Citation
Extreme sensitivity to ultraviolet light in the fungal pathogen causing white-nose syndrome of bats

Material Information

Title:
Extreme sensitivity to ultraviolet light in the fungal pathogen causing white-nose syndrome of bats
Series Title:
Nature communications
Creator:
Palmer, Jonathan M.
Drees, Kevin P.
Foster, Jeffrey T.
Lindner, Daniel L.
Publisher:
Springer Nature
Publication Date:
Physical Description:
1 online resource

Subjects

Subjects / Keywords:
Bats -- Pathogens ( lcsh )
Genre:
serial ( sobekcm )

Notes

Abstract:
Bat white-nose syndrome (WNS), caused by the fungal pathogen Pseudogymnoascus destructans, has decimated North American hibernating bats since its emergence in 2006. Here, we utilize comparative genomics to examine the evolutionary history of this pathogen in comparison to six closely related nonpathogenic species. P. destructans displays a large reduction in carbohydrate-utilizing enzymes (CAZymes) and in the predicted secretome (~50%), and an increase in lineage-specific genes. The pathogen has lost a key enzyme, UVE1, in the alternate excision repair (AER) pathway, which is known to contribute to repair of DNA lesions induced by ultraviolet (UV) light. Consistent with a nonfunctional AER pathway, P. destructans is extremely sensitive to UV light, as well as the DNA alkylating agent methyl methanesulfonate (MMS). The differential susceptibility of P. destructans to UV light in comparison to other hibernacula-inhabiting fungi represents a potential “Achilles’ heel” of P. destructans that might be exploited for treatment of bats with WNS.
Original Version:
Volume 9

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:
K26-05397 ( USFLDC DOI )
k26.5397 ( 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

ARTICLEExtremesensitivitytoultravioletlightinthefungal pathogencausingwhite-nosesyndromeofbatsJonathanM.Palmer 1,KevinP.Drees2,JeffreyT.Foster 2,3&DanielL.Lindner 1Batwhite-nosesyndrome(WNS),causedbythefungalpathogen Pseudogymnoascus destructans ,hasdecimatedNorthAmericanhibernatingbatssinceitsemergencein2006. Here,weutilizecomparativegenomicstoexaminetheevolutionaryhistoryofthispathogen incomparisontosixcloselyrelatednonpathogenicspecies. P.destructans displaysalarge reductionincarbohydrate-utilizingenzymes(CAZymes)andinthepredictedsecretome (~50%),andanincreaseinlineage-speci cgenes.Thepathogenhaslostakeyenzyme, UVE1,inthealternateexcisionrepair(AER)pathway,whichisknowntocontributetorepair ofDNAlesionsinducedbyultraviolet(UV)light.ConsistentwithanonfunctionalAER pathway, P.destructans isextremelysensitivetoUVlight,aswellastheDNAalkylatingagent methylmethanesulfonate(MMS).Thedifferentialsusceptibilityof P.destructans toUVlight incomparisontootherhibernacula-inhabitingfungirepresentsapotential “ Achilles ’ heel ” of P.destructans thatmightbeexploitedfortreatmentofbatswithWNS. DOI:10.1038/s41467-017-02441-z OPEN 1CenterforForestMycologyResearch,NorthernResearchStation,USForestService,Madison,WI53726,USA.2DepartmentofMolecular,Cellular,and BiomedicalSciences,UniversityofNewHampshire,Durham,NH03824,USA.3Presentaddress:PathogenandMicrobiomeInstitute,NorthernArizona University,Flagstaff,AZ,USA.CorrespondenceandrequestsformaterialsshouldbeaddressedtoD.L.L.(email: dlindner@fs.fed.us )NATURECOMMUNICATIONS| (2018) 9:35 |DOI:10.1038/s41467-017-02441-z|www.nature.com/naturecommunications1 1234567890

PAGE 2

White-nosesyndromeofbats(WNS)continuestodecimatehibernatingbatpopulationsinNorthAmerica. Thefungaldiseasewas rstdocumentedin2006in easternNorthAmerica(NewYork)andthefungushasadvanced westacrossthecontinent,mostrecentlybeingdetectedonthe WestcoastofNorthAmerica1.WNScanresultin > 90%mortalityinlocalhibernatingbatpopulations2andiscausedbythe psychrophilicfungus Pseudogymnoascusdestructans3 4.Thefungushasastricttemperaturegrowthrangeof~4 – 20Cand thereforecanonlyinfectbatsduringhibernation5.WNSisnota systemicinfection,butratherischaracterizedby P.destructans colonizationoftheskinofhibernatingbats,whichmanifestsas cuppingskinerosionsbasedonhistopathology6.Frequentarousal fromhibernation,depletionoffatreserves,anddehydration appeartocontributetomortalityininfectedindividuals7 – 10. P. destructans hasbeenfoundthroughoutEurasiaandoccasionally causesmildWNSsymptoms;however,nomassmortalityevents havebeenobservedinEurasia11. P.destructans hasspreadina “ bullseye ” patterninNorthAmericaandhasonlybeenfoundin environmentswhereWNS-infectedbatsarefound,strongly suggestingthatthefungusisnotnativetoNorthAmericaand representsaclassicexampleofanintroducedpathogendecimatinganavepopulation12 13. P.destructans isamemberofthe under-studiedPseudeurotiaceaefamily;whilerecentstudieshave resultedinseveraldraftgenomeassemblieswithinthisgroup, muchofthebiologyofthefungalpathogenanditsrelatives remainsunknown. Here,wepresentfunctionallyannotatedgenomesfor P. destructans ,aswellassixcloselyrelatednonpathogenic Pseudogymnoascus species.Comparisonofthepathogenwiththenonpathogenicspeciesprovidesanopportunitytoobtaininsightinto theoriginsandadaptationsofthefungalpathogenofWNS. Results Sequencing,assembly,andannotation .Recentphylogenetic workthataimedatidenti cationandresolutionofcloselyrelated speciesto P.destructans resultedinmovingthisspeciesfrom genus Geomyces to Pseudogymnoascus4.Moreover,samplingfor fungalisolatesfromhibernacularsoilresultedinidenti cationof othercloselyrelated Pseudogymnoascus speciesthatarenot knowntobepathogenic12.Wepreviouslydescribedthehybrid genomeassemblyof P.destructans14;inaddition,wechosesix closelyrelatedspeciesandsequencedtheseisolatesusingIllumina chemistry,resultingin~250coverageforeachgenome(strains listedinSupplementaryTable 1 ).Allseven Pseudogymnoascus genomeswerewithintheexpectedgenomesizeforhaploid ascomycetes(~30 – 36MB)with~50%GCcontent(Table 1 ). Genomeannotationwascompletedusingfunannotatev0.1.8 ( https://github.com/nextgenusfs/funannotate ),resultingin9335 protein-codinggenomemodelsfor P.destructans andarangeof 10,252 – 11,033proteingenemodelsforthenonpathogenic Pseudogymnoascus species(Table 1 ).Genomefunctionalannotation (see “ Methods ” section)wasaddedtoeachgenemodel;complete functionalannotationisavailableinSupplementaryData 1 – 7 Orthologyandevolutionaryphylogeny .Duetotheimportance ofWNS,aswellasinterestincold-tolerantfungi,therehas recentlybeenseveraldraftgenomessequencedinthegenus Pseudogymnoascus .Todelineaterelationshipsbetweenthese genomes,weacquiredall Pseudogymnoascus genomesdeposited inNCBI,aswellasseveraloutgroupspecies: Aspergillusnidulans Botrytiscinerea Fusariumfujikuroi Penicilliumchrysogenum and Neurosporacrassa (SupplementaryTable 2 ).Single-copy BUSCOorthologs15wereextractedusingPhyloma( https:// github.com/nextgenusfs/phyloma )andwereusedtogeneratea maximumlikelihoodphylogenyinRAxMLv8.29(PROTGAMMALG;1000bootstrapreplicates)from822concatenatedgene models.Toestimatetimeofevolutionarydivergence,nodecalibrationsfromBeimfordeetal.16wereusedinr8sv1.8017; Leotiomycetes – Sordariomycetes267 – 430MYA,Eurotiomycetes 273 – 537MYA,Sordariomycetes207 – 339MYA,andPezizomycotina400 – 583MYA(SupplementaryData 8 ).Thisanalysis illustratesthatthenonpathogenic Pseudogymnoascus species sequencedhereareamongtheclosestknownrelativesof P. destructans (Fig. 1 ).Additionally,thesedatasuggestedthatthe lastknowncommonancestorof P.destructans diverged approximately23.5MYA.Theoldestknownchiropteranspecies inthefossilrecord, Palaeochiropteryx ,wasestimatedtohavelived 50 – 40MYA18 19,whilethemostrecentadaptivespeciationof Eurasian Myotis batspeciesoccurred9 – 6MYAandtheNorth American Myotis speciesemergedmorerecentlyat6 – 3.2MYA20. Thus,Eurasian Myotis specieswerepresentatthetimewhen P. destructans divergedfromitsrelatives,suggestingthatitcould havecoevolvedalongsidemodern-dayEurasianchiropteran species. ProteinOrtho521wasusedtoidentify3949single-copy orthologousgroupsbetweenseven Pseudogymnoascus species usedinthisstudy,representingonly42.3%oftheprotein-coding genesin P.destructans .Relationshipsamongthesespecieswere inferredfromamaximumlikelihoodRAxMLphylogenybasedon theconcatenatedalignmentof500orthologousproteinsusing Botrytiscinerea asanoutgroup(Fig. 2 ).Theorthologous proteomeanalysisidenti ed1934uniqueproteinsin P. destructans thatwerenotfoundinanyofthenonpathogenic Pseudogymnoascus species(Fig. 2 andTable 1 ),making P. destructans thespecieswiththemostuniqueproteinsinour study.Toidentifyorthologousproteinsunderpositiveselection, d N /d S ratiosforallorthologousgroupswerecalculatedusingthe codemlM0modelfromPAML22.LikelihoodratiotestscomparingtheM1/M2models,aswellasM7/M8modelswith signi canceat < 0.05werecalculatedtovalidatetheestimated d N/ d S ratiosgreaterthan1,ratiosthatsuggestpositiveselectionis occurring22.Forty-sixorthologousgroupsdisplayedevidenceof positiveselection.Ofthoseorthologousgroupsunderpositive Table1Genomesummarystatisticsof P.destructans andnonpathogenic Pseudogymnoascus speciesSpeciesIsolateAssembly size Largest scaffold Average scaffold Num scaffolds Scaffold N50 Percent GC Repetitive DNA Num genes Num proteins Num tRNA Unique proteins P.destructans 20631-2135,818,2012,552,699431,545831,168,63749.24%38.17%957593352401934 P .sp.23342-1-I123342-1-I132,900,320990,41233,640978289,62649.94%7.35%10,91410,7621521275 P .sp.24MN1324MN1330,179,533129,01910,675282724,07850.17%3.45%10,36810,269991090 P .sp.WSF3629WSF362935,517,105741,10686,839409189,86448.85%10.78%11,19311,0331601053 P .sp.3VT503VT0533,292,640361,48038,893856115,50849.09%15.94%10,39310,252141795 P .sp.5NY805NY0832,206,663867,88856,403571205,17249.74%10.22%10,66910,514155480 P.verrucosus UAMH1057930,174,8561,768,408199,833151446,34250.36%4.60%10,71510,573142515 ARTICLENATURECOMMUNICATIONS|DOI:10.1038/s41467-017-02441-z2NATURECOMMUNICATIONS| (2018) 9:35 |DOI:10.1038/s41467-017-02441-z|www.nature.com/naturecommunications

