Bidon T, Janke A, Fain SR, Eiken HG, Hagen SB, Saarma U, Hallström BM, Lecomte N, Hailer F. 2014. Brown and polar bear Y chromosomes reveal extensive male-biased gene flow within

Article

FastT

rackBrown and Polar Bear Y Chromosomes Reveal ExtensiveMale-Biased Gene Flow within Brother LineagesTobias Bidon1 Axel Janke12 Steven R Fain3 Hans Geir Eiken4 Snorre B Hagen4 Urmas Saarma5

Bjorn M Hallstrom16 Nicolas Lecomte7 and Frank Hailer1

1Biodiversity and Climate Research Centre (BiK-F) Frankfurt am Main Germany2Goethe University Frankfurt Institute for Ecology Evolution amp Diversity Frankfurt am Main Germany3National Fish and Wildlife Forensic Laboratory Ashland OR4Bioforsk Norwegian Institute for Agricultural and Environmental Research Svanvik Norway5Department of Zoology Institute of Ecology and Earth Sciences University of Tartu Tartu Estonia6Science for Life Laboratory School of Biotechnology KTH Stockholm Sweden7Canada Research Chair in Polar and Boreal Ecology Department of Biology University of Moncton Moncton Canada

Corresponding author E-mail tobiasbidonsenckenbergde frashaigmxnet

Associate editor David Irwin

Abstract

Brown and polar bears have become prominent examples in phylogeography but previous phylogeographic studies reliedlargely on maternally inherited mitochondrial DNA (mtDNA) or were geographically restricted The male-specific Ychromosome a natural counterpart to mtDNA has remained underexplored Although this paternally inherited chro-mosome is indispensable for comprehensive analyses of phylogeographic patterns technical difficulties and low vari-ability have hampered its application in most mammals We developed 13 novel Y-chromosomal sequence andmicrosatellite markers from the polar bear genome and screened these in a broad geographic sample of 130 brownand polar bears We also analyzed a 390-kb-long Y-chromosomal scaffold using sequencing data from published maleursine genomes Y chromosome evidence support the emerging understanding that brown and polar bears started todiverge no later than the Middle Pleistocene Contrary to mtDNA patterns we found 1) brown and polar bears to bereciprocally monophyletic sister (or rather brother) lineages without signals of introgression 2) male-biased gene flowacross continents and on phylogeographic time scales and 3) male dispersal that links the Alaskan ABC islands popu-lation to mainland brown bears Due to female philopatry mtDNA provides a highly structured estimate of populationdifferentiation while male-biased gene flow is a homogenizing force for nuclear genetic variation Our findings highlightthe importance of analyzing both maternally and paternally inherited loci for a comprehensive view of phylogeographichistory and that mtDNA-based phylogeographic studies of many mammals should be reevaluated Recent advances insequencing technology render the analysis of Y-chromosomal variation feasible even in nonmodel organisms

Key words Y chromosome phylogeography bears introgression SNP microsatellite

IntroductionPhylogeography describes the origin of genetic variationamong closely related lineages tracing the geographic distri-bution of genetic variation through time and space (Avise2000 Hewitt 2000) Historically phylogenetic and phylogeo-graphic research has relied heavily on mitochondrial DNA(mtDNA) with the brown bear (Ursus arctos) as an exten-sively studied example (Taberlet et al 1998 Purvis 2005Davison et al 2011) Advantages of analyzing mtDNA includeits high mutation rate availability of markers high copynumber lack of recombination and its haploid natureHowever the typically maternal inheritance of mtDNAimplies that signatures of male-mediated dispersal cannotbe detected An approach to further investigate phylogeo-graphic patterns is to analyze independently and biparentallyinherited autosomal loci in a multilocus framework Howeverrecombination hampers inferences of haplotypes over long

genomic regions limiting the resolution that is available fromindividual autosomal loci

The only other haploid fraction of the mammalian genomeis the male-specific Y chromosome Due to its lack of recom-bination except for the small pseudoautosomal regions hap-lotypes can be inferred over extended genomic regionsproviding a high-resolution view of patrilineal evolutionaryhistory Also both mtDNA and the Y chromosome exhibitfaster lineage sorting than nuclear loci facilitating the detec-tion of population structuring (Avise 2000) The male-specificsection of the Y chromosome therefore provides an essentialcomplement to data from maternally inherited mtDNA andbiparentally inherited loci giving insight into the history ofuniquely male-inherited lineages Y-linked variation allows thedetection of potentially contrasting patterns of male andfemale gene flow (Chan et al 2012) This is particularly rele-vant in many mammals where males typically disperse much

The Author 2014 Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution All rights reserved For permissions pleasee-mail journalspermissionsoupcom

Mol Biol Evol 31(6)1353ndash1363 doi101093molbevmsu109 Advance Access publication March 25 2014 1353

by guest on May 27 2014

httpmbeoxfordjournalsorg

Dow

nloaded from

farther than females (Pusey 1987) Along with other loci Y-linked variation has therefore provided a backbone for ourunderstanding of phylogeography in humans (Hughes andRozen 2012 Wei et al 2013) canids (Brown et al 2011Sacks et al 2013) and domesticated animals (Meadowset al 2006 Lippold et al 2011)

Despite these qualities very little data is available frommammalian nonprimate Y chromosomes in part because ithas been disregarded from many genome sequencing projectsdue to its repetitive nature (Willard 2003) In addition othertechnical challenges such as avoiding co-amplification of ho-mologous X-chromosomal regions have hampered the anal-ysis of paternally inherited markers in natural populations(Greminger et al 2010) The Y chromosome thus representsan understudied part of the mammalian genome with a largepotential to add valuable information to our understandingof phylogeography In the era of genomics it is now feasible toidentify large regions on the Y chromosome and developmale-specific markers for studies of evolutionary history

Brown and polar (U maritimus) bears have been modelspecies in phylogeography since the early 1990s (Cronin et al1991 Taberlet and Bouvet 1994 Kohn et al 1995 Paetkauet al 1998 Taberlet et al 1998 Hewitt 2000 Waits et al 2000)in part because these species are widely dispersing and pro-vide the advantage of being distributed over large parts of theNorthern hemisphere Polar bears exhibit low levels of popu-lation differentiation at biparentally inherited and mitochon-drial markers throughout their range (Paetkau et al 1999Cronin and MacNeil 2012 Miller et al 2012 Campagnaet al 2013) Brown bears in contrast show considerable phy-logeographic structuring at mitochondrial markers (Davisonet al 2011 Edwards et al 2011 Hirata et al 2013 Keis et al2013) and population structuring can also be discerned atbiparentally inherited microsatellites (Paetkau et al 1997Tammeleht et al 2010 Kopatz et al 2012) Most mtDNAclades are confined to certain geographical regions and arenot shared between continents although one brown bearclade is widespread throughout Eurasia and extends intoNorth America (Korsten et al 2009 Davison et al 2011)Surprisingly all range-wide phylogeographic studies onbrown bears have so far relied on mtDNA Studies of autoso-mal markers were regionally restricted to either NorthAmerica or Eurasia (Paetkau et al 1997 Tammeleht et al2010 Kopatz et al 2012 Cahill et al 2013) and no phylogeo-graphic study of Y chromosome markers in bears existsHowever analysis of male-specific markers is crucial to un-derstand bear evolution in the light of their well-documentedmale-biased dispersal (McLellan and Hovey 2001 Zedrosseret al 2007)

