuation factor in the post-synaptic impulse response function particularly the average synaptic gain variable . Specifically , we assume a temperature dependence in terms of a Q10 factor as follows: h X ( t ) = Q 10 , s y n ( T - T 0 ) / 10 G X g X t e - g X t ; t ≥ 0 . ( 4 ) Here , T0 is the baseline temperature which is 31°C in the experiments . This temperature dependence attenuates the average synaptic gain and thus reduces the average PSP ( Fig 4 ) which makes up the average membrane potential of the population to which it provides excitation or inhibition . For excitatory and slow inhibitory interneurons , their average membrane potentials are solely contributed by the average PSP from pyramidal cell population , thus , are also attenuated and consequently yield reduced firing frequency . For the pyramidal cell population and fast inhibitory interneurons , negative inhibitory PSP contributes to their average membrane potential . If the weighted ( in terms of synaptic connections ) effect of temperature on inhibitory PSP is less than that on excitatory PSP , a net decrease in average membrane potential results . With the parameter values chosen in the model ( Table 1 ) , this is more likely the case . In Fig 5 , we can see that as Q10 , syn is increased from unity , frequency of discharges during cooling is decreased until termination . However , the value of Q10 , syn at which termination is nearly achieved ( Q10 , syn = 1 . 085 ) does not significantly attenuate PSP magnitude ( Fig 4 ) , consequently the magnitude of isolated discharges . In contrast , persistent discharges were observed during cooling in the experiments ( Fig 1 ) . These are suggestive that another mechanism is involved . To model persistent discharges during the cooling period , we conjecture that the reduction in the average frequency of firing caused by the first temperature dependence should be compensated . This can be achieved through the firing response function negating the effect of Q10 , syn ( see Discussion ) . A second temperature dependence is thus put forward involving a reciprocal Q10 factor multiplied to the average membrane potential: S ( v ) = 2 e 0 1 + e ( v t h - Q 10 , i n t - ( T - T 0 ) / 10 v σ t h ) . ( 5 ) Fig 4 illustrates the effect of this temperature dependence in the original firing response curve . The modified firing response curve is translated to the left and has steeper slope . In summary , two temperature parameters are introduced in this study—Q10 , syn and Q10 , int . The latter part of this study also looks at the possibility that Q10 , syn varies for different populations in their respective PSP generation . Since the cooling experiments were performed on five rats , model parameters were estimated per rat using three cooling temperatures . Modified from [52] , the objective function involved in the estimation is given by J ( θ ) = ∑ E I D I + E E f f M a g + P ( θ ) , ( 6 ) where Ex is the mean absolute percentage error ( MAPE ) of feature x computed as \|xmodel − xdata\| / \|xdata\| . The summation is over the three cooling experiments per rat . The features used in the estimation are the average inter-dischrage interval ( IDI ) over the one-minute series and the effective magnitude ( EffMag ) of epileptic discharges . IDI is computed as I D I = 1 N D ∑ i = 1 N D - 1 t i + 1 - t i , ( 7 ) where ti is a time at which a discharge ( exceeding three standard deviations of the activity ) occurs , and ND is the number of discharges within the one-minute activity . EffMag , on the other hand , is defined as E f f M a g = P 99 - P 1 , ( 8 ) where Pn denotes nth percentile of the activity . A penalty term P ( θ ) is also included in the objective for the estimation of the temperature parameters of the model from the epileptic discharge activity during cooling: P ( θ ) = K ( [ max { v D C } - max { v B C } ] + + [ min { v B C } - min { v D C } ] + ) , ( 9 ) where [⋅]+ = max{0 , ⋅} , {v} is the simulated discharge activity centered with respect to the baseline , and K is penalty strength set to 1000 . This term imposes the constraint that the range of discharge activity during cooling ( DC ) is contained within the range of the discharge activity before cooling ( BC ) , that is , epileptic discharges are indeed suppressed during cooling . Since the model is stochastic , ten different simulations were taken for each set of parameters from which the MAPE is computed against the experimental data . Finally , after we are able to narrow down the parameter space to optimize the objective function , a global search is employed [53] . We used Dividing Rectangle ( DiRect ) method [54] , a deterministic global optimization method that is less computationally expensive than stochastic evolutionary methods such as Genetic Algorithm which was used in [52] . Moreover , estimation was performed using a one-minute steady-state activity in contrast to dynamic estimation procedures such as Kalman Filtering [55] and Dynamic Causal Model [56] . A two-part estimation is performed for each experiment . The first part estimates the parameters of the Wendling model ( no temperature-dependent parameters ) that describes the activity of epileptic discharges before cooling . The second part estimates the temperature-dependent parameters ( Q10 factors ) during cooling using the result of the first part describing the pathological activity of the brain . Simultaneous estimation of all model parameters ( before and during cooling ) can be done , however , the two-part approach circumvents search issues in high-dimensional space . Furthermore , to address possible over-fitting , estimation of the Q10 values was done using the first 40 seconds of the one-minute activity during cooling . The next 20 seconds of the activity were used for validating the model estimates from which statistical tests are performed . It is generally accepted that epileptic activity results from changes in excitation-inhibition ratio . In the neural mass model , keeping the average excitation gain constant , excitation-to-inhibition ratio increases as GSIN or GFIN is decreased thereby simulating epileptic discharge activity . Exploration of the model shows that EEG recordings from the animal model of epilepsy used in the study is best explained by high average fast inhibitory gain GFIN and low average slow inhibitory gain GSIN ( Table 2 ) . This is consistent with previous findings that epileptic activity can arise when dendritic inhibition is impaired [34] . Fig 6 ( a ) shows a reproduction of epileptic discharge activities before cooling for two of the five rats . We observe that lower values of GSIN reproduce a discharge activity that is asymmetric with respect to baseline while higher values of GSIN reproduce a discharge activity that tends to be symmetric with respect to baseline . On the other hand , increasing both GSIN and GFIN reduces the frequency of epileptic discharges by effectively reducing the average membrane potential of the primary cell population which is basically the simulated EEG . The estimation of average slow inhibitory gain and fast inhibitory gain of Wendling et al . model was aimed to reproduce epileptic discharge activity recorded from the animal model of epilepsy used . Next , we estimate the parameters involved in the temperature dependence of the model from the activity during which focal cooling is applied in the epileptic brain area . To assess our temperature-dependent formulation , three models were estimated from the experimental data namely: SYN ( synaptic mechanism only: estimate Q10 , syn with Q10 , int = 1 . 0 ) , INT ( intrinsic mechanism only: estimate Q10 , int with Q10 , syn = 1 . 0 ) , and SYN_INT ( synaptic and intrinsic mechanisms: estimate Q10 , syn and Q10 , int ) . The results of the estimation were compared to no-temperature dependence ( NTD ) model ( Q10 , syn = 1 . 0 , Q10 , syn = 1 . 0 ) . As discussed earlier , SYN captures changes in the frequency of epileptic discharges but not their magnitude ( Fig 5 ) . On the other hand , INT , as expected , yields estimates that are almost unity ( like in the case of NTD ) since the model does not have anything to compensate for having Q10 , syn = 1 . 0 , i . e . no changes in average PSP yield no changes in the average firing rate . These suggest that temperature dependence in the post-synaptic impulse response function or firing response function alone does not capture the effect of cooling on the epileptic discharges ( Fig 7 ) . In fact , when both functions have temperature dependence as formulated ( SYN_INT ) , we see that suppression of epileptic discharges is reproduced . Fig 8 shows the boxplots of the mean absolute percentage error ( MAPE ) of the different models from fifteen cooling experiments . Note that the MAPE are computed from the last twenty seconds of the epileptic discharge activity during cooling which is apart from that used for the estimation ( see Materials and Methods ) . A Wilcoxon signed rank test shows that SYN_INT is significantly different from NTD model ( p = 0 . 0034 ) . It is also interesting to look at the estimated values of Q10 , syn and Q10 , int using SYN_INT model ( Table 3 ) . We can clearly see that Q10 , int is only slightly less than Q10 , syn . This is consistent in all estimations performed from experiments on five rats . We also performed estimation of Q10 factors from each cooling experiment per rat where we find cases in which Q10 , int is slightly greater than Q10 , syn . These cases correspond to experiments where there are slight increases in the frequency of epileptic discharges during cooling . However , in the results that we present here , Q10 factors are estimated from three cooling experiments per rat which yield Q10 , int values that are all slightly less than Q10 , syn . Fixing Q10 , syn at 1 . 8 , we vary Q10 , int from 1 . 0 to 2 . 0 at intervals of 0 . 01 and performed ten simulations of SYN_INT model with different random generator seeds . We find that the magnitude and frequency of simulated activity during cooling exhibit bifurcation behavior for different temperatures ( Fig 9 ) . There are three apparent bifurcation regions found for cooling temperatures 15°C and 20°C . From baseline activity , a bistable region occurs at around Q10 , int = 1 . 5 and vanishes at around Q10 , int = 1 . 66 going back to baseline activity until a sudden transition to discharge activity at around Q10 , int = 1 . 8 which is the same value at which Q10 , syn is fixed . The results of estimation from experiments lie around the third region where Q10 , int values are only slightly less than Q10 , syn values . This region corresponds to termination of epileptic discharges or suppression of epileptic discharges to a fixed magnitude . The bistable region , on the other hand , correspond to two possible activities depending on initial condition of the simulation- a baseline activity and an activity characterized by low-amplitude high frequency oscillations . This region , however , was not realized in the experiments . Hypothetically though , this suggests that seizure may occur with cooling when the compenstatory mechanism that involves the intrinsic excitability of neurons operates with Q10 , int values in this region . This bistable region vanishes at weaker cooling temperatures ( Fig 10 ) indicating that such possibility of seizure may be prevented . Similar pattern of bifurcation is also observed with a bistability region that is wider at higher values of Q10 , syn and vanishes at lower values of Q10 , syn ( Fig 10 ) . To gain more insight about the bifurcation behavior observed in the model , we performed a numerical continuation of the deterministic version of the model ( standard deviation of input is zero ) using MatCont [57] . Similarly , we fixed Q10 , syn at 1 . 8 . Continuing from a fixed point with Q10 , int = 1 . 0 , two saddle node bifurcations are found at around Q10 , int = 1 . 7996 and Q10 , int = 1 . 1702 ( Fig 11 ( a ) ) . From the second bifurcation point , a Hopf bifurcation is found at around Q10 , int = 1 . 566175 with negative first Lyapunov coefficient . This implies that a stable fixed point transitions into a stable limit cycle . These bifurcation points explain the observed bistable region in the original stochastic model above where low-amplitude high-frequency oscillations or a baseline activity can be observed depending on the initial state of the system . ( Note that stationary state in the noiseless model corresponds to baseline activity in the stochastic model . ) Furthermore , continuing from the Hopf bifurcation point , a limit point of cycles ( LPC ) is found at around Q10 , int = 1 . 68 . A LPC is a saddle node bifurcation for periodic orbits where two limit cycles coalesce and annihilate each other . This explains the recovery of stationary state until the first bifurcation point at which the system exits the bistable region and goes back to stable periodic orbits ( discharge activity ) . The transition point observed in the stochastic model ( termination to suppression of discharge activity ) is then a sudden jump from baseline activity resulting in a magnitude of suppressed discharge activity that is proportional to the width of the hysteresis loop for a particular temperature and does not gradually increase from the magnitude of a baseline activity . At weaker cooling temperatures , such bifurcation is not observed at least in the physiologically explicable region of Q10 values . We also explored the possibility that cooling has differential effect on PSP generation of different neuronal populations . We investigate this by assuming that Q10 , syn is not homogeneous for different populations with different average synaptic gains . ( Q10 , int is not differentiated across different subpopulations as we assumed that the temperature effect is the same across different populations in their intrinsic excitability mechanisms . ) SYN_INT assumes homogeneous effect of cooling across different populations . Two more models were estimated to account for the possibility of such differential effect of cooling . In EXC_INH , we assume differential effect of cooling on excitatory and inhibitory PSP generation involving production of glutamate and GABA respectively . In EXC_SIN_FIN , we further assume differential effect of cooling on slow and fast inhibitory PSP generation involving slow GABA and fast GABA respectively . Estimation of these two models were also found to yield significant difference from NTD ( p = 0 . 0034 and p = 0 . 0034 respectively ) . The two models however are not significantly different from SYN_INT ( p > 0 . 01 , Fig 8 ) . It is interesting to note that EXC_SIN_FIN is able to capture termination of epileptic discharges from rat 1 under cooling temperature of 15°C which is roughly captured using SYN_INT or EXC_INH . Estimated Q10 values in Table 3 present some general observations . In EXC_INH model , Q10 , syn values are now slightly less than Q10 , int values except for rat 1 in which termination of epileptic discharges was observed . In EXC_SIN_FIN , higher Q10 , syn , FIN values were estimated especially with rats 3 and 4 . On the other hand , lower Q10 , syn , EX values are observed for rats 1 and 2 in which termination of epileptic discharges were found while lower Q10 , syn , SIN values for rats 3 and 4 in which epileptic discharges are only suppressed during cooling . These observations suggest that termination or suppres sion of epileptic discharges can result from different synaptic responses of different neuronal populations to cooling . Figs 12 and 13 show how the different models reproduce termination or suppression of epileptic discharges in rats 4 and 1 , respectively . Finally , it can also be observed that the estimated Q10 values are between 1 . 7 and 2 . 0 except those estimated from rat 5 in which case the estimated values are less than 1 . 2 . The estimation result from rat 5 can be substantiated by observing the activities during cooling of rat 5 at different temperatures showing less evidence of suppression of epileptic discharges ( Fig 1 ) . Our study confirms the ability of Wendling et al . model to capture different brain activities particularly epileptic discharge activity induced in the animal model of epilepsy used . After a brute-force search in the GSIN and GFIN space ( with GPY = GEX = 5 . 0 ) , we find that the epileptic discharge activities from our animal model of epilepsy are best estimated in the range [24 . 0 , 31 . 0] mV for GSIN and [80 . 0 , 110 . 0] mV for GFIN , the latter of which is not explored in the original model . Alternatively , we can keep GFIN in physiological range [40 . 0 , 60 . 0] mV but would entail that the number of synaptic connections from fast interneurons to pyramidal cells is twice than the standard value or that the maximum average firing rate of fast inhibitory interneurons is twice than that of the others ( see Eq 3 ) . This is still consistent with the findings of Wendling et al . [34] suggesting that impaired dendritic inhibition alters excitation-inhibition balance giving rise to rhythmic discharge activity capturing the effect of Penicillin G potassium in cortical tissues inhibiting GABA receptors [58] . Nevertheless , the estimated parameters indicate that our animal model of epilepsy can be best explained by much lower dendritic inhibition and much higher perisomatic inhibition compared to the standard range of values reported . High GSIN values in fact supports [59] which reported high somatic inhibition together with impaired dendritic inhibition in experimental epilepsy . Meanwhile , asymmetric epileptic discharge activities with respect to baseline activity as seen from experiments with rat 3 can be reproduced with lower value of GSIN ( 25 . 012 mV ) and higher value of GSIN ( 101 . 44 mV ) . On the other hand , symmetric discharge activity with respect to baseline is observed when dendritic inhibition is increased . Fig 6 ( b ) illustrates that this symmtery ( asymmetry ) of the discharge activity ( which is the summation of the PSP from excitatory and inhibitory interneuronns ) is largely due to the PSP response of excitatory interneurons showing faster ( slower ) repolarization while the PSP responses of the inhibitory interneurons do not show significant changes . Neurotransmitters play a central role in the generation of PSP [60] . They are released in response to Ca2+ influx after depolarization of pre-synaptic terminal and bind to their receptor molecules at the post-synaptic membrane opening or closing ion channels thereby generate excitatory or inhibitory PSP . It has long before suggested that neurotransmitter release has temperature dependence which causes changes in PSP generation [61] . This was confirmed by experimental observations of reduced efficacy of neurotransmitter vesicle release and reduced extracellular glutamate concentration during cooling [38 , 40] that imply lower neurotransmitter concentration at the synapses to bind at the post-synaptic receptor and generate PSP . In light of this , it was straightforward to assume a temperature dependence on the post-synaptic impulse response function in a neural mass model . Similar to temperature-dependent formulation of Hodgkin-Huxley type neurons [62 , 63] , temperature dependence in Wendling et al . model is modelled using a temperature coefficient given by a Q10 factor . This factor accounts for mean-field effect of temperature to several processes occurring during PSP generation across the neuronal population . For example , diffusion of neurotransmitters , Ca2+ , and receptor proteins [64–66] are slowed down at different rates at decreasing temperatures affecting efficacy of neurotransmitter vesicle release and the binding of neurotransmitters at the post-synaptic terminal receptors which regulate the activities of specific ion channels . Simply , Q10 , syn is added to the post-synaptic impulse response function and can be interpreted as direct attenuation of the average post-synaptic gain of the population or synaptic conductance of one neuron . This yields lower average PSP values when temperature is decreased from a baseline temperature . This decreases or increases the average membrane potential of the populations to which the population provides excitation or inhibition respectively . Reduced average membrane potential yields lower frequency of firing . In fact , we saw that termination of epileptic discharges results when the firing frequency approaches zero with Q10 , syn ≈ 1 . 085 with nonsignificant decrease in the magnitude of isolated discharges . In contrast , what was actually observed from experiments is that epileptic discharges are persistent during cooling but suppressed in magnitude . This is not reproduced by the model because of the nonlinearity of the firing response function . A Q10 value of 1 . 085 does not significantly suppress the magnitude of discharges but its effect on attenuated PSP responses significantly lowers the firing rate of the receiving population . Interestingly , in some cases in the experiment , slight increases in frequency of epileptic discharges were observed ( Fig 1 ) . These lead us to assume that a concomitant mechanism plays a role during cooling which may involve the intrinsic excitability mechanism of neurons compensating for the effect of reduced PSP on the average firing activity of the populations . Thus , a reciprocal Q10 factor was formulated as put forward in ( Eq 10 ) . Similarly , the Q10 factor involved here accounts for mean-field effect of temperature to several processes occurring during action potential generation such as diffusion of ions and ion channel gating across the neuronal population . Fig 4 shows how the average firing rate is compensated by the second temperature dependence . The firing frequency of a positive average membrane potential in the original firing response curve corresponds to an increased firing frequency at the same value of average membrane potential in the temperature dependent curve . The effect is opposite for negative average membrane potentials and rather minimal . A direct physiological interpretation of this mechanism can be examined if we write the equation in its equivalent form S ( v ) = 2 e 0 1 + e ( Q 10 , i n t ( T - T 0 ) / 10 v t h - v Q 10 , i n t ( T - T 0 ) / 10 σ t h ) , ( 10 ) where the Q10 factors are now with the parameters vth and σth . Recall that vth is the average threshold of firing of neurons and σth is the variability in the thresholds of excitation of neurons . This then implies that as a compensatory mechanism , cooling lowers both the average and variance of the distribution of the firing thresholds of neurons in the population . Hence , even if the average PSP is reduced resulting in lower average membrane potential , epileptic discharges can still be persistent since lower average threshold of firing allows for subthreshold activity before cooling to become suprathreshold during cooling . This can be seen as a form of homeostasis in the firing activity of the neuronal population involving both synaptic and intrinsic excitability mechanisms . Surprisingly , the combined mechanisms result in suppression of epileptic discharges in terms of magnitude which is not captured if we assume temperature dependence in the post-synaptic impulse response function alone . This is because higher Q10 , syn values now significantly reduce PSP responses ( Fig 11 ( b ) ) but the effect of which is compensated by the reciprocal of Q10 , int . Slight increase in frequency of discharges observed in some of the experiments can be realized if Q10 , int is made slightly greater than Q10 , syn . Further increasing Q10 , int proportionately increases the frequency of discharge activity ( Fig 9 ) . Reduced threshold potential of firing during cooling has been reported on an early experiment with squid axons [27] . Experiments with mammalian brains [28 , 67 , 68] reported that cooling depolarizes cell membrane potential and increases input resistance . In [68] , Volgushev et al . noted that cooling-induced depolarization of cell membrane occurs with an even higher gradient giving a marked decrease in the difference between the spiking threshold and the actual resting membrane potential . Thus , cooling brings the cells closer to spiking threshold , increasing excitability and decreasing variability in excitation levels across neuronal population . They proposed that such cooling-induced depolarization of the cell membrane may be attributed mainly by reduction of partial K+ conductance . Variability in threshold potential of firing has also been reported to increase with recent spiking activity [69] . We suppose that the opposite happens during cooling . As discussed earlier , cooling can decrease average firing rate of neurons which can imply less recent spiking activity . Henze and Buzsaki [69] suggested that prior action potentials cause Na+ channel inactivation that recovers with approximately a one-second time constant , increasing action potential threshold during this period . On the other hand , a study by Yu et al . [70] suggests that firing threshold variability can be explained by backpropagation of action potentials . Moreover , cooling was shown to strongly inhibit A-type K+ channels [71] in DRG neurons while these channels are reported to regulate action potential backpropagation in CA1 pyramidal neurons [72] . This might be in conflict with our finding that cooling reduces variability in firing thresholds since inhibited A-type K+ channels enhance backpropagating action potentials which in turn increases variability in firing thresholds . Then again , it is also possible that a net decrease in action potential backpropagation results as cooling can attenuate other critical factors such as density of Na+ at axon initial segment [73] and ion transport at nodes of Ranvier [74] . Estimation of the model from cooling experiments indicated that Q10 , int is only slightly less than Q10 , syn . This means that during cooling , the intrinsic excitability mechanisms of neurons just balance out the effect of temperature change on PSP generation . At first , it seemed that when Q10 , int ≈ Q10 , syn , discharge activity is suppressed but not terminated and when Q10 , int ⪇ Q10 , syn , discharge activity is terminated . To verify this generalization , we simulated the model for different values of Q10 , int fixing Q10 , syn = 1 . 8 . This led us to discover bifurcation patterns in the model which were confirmed using numerical continuation on the noiseless version of the model . First , we have verified that when Q10 , int ≈ Q10 , syn , discharge activity is suppressed but not terminated . At this point , the intrinsic mechanism “fully” compensates the effect of the synaptic mechanism resulting to a discharge activity that has approximately the same frequency but reduced in magnitude ( Fig 11 ( b ) ) . However , we found out that when Q10 , int ⪇ Q10 , syn , discharge activity is terminated only up to a certain value of Q10 , int and a seizure activity can arise with a wide range of intermediate Q10 , int values . As far as the authors are knowledgeable , there has been no report that seizure activity was ever observed in focal cooling of epileptic discharges . Moreover , Q10 , int values are not interpretable in terms of how intrinsic firing mechanisms can give rise to such values which would allow experiments to verify such finding . In theory , this should guide the design of implantable cooling devices which would necessitate a feedback control law to terminate cooling when a possible seizure can arise . Similar bifurcation patterns were observed for arbitrary values of Q10 , syn other than 1 . 8 . Our estimation results indicated Q10 values around 1 . 8 which was , surprisingly , also reported in previous studies involving voltage-gated Na+ channel ( VGNC ) dynamics [13] . Then again , in vitro studies [68 , 75] suggest that involvement of VGNC might be ruled out as abortion of epileptiform discharges were seen to be associated with a depolarization block . Perfect depolarization is against changes in the gating property as initially hypothesized , i . e . , cooling is not inducing a liquid phase transition in phospholipid bilayer of the membrane thereby distorting the channel’s property , rather through other mechanisms . Another interesting study by Motamedi et al . [75] with an in vitro epilepsy model showed that cooling has differential effect on the firing rates of pyramidal cells and interneurons . This actually motivated the models where we included more temperature dependent parameters to investigate possible differential effect of cooling on PSP generation . This relies on the assumption that cooling may have differential effect on different neurotransmitters responsible for generating PSP . However , in this study , the model parameters were estimated from in vivo EEG recordings which have clear departures from the aforementioned in vitro study . We can speculate though that it may be possible to reproduce such differential effect of cooling on the activity of pyramidal cells and inhibitory interneurons if we had isolated EEG recordings from pyramidal cell population and interneuronal population activities and from which we could estimate the model parameters with an appropriate modification of the objective function ( Eq 9 ) . Nevertheless , when the effect of cooling on inhibitory interneurons is much less than on excitatory interneurons , reduced average membrane potential of pyramidal cell population results and consequently , reduction in the average firing frequency of the population is observed as reported in the study . The results presented in this paper only considered the steady-state effect and does not include transient dynamics of cooling on epileptic discharges although some experiments have noted the effect of rate of cooling on termination of epileptic discharges . For instance , an in vitro study [44] reported that during slow cooling , epileptic discharges persist with decreasing amplitude until termination is achieved with further temperature drop . In contrast , rapid cooling achieves immediate termination of the discharges . The gradual decrease in amplitude of epileptic discharges during slow cooling can be captured by the model using an appropriate model for temperature dynamics ( e . g . Newton’s Law of Cooling ) . In its present form , immediate termination of discharges by rapid cooling can be explained by our model as a case where Q10 , int ⪇ Q10 , syn , i . e . reduction in average and variance of firing thresholds across neuronal population is not able to compensate reduction in discharge frequency due to reduced average membrane potential resulting from attenuation of post-synaptic activity . Alternatively , such transient effect may be modeled by a Q10 that decays from a non-steady state value to a steady state value proportional to the rate of cooling . In most in vitro studies that we reviewed , steady-state termination of epileptic discharges was achieved using either slow or rapid cooling down to a constant temperature . In contrast , termination may not be always possible in in vivo setting . We surmise that the compensatory mechanism put forward by the model is more concomitant in in vivo than in in vitro environment . Recent studies on epilepsy and epilepsy models have involved the role of non-neuronal cells such as astrocytes and microglia in mechanisms of seizure development such as reactive astrogliosis , glial-mediated inflammation , and Ca2+ signalling dysfunction [76 , 77] . It may also be possible that cooling can attenuate activation of both neuronal and non-neuronal cells that will consequently impair their involvement in one or several hyperexcitability mechanisms . While there have been recent attempts at modelling the interaction of neuronal and non-neuronal cells [78 , 79] , formulation of temperature dependence on the models may involve multimodal recordings other than EEG ( extracellular GABA and glutamate concentrations , cerebral blood flow ) in focal brain cooling experiments to estimate the model parameters . This is an interesting direction which we hope to pursue in the future .	Focal cooling of the epileptic brain region has been shown to consistently suppress epileptic activity and it is hoped that this treatment can be developed in the future into an implantable cooling device . However , it is still not clearly understood how cooling suppresses epileptic activity . This study uses a computational approach to identify and investigate possible mechanisms . First , we used a neural mass model to reproduce epileptic discharge activity . Next , we simulate the effect of cooling by introducing temperature dependence in the model . Based from evidences reported from in vitro and in vivo studies , we formulated two temperature-dependent mechanisms that can reproduce the effect of cooling on the epileptic discharge activity . Parameter estimation of the model was performed using EEG recordings of focal brain cooling experiments with rats in vivo . Our model involves a synaptic mechanism that results in a reduced frequency of discharges and an intrinsic excitability mechanism that compensates such reduction in frequency of discharges resulting in persistent discharges during cooling but suppressed in magnitude . The temperature dependence is in the form of Q10 temperature coefficients which determine whether suppression or termination of epileptic discharges can be achieved .	[ "Abstract", "Introduction", "Materials", "and", "methods", "Results", "Discussion" ]	[ "medicine", "and", "health", "sciences", "ecology", "and", "environmental", "sciences", "action", "potentials", "membrane", "potential", "brain", "electrophysiology", "electrophysiology", "neuroscience", "clinical", "medicine", "ganglion", "cells", "brain", "mapping", "interneurons", "bioassays", "and", "physiological", "analysis", "electroencephalography", "neuroimaging", "research", "and", "analysis", "methods", "ecological", "metrics", "epilepsy", "imaging", "techniques", "animal", "cells", "clinical", "neurophysiology", "electrophysiological", "techniques", "cellular", "neuroscience", "cell", "biology", "ecology", "pyramidal", "cells", "neurology", "physiology", "neurons", "q10", "temperature", "coefficient", "biology", "and", "life", "sciences", "cellular", "types", "neurophysiology" ]	2017	Differential temperature sensitivity of synaptic and firing processes in a neural mass model of epileptic discharges explains heterogeneous response of experimental epilepsy to focal brain cooling
Chlamydia trachomatis is an important human pathogen that replicates inside the infected host cell in a unique vacuole , the inclusion . The formation of this intracellular bacterial niche is essential for productive Chlamydia infections . Despite its importance for Chlamydia biology , a holistic view on the protein composition of the inclusion , including its membrane , is currently missing . Here we describe the host cell-derived proteome of isolated C . trachomatis inclusions by quantitative proteomics . Computational analysis indicated that the inclusion is a complex intracellular trafficking platform that interacts with host cells’ antero- and retrograde trafficking pathways . Furthermore , the inclusion is highly enriched for sorting nexins of the SNX-BAR retromer , a complex essential for retrograde trafficking . Functional studies showed that in particular , SNX5 controls the C . trachomatis infection and that retrograde trafficking is essential for infectious progeny formation . In summary , these findings suggest that C . trachomatis hijacks retrograde pathways for effective infection . With 100 million new infections per year , Chlamydia trachomatis is the most frequently sexually transmitted bacterial pathogen world-wide [1] . C . trachomatis replicates inside a membrane-bound vacuole , the inclusion , and has a unique cycle of development , alternating between two distinct bacterial forms . The elementary body ( EB ) is spore-like , infectious but non-dividing . In contrast , the reticulate body ( RB ) is non-infectious but replicative . After internalization of the EB , the bacteria are found inside the inclusion , which is segregated from the lysosomal degradation pathway . EBs then differentiate into RBs , which replicate inside the growing inclusion . At mid-infection time points the inclusion is packed with replicating RBs that start to re-differentiate into EBs [2] . The surrounding inclusion membrane is the interface between the bacteria and the host cell . This membrane is actively modified by insertion of bacterial proteins and is not permissive for diffusion of molecules of 520 Da and larger [3] . It contains classical bacterial inclusion proteins of the Inc-protein family as well as non-classical Inc proteins [4] . Furthermore , a growing number of cellular proteins have been described to associate with the Chlamydia inclusion , but a global picture of proteins contributing to the inclusion is currently missing . Membranes compartmentalize the eukaryotic cell into different organelles , including those of the secretory pathway and the endo-lysosomal system . In the secretory pathway , cargo is modified to address it to and then to transport it to its designated destination . The endo-lysosomal system functions in internalization of molecules from the plasma membrane ( PM ) or the extracellular space , followed by sorting of these molecules either for degradation in the lysosomes or for retrograde transport to different organelles , including the Golgi apparatus ( GA ) . The human retromer is a multi-protein complex essential for recycling of cargo receptors into the tubular endosomal network and transports them to the trans-Golgi network ( TGN ) [5] . In human cells , the retromer consists of a membrane-deforming and a cargo recognition subcomplex , which are composed of the sorting nexins ( SNX ) 1 , 2 , 5 , 6 and the vacuolar protein sorting-associated proteins ( VPS ) 26 , 29 , 35 , respectively [6] . On endosomes , SNX dimers bind to phosphatidylinositol phosphates ( PIPs ) via their phox homology ( PX ) -domains . Additionally , these SNXs contain a Bin-Amphiphysin-Rvs ( BAR ) domain that recognizes membranes with high curvature and induces membrane tubulation , which is thought to support sorting of retrograde receptors out of the endo-lysosomal pathway [7] . Interaction with the cargo recognition subcomplex eventually leads to vesicle formation and the enclosed cargo is transported along microtubules to the TGN [8 , 9] . Proteomic studies of phagosomes isolated using latex-beads have greatly increased our knowledge about the biogenesis and function of these organelles [10–12] . Furthermore , the biochemical purification of vacuoles containing Salmonella enterica , Mycobacterium avium , Rhodococcus equi and Legionella pneumophila also fostered our understanding of the host cell protein composition of these unique intracellular compartments [13–17] . Here , we describe a two-step protocol for the isolation of high purity C . trachomatis serovar L2 inclusions at mid-cycle . Using LC-MS/MS based proteomics combined with ss isotope labeling by amino acids in cell culture ( SILAC ) , we identified 351 host cell proteins that are significantly enriched in the proteome of isolated inclusions , representing the host cell-derived Chlamydia inclusion proteome . Enrichment analysis of this data showed that the C . trachomatis inclusion is a complex intracellular compartment that interacts with components of the retromer . Confocal studies confirmed the recruitment of SNX1 , 2 , 5 and 6 to the inclusion and further suggested that the retromer subcomplexes are at least partially separated at the inclusion membrane . Functional analyses of the retromer by RNA interference and by treatment with Retro-2 , an inhibitor of retrograde transport of toxins and viruses , revealed that knockdown of SNX5 resulted in an increase in infectious progeny whereas Retro-2 treatment inhibited the formation of infectious bacteria . Taken together , these results show a previously unknown association of SNXs with C . trachomatis inclusions and provide evidence for a new role of SNXs during bacterial infections that appears to be independent of the classical SNX-BAR retromer complex . We established an isolation method for C . trachomatis inclusions at mid-infection time points , based on a two-step protocol originally described for the isolation of Legionella-containing vacuoles from amoebae ( Fig 1A ) [16] . Infected HeLa cells were lysed and the obtained cell lysate containing inclusions was separated on a self-forming Percoll gradient . Gradient fractions were taken and analyzed for presence of bacterial and cellular proteins by immunoblotting and for presence of intact inclusions by phase contrast microscopy ( S1A and S1B Fig ) . The high density fractions harboring intact inclusions ( S1A and S1B Fig ) were collected , pooled and further purified by magnet assisted cell sorting ( MACS ) using an antibody specific for IncA , a bacterial transmembrane protein located in the inclusion membrane [18] . Presence and numbers of inclusions were monitored by phase contrast microscopy ( Fig 1A and 1B ) . Counting of visually intact inclusions at each purification step showed that ~50% of C . trachomatis inclusions present in the cell lysate could be isolated ( Fig 1B ) . The purity of the different fractions was assessed by immunoblotting , using antibodies specific for marker proteins of different cellular compartments and for chlamydial proteins ( Fig 1C ) . Lysate of infected and uninfected HeLa cells showed presence of organelles such as the nucleus , endoplasmic reticulum ( ER ) , lysosomes , mitochondria , cytosol and the PM ( Fig 1C ) . After separation by Percoll gradient , inclusions were enriched as indicated by an increase in IncA and Hsp60 signals , accompanied by a decrease in signals for cellular compartments . MACS purification resulted in a fraction that contained chlamydial inclusions that were nearly completely devoid of cellular contaminants as monitored by immunoblotting ( Fig 1C ) . Obtained inclusion fractions were then analyzed by electron and fluorescence microcopy ( Fig 1D and 1E ) . Transmission electron microscopy ( TEM ) demonstrated the presence of inclusions that contained both bacterial forms surrounded by the inclusion membrane ( Fig 1D ) . To validate the presence of cellular proteins in the isolated inclusion fraction , inclusions were purified from cells expressing a Rab11A-eGFP fusion protein that is known to be associated with C . trachomatis inclusions [19] . Immunofluorescence ( IF ) staining and confocal microscopy of isolated inclusions revealed that Rab11A-eGFP signal co-localized with IncA in a rim-like pattern ( Fig 1E ) . In summary , these data show that we are able to isolate C . trachomatis inclusions at mid-infection time points . To identify host cell proteins specifically associated with isolated C . trachomatis inclusions , SILAC was applied [20] . Using this method , we were able to control for non-specific , co-purifying proteins during the isolation procedure ( Fig 2A ) . The proteins that are bona fide constituents of the inclusion were expected to have a high ratio of L label vs . H label ( SILAC ratio ) of one peptide species , whereas contaminants were expected to have SILAC ratios close to 1 in the inclusion fraction ( Fig 2A ) . The abundance of inclusion-associated proteins in enriched fractions and proteins in total cell lysates was calculated using iBAQ ( intensity based absolute quantification ) which estimates the abundance of proteins based on the sum of peak intensities of all peptides matching to a specific protein , divided by the number of theoretically observable peptides [21] . Despite limited accuracy , this method provides additional information especially for highly abundant proteins in addition to the SILAC based exclusion approach . Based on this method , we quantified the relative contribution of each protein to the total proteome of the lysate and the inclusion using sum total normalization for the proteins in each fraction . Only proteins that passed the SILAC exclusion approach were considered for the inclusion proteome . The quotient of the values for the inclusion and the lysate resulted in the enrichment score for proteins which were overlapping in the two datasets ( iBAQ enrichment score ) ( Fig 2B and S1 Text ) . For proteins that were not found in our lysate proteome , we used a recently published very high coverage dataset of the HeLa proteome [22] for approximation of the protein abundance in the cell lysate . We performed experiments in three biological replicates . Analysis of the raw data by MaxQuant resulted in the identification of 1400 host cell proteins in the inclusion fraction ( Fig 2C ) and 2002 host cell proteins in the cell lysate . To characterize potential organellar contaminants , subcellular localization data of all proteins in the inclusion fraction was retrieved from UniprotKB [23] and annotations were plotted according to their SILAC ratios ( Fig 2D ) . This data clearly showed that proteins from mitochondria , the nucleus and the PM appeared at SILAC ratios of 1 and lower , and therefore are most likely contaminants of the inclusion fraction . The majority of proteins annotated with the terms cytoplasmic vesicle , ER , ER-Golgi intermediate compartment ( ERGIC ) , GA and lysosome were separated from the contaminants with a SILAC ratio above 1 . 5 , demonstrating an enrichment of these proteins in the inclusion isolation procedure of infected cells vs . uninfected cells ( Fig 2D ) . Statistical testing based on the SILAC ratio distribution in the lysate and in the inclusion fractions revealed 351 host proteins that were significantly enriched in the inclusion fraction , of which 253 were highly reliable due to the presence of high ratios in all three replicates , resulting in small multiplicity adjusted p values of below 0 . 01 ( S2A Fig ) . An additional 98 proteins were qualified as enriched with reduced statistical confidence ( multiplicity adjusted p value < 0 . 05 , S2B Fig ) . These 351 host proteins are thus considered to be inclusion associated ( S1 Table ) . Of the approximately 50 host proteins known to be recruited to Chlamydia inclusions , 23 were identified in our analysis ( S2 Table ) . These proteins included 14-3-3 ß , CERT , VAP-A , VAP-B , Rab1 , Rab6A , Rab11A and Rab14 [19 , 24–27] . These known inclusion-associated proteins were distributed across the SILAC ratios , further increasing our confidence in the generated inclusion proteome data set ( Fig 2C ) . We next validated the obtained data by confocal microscopy . To this end , 26 newly found inclusion-associated proteins with different SILAC ratios were chosen . Proteins of interest were either detected after ectopic expression of tagged fusion proteins or by visualizing endogenous proteins using specific antibodies ( S3 and S4 Figs ) . Non-fused eGFP was used as control . Localization of these proteins in infected cells was assessed after IF staining counterstained with an IncA-specific antibody to visualize the inclusion membrane and were then analyzed by laser scanning confocal microscopy ( LSCM ) ( Figs 2E and S3 and S4 ) . To confirm the presence of the fluorescently tagged proteins in the inclusion fraction , inclusions were also isolated from cells transiently expressing the respective fusion proteins ( Figs 2E and S3 ) . In total , 26 proteins were included in the validation process . From these 26 proteins , 19 proteins were validated positively , either by inclusion isolation or by immunofluorescence microscopy . Among these positive hits were YFP-RAB3D wild-type , VCP-eGFP , eGFP-SYNGR2 , eGFP-Rab8A , GFP-Syntaxin 7 , STIM1 and Sec22b . As expected , no co-localization of eGFP was observed ( S3 Fig ) . Five proteins were evaluated as false-positive including eGFP-Cofilin-1 , Sequestosome-1 and Arginase-1 ( S3 and S4 Figs ) . For two proteins the localization to the inclusion as monitored by fluorescence microscopy was ambiguous ( S3 and S4 Figs ) . Furthermore , recruitment of Rab3D appears to be an active process , as the dominant negative form of Rab3D ( YFP-RAB3D T36N ) was not found at the inclusion ( Fig 2E ) . Taken together , we have identified 351 host cell proteins that are significantly enriched in the isolated inclusion fraction and thus contribute to the host cell-derived inclusion proteome . Based on this core host cell-derived inclusion proteome , we analyzed the contribution of cellular organelles to the proteome of isolated inclusions . Subcellular localization data of the identified proteins was retrieved from UniprotKB to calculate the relative contribution of different organelle types to the obtained proteomes . We observed a clear enrichment of proteins annotated as components of the ER , the PM , the ERGIC , the GA , endosomes and cytoplasmic vesicles ( Fig 3A ) . As expected , relative depletion was seen for proteins annotated as nuclear and mitochondrial ( Fig 3A ) . Next , we performed a gene ontology ( GO ) enrichment analysis based on GO of biological processes ( GOBP ) ( S3 Table ) . The most highly enriched single term apart from ER-specific processes was `establishment of protein localization´ ( GO:0045184 ) with a p value of 3 . 94 x 10–13 and a total of 86 proteins contributing to this category . Proteins from this term were analyzed for specific complexes of interacting proteins using STRING 9 . 1 [28] . This interaction map revealed four clusters of highly interacting proteins including a cluster composed of the SNX-BAR retromer , a complex involved in retrograde trafficking from endosomes to the TGN ( Fig 3B ) . The most granular ( i . e . highly resolved ) GO term apart from ER-related processes was `vesicle-mediated transport´ ( p = 1 . 66 x 10–10 , GO:0016192 , n = 58; n = 72 including child terms ) . To further characterize these trafficking pathways that are putatively involved in the function of the inclusion , we analyzed the contribution of proteins involved in anterograde and retrograde transport to the proteome ( Fig 3C ) . Proteins involved in retrograde trafficking constitute 39% of these proteins , with retrograde transport from endosomes to the GA being the largest group within the retrograde trafficking group ( 17% of total ) . Strikingly , components of the human retromer were highly enriched in the host cell-derived inclusion proteome compared to total cell lysates , including proteins of the SNX family and the retrograde-transport cargo protein Ci-M6PR , which are among the 25% most highly enriched proteins ( Fig 3D ) . In summary , the host cell-derived proteome of C . trachomatis inclusions reveals a complex intracellular compartment enriched for SNX-BAR retromer and suggests that the inclusion interacts with multiple cellular trafficking pathways , including this retrograde transport pathway . Based on the high enrichment of retromer components on C . trachomatis inclusions , we performed IF studies using antibodies specific for SNX1 , SNX2 , VPS35 and Ci-M6PR to confirm localization of these proteins to the inclusion using LSCM ( Figs 4A and S5A ) . SNX5 and SNX6 localizations were analyzed after ectopic expression of eGFP-SNX fusion proteins ( Figs 4B and S5B ) . In uninfected HeLa cells , signals for SNX1 and SNX2 , were found in punctuated structures in the cytosol consistent with the reported endosomal localization of these SNXs ( S5A Fig ) . In contrast , in C . trachomatis-infected HeLa cells , SNX1 , SNX2 , eGFP-SNX5 and eGFP-SNX6 were detected as a rim-like staining pattern that partially co-localized with the bacterial inclusion marker , IncA ( Fig 4A and 4B ) . Recruitment of these SNXs was specific , as other members of the SNX family ( SNX3 and SNX12 ) did not co-localize with the inclusion membrane ( S6 Fig ) . Furthermore , these SNXs were also found in IncA-positive fibers emanating from the inclusion body ( Fig 4C ) . Interestingly , VPS35 and Ci-M6PR did not show a rim-like inclusion-staining pattern , but rather were depicted as small punctuated structures adjacent to the inclusion , suggesting that the membrane-deforming and receptor-recognition subcomplex of the human retromer are at least partially disconnected at the inclusion ( Fig 4A ) . To confirm the separation of these two subcomplexes , SNX2 and VPS35 were simultaneously localized in infected and uninfected cells ( Figs 4D and S7 ) . Interestingly , at the inclusion , a separation of the two signals was observed . Co-localization of the two signals in defined punctuated structures at the inclusion was rarely seen ( Fig 4D ) . In contrast , in uninfected cells , signals for both subcomplexes were clearly co-localized ( S7 Fig ) . Pearson's correlation coefficient also suggested only a moderate co-localization of the two signals at the inclusion , whereas a strong correlation was detected in punctuate-structures in the cytoplasm of either infected or uninfected cells ( S7 Fig ) . To avoid artifacts due to overexpression of eGFP-SNX2 , we also performed experiments in cells expressing eGFP-VPS35 and stained for endogenous SNX2 ( S7B Fig ) , confirming that the retromer subcomplexes do not co-localize at the inclusion , indicating separation or dissociation of the retromer complex . No difference in protein abundance for all tested retromer components was detected in C . trachomatis-infected cells compared to control cells ( Fig 4E ) . These observations demonstrate that during C . trachomatis infection SNX-BAR proteins become recruited to the inclusion and the localization of the two retromer subcomplexes is dramatically changed . Given that SNX-BAR proteins of the retromer are recruited to the C . trachomatis inclusion at 24 h p . i . , we tested whether knockdown of retromer components by RNA interference ( RNAi ) affects C . trachomatis infection including inclusion formation and development of infectious EBs . We used pools of small-interfering RNAs ( siRNAs ) to target SNX1 , 2 , 5 and 6 . Silencing of these proteins did not affect the formation of inclusions as analyzed by inclusion size and numbers ( Fig 5A and 5B ) . Interestingly , silencing of SNX5 resulted in a clear increase in infectious EBs compared to control transfections ( Fig 5C ) . SNX1 , 2 and 6 knockdown also increased infectious progeny , albeit only marginally ( Fig 5C ) . Genome copy numbers upon silencing of the different SNX proteins were slightly affected , showing the strongest increase in genome copy numbers in SNX5 knockdown cells ( Fig 5D ) . Immunoblotting confirmed that upon knockdown , the targeted SNX-BAR proteins were drastically reduced compared to control treated cells ( S8A Fig ) . We confirmed published data that silencing of SNX5 also resulted in a decrease in protein level of SNX1 ( S8A Fig ) . To elucidate if the observed increase in infectious progeny in SNX5 knockdown cells is dependent on co-regulating the abundance of the other SNX proteins , we silenced SNX5 in combination with SNX1 , 2 or 6 and measured infectious progeny formation ( S8B Fig ) . Number of infectious bacteria was increased under all combinational knockdown conditions compared to control , suggesting that other SNX proteins do not contribute to the observed increase in infectious progeny formation in SNX5 knockdown cells . Knockdown efficiencies in these double knockdown cells were confirmed by immunoblotting ( S8A and S8C Fig ) . Taken together , these results suggest that individual SNX-BAR proteins might have distinct functions in addition to controlling the retrograde transport of specific receptors . SNX5 in particular might be a rate-limiting factor and involved in intracellular replication of C . trachomatis , most likely independently of the other SNX-BAR retromer components . Retro-2 was identified in a high-throughput screen for small molecules that inhibit the toxicity of the plant toxin ricin in cell culture and was additionally found to efficiently protect cells from secreted bacterial toxins , including Shiga-like toxin and cholera toxin by inhibiting retrograde trafficking of these toxic agents from the endosomes to the GA or the ER without affecting trafficking of endogenous cellular retrograde-transport cargo proteins including Ci-M6PR [29] . SNX1 , SNX2 and eGFP-SNX5 recruitment to the inclusion was detected starting from 12 h p . i . Interestingly , association of eGFP-SNX6 with the inclusion was detected slightly later ( S9 Fig ) . At 16 h p . i . all inclusions were positive for the four different SNX proteins , coinciding with the expansion of the inclusion ( S9 Fig ) . Taking this into account , we treated cells prior to SNX recruitment ( 8 h p . i . ) with different concentrations of Retro-2 and assessed the formation of infectious EBs by re-titration at 48 h p . i . Treatment of C . trachomatis-infected cells with Retro-2 resulted in a dose-dependent decrease by more than one order of magnitude in EB numbers compared to the vehicle control ( Fig 6A ) . Reducing the treatment duration from 40 h to 28 h by shifting the time point of Retro-2 addition to 20 h p . i . still showed a decrease in infectious progeny formation albeit to a much lesser extent ( S10 Fig ) . The progression of the chlamydial developmental cycle was not affected as EB formation peaked at 48 h p . i . under both conditions , even though fewer EBs were recovered from the Retro-2 treated sample ( Fig 6B ) . Retro-2 treatment reduced the size of C . trachomatis inclusions at 24 h and 48 h p . i . by about 40% without changing the shape of the inclusions ( S11 Fig ) . Pretreatment of EBs with high Retro-2 concentrations ( 200 μM ) before infection did not reduce infectious progeny compared to vehicle control ( Fig 6C ) and numbers of bacterial genomes were only slightly affected by the inhibitor ( Fig 6D ) . To elucidate the effect of Retro-2 treatment on induction of chlamydial persistence , the ultrastructure of Retro-2 treated and control infected cells were determined by electron microscopy ( Fig 6E ) . No signs of persistence in Retro-2-treated infections , as characterized by the appearance of larger aberrant Chlamydia forms were observed . Quantification of bacterial numbers confirmed that Retro-2 treatment affects replication of the bacteria which is in agreement with Retro-2 effects on genome copy numbers ( S12 Fig and Fig 6D ) . Interestingly , we also detected a slight increase in numbers of intermediate bodies and ghosts in C . trachomatis inclusion grown in Retro-2 treated cell cultures compared to solvent control ( S12 Fig ) . A recovery assay in which infected cells were treated with Retro-2 from 8–48 h p . i . , followed by removal of the inhibitor and additional incubation for 48 h in the absence of the inhibitor , confirmed that Retro-2 does not induce chlamydial persistence ( Fig 6F ) . These experiments demonstrated that treatment of C . trachomatis infected cultures with Retro-2 strongly reduced the number of infectious bacteria at 48 h p . i . and upon removal the number of infectious bacteria remained on a low level . In contrast , the bacteria nearly completely recovered after removal of the well-known persistence inducer , penicillin G ( Fig 6F ) . In summary , our data show that C . trachomatis infections are Retro-2 sensitive resulting in smaller inclusions with slightly less bacteria inside , but with a strong defect in the generation of infectious EBs without induction of persistence . We have shown that SNX5 and Retro-2 act on C . trachomatis infections , albeit with opposite effects on the bacteria . To further determine which effect is dominant , cells were treated with siRNA pools specific for SNX5 , SNX1 and luciferase . Luciferase was used as non-targeting control while SNX1 knockdown served as additional control , as it did not significantly increase the EB numbers ( Fig 5C ) . Infected knockdown cells were either treated with a single dose of Retro-2 at 8 h p . i . or mock-treated . Infectious progeny number was determined 48 h p . i . ( Fig 6G ) . As expected , in vehicle-treated SNX5 knockdown cells , the characteristic increase in EB numbers upon knockdown of SNX5 was observed ( Fig 6G ) . Interestingly , this increase in EB numbers in comparison to SNX1 knockdown and non-targeting control was lost upon Retro-2 treatment ( Fig 6G ) . To assess whether Retro-2-sensitive retrograde transport is involved in recruiting SNX proteins to the inclusions , the localization of SNX proteins after Retro-2 treatment was analyzed at 12 h , 16 h and 24 h p . i . by confocal microscopy . In these imaging studies , no change in SNX localization was observed ( S13 Fig ) . These data show that the increase in numbers of infectious EB after the silencing of SNX5 is Retro-2 sensitive whereas recruitment of SNX proteins to the inclusion appears to be Retro-2 insensitive . The previous inability to isolate Chlamydia inclusions enforced severe experimental constraints and impeded progression in our comprehension of virulence mechanisms and the development of novel anti-chlamydial therapies . For example , recruitment of cellular proteins to the inclusion could only be addressed by microscopy . Direct biochemical evidence for the association of these factors with the inclusion membrane was therefore missing . To overcome this limitation , we established a method to isolate C . trachomatis inclusions at 24 h p . i . and analyzed isolated inclusions using a quantitative proteomics approach to decipher the host-derived C . trachomatis inclusion proteome . We used the recently described protocol for the isolation of LCV from D . discoideum [16] as a starting point , but due to the fragile nature of the C . trachomatis inclusion , this protocol was heavily modified . As a result , we retained a two-step protocol but started with a Percoll-based gradient followed by immuno-magnetic separation using an IncA-specific antibody . One of the critical steps in the isolation protocol was the lysis of the infected host cells . We carefully tested different buffer and infection conditions , but the majority of inclusions were ruptured at this step resulting in a maximum recovery of 15% of the calculated initial numbers of inclusions . The yield in the following steps ( gradient and MACS ) was about ~50% amounting to a total recovery rate of about 8% . This recovery rate is in the range or even slightly higher than the yields obtained for Legionella containing vacuole isolations [16 , 30] . The second challenge was to find an optimal strategy for initial purification of the visually intact inclusions from cellular debris . We used isopycnic density gradient centrifugation to separate inclusions from host cell debris . We recovered the majority of inclusions in solution by fractionation of the gradient , but apparently the buoyant density of inclusions is very diverse , distributed across the range of densities of intracellular organelles , thus a subpopulation escaped our analysis which was distributed over the whole gradient without apparent peaks . It seems likely that these are inclusions that either contained large amounts of glycogen [31] or lipid droplets which are known to be translocated into the lumen of inclusions [32] . This translocation could have a considerable effect on their overall density . This speculation is supported by the absence of markers for lipid droplets in our proteome analysis . Moreover , we detected inclusions ranging in size from 3 μm up to 10 μm , representing the majority of expected inclusion sizes , possibly with a slight bias towards smaller inclusions , which could result from an increased fragility of larger inclusions . The high sensitivity of modern LC-MS/MS-based proteomics demands an experimental design which includes a strategy to distinguish between bona fide components of the isolated compartment as well as co-purified contaminations . To this end , we used a SILAC-based exclusion approach in combination with label-free absolute quantification . A similar method was successfully used in a recent study to identify contaminants in purified latex bead-containing phagosome preparations [33] . Underlining the success of the purification and SILAC exclusion approach , we found a significant proportion of previously reported inclusion-associated proteins in our dataset . To further investigate the sensitivity of our assay , we ranked the proteins detected in a deep proteome of HeLa cells [22] by the iBAQ value of tryptic peptides , to see if highly abundant proteins are over-represented in the overlap with previously known inclusion-associated proteins ( S14 Fig ) . Our limit for reliable detection of proteins with more than one peptide is slightly above the median iBAQ intensity in the HeLa cell lysate ( S14 Fig ) . This is satisfying , considering the technical difficulties due to massive amounts of bacterial peptides present in our samples . However , based on these data , the true number of inclusion-associated proteins might be significantly higher than what we report here , probably around two times greater than the reported number based on known host proteins associated with inclusions . Furthermore , the SILAC exclusion approach has also some limitations , for example with proteins that have a high dissociation constant , which reduces the SILAC ratio due to exchange of L- for H-labeled proteins during the extended incubation time in cell lysate before MACS separation , thereby increasing the number of false negative classifications . These factors influence the number of reported proteins , but are all likely to reduce the reported number rather than to lead to false positives . Whereas originally the inclusion was thought to be a separated compartment that acts as a niche devoid of host proteins [34] , this picture has changed dramatically in recent years as indicated by the extensive interaction with cellular organelles and recruitment of specific proteins , often mediated by bacterial effectors , which was first described for 14-3-3 ß [27] . Interestingly , proteins annotated as nuclear , mitochondrial and lysosomal were significantly depleted in the Chlamydia inclusion proteome . Proteins assigned to other cellular organelles contributed significantly to the inclusion proteome , suggesting the inclusion is embedded in the intracellular trafficking network of the host cell . This conclusion supports the view that the C . trachomatis inclusion is a complex intracellular trafficking platform that exploits different pathways to foster optimal intracellular growth , rather than that of an isolated niche . For an obligate intracellular pathogen that lacks a number of genes for the biosynthesis of essential nutrients , this integration into the host cell organellar network seems reasonable to secure intracellular survival [35] . We noted redundancy in interactions which could reflect robustness of the intracellular lifestyle , which is further supported by the fact that C . trachomatis can infect and grow in an array of different cell types . Detailed analysis of the host cell-derived inclusion proteome showed that C . trachomatis inclusions interact with the retromer , an important complex regulating retrograde transport of different cellular receptors and a pathway also hijacked by bacterial and plant toxins and distinct viruses to intoxicate and infect cells [6 , 36–38] . In Chlamydia-infected cells , the SNX-BAR proteins SNX1 , 2 , 5 and 6 , are recruited to the inclusions decorating the inclusion in a rim-like staining pattern and are additionally found on IncA-laden fibers emanating from the inclusion body . In this context , it is interesting to note that Salmonella enterica serovar Typhimurium acquire SNX1 and SNX3 , and SNX1 is found on spacious vacuole-associated tubules early in the infection process [39 , 40] . In uninfected cells , the PX and BAR domains of SNX-BAR proteins target these proteins to phosphoinositide-enriched , high-curvature membranes [41 , 42] . Phosphatidylinositol-4-phosphate ( PI4P ) has also been detected in the inclusion membrane by expression PIP-sensitive reporter proteins [43] . Whether the detected PI4P or additional bacterial proteins such as Inc proteins that are present in the inclusion membrane are involved in recruiting the SNX-BAR proteins to the inclusions is currently not known . Interestingly , the cargo recognition subcomplex of retromer showed only a punctual localization at the inclusion membrane . Consequently , there is partial separation of the two retromer subcomplexes at the inclusion membrane but not in other locations of infected cells . These observations support recent findings on the structure and function of the cellular retromer . Firstly , whereas the retromer complex is a stable hetero-pentamer in yeast cells , this association is much more transient in mammalian cells [44] and secondly , the two subcomplexes and the individual SNX-BAR proteins are involved independently of each other in trafficking of distinct cargo [45–47] . Functional analysis of SNX-BAR proteins using RNAi showed that in particular SNX5 knockdown resulted in an increase in infectious progeny . This may indicate that SNXs , and in particular SNX5 , become segregated by recruitment to the C . trachomatis inclusion , thereby affecting the cellular retrograde trafficking pathways . The activity of the retromer complex has often been linked to processes controlling the sorting of cellular receptors including the epidermal growth factor receptor ( EGFR ) and M6PR [48 , 49] . SNX5 in particular has been implicated in EGFR trafficking and signaling in uninfected cells [48] . For C . trachomatis infections it has recently been demonstrated that EGFR activity is important for maturation of the inclusion by controlling calcium signaling and actin remodeling [50] . In light of these and our findings it is tempting to speculate that SNX5 recruitment to the inclusion alters e . g . EGFR transport and signaling inside the cells which in turn triggers calcium release and F-actin rearrangements . These changes then support the development of a proper C . trachomatis inclusion and are thus important for a successful infection . Alternatively , distinct SNX-BAR proteins control a currently not well-defined Retro-2-sensitive retrograde trafficking pathway that delivers distinct nutrients to the bacteria or alternatively could be related to factors controlling innate immunity . The idea of an innate immunity-related function of the retromer is further supported by the recently published observation in Drosophila that retromer can also control the Toll pathway [51] . The observed sensitivity towards the retrograde inhibitor Retro-2 also supports the view that retrograde transport is important for C . trachomatis progeny formation . The molecular target of Retro-2 is currently unknown but treatment results in displacement of the three t-SNAREs syntaxin ( Stx ) 5 , 6 and 16 from membranes of the Golgi apparatus . These t-SNAREs are essential for retrograde transport of different cargo molecules to the TGN [52] . Interestingly , the localization of Stx6 to the inclusion has also been documented using microscopy and lack of Stx6 slightly but significantly reduced C . trachomatis infectious progeny [53 , 54] . Whether the strong inhibitory effect of Retro-2 treatment on C . trachomatis growth and infectious progeny formation is a result of mislocalization of different t-SNAREs from the inclusion or if additional proteins are also targeted by the treatment remains to be determined . Experiments are in progress to address Retro-2 dependent changes on a global level to determine these factors , which will potentially identify the molecular target of Retro-2 and might also uncover novel functions of the evolutionarily highly conserved retromer complex . In summary , we have deciphered the core host cell-derived proteome of the C . trachomatis inclusion 24 h p . i . by quantitative proteomics of isolated inclusions . This data set describes the inclusion as a highly complex and interactive compartment that amongst others recruits proteins normally forming the membrane-binding subcomplex of the cellular SNX-BAR retromer . Of the subset of SNX-BAR proteins , SNX5 controlled the formation of infectious Chlamydia progeny in a Retro-2 sensitive pathway highlighting the importance of distinct SNX-BAR proteins and the retrograde transport for C . trachomatis infections . Thus , the development of a technique to isolate Chlamydia inclusions fosters our understanding of the inclusion composition , the contribution of cellular factors to inclusion formation and maintenance . This may pave the way for the development of axenic culture conditions and novel anti-chlamydial strategies . HeLa cells were grown in Roswell Park Memorial Institute medium ( RPMI , Gibco ) 1640 supplemented with 10% fetal calf serum ( FCS , Biochrom ) at 37°C and 5% CO2 in a humidified incubator . The cells were routinely tested for Mycoplasma contamination via polymerase chain reaction ( PCR ) using the VenorGeM kit ( Biochrom ) according to manufacturer’s instructions . C . trachomatis L2 lymphatic isolate 434 Bu ( ATCC: VR-902B ) was propagated in HeLa cells . For more details on infections , determination of infectious progeny formation , the quantification of relative bacterial genome copy number , infection recovery assay , bacterial morphology assay and measurement of inclusion size , see S1 Text . For plasmid transfections , HeLa cells were grown to 80% confluency and transfected with Lipofectamine 2000 reagent ( Invitrogen ) according to manufacturer’s instructions . For knockdown of target host cell proteins , HeLa cells were transfected with pools of target specific siRNAs as described in S1 Text . For the standard procedures TEM , IF , SDS-PAGE , immunoblotting , molecular cloning as well as used reagents , plasmids and oligonucleotides , see S1 Text . HeLa cells were infected with C . trachomatis ( MOI 4 ) at 70–90% confluence . For standard isolations , 6 x 107 cells were used . All steps were done on ice or in a cold room at 4°C . Cells were washed once with PBS and subsequently with ice cold HSMG buffer ( 20 mM HEPES , 250 mM sucrose , 1 . 5 mM MgCl2 , 0 . 5 mM EGTA , pH 7 . 4 ) . Cells were scraped into 6 ml lysis buffer ( 33% Percoll solution ( Sigma ) , HSMG ) supplemented with cOmplete EDTA free protease inhibitors ( Roche ) . Lysis was performed by repeated passage through a ball homogenizer ( Isobiotech ) using 16 μm clearance and 11–13 passages . The lysate was then separated on a self-forming Percoll gradient in a total volume of 16 ml by centrifugation at 35’000 x g for 30 minutes at 4°C ( Beckmann RC-6 with Thermo Scientific F21-8x50y rotor ) . The lower 6 ml of the gradient were either used for MACS purification or crude inclusions were diluted six fold in HSMG and pelleted at 1500 x g for 10 minutes , followed by another wash and centrifugation at 1200 x g for 10 minutes . For MACS separation , crude inclusions were incubated with rabbit αIncA ( 1:1000 ) antibody [55] for 1 . 5 h at 4°C , followed by incubation with MACS secondary goat anti-rabbit antibody ( 1:100 , Miltenyi ) for another 1 . 5 hours . Inclusions were mixed gently by inversion every 30 minutes . The crude inclusions were loaded on a MACS LS separation column ( Miltenyi ) column in steps of 2 ml and washed with three times the input volume of HSMG buffer . Inclusions were then eluted with 3 ml HSMG buffer after removal of the magnet , aided by gentle pushing using the supplied plunger . Counting of inclusions , the small scale isolation procedure for validation and processing of inclusions for IF and TEM are described in supporting information ( S1 Text ) . For SILAC experiments , cells were grown in SILAC DMEM ( PAA ) containing dialyzed FCS ( Biochrom ) , supplemented with H labeled L-arginine ( 13C615N4 ) and L-lysine ( 13C615N2 ) ( Silantes ) or non-labeled amino acids ( L ) . Inclusions were isolated as described above but H labeled mock infected cells were mixed with equal amounts with L labeled infected cells prior to cell lysis . Inclusion samples were prepared for LC-MS/MS . 10% of the sample was used for direct injection after desalting . The remaining peptides were separated by strong anion exchange chromatography into 6 fractions before desalting , followed by LC-MS/MS . Lysate samples were prepared for LC-MS/MS without pre-fractionation . For more details , see S1 Text . Tryptic peptides were analyzed using a data dependent method on a Q Exactive mass spectrometer ( Thermo ) coupled to a Ultimate 3000 nHPLC ( Dionex ) for separation by reverse phase chromatography . The resulting . raw files were analyzed in MaxQuant 1 . 3 . 0 . 5 [56] . Protein groups that had less than two unique + razor peptides in at least one experiment were filtered . See S1 Text for more details on SILAC enrichment analyses , abundance analyses using iBAQ and further bioinformatics analyses .	The important human pathogen Chlamydia trachomatis causes 100 million new infections each year world-wide . It replicates inside the infected host cell in a unique vacuole , the inclusion . Currently , the nature , and specifically the protein composition of the inclusion , is poorly defined . Here , we described the host cell-derived inclusion proteome by quantitative proteomics using a newly established method to purify inclusions from infected epithelial cells . We showed that the inclusion is a complex intracellular trafficking platform that is well embedded into the organellar network and interacts with host cells’ antero- and retrograde trafficking pathways . Particularly , SNX1 , 2 , 5 and 6 , components of the retromer , are recruited to the inclusion and seem to control the infection . We found also that retrograde trafficking is essential for Chlamydia progeny formation . Our study provides new insights into how the obligate intracellular bacterium C . trachomatis interacts with the eukaryotic host cell and subverts host cell functions for productive infection .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[]	2015	The Proteome of the Isolated Chlamydia trachomatis Containing Vacuole Reveals a Complex Trafficking Platform Enriched for Retromer Components
The epileptic network is characterized by pathologic , seizure-generating ‘foci’ embedded in a web of structural and functional connections . Clinically , seizure foci are considered optimal targets for surgery . However , poor surgical outcome suggests a complex relationship between foci and the surrounding network that drives seizure dynamics . We developed a novel technique to objectively track seizure states from dynamic functional networks constructed from intracranial recordings . Each dynamical state captures unique patterns of network connections that indicate synchronized and desynchronized hubs of neural populations . Our approach suggests that seizures are generated when synchronous relationships near foci work in tandem with rapidly changing desynchronous relationships from the surrounding epileptic network . As seizures progress , topographical and geometrical changes in network connectivity strengthen and tighten synchronous connectivity near foci—a mechanism that may aid seizure termination . Collectively , our observations implicate distributed cortical structures in seizure generation , propagation and termination , and may have practical significance in determining which circuits to modulate with implantable devices . Localization-related epilepsy causes seizures that arise from one or more abnormal islands of cortical tissue in the neocortex or mesial temporal structures [1] . In more severe cases , seizures with focal onset secondarily generalize , as pathologic activity spreads across the brain [2] . Localization-related epilepsy represents ≈80% of epilepsy cases and is often resistant to medication [3] . For drug-resistant patients , the only treatment options are implantable devices , or more traditionally resective surgery to remove enough cortical tissue in the epileptic network to decrease seizure frequency , while preserving brain tissue responsible for eloquent function . In surgical cases where discrete lesions associated with seizure onset ( ‘foci’ ) are not evident on an MRI , only ≈40% remain seizure-free post-surgery [3] . The modest outcome associated with these procedures has lead investigators to further explore spatial distributions of epileptic activity using multiscale neural signals in ECoG and sub-millimeter μECoG to more accurately localize where seizures start and how their pathologic activity spreads [4–9] . These approaches have spurred a paradigm shift from localizing just the foci towards informing interventions by mapping structural and functional connectivity of the whole epileptic network . The notion of an epileptic network stems from the idea that pathologic functional connections and/or disconnections disrupt neural function , producing rhythmic motor activity , altered cognition , or abnormal sensation . Functional connections are time-dependent [10] communication pathways between neural populations that are measured by statistical relationships between electrode sensor ( node ) time series [11] , and that evolve according to brain state to produce behavior . The seizure state was originally considered to be hypersynchronous , or composed predominantly of strong functional connections . In contrast , a significant body of recent work presents compelling evidence that complex changes among strong ( synchronized ) and weak ( desynchronized ) network nodes accompany seizure dynamics [12–17] . The state-space of these dynamics are well described at the sensor level using measures of node centrality [18 , 19] . However , epileptic network architecture at the basic sub-unit of individual connections is poorly understood , but tremendously powerful for discriminating fine-grain network changes that drive seizure dynamics . Understanding the interplay between individual functional connections in the epileptic network is critical to answer questions goading clinical epileptologists and translational researchers: Where do seizures start ? Can the epileptic network be modulated therapeutically ? What can these methods reveal about the underlying neurophysiologic mechanisms ? Progress in addressing these questions requires methods to track time-dependent functional connections within the epileptic network and understand their relative strengths and weaknesses , which in network terms are collectively referred to as the network’s geometric structure . Such methods would not only shed light on geographical dysfunction of epileptic foci , but also the disruption of normal brain tissue that is recruited during seizure events . We hypothesize that the epileptic network achieves dysfunction and drives seizure activity by reconfiguring network connections during key network states that are clinically described as seizure generation , propagation , and termination . Our network reconfiguration hypothesis is informed by recent work demonstrating that human brain networks dynamically reorganize prior to changes in behavior [20 , 21] . During pathologic events , reconfiguration in epileptic networks may involve a redistribution of metabolic resources between strong and weak connections , supporting distinct network functions [22 , 23] . Our results support this hypothesis , demonstrating that the epileptic network can be characterized by hubs of persistent strong connections surrounded by rapidly reconfiguring weak connections that drive seizure processes . Do functional connectivity patterns significantly change as a seizure progresses ? To answer this question , we developed a new method to uncover network states , defined by unique patterns of sensor-sensor functional connectivity between T time windows ( Fig 2 ) . We define a network state to be the set of all configuration vectors that exhibit a similar pattern of functional connectivity , more formally known in the network science literature as “network geometry” . To quantify geometric similarity , we calculated the Pearson correlation coefficient between configuration vectors extracted from all possible pairs of T time windows . This procedure produced a symmetric T × T configuration-similarity matrix ( Fig 2c ) . We next ask whether clusters of time windows exhibit similar configuration patterns indicative of independent network states ( Fig 2d ) . To test for distinct states in each epileptic event , we used an unsupervised clustering approach for networked data—community detection—that maximizes a modularity quality function Q obtained from the configuration-similarity matrix ( see Materials ) . In this approach , a structural resolution parameter γ can be tuned to maximize the reliability of state estimates; we separately tuned this parameter for each seizure and pre-seizure epoch in each patient ( see S1 Text ) . This procedure assigns each time window to a community ( or state ) , and each state is composed of time windows that exhibit similar network geometry . Note that these time windows need not be temporally contiguous . We found that the epileptic network transitions through a variety of network states during pre-seizure and seizure epochs ( Fig 3a–3b ) . A comprehensive summary of epoch and state durations for each patient can be found in Table A in S1 Text . The existence of epileptic state transitions support the notion of a dynamically reconfiguring network . To quantify reconfigurability of the epileptic network , we measured the network flexibility , or rate the of state change in each epoch ( Fig 3c ) . We found that pre-seizure epochs display significantly higher flexibility ( μ = 0 . 665±0 . 205 ) than seizure epochs ( μ = 0 . 274±0 . 165 ) ( paired-samples t-test; t87 = −14 . 12 , p = 2 . 2 × 10−16 ) , indicating that the epileptic network transitions between states more slowly through seizure epochs than through pre-seizure epochs . Furthermore , pre-seizure epochs consisted of many short-duration states , while seizure epochs consisted primarily of 3 long states that occupy ≈87% of seizure duration ( Fig 3d ) . The three largest pre-seizure states occupied approximately 75% of the epoch . Together , these results support the possibility that rapid changes in network geometry in pre-seizure epochs lead to seizures , and once there , the network undergoes slower geometric changes through 3 main dynamic states . To fairly assess differences in seizure and pre-seizure states , we retained the 3 longest network states from seizure ( S0 , S1 , S2 ) and pre-seizure epochs ( PS0 , PS1 , PS2 ) for the following analyses . In the previous section , we observed that seizures progress through distinct states characterized by different functional connectivity patterns . To understand how these patterns differ , we used a two-pronged approach , examining ( i ) the strength of functional connections and ( ii ) the pattern of functional connections in different network states ( Fig 4a ) . For simplicity , we report the strength of functional connections as a fraction of the total strength , and we refer to this quantity as the connection density . Similarly , to characterize the pattern of functional connections , we examine the relative prevalence of synchronized ( strong ) versus desynchronized ( weak ) connections , and we refer to this quantity as the connection type index . The functional connection density measures the average connection strength in the network , where greater connection density indicates increased global network synchrony . We computed connection density by averaging the distribution of all connection strengths over all time windows in the given network state . We performed a one-way ANOVA to compare the effect of pre-seizure and seizure network states on connection density . We observed a significant effect of network state on connection density ( F5 , 474 = 21 . 34 , p < 2 × 10−16 ) . Post-hoc analysis using Tukey’s honest significant difference test ( HSD ) to control for a family-wise rejection error rate of 5% ( FWER = 5% ) revealed a significant increase of connection density in each seizure state compared to any pre-seizure state . During the seizure , connection density increased between S0 ( μ = 0 . 304±0 . 051 ) and S1 ( μ = 0 . 333±0 . 58 ) ( padj = 0 . 014 ) , and S0 and S2 ( μ = 0 . 338±0 . 052 ) ( padj = 0 . 002 ) , but did not significantly change between S1 and S2 ( padj = 0 . 995 ) . Differences in connection density between the pre-seizure states ( PS0 , PS1 , PS2 ) were not significant . These results suggest synchronization increases as the network transitions from pre-seizure to seizure states . While we observed an increase in global synchrony as seizures begin and progress , it is unclear whether this increase accompanies a change in functional connectivity pattern , and particularly in a switch from relative desynchronization ( weak connectivity ) to synchronization ( strong connectivity ) . To type individual connections as strong or weak , we ( 1 ) compiled a distribution of all functional connections over all time windows across each event ( encompassing the pre-seizure and seizure epoch ) , and ( 2 ) determined thresholds for connection type based on rank percentile , where strong ( weak ) connections were stronger ( weaker ) than 95% of all connections . Based on connection type assignments in each epoch , we found the total number of strong ( Cs ) and weak ( Cw ) connections over all time windows in each network state and computed the connection type index as C s - C w C s + C w . A strong type dominant network has a connection type index between 0 and +1 , where +1 implies all connections are strong , while a weak type dominant network has a connection type index between 0 and −1 , where −1 implies all connections are weak . To determine the effect of network state on connection type index ( Fig 4b ) , we conducted a one-way ANOVA test . We observed a significant effect of network state on connection type index ( F5 , 474 = 70 . 41 , p < 2 × 10−16 ) . Post-hoc analysis using Tukey’s HSD ( FWER = 5% ) indicated a significant change from weak type dominance during any pre-seizure state towards strong type dominance during seizure states . During the seizure , connection type index increased between S0 ( μ = 0 . 023±0 . 498 ) and S1 ( μ = 0 . 437±0 . 417 ) ( padj < 2 × 10−16 ) , and S0 and S2 ( μ = 0 . 512±0 . 447 ) ( padj < 2 × 10−16 ) , but did not significantly change between S1 and S2 . Differences of connection type index between the pre-seizure states ( PS0 , PS1 , PS2 ) ( μ ≈ −0 . 401 ) were not significant . Chronologically , the network is persistently desynchronized during the pre-seizure epoch , is driven to a quasi-synchronized seizure generation state S0 , and remains persistently synchronized as the seizure progresses through S1 and S2 . A predominance of weak connections during a persistently desynchronized pre-seizure epoch coincides with earlier findings of improved network flexibility to reorganize during the same epoch . Unremarkable change in weak connection type dominance during the pre-seizure epoch suggests that the network simply redistributes weak connections amongst different nodes during this period . A critical transition to seizure generation during state S0 is accompanied by synchronization towards more evenly distributed strong and weak connection types . As network flexibility decreases during the seizure , connections become more strong type dominant . To better understand how the network evolves through the desynchronized and synchronized states , we next study the impact of local , geographical changes in network geometry . In the preceding analyses , we demonstrated that the epileptic network displays weak type dominant connectivity during pre-seizure epochs and undergoes synchronizes to strong type dominance as the seizure initiates and progresses through 3 primary states . However , our approaches did not address whether these reconfigurations are spatially localized or distributed , and how they relate to seizure foci . To address these questions , we leveraged routine clinical procedures: A team of neurologists successfully identified the sensors on the seizure onset zone ( SOZ ) based on visual inspection of the intracranial recordings in 15 patients across a total of 50 seizures . We used this information to map connections in each seizure state to physical electrode locations in stereotaxic space ( Fig 5a ) . To quantify spatial localization of connectivity relative to seizure foci and examine the role of network region in pre-seizure and seizure dynamics , we delineated the following three geographic types: ( i ) connections between nodes within the SOZ ( SOZ-SOZ ) , ( ii ) connections between nodes outside the SOZ ( OUT-OUT ) , and ( iii ) connections between one node within the SOZ and one node outside the SOZ ( SOZ-OUT ) ( Fig 5b ) . We performed a two-way ANOVA test to compare the effects of geography and network state on connection strength . We observed a significant main effect of geography on connection strength ( F2 , 882 = 158 . 501 , p < 2 × 10−16 ) and a significant main effect of network state on connection strength ( F5 , 882 = 26 . 394 , p < 2 × 10−16 ) . We also observed significant interactions between geography and network state ( F10 , 882 = 2 . 871 , p = 0 . 002 ) . Post-hoc analysis on the interactions using Tukey’s HSD ( FWER = 5% ) identified persistently stronger connection strength amongst SOZ-SOZ connections ( μ ≈ 0 . 393±0 . 140 ) relative to OUT-OUT ( μ ≈ 0 . 282±0 . 059 ) and SOZ-OUT ( μ ≈ 0 . 284±0 . 059 ) connections in every network state ( padj < 1 × 10−3 ) . Connections in the SOZ-SOZ group were modestly strengthened during S0 relative to PS0 and PS2 ( padj < 0 . 05 ) , were greatly strengthened during S1 and S2 relative to any pre-seizure state ( padj < 1 × 10−3 ) , and during the seizure only strengthened between S0 to S2 ( padj < 0 . 05 ) . However , connection strengths in the SOZ-SOZ group did not significantly vary between pre-seizure states . Similarly , SOZ-OUT and OUT-OUT group did not significantly vary between any network states . These results suggest that SOZ-SOZ connections are persistently the strongest of all network connection types during pre-seizure and seizure epochs . Upon seizure generation SOZ-SOZ connections strengthen incrementally , and then substantially as seizures progress . Nuancing our description of global network connectivity during pre-seizure and seizure epochs , which demonstrates a progression from desynchronization to synchronization over time , our results demonstrate that ( i ) desynchronization during pre-seizure states is primarily localized to SOZ-OUT and OUT-OUT connections , and ( ii ) resynchronization is primarily localized to SOZ-SOZ connections . Intuitively , desynchronous SOZ-OUT and OUT-OUT connections that frequently re-wire drives heightened network flexibility during pre-seizure epochs and synchronous SOZ-SOZ connections disrupts network flexibility during the seizure . To investigate the sensitivity and specificity of connection strength as a measure for identifying SOZ-SOZ connections , we employed receiver operating characteristic ( ROC ) analysis during pre-seizure and seizure epochs ( Fig 5c ) . The ROC analysis evaluates the sensitivity and specificity of connections belonging to the SOZ-SOZ type as connection strength threshold is incrementally raised . We evaluate performance in detecting SOZ-SOZ connections by computing the area under the ROC curve ( AUC ) ranging from 0 to +1 , where values of +1 imply low sensitivity and false positives with high specificity and true positives . To assess significance of the AUC , we bootstrapped confidence intervals ( α = 0 . 05 ) by re-assigning sensors to the SOZ uniformly at random without replacement 10000 times for each network state in both epochs . During seizure epochs , we found that S2 was most effective at predicting SOZ-SOZ connections based on AUC ( μ = 0 . 849±0 . 169 ) with significant AUC values in 32 of 50 seizures . Conversely S0 was least effective at predicting SOZ-SOZ connections ( μ = 0 . 773±0 . 238 ) with significant AUC values across 26 of 50 seizures . During pre-seizure epochs , SOZ-SOZ connections were similarly predictable across PS0 ( μ = 0 . 709±0 . 268 ) ( significant in 25 of 50 ) , PS1 ( μ = 0 . 722±0 . 257 ) ( significant in 25 of 50 ) , and PS2 ( μ = 0 . 754±0 . 238 ) ( significant in 27 of 50 ) . These results suggest that connection strength may be used to predict SOZ-SOZ connections during pre-seizure epochs with precision , but has better performance during more synchronized states such as S2 compared to less synchronized states such as S0 . Thus far we have seen how connectivity associated with the SOZ synchronizes the epileptic network during seizures . However , it is unclear whether involvement from the broader epileptic network aids or disrupts pre-seizure and seizure dynamics . We first hypothesized that changes in network geometry are not limited to redistribution of connection strengths , but may also involve topographical changes in connection lengths accompanying changes in functional network anatomy . In a sample of pre-seizure and seizure states , we observed clustering of strong connections while weak connections distributed more broadly ( Fig 5a ) . To test our hypothesis , we restricted our analysis to connections within electrode grids with uniformly spaced nodes in 8 × 8 , 8 × 6 , 6 × 6 , or 4 × 6 configurations ( in 75 seizures over 19 patients ) and computed average Spearman’s rank correlation coefficient between connection length and connection strength over all time windows of each network state ( Fig 6a ) . A more positive ( negative ) correlation coefficient indicated stronger connections were longer ( shorter ) . A one-way ANOVA test was conducted to compare the effect of pre-seizure and seizure network states on correlation between connection length and connection strength . We observed a significant effect of network state on correlation ( F5 , 444 = 9 . 348 , p = 1 . 76 × 10−8 ) . Post-hoc analysis using Tukey’s honest significant difference test ( HSD ) to control for a family-wise rejection error rate of 5% ( FWER = 5% ) revealed significant increase in negative correlation between connection length and strength in S0 ( μ = −0 . 225±0 . 92 ) compared to PS2 ( μ = −0 . 182±0 . 091 ) ( padj < 0 . 05 ) but not PS0 ( μ = −0 . 188±0 . 100 ) or PS1 ( μ = −0 . 189±0 . 092 ) . Connection length is significantly more negatively correlated with connection strength in S1 ( μ = −0 . 258±0 . 081 ) and S2 ( μ = −0 . 242±0 . 086 ) compared to any pre-seizure state ( padj < 0 . 01 ) . There was no significant change in correlation between pre-seizure states or seizure states . In summary , we found that stronger connection strengths are present in connections with shorter lengths , regardless of network state . During seizures , reorganization in the epileptic network leads to further lengthening of weaker connections and shortening of stronger connections . Coinciding with the earlier finding that seizure generation involves quasi-synchronization of the network , we find a modest shortening of strong connections relative to the pre-seizure period . As seizures progress , synchronous connections tighten to more local regions , while desynchronous connections stretch further into the broader epileptic network . Intuitively , complex reconfiguration of functional brain networks can accompany changes in cognitive state or changes in behavior . Prior fMRI studies have explored such reconfiguration in whole-brain networks constructed from data acquired during motor skill learning [20] and as task states change [25] , and in networks impacted by stroke [26 , 27] . In contrast , here we explore the reconfiguration of a local area and use higher resolution ECoG data to map the fine-scale temporal dynamics of reconfiguration processes . In this study , we developed and exercised a novel method for distinguishing brain states based on differences in time-dependent functional network geometry . Our approach expands upon previous notions of state-space in dynamic epileptic networks [18 , 19] , by tracking changes between node pairs ( connections ) rather than in node importance ( centrality ) . An important advantage associated with this technique is that network reorganization can be studied without a priori knowledge of specific topological structure , such as small-worldness [16] . Rather , time-dependent changes in connectivity are based simply on similarities in signal statistics . We applied our technique to a set of human ECoG recordings , and extracted network dynamics during seizure and pre-seizure epochs . We found that seizures exhibit at least three network states ( S0 , S1 , S2 ) and that the epileptic network progresses through these states more slowly in comparison to the period preceding seizure generation . Our results are in line with prior work that has shown more frequent state changes during the interictal period in comparison to seizures [19] . Next , we provide a mechanistic explanation of how state changes operate with strong and weak regimes of connectivity to drive seizures through neurologically-defined onset , propagation and termination states ubiquitous in clinical descriptions . Our analytical approach utilizes the distribution of functional connection strengths to characterize connections as “strong” ( synchronous ) or “weak” ( desynchronous ) , rather than simply stating that two sensors are functionally “connected” or “not connected” . Mathematically , this focus corresponds to a study of network geometry as opposed to network topology . A primary advantage of the weighted network approach is the ability to separate connections into classes that differ in strength . Evidence suggests that strong and weak connections play different roles in supporting cognitive function [23 , 28] . Traditional thought is that strong connections represent primary communication pathways between brain areas . However , recent work demonstrates that weak connections support increased network efficiency and may play a large role in distinguishing pathologic [29] and healthy [23 , 30] network states . From a dynamical perspective , strong connections may persistently engage throughout neurophysiological processes , whereas weak connections may engage transiently to enable brain state transitions . Prior work has speculated that the epileptic network is connected at the beginning of the seizure , disconnected in the middle , and finally reconnected at the end [12 , 14 , 16] . However , our results suggest that a more accurate way to address this hypothesis is to consider the strength of functional connections and disambiguate slower temporal dynamics occurring at each node , independently , which may elevate spurious connectivity between disconnected regions . Using a weighted connectivity approach , we find that connections in the epileptic network have more weak than strong connections during PS0 , PS1 , and PS2 , states preceding the electrographic seizure onset , a near balance of strong and weak connections during S0 , which corresponds to seizure generation , and more strong than weak connections during S1 and S2 states representing seizure progression and termination . It is possible that clinician subjectivity in marking the time of seizure onset may explain our result of disconnectivity before seizure generation , which contrasts with prior reports of a disconnected network at either seizure onset or mid-seizure [12 , 14 , 16] . Our method places greater emphasis on connectivity derived from faster activity by reducing contribution from slower dynamics ( see Materials and Methods ) and corroborates clinical belief that seizure generation during S0 involves a gradual transition from desynchronous to synchronous connectivity , which peaks during the termination phase of the seizure ( S2 ) . Mechanistically , the weak connectivity that we observe preceding seizure generation benefits from high network flexibility to drive seizure generation through a rapid reorganization of weaker connections in the epileptic network . As seizures initiate and progress , the epileptic network redistributes weak connectivity to strong connectivity while network flexibility is concurrently diminished . In relation to prior work that demonstrates a propensity for the epileptic network to follow a recurring pattern of state transitions during seizures [19] , our results suggest that weak connectivity preceding the seizure drives the network to a more predictable series of increasingly synchronized states during seizures . Next , we explore beyond global network structure and discuss how regional connectivity dynamics provide further insight on network drivers of seizure evolution . While temporal network structure provides rich information regarding seizure states , it does not directly provide information regarding the spatial processes involved in seizure dynamics . We therefore complemented the temporal network approach by incorporating information about sensor role either within or outside the seizure onset zone and sensor location in Euclidean space . Our results demonstrate that these additional spatial features provide new insights into potential neurophysiological mechanisms involved in seizure generation , and may inform the development of clinical tools for objectively isolating the seizure onset zone directly from seizure or pre-seizure data . Prior work has demonstrated high synchronization within the seizure onset zone during interictal epochs [31 , 32] . However , the temporal dynamics and geometrical roles of these two sets of areas has remained elusive . Our results elucidate the role played by seizure onset regions during seizures and the accompanying recruitment of the surrounding epileptic network during termination . Clear isolation of the seizure onset zone exists in pre-seizure periods , suggesting the potential to identify foci , niduses of seizure generation , within the network from inter-ictal data . Critically , connectivity within the onset zone strengthens during early seizure periods ( S1 ) and intensifies as seizures progress ( S2 ) and terminate ( S3 ) , suggesting that the onset zone drives the transition from global desynchronization to synchronization during seizure generation and persists in this functional role through the entire seizure . Such a mechanism also points to a role of the onset zone in seizure termination , potentially in tandem with topographical mechanisms , which we discuss in the next section . Our observation that stronger connections are typically short and weaker connections are typically long , is consistent with results from two lines of research: ( i ) functional studies in healthy individuals that utilize other imaging modalities such as fMRI [23] and ( ii ) structural connectivity studies in non-human primates that utilize tract tracing techniques [22] . In epilepsy , prior work has shown hubs of connectivity proximal to the seizure onset zone [18 , 33 , 34] , however their role in seizure dynamics was previously unknown . We show that seizure generation leads to further shortening of stronger connections and lengthening of weaker connections , suggesting that stronger connections are physically tightening , perhaps into more functionally cohesive portions of cortex during seizures . We speculate that the tightening of stronger connections to localized sub-networks might act as a control mechanism to quench disruptive network activity that may have built-up over many hours prior to the seizure through increasing frequency of epileptiform discharges [24] or facilitate previously described compensatory mechanisms [18] . Conversely , weaker connections are longer during seizure periods than pre-seizure periods and could be a vehicle for spreading desynchronous activity broadly . We have seen that the dynamical processes that propel epileptogenic networks into seizures can be complex and are poorly understood . Yet , clinicians rely on visual inspection to describe spatial and temporal properties of seizures . The lack of standardized clinical measures to mark epileptic events calls for the development of automated methods . The network analysis tools we have built , while generally applicable to any dynamic network , can parse seizure states , localize driver ‘foci’ of seizures , and characterize how seizures progress and terminate . This interpretation can be translated into useful clinical tools to identify dysfunctional anatomical regions that drive the epileptic network and may be particularly amenable to local interventions , such as surgery or device placement . Of interest , seizure driving ‘foci’ were equally present in the half of our study patients who did not have focal lesions on brain imaging , compared to those patients with lesions demonstrated on MRI . We plan a more detailed study in the future to correlate mapping of these seizure-driving regions with brain resection and outcome . Currently , our tools employ community detection techniques to identify gross changes in the meso-scale architecture of network structure across time . The observed meso-scale reconfiguration processes may be accompanied by region-specific trends in reconfiguration between the epileptic network and surrounding healthy networks . A remaining gap is understanding how functional dynamics map to structural features of the epileptic network using fiber-tracking techniques to describe how seizures start and then spread through white-matter . Additionally , this work could be used to address cellular mechanisms by considering micro-scale reconfigurations . Recent studies suggest that epileptic networks in the neocortex may be composed of distributed micro-domains on the scale of a few cortical columns generating high frequency oscillations and micro-seizures that coalesce in a network during seizure generation and termination [6] . While of great interest , these studies are currently limited by the lack of appropriate implantable high-resolution sensors capable of covering clinically relevant areas sufficiently to yield comprehensive high-resolution maps . Further development of dynamic community detection methods to identify and track reconfiguration within network sub-regions at both the meso and micro-scales may help delineate healthy and pathologic networks and uncover mechanisms of network recruitment . An important clinical consideration related to this work is the impact of sampling error inherent in any intracranial implantation procedure on our results . Any technique used to map epileptic networks , subdural electrode strips and grids , more distributed “Stereo EEG” implantations , and combinations of these two approaches , usually yield incomplete representations of epileptic networks . It is not possible to fully record from the entirety of cortex in affected patients . In some cases this might mean that neither seizure onset zones nor all regions of seizure spread are fully delineated . Despite this incomplete representation , the presence of three clear states defining seizures in each of the patients presented above , and their objective and independently determined relationship to the seizure onset zone suggest that our findings are important and real . With further validation on a larger number of patients with both lesional and non-lesional epilepsies , we hope to demonstrate the utility of our method to define functional components of epileptic networks . Our method shows promise for informing epilepsy surgery and for placing devices into regions that drive seizure generation and termination . Future work will focus on using these methods to compare competing approaches for localizing epileptic networks , such as subdural and stereo EEG . It is intuitively plausible that each will have advantages in recording components of epileptic networks in different types of localization-related epilepsy . All patients included in this study gave written informed consent in accord with the University of Pennsylvania Institutional Review Board and Mayo Clinic Institutional Review Board for inclusion in this study . Twenty-one patients undergoing surgical treatment for medically refractory epilepsy believed to be of neocortical origin underwent implantation of subdural electrodes to localize the seizure onset zone after noninvasive monitoring was indeterminate . De-identified patient data was retrieved from the online International Epilepsy Electrophysiology Portal ( IEEG Portal ) [35] . ECoG signals were recorded and digitized at either 512 Hz ( Hospital of the University of Pennsylvania , Philadelphia , PA ) or 500 Hz ( Mayo Clinic , Rochester , MN ) sampling rate . Surface electrode ( Ad Tech Medical Instruments , Racine , WI ) configurations , determined by a multidisciplinary team of neurologists and neurosurgeons , consisted of linear and two-dimensional arrays ( 2 . 3 mm diameter with 10 mm inter-contact spacing ) and sampled the neocortex for epileptic foci ( depth electrodes were first verified as being outside the seizure onset zone and subsequently discarded from this analysis ) . Signals were recorded using a referential montage with the reference electrode , chosen by the clinical team , distant to the site of seizure onset and spanned the duration of a patient’s stay in the epilepsy monitoring unit . We analyzed a total of 88 seizure events , including simple partial , complex partial , and secondarily generalized , stemming from neocortical foci in this study . Seizure onset time and localization were defined by the point of earliest electrographic change ( EEC ) and annotated and marked by a team of practicing epileptologists [24] . ECoG signal directly preceding each seizure and equal in duration to that seizure was also extracted for balanced comparison and labeled as pre-seizure . Signals from each epoch were divided into 1-second , non-overlapping , wide-sense stationary time windows in accord with other studies [16] and subsequently pre-processed . To test the biasing effect of high-amplitude spiking on signal connectivity measurements , we also investigated windows 0 . 5-seconds in duration to sample more of the non-biasing temporal space and found similar results . In each time window , signals were re-referenced to the common average reference [16 , 36] to account for variation in reference location across patients and to avoid broad field effects that may bias connectivity measurements erroneously in the positive direction . Each window was filtered at 60 Hz to remove line-noise , and low-pass and high-pass filtered at 120 Hz and 1 Hz , respectively , to account for noise and drift . To correct for correlated signal dynamics for each individual node , we pre-whiten signals in each window and reduce autocorrelation effects for time lags greater than zero . This accomplishes two goals: ( i ) flattening of the signal power spectrum to enhance higher-frequency content that contains local neural population dynamics , and ( ii ) decreases the influence of independent node dynamics when computing correlation-based connectivity measurements [36–39] . Dynamic functional networks were formed by applying a normalized cross-correlation similarity function ρ between the time series of two sensors in the same time window using the formula ρ xy ( k ) = E [ ( x k ( t ) - μ x k ) ( y k ( t + τ ) - μ y k ) ] ( 1 ) where x and y are signals from one of N sensors or network nodes , k is one of T non-overlapping , one-second time windows , and xk = yk = 0 . The NxNxT similarity matrix is also known as a network adjacency matrix A ( Fig 1a ) . In our weighted network analysis approach , we retain and analyze all possible connection weights between nodes . Network states , or temporal changes in network geometry , was tracked separately in each epoch by clustering the configuration-similarity matrix through a community detection technique known as modularity optimization . We construct the configuration-similarity matrix by first unraveling A to a network evolution matrix A ^ describing the weights of N ( N - 1 ) 2 connections across T time windows . Using a Pearson correlation coefficient to measure similarity , we transform A ^ to a fully-connected TxT configuration state adjacency matrix S . The configuration adjacency matrix is partitioned into communities by maximizing the modularity index Q[40] using a Louvain-like locally greedy algorithm [41] . We employed a Newman-Girvan null model [42 , 43] and adaptively determined an optimal structural resolution parameter γ per seizure ( see S1 Text and [44] for a more detailed discussion of resolution parameters in modularity maximization ) . We used a consensus partition method with 1000 optimizations per run until we obtained consistent community partitioning [44 , 45] . The three longest communities ( clusters , or network states ) from each seizure were selected for further analysis and re-labeled in order of median temporal occurrence for population-level comparison . Connections were classified as strong or weak based on thresholds determined by the distribution of connection strengths for each epoch separately for each seizure . The strong ( weak ) connections must be >95% ( <5% ) of all connection strengths . To measure the dominance of strong or weak connections , we defined the connection type index as B = C s - C w C s + C w ( 2 ) where Cs and Cw are the average number of strong and weak connections over possible connections and number of time windows . Connection topography metrics were computed for only within-grid electrodes , ignoring all other non-grid electrodes such that inter-electrode spacing in all analyses was kept constant . We related connection strength to the two-dimensional physical distance between nodes ( electrode sensors ) of that connection in millimeters .	Localization-related epilepsy is a debilitating condition where seizures begin in dysfunctional brain regions , and is often resistant to medication . The challenge for treating patients is mapping connections between cortical structures that vary with time and drive seizure dynamics . While it is well known that whole-brain functional architecture reconfigures during tasks , we hypothesize that epileptic networks reconfigure at the meso-scale leading to seizure generation , propagation , and termination . We develop new methods to track dynamic network reconfiguration amongst connections of different strength as seizures evolve . Our results indicate that seizure generation is primarily driven by rapidly reorganizing weak connections that drive stronger connections to further strengthen and topographically tighten as seizures progress and terminate . These findings may have practical clinical implications for targeting specific connections with implantable , therapeutic devices to control seizures .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[]	2015	Dynamic Network Drivers of Seizure Generation, Propagation and Termination in Human Neocortical Epilepsy