PAGE 3

selection,only14containedaproteinfrom P.destructans (SupplementaryData 9 ).However,functionalannotationfor theseorthologousgroupsdidnotprovidesuf cientinformation topredictafunctionforanyoftheseorthologs. Genome-levelcomparisons .Carbohydrate-activatingenzymes (CAZymes)areagroupofproteinsinvolvedinthebreakdown and/orutilizationofcarbon.Genomesoffungiassociatedwith plants,eitherplantpathogensordecomposers,generallyharbora greaternumberofCAZymesthanthosecausingdiseaseinanimals23.Thesixnonpathogenic Pseudogymnoascus specieson averageharbored493CAZymes,rangingfrom463to544; however,thegenomeof P.destructans containsonly~36%ofthe averagenumberofCAZymes(179intotal)forthegroup(Fig. 2 ; SupplementaryData 10 ).TheCAZymescanbefurtherbroken 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580 600 Terreneuvian Series2 Series3 Furongian Lower Middle Upper LlandoveryWenlock Ludlow PridoliLower Middle Upper Mississippian Pennsylvanian Cisuralian Guadalupian Lopingian Lower Middle Upper Lower Middle Upper Lower Upper Paleocene Eocene Oligocene Miocene Pliocene Pleistocene Holocene EdiacaranCambrian Ordovician SilurianDevonianCarboniferousPermianTriassicJurassicCretaceousPaleogeneNeogene Quaternary Pseudogymnoascus sp. VKMF103 Pseudogymnoascus sp. VKMF4519 Pseudogymnoascus verrucosus Pseudogymnoascus sp. 24MN13 Pseudogymnoascus sp. 04NY16 Pseudogymnoascus sp. 03VT05 Pseudogymnoascus sp. 05NY08 Pseudogymnoascus sp. WSF3629Pseudogymnoascus destructansPseudogymnoascus sp. VKMF4517 Pseudogymnoascus sp. VKMF4515 Pseudogymnoascus sp. BL308 Pseudogymnoascus sp. 233421I1 Pseudogymnoascus sp. VKMF4520 Pseudogymnoascus sp. VKMF4518 “Pseudogymnoascus pannorum” M1372 Pseudogymnoascus sp. VKMF4281 Pseudogymnoascus sp. VKMF4513 Pseudogymnoascus sp. VKMF4246 Pseudogymnoascus sp. BL549 Pseudogymnoascus sp. VKMF3775 “Pseudogymnoascus pannorum” ATCC1622 2 Pseudogymnoascus sp. VKMF3557 Pseudogymnoascus sp. VKMF4514 Pseudogymnoascus sp. VKMF4516 Pseudogymnoascus sp. VKMF3808 Botryotinia fuckeliana Sclerotinia sclerotiorum Fusarium fujikuroi Neurospora crassa Aspergillus nidulans Pencillium chrysogenum MYAPseudogymnoascus Leotiomycetes Helotiales Sordariomycetes Eurotiomycetes100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 97 100 100 100 ~ 23.5 MYA Fig.1 Maximumlikelihoodphylogenyof Pseudogymnoascus draftgenomesandseveraloutgroupAscomycetescalibratedusingfossilevidence.Thesix nonpathogenic Pseudogymnoascus speciessequencedhereareamongtheclosestknownrelativesof P.destructans P.destructans wasestimatedtohave divergedfromitslastcommonancestoraround23.5MYA.Nodesupportvaluesarederivedfrom1000bootstrapreplicates NATURECOMMUNICATIONS|DOI:10.1038/s41467-017-02441-zARTICLENATURECOMMUNICATIONS| (2018) 9:35 |DOI:10.1038/s41467-017-02441-z|www.nature.com/naturecommunications3

