scieee AI-readable full text Open interactive document viewer

Quaternary geomorphological and climatic changes associated with the diversification of Iberian freshwater fishes: The case of the genus Cobitis (Cypriniformes, Cobitidae)

Corral-Lou, Andrea,Perea, Silvia,Perdices, Anabel,Doadrio, Ignacio

Abstract

We thank P. Garzón, I. Doadrio Jr, J.L. González, and G. González for collecting almost every sample from 2009 to 2010 under the project “Basis for the monitoring of Spanish Freshwater Fishes”, supported by the Ministerio de Medio Ambiente y Medio Rural y Marino (MARM), and C. Marcos, J.C Velaco, and G. González for their help in the project Life13nat/es/000772. This study was also supported by a management commission funded by the Duero Hydrographic Confederation in Spain (Análisis y estudios filogeográficos y de variabilidad genética de endemismos ibéricos de ciprínidos: Life13 nat/es/000772 “Actuaciones para la protección y conservación de ciprínidos ibéricos de interés co munitario”) and by grant IND2017/AMB-7699 to AC- L, funded by the Community of Madrid (Spain) through its Industrial Doctorate program.

Full text

Ecology and Evolution. 2022;12:e8635.   | 1 of 22 https://doi.org/10.1002/ece3.8635 www.ecolevol.org Received:22December2021 | Revised:20January2022 | Accepted:25January2022 DOI: 10.1002/ece3.8635 RESEARCH ARTICLE Quaternary geomorphological and climatic changes associated with the diversification of Iberian freshwater fishes: The case of the genus Cobitis (Cypriniformes, Cobitidae) Andrea CorralLou1,2 | Silvia Perea1,3 | Anabel Perdices1 | Ignacio Doadrio1 ThisisanopenaccessarticleunderthetermsoftheCreativeCommonsAttributionLicense,whichpermitsuse,distributionandreproductioninanymedium, providedtheoriginalworkisproperlycited. ©2022TheAuthors.Ecology and EvolutionpublishedbyJohnWiley&SonsLtd. 1BiodiversityandEvolutionaryBiology Department,MuseoNacionaldeCiencias Naturales,CSIC,Madrid,Spain 2ConsultoresenBiologíadela ConservaciónS.L.,Madrid,Spain 3InstitutodeBiología,Departamentode Zoología,UniversidadNacionalAutónoma deMéxico,CiudaddeMéxico,México Correspondence AndreaCorral-Lou,Biodiversityand EvolutionaryBiologyDepartment,Museo NacionaldeCienciasNaturales,CSIC, JoséGutiérrezAbascal,2;28006Madrid, Spain. Email:[email protected] Funding information IndustrialDoctorateofCommunityof Madrid,Grant/AwardNumber:IND2017/ AMB-7699;DueroHydrographic Confederation,Grant/AwardNumber: Life13nat/es/000772 Abstract WestudiedthepopulationgeneticstructureofCobitis vettonica,anendangeredfreshwaterfishspeciesendemictotheIberianPeninsula,inordertoproposeabiogeographicmodeloftheresponsesofspeciestothemultiplechangesthatoccurredin theIberianhydrologicalsystemduringtheQuaternaryperiod.Wealsodeciphered therelationshipofC. vettonica with its sister species C. paludica,particularlyinsympatricareas,andprovidegeneticinformationforconservationpurposes.Toachieve thisgoal,weanalyzedbothmitochondrialandnucleardata(thecytochrome band the nuclear recombination activating 1genes)andabatteryofsingle-nucleotidepolymorphisms(SNPs)of248individualsofC. vettonica or C. paludicafrom38localities, includingsomesympatricones,coveringtheentiredistributionareaofC. vettonica. Wehighlighttheimportantroleplayedbythehydrogeomorphologicalprocessesand climaticchangesthatoccurredintheIberianPeninsuladuringtheQuaternaryonboth thepopulationstructureofC. vettonicaanditsrelationshipwithitssisterspeciesC. paludica.Ourresultssupportthegeneticintrogressionofpopulationsattheeastern limitofthedistributionofC. vettonica.Furthermore,wepostulategeneticintrogressioninsympatricareas.Finally,weproposetheestablishmentorexpansionoffour OperationalConservationUnits(OCUs)forC. vettonica,andhighlightthethreatfaced byitspopulationsduetothelowlevelofgeneticdiversitydetectedforsomeofits populationsandgeneticintrogressionwithC. paludica,whichcouldeventuallydisplaceC. vettonica,resultinginalossofdiversityinthisspecies. KEYWORDS Cobitis,hybridization,IberianPeninsula,populationstructureandconservation,quaternary TAXONOMY CLASSIFICATION Populationgenetics 2 of 22 | CORRALLOU et AL. 1 | INTRODUCTION The Quaternary is a geological period characterized by glacial– interglacial cycles that have dominated the global climate since 2.58Myatothepresentera(Gibbardetal.,2010;Pillans&Naish, 2004),whichhavehaddrasticconsequencesontheevolutionofthe biotaofmanyregions.Duetoadvancesinthefieldofphylogeography,wehaveabetterunderstandingoftheresponsesoforganismstoPleistoceneevents(Hewitt,2004;Weiss&Ferrand,2007). ClassicalstudiessupporttheroleofMediterraneanpeninsulasas refugesforfaunaduringtheQuaternary,whichprovidedthestock forrecolonizationsofnorthernandcentralEurope(Hewitt,1996; Taberlet&Bouvet,1994).TheIberianPeninsulaisconsideredoneof themostimportantrefugesduringthisperiod,asshownbyseveral studies,mainlyofterrestrialfauna(Querejetaetal.,2017;Valdiosera etal.,2008). Primaryfreshwaterfishes(i.e.,fishesstrictlyintolerantofsaltwater; Myers, 1966) have limited dispersal abilities and are often restrictedtospecifichydrographicbasins.Dispersalbetweenbasins during the Quaternary was possible mainly through downstream connectionscausedbythedecreaseinsealevelduringcoldperiods orbyupstreampiracy(Corral-Louetal.,2019;Mesquitaetal.,2005; Perea&Doadrio,2015).TheformationoftheIberianfluvialnetwork alsoculminatedduringthisperiod,whichaffectedtheregion'shydrogeomorphologyandthereforethecurrentevolutionarypatterns foundinprimaryfreshwaterfauna(Alonsoetal.,2007;Corral-Lou etal.,2021;Paisetal.,2012;Pereaetal.,2016).Likewise,thecurrent populationstructureofmanyIberianfreshwaterfisheshasbeenattributedtotheinteractionofvariousnaturalfactorsthatoccurred duringtheQuaternarysuchasthedryingupofbodiesofwater,sea- levelfluctuations,tsunamis,streampiracy,isolationofbasins,hydrogeomorphologicalchanges,andsecondarycontactoftwodifferent basins(Casal-Lópezetal.,2017;Corral-Louetal.,2019;Ganteetal., 2009;Pereaetal.,2016). Despiteeffortsmadeinthelastseveraldecades,thereisstill muchtobelearnedabouttheimpactofQuaternarychangesonthe evolutionary processes and patterns of diversification of Iberian freshwater fish populations. Especially those species with a restricteddistributionrangesincetheywereprobablymoreaffected byclimaticandgeologicalchangesduringtheQuaternarythanspecieswithlargerranges.Inordertoaddresssomeopenevolutionary questions, such as the role of hydrogeomorphologic changes and genetic introgression in species evolution, we analyzed populationsofCobitis vettonicaasacasestudy.Thisspeciesisanendangered Iberian freshwater fish whose distribution is restricted toafewriversintheTagusandDuerobasinsinthemid-western IberianPeninsula(Figure1).ItinhabitsriverswithlowpHandwater hardness levels, andgravel androckybottoms,andismorecommonlyfoundintheheadwatersoftheserivers(Carmonaetal.,1999; Collares-Pereiraetal.,2021;Doadrioetal.,2011;Perdices&Coelho, 2020).Incontrast,thesisterspeciesofC. vettonica,Cobitis paludica (Doadrio&Perdices,2005;Perdices&Doadrio,2001),isageneralistthatinhabitsmostoftheIberiandrainagesincludingTagusand Douro,withapreferenceforstreamsclosetothemainchannelwith highsuspendedsolids,highwaterhardness,lowtransparency,low currentvelocityandmuddybottoms(Carmonaetal.,1999;Doadrio etal.,2011).SympatriczonesofC. vettonicaandC. paludicahave beenreportedinthewesternlimitsofthe distributionareaofC. vettonicabutgeneticintrogressionbetweenthemhasnotbeenreported(Perdices&Coelho,2020).However,insomeeasternpopulationstherehasbeenmentionofgeneticintrogressionbetweenC. vettonicaandC. paludica,butnomoreinformationhasbeendetailed (Doadrio et al., 2011, 2021). Sister species of Iberian freshwater fishesgenerallyhaveanallopatricdistributionthatwasestablished mainlybeforetheQuaternary(Doadrio,1988;Sousa-Santosetal., 2019).Somespeciesnowshowpatternsofsympatryincertainareas as a consequence of secondary contacts during the Quaternary, whichhasledtogeneticintrogression,asisthecaseofsomespecieswithinthegeneraofLuciobarbusandPhoxinus(Corral-Louetal., 2019;Denysetal.,2013;Machordometal.,1990).Inthecaseof C. vettonicaandC. paludica,phylogeneticstudieshaveshownthey diverged during the Plio-Pleistocene (Doadrio & Perdices, 2005; Sousa-Santosetal.,2014).However,boththeoriginofthesympatric zonesbetweenC. vettonicaandC. paludicaandtheplausiblegenetic introgressionbetweentheminthewesterndistributionareaofC. vettonicaareunknown.Allthesepointsmakefisheswithrestricted distributionareasasexcellentmodelstodeciphertheQuaternary effectsonthediversificationoffreshwater-restrictedtaxawithina glacialrefugeastheIberianPeninsula(Ganteetal.,2009). PhylogeographicstudiesofIberianprimaryfreshwaterfishes have mainly used mitochondrial genes combined with nuclear markerssuchasmicrosatellitesorintrons(Casal-López&Doadrio, 2018;Corral-Louetal.,2019;Gonzalezetal.,2014,2018;Perea &Doadrio,2015).Inrecentyears,phylogeographicandpopulation genetics studies have taken advantage of next-generation sequencing(NGS)technologiestobroadlyscreenthegenomeat highresolution,yetsomechallengesremainintheanalysisofNGS FIGURE 1 Samplinglocalitiesincludedinthisstudy.Theorange shadingdelimitstheDueroBasin,andthegreenshadingdelimits theTagusBasin.NumberscorrespondtothoselistedinTable1 | 3 of 22 CORRALLOU et AL. datasuchasthehandlingoflarge-scaleandcomplexdata,theupstreamprocess of pipeline,andthelimitednumber ofavailable reference genomes (McCormack et al., 2013; Tan et al., 2019). Combined analyses of single-nucleotide polymorphisms (SNPs) with mitochondrial and nuclear markers have proven useful to providemorecompleteandmoredetailedphylogeographicand biogeographic models of the past and present relationships of populationsofvariousspecies (Corral-Louetal.,2021;Mendes etal.,2019;Zarraonaindiaetal.,2012). The main aim of this study is to provide a robust biogeographicmodelforthespeciesC. vettonicathroughoutitsentire distributionasawitnessoftheevolutionoftheIberianPeninsula throughouttheQuaternary.Wealsodeciphertherelationshipof C. vettonica with its sister species C. paludicaatthelimitsofits distributionarea.Inaddition,weassessthepopulationstructure andgeneticdiversityofthespeciesacrossitsdistributiontorevise the Operational Conservation Units (OCUs; Doadrio et al., 1996)previouslyestablishedforC. vettonica(Doadrioetal.,2021) ortoestablishnewones.MoredetailedandaccurateOCUsand decipheringtherelationshipwithitssisterspeciesthroughoutits entirerangearekeyfortheeffectivemanagementandconservationoftheendangeredC. vettonica.Toachievethesegoals,we analyzed the mitochondrial marker cytochrome b(MT-CYB), the nuclearmarkerrecombination activating gene 1(RAG1),andaset ofSNPsobtainedthroughnext-generationsequencingofpopulationsthroughouttheentiredistributionrangeofC. vettonicaand someofC. paludicafromadjacentsub-basins. 