The GI tract is preferentially targeted during acute/early HIV-1 infection . Consequent damage to the gut plays a central role in HIV pathogenesis . The basis for preferential targeting of gut tissues is not well defined . Recombinant proteins and synthetic peptides derived from HIV and SIV gp120 bind directly to integrin α4β7 , a gut-homing receptor . Using both cell-surface expressed α4β7 and a soluble α4β7 heterodimer we demonstrate that its specific affinity for gp120 is similar to its affinity for MAdCAM ( its natural ligand ) . The gp120 V2 domain preferentially engages extended forms of α4β7 in a cation -sensitive manner and is inhibited by soluble MAdCAM . Thus , V2 mimics MAdCAM in the way that it binds to α4β7 , providing HIV a potential mechanism to discriminate between functionally distinct subsets of lymphocytes , including those with gut-homing potential . Furthermore , α4β7 antagonists developed for the treatment of inflammatory bowel diseases , block V2 binding to α4β7 . A 15-amino acid V2 -derived peptide is sufficient to mediate binding to α4β7 . It includes the canonical LDV/I α4β7 binding site , a cryptic epitope that lies 7–9 amino acids amino terminal to the LDV/I , and residues K169 and I181 . These two residues were identified in a sieve analysis of the RV144 vaccine trial as sites of vaccine -mediated immune pressure . HIV and SIV V2 mAbs elicited by both vaccination and infection that recognize this peptide block V2-α4β7 interactions . These mAbs recognize conformations absent from the β- barrel presented in a stabilized HIV SOSIP gp120/41 trimer . The mimicry of MAdCAM-α4β7 interactions by V2 may influence early events in HIV infection , particularly the rapid seeding of gut tissues , and supports the view that HIV replication in gut tissue is a central feature of HIV pathogenesis . Gut associated lymphoid tissue ( GALT ) is a primary target for HIV and SIV , particularly in the early weeks of infection [1 , 2] . Within days after transmission , high levels of proviral DNA can be isolated from GALT [2–4] . Subsequently , gut CD4+ T cells are severely depleted , and the structural integrity of GALT is to a large extent irreversibly damaged in a way that is thought to contribute to chronic immune activation [5–7] . Administration of anti-retroviral therapy ( ART ) , even shortly after infection , fails to fully reverse this damage [8] . These early events in infection are believed to contribute in a significant way to the immune dysfunction that characterizes HIV disease [5] . Thus , the early seeding of gut tissues plays a central role in HIV pathogenesis . We and others have demonstrated that both HIV and SIV recombinant envelope proteins directly bind integrin α4β7 ( α4β7 ) , a gut homing receptor [9–12] , while notably , some studies have failed to detect this interaction [13–15] . α4β7 is not required for viral entry [15–17] . However , our findings raise the possibility that there exists a link between the gut-tropic aspect of HIV infection and this physical interaction . It is possible that α4β7 functions simply as an attachment factor [18] . However , gp120 binding to α4β7 , like mucosal addressin cellular adhesion molecule ( MAdCAM ) transduces signals to primary CD4+ T cells , suggesting that such signals may be relevant to infection in vivo [9 , 17 , 19] . In this regard we have recently reported that MAdCAM delivers a signal to CD4+ T cells that promotes cellular activation and viral replication [19] . α4β7 is expressed on the cell-surface membrane of a number of cellular subsets including most naive CD4+ T cells and a subset of memory CCR5+/CD4+ T cells [20] . Similar to each of the 24 human integrins , α4β7 is a heterodimer . It is comprised of a 180 kDa α4 chain [21] and a 130 kDa β7 chain [22] . α4β7 is structurally dynamic and can adopt at least three conformational states , two of which are extended , and competent to mediate lymphocyte adhesion [23] . Transition between conformations is tightly controlled intracellularly , which provides a regulatory mechanism for α4β7 activity [24] . The normal function of α4β7 involves binding to two adhesion receptors , MAdCAM and vascular addressin cellular adhesion molecule ( VCAM ) , along with the alternatively-spliced III connecting segment ( CS ) fragment of fibronectin [25] . Importantly , α4β7 is the only integrin capable of binding to MAdCAM [26] . In healthy adults MAdCAM is expressed on follicular dendritic cells in gut tissues [27 , 28] and on endothelial cells lining the lumen of high endothelial venules in GALT and the gut lamina propria [29–31] . The specificity of MAdCAM for α4β7 , along with its tissue specific expression are the two factors that define α4β7 as the gut homing integrin receptor . There is growing evidence that α4β7 plays a significant role in the pathogenesis of HIV disease . It has been shown that α4β7high CD4+ memory T cells are preferentially infected during both HIV and SIV acute infection [8 , 32] . Additionally , the frequency of α4β7high CD4+ memory T cells is correlated with risk of acquisition in both SIV and HIV [8 , 33] , and in HIV this association was shown to be independent of other markers of cellular activation [8] . Sexually transmitted diseases ( STDs ) , which are associated with increased risk of acquisition of HIV , increase the frequency of α4β7high CD4+ memory T cells in both genital mucosa and blood [34 , 35] . In one recent study of HIV infected women , pre-infection levels of peripheral blood α4β7high CD4+ memory T cells correlated with the rate of CD4+ T cell decline post-infection [8] . In an SIV rhesus macaque model , a substantial proportion of animals pretreated with an anti α4β7 monoclonal antibody ( mAb ) were protected from infection following repeated low-dose vaginal challenge [36] . The same mAb , when combined with ART , promoted durable immune-mediated control of viremia in SIV infected macaques after all forms of therapy were withdrawn [37] . Taken together , these findings demonstrate the importance of α4β7-expressing cells in HIV/SIV infection , and also in the ensuing host immune response , and underscore the need for a more complete understanding the role of α4β7 in the pathogenesis of HIV/SIV disease . Previous studies have demonstrated that the carboxy-terminal region of the V2 domain of gp120 interacts with α4β7 [9 , 10 , 12] . Using site directed mutagenesis of gp120 we reported that a tripeptide motif leucine-aspartic acid-valine or isoleucine at positions 179–181 ( L179D180V/I181 ) in the V2 domain plays a central role in this interaction . This tripeptide motif is similar to critical binding epitopes in the natural ligands of α4β7 . MAdCAM encodes a leucine-aspartic acid-threonine ( LDT ) , VCAM encodes an isoleucine-aspartic acid-valine ( IDV ) , and the IIICS fragment of fibronectin encodes leucine-aspartic acid-valine ( LDV ) . The key feature of each is a core aspartic acid flanked by an aliphatic residue . In each of these natural ligands the aspartic acid coordinates with a Mg++ ion that sits in the metal ion dependent adhesion site ( MIDAS ) of β7 . Mg++ ion coordination is a strict requirement for ligand binding [23 , 38] . Cardozo and colleagues identified nearby amino acids , glutamine-arginine-valine ( QRV ) ( 170–172 ) that also influence V2-α4β7 interactions , demonstrating that the binding site is not limited to the LDV/I tripeptide [12] . These two regions of V2 are flanked by potential N-linked glycosylation sites ( PNGs ) , and removal of these flanking PNGs can enhance binding of recombinant gp120 to α4β7 [16] . It is unknown whether this enhancement is due to relief from steric hindrance , or allosteric changes in the α4β7 binding epitope . High-resolution cryo-electron microscopy and X-ray diffraction analyses indicate that the three V2 domains appear at the apex of the trimeric envelope spike [39–41] . In these structures , as well as in structures derived from scaffolded V1/V2 proteins and monomeric gp120 subunits , the region of V2 from 170–181 appears in the context of a β strand or β barrel [40–45] . In high-resolution structures , the LDV/I appears partially or fully buried in a way that would seem to make it inaccessible to α4β7 . Thus , it is reasonable to conclude that the context in which α4β7 binds to V2 must involve either a rearrangement of these structures , or an alternative presentation of V2 . In this study , we characterized the physical interaction between the HIV envelope and α4β7 reasoning that such information could provide valuable insight regarding the role of α4β7-expressing cells in HIV pathogenesis . In this report , we demonstrate that a region near the carboxy-terminus of gp120 V2 appears to mimic , to a significant degree , the way in which MAdCAM engages α4β7 . In this regard , MAdCAM utilizes dynamic and tightly regulated changes in the conformation of α4β7 to regulate α4β7-expressing lymphocyte access to both gut inductive and effector sites . Thus , this mimicry may provide HIV a mechanism to access gut tissues in a relatively efficient way , and argues that viral replication in gut tissues is central to HIV pathogenesis . One consequence of this mimicry is that drugs developed to antagonize MAdCAM-α4β7 interactions could also disrupt V2-α4β7 interactions . In addition , we find that a subset of HIV and SIV V2 antibodies derived from both infected subjects and vaccine recipients can effectively block V2 α4β7 interactions . Several of the vaccine -elicited weakly neutralizing mAbs have been linked with protection from infection . Rather than binding to the closed trimeric spike that is the primary target of broadly neutralizing antibodies these mAbs recognize an alternative conformation of the V2 region . This suggests that α4β7 also recognizes an alternative form of V2 , that is conserved in both HIV and SIV . To characterize the interaction between gp120 and α4β7 we employed two assays . We developed a novel surface-plasmon resonance ( SPR ) based assay that utilized dextran surfaces coated with recombinant envelope ( env ) proteins , V1/V2 scaffolds , or synthetic V2 cyclic peptides . The analyte that we reacted with these surfaces was a recombinant soluble α4β7 heterodimer in which the carboxy-terminal transmembrane and cytoplasmic tail domains of both chains were removed and replaced by short peptides that function as an “α4 chain acid-β7 chain base coiled-coil clasp” [46] . This acid-base clasp was joined by a disulfide bond that served to stabilize the heterodimer . In one iteration of this assay we employed short linear peptides derived from V2 as competitive inhibitors . The second assay we employed was a static adhesion assay based on the method developed by Peachman and colleagues , in which RPMI8866 cells , that express α4β7 on the cell surface , were allowed to adhere to the recombinant env proteins , V1/V2 scaffolds or synthetic V2 cyclic peptides ( S1A Fig ) . The α4β7-expressing RPMI8866 cell line was derived from a human B cell lymphoma , and expresses α4β7 , but no detectable CD4 or CCR5 . Cells were grown in media containing retinoic acid , which increased both levels of expression , and clustering of α4β7 ( S2B Fig ) . In some assays we included anti-integrin and anti-gp120 mAbs as adhesion inhibitors . The SPR assay allowed us to evaluate the kinetics of integrin-gp120 binding , while the cell-based assay measured adhesion between two multivalent surfaces . Previous studies describing the interaction between gp120 and α4β7 have demonstrated , in a qualitative way , the specific interaction between these two proteins , without establishing an estimate of affinity [9–12] . We reasoned that a quantitative comparison of the binding kinetics of gp120 vs . MAdCAM to α4β7 would help determine if α4β7-gp120 interactions mimic α4β7-MAdCAM interactions . We carried out a kinetic analysis using the SPR assay noted above . Soluble α4β7 ( analyte ) was passed over surfaces coupled with either recombinant gp120 or a MAdCAM-Ig fusion protein ( ligands ) . A recombinant A244 gp120 ( subtype A/E produced by Global Solutions for Infectious Diseases ( GSID ) ) was employed in these assays . We initially measured binding kinetics in the presence of a buffer containing 1mM MnCl2 in order to uniformly stabilize α4β7 in an extended/activated conformation ( discussed below ) . Under these conditions MAdCAM and A244 gp120 demonstrated comparable high affinities ( KD ( nM ) ) of 0 . 597 and 7 . 140 , respectively ( Fig 1A and 1B ) . These affinities are generally comparable to gp120-soluble CD4 binding kinetics ( e . g 22 nM KD ) [47] . When MnCl2 was removed , affinity for both MAdCAM and gp120 fell below the detection limit of this assay ( Fig 1C and 1D ) . This requirement for Mn++ is consistent with our previous report that gp120 , similar to MAdCAM , interacts with an extended conformation of α4β7 [9] . This observation suggests that gp120 is likely to engage α4β7 only on cells with an enhanced potential to traffic to the gut . We next replaced A244 gp120 with a synthetic cyclic 42 amino acid peptide fragment ( cV2 ) derived from the V2 domain ( AA 157–196 ( HXB2 numbering ) ) of 92TH023 gp120 ( subtype A/E ) , in which N and C termini were joined by a disulfide bond , and the C terminus was biotinylated to facilitate coupling to NeutrAvidin coated biosensor chips . The 92TH023 V2 sequence is nearly identical to that of A244 gp120 V2 ( S1A Fig ) . The affinity of this peptide ( cV2 92TH023 ) for α4β7 ( KD ( nM ) 1 . 150 ) was close to that of A244 gp120 ( Fig 1E ) , demonstrating that a cV2 is sufficient to mediate the high-affinity interaction shown in Fig 1B . Moreover , it indicates that this high-affinity interaction does not require the glycans that decorate the V2 of GSID A244 gp120 , or any bridging protein [13] . When we replaced α4β7 with α4β1 affinity for cV2 92TH023 was reduced by >8000 -fold ( ( KD ( nM ) 9710 ) ( Fig 1F ) , demonstrating binding specificity and the fact that V2 , like MAdCAM , preferentially binds to α4β7 . We extended this analysis by measuring binding kinetics for a cV2 peptide derived from C06980v0c22 ( subtype C ) and a recombinant gp120 BG505 protein ( subtype A ) ( Fig 1G and 1H ) . α4β7 bound to each with similar high affinities , consistent with the conserved nature of this interaction across HIV clades as we had originally reported [9] . The V2 domains of HIV and SIV diverge both in primary sequence , as well as in length and the typical number of disulfide bridges ( 1 vs 2 ) , suggesting significant structural differences between them . Using a recombinant gp120 derived from SIVmac766 we obtained an α4β7 affinity ( KD ( nM ) ) of 104 ( Fig 1I ) . This is ~10-fold lower than that observed for HIV A244 gp120 . It is possible that this reduction in affinity reflects differences in the primary amino acid sequences of human vs . rhesus macaque α4β7 ( Figs 2A and S1A ) , such that SIV gp120s might exhibit a higher affinity for a rhesus macaque versus a human derived α4β7 protein . Detailed α4β7 binding parameters for each of the analytes appears in S1 Table . Similarities between MAdCAM and HIV gp120 with respect to affinity , cation-dependence , and preference for binding α4β7 over α4β1 suggests that gp120 V2 mimics MAdCAM in the manner in which it engages α4β7 . A number of α4β7 antagonists , developed to treat inflammatory bowel disease ( IBD ) act by occupying MAdCAM binding sites on α4β7 . To the extent that gp120 uses these same sites , such antagonist should also interfere with gp120 adhesion . These antagonists include a class of small molecule mimetics that resemble the Leu-Asp binding motif present in the CC′ β strands of MAdCAM IgSF domain 1 [48] . They compete directly with MAdCAM by binding to key residues in the ligand binding groove formed by the interface between α4 and β7 ( Fig 2A ) . Vedolizumab is a mAb antagonist of α4β7 with a different and unique mechanism of action ( MOA ) . It binds exclusively to the specificity-determining loop ( SDL ) in β7 ( Fig 2A ) [23] . The SDL mediates secondary interactions with a negatively charged DE loop in MAdCAM IgSF domain 2 ( Fig 2B ) [23 , 49–52] . We evaluated the capacity of a small molecule LDV mimetic , ELN-475772 , and vedolizumab to interfere with gp120-α4β7 interactions using the RPMI8866-based adhesion assay described above . As expected , both ELN-475772 and vedolizumab blocked adhesion of α4β7 to immobilized MAdCAM ( Fig 2C and 2E ) . We then tested their ability to block α4β7-mediated adhesion to the cV2 92TH023 . The anti-α4 mAb 2B4 , which inhibits most α4-ligand interactions , along with human IgG were employed as specificity controls . Both ELN-475772 and vedolizumab inhibited α4β7 adhesion to V2 by >90% ( Fig 2C and 2E ) . Because the inhibitory MOAs of ELN-475772 and vedolizumab involve direct interactions with two discreet MAdCAM binding sites on α4β7 , these results suggest that soluble MAdCAM would compete with gp120 V2 in binding to α4β7 . Of note , soluble MAdCAM-Ig blocked V2-α4β7 adhesion ( Fig 2C and 2E ) . We conclude that , at least in a general way , gp120 V2 effectively mimics MAdCAM in the manner in which it engages α4β7 . α4β7 activity is regulated by its conformation . Intracellular signaling events modulate the ectodomain to reversibly transition between bent and extended conformations ( Fig 2D ) . In addition , the headpiece , which mediates ligand interactions , can assume closed , intermediate , and open conformations . Transitions between these conformations allows α4β7 to mediate both rolling and firm adhesion [23] . Rolling adhesion is associated with lower affinity binding to MAdCAM , while firm adhesion is associated with a higher affinity interaction . Manipulation of Ca++ , Mg++ , and Mn++ concentrations provides a way to manipulate the affinity of α4β7-MAdCAM interactions in vitro [38] . Strength of adhesion is highest in Mn++ , followed by Mg++ > Mg++/Ca++ , > Ca++ . We asked whether the pattern of V2 adhesion to α4β7 was similar to that mediated by MAdCAM in buffers containing Mn++ vs Mg++ , vs low cations ( Ca++ ) . Adhesion of both MAdCAM and cV2 92TH023 was strongest in the presence of MnCl2 and reduced in the presence of MgCl2 by ~5-fold and 2 . 5-fold respectively ( Figs 2F and S3 ) . In low-cation conditions MAdCAM adhesion appeared close to background levels; however , we were still able to detect residual cV2 92TH023 adhesion . Overall , V2 adhesion to α4β7 responded to divalent cations in a similar manner as did MAdCAM . This is particularly noteworthy insofar as the dynamic changes in affinity required to mediate both rolling and firm adhesion reflect the unique and highly specialized nature of α4β7-MAdCAM interactions [23] . The way in which both HIV and SIV gp120 mimic this highly specialized interaction argues that it provides them a selective advantage . The conformation of the V2 domain of gp120 is dynamic , and consequently it was deleted from the recombinant proteins used to generate the initial high-resolution gp120-mAb cocrystals [53 , 54] . Instead , V2 structures were obtained by grafting V1/V2 fragments onto scaffolds derived from unrelated proteins . These scaffolds stabilized V1/V2 in a way that , in complex with conformation-dependent V2 mAbs , allowed for the derivation of high resolution structures . The first V2 structure was obtained by presenting V1/V2 on a scaffold termed 1FD6 in complex with the broadly neutralizing , glycan dependent monoclonal antibody PG9 [43] . In this context , V2 adopted a Greek key β sheet structure ( Fig 2A ) . Another study in which the same 1FD6-V1/V2 protein was complexed with mAb 830A provided additional detail and revealed V2 in a related β- barrel conformation [42] , which is consistent with what has been observed in pre-fusion env trimers [40 , 41] . However , when a linear V2 peptide is left unconstrained it can adopt α helical structure [55] . This is the case for a crystal structure of mAb CH58 in complex with a linear V2 peptide derived from HIV isolate 92TH023 gp120 [56] . CH58 does not exhibit broad potent neutralizing activity and we refer to here as weakly- neutralizing . It recognizes a helix structure ( Fig 3A ) and is noteworthy insofar as it was generated from an uninfected immunized individual who participated in the RV144 vaccine trial . It recognizes an epitope that maps within a short region of V2 ( AA 168–181 ) , that includes two residues , K169 and I181 , identified by sieve analysis as sites of vaccine elicited immune pressure in the RV144 trial ( Fig 3B ) . Additional V2-specific mAbs that recognize helical structures in this same region have subsequently been described . We report here the cocrystal structure of a V2 peptide with one such mAb , Mk16C2 , that was generated from a gp120 immunized rabbit ( Fig 3A and S2 Table ) . It binds to the same helical structure as CH58 but approaches V2 from the opposite side . Of note , helix -preferring V2 antibodies are not limited to vaccine elicited immune responses . mAb CAP228-16H , which was derived from an HIV-infected subject , recognizes a V2 helix structure that is strikingly similar to that recognized by CH58 [57] . Thus , mAbs reacting with the region of V2 from AA 153–194 recognize at least two distinct types of epitopes: those like PG9 and 830A that recognize a constrained β- sheet , and those like CH58 that recognize a less constrained helical conformation . In an ELISA assay mAbs 830A , CH58 , CAP228-16H and Mk16C2 bind to cV2 92TH023 ( S1B Fig ) , indicating that this cyclic peptide is sufficiently long and flexible to present the epitopes recognized by each of these mAbs . Although the core epitopes of CH58 and 830A differ , the C-terminal end of the CH58 epitope overlaps the 830A epitope ( Fig 3B ) , which is consistent with our observation that CH58 can compete with 830A ( S1C Fig ) . We evaluated the ability of each of these mAbs to inhibit α4β7 interactions with V2 using the adhesion assay described above . mAbs 2B4 and VRC01 were employed as specificity controls . CH58 and CAP228-16H inhibited α4β7 adhesion to V2 by >90% ( Fig 3C ) . Mk16C2 inhibited adhesion less effectively . Of note , 830A failed to inhibit adhesion in a significant way . Thus , while a V2 mAb ( CH58 ) that recognizes a helical structure interfered with α4β7 -mediated adhesion , mAb 830A , that shows preference for the β strand , failed to show detectable interference . Rao and colleagues have recently reported that adhesion of recombinant gp120 to α4β7 requires the partial enzymatic removal of glycans [58] . This is consistent with our finding that removal of several potential N-glycosylation sites ( PNGs ) within V2 can enhance gp120 binding to α4β7 [16] . In agreement with these reports we find that GSID A244 gp120 required limited deglycosylation with PNGase , under nonreducing conditions , in order to mediate α4β7 adhesion ( Fig 4A ) . This adhesion , similar to the adhesion of α4β7 to cV2 92TH023 , was efficiently inhibited by ELN-475772 , CH58 , and CAP228-16H ( >90% ) ( Figs 4A and S4 ) . Again , Mk16C2 was less effective ( ~47% reduction ) . We also evaluated α4β7 adhesion to a BG505 SOSIP gp120/41 trimer and monomeric SIVmac766 gp120 . Unlike the two monomeric gp120s , the SOSIP stabilized protein failed to mediate adhesion after removal of glycans ( Figs 4B and S5 ) . Inability to adhere to α4β7 is not a consequence of the primary sequence of BG505 V2 since a BG505 cyclic V2 ( cV2 BG505 ) was able to mediate α4β7 adhesion ( Figs 4C and S6 ) . These finding are in agreement with the observations of Rao and colleagues . Of note , in an SPR -based assay , CH58 binds to the cV2 BG505 but not to the BG505 SOSIP ( Fig 4D ) . Given that CH58 , which recognizes a helix , efficiently blocks V2 adhesion to α4β7 , we hypothesized that the failure of the BG505 SOSIP trimer to engage α4β7 reflects an underlying conformational constraint on V2 mediated by the SOSIP stabilization strategy in which this constraint precludes the formation of a structure required for α4β7 reactivity . To address whether the β- barrel conformation of V2 is incompatible with α4β7 -reactivity , we inserted the V1/V2 sequences of 92TH023 into two scaffolds . The first scaffold , termed 1FD6 , has been previously shown to constrain V1/V2 in a way that increases its propensity to form a β- barrel [42] . The second scaffold , termed tag , consists of V1/V2 that is untethered at the C-terminus , allowing it to adopt an unconstrained , CH58 -reactive , helical conformation [42] [PMID: 27707920] . We found that the deglycosylated 92TH023 V1/V2 tag scaffold showed ~5x greater adhesion than did deglycosylated 92TH023 V1/V2 1FD6 , which mediated only minor levels of α4β7 adhesion ( Fig 4E ) . We conclude that α4β7 -reactivity requires a degree of V1/V2 flexibility that is not present in the recombinant BG505 SOSIP trimer . Our results suggest that this is due to constraints placed on V2 by other sequences encoded in the closed trimer . However , we cannot rule out interference by PNGase resistant glycans . The data presented above suggests that α4β7 recognizes a structure distinct from the closed spike on virions that is the target of many well characterized neutralizing mAbs . However , it is well established that env appearing on virions is conformationally heterogeneous [59] . We asked whether , among these various env conformations was one that is α4β7 -reactive . To address this question , we employed magnetic nanoparticles ( MNPs ) coated with either α4β7 or V2 mAbs CH58 , PG9 , and 830A to capture 92TH023 virions derived from primary CD4+ T cells ( Fig 5 ) . mAb 2G12 was employed as a positive control . After extensive washing , virion capture was measured by a Luminex -based p24 detection assay [60] . In three independent experiments CH58 , PG9 , and 2G12 each captured ~5X greater amounts of virus than non-specific IgG . Capture by mAb 830A was less efficient ( ~3X over IgG ) ( Fig 5A ) . The capacity of mAb CH58 , which recognizes a helical structure and inhibits V2 adhesion to α4β7 is able to capture virions suggesting that these virions present an α4β7 -reactive form . To test this directly we incubated virions with α4β7- MNPs in the absence or presence of increasing amounts of the α4β7 inhibitor ELN-475772 . In three independent experiments α4β7- MNPs captured virus , and this capture was inhibited by ELN-475772 in a dose -dependent manner ( Fig 5B ) . Thus , 92TH023 virions derived from primary CD4+ T cells present gp120 in an α4β7 -reactive form . Liao and colleagues demonstrated that a 15 AA peptide corresponding to residues 168–181 of V2 adopts a helical structure when complexed with mAb CH58 ( Fig 3A ) [56] . Using the SPR assay described above we asked whether a similar 15 AA peptide was sufficient to bind to α4β7 . Soluble α4β7 was passed over a cV2 92TH023 coated surface in the absence or presence of 8 overlapping linear 15 amino acid peptides . These peptides corresponded to sequences in an HIV-1 subtype B consensus V2 domain ( Fig 6A ) . Peptide H43 ( Q170KEYALFYKLDVVPI184 ) , which closely aligns with the peptide employed by Liao and colleagues , inhibited α4β7 -binding by >90% ( Fig 6B ) . This peptide includes both the canonical L179D180 α4β7 binding site and a Q170K171E172 that aligns with the QRV cryptic α4β7 epitope identified by Cardozo and colleagues [12] . We repeated this analysis but substituted cV2 92TH023 with A244 gp120 and obtained a similar result ( Fig 6C ) . We then competed mAbs CH58 and CAP228-16H with these same peptides ( Fig 6D and 6E ) . Peptides H42 ( R166DKVQKEYALFYKLD180 ) and H43 each partially inhibited mAb CH58 binding and strongly inhibited ( >90% ) CAP228-16H binding . Taken together these results suggest that inhibition by peptide H43 involves direct binding to α4β7 . To rule out allosteric inhibition ( i . e . peptide H43 binding directly to , and altering the conformation of V2 ) , we carried out a similar peptide inhibition assay with immobilized MAdCAM and determined that H43 inhibited α4β7 binding to MAdCAM by >90% ( Fig 6F ) . This result is best explained by direct competition between the Leu179-Asp180 in peptide H43 and the critical Leu41-Asp42 encoded within MAdCAM IgSF domain 1 ( Fig 6G ) . However , we believe it is very likely that other residues in H43 also engage α4β7 . We conclude that an epitope contained within a linear peptide corresponding to residues 170–181 of V2 binds directly to α4β7 . This same region of V2 overlaps the epitopes recognized by CH58 and CAP228-16H . The α4β7 binding epitope of V2 is conserved in SIV as exemplified by the specific affinity of SIVmac766 gp120 for human α4β7 ( Fig 1I ) . To determine whether this region in SIV V2 is also involved in α4β7 -binding , we carried out a similar competition binding experiment as described above in Fig 6 , but utilized SIVmac239 derived V2 overlapping peptides , and immobilized SIVmac766 gp120 , along with SIVsmE660-CR51 gp120 ( Fig 7A ) . Of the eight SIV peptides we employed only S46 showed strong inhibition ( >90% ) ( Fig 7B ) . S46 is the SIVmac239 V2 domain peptide that corresponds to HIV peptide H43 ( Fig 7C ) . Peptides H43 and S46 show limited sequence identity , but notably 5 residues: K171/183 , E172/184 , Y173/185 , Y177/190 , D180/193 , appear to be conserved ( Fig 7C ) . We next asked whether SIV V2 mAbs could inhibit α4β7 adhesion to V2 . Five V2 mAbs , ITS03 , ITS09 , ITS12 . 01 , ITS41 , and NCI09 were evaluated using the same strategy outlined above for HIV V2 mAbs ( see Fig 3C above ) . We also included ITS13 , a V1 mAb and VRC01 an HIV CD4 binding-site mAb as reagent controls . Mapping studies of these V2 mAbs have been described in detail elsewhere [61] and are summarized in Fig 8A . The epitope for mAb ITS12 . 01 spans SIV V2 residues 187–197 and falls within S46 . It includes the key Asp that is conserved in HIV , SIV and MAdCAM IgSF domain 1 . However , ITS12 . 01 did not block α4β7 adhesion to SIVmac766 gp120 ( Fig 8B ) . Instead , ITS03 and NCI09 , which recognize residues NH2-terminal to the ITS12 . 01 epitope inhibited adhesion most efficiently ( >90% ) . ITS09 and ITS41 also inhibited adhesion , but to a lesser extent , while NCI05 , which maps to a region overlapping the ITS09 epitope failed to inhibit adhesion in a detectable way . Of note , the epitopes for both ITS03 and NCI09 do not include the canonical Leu-Asp binding site ( Ala192-Asp193 in SIV ) that is conserved in HIV V2 and MAdCAM ( Fig 8A ) . Both of these mAbs do however overlap with the corresponding region of HIV V2 , that includes the epitopes for CH58 and CAP228-16H , both of which inhibit α4β7 adhesion to HIV V2 . In summary , a 15-amino acid linear peptide derived from SIV V2 ( AA 183–197 ) inhibits SIV gp120 binding to α4β7 , and SIV mAbs whose epitopes overlap this peptide also inhibit α4β7 adhesion . One potential explanation for the conservation of α4β7 -reactivity between HIV and SIV , despite the divergent V2 sequences represented by peptides H43 and S46 ( < 50% sequence similarity ) is that the secondary structures of these two peptides share common features [62] . To address this possibility , we screened SIV mAbs for cross-reactivity with HIV V2 and found that ITS03 , one of the SIV mAbs that blocked α4β7 binding most effectively , reacted with relatively high-affinity to HIV cV2 92TH023 ( KD ( nM ) ) 0 . 105 ) ( Fig 8C ) . This cross-reactivity is consistent with shared secondary structure between the α4β7 -binding epitopes localized within the V2 regions of HIV and SIV gp120 . SIV infected macaques treated with a combination of ART and a recombinant rhesus anti α4β7 -mAb ( Rh-anti-α4β7 ) were able to durably control viremia at relatively low levels following treatment interruption [37] . Although these animals failed to mount neutralizing antibody responses , we noted that following withdrawal of ART , 8/8 generated anti-V2 antibody responses , while only 3/7 control animals administered normal IgG and ART , generated similar responses . Mapping studies indicated that this response was strongly focused on the region of V2 corresponding to S43 , the region of V2 that includes the NCI09 and ITS03 epitopes . These V2 responses persisted for at least 50 weeks , despite low plasma viremia [37] . Because NCI09 and ITS03 block adhesion to gp120 , we asked whether these polyclonal antibody responses included antibodies that could also inhibit α4β7 adhesion to gp120 . Serum IgG was purified from three ART + anti- α4β7 mAb -treated animals , RLn12 , RDa15 , and RId14 , by protein G affinity chromatography , and along with normal rhesus macaque IgG was evaluated in an α4β7 adhesion assay . While normal rhesus IgG showed minimal inhibition of α4β7-adhesion to SIVmac766 gp120 , the sera from all three ART + α4β7 treated macaques inhibited adhesion in a dose dependent manner ( Fig 9A ) . We then asked whether these sera contained antibodies with specificities similar to the SIV V2 mAbs ( described above in Fig 8 ) that inhibited α4β7 adhesion to gp120 . Using an SPR -based assay , SIVmac766 gp120 coated surfaces were preincubated with SIV V2 mAbs ITS03 , ITS09 . 01 , ITS12 . 01 or NCI09 . A surface preincubated with RLn12 serum served as a positive control . Surfaces were then reacted with RLn12 serum . As expected , pre-bound RLn12 serum inhibited RLn12 binding ( ~70% reduction ) ( Fig 9B ) . ITS03 , ITS09 . 01 and NCI09 -mediated similar levels of inhibition . ITS12 . 01 , which showed minimal inhibition of α4β7 adhesion ( Fig 8B ) , failed to inhibit RLn12 binding ( Fig 9B ) . We conclude that antibody responses generated in SIV infected macaques treated with ART and Rh-anti-α4β7 included V2 antibodies that target the α4β7 binding epitope of SIV V2 . In this study we report that the way in which α4β7 interacts with the V2 region of gp120 shares key features with the interaction between α4β7 and its natural ligand , MAdCAM . This apparent mimicry may have important implications in HIV pathogenesis , particularly in regard to the role of the gut in the development of HIV disease . It may also impact anti-V2 loop immune responses in both infected and vaccinated subjects . One consequence of this mimicry is that antagonists developed to treat IBD interfere with V2-α4β7 interactions . The V2 domains of HIV and SIV gp120 vary in both length and sequence identity . Yet we find that V2s from three subtypes of HIV , as well as a V2 from SIV retain the capacity to bind to α4β7 , suggesting that this interaction is a general property across HIV subtypes . Other studies failed to detect a specific interaction between α4β7 and gp120 [13 , 14] . An explanation for this discrepancy likely reflects two variables . First , as we and others have shown , the addition of glycans can reduce the interaction between recombinant gp120 with α4β7 [16 , 58] . It is likely that excess amounts of complex carbohydrate and sialic acid moieties that are added to gp120s expressed in cell lines contribute to this inhibitory effect . Of note , we demonstrated that highly purified ( >95% ) cyclic V2 loop peptides that lack glycans bind α4β7 with high affinity in both SPR -based assays and in a cell -based adhesion assay . Removal of glycans in order to observe α4β7 reactivity in not an absolute requirement insofar as we were able to capture virions derived from primary CD4+ T cells with α4β7 coated nanoparticles . The second variable that may influence the sensitivity of α4β7 binding assays involves the expression level , and state of α4β7 on cell surfaces . Many integrins rely on complex avidity effects and clustering in order to engage ligands . It is likely that the surface density of α4β7 plays a key role in its interaction with gp120 . The specific affinity of α4β7 for gp120 is comparable to that of MAdCAM . Among the 24 integrins expressed in humans , the α4β7 heterodimer is distinct in its ability to mediate both lymphocyte rolling , and firm adhesion , which reflects the highly specialized nature of MAdCAM-α4β7 interactions . These two functions are achieved by dynamic changes in the overall structure of the heterodimer . In this regard , it is notable that the V2 region of gp120 , despite its variable sequence , is able to mimic the binding of α4β7 to MAdCAM . The evidence for mimicry comes from two observations . First , the manner in which V2 depends upon divalent cations to engage α4β7 tracks closely with the way cations are used by MAdCAM . To bind α4β7 , MAdCAM utilizes divalent cations . A Mg++ ion sits in the MIDAS of β7 and coordinates with an Asp in the ligand ( Asp42 in MAdCAM ) . Mn++ occupies the MIDAS in a more stable way , so that replacing Mg++ with Mn++ results in an apparent increase in affinity . This pattern holds for gp120 V2 , indicating that V2 appears to interact with the different conformations of α4β7 in the same way that MAdCAM does . Importantly , the conformational state ( inactive , intermediate or active ) of α4β7 is responsive to both intracellular and external cues that are linked to cellular signals generated during inflammatory responses [63 , 64] . The ability of V2 to discriminate between different forms of α4β7 provides a mechanism to distinguish between different subsets of lymphocytes , including those with high potential to home to GALT . Given the propensity of HIV to replicate in GALT , it is tempting to link the preferential infection and depletion of α4β7high CD4+ T cells in the very early stages of infection [8 , 32 , 65] , with V2-α4β7 interactions . However , such a link has not yet been established . The second line of evidence that supports the proposition that V2 mimics the binding characteristics of MAdCAM comes from our demonstration that α4β7 antagonists that were developed to block binding to MAdCAM , also block binding to V2 . The incidence and prevalence of IBD is increasing throughout the world [66] , and consequently there has been a concerted effort to develop effective treatments , including drugs that target α4β7 . To this end , detailed structural characterizations of both MAdCAM and α4β7 have been employed in the rational design of small molecule LDV mimetics [67] . These mimetics bind with precision to the MAdCAM binding site on α4β7 , which lies within a ~ 40 Å long , 10 Å deep groove formed by the α4-β7 interface ( Fig 2A ) [23] . By showing that one of these mimetics competes with V2 we conclude that the aliphatic amino acid-Asp motifs conserved in both HIV and SIV fit into this groove and engage α4β7 in a way that , at least partially , mimics the Leu41-Asp42 encoded in the MAdCAM CC' loop of IgSF domain 1 ( Figs 2B and 6G ) [52 , 68] . These results suggest that the carboxy-terminus of V2 and this IgSF domain 1 CC' loop can adopt similar conformations . Evidence for the conserved nature of this structure comes from our observation that one SIV V2 mAb , ITS03 , whose epitope maps close to the α4β7 binding site , blocks α4β7 adhesion to V2 and also cross-reacts with an HIV subtype A/E V2 . This raises the intriguing possibility that , with additional screening , one might identify a V2 mAb that cross-reacts with MAdCAM . Indeed , other regions of the HIV envelope have shown to mimic “self” epitopes [69] . Vedolizumab is a unique α4β7 antagonist . It binds exclusively to the SDL of β7 in the context of α4β7; however , structural constraints preclude it from binding to αEβ7 [23] . It inhibits V2-α4β7 adhesion , which further supports the idea that V2 mimics MAdCAM . However , the mechanism by which vedolizumab , and its parent mAb , Act-1 [70] , interfere with MAdCAM binding is less well defined than that for LDV mimetics . The key to understanding this mechanism is understanding the role of MAdCAM IgSF domain 2 . Mutagenesis or deletion of the charged C'-E loop of MAdCAM IgSF domain 2 abrogates α4β7 binding [49 , 50] . Docking experiments carried out in silico indicate that this loop comes in close proximity to the β7 SDL and may contact α4β7 directly ( Fig 2A and 2B ) . By binding to the SDL , vedolizumab is likely to prevent direct interactions with the DE loop and/or sterically interfere with interactions between IgSF domain 1 and the binding groove at the interface between α4 and β7 . As such , the α4β7 binding footprint on MAdCAM spans two IgSF domains , and interactions with two loops that are separated by ~18-36Å that are both involved in binding to α4β7 . This raises a question regarding our demonstration that vedolizumab blocks adhesion between V2 and α4β7 . We show that a limited region within a ~40 AA cV2 peptide appears to engage the binding groove . The limited size of these cV2 peptides suggests that simultaneous interactions between these peptides and both the binding groove and the SDL are unlikely . Thus , the mechanism by which vedolizumab blocks V2 binding remains unclear and requires further investigation . Although the evidence for molecular mimicry outlined above is strong , the extent of this mimicry is likely to be limited . Because the interaction between MAdCAM and α4β7 facilitates both rolling and firm adhesion of lymphocytes along the endothelium , it encompasses a high degree of complexity that may not be entirely reflected in the interaction between V2 and α4β7 . Moreover , the region of V2 involved in binding to α4β7 is variable which makes it unlikely that it could retain the functional complexity inherent in the interactions that occur between MAdCAM and α4β7 . We found that certain weakly neutralizing V2 mAbs elicited from both infection ( CAP228-16H ) and vaccination ( CH58 ) could inhibit V2 -mediated adhesion to α4β7 . In this regard , there is a growing interest among HIV vaccine researchers regarding the potential utility of weakly neutralizing , or “functional” antibodies . This stems in large measure from the results of the RV144 Phase III vaccine trial in Thailand , where risk of acquisition was found to correlate inversely with weakly neutralizing V1/V2 antibodies [71–75] . Subsequently , a molecular sieve analysis of viral quasi-species in vaccinated individuals who became infected showed that residues at positions 169 and 181 within the V2 region were subject to immune pressure around the time of infection ( Fig 3B ) [76] . These residues fall around the binding epitopes for both CH58 and α4β7 . Follow-up studies suggested that antibody effector functions might contribute to the observed RV144 risk-reduction [77 , 78] . Our finding that mAb CH58 , which was derived from an RV144 vaccine recipient , block V2 -mediated adhesion to α4β7 raise the possibility that antibody activities distinct from both neutralization and Fc -mediated effector functions might contribute to the efficacy of HIV vaccines . Supporting this concept , we found that mAb NCI09 , which was derived from a macaque administered an SIVmac251-based vaccine designed to mimic the RV144 vaccine [79] , also blocked V2 -mediated adhesion to α4β7 . As in RV144 , reduced risk of infection in animals administered this vaccine was correlated with weakly neutralizing V2 antibodies [79] . A more complete description of this antibody and its activities are described elsewhere ( Franchini et al . , in preparation ) . How weakly neutralizing antibodies that block V2 -mediated adhesion to α4β7 might contribute to reduced risk of infection is unknown . We previously reported that the V2 region of gp120 can deliver cellular signals through α4β7 [9] . In this regard , integrins including α4β7 are key components in integrin associated complexes that are able to modulate biochemical pathways and reorganize both cell-surface receptors and the actin-cytoskeleton [63 , 80] . In addition , α4β7 can deliver costimulatory signals to CD4+ T cells that impact cell activation , proliferation and apoptosis [81 , 82] . We recently reported that MAdCAM -mediated costimulation supports HIV replication in α4β7high CD4+ T cells [19] . Whether V2 signaling through α4β7 can similarly facilitate HIV replication requires additional investigation . Such information will help determine whether this type of signaling could facilitate HIV transmission/replication and whether antibodies that interfere with this signal can reduce the risk of infection . The nature of the epitopes recognized by mAbs that inhibit the interaction between V2 and α4β7 should help us identify the context in which these interactions occur in vivo . It is noteworthy that these weakly neutralizing mAbs recognize an epitope that is structurally distinct from the β- strand that is presented on the closed BG505 SOSIP trimer . As such , broadly neutralizing mAbs that do recognize epitopes on the closed BG505 SOSIP trimer are unlikely to be effective inhibitors of V2-α4β7 interactions . Instead , we show that a unique α4β7 -reactive conformation of V2 is formed when it is relieved from constraints mediated by other domains of gp120/41 . Although the context in which V2 engages α4β7 remains to be determined , evidence is accumulating that α4β7 -expressing cells play an important role in the early stages of HIV/SIV infection . The recent demonstration by Sivro and colleagues that α4β7high CD4+ T cells are preferentially depleted from gut tissues as early as Fiebig I/II provides strong evidence that these cells serve as prime early targets for infection following transmission , an observation that is consistent with our demonstration that an α4β7 mAb protects macaques from vaginal challenge [36] . However , we cannot exclude the possibility that α4β7 plays a broader role in HIV pathogenesis . When we combined this same α4β7 mAb with ART , SIV infected animals were able to control viremia in a sustained way [37] . In trying to identify the underlying mechanism of control we reported , among other findings , that each of the controlling animals generated anti-V2 specific antibody responses that mapped to the region of SIV V2 that corresponds to the CH58 epitope in HIV V2 . In this report , we show that these serum V2 antibodies can block V2 -mediated adhesion to α4β7 . This finding underscores the need to further explore the role of V2 α4β7 interactions in HIV pathogenesis . In conclusion , the way that the V2 region of gp120 engages α4β7 shares key features with the way that MAdCAM , a receptor expressed primarily in gut tissues , engages α4β7 . One consequence of this apparent mimicry is that antagonists developed to treat IBD by interfering with the interaction between MAdCAM and α4β7 also interfere with V2-α4β7 interactions . The nature of the epitope in V2 that engages α4β7 appears to involve a structure that is not present in a recombinant SOSIP trimer designed to mimic the closed trimeric spike that is the target of most broadly neutralizing antibodies . Antibodies that target this epitope and block V2-α4β7 interactions are not themselves broadly neutralizing , although the structure that they recognize is conserved across clades and is present in SIV V2 . These findings suggest that mimicry of MAdCAM-α4β7 interactions by V2 may influence early events in HIV infection and replication in GALT . Generation of mAb Mk16C2 was approved and carried out under animal use protocol A-1896 by the Institutional Animal Care Committee ( IACUC ) of the University of Massachusetts Medical School . The University of Massachusetts Medical School is fully accredited by Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) and has an Animal Welfare Assurance on file with the Office of Laboratory Animal Welfare ( OLAW ) . Assurance number: A-3306-01 . RPMI8866 cells , a human B lymphoma cell line that constitutively expresses α4β7 was purchased from Sigma-Aldrich . Cells were maintained in RPMI-1640 ( Lonza ) containing 10% heat inactivated fetal bovine serum ( Gibco ) , 2% penicillin/streptomycin/glutamine ( Gemini Bio-Products ) and 0 . 1% retinoic acid ( RA ) . Cells were cultured for a minimum of 7 days prior to use in adhesion assays . RA was obtained from Sigma Chemical . CH58 , human mAb isolated from RV144 vaccinated individuals , and VRC01 were provided by the NIAID AIDS reagent program . CAP228-16H mAb was generated in the laboratory of Dr . Lynn Morris ( CAPRISA ) [57] . The 830A mAb was provided by Dr . Susan Zolla-Pazner ( Mt . Sinai Medical School ) [42] . Rabbit Mk16C2 mAb ( provided by Dr . Shan Lu , University of Massachusetts Medical School ) was isolated from a rabbit that received a gp120-JRFL DNA prime- protein boost immunization , using V2-peptide specific single B cell sorting , and produced by transfection of cloned Ig genes in 293F cells at the University of Massachusetts Medical School [83] . The SIV mAbs ITS03 , ITS09 . 01 , ITS12 . 01 , ITS13 , and ITS41 were provided by Dr . Rosemarie Mason , NIAID VRC [61] . NCI09 was provided by Genoveffa Franchini and produced in the laboratory of Rosemarie Mason in a manner identical to mAbs mAbs ITS03 , ITS09 . 01 , ITS12 . 01 , ITS13 , and ITS41 [61] . Rhesus macaque serum from animals RLn12 , RDa15 , RId14 was provided by Dr . Aftab Ansari ( Emory University ) [37] , was obtained after 72 weeks post-infection , and more than 40 weeks after the last infusion of anti-α4β7 mAb . Serum was purified by protein G column chromatography and dialyzed into HBS . Cyclic V2 peptides with > 90% purity and having an amino-terminal biotin , derived from 92TH023 , BG505 and C06980v0c22 were supplied by JPT Peptide Technologies . Linear HIV and SIV 15 amino acid peptides were obtained from the NIAID AIDS Reagent Repository , or from Biopeptide Co . and supplied at >90% purity . Scaffolds V1/V2 92TH023-1FD6 and V1/V2 92TH023-Tag were constructed , expressed and purified as described elsewhere [42] . CHO cell derived A244 gp120 ( Lot 26539–1 ) was provided by Global Solutions for Infectious Diseases ( San Francisco , CA ) . Purification employed an anti-gD immunoaffinity resin , followed by both cation and anion exchange chromatography steps . Purity was estimated at 97 . 1% . CHO cell derived SIVmac766 gp120 was provided by Advanced Biotechnologies Laboratories . BG505 SOISP trimer was generously provided by Dr . Paolo Lusso , LIR/NIAID . Vedolizumab was obtained from the NIH Clinical Center Pharmacy Department . Human integrin α4 mAb 2B4 , MAdCAM-Ig , soluble α4β7 and α4β1 were obtained from R&D Systems . The LDV mimetic ELN-475772 was provided by ELAN Pharmaceuticals [84] . Conjugation of HRP to mAbs was carried out using a LYNX HRP conjugation kit obtained from Bio-Rad , using the manufacturer’s instructions . Prior to use purified gp120 and gp140 proteins utilized in the α4β7 adhesion assays were first treated with a deglycosylation protocol [58] . Purified proteins were treated with 500U of PNGase F ( NEB ) per 20 μg of protein under non-denaturing conditions ( 1X GlycoBuffer 2 ( NEB ) , 5mM DTT , and water ) at 37°C for 3 hours . Experiments were performed using a Biacore 3000 instrument ( GE Life Sciences ) using CM4 or CM5 sensor chips . The data were evaluated using BIAevaluation 4 . 1 software ( GE Life Sciences ) . The chip surface was activated by injecting 35 μl of a 1:1 mixture of 0 . 05 M N-hydroxysuccinimide and 0 . 2 M N-ethyl-N- ( dimethylaminopropyl ) carbodiimide at 5 μl/min . NeutrAvidin , HIV gp120 or Hu-MAdCAM-Ig ( R&D Systems ) at concentrations of 5 μg/ml in 10mM NaOAc , pH 4 . 5 , were immobilized to approximately 750 resonance units ( RU ) . After the proteins were immobilized to the desired densities , unreacted sites on each surface were blocked with 35 μl of 1 M Tris-HCl ( pH 8 . 0 ) . Biotinylated cyclic V2 peptides ( 1 μg/ml in 20 mM Tris-HCl , pH 8 . 0 ) were bound to the NeutrAvidin surfaces to densities of approximately 250–300 RU . One surface was activated and blocked without ligand to act as a control surface for non-specific binding of the soluble ligand . Any binding was subsequently subtracted from the remaining surfaces . Running buffer was HBS ( pH 7 . 4 ) , 0 . 01 mM CaCl2 , either 1 mM MgCl2 or MnCl2 , 0 . 005% Tween-20 , 0 . 05% soluble carboxymethyl-dextran . Binding experiments were carried out at a flow rate of 25 μl/min at 25°C . After a 2 min injection , the surface was washed for an additional 2 min in running buffer to follow dissociation of the bound ligand from the surface . The surfaces were regenerated by multiple injections of 4 . 5 M MgCl2 at a flow rate of 100 μl/min . Inhibition of α4β7 or anti-V2 loop antibodies by linear V2-loop peptides was carried out by pre-incubating the proteins with the peptides in running buffer at the indicated concentrations for 2 hours at room temperature prior to passing them over the prepared surfaces as described above . Antibodies were diluted to the indicated concentrations in running buffer prior to being sequentially passed over the surface bound cyclic peptides as described above . The resulting sensorgram series were analyzed using the BiaEvaluation 4 . 1 software ( GE Life Sciences ) and fitted using a 1:1 Langmuir binding model to determine the kinetic rate and affinity constants . The Fab fragment of rabbit mAb Mk16C2 ( provided by Dr . Shan Lu , Univ . of Massachusetts Medical School ) was prepared by papain digestion as described ( PMID: 19913488 and 20622876 ) . Briefly , the IgG molecule was mixed with papain ( Worthington , Lakewood , NJ ) at a 20:1 molar ratio in 100 mM Tris ( pH 6 . 8 ) with 1 mM cysteine hydrochloride and 4 mM EDTA . The mixture was incubated for 1 hour at 37°C and the reaction was stopped by 10 mM iodoacetamide . The Fab fragment was separated from the Fc fragment and the undigested IgG by a protein A column and further purified by size exclusion chromatography . The Fab fragment was then concentrated to about 10 mg/ml for crystallization . The 15mer V2ConB peptide ( RDKVQKEYALFYKLD ) was dissolved in water and mixed with Fab of rabbit mAb Mk16C2 in excess at a 10:1 molar ratio . Crystallizations conditions were screened and optimized using the vapor diffusion hanging drop method . Well-diffracted crystals of Mk16C2 Fab/V2ConB complex were obtained with a well solution of 23% polyethylene glycol 3350 , 0 . 2 M LiCL , 0 . 1 M 2-ethanesulfonic acid ( MES ) pH 6 . 5 , and soaked briefly in the crystallization solution with an additional 20% glycerol before being flash frozen in liquid nitrogen . X-ray diffraction data sets were collected at the synchrotron beamline 14–1 of Stanford Synchrotron Radiation Lightsource ( SSRL ) of Stanford Linear Accelerator Center ( SLAC ) National Accelerator Laboratory . All data sets were processed using the XDS ( PMID: 20124692 ) , and structures determined by molecular replacement using another rabbit mAb R56 Fab structure ( PDB ID 4JO1 ) as the initial model . Cycles of refinement for each model were carried out in COOT ( PMID: 15572765 ) and Phenix ( PMID: 20124702 ) . Final structural analyses were carried out using ICM and figures were generated using PyMOL ( pymol . org ) and ICM ( www . molsoft . com ) . The antigen-antibody interactions described in Fig 3A are calculated by PDBePISA ( EMBL-EBI ) . Coordinate and structure factor of the complex have been deposited in the Protein Data Bank under PDB ID 6CEZ . The binding of α4β7 expressed by the RPMI8866 cell line to MAdCAM , HIV A244 gp120 , HIV cyclic V2 peptides , and SIVmac766 gp120 gp120 in the absence or presence of vedolizumab or 2B4 or ELN-475772 to MAdCAM , HIV A244 gp120 were analyzed by an adhesion assay ( adapted from KK Peachman et al . , [11] ( S2A Fig ) . This assay was modified by culturing RPMI8866 cells in media containing 1μM RA for at least 7 days prior to use in adhesion assays . Inclusion of RA increases adhesion to cV2 peptides , gp120 and MAdCAM ( S2C Fig ) . Briefly , triplicate wells of a 96-well flat-bottom polypropylene plate ( Greiner Bio-One ) were coated overnight at 4°C with 100 μl of 2 μg/ml of MAdCAM-1 ( R & D Systems ) or 100 μl of 2 μg/ml NeutrAvidin or 100 μl of 0 . 5–2 . 0 μg of deglycosylated SIV and HIV gp120 diluted in 50mM bicarbonate buffer , pH 9 . 6 . The NeutrAvidin-coated plates were then incubated with biotinylated cyclic V2 peptides ( 5 μg/ml in bicarbonate buffer ) for 1 hour at 37°C . The solution from the plates was discarded and the plates were then blocked with blocking buffer ( 25mM Tris , 2 . 7mM potassium chloride , 150 mM sodium chloride , 0 . 5% BSA , 4mM manganese chloride , pH 7 . 2 ) for 1 hour at 37°C . The solution was discarded and plates were manually washed 4 times with blocking buffer . After blocking and washing the plate , RPMI8866 cells in a volume of 50 μl/well were pre-incubated for 40 min at 37°C with sample buffer in the absence or presence of 10 μg/ml of vedolizumab ( α4β7 mAb ) or 2B4 ( α4 mAb ) or ELN-475772 ( α4β7/α4β1 dual inhibitor ) . Plates were then incubated with 50 μl/well of 2x105 RPMI8866 cells at 37°C ( 5% CO2 ) for 1 hour , washed 5 times with PBS followed by the addition of 100 μl of RPMI-1640 containing 1% FBS , 1% pen/strep/glutamine , 25mM HEPES with 10 μl/well of AlamarBlue dye . Fluorescence ( excitation 560 nm and emission 590 nm ) was measured immediately after the addition of the AlamarBlue dye for 8 hours . 92TH023 virions were captured with 15 nm magnetic nanoparticles ( MNPs ) coupled to α4β7 or 2G12 , CH58 , PG9 , and 830A mAbs as previously described [85] . Briefly , carboxyl-terminated iron oxide nanoparticles ( Ocean Nanotech , San Diego ) via two step carbodiimide reaction were coated with 1mg of anti-gp120 mAbs or recombinant soluble α4β7 according to manufacturer’s protocol . Virus preparations were derived from primary CD4+ T cells infected with an infectious molecular clone derived from 92TH023 , using standard conditions . To capture virions , MNPs coated with mAbs ( 3 . 9 x1012 ) in 60μl were incubated with viral preparation ( 33 ng/ml based on p24 content ) for 1hour at 37°C . Captured virions were separated on MACS magnetic columns attached to an OctoMacs magnet ( Miltenyi Biotech ) washed 4 times with 600 μl wash buffer ( 0 . 5% bovine serum albumin , 2mM EDTA in PBS ) , eluted in 200 μl PBS and analyzed on Luminex X200 for p24 content using a dynamic immunofluorescent cytometric bead assay [60] . In experiments with α4β7-MNPs , virions were incubated in the absence or presence of increasing amounts ( 1 . 25nM , 12 . 5nM , 125nM ) of the α4β7 inhibitor ELN-475772 . Incubation and washing were performed in complete medium with 1mM MnCl2 . Triplicate wells of a 96-well flat-bottom polypropylene plate ( Greiner Bio-One ) were coated with biotinylated cyclic V2 peptides , deglycosylated SIV and HIV gp120 as described above . After blocking and washing , plates were incubated with 20–100 μg/ml of the designated anti-V2 mAbs or 2 . 5–10 . 0 μg of protein G purified IgG from sera drawn from SIV infected rhesus macaques ( RLn12 , RDa15 , RId14 ( provided by Dr . Aftab Ansari , Emory University School of Medicine ) ) in sample buffer ( 25 mM Tris , 2 . 7 mM KCl , 150 mM NaCl , 4 mM manganese chloride , 1% fetal bovine sera , pH 7 . 2 ) for 1 hour at 37°C . RPMI8866 cells were pre-incubated for 40 min at 37°C with sample buffer . Plates were then incubated with 50 μl/well of 2x105 cells at 37°C ( 5% CO2 ) for 1 hour , washed 5 times with PBS followed by the addition of 100 μl of RPMI-1640 containing 1% FBS , 1% pen/strep/glutamine , 25mM HEPES with 10 μl/well of AlamarBlue dye ( Invitrogen ) . Following the addition of AlamarBlue dye ( excitation 560 nm and emission 590 nm ) fluorescence was measured for a period of 8 hours . 92TH023 V1/V2 was cloned into both 1FD6 ( constrained ) and tag ( unconstrained ) scaffolds . Plates were then coated with 0 . 5 , 1 . 0 , and 2 . 0 μg of either V1/V2 92TH023 1FD6 or V1/V2 92TH023 tag scaffolds followed by addition of RPMI8866 cells . The plates were washed and 100 μl of media and 10 μl of AlamarBlue dye was added to each well as described above . Fluorescence ( excitation 560 nm and emission 590 nm ) was measured immediately after the addition of the AlamarBlue dye . Briefly , Corning Costar 96-Well plates were coated with NeutrAvidin at 4°C overnight . Wells were washed six times with wash buffer ( water , 1 mM MnCl2 , and 1X plate wash buffer ) using Microplate Washer ELx50 , ( Bio Tek Instruments ) , and then blocked with blocking buffer ( HBS , 5% bovine serum albumin ) for 1 hour at room temperature . Plates were coated with 1 μg/ml biotinylated cyclic V2 peptides in binding buffer ( HBS , 1 mM MnCl2 ) for 1 hour at room temperature . The plates were washed six times with wash buffer and HRP-conjugated anti-V2 mAb was added to wells for 1 hour at room temperature . After washing , 100 μl/well of substrate was added for 10 min to develop color at room temperature in the dark . Plates were read at OD450 nm using an EnSpire Multimode Plate Reader ( PerkinElmer ) . 10 , 20 , 50 , 80 and 160 ng of the CH58 mAbs in binding buffer ( HBS , 1 mM MnCl2 ) were incubated with 50 ng of HRP-conjugated 830A mAb and then added to a plate coated with 100 ng of cV2 92TH023 . Substrate was added using conditions specified by the manufacturer . Plates were read at OD450 nm using an EnSpire Multimode Plate Reader , PerkinElmer . RPMI8866 cell line cultured in the presence or absence of RA were seeded at 1×105/well in Poly-d-lysine coated glass bottom dishes ( MatTek ) with cover glasses and incubated overnight at 37°C . The cells were fixed with 2% paraformaldehyde ( PFA ) , blocked with 1% BSA , and stained with an anti-β7 PE mAb ( BD Biosciences ) or an IgG2a-PE isotype control mAb ( R & D Systems ) . Stained RPMI8866 cells were microscopically analyzed using a Leica SP8 confocal microscope ( Leica Microsystem , Inc . ) and images were processed with Leica LAS AF software ( Leica Microsystem , Inc . ) and Imaris software v . 9 . 0 . 1 64x ( Bitplane AG ) .	HIV is gut-tropic . Disruption of GALT plays an important role in HIV -mediated immune dysfunction . α4β7 is a receptor that facilitates homing of lymphocytes to GALT . α4β7high CD4+ T cells are early targets of HIV infection . The HIV envelope protein gp120 binds to α4β7; however , a link between this interaction and the preferential infection of α4β7high CD4+ T cells has not been established . Here we report an apparent gp120 mimicry of MAdCAM , a natural ligand of α4β7 . That the HIV envelope protein mimics an adhesion receptor expressed primarily in gut tissues supports the concept of a central role for GALT in HIV pathogenesis . Antibodies derived from vaccination and infection were evaluated for their capacity to inhibit gp120-α4β7 interactions . Antibodies that inhibit binding target a region of the V2 domain that has been linked with reduced risk of acquisition in the RV144 vaccine trial . These antibodies recognize epitopes that do not appear on the closed gp120/41 trimer . Instead , they recognize structures that form when V2 is allowed to fold in a less constrained way . We show that such epitopes can appear on virions . Thus , an alternative structure of the V2 domain may facilitate the trafficking of HIV to GALT .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "blood", "cells", "medicine", "and", "health", "sciences", "chemical", "characterization", "immune", "physiology", "immune", "cells", "pathology", "and", "laboratory", "medicine", "pathogens", "immunology", "ions", "microbiology", "vertebrates", "viral", "structure", "animals", "mammals", "retroviruses", "viruses", "immunodeficiency", "viruses", "primates", "rna", "viruses", "antibodies", "cations", "old", "world", "monkeys", "research", "and", "analysis", "methods", "physical", "chemistry", "immune", "system", "proteins", "white", "blood", "cells", "monkeys", "animal", "cells", "proteins", "medical", "microbiology", "hiv", "microbial", "pathogens", "t", "cells", "chemistry", "binding", "analysis", "virions", "siv", "macaque", "biochemistry", "eukaryota", "cell", "biology", "virology", "viral", "pathogens", "physiology", "biology", "and", "life", "sciences", "cellular", "types", "physical", "sciences", "lentivirus", "amniotes", "organisms" ]	2018	Select gp120 V2 domain specific antibodies derived from HIV and SIV infection and vaccination inhibit gp120 binding to α4β7
Prediction of antibiotic resistance phenotypes from whole genome sequencing data by machine learning methods has been proposed as a promising platform for the development of sequence-based diagnostics . However , there has been no systematic evaluation of factors that may influence performance of such models , how they might apply to and vary across clinical populations , and what the implications might be in the clinical setting . Here , we performed a meta-analysis of seven large Neisseria gonorrhoeae datasets , as well as Klebsiella pneumoniae and Acinetobacter baumannii datasets , with whole genome sequence data and antibiotic susceptibility phenotypes using set covering machine classification , random forest classification , and random forest regression models to predict resistance phenotypes from genotype . We demonstrate how model performance varies by drug , dataset , resistance metric , and species , reflecting the complexities of generating clinically relevant conclusions from machine learning-derived models . Our findings underscore the importance of incorporating relevant biological and epidemiological knowledge into model design and assessment and suggest that doing so can inform tailored modeling for individual drugs , pathogens , and clinical populations . We further suggest that continued comprehensive sampling and incorporation of up-to-date whole genome sequence data , resistance phenotypes , and treatment outcome data into model training will be crucial to the clinical utility and sustainability of machine learning-based molecular diagnostics . At least 700 , 000 deaths annually can be attributed to antimicrobial resistant ( AMR ) infections , and , without intervention , the annual AMR-associated mortality is estimated to climb to 10 million in the next 35 years [1] . As most patients are still treated based on empirical diagnosis rather than confirmation of the causal agent or its drug susceptibility profile , development of improved , rapid diagnostics enabling tailored therapy represents a clear actionable intervention [1] . The Cepheid GeneXpert MTB/RIF assay , for example , has been widely adopted for rapid point-of-care detection of Mycobacterium tuberculosis ( TB ) and rifampicin ( RIF ) resistance [2] , and the SpeeDx ResistancePlus GC assay used to detect both Neisseria gonorrhoeae and ciprofloxacin ( CIP ) susceptibility was recently approved for marketing as an in vitro diagnostic in Europe . Molecular assays offer improved speed compared to gold-standard phenotypic tests and are of particular interest because of their promise of high accuracy for the prediction of AMR phenotype based on genotype [2 , 3] . Approaches for predicting resistance phenotypes from genetic features include direct association ( i . e . , using the presence or absence of genetic variants known to be associated with resistance to infer a resistance phenotype ) and the application of predictive models derived from machine learning ( ML ) algorithms . Direct association approaches can offer simple , inexpensive , and often highly accurate resistance assays for some drugs/species [2] and may even provide more reliable predictions of resistance phenotype than phenotypic testing [4–6] . However , these approaches are limited by the availability of well-curated and up-to-date panels of resistance variants , as well as the diversity and complexity of resistance mechanisms . ML strategies can facilitate modeling of more complex , diverse , and/or under-characterized resistance mechanisms , thus outperforming direct association for many drugs/species [7–9] . With the increasing speed and decreasing cost of sequencing and computation , ML approaches can be applied to genome-wide feature sets [8 , 10–18] , ideally obviating the need for comprehensive a priori knowledge of resistance loci . While prediction of antibiotic resistance phenotypes from ML-derived models based on genomic features has become increasingly prominent as a promising diagnostic tool [8 , 11–15 , 17] , there has been no systematic evaluation of factors that may influence performance of such models and their implications in the clinical setting . The extent to which ML model accuracy varies by antibiotic is unclear , as is the impact of sampling bias on model performance . It is further unclear what the most relevant resistance metric ( i . e . , minimum inhibitory concentration [MIC] or categorical report of susceptibility ) for such a diagnostic might be and how amenable different species might be to genotype-to-phenotype modeling of antibiotic resistance . We used set covering machine ( SCM ) [19] and random forest ( RF ) [20] classification as well as RF regression algorithms to build and test predictive models with seven gonococcal datasets for which whole genome sequences ( WGS ) and ciprofloxacin ( CIP ) and azithromycin ( AZM ) MICs were available . AZM is currently part of the recommended treatment regimen for gonococcal infections , and with the development of resistance diagnostics , CIP may represent a viable treatment option [21–23] . While the majority of CIP resistance in gonococci can be attributed to gyrA mutations , AZM resistance is associated with more diverse and complex resistance mechanisms [23 , 24] , offering an opportunity to evaluate ML methods across drugs with distinct pathways to resistance . The range of datasets and sampling frames enables assessment of sampling bias on model reliability . Further , the availability of MICs , as well as distinct European Committee on Antibiotic Susceptibility Testing ( EUCAST ) and Clinical and Laboratory Standards Institute ( CLSI ) breakpoints , for these drugs allows for evaluation of predictive models based on different resistance metrics . Finally , extension of these analyses to Klebsiella pneumoniae and Acinetobacter baumannii datasets for which WGS and CIP MICs were available allows for assessment of model performance for the same drug in species with open pangenomes [25 , 26] , which may be more difficult to model given the increased genomic diversity and potential resistance mechanism diversity and complexity [27] . Our results demonstrate that using ML to predict antibiotic resistance phenotypes from WGS data yields variable results across drugs , datasets , resistance metrics , and species . While more comprehensive assessment of different methods will be required to build the most accurate and reliable models , we suggest that tailored modeling for individual drugs , species , and clinical populations may be necessary to successfully leverage these ML-based approaches as diagnostic tools . We further suggest that continuing surveillance , isolate collection , and reporting of WGS , MIC phenotypes , and treatment outcomes will be crucial to the sustainability of any such molecular diagnostics . Given the distinct MIC distributions and distinct pathways to resistance for CIP and AZM in gonococci , these two drugs enable evaluation of drug-specific performance of ML-based resistance prediction models . CIP MICs in surveys of clinical gonococcal isolates are bimodally distributed , with the majority of isolates having MICs well above or below the non-susceptibility ( NS ) breakpoints , while the majority of reported AZM MICs in gonococci are closer to the NS breakpoints ( https://mic . eucast . org/Eucast2 ) . These trends were recapitulated in the gonococcal isolates assessed here ( Fig 1A and 1B ) . Further , the vast majority of CIP resistance in gonococci observed to date is explained by mutations in gyrA and parC and has spread predominantly through clonal expansion , generally resulting in MICs ≥ 1 μg/mL [23 , 28] . In contrast , AZM resistance in gonococci has arisen many times de novo through multiple pathways , many of which remain under-characterized and are associated with lower-level resistance [23 , 28 , 29] . As expected , the GyrA S91F mutation alone predicts NS to CIP by both EUCAST and CLSI breakpoints in the aggregate gonococcal dataset assessed here with ≥98% sensitivity and ≥99% specificity ( S1 Table ) . AZM NS showed lower values for these metrics , indicating it was not as well explained by known resistance variants , with extensive contributions from uncharacterized mechanisms and/or multifactorial interactions ( S2 Table ) . We next trained and evaluated ML-based predictive models for CIP and AZM resistance in gonococci ( S3 Table ) . By all ML methods and breakpoints , CIP NS was predicted with significantly higher balanced accuracy ( bACC ) than AZM NS in the aggregate gonococcal dataset ( P < 0 . 0001 , Fig 1C and 1D , S4 and S5 Tables ) : CIP NS was predicted with mean bACC ≥93% across all methods , breakpoints , and datasets , whereas mean bACC for AZM NS classification ranged from 57% to 94% ( S4 and S5 Tables ) . Variation in model performance across antibiotics has been attributed to different proportions of susceptible ( S ) and NS isolates [7 , 14 , 15]; however , by the EUCAST breakpoints , the aggregate gonococcal dataset as well as some of the individual datasets had nearly identical proportions of CIP and AZM susceptible and non-susceptible isolates , demonstrating that variable representation of S and NS isolates alone cannot explain reduced performance of AZM models compared to CIP . We tested whether the poorer performance for AZM may be attributable to the large fraction of isolates with MICs around the breakpoint . Removing strains with AZM MICs that were ≤2 doubling dilutions of the NS breakpoints from the aggregate gonococcal dataset ( S6 Table ) yielded AZM MIC distributions similar to those of CIP ( S1A and S1B Fig ) . Analysis of this restricted dataset resulted in higher performance of SCM and RF AZM NS classifiers compared to those trained and tested on the full aggregate gonococcal dataset ( S1C Fig ) . However , bACC of AZM classifiers trained and tested on the restricted datasets was still significantly lower than bACC of the CIP NS classifiers ( P < 0 . 0001 and P < 0 . 003 for classifiers based on the EUCAST and CLSI breakpoints , respectively ) , suggesting that both MIC distribution and additional drug-specific factors can influence performance of resistance classifiers . The diversity of resistance mechanisms for AZM in gonococci offers an opportunity to evaluate the effects of sampling bias on model performance . The sampling frames for the seven gonococcal datasets ranged geographically from citywide to international and temporally from a single year to >20 years , and several datasets were enriched for AZM resistance [11 , 30] ( Table 1 ) . The distributions of both AZM MICs and known resistance mechanisms across datasets ( Fig 1B , S2 Table ) and the variable performance of AZM resistance models across datasets ( S5 Table ) suggest that AZM resistance mechanisms are differentially distributed across the sampled clinical populations . Further , the higher performance of many SCM and RF-based AZM classifiers on training data compared to test sets ( S5 Table ) suggests that potentially due to a lack of signal , AZM models are incorporating substantial noise or confounding factors , which may be population-specific . To assess the impact of sampling on model reliability , the performance of RF classifiers in prediction of AZM NS phenotypes were compared across multiple training and testing sets . These include classifiers trained on subsamples of isolates from a single dataset , classifiers trained on the aggregate gonococcal dataset , and classifiers trained on the aggregate gonococcal dataset excluding isolates from the same dataset as the testing set ( S6 Table ) . Given the low representation of AZM NS strains by the CLSI breakpoint in many datasets , these analyses were only performed using the EUCAST breakpoint . While it may be assumed that increased availability of paired genomic and phenotypic resistance data from a broader range of clinical populations will facilitate more accurate and reliable modeling [13] , our results demonstrate that in predicting AZM resistance phenotypes for isolates from most datasets ( with the exception of datasets 2 and 5 ) , performance of classifiers trained on the aggregate dataset was not significantly better than performance of classifiers trained only on isolates from the dataset from which the test isolates were derived ( P < 0 . 0001 and P = 0 . 002 for datasets 2 and 5 , respectively , P = 0 . 008 for dataset 3 , where the classifiers trained on the aggregate dataset had lower bACC than classifiers trained only on isolates from dataset 3 , and P > 0 . 234 for all other datasets , Fig 2A ) . Further , there was substantial variation in performance of models trained on the aggregate dataset across testing sets , with models achieving significantly higher bACC for strains from datasets 3 and 4 than for strains from dataset 2 ( P < 0 . 0009 , Fig 2A ) , perhaps reflecting enrichment for AZM NS in these former datasets ( Table 1 ) . Additionally , with the exception of dataset 5 , performance of AZM resistance classifiers trained only on isolates from the dataset from which the test isolates were derived was significantly higher than performance of classifiers trained on the aggregate dataset excluding isolates from the test dataset ( P = 0 . 537 for dataset 5 , P < 0 . 0005 for all other datasets , Fig 2A ) . Performance of RF classifiers trained and tested on dataset 2 was limited by low specificity , which was improved in models trained on the aggregate dataset ( Fig 2B ) . The low specificity achieved by RF classifiers trained and tested on this dataset is likely due to the low representation of S strains , most of which were within one doubling dilution of the NS breakpoint ( Fig 2C ) , and thus the more comprehensive representation of negative ( S ) data in the aggregate training set was associated with improved specificity . Conversely , performance of RF classifiers trained and tested on dataset 5 was more limited by low sensitivity , which was improved in models trained on the aggregate dataset ( Fig 2B ) . This dataset had a low representation of strains with high AZM MICs ( Fig 2D ) , and thus the more comprehensive representation of positive ( NS ) data in the aggregate training set was associated with improved sensitivity in predicting AZM NS for these strains . For both SCM and RF-C AZM resistance models across all datasets , there was a significant positive correlation between the ratio of model sensitivity to model specificity and the ratio of NS to S strains in the dataset ( Pearson r > 0 . 98 , P < 0 . 0001 [Pearson correlation] for both SCM and RF-C , S2A Fig ) . On the other hand , while representation of strains with higher AZM MICs was also observed in other datasets ( i . e . , datasets 1 , 6 , and 7 ) and was similarly reflected in the sensitivity-limited performance of RF classifiers trained and tested on these datasets ( S5 Table ) , AZM NS prediction accuracy for strains from these datasets was not improved by training classifiers on the aggregate dataset . Further , even after down-sampling two of the datasets with the most disparate MIC distributions , sample sizes , and model performance ( datasets 2 and 4 ) such that the number of strains and AZM MIC distributions were identical between the two datasets ( S2B Fig ) , there was still a significant difference in AZM NS prediction accuracy of models trained and tested on these different datasets ( S2C Fig , P < 0 . 004 ) . Together , these results demonstrate that resistance model performance may be strongly associated with the distributions of both resistance phenotypes and genetic features and thus can be highly population-specific . Gonococcal CIP and AZM MICs were dichotomized by both EUCAST and CLSI breakpoints to assess the impact of variation in MIC breakpoints on model performance . As the EUCAST and CLSI breakpoints for CIP in gonococci are within a single doubling dilution and the vast majority of isolates have much lower or higher CIP MICs ( Fig 1A ) , >99% of isolates in the aggregate dataset were consistently S or NS by both breakpoints . Of the 23 isolates with MICs between the two breakpoints , 18 had MICs derived from Etests of 0 . 032 μg/mL or 0 . 047 μg/mL , making their classification relative to the EUCAST breakpoint of 0 . 03 μg/mL ambiguous . In contrast , the EUCAST and CLSI breakpoints for AZM in gonococci are separated by two doubling dilutions , and for many isolates , the AZM MIC was within this range ( Fig 1B ) . As such , only 67% of isolates in the aggregate dataset were consistently S or NS by both breakpoints . CIP NS classifier performance was either identical or nearly identical for both breakpoints in the aggregate and most individual gonococcal datasets ( Fig 3A ) . In contrast , the bACC of AZM NS prediction by both SCM and RF classifiers based on the CLSI breakpoint was significantly higher than for those based on the EUCAST breakpoint across all gonococcal datasets assessed by both breakpoints ( P < 0 . 0001 , Fig 3B ) . To assess the performance of MIC prediction models relative to binary S/NS resistance phenotype classifiers , RF-mC and RF-R models were trained and evaluated for CIP and AZM MIC prediction in gonococci . Average exact match rates between predicted and phenotypic MICs ranged from 64–86% and 54–78% by RF-mC and RF-R , respectively , for CIP , and from 24–60% and 45–65% , respectively , for AZM ( S4 and S5 Tables ) . Average 1-tier accuracies ( the percentage of isolates with predicted MICs within one doubling dilution of phenotypic MICs ) were substantially higher but also varied widely across datasets and between the two MIC prediction methods ( ranging from 82%-96% and 76–87% by RF-mC and RF-R , respectively , for CIP , and from 73–94% and 73–83% , respectively , for AZM; S4 and S5 Tables ) . There was no consistent or significant relationship across the different datasets between MIC prediction accuracy ( exact match or 1-tier accuracy ) and bACC for either drug by either MIC prediction method ( Fig 3C–3F ) . Further , for both drugs by both breakpoints in the aggregate gonococcal dataset , binary RF-C models had equivalent or significantly higher bACC than RF-mC and RF-R MIC prediction models ( P > 0 . 175 for AZM NS by the CLSI breakpoint by RF-C compared to RF-mC or RF-R , P < 0 . 017 for all others , S4 and S5 Tables ) . Increasing genomic diversity , or an increasing ratio of genomic features ( e . g . , k-mers ) to observations ( e . g . , genomes ) , may present an additional challenge for ML-based prediction of antibiotic resistance [12] . To investigate ML-based antibiotic resistance prediction across species with different levels of genomic diversity , SCM and RF-C were used to model CIP NS in K . pneumoniae and A . baumannii , two species with genomic diversity ( i . e . , ratio of unique 31-mers to number of genomes ) several times that of gonococci ( Fig 4A and 4B ) . SCM classifiers trained on and used to predict CIP NS for K . pneumoniae achieved significantly lower accuracy than all of the gonococcal datasets ( P < 0 . 0001 , Fig 4C ) , while SCM classifiers trained on and used to predict CIP NS for A . baumannii achieved significantly lower accuracy than gonococcal datasets 3–5 and 7 ( P < 0 . 033 ) and roughly equivalent accuracy to gonococcal datasets 1–2 and 6 , as well as the aggregate gonococcal dataset ( P > 0 . 059 , Fig 4C ) . The performance of RF-C models was significantly lower for both K . pneumoniae and A . baumannii compared to all gonococcal datasets ( P < 0 . 0001 , Fig 4D ) . While the SCM classifiers for CIP NS in K . pneumoniae performed significantly better on the training sets than the testing sets ( S4 Table , P < 0 . 0001 ) , indicating that these models may be overfitted , there was no significant difference between RF-C model performance on training and testing sets for either K . pneumoniae or A . baumannii ( P > 0 . 194 ) , suggesting that overfitting alone cannot explain the variable classifier performance across different species . Down-sampling K . pneumoniae and A . baumannii to match the CIP MIC distributions of the gonococcal datasets was infeasible due to the narrow range of MICs tested for the former two species ( S7 Table ) . However , even after down-sampling to equalize the number of S and NS strains within each dataset ( S6 Table , S3A and S3B Fig ) , performance of K . pneumoniae and A . baumannii CIP NS classifiers was still significantly lower than that of gonococcal CIP NS classifiers , with the exception of SCM classifiers based on the down-sampled K . pneumoniae dataset , which performed roughly equivalently to SCM classifiers based on gonococcal datasets 2 and 6 ( P > 0 . 07 for the SCM classifiers based on the down-sampled K . pneumoniae dataset compared to SCM classifiers based on gonococcal datasets 2 and 6; P < 0 . 0004 for all other comparisons , S3C Fig ) . Direct association based on GyrA codon 83 mutations ( equivalent to codon 91 in gonococci ) alone predicted CIP NS in K . pneumoniae with 86% sensitivity and 99% specificity , and thus had a marginally higher bACC ( 92 . 5% ) than for the SCM classifiers and a substantially higher bACC than the RF classifiers . Similarly , for A . baumannii , GyrA codon 81 mutations ( equivalent to codon 91 in gonococci ) alone predicted CIP NS in with 97% sensitivity and 98% specificity , and thus with a roughly equivalent bACC ( 97 . 5% ) to the SCM classifiers and a substantially higher bACC than the RF classifiers . Genotype-based resistance diagnostics have largely focused more on evaluating the presence of resistance determinants and less on predicting the susceptibility profile of a given isolate [8] . However , in clinical settings where the empirical presumption is of resistance , prediction that an isolate is susceptible to an antibiotic may be more important in guiding treatment decisions . As such , the clinical utility of a genotype-based resistance diagnostic may be determined by its capacity to accurately predict susceptibility phenotype for multiple drugs . While variable performance of ML-based predictive models has been observed across different drugs [7 , 8 , 10 , 11 , 14 , 15] , it has often been attributed to dataset size and/or imbalance [7 , 14 , 15] . Further , while it is more difficult to predict resistance phenotypes from genotypes for drugs that are associated with unknown , multifactorial , and/or diverse resistance mechanisms than for drugs for which resistance can largely be attributed to a single variant [14 , 30] , this caveat has been presented specifically as a limitation of models based on known resistance loci in comparison to unbiased machine learning-based MIC prediction using genome-wide feature sets [14] . However , by comparing performance of predictive models based on genome-wide feature sets between CIP and AZM across multiple gonococcal datasets , we showed that even with relatively large and phenotypically balanced datasets , ML algorithms cannot necessarily be expected to successfully model complex and/or diverse resistance mechanisms , particularly given that the representation of these resistance mechanisms in training datasets is a priori unknown . As a high proportion of reported AZM MICs in gonococci are within 1–2 doubling dilutions of the NS breakpoints , it is possible that the inferior performance of AZM classifiers is partly attributable to errors and/or variations in MIC testing . However , given the noise of phenotypic MIC testing even with standardized protocols [31] , this may be an inherent limitation of NS classifiers when low-level resistance is common . Further , while we show that removing strains with MICs ≤2 doubling dilutions from the breakpoints improved AZM classifier performance compared to AZM models trained and tested on the full dataset , performance of AZM classifiers trained and tested on this restricted dataset was still significantly lower than that of CIP classifiers , suggesting that additional drug-specific factors , such resistance mechanism diversity and/or complexity , can constrain classifier performance . Sampling bias presents a substantial challenge in any predictive modeling , and sampling from limited patient demographics or during limited time periods may have considerable effects on the distributions of resistance phenotypes and resistance mechanisms [32 , 33] . For example , in TB , the RpoB I491F mutation that has been associated with failure of commercial RIF resistance diagnostic assays , including the GeneXpert MTB/RIF assay , reportedly accounted for <5% of TB RIF resistance in most countries , but , in Swaziland was found to be present in up to 30% of MDR-TB [34] . Further , as the focus with statistical classifiers is building models from feature sets that can accurately predict an outcome , rather than understanding the association between each of the features and the outcome , potential confounding effects from factors such as population structure [35–37] or correlations among resistance profiles of different drugs [13] are rarely considered . By comparing performance of AZM NS classifiers across multiple training and testing sets , we showed significant variation in performance of classifiers trained on a large and diverse global collection across testing sets from different sampling frames . In some cases of imbalanced datasets , models trained on datasets with a more comprehensive representation of resistance phenotypes improve prediction accuracy . Our results further demonstrate that the direction of dataset imbalance ( i . e . , the ratio of NS to S strains ) is significantly correlated with the direction of model performance ( i . e . , the ratio of sensitivity to specificity ) , suggesting that , for example , optimizing sensitivity of predictive models for drugs with low prevalence of NS strains may require substantial enrichment of NS strains and/or down-sampling of S strains . However , while differential classifier performance among different datasets may be partially attributable to differential MIC distributions , our results also show variable classifier performance between datasets even in the case of identical MIC distributions ( and sample size ) and further suggest that heavier sampling across more geographic regions cannot necessarily be expected to significantly improve model performance , as models trained on the aggregate global gonococcal dataset did not improve prediction accuracy for most datasets . This , together with decreased performance when excluding isolates from the dataset from which the isolates being tested were derived , suggests that factors such as population-specific resistance mechanisms , genetic divergence at resistance loci , and/or confounding effects may constrain model reliability across populations , particularly in the case of drugs like AZM with complex and/or diverse resistance mechanisms , where a substantial portion of the model may be overfit , or based on confounding factors or noise , rather than biologically-meaningful resistance variants . Further , it should be noted that MIC testing methods varied between some datasets ( and between strains within dataset 5 ) , and such variations may represent an additional confounding factor influencing classifier performance . Thus , both incorporation of methods to correct for potentially confounding factors , such as population structure , as have been introduced for genome-wide associate studies [35–37] , and increased availability of paired WGS and antibiotic susceptibility data produced by consistent standardized protocols may improve reliability of machine learning-based prediction of antibiotic resistance across different populations . While measurement of MICs is vital for surveillance and investigation of resistance mechanisms , resistance breakpoints that relate in vitro MIC measurements to expected treatment outcomes inform clinical decision-making . However , standard breakpoints for NS to a given drug in a given species are often informed less by treatment outcome data , but rather factors such as pharmacokinetics and MIC distributions that can fail to account for a variety of intra-host conditions that could influence drug efficacy [38–41] . Recent studies have shown that isolates that are classified as susceptible by standard breakpoints but have higher MICs are associated with a greater risk of treatment failure than isolates with lower MICs [42] . Further , resistance breakpoints and testing protocols can vary across different organizations , and thus incongruence across phenotypic information included in the training data may introduce additional sources of error in predictive modeling . By comparing performance of predictive models of CIP and AZM NS based on EUCAST and CLSI breakpoints , we demonstrated breakpoint-specific performance of models . For CIP , such breakpoint-specific performance is likely largely attributable to variations in MIC testing protocols and thus ambiguous classification of some strains by the EUCAST breakpoint . On the other hand , the substantially lower performance of all AZM models based on the EUCAST breakpoint compared to those based on the CLSI breakpoint suggests that many isolates with AZM MICs between the two breakpoints lack genetic signatures that contribute to high model performance . While the clinical relevance of AZM MICs between these two breakpoints in gonococci is unclear , these isolates may be more likely to be associated with AZM treatment failure than isolates with lower MICs , and thus evaluation of classifiers using only higher breakpoints may misrepresent their diagnostic value , particularly in the absence of sufficient treatment outcome data . Models that predict MICs provide more refined output than a binary classifier but generally achieve low rates of exact matches between phenotypic and predicted MICs and even fairly variable 1-tier accuracies [14 , 15 , 30] . Given the noise in phenotypic MIC testing [31] and the potential lack of discriminating genetic features between isolates with MICs separated by 1–2 doubling dilutions [14] , MIC prediction models may be unlikely to provide much better resolution than binary S/NS classifiers . Even if MIC predictions could provide additional resolution , the most important criterion of such a diagnostic would likely still be its ability to correctly predict resistance phenotypes relative to a clinically relevant breakpoint . Thus , performance of MIC prediction models with respect to breakpoints may be the biggest determinant of their diagnostic utility . By building MIC prediction models for CIP and AZM in gonococci , we observed low rates of exact matches between phenotypic and predicted MICs and variable 1-tier accuracies , with no relationship between 1-tier accuracy and categorical agreement ( i . e . , prediction accuracy relative to NS breakpoints ) . Further , binary classifiers performed equivalently or better than MIC prediction models . Bacterial species with high genomic diversity ( e . g . , open pangenomes ) present additional challenges to ML-based prediction of antibiotic resistance . Increased resistance mechanism complexity and greater inter-isolate variation in resistance mechanisms require more intensive sampling to capture a significant portion of the resistome [27] . On the technical side , even for heavily sampled species , when using whole genome feature sets , the number of genetic features ( e . g . , k-mers or SNPs ) will always be much larger than the number of observations ( isolates ) , increasing the risk of overfitting ( a situation that arises with so-called ‘fat data’ , [12] ) . This raises concern in species with open pangenomes , as the ratio of genetic features to the number of genomes is larger and the number of unique genetic features per number of genomes does not plateau . By comparing classifier performance in predicting CIP NS across gonococci , K . pneumoniae , and A . baumannii , we show that classifiers generally did not perform as well for species with open genomes ( K . pneumoniae or A . baumannii ) as for gonococci . Further , while a single GyrA mutation could explain the majority of CIP NS across all species evaluated here , unlike in gonococci and A . baumannii where this mutation explained ≥97% of CIP NS , 14% of CIP NS in K . pneumoniae could not be explained by this mutation , suggesting increased CIP resistance mechanism diversity and/or complexity in this species . Increased sampling , different methods , and/or finer tuning of hyperparameters may yield increased prediction accuracy for drug resistance in species with open genomes . For example , Nguyen et al . , 2018 reported a mean bACC of 98 . 5% ( average VME and ME rates of 0 . 5% and 2 . 5% , respectively ) using a decision tree-based extreme gradient boosting regression model to predict CIP MICs for the K . pneumoniae strains assessed here [14] , and adjusting for confounding factors such as population structure or variation in MIC testing method may yield more consistent prediction accuracies across species . However , our results demonstrate clear variation in potential limitations of genotype-to-resistance-phenotype models across different species . Given the biological and epidemiological disparities associated with resistance to different drugs in different clinical populations and bacterial species , and their evident impact on performance of predictive models , successful implementation of genotype-based resistance diagnostics will likely require sustained comprehensive sampling to ensure representation of complex , diverse , and/or novel resistance mechanisms , customized modeling , and incorporation of feedback mechanisms based on treatment outcome data . Further evaluation of additional ML methods and datasets may reveal more quantitative requirements and limitations associated with the application of genotype-to-resistance-phenotype predictive modeling in the clinical setting . See Table 1 for details of the datasets assessed and S7 Table for per-strain information . All gonococcal datasets contained a minimum of 200 isolates with WGS ( Illumina MiSeq , HiSeq , or NextSeq ) and MICs available for both CIP and AZM ( by agar dilution and/or Etest ) . Isolates lacking CIP and AZM MIC data were excluded . MIC testing methods are indicated in S7 Table . K . pneumoniae and A . baumannii datasets were selected based on the availability of isolates collected during a single survey that were tested for CIP susceptibility and whole genome sequenced using consistent platforms ( in both cases , the BD-Phoenix system and either Illumina MiSeq or NextSeq ) . MIC data were obtained from the associated publications , except in the cases of dataset 1 ( NCBI Bioproject PRJEB10016; see S7 Table ) and dataset 9 , which were obtained from the NCBI BioSample database ( https://www . ncbi . nlm . nih . gov/biosample ) . Raw sequence data were downloaded from the NCBI Sequence Read Archive ( https://www . ncbi . nlm . nih . gov/sra ) . Genomes were assembled using SPAdes [43] with default parameters , and assembly quality was assessed using QUAST [44] . Contigs <200 bp in length and/or with <10x coverage were removed . Isolates with assembly N50s below two standard deviations of the dataset mean were removed . Previously identified genetic loci associated with reduced susceptibility to CIP or AZM in gonococci are indicated in S1 and S2 Tables , respectively . The sequences of these loci were extracted from the gonococcus genome assemblies using BLAST [45] followed by MUSCLE alignment [46] to assess the presence or absence of known resistance variants . The presence or absence of quinolone resistance determining mutations in gyrA was similarly assessed in K . pneumoniae and A . baumannii assemblies . Presence or absence of gonococcal AZM resistance mutations in the multi-copy 23S rRNA gene was assessed using BWA-MEM [47] to map raw reads to a single 23S rRNA allele from the NCCP11945 reference isolate ( NGK_rrna23s4 ) , the Picard toolkit ( http://broadinstitute . github . io/picard ) to identify duplicate reads , and Pilon [48] to determine the mapping quality-weighted percentage of each nucleotide at the sites of interest . Predictive modeling was carried out using SCM and RF algorithms , implemented in the Kover [11 , 12] and ranger [49] packages , respectively . K-mer profiles ( abundance profiles of all unique words of length k in each genome ) were generated from the assembled contigs using the DSK k-mer counting software [50] with k = 31 , a length commonly used in bacterial genomic analysis [11 , 12 , 35 , 51] . For each dataset , 31-mer profiles for all strains were combined using the combinekmers tool implemented in SEER [35] , removing 31-mers that were not present in more than one genome in the dataset . Final matrices used for model training and prediction were generated by converting the combined 31-mer counts for each dataset into presence/absence matrices . For each SCM binary classification analysis ( using S/NS phenotypes based on the two different breakpoints for each drug ) , the best conjunctive and/or disjunctive model using a maximum of five rules was selected using five-fold cross-validation , testing the suggested broad range of values for the trade-off hyperparameter of 0 . 1 , 0 . 178 , 0 . 316 , 0 . 562 , 1 . 0 , 1 . 778 , 3 . 162 , 5 . 623 , 10 . 0 , and 999999 . 0 to determine the optimal rule scoring function ( http://aldro61 . github . io/kover/doc_learning . html ) . In order to assess binary classification across multiple methods , RF was also used to build binary classifiers ( RF-C ) using S/NS phenotypes . Further , to compare performance of binary classifiers to MIC prediction models , RF was used to build multi-class classification ( RF-mC ) and regression ( RF-R ) models based on log2 ( MIC ) data . For all RF analyses , forests were grown to 1000 trees using node impurity to assess variable importance and five-fold cross-validation to determine the most appropriate hyperparameters ( yielding the highest bACC or 1-tier accuracy for NS- or MIC-based models , respectively ) , testing maximum tree depths of 5 , 10 , 100 , and unlimited and mtry ( number of features to split at each node ) values of 1000 , 10000 , and either √p or p/3 , for classification and regression models , respectively , where p is the total number of features ( 31-mers ) in the dataset . While a grid search would enable assessment of more combinations of different hyperparameter values and thus finer tuning of hyperparameters , such an approach is computationally prohibitive on datasets of this size . To standardize reported MIC ranges across datasets , CIP MICs ≤0 . 008 μg/mL or ≥32 μg/mL were coded as 0 . 008 μg/mL or 32 μg/mL , respectively , and AZM MICs ≤0 . 008 μg/mL or ≥32 μg/mL were coded as 0 . 03 μg/mL or 32 μg/mL , respectively . The set of SCM and RF analyses performed are indicated in S3 and S6 Tables . For each of the seven individual gonococcal datasets , as well as the aggregate gonococcal dataset ( all gonococcal datasets combined , removing duplicate strains ) and the K . pneumoniae and A . baumannii datasets , training sets consisted of random sub-samples of two-thirds of isolates from the dataset indicated ( maintaining proportions of each resistance phenotype from the original dataset ) , while the remaining isolates were used to test performance of the model . Each set of analyses ( for each combination of dataset/drug/resistance metric/ML algorithm ) was performed on 10 replicates , each with a unique randomly partitioned training and testing set . For all gonococcal datasets , separate models were trained and tested using the EUCAST [52] and CLSI [53] breakpoints for NS to CIP . Four of the N . gonorrhoeae datasets had insufficient ( <15 ) NS isolates by the CLSI breakpoint for AZM non-susceptibility and thus were only assessed at the EUCAST AZM breakpoint . CIP MICs for the K . pneumoniae isolates were not available in the range of the EUCAST breakpoint ( 0 . 25 μg/mL ) , and thus only the CLSI breakpoint for NS ( >1 μg/mL ) was assessed . For A . baumannii , the EUCAST and CLSI breakpoints for ciprofloxacin NS are the same ( >1 μg/mL ) . Due to the very limited range of MICs within the BD-Phoenix testing thresholds and thus the CIP MICs available for K . pneumoniae and A . baumannii , predictive models based on MICs were not generated for these species . For analyses in S6 Table where datasets were down-sampled to equalize MIC distributions between datasets or the number of S and NS strains within datasets , the required number of strains from the over-represented class ( es ) were selected at random for removal . Model performance was assessed by sensitivity ( 1 –VME rate ) , specificity ( 1 –ME rate ) , and aggregate bACC ( the average of the sensitivity and specificity [54] ) . bACC was used as an aggregate measure of model performance as , unlike metrics such as raw accuracy , error rate , and F1 score , it provides a balanced representation of false positive and false negative rates , even in the case of dataset imbalance . For MIC prediction models , the percentage of isolates with predicted MICs exactly matching the phenotypic MICs ( rounding to the nearest doubling dilution , in the case of regression models ) , as well as the percentage of isolates with predicted MICs within one doubling dilution of phenotypic MICs ( 1-tier accuracy ) , were also assessed . In order to account for variations in MIC testing methods and thus in the dilutions assessed , criteria for exact match rates and 1-tier accuracies were relaxed to include predictions within 0 . 5 doubling dilutions or 1 . 5 doubling dilutions , respectively , of the phenotypic MIC . Mean and 95% confidence intervals for all metrics were calculated across the 10 replicates for each analysis . Differential model performance between datasets or methods was evaluated by comparing mean bACC between sets of replicates by two-tailed unpaired t-tests with Welch’s correction for unequal variance ( α = 0 . 05 ) . Unless otherwise noted , all P-values are derived from these unpaired t-tests . Relationships between MIC prediction accuracy and bACC and between dataset imbalance and model performance were assessed by Pearson correlation ( α = 0 . 05 ) .	Machine learning-based prediction of antibiotic resistance from bacterial genome sequences represents a promising tool to rapidly determine the antibiotic susceptibility profile of clinical isolates and reduce the morbidity and mortality resulting from inappropriate and ineffective treatment . However , while there has been much focus on demonstrating the diagnostic potential of these modeling approaches , there has been little assessment of potential caveats and prerequisites associated with implementing predictive models of drug resistance in the clinical setting . Our results highlight significant biological and technical challenges facing the application of machine learning-based prediction of antibiotic resistance as a diagnostic tool . By outlining specific factors affecting model performance , our findings provide a framework for future work on modeling drug resistance and underscore the necessity of continued comprehensive sampling and reporting of treatment outcome data for building reliable and sustainable diagnostics .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "antimicrobials", "united", "states", "medicine", "and", "health", "sciences", "pathology", "and", "laboratory", "medicine", "statistics", "pathogens", "drugs", "geographical", "locations", "microbiology", "neisseria", "gonorrhoeae", "organisms", "north", "america", "antibiotic", "resistance", "mathematics", "forecasting", "antibiotics", "artificial", "intelligence", "genome", "analysis", "pharmacology", "klebsiella", "bacteria", "bacterial", "pathogens", "research", "and", "analysis", "methods", "computer", "and", "information", "sciences", "antimicrobial", "resistance", "genomics", "neisseria", "medical", "microbiology", "mathematical", "and", "statistical", "techniques", "microbial", "pathogens", "people", "and", "places", "genetics", "microbial", "control", "biology", "and", "life", "sciences", "klebsiella", "pneumoniae", "physical", "sciences", "computational", "biology", "statistical", "methods", "machine", "learning" ]	2019	Evaluation of parameters affecting performance and reliability of machine learning-based antibiotic susceptibility testing from whole genome sequencing data
