www.nature.com/hdy ARTICLE OPEN Spatial genetic structure of European wild boar, with inferences on late-Pleistocene and Holocene demographic history ✉ Joost F. de Jong 1, Laura Iacolina 2,3 , Herbert H. T. Prins 4, Pim van Hooft1, Richard P. M. A. Crooijmans5, Sip E. van Wieren1,23, Joaquin Vicente Baños , Eric Baubet , Seán Cahill8, Eduardo Ferreira 9, Carlos Fonseca9, Peter M. Glazov 10, Ida Jelenko Turinek11, 6 7 Victor M. Lizana Martín 12,13, András Náhlik 14,15, Boštjan Pokorny16,17, Tomasz Podgórski18,19, Nikica Šprem20, Rauno Veeroja21, Ronald C. Ydenberg22 and Hendrik-Jan Megens5 © The Author(s) 2023 European wildlife has been subjected to intensifying levels of anthropogenic impact throughout the Holocene, yet the main genetic partitioning of many species is thought to still reflect the late-Pleistocene glacial refugia. We analyzed 26,342 nuclear SNPs of 464 wild boar (Sus scrofa) across the European continent to infer demographic history and reassess the genetic consequences of natural and anthropogenic forces. We found that population fragmentation, inbreeding and recent hybridization with domestic pigs have caused the spatial genetic structure to be heterogeneous at the local scale. Underlying local anthropogenic signatures, we found a deep genetic structure in the form of an arch-shaped cline extending from the Dinaric Alps, via Southeastern Europe 1234567890();,: and the Baltic states, to Western Europe and, finally, to the genetically diverged Iberian peninsula. These findings indicate that, despite considerable anthropogenic influence, the deeper, natural continental structure is still intact. Regarding the glacial refugia, our findings show a weaker signal than generally assumed, but are nevertheless suggestive of two main recolonization routes, with important roles for Southern France and the Balkans. Our results highlight the importance of applying genomic resources and framing genetic results within a species’ demographic history and geographic distribution for a better understanding of the complex mixture of underlying processes. Heredity (2023) 130:135–144; https://doi.org/10.1038/s41437-022-00587-1 INTRODUCTION the ranges and population sizes of wildlife across Europe In human-dominated landscapes, the genetic variation of wildlife (Apollonio et al. 2010), whereas changes in land use and increased is shaped not only by natural demographic forces, but also by presence of infrastructures and barriers to animal movement led anthropogenic factors. In Europe, the Last Glacial Maximum (LGM), to fragmentation and loss of connectivity (Apollonio et al. 2010; when ice sheets reached their most recent maximum extent ~27k- Deinet et al. 2013; Koemle et al. 2018). On top of this, another 19k years ago (Clark et al. 2009), is considered a major anthropogenic impact was asserted via translocation and hybri- demographic force, as it restricted most wildlife populations to dization events with domestic relatives or introduced (sub)species, southern refugia for several thousand years and left genetic (e.g., Cervus spp. and Capreolus spp.; Putman et al. 2011; Iacolina signatures that are still detectable today (Hewitt 1999, 2004). et al. 2019; De Jong et al. 2020). Although shorter, a similar force was asserted by humans during One of the European mammals strongly affected by humans is the last few centuries, as overexploitation and eradication reduced the wild boar (Sus scrofa). Although currently widespread, 1 Wildlife Ecology & Conservation Group, Wageningen University, Droevendaalsesteeg 3A, 6708 PB Wageningen, The Netherlands. 2Department of Chemistry and Bioscience, Aalborg University, Frederik Bajers Vej 7H, 9220 Aalborg, Denmark. 3Faculty of Mathematics, Natural Sciences and Information Technologies, University of Primorska, Glagoljaška 8, SI-6000 Koper, Slovenia. 4Animal Sciences Group, Wageningen University, De Elst 1, 6708 WD Wageningen, The Netherlands. 5Animal Breeding and Genomics Group, Wageningen University, Droevendaalsesteeg 1, 6708 PD Wageningen, The Netherlands. 6SaBio, Instituto de Investigación en Recursos Cinegéticos -IREC (CSIC-UCLM-JCCM), Ronda de Toledo 12, 13071 Ciudad Real, Spain. 