ebook img

Genomic Mining of Phylogenetically Informative Nuclear Markers PDF

26 Pages·2016·2.66 MB·English
by  
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Genomic Mining of Phylogenetically Informative Nuclear Markers

RESEARCHARTICLE Genomic Mining of Phylogenetically Informative Nuclear Markers in Bark and Ambrosia Beetles DarioPistone,SigridMugu,BjarteHenryJordal* DepartmentofNaturalHistory,TheUniversityMuseum,UniversityofBergen,PB7800,NO-5020Bergen, Norway *[email protected] a11111 Abstract Deeplevelinsectrelationshipsaregenerallydifficulttoresolve,especiallywithintaxaofthe mostdiverseandspeciesrichholometabolousorders.Inbeetles,themajordiversityoccurs inthePhytophaga,includingcharismaticgroupssuchasleafbeetles,longhornbeetlesand weevils.Barkandambrosiabeetlesarewoodboringweevilsthatcontribute12percentof OPENACCESS thediversityencounteredinCurculionidae,oneofthelargestfamiliesofbeetleswithmore than50000describedspecies.PhylogeneticresolutioningroupsofCretaceousagehas Citation:PistoneD,MuguS,JordalBH(2016) GenomicMiningofPhylogeneticallyInformative provenparticularlydifficultandrequireslargequantityofdata.Inthisstudy,weinvestigated NuclearMarkersinBarkandAmbrosiaBeetles. 100nucleargenesinordertoselectanumberofmarkerswithlowevolutionaryratesand PLoSONE11(9):e0163529.doi:10.1371/journal. highphylogeneticsignal.APCRscreeningusingdegenerateprimerswasappliedto26dif- pone.0163529 ferentweevilspecies.Weobtainedsequencesfrom57ofthe100targetedgenes. Editor:BenJMans,OnderstepoortVeterinary Sequencesfromeachnuclearmarkerwerealignedandexaminedfordetectingmultiple Institute,SOUTHAFRICA copies,pseudogenesandintrons.Phylogeneticinformativeness(PI)andthecapacityfor Received:April13,2016 reconstructionofpreviouslyestablishedphylogeneticrelationshipswereusedasproxies Accepted:September10,2016 forselectingasubsetofthe57amplifiedgenes.Finally,weselected16markerssuitable Published:September26,2016 forlarge-scalephylogeneticsofScolytinaeandrelatedweeviltaxa. Copyright:©2016Pistoneetal.Thisisanopen accessarticledistributedunderthetermsofthe CreativeCommonsAttributionLicense,which permitsunrestricteduse,distribution,and reproductioninanymedium,providedtheoriginal authorandsourcearecredited. Introduction DataAvailabilityStatement:Allrelevantdataare Inthepostgenomicera,obtainingwellresolvedandhighlysupportedmolecularphylogeniesof withinthepaperanditsSupportingInformation files. hyper-diverseeukaryoticlineagescontinuestorepresentamajorchallenge.Previousattempts oninvestigatingphylogeneticrelationshipsinbeetleshavedemonstratedrecurrentproblemsin Funding:Thisprojectwasfundedbygrant214232/ resolvingdeeperrelationshipssuchasthosebetweenthefourbeetlesuborders,butalsomuch F20fromtheNorwegianResearchCouncil.The funderhadnoroleinstudydesign,datacollection youngerdivergences[1–4].Oneofthemostproblematicgroupsincludestheweevils,where andanalysis,decisiontopublish,orpreparationof themajorityoftribesandsubfamiliesremainunresolveddespiteconsiderableeffortsinassem- themanuscript. blingmoleculardata[5–8].BarkandambrosiabeetlesinthesubfamilyScolytinaerepresenta CompetingInterests:Theauthorshavedeclared weevillineagewheremuchefforthasbeeninvestedindevelopingmolecularmarkersforphylo- thatnocompetinginterestsexist. geneticanalysis[9,10].Nevertheless,resolutionbetweenmanyCretaceousrelationships PLOSONE|DOI:10.1371/journal.pone.0163529 September26,2016 1/26 GenomicMininginBarkandAmbrosiaBeetles remainsratherlow[11],emphasizingthescarcenessofmolecularmarkerstoresolvethispar- ticularphylogeny. Sofar,thevastmajorityofphylogeneticstudiesonbeetleswerebasedonmarkerssuchas ribosomalRNAsandmitochondrialcytochromeoxidaseIandIIgenes[8,12–15].Withthe exceptionofnuclearribosomalgenes(18sand28srRNAs)aremostmarkersusefulforthereso- lutionofCenozoicdivergences,showinglackofphylogeneticsignalforCretaceoustimeframes [10].Inthelastyears,agrowingnumberofphylogeneticstudiesonbeetleshavestartedto includenuclearproteincodinggenes,especiallyEF-1α,CAD,ArgK,andwingless[11,16,17], whicharealsowidelyusedinotherinsecttaxa[18–21].However,arelativelylimitedamount ofworkhasbeendonetodiscoverandselectadditionalnucleargenesforbeetlesystematics [22,23–25],andallstudiestodatewerebasedonlessthan10molecularmarkers[26,27]. Therefore,obtainingahighdegreeofphylogeneticresolutioninbeetlesisdifficult;adirectcon- sequenceofhighspeciesdiversityandalimitednumberofinformativemarkers. Thefirststudiesontheutilityofproteincodinggenesininsectsystematicsdatebackto morethan20yearsago[28–30].Theadvancementofinsectphylogenieshaslargelybeen drivenbythedevelopmentofnewmarkersinLepidoptera[31].Atpresent,dozensofnuclear markerscanbechosentoinvestigateLepidopteraphylogenyatvariousranks[18,32–36]. Hymenopteraisanothergroupwhereaconsistentnumberofnuclearmarkershavebeendevel- oped[37–39].Althoughsimilarstudieshavebeencarriedoutinotherinsectgroupssuchas Diptera[40–42],themajorityoftheremaininginsectorderspresentasituationmoresimilarto Coleopterawithfewpublishedmarkersconservedacrossdifferentfamilies[43,44].Thus, increasingthenumberofphylogeneticcharactersfromproteincodingnucleargenesisofman- datoryimportanceforachievingrobustphylogenetichypothesesinbeetlesystematics. Recently,theadventofnextgenerationsequencing(NGS)technologieshascontributedto additionalground-breakingadvancementsinthesystematicsfield,profoundlyincreasingthe levelofresolutioncomparedtopreviousphylogeniesbasedonsingleorfewgenes[45].Geno- micandtranscriptomicdataobtainedfromNGSbasedresearchhasledtopredictiveinsect phylogenies,whichnowmoreclearlyrevealkeyeventsininsectevolutionaryhistory[46–50]. Newdevelopmentsbasedonultra-conservedelements(UCEs)orRAD-sequencingwill increaseresolutionalsoatlowertaxonomicranksininsects[51,52].However,thebenefitsof NGSaregenerallycounterbalancedbythehighcostandcomputationallydemandinganalyses ofsuchhighthroughputdata.Theutilityoffewwell-characterizedmarkersshouldnotbe underestimatedastheyrepresentarapidandcosteffectiveapproachforresolvingsmallscale phylogenies. BarkandambrosiabeetlesinthesubfamilyScolytinaeconstituteagroupofhighlyderived, smallwoodboringweevilscapableofexcavatinggalleriesintodifferentpartsofdeadtrees, shrubsandbushes,aswellasinlianasandotherplanttissuesindifferentforesthabitats throughouttheworld[53].Scolytinaeisgenerallyregardedasawell-supportedcladeofmore than6000describedspeciesrepresentingapproximately12percentoftheentirediversityin thefamilyCurculionidae[5,54,55].Atremendousvariabilityinlifecycles,reproductivestrate- gies,matingsystems,hostplantsinteractions,feedingbehaviorandecologyhasbeendocu- mented[56,57],whichmakesthisgroupofbeetlesparticularlyinterestingtostudyina phylogeneticallycomparativecontext.PhylogeniesofScolytinaehavesofarreliedonacombi- nationoffivemolecularmarkers(onemitochondrialandfournucleargenes)andeventually morphologicalcharacters.GiventhehighdiversityofScolytinespecies,additionaldataare neededtoobtainsufficientresolutionatdeepernodes. Inordertoselectnewphylogeneticmarkers,100differentnucleargeneswerescreenedby PCRusingdegenerateprimersandtestedinarestrictedbutrepresentativegroupofScolytinae andotherweevils.Withtheaimofdevelopingslowlyevolvinggenes,thepropertiesofeachgene PLOSONE|DOI:10.1371/journal.pone.0163529 September26,2016 2/26 GenomicMininginBarkandAmbrosiaBeetles fragmentwereevaluatedbasedonPCRamplificationandsequencingsuccessandtheirphyloge- neticperformance.Thisstudyreportsonthedevelopmentandutilityof16novelmarkersfor weevils,withaparticularfocusonbarkandambrosiabeetlesinthesubfamilyScolytinae. MaterialsandMethods We included18speciesofbarkandambrosiabeetlesand8additionalweevilsfromothersub- familiesforprimerscreening(Table 1andS1Table).Thesebeetleswerecollectedbyoneofthe authors(BHJ)duringfieldworkintropicalforests(1998–2012).Collectionpermitswere requestedfromauthoritiesinUganda,Tanzania,Cameroun,SouthAfricaandMadagascar. Ethicalguidelineswerefollowed.VoucherspecimensaredepositedintheColeopteracollection oftheUniversityMuseumofBergen,UniversityofBergen,Norway.Allweevils,Platypodinae andScolytinaespeciesusedinthisstudywerepreviouslydescribedinotherphylogeneticstud- ies[7,11,58]. Theprocedureforprimerselectioncanbesummarizedasfollows:1)putativelysinglecopy expressedsequencetags(ESTs) longerthan800basepairswereselectedinGenBankfortwo differentbeetlespecies,Tribolium castaneumandDendroctonusponderosae;2)preliminary Table1. Weevilspeciesincludedinthisstudy. Species Code Subfamily Tribe Country Brentidaesp. BrBre05 Brentidae(familiy) Brentinae Cameroon Mesitesfusiformis CsMes01 Cossoninae Cossonini Spain Pselactussp. CsPse01 Cossoninae Onycholipini Portugal(Madeira) Larinussp. ClLar01 Lixinae Cleonini Russia Portheteshispidus MoPor01 Molytinae Amorphocerini South-Africa Platypusimpressus PlPla07 Platypodinae Platypodini Tanzania Triozastusmarshalli PlTri02 Platypodinae Platypodini Cameroon Chaetastustuberculatus TsCha02 Platypodinae Tesserocerini Cameroon Pityophthorusmicrographus CoPit01 Scolytinae Corthylini Norway Diamerusinermis/D.hispidus DiDia03/DiDia04 Scolytinae Diamerini Tanzania/Madagascar Dryocoetesautographus DrDry01 Scolytinae Dryocoetini Russia Ozopemonuniseriatus DrOzo02 Scolytinae Dryocoetini PapuaNewGuinea Hylastesattenuatus HtHyt06 Scolytinae Hylastini Sweden Hylesinusvarius HlHyl02 Scolytinae Hylesinini Sweden Kissophagushederae HlKis01 Scolytinae Hylesinini Austria Chaetopteliusvestitus ToCha01 Scolytinae Hylurgini Morocco Dendroctonusterebrans/D.micans ToDen02/ToDen01 Scolytinae Hylurgini USA Tomicuspiniperda ToTom01 Scolytinae Hylurgini Norway Acanthotomicussp. IpAca01 Scolytinae Ipini Cameroon Pityogenesquadridens IpPit03 Scolytinae Ipini Sweden Premnobiuscavipennis PrPre01 Scolytinae Premnobiini SierraLeone Camptocerusaenipennis ScCam02 Scolytinae Scolytini Guyana Cnemonyxvismiacolens ScCne01 Scolytinae Scolytini Guyana Scolytusintricatus ScScl02 Scolytinae Scolytini CzechRepublic Xyleborusaffinis XyXyl00 Scolytinae Xyleborini Cameroon Xyleborusmonographus XyXyl03 Scolytinae Xyleborini CzechRepublic Degenerateprimersweredesignedonconservedregionsinthealignmentofinsectnucleotidesequencesthatwereavailablefromgenomicand transcriptomicsources.Twoormoreconsecutivedegeneratesiteswerepreferentiallyavoidedaswellastheuseofcompletelydegeneratesites(N).Atotal of274primersweredesigned(Table2-onlysuccessfulprimersreported). doi:10.1371/journal.pone.0163529.t001 PLOSONE|DOI:10.1371/journal.pone.0163529 September26,2016 3/26 GenomicMininginBarkandAmbrosiaBeetles BLASTsearcheswereperformedtodiscardunsuitablemarkers,basedontheevidenceformul- tipleparalogouscopies(e.g.largegenefamilies)orambiguousgenomiccharacterization(e.g. similarmatchingvaluesfordifferentproteins);3)availablesequencesforeachselectedgene werealigned,includingannotatedgenomicandtranscriptomicsequencesfrommodelorgan- isms(e.g.Drosophilamelanogaster,ApismelliferaandBombyxmori)todetermineintron-exon structure;4)degenerateprimersweredesigned;5)aPCRscreeningwasrunandproductswith theexpectedcorrectsize(albeithighlyvariableduetopresenceofintrons)weresequenced;6) markersreachingaminimumPCRandsequencingsuccessof20%wereusedtoreconstruct singlegenephylogenies(Bayesian)andtreeswerecomparedtopreviouslyestablishedand well-supportedclades[5,7,10,11]. DNAwasextractedfromindividualspecimensusingDNeasyBlood&Tissuekit(Qiagen) followingthemanufacturer’s instructions.ThePCRreactionmixturecontained2.5μl10xPCR buffer(Qiagen),inwhichthefinalconcentrationofMgCl was2.0mM,200μMofeachdNTP 2 (SigmaAldrich),0.5μMofeachprimer,0.125unitsHotStartTaq1DNApolymerase(Qia- gen),2μlDNA,withwateraddedtoafinalvolumeof25μl.Anegativecontrol(sterilewater) wasincludedineachtest.ThePCRwasperformedusingaS1000TMThermalCycler(BIO-RAD Laboratories,Inc.).Threestandardcycleprogramswereusedfortheinitialscreening:denatur- ationstepat95°Cfor5minutes,35cyclesof30secondsat95°C,30secondsat48,52and58°C, 60secondsat72°C,andfinally5minutesextensionat72°C.Furtheroptimizationincludeda gradientofannealingtemperaturesintherangeof44–62°C,modulatingtheextensiontime dependingontheexpectedPCRproductlength,andMgCl concentration.We alsoconsidered 2 twodifferenttouch-downPCRprotocolsfortwoofthesegenes(seeTable 2fordetails). PCRproductsweresequencedwiththesameprimersasthoseusedforamplification.DNA sequencesofbothstrandswereobtainedusingtheBigDyeTerminator cyclesequencingready reactionkit(AppliedBiosystemsInc.)usinganautomatedDNAsequencer(AppliedBiosys- temsPrism3700)followingthemanufacturer’s instructions. AllobtainedsequencesweresubmittedtoBLASTanalyses,acceptingacorrectgenetargetif thecutoffvaluewasbelow1E-4.Allsequencesforeachgenewerealignedwithotherinsect sequencesforapreliminaryNJanalysisinPAUP(cid:3)4.0[59]todetectdeviantsequences.The sequenceswerecheckedbyeyeandusingBioedit7.2.5[60]andMAFFT[61]toaligngenefrag- mentswithcomplexstructure,causedeitherbytothepresenceofindelsofcodingtriplets,or lessfrequentlybylongintronsmarkedbyunusualexon-intronborderssuchasthemostcom- monalternativesplicesiteGC—AG[62]. Intronsweretrimmedandthecodingfragmentsweretranslatedintoaminoacidsequences usingBioedit7.2.5tocheckfortranslationalerrors(stopcodons).Allthesepreliminaryanaly- seshadthepurposeofdetectingpseudogenesorearlysignsofpossibleparalogs(e.g.high degreeofaminoacidsubstitutions).Inaddition,theaminoacidsequencesoftheselected markerswereexaminedinOrthoDBv9toassessgeneorthology[63,64].Theorthologyfor eachgenewasconfirmedbyclusteroforthologousgroups(COGs)comparisonamongarthro- podsequencesinthedatabase.Ambiguousnucleotidepositionsinthecodingregionthatwere difficulttoalignweretentativelyexcluded(inArr2andIap2)tocreateanalternativealignment forcomparisons(seeresultsanddiscussion). Phylogeneticanalyseswereperformedonunambiguouslyalignedsequencesobtainedfroma minimumof5species.PhylogeneticinferencewasbasedonBayesianandmaximumparsimony analyses,thelatterasimplementedinPAUP(cid:3)4.0.Nodesupportintheparsimonyanalyseswas estimatedbybootstrapanalysesusing20randomadditionsofheuristicsearchesforeachof200 bootstrapreplicates.BayesianphylogeneticanalyseswereperformedinMrBayes3.2[65].The mostappropriatemodelforbasefrequenciesandsubstitutionrateswasdeterminedbyjModelT- est[66],usingtheAkaikeinformationcriterion(AIC).MrBayessearcheswererunforeachgene PLOSONE|DOI:10.1371/journal.pone.0163529 September26,2016 4/26 GenomicMininginBarkandAmbrosiaBeetles Table2. Primersequencesandannealingtemperatureforthenuclearmarkersselectedinthisstudy. Furthermore,primersforadditionalgenesfor lowerlevelphylogeneticsarereported. Geneacronym Primerforward(5’-3’) Primerreverse(5’-3’) AnnealingT˚C EF2 CGTTTCTAYGCBTTYGGHCGTG CCYTCYTTRGTGGCCCAYTGG TD58(10cy)44(25cy) ATGATGGGYCGTTAYGTWGARGC TD58(10cy)44(25cy) Hsp70 CAAGCYGACATGAAGCAYTGGCC CGGGTGATGGAGGTGTAGAARTC 58 GAYGGTATCTTYGARGTMAAGTC CGRCCYTTGTCRTTRGTGATGG 55 CCNC ATGGCTGGMAAYTTTTGGCARAG TCGAGCAGATARAAYTCRCAYTC 52 HDACRpd3 ATGAARCCSCACMGSATAMGSATGAC GTAGTCGTTRTARGGSAGYTCRTTGGC 53 GCCACSGAAGTYTCRTASGTCCA 53/50 Arr2 CGYGARGAGGAYGARGTYATGGG ACCATSGTRACYTCGCAATGYTGCAC 52 CTCAAARACKATRTTGTCGTCRTCGTC 52 Iap2 TGGAAYTAYGGRGACCAAGTRATGGC CCATCKGGCRTGYTCYGTCCAWGGATC 52 PABP1 CCRATTCGYATYATGTGGTC GAARGCRACAAAWCCRAAWCC 50 Prp1 ATGTCSGCKACTYTRGAYGCWGG GGRTASGTGTTRTCYTGCATYTC 44 CTR9 GAAGGYGATAARATGGAWCARGC TCGAAACAYTGKGCKGCATTTTC 52 RCC1 GGKTGYAATGACGARGGSGC CGGCCCAATTGTCCYTGYTC 52 SOD1 TCCACATYCAYGARTTYGGGG CCTTKKCCCAAATCATCMGG TD52(10cy)46(25cy) TPI CGHAAATTCGTWGTYGGWGGHAACTGG CKGARCCYCCRTATTGRATTC 50 GGTGGHAACTGGAARATGAACGG 52 ADA2 GAYATGYTDGAYGTVCATGC ACAGGRCCRGCTTCRCCRCAATG 52 AARTTYAATGCCAAATAYAAYCC GGWCCRGCTTCACCRCARTGWGG 48/52 UBA5 TTGGKAGYGTAACWGCRGAAATG ATATGGCCWGARACSGCRTTTTC 52 Cda4 TACGARGARTGGGTKGGRGARATG AACCAATTMGTRTGRAASGGCATC 48 FEN1 GARGCCCCYTGYGARGCKGARGC TCACCATGCCYTCYTCRTCMGG 48 ACTB CTGAAGCCCCMTTGAACCCMAAGGC GAGATCCACATCTGYTGGAARGTGG CXorf56 GAAGYATTGCRTGTTCSGAYAC GTCACMGAACTGAAYTTKCCC eRF1 GTTGGCAGATGAATTTGGAACRGC CCRAABAGAGCTCCRTTACCATCC U2AF ATYGCTGGATTWAAYGGRATGC TCTCKTCTRTGRTACTTRTCSGGWTC MAD YAAYTTYCCWGCYATGRTWCC ACACCRTGRTTYTTWGCWCC mp20 GACAAGGARGCCCARGARTGGATCG TCCCACAGRTCAACTGTYTGGAARAC GGTCCGGGCCCAYTCRGRGTGCYTGTTAGG 5MP CATGACKTTTATGMGKGCKTTC CTTCYTCRGCGTTTTGWAGCC Pi4k TGYTGYCCKTGYTGYTTYGG TGGTAYGGRTASGCYCGCC Gel GAYGAGGGCSGGWTCSGCWGC AGGATRAAGCARTCRCCTTTGTTC C1-THF CATYTRACYGGYGAYATYCATGC ACAGCYCCYGTKGCYCCCAAATC alpha-Spec CAYGCHAATGCWTTCCATCARTGG GGYTGKCCYTCYTCWACCATYGG AATS CATCAYACGTTTTTTGAGATG GCATGRTCNGCTAARACNCGRTARGCC Hsp90 GATCATCAATATSTTCTACTC TCTCCGGTGATGWARTAGATG dldE3 GGRGAYTGTATWCATGGRCC GCYTCRTTRATBARTTCRCC CATCCWGAAGTKGGMTGGGTKGG Mpgt AAACCSCTGTTYCCMGTTGCKGG GCMGTTTTYAACTGSGACCACC NaK GGYGGTTTCGCSWTGYTGYGTGGATCGG GCGACGATGATACCGATCARGAAGATGACAGC Fbox11 AATGCWTTRGCTGGWATYTGGG CCRCCRTGYTGACCRTGRTG UDE AAGCCRGACACCGTWCCCGG CTGGCWTCRGGRCTGTACGCCC GTPbp ATTARAAYGTAKCCATCGTTRCCCC GTGTTGATAATWGASGACTTGCC CatL CACATTTACACTTTYAACCCRATG ACCARCTGTTYTTMACCARCCAGTA TpC CTTCCCSCMGARCARATYGCCG CCTCSCCRGTCATCATCTCCATG PGI GGCCCSCTKATGGTRACCGAAGC CCCAGCTCCACKCCCCATTGGTC (Continued) PLOSONE|DOI:10.1371/journal.pone.0163529 September26,2016 5/26 GenomicMininginBarkandAmbrosiaBeetles Table2. (Continued) Geneacronym Primerforward(5’-3’) Primerreverse(5’-3’) AnnealingT˚C AcCoA GGTGTACTGCKGAYATTGGYTGGATCAC GGAAACSCAGCMGCKCCWGGYTTCAT CATCAGRTGYCCKGASACGTTYARCAT Ucdk GAGCACKGTWTGCAARCGYATWATGG CCYCTWGGAATRATRACATCAGC PPO1 AAYCTSCACCAYTGGCAYTGGC CGGAASGTSCKCTCRAASGG Prp6 AATCCSAATCATCCWCCGGCKTGG TTCTTCCAGYTTRGCSGCRGTWGTCC Mxp TAMGSACRGCSTAYACSAACAC CGCTTGTGYTTCATSCKCCG Npl4 CTCGYTGYGTSCAYTGCTC TCGCGCACYAGCGCCATRCAYTG Cam1 GAYGGMGATGGCACRATYACTACC TCRTAATTGACCTGACCGTCRCC STX1A ATGACYAARGAYAGATTRGCRGC GCCATRTCCATRAACATRTCRTG TP120b TWGGRAATGTCAAYGTYTC AAGCTCAACCCKCKCCACATCC CHS1 CATATMTTYTTCGAYGAYGC CAACGATCYTCKCCYTGATC DDX49 AARGCTATACGARGAYCCWTATGG TGCCTGCYCTAGCWGTYCTYCC GTF2H3 CTCGCATTTGATGCAGAAGGC CARATYGGRCTAAACTTGCA IF3 ACTCGCTYTACAAAATGTTGGG CTTTSGTRTCGGCRATATGRATC TIF6 GACACRATWCCSGTGGTSCATGC CTACCWCARTTWACYGTTCC IDH TACAAYGTWGGAATWAARTGTGC CAMACAAARCCYCCYTCMGATTTC Ecr GAAGTKATGATGTTCMGRATGGC GAWGCACATYTCDGARTTYTG doi:10.1371/journal.pone.0163529.t002 separatelyandforconcatenateddatasets(8109bp–2702aa)usingthesuggestedmodelsfor eachgenepartitionandamixedmodelforaminoacidsubstitution.Inbothcases,thesearch consistedof2000000generationswithtwoindependentruns,eachwithfoursimultaneous chains,andtreessampledevery1000generations.Theconvergencediagnostics(SDSF, PSRF) andparametersampleplotswereevaluatedusingthesoftwareTracer1.6[67]. Anindirectmeasureofthephylogeneticsignalineachmarkerwasassessedthroughtopo- logicalcongruencewithpreviouslywelldocumentedclades[5–7,10,11,68]whichwereused toderiveaschemeofthecurrentclassificationofCurculionoidea(Fig1).Thesecladesbelong tosixtribesofScolytinae(A=DryocoetiniincludingXyleborini,B=Ipini,C=Hylurgini +Hylesinini,D=Scolytini)andthesubfamilyPlatypodinae(E).Rootingofthetreeswas Fig1.SchematictreeshowingwellsupportedrelationshipsbetweentribeswithinthesubfamilyScolytinaeandother weevilfamiliesandsubfamiliesconsideredinthisstudy. doi:10.1371/journal.pone.0163529.g001 PLOSONE|DOI:10.1371/journal.pone.0163529 September26,2016 6/26 GenomicMininginBarkandAmbrosiaBeetles dependentonthesequencesavailable,andusedinthefollowingorder:1)Brentidae,2)Platy- podinae,3)Cossoninae,MolytinaeandLixinae,4)Scolytini[5,6]. Basicpropertiesofeachgene,includingtheoverallmeandivergenceofsequences(p-dis- tance)andthevariationinfirst,secondandthirdpositions,werecalculatedforeachgenefrag- mentusingMEGA6.0[69].Parsimonyinformativesiteswerecalculatedtogetherwiththe homoplasyandretentionindices(respectivelyHIandRI–S2Table) usingPAUP(cid:3)4.0.Aphylo- geneticinformativenessprofile(PI)wasobtainedforeachgeneusingPhyDesign[70],anon- lineprogramdevelopedfromapreviousstudy[71].Substitutionratesforeachpositionwere calculatedusingHyPhyimplementedinPhyDesign,selectingaK2Pmodel(basefrequen- cies=0.25,transitions=2,transversions=1).TheinputtimetreewasobtainedusingBeast v1.8.2[72],withtopologyconstraintsfollowingpreviouslypublishedphylogeniesofweevils andScolytinae[5,6,11].Thetreewasreconstructedusingaconcatenateddatasetof16genes, usingaGTR+I+Γmodelforeachgenepartition,andaYulespeciationprocess.We selectedan uncorrelatedlognormalrelaxedmolecularclockanduseddefaultpriorsassuggestedbythe authors(seeXMLS1fileinSupplementaryinformation).Twocalibrationpointswereused: 116MaforthenodesubtendingScolytinaeandotherweevilsubfamilies,and30Maforclade A(Dryocoetini+Xyleborini). Results Sequenceswereobtainedfor57differentgenes,whereas43primersetsneveramplifiedthecor- rectgene.Atotalof798sequenceswereobtained,butonly510ofthese(64%)wereunambigu- ouslycharacterizedasbeetleorthologsinBLASTNsearch.Amongtheremaining288 sequences,53wereidentifiedasnon-beetlesequences(mainlyfrombacteria,fungiornema- todesassociatedwithbeetles)withdifferentdegreeofconfidenceingeneidentity.Theremain- ing235sequencesresultedinunreadableorpoorqualitysequenceswithoutaclearmatchin GenBank(Evalue>1E-4,querycoverage<30%and/orlessthan30%identity). Theevaluationofthe57markerswithreadablesequenceswasbasedonthenumberof sequencesobtainedandtheirphylogeneticperformance.Whenonlyoneortwosequences wereobtainedforagene(e.g.cathepsinL,troponinC,acetylcoenzimaAsynthetase,maxillope- dia,calmodulin1),thephylogeneticutilitywasnotpossibletoassess.Otherexcludedmarkers producedahighernumberofsequences,suchasodorantbindingprotein(8sequences)andgly- cosidehydrolasefamily31(11),butthesewerelargelyunalignable.Anothergroupoffailed markersproducedsequencesfromnon-targetorganisms,suchas6-phosphogluconatedehydro- genaseoffungi,orphosphoglucoseisomeraseofbacteria.Atotalof23geneswerediscardeddue tolowamplificationrates,highlevelsofnon-beetleamplification,orgenerallylowdegreeof geneorthology. Theremaining34genesshoweddifferingdegreeofPCRandsequencingsuccess(from5to 26sequencesobtained),andwerefurtherevaluatedbasedontheircapacitytorecoverknown relationshipsatvarioustaxonomiclevels.Eighteenofthesemarkerswerefoundinsufficiently informativeforhigherlevelphylogenetics,becausenomorethantwoofthepredefinedclades werereconstructedcorrectly.However,mostdiscardedmarkersneverthelessrevealedsome phylogeneticutilityatlowertaxonomiclevel;includingpopulations(seeS3Table forfurther details). We selected16genesthatrevealedarelativelyhighandstablePCRandsequencingsuccess (from50to100%)asthebestcandidatesforScolytinaephylogenetics(Table 3).Alltheverified sequencesobtainedinthisstudyweredepositedinGenBankdatabaseundertheaccession numbersKX160539—KX160803(S1Table).ThespeciesXyleborusaffiniswasthemostsuc- cessfulinPCRandsequencing(15outof16possiblesequencesobtained);theothersamples PLOSONE|DOI:10.1371/journal.pone.0163529 September26,2016 7/26 GenomicMininginBarkandAmbrosiaBeetles Table3. PCRandsequencingsuccessfor16selectedgenes. GENEACRONYM A B C D E F G H Total(%) PABP1 4 3 6 3 3 4 2 1 26(100%) TPI 4 2 6 - 2 2 2 - 18(69%) UBA5 3 3 5 3 2 2 1 1 20(77%) Iap2 3 3 1 2 1 4 2 - 16(62%) SOD1 2 1 4 3 2 3 1 - 16(62%) Prp1 3 3 5 1 3 1 2 - 18(69%) ADA2 3 2 2 2 3 - 2 - 14(54%) CTR9 2 2 4 2 - 1 2 - 13(50%) CCNC 4 2 5 2 2 2 2 1 20(77%) Cda4 2 1 4 - 3 1 1 1 13(50%) HDACRpd3 3 1 4 - 2 2 1 - 13(50%) Arr2 4 2 4 3 3 2 2 - 20(77%) FEN1 3 2 4 2 1 - 2 1 15(58%) EF2 2 2 3 2 3 - 2 - 14(54%) Hsp70 1 1 5 2 1 1 2 1 14(54%) RCC1 2 2 4 - 2 2 1 - 13(50%) Thenumberofsequencesobtainedwasreportedforthefollowinggroups:A=Xyleborini+Dryocoetini,B=Ipini,C=Hylurgini+Hylesinini,D=Scolytini, E=Platypodinae,F=otherCurculionidaesubfamilies,G=otherScolytinae,H=Brentidae. doi:10.1371/journal.pone.0163529.t003 variedconsiderablyinthisrespectwithonly4sequencesobtainedforLarinussp.(S1Table). Thetotalfragmentlength,thepresenceoflength-variableregions,andthenumberandposi- tionofintrons,weremappedontheannotatedgenomesofT.castaneumandD.ponderosae (eventuallytranscriptomicandgenomicdataofotherinsectspecies)tocreateamapofthe genestructure(Fig2;seealsoTable 4). OrthoDBanalysesshowedthat12outof16genesselectedinthisstudyarepresentinsingle copyinmorethan70%ofthearthropodspeciescurrentlyinthedatabase(133).PABP1and UBA5areinsinglecopyin96%ofthesespecies,followedbyHDACRpd3(95%),CCNC(94%), Prp1(92%),TPI,CTR9andFEN1(90%),Cda4(89%),EF2(84%),RCC1(81%)andADA2 (74%).Onlyfivegenesarefrequentlyinmulti-copystatusinarthropodgenomes:Hsp70(single copyonlyin2%ofthespeciesinthedatabase),Arr2(4.5%),Iap2(8.3%)andSOD1(22%). ThebestevolutionarymodelforthemajorityofthegeneswasGTR+I+Γ,exceptforSOD1 andIap2inwhichSYM+I+ΓandGTR+Γwereselected.Bayesiananalysisoftheconcatenated nucleotideandaminoaciddatafrom16genesshowedawellresolvedtreetopology(S1Fig) withallexpectedcladesrecoveredwithmaximumsupport,exceptScolytini(pp=0.75).The overalltreetopologywascorrectwiththeexceptionoffourweevilspeciesthatwerenested insideScolytinaeasthesisterlineagetoHylurgini(weaklysupportedintheaminoacidanaly- sis).Parsimonyanalysesoftheconcatenateddatasetrevealedsimilarresultsbothforthenucle- otideandaminoaciddatasets,withallmajorcladesrecoveredwithmediumtohighbootstrap support.However,thesub-familyScolytinaewasnotmonophyleticinrespecttotheother advancedweevilspecies(S2Fig). Singlegeneanalysesresultedinpartiallyresolvedphylogenies,mainlyrecoveringamono- phyleticScolytinae,themajorityofthepredefinedsubgroupsofScolytinae(A-B-C-D),andthe subfamilyPlatypodinae(Fig3).Allselectedgenesenabledthecorrectreconstructionofthe mostrecentclade(A),with3genesobtainingthecorrectsistergroup(B).Noneoftheselected genesshowedhighdegreeofincongruencethatreceivedhighnodesupport.Overallmean PLOSONE|DOI:10.1371/journal.pone.0163529 September26,2016 8/26 GenomicMininginBarkandAmbrosiaBeetles Fig2.StructureofthePCRamplifiedgenefragments.Thegraphicsillustrateintron-exonpatternsin16markers withcodingregionsshownasblackbarsandintronsasthinblacklines.Lengthvariablecodingregions(indels)were coloredinlightgrey(Iap2andArr2). doi:10.1371/journal.pone.0163529.g002 PLOSONE|DOI:10.1371/journal.pone.0163529 September26,2016 9/26 GenomicMininginBarkandAmbrosiaBeetles Table4. Geneinformation. Acronym nucs aa Intron Intronrange(perintron) PABP1 435 145 0 - TPI 547 182 0–2 (457–51)(237–48) UBA5 348 116 1 (94–48) Iap2 672* 224* 1 (1131–50) SOD1 213 71 0 - Prp1 582 194 0–1 (258–55) ADA2 624 208 2 (70–39)(105–53) CTR9 627 209 0–1 (81–59) CCNC 384 128 3 (200–69)(134–49)(71–58) Cda4 410 136 0–3 (68–51)(63–56)(53) HDACRpd3 858 286 3–5 (69–53)(70–54)(165–48)(564–54)(66–55) Arr2 501* 167* 0–3 (110–51)(84–53)(158–55) FEN1 417 139 1–3 (63–46)(55–42)(93–44) EF2 621 207 1–2 (398–183)(702–84) Hsp70 567 189 0–2 (61-?)(317–187) RCC1 303 101 0–1 (250–51) Foreachmarker,thelengthofthesequencedcodingregionisgivenasthenumberofnucleotidesand aminoacids,togetherwiththenumberandlengthofintron(s).Thesymbol*indicatesgeneswithsequence lengthvariabilityduetoexonicindels. doi:10.1371/journal.pone.0163529.t004 divergenceinnucleotidesequenceswasreportedforeachcodonpositionforeachgene(S3 Fig). SelectedgenesforScolytinaephylogeny Polyadenylatebindingprotein1(PABP1). PABP1wasthemostsuccessfulmarker,with sequencesobtainedfromall26species.Theamplifiedfragmentwas435bplong,containedno introns,andtranslatedinto145aminoacids.Thephylogeneticanalysesrecoveredalmostall pre-definedclades(Fig3a),butonlytwoofthemwerehighlysupported(B,pp=0.98; E,pp=1).ThetribeScolytiniwasplacedoutsideapolytomyincludingtheremainingspecies ofScolytinae,thesubfamilyPlatypodinaeandthevariousotherweevilsubfamilies.Noclear evidenceofparalogsemergedfromtheanalyses.Preliminarystudiesindicatedincreasedphylo- geneticperformancewithbroadertaxoncoverage. Triose-phosphateisomerase(TPI). Acombinationoftwoprimerpairs(twoforward,one reverse)resultedin67%PCRamplificationandsequencingsuccess.Thealignedfragments consistedof547bpafterremovalofintrons,whichtranslatedinto182aminoacids.Two intronswerelocatedinthisgenefragment(Fig2,Table 4).Thephylogenybasedonthismarker confirmedthemonophylyofPlatypodinae(pp=1),whileScolytinaeformedalargepolytomy includingtwoadvancedweevilspecies.FurthermorewasCossoninaemonophyletic(pp=1), inadditiontoonescolytinesubgroup(A,pp=1),andsubgroupCalmostso(Fig3b). Ubiquitin-likemodifieractivatingenzyme5(UBA5). TheUBA5genefragmentis348 bplongandtranslatedinto116aminoacids.Itwasamplifiedfrom20differentspecies(77%) inallmaincladesandcontainedoneshortintroninallspecies.Thephylogenyrecoveredthe monophylyofcladesAandEwithhighnodesupport(pp=0.99and1,respectively)while cladeD(pp=1)hadScolytusintricatusexcluded.CladeBandCwereweaklysupported (pp<0.95)andKissophagushederaewasnotincludedinHylurgini(Fig3c). PLOSONE|DOI:10.1371/journal.pone.0163529 September26,2016 10/26

Description:
PLOS ONE | DOI:10.1371/journal.pone.0163529 September 26, 2016 . This study reports on the development and utility of 16 novel markers for . logical congruence with previously well documented clades [5–7, 10, 11, 68] which
See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.