Chikungunya virus is a vector-borne alphavirus transmitted by the bites of infected female Ae . aegypti and Ae . albopictus . In Brazil between 2014 and 2016 almost 320 thousand autochthonous human cases were reported and in Florida numerous imported CHIKV viremic cases ( > 3 , 800 ) demonstrate the potential high risk to establishment of local transmission . In the present study , we carried out a series of experiments to determine the viral dissemination and transmission rates of different Brazilian and Florida populations of Ae . aegypti and Ae . albopictus at 2 , 5 , and 13 days post-infection for the emergent Asian genotype of CHIKV . Our results show that all tested populations of Ae . aegypti and Ae . albopictus have a high proportion ( > 0 . 80 ) of individuals with disseminated infection as early as 2 days-post exposure . We found no significant treatment effects of mosquito population origin effects on viral dissemination rates . Transmission rates had a heterogeneous pattern , with US Ae . aegypti and Brazilian Ae . albopictus having the highest proportion of individuals with successful infection ( respectively 0 . 50 and 0 . 82 as early as 2 days-post infection ) . Model results found significant effects of population origin , population origin x species , population origin x days post-infection and population origin x species x days post infection . Chikungunya fever is a vector-borne viral disease that originated in Africa and is caused by a virus ( CHIKV; family Togaviridae , genus Alphavirus ) transmitted by the bites of infected female Aedes mosquitoes , mainly Ae . aegypti and Ae . albopictus [1] . There are three genotypes of CHIKV , which apparently evolved independently in distinct geographic regions: Asian , West African , and East/Central/South African ( ECSA ) [2] . CHIKV is widespread worldwide and poses as a major public health problem in tropical and subtropical regions [3–6] . In the Americas , autochthonous transmission of CHIKV was first detected in St . Martin Island in October 2013 and quickly spread throughout the Americas in the following months [7–9] . The initial spread of autochthonous cases in the Americas was due to the Asian genotype , but the ECSA genotype was also detected circulating in Brazil in 2014 [10] . To date , local transmission of CHIKV has been documented in over 43 countries with more than 1 , 000 , 000 confirmed cases , where Brazil reported 314 , 834 until the 15th epidemiological week of 2017 [11–12] . Aedes aegypti and Ae . albopictus are the main vectors of CHIKV , and both are highly invasive species and closely associated with the human peridomestic environment [13 , 14 , 6] . Aedes aegypti is highly anthropophilic and exhibits endophilic behavior and is mostly associated with high human density . In contrast , Ae . albopictus shows an eclectic feeding behavior , preferentially feeding and resting in the peridomicile and is more common in vegetated and urban/urban forest transition habitats , especially where it is sympatric with Ae . aegypti [15–19] . In Africa , CHIKV is maintained via an enzootic cycle involving several species of arboreal mosquitoes , including Ae . africanus and Ae . furcifer , and non-human primates [20] . Epidemic transmission is maintained mainly by Ae . aegypti in urban environments , but a single-base mutation in a strain of the ECSA genotype during the outbreak in La Réunion Island enhanced vector competence of Ae . albopictus [21 , 22] . A second mutation is associated with enhanced vector competence of Ae . albopictus during an outbreak in Kerala , India [23] . In fact , the acquisition of second-step Ae . albopictus-adaptive mutations by CHIKV strains might indicate even more efficient transmission by this invasive vector [24] . Vector competence studies are important to determine the potential of resident mosquito populations to transmit CHIKV . Vector competence is a phenotypic parameter that describes the ability of the vector to become infected , replicate and transmit a pathogen [25 , 26] . Moreover , vector competence depends on vector and viral genetic characteristics [27] and environmental factors such as ambient temperature and diurnal temperature range [28–32] . It has been shown that vector competence of Ae . aegypti and Ae . albopictus for CHIKV is a complex interaction dependent on vector population , virus strain and temperature [33 , 34] . The vector competence of Ae . aegypti for dengue virus ( DENV ) has been shown to have high variability and heterogeneity whether it is analyzed at city [35] , country [36] or continental level [37] . Previous studies of CHIKV have characterized variation in vector competence among CHIKV genotypes , extrinsic incubation temperature , and geographic populations of Ae . aegypti and Ae . albopictus , and species-specific differences . In Florida , Ae . aegypti and Ae . albopictus were highly susceptible to infection and viral dissemination to ECSA and Asian genotypes of CHIKV , with some variation between strains [38 , 39] . Pesko et al . ( 2009 ) [40] evaluated vector competence of Ae . aegypti and Ae . albopictus from Florida for infection with a La Réunion island ECSA isolate of CHIKV . Although both species were susceptible to high CHIKV doses , Ae albopictus was more susceptible to infection than Ae . aegypti . Richards et al . ( 2010 ) [33] assessed the effect of extrinsic incubation temperature on vector competence of Florida mosquitoes for CHIKV isolates from La Réunion and found highest infection , dissemination , and transmission rates in Ae . albopictus than in Ae . aegypti and Culex quinquefasciatus , but no effect on the extrinsic incubation period . Vega-Rúa et al . ( 2014 ) [31] working with three CHIKV genotypes and 35 populations of Ae . aegypti and Ae . albopictus mosquitoes from 10 American countries showed that all Aedes populations tested were susceptible to CHIKV infection by all three genotypes . However , CHIKV transmission was heterogeneous in American Ae . aegypti and Ae . albopictus populations , ranging from 11 . 1% to 96 . 7% . In this study , the Aedes populations from Rio de Janeiro showed high transmission rates , and Ae . albopictus from Florida were more competent vectors than Ae . aegypti . Although Ae . aegypti is considered the primary epidemic vector of CHIKV and Ae . albopictus a potential vector in some areas [2 , 21 , 31] , heterogeneous vector competence of both species may alter risk of disease transmission , as evidenced by the participation of Ae . albopictus in the outbreak in La Réunion Island [21] . Studies comparing vector competence in American populations of both species are necessary in a scenario where travel and global trade in endemic regions have increased the risk for spread of CHIKV , as evidenced by its introduction in the Americas [41] . Also , there is a real risk for the introduction of CHIKV strains with adaptive mutations to enhance vector competence of Ae . albopictus , an invasive species which is widespread in the Americas [24 . With the aim to shed light on the causes and consequences of geographical variations in the transmission of arboviruses of public health concern , we carried out an experiment to determine the dissemination and transmission rates of Brazilian and Florida populations of Ae . aegypti and Ae . albopictus for the emergent Asian genotype of CHIKV . Chikungunya virus ( Asian lineage , GenBank accession: KJ451624 ) used was isolated from the serum of an infected human in the British Virgin Islands in 2013 by other investigators . Subsequently , this isolate was archived with the Centers for Disease Control and Prevention . We requested an isolate of this virus for use in this study and so the sample was already present in an already-existing collection ( Centers for Disease Control and Prevention , Arboviral Diseases Branch ) . The virus sample was anonymized and Institutional Review Board approval was not needed for receipt and use of the sample in this study . No entomological gathering was done on private land or in private residence for this study . The Ae . aegypti and Ae . albopictus populations used in this experiment were collected in Rio de Janeiro ( RJ ) and Macapá ( MC ) —Brazil , Key West ( KW ) and Okeechobee ( OK ) , Florida—United States ( Fig 1 , Table 1 ) . All gathering of entomological samples were done on public land . We chose collection sites based on allopatric Ae . aegypti to Ae . albopictus ( MC and KW ) and sympatric populations ( RJ and OK ) . Some of these areas report local transmission of chikungunya cases ( RJ and MC ) while others are located near regions in Florida where local transmission has occurred ( Miami-Dade , Palm Beach , St . Lucie , and Broward Counties ) ( KW and OK ) [12 , 42] . In Brazil , eggs of both species were obtained from oviposition traps during a routine entomological survey . Aedes albopictus from a sympatric population ( RJ ) were obtained at the Oswaldo Cruz Foundation campus , in the Manguinhos neighborhood in March 2015 from 50 oviposition traps using methods described elsewhere [16] . Aedes aegypti eggs from an allopatric population ( MC ) were collected with oviposition traps by personnel from the Amapá State Health Secretary in May 2015 . In United States , eggs of allopatric Ae . aegypti ( KW ) were collected in March 2015 with oviposition traps by personnel of Florida Keys Mosquito Control District . Immatures of sympatric Ae . albopictus ( OK ) were collected from tires in October 2015 . Field-collected mosquitoes ( eggs or larvae ) were reared in pans containing 1 L of tap water ( 100 larvae per pan ) to adulthood on a diet with 0 . 6 g of equal amounts of brewer’s yeast and lactalbumin . Mosquitoes were held in a climate controlled room at 26–28°C and a photoperiod of 14:10 hours light:dark . Upon pupation , pupae were collected daily and placed in vials with a cotton seal until eclosion after which adult mosquitoes were identified to species . Adults were transferred to 0 . 3m3 cages and provided with 10% sucrose solution and water from cotton wicks and allowed to feed on bovine blood once per week using an artificial feeding system with hog intestine membranes . Females and males were held together for eleven days after which females were transferred to cylindrical cages ( ht . by dia . , 10 cm by 10 cm , 50 females/cage ) with mesh screening one day before being fed CHIKV infected blood . The F2 ( Okeechobee ) and F3 ( Rio de Janeiro , Macapá and Key West ) generations progeny of field-collected Ae . aegypti and Ae . albopictus were used for the CHIKV infection study . The strain of CHIKV ( Asian lineage , GenBank accession: KJ451624 ) used was isolated from the serum of an infected human in the British Virgin Islands in 2013 . The Centers for Disease Control and Prevention was the source of the virus strain used in this study . The CHIKV isolate was passaged twice in culture using African green monkey ( Vero ) cells and viral titer was determined in 6-well plates seeded with Vero cells ( American Type Culture Collection , ATCC ) by plaque assay using a modified procedure by Kaur et al . ( 2016 ) [43] . For preparation of the virus suspension , monolayers of Vero cells were inoculated with dilute stock CHIKV at a multiplicity of infection of 0 . 1 followed by a one-hour incubation at 37°C and 5% carbon dioxide atmosphere . The American Type Culture Collection was the source of Vero cells used in this study . After the inoculation procedure , each flask received 24 ml media ( M199 medium supplemented with 10% fetal bovine serum , penicillin/streptomycin and mycostatin ) and was left to incubate for an additional 47-hours . Adult females aged 10–11 days were offered CHIKV infected defibrinated bovine blood ( Hemostat , Dixon , CA ) using an artificial feeding system with hog intestine membranes ( Hemotek , Lancashire , United Kingdom ) . Samples of blood were taken of the virus-blood suspension at the time of feeding to determine the concentration of CHIKV ingested by the adult mosquitoes . Blood meal titers ranged from log10 7 . 3 to 8 . 3 plaque forming unit equivalents ( pfue ) /mL . Fully engorged females were held in cylindrical cages along with an oviposition substrate and maintained at a 14:10 hour light:dark photoperiod and 28°C . Virus transmission potential using saliva assays was determined at 2 , 5 , and 13 days after feeding on infected blood . Mosquitoes were deprived of sucrose for 1-day and then individually transferred to plastic tubes fitted with a removable screen lid ( 37-mL 8 by 3 cm ) . Honey was dyed with blue food coloring ( McCormick ) and impregnated on filter paper ( 1 cm diameter ) and fastened to the inside lid of the tube . Mosquitoes that fed on the honey deposited saliva and the blue food coloring was visualized in the crop with aid of an incandescent flashlight . Mosquitoes were examined for blue in the crop after 24 and 48-hours during the transmission assay . Only mosquitoes that fed on honey were used to assess transmission potential . Additionally , saliva was collected from another subset of mosquitoes in capillary tubes with immersion oil as described previously [44 , 32 , 39] . Mosquitoes were stored at -80°C after the transmission assay and later dissected to test the legs and saliva for the presence of CHIKV RNA by qRT-PCR [32] . The sequence of primers targeting a nonstructural polyprotein gene was as follows: forward , 5'-GTACGGAAGGTAAACTGGTATGG-3': reverse , 5'-TCCACCTCCCACTCCTTAAT-3' . The probe sequence was: 5'-/56-FAM/TGCAGAACCCACCGAAAGGAAACT/3BHQ_1/-3' ( Integrated DNA Technologies , Coralville , IA ) . Detection of CHIKV RNA in the legs of a mosquito is considered a proof that the virus infection has disseminated from the midgut , and we use the number of mosquitoes with a disseminated infection over the number of mosquitoes fully engorged on a viraemic blood-meal , as the virus dissemination rate . Detection of CHIKV RNA in mosquito saliva is considered a proof that the mosquito can transmit virus when feeding , and we use the proportion of mosquitoes with virus in saliva among all mosquitoes with a disseminated infection as our expression of transmission rate . For each mosquito , legs were triturated in 1 . 0 mL of media ( GIBCO Media 199 ) . Saliva from mosquitoes was combined with 300 μL of media . RNA isolation on a 140 μL sample of mosquito legs and saliva homogenate was achieved using the QIAamp viral RNA mini kit ( Qiagen , Valencia , CA ) and eluted in 50 μL of buffer according to the manufacturer’s protocol . Viral RNA was detected using the Superscript III One-Step qRT-PCR with Platinum Taq kit by Invitrogen ( Invitrogen , Carlsbad , CA ) using methods described elsewhere [32 , 39] . Quantitative RT-PCR was performed with the CFX96 Real-Time PCR Detection System ( Bio-Rad Laboratories , Hercules , CA ) with the following program: 50°C for 30 minutes , 94°C for 2 minutes , 39 cycles at 94°C for 10 seconds and 60°C for 1 minute , and 50°C for 30 seconds . The expression of viral titer in mosquito-derived samples used a standard curve method comparing cDNA synthesis for a range of serial dilutions of CHIKV in parallel with plaque assays of the same dilutions of virus , expressed as plaque forming unit equivalents ( pfue ) /ml [45] . We were interested in analyzing the relationship between the presence or absence of CHIKV in the legs and saliva ( dependent variables ) and the following independent variables: mosquito species ( Ae . aegypti and Ae . albopictus ) , population origin ( Brazil and USA ) , days post-infection ( dpi , 2 , 5 and 13 ) , and a three-way interaction of species by population origin by days post-infection . Exploratory analyses were done using chi-square tests to verify possible relationships between both dependent variables ( presence or absence of CHIKV in the legs and saliva ) and each of the independent variables . We modeled this relationship using two separate binomial generalized linear models: one focused on the viral dissemination to the legs , and the other focused on the viral infection of saliva . To account for numerical problems in the viral dissemination binomial model , we used a Firth's Bias-Reduced Logistic Model [46] . We also analyzed the relationship between the viral titer of legs and saliva and the aforementioned main effects using a Gaussian generalized linear model . All analyses were done using R [47] and RStudio [48] , with the libraries ggplot2 [49] , logistf [46] and lsmeans [50] . Chikungunya virus ( Asian lineage , GenBank accession: KJ451624 , repository: Centers for Disease Control and Prevention ) . Chikungunya virus dissemination rates were measured by the proportion of mosquitoes that had infected legs from the total that fully engorged on infected blood . A total of 358 Aedes mosquitos were tested for disseminated infection ( 172 Ae . aegypti and 186 Ae . albopictus ) . Overall , our results showed the proportion of individuals of both species with disseminated infection significantly increased with each of the days post-infection analyzed ( 2-dpi , 0 . 847 ± 0 . 034; 5-dpi , 0 . 977 ± 0 . 013; and 13 dpi , 0 . 984 ± 0 . 011 ) ( χ2 = 24 . 35 , df = 2 , p<0 . 0001 ) . Aedes aegypti had higher dissemination rates than Ae . albopictus ( mean ± SE , 0 . 960 ± 0 . 014 and 0 . 919 ± 0 . 020 , respectively ) , although not significant ( χ2 = 2 . 09 , df = 1 , p = 0 . 148 ) . Both US and Brazilian populations of Ae . aegypti ( 0 . 976 ± 0 . 016 and 0 . 946 ± 0 . 023 , respectively ) had higher dissemination rates when compared to Ae . albopictus ( 0 . 915 ± 0 . 028 and 0 . 922 ± 0 . 028 , respectively ) , but this difference was also not significant ( χ2 = 0 . 03 , df = 1 , p = 0 . 857 ) . When analyzing the dissemination rates per species , population origin and days post-infection interaction , Ae . aegypti reached 100% of individuals at the 5th and 13th days , but the US population had higher dissemination rates at the 2nd day when compared to the Brazilian population ( 0 . 913 ± 0 . 06 and 0 . 814 ± 0 . 07 , respectively ) ( Fig 2 ) . These differences , however , were not significant ( χ2 = 0 . 07 , df = 2 , p = 0 . 961 ) . For Ae . albopictus , both US and Brazilian populations had similar dissemination rates at the 2nd day ( 0 . 843 ± 0 . 065 and 0 . 827 ± 0 . 071 ) . At the 5th day , the US population had a lower dissemination rate when compared to the Brazilian population ( 0 . 906 ± 0 . 052 and 1 . 0 , respectively ) . At the 13th day , the A . albopictus US population had a higher dissemination rate ( 1 . 0 ) than the Brazilian population ( 0 . 933 ± 0 . 046 ) . The dissemination rate did not significantly differ between population origins ( χ2 = 0 . 36 , df = 2 , p = 0 . 834 ) ( Fig 2 ) . The three-way interaction Firth's bias-reduced logistic model results show that none of the main effects or the interactions were significant for disseminated infection rates ( Table 2 ) . When analyzing the viral titers in the mosquito legs , Gaussian model results show that days post-infection had a significant positive effect , and the interaction of species and population origin had a significant negative effect ( Table 2 ) . Overall , both populations of Ae . aegypti had lower levels of viral titer ( expressed in log10 pfue/mL ) in their legs at 2nd day post-infection , which increased and peaked at the 5th and 13th days ( US; 2nd day = 2 . 884 ± 0 . 453 , 5th day = 4 . 289 ± 0 . 179 and 13th day = 4 . 131 ± 0 . 053; and Brazilian 2nd day = 2 . 668 ± 0 . 411 , 5th day = 3 . 610 ± 0 . 277 and 13th day = 4 . 119 ± 0 . 110 ) . The same pattern was observed for Ae . albopictus for both US ( 2nd day = 2 . 060 ± 0 . 290 , 5th day = 4 . 074 ± 0 . 263 and 13th day = 3 . 676 ± 0 . 244 ) and Brazilian populations ( 2nd day = 3 . 086 ± 0 . 362 , 5th day = 3 . 971 ± 0 . 241 and 13th day = 3 . 988 ± 0 . 183 ) ( S1 Fig ) . Chikungunya virus infection rates were measured by the proportion of mosquitoes that had infected saliva from the total that presented viral dissemination . A total of 224 Aedes mosquitoes that had positive leg infections were tested for saliva infection ( 107 Ae . aegypti and 117 Ae . albopictus ) . Overall , we found a significant effect of days post-infection and infection rates when analyzing both species ( χ2 = 8 . 88 , df = 2 , p<0 . 05 ) ( Fig 3 ) . The infection rates reached a peak at the 5th day post-infection and decreased at the 13th day ( 2-dpi , 0 . 415 ± 0 . 068; 5-dpi , 0 . 500 ± 0 . 050; and 13-dpi , 0 . 274 ± 0 . 053 ) . We also found a significant relationship between infection rates per species and population origin ( χ2 = 11 . 55 , df = 1 , p<0 . 0001 ) ; US Ae . aegypti had higher infection rates when compared to the Brazilian ( Ae . aegypti , 0 . 5 ± 0 . 068; Ae . albopictus , 0 . 264 ± 0 . 061 ) . For Ae . albopictus , the US population had lower infection rates when compared to Brazilian conspecifics ( 0 . 245 ± 0 . 057 and 0 . 6 ± 0 . 063 , respectively ) . The analysis of infection rates per species , population origin and days post-infection for Ae . aegypti showed that the US population had similar rates in all days ( 2-dpi , 0 . 5 ± 0 . 166; 5-dpi , 0 . 52 ± 0 . 101; 13-dpi , 0 . 473 ± 0 . 117 ) . The Brazilian population had a lower infection rate when compared with the US population at all day’s post-infection ( 0 . 1 ± 0 . 1 , 0 . 391 ± 0 . 2 and 0 . 104 ± 0 . 091 , respectively ) , although this difference was not significant ( χ2 = 1 . 32 , df = 2 , p = 0 . 67 ) . For Ae . albopictus , the US population had a lower infection rate at the 2nd and 13th days ( 0 . 125 ± 0 . 085 and 0 . 176 ± 0 . 095 , respectively ) and higher infection rates at the 5th day ( 0 . 375 ± 0 . 1 ) . The Brazilian population however had high infection rate at the 2nd day ( 0 . 823 ± 0 . 095 ) , decreasing at the 5th day ( 0 . 692 ± 0 . 092 ) and finally decreasing further at the 13thday ( 0 . 235 ± 0 . 106 ) . The Brazilian population had a higher infection rate at all day’s post-infection when compared to the US population , but this difference was not significant ( Fig 2 , χ2 = 3 . 05 , df = 2 , p = 0 . 238 ) ( Fig 3 ) . The three-way interaction logistic model results showed a significant effect of population origin , and the interactions between population origin x species , population origin x days post-infection and the three-way interaction of population origin x species x days post infection were significant for saliva infection rates ( Table 3 ) . The Gaussian model to analyze the viral titer in the saliva of the tested mosquitoes did not detect significant main effects or interactions of the treatment factors ( Table 3 ) . The US population of Ae . aegypti had similar levels of viral titer in the saliva at all three time-points tested ( respectively 1 . 794 ± 0 . 593 , 1 . 516 ± 0 . 247 and 1 . 351 ± 0 . 171 pfue/mL ) , while the Brazilian population had a peak at the 5th day and decreasing at the 13th day ( respectively 1 . 659 ± 0 . 376 and 1 . 300 ± 0 . 429 pfue/mL ) . The US population of Ae . albopictus had higher viral titer in their saliva at the 2nd day , decreasing with each passing time point ( 2 . 290 ± 0 . 730 , 1 . 036 ± 0 . 247 and 0 . 810 ± 0 . 228 pfue/mL ) . For the Brazilian population of this species , viral titer peaked at 5th days , decreasing at the 13th ( respectively 1 . 095 ± 0 . 140 , 1 . 501 ± 0 . 245 and 1 . 058 ± 0 . 458 pfue/mL ) ( S2 Fig ) . This study tested the vector competence of two populations of Ae . aegypti and Ae . albopictus from Brazil and Florida for an emergent Asian lineage of CHIKV . We carried out a series of experiments to determine two fundamental characteristics of this phenotypic trait: viral dissemination into the haemocoel of the tested mosquitos and saliva infection . These measurements characterize midgut and salivary gland barriers and are determinants of the vector competence of a mosquito population [26] . While viral dissemination indicates its propagation in the midgut and subsequent spread of the infection to other tissues , saliva infection is needed for the mosquito to successfully transmit the arbovirus by bite to a vertebrate host . Our results shed light on important questions regarding vector competence of Aedes mosquito populations of the Americas . The lack of statistical significance when comparing species and populations shows that viral dissemination occurs equally in these treatment conditions . In fact , more than 90% of all individuals have successful viral dissemination in their bodies , despite heterogeneity in species and population origin . This conclusion is further supported by the model results , which shows that none of the tested effects and interactions were statistically significant . Because high rates of disseminated infection were observed under these conditions , we had greater potential to detect treatment-dependent reductions in disseminated infection and less ability to identify treatment enhanced disseminated infection . In our study , viral dissemination occurred rapidly , with around 85% of all individuals with positive legs at the 2nd day post-infection , and more than 98% of mosquitoes tested positive at the 13th day-post infection . Rapid viral dissemination together with a short extrinsic incubation period , as observed by saliva infection assays , may have important consequences for CHIKV epidemiology , especially given that both these Aedes species exhibit gonotrophic discordance [51 , 52] . For instance , females will remain infectious for longer periods during the adult stage after ingesting CHIKV than pathogens with longer EIPs . Moreover , mosquito adult survival , EIP and host feeding strongly contribute to vectorial capacity which describes the number of infective bites received daily by a single host [53 , 6] . A more thorough analysis showed that both populations of Ae . aegypti had similar levels of viral dissemination , reaching 100% of all tested individuals at the 5th day post-infection . For Ae . albopictus , we found a similar pattern with an increasing proportion of individuals with disseminated infection with each passing day post-infection . However , only the US population reached 100% of individuals with disseminated infection . This high number of individuals of both species and populations with disseminated infection might suggest a lack of substantial midgut escape barriers for the CHIKV strain used [31] . It is unclear whether differences in disseminated infection rates may be observed among these invasive Aedes mosquitoes if lower titer CHIKV infected blood were ingested . Studies have shown differences in susceptibility of Aedes vectors to CHIKV depending the dose of virus ingested [54 , 55 , 39] . Differences in susceptibility of Ae . aegypti and Ae . albopictus from Florida to infection and transmission of two lineages of CHIKV ( Indian Ocean and Asian genotype ) were tested [39] . In this study , Ae . aegypti tested with a lower dose of CHIKV Asian genotype in two different temperatures ( 25°C and 30°C ) did not have significant differences in viral dissemination and transmission ( 100% to 40% and 33 . 3% to 0% , respectively ) . The low infection rates were attributed to a relatively low dose of CHIKV in blood meals ( 5 . 8 log10 pfue/ml ) . On the other hand , all populations of Ae . aegypti and Ae . albopictus presented higher susceptibility to infection and transmission for these two tested lineages of CHIKV at high titers [39 , 54] determined the relative susceptibility of selected strains of Ae . aegypti and Ae . albopictus fed on a viremic monkey to infection with Southeast Asian strain of CHIKV . The results showed that strains of Ae . albopictus , regardless of their geographical origin , were more susceptible to infection ( range , 72–97% ) and dissemination ( 36–80% ) with CHIKV than Ae . aegypti ( infection rate , 12–25% and dissemination 8–25% ) even though some strains presented lower infection rates in mosquitoes that ingested the lower dose ( 104 . 2–4 . 6 pfu/ml ) . Coffey et al . ( 2014 ) [55] summarizes numerous chikungunya virus infection with Ae . aegypti and Ae . albopictus , stating lower and higher doses used in infected blood meals . In this review , the authors showed that infection , dissemination , and transmission rates of both Aedes vectors can vary according to the geographic sources of mosquitoes and the titer of the ingested bloodmeal . For instance , using bloodmeal titers of > 7 log10 pfu/ml ( high dose ) presented 80% of Ae . aegypti from all locations develop disseminated infection . For Ae . albopictus , more than half became infected or develop disseminated infection . The infection and dissemination rates for US Ae . albopictus are dose-dependent and seem to increase with the titer of the ingested bloodmeal [21 , 40 , 55] . Vega-Rúa et al . ( 2014 ) [31] assessed 35 American Ae . aegypti and Ae . albopictus for three CHIKV genotypes with the titer of 107 . 5 pfu/ml , including mosquitoes populations from Brazil and Florida . Their study demonstrated that all 35 populations of both Aedes vectors were susceptible to CHIKV infection by all genotypes tested and that CHIKV transmission efficiency was highly heterogeneous in American mosquitoes ranging from 11 . 1% to 96 . 7% . Indeed , Ae . albopictus from Rio de Janeiro showed high transmission efficiencies even between geographically close populations , i . e . , with some populations being able to transmit infectious viral particles as early as 2 days post-infection . However , the vector competence of Ae . aegypti and Ae . albopictus from Vero Beach was not tested for the Asian lineage of CHIKV , but for Indian Ocean and ancestral ECSA genotypes showed that transmission efficiencies were low ( <30% ) . The proportion of individuals with saliva infection was substantially lower than those with viral dissemination , suggesting salivary gland barrier ( s ) [31 , 39] . Interestingly , US Ae . aegypti had almost twice as many infected individuals when comparing with the Brazilian population . A contrasting relation was observed for Ae . albopictus , with the Brazilian population reaching 60% of infected individuals against 24 . 5% from the US population . Thus , observed inherent differences in mosquito-virus interactions for both Ae . aegypti and Ae . albopictus might depend on geographic origin , which might impact disease transmission and contribute to its establishment in areas endemic for DENV and/or ZIKV . It is not clear whether heterogeneity exists in other traits that compose vector capacity , such as adult survival and biting rates , adult density , feeding behavior , and others , which would further influence CHIKV transmission and epidemiology in such areas [6] . Also , we observed that saliva infection declined with length of infection suggesting impaired transmission efficiency among older mosquitoes , most likely attributable to virus modulation of the infection as observed in other studies [56 , 57] . Further studies on vector competence of Ae . aegypti and Ae . albopictus should be done to analyze the heterogeneity of dissemination and transmission of CHIKV among different populations of endemic or receptive areas for this arbovirus using a range of viral titers .	Chikungunya is considered a serious mosquito-borne disease in many tropical and subtropical countries throughout the world . It is already an epidemic disease in Brazil and poses as a potential risk in Florida . It is mainly transmitted by mosquitoes Aedes aegypti and Aedes albopictus . These mosquito species are common and abundant throughout much of the year in Brazil and Florida . In this study , we determined two components of vector competence from Brazilian and Florida populations of both mosquitoes to the emergent Asian genotype of chikungunya virus: viral dissemination and transmission rates . Both Aedes populations exhibited a high proportion of disseminated infection as early as two days after ingestion of chikungunya virus infected blood . Transmission efficiency was higher in Ae . aegypti from Florida and Ae . albopictus from Brazil . Our findings suggest that mosquito-virus interactions of both Ae . aegypti and Ae . albopictus may vary by geographic population , which may impact public health measures and should be considered during outbreaks of this arboviral disease .	[ "Abstract", "Introduction", "Materials", "and", "methods", "Results", "Discussion" ]	[ "united", "states", "invertebrates", "medicine", "and", "health", "sciences", "body", "fluids", "pathology", "and", "laboratory", "medicine", "togaviruses", "chikungunya", "infection", "pathogens", "tropical", "diseases", "microbiology", "geographical", "locations", "saliva", "animals", "alphaviruses", "viruses", "north", "america", "chikungunya", "virus", "rna", "viruses", "neglected", "tropical", "diseases", "insect", "vectors", "florida", "infectious", "diseases", "south", "america", "aedes", "aegypti", "medical", "microbiology", "microbial", "pathogens", "disease", "vectors", "insects", "brazil", "arthropoda", "people", "and", "places", "mosquitoes", "eukaryota", "anatomy", "viral", "pathogens", "physiology", "biology", "and", "life", "sciences", "viral", "diseases", "species", "interactions", "organisms" ]	2018	Chikungunya virus vector competency of Brazilian and Florida mosquito vectors
A pandemic-capable influenza virus requires a hemagglutinin ( HA ) surface glycoprotein that is immunologically unseen by most people and is capable of supporting replication and transmission in humans . HA stabilization has been linked to 2009 pH1N1 pandemic potential in humans and H5N1 airborne transmissibility in the ferret model . Swine have served as an intermediate host for zoonotic influenza viruses , yet the evolutionary pressure exerted by this host on HA stability was unknown . For over 70 contemporary swine H1 and H3 isolates , we measured HA activation pH to range from pH 5 . 1 to 5 . 9 for H1 viruses and pH 5 . 3 to 5 . 8 for H3 viruses . Thus , contemporary swine isolates vary widely in HA stability , having values favored by both avian ( pH >5 . 5 ) and human and ferret ( pH ≤5 . 5 ) species . Using an early 2009 pandemic H1N1 ( pH1N1 ) virus backbone , we generated three viruses differing by one HA residue that only altered HA stability: WT ( pH 5 . 5 ) , HA1-Y17H ( pH 6 . 0 ) , and HA2-R106K ( pH 5 . 3 ) . All three replicated in pigs and transmitted from pig-to-pig and pig-to-ferret . WT and R106 viruses maintained HA genotype and phenotype after transmission . Y17H ( pH 6 . 0 ) acquired HA mutations that stabilized the HA protein to pH 5 . 8 after transmission to pigs and 5 . 5 after transmission to ferrets . Overall , we found swine support a broad range of HA activation pH for contact transmission and many recent swine H1N1 and H3N2 isolates have stabilized ( human-like ) HA proteins . This constitutes a heightened pandemic risk and underscores the importance of ongoing surveillance and control efforts for swine viruses . Numerous influenza A viruses ( IAVs ) exhibiting great diversity circulate in various host species , yet few strains evolve the traits necessary to jump between species and sustain an epidemic . The largest pool of IAVs is maintained in a global reservoir of wild aquatic birds [1] . Between them , avian IAVs include 16 hemagglutinin ( HA ) and 9 neuraminidase ( NA ) antigenic subtypes in many combinations [2 , 3] , which are designated H1-H16 and N1-N9 , respectively . The H17N10 and H18N11 subtypes have recently been discovered in bats [4 , 5] , although the extent to which these viruses are capable of transmission to other species is not yet known . Over the past century , a few avian IAVs have been able to establish long-term epidemics in domestic poultry and swine with human infection resulting from close contact with infected animals [6] . Avian H5 , H7 , and H9 subtypes caused outbreaks in poultry and have resulted in spillover into humans with limited human-to-human transmission . Domestic swine have proven highly capable of serving as bridging hosts for the adaptation of avian IAVs to replication in humans [1] . Swine are susceptible to many avian and human strains and can serve as a mixing vessel for the reassortment of 8 gene segments from different IAVs [7] . The current IAV epidemics in domestic swine herds are caused by H1N1 , H1N2 , and H3N2 strains . IAV diversity in swine is increased by antigenic drift [8 , 9] . In 2009 , a swine H1N1 virus emerged in humans and rapidly spread globally , causing a pandemic within months [10 , 11] . The 2009 pandemic H1N1 ( pH1N1 ) virus is the result of reassortment between viruses containing genes from classical swine IAVs ( HA gene ) , Eurasian swine IAVs ( NA and M genes ) , and triple-reassortant swine IAVs that contain internal genes derived from swine , human , and avian influenza viruses ( the M , NS , and NP genes are derived from classical North American swine IAV , the PB1 gene from a human IAV , and the PB2 gene from an avian IAV ) [11] . This complex reassortment of pH1N1 in pigs suggests that an optimized gene constellation helps promote the emergence of a human pandemic virus . Two well-established properties linked to IAV interspecies adaptation include polymerase activity and receptor binding [12] . Avian and human IAVs replicate efficiently at approximately 41°C and 33°C , respectively [13] . These temperatures correspond , respectively , to those of the avian enteric and human upper respiratory tracts . Mammalian adaptation has been linked to mutations in the PB2 protein at positions 591 , 627 , and 701 that increase polymerase activity and replication at the lower temperature [13–16] . During IAV entry , the HA protein binds sialic acid–containing receptors on the cell surface , the virus is internalized by endocytosis , and the HA protein is triggered by acidic pH to undergo irreversible structural changes that cause membrane fusion and release of the viral genome into the host cell cytosol [17] . Human- and avian-adapted IAVs bind preferentially to α-2 , 6- and α-2 , 3-linked sialic acid receptors , respectively , which are differentially expressed in various hosts , cells , and tissues [18–22] . Accordingly , human influenza viruses have shown a tendency to bind more strongly to α-2 , 6-containing non-ciliated cells in mammalian trachea and bronchi as well as type I pneumocytes in the lungs; whereas , avian influenza viruses appear to have a preference for binding α-2 , 3-containing ciliated cells and type II pneumocytes [18 , 19] . Both α-2 , 6- and α-2 , 3-linked sialic acid receptors have been extensively detected in pigs , having a respiratory distribution similar to that observed in humans [20] . Adaptation of IAVs to mammalian receptor binding has been associated with mutations in the receptor-binding pocket that switch receptor specificity to α-2 , 6 at HA positions 190 , 225 , 226 , and 228 ( H3 numbering ) [23–29] . Swine express both α-2 , 3- and α-2 , 6 forms of sialic acid receptors , permitting entry by IAVs with either receptor specificity [20 , 30 , 31] and thereby facilitating their role as bridging hosts . A third trait linked to interspecies adaptation of IAVs is HA acid stability , which is commonly defined as the activation pH at which irreversible HA conformational changes are triggered [32 , 33] . Avian influenza viruses tend to have relatively unstable HA proteins that are triggered at a higher activation pH than are those of human- and ferret-adapted IAVs [34–41] . This may not be universally true , as six H1N1 IAVs isolated from ducks and coots between 1976 and 1980 have HA activation pH values similar to those of human seasonal IAVs [42] . For H5N1 , a relatively unstable HA ( activation pH 5 . 6–6 . 0 ) is necessary for efficient replication and transmissibility in avian hosts [34 , 38] . In ferrets , a more stable HA protein ( activation pH < 5 . 6 ) is needed for efficient upper respiratory tract growth and airborne transmissibility of H5N1 [36 , 37 , 43 , 44] and pH1N1 viruses [39] . Furthermore , a stable HA protein has also been linked to pH1N1 pandemic potential and adaptation to humans [39 , 45] . The permissible range of HA activation pH in pigs and the pathways available for interspecies adaptation of this property may depend in part on the virus genetic constellation . Swine influenza viruses are highly diverse and have a complex evolutionary history in North America and Eurasia [46 , 47] . H1N1 , H1N2 , and H3N2 subtypes are currently endemic in pigs [48] . With respect to H1 viruses , the classical lineage was first isolated in North America in 1930 [49] , before which the clinical symptoms of influenza in pigs were described during the 1918 Spanish influenza pandemic [50] . The classical lineage ( H1α cluster ) remained dominant until the emergence of triple-reassortant swine viruses around 1998 [46] , after which there was a dramatic increase in swine influenza virus diversity [8] . H1β swine viruses were first detected in 2001–2002 , H1δ ( or “seasonal human-like” swine H1 ) in 2003–2005 , and H1γ strains in 1999–2000 [51] . Despite recognition that HA stabilization may be necessary for the adaptation of avian-like IAVs to ferrets and humans , the role of swine in this process is unknown . Pre-2009 swine H1 virus isolates have HA activation pH values ranging from pH 5 . 4 to 6 . 0 , early human 2009 pH1N1 isolates have HA activation pH values of 5 . 5 to 5 . 6 , and subsequent human-adapted 2010–2012 human isolates range from 5 . 2 to 5 . 4 [39 , 42] . These observations suggest an unstable HA may become stabilized to intermediate stability in pigs before the virus jumps to humans and HA becomes further stabilized . Other pathways may be possible , yet have not been investigated . To determine the importance of HA stability for replication in and transmission from swine , we measured the HA activation pH values of recently isolated H1 and H3 swine IAVs and evaluated experimentally the impact of HA activation pH on replication in swine , as well as on swine-to-swine and swine-to-ferret transmission . The results show swine permit replication and transmission by influenza viruses varying widely in acid stability . We measured the HA activation pH for 14 swine H1N1 and H1N2 IAVs isolated in 2009–2016 ( Fig 1 ) using a syncytia assay . These viruses contained HA genes of β , γ , δ1 , γ-pandemic-like , and pandemic lineages [52] . One of the isolates with the least HA stability , sw/NE/4D-0114-P14/2014 ( pH 5 . 9 ) , contains a pandemic-lineage HA gene ( S1 Table ) . This does not preclude swine from being infected with or transmitting influenza viruses containing HA proteins that are more acid stable . For example , the 2 contemporary swine H1 isolates with the greatest HA stability , sw/NE/4G-0314-P18/2014 ( pH 5 . 1 ) and sw/GA/1E-0214-P26/2014 ( pH 5 . 2 ) , also contain HA genes from the human pandemic lineage . Overall , the contemporary swine H1 viruses had HA activation pH values ranging from 5 . 1 to 5 . 9 ( S1 Table ) , which is a broader and significantly lower pH range ( P < 0 . 05 ) than that of pre-2009 swine H1 isolates , for which the HA activation pH values ranged from 5 . 4 to 6 . 0 ( Fig 1 ) . We next measured the HA activation pH for 57 contemporary swine H3N2 IAVs ( Fig 1 ) . These viruses had HA activation pH values ranging from 5 . 3 to 5 . 8 ( S2 Table ) , overlapping on the lower end with human seasonal and pandemic H3 IAVs [35] . Overall , the data show that swine support infection by contemporary H1 and H3 IAVs that have a broad range of HA stability and are in many cases acid stable ( activation pH < 5 . 5 ) . To determine experimentally the permissible range of HA activation pH for IAV replication and transmission in swine , we selected 3 previously characterized pH1N1 viruses that have a common A/TN/1-560/2009 ( pH1N1 ) backbone: WT ( pH 5 . 5 ) , HA2-R106K ( 5 . 3 ) , and HA1-Y17H ( 6 . 0 ) [39] . The two mutant viruses differ from the WT virus by a single amino-acid residue in their HA and have WT-like properties of expression , cleavage , and preferential α-2 , 6-linked sialic acid binding . All 3 viruses also had similar replication kinetics in MDCK , A549 , and NHBE cells . Compared to WT and R106K , Y17H with a destabilized HA protein ( pH 6 . 0 ) had reduced replication and was a loss-of-function mutant for airborne transmission in ferrets [39] . A/TN/1-560/2009 ( pH1N1 ) was isolated during the initial stages of the 2009 pandemic; thus , its lineage demonstrated pandemic capability in humans [53] . On multiple occasions , starting in 2009 , pH1N1 viruses were transmitted back to swine [54] , showing that the lineage retained an ability to infect swine . A/TN/1-560/2009 and the related A/CA/04/2009 replicated efficiently in ferrets and could be transmitted to them by the airborne route [39 , 55] . Ferrets have airway characteristics similar to those of humans , and this animal model is widely accepted to be well suited for studying the pathogenicity and transmissibility of IAV as observed in humans [56] . Because A/TN/1-560/2009 ( pH1N1 ) and closely related viruses replicate in and are transmitted in swine , ferrets , and humans , we considered this genetic backbone to be well suited for studying interspecies transmission . In 2 separate experiments , we intranasally inoculated a total of 8 pigs with pH1N1 WT , Y17H , or R106K viruses: 5 to be used for daily measurements of the viral load in nasal swabs and 3 in which to examine the tissue titers , histopathology , and inflammatory responses at 3 days post-inoculation ( dpi ) . All of the pigs seroconverted , with no significant differences being observed between the virus groups in terms of serum antibody titers at 14 dpi ( S1A Fig ) . Compared to WT virus , the stabilized R106K mutant displayed no statistically significant differences with respect to the viral load in nasal swabs , tracheal homogenates , or tracheobronchoalveolar lavage ( TBAL ) fluid ( Fig 2 ) . Histopathologic analyses also showed similar spread by WT and R106K viruses in the nasal turbinates and trachea ( Table 1 ) . In contrast , the R106K mutant showed greater spread in the lungs by immunohistochemistry , perhaps explaining in part why lung viral loads in the R106K group were an average of 30-fold higher ( P < 0 . 05 ) than in the WT group ( Fig 2D ) . Compared to WT virus , the destabilized Y17H mutant yielded peak nasal swab titers that were delayed by 3 days and reduced 30-fold ( P < 0 . 05 ) ; Y17H viral loads were also significantly reduced in TBAL fluid and lung homogenates ( Fig 2 ) . Two of the three Y17H-infected pigs were negative for NP by immunohistochemical staining ( IHC ) in their nasal turbinates , tracheae , and lungs . In contrast , all tissues were positive in WT-infected piglets except for the trachea of 1 animal ( Table 1 ) . All 3 viruses were mildly pathogenic , which is consistent with other studies on pH1N1 infection in pigs [55 , 57] . The animals were monitored daily for the following signs , which were not observed in any of the pigs: biting , aggression , squealing , increased scent marking , restless/constant walking and slipping , self-mutilation , diarrhea , weight loss , and open-mouthed breathing/gasping . For pigs infected with any of the 3 viruses , we found no notable lesions in the lungs or trachea , except in 1 animal infected with R106K ( Table 1 and S2 Fig ) . This pig had attenuated epithelium ( damaged and lost columnar epithelium that was replaced by a thin flattened epithelium covering the basement membrane ) in some bronchioles as well as cell debris in some alveoli and bronchioles . It is possible that some lesions may not have been included in the analyses because of the large sizes of the tissues . All 3 pigs inoculated with WT virus had damage to their nasal turbinates , characterized by multifocal ulcerated areas containing granulocytic inflammation . In contrast , similar observations were made in only a third of pigs infected with R106K or Y17H . Compared to the Y17H- or PBS-inoculated groups , WT- and R106K-infected pigs had more infiltrating cells in their TBAL fluid ( S3A Fig ) , as well as increased levels of mRNAs encoding proinflammatory cytokines ( i . e . , IL-1β and IL-6 ) and chemokines ( i . e . , MIP2α and MCP1 ) ( S3B Fig ) . Cellular infiltration and transcription of inflammatory genes in the lungs of pigs inoculated with Y17H virus were minimal . Overall , the HA-stabilizing mutation R106K ( pH 5 . 3 ) supported pH1N1 growth , spread , and pathogenicity comparable to those observed with WT virus ( pH 5 . 5 ) , whereas the HA-destabilizing Y17H mutation ( pH 6 . 0 ) resulted in delayed or reduced virus growth , spread , and pathogenicity . To study swine-to-swine and swine-to-ferret transmission , we co-housed naïve pigs ( the contact pig group ) in the same pen as donor pigs and positioned cages of naïve ferrets approximately 30 cm from the pigpen ( the ferret group ) . We collected nasal swabs from swine daily and nasal washes from ferrets every other day . There was 100% transmission of all 3 viruses ( 3/3 ) in both the contact pig and ferret groups ( Table 2 ) , as assessed by positive viral titers from piglet nasal swabs and ferret nasal washes ( Fig 3 ) and by serum antibody titers 2 weeks post-infection ( S1 Fig ) . The average peak titers were similar for all 3 viruses in both host species ( Fig 3 ) . For WT virus transmitted by contact transmission to pigs and by airborne transmission to ferrets , the average times of first detection and peak infection were approximately 3 and 4 days after donors were infected , respectively ( Table 2 ) . The transmission timing of the stabilized R106K virus was similar to that of WT virus ( P > 0 . 2 ) . In contrast , the transmission of the destabilized Y17H virus was significantly delayed ( P < 0 . 04 ) , with the first detection and peak of infection occurring at averages of 5 . 0 and 7 . 3 days after donor inoculation , respectively , in pigs and 6 . 3 and 9 . 0 days after donor inoculation , respectively , in ferrets ( Table 2 ) . To determine if HA stability changed in animals , we measured the HA activation pH of isolates from pig nasal swabs and ferret nasal washes . In the WT-infected/exposed groups of pigs and ferrets , the HA activation pH averaged 5 . 5 ( the input value ) and ranged from 5 . 45 to 5 . 60 ( Fig 4A and 4D ) . In the R106K-infected/exposed groups , the HA activation pH averaged 5 . 4 ( range , 5 . 30–5 . 55 ) in pigs and 5 . 3 ( range , 5 . 20–5 . 40 ) in ferrets ( Fig 4B and 4E ) . Thus , HA stability phenotypes were maintained in pigs and ferrets infected with WT and R106K viruses . In the Y17H groups , the HA activation pH ranged from 5 . 55 to 6 . 00 in donor pigs ( average , 5 . 83 ) , from 5 . 55 to 5 . 93 in contact pigs ( mean , 5 . 77 ) , and from 5 . 47 to 5 . 63 in ferrets ( mean , 5 . 52 ) ( Fig 4C and 4F ) . As the input HA activation pH for Y17H was 6 . 0 , average decreases of 0 . 2 and 0 . 5 pH units were associated with adaptation and transmission in swine and ferrets , respectively . For swine-to-ferret transmission of Y17H , the measured HA activation pH values for ferret recipients were initially 5 . 5 , 5 . 6 , and 5 . 9 ( Fig 4C ) . Thus , 2 of the 3 transmission events to ferrets were associated with HA stabilization at 5 . 5 to 5 . 6 , a stability of pH1N1 that is associated with airborne transmissibility in ferrets and human pandemic potential [39] . In the third ferret , a relatively unstable virus ( pH 5 . 9 ) was transmitted from a pig to the ferret before becoming stabilized at pH 5 . 4 ( Fig 4C ) . It is unknown whether the virus was transmitted by large droplets over a short range or by smaller aerosols capable of traveling longer distances . Space constraints in our animal facility prevented wide ( > 1 m ) separation of the ferret cages from the pig pens . From pig nasal swab and ferret nasal wash samples , we sequenced the HA , NA , and M genes , as all 3 genes may alter the HA activation pH [38 , 58 , 59] . Consistent with their maintaining an HA activation pH of approximately 5 . 5 , the WT-virus groups showed little nucleotide sequence variation ( Fig 5 ) , had only minor populations of HA gene variants , and had no populations of NA or M gene variants ( S4 Fig ) . The activation pH values for the R106K groups in pigs increased by an average of 0 . 1 pH units and remained at an average of 5 . 3 in ferrets ( Fig 4E ) . Accordingly , the R106K groups exhibited relatively little sequence variation ( Fig 5 ) . Each animal had some populations of one or more minor variants , but these generally did not increase in abundance over time ( S5 Fig ) . No variant with a K253R mutation in the NA protein was found in pigs , whereas the mutation had an abundance of 95% in one R106K-group ferret on day 3 , although this decreased to 68% of the virus population within 2 days , suggesting that it is not a preferred mutation . Indeed , it may have limited structural or functional impact on the NA protein , as K253 is a surface residue located on the bottom of the NA head at a turn between two β-strands located distal to both the enzyme active site and the adjacent protomers of the tetramer [60] . The HA sequence variation averaged approximately 2% in the Y17H groups ( Fig 5 ) , and subpopulations of viruses containing HA protein mutations emerged ( S6 Fig ) . Minor populations of the NA mutations V46A and K253R were detected in 1 donor pig and 1 contact pig , respectively , but were not associated with transmission . The following five HA mutations located in the stalk region were transmitted to or emerged in contact pigs and/or ferrets: HA1-H17Y ( reversion ) , HA2-V55I , HA2-R106K , HA2-K153E , and HA2-V192A ( Fig 6 ) . In both experiments , a proportion of the HA1-H17Y reversion mutation was detected in pigs ( donor and contact ) and ferrets . HA1-Y17 forms a hydrogen bond with the fusion peptide backbone ( Fig 6E ) , stabilizing the HA protein by approximately 0 . 5 pH units ( Fig 7 ) . In experiment 1 , HA2-K153E emerged in each of the 3 donor pigs and in 1 of the contact pigs ( S6 Fig ) . The HA2 residue K153 is located in helix G of the membrane-proximal region . The K153 sidechain may exert electrostatic repulsion with HA2-H26 , which is located in the center of one of the two β-strands attached to the fusion peptide ( Fig 6F ) . We generated reverse-genetics viruses containing mutations associated with transmission in the Y17H groups . A K153E mutation reduced the HA activation pH by approximately 0 . 2 units on the backgrounds of WT pH1N1 , Y17H , and Y17H/R106K ( Fig 7 ) . In experiment 2 , HA2-V55I was a minor population in 1 of the 2 donor pigs , was more than 90% abundant in the contact pig , and reached more than 99% abundance in each of the 3 ferrets ( S6 Fig ) . The HA2 residue V55 is located in helix A , which buttresses the central coiled-coil ( Fig 6C ) . The packing of residue 55 into its pocket may be stabilized by a V55I mutation , which reduced the HA activation pH by approximately 0 . 2 to 0 . 3 units ( Fig 7 ) . Some isolates contained HA2-V192A substitutions , but their proportions remained below 30% , suggesting that these substitutions did not improve fitness . Minor populations of HA2-R106K were also detected in both experiments . The R106K mutation stabilizes the HA protein by 0 . 2 to 0 . 3 pH units , perhaps by reducing electrostatic repulsions at the core of the central triple-stranded coiled-coil at the hinge between helices C and D ( Figs 6D and 7 ) . Overall , mutations associated with Y17H adaptation in pigs and ferrets were found to stabilize the HA protein , consistent with the average HA activation pH for this group shifting from 6 . 0 to 5 . 8 in pigs and 5 . 5 in ferrets ( Fig 4 ) . In contrast , the R106K virus , which had an HA stabilized at pH 5 . 3 , retained a stable genotype and phenotype , along with wild-type–like replication and transmissibility in pigs and ferrets by contact and airborne routes , respectively . Pigs appear to be well suited as an intermediate host within which zoonotic influenza genes may undergo initial mammalian adaptation in advance of a human pandemic . Humanizing adaptations may arise after the reassortment of human and animal influenza viruses , as was the case in the 3 previous pandemics in 1957 , 1968 , and 2009 [61] . A pandemic virus may also arise after direct adaptation of a swine , avian , or other zoonotic virus in humans . Pigs tolerate the 2 previously known humanizing adaptations required for pandemic potential: a switch in receptor-binding specificity from α-2 , 3- to α-2 , 6-linked sialic acid and a decrease in the temperature required for optimal polymerase activity from approximately 41°C to 33°C [54 , 55] . A third molecular adaptation has recently been discovered to be necessary for airborne transmissibility of H5 influenza viruses in ferrets and pH1N1 pandemic potential in humans: stabilization of the HA protein [36 , 39 , 62] . The observed HA activation pH values of H1 and H5 viruses in avian hosts typically , but not always , range from pH 5 . 5 to 6 . 0 , whereas those of human-adapted H1 , H2 , and H3 viruses range from approximately 5 . 1 to 5 . 6 [33 , 35 , 42 , 63] . Thus , the HA proteins of avian influenza viruses generally appear to be less stable than those of human influenza viruses . We measured the HA activation pH values of circulating H1 and H3 swine influenza viruses and found them to range from 5 . 1 to 5 . 9 . We also investigated the capacities of engineered pH1N1 viruses with HA activation pH values of 5 . 3 , 5 . 5 , and 6 . 0 to replicate in pigs and be transmitted from pigs to pigs by contact and from pigs to ferrets by the airborne route . All 3 viruses replicated in 100% of the inoculated pigs and were transmitted with 100% efficiency to contact pigs and ferrets . After transmission , only the input virus with an HA activation pH of 6 . 0 was genetically unstable , resulting in average HA activation pH values after transmission that ranged from 5 . 3 to 5 . 9 in pigs and from 5 . 2 to 5 . 6 in ferrets . When these results were combined with surveillance data , pigs were found to support a broad range of HA activation pH ( 5 . 1–5 . 9 ) , consistent with the notion that this species serves as a bridging host for HA stabilization , similar to its role with respect to receptor binding and polymerase activity . The cohousing of pigs , both in this experiment and in commercial facilities , may also contribute to a broad range of tolerable HA activation pH for virus transmission . Cohousing of animals allows contact transmission , which was previously shown to be more permissive than airborne transmission between ferrets for the destabilized Y17H mutant [39] . Comparing the two experiments , swine in the Y17H group were able to transmit with 100% efficiency to ferrets by the airborne route , whereas ferret donors promoted less efficient airborne transmission to ferrets ( 25% ) . Given that human influenza A viruses prefer a stable HA protein but many zoonotic viruses have an HA protein that is relatively unstable [35 , 39] , it is important to identify the evolutionary pathways by which an unstable HA could acquire increased stability . The 2009 pH1N1 pandemic virus most likely evolved by a stepwise stabilization pathway , as HA acid stability progressively increased during the evolution of H1N1 from swine precursors ( pH 5 . 5–6 . 0 ) and early 2009 human isolates ( pH 5 . 5 ) to later human isolates ( pH 5 . 2–5 . 4 ) [35 , 39 , 45 , 64 , 65] . The surveillance results reported here show that post-2009 swine H1 viruses have a broad range of HA activation pH ( 5 . 1–5 . 9 ) , and experimental infections in pigs show that a pH1N1 virus with an HA activation pH of 5 . 3 is readily transmitted between swine and from swine to ferrets without attenuation , fixed mutations , or a change in its HA acid stability . Thus , HA stabilization may also occur by a wide-range crossover pathway in which the HA protein becomes stabilized in swine without becoming attenuated . In a recent study , we found that direct inoculation of a pH1N1 virus with an unstable HA protein ( pH 6 . 0 ) into ferrets can result in HA stabilization within the ferret host [39] . If these results can be extended to humans , then a third pathway to HA stabilization is direct adaptation in humans . HA stabilization was required for the direct adaptation of airborne transmissibility in ferrets by avian H5 viruses containing mutations conferring α-2 , 6 receptor–binding specificity [36 , 37 , 62] , further demonstrating the importance of this molecular property in interspecies adaptation . Similarly , the 1957 and 1968 pandemic viruses that emerged after reassortment had stable HA proteins ( with activation pH values of approximately 5 . 1 ) [35]; however , the HA activation pH values of the prepandemic precursor viruses are unknown . For pre-2009 H1N1 swine IAVs , the ranges of observed HA activation pH values from classical , Eurasian avian-like , and triple-reassortant lineages were measured as 5 . 5 to 5 . 8 , 5 . 5 to 6 . 0 , and 5 . 4 to 5 . 7 , respectively [39 , 42] . Thus , the overall observed range is intermediate to unstable at pH 5 . 4 to 6 . 0 . Here , we measured the HA activation pH values of swine IAVs isolated in 2009 or later . Postpandemic H1N2 viruses ( pH 5 . 5–5 . 9 ) had HA acid stability similar to that of prepandemic swine IAVs , whereas postpandemic H1N1 ( pH 5 . 1–5 . 7 ) and H3N2 ( pH 5 . 3–5 . 8 ) were shifted to lower values that overlap at the bottoms of their ranges with those viruses associated with adaptation to humans [39 , 45 , 64 , 65] . With respect to HA activation pH , contemporary swine H1N1 and H3N2 IAVs most likely pose a greater pandemic risk than do pre-2009 swine viruses , as isolates of these strains that contain stable ( human-like ) HA proteins appear to retain fitness . However , it should be noted that pandemic potential is also determined by the antigenic distance of a given zoonotic virus from those that have circulated in humans . The present findings should heighten appreciation of the threat posed by swine IAVs in terms of causing a future pandemic in humans . Evidence is growing that HA stabilization plays a key role in interspecies adaptation and human pandemic potential [35–37 , 39 , 43 , 44] , and here we found that contemporary swine IAVs with a stable HA protein remain fit in swine . Swine also support the other 2 known molecular properties associated with human pandemic potential , namely α-2 , 6-receptor binding specificity and efficient polymerase activity at 33°C [55] . Given the historical importance of swine IAVs in the emergence of pandemic influenza , surveillance of circulating swine IAVs should be intensified and should include measuring molecular properties associated with human pandemic potential , in addition to gene sequencing and monitoring potential drift away from cross-reactivity with human IAV antibody responses . Close monitoring of IAVs circulating in swine production systems would not only enable further analysis of the natural evolution of swine IAV strains but also help us anticipate the emergence of viruses with pandemic potential , enabling enhanced preparation for and prevention of swine-to-human transmission . Madin-Darby Canine Kidney ( MDCK ) , African green monkey kidney ( Vero ) , and human embryonic kidney ( HEK 293T ) cells were obtained from the American Type Culture Collection . MDCK and Vero cells were maintained in Dulbecco’s modified Eagle’s medium ( MEM ) supplemented with 5% fetal bovine serum ( FBS ) and 1% penicillin-streptomycin at 37°C in 5% CO2 . 293T cells were maintained in Opti-MEM containing 10% FBS at 37°C in 5% CO2 . A/Tennessee/1-560/2009 recombinant viruses were generated by reverse genetics with pHW2000 plasmids , each containing an individual gene , being transfected into co-cultures of MDCK and 293T cells as described previously [66] . Amino-acid changes were introduced into the pHW2000-HA plasmid by using the QuikChange site-directed mutagenesis kit ( Stratagene , Cedar Creek , TX ) in accordance with the manufacturer's instructions . The Y17H and R106K mutants were previously described [39] . Virus stocks were prepared in MDCK cells and titrated by plaque assay . Virus identity and the absence of unintended mutations were confirmed by Sanger sequencing and next-generation sequencing as shown previously [39] . The contemporary swine H1 and H3 influenza viruses described in S1 and S2 Tables were obtained from the repository at St . Jude Children’s Research Hospital ( St . Jude ) and propagated in MDCK cells . Animal experiments were conducted in an ABSL2+ facility in compliance with the NIH and the Animal Welfare Act and with the approval of the St . Jude Animal Care and Use Committee , protocol number 464 . Food and water were provided ad libitum to all animals . Animals were observed daily for signs of diseases or stress . The 3-week-old piglets ( Midwest Research Swine , Glencoe , MN ) and 5-month-old male ferrets ( Triple F farms , Sayre , PA ) tested negative for IAVs . Two animal experiments were performed . In both experiments , 1 . 4 × 106 PFU of virus in PBS was intranasally inoculated using a spray bottle . In experiment 1 , we inoculated 3 donor pigs and 1 day later co-housed 2 naïve contact pigs in the pen . Nasal swabs were collected daily for 11 days for viral titer determination , and serum was collected on day 15 of the experiment for seroconversion testing . In experiment 2 , we inoculated 5 pigs and 1 day later introduced 1 naive contact pig into the pen and positioned 3 naïve ferrets in cages approximately 30 cm from the pen , thereby allowing the exchange of droplets and aerosol particles . On day 3 of the experiment , 48 h after naïve animals were added , 3 of the directly inoculated pigs were euthanized by intracardiac administration of Euthasol solution ( sodium pentobarbital and sodium phenytoin ) under anesthesia and exsanguinated . The lungs and trachea were washed with 50 mL PBS containing 2 mM EDTA , and the BALF was harvested for virus titration and cell counting . Nasal turbinates , tracheae , and lungs were collected then homogenized in PBS in the Qiagen Tissue Lyser II , and a TCID50 titration was performed in MDCK cells . Tissues were also used for analyses of cytokine and chemokine mRNA expression and for histopathologic analysis . Swine respiratory samples were collected 3 days after infection . Whole lungs , tracheae , and nasal turbinates were fixed in 10% neutral-buffered formalin , embedded in paraffin , and sectioned . Sections on slides were stained with hematoxylin and eosin or with polyclonal anti–influenza NP antibody and examined by light microscopy in a blinded fashion by a pathologist according to common guidelines . RNAlater-preserved swine tissues were homogenized , and total RNA was extracted using the RNeasy Mini Kit ( Qiagen , Germantown , MD ) . The levels of IFN-α , IL-1β , IL-6 , IL-8 , MIP2α , and MCP1 mRNA were analyzed by semiquantitative real-time PCR analysis on a 7500 Fast Real-Time PCR system ( Applied Biosystems , Waltham , MA ) . Briefly , mRNA was reverse transcribed using oligo-dT primers and the SuperScript III First-Strand Synthesis System ( Invitrogen , Carlsbad , CA ) . The resulting cDNA was analyzed with specific primers and the QuantiTect SYBR green PCR master mix ( Qiagen ) in accordance with the manufacturer’s instructions . The primers for swine housekeeping ( 18S ) and cytokine genes were described previously [67] . Samples were analyzed in triplicate . After normalization to 18S , the fold-change ratio of expression in virus-infected to that in control samples was calculated for each gene by using the ΔΔCt method and expressed as 2−ΔΔCt . Seroconversion was tested 2 weeks after virus inoculation or contact . Serum was treated with receptor-destroying enzyme ( Denka Seiken , Campbell , CA ) overnight at 37°C to destroy nonspecific inhibitors , heat-inactivated at 56°C for 30 min , and tested by a hemagglutination inhibition ( HI ) assay with A/Tennessee/1-560/2009 WT virus and 0 . 5% turkey red blood cells ( Rockland Immunochemicals Inc . , Limerick , PA ) . The HI titer was determined as the reciprocal of the highest serum dilution that completely inhibited hemagglutination . Deep amplicon sequencing was used to determine non-synonymous variations in the HA , NA , and M genes of viruses isolated from unpassaged pig nasal swab and ferret nasal wash samples . Two-step reverse transcription–PCR ( RT-PCR ) , DNA library preparation , and genomic sequence analysis were performed as previously described [39 , 68] . Briefly , viral RNA was extracted using TRIzol ( Ambion , Carlsbad , CA ) , and cDNA was synthesized via reverse transcriptase PCR with the SuperScript III First-Strand Synthesis System ( Invitrogen ) . Influenza A virus HA , NA , and M gene segments were separately amplified using Phusion High-Fidelity PCR Master Mix with HF Buffer ( New England BioLabs , Ipswich , MA ) and specific primers . PCR amplicons were purified with the QIAquick Gel Extraction Kit ( Qiagen ) and prepared using the Nextera XT cDNA Library Prep Kit ( Illumina , San Diego , CA ) in accordance with the manufacturer’s protocol . High-throughput paired-end sequencing was performed using a 2 × 150-bp cycle on an Illumina MiSeq platform . Data analysis was performed using CLC Genomics Workbench 8 software ( CLC Bio , Aarhus , Denmark ) and a custom Quality-Based Variant Detection pipeline . The variants were called if they met the predefined quality scores and were present in both forward and reverse reads at equal ratios . In addition , the minimum variant read frequency was set at 5% , and variants had to be supported by a minimum of 10 reads . All segments sequenced were completely and equally covered . Heat maps were assembled using Excel ( Microsoft Office Professional Plus 2010 ) . The mean variation frequency for each sample/time point was calculated by using positions that are variable in at least one of the examined samples/reads . If a site had a non-synonymous mutation with a frequency greater than 5% in any of the samples , the proportion of the variant was included in the numerator; the denominator was the number of variable sites compared to the inoculum . HA activation pH values were determined by syncytium and acid inactivation assays [69 , 70] . For the syncytium assay , Vero cells were infected with the recombinant viruses or nasal swab/wash samples and appropriate control viruses for 1 h . At 18 to 24 h after infection , HA-expressing cells were treated with 5 μg/mL l-tosylamido-2-phenylmethyl chloromethyl ketone ( TPCK ) -treated trypsin ( Worthington Biochemical , Lakewood , NJ ) for 15 min and pH-adjusted PBS buffers for 5 min at 37°C . Cells were incubated in MEM containing 5% FBS for 3 h at 37°C . Cells were fixed and stained with a Protocol Hema 3 kit ( Fisher Scientific , Kalamazoo , MI ) , and syncytium formation was observed by light microscopy . The pH of activation was determined as the highest pH value at which syncytia were observed . To measure the effect of acid exposure on in vitro inactivation , 10 μL of nasal swab/wash samples or virus stocks were diluted in 990 μL of pH-adjusted PBS solutions and incubated for 1 h at 37°C . The remaining infectious virus titer was then determined by TCID50 titration . The curves were fitted to an asymmetric ( 5-parameter ) regression model , and the pH50 values were determined as the point at which a 50% change between the maximum and baseline was observed . Student’s t-test , 1-way ANOVA followed by a Tukey post-hoc test , and 2-way ANOVA with the Bonferroni test were used to compare groups . P-values of less than 0 . 05 were considered significant . All statistical analyses were performed with GraphPad Prism 5 software .	Many genetically diverse influenza viruses circulate among wild aquatic birds . Occasionally , one causes an outbreak in domestic poultry , swine , other animals , or humans . Zoonotic influenza viruses rarely cause a human pandemic . The hemagglutinin ( HA ) surface glycoprotein is the major surface antigen . A pandemic-capable virus requires an HA protein immunologically unseen by most people . During entry , HA binds sialic-acid-terminating receptors . A switch of preference from avian- to human-preferred receptor form is necessary but not sufficient for pandemic potential . What else is needed ? After receptor binding , the virus is endocytosed . Acidification triggers the HA protein to undergo irreversible structural changes that cause membrane fusion , allowing genome delivery . Avian viruses tend to have unstable ( easy to activate ) HA proteins while humanized viruses prefer greater stability . Here , we show pigs permit replication and contact transmission of both stable and unstable HA proteins to recipient pigs . Thus , swine may accommodate both human-like and avian-like HA stability and receptor-binding properties , which have been linked to pandemic potential . Many swine viruses isolated since 2009 contain HA proteins already humanized with respect to receptor binding and stability . One could cause the next pandemic if it becomes immunologically distinct from human influenza viruses .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "medicine", "and", "health", "sciences", "microbial", "mutation", "pathology", "and", "laboratory", "medicine", "viral", "transmission", "and", "infection", "pathogens", "microbiology", "vertebrates", "orthomyxoviruses", "animals", "mammals", "viruses", "rna", "viruses", "h1n1", "swine", "ferrets", "influenza", "a", "virus", "medical", "microbiology", "microbial", "pathogens", "viral", "replication", "influenza", "viruses", "virology", "viral", "pathogens", "biology", "and", "life", "sciences", "amniotes", "organisms" ]	2017	H1N1 influenza viruses varying widely in hemagglutinin stability transmit efficiently from swine to swine and to ferrets
The surface proteins of human influenza A viruses experience positive selection to escape both human immunity and , more recently , antiviral drug treatments . In bacteria and viruses , immune-escape and drug-resistant phenotypes often appear through a combination of several mutations that have epistatic effects on pathogen fitness . However , the extent and structure of epistasis in influenza viral proteins have not been systematically investigated . Here , we develop a novel statistical method to detect positive epistasis between pairs of sites in a protein , based on the observed temporal patterns of sequence evolution . The method rests on the simple idea that a substitution at one site should rapidly follow a substitution at another site if the sites are positively epistatic . We apply this method to the surface proteins hemagglutinin and neuraminidase of influenza A virus subtypes H3N2 and H1N1 . Compared to a non-epistatic null distribution , we detect substantial amounts of epistasis and determine the identities of putatively epistatic pairs of sites . In particular , using sequence data alone , our method identifies epistatic interactions between specific sites in neuraminidase that have recently been demonstrated , in vitro , to confer resistance to the drug oseltamivir; these epistatic interactions are responsible for widespread drug resistance among H1N1 viruses circulating today . This experimental validation demonstrates the predictive power of our method to identify epistatic sites of importance for viral adaptation and public health . We conclude that epistasis plays a large role in shaping the molecular evolution of influenza viruses . In particular , sites with , which would normally not be identified as positively selected , can facilitate viral adaptation through epistatic interactions with their partner sites . The knowledge of specific interactions among sites in influenza proteins may help us to predict the course of antigenic evolution and , consequently , to select more appropriate vaccines and drugs . Influenza A is among the most extensively studied viruses , owing to its importance as a human pathogen [1]–[6] . With a large , public database of genetic sequences , influenza viruses also offer a model system for studying molecular evolution in general . The evolution of influenza viruses is characterized by frequent reassortment events within subtypes [3] , [7] as well as high rates of amino-acid substitutions in the viral surface proteins hemagglutinin ( HA ) and neuraminidase ( NA ) [8]–[10] . Such high evolutionary rates reflect both the poor fidelity of the viral polymerase [10] , and the strong selection pressures to evade the human immunity [8] , [9] , [11]–[13] and , more recently , to develop drug resistance [14]–[16] . Numerous experimental studies and statistical analyses of genetic and antigenic data have identified sets of residues in HA and NA proteins , the so called epitopes , that are bound by human antibodies [17]–[20] . As a consequence , the epitopic sites tend to evolve especially quickly , in order to evade immunity [8] , [21] , [22] . Moreover , several recent studies have suggested lists of amino acids at specific residues in HA that evolved under positive selection over the past 40 years [23]–[25] . In addition to escaping human antibodies , several other selective forces act on hemagglutinin . As with any functional protein , HA must maintain its stability and its function – namely , to bind the sialic acid receptor of host cells and subsequently mediate membrane fusion [17] , [18] , [20] , [26] . Thus , antibody escape mutations must not compromise these properties . Yet , numerous studies of protein evolution in vitro [27]–[29] as well as studies in bacteria [30] and viruses [20] , [31] , [32] have shown that beneficial mutations are often pleiotropic: in addition to their original beneficial effect , they cause some , usually negative , side effects on other protein properties , such as stability [28] , [33] . These negative effects can typically be alleviated or compensated by other mutations , making certain combinations of mutations substantially more beneficial than single mutations alone [34] , [35] . This phenomenon is known as positive epistasis between mutations [36] . Epistasis can also be negative , if a combination of mutations confers a smaller fitness gain than would be expected under additive effects of the individual mutations [36] . Epistasis is commonplace in eukaryotes [37]–[39] , bacteria [30] , [40] , [41] , and viruses [31] , [34] , [35] , [42] , and it plays an important role in the evolution of immune escape and drug resistance in various pathogens [35] , [43]–[46] including influenza [16] , [32] . Surprisingly , the extent of epistatic interactions in influenza proteins has not been systematically quantified or utilized . Yet , the knowledge of such interactions might provide a powerful tool for predicting future antigenically important substitutions and , consequently , for selecting better vaccine strains . Numerous methods have been developed for detecting epistasis between mutations , based on sampled genetic sequences [47] . Early methods were based on the idea that co-evolving pairs of sites in a protein should leave a typical signature in a sequence alignment , which can be detected using quantities such as mutual information [48]–[50] . However , such methods ignore the phylogenetic relationships among sequences and so are justified only if the divergence times between samples are very large [51] . Various corrections for the phylogenetic non-independence have been proposed [52]–[54] , and their performance has been shown to be satisfactory in some cases [55]–[57] . Nevertheless , methods that explicitly take account of the phylogeny are preferable [58] . Several such methods have been proposed recently [42] , [59]–[66] . Most of them attempt to detect unusually frequent co-occurrences of substitutions at pairs of sites on individual branches of the phylogeny . This approach is conservative since it detects only those positively epistatic pairs of sites for which a mutation at one site increases the beneficial effect of a mutation at the second site so dramatically that one mutation could not fix without the other one [42] . Such strong epistasis can occur , for example , when one mutation confers a strongly deleterious effect that is compensated by a second mutation . However , a mutation at one site in a protein may lead to only a moderate increase in the beneficial effect of a mutation at another site , so that the latter substitution occurs at an accelerated rate , but it does not necessarily appear exclusively on the same branch of the phylogeny [59] , [62] , [64] , [65] . In other words , substitutions at positively epistatic pairs of sites are likely to be temporally clustered [67] . In this paper , we exploit this idea to design an “epistasis statistic” that allows us to detect a broad class of epistatically interacting pairs of sites . In essence , for each ordered pair of sites in a protein we measure the amount of phylogenetic time that typically elapses between a substitution at the first site and a subsequent substitution at its partner site . The epistasis statistic is defined as a decreasing function of this time interval . Thus , pairs in which the substitution rate at the second site tends to be increased after a substitution at the first site will have a larger value of the statistic . We obtain the null distribution of this statistic for all pairs simultaneously , by randomly shuffling the identities of substitutions on the phylogeny . We show that the number of site pairs in the surface proteins of the human influenza A/H3N2 virus with large values of the epistasis statistic significantly exceeds the null expectation—thus , influenza surface proteins evolve under substantial positive epistasis . We characterize the epistatically interacting sites we have inferred in terms of their overall patterns of evolution , protein locations , and functional significance . For type-1 neuraminidase , we compare the identities of the epistatic sites we have inferred with those that have been experimentally verified . We discuss the implications of our results both for practical issues surrounding influenza's antigenic drift and drug resistance , and for broader issues surrounding protein evolution in general . We reconstructed the phylogenetic trees for HA and NA proteins ( subtypes H3N2 and H1N1 ) and inferred the nucleotide sequences at internal nodes by maximum likelihood as described in “Materials and Methods” . In order to detect pairs of codon sites in a protein that have evolved under positive epistasis we used the “epistasis statistic” described in “Materials and Methods” . Briefly , the epistasis statistic considers an ordered pair of sites , the first of which is called the “leading site” and the second is called the “trailing site” . The epistasis statistic tends to be large for pairs of sites in which a non-synonymous substitution at site tends to quickly follow a non-synonymous substitution at site , and for which substitutions at the trailing site occur in multiple lineages ( see schematic in Figure 1 ) . We measure time between a pair of non-synonymous substitutions as the number of synonymous substitutions that occur between them . Since we are interested in positive epistasis and would like to detect only those pairs of substitutions in which the second substitution is beneficial , we excluded all substitutions at terminal branches , because many such substitutions are likely to be deleterious . We also discarded all sites that experienced fewer than two substitutions at the internal branches ( Table 1 ) . The epistasis statistic depends on the parameter that sets the timescale over which substitutions contribute information . Pairs of substitutions that are separated by times much shorter than contribute significantly to the epistasis statistic , whereas pairs of substitutions that are separated by times much longer than do not . We set to be equal to the average time , measured in the number of synonymous substitutions , that elapses between two non-synonymous substitutions randomly sampled from a phylogeny ( see Text S1 and Table 1 ) . For each phylogeny ( H1 , N1 , H3 , and N2 ) and its corresponding value of , we computed the epistasis statistic for all qualifying ordered pairs of sites and , for each such pair , we computed the distribution of the epistasis statistic under the non-epistatic null hypothesis ( see “Materials and Methods” for details ) . We then selected all pairs of sites whose nominal was smaller or equal to 0 . 01 . In H3 , we identified 333 site pairs with a nominally significant epistasis statistic; we identified 225 such pairs in H1 , 205 such pairs in N1 , and 188 such pairs in N2 ( see Table 1 , and Table S1 ) . Examples of epistatic site pairs in HA and NA are shown in Figure 2 and Figure 3 . We computed the false discovery rate ( FDR ) as well as the overall for the observed number of significant pairs ( see “Materials and Methods” ) . Although the FDR in all proteins was high , around 60% , the observed number of nominally significant pairs was much larger than would be expected by chance ( , see Table 1 ) . Reducing the nominal cutoff somewhat reduced the FDR but also disproportionately reduced the number of inferred positives ( see Figure S1 ) . We tested the sensitivity of our method with respect to the choice of the timescale parameter , in the range from to , as well as to uncertainty in phylogeny and internal node reconstruction . The results remained qualitatively similar to those reported here ( see Text S1 , Figure S2 , and Table S4 ) . As a negative control , we performed 100 simulations in which sites evolved independently ( i . e . without epistasis ) along a given phylogenetic tree ( see “Materials and Methods” ) . In 52 out of 100 simulations , the number of significant pairs at the cutoff 0 . 01 was smaller than expected , in 47 cases this number was larger that expected , but not significantly so . In only one simulation out of 100 was this number larger than expected and significant ( ) . We are therefore confident that our method indeed detects epistatic pairs of sites , and it does not systematically report more false positives than our FDR computation indicates . Having obtained a list of pairs of sites with putative epistatic interactions ( Table S1 , Figure 2 and Figure 3 ) , we inspected the properties of these pairs , compared to an appropriate null set . In particular , we compared the true , epistatic pairs to the pairs that had appeared as nominally significant in the 400 “fake data sets” produced by permutation ( see “Materials and Methods” ) . Thus , we asked whether the pairs that we detected as epistatic differed systematically from the false positive pairs . We investigated three types of properties: the average dN/dS value at epistatically interacting sites , their location in the protein with respect to known epitopes ( for H3N2 only ) , and the distances between interacting sites . For comparison of physical and linear distances we also excluded sites that were not present in the resolved crystal structure ( see “Materials and Methods” ) . The results are summarized in Table 2 and Table 3 , and discussed below . The dN/dS values at the leading sites among the putative epistatic pairs were not significantly larger , on average , than at the leading sites among pairs identified under permutation . However , the average dN/dS value at the leading sites was less than one , which is usually interpreted as evidence of purifying selection . Therefore , without an analysis of epistasis , many of the leading sites would not have been identified as experiencing positive selection , even though they may play a critical role in facilitating adaptation in co-ordination with substitutions at their partner sites . By contrast to the leading sites , the dN/dS values at the trailing sites were significantly larger than the null expectation , and they exceeded one on average ( with the exception of N1 ) . Thus , the trailing sites exhibit the characteristic signature of positive selection [68] , even though the positive selection they experience was likely made possible ( or , at least , more likely ) by preceding substitutions at their corresponding leading sites . In other words , many of the positively selected substitutions that have occurred in HA and NA may have been facilitated by previous substitutions at epistatically interacting sites . In previous work , we have identified 25 sites in H3 at which certain specific amino acids evolved under directional positive selection [25] . Interestingly , 22 of these sites appear to be involved in epistatic interactions: 10 sites appear as leading sites , 7 sites appear as trailing , and 5 sites appear as both . We also studied the location of epistatic sites with respect to the known antigenic regions of the influenza surface proteins , for H3 and N2 . In both proteins , the leading site in an epistatic pair was more likely to fall within an antigenic epitope than under the null expectation ( and significantly so in H3 ) . The trailing site was slightly more likely to fall outside of known epitopes , despite the fact that the dN/dS ratio was typically greater than 1 at such sites . Thus , pairs of sites in which the leading site was in an epitope and the trailing site was not in an epitope , were typically overrepresented ( although not significantly so ) . This suggests that the leading sites may often be directly involved in antigenic escape and the trailing sites may subsequently compensate for deleterious ( e . g . destabilizing ) side effects of the initial mutation . In some cases , however , both the leading and trailing sites of an epistatic pair fall within epitopes ( 47% of pairs in H3 , 9% of pairs in N2 ) . In such cases , for H3 , the leading and trailing sites were significantly more likely to fall in different epitopes from each other , than expected – suggesting that substitutions across multiple epitopes may be particularly important for antigenic escape , at least in hemagglutinin of the H3 subtype . This observation reflects the widely held belief that antigenic change in hemagglutinin typically requires multiple substitutions spread across multiple epitopes [21] , [26] . How far apart are the leading and trailing sites of epistatic pairs ? Surprisingly , neither the average linear ( sequence ) distance nor the physical distance between the leading and the trailing sites in an epistatically interacting pair was significantly smaller than would be expected among false positive pairs . Finally , we investigated the timing of consecutive substitutions at the leading and trailing sites in epistatic pairs . On average , both across pairs and across consecutive substitutions , a substitution at a leading site in H3 was followed by a consecutive substitution at its corresponding trailing site 3 . 7 years later ( see Text S1 for details ) . Similarly , in H1 the mean time between consecutive substitutions was 5 . 8 years; 4 . 4 years in N1; and 4 . 2 years in N2 . In all cases , the mean time between consecutive substitutions exceeds two years – which suggests that the observation of a substitution at the leading site of a known epistatic pair may provide useful predictive value for anticipating a subsequent substitution at its corresponding trailing site , within the time-frame required for selecting a seasonal vaccine strain [69] . Our analyses of substitution patterns suggest that positive epistasis is prevalent among sites in HA and NA . However , it is important to verify experimentally that the identified pairs of sites indeed show non-additive fitness effects . Fortunately , such verification has recently been performed for two specific pairs of sites in type-1 neuraminidase . Currently circulating variants of the seasonal H1N1 subtype are resistant to the drug oseltamivir , which inhibits neuraminidase [15] . Resistance to this drug is conferred by the mutation at site 275 , which is referred to as the “H274Y” mutation in the literature [15] . However , this mutation is known to be strongly deleterious in the absence of the drug [70] . Recently , Bloom et al . demonstrated that mutations R222Q and V234M restore the drug-resistant mutant's fitness in vitro [16] , for seasonal H1N1 . They also observed that mutations R222Q and V234M were fixed in the seasonal H1N1 population prior to the emergence of the H275Y mutation , and thus they likely acted as epistatic “permissive mutations” for the emergence of drug resistance in competent viruses . Our statistical analysis of epistasis in N1 , based on patterns of sequence evolution alone , is remarkably concordant with the experimental findings of Bloom et al . In particular , our analysis indicates that sites 222 and 234 interact strongly with site 275 ( see Table S1 ) . Moreover , among the top 10 most significantly epistatic pairs in N1 there are 6 other pairs that involve the drug-resistance site 275 as the trailing site; the leading sites in these pairs are 214 , 287 , 329 , 354 , 382 , and 344 . In all cases the subsequent mutation at site 275 is . Therefore , aside from sites 222 and 234 , our analysis predicts that these six additional sites may be permissive mutations that , in combination with H275Y , produce competent , drug-resistant viruses . Although no epistasis between two of these sites ( 214 and 382 ) and site 275 was found experimentally [16] , one of the mutations ( D344N ) has subsequently been shown to help counteract the decrease in total surface-expressed activity associated with the mutant neuraminidase ( [71] and Jesse Bloom , personal communication ) , and it , along with 224 and 234 , may have played a role in the emergence of oseltamivir resistance in seasonal H1N1 viruses before 2009 . Although further experimental validation is required , the remarkable concordance between our statistical inferences and experimentally verified epistatic interactions [16] suggests that patterns of sequence evolution contain extremely useful information about a protein's fitness landscape . In the case of oseltamivir resistance , this information is highly specific and of significant import to public health . The oseltamivir resistance mutation , H275Y , was known in advance of the drug's widespread introduction . Moreover , this prior knowledge was used by Bloom et al . [16] to focus their experimental search for an epistatic partner to site 275 . Nonetheless , our method of identifying epistatic pairs from sequence data implicates site 275 – without any prior knowledge of its role in drug resistance – as extremely important in the adaptive evolution of N1 , especially in combination with sites 222 , 234 , and six other leading sites ( Table S1 ) . This demonstrates the practical , predictive power of our method for inferring the specific , epistatic interactions that shape viral adaptation . Thus , our method may , in the future , help us identify sites important for drug resistance or antigenic drift , even when no prior experimental data are available . Finally , we note that our analysis implicates sites 222 and 234 , which have been verified as important epistatic partners of the oseltamivir resistance site 275 , as significant epistatic leading sites even when we restrict our data set to those viral isolates prior to the introduction of oseltamivir . In particular , based on sequence data prior to 2001 , our method identifies sites 222 and 234 as participating in epistatic interactions with sites other than 275 ( see Table S1 ) . Thus , sites 222 and 234 may be structurally important and experience epistatic interactions even in the absence of selection for oseltamivir resistance . We have developed a statistical method to detect positive epistasis between pairs of sites in a protein , based on patterns of thoroughly-sampled sequence variation . The essential idea underlying this method is simple: a substitution at one site should rapidly follow a substitution at another site if the sites interact epistatically . We applied this method to identify putative epistatic pairs in the influenza surface proteins hemagglutinin and neuraminidase , and we found a highly significant number of interacting pairs . We characterized the properties of the leading and trailing sites identified as epistatic . Finally , we validated our approach by comparison to experimentally verified epistatic interactions in neuraminidase , with significant implications for public health . This study sheds some light on methodological and empirical questions in molecular evolution generally , as well as practical questions about influenza viral evolution in particular . Methodologically , it is instructive to compare our approach to identifying epistasis with other techniques in the literature . Over very long timescales , interacting sites in a protein have been identified by inspecting multiple sequence alignments , ignoring the phylogenetic relationship among the sequences being compared . Such an approach is justifiable over timescales so long that each site may be treated independently , and indeed it has proven successful at identifying epistasis in proteins conserved across all domains of life [51] . However , such techniques are not justified for shorter timescales , because correlations between sites may arise simply as the result of linkage and shared ancestry [58] . Although techniques exist to control for phylogeny in such tests [52]–[54] , it is preferable to leverage the phylogeny in the design of a more powerful statistic for epistasis – which is the approach we have taken here . Even among the techniques that account for phylogeny , methods differ in their power to detect epistasis . Some methods will be more powerful in some contexts , and others in other contexts – depending upon the structure of epistasis among sites , the selection coefficients involved , and the density of sampling . Most existing methods that utilize phylogenetic information assume that epistatic substitutions will co-occur along the same branch of the phylogeny [42] , [60] , [63] , [64] . This assumption will not always be met , however , if the selective advantage conferred by a substitution at the trailing site is only moderate; in such cases , substitutions at trailing sites will occur at an accelerated rate but they may likely fall on subsequent branches in the phylogeny . To demonstrate this point , we applied the method of Poon et al [64] , implemented in the HyPhy package [72] , to the same data set of influenza sequences . That method detected 4 to 10 times fewer epistatically interacting pairs of sites than our method did , at the same false discovery rate ( see Text S1 and Tables S2 and S3 ) . Importantly , the method by Poon et al . failed to detect epistatic interactions between sites 222 and 234 and the drug-resistance site 275 in neuraminidase subtype N1 , even though those pairs were highly ranked by our method and those epistatic interactions were confirmed experimentally . Although a thorough comparison between various existing methods is beyond the scope of this paper , we believe that the additional power of our method to detect epistasis in the influenza data arises because we allow for time lags between substitutions at interacting sites . The epistasis statistic developed here is admittedly ad-hoc , compared to systematic , likelihood-based methods for jointly inferring phylogeny and epistasis under Markov substitution models [73]–[75] . At the same time , the vast dimensionality associated with substitution models incorporating pairwise epistasis , of order for a sequence of length , is daunting; whereas the frequentist statistic defined here seems to perform quite well . The strong performance of our approach likely arises from our ability to infer ancestral states reliably , due to the high-resolution sampling of influenza sequences . Our method has several important shortcomings . One drawback is that it requires a large number of substitutions per site in order to discriminate between truly interacting site pairs and pairs that sustain substitutions in close succession just by chance . Moreover , even if the protein evolves rapidly , as influenza surface proteins do , the false discovery rate is still very high . Our method will likely perform much worse for proteins that evolve slowly or have been sampled sparsely . Two other concerns are problematic for our approach , as well as most other methods of detecting epistasis from phylogenetic data . Such approaches generally suffer from the inability to weed out spuriously epistatic pairs , which leads to high false discovery rates . There are at least two sources of spuriously significant pairs: hitchhiking and coordinated temporal variation in selection pressures across sites . Imagine , for example , that sites and interact epistatically and that multiple substitutions at site in independent lineages rapidly follow a single substitution at site . Then site pair would be detected by our method . However , if the variant that carries the leading mutation at site also , by chance , happens to carry a mutation at site ( which is not epistatic with ) , then mutation hitchhikes to fixation together with mutation and so the site pair may also be detected as epistatic . In fact , mutation at site may be advantageous while mutation at site may be a neutral or slightly deleterious mutation that hitchhikes to fixation together with , but then “permits” the beneficial mutation at site . It may be possible to reduce the false discovery rate by designing statistics that consider only those site pairs for which consecutive substitutions involve multiple independent substitutions at the leading site as well as the trailing site . Coordinated temporal variation in selection pressures across sites is another source of potential false positives under this and other tests of epistasis . Consider , for example , sites 391 and 73 in H3 illustrated in Figure 2 . Substitutions at site 73 appear to quickly follow substitutions at site 391 in the early 1990's . Apart from epistasis , the clustering of substitutions at these two sites could be explained if both sites independently experienced positive selection during this time period , and otherwise negative selection . However , if this explanation were the dominant one for the observed clustering of substitutions , then , for each nominally significant ordered pair of sites , we would expect its inverse pair also to be nominally significant , on average . Yet , we do not find a single nominally significant pair whose inverse pair is also nominally significant , even though consecutive substitutions do occur in the direct and reverse order ( see for example , Figure 2 and Table S1 ) . It is unlikely that this observation is caused by insufficient sampling . Indeed , in H3 , there are typically more than 6 substitutions ( at internal branches ) at either a leading or trailing site in an identified epistatic pair , and similarly for the other proteins . Moreover , many sites appear in our lists as both leading and trailing . Thus , leading and trailing sites exhibit similar number and pattern of substitutions , and there is plenty of power to detect a significant epistasis statistic in both directions . This suggests that the excess of significant pairs we observe is likely caused by epistasis , rather than coordinated temporal variation in selection pressures . Another shortcoming of our method is that it aims to detect epistasis only between pairs of sites , whereas interactions among residues in a protein are certainly more complex . This may be a cause of our large false discovery rate . Imagine , for example , three sites , , , and , such that pairs and interact epistatically , but the pair does not . If substitutions at site quickly follow substitutions at site and if substitutions at site quickly follow substitutions at site , our method may detect the pair as epistatic , even though there is no direct epistatic interaction between these sites . Indeed , in our list of putatively epistatic pairs , we find 133 of such “circular” triplets in H3 , 41 triplets in N2 , 81 triplets in H1 , and 71 triplets in N1 . In order to discriminate truly epistatically interacting site pairs from spurious pairs , it may be possible to modify the Bayesian graphical models recently used for detecting epistasis in HIV [64] , [66] to incorporate a time lag between consecutive substitutions . Finally , although our method detects epistatic interactions between pairs of sites , it does not determine which specific mutations at those sites were epistatic . Extending our permutation technique to incorporate the information about specific mutations may prove difficult , but in many cases it is unnecessary . Often we can a posteriori identify the specific mutations that led to a significant value of the epistasis statistic for a pair of sites . For example , the drug-resistance site 275 is identified as trailing with many leading sites in N1 ( see Table S1 ) , but the specific substitutions at site 275 are all in fact identical: H275Y . Methodological issues aside , our results on epistasis in HA and NA have several important practical implications for our understanding of influenza evolution . We have demonstrated that our method reliably infers a critically important oseltamivir resistance site , as well as the associated leading sites at which initial mutations are required for the production of a viable , drug-resistant virus . Remarkably , we can identify some of the leading sites ( 222 and 234 ) even when restricted to sequence data prior to the introduction of the drug . This degree of specificity and accuracy may prove helpful in preparing for resistance to other drugs that may be developed , or in predicting the emergence of oseltamivir resistance in the recent type-1 swine neuraminidase responsible for the 2009-10 influenza pandemic . In addition to NA , we have also detected substantial amounts of epistasis in HA , including in the known epitopes , likely associated with antigenic drift . Knowledge of specific pairs of sites that interact epistatically in HA may improve our ability to predict future antigenic variants , and thus to calibrate vaccine strain choices accordingly . Previous studies of HA antigenic evolution have focused almost exclusively on those sites with the strongest signatures of positive selection , e . g . elevated dN/dS ratios [8] , [21] , [22] . However , our results suggest that this approach will inevitably miss many sites of genuine importance to adaptation , and will implicate others that are not directly involved in antigenic escape . In particular , we have seen that the leading site of an epistatic pair often falls within an epitope , but it also often exhibits . In contrast , the trailing site typically falls outside of an epitope and it exhibits significantly elevated . This observation appears counter to our expectation that epitopic sites have elevated dN/dS values and non-epitopic sites have depressed dN/dS values . However , not all epitopic sites experience elevated dN/dS values at all times because different epitopes may be immunodominant at different times [76] , [77] . Thus , an average dN/dS value at a site may be well below 1 even if this site occasionally evolves under strong immune selection [78] . The patterns of epistatic interactions we have detected suggest the the following speculative model for the evolution of influenza surface proteins . If an epitope is immunodominant at a certain period of time , the pressure for antigenic escape is so strong that the leading site , of antigenic importance , substitutes despite a negative side-effect ( e . g . diminished protein stability or function ) . This side-effect is subsequently compensated by a substitution at another , relatively unconstrained , site , in an epitope or not . It is possible that such unconstrained sites may act as “global suppressors” , i . e . , compensate the destabilizing effects of many mutations [79] , [80] . If there is a constant need for compensation ( caused by antigenically important mutations ) , such compensating sites will continually evolve under positive selection and will exhibit . Under this scenario , the roles of the two sites would both be misinterpreted by an analysis based on dN/dS alone that neglects epistasis . In particular , our observations imply that mutations at sites with may sometimes be extremely important for antigenic adaptation , even though they have been largely ignored in compilations of antigenically relevant sites [8] , [11] , [21] . Conversely , some mutations at sites with may be unimportant for antigenicity per se , but are positively selected simply to compensate for prior antigenic escape mutations with deleterious side effects . Another potential mechanism of epistasis in influenza surface proteins could be the dynamic balance between mutations that simultaneously influence receptor binding avidity and antigenicity , as suggested recently by Hensley et al . [81] . Some of the epistatic pairs that we detect consist of an apparently neutral but “permissive” mutation at the leading site followed by a highly advantageous mutation at the trailing site , such as the pair of mutations V234M and H274Y in N1 previously identified by Bloom et al [16] and also detected by our method ( Table S1 ) . This observation is consistent with the idea that neutral or nearly neutral substitutions can facilitate adaptation at partner sites that might not otherwise have been available—a concept that has received much attention in theoretical studies of adaptation [82] , [83] and of influenza evolution in particular [84] . Finally , we have observed that epistatic residues do not tend to be significantly closer to each other in the folded protein structure than would be expected by chance – a result that we expect to hold generally , and which suggests that structural influences on epistasis are probably not as straightforward as simple proximity of residues . In the future , it will be important to investigate , by computation or experiment , whether epistatic partner sites are compensating for protein stability even if they are distant from each other in a folded protein structure . We downloaded all HA and NA coding region sequences of human influenza A virus subtypes H1N1 and H3N2 that were available in the NCBI's Influenza Virus Resource [85] in June 2010 . The amino acid sequences were aligned using Clustal W ver . 1 . 83 [86] and the alignments were reverse translated using PAL2NAL [87] . Occasional gaps in the alignments were filled if more than 70 percent of sequences agreed on the nucleotide at the gap position; otherwise the sequence with a gap was excluded from further analysis . To test some aspects of our method , we used a smaller HA data set ( subtype H3N2 ) which was downloaded in April 2009 . To investigate whether our method detected any of the known epistatic site pairs in type-1 NA prior to the introduction of oseltamivir , we also used a truncated data set of N1 sequences with all sequences isolated subsequent to 2001 removed . All used alignments are available upon request . In computing the epistasis statistic we excluded all substitutions at terminal branches , and we discarded all sites that experienced fewer than 2 substitutions at the internal branches . We also downloaded the HA and NA crystal structures from the RCSB Protein Data Bank . In computing the linear ( sequence ) and physical distances between residues we excluded all residues that were not resolved in the crystal structures . We used the distance between the alpha-carbon atoms as a proxy for the physical distance between residues . We reconstructed the maximum likelihood phylogenetic trees for HA and NA using PHYML [88] under the GTR substitution model with the four-category discrete approximation of the gamma distribution for the substitution rates . We reconstructed the nucleotide sequences at the internal nodes of the phylogeny using maximum likelihood algorithm in PAUP* 4 . 0b10 [89] . For each codon site , we identified whether it experienced at least one synonymous and/or non-synonymous substitution on each branch of the reconstructed phylogeny . In those rare cases in which a codon experienced more than one substitution of the same kind ( synonymous or non-synonymous ) on a branch , we did not record the number of substitutions , in order to simplify computations . Consider an ordered pair of sites in the protein of interest . In order to detect a positive epistatic interaction for this pair , we designed a statistic that detects the acceleration of non-synonymous substitutions at site , which we call the trailing site , after the occurrence of a non-synonymous substitution at site , which we call the leading site . First , we obtained a strict temporal order in which non-synonymous substitutions at sites and occurred on the phylogenetic tree . Such an order is not actually known if the phylogeny contains one or more branches on which both sites have experienced a non-synonymous substitution . We say that such branches are temporally unresolved with respect to the pair . Since we do not know in which order the sites in the pair have experienced substitutions on a temporally unresolved branch , we assume that both orders are equiprobable . If there are branches on the phylogenetic tree that are temporally unresolved with respect to the pair , there are a total of equally likely distinct strict temporal orders of substitutions on the tree with respect to this pair . Next , for each strict temporal order of substitutions , , at sites and , we find all pairs of substitutions that are consecutive along the tree . Substitution at the trailing site ( in this case ) and substitution at the leading site ( in this case ) form a consecutive pair if has occurred in the lineage ancestral to and no other substitution at either site has occurred in the lineage between them . This notion is illustrated in Figure 1 . If is a consecutive pair , we also say that substitution is consecutive to substitution . For each consecutive pair , substitution is called initial and substitution is called subsequent . Before defining the epistasis statistic we introduce some notation . We denote the fact that branch is ancestral to branch by ( “ precedes ” ) or by ( “ follows ” ) ; if and denote the same branch , we naturally write . We denote the number of synonymous substitutions that occurred on branch by . We measure time between the initial substitution and the subsequent substitution of a consecutive pair as the expected number of synonymous substitutions that occurred between them . More precisely , if substitutions and occurred on branches and respectively , then ( 1 ) The sum in this expression is taken over all branches on the lineage connecting branches and . Note that can be zero if no synonymous substitutions occurred between substitutions and . Let denote the set of all consecutive substitution pairs at site pair found on the phylogenetic tree with the order of substitutions . We define the epistasis statistic as ( 2 ) where is a time-scale parameter that we specify in the “Results” section . The choice of an exponential function of is arbitrary . We expect that any monotonic decreasing function of would yield similar results . Note that if , i . e . if sites and never experience a non-synonymous substitution on the same branch , the strict temporal order of substitutions with respect to the ordered pair is unique . In this simple case , the epistasis statistic is equivalent to If we take the time-scale parameter to be infinite , the statistic simply equals the number of consecutive substitutions for the ordered pair . We also define if all sets are empty for the pair . In other words , the epistasis statistic is zero for pairs of sites that never experience consecutive substitutions . The value of the epistasis statistic is large if substitutions at the trailing site often follow substitutions at the leading site and if the time-lag between initial substitutions at the leading site and subsequent substitutions at the trailing site is typically small ( compared to ) . We therefore expect that pairs of sites that evolve under positive epistasis will have a larger value of the epistasis statistic . If there were no epistatic interactions between sites and no temporal variation in selection pressures then we would expect the non-synonymous substitutions at each site to be distributed randomly on the phylogenetic tree . In order to obtain the distributions of the epistasis statistic under this null hypothesis for all ordered pairs of sites , we utilize the following straightforward permutation procedure . We shuffle all non-synonymous substitutions on the phylogenetic tree while keeping two sets of marginal quantities preserved: ( a ) for each branch of the phylogeny , we preserve the number of non-synonymous substitutions that occurred on that branch and ( b ) for each site , we preserve the total number of non-synonymous substitutions that occurred at that site on the tree . Condition ( a ) ensures that any possible temporal biases in the sampling of viral isolates , which would apply equally to all sites , are preserved in the null distribution . Condition ( b ) ensures that the overall selective constraint on each site is preserved . Synonymous substitutions are unaffected by this shuffling procedure . Although this permutation procedure is conceptually simple , its computational implementation is challenging . A priori , it is unclear how to efficiently sample the space of possible substitution configurations while preserving the aforementioned marginals . This problem can be rephrased as follows . We can represent the phylogenetic tree with non-synonymous substitutions as an matrix , where is the number of branches on the phylogeny and is the number of sites , so that each cell in the matrix is either 1 or 0 depending on whether or not the given site experienced a non-synonymous substitution on the given branch . Thus , we would like to randomly permute the entries of this matrix while preserving the row and column sums . This problem is equivalent to the problem of obtaining the null distribution for the matrix of associations between individuals across a set of observations , which has been extensively studied in the ecology literature [90]–[92] . The method typically employed in ecology to sample the space of matrices that satisfy the constraint on the marginals is called the “swap method” and is based on the idea of swapping the entries of certain specific submatrices in a way that does not violate the constraints . This method , although computationally efficient , generates matrices that are not independent [91] . An alternative “fill method” permutes the matrix entries and simply discards those resulting matrices that do not satisfy the constraints [92] . This method can be prohibitively computationally expensive if many matrices are discarded , but it guarantees independent sampling . We employed the “fill method” and found that only about between 0 . 05% and 5% of matrices are accepted , yet this did not present a serious computational limitation . We generated valid permutations per protein , which required about 10 minutes on a desktop computer . We compute the value of the epistasis statistic and its associated nominal for many thousands of site pairs . We therefore need to quantify the fraction of false positives among the observed nominally significant pairs [93] . Because the values of the epistasis statistic for different site pairs are not independent , we estimate the distribution of the number of false positives in the data through bootstrap by designating 400 out of permutations generated by the procedure described above as “fake data sets” . For each such fake data set , which represents one draw from the null hypothesis , we computed the number of nominally significant site pairs . This allowed us to estimate the full distribution of the number of false positives in the data . In particular , we recorded the expected number of false positives in our data , which is typically referred to as “false discovery rate” ( FDR ) , and overall for the total number of positives actually observed . To ensure that our method does not detect epistatic interactions when there are none , we have performed detailed simulations of sequence evolution along a phylogenetic tree , with independent sites , as described in [25] . Briefly , the simulation algorithm takes as input a phylogenetic tree with branch lengths equal to the number of nucleotide substitutions and the nucleotide sequence at the root of the tree; it outputs the nucleotide sequences at all internal and terminal nodes . The sequence at a node is generated recursively , given that the sequence at the parental node is already known , using the following stochastic procedure . If the branch length connecting the focal node to the parental node is , then mutations are randomly distributed along the parental sequence proportionally to the entries in the nucleotide mutation matrix and the codon-specific dN/dS values . We used the HA phylogenetic tree to perform this simulation as well as to infer the nucleotide mutation matrix and the codon-specific dN/dS values [25] . Using this simulation algorithm , we generated 100 independent sequence data sets . On each of them , we performed the analysis described above with permutations , 100 of which were considered “fake data sets” .	Epistasis describes non-additive interactions among genetic sites: the consequence of a mutation at one site may depend on the status of the genome at other sites . In an extreme case , a mutation may have no effect if it arises on one genetic background , but a strong effect on another background . Epistatic mutations in viruses and bacteria that live under severe conditions , such as antibiotic treatments or immune pressure , often allow pathogens to develop drug resistance or escape the immune system . In this paper we develop a new phylogenetic method for detecting epistasis , and we apply this method to the surface proteins of the influenza A virus , which are important targets of the immune system and drug treatments . The authors identify and characterize hundreds of epistatic mutations in these proteins . Among those identified , we find the specific epistatic mutations that were recently shown , experimentally , to confer resistance to the drug Tamiflu . The results of this study may help to predict the course of influenza's antigenic evolution and to select more appropriate vaccines and drugs .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "evolutionary", "biology/microbial", "evolution", "and", "genomics", "evolutionary", "biology/evolutionary", "and", "comparative", "genetics", "public", "health", "and", "epidemiology/infectious", "diseases", "evolutionary", "biology", "public", "health", "and", "epidemiology/immunization" ]	2011	Prevalence of Epistasis in the Evolution of Influenza A Surface Proteins