7Office Français pour la Biodiversité, Monfort 01330 Birieux, France. 8Estació Biològica de Can Balasc, Consorci del Parc Natural de la Serra de Collserola, Ctra. de l’Església 92, 08017 Barcelona, Spain. 9Department of Biology & CESAM - Centre for Environmental and Marine Studies, University of Aveiro, Portugal e de Aveiro, Portugal. 10Institute of Geography, Russian Academy of Sciences, Staromonetny per. 29, Moscow 119017, Russia. 11Ministry of Environment and Spatial Planning, Dunajska 48, 1000 Ljubljana, Slovenia. 12Servicio de Análisis, Investigación y Gestión de Animales Silvestres (SAIGAS), Facultad de Veterinaria, Universidad Cardenal Herrera-CEU, CEU Universities, C/Tirant lo Blanc 7, 46115, Alfara del Patriarca, Valencia, Spain. 13Wildlife Ecology & Health group (WE&H), Universitat Autònoma de Barcelona (UAB), Edifici V, Travessera del Turons, 08193 Bellaterra, Barcelona, Spain. 14University of Sopron, Bajcsy-Zsilinszky u. 4., H-9400 Sopron, Hungary. 15Sapientia Hungarian University of Transylvania, 400112 Cluj-Napocastr. Matei Corvin nr. 4Romania. 16Faculty of Environmental Protection, Trg mladosti 7, 3320 Velenje, Slovenia. 17Slovenian Forestry Institute, Večna pot 2, 1000 Ljubljana, Slovenia. 18Mammal Research Institute, Polish Academy of Sciences, Stoczek 1, 17-230 Białowieża, Poland. 19Department of Game Management and Wildlife Biology, Faculty of Forestry and Wood Sciences, Czech University of Life Sciences, Kamýcká 129, 165 00 Praha 6, Czech Republic. 20Department of Fisheries, Apiculture, Wildlife Management and Special Zoology, Faculty of Agriculture, University of Zagreb, Svetošimunska cesta 25, 10000 Zagreb, Croatia. 21Estonian Environment Agency, Rõõmu tee 6, 50705 Tartu, Estonia. 22Centre of Wildlife Ecology, Simon Fraser University, Burnaby, BC V5A 1S6, Canada. 23Deceased: Sip E. van Wieren. Associate editor: Xiangjiang Zhan. ✉email: lauraiacolina@gmail.com Received: 23 February 2021 Revised: 13 December 2022 Accepted: 14 December 2022 Published online: 13 January 2023 J.F. de Jong et al. 136 populations were decimated or eradicated until World War II structure, and (ii) how and to what extent the spatial genetic across large part of Europe and later supplemented with structure still holds signatures of the LGM. Given the historic and individuals from other regions (Apollonio et al. 2010). Additionally, contemporary intense human influence on European wild boar, the species has experienced genetic introgression from domestic we predicted to detect inbreeding, hybridization and transloca- pigs (S. s. domesticus) in many areas of its range, although the tion events in numerous wild boar populations. Depending on the degree of genetic ‘pollution’ varies greatly among locations intensity of these anthropogenic influences, we expected (Goedbloed et al. 2013a; Iacolina et al. 2018, 2019). Genetic alteration of the spatial continental-wide genetic structure studies have shown that, in the absence of barriers to the species, through weakening or erasing of natural signatures such as the observed hybridization patterns and genetic discontinuities Isolation by Distance, and—in case of geographic barriers – are likely to be caused by translocation events (Vernesi et al. 2003) Isolation by Resistance patterns. For the LGM specifically, we or local extinctions (Ferreira et al. 2009; Nikolov et al. 2009; expected to find genetic signatures reminiscent of refugia, Goedbloed et al. 2013b). However, these alterations of genetic recolonization areas and (a) suture zone(s). An overview of our variation were found to be minor compared to signatures caused hypotheses is presented in Table S1. by the LGM (Scandura et al. 2008, 2011a). The demographic history inferred from whole genome sequences shows that wild boar populations underwent a marked MATERIALS AND METHODS decline during the Late Pleistocene, reaching the lowest levels Sample collection and genotyping around the LGM (Groenen et al. 2012). In line with this, mismatch To obtain a good coverage of the European continent, we combined distributions of mitochondrial DNA (mtDNA) sequences of publicly available SNP data on European