2 | MATERIALS AND METHODS 2.1 | Sampling, DNA extraction, amplification and sequencing Wesampled 204individualsofC. vettonicaand60ofC. paludica fromatotalof38localities(Figure1;Table1).The27localities(23in theTagusand4intheDuero)inwhichC. vettonicawasfoundcover theentireknowndistributionareaofthespecies,andincludesits typelocality(ÁrragoRiver,TagusBasin).ThesamplesofC. paludica werecollectedfrom13localities,includingitstypelocality(Tiétar River,TagusBasin):ofthese,11wereeitheradjacenttothoseofC. vettonicaorinsympatriclocalitiesintheTagusBasin,andtwowere intheDueroBasin(Figure1;Table1).Tissuesampleswereobtained from the DNA and Tissue Collection at the National Museum of NaturalSciencesofMadrid(MNCN–CSIC;TableS1).Sequencesof MT-CYBofC. vettonicaandC. paludicaavailableinGenBankwere alsoincludedinthestudy(33ofC. vettonicaand7ofC. paludica; TableS1). For each individual, DNA was extracted from ventral fin tissue using the Qiagen DNeasy® Blood and Tissue Kit (Qiagen, Inc., Valencia, CA, USA), following the manufacturer's protocol. Polymerasechainreaction(PCR)wasusedtoamplify1140bpofMT- CYBfrom208individuals(164ofC. vettonicaand44ofC. paludica), and1500bpofRAG1from90individuals(66ofC. vettonicaand24 ofC. paludica)(TableS1).Amplificationswereperformedfollowing theprotocoldescribedbyDoadrioandPerdices(2005)forMT-CYB, andCorral-Louetal.(2021)forRAG1. All sequences (the new ones and the ones downloaded from GenBank;TableS1)forMT-CYB(248sequences,240fromthepresentstudyand40fromGenBank;TableS1)andRAG1(90sequences fromthepresentstudy;TableS1)werealignedusingMAFFT(Katoh & Standley, 2013), as implemented in Geneious 10.1.3 (https:// www.geneious.com;Kearseetal.,2012),andthenmanuallyexamined. Alleles of RAG1were separated using the PHASEalgorithm (Stephens&Donnelly,2003),asimplementedinDnaSPv.6.10.01 (Rozasetal.,2017). 2.2 | Genotyping and SNP filtering FortheSNPstudy,weselectedeightpopulationsofeachC. vettonicaandC. paludica(Table1)basedonthestructureobserved inapreviousstudyusingmitochondrialdataandavailableDNA (Doadrioetal.,2021).Atotalof106individuals(60ofC. vettonica and46ofC. paludica;TableS1)wereusedtopreparelibrariesfor doubledigestrestrictionsite-associatedDNAsequencing(ddRAD- seq)followingtheprotocoldescribedbyKessetal.(2016).AllDNA sampleswereusedasaninputforacustomlibrarypreparation protocol.Librariesweredual-indexedforpostsequencingdemultiplexing.ThesampleswereruninaNovaSeq6000PE150lane. Trimmomatic0.36(Bolgeretal.,2014)wasusedtoremoveadapters (ILLUMINACLIP option). Using the process_radtags program inSTACKS2.4(Catchenetal.,2013),allreadsweretruncatedto thesamelengthof95bp,andlow-qualityreadswereremoved usingthe-qparameteraccordingtophred33system.Adenovo_ mapanalysiswasalsoperformedinSTACKS,inwhichdifferent programswereruntoassemblelociineachindividual(ustacks), buildacatalogue(cstacks),matchallgenericsamplesagainstthe catalogue(sstacks),andreconstructlociusingR2readsandidentifySNPsusingthemetapopulationinformation(gstacks).Priorto runningtheSTACKSmodules,severaltestswereusedtoidentify whichparametersmaximizedthenumberofSNPsobtainedinat least 80% of the individuals (r80 rule; Paris et al., 2017). Since SNPswereobtainedforbothspecies,weexploredparametersin aconservative wayinthede_novomapsmoduleofstacks.The parameter m (i.e., the minimum depth of coverage required to createastack)wassetto5duetothedepthofcoveragevalues obtained.TheMparameter(i.e.,numberofmismatchesallowed betweenstackswithinindividuals)hastobecarefullyset.Ifitis settoohigh,paralogousornonhomologouslocicanbeincorrectly mergedintothesamelocusandifitissettoolow,homologous locicanbelost.Forthisreason,theparameterMwasexplored between3and5.FollowingtheindicationsofParisetal.(2017), thevaluesofn(i.e.,thenumberofmismatchesallowedbetween stacksbetweenindividuals)wereexploredfrom2to4(TableS2). Finally,theselectedparameterswerem =5,M =3,andn = 2. The 4 of 22 | CORRALLOU et AL. TABLE 1 Informationonthesamplinglocalitiesincludedinthepresentstudy No Species River Basin Subbasin Locality Province mtDNA nDNA SNPs 1Sympatricarea Alfrividas Tagus Ponsul Alfrividas Portugal 4/1 – – 2C. vettonica Aravil Tagus Aravil Alcafozes Portugal 1– – 3C. vettonica Aravil Tagus Aravil CegonhasNovas Portugal 4 – – 4C. vettonica Erjas Tagus UpperErjas Valverdedel Fresno Spain 12 9? 5 5C. vettonica SanMartin Tagus UpperErjas SanMartínde Trevejo Spain 6 1 – 6C. vettonica Trevejana Tagus UpperErjas Cilleros Spain 4 4 – 7C. vettonica Erjas Tagus Middle Erjas Termasde Monfortinho Portugal 6– – 8C. vettonica Arades Tagus Middle Erjas Salvaterrado Extremo Portugal 2– – 9C. vettonica Erjas Tagus LowerErjas Salvaterrado Extremo Portugal 2– – 10 Sympatricarea Erjas Tagus LowerErjas CabeçaQueimado Portugal 1/1 – – 11 Sympatricarea Erjas Tagus LowerErjas AzenhadoRoque Portugal 4/1 – – 12 Sympatricarea Erjas Tagus LowerErjas Serrinha Portugal 2/1 – – 13 C. vettonica Árrago Tagus Western Alagón CadalsodeGata Spain 15 3? – 14 C. vettonica Gata Tagus Western Alagón Gata Spain 19 5 5 15 C. vettonica SanBlas Tagus Western Alagón Gata Spain 1– – 16 C. vettonica Acebo Tagus Western Alagón Hoyos Spain 3 1 – 17 C. vettonica Mayas Duero Águeda Descargamaría Spain 2– – 18 C. vettonica Mayas Duero Águeda ElSahugo Spain 20 410 19 C. vettonica Águeda Duero Águeda LaHerguijuela Spain 1– – 20 C. vettonica Turones Duero Águeda LaBouza Spain 1 1 – 21 C. vettonica Alagón Tagus Eastern Alagón SantiIbañezdela Sierra Spain 14 9 4 22 C. vettonica Francia Tagus Eastern Alagón NavadeFrancia Spain 14 5 8 23 C. vettonica Cuerpode Hombre Tagus Eastern Alagón Sotoserrano Spain 19 910 24 C. vettonica Ladrillar Tagus Eastern Alagón LaRebollosa Spain 1– – 25 C. vettonica Hurdano Tagus Eastern Alagón VegasdeCoria Spain 1 1 – 26 C. vettonica Caparro Tagus Eastern Alagón Zarzade Granadilla Spain 23 78 27 C. vettonica Jerte Tagus Eastern Alagón Navaconcejo Spain 15 510 28 C. paludica Alvito Tagus Ocreza MonteGordo Portugal 1– – 29 C. paludica Aurela Tagus Salor Santiagode Alcántara Spain 5 5 9 30 C. paludica Toula Tagus Aravil Alcafozes Portugal 1– – 31 C. paludica Aravil Tagus Aravil CegonhasNovas Portugal 1– – 32 C. paludica Salor Tagus Salor Membrío Spain 2– 2 33 C. paludica Tamuja Tagus Almonte Trujillo Spain 4 – 4 | 5 of 22 CORRALLOU et AL. programpopulations,alsoinSTACKS,wasusedtofiltertheSNPs usingthefollowingparameters:-p16-r0.80,– min_maf =0.05, – max_obs_het =0.75,and– write_random_snp. TwomatricesfortheSNPswerebuilt.Theallowedpercentage ofmissingdataperlocusandperindividualwas30%(TableS3).The firstmatrixincludedthepopulationsofC. vettonicaanddatafrom60 individualsfrom8localitiesand4000polymorphicloci.Thesecond includedpopulationsofbothC. vettonicaandC. paludica,referred toasC. vettonica + C. paludica,anddatafrom106individualsfrom 16localitiesand4538polymorphicloci. 2.3 | Phylogeny Aphylogenetictreewasconstructedtoassessrelationshipsbased onthecollapsedMT-CYBhaplotypesmatrix.TheselectedsubstitutionmodelswereSYM,HKY+I,andGTR+Gforthefirst,second,andthirdposition,respectively,basedontheresultsobtained inPartitionFinder2(Lanfearetal.,2017)usingtheAkaikeinformationcriterion(AIC;Akaike,1974).Theanalysiswasimplementedin MrBayesv3.2(Ronquistetal.,2012),withtwosimultaneousindependentrunseachwithfourMarkovchainMonteCarlo(MCMC), whichwererunfor5× 107generations.Thefirst25%ofgenerationswereremovedasburn-in.Posteriorprobability(pp)values wereusedtoassessthereliabilityofthephylogenetichypothesis. TwosequencesofotherspeciesofCobitiswereusedasoutgroups: C. bilineata(EF605321.1)andC. zanandreai(AF263089.1).Thegenetic divergencebetweenthelineagesobtained intheMT-CYB analysis was evaluated through uncorrected p-distances with 1000permutationsusingMEGAv.6.0(Tamuraetal.,2013).For the SNP data, phylogenetic relationships were evaluated using thecompleteSNPmatrix(C. vettonica + C. paludica)throughthe MaximumLikelihood(ML)methodintheRaxMLprogramimplementedinCIPRESScienceGatewayv3.3.Theevolutionarymodel selectedwasASC_GTRGAMMA,asrecommendedintheprogram manual for data that contain only variable sites, and the Lewis ascertainmentbiascorrectionwasused(Stamatakis,2016).The robustness of the tree was evaluated with 1000 bootstrap (b) replicates. 2.4 | Genetic structure To examine the genetic structure of all studied populations of C. vettonica and C. paludica, haplotype networks for the two genes wereconstructedusingthemedian-joiningalgorithm(Bandeltetal., 1999)implementedinPopART(Leigh&Bryant,2015).Analysisof molecularvariance(AMOVA)wasusedtodeterminethesourceof thegeneticvariationinMT-CYBforthepopulationsidentifiedasC. vettonica(excludingV4subset,seebelow)usingdifferentgroupings basedonbasin,sub-basin,andevolutionarylineage.TheseanalyseswereimplementedinArlequinv.3.11(Excoffier&Lischer,2010) with 10,000 permutations. Population differentiation in terms of Φst(Hudsonetal.,1992)betweenallpopulationpairs(exceptthose representedbyonlyoneindividual)forthecompletedataset(C. vettonicaandC. paludica)wasalsocalculatedinArlequinv.3.11.Forthe localitiesinwhichmitochondrialhaplotypesofbothC. vettonicaand C. paludicawerefound,theΦstwascalculatedbytreatingtheindividualsofeachspeciesasseparatepopulations. FortheSNPsdata,onlyneutrallociweretakenintoaccountto studythepopulationstructureinordertoavoidthebiasthatcould becausedbycandidatelociforselection.Forthisreason,weevaluated the presence of loci under selection with BayeScan v.2.0 (Foll & Gaggiotti, 2008), using the default parameters except for thepriorodds(priorodds=100;Lotterhos&Whitlock,2014).We removedthelociunderselectionandconstructedtwonewmatricesfortheneutralloci,keeping3998lociintheC. vettonicamatrix and4500lociintheC. vettonica + C. paludicamatrix.STRUCTURE (Pritchard et al., 2000) was used to assess the genetic structure basedonthetwoSNPmatrices.Themostprobablenumberofsubpopulations(K)foreachanalysiswasequaltothenumberofpopulationsstudied+1(K =9andK =17,respectively).Weperformed10 independentsimulationsforeachKwithaburn-inlengthof50,000 and50,000MCMCrepetitionsaftertheburn-in.Themostprobable number of subpopulations for each matrix was estimated by takingintoaccounttheresultsofboththeΔK(Evannoetal.,2005) andthePuechmaillemethod(Puechmaille,2016).Thewebserver StructureSelector(Li&Liu,2018)wasusedtomaketheseestimations.WeselectedKbasedonbothmethodsbecauseagreaterprobabilityofK =2existswhenthestructureisanalyzedwiththeΔK No Species River Basin Subbasin Locality Province mtDNA nDNA SNPs 34 C. paludica Almonte Tagus Almonte Jaraicejo Spain 5 5 5 35 C. paludica Ibor Tagus Lower Tagus CastañardeIbor Spain 3– 3 36 C. paludica Tiétar Tagus Tiétar LaIglesueladel Tiétar Spain 11 5 6 37 C. paludica Yeltes Duero Huebra- Yeltes SanMartínde Yeltes Spain 5 5 8 38 C. paludica Huebra Duero Huebra- Yeltes Elcubodedon Sancho Spain 9 4 9 Notes: Alsodetailedarethenumberofindividualsstudiedforeachofthemarkertypes:mtDNA(MT-CYB),nDNA(RAG1),andSNPs.Underthe mtDNAcolumn,foreachsympatricarea,thetwonumbersindicatethenumberofindividualsanalyzedforC. vettonicaandC. paludica,respectively. TABLE 1 (Continued) 6 of 22 | CORRALLOU et AL. method(Janesetal.,2017),andsimilarresultsfordifferentmethods indicateaclearsignal(Puechmaille,2016). 2.5 | Genetic diversity and demography DifferentgeneticdiversityparameterswereestimatedforbothMT- CYBandRAG1inDnaSPforeachriver,sub-basin,species,evolutionarylineage,andOCUs.Thestudiedgeneticdiversityparameters werehaplotypeornuclearallelesnumber(h or aforMT-CYBand RAG1,respectively),haplotypeorallelicdiversity(Hd or adforMT- CYBandRAG1,respectively),nucleotidediversity(π),andnumberof polymorphicsites(S).ForMT-CYB,thepopulationswithsharedhaplotypesbetweenC. vettonicaandC. paludicawereanalyzedconsideringonlytheindividualsidentifiedasC. vettonica.FortheSNPdata, adifferentsetofgeneticdiversityparameterswasestimatedusing the populationsprograminSTACKS2.4(Catchenetal.,2013).These parameterswereevaluatedforthetwomatricesindependently(C. vettonicaandC. vettonica + C. paludica);however,onlytheresultsfor C. vettonicaareshown.Inaddition,athirdmatrixwasstudiedtaking intoaccountthepopulationsofbothspecies,exceptthepopulations oftheAurelaandSalorriversofC. paludica,duetothecloserelationshipthattheselocalitiesshowedwiththelocalitiesoftheErjas sub-basinofC. vettonica. ToevaluatethedemographyofpopulationsofC. vettonica,deviationsfromamodelofmutation-driftequilibriumforMT-CYB were testedusingFu'sFs(Fu,1997)andTajima'sD(Tajima,1989),asimplementedinArlequinv.3.11. 2.6 | Divergence times and niche modeling DivergencetimesamongthepopulationsofC. vettonica were estimatedusingarelaxedlognormalclockandacoalescentmodelonthe collapsedMT-CYBhaplotypesmatrixinBEASTv1.8.4(Drummond etal.,2012).Themolecularclockwascalibratedusing,asanormal prior,anevolutionaryrateof0.34%divergenceperlineagepermillionyears(Doadrio&Perdices,2005),andthesubstitutionmodel usedwasGTR+ I. ToidentifythepotentialnicheofC. vettonicainbothpastand presentscenarios,ecologicalnichemodels,asappliedinthesoftwareMaxEntv.3.4.1(Phillipsetal.,2017),wereusedtopredicta meanhabitatsuitabilityvalueforeachlocality.Toselectthemost appropriateofthe19accessiblebioclimaticvariablesdownloaded from the WorldClim dataset (Fick & Hijmans, 2017) for the final analysis,correlationandcollinearityanalysesandpreliminaryanalyseswerefirstperformed.Threevariablesbasedontheecological characteristics of the species were chosen for the final analysis: Isothermality (BIO3), Precipitation of Driest Month (BIO14), and PrecipitationSeasonality(BIO15).Thedatamatrixonthepresence ofthespeciesincludedonlythestudiedpopulationsofC. vettonica thatdonothaveanyputativegeneticintrogressionwithC. paludica. Thenichemodelingwasmadeforthreeperiodsoftime:Present(the years1970–2000),LastGlacialMaximum(LGM;~22,000yearsbeforepresent,BP),andLastInterglacial(LIG;~120,000–140,000BP). ThequalityofthemodelwasevaluatedbytheAreaUndertheCurve (AUC)derivedfromtheReceiverOperatingCharacteristic(ROC). 2.7 | Ancestral area reconstructions TheancestralareasofC. vettonicawerereconstructedusingboth theS-DIVA(Yuetal.,2010)andS-DEC(Beaulieuetal.,2013;Ree & Smith, 2008) methods implemented in RASP v.4.2 (Yu et al., 2015,2020).TheestablishedareaswerebasedonthecurrenthydrographicsubdivisionoftheIberianPeninsula.Duetothegenetic differentiationdetectedforMT-CYBin the Alagón sub-basin,this areawasdividedintoaneasternandawesternpart.Thecodefor theareaswasasfollows:A:Águeda,B:WesternAlagón,C:Eastern Alagón,D:Aravil,E:Erjas,F:Ponsul,G:Salor,H:Ocreza,I:Aurela,J: HuebraandYeltes,K:TamujaandAlmonte,L:Ibor,andM:Tiétar.We usedtheBEAST-derivedtreesastheinputfilesforRASP.Weeliminated50%ofthetotalinitialtrees,andused100randomtreesfor theanalysis.Ancestralrangeswerelimitedtoincludenomorethan threeadjacentareas.OnlythereconstructedareasforC. vettonica areshownasthislineageisthefocusofthepresentstudy. 3 | RESULTS 3.1 | Phylogeny ThephylogeneticanalysisofMT-CYBrecoveredtwomainlineages correspondingtothespeciesC. vettonica(S1)andC. paludica(S2) withhighsupport(pp=1;Figure2).ThespeciesC. vettonica(S1) was not monophyletic since many individuals identified morphologicallyasC. vettonicafromtheAlagónsub-basinpopulationswere foundinbothlineages(i.e.,Jerte,Francia,Alagón,Caparro,Cuerpo deHombre[Cdh],andGatarivers).ThelineageofC. paludica(S2) wasalsonotamonophyleticgroupsinceoneindividualidentified morphologicallyasC. paludicafromtheSalorsub-basinwasincluded intheC. vettonicalineage(S1).WithinC. vettonica(S1),therewere twowell-differentiatedlineages(Figure2).ThefirstgroupedpopulationsfromtheAlagónsub-basinintheTagusBasin(Árrago,Gata, SanBlas,Acebo,Alagón,FranciaCdh,Ladrillar,Hurdano,Caparro, andJerterivers)andthosefromtheÁguedasub-basinintheDuero Basin(Turones,Mayas,andÁguedarivers).Withinthislineage,these populationsgroupedintotwodistinctlineages,V1andV2.ThepopulationsfromtheeasternAlagónsub-basin(Alagón,Francia,Cdh, Ladrillar,Hurdano,Caparro,andJerterivers)formedtheV2lineage (pp=0.99),whilethosefromthewesternAlagón(Árrago,Gata,San Blas,andAceborivers)andÁguedasub-basins(Turones,Mayas,and Águeda rivers),and one individual fromJerteRiver(fromeastern Alagónsub-basin)thatwasnotincludedinV2,formedtheV1lineage.ThephylogeneticrelationshipsofthesequencesincludedinV1 werenotresolved.However,withinV2,wedetectedanotherhighly | 7 of 22 CORRALLOU et AL. supportedmonophyleticgrouping(pp=1)inwhichonlythesingle individual studied from Ladrillar River was excluded. The second well-differentiatedlineagewithinC. vettonica(V3;pp=1)included populations from the Ponsul, Aravil, and Erjas sub-basins located alongthewesternlimitofthespecies’distributionrange,andone individualfromSalorRiver(Figure2). GeneticdistancesamongthethreelineagesdetectedinC. vettonica(V1,V2,andV3)rangedfrom0.3%to1.2%(Table2). TheunrootedphylogenetictreeofC. vettonicaandC. paludica basedonthe4538SNPswaslargelycongruentwiththeresultsobtainedinthephylogeneticreconstructionforMT-CYB,althoughwith someslightdifferences(Figure3).Theindividualsofthepopulations analyzed for SNPs belonging to the V1 lineage (i.e., Mayas from Águedasub-basin,DueroBasinandGatafromwesternAlagónsub- basin,TagusBasin)werewelldifferentiatednotonlyfromtherest oftheindividualsanalyzed(b>90)butalsofromeachother(b>90 and75< b >90,respectively).ThosebelongingtotheV3lineage (ErjasRiver),whichwereseparatedfromtherestofthepopulations ofC. vettonica,weremorerelatedtoindividualsofC. paludicafrom AurelaRiver,atributaryoftheTagusBasinthatisgeographically closetotheErjassub-basin.Likewise,individualsofC. vettonicafrom theCaparroriverpopulation(belongingtoV2andS2)weremore related to those of C. paludica from the populations in the Salor, Tamuja,Ibor,Huebra,andYeltesrivers.Theremainingpopulations ofC. vettonicaincludedintheSNPanalysisbelongingtotheV2and S2lineages(Alagón,Cdh,Jerte,andFranciarivers)didnotshowa clearphylogeneticstructurewithseveralintermediateindividuals. 3.2 | Genetic structure ThehaplotypenetworkforMT-CYBwasconsistentwiththephylogenetic reconstruction for MT-CYB. The network showed two main groups separated by 20 mutational steps corresponding to the species C. vettonica and C. paludica (S1 and S2, respectively; Figure4).ForC. vettonica(S1),thenetworkwasmoreinformative FIGURE 2 PhylogenetictreebasedonBayesianinferenceforthemitochondrialmarkerMT-CYB.Posteriorprobabilityvaluesare indicatedabovebranches.S1:Cobitis vettonica.Theverticalcoloredbarsrepresentthethreemainlineages(V1–V3)ofthespeciesandtheir localities(seelegend).S2:Cobitis paludica.Also,themapisshownindicatingthesamplinglocationsreferencedinTable1.Thecolorofthe samplingpointindicatesthedetectedlineageineachpopulationbasedontheresultsofthephylogenetictree TABLE 2 Uncorrectedabsolutegeneticdistancesbetweenthe mainlineagesofC. vettonica(V1,V2,andV3)andC. paludica(S1). Thestandarddeviationisshownontheupperdiagonal V1 V2 V3 S2 V1 0.002 0.003 0.005 V2 0.003 0.003 0.004 V3 0.011 0.012 0.005 S2 0.022 0.025 0.026 8 of 22 | CORRALLOU et AL. thanthephylogeneticreconstructionsincetherewerethreewell- differentiatedhaplogroups(V1,V2,andV3).Thefirsthaplogroup (V1)clusteredpopulationsfromtheÁguedasub-basinintheDuero BasinandintheTagusBasin,thosefromthewesternpartofAlagón sub-basin, plus one individual belonging to Jerte River (eastern Alagónsub-basin)(Figure4).Thesepopulationscorrespondedtothe FIGURE 3 UnrootedMLphylogenetic treebasedonthe4538SNPloci. Circlesandsquaresindicateindividuals morphologicallyidentifiedasC. vettonica andC. paludica,respectively.Thenumbers refertothoseassignedtothepopulations inTable1.Thebootstrapvalues(b)are representedwithalineperpendicular tothebranchfor75< b >90,andtwo linesforb>90.Thearrowshighlight thepopulationsofC. vettonicafromthe Caparro(→)andupperErjas(←)rivers. Abbreviation:R.forriver.Also,themapis shownindicatingthesamplinglocations referencedinTable1 FIGURE 4 HaplotypenetworkforthemitochondrialmarkerMT-CYB.Thelocalitiesareindicatedbydifferentcolorsandnumbers(see Table1).Mutationalstepsarerepresentedasfollows:oneshortlinefortwosteps,twoshortlinesforthreesteps,oracirclewiththe numberofstepsindicatedforfourormoresteps.V4*refersonlytothoseindividualsidentifiedasC. vettonicaandfoundintheS2lineage (individualsindicatedbypinkarenotincludedinV4*).Inthelegend,populationsidentifiedasC. vettonicaarerepresentedbycircles,and thoseidentifiedasC. paludica,bysquares.Populationsthatcontainindividualswithcontrastingmorphologicalandgeneticidentifications areunderlined.Abbreviations:R.forriver;S.forsub-basinandB.forbasin | 9 of 22 CORRALLOU et AL. nondifferentiatedlineageobtainedinthephylogeneticreconstruction(i.e.,V1inFigure4).Thesecondhaplogroup(V2)wascomposed ofalltheeasternAlagónsub-basinpopulationsintheTagusBasin, correspondingtotheV2lineageobtainedinthephylogeneticreconstruction.Likewise,thethirdhaplogroup(V3),correspondingto theV3lineageinthephylogeneticreconstruction,includedpopulationsfromthewesternlimitofthedistributionrangeofC. vettonica intheTagusBasin,namelythosefromthePonsul,Aravil,andErjas sub-basins,plusoneindividualfromSalorRiver.Theindividualfrom SalorRiver(identifiedasC. paludica)presentedthemostcommon haplotypeofthehaplogroup.HaplogroupsV1andV2wereseparatedbythree/fourmutationalsteps,andV3wasseparatedfromV1 by10mutationalsteps.Inthegroupcomprisinghaplotypesfound mainly in C. paludica (S2), a subset was composed of individuals identifiedasC. vettonicafromtheAlagónandErjassub-basins(V4; referringtoonlythoseindividualsidentifiedmorphologicallyasC. vettonicabutthataregeneticallyclosertoC. paludicaaccordingto themitochondrialdata)alongwithotherindividualsofC. paludica fromotherTagusandDuerotributaries.However,noevidenceofa geographicstructurewasfoundwithinS2.Thehaplotypenetwork forthenuclear marker(RAG1) showeda singlegroup inwhichall studiedpopulationsofbothspecieswererepresentedandtherefore wasoflowresolution(Figure5). AccordingtotheAMOVAswiththeMT-CYBdata,thehighest percentageofvarianceamonggroupswaswhenthegroupingswere comprisedofthe lineagesfoundinthepreviousanalysis (V1,V2, V3,andV4;Table3).Thepercentageofvarianceexplainedbetween basins(DueroandTagus)wasnotsignificant,withmostofthevariation explained by differences between and within populations. Populationdifferentiationbetweenlocalitieswithineachlineageof C. vettonicaforMT-CYBwaslow(Φst <0.34),andnosignificantdifferenceswerefound,exceptfortheAceboRiver,whichshoweda highlevelofdifferentiationwiththeÁrragoriver(Φst >0.74;Table S4).HighvaluesofΦstweredetectedbetweenpopulationsfromdifferentlineages(Φst >0.75). TheresultsoftheSNPstructureanalysesshowedthatthemost probablenumberofsubpopulations(geneticgroups)forC. vettonicamatrixwasK =4andK =2forPuechmailleandΔKmethods, respectively(TableS5).ForK =4,onlythreepopulationsshowed verylittleadmixture:MayasRiverintheDueroBasin(firstgroup,in yellow)andCaparroandErjasriversintheTagusBasin(secondand thirdgroups,inblueandpurple,respectively)(Figure6).Theother populations(Alagón,Cdh,Jerte,andFrancia)showeddifferentdegreesofadmixtureofthesecond(blue)andfourth(orange)groups. GataRiverhadasimilarlevelofadmixtureofthefirst(yellow)and fourth(orange)groupswithasmallcontributionfromthethird(purple).TheresultsforK =2wereconsistentwiththoseobtainedfor K =4.MayasandGatabelongedtothefirstgroup(orange),Caparro tothesecondgroup(blue),andtherestofthepopulationspresented amixtureofbothgroups.Whenweanalyzedthepopulationsofboth species,themostprobablenumberofsubpopulationswasK =6and K =2forPuechmailleandΔKmethods,respectively(TableS5).For K =6,MayasandGatacomprisedthefirstgroup(orange),withGata havingasmallcontributionfromthethirdgroup(purple)inallindividuals(Figure6).Alagón,Cdh,Jerte,Francia,andCaparroshowed differentdegreesofadmixtureofthefirst(orange)andsecond(blue) groups,withCdhandFranciaalsohavingasmallproportionofthe thirdgroup(purple)andJertehavingasmallproportionofthefourth group(pink).Thesecondgroup(blue)waspresentinallpopulations ofC. paludica.TheupperErjasRiverwasassignedtoasinglegenetic group(Group3;purple),whichalsocomprisedapproximately29% and15%ofthetotalcompositionoftheAurelaandSalorrivers,respectively,withverysmallcontributionsfromthesecond(blue)and fourth(pink)groupsinsomeindividuals.TheresultsforK = 2 were consistentwiththoseobtainedforK =6.MayasandGatabelonged tothefirstgroup(orange),Ibor,Tamuja,Huebra,andYeltestothe secondgroup(blue),andtherestofthepopulationspresentedamixtureofbothgroups. 3.3 | Genetic diversity and demography Overallgeneticdiversityparametersforallindividualsidentifiedas C. vettonicabasedonMT-CYBwerehigh(Table4).ExcludingpopulationsofC. vettonicainwhichonlyoneindividualwasstudied,Hd rangedfrom0to1,andhrangedfrom1to7.Thepopulationswith thelowestvaluesofgeneticdiversitywereMayas,Acebo,Arades, andCaparrowithonlyonehaplotypeeach(Table4).Themostvariablepopulationsforallofthediversityparameterswerethelower andmiddleErjas,Alagón,andGata.Thesub-basinswiththehighest FIGURE 5 HaplotypenetworkforthenuclearmarkerRAG1.Thecolorscorrespondtothedifferentlocalities,exceptpink,whichisused foralllocalitiesofC. paludica.Mutationalstepsarerepresentedasfollows:nolineforonestepandoneshortlinefortwosteps.Inthe legend,thepopulationsidentifiedasC. vettonicaarerepresentedbycirclesandthoseidentifiedasC. paludicaarerepresentedbysquares. NumberscorrespondtothoseusedinTable1 16 of 22 | CORRALLOU et AL. smallnumberofindividualsfromtheriversofthewesternAlagón sub-basin(assuggestedinourancestralareareconstructions). 4.1.2 | (V2)EasternAlagónsub-basin Thisgroup,composedofthepopulationsfromtheeasternAlagón sub-basin,wasnotmonophyleticbasedonMT-CYB,andintheSNP analysis,presentedahighlevelofadmixturewithC. paludica. These resultssuggestextensivegeneticintrogressionofC. vettonica with its sister species C. paludica,exceptinHurdanoandLadrillarrivers, whereonlyoneindividualfromeachwasstudied.Therefore,regardlessofthedifferentlevelsofgeneticintrogression,weproposethat thesepopulationsallbeassignedtotheOCUthatwasrecentlyestablishedbyDoadrioetal.(2021)(OCUIII;Figure10). 4.1.3 | (V3)Ponsul,Aravil,andErjassub-basins According to the MT-CYB analyses, the Ponsul, Aravil, and Erjas sub-basin populations constituted a well-differentiated independentlineageofC. vettonica,supportingevidencefromotherstudies thatsuggesttheuniquenessoftheErjaspopulations(Doadrioetal., 2011,2021).TheupperErjaswasalsoanindependentgroup,accordingtotheSNPanalyses,indicatingitspresent-dayisolationwith respecttotherestofthepopulationsintheAlagónsub-basinand DueroBasin.Forthesereasons,weincludetheupperErjas,Ponsul, andAravilsub-basinsintheOCUpreviouslyestablishedforonlythe Erjassub-basin(Doadrioetal.,2021),therebyexpandingthisOCU’s areaofcoverage(OCUIV;Figure10). 4.2 | Hybridization Hybridizationduetointrogressionasaresultofbothanthropicand naturalfactorshasbeendetectedforotherMediterraneancyprinid and cobitid species (Almodóvar et al., 2012; Bohlen & Ráb, 2001; Choleva et al., 2014; Corral-Lou et al., 2019; Cunha et al., 2004; Denysetal.,2013;Jankoetal.,2005;Pereaetal.,2016;Tancioni etal.,2013).WithintheIberianPeninsula,themajorityofthenaturalhybridpopulationsoriginatedduringtheQuaternaryperiodasa consequenceofhydrographicalrearrangementsand/orPleistocene glaciation–deglaciationcycles(Almodóvaretal.,2012;Gante,2009). Consistentwiththis,wepresentstrongevidenceconfirmingtheintrogressionpreviouslyreportedintheeasternAlagónpopulationsof C. vettonicabyDoadrioetal.(2011),Doadrioetal.(2021),andalso postulatethegeneticintrogressionofpopulationsinsympatricareas (Ponsul,Aravil,andlowErjas;seebelow).Supportingthishypothesis isthelocationofthesepopulationsoutsidethepotentialrangeofthe speciesbasedontheecologicalnichemodels.Theintrogressionof evenafewlociduetohybridizationcanleadtoadaptivedivergenceif thelociarebetteradaptedtoothernichesandpersistinthegenome (Abbottetal.,2013). 4.2.1 | (V1)WesternAlagónsub-basinand DueroBasin TheGatapopulationwithinthewesternAlagónsub-basinwasthe onlyonethatpresentedamitochondrialhaplotypeofC. paludica. However,intheSNPanalysis,noneoftheindividualsshowedadmixturewithC. paludica,andtheyalsoresolvedasaphylogenetically isolatedgroup.Therefore,atpresent,thereisnoevidencetosupportgeneticintrogressionwithC. paludicainthisarea. 4.2.2 | (V2)EasternAlagónsub-basin WefoundstrongevidenceofgeneticintrogressionbetweenC. paludicaandC. vettonicaintheeasternAlagónsub-basin,inlinewith previous studies (Doadrio et al., 2011, 2021). These populations notonlyshowedahighlevelofadmixturewithC. paludicabutalso highvaluesofgeneticdiversityintheSNPanalyses.Indeed,they hadthehighestvaluesofallthepopulations,aswouldbeexpected forhybridspopulations.Thehighdiversityvaluesobtainedforthe SNPdatawerenotcongruentwiththelowonesobtainedforMT- CYB.However,thisobservationisnotsurprisingashybridstypically cannotbedetectedbymitochondrialmarkersduetotheirmaternal inheritance. ThecaseofthepopulationfromCaparroRiverisparticularlyinteresting.ItwastheonlyeasternAlagónsub-basinpopulationthat showed a closer phylogenetic relationship with populations of C. paludicathanwiththoseofC. vettonicaandahigherdegreeofgeneticintrogression.Thisfactcouldbeexplainedbytwohypotheses. ThefirstisthatthesourcepopulationwasC. paludicaand,through afoundereffect,afewindividualsofC. vettonicaestablishedthemselvesinthislocality,givingrisetoapopulationofC. paludicaintrogressedbyC. vettonica.Thesecondhypothesisisthatthesource populationwasC. vettonica,butitwasdisplacedbyC. paludicaafter thisspecies’arrivalduetoitsbetteradaptedgenome(Abbottetal., 2013). 4.2.3 | (V3)Ponsul,Aravil,andErjassub-basins Accordingtothemitochondrialanalyses,thePonsul,Aravil,andErjas sub-basinpopulationsofC. vettonicawererelatedtoapopulation ofC. paludicafromSalorRiverwhosemostcommonhaplotypewas alsosharedbyallofthesepopulationsofC. vettonica.Inaddition,the resultsoftheSNPanalysisrevealedastrongrelationshipbetween theErjaspopulation(upperErjas)withthoseintheSalorandAurela sub-basins,particularlythelatterwithwhichitwasmostcloselyrelatedphylogenetically.Therelationshipbetweenthesepopulations wasmoreevidentwhentheAurelaandSalorpopulationswereremovedfromtheSNPgeneticdiversityanalysis,whichresultedinan increaseinthenumberofprivateallelesfromeightto141.However, intheSNPstructureanalysis,onlyoneindividualbelongingtothe Erjaspopulationhadaverylowcontributionofthegeneticgroup | 17 of 22 CORRALLOU et AL. assignedtoC. paludica.Altogether,theresultsdonotsupportthe presenceofgeneticintrogressionintheupperErjas,wherenomitochondrialhaplotypesofC. paludicawerefound.Moreover,higher valuesofgeneticdiversitywouldbeexpectedfortheSNPdataif therewereintrogressions,asoccursintheeasternAlagónsub-basin populations. However,wepostulategeneticintrogressioninthePonsul,Aravil, andlowerErjassub-basin,wherebothspeciesarepresent(Perdices &Coelho,2020).Thispostulationisbasedontheirlocationoutside theareaindicatedbytheecologicalnichemodelsandthefactthat introgression by hybridization is a common phenomenon that has beendetectedfornotonlyC. vettonicaandC. paludicabutalsoother speciesofEuropeanCobitis(Bohlen&Ráb,2001;Jankoetal.,2007; Ráb&Slavík,1996).Furthermore,thisgeneralpatternhasoftenbeen observedintheIberianPeninsulafortwosympatricspeciesasaconsequenceofsecondarycontactafterdiversification(Corral-Louetal., 2019;Doadrioetal.,2021;Pereaetal.,2016,2021). Withinthisgroup,hybridlocalitiesseemtoberestrictedtothe lowerreachesofrivers,whiletheupperreachesareinhabitedby onlyC. vettonica(Perdices&Coelho,2020).Thehabitatpreferences ofthetwospecies,C. paludicainthemiddleandlowerreaches,and C. vettonicaintheupperreaches(Doadrioetal.,2011),likelyexplains therestricteddistributionofthehybrids. 4.3 | Biogeography Accordingtoourdivergencetimeestimations,thetimingofthese divergencesagreeswithacommonperiodofpopulationdifferentiationformanyIberianfreshwaterfishspecies,whichinturncoincideswiththeculminationoftheformationoftheIberiandrainage networkandthePleistoceneglacialcycles(Casal-Lópezetal.,2017; Corral-Louetal.,2019,2021;Paisetal.,2012;Pereaetal.,2016). Althoughgeologicalinformationforthestudyareaisscarce,several tectonicandclimaticeventsareknowntohaveoccurredduringthe Quaternary near the study area, which influenced the hydrogeomorphologyoftheaquaticnetworksuchasthefluvialcapturephenomena,thepresenceofpaleoglaciersduringthelastglacialcycle, andchangesinthetrajectoryofsomerivers(Benitoetal.,2003; Carrascoetal.,2013,2015;Goyetal.,2020).Therefore,wepropose severalhypothesesbasedonthebiogeographyofthespecies,which mayrevealthehydrogeomorphologicalevolutionofthestudyarea withintheIberianPeninsula. TheearliestdivergenceinC. vettonicawasthePonsul-Aravil-Erjas lineage(V3),whichsplitfromthetwoAlagónsub-basinlineages(V1- V2)duringthePleistocene(Calabrianperiod,~250,000Mya).Based ontheancestralareareconstructions,thediversificationofthese lineagesoccurredintheareacurrentlycomprisingboththeErjasand Alagónsub-basins.ThisimpliesthattheErjasacquireditscurrent configurationduringthisperiodoftime,interruptinganygeneflow thatmayhavebeenestablishedbyconnectionsbetweentributaries ontherightbankofErjasRiverandtheleftbankofÁrragoRiver. Subsequently,followingtheisolationanddiversificationoftheErjas sub-basinpopulationfromtherestofthepopulations,individuals from this sub-basin dispersed to downstream areas of the Tagus Basin(reachingPonsul,AravilandSalorsub-basins).Thesedispersal eventscouldhavetakenplaceasaconsequenceofclimaticchanges associatedwiththeLGM,assupportedbyournichemodelswhich indicated the Ocreza, Nisa, Sever, Ponsul, Aravil, and Salor sub- basinsasmoresuitablehabitatsthantherestoftheriversinthe currentdistributionrangeofthespecies.Theseeventsgaveriseto boththecolonizationofnewareastothewest(PonsulandAravil sub-basins)andthesympatrywithC. paludica,whichledtointrogressionevents. Thedivergenceofthetwootherlineagescomprisingthewestern andtheeasternAlagónsub-basinpopulations(V1andV2)alsooccurredduringthePleistocene(Chibanianperiod,~80,000Mya).The complexgeologyoftheAlagónsub-basinisevidentbytheanomalouscurrenttrajectoryofsomeofitsrivers(Carrasco&Pedraza, 1991;DíezHerrero,2003;Goyetal.,2020;Jiménez,1994;Schnabel &Gutiérrez,2014).Forinstance,thetwobanksoftheAlagónsub- basin,representedbythetwomitochondriallineages,arecurrently joinedbythemouthoftheÁrragoRiverintheAlagónRiver,that is,themainriverofthewesternandtheeasternAlagónsub-basin, respectively.Theunionbetweenthetwobanksoccurredrelatively recentlyasaconsequenceoftheupriveractionoftheAlagónRiver, whichcapturedtheCaparroRiverthateventuallyflowedintothe Tagus through the Fresnedosa riverbank (Schnabel & Gutiérrez, 2014).Giventhiscontext,weproposetwoscenariosforthedivergenceofV1andV2.Inthefirst,thetwolineagesdivergedpriorto theunionofthetwobanksandlatergeneticflowbetweenpopulationswasnotpossibleduetothepoorhabitatconditionsforC. vettonicainthelowerAlagónsub-basin.Inthesecond,theydiverged aftertheunionofthebanksduetoclimaticchangesduringtheLGM. According to our niche models, the optimal ecological conditions forC. vettonicawerelocatedfurtherdownstreamand,aspredicted byphylogeographytheory,thelatitudinaldistributionoforganisms retractedsouthwards(Ehrichetal.,2007;Hewitt,1996;Rodríguez etal.,2011).Onceoptimalconditionsreturnedtotheriverheadwaters,thespecieswentbackupstreamtowardboththeeastandwest marginsoftheAlagónsub-basin,andsubsequentlydiversified. AfterthedivergenceofthetwobanksoftheAlagón,thepopulations in the Duero and western Alagón sub-basin diverged. SeveralstudieshavereportedonthegeneticstructureoffreshwaterfaunaassociatedwithconnectionsbetweentheTagusBasin and adjacent basins (e.g., Alagón, Alberche and Lozoya rivers; Carmonaetal.,2000;Casas-Sainz&DeVicente,2009;Doadrio, 1988;Pérez-González,1980;Sousa-Santosetal.,2007).Although recent connections may be explained as a consequence of the tectonic activity in the region (Goy et al., 2020), there are no geologicalstudiesthatsupportarecentconnectionbetweenthe westernAlagónsub-basin(TagusBasin)andtheÁguedasub-basin (DueroBasin),despitethecloseproximityofsomeoftheirriver headwaters(insomecases,aslittleas150moflineardistance). 18 of 22 | CORRALLOU et AL. A connection, however, has been hypothesized to explain the co-occurrence of individuals with the mitochondrial genome of Squalius carolitertii(distributedintheDueroBasin)andthoseof S. pyrenaicus(distributedintheTagusBasin)intheÁrragoRiver (Pereaetal.,2021).ThecloseproximityofMayas(DueroBasin) andÁrrago(TagusBasin)rivers,whosecoursesareonlyseparated by~150m,isparticularlynotable.PiracyeventsthatoccurredbetweenthesetworiversinthePleistocenemayexplaintheclose mitochondrialrelationshipofthepopulationsinhabitingthem,as suggestedbythedispersaleventestimatedinourancestralarea reconstructionanalysis.LaterisolationoftheDueroandTagushydrologicalbasinsthenledtothedivergenceofthesepopulations, ascorroboratedbytheSNPanalysis. 4.4 | Conservation ThehabitatsofC. vettonicahavebeenandcontinuetobethreatenedbythemaincausesofbiodiversityloss(e.g.,overexploitation, waterpollution,flowmodification,habitatdestructionanddegradation,andtheintroductionofinvasivespecies)(Doadrioetal.,2011; Dudgeon,2019;Sousa-Santosetal.,2014).Giventhesethreats,the populationsinOCUI(Águedasub-basin)areparticularlyvulnerable astheyhadthelowestgeneticdiversityvaluesforbothMT-CYBand theSNPs,andonlyoneMT-CYBhaplotypesharedbyallthepopulations.Thislowlevelofgeneticdiversitymakesthemmoresensitivetoextrinsicchangesandthereforeatgreaterriskofextinction (Frankhametal.,2002).Inaddition,populationsinOCUIareshrinkingasaresultoftheformationofdamsalongtheÁguedasub-basin, andinsomelocalities(i.e.,Turones),theyhaveevendisappeared. Thespecies’distributioninthisareaisnowrestrictedtoafewtributariesofMayasRiver.AlthoughthegeneticdiversityvaluesofpopulationsinOCUIIwerehigherthanthoseofOCUI,thepossibilityof abottleneckfollowedbyapopulationexpansioncannotberejected forthesepopulations,despiteadecreaseinthenumberofindividualsobservedinrecentyears(Doadrioetal.,2011).Exoticspecies, suchasLepomis gibbosusandMicropterus salmoides,havecolonized theupperpartsoftheriverscoveredbyOCUII,whichmaybeone ofthemaincausesforthedeclineinthenumberofC. vettonicain theserivers.Otherpotentialcausesofthisdeclineremainunknown. For OCU IV, there was no genetic distinction between the non-introgressed(upperErjas)andputativelyintrogressed(Ponsul, Aravil,andlowerErjas)populations.Despitethis,anymanagement planmusttakeintoconsiderationthispotentialintrogressionasthe indiscriminate mixing of individuals from these two groups could leadtogeneticintrogressionwithC. paludicathroughouttheentire areaofOCUIV.Cobitis paludica,whichismoreofageneralistthan C. vettonica,iswidelydistributedthroughouttheIberianPeninsula, occupyingagreatvarietyofecologicalniches(Doadrioetal.,2011). Thus, genetic introgression by C. paludica in sympatric localities would likelyprove disadvantageous forC. vettonicaingeneral,as twoofthefourOCUs(IIIandIV)wouldbeaffectedbythisphenomenon.ItcouldleadtotheextinctionofC. vettonicaintheseareasor toadaptivevariations,resultinginamajorlossofthegeneticdiversityofthisspecies. ACKNOWLEDGMENTS WethankP.Garzón,I.DoadrioJr,J.L.González,andG.Gonzálezfor collectingalmosteverysamplefrom2009to2010undertheproject “BasisforthemonitoringofSpanishFreshwaterFishes”,supportedby theMinisteriodeMedioAmbienteyMedioRuralyMarino(MARM), andC.Marcos,J.CVelaco,andG.Gonzálezfortheirhelpintheproject Life13nat/es/000772.WegreatlythankL.Alcarazforlaboratoryassistance.WearegratefulforthevaluablesuggestionsandEnglishediting performedbyM.Modrell.ThisstudywasalsosupportedbyamanagementcommissionfundedbytheDueroHydrographicConfederationin Spain(Análisisyestudiosfilogeográficosydevariabilidadgenéticade endemismosibéricosdeciprínidos:Life13nat/es/000772“Actuaciones paralaprotecciónyconservacióndeciprínidosibéricosdeinteréscomunitario”)andbygrantIND2017/AMB-7699toAC-L,fundedbythe Community of Madrid (Spain) through its Industrial Doctorate program. We acknowledge the Santander Supercomputacion support groupattheUniversityofCantabriaforprovidingaccesstothesupercomputerAltamiraattheInstituteofPhysicsofCantabria(IFCA-CSIC), amemberoftheSpanishSupercomputingNetwork,whichwasusedto performsimulations/analyses. CONFLICT OF INTEREST Theauthorsdeclarenoconflictofinterest. AUTHOR CONTRIBUTIONS Andrea CorralLou: Conceptualization (lead);Datacuration (lead); Formal analysis (lead); Methodology (equal); Resources (lead); Software (lead); Validation (lead); Visualization (lead); Writing – originaldraft(lead);Writing–review&editing(lead).Silvia Perea: Conceptualization (equal); Data curation (equal); Formal analysis (supporting); Methodology (supporting); Supervision (supporting); Writing–review&editing(equal).Anabel Perdices:Resources(supporting);Supervision(supporting);Writing–review&editing(supporting). Ignacio Doadrio: Conceptualization (lead); Data curation (equal);Fundingacquisition(lead);Investigation(lead);Methodology (lead); Resources (equal); Supervision (lead); Validation (lead); Visualization(equal);Writing–review&editing(equal). DATA AVAILABILITY STATEMENT The new sequences of the mitochondrial (MT-CYTB; OM234794- OM235001) and nuclear (RAG1; OM235002-OM235091) markers obtainedfromthisstudyareavailableinGenBank.TheSNPdatahave beendepositedinVCFformatinFigshare(https://doi.org/10.6084/ m9.figshare.18778148.v1). ORCID Andrea CorralLou https://orcid.org/0000-0003-0895-1305 Silvia Perea https://orcid.org/0000-0003-0436-8577 Anabel Perdices https://orcid.org/0000-0003-0980-215X Ignacio Doadrio https://orcid.org/0000-0003-4863-9711 | 19 of 22 CORRALLOU et AL. REFERENCES Abbott,R.,Albach,D.,Ansell,S.,Arntzen,J.W.,Baird,S.J.E.,Bierne,N., Boughman,J.,Brelsford,A.,Buerkle,C.A.,Buggs,R.,Butlin,R.K., Dieckmann,U.,Eroukhmanoff,F.,Grill,A.,Cahan,S.H.,Hermansen, J. S., Hewitt, G., Hudson, A. G., Jiggins, C., … Zinner, D. (2013). Hybridizationandspeciation.Journal of Evolutionary Biology,26(2), 229–246.https://doi.org/10.1111/j.1420-9101.2012.02599.x Akaike, H. (1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control,19(6),716–723.https://doi. org/10.1109/TAC.1974.1100705 Almodóvar,A.,Nicola,G.G.,Leal,S.,Torralva,M.,&Elvira,B.(2012). NaturalhybridizationwithinvasivebleakAlburnus alburnusthreatens thesurvivalofIberianendemiccalandinoSqualiusalburnoides complex and Southern Iberian chub Squalius pyrenaicus. Biological Invasions, 14(11), 2237–2242. https://doi.org/10.1109/ TAC.1974.1100705 Alonso, J. L., Pulgar, F. J. Á., & Pedreira, D. (2007). The relief of the CantabrianMountains.Teaching Earth Sciences [In Spanish],15(2), 151–163.https://doi.org/10.1109/TAC.1974.1100705 Bandelt,H.J.,Forster,P.,&Röhl,A.(1999).Median-joiningnetworksfor inferringintraspecificphylogenies.Molecular Biology and Evolution, 16(1), 37–48. https://doi.org/10.1093/oxfordjournals.molbev. a026036 Beaulieu, J. M., Tank, D. C., & Donoghue, M. J. (2013). A Southern Hemisphere origin for campanulid angiosperms, with traces of thebreak-upofGondwana.BMC Evolutionary Biology,13(1),1–17. h t t p s : / / d o i . o r g / 1 0 . 1 1 8 6 / 1 4 7 1 - 2 1 4 8 - 1 3 - 8 0 Benito, G., Sopeña, A., Sánchez-Moya, Y., Machado, M. J., & Pérez- González,A.(2003).PalaeofloodrecordoftheTagusRiver(centralSpain)duringtheLatePleistoceneandHolocene.Quaternary Science Reviews, 22(15–17), 1737–1756. https://doi.org/10.1016/ S 0 2 7 7 - 3 7 9 1 ( 0 3 ) 0 0 1 3 3 - 1 Bohlen, J., & Ráb, P. (2001). Species and hybrid richness in spined loachesofthegenusCobitis(Teleostei:Cobitidae),withachecklistofEuropeanformsandsuggestionsforconservation.Journal of Fish Biology,59,75–89.https://doi.org/10.1111/j.1095-8649.2001. t b 0 1 3 8 0 . x Bolger, A. M.,Lohse,M.,& Usadel,B. (2014).Trimmomatic: aflexible trimmerforIlluminasequencedata.Bioinformatics,30(15),2114– 2120.https://doi.org/10.1093/bioinformatics/btu170 Buj, I., Marčić, Z., Ćaleta, M., Šanda, R., Geiger, M. F., Freyhof, J., Machordom, A., & Vukić, J. (2017). Ancient connections among theEuropeanriversandwatershedsrevealedfromtheevolutionaryhistoryofthegenusTelestes(Actinopterygii;Cypriniformes). PLoS One, 12(12), e0187366. https://doi.org/10.1371/journ al.pone.0187366 Carmona, J. A., Doadrio, I., Márquez, A. L., Real, R., Hugueny, B., & Vargas,J.M.(1999).DistributionpatternsofindigenousfreshwaterfishesintheTagusRiverbasin,Spain.Environmental Biology of Fishes,54(4),371–387.https://doi.org/10.1023/A:1007535200837 Carmona, J. A., DomÍnguez, J., & Doadrio, I. (2000). Congruence betweenallozymeandcytochromebgenesequencedatainassessing geneticdifferentiation withintheIberianendemicChondrostoma lemmingii(Pisces:Cyprinidae).Heredity,84(6),721–732.https://doi. org/10.1046/j.1365-2540.2000.00720.x Carrasco, R. M., & Pedraza, J. (1991). In the Jerte valley [In Spanish]. Cuaternario Y Geomorfología, 5, 15–25. https://doi. org/10.1046/j.1365-2540.2000.00720.x Carrasco,R.M.,Pedraza,J.,Domínguez-Villar,D.,Villa,J.,&Willenbring, J. K. (2013). The plateau glacier in the Sierra de Béjar (Iberian Central System) during its maximum extent. Reconstruction and chronology.Geomorphology,196, 83–93.https://doi.org/10.1109/ TAC.1974.1100705 Carrasco, R. M., Pedraza, J., Dominguez-Villar, D., Willenbring, J. K., & Villa, J. (2015). Sequence and chronology of the Cuerpo de Hombrepaleoglacier(IberianCentralSystem)duringthelastglacial cycle. Quaternary Science Reviews, 129, 163–177. https://doi. org/10.1016/j.quascirev.2015.09.021 Casal-López,M.,&Doadrio,I.(2018).TheMessinianimprintonthe evolutionoffreshwaterfishesofthegenusLuciobarbusHeckel, 1843 (Teleostei, Cyprinidae) in the western Mediterranean. Journal of Biogeography, 45(7), 1593–1603. https://doi. org/10.1111/jbi.13244 Casal-López,M.,Perea,S.,Sousa-Santos,C.,Robalo,J.I.,Torralva,M., Oliva-Paterna, F. J. & Doadrio, I. (2017). Paleobiogeography of anIberianendemicspecies,Luciobarbussclateri(Günther,1868) (Actinopterygii,Cyprinidae),inferredfrommitochondrialandnuclearmarkers.JournalofZoologicalSystematicsandEvolutionary Research,56(2),127-147.https://doi.org/10.1111/jzs.12192 Casas-Sainz,A.M.,&DeVicente,G.(2009).Onthetectonicoriginof Iberiantopography.Tectonophysics,474(1–2),214–235.https://doi. org/10.1016/j.tecto.2009.01.030 Catchen,J.M.,Hohenlohe,P.A.,Bernatchez,L.,Funk,W.C.,Andrews,K. R.,&Allendorf,F.W.(2013).Stacks:Ananalysistoolsetforpopulationgenomics.Molecular Ecology,22(11),3124–3140.https://doi. org/10.1111/mec.12354 Choleva,L.,Musilova,Z.,Kohoutova-Sediva,A.,Paces,J.,Rab,P.,&Janko, K.(2014).Distinguishingbetweenincompletelineagesortingand genomicintrogressions:completefixationofallospecificmitochondrialDNAinasexuallyreproducingfish(Cobitis;Teleostei),despite clonalreproductionofhybrids.PLoS One,9(6),e80641.https://doi. org/10.1371/journal.pone.0080641 Collares-Pereira,M.J.,Alves,M.J.,Ribeiro,F.,Domingos,I.,Almeida,P. R.,DaCosta,L.,Gante,H.,Filipe,A.F.,Aboim,A.,Rodrigues,P.M., &Magalhães,M.F.(2021).Guide to freshwater fish and migrants from mainland Portugal [In Portuguese](p.292).EdiçõesAfrontamento. Corral-Lou,A.,Perea,S.,Aparicio,E.,&Doadrio,I.(2019).Phylogeography andspeciesdelineationofthegenusPhoxinusRafinesque,1820 (Actinopterygii: Leuciscidae) in the Iberian Peninsula. Journal of Zoological Systematics and Evolutionary Research, 57(4), 926–941. https://doi.org/10.1111/jzs.12320 Corral-Lou,A.,Perea,S.,&Doadrio,I.(2021).HighgeneticdifferentiationintheendemicandendangeredfreshwaterfishAchondrostoma salmantinumDoadrioandElvira,2007fromSpain,asrevealedby mitochondrialandSNPmarkers.Conservation Genetics,22(4),585– 6 0 0 .  h t t p s : / / d o i . o r g / 1 0 . 1 0 0 7 / s 1 0 5 9 2 - 0 2 1 - 0 1 3 8 1 - y Cunha, C., Coelho, M. M., Carmona, J. A., & Doadrio, I. (2004). PhylogeographicalinsightsintotheoriginsoftheSqualius alburnoides complexviamultiplehybridizationevents.Molecular Ecology,13(9), 2807–2817.https://doi.org/10.1111/j.1365-294X.2004.02283.x Denys,G.P.J.,Dettai,A.,Persat,H.,Doadrio,I.,Cruaud,C.,&Keith,P. (2013).StatusoftheCatalanchubSqualius laietanus(Actinopterygii, Cyprinidae) in France: Input from morphological and molecular data.Knowledge and Management of Aquatic Ecosystems,408,04. https://doi.org/10.1051/kmae/2013039 Díez Herrero, A. (2003). Geomorfología e Hidrología fluvial del río Alberche. Modelos y SIG para la gestión de riberas. Serie Tesis Doctorales2. Publicacionesdel Instituto Geológico yMinerode España,Madrid.587 pp. Doadrio,I.(1988).DelimitationofareasintheIberianPeninsulaonthe basisoffreshwaterfishes.Bonner Zoologische Beiträge,39,  1 1 3 – 1 2 8 . Doadrio,I.,Carmona,J.A.,&Machordom,A.(2002).Haplotypediversity andphylogeneticrelationshipsamongtheIberianbarbels(Barbus, Cyprinidae) revealtwoevolutionarylineages.Journal of Heredity, 93(2),140–147.https://doi.org/10.1093/jhered/93.2.140 Doadrio,I.,Perdices,A.,&Machordom,A.(1996).Allozymicvariationof theendangeredkillifishAphanius iberusanditsapplicationtoconservation.Environmental Biology of Fishes,45(3),259–271. Doadrio,I.,&Perdices,A.(2005).Phylogeneticrelationshipsamongthe Ibero-Africancobitids(Cobitis,Cobitidae)basedoncytochromeb 20 of 22 | CORRALLOU et AL. sequencedata.Molecular Phylogenetics and Evolution,37(2),484– 493.https://doi.org/10.1016/j.ympev.2005.07.009 Doadrio,I.,Perea,S.,&Corral-Lou, A. (2021).Genetic characterization and identification of Conservation Units in endemic fish of community interest[InSpanish].Retrievedfromhttp://hdl.handle.net/10261/ 241564 Doadrio,I.,Perea,S.,Paloma,G.H.,&González,J.L.(2011).Ictiofauna continental española: bases para su seguimiento, Ministerio de Medio Ambiente y Medio Rural y Marino. Centro de Publicaciones.DGMedio NaturalyPolíticaForestal,MARM. Drummond,A.J.,Suchard,M.A.,Xie,D.,&Rambaut,A.(2012).Bayesian phylogeneticswithBEAUtiandtheBEAST1.7.Molecular Biology and Evolution,29(8),1969–1973.https://doi.org/10.1093/molbev/ mss075 Dudgeon,D.(2019).Multiplethreatsimperilfreshwaterbiodiversityin theAnthropocene.Current Biology,29(19),R960–R967.https://doi. org/10.1016/j.cub.2019.08.002 Ehrich, D., Gaudeul, M., Assefa, A., Koch, M. A., Mummenhoff, K., Nemomissa, S., Consortium, I., & Brochmann, C. (2007). Genetic consequences of Pleistocene range shifts: Contrast between the Arctic, the Alps and the East African mountains. Molecular Ecology, 16(12), 2542–2559. https://doi. org/10.1111/j.1365-294X.2007.03299.x Elith,J.,H.Graham,C.,P.Anderson,R.,Dudík,M.,Ferrier,S.,Guisan,A., J.Hijmans,R.,Huettmann,F.,R.Leathwick,J.,Lehmann,A.,Li,J.,G. Lohmann,L.,A.Loiselle,B.,Manion,G.,Moritz,C.,Nakamura,M., Nakazawa,Y.,McC.M.Overton,J.,TownsendPeterson,A.,…E. Zimmermann,N.(2006).Novelmethodsimprovepredictionofspecies’distributionsfromoccurrencedata.Ecography,29(2),129–151. https://doi.org/10.1111/j.2006.0906-7590.04596.x Evanno,G.,Regnaut,S.,&Goudet,J.(2005).Detectingthenumberof clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology, 14(8), 2611–2620. https://doi. org/10.1111/j.1365-294X.2005.02553.x Excoffier,L.,&Lischer,H.E.(2010).Arlequinsuitever3.5:Anewseries ofprogramstoperformpopulationgeneticsanalysesunderLinux andWindows.Molecular Ecology Resources,10(3),564–567.https:// doi.org/10.1111/j.1755-0998.2010.02847.x Fick,S.E.,&Hijmans,R.J.(2017).WorldClim2:New1-kmspatialresolutionclimatesurfacesforgloballandareas.International Journal of Climatology,37(12),4302–4315.https://doi.org/10.1002/joc.5086 Foll,M.,&Gaggiotti,O.(2008).Agenome-scanmethodtoidentifyselectedlociappropriateforbothdominantandcodominantmarkers:ABayesianperspective.Genetics,180(2),977–993.https://doi. org/10.1534/genetics.108.092221 Frankham,R.,Ballou,S.E.J.D.,Briscoe,D.A.,&Ballou,J.D.(2002). Introduction to conservation genetics.CambridgeUniversityPress. Fu,Y.X.(1997).Statisticaltestsofneutralityofmutationsagainstpopulation growth, hitchhiking and background selection. Genetics, 147(2),915–925.https://doi.org/10.1093/genetics/147.2.915 Gante,H.M.F.(2009).A role for introgressive hybridization in the evolution of freshwater fishes: The casestudy of Iberian Barbus (Teleostei, Cyprinidae).ArizonaStateUniversity. Gante, H. F., Micael, J., Oliva-Paterna, F. J., Doadrio, I., Dowling, T. E., & Alves, M. J. (2009). Diversification within glacial refugia: tempo and mode of evolution of the polytypic fish Barbus sclateri. Molecular Ecology, 18(15), 3240–3255. https://doi. org/10.1111/j.1365-294X.2009.04264.x Gibbard,P.L.,Head,M.J.,Walker,M.J.,&SubcommissiononQuaternary Stratigraphy(2010).FormalratificationoftheQuaternarySystem/ Period and the Pleistocene Series/Epoch with a base at 2.58 Ma. Journal of Quaternary Science, 25(2), 96–102. https://doi. org/10.1002/jqs.1338 Gonzalez,E.G.,Cunha,C.,Ghanavi,H.R.,Oliva-Paterna,F.J.,Torralva, M.,&Doadrio,I.(2018).Phylogeographyandpopulationgenetic analyses in the Iberian toothcarp (Aphanius iberus Valenciennes, 1846)atdifferenttimescales.Journal of Heredity,109(3),253–263. https://doi.org/10.1093/jhered/esx076 Gonzalez,E.G.,Pedraza-Lara,C.,&Doadrio,I.(2014).Geneticdiversity andpopulationhistoryoftheendangeredkillifishAphanius baeticus. Journal of Heredity,105(5),597–610.https://doi.org/10.1093/ jhered/esu034 Goy,J.L.,Cruz,R.,Martínez-Graña,A.,Valdés,V.,&Yenes,M.(2020). Geomorphological map and quaternary landscape evolution of theMonfragüePark(Cáceres,Spain).Sustainability,12(23),10099. https://doi.org/10.3390/su122310099 Hewitt,G.M.(1996).Somegeneticconsequencesoficeages,andtheir roleindivergenceandspeciation.Biological Journal of the Linnean Society,58(3),247–276.https://doi.org/10.1111/j.1095-8312.1996. t b 0 1 4 3 4 . x Hewitt,G.M.(2004).Geneticconsequencesofclimaticoscillationsin the Quaternary. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences,359(1442),183–195.https:// doi.org/10.1098/rstb.2003.1388 Hudson,R.R.,Boos,D.D.,&Kaplan,N.L.(1992).Astatisticaltestfordetectinggeographicsubdivision.Molecular Biology and Evolution,9(1), 138–151.https://doi.org/10.1093/oxfordjournals.molbev.a040703 Janes,J.K.,Miller,J.M.,Dupuis,J.R.,Malenfant,R.M.,Gorrell,J.C., Cullingham, C. I., &Andrew,R.L. (2017).TheK=2conundrum. Molecular Ecology, 26(14), 3594–3602. https://doi.org/10.1111/ mec.14187 Janko, K.,Culling, M. A.,Rab, P., &Kotlik,P.(2005). Iceage cloning– comparisonoftheQuaternaryevolutionaryhistoriesofsexualand clonalformsofspinyloaches(Cobitis;Teleostei)usingtheanalysis ofmitochondrialDNAvariation.Molecular Ecology,14(10),2991– 3004.https://doi.org/10.1111/j.1365-294X.2005.02583.x Janko,K.,Flajšhans,M.,Choleva,L.,Bohlen,J.,Šlechtová,V.,Rábová,M., Lajbner,Z.,Šlechta,V.,Ivanova,P.,Dobrovolov,I.,Culling,M.,Persat, H.,Kotusz,J.,&Ráb,P.(2007).DiversityofEuropeanspinedloaches (genusCobitisL.):anupdateofthegeographicdistributionofthe Cobitis taeniahybridcomplexwithadescriptionofnewmolecular toolsforspeciesandhybriddetermination.Journal of Fish Biology, 71,387–408.https://doi.org/10.1111/j.1095-8649.2007.01663.x Jiménez,A.S.D.C.(1994).Themountain-lowersurfacecontactinthe southern slope of the Sierra de Gata (Spanish Central System, Cáceres)[InSpanish].Cuaternario Y Geomorfología,8(1–2),45–55. Katoh, K., & Standley, D. M. (2013). MAFFT multiple sequence alignmentsoftwareversion7:Improvementsinperformanceandusability.Molecular Biology and Evolution,30(4),772–780.https://doi. org/10.1093/molbev/mst010 Kearse,M.,Moir,R.,Wilson,A.,Stones-Havas,S.,Cheung,M.,Sturrock, S., Buxton, S., Cooper, A., Markowitz, S., Duran, C., Thierer, T., Ashton,B.,Meintjes,P.,&Drummond,A.(2012).GeneiousBasic: Anintegratedandextendabledesktopsoftwareplatformforthe organizationandanalysisofsequencedata.Bioinformatics,28(12), 1647–1649.https://doi.org/10.1093/bioinformatics/bts199 Kess,T.,Gross,J.,Harper,F.,&Boulding,E.G.(2016).Low-costddRAD methodofSNPdiscoveryandgenotypingappliedtotheperiwinkleLittorina saxatilis. Journal of Molluscan Studies,82(1),104–109. https://doi.org/10.1093/mollus/eyv042 Lanfear, R., Frandsen, P. B., Wright, A. M., Senfeld, T., & Calcott, B. (2017).PartitionFinder2:Newmethodsforselectingpartitioned modelsofevolutionformolecularandmorphologicalphylogenetic analyses.Molecular Biology and Evolution,34(3),772–773.https:// doi.org/10.1093/molbev/msw260 Leigh,J.W.,&Bryant,D.(2015).popart:full-featuresoftwareforhaplotypenetworkconstruction.Methods in Ecology and Evolution,6(9), 1110–1116.https://doi.org/10.1111/2041-210X.12410 Li,Y.L.,&Liu,J.X.(2018).StructureSelector:Aweb-basedsoftwareto selectandvisualizetheoptimalnumberofclustersusingmultiple methods. Molecular Ecology Resources,18(1),176–177.https://doi. org/10.1111/1755-0998.12719 | 21 of 22 CORRALLOU et AL. Lotterhos,K.E.,&Whitlock,M.C.(2014).Evaluationofdemographic historyandneutralparameterizationontheperformanceofFST outlier tests. Molecular Ecology, 23(9), 2178–2192. https://doi. org/10.1111/mec.12725 Machordom, A., Berrebi, P., & Doadrio, I. (1990). Spanish barbel hybridization detected using enzymatic markers: Barbus meridionalis Risso × Barbus haasi Mertens (Osteichthyes, Cyprinidae). Aquatic Living Resources, 3(4), 295–303. https://doi.org/10.1051/ alr:1990030 Marić, S., Stanković, D., Šanda, R., Ćaleta, M., Čolić, S., Šukalo, G., & Snoj, A. (2019). Genetic characterisation of European mudminnow (Umbra krameri) populations from the Sava River system. Knowledge & Management of Aquatic Ecosystems,420,46.https:// doi.org/10.1051/kmae/2019035 Marić,S.,Stanković,D.,Wanzenböck,J.,Šanda,R.,Erős,T.,Takács, P.,Specziár,A.,Sekulić,N.,Bănăduc,D.,Ćaleta,M.,Trombitsky, I.,Galambos,L.,Sipos,S.,&Snoj,A.(2017).Phylogeographyand populationgeneticsoftheEuropeanmudminnow(Umbra krameri) with a time-calibrated phylogeny for the family Umbridae. Hydrobiologia, 792(1), 151–168. https://doi.org/10.1007/s1075 0 - 0 1 6 - 3 0 5 1 - 9 McCormack,J.E.,Hird,S.M.,Zellmer,A.J.,Carstens,B.C.,&Brumfield, R.T.(2013).Applicationsofnext-generationsequencingtophylogeographyandphylogenetics.Molecular Phylogenetics and Evolution, 66(2),526–538.https://doi.org/10.1016/j.ympev.2011.12.007 Mendes,S.L.,Coelho,M.M.,&Sousa,V.C.(2019).PopulationgenomicsdatasupportsintrogressionbetweenWesternIberianSqualius freshwaterfishspeciesfromdifferentdrainages.bioRxiv,585687. Mesquita, N., Hänfling, B., Carvalho, G. R., & Coelho, M. M. (2005). Phylogeography of the cyprinid Squalius aradensis and implicationsforconservationoftheendemicfreshwaterfaunaofsouthern Portugal. Molecular Ecology, 14(7), 1939–1954. https://doi. org/10.1111/j.1365-294X.2005.02569.x Myers,G.S.(1966).DerivationofthefreshwaterfishfaunaofCentral America.Copeia,766–773.https://doi.org/10.2307/1441405 Pais, J., Cunha, P. P., Pereira, D., Legoinha, P., Dias, R., Moura, D., González-Delgado, J. A. (2012). The Paleogene and Neogene ofwesternIberia(Portugal):aCenozoicrecordintheEuropean Atlantic domain. In The Paleogene and Neogene of Western Iberia (Portugal)(pp.1–138).SpringerBriefsinEarthSciences, Springer,SeriesID:8897,158pp.https://doi.org/10.1007/978-3- 6 4 2 - 2 2 4 0 1 - 0 _ 1 Paris,J.R., Stevens, J.R., &Catchen, J. M.(2017).Lost in parameter space: A road map for stacks. Methods in Ecology and Evolution, 8(10),1360–1373.https://doi.org/10.1111/2041-210X.12775 Perdices, A., & Coelho, M. M. (2020). First records of the Vettonian spinedloachCobitis vettonicainPortugalwithupdateonitsIberian distribution.Fishes in Mediterranean Environments,2020(002),15. https://doi.org/10.29094/FiSHMED.2020.002 Perdices,A.,&Doadrio,I.(2001).ThemolecularsystematicsandbiogeographyoftheEuropeancobitidsbasedonmitochondrialDNA sequences.Molecular Phylogenetics and Evolution,19(3),468–478. https://doi.org/10.1006/mpev.2000.0900 Perea,S.,&Doadrio,I.(2015).Phylogeography,historicaldemography and habitat suitability modelling of freshwater fishes inhabiting seasonallyfluctuatingMediterraneanriversystems:acasestudy using the Iberian cyprinid Squalius valentinus. Molecular Ecology, 24(14),3706–3722. Perea,S.,Cobo-Simon,M.,&Doadrio,I.(2016).Cenozoictectonicand climatic events in southern Iberian Peninsula: Implications for theevolutionaryhistoryoffreshwaterfishofthegenusSqualius (Actinopterygii,Cyprinidae).Molecular Phylogenetics and Evolution, 97,155–169.https://doi.org/10.1016/j.ympev.2016.01.007 Perea,S., Sousa-Santos, C., Robalo, J., &Doadrio, I. (2021).Historical biogeographyoftheIberianPeninsula:Multilocusphylogenyand ancestralareareconstructionforthefreshwaterfishgenusSqualius (Actinopterygii,Leuciscidae).Journal of Zoological Systematics and Evolutionary Research, 59(4), 858–886. https://doi.org/10.1111/ jzs.12464 Pérez-González, A. (1980). Geology and stratigraphy of the Aridos deposits in the alluvial plain of Arganda (Madrid) [In Spanish]. In M. Santonja, N. López Martínez, & A. Pérez-González (Eds.), Ocupaciones achelenses en el valle del Jarama (ArgandaMadrid)(pp. 49–61).PublicacionesdelaExcelentísimaDiputaciónProvincialde Madrid. Phillips,S.J.,Anderson,R.P.,Dudík,M.,Schapire,R.E.,&Blair,M.E. (2017).Openingtheblackbox:Anopen-sourcereleaseofMaxent. Ecography,40(7),887–893.https://doi.org/10.1111/ecog.03049 Pillans, B., & Naish, T. (2004). Defining the quaternary. Quaternary Science Reviews,23(23–24),2271–2282.https://doi.org/10.1016/j. quascirev.2004.07.006 Pritchard,J.K.,Stephens,M.,&Donnelly,P.(2000).Inferenceofpopulationstructureusingmultilocusgenotypedata.Genetics,155(2), 945–959.https://doi.org/10.1093/genetics/155.2.945 Puechmaille, S. J. (2016). The program structure does not reliably recover the correct population structure when sampling is uneven: Subsampling and new estimators alleviate the problem. Molecular Ecology Resources, 16(3), 608–627. https://doi. org/10.1111/1755-0998.12512 Querejeta, M., Fernández-González, A., Romero, R., & Castresana, J. (2017). Postglacial dispersal patterns and mitochondrial genetic structure of the Pyrenean desman (Galemys pyrenaicus) in the northwesternregionoftheIberianPeninsula.Ecology and Evolution, 7(12),4486–4495.https://doi.org/10.1002/ece3.3034 Ráb, P., & Slavík, O. (1996). Diploid-triploid-tetraploid complex of the spinedloach,genusCobitisinPšovkaCreek:Thefirstevidenceof newspeciesofCobitisintheichthyofaunaoftheCzechRepublic. Acta Universitatis Carolinae. Biologica,39,201–214. Ree,R.H.,&Smith,S.A.(2008).Maximumlikelihoodinferenceofgeographicrangeevolutionbydispersal,localextinction,andcladogenesis. Systematic Biology,57(1),4–14.https://doi.org/10.1080/10635 150701883881 Rodríguez,R.,Ramirez,O.,Valdiosera,C.E.,Garcia,N.,Alda,F.,Madurell- Malapeira,J.,&Dalen,L.(2011).50,000yearsofgeneticuniformity inthecriticallyendangeredIberianlynx.Molecular Ecology,20(18), 3785–3795.https://doi.org/10.1111/j.1365-294X.2011.05231.x Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D. L., Darling, A., Höhna,S.,Larget,B.,Liu,L.,Suchard,M.A.,&Huelsenbeck,J.P. (2012).MrBayes3.2:efficientBayesianphylogeneticinferenceand modelchoiceacrossalargemodelspace.Systematic Biology,61(3), 539–542.https://doi.org/10.1093/sysbio/sys029 Rozas, J., Ferrer-Mata, A., Sánchez-DelBarrio, J. C., Guirao-Rico, S., Librado, P., Ramos-Onsins, S. E., & Sánchez-Gracia, A. (2017). DnaSP6:DNAsequencepolymorphismanalysisoflargedatasets. Molecular Biology and Evolution, 34(12), 3299–3302. https://doi. org/10.1093/molbev/msx248 Schnabel,S.,&Gutiérrez,Á.G.(2014).AdvancesinGeomorphologyin Spain2012-2014.InS.Schnabel&Á.G.Gutiérrez(Eds.),Actas de la XIII Reunión Nacional de Geomorfología. Cáceres, Universidad de Extremadura– Sociedad Española de Geomorfología. Seifertová,M.,Bryja,J.,Vyskočilová,M.,Martínková,N.,&Šimková,A. (2012).MultiplePleistocenerefugiaandpost-glacialcolonizationin theEuropeanchub(Squalius cephalus)revealedbycombineduseof nuclearandmitochondrialmarkers.Journal of Biogeography,39(6), 1024–1040.https://doi.org/10.1111/j.1365-2699.2011.02661.x Sousa-Santos,C.,Collares-Pereira,M.J.,&Almada,V.(2007).Reading the history of a hybrid fish complex from its molecular record. Molecular Phylogenetics and Evolution,45(3),981–996.https://doi. org/10.1016/j.ympev.2007.05.011 Sousa-Santos, C., Jesus, T. F., Fernandes, C., Robalo, J. I., & Coelho, M.M.(2019). Fishdiversification atthepaceof geomorphological changes: Evolutionary history of western Iberian Leuciscinae 22 of 22 | CORRALLOU et AL. (Teleostei: Leuciscidae) inferred from multilocus sequence data. Molecular Phylogenetics and Evolution, 133, 263–285. https://doi. org/10.1016/j.ympev.2018.12.020 Sousa-Santos, C., Robalo, J. I., & Pereira, A. (2014). Threatened fishes of the world: Cobitis vettonica Doadrio & Perdices, 1997 (Cobitidae).Croatian Journal of Fisheries,72(4),174–175.https://doi. org/10.1109/TAC.1974.1100705 Stamatakis,A. (2016). The RAxML v8. 2. X Manual. Heidleberg Institute for Theoretical Studies. Retrieved from https://cmeh-itsorg/exeli xis/resource/download/NewManualpdf https://doi.org/10.1109/ TAC.1974.1100705 Stephens,M.,&Donnelly,P.(2003).Acomparisonofbayesianmethods forhaplotypereconstructionfrompopulationgenotypedata.The American Journal of Human Genetics,73(5),1162–1169.https://doi. org/10.1086/379378 Swets, J. A. (1988). Measuring the accuracy of diagnostic systems. Science, 240(4857), 1285–1293. https://doi.org/10.1109/ TAC.1974.1100705 Taberlet,P.,&Bouvet,J.(1994).MitochondrialDNApolymorphism,phylogeography,andconservationgeneticsofthebrownbearUrsus arctosinEurope.Proceedings of the Royal Society of London. Series B: Biological Sciences, 255(1344), 195–200. https://doi.org/10.1098/ rspb.1994.0028 Tajima,F.(1989).StatisticalmethodfortestingtheneutralmutationhypothesisbyDNApolymorphism.Genetics,123(3),585–595.https:// doi.org/10.1093/genetics/123.3.585 Tamura, K., Stecher, G., Peterson, D., Filipski, A., & Kumar, S. (2013). MEGA6: Molecular evolutionary genetics analysis version 6.0. Molecular Biology and Evolution, 30(12), 2725–2729. https://doi. org/10.1093/molbev/mst197 Tan,M.P.,Wong,L.L.,Razali, S.A.,Afiqah-Aleng,N.,MohdNor,S.A., Sung,Y.Y.,VandePeer,Y.,Sorgeloos,P.,&Danish-Daniel,M.(2019). Applicationsofnext-generationsequencingtechnologiesandcomputational tools in molecular evolution and aquatic animals conservation studies: A short review. Evolutionary Bioinformatics, 15, 1176934319892284.https://doi.org/10.1177/1176934319892284 Tancioni,L.,Russo,T.,Cataudella,S.,Milana,V.,Hett,A.K.,Corsi,E., &Rossi,A.R.(2013).TestingspeciesdelimitationsinfourItalian sympatricleuciscinefishesintheTiberRiver:Acombinedmorphologicalandmolecularapproach.PLoS One,8(4),e60392.https://doi. org/10.1371/journal.pone.0060392 Valdiosera,C.E.,García-Garitagoitia,J.L.,Garcia,N.,Doadrio,I.,Thomas, M.G.,Hänni,C.,&Götherström,A.(2008).Surprisingmigration andpopulationsizedynamicsinancientIberianbrownbears(Ursus arctos).Proceedings of the National Academy of Sciences of the United States of America, 105(13), 5123–5128. https://doi.org/10.1073/ pnas.0712223105 Weiss,S.,&Ferrand,N.(2007).Phylogeography of southern European refugia(pp.341–357).Springer. Wetjen,M.,Hübner,D.,Seehausen,O.,&Schulz,R.(2020).GeneticdiversityofendangeredChondrostomanasusintheRiverRhinesystem:Conservationgeneticsconsiderationsonstockingandreintroduction.Knowledge & Management of Aquatic Ecosystems,421,25. https://doi.org/10.1051/kmae/2020016 Yu, Y., Blair, C., & He, X. (2020). RASP 4: ancestral state reconstruction tool for multiple genes and characters. Molecular Biology and Evolution, 37(2), 604–606. https://doi.org/10.1093/molbev/ msz257 Yu,Y.,Harris,A.J.,Blair,C.,&He,X.(2015).RASP(ReconstructAncestral StateinPhylogenies):atoolforhistoricalbiogeography.Molecular Phylogenetics and Evolution, 87, 46–49. https://doi.org/10.1016/j. ympev.2015.03.008 Yu, Y., Harris, A. J., & He, X. (2010). S-DIVA (Statistical Dispersal- Vicariance Analysis):A tool forinferring biogeographic histories. Molecular Phylogenetics and Evolution,56(2),848–850.https://doi. org/10.1016/j.ympev.2010.04.011 Zarraonaindia, I., Iriondo, M., Albaina, A., Pardo, M. A., Manzano, C., Grant,W.S.,Irigoien,X.,&Estonba,A.(2012).MultipleSNPmarkersrevealfine-scalepopulationanddeepphylogeographicstructureinEuropeananchovy(Engraulis encrasicolusL.).PLoS One,7(7), e42201.https://doi.org/10.1371/journal.pone.0042201 SUPPORTING INFORMATION Additional supporting information may be found in the online versionofthearticleatthepublisher’swebsite. How to cite this article:Corral-Lou,A.,Perea,S.,Perdices,A., &Doadrio,I.(2022).Quaternarygeomorphologicaland climaticchangesassociatedwiththediversificationofIberian freshwaterfishes:ThecaseofthegenusCobitis (Cypriniformes,Cobitidae).Ecology and Evolution,12,e8635. https://doi.org/10.1002/ece3.8635