The AID / APOBEC genes are a family of cytidine deaminases that have evolved in vertebrates , and particularly mammals , to mutate RNA and DNA at distinct preferred nucleotide contexts ( or “hotspots” ) on foreign genomes such as viruses and retrotransposons . These enzymes play a pivotal role in intrinsic immunity defense mechanisms , often deleteriously mutating invading retroviruses or retrotransposons and , in the case of AID , changing antibody sequences to drive affinity maturation . We investigate the strength of various hotspots on their known biological targets by evaluating the potential impact of mutations on the DNA coding sequences of these targets , and compare these results to hypothetical hotspots that did not evolve . We find that the existing AID / APOBEC hotspots have a large impact on retrotransposons and non-mammalian viruses while having a much smaller effect on vital mammalian genes , suggesting co-evolution with AID / APOBECs may have had an impact on the genomes of the viruses we analyzed . We determine that GC content appears to be a significant , but not sole , factor in resistance to deaminase activity . We discuss possible mechanisms AID and APOBEC viral targets have adopted to escape the impacts of deamination activity , including changing the GC content of the genome . The AID/APOBEC family of cytidine deaminases have important functions in both intrinsic and adaptive immunity . AID is expressed primarily in germinal center B cells as part of the adaptive immune response [1] , whereas the APOBECs act primarily in the intrinsic immune response in various cell types ( reviewed in [2] ) . These mutagenic enzymes act mostly upon single-stranded DNA [3] converting Cytosine in DNA or RNA to Uracil , or for AID , a methylated Cytosine to Thymine in vitro [4] . In the absence of further editing , the resulting U-G mismatch in DNA will often be replicated over to create a C to T transition mutation [5] . AID in particular relies on downstream non-canonical DNA repair pathways to introduce further mutations [6] . Although AID-mediated mutations occur at rates approximately 106 higher than background [7 , 8] , they occur almost entirely in the Immunoglobulin genes that code for the B-cell receptor [9 , 10] . At the same time , DNA editing mechanisms can potentially target the host genome ( reviewed in [11] ) and evidence of AID and APOBEC-mediated mutagenesis has been identified in many human cancers [12] . Thus , there appears to be a tradeoff between the benefits and the potential for self-damage of APOBEC-mediated mutations , yet the details on how this tradeoff is achieved both on a biochemical and evolutionary level are still not well understood . AID has been proposed to be the most ancestral member of the APOBEC family [13] and is found in all jawed vertebrates . Duplication and diversification of AID have created the APOBEC family , which edit both RNA and DNA . All mammals additionally have at least 4 APOBEC genes , APOBECs1-4 [14 , 15] . APOBEC1 deaminates the mRNA of the Apolipoprotein B gene ( ApoB ) at a specific site to introduce a stop codon in humans and mice [16 , 17] . Zebrafish APOBEC2 appears to maintain embryonic development and other functions in development including retina regeneration [18–20] , while the role of human APOBEC2 is currently unknown . The function of the APOBEC4 gene also still not known [2] . APOBEC3 has diversified considerably during mammalian evolution . Whereas mice have a single APOBEC3 , humans have seven APOBEC3 variants ( A-H ) [21] . The APOBEC3 genes participate widely in innate immunity by mutating retrotransposons [22] , exogenous viruses [23] , and endogenous viruses [24] . Human APOBEC3s differ in their restriction capabilities . Perhaps the best understood example of APOBEC restriction is against HIV , which is targeted by human APOBEC3G . The HIV genome in turn encodes a counter-defense in the form of the vif gene , which encodes a protein that targets APOBEC3G for degradation . Experiments in vitro using vif-deficient viruses showed that HIV was highly restricted and mutated by APOBEC3G [25] , and later APOBEC3F [26] . Although APOBEC deamination appears effective in inhibiting HIV in the absence of vif , other studies suggest that HIV has evolved its own deamination hotspots , thus co-opting APOBEC mutagenesis to increase the likelihood of beneficial mutations [27] . Thus , there is a conflict between HIV adopting APOBEC3G mutation via hotspots compared to the ability of APOBEC to induce lethal hypermutation , although it is not currently established to what extent HIV favors either . Anti-viral activity for human APOBEC3G and other mammalian APOBEC3s has been observed across a wide range of viruses including Human Papillomaviruses such as HPV-16 [28] , Adeno-associated virus ( AAV ) [29] , Torque Teno Virus ( TTV ) [30] , Human Herpes-Simplex Virus 1 ( HSV-1 ) [31] , Human T-lymphotropic virus 1 ( HTLV-1 ) [32] and Simian Foamy Virus ( SFV ) [33] . AID has been shown to deaminate the Hepatitis B Virus ( HBV ) [34] and C virus ( HCV ) in vitro [35] . In addition , APOBECs are thought to restrict human endogenous retroviruses ( HERV ) [24] . Furthermore , human APOBEC3s are capable of mutating the Long Terminal Repeat ( LTR ) retrotransposons such as the LINE1 ( L1 ) element [36] . In particular , human APOBEC3B , human APOBEC3C , and human APOBEC3F strongly inhibit L1 retrotransposition in vitro , while human APOBEC3G and human APOBEC3H weakly inhibit this same activity [37] . This phenomenon has also been demonstrated with APOBEC3s in reptiles [38] . Homologs of human APOBEC3B have been shown to restrict L1 retrotransposition in other primates as well , where greater L1 activity was correlated with lower human APOBEC3B expression levels and diversity [39] . AID and the human APOBEC3s differ not only by their targets which include exogenous and endogenous viruses , but also their preferred deamination context in DNA or RNA . The relation between these preferred contexts and their biological targets is still poorly understood . Although members of the APOBEC family all mutate C to U , individual preferences vary . AID , for example , preferentially deaminates the hotspot WRC ( W = A/T , R = A/G , C is mutated ) while avoiding the coldspot SYC ( S = G/C , Y = C/T ) . APOBEC3G , on the other hand , preferentially mutates CCC , an AID coldspot [6 , 40] . Additionally , in the context of mutations targeting HERV , it has been shown that the +1 position is also important for APOBEC3F and APOBEC3G leading to more complex motifs , respectively TTCA and CCCA [41] . Preferences for AID/APOBEC targeting are summarized in Table 1 . Interestingly , individual APOBEC preferences can be changed via minimal amino acid changes in a hotspot recognition loop [42] . The AID/APOBEC enzymes must maintain a delicate balance of ensuring deamination is sufficient for key functions such as antibody diversification and restriction of viral and retrotransposon activity , while avoiding detrimental targeting of the host genome . Achieving this balance is accomplished via several mechanisms . Firstly , individual AID/APOBECs are expressed in distinct cell types . For example , AID is expressed primarily in Germinal Center B cells , whereas APOBEC3G is expressed in several cell types , particularly T cells [53] . Additionally , there may be intracellular regulation . Thus , in Germinal Center B-cells , AID is mainly found in the cytoplasm but its activity levels , are modulated in part by its Nuclear Localization Signal ( NLS ) [54] . Although APOBEC3B also contains an NLS domain , this type of regulation has still not been investigated for other APOBECs [55] . Although AID is actively targeted to the antibody ( Immunoglobulin ) loci in B-cells , it also has the potential to mutate many “off-target” locations throughout the entire genome [56] . These targets include known oncogenes such as Bcl6 , Myc , and others . Mistargeted AID is also linked to poor prognosis in Chronic lymphocytic leukemia ( CLL ) [57] and other B Cell Lymphomas [58] . Furthermore , the mutation pattern of human APOBEC3B , NTC , has been strongly correlated to so-called “kataegis” mutations in breast cancer , characterized by short inter-mutation distances [12] . Although the mechanism of action and physical structure of several APOBECs are reasonably well characterized , it is not well understood how their specific hotspots were established during APOBEC evolution . We investigate the hypothesis that the APOBECs evolved their mutational preferences to control the impact of deaminase activity by evolving preferences that induce the most damage to their intended targets such as exogenous virus genomes , while causing the least damage to the host genes . We describe a bioinformatics-based approach examining the impacts of motifs , both known and hypothetical , on various genomes . We demonstrate that the current APOBEC preferences are shaped by their ability to restrict their targets , though these observed vulnerabilities may be mediated in large part by the GC content of the native sequence . Our motivation is to quantify the impact of deamination at motifs , both known and hypothetical , across many different potential targets . Previous structural descriptions of APOBEC mutagenesis have indicated the role of a recognition pocket in the -2 and -1 positions on the single-stranded DNA substrate preceding the targeted Cytosine [59] . We therefore examined the effects of mutation at each of the 16 different NNC hypothetical hotspots , referred throughout as NNC motifs ( N = any nucleotide ) to NNT , consistent with the AID/APOBEC deamination mechanism in the absence of further repair . We measure a genome’s susceptibility to mutation by extending a method first introduced by Karlin [60] ( see: “Calculating susceptibility” section in Methods for a detailed definition , including discussion of necessary controls ) . Briefly , the model we use assesses the impact of a particular hotspot or motif ( for example , AGC ) on a given gene coding sequence by first counting the number of observed motifs in both orientations , then counting the number of nonsynonymous ( changing the amino acid ) mutations on the sequence that would occur by mutating C to T in every observed motif ( or G to A in the reverse orientation ) ( Fig 1A ) . In quantifying a coding sequence’s susceptibility to mutations in a particular motif , we assumed that nonsynonymous mutations have more potential to be deleterious . Given the complexity of nucleotide sequence evolution , there are many factors this simple model does not account for , such as the GC content of the genome , or the impact of hypothetical mutations other than from C to T . To address these considerations , we also evaluate several alternative models ( see: “Alternative models and controls corroborate observed susceptibilities” section below ) . To illustrate our method , we show an example in Fig 1 . Here , the target sequence has 4 AGC motifs , with two on the forward strand , and two on the reverse strand ( appearing as the reverse complement of AGC , GCT ) of which 3 cause nonsynonymous mutations and 1 is synonymous when mutated from C/G to T/A . This sequence therefore has a nonsynonymous mutation fraction of 0 . 75 , equivalent to the probability that a mutation at an AGC site causes an amino acid change ( Fig 1B ) . This fraction was then compared to a null model consisting of 1000 randomized sequences that preserve the amino acid sequence , but where synonymous codons are chosen randomly . For each randomized sequence in the null model , the nonsynonymous mutation fraction was calculated . Under the model , if a gene is susceptible to a particular NNC motif , then it would have a higher incidence of nonsynonymous mutations than the majority of randomized versions from the null model . We therefore refer to the percentile of randomized sequences having a lower nonsynonymous mutation fraction than the wild-type sequence as that gene’s susceptibility for a given motif ( Fig 1C ) . It is in fact equivalent to a P-value for susceptibility . Thus , a particular motif is susceptible to mutation if the percentile is close to 1 . Conversely , a gene with a low susceptibility , closer to 0 , can be considered resistant to that same motif . As an example , if we are considering AGC to AGT mutations , if a gene has a nonsynonymous mutation fraction of 0 . 75 but in the null model , 1 out of 4 permutations have an even lower nonsynonymous mutation fraction , this gene’s sequence would be at the 25th percentile for the nonsynonymous mutation fraction of the AGC motif ( Fig 1D ) . We next examined how mutational susceptibility of retrotransposons and viral genomes was affected by different APOBEC preferences . We defined a gene set as a group of genes with a common biological function , for example , a set of housekeeping genes , or the genes in a particular virus . Starting with the individual susceptibility measures for each gene , we calculated the average susceptibility for each gene set ( Fig 2A ) . As an example , Fig 2B shows the susceptibility values for the Epstein-Barr Virus ( EBV ) , a herpesvirus that is tropic to B-cells , potentially exposing it to AID , which has a strong preference for mutating AGC motifs . We reasoned that under a null model , if susceptibility is equivalent to a P-value , then random percentiles would assume a uniform distribution from 0 to 1 , and would have a median and mean of 0 . 5 . In contrast , the average susceptibility of the EBV genome is approximately 0 . 2 and for the set of human housekeeping genes it is 0 . 4 ( Fig 2C ) , suggesting human housekeeping genes are slightly less resistant to the impact of AGC deamination . As each coding sequence susceptibility score can be treated as a P-value , to assess the susceptibility of each gene set statistically we combined the individual P-values using an unweighted Z-transform approach , which is more powerful than the standard Fisher method [61] . We found this approach was capable of detecting significance even for small gene sets , such as HIV or Adeno-Associated Virus 2 ( AAV2 ) , which have only 7 and 8 genes respectively but show significant results for several motifs ( S1 Table ) . In the example of Fig 2B , the susceptibility of the EBV genome to the AGC motif was statistically lower than a null median of 0 . 5 ( P < 10−49 , S1 Table ) and in general EBV displays statistical resistance to 13 of the 16 NNC motifs we examined ( using a cutoff of P < 3 . 125 × 10−3 for significance , representing Bonferroni corrected P = 0 . 05 ) . We applied the same z-score method to determine the statistical vulnerability of a gene set , defining it as the combination of the scores ( 1-susceptibility ) to obtain a P-value ( S2 Table ) . To quantify the impact of deamination on various genomes we analyzed a diverse set of genomes , including viruses previously reported as AID or APOBEC targets , sets of housekeeping genes from mammalian ( mouse and human ) genomes and , as controls , genomes from viruses of non-vertebrate hosts that have evolved in the absence of AID/APOBECs ( see Methods section “Data Sources” , and Table 2 ) . For each of these gene sets we were interested in quantifying the impact of C>T mutations at the reported hotspots for AID/APOBEC ( Table 1 ) . Since we wanted to gain insight into why these particular hotspots evolved , we considered all motifs of the form NNC ( N = any nucleotide ) , which includes 10 hotspots targeted by AID/APOBEC and 6 hypothetical hotspots such as GCC that have not evolved specificity within the AID/APOBEC family . We measured susceptibility values ( as described above ) for every gene within each gene set , calculating the mean susceptibility in each case . This was repeated for all 16 NNC motifs , thus obtaining 16 susceptibility values for each gene set . We observed that one of these gene set clusters ( indicated by the red box labeled “Resistant” in Fig 3 ) has high overall resistance ( i . e . , low susceptibility ) to NNC motifs . This resistant cluster includes many viral genomes , such as Human Herpesvirus 1 [31] , and hepatitis C Virus [35] . Curiously , this cluster includes the Epstein-Barr Virus , suggesting that this virus may have evolved resistance to mutations by both APOBECs and AID . Although to our knowledge no studies have demonstrated a direct mechanism of AID restricting EBV experimentally , the virus is B-cell tropic and would presumably be exposed to AID deamination in germinal centers and to an extent in extrafollicular compartments [62] . Gene sets of host ( mouse and human ) housekeeping genes also show overall resistance to mutations in NNC motifs . Housekeeping genes will , by definition , be co-expressed with AID and those APOBEC genes wherever they are expressed . Although little is known about the mechanisms by which our host genomes avoid APOBEC mutations , the evidence from AID in B-cells shows that mutations indeed occur genome-wide but are repaired , although the repair process is imperfect [56] . Our analysis here suggests that this mutational behavior ( perhaps including APOBEC ) may have created selective pressures on the genome to minimize this potentially hazardous activity . The cluster of gene sets in the blue box of Fig 3 and labeled as Vulnerable had higher mean susceptibility scores across all NNC motifs . The cluster includes a human endogenous retrovirus ( HERV ) , the human ( LINE-1 ) and mouse retrotransposon ( LINE-1 and MusD ) coding regions , as well as the control genomes , which include viruses of invertebrate hosts ( oyster , plants ) that would not have co-evolved with AID or APOBEC . Surprisingly , there are several virus gene sets that are targeted by APOBEC and that we did not initially expect to be vulnerable , including HIV , Simian Foamy Virus ( SFV ) , Hepatitis B Virus , Murid Herpesvirus 4 and Human Papillomavirus 16 . HIV’s apparent vulnerability is surprising since HIV is a well-known target of human APOBEC3G . However , as described in the Introduction , HIV encodes vif , an APOBEC3G inhibitor , which may obviate the need for sequence-level avoidance . Other unknown defense mechanisms for these targeted viruses may also explain their apparent vulnerability . Mouse APOBEC3 is incapable of impairing MHV68 ( Murid Herpesvirus 4 ) viral replication , for example , whereas several human APOBEC3s are capable of restriction , suggesting an alternate , unknown anti-mAPOBEC defense may be present [63] . Furthermore , recent work on HIV has shown that APOBEC3G may in fact help the virus mutate and create escape variants [64] , which may in turn explain its relatively high level of susceptibility . A similar explanation may hold for human papillomavirus ( we analyzed HPV16 ) , which recent studies have shown to be extensively mutated in vivo and in vitro [28] . In light of these phenomena , our results suggest that even if a virus targeted by APOBEC is found to be susceptible to APOBEC hotspots , the virus itself may contain an alternate method of escape such as vif , subversion of APOBEC for evolutionary purposes , or other factors . Considering now the hierarchical clustering results for the 16 NNC motifs ( vertical dendrogram of Fig 3 ) we observed that many existing ( or “canonical” ) AID/APOBEC hotspots , including 7 of the 10 AID/APOBEC hotspots are contained within the same cluster ( right-most 7 motifs in Fig 3 , boxed in orange ) . This cluster contains 3 of the 4 canonical AID hotspots AGC , AAC , and TGC ( only excluding TAC ) , as well as 3 of the canonical human APOBEC3B NTC hotspots ( ATC , GTC , and TTC , only excluding CTC ) , and the mouse APOBEC3 hotspot TCC , but excludes the human APOBEC3G hotspot CCC . In light of the contrast observed , the existence of the two clusters , resistant and vulnerable , suggests that AID/APOBEC hotspots cause high mutational damage to exogenous viruses and native transposable elements . At the same time , we have observed that many of the genes targeted ( intentionally in the cases of viruses and unintentionally in the case of host housekeeping genes ) show resilience to the impact of these mutations at hotspots . This resilience may have evolved as an avoidance strategy to minimize mutations if the hotspots that APOBEC evolved do not cause high collateral damage to the host genome . GC bias is an important evolutionary constraint in many viruses , and may play important APOBEC-independent roles such as protection from insertions [65] . Although we demonstrate later that the GC content of the gene sets considered correlates strongly with susceptibility ( see next section “GC content of genomes predicts deaminase hotspot susceptibility” ) , we sought to quantify the extent to which native GC content contributed to the observed susceptibilities . To that end , for each of the gene sets considered , we generated new susceptibility scores that controlled for the GC content by defining a new null model that used the GC content of the gene set being analyzed to select codons ( see Methods ) . Using the same unweighted z-score calculation as described above to determine statistical significance , we observed that for the high GC-content EBV genome , 9 out of the 16 NNC trinucleotide contexts that we examined were still statistically resistant ( compared to 13 out of 16 in the uncorrected case ) . For the vulnerable OsHV genome , 6 out of 16 NNC trinucleotides were statistically vulnerable ( compared to 10 out of 16 in the uncorrected case ) . We confirmed these trends again by recalculating susceptibilities for each of the gene sets , now using the GC-corrected null model ( S2 and S3 Tables ) . We visualize these trends with another heatmap similar to Fig 3 which maintains the ordering of the columns and rows for ease of comparison ( S1B Fig ) . Although the trends in susceptibility are similar to the non-GC-corrected case , the differences between vulnerable and resistant gene sets are reduced , suggesting that GC content explains a great deal , though not all , of the observed resistance to APOBECs . Additionally , there is a question of whether the total number of nonsynonymous mutations should be minimized , or whether it is more appropriate to examine the nonsynonymous mutation fraction , which normalizes the number of motifs . There are two possible scenarios to consider for the frequency of mutation . In one scenario , a gene may only be exposed to the mutagen a few times during its activity . This is observed in B-cells , where AID is exposed to a gene possibly only once per cell cycle [8] . In this scenario , because the number of mutations is limiting , the probability that a mutagenic event causes an amino acid change is reduced by minimizing the fraction of nonsynonymous mutations , i . e . the number of nonsynonymous mutations divided by the total number of motifs . If a mutagen is exposed to a sequence only rarely , as in the case of AID , a coding sequence could minimize the probability of each mutagen changing its sequence by having many hotspots in the silent position . In the second scenario , mutations are not rate-limiting and the absolute number of hotspots causing nonsynonymous mutations would need to be minimized rather than the fraction . We found that tests conducted under the second scenario show similar results , albeit with slightly weaker trends ( see section below “GC content of genomes predicts deaminase hotspot susceptibility“ ) . Thus , our results are clearest in the first scenario ( fraction of nonsynonymous mutations ) assuming it is more indicative of susceptibility to mutagenesis . We conclude that examining the fraction is more instructive in determining trends ( S2A Fig ) . Additionally , as a control , we applied this model to different mutation profiles rather than the standard C to T . We chose every type of transition mutation ( NNA to NNG , NNT to NNC , and NNG to NNA , which is complementary to NNC to NNT ) . In the cases of NNA to NNG and NNT to NNC ( S2B and S2C Fig ) , we observe trends that are the opposite of our findings as described in the main text . Averaged across all 16 hypothetical NNA -> NNG hotspots , the susceptibility is lower in the oyster herpes virus by 0 . 18 ( P = 0 . 011 , 2 tailed t-test ) relative to the housekeeping genes , though still prevalent but not as strong for the L1 elements ( difference = 0 . 10 , P = 0 . 24 ) . Similarly , with all 16 NNT to NNC T to C mutations , L1 ( difference = 0 . 12 , P = 0 . 048 ) and oyster herpes viruses ( difference = 0 . 198 , P = 0 . 0063 ) show lower susceptibility to hotspots relative to human housekeeping genes . This suggests that mutagenesis at NNC to NNT hotspots is preferable for host genomes compared to these other patterns of mutation . We applied the same analysis as we did in Fig 3 , to the hypothetical mutations of NNG to NNA ( S2D Fig ) . Because we consider motifs on both strands , NNG to NNA is equivalent to CNN to TNN . For NNG to NNA we observed some similarity to the NNC to NNT case . Again there appear to be groups of resistant and vulnerable gene sets and these overlap somewhat with the NNC to NNT case , although the clusters are less clearly separated . If the average susceptibility of each gene set ( rows in Figs 3 and S2D ) were the key feature separating the vulnerable and resistant clusters , then we would expect similar clusters for both NNC to NNT and CNN to TNN . However , the clusters in each case are not the same suggesting that hotspot ( column- ) specific features that are unique to each cluster , are also important . Although considering all 12 possible types of mutation is beyond the scope of this study , we did consider the three transition mutations other than NNC to NNT ( S2B–S2D Fig ) and one transversion ( NNC to NNG , S2E Fig ) that is directly comparable to Fig 3 . Concerning the latter case , we found some similarities to the NNC to NNT transition case in that all control and retrotransposon gene sets clustered together in a predominantly vulnerable group . Clearly , some similarity is to be expected since most mutations at the third position of a codon are synonymous , and the NNC contexts are shared between the two cases . When we analyzed susceptibilities using the total number of nonsynonymous replacements rather than the fraction of replacements , we observed greater variability across the 16 NNC motifs ( S2A Fig ) . In spite of this , at the highest level the same two clusters of vulnerable and resistant gene sets emerge . However , under this assumption many of the gene sets in the resistant cluster , such as the mammalian housekeeping genes and the Epstein-Barr virus genome here show high vulnerability to GC rich motifs . These motifs include the APOBEC3G hotspot CCC , and the hypothetical hotspots GCC and GGC , all three of which cluster together . These gene sets simultaneously show resistance to the AT rich motifs ATC , TTC , TAC , and AAC , which are also clustered together . Varying GC content of the gene sets may explain these discrepancies . A gene with a very high GC content would contain more GC-rich motifs and fewer AT-rich motifs ( and therefore a higher raw nonsynonymous mutation count ) by chance . However , it is surprising that the trends are similar to our observed susceptibilities ( Fig 3 ) , since our definition of susceptibility examines only the nonsynonymous mutation fraction instead of raw count , “normalizing” for the GC content . Our analysis from here on thus examines the model of susceptibility that calculates the percentile of nonsynonymous mutation fraction compared to random sequences . It is not immediately clear how GC content will also affect the nonsynonymous mutation fraction that is used to calculate susceptibility ( Fig 3 ) . However , a gene set with varying GC content may adopt a codon usage that inherently favors APOBEC hotspots at positions that cause synonymous mutations . We therefore compared the GC content of our vulnerable and resistant gene sets and found a strong statistical difference , with resistant gene sets having much higher GC content ( Fig 4A , 2-tailed t-test , P <1 . 0−5 ) . Furthermore , we observed that the correlation of susceptibility with GC content was strong for some motifs and not others . For many of the 16 motifs we examined , a higher GC content correlated strongly with low susceptibility to that motif as demonstrated by very negative correlation coefficients ( Fig 4B ) . For example , the AT-rich hotspot TTC showed a strong correlation ( Fig 4C ) , but for the GC-rich hotspot CCC the trend is not as apparent ( Fig 4D ) . There is zero correlation between susceptibility of the motif GGC ( which is not an observed AID/APOBEC hotspot ) and GC content . Finally , to further corroborate these trends we have assessed the correlations between GC content and the alternate definition of motif susceptibility which considers the absolute number of nonsynonymous mutations rather the fraction . We observe again that a low GC content confers stronger susceptibility ( negative correlation ) to motifs of the WNC ( W = A/T , N = any nucleotide ) motif ( Fig 4E ) , consistent with our findings to be discussed below ( see section: “Retrotransposons and viruses show great discrepancy in resilience to biological deaminase hotspots . ” ) . Our analysis suggests that GC content is strongly predictive of existing APOBEC hotspot susceptibility , with the exception of the human APOBEC3G hotspot CCC . The results shown in Fig 3 suggested that the current APOBEC hotspot preferences may have evolved to mutate particular targets while minimizing self-damage . To explore this further , we asked which of the 16 possible NNC motifs were best in discriminating host against parasitic ( viral , transposable element ) genomes . We proceeded by performing all pairwise comparisons for each gene set in the resistant cluster to each gene set in the vulnerable cluster , noting the differences in the susceptibility scores for each of the 16 NNC motifs . Since we observed 10 resistant gene sets and 12 vulnerable gene sets ( Fig 3 ) , this resulted in 120 individual comparisons . For each comparison the 16 NNC motifs were sorted by the difference in susceptibility of the resistant to the vulnerable genome ( resistant—vulnerable ) . As an example , when we compared human housekeeping genes ( resistant ) to L1 retrotransposon elements ( vulnerable ) we discovered that the housekeeping gene sets had an average susceptibility to the motif ATC of 0 . 39 , whereas the average susceptibility for the L1 elements was 0 . 872 ( S4 Table ) . Thus , with a difference of -0 . 482 , the APOBEC3A/B hotspot ATC shows the strongest contrast out of any of the 16 motifs and is ranked the highest , whereas the motif GAC has a difference of +0 . 321 and is ranked the lowest . The rankings suggest that ATC , rather than GAC , might evolve to be a preferable motif due to its capability for damaging L1 ORFs more so than impacting host genes . By analyzing all 16 motifs across 120 comparisons , we found that the rankings of human APOBEC3B ( NTC , N = any nucleotide ) and AID ( WRC , W = A/T , R = A/G ) hotspots in particular , were significantly lower than expected by chance ( see Methods , bootstrapped P = 0 . 0071 and P = 0 . 0002 respectively ) . We further calculated , for each motif , the average difference across all pairwise comparisons of vulnerable and resistant gene sets , again ranking each motif by the overall difference in susceptibility . For 9 out of the 16 motifs we examined , the susceptibility of the vulnerable genes was strongly lower than that of the resistant gene sets to a similar degree , a difference that was highly significant ( t-test , P < 2 . 2 × 10−16 for all 9 motifs , Fig 5A columns shown in orange box ) . Curiously , these 9 motifs which demonstrate the greatest difference in mean susceptibilities also include all 8 motifs defined by the motif WNC ( W = A/T ) . Although WNC itself is not necessarily a canonical preference seen in an existing cytidine deaminase , this set of 9 significant motifs still contains considerable overlap with existing cytidine deaminases , including all AID hotspots ( WRC , where R = A/G ) , three of the four possible APOBEC3A/B hotspots ( ATC , TTC and GTC ) , the human APOBEC3C hotspot ( TTC ) and the murine APOBEC3 hotspot ( TCC ) . As an alternative method to confirm the significantly lower susceptibility of these 9 motifs ( WNC and GTC ) , we compared the average rank of the 16 NNC motifs across all of our 120 pairwise comparisons , noting that the average rank of the 9 motifs were all statistically lower than expected by chance , whereas the average ranks of the other 7 motifs were higher than this ( S5 Table , BH-corrected P < 10−4 , one-sample MWU ) . Additionally , although it was not possible to include degenerate motifs such as WRC and NTC in our original analysis due to the non-independence with other observed motifs such as ATC , we did calculate susceptibility scores for these two motifs separately , as described next . The 9 motifs we identified here as having lower susceptibility ( Fig 5A ) includes all 7 motifs that we previously indicated as distinguishing APOBEC-resistant and vulnerable gene sets based on the clustering of motif susceptibility ( Fig 3 , orange box ) . These 7 motifs cluster primarily as a result of their biological ability to distinguish vulnerable and resistant gene sets . Additionally , the two degenerate motifs of respectively AID ( WRC ) and hAPOBEC3B ( NTC ) showed large differences between vulnerable and resistant motifs as well ( shown in white in Fig 5A ) . Both results suggest that the motifs that most strongly distinguish APOBEC-resistant and vulnerable genomes tend to be the existing hotspots , rather than hypothetical hotspots not associated with AID or APOBEC . We observed that broadly , the vulnerable cluster ( Fig 3 , blue box ) contains two qualitatively distinct families of genes , notably viruses and transposable element clusters . Similarly , the resistant cluster includes endogenous host genes such as the housekeeping gene set , and also viruses such as EBV and HSV-1 , that have likely co-evolved with the APOBECs and may have acquired low APOBEC susceptibility as a consequence of this co-evolution . We assumed that the preferred APOBEC hotspots would be those that maximally damage their intended targets in intrinsic immunity , i . e . retrotransposons and vulnerable viruses , while minimizing damage to the host genome . With this assumption in mind , we looked more closely at two particular comparisons relevant to APOBEC evolution in humans . We first examined human housekeeping genes against human L1 elements , as the L1 elements are a major biologically significant target for APOBECs [39 , 66] . Secondly , we compared two viruses: the Epstein-Barr virus ( Human Herpesvirus 4 ) from the resistant cluster , against the Ostreid Herpesvirus , which we used as a control genome since its host is invertebrate and therefore not expressing any APOBEC genes , and is categorized in the vulnerable cluster . In the first comparison ( Human housekeeping vs Human L1 ) we found that the top 3 motifs in terms of susceptibility differences are all APOBEC3B TC hotspots: ATC , GTC , and TTC ( Fig 5B , S4 Table ) . Interestingly , the observed difference does not arise because human housekeeping genes are particularly resistant to these motifs , since they have mean susceptibilities of 0 . 39 , 0 . 51 , and 0 . 43 respectively , which is close to that expected by chance ( 0 . 5 ) . However , retrotransposons are particularly susceptible to these hotspots , with susceptibilities ranging from 0 . 87 to 0 . 91 for these three hotspots , where 1 represents maximum susceptibility . These high values suggest that human APOBEC3B hotspots may have evolved due to a particularly high capability of inducing nonsynonymous mutations in L1 elements , rather than low susceptibility to these hotspots in the host genome . We verified this effect for the AID hotspot WRC on some human non-housekeeping genes as well , by comparing the susceptibilities of our human housekeeping genes to a set of B-cell specific genes [67] . We used the list of B-cell specific genes listed in this study to obtain their coding sequences from the UCSC genome browser and calculated their susceptibilities , using the longest sequence if multiple isoforms were available for a single gene . For the B-cell specific genes , the susceptibilities to the 4 WRC hotspots were even lower than housekeeping genes: ( AGC: difference = 0 . 10 , P = 0 . 0075 ) , ( AAC: difference = 0 . 1 , P = 0 . 000038 ) , ( TAC: difference = 0 . 12 , P = 8 . 8 × 10−9 ) , ( TTC: difference = 0 . 07 , P = 0 . 00313 ) , suggesting that B-cell genes are under even more pressure to avoid deleterious mutations at AID hotspots than other genes . In the second case ( Fig 5C , S6 Table ) , we compared the APOBEC-resistant Epstein-Barr virus ( EBV ) to the Ostreid Herpesvirus ( OsHV ) . The most highly ranked motifs in terms of susceptibility differences again included many A/T-rich motifs , with 8 out of the 9 top motifs again having the motif WNC . Furthermore , the two motifs showing the strongest difference are the hotspots ATC and TTC ( collectively the motif WTC ) , which are human APOBEC3A/APOBEC3B hotspots . Many of the AID hotspots that EBV might be expected to evolve resistance to ( given that EBV is a B-cell tropic virus ) also show strong contrast , with the hotspots TAC and AAC ranked 4th and 5th and TGC and AGC being ranked 8th and 9th respectively . These differences , in contrast to the comparison between housekeeping genes and L1 elements , arise not only due to the high susceptibility of OSHV to motifs such as ATC ( 0 . 82 ) and TTC ( 0 . 83 ) , but also due to EBV’s strong resistance , i . e . low susceptibility ( 0 . 21 for ATC , 0 . 277 for TTC ) . These low susceptibilities to APOBEC hotspots in EBV , together with our observation that AID/APOBEC hotspots again show the strongest contrast between resistant and vulnerable viruses , suggest that viral genomes such as EBV have adopted resistance to this particular form of mutagenesis . In general , we note that motifs of the form WNC , and more specifically , WTC ( W = A/T ) appear to show the largest difference in vulnerabilities between housekeeping genes and retrotransposons , as well as between resistant virus and vulnerable genes . While WNC itself is not necessarily a canonical AID or APOBEC hotspot , it is a superset of the well-characterized AID hotspot WRC [1] . Furthermore , WTC overlaps with the hotspot NTC seen in human APOBEC3A/APOBEC3B . Thus , the fact that many of the motifs with the largest differences correspond to existing AID/APOBEC hotspots supports a hypothesis that the ability of AID/APOBECs to act effectively against viruses and retrotransposons while avoiding damage to the host is an evolutionary advantage for the host . Many factors play an important role in shaping virus nucleotide evolution [68 , 69] . We next describe how we investigated in particular the effects of codon bias , GC content , CpG dinucleotide frequency , selective pressure , and AID hotspot targeting of a hypothetical or actual virus sequence to validate our model . We constructed new gene sets of 250 genes of completely random open reading frames that are 1500 nucleotides long , beginning with the methionine codon , ending with a stop codon , and with random non-stop codons in between . For the random codons , each nucleotide was chosen randomly , with either 40% , 50% , or 60% GC content . We calculated the susceptibility scores of these random sequences to validate susceptibility independent of evolutionary effects . Under uniform distribution of C and G ( 50% GC content ) , meaning each of the four bases is equally likely to be chosen , we observe that the median susceptibility to the canonical hotspots NTC ( hA3B ) , CCC ( hA3G ) , and WRC ( AID ) is 0 . 5 ( S3A Fig ) . With random sequences of varying GC content , however , the susceptibility changes . Randomized sequences of low ( 40% ) GC content showed high susceptibility to the hotspots NTC and WRC , and similarly , high GC content sequences showed low susceptibility ( 60% ) . We reaffirmed this by creating new randomized sequences of the resistant , high-GC content EBV and vulnerable , low-GC content OsHV , replacing each codon in each sequence with a synonymous one weighted by the GC parameter ( a similar weighing description is given in the Methods under “Calculating Susceptibility” , though here it is applied to the baseline sequences rather than the null model ) and observed similar trends , suggesting a strong role of wildtype GC content in explaining APOBEC resistance . We note , however , that this effect is not present for the hotspot CCC . We corroborated these effects by looking at the effect of GC content on codons for existing protein sequences . To determine the effects of native GC content on determining APOBEC hotspot resistance , we additionally calculated susceptibility under a GC-controlled null model ( see Methods ) , weighing the GC parameter to be the mean GC content of the entire gene set . This meant that each gene sequence would now be compared to random synonymous sequences with a chosen GC content . Using this model , we found that changing the GC content of the random sequences affects the observed susceptibilities , reaffirming the importance of GC content in determining susceptibility . If a baseline ( wild type ) coding sequence is being compared to random sequences with higher GC content , the baseline sequence will tend to show higher susceptibilities to NTC and WRC hotspots ( suggesting that lower GC content is more vulnerable to deamination ) . Despite these trends , the vulnerable OsHV sequence still has higher susceptibilities to NTC and WRC than EBV when GC is corrected ( S3B Fig ) , although are not significantly so ( NTC P-val: 0 . 18 , WRC P-val: 0 . 34 , BH corrected t-test ) . L1 elements , however , have their apparent high susceptibilities to NTC ( the hotspot that hAPOBEC3B has which restricts them ) neutralized when corrected for in this manner . Thus , high impact of these hotspots on L1 may be a beneficial effect of their inherent low GC content . There is little change in the susceptibilities of HIV to the NTC and WRC degenerate motifs ( NTC P-val: 0 . 93 , WRC P-val: 0 . 82 ) when we apply this correction , suggesting that while low GC genomes can be resistant to AID/APOBEC hotspots , there can be genomes for which their GC content does not contribute strongly to their vulnerability . Additionally , we obtained codon usage tables of the coding sequences of OsHV , EBV , and our human housekeeping genes and L1 elements using Emboss cusp ( http://www . bioinformatics . nl/cgi-bin/emboss/cusp ) . When we compared OsHV to EBV , we regenerated sequences in each genome using the codon usage table of the other and calculated the susceptibility of the new sequence . We observed that OsHV , when shuffled with codons with probability proportional to the codon usage bias of EBV , adopts a susceptibility much closer to that of EBV , and vice versa ( S3C Fig ) . We observe similar effects when we exchange the codon biases of HK and L1 elements ( two mammalian gene sets that we assume would be differentially targeted by APOBECs ) and calculate susceptibilities ( S3C Fig ) . We conclude that the nature of APOBEC hotspot susceptibility is also influenced by codon bias for both virus and vertebrate genes . However , note that the results described above using the GC-corrected model in effect creates a model with codon bias according to wildtype GC content . Thus , it is most likely that our results here using codon bias changes ( S3C Fig ) are correlated with those using GC-content changes ( S3B Fig ) . Additionally , codons within the same amino acid can be classified as “intra-codon” or “inter-codon” depending on whether they are a NNC with C to T at the silent position or not . For example , AGC is an intra-codon for serine , since AGC to AGT does not change its amino acid , but AGA would be an inter-codon . Statistical tests of all intra- and inter-codons comparing the vulnerable OsHV and resistant EBV show that OsHV has statistically fewer intra-codons than EBV ( difference = 0 . 182 , Fisher Test P < 2 . 2 × 10−16 , 2-tailed Fisher test ) , providing additional evidence for the role of codon bias in hotspot susceptibility . Finally , our findings on codon bias were corroborated by the trinucleotide distribution of resistant and vulnerable gene sets ( “Clustering algorithm” , under Methods ) . Selective pressure was also modeled by taking the original EBV sequence , a resistant genome , and inducing mutations at different positions with probability weighed to either favor nonsynonyous or synonymous mutations , according to a parameter we use to approximate the dN/dS ratio . dN/dS represents the ratio of the number of observed mutations at nonsynonymous sites per site , to the number of mutations at synonymous sites per site . Specifically , mutations occurring in the first two positions of a random codon almost always represent nonsynonymous mutations and mutations in the third position are typically ( but not always ) synonymous . Therefore , we mutated positions along the EBV sequence with probability weighed by a parameter . This parameter is the ratio of mutation at the first two positions to the third , and acts under the assumption that each site is equally likely to be mutated and that this ratio of mutations is correlated to dN/dS . For example , if this parameter is equal to 2 , then a mutation at each position of the codon is equally likely , since 2/3 of the mutations would be nonsynonymous , affecting one of the first two positions , and the remaining mutations would affect the third nucleotide , which is often synonymous . The corresponding value for the dN/dS ratio would be closer to 1 , as there are twice as many nonsynonymous sites as synonymous . Under these assumptions , dN/dS is roughly half of our selective pressure parameter . For values of this parameter being 0 . 25 ( red ) , 1 ( green ) , 4 ( blue ) , and 8 ( purple ) and for up to 500 mutations over time we calculated susceptibilities for all 16 NNC motifs ( S3D Fig ) . Over time , averaged across all these motifs , the low susceptibility changes to a higher susceptibility , which may be explained by a change in GC content to neutral ( or 50% ) as each of the 12 mutation types ( from one base to another ) are equally likely . If dN/dS is high , however , the change is smaller . Indeed , the susceptibilities between the observed EBV sequences and EBV after 500 mutations when our selective pressure parameter = 8 ( approx . dN/dS = 4 ) is not statistically significant except for the motifs GCC and GGC ( GCC: P = 0 . 0004694 , GGC: P = 0 . 0053; G2-tail t-test , BH corrected ) . The difference however between EBV after 500 mutations under high selective pressure ( parameter = 8 , approx . dN/dS = 4 ) and low ( parameter = 0 . 25 , approx . dN/dS = 0 . 125 ) is significant for many motifs including AAC ( P = 0 . 0019 ) , ACC ( P = 0 . 0018 ) , ATC ( P = 0 . 0018 ) , CGC ( P = 0 . 0019 ) , GAC ( P = 0 . 025 ) , TAC ( P = 0 . 00039 ) , TCC ( P = 0 . 0377 ) , and TTC ( P = 0 . 0018 ) . Based on our simulations , high selective pressure appears to play a modest role in maintaining APOBEC hotspot resistance . We note that many mammalian herpesviruses display resistance ( Fig 3 ) , and if APOBEC deamination is deleterious to survival of the virus , then there would be evolutionary pressure on the virus to change in such an environment . However , this may be mediated simply by changing the GC content of the virus . Finally , there are other models of APOBEC/AID activity at hotspots . One of these is the S5F model of Yaari et al . [70] which inferred the mutability and substitution profile during somatic hypermutation among all 1024 possible 5-mers that targeted silent positions and which are therefore assumed to be independent of the effects of selection . Although somatic hypermutation includes not only AID activity , but also non-canonical Base Excision Repair and Mismatch Repair , the model is arguably the most comprehensive available for somatic hypermutation . To investigate the impacts of this activity on a gene’s susceptibility over time , we used this model and induce mutations at each position to the vulnerable OsHV and HIV genomes , resistant EBV genome , and random sequences , with probabilities proportional to the S5F model . This model includes the effects of base excision and mismatch repair which induce mutations beyond that which occur at hotspots , so we also simulate mutations using the same frequencies but only at C to T . We induce a number of mutations equal to up to 60% of the sequence length ( to account for different gene lengths ) . Our results show that when we consider susceptibility to the AID hotspot WRC , AID targeting decreases susceptibility to WRC motifs ( S4A Fig ) , highlighting the possibility that mutations at non-hotspot sites can decrease the susceptibility of a gene . In the case where we apply S5F , but only mutating C to T ( and thus also G to A on the opposite strand ) , we mutate up to 60% of C and G positions instead per gene . Under this mutational model , that ignores the DNA repair mechanisms that act downstream of AID , we observe the opposite trend , namely that susceptibilities to the WRC motif instead increase as mutations accumulate ( S4B Fig ) . To extend this analysis beyond AID and somatic hypermutation , we also modeled the activity of two other APOBECs . We performed a simple simulation of APOBEC mutation at TC ( hAPOBEC3B ) and CCC ( hAPOBEC3G ) hotspots and then calculated the changes in susceptibility to those motifs accordingly . When we mutate the resistant EBV sequence at the hotspot NTC , susceptibility increases to a high level ( S4C Fig ) . Interestingly , we did not see a change in susceptibility to the other hotspot CCC ( S4D Fig ) for all genomes that we examined . This suggests that regardless of the number of hotspots that are at nonsynonymous or synonymous positions , susceptibility to hotspots can be altered by GC content . The fact that CCC susceptibility , a hotspot which appears neutral to GC content ( S3A Fig ) , does not change by this mechanism lends support for this idea . Finally , the CpG dinucleotide has biological significance as a marker for DNA methylation and therefore gene activation and inactivation , especially in mammals [71] . We calculate another susceptibility measure that properly controls for this frequency ( See: “Controlling for CpG dinucleotide content” under Methods ) . As shown in S5 Fig , when we consider this , most of the gene sets clustered similarly to the uncorrected method ( Fig 3 ) . As far as clustering of the motifs is concerned , we obtained one cluster of 6 motifs ( rightmost dendrogram of S5 Fig ) , containing all 4 ANC motifs including two AID hotspots , AGC and AAC , which strongly overlaps with the cluster of motifs obtained without correcting for CpG ( as described above for Fig 3 , which contains 3 of the 4 ANC motifs , 3 of 4 TNC motifs , and GTC ) . These results suggest that the impact of these motifs is still similar when measuring susceptibility controlled for CpG . The gene set clusters also remain the same as the clusters without controlling for CpG content , with the exception of two formerly vulnerable gene sets , MHV68 and HERV which now cluster within the resistant gene sets . These gene sets , respectively a mammalian virus and set of retrotransposons , are co-expressed in environments alongside APOBECs , and in the case of HERV , is strongly edited by them [24] . Since accounting for CpG does not appear to affect the results very strongly , we do not control for this motif in the analyses , although this observation suggests that adapting CpG usage can be a mechanism of reducing effective susceptibility for a subset of viruses , particularly HERV . Altogether our results suggest that resistance to APOBEC hotspots is a natural consequence of GC content , and provides an explanation of how , under selective pressure , a coding sequence may adopt resistance to APOBECs . These findings are consistent with the results described above ( Fig 4 ) where we found a strong correlation between a genome’s GC content and its motif susceptibility . High GC content , for example , is a feature of several herpesviruses , and a previous study proposes that GC content may play a previous protective role against retrotransposon insertion in these viruses [65] . While we are not suggesting that gene sequences always adopt or evolve their GC content solely for the purposes of APOBEC resistance , we do show that GC content adjustment may confer additional defense against the consequences of restriction by most APOBECs , with the notable exception of the hAPOBEC3G hotspot CCC . We have introduced a method of quantifying the susceptibility of gene coding sequences to mutational hotspots . We used this method to examine differences in susceptibilities across many host and pathogen genomes and tested both known and hypothetical cytidine deaminase targets . We observed that the known hotspots , particularly the APOBEC hotspots , have a high capacity for inducing non-synonymous mutations in retrotransposons and viruses , while minimizing damage to host genes . Although it needs further investigation , our data suggests a possibility that this selectiveness has been important in shaping AID/APOBEC hotspot preferences during vertebrate evolution . Susceptibility is potentially influenced by many non-independent factors , including GC content , amino acid usage , and codon bias . Our basic measure of susceptibility is neutral to GC content and codon bias , allowing us to quantify the extent that these factors can evolve to influence susceptibility and its counterpart , resistance . In some cases GC content , via its effects on codon usage , strongly influences susceptibility , while in other cases the effect is less clear ( Fig 4A–4D ) . We expect that a C/G rich hotspot such as the APOBEC3G hotspot CCC should be correlated to the genome’s GC content and indeed , many GC-rich herpesviruses such as HSV-1 ( 68% GC ) and EBV ( 60% GC ) show high susceptibility to CCC ( as well as to the hypothetical hotspots GGC and GCC ) when we use an alternate definition of susceptibility which considers the absolute number of nonsynonymous C to T mutations , rather than the fraction ( S2A Fig ) . However , the situation is complex since we found cases where GC-rich genomes showed resistance to even GC-rich motifs such as CGC ( S2A Fig ) . We conclude that evolving the optimal fraction of nonsynonymous motifs is more important than the absolute number in order to resist a broad variety of APOBEC mutations , since it is the fraction that is minimized across the resistant gene sets . Alternatively , the motif count ( as opposed to the fraction ) may still play a role in achieving susceptibility and it is possible a virus may evolve a high GC content to reduce the potential damage of A/T rich hotspots such as TC ( human APOBEC3A/APOBEC3B ) and TTC ( human APOBEC3F ) . Additionally , tetranucleotide motifs for APOBEC3F and APOBEC3G have been observed [41] where the +1 position is also important , and although we have done the analysis for NNC motifs only , future work might include more complex motifs as well . We have controlled for the frequency of the CpG motif ( see under Methods: “Controlling for CpG dinucleotide content” , S5 Fig ) , however an extended analysis ( that we leave for future work ) might also consider the possibility that resistance is a consequence of conservation of the CpG motif itself . CpG sites in vertebrate hosts will often be methylated and are prone to spontaneous deamination . At particular loci , deamination of CpGs may be disadvantageous and evolutionarily selected against . Interestingly , AID and some APOBEC enzymes including APOBEC1can deaminate methylated Cs [4] . Thus , there may be some interplay between spontaneous and AID/APOBEC mediated deamination of methylated CpG sites . Given the additional constraints imposed by the coding region sequence , this interplay is potentially quite complex , and we therefore leave the corresponding analysis for future work . A high GC content appears to confer resistance to deamination by many of the naturally occurring cytidine deaminases ( including hAPOBEC3B , characterized by the hotspot TC and AID , characterized by the hotspot WRC ) . Additionally , the diversification of APOBEC3 hotspots observed in primates may favor a wider spectrum of intrinsic immune responses , as we have shown that by the same metric , genomes with high GC content remain vulnerable to hAPOBEC3G’s CCC hotspot . However , even when accounting for the endogenous GC content of the gene sets considered , we observed similar , albeit less significant overall trends ( S3 Table ) . Thus we have shown here that there is an advantage to having a high GC content in that it tends to minimize the impact of most APOBEC hotspots . In the future deaminase resistance might also be validated experimentally , for example , by examining the viral fitness of vulnerable or low-GC content herpesviruses in APOBEC-deficient cell lines . One of our important observations was that retrotransposon ORFs from the L1 families had very high susceptibility scores , with some having average susceptibility as high as 0 . 9 . As our definition of susceptibility is based on the fraction of mutations that cause amino acid changes , one potential explanation for high susceptibility is that past mutations in hotspots have more frequently caused synonymous mutations rather than nonsynonymous ones . This bias would ( as expected ) cause the remaining hotspots to become enriched for potential nonsynonymous mutations . Also , it would be incompatible with the observation that APOBEC restriction of retrotransposons is usually deaminase-independent , as has been shown in cell culture experiments [72] . However , this observation is in dispute , as in vivo and in vitro assays have shown that hAPOBEC3A can deaminate exposed single-stranded LINE1 DNA [73] . Recent results show that older retrotransposons are often heavily edited for deactivation and that high retrotransposon editing can be beneficial in terms of genome diversification and potential exaptation [66 , 74] . Additionally , a recently published study provides phylogenetic and experimental evidence suggesting that even throughout primate APOBEC3A evolution , L1 elements were unable to evolve resistance to APOBEC3A restriction , which may explain our high observed susceptibilities of L1 ORFs to APOBEC3 hotspots [75] . Thus , a high L1 ORF hotspot susceptibility may be beneficial to the host by driving exaptation and new functionality . One possible implication that arises from our results is that the evolved hotspot preferences may have in turn shaped the genomes of viruses that target those cell types where the deaminases are expressed , suggesting a possible arms race ( past or ongoing ) between viruses and the vertebrate intrinsic immune systems . It is important to note that while the susceptibility measure may reflect the evolutionary impact of deaminases , it might not always implicate AID/APOBEC as an active agent . The low susceptibilities to canonical hotspots that we have observed in virus genomes that are likely to have ongoing exposure to AID/APOBEC , such as EBV and HSV-1 ( Fig 3 ) suggest that this feature may be evolutionarily advantageous . We observed that HIV was a vulnerable gene set with high susceptibilities , yet the HIV vif protein neutralizes APOBEC3G , targeting it for degradation , which may abrogate the need to evolve resistance at the genome level . Similarly , another herpesvirus we examined , MHV 68 ( a strain of Murid Herpesvirus 4 ) , showed high susceptibility to APOBEC hotspots including mouse APOBEC3 hotspots defined by the motif TTC . A recent study showed that the mouse APOBEC3 does not restrict this virus either in cell culture or in vivo [63] , suggesting the possible existence of an as yet unidentified APOBEC evasion mechanism . MHV68 , alongside the transposon-like HERV elements , also show reduced susceptibility when we control for CpG dinucleotide motif frequency ( S5 Fig ) . Thus , for some genomes the CpG dinucleotide frequency appears to be an important constraint on codon selection which may also provide reduced susceptibility to APOBEC hotspot mutations . Thus , an apparently vulnerable genome may still adapt some level of resistance to APOBECs by adjusting its CpG usage accordingly , although we only observed this for two vulnerable viral genomes . We demonstrate that a high GC content and the corresponding observed codon biases are a mechanism by which resistance is conferred ( Fig 4 and S6 Fig ) . Not all hotspot resistance is explained by GC content , and we observe that varying the GC content of random sequences does not change susceptibility to the hAPOBEC3G hotspot CCC . Under our model of susceptibility , which is based upon the percentile of the fraction of nonsynonymous mutations , CCC may be robust to changes in GC content . For example , deamination at a CCC hotspot codon to CCT might be synonymous , whereas if the hotspot were on the opposite strand ( and in frame ) forming a GGG codon , then deamination to AGG will be nonsynonymous . Since the frequency of both CCC and GGG codons will presumably change linearly with GC content then the fraction of nonsynonymous mutations should remain approximately unchanged for different GC levels . Although this argument applies only when the hotspot is in frame , in practice we find that CCC hotspots are indeed robust to changes in GC content regardless ( S3A Fig ) . Thus , there may be alternating roles for the different APOBEC hotspots , with human APOBECs that prefer NTC to be primed to target low GC content virus sequences and retrotransposons , and human APOBEC3G which may have evolved the CCC hotspot to be robust to the protection a high GC content may confer to other hotspots . Notably , human APOBEC3G is known to restrict HSV-1 in vitro [31] , which is a high GC content virus [65] . Due to the differences in susceptibility to the hAPOBEC3G hotspot CCC and the hAPOBEC3B hotspot NTC , there may be an evolutionary advantage in there being multiple APOBEC hotspots since this increases the breadth of viruses ( meaning variability in GC content ) that the host can optimally restrict . This result is consistent with previous suggestions that the diversification of APOBEC3 genes has arisen due to an arms race against a wide variety of viral pathogens , given that many of the diversified APOBEC3 genes are under positive selection [2 , 76 , 77] . Our model and definition of susceptibility examines the impact of mutations on both strands , weighing both equally . This model has been sufficient to demonstrate overall trends and differences between a variety of mammalian and non-mammalian genes and viruses . However , in the particular case of retroviruses , APOBEC deamination is biased in that mutations are predominantly G-to-A on the plus strand , as has been observed for HIV and HTLV [78] . Thus , an alternative model that might be considered would examine susceptibility only on one strand . Under this scenario , differences in susceptibility between different retroviruses and among viruses from different clinical isolates might be used as additional evidence of an ongoing arms race between the deaminase and the retrovirus . Although mutations are likely to be predominantly deleterious , viruses in particular need to maintain variation at the population level , and often within an individual host to escape host immunity . For HIV in particular , there is growing evidence that APOBEC-mediated mutagenesis may actually contribute to the generation of escape variants [27 , 64] . Based on the idea that there is an optimal mutation level for viruses to propagate [79] , it is conceivable that APOBEC susceptibility may benefit many viruses , particularly if we were to consider individual genes ( that contain epitopes , for example ) rather than entire genomes as we have done here . Many mammalian genes often showed great differences in susceptibilities to different hotspots as well . In the future , predicting which genes are particularly susceptible may illuminate the overall impact of AID/APOBEC mutagenesis on the mammalian genome as well . Our results suggest that the current AID/APOBEC hotspots are effective in mutating unadapted viruses and retrotransposons , while minimizing collateral damage . However , several of the viruses that we expected would be vulnerable are resistant and vice versa ( such as vulnerable HIV , MHV4 , and SFV , which are all mammalian viruses ) . These features may be explained by alternative APOBEC evasion mechanisms , such as the HIV gene vif . A similar mechanism was recently proposed to exist for MHV4 [63] . We expect future work , that uses more sophisticated models of sequence evolution and mutation and on more thorough strains of viruses , to elucidate these differences further . We describe a model that calculates the susceptibility , UH , to motif H , from NNC to NNT , a wildtype coding sequence ( denoted as swt ) , that contains L codons as follows . Given swt , we generate a vector: CL= ( c1 , c2 , …cL ) where ci is the ith codon in swt . We generate a set W , of N ( in practice 1000 ) random sequences: W = ( s1 , s2 , … sN ) where each sequence sj , in W is a sequence generated from swt by randomizing codon usage as follows . For each sequence sj , for each codon ci in CL we replace the codon with another synonymous codon chosen either with uniform probability or a biased probability derived from a GC-bias parameter ( see next paragraph ) . For each sequence sj , the k occurrences of ( real or hypothetical ) hotspot motif H are counted . Then for each sequence sj we associate with it a vector Hk = ( h1 , h2 , … hk ) , where ht is the position of the tth occurrence of hotspot motif H on sequence sj , in both orientations . We define a function f ( ht ) which is 1 if changing the nucleotide at position ht from C to T ( or G to A ) , causes an amino acid change , and is 0 otherwise . Then the nonsynonymous mutation fraction of sequence sj is: M ( sj ) = ∑t=1kf ( ht ) k The susceptibility , UH of wildtype sequence swt to the hotspot H , is calculated as: UH = ∑l=1NR ( l ) N Where R ( l ) = 1 if M ( Sl ) < M ( swt ) , ( M ( swt ) being the nonsynonymous mutation rate of the wild type sequence ) , and 0 otherwise . Additionally , we may generate random sequences taking into account GC content , by assigning a weight to each synonymous codon . Given a GC content ( fraction ) of p , and a set of synonymous codons for a particular amino acid ( e . g . GAG and GAA for Glu ) , a weight is derived for each codon based on the GC content of the codon . Within each codon , we assume that G and C nucleotides are chosen according to GC content with probability p , whereas A and T nucleotides are chosen with probability 1-p . The codon weight is the product of the three corresponding probabilities . Thus , for Glu , the codon GAG would be assigned a weight of p× ( 1-p ) ×p = ( 1-p ) p2 , whereas the codon GAA would be assigned the weight p ( 1-p ) 2 . Codons are then selected with probability equal to a normalized vector of these weights . Note that if p = 0 . 5 , each synonymous codon is selected with equal probability . As described herein , susceptibility is impacted in many cases by GC content ( Fig 4 ) . We evaluated the larger gene sets on a gene-by-gene basis to assess the impact of GC content on susceptibility . These large gene sets include the human and mouse housekeeping gene sets and several virus genomes ( S6 Fig ) . Although not every hotspot we examined demonstrated a strong relationship , the TTC motif had the strongest mean correlation among the 16 NNC motifs we examined , in the human and mouse housekeeping gene sets ( Spearman correlation = -0 . 728 and -0 . 486 respectively , S6A and S6B Fig ) . For many of the viral genomes we analyzed , the GC content fell within a very narrow range and/or had very few genes , leading to weak correlations . We therefore grouped the genomes of EBV , HSV-1 , Murid Herpesvirus 2 , and the Ostreid herpesvirus , as well as many non-herpesviruses with 10 or fewer genes including HIV , Adenoviruses , and others to obtain a wider range of GC content across many genomes . By analyzing all the genes together we observed a trend similar to that for the housekeeping genes , where GC content of the gene correlates with the motif susceptibility ( -0 . 493 for the TTC hotspot , S6C Fig ) . Because GC content influences susceptibility in some cases , but not others ( Fig 4B ) , we chose in the first instance to keep our analysis neutral and we generated our null model without any constraints on GC , uniformly choosing each synonymous codon at random . We subsequently compared our susceptibility results to the GC content of the corresponding gene set to determine the extent to which GC content determines susceptibility ( Fig 4B ) . To control for CpG dinucleotide frequency we are interested in calculating susceptibility for each NNC motif conditioned on the CpG frequency in the original sequence . It is usually impractical to compute this conditional probability empirically . Controlling for this condition might be achieved by considering only those randomly generated sequences that have , by chance , produced the same number of CpG motifs as the original sequence . However , in practice the fraction of random sequences that exactly match the CpG motif count is often very small , particularly if the CpG content is already skewed , making it unlikely that such exact matches are produced by the random model . This may lead to very small numbers , making an approximation preferable . Thus , we approximated the joint distribution of CpG dinucleotide frequency and susceptibility , assuming the joint distribution is a bivariate normal distribution . The corrected susceptibility is simply the percentile of the approximated Gaussian conditional on the CpG frequency in the analyzed sequence . Correcting for CpG frequency made little difference to our results . We obtained protein-coding sequences from a variety of biological sources , and broadly speaking our data sources are categorized into three groups: Known or suspected AID/APOBEC targets including viruses and repeat elements; non-targeted viruses; and putative unintentional or collateral targets of AID/APOBEC which include host genes . Known virus targets that we analyzed included HSV-1 , AAV-2 , HERV ( specifically the HCML-ARV retrovirus ) and SFV . Since papillomaviruses are highly diverse we selected a well- characterized strain with medical relevance , which was HPV16 due to its link to cervical cancer . We also analyzed EBV , due to its shared B-cell tropism with AID . MHV68 , a close relative of the other herpesviruses we studied , is not suspected to be restricted by mAPOBEC3 although multiple human APOBECs were capable of restricting the virus [63] , so to test the possibility that mAPOBEC3 hotspots may have impacted murid herpesvirus coding sequence vulnerability or vice versa we included both Murid herpesvirus 2 and MHV68 . As controls , we included several viruses unlikely to be restriction targets , due to not being vertebrate or mammalian viruses , which consist of the Ostreid herpesvirus , as well as two plant viruses , the cherry and strawberry viruses ( Table 2 ) . These controls were evaluated by substituting other potential controls consistent with the same rationale , i . e . they are not restricted or exposed to AID/APOBEC . We also analyzed a set of potential mutational targets in the mouse genome , namely a data set that is enriched in potentially oncogenic AID off-target genes [56] and sets of housekeeping genes from the human and mouse genomes ( see below ) . Altogether we examined a total of 22 distinct gene sets . We wanted to avoid including an overwhelming number of controls , but at the same time we checked whether those we did include were unusual . Therefore , additional control viruses including plant viruses were obtained and individually verified by including these viruses in the clustering analysis ( Fig 3 ) and determining whether they clustered consistently with the viruses originally used as controls . In each case these clustered with our previous control viruses , demonstrating the validity of the original choices ( S7 Fig ) . Additional controls used to validate our current control set included: Tobacco mosaic virus ( NC_001367 ) , Grapevine leafroll associated virus 1 and 3 ( NC_016509 , NC_004667 ) , Citrus tristeza Virus ( NC_001661 ) , Carrot yellow leaf Virus ( NC_013007 ) , and Fly C virus ( NC_001834 ) . Housekeeping genes for the vertebrate zebrafish , which lacks APOBEC3 , and the invertebrate C . elegans were obtained from a RT-PCR array ( http://www . sabiosciences . com/rt_pcr_product/HTML/PAZF-000Z . html ) and a study of C . elegans housekeeping genes respectively [2 , 80] . The list of human housekeeping genes were identified in a previous study [81] . In cases where a gene had multiple splice variants , we used the longest sequence available ( UCSC Genes track/ knownGenes table via the UCSC genome browser query tool ) . For the set of mouse housekeeping genes , we used a previously published list which assigned probabilities of being housekeeping to each gene [82] . We used a threshold probability of 0 . 75 in order to obtain a gene set of comparable size to our human housekeeping gene set . Again , the longest sequence was used in case one gene had several coding sequences . L1 sequences for both mouse and human were downloaded from the UCSC genome browser database under the RepeatMasker table from the mm10 and hg19 databases respectively . Sequences were filtered with a minimum of 5500 base pairs , to ensure only full length elements , which are canonically around 6000 bp long . ORFs were extracted using EMBOSS getorf , with a minimum length of 1000 nucleotides finding all sequences between start and stop codons . These ORFs were separated based on their family and all ORFs were subsequently separated by family . In total 14 families were identified , including members from the L1PA and L1PB families identified in [52] . For each family , sequences were clustered . Representative sequences were processed to reduce statistical redundancies between repetitive sequences using cd-hit with default parameters [83] . The final L1 set was obtained by merging the representative sequences from each cluster . The data being clustered was a matrix of the average susceptibility scores , which range from 0 to 1 , of all 16 NNC motifs in the columns , and gene sets ( e . g . L1 , Housekeeping genes , etc . ) in the rows . Thus , both rows and columns were clustered using standard hierarchical agglomerative clustering and average linkage , using the Euclidean distance metric as the distances between each row and column . These clusters were computed and displayed using the heatmap . 2 function in the R library gplots . We further validated these clusters using a k-means approach , showing that when we tested across a range of values for k , 2 clusters represents an optimal separation of the gene sets , indicated by the largest decline in intra-cluster sum of squares ( S8A Fig ) . As an alternative approach , we used Principal Component Analysis ( PCA ) . Here , one principal component accounts for a large amount of the overall trend in motif susceptibility ( S8B Fig ) , and the clusters of gene sets are consistent for the different methods and separate primarily along the first principal component ( S8C Fig ) . Codon bias is likely correlated with the more general measure of trinucleotide ( NNN ) frequency . We confirmed this by calculating the trinucleotide frequency usage of all of our gene sets and clustering them similarly ( S9 Fig ) . The occurrence of the trinucleotide AAA was skewed for the L1 elements for human and mouse , so it was excluded as an outlier . We observed the same clusters as before in Fig 3 , except that Hepatitis B virus , and HERV , which under susceptibility are grouped with the vulnerable clusters that include L1 elements among others , now appears in the resistant cluster that includes housekeeping genes and mammalian viruses . To assess statistically whether the rankings of certain groups of the 16 possible NNC ( for example , the 4 human APOBEC3B NTC hotspots of ATC , CTC , GTC , and TTC ) motifs were significantly high or low , we performed a pairwise comparison between each of the 10 resistant and the 12 vulnerable genomes of Fig 3 . For each comparison , we ranked the motifs by the difference in average susceptibilities between the resistant and the vulnerable gene set ( resistant-vulnerable ) , noting the ranks of the 4 motifs in each comparison . We assessed significance by sampling one resistant and one corresponding vulnerable gene set chosen at random , until one of each resistant gene set was sampled , and ensuring no gene set was sampled more than once . In each comparison we noted the ranks of the 4 motifs , summed them and compare the sum to a random null model selection of ranks , taken 4 at a time uniformly from the numbers 1 through 16 without replacement . For example , if examining the statistical significance of the motif NTC , the ranks of the hotspots ATC , CTC , GTC , and TTC , for a particular resistant vs vulnerable gene set , were 1 , 3 , 5 , and 9 , these would be compared to a random selection of numbers 1–16 which could be 3 , 8 , 9 , 15 . The sums of these ranks would be 18 and 35 respectively . We then summed the ranks of the 4 hotspots across all our comparisons and compared this to the sum of null model ranks . Repeated 10000 times , our bootstrapped P-value was the fraction of times the sum of observed ranks found to be greater than the sum of the random ranks . If this fraction ( equivalent to a P-value ) were greater than 0 . 05 , this demonstrated that the observed ranks were not significant .	The APOBEC family of cytidine deaminases are important enzymes in most vertebrates . The ancestral member of this gene family is activation induced deaminase ( AID ) , which mutates the Immunoglobulin loci in B Cells during antibody affinity maturation in jawed vertebrates . The APOBEC family has expanded particularly in the mammals and in primates , where they have evolved roles in restriction of viruses and retrotransposons . Biochemical studies have established that AID preferentially targets “hotspots” such as AGC and avoids “coldspots” such as CCC . Other APOBECs have evolved distinct hotspots . For example , APOBEC3G , which targets retroviruses including HIV , has evolved to target the motif CCC as a hotspot , but it is unclear why . Here we ask why the AID/APOBEC cytidine deaminases evolved their particular mutational hotspots . Our results show that a wide range of unrelated genes including mammalian LINE1 ORFs and non- mammalian ( ancestral-like ) viruses are highly susceptible to mutations in APOBEC hotspots and less susceptible to the hypothetical non-APOBEC hotspots . On the other hand , mammalian viruses tend to exhibit low susceptibility to the same APOBEC hotspots , suggesting these viruses have co-evolved to minimize potential damaging mutations , and that the native GC content plays a large role in this behavior .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "microbial", "mutation", "medicine", "and", "health", "sciences", "retrotransposons", "pathology", "and", "laboratory", "medicine", "pathogens", "microbiology", "retroviruses", "viruses", "immunodeficiency", "viruses", "mutation", "rna", "viruses", "dna", "viruses", "genetic", "elements", "sequence", "motif", "analysis", "mammalian", "genomics", "epigenetics", "dna", "herpesviruses", "chromatin", "dna", "methylation", "research", "and", "analysis", "methods", "sequence", "analysis", "epstein-barr", "virus", "lentivirus", "bioinformatics", "chromosome", "biology", "medical", "microbiology", "hiv", "gene", "expression", "microbial", "pathogens", "chromatin", "modification", "dna", "modification", "animal", "genomics", "biochemistry", "cell", "biology", "nucleic", "acids", "database", "and", "informatics", "methods", "viral", "pathogens", "genetics", "transposable", "elements", "biology", "and", "life", "sciences", "genomics", "mobile", "genetic", "elements", "organisms" ]	2017	The preferred nucleotide contexts of the AID/APOBEC cytidine deaminases have differential effects when mutating retrotransposon and virus sequences compared to host genes
New therapeutics are needed for neglected tropical diseases including Human African trypanosomiasis ( HAT ) , a progressive and fatal disease caused by the protozoan parasites Trypanosoma brucei gambiense and T . b . rhodesiense . There is a need for simple , efficient , cost effective methods to identify new molecules with unique molecular mechanisms of action ( MMOAs ) . The mechanistic features of a binding mode , such as competition with endogenous substrates and time-dependence can affect the observed inhibitory IC50 , and differentiate molecules and their therapeutic usefulness . Simple screening methods to determine time-dependence and competition can be used to differentiate compounds with different MMOAs in order to identify new therapeutic opportunities . In this work we report a four point screening methodology to evaluate the time-dependence and competition for inhibition of GSK3β protein kinase isolated from T . brucei . Using this method , we identified tideglusib as a time-dependent inhibitor whose mechanism of action is time-dependent , ATP competitive upon initial binding , which transitions to ATP non-competitive with time . The enzyme activity was not recovered following 100-fold dilution of the buffer consistent with an irreversible mechanism of action . This is in contrast to the T . brucei GSK3β inhibitor GW8510 , whose inhibition was competitive with ATP , not time-dependent at all measured time points and reversible in dilution experiments . The activity of tideglusib against T . brucei parasites was confirmed by inhibition of parasite proliferation ( GI50 of 2 . 3 μM ) . Altogether this work demonstrates a straightforward method for determining molecular mechanisms of action and its application for mechanistic differentiation of two potent TbGSK3β inhibitors . The four point MMOA method identified tideglusib as a mechanistically differentiated TbGSK3β inhibitor . Tideglusib was shown to inhibit parasite growth in this work , and has been reported to be well tolerated in one year of dosing in human clinical studies . Consequently , further supportive studies on the potential therapeutic usefulness of tideglusib for HAT are justified . New medicines are needed for neglected tropical diseases ( NTDs ) . Drug discovery and development in all disease areas is inefficient due to high failure rates that increase the costs . The impact of the high failure rates on NTDs is very significant as a result of limited resources afforded to these diseases because of a lack of commercial markets for the medicines [1] . There is an urgent need for cost effective methods and strategies that expedite opportunities to identify new medicines for NTDs [1] . As part of our on-going efforts to identify new treatment molecules for human African trypanosomiasis ( HAT; also known as sleeping sickness ) we have been investigating approaches to identify new molecular mechanisms of action ( MMOAs ) early in the discovery process . HAT is a deadly infection affecting mostly impoverished areas in rural sub-Saharan Africa , caused by the protozoan Trypanosoma brucei . There are currently an estimated 30000 cases of HAT , with approximately 70 million people in 36 African countries at risk [2] . At this time , five medicines are available for HAT , all of which must be administered intravenously or intramuscularly , and three of these treat late-stage infections , when the parasite has crossed the blood-brain barrier . Notable limitations of these treatments include ineffectiveness , long periods of drug administration , cost , toxicity , and other severe medical side effects [3] . Consequently , there is a critical need to discover and develop safer , more effective and less expensive therapies for HAT . There is hope that this unmet medical need will be addressed with the promising candidates currently in development , fexinidazole and SCYX-7158 [4] . However , there is still a need for back-up strategies for these compounds in case they do not meet the promise and to also address potential resistance . Drug discovery and development is facilitated by an understanding of how drugs work ( i . e . their mechanism of action ) . How a drug works is a key feature of the pharmacological response , the therapeutic index , and the drug’s therapeutic usefulness [5–7] . Knowledge of a drug’s mechanism will also facilitate optimization and clinical development . It has been long recognized that pharmacological action begins with an interaction between two molecules . Ehlrich noted in 1913 that a substance will not work unless it is bound “corpora non agunt nisi fixata” [8] . However , binding alone is not sufficient for a medicine to produce an effective and safe response that is therapeutically useful . The molecular mechanism of action ( MMOA ) through which a medicine connects binding to the response is also important [9 , 10] . For example , two similarly structured molecules can bind to an enzyme with similar affinity; however , one will bind as a substrate in a manner suitable for the catalytic reaction while the other will inhibit the reaction . For receptors , two similarly structured molecules can bind to a receptor with similar affinity; however , one will initiate the response ( agonist ) whereas the other will block the response ( antagonist ) . Understanding drug action at the molecular level can facilitate the rational design of new medicines as well as provide opportunities to identify new therapies . Understanding the MMOA can provide opportunities to identify new molecules that are differentiated from previous molecules , molecules that can be repurposed for new indications or molecules that have been previously overlooked . A detailed understanding of the MMOA is also important for interpretation of in vitro/ in vivo correlations in target validation studies and understanding pharmacokinetic/pharmacodynamics ( PK/PD ) relationships . Two important features of MMOA that have been shown to differentiate medicines are binding kinetics and binding competition . The binding kinetics are the rate at which a molecule binds ( association rate ) and debinds ( dissociation rate ) . A reaction with a slow dissociation rate can be functionally irreversible when the dissociation rate is sufficiently slow or covalent . Competition occurs when two molecules compete for the same binding site and will result in decreased fractional occupancy of the drug bound to the target . The decrease in fractional occupancy due to competition can be overcome by increasing the concentration of the drug . The decrease in fractional occupancy due to competition can also be overcome with slow dissociation kinetics and irreversibility . This pharmacological behavior is described as insurmountable drug action . Many examples demonstrate the important role of binding kinetics in effective drug action [9 , 11 , 12] . Aspirin is an irreversible inhibitor of prostaglandin H2 synthases ( also known as cyclooxygenase , COX ) , whereas ibuprofen is a rapidly reversible inhibitor of these enzymes with a fast dissociation rate [13 , 14] . The irreversibility of aspirin contributes to its usefulness for prevention of atherothrombotic disease [15 , 16] and differentiates aspirin from ibuprofen . Irreversibility can be achieved by covalent binding as well as long residence times in a system not at equilibrium to provide insurmountable pharmacological behavior [17] . Slow dissociation kinetics in a system not at equilibrium contributes to the use-dependence behavior of channel blockers [18] and the insurmountable behavior of many receptor blockers , including the well-documented , angiotensin receptor blockers [19 , 20] . These examples illustrate some of the advantages to time-dependent behavior including a greater inhibition of activity and longer lasting pharmacodynamic behavior and target occupancy enabling administration of lower doses and in some cases greater durability . These mechanistic behaviors contribute to the effectiveness and utility of many anti-infectives including the irreversible inhibitor , penicillin [21] , and isoniazid [22 , 23] . This behavior also contributes to the effectiveness of many other medicines including lapatinib , tiotropium , and candesartan to name a few [9 , 11] . For completeness it must be noted that long-residence time/irreversibility is not suited for all system . When there are liabilities due to mechanism-based toxicity ( on-target toxicity ) , long residence time/irreversible behavior is not appropriate [5 , 9] . Competition of a drug with endogenous substrate for binding will reduce the fractional occupancy and may result in a loss of effectiveness . This will require higher concentrations to achieve the same effect and thereby decrease the selectivity , increase the potential for toxicity as well as provide a challenge for pharmaceutical development to administer the drug at a sufficient dose and concentration at the site of action to achieve efficacy . Competition with endogenous substrates is particularly relevant for protein kinases where ATP competitive inhibitors must compete with high concentrations of endogenous ATP for binding and inhibition of the kinase activity . The physiological concentrations of ATP are estimated to be in excess of 1 mM , which will result in a ≈ 100 fold shift in IC50 for inhibitors with a Km for ATP of 10 μM ( see theoretical explanation below ) . Importantly , mechanisms that avoid competition with ATP have been identified , including slow dissociation kinetics ( long residence times ) and non-competitive mechanisms . Wilson and coworkers recently demonstrated that the MMOA of the first approved kinase inhibitor , Gleevec ( imatinib mesylate ) , involves time-dependent binding that is important to its action and selectivity [24] . This time-dependent , mechanistic behavior was not identified until over a decade after the drug had been discovered . In general , the many molecular mechanistic features important to the action of first in class drugs are identified long after a drug is discovered , and consequently , not used to inform optimization and development [6] . Accordingly , a simple , cost-effective method to identify MMOAs for lead compounds will be valuable to R&D , notably in resource limited diseases , such as NTDs . Towards this goal , we have used a simple four point method to characterize the MMOA of compounds for time-dependence and competition as part of our efforts to identify novel and therapeutically useful protein kinase inhibitors of T . brucei . The method involves measuring activity at one inhibitor concentration with and without a preincubation ( 30 min in this case ) to determine time-dependence , and at two substrate concentrations ( 0 . 5 Km and 5x Km in this work ) to determine competition . A shift in the time dependence indicates that the inhibitor binding does not reach equilibrium in the time frame of the experiment . The inability to rapidly reach equilibrium can be due to a multi-step binding mechanism ( Fig 1 ) . A loss of activity with higher substrate concentrations is consistent with a substrate competitive mechanism of inhibition . A lack of shift in inhibition can be considered as noncompetitive , such as inhibitor binding in a different site . However , a combination of noncompetitive behavior and time-dependence can be diagnostic of a molecule with insurmountable pharmacological behavior due to slow binding kinetics ( including irreversibility ) that prevents competition . For example , an irreversible inhibitor will appear noncompetitive even when it is bound in the substrate binding site . The utility of the four point MMOA method was demonstrated with GSK3β isolated from T . brucei ( TbGSK3β ) and the identification of tideglusib ( NP-12 , NP031112 ) as a time-dependent , competitive inhibitor of TbGSK3β . Further evaluation of tideglusib showed irreversible behavior against the enzyme and inhibition of parasite proliferation at low micromolar concentrations . Tideglusib , previously evaluated in the clinic , has a good safety profile in humans [25 , 26] and therefore , warrants further studies to evaluate its potential for HAT . Altogether this work demonstrates a method for rapidly evaluating MMOA that can help identify opportunities for new NTDs . Reagents , unless noted , were purchased from Sigma-Aldrich ( Saint Louis , Missouri ) . GSM peptide ( sequence = RRRPASVPPSPSLS RHS ( pS ) HQRR , where pS is a phosphorylated serine residue ) [27] was purchased from EMD Millipore Corporation ( Temecula , California ) . ADP-Glo kit was purchased from Promega Corporation ( Madison , Wisconsin ) . Tideglusib and GW8510 were purchased from Sigma-Aldrich ( Saint Louis , Missouri ) . The short-form of GSK3β from T . brucei was expressed in E . coli with a C-terminal His tag and purified to homogeneity as described in Materials and Methods . Conditions for the enzyme assay were optimized for linearity with time and enzyme concentration using GSM as a phosphoryl acceptor and measuring ADP production using Promega ADP-Glo reagents . The substrate versus rate plots showed saturable kinetics for the two substrates GSM and ATP , with Kms of 23 μM and 21 μM , respectively and Vmax of 31 and 34 s-1 in the two studies ( Fig 2 ) . IC50 for inhibition TbGSK3β was determined as a function of preincubation time ( Fig 3 ) . Preincubated reactions were preincubated with inhibitor , TbGSK3β , GSM peptide , and buffer at room temperature; after 30 minutes reactions were initiated with the addition of ATP . Reactions with 0 min preincubation were initiated with the addition of TbGSK3β . Reactions were run for 5 min at room temperature , and stopped at 80°C . ADP product formation was measured by ADP-Glo kit . As shown in Fig 3 a shift in IC50 with preincubation time was observed for tideglusib , but not for GW8510 . The IC50 values for tideglusib were 43 nM and 173 nM for preincubated reactions and non-preincubated reactions , with standard errors of 6 . 5 nM ( N = 2 ) and 22 . 6 nM ( N = 5 ) , respectively . The IC50 values for GW-8510 were 14 . 5 nM and 15 . 1 nM for preincubated reactions and non-preincubated reactions , respectively . The four point methodology was demonstrated for tideglusib and GW8510 ( Fig 4 ) . Tideglusib decreased the enzyme activity following preincubation ( gray bars no preincubation; black bars 30 min preincubation ) at both 10 μM and 100 μM ATP ( Fig 4A and 4C , respectively ) . On the other hand there was no effect of GW8510 preincubation on the TbGSK3β activity at 10 μM and 100 μM ATP ( Fig 4B and 4D , respectively ) . These results can be interpreted that GW-8510 binds to TbGSK3β corresponding to a one-step model , to from an EI complex ( Fig 1 ) while tideglusib follows a two-step model , forming an initial EI complex that rearranges with time to a more stable E’I complex ( Fig 1 ) . In ATP competition studies the percent inhibition of activity by GW8510 was less at higher ATP concentrations ( 100 μM ) than lower ATP ( 10 μM ) , 28% vs 67% , respectively irrespective of preincubation time ( Fig 4B and 4D ) . Tideglusib’s activity was also decreased at when there was no preincubation but not when there was a preincubation ( 100 μM ) ( Fig 4A and 4C ) . This suggested that 1 ) tideglusib’s initial binding to form EI was competitive with ATP , and 2 ) that there is a time-dependent transition from the EI state to a more stable E’I state and is not sensitive to ATP competition . Due to this transition , tideglusib is more potent than GW8510 at high ATP concentrations following preincubations . We did not observe a shift in tideglusib activity at higher concentrations [80 μM] of the peptide substrate , GSM ( supplementary , S1 Fig ) indicating that the tideglusib does not interact with the peptide substrate binding site . Further evaluation of time dependence under more physiological conditions of no preincubation and higher ATP concentration was evaluated using enzyme progress curves . Progress curves measure the progress of the enzyme reaction with time [28] . In these reactions the time-dependent inhibition of TbGSK3β by tideglusib and GW8510 was evaluated under conditions in which inhibitor , 100 μM ATP and 80 μM GSM were combined in buffer and the reactions started with TbGSK3β ( Fig 5 ) . The reactions were stopped by heating at the specific times ( 5 , 10 , 20 , 30 and 60 min ) . The progress of the reaction , as measured by product formation ( ADP ) with no inhibitor ( DMSO control ) , was linear with time . When tideglusib was added , the percent inhibition was greater at 30 min as compared to 5 min , consistent with time-dependent loss of activity . In contrast , the effect of GW8510 on the rate of product formation was similar at all time points , consistent with rapid equilibrium reversible inhibition . Also of interest with tideglusib is that the slope of the progress curves approached zero at the later incubation times in a concentration dependent manner , suggesting complete inhibition of the enzyme at the later time points at all concentrations , behavior consistent with irreversible inhibition ( Fig 5A ) . Tideglusib and GW8510 were evaluated for irreversibility by measuring the TbGSK3β activity following a 100-fold dilution with reaction buffer . In these experiments tideglusib and GW8510 [100 nM] were preincubated with 50 nM TbGSK3β for 30 min at which time the reaction was diluted 100 fold into a solution containing 100 μM ATP and 80 μM GSM . The activity was then measured for up to two hours ( Fig 6 ) . The activity was compared to a DMSO control ( also preincubated for 30 min and activity measure up to 2 hr ) . The final concentration of tideglusib and GW8510 following dilution was 1 nM . Inhibition of enzyme activity was retained following dilution of tideglusib whereas enzyme activity was recovered in the reactions with GW8510 . The results are consistent with reversible behavior of GW8510 and irreversible behavior of tideglusib . This experiment does not distinguish between a very slow dissociation rate or covalent binding . As discussed below , tideglusib was previously demonstrated to have irreversible inhibition against human GSK3β with an IC50 of 5 nM following preincubation , with no evidence for covalent binding [29] . These investigators concluded in that work that tideglusib was a functionally irreversible inhibitor [29] . The results shown in Fig 6 are consistent with similar behavior for TbGSK3β . The four point MMOA assay was used to characterize molecules that had previously been identified as TbGSK3β inhibitors . The concentrations chosen for the MMOA assays were determined from the IC50 in the HTS screens . A good starting concentration for the time-dependent studies is the IC25 ( concentration at 25% inhibition ) , since time-dependent inhibition will increase with preincubation time . For the competition analysis , use of an inhibitor concentration equal to the IC50 at substrate Km ( theoretically 2x Ki ) will allow observation of the decrease in inhibition due to higher substrate concentrations ( Fig 5 ) . In practice , concentrations must be chosen that will allow the change to be observed and will be specific to each inhibitor . The activities of these molecules were not time dependent since they were unchanged after 0 and 30 minutes incubation ( Table 1 ) and all were found to be competitive with ATP since the percent inhibition was less using 100 μM versus 10 μM ( Fig 7 ) . Tideglusib was evaluated in T . brucei parasite assays as previously described in Materials and Methods . The concentration of inhibitor at which trypanosome proliferation was inhibited by 50% ( GI50 ) was 2 . 3 μM ( Fig 8 ) . In comparison , Ojo and coworkers reported the activity of GW-8510 in a similar 48 hr parasite growth assay with T . brucei brucei strain 427 to have an EC50 of 119 nM [30] . The effect of tideglusib on trypanosome viability after short term exposure was evaluated to determine if the molecule is cidal or cytostatic . During the short 6 hour treatment at high trypanosome density ( 5 x 105/mL ) , tideglusib ( at 5 or 10 μM ) markedly slowed trypanosome growth ( Fig 9B ) . In fact at 10 μM tideglusib , proliferation was entirely halted . However , dilution of tideglusib ( or DMSO ) from the media no longer arrested trypanosome proliferation , with a slight lag in recovery after 10 μM tideglusib exposure . After the initial lag in recovered growth , the cells grew with a normal rate of approximately 11 doublings in 48 hours ( Fig 9C ) . This is in contrast to the control , pentamidine , which killed all observed trypanosomes within 24 hours of the 6 hour exposure ( Fig 9C ) . These results show that tideglusib was cytostatic at high trypanosome cell densities . Trypanosomes treated with 10 μM tideglusib did demonstrate a slight lag in normal proliferation rates after the drug pressure was removed ( Fig 9C ) . We conclude that tideglusib is better at controlling proliferation during short exposure , but pentamidine wins over the long run . As part of our on-going efforts to identify new treatment molecules for NTDs we have been investigating approaches to identify new MMOAs early in the discovery process . Currently the identification of MMOA occurs after a molecule has been identified as a lead or clinical candidate using rigorous biochemical and pharmacological methods . Previous analyses have suggested that an optimal MMOA can provide an improved therapeutic index , however the challenge is to identify compounds with an optimal MMOA [6 , 31] . One approach to accomplish early identification of MMOA is to identify clusters of compounds with different MMOAs . MMOA identification would be used in addition to other factors to cluster molecules , such as activity , chemical structure and physical properties . Representatives from these compound clusters will then be evaluated empirically in phenotypic screens to identify actives and new opportunities . Accordingly , simple , efficient methods are required to easily identify molecules with different mechanistic features . We report here a simple , efficient four point MMOA screening method for evaluating compounds for time-dependent activity and competitive inhibition . In practice , this four point MMOA method can be applied at any stage in the drug discovery process . In early discovery , hits are identified in screens at one or a few concentrations and confirmed with IC50 determinations . The IC50 concentration will inform the concentration to use for the four point MMOA method . It should be noted that this method can be employed with any enzyme , regardless of the number of substrates by using two concentrations of the substrate of interest ( one around Km and the other above Km ) . It is worth mentioning that this four point MMOA method is a simplification of more rigorous methods that are common practice in biochemistry and pharmacology labs . Typically studies to look for time-dependence and competition are accomplished by investigating complete dose-response curves whereby the results inform a detailed understanding of the molecular interaction . In preparation for these definitive studies , smaller experiments are used to establish conditions . We have found that the four point MMOA studies are suitable for screening purposes and can be sufficiently reliable to differentiate compounds . In previous work , we applied a similar strategy to differentiate kinetics for antagonists binding to CCR5 using IC50 measures in receptor competition assays with and without preincubation [32] . Guo et al have recently described a dual point competition association assay for assessing ligand-receptor binding kinetics . Their approach was found to be useful for ranking molecules [33] . The four point MMOA method described here is a simpler method as compared to both of these , requiring less data points , thereby resulting in more efficient use of time and resources . Interesting findings from the four point MMOA method can be evaluated in more detailed assays as warranted . Using the four point MMOA method , tideglusib was determined to be a time-dependent inhibitor of TbGSK3β . Its effectiveness increased with time until its inhibition activity was similar to GW8510 , a well-documented potent TbGSK3β inhibitor [30] ( Figs 3 and 4 ) . Both tideglusib and GW8510 competed with ATP for binding to TbGSK3β without preincubation . Interestingly , tideglusib’s inhibition increased with preincubation time ( Figs 3 and 4 ) , incubation time ( Fig 5 ) and showed non-competitive behavior following preincubation ( Fig 4 ) . In comparison , the MMOA of GW8510 was competitive following preincubation ( Fig 4 ) . These results show tideglusib to have a different mechanistic profile for TbGSK3β than GW8510 . Tideglusib shows a time dependent switch from competitive to non-competitive behavior . This type of behavior is seen with other slow-dissociating molecules , which has been termed insurmountable and is a property of successful medicines [5 , 17 , 19 , 20 , 34] . We also used the four point MMOA method to characterize other kinase inhibitors that were previously identified in screens as TbGSK3β inhibitors ( Fig 7 , Table 1 ) . All were found to behave mechanistically as non-time dependent , competitive inhibitors . Adjusting the enzyme assays to account for these mechanisms showed that GW8510 and tideglusib had similar activity for TbGSK3β under more physiological conditions of preincubation and 100 μM ATP ( Fig 4 ) . However , this similarity did not translate to cellular activity , where GW-8510 was more active than tideglusib ( 0 . 12 μM to 2 . 3 μM , respectively [29] ) . These data suggests that factors other than the TbGSK3β MMOA differentiate the effect of these molecules against T . brucei , and demonstrates the value of early MMOA determination to help understand target specific hypotheses . Many factors may contribute to the difference in activity including poor solubility of tideglusib that limits cellular activity and/or that the cellular activity of GW-8510 maybe due in part to a non-TbGSK3β mechanism . While further studies are needed to evaluate these issues , the understanding of their different TbGSK3β MMOAs will assist in designing and correctly interpreting the data . The MMOA for tideglusib inhibition of TbGSK3β identified with the four point MMOA assay is consistent with that observed for inhibition of human GSK3β as reported by Dominquez and coworkers [29] . Using a more sophisticated analysis they also reported a binding mechanism which had an ATP-competitive component [29] . Furthermore , their studies on time-dependence led to the conclusion that tideglusib is a functionally irreversible inhibitor of human GSK3β and that the duration of the pharmacological effect caused by this behavior may be exploited to maximize its therapeutic potential . The work described here for TbGSK3β inhibition by tideglusib shows that the mechanism of inhibition is similar against the enzymes from human and T . brucei . In humans , GSK3β is a regulatory kinase for over 40 different proteins in a variety of pathways , and has been implicated in a number of diseases including: Type II diabetes ( Diabetes mellitus type 2 ) , Alzheimer's Disease , inflammation , cancer , and bipolar disorder [26] . Furthermore , studies have shown that one of the T . brucei GSK3 homologs ( TbGSK3 short ) , is necessary for cell growth and viability and thus may serve as a potential drug target for the treatment of HAT [30] . In a recent clinical trial designed to investigate safety and efficacy , tideglusib administered in escalating doses of up to 1000 mg/day to 30 Alzheimer’s patients for 6 weeks was generally well tolerated [25] . Tideglusib was also well-tolerated in phase II studies of Alzheimer’s disease and progressive supranuclear palsy [35 , 36] . Blood levels for tideglusib were reported in mice of approximately 2 . 25±1 . 55 μg/ml ( 6 . 7 μM ) after oral administration [37] . In summary , we used the four point MMOA method to evaluate the MMOA for T . brucei GSK3β inhibitors and observed tideglusib to have a time-dependent mechanism which differentiated it from rapidly reversible inhibitors , such as GW8510 . Tideglusib was shown to inhibit parasite growth in this work and has been reported to be well tolerated in one year of dosing in human clinical studies [35–37] . Consequently , further supportive studies on the potential therapeutic usefulness of tideglusib for HAT are justified .	Drug discovery for neglected tropical diseases must use efficient methods due to limited resources . One preferred drug discovery strategy is target-based drug discovery . In this strategy it is assumed that drug action begins with binding of a drug to its target . However , while binding is required , it is not sufficient to describe all the molecular interactions that translate binding to a therapeutically useful response . The contribution of aspects of the molecular mechanism of action ( MMOA ) such as time-dependence and substrate competition can influence concentration response relationships . To address this , a four point MMOA methodology was developed to evaluate time-dependence and substrate competition . We used this method to evaluate the MMOA for T . brucei GSK3β inhibitors , and observed tideglusib to have a time-dependent , ATP-competitive mechanism that differentiated it from rapidly reversible inhibitors , such as GW8510 . Adjusting the enzyme assays to account for these mechanisms showed that GW8510 and tideglusib had similar activities for TbGSK3β . However , this similarity did not translate to cellular activity , where GW-8510 was more active than tideglusib ( 0 . 12 μM to 2 . 3 μM , respectively ) . These data suggest that factors other than TbGSK3β MMOA differentiate the effect of these molecules against T . brucei .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "medicine", "and", "health", "sciences", "enzymology", "parasitic", "protozoans", "protozoans", "pharmaceutics", "pharmacology", "enzyme", "inhibitors", "chemical", "dissociation", "competitive", "inhibitors", "behavior", "chemistry", "drug", "therapy", "drug", "discovery", "biochemistry", "trypanosoma", "drug", "research", "and", "development", "biology", "and", "life", "sciences", "chemical", "reactions", "physical", "sciences", "trypanosoma", "brucei", "gambiense", "drug", "interactions", "organisms" ]	2016	A Four-Point Screening Method for Assessing Molecular Mechanism of Action (MMOA) Identifies Tideglusib as a Time-Dependent Inhibitor of Trypanosoma brucei GSK3β
The oscillating concentration of intracellular calcium is one of the most important examples for collective dynamics in cell biology . Localized releases of calcium through clusters of inositol 1 , 4 , 5-trisphosphate receptor channels constitute elementary signals called calcium puffs . Coupling by diffusing calcium leads to global releases and waves , but the exact mechanism of inter-cluster coupling and triggering of waves is unknown . To elucidate the relation of puffs and waves , we here model a cluster of IP3R channels using a gating scheme with variable non-equilibrium IP3 binding . Hybrid stochastic and deterministic simulations show that puffs are not stereotyped events of constant duration but are sensitive to stimulation strength and residual calcium . For increasing IP3 concentration , the release events become modulated at a timescale of minutes , with repetitive wave-like releases interspersed with several puffs . This modulation is consistent with experimental observations we present , including refractoriness and increase of puff frequency during the inter-wave interval . Our results suggest that waves are established by a random but time-modulated appearance of sustained release events , which have a high potential to trigger and synchronize activity throughout the cell . Transient and repetitive increases in the concentration of cytosolic Ca2+ are ubiquitous chemical cues in a cell . They are crucial for neuronal adaptation , cell growth and myocyte function , to name a few examples . The formation of complex intracellular release patterns plays an important role in cell communication , since Ca2+ achieves its functional specificity by differential signaling in space , time , and amplitude [1] . It is therefore a purpose of numerous research studies to understand the systemic generation of cytosolic Ca2+ signals [2] . In many non-excitable cells , increasing the concentration of inositol 1 , 4 , 5-trisphosphate ( IP3 ) triggers Ca2+ release from the endoplasmic reticulum ( ER ) by activating intracellular IP3 receptor ( IP3R ) channels in the ER membrane . Opening of IP3R channels is induced by the binding of IP3 and Ca2+ [3] and therefore Ca2+ released from open channels and diffusing through the cell can recruit further IP3R channels to open . If the IP3 stimulation is modest , Ca2+ is released from spatially confined clusters of intracellular Ca2+ channels [4] . Molecular interactions within a cluster lead to coherent opening of its channels and result in local elementary events called puffs [5]–[7] . The small number of IP3R channels involved in a puff ( 10 ) and the random appearance of puffs suggest their spontaneous generation by microscopic fluctuations , which has been related to classical excitability in activator-inhibitor systems [8] , [9] . For larger stimulation , however , Ca2+ forms more regular spatio-temporal waves or whole-cell oscillations involving release by multiple clusters and many channels [10]–[12] . The transition from small localized release to Ca2+ waves across cells has been addressed by a large number of experimental and theoretical studies , see e . g . [8] , [13]–[20] . The interest particularly concerns the mechanism for the generation of waves or global oscillations , because it controls the oscillation frequency , which is known to regulate important cellular functions [21] . Ca2+ diffusion coupling is responsible for communication between clusters [11] , [22] , [23] , thus mediating the synchronization of clusters into oscillations and the propagation of waves , but it is less clear what are the respective roles of increased IP3 stimulation and Ca2+ diffusion in the puff-to-wave transition . Enhanced synchronization with increasing [IP3] can in principle be mediated by two IP3 dependent scenarios . In a first scenario ( i ) the excitability of clusters grows with [IP3] so that a given amount of Ca2+ , e . g . diffusing from an active cluster , triggers puffs more frequently . Cluster excitability is expected to increase with rising IP3 concentration because the number of channels with bound IP3 increases . In a second scenario ( ii ) the amount of Ca2+ diffusing from an active cluster to clusters in its proximity increases with [IP3] for instance because of larger release current or longer release . In both scenarios the likelihood of propagation of activity from cluster to cluster increases with [IP3] . Recent modeling has often assumed that puffs are stereotyped events with a relatively constant amplitude and life time . Therefore , several studies focused on scenario ( i ) , which implicates that for larger stimulation the frequency of puffs increases . Furthermore , it has been suggested that the occurrence of a supercritical number of puffs during a short time interval leads to nucleation of a wave from a cellular subdomain [24]–[26] . Taken together , it follows that wave nucleations occur more often when the puff frequency increases , i . e . , for larger stimulation and excitability . Several aspects of intracellular Ca2+ waves are in very good accord with this explanation of the puff-to-wave transition , particularly the large variability in the inter wave interval ( IWI ) [26] . However , other important features of global releases , including their long lifetime and the extended refractoriness , are difficult to accommodate with present models and a realistic biomathematical model showing wave nucleation and the mentioned features remains to be devised . Here we pursue a new model of intracellular Ca2+ release that exhibits the puff-to-wave transition and the randomness of global releases and , in addition , displays the previously unexplained facets of Ca2+ waves . Our model is based on excitable dynamics of a single cluster that goes beyond the activator-inhibitor schemes . To achieve its complexity , the model incorporates the possibility of slowly decaying release from a cluster [27] . This slow release phase involves a residual domain , i . e . , Ca2+ which is transiently present in the cluster domain after a channel has closed . Then , the residual domain provokes perpetual reopening of channels in the cluster . This particularly happens when the timescale of recovery from inhibition , , is shorter than the typical decay time of the residual domain , , so that channels lose inhibiting Ca2+ before [Ca2+] is below activating levels . It was found in [27] , that in this case termination of release necessitates a further negative regulation , different from inhibition , and it was shown that this “inactivation” can be provided by transient unbinding of IP3 from the receptor . This previously overlooked effect originates from allosteric coupling between inhibiting Ca2+ binding and IP3 binding , which is a hallmark of IP3R gating [28] . To address the puff-to-wave transition in the context of complex cluster dynamics , we here analyze in detail the time courses of release in their dependence on IP3 concentration . We are using the most complete computer simulation of intracluster [Ca2+] that is presently available . It allows us to incorporate the effects of residual Ca2+ transients on channel opening . We find that with increased IP3 concentration the number of IP3 molecules bound to receptors ( excitability , scenario i ) and the amplitude of release from a cluster only slightly increase , while the lifetime of the signal increases dramatically ( scenario ii ) . This growing event lifetime can be related to the former finding of perpetual re-opening due to residual Ca2+ in the domain . Crucially , the mean lifetime increases because of the appearance of a subset of sustained release events at increased [IP3] , which thus constitutes a stimulation-dependent transition in the release pattern . We further find that this effect pertains to the increased variability in the number of channels that have IP3 bound . It is then likely , that , for a subset of events , the lifetime of residual domains increases because of larger total flux from the cluster , while typically remains constant [29] . As a result , for sufficiently increased stimulation is larger than and release becomes extended for several seconds until termination by IP3 loss . The transition to a qualitatively altered regime in our cluster model generates surprisingly many dynamical and statistical features that are in excellent agreement with the experimental descriptions of global releases . The modification in release shape and duration can be related to a study of temporal profiles of Ca2+ liberation in Xenopus oocytes in response to [IP3] step increases , which reported two distinct phases [30] . At moderate [IP3] , fluorescence recordings exhibit a fast but short release flux . For larger [IP3] , however , Ca2+ efflux involves first a time course similar to that observed for small [IP3] followed by a second phase of small amplitude release lasting several seconds . The fast and slow processes correspond well to the dynamics in the excitable system described by our model . The prolongation of release in the simulations and the comparison to experiments therefore hint at scenario ( ii ) , i . e . , it is the magnitude of Ca2+ release that predominantly grows with stimulation . This leads to a new explanation for the puff-to-wave transition , in which generation of waves emerges from the complex and stochastic excitable dynamics of a single cluster . The fact that sustained events in our computations resemble waves in their fluorescence traces suggests to identify these events with global release . In our simulations we find accumulating amounts of Ca2+ in the local domain during sustained events , which endows them with higher potential to establish synchronized release of multiple clusters . Thus , although a Ca2+ wave is a spatially organized event involving many clusters , the capacity of triggering or participating in a wave is a distinctive feature of long-lasting events in our single cluster model , which , in this sense , qualifies them as global events or waves . Propagation of waves , from our point of view , is probabilistic and depends on the positions and distances of adjacent channel clusters , but it is fundamentally the change in release time course that primes the channels for global release and that enables the synchronous opening of clusters . As a distinctive feature of large- [IP3] simulations we find that the IP3 unbindings and rebindings become very frequent for long-lasting , wave-like events . During release many channels may lose IP3 synchronously , while subsequently to the termination of release they rebind IP3 . These dynamics of IP3 binding/unbinding , synchronized to the appearance of wave-like release , causes a modulation of the number of activatable channels over timescales of minutes . The frequency of puffs occurring between waves is then strongly modulated and exhibits the refractoriness and increase of puff frequency known from experiments [11] , [31] . To the best of our knowledge , this crucial feature of Ca2+ waves has not been addressed in modeling studies so far . It is important to note that puffs in our simulations cannot be compared directly to those of experiments with EGTA-loaded cells [32] , [33] . Residual Ca2+ domains are suppressed by EGTA buffer , so that sustained release may not be observed in this setup [27] , [34] . Appearance of long-lasting events in the present simulations without exogenous buffer is therefore not in contradiction to the short puffs occurring in experiments for large [IP3] and with EGTA loading . EGTA thus prevents residual Ca2+ domains , appearance of prolonged release and , consequently , unbinding of IP3 . Nevertheless , occasional stochastic IP3 binding/unbinding may occur also in this situation and lead to puff variability . This finding may also explain part of the fluctuating puff amplitudes found in recent experiments with EGTA loaded SH-SY5Y cells [33] but will not be discussed in the present study . The article is organized as follows: We first describe the basic components of our model , consisting of Markovian gating of the IP3R channels and a reaction-diffusion system for the evolution of cytosolic concentrations of Ca2+ and Ca2+ binding proteins . We also summarize our numerical method . In contrast to our earlier publication [27] , we here employ the finite element method ( FEM ) for three-dimensional concentration fields and a hybrid scheme to couple local Ca2+ concentrations and channel gating states [35] . We then discuss our main findings based on simulation traces that typically cover several thousands of seconds simulation time . Here , we describe the distributions of event lifetimes and their dependence on IP3 concentration . We then analyze the role of IP3 unbinding for the lifetimes and discuss the stochastic variability . Finally , we suggest that the large number of long-lasting events for large IP3 concentration can be related to the frequency of global events . We here define a global event by an individual cluster's capacity to trigger adjacent clusters . In the final part of the paper , we draw a relation of such events to wave generation and dynamics in the inter-wave period and compare our results to experimental recordings . Experiments were performed on immature oocytes obtained from Xenopus laevis as described previously [4] , [13] . Frogs were anaesthetized by immersion in a 0 . 15% aqueous solution of MS-222 ( 3-aminobenzoic acid ethyl ester ) for 15 min , and small pieces of ovary removed by surgery following procedures approved under UCI IACUC ( University of California , Irvine , Institutional Animal Care and Use Committee ) protocol 1998-1337 . In this section we will describe the components of our mathematical model for channel gating , ion and buffer diffusion , and chemical reactions . The model consists of a Markov chain for IP3R channel states [36] and partial differential equations for spatial concentration fields ( Ca2+ and buffers ) . For numerical implementation , the two stochastic and deterministic paradigms are coupled by the hybrid method introduced in [35] . Experimental results on Ca2+ imaging of puffs and waves in Xenopus oocytes are previously unpublished data acquired during experiments described in [11] . Full experimental methods are given in that paper . To investigate the collective behavior of the IP3R channels , we group synchronous channel openings in collective events . We define a collective event based on intervals , during which the spatially averaged [Ca2+] in the cluster vicinity ( 500 nm box ) exceeds a threshold value of 0 . 1 µM . All events containing only a single channel opening were filtered out . Using the spatial average of [Ca2+] to define a collective event serves to facilitate comparison to experiments on Ca2+ puffs and waves in cells with dye buffer , where the number of open channels underlying a release event is not directly known . Following our event definition , we determine its lifetime or duration from the interval in which the Ca2+ concentration exceeds 0 . 1 µM . Simulations as those in Fig . 2 show that for low [IP3] we find short events lasting up to one or two seconds at most . The average duration of release events for [IP3] = 10 nM is about 0 . 5 s , roughly equal to the duration of puffs in experiments on Xenopus oocytes [10] . However , for high [IP3] we can observe both the short fast-decaying events as well as events characterized by sustained release lasting up to 10 s , akin to release waves observed in the same cell type . Fig . 3 shows the distribution of event durations for different [IP3] , as well as the average event duration depending on [IP3] . For all IP3 concentrations , the majority of events ( 60% ) was shorter than one second . However , while for low [IP3] all events are shorter than 3 s , for increasing [IP3] the distribution develops a wide shoulder . This results in a qualitative change in the distribution accompanied by increased average and increased variance of event durations ( Fig . 3 inset ) . Interestingly , models of release termination by inhibiting Ca2+ binding would suggest that larger release amplitudes could only accelerate termination [42] . However , in Fig . 3 we observe both , larger amplitudes and lifetimes , at higher [IP3] . To understand this behavior we will now study the IP3 binding to the receptors . A notable feature of the simulations in Fig . 2 is that the number of activatable channels ( i . e . channels that have bound a sufficient amount of IP3 to be able to open ) is fluctuating both for low ( D ) and for high ( H ) IP3 concentration . Unbinding of IP3 during release was first observed in a model of sustained Ca2+ release from IP3R channel clusters [27] . Note that the IP3 concentrations used in the prior and in the present work exceed the value of the dissociation constant by far ( see Table 1 ) and hence saturated IP3 binding sites , i . e . , 16 activatable channels , could be expected . We will now discuss how the impact of varying IP3 concentration on event lifetime is mediated by the dynamics of IP3 binding and unbinding , i . e . the dynamics of the number of activatable channels . Events with a higher number of participating channels will generally last longer ( see below ) . A simple approach to elucidate this relation is to count the number of activatable channels at the beginning of each event . The distribution of this number of activatable channels is shown for three different IP3 concentrations in Fig . 4 . Here we can find similar features as before: for increasing [IP3] the histogram gets skewed to the right leading to a much increased shoulder . Both the average and the standard deviation of the distribution increase with [IP3] ( inset ) . However , the increase of the average number of activatable channels is relatively small , rising from four channels at [IP3] = 10 nM to about five channels at 80 nM . On the other hand , the variability doubles for the same range of [IP3] . Thus , it is plausible , that the increased variability determines the appearance of extended events . What causes the increase of variance in the number of activatable channels with increasing [IP3] ? This effect can be understood with the help of Fig . 2 , D and H . First , the number of activatable channels generally decreases during release because of dissociation of IP3 , and it does so with much larger magnitude for long events found at large [IP3] ( see Fig . 2 H ) . Then , after event termination , the cluster is slowly reactivated by IP3 rebinding . The resulting rise in the number of activatable channels can be understood as steadily increasing cluster excitability . Hence , at some point , random fluctuations will trigger a new event , resetting some of the channels back to an unactivatable state and closing the circle . While for low [IP3] , the speed of reactivation is too slow to accumulate a high number of activatable channels , the higher reactivation speed for large [IP3] sometimes allows to reach a nearly fully activatable cluster and thus causes the large variability . Fig . 5 A shows that there is indeed a strong correlation between the number of activatable channels and the event duration . Events starting with only a low number ( below 4 or 5 ) of activatable channels reliably terminate fast . For an intermediate number of activatable channels ( 6 to 9 ) , there still is a strong stochastic variability of event durations . Finally , for a sufficiently high number almost all events last longer than a few seconds . In other words: a high number of activatable channels is necessary to produce a long event . In the intermediate regime ( 6 to 9 activatable channels ) , the time course of the events furthermore depends on the details of the channels' states , e . g . whether activatable channels have 3 or 4 subunits with bound IP3 ( data not shown ) . Taken together , these findings suggest that the increase of release lifetime seen in Fig . 3 is primarily mediated by the increase of variability for larger [IP3] , and not by the increase of mean number of activatable channels ( Fig . 4 ) . The nonlinear increase of duration with the number of activatable channels leads to a substantial increase of lifetime for higher [IP3] . We may also ask for the reason of the substantial loss of IP3 for some release events . If the number of channels opening synchronously is large , termination by inhibitory Ca2+ binding ( i . e . subunit transitions from X110 to X111 ) is often incomplete , mostly because the residual Ca2+ remaining locally after each channel closing is large enough to reopen the channel [27] . The timescale for recovering from inhibition , , is given by the inverse of the rate s−1: s . Hence , during a single long-lasting event ( i . e . longer than ) , a channel can easily switch back and forth between inhibited , open and resting states . Because inhibition does not guarantee termination for those events , inactivation of channels by IP3 dissociation is needed . This can clearly be seen from Fig . 5 B . The long-lasting events will result in a cluster configuration where only a few or even no channel at all remain activatable at the end of release events . For small [IP3] , however , the short event durations leave less probability for IP3 unbinding . Thus , those short events are a consequence of sufficient inhibition and little reopening probability , because with a smaller number of open channels less Ca2+ is extruded and residual Ca2+ domains are smaller [34] . The experimental observations of puffs and waves , as well as the diversity of the simulated release events for high [IP3] in terms of duration and event shape , call for an attempt to classify the observed release events into two different categories . The first category shall contain the short-termed puffs , observable for all concentrations of IP3 . The other category shall contain the long events showing low channel activity tails and termination by IP3 -unbinding , which we identify as waves . A simple classification criterion for the events is given by their potential to trigger another event at a close-by , imaginary cluster , i . e . their potential to support a Ca2+ release wave . Therefore , we classify a release event as wave if it induces a local [Ca2+] of at least 0 . 25 µM ( corresponding to the dissociation constant of activation , , see Table 1 ) in the entire domain . Otherwise the event is classified as puff . For a detailed discussion of the classification criterion please refer to the supplemental information . In Fig . 2 G , waves are indicated by a shaded background . Interestingly , the Ca2+ signals between 250 s and 275 s in Fig . 2 G clearly differ from the homogeneous signals produced by waves . With the above wave criterion , this burst of channel openings is identified as a sequence of distinct puffs . To compare the temporal structure of the release events , we averaged the time courses of the number of open channels over all events for each value of [IP3] . We then normalized the average time courses for different [IP3] to a peak value of 1 . Fig . 6 displays the average time courses for 10 nM and 70 nM concentrations . The solid red line shows the resulting average puff for small IP3 concentration . The fact , that for [IP3] = 10 nM we solely observed puffs , is reflected by a simple exponentially decreasing average number of open channels representing puff termination by Ca2+ inhibition on a fast time scale . However , the dotted blue line shows that for large IP3 concentration an increased second release phase appears , while for times below 200 ms no deviation from the profile of puffs occurs . Furthermore , Fig . 6 presents averaging of the same events for 70 nM concentration of IP3 , but now with two groups separating into puffs and waves according to our criterion described above . The solid blue ( puffs ) and green ( waves ) curves indicate that our classification between puffs and waves worked out well , as the average event shape for puffs at [IP3] = 70 nM shows a decay very similar to that of puffs for [IP3] = 10 nM . Analyzing the event shape for waves , we can clearly distinguish the two termination mechanisms: the first 200 ms are dominated by inhibition , while afterwards the slow IP3 unbinding takes over . A prominent feature of our model for large [IP3] is that here puffs and waves coexist . Occurrences of waves or global oscillation require the interaction of several clusters , and this interaction may affect dynamical aspects of the release , including oscillation period and variability [53] . Nonetheless , consideration of local dynamics generally allows revealing insights into the global aspects and the local dynamics often keep a dominating influence in many oscillating systems . Thus , having distinguished waves from puffs , we now interpret the repeated occurrence of wave-like events as a slow oscillation in the local dynamics and identify the period of global oscillations by the IWI in our simulations . Fig . 7 shows averages for IWI and inter puff interval ( IPI ) depending on [IP3] . Similar as shown by experiments [11] , our model predicts decreasing IWI for increasing [IP3] . The wave periods are generally in the same range as the periods of global oscillation measured in various cell types [11] , [12] , [52] . Recent experiments by Thurley et al . [55] showed exponential dependence of average period of global oscillations with stimulation , which is consistent with the scaling of IWI data in Fig . 7 . Furthermore , there is a linear relation between the average and standard deviation of the IWI in our simulation results ( inset ) . The slope of the regression line is 0 . 88 in simulations , which is similar to what was found for several cell types including astrocytes [26] . For HEK cells [26] , [55] , Hepatocytes [54] and Xenopus oocytes [4] , [13] , smaller variability was measured . This can possibly be understood from more complex effects of coupling , where synchronization can cause higher regularity of periodic dynamics [53] . In [11] , [24] it was shown that in Xenopus oocytes puffs appear in the phase between succeeding global waves . Exemplary fluorescence traces from experiments with this cell type are shown in Fig . 8A . Here a wave can be identified as a group of large events that appear within temporal proximity at neighboring cluster sites . The remaining events are then identified as puffs and their amplitude has been obtained as peak fluorescence level from traces such as shown in the figure . These amplitudes are plotted in Fig . 8B versus the phase during the IWI , where the phase has been defined as the ratio of time that has passed since the last wave and the total difference between preceding and next wave . Fig . 8C shows for comparison the corresponding puff events from numerical simulations . As seen from Fig . 7 , the three data sets for high [IP3] are quite similar , suggesting to pool them for better statistics . It is evident from the two plots , that events are absent at the very early phase after waves and that large-amplitude events occur only towards the end of the interval . To assess quantitatively the data , we have statistically evaluated Fig . 8B , C by pooling puffs from phase intervals of 0 . 1 and determining the number of events for each bin as well as the mean and standard deviation of their peak amplitudes . The connected dots in Fig . 8D show clearly that the frequency of puffs in experiments is strongly diminished in the early time after each wave . Similarly , we can calculate the temporal distribution of puffs in between waves from simulations . The bars in Fig . 8D are a stacked histogram of those phases for all data sets with high [IP3] . From what was observed above , in our model a wave will likely result in a transiently unactivatable cluster with few IP3 bound ( compare Fig . 2H and 5B ) . Thus , there is a refractory time where no puffs can arise , which lasts to about 30% of the IWI . The decrease of puff density at the last two bars at the end of the IWI is likely caused by the fact , that a wave also needs some preceding “silent” phase where the accumulation of activatable channels is not interrupted and partially reset by puffs . A similar decrease of puff frequency just before the wave onset is present in the experimental data . It is more clearly seen for cells with moderate and large IWI , as evident from Fig . 3 in [11] . We finally compare the evolution of puff peak amplitudes for both experimental and simulation data ( Fig . 8E ) . There is very good agreement for behavior of amplitudes ( solid lines ) and standard deviations ( dashed lines ) . The refractory effect is much smaller for the amplitude of puffs than for their frequency . However , there is a noticeable increase in standard deviation from close to 0 . 2 to 0 . 5 . More importantly , there is a small number of large amplitude events that are absent in the first half of the IWI but are present in the second half ( see B and C ) . These events represent the large elementary releases that possess the potential to synchronize to a global release wave . In this paper we have modeled Ca2+ release from clusters of intracellular channels where unbinding of IP3 occurs during release . The possibility of such unbinding appearing even if surrounding IP3 concentrations seem saturating was discovered recently in numerical simulations using a very simple model of spatial coupling of channels [27] . While we have used the mean-field description of Ca2+ concentration within the cluster in the prior publication , we here use the FEM with a locally refined spatial grid and adaptive time steps to accurately calculate the complex spatio-temporal Ca2+ and buffer distributions . In the present publication we focus on the changes of release dynamics and the role of IP3 unbinding for different IP3 concentrations . Our main finding is that there is a complex dynamics in the fraction of subunits bound to IP3 . The character of this dynamical variation depends on the IP3 concentration . For small [IP3] there are relatively rare and asynchronous unbindings during the active release phase ( i . e . , during puffs ) so that IP3 unbinding contributes little to termination of puffs , puff dynamics and lifetime . Typically , during a puff at most one channel becomes unactivatable because of IP3 loss . Nevertheless , the frequent occurrence of puffs leads to accumulated unbinding , so that the number of activatable channels is much smaller than the number of channels in the cluster . Additionally , because of the stochastic nature of unbinding and rebinding , there is a variability of the number of activatable channels at the beginning of each puff . As a result , the dynamics of IP3 loss and rebinding modulates the number of channels that open during a puff , which may contribute significantly to the puff amplitude variability observed in recent experiments [31] , [33] . Note , however , that in recent studies of puffs cells were loaded with EGTA , which suppresses residual domains and thus affects IP3 binding . Therefore , comparison of our current simulations to these findings is not straightforward . The IP3 dynamics at large [IP3] is very different from that observed at small [IP3] . Most importantly , we find a fraction of release events that last much longer than the typical puff . These events exist because , for higher IP3 concentration , a larger number of activatable channels in a cluster is sometimes present and hence larger Ca2+ domains and a reduced probability of channel closing follow . We further find that many channels synchronously dissociate IP3 during the long-lasting release events . It had been shown in [27] that this behavior can be caused by the extended release during long events or waves , during which most channels repeatedly enter the inhibited state . This increasingly exposes the channels to IP3 unbinding , so that most of the channels lose IP3 . This leads to the paradoxical situation , that for larger IP3 concentration more channels dissociate IP3 than for small concentration of IP3 . Eventually , the loss of IP3 causes termination of release because of reduction of activatable channel numbers over a timescale of several seconds . Subsequently , a few seconds after termination , IP3 rebinds to the receptors and appropriates them for the next opening . It is interesting to note , that , because of growing dissociation , the increase of [IP3] from 10 nM to 80 nM is accompanied by only a slight increase of the typical number of activatable channels from 4 to 5 ( Fig . 4 , inset ) . However , it is also pertinent that the variability in activatable numbers at the beginning of an event is increasing with [IP3] so that for large [IP3] sometimes a number of 8 or more channels are available for opening . Statistically speaking , these are the events of long lifetime and a temporal pattern of striking resemblance with global release in Xenopus oocytes [10] , [24] , [30] . Parker and coworkers have shown that for small IP3 concentrations typically only a short release event is observed . For larger IP3 concentration , some release events consist of two phases: a short high-amplitude spike and a small-amplitude and slowly decaying second release phase , similar to what is observed in our simulations during long-lasting events . The similarities in the temporal evolution gave us the cue to consider sustained release events in our simulations as global releases or waves . For further comparison with experiments , we need to classify events based on a quantitative criterion and we here use the potential of an event to trigger release in neighboring clusters as that quantity . This enables us to identify the two types of events as global release or waves and as the small Ca2+ puffs that occur in between waves . In our model , the occurrence of puffs between waves is modulated by the slow recovery of IP3 binding to the receptors . We compare our results in detail to experimental statistics from waves in Xenopus oocytes . Both experiments and simulations show almost no puffs in the early phase after each wave , but a strong increase of puff frequency during the IWI . In contrast , the increase of puff amplitudes during the IWI is modest , while a small number of large amplitude events occurs towards the end of the IWI . To our knowledge , we present the first model that exhibits the increase of puff frequency before waves , which is a hallmark of Ca2+ stochasticity . It is questionable whether other possible refractory mechanisms , including ER depletion of Ca2+ , can generate a similar effect . We here argue that the frequency of global release events derives from the frequency of large events as defined in our model . The fact that in oocytes not every Ca2+ puff causes a wave can be immediately understood , since the stochastic variability of Ca2+ release events produces only a subset of events that can lead to global release . The resulting period of waves is indeed in the range of experimental observations in many cell types . We have here assumed that the period of waves is set by a single cluster site in the cell that emanates the waves . This assumption may be justified for some cell types , where waves appear repeatedly at focal sites [11] . For other cell types it may be more realistic to assume that every cluster site can initiate a wave and the period of waves is then set by the first long release event in any of the clusters and the period of waves or global oscillation becomes shorter than the IWI of the individual cluster . At the center of the effects described in this paper is the repeated burst-like opening of channels because of residual Ca2+ and the subsequent IP3 unbinding during release . How realistic is this scenario ? IP3 unbinding depends first on the respective dissociation constants for IP3 binding , specifically the constants and . The values of and in the DYK model , as obtained from fitting to patch clamp experiments , differ by orders of magnitude , which reflects the fact that the open probability peak of the IP3R channel moves to larger Ca2+ concentrations with increasing [IP3] ( see [27] for a detailed discussion ) . This contrast in dissociation constant was incorporated in many models of Ca2+ release . However , since the time a channel spends in the inhibited state is relatively short , it is also necessary for an effect that the unbinding rate be sufficiently large . Note that this rate constant is not directly observable from experiments and was in the present study chosen to the order of 1 s−1 to allow unbinding during long release events ( duration 1 s ) . It should also be noted that IP3 unbinding during occupancy of the inhibited state is not a unique property of the DYK scheme . Other models of IP3 gating also allow dissociation of IP3 from the inhibited configuration . This process presumably reflects the shift of the open probability curve with growing [IP3] and should therefore be universal for IP3 gating models . This particularly holds for sequential binding models [56] and newer models that are based on channel states and not subunit states [29] . Finally , we would like to suggest further experimental studies that could help to validate our approach . One experimental verification could result from the model's prediction of long duration events even in absence of coupling to other clusters . These long puffs do so far not appear in experiments since addition of exogenous buffer EGTA is used to prevent waves at large IP3 concentration . In contrast , our finding could be tested from experiments on genetically engineered cells that only possess one cluster . Further indirect evidence may be obtained from comparison to experiment with repeated stimulations with IP3 and Ca2+ which were described for oocytes [10] and , more recently , for Purkinje cells [57] . Such experimental protocols require simulations with fixed initial conditions for comparison , which is different from our current focus on long-time simulations . Another debated issue of calcium oscillations regards the occurrence of concomitant oscillations of free IP3 concentrations . It is puzzling that in some cells [IP3] oscillates together with [Ca2+] while in other cells this is not the case [58] . This [IP3] oscillation has been attributed to further metabolic processes , where changes in Ca2+ concentration affect the synthesis and degradation of [IP3] [59] . However , our model provides an alternative explanation in that for certain situations the unbinding of IP3 from the receptors during a wave could lead to larger free IP3 concentrations and rebinding to lower concentrations . Quantitative comparison of our predictions of this oscillation mechanism may be possible with a more detailed knowledge of concentrations of IP3 receptors in different cell types and provide an explanation for the presence or absence of concomitant [IP3] oscillations independent of possible metabolic IP3 processes .	Intracellular calcium oscillations and waves are paramount cellular signals . The frequency of global release events regulates , for example , expression of genes . Knowledge about the mechanism of oscillations and the factors that determine their frequency is crucial when aiming at the control of downstream processes . Many experimental and modeling studies have demonstrated that a calcium cycle consists of both deterministic and stochastic components , but the respective mechanisms are under debate . Here we aim to clarify both components by analyzing calcium release in Xenopus oocytes and a computational model for a cluster of IP3 receptor channels . Just as in calcium fluorescence traces , in the computed sequences some of the events are prolonged releases lasting for several seconds . We find that synchronized unbinding and rebinding of IP3 cause this modulation in time . Our experimental and computational data show agreement in many properties including wave period , extended refractoriness , and release amplitude . Our analysis suggests that global calcium concentrations are stochastically oscillating because of a modulated but random appearance of high-release events . Thus our approach integrates both deterministic properties and stochasticity of waves , and reveals key control parameters of calcium oscillations .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "ion", "channel", "gating", "ion", "channels", "systems", "science", "mathematics", "stochastic", "processes", "calcium", "signaling", "computer", "and", "information", "sciences", "proteins", "calcium", "channels", "nonlinear", "dynamics", "biophysics", "probability", "theory", "physics", "biochemistry", "signal", "transduction", "cell", "biology", "calcium-mediated", "signal", "transduction", "biology", "and", "life", "sciences", "physical", "sciences", "cell", "signaling" ]	2015	Modulation of Elementary Calcium Release Mediates a Transition from Puffs to Waves in an IP3R Cluster Model
Genome-scale datasets have been used extensively in model organisms to screen for specific candidates or to predict functions for uncharacterized genes . However , despite the availability of extensive knowledge in model organisms , the planning of genome-scale experiments in poorly studied species is still based on the intuition of experts or heuristic trials . We propose that computational and systematic approaches can be applied to drive the experiment planning process in poorly studied species based on available data and knowledge in closely related model organisms . In this paper , we suggest a computational strategy for recommending genome-scale experiments based on their capability to interrogate diverse biological processes to enable protein function assignment . To this end , we use the data-rich functional genomics compendium of the model organism to quantify the accuracy of each dataset in predicting each specific biological process and the overlap in such coverage between different datasets . Our approach uses an optimized combination of these quantifications to recommend an ordered list of experiments for accurately annotating most proteins in the poorly studied related organisms to most biological processes , as well as a set of experiments that target each specific biological process . The effectiveness of this experiment- planning system is demonstrated for two related yeast species: the model organism Saccharomyces cerevisiae and the comparatively poorly studied Saccharomyces bayanus . Our system recommended a set of S . bayanus experiments based on an S . cerevisiae microarray data compendium . In silico evaluations estimate that less than 10% of the experiments could achieve similar functional coverage to the whole microarray compendium . This estimation was confirmed by performing the recommended experiments in S . bayanus , therefore significantly reducing the labor devoted to characterize the poorly studied genome . This experiment-planning framework could readily be adapted to the design of other types of large-scale experiments as well as other groups of organisms . To understand the functions of gene products and the interplay between them , significant effort has been spent on performing and analyzing genome-wide expression profiling experiments . Compared to traditional experiments that study protein functions on the single-gene scale , modern high-throughput techniques efficiently characterize expression of the whole genome . One of the most popular techniques is the gene expression microarray , with thousands of expression profiles available for the commonly-studied species . For example , in the Gene Expression Omnibus repository , over 150 datasets comprised of 2400 conditions were available for Saccharomyces cerevisiae as of 2007 [1] , with data continuing to appear at an enormous rate . These large scale data have been used to accurately predict gene functions [2]–[4] , protein-protein physical interactions [5] and functional relationships for yeast [6] and other model organisms [7] , [8] , as well as human [9] . On the other hand , new genomes are being sequenced at an exponentially growing rate [10] , with more than 2 , 200 genome sequencing projects completed or ongoing to date . These sequencing efforts accelerate our understanding on diverse species , but identifying the gene sequence is not sufficient to define the biological role of its product , and functional annotation of these genomes lags far behind sequencing . Many of these newly sequenced species are amenable to further experimental study in the lab . The lack of such functional annotation is partly due to the fact that experiments in poorly-studied species are still mainly based on expertise experience or heuristic trials , rather than using a systematic approach based on comparative functional genomics . Although the heuristic approach is useful in directing specific experiments , it is often far from optimal for a systematic functional annotation of all proteins ( or at least the majority ) in a newly-sequenced genome . Furthermore , experiments that target a specific biological process may also provide accurate functional signal for additional pathways . For example , hyperosmotic shock datasets not only elucidate stress responses , these experiments provide information on regulation of DNA replication initiation because of the cell cycle arrest that occurs under this condition . This functional coverage information is often implicit . We propose here that systematic analysis and quantification of this information in a well-studied species could be the foundation of a systematic experimental design scheme in related poorly-studied species . In recent years , computationally directed experiments have been applied to different fields . The most prominent application domain is the prediction of protein function with follow-up in vivo tests . For example , the prediction results of an ensemble of three algorithms have been used to direct experiments to find genes required for mitochondrial biogenesis [2] . Experiments that detect physical or genetic interactions have also been directed through computationally integrating quantitative genetic interactions and TAP-MS data [11] . The recent development of the Robot Scientist “Adam” marks the state-of-art pipeline of computationally directed studies , which generate hypotheses and experimentally test them [12] . However , these computational efforts have not been extended to direct experiments in a poorly studied species for functional annotation based on existing knowledge in a well characterized species . In this paper , we developed a systematic approach to recommend experiments for functional annotation of S . bayanus ( a poorly-studied yeast species ) based on the wealth of available gene expression data in the model organism S . cerevisiae ( baker's yeast ) . The system identifies experiments that are informative of genes participating in each function and then uses an optimized combination of the predictive power of each experimental treatment with the coverage overlap between treatments to rank a list of experiments that are able to predict the maximum spectrum of biological processes using the minimum number of arrays . Based on functional analysis we estimated that experimentalists can achieve similar functional coverage in the same or related species with less than 10% of the arrays . We further carried out these recommended experiments in S . bayanus and the resulting arrays achieved similar functional coverage to all the existing S . cerevisiae arrays with a 10 fold reduction in labor . Our approach is readily adaptable to other sequencing-based measurements of expression and to measurements of protein and metabolite levels , and is potentially applicable to other large-scale experiment types . Genes play individualized roles in the cell and one gene product can be involved in several different biological processes . Hence for a given experimental treatment or genetic perturbation , we would expect that genes of some functional groups respond more strongly than others . Thus , different datasets are more or less informative of particular processes , including processes that are not necessarily the direct target of the experiment's design . This information , i . e . , the informativeness of a dataset when used to predict certain biological process , could be used to select experimental treatments to target certain biological process . However , this information is often implicit and must be quantified statistically . The Brem et al . , 2005 dataset , for example , represents the progeny from an outcross between two strains , executed with the goal of using the resulting expression profiles as phenotypic traits in genetic mapping . It performs well in predicting a wide range of biological processes , including terms not directly related to genetic crossing such as electron transport and sulfur metabolic process ( Figure 2A ) . Our method can rank the candidate experiments according to their informativeness , or how much information each experiment provides on telling whether a gene is related to a certain biological process . We propose to access such informativeness by assessing the predictive performance of a machine learner that uses the data in the experiment under consideration to predict proteins involved in that process . The intuition is that the machine learner will achieve higher accuracy if the training data provides more information about the specific process ( at least in terms of functional annotation ) , and thus this experiment is likely to be highly effective in interrogating this functional group in the evolutionarily related , less well-studied organism . To do this , we use Support Vector Machine ( SVM ) , a state-of-the-art machine learning algorithm [4] , [13] , though the method can be used with any machine learning approach . We used bootstrap cross-validation [14] to characterize the performance of our S . cerevisiae microarray data collection in predicting Gene Ontology ( GO ) biological process ( BP ) terms that are annotated with 10 to 500 genes in S . cerevisiae ( Figure 2B ) . We employed several different measurements , including AUC ( area under the receiver operating characteristic curve ) , which characterizes the overall ability of a dataset resulting from an experimental treatment to predict function for proteins in a certain biological process; and precision ( accuracy ) at 1 percent , 10 percent , 50 percent and 80 percent recall , which focuses more on discovering new genes . It is important to note that these measures assess the ability of experimental treatments to interrogate participants in specific biological processes . Although these measures are influenced by the quality of data as well , we found that they are sensitive to different experimental treatments , because these measures are highly correlated across GO terms in the same treatment between S . cerevisiae and S . bayanus ( see below ) . If a different property of experimental design is desired , such as ability to assess regulatory interactions or binding partners , this particular characteristic can be optimized through the same evaluation methodology . With this information , experimental treatments can be ranked by how effective they are in predicting a given biological process . We make this method available to the scientific community through our interactive website , where the users can search for the most relevant experiment ( s ) for the biological process of interest . An important phenomenon we observed through the function-dataset informativeness analysis is that some biological processes are well represented in many datasets , while others are only reflected in a small fraction of datasets ( Figure 2B ) . For example , signals for the group of ribosome- and translation-related biological processes are present in the majority of the S . cerevisiae expression datasets . Metabolism , mitosis , carbohydrate , and amino acid-related terms are also well represented by many datasets . However , most of the biological processes , for example , transcription , cell cycle , stress and transport-related terms , are only detectable in particular datasets . Datasets that have the best overall performance in our analysis showed this same range of variability in the terms that they could cover ( Figure 2A ) . For example , transcription from RNA polymerase I promoter ( GO:0006360 ) and electron transport ( GO:0006118 ) are reflected in most of the top 10 datasets . On the other hand , response to toxin ( GO:0009636 ) has strong signal only in the Chitikila02 dataset [15] and the Brem02 set1 data [16] , and function of proteins in peroxisome organization and biogenesis ( GO:0007031 ) is well represented only in the Boer05 dataset , which profiles the expression pattern of a leu3 mutant strain [17] . The phenomenon of different functional sensitivity of different expression datasets is consistent with previous studies using different machine leaning methods to estimate informativeness [1] , [18] . Therefore , both the accuracy and the redundancy between datasets should be considered for planning experiments . To quantify the overlap in information between datasets , we calculated pair-wise conditional mutual information ( CMI ) . This CMI analysis is highly informative of functional redundancies and therefore is critical to our experiment planning system as shown below . Intuitively , CMI quantifies the overlapping information between datasets in predicting functions . The CMI analysis effectively identified datasets that result from similar experimental treatments ( Figure 3 ) . For example , the Brem et al . 2002 , Brem et al . , 2005 and Yvert et al . , 2003 datasets have very high mutual information , and in fact are overlapping subsets of the same type of experiment [16] , [19] , [20] . Less obvious is identification of datasets that are different in their treatments but essentially targeting the same biological processes , for example , the Tai et al . 2005 and Boer et al . , 2005 datasets have very high mutual information , although the former is a nutrient limitation treatment , while the latter studies the expression in leucine auxotroph mutants . This overlap is likely because both experiments were performed in chemostat culture which may induce similar responses in yeasts . [17] , [21] . Less obvious relationships identified by the CMI analysis are those among datasets that do not directly share the same or similar experiment treatments but still contain high mutual information . For example , a set of cell cycle-related experiments are clustered together by their high mutual information , including Spellman98 cyclin [22] and the two technical replicates from Iyer , et al . 2001 [23] ( Figure 3 ) . These experiments altered key cell cycle regulators including the cyclins and the transcriptional regulator MBF/SBF . Although these experiments do not analyze a time course of synchronized cells , they measure the transcriptional response to these key regulators and so are very informative about gene expression regulation in the cell cycle . Similarly , stress-response experiments Rutherford01 [24] , Fernandes04 [25] and Gasch00 HOresponse [26] are clustered together , despite the fact that they represent diverse experimental treatments , including iron homeostasis , hydrostatic pressure and hydrogen peroxide . The CMI quantification allows us to statistically identify redundant datasets and avoid such redundancies in experimental recommendations for the less-studied species . Large-scale microarray experiments designed for characterization of the whole genome tend to be of very high accuracy but often include a large number of arrays . For example , among the top performing datasets in predicting gene functions are the Brem et al . , 2005 experiments designed to detect segregation of expression patterns in an outcross [19] and the Hughes et al . , 2000 experiments designed for genome-wide mutation analysis [27] . They include 130 and 300 arrays respectively . Although both datasets are among the top recommended experiments , intensive labor is required to repeat these experiments in a new species . We attempted to minimize the number of arrays in these large-scale microarray experiments while retaining their function prediction capability . Through randomized selection of subsets of datasets , we could estimate the accuracy versus number of arrays included in the subsets . Surprisingly , a rather small fraction of the arrays ( 25–40 ) can achieve very similar performance in overall function prediction to the entire datasets ( Figure 4A and 4B ) . Additional arrays only add to marginal improvement in performance . Therefore only a small proportion of the arrays of these very large scale experiments are required for our experiment planning system . Of course , this does not mean that these additional data are not biologically relevant , in fact , for genetic linkage experiments the entire dataset is informative . Rather , a subset of these experiments of optimized size can be used for this specific goal of functional annotation; if a different biological question is important , the size of the appropriate subset or entire dataset can be estimated specifically for that question ( e . g . regulatory relationship prediction ) . Biological processes differ in their sensitivity to the number of arrays required for reasonable assessment of each process ( Figure 4C ) . For example , we could accurately predict the biological process ‘ribosome biogenesis and assembly’ ( GO:0007046 ) with only about 15 arrays from the Brem et al . , 2005 [19] dataset . Additional arrays add no improvement in predicting this term . Predictions for many of the other biological processes have different sensitivity to the number of arrays . For example , ‘cellular lipid catabolic process’ ( GO:00044242 ) and ‘histidine metabolic process’ ( GO:0006547 ) could be captured with relatively small number of arrays . On the contrary , ‘co-factor biosynthetic process’ ( GO:0051188 ) and ‘G1 phase of mitotic cell cycle’ ( GO:0000080 ) require a large number of arrays to be well characterized . There are also terms like ‘cell growth’ ( GO:0016049 ) , which cannot be captured even using the maximum number of array we tested . For this case , increasing the number of arrays is meaningless . The requirement for the number of arrays on a per-biological process basis was calculated and provided though our online searchable system . For the general recommendation process , we define the minimum number of arrays based on the average AUC across all GO functional SLIM terms [28] and therefore guarantee the overall performance . Our experiment planning system flexibly leverages both the accuracy of each experimental treatment in capturing different functions and the overlap in information between them . We determined the overall accuracy of a dataset by its average AUC across GO functional SLIM terms ( as listed in Figure 2B ) , which are terms curated by biologists to represent functions specific enough for experimental characterization , but which do not have any parent terms satisfying this criterion [28] . The redundancy between datasets was quantified by pair-wise conditional mutual information as described above . A trade-off factor ( α ) between accuracy and redundancy , where a higher value means more weight on accuracy and vice versa , allows flexibility in the experiment design process . In our study , α was optimized through cross-validation; in the web-interface , the users can optimize this factor according to their specific preferences . Recommendation of datasets for functional annotation requires leverage between data precision and redundancy . We applied bootstrap cross-validation [14] to evaluate the ability of the selected set of data in predicting different functions ( Figure 5A ) . We found that we could optimize the function prediction capability of the top 10 datasets by a trade-off factor α = 0 . 9 ( Figure 5B ) . Function prediction by SVM maps the original data into feature space , which relies on the accuracy of the datasets and penalized by the redundancy in information between them . Thus an adjustable trade-off factor is necessary and provides flexibility for the experiment recommendation process . We estimated how effective our approach is in reducing the number of experiments to characterize the overall functionality of the S . bayanus proteins based on the S . cerevisiae gene expression compendium . We integrated the information from our analysis of the minimum number of arrays in the very-large-scale microarray experiments , and the information from our accuracy and redundancy analysis . This gives us an ordered list of experiments in S . cerevisiae , including the number of microarrays that need to be completed in the very large experiments . We experimentally generated an S . bayanus expression data compendium based on the experiments proposed by our system ( GEO accession GSE16544 ) . The list of highly informative experiments ( 250 arrays ) included cell cycle progression , meiosis , diauxic shift , nutrient limitation , stress conditions , and outcross progeny . The S . bayanus data we generated are highly informative for diverse biological processes ( Figure 6 ) . As no S . bayanus functional annotation exists , to assess the coverage of S . bayanus experimental data , we use the gene ontology annotations from S . cerevisiae orthologs of the S . bayanus genes . This is a conservative measurement because not all orthologs are conserved in function , but as most genes are likely to be conserved at least on the level of functional annotations , this measure should provide a reasonable lower bound on performance . We used bootstrapping and a linear SVM classifier to estimate the accuracy of the expression data in functional annotation of GO functional SLIM terms [28] . Interestingly , the ‘informativeness’ is highly similar between matched experiments between S . bayanus and S . cerevisiae , further supporting the validity of our approach ( Figure 6 ) . We find that our dataset of 250 S . bayanus arrays predicts gene function with an average AUC of 0 . 74 , which is very close to the AUC ( 0 . 75 ) of predictions made with a set of 2547 S . cerevisiae arrays ( Figure 7A ) . This is very similar to the theoretical analysis , where we estimated that less than 10% of the arrays of the total 2569 arrays available in the S . cerevisiae repository as of 2007 would achieve similar performance . Such performance requires selection of specific experimental treatments – computational simulation shows that selection of random subsets of experiments from the repository substantially decreases overall accuracy ( Figure 7 ) . This indicates that the experiment planning scheme can significantly reduce the human and technical resources necessary to characterize a newly sequenced species by providing effective guidance for the most informative sets of experiments for functional annotation based on related model organisms or other well-studied species . To further validate our approach , we also compared the performance of individually matched dataset pairs between S . bayanus and S . cerevisiae for all GO terms with more than 30 genes annotated to each . The correlation and the similar range of AUC between the two sets ( Figure 6 ) further supports the recommendation of experiments based on closely-related model organisms . Signals for most of the biological processes are very well represented in our S . bayanus expression compendium , to an extent comparable to the theoretical maximum in S . cerevisiae ( Figure 7A ) . However , a small set of biological processes , including aging , ion homeostasis , hyperosmotic shock , auxotroph starvation , and alternative carbon sources were not well-captured by the experiments , with an AUC less than 0 . 65 . Thus , we used the system to suggest a second round of experiments targeting these particular processes again based on the accuracy and redundancy analysis . The second round experiments included 54 arrays covering 11 biological treatments carried out in S . bayanus ( Text S1 ) . On average we gained a 0 . 006 ( 2 . 5% over random ) improvement in AUC over all GO biological process terms with 10 to 300 genes annotated to each . This minor improvement indicates the saturation of the ability to predict functions based on expression data and transfer of annotation through homology . However , we observed an average of 0 . 012 ( 10% over random ) improvement in AUC in the targeted GO categories , which were poorly predicted in the first round . Five out of seven of the targeted GO categories achieved significantly improved AUC ( Figure 7B ) . Therefore the second round recommends experiments that provide relatively orthogonal information to the first round , indicating the ability of our experimental planning system to extract the information contained in the existing data and to direct further specific experiments . The top improved terms during the second round ( Text S1 ) are well-explained by the additional datasets . For example , we observed a 52% improvement ( 0 . 071 ) in AUC for “double-strand break repair via nonhomologous end joining” ( GO:0006303 ) and a 56% improvement ( 0 . 111 ) for “DNA catabolic process , endonucleolytic ) , most likely due to the addition of MMS and zeocin DNA damage datasets . Starvation experiments might explain the 177% improvement in “nitrogen utilization” ( GO:0019740 ) and “histidine biosynthetic process” ( GO:0000105 ) . Experiments of alternating carbon sources , particularly glycerol , lead to a 35% improvement in our prediction power on “hexose biosynthetic process . ” These observations suggest that our evaluation scheme is well in accordance to the established knowledge in this field . In this paper , we propose that existing genome-scale data in model organisms could facilitate the planning of experiment treatment in poorly-studied species . We demonstrate the feasibility of this approach by planning microarray experiments for a relatively poorly studied yeast species S . bayanus based on an available gene expression data repository for model organism S . cerevisiae . In this framework , we recommend an ordered list of experiments targeting the overall functional annotation of S . bayanus proteins as well as experiments targeting specific biological processes . We also detected the minimum number of arrays to achieve satisfactory performance for some very large-scale microarray experiments . Our framework results in a substantial reduction in the resources we need to characterize the functions of a poorly-studied genome . Our work represents the first attempt for large-scale experiment design of a relatively poorly studied organism based on available data in a related organism , rather than existing functional genomics data in that species , which in the case of S . bayanus does not exist . This method is complementary to designing experiments based on expert knowledge or intuition , which is irreplaceable when targeting in-depth aspects of biology but is likely not to be optimal for generating a large compendium for functional annotation . Of course , after such initial functional annotation , carefully designed experiments will be necessary to ascertain specific relationships within functions and to further explore the functional space . This task should be facilitated greatly by the availability of the initial functional annotations generated based on the experiment design system and the resulting data compendium . The experiment design system is adaptable , and can be extended to other related species groups . The current analysis is restricted to S . cerevisiae and S . bayanus , which are separated from each other by 20 million years . There are thousands of genome projects finished or ongoing , and several model organisms with large amount of genome-scale data . Analysis of these genome-scale data could be used to design experiments for the poorly-studied related species . The 20 millions years distance between S . cerevisiae and S . bayanus is comparable to the sequence divergence between human and mouse [29] . On the other hand , comparative genomics often focuses on less diverged groups , for example , the Drosophila species subgroup or the other sensu stricto yeast species . The extendibility to further related groups , however , remains to be validated by future investigation in the intelligent experiment design field . Nevertheless , currently , GO annotations are often transferred between species of vast distance based on sequence alone . Experiment recommendations such as those described here provide a complementary approach to the current annotation scheme . Indeed , expression patterns of the majority of genes are conserved across species over vast distances ( e . g . from human to mouse , and from Candida to S . cerevisiae ) [30] , [31] , suggesting the likely value of applying such experimental design methods across further distances . Applying the experiment design system could not only facilitate the annotation of these genomes but also provide invaluable resources for cross-species expression comparison . In addition , our current study focuses on the prediction of biological processes . The same approach could be extended to molecular function , cellular component and pathway predictions , as well as predictions of particular types of relationships among proteins ( e . g . physical , regulatory interactions , etc ) . The basic framework could also be extended to other data types , more complex data , and higher organisms . Our current work focuses on microarray data because it is currently the most abundant functional genomics data source for most organisms . This methodology can be readily applied to sequencing-based measurements of expression and to measurements of protein and metabolite level . We expect that as more data of these types become available , applications of this and similar methods will become more common . An extension of this methodology could be developed for other types of large-scale experimental methods , including yeast two hybrid , affinity precipitation , chromatin IP datasets , etc . These data , like microarrays , are often readily available in diverse well-characterized model organisms . Furthermore , data from several model organisms could be integrated together so that more confident experiment planning system could be established . Novel methodology that integrates both sequence data and information from these types of large-scale datasets will ultimately allow us to more accurately and quickly understand the differences and similarities of functions between species . We collected an extensive compendium of S . cerevisiae microarray datasets from diverse sources [32]–[36] . This compendium includes 125 datasets with 2569 arrays . A complete list of publications for these datasets is available on the website supporting this publication http://exprecommender . princeton . edu . To allow reasonable comparison between datasets , we carried out the following normalization steps . For each raw dataset , genes that are represented in less than half of the arrays were removed , and missing values were inserted using KNNimpute [37] with K = 10 , Euclidean distance . Technical replicates are averaged , resulting in datasets with each gene followed by a vector representing its expression values in a series of arrays . Based on the analysis of experiment performance , we observed that some of the biological processes have strong signal in a wide range of experiments , but others are only sensitive to one or a very limited set of experiments . Furthermore , two very accurate datasets may interrogate a highly overlapping set of processes , thus providing largely redundant information in terms of functional annotation . Therefore , when designing a set of microarray experiments for global function profiling , we should not only consider the accuracy of each experiment in predicting function , but also weigh the overlap in information between datasets . We found using a trade-off factor between the two could allow flexibility in experimental design for different applications ( see Results ) . In the following section , we will describe our measurements of accuracy and redundancy and the combination of the two . To leverage the trade-off between accuracy and mutual information , we introduced a trade-off factor α , which linearly combines the two factors: Where P ( X ) is the overall precision of the dataset X in consideration , k is the number of experiment selected before X . This approach allows iterative selection of datasets . Therefore it is suitable for selecting datasets in a species with several experiments available already , which is more likely to be true in the real-world situations . We provide this feature on our website , allowing the user to select the experiments already performed and the desired tradeoff value , so that our system can recommend additional experiments . We identified GO functional SLIM biological processes that are weakly represented in the first round datasets ( below 0 . 65 in AUC and with a minimum of 30 genes annotated to it in S . cerevisiae ) . This list included 7 biological processes in total ( Figure 6 ) . Experimental treatments that best cover each of these biological processes were ranked by accuracy and carried out in S . bayanus in the second round . Based on the recommendations , we carried out the experiments in S . bayanus . The resulting array data are accessible from the GEO database with accession ID GSE16544 . For both first round and second round datasets , we borrowed annotations from S . cerevisiae through orthology and applied bootstrap cross-validation to estimate the error rates as describe in S . cerevisiae accuracy estimation . We developed a website to facilitate exploration of the functional analysis of the microarray data and our recommendation of datasets for yeast species related to S . cerevisiae . This website supports searchable recommendations for datasets targeting specific biological processes and ones targeting the entire functionality of the genome given existing datasets in poorly-studied species . In addition , the number of arrays ( on a per biological process basis ) required for the large-scale datasets is also searchable by the users . This website is publicly available at http://exprecommender . princeton . edu .	Microarray expression experiments allow fast functional profiling of an organism's entire genome and significant efforts are devoted to analyzing the resulting data . Available genome sequences are also increasing quickly . However , it is unexplored how to use available functional genomics data to direct large-scale experiments in newly sequenced but poorly studied species . In this paper , we propose a strategy to systematically plan experimental treatments in the poorly studied species based on their model organism relatives . We consider both the accuracy of the datasets in capturing different biological processes and the redundancy between datasets . Quantifying the above information allows us to recommend a list of experimental treatments . We demonstrate the efficacy of this approach by designing , performing and evaluating S . bayanus microarray experiments using an available S . cerevisiae data repository . We show that this systematic planning process could reduce the labor in doing microarray experiments by 10 fold and achieve similar functional coverage .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "genetics", "and", "genomics/genomics", "genetics", "and", "genomics/gene", "expression", "genetics", "and", "genomics/functional", "genomics" ]	2010	Systematic Planning of Genome-Scale Experiments in Poorly Studied Species
Retroviruses are not expected to encode miRNAs because of the potential problem of self-cleavage of their genomic RNAs . This assumption has recently been challenged by experiments showing that bovine leukemia virus ( BLV ) encodes miRNAs from intragenomic Pol III promoters . The BLV miRNAs are abundantly expressed in B-cell tumors in the absence of significant levels of genomic and subgenomic viral RNAs . Using deep RNA sequencing and functional reporter assays , we show that miRNAs mediate the expression of genes involved in cell signaling , cancer and immunity . We further demonstrate that BLV miRNAs are essential to induce B-cell tumors in an experimental model and to promote efficient viral replication in the natural host . Virally encoded miRNAs were first identified in DNA viruses . Most of these miRNAs are generated from endonucleolytic cleavage of long viral transcripts . Retroviruses were initially assumed not to encode miRNAs because of the potential self-cleavage of their RNA genome . Human immunodeficiency virus ( HIV ) , West Nile virus ( WNV ) , Dengue virus ( DENV ) , hepatitis A virus ( HAV ) and Ebola virus ( EBOV ) potentially express small non-coding RNAs [1–4]; however , no significant amounts of viral miRNAs have been identified by high-throughput sequencing of RNA from cultured cells infected with these different viruses [5–7] . Recently , four retroviruses , i . e . , bovine leukemia virus ( BLV ) , bovine foamy virus ( BFV ) , avian leucosis virus ( ALV-J ) and simian foamy virus ( SFV ) were shown to express viral miRNAs in infected cells [8–11] . ALV-J encodes a single viral miRNA ( XSR ) from a Pol II-transcribed precursor that is processed by the canonical biogenesis pathway . ALV-J thus tolerates some Drosha and Dicer cis-cleavage of the viral genome . In contrast , BLV and BFV massively express viral miRNAs driven by Pol III promoters embedded within their proviral genome . These viruses avoid unproductive cleavage of their genomic RNA because only the subgenomic Pol III transcripts are processed into miRNAs . The BFV long terminal repeat U3 region is transcribed into a 122-nucleotide pri-miRNA that is subsequently cleaved into two pre-miRNAs and processed in three viral miRNAs . The BLV genome encodes a cluster of five miRNA hairpins from a proviral region lacking significant open reading frames located just 3' to the envelope gene . These hairpins undergo maturation into ten miRNAs in a Drosha-independent pathway [8 , 12 , 13] . BLV naturally infects the bovine species ( Bos taurus ) , zebu ( Bos indicus ) and water buffalo ( Bubalus bubalis ) . BLV infection remains mostly asymptomatic in a large majority of infected animals . Only approximately one-third of BLV-infected cows will develop benign lymphocytosis . In a minority of cases ( approximately 5% ) , BLV infection leads to B-cell leukemia/lymphoma after long latency periods ( 7–10 years ) . Experimental inoculation of sheep with BLV recapitulates the different steps of oncogenesis with higher frequencies ( up to 100% ) and relatively short latency periods ( approximately 2 years ) [14] . Remarkably , BLV persists and replicates in vivo in the absence of significant levels of viral mRNA transcription [15–17] . In contrast , BLV miRNAs are massively expressed in the infected cells , representing up to approximately 40% of all cellular miRNAs [12] . Because ribonucleic acids are likely less immunogenic than viral antigens , BLV miRNAs could thus modify the cell fate and concomitantly escape from immune recognition . Although miRNAs encoded by BLV are massively expressed , their biological significance in viral replication and pathogenesis is still unknown . In this report , we specifically addressed this question in the BLV system using a reverse genetics approach . To investigate the role of the miRNAs in the natural host , a BLV provirus isogenic to a wild-type molecular clone but devoid of the intergenic sequences located between the env and R3 genes ( pBLV-ΔmiRNA ) was constructed ( Fig 1A ) and inoculated into 3 calves . All three animals produced specific antibodies against BLV gp51 envelope protein and became persistently infected by the deleted virus , demonstrating that the miRNAs are dispensable for infectivity . Serial PCR amplifications of peripheral blood mononuclear cell ( PBMC ) DNA with primers flanking the miRNA region and subsequent sequencing confirmed that pBLV-ΔmiRNA was indeed infectious in vivo ( Fig 1B ) . Quantitative RT-qPCR further revealed the lack of viral miRNA expression in these animals ( B1-3p , B2-5p , B3-3p , B4-3p and B5-5p ) , as in a control bovine B-lymphocyte cell line ( BL3 ) ( Fig 1C ) . In contrast , the miRNA copy numbers measured in primary PBMCs from wild-type infected calves ( Bo-WT ) were similar to the levels of BL3 cells transduced with a lentiviral vector expressing the BLV miRNAs ( BL3-miRNA , Fig 1C ) . Because a minimal heptameric seed sequence is sufficient for activity , the ten BLV miRNAs are predicted to substantially modify the host transcriptome . To address species-specificity , cell type adequacy and biological relevance , gene expression was characterized in parallel in primary PBMCs and miRNA-transduced BL3 lymphocytes . High-throughput RNA sequencing was performed with pLenti-miRNA-transduced BL3 cells ( BL3-miRNA ) or control pLenti-transduced BL3 ( BL3 ) and with PBMCs from wild-type BLV ( Bo-WT ) , pBLV-ΔmiRNA ( Bo-ΔmiRNA ) or mock-infected calves . Fig 2A illustrates the expression coverage profile corresponding to the granzyme A ( GZMA ) locus . GZMA mRNA was expressed in control lymphocytes ( BL3 ) but not in pLenti-miRNA-transduced BL3 cells ( BL3-miRNA ) . Consistently , GZMA transcripts were present in PBMCs from pBLV-ΔmiRNA-inoculated animals ( Bo-ΔmiRNA ) and uninfected controls but were almost undetectable in Bo-WT calves ( Fig 2A ) . The combined analysis of BL3 cells and PBMCs thus identified mRNAs associated with the presence of the miRNAs in B lymphocytes and in vivo . Panel B of Fig 2 provides the top list of mRNAs that match these criteria . Gene ontology analysis using MSigDB software ( Molecular Signature DataBase from the Broad Institute ) identified a network of the most enriched pathways that included cell signaling , cancer genes and immune response ( Fig 2C ) . The expression networks of statistically significant and biologically relevant genes revealed intramodular hubs in consensus modules ( Fig 2C ) . Based on their connection frequency , the following six hub genes were selected for further functional analysis ( boxes on Fig 2C ) : PPT1 ( palmitoyl-protein thioesterase 1 ) , ANXA1 ( annexin A1 ) , GZMA ( granzyme A ) , PIK3CG ( phosphatidylinositol-4 , 5-bisphosphate 3-kinase , catalytic subunit gamma ) , FOS ( FBJ murine osteosarcoma viral oncogene homolog ) and MAP2K1 ( mitogen-activated protein kinase kinase 1 or MEK1 ) . BLV miRNA functionality was evaluated in BL3 cells by reporter constructs containing gene cDNAs cloned downstream of the Renilla luciferase gene . BLV miRNAs silenced the luciferase activity of reporter plasmids containing the full-length cDNAs of FOS , GZMA and PPT1 ( compare BL3-miRNA with BL3 on Fig 3A ) . As a control for specificity , luciferase expression of a backbone reporter lacking cDNAs was unaffected by BLV miRNAs ( S3 Fig ) . In contrast , ANXA1 , MAP2K1 and PIK3CG reporters were not down-regulated under similar conditions , indicating that these genes were not direct miRNA targets ( Fig 3A ) . The FOS , GZMA and PPT1 mRNAs were predicted by Sfold and STarMiRNA software to be direct interactors of miRNA B4-3p ( Fig 3G and S1 Fig ) . Mutation of the predicted miRNA binding sites in the full-length cDNAs of FOS , GZMA and PPT1 abrogated miRNA silencing ( Fig 3B ) . To test the selective effect of BLV miRNA B4 on the FOS , GZMA and PPT1 mRNAs , HEK293T cells were transfected with a pSUPER plasmid encoding only miRNA B4 ( Fig 3C–3F ) . The levels of miRNA expression achieved in HEK293T cells were biologically relevant as they were similar to those measured in PBMCs ( S2 Fig ) . BLV miRNA B4 silenced the luciferase activity of reporter plasmids containing the full-length cDNAs of FOS , GZMA and PPT1 ( compare pSUPER with pSUPER-B4 on Fig 3C ) . Mutation of the predicted miRNA binding sites in the full-length cDNAs of FOS , GZMA and PPT1 abrogated the miRNA B4 silencing ( Fig 3D ) . Insertion of the predicted target ( Fig 3E ) but not of a mutated version ( Fig 3F ) in the 3'UTR of the luciferase gene recapitulated the silencing activity of the miRNAs . These functional reporter assays thus identified direct ( FOS , GZMA and PPT1 ) and indirect ( ANXA1 , MAP2K1 and PIK3CG ) targets among biologically relevant genes whose expression is affected by BLV miRNAs in vivo . To evaluate the role of the miRNAs in viral fitness , 12 calves were inoculated either with the wild-type molecular clone or with the recombinant pBLV-ΔmiRNA . The proviral loads profile indicated that the viral burden at 28–30 days post-inoculation was significantly lower in pBLV-ΔmiRNA infected animals compared to wild-type levels ( Fig 4A Mann-Whitney test , p = 0 . 04 ) . The integrity of the miRNAs was also important to maintain stable proviral loads in the long term ( i . e . , at 252–254 days on Fig 4A; Mann Whitney test , p = 0 . 03 ) . Nested PCR and sequencing indicated the absence of reversion of the mutation at any time post-infection ( S4 Fig ) . Consistently , spontaneous ex-vivo transcription of viral mRNAs was reduced in the animals infected with pBLV-ΔmiRNA compared to wild-type levels ( Fig 4B Mann Whitney test , p = 0 . 009 , p = 0 . 004 and p = 0 . 002 for GAG , ENV and TAX , respectively ) . miRNAs were abundantly expressed in PBMCs ( Fig 4C ) as well as in the plasma ( Fig 4D ) of wild-type BLV-infected calves and correlated with proviral loads but were undetectable in the pBLV-ΔmiRNA deletants . The expression of miRNA-regulated transcripts ANXA1 , FOS , GZMA , MAP2K1 , PPT1 and PIK3CG further validated the transcriptomic profiling data in a larger series of animals ( Fig 4E ) . Together , these observations demonstrate that ablation of the BLV miRNAs impacts replication fitness in the natural host . Due to the low frequency of tumor development and the long latency period , the evaluation of the role of miRNAs in oncogenesis in the bovine species would take decades . Although sheep are not naturally infected by BLV , this model recapitulates the main characteristics of bovine leukemia-lymphoma and allows one to address this important question . Replication efficacy of the wild-type and pBLV-ΔmiRNA viruses was initially similar during early stages post-inoculation ( NS , Fig 5A ) . At later times , the miRNA deletant was unable to maintain wild-type replication ( p = 0 . 04 according to the Mann Whitney test ) . The levels of miRNA expression in PBMCs and in the plasma were similar to those measured in the calves ( Fig 5B and 5C ) . Since BLV miRNAs are abundantly expressed in the plasma , GZMA transcription could be affected in other cell types ( CTLs and NK cells ) . Cell sorting and RT-qPCR demonstrated that miRNA-associated GZMA down-regulation is restricted to the B cell lineage ( S4 Fig ) . As in the bovine species , replication of the pBLV-ΔmiRNA deletant in sheep occurred in the absence of reversion ( S5 Fig ) . Fifty percent of the sheep infected with wild-type BLV developed leukemia-lymphoma within 22 months , confirming previous rates reported in the literature ( Fig 5D ) [18] . Compared to the high incidence of leukemia-lymphoma in the sheep inoculated with parental BLV , none of the animals infected with pBLV-ΔmiRNA displayed any clinical sign of oncogenicity ( Fig 5D ) . These data thus reveal a dramatic suppression of oncogenicity subsequent to the loss of BLV miRNAs . The high abundance of BLV miRNAs in tumors suggested a possible role in viral replication and oncogenesis [8 , 12] . In this report , we have provided mechanistic and functional evidence demonstrating that the BLV miRNAs modify host gene expression to promote viral persistence and induce pathogenicity . The goal of the transcriptomic analysis was to identify the global changes associated with the BLV miRNAs in the natural host . The rationale was based on the merged analysis of a well-characterized cell system ( BL3 bovine cells transduced with miRNAs ) and primary PBMCs isolated from infected cows . The main reason of this strategy is that unbiased isolation of BLV-infected B cells from an infected animal is technically still impossible . Therefore , RNA sequencing was performed in parallel with miRNA or mock-transduced BL3 cells and primary PBMCs from cows inoculated with wild-type or recombinant pBLV-ΔmiRNA viruses . The objective was not to produce an exhaustive list of direct RNA targets but rather monitor the global effect of the miRNAs . High-throughput sequencing of RNA isolated from immunoprecipitated Ago-miRNA-mRNA complexes would identify only direct BLV miRNA targets . Transcriptomic profiling by high throughput RNA sequencing identified genes whose expression is affected by miRNAs in B-lymphocytes and primary PBMCs ( Fig 2B ) . This approach revealed a list of biologically relevant genes that are directly ( GZMA , FOS , PPT1 ) or indirectly ( ANXA1 , MAP2K1 and PIK3CG ) targeted by the BLV miRNAs ( Fig 3 ) . In contrast , peroxidasin homolog ( PXDN ) identified by miRNA/RNA duplex prediction analyses and reporter assays was not a biologically relevant miRNA target in vivo [8 , 12] . Similarly , tumor suppressor HMGbox transcription factor 1 ( HBP1 ) whose expression is halved in sheep tumor cells [12] is unaffected in bovine B-lymphocytes expressing the BLV miRNAs ( S6 Fig ) . Gene ontology analysis revealed intramodular hubs in two consensus modules that relate to apoptosis and immunity ( GZMA , PPT1 and ANXA1 ) or cell signaling and oncogenesis ( FOS , MAP2K1 and PIK3CG ) . In particular , the serine protease GZMA expressed mostly by natural killer ( NK ) cells and cytotoxic T-lymphocytes ( CTL ) but also by B cells under inflammatory conditions [19] , induces caspase-independent apoptosis [20 , 21] . Knockout of GZMA in mouse models sensitizes to the animals viral infections , indicating a central role in immunity [22] . Inactivation of GZMA by Kaposi's sarcoma-associated herpes virus ( KSHV ) miRNA-K12-6 further supports the significance of this target in other viral life cycles [23] . Our data ( S4 Fig ) show that GZMA expression is reduced by BLV miRNAs in B-lymphocytes but not in non-B cells . It remains nevertheless possible that microRNAs are transiently transferred via the immunological synapse upon contact with CTL or NK cells . Another important BLV miRNA target is FOS , which mediates the primary response to B-cell receptor signaling upon dimerization with c-JUN in the AP1 complex [24–26] . Interestingly , Tcl1-directed inhibition of AP-1 transcriptional activity is associated with a CLL-like disease in mice [27] . Similarly , transgenic mice overexpressing miR-29 , a cellular miRNA sharing the BLV miRNA B4-3p seed sequence and repressing the FOS transcript [28] , develop B-cell tumors [29] . FOS expression is also repressed by KSHV-encoded miRNA-K12-11 in lymphoma B-cells [23 , 30] , illustrating common pathways of virus-induced oncogenesis . A third hub transcript that directly interacts with BLV miRNAs is PPT1 , a glycoprotein involved in the catabolism of lipid-modified proteins during lysosomal degradation . PPT1 removes thioester-linked fatty acyl groups from cysteine residues and modulates tumor necrosis factor alpha ( TNFα ) signaling [31] . Because PPT1-deficient fibroblasts are partially resistant to TNF-induced cell death , BLV miRNAs could promote survival via this pathway [31] . RNA sequencing also revealed a series of genes whose expression is modified by pathways not involving direct seed-to-target interactions such as ANXA1 , a mediator of macrophage phagocytosis . Together , the experimental data from this report and evidence from the literature highlight common mechanisms of transformation shared by other DNA viruses or those involved in human leukemia . Assessing the role of viral miRNAs in cell lines faces objections of biological relevance in vivo . Transduction of BLV miRNAs into the BL3 cell line was not associated with alterations of cell proliferation or apoptosis ( S7 Fig ) . BLV miRNAs do not directly impact on viral mRNAs or proteins in vitro ( S8 Fig ) . In contrast , ablation of the miRNAs had significant effects on viral replication in the natural host ( Fig 4 ) . Compared to the wild-type controls , the proviral loads associated with miRNA deletants were indeed significantly lower during the primary infection and decreased regularly in the long-term . Whether miRNAs deletion will ultimately lead to an abortive infection remains to be determined but is predicted to require extensive periods of several years . Notwithstanding , a decrease in viral burden correlates with a reduction in viral transmission as well as pathogenicity . Consistently , our preliminary evidence indicates that the miRNA deletant does not transmit from cow-to-calf ( S9 Fig ) . Considering the low frequency ( 5% ) and long latency period ( 7–10 years ) of tumor development in the natural host , assessing the role of the miRNAs in the bovine species is an interesting but almost unanswerable question . Nevertheless , observations in the sheep experimental model indicate that BLV miRNAs contribute to leukemogenesis ( Fig 5 ) . Because sheep develop a leukemia/lymphoma that closely mimics natural pathogenesis , BLV miRNAs are speculated to be important mediators of oncogenesis in the bovine species . It should be mentioned that the reduced proviral load observed in the pBLV-ΔmiRNA-infected animals is likely to cause reduced pathogenicity . In fact , it is not possible to dissociate both parameters in vivo since development of leukemia positively correlates with proviral loads in the PBMCs [32 , 33] . Whether BLV miRNAs exert a direct oncogenic role by targeting specific cellular transcripts independently of viral replication remains to be demonstrated . A role of viral miRNAs in pathogenesis has important consequences for the safety of an anti-BLV vaccine that is currently under development [34] . Considering the evidence reported here , a safe live-attenuated vaccine should also include a deletion of the miRNAs to limit the risk of disease . Another consequence of this report relates to human health , although this topic is still controversial [35 , 36] . Because viral ( BLV miRNA B4 ) and cellular ( miRNA-29 ) miRNAs share identical seed sequences , there is also a potential threat for zoonotic-induced oncogenesis , as recently suggested by the association of BLV and breast cancer [37] . Plasmid pBLV-WT ( GenBank: JC613347 . 1 ) contains an infectious and pathogenic wild-type BLV provirus ( strain 344 ) . An isogenic molecular clone ( pBLVΔ-miRNA ) was constructed by deleting pBLV-WT of the miRNA coding region ( nucleotides 6170 to 6736 according to the BLV reference genome NC_001414 . 1 ) ( see S1 Text ) . Calves and sheep were inoculated by subcutaneous injection of pBLV-WT or pBLVΔ-miRNA as described in reference [38] . At regular intervals of time , heparinized blood was collected by jugular venipuncture and an aliquot of plasma was cryopreserved . PBMCs were isolated by Lymphoprep density gradient centrifugation ( Stemcell technologies ) , resuspended in fetal calf serum containing 10% DMSO ( Sigma-Aldrich ) and frozen at -80°C or in liquid nitrogen . The BLV sequences corresponding to the miRNA region ( nucleotides 6170 to 6759 of the reference BLV genome NC_001414 . 1 ) were inserted into a lentiviral vector ( pLenti6 , Life technologies ) to construct pLenti-miRNA . HEK 293T cells were transfected using lipofectamine 2000 ( Life Technologies ) with pLenti6 or pLenti-miRNA vectors together with pCAG-HIVgp and pCMV-VSV-G-RSV-Rev ( provided by Masahiro Fujii , Niigata University , Japan ) . Forty-eight hours after transfection , the supernatant was collected and incubated with BL3 cells in presence of polybrene ( 10μg/ml ) . BL3 is a bovine B cell line with lymphoblastic morphology derived from a naturally occurred lymphosarcoma [39] . After selection with blasticidin ( 10μg/ml ) , correct miRNA expression in BL3-miRNA cells was verified by RT-qPCR . Total RNA was isolated from 3 independent batches of BL3 and BL3-miRNA cells using the miRNANeasy kit ( Qiagen ) . In parallel , RNA was also extracted in triplicate from bovine PBMCs isolated from pBLV-WT , pBLV-ΔmiRNA and mock- infected calves ( 3 animals in each category ) . Proviral loads in 100 PBMCs of these animals were comparable: 4 , 0 . 4 and 5 . 5 copies ( pBLV-WT ) and 2 . 3 , 1 . 5 and 1 . 7 copies ( pBLV-ΔmiRNA ) . Truseq Stranded mRNA libraries were prepared according to the manufacturer’s instructions ( Illumina ) and indexed 100bp paired-end runs were acquired on a HiSeq2000 Illumina sequencer . After mapping of the reads with the Tophat software , differential expression analysis was performed with Cufflinks . Details on plasmid constructions are provided in the Supporting Information file S1 Text . After extraction of BL3 RNA , reverse transcription and PCR amplification , a series of luciferase reporter plasmids ( ANXA1 , GZMA , FOS , MAP2K1 , PIK3CG and PPT1 ) were constructed by inserting the full-length bovine cDNA sequences into psiCHECK2 ( Promega ) . Mutations of the predicted target sequences in these reporter plasmids were introduced using the Q5 directed mutagenesis kit ( NEB ) . Reporter constructs ( 500ng ) were transfected into 200 , 000 BL3 or BL3-miRNA cells using the Neon Transfection System ( ThermoFisher Scientific ) . Electroporation parameters included 3 pulses of 1350V during 10ms . Effector plasmid pSUPER-miRNA was constructed by cloning the five miRNA hairpins ( nucleotides 6170 to 6759 of the BLV reference genome NC_001414 . 1 ) into BglII and HindIII sites of expression vector pSUPER ( Oligoengine ) . pSUPER-B4 only contains the miRNA-B4 hairpin ( nucleotides 6484 to 6664 ) . Reporter ( 100ng ) and effector ( 500ng ) constructs were transfected with Lipofectamine 2000 into 200 , 000 HEK 293T cells ( 293T/17; ATCC CRL-11268 ) in a 24 well plate . After 24 hours , cells were lysed and luciferase activities were measured using the Dual-Glo Luciferase Assay System ( Promega ) according to the manufacturer’s instructions . Assays were carried out in independent biological triplicates . See S1 Text for primers used for cloning . DNA was extracted from PBMCs using DNeasy Blood and Tissue kit ( Qiagen ) . BLV sequences were PCR amplified using pol gene sequence-specific primers 5'-GAAACTCCAGAGCAATGGCATAA-3' and 5'-GGTTCGGCCATCGAGACA-3' . As reference for quantification , β-actin was amplified with oligonucleotides 5'-TCCCTGGAGAAGAGCTACGA-3’ and 5’-GGCAGACTTAGCCTCCAGTG-3' . DNA was amplified by real-time quantitative PCR in a Roche light cycler using MESA green master mix ( Eurogentec ) . The thermal protocol was initiated by a 5 min denaturation step at 95°C , followed by 45 cycles ( 15 sec at 95°C , 20 sec at 60°C , 40 sec at 72°C ) and terminated by a melting curve . PCR efficiencies were calculated for each sample using 100ng , 33ng and 11ng of DNA . Standard curves were generated using PCR4topo vectors ( Life Technologies ) containing the corresponding pol or actin amplicon . Proviral load was calculated , as an average of the three dilutions , from the number of proviral copies divided by half of the number of actin copies and expressed as number of proviral copies per 100 of PBMCs . One hundred nanograms of genomic DNA were used for the first round of PCR using primers ( Fw 5’-GCTTGACCTCTCGCCTTTTA-3’; Rv GTGCCGAGGTGGAAATAGAA ) , Phusion hot start II High Fidelity DNA polymerase and High Fidelity buffer ( NEB ) . Thermal cycling conditions were: 30 sec at 98°C; 35 cycles ( 98°C 5 sec , 63°C 10 sec , 72°C 30 sec ) ; 2 min at 72°C . One microliter of a 10-fold dilution of first round PCR was used as template for the second round PCR using primers ( nFw 5’-CCCCTAAACCCGATTCTGAT-3’; nRv 5’-GGGCTTGTTACATGGGAAGA-3’ ) . Thermal protocol for nested PCR were: 30 sec at 98°C , 35 cycles ( 5 sec at 98°C , 10 sec at 62°C , 30 sec at 72°C ) , 2 min at 72°C . Actin gene DNA was amplified with primers ( Fw 5’-TCCCTGGAGAAGAGCTACGA-3’ and Rv 5’-GGCAGACTTAGCCTCCAGTG-3’ ) from 100 ng of genomic DNA according to the cycling protocol: 30 sec at 98°C followed by 35 cycles of ( 5 sec at 98°C /10 sec at 64°C / 30 sec at 72°C ) and terminated by 2 min at 72°C . PBMCs were cultivated during 24 hours in RPMI medium supplemented with L-glutamine , antibiotics , 10% fetal calf serum ( FCS ) , 0 . 2μM of phorbol 12-myristate 13-acetate ( PMA ) , 0 . 5μM of ionomycin , 0 . 1μM of concanamycin A ( Sigma Aldrich ) . HEK293T were transfected with plasmids pBLV-WT or pBLVΔ-miRNA using lipofectamine 2000 ( Life Technologies ) and cultivated during 48 hours in RPMI medium supplemented with L-glutamine , antibiotics and 10% FCS . Total RNA was extracted from ex vivo cultivated PBMCs and transfected HEK293T using the miRNeasy kit ( Qiagen ) . After digestion with Turbo DNAse ( Life Technologies ) , cDNA was synthesized with random hexamers using the Reverse Transcriptase Superscript III ( Life Technologies ) . cDNA was amplified by PCR with primers for HPRT ( 5’-GGTCAAGAAGCATAAACCAAAG-3’ and 5’-AAGGGCATATCCCACAACAAAC-3’ ) , Tax/Rex ( 5’-GCGTTTGCTGAAAGCCTTCAA-3’ and 5’-GGGCAGGCATGTAGAGAGTG-3’ ) , Gag ( 5’-TCCCTTTCTCATCACGTTCC-3’ and 5’-GTGGGGGTGAATGGTGTAAC-3’ ) and Env ( 5’-CTATCCGGCAGCGGTCAG-3’ and 5’-GAGGAGAGTAAGAGTGAGACTTACCC-3’ ) in a Roche light cycler using MESA green master mix ( Eurogentec ) according to the protocol: 5min denaturation at 95°C followed by 45 cycles ( 15 sec 95°C , 20 sec 60°C , 40 sec 72°C ) and terminated with a melting curve . PCR efficiency was calculated for each sample using three serial dilutions of input DNA ( 100ng , 33ng and 11ng ) . Relative BLV/HPRT expression was calculated using the delta-delta Ct method . Total RNA was extracted from uncultured bovine PBMCs using the miRNeasy kit ( Qiagen ) and reversed transcribed into cDNA with the Reverse Transcriptase Superscript III ( Life Technologies ) . Three cDNA dilutions ( 1x , 3x and 9x ) were amplified by real-time quantitative PCR in a Roche light cycler using MESA green master mix ( Eurogentec ) according to the protocol: 5 min denaturation at 95°C followed by 45 cycles of PCR ( 15 sec at 95°C , 20 sec at 60°C , 40 sec at 72°C ) and terminated with a melting curve . Primers used for PCR amplification are provided in the Supporting Information file S1 Text . After calculation of PCR efficiencies , mRNA expression relative to HPRT was using the delta-delta Ct method . cDNA was amplified for HPRT house-keeping gene transcript and for ANXA1 , FOS , GZMA , MAP2K1 , PIK3CG and PPT1 . Total RNA of HEK293T , YR2 and PBMCs was extracted using the miRNeasy kit ( Qiagen ) , incubated with Turbo DNAse kit ( Life Technologies ) and reverse transcribed using the Taqman microRNA Reverse transcription kit ( Life Technologies ) and stem-loop custom BLV microRNA specific primers ( Life Technologies ) . To generate HPRT cDNA , 1 . 5 pmol of primer 5’-AAGGGCATATCCCACAACAAAC-3’ was added at the end of the reverse transcription step and incubated for 10 min at 50°C . cDNA was PCR amplified in a Roche light cycler with HPRT primers ( 5’-GGTCAAGAAGCATAAACCAAAG-3’ and 5’-AAGGGCATATCCCACAACAAAC-3’ ) and custom Taqman BLV microRNA probes ( Life Technologies ) using MESA green master mix ( Eurogentec ) or Taqman Universal Master Mix II ( Life Technologies ) , respectively . The thermal protocol used for microRNA quantification started with a denaturation step at 95°C for 10min followed by 45 PCR cycles ( 15 sec at 95°C , 60 sec at 60°C ) . For HPRT , PCR cycles included 5 min denaturation at 95°C followed by 45 cycles of PCR ( 15 sec at 95°C , 20 sec at 60°C , 40 sec at 72°C ) and terminated with a melting curve . PCR efficiencies were calculated for each sample using three cDNA dilutions . Expression of BLV microRNA to relative to HPRT was calculated using the delta-delta Ct method . Frozen plasmas were incubated for 10 min at 37°C , centrifuged during 1 hour at 20 , 000g and filtrated on 0 . 1μm pores . A spike-in control ( C . elegans miR-39 ) was added to the plasma for absolute quantification according to the miRNeasy kit’s user manual . RNA was extracted using the miRNeasy Serum/Plasma kit ( Qiagen ) and reverse transcribed using the Taqman microRNA Reverse transcription kit ( Life Technologies ) with stem-loop custom microRNA primers ( bovine let-7a and miR-92a , BLV miRNAs and C . elegans miR-39 , Life Technologies ) . cDNA was synthesized using custom Taqman probe for each miR ( Life Technologies ) . Three cDNA dilutions ( 1x , 3x and 9x ) were amplified in a Roche light cycler using Taqman Universal Master Mix II ( Life Technologies ) according to the protocol: 10 min at 95°C followed by 45 PCR cycles ( 15 sec at 95°C , 60 sec at 60°C ) . Expression of BLV miRNAs relative to the miR-39 spike-in control was calculated by the using delta-delta Ct method . Animal experimentation was conducted in accordance with the most recent national and international guidelines for animal care and use and following the directive 2010/63/UE of the European parliament . Handling of animals and experimental procedures were reviewed and approved by INTA´s Institutional Committee for Care and Use of Experimental Animals ( CICUAE-INTA ) under protocol number 35/2010 and by PIWet Committee for Care and Use of Experimental Animals under the protocol number 1515 . Statistical tests were performed using GraphPad Prism and Microsoft Excel . The symbols , , * and NS were used when p<0 . 001 , p<0 . 01 , p<0 . 05 and Not Significant , respectively .	The recent discovery of miRNA expression by retroviruses is a matter of active debate and some controversy . Several retroviruses including human immunodeficiency virus ( HIV ) potentially encode functional small non-coding RNAs that are nevertheless undetectable in vivo . While these miRNAs regulate multiple host and viral processes in vitro , their contribution to infection and pathogenesis remains largely unknown . Using next-generation sequencing and reverse genetics , we provide mechanistic and functional evidence demonstrating that BLV miRNAs are functional genetic elements that regulate replication and contribute to oncogenesis in suitable animal models .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "blood", "cells", "livestock", "medicine", "and", "health", "sciences", "luciferase", "immune", "cells", "ruminants", "gene", "regulation", "enzymes", "immunology", "rna", "extraction", "enzymology", "vertebrates", "microbiology", "animals", "mammals", "organisms", "micrornas", "forms", "of", "dna", "molecular", "biology", "techniques", "dna", "extraction", "techniques", "research", "and", "analysis", "methods", "white", "blood", "cells", "sheep", "artificial", "gene", "amplification", "and", "extension", "animal", "cells", "proteins", "gene", "expression", "oxidoreductases", "viral", "replication", "complementary", "dna", "molecular", "biology", "agriculture", "biochemistry", "rna", "antibody-producing", "cells", "cell", "biology", "nucleic", "acids", "b", "cells", "virology", "genetics", "biology", "and", "life", "sciences", "cellular", "types", "non-coding", "rna", "amniotes", "polymerase", "chain", "reaction" ]	2016	Bovine Leukemia Virus Small Noncoding RNAs Are Functional Elements That Regulate Replication and Contribute to Oncogenesis In Vivo
Correct developmental timing is essential for plant fitness and reproductive success . Two important transitions in shoot development—the juvenile-to-adult vegetative transition and the vegetative-to-reproductive transition—are mediated by a group of genes targeted by miR156 , SQUAMOSA PROMOTER BINDING PROTEIN ( SBP ) genes . To determine the developmental functions of these genes in Arabidopsis thaliana , we characterized their expression patterns , and their gain-of-function and loss-of-function phenotypes . Our results reveal that SBP-LIKE ( SPL ) genes in Arabidopsis can be divided into three functionally distinct groups: 1 ) SPL2 , SPL9 , SPL10 , SPL11 , SPL13 and SPL15 contribute to both the juvenile-to-adult vegetative transition and the vegetative-to-reproductive transition , with SPL9 , SP13 and SPL15 being more important for these processes than SPL2 , SPL10 and SPL11; 2 ) SPL3 , SPL4 and SPL5 do not play a major role in vegetative phase change or floral induction , but promote the floral meristem identity transition; 3 ) SPL6 does not have a major function in shoot morphogenesis , but may be important for certain physiological processes . We also found that miR156-regulated SPL genes repress adventitious root development , providing an explanation for the observation that the capacity for adventitious root production declines as the shoot ages . miR156 is expressed at very high levels in young seedlings , and declines in abundance as the shoot develops . It completely blocks the expression of its SPL targets in the first two leaves of the rosette , and represses these genes to different degrees at later stages of development , primarily by promoting their translational repression . These results provide a framework for future studies of this multifunctional family of transcription factors , and offer new insights into the role of miR156 in Arabidopsis development . Shoot development in plants can be divided into several more-or-less discrete phases based on the character of the lateral organs produced during each phase [1 , 2] . These phases consist of a juvenile vegetative , adult vegetative , and a reproductive phase , along with transition periods during which the shoot produces organs of intermediate identity . miR156 , and the closely related miRNA , miR157 , are the master regulators of the transition from the juvenile to the adult phase of vegetative development ( vegetative phase change ) [3 , 4] . miR156/miR157 are expressed at high levels in organs produced early in shoot development , where they repress the expression of their targets , SQUAMOSA PROMOTER BINDING PROTEIN ( SBP ) transcription factors [3 , 5–9] . Vegetative phase change is initiated by a decline in the expression of miR156/157 and the consequent increase in the expression of SBP genes in newly formed organs [7] . These SBP genes promote the development of adult vegetative traits and also promote floral induction in some species [10] . Although this model of shoot development is supported by studies in a number of herbaceous and woody plants , a detailed understanding of the function of individual SBP genes in vegetative and reproductive phase change is still lacking . Arabidopsis has 16 SBP-LIKE ( SPL ) genes , 10 of which are targeted by miR156 [3 , 5 , 11–14] . These genes can be grouped into 5 clades based on the amino acid sequence of their conserved DNA binding domain—SPL3/SPL4/SPL5 , SPL9/SPL15 , SPL2/SPL10/SPL11 , SPL6 and SPL13A/B [12 , 13 , 15] . The phenotype of transgenic plants constitutively over-expressing miR156 reveals that , as a group , these genes control many aspects of Arabidopsis development and physiology , including the timing of vegetative phase change and floral induction , the rate of leaf initiation , shoot branching , anthocyanin and trichome production on the inflorescence stem , stress responses , carotenoid biosynthesis , and shoot regeneration in tissue culture and lateral root development [3 , 6 , 7 , 16–28] . However , the role of individual SPL genes in these processes is still poorly understood . The function of individual SPL genes has been investigated primarily by characterizing the phenotypes of plants expressing miR156-resistant versions of these genes under the regulation of their own promoter , or the constitutively expressed Cauliflower Mosaic Virus 35S promoter . These over-expression phenotypes suggest that SPL2 , SPL9 , SPL10 , SPL11 , SPL13 , and SPL15 control a variety of processes in root and shoot development [7 , 17 , 19 , 21 , 23 , 24 , 29–31] whereas SPL3 , SPL4 and SPL5 primarily promote floral induction and/or floral meristem identity [3 , 16 , 18 , 32] . However , in most cases , it is unknown if this over-expression phenotype reflects the normal function of these genes because their loss-of-function phenotypes have not been characterized , or are not readily apparent . The best-characterized members of this family are SPL9 and SPL15: over-expression of these genes accelerates vegetative phase change and delays the rate of leaf initiation , whereas loss-of-function mutations have the opposite phenotype [7 , 17 , 30 , 33] . Consistent with their sequence similarity , the spl9 spl15 double mutant has a stronger phenotype than either single mutant , although this phenotype is relatively mild compared to the phenotype of plants over-expressing miR156 [33] . This indicates that other targets of miR156 also have important functions in vegetative phase change . However , it is unknown of these functions are shared more-or-less equally by all members of this gene family or are the property of one or a few genes . Over-expression of SPL3 accelerates abaxial trichome production and produces early flowering , but a loss-of-function mutation in this gene has no obvious phenotype [3 , 32] . Similarly , plants over-expressing SPL10 have a reduced rate of leaf initiation and undergo vegetative phase change precociously , but an spl10 mutation has no vegetative phenotype [7 , 17] . The loss-of-function phenotypes of other members of this gene family remain to be determined . Here we describe the temporal and spatial expression patterns of the transcripts of miR156-regulated SPL genes , the expression patterns of miR156-sensitive and miR156-resistant translational reporters for these genes , and the phenotypes loss-of-function mutations in these genes , individually , and in combination . These results provide a detailed picture of the function of miR156-regulated SPL genes in Arabidopsis and the role of miR156 in their regulation . In addition to defining the developmental functions of miR156 , we show that translational repression is more important for the function of miR156 than previously thought . The level of miR156 decreases dramatically in the shoot apex of Arabidopsis seedlings early in development [34] . To determine how this decrease affects the abundance of the transcripts of individual SPL genes , we used qRT-PCR to measure the level of miR156 and its direct targets in shoot apices over a 5-week period . Plants were grown in a 10hr light:14hr dark short day ( SD ) photoperiod to delay flowering , which occurred 5 weeks after planting based on the increase in the expression of the floral marker , AP1 ( Fig 1A ) . The level of miR156 decreased by about 90% from 1 week to 3 weeks , and declined very little after this time ( Fig 1A ) . SPL3 , SPL9 and SPL13 mRNA increased 2-fold from 1 to 3 weeks , whereas SPL2 , SPL4 , SPL6 , SPL10 , SPL11 and SPL15 transcripts increased less than 2-fold or remained constant during this period ( Fig 1A ) . The transcripts of all of these genes increased significantly between 4 and 5 weeks , coincident with the increase in AP1 expression ( Fig 1A ) . We then used in situ hybridization to examine the spatial distribution and the relative abundance of these SPL transcripts in the shoot apices of 3-week-old plants grown in SD , when the level of miR156 was near its minimum . SPL3 , SPL9 , SPL13 and SPL15 transcripts were uniformly expressed in the shoot apical meristem and in leaf primordia , but the transcripts of SPL2 , SPL4 , SPL5 , SPL6 , SPL10 and SPL11 could not be detected using the same approach ( Fig 1B ) . To obtain a comprehensive picture of the expression pattern of these SPL genes and the contribution of miR156 to this pattern , we produced transgenic plants expressing miR156-sensitive ( sSPL ) and miR156-resistant ( rSPL ) fusion proteins tagged with ß-glucoronidase ( GUS ) . With the exception of SPL9 and SPL11 , GUS was inserted within a genomic fragment that extended from the gene upstream to the gene downstream of the SPL locus ( S1 Fig ) . The SPL10 and SPL11 constructs do not include the 3'UTR and the sequence 3' of these genes . miR156-resistant constructs were generated by introducing mutations at the miR156 target site that did not alter the amino acid sequence ( S1 Table ) . Only phenotypically wild-type lines were saved in the case of plants transformed with sSPL constructs , to ensure that the transgene was not over-expressed . For each construct , 6–12 lines homozygous for single insertion sites were identified , and these were then screened for GUS activity under LD conditions . Two lines that expressed GUS at an intermediate level relative to the range for each construct , and in the most frequent pattern , were saved for further analysis . Unless otherwise specified , sSPL and rSPL refer to these translational fusions . The expression of sSPL and rSPL reporters was examined at 1 , 2 , and 3 weeks in plants growing in SD , and at 3 weeks in plants growing in LD ( Fig 2 ) . In SD , rSPL3 and rSPL9 were strongly expressed throughout leaf development in all rosette leaves , rSPL2 , rSPL6 , rSPL10 , rSPL11 , rSPL13 and rSPL15 were expressed in leaf primordia but not in fully expanded leaves , and rSPL4 and rSPL5 were undetectable . The miR156-sensitive versions of these constructs had a much more restricted expression pattern . With the exception of sSPL3 , sSPL9 , and sSPL13 , all of these reporters were undetectable in rosette leaves ( Fig 2 ) . sSPL3 , sSPL9 , and sSPL13 were not expressed in leaves 1 and 2 , but were expressed in all subsequent rosette leaves , although at a much lower level , and for a shorter time in leaf development than the rSPL reporters; this latter observation suggests that the abundance of miR156 increases as leaves expand , as has been reported in rice [35] . These results are consistent with the expression patterns of the transcripts these proteins ( Fig 1 ) , and indicate that most miR156-regulated SPL genes are transcribed throughout the vegetative phase of development in similar patterns , but are strongly and constitutively repressed during this phase by miR156 . miR156 completely represses the expression of all of these genes in leaves 1 and 2 , and represses their expression to varying degrees later in shoot development . Three weeks after planting , the sSPL lines growing in SD were still vegetative , whereas the sSPL lines growing in LD were in the early stages of inflorescence development . At this stage , LD and SD plants had the same relatively low level of GUS expression in rosette leaves , but both the sSPL and the rSPL plants had GUS activity in the developing inflorescence ( Fig 2 ) . Some sSPL reporters were expressed at a lower level in than the corresponding rSPL reporter ( e . g . , sSPL2 , sSPL10 ) , but most sSPL and rSPL reporters were expressed at essentially the same level the inflorescence primordium . With the exception of sSPL15 and rSPL15 , all reporters were expressed throughout inflorescence development and , in most cases , the expression patterns of the corresponding rSPL and sSPL reporters were nearly identical ( S2 Fig ) . In contrast , sSPL15 and rSPL15 were only expressed during early stages of inflorescence development . These observations suggest that miR156 plays a minor role in the regulation of SPL activity during and after the floral transition , and that SPL15 is important for floral induction and/or the floral meristem identity transition , but not for later stages of inflorescence development . Six of the rSPL lines ( rSPL2 , rSPL9 , rSPL10 , rSPL11 , rSPL13 and rSPL15 ) had a phenotype that resembled the phenotype of plants with reduced levels of miR156 , demonstrating that these GUS-fusion proteins are functional ( Fig 3A and 3B ) . All of these lines had a reduced rate of leaf initiation , and the angle between the leaf blade and the petiole in leaves 1 and 2 was less acute than in the corresponding sSPL line ( Fig 3A and 3B ) . In LD , these lines had fewer juvenile leaves ( leaves without abaxial trichomes ) than Col ( Fig 3C ) . In contrast , rSPL3 , rSPL4 , rSPL5 , and rSPL6 had little or no effect on the rate of leaf initiation and were not obviously different from Col ( Fig 3A and 3C ) . SPL3 , SPL4 and SPL5 are relatively small proteins , and we were concerned that the GUS fusion might disrupt their activity . To address this issue , we produced lines expressing rSPL3 without a GUS fusion . In comparison to a previously characterized 35S::rSPL3 line [3] , which expresses SPL3 at approximately 1 , 700 times the normal level , these three lines had between 15–40 times the normal level of the SPL3 transcript ( S3 Fig ) . The 35S::rSPL3 line is early flowering and has slightly accelerated abaxial trichome production [3] , but lines expressing rSPL3 under the control of its endogenous regulatory sequences were not significantly different from Col with respect to abaxial trichome production , leaf number , or flowering time ( Table 1 , Experiment 1 ) . These results suggest that SPL2 , SPL2 , SPL9 , SPL10 , SPL11 , SPL13 and SPL15 promote vegetative phase change whereas SPL3 , SPL4 , SPL5 , and SPL6 do not contribute significantly to this process . These results also suggest that SPL3 does not normally promote in floral induction in either LD or SD . The phenotype of plants expressing miR156-resistant SPL genes reveals the processes that these genes are capable of regulating , but does not necessarily reveal the processes in which they are actually involved because these transgenes may not be transcribed in a completely normal pattern or at a completely normal level due to their position in the genome or the presence of multiple T-DNAs at each insertion site [36 , 37] . To determine the normal functions of miR156-regulated SPL genes we therefore characterized the phenotypes of loss-of-function mutations in these genes ( Fig 4 ) . The mutations used for this analysis were generated by several different methods in several different ecotypes , and were introgressed into Col so that their phenotype could be compared . T-DNA ( spl2-1 , spl5-1 , spl9-2 , spl9-4 , spl15-1 , spl15-2 ) and CRISPR-Cas9 ( spl10-1 , spl10-2 , spl10-3 ) alleles generated in Col were crossed to Col 3 times before use . T-DNA alleles generated in Ws ( spl3-1 , spl11-1 ) , and EMS-induced alleles generated in Col ( spl4-1 , spl4-2 ) or Ler ( spl13-1 , spl13-2 , spl13-3 ) were crossed to Col 6 to 8 times before use . All of the T-DNA mutations significantly reduced mRNA production ( S4 Fig ) and are likely to be null alleles . Different alleles of spl9 , spl13 , and spl15 had similar effects on abaxial trichome production and produced similar phenotypes in combination with each other ( S3 Table , Experiments 1 , 2 , and 3 ) , so we only used one allele of these genes for the generation of multiple mutant lines . Although we only identified one allele of SPL2 , SPL3 , SPL5 and SPL11 that significantly affected the expression of these genes , all of these alleles were RNA null , had no obvious phenotype ( spl2 , spl3 , spl5 ) or a very weak phase change phenotype ( spl11 ) on their own , and had the expected phenotype in combination with other spl mutations . Consequently , we believe that this phenotype accurately reflects the function of these genes . We were unable to identify T-DNA mutations that reduce the expression of SPL4 so we used TILLing to identify mutations in this gene . Three missense mutations were selected for further study ( Fig 4 ) . All of these mutations affected highly conserved amino acids , but none had an obvious mutant phenotype . To determine if these mutations affect the function of SPL4 , we took advantage of the observation that plants over-expressing an SPL4 transcript without a miR156 target site ( 35S::SPL4Δ ) are early flowering [3] . 35S::SPL4Δ constructs containing each mutation ( 35S::SPL4Δm ) were transformed into Col . A comparison of the flowering time of T1 plants transformed with 35S::SPL4Δ and these 35S::SPL4Δm constructs revealed that the G541-to-A mutation produced the largest reduction in SPL4 activity ( S2 Table ) , so we used this allele ( spl4-1 ) for all of the analyses described here . Identifying mutations that block the activity of SPL13 was problematic because there are two copies of this gene in Col , SPL13A ( AT5G50570 ) and SPL13B ( AT5G50670 ) . These genes reside within a 33kb tandem duplication that is so recent that there are no polymorphisms between the duplicated segments . To determine if this duplication exists in other accessions of Arabidopsis , we used qPCR to compare the amounts of SPL13 and SPL9 DNA in Col and several other ecotypes . SPL13 and SPL9 were present in equal amounts in Ler , Bak-2 and Voeran , but Col had twice as much SPL13 as SPL9 DNA ( S5 Fig ) . This result suggests that the duplication in Columbia is relatively recent , and also suggests that the basal number of miR156-regulated SPL genes in Arabidopsis is 10 , not 11 , as is commonly reported . Importantly , the evidence that Ler has a single SPL13 gene meant that we could use mutations generated in this accession to obtain spl13 loss-of-function alleles . 40 EMS-induced mutations of SPL13 have been identified in Ler ( Martin et al , 2009 ) . Two of these ( spl13-1 , spl13-3 ) introduce stop codons near the 5' end of the gene , and a second ( spl13-2 ) is a mutation in the splice donor site in the third intron ( Fig 4 ) . We used spl13-1 for all of our analyses because it was the most highly introgressed mutation at the start of these experiments . A line containing T-DNA insertion in the first exon of SPL10 has been identified [38] , but this insertion does not reduce the SPL10 transcript , and does not have an obvious phenotype , making it difficult to know if it has an effect on SPL10 activity . Furthermore , it was impractical to recombine this mutation with mutations in its close paralog , SPL11 , because SPL10 and SPL11 are only 1 . 6 kb apart . Consequently , mutations in SPL10 were created using CRISPR-Cas9 . spl10-2 and spl10-4 were generated in Col , and spl10-3 was produced in a Col line containing spl11-1 in order to generate an spl10 spl11 double mutant . Sequencing of several potential off-target genes revealed no mutations in these genes . Fortuitously , spl10-2 and spl10-3 have an identical 35 bp deletion that produces a premature stop codon upstream of the SBP box . spl10-4 has a 1 bp-deletion followed by 30 bp insertion , which also results in a premature stop codon upstream of the SBP-box domain ( Fig 4 ) . We did not perform a detailed analysis of loss-of-function mutations of SPL6 because we were unable to identify T-DNA insertions or EMS-induced mutations that unambiguously reduced the activity of this gene . TILLing produced 3 missense mutations in highly conserved amino acids that were predicted to have a damaging effect on protein function ( Fig 4 ) , but these mutations did not have an obvious phenotype . Over-expression of an rSPL6 construct in transgenic plants under the regulation of the 35S promoter slightly accelerated vegetative phase change , but this effect was so weak that we did not believe that we would be able to assess the effect of these missense mutations on SPL function using this approach . The lack of a strong gain-of-function phenotype suggests that SPL6 may not play a critical role in regulating vegetative morphogenesis . Lines containing combinations of the mutations described above were generated by intercrossing mutant lines and identifying plants homozygous for the relevant alleles in F2 families , using allele-specific PCR ( S4 Table ) . As we were generating these stocks we discovered that spl2-1 is semi-sterile in heterozygous but not homozygous condition , and could not be recombined with spl4-1 . This behaviour is characteristic of reciprocal translocations , and suggests that spl2-1 contains a translocation with breakpoints near SPL2 on chromosome 5 and SPL4 on chromosome 1 . Phylogenetic analysis demonstrates that miR156-targetted SPL genes in Arabidopsis fall into 5 clades: SPL3/SPL4/SPL5 , SPL2/SPL10/SPL11 , SPL9/SPL15 , SPL6 , and SPL13 [26 , 33 , 39] ( Fig 4 ) . We focused on SPL2 , SPL9 , SPL10 , SPL11 , SPL13 and SPL15 because the phenotype of plants expressing miR156-resistant versions of these genes indicated that they have a significant role in vegetative development . Because miR156-resistant transgenes and lack-of-function mutations in SPL3 , SPL4 , and SPL5 did not have obvious phenotypes , we only conducted detailed analyses of the spl3-1 spl4-1 spl5-1 ( spl3/4/5 ) triple mutant . This triple mutant displayed no significant delay in vegetative phase change in LD and under two different SD conditions—10 hrs light:14 hrs dark and 8 hrs light:16 hrs dark ( Table 1 , Experiment 2 ) . We performed all subsequent SD experiments using a 10 hrs light:14 hrs dark photoperiod because this photoperiod significantly delays flowering but does not produce a major delay in vegetative phase change . Individually , spl9-4 , spl11-1 and spl13-1 ( hereafter , spl9 , spl11 , spl13 ) produced a small increase the number of juvenile leaves in both LD and SD , whereas spl2-1 , spl10-2 , and spl15-1 ( hereafter , spl2 , spl10 , spl15 ) had no obvious effect on juvenile leaf number ( Table 1 , Experiment 3 ) . Plants mutant for more than one of these genes had much stronger phenotypes however . The strongest interaction we observed was between spl9 and spl13 . In LD , spl9/13 had 6–8 more juvenile leaves and 3–5 more rosette leaves than Col , and in SD it had 15–16 more juvenile leaves and 6–7 more rosette leaves than Col ( Table 1 , Experiment 3 , Experiment 4 ) . The addition of spl15 ( spl9/13/15 ) produced a further delay in vegetative phase change and a larger increase in rosette leaf number ( Table 1 , Experiment 3 , Experiment 4 ) . In LD , spl9/13/15 produced 11–14 more juvenile leaves and 9–16 more rosette leaves than Col , and in SD it produced 16–24 more juvenile leaves and 6–17 more rosette leaves than Col . These genotypes either had no effect on flowering time , or produced a very small delay in flowering , indicating that their effect on rosette leaf number is largely attributable to an increase in the rate of leaf initiation . These results suggest that SPL9 , SPL13 and SPL15 strongly promote the juvenile phase and delay leaf initiation . Mutations in spl2 , spl10 and spl11 interacted weakly with each other and with spl9 , spl13 and spl15 ( S3 Table , Experiment 4 ) . For example , the spl2/9/11/15 quadruple mutant did not produce significantly more juvenile leaves or rosette leaves than spl9/15 in either LD or SD ( Table 1 , Experiment 3 ) . spl9/11/13/15 was not significantly different from spl9/13/15 in LD , and had only a slightly stronger phenotype than spl9/13/15 in SD ( Table 1 , Experiment 3 ) . spl2 interacted more strongly with spl9/13/15: the spl2/9/13/15 quadruple mutant produced significantly more juvenile leaves and rosette leaves than spl9/13/15 in both LD and SD , although its phenotype was still much less severe than 35S::MIR156A ( Table 1 , Experiments 3 and 4 ) . Adding spl11 to the spl2/9/13/15 quadruple mutant ( i . e . spl2/9/11/13/15 ) produced a further increase in juvenile leaf number and rosette leaf number , and adding both spl10 and spl11 ( spl2/9/10/11/13/15 ) produced a vegetative phenotype that was more severe than that of 35S::MIR156A ( Table 1 , Experiment 4 ) . The spl2/3/5/9/11/13/15 sextuple mutant was not significantly different from spl2/9/11/13/15 ( Table 1 , Experiment 4 ) . These results provide additional evidence that SPL9 , SPL13 and SPL15 play dominant roles in vegetative phase change , but reveal that SPL2 , SPL10 and SPL11 also contribute to this developmental transition . Together , these 6 genes account for the effect of miR156 on vegetative phase change . Leaf number cannot be used to measure flowering time in spl mutants because most of these mutations accelerate the rate of leaf initiation ( Fig 5A ) [17] . The major exception is spl3/4/5 , which has no effect on the rate leaf initiation ( S6 Fig ) . The effect of spl mutations on flowering time was therefore determined by recording the date of the first open flower . Many genotypes displayed small and sometimes statistically significant differences in flowering time relative to Col ( Table 1 , Experiments 2 , 3 and 4 ) , but is difficult to know if these differences are meaningful because we have observed similar variation between different stocks of Col; furthermore , some single and multiple mutant lines flowered earlier than Col , which is the opposite of the expected effect and is inconsistent with the phenotype of higher order mutant combinations and 35S::MIR156A . However , certain combinations of mutations produced consistent effects on flowering time , which we believe accurately reflect the role of these genes in floral induction . The phenotype of plants expressing 35S::MIR156A reveals the overall contribution of miR156-regulated SPL genes to flowering time . This transgene consistently produced a 7–8 day delay in flowering in LD , but produced a more variable delay in flowering in SD , ranging from 5 to 12 days in different experiments ( Table 1 , Experiments 3 and 4 ) . None of the single mutants flowered later than Col in either LD or SD . In LD , the most significant interactions occurred between spl15 and other genotypes . Under these conditions , spl9/15 flowered 2 days later than Col , and spl9/13/15 flowered 4–5 days later than Col ( Table 1 , Experiments 3 and 4 ) , In contrast , spl9/13 only flowered 1 day later than Col , and spl2/9/13 was not significantly different from spl9/13 ( Table 1 , Experiments 3 and 4; S3 Table , Experiment 4 ) . The addition of spl2 , spl10 and spl11 to spl9/13/15 ( spl2/9/10/11/13/15 ) produced a delay in flowering time equal to that of 35S::MIR156A ( Table 1 , Experiment 4 ) . We obtain different results in SD . In SD , most genotypes flowered earlier , or at approximately the same time as Col . spl9/13/15 flowered 2 days later that Col in one experiment ( Table 1 , Experiment 3 ) , but it did not flower significantly later than Col in a second experiment ( Table 1 , Experiment 4 ) . Similarly , spl2/9/11/13/15 flowered significantly later than Col and spl9/13/15 in one experiment ( Table 1 , Experiment 3 ) , but did not flower significantly later than these genotypes in a second experiment ( Table 1 , Experiment 4 ) . Flowering was only consistently delayed in SD in the spl2/9/10/11/13/15 hextuple mutant , which flowered later than 35S::MIR156A in both LD and SD . Thus , SPL2 , SPL9 , SPL10 , SPL11 , SPL13 , and SPL15 all promote floral induction , and together explain the effect of 35S::MIR156A on this process . These results also suggest that SPL15 plays a more important role in floral induction than other SPL genes in LD . spl3/4/5 did not flower significantly later than Col in either LD or SD ( Table 1 , Experiment 2 ) . Furthermore , adding spl3 and spl5 to spl2/9/11/13/15 ( spl2/3/5/9/11/13/15 ) did not produce an additional delay in flowering time beyond that observed for spl2/9/11/13/15 ( Table 1 , Experiment 4 ) . This result is consistent with the phenotype of the rSPL3 transgenic lines ( Table 1 , Experiment 1 ) , and demonstrates that these genes are not required for floral induction . After floral induction , Arabidopsis makes several cauline leaves and co-florescence branches before transitioning to the floral phase of inflorescence development . This so-called "floral meristem identity" transition is promoted by LFY and AP1 [40–46] , which are bound by SPL3 and SPL9 in vivo and are up-regulated in plants over-expressing these genes [16 , 18] . The effect of spl mutations on the floral meristem identity transition was measured by counting the number of cauline leaves [45 , 47] . No single or double mutant had a significant effect on cauline leaf number in LD . However , spl3/4/5 consistently produced approximately 1 extra cauline leaf ( Table 1 , Experiment 2 ) and spl2/9/11/13/15 and other genotypes containing these mutations produced two extra cauline leaves in LD ( Table 1 , Experiment 4; Fig 5C ) , which was identical to the effect of 35S::MIR156A . In SD , spl9/13 produced the same number of cauline leaves as Col , but 35S::MIR156A and other multiple mutants only produced 5 cauline leaves—4 less than the number of cauline leaves in Col ( Table 1 , Experiment 4 ) . These results indicate that SPL2 , SPL3 , SPL4 , SPL5 SPL9 , SPL10 , SPL11 , SPL13 , and SPL15 promote the floral meristem identity transition in LD , and that SPL2 , SPL10 , SPL11 , and/or SPL15 inhibit this transition in SD . We explored the molecular basis for these phenotypes by examining the effect of spl2/9/11/13/15 and spl3/4/5 on the expression of genes that regulate flowering time ( SOC1 , MIR172B ) and the floral meristem identity transition ( LFY , AP1 , FUL ) in the shoot apices of 11 day-old shoots grown in LD . spl2/9/11/13/15 reduced the abundance of LFY , AP1 , FUL , and SOC1 transcripts by about 50% , whereas spl3/4/5 reduced the expression of LFY and AP1 , but had little effect on the expression of FUL or SOC1 ( Fig 6 ) . This result is consistent with the developmental phenotypes of these genotypes , and indicates that SPL2 , SPL9 , SPL11 , SPL13 and SPL15 promote both floral induction and floral meristem identity , whereas SPL3 , SPL4 and SPL5 primarily promote floral meristem identity . miR156 is thought to repress floral induction by repressing the expression of miR172 [7] , thus elevating the expression of miR172-regulated AP2-like transcription factors , which repress the expression of floral activators , such as FT and SOC1 [48–50] . miR156 modulates the level of miR172 through its effect on the expression MIR172B [51] . MIR172B is a direct target of SPL9 [7] . To determine if other SPL genes also regulate the expression of MIR172B , we examined the abundance of pri-miR172b and miR172 in the shoot apices of 16-day old spl mutants grown in SD . spl2 and spl9 had normal levels of pri-miR172b , but spl11 , spl13 and spl15 all had slightly reduced amounts of this transcript ( Fig 7A ) . spl2/13 , spl9/13 , spl9/15 , and spl9/13/15 all had between 30–50% of the normal amount of pri-miR172b , and higher order mutants had approximately 20% of the normal pri-miR172b levels ( Fig 7A ) . In contrast to their effect on pri-miR172b , most of these mutant lines only produced a small decrease in the level of miR172 ( Fig 7B ) , possibly because of feedback regulation of other miR172 genes by the AP2-like transcription factors targeted by this miRNA [48] . The biggest decrease in miR172 was observed in spl2/9/11/1315 , which had approximately 60% of the normal level of this transcript . Thus , SPL2 , SPL9 , SPL11 , SPL13 and SPL15 all promote the expression of MIR172B , and all of these genes must be repressed to produce a significant reduction in the level of miR172 . To determine if the effect of these mutations on miR172 is functionally significant , we examined the expression of SPL3 and SPL5 . These SPL genes are downstream of miR172 and are repressed by the miR172-regulated AP2-like transcription factors , TOE1 and TOE2 [7 , 51] . SPL3 and SPL5 transcripts were reduced by approximately 30–40% in spl9/13/15 , spl2/9/13/15 , and spl9/11/13/15 , and by an even greater amount in spl2/9/11/13/15 ( Fig 7C ) . These mutants had similar effects on miR172 and SPL5 transcripts , but SPL3 transcripts were present at a much lower level in spl2/9/11/13/15 than was expected from the effect of this genotype on miR172 . At present , we have no explanation for this effect . In any case , these results suggest that in addition to directly targeting the transcripts of SPL3 , SPL4 and SPL5 , miR156 represses the transcription of these genes by elevating the expression of AP2-like transcription factors . miR156-regulated SPL genes have been reported to repress lateral root development in Arabidopsis [31] . To explore the function of SPL genes in root development , we examined the expression of the rSPL and sSPL reporters in the roots of 12-day-old plants . With the exception of rSPL4 and rSPL5 , all rSPL reporters were expressed in the root ( S7 Fig ) . rSPL2 and rSPL11 were most strongly expressed at the root tip , rSPL3 was strongly expressed outside the root tip , rSPL15 was expressed most strongly in the stele of the primary root , and rSPL6 , rSPL9 , rSPL10 and rSPL13 were expressed in both the root tip and in more mature parts of the root . In contrast , the only miR156-sensitive constructs that were expressed in the root were sSPL6 , sSPL9 and sSPL11 . sSPL6 was expressed in the same pattern as rSPL6 , but at a lower level . sSPL9 was expressed in the stele of the primary root , but not in lateral roots , whereas sSPL11 was expressed exclusively at the tip of the primary and lateral roots . Thus , most SPL genes are transcribed in roots but their expression is strongly repressed by miR156 for at least two weeks after germination . Variation in the capacity for adventitious root production is often used as a marker of shoot maturation in woody plants; in general , cuttings from juvenile nodes root more readily than cuttings from adult nodes [52] . To determine if SPL genes regulate adventitious root development , we removed the root system of wild-type and mutant seedlings in order to induce root production from the base of the hypocotyl ( Fig 8A ) . Plants transformed with 35S:MIR156A , as well as hypocotyls of the spl2/9/11/13/15 mutant , produced the same number of adventitious roots as wild-type plants ( Fig 8B ) . This is not surprising because miR156 is already present at very high levels in young seedlings , and SPL gene expression is strongly repressed during this phase ( Fig 2 ) . In contrast , plants transformed with 35S::MIM156—which causes an increase in SPL expression [53]—produced significantly fewer adventitious roots than wild-type plants ( Fig 8B ) . We conclude that miR156-targetted SPL genes inhibit adventitious root development . In our original analysis of vegetative phase change in Arabidopsis [57] , we found that juvenile leaves can be divided into two classes based on their morphology and their sensitivity to gibberellin . Specifically , we found that the identity of the first two rosette leaves is distinct from , and much more stable than , the identity of subsequent juvenile leaves . The results presented here provide a molecular explanation for this observation . We found that miR156 is expressed at very high levels in leaves 1 and 2 and completely blocks the expression of all SPL genes in these leaves . As the level of miR156 declines in subsequent leaves , the expression of some SPL genes increases significantly while others remained strongly repressed . During this latter phase , factors that promote the expression of SPL genes , such as GA [54 , 56] or floral inductive signals [11 , 58] , are capable of increasing SPL activity above the threshold set by miR156 , enabling phase transitions to occur . During this latter phase , miR156 may act to fine-tune the expression of some of its targets ( e . g . SPL3 , SPL9 , SPL13 and SPL15 ) , and to set a threshold for the expression of other targets ( e . g . SPL2 , SPL10 and SPL11 ) . This latter function would ensure that only factors that strongly promote SPL transcription produce functionally significant changes in SPL activity . This might be important for preventing transient increases in SPL activity from prematurely promoting floral induction , for example . Our results indicate that miR156 does not play a direct role in floral induction because the abundance of miR156 does not change significantly during this process . However , miR156 could regulate this process indirectly , by ensuring that floral induction only occurs under appropriate environmental conditions . miR156 represses SPL gene expression by cleaving SPL transcripts [3 , 59–61] and by promoting their translational repression [8 , 62 , 63] , but the relative importance of these activities is still unknown . It is therefore significant that the steep decline in miR156 levels early in shoot development is not accompanied by a corresponding increase in SPL transcripts . This could either mean that miR156 represses SPL expression primarily by translational repression , or that the amount of miR156 is sufficient to maximally induce the cleavage of most SPL transcripts , even when this miR156 is present at relatively low levels . However , this latter explanation implies that variation in the abundance of miR156 is functionally irrelevant because it does not produce a change in gene expression , and this is not the case; GUS expression from the miR156-sensitive reporters for SPL3 , SPL9 , and SPL13 increased substantially during shoot development in a miR156-dependent fashion . We interpret the relatively small increase in SPL transcript levels as evidence that miR156 regulates the expression of most of its targets primarily through its effect on translation , rather than via its effect on transcript stability . SPL2 , SPL9 , SPL10 , SPL11 , SPL13 , and SPL15 have overlapping functions , and together promote vegetative traits associated with the adult phase . Inappropriate expression of any one of these genes early in shoot development results in the precocious expression of traits that are normally expressed later in development , while the combined loss of these genes prolongs the expression of juvenile traits , and produces a phenotype that is essentially indistinguishable from that of plants constitutively expressing miR156 . However , the phenotypes of plants lacking subsets of these genes demonstrate that SPL9 , SPL13 and SPL15 play a much larger role in vegetative development than SPL2 , SPL10 and SPL11 . This is at least partly explained by the level at which these genes are expressed in the shoot apex . Both the gain-of-function and the loss-of-function phenotypes of these six genes indicate that they have closely related functions , but SPL9 , SPL13 and SPL15 are much more highly expressed in the vegetative shoot than SPL2 , SPL10 and SPL11 . While the miR156-sensitive constructs for these genes are expressed at very different levels during vegetative development , the miR156-resistant constructs for SPL2 , SPL9 , SPL10 , SPL11 and SPL13 are expressed at roughly the same level in the shoot apex . This observation suggests that the differential expression of these genes is due to their differential sensitivity to miR156 . In contrast to these genes , SPL3 , SPL4 and SPL5 do not have dramatic effects on vegetative morphology . This is particularly surprising in the case of SPL3 because it is highly expressed in the rosette . Over-expression and under-expression of miR156 affects the response of Arabidopsis to heat stress [21] and salt stress [22 , 64] , and it may be that SPL3 regulates these physiological processes rather than shoot morphogenesis . Previously , we suggested [57] that the timing of vegetative phase change is regulated independently of leaf initiation because mutations in ALTERED MERISTEM PROGRAMMING 1 ( AMP1 ) and PAUSED ( PSD ) increase ( amp1 ) or decrease ( psd1 ) the number of juvenile leaves without changing the timing of vegetative phase change . Instead , the effect of these mutations on juvenile leaf number appeared to be attributable to an increase ( amp1 ) or a decrease ( psd ) in the rate of leaf initiation . However , the tight linkage between the timing of vegetative phase change and rate of leaf initiation in plants with elevated or reduced levels of SPL gene expression ( this report; [17] ) suggests that this hypothesis needs to be re-evaluated . In particular , the evidence that AMP1 promotes miRNA-mediated translational repression [65] raises the possibility that the effect of amp1 on juvenile leaf number could be attributable to the effect of this mutation on miR156 activity , rather than being an indirect effect of the accelerated rate of leaf initiation in this mutant . On the other hand , the effect of amp1 on vegetative phase change is inconsistent with its proposed role in miRNA-mediated translational repression , at least with respect to miR156 . Mutations that interfere with the activity of miR156 , such as ago1 , sqn and suo , reduce the number of juvenile leaves [62 , 66 , 67] , whereas amp1 has the opposite phenotype [57] . Indeed , the phenotype of amp1 is more consistent with an increase in miR156 activity than with a decrease in miR156 activity . Further studies will be necessary to determine if the effect of psd and amp1 on vegetative phase change is an indirect result of their effect on leaf initiation , or reflects a more direct role in this process . In Col , miR156-regulated genes are less important for floral induction than they are for vegetative phase change . Under LD , 35S::MIR156A and spl2/9/10/11/13/15 plants only flowered 8 days later than normal but produced 22 additional juvenile leaves; under SD , they also flowered 8 days later than normal but produced more than 60 additional juvenile leaves . These genes may be more important for flowering in other ecotypes , however . Col has relatively low levels of the floral repressor , FLC [68] , because it possesses a non-functional allele FRI , which is required for the expression of FLC [69] . In Arabis alpina and Cardamine flexuosa [70 , 71] , FLC acts together with miR156 to repress flowering; plants in which both of these factors are expressed at high level are extremely late flowering . Arabidopsis ecotypes with functional alleles of FRI have relatively high levels of FLC [72 , 73] and it will be important to determine if miR156-regulated SPL genes are more important for floral induction in these ecotypes . The extent to which SPL genes are required for floral induction also appears to be strongly dependent on environmental conditions . Both we and Wang et al [18] found that 35S::MIR156 had a relatively small effect on flowering time SD , whereas Schwab et al [6] reported that 35S::MIR156 flowers at about 7 months in SD . This difference is unlikely to be attributable to variation in the strength of the 35S::MIR156 transgenes used in these experiments because the phenotype of our 35S::MIR156 line was nearly identical to the spl2/9/10/11/13/15 mutant , implying that this 35S::MIR156 transgene completely , or nearly completely , eliminates SPL activity . This variability suggests that the effect of 35S::MIR156 on flowering time under SD is strongly dependent on environmental factors other than photoperiod , such as light quantity and quality , temperature , water availability etc . Arabidopsis is extraordinarily sensitive to minor variation in environmental conditions [74] , and it may be that SPL genes only play a major role in floral induction in Col when all of the environmental factors that positively regulate this process are absent . A summary of the role of miR156-regulated SPL genes in flowering is shown in Fig 9 . Many studies have focused on the role of SPL3 , SPL4 and SPL5 in floral induction because these genes are strongly up-regulated during floral induction and cause early flowering when expressed under the regulation of the constitutive CaMV 35S promoter [3 , 11 , 32 , 51 , 55 , 56 , 58] . Although spl3/4/5 mutants consistently had extra cauline leaves , they displayed little or no delay in flowering time under both LD and SD , and had no effect on the expression of the flowering time genes , MIR172B and SOC1 . This latter observation is consistent with previous studies indicating that SPL3 , SPL4 and SPL5 are downstream of SOC1 , miR172 , and the flowering time regulator , FT [51 , 75 , 76] . The inflorescence phenotype of spl3/4/5 is explained by the effect of this genotype on the floral meristem identity genes LFY , AP1 and FUL . We found that the spl3/4/5 triple mutant has reduced levels of the transcripts of these three genes . This is consistent with previous studies showing that LFY , AP1 and FUL transcripts are elevated in plants over-expressing SPL3 , and with the evidence that SPL3 binds to the promoters of these three genes [16 , 18 , 51 , 56] . As is the case with SPL3 , SPL4 and SPL5 [3 , 32] , over-expression of LFY , AP1 and FUL accelerates flowering , but loss-of-function mutations in these genes are not late flowering [41 , 77–80] . These and many other studies demonstrate that floral induction is distinct from the floral meristem identity transition . Floral induction involves changes in many different aspects of shoot development including the growth and morphogenesis of rosette leaves , stem elongation , and a change the identity of the lateral organs produced by the shoot apical meristem [81 , 82] . The floral meristem identity transition is the latter of these processes [83] . Over-expression of genes involved in the floral meristem identity transition , such as LFY or AP1 , forces the vegetative meristem to become an inflorescence meristem , resulting in early flowering . Similarly , over-expression of SPL3 , SPL4 , and SPL5 accelerates the floral meristem identity transition , but it is apparent from their loss-of-function phenotype that these genes do not play a general role in floral induction . Our results are consistent with the observation that the ortholog of SPL3/4/5 in Antirrhinum majus , SBP1 , acts after floral induction to promote the floral meristem identity transition [84] . Other miR156-regulated SPL genes are required for both floral induction and floral meristem identity ( Fig 9 ) . As in vegetative phase change , SPL9 , SPL13 and SPL15 play dominant roles in both of these processes , but SPL2 , SPL10 and SPL11 also make important contributions , particularly in SD . This is evident from the observation that flowering was only delayed significantly in SD in genotypes that lacked SPL9 , SPL15 , and either SPL2 or SPL11 . SPL2 and SPL11 also contribute to floral induction in LD , but their effect is relatively modest under these conditions . SPL9 is bound to the promoters of the flowering time genes MIR172B and SOC1 in vivo , and promotes their expression when it is over-expressed [7 , 18] . Although loss of SPL9 does not have a major effect on the expression of MIR172B and SOC1 , their expression was significantly reduced in spl2 , 9 , 11 , 13 , 15 mutants . Together , these results suggest that SPL2 , SPL9 , SPL11 , SPL13 and SPL15 directly promote the transcription of these genes . The possibility that SPL2 , SPL9 , SPL11 , SPL13 and SPL15 promote MIR171B and SOC1 transcription by a different mechanism should also be considered . For example , SPL9 blocks the dimerization of the TCP4 and CUC2 transcription factors by binding to TCP4 [85] , and there is evidence that it regulates the response of plants to GA by interfering with the activity of the DELLA protein , RGA [54] . However , if SPL2 , SPL9 , SPL11 , SPL13 and SPL15 act primarily by modulating the activity of other transcription factors , one would expect the dimerization domain in these functionally redundant proteins to be highly conserved , and this is not the case . SPL9 interacts with TCP4 [85] and RGA [54] via its C-terminal region , and this region is highly variable between SPL2 , SPL9 , SPL11 , SPL13 and SPL15 . The most highly conserved region of these proteins is their DNA-binding domain , the SBP-box . For this reason , we suspect that these SPL proteins act primarily as direct transcriptional activators or repressors . Adventitious root production is increased in plants with elevated levels of miR156 , such as the Teopod/Corngrass mutants of maize [86] or tobacco transformed with 35S::miR156 [87] , suggesting that SPL genes normally inhibit this process . We found that elevated levels of miR156 have no effect on adventitious root production in the hypocotyl , but reducing miR156 activity inhibits this process , implying that SPL proteins inhibit adventitious root production , just as they inhibit lateral root production in the primary root [31] . SPL expression increases in successive nodes of woody plants [9] , so this result may provide an explanation for the correlation between shoot age and the loss of rooting capacity in these plants [52 , 88] . Unfortunately , we were unable to investigate whether variation in the capacity for adventitious root production is a marker for vegetative phase change or reproductive phase change because the short internodes of an Arabidopsis rosette make it difficult to examine adventitious root production at different stages of vegetative development . SPL gene expression increases during both vegetative phase change and floral induction , and there may be a threshold level of SPL gene expression required to repress adventitious root development . This is an important question to answer because rooting ability determines the ease with which many horticulturally-important species can be propagated . The ability to SPL expression using exogenous factors could be of considerable practical importance . SPL genes arose early in plant evolution , and are present in multiple copies in all land plants examined to date [15 , 89] . We focused on the roles of miR156-regulated SPL genes in shoot and root morphogenesis , but these genes are involved in many other aspects of plant biology as well . The reporter lines and mutant stocks described here will be useful for defining the full range of their function , and the role of miR156 in sculpting their activity . All of the stocks used in this study were in a Col genetic background . Mutations that were originally generated in a different genetic background ( spl3-1 , spl11-1 , spl13-1 , spl13-2 , spl13-3 ) were crossed to Col 6 or more times . spl2-1 ( SALK_022235 ) , spl3-1 ( FLAG_173C12 ) , spl4-1 ( CS90956 ) , spl4-2 ( CS88228 ) , spl4-3 ( CS96315 ) , spl5-1 ( SAIL_265_D02 ) , spl6-1 ( CS90560 ) , spl6-2 ( CS93521 ) , spl6-3 ( CS92990 ) , spl9-4 ( SAIL_150_B05 ) , spl11-1 ( FLAG_422H07 ) and spl15-1 ( SALK_074426 ) were obtained from the Arabidopsis Biological Resource Center ( Ohio State University , Columbus , OH ) . spl13-1 ( line 2754 ) , spl13-2 ( line 3697 ) and spl13-3 ( line 6746 ) were obtained from the TILLer service ( Carlos Alonso Blanco , Centro Nacional de Biotecnologia , Madrid , Spain ) . Seeds were sown on Farfard #2 potting soil , placed at 4°C for 2 to 3 days , and grown at 22°C in Conviron growth chambers under either long days ( 16 hrs light/8 hrs . dark; 95 μmol m-2 s-1 ) or short days ( 10 hrs . light/ 14 hrs . dark; 180 μmol m-2 s-1 ) using a 5:3 combination of white ( USHIO F32T8/741 ) and red-enriched ( Interlectric F32/T8/WS Gro-Lite ) fluorescent lights . As indicated in the Results , one experiment was performed with plants growing in at 8 hrs . light/ 14 hrs . dark; 180 μmol m-2 s-1 ) . Plant age was measured from the date seeds were transferred to the growth chamber . For analyses of root development , plants were grown on agar in petri dishes on 1/2 strength Murashige and Skoog medium under long day conditions . Sugar was omitted from the medium because it affects the expression of miR156 [90 , 91] . The miR156-sensitive and miR156-resistant SPL-GUS fusion lines were constructed by placing the GUS gene from pCAMBIA3301 or the GUS+ gene from pCAMBIA1305 , at the 5' or 3' end of the coding sequence of different SPL genes ( S1 Fig ) . In all but two cases , the construct consisted of the genomic sequence extending from gene upstream of the SPL gene to the gene downstream of the SPL gene . The only exceptions were SPL10 and SPL11 , which were constructed according to the strategy described in Yang et al [92] , and only extend to the end of the coding region . These constructs were inserted into pCAMBIA3300 or pCAMBIA3301 [93] and then transformed into Col by floral dipping [94] . The primers used in making these constructs are listed in S4 Table . Transgenic plants were identified using Basta resistance , and their T2 progeny were screened to identify lines segregating 3:1 for the transgene . The T3 progeny of T2 plants were then screened to identify lines homozygous for the insertion . Plants were fixed in 90% acetone on ice for 10 minutes , and were washed with GUS staining buffer ( 4mM potassium ferrocyanide and 4mM potassium ferricyanide in 0 . 1 M PO4 buffer ) , and stained overnight at 37° in 2mM X-Gluc in GUS staining buffer . in situ hybridization was performed on 21-day old plants , which were processed according to Xu et al [46] . The spl10 CRISPR-cas9 mutant lines were generated with the guide RNA ( 5’-GGT ACC TCG AGA GCT GTG GA-3’ ) using protocols described previously [95 , 96] . Primers flanking the guide RNA ( 5’-AGG ACA AAC GAT GCA ATC TTG-3’ , 5’-TTT TCT TCC GAG CAA CAA CAG-3’ ) were used to verify the mutations in the spl10-2 , spl10-3 and spl10-4 alleles . RNA was extracted from the shoot apices of plants grown under SD or LD conditions , as indicated in the text . Shoot apices were harvested by removing the cotyledons and all leaves larger that 5 mm . Total RNA was isolated using Trizol ( Invitrogen ) , and was then treated by RNase-free DNase ( Ambion ) following the manufacturer's instructions . 600 ng RNA was used for reverse transcription of miR156 , miR272 and SnoR101 , using miR156 , miR172 and SnoR101-specific RT primers . Quantification of miR156 and miR172 was performed according to [97] . Quantification of miR172b was performed according to [7] . qPCR reactions were run in triplicate and the results were averaged to produce the value for 1 biological replicate; the data presented here are the average of 2–4 biological replicates . Primers used for RT-PCR and qRT-PCR are listed in S4 Table . Plants were grown on 1/2 MS under LD conditions for 6 days , and primary roots were then removed with a scalpel . Plants were then etiolated for 2 days in darkness and returned to LD conditions . Adventitious roots were analysed 7 days after return to LD conditions .	In Arabidopsis , miR156 acts by repressing the expression of 10 SQUAMOSA PROMOTOR BINDING PROTEIN-LIKE ( SPL ) genes . The phenotype of plants over-expressing miR156 demonstrates that these genes control many aspects of plant development and physiology , but the functions of individual miR156-regulated SPL genes , and how their expression is regulated by miR156 , are largely unknown . We addressed these questions by determining the phenotypes of loss-of-function mutations in these genes individually and in combination , and by comparing the expression patterns and the phenotypes of miR156-sensitive and miR156-resistant reporters for these genes . Our results reveal the unique and shared functions of the members of this gene family , and demonstrate that miR156 plays different roles in the regulation of SPL gene expression at different times in development .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "plant", "anatomy", "floral", "meristem", "gene", "regulation", "brassica", "plant", "physiology", "mutation", "plant", "science", "model", "organisms", "plants", "flowering", "plants", "research", "and", "analysis", "methods", "arabidopsis", "thaliana", "gene", "expression", "leaves", "plant", "and", "algal", "models", "phenotypes", "genetics", "biology", "and", "life", "sciences", "meristems", "organisms" ]	2016	Developmental Functions of miR156-Regulated SQUAMOSA PROMOTER BINDING PROTEIN-LIKE (SPL) Genes in Arabidopsis thaliana
Scabies is an infectious skin disease caused by the mite Sarcoptes scabiei and has been classified as one of the six most prevalent epidermal parasitic skin diseases infecting populations living in poverty by the World Health Organisation . The role of the complement system , a pivotal component of human innate immunity , as an important defence against invading pathogens has been well documented and many parasites have an arsenal of anti-complement defences . We previously reported on a family of scabies mite proteolytically inactive serine protease paralogues ( SMIPP-Ss ) thought to be implicated in host defence evasion . We have since shown that two family members , SMIPP-S D1 and I1 have the ability to bind the human complement components C1q , mannose binding lectin ( MBL ) and properdin and are capable of inhibiting all three human complement pathways . This investigation focused on inhibition of the lectin pathway of complement activation as it is likely to be the primary pathway affecting scabies mites . Activation of the lectin pathway relies on the activation of MBL , and as SMIPP-S D1 and I1 have previously been shown to bind MBL , the nature of this interaction was examined using binding and mutagenesis studies . SMIPP-S D1 bound MBL in complex with MBL-associated serine proteases ( MASPs ) and released the MASP-2 enzyme from the complex . SMIPP-S I1 was also able to bind MBL in complex with MASPs , but MASP-1 and MASP-2 remained in the complex . Despite these differences in mechanism , both molecules inhibited activation of complement components downstream of MBL . Mutagenesis studies revealed that both SMIPP-Ss used an alternative site of the molecule from the residual active site region to inhibit the lectin pathway . We propose that SMIPP-Ss are potent lectin pathway inhibitors and that this mechanism represents an important tool in the immune evasion repertoire of the parasitic mite and a potential target for therapeutics . Scabies is an infectious skin disease caused by the mite Sarcoptes scabiei and has been classified as one of the six most prevalent epidermal parasitic skin diseases infecting populations of the world living in poverty by the World Health Organisation [1] . A quintessential feature of the scabies infection is the broken epidermal tissue resulting from the mite burrowing into the host epidermis and patient scratching . The tissue damage and release of antigens and excretory products from the mite trigger activation of host complement components in the burrow , where the mite ingests them [2] , [3] . The role of complement , a pivotal component of innate immunity , as an important defence against invading pathogens has been well documented and many parasites have an arsenal of anti-complement defences [4] , [5] . To avoid complement-mediated mite gut damage , S . scabiei has evolved an intricate set of complement inhibitors [3] . We previously reported on a family of scabies mite proteolytically inactive serine protease paralogues ( SMIPP-Ss ) thought to be implicated in host defence evasion [3] , [6] . Further studies revealed that at least five members of the SMIPP-S family function as complement inhibitors [7] . Representative SMIPP-Ss of each clade within the thirty-three member family were localised to the mite gut and mite faeces in host skin [8] . The co-localisation of complement components in the mite gut makes this is an appropriate location for scabies mite anti-complement molecules [2] , [3] . We have since shown that two family members , SMIPP-S D1 and I1 ( D1 and I1 respectively ) , have the ability to bind the complement components C1q , mannose binding lectin ( MBL ) and properdin and are capable of inhibiting all three pathways [3] . Crystal structures of D1 and I1 revealed that occlusion of the S1 subsite of both the SMIPP-Ss by a conserved tyrosine residue blocked potential protein or peptide substrate access [9] . It was therefore concluded that the SMIPP-Ss had “lost” the ability to bind substrates in a classical fashion and had evolved alternative interaction sites . Screening of a phage display library , used to identify peptide substrates of several chymotrypsin-like serine proteases , found no evidence of binding by the SMIPP-Ss , supporting the above hypothesis [9] . Overlaying of thirty-three SMIPP-S sequences , on the two crystal structures , revealed areas of high conservation containing surface exposed residues , which could represent potential functional sites for protein-protein interaction . Interestingly , the regions of conservation were found to exist not on the side of the surface containing the defunct canonical active site , but on the opposite side of the molecule [9] . Although previous studies have demonstrated that SMIPP-Ss can inhibit the classical , alternative and lectin pathways [3] the focus of this study is the inhibition of the lectin pathway . The recent identification of a novel peritrophin localised in the mite gut that appears to trigger activation of the lectin pathway [2] suggests that this pathway would be a major target for inhibition by scabies mite defence proteins , amongst which are the SMIPP-Ss . To elucidate the inhibitory mechanism employed by these scabies mite proteins , we now investigated the functional interaction of D1 and I1 with lectin pathway components and used mutagenesis studies to identify functionally relevant residues . These studies demonstrated that D1 and I1 are potent inhibitors of the lectin pathway and appear to bind MBL , thereby suppressing the activation of the lectin pathway . Interestingly , these two SMIPP-Ss seem to inhibit lectin pathway activation by different mechanisms , facilitated through conserved lysine residues on the SMIPP-S surface . Normal human serum ( NHS ) was prepared from the blood of eight healthy donors . Informed written consent was obtained from all blood donors according to the principles of ethical conduct stated in the “National Statement on Ethical Conduct in Human Research” outlined by the Australian National Health and Medical Research Council . The protocol was approved by the medical ethics committee of the QIMR Berghofer Medical Research Institute . Recombinant SMIPP-S protein expression in Pichia pastoris and purification were carried out as previously described [3] . Briefly , mature SMIPP-S protein secreted from P . pastoris was purified by hydrophobic interaction chromatography on HiTrap phenyl-Sepharose columns ( GE Healthcare ) followed by dialysis and ion exchange chromatography on HiTrap SP-Sepharose FF columns ( GE Healthcare ) . Purified protein was concentrated in 10 kDa centrifugal filters ( Amicon Ultra , Millipore ) . Centrifugal flow through material was also collected ( microfiltrate ) and used as a control in assays . Molecular masses and purity of recombinant proteins were confirmed using SDS-PAGE with Coomassie Blue R-250 staining . The purity of the recombinant proteins used in this study were comparable to the protein produced in previous studies [3] . Mutations were introduced in the cDNA sequence of D1 ( YvT004A06; GenBank accession no . AY333085 ) and I1 ( Yv6023A04; GenBank accession no . AY333081 ) by site directed mutagenesis . Residues were substituted with alanine or glutamine as outlined in Table 1 . All mutants were cloned into the vector pPICZαA ( Invitrogen ) using primers outlined in Supporting Information Tables S1 and S2 and constructs were transformed into P . pastoris strain KM71H . Mutations in SMIPP-S constructs were confirmed by ABI PRISM BigDye Terminator 3 . 1 . Mutant proteins were expressed , purified and concentrated as described above . Protein samples were buffer exchanged to assay buffer using Zeba Desalt columns ( Thermo Fischer Scientific , Australia ) . The same volume of SMIPP-S microfiltrate was also buffer exchanged and used in the assay as a control ( microfiltrate ) . Complement proteins and antibodies used were Human MBL ( Statens Serum Institute , Copenhagen , Denmark ) , recombinant MASP-2 , anti-MASP-1 , anti-MASP-2 ( Sapphire Biosciences , Australia ) and recombinant MASP-1 ( Life Research , Australia ) . Normal human serum was prepared from the blood of eight healthy donors . All incubation steps were carried out at room temperature ( RT ) in 50 µl of assay buffer , except washing and blocking , in 250 µl of solution . Microtitre plates ( Maxisorp , Nunc ) were incubated overnight at 4°C with coating buffer ( 50 mM sodium carbonate , pH 9 . 6 ) , 100 µg/ml mannan ( Sigma-Aldrich ) and 1% ( w/v ) BSA ( negative control ) and incubated with blocking buffer ( 1% [w/v] BSA in PBS ) for 2 hr . To analyse the lectin pathway , 2% ( v/v ) NHS , was incubated in GVB2+ buffer ( 5 mM veronal buffer pH 7 . 35 , 140 mM NaCl , 0 . 1% [w/v] gelatin , 1 mM MgCl2 , 0 . 15 mM CaCl2 ) for 20 min ( for detection of C4b and C3b ) or 1 hr ( for detection of MBL and C9 ) at 37°C . NHS was pre-incubated for 15 min at RT with various concentrations of SMIPP-S protein , SMIPP-S microfiltrate or BSA , as a negative control , before addition to the microtiter plate . Complement activation was assessed by detection of deposited complement proteins using antibodies against C4c and C3d ( Dako ) , MBL ( R&D Systems ) and C9 ( Complement Technology ) diluted in blocking buffer . After 1 hr incubation with the primary antibody , HRP-conjugated secondary antibodies against IgG ( Dako ) were diluted in blocking buffer and added for 30 min ( for C4b and C3b detection ) or 1 hr ( for MBL and C9 ) . Bound enzyme was detected using o-1 , 2-phenylenediamine dihydrochloride tablets ( OPD , Dako ) in presence of hydrogen peroxide . Absorbance was measured at 490 nm . The absorbance value obtained in the absence of SMIPP-S was defined as 100% . Each incubation step was followed by extensive washing using washing buffer ( 50 mM Tris-HCl pH 8 . 0 , 150 mM NaCl and 0 . 1% [v/v] Tween-20 ) . Complement activity for protein samples was calculated against the microfiltrate control ( taken as 100% ) and BSA was calculated against the PBS control ( taken as 100% ) . All incubation steps were carried out at RT in 50 µl of assay buffer , except washing and blocking in 250 µl of solution . Microtitre plates ( Maxisorp , Nunc ) were incubated overnight at 4°C with coating buffer ( as above ) containing SMIPP-S proteins ( 10 µg/ml or 0–20 µg/ml ) , mannan ( 100 µg/ml ) ( positive control ) , or 1% ( w/v ) BSA ( negative control ) . After washing and blocking for 2 hr at RT , wells were incubated for 2 hr at RT with MBL ( 0–20 µg/ml ) , rMASP-1 ( 1 µg/ml ) or rMASP-2 ( 1 µg/ml ) in HEPES buffer ( 50 mM HEPES pH 7 . 4 , 100 mM NaCl , 2 mM CaCl2 ) . Bound complement protein was detected by incubation with specific antibody against MBL , MASP-1 , or MASP-2 for 1 hr at RT , followed by incubation with HRP-conjugated secondary antibody for 1 hr at RT . Bound HRP was detected as described for the deposition assays . All incubation steps were carried out at RT in 50 µl of assay buffer , except washing and blocking in 250 µl of solution . Microtitre plates ( Maxisorp , Nunc ) were incubated overnight at 4°C with coating buffer ( as above ) containing SMIPP-S proteins ( 0–20 µg/ml ) or 1% ( w/v ) BSA ( negative control ) . After washing and blocking for 2 hr at RT , wells were incubated for 2 hr at RT with purified MBL ( 10 µg/ml ) in HEPES buffer . MASP-1 or MASP-2 protein was detected by incubation with specific antibody for 1 hr at RT , followed by incubation with HRP-conjugated secondary antibody for 1 hr at RT . Bound HRP was detected as described for the deposition assays . Circular dichroic spectra were acquired on a Jasco J-815 CD/ORD spectrometer at 20°C in a quartz cell of 0 . 1 cm path length . Data were collected at 0 . 5 nm intervals with five scans taken and averaged . The protein concentration used was 0 . 3 mg/ml in 50 mM Tris , pH 7 . 5 . Statistical significance was calculated by two-way ANOVA and Bonferroni post test ( GraphPad Prism ( 6 ) , GraphPad Software Inc . USA ) . Values of p<0 . 05 were considered significant . The deposition of MBL and the downstream components , C4b , C3b and C9 were measured for the lectin pathway following treatment with purified D1 and I1 protein ( figure 1 ) . The purity of the recombinant proteins used in this study were comparable to previous studies [3] . D1 inhibited the deposition of MBL from the lectin pathway by 70% at 10 µg/ml , while I1 inhibited MBL deposition by 80% at 50 µg/ml ( figure 1A , 1B ) . The SMIPP-Ss also significantly inhibited deposition of complement components downstream of MBL in the lectin pathway , but to a lesser extent than the impact on MBL deposition . This suggests that the main molecular target for the SMIPP-Ss is MBL . Levels of deposition of the MBL complex were found to be very sensitive to the presence of SMIPP-Ss , thus it is reasonable to expect that these mite proteins might bind to the complex . The complex includes MBL and MBL-associated serine proteases ( MASPs ) . To determine whether the SMIPP-Ss bound to the MBL:MASP complex or to the MASP proteases , direct binding assays were performed . A dose dependent binding to MBL was observed for both the SMIPP-Ss ( figure 2A , 2B ) . However , similar experiments with recombinant MASP-1 and MASP-2 did not show any binding ( figure 2C , 2D ) . These results suggest that the binding of SMIPP-Ss to the MBL complex is mediated via MBL and not the MASP-1 or MASP-2 . It is possible that the binding of the SMIPP-Ss to the MBL complex could result in conformational changes , causing either a loss of some functional components ( such as the MASPs ) or inhibition of the activation of the MASPs in the complex . To elucidate this , we investigated whether SMIPP-S binding results in a release of MASP-1 and/or MASP-2 from the MBL complex . When specific MASP-1 or MASP-2 antibodies were used to monitor integrity of the MBL complex after binding to both the SMIPP-Ss , MASP-1 was still detectable in the complex ( figure 3A ) . In contrast , while MASP-2 was still detectable after I1 treatment , it was absent after D1 treatment ( figure 3B ) . These observations indicate differences in the mechanism of action of the two SMIPP-Ss . Whereas D1 appears to mediate inhibition of the lectin pathway by causing the release of MASP-2 from the complex , I1 apparently inhibits the pathway without causing the release of either protease . To pinpoint the MBL binding site on the SMIPP-S molecules , regions of conservation on the SMIPP-S surface containing surface exposed conserved residues were investigated . These regions were previously predicted as potential interaction sites in structural studies [9] . Preliminary mutagenesis studies suggested that residue K103 in D1 and K108 in I1 and the surrounding regions were important for the inhibitory effects on complement ( data not shown ) . Sequence alignment of the thirty-three SMIPP-Ss revealed that residue K103 in D1 and K108 in I1 are conserved surface exposed residues that align in the sequence of all members of the SMIPP-S family ( figure 4 ) . Within the tertiary structure , the residue is centrally located within a cluster of conserved surface-exposed residues in its respective SMIPP-S . Focusing on K103 and K108 , single point mutations with alanine or glutamine ( D1-K103A , D1-K103Q , I1-K108A and I1-K108Q ) were used to assess if the residue and/or its charge were functionally relevant in SMIPP-S inhibitory activity . Other conserved and non conserved residues identified in the tertiary structure as being surface exposed and in close proximity to K103 and K108 were also targeted for subsequent mutagenesis studies . Mutants described in this study are shown in Table 1 . All substitution mutants were assessed for their ability to inhibit deposition of complement components . D1 mutants 2 and 3 showed a significant loss of inhibitory activity , as judged by MBL deposition , with no significant differences seen in the other mutants compared to the wild type ( figure 5A ) . All I1 mutants showed a significant reduction in inhibitory activity , with the greatest loss seen in mutant I1-K108A ( figure 5B ) . Collectively these results show that the residues targeted in D1 mutants 2 and 3 and I1 mutants K108A and 4 are involved in preventing the deposition of MBL by these molecules . To confirm that the secondary structure of the SMIPP-S mutant protein circular dichroism spectroscopy was performed with D1 mutant 2 and I1 mutant 4 . These two mutant proteins were chosen as they had yielded the most interesting results for each SMIPP-S and were considered suitable for future assays . Both mutants displayed comparable secondary structure characteristics to the related wild type protein ( data not shown ) . Verification of the integrity of the SMIPP-S structural conformation in the mutants confirmed the functional relevance of the residues targeted for mutation in inhibitory activity . The recent finding that a mite gut wall peritrophin molecule stimulates the activation of the lectin pathway of complement [2] would suggest that this pathway is the most likely target of mite defence molecules , such as the SMIPP-Ss . The two SMIPP-S proteins , prepared as described here , were found to be strong inhibitors of lectin pathway , yielding 50% inhibition of the deposition of MBL with as little as 5 µg/ml D1 ( 191 nM ) and 15 µg/ml I1 ( 589 nM ) . As the in vivo concentrations of SMIPP-Ss in the mite gut and complement proteins , such as MBL , at the infection site have yet to be determined , it is difficult to determine how related these concentrations are physiologically . Most likely in vivo the mite molecules are highly concentrated at the local level within the confined mite gut and the burrows . The assays shown are done under in vitro conditions with recombinant molecules that were expressed and folded in vitro . What percentage of the recombinant molecules is biologically active is unknown . Given that , compared to lectin pathway studies with other parasite recombinant MBL inhibitors these concentrations are comparable [10] . The action of the SMIPP-Ss at the level of MBL in the lectin pathway suggested that MBL is a major target and central to the SMIPP-S inhibitory mechanism . Human MBL is a collagen-like structure comprised of a C-terminal moiety containing carbohydrate recognition domains and a N-terminal moiety containing a collagen-like domain referred to as the stalk [11] . The collagen stalk is the binding site of MASPs , which activate downstream complement components [11] . Given that no direct binding to MASPs was evident , other mechanisms may be operating . Interestingly , in the presence of D1 , MASP-2 was no longer detectable in the complex , suggesting that either D1 binds at the same location on the MBL collagen stalk as MASP-2 or in close proximity resulting in the release of MASP-2 from the complex . Release of MASP-2 from the MBL/MASP complex would most likely prevent its activation and subsequent downstream production of the C3 convertase . By contrast , I1 did not release MASP-1 or MASP-2 from the complex . However , the deposition assays demonstrated that lectin pathway activation was inhibited from the level of MBL ( figures 1A , 1B ) . Based on these data it seems reasonable to assume that I1 binds to an undefined site in the MBL collagen stalk , possibly disrupting MBL conformation or access for the MASP substrates . Evidently , the mechanisms employed by the two SMIPP-Ss differ , but both disrupt MASP activation . A preference for collagen binding is supported by previous studies demonstrating direct binding by SMIPP-Ss to the collagen stalk of C1q [3] . Given the high structural and sequence homology shared by MBL and C1q , one can assume that this same region would be the SMIPP-S binding region in MBL . To further define this binding interaction , mutagenesis studies investigating functional residues involved in lectin pathway inhibition by D1 and I1 were conducted . Preliminary studies identified a specific region on the surface of each SMIPP-S as a potential interaction site . Focusing on these regions , a residue common to both SMIPP-Ss , aligning in the sequence , was identified , K103/K108 . Given the initial thoughts that functionally relevant regions and residues could be conserved between the SMIPP-Ss , residues K103 ( D1 ) and K108 ( I1 ) were investigated . To assess residue importance , a single point mutation to alanine or glutamine was introduced at this position and tested in lectin pathway deposition assays . For D1 , the single point mutation to either an alanine or glutamine ( D1-K103A , D1-K103Q ) residue did not result in the loss of inhibitory activity . The lack of change in the effect on pathway activation suggests that either K103 is not an important functional residue , or that targeting K103 alone was not enough to disrupt the inhibitory activity in D1 . For I1 , both single point mutations ( I1-K108A , I1-K108Q ) were shown to have caused a significant reduction of inhibitory activity . Comparative analysis showed that the alanine mutant displayed the greater effect on function . Since a K>A mutation results in a loss of positive charge and any potential side chain hydrogen bond formation , both factors ( charge and hydrogen bonding ) are functionally relevant for inhibition of the lectin pathway by I1 . The I1 results indicate that K108 is a pivotal residue in the binding mechanism for this protein . We then explored whether additional surface exposed residues identified in the tertiary structure as being surface exposed and in close proximity to K103 and K108 were also involved in the binding mechanism . Examination of the D1 structure showed that K103 resides in a ridge-like formation of polar residues on the D1 surface . Mutations targeted the regions flanking K103 along the ridge ( mutant 2 ) and spanning the ridge ( mutant 3 ) . Residue K108 also sits in a smaller ridge-like formation hence only two additional residues ( K10 and Q11 ) were mutated . Results obtained with constructs 2 and 3 confirmed the relevance of the targeted region in D1 . Given that the inhibitory activity by mutants 2 and 3 were similar , it would appear that the residues common to these mutants ( L31 , K104 and K225 ) may be important in mediating binding to MBL . For I1 , the K108 residue appears to be the most important residue since additional substitution of the K10 and Q11 residues in mutant 4 decreased the effect of the K108A mutation on MBL deposition . Future mutagenesis studies investigating the region may help to narrow down critical sites . Given the above evidence , two potential binding scenarios could be considered: ( 1 ) SMIPP-Ss bind to MBL , releasing the MASPs from the MBL complex; or ( 2 ) the SMIPP-Ss bind to MBL and confer a disturbance to the conformation of the MBL complex disrupting MASP activation . Binding to MBL could be facilitated through a negatively charged region in the collagen stalk recently identified as being responsible for enhancing phagocytosis [12] . This region is in close proximity to the MASP binding site . Once a MBL/MASP-2 complex is formed , the MASP-2 would autoactivate when MBL binds to ligands [13] , [14] . This activation is thought to involve a conformational change in the collagen stalk of MBL , which impacts on the MASP [15] , [16] . The SMIPP-S lysine residues may act as a positively charged patch , which can engage with the negatively charged region on the MBL stalk . As the MBL stalk binding region in D1 is larger than I1 , the D1 binding region may extend into the MASP binding region . This , in turn , could explain why D1 binding results in release of MASP-2 – a phenomenon , which does not occur with I1 . The fact that the binding site for MASP-1 and MASP-2 are not identical explains why only MASP-2 was released [17] . Importantly , either scenario results in a suppression of MASP activation by SMIPP-Ss , and subsequently , suppressed activation of downstream complement components . Given that the inhibitory mode of action appears to differ between the two SMIPP-Ss future studies evaluating if D1 and I1 act synergistically , perhaps even at lower concentrations , will be of great interest . Previous synergistic studies of scabies mite serpins , that are also anti-complement molecules , did demonstrate synergistic activity [18] . If this is the case with the SMIPP-Ss , a multivalent drug design may need to be considered . This study provides evidence supporting the conclusion that SMIPP-Ss inhibit the lectin pathway through binding to MBL . Notably , the results indicate that the mode of action of individual SMIPP-Ss could be different . A family of proteins that target complement pathway ( s ) by different mechanisms would represent a highly sophisticated and diverse immune evasion strategy by this parasite . Further studies will help to clarify the roles of SMIPP-Ss in immune evasion and open avenues to the design of novel therapeutics .	Scabies is a skin infection caused by parasitic scabies mites . There are an estimated 300 million cases globally , with the majority of infections occurring in the world's poorest communities . In Australia , scabies is common in remote Indigenous communities where the infection rate is 16 times higher than the non-Indigenous population . Current treatments have remained relatively unchanged for years and consequently treatment resistance has inevitability emerged . Despite scabies being a well known and frequent infectious skin disease , scabies research has been neglected , resulting in a lack of basic scabies mite biological data . As a result no new therapeutics have been developed . Our research seeks to understand the relationship between the parasite and the human host and one key area of interest is how mites avoid destruction and survive in human skin . We have determined that to survive an attack by the skin's immune defence system the mites release counter defensive proteins that inhibit the skin's defences from activating . This strategy allows the mites to survive in the skin , reproduce and to establish an infection . With this information we can design therapeutics that target these mite proteins , allow the skin to mount an attack and potentially reduce infection .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "dermatology", "infectious", "diseases", "innate", "immune", "system", "medicine", "and", "health", "sciences", "skin", "infections", "clinical", "immunology", "immunity", "sexually", "transmitted", "diseases", "scabies", "neglected", "tropical", "diseases", "biology", "and", "life", "sciences", "immunology", "tropical", "diseases", "parasitic", "diseases", "immunomodulation", "immune", "system" ]	2014	Scabies Mite Inactive Serine Proteases Are Potent Inhibitors of the Human Complement Lectin Pathway
Admixture—the mixing of genomes from divergent populations—is increasingly appreciated as a central process in evolution . To characterize and quantify patterns of admixture across the genome , a number of methods have been developed for local ancestry inference . However , existing approaches have a number of shortcomings . First , all local ancestry inference methods require some prior assumption about the expected ancestry tract lengths . Second , existing methods generally require genotypes , which is not feasible to obtain for many next-generation sequencing projects . Third , many methods assume samples are diploid , however a wide variety of sequencing applications will fail to meet this assumption . To address these issues , we introduce a novel hidden Markov model for estimating local ancestry that models the read pileup data , rather than genotypes , is generalized to arbitrary ploidy , and can estimate the time since admixture during local ancestry inference . We demonstrate that our method can simultaneously estimate the time since admixture and local ancestry with good accuracy , and that it performs well on samples of high ploidy—i . e . 100 or more chromosomes . As this method is very general , we expect it will be useful for local ancestry inference in a wider variety of populations than what previously has been possible . We then applied our method to pooled sequencing data derived from populations of Drosophila melanogaster on an ancestry cline on the east coast of North America . We find that regions of local recombination rates are negatively correlated with the proportion of African ancestry , suggesting that selection against foreign ancestry is the least efficient in low recombination regions . Finally we show that clinal outlier loci are enriched for genes associated with gene regulatory functions , consistent with a role of regulatory evolution in ecological adaptation of admixed D . melanogaster populations . Our results illustrate the potential of local ancestry inference for elucidating fundamental evolutionary processes . Characterizing the biological consequences of admixture—the mixing of genomes from divergent ancestral populations—is a fundamental and important challenge in evolutionary genetics . Admixture has been reported in a variety of natural populations of animals [1 , 2] , plants [3–5] and humans [6 , 7] , and theoretical and empirical evidence suggests that admixture may affect a diverse suite of evolutionary processes . Individuals’ ancestry can affect disease susceptibility in admixed populations , and inferring and correcting for sample population ancestries is a common practice in human genome wide association studies [8–10] . More generally , admixture has the potential to influence patterns of genetic variation within populations [11 , 12] , to introduce novel adaptive [13 , 14] and deleterious variants [7 , 15 , 16] , as well as to disrupt epistatic gene networks [17 , 18] . Therefore , developing a comprehensive understanding of the extent of admixture in natural populations and resulting mosaic genome structures is essential to furthering our understanding of a variety of evolutionary processes . Estimating genome-wide ancestry proportions has become a common practice in population genetic inference . For example , the program STRUCTURE [19] , originally released in 2000 , uses a Bayesian framework to model the ancestry proportions of individuals derived from any number of source populations based on genotype data at a set of unlinked genetic markers . More recently , this model for ancestry proportion estimation has been extended to cases where individual genotypes are not known , but can be studied probabilistically using low-coverage sequencing short read sequencing data [20] , which is an important step towards accommodating modern sequencing practices . Additionally , Bergland et . al . [21] developed a method for estimating ancestry proportions in pooled population samples of relatively high ploidy ( i . e . 40–250 distinct chromosomes ) from short read sequencing data . In general , it is straightforward to estimate genome-wide ancestry proportions using a number of sequencing strategies and applications . It is substantially more challenging to accurately estimate local ancestry ( LA ) at markers distributed along the genome of a sample . Nonetheless , analyses of LA have the potential to yield more nuanced insights into our understanding of the evolutionary processes affecting ancestry proportions across the genome . One of the first LA inference ( LAI ) methods was an extension of the STRUCTURE [19] framework that modeled the correlation in ancestry among markers due to linkage . Because the ancestry at each locus is not observed , Falush et al . [22] suggested that a hidden Markov model ( HMM ) is a straightforward means of inferring the ancestry states at each site in the genome ( which are unobserved ) based on observed genotype data distributed along a chromosome . Most subsequent LAI methods have also used an HMM framework , and the majority are geared towards estimating LA in admixed human populations ( e . g . [23 , 24] ) . Consequently , most existing LAI methods are limited to diploid genomes with high quality genotype calls . Furthermore , many methods require phased reference panels [24 , 25] , and require the user to provide an estimate of , or make implicit assumptions about , the number of generations since the initial admixture event [2 , 23–25] . This is straightforward with human population genomic samples , where abundant high quality genotyped samples are available and for which well-documented demographic histories are sometimes known . However for most other species , demographic histories are less well characterized , and assumptions about admixture times may bias the result of LAI methods . A number of approaches exist to estimate the time since admixture based on well characterized ancestry tract length distributions [26–29] but in general , these parameters are unknown prior to LAI . Conversely , another class of methods can be used to estimate the time of admixture based on the decay of linkage disequilibrium without performing LAI [30–32]; however as with LAI procedure above , these approaches are also limited to diploid genotype data . We may therefore expect to improve LAI by simultaneously estimating LA and demographic parameters ( e . g . admixture time ) . Furthermore , in the majority of sequencing applications , relatively low individual sequencing coverage is often used to probabilistically estimate individual and population allele frequencies ( e . g . [33] ) but these data are often not sufficient to determine high confidence genotypes that are required for existing LAI applications . Hence , there is a clear need for a general LAI method that can accommodate genotype uncertainty and requires less advanced knowledge of admixed populations’ demographic histories . Here , we introduce a framework for simultaneously estimating LA using short read pileup data and the time of admixture within a population . Briefly , as with many previously proposed LAI methods , we model ancestry across the genome of a sample as a HMM . We estimate LA by explicitly modeling read counts as a function of sample allele frequencies within an admixed population . Our method is generalized to accommodate arbitrary sample ploidies , and is therefore applicable to haploid ( or inbred ) , diploid , tetraploid , as well as pooled sequencing applications . We show that this approach accurately infers the time since admixture when data are simulated under the assumed model . Furthermore , our method yields accurate LA estimates for simulated datasets , including samples of high sample ploidy and including evolutionary scenarios that violate the assumptions of the neutral demographic model . In comparisons between ours and an existing LAI method , WINPOP [34] , we find that our approach offers a significant improvement and is accurate over longer time scales . Furthermore , we demonstrate , using a published dataset , that even state-of-the-art LAI methods can be significantly impacted by assumptions about the time since admixture , and that our method provides a solution to this problem . Finally , we apply this method to a Drosophila melanogaster ancestry cline on the east coast of North America . This species originated in sub-Saharan Africa , and approximately 10 , 000–15 , 000 years ago a subpopulation expanded out of the ancestral range . During this expansion , the derived subpopulation experienced a population bottleneck that resulted in decreased nucleotide polymorphism , extended linkage disequilibrium within the derived population and substantial genetic differentiation between ancestral and derived populations [2 , 35–39] . Hereafter , the ancestral population will be referred to as “African” and the derived population as “Cosmopolitan” . Following this bottleneck , descendant populations of African and Cosmopolitan D . melanogaster have admixed in numerous geographic regions [2 , 11 , 21] . Of particular relevance to this work , North America was colonized recently by a population descendent from African individuals from the South , and by a population descendent from cosmopolitan D . melanogaster in the North [11 , 21 , 38] . Where these populations encountered each other in eastern North America , they form an ancestry cline where southern populations have a greater contribution of African ancestry than northern populations [21] . Previous work on these ancestry clines has shown that ancestry proportions vary across populations with increasing proportions of cosmopolitan alleles in more temperate localities . Evidence suggests spatially varying selection affects the distribution of genetic variants [40–45] . Furthermore , strong epistatic reproductive isolation barriers partially isolate individuals from northern and southern populations along this ancestry cline [46 , 47] . This may be generally consistent with recent observations of ancestry-associated epistatic fitness interactions within a D . melanogaster population in North Carolina [17] , and with the observation of widespread fitness epistasis between populations of this species more generally [48] . There is therefore good reason to believe that natural selection has acted to shape LA clines that are tightly linked to selected mutations in these D . melanogaster populations . Here , we show that African ancestry in North American D . melanogaster populations is negatively correlated with recombination rates , consistent with more efficient selection against foreign ancestry in high recombination rate regions of the genome . We also find that the X chromosome displays a higher rate of LA outlier loci , potentially consistent with a greater role of the X chromosome in clinal adaptation . Clinal loci are disproportionately likely to be associated with high level gene regulatory protein complexes , and may play important roles in ecological divergence between African and Cosmopolitan D . melanogaster populations . Furthermore , we identify numerous loci with decreased African ancestry across all populations , which suggests that these alleles that are disfavored on predominantly cosmopolitan genetic backgrounds . This subset of loci is enriched for genes related to oogenesis , potentially consistent with epistatic interactions that affect female reproductive success in these populations . Although admixed populations often are diploid , we derived a general model of ploidy in which the individual has n gene copies at each locus , i . e . for diploid species n = 2 . In practice , sequences are often obtained from fully or partially inbred individuals ( e . g . [39 , 49] ) , which represent only a single uniquely derived chromosome . It is also common to pool individuals prior to sequencing for allele frequency estimation , so called pool-seq ( e . g . [21 , 40 , 42 , 50–53] ) . If the pooling fractions are exactly equal , such a sample of b diploid individuals can be treated as a sample from a single individual with ploidy n = 2b . Although that requirement is restrictive , pool-seq has been experimentally validated as a method for accurate allele frequency estimation—i . e . alleles are approximately binomially sampled from the sample allele frequencies [54] . We therefore aimed to derive a model that can accommodate arbitrary sample ploidies . In the model , we assumed that the focal population was founded following a single discrete admixture event between two ancestral subpopulations , labeled 0 and 1 , with admixture proportions 1-m and m , respectively , at a time t generations in the past . We modeled emission probabilities such that the method can work directly on read pileup data , rather than high quality known genotypes . Briefly , in our model , we specify an HMM {Hv} with state space S = {0 , 1 , … , n} , where Hv = i , i ∈ S , indicates that in the vth position i chromosomes are from population 0 and n–i chromosomes are from population 1 . In other words , this HMM enables one to estimate what ancestry frequencies are present at a given site along a chromosome within a sample . Importantly , we designed this method to simultaneously estimate the time of admixture , which is related to the correlation between ancestry informative markers along a chromosome . See Methods for a complete description of the HMM including the emissions and transition probability calculations . The source code and manual are available at https://github . com/russcd/Ancestry_HMM . For this model , it is assumed that the number of chromosomes present in a sample , n , is known and that the global ancestry proportion , m , is known . As there are many methods for accurately estimating m in a wide variety of contexts implemented in standard population genetic analysis pipelines [19 , 20] , we believe this assumption is not too restrictive . In order to test our method with data of known provenance , we also developed an approach for simulating chromosomes sampled from admixed populations . Briefly , we first simulated genetic diversity consistent with ancestral populations using a coalescent simulation method [55] . We then generated ancestry tracts consistent with admixture models developed to test our inference method using the forward-time admixture simulation program , SELAM [56] . We retained a portion of each coalescent-generated population to serve as a reference panel for allele frequency and LD estimation . We then took the remaining chromosomes and placed them on the appropriate ancestry tracts in admixed chromosomes . Finally , we generated read counts for these chromosomes , or pools of chromosomes for samples with ploidy greater than one , via binomial sampling from the genotype frequencies of the sample . Implicitly , this procedure assumes that the allele frequencies in the reference panel and the admixed individuals whose ancestry is from a given reference panels are equivalent . For large , well-mixed populations such as those of D . melanogaster , this is likely to be a reasonable assumption . Nonetheless , below we assess the impact of differences in the ancestral allele frequencies for plausible demographic models in this species . Within an admixed population , there are two sources of LD . LD that is induced due to the correlation of alleles from the same ancestry type ( i . e . admixture LD ) , and LD that is present within each of the ancestral populations ( ancestral LD ) . Admixture LD , is the signal of LA that we seek to detect using the HMM . The second type , ancestral LD , limits the independence of the ancestral information captured by each marker , and is expected to confound HMM-based analyses , particularly as we aimed to estimate the time since admixture within this framework . We therefore sought to quantify the effect of ancestral LD by discarding one of each pair of sites in LD within either ancestral population . We found that ancestral LD tends to increase admixture time estimates obtained using our method , and we decreased the cutoff of the LD parameter , \|r\| , by 0 . 1 until the time estimates obtained for single chromosomes were unbiased with respect to the true time since admixture . We found that \|r\| ≤ 0 . 4 fit this criterion , although for relatively ancient admixture events with highly skewed ancestry proportions—i . e . m < 0 . 1 or m > 0 . 9—some residual bias was apparent in the estimates of admixture time ( Fig 1 ) . This reflects the fact that the SMC’ ancestry tract distribution performs poorly with highly skewed ancestry proportions and especially for long times since admixture [57] . Fig 1 also reveals a striking difference between otherwise equivalently skewed admixture proportions . For example when m = 0 . 1 , there was a much larger effect of ancestral LD than when m = 0 . 9 . This is due to differences in the variability of LD within the ancestral populations . That is , due to the strong population bottleneck , cosmopolitan D . melanogaster populations have substantially more LD and fewer polymorphic sites than African D . melanogaster populations . Because the time estimation procedure appears to be sensitive to the amount of ancestral LD present in the data , simulations of the type we described here may be necessary to determine what \|r\| cutoffs are required to produce unbiased time estimates given the ancestral LD of the populations in a given analysis using this method . We next sought to quantify the accuracy of our approach across varying sample ploidies and times since admixture ( Fig 2 ) . Especially for moderate and short admixture times ( i . e . 0–500 generations ) , our method performed well for all ploidies considered and we were able to accurately recover the correct admixture time with relatively little bias . However , as true admixture time increases , the time estimates for pooled samples become significantly less reliable and show a clear negative bias . Nonetheless , across the range of times presented in Fig 2 , samples of ploidy one and two showed little bias , and we therefore believe our method will produce sufficiently accurate admixture time estimates for a wide variety of applications . All measures of accuracy decrease with increasing time since admixture ( Fig 2 ) . However , even for relatively long times since admixture—2000 generations—and for large sample ploidies , the mean posterior error remained relatively low for all ancestry proportions and for long times since admixture . This indicates that this approach may be sufficiently accurate for a wide variety of applications , sequencing depths , and sample ploidies . Nonetheless , the proportion of sites within the 95% credible interval decreased with larger pool sizes and it is clear that for larger pools the posterior credible interval tends to be too narrow . Therefore , correcting for this bias may be necessary for applications that are sensitive to the accuracy of the credible interval . An important consideration is that estimates of t will be reliable only if the local recombination rates are known with reasonably high accuracy [58] . In many species , an accurate broad-scale map is available . However , fine-scale variation in recombination rates has only been documented for a few model species . Therefore , for relatively short to moderate times since admixture , error in the genetic map is expected to have a limited impact on date estimates . However , for longer times since admixture , this factor has the potential to bias estimates of t [58] , particularly in species with large variance in local recombination rates ( e . g . due to hotspots ) . Since D . melanogaster has one of the best recombination maps currently available in any species [59] and because we do not aim to estimate time in our applications , we do not believe this will heavily impact the analyses we present below . However , for most applications , it will be necessary to consider the impact of error in the assumed genetic map to accurately interpret estimates of t obtained using this method . We emphasize that this challenge is not unique to this application , but will impact virtually all ancestry estimation methods that rely on a genetic map for estimating the time since admixture . As described above , estimates of the time of admixture demonstrate an apparent bias in pools of higher ploidy ( Fig 2 ) . Specifically , time tends to be slightly overestimated for relatively short admixture times and underestimated at relatively long admixture times . This is particularly apparent at highly skewed ancestry proportions . Given that this bias is primarily evident in pools of 10 to 20 individuals , we hypothesized that it might be due to the non-independence of ancestry tracts among chromosomes , which should tend to disproportionately affect samples of higher ploidy because all ancestry breakpoints are assumed to be independent in our model . To test this , we simulated genotype data from independent and identically distributed exponential tract lengths as is assumed by our model . When we ran our HMM on this dataset , we found that no bias is evident for simulations of up to 2000 generations ( S1 Fig ) , indicating that the primary cause of this bias was violations in the real data of the independence of ancestry tracts that we assumed when computing the transition probabilities . However , it should be possible to quantify and correct for this bias in applications of this method that aim to estimate the time since admixture . The transition probabilities of this HMM depend on knowledge of the population size . In practice , this parameter is unlikely to be known with certainty . Hence , to assess the impact of misspecification of the population size , we performed simulations using a range of population sizes that span three orders of magnitude ( N = 100 , 1000 , 10000 , and 100000 ) . All analyses presented here were conducted by applying our HMM to haploid and diploid samples , but qualitatively similar results hold for samples of larger ploidy . We then analyzed these data assuming the default population size , 10000 , is correct . For relatively short times since admixture , there was not a clear bias for any of the true population sizes considered . However , at longer true admixture times , estimated admixture times for both N = 100 and N = 1000 asymptote at a number of generations near to the population sizes . This result reflects the fact that smaller populations will tend to coalesce at a portion of the loci in the genome relatively quickly , and ancestry tracts cannot become smaller following coalescence . Nonetheless , the accuracy of LAI remained high even when time estimates were unreliable ( S2 Fig ) for the tested marker densities and patterns of LD . Furthermore , in some cases it should be straightforward to determine if a population has coalesced to either ancestry state at a large portion of the loci in the genome , potentially obviating this issue . A more subtle departure from the expectation was evident for population sizes that are larger than we assumed in analyzing these data ( S2 Fig ) . This likely reflects the fact that the probability of back coalescence to the previous marginal genealogy to the left after a recombination event is inversely related to the population size . Hence , the rate of transition between ancestry types is actually slightly higher in larger populations where back coalescence is less likely than we assumed during the LAI procedure . This produced a slight upward bias in the estimates of admixture time when the population was assumed to be smaller than it is in reality . However , this bias appears to be relatively minor , and we expect that time estimates obtained using this method will be useful so long as population sizes can be approximated to within an order of magnitude . Of course , this bias is not unique to our application , and it will affect methods that aim to estimate admixture time after LAI as well . That is , estimating the correct effective population size is an inherent problem for all admixture demographic inference methods . Although it is clear that accurately estimating relatively ancient admixture times is challenging in higher ploidy samples , we sought to determine the limits of our approach for LAI and time estimation for longer admixture times for haploid sequence data . Because of rapid coalescence in smaller samples ( see above ) , we performed admixture simulations with a diploid effective population size of 100 , 000 . It is clear that there is a limit to the inferences that can be made directly using our method . Like the higher ploidy samples , time estimates for haploid samples departed from expectations shortly after 2 , 000 generations since admixture ( S3 Fig ) . Nonetheless , the magnitude of this bias is slight , and it is likely that it could be corrected for when applying this method even for very ancient admixture events . For all admixture times considered , LAI remained acceptably accurate despite the slight bias in time estimates ( S3 Fig ) . One question is what effect varying the reference panel sizes will have on LAI inference using this method . We therefore compared results from reference panels of size 10 chromosomes with those from panels of size 100 chromosomes ( S4 Fig ) . As with results obtained for reference panels of size 50 , panels of size 100 were sufficient to accurately estimate admixture time and LA over many generations since admixture . Whereas , when panel sizes were just 10 chromosomes , time estimates were clearly biased and the result was variable across ancestry proportions ( S4 Fig ) . However , since there was a strong correlation between true and estimated admixture times even with relatively small panel sizes , it may therefore be possible to infer the correct time by quantifying this bias through simulation and correcting for it . Furthermore , although LAI is clearly less reliable with smaller panels , these results are not altogether discouraging and this approach , in conjunction with modest reference panels may still be effective for some applications . Ultimately , there are three reasons why allele frequencies in the reference panels and in the admixed population panel would be expected to differ beyond that expected from binomial samples with the same mean . First , some amount of genetic drift may have occurred in the ancestral population and in the admixed population in the time since the admixed population was founded . Second , in some cases , it is infeasible to sample the ancestral population of an admixed group , and a genetically divergent population must suffice as the reference panel if this method is to be used . Third , divergent selection may quickly modify allele frequencies between admixed and ancestral populations . Hence , genetic divergence between reference and admixed populations may be an important challenge for this method . To address this , we simulated the second scenario , where increasingly divergent populations are used as the reference panels to study admixed populations . In order to make this relevant to the application to D . melanogaster populations , below , we selected times for divergence that might be consistent with differences across continental populations in Sub-Saharan Africa and in Cosmopolitan populations . Although time estimates obtained using this approach are weakly positively biased with increasing divergence between the ancestral population and reference panels , the accuracy of this LAI method is largely unaffected ( S5 Fig ) . Hence , for biological scenarios potentially consistent with those of D . melanogaster ancestral populations , we do not expect this challenge to strongly bias our method . Nonetheless , in applications to other populations , with potentially differently structured ancestral populations , it would be necessary to examine the effects of this bias in detail . In a wide variety of pool-seq applications , samples are pooled in larger groups than we have considered above ( e . g . [40 , 50 , 52] ) . We are therefore interested in determining how our method will perform on pools of 100 individuals . Towards this , we performed simulations as before , but we designed our parameters to resemble those of the pooled sequencing data that we analyze in the application of this method below . Specifically , we simulated data with a mean sequencing depth of 25 , a time since admixture of 1500 generations , and an ancestry proportion of 0 . 8 . Consistent with results for ploidy 20 , we found that time tends to be dramatically underestimated ( i . e . the mean estimate of admixture time was 680 generations ) . However , when we provided the time since admixture , our method produced reasonably accurate LAIs for these samples . Although the posterior credible interval was again too narrow , the mean posterior error was just 0 . 053 when expressed as an ancestry frequency , indicating that this approach can produce LA estimates that are close to their true values for existing sequencing datasets ( e . g . Fig 3 ) . However , the HMM’s run time increases dramatically for higher ploidy samples and higher sequencing depths , a factor that may affect the utility of this program for some analyses . Nonetheless , for more than 36 , 000 markers , a sample ploidy of 100 and a mean sequencing depth of 25 , the average runtime was approximately 42 hours . In contrast , for the same set of parameters , but where individuals are sequenced and analyzed as diploids , the mean runtime was just 8 minutes ( See S1 Table for a comparison of run times across many parameter sets ) . An important concern is that many biologically plausible admixture models would violate the assumptions of this inference method . In particular , continuous migration and selection acting on alleles from one parental population are two potential causes of deviation from the expected model in the true data . To assess the extent of this potential bias , we performed additional simulations . First , we considered continuous migration at a constant rate that began t generations prior to sampling . In simulations with continuous migration , additional non-recombinant migrants enter the population each generation . Relative to a single pulse admixture model , this indicates that the ancestry tract lengths will tend to be longer than those under a single pulse admixture model in which all individuals entered at time t . Indeed , we found that admixture times tended to be underestimated with models of continuous migration . However , the accuracy of LAI remained high across all situations considered here ( Table 1 ) , indicating that the LAI aspect of this approach may be robust to alternative demographic models . In the second set of simulations , we considered additive selection on alleles that are perfectly correlated with local ancestry in a given region ( i . e . selected sites with frequencies 0 in population 0 and frequency 1 in population 1 ) , and experience relatively strong selection ( selective coefficients were between 0 . 005 and 0 . 05 ) . We placed selected sites at 2 , 5 , 10 and 20 loci distributed randomly across the simulated chromosome , where admixture occurred through a single pulse . Ancestry tracts tend to be longer immediately surrounding selected sites , and we therefore expected admixture time to be underestimated when selection is widespread . When the number of selected loci was small , time estimates were nearly unbiased ( Table 2 ) , suggesting that our approach can yield reliable admixture time estimates despite the presence of a small number of selected loci ( i . e . 2 selected loci on a chromosome arm ) . However , with more widespread selection on alleles associated with local ancestry , time estimates showed a downward bias that increased with increasing numbers of selected loci . This is likely because selected loci will tend to be associated with longer ancestry tracts due to hitchhiking . However , the accuracy of the LAI remains high for all selection scenarios that we considered here , further indicating that our method can robustly delineate LA , even when the data violate assumptions of the inference method ( Tables 1 and 2 ) . We next compared the results of our method to those of WinPop [34] . Because WinPop accepts only diploid genotypes , we provided this program diploid genotype data . However , for these comparisons , we still ran our method on simulated read pileups with the mean depth equal to 2 . WinPop was originally designed for local ancestry inference in very recently admixed populations . As expected , WinPop performed acceptably for very short admixture times , but rapidly decreased in performance with increasing time ( S6 Fig ) . However , by default , WinPop removes sites in strong LD within the admixed samples , which includes ancestral LD , but also admixture LD—the exact signal LAI methods use to identify ancestry tracts . We therefore reran WinPop , but instead of pruning LD within the admixed population , we removed sites in strong LD within the ancestral populations as described above in our method . With this modification , WinPop performs nearly as well as our method , but remains slightly less accurate especially at longer admixture times ( S6 Fig ) . This difference presumably reflects the windowed-based approach of WinPop . At longer times since admixture a given genomic window may overlap a breakpoint between ancestry tracts . Although the performance is nearly comparable with this modification , we emphasize that our method enables users to estimate the time since admixture , where this must be supplied for WinPop , and allows for LAI on read pileups , therefore incorporating genotype uncertainty into the LAI procedure . Indeed our method is more accurate at longer timescales even when supplied with considerably lower quality read data . However WinPop supports LAI with multiple ancestral populations , which our method currently does not ( but see Conclusions ) . Furthermore many LAI algorithms utilize haplotype information , which may be particularly valuable in populations where LD extends across large distances as in e . g . human populations . Given the strong interest in studying admixture and local ancestry in human populations ( e . g . [22–25] ) , it is useful to ask if our method can be applied to data consistent with admixed populations of humans . Towards that goal , we simulated data similar to what would be observed in admixture between modern European and African lineages and applied our HMM to estimate admixture times and LA . We found that our method can accurately estimate admixture times for relatively short times since admixture , however , substantially more stringent LD pruning in the reference panels is necessary to produce unbiased estimates ( Fig 4 ) . This may be expected given that linkage disequilibrium extends across longer distances in human populations than it does in D . melanogaster . In other words , the scales of ancestral LD and admixture LD become similar rapidly in admixed human populations . Furthermore , this approach yields accurate time estimates for shorter times since admixture than with genetic data consistent with D . melanogaster populations . For a relatively short time since admixture , around 100 generations , it is possible to obtain accurate and approximately unbiased estimates of the admixture time over a wide range of ancestry proportions , indicating that this method may be applicable to recently admixed human populations as well ( Fig 4 ) . Nonetheless , this result underscores the need to examine biases associated with LD pruning in this approach prior to application to a given dataset . To demonstrate that assumptions about the number of generations since admixture have the potential to bias LAI , we analyzed a SNP-array dataset from Greenlandic Inuits [60 , 61] . The authors had previously noted a significant impact of t on the LAI results produced using RFMix [24] , which we were able to reproduce here for chromosome 10 ( S7 Fig ) . Indeed , even for comparisons between t = 5 and t = 20 , both of which may be biologically plausible for these populations , the mean difference in posterior probabilities between samples estimated using RFMix was 0 . 0903 ( S7 Fig ) . However , when we applied our method to these data , a clear optimum from t was obtained at approximately 6–7 generations prior to the present , which is close to the plausible times of admixture for these populations ( S7 Fig ) . This comparison therefore demonstrates that even relatively minor changes in assumptions of t have the potential to strongly impact LAI results , and underscores the importance of simultaneously performing LAI while estimating t . However , these results also indicate that our method may not be robust in situations where the background LD is high and ancestry informative markers are neither common nor distributed evenly across the genome . When we compared the results of our method at t = 5 and at t = 20 , we also obtained differences in the mean posterior among individuals as with RFMix . However , one notable difference is that the mean posterior difference using RFMix has a particularly high variance and therefore higher mean error ( S7 Fig ) , but actually a lower median difference than we found using our method . There are likely two causes for differences observed in the mean ancestry posterior among individuals . First , the datasets considered were generated with a metabochip SNP-chip [62] , which contains a highly non-uniform distribution of markers across the genome . Second , the ancestral LD in the Inuit population is extensive [61] , and we could only retain a relatively small proportion of the markers after LD pruning in the reference panels . These results therefore also underscore the challenges of LAI when the signal to noise ratio is low as may be the case in some human populations , for which LD is extensive , and for some sequencing strategies . Although in general it is straightforward to estimate m from genome-wide data , in some cases this parameter may be misestimated prior to LAI . We therefore sought to quantify this potential effect by performing LAI after supplying incorrect values of m . In general , we found that values close to the true range , i . e . within 0 . 05 of the true m , tend to yield reasonably accurate time estimates . However , increasingly incorrect values produce sharply downwardly biased time estimates and this effect is especially pronounced for highly skewed true m ( S8 Fig ) . As could be expected given the robustness of LAI to many perturbations ( above ) , when the incorrect t is supplied to the program , the LA results remain reasonable . However it is worth noting that the penalty appears to be greatest when t is too small rather than too large ( S9 Fig ) . Although this is not a primary focus for this work , for some users it may be of interest to construct confidence intervals for estimates of t . We recommend the block bootstrap as the preferred method for estimating confidence interval for t , and we have written a script that will produce these ( available on the github page for this project: https://github . com/russcd/Ancestry_HMM ) . Simulations confirm that this can produce confidence intervals overlapping the true t ( S10 Fig ) , but bias in t estimates for higher ploidy samples may still be apparent in some cases . Given their effects suppressing recombination in large genomic regions , chromosomal inversions may be expected to strongly affect LAI [2 , 63] . Although we attempted to limit the impact of chromosomal inversions by eliminating known polymorphic arrangements from the reference panels ( see methods ) , many known inversions are present within the pool-seq samples we aimed to analyze [64] . We therefore focused on known inverted haplotypes within the DGPR samples [63 , 65–67] , which are comprised of inbred individuals , and therefore phase is known across the entire chromosome . In comparing LA estimates between inverted and standard arrangements , it is clear that chromosomal inversions can substantially affect LA across the genomes ( Fig 5 ) . In general , the chromosomal inversions considered in this work originated in African populations of D . melanogaster [63] , and consistent with this observation , most inversion bearing chromosomes showed evidence for elevated African ancestry . This was particularly evident in the regions surrounding breakpoints , where recombination with standard arrangement chromosomes is most strongly suppressed . Importantly , this pattern continued outside of inversion breakpoints as well , consistent with numerous observations that recombination is repressed in heterokaryotypes in regions well outside of the inversion breakpoints in Drosophila ( e . g . [2 , 63 , 68] ) . In ( 3R ) Mo is an exception to this general pattern of elevated African ancestry within inverted arrangements ( Fig 5 ) . This inversion originated within a cosmopolitan population [63] , and has only rarely been observed within sub-Saharan Africa [69 , 70] . Consistent with these observations , In ( 3R ) Mo displayed lower overall African ancestry than chromosome arm 3R than standard arrangement chromosomes . Although chromosomal inversions may affect patterns of LA in the genome on this ancestry cline , we believed including chromosomal inversions in the pool-seq datasets would not heavily bias our analysis of LA clines . Inversions tend to be low frequency in most populations studied [64] , and because they affect LA in broad swaths of the genome—sometimes entire chromosome arms—including inversions is unlikely to affect LA cline outlier identification which appears to affect much finer scale LA ( below ) . Furthermore , inversion breakpoint regions were not enriched for LA cline outliers in our analysis ( S2 Table ) , suggesting that inversions have a limited impact on overall patterns of local ancestry on this cline . Nonetheless , the LAI complications associated with chromosomal inversions should be considered when testing selective hypotheses for chromosomal inversions as genetic differentiation may be related to LA , rather than arrangement-specific selection in admixed populations such as those found in North America . Finally , we applied our method to ancestry clines between cosmopolitan and African ancestry D . melanogaster . Genomic variation across two ancestry clines have been studied previously [21 , 38 , 40 , 52] . In particular , the cline on the east coast of North America has been sampled densely by sequencing large pools of individuals to estimate allele frequencies , and previous work has shown that the overall proportion of African ancestry is strongly correlated with latitude [21] . Consistent with this observation , we found a significant negative correlation for all chromosome arms between proportion of average African ancestry and latitude ( rho = -0 . 891 , -0 . 561 , -0 . 912 , -0 . 913 , and -0 . 755 , for 2L , 2R , 3L , 3R , and X respectively ) . Although global ancestry proportions have previously been investigated in populations on this ancestry cline [21 , 38] , these analyses neglected the potentially much richer information in patterns of LA across the genome . We therefore applied our method to these samples . Because of the relatively recent dual colonization history of these populations and subsequent mixing of genomes , a genome-wide ancestry cline is expected [21] . However , loci that depart significantly in clinality from the genome-wide background levels may indicate that natural selection is operating on a site linked to that locus . Previously Pool ( 2015 ) found that regions of low recombination are disproportionately enriched for African ancestry in the Raleigh , NC population [17] . Here , we find a similar pattern and we further find that is replicated across all populations that were assayed on this ancestry cline . Specifically , in all populations studied the proportion of African ancestry is significantly negatively correlated with local recombination rates ( Fig 6 ) . Ultimately , this correlation may have two causes . First , if selection is more efficient at purging African alleles in high recombination regions , these loci will tend to be removed preferentially in those genomic regions . An alternative explanation is that introgressing African alleles that are favored by selection would tend to bring larger linkage blocks along with them in the predominantly low recombination regions . Regardless of the specific source of natural selection , a neutral admixture model would not predict this robust correlation between LA and recombination rates within all populations , indicating that natural selection has played an important role in shaping LA on this ancestry cline . Previous studies have found that heterogeneity in the genome with respect to ancestry informative markers may impact the accuracy of LAI [71] . To assess this possibility , we computed the mean difference between posterior mean estimates for the two samples from Florida and between the two samples from Maine . Importantly , because these pooled samples were created using different isofemale lines [40] , this is a conservative test of our method since there will be true biological differences as well as stochastic sequencing differences between replicates from each population . We found no correlation between the mean difference of the posterior means and local recombination rates ( P = 0 . 2353 and P = 0 . 7529 , Spearman’s rank correlation for Florida and Maine respectively ) , indicating that the correlation observed between local recombination rates and LA is unlikely to be an artifact of differential accuracy of LAI in different genomic regions . However , it should be acknowledged that in some genomic regions it maybe challenging to unambiguously infer LA [17 , 71] . Selection within admixed populations may take several distinct forms . On the one hand , loci that are favorable in the admixed population—either because they are favored on an admixed genetic background , enhance reproductive success in an admixed population , or are favorable in the local environment—will tend to achieve higher frequencies , and we would expect these sites to have a more positive correlation with latitude than the genome-wide average . Conversely , loci that are disfavored within the admixed population may be expected to skew towards a more negative correlation with latitude . Although it is not possible to distinguish between these hypotheses directly , a majority of evidence suggests that selection has primarily acted to remove African ancestry from the largely Cosmopolitan genetic backgrounds found in the Northern portion of this ancestry cline . First , abundant evidence suggests pre-mating isolation barriers between some African and cosmopolitan populations [72–74] . Second , there is strong post-mating isolation between populations on the ends of this cline [46 , 47] . Third , we report here a strong negative correlation between LA frequency and local recombination rates ( above ) . Finally , circumstantially , the local environment on the east coast of North America is perhaps most similar to the environment of Cosmopolitan compared to African ancestral populations , which further suggests that Cosmopolitan alleles are likely favored through locally adaptive mechanisms . For these reasons , we therefore examined loci that are outliers for a negative partial correlation with latitude , as this is the expected pattern for African alleles that are disfavored in more temperate populations . In other words , the outlier regions show a significantly stronger negative correlation between local African ancestry and latitude than the chromosome arm does on average . There is an ongoing debate about the relative merits of an outlier approach versus more sophisticated models for detecting and quantifying selection in genome-wide scans . We believe that the difficulties of accurately estimating demographic parameters for this ancestry cline make the outlier approach most feasible for our purposes . Using our outlier approach , we identified 80 loci that showed the expected negative partial correlation with latitude ( Fig 7 ) . Although the specific statistical threshold that we employed is admittedly arbitrary , given the strength of evidence indicating widespread selection on local ancestry in this species ( above ) , we expected that the tail of the LA cline distribution would be enriched for the genetic targets of selection . Due to the differences in inheritance , evolutionary theory predicts that selection will operate differently on the X chromosome relative to autosomal loci . Of specific relevance to this work , the large-X effect [75 , 76] is the observation that loci on the X chromosome contribute to reproductive isolation at a disproportionately high rate . Additionally , and potentially the cause of the large-X effect , due to the hemizygosity of X-linked loci , the X chromosome is expected to play a larger role in adaptive evolution , the so-called faster-X effect [77] . There is therefore reason to believe that the X chromosome will play a significant role in genetically isolating Cosmopolitan and African D . melanogaster . Consistent with a larger role for the sex chromosomes in generating reproductive isolation or selective differentiation between D . melanogaster ancestral populations , we found that that the X chromosome has a lower mean African ancestry proportion than the autosomes in all populations . Furthermore , the X displays a stronger correlation between local recombination rates and the frequency of African ancestry than the autosomes in all 14 populations samples , potentially indicating that selection has had a disproportionately strong effect shaping patterns of local ancestry on this chromosome than on the autosomes . In addition , the X has a significantly higher rate of outlier LA clinal loci than the autosomes ( 23 LA outliers on the X , 57 on the Autosomes , p = 0 . 0341 , one-tailed exact Poisson test ) . Although consistent with evolutionary theory , differences between autosomal arms and the X chromosome may also be explained in part by differences in effective recombination rates on the X chromosome than the autosomes , differences in power to identify LA clines associated with chromosome arm specific patterns , or by the disproportionately larger number of chromosomal inversions on the autosomes than on the X chromosome in these populations [64 , 69] . Distinguishing between this hypothesis and confounding factors will be central to determining whether key results from speciation research are replicated in much more recently diverged populations . We next applied gene ontology analysis to the set of outlier genes to identify common biological attributes that may suggest more specific organismal phenotypes underlying LA clinal outliers . In total , we identified seven GO terms that remained significant after applying a 5% FDR correction ( S3 Table ) . These GO terms reflect the presence of two primary clusters of genes . The first , which corresponds broadly to histone acetylation , may be related to chromatin remodeling and therefore is expected to effect gene expression levels across a large number of loci . Previous work focused on this ancestry cline has identified chromatin remodeling genes as a potentially important component locally adaptive variation on this ancestry cline [78] . This may indicate that this previous efforts to identify spatially varying selection in these populations may have been detecting selection on local ancestry components associated with ecological adaptation in ancestral populations . The second GO cluster , eukaryotic translation initiation factor 2 complex , also appears to implicate a central role of clinal LA outliers on the regulation of gene expression . One plausible explanation of these observations is that gene expression , particularly high level regulation of gene expression , may be especially likely to contribute to epistatic interactions as these proteins will inherently interact with a diverse set of loci throughout the genome . Given that two distinct gene clusters related to gene expression are identified by this analysis , gene expression would appear to be a plausible candidate phenotype to investigate in future work on ecological divergence and isolating factors in admixed D . melanogaster populations . Testing this prediction empirically through expression profiling may therefore offer fruitful grounds of understanding the earliest stages of reproductive isolation . Another subset of loci that we may wish to identify using these data are those that contribute to reproductive isolation between African and Cosmopolitan D . melanogaster populations and would therefore be removed by selection from most populations on this ancestry cline . Although it is possible that Cosmopolitan alleles would be disfavored in an admixed background as well , because these populations are predominantly Cosmopolitan , we expect that the majority of selection on negatively epistatically interacting loci would remove African alleles from populations . To identify these loci , we first computed the mean African ancestry across all populations , and we then identified the subset of loci that were in the lowest 5% tail . From those loci , we selected the loci minima from adjacent genomic windows ( see Methods , Fig 8 ) , and we obtained a total of 84 local ancestry outliers . As with the clinal outlier analysis above , to identify commonalities in the types of loci identified by this analysis , we performed GO analysis on the set of loci that are outliers for the mean proportion of African ancestry . After a 5% FDR correction , there are again several gene clusters that are significantly enriched in this set of outlier loci ( S4 Table ) . Of particular interest is the GO term oogenesis , which may indicate that female reproduction is affected during admixture between cosmopolitan and African populations of D . melanogaster . This finding is particularly interesting in light of the fact that female fertility is strongly affected when autosomal chromosomes from one end of this ancestry cline are made homozygous on a genetic background carrying the X chromosome from the other end of this ancestry cline [47] . Hence , the effects of combining divergence ancestry types on female fertility , and specifically the genetic basis of oogenesis , may be an appealing phenotype to characterize in detail in attempting to clarify the genetic effects that isolate African and Cosmopolitan D . melanogaster populations . Given the abundance of evidence supporting a role for pre-mating isolation barriers between African and Cosmopolitan flies [72–74] , we are interested in highlighting genes potentially related to behavioral isolation between ancestral populations of D . melanogaster . Consistent with this observation , one of the strongest LA cline outliers , egh , has been conclusively linked to strong effects on male courtship behavior using a variety of genetic techniques [79] . Additionally , gene knockouts of CG43759 , another LA cline outlier locus , have strong effects on inter-male aggressive behavior [80] , and may also contribute to behavioral differences between admixed individuals . These loci are therefore appealing candidate genes for functional follow-up analyses , and illustrate the power of this LAI approach for identifying candidate genes that are potentially associated with well characterized phenotypic differences between ancestral populations . The Pennsylvania population included in this study has been sampled extensively , including several paired fall and spring samples across three consecutive years . Previously , Bergland et al . [40] identified numerous SNPs that showed recurrent and rapid seasonal frequency changes in the Pennsylvania populations included in this study . They concluded that these sites are experiencing recurrent selection associated with recurrent environmental seasonal changes . To determine if LA across the D . melanogaster genome might also experience selection associated with seasonal frequency shifts , we searched for loci that showed a strong recurrent seasonal shift in LA . However , we identified fewer significantly seasonal sites than we would expect to by chance ( the proportion of significant site at the alpha = 0 . 05 level of significance is 0 . 041 ) . Furthermore , after applying a false discovery rate correction [81] , there are no sites that are significantly seasonal at the q = 0 . 1 level . Collectively , these results indicate that LA within the Pennsylvania populations of D . melanogaster remains remarkably stable during seasonal environmental cycles . Although this observation may , to a first approximation , appear to be at odds with the results reported in Bergland et al . [40] , we believe that it is consistent with the model proposed in that work . Specifically , the authors suggested that long term balancing selection may maintain these seasonally favorable polymorphisms in diverse D . melanogaster populations and even in the ancestors of D . melanogaster and D . simulants [40] . We therefore may expect that these polymorphisms will be maintained at similar frequencies in African and Cosmopolitan populations . Hence , although the seasonal SNPs change rapidly in frequency between spring and fall [40] , the LA at these sites can remain stable during seasonal fluctuations . A growing number of next-generation sequencing projects produce low coverage data that cannot be used to unambiguously assign individual genotypes , but which can be analyzed probabilistically to account for uncertainty in individual genotypes [82–84] . However , most existing LAI methods require genotype data derived from diploid individuals . Hence , there is an apparent disconnect between existing LAI approaches and the majority of ongoing sequencing efforts . In this work , we developed the first framework for applying LAI to pileup read data , rather than genotypes , and we have generalized this model to arbitrary sample ploidies . This method therefore has immediate applications to a wide variety of existing and ongoing sequencing projects , and we expect that this approach and extensions thereof will be valuable to a number of researchers . Although evaluating this application is beyond the scope of this work , one particularly enticing potential use of this method is LAI in ancient DNA samples for which sequencing depths often preclude accurate genotype calling . Importantly , it would be straightforward to model site-specific errors in this framework , which could be particularly important for ancient DNA applications [6] . For many applications , a parameter of central biological interest is the time since admixture began ( t ) . A wide variety of approaches have been developed that aim to estimate t and related parameters in admixed populations [26 , 28–31 , 85 , 86] . Many of these methods are based on an inferred distribution of tract lengths , however , inference of the ancestry tract length distribution is associated with uncertainty that is typically not incorporated in currently available methods for estimating t . Furthermore , incorrect assumptions regarding t have the potential to introduce biases during LAI . Hence , it is preferable to estimate demographic parameters such as the admixture time during the LAI procedure . Nonetheless , as noted above , although LAI using our method is robust to many deviations from the assumed model , admixture time estimates are sensitive to a variety of potential confounding factors and examining the resulting ancestry tract distributions after LAI may be necessary to confirm that the assumed demographic model provides a reasonable fit to the data . To our knowledge , this is the first method that attempts to simultaneously link LAI and population genetic parameter estimation directly , and we can envision many extensions of this approach that could expand the utility of this method to a broad variety of applications . For example , it is straightforward to accommodate additional reference populations ( e . g . by assuming multinomial rather than binomial read sampling ) . Alternatively , any demographic or selective model that can be approximated as a Markov process could be incorporated—in particular , it is feasible to accommodate two-pulse admixture models and possibly models including ancestry tracts that are linked to positively selected sites . Such methods can be used to construct likelihood ratio tests of evolutionary models and for providing improved parameter estimates . We model the ancestry using an HMM {Hv} with state space S = {0 , 1 , … , n} , where Hv = i , i ∈ S , indicates that in the vth position i chromosomes are from population 0 and n–i chromosomes are from population 1 . In the following , to simplify the notation and without loss of generality , we will omit the indicator for the position in the genome as the structure of the model is the same for all positions of equivalent ploidy . We assume each variant site is biallelic , with two alleles A and a , and the availability of reference panels from source populations 0 and 1 with total allelic counts C0a , C1a , C0A , and C1A , where the two subscripts refer to population identity and allele , respectively . Also , C0 = C0A + C0a and C1 = C1A + C1a . Finally , we also assume we observe a pileup of r reads from the focal population , with rA and ra reads for alleles A and a respectively ( r = rA + ra ) . The emission probability of state i ∈ S of the process is then defined as ei = Pr ( rA , C0A , C1A \| r , C0 , C1 , H = i , ε ) , where ε is an error rate . This probability can be calculated by summing over all possible genotypes in the admixed sample and over all possible population identities of the reads , as explained in the following section . The probability of obtaining r0 ( = r–r1 ) reads , in the admixed population , from chromosomes of ancestry 0 , given r and the hidden state H = i , and assuming no mapping or sequencing biases , is binomial , r0\|H=i , n , r∼Bin ( r , i/n ) ( 1 ) These probabilities are pre-computed in our implementation for all possible values of i ∈ S and r0 , 0 ≤ r0 ≤ r . Similarly , for the reference populations , for j = 0 , 1 , CjA\|Cj , fj∼Bin ( Cj , fj ) ( 2 ) where fj is the allele frequency of allele A in population j . The analogous allelic counts in the admixed population , denoted CM0a , CM1a , CM0A , and CM1A , are unobserved ( only reads are observed for the admixed population ) , but are also conditionally binomially distributed , i . e . : CM0A\|H=i , f0∼Bin ( i , f0 ) andCM1A\|H=i , n , f1∼Bin ( n−i , f1 ) ( 3 ) Finally , in the absence of errors , and assuming no sequencing or mapping biases , the conditional probability of obtaining r0A reads of allele A in the admixed population is r0A\|H=i , r0 , CM0A∼Bin ( r0 , CM0A/i ) ( 4 ) It should be noted that because we are explicitly modeling the process of sampling alleles from the population ( Eq 3 ) and the process of sampling reads conditional on the sample allele frequencies ( Eq 4 ) , that this approach corrects for the increased variance associated with two rounds of binomial sampling in poolseq applications that has been reported previously ( e . g . , in [52] ) . This probability can be expanded to include errors , in particular assuming a constant and symmetric error rate ε between major and minor allele , and assuming all reads with nucleotides that are not defined as major or minor are discarded , we have r0A\|H=i , ro , CM0A , ε∼Bin ( r0 , ( 1−ε ) CM0A/i+ε ( 1−CM0A/i ) ) . ( 5 ) Using these expressions , and integrating over allele frequencies in the source populations , we have Pr ( r0A , C0A , \|r0 , C0 , n , H=i , ε ) =∫01∑k=0iPr ( r0A\|H=i , r0 , CM0A=k , ε ) Pr ( CM0A=k\|H=i , f0 ) p ( f0 ) df0=C0 ! i ! ( C0−C0A ) ! C0A ! ( C0+i+1 ) ! ∑k=0iPr ( r0A\|H=i , r0 , CM0A=k , ε ) ( C0−C0A+i−k ) ! ( C0A+k ) ! ( i−k ) ! k ! ( 6 ) assuming a uniform [0 , 1] distribution for f0 . A similar expression is obtained for Pr ( r1A , C1A , \|r1 , C1 , n , H = i , ε ) , assuming f1 ∼ U[0 , 1] , and these expressions combine multiplicatively to give Pr ( rA , C1A , , C0A , \|r0 , C0 , r1 , C1 , n , H=i , ε ) =∑r0A=max{0 , rA−r1}min{r0 , rA}Pr ( r0A , C0A , \|r0 , C0 , n , H=i , ε ) Pr ( r1A=rA−r0A , C1A , \|r1 , C1 , n , H=i , ε ) , ( 7 ) and the emission probabilities become Pr ( rA , C0A , C1A\|r , C0 , C1 , H=i , ε ) =∑r0=0rPr ( r0\|H=i , n , r ) Pr ( rA , C1A , , C0A , \|r0 , C0 , r1=r−r0 , C1 , n , H=i , ε ) ( 8 ) Alternatively , if the sample genotypes are known with high confidence , i . e . CMA = CM0A + CM1A is observed , the emission probabilities are the defined as Pr ( CMA , C0A , C1A\|C0 , C1 , n , H=i ) = ( C0C0A ) ( C1C1A ) ∑k=max{CMA−i , 0}min{n−i , CMA}∫01 ( n−ik ) ( f0 ) C0A+k ( 1−f0 ) C0+n−i−C0A−kdf0∫01 ( iCMA−k ) ( f1 ) CMA−k+C1A ( 1−f1 ) C1+i−C1A−CMA+kdf1=∑k=max{CMA−i , 0}min{n−i , CMA}C0 ! C1 ! i ! ( n−i ) ! ( CMA+C1A−k ) ! ( C0A+k ) ! ( C1−CMA−C1A+i+k ) ! ( C0−C0A−i−k+n ) ! ( C0−C0A ) ! C0A ! ( C1−C1A ) ! C1A ! ( CMA−k ) ! k ! ( k+i−CMA ) ! ( n−k−i ) ! ( n−i+C0+1 ) ! ( i+C1+1 ) ! ( 9 ) These emissions probabilities are sometimes substantially faster to compute than those for short read pileups , especially when sequencing depths are high . However , the genotypes must be estimated with high accuracy for this approach to be valid . For applications with low read coverage , or with ploidy >2 for which many standard genotype callers are not applicable , it is usually preferable to use the pileup-based approach described above . We assume an admixed population , of constant size , with N diploid individuals , in which a proportion m of the individuals in the population where replaced with migrants t generations before the time of sampling . Given these assumptions , and an SMC’ model of the ancestral recombination graph [87] , the rate of transition from ancestry 0 to 1 , along the length of a single chromosome , is λ0=2Nm ( 1−e−t2N ) ( 10 ) per Morgan [57] . Similarly , the rate of transition from ancestry 1 to 0 on a single chromosome is λ1=2N ( 1−m ) ( 1−e−t2N ) ( 11 ) per Morgan . Importantly , because these expressions are based on a coalescence model , they account for the possibility that a recombination event occurs between two tracts of the same ancestry type and the probability that the novel marginal genealogy will back-coalesce with the previous genealogy [57] . Both events are expected to decrease the number of ancestry switches along a chromosome and ignoring their contribution will cause overestimation of the rate of change between ancestry types between adjacent markers . The transition rates are in units per Morgan , but can be converted to rates per bp , by multiplying with the recombination rate in Morgans/bp , rbp within a segment . The transition probabilities of the HMM for a single chromosome , P ( l ) = {Pij ( l ) } , i , j ∈ S , between two markers with a distance l between each other , is then approximately P ( l ) =[1−λ0rbpλ0rbpλ1rbp1−λ1rbp]l ( 12 ) using discrete distances , or P ( l ) =[λ1λ0+λ1+λ0λ0+λ1e−rbpl ( λ0+λ1 ) λ0λ0+λ1−λ0λ0+λ1e−rbpl ( λ0+λ1 ) λ0λ0+λ1+λ1λ0+λ1e−rbpl ( λ0+λ1 ) λ1λ0+λ1−λ1λ0+λ1e−rbpl ( λ0+λ1 ) ] ( 13 ) using continuous distances along the chromosome . Here , we use the continuous representation for calculations . We emphasize that the assumption of a Markovian process is known to be incorrect [57] , in fact admixture tracts tend to be more spatially correlated than predicted by a Markov model , and the degree and structure of the correlation depends on the demographic model [57] . Deviations from a Markovian process may cause biases in the estimation of parameters such as t . The Markov process defined above is applicable to a single chromosome . We now want to approximate a similar process for a sample of n chromosomes from a single sequencing pool . The true process is quite complicated , and we choose for simplicity to approximate the process for n chromosomes sampled from one population , as the union of n independent chromosomal processes . We will later quantify biases arising due to this independence assumption using simulations . Under the independence assumption , the transition probability from i to j is simply the probability of l transitions from state 1 to state 0 in the marginal processes and j–i + l transitions from state 0 to state 1 , summed over all admissible values of l , i . e . , Pr ( Hv+k=j\|Hv=i ) =∑l=max{0 , i−j}min{n−j , i} ( n−ij−i+l ) ( P01 ( k ) ) j−i+l ( 1−P01 ( k ) ) n−j+i−l ( il ) ( P10 ( k ) ) l ( 1−P10 ( k ) ) i−l ( 14 ) Although this procedure can be computationally expensive when there are many markers , read depths are high , and especially when n is large , in our implementation , we reduce the compute time by pre-calculating and storing all binomial coefficients . A parameter of central biological interest , that is often unknown in practice , is the time since the initial admixture event ( t ) . We therefore use the HMM representation to provide maximum likelihood estimates of t using the forward algorithm to calculate the likelihood function . As this is a single parameter optimization problem for a likelihood function with a single mode , optimization can be performed using a simple golden section search [88] . Default settings for this optimization in our software , including the search range maxima defaults , tmax and tmin , are documented in the C++ HMM source code provided at https://github . com/russcd/Ancestry_HMM . After either estimating or providing a fixed value of the time since admixture to the HMM , we obtained the posterior distribution for all variable sites considered in our analysis using the forward-backward algorithm , and we report the full posterior distribution for each marker along the chromosome . To validate our HMM , we generated sequence data for each of two ancestral populations using the coalescent simulator MACS [55] . We sought to generate data that could be consistent with that observed in Cosmopolitan and African populations of D . melanogaster , which has been studied previously in a wide variety of contexts [2 , 11 , 35–37] . We used the command line “macs 400 10000000 -i 1 -h 1000 -t 0 . 0376 -r 0 . 171 -c 5 86 . 5 -I 2 200 200 0 -en 0 2 0 . 183 -en 0 . 0037281 2 0 . 000377 -en 0 . 00381 2 1 -ej 0 . 00382 2 1 -eN 0 . 0145 0 . 2” to generate genotype data . This will produce 200 samples of ancestry 0 and 200 samples of ancestry 1 on a 10mb chromosome—i . e . this should resemble genotype data for about half of an autosomal chromosome arm in D . melanogaster . Unless otherwise stated below , we then sampled the first 50 chromosomes from each ancestral population as the ancestral population reference panel , whose genotypes are assumed to be known with low error rates . The sample size was chosen because it is close to the size of the reference panel that we obtained in our application of this approach to D . melanogaster ( below ) . To evaluate the performance of our method on data consistent with human populations , we simulated data that could be consistent with that observed for modern European and African human populations . Specifically , we simulated the model of [89] using the command line “macs 200 1e8 -I 3 100 100 0 -n 1 1 . 682020 -n 2 3 . 736830 -n 3 7 . 292050 -eg 0 2 116 . 010723 -eg 1e-12 3 160 . 246047 -ma x 0 . 881098 0 . 561966 0 . 881098 x 2 . 797460 0 . 561966 2 . 797460 x -ej 0 . 028985 3 2 -en 0 . 028986 2 0 . 287184 -ema 0 . 028987 3 x 7 . 293140 x 7 . 293140 x x x x x -ej 0 . 197963 2 1 -en 0 . 303501 1 1 -t 0 . 00069372 -r 0 . 00069372” . Admixture between ancestral populations was then simulated as described below . Although it is commonly assumed that admixture tract lengths can be modeled as independent and identically distributed exponential random variables ( e . g . [26 , 29] and in this work , above ) , this assumption is known to be incorrect as ancestry tracts are neither exponentially distributed , independent across individuals , nor identically distributed along chromosomes [57] . We therefore aim to determine what bias violations of this assumption will have on inferences obtained from this model . Towards this , we used SELAM [56] to simulate admixed populations under the biological model described above . Because this program simulated admixture in forward time , it generates the full pedigree-based ancestral recombination graph , and is therefore a conservative test of our approach relative to the coalescent which is known to produce incorrect ancestry tract distributions for short times [57] . Briefly , we initialized each admixed population simulation with a proportion , m , of ancestry from ancestral population 1 , and a proportion 1-m ancestry from ancestral population 0 . Unless otherwise stated , all simulations were conducted with neutral admixture and a hermaphroditic diploid population of size 10 , 000 . We then assigned the additional , non-reference chromosomes from the coalescent simulations , to each ancestry tract produced in SELAM simulations according to their local ancestry along the chromosome . In this way , each chromosome is a mosaic of the two ancestral subpopulations . See , e . g . [2] , for a related approach for simulating genotype data of admixed chromosomes . Correlations induced by LD between markers within ancestral populations violates a central assumption of the Markov model framework . Although it may be feasible to explicitly model linkage within ancestral populations ( e . g . , [24 , 25] ) , when ancestral populations have relatively little LD , such as those of D . melanogaster , another effective approach is to discard sites that are in strong LD in the ancestral populations . Hence , to avoid this potential confounding aspect of the data , we first computed LD between all pairs of markers within each reference panel that are within 0 . 01 centimorgans of one another . We then discarded one of each pair of sites where \|r\| in either reference panel exceeded a particular threshold , and we decreased this threshold until we obtained an approximately unbiased estimate of the time since admixture estimates of the HMM . This approach differs from a previous method , WinPop [34] , where LD is pruned from within admixed samples ( see also below ) . We first identified all sites where the allele frequencies of the ancestral populations differ by at least 20% within the reference panels . We excluded weakly differentiated sites to decrease runtime and because these markers carry relatively little information about the LA at a given site . Then , to generate data similar to what would be produced using Illumina sequencing platforms , we simulated allele counts for each sample , by first drawing the depth at a given site from a Poisson distribution . In most cases and unless otherwise stated , the mean of this distribution is set to be equal to the sample ploidy . We did this to ensure equivalent sequencing depth per chromosome regardless of pooling strategy , and because this depth is sufficiently low that high quality genotypes cannot be determined . We then generated set of simulated aligned bases via binomial sampling from the sample allele frequency and included a uniform error rate of 1% for both alleles at each site . Unless otherwise stated , we simulated a total of 40 admixed chromosomes . The HMM can perform LAI on more than one sample at a time , and we therefore included all samples when running it . Hence , we used 40 haploid , 20 diploid , 4 pools of 10 chromosomes , and 2 pools of 20 chromosomes for most comparisons of accuracy reported below , unless otherwise stated . It is worth noting that it is possible to jointly analyze distinct samples from the same subpopulation that have been sequenced at different ploidies . To investigate the effects of allele frequency differences between reference populations and admixed populations , we performed coalescent simulations using the software MACS [55] , using the command line “ . /macs 500 10000000 -i 1 -h 1000 -t 0 . 0376 -r 0 . 171 -c 5 86 . 5 -I 8 100 100 50 50 50 50 50 50 0 -en 0 2 0 . 183 -en 0 . 0037281 2 0 . 000377 -en 0 . 00381 2 1 -ej 0 . 00382 2 1 -eN 0 . 0145 0 . 2 -ej 0 . 0005 3 2 -ej 0 . 000500001 4 1 -ej 0 . 001 5 2 -ej 0 . 001000001 6 1 -ej 0 . 002 7 2 -ej 0 . 002000001 8 1” . This might be expected to produce populations that are differentiated similarly to how populations of D . melanogaster would be across European populations or between populations in Central Africa . We then substituted the increasingly divergent populations for the reference panel . All allele frequency differences and LD pruning were performed as described above on each of the substitute reference panels . To evaluate the performance of the HMM , we computed four statistics . First , we compute the proportion of sites where the true state is within the 95% posterior credible interval , where ideally , this proportion would be equal to or greater than 0 . 95 . As this HMM has discrete states , there are many ways the 95% credible interval could be defined . In light of the fact that the credible interval tends to be narrow ( Results ) , we defined the interval to include all states that are overlapped , by any amount , in the 95% confidence interval of the posterior distribution . Second , we compute the mean posterior error , the average distance between the posterior distribution of hidden states and the true state E=∑v=0S∑i=0np ( Hv=i\|r ) \|i−Iv\|Sn Here S is the total number of sites , Iv is the true state at site v , and r is all the combined read data . Third , we also report the proportion of sites where the maximum likelihood estimate of the hidden state is equal to the true ancestry state . Finally , as an indicator of the specificity of our approach , we also report the average width of the 95% credible interval . A potential issue with this framework is that the assumptions underlying the transition matrixes and related time of admixture estimation procedure is likely to be violated in a number of biologically relevant circumstances . We therefore simulated populations wherein individuals of ancestral population 1 began entering a population entirely composed of individuals from ancestral population 0 , at a time t generations before the present , at a constant rate that is sustained across all subsequent generations until the time of sampling . That is , additional unadmixed individuals of ancestry 1 migrate each generation from t until the present . Natural selection acting on admixed genetic regions has been inferred in a wide variety of systems ( e . g . [5 , 7 , 13 , 17 , 18] ) , and is expected to have pronounced effects on the distribution of LA among individuals within admixed populations . Here again , this aspect of biologically realistic populations will tend to violate central underlying assumptions of the model assumed in this work . Towards this , we simulated admixed populations with a single pulse of admixture t generations prior to the time of sampling . We then incorporated selection at 2 , 5 , 10 , and 20 loci at locations uniformly distributed along the length of the chromosome arm . All selected loci were assumed to be fixed within each ancestral population . Selection was additive and selective coefficients were assigned based on a uniform [0 . 005 , 0 . 05] distribution to either ancestry 0 or 1 alleles with equal probability . As above , these simulations were conducted using SELAM [56] . For both selected and continuous migration simulations , we then performed the genotype and read data simulation procedure , and reran our HMM as described above . We performed 10 simulations for each treatment . We next sought to compare our method to a commonly used local ancestry inference method , WinPop [34] . Towards this , we again simulated data using MACS and SELAM as described above . For these comparisons , the initial ancestry contribution was 0 . 5 and the number of generations since admixture varied between 5 and 1000 . For comparison , we supplied WinPop and our program the correct time since admixture and ancestry proportions , as these are required parameters for WinPop . We also supplied the program with genotypes rather than read counts , another requirement of WinPop , whereas we supplied our HMM with read data simulated as described above . We then ran WinPop under default parameters , and we also reran WinPop using LD pruning within the reference panels , as we do in our method , instead of the default LD pruning implemented in WinPop . To demonstrate that LAI methods can be biased by the arbitrary selection of the time since admixture , we analyzed a dataset of SNP-array genotype data from Greenlandic Inuits . These data are described in detail elsewhere [60 , 61] . This population has received some admixture from a European source population , and the authors had previously used RFMix [24] to perform LAI , and found some sensitivity to the assumed time since admixture ( J . Crawford pers . Comm . ) . We analyzed data from chromosome 10 using RFMix v1 . 5 . 4 [24] as described in Moltke et al . [61] assuming admixture occurred either 5 or 20 generations ago . We subsequently analyzed chromosome 10 using our HMM including the genotype-analysis emissions probabilities and assuming a genotype error rate of 0 . 2% . For our analysis we identified the LD cutoff that is appropriate for these data as described above . To generate reference panels , we used a subset of the high quality D . melenaogaster assemblies that have been described previously in Pool et al . ( 2012 ) and Lack et al . ( 2015 ) . As in the local ancestry analysis of Pool ( 2015 ) , we used the French population . For our African reference panel , we selected a subset of the Eastern and Western African populations ( CO , RG , RC , NG , UG , GA , GU ) and grouped them into a single population for the purposes of our analysis . We elected to combine populations so that we would have a larger reference panel of African populations for this analysis , this solution may be justified because these D . melanogaster populations are only weakly genetically differentiated [2 , 21 , 90] , particularly after common inversion-bearing chromosomes are removed from analyses . Specific individuals were selected for inclusion in the African reference panel if previous work found they have relatively little cosmopolitan ancestry ( i . e . , below 0 . 2 genome-wide in [2] ) . Because of their powerful effects on recombination , chromosomal inversions are known to have substantial impacts on the distribution of genetic variants on chromosomes containing chromosomal inversions in D . melanogaster [2 , 63] . For this reason , we removed all common inversion-bearing chromosome arms from the reference populations [91] . Nonetheless , it is clear that chromosomal inversions are present in the pool-seq samples [64] . Although the inversions certainly violate key assumptions of our model—particularly the transmission probabilities—given that our approach is robust to a many perturbations , we expect the LA within inverted haplotypes can be estimated with reasonable confidence , and the overall LAI procedure will still perform adequately with low frequencies of chromosomal-inversion bearing chromosomes present within these samples . Although these reference populations are believed to have relatively little admixture , some admixture is likely to remain within these samples [2] . To mitigate this potential issue , we first applied our HMM to each reference population using the genotype-based emissions probabilities ( above ) . Calculated across all individuals , we found that our maximum likelihood ancestry estimates were identical with those of Pool et al . ( 2012 ) at 96 . 2% of markers considered in our analysis . The differences between the results of these methods may reflect differences in the methodology of LAI or differences in the reference panels . Nonetheless , the broad concordance suggests the two methods are yielding similar overall results . We masked all sites where the posterior probability of non-native ancestry was greater than 0 . 5 within each reference individual’s genome . These masked sequences were then used as the reference panel for the analyses of pool-seq data below . We acquired pooled sequencing data from six populations from the east coast of the United States . The generation of these samples , sequencing data , and accession numbers are described in detail in [21 , 40] . Briefly , the samples are comprised of individuals drawn from natural populations and sequenced in relatively large pools of 66–232 chromosomes . We aligned all data using BWA v0 . 7 . 9a-r786 [92] using the ‘MEM’ function and the default program parameters . For all alignments , we used version 5 of the D . melanogaster reference genome [93] in order to make our analysis and coordinates compatible with the Drosophila genome nexus [91] . We then realigned all reads using the indelrealigner tool within the GATK package [84] , and we extracted the sequence pileup using samtools mpileup v1 . 1 [94] using the program’s default parameters . We extracted sites at ancestry informative positions within the reference panels , where we required that the reference panel have a minimum of 50% of individuals with a high quality genotype call in both Cosmopolitan and African reference populations . As above , ancestry informative sites were defined as those with a minimum of 20% difference in allele frequencies between the reference panels used , and we retained only ancestry informative sites for our analyses . We then produced global ancestry estimates for each chromosome arm separately for each sample using the method of Bergland et al . ( 2016 ) . We ran our HMM for each chromosome arm and each population , and we provided the program this estimate of the ancestry proportion and the time since admixture , 1593 generations [17] . We elected to provide the time since admixture because we have found that this parameter is difficult to estimate in relatively large pools ( see Results ) . However , the program can accurately estimate LA in high ploidy samples even when the time since admixture cannot be estimated correctly ( see Results ) . To assess the correlation between local recombination rates and LA in the genome , we computed Spearman’s rank sum correlation between the proportions African ancestry and the local recombination rates in windows of 100 ancestry informative markers . As above , we used the recombination rate estimates of [59] . We estimated confidence intervals using 1000 block-bootstrap samples using window sizes of 100 SNPs . To determine if there are systematic biases in LAI across the genome , we computed the mean difference in genomic windows between LA estimates for two samples form Maine and between two samples from Florida . We assessed evidence for systematic biases through the correlation between local recombination rates and differences in local ancestry inference using Spearman’s rank sum correlation . To detect loci that show evidence for steeper ancestry clines than the genomic average , we first computed the Spearman’s rank correlation between mean ancestry proportions and latitude for each chromosome arm separately . Then , for each site for which we obtained a posterior ancestry distribution for all samples , we computed the partial Spearman’s rank correlation between the posterior ancestry mean and latitude while correcting for the correlation between latitude and the overall ancestry proportion . We then computed the probability of obtaining the observed partial correlation in R , which implements the approach of [95] , and we retained those sites where the probability of the partial correlation between local ancestry and latitude was less than 0 . 005 as significant in our analysis . Although this cutoff is arbitrary , given the strong evidence for local adaptation and reproductive isolation in these populations [46 , 47 , 96] , the tail of the LA cline distribution will likely be enriched for sites experiencing selection on this ancestry gradient . Due to linkage , adjacent sites show strong autocorrelation . We therefore selected the local optima for a given clinally significant LA segment ( i . e . a tract where all positions are significantly correlated with latitude at our threshold ) and retained these for analyses of outlier loci . Finally , to further reduce the effect of autocorrelation , we retained only those local optima for which no other optimum had a stronger correlation with latitude within 100 , 000bp on either side on the site . To identify loci with a disproportionately low proportion of African ancestry across this ancestry cline , we computed the mean African ancestry across all populations . We then selected those sites in the lowest 5% tail on each chromosome arm and selected only the local minima within 100kb windows on either side of a selected locus . We performed Gene-ontology ( GO ) analyses on outlier SNPs using Gowinda [97] , where the background set of SNPs was all positions at which we obtained a posterior distribution in all samples ( i . e . the set on which we obtained estimates of the posterior probability of African ancestry ) . We ran the program using default parameters , except that we included all genes within 10000bp of a focal SNP , and we performed 1e6 total GO simulations . To identify recurrent seasonal changes in the local ancestry , we followed an approach similar to [40] . Specifically , we fit a generalized linear model of the form MeanPosteriorAncestry∼Season+ε We then recorded the estimated effect size , and probability of the observed correlation for each site in the genome at which we obtained a posterior ancestry distribution in all samples considered . To correct for multiple testing , we applied a false discovery rate correction [81] to the resulting p-value distribution .	When divergent populations hybridize , their offspring obtain portions of their genomes from each parent population . Although the average ancestry proportion in each descendant is equal to the proportion of ancestors from each of the ancestral populations , the contribution of each ancestry type is variable across the genome . Estimating local ancestry within admixed individuals is a fundamental goal for evolutionary genetics , and here we develop a method for doing this that circumvents many of the problems associated with existing methods . Briefly , our method can use short read data , rather than genotypes and can be applied to samples with any number of chromosomes . Furthermore , our method simultaneously estimates local ancestry and the number of generations since admixture—the time that the two ancestral populations first encountered each other . Finally , in applying our method to data from an admixture zone between ancestral populations of Drosophila melanogaster , we find many lines of evidence consistent with natural selection operating to against the introduction of foreign ancestry into populations of one predominant ancestry type . Because of the generality of this method , we expect that it will be useful for a wide variety of existing and ongoing research projects .	[ "Abstract", "Introduction", "Results", "and", "Discussion", "Methods" ]	[ "invertebrates", "markov", "models", "population", "genetics", "ploidy", "animals", "invertebrate", "genomics", "animal", "models", "drosophila", "melanogaster", "model", "organisms", "mathematics", "inversions", "experimental", "organism", "systems", "population", "biology", "drosophila", "research", "and", "analysis", "methods", "chromosome", "biology", "chromosomal", "aberrations", "hidden", "markov", "models", "genetic", "loci", "insects", "probability", "theory", "animal", "genomics", "arthropoda", "cell", "biology", "genetics", "biology", "and", "life", "sciences", "physical", "sciences", "genomics", "evolutionary", "biology", "organisms", "chromosomes" ]	2017	A Hidden Markov Model Approach for Simultaneously Estimating Local Ancestry and Admixture Time Using Next Generation Sequence Data in Samples of Arbitrary Ploidy
Differentiation of extracellular Leishmania promastigotes within their sand fly vector , termed metacyclogenesis , is considered to be essential for parasites to regain mammalian host infectivity . Metacyclogenesis is accompanied by changes in the local parasite environment , including secretion of complex glycoconjugates within the promastigote secretory gel and colonization and degradation of the sand fly stomodeal valve . Deletion of the stage-regulated HASP and SHERP genes on chromosome 23 of Leishmania major is known to stall metacyclogenesis in the sand fly but not in in vitro culture . Here , parasite mutants deficient in specific genes within the HASP/SHERP chromosomal region have been used to investigate their role in metacyclogenesis , parasite transmission and establishment of infection . Metacyclogenesis was stalled in HASP/SHERP mutants in vivo and , although still capable of osmotaxis , these mutants failed to secrete promastigote secretory gel , correlating with a lack of parasite accumulation in the thoracic midgut and failure to colonise the stomodeal valve . These defects prevented parasite transmission to a new mammalian host . Sand fly midgut homogenates modulated parasite behaviour in vitro , suggesting a role for molecular interactions between parasite and vector in Leishmania development within the sand fly . For the first time , stage-regulated expression of the small HASPA proteins in Leishmania ( Leishmania ) has been demonstrated: HASPA2 is expressed only in extracellular promastigotes and HASPA1 only in intracellular amastigotes . Despite its lack of expression in amastigotes , replacement of HASPA2 into the null locus background delays onset of pathology in BALB/c mice . This HASPA2-dependent effect is reversed by HASPA1 gene addition , suggesting that the HASPAs may have a role in host immunomodulation . Kinetoplastid parasites of the genus Leishmania cause a diverse spectrum of mammalian infectious diseases , the leishmaniases , ranging from cutaneous and mucosal pathologies to potentially fatal visceral infections [1] . Endemic human cases have been reported on all continents except Australia and Antarctica [2] . Leishmania parasites are transmitted by female phlebotomine sand flies ( Diptera: Psychodidae: Phlebotominae ) of the genera Phlebotomus ( Old World ) and Lutzomyia ( New World ) [3] . All mammalian-infective Leishmania species ( spp . ) belong to two characterized subgenera , L . ( Leishmania ) [4] and L . ( Viannia ) [5] . Within the sand fly , parasites of both subgenera undergo metacyclogenesis , a series of morphological and functional changes that produce mammalian-infective metacyclic promastigotes . This process can be triggered , at least in vitro , by various factors including nutrient depletion , reduction of pH and tetrahydrobiopterin levels [6 , 7] . As recently described [8] , mammalian-resident intracellular amastigotes transform into midgut adapted , proliferative promastigotes post blood meal ( PBM ) and these in turn produce nectomonads ( synonymous with elongated nectomonads [9 , 10] ) that mediate midgut attachment [11] . Nectomonads then transform into proliferative leptomonads ( synonymous with short nectomonads [9 , 10] ) , which produce promastigote secretory gel ( PSG ) containing filamentous proteophosphoglycan ( fPPG ) [12 , 13] . Finally , leptomonads transform into the mammalian-infective , midgut-detached metacyclics [14] and haptomonads which colonise and degrade the stomodeal valve ( SV ) [15 , 16] . While little is known about the molecular regulation of metacyclogenesis , several genes have been identified that are specifically expressed in late stages of the process . The best-characterised of these are found at the L . ( L . ) major cDNA16 locus on chromosome 23 and are termed the L . ( Leishmania ) species-specific HASP ( hydrophilic acylated surface protein ) and SHERP ( small hydrophilic endoplasmic reticulum-associated protein ) genes ( HASPA1 , SHERP1 , SHERP2 , HASPB and HASPA2; [17 , 18] ) . The locus is conserved in other L . ( Leishmania ) species but divergent in L . ( Viannia ) species [19] . The three HASPs are highly related proteins with identical N- and C-terminal regions . HASPB , the best characterized of these proteins , is trafficked to and tethered at the parasite cell surface by co-translational N-myristoylation and post-translational palmitoylation at its N-terminal SH4 domain [20 , 21] . HASPB contains extensive amino acid repeat domains in its central region and these show inter- and intraspecific variations while bearing some resemblance to peptidoglycan and immunoglobulin-binding domains of several bacterial surface proteins [22–24] . HASPB is specifically expressed in L . ( L . ) major metacyclics and amastigotes , but is only detectable in amastigotes in L . ( L . ) mexicana [19 , 23 , 25] . The HASPA genes , which do not encode amino acid repeats , have identical 5’ untranslated regions ( UTRs ) and open reading frames ( ORFs ) but different 3’UTR sequences . These contribute to distinct mRNA expression patterns: HASPA2 mRNA is expressed early in procyclics and peaks in metacyclics , while HASPA1 mRNA is upregulated in metacyclics and amastigotes [17 , 24 , 26] . SHERP , a small membrane associated protein , is expressed predominantly in metacyclic parasites , where it localizes to the cytosolic faces of the endoplasmic reticulum and mitochondrion [27] and can fold in the presence of membrane phospholipids , supportive of a role in protein-protein interactions [28] . The in vitro binding of SHERP to vacuolar H+-ATP synthase components involved in subcellular compartment acidification has led to the hypothesis that SHERP may impact on parasite autophagocytosis [28] , a process shown to be essential for metacyclogenesis [29] . Genetic deletion of the whole cDNA16 locus in L . ( L . ) major , by homologous recombination , generated mutants that were stalled in metacyclogenesis within the sand fly , predominately in the nectomonad stage [30] . By contrast , the same mutants showed no significant phenotype when maintained in in vitro culture with the culture-generated metacyclics proving more virulent than the parental parasite line ( Friedlin V1; FVI ) in BALB/c mice [31] . In the same study , episomal-replacement of the full cDNA16 locus into the null background led to unregulated HASP and SHERP overexpression and avirulence [31] . Conversely , more recent reintegration of the whole cDNA16 locus into its former location on chromosome 23 re-established parental line gene regulation and rescued metacyclogenesis in vivo [30] . The replacement of an episomal HASPB copy alone into the null background suggested that HASPB was key for the completion of metacyclogenesis in vivo [30] . However , since unregulated episomal expression can cause misleading phenotypes in Leishmania , verification of this observation by HASPB reintegration into the original cDNA16 locus became essential . The main focus here was to investigate the contribution of the individual HASPs and SHERP to metacyclogenesis within the sand fly midgut and to host transmission , utilising a broad panel of newly-validated genetic mutants , generated and rigorously tested in vitro for this study . While these aims were not fully met , passaging all mutant lines through sand flies to investigate the respective contribution of the HASP and SHERP proteins to metacyclogenesis in vivo revealed clear differences in gene expression and parasite behaviour when comparing in vitro and in vivo conditions . This stimulated a first investigation into the potential impact of midgut factors on gene regulation in Leishmania . Further experiments with a sub-group of mutants confirmed that completion of metacyclogenesis , PSG formation , SV colonisation and parasite-to-host transmission are dependent on HASP and SHERP genes . Use of these mutants also allowed us , for the first time , to address the previously established differences in HASPA1 and HASPA2 mRNA expression at the protein level and to investigate the respective contribution of HASPA1 and HASPA2 to amastigote virulence in vivo . For this study , a total of seventeen new L . ( L . ) major HASPA1 , SHERP , HASPB and/or HASPA2 replacement lines were generated by homologous recombination of newly synthesized gene replacement constructs ( S1 Fig ) into the original chromosomal location of the cDNA16 locus within the null background of the previously characterized cDNA16 double deletion mutant ( cDNA16 dKO [31]; Tables 1 and S1 ) . The alternative approach of targeted gene disruption/deletion of individual genes was not possible technically due to high levels of sequence identity and repetition within the cDNA16 locus . For clarity , and due to strong similarities in mutant phenotypes , the data for only 6 representative gene-replacement lines are shown and discussed here ( HASPB sKI , SHERP sKI , HASPA1 sKI , HASPA2 sKI , HASPA1/2 sKI & HA1/2+S2/HB sKI; Table 1 ) . Additional data from other lines and clones can be viewed in the supplementary files . L . ( L . ) major FVI ( the wild type parental line ) and the previously characterized cDNA16 dKO and full cDNA16 locus single replacement mutant lines ( cDNA16 sKI [30] ) served as controls in all experiments . Briefly , newly generated HASP and/or SHERP replacement lines were rigorously tested in vitro to ensure correct construct integration and regulated expression in the former cDNA16 locus , using a series of standard protocols , and selected clones ( at least two per genotype ) were then used for further analyses . Clones were initially screened by PCR ( S2 Fig ) , followed by Southern blot and qPCR analysis to ensure correct integration of HASP and SHERP gene replacement constructs ( Figs 1A , 1B , 1C and S2–S5 ) . For Southern blot analysis , genomic DNA ( gDNA ) from selected clones was SacI digested , size separated and probed with suitable DIG-labelled DNA fragments ( Figs 1B and S4 ) . The HASP probe hybridized to HASPA2 , HASPA1 and HASPB in FVI , detecting the previously observed 7 . 6 Kb , 4 . 3 Kb and 2 . 2 Kb fragments , respectively [30] . The HASPA1 sKI , HASPA1/2 sKI , HASPA2 sKI and HASPB sKI mutants also showed single fragments of the expected sizes ( 6 . 1 Kb , 9 . 3 Kb , 7 . 5 Kb and 6 . 7 Kb , respectively ) , while no fragments were observed in the cDNA16 dKO and SHERP sKI mutant lines . The HA1/2+S2/HB sKI mutant line , containing all the cDNA16 locus gene types , showed two expected fragments: one of 9 . 3 Kb equivalent to that detected in the HASPA1/2 sKI mutant line ( due to integration of the same HASPA1/2 construct ) ; the second of 2 . 2 Kb , matching the HASPB fragment detected in FVI , as expected . The SHERP probe confirmed the previously observed 1 . 8 Kb and 1 . 6 Kb fragments in FVI [30] and hybridized to an expected 4 . 2 Kb fragment in SHERP sKI and to an expected 1 . 6 Kb fragment in HA1/2+S2/HB sKI , matching one of the SHERP fragments in FVI . The BSD ( blasticidin resistance ) gene [32] was only detected in the HASPA1 and SHERP/HASPB ( S2/HB ) constructs; a single fragment each in the HASPA1 sKI ( 6 . 1Kb ) and in the HA1/2+S2/HB sKI ( 2 . 6 Kb ) mutant lines as expected , with no BSD hybridising fragments in FVI and cDNA16 dKO . The NEO ( neomycin resistance ) gene [33] , present in the HASPA1/HASPA2 ( HASPA1/2 or HA1/2 ) , HASPA2 , HASPB and SHERP constructs , generated fragments of the same size as those detected by the HASP and SHERP probes since there was no SacI restriction site between the gene of interest and the resistance marker cassette ( Figs 1A and S3 ) . The Southern analysis also excluded the presence of multi-copy episomal constructs in the parasite lines . Gene copy number in selected clones was verified by gene-specific qPCR ( Figs 1C and S5 ) . The Na/H antiporter-like protein gene , present as a single copy on chromosome 23 , was used as control for data normalization . A value of 1 ( ±0 . 2 ) was predicted for all targets ( HASPA1 , SHERP , HASPB and HASPA2 gene constructs ) present as single gene copies . Only HASPA1/2 sKI and HA1/2+S2/HB sKI , which contained both HASPA1 and HASPA2 in the HASPA1/2 ( HA1/2 ) construct were expected to generate a value of 2 ( ±0 . 5 ) for HASPA . As shown in Fig 1C ( see also S5 Fig ) , the expected values were obtained for each target gene within the mutant lines . Two to three selected clones from each line were then passaged through BALB/c mice to restore parasite infectivity and stable , regulated gene construct expression ( which is known to diminish for the HASP and SHERP genes following prolonged parasite passage in culture [21] ) . Immunoblotting was used to verify stage-specific protein expression from the integrated HASP and SHERP genes , using whole lysates generated from low-passage parasites cultured over a 6 day time course from day 2 to day 7 post inoculum ( p . i . ) , in comparison with HASPB , HASPA and SHERP expression patterns in FVI ( Fig 1D ) . While only one clone per line is shown here , additional clones showed comparable results for each parasite line , respectively ( S6 Fig ) , as demonstrated in S7 Fig for HASPA2 sKI and HASPA1/2 sKI . Due to its small molecular mass and biophysical properties , SHERP is inherently difficult to transfer onto nitrocellulose , reducing blot quality . However , the results for HASPB and SHERP showed that integration of the HASPB and SHERP constructs into the cDNA16 locus was sufficient to reproduce the parental line ( FVI ) protein expression patterns . The unexpected differences in HASPA expression between HASPA1 sKI , HASPA1/2 sKI and HASPA2 sKI mutant lines are discussed further below . The constitutively expressed L . ( L . ) major N-myristoyltransferase protein ( NMT; [36] ) served as a loading control in all protein analyses . Selected mutants were further tested for any growth defects due to genetic manipulation . Low-passage parasites were inoculated at 105 parasites/ml into M199 medium and cell numbers monitored every 24 hr for 7 days . This growth assay was performed in 3–4 consecutive repeats , inoculating fresh M199 medium to a final concentration of 105 parasites/ml using the previous cultures at late log-phase ( day 3 p . i . ) . No significant fitness defect was observed in any of the mutant lines growing as in vitro promastigotes in comparison to the parental line ( FVI; Figs 1E and S8 ) . All selected clones reached stationary phase by day 3–4 p . i . , correlating with peak parasite density , followed by a similar rate of decline in live parasite density to day 7 p . i . with no statistically significant differences . Since HASPA1 and HASPA2 have identical ORFs , it is impossible to distinguish these two proteins by antibody probing in wild type parasites [26] . Generation of individual HASPA1 and HASPA2 replacement mutant lines in this study allowed analysis of protein expression from these two genes , previously only possible at the mRNA level [17 , 18] . The immunoblot data from promastigotes ( Figs 1D , S6 and S7 ) showed a similar HASPA expression in HASPA2 sKI as in FVI and cDNA16 sKI , peaking at day 7 p . i . when cultures were enriched in metacyclics . This correlated with the previously established HASPA2 mRNA expression pattern in FVI [17 , 18] , although the correlation was less clear for HASPA2-only combination mutants with HASPB and/or SHERP ( S6 Fig ) . In contrast , HASPA protein was not detectable in HASPA1 sKI and in HASPA1-only combination mutants cultured up to 7 days ( Figs 1D and S6 ) regardless of the clone tested , while low level mRNA expression had previously been detected in metacyclic cells [17 , 18] . Conversely , immunoblots of whole lysates of intracellular amastigotes purified from skin lesions of BALB/c mice showed expression of HASPA in the HASPA1 sKI amastigotes , but no detectable HASPA expression in the HASPA2 sKI amastigotes ( Fig 2E ) . Overall , this analysis has for the first time demonstrated that protein expression from the individual HASPA genes is stage-specific in L . ( L . ) major , with HASPA1 expressed in amastigotes and HASPA2 in promastigotes . Interestingly , the presence of both HASPA1 and HASPA2 in the same construct appeared to enhance and deregulate HASPA expression in HASPA1/2 sKI and HA1/2+S2/HB sKI promastigotes ( Fig 1D; and in other HASPA1-HASPA2 combination mutants with HASPB or SHERP; S6 Fig ) regardless of the clone tested ( S7 Fig ) , suggesting that either HASPA1 contributes to HASPA expression in early culture stages in the presence of HASPA2 or that the HASPA2 gene has lost parental line regulation in the HASPA1/2 construct ( Figs 1D and S6 ) . The latter explanation implicates cDNA16 locus structural constraints as important for correct HASPA regulation . This is a valid hypothesis since the HASPA1/2 construct was assembled with the same DNA fragments as the HASPA1 and HASPA2 constructs ( S1 Fig ) , the sequence was verified post construct assembly and the different HASPA expression patterns in HASPA1 sKI and HASPA2 sKI correlated with the previously established mRNA expression patterns [17 , 18] . However , this unexpected HASPA1/2 mutant phenotype did not result in an unexpected phenotype in vivo . Our previous work had shown that deletion of the full cDNA16 locus enhanced footpad lesion development in BALB/c mice following inoculation with cDNA16 dKO as compared to FVI promastigotes [31] . Due to our new observations on stage specific expression of HASPA1 and 2 in amastigotes and promastigotes , respectively , we wanted to revisit the original in vivo phenotype and address whether deletion and replacement of these genes affected footpad pathology . For that purpose , two clones from each of seven parasite lines ( FVI , cDNA16 dKO cDNA16 sKI , HASPA1 sKI , HASPA2 sKI , HASPA1/2 sKI and HA2+S2/HB sKI; refer to S9 Fig for clone ID ) were grown in vitro to late-stationary phase ( day 7 p . i . ) and assessed for the presence of metacyclic parasites . Given the lack of metacyclic-specific markers in L . ( L . ) major other than SHERP and HASPB , the focus of this study , the efficiency of metacyclogenesis in vitro was monitored by two complementary approaches: by morphometric analysis of fixed parasites ( Figs 2A and S10 ) , in comparison with existing sand fly data , and by metacyclic purification following peanut lectin agglutination ( PNA , Fig 2B ) . Morphometry showed that all mutant lines had generated metacyclics ( Figs 2A and S10 ) with FVI and HASPA1 sKI being the most efficient ( >70% ) and cDNA16 dKO and HASPA2 sKI the poorest ( <55% ) . Statistical significant differences occurred only between: FVI and cDNA16 dKO , P = 0 . 003; FVI and HASPA2 sKI , P = 0 . 003; HASPA1 sKI and cDNA16 dKO , P = 0 . 011; HASPA1 sKI and HASPA2 sKI , P = 0 . 014 . However , metacyclics from all cultured lines and clones were similar in size and morphology ( S10 Fig ) , suggesting no difference in metacyclic quality . Leptomonads were present at a similar percentage ( ~20% ) in all cultured parasites tested . Metacyclic enrichment by PNA is subject to some losses due to passive entrapment of the highly motile parasites in the lectin aggregates . Despite the observed reduction in metacyclic numbers for some lines as compared to morphometry ( Fig 2A and 2B ) , overall the agglutination test did not reveal any significant differences in the capacity of the different parasite lines to undergo metacyclogenesis ( Fig 2B ) , as verified by morphometric analysis ( S10 Fig ) . Comparing three replicates , FVI generated a mean of 55 . 8% , cDNA16 dKO 55 . 2% , cDNA16 sKI 57 . 7% , HASPA1 sKI 47 . 9% , HASPA1/2 sKI 51 . 5% , HASPA2 sKI 58 . 2% and HA2+S2/HB sKI 47 . 2% metacyclics in day 7 cultures . Based on these findings , the numbers of metacyclic parasites inoculated into the footpads of BALB/c mice were very similar for all clones tested . Given the demonstrated consistency of metacyclogenesis in these clones , and to allow direct comparison with previous observations , cultured promastigotes were harvested and injected at 3x107 parasites per BALB/c footpad without prior metacyclic purification . Disease progression for two independent clones per mutant line ( Figs 2C , S9a and S9c ) was tested by weekly footpad measurements followed by limiting dilution assay ( LDA ) of footpad homogenates once ~2 mm lesion diameter was reached ( Figs 2D , S9b and S9d ) . This analysis showed that both cDNA16 dKO clones were fast to develop severe lesions ( within 5 weeks p . i . ) , as previously observed [31] , although this was not due to elevated numbers of parasites within the lesions ( Figs 2C , 2D and S9 ) . The parental line ( FVI ) clones required ~8 weeks p . i . to reach the same level of lesion development , while cDNA16 sKI required ~6–7 weeks . Interestingly , both HASPA2 sKI clones showed a delayed onset of footpad lesion development compared to FVI , which was also repeatedly observed during routine passaging of parasites through BALB/c mice , using these two distinct clones . HASPA2 sKI-infected BALB/c mice took significantly longer ( >10 weeks p . i . ; P<0 . 001 ) to produce comparable footpad lesions to those observed in the other mutants . Conversely , lesion development in HASPA1 sKI and HASPA1/2 sKI infected footpads had a similar time course to cDNA16 sKI ( ~6–7 weeks ) . Lesion development in HA2+S2/HB sKI was comparable to FVI , suggesting that addition of a HASPB and/or SHERP copy restored FVI virulence levels . Further investigation will be required to resolve the mechanisms whereby the HASPs and SHERP contribute to disease outcome . Sádlová et al . [30] showed that stalling of metacyclogenesis due to cDNA16 locus deletion is only observed in the sand fly vector and not in culture . In this new study , each of the L . ( L . ) major mutant lines ( Tables 1 and S1 ) was fed independently at 106 early log-phase promastigotes/ml blood to P . ( P . ) papatasi and/or P . ( P . ) duboscqi , both L . ( L . ) major-specific vector species [8] . A total of 2 , 736 sand flies ( S2 Table ) were sampled at different time points ( day 2 , 5 , 9 and 12 PBM or at day 6 and 12 PBM only ) and infection loads , parasite localization and parasite morphology assessed ( Figs 3–5 and S11–S14 ) . Infection load data by microscopy ( Figs 3A and S11 ) revealed that P . ( P . ) duboscqi supports significantly higher ( P<0 . 001 ) parasite numbers in the midgut than P . ( P . ) papatasi . In P . ( P . ) papatasi , significant differences in parasite load were observed between FVI and the three mutant lines tested , cDNA16 dKO , HASPB sKI and SHERP sKI , after blood meal defecation ( day 5 , 9 and 12 PBM; P< 0 . 001; Figs 3A and S11 ) . While FVI , cDNA16 dKO and HASPB sKI showed significantly increased parasite loads from day 2 PBM to day 12 PBM ( P<0 , 001; P = 0 . 009; P = 0 . 032 , respectively ) , SHERP sKI survived comparatively poorly ( ~40% infected at day 12 PBM ) with significantly decreased persistence of infection from day 2 PBM to day 12 PBM ( P = 0 . 027 ) . Interestingly , SHERP sKI survival in P . ( P . ) duboscqi was not affected ( ~85% at day 12 PBM ) and developed as well as other mutant lines tested ( Figs 3A and S11 ) . In general , parasite lines survived well in P . ( P . ) duboscqi , showing significant increases in parasite numbers ( P≤0 . 002 ) from day 6 PBM to day 12 PBM ( S11 Fig ) . The microscopically evaluated parasite loads were verified for day 12 PBM samples by qPCR for 30 infected female sand flies per parasite line ( Figs 3B and S12 ) . Although microscopy tended to underestimate infection loads compared to qPCR , there was generally good correlation between light microscopic and qPCR data with the exception of cDNA16 dKO and SHERP sKI in P . ( P . ) papatasi , which showed significantly higher parasite loads by qPCR than by microscopy ( Figs 3B and S12 ) . However , while microscopic analysis only evaluates live parasites , qPCR does not discriminate between live and dead parasites containing genomic DNA . No significant differences were established between parasite lines tested in P . ( P . ) duboscqi at day 12 PBM by qPCR with the exception of FVI and HASPA1 sKI compared to HASPA2 sKI ( P = 0 . 01 and 0 . 004 , respectively ) and HA1/2+S2/HB sKI ( P = 0 . 03 and 0 . 01 , respectively ) . In parallel , parasite localization was assessed during the course of sand fly infection . Parasites were generally observed in the endoperitrophic space by day 2 PBM and within the midgut lumen by day 5/6 PBM after blood meal defecation ( Figs 4 , S13 & S3–S5 Tables ) . On rare occasions , blood meal remnants were present in the AMG and hindgut by day 5/6 PBM . Invasion of the TMG ( thoracic midgut ) was observed to varying degrees by day 5/6 PBM and this increased in frequency and intensity by day 9 and 12 PBM ( Figs 4 , S13 & S3–S5 Tables ) . Infections with FVI and cDNA16 sKI concentrated strongly in the TMG by day 9 and 12 PBM and these were accompanied by TMG distention . All other mutant lines tested did not show this distension effect , with parasites being either evenly spread from the cardia to the posterior of the AMG or largely found in the AMG by day 12 PBM . An in vitro assay was used to demonstrate that parasite osmotaxis was not significantly compromised in any of the mutant lines compared to FVI; this could , therefore , be excluded as a factor affecting parasite accumulation in the TMG ( Fig 5A ) . The observed TMG distention in late stage infection in FVI and cDNA16 sKI parasites only was indicative of the presence of PSG , an attachment matrix for leptomonads and nectomonads but proposed to be traversable by metacyclics ( communication by Matthew E . Rogers ) . Mutant line infections lacked this gel , as observed by light microscopy ( Fig 5B ) , suggesting compromised PSG generation in the mutant lines in vivo . To further investigate this observation , PSG was extracted from infected sand flies by pooling 10 infected midguts per infecting line . Samples were dot-blotted to activated nitrocellulose membranes and probed with the LT15 antibody that recognises L . major fPPGs , major components of the PSG [38] . PSG was only detected in FVI and cDNA16 sKI , but was undetectable in the other mutants tested . While parasites from all tested lines reached the cardia , significant differences were observed in the efficiency of SV colonization . FVI and cDNA16 sKI were the only lines tested to efficiently colonize the SV ( Figs 4 , 5B and 5C ) . While SHERP sKI and HASPB sKI were observed to attach at low numbers to the SV in P . ( P . ) papatasi ( 29 . 4% and 38 . 4% of analysed infected sand flies , respectively ) , mutant lines infecting P . ( P . ) duboscqi colonized the SV only weakly ( <5% of cases; Fig 4 ) , probably due to reduced haptomonad generation , the only parasite forms that attach to the SV . This hypothesis could not be verified due to the lack of any haptomonad specific markers . Parasite morphology was analysed on Giemsa-stained gut smears from day 5/6 , 9 and 12 PBM , using measurements of flagellum length , cell body length and width with separate analysis in AMG and TMG . This analysis showed significant differences in midgut metacyclogenesis between L . ( L . ) major lines at different time points ( Figs 6 and S14 ) . Only FVI and cDNA16 sKI produced metacyclics efficiently by day 12 PBM ( P<0 . 001 ) , compared to all other lines tested in the same vector species , although FVI metacyclic generation was significantly more efficient in P . ( P . ) duboscqi by day 12 PBM than in P . ( P . ) papatasi ( P<0 . 001 ) . These observations confirm that all other mutant lines tested could not complete metacyclogenesis in vivo , although they did so in vitro ( Fig 2A and 2B ) . In addition , metacyclics were preferentially present in the TMG , while leptomonads were equally represented in the AMG and TMG ( Figs 6 and S14 ) and nectomonads were preferentially found in the AMG . Thus all mutant lines , except cDNA16 sKI , resembled the cDNA16 null background with very few metacyclic-like parasites present at day 12 PBM and no gradient of differentiated parasites towards the TMG , correlating with the lack of parasite accumulation in the TMG and PSG-deficiency in these lines . Leptomonad generation in P . ( P . ) duboscqi showed no significant differences at day 12 PBM between all tested lines , except for HASPA1 sKI , which generated leptomonads very inefficiently ( P<0 . 001 , compared to all other lines tested in the same vector species ) . In P . ( P . ) papatasi , differences in leptomonad generation were only observed in SHERP sKI ( P<0 . 001 , compared to all other lines tested in the same vector species ) , which produced leptomonads as inefficiently as HASPA1 sKI in P . ( P . ) duboscqi . Differences in leptomonad generation between FVI and all other mutant lines by day 12 PBM were more pronounced in P . ( P . ) papatasi . Overall , no mutant line tested rescued the full parental line ( FVI ) phenotype , with the exception of cDNA16 sKI . To further characterize HASPB and SHERP expression in the newly generated mutant lines , parasites derived either from culture or from sand fly midguts were fixed , antibody labelled for HASPB or SHERP and analysed by confocal microscopy . Both proteins were clearly detected in metacyclics derived from cultured FVI , cDNA16 sKI , HASPB sKI and SHERP sKI , respectively , while cDNA16 dKO promastigotes did not show fluorescence , as expected ( Fig 7 ) . FVI and cDNA16 sKI recovered from sand fly midguts were positive for HASPB and SHERP expression , too . However , unlike in vitro , HASPB sKI and SHERP sKI from sand fly midguts produced no detectable HASPB or SHERP signal , respectively . These observations suggest that either the gene regulation observed in these mutants in vitro is not replicated in vivo or that parasite differentiation is compromised in vivo , leading to a loss of HASPB and SHERP expression . To investigate whether the lack of HASPB and SHERP expression was due to altered regulation at the protein or mRNA level , qRT-PCR was performed on midgut and culture-derived parasite mRNA . The HASPB mRNA levels of FVI and cDNA16 sKI , both having completed metacyclogenesis in vivo , showed elevated HASPB mRNA levels at day 6 PBM in vivo , subsequently declining towards day 12 PBM ( Fig 8 ) . Interestingly , HASPB sKI from midguts , which had not produced a detectable fluorescent HASPB signal ( Fig 7 ) , showed the same expression pattern , although total mRNA levels were lower than in FVI at all time-points , despite statistical compensation for the two HASPB copies in FVI compared to the one copy in HASPB sKI . In culture derived parasites , FVI and HASPB sKI showed similar expression patterns with peak expression of HASPB mRNA at day 7 p . i . , although initial mRNA levels were higher ( ~1 . 39-fold ) at day 3 p . i . in HASPB sKI compared to FVI . HASPB mRNA expression in cDNA16 sKI had already peaked at day 5 p . i . and declined by day 7 p . i . The downregulation of HASPB mRNA in vivo towards day 12 PBM was unexpected , since HASPB is expressed in metacyclics and amastigotes , and contradicts the in vitro observed mRNA expression pattern , which peaked at the final time-point ( day 7 p . i . ) , when metacyclics were at their densest ( Fig 8 ) . For HASPA mRNA analysis , the HASPA2 sKI mutant was used , given that HASPA2 is known to be upregulated in early log-phase growth [18] . In the case of HASPA , the expression patterns of cDNA16 sKI and HASPA2 sKI were comparable to FVI in vitro , but distinct from FVI in vivo , while similar between cDNA16 sKI and HASPA2 sKI ( Fig 8 ) . In the case of SHERP , cDNA16 sKI and SHERP sKI had similar mRNA expression patterns both in vitro and in vivo , but these were distinct from FVI in both conditions ( Fig 8 ) . Overall , these results suggest that dysregulation of gene expression is not responsible for the lack of HASPB and SHERP detection in sand fly derived parasites , since the single-replacement lines either resembled FVI and/or cDNA16 sKI , which both develop normally in vivo . Since it has previously been reported that culture conditions can influence promastigote development [39] , we investigated whether the 20% FCS supplemented M199 medium used for in vitro culture influenced HASPB and SHERP expression , as compared to growth in 5% sucrose , mimicking the sugar-rich plant sap that provides nutrients in the sand fly midgut after blood meal defecation in vivo . Parasites grown for 2 days in M199 were washed and suspended in 5% sucrose/PBS solution , prior to collection of protein samples every 24 hours of the growth cycle . Comparative immunoblots of the whole parasite lysates ( S15 Fig ) provided no evidence for downregulation of HASPB and SHERP in HASPB sKI and SHERP sKI , respectively , when grown in 5% sucrose conditions , thereby excluding culture conditions as the source of differential HASPB and SHERP expression . An alternative explanation for the observed differences in HASP and SHERP expression from replacement constructs in the sand fly could include a role for vector-derived regulatory factors . To further investigate this hypothesis , selected L . ( L . ) major lines were incubated either with ( + ) or without ( - ) homogenized midguts of uninfected blood-fed sand flies harvested at day 6 and 12 PBM . Parasite growth was monitored every 24 h and lysates harvested at day 6 p . i . were immunoblotted and analysed using ImageJ . This analysis showed that the addition of midgut homogenate affected parasite growth in culture ( Fig 9A ) . All tested lines grew more slowly with day 6 PBM midgut homogenate as compared to negative controls , failing to reach stationary phase by day 6 p . i . Conversely , parasite growth rates with day 12 PBM midgut homogenates were comparable to negative controls until day 3 p . i . when the negative controls reached stationary phase with subsequent decline in cell numbers . By contrast , the day 12 homogenate-supplemented parasite population continued to expand slowly until day 6 p . i . All tested lines showed a significant increase in growth when day 12 homogenates were compared to day 6 homogenates ( FVI: P = 0 . 027; cDNA16 dKO: P = 0 . 016; cDNA16 sKI: P = 0 . 013; HASPA2 sKI: P = 0 . 015 ) . FVI and cDNA16 sKI grown with day 6 PBM homogenate also showed a reduction of detectable HASPA and HASPB compared to the negative control ( FVI: 2 . 7-fold and 1 . 1-fold; cDNA16 sKI: 2 . 2-fold and 3 . 8-fold , respectively; Fig 9B and 9C ) . Cultures grown with day 12 midgut homogenate showed more limited reduction in HASPA and HASPB levels compared to the negative control ( FVI: 1 . 5-fold and 1 . 1-fold; cDNA16 sKI 1 . 4-fold and 1 . 2-fold , respectively ) , although these differences could be a consequence of the observed differences in parasite growth and potential slowing of metacyclogenesis . HASPA2 sKI and HASPB sKI were also tested in this way ( Fig 9D ) . Compared to FVI and cDNA16 sKI , HASPA2 sKI showed a similar response in cultures spiked with day 6 and 12 PBM midgut homogenates , while HASPB sKI showed increased HASPB expression in response to both midgut homogenates ( Day 6 PBM: 1 . 45-fold; Day 12 PBM 1 . 2-fold ) . Since completion of metacyclogenesis , PSG plug formation and SV degradation have been hypothesised to be essential for successful parasite transmission , we wanted to experimentally confirm the failure of our metacyclogenesis-impaired mutant lines to be transmitted in vivo . Due to the complexity of these experiments , we were limited to testing only a small subset of mutants: cDNA16 dKO and HASPB sKI ( both unable to complete metacyclogenesis , produce PSG or colonize the SV in P . ( P . ) duboscqi ) as representative lines hypothesised to be non-transmissible; and cDNA16 sKI and FVI ( competent for metacyclogenesis , PSG production and SV colonisation ) as parasite lines predicted to be successfully transmitted to a suitable host . These experiments were conducted at the National Institute of Health ( NIH ) , USA , using P . ( P . ) duboscqi , the same vector species used at the Charles University in Prague , CZ . Due to technical issues , the original FVI line was replaced with the FVI line available at the NIH ( FVI ( NIH ) ) ; both FVI lines are derived from the same original parent . Infection quality and metacyclogenesis progression in sand flies were monitored by parasite counting and morphometry , using light microscopy over a 14 day course PBM . FVI ( NIH ) presented similar infection loads ( Fig 10A ) as previously observed by qPCR for P . ( P . ) duboscqi in Prague ( Fig 3B ) . While showing on average lower infection loads compared to FVI , HASPB sKI also showed comparable results between previous qPCR data and dissection ( Figs 3B and 10A ) . cDNA16 dKO and cDNA16 sKI showed weaker infections compared to previous qPCR results ( Figs 3B and 10A ) . The frequency ( % ) of metacyclics per midgut parasite load was also assessed , showing that only FVI ( NIH ) and cDNA16 sKI produced metacyclics efficiently in the vector ( Fig 10B ) , as previously observed in the P . ( P . ) duboscqi colony from Prague ( Fig 6 ) , although FVI ( NIH ) was more efficient than cDNA16 sKI . For the transmission experiments , sand flies infected with FVI ( NIH ) , cDNA16 dKO , cDNA16 sKI or HASPB sKI were exposed on day 14 PBM to both ears of anaesthetised naïve mice . Genomic DNA was collected from the exposed ears and subjected to qPCR for parasite detection . As expected , the results showed successful parasite transmission for cDNA16 sKI and FVI ( NIH ) lines ( 37 . 5% and 27 . 