mainland wild boar (Iacolina European wild boar show signs of demographic expansion et al. 2016) with newly collected samples from previously under- represented regions, and with genetically distinct populations as expected to have occurred when climate became more favorable reference (Table S2). All samples were collected within the frameworks after the LGM (Scandura et al. 2008; Alexandri et al. 2012), of national game management and population control programs although strong demographic fluctuations were not detected in according to national laws. every country across Europe (Kusza et al. 2014). Additionally, For detection of pig introgression, we incorporated 140 pig samples, mtDNA haplotype diversity generally decreases with latitude which were collected within the framework of the PigBioDiv project (see (Vilaça et al. 2014), with higher levels in southern Italy and lower in Megens et al. 2008). These pig samples came from four standardized northwestern Europe. This pattern is indicative of the leading- domestic breeds (Large White, Landrace, Angler Sattle and Pietrain), three edge expansion model, in which most of the recolonization is Spanish (Negro Iberico, Manchado and Retinto) and three Italian undertaken by descendants of the northernmost populations of indigenous breeds (Calabrese, Casertana and Cinta Senese). refugia (Hewitt 1999; Nykänen et al. 2019). Moreover, each We isolated DNA following the Gentra Puregene Blood kit protocol (Qiagen, Venlo, the Netherlands). Samples were genotyped using the putative refugium has its own unique set of haplotypes, whilst PorcineSNP60 DNA Analysis Kit beadchip (https://emea.illumina.com/ most of the haplotypes observed in northern regions are shared products/by-type/microarray-kits/porcine-snp60.html). After excluding with one or several refugia (Scandura et al. 2008; Alves et al. 2010; duplicates and individuals with call rates below 0.95, we retained 464 Goedbloed et al. 2013b; Vilaça et al. 2014; Veličković et al. 2015). animals from 23 different countries in mainland Europe (Table S2). Lastly, eastern European wild boar, like pigs, have 2n = 38 Additionally, we had 15 wild boar from Israel and the Greek island of chromosomes, whereas due to Robertsonian fusion of chromo- Samos (Near East lineage) and 33 samples from the East-Asian lineage somes 15 and 17, western wild boar typically have 2n = 36 (Korea, Japan, eastern Russia and China) for comparison. Since two (Rejduch et al. 2003)—suggesting the presence of at least two versions of the PorcineSNP60 beadchip are available (v1 and v2), and some of the previously published data were genotyped with version v1, we different European clades. Nevertheless, it remains unclear what verified that genotypes of both beadchip versions aligned and were hence exactly the contribution of each refugium was to the recoloniza- compatible (see Fig. S1 for additional information) by means of a Principal tion of northern Europe and where and when wild boar Coordinate Analysis (PCoA) on pairwise genetic distances in Adegenet populations from different refugia met to form hybrid zones. 2.1.0 (Jombart and Ahmed 2011). Although the LGM has been recognized to have left the strongest signature on wild boar genetic composition (e.g., Scandura et al. 2008, 2011a), human manipulation throughout Quality control the Holocene and in particular the last two or three centuries have We focused our analyses on the 26,342 autosomal SNPs that occurred on both PorcineSNP60 beadchip v1 and v2, had call rates above 0.95 and also contributed to shaping the species’ genetic structure minor allele frequency (MAF) above 0.025. Finally, to adhere to the (Scandura et al. 2011b; Goedbloed et al. 2013b). While artificial assumption of neutrality, for all analyses except runs of homozygosity infrastructures appear to have limited influence on wild boar (ROH), we excluded intragenic SNPs (remaining: 15,296 SNPs). dispersal and, consequently, on its genetic structure (Frantz et al. We used PLINK (Purcell et al. 2007; Chang et al. 2015) to remove SNPs 2012; Mihalik et al. 2018), urban environments are becoming with strong dependency (LD) (function