PAGE 4

downintoenzymeclasses/modules.Incomparisontotheaverage ofthenonpathogenic Pseudogymnoascus species,thegenomeof P.destructans harbors~69%(22)ofauxiliaryactivity(AA) enzymes,~17%(4)ofproteinscontainingcarbohydrate-binding modules(CBM),~20%(13)ofthecarbohydrateesterases(CE), ~30%(82)oftheglycosidehydrolases(GH),~10%(2)ofthe pectinlyases(PL),and~95%(52)oftheglycosyltransferases (GT).Moreover,therewerenospeci cCAZyfamiliesthatwere expandedin P.destructans thatwouldindicateenhancedability toutilizecertaincarbohydratesources.Duetothelargedecrease inCAZymes,wehypothesizedthat P.destructans wouldhave limitedgrowthoncomplexand/ordifferentcarbonsourcesin comparisontothenonpathogenic Pseudogymnoascus species.To testthishypothesis,wecomparedgrowthon190carbonsources usingamodi edBiologPhenotypeArrayPlatform.Consistent withthelargereductioninCAZymesinitsgenome, P.destructans wasunabletoutilizemanycarbonsourcesthatthenonpathogenic Pseudogymnoascus speciescouldreadilyutilize(Fig. 3 SupplementaryData 11 ).Thesedataareconsistentwitharecent reportcomparinggrowthcharacteristicsof P.destructans to closelyrelatedsoilfungi24.ThereducedCAZymerepertoireisa sharedcharacteristicofotherfungalpathogensofanimals, includingthedermatophytes,thetruefungaldimorphicpathogens,aswellasskin-inhabitingyeasts23.Presumably,thereductioninCAZymesre ectsachangeincarbonavailability/ acquisitionduringtheevolutionfromanancestralsaprobetoa batpathogen,i.e., P.destructans mayhaveshedCAZymepathwaysnecessaryforlivinginsoil/sedimentsbutarenotrequired forgrowthonhibernatingbats. Agroupofsubtilisinserineendopeptidases(MEROPSfamily S08A)of P.destructans havepreviouslybeendescribedasPdSP1, PdSP2,25andoneofthem(PdSP2;destructin-1)wassubsequentlyshowntodegradecollagen,suggestingapotentialrolein colonizationofbatskin26.UsingtheMEROPSproteasedatabase, 230predictedproteaseswereidenti edfromthe P.destructans genome,whichrepresentsa28%reductioncomparedtothe averageofthenonpathogenic Pseudogymnoascus species(average: 321;range:308 – 330)(Fig. 2 ;SupplementaryData 12 ).The subtilisinproteasesappeartobeconservedinthe Pseudogymnoascus speciesanalyzedhereandthereforeisconsistentwiththe observationfromVranetal.27thatcollagen-degradingenzymes arepresentinseveralnonpathogenicsoilmicroorganisms.While proteasesarelikelytobeinvolvedincolonizationofbatskin tissueby P.destructans ,thepreviouslyidenti edsubtilisin proteases(e.g.,PdSP2;destructin-1)donotappeartohave evolvedspeci callyforthispurpose.Moreover,PdSP2has recentlybeenshowntobemorehighlyexpressedinlaboratory culturemediumthanduringWNS28 29,indicatingthatitmaynot playalargeroleduringpathogenesis. Secretedproteinsaregenerallyimportantforfungiasdigestion ofnutrientsusingsecretedenzymesoccursoutsidethefungalcell. Recently,fungaleffectorshavebeendescribedinplantpathogens (someofwhichcanenteranimalcells),anexampleofthe intricatecompetitivearmsracebetweenpathogenandhost (reviewedinrefs.30 31).Thepredictedsecretomeof P.destructans isalsoreducedbymorethan50%incomparisontothe nonpathogenic Pseudogymnoascus species(452vs.averageof 970secretedproteins;range:848 – 1033)(Fig. 2 ).Thetrendof fewersecretedproteinsinfungalpathogensofanimalshasbeen previouslydescribed32 33,withonehypothesisbeingthatlosing unnecessarysecretedproteinsisanevolutionarystrategytoevade vertebrateimmunesystems32.Previousreportsindicatethat fungalsecretomescanbelineagespeci c,consistentwiththe observationthat56ofthe362predictedsolublesecretedproteins in P.destructans (~15%)donothaveorthologsinthe nonpathogenic Pseudogymnoascus species. Tovisuallydepictthefunctionalvariationbetweenspecies,we generatedcountdataofInterProScanDomainandPfamdomains foreachgenomeinourstudy(SupplementaryData 13 – 14 ).The resultingmatriceswerevisualizedusinganonmetricmultidimensionalscaling(NMDS)ordination,whichconsistently identi ed P.destructans asdistinctfromthenonpathogenic Pseudogymnoascus species(SupplementaryFig. 2 ).Toidentify functionalgroupsandprocessesthatwereenriched(positivelyor negatively),whichcouldhelpexplaintheshiftinfunctional domainsbetween P.destructans andthenonpathogenic 0200040006000800010,000 tRNA Orthologs Unique Rest Number of proteins 0100200300400500600 Auxiliary activities Carboydrate binding module Carbohydrate esterases Glycoside hydrolases Glycosyl transferases Polysacccharide lyases Number of proteins 0100200300 Number of proteins Mixed proteases Cysteine proteases Glutamic proteases Aspartic proteases Metallo proteases Threonine proteases Aspartic proteases Peptidase inhibitors 02505007501000 Soluble Number of proteins TM Domain P. destructans P. sp. 05NY08 P. sp. 03VT05 P. sp. WSF 3629 P. verrucosus P. sp 24MN13 P. sp. 23342-1-I1 Botrytis cinerea Gene modelsCAZymesProteasesSecreted proteins 100 100 100 100 100 100 1 100 Fig.2 Comparativegenomicanalysesoforthologousproteinsof Pseudogymnoascusdestructans andsixcloserelatives.Maximumlikelihoodphylogeny illustratesrelationshipsbetweenfungiusedinthisstudy.Genome-levelcomparisonsmadebasedonnumberofgenemodels,numberofCAZymes, numberofproteases,andthenumberofsecretedproteinsasdescribedinMethods ARTICLENATURECOMMUNICATIONS|DOI:10.1038/s41467-017-02441-z4NATURECOMMUNICATIONS| (2018) 9:35 |DOI:10.1038/s41467-017-02441-z|www.nature.com/naturecommunications

PAGE 5

Pseudogymnoascus species,weusedaGeneOntology(GO)term enrichmentmethod.Asexpectedandconsistentwiththe ordinations,noenrichmentofGOtermswasfoundforanyof thesixnonpathogenic Pseudogymnoascus species.Twoverybroad biologicalprocesses(BP)wereenrichedin P.destructans :cellular processes(GO:0009987)andcellularmetabolicprocesses (GO:00044327)(SupplementaryTable 3 ).Ontheotherhand, somemorespeci cBPtermswereunderrepresentedin P. destructans :transmembranetransport(GO:0055085),carbohydratemetabolicprocess(GO:005975),andoxidation – reduction process(GO:0055114).Thisisconsistentwithreductionsinthe secretome,aswellasCAZymesusingalternatemethodsas previouslydiscussed.Additionally,theGOenrichmentidenti ed atrendinunderrepresentedBPtermsrelatedtotranscription (GO:0006351,GO:0097659,andGO:0032774)(Supplementary Table 3 ).Generally,thesedatasuggestthatmajorgenome-level differencesbetween P.destructans andthenonpathogenic Pseudogymnoascus speciesaredrivenbythelargedifferencesin carbohydrateutilizationenzymes,aswellasareductioninthe putativesecretome. Ahallmarkofmanyascomycetefungiistheirabilityto producebioactivesmallmolecules,suchasthepharmaceuticals lovastatinandpenicillin34.Thesesmallmoleculesarealsoknown assecondarymetabolitesandmanyareproducedbythe coordinatedeffortofgeneclusters(genesphysicallylocatedin proximitytoeachotheronachromosome).Secondarymetaboliteshavebeenhypothesizedtobeinvolvedinnicheexploitation byfungi,includingevasionofvertebrateimmunesystems35. AntiSMASHv3.036wasusedtopredictsecondarymetabolite geneclustersfromeachgenome.BasedonantiSMASHprediction, P.destructans harbors14predictedsecondarymetabolite geneclusters,whiletheaverageforthenonpathogenic Pseudogymnoascus specieswas27geneclusters,witharangeof14 – 36 (SupplementaryTable 4 ,SupplementaryData 15 – 21 ).Despite havingfewerputativesecondarymetabolismclusters,the P. destructans genomeencodesforthetwonon-ribosomalpolyketidesynthetaseenzymesrequiredtoproducetheironscavengingsiderophoresthathavepreviouslybeencharacterized chemically:ferrichromeandtriacetylfusarinineC37(SupplementaryData 15 ).Additionally, P.destructans harborsaputative melaninclusterthathasbeenshowntobeinvolvedin pathogenicityinotherfungisuchastheopportunistichuman pathogen Aspergillusfumigatus ;fourofthesixnonpathogenic Pseudogymnoascus speciesalsoharborasimilarmelanin-like cluster,whichin A.fumigatus isphysicallylocatedinthecellwall ofsporesofthefungusandhasbeenhypothesizedtofunctionas protectionfromultraviolet(UV)lightandreactiveoxygenspecies (ROS)damage38 39.Thesecondarymetabolitearsenalofthe Pseudeurotiacaefungiinthisstudyislessthansomeothergroups suchastheEurotiomycetes,whichcanhave70ormore secondarymetabolismclusters.However,severalofthesecondary metabolitesproducedby P.destructans warrantfuturestudyas theycouldcontributetothedevelopmentandpersistenceofWNS inbats. Aninterestingcharacteristicofthe P.destructans genomeis thatitcontainsalargeexpansionofrepetitiveDNAsequences, accountingfor38.17%ofthegenome,asigni cantexpansionin comparisontothenonpathogenic Pseudogymnoascus species (Table 1 ).Coincidingwiththisrepeatexpansionisareduction intotalgenemodels,aswellasageneraltrendtowardareduction inseveralgenefamilies,drivenlargelybyamassive reductioninCAZymes,secretedproteins,andproteases.Atthe sametime, P.destructans containsmorelineage-speci cgenes thanthenonpathogenic Pseudogymnoascus species(Fig. 2 ).The observationofexpandedrepetitiveelementsandincreasing numbersoflineage-speci cgenesisreminiscentofthelineageP. destructans P. sp. 23342-1-I1 P. sp. 05NY08P. sp. 03VT05P.sp. 24MN13 P. verrucosus P. sp. WSF3629 0 0.5 1.0 1.5D,L-Malic acidL-Malic acid Glycyl-L-Aspartic acidL-Valine PalatinoseL-Aspartic acidL-Ornithine Glycine Gelatin b-methyl-D-GalactosideD-Mannitol a-keto-Valeric acid Pectin g-CyclodextrinL-Glutamic acid Tween 20 Propionic acid Citric acidD-Melezitose ArbutinL-AlaninamideL-SerineL-Arginine Acetic acid a-keto-Butyric acidD-Lactic acidL-ProlineD-Saccharic acidL-IsoleucineL-Phenylalanine Acetamide Amygdalin methyl ester Malonic acid a-methyl-D-Glucoside a-keto-Glutaric acid Succinic acid bromo Succinic acid p-hydroxy phenyl Acetic acid Tricarballylic acid 2-Aminoethanol g-amino-Butyric acid Sorbic acid Inulin b-methyl-D-Glucuronic acidL-Galactonic acid-g-LactoneD-Galacturonic acid Fumaric acid mono methyl Succinate Xylitol Lactulose i-Erythritol 4-hydroxy Benzoic acid Glycyl-L-Glutamic acid Glycyl-L-ProlineD-Glucuronic acidD-Gluconic acidD-Arabitol Salicin m-hydroxy-phenyl Acetic acidD-Sorbitol MaltorioseD-Raffinose Maltose b-methyl-D-GlucosideD-Ribose StachyoseL-Sorbose Quinic acid Glycogen m-Inositol Melibionic acidD-Melibiose TuranoseD-GalactoseD-Xylose DextrinD-Lactose L-Arabinose N-acetyl-N-euraminic acidD-Cellobiose Adonitol SucroseL-GlutamineL-AsparagineL-AlanineL-alanyl-GlycineL-Rhamnose LaminarinD-MannoseD-TrehaloseL-Pyroglutamic acid Caproic acid Tween 80 Tween 40 Sebacic acidD-GlucosamineD-Fructose Gentiobiose N-acetyl-D-GlucosamineD-Glucose Fig.3 HeatmapofBiologphenotypicmicroarraystestingfungalgrowth. Carbonutilizationfrom190sourceswastestedtoassessthegrowthof eachofseven Pseudogymnoascus species.Growthwasquanti edusing ImageJColonyAreaplug-inafter15Cincubationfor7daysforthe nonpathogenic Pseudogymnoascus speciesandafter14daysfor P. destructans .Growthforeachcarbonsourceispresentedasnormalizedto growthonglucose.Thesedataarederivedfromasinglebiologicalreplicate ( n = 1) NATURECOMMUNICATIONS|DOI:10.1038/s41467-017-02441-zARTICLENATURECOMMUNICATIONS| (2018) 9:35 |DOI:10.1038/s41467-017-02441-z|www.nature.com/naturecommunications5

PAGE 6

speci cregions/chromosomesin Fusariumoxysporum40.However,incontrastto F.oxysporum ,analysisoftherepetitive sequencesacrossthe P.destructans genomesuggestsauniform distribution(SupplementaryFig. 2 A)andlineage-speci cgenes werenotcolocalizedwithrepetitiveregions(Supplementary Fig. 2 B).SimilarlytofungalpathogensintheOnygenales, P. destructans lacksproteinscontainingthefungalcellulose-binding domain(CBM18)41;however,thisdoesnotseemtobetiedto pathogenicityas5ofthe6nonpathogenic Pseudogymnoascus speciesalsolackthisfunctionaldomain.Wewereunableto identifyproteinfamiliesthatwereexpandedin P.destructans somewhatreminiscentof Coccidioides ,whichhaveveryfew expandedproteinfamilies41.Thisisinstarkcontrasttothe amphibianchytridpathogens( Batrachochytriumdendrobatidis and Batrachochytriumsalamandrivorans ),whichhaveundergone anexpansionofproteasegenefamilies,aswellastheirgenomes enroutetopathogenicity42.Giventheevolutionarydistance betweenthechytridsandascomycetefungi,itisperhapsnot surprisingthatthesedevastatingfungalpathogenshavedifferent evolutionarytrajectories. LightandDNArepairpathways .Takentogether,ourdata suggestthat P.destructans hasbeenapathogenofbatsfor millionsofyearsandthushaslikelycoevolvedintheabsenceof light.Mostorganismsthathavebeenfoundintheabsenceoflight maintaintheabilitytorepairDNAcausedbyUVlightradiation, includingmostmicrobesthathavebeenisolatedfromhibernacula43 44.Moreover,ametagenomicsanalysisofacaveecosystem identi edanoverrepresentationofDNArepairenzymes,despite theabsenceofUVlight45.Wechallengedthefungalspecies studiedherewithfourdifferentDNA-damagingagentsthat includedUV,methylmethanesulfonate(MMS),4-nitroquinoline (4-NQO),andcamptothecin(CPT).Whileweobservedslight differencesinsensitivityamongspecies(andevenbetweenisolatesof P.destructans )with4-NQOandCPT,themostdramatic differentialsensitivitywasseenwithUVlightandMMS(Fig. 4 a). TofurthercharacterizethesensitivitytoUVlight,weemployeda quantitativeconidialsurvivalassayandexposedthefungitothree differentwavelengthsofUVlightandthreedifferentexposure levels(Fig. 4 b).Thesedatashowthatnoneofthefungitested weresensitivetoUV-A(366nm)lightatthedosagetested,while P.destructans isdifferentiallysensitivetothehigher-energyUV-B (312nm)andUV-C(254nm).Remarkably,alowdoseof5mJ/ cm2exposureofUV-Clightresultedinonly~15%survival,while a10mJ/cm2UV-Cexposureresultedin < 1%survivalof P. destructans (Fig. 4 b). 20631-21 (MAT1 – 1) CCF3942 (MAT1 – 2) CCF4125 (MAT1 – 1) CCF3941 (MAT1 – 1) CCF4124 (MAT1 – 2)P. destructansControl 25 mJ UV254 nm25 M CPT 0.01% MMS 0.5 g/ml 4-NQO 104103102101100104103102101100104103102101100104103102101100104103102101100 P. sp. WSF3629 P. sp. 23342 – 1 – I1 P. sp. 3VT5 P. sp. 5NY8 P sp. 24MN13 P. verrucosus 400 nm 750 nm 500 nm 600 nm Visible light UV light254 nm 312 nm 366 nmIntensity 5 mJ10 mJ25 mJ 0 10 20 30 40 50 60 70 80 90 100 110 120 UV-C light (254 nm)Percent survival (CFU/control) 4 mJ8 mJ20 mJ 0 10 20 30 40 50 60 70 80 90 100 110 120 UV-B light (312 nm)Percent survival (CFU/control) 70 mJ175 mJ350 mJ 0 10 20 30 40 50 60 70 80 90 100 110 120 UV-A light (366 nm)Percent survival (CFU/control)P. destructans P. verrucosus P. sp. WSF3629 P. sp. 3VT5 P. sp. 5NY8 P. sp. 23342-1-I1 P. sp. 24MN13 UV-C (254 nm) sensitivityUV-B (312 nm) sensitivityUV-A (366 nm) sensitivityNon-pathogenic Pseudogymnoascusa b Fig.4 Sensitivityof Pseudogymnoascus speciestoDNA-damagingagents. a QualitativeplateassaymeasuringtheeffectsoffourDNA-damagingagentson thegrowthofeachfungalspecies(4-nitroquinoline(4-NQO),camptothecin(CPT),methylmethanesulfonate(MMS),and254-nmultravioletlight(UV C)).Fungalsporeswereseriallydiluted,inoculatedonappropriatemedium,andgrowthwasquanti edafter7-dayincubationat15C.Multipleisolatesof P.destructans weretestedalongsidethenonpathogenic Pseudogymnoascus species. b Aquantitativeassayusingcolony-formingunits(CFUs)tomeasure survivalofeachfungusunderdifferentwavelengthsofUVlight(254nm(UV-C),312nm(UV-B),and366nm(UV-A)).CFUassayswereconductedin biologicaltriplicate( n = 3)anderrorbarsrepresentstandarddeviations ARTICLENATURECOMMUNICATIONS|DOI:10.1038/s41467-017-02441-z6NATURECOMMUNICATIONS| (2018) 9:35 |DOI:10.1038/s41467-017-02441-z|www.nature.com/naturecommunications

PAGE 7

UVlightdamagesDNAbyinducingtheformationof pyrimidinedimers(cyclobutanedimersand6-4photoproducts), whileMMSalkylatesguanineandadeninenucleotides46. Damagedormodi ednucleotidesresultinmispairingand/or replicationforkblockageandthereforecanresultinmutations. Camptothecin ’ smodeofactionissimilarasitinhibits topoisomeraseIfunction,whichresultsinstalledreplication forks;thus,lethalmutationscanaccumulateduringSphase47.On theotherhand,4-NQOinducessinglebasepairmutationswitha biastowardguaninetothyminetransversions48andthuscan causedirectmutagenesis.TocombatDNAdamage,organisms haveemployedseveralDNArepairpathways,including 0 mJ5 mJ10 mJ 0 10 20 30 40 50 60 70 80 90 100 110 120 130 P. verrucosus 0 mJ5 mJ10 mJ 0 10 20 30 40 50 60 70 80 90 100 110 120 130 P. sp. 24MN13 0 mJ5 mJ10 mJ 0 10 20 30 40 50 60 70 80 90 100 110 120 130 P. sp. 05NY08 0 mJ5 mJ10 mJ 0 10 20 30 40 50 60 70 80 90 100 110 120 130 P. sp. 03VT05 0 mJ5 mJ10 mJ 0 10 20 30 40 50 60 70 80 90 100 110 120 130 P. sp. 23342-1-I1 0 mJ5 mJ10 mJ 0 10 20 30 40 50 60 70 80 90 100 110 120 130 Percent survival (CFU/control)P. destructans CPD photolyases I DASH cryptochromes Animal cryptochromes 6-4 photolyases CPD photolyases II Plant cryptochromes CPD photolyases III UV-C dose(254 nm) UV-C UV-C + 1 h UV-A (366 nm)N. crassa CRY1Rice cryDASHArabidopsis cyr3Zebrafish cryDASHUstilago CRY1Zebrafish cry4Zebrafish cry2Zebrafish cry3Zebrafish cry1aHuman cry2Human cry1Mouse cry2B. cinerea CRY1S. sclerotiorum CRY1G. lozyensis CRY1P.sp. 24MN13 CRY1P. verrucosus CRY1P.sp. WSF3629 CRY1P.sp. WSF3629 CRY2P.sp. 3VT5 CRY1P.sp. 5NY8 CRY1P. destructans CRY1P.sp. 23342–1 CRY2P.sp. 23342–1 CRY1P.sp. 5NY8 Phr1P. verrucosus Phr1P.sp. 24MN13 Phr2P.sp. 24MN13 Phr1B.cinerea Phr1S.cerevisiae Phr1F.graminearum Phr1N.crassa PhrA.nidulans CryAUstilago Phr1Agrobacterium phrAMesorhizobium phrAXanthomonas phrATomato cry2Arabidopsis cyr2Rice cyr2Tomato cry1Arabidopsis cyr1Rice cry1Rice CPDIIArabidopsis CPDIIRice 6–4phrArabidopsis 6–4phrZebrafish 6–4phra b Fig.5 Bluelight-mediatedphotoreactivationDNArepairisnotfunctionalin P.destructans a Characterizationofphotolyasesin Pseudogymnoascus species basedonproteinalignmentandmaximumlikelihoodphylogenyofwell-characterizedenzymes. Pseudogymnoascusdestructans lacksacanonicalCPD photolyaseI,althoughitharborsacyr-DASHphotolyase(VC83_00225).Thegenomesof Pseudogymnoascusverrucosus P.sp.24MN13, and P.sp.05NY08 containaCPDphotolyaseI,whereasallspeciesstudiedcontainatleastonecry-DASHortholog.Proteinsidenti edinref.66wereusedforphylogenetic comparisonofDNAphotolyases. b PhotoreactivationexperimentscomparingsurvivalofgerminatingconidiaexposedtovaryingdosesofUV-C(254nm) followedbytreatmentwithUV-A(366nm)for1hindicatethatonly P.verrucosus and P.sp.24MN13 displayincreasedsurvivalattributedtotheactivityof CPDphotolyaseI.Nochangeinsurvivalwasdetectedin P.destructans ortheotherthreenonpathogenic Pseudogymnoascus speciestested.Interestingly, P. sp.05NY08 didnotshowanincreaseinCFUsurvivaldespitethefungusharboringaCPDphotolyaseIenzyme.Asimilarphenomenonoccursin Aspergillus nidulans ,wheretherewasnochangeinCFUsurvivalinphotoreactivationassaysinwild-typebackground,despitethepresenceofthefunctionalCryAgene (CPDphotolyaseI)67.CFUassayswereperformedinbiologicaltriplicate( n = 3)anderrorbarsrepresentstandarddeviation NATURECOMMUNICATIONS|DOI:10.1038/s41467-017-02441-zARTICLENATURECOMMUNICATIONS| (2018) 9:35 |DOI:10.1038/s41467-017-02441-z|www.nature.com/naturecommunications7

PAGE 8

photoreactivation,baseexcisionrepair(BER),nucleotideexcision repair(NER),double-strandbreakrepair,mismatchrepair,as wellasalternateexcisionrepair(AER)(reviewedinrefs.49 – 51). Using ssionyeast( Schizosaccharomycespombe )asamodel organism,weextracted169proteinsannotatedasinvolvedin DNArepairfrom www.pombase.org andqueriedthemagainst the Pseudogymnoascus proteomesusingBLAST(Supplementary Data 22 ).Sixputativehits,UVE1,MAG1,MAG2,CENP-X, SWI5,andPIF1,wereidenti edonthebasisthattheywere absentin P.destructans butpresentinmostofthesix nonpathogenic Pseudogymnoascus .Tovalidatethatorthologous proteinswereinfactmissingasopposedtobeingmisannotated, wequeriedthecorrespondingPfamhiddenMarkovmodel (HMM)pro lesfrom( http://www.pfam.org/ )usinganexhaustive HMMmodelsearchinPhyloma( https://github.com/nextgenusfs/ phyloma ),whichcanidentifytruncatedand/orunannotated genes.TheseresultsindicatedthatMAG1,MAG2,andCENP-X orthologsweremissingfromall Pseudogymnoascus genomes (whichisconsistentwiththelowscoresoftheBLASThits),while partialmatcheswerefoundforSWI5andPIF1,indicatingeither missedgenemodelsduringannotationornonfunctionalproteins. Finally,wewereunableto ndanytraceofUVE1inthe P. destructans genome,despite ndingclearhomologsinthesix nonpathogenic Pseudogymnoascus species,suggestingthatUVE1 playsalargeroleintherepairofUV-damagedDNAin Pseudogymnoascus P.destructans harborsaputativeDNAphotolyase (VC83_00225);thus,photoreactivationcouldbeamechanism thefungususestorepairUV-inducedDNAlesions.However, DNAphotolyasesrequirelightforfunctionandthereforeare unlikelytobeamajorcontributortoDNArepairinhibernacula. Preliminarytestingofsurvivabilityof P.destructans incubatedin eitherlightordarkindicatedthatlightwasinsuf cienttorepair DNAlesions(SupplementaryFig. 3 ).Thereareseveralknown classesofDNAphotolyasesinfungi,includingthecanonicalCPD photolyaseIfamilyshowntodirectlyrepaircyclobutanedimers, aswellasacyr-DASHfamilyshowntobeinvolvedinlightsensingphenotypes52 53.Amaximumlikelihoodphylogenyshows that P.destructans VC83_00225isamemberofthecry-DASH familyofDNAphotolyasesandisthusunlikelytocontributeto directphotorepairofcyclobutanedimers(Fig. 5 a).Tobe thorough,wetestedphotoreactivationinthelaboratoryand foundnophenotypein P.destructans ,whiletwoofthe nonpathogenic Pseudogymnoascus speciesthatharboredatleast oneCPDphotolyaseI( P.verrucosus and P.sp.24MN13 )showed increasedsurvivalwhenexposedtoUV-A(366nm)lightafter DNAdamage(Fig. 5 b). In ssionyeast,UVE1isakeycomponentoftheAERpathway; therefore,ourdatasuggestthatmostUV-damagedDNAin Pseudogymnoascus isrepairedthroughAER.Asmallpercentage ofrepaircanbeattributedtophotoreactivationin P.verrucosus and P.sp.24MN13 ;however,thisappearstonotoccurin P. destructans asitlacksacyclobutane-dimerphotorepairenzyme. The P.destructans genomeharborsclearhomologsforthemajor enzymesinvolvedinNERpathway(SupplementaryData 22 ),and thus,moreworkisneededtodeterminewhythegeneralNERand BERpathwaysareunabletocompensateforlossofAERin P. destructans .Nevertheless,theextremesensitivitytoUV-Clightin P.destructans representsageneticpathwaythatcouldbe exploitedforactivemanagementofWNSofbats.Currently, UV-A(366nm)lightisbeingusedasaWNSnoninvasive eld diagnostictoolbecausethecharacteristiccuppingskinlesions uoresceunderthiswavelength54.WhileUV-Alighttestedunder laboratoryconditionshadnoeffectonsurvivalof P.destructans (Fig. 4 ),theuseofUV-Aasadiagnostictoolsuggeststhattreating individualbatswithadosageofUV-Clightisfeasible.The relativelylowdose(~10mJ/cm2)ofUV-Clightrequiredtokill P. destructans conidiacouldbeappliedtobatsinafewsecondsof exposurefromaportablelightsource.Moreworkisneededto understandthephysiologicaleffectofUV-Clightonbats; however,itisencouragingthattreatmentofmammalfungalskin/ nailinfectionswithUV-Clighthasbeenusedfordermatophyte fungicausingonychomycosis55,aswellaswoundinfections causedby Candidaalbicans56. Discussion WNSrepresentsoneofthemostseverewildlifediseasesever recorded.Currently,therearenopracticaltreatmentoptionsfor containingthespreadofWNSinNorthAmericaandthusthe fungushasmovedrapidlyacrossthecontinent.Survivorpopulationsofaffectedbatspeciescanstillbefoundnearthedisease outbreakepicenter,sothatthereishopethatspeciesextinction willbeavoided,albeitwithaprolongedrecovery57.However,the long-termeffectsonecosystemsinvolvingbatswillnotbe understoodfordecades.Comparativegenomicsanalysespresentedheresuggestthat P.destructans islikelyatruefungal pathogenofbats,evolvingalongsideEurasianbatspeciesfor millionsofyears.Theannotatedgenomeof P.destructans in additiontoseveralnonpathogeniccloselyrelatedspeciesprovides aframeworkforunderstandingthepathobiologyofWNS.The serendipitousdiscoverythat P.destructans lacksaUVE1homolog andisthereforeextremelysensitivetopyrimidinedimerinducing DNA-damagingagentsisavulnerabilitythatcouldbeexploited forWNSmanagement. MethodsGrowthconditions,DNAextraction,andsequencing .Fungalstrainsweregrown inliquid-stationaryculture58forseveraldaysat15C,myceliawerelyophilized, andsubsequentlygenomicDNA(gDNA)wasextractedaspreviouslydescribed59. High-molecular-weightgDNAwascon rmedbygelelectrophoresis.ForsequencingontheIonTorrentPersonalGenomeMachine(PGM),arandom400-bpsize fractionatedlibrarywasconstructedusingIonPlusFragmentLibraryKit (#4471257),templatedusingtheIonPGMTemplateOT2400Kit(#4479878), loadedona318v2chip(#4484354,andsequencedusingtheIon400bpSequencing Kit(#4482002)(allkitswereusedaccordingtothemanufacturer ’ srecommendations).LibraryconstructionforsequencingusingtheIlluminaGAxIIandIllumina MiSeqwasdoneusingtheNEBNextLibraryPrepKit,puri edusingQIAQuick CartridgeKit,andsizefractionatedusingE-gel2%sizeselectgelelectrophoresis system(ThermoFisher).ForobtainingRNA-seqreadstoassistwithgenome annotation,fungiweregrowninliquid-shakingculturesat15Cfor3days,mycelia werelyophilized,totalRNAwasextractedusingTriZol(Invitrogen),andpolyadenylatedRNAwasselectedusingtheDynaBeadsPolyACleanupKit(Ambion). Thepoly-ARNAwasthenusedtomakelibrariesusingtheIonRNA-seqKit2.0 andsequencedontheIonTorrentPGMusingIon200bpSequencingKit (#4482006). Genomeassembly .Initialattemptstobuildahigh-qualityassemblyfor P. destructans usingpaired-endMiSeq(2250bp)dataresultedinheavilyfragmentedassemblieswithseveraldifferentassemblysoftware(DiscovarDeNovo, Abyss,Spades,CLC,etc.).Wegeneratedahighlycontiguousassemblyusinga combinationofPacBioreads,MiSeq,Roche454mate-pairreads,andSangerend sequencesfroma100-kbBAClibrary14.For veofthenonpathogenic Pseudogymnoascus species(UAMH10579,03VT05,05NY08,WSF3629,and23342-1-I1), acceptableassembliesweregeneratedusingthedenovoassemblerinCLCGenomicsWorkbench7.5(Qiagen)usingpaired-endMiSeqreads(2300bp).For Pseudogymnoascussp. 24MN13,thepaired-endIllumina(2100bp;GAIIx)data werefoundtohavesomebacterialcontamination,thus,wegeneratedasecond sequencinglibraryfromacleangDNAextractionfortheIonTorrentPGM,which wasusedforaninitialassemblyandscaffoldingusingthepaired-endIllumina reads.ContaminationwasremovedfromtheassembliesusingDeconSeq60andwas subsequentlyerrorcorrectedusingPilon61.Smallrepetitivecontigsfromeach assemblywereidenti edusingMummer62andwereremovediftheywere95% identicalover95%oftheirlengthwithothercontigsintheassembly(achieved using “ funannotateclean ” command). Genomeannotationandfunctionalcharacterization .Detailedinformationon annotationandfunctionalcharacterizationispresentedinSupplementary Methods. ARTICLENATURECOMMUNICATIONS|DOI:10.1038/s41467-017-02441-z8NATURECOMMUNICATIONS| (2018) 9:35 |DOI:10.1038/s41467-017-02441-z|www.nature.com/naturecommunications

PAGE 9

Biologphenotypticmicroarray Pseudogymnoascus speciesweretestedforthe abilitytoutilize190differentcarbonsourcesintheBiologPhenotypticMicroarray platform(platesPM1andPM2A).Fungalsporesatadensityof1104sporeswere inoculatedinto100 lofcarbon-freeminimalmedium(MMlackingglucose63)and thensubsequentlyaliquotedinto96-wellBiologplates(PM1andPM2A).Cultures weregrownat15Candthenonpathogenic Pseudogymnoascus specieswere incubatedfor7days,while P.destructans wasincubatedfor14days.Theplates werethenphotographedandgrowthwasquanti edusingtheFiji/ImageJplug-in ColonyArea1.564.Relativegrowthwasthennormalizedtogrowthonglucoseand plottedinheatmap.2fromthegplotspackageinR65. DNAdamageassays .Conidiawereharvestedfromfungalisolatesfromsolid mediumagarplatesusingsterile0.01%Tween-80water,puri edby lteringover sterilemiracloth(EMDMillipore#475855),andenumeratedusingahemocytometer.Forspotplateassays,conidiawereseriallydilutedand5 lwaspipetted ontothesurfaceofglucoseminimalagarmedium63.TreatmentwithUVlightwas accomplishedbyexposingthespottedagarplateto25mJ/cm2ofUV-C(254nm) lightusingaUVcross-linker(UVPCL-1000).SensitivitytochemicalDNA mutagenswasdonebypreparingGMMagarmediumthatcontained25 MCPT, 0.01%MMS,or0.5 g/mlof4-nitroquinolone(4-NQO).Controlandtreatedplates wereincubatedat15Cfor1weekpriortoimaging.Colony-formingunit(CFU) assaysweredonebyspreadplating~50conidiaonthesurfaceofaGMMagarplate andweresubsequentlyexposedtoUVlightinaUVcross-linker,andsurviving colonieswerecountedafterincubationat15Cfor1week.Quanti cationof photoreactivationforeachspecieswasdoneusingaverysimilarCFUassay comparingvaryingUV-Ctreatmentsfollowedby1-hexposuretoUV-A(366nm) lightinaUVcross-linker.Conidiawereallowedtogerminatefor24hinthedark priortoUV-C(254nm)treatmentsandsubsequentUV-A(366nm)treatment. Survivingcolonieswerecountedafterincubationinthedarkatanappropriate temperature(15Cfor P.destructans and25Cfornonpathogenic Pseudogymnoascus species).AllCFUassaysweredoneinbiologicaltriplicates( n = 3)for eachexperimentalcondition. Dataavailability .SequencingdataandgenomeassembliesareavailableviaNCBI underBioProjectPRJNA276926andSRAprojectSRP055906.Allotherrelevant datasupportingthe ndingsofthestudyareavailableinthisarticleanditsSupplementaryInformation les,orfromthecorrespondingauthoruponrequest.Received:29March2017Accepted:30November2017 References1.Lorch,J.M.etal.Firstdetectionofbatwhite-nosesyndromeinwesternNorth America. mSphere 1 ,e00148-16(2016). 2.Frick,W.F.etal.Anemergingdiseasecausesregionalpopulationcollapseofa commonNorthAmericanbatspecies. Science 329 ,679 – 682(2010). 3.Gargas,A.,Trest,M.T.,Christensen,M.,Volk,T.J.&Blehert,D.S. Geomyces destructans sp.nov.associatedwithbatwhite-nosesyndrome. Mycotaxon 108 147 – 154(2009). 4.Minnis,A.M.&Lindner,D.L.Phylogeneticevaluationof Geomyces andallies revealsnocloserelativesof Pseudogymnoascusdestructans ,comb.nov.,inbat hibernaculaofeasternNorthAmerica. FungalBiol. 117 ,638 – 649(2013). 5.Verant,M.L.,Boyles,J.G.,Waldrep,W.,Wibbelt,G.&Blehert,D.S. Temperature-dependentgrowthof Geomycesdestructans ,thefungusthat causesbatwhite-nosesyndrome. PLoSONE 7 ,e46280(2012). 6.Cryan,P.M.,Meteyer,C.U.,Boyles,J.G.&Blehert,D.S.Wingpathologyof white-nosesyndromeinbatssuggestslife-threateningdisruptionofphysiology. BMCBiol. 8 ,135 – 135(2010). 7.Reeder,D.M.etal.Frequentarousalfromhibernationlinkedtoseverityof infectionandmortalityinbatswithwhite-nosesyndrome. PLoSONE 7 ,e38920 (2012). 8.Lorch,J.M.etal.Experimentalinfectionofbatswith Geomycesdestructans causeswhite-nosesyndrome. Nature 480 ,376 – 378(2011). 9.Verant,M.L.etal.White-nosesyndromeinitiatesacascadeofphysiologic disturbancesinthehibernatingbathost. BMCPhysiol. 14 ,10(2014). 10.Warnecke,L.etal.InoculationofbatswithEuropean Geomycesdestructans supportsthenovelpathogenhypothesisfortheoriginofwhite-nosesyndrome. Proc.Natl.Acad.Sci.USA 109 ,6999 – 7003(2012). 11.Puechmaille,S.J.etal.White-nosesyndrome:isthisemergingdiseaseathreat toEuropeanbats? TrendsEcol.Evol. 26 ,570 – 576(2011). 12.Lorch,J.M.etal.Aculture-basedsurveyoffungiinsoilfrombathibernacula intheeasternUnitedStatesanditsimplicationsfordetectionof Geomyces destructans ,thecausalagentofbatwhite-nosesyndrome. Mycologia https://doi. org/10.3852/12-207 (2012). 13.Lorch,J.M.etal.Distributionandenvironmentalpersistenceofthecausative agentofwhite-nosesyndrome, Geomycesdestructans ,inbathibernaculaofthe easternUnitedStates. Appl.Environ.Microbiol. 79 ,1293 – 1301(2013).14.Drees,K.P.etal.UseofmultiplesequencingtechnologiestoproduceahighqualitygenomeofthefungusPseudogymnoascusdestructans,thecausative agentofbatwhite-nosesyndrome. GenomeAnnounc 4 ,e00445 – 16(2016). 15.Simo,F.A.,Waterhouse,R.M.,Ioannidis,P.,Kriventseva,E.V.&Zdobnov,E. M.BUSCO:assessinggenomeassemblyandannotationcompletenesswith single-copyorthologs. Bioinformatics 31 ,3210 – 3212(2015). 16.Beimforde,C.etal.EstimatingthephanerozoichistoryoftheAscomycota lineages:combiningfossilandmoleculardata. Mol.Phylogen.Evol. 78 ,386 – 398 (2014). 17.Sanderson,M.J.r8s:inferringabsoluteratesofmolecularevolutionand divergencetimesintheabsenceofamolecularclock. Bioinformatics 19 301 – 302(2003). 18.Novacek,M.J.Evidenceforecholocationintheoldestknownbats. Nature 315 140 – 141(1985). 19.Springer,M.S.,Teeling,E.C.,Madsen,O.,Stanhope,M.J.&deJong,W.W. Integratedfossilandmoleculardatareconstructbatecholocation. Proc.Natl. Acad.Sci.USA 98 ,6241 – 6246(2001). 20.Ruedi,M.&Mayer,F.Molecularsystematicsofbatsofthegenus Myotis (Vespertilionidae)suggestsdeterministicecomorphologicalconvergences. Mol. Phylogenet.Evol. 21 ,436 – 448(2001). 21.Lechner,M.etal.Orthologydetectioncombiningclusteringandsyntenyfor verylargedatasets. PLoSONE 9 ,e105015(2014). 22.Yang,Z.PAML4:phylogeneticanalysisbymaximumlikelihood. Mol.Biol. Evol. 24 ,1586 – 1591(2007). 23.Zhao,Z.,Liu,H.,Wang,C.&Xu,J.-R.Correction:comparativeanalysisof fungalgenomesrevealsdifferentplantcellwalldegradingcapacityinfungi. BMCGenom. 15 ,6(2014). 24.Wilson,M.B.,Held,B.W.,Freiborg,A.H.,Blanchette,R.A.&Salomon,C.E. Resourcecaptureandcompetitiveabilityofnon-pathogenic Pseudogymnoascus spp.and P.destructans ,thecauseofwhite-nosesyndromeinbats. PLoSONE 12 ,e0178968(2017). 25.Pannkuk,E.L.,Risch,T.S.&Savary,B.J.Isolationandidenti cationofan extracellularsubtilisin-likeserineproteasesecretedbythebatpathogen Pseudogymnoascusdestructans PLoSONE 10 ,e0120508(2015). 26.O ’ Donoghue,A.J.etal.Destructin-1isacollagen-degradingendopeptidase secretedbyPseudogymnoascusdestructans,thecausativeagentofwhite-nose syndrome. Proc.Natl.Acad.Sci.USA 112 ,7478 – 7483(2015). 27.Vran,B.,Hntkov,Z.&Lettl,A.Occurrenceofcollagen-degrading microorganismsinassociationsofmesophilicheterotrophicbacteriafromvarioussoils. FoliaMicrobiol. 33 ,458 – 461(1988). 28.Reeder,S.M.etal. Pseudogymnoascusdestructans transcriptomechanges duringwhite-nosesyndromeinfections. Virulence 0 ,https://doi.org/10.1080/ 21505594.2017.1342910(2017). 29.Field,K.A.etalThewhite-nosesyndrometranscriptome:activationofantifungalhostresponsesinwingtissueofhibernatinglittlebrown Myotis PLoS Pathog. 11 ,e1005168(2015). 30.Petre,B.&Kamoun,S.Howdo lamentouspathogensdelivereffectorproteins intoplantcells? PLoSBiol. 12 ,e1001801(2014). 31.Kale,S.D.&Tyler,B.M.Entryofoomyceteandfungaleffectorsintoplantand animalhostcells. Cell.Microbiol. 13 ,1839 – 1848(2011). 32.Krijger,J.-J.,Thon,M.R.,Deising,H.B.&Wirsel,S.G.R.Compositionsof fungalsecretomesindicateagreaterimpactofphylogenetichistorythan lifestyleadaptation. BMCGenom. 15 ,722(2014). 33.Kim,K.-T.etal.Kingdom-wideanalysisoffungalsmallsecretedproteins (SSPs)revealstheirpotentialroleinhostassociation. Front.PlantSci. 7 ,186 (2016). 34.Keller,N.P.,Turner,G.&Bennett,J.W.Fungalsecondarymetabolism-from biochemistrytogenomics. Nat.Rev.Microbiol. 3 ,937 – 947(2005). 35.Scharf,D.H.,Heinekamp,T.&Brakhage,A.A.Humanandplantfungal pathogens:theroleofsecondarymetabolites. PLoSPathog. 10 ,e1003859 (2014). 36.Weber,T.etal.antiSMASH3.0-acomprehensiveresourceforthegenome miningofbiosyntheticgeneclusters. NucleicAcidsRes. 43 ,W237 – W243 (2015). 37.Mascuch,S.J.etal.Directdetectionoffungalsiderophoresonbatswithwhitenosesyndromevia uorescencemicroscopy-guidedambientionizationmass spectrometry. PLoSONE 10 ,e0119668(2015). 38.Jahn,B.etal.Isolationandcharacterizationofapigmentless-conidiummutant of Aspergillusfumigatus withalteredconidialsurfaceandreducedvirulence. Infect.Immun. 65 ,5110 – 5117(1997). 39.Langfelder,K.,Streibel,M.,Jahn,B.,Haase,G.&Brakhage,A.A.Biosynthesis offungalmelaninsandtheirimportanceforhumanpathogenicfungi. Fungal Genet.Biol. 38 ,143 – 158(2003). 40.Ma,L.-J.etal.Comparativegenomicsrevealsmobilepathogenicity chromosomesin Fusarium Nature 464 ,367 –373(2010). NATURECOMMUNICATIONS|DOI:10.1038/s41467-017-02441-zARTICLENATURECOMMUNICATIONS| (2018) 9:35 |DOI:10.1038/s41467-017-02441-z|www.nature.com/naturecommunications9

PAGE 10

41.Sharpton,T.J.etal.Comparativegenomicanalysesofthehumanfungal pathogens Coccidioides andtheirrelatives. GenomeRes. 19 ,1722 – 1731 (2009). 42.Farrer,R.A.etal.Genomicinnovationslinkedtoinfectionstrategiesacross emergingpathogenicchytridfungi. Nat.Commun. 8 ,14742(2017). 43.Arrage,A.A.,Phelps,T.J.,Benoit,R.E.&White,D.C.Survivalofsubsurface microorganismsexposedtoUVradiationandhydrogenperoxide. Appl. Environ.Microbiol. 59 ,3545 – 3550(1993). 44.Snider,J.,Goin,C.,Miller,R.,Boston,P.&Northup,D.Ultravioletradiation sensitivityincavebacteria:evidenceofadaptationtothesubsurface? Int.J. Speleol. 38 ,11 – 22(2009). 45.Ortiz,M.etal.Makingalivingwhilestarvinginthedark:metagenomic insightsintotheenergydynamicsofacarbonatecave. ISMEJ. 8 ,478 – 491 (2014). 46.Beranek,D.T.Distributionofmethylandethyladductsfollowingalkylation withmonofunctionalalkylatingagents. Mutat.Res. 231 ,11 – 30(1990). 47.Fiorani,P.&Bjornsti,M.A.MechanismsofDNAtopoisomeraseI-inducedcell killingintheyeast Saccharomycescerevisiae Ann.N.Y.Acad.Sci. 922 ,65 – 75 (2000). 48.Downes,D.J.etal.Characterizationofthemutagenicspectrumof 4-nitroquinoline1-oxide(4-NQO)in Aspergillusnidulans bywholegenome sequencing. G3 4 ,2483 – 2492(2014). 49.Yasui,A.Alternativeexcisionrepairpathways. ColdSpringHarb.Perspect.Biol 5 ,a012617(2013). 50.Dexheimer,T.S. DNARepairofCancerStemCells 19 – 32(Springer, Netherlands,2012). 51.Marteijn,J.A.,Lans,H.,Vermeulen,W.&Hoeijmakers,J.H.J.Understanding nucleotideexcisionrepairanditsrolesincancerandageing. Nat.Rev.Mol.Cell Biol. 15 ,465 – 481(2014). 52.Cohrs,K.C.&Schumacher,J.Thetwocryptochrome/photolyasefamily proteinsful lldistinctrolesinDNAphotorepairandregulationofconidiation inthegraymoldfungus Botrytiscinerea.Appl.Environ.Microbiol. 83 AEM.00812 – 17(2017). 53.Froehlich,A.C.etal.Geneticandmolecularcharacterizationofa cryptochromefromthe lamentousfungus Neurosporacrassa. Eukaryot.Cell 9 738 – 750(2010). 54.Turner,G.G.etal.Nonlethalscreeningofbat-wingskinwiththeuseof ultraviolet uorescencetodetectlesionsindicativeofwhite-nosesyndrome. J. Wildl.Dis. 50 ,566 – 573(2014). 55.Dai,T.,Tegos,G.P.,Rolz-Cruz,G.,Cumbie,W.E.&Hamblin,M.R. UltravioletCinactivationofdermatophytes:implicationsfortreatmentof onychomycosis. Br.J.Dermatol. 158 ,1239 – 1246(2008). 56.Dai,T.etal.Ultraviolet-Clightfortreatmentof Candidaalbicans burn infectioninmice. Photochem.Photobiol. 87 ,342 – 349(2011). 57.Langwig,K.E.etal.Resistanceinpersistingbatpopulationsafterwhite-nose syndromeinvasion.P hilos.Trans.R.Soc.Lond.BBiol.Sci. 372 ,20160044 (2017). 58.Champe,S.P.&el-Zayat,A.A.Isolationofasexualsporulationhormonefrom Aspergillusnidulans J.Bacteriol. 171 ,3982 – 3988(1989). 59.Palmer,J.M.etal.Molecularcharacterizationofaheterothallicmatingsystem in Pseudogymnoascusdestructans ,thefunguscausingwhite-nosesyndromeof bats. G3 4 ,1755 – 1763(2014). 60.Schmieder,R.&Edwards,R.Fastidenti cationandremovalofsequence contaminationfromgenomicandmetagenomicdatasets. PLoSONE 6 ,e17288 (2011). 61.Walker,B.J.etal.Pilon:anintegratedtoolforcomprehensivemicrobialvariant detectionandgenomeassemblyimprovement. PLoSONE 9 ,e112963(2014). 62.Kurtz,S.etal.Versatileandopensoftwareforcomparinglargegenomes. GenomeBiol. 5 ,R12(2004). 63.Shimizu,K.&Keller,N.P.GeneticinvolvementofacAMP-dependentprotein kinaseinaGproteinsignalingpathwayregulatingmorphologicalandchemical transitionsin Aspergillusnidulans Genetics 157 ,591 – 600(2001). 64.Guzmn,C.,Bagga,M.,Kaur,A.,Westermarck,J.&Abankwa,D.ColonyArea: anImageJplugintoautomaticallyquantifycolonyformationinclonogenic assays. PLoSONE 9 ,e92444(2014). 65.RCoreTeam. R:ALanguageandEnvironmentforStatisticalComputing https://www.r-project.org/ (Vienna,Austria,2016). 66.Tagua,V.G.etal.FungalcryptochromewithDNArepairactivityrevealsan earlystageincryptochromeevolution. Proc.Natl.Acad.Sci.USA 112 15130 – 15135(2015).67.Bayram,O.,Biesemann,C.,Krappmann,S.,Galland,P.&Braus,G.H.More thanarepairenzyme: Aspergillusnidulans photolyase-likeCryAisaregulator ofsexualdevelopment. Mol.Biol.Cell 19 ,3254 – 3262(2008).AcknowledgementsFundingwasprovidedtoJ.M.P.andD.L.L.bytheUSFish&WildlifegrantF14PG00132 andtoK.P.D.andJ.T.F.byUSFish&WildlifegrantF14AP00644.Additionalfunding andsupportforJ.M.P.andD.L.L.wereprovidedbytheUSForestService,Northern ResearchStation.AuthorcontributionsJ.M.P.,K.P.D.,J.T.F.,andD.L.L.conceivedthestudy;K.P.D.andJ.T.F.performedwholegenomesequencing,K.P.D.conductedhybridgenomeassemblyof P.destructans ,JMP conductedgenomeassembly,annotation,genome-levelcomparisons,andconducted laboratoryexperiments;andJ.M.P.wrotethepaperwithinputfromK.P.D.,J.T.F.,andD. L.L.AdditionalinformationSupplementaryInformation accompaniesthispaperat https://doi.org/10.1038/s41467017-02441-z Competinginterests: Theauthorsdeclarenocompeting nancialinterests. Reprintsandpermission informationisavailableonlineat http://npg.nature.com/ reprintsandpermissions/ Publisher'snote: SpringerNatureremainsneutralwithregardtojurisdictionalclaimsin publishedmapsandinstitutionalaf liations. OpenAccess ThisarticleislicensedunderaCreativeCommons Attribution4.0InternationalLicense,whichpermitsuse,sharing, adaptation,distributionandreproductioninanymediumorformat,aslongasyougive appropriatecredittotheoriginalauthor(s)andthesource,providealinktotheCreative Commonslicense,andindicateifchangesweremade.Theimagesorotherthirdparty materialinthisarticleareincludedinthearticle ’ sCreativeCommonslicense,unless indicatedotherwiseinacreditlinetothematerial.Ifmaterialisnotincludedinthe article ’ sCreativeCommonslicenseandyourintendeduseisnotpermittedbystatutory regulationorexceedsthepermitteduse,youwillneedtoobtainpermissiondirectlyfrom thecopyrightholder.Toviewacopyofthislicense,visit http://creativecommons.org/ licenses/by/4.0/ TheAuthor(s)2017 ARTICLENATURECOMMUNICATIONS|DOI:10.1038/s41467-017-02441-z10NATURECOMMUNICATIONS| (2018) 9:35 |DOI:10.1038/s41467-017-02441-z|www.nature.com/naturecommunications