With regard to bear phylogeny reliance on mtDNA alonehas proven problematic Polar bear mtDNA sequences arenested within the genetic diversity of brown bears resultingin a paraphyletic matrilineal relationship (Cronin et al 1991Lindqvist et al 2010) Although mtDNA is expected to attainreciprocal monophyly faster than nuclear loci (Avise 2000)recent studies utilizing autosomal markers have shown thatextant brown and polar bears comprise distinct sister lineagesat the species tree level and that their divergence occurred

earlier than previously estimated (Hailer et al 2012 Milleret al 2012 Cahill et al 2013) Therefore brown bear paraphylyfor mtDNA is likely a consequence of past introgressive hy-bridization with polar bears (Edwards et al 2011 Hailer et al2012 Miller et al 2012 Hailer et al 2013)

We mined a recently sequenced polar bear genome anddeveloped 13 male-specific markers to sequence 53 kb of theY chromosome and to analyze microsatellite variation in abroad geographic sample of 130 brown and polar bears fromacross Europe Asia and North America We also analyzed a390-kb-long genomic Y-chromosomal scaffold in availablebrown polar and American black bear genomes Thesedata allowed us to investigate 1) whether introgression be-tween brown and polar bears can be detected at Y chromo-some markers 2) whether the male lineage shows lessgeographic structuring than the maternal lineage and 3)the relative intraspecific clade depth of mtDNA and the Ychromosome

Results

Y Chromosome Phylogeny and Lack of IntrogressionSignals

Male-specific sequence data revealed that brown and po-lar bears carry differentiated species-specific Y chromosomeseach exhibiting a closely related group of haplotypes (fig 2A)The clear separation and reciprocal monophyly of brownpolar and American black bear (U americanus) Y chromo-somes was further supported by Bayesian phylogenetic anal-yses (fig 2B) with high statistical support (Pgt 095) for allmajor nodes

In 3078 bp of Y chromosome sequence analyzed in90 brown 40 polar and 4 black bears (fig 1 and table 1)we found over 75 of the variable sites among species (31-kbdata set solid lines in fig 2A) Only a small portion of se-quence polymorphism was intraspecific We encounteredeight haplotypes five within brown two within polarand one within black bears These haplotypes were definedby a total of 21 segregating sites 10 of which discriminatebetween brown and polar bears 9 between brown and blackbears and 13 between polar and black bears Brown and polarbears each showed one abundant haplotype that was dom-inant in all populations across their ranges Haplotype BR11was found in 94 of brown bears and PO11 in 90 of polarbears (fig 2A) Two haplotypes found in brown bears fromthe ABC islands (BR5) and the Alaskan mainland (BR4)formed a joint lineage indicative of a geographically informa-tive clustering Additional rare haplotypes in brown bearswere found in two individuals from Kamchatka (BR2) andin one individual from the Ural Mountains (BR3) In polarbears the rare haplotype PO2 was found in three individualsfrom Alaska and in one from Western Greenland (fig 1)Results for four black bear males are described in the supple-mentary material Supplementary Material online

Increasing sequence length by ~70 (adding 2216 bp53-kb data set dotted lines in fig 2A) for 63 individualschosen to represent most populations (supplementary tableS1 Supplementary Material online) increased the resolution

1354

Bidon et al doi101093molbevmsu109 MBE by guest on M

ay 27 2014httpm

beoxfordjournalsorgD

ownloaded from

among species and revealed additional rare haplotypes inbrown bears (BR12 BR13) polar bears (PO12) and blackbears (BL2) The general patterns were not substantially chan-ged compared with the 31-kb data set and still one singlehaplotype remained dominant across the distribution ranges

in each species (BR11PO11) Reflecting the few polymorphicsites found within species nucleotide diversity (pplusmn SD)was low in brown (000007 plusmn 000002) and polar bears(000003 plusmn 000002) (table 2)

Using a Bayesian approach we estimated the timing ofthe split between brown and polar bear male lineages(TMRCA (BP)) This was based on 5197 bp of Y-chromosomalsequence using the spectacled bear (Tremarctos ornatus) asoutgroup Assuming 6 Ma for the split from the spectacledbear (a calibration based on the fossil record Wayne et al1991) we estimated a TMRCA (BP) of ~112 Ma (fig 2B) Wealso constrained the analysis to a pedigree based Y-specificmutation rate (30 108sitegeneration [Xue et al 2009]rendering 30 109siteyear with a generation time esti-mate for bears of 10 years) and obtained estimates ofTMRCA (BP) of ~043 Ma (supplementary table S2Supplementary Material online) The absolute timing of thesplit therefore depended strongly on the calibration prior(ie divergence time of the outgroup or substitution rate)Additional calibration scenarios from previous studies are ex-amined in the supplementary material SupplementaryMaterial online Our data consistently recovered thebrownpolar bear split to be ~80 of the age of the oldersplit from the black bear lineage indicating that the diver-gences among different ursine species occurred relativelyshortly after each other We note however that the designof our Y sequence fragments targeted regions exhibiting nu-cleotide differences between one polar and one brown bearindividual which could lead to an upward ascertainment biaswith regard to the magnitude of the brownpolar bear diver-gence (discussed later) Nevertheless all variable sites on theblack bear branch (fig 2A) were newly discovered in our se-quencing data confirming the divergence of the black bearlineage with respect to brown and polar bears

The findings of species-specific groups of haplotypes andthe lack of haplotype sharing among species (fig 2A) revealedno signal of recent Y-chromosomal introgression In contrastanalysis of a 642-bp fragment of the mtDNA control region ofthe same samples showed polar bears nested within the var-iation of all brown bears (fig 2C) as expected for this locus

Table 1 Sample Size (n) Number of Haplotypes (H) and HaplotypeDiversity (HD) Based on the Combination of 31-kb Y-ChromosomalSequence and Six Microsatellites

Species and Population (abbreviation) n H HD

Brown bear 90 41a 096 plusmn 001

Central Europe (C-EU) 14 8 089 plusmn 006

Northern Europe (N-EU) 10 4 073 plusmn 012

Western Asia (W-AS) 8 7 096 plusmn 008

Ural Region 5 5

Central Siberia 3 2

East Asia (E-AS) 29 12 084 plusmn 005

Far East 4 4

Kamchatka 25 9

North-West America (NW-A) 10 6 084 plusmn 010

Alaska 7 4

ABC Mainland 2 1

North-Western USAIdaho 1 1

ABC islands (ABC) 11 5 082 plusmn 008

Canada (CAN) 8 2 025 plusmn 018

Polar bear 40 17a 083 plusmn 006Atlantic (ATL) 4 3 083 plusmn 022

Eastern Greenland 2 1

Iceland 1 1

Franz Josef Land 1 1

Alaska (AK) 19 7 072 plusmn 010

Western Greenland (W-GR) 8 5 079 plusmn 015

Baffin Bay 7 4

Kane Basin 1 1

Davis Strait (DS) 9 6 089 plusmn 009

Black bear 4 4 100 plusmn 018Alaska zoo Oregon Montana Vermont 4 4

aSum of haplotypes across populations is larger than the number of haplotypes perspecies due to haplotype sharing

FIG 1 Geographical distribution of analyzed bear samples Circle area is proportional to the number of individuals Some sampling localities (italics)were combined into groups (see table 1) Brown brown bears blue polar bears black black bears

1355

Male-Biased Gene Flow in Bears doi101093molbevmsu109 MBE by guest on M

ay 27 2014httpm

beoxfordjournalsorgD

ownloaded from

Y Chromosome Phylogeography of Bears

On the Y chromosome we found a maximum of three var-iable sites separating different brown bear haplotypes (eg thedifference between BR3 and BR5) but 14 substitutions be-tween brown and black bears (53-kb dataset not countingsites in microsatellite-like regions see m in fig 2A) The

intraspecific divergences relative to the outgroup obtainedfrom Bayesian analyses amounted to 27 for the Y-chromo-somal data and 59 for mtDNA control region data (fig 2Band C) Similarly estimates of mean (plusmn SE) among-group ge-netic distances from mtDNA control region sequencesshowed that divergence between two major brown bear

FIG 2 Phylogenetic relationships of bears for Y-chromosomal and mitochondrial markers (A) Parsimony network of Y chromosome sequences Solidlines variation in 31 kb dashed lines variation from additional 22 kb (total 53 kb) Circle area is proportional to number of individuals small opencircles inferred intermediate haplotypes lines represent single mutational steps Inset boxes number of individuals per population Asterisks haplotypesfound only in the 53-kb data set (individuals with these haplotypes have the respective common haplotypes in the 31-kb data set) Insertionsdeletionsof repeat units in microsatellite-like regions counted as number of repeat unit changes () Population abbreviations as in table 1 (B) Maximum cladecredibility tree of Y chromosomal sequence (5197 bp) based on a divergence of the spectacled bear 6 Ma Bold median divergence in Ma (95 highestposterior density intervals in brackets) Numbers below nodes posterior support gt095 (C) Maximum clade credibility tree of mtDNA control regiondata Sampling covers all major matrilineal brown bear clades (Davison et al 2011) (collapsed into triangles) and polar bears (clade 2B) are nested withinbrown bear variation Asterisks divergence times obtained from complete mtDNA sequences (Hirata et al 2013) Numbers below nodes posteriorprobabilities Below (B) and (C) brown bear clade depth (relative to the divergence from black bears) is indicated (D) NeighborNet network based on a~390 kb Y-chromosomal fragment from 12 polar bears 2 brown bears and 1 black bear Numbers on branches denote numbers of variable sites Withinpolar bears two haplogroups were identified corresponding to the haplotypes PO11 and PO2 in figure 2A

1356

Bidon et al doi101093molbevmsu109 MBE by guest on M

ay 27 2014httpm

beoxfordjournalsorgD

ownloaded from

mtDNA clades (1 and 3a) (0036 plusmn 0007) amounted to57ndash60 of the mean distance between brown and blackbears (0064 plusmn 0009 for clade 1 and 0061 plusmn 0009 for clade3a) Thus a considerable reduction in phylogeographic struc-turing of the patriline was detected in comparison to theestablished matrilineal pattern where deeply separatedmtDNA clades most of which are region-specific are foundwithin brown bears

This discrepancy in clade depth between the matri- andpatriline was also obvious when analyzing a ~390-kb Y-chro-mosomal scaffold (scaffold number 297) from 14 publishedmale bear genomes (Miller et al 2012) along with the corre-sponding sequence from a male brown bear from northernNorway (supplementary table S3 Supplementary Materialonline) This alignment of 2 brown 12 polar bears and 1black bear identified gt1000 high-quality variable sitesmost of them distinguishing between the three bear species(fig 2D) In this data set the divergence between the twobrown bear individuals (one from Norway and one from theABC islands) was ~5 of the divergence of these to one blackbear individual (36 substitutions between the two brownbears 752ndash758 substitutions between brown and blackbears) compared with ~20 between the divergence of allbrown bears from the black bear based on whole mitochon-drial sequences (Lindqvist et al 2010)

The shallow clade depth on the brown bear Y chromo-some could result from population expansion of one Y line-age that has replaced other clades The pattern is alsoconsistent with positive selection favoring a particular Y var-iant and male-mediated gene flow spreading this variantacross the range To disentangle the effects of backgroundselection genetic hitchhiking and recent population growthwe calculated four summary statistics to test for deviationsfrom neutral expectations In brown bears all estimates weresignificant and negative (Tajimarsquos D =194 Plt 001 Fu andLirsquos D =301 Plt 005 F =313 Plt 005 FursquosFS =4659 Plt 001 table 2) consistent with all three selec-tivedemographic processes The values calculated for polarbears were not significantly different from neutral expecta-tions (Tajimarsquos D =116 Pgt 01 Fu and Lirsquos D =142Pgt 005 F =152 Pgt 005 Fursquos FS =0649 Pgt 01table 2) Haplotype configuration tests (Innan et al 2005)did not allow us to distinguish between signals of populationstasis (g = 0) population growth (g = 2 g = 10) or selection inbrown bears because no tested scenario differed significantlyfrom neutral expectations (cumulative Pgt 005 for all tests)

In addition to sequence data we developed and analyzedsix faster evolving male-specific microsatellites to obtain ahigh-resolution data set (fig 3 and supplementary figsS1ndashS4 Supplementary Material online) Although the overallY-chromosomal haplotypic variability was high (table 1) andwe observed a ratio of haplotypes to individuals of gt40branches between haplotypes were short and defined by fewmutational steps (fig 3 and supplementary materialSupplementary Material online) Except for a group of threehaplotypes found in Central European brown bears (fig 1)and a group of 13 brown bears from eastern Asia(Kamchatka) exhibiting five differentiated haplotypes all pop-ulations contained haplotypes that were distributed acrossthe network (fig 3A)

In polar bears male-specific sequence data showed fewrare mutations (fig 2A) and even when combined with mi-crosatellites one haplotype was found to be abundant acrossmuch of the range (fig 3B) From analysis of molecular var-iance (AMOVA) we obtained estimates of the proportion ofvariation among all populations of 028 for brown and 016 forpolar bears (supplementary tables S4 and S5 SupplementaryMaterial online) This is consistent with results from autoso-mal microsatellite markers which show stronger populationdifferentiation in brown than in polar bears (Cronin andMacNeil 2012)

ABC Islands Brown BearsmdashEvidence for Male-Mediated Gene Flow from the Mainland

The Alaskan ABC islands are inhabited by brown bears thatare unique in the close relatedness of their maternal lineage topolar bears All polar and ABC islands brown bear samplesincluded in our study show this expected relationship(fig 2C) For the Y chromosome we found five haplotypesamong 11 ABC islands brown bears (fig 3A) all clusteringwith brown rather than polar bears (fig 2A) One haplotypewas shared with individuals from Canada and another withindividuals from northwest America and western Asia (fig 1)Nonsignificant differentiation from brown bears on the adja-cent North American mainland (ABCNW-A FST = 002Pgt 005 supplementary table S4 Supplementary Materialonline) but significant differentiation from all other popula-tions further confirmed the connectivity by male-mediatedgene flow This gene flow is evidently substantial enough tomaintain a high level of variability on the ABC islands wefound five haplotypes in 11 ABC islands individuals

Table 2 Summary Statistics Based on 53-kb Y-Chromosomal Sequence

Species n H fH S p plusmn SD (104) hW (104) Tajimarsquos D D F FS

Brown bear 44 6a 084 6 07 plusmn 02 26 plusmn 13 194b301b

313b4659b

Polar bear 15 2a 093 1 03 plusmn 02 06 plusmn 06 116 142 152 0649

NOTEmdashSample size (n) number of haplotypes (H) the frequency of the dominant haplotype (fH) number of segregating sites (S) nucleotide diversity () Wattersonrsquos W (persite) Tajimarsquos D Fu and Lirsquos D and F and Fursquos FS are givenaIndividuals with haplotypes BR4 and PO2 (fig 2A) were only represented in the 31-kb data set (supplementary table S6 Supplementary Material online) hence these haplotypesare not counted herebPlt 005

1357

Male-Biased Gene Flow in Bears doi101093molbevmsu109 MBE by guest on M

ay 27 2014httpm

beoxfordjournalsorgD

ownloaded from

(haplotype diversity HD = 082 table 1) which is similarly highas the variability of all brown bears combined (HD = 096table 1)

DiscussionPhylogeographic research has relied heavily on maternallyinherited mtDNA but male-biased dispersal in many mam-mals implies that mtDNA provides a highly structured (phi-lopatric) estimate of population differentiation comparedwith paternally and biparentally inherited loci Modern se-quencing techniques now allow the generation of extensivegenomic data enabling large-scale identification and analysisof sequences from the male-specific Y chromosome(Bachtrog et al 2011 Wei et al 2013) This chromosome isespecially interesting for evolutionary studies because it allowsthe inference of high-resolution haplotypes from long se-quences avoiding analytical challenges posed by interchro-mosomal recombination Our analysis of newly developedY-linked markers in comparison to results from maternallyinherited mtDNA revealed a large impact of sex-biased geneflow on phylogeographic structuring and enabled us to ex-amine phylogeny and introgression in brown and polar bears

Speciation and Introgression

The Y chromosome phylogeny of brown and polar bear lin-eages resembles the topology of species trees reconstructedfrom biparentally inherited autosomal markers (Hailer et al2012 Miller et al 2012 Cronin et al 2013) where the speciesconstitute distinct sister (or rather brother) lineages withblack bears clustering outside their variation (fig 2B) Thiscontrasts with the pattern obtained from maternally in-herited mtDNA where polar bears cluster within the varia-tion of brown bears rendering the latter paraphyletic (Croninet al 1991 Edwards et al 2011) (fig 2C)

The timing of the split between brown and polar bears hasbeen the subject of recent debates with inferred dates rang-ing from ~160000 to ~5 million years (Lindqvist et al 2010

Edwards et al 2011 Hailer et al 2012 Miller et al 2012 Cahillet al 2013 but see Ho et al 2008 and Davison et al 2011 foreven younger estimates depending on the calibration methodused) Compared with the mtDNA divergence estimate of~160000 years between polar and brown bears (Lindqvistet al 2010 Edwards et al 2011 Hirata et al 2013) divergencetimes for the Y chromosome (gt043 Ma supplementary tableS2 Supplementary Material online) are much older confirm-ing earlier suggestions that mtDNA has been introgressed(Hailer et al 2012 2013 Miller et al 2012 Cahill et al 2013)Compared with divergence times estimated from autosomaldata our 112 Ma estimate for brownpolar bear Y chromo-somes (fig 2B scenario B in supplementary table S2Supplementary Material online) is older than a divergencetime estimate from introns of ~034-093 Ma (Hailer et al2012) but younger than the 4ndash5 Ma estimate by Milleret al (2012) from genomic data When based on a rate cal-ibration from human Y chromosomes (scenario D in supple-mentary table S2 Supplementary Material online) ourestimate of the Y chromosome divergence (043 Ma) fallsinto the Middle Pleistocene resembling the estimate ofHailer et al (2012) In summary Y chromosome evidencesupport the emerging understanding of brown and polarbears as distinct evolutionary lineages that started to divergeno later than the Middle Pleistocene at least several hundredsof thousands years ago

Although incomplete lineage sorting can hamper definiteconclusions brown and polar bears likely carry introgressedalleles at mtDNA and autosomal loci (Hailer et al 2012 Milleret al 2012 Cahill et al 2013) Current hybridization levelshowever appear to be low (Cronin and MacNeil 2012Hailer et al 2012) Our findings of species-specific groups ofY chromosome haplotypes and a lack of haplotype sharingamong species revealed no signal of patrilineal introgressionReduced introgression of Y chromosomes has been reportedpreviously (eg Geraldes et al 2008) and can arise from severalmechanisms random effects of lineage sorting sex-biased hy-bridization reduced hybrid fitness of the heterogametic sex

FIG 3 Statistical parsimony networks of Y chromosome haplotypes inferred from unweighted combination of 31-kb sequence data and six micro-satellites for (A) brown bears and (B) polar bears Rare haplotype names as in figure 2A population abbreviations as in table 1

1358

Bidon et al doi101093molbevmsu109 MBE by guest on M

ay 27 2014httpm

beoxfordjournalsorgD

ownloaded from

due to genomic incompatibilities (Haldanersquos rule) or lowerintrogression rates at markers exhibiting high intraspecificgene flow (Petit and Excoffier 2009)

Variability on the Y Chromosome

Most variable sites on the Y chromosome in bears were foundamong species while only relatively little intraspecific se-quence variation was encountered The latter is compatiblewith the generally low intraspecific variability observed onmammalian Y chromosomes including field voles elephantschamois and humans (Hellborg and Ellegren 2004 Roca et al2005 Perez et al 2011 Wilson Sayres et al 2014) Nakagomeet al (2008) compared Y X and mtDNA phylogenies andvariability in bears based on single representations per speciesThey found a lower than expected Y-chromosomal substitu-tion rate within Ursinae as compared with the deeper nodesof the tree possibly mirroring our findings of low variability onthe Y chromosomes of brown and polar bears After applyinga standard correction factor of four to account for the smallereffective population size of the Y chromosome (but seeChesser and Baker 1996) variability on the brown bear Ychromosome was ~10 of that on the autosomes (datafrom Hailer et al 2012) As shown for other mammals(Hellborg and Ellegren 2004) this discrepancy between theY chromosome and autosomes exists despite higher malethan female mutation rates Low intraspecific variability onthe Y chromosome can be explained by its haploid and unipa-rental inheritance reproductive skew among males male-biased dispersal demographic history but also by selectionor a combination of these (Chesser and Baker 1996Charlesworth and Charlesworth 2000 Petit et al 2002Wilson Sayres et al 2014)

In polar bears Y-linked variability patterns did not deviatesignificantly from neutral expectations (table 2) In brownbears the deviation was significant with most of the appliedtests showing an excess of rare mutations (table 2) consistentwith population growth andor positive selection Howeverhaplotype configuration tests did not necessitate a history ofongoing or recent positive selection on the Y chromosome inbrown bears Based on SNPs from the nuclear genome Milleret al (2012) found a long-term decline in brown bear effectivepopulation size particularly since the Eemian interglacialGenome-wide data thus do not indicate recent populationgrowth reinforcing the particular evolutionary history of theY chromosome in brown bears

Despite overall low levels of intraspecific variation on the Ychromosome our analysis of long scaffold sequences (fig 2D)illustrates that application of modern genomic techniquescan nevertheless recover large numbers of polymorphicsites on the Y chromosome enabling high-resolutioninferences

Phylogeographic Structuring

mtDNA control region data show pronounced phylogeo-graphic structuring in brown bears with 1) deeply separatedclades and 2) clades which are geographically restricted(Davison et al 2011) (fig 2C) The Y chromosome is predicted

to be a geographically informative marker that shows differ-ences among populations because of strong genetic drift inthe patriline (Petit et al 2002) However we observed neitherof the abovementioned signals at paternally inherited mar-kers no deep intraspecific divergences were found and overevolutionary time scales male-biased gene flow has distrib-uted genomic variation across and among continentsCompared with mitochondrial control region data brownbear Y chromosomes showed shallow intraspecific diver-gences relative to the divergence from black bears with fewsubstitutions differentiating among Y-chromosomal haplo-types Despite limited sample numbers because to dateonly few male bear genomes have been sequenced ascertain-ment bias-free scaffold data confirm the main conclusionsfrom our sequence data First patrilineal genomic divergenceswithin brown and polar bears were considerably shallowerthan for mtDNA Second the 390-kb data set recovered thesame two groups of polar bear Y haplotypes that correspondto PO11 and PO2 Finally brown bear sequences were sepa-rated from each other by small genetic distances Althoughincreased sampling and sequencing of longer fragmentsmight recover additional clades our conclusions are not im-pacted by a strong ascertainment bias (Brumfield et al 2003)On deeper phylogenetic scales however we note that thedivergence of the black bear Y chromosome was likely under-estimated in our 31- and 53-kb data sets

The observed discrepancy between the matri- andpatriline can be due to effects of demography and selec-tion on the Y chromosome In addition mtDNA canshow signals of mutational saturation (Ingman andGyllensten 2001) and purging of slightly deleterious mu-tations due to purifying selection (Subramanian et al2009) leading to a time dependency of evolutionaryrates for mtDNA (Ho et al 2008) Whole mtDNA datafrom Lindqvist et al (2010) show relative to the diver-gence from black bears a shallower clade depth in brownbears compared with data from the control regionHowever our analysis of longer sequences from Y scaffolddata confirmed the weaker structuring of the patrilinethan the matriline Whichever the mechanism(s) a re-duced phylogeographic structuring on the Y comparedwith well differentiated mtDNA clades has also beenfound in other species for example shrews chamoisand gibbons (Lawson Handley et al 2006 Perez et al2011 Chan et al 2012)

Despite known uncertainties with regard to absolute agesour Bayesian phylogenetic analyses suggested that the mostbasal divergence of brown bear Y haplotypes considerablypredates the last glacial maximum with plausible dates reach-ing into the Middle Pleistocene (95 highest posterior den-sity 019ndash061 Ma fig 2B) This suggests that one Ychromosome lineage (BR11) has been maintained for along time and at a high frequency throughout Eurasia andNorth America While selection may therefore have contrib-uted to the shallow Y-chromosomal clade depth withinbrown bears our data are also consistent with a purely de-mographic scenario involving extensive male gene flowacross large geographical distances Indeed analysis of a

1359

Male-Biased Gene Flow in Bears doi101093molbevmsu109 MBE by guest on M

ay 27 2014httpm

beoxfordjournalsorgD

ownloaded from

390-kb-long Y-chromosomal fragment showed that twobrown bears from populations as far away from each otheras Norway and the Alaskan ABC islands carried highly similarY chromosomes (fig 2D) This pattern in brown bears coverseven larger geographic areas (throughout Eurasia and NorthAmerica) than analogous findings from humans where the Y-chromosomal lineage of Genghis Khan founder of theMongol Empire was spread across much of Asia (Zerjalet al 2003)

Our discovery of distinct Y-chromosomal haplotypeson Kamchatka mirrors previous findings of distinctmtDNA lineages (Korsten et al 2009) highlighting the com-plex biogeography of this peninsula Besides this clear signalfrom Kamchatka brown bear populations in general con-tained a mix of different Y chromosome lineages with themost closely related lineages of a given haplotype being lo-cated in a different geographic region This lack of pro-nounced patrilineal geographic structuring is an expectedconsequence of male-mediated gene flow and contrastsstrongly with the picture from mtDNA where popula-tions tend to contain region-specific lineages (Davison et al2011)

In polar bears we observed weak population structuringand no clear evidence of past phylogeographic barriers on theY chromosome This is similar to patterns from maternallyand biparentally inherited markers (Paetkau et al 1999Cronin et al 2006 Miller et al 2012 Campagna et al 2013)reflecting the large dispersal distances described for polarbears

Male-Biased Gene Flow and the Alaskan ABC IslandsBears

We provide the first direct evidence for male-mediatedgene flow between the mainland and the Alaskan ABC is-lands which host a population of bears that has long been ofinterest to evolutionary biologists due to the close matrilinealrelationship to extant polar bearsmdashthe extant polar bearmatriline is the sister lineage of the ABC clade (Cronin et al1991 Davison et al 2011) The absence of mainland brownbear mtDNA haplotypes on the ABC islands and viceversa shows that female-mediated gene flow is effectivelyzero However nuclear microsatellites (Paetkau et al 1998)and comparisons of autosomal versus X chromosomevariation (Cahill et al 2013) demonstrated that ABC bearsare not isolated from continental brown bear populationspostulating that connectivity between the ABC islands andthe mainland stems from male-mediated gene flow We hereshow that male-mediated gene flow is connecting the ABCislands to the North American mainland and that this geneflow is substantial enough to maintain appreciable geneticvariability in this island population Cahill et al (2013) sug-gested an initial polar bear ancestry of ABC islands brownbears followed by extensive male-biased immigration ofmainland brown bears Based on this scenario the fact thatwe found no polar bear Y chromosomes on the ABC islandsindicates a replacement of the original polar bear Ychromosomes

Phylogeography Insights from Matri- and PatrilinealMarkers

Since its conception the field of phylogeography has realizedthe importance of sampling several statistically independentloci (reviewed in Avise 2000) but problems related to discov-ering intraspecific variability on the Y chromosome (Hellborgand Ellegren 2004 Luo et al 2007) have long hampered theapplication of patrilineal markers in nonmodel speciesNevertheless some studies have revealed similar paternaland maternal structuring (Hellborg et al 2005) while othersrecovered discordant signals (Boissinot and Boursot 1997Roca et al 2005 Pidancier et al 2006 Perez et al 2011)Inference of the mechanism(s) that could have led to differ-ences in genetic structuring between the matri- and patrilineis generally not straightforward because the effects ofdemography and selection are difficult to disentangle(Lawson Handley et al 2006 Pidancier et al 2006Nakagome et al 2008 Perez et al 2011) even in humans(Wilson Sayres et al 2014) Regardless whether demographyor selection are the ultimate cause a weaker paternal thanmaternal structuring is indicative of gene flow among popu-lations implying that mtDNA alone in such cases overesti-mates population structuring

Conclusions

Bears are a prominent and widely cited example in phylogeo-graphy with range-wide signals of pronounced populationstructuring reported for brown bear mtDNA (Davison et al2011) We reexamined this paradigm using paternally in-herited markers In strong contrast to mtDNA data shallowdivergences and lack of pronounced geographic structuring ofbrown bear Y chromosomes were found mtDNA-basedinferences have thus overestimated phylogeographic struc-turing due to extensive male gene flow on regional andrange-wide scales Nevertheless various adaptive traits havebeen linked to mtDNA (Ballard and Rand 2005) and themtDNA of an individual may have important consequencesfor its phenotype and local adaptation Phylogeographicstructuring of the brown bear matriline into regional assem-blages could therefore be adaptively significant Our findingshighlight that evolutionary patterns inferred from mtDNAdespite its popularity are not representative of the entiregenome and that phylogeographic histories of many speciesmay need to be reevaluated Y-chromosomal data are essen-tial in any phylogeographic analyses of mammalsmdasheven inpresumably well-studied species such as bears

Materials and Methods

Identification of Y-Chromosomal Markers

A whole genome sequence assembly of a male polar bear (Liet al 2011) was used to identify putative Y-chromosomalscaffolds by searching for matches with the sequences ofknown Y-linked genes (SMCY ZFY SRY UBEY RMBY) Weidentified five scaffolds from ~19 to ~390 kb in length (scaf-fold numbers 297 318 369 579 605) These scaffolds wereextracted and compared with the corresponding sequencesin a male brown bear (accession numbers CBZK010000001ndash

1360

Bidon et al doi101093molbevmsu109 MBE by guest on M

ay 27 2014httpm

beoxfordjournalsorgD

ownloaded from

CBZK010000005) in order to identify genomic regionscontaining either variable sites or microsatellite motifsrespectively between the two individuals To decrease thepossible ascertainment bias in the subsequent applicationof the markers in samples from different species and popu-lations we did not type these variable sites but we designedand sequenced 11 polymerase chain reaction (PCR) frag-ments around them with lengths of at least 500 bp (529ndash1216 bp) All variable sites on the black bear branch andmost variable sites within brown and polar bears respectivelywere newly discovered by this sequencing approach (supple-mentary table S6 Supplementary Material online) All butthree variable sites between brown and polar bears howeverwere known from the ascertainment panel Y-chromosomalsequences for each haplotype can be accessed at the EMBLdata archive (accession numbers HG423284ndashHG423309)The scaffold sequences were then mined for di- and tetranu-cleotide microsatellites that exhibited at least five uninter-rupted repeat units Primers for nine microsatellite markersare shown in supplementary table S9 SupplementaryMaterial online Allele size data can be accessed at theDRYAD repository (httpdoiorg105061dryad3p21q)

PCR fragments obtained from brown polar and blackbears were then evaluated for their male specificity This as-sessment resulted in seven sequence fragments and nine mi-crosatellite markers that were ultimately used(supplementary tables S1 and S6 Supplementary Materialonline) Male specificity was ensured throughout all experi-ments by consistently including female DNA controls Seesupplementary tables S7ndashS9 Supplementary Material onlinefor details on PCR conditions sequencing and fragmentanalysis

Sampling and DNA Extraction

Tissue and DNA samples from 90 male brown and 40 malepolar bears were included in this study covering large parts oftheir distribution ranges (fig 1 table 1 and supplementarytable S1 Supplementary Material online) For comparison wealso analyzed four American black bear samples coveringtheir two previously described mitochondrial clades (supple-mentary fig S4 Supplementary Material online) and a malespectacled bear as outgroup for divergence time estimationsAll tissue samples originated from animals legally hunted forpurposes other than this study or from zoo individualsIndividuals with unknown sex were tested as in Bidon et al(2013) DNA was extracted using a modified Puregene(Qiagen Hilden Germany) DNA salt extraction protocol orDNeasy Tissue kit (Qiagen)

Analysis of Y-Chromosomal Scaffold Sequences

Genomic sequence data was used from 12 male polar bears1 male brown bear and 1 male black bear (Miller et al 2012)plus 1 male brown bear from Northern Europe (supplemen-tary table S3 Supplementary Material online) Short readswere mapped to a gt390-kb-long putative Y-linked scaffoldfrom a male polar bear (Li et al 2011) (scaffold 297)Consensus sequences were determined for every individual

using Geneious 616 (Biomatters Auckland New Zealand)calling ldquordquo for regions without coverage and ldquoNrdquo for bases witha Phred quality score lt20 Consensus sequences of the 15individuals were aligned and single-nucleotide variants deter-mined in regions with coverage for all individuals All variantswere manually checked in the alignment and we excluded allsites that contained insertionsdeletions or ambiguous basesAdditionally variants within 5 nt of ambiguous sites ( and Nrespectively) variants directly adjacent to each other andvariants in microsatellite regions were excluded in order toaccount for sequencing and alignment errors

Data Analysis

PCR products were sequenced or subjected to fragment anal-ysis (microsatellites) Sequences were aligned and edited inGeneious 562 (Biomatters Auckland New Zealand) andallele sizes were determined using Genemapper 40(Applied Biosystems Life Technologies GmbH DarmstadtGermany) To infer phylogenetic relationships among haplo-types networks were estimated using statistical parsimony asimplemented in TCS 121 (Clement et al 2000) with theconnection limit set to 095 for sequence data or fixed at50 steps for microsatellite haplotypes For the combined anal-ysis of sequence and allele size polymorphisms data from allY-linked markers were combined into one compound haplo-type per individual A haplotype distance matrix was calcu-lated from allele sizes with GenoDive 20b23 (Meirmans andVan Tienderen 2004) assuming a strictly stepwise mutationmodel with single repeat unit changes counted as one mu-tational step Analyses of polymorphic sites and other sum-mary statistics nucleotide diversity p tests for signals ofdemography and selection (Tajima 1989 Fu and Li 1993 Fu1997) and analysis of molecular variance (AMOVA) weredone in DnaSP v510 (Librado and Rozas 2009) andArlequin 35 (Excoffier and Lischer 2010) Haplotype configu-ration tests were performed in haploconfig and haplofreq(Innan et al 2005) with theta values obtained from thenumber of segregating sites (Wattersonrsquos theta) and nucleo-tide diversity (p) respectively and simulating different pop-ulation expansion scenarios ( = 138 037 growth rate g = 02 10 a = 10000 n = 44 s = 6) Different weighting schemeswere applied to sequence and microsatellite markers as inBrown et al (2011) Estimates of mean (plusmnSE) among-groupdistances were obtained in MEGA5 (Tamura et al 2011)SplitsTree4 (Huson and Bryant 2006) was used to calculatea NeighborNet network for the 390-kb-long data set Bayesianphylogenetic analyses and divergence time estimations wereperformed in Beast v174 (Drummond et al 2012)

Supplementary MaterialSupplementary material figures S1ndashS4 and tables S1ndashS9 areavailable at Molecular Biology and Evolution online (httpwwwmbeoxfordjournalsorg)

Acknowledgments

The authors thank N Schreck D Herbert and C Tobiassenfor assistance U Arnason M Balint EW Born C Nowak

1361

Male-Biased Gene Flow in Bears doi101093molbevmsu109 MBE by guest on M

ay 27 2014httpm

beoxfordjournalsorgD

ownloaded from

M Onucsan K Skırnisson and F Zachos for providingsamples and the editor and three anonymous reviewersfor insightful comments This work was supported byHessersquos ldquoLOEWE Landes-Offensive zur EntwicklungWissenschaftlich-okonomischer Exzellenzrdquo by the Arthurund Aenne Feindt-Stiftung the Estonian Research Council(IUT-2032 ESF-8525) and the European Union through theEuropean Regional Development Fund (Centre of ExcellenceFIBIR) Jon Baldur Hlıberg kindly provided the bear paintingsThe findings and conclusions in this article are those of theauthor(s) and do not necessarily represent the views of theUS Fish and Wildlife Service

ReferencesAvise JC 2000 Phylogeography the history and formation of species

Cambridge (MA) Harvard University PressBachtrog D Kirkpatrick M Mank JE McDaniel SF Pires JC Rice W

Valenzuela N 2011 Are all sex chromosomes created equalTrends Genet 27350ndash357

Ballard JWO Rand DM 2005 The population biology of mitochondrialDNA and its phylogenetic implications Annu Rev Ecol Evol Syst 36621ndash642

Bidon T Frosch C Eiken HG Kutschera VE Hagen SB Aarnes SG FainSR Janke A Hailer F 2013 A sensitive and specific multiplex PCRapproach for sex identification of ursine and tremarctine bears suit-able for non-invasive samples Mol Ecol Resour 13362ndash368

Boissinot S Boursot P 1997 Discordant phylogeographic patterns be-tween the Y chromosome and mitochondrial DNA in the housemouse selection on the Y chromosome Genetics 1461019ndash1034

Brown SK Pedersen NC Jafarishorijeh S Bannasch DL Ahrens KD WuJ-T Okon M Sacks BN 2011 Phylogenetic distinctiveness of MiddleEastern and Southeast Asian village dog Y chromosomes illuminatesdog origins PLoS One 6e28496

Brumfield RT Beerli P Nickerson DA Edwards SV 2003 The utility ofsingle nucleotide polymorphisms in inferences of population historyTrends Ecol Evol 18249ndash256

Cahill JA Green RE Fulton TL Stiller M Jay F Ovsyanikov N SalamzadeR John J Stirling I Slatkin M et al 2013 Genomic evidence for islandpopulation conversion resolves conflicting theories of polar bearevolution PLoS Genet 9e1003345

Campagna L Van Coeverden de Groot PJ Saunders BL Atkinson SNWeber DS Dyck MG Boag PT Lougheed SC 2013 Extensive sam-pling of polar bears (Ursus maritimus) in the Northwest Passage(Canadian Arctic Archipelago) reveals population differentiationacross multiple spatial and temporal scales Ecol Evol 33152ndash3165

Chan Y-C Roos C Inoue-Murayama M Inoue E Shih C-C Vigilant L2012 A comparative analysis of Y chromosome and mtDNA phy-logenies of the Hylobates gibbons BMC Evol Biol 12150

Charlesworth B Charlesworth D 2000 The degeneration of Y chromo-somes Philos Trans R Soc Lond B Biol Sci 3551563ndash1572

Chesser RK Baker RJ 1996 Effective sizes and dynamics of uniparentallyand diparentally inherited genes Genetics 1441225ndash1235

Clement M Posada D Crandall KA 2000 TCS a computer program toestimate gene genealogies Mol Ecol 91657ndash1660

Cronin MA Amstrup SC Garner GW 1991 Interspecific and intraspe-cific miochondrial DNA variation in North American bears (Ursus)Can J Zool 692985ndash2992

Cronin MA Amstrup SC Scribner KT 2006 Microsatellite DNA andmitochondrial DNA variation in polar bears (Ursus maritimus) fromthe Beaufort and Chukchi seas Alaska Can J Zool 660655ndash660

Cronin MA MacNeil MD 2012 Genetic relationships of extant brownbears (Ursus arctos) and polar bears (Ursus maritimus) J Hered 103873ndash881

Cronin MA McDonough MM Huynh HM Baker RJ 2013 Geneticrelationships of North American bears (Ursus) inferred from

amplified fragment length polymorphisms and mitochondrialDNA sequences Can J Zool 91626ndash634

Davison J Ho SYW Bray SC Korsten M Tammeleht E Hindrikson MOslashstbye K Oslashstbye E Lauritzen S-E Austin J et al 2011 Late-Quaternary biogeographic scenarios for the brown bear (Ursusarctos) a wild mammal model species Quat Sci Rev 30418ndash430

Drummond AJ Suchard MA Xie D Rambaut A 2012 Bayesian phylo-genetics with BEAUti and the BEAST 17 Mol Biol Evol 291969ndash1973

Edwards CJ Suchard MA Lemey P Welch JJ Barnes I Fulton TL BarnettR OrsquoConell TC Coxon P Monaghan N et al 2011 Ancient hybrid-ization and an Irish origin for the modern polar bear matriline CurrBiol 211251ndash1258

Excoffier L Lischer HEL 2010 Arlequin suite ver 35 a new series ofprograms to perform population genetics analyses under Linux andWindows Mol Ecol Resour 10564ndash567

Fu Y-X 1997 Statistical tests of neutrality of mutations against popula-tion growth hitchhiking and background selection Genetics 147915ndash925

Fu Y-X Li W-H 1993 Statistical tests of neutrality of mutations Genetics133693ndash709

Geraldes A Carneiro M Delibes-Mateos M Villafuerte R Nachman MWFerrand N 2008 Reduced introgression of the Y chromosome be-tween subspecies of the European rabbit (Oryctolagus cuniculus) inthe Iberian Peninsula Mol Ecol 174489ndash4499

Greminger MP Krutzen M Schelling C Pienkowska-Schelling AWandeler P 2010 The quest for Y-chromosomal markers - meth-odological strategies for mammalian non-model organisms Mol EcolResour 10409ndash420

Hailer F Kutschera VE Hallstrom BM Fain SR Leonard JA Arnason UJanke A 2013 Response to comment on ldquoNuclear genomic se-quences reveal that polar bears are an old and distinct bear lineagerdquoScience 3391522ndash1522

Hailer F Kutschera VE Hallstrom BM Klassert D Fain SR Leonard JAArnason U Janke A 2012 Nuclear genomic sequences revealthat polar bears are an old and distinct bear lineage Science 336344ndash347

Hellborg L Ellegren H 2004 Low levels of nucleotide diversity in mam-malian Y chromosomes Mol Biol Evol 21158ndash163

Hellborg L Gunduz I Jaarola M 2005 Analysis of sex-linked se-quences supports a new mammal species in Europe Mol Ecol 142025ndash2031

Hewitt G 2000 The genetic legacy of the Quaternary ice ages Nature405907ndash913

Hirata D Mano T Abramov AV Baryshnikov GF Kosintsev PS VorobievAA Raichev EG Tsunoda H Kaneko Y Murata K et al 2013Molecular phylogeography of the brown bear (Ursus arctos) inNortheastern Asia based on analyses of complete mitochondrialDNA sequences Mol Biol Evol 301644ndash1652

Ho SYW Saarma U Barnett R Haile J Shapiro B 2008 The effect ofinappropriate calibration three case studies in molecular ecologyPLoS One 3e1615

Hughes JF Rozen S 2012 Genomics and genetics of human and primateY chromosomes Annu Rev Genomics Hum Genet 1383ndash108

Huson DH Bryant D 2006 Application of phylogenetic networks inevolutionary studies Mol Biol Evol 23254ndash267

Ingman M Gyllensten U 2001 Analysis of the complete human mtDNAgenome methodology and inferences for human evolution J Hered92454ndash461

Innan H Zhang K Marjoram P Tavare S Rosenberg NA 2005 Statisticaltests of the coalescent model based on the haplotype frequencydistribution and the number of segregating sites Genetics 1691763ndash1777

Keis M Remm J Ho SYW Davison J Tammeleht E Tumanov IL SaveljevAP Mannil P Kojola I Abramov AV et al 2013 Completemitochondrial genomes and a novel spatial genetic methodreveal cryptic phylogeographical structure and migration patternsamong brown bears in north-western Eurasia J Biogeogr 40915ndash927

1362

Bidon et al doi101093molbevmsu109 MBE by guest on M

ay 27 2014httpm

beoxfordjournalsorgD

ownloaded from

Kohn M Knauer F Stoffella A Schroder W Paabo S 1995 Conservationgenetics o the European brown bearmdasha study using excrementalPCR of nuclear and mitochondrial sequences Mol Ecol 495ndash103

Kopatz A Eiken HG Hagen SB Ruokonen M Esparza-Salas R Schregel JKojola I Smith ME Wartiainen I Aspholm PE et al 2012Connectivity and population subdivision at the fringe of a largebrown bear (Ursus arctos) population in North Western EuropeConserv Genet 13681ndash692

Korsten M Ho SYW Davison J Pahn B Vulla E Roht M Tumanov ILKojola I Andersone-Lilley Z Ozolins J et al 2009 Sudden expansionof a single brown bear maternal lineage across northern continentalEurasia after the last ice age a general demographic model formammals Mol Ecol 181963ndash1979

Lawson Handley LJ Berset-Brandli L Perrin N 2006 Disentangling rea-sons for low Y chromosome variation in the greater white-toothedshrew (Crocidura russula) Genetics 173935ndash942

Li B Zhang G Willerslev E Wang J 2011 Genomic data from the PolarBear (Ursus maritimus) Gigascience [cited 2014 Mar 7] Availablefrom httpdxdoiorg105524100008

Librado P Rozas J 2009 DnaSP v5 a software for comprehensive analysisof DNA polymorphism data Bioinformatics 251451ndash1452

Lindqvist C Schuster SC Sun Y Talbot SL Qi J Ratan A Tomsho LPKasson L Zeyl E Aars J et al 2010 Complete mitochondrial genomeof a Pleistocene jawbone unveils the origin of polar bear Proc NatlAcad Sci U S A 1075053ndash5057

Lippold S Knapp M Kuznetsova T Leonard JA Benecke N Ludwig ARasmussen M Cooper A Weinstock J Willerslev E et al 2011Discovery of lost diversity of paternal horse lineages using ancientDNA Nat Commun 2450

Luo S-J Johnson WE David VA Menotti-Raymon M Stanyon R Cai QXBeck T Yuhki N Pecon-Slattery J Smith JLD et al 2007Development of Y chromosome intraspecific polymorphic markersin the Felidae J Hered 98400ndash413

McLellan BN Hovey FW 2001 Natal dispersal of grizzly bears Can J Zool79838ndash844

Meadows JRS Hanotte O Drogemuller C Calvo J Godfrey R Coltman DMaddox JF Marzanov N Kantanen J Kijas JW 2006 Globally dis-persed Y chromosomal haplotypes in wild and domestic sheepAnim Genet 37444ndash453

Meirmans PG Van Tienderen PH 2004 GENOTYPE and GENODIVEtwo programs for the analysis of genetic diversity of asexual organ-isms Mol Ecol Notes 4792ndash794

Miller W Schuster SC Welch AJ Ratan A Bedoya-Reina OC Zhao FKim HL Burhans RC Drautz DI Wittekindt NE et al 2012 Polar andbrown bear genomes reveal ancient admixture and demographicfootprints of past climate change Proc Natl Acad Sci U S A 109E2382ndashE2390

Nakagome S Pecon-Slattery J Masuda R 2008 Unequal rates of Ychromosome gene divergence during speciation of the familyUrsidae Mol Biol Evol 251344ndash1356

Paetkau D Amstrup SC Born EW Calvert W Derocher AE Garner GWMessier F Stirling I Taylor MK Wiig Oslash et al 1999 Genetic structureof the worldrsquos polar bear populations Mol Ecol 81571ndash1584

Paetkau D Shields GF Strobeck C 1998 Gene flow between insularcoastal and interior populations of brown bears in Alaska Mol Ecol71283ndash1292

Paetkau D Waits LP Clarkson PL Craighead L Strobeck C 1997 AnEmpirical Evaluation of Genetic Distance Statistics UsingMicrosatellite Data From Bear (Ursidae) Populations Genetics 1471943ndash1957

Perez T Hammer SE Albornoz J Domınguez A 2011 Y-chromosomephylogeny in the evolutionary net of chamois (genus Rupicapra)BMC Evol Biol 11272

Petit E Balloux F Excoffier L 2002 Mammalian population genetics whynot Y Trends Ecol Evol 1728ndash33

Petit RJ Excoffier L 2009 Gene flow and species delimitation TrendsEcol Evol 24386ndash393

Pidancier N Jordan S Luikart G Taberlet P 2006 Evolutionary history ofthe genus Capra (Mammalia Artiodactyla) discordance betweenmitochondrial DNA and Y-chromosome phylogenies MolPhylogenet Evol 40739ndash749

Purvis A 2005 Phylogeny and conservation Cambridge CambridgeUniversity Press

Pusey A 1987 Sex-biased dispersal and inbreeding avoidance in birdsand mammals Trends Ecol Evol 2295ndash299

Roca AL Georgiadis N OrsquoBrien SJ 2005 Cytonuclear genomic dissocia-tion in African elephant species Nat Genet 3796ndash100

Sacks BN Brown SK Stephens D Pedersen NC Wu J-T Berry O 2013 Ychromosome analysis of dingoes and southeast asian village dogssuggests a neolithic continental expansion from Southeast Asia fol-lowed by multiple austronesian dispersals Mol Biol Evol 301103ndash1118

Subramanian S Denver DR Millar CD Heupink T Aschrafi A Emslie SDBaroni C Lambert DM 2009 High mitogenomic evolutionary ratesand time dependency Trends Genet 25482ndash486

Taberlet P Bouvet J 1994 Mitochondrial DNA polymorphism phylo-geography and conservation genetics of the brown bear Ursusarctos in Europe Proc R Soc Lond B Biol Sci 255195ndash200

Taberlet P Fumagalli L Wust-Saucy A Cosson J 1998 Comparativephylogeography and postglacial colonization routes in EuropeMol Ecol 7453ndash464

Tajima F 1989 Statistical method for testing the neutral mutation hy-pothesis by DNA polymorphism Genetics 123585ndash595

Tammeleht E Remm J Korsten M Davison J Tumanov I Saveljev AMannil P Kojola I Saarma U 2010 Genetic structure in large con-tinuous mammal populations the example of brown bears in north-western Eurasia Mol Ecol 195359ndash5370

Tamura K Peterson D Peterson N Stecher G Nei M Kumar S 2011MEGA5 molecular evolutionary genetics analysis using maximumlikelihood evolutionary distance and maximum parsimony meth-ods Mol Biol Evol 282731ndash2739

Waits L Taberlet P Swenson JE Sandegren F Franzen R 2000Nuclear DNA microsatellite analysis of genetic diversity and geneflow in the Scandinavian brown bear (Ursus arctos) Mol Ecol 9421ndash431

Wayne RK Van Valkenburgh B OrsquoBrien SJ 1991 Molecular distanceand divergence time in carnivores and primates Mol Biol Evol 8297ndash319

Wei W Ayub Q Chen Y McCarthy S Hou Y Carbone I Xue Y Tyler-Smith C 2013 A calibrated human Y-chromosomal phylogenybased on resequencing Genome Res 23388ndash395

Willard HF 2003 Tales of the Y chromosome Nature 423810ndash813Wilson Sayres MA Lohmueller KE Nielsen R 2014 Natural selection

reduced diversity on human y chromosomes PLoS Genet 10e1004064

Xue Y Wang Q Long Q Ng BL Swerdlow H Burton J Skuce C Taylor RAbdellah Z Zhao Y et al 2009 Human Y chromosome base-sub-stitution mutation rate measured by direct sequencing in a deep-rooting pedigree Curr Biol 191453ndash1457

Zedrosser A Stoslashen O-G Saeligboslash S Swenson JE 2007 Should I stay orshould I go Natal dispersal in the brown bear Anim Behav 74369ndash376

Zerjal T Xue Y Bertorelle G Wells RS Bao W Zhu S Qamar R Ayub QMohyuddin A Fu S et al 2003 The genetic legacy of the MongolsAm J Hum Genet 72717ndash721

1363

Male-Biased Gene Flow in Bears doi101093molbevmsu109 MBE by guest on M

ay 27 2014httpm

beoxfordjournalsorgD

ownloaded from