3% , respectively , of analysed ear biopsies positive for parasite DNA ) , while cDNA16 dKO and HASPB sKI showed no evidence of transmission ( Fig 10C ) . Interestingly , cDNA16 sKI was more efficiently transmitted than the FVI ( NIH ) control , despite the higher parasite loads and metacyclic content detected on average in FVI ( NIH ) . This may relate to the higher virulence of cDNA16 sKI infections in mice ( Fig 2A and 2B ) . However , for successful transmission , the arbitrary nature of sand fly feeding under these experimental conditions must also be considered . Since feeding success does not correlate with transmission success , this variable is difficult to quantify [40] . Either way , previous use of the FVI ( NIH ) line in this setting suggested that transmission success is commonly higher and comparable to cDNA16 sKI . This study builds on our previous findings that the cDNA16 locus is essential for completion of metacyclogenesis in the sand fly midgut and that episomal expression of a HASPB gene copy can partially rescue the parental phenotype [30] . Here , we aimed to determine whether metacyclogenesis requires one gene from the cDNA16 locus , a subset of the HASP and/or SHERP genes , or the whole cDNA16 locus under parental gene regulation signals . Our in depth analysis has demonstrated a more complex picture . The mRNA data reported here suggest that correctly regulated HASPB expression is the important event in metacyclogenesis in L . ( L . ) major , with cDNA16 sKI showing a similar expression pattern as FVI in vivo ( Fig 8 ) . This interpretation is further supported by the observed differences in HASPA and SHERP expression between the FVI and mutant lines , including cDNA16 sKI , suggesting that these genes are less likely to be required for completion of metacyclogenesis . The high expression of HASPB protein in L . ( L . ) major metacyclics [23] and the observation by Sádlová et al . ( 2010 ) that unregulated episomal expression of HASPB promoted metacyclogenesis completion further support this hypothesis [30] . Our data cannot directly prove that HASPB is the key player in metacyclogenesis in vivo , however , as the HASP and SHERP genes re-integrated alone into the cDNA16 locus are not expressed in the sand fly midgut . This is in contrast to the in vitro data collected during characterization of the same mutant lines , which show convincingly that the HASP and SHERP constructs are expressed and stage-regulated as in wild type parasites . Interestingly , replacement of the whole cDNA16 locus into its former location does recover the parental line phenotype in vivo , including detectable HASPB and SHERP product expression ( Fig 7 ) , while the step-by-step replacement of HASP and SHERP genes does not ( see HA1/2+S2/HB sKI ) , suggesting that either the order and genomic context of the HASP and SHERP genes are vital to their correct regulation in vivo but not in vitro , or that trans-factors present in vivo play a role in parasite gene regulation in the sand fly . A confounding factor here is the observed weak correlation between mRNA and protein abundance in Leishmania species [41] . In trypanosomes , mRNAs can be stored in cytoplasmic RNA granules until translation , potentially accounting for this discrepancy [42] . Recently , formation of mRNA granules has also been described for Leishmania under stress conditions , such as starvation [43] , previously demonstrated to be a trigger for metacyclogenesis in vitro [6 , 7] . In the case of stage-regulated genes in Leishmania , as in other protozoan species [44] , mRNA up-regulation can occur at a developmental stage prior to protein expression [41] . Although not definitively shown in Leishmania , it is feasible that these mRNAs are stored in RNA granules prior to entry into the translational machinery . Such sequestration could accelerate adaptational responses in dynamically changing environments , such as in the sand fly midgut , thereby promoting parasite survival . It is not known whether HASP and SHERP mRNAs are stored in RNA granules nor what signals might operate in vivo to initiate their subsequent translation . However , one testable hypothesis might propose that HASP and SHERP mRNAs are indeed sequestered cyoplasmically and “bypassed” in vitro due to a lack of midgut-specific signals in culture , thereby explaining the observed differences in HASP and SHERP mRNA and product expression in vitro and in vivo . The above observations suggest that HASP and SHERP function could be related to cellular mechanisms required for survival and development under sand fly midgut conditions . In this dynamic environment , alterations in pH , temperature , amino acid and digestive enzyme content occur during parasite differentiation , changes that are only partially reproduced under culture conditions . The importance of in vivo passaging to restore virulence in L . ( L . ) major and other Leishmania spp . has been previously demonstrated [40 , 45 , 46] . Prolonged in vitro culture frequently causes parasite avirulence [47] , which can be partially restored by repeated passaging through a susceptible mammalian host or sand fly vector [45] . Similar effects have been reported in other parasitic systems: culture-adapted Trypanosoma brucei bloodstream parasites , for example , have a 1 , 000x lower antigen switching rate for their variable surface glycoproteins ( VSG ) than in natural isolates [48 , 49] . Interestingly , the addition of crude uninfected blood-fed midgut homogenates to cultured Leishmania promastigotes in this current study indicated potential regulatory and temporal influences on parasite growth and HASPA and HASPB expression in the period following the blood meal time ( Fig 9 ) . Considering that the midgut is subject to dynamic changes over the course of Leishmania infection , reflecting the transition from blood meal to sugar digestion , our data suggest that Leishmania could utilise changes in the midgut content to drive its gene regulation and subsequent proliferation and differentiation . Well-established examples of this type of parasite/host interaction can be found in other kinetoplastid species such as T . brucei , during differentiation into stumpy forms [50 , 51] and trypomastigotes [52–54] and during gametocytogenesis in Plasmodium spp . [55] . It is also possible that differences in mutant gene expression are chromosomal context dependent . The observation that the LmjcDNA16 sKI line is the only one of the range of mutants generated to express HASPB and SHERP in vivo ( Fig 7 ) and rescue metacyclogenesis ( Figs 4 and 6 ) , as described above , suggests that the order and context in which the HASP and SHERP genes occur in the cDNA16 locus are relevant for correct in vivo regulation . Accurate polycistronic RNA processing for the production of mature mRNAs in Leishmania requires the correct positioning of downstream gene 5’-splice acceptor sites relative to the upstream gene polyadenylation sites [56] . Clearly our constructs , with their flanking sequences and DHFR-flanked antibiotic-resistance genes , are expressed appropriately in vitro but not in vivo , suggesting that other factors are important in regulating expression from our constructs in the sand fly midgut . While further investigation into HASP and SHERP function is highly desirable , the lack of HASP and SHERP mutant phenotypes in vitro hampers further rapid investigation due to the complexity of parasite manipulation in the sand fly midgut and the limiting amounts of biological material recoverable from infected sand flies . Sádlová et al . [30] showed that cDNA16 dKO parasites grown in vitro secrete fPPG , constituent of the PSG , into the culture medium . By contrast here , only FVI and cDNA16 sKI secrete detectable amounts of fPPG in the sand fly midgut ( Fig 5C ) leading to formation of PSG that is detectable by light microscopy ( Fig 5B ) . While further analysis using a panel of relevant antibody probes would be desirable , our current data suggest that PSG is not produced in the cDNA16 mutant lines in vivo . It is unclear why fPPG secretion and PSG formation should be impaired in vivo in the sand fly midgut but not in vitro , and how the HASPs and SHERP contribute to this process . The PPG synthetic pathway , while not completely described in Leishmania , involves secretion via the ER and Golgi in other eukaryotic cells [57] , two cellular compartments that the HASPs and SHERP do not appear to enter . Rather , N-myristoylated HASPs are palmitoylated on the cytosolic face of the Golgi , prior to transport to the plasma membrane [20] , while SHERP associates intracellularly as a peripheral membrane protein [27] , interacting in vitro with a sub-unit of a vacuolar type H ( + ) -ATPase that functions in acidification [28] . There is no evidence to suggest that SHERP affects secretion via the ER/Golgi route , but it cannot be excluded that this small protein influences fPPG synthesis and/or secretion indirectly by its interactions with intracellular membrane constituents [27] . Alternatively , the lack of fPPG secretion could be a consequence of incomplete metacyclogenesis in the HASP and SHERP mutants preventing maturation of fPPG-secreting leptomonads . This seems unlikely , given that leptomonad forms were observed by morphometry in comparable numbers in HA1/2+S2/HB sKI , FVI and cDNA16 sKI , while PSG formation in the TMG was only observed in FVI and cDNA16 sKI infections . Since there are no molecular markers for mature leptomonads , it cannot be excluded that the parasite forms observed by morphometry were functionally immature , however . Further investigation is required to determine the relationship between HASP and SHERP deletion and the lack of PSG formation . PSG has been shown to play a role in TMG colonization in the sand fly [13 , 58] , which could explain why several mutant lines lacking fPPG failed to establish mature infections and to accumulate in the TMG , in particular , in P . ( P . ) papatasi . The PSG plug forces sand flies to regurgitate prior to blood meal intake , supporting parasite transmission into the host skin [38] . Once in the skin , glycans donated by the fPPGs promote the recruitment of neutrophils and macrophages [38] , prior to macrophage invasion [59 , 60] . Further , PSG induces the alternative activation of macrophages , promoting arginase-1 activation and antagonising nitric oxide synthase 2 ( NOS2 ) , thereby facilitating parasite survival [60] . Therefore , a lack of PSG may impair sand fly transmission and reduce the likelihood of parasite survival in the mammalian host . While we were able to show impaired transmission in our PSG-deficient mutant lines in vivo , the same lines cultured in vitro were all able to establish persistent infection in BALB/c mice following sub-cutaneous high dose needle infection of 107 late-stage parasites . Interestingly , several PSG-deficient lines , in particular cDNA16 dKO , HASPA1 sKI and HASPA1/2 sKI , were more virulent in BALB/c mice than FVI , while other lines were significantly less virulent ( e . g . HASPA2 sKI ) . However , it must be noted that in vitro , these parasite lines were able to complete metacyclogenesis and to secrete fPPG [30] . In vivo-derived parasites from these PSG-deficient lines show a lack of metacyclic generation , suggesting that parasites would not be infective , even if they were transmitted from the sand fly to a mammalian host . Since our data showed that our mutant lines are not transmissible , with the exception of cDNA16 sKI , this question could not be addressed in a natural transmission scenario . Even so , our results confirmed experimentally that completion of metacyclogenesis , PSG secretion and SV colonization , which are all hallmarks of mature parasite infections in sand flies , are essential for successful parasite transmission , as is the presence of the cDNA 16 gene locus . On artificial challenge in BALB/c mice with our mutant lines , significant differences were observed in pathology between mutant lines . As previously observed by McKean et al . ( 2001 ) , the full cDNA16 locus deletion mutant ( cDNA16 dKO ) caused faster footpad swelling than the parental line ( FVI; Fig 2C ) . Interestingly , the phenotype of the full cDNA16 locus replacement line ( cDNA16 sKI ) was intermediate , suggesting gene dose dependency of the observed phenotype . Since HASPA1 sKI has a similar phenotype to cDNA16 sKI , while HA2+S2/HB sKI has a similar phenotype to FVI , this suggests that the cDNA16 sKI intermediate phenotype is HASPA1 gene dependent . Although not directly demonstrated , these data suggest that the deletion of the cDNA16 locus products increases inflammatory responses ( either directly or by pleiotropic effects ) , a phenotype moderated by re-introduction of one copy of the locus containing half the wild type copy number of HASP and SHERP genes . This hypothesis is supported by the observation in our previous study [31] that an episomal cDNA16 replacement mutant , overexpressing HASP and SHERP , was avirulent . While HASPA1 sKI showed a similar phenotype as cDNA16 sKI , the HASPA2 sKI mutant line showed a significantly delayed onset of lesion formation , despite equivalent inoculum and in vitro capacity for metacyclic formation ( Fig 2A and 2B ) . This observation was of interest because HASPA2 protein expression is promastigote-specific , while HASPA1 protein expression is amastigote-specific ( Figs 1D and 2E ) , correlating with the mRNA expression data for these genes [17] . Introducing a HASPA1 copy with a HASPA2 copy ( HASPA1/2 sKI ) caused a similar phenotype as HASPA1 sKI ( Fig 2C ) . Interestingly , the parasite burden per footpad at 2 mm swelling was similar for all tested lines; only the time point when these were reached varied between the lines . These observations could indicate functional differences in the HASPA proteins during the parasite life cycle , potentially affecting initial metacyclic parasite survival post inoculation or amastigote proliferation rather than disease progression per se; a surprising conclusion given their identical protein sequences . Based on the results presented here , we confirm a role for both the HASP and SHERP proteins in Leishmania metacyclogenesis in the sand fly , leading to parasite transmission , and propose a subsequent role for the HASP proteins in establishing infection in the mammalian host . We also show for the first time that sand fly midgut extracts collected post-blood meal affect parasite behaviour in vitro over time as the midgut luminal content changes . While biochemical fractionation and in depth analysis of the midgut lysates are now required to advance these studies , they may offer a novel approach to simulate in vivo phenotypes in vitro while also affirming the importance of vector-parasite interactions in vivo . L . ( L . ) major Friedlin V1 ( MHOM/IL/81/Friedlin/VI; FVI ) was used as a parental line/wild type in this study , as for all previous work on this locus . The previously described L . ( L . ) major 4 . 8 cDNA16 double deletion ( cDNA16 dKO; ΔcDNA16::HYG/ΔcDNA16::PAC; [31] ) and cDNA16 single replacement lines ( cDNA16 sKI; ΔcDNA16::HYG/ΔcDNA16::PAC/ΔPAC::cDNA16+ NEO [30] ) were used as controls . New L . ( L . ) major HASP and SHERP replacement lines were generated by homologous recombination with newly synthesized linear DNA constructs into the former cDNA16 locus within the null or other mutant backgrounds , using the nucleofection kit ( Amaxa ) according to the suppliers’ guidelines . The DNA constructs contained one or two HASP and/or SHERP gene ( s ) with their native 5’ and 3’ UTRs and a selectable antibiotic resistance marker gene ( either NEO or BSD ) . Correct genomic integration of DNA constructs was verified by Southern blot and quantitative PCR ( qPCR ) followed by parasite passage through BALB/c mice ( described below ) . Promastigotes of all L . ( L . ) major lines were routinely cultured in Medium 199 ( M199 ) ( supplemented with 20% v/v heat-inactivated foetal calf serum , 1% v/v penicillin-streptomycin ) as described [31] . For parasite in vitro differentiation , parasites grown until late log-phase in M199 were harvested , washed and suspended in 5% sucrose/PBS ( described below ) . Metacyclic purification by peanut agglutination ( PNA; Sigma ) was originally described elsewhere [61] . We used 50 μg/ml PNA for 15 min . at room temperature with regular agitation of parasite suspensions to separate agglutinated promastigote forms from metacyclics by slow centrifugation ( 175xg ) . Metacyclic ratios in day 7 cultures were established by pre- and post-agglutination counts on a haemocytometer . The protocol for Southern analysis has been described elsewhere [30] . Genomic DNA ( gDNA ) was extracted using DNeasy Blood & Tissue columns ( Qiagen ) , SacI-digested , separated by 0 . 8% agarose gel electrophoresis and blotted onto positively charged nitrocellulose membranes . Digoxigenin ( DIG ) labelled probes were used for detection with the DIG-development system ( Roche ) . Whole parasite lysates in Laemmli buffer ( 20 ml 0 . 5 M Tris-HCl [pH 6 . 8] , 3 . 08 mg DTT , 40 ml SDS [10%] , 50 mg Bromophenol Blue , 20 ml Glycerol [100%] and sterile Milli-Q water ( MQH2O ) to 100 ml ) were separated by SDS-PAGE as described [31] and blots analysed using affinity-purified polyclonal rabbit antibodies against SHERP [27] , HASPB ( 336; [31] ) or non-affinity-purified polyclonal HASP antibodies for HASPA-detection . A secondary anti-rabbit HRP antibody ( Sigma ) was used with ECL Prime ( Amersham ) for detection . A polyclonal antibody against L . ( L . ) major N-myristoyltransferase ( NMT; [36] ) was used as a loading control . In case of the growth condition assay , band intensity on immunoblots were analysed using ImageJ . Values were normalized first for each gene individually ( HASPA2 , HASPB and NMT , respectively ) against negative control to compensate variations within the image . Then normalized HASPA and HASPB values were normalized against the corresponding NMT loading control to adjust for loading variations prior to calculating the fold-difference between ( + ) and ( - ) conditions . PSG was extracted from 10 infected midguts pooled into 50 μl PBS , as described [60] . Whole lysates of debris pellets and 6x spun PSG containing supernatants were blotted on to activated nitrocellulose membranes using an adapted protocol [60] . Dot blots were probed for PSG with the LT15 antibody against phosphoglycan disaccharide repeats [PO4-6Galb1-4Mana1-]x [38] and treated as for the immunoblots using an anti-mouse HRP secondary antibody ( Sigma ) . The osmotaxis assay was adapted from Leslie et al . [62] and Oliveira et al . [63] . Briefly , plain glass capillary tubes ( 75 mm length , 0 . 8 inner/1 mm outer diameter ) were filled with wash and incubation ( WIS ) buffer ( 30 mM β-glycerophosphate disodium salt , 87 mM NaCl , 27 mM KCl , 2 mM CaCl2 , 2 mM MgCl2 , 0 . 004% enriched Bovine Serum Albumin [pH 7 . 1] ) containing 1% agarose ± 100 mM of sucrose , leaving exactly 1 cm void ( ~5 μl ) . Once set , the void was filled with WIS buffer . The glass capillaries were equilibrated in WIS buffer for ~30 min . at room temperature on a rocking table . Parasites were grown to late log-phase/early stationary-phase in M199 , harvested , washed twice in WIS buffer and suspended to a final concentration of ~2 . 5x107 cells/ml in WIS buffer . The equilibrated glass capillary tubes were dipped into the parasite suspension at a slight angle ( 6 capillary tubes with 100 mM sucrose and 6 without sucrose were used per strain ) and incubated at 26°C for 1 h . The WIS buffer in the capillary void was removed , mixed with 1% formaldehyde in saline solution and applied to a haemocytometer for parasite counting . At the Charles University , Prague , sand flies were maintained at 26°C and high humidity ( 75% ) on 50% sucrose solution and a 14 h light/10 h dark photoperiod as described [64] . Sand fly infections and analysis were carried out as described by Sádlová et al . [30] . For infections , colony-bred Phlebotomus ( Phlebotomus ) papatasi and P . ( P . ) duboscqi ( Turkey and Senegal strains , respectively; reared at the Department of Parasitology , Charles University , Prague ) were fed on infected ( 106 parasites/ml ) heat-inactivated rabbit blood through a chick-skin membrane for up to 2 h in the dark . Unfed sand flies were separated from engorged females , which were sampled at day 2 , 5 , 9 and 12 post blood meal ( PBM ) or at day 6 and 12 PBM only . Light microscopic analysis of dissected midguts was used to establish parasite localization ( endoperitrophic space , AMG to cardia , attached to the stomodeal valve either weakly or strongly ) and parasite load per midgut for all dissection points; scoring adapted from Myšková et al . [65] as light ( <100 parasites/gut ) , moderate ( 100–1000 parasites/gut ) , heavy ( >1000 parasites/gut ) and very heavy ( >>1000 parasites/gut ) . Dissected midguts were scored by two independent researchers . In addition , 30 infected female sand flies from day 12 PBM were individually frozen at -20°C in buffer for DNA extraction and parasite loads were scored by qPCR as described [30] . Sand fly infection experiments were repeated at least three times per tested line . At the National Institute of Health ( NIH ) , Rockville , USA , sand flies were maintained in similar conditions . For infection , colony-bred 2- to 4-day-old P . ( P . ) duboscqi females ( Mali strain; reared at the Laboratory of Malaria and Vector Research , NIAID ) were infected by artificial feeding through a chick skin membrane on defibrinated rabbit blood ( Spring Valley Laboratories , Sykesville , MD ) containing 350 units/ml penicillin , 350 μg/ml streptomycin , and 3–4 x 106 procyclics or amastigotes/ml ( P1-P5 ) from four L . major lines ( FVI , cDNA16 dKO , cDNA16 sKI and HASPB sKI ) initially isolated from BALB/c footpad lesions . After 3 h of feeding in the dark , fully blood-fed sand flies were separated . For the initial experiment , the total number of parasites and percent metacyclics per midgut were established at different days after infection ( D2 , D7 , D9 and D14; data on for D14 shown in Fig 10 ) for each group . Thereafter , in two independent repeats , the total number of parasites and the percentage of metacyclics were determined on the day of transmission . After dissection , each midgut was placed in 50 μl of PBS in a microcentrifuge tube , macerated with a plastic pestle ( Kimble Chase ) and parasites counted using a haemocytometer; metacyclic forms were distinguished by morphology and movement [66] . Quantitative PCR was performed on genomic DNA samples of selected new mutant clones to establish integrated copy number using Fast SYBR Master Mix ( Applied Biolscience ) in the OneStep qPCR system ( Life technologies ) with the OneStep software v . 2 . 2 . 2 according to the supplier’s guidelines . Target genes were detected with gene-specific primers and results were normalized against the Na/H antiporter-like protein gene ( see S6 Table ) . The protocol for detection of parasite load is described elsewhere [30 , 65] . Primers specific for kinetoplast minicircle DNA were used for parasite detection as described by Mary et al . [67] ( S6 Table ) . Parasites were grown for 3 days in 10 ml complete M199 at 26°C , pelleted and washed twice in sterile PBS at room temperature . Parasites resuspended in 10 ml 5% sucrose/PBS were incubated for an additional 4 days at 26°C . 107 parasites were collected every 24 h and lysed in Laemmli buffer for Western analysis . Sand flies were fed on heat inactivated uninfected rabbit blood through a chick-skin membrane for up to 2 h in the dark . 50 blood fed midguts were dissected at day 6 and 12 PBM into 200 μL M199 + Amikin ( 250 μg/ml ) + penicillin ( 60 μg/ml ) + fluorocytosin ( 1 . 5 mg/ml ) . The midguts were homogenised and filter through a 0 . 22 μm filter spinning column ( Ultrafree—MC , GV dutapore ) . The midgut extract was added to 4 ml of M199 and 1 ml was aliquoted into culture tubes . M199 was chosen as the medium of choice given its high serum content , which reduced the impact of additional protein introduced via the midgut homogenate . FVI , LmjcDNA16 dKO , LmjcDNA16 sKI and LmjHASPB sKI were inoculated into the 1 ml medium , respectively , to a final concentration of 105 parasites/ml and left to grow for 6 days . Parasite density was established by parasites counting on a haemocytometer . At day 6 p . i . , parasites were pelleted and washed twice in PBS before lysis in 50 μl 1x Laemmli buffer at 95°C for 10 min . As described previously [30] , midguts were divided into AMG and TMG , smeared on to glass slides , fixed with 100% methanol and stained with Giemsa . 160 randomly chosen parasites were imaged ( using a 100x oil-immersion objective , Olympus BX51 fluorescent microscope , Olympus DP70 camera ) per midgut section per day per strain on at least three randomly chosen gut smear slides . For cultured parasites , cell pellets were washed twice in saline solution ( 0 . 9% NaCl ) , applied to polylysine slides for 15 min . and the excess suspension tipped off . Glass slides were air dried , fixed with 100% methanol , rinsed with water , stained with Giemsa and analysed by microscopy . 100 randomly chosen parasites per slide were imaged ( using a 63x oil-immersion objective , Zeiss Axioplan microscope , Optronics 60800 camera system ) . For immunofluorescence analysis , cultured parasites were harvested , washed , fixed in 4% formaldehyde and applied to polylysine glass slides as described [21 , 68] . Parasites derived from sand flies were analysed in 100% methanol fixed gut smears as above . Cells were permeabilized with 0 . 2% Triton-X 100/PBS and blocked with Image-iT FX signal enhancer ( Invitrogen ) prior to analysis of HASPB or SHERP expression using polyclonal anti-HASPB ( 336; [32] ) and anti-SHERP [27] antibodies , respectively , followed by Alexa Fluor 488 Dye ( Invitrogen ) secondary antibody . Samples were mounted either in Vectashield or Mowviol with 4' , 6-diamidino-2-phenylindole ( DAPI; Vector ) and imaged with Zeiss LSM 510 or 710 META confocal microscopes . Flagellum and cell body length and width were measured with Image J [69] . Parasites were classified into four groups adapted from Walters ( 1993 ) and Ciháková and Volf ( 1997 ) : ( i ) procyclics < 14 μm body length ≥ 4 μm , flagellum shorter than cell body; ( ii ) nectomonads: body length ≥ 14 μm; ( iii ) leptomonads: body length < 14 μm and flagella length < 2 times body length and ( iv ) metacyclics: body length < 14 μm and flagella length ≥ 2 times body length . Paramastigotes and haptomonads were not considered . In vivo , haptomonad presence or absence was determined by the state of SV colonization . Experiments routinely used 6–8 weeks old BALB/c mice ( Harlan Laboratories , UK ) . For footpad infections , 3x 107 late-stage stationary parasites were injected in 30 μl PBS ( adapted from Depledge et al . [70] ) . For routine parasite passage to re-establish virulence , amastigotes were harvested from draining lymph nodes 8–10 weeks post footpad infection and inoculated into M199 for incubation at 26°C . To establish parasite virulence , footpads of groups of 5 female BALB/c mice were infected , ears were marked and mice placed randomly in cages . Mice were chosen blindly for footpad measurement once a week until footpad lesion reached 2 mm , at which point they were sacrificed . Parasite burdens per footpad were established after mouse-scarification by a limiting dilution assay ( LDA ) of infected footpads , adapted from Titus et al . [71] and Lima et al . [72] and calculated by the online Leishmania LDA analysis tool available on the Imperial College London homepage: ( http://wwwf . imperial . ac . uk/theoreticalimmunology/llda/ ) [73] . Amastigotes were isolated using a protocol adapted from Paape and Aebischer ( 2011 ) [74] . BALB/c mice were infected at the base of the tail on both sides by needle inoculation of 3x107 late stationary parasites . After lesions developed ( 8–10 weeks p . i . ) , mice were sacrificed and lesion material excised with a scalpel , weighed , forced through a 70 μm cell strainer into homogenization buffer ( 20 mM HEPES-KOH , pH 7 . 3 , 0 . 25 M sucrose supplemented with cOmplete Mini proteinase inhibitor cocktail [Roche] ) and washed once in homogenization buffer . Amastigotes were released by forcing the cell suspension through a 25-gauge needle . Nuclei were removed by centrifugation at 100xg for 2 min . The supernatants were loaded onto a discontinuous sucrose gradient: 20 , 40 , and 60% ( w/w ) sucrose in HEPES saline ( 30 mM HEPES-KOH , pH 7 . 3 , 0 . 1 M NaCl , 0 . 5 mM CaCl2 , 0 . 5 mM MgCl2 ) [75] , centrifuged for 25 min . at 700xg . Amastigotes were isolated from the 40/60% sucrose interface , diluted in PBS and washed once in PBS . Experiments at the NIH were performed using 6- to 8-week-old BALB/c mice ( Charles River Laboratories Inc . ) , maintained under pathogen-free conditions . To assess the capacity of mutants to transmit to BALB/c mice , 10–20 ( as available ) infected P . ( P . ) duboscqi females from each group ( FVI , cDNA16 dKO , cDNA16 sKI and HASPB sKI ) on day 14 PDM were exposed to each mouse ear in the first experiment . Thereafter , validation of transmission was carried out for the groups where mature infections containing metacyclics were observed ( FVI and cDNA16 sKI ) . Mice were anesthetized intraperitoneally with a mixture of ketamine ( 100 mg/kg ) and xylazine ( 10 mg/kg ) . Infected sand flies were placed in vials with a meshed surface and applied to the ears using custom-made clamps . The flies were allowed to feed for two hours in the dark . Mouse ears were removed two hours after exposure to infected flies and frozen at -70°C until processed . Total genomic DNA was extracted using the DNeasy tissue kit following the manufacturer’s protocol ( Qiagen ) . A total of 50 ng of sample DNA was amplified in triplicate by real time PCR ( Biorad c1000 thermal cycler and cfx96 real time system ) using primers JW11 and JW12 [76] together with a Ld3C6 fluorescent probe [77] targeting kinetoplast minicircle DNA . Parasite numbers were determined by the real time system software based on a standard curve of serially diluted L . major ( WT Friedlin V1 ) DNA . The cut-off was based on values obtained for naïve DNA controls . Animal experiments in York were approved by the University of York Animal Procedures and Ethics Committee and performed under UK Home Office license ( ‘Immunity and Immunopathology of Leishmaniasis’ Ref # PPL 60/4377 ) . All animal experimental procedures at the NIH were reviewed and approved by the National Institute of Allergy and Infectious Diseases ( NIAID ) Animal Care and Use Committee under animal protocol LMVR4E . The NIAID DIR Animal Care and Use Program complies with the Guide for the Care and Use of Laboratory Animals and with the NIH Office of Animal Care and Use and Animal Research Advisory Committee guidelines . All statistical analysis was done with the SPSS software v . 20-22 ( IBM ) or GraphPad Prism 5 . P<0 . 05 was considered to be significant . Categorical data were analysed by χ2 test . Normally distributed continuous data were analysed by one-way ANOVA and post-hoc Tukey multiple comparison test . Non-normally distributed data were normalized by log10 or square root transformation , as appropriate , and submitted to parametric analysis where possible . Otherwise , non-normally distributed data were analysed by Kruskal-Wallis and post-hoc Dunn’s multiple comparison test . Growth assay and footpad lesion data were analysed by repeat measure ANOVA and post-hoc Tukey multiple comparison test or by Freidman test and post-hoc Dunn’s multiple comparison test .	Millions of people around the world are at risk of infection with single-celled Leishmania parasites that cause a wide range of infectious diseases of the immune system , the leishmaniases . There is no effective vaccine for these infections while available drugs have toxic side-effects and resistance is an increasing problem . Human infection occurs through the bite of infected blood-feeding sand flies . Leishmania parasites live in the sand fly gut , where they complete a complex series of developmental changes to become infective to mammals . The parasites also modify the insect gut to promote their own transmission . Little is known about the molecular regulation of these processes . Recently , we showed that deletion of a small group of related genes from the parasite prevented completion of its development in the sand fly , suggesting a role for these sequences in transmission to the host . This study clarifies the expression pattern of these genes during parasite development and shows that the observed stalling of development is accompanied by changes in parasite-sand fly gut interactions and a loss of parasite transmission . These target genes also influence disease development in the mammalian host , confirming critical roles for their encoded proteins throughout the parasite life cycle .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "medicine", "and", "health", "sciences", "microbiology", "sand", "flies", "cloning", "parasitic", "diseases", "parasitic", "protozoans", "protozoan", "life", "cycles", "developmental", "biology", "protozoans", "leishmania", "molecular", "biology", "techniques", "insect", "vectors", "promastigotes", "research", "and", "analysis", "methods", "epidemiology", "life", "cycles", "molecular", "biology", "genetic", "loci", "disease", "vectors", "amastigotes", "leishmania", "major", "genetics", "biology", "and", "life", "sciences", "protozoology", "organisms" ]	2017	Leishmania HASP and SHERP Genes Are Required for In Vivo Differentiation, Parasite Transmission and Virulence Attenuation in the Host
Antimonials are still being used for visceral leishmaniasis ( VL ) treatment among HIV co-infected patients in East-Africa due to the shortage of alternative safer drugs like liposomal amphotericin B . Besides tolerability , emergence of resistance to antimonials is a major concern . This study was aimed at assessing the clinical outcome of VL-HIV co-infected patients when treated with sodium stibogluconate ( SSG ) . Retrospective patient record analysis of VL-HIV co-infected patients treated at a clinical trial site in north-west Ethiopia was done . Patients with parasitologically confirmed VL and HIV co-infection treated with SSG were included . The dose of SSG used was 20 mg Sb5 ( pentavalent antimony ) /kg and maximum of 850 mg Sb5 for 30 days . The clinical outcomes were defined based on the tissue aspiration results as cure or failure , and additionally the safety and mortality rates were computed . The study included 57 patients treated with SSG and by the end of treatment only 43 . 9% of patients were cured . The parasitological treatment failure and the case fatality rate were 31 . 6% and 14 . 0% respectively . SSG was discontinued temporarily or permanently for 12 ( 21 . 1% ) cases due to safety issues . High baseline parasite load ( graded more than 4+ ) was significantly associated with treatment failure ( odds ratio = 8 . 9 , 95% confidence interval = . 5-51 . 7 ) . SSG is not only unsafe , but also has low effectiveness for VL-HIV patients . Safe and effective alternative medications are very urgently needed . Drug sensitivity surveillance should be introduced in the region . Visceral leishmaniasis ( VL ) , also called kala-azar , is a vector-borne disseminated infection caused by the Leishmania donovani spp . complex , a protozoan parasite that predominantly affects tissue macrophages . Overt disease is lethal without treatment . The zoonotic form , with dogs as the main reservoir , is caused by Leishmania infantum and is found mainly in the Mediterranean basin and Latin America . The anthroponotic form is caused by Leishmania donovani and is prevalent in the Indian subcontinent ( with estimated 300 , 000 cases/year ) and East Africa ( 30 , 000 cases/year ) . Within East Africa , Sudan and Ethiopia are most affected [1] . Like most other neglected tropical diseases , VL has traditionally ranked low on the national and international health agenda [2]–[5] . HIV has been identified as one of the emerging challenges for VL control . HIV infection dramatically increases the risk of progression from asymptomatic Leishmania infection to VL and VL accelerates HIV disease progression [6] , [7] . At the global level , the highest burden of VL-HIV co-infection is found in north-west Ethiopia , where up to 40% of VL patients can be co-infected with HIV [8] . Treatment of HIV co-infected individuals poses particular challenges with poor treatment response and recurrent relapse [9]–[12] . For over four decades , antimonials have been in use and still are the mainstay of therapy in East-Africa for immunocompetent individuals either alone or in combination with paromomycin [13] . Several studies demonstrated more than 90% effectiveness among VL patients without HIV [11] , [14]–[17] . However , following reports of high rates of toxicity of antimonials in HIV-infected individuals [18] , liposomal amphotericin B has been recommended as the preferential first line treatment for this patient group [13] , [19] . Although the cost of liposomal amphotericin B has undergone substantial reductions for low income countries , there remain issues of availability of this drug in the public sector in several resource-constrained settings in East-Africa [20] , [21] . In Ethiopia , the country with the highest co-infection rate globally , liposomal amphotericin B is in short supply . As a result , clinicians often still have to resort to antimonials for treating co-infected individuals , reserving the available liposomal amphotericin B for the most complicated or severe cases . Besides poor tolerability , emergence of drug resistance is a major concern with antimonials , as has been witnessed in Bihar , India [22] . Whereas initial studies on antimonials reported relatively low rates of treatment failure in HIV-infected individuals in East-Africa [17] , more recent retrospective cohort studies conducted in Ethiopia seemed to indicate relatively high failure rates [11] , [14] . However , these reports came from settings where parasitological confirmation of diagnosis and treatment response was not systematically performed . Moreover , with the roll-out of Anti-Retroviral Therapy ( ART ) in East-Africa and better survival of HIV-infected patients , the VL patient population has gradually changed , with an increasing proportion of HIV patients presenting with VL relapse [23] . Data on efficacy and tolerability of antimonials in this patient population are very scarce . Hereby , we report on the clinical outcomes in a cohort of adult VL-HIV co-infected patients recruited at the Leishmaniasis Research and Treatment Center ( LRTC ) of University of Gondar ( UoG ) Hospital , Ethiopia . Ethical approval was obtained for this study from the institutional review board of the University of Gondar . All data were analyzed anonymized . The study was conducted at the LRTC in the UoG that is located in north-west Ethiopia close to the main VL endemic focus in the country . The LRTC was founded by the Drugs for Neglected Diseases initiative ( DNDi ) , and is now part of the governmental health system and is used as a clinical trial site . In addition to VL research , the center provides free VL treatment and care to all patients with leishmaniasis . Patients present to the center either spontaneously or are referred from other health institutes in the catchment area . Several international aid agencies including the World Health Organization ( WHO ) , Médecins Sans Frontières ( MSF ) and DNDi support VL treatment and care within Ethiopia , but shortage of anti-leishmanial drugs remains a frequent problem in clinical practice . Pentavalent antimonials and liposomal forms of amphotericin B , and more recently paromomycin , are the main drugs to treat VL . The combination regimen SSG and paromomycin was started as first line therapy for non-HIV VL cases at the site in September 2012 . While liposomal amphotericin B is recommended for VL in HIV patients , due to the inadequate supply , it is often reserved for more severe cases such as patients with organ dysfunction . Miltefosine is infrequently available . As a result , most HIV co-infected VL patients are being treated with SSG . In November 2011 the LRTC initiated a clinical trial on the use of secondary prophylaxis to prevent relapse in HIV co-infected VL patients that is currently ongoing ( http://clinicaltrials . gov/show/NCT01360762 ) . All VL-HIV co-infected patients presenting at LRTC are screened for enrolment in this clinical trial and their findings and initial treatment responses are documented in individual patient record files . We reviewed these patient records and considered those who were initially started on treatment with antimonials between November 2011 and January 2013 as eligible for the present retrospective cohort study on safety and effectiveness of antimonials . A VL diagnosis was suspected based on the WHO clinical case definition: prolonged fever , weight loss and splenomegaly in a patient from an endemic area or with a travel history . In patients who met this case definition we confirmed the VL diagnosis by microscopic detection of the parasite in tissue aspirates ( spleen , bone marrow , lymph node ) . The parasite load was graded as 6+: >100 parasites per field; 5+: 10–100 parasites per field; 4+: 1–10 parasites per field; 3+: 1–10 parasites per 10 fields; 2+: 1–10 parasites per 100 fields; 1+: 1–10 parasites per 1000 fields and 0: 0 parasites per 1000 fields as viewed with a 100x oil-immersion lens [13] . Sodium stibogluconate ( SSG ) ( at a dose of 20 mg Sb5 ( pentavalent antimony ) /kg for 30 days ) was used for the initial treatment of these patients . The maximum SSG dose used was 850 mg Sb5/day , to avoid the toxicity related to high doses in HIV infected patients [24] , [25] . All patients were admitted to the LRTC and monitored for their treatment response using clinical and laboratory parameters . The main clinical parameters monitored during treatment were fever , appetite , fatigue , weight and spleen size . In addition , the patients were checked daily for new complaints especially related to adverse reactions . Blood chemistry and hematology tests were done weekly and a test of cure ( ToC ) by tissue aspiration and microscopic evaluation for parasites was done systematically at the end of SSG treatment ( day 30 ) . Electrocardiography was done based on the presence of cardiac symptoms . Treatment was extended for patients with a positive ToC at the end of SSG treatment . Those who i . showed a clinical response , ii . had a parasite load reduction of two log or more ( “slow responders” ) and iii . a parasite grade below 4 received a treatment extension with the same drug , while those who had less than two logs parasite reduction ( “non-responders” ) or a parasite grade of 4 or above were treated with an alternative drug ( most commonly liposomal amphotericin B with or without miltefosine ) . While this was the principle followed , the shortage of medications often affected the practice . Depending on the parasite grade ( high or low ) , ToC was repeated on day 15 or day 30 of the treatment extension with further extension of treatment when ToC was positive . At LRTC provider-initiated testing and counselling for HIV is offered to all hospitalized patients . The HIV diagnosis is based on the national algorithm with two serial positive rapid test results; KHB ( Shanghai Kehua Bio-engineering Co-Ltd , Shanghai , China ) followed by STAT-PAK™ ( Chembio HIV1/2 , Medford , New York , USA ) . In case of discrepancy between the two tests , Uni-Gold™ ( Trinity Biotech PLC , Bray , Ireland ) is used as a tie breaker . As VL is considered a stage IV-defining illness in HIV patients [13] , [19] , all patients are given ART as soon as they were stabilized from their acute illnesses . ART regimens follow the national guidelines: tenofovir-lamivudine-efavirenz; zidovudine-lamivudine-efavirenz; or zidovudine-lamivudine-nevirapine [26] . Second-line ART consists of protease inhibitor-based combination regimens . Initial cure is defined as a negative ToC at the end of the standard treatment [13] . A positive ToC indicates treatment failure , which consists of two types: slow-response or non-response . Clinical improvement ( resolution of symptoms and signs such as fever , improvement in appetite , weight gain , gaining physical strength and regression of the spleen ) with a reduction in parasite load by more than two logs without complete parasite clearance is defined as slow-response . If the parasite load remains the same or if the reduction is less than two logs , this is defined as non-response . SSG discontinuation as an outcome refers to permanent discontinuation of SSG due to intolerance . Besides the initial treatment outcomes ( at the end of the standard 30 days SSG treatment ) , treatment outcomes were also ascertained at the end of the entire treatment course ( end of treatment outcomes ) , integrating eventual SSG extensions , or treatment changes for reasons of SSG intolerance or initial treatment failure . Adverse reactions that required discontinuation ( temporary or permanent ) of the treatment were considered serious adverse drug reactions and included in the analysis . Reasons for interruption included clinical or biochemical pancreatitis; acute renal injury; or other severe conditions that were considered probably related to SSG like cardiotoxicity or worsening of bone marrow function . Clinical pancreatitis was defined as persistent vomiting , abdominal pain and raised serum amylase levels . Biochemical pancreatitis was defined as asymptomatic grade 4 serum amylase levels ( using the Common Terminology Criteria for Adverse Events ( CTCAE v 4 . 0 ) system [27] . Acute renal injury requiring SSG interruption was defined as acute increase of serum creatinine to more than 2 mg/dl ( normal values 0 . 6 to 1 . 1 mg/dl ) . While some patients with toxicity improved after a temporary interruption ( less than five days ) and were able to continue with SSG , those with poor/slow recovery were shifted to other treatment regimens . We used a structured data collection format to extract information from the individual patient chart records . Socio-demographic data , weight , episode of leishmania , CD4 count , hematology and blood chemistry values and treatment outcome data were recorded . Count data were summarized as frequency ( % ) , and numerical variables as median values with Inter-Quartile Range ( IQR ) . To assess associations we computed Crude Odds Ratios ( OR ) with a 95% Confidence Interval ( CI ) . For individuals with either initial cure or failure , the medians of within patient changes during treatment in laboratory and clinical parameters were compared using the Wilcoxon rank-sum test . P-values<0 . 05 were considered statistically significant . Statistical analysis was done using STATA 11 software . From a total of 84 HIV co-infected VL patients treated at LRTC during the study period , 57 received SSG as first line regimen . The others were treated with liposomal amphotericin B ( 6 patients ) , liposomal amphotericin B plus miltefosine ( 7 patients ) or the SSG and paromomycin ( 14 patients ) combination regimen , see Figure 1 . Of the 57 patients treated with SSG , 56 ( 98 . 2% ) were male; the median age was 32 ( IQR 28–36 ) years . While 33 ( 57 . 9% ) were admitted for primary VL , 24 ( 42 . 1% ) were relapses , and out of them 12 ( 50% ) had more than two previous VL episodes . The majority were malnourished , with huge spleen and anemia ( Table 1 ) . Most patients ( n = 36; 63 . 2% ) were already on ART at the time of VL diagnosis . The median CD4 count was 61 ( IQR 35–101 ) cells/µL , and 39 ( 73 . 6% ) of them had a CD4 count less than 100 cells/µl . Most patients ( n = 48; 90 . 6% ) had a CD4 count less than 200 cells/µl . The leishmania parasite load in tissue aspirates was graded as +4 or above for 41/56 ( 73 . 2% ) individuals ( Table 1 ) . Amongst the 57 patients starting antimonial treatment , SSG was permanently discontinued due to intolerance in five ( 8 . 8% ) cases . In addition , seven ( 12 . 3% ) patients temporarily interrupted SSG ( for a maximum of five days ) due to adverse reactions but re-instituted after stabilization . At the end of the standard SSG treatment course , 25 ( 43 . 9% ) achieved initial cure while eight ( 14 . 0% ) patients died during SSG therapy . One patient left the hospital against medical advice . There were 18 ( 31 . 6% ) cases with initial treatment failure ( positive ToC ) . Ten of these were slow responders while the eight others were non-responders . For the 18 cases with parasitologically confirmed SSG treatment failure , treatment was extended beyond one month . A repeat course of SSG was used for 14 of these patients ( for an additional 15 to 90 days ) . The other 4 were given liposomal amphotericin B ( n = 2 ) or a liposomal amphotericin B and miltefosine combination ( n = 2 ) . For the five cases that permanently discontinued SSG due to intolerance , liposomal amphotericin B was given and four of them eventually got cured . At the global assessment made at the end of drug treatment , 43 of the 57 ( 75 . 4% ) patients had achieved parasite clearance and were considered cured , three ( 5 . 3% ) had treatment failure and nine ( 15 . 8% ) died . The outcome was unknown for two ( 3 . 5% ) individuals . Serious adverse reactions were observed in 14 of the 57 patients ( 12 in the first month and 2 more during extension of treatment ) mainly due to pancreatitis ( n = 3 ) , renal failure ( n = 3 ) or both ( n = 6 ) . Cardiotoxicity and the combination of hepatitis and bone marrow suppression were observed in individual cases . Out of the nine deaths , five occurred after developing both pancreatitis and renal failure , and one after acute renal failure . The other three deaths were related to the presence of additional co-morbidities ( malnutrition , sepsis and tuberculosis ) . Table 2 shows the key clinical and laboratory parameters in relation with treatment response . While tissue parasite grading decreased more significantly in cured patients , there was a more pronounced increase in hematocrit and a greater reduction in spleen size in failure cases . Although VL relapse cases and those on ART were more likely to fail , only high tissue parasite grading was significantly associated with treatment failure ( OR = 8 . 9 , CI = 1 . 5–51 . 7 ) , Table 3 and Figure 2 ) . Individuals with low CD4 counts ( <100 cells/µl ) , malnourished patients ( BMI <18 . 5 kg/m2 ) , patients presenting with primary VL , patients having a large spleen size ( ≥10 cm ) on admission and a higher tissue parasite load ( ≥4 ) were more likely to die , but these differences did not reach statistical difference ( Table 3 ) . This study showed high toxicity and very low effectiveness of SSG when used as a primary treatment option for VL-HIV co-infection . A 30-days drug course of SSG led to treatment failure in close to one out of three patients , an assessment based on parasitological data . SSG-related toxicity was common , probably contributing to death in six cases ( i . e . 75% of the deaths ) . These findings indicate the urgent need of wider availability of alternative and safer drugs such as liposomal amphotericin B for this specific population . Previous studies from the region have reported initial failure rates of SSG in the range of 2 . 3% to 14 . 1% [11] , [14] , [16] . Several factors might have contributed to the higher frequency observed in this study . In contrast to most other studies , proportionally more patients on ART and/or ( multiple ) VL relapse cases , identified as risk factors for treatment failure , were included [23] . High baseline tissue parasite grade – the strongest risk factor in our study – was also relatively common . Third , the parasitological response was not systematically assessed in the other studies , which could have led to under-diagnosis of failure . Moreover , a maximum dose of 850 mg SSG was used in this treatment center for this group of patients due to the experiences of exacerbated toxicity with higher doses in other regions with a similar patient population [25] . Whether parasite drug resistance , as observed in India [22] , played a role is currently unclear . Although the change in parasite grading was lower in failing patients compared to cured ones , the difference was not statistically significant . Drug sensitivity testing of the isolated parasites is currently underway . Safety of SSG has always been a concern . However , limited studies addressed the experience in the treatment setting concerning the adverse effects of SSG in the region . Vomiting has been reported as a common adverse event in most of the studies ranging from 8% to 38% [14] , [16] , [17] , though the underlying cause of the vomiting is not clearly addressed . A quarter of the patients in this study were suffering from adverse reactions that required either temporary or permanent discontinuation of the SSG . Pancreatitis and renal failure were found to be the serious adverse reactions that were probably the main causes of death necessitating the need to close monitoring during treatment . This implies that SSG should only be used in a setting where monitoring of renal functions and pancreatitis is possible . As electrocardiographic monitoring was not systematically done , the cardiotoxicity might be under estimated . The case fatality rate during treatment ( 14% ) is in the previously observed range , ( 6 . 8%–33 . 3% ) [11] , [28] . The risk factors for mortality identified in our study are in line with the previous observations . The strengths of this study were that it was conducted in a dedicated leishmania treatment and research center where tissue aspiration is routinely performed in HIV co-infection , and standardized treatment protocols and data collection tools are in place . Limitations include the relatively small sample size , and the missing information for some laboratory tests . Moreover , HIV-1 viral load testing and electrolytes measurements could not be conducted . The small sample size in this study did not allow for a thorough study of risk factors for treatment failure , and did preclude any control for confounding in multivariable analysis . We observed that patients with high parasite load tended to respond slowly or not . Extension of treatment ( using antimonials or liposomal amphotericin B ) beyond 30 days helped to increase the cure rate from 44% to 75% . Patients who received prolonged duration of treatment seem to have tolerated SSG . These few patients were also with better clinical situations ( higher hemoglobin and better spleen regression ) which can be a source of bias . High SSG toxicity was observed with the daily maximal dose limited to 850 mg . It should be noted that higher dose and prolonged therapy could be at the expense of increasing toxicity . Higher daily dose of antimonials should be discouraged for HIV co-infected patients given the high toxicity observed at the current dose . Case-by-case decisions on dose and duration of therapy need to take into consideration the parasite load at the time of diagnosis and safety issues . The small sample size in this study did not allow for a thorough study of risk factors for treatment failure , and did preclude any control for confounding in multivariable analysis . We observed that patients with high parasite load tended to respond slowly or not . Extension of treatment ( using antimonials or liposomal amphotericin B ) beyond 30 days helped to increase the cure rate from 44% to 75% with minimal additional toxicity . Given the small sample size , the safety of extending SSG in individuals tolerating the first month of SSG remains to be confirmed . On the other hand , high SSG toxicity was observed with the daily maximal dose limited to 850 mg . Consequently , a higher daily dose of antimonials should in general be discouraged for HIV co-infected patients . Case-by-case decisions on dose and duration of therapy need to take into consideration the parasite load at the time of diagnosis and safety issues . Patients with VL relapse and those developing VL while on ART tended to fail SSG treatment . With further expansion and access to ART in this region , such difficult to treat patients might gradually become more prevalent . Additional studies focusing on treatment in this patient group should be conducted . Despite ART , these patients had profound immune deficiency as seen from their CD4 cell level . Additionally , most patients failed SSG treatment while on ART showing the need for additional treatment strategies . Early screening and treatment or primary prophylaxis are possible options yet to be explored . The reason behind the better response in hematocrit and spleen regression in the treatment failure group may be related to the longer treatment period . Patients with better hematocrit may survive longer than the severely anemic ones , and face repeated relapses . This study re-confirms that SSG is not safe in patients with HIV co-infection and additionally shows that its effectiveness is low - and potentially declining - in the region [29] . However , despite the recommendations in the national and international guidelines , SSG remains in use due to shortage of alternative safe and effective medications . Even with the price reductions and donation programs of liposomal amphotericin B , there is still a limited and irregular supply in Ethiopia . We call on all stakeholders to urgently take measures to ensure a stable access to liposomal amphotericin B for all high risk groups – including HIV co-infected patients – in which treatment with antimonials leads to unacceptable high rates of toxicity and/or failure . A clinical trial evaluating high dose liposomal amphotericin B or a combination with miltefosine in HIV co-infection is about to be initiated in Ethiopia . If found effective , ensuring the availability of these drugs should be a priority . This study provides evidence of high treatment failure ( failure of parasite clearance ) and toxicity in HIV co-infected VL patients treated with SSG . Assuring availability of safer and more efficacious treatment options for HIV co-infected VL patients in this region needs urgent attention . Furthermore , the need for drug resistance surveillance is called upon .	The co-infection of VL and HIV is a very challenging clinical problem especially in the East-Africa region . Though liposomal amphotericin B is the recommended treatment option for VL-HIV co-infection , it is often not available in practice in Ethiopia . Thus several patients are still being treated with antimonials that are infamous for their toxicity . In this study , we describe the results of such antimonial treatment in a series of 57 patients . The effectiveness of antimonials was found to be low and , in comparison to previous studies , declining . There is an urgent need to assure the availability of safer and more effective alternative medications for VL-HIV . It seems also wise to start a surveillance scheme for drug susceptibility in leishmania parasites as our results may relate to emerging antimonial resistance in the East-Africa region .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "infectious", "diseases", "veterinary", "diseases", "zoonoses", "medicine", "and", "health", "sciences", "leishmaniasis", "neglected", "tropical", "diseases", "biology", "and", "life", "sciences", "tropical", "diseases", "veterinary", "science" ]	2014	High Parasitological Failure Rate of Visceral Leishmaniasis to Sodium Stibogluconate among HIV Co-infected Adults in Ethiopia
Gliotoxin , and other related molecules , are encoded by multi-gene clusters and biosynthesized by fungi using non-ribosomal biosynthetic mechanisms . Almost universally described in terms of its toxicity towards mammalian cells , gliotoxin has come to be considered as a component of the virulence arsenal of Aspergillus fumigatus . Here we show that deletion of a single gene , gliT , in the gliotoxin biosynthetic cluster of two A . fumigatus strains , rendered the organism highly sensitive to exogenous gliotoxin and completely disrupted gliotoxin secretion . Addition of glutathione to both A . fumigatus ΔgliT strains relieved gliotoxin inhibition . Moreover , expression of gliT appears to be independently regulated compared to all other cluster components and is up-regulated by exogenous gliotoxin presence , at both the transcript and protein level . Upon gliotoxin exposure , gliT is also expressed in A . fumigatus ΔgliZ , which cannot express any other genes in the gliotoxin biosynthetic cluster , indicating that gliT is primarily responsible for protecting this strain against exogenous gliotoxin . GliT exhibits a gliotoxin reductase activity up to 9 µM gliotoxin and appears to prevent irreversible depletion of intracellular glutathione stores by reduction of the oxidized form of gliotoxin . Cross-species resistance to exogenous gliotoxin is acquired by A . nidulans and Saccharomyces cerevisiae , respectively , when transformed with gliT . We hypothesise that the primary role of gliotoxin may be as an antioxidant and that in addition to GliT functionality , gliotoxin secretion may be a component of an auto-protective mechanism , deployed by A . fumigatus to protect itself against this potent biomolecule . Gliotoxin , which has a molecular mass of 326 Da and is an epipolythiodioxopiperazine ( ETP ) , contains a disulphide bridge of unknown origin and has been shown to play a significant role in enabling the virulence of Aspergillus fumigatus [1]–[3] . The cytotoxic activity of gliotoxin is generally mediated by direct inactivation of essential protein thiols [4] and by inhibition of the respiratory burst in neutrophils by disrupting NADPH oxidase assembly , thereby facilitating in vivo fungal dissemination [5] , [6] . The enzymatic machinery responsible for gliotoxin biosynthesis , and metabolism , is encoded by a multi-gene cluster in A . fumigatus which is coordinately expressed during gliotoxin biosynthesis [7] , [8] . This cluster encodes gliP , a bimodular nonribosomal peptide synthetase ( NRPS ) which has been conclusively shown to be responsible for the biosynthesis of a Phe-Ser dipeptide , a gliotoxin precursor , by gene disruption ( ΔgliP mutant ) [9]–[12] . In fact , disruption of gliP within the gliotoxin biosynthetic cluster has resulted in the effective inhibition of all cluster gene expression in a ΔgliP mutant [9] . A putative transporter , encoded by gliA , has been shown to facilitate gliotoxin efflux , and increased tolerance to exogenous gliotoxin , when expressed in Leptosphaeria maculans [13] . sirA is a gliA ortholog in this organism and L . maculans ΔsirA was more sensitive to exogenous gliotoxin and sirodesmin than wild-type , however restoration of sirA in the mutant led to greater tolerance towards these metabolites [13] . Bok et al . [14] have demonstrated that disruption of a fungal Zn ( II ) 2-Cys ( 6 ) binuclear cluster domain transcription factor ( gliZ ) results in the complete inhibition of all gliotoxin cluster gene expression and effective diminution of gliotoxin production [14] . Although GliP has been shown to activate and condense L-Phe and L-Ser to form a precursor diketopiperazine moiety , no information relating to subsequent modification ( e . g . , thiolation ) is available [9]–[12] , [15] and it is also unclear if A . fumigatus might need to protect itself against potential gliotoxin cytotoxicity [13] . Interestingly , addition of gliotoxin ( up to 5 µg/ml ) to A . fumigatus ΔgliP resulted in the up-regulation of selected gene expression ( gliI , J , T and N ) within the gli cluster and Cramer et al . [9] noted complete activation of the gene cluster ( except gliP ) following gliotoxin exposure ( 20 µg/ml ) . However , exposure of wild-type A . fumigatus Af293 to gliotoxin ( 20 µg/ml ) , for 24 h , did not result in any significant alteration in gliotoxin cluster expression [9] . The biological significance of these observations is unclear , apart from implying a role for gliotoxin in the regulation of the gli cluster in the absence of gliotoxin production . It has recently been demonstrated that gliotoxin and sporidesmin , also an ETP toxin containing a disulphide bridge , are both substrates and inactivators of glutaredoxin ( Grx1 ) [16] . These authors also confirmed that the intact disulphide form of these ETP moieties was essential for Grx1 inactivation and that prior reduction of sporidesmin , using glutathione , prevented subsequent Grx1 inactivation . Oxygen presence was also required for Grx1 inactivation by sporidesmin and mass spectrometric analysis confirmed the formation of mixed disulphides between one molecule of Grx1 and either gliotoxin or sporidesmin , respectively . Combined , these data suggest interplay between oxygen availability and selective protein inactivation in the presence of oxidised ETP-type molecules . This indirectly suggests either a protective , or neutral , involvement of the oxidised forms of gliotoxin or sporidesmin in protecting against the deleterious effects of oxygen by selective protein inactivation . In mammalian cells it has been demonstrated that the oxidized form of gliotoxin is actively concentrated in a glutathione-dependent manner and that it then exists within the cell almost exclusively in the reduced form [17] . As glutathione levels fall due to apoptosis , the oxidized form of gliotoxin effluxes from the cell where the cytocidal effects of gliotoxin are perpetuated in a pseudocatalytic manner . Conversely , it has been shown that gliotoxin may substitute for 2-cys peroxiredoxin activity in HeLa cells by accepting electrons from NADPH via the thioredoxin reductase–thioredoxin redox system to reduce H2O2 to H2O . In this way , nanomolar levels of gliotoxin may actually protect against intracellular oxidative stress [18] . Although the cytotoxic effects of gliotoxin on mammalian cells have been extensively investigated , and yeast have been deployed as a model system to study this interaction [19] , no direct investigation of any self-protective mechanism used by A . fumigatus against this intriguing molecule has been undertaken . Here , we demonstrate that deletion of gliT results in transformants which cannot grow in the presence of even modest levels of exogenous gliotoxin and that exogenous gliotoxin up-regulates gene expression within the gliotoxin cluster , especially that of gliT . We propose that GliT is the key cellular defence against gliotoxin in A . fumigatus and that this finding yields a new selection marker system for detecting transformation . ΔgliT mutants were generated by transformation of A . fumigatus strains ATCC46645 and ATCC26933 , respectively , as described in Materials and Methods , using the bipartite marker technique and pyrithiamine selection , with modifications [20] , [21] ( Figure S1 ) . Deposition number: IMI CC 396691 ( CABI Bioscience Centre , Egham , Surrey , UK ) . These two strains were chosen because ATCC26933 is a gliotoxin producer , whereas ATCC46645 lacks significant gliotoxin production using the Minimal Media described in Materials and Methods ( see below ) . Complementation of gliT mutant strains was carried out as described in Materials and Methods and Figure S1 ) . Complemented strains ( gliTC ) ( Deposition number: IMI CC 396692 ) exhibited wild-type like features in all subsequent experiments , demonstrating that the occurrence of a single ectopic integration of a gliT fragment is insignificant in the A . fumigatus ATCC26933 background . ΔgliT protoplasts grew and regenerated mycelia perfectly in the absence of gliotoxin ( Figure 1A ) . The ΔgliT strain grew at identical rates to wild-type ( data not shown ) . However , ΔgliT protoplasts were unable to grow in the presence of gliotoxin ( 10 µg/ml ) ( Figure 1A ) whereas exogenous gliotoxin had no effect on wild-type growth . Subsequent phenotypic analysis of A . fumigatus ATCC46645 , ATCC26933 , and respective ΔgliT conidia ( ΔgliT46645 and ΔgliT26933 ) demonstrated that gliotoxin ( 5 µg/ml ) significantly inhibited ΔgliT growth on minimal medium and completely inhibited ΔgliT growth on both AMM and Sabouraud medium ( gliotoxin , 10 µg/ml ) ( Figure 1B & C; p<0 . 0001 and Figure S2 ) . Moreover , germination rates of ΔgliT strains were comparable to those of wild-type A . fumigatus , even in the presence of gliotoxin up to 10 µg/ml . These results clearly indicated that ΔgliT was highly sensitive to exogenous gliotoxin . Consequently , ΔgliT46645 and ΔgliT26933 mutant complementation was carried out by introducing gliT only ( no antibiotic resistance gene ) to complement ΔgliT with selection in the presence of gliotoxin ( 10 µg/ml ) . Transformants , which had recovered resistance to exogenous gliotoxin , were confirmed by Southern analysis to have an intact and functional copy of gliT present ( Figure S1 ) . This result confirms that gliT confers resistance to gliotoxin in A . fumigatus and that ΔgliT mutants have significant potential for future functional genomic studies involving A . fumigatus since gene deletions in this strain are selectable by gliT reintroduction , with selection in the presence of gliotoxin . Remarkably , addition of reduced glutathione ( GSH; 20 mM ) ) to test plates completely abolished the cytotoxic effects of exogenous gliotoxin which indicated that gliT loss resulted in depletion of intracellular GSH , when exposed to gliotoxin , or that only the oxidized form of gliotoxin is imported into A . fumigatus ( Figure 1B & C ) . Prior reduction of gliotoxin , using 50 mM NaBH4 , resulted in a statistically significant inhibitory effect of gliotoxin on growth of ΔgliT26933 ( p<0 . 05 ) ( Figure 1B & C ) . NaBH4 was selected as reductant as it avoided complications associated with the introduction of additional thiols , or GSH , and the formation of gliotoxin conjugates , which may have resulted from GSH , DTT or β-mercaptoethanol-mediated reduction . It was also observed that GSH presence ( 8 mM ) partially alleviated the growth inhibitory effects of gliotoxin ( with or without prior reduction; p<0 . 01 and p<0 . 005 , respectively ) ( Figure 1C ) . However , wild-type levels of growth were only achieved in the presence of 20 mM GSH ( Figure 1B ) . The enhanced GSH-mediated alleviation of gliotoxin-induced cytostatic effects observed in ΔgliT , strongly suggest that depletion of intracellular glutathione may be a consequence of gliT loss . GSH-mediated relief of A . fumigatus ΔgliT growth inhibition , by exogenous NaBH4-reduced gliotoxin , indicates that intracellular GSH depletion plays a role in the inhibitory effect of gliotoxin- and not that GSH is merely acting to reduce exogenously added gliotoxin and prevent uptake ( Figure 1B & C ) . Exogenous gliotoxin or reduced gliotoxin had no effect on growth of ΔgliZ and gliZc ( gliZ complemented strain ) [14] ( kind gifts from Professor Nancy Keller , University of Wisconsin-Madison ) and an identical pattern was observed in the presence of GSH ( data not shown ) . Moreover , A . fumigatus ΔgliT did not exhibit any phenotype when exposed to either H2O2 or phleomycin ( data not shown ) . A . fumigatus gliTC strains were resistant to exogenous gliotoxin ( Figure 1D ) . gliZ , A and G encode the gliotoxin cluster transcription factor , transporter and a putative glutathione s-transferase ( generally a detoxification enzyme ) , respectively , and all are conceivably involved in protection against gliotoxin toxicity [3] , [8] , [22] . Northern analysis showed that expression of these 3 genes plus gliT , from the gliotoxin gene cluster , was induced in A . fumigatus ATCC46645 within 3 h following gliotoxin ( 5 µg/ml ) addition at 21 h ( Figure 2A ) . No gliT expression was detectable in ΔgliT whereas the expression of all other genes was identical to the wild-type , including the continued absence of expression at 24 h in the absence of added gliotoxin ( Figure 2A ) . Expression of gliT was restored in pyrithiamine-resistant A . fumigatus gliTC derived from both ATCC46645 and ATCC26933 strain backgrounds , which unambiguously confirms restoration of gliT expression in complemented strains ( Figure 2B ) . Moreover , gliT expression was inducible by addition of gliotoxin ( 5 µg/ml ) , as had been observed in both wild-type strains , thereby convincingly demonstrating that the wild-type phenotype had been entirely restored ( Figure 2B ) . As noted above , no significant growth inhibition of A . fumigatus ΔgliZ in particular , or gliZc , was observed in the presence of gliotoxin or reduced gliotoxin ( Figure 1C ) . These observations further confirm the minimal role played by any other component of the gli gene cluster in protection against gliotoxin presence since gliZ absence results in complete cluster attenuation [14] . Significantly , Northern analysis confirmed gliotoxin-induced gliT expression in ΔgliZ , which indicates the independent regulation of gliT with respect to other gli cluster components , such as gliG and gliA which are not expressed by A . fumigatus ΔgliZ following exposure to gliotoxin ( Figure 2C ) . These observations are in complete accordance with proteomic data which demonstrated a threefold up-regulation of GliT expression ( 33% sequence coverage ) in A . fumigatus ATCC26933 , and the absence of detection of any other gli cluster component , following 3 h exposure to exogenous gliotoxin ( 14 µg/ml ) ( Figure 2D and Figure S3 ) . Sequence analysis of the 5′ and 3′ regions adjacent to the original gliT locus in A . fumigatus ΔgliT26933 confirmed that gliF was intact but revealed two mutations ( C23R and E160G ) in the open reading frame of a gene ( AFUA_6G09745; identified as a conserved hypothetical protein at http://www . cadre-genomes . org . uk ( but here termed gliH ) , located 3′ with respect to the gliT locus . Although expression of gliF and gliH was confirmed by RT-PCR in A . fumigatus ΔgliT26933 ( Figure 2E ) , there was concern that the altered sequence of gliH may have resulted in a mutant enzyme , which could possibly have also contributed to gliotoxin sensitivity in ΔgliT26933 . However , A . fumigatus ΔgliH26933 grew in the presence of gliotoxin ( 10 µg/ml ) ( Figure 1D ) which completely eliminated the possibility that this gene , located adjacent to gliT in the A . fumigatus genome , contributed to gliotoxin resistance and established , beyond question , the key role of gliT in mediating resistance to exogenous gliotoxin . A . fumigatus gliHC ( Figure S1 ) was also resistant to exogenous gliotoxin , as expected ( Figure 1D ) . Gliotoxin ( 580 ng/ml ) was detectable in organic extracts from A . fumigatus ATCC26933 but not ΔgliT26933 cultures , grown under identical conditions , by RP-HPLC and LC-MS analysis ( Figure 3 ) . Gliotoxin production was recovered in A . fumigatus ATCC26933 gliTC ( Figure S4 ) Interestingly , ΔgliT26933 exhibited an identical phenotype to ΔgliT46645 which was generated from A . fumigatus ATCC46645 , yet gliotoxin production was undetectable , under the culture conditions employed , in both A . fumigatus ATCC46645 and ΔgliT46645 indicating that sensitivity to exogenous gliotoxin is not associated with a de novo gliotoxin biosynthetic capacity . A metabolite with retention time ( Rt ) = 11 . 7 min ( A220 nm ) was apparent in ΔgliT26933 extracts which was absent in wild-type extracts ( Figure 3 ) . This material was purified to assess any growth inhibitory effect , however when added to AMM cultures of ΔgliT or wild-type no alteration of growth rates was observed ( data not shown ) . High resolution LC-ToF MS analysis of the metabolite ( from Figure 3B ) confirmed the presence of a molecular ion with a mass of 279 . 0796 m/z ( ( M+H ) + ) ( Figure S4 ) . This accurate mass value ( 279 . 0796 m/z ) corresponded to a predicted molecular formula of C13H15N2O3S for the ion whereby the calculated exact mass for C13H15N2O3S + H+ was 279 . 0798 Da using Agilent Technologies Masshunter workstation software . This result suggests that a monothiol form of gliotoxin could have been secreted from A . fumigatus ΔgliT26933 . A molecular species of m/z 279 , which yielded daughter ions of m/z 261 . 1 , 231 . 0 and 203 . 1 , upon MS2 analysis , was also detected by LC-MS analysis of the purified gliotoxin-related metabolite from A . fumigatus ΔgliT26933 ( Figure S4 ) . Gliotoxin was not produced by A . fumigatus ΔgliH26933 ( Figure S4 ) which strongly supports a role for this gene in gliotoxin biosynthesis or secretion , but not protection against exogenous gliotoxin . This result was further consolidated whereby no gliotoxin production was detectable , by HPLC-DAD or LC-MS , in A . fumigatus ΔgliT26933gliH ( data not shown ) , which was generated by restoration of the fully intact gliH in gliT-deficient A . fumigatus ( Figure S1 ) . Recombinant GliT was expressed in , and purified by differential extraction from , E . coli with a yield of approximately 5 . 7 mg per gram of cells . However the protein was completely insoluble and was refractory to any attempts at refolding for activity analysis ( data not shown ) . SDS-PAGE analysis confirmed a subunit molecular mass of 38 kDa for recombinant GliT ( Figure S5 ) , which appears to migrate as a dimer under non-reducing conditions ( Figure S5 ) , and protein identity was unambiguously confirmed by MALDI-ToF MS whereby peptides ( following tryptic digestion ) were identified yielding 21% sequence coverage ( Figure S6 ) . Immunoaffinity purification of GliT-specific human IgG was achieved by incubation of human sera with Sepharose-coupled recombinant GliT . The specificity of this GliT-specific human IgG was confirmed by the successful detection of native GliT in both A . fumigatus cell lysates , and partially-purified extracts of A . fumigatus ( Protocol S1; Figure 4 ) . Notably , GliT was not detectable in A . fumigatus ΔgliT ( Figure 4 ) . Previous hypotheses have suggested that GliT may only exhibit gliotoxin oxidase activity ( responsible to disulphide bridge closure during biosynthesis ) ( 3 , 8 , 22 ) . However , following gliotoxin induction of A . fumigatus ATCC46645 , enhanced GliT activity was evident in cell lysates and native GliT was partially purified by ammonium sulphate precipitation and ion-exchange chromatography ( Figure S7 ) . Data presented in Figure 5A confirm that partially-purified native GliT specifically catalyses the NADPH-mediated reduction of oxidized gliotoxin , whereby NADPH oxidation is only evident in the presence of both gliotoxin ( 9 µM ) and GliT-containing lysates . Hence , GliT appears to exhibit gliotoxin reductase activity which can catalyse disulphide bridge cleavage , at concentrations up to 9 µM gliotoxin ( Figure 5B ) . This activity is inhibited at higher gliotoxin concentrations ( >12 µM ) . Not unexpectedly , A . fumigatus cell extracts appear to contain basal NADPH oxidase activity which yields background , non-specific NADPH oxidation ( Figure 5A ) . Thus , A . fumigatus ATCC46645 and ΔgliT lysates , generated without prior gliotoxin induction of GliT expression , exhibit near-identical activity . However , significantly greater gliotoxin reductase activity ( 2:1 ) was apparent in A . fumigatus ATCC46645 , than ΔgliT , cell lysates following gliotoxin exposure ( Figure 5C ) . Immunoprecipitation of GliT from partially purified A . fumigatus cell lysates ( Figure S7 ) using human IgG [anti-GliT] resulted in a 51% reduction of gliotoxin reductase ( NADPH oxidase ) activity ( Figure 5D ) , in complete accordance with data in Figure 5C , further confirming enzyme specificity . Interestingly , GliT activity was not enhanced in the presence of thioredoxin from Spirulina sp . , in activity assays , which indicates that GliT is specific for gliotoxin reduction and that it may operate independently of cellular thioredoxin reductase/thioredoxin systems . Expression of GliT in A . fumigatus was further explored by fluorescence confocal microscopy . Data in Figure S8A-C confirm transformation of A . fumigatus ΔgliT46645 and that gliT-gfp expression is enhanced by gliotoxin addition . As shown in Figure S8A , it appears that low-level GliT expression is evident throughout mycelia without gliotoxin addition . However , following mycelial exposure to gliotoxin ( 5 µg/ml ) , an enhancement of GliT expression in the cytoplasm , and in nuclei , as shown by fluorescence intensities ( Figure S8B & C ) , is observed - which is in complete agreement with proteomic , molecular and enzyme activity observations . Expression of GliT-GFP fusion protein completely restored gliotoxin resistance ( 10 µg/ml ) , although colonies appeared white ( Figure S9 ) . The concordance of these data lead us to conclude that a GliT-mediated gliotoxin reductase activity is induced by exposure of A . fumigatus to gliotoxin . A prerequisite for testing A . fumigatus ΔgliT virulence was to evaluate the utility of our G . mellonella infection model . To this end , assessment of the relative virulence of A . fumigatus ΔgliZ and corresponding wild-type in G . mellonella , in either the presence or absence of added gliotoxin , was assessed ( Figure S10 ) . Here , all Galleria exposed to A . fumigatus ΔgliZ were alive at 24 h and the wild-type strain exhibited greater virulence than ΔgliZ ( 60% ( 12/20 ) versus 20% ( 4/20 ) mortality , respectively ) , at 48 h post-inoculation , thereby confirming the utility of the model system for detection of alteration in virulence associated with gliotoxin production . To assess now the relative contribution of gliT to virulence of A . fumigatus we compared the survival of larvae of the greater wax moth G . mellonella following infection with 106 conidia/larvae of A . fumigatus ATCC26933 and gliTC to that of larvae ( n = 20 ) infected with the same dose of ΔgliT26933 ( Figure S10 ) . For all groups of infected larvae , 100% mortality was recorded after 72 h and the degree of melanisation was not distinguishable between these groups . Also , pretreatment of larvae with gliotoxin ( 0 . 5 µg/larva in 20 µl ) did not lead to an attenuation of virulence of ΔgliT ( Figure S10 ) . Notably , similar results were obtained using ATCC46645 and ΔgliT46645 strains ( data not shown ) . These results clearly show that , gliT has a minimal , if any , role to play in the virulence of A . fumigatus employing a Galleria model . Reintroduction of gliT into A . fumigatus ΔgliT was selected for in the presence of gliotoxin and no additional selection marker was required ( Figure S1 and Figure 1D ) . To further test the ability of gliT to confer resistance to gliotoxin , and its future potential as a selection marker gene , we introduced gliT into A . nidulans which does not produce gliotoxin and neither does it contain any genes involved in gliotoxin biosynthesis [22] , [23] . The absence of gliT , and cognate gene expression , in A . nidulans was confirmed by Southern and Northern analysis ( Figure 6A & B ) . Subsequent transformation of A . nidulans with A . fumigatus-derived gliT resulted in the generation of three transformants ( AngliT 1 , 2 and 3 ) ( Deposition number: IMI CC 396693 ) , which were shown by Northern analysis to express gliT to different extents ( Figure 6B ) . This led to acquisition of resistance to high levels of exogenous gliotoxin ( 50 µg/ml ) ( Figure 6C ) thereby confirming the key role of gliT in protection against gliotoxin toxicity in gliotoxin-naïve fungi . The gliT coding sequence was also transformed into the genetically distant yeast , S . cerevisiae BY4741 , under control of the constitutive SSA2 promoter [24] in plasmid pC210 . As can be seen in Figure 6D , yeast transformed with plasmid-encoded gliT were capable of growth in the presence of gliotoxin ( 16 and 64 µg/ml , respectively ) depending on whether minimal or rich media was used to support growth , while those transformed with empty vector were unable to grow , irrespective of what media conditions were used . These observations further confirm the pivotal role of gliT in mediating resistance to gliotoxin , even in fungal species which do not normally contain the gene or biosynthesise gliotoxin . Studies into the biosynthesis and pathogenicity of gliotoxin have attracted significant recent attention , stimulated in part by the plethora of fungal genome data now emerging [3] , [22] . Here , we demonstrate for the first time that disruption of gliT , found within the gliotoxin biosynthetic cluster , but subject to differential regulation , completely sensitizes A . fumigatus to exogenous gliotoxin , and abolishes gliotoxin secretion . The possibility that genes adjacent to gliT in the gliotoxin gene cluster ( gliF or gliH ) play a role in auto-protection is excluded . Thus , we have elucidated a key cellular protective mechanism against the hitherto unknown , potent auto-toxicity of gliotoxin in A . fumigatus . Exposure of A . fumigatus ΔgliT to gliotoxin appears to result in depletion of intracellular GSH since the inhibitory phenotype can be completely relieved by GSH supplementation . Furthermore , we demonstrate the enzymatic functionality of GliT as a gliotoxin reductase and that GliT reactivity is evident in human sera . We demonstrate that gliT confers resistance to exogenous gliotoxin , independently of the extent of gliT expression , following transformation in naïve hosts , A . nidulans and S . cerevisiae . Finally , identification of gliT complementation in A . fumigatus ΔgliT46645 and 26933 , respectively , was selected for in the presence of gliotoxin which supports a selection marker role for gliT in A . fumigatus transformation experimentation . To date , the potential requirements for self-protection against gliotoxin , in A . fumigatus , have not been studied . The ETP toxin , sirodesmin , is produced by the fungus Leptosphaeria maculans with biosynthesis encoded by a multigene cluster similar to that responsible for gliotoxin production in A . fumigatus [13] . Deletion of the sirodesmin transporter gene , sirA , in L . maculans led to increased sensitivity to exogenous sirodesmin and gliotoxin , however the A . fumigatus gliotoxin transporter , GliA , was shown to confer resistance to exogenous gliotoxin ( 10 µM ) , but not sirodesmin , in L . maculans ΔsirA . Interestingly , production and secretion of sirodesmin was actually increased by 39% in L . maculans ΔsirA compared to wild-type and resulted in speculation as to the presence of alternative toxin efflux mechanisms [13] . Based on our observations , we hypothesise that in addition to the likely role of gliA in gliotoxin efflux in A . fumigatus , GliT may play an essential role in the auto-protective strategy against the deleterious effects of the ETP toxin . Moreover , we predict that gliT orthologs in other fungi [22] may play similar , if not identical roles . Our results indicate that absence of GliT may lead to accumulation of intracellular gliotoxin which is reduced , non-enzymatically , by GSH , analogous to the situation in animal cells as demonstrated by Bernardo et al . [17] . The concomitant depletion of intracellular GSH levels , allied to the cytotoxicity of reduced gliotoxin , results in strong growth inhibition , possibly mediated by disruption of the cellular redox status and significant protein modification by gliotoxin . This conclusion is strongly supported by the observation that addition of GSH , during exposure of A . fumigatus ΔgliT to gliotoxin , effectively completely reverses the cytostatic effects of gliotoxin . While we cannot exclude the possibility that added GSH is merely reducing exogenously added gliotoxin and preventing import of the reduced form , it is clear from Figure 1 that addition of NaBH4-reduced gliotoxin results in significant growth inhibition of A . fumigatus ΔgliT ( p<0 . 05 ) . The observed alleviation of this inhibition ( by NaBH4-reduced gliotoxin ) , in the presence of added GSH , supports the proposal that intracellular GSH depletion is a consequence of gliT disruption , when growth occurs in the presence of exogenous gliotoxin . Addition of gliotoxin ( up to 20 µg/ml ) for 24 h resulted in the complete up-regulation of the gene cluster ( except gliP ) in A . fumigatus ΔgliP , but not in A . fumigatus wild-type [9] . We demonstrate that exposure to exogenous gliotoxin for 3 h does induce GliT expression in A . fumigatus wild-type at the transcript and protein level , in fact these data also represent the first confirmed identification of a protein encoded by the gliotoxin biosynthetic cluster . The discrepancy , possibly due to 3 versus 24 h experimental windows , nonetheless , indicates differential GliT expression relative to other gli genes . Disruption of gliZ , the transcriptional regulator of the gliotoxin biosynthetic cluster , has been shown to result in abolition of gliotoxin production and loss of gliotoxin cluster gene expression [14] . Our data demonstrate that although growth of A . fumigatus ΔgliZ and gliZc is unaffected by exogenous gliotoxin , gliZ expression is up-regulated in response to exogenous gliotoxin exposure in A . fumigatus ATCC46645 , but to a lesser extent than that of gliT ( Figure 2 ) . In addition , we have shown that gliT expression is induced by gliotoxin addition to liquid cultures of A . fumigatus ΔgliZ thereby confirming the independent regulation of gliT expression to other cluster components ( e . g . , gliA and gliG ) . In combination , these observations further confirm the minimal role played by any other component of the gli gene cluster in protection against gliotoxin presence since gliZ absence results in complete cluster attenuation [14] , except for gliT . A thioredoxin system in A . nidulans has recently been described whereby a thioredoxin mutant exhibited decreased growth , impaired reproductive function and altered catalase activity [25] . These authors also identified a thioredoxin reductase ( termed AnTrxR ) which functions to regenerate reduced thioredoxin in A . nidulans . Our BLAST analysis indicates minimal identity between GliT and AnTrxR as well as between GliT and a second putative thioredoxin reductase in A . fumigatus ( Genbank accession number: EAL85952; 30% identity ) . This strongly indicates distinct functionality of gliT and confirms that alternative thioredoxin reductase activities cannot compensate for loss of gliT in A . fumigatus . It further appears unlikely that thioredoxin is involved in mediating GliT activity since no thioredoxin reductase present in A . fumigatus cell lysates appears capable of compensating for GliT absence . Consequent to its bioinformatic classification as a thioredoxin reductase , GliT has been predicted by many authors to encode disulphide bond formation in gliotoxin and to play a role in gliotoxin biosynthesis [3] , [8] , [22] . While this ‘gliotoxin oxidase’ activity cannot be ruled out completely , our demonstration that GliT exhibits gliotoxin reductase activity ( Figure 5 ) suggests that direct gliotoxin reduction is a pre-requisite for secretion from A . fumigatus via a GliT-mediated pathway or as a component of the auto-protective mechanism deployed against exogenous gliotoxin secreted by adjacent fungi in the environment ( Figure 7 ) . This hypothesis is firmly supported by the absence of gliotoxin secretion in A . fumigatus ΔgliT26933 . Given the potential of reduced gliotoxin to thiolate cellular proteins , we speculate that reduced gliotoxin may be sequestered into intracellular vesicles where it is converted to the oxidized form , by an unidentified activity , prior to release from the cell by an exocytotic mechanism complementary to GliA-mediated efflux ( Figure 7 ) . It remains possible that GliT-mediated gliotoxin oxidase activity may be associated with disulfide bridge closure during gliotoxin biosynthesis when intracellular levels of gliotoxin can be regulated more precisely by the organism . Thus , GliT could be necessary to maintain a balance between reduced and oxidised gliotoxin in A . fumigatus . The detection of a molecular ion , with a molecular mass corresponding to a monothiol form of gliotoxin , in culture supernatants from A . fumigatus ΔgliT is interesting , and we hypothesize that this metabolite may represent a breakdown product of gliotoxin . Future work will involve purification and complete characterization of this molecule . The observation that GliT-specific IgG was present in human sera was unexpected and implies that GliT is either present in inhaled conidia or is expressed during abortive conidial germination in immunocompetent individuals . However , our observation suggests that the option of using normal human sera as a source of immunoaffinity antibodies , following Ig isolation and purification using a recombinant antigen ( e . g . , GliT ) , represents a novel approach for readily obtaining monospecific antisera against antigenic A . fumigatus proteins . The animal model system deployed herein appears to distinguish between virulence diminution associated with lack of gliotoxin production , since inoculation with A . fumigatus ΔgliZ resulted in reduced Gallerial mortality than exposure to wild-type A . fumigatus . This result extends previous observations with respect to the potential avirulence of A . fumigatus ΔgliZ [14] . However , the relatively equivalent virulence observed for A . fumigatus wild-type and ΔgliT , whereby the latter does not produce gliotoxin is somewhat at variance with the A . fumigatus ΔgliZ findings . We suggest that alterations in the levels of additional metabolites in A . fumigatus ΔgliZ , as noted in [14] , or a possible cytotoxic role in G . mellonella for the putative monothiol form of gliotoxin secreted by A . fumigatus ΔgliT may account for this dichotomy . Our demonstration that gliT is expressed independently of other cluster components implies that previous virulence model experimentation , involving gliP- and gliZ –deficient mutants [9]–[12] , may require interpretation in light of the possibility of independently regulated gliT expression , or GliT functionality . Indeed , if it is ever demonstrated that gliT expression occurs in the absence of gli cluster expression/gliotoxin biosynthesis ( as has been demonstrated herein for A . fumigatus ΔgliZ ) , or is regulated by factors other than exposure to exogenous gliotoxin , then consideration may need to be given to this phenomenon in future studies . This consideration is based on the fact that independent regulation of gliT may have enabled acquisition of functionality beyond a role in gliotoxin biosynthesis or auto-protection . Genetic modification of filamentous fungi for the improved production of food additives , industrial enzymes or pharmaceuticals is an ongoing requirement of the biotechnological industry [26] , [27] . Antibiotic-producing fungi are continually subjected to strain improvement , with a concomitant requirement for new selection markers , to increase product yield and decrease the level of unwanted side-products [28] . Our observation that gliT complementation in A . fumigatus can be selected for in the presence of gliotoxin , without the use of conventional selection markers , and that transformation of A . nidulans and S . cerevisiae with gliT confers enhanced resistance to gliotoxin offers the possibility of using the gliT/gliotoxin combination to select for fungal transformation . Moreover , acquired gliotoxin resistance in A . nidulans and S . cerevisiae resulting from gliT presence , underpins the important role played by this gene in mediating resistance to exogenous gliotoxin . Gliotoxin isolated from cultures of a marine fungus from the genus Pseudallescheria has been shown to possess both anti-bacterial and free-radical scavenging capability whereby an MIC50 of 1 µg/ml was observed against methicillin-resistant Staphylococcus aureus [29] . Gliotoxin may also provide a competitive advantage for A . fumigatus when grown in the presence of other fungi [30] . In this regard , gliotoxin production has been detected when A . fumigatus was co-cultured , at both 30 and 37°C , with a range of other Aspergillus spp . , leading the authors to speculate that co-expression of resistance genes may allow toxin producers to resist the effects of their own biological arsenal in competitive co-culture situations [30] . The parallel between this supposition , and our observation of GliT-mediated resistance to exogenous gliotoxin , is vivid . The vast majority of literature surrounding the role of gliotoxin in A . fumigatus focuses on its function as a cytotoxic molecule which has deleterious effects on cells within infected individuals and exhibits anti-microbial activity [5] , [6] , [9]–[12] , [29] , [30] . However , based on our observations and significant other literature [16] , [18] , [31] , a credible alternative hypothesis is that gliotoxin may actually be part of the intracellular antioxidant defense system within A . fumigatus , and is a molecule , analogous to thioredoxin or 2-cys peroxiredoxin , which may undergo rapid changes in redox status to buffer against specific exogenous or endogenous oxidants . In other words , the cytotoxic effects of gliotoxin in infected host cells may actually be an indirect consequence of its role within A . fumigatus . This alternative hypothesis is not without support . Firstly , Watanabe et al . [31] have shown that the cytotoxicity of A . fumigatus culture filtrates was significantly attenuated , or absent , when cultures were grown under reduced aerobic or anaerobic conditions . Interestingly , gliotoxin production was detectable by GC-MS analysis from aerobic but not in reduced aerobic culture supernatants . Although Watanabe et al . concluded that their results indicated that gliotoxin production is increased to facilitate fungal pathogenicity ( mimicking the aerobic lung environment ) , an alternative conclusion , which is in accordance with our thinking , is that gliotoxin production is actually elevated to cope with increased oxygen levels and that secretion of gliotoxin forms part of the gliotoxin homeostasis control mechanism within A . fumigatus to prevent the side-effect of intracellular oxidative stress . As noted earlier , in animal cells it has been shown that gliotoxin may substitute for 2-cys peroxiredoxin activity in HeLa cells by accepting electrons from NADPH via the thioredoxin reductase–thioredoxin redox system to reduce H2O2 to H2O . In this way , nanomolar levels of gliotoxin may actually protect against intracellular oxidative stress [18] . Additionally , as demonstrated by Srinivasan et al . [16] , oxidized gliotoxin facilitates selective protein inactivation in the presence of molecular oxygen which , we hypothesise , could prevent global intracellular damage due to resultant reactive oxygen species . Moreover , a protective role for gliotoxin against environmental stress in A . fumigatus has been considered [2] , [13] . Our observations and consequent hypothesis now provide a vehicle to explore this proposal . In summary , we have demonstrated that GliT plays a major auto-protective role against gliotoxin toxicity in A . fumigatus which points to alternative gliotoxin functionality in A . fumigatus . From a utilitarian viewpoint , gliT/gliotoxin sensitivity represents a potential new selection marker strategy for fungal transformation . The trans-fungal implications of our observations remain to be explored . This study was conducted according to the principles expressed in the Declaration of Helsinki . Ethical permission was obtained from The Ethics Committee of NUI Maynooth for the use of human serum specimens . Anonymous serum specimens were obtained with the signed agreement of the Irish Blood Transfusion Service . In general , A . fumigatus strains ( Table 1 ) were grown at 37°C in Aspergillus minimal media ( AMM ) . AMM contained 1% ( w/v ) glucose as carbon-source , 5 mM ammonium tartarate as nitrogen-source , and trace elements according to Pontecorvo et al . [32] . Liquid cultures were performed with 200 ml AMM in 500 ml Erlenmeyer flasks inoculated with 108 conidia . For growth assays , 104 conidia of the respective strains were point inoculated on AMM plates , containing the relevant supplements and incubated for 48 h at 37°C . TOPO TA cloning system ( Invitrogen ) and TOP10 E . coli cells ( F-mcrA Δ ( mrr-hsdRMS-mcrBC ) φ80lacZΔM15 ΔlacX74 recA1 araD139 galU galK Δ ( ara-leu ) 7697 rpsL ( StrR ) endA1 nupG ) were used for general plasmid DNA propagation and A . fumigatus genomic DNA was purified using a ZR Fungal/Bacterial DNA Kit ( Zymoresearch ) . For generating ΔgliT mutant strains , the bipartite marker technique was used [20] . Briefly , A . fumigatus strains ATCC46645 and ATCC26933 were co-transformed with two DNA constructs , each containing an incomplete fragment of a pyrithiamine resistance gene ( ptrA ) [21] fused to 1 . 2 kb , and 1 . 3 kb of gliT flanking sequences , respectively . These marker fragments shared a 557 bp overlap within the ptrA cassette , which served as a potential recombination site during transformation . During transformation , homologous integration of each fragment into the genome flanking gliT allows recombination of the ptrA fragments and generation of the intact resistance gene at the site of recombination . Two rounds of PCR generated each fragment . First , each flanking region was amplified from ATCC46645 genomic DNA using primer ogliT1 and ogliT4 for flanking region A ( 1 . 3 kb ) , and ogliT-2 and ogliT-3 for flanking region B ( 1 . 2 kb ) . Subsequent to gel-purification , the fragments were digested with SpeI and HindIII , respectively . The ptrA selection marker was released from plasmid pSK275 ( a kind gift from Sven Krappmann , Goettingen , Germany ) by digestion with SpeI and HindIII , and ligated with the two flanking regions A and B described above . For generation of ΔgliT , two overlapping fragments were amplified from the ligation products using primers ogliT-5 and optrA-2 for fragment C ( 2 . 6 kb ) and primers ogliT-6 and optrA-1 for fragment D ( 2 . 2 kb ) . Subsequently ATCC46645 and ATCC26933 were transformed simultaneously with the overlapping fragments C and D . In the generated mutant allele of ΔgliT-ptrA the deleted region comprises amino acids 1–325 of gliT . For reconstitution of the ΔgliT strain with a functional gliT copy , a 3 . 2 kb PCR fragment , amplified using primers ogliT-5 and ogliT-6 , was subcloned into pCR2 . 1-TOPO ( Invitrogen ) . The resulting 7 . 1 kb pgliT was linearised with AatII and used to transform A . fumigatus ΔgliT protoplasts . Taking advantage of the decreased resistance of the ΔgliT mutant to exogenous added gliotoxin ΔgliT protoplasts were transformed with pgliT and screened for wild-type resistance to gliotoxin for genetic complementation . Positive deletion- and reconstituted- strains were screened by Southern analysis ( Figure S1 ) and DIG-hybridisation probes were generated using primers ogliT-5 and ogliT-4 . To obtain knock-out constructs for the deletion of gliH a 5′ flanking region with oligos ogliH1 and ogliH4 was amplified . For the 3′ flanking region a PCR with oligos ogliH2 and ogliH3 was performed . Amplicons were digested with SpeI and HindIII , respectively . Resulting fragments were ligated to a ptrA cassette , released from pSK275 via SpeI and HindIII digest . Final PCRs were obtained using oligos ogliH5/optrA2 and ogliH6/optrA1 and used for transformation . To complement ΔgliH and ΔgliT26933 with a functional copy of gliH , oligos ogliT7 and M13 were used to amplify a PCR-fragment using pgliT as template . This fragment digested with EcoRI and SacII was cloned into pBS-KS ( Stratagene ) , resulting in pgliH . Together with pAN7-1 [33] , pgliH was used to complement A . fumigatus ΔgliH and ΔgliT26933 . GliT was C-terminally fused in frame to gfp ( green fluorescent protein ) to determine its subcellular localisation . To this end , a fragment containing gliT was amplified using oligos ogliT-5-SphI and ogliT-16 . The resulting 2 . 2 kb fragment was sub-cloned into pCR2 . 1-TOPO ( Invitrogen ) and sequenced . Via SphI digest a fragment containing the gliT promoter region and the coding sequence was released and cloned into the corresponding SphI site of pgfp , resulting in pgliTgfp . To obtain pgfp , a gfp containing fragment was released from pUCG-H [34] via SmaI and SacI and subcloned into the corresponding EcoRV and SacI sites of pGEM5zf+ ( Promega ) . The plasmid pgliTgfp was used to transform ΔgliT protoplasts via co-transformation using a phleomycin resistance gene . Phleomycin resistant transformants carrying an in-frame gliT-gfp fusion were used to localize GliT using fluorescence microscopy . Positive , GliT-GFP harbouring strains were screened by Southern analysis and hybridization probes were generated using oligos ogliT-7 and ogliT-8 . A . fumigatus transformation was carried out according to Tilburn et al . [35] . In order to obtain homokaryotic transformants , colonies from single homokaryotic spores were picked and single genomic integration was confirmed by PCR ( data not shown ) and Southern blot analysis . RNA was isolated using TRI-Reagent ( Sigma-Aldrich ) . Equal concentrations of total RNA ( 10 µg ) were size-separated on 1 . 2% agarose-2 . 2 M formaldehyde gels and blotted onto Hybond N+ membranes ( Amersham Biosciences ) . The hybridisation probes used in this study were generated by PCR using primers ogliA1 and ogliA2 for AFUA_6G09710 , ogliG7 and ogliG8 for AFUA_6G09690 , ogliT7 and ogliT8 for AFUA_6G09740 , and ogliZ1 and ogliZ2 for AFUA_6G09630 . All primers used in this study are listed in Table 2 . A . fumigatus ATCC26933 was cultured ( n = 3 ) for 21 h in Sabouraud media followed by gliotoxin addition for 3 h ( final concentration: 14 µg/ml ) . Control cultures ( n = 3 ) , where gliotoxin was not added , were also performed . Mycelia were harvested , lysed and subject to MALDI-ToF mass spectrometric analysis as previously described [36] and Imagemaster analysis ( GE Healthcare ) . To analyze gliotoxin , or related metabolite production , A . fumigatus wild-type and mutant strains were grown up at 37°C for 72 h in Czapeks-Dox . Supernatants were chloroform extracted overnight and fractions were lyophilized to complete dryness . Samples were resolubilised in MeOH and analysed using a reversed phase HPLC as described in [37] and LC-MS ( Agilent 6340 ETD LC-MS system ) . Samples ( 1 µl ) were loaded onto a Zorbax 300SB C-18 Nano-HPLC Chip ( 150 mm×75 µm , Agilent ) with 0 . 1% ( v/v ) formic acid ( 0 . 6 µl/min ) , and compounds eluted by an increasing 0 . 1% ( v/v ) formic acid , acetonitrile gradient ( 90% ( v/v ) final ) . Eluted compounds were directly ionised and analysed by ion trap mass spectrometer ( Agilent ) . For each round of MS the two most abundant compounds were automatically selected for MSn analysis . Gliotoxin was identified by its whole mass of 326 . 4 m/z and its characteristic MSn fragmentation pattern ( 263 , 245 and 227 m/z ) . LC-ToF analysis was performed using an Agilent HPLC 1200 series using electrospray ionisation inputted into a ToF ( Agilent ) . LC separation was via an XDB C18 column ( 4 . 6×150 mm ) using a water/acetonitrile ( both containing 0 . 1% ( v/v ) formic acid ) gradient at a flow rate of 0 . 5 ml/min . The gradient was started at 50% ( v/v ) acetonitrile , which was increased to 100% acetonitrile in 10 min; 100% acetonitrile was maintained for 5 min before the gradient was returned to starting conditions . Spectra were collected at 0 . 99 spectra per second . The gliT sequence was amplified from cDNA using primers incorporating terminal XhoI and HindIII sites to facilitate downstream cloning . PCR products were cloned into the pCR2 . 1 cloning vector ( Invitrogen , Carlsbad , CA , USA ) according to the manufacturer's instructions . gliT was subsequently cloned into the pProEX-Htb expression vector ( Invitrogen ) . Ligations were performed using Quickstick ligase ( Bioline , London , UK ) according to the manufacturer's instructions . pPXAgliT , the resultant expression vector was transformed into E . coli strain BL21 by standard protocols . Expression of GliT was induced by the addition of isopropyl β-D-thiogalactoside ( IPTG; to 0 . 6 mM ) and monitored by SDS-PAGE and Western blot analysis . Recombinant GliT purification was undertaken by differential extraction . Protein concentrations were determined using the Bradford method [38] with bovine serum albumin as a standard . A . fumigatus ATCC46645 mycelia were ground in liquid nitrogen and lysed in ice-cold lysis buffer as described [36] following incubation with gliotoxin ( 10 µg/ml ) for 3 h ) . Following centrifugation ( 12 , 000 g; 30 min ) , the lysate supernatant ( 176 ml ) was ammonium sulphate precipitated ( 10 , 20 , 50 and 70% ammonium sulphate ) . The 50% pellet was resuspended in 20 mM Bis-Tris propane pH 7 . 6 and dialysed three times against 50 volumes of the same buffer at 4°C . The dialysate was centrifuged ( 12 , 000 g; 20 min ) and filtered ( 0 . 45 µm ) to remove particulates . The dialysate was loaded onto an equilibrated Q-Sepharose ion-exchange ( IEX ) column ( 4 ml ) at a flow rate of 1 ml/min . The column was washed with 20 mM Bis-Tris propane pH 7 . 6 before bound protein was eluted using an NaCl gradient ( 0 . 5 M final ) . Absorbance detection was at 280 nm and 454 nm . Collected fractions were subjected to SDS-PAGE , Western blot and activity analysis for GliT . Serum specimens ( provided by the Irish Blood Transfusion Service , Dublin , Ireland according to institutional guidelines ) containing high titer IgG [anti-GliT] were pooled , diluted 1 in 4 in PBS , and applied to a GliT-Sepharose affinity column ( 0 . 5 ml ) , prepared as per manufacturer's instructions . After removal of unbound proteins by PBS washing , immobilised IgG [anti-GliT] was eluted using 100 mM glycine pH 2 . 8 , followed by immediate neutralization using 100 mM Trizma base pH 8 . 3 . Resultant immunoaffinity purified ( IAP ) IgG [anti-GliT] was used to detect native GliT by Western analysis . A . fumigatus ATCC46645 mycelia were ground in liquid nitrogen and lysed in ice-cold lysis buffer and bead-beating as described elsewhere [36] . Following centrifugation ( 12 , 000 g; 30 min ) , the lysate supernatants were used to determine gliotoxin reductase activity ( ΔA340 nm ) in the presence of gliotoxin ( 9 µM ) and NADPH ( 200 µM ) at pH 7 . 2 ( a modified version of Hill et al . [39] ) . A . fumigatus cell lysates were also subjected to ion-exchange chromatography and a pooled IEX fractions ( 250 µl ) incubated with IAP human IgG [anti-GliT] ( 100 µl ) followed by Protein A-Sepharose addition and centrifugation ( 10 , 000 g; 10 min ) . Supernatant activity analysis as described above . A . fumigatus gliTgfp and ATCC46645 mycelia were grown in cell culture six well plates ( Corning Inc . ) for 21 h before induction with ( or without ) gliotoxin ( 5 µg/ml ) . Mycelia were removed from the wells and centrifuged ( 12 , 000 g; 5 min ) . Supernatants were stored while pellets were resuspended in DAPI staining solution and incubated ( 5 min ) at room temperature . The stained mycelia were centrifuged and washed with deionised H2O before resuspension in the original supernatant . Aliquots of these preparations were analysed for GliT-GFP presence and DAPI fluorescence on an Olympus Fluoview 1000 confocal microscope . G . mellonella larvae ( n = 10 ) were inoculated into the hind pro-leg with 106 A . fumigatus conidia in 20 µl ( per larva ) [37] . In addition , one cohort of larvae was pre-treated with gliotoxin ( 0 . 5 µg/larva in 20 µl ) . Control treatments were included to ensure that neither the injection procedure , or the incubation period , were responsible for any mortality observed . These controls involved G . mellonella larvae injected with 20 µl of sterile PBS or gliotoxin alone . G . mellonella larvae were placed in Petri-dishes and incubated in the dark at 30°C . Mortality rates were recorded for 72 h post-injection . Mortality was assessed based on lack of movement in response to stimulation and discolouration ( melanisation ) of the cuticle . To introduce gliT in A . nidulans TRAN , a plasmid containing gliT coding sequence under the control of a constitutive otef [40] promoter was used . Therefore , a 1 . 1 kb fragment containing gliT was amplified using ogliT-BglII and ogliT-NotI and subcloned into pCR2 . 1-TOPO ( Invitrogen ) . A 0 . 9 kb fragment containing an otef promoter was released via BamHI/KpnI digest from plasmid pUCG-H [34] and cloned into the respective sites into pGliT-BglII-NotI . Transformation was performed as described for A . fumigatus . The S . cerevisiae strain used in this study was BY4741 ( MATa his3D1 leu2D0 met15D0 ura3D0 ) and was purchased from Euroscarf . Rich and minimal yeast medium was as described in [41] , and gliotoxin was added to the desired concentration to cooled molten agar . To monitor the effects of GliT expression in S . cerevisiae gliT was amplified from A . fumigatus ( ATCC46645 ) using PCR with primers Sc-gliT-F and Sc-gliT-R ( Table 2 ) , and cloned into the yeast shuttle vector pC210 [42] . Plasmids pC210 harbors the SSA1 coding sequence under control of the constitutive SSA2 promoter . Following digestion of pC210 with NdeI and SphI to remove the SSA1 coding sequence , similarly digested gliT PCR product was ligated into pC210 to create pC-GliT . Thus , pC-GliT harbors A . fumigatus gliT under control of the strong constitutive S . cerevisiae SSA2 promoter . The integrity of pC-GliT was confirmed by sequencing . To test the sensitivity of yeast to gliotoxin , BY4741 harboring either vector alone ( pRS315 ) or pC-GliT was grown to mid-exponential phase ( 3×106 cells/ml ) . Cells were harvested and resuspended in rich medium to a concentration of 5×106 cells/ml . Cells were serially diluted and were plated onto rich or minimal medium containing the desired concentration of gliotoxin , using a multi-pronged replicator . Plates were incubated at 30°C for 48 h with further monitoring of plates at room temperature for 72 h . The proteins named herein are available at Genbank under the following Accession numbers: GliA ( AAW03302 ) ; GliF ( AAW03300 ) ; GliG ( AAW03304 ) ; GliH/AFUA_6G09745 ( EAL88826 ) ; GliT ( AAW03299 ) and GliZ ( AAW03310 ) .	The pathogenic fungus Aspergillus fumigatus causes disease in immunocompromised individuals such as cancer patients . The fungus makes a small molecule called gliotoxin which helps A . fumigatus bypass the immune system in ill people , and cause disease . Although a small molecule , gliotoxin biosynthesis is enabled by a complex series of enzymes , one of which is called GliT , in A . fumigatus . Amazingly , nobody has really considered that gliotoxin might be toxic to A . fumigatus itself . Here we show that absence of GliT makes A . fumigatus highly sensitive to added gliotoxin and inhibits fungal growth , both of which can be reversed by restoring GliT . Neither can the fungus make or release its own gliotoxin when GliT is missing . We also show that gliotoxin sensitivity can be totally overcome by adding glutathione , which is an important anti-oxidant within cells . We demonstrate that gliotoxin addition increases the production of GliT , and that GliT breaks the disulphide bond in gliotoxin which may be a step in the pathway for gliotoxin protection or release from A . fumigatus . We conclude that gliotoxin may mainly be involved in protecting A . fumigatus against oxidative stress and that it is an accidental toxin .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "biotechnology/biocatalysis", "infectious", "diseases/respiratory", "infections", "microbiology/immunity", "to", "infections", "chemical", "biology/biocatalysis", "biotechnology/applied", "microbiology", "biotechnology/chemical", "biology", "of", "the", "cell", "infectious", "diseases/fungal", "infections", "biotechnology/protein", "chemistry", "and", "proteomics", "microbiology/applied", "microbiology", "chemical", "biology/protein", "chemistry", "and", "proteomics", "microbiology/medical", "microbiology" ]	2010	Self-Protection against Gliotoxin—A Component of the Gliotoxin Biosynthetic Cluster, GliT, Completely Protects Aspergillus fumigatus Against Exogenous Gliotoxin

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.