Extended survival of Pleistocene Siberian wolves into the early 20th century on the island of Honshū

•Generated 3.7✕ nuclear genome of the extinct Honshu wolf•The Honshū wolf belonged to lineage Siberian Pleistocene wolves•There was gene flow between wolves and Japanese dogs The or one most distinct gray subspecies due its small stature endemicity islands Honshū, Shikoku, Kyūshū. Long revered as a guardian farmers travellers, it persecuted from 17th century following rabies epidemic, which led extinction in early 20th century. To better understand evolutionary history, we sequenced 19th specimen an average depth coverage 3.7✕. We find were closely related that previously believed have gone Late Pleistocene, thereby extending survival this ancient until also detected significant wolf, corroborating previous reports on dog interbreeding. Researchers long debated origin present-day Eurasian wolves, well their relationships each other. While is now clear analyses based genome-scale data sets they are reciprocally monophyletic sister clades thus not derived any yet identified extant (Freedman et al., 2014Freedman A.H. Gronau I. Schweizer R.M. Ortega-Del Vecchyo D. Han E. Silva P.M. Galaverni M. Fan Z. Marx P. Lorente-Galdos B. al.Genome sequencing highlights dynamic history dogs.PLoS Genet. 2014; 10: e1004016Crossref PubMed Scopus (381) Google Scholar), what population domesticated, ancestral homeland common ancestors, remains (Larson Bradley, 2014Larson G. Bradley D.G. How much years? advent canine genomics.PLoS e1004093Crossref (35) Scholar). A further open question relationship both lineages other groups. One key example enigmatic Beringian may specialized hunting megafauna (Leonard 2007Leonard J.A. Vilà C. Fox-Dobbs K. Koch P.L. Wayne R.K. Van Valkenburgh Megafaunal extinctions disappearance ecomorph.Curr. Biol. 2007; 17: 1146-1150Abstract Full Text PDF (164) several whose genomes recently been subfossil materials, found represent fall outside modern (Ramos-Madrigal 2020Ramos-Madrigal J. Sinding M.H.S. Carøe Mak S.S. Niemann Castruita J.A.S. Fedorov S. Kandyba A. Germonpré Bocherens H. al.Genomes Wolves Uncover Multiple Extinct Wolf Lineages.Curr. 2020; (In press)Abstract (16) Scholar, Skoglund 2015Skoglund Ersmark Palkopoulou Dalén L. Ancient reveals divergence domestic ancestors admixture into high-latitude breeds.Curr. 2015; 25: 1515-1519Abstract (188) Indeed, studies mitochondrial (Koblmüller 2016Koblmüller Dabad Ramirez O. Marques-Bonet T. Leonard Whole illuminate intercontinental dispersals grey (Canis lupus).J. Biogeogr. 2016; 43: 1728-1738Crossref (46) Scholar; Loog 2020Loog Thalmann Schuenemann V.J. Perri Witt K.E. Samaniego Velasco M.S. Lundstrøm I.K. DNA suggests trace expansion Beringia.Mol. Ecol. 29: 1596-1610Crossref (47) Scholar) (Loog recovered suggest lost replaced by around Pleistocene-Holocene transition Siberia - possibly associated with upon depended curious observation such lack evidence for these breeds (with exception some Asian Arctic sled (Sinding 2020Sinding Gopalakrishnan Ramos-Madrigal de Manuel Pitulko V.V. Kuderna Feuerborn T.R. Frantz L.A. Vieira F.G. al.Arctic-adapted emerged at Pleistocene–Holocene transition.Science. 368: 1495-1499Crossref (42) This particularly striking given widespread coexisting species within crown Canids (Gopalakrishnan 2018Gopalakrishnan M.-H.S. Montero M.de Serres al.Interspecific shaped evolution genus Canis.Curr. 2018; 28: 3441-3449Abstract (64) must physically isolated during Last Glacial Maximum (LGM), prior replacing them after 14,000 years ago likely Holocene turn raises obvious, currently unanswered, question, where isolation could occured? archipelago potential candidate LGM refugium dogs, land bridges Korean peninsula Japan's largest island, formed beginning (Ohshima, 1990Ohshima straits Islands late-Quaternary.Quat. Res. (Daiyonki-kenkyu). 1990; 193-208Crossref (147) Hokkaidō, second northernmost island Japan, connected continent periods low sea level, occurred instance Until century, Japan inhabited two highly phenotypically endemic subspecies: lupus hodophilax), Ezo hattai). be Kyūshū, habitat restricted Hokkaidō Sakhalin (Figure 1) (Ishiguro 2009Ishiguro N. Inoshima Y. Shigehara Mitochondrial analysis hodophilax Temminck, 1839) comparison representative haplotypes.Zoolog. Sci. 2009; 26: 765-770Crossref (18) among smallest world appreciated medieval killing crop-destroying wildlife (Fritts 2003Fritts S.H. Stephenson R.O. Hayes R.D. Boitani humans.in: David Mech Luigi Wolves: Behavior, Ecology, Conservation. University Chicago Press, 2003: 289-316Google epidemic caused increase attacks, setting human persecution motion, culminated 1905 (Walker, 2009Walker B.L. Lost Japan. Washington 2009Google deep Tsugaru strait major zoogeographical barrier islands, known Blakiston's Line (Dobson, 1994Dobson Patterns distribution mammals.Mammal Rev. 1994; 24: 91-111Crossref (70) As result, fauna snow macaques (Macaca fuscata) black bears (Ursus thibetanus), has similarities Southeastern Asia, while includes Ussuri brown bear arctos lasiotus), resembles biological diversity Northeastern Asia. consequence barrier, there no overlap habitats colonized Siberia, respectively. exact phylogenetic placement speculative, apart osteological comparisons noting morphological differences (Imaizumi, 1970Imaizumi ニホンオオカミの系統的地位について.J. Mammalogical Soc. Jpn. 1970; 5: 27-32Google Ishiguro 2010Ishiguro Ichikawa Kato Osteological genetic hattai) hokkaido Japan.Zoolog. 2010; 27: 320-324Crossref (9) only (Matsumura 2014Matsumura Reconstructing colonization genome.Mol. Phylogenet. Evol. 80: 105-112Crossref (21) Koblmüller These basal all North American clade. however marker, does allow quantification admixture, especially interest populations link wolves. subspecies, reassess test hypothesis resequenced sample provided Natural History Museum, London, coverage. had shot wild 1800s Chichibu District, Kotsuki, Northwest Tokyo, confirmed authenticity misincorporation read length patterns S1) determined chromosomal sex male S2). First, investigated historic whole-genome using NGSadmix 2A, see Figure S3). indistinguishable profiles panel (see Table S1), across tested numbers assumed populations. In contrast populations, contributed substantially genome. Irrespective number ancestry clusters inferred analysis, always derives majority same cluster S4). explore landscape used D-statistics formally provide support excess allele sharing Greenland Chinese 2B). already observed shared so investigate might more genetically similar than created scatterplot H3, Portuguese H1, H2 results symmetrically share alleles do wolf. explanation substantial East (Fan 2016Fan Wang Armero A.S. Pollinger Ortega Del-Vecchyo al.Worldwide genomic variation wolves.Genome 163-173Crossref (125) All individuals included significantly having closest affinity Additionally, qpWave showed can explained single migration stream equally needs lineage. therefore hypothesize our individual admixed introgression (Skoglund likewise shown contribution order robustly identify structure samples, haplotype-aware clustering tool fineSTRUCTURE (Lawson 2012Lawson D.J. Hellenthal Myers Falush Inference dense haplotype data.PLoS 2012; 8: e1002453Crossref (622) Tables S2 S3), low-coverage (Martiniano 2016Martiniano R. Caffell Holst Hunter-Mann Montgomery Müldner McLaughlin R.L. Teasdale M.D. van Rheenen W. Veldink J.H. al.Genomic signals continuity Britain before Anglo-Saxons.Nat. Commun. 7: 10326Crossref (61) Martiniano 2017Martiniano Cassidy L.M. Ó’Maoldúin N.M. Manco Fidalgo Pereira Coelho M.J. Serra al.The genomics archaeological west Iberia: investigation substructure imputation haplotype-based methods.PLoS 2017; 13: e1006852Crossref dendrogram similarity matrix, positioned clade three – Tumat, Yana, Bunge-Toll, earlier findings 3A, S5). verify performed unsupervised dimension reduction principal component (PCA). incline together along first (PC1). Among placed components 3B, S6). examine eight putatively populations: chromosomes subset then painted best fitting haplotypes remaining individuals. resulting chromosome paintings input GLOBETROTTER (Hellenthal 2014Hellenthal Busby G.B.J. Band Wilson J.F. Capelli atlas history.Science. 343: 747-751Crossref (467) uses information describe date events involving pre-defined (surrogate populations) leading (target population). requires multiple target infer dates, unable use population. Instead, chose run potentially detect local Using surrogate estimated described mixture 93% 7% scenario event, occurring approximately 25 generations ago, 9% 91% 100% dog. preliminary indicate genomes, large confounds results. Further larger sizes needed positively determine breeds. Finally, Markov chain Monte Carlo algorithm implemented SOURCEFIND, modeled analysis—Japanese wolves—as seven i.e. except being modeled. painting split 100 subsections, subsection assigned counterpart method, partitioned 52% 47% 1% Furthermore, 15% Japanese/Korean breed but 3C). above, inference ill-defined computationally challenging, inclusion necessary obtain statistically sound estimates That said, study documented show insubstantial deem unlikely ancestor predate LGM. However, made unexpected discovery sampled hybrid dogs. now, thought Holocene, strong hodophilax) rather thousands years, descendants went about ago. last kind centuries persecution, resulted drastic decline extent lower actively hunted. It sequence analyze additional specimens, those decline, accurate representation makeup high proportion variants hinders ability quantify remain candidates refugia, covered southern Analyzing understudied resolve mystery absent specimen, nature limits information, requests, inquiries should directed will fulfilled Lead Contact, Jonas ([email protected]). did generate new materials. project accession reported paper ENA: PRJEB41490. ERS5374233. methods accompanying Transparent supplemental file. research funded European Union's Horizon 2020 innovation programme under grant agreement no. 676154 (ArchSci2020). authors thank Danish National High-throughput Sequencing Center assistance generating data. London Museum curators Louise Tomsett Richard Sabin assisting collection. M.T.P.G, M.-H.S.S. conceived study. lab work. J.N, S.G. bioinformatic analysis. J.R.-M. computation expertise. J.N., M.-H.S.S., S.G., N.W., M.T.P.G. supervised J.R.-M., N.Y., interpreted wrote manuscript authors. approved manuscript. declare conflict interest. Download .pdf (1.41 MB) Help pdf files Document S1. methods, Figures S1–S6, S1–S3 .xlsx (.02 xlsx Data List published unpublished canid reference 2 3 (.01 S2. Population assignments finestructure clustering,