University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln James Van Etten Publications Plant Pathology Department 11-2014 Chlorovirus ATCV-1 is part of the human oropharyngeal virome and is associated with changes in cognitive functions in humans and mice Robert H. Yolken Johns Hopkins School of Medicine, [email protected] Lorraine Jones-Brando Johns Hopkins School of Medicine David D. Dunigan University of Nebraska-Lincoln, [email protected] Geetha Kannan Johns Hopkins School of Medicine Faith Dickerson Sheppard Pratt Health System See next page for additional authors Follow this and additional works at:http://digitalcommons.unl.edu/vanetten Part of theGenetics and Genomics Commons,Plant Pathology Commons, and theViruses Commons Yolken, Robert H.; Jones-Brando, Lorraine; Dunigan, David D.; Kannan, Geetha; Dickerson, Faith; Severance, Emily; Sabunciyan, Sarven; Talbot, C. Conover Jr.; Prandovszky, Emese; Gurnon, James R.; Agarkova, Irina V.; Leister, Flora; Gressitt, Kristin L.; Chen, Ou; Deuber, Bryan; Ma, Fangrui; Pletnikov, Mikhail V.; and Van Etten, James L., "Chlorovirus ATCV-1 is part of the human oropharyngeal virome and is associated with changes in cognitive functions in humans and mice" (2014).James Van Etten Publications. 8. http://digitalcommons.unl.edu/vanetten/8 This Article is brought to you for free and open access by the Plant Pathology Department at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in James Van Etten Publications by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. Authors Robert H. Yolken, Lorraine Jones-Brando, David D. Dunigan, Geetha Kannan, Faith Dickerson, Emily Severance, Sarven Sabunciyan, C. Conover Talbot Jr., Emese Prandovszky, James R. Gurnon, Irina V. Agarkova, Flora Leister, Kristin L. Gressitt, Ou Chen, Bryan Deuber, Fangrui Ma, Mikhail V. Pletnikov, and James L. Van Etten This article is available at DigitalCommons@University of Nebraska - Lincoln:http://digitalcommons.unl.edu/vanetten/8 Chlorovirus ATCV-1 is part of the human oropharyngeal virome and is associated with changes in cognitive functions in humans and mice RobertH.Yolkena,1,LorraineJones-Brandoa,DavidD.Duniganb,GeethaKannanc,FaithDickersond,EmilySeverancea, SarvenSabunciyana,C.ConoverTalbotJr.e,EmesePrandovszkya,JamesR.Gurnonb,IrinaV.Agarkovab,FloraLeistera, KristinL.Gressitta,OuChena,BryanDeubera,FangruiMab,MikhailV. Pletnikovc,andJamesL.VanEttenb,1 aStanleyDivisionofDevelopmentalNeurovirology,DepartmentofPediatrics,cDepartmentofPsychiatryandBehavioralSciences,andeInstituteforBasic BiomedicalSciences,JohnsHopkinsSchoolofMedicine,Baltimore,MD21205;bNebraskaCenterforVirologyandDepartmentofPlantPathology,University ofNebraska,Lincoln,NE68583-0900;anddDepartmentofPsychology,SheppardPrattHealthSystem,Baltimore,MD21205 ContributedbyJamesL.VanEtten,October3,2014(sentforreviewAugust9,2014;reviewedbyJoramFeldonandAllanV.Kalueff) Chloroviruses (family Phycodnaviridae) are large DNA viruses In the process of analyzing whole genome sequences obtained knowntoinfectcertaineukaryoticgreenalgaeandhavenotbeen fromunfractionatedsamplesoftheoropharynxfromhealthyindi- previously shown to infect humans or to be part of the human viduals participating in a study that involved the assessment of virome. We unexpectedly found sequences homologous to the cognitive functioning, we unexpectedly discovered a substantial chlorovirus Acanthocystis turfacea chlorella virus 1 (ATCV-1) in numberofsequencereadsverysimilartovirusAcanthocystisturfa- ametagenomicanalysisofDNAextractedfromhumanoropharyn- ceachlorellavirus1(ATCV-1),amemberofthegenusChlorovirus geal samples. These samples were obtained by throat swabs of (family Phycodnaviridae). This family of algae-infecting viruses is adults without a psychiatric disorder or serious physical illness common in aqueous environments but not previously thought to whowereparticipatinginastudythatincludedmeasuresofcog- nitivefunctioning.ThepresenceofATCV-1DNAwasconfirmedby infect humans or animals or to inhabit human mucosal surfaces quantitativePCRwithATCV-1DNAbeingdocumentedinoropha- (13). Viruses that cross kingdoms are rare; however, some plant ryngealsamplesobtainedfrom40(43.5%)of92individuals.The viruses can replicate in both their plant host as well as an in- presence of ATCV-1 DNA was not associated with demographic vertebratevector.However,thereisonereportindicatingapossible variablesbutwasassociatedwithamodestbutstatisticallysignif- algal-infectingvirusassociatedwithhumans.Inthisreport,cervico- icantdecreaseintheperformanceoncognitiveassessmentsofvisual vaginal secretion samples contained virus-like particles, and these processingandvisualmotorspeed.Wefurtherexploredtheeffectsof samplesinhibitedthepropagationofcertainalgalcultures,consis- ATCV-1inamousemodel.TheinoculationofATCV-1intotheintes- tentwiththepresenceofaviruscapableofinfectingalgae(14). tinaltractof9–11-wk-oldmiceresultedinasubsequentdecreasein performance in several cognitive domains, including ones involving Significance recognitionmemoryandsensory-motorgating.ATCV-1exposurein micealsoresultedinthealteredexpressionofgeneswithinthehip- pocampus.Thesegenescomprisedpathwaysrelatedtosynapticplas- Humanmucosalsurfacescontainawiderangeofmicroorganisms. ticity, learning, memory formation, and the immune response to The biological effects of these organisms are largely unknown. viralexposure. Large-scale metagenomic sequencing is emerging as a method to identify novel microbes. Unexpectedly, we identified DNA | | | chlorovirusATCV-1 infection cognitivefunctioning sequences homologous to virus ATCV-1, an algal virus not | oropharyngealvirome metagenomicsequencing previously known to infect humans, in oropharyngeal samples obtainedfromhealthyadults.ThepresenceofATCV-1wasasso- Mucosal sites of humans such as the oral pharynx and in- ciatedwithamodestbutmeasurabledecreaseincognitivefunc- tioning.ArelationshipbetweenATCV-1andcognitivefunctioning testinal tract are not sterile but contain many micro- wasconfirmedinamousemodel,whichalsoindicatedthatex- organisms including bacteria, viruses, and fungi. It has recently posuretoATCV-1resultedinchangesingeneexpressionwithin become apparent that individuals differ in the composition of the microbialfloraattheirmucosalsites,characterizedasthe“micro- thebrain.Ourstudyindicatesthatvirusesintheenvironmentnot biota” comprising the “microbiome.” Studies in both humans and thoughttoinfecthumanscanhavebiologicaleffects. animal models have shown that the microbiome can affect bio- logical functions, including cognitive performance (1, 2). Bacterial Authorcontributions:R.H.Y.,L.J.-B.,M.V.P.,andJ.L.V.E.designedresearch;L.J.-B.,G.K., F.D.,E.S.,S.S.,J.R.G.,I.V.A.,F.L.,K.L.G.,O.C.,B.D.,andM.V.P.performedresearch;R.H.Y. and fungal components of the oral microbiome have been the andL.J.-B.supervisedtheoverallperformanceoftheanalyses;D.D.D.andF.M.mapped subject of several prior studies. These studies generally rely on thevirusgenes;G.K.performedtheanimalinfectionandbehaviorexperiments;F.D. PCR-based amplification and subsequent analysis of ribosomal supervisedthecollectionofthehumansamples;E.S.supervisedtheprocessingofthe RNA sequences conserved among most species (e.g., refs. 3–6). humansamplesandthemeasurementoftheantibodiesinthemousesamples;S.S.su- pervisedtheexperimentsrelatedtohighthroughputsequencing;C.C.T.andE.P.per- Analysis of viral components in the microbiome (virome) has re- formedtheanalysisofgeneexpression;J.R.G.testedforinfectiousvirusandproduced ceived less attention, largely because there are no universal target thevirus;I.V.A.testedforinfectiousvirus;F.L.performedtheexperimentsrelatedtoviral regions suitable for PCR amplification across different viral taxa. DNAdetection;K.L.G.performedtheantibodymeasurementstudies;O.C.performedthe experimentsrelatedtohighthroughputsequencing;B.D.producedthevirus;M.V.P. This problem can be overcome by whole genome sequencing in supervisedtheanimalinfectionandbehaviorexperiments;D.D.D.,C.C.T.,E.P.,andF.M. whichviralparticlesareseparatedfromtherestofthesample,se- analyzeddata;andR.H.Y.,L.J.-B.,D.D.D.,M.V.P.,andJ.L.V.E.wrotethepaper. quenced,andthenbioinformaticallyidentified.Suchanalyseshave Reviewers:J.F.,TheSwissFederalInstituteofTechnology;andA.V.K.,TulaneUniversity. identifiedbacteriophagesasthe most abundant component of the Theauthorsdeclarenoconflictofinterest. humanoralmicrobiome(6–9).Theviromecanalsobeevaluatedby 1Towhomcorrespondencemaybeaddressed.Email:[email protected]@ sequencing unfractionated samples. This approach has the advan- gmail.com. tage of identifying viruses that might be in low concentrations or Thisarticlecontainssupportinginformationonlineatwww.pnas.org/lookup/suppl/doi:10. evadedetectionbystandardfractionationmethods(10–12). 1073/pnas.1418895111/-/DCSupplemental. 16106–16111 | PNAS | November11,2014 | vol.111 | no.45 www.pnas.org/cgi/doi/10.1073/pnas.1418895111 The surprising discovery of this apparent human–ATCV-1 association led us to conduct further investigations in humans andinmicebyusinghighthroughputsequencingmethods,PCR procedures,andimmunologicalanalyses.ThepresenceofATCV-1 genomesintheoropharynxofmanyindividualswasevaluatedin a Baltimore, Maryland-based cohort, and correlations were dis- covered between the presence of ATCV-1 and cognitive per- formance. These results prompted us to explore the ability of ATCV-1 to infect mice under experimental conditions and to study the effect of ATCV-1 infection on cognitive performance andbraingene expression. Results SAG 3.83_ATCV-1 Detection of ATCV-1 DNA in Human Pharyngeal Samples. Meta- genomic sequencing was performed on DNA extracted from oro- pharyngeal samples obtained from 33 adult individuals without a known psychiatric disorder or physical illness. The demographic characteristicsoftheseindividualsarereportedinTableS1A.The viralfractionofthesemetagenomicanalysesrevealedawiderange of virusesconsistent with human and bacterial components of the oropharyngeal cavity. However, there were an unexpected signifi- cant number of sequences that resembled chlorovirus ATCV-1. IndividualswithandwithoutdetectableATCV-1sequencesintheir oropharyngeal samples did not differ significantly in terms of the demographicvariablesof age, sex, race, educational level, level of maternaleducation,cigarettesmoking,basalmetabolicindex(BMI) score,ahistoryoftraveloutsideofNorthAmerica,orplaceofbirth. Fig.1. ChlorovirusATCV-1genomeshowingthegeneblockdistributions The sequence reads mapped to many ATCV-1 sites located [blue arrows, protein coding sequence (CDS); red arrows, tRNAs] on each strand of the genome. Histograms in black indicate the G+C distribution throughouttheviral genome(Fig.1andTableS2). along the genome; colored histograms (green, magenta) indicate the GC A quantitative PCR (qPCR) procedure with a fluorescent- skewofthegenome.Themostinnercircleindicatesthegenomemappo- labeledprobe(Taqman)wasdevelopedtoallowthroatsofmore sitionwiththestartpositionat“12o’clock.”Theviralgenomeisalinear individuals to be tested for ATCV-1 DNA. The assay relied on dsDNA,butisrepresentedhereasacircleforconvenienceofpresentation. primers directed at ATCV-1 gene z100l. The sensitivity of the Control throat swab deep sequencing consensus sequence reads are assay was ∼10 copies of target DNA based on standard curves matched to ATCV-1, and two experiments (17 and 16 individuals per ex- generated from purified ATCV-1 DNA. The qPCR assay periment) arerepresentedbytheblacklines connectingthegene blocks. detected ATCV-1 DNA in all 10 individuals with at least two BLAST hits, 61; Query, ATCV-1; Subject, human throat swab chlorovirus sequence reads homologous to ATCV-1 and in 12 of the 14 consensussequencereads(52). individualswhohadonesequencereadhomologoustoATCV-1. TheTaqmanassaywasnegativewhentestedwitheitherhuman the Wechsler Adult Intelligence Scale (WAIS) Information DNA or extracts from buffer solutions. Furthermore, the assay subtest,atestofgeneralknowledge. wasnegative,withDNAextractedfromeithertheATCV-1host Theoddsratiosdefiningtheassociationbetweenthepresence Chlorella heliozoae or with DNA from two other chloroviruses, of oropharyngeal ATCV-1 DNA and low performance on the PBCV-1 and CVM-1, and their hosts Chlorella variabilis and Micractiniumconductrix,respectively. cognitivetestsweresignificantlycorrelated.AsdepictedinFig.2, thepresenceofATCV-1oropharyngealDNAwasassociatedwith The Taqman assay was used on oropharyngeal samples from low performance on Trails A with an odds ratio of 5.2 (95% 92individuals,includingthe33individualstestedabove.Overall ATCV-1–like DNA was detected in 40 (43.5%) of the 92 sam- confidence interval, 1.63–16.7; P < 0.005) and low performance on the RBANS Total Score with an odds ratio of 4.3 (95% Y ples.IndividualswithorwithoutdetectableATCV-1DNAwere G similar in terms of demographic variables (Table S1B). No confidenceinterval,1.4–12.8;P<0.01).WithintheRBANSthere OLO ATCV-1DNAwasdetectedinbloodsamplesobtainedfromthe was a strong association between oropharyngeal ATCV-1 DNA OBI stuBdeycianudsievitdhuealisndbiyvitdhueaqlsPiCnRthaesssatuy.dy cohort were also partici- a9n5%d lcoownfpideerfnocremianntecrevaoln,1t.h7e–3a7t.t6e;nPtio<n0d.0o0m8)a.inTh(eosdedassrsoatciioa,ti8o.n0s; MICR patinginastudyofcognitivefunctioning(15),weexaminedthe were independent of the covariates of age, sex, race, socioeco- association between detection of ATCV-1 DNA and perfor- nomic status, educational level, place of birth, and current ciga- mance on a battery of cognitive tests. A significant association rettesmoking.Nosignificantdifferencesoccurredwithalowlevel occurredbetweenthepresenceoforopharyngealATCV-1DNA ofperformance onthe other RBANS domains or on thetest of andalowerlevelofperformanceontheTrailMakingTestPart information(allP>0.1). A(TrailsA),atestofvisualmotorspeed(P<0.002),aswellas thetotalscoreoftheRepeatableBatteryfortheAssessmentof EffectofATCV-1onMouseBehaviorandCognition.A series of be- Neuropsychological Status (RBANS) (P < 0.014) (Table 1). havior tests were performed on an equal number of male and Within the RBANS test, there were statistically significant dif- femalemicegavagedwitheitherC.heliozoaealone(control,n= ferences between those who had detectable oropharyngeal 20) or C. heliozoae infected with ATCV-1 at a multiplicity of ATCV-1DNAandthosewhodidnotinthedomainsofdelayed infection of 10 per cell for 5 h (C. heliozoae/ATCV-1 exposed, memory(P<0.039)andattention(P<0.011).Thesedifferences n = 30). The behavior tests were started 6 wk postinoculation. were independent of the covariates of age, sex, race, socioeco- Anopenfieldtestanddark–lightboxwereusedtoevaluatethe nomic status, educational level, place of birth, and current cig- effects of viral exposure on general locomotor activity and anx- arettesmoking.Ontheotherhand,nodifferenceswereobserved iety(16,17).Nosignificantgroupdifferencesoccurredineither between the presence/absence of ATCV-1 DNA and scores on test. The effect of ATCV-1 exposure on learning and memory Yolkenetal. PNAS | November11,2014 | vol.111 | no.45 | 16107 Table1. AssociationbetweenATCV-1oropharyngealDNAandperformanceoncognitivetests CognitiveTest ATCV-1DNAdetected,n=40 ATCV-1DNAnotdetected,n=52 Overallcohort,n=93 Pvalue TrailsA,scaledscore 38.2(12.4) 46.7(11.7) 43.0(12.7) <0.002 WAISIII,Informationsubtest,scaledscore 10.8(2.7) 10.8(2.6) 10.8(2.6) NS RBANS TotalScore 81.3(11.9) 85.4(11.5) 83.6(11.8) <0.014 AttentionIndex 91.4(17.5) 98.5(14.5) 95.4(16.2) <0.011 DelayedMemoryIndex 85.2(11.7) 88.3(9.9) 87.0(10.8) <0.039 ImmediateMemoryIndex 85.8(15.5) 89.3(14.5) 87.8(14.9) NS Visuospatial/ConstructionalIndex 72.6(9.1) 74.4(10.6) 73.6(9.9) NS LanguageIndex 93.3(17.0) 94.8(17.0) 94.2(16.9) NS Valueslistedaremeans(standarddeviations).Pvaluescalculatedbylinearregressionadjustedforage,sex,race,educationallevel,maternaleducation, cigarettesmoking,andplaceofbirth.NSindicatesP>0.1. was then tested using the Y-maze test to evaluate spatial rec- AntibodiestoATCV-1inExposedMice.Followingthecompletionof ognitionmemory(17,18).MiceexposedtoATCV-1performed thebehavioralstudies,therewereatotalof47micefromwhich atalowerlevelonthistestcomparedwithcontrolmice(Fig.3A). serum could be obtained for antibody testing ∼6 mo following ANOVAofthepercentageofentriestothepreviouslyblockedarm oral inoculation. This set included 28 of the 30 mice that had showedasignificanteffectofgroup,F(1,49)=6.4,P=0.015.This beeninoculatedwithATCV-1–infectedC.heliozoae(15females, difference could not be explained by lower locomotor activity in 13 males; 2 males died before testing) and 19 mice inoculated ATCV-1–exposedmice,asbothgroupshadsimilarnumbersoftotal with C. heliozoae alone (10 males, 9 females; 1 female died be- entrancesovertheobservationalperiod:19.8±1.45forthecontrol fore testing). We found detectable antibodies to ATCV-1 by group and 19.5 ± 1.39 for the ATCV-1 group. The effects of enzyme immunoassay (ELISA) in 10 of the 28 mice exposed to ATCV-1 exposure on recognition of a novel object or a novel lo- ATCV-1 (5 males and 5 females) but in none of the 19 mice cationwerethenevaluated.Thesetestsevaluatedifferentaspectsof exposedtoC.heliozoae alone(P<0.0033,Fisher’sexacttest). recognitionmemoryinmice(19).Nodifferenceswereobservedin ThepresenceofantibodiestoATCV-1proteinswasexaminedby baselineexploratoryactivityoftheobjectsduringtrainingbetween Western blot in 12 available blood samples from mice exposed to the ATCV-1–exposed and control group. In contrast, compared ATCV-1inthisstudy.Offivetestedsamplesthatwerepositiveby with control mice, ATCV-1–exposed mice spent significantly less ELISA,fouralsoreactedtomultipleATCV-1proteinsbutnotto timeexploringthenovelobject,F(1,48)=62.3,P<0.001(Fig.3B), C. heliozoae proteins (Fig. S9). One of the predominant proteins orthenovellocationofthesameobject,F(1,47)=7.75,P=0.008 recognized was tentatively identified as the ATCV-1 major capsid (Fig.3C).Thepassiveavoidancetestwasusedtoevaluatememory protein.NoreactiontoATCV-1proteinsoccurredinWesternblots to aversive stimuli in mice (20). No significant effect of virus ex- with the seven samples that were seronegative with ELISAs. Sera posurewasobservedinthistest. obtained from mice exposed to C. heliozoae in the absence of TheeffectsofATCV-1exposureontheacousticstartleandits ATCV-1didnotreactwithATCV-1proteins. prepulseinhibition (PPI)wereassessed.Thesearetranslational Discussion measures of sensorimotor gating that are impaired in patients with neurological and psychiatric illnesses (21). No group dif- Metagenomic sequencing of DNA extracted from human oro- ferencewasfoundinthebaselineacousticstartleresponse,and pharyngeal samples identified sequences homologous to chlor- both groups exhibited comparable amplitudes (in mV): 193.1 ± ovirus ATCV-1 in 14 of 33 (42.4%) individuals from a cohort of 8.4forcontrolmiceand212.4±8.6forATCV-1–exposedmice. adults without a known psychiatric disorder or physical illness However, ATCV-1 exposure impaired PPI in mice. Two-way who were living in an urban area in the United States. DNA repeatedmeasuresANOVAwithtreatmentasabetween-subject sequences found in study individuals mapped to many sites on factor and PPI as a within-subjects factor indicated a significant theATCV-1genomeandincludedmanyviralgenes(Fig.1).The groupeffect,F(1,45)=6.9,P=0.015,andasignificanteffectoftype finding of ATCV-1 sequences by metagenomic sequencing was of prepulse, F(4, 234) = 98.2, P < 0.001 (Fig. 3D). No Group × confirmed by a sequence-specific (gene z100l) qPCR assay that Prepulse interaction was found. Post hoc test results revealed that detected ATCV-1 sequences in oropharyngeal samples from 40 compared with control micethe ATCV-exposed grouphad signifi- of 92 (43.5%) individuals from the same study cohort. There cantlyreducedPPIaveragedacrossallprepulses(P<0.05)(Fig.3E). werenodifferencesbetweenindividuals,withorwithoutATCV-1 sequences, with respect to demographic variables, such as age, Hippocampus Gene Expression in ATCV-1–Exposed Mice. Gene ex- sex,race,socioeconomicstatus,cigarettesmoking,travelhistory, pression in the hippocampus of mice was evaluated to compare orplaceofbirth. ATCV-1 exposure (26 wk after ATCV-1 exposure) and control Asnotedintheintroduction,ATCV-1isamemberofthegenus mice.Thehippocampuswasselectedbecauseitcontainspathways Chlorovirus, family Phycodnaviridae. Viruses in the phycodnavirus essentialforlearning,memory,andbehavior(22,23).Exposureto family, together with those in the Poxviridae, Iridoviridae, Ascovir- ATCV-1 was associated with a significant up-regulation or down- idae,Asfarviridae,Mimiviridae,andMarseilleviridae,areproposedto regulation of 1,285 individual genes (Fig. S1), which could be haveacommonevolutionaryancestorandareoftenreferredtoas constructed into 34 functional pathways that met the inclusion nucleocytoplasmic large DNA viruses (24–26). Chloroviruses have criteria(TableS3).Pathwayswithmultiplecomponentsrelevantto largedsDNAgenomes(290–370kb)thatencodeupto410proteins theresponseofthemicetoATCV-1exposureandthedevelopment and many tRNAs (13). Chloroviruses infect certain unicellular, ofcognitivechangesfollowinginfectionincludepathwaysrelatedto eukaryotic, exsymbiotic chlorella-like green algae, called zoochlor- dopamine receptor signaling (Fig. S2), cyclin-dependent kinase 5 ellae.ZoochlorellaeareassociatedwiththeprotozoanParamecium (CDK5)signaling(Fig.S3),antigenpresentation(Fig.S4),immune bursaria, the coelenterate Hydra viridis, the heliozoon A. turfacea, cell adhesion (Figs. S5 and S6), and eukaryotic initiation factor 2 and other freshwater and marine invertebrates and protozoans (EIF2)(Fig.S7)withcolorcodinglegend(Fig.S8). (27, 28). Three such zoochlorellae are Chlorella NC64A (renamed 16108 | www.pnas.org/cgi/doi/10.1073/pnas.1418895111 Yolkenetal. RBANSAttention,bothofwhichmeasurevisualprocessingand visualmotorspeed. The association between the level of ATCV-1 and decreased performanceoncognitivetestswasindependentofdemographic variables (Tables S1A and S1B). Also, there was no association between the level of oropharyngeal ATCV-1 DNA and WAIS Information, indicating that the association between ATCV-1 DNA and the other cognitive domains was not due to level of general knowledge or educational background. Studies of asso- ciationsbetweenenvironmentalfactorsandcognitiveperformance in humans must always be interpreted with caution because it is possiblethatexposuretoasingleinfectiousagentmightbeassoci- ated with unmeasured exposures, such as other infectious agents, heavy metals, pollutants, or other environmental factors that also canbeassociatedwithalterationsincognitivefunctioning(33). For this reason, we developed a mouse model for assessing behavior and cognitive functioning following ATCV-1 exposure Fig.2. OddsofdetectingATCV-1inthepharynxbypercentileofscoreon bytheoralroute.Asagroup,miceinoculatedbytheoralroute cognitivetesting.Barsrepresentthemeanand95%confidenceintervalodds showed signs of impairment in new object or new location rec- ofdetectingATCV-1DNAintheoropharynxinindividualswiththeindicated ognitionmemoryandsensory-motor gating asmeasuredbyPPI test.Theoddsratiosareadjustedforthedemographicvariablesofage,sex, severalmonthsafter asingleviral exposure(Fig.3). race, maternal education, educational status, and place of birth in the ExposureofmicetoATCV-1alsoresultedinchangesingene UnitedStates.TrailsAandInformationareseparatetestsandnotpartof theRBANS.**P<0.005,*P<0.01,adjustedforthesamecovariates. expressionwithinthehippocampus,theregionofthebrainmost associated with spatial memory and navigation (22, 34, 35). Of particularinterestinlightofthecognitiveassayswerealterations C. variabilis), Chlorella Pbi (renamed M. conductrix), and C. helio- intheCdk5pathway(Fig.S3)becausethispathwayiscentralto zoae,thehostforATCV-1.ATCV-1isarepresentativeofagroupof learning and memory formation (36). Also of note were differ- viruses that infect C. heliozoae and collectively are referred to as encesinexpressionofgenesinthedopaminepathway(Fig.S2) SAGviruses. because perturbation of the dopamine system impairs novel Genomesfrom41chlorovirusesinfectingthesethreehostshave objectrecognitionandPPIinrodents(37,38),asdoalterations beensequenced,assembled,andannotated(29).Collectively,the41 in the pathway of EIF2 because this factor is central to the virusesencodegenesfrom632proteinfamilies,whereasanygiven controloflong-termsynapticplasticityandmemorystorage(39). virusonlyhas330–410protein-encodinggenes.Thus,thegenomic Althoughitisdifficulttodirectlyrelatethecognitivetestsused diversity among these viruses is large. Furthermore, the viruses inourhumanstudywithmemorytestsinmice,itisstrikingthat encodesomeunusualproteinsthatmightinfluencebrainfunction includingpotassiumionandaquaglyceroporinchannels,potassium andcalciumtransporters,polyaminemetabolicenzymes,ahistone A B C methyltransferase, and numerous sugar enzymes including several gw(teyPlloyplFCacrieoclUhdaslsywlol)wlyrthroipeat1vhneni–rrs1tuftieh0mstr0eeeairyslsslPeiaalsFairr.tsUeeerhsic/niomogmhftLhmi.aensiTodrnihtghseeyionnmuvoisbiruaniuosnlsatdeiwnscsdaocptefawhrnaap,nstleeaaor,lqttsuhaieonnth-ufdfergoocwhrtumegztihionhtoeogaurcvsthuelntaonhirrtoees- % Novel Arm Entry12345000000 * % Time with Novel Object12345678000000000 * % Time Loca(cid:415)on at Novel 468200000 * evidence that the zoochlorellae grow free of their hosts in D E indigenouswaters. 100 100 To our knowledge, this is the first report of chlorovirus gene Y sequencesintheoralpharynxofhumans.Therehavebeenseveral 80 80 * LOG rweecreentnoretpmoretnstioofnoerda.lTvhireormeeasr,ebtuwtoAeTxCplVan-1atoiornosthfoerrtchhilsoaropvpiaruresenst % PPI 4600 % PPI 4600 ROBIO C absence. First, in one report the authors specifically looked for 20 20 MI knownhumanviralpathogens(30),andsotheywouldhavemissed 0 0 p4 p8 p12 p16 p20 thechloroviruses.Second,inthreeotherreports,thesampleswere Pre-pulse(dB) filteredthroughboth0.45-μmand0.2-μmfilters;thematerialthat passed through both filters was used to extract DNA. These Fig.3. BehavioraleffectsoforalATCV-1exposure.Micewereorallyinfec- researcherswereprimarilyevaluatingbacteriophages(7,8,31).The tedwithC.heliozoaealone(openbars)orwithATCV-1–infectedC.helio- zoae(solidbars)asdescribedinthetext.(A)Spatialrecognitionmemory;the icosahedral-shapedchlorovirusesare190nmindiameterandhave y axis displays the percentage of the previously blocked (i.e., novel) arm a 34-nm spike structure protruding from one unique vertex (32). entries;*P=0.015measuredbyone-wayANOVA.(B)Novelobjectrecog- Therefore,itislikelythatATCV-1andmostchloroviruseswouldbe nition;theyaxisdepictsthepercentageoftimespentexploringthenovel trappedona0.2-μmfilter. object;*P<0.001measuredbyone-wayANOVA.(C)Placerecognitionmemory The fact that the individuals in our study population were recognition;theyaxisdepictsthepercentageoftimespentexploringthenew participating in a project, which included measures of cognitive locationofthefamiliarobject;*P<0.008measuredbyone-wayANOVA.(D) functioning,allowedustoexaminetheassociationofdetectable ImpairedPPI;micewereexposedtopresentationofpulsealone(120dB)and prepulse–pulsecombinationsacrossdifferentprepulseintensities;forexample, ATCV-1 DNA in the pharynx and performance on a range of p4indicatespairingoftheprepulse(4dBabovethebackgroundnoiseof70dB) cognitive tests. Surprisingly, the presence of ATCV-1 DNA in withthepulsealone(120dB)(seethetextformoredetails);theyaxisdisplays the oropharynx was associated with modest but statistically sig- the percentage of PPI. (E) Impaired average PPI; the y axisdisplays the per- nificantdecreasesinperformanceontestsincludingTrailsAand centageofPPI;*P<0.015measuredbyposthoctest. Yolkenetal. PNAS | November11,2014 | vol.111 | no.45 | 16109 exposure to ATCV-1 in both humans and mice was associated includingage,self-reportedrace,levelofhighesteducationalattainment,level withdecrementsinperformanceontaskscallingforvisualspatial ofmaternaleducation,andcurrentuseofcigaretteswereobtainedfromall abilities (Figs. 2 and 3). More detailed cognitive assessment of participants.Alloftheparticipantsprovidedwritteninformedconsentfollowing explanationofthestudygoalsandprocedures.Thestudywasapprovedbythe humansandmiceexposedtoATCV-1mightbetterdefinethese Institutional Review Boards of Johns Hopkins University and Sheppard associations and the relationship between exposure to ATCV-1 PrattHospital. andcognitivefunctioning. Clinicalsamples.Oropharyngealsampleswereobtainedbyswabbingtheback There are several questions relating to ATCV-1 exposure in ofthethroatusingsterilecottonswabs(BBLCultureSwabs,BectonDickinson). humans that remain to be addressed. One concerns the source Onthedayofcollection,theswabsweretransportedfromthecollectionsite of the acquisition of ATCV-1 in the virome. ATCV-1–like totheprocessinglaboratoryatroomtemperatureandthenfrozenuntil viruses are common in inland waters such as those around Bal- processedfurther,asdescribedbelow. timore, so exposure to these water sources would be relatively Cognitivetesting.Alloftheparticipantsunderwentabatteryofcognitivetests, common. The factors involved in the acquisition of ATCV-1 in aspreviouslydescribed(45).TheseincludedtheRBANS(47),TrailsA(48),and theInformationsubtestoftheWAISIII(49).Detailsofthesetestsarede- the oropharynx following exposure will be the subject of addi- scribedinSIMaterialsandMethods. tional investigations. ATCV-1 is the reference genome for the SAG virus group, but we have previously sequenced 12 other Sample Processing. DNA was extracted from throat swabs using Qiagen’s SAG viruses and there are 2–3 distinct clades (29). Therefore, GentraPuregeneBuccalCellKit.Thecollectionbrushheadsfromtheswab theexactnatureofSAGvirusescapableofcolonizingthehuman endswereexcisedandincubatedat65°Covernightinthekitcelllysisso- oropharynxandothermucosalsitesisalsoanimportantsubject lution. The manufacturer’s instructions were followed with some minor forfuturestudies. changes to the protocol as follows: (i) During the isopropyl alcohol pre- Another set of questions concerns the mechanisms by which cipitation,2μLof5mg/mLlinearacrylamide(LifeTechnologies),achemically ATCV-1 might be associated with alterations in cognitive func- synthesized reagent, was substituted for the glycogen carrier to minimize possible contamination by reagents extracted from natural sources; (ii) fol- tioning.Althoughtheexactmechanismsofthebehavioralchanges lowingthisprecipitationstep,incubationonicewasaddedfor15min;(iii) intheexposedhostremainunclear,theobservedcognitivedeficits centrifugationfollowingthe70%(vol/vol)ethanolwashwasextendedto5min areunlikelytoberelatedtosicknessbehavior,asnoovertsignsof from1min;and(iv)finalelutionofDNAwasreducedfrom100μLto30μL. malaisewerenotedinexposedmice.Similartoanumberofother microbialinfections,wethinkthatbothdirectandindirecteffects Metagenomic Sequencing. DNA samples from 33 individuals (two in- of pathogens could play a role (e.g., refs. 40–42). The finding of dependentexperimentswith17and16individuals)wereanalyzedbymet- alterations in several pathways involved in antigen processing agenomicsequencing.Thedemographicinformationontheseindividualsis andimmunecellfunctioninginthehippocampusofmiceexposed reported in Table S1A. The method used for sequencing is detailed in SI to ATCV-1 (Table S3 and Figs. S4–S6) suggests that immune MaterialsandMethods. mechanisms maybe involved, ashavebeendocumentedinother biologicalsystems(43).Wefoundevidenceofanimmuneresponse qPCRAnalysis.Oropharyngealsamplesfromalargersetofindividuals(n=92) were tested by a qPCR system. These individuals included the 33 individuals to ATCV-1 in about 35% of mice exposed to ATCV-1 when evaluated by the initial metagenomic sequencing method. The demographic measured6mofollowingasingleexposure.Thus,ourserological informationrelatedtotheseindividualsispresentedinTableS1B.Themethod andgeneexpressiondataimplicateimmuneresponsetoACTV-1 ofsamplecollectionandDNAextractionwasidenticaltothatusedinthemet- asamechanismunderlyingthecognitivedeficits.Itisconceivable agenomic sequencing. The qPCR was performed using the 5′–3′ exonuclease that immune activation produced secretion of proinflammatory activityofThermusaquaticuspolymerase(Taqman)(50).Targetprimerswere cytokinesthataffectedneuronalfunctioning,leadingtobehavioral directedattheportionofthegenomeencodingATCV-1proteinZ100L.This abnormalities.Inthiscontext,bothsharedanduniqueprofilesof targetregionwasselectedbecausetheprimersdidnothaveappreciableho- cytokine up-regulation have been shown for various microbial mologywithanyotherviruses,bacteria,oreukaryoticorganisms. infections (Borna virus vs. Toxoplasma), and it is plausible that ThemethodusedfortheTaqmanassaywasasfollows:A20×assaymix wasmadeusingforwardprimer5′-GCAATTCCGATAGTAATGGTCA-3′, differential neurobehavioral outcomes of different microbial infections may be at least in part explained by unique “sig- reverseprimer5′-CTTGTTTGGCCTTTCACAAA-3′,andprobe/56-FAM/AG TAAACCCACACCCTTTGGTAGCCA/36-TAMSp/containing18-μMprimers natures”ofcytokine expression(44). anda4-μMprobe.A20-μLreactionvolumewasusedusing10μLof2×Gene Earlier blood samples were not obtained from this cohort of ExpressionMasterMix(AppliedBiosystems)and1μLofthe20×assaymix micetoavoidaffectingthebehavioraltests.Furtherstudiesofthe withtheremainingvolumeconsistingofinputDNAandsterilewater.The kinetics of ATCV-1 infection and the immune response to in- qPCRreactionprofileconsistedofonecycleof10minat95°Cfollowedby fectionarethuswarranted.OurstudiesdocumentthatATCV-1is 45cyclesof15sat95°Cand1minat60°C.Reactionswereperformedin partofthehumanviromeandisassociatedwithcognitivechanges a Stratagene Mx3005P Thermocycler (Agilent Technology). Results were in humans and experimentally infected animals. An increased quantified with a standard curve generated from the testing of 10-fold dilutionsofaplasmidcreatedtocontainthetarget.Testingofthisstandard understandingoftheroleofATCV-1andrelatedvirusesmaylead curveindicatedthat10copiesofATCV-1DNAcouldbereliablydetectedin toanewunderstandingoftheroleoftheoropharyngealviromein theqPCRreaction.Asample,whichcontained≥10copiesofATCV-1geno- humanhealthandcognition. micDNA,wasconsideredtocontainATCV-1DNA.DNAextractedfromre- lated chloroviruses PBCV-1 and CVM-1 (13, 29) and human DNA did not MaterialsandMethods produceproductswiththisassay. StudiesinHumans. Studypopulation.Thestudycohortconsistedof92individualslivingintheBal- StatisticalAnalysisofViromeStudies.Thedemographiccorrelatesforthe timore,Marylandmetropolitanareawhodidnothavecurrentorpastpsychiatric presence of ATCV-1 DNA were calculated by χ2 analysis for categorical disordersandwhodidnothaveaseriousmedicalillnessthatwouldbelikelyto variablessuchassex,race,cigarettesmoking,historyoftraveloutsideof affect cognitive performance. The overall characterization of the study pop- NorthAmerica,andplaceofbirthandbytwo-wayanalysisofvariancefor ulation including the methods of recruitment and measures used for their continuousvariablessuch asage,levelofeducation,BMIscore,andma- characterizationwaspreviouslydescribed(45).Controlindividualswereenrolled ternaleducation,thelatterbeingusedasamarkerofsocioeconomicstatus aftertheywerescreenedtoruleoutthepresenceofcurrentorpastpsychiatric (51).Individualperformancesonthecognitivetestsdescribedabovewere disorders with the Structured Clinical Interview for Diagnostic and Statistical furthercomparedwiththeperformanceofindividualswhodidordidnot ManualofMentalDisorders,FourthEditionAxisIDisorders–NonpatientEdition havedetectableATCV-1genomesintheirpharynxusinglinearregression (46).Participantsalsometthefollowingcriteria:proficientinEnglish,nohistory modelsincorporatingthecovariatesofage,sex,race,educationallevel,ma- ofi.v.substanceabuse,absenceofmentalretardation,absenceofHIVinfection, ternal education, cigarette smoking, and place of birth. Logistic regression absenceofaseriousmedicaldisorderthatwouldaffectcognitivefunctioning, modelswereusedtocalculatetheoddsratios,whichdefinetheassociation and no indication of alcohol or substance use disorder. Demographic data betweenthepresenceofATCV-1DNAinthethroatwithlowperformanceon 16110 | www.pnas.org/cgi/doi/10.1073/pnas.1418895111 Yolkenetal. thecognitivetestsasdefinedabove,usingthesamecovariatesthatwereused for 5 h (C. heliozoae/ATCV-1), pelleted (3,800 × g × 5 min, 4 °C), and then inthelinearregressionmodels.Tohaveadequatestatisticalpower,analysesof resuspended in 0.4× PBS. Mice were gavaged with 0.2 mL of either cognitivefunctioningwereonlyperformedonthelargercohort(n=92),on C. heliozoae/ATCV-1 (n = 30) or C. heliozoae control preparations con- whomATCV-1DNAwasdetectedbytheqPCRmethoddescribedabove. taining∼4×107cells(n=20),withbothgroupsequallydividedbetween malesandfemales. MouseModelofInfection.Animal model studies were performed to de- ThetestsusedtomonitorthebehaviorofthemiceexposedtoATCV-1,as terminetheeffectofATCV-1exposureviatheoralrouteoncognitionand wellastheproceduresusedforRNA extraction,microarray analysis, data otherbehaviors.Theconditionsofviralgrowthandmouseinoculation normalization and statistical analysis for microarray transcriptomics, weredeterminedbyasetofpreliminaryexperiments.Allprotocolswere pathwayanalysis,andmeasurementofantibodiestoATCV-1andrelated approved by the Animal Care and Use Committee at Johns Hopkins chloroviruses in mouse blood samples are described in SI Materials University. andMethods. Atotalof50C57BL/6maleandfemalemice(CharlesRiverLaboratories) wereusedtoevaluatetheeffectofATCV-1exposureoncognitionand ACKNOWLEDGMENTS.TheauthorsthankFabriceCasseus,JoshuaCrawford, behavior. Mice were housed five per cage (28.3 cm length × 17.4 cm RossMcFarland,andChunXiaYangfortheirexcellenttechnicalassistance width×13cmheight)unlessseparatedduetofighting.Animalshadfree with the mice studies. This work was supported by grants from the accesstofoodandwateratalltimes.Micewereinoculatedat9–11wkof Stanley Medical Research Institute, National Institute of Mental Health P50SilvioO.ConteCenteratJohnsHopkins(MH-94268),NationalScience age,asdescribedbelow. Foundation–Experimental Program to Stimulate Competitive Research Grant EPS-1004094, and a P20-RR15635 grant from the Centers of Bio- ExposuretoATCV-1.C.heliozoaehostalgaewereeitheruninfected(C.heliozoae medicalResearchExcellenceprogramoftheNationalCenterforResearch control)orinfectedwithATCV-1atamultiplicityofinfectionof10PFUpercell Resources. 1. Al-AsmakhM,AnuarF,ZadjaliF,RafterJ,PetterssonS(2012)Gutmicrobialcom- 27.PröscholdT,DarienkoT,SilvaPC,ReisserW,KrienitzL(2011)Thesystematicsof munitiesmodulatingbraindevelopmentandfunction.GutMicrobes3(4):366–373. Zoochlorella revisited employing an integrative approach. Environ Microbiol 2. DavariS,TalaeiSA,AlaeiH,SalamiM(2013)Probioticstreatmentimprovesdiabetes- 13(2):350–364. induced impairment of synaptic activity and cognitive function: Behavioral and 28.EstebanGF,FenchelT,FinlayBJ(2010)Mixotrophyinciliates.Protist161(5):621–641. electrophysiologicalproofsformicrobiome-gut-brainaxis.Neuroscience240:287–296. 29.JeanniardA,etal.(2013)Towardsdefiningthechloroviruses:Agenomicjourney 3. CostelloEK,etal.(2009)Bacterialcommunityvariationinhumanbodyhabitatsacross throughagenusoflargeDNAviruses.BMCGenomics14:158. spaceandtime.Science326(5960):1694–1697. 30.NakamuraS,etal.(2009)Directmetagenomicdetectionofviralpathogensinnasal 4. NasidzeI,LiJ,QuinqueD,TangK,StonekingM(2009)Globaldiversityinthehuman andfecalspecimensusinganunbiasedhigh-throughputsequencingapproach.PLoS salivarymicrobiome.GenomeRes19(4):636–643. ONE4(1):e4219. 5. SegataN,etal.(2012)Compositionoftheadultdigestivetractbacterialmicrobiome 31.WillnerD,etal.(2011)Metagenomicdetectionofphage-encodedplatelet-binding basedonsevenmouthsurfaces,tonsils,throatandstoolsamples.GenomeBiol13(6): factorsinthehumanoralcavity.ProcNatlAcadSciUSA108(Suppl1):4547–4553. R42. 32.CherrierMV,etal.(2009)Anicosahedralalgalvirushasacomplexuniquevertex 6. WadeWG(2013)Theoralmicrobiomeinhealthanddisease.PharmacolRes69(1): decoratedbyaspike.ProcNatlAcadSciUSA106(27):11085–11089. 137–143. 33.LiuJ,LewisG(2014)Environmentaltoxicityandpoorcognitiveoutcomesinchildren 7. PrideDT,etal.(2012)Evidenceofarobustresidentbacteriophagepopulationre- andadults.JEnvironHealth76(6):130–138. vealedthroughanalysisofthehumansalivaryvirome.ISMEJ6(5):915–926. 34.CabralHO,etal.(2014)Oscillatorydynamicsandplacefieldmapsreflecthippocampal 8. PrideDT,SalzmanJ,RelmanDA(2012)Comparisonsofclusteredregularlyinterspaced ensembleprocessingofsequenceandplacememoryunderNMDAreceptorcontrol. shortpalindromicrepeatsandviromesinhumansalivarevealbacterialadaptationsto Neuron81(2):402–415. salivaryviruses.EnvironMicrobiol14(9):2564–2576. 35.LyonL,SaksidaLM,BusseyTJ(2012)Spontaneousobjectrecognitionanditsrelevance 9. Robles-SikisakaR,etal.(2013)Associationbetweenlivingenvironmentandhuman toschizophrenia:Areviewoffindingsfrompharmacological,genetic,lesionand oralviralecology.ISMEJ7(9):1710–1724. developmentalrodentmodels.Psychopharmacology(Berl)220(4):647–672. 10.WommackKE,etal.(2012)VIROME:Astandardoperatingprocedureforanalysisof 36.ShahK,LahiriDK(2014)Cdk5activityinthebrain—Multiplepathsofregulation. viralmetagenomesequences.StandGenomicSci6(3):427–439. JCellSci127(Pt11):2391–2400. 11.SolonenkoSA,et al.(2013)Sequencingplatformandlibrary preparationchoices 37.vandenBuuseM(2010)Modelingthepositivesymptomsofschizophreniainge- impactviralmetagenomes.BMCGenomics14:320. netically modified mice: Pharmacology and methodology aspects. Schizophr Bull 12.WylieKM,WeinstockGM,StorchGA(2013)Viromegenomics:Atoolfordefiningthe 36(2):246–270. humanvirome.CurrOpinMicrobiol16(4):479–484. 38.ValsamisB,SchmidS(2011)Habituationandprepulseinhibitionofacousticstartlein 13.VanEttenJL,DuniganDD(2012)Chloroviruses:Notyoureverydayplantvirus.Trends rodents.JVisExp55(55):e3446. PlantSci17(1):1–8. 39.Costa-MattioliM,SonenbergN(2006)Translationalcontroloflong-termsynaptic 14.StepanovaOA,SolovyovaYV,SolovyovAV(2011)Resultsofalgaevirusessearchin plasticityandmemorystoragebyeIF2alpha.CritRevNeurobiol18(1-2):187–195. humanclinicalmaterial.UkrainicaBioorganicaActa2:53–56. 40.YolkenRH,TorreyEF(1995)Viruses,schizophrenia,andbipolardisorder.ClinMi- 15.DickersonF,etal.(2014)Associationbetweencytomegalovirusantibodylevelsand crobiolRev8(1):131–145. cognitivefunctioninginnon-elderlyadults.PLoSONE9(5):e95510. 41.PletnikovMV,MoranTH,CarboneKM(2002)Bornadiseasevirusinfectionofthe 16.BourinM,HascoëtM(2003)Themouselight/darkboxtest.EurJPharmacol463(1-3): neonatalrat:Developmentalbraininjurymodelofautismspectrumdisorders.Front Y 55–65. Biosci7:d593–d607. G O 17.PletnikovMV,etal.(2008)InducibleexpressionofmutanthumanDISC1inmiceis 42.KannanG,PletnikovMV(2012)Toxoplasmagondiiandcognitivedeficitsinschizo- L O associatedwithbrainandbehavioralabnormalitiesreminiscentofschizophrenia.Mol phrenia:Ananimalmodelperspective.SchizophrBull38(6):1155–1161. BI Psychiatry13(2):173–186,115. 43.KohmanRA,RhodesJS(2013)Neurogenesis,inflammationandbehavior.BrainBehav RO 18.MelnikovaT,etal.(2006)Cycloxygenase-2activitypromotescognitivedeficitsbutnot Immun27(1):22–32. MIC increasedamyloidburdeninamodelofAlzheimer’sdiseaseinasex-dimorphicpat- 44.StewartAM,etal.(2014)Cytokineandendocrineparametersinmousechronicsocial tern.Neuroscience141(3):1149–1162. defeat:Implicationsfortranslational‘cross-domain’modelingofstress-relatedbrain 19.AntunesM,BialaG(2012)Thenovelobjectrecognitionmemory:Neurobiology,test disorders.BehavBrainRes,pii:S0166-4328(14)00549-X,10.1016/j.bbr.2014.08.037. procedure,anditsmodifications.CognProcess13(2):93–110. 45.DickersonF,etal.(2008)Associationbetweencognitivefunctioning,exposureto 20.TayebiMeybodiK,VakiliZarchA,ZarrindastMR,DjahanguiriB(2005)Effectsofultra- HerpesSimplexVirustype1,andtheCOMTVal158Metgeneticpolymorphismin lowdosesofmorphine,naloxoneandethanolonmorphinestate-dependentmemory adultswithoutapsychiatricdisorder.BrainBehavImmun22(7):1103–1107. ofpassiveavoidanceinmice.BehavPharmacol16(3):139–145. 46.FirstMB,SpitzerRL,GibbonM,WilliamsJBW(2002)StructuredClinicalInterviewfor 21.SwerdlowNR,BraffDL,TaaidN,GeyerMA(1994)Assessingthevalidityofananimal DSM-IV-TRAxisIDisorders,ResearchVersion,PatientEdition.(SCID-I/P)(Biometrics modelofdeficientsensorimotorgatinginschizophrenicpatients.ArchGenPsychiatry Research,NewYorkStatePsychiatricInstitute,NewYork). 51(2):139–154. 47.RandolphC(1998)RBANSManual—RepeatableBatteryfortheAssessmentofNeu- 22.ChenG,KingJA,BurgessN,O’KeefeJ(2013)Howvisionandmovementcombinein ropsychologicalStatus(PsychologicalCorporation,SanAntonio,TX). thehippocampalplacecode.ProcNatlAcadSciUSA110(1):378–383. 48.ReitanRM(1979)ManualforAdministrationofNeuropsychologicalTestBatteriesfor 23.TsienJZ,etal.(2013)Oninitialbrainactivitymappingofepisodicandsemantic AdultsandChildren(ReitanNeuropsychologyLaboratory,Tucson,AZ). memorycodeinthehippocampus.NeurobiolLearnMem105:200–210. 49.WechslerD(1997)WechslerAdultIntelligenceScaleIII(PsychologicalCorporation, 24.IyerLM,AravindL,KooninEV(2001)Commonoriginoffourdiversefamiliesoflarge SanAntonio,TX). eukaryoticDNAviruses.JVirol75(23):11720–11734. 50.BergalloM,etal.(2009)RealtimePCRTaqManassaysfordetectionofpolyomaviruses 25.IyerLM,BalajiS,KooninEV,AravindL(2006)Evolutionarygenomicsofnucleo- KIVandWUVinclinicalsamples.JVirolMethods162(1-2):69–74. cytoplasmiclargeDNAviruses.VirusRes117(1):156–184. 51.Elo IT (2009) Social class differentials in health and mortality: Patterns and ex- 26.YutinN,WolfYI,RaoultD,KooninEV(2009)Eukaryoticlargenucleo-cytoplasmic planations.Comparativeperspective.AnnuRevSociol35:553–572. DNA viruses: Clusters of orthologous genes and reconstruction of viral genome 52.GrantJR,ArantesAS,StothardP(2012)Comparingthousandsofcirculargenomes evolution.VirolJ6:223. usingtheCGViewComparisonTool.BMCGenomics13:202. Yolkenetal. PNAS | November11,2014 | vol.111 | no.45 | 16111 Supporting Information Supporting Information Corrected February 20, 2015 Yolken et al. 10.1073/pnas.1418895111 SIMaterialsandMethods Novelobjectrecognition.Thenovelobjectrecognitiontestwasused Thetestsusedtomonitorcognitivebehavior,thebehaviorof toassessrecognitionmemory(8).Briefly,micewerehabituated themiceexposedtoATCV-1,aswelltheproceduresusedfor for4dtoanemptymousecage(28.3cmlength×17.4cmwidth× metagenomic sequencing, RNA extraction, microarray analysis, 13cmheight)for10mineachdayaspreviouslydescribed(9,10). data normalization and statistical analysis for microarray tran- Onday 5, two identical objectswere placed onopposite ends of scriptomics,pathwayanalysis,andmeasurementofantibodiesto theemptycage,andthemousewasallowedtofreelyexplorethe ATCV-1andrelatedchlorovirusesinmousebloodsamplesfollow. objectsfor10min.Aftera1-hdelay,duringwhichthemousewas heldinitshomecage,oneofthetwofamiliarobjectswasreplaced CognitiveTesting.All of the participants underwent a battery of withanovelone,andthemousewasallowedtofreelyexplorethe cognitive tests, as previously described (1). These included the familiarandnovelobjectfor5min.Thepercenttimenearthenovel Repeatable Battery for the Assessment of Neuropsychological objectwascalculatedasthetimenearthenovelobjectdividedby Status (RBANS) (2), Trail Making Test Part A (Trails A) (3), thetotaltimeneareitherobject. and the Information subtest of the Wechsler Adult Intelligence Novellocationrecognition.Novellocationrecognitionwasusedto Scale (WAIS)III (4). assessspatialrecognitionmemory.Briefly,micewerehabituated TheRBANSconsistsof12subteststhatareusedtocalculate for4dtoanemptymousecagefor10mineachday.Onday5, five index scores and a total score. Test indices are Immediate twoidenticalobjectswereplacedonoppositeendsoftheempty Memory (comprising List Learning and Story Memory tasks), cage,andthemousewasallowedtofreelyexploretheobjectsfor Visuospatial/Constructional (comprising Figure Copy and Line 10min.Aftera1-hdelay,oneobjectwasmovedtoadifferent locationinthecage,andthemousewasallowedtoexplorefor Orientation tasks), Language (comprising Picture Naming and 5min.Thepercenttimeneartheobjectatthenovellocationwas Semantic Fluency tasks), Attention (comprising Digit Span and calculated as the time near the novel location divided by total Coding tasks), and Delayed Memory (comprising List Recall, time near novel and old location. Story Recall, Figure Recall, and List Recognition tasks). Each Sensorimotorgating.SensorimotorgatingwasassessedusingPPIof index score is expressed as an age-adjusted standard score with a mean of ∼100 and an SD of ∼15. The index scores were the acoustic startle (San Diego Instruments Inc.). Mice were acclimatizedtoa70-dBbackgroundnoisefor5min.Theywere combined to yield an RBANS Total score, which is a measure then given 10 presentations each of a 120-dB pulse and 0-dB of overall cognitive functioning. Trails A requires an individual pulse. This was followed by 5–6 presentations in randomized to draw lines sequentially connecting 25 encircled numbers dis- order of a 120-dB pulse, 0-dB pulse, or the following prepulses tributed on a sheet of paper; the score is based on the time to followed by the 120-dB pulse: 74, 78, 82, 86, and 90 dB. The completethetask.TrailsAisatestofvisualscanningandmotor intervalsbetweeneachpresentationvariedfrom10to19s.PPI% speed.TheInformationsubtestoftheWAISisatestofgeneral wascalculatedby[100–(meanstartleamplitudeofeachprepulse/ knowledge, including questions about geography and literature. meanstartleamplitudeof120dBpulse)]×100.MeanPPI%was Forthelattertwotests,scoresareexpressedasanage-adjusted calculated by averaging all PPI% values for presentations of all scaledscorewithameanof10andanSDof3.Forthepurposesof prepulsesforeachexperimentalgroup. calculatingoddsratios,alowperformanceontheRBANStestswas Passiveavoidance.Associativelearningandmemorywereevaluated defined as less than or equal to 80 and low performance on the usinga2-dpassiveavoidancetest(SanDiegoInstrumentsInc.). othertestsasperformancebelowthe25thpercentile(4,5). Onday1,micewereplacedinalitcompartmentwiththegateto thedarkcompartmentclosed.Aftera30-sdelay,thegateopened, MonitortheBehaviorofMice.Behavioraltestswereperformedon allowingthemousetocrosstothedarkcompartment.Oncethe miceinoculatedwithChlorellaheliozoae/ATCV-1(exposed)and mousecrossedover,thegateautomaticallyclosed,andaftera3-s C.heliozoae(control)between6 and 22 wkpostinoculation.The delay, a 0.3-mA shock was administered for 3 s. Twenty-four tests were performed in the following order: novelty-induced ac- hourslater, the mouse was again placed inthe litcompartment tivityinopenfield,Y-maze,objectplacement,dark–lightbox,new withthe gateshut. Aftera5-sdelay, thegate opened.Thetrial object recognition or location, prepulse inhibition (PPI) of the endedeitherwhenthemousecrossedtothedarkcompartmentor acousticstartle,andpassiveavoidance. once10minelapsed.Oneachday,thelatencyinsecondsforthe Novelty-inducedactivity.Novelty-induced activity inthe open field mouse to cross from the light to dark compartment was auto- was assessed over a 30-min period using activity chambers with matically recordedandused inthe analysis. infrared beams (San Diego Instruments Inc.), as previously de- scribed(6). StatisticalAnalysesofBehavioralStudies.Thebehavioraldatawere Spatial recognition in the Y-maze. Spatial recognition memory was analyzedwithone-wayanalysisofvariance(ANOVA)foralltests evaluated in a Y-maze as described by Melnikova et al. (7). In exceptPPI.ThePPIdatawereanalyzedusingtwo-wayrepeated brief,onearmofthemazewasblockedandamousewasallowed measures ANOVA with treatment as a between-subject factor to freely explore the two open arms for 5 min. After a 20-min andPPIasawithin-subjectsfactor.Wedidnotusesexofanimals delay, the block was removed and the mouse was allowed to as an independent variable, as our analyses detected no sex- freelyexploreallthreeopenarmsfor5min.Thepercentageof dependenteffectsinthetestsdescribed.Ifthedatadidnotpass timeandvisitsintothenovel(previouslyblocked)armduringthe tests for normality or equal variance, the data were rank-trans- first2 minof the5-min trial was analyzed. formedbefore further statistical analysis. Dark–light box. Anxiety was evaluated using a dark–light box (Colbourn Instruments). Mice were placed in the transparent Metagenomic Sequencing. DNA samples from 33 individuals (two sideof the boxandallowed to freelymove between the dark and independentexperimentswith17and16individuals)wereanalyzed lightchambersfor5min.Thelatenciestocrossbetweenchambers bymetagenomicsequencing.Thedemographicinformationon wereautomaticallyrecordedusingGraphicStatev3.03. these individuals is reported in Table S1A. The method used Yolkenetal.www.pnas.org/cgi/content/short/1418895111 1of9 forthesequencingispresentedasfollows.Atotalof75–100ng differentially expressedgenesunder variousconditions,asingle of DNA was used for paired-end library generation using the expression value was assigned for each transcript, including all its NugenUltralowDRMultiplexSystem(NuGEN)followingthe exons. One-wayANOVAwasused todetect statisticallysignificant manufacturer’s instructions. The libraries were purified and changesingeneexpressionbetweensamplesfrommiceinoculated analyzed on the Bioanalyzer (Agilent Technologies) to con- with C. heliozoae/ATCV-1 and control mice inoculated with C. firm size and concentration. The purified libraries were se- heliozoaealone.Transcriptswithadifferenceofatleast2SDsin quencedusingIlluminaHiSeq,whichgenerated∼200,000,000 either direction between the ATCV-1–exposed and control mice paired-end reads of 100 nucleotides. were selected for pathway analyses (12). Sequencereadswerefilteredtoremovelow-qualitysequences, resulting in a minimum length of 60 nucleotides. To evaluate PathwayAnalysis.Network, function, andpathway analyses were putative viruses associated with the human throat sequence generatedusingIngenuityPathwaysAnalysis(IngenuitySystems), reads,human,bacteria,fungi,andparasitesequencereadswere whichfacilitatestheinterpretationofmicroarraydatabygrouping removedbybioinformaticfilteringasfollows:sequencereadswith differentially expressed genes into known functional pathways. homology to human samples were removed in two stages. The These analyses identified statistically increased representations firststageusedtheprogramBowtie(bowtie-bio.sourceforge.net/ ofthedifferentiallyexpressedgenesinbiologicallyrelevantpro- index.shtml). A sliding window approach was used to align a cesses(13).Genesthatshoweddifferentialexpressionofgreater 40-base-pairsubsequence fromthereadstothehumangenome than2SDsbetweeninoculatedandcontrolmicewerecompared Build 37 (www.snpedia.com/index.php/GRCh37). During each with those genes that did not, using the Fisher’s exact test to iterationofthisprocedure,readsmappingtothehumangenome identify the differentially expressed genes’ pathways for review were removed from the analysis and subsequences used for ofpotentialbiologicalfunction.BasedonitscuratedKnowledge alignment were offset by five bases. The second filtering of hu- Base (MAP Molecule ActivityPredictor; www.ingenuity.com/ man sequences used CLC Genomics Workbench Version 6 products/ipa/ipa-summer-release-2014), Ingenuity Pathways (www.clcbio.com) using a reference set of sequences based on Analysisfurtherpredictedwhetherthegenes’observedlevelsof the human genome Build 37 with the following settings: length altered transcription were in accordance with regulatory rela- fraction, 0.4; similarity, 0.4. Sequence reads that were not re- tionships from the literature; for example, elevated expression moved by this subtraction were filtered sequentially to remove of gene A (a known inhibitor of gene B) is observed together bacterial, fungal, and protozoan sequences by matching to ap- with reduced expression of gene B. Due to the exploratory na- propriate National Coalition Building Institute Reference Se- ture of this study, pathways with P values ≤ 0.05 were selected quence(RefSeq)databases(www.ncbi.nlm.nih.gov/refseq/).The for inclusion. remaining sequences, which consisted of <1% of the starting se- quences, were then mapped to the RefSeq complete set of viral MeasurementofAntibodiestoATCV-1andRelatedChlorovirusesin genomes (ftp://ftp.ncbi.nlm.nih.gov/refseq/release/viral/) using CLC MouseBloodSamples. GenomicsWorkbenchVersion6withthefollowingsettings:length Enzymeimmunoassays.IgG antibodies to ATCV-1were measured fraction, 0.8; similarity, 0.8. Sequences homologous to ATCV-1 by ELISA using variations of previously described procedures. (reference sequence NC_008724.1) by this analysis were further Highly purified virion stocks containing ∼1011 PFU/mL were mappedtotheATCV-1genomeusingCGView(11). diluted1:1,000in50μLcarbonatebufferandcoatedovernightat 4°Con96-wellpolystyreneflatbottomMaxiSorpplates(Nunc; RNA Extraction. Following completion of the behavioral experi- Thermo Fisher Scientific). Plates were blocked for 1 h at 37 °C ments,themicewerekilledandbrainswereremovedandplaced with Starting Block (Thermo Scientific). Plates were then in- on ice. The hippocampus was dissected, placed into RNAlater cubated with a 1:1,000 dilution of the mouse test serum in du- RNAstabilizationreagent(Qiagen),andstoredat−80°C.Total plicatewells,incubatedfor1hat37°C,washed,andincubated RNA was isolated from either the left or right hippocampus with peroxidase-conjugated goat–anti-mouse IgG for 45 min at usingmiRNeasyQiagenminikit(cat.no.217004)followingthe 37 °C (Southern Biotech). A 2,2’-azino-di-(3-ethylbenzthiazo- manufacturer’s protocol. To remove genomic DNA carryover, line-6-sulfonate) hydrogen peroxide solution (KPL Protein Re- RNA samples were treated with DNase for 20 min at 37 °C search Products) was added for color development, and using a TurboDNA-free kit from Ambion (cat. no. AM1907). absorbance was measured at 405 nm, with a reference wave- Samples were assessed for RNA quality and concentration by length of 490 nm, in an automated microtiter plate reader TapeStation 2200(Agilent Technologies). Based on these mea- (Molecular Devices), with the results expressed as absorbance surements,24sampleswereselectedforanalysis.Theseincluded units.AsamplewasconsideredpositiveforantibodiestoATCV-1 samples from 16 mice gavaged with C. heliozoae/ATCV-1 and ifitgeneratedasignalinthewellscoatedwithATCV-1thatgave from eight mice gavaged with C. heliozoae alone. anabsorbancevalueofatleast0.4units. MeasurementofantibodiesbyWesternblotting.C. heliozoae infected Microarray Analysis. RNA transcript levels were quantified by with ATCV-1 and uninfected C. heliozoae were added to SDS microarray analyses. RNAs were amplified into cDNA and Laemmlibufferandheatedto95°C,diluted1:10,andloadedon biotinylated by in vitro transcription with Affymetrix reagents, a precast NuPAGE Novex 4–12% Bis·Tris gel (Life Technolo- using the Whole Transcript Sense Target Labeling protocol as gies). Proteins were resolved through SureLock gel electropho- describedintheAffymetrixmanual(www.affymetrix.com/support/ resis,andthegelswerestainedwithSimplyBlueSafeStain(Life technical/product_updates/wt_1_1_assay.aff). Biotinylated cDNAs Technologies) to visualize protein amounts. Proteins were were purified, fragmented, and subsequently hybridized to Affy- transferred to nitrocellulose using iBlot dry transfer technol- metrixGeneChipMouseGene2.0STarrays. ogy (Life Technologies). Membranes were incubated over- night at 4 °C with Starting Block (Thermo Scientific). Data Normalization and Statistical Analysis for Microarray Tran- Following 1 h incubation with a 1:1,000 dilution of the test scriptomics.AffymetrixCELfiles,containingtherawGeneChipdata, sampleofmouseserum,theblotswerewashedandincubatedfor were prepared using GeneCp Command Console software. These 45 min with goat–anti-mouse IgG conjugated to alkaline phos- datawereextractedandnormalizedwithGenomicsSuitev6.6(Partek phatase (Southern Biotech). Following another incubation, the Inc.) software using the Robust Multichip Analysis algorithm. One blots were washed and developed with an alkaline phosphatase- C.heliozoaeinoculatedcontrol(IDno.650)hadRNAthatwas based conjugation kit (BioRad Life Science). For reference, we largelydegradedandwasomittedfromthefinalanalyses.Todetect also used a 1:500 dilution of a rabbit polyclonal antibody gener- Yolkenetal.www.pnas.org/cgi/content/short/1418895111 2of9
Description: