columnswww.avtt3.com

ORIGINALPAPERAneedleinahaystack:amultigeneTaqManassayforthedetectionofAsiangypsymothsinbulkpheromonetrapsamplesDonStewart.
AudreyNisole.
AbdelmadjidDjoumad.
RezaZahiri.
JosyanneLamarche.
RogerC.
Levesque.
RichardC.
Hamelin.
MichelCussonReceived:29August2018/Accepted:10February2019/Publishedonline:7March2019TheAuthor(s)2019AbstractTheAsiangypsymoth(AGM)isconsid-eredaveryseriousinvasivethreatinNorthAmerica.
Forthisreason,itissubjectedtoabio-surveillanceprogramthatincludesanextensivenetworkofpheromonetraps.
Forregulatorypurposes,theterm''AGM''designatesagroupofAsianLymantriaspeciesandsubspecies,comprisingtwoL.
disparsubspecies(L.
d.
asiaticaandL.
d.
japonica),andthreecloselyrelatedspecies(L.
umbrosa,L.
albescensandL.
postalba).
ThesemothsareattractedtothesamepheromoneastheEuropeangypsymoth(EGM),L.
dispardispar,whichisalreadyestablishedinNorthAmericaandtypicallymakesupthebulkofmothscaughtingypsymothpheromonetraps.
Thesediffer-entLymantriataxaaredifculttodistinguishfromoneanotherusingmorphologicalcharactersalone.
Here,wedesignedaTaqMantriplexassaycapableofdetectingAGMinbulkpheromonetrapsamples.
TheassaytargetsSNPsfoundinthreedifferentmitochondrialgenes.
UsingaDNAdilutionseries,weshowthattheassaycandetectAGMtaxaatAGM:EGMdilutionratiosC1:1000.
TheassaywasvalidatedusingbatchDNAextractionsofmothlegstestedata1:100AGM:EGMlegratio,aproportionthatisaroundtheoperationallimitforasinglepheromonetrap.
Theassayprovidedcorrectidenti-cationforallAGMtaxatested.
AnexperimentexaminingtheintegrityofDNAextractedfromgypsymothsleftinpheromonetrapsundereldconditionsforupto4monthsindicatedthatDNAqualityremainssufcient,duringthatperiod,forthepresentassaytoremainaccurate.
KeywordsAsiangypsymothLymantriadisparMolecularidenticationBulkassayTaqManElectronicsupplementarymaterialTheonlineversionofthisarticle(https://doi.
org/10.
1007/s10530-019-01943-9)con-tainssupplementarymaterial,whichisavailabletoauthorizedusers.
D.
StewartA.
NisoleA.
DjoumadJ.
LamarcheM.
Cusson(&)LaurentianForestryCentre,CanadianForestService,NaturalResourcesCanada,QuebecCity,QC,Canadae-mail:michel.
cusson@canada.
caR.
ZahiriOttawaPlantLaboratory,EntomologyUnit,CanadianFoodInspectionAgency,Ottawa,ON,CanadaR.
C.
LevesqueR.
C.
HamelinM.
CussonInstitutdeBiologieIntegrativeetdesSyste`mes(IBIS),UniversiteLaval,Quebec,QC,CanadaR.
C.
HamelinDepartmentofForestandConservationSciences,FacultyofForestry,UniversityofBritishColumbia,Vancouver,BC,Canada123BiolInvasions(2019)21:1843–1856https://doi.
org/10.
1007/s10530-019-01943-9(0123456789().
,-volV)(0123456789().
,-volV)IntroductionThegypsymoth,Lymantriadispar(Linnaeus)(Lepi-doptera:Erebidae:Lymantriinae),isconsideredamajorthreattoNorthAmerica'sforestresourcesandeconomy.
Thisspeciescomprisesthreerecognizedsubspecies,includingtheEuropeangypsymoth(EGM),L.
dispardispar(Linnaeus),whichisfoundovermostofcontinentalEuropeandnorth-westernAfrica(PogueandSchaefer2007).
EGMwasacci-dentallyintroducedinMassachusettsin1869(Lieb-holdetal.
1989)andhassincebecomeestablishedinnorth-easternNorthAmerica,whereitisresponsibleforrecurrentandsubstantialdefoliationresultinginimportanteconomiclosses(Bradshawetal.
2016).
Theothertwosubspecies,L.
disparasiaticaVnu-kovskijandL.
disparjaponica(Motschulsky),arefoundthroughoutmainlandeasternAsiaandJapan,respectively(PogueandSchaefer2007).
AlongwiththreecloselyrelatedLymantriaspecies(L.
umbrosaButler,L.
albescensHoriandUmeno,andL.
postalbaInoue),theyformacomplexinformallyknownasAsiangypsymoth(AGM).
Asagroup,theseinsectsareconsideredamuchmoreseriousthreatthantheEuropeansubspecies,inlargepartbecauseAGMfemales,unlikethoseofL.
dispardispar,areight-capable,increasingtheprobabilityofarapidinvasionfollowinganuncheckedaccidentalintroduction.
Suchintroductionshaveindeedoccurredinthepast,buteradicationoperationsappeartohavebeensuccessful(APHIS2016;Nealis2002,2009).
Molecularmarkersanddiagnostictoolstargetingthegypsymothhavebecomeimportanttohelppreventinvasionsviaearlyintervention(ArimotoandIwai-zumi2014;ArmstrongandBall2005;Armstrongetal.
2003;BallandArmstrong2006;Bogdanowiczetal.
1993,1997,2000;Boykinetal.
2012;Chenetal.
2013,2015;deWaardetal.
2010;GarnerandSlavicek1996;HarrisonandODell1989;Islametal.
2015;Kangetal.
2015,2017;Keenaetal.
2008;Lackovicetal.
2015;Pfeiferetal.
1995;Qianetal.
2014;Stewartetal.
2016;Wuetal.
2015).
Giventhateggmassesandlarvaeofgypsymothsubspeciesareimpossibletodistinguishfromoneanotherandfromthoseofothercloselyrelatedspecies,werecentlydevelopedasuiteofTaqManassaysthatenablestherapidandaccurateidenticationof10Lymantriaspeciesandsubspeciesfrominterceptedeggmassesandolderlifestages(Stewartetal.
2016).
Theseassaysweredesignedfortheprocessingofindividualsam-ples,regardlessoftheirlifestage,duringinspectionorsurveys.
However,theyarenotoptimalforprocessingbulkpheromonetrapsamples,inasmuchasthecaptureofanAGMisarareeventinNorthAmerica,wherespecimensaretypicallyallL.
dispardispar.
DetectionofasingleAGMspecimeninsuchsampleswouldbelaboriousifallmothsweretobeassayedindividually.
Forthisreason,onlyafractionofthemothscaughtinagiventrapareprocessedforidenticationincurrentsurveyprocedures,whichcouldresultinafailuretodetectthepresenceofAGM(i.
e.
,falsenegatives).
Totacklethisissue,wedevelopedatriplexTaqManassaythatcanbeusedonbulksamples,withtheobjectiveofachievingsensitivitylevelssufcienttodetectoneAGMinabackgroundof1000EGMs.
Wemodiedourpreviouslypublishedsuiteofassays(Stewartetal.
2016)toensurethatAGMspeciesandsubspeciescouldbediscriminatedagainstEGMinasinglereaction.
Thisrequiredthatweidentifymarkersenablingthedesignofnotonlydiscriminatoryprobes,butalsodiscriminatoryprimersforeachofthethreesubassaystargetingdifferentAGMtaxa.
Thetoolwedevelopedwillproveveryusefultoperformlarge-scalescreeningforAGMadultsintrapsandprovideearlywarningofingress.
MaterialsandmethodsSNPvalidationAscanoftheSNPdatapresentedinDjoumadetal.
(2017)enabledtheidenticationofmarkersincytochromeb(cytb)andNADH-ubiquinoneoxidore-ductasechain1(ND1)potentiallysuitablefordistin-guishingthetwoAsianL.
disparsubspecies(treatedheretogether)andL.
umbrosa,respectively,fromtheotherAGMtaxa.
However,selectionoftheseSNPswasbasedonaverysmallsampleofmitochondrialgenomesequences.
Toincreasecondenceinthesuitabilityofthesemarkersforspeciesdiscrimination,sequencesgleanedfromDjoumadetal.
(2017)werealignedwithcorrespondingcytbandND1sequencesfromL.
dispar[69cytbsequences(Wuetal.
2015);Popset756785935)and19ND1sequences(LinandZhu,Genbanksubmissions,accessionnumbersJX133717–JX133736)].
WealsoampliedandsequencedtheappropriateregionoftheND1gene1231844D.
Stewartetal.
fromadditionalL.
umbrosaspecimensprovidedbyRZ,usingprimersandPCRconditionsdescribedinDjoumadetal.
(2017).
ThecytochromecoxidaseI(COI)markerusedfordiscriminationofL.
albescensandL.
postalba(treatedheretogether)fromotherAGMtaxawerethesameasthoseemployedbyStewartetal.
(2016).
SourcesofspecimensusedForthepurposeofdevelopingtheassayandconduct-inginitialdilutiontests,weusedDNAextractsfromLymantriaspecimensgeneratedinthecontextofanearlierstudy(Stewartetal.
2016);seeTable1fordetailsonsampleoriginsandproviders.
Forassayvalidation,freshL.
dispardisparmalemothswerecollectedinCanadausingpheromonetraps(milk-cartontype;DistributionSolida,Saint-Ferreol-les-neiges,Canada)containingaVaportapeIIinsecticidalstrip(Dichlorvos10%,HerconEnvironmental,Emigsville,USA)andagypsymothpheromonedispenser(()Disparlure,TreceInc.
,Adair,USA).
Thetrapsweresuspendedfromtreesataheightof*1.
5m,from15Juneto15September2017.
Mothswerecollectedandstoredinglassineenvelopesat-20°Cuntilused.
OnehundredL.
dispardisparmothsweresampledfromeachoffourcollectionsites,threeintheprovinceofQuebecandoneintheprovinceofOntario,foratotalof400moths.
Twoparallelassayswerecon-ductedforeachofthefoursetsof100moths.
Intherstassay,weusedonelegfromeachL.
dispardisparspecimen,whileforthesecondassayweusedanotherlegfromthesame100specimens,plusasingleAGMleg(fromeitherL.
disparjaponica,L.
disparasiatica,L.
albescensorL.
umbrosa;seeTable2forsources).
Thelegswerecollectedin2mLround-bottomEppendorfsafe-locktubesandplacedat-20°Cuntilused.
DNAbatchextractionDNAwasextractedfromeachbatchoflegs(30–50mgoftissue)usingtheDNeasyBloodandTissueMiniKit(Qiagen,Valencia,CA).
A5mmstainlesssteelbeadwasaddedtothemothlegsineach2mLtube.
Thetubeswerethenash-frozeninliquidnitrogenandplacedat-80°CinaprechilledTissuelyserblock.
LiquidnitrogendisruptionwasusedtomaximizethelikelihoodthatalllegswouldbesufcientlydisruptedtoproducesimilaryieldsofDNA,aswewerenotcondentthatwewouldachievethisgoalwiththemicropestlemethod.
ThesamplesweredisruptedusingaQiagenTissuelyser(Qiagen,Valencia,CA)ataspeedof26Hzfor60stoproduceanepowder.
BufferATL(1080lL)andproteinaseK(120lL)werethenaddedtoeachtube,followedbyvortexing.
Thesampleswereincubatedovernightat56°Cwithconstantgentlemixing,usingaRobbinsScienticModel400Hybridizationovensetat4rpm.
Afterincubation,sampleswerevortexedandtrans-ferredtoindividual15mLFalcontubes.
BufferALTable1LymantriaspecimensusedforassaydevelopmentSpecimenIDSpecies/subspeciesCountryoforiginRegionSupplierCFS-0008L.
dispardisparRussiaKrasnoyarskM.
KeenaCFS-0009L.
disparasiaticaRussiaPrimorskyKraiM.
KeenaCFIA-LEP-1724L.
umbrosaJapanHokkaidoD.
HoldenCFIA-LEP-1696L.
albescensJapanOkinawaIslandD.
HoldenTable2AsiangypsymothspecimensusedforassayvalidationSpecimenIDaSpecies/subspeciesCountryoforiginRegionSupplier–L.
disparjaponicaJapanHonshuM.
Keena–L.
disparasiaticaChinaBeijingM.
KeenaCFIA-LEP-0452L.
umbrosaJapanHokkaidoR.
ZahiriCFS-0016L.
albescensJapanOkinawa-JimiislandD.
HoldenaL.
disparasiaticaandL.
disparjaponicaspecimenscamefromlabcoloniesandwerenotgivenanindividuallabel123Aneedleinahaystack:amultigeneTaqManassay1845(1200lL)wasaddedtothetubesandvortexed,followedbyadditionof1200lLof100%ethanolandvortexing.
Foreachsample,thelysatewassplitandtransferredtotwo2mLEppendorfsafe-locktubesandcentrifugedat5000rpmfor1mintopelletthedebris.
Theclearedlysateswerethenaddedtotwospincolumns(DNeasyMinispincolumns;Qiagen,Valen-cia,CA)inthreestepsof600lL,andcentrifugedat8000rpmfor1minaftereachadditionoflysate.
ThewashstepswithbuffersAW1andAW2wereperformedaccordingtothemanufacturer'sinstruc-tions.
Foreachsample,DNAwaselutedfromthetwospincolumnsusing150lLofbufferAEeach.
TheelutedDNAwasrecoveredandthenpassedthroughthecolumnasecondtime;eluatesfromthetwocolumnsweresubsequentlycombinedforanalvolumeof300lLDNApersample.
TheDNAwasquantiedonaNanodropND-100spectrophotometer(ThermoFisherScientic).
PrimerandprobedesignPrimersandprobesweredesignedusingtheOligoExplorerprogram(http://www.
genelink.
com/tools/gl-downloads.
asp),exceptfor''lockednucleicacid''(LNA)probes,whichweredesignedusingtheIDTBiophysicsdesigntool(http://biophysics.
idtdna.
com/).
Thelatterwereemployedtoincreasesensitivityandspecicityofannealingincaseswhereprobe-baseddiscriminationwasprovidedbyasingleSNP;atripletofLNAnucleotidessurroundingasinglebasemis-matchsitetendstomaximizeprobespecicity(Owczarzyetal.
2011).
Similarly,weused''Ampli-cationRefractoryMutationSystem''(ARMS)pri-merswheneveronlyonediscriminatorySNPwasavailableforprimerdesign.
Withthisapproach,anarticialARMSSNPcanbeaddedadjacenttoanexisting30-endSNP,ateitherposition2or3oftheprimer,tosignicantlyincreasethediscriminatoryabilityofthatprimer(Newtonetal.
1989).
ARMSprimercombinationsweretestedagainstnon-ARMSprimerpairstodeterminethecombinationthatgavethebestdiscriminationagainstnon-targetspecieswhilehavingaminimaleffectonthespecicityofthetargetspeciesamplication.
PrimerandprobesequencesareprovidedinTable3.
qPCRconditionsforcopynumberdeterminationOptimalperformanceoftheTaqManassaydescribedhererequiresthatDNAconcentrationsbenormalized.
Tothisend,wequantiedCOIcopynumberusingaquantitativereal-timePCR(qPCR)approachandprimerstargetingCOI(Table3).
qPCRwasper-formedusinganAppliedBiosystems7500FastRealTimePCRmachine.
96-wellMicroAmpFastOpticalreal-timePCRplateswereused(AppliedBiosystems,FosterCity,CA)withMicroAmpOpticalAdhesiveFilm(AppliedBiosystems,FosterCity,CA).
The10lLamplicationreactionscontained1lLofDNAand500nMofprimers.
PCRconditionswereasfollows:a15-mininitialdenaturationstepat95°Cforenzymeactivation,followedby50cyclesof95°C,15s;58°C,30s;65°C,90s.
PCRwasperformedusingtheQiagenQuantitectSYBRGreenPCRKit(Qiagen,Valencia,CA).
Twotechnicalreplicateswererunforeachsample.
DNAcopynumberwascalculatedusingthelinearregressionofefciency(LRE)approach(Rutledge2011).
TaqManconditionsforAGMdetectioninbulksamplesTheTaqManassayswereperformedusingthesameequipmentasdescribedabove.
TheassaysweretestedinsimplexandtriplexformatsusingtheQuantitectMultiplexPCRNoRox2XMastermix(Qiagen,Valencia,CA).
Finalprimerandprobeconcentrationsweresetat500nMand100nM,respectively,ina10lLreactionvolume.
TheL.
disparasiatica/L.
disparjaponicacytbprobewasanLNAprobe,withauorescein(6-FAM)labelatthe50endandanIowaBlackFQquencheratthe30end.
TheL.
umbrosaND1probewasanLNAprobewithaHEXlabelatthe50endandanIowaBlackFQquencheratthe30end.
TheL.
albescens/L.
postalbaCOIprobewaslabelledatthe50endwithCy5andwasadoublequenchedTaoprobewithanIowaBlackRQquencheratthe30end(IntegratedDNATechnologies,Coralville,IA).
TwolLofL.
dispardisparDNA(equivalentto500,000COIcopies)wasusedforeachreaction.
PCRcondi-tionsfortheTaqManreactionswereasfollows:a15-mininitialdenaturationstepat95°Cforenzymeactivation,followedby45cyclesof95°C,15s;60°C,90s.
Threetechnicalreplicateswererunforeachsample.
1231846D.
Stewartetal.
Table3PrimersandprobesusedforAGMdetectioninbulksamplesPrimer/probeusePrimer/probenamePrimer/probesequencePrimer/probelength(bp)Primer/probeTm(°C)Ampliconsize(bp)DNAquancaonCOIBulkGENF306-3271TAGAAAATGGAGCAGGAACAGG2259.
9COIBulkGEN17WR399-4241TGATGAAATWCCAGCTAAGTGAAGAG2659.
9118LalbCOIF497-5202CCTTTATTTGTTTGAAGAGTAGGT2455.
2LalbCOIR562-5852GTTAATAATATTGTAATAGCACCC2452.
888DeteconofL.
umbrosaLumbND13C/AF233-262TATTTTTCTCCTGTATTAGCTTTTGATA32857.
9LumbND12C/TR375-395GAATTAGAAGACCATCCTGTC2158.
9127LumbND1T340-358R/C(HEX–LNA)4AC+A+CTATAAA+C+T+CCAAAAC51958.
9DeteconofL.
albescens&L.
postalbaLalbCOIF3G/AF491-520CAAATACCTTTATTTGTTTGAAGAGTAAGT3056.
6LalbCOIR2C/AR562-590GGTCAGTTAATAATATTGTAATAGCACAC2958.
4100LalbCOIT525-552(TAO–CY5)4CAGCTTTCCTTCTACTTTTATCATTACC2859.
9DeteconofL.
dispara.
+j.
6Ldajcytb3A/TF120-150GGATCTTTGTTAGCTTTATGTTTAATTTCC3058.
6LdajcytbR339-369TCCAATTATTCATGTTTGTTTTAAATTAAAA3151.
4240LdajcytbT262-273(FAM–LNA)4CT+CT+TC+A+C+G+CT1158.
71TheseprimerswillamplifyDNAfromL.
dispardispar,L.
disparasiatica,L.
disparjaponicaandL.
umbrosa.
TheycanbeusedtoquantifybulkDNAextracts,whichwilltypicallycontainonlyormostlyL.
dispardisparDNA2TheseprimersareusedtoquantifyL.
albescens/L.
postalbaDNA3SequencescontainingaredcharacterareARMSprimers,wheretheredletterrepresentstheintroducedmismatch4Inbrackets:probetype5Inprobesequences(blueletters),basesprecededbya''''signareLNAbases6L.
dispara.
j.
=L.
disparasiaticaandL.
disparjaponica123Aneedleinahaystack:amultigeneTaqManassay1847AssessingDNAintegrityinpheromonetrapsamplesSincetheassaywedevelopedismeanttobeusedonsamplescontainingmothsthathaveresidedinpheromonetrapsforvariousperiodsoftimeundereldconditions,wewantedtodeterminewhetherthequalityofDNAextractedfromsuchmothswassufcienttoensurereliablemolecularidentication.
BecausewedidnothaveaccesstofreshAGMspecimens,weranthisexperimentusingL.
dispardisparmothsonly,withtheassumptionthatDNAstabilityindeadmothswillbecomparableforallgypsymothsubspecies.
FreshlykilledL.
dispardisparmalemothswereobtainedfromInsectProductionServices(NaturalResourcesCanada,SaultSte.
Marie,Ontario)andplacedinsixdeltabi-directionalinsectstickytraps(BetterWorldManufacturing,Fresno,California)onMay20,2016,atadensityof5–6moths/trap.
Thetrapsweresuspendedonsprucetrees*3maboveground,inasemi-woodedareaontheLaurentianForestryCentrepropertyinQuebecCity,Canada.
Mothswerethencollectedonamonthlybasis,betweenMayandSeptember,andfrozenat-20°Cuntilused;time-zerosampleswereplaceddirectlyinthefreezer.
DNAwasextractedfromtwolegs(orheadplusthorax,iflegswereabsent),usingtheQiagenDNeasyBloodandTissueMiniKitprotocol,accord-ingtothemanufacturer'sinstructions.
Thetissuedisruptionstepwasperformedin180lLbufferATLusingamicro-pestle.
ThePCRprimersusedarethosedesignedbyStewartetal.
(2016)forEGMdetection(i.
e.
,subas-says3A,4Aand4B).
OnepairisspecictotheL.
dispardisparmitochondrialCOIgenewhiletheotherisspecictothenuclearFS1marker(Table4).
DNAwasquantiedanddilutedto2.
5ng/lL,andqPCRreactionswererunusing5ngofgenomicDNA.
Ctvaluesforbothmitochondrial(COI)andnuclear(FS1)DNAsweremonitoredovertime,andtheCtratiocalculatedtodetermineifmitochondrialDNAwasmorestablethannuclearDNA,asexpected.
TheqPCRconditionsweresimilartothosedescribedaboveforcopynumberdetermination(initial15mindenatura-tionstepat95°C,followedby50cyclesof95°C,15s;50°C,30s;65°C,90s).
ResultsAssaydevelopmentToachieveourgoalofdesigningamultiplexassaythatcouldbeconductedinasingletube,wetreatedasasingletaxoneachoftwotaxonpairs,namelyL.
disparasiaticaandL.
disparjaponica,andL.
albescensandL.
postalba,giventhehighwithin-pairsimilarityobservedintheirmitogenomes(Djoumadetal.
2017;Djoumadetal.
unpublisheddata).
BulktriplexAsubassayAmongtheprimersandprobesdevelopedbyStewartetal.
(2016),thesetdevelopedforL.
albescens/L.
postalba,targetingCOI,wasdirectlytransferabletothisassayasitmetourrequirementsforspecies-specicityanddualprimer/probe-baseddiscrimina-torycapabilities.
SpeciesdiscriminationwashereprovidedbytwoARMSprimersandaprobe(fordetails,seeFig.
2underthe''BulkTriplexA''tabofOnlineResource1).
However,fortheothertwosubassays(seebelow),newprimersandprobeshadtobedesignedfromalternativemarkergenes,assuf-cientdiscriminatorySNPsforprimerdevelopmentcouldnotbefoundintheCOIgene.
Table4PrimersusedforL.
dispardisparDNAamplication,forthepurposeofassessingDNAintegrityovertimePrimernamePrimersequencePrimerlength(bp)PrimerTm(°C)LddCOI8RF487-513TGATCAAATACCTTTATTTRTTTGAAG2753.
2/51.
0LddCOIR574-603GTATTTAAATTTCGGTCAGTTAATAATATT3052.
5FS1F2-16GATGGTGGGTGTCGT1554.
9FS1R176-200GATTCATCTGATCCTGATAATTCAT2554.
91231848D.
Stewartetal.
BulktriplexBsubassayFortheL.
disparasiatica/L.
disparjaponicataxonpair,analysisofthemitochondrialgenomesequencesreportedbyDjoumadetal.
(2017)ledtotheidenti-cationofdiscriminatorySNPsinthecytbgene.
Comparisonofthistargetsequencewithequivalentsequencesgleanedfrompublicdatabasesindicatedthatinter-individualsequencehomogeneityofthecytbgeneinL.
disparasiaticaandL.
disparjaponicawassufcienttoyielddiscriminatoryamplication;dis-criminationwashereprovidedbyaforwardARMSprimerandanLNAprobe(Fig.
1,alignmentprovidedasanexample;seealsoFig.
3underthe''BulkTriplexB''tabofOnlineResource1).
BulktriplexCsubassayWithrespecttoL.
umbrosa,ananalyticalapproachsimilartotheonedescribedabovefortheselectionofcytbpointedtoND1asthemostlikelysourceofSNPsthatwouldenablediscriminationofthisspeciesfromothertaxa.
AsnoL.
umbrosaND1sequencesotherthantheonewehadreported(Djoumadetal.
2017)wereavailableinpublicdatabases,inter-individualsequencestabilitywasconrmedthroughthesequenc-ingofthisregionfromadditionalspecimens.
Forthissubassay,discriminationwasprovidedbytwoARMSprimersandanLNAprobe(seeFig.
2under''BulkTriplexC''tabofOnlineResource1).
InitialtestsofthebulkassayusingDNAdilutionseriesAsarstassessmentoftheperformanceofthebulkassay,testswereconductedinbothsimplexandtriplexformats,usingDNAdilutionseries.
Forexample,1000,100,10and0COIcopiesofL.
albescensDNAwereassayedaloneorincombinationwith1millionCOIcopiesofL.
dispardisparDNA,generatingdilutionratiosof1:103,1:104,1:105and0:106.
AmplicationcurvesobservedinthepresenceofL.
dispardisparDNAweresimilartothosegeneratedbyL.
albescensDNAalone,atallthreenon-zerodilutionstested(Fig.
2a;simplexformat),indicatingthatthisFig.
1AlignmentoftheportionofthecytbgeneusedtodesignprimersandprobesfordevelopmentofaTaqManassaythatcandetectDNAfromeitherLymantriadisparasiaticaorL.
disparjaponica,whenmixedwithDNAfromL.
dispardisparand/orotherLymantriaspecies.
Sequencesshownatthetopofthegreybarsare(fromlefttoright)thoseoftheforwardprimer,probeandreverseprimer.
Redletters:ARMSbaseinforwardprimerandLNAbasesinprobe.
Left-mostcolumn:Genbankaccessionnumbers;cytbsequencesforL.
albescens,L.
postalba,L.
mathura,L.
monachaandL.
xylinaweregeneratedwithinthecontextofaseparate,ongoingstudy(Djoumadetal.
,unpub-lisheddata)123Aneedleinahaystack:amultigeneTaqManassay18491231850D.
Stewartetal.
subassaycoulddetect1L.
albescensspecimenagainstabackgroundof100,000L.
dispardisparmoths.
Noamplicationwasobservedwhen106COIcopiesofL.
dispardisparDNAwereassayedalone,indicatingthattheprimersandprobedesignedforthissubassayarehighlyspecictoL.
albescens.
Similarresultswereobtainedintriplexmode(seeTable1under''BulkTriplexA''tabofOnlineResource1).
TheBulkTriplexBsubassay,targetingL.
disparasiaticaandL.
disparjaponica,providedcompletediscriminationagainstL.
dispardisparata1:1000dilution(i.
e.
,103:106),butitsperformancegraduallydeclinedathigherdilutions(Fig.
2b;simplexformat).
Compara-bleresultswereobtainedintriplexformataswellasintestsinvolvingalternativedilutionseriesruninsimplexformat(see''BulkTriplexB''tabofOnlineResource1).
Finally,theBulkTriplexCsubassay,whichtargetsL.
umbrosa,showedadegreeofsensitivitysimilartothatobtainedwiththeBulkTriplexAsubassay(Fig.
2c;simplexformat).
Again,assayperformancewasequallygoodwhenrunintriplexformat(seeTable1undertheBulkTriplexCtabofOnlineResource1).
AssayvalidationTovalidatethepresentassayusingDNAextractsfromdifferentmoths(asopposedtovaryingproportionsofDNAused,asintheabovedilutionseries),weusedDNAextractedfrom100L.
dispardisparlegs(1leg/moth),eitheraloneorincombinationwithalegfromanAGMspecimen(L.
disparasiatica,L.
disparjaponica,L.
albescensorL.
umbrosa),resultingina1:100AGM:EGMlegratio.
Noamplicationwasobservedforanyofthefournegativecontrols(L.
dispardisparalone,obtainedfromfourdifferentcollectionsites;106COIcopies),whereaseachofthefourtestassaysgeneratedapositiveamplication(Fig.
3;seealsoOnlineResource2).
TodeterminewhetheramixtureofAGMtaxapresentinagivensamplecouldbedetectedandidentiedinasingleassay,threeoftheaboveDNAextractswerecombinedinequalaliquots,effectivelydilutingcopynumberforL.
umbrosa,L.
albescensandL.
disparasiaticabyafactorofthree.
Allthreesamplesgeneratedaspecies-specic,positiveamplication(OnlineResource2).
DNAintegrityinpheromonetrapsamplesDNAextractedfromL.
dispardisparmothsthatwereleftinpheromonetrapsundereldconditionsfordifferentperiodsoftime,wasassessedforitsreliabil-ityinanassaysuchastheonereportedhere,usingqPCRprimersdevelopedforEGMdetection(Stewartetal.
2016).
Allofthesamplesatallofthetimepoints,forbothmarkerstested,yieldedpositiveresultsforthepresenceofEGM(Table5).
Overall,theDNAappearstohaveremainedingoodconditionoverthe4monthstheinsectswereleftinthetraps,asjudgedbytherelativestabilityoftheCtvaluesobtainedforbothCOIandFS1overtime(Table5).
Nonetheless,aslightincreaseinvariationinCtvaluesbecameapparentinmonths3and4(T3andT4),withtwosamplesshowingsignicantshifts(T3-2,COIandFS1;T4-2,FS1;Table5),suggestingtheDNAinthesemothshadbegundegradingatthetimetheyweresampled,albeitnottoadegreesufcienttojeopardizetheassay.
ThegreaterCOI:FS1ratiorecordedforthesetwosamples(Table5)alsopointedtothehigherstabilityofmitochondrialDNA,versusnuclearDNA,aspredicted.
DiscussionDetectinganextremelyrareinvasivespeciesagainstabackgroundofanabundant,alreadyestablishedcloserelativeisakintolookingforaneedleinahaystack.
Forexample,networksofpheromonetrapsdeployedforthedetectionofAGMcaneasilyproducesituationswhereanAGMindividual,ifcaught,isgreatlyoutnumberedbylocalEGMs.
Unlikeothermolecularassaysdesignedbyotherstodetectasinglespeciesinalargebackgroundofanotherspecies(e.
g.
,Zinketal.
2017,2018),theassayreportedherehadtoovercometheadditionalhurdleofdistinguishingtaxawithintheAGMcomplex.
Tothisend,wedevelopedindepen-dentsetsofqPCRprimersandTaqManprobesforAGMspecies/subspecies,usingSNPsidentiedbFig.
2AssessmentofthesensitivityofeachTaqMansubassaythroughdilutionoftargetspeciesDNA.
Adilutionseriesof1000,100and10COIcopiesofaL.
albescens(LYM01sample),bL.
disparasiatica(CFS9sample),andcL.
umbrosa(CFIA-LEP1724sample)wasruninsimplexmode,with(red)andwithout(blue)abackgroundofonemillionCOIcopiesofL.
dispardispar(CFS8sample).
Twotechnicalreplicatesareshownforeachassay123Aneedleinahaystack:amultigeneTaqManassay1851earlierinthreedistinctmitochondrialgenes(Djoumadetal.
2017).
ThisapproachenabledthedevelopmentofahighlysensitivetriplexTaqManassaycapableofdetectingasinglespecimenofL.
umbrosaorL.
albescens(orL.
postalba)inabackgroundofupto100,000L.
dispardisparmoths,andasinglespecimenofL.
disparasiaticaorL.
disparjaponicainabackgroundofatleast1000L.
dispardisparmoths,basedonDNAdilutionseries(Fig.
2).
Inthevalida-tiontestsweconductedusingDNAextractedfrombatchesofmothlegs,theabove-mentionedAGMtaxawerealleasilydetectedwhenpresentina1:100AGM:EGMlegratio(Fig.
3),aproportionbelievedtobenearthemaximumtobeeverprocessedopera-tionally.
Ofcourse,giventhatDNAextractionefciencyforsuchsamplesmaynotbeuniformamonglegs,theproportionofDNAfromasingleAGMlegcould,bychance,belowerthanthatsuggestedbythe1:100legratio.
However,giventhehighsensitivityofthisassay(C1:1000AGM:EGM,asindicatedbythedilutionseries),thelikelihoodofafailedAGMdetectionisverylow.
OuroperationalgoalindevelopingthisassaywasquitedifferentfromtheonewepursuedindesigningourearliersuiteofTaqManassays(Stewartetal.
2016),whichtargetedindividualsamplesofimmaturestagesofgypsymoths.
Inanefforttostreamlinethebulkassay,wedidnotattempttoidentifyspeciesotherthanthoseconsideredmembersoftheAGMcomplex.
However,welookedformarkersthatwouldenableAGMdiscriminationagainstbothEGMandotherLymantriaspeciesthatcouldaccidentallyndtheirwayintogypsymothpheromonetraps,includingthoseconsidered''OTLS''(''otherthreateningLymantriaspecies'')inourearlierwork(Stewartetal.
2016);itshouldbenotedthattheseOTLSmothsareFig.
3ValidationofthetriplexTaqManassayusingDNAextractedfrommothlegs.
Theresultsofeightindependentassays,eachrunintriplexmode,arecombinedhereonasinglegraphforcomparativepurposes.
Foreachoffour''negativecontrol''assays,weusedDNAextractedfrom100legsobtainedfrom100L.
dispardisparspecimenscollectedatoneoffourCanadianlocations(Notre-Dame-des-Laurentides,Quebec—NDL;Denwood,ON—DEN;Victoriaville,QC—VIC;QuebecCity,QC—QUE),whileforeachofthefour''test''assays,weusedDNAextractedfrom100legsobtainedfromthesamesetsof100L.
dispardisparspecimens,plusasingleAGMlegobtainedfromeitherL.
disparjaponica,L.
disparasiatica,L.
albescensorL.
umbrosa(seeTable2forsources).
Noamplicationwasobservedforthefournegativecontrols(L.
dispardisparalone),whereaseachofthefourtestassaysgeneratedapositiveamplication.
Differencesinuorescencelevels(Fmax)aretheresultofdifferentuorophoresbeingusedforprobedesign:FAMhasthehighestuorescence,followedbyCy5andHEX.
Twotechnicalreplicatesareshownforeachassay1231852D.
Stewartetal.
morphologicallydistinguishablefromAGMandEGM,althoughthisisnottrueofallLymantriaspecies.
Inaddition,wehadtomakesurethatspeciesdiscrimination,forthepresentassay,wasprovidedbybothprimersandprobes(asopposedtoprobe-only-baseddiscrimination)soastoavoidsituationswhereTable5DNAextractionandqPCRresults(Ctvalues)fortheexperimentassessingDNAintegrityinmothsheldundereldconditionsfordifferentlengthsoftimeSampleIDaCollectiondate(2016)StartingmaterialDNAconc.
(ng/lL)b260:280cMeanCtCOIMeanCtFS1RatioCOI:FS1dT0-120May2legs7.
71.
3814.
2320.
6183.
71T0-220May2legs8.
11.
3714.
6620.
7769.
09T0-320May2legs9.
81.
3913.
9120.
79117.
48T0-420May2legs5.
31.
3413.
8120.
71119.
35Average14.
1520.
7297.
41CV2.
7%0.
4%25.
7%T1-121June2legs6.
81.
2415.
4621.
4764.
13T1-221June2legs8.
91.
4215.
9522.
77112.
91T1-321JuneHead/thorax55.
41.
8516.
5621.
2425.
62T1-421June2legs11.
31.
5215.
8721.
1238.
08T1-521June2legs9.
21.
5914.
4220.
5670.
42T1-621JuneHead/thorax38.
51.
7215.
5921.
5461.
47Average15.
6421.
4562.
10CV4.
5%1.
9%48.
64%T2-121July2legs8.
11.
3614.
7320.
7263.
44T2-221July2legs10.
51.
6816.
1122.
5385.
51T2-421July2legs9.
01.
9614.
1520.
3473.
01T2-521July2legs4.
53.
0914.
2020.
0356.
58T2-621July2legs8.
42.
0414.
8420.
5752.
94Average14.
8120.
8466.
30CV5.
3%5.
4%19.
9%T3-122Aug2legs5.
01.
3916.
2522.
3166.
51T3-222Aug2legs10.
71.
7519.
0227.
40332.
68T3-422Aug2legs4.
52.
3214.
8821.
4595.
30T3-522AugHead/thorax62.
01.
8516.
5920.
8819.
58T3-622Aug2legs4.
02.
3515.
9121.
7256.
43Average16.
5322.
75114.
10CV9.
3%13.
4%109.
7%T4-220SeptHead/thorax4.
23.
6315.
9327.
503038.
10T4-420SeptHead/thorax27.
21.
9816.
1822.
3270.
63T4-520Sept2legs6.
71.
7815.
8522.
0272.
12T4-620Sept2legs6.
32.
8114.
7120.
8470.
02Average15.
6723.
17812.
72CV4.
2%12.
8%182.
5%aSampleID:T0toT4refertosamplingtimes,while-1to-6refertotrapnumber,exceptforT0samples,whichwerenotplacedintraps,buttransferreddirectlytothefreezer.
Samplesweretakenfromtrap#3onlyonceasbirdsenteredthetrapaftertheJunesamplingdateanddestroyedallmothsbThisconcentrationwasnormalizedto2.
5ng/lLforrunningtheassayc260:280purityratiodCalculatedusingtheformula:2^(CtFS1-CtCOI)123Aneedleinahaystack:amultigeneTaqManassay1853allDNAsinagivenbulksampleareampliedevenly,resultinginthedrowningofthesignalproducedbyaprobeannealingtolow-abundanceAGM-relatedDNAs.
ExamplesofotherassaysdesignedforthedetectionofinvasivespeciesinbulkpheromonetrapsamplesincludetheonedevelopedforthedetectionoftheinvasivenoctuidHelicoverpaarmigeraamonglargenumbersofthenativecongenerH.
zea(Zinketal.
2017),andanotheronethattargetsthesilverYmoth,Autographagamma,inabackgroundofseveralcloselyrelatedspecies(Zinketal.
2018).
Forthesetwoassays,theauthorsoptedfortheuseofadifferent,morerecentPCRtechnologyknownasdropletdigitalPCR(ddPCR),largelybecauseofitshighreportedsensitivity(Milburyetal.
2014).
AlthoughddPCRclearlyprovidedexcellentperformanceforthesetwoassays,thelevelofsensitivityweachievedinthepresentTaqManqPCRassayisatleastasgoodasthatreportedbythisgroup(1:1000,Zinketal.
2017;1:200,Zinketal.
2018).
TheqPCRTaqManassayalsohastheadvantageofrequiringatypeofthermocyclerthatislessexpensiveandmoreubiquitousthantheddPCRequipmentemployedbyZinketal.
(2017,2018).
Inaddition,theworkinvolvedindataanalysisandinterpretationforthepresentassayissimplerandmoreeasilyautomatedthanitisfortheddPCRtechnology.
Theresultsreportedhereforthecytbsubassay(L.
disparasiatica/L.
disparjaponica),pointtothesomewhatlowerdiscriminatorypowerofthisassayagainstL.
dispardisparascomparedtotheothertwosubassays(Fig.
2),albeitatalevelthatremainsmorethanadequate(1:1000AGM:EGM)foroperationalbulkprocessing.
SeveralfactorsarelikelyinvolvedinthecollapseoftheAGM-positivecurvethatweobserveatdilutions[1:1000.
First,mitochondrialsequencesofL.
dispardisparshowgreateroverallsimilaritytothoseofL.
disparasiaticaandL.
disparjaponicathantothoseofL.
umbrosaandL.
albescens,resultinginslightlylowerprimer-baseddiscrimina-tion.
Compoundingthiseffectisthesignicantlylargeramplicongeneratedbythissubassay(240bp)ascomparedtothoseproducedbytheothertwosubas-says(100and127bp;seeTable3).
Nonetheless,althoughtheprimersspecictotheL.
disparasiatica/L.
disparjaponicataxonpairlikelygenerateanampliconinthepresenceofveryhighexcessL.
dispardisparDNA,thespecicLNAprobeisnotexpectedtoefcientlyannealtoit.
ThemitochondrialgenomesofL.
disparasiaticaandL.
disparjaponicadisplayastrikinglyhighdegreeofsequencesimilarity(Djoumadetal.
2017),makingitverydifculttodesignaTaqManassaythatcoulddistinguishonesubspeciesfromtheotherinthecontextofaproceduremeanttoprocessbulksamples.
Inaddition,tokeepthepresentassaysimple(i.
e.
,singletubemultiplexqPCR),anabilitytodistinguishthetwoAsiansubspecieswouldhaverequiredthatweincreasetheassay'smultiplexingcapabilitytofour.
Forthesereasons,weoptedforthesimplertaxoncombinationusedhere,withanunderstandingthat,shouldanygiventrapsamplegenerateapositiveL.
disparasiatica/L.
disparjaponicasignal,theexactsubspeciesinvolvedcouldsubsequentlybeidentiedthroughananalysisofDNAextractsfromindividualmothsusingourearlierassay(Stewartetal.
2016).
Gypsymothpheromonetrapsaretypicallyemptiedoftheircontentonlyattheendofthetrappingseason.
Asaconsequence,mothsrecoveredattheendoftheightperiodmayhavespentupto3–4monthsintheeld,exposedtoenvironmentalconditionsthatcouldbedetrimentaltoDNAintegrity.
Giventhatcompro-misedDNAqualitycouldseriouslyjeopardizethereliabilityofthepresentbulkassay,weassessedmothDNAstabilityundereldconditions,usingL.
dispardisparmothsasAGMsurrogates.
OurresultspointtoadegreeofDNAintegritythatwouldbesufcientforthebulkassaytoprovideaccuratespeciesdetectionwhenrunonmothsthathavespentupto4monthsintheeld,atleastunderconditionssimilartothosetestedhere(Table5;OnlineResource3).
However,someofthesamplescollectedinthesecondhalfoftheexperimentalperioddisplayedasomewhatdelayedCtrelativetoearlysamples,suggestingthattheirDNAmayhavebeguntolosesomeofitsinitialquality.
ToavoidDNAdegradationaftereldcollection,wethereforerecommendthatsamplesbefrozenuntilprocessedforspeciesidentication.
ItisimportanttonotethatthepresentassaywasdesignedtoidentifyAGMspecimensoriginatingfromtheirmostlikelysourceswheninterceptedinNorthAmerica,i.
e.
,Japan,SouthKorea,easternChinaandtheRussianFarEast.
Indeed,giventhatourassayreliesonmitochondrialmarkers,itwillnotdetectspecimensfromcentralAsiaiftheypossessanL.
dispardisparmitochondrialhaplotype,eventhoughtheirnucleargenomesandsomeoftheirbiologicaltraits(e.
g.
,femaleightcapability)maydisplay1231854D.
Stewartetal.
greatersimilaritytoL.
disparasiaticathantoL.
dispardispar,asearlierstudieshavesuggested(Djoumadetal.
2017;Keenaetal.
2008;Picqetal.
2018;Stewartetal.
2016;Wuetal.
2015).
However,becausecountries/regionswherepopulationsshowingthistypeofinter-subspeciesintrogressionarerelativelyfarfromthosementionedabove,thelikelihoodofsuchAGM-likehybridsndingtheirwaytoNorthAmericaiscurrentlylow.
Nonetheless,wearenowconductingworkaimedatidentifyingnuclearmarkersthatcouldbeaddedtoourcurrentpanelandusedtoidentifygypsymothsfromcentralAsia.
Finally,itshouldbepointedoutthat,withtheadventofthree-channelportableqPCRdevices,theassaydescribedherecouldalsobeusedforon-sitescreeningofgypsymotheggsamplesinterceptedduringshipinspections,wheneverarapidanswertothequestion''isthisAGM''mayberequired.
Suchsamplescouldsubsequentlybere-processedusingtheStewartetal.
(2016)suiteofassaysforamoredetailedanalysis.
AcknowledgementsWethankM.
Keena(USDAForestService),S.
Picq(NaturalResourcesCanada;NRCan),andD.
Holden(CanadianFoodInspectionAgency;CFIA)forprovidingspecimensusedinthisstudy.
WearealsogratefultoV.
Nealis(NRCan),whobroughttoourattentiontheneedforanAGMbulkassay.
Finally,wethankJ.
Delisle,A.
LabrecqueandS.
Picq(NRCan)forprovidingtheweatherdatapresentedintheOnlineResource3.
ThisworkwasfundedbyGenomeCanada,GenomeBritishColumbia,GenomeQuebec,theCanadianForestService(GenomicsResearchandDevelopmentInitiative,GRDI)andtheCFIA,throughaGenomicsApplicationsPartnershipProgram(GAPP6102;GenomeCanada)grant.
OpenAccessThisarticleisdistributedunderthetermsoftheCreativeCommonsAttribution4.
0InternationalLicense(http://creativecommons.
org/licenses/by/4.
0/),whichpermitsunre-stricteduse,distribution,andreproductioninanymedium,providedyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinktotheCreativeCom-monslicense,andindicateifchangesweremade.
ReferencesArimotoM,IwaizumiR(2014)IdenticationofJapaneseLy-mantriaspecies(Lepidoptera:Lymantriineae)basedonPCR–RFLPanalysisofmitochondrialDNA.
ApplEntomolZool49:159–169ArmstrongKF,BallSL(2005)DNAbarcodesforbiosecurity:invasivespeciesidentication.
PhilosTransRSocBLond360:1813–1823ArmstrongKF,McHughP,ChinnW,FramptonER,WalshPJ(2003)TussockmothspeciesarrivingonimportedusedvehiclesdeterminedbyDNAanalysis.
NZPlantProt56:16–20BallSL,ArmstrongKF(2006)DNAbarcodesforinsectpestidentication:atestcasewithtussockmoths(Lepidoptera:Lymantriineae).
CanJForRes36:337–350BogdanowiczSM,WallnerWE,BellJ,OdellTM,HarrisonRG(1993)Asiangypsymoths(Lepidoptera:Lymantriineae)inNorthAmerica:evidencefrommoleculardata.
AnnEntomolSocAm86:710–715BogdanowiczSM,MastroVC,PrasherDC,HarrisonRG(1997)MicrosatelliteDNAvariationamongAsianandNorthAmericangypsymoths(Lepidoptera:Lymantriineae).
AnnEntomolSocAm90:768–775BogdanowiczSM,SchaeferPW,HarrisonRG(2000)Mito-chondrialDNAvariationamongworldwidepopulationsofgypsymoths,Lymantriadispar.
MolPhylogenetEvol15:487–495BoykinLM,ArmstrongKF,KubatkoL,DeBarroP(2012)Speciesdelimitationandglobalbiosecurity.
EvolBioinfo8:1–37BradshawCJAetal(2016)Massiveyetgrosslyunderestimatedglobalcostsofinvasiveinsects.
NatCommun7:12986ChenF,ShiJ,LuoY-Q,SunS-Y,PuM(2013)Geneticchar-acterizationofthegypsymothfromChina(Lepidoptera,Lymantriineae)usingintersimplesequencerepeatsmarkers.
PLoSONE8:e73017ChenF,LuoY,KeenaMA,WuY,WuP,ShiJ(2015)DNAbarcodingofgypsymothsfromChina(Lepidoptera:Ere-bidae)revealsnewhaplotypesanddivergencepatternswithingypsymothsubspecies.
JEconEntomol109:366–374deWaardJR,MitchellA,KeenaMA,GopurenkoD,BoykinLM,ArmstrongKFetal(2010)TowardsaglobalbarcodelibraryforLymantria(Lepidoptera:Lymantriinae)tussockmothsofbiosecurityconcern.
PLoSONE5:e14280DjoumadA,NisoleA,ZahiriR,FreschiL,PicqS,Gundersen-RindalDE,SparksME,DewarK,StewartD,MaarouH,LevesqueRC,HamelinRC,CussonMC(2017)Compar-ativeanalysisofmitochondrialgenomesofgeographicvariantsofthegypsymoth,Lymantriadispar,revealsapreviouslyundescribedgenotypicentity.
SciRep7:14245GarnerKJ,SlavicekJM(1996)Identicationandcharacteri-zationofaRAPD-PCRmarkerfordistinguishingAsianandNorthAmericangypsymoths.
InsectMolBiol5:81–91HarrisonRG,ODellTM(1989)MitochondrialDNAasatracerofgypsymothorigins.
In:WallnerWE,McManusKA(eds)Proceedings,Lymantriidae:acomparisonoffeaturesofNewandOldWorldtussockmoths.
USDAForestSer-vice,NortheasternForestExperimentStation,Broomall,pp265–273IslamMS,BarrNB,BraswellWE,MartinezM,LedezmaLA,MolongoskiJetal(2015)Amultiplexreal-timePCRassayforscreeninggypsymoths(Lepidoptera:Erebidae)intheUnitedStatesforevidenceofanAsiangenotype.
JEconEntomol108:2450–2457123Aneedleinahaystack:amultigeneTaqManassay1855KangTH,LeeK-S,LeeH-S(2015)DNAbarcodingoftheKoreanLymantriaHu¨bner,1819(Lepidoptera:Erebidae:Lymantriinae)forquarantineinspection.
JEconEntomol108:1596–1611KangTH,HanSH,LeeHS(2017)GeneticstructureanddemographichistoryofLymantriadispar(Linnaeus,1758)(Lepidoptera:Erebidae)initsareaoforiginandadjacentareas.
EcolEvol7:9162–9178KeenaMA,CoteMJ,GrinbergPS,WallnerWE(2008)WorlddistributionoffemaleightandgeneticvariationinLy-mantriadispar(Lepidoptera:Lymantriidae).
EnvironEntomol37:636–649LackovicN,BertheauC,StaufferC,PernekM,AvtzisD(2015)GeneticsplitbetweencoastalandcontinentalpopulationsofgypsymothseparatedbyDinaricAlps.
JApplEntomol139:721–726LiebholdA,MastroV,SchaeferPW(1989)LearningfromthelegacyofLeopoldTrouvelot.
BullEntomolSocAm35:20–22MilburyCA,ZhongQ,LinJ,WilliamsM,OlsonJ,LinkDR,HutchisonB(2014)DetermininglowerlimitsofdetectionofdigitalPCRassaysforcancer-relatedgenemutations.
BiomolDetectQuant1:8–22NealisVG(2002)GypsymothinCanada:casestudyofaninvasiveinsect.
In:ClaudiR,NantelP,Muckle-JeffsE(eds)AlieninvadersinCanada'swaters,wetlands,andforests,pp151–159NealisVG(2009)Stillinvasiveafteralltheseyears:keepinggypsymothoutofBritishColumbia.
ForChron85:593–603NewtonCR,GrahamA,HeptinstallLE,PowellSJ,SummersC,KalshekerNetal(1989)AnalysisofanypointmutationinDNA.
Theamplicationrefractorymutationsystem(ARMS).
NucleicAcidsRes17:2503–2516OwczarzyR,YouY,GrothCL,TataurovAV(2011)Stabilityandmismatchdiscriminationoflockednucleicacid-DNAduplexes.
Biochemistry50:9352–9367PfeiferTA,HumbleLM,RingM,GrigliattiTA(1995)Char-acterizationofgypsymothpopulationsandrelatedspeciesusinganuclearDNAmarker.
CanEntomol127:49–58PicqS,KeenaM,HavillN,StewartD,PouliotE,BoyleB,LevesqueRC,HamelinRC,CussonM(2018)Assessingthepotentialofgenotyping-by-sequencing-derivedsinglenucleotidepolymorphismstoidentifythegeographicori-ginsofinterceptedgypsymoth(Lymantriadispar)speci-mens:aproof-of-conceptstudy.
EvolAppl11:325–339PogueMG,SchaeferPW(2007)AreviewofselectedspeciesofLymantriaHu¨bner[1819](Lepidoptera:Noctuidae:Lymantriinae)fromsubtropicalandtemperateregionsofAsia,includingthedescriptionofthreenewspecies,somepotentiallyinvasivetoNorthAmerica.
USDAForestSer-vice,WashingtonQianL,AnY,SongJ,XuM,YeJ,WuCetal(2014)COIgenegeographicvariationofgypsymoth(Lepidoptera:Lymantriineae)andaTaqManPCRdiagnosticassay.
DNABarcodes2:10–16RutledgeRG(2011)AJavaprogramforLRE-basedreal-timeqPCRthatenableslarge-scaleabsolutequantication.
PLoSONE6:e17636StewartD,ZahiriR,DjoumadA,FreschiL,LamarcheJ,HoldenD,CervantesS,OjedaDI,PotvinA,NisoleA,BeliveauC,CapronA,KimotoT,DayB,YuehH,DuffC,LevesqueRC,HamelinRC,CussonM(2016)Amulti-speciesTaq-ManPCRassayfortheidenticationofAsiangypsymoths(Lymantriaspp.
)andotherinvasivelymantriinesofbiosecurityconcerntoNorthAmerica.
PLoSONE11:e0160878USDAPestalert(2016)Asiangypsymoth.
APHIS81-35-027leaet.
USDAAnimalandPlantHealthInspectionService,RiversidePark,MD.
https://www.
aphis.
usda.
gov/publications/plant_health/content/printable_version/fs_phasiangm.
pdf.
Accessed6Mar2019WuY,MolongoskiJJ,WinogradDF,BogdanowiczSM,LouyakisAS,LanceDRetal(2015)Geneticstructure,admixtureandinvasionsuccessinaHolarcticdefoliator,thegypsymoth(Lymantriadispar,Lepidoptera:Erebidae).
MolEcol24:1275–1291ZinkFA,TembrockLR,TimmAE,FarrisRE,PereraOP,GilliganTM(2017)AdropletdigitalPCR(ddPCR)assaytodetectHelicoverpaarmigera(Lepidoptera:Noctuidae)inbulktrapsamples.
PLoSONE12:e0178704ZinkFA,TembrockLR,TimmAE,GilliganTM(2018)AddPCRassayforidenticationofAutographagamma(Noctuidae:Plusiinae)inbulktrapsamples.
JEconEnto-mol111:1490–1495Publisher'sNoteSpringerNatureremainsneutralwithregardtojurisdictionalclaimsinpublishedmapsandinstitutionalafliations.
1231856D.
Stewartetal.