Citation

Material Information

Title:
Enrichment of beneficial bacteria in the skin microbiota of bats persisting with white-nose syndrome
Series Title:
Microbiome
Creator:
Lemieux-Labonté, Virginie
Simard, Anouk
Willis, Craig K. R.
Lapointe, François-Joseph
Publisher:
BMC
Publication Date:
Language:
English
Physical Description:
1 online resource

Subjects

Subjects / Keywords:
Bats -- Mortality ( lcsh )
White-nose syndrome ( lcsh )
Genre:
serial ( sobekcm )

Notes

Abstract:
Background Infectious diseases of wildlife are increasing worldwide with implications for conservation and human public health. The microbiota (i.e. microbial community living on or in a host) could influence wildlife disease resistance or tolerance. White-nose syndrome (WNS), caused by the fungus Pseudogymnoascus destructans (Pd), has killed millions of hibernating North American bats since 2007. We characterized the skin microbiota of naïve, pre-WNS little brown bats (Myotis lucifugus) from three WNS-negative hibernation sites and persisting, previously exposed bats from three WNS-positive sites to test the hypothesis that the skin microbiota of bats shifts following WNS invasion. Results Using high-throughput 16S rRNA gene sequencing on 66 bats and 11 environmental samples, we found that hibernation site strongly influenced the composition and diversity of the skin microbiota. Bats from WNS-positive and WNS-negative sites differed in alpha and beta diversity, as well as in microbiota composition. Alpha diversity was reduced in persisting, WNS-positive bats, and the microbiota profile was enriched with particular taxa such Janthinobacterium, Micrococcaceae, Pseudomonas, Ralstonia, and Rhodococcus. Some of these taxa are recognized for their antifungal activity, and specific strains of Rhodococcus and Pseudomonas are known to inhibit Pd growth. Composition of the microbial community in the hibernaculum environment and the community on bat skin was superficially similar but differed in relative abundance of some bacterial taxa. Conclusions Our results are consistent with the hypothesis that Pd invasion leads to a shift in the skin microbiota of surviving bats and suggest the possibility that the microbiota plays a protective role for bats facing WNS. The detection of what appears to be enrichment of beneficial bacteria in the skin microbiota of persisting bats is a promising discovery for species re-establishment. Our findings highlight not only the potential value of management actions that might encourage transmission, growth, and establishment of beneficial bacteria on bats, and within hibernacula, but also the potential risks of such management actions.

Record Information

Source Institution:
University of South Florida
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
K26-05345 ( USFLDC DOI )
k26.5345 ( USFLDC Handle )

USFLDC Membership

Aggregations:
Karst Information Portal

Postcard Information

Format:
Serial

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

RESEARCHOpenAccess Enrichmentofbeneficialbacteriaintheskinmicrobiotaofbatspersistingwithwhite-nosesyndromeVirginieLemieux-Labont1,4* ,AnoukSimard2,4,CraigK.R.Willis3andFranois-JosephLapointe1,4AbstractBackground:Infectiousdiseasesofwildlifeareincreasingworldwidewithimplicationsforconservationandhumanpublichealth.Themicrobiota(i.e.microbialcommunitylivingonorinahost)couldinfluencewildlifediseaseresistanceortolerance.White-nosesyndrome(WNS),causedbythefungusPseudogymnoascusdestructans(Pd),haskilledmillionsofhibernatingNorthAmericanbatssince2007.Wecharacterizedtheskinmicrobiotaofnave,pre-WNSlittlebrownbats(Myotislucifugus)fromthreeWNS-negativehibernationsitesandpersisting,previouslyexposedbatsfromthreeWNS-positivesitestotestthehypothesisthattheskinmicrobiotaofbatsshiftsfollowingWNSinvasion.Results:Usinghigh-throughput16SrRNAgenesequencingon66batsand11environmentalsamples,wefoundthathibernationsitestronglyinfluencedthecompositionanddiversityoftheskinmicrobiota.BatsfromWNS-positiveandWNS-negativesitesdifferedinalphaandbetadiversity,aswellasinmicrobiotacomposition.Alphadiversitywasreducedinpersisting,WNS-positivebats,andthemicrobiotaprofilewasenrichedwithparticulartaxasuchJanthinobacterium,Micrococcaceae,Pseudomonas,Ralstonia,andRhodococcus.Someofthesetaxaarerecognizedfortheirantifungalactivity,andspecificstrainsofRhodococcusandPseudomonasareknowntoinhibitPdgrowth.Compositionofthemicrobialcommunityinthehibernaculumenvironmentandthecommunityonbatskinwassuperficiallysimilarbutdifferedinrelativeabundanceofsomebacterialtaxa.Conclusions:OurresultsareconsistentwiththehypothesisthatPdinvasionleadstoashiftintheskinmicrobiotaofsurvivingbatsandsuggestthepossibilitythatthemicrobiotaplaysaprotectiveroleforbatsfacingWNS.Thedetectionofwhatappearstobeenrichmentofbeneficialbacteriaintheskinmicrobiotaofpersistingbatsisapromisingdiscoveryforspeciesre-establishment.Ourfindingshighlightnotonlythepotentialvalueofmanagementactionsthatmightencouragetransmission,growth,andestablishmentofbeneficialbacteriaonbats,andwithinhibernacula,butalsothepotentialrisksofsuchmanagementactions.Keywords:Skinmicrobiota,White-nosesyndrome,Myotislucifugus,Endangeredspecies,Conservationandmanagement,16SrRNA *Correspondence:virginie.lemieux-labonte@umontreal.ca1DpartementdeSciencesBiologiques,UniversitdeMontral,CP6182,SuccursaleCentre-ville,Montral,QubecH2V2S9,Canada4QuebecCentreforBiodiversityScience,CP6182,SuccursaleCentre-ville,Montral,QubecH2V2S9,CanadaFulllistofauthorinformationisavailableattheendofthearticle TheAuthor(s).2017OpenAccessThisarticleisdistributedunderthetermsoftheCreativeCommonsAttribution4.0InternationalLicense(http://creativecommons.org/licenses/by/4.0/),whichpermitsunrestricteduse,distribution,andreproductioninanymedium,providedyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinktotheCreativeCommonslicense,andindicateifchangesweremade.TheCreativeCommonsPublicDomainDedicationwaiver(http://creativecommons.org/publicdomain/zero/1.0/)appliestothedatamadeavailableinthisarticle,unlessotherwisestated.Lemieux-Labontetal.Microbiome (2017) 5:115 DOI10.1186/s40168-017-0334-y

PAGE 2

BackgroundInfectiousdiseasesofwildlifeareontheriseworld-widewithdramaticconsequencesforwildlifeconser-vationandhumanpublichealth[1–3].InNorthAmerica,insectivorousbatsprovideimportantecosystemservicesbylimitinginsectpestsandpotentiallysavingbillionsofdollarsannuallyforagriculture[4,5].However,anumberofecologicallyimportantspeciesarethreatenedbywhite-nosesyndrome(WNS).Thisskindisease,causedbythefungusPseudogymnoascusdestructans(Pd)[6,7],haskilledmillionsofNorthAmericanbatssince2006[8].White-nosesyndromeinvolvesinvasionofexposedskinbyPd,andthediseaseisdefinedbycup-shapederosionsandulcerationsonthetissueoftheflightmembranes(wingsandtail),ears,andmuzzle[9].Infectionoftheflightmembranesisthoughttobethemostpathologicallysignificantaspectoftheinfectionbe-causethistissueisinvolvedinfluidbalance,thermoregu-lation,andgasexchange[10].Pdinvadeshairfolliclesandsebaceousandapocrineglands[9].Thislikelydisruptssecretionsthatcontributetoskinintegrity[11,12]withconsequencesfordefenseagainstpathogensandimport-antskincommensalmicroorganisms[13].Hibernatingbatssurvivethewinteronjustafewgramsofstoredfatbyusingprolongedenergy-savingboutsoftorporcharacter-izedbydramaticallyreducedbodytemperaturesandmetabolism[14–16].Pdisadaptedforgrowthatthelowtemperaturecharacteristicofbatskinduringtorpor[6],andinfectioncauseshibernatingbatstowarmuptoofrequentlyduringwinteranddepletetheirfatreserves[17,18].Theimmunesystemisdownregulatedduringhibernation[19–21]which,inturn,facilitatesinfection.SevenspeciesofbatshavesufferedimpactsfromWNSinNorthAmerica[22]butnotallbatspeciesareequallyaffected[23,24].Ithasbeensuggestedthatenvironmentalconditionsinsidehibernacula,physiology,andbehaviorcouldallplayaroleinthevariabletoleranceof,orresistanceto,infectionwithPdamongspecies[22,24–26].InCanada,thenorthernlong-earedbat(Myotisseptentrionalis),thelittlebrownbat(Myotislucifugus),andthetricoloredbat(Perimyotissubflavus)arelistedasfederallyen-dangered[22]duetomortalityratesof75–90%dur-ingtheseveral-yearinvasionstageofthedisease[27].DespiteextremelyhighmortalityduringtheepidemicstageofWNS,somehibernatingcoloniesofatleastonehighlyvulnerablespecies(e.g.,M.lucifugus)seemtohavepersistedfollowingdiseaseinvasion[24,28–30]withcolonycountsstabilizingatabout5to30%oftheirinitialsize[24,28].Recently,itwasobservedthatinten-sityofinfectionwithPd,basedonswabsofbatforearmsandquantitativePCR,wassignificantlylowerforpersist-ingcoloniesinwhichPdhadbecomeestablished,comparedtocoloniesinthemidstoftheepidemicphaseandmassivedeclines[31].Onemechanismthatcouldexplainthispatternisafundamentalshiftinthemicro-bialcommunitylivingonbatskinduetoselectionforPdantagonists.StrongselectionformicrobialtaxathatinhibitPdcouldprovideresistancetothefungusandincreasebatsurvival.Animalskinisanecosysteminhabitedbyhighlyvariableandcomplexcommunitiesofmicroorganisms[13].Thiscommunity,calledmicrobiota,canbedividedintoaresidentflora,definedasarelativelystableassemblageinsizeandcomposition,andatransientflora,acquiredfromthelocalenvironmentandthatonlytemporarilycolonizestheskin[32].Ahealthyskinmicrobiotacandirectlycontributetohostfitnessbyoccupyingpathogenadhesionsitesandproducingpathogeninhibitors[13,33].Competitiveinteractionsbetweenbeneficialandpathogenicskinmicrobesarehypothesizedtoplayaroleindiseasedynamicsforwildanimals[34].Forexample,thebacteriumJanthinobacteriumlividum,whichlivesonsalamanderskin,appearstoconferresistancetothedevastatingfungalpathogenBatrachochytriumdendrobatidis[35]andcouldexplainwhysomesala-manderpopulationsdeclinewhileothersdonot.Re-cently,astrainofthebacteriumPseudomonasfluorescensisolatedfromtheskinofabatspeciesthoughttoberesistanttoWNS(Eptesicusfuscus)wasshowntoinhibitPdgrowthinvitro[36]aswellasinvivoforM.lucifugus[37].IthasbeenhypothesizedthatWNScouldcauseashiftinmicrobiotacommunitiesoftheskin[38],andthiscouldbeonemechanismunderlyingresist-anceinpersistingbats.However,itcouldalsohavenegativeconsequencesforbatpopulationsifashiftinthemicrobiotamakesiteasierforopportunisticpathogensotherthanPdtoinvadetheskin.AdetailedcharacterizationoftheskinmicrobiotaforWNS-positiveandWNS-negativebatsis,therefore,neededtofullyunderstandpotentialimplicationsofskinmicrobialcommunitiesinthecontextofWNS.Duetoitsdirectexposuretothelocalenvironment,theskinmicrobiotaismoredynamicandshouldbemorestronglyinfluencedbytheenvironment,thanthegutmicrobiota[39].Environmentandhostspeciesarestrongpredictorsofvariationintheskinmicrobiotaamongbats[40–42].However,onestudy[43]foundastronginfluenceofsiteandhabitattypeontheskinmicrobiotaof13batspeciesinthewesternUSAbutwasnotabletodetectaninfluenceofhostspeciesorsex.Forbatsandamphibians,thelocalenvironmentappearstoactasareservoirforskinmicrobiota,whilecondi-tionsontheskinmayleadtoselectionfavoringorenrichingparticulartaxa[40,44,45].Consequently,hostandlocalenvironmentalfactorsappeartointeractLemieux-Labontetal.Microbiome (2017) 5:115 Page2of14

PAGE 3

closelytoshapetheskinmicrobiota.Thissuggeststhattheskinmicrobiotacouldexhibitdramatictemporalvariationforspeciescharacterizedbyseasonalshiftsinphysiologyandhabitatselection.Batsexhibitenormouschangesinmetabolismandhabitatselectionbetweenwinterandsummer.Therefore,tofullycharacterizetheskinmicrobiotaanditsrelevancetoWNS,batsmustbesampledattheappropriatetimeduringhibernation.Severalstudieshavereportedontheskinmicro-biotaofNorthAmericanbats,but,todate,thesehaveinvolvedindividualsnotyetaffectedbyWNS[42,43]orhavebeenbasedonrelativelysmallsam-plesizes[38,40].OurobjectivewastounderstandthepotentialinteractionbetweenPdandtheskinmicrobiotaofbatsbycomparingindividualsfromWNS-positiveandWNS-negativeregions.Weusedhigh-throughput16SampliconsequencingtocharacterizethecompositionanddiversityoftheskinmicrobiotaofM.lucifugussampledfromWNS-positive(Qubec)andWNS-negative(Manitoba)hi-bernaculainthenorthernpartofthisspecies’range,inCanada.Wetestedtwopredictionsofthehypoth-esisthatWNSiscausingselectionfavoringPdantagonistsontheskinmicrobiotaofbatsintheaffectedregion.First,wepredictedthatbatspersist-inginWNS-affectedsiteswouldexhibitreduceddi-versityoftheirmicrobiotaconsistentwithstrongselectionforasubsetofpre-WNSmicrobialspecies[38].Second,wepredictedthatthemicrobiotaofpersistingbatsfromWNS-affectedsiteswouldshowaproportionalincreaseinantifungal/anti-Pdbacter-ialspeciessuchasthoseidentifiedinpreviousstudies[36,37,46].Wealsotestedthethirdhypoth-esisthatvariationintheskinmicrobiotaofhiberna-tingbatsrelatestoenvironmentalvariationinthemicrobialcommunityofagivencave.Wepredictedthatthediversityandcompositionofthemicrobialcommunitylivingonindividualbatswouldbesimilartothatfoundonsubstratesinthelocalen-vironmentoftheirhibernaculumandwoulddifferfromthatonbatsandthelocalenvironmentinotherhibernacula.MethodsSamplingandethicsDuringwinter2015–2016,wesampledtheskinmicro-biotaof33M.lucifugusfromthreeWNS-negativehiber-naculaincentralManitoba(Canada)about50kmnorthofthetownofGrandRapids(5330N;9924W)andanother33individualsfromthreesitesknowntobeWNS-positivesince2010inQubec(Canada)within60kmnorthofGatineaucity(4528N;7542W).Thetemperaturewithinsitesrangedfrom3to7Catsamplingtime.SiteandpopulationinformationarespecifiedinTable1.Batsinagivenhibernaculumwerealwayssampledfromwithinthesamearea(i.e.,room,gallery,corridor).Weselectedbatsatrandomfromamongthosewecouldreachfromtheground.Littlebrownbatsarehighlygre-gariousduringhibernation,andmostindividualsspendatleastpartoftheirtimehuddlingorclusteringduringhibernation.Wedefinedbatsindirectcontactwitheachotherasbeingmembersofthesamecluster.Sixty-fourofthe66batswesampledwereclusteringwithotherbats,andclustersizesrangedinsizefrom2to11indi-viduals.WesampledtwobatsatEmeraldthatwereroostingsolitarily(Additionalfile1).Weswabbed11individualbatspersite.Sampleswerecollectedbyswab-binginlinearstrokesthebackandforearmofeachbatfor20swithasterileWhatmanOmniswab(FisherScientific)soakedinsterile0.15MNaCl[41].SwabtipswereejectedintoMoBioPowersoilDNAisolationKittubes(MoBioLaboratories),whichweretransferredto20Cwithin24hofsamplinguntilDNAextraction.Localenvironmentsampleswerealsocollectedbyswabbingcavewallsadjacenttoclustersofsampledbatsfor20sinlinearstrokes(approx.5cm).Asanegativecontrol,ahumidifiedsterileswabwasexposedtoopenairfor20s,priortoejectingitstipintoaMoBiotube.Batsarevulnerabletodisturbanceduringhibernation,andwewerecarefultominimizetheimpactofourvisits.Onlytwopeopleenteredhibernaculaforsampling,andbatswerenothandledduringswabbingsowedidnotdeterminetheirsex.Apreviousstudyestablishedthatsexwasnotasignificantpredictoroftheexternal Table1M.lucifugushibernaculumsitesinformationinManitobaandQubecprovincesSitesProvinceHibernaculumRockPre-WNScountTotalcount2015–2016SamplingdatesAbyssManitobaCaveDolomiteNA39908/02/2016Dale’sManitobaCaveDolomiteNA38508/02/2016MicrowaveManitobaCaveDolomiteNA3009/02/2016EmeraldQubecMinePyroxenite735a1804/03/2016LaflcheQubecCaveCalcite450a15523/11/2015LamesQubecCaveCalciteUnknownb10524/11/2015a2009–2010bFirstcountof96batswasin2012–2013,afterthearrivalofWNSintheareaLemieux-Labontetal.Microbiome (2017) 5:115 Page3of14

PAGE 4

microbiotaofbats[43].Therefore,differencesamonghibernaculaarelikelytoreflecttheinfluenceofthelocalhabitat(e.g.,differencesintemperature,humidity,andenvironmentalbacteria),ratherthandifferenceinsexratioamongsites.AllmethodswereapprovedbytheAnimalWelfareandEthicsCommitteeatUniversitdeMontral(ProtocolNumber#16-015)andtheUniversityofWinnipegAnimalCareCommittee(ProtocolNumberAEO5639).DNAextraction,amplification,andsequencingBacterialgenomicDNAwasextractedfromeachswabusingtheMoBioPowersoilDNAisolationKitaccordingtothemanufacturer’sprotocol.Extractionswererandom-izedforsiteandregiontoavoiddetectingfalsepatterns[47].Extraction,amplificationblanks,andtheHM-782DHumanMicrobiomeProjectmockcommunity(BEIResources)werealsoincludedtodetectpossiblecontam-inationandassesssequencingaccuracy[47,48].Amplificationandsequencingwerethenperformedaspreviouslydescribed[49].Librarieswerepreparedusingatwo-stepPCR.ThefirstPCRamplifiedthehypervariableregionV4ofthe16SsmallsubunitribosomalgenewithforwardprimerU515_f:ACACGACGCTCTTCCGATCTYRYRGTGCCAGCMGCCGCGGTAAandreversepri-merE786_R:CGGCATTCCTGCTGAACCGCTCTTCCGATCTGGACTACHVGGGTWTCTAAT[50].Twomi-crolitersofextractedDNA(equivalentDNAamountbysample)wasaddedtothePCRreactioncontaining14.25lofsterilewater,5lHFbuffer,0.5lDNTPs,0.25lPhusionHigh-FidelityDNAPolymerase(NewEnglandBiolabsInc.),and1.5lofforwardandreverseprimers.AmplificationswereperformedwithaMastercyclerNexusGSX1(Eppendorf)underthefol-lowingconditions:initialdenaturationat98Cfor30s;30cyclesalternating98Cfor25s,40sat54C,35sat72C,andfinalelongationstepfor1minat72C.EachsamplewasamplifiedinquadruplicateandpooledtolimitpossiblePCRartifacts.AllPCRproductswerethenpurifiedbyPCRpurificationAgencourtAMPureXP(BeckmanCoulter).ThesecondPCRstepconsistedofaddingprimerscontainingabarcode(index)andIlluminaadaptersequencestoeachDNAamplicon.Todoso,4lofthefirststepamplificationproductwasaddedtoaPCRreactioncontaining10.25lofsterilewater,5lHFbuffer,0.5lDNTPs,0.25lPhusionHigh-FidelityDNAPolymerase,and2.5lofforwardprimerPE-III-PCR-F:AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCTandreversepri-merPE-III-PCR-001-096:CAAGCAGAAGACGGCATACGAGATNNNNNNNNNCGGTCTCGGCATTCCTGCTGAACCGCTCTTCCGATCT(Nindicatingtheuniquebarcode)[51].Indexingwasperformedunderthefollow-ingthermalconditions:initialdenaturationat98Cfor30s,7cyclesalternating98Cfor30s,30sat83C,andfinally30sat72C.Thissecondamplificationwasperformedintriplicate.SampleswerepooledandpurifiedwiththePCRpurificationAgencourtAMPureXP(BeckmanCoulter).Qubit2.0Fluorometer(Invitrogen)wasusedtomeasuretheDNAconcentrationofeachsam-ple.Indexedsampleswerethenpooledtoobtainafinalconcentrationrangebetween10and20ng/l.DNAwasnextdilutedanddenaturedaccordingtothemanufac-turer’sprotocolforpaired-endsequencingusingMiSeqReagentKitv2(500cycles)2250bponMiSeq(Illumina).DataanalysisWeamplified4,072,792sequencesclassifiedinto13,224operationaltaxonomicunits(OTUs)fromthe66swabsofbatskinand11environmentalsamples(oneortwopersite).Atotalof3,729,096sequencesclassifiedin11,812OTUswereamplifiedfrombatsampleswithameanof56,501sequencespersample(range9920–100,812).Atotalof343,696sequencesclassifiedin9302OTUswereobtainedfromthe11environmentalsamples,withameanof31,245se-quencespersample(range10,325–73,877).Wewereabletomatchallexpectedsequencesinthemockpositivecontrol,exceptforHelicobacterpylori,whichgenuswasnonethelessthemostabundantinthecompositionaldata(seeAdditionalfile2).Thegeneraorfamiliesofthe20expectedmocktaxawerealsothemostabundantinthemockprofile.Wedetected36falsepositives,withverylowabundances(<0.3%)(seeAdditionalfile2).AfterfilteringoutOTUswithabundancevaluessmallerthan3,samplingcontrols,extractioncontrols,andlibrarynegativecontrolsweredominatedbyHalomonas(5–75%,meanof56%)andShewanellagenera(1–26%,mean18%)(seeAdditionalfile3).Preclustering,qualityfiltering,primerremoval,mer-gingofrawsequences,andpostclusteringdereplicationwereperformedwiththeSmileTrainscripts[52]for16SdataprocessingusingUSEARCHv.7.0.1090[53].Distribution-basedclusteringusingthedbOTUcalleralgorithmwasperformedtoclustersequencesintoOTUsbyconsideringthedistributionofDNAsequencesacrosssamplesanddistancesbetweensequences[51].ThecorrespondingOTUtable,providingabundancesofbacterialtaxainthedifferentsampleswasassignedwithQIIMEversion1.9.0[54]usingGreenGenesdatabaserelease13_5[55].Foralphadiversityandcompositionalanalysisofbatsamples,mitochondrialandchloroplasticDNAsequences,aswellasOTUswithabundancevaluessmallerthan3,werefilteredout,leaving3,716,672sequencesclassifiedinto9897OTUs.Inaddition,thegeneraHalomonasandShewanella,presentinnegativeLemieux-Labontetal.Microbiome (2017) 5:115 Page4of14

PAGE 5

controls,werefilteredoutfromallbatsamplesforcom-positionalanalysis,resultingin3,145,399sequencesclas-sifiedinto9575OTUs.Thediversityoftheskinmicrobialcommunity(alphadiversity)ofeachsamplewascomputedusingtheShannonindex[56].TheShannonindex,whichincludesbothOTUrichnessandevenness,wasselectedduetoitsreducedsensitivitytosampledepthdifferences[49,57](Additionalfile4).Rversion3.1.3[58]wasusedforallstatisticalanalyses.Log-transformedalphadiversityvalueswerecomparedbetweenWNS-positiveandWNS-negativeregions,usingalinearmixed-effectmodel(lme()function),andsignificancewastestedwithanova.lme()ofthenlmepackage[59].Hibernaculumandclusterswereincludedasarandomeffect.VariationindiversityamongsiteswithintheWNS-positiveandWNS-negativeregionswastestedusingaone-wayANOVA(functionaov())andposthocTukeytest(func-tionTukeyHSD())ofthepackagestats[58].Thechangeindiversityamongskinmicrobialcommunity(betadiversity)wascalculatedamongskinmicrobiotasamplesandenvironmentalsamples.Twodistinctphylogeneticdistances,unweightedUniFrac(qualitative)andweightedUniFrac(quanti-tative)[60,61],werecomputedonrarefieddata,assuchmeasurescouldbesensitivetodifferencesinsequencingdepth[62,63].UniFracdistanceswerecomputedfrombatsamplesrarefiedat9886se-quences/sampleandfromenvironmental+batsam-plesrarefiedat9898sequences/sampleafterretrievingOTUsinlowabundance(<3sequences).Computationswereperformedwiththephyloseqpackage[64].Allbetadiversityresultswerevisual-izedwithprincipalcoordinatesanalysis(PCoA)[65]usingtheordinate()function.TheUniFracdistancematrixwascheckedwithis.euclid()functionoftheade4package[66]priortotheordinationtoensurethatalldistanceswereEuclidianandproperlyrepre-sentedbyPCoA[67].Whenrequired,square-roottransformationswereappliedtoobtaindistancematricessatisfyingtheEuclidiancondition.Allphylogeny-basedUniFracdistanceswerecalculatedusingaphylogenetictreeconstructedwithFastTree2.1.8[68].Toassesstheinfluenceofexplanatoryvariablesonthemicrobiotacomposition,weuseddistance-basedredun-dancyanalysis(db-RDA),amethodintendedtoconductaredundancyanalysis(RDA)ondistancematrices[69].ItiscomputedbyfirstdecomposingUniFracdistances(weightedorunweighted)intoprincipalcoordinatesandthenapplyingRDAtothecorrespondingprincipalcoor-dinatesusingthecapscale()functionoftheRpackagevegan[70].Fourdistinctmodelswereconstructedtotesttherelativeimportanceof(1)WNSstatus(i.e.,WNS-negativevs.WNS-positive),(2)samplingsites(i.e.,thesixdifferenthibernacula),(3)typesofsamples(i.e.,batsamplesvs.localenvironmentsamples),and(4)clusters(i.e.,batclusterswithineachhibernaculum).Tobetterunderstandtherelationshipsamongexplanatorymodelsinthevariationofthemicrobialassemblages,partialdb-RDAwasalsocomputed[71].ThisformofRDAallowsforexplorationofthecontributionofanexplanatoryvariablemodelwhilecontrollingforotherexplanatorymodels.AdjustedR-squared(R2)values[72]werecalcu-latedtocomparetheexplanatorypowerofsuchmodelscontainingdifferentnumbersofvariables.Significanceofdb-RDAandpartialdb-RDAwastestedvia9999permu-tationswiththeanova.cca()functionoftheRpackagevegan.ThemicrobiotacompositionwasexploreddowntogenusleveltoassessdifferencesamonghibernaculaandbetweenWNS-positiveandWNS-negativesites.Toemphasizethesedifferences,IndicatorValuetests(IndVal)[73]wereperformedonrelativeabundancedata,usingthe26taxawitharelativeabundancelargerthan1%fortheanalysis.TheIndValindicatorvalueisbasedonthecomparisonofoccurrencesandabun-dancesoftaxaacrosspredefinedgroupsofbats(e.g.,groupedbysitesorWNSstatus).Theanalysisforanygiventaxonisnotinfluencedbyothertaxapresentinthedataset.Itprovidesanindexrangingbetween0and1,themaximumvalueindicatingataxonexclusivelypresentinonegroup.IndValiscalculatedastheproductofA(specificity,i.e.,theprobabilitythatasitebelongstothegroupgiventhefactthatagivenspeciesisfoundinthatsite)andB(fidelity,i.e.,theprobabilityoffindingagiventaxonatasitewhenthesitebelongstothatgroup)[73].Themultipatt()functionoftheRpackageindicspe-cies[74]wasusedtocomputeindicatorvalues,andsignificancewasassessedwith9999permutationsofobjectbetweengroups.Thep.adjust()functionoftheRpackagestatswasusedtocorrectpvaluesformultiplecomparisons[75].Acorrectedpvaluethresholdof0.05wasconsideredsignificantinalltests,andonlysignifi-canttaxawithaspecificityofA0.4wereretainedasindicators.ResultsAlphadiversityinWNS-positiveandWNS-negativeregionsAftercontrollingforsitesandbatclustersusingalinearmixed-effectsmodel,wefoundsignificantdifferencesinShannondiversitybetweenourpooledsetofWNS-positivehibernaculainQubecandWNS-negativehibernaculainManitoba(F1,4=16.27,p0.05)(Fig.1a).WNS-positivesiteshadsignificantlylowerShannondi-versitythanWNS-negativesites(Fig.1a).WealsofoundsignificantvariationinalphadiversitybetweensomeofLemieux-Labontetal.Microbiome (2017) 5:115 Page5of14

PAGE 6

thehibernaculaintheWNS-negativeregion(ANOVA:F2,30=22.84,p0.001),whereasallthethreeWNS-po-sitivesitesinQubecwerestatisticallyindistinguishablefromeachotherandhadrelativelylowalphadiversity(ANOVA:F2,30=0.96,p=0.395)(Fig.1b).WithintheWNS-negativeregioninManitoba,Abysscaveharboredparticularlyhighalphadiversityandwassignificantlydif-ferentfromDaleÂ’s(TukeyÂ’s;p0.001)andMicrowave(TukeyÂ’s:p0.001).BetadiversityanalysisofmicrobialcommunityassemblageWefirstusedbetadiversityanalysistoexplorecompos-itionaldifferencesamongskinmicrobiotasamplesalone,thatis,afterremovingallenvironmentalsamplesfromtheanalysis.ThePCoA,basedonunweightedUniFrac,revealedaclearseparationbetweenWNS-positivesitesinQubecandWNS-negativesitesinManitoba,andalsogroupedsamplesfromwithinthesamehibernacu-lum(Fig.2a).ThispatternwasnotobservedwithweightedUniFrac(Fig.2b),whichimpliesthataccount-ingfordifferentialabundances(weightedUniFrac),andnotjustthepresence/absenceofbacterialOTUsbetweensamples(unweightedUniFrac),affectedourresults.However,thefirstprincipalaxes,accountingfor20.1%ofthevariationinthedata,supporttheseparationofmicrobiotasamplesaccordingtoWNSstatus(Fig.2a).Inordertobetterrelatethesepatternstodifferentvar-iables,weusedadistance-basedredundancyanalysis(db-RDA)tocomputefromUniFracdistancesamongskinmicrobiotasamplesusingthreedistinctexplanatorymodels:(1)WNSstatus,(2)samplingsites(hibernacula),and(3)clusters(batclusterswithineachhibernaculum).UnweightedUniFracdistancesrevealedthateachofthesemodelsexplainedasignificantfractionofmicro-biotacommunityvariation(Table2).TheWNSstatusmodelexplained8%,sitesexplained22%,andbatclusterexplained28%ofmicrobiotacommunityvariationamongsamples.WeightedUniFracdistancesaccountingforabundanceoftaxarevealedsimilarpatternswithWNSstatusexplaining14%,sitesexplaining26%,andclusterexplaining30%ofthevariationinthemicrobiotasamples(Table2).Inlightoftheseresults,weconductedpartialRDAtobetterdistinguishtherelativeinfluenceofourthreeexplanatorymodels(Table2).ThisanalysisrevealedthatWNSstatusexplainednovariationaftercontrollingforsitesand/orclustering.Similarly,thesitesmodelexplainednoneofthevariationaftercontrollingforbatcluster.Theclustermodel,ontheotherhand,explainedasignificantfractionofvariationinmicrobiotacommunityaftercontrollingforsites,bothwithweightedandunweightedUniFracdistances.Thesere-sultssuggestthatbatclusteringwithineachhibernacu-lumexertsstronginfluenceonthecompositionoftheskinmicrobiota.ControllingforWNSstatus,however,greatlyreducedthevariationexplainedbysitesandclustermodelsalone,regardlessofUniFracdistances(unweightedorweighted).Takentogether,theresultsof WNS-WNS+ 0.50.60.70.80.9WNS statuslogShannon WNS-WNS+ AbyssDale'sMicrowaveEmeraldLaflcheLames 0.50.60.70.80.9SiteslogShannon WNS-WNS+a b b c c c a b ab Fig.1AlphadiversityofM.lucifugusskinmicrobiotainWNS-positiveandWNS-negativesitesinCanada.DistributionofalphadiversitywithingroupsasestimatedbytheShannonindexforahibernaculapooledbyWNSstatus(positivevs.negative)andballsixhibernaculasampledinthestudy.Errorbarsrepresentstandarddeviations.Significantdifferencesinalphadiversityamonggroupsareindicatedbydifferentlettersaccordingtolinearmixedmodeleffect,ANOVA,andTukeyÂ’stest(p0.05)Lemieux-Labontetal.Microbiome (2017) 5:115 Page6of14

PAGE 7

simpleandpartialdb-RDAanalysessuggestthatsites,combinedwithalocaleffectofbatclusteringwithinsites,contributetoshapingtheskinmicrobiota,whereasWNSstatushavemuchlessinfluence.WenextusedunweightedandweightedUniFractoexploretherelationshipbetweenskinmicrobiotasam-plesandenvironmentalsamplescollectedateachsite.ThefirstPCoA,basedonunweightedUniFracdistances,groupedskinmicrobiotasamplesandenvironmentalsamplesbysites,withsomeoverlapbetweenthem(Fig.3a).ThesecondPCoAbasedonweightedUniFracdistancesrevealedadifferentpattern,however.Inthatcase,thePCoAplotclearlydistinguishedbetweenenvir-onmentalsamplesandskinmicrobiotasamplescollectedatallsites(Fig.3b),exceptforthreebatsamplesfromAbyssandonefromDaleÂ’s.TheresultsofPCoAsuggestthatthepresence/absenceofOTUsinbatskinsamplesisinfluencedbythelocalenvironmentwithineachhibernaculum.Yet,thesameanalysesalsorevealdiffer-encesinabundancepatternsofsomeOTUsinbatskinsamplescomparedtolocalenvironmentalsamples.WethenusedRDAtoexploretheinfluenceofsite(hi-bernaculum)andsampletype(batvs.environmentsam-ples)onvariationinmicrobialcommunityassemblage.Bothofthesemodelsweresignificant(Table3),butthesitesmodelaccountedforthemostvariationinthedata,explaining18%ofthemicrobialcommunityvariationinbothUniFracdistancesemployed(Table3).Thesampletypemodelonlyexplained5%ofmicrobialvariationforunweightedUniFracdistancesand8%forweighteddis-tances.ThehigherexplanatorypoweroftheweightedUniFracmodelindicatesthatdifferencesobservedbe-tweenenvironmentalsamplesandbatsamplespartlyde-pendontherelativeabundanceofeachtaxonwithinthecorrespondingmicrobialcommunities.ThisisconsistentwithpatternsrevealedbythePCoAplots(Fig.3). ab Fig.2PrincipalcoordinateanalysisofM.lucifugusskinmicrobiotainWNS-positiveandWNS-negativesites.aPrincipalcoordinateanalysisofunweightedUniFracdistances.bPrincipalcoordinateanalysisofweightedUniFracdistances.EachpointrepresentsasamplefromanindividualbathibernatinginoneofthesixdifferenthibernaculathatdifferedinWNSstatusLemieux-Labontetal.Microbiome (2017) 5:115 Page7of14

PAGE 8

Weusedapartialdb-RDAtobetterunderstandthere-lationshipbetweenthetwoexplanatorymodels,andtheirinfluenceonthecompositionofmicrobialcommu-nities.Inbothcases,whenvariationofonemodelwascontrolledforusingpartialdb-RDA,theabilityofthemodelstoexplainvariationinmicrobialcommunitycompositionwasslightlyincreasedby1%(Table3).Theseresultssuggestthatbothsampletypeandsitesmodelshadanon-redundantinfluenceonmicrobialcommunityvariationandthatthelocalenvironmentisanimportantfactorexplainingskinmicrobiotapatternsofhibernatingbatsinourstudysites.TaxonomicindicatorsofWNSstatusWefoundthatthemostabundantbacterialtaxaweresharedamongallhibernacula,butwealsoidentifiedindicatortaxapresentmoreoftenandmoreabundantatparticularsites.Atthephylumlevel,thedominanttaxaac-countingtogetherfor86to98%ofoverallprofilesinagivencavewereActinobacteria(23to53%),Proteobacteria(24to51%),andBacteroidetes(6to38%)(Additionalfile5).Attheclasslevel,thesixprincipaltaxaaccountingfor80to98%ofthetotalabundancewereActinobacteria,Gammaproteobacteria,Flavobacteriia,Sphingobacteriia,Alphaproteobacteria,andBetaproteo-bacteria(Additionalfile6).GeneralistgenerasuchasArthrobacter,Chryseobacter-ium,Flavobacterium,Intrasporangiaceae,Pedobacter,Mycoplana,Pseudonocardiaceae,Ralstonia,Rhodococ-cus,Sinobacteraceae,andSphingobacteriumwereidenti-fiedfromallsites(Additionalfile7).Significantrepresentativeswerefoundamongthe26moreabun-danttaxarepresentingmorethan1%ofthetotalcom-positionprofile(Fig.4,Additionalfile8).AmongthemoreabundanttaxaidentifiedatthreeWNS-positivesitesinQubec,onlyPseudomonasandAcinetobacterwereindicatorsofonesite(Fig.4,Additionalfile8).Ontheotherhand,themoreabundanttaxaatthreeWNS-negativesitesinManitoba,Knoellia,Brucellaceae:Other,Microbacterium,andPseudomonadaceaewereallindica-torsoftheMicrowavesite(Fig.4,Additionalfile8).ThelargestindicatorvaluewasobtainedforNitrosovibrioattheAbysssite.CytophagaceaeandFlavobacteriaceaewerealsoassociatedwithAbyss.Representativetaxawereidentifiedfromallhibernacula,exceptforEmerald,LaflcheandDaleÂ’s.WecomparedskinmicrobiotaprofilesbasedonWNSstatusinordertohighlightpossibledifferencesinmicrobialcompositionrelatedtothefungaldisease.Here,again,someofthemostabundanttaxasuchasPedobacterandIntrasporangiaceaewerenotsignifi-cantrepresentativesastheywereidentifiedinbothareasinsimilarrelativeabundance(Additionalfile7).However,alargenumberofsignificantindicatorsweredetected,somewithhighindicatorvalues.AtWNS-negativesitesinManitoba,significantindicatorswereKnoellia,Brucellaceae:Other,Nitrosovibrio,Flavobacteriaceae,Enterobacteriaceae,Microbacterium,Sphingobacterium,Cytophagaceae,Chryseobacterium,andXanthomonadaceae(Fig.5,Additionalfile9).Ontheotherhand,significantindicatorsofWNS-positivesitesinQubecwereRalstonia,Janthinobacterium,Rhodococcus,Micrococcaceae,andPseudomonas(Fig.5,Additionalfile9).DiscussionWecomparedtheskinmicrobiotaofbatsfromWNS-positiveandWNS-negativesitestobetterunderstandtheroleofthemicrobiotaasafactorinthehost-pathogeninteractionassociatedwithWNS.WefoundsupportforthehypothesesthatWNShascontributedchangesintheskinmicrobiotaforbatsthatarepersist-inginaffectedregionsandthattheskinmicrobiotaisstronglyinfluencedbythelocalenvironmentwithinhibernacula.Althoughwecannotruleouttheroleof Table2db-RDAofunweightedandweightedUniFracdistancesofM.lucifugusskinmicrobiotasamplesModelTestAdjustedR2Fstatisticdb-RDAunweightedUniFracWNSGlobaltest0.0850***7.0354Partialtest:sites00Partialtest:clusters00SitesGlobaltest0.2204***4.6754Partialtest:clusters00Partialtest:WNS0.1354***3.7798ClustersGlobaltest0.2840***2.9829Partialtest:sites0.0637***1.6676Partialtest:WNS0.1989***2.4813db-RDAweightedUniFracWNSGlobaltest0.1413***11.7000Partialtest:sites00Partialtest:clusters00SitesGlobaltest0.2552***5.4543Partialtest:clusters00Partialtest:WNS0.1138***3.4457ClustersGlobaltest0.3039***3.1833Partialtest:sites0.0491***1.5286Partialtest:WNS0.1623***2.2433WNS,sites,andclustersmodelredundantvariationwithUniFracbetadiversityvariationamongM.lucifugusskinmicrobiota.Globaltestforonemodelredundantvariationonmicrobialcommunitywhereasthepartialtestforthemodelcontrollingvariationfromtheothermodel.***p0.001.Totalinertiaofresponsevariablematrixis0.21286forunweightedUniFracdb-RDAand0.10555forweightedUniFracdb-RDALemieux-Labontetal.Microbiome (2017) 5:115 Page8of14

PAGE 9

geographicvariationconfoundedherewithWNSstatuswebelievethatourresultsaremoreconsistentwiththeproposalthatWNShasledtoashiftinthemicrobiotaofbatsinhabitingWNS-positivesites.Forone,ouranalysesbasedonweightedUniFracdistancesshownoclusteringenvironmentalsamplesbyprovincewhenanalyzedtogetherwithbatsamples,supportingthatre-gionisnotamajordriverofmicrobiotacommunities.Second,becausePdaffectsandinteractswiththeskinsodirectlyandbecausehigherlevelsofbacteriaknowntoinhibitPdgrowthinvitro[36,46]andinvivo[37]wereobserved,itseemsmorelikelythatWNSstatus,andnotgeography,explainsthecompositionalpatterns.Aspre-dictedbyourfirsthypothesis,thediversityoftheskinmicrobiotawasindeedsmalleratWNS-positivesitescomparedtoWNS-negativesites,whichisconsistentwithashiftinmicrobiotacausedbyPd.Inaddition,WNSstatuswasastrongpredictorofvariationinShannondiversityvaluesacrosssites.Apreviousstudyontricoloredbats(P.subflavus)affectedbyWNSalsorevealedatrendforlowerdiversityvaluesatWNS-positivesites[38],asshownhereforlittlebrownbatspersistingafterWNSinvasion.PhylogeneticbetadiversityanalysiswasalsoconsistentwithselectiononthemicrobiotabyPdandWNS.Futurestudies,asses-singdiversityofthemicrobiotaonbatsfromthesamesites,beforeandafterPdinvasion,wouldhelpresolvetheWNSinfluenceinmicrobiotadiversitypatternsofpersistingbats.Atthecompositionallevel,theskinmicrobiotaofhi-bernatinglittlebrownbatsisdominatedbytheclassesActinobacteria,Gammaproteobacteria,Flavobacteriia,Alphaproteobacteria,andBetaproteobacteria,apatternconsistentwithpreviousinvestigationoftheskinmicrobiotainseveralspeciesofbats[42]andparticularlyM.lucifugus[40].OuranalysisalsoidentifiedSphingobacteriiaasapredominantclass.IndValanalysisrevealedthatoneinterestinggenus,Rhodococcus,was ab Fig.3PrincipalcoordinateanalysiscomparinglocalenvironmentsitessamplesandM.lucifugusskinmicrobiotainCanada.aPrincipalcoordinateanalysisofunweightedUniFracdistances.bPrincipalcoordinateanalysisofweightedUniFracdistances.EachpointrepresentsasinglesampleLemieux-Labontetal.Microbiome (2017) 5:115 Page9of14

PAGE 10

significantlymoreabundantinskinmicrobiotasamplescollectedatWNS-positivesitesinQubec.Thisgenushaspreviouslybeenidentifiedonbats[38,76]andisknownforitsantifungalactivity[77,78].Mostinterest-ing,avolatileorganicchemicalproducedbyR.rhodo-chrousstrainDAP96253hasbeenshowntoinhibitPdgrowthinvitro[46].Severalothergenera,reportedasantifungalagents,werealsoidentifiedassignificantindi-catorsofbatsamplescollectedatWNS-positivesites.Namely,PseudomonaswasenrichedatallWNS-positivesites,whereasAcinetobacterwasenrichedatasingleWNS-positivesite.Bothtaxaareknownfortheiranti-fungalactivity[79,80]andhavebeenpreviouslyidenti-fiedontheskinofNorthAmericanbats[38,40].Moreover,onestrainofPseudomonasfluorescenshasbeenshowntoinhibitPdgrowthinvitroandreducedis-easeseverityandimprovesurvivalofbatswithWNSinalaboratorychallengeexperiment[36,37].Anotherlesser-knownantifungalbacterialgenus,Janthinobacter-ium,wasalsoidentifiedasasignificantrepresentativeatWNS-positivesitesinQubec.SomespeciesfromthesamegenusisolatedfromtheskinofwildamphibiansconferresistanceagainstthefungalpathogenBatracho-chytriumdendrobatidis[35,79].Enrichmentofmultipletaxawithpotentialantifungalandanti-PdactivityinbatspersistingfollowingWNSinvasionisconsistentwithoursecondhypothesisthattheskinmicrobiotaofbatspro-videsamechanismforresistanceto,ortoleranceof,Pdinfection.Furtherstudiesshouldfocusonanyfunctionalinfluenceofthesebacteriaonthehost-pathogeninter-actionbetweenbatsandPd.Consistentwithourthirdhypothesis,wefoundthattheskinmicrobiotaofhibernatinglittlebrownbatsisre-latedtothemicrobialcommunitycompositionofthenearbyenvironmentalsubstrates.Thatis,bacterialivingonbatsandbacterialivingonadjacentcavewallsareverylikelyexchangedbycontact.However,batskinsam-plesandlocalenvironmentalsampleswerebynomeansidenticalintheircompositionalprofiles,indicatingthatmicrobialcommunitiesontheskinofhibernatingbatsareprobablynotregulatedinthesamewayasinthe Fig.4M.lucifugusskinmicrobiotataxaindicatorofthesixhibernaculaofdifferentWNSstatusinCanada.ThesitesfromWNS-negative(Manitoba,Canada)andWNS-positive(Qubec)regionsarepresented.ThesignificantindicatorswereidentifiedbyIndValanalysisamongthe26moreabundanttaxarepresentingmorethan1%oftotalabundance.Starsindicatehibernaculawithsignificantrepresentativetaxa.*IndVal>0.60,**IndValL0.75,***IndVal0.89 Table3db-RDAofunweightedandweightedUniFracdistancesamonglocalenvironmentandbatskinmicrobiotasamplesModelTestAdjustedR2Fstatisticdb-RDAunweightedUniFracSitesGlobal0.1804***4.3457Partial:type0.1934***4.8220TypeGlobal0.0473***4.7786Partial:sites0.0604***6.6491db-RDAweightedUniFracSitesGlobal0.1763***4.2544Partial:type0.1892***4.8584TypeGlobal0.0752***7.1866Partial:sites0.0881***9.5141SitesandtypemodelredundantvariationwithUniFracbetadiversityvariationamongM.lucifugusskinmicrobiotaandsiteenvironmentalmicrobialassemblage.Globaltestforonemodelredundantvariationonmicrobialcommunitywhereasthepartialtestforthemodelcontrollingvariationfromtheothermodel.***p0.001.Totalinertiaofresponsevariablematrixis0.21818fordb-RDAunweightedUniFracand0.22375db-RDAweightedUniFracLemieux-Labontetal.Microbiome (2017) 5:115 Page10of14

PAGE 11

environment(Additionalfile10).Theseresultsarecon-sistentwithotherstudiesofbats[40]orfrogs[44,45]showingthatskinmicrobiotaassemblagesdonotexactlymirrorthemicrobialcommunitiesintheimmediateen-vironment.Althoughwedidnotdetectanyrelatedvari-ationwithinsitesofmicrobialcommunitiesoftheskinandthatofthesubstrates,wefoundthatindividualbatsstronglydifferacrosssites.Moreover,thetendencyforhibernatinglittlebrownbatstocluster,ofteninlargegroupsofhundredstothousandsofindividuals,islikelytoreducevariationinthemicrobiotaamongindividualsbecauseoftransferwithinclusters,asshownbyouranalysis.Homogenizationoftheskinmicrobiotabyclosecontactamongindividualshasalsobeenobservedinpreviousstudiesofbatsandhumans[41,81].Takentogether,theseresultssuggestthatbatpopulationscoulddifferintheirsusceptibilitytoWNSdependingonthemicrobialcommunityintheirimmediateenvironment,theirrelianceonclusteringbehavior,andthepotentialforclusteringtohomogenizethebacterialcommunity.Inthisstudy,wedidnotattempttoquantifythepotentialinfluenceofabioticvariables,suchaspH,temperature,andhumidity,andconsideredthesefactorsaspossiblecontributorstositeeffects.Itwouldbeinterestinginfuturestudiestoanalyzethesefactorsseparatelytounderstandtheirrelativeinfluenceonthemicrobialcommunityonbatsandintheenvironment.Temporalvariationmayalsohaveinfluencedthecompositionalpatternsobservedinthisstudy,butourexperimentalprotocolwasdesignedtoreducethiseffectasmuchaspossible.Allsitesweresampledwithinarelativelyshortperiodoftime(lessthan3.5months),andweavoidedthestartofhibernationwhentheexperi-enceofbatspriortohibernationmightbeexpectedtomorestronglyinfluencetheirskinmicrobiota.Moreover,themicrobiotaonbatsorintheenvironmentforthesingleWNS-positivesitewesampledinMarch(Emerald)wasnotdifferentfromthetwoWNS-positivesiteswesampledinNovember(LamesandLaflche).ConclusionsThisstudyhighlightstheroleofskinmicrobiotaforwild-lifepopulationhealth,conservation,andmanagementinthefaceofemerginginfectiousdiseases.Theenrichmentofpotentiallybeneficialbacteriainskinmicrobiotasam-plescollectedatWNS-positivehibernaculaisanencour-agingdiscoveryfortheprospectofbatpopulationrecoveryafterWNSbecomesendemicinagivenregion.Thisfindinghighlightsthepotentialvalueofmanagementactionsthatmightencouragetransmission,growth,andestablishmentofbeneficialbacterialtaxaonbatsandwithinhibernacula[37].However,ourfindingsalsohigh-lightapotentialriskofsomeproposedmanagementac-tions.ConsiderablefundingandtimeiscurrentlybeingdevotedtodevelopmentandtestingofpotentialchemicalorbiologicaltreatmentsforWNSthatcouldbeappliedtohibernatingbatsorhibernaculumsubstrates.OurresultsnotonlysupportpreviousworkhighlightingthepotentialofsomebacteriaasbiologicalcontrolagentsforPd(e.g.,[36,37,46])butalsohighlightapotentialriskofbiologicalorchemicaltreatments.Treatmentsthatdisrupttheskinmicrobiotaorattenuateselectionforabeneficialskincommunitycouldcausemoreharmthangoodforrecov-eryofbatpopulationsandtheestablishmentofstable, Fig.5M.lucifugusskinmicrobiotataxaindicatorofWNS-positive(Qubec)andWNS-negativeregions(Manitoba)inCanada.Significantindicatorswerefoundamongthe26moreabundanttaxarepresentingmorethan1%oftotalabundancewithIndValanalysis.Starsindicateregionswithsignificantrepresentativetaxa.**IndVal0.75,***IndVal0.89Lemieux-Labontetal.Microbiome (2017) 5:115 Page11of14

PAGE 12

long-termresistancetoWNSinthewild.Thus,anim-portantcomponentoftestinganypotentialtreatmentforWNSshouldbetoconfirmthatitisselectiveforPdandhasminimalnegativeimpactsonthewholenon-targetmicrobiotaonbatsorinhibernacula.AdditionalfilesAdditionalfile1:Clustersofbatssampledwithineachhibernaculumandcodedasdummyvariablesfordb-RDAanalysis.Additionalfile2:Positivecontrolmockcommunityanalysis.Sequencesetcomparisonsofthemockcommunitytowhatisexpected.File1showsthematchingsequencesandrelatedtaxaidentified.File2showsthetaxacompositionofthemockinrelativeabundance,thematchingtaxaatthegenusorfamilylevel,andthefalsepositivetaxa.Additionalfile3:Maintaxarelativeabundance(>0.1%)innegativecontrolsamples.File1presentsDNAextractioncontrolsamples,file2thelibrarycontrol,file3thenegativesitecontrols,andfile4presentsallcontrolstogether.Additionalfile4:Rarefactioncurvesofalphadiversitycalculatedonmultiplerarefieddatatableforeachofthe66batskinmicrobiotasamplesand11environmentalsamples.(A)Shannondiversityofbatskinsamples.(B)Overallrichness(OTUsobserved)ofbatskinsamples.(C)Shannondiversityofenvironmentalsamples.(D)Overallrichness(OTUsobserved)ofenvironmentalsamples.Additionalfile5:Mainphylaidentifiedinbatskinmicrobiotasamples.The8moreabundantphylaacrossallhibernaculaareprovided.Additionalfile6:Mainclassesidentifiedinbatskinmicrobiotasamples.The6moreabundantclassesacrossallhibernaculaareprovided.Additionalfile7:Majorbacterialtaxaidentifiedinbatskinmicrobiotasamples.The16moreabundanttaxaacrossallhibernaculaareprovided.Starsrepresentsignificantindicatortaxa.*IndVal<0.50,**IndVal0.50,***IndVal0.89.Additionalfile8:M.lucifugusskinmicrobiotataxaindicatortestandrelatedassociationmeasure(A,B)ofsixhibernaculumgroupswithdifferentWNSstatusinCanada.Indicatorvaluetestswerecomputedwiththemultipatt()functionoftheindicspeciespackageinR.OnlytaxawithA0.4wereretainedasindicators.A,thespecificity,istheprobabilitythatasitebelongstothegroupgiventhefactthatthespeciesisfoundandB,thefidelity,istheprobabilityoffindingagiventaxonwhenthesitesbelongtothatgroup.*p0.05,**p0.01,***p0.001.Additionalfile9:M.lucifugusskinmicrobiotataxaindicatorandrelatedassociationmeasure(A,B)ofWNS-positive(Qubec)andWNS-negative(Manitoba)sitesinCanada.Indicatorvaluetestswerecomputedwiththemultipatt()functionoftheindicspeciespackageinR.OnlytaxawithA0.4wereretainedasindicators.A,thespecificity,istheprobabilitythatasitebelongstothegroupgiventhefactthatthespeciesisfoundandB,thefidelity,istheprobabilityoffindingagiventaxonwhenthesitesbelongtothatgroup.*p0.05,**p0.01,***p0.001.Additionalfile10:OTUtableresultingfromtheanalysisof66batskinmicrobiotasamplesand11environmentalsamples.AbbreviationsPd:Pseudogymnoascusdestructans;WNS:White-nosesyndromeAcknowledgementsWethankPascalSamson,JocelynCaron,ValrieSimard,ArianeMass,GuillaumeTremblay,KaleighNorquay,AnaBreit,DrewSippell,ManonGagn,andArmandYargeaufortheirassistancewiththesampling,JulieMarleauforhelpwiththeexperiments,NicolasTromasforhelpwiththesequencetreatment,andQuinnFletcherforassistanceinthedataanalysis.ThefollowingreagentwasobtainedthroughBEIResources,NIAID,NIHaspartoftheHumanMicrobiomeProject:GenomicDNAfromMicrobialMockCommunityB(Even,LowConcentration),v5.1L,for16SrRNAGeneSequencing,HM-782D.FundingThisresearchwasfundedbyDiscoveryGrantsfromtheNaturalSciencesandEngineeringResearchCouncil(NSERCCanada)OGP0155251toFJLandRGPIN-2015-04437toCKRWandexcellenceawardsfromtheQuebecCentreforBiodiversityScience.Thefundershadnoroleinthestudydesign,datacollectionandanalysis,decisiontopublish,orpreparationofthemanuscript.AvailabilityofdataandmaterialsThedatasetsgeneratedandanalyzedduringthecurrentstudyareavailableintheFigsharerepository.RawsequencedataandmetadataareavailableatDOI:https://figshare.com/s/623a1e47b4bed20459a7andDOI:https://figshare.com/s/74d9497a792f9c0c76df.Authors’contributionsVLLdesignedtheexperiments,performedthesampling,carriedouttheexperiments,analyzedthedata,andwrotethemanuscript.CKRWandAShelpedinthesamplinglogisticandwrotethemanuscript.FJLdesignedtheexperimentsandwrotethemanuscript.Allauthorsreadandapprovedthefinalmanuscript.EthicsapprovalandconsenttoparticipateAllmethodswereapprovedbytheAnimalWelfareandEthicsCommitteeatUniversitdeMontral(ProtocolNumber#16-015)andtheUniversityofWinnipegAnimalCareCommittee(ProtocolNumberAEO5639).ConsentforpublicationNotapplicable.CompetinginterestsTheauthorsdeclarethattheyhavenocompetinginterests.Publisher’sNoteSpringerNatureremainsneutralwithregardtojurisdictionalclaimsinpublishedmapsandinstitutionalaffiliations.Authordetails1DpartementdeSciencesBiologiques,UniversitdeMontral,CP6182,SuccursaleCentre-ville,Montral,QubecH2V2S9,Canada.2Directiondel’expertisesurlafauneterrestre,l’herptofauneetl’avifaune,MinistredesForts,delaFauneetdesParcs,Qubec,Canada.3DepartmentofBiologyandCentreforForestInterdisciplinaryResearch,UniversityofWinnipeg,Winnipeg,Manitoba,Canada.4QuebecCentreforBiodiversityScience,CP6182,SuccursaleCentre-ville,Montral,QubecH2V2S9,Canada.Received:27March2017Accepted:28August2017 References1.FisherMC,HenkDA,BriggsCJ,BrownsteinJS,MadoffLC,McCrawSL,etal.Emergingfungalthreatstoanimal,plantandecosystemhealth.Nature.2012;484(7393):186–94.https://doi.org/10.1038/nature10947.2.DaszakP,CunninghamAA,HyattAD.Emerginginfectiousdiseasesofwildlife-threatstobiodiversityandhumanhealth.Science.2000;287(5452):443–9.https://doi.org/10.1126/science.287.5452.443.3.JonesKE,PatelNG,LevyMA,StoreygardA,BalkD,GittlemanJL,etal.Globaltrendsinemerginginfectiousdiseases.Nature.2008;451(7181):990–3.https://doi.org/10.1038/nature06536.4.BoylesJG,CryanPM,McCrackenGF,KunzTH.Economicimportanceofbatsinagriculture.Science.2011;332(6025):41–2.https://doi.org/10.1126/science.1201366.5.MaineJJ,BoylesJG.Batsinitiatevitalagroecologicalinteractionsincorn.ProcNatlAcadSci.2015;112(40):12438–43.https://doi.org/10.1073/pnas.1505413112.6.GargasA,TrestMT,ChristensenM,VolkTJ,BlehertDS.Geomycesdestructanssp.nov.associatedwithbatwhite-nosesyndrome.Mycotaxon.2009;108(1):147–54.https://doi.org/10.5248/108.147.7.LorchJM,MeteyerCU,BehrMJ,BoylesJG,CryanPM,HicksAC,etal.ExperimentalinfectionofbatswithGeomycesdestructanscauseswhite-Lemieux-Labontetal.Microbiome (2017) 5:115 Page12of14

PAGE 13

nosesyndrome.Nature.2011;480:376–8.https://doi.org/10.1038/nature10590.8.U.S.Fish&WildlifeService.NorthAmericanbatdeathtollexceeds5.5millionfromwhite-nosesyndrome.In:Newsrelease.2012.http://www.batcon.org/pdfs/USFWS_WNS_Mortality_2012_NR_FINAL.pdf.Accessed28Dec2016.9.MeteyerCU,BucklesEL,BlehertDS,HicksAC,GreenDE,Shearn-BochslerV,etal.Histopathologiccriteriatoconfirmwhite-nosesyndromeinbats.JVetDiagnInvestig.2009;21(4):411–4.https://doi.org/10.1177/104063870902100401.10.CryanPM,MeteyerCU,BoylesJG,BlehertDS.Wingpathologyofwhite-nosesyndromeinbatssuggestslife-threateningdisruptionofphysiology.BMCBiol.2010;8:135.https://doi.org/10.1186/1741-7007-8-135.11.CorteseTA,NicollPA.Invivoobservationsofskinappendagesinthebatwing.JInvestDermatol.1970;54:1–10.https://doi.org/10.1111/1523-1747.ep12551469.12.SiskMO.Astudyofthesudoriparousglandsofthelittlebrownbat,Myotislucifuguslucifugus.JMorphol.1957;101:425–55.https://doi.org/10.1002/jmor.1051010303.13.GriceEA,SegreJA.Theskinmicrobiome.NatRevMicrobiol.2011;9(4):244–53.https://doi.org/10.1038/nrmicro2537.14.CzenzeZJ,WillisCKR.Warmingupandshippingout:cuesforarousalandemergenceinhibernatinglittlebrownbats(Myotislucifugus).JCompPhysiolB.2015;185:575–86.https://doi.org/10.1007/s00360-015-0900-1.15.CzenzeZJ,ParkAD,WillisCKR.Stayingcoldthroughdinner:cold-climatebatsrewarmwithconspecificsbutnotsunsetduringhibernation.JCompPhysiolB.2013;183(6):859–66.https://doi.org/10.1007/s00360-013-0753-4.16.JonassonKA,WillisCKR.Hibernationenergeticsoflittlebrownbats.JExpBiol.2012;215:2141–9.https://doi.org/10.1242/jeb.066514.17.ReederDM,FrankCL,TurnerGG,MeteyerCU,KurtaA,BritzkeER,etal.Frequentarousalfromhibernationlinkedtoseverityofinfectionandmortalityinbatswithwhite-nosesyndrome.PLoSOne.2012;7(6):e38920.https://doi.org/10.1371/journal.pone.0038920.18.WarneckeL,TurnerJM,BollingerTK,LorchJM,MisraV,CryanPM,etal.InoculationofbatswithEuropeanGeomycesdestructanssupportsthenovelpathogenhypothesisfortheoriginofwhite-nosesyndrome.ProcNatlAcadSci.2012;109(18):6999–7003.https://doi.org/10.1073/pnas.1200374109.19.MooreMS,ReichardJD,MurthaTD,ZahediB,FallierRM,KunzTH.Specificalterationsincomplementproteinactivityoflittlebrownmyotis(Myotislucifugus)hibernatinginwhite-nosesyndromeaffectedsites.PLoSOne.2011;6(11):e27430.https://doi.org/10.1371/journal.pone.0027430.20.BoumaHR,CareyHV,KroeseFG.Hibernation:theimmunesystematrest?JofLeukocBiol.2010;88(4):619–24.https://doi.org/10.1189/jlb.0310174.21.GeiserF.Metabolicrateandbodytemperaturereductionduringhibernationanddailytorpor.AnnuRevPhysiol.2004;66:239–74.https://doi.org/10.1146/annurev.physiol.66.032102.115105.22.FrickWF,PuechmailleSJ,WillisCKR.White-nosesyndromeinbats.In:VoigtCC,KingstonT,editors.BatsintheAnthropocene:conservationofbatsinachangingworld.Cham,Heidelberg,NewYork,Dordrecht,London:SpringerInternationalPublishing;2016.p.245–62.23.TurnerGG,ReederDM,ColemanJTH.Afive-yearassessmentofmortalityandgeographicspreadofwhite-nosesyndromeinNorthAmericanbats,withalooktothefuture.BatResNews.2011;52:13–27.24.LangwigKE,FrickWF,BriedJT,HicksAC,KunzTH,KilpatrickAM.Sociality,density-dependenceandmicroclimatesdeterminethepersistenceofpopulationssufferingfromanovelfungaldisease,white-nosesyndrome.EcolLett.2012;15(9):1050–7.https://doi.org/10.1111/j.1461-0248.2012.01829.x.25.WillisCKR,WilcoxA.Hormonesandhibernation:possiblelinksbetweenhormonesystems,winterenergybalanceandwhite-nosesyndromeinbats.HormBehav.2014;66(1):66–73.https://doi.org/10.1016/j.yhbeh.2014.04.009.26.PuechmailleSJ,FrickWF,KunzTH,RaceyPA,VoigtCC,WibbeltG,etal.White-nosesyndrome:isthisemergingdiseaseathreattoEuropeanbats?TrendsEcolEvol.2011;26:570–6.https://doi.org/10.1016/j.tree.2011.06.013.27.COSEWIC.COSEWICAssessmentandStatusReportonthelittlebrownMyotislucifugus,NorthernMyotisseptentrionalisandtri-coloredbatPerimyotissubflavusinCanada.CommitteeontheStatusofEndangeredWildlifeinCanada.2013.http://www.registrelep-sararegistry.gc.ca/document/default_e.cfm?documentID=1323.Accessed4Jan2017.28.FrickWF,PuechmailleSJ,HoytJR,NickelBA,LangwigKE,FosterJT,etal.DiseasealtersmacroecologicalpatternsofNorthAmericanbats.GlobEcolBiogeogr.2015;24(7):741–9.https://doi.org/10.1111/geb.12290.29.DobonyCA,HicksAC,LangwigKE,vonLindenRI,OkoniewskiJC,RainboltRE.Littlebrownmyotispersistdespiteexposuretowhite-nosesyndrome.JofFishWildlManag.2011;2(2):190–5.https://doi.org/10.3996/022011-JFWM-014.30.MasloB,ValentM,GumbsJF,FrickWF.Conservationimplicationsofamelioratingsurvivaloflittlebrownbatswithwhite-nosesyndrome.EcolAppl.2015;25(7):1832–40.https://doi.org/10.1890/14-2472.1.31.LangwigKE,HoytJR,PariseKL,FrickWF,FosterJT,KilpatrickAM.Resistanceinpersistingbatpopulationsafterwhite-nosesyndromeinvasion.PhilosTransRSocBBiolSci.2017;372(1712):20160044.https://doi.org/10.1098/rstb.2016.0044.32.PricePB.Thebacteriologyofnormalskin;anewquantitativetestappliedtoastudyofthebacterialfloraandthedisinfectantactionofmechanicalcleansing.JInfectDis.1938;63(3):301–18.http://www.jstor.org/stable/30088420.33.RothRR,JamesWD.Microbialecologyoftheskin.AnnuRevMicrobiol.1988;42:441–64.https://doi.org/10.1146/annurev.mi.42.100188.002301.34.BeldenLK,HarrisRN.Infectiousdiseasesinwildlife:thecommunityecologycontext.FrontEcolEnviron.2007;5(10):533–9.https://doi.org/10.1890/060122.35.BruckerRM,HarrisRN,SchwantesCR,GallaherTN,FlahertyDC,LamBA,etal.Amphibianchemicaldefense:antifungalmetabolitesofthemicrosymbiontJanthinobacteriumlividumonthesalamanderPlethodoncinereus.JChemEcol.2008;34(11):1422–9.https://doi.org/10.1007/s10886-008-9555-7.36.HoytJR,ChengTL,LangwigKE,HeeMM,FrickWF,KilpatrickAM.BacteriaisolatedfrombatsinhibitthegrowthofPseudogymnoascusdestructans,thecausativeagentofwhite-nosesyndrome.PLoSOne.2015;10(4):e0121329.https://doi.org/10.1371/journal.pone.0121329.37.ChengTL,MayberryH,McGuireLP,HoytJR,LangwigKE,NguyenH,etal.Efficacyofaprobioticbacteriumtotreatbatsaffectedbythediseasewhite-nosesyndrome.JApplEcol.2016;https://doi.org/10.1111/1365-2664.12757.38.LueschowSR.EffectofPseudogymnoascusdestructansonmicrobialcommunitycompositiononbats.Ph.D.diss.,WesternIllinoisUniversity.2015.http://search.proquest.com/docview/1758623705?accountid=12543.Accessed30Dec2016.39.Romano-BertrandS,Licznar-FajardoP,ParerS,Jumas-BilakE.Impactdel’environnementsurlesmicrobiotes:focussurl’hospitalisationetlesmicrobiotescutansetchirurgicaux.RevueFrancophonedesLaboratoires.2015;2015(469):75–82.https://doi.org/10.1016/S1773-035X(15)72824-8.40.AvenaCV,ParfreyLW,LeffJW,ArcherHM,FrickWF,LangwigKE,etal.Deconstructingthebatskinmicrobiome:influencesofthehostandtheenvironment.FrontMicrobiol.2016;7:1753.https://doi.org/10.3389/fmicb.2016.01753.41.Lemieux-LabontV,TromasN,ShapiroBJ,LapointeFJ.Environmentandhostspeciesshapetheskinmicrobiomeofcaptiveneotropicalbats.PeerJ.2016;4:e2430.https://doi.org/10.7717/peerj.2430.42.WinterAS,KimbleJC,YoungJM,BuecherDC,ValdezEW,HathawayJJM,etal.ExternalbacterialdiversityonbatsinthesouthwesternUnitedStates:changesinbacterialcommunitystructureaboveandbelowground.PeerJPreprints.2016;4:e2493v1.https://doi.org/10.7287/peerj.preprints.2493v1.43.KooserA,KimbleJC,YoungJM,BuecherDC,ValdezEW,Porras-AlfaroA,etal.ExternalmicrobiotaofwesternUnitedStatesbats:doesitmatterwhereyouarefrom?bioRxiv.2015;017319.doi:https://doi.org/10.1101/017319.44.LoudonAH,WoodhamsDC,ParfreyLW,ArcherH,KnightR,McKenzieV,etal.Microbialcommunitydynamicsandeffectofenvironmentalmicrobialreservoirsonred-backedsalamanders(Plethodoncinereus).ISMEJ.2014;8(4):830–40.https://doi.org/10.1038/ismej.2013.200.45.WalkeJB,BeckerMH,LoftusSC,HouseLL,CormierG,JensenRV,etal.Amphibianskinmayselectforrareenvironmentalmicrobes.ISMEJ.2014;8(11):2207–17.https://doi.org/10.1038/ismej.2014.77.46.CornelisonCT,KeelMK,GabrielKT,BarlamentCK,TuckerTA,PierceGE,etal.Apreliminaryreportonthecontact-independentantagonismofPseudogymnoascusdestructansbyRhodococcusrhodochrousstrainDAP96253.BMCMicrobiol.2014;14:246.https://doi.org/10.1186/s12866-014-0246-y.47.SalterSJ,CoxMJ,TurekEM,CalusST,CooksonWO,MoffattMF,etal.Reagentandlaboratorycontaminationcancriticallyimpactsequence-Lemieux-Labontetal.Microbiome (2017) 5:115 Page13of14

PAGE 14

basedmicrobiomeanalyses.BMCBiol.2014;12:87.https://doi.org/10.1186/s12915-014-0087-z.48.GlassingA,DowdSE,GalandiukS,DavisB,ChiodiniRJ.InherentbacterialDNAcontaminationofextractionandsequencingreagentsmayaffectinterpretationofmicrobiotainlowbacterialbiomasssamples.GutPathog.2016;8:24.https://doi.org/10.1186/s13099-016-0103-7.49.PreheimSP,PerrottaAR,FriedmanJ,SmilieC,BritoI,SmithMB,etal.Computationalmethodsforhigh-throughputcomparativeanalysesofnaturalmicrobialcommunities.MethodsEnzymol.2013;531:353–70.https://doi.org/10.1016/B978-0-12-407863-5.00018-6.50.CaporasoJG,LauberCL,WaltersWA,Berg-LyonsD,LozuponeCA,TurnbaughPJ,etal.Globalpatternsof16SrRNAdiversityatadepthofmillionsofsequencespersample.ProcNatlAcadSciUSA.2011;108:4516–22.https://doi.org/10.1073/pnas.1000080107.51.PreheimSP,PerrottaAR,Martin-PlateroAM,GuptaA,AlmEJ.Distribution-basedclustering:usingecologytorefinetheoperationaltaxonomicunit.ApplEnvironMicrobiol.2013;79(21):6593–603.https://doi.org/10.1128/AEM.00342-13.52.AlmLaboratories.SmileTrain.https://github.com/almlab/SmileTrain/wiki/.Accessed1July2016.53.EdgarRC.SearchandclusteringordersofmagnitudefasterthanBLAST.Bioinformatics.2010;26(19):2460–1.https://doi.org/10.1093/bioinformatics/btq461.54.CaporasoJG,KuczynskiJ,StombaughJ,BittingerK,BushmanFD,CostelloEK,etal.QIIMEallowsanalysisofhigh-throughputcommunitysequencingdata.NatMethods.2010;7(5):335–6.https://doi.org/10.1038/nmeth.f.303.55.DeSantisTZ,HugenholtzP,LarsenN,RojasM,BrodieEL,KellerK,etal.Greengenes,achimera-checked16SrRNAgenedatabaseandworkbenchcompatiblewithARB.ApplEnvironMicrobiol.2006;72(7):5069–72.https://doi.org/10.1128/AEM.03006-05.56.ShannonCE.Amathematicaltheoryofcommunications.BellSystTechJ.1948;27(3):379–423,623–56.https://doi.org/10.1002/j.1538-7305.1948.tb01338.x.57.HaegemanB,HamelinJ,MoriartyJ,NealP,DushoffJ,WeitzJS.Robustestimationofmicrobialdiversityintheoryandinpractice.ISMEJ.2013;7(6):1092–101.https://doi.org/10.1038/ismej.2013.10.58.RDevelopementCoreTeam.R:alanguageandenvironmentforstatisticalcomputing.Vienna,Austria:RFoundationforStatisticalComputing;2016.59.PinheiroJ,BatesD,DebRoyS,SarkarD,RCoreTeam.nlme:linearandnonlinearmixedeffectsmodels.Rpackageversion3.1–131.2017.https://CRAN.R-project.org/package=nlme.Accessed19July2017.60.LozuponeCA,KnightR.UniFrac:anewphylogeneticmethodforcomparingmicrobialcommunities.ApplEnvironMicrobiol.2005;71(12):8228–35.https://doi.org/10.1128/AEM.71.12.8228-8235.2005.61.LozuponeCA,HamadyM,KelleyST,KnightR.Quantitativeandqualitativediversitymeasuresleadtodifferentinsightsintofactorsthatstructuremicrobialcommunities.ApplEnvironMicrobiol.2007;73(5):1576–85.https://doi.org/10.1128/AEM.01996-06.62.LozuponeCA,LladserME,KnightsD,StombaughJ,KnightR.UniFrac:aneffectivedistancemetricformicrobialcommunitycomparison.ISMEJ.2011;5(2):169–72.https://doi.org/10.1038/ismej.2010.133.63.WeissS,XuZZ,PeddadaS,AmirA,BittingerK,GonzalezA,etal.Normalizationandmicrobialdifferentialabundancestrategiesdependupondatacharacteristics.Microbiome.2017;5:27.https://doi.org/10.1186/s40168-017-0237-y.64.McMurdiePJ,HolmesS.phyloseq:anRpackageforreproducibleinteractiveanalysisandgraphicsofmicrobiomecensusdata.PLoSOne.2013;8(4):e61217.https://doi.org/10.1371/journal.pone.0061217.65.GowerJC.Somedistancepropertiesoflatentrootandvectormethodsusedinmultivariateanalysis.Biometrika.1966;53(3/4):325–38.https://doi.org/10.2307/2333639.66.DrayS,DufourAB.Theade4package:implementingthedualitydiagramforecologists.JStatSoftw.2007;22(4):1–20.10.18637/jss.v022.i04.67.GowerJC,LegendreP.MetricandEuclideanpropertiesofdissimilaritycoefficients.JClassif.1986;3(1):5–48.https://doi.org/10.1007/BF01896809.68.PriceMN,DehalPS,ArkinAP.FastTree2—approximatelymaximum-likelihoodtreesforlargealignments.PLoSOne.2010;5(3):e9490.https://doi.org/10.1371/journal.pone.0009490.69.LegendreP,AndersonMJ.Distance-basedredundancyanalysis:testingmultispeciesresponsesinmultifactorialecologicalexperiments.EcolMonogr.1999;69(1):1–24.https://doi.org/10.2307/2657192.70.OksanenJ,BlanchetG,KindtR,LegendreP,MinchinPR,O’HaraRB,etal.vegan:CommunityEcologyPackage.Rpackageversion2.4–1.2016.http://CRAN.R-project.org/package=vegan.Accessed29Oct2016.71.DaviesPT,TsoMKS.Proceduresforreduced-rankregression.ApplStat.1982;31(3):244–55.https://doi.org/10.2307/2347998.72.EzekielM.Methodsofcorrelationanalysis.NewYorkandLondon:Wiley;1930.73.DufrneM,LegendreP.Speciesassemblagesandindicatorspecies:theneedforaflexibleasymmetricalapproach.EcolMonogr.1997;67(3):345–66.https://doi.org/10.2307/2963459.74.DeCceresM,LegendreP.Associationsbetweenspeciesandgroupsofsites:indicesandstatisticalinference.Ecology.2009;90(12):3566–74.https://doi.org/10.1890/08-1823.1.75.HolmS.Asimplesequentiallyrejectivemultipletestprocedure.ScandJStatist.1979;6(2):65–70.http://www.jstor.org/stable/4615733.76.VoigtCC,CaspersB,SpeckS.Bats,bacteria,andbatsmell:sex-specificdiversityofmicrobesinasexuallyselectedscentorgan.JMammal.2005;86(4):745-749.doi:http://dx.doi.org/10.1644/1545-1542(2005)086[0745:BBABSS]2.0.CO;2.77.ChibaH,AgematuH,KanetoR,TerasawaT,SakaiK,DobashiK,etal.Rhodopeptins(Mer-N1033),novelcyclictetrapeptideswithantifungalactivityfromRhodococcussp.I.Taxonomy,fermentation,isolation,physico-chemicalpropertiesandbiologicalactivities.JAntibiot.1999;52(8):695–9.78.NakayamaK,KawatoHC,InagakiH,NakajimaR,KitamuraA,SomeyaK,etal.Synthesisandantifungalactivityofrhodopeptinanalogues.2.Modificationofthewestaminoacidmoiety.OrgLett.2000;2(7):977–80.https://doi.org/10.1021/ol005630k.79.LauerA,SimonMA,BanningJL,LamBA,HarrisRN.Diversityofcutaneousbacteriawithantifungalactivityisolatedfromfemalefour-toedsalamanders.ISMEJ.2008;2(2):145–57.https://doi.org/10.1038/ismej.2007.110.80.LiuCH,ChenX,LiuTT,LianB,GuY,CaerV,etal.StudyoftheantifungalactivityofAcinetobacterbaumanniiLCH001invitroandidentificationofitsantifungalcomponents.ApplMicrobiolBiotechnol.2007;76(2):459–66.https://doi.org/10.1007/s00253-007-1010-0.81.SongSJ,LauberC,CostelloEK,LozuponeCA,HumphreyG,Berg-LyonsD,etal.Cohabitingfamilymemberssharemicrobiotawithoneanotherandwiththeirdogs.elife.2013;2:e00458.https://doi.org/10.7554/eLife.00458. • We accept pre-submission inquiries  Our selector tool helps you to nd the most relevant journal We provide round the clock customer support  Convenient online submission Thorough peer review Inclusion in PubMed and all major indexing services  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step: Lemieux-Labontetal.Microbiome (2017) 5:115 Page14of14