indep-pairwise, window of 50 SNPs, increasingly used by this species (Cahill et al. 2012; Stillfried et al. step size 5). Two filtering procedures were used, a more relaxed r2 < 0.5 for 2017). However, the main anthropogenic drivers affecting genetic ROH analyses (21,261 intra- and intergenic SNPs remaining) and a more diversity were identified as hunting, translocations and reintro- stringent r2 < 0.2 (9761 intergenic SNPs remaining) to investigate genetic ductions (Vernesi et al. 2003; Scandura et al. 2011a), together with variation and structure. Additionally, we used the same software to calculate, separately for each sampling location, relatedness among farming practices that led to domestic pig x wild boar hybridiza- individuals as an identity-by-descent score using subsets of SNPs with tion (Goedbloed et al. 2013a; Iacolina et al. 2018). Studies based on MAF > 0.10. For subsequent analyses, we retained only non-related both microsatellites (Vernesi et al. 2003; Scandura et al. individuals (identity-by-descent score <0.183). This filtering procedure 2008, 2011a; Ferreira et al. 2009) and SNPs (Goedbloed et al. led to a dataset of 330 wild boar from 56 sampling locations across 2013b; Iacolina et al. 2016) showed a connection between human- mainland Europe (minimum, median and maximum sample size per mediated wild boar movement and the observed genetic location: 1, 5 and 26, respectively). structure at local scale. We analyzed genome-wide SNP data of European wild boar to Hybridization and inbreeding (a) assess to what extent European wild boar populations are To estimate the intensity of introgression from pigs, we used the hybrid subject to pig hybridization, translocation and inbreeding, and to detection technique called ‘PCoA projection’ (see McVean 2009), where (b) delineate the continental wide spatial genetic structure of wild observed wild boar genotypes are projected on a polar axis that opposes boar. In so doing, we ultimately aimed to infer (i) to what extent pig genotypes and simulated ‘pure’ wild boar genotypes. Genotypes of anthropogenic influences have altered the natural spatial genetic pure wild boar were simulated through application of the R 3.4.2 (R Core Heredity (2023) 130:135 – 144 J.F. de Jong et al. 137 Team 2018) base function rbinom (‘n’ = 20 simulated pure wild boar, bias of uneven sampling on ordination (McVean 2009; DeGiorgio and ‘size’ = 2 alleles, ‘prob’ = a vector with, for each locus, the MAF observed at Rosenberg 2013), we randomly selected a maximum of five individuals per population level, which, unless the whole population is hybridized, sampling location for the PCoA. represents a wild boar signature). Because of the existence of genetic Lastly, to examine spatial patterns, we calculated and mapped structure among populations (see Results), MAFs were estimated multilocus observed heterozygosity (MLH) using R base functions. As a separately per genetic cluster (four clusters: Iberia, southern France, Italy, further exploration of patterns of gene flow, we examined allelic clines in and rest of Europe) (Fig. S2). To subsequently estimate genetic distances European wild boar. We visualized the spatial distribution of the mean among pigs, the simulated pure wild boar and the observed wild boar frequency of alleles typical for focal regions (selection of SNP loci for samples, we calculated Hamming pairwise genetic distances using the R which MAF within the focal region is larger than 0.5). We then package poppr 2.8.0 (Kamvar et al. 2014). Subsequently, we did a PCoA interpolated the mean allele frequencies over the map of Europe (see analysis using the R package ape 3.0 (Paradis et al. 2004). In accordance Fig. S6 for the spatial distribution of the wild boar samples used for with McVean (2009), the projected ordination positions of the observed interpolation), using the function autoKrige from the R package automap wild boar samples towards the pig source population were then taken as a 1.0–14 (Hiemstra et al. 2009). proxy for percentage of pig ancestry. Furthermore, following Goedbloed et al. (2013a), we further explored pig introgression by examining the distribution of alleles that are uncommon in European wild boar (MAF < 0.025), but abundant in pigs (MAF > 0.225). We studied whether RESULTS these putative pig alleles had a higher frequency in the hybrids detected Hybridization and inbreeding by PCoA projection. Additionally, we examined the distribution of these A spatial map of the projection scores of samples along the axis alleles along the genome, thereby verifying clustering of pig alleles in discriminating between wild boar and pigs in the PCoA showed certain regions of the genome (Fig. S3), which is indicative of introgressed that wild boar x pig hybrids occur in multiple countries across the haplotypes (Goedbloed et al. 2013a). continent (Fig. 1A). Using the PCoA projections as a proxy for pig To study the level of inbreeding, we used the fraction of the genome ancestry (Fhybrid) of the 330 unrelated European wild boar containing ROH segments (FROH). FROH can be reliably estimated with investigated, we estimated that 22 (7%), 20 (3%) and 6 (2%) 10,000 SNPs or more, if targeted at large ROH segments (Kardos et al. 2018). We used the 21,287 SNPs with relaxed LD filtering and focused on samples had Fhybrid values of 0.063–0.125, 0.125–0.250, and >5 Mb regions (common ancestor maximum 10 generations ago; Kardos >0.250, respectively. These putative hybrids stemmed mainly et al. 2018) that had a minimum of 50 SNPs with a mean density of at least from wild boar populations in northwestern Europe, southern 1 SNP per 150 kb and a maximum inter-SNP distance of 500 kb (Fig. S4). We Switzerland, Italy, and the Balkans. This result was in line with the detected ROHs using the homozyg function in PLINK and criteria of 70 observed occurrence of clusters of alleles abundant in pigs, but consecutive SNPs, with no heterozygotes allowed (Howrigan et al. 2011; rare in European wild boar, in the genomes of the putative hybrids Ferenčaković et al. 2013). We excluded wild boar samples for which (Fig. S3). The median PCoA scores for Iberian, Italian and French FROH>5Mb was larger than 0.125 as that is the expected level of autozygosity wild boar were 0.15, 0.08 and 0.10, respectively (Fig. S5), indicating for offspring of 2nd order relatives. a high genetic similarity with domestic pigs, whereas in the Balkans and eastern Europe we additionally observed an influence Spatial genetic structure from Asiatic lineages (Fig. 2). We examined spatial genetic structure through PCoA and ADMIXTURE We detected ROHs longer than 5 Mb (inbreeding event < 10 1.3.0 (Alexander et al. 2009), the latter plotted against a European map of generations ago) in populations all across the continent historic forest cover (year: 1850, Kaplan et al. 2009). To prevent inbreeding (Fig. 1B) over a total of 662 Mb scanned regions, equal to and hybridization from confounding the genetic structure analysis, we excluded inbred (FROH > 0.0625) and hybrid wild boar (Fhybrid > 0.0625) 29% of the wild boar autosomal genome (NCBI 2018). In 36 from these analyses (see Fig. S5 for the impact on the PCoA). ADMIXTURE (11%) of the 330 sampled unrelated wild boar, FROH>5Mb was 1.3.0 tool assesses the most likely number of genetic clusters and larger than 0.125. Additionally, 47 wild boar had FROH>5Mb subsequently the individual cluster membership proportions for each between 0.063 and 0.125. Most of these inbred individuals individual by means of a maximum likelihood approach. To decrease the were found in just a few sampling locations, namely the Fig. 1 Recent anthropogenic impact on European wild boar. A Recent pig hybridization. Fhybrid is the fraction of the genome of domestic pig origin, as determined by PCOA projection. Fhybrid of 0.250, 0.125, 0.0625 are the expected values for a 2nd, 3rd and 4th generation hybrid. B Recent inbreeding. FROH > 5 Mb is the proportion of the genome that contains Runs of Homozygosity larger than 5 Mb. ROHs longer than 5 Mb stem from common ancestors typically less than 10 generations ago. To reduce the overlap of samples from the same population, the sample locations are jittered. FROH values of 0.125, 0.0625 and 0.03125 are the expected values for offspring of half siblings, full cousins and half cousins. Please note, however, that these values can also be obtained through accumulation of inbreeding via multiple, distant common ancestors. Heredity (2023) 130:135 – 144 J.F. de Jong et al. 138 Fig. 2 Allelic clines in European wild boar. Spatial distribution of the mean frequency of alleles typical for focal regions (selection of SNP loci for which Minor Allele Frequency (MAF) within the focal region is larger than 0.5): (A) the Iberian Peninsula; (B) southern France; (C) the Italian Peninsula; (D) southeastern Europe; (E) northwestern Balkans; and (F) the Far East. The spatial distribution of the samples underlying the interpolations is shown in Fig. S6. Color scales are calibrated for each region specific selection set of SNP loci. populations of northeastern Spain, The Netherlands, western the rest of Europe (Fig. 3B). In both PCoA analyses, wild boar from Germany, southern Switzerland and Greece. Italy had a central position, near the origin of the axes, showing that the main axes could not explain the genetic variation of the Spatial genetic structure Italian peninsula. Concordantly, ADMIXTURE analysis could not The first axis of the PCoA of European wild boar distinguished assign Italian wild boar to any cluster at K = 5, the most likely those from western and eastern Europe (Fig. 3A). On the second number of clusters (Fig. S7). axis, wild boar from the Iberian peninsula were separated from the The first and second axes of both PCoAs were correlated with rest of western Europe. A subsequent PCoA, without Iberian wild longitude and latitude, respectively (Spearman rank correlation, boar, showed, again, separation of western and eastern European all p < 0.05, whether with or without the Iberian peninsula, see wild boar. The second axis distinguished wild boar of northeastern Fig. 3). In addition, there was a significant and strong correlation Europe (northeastern Germany, northern Poland, Estonia, Finland between geographic and genetic distance of wild boar popula- and Russia – Kaliningrad region and Central Federal District) from tions (rmantel = 0.73 and rmantel = 0.63 with and without Iberian Heredity (2023) 130:135 – 144 J.F. de Jong et al. 139 Fig. 4 Multilocus Heterozygosity (MLH) of wild boar across Europe. To reduce the overlap of samples from the same population, the sample locations are jittered. MLH values are classified in 10 intervals with an equal number of observations (i.e. deciles). Legend numbers represent the lower boundaries of the intervals. eastern group (comprising populations from Poland eastwards), nearby wild boar from southeastern Germany clustered with the western group (namely Germany, France, Belgium and the Netherlands). Samples from Austria appeared to be a contact zone between these two European clusters (PCoA projection at the center of the axis, and mixed ADMIXTURE assignment probabilities, Fig. S7). The second axis of the PCoA showed low genetic dissimilarity among distant wild boar populations, with wild boar from Greece and Bulgaria projecting closer to northeastern European populations than animals from the Carpathians or the Dinaric Alps. Overall, our results show the characteristics of a fragmented cline extending from the Dinaric Alps via southeastern Europe and the Carpathians, to north- eastern Europe, and, from there, to western Europe and, ultimately, the Iberian peninsula. The spatial distribution of alleles typical for putative refugial areas, or source populations (MAF > 0.5), showed insightful continental-wide patterns (Fig. 2). Alleles typical for the south- Fig. 3 The population genetic structure of European wild boar. PCoA biplots of Hamming genetic distances among wild boar, eastern (Greece and Bulgaria) and the southwestern regions superimposed on the map of Europe, for wild boar samples of (A) (Iberian peninsula, and southwestern France) gradually decreased the whole European mainland, and (B) the European mainland in frequency towards northern latitudes (Fig. 2). In contrast, Italian without the Iberian Peninsula. Inbred and hybrid wild boar are alleles had uniformly low frequency outside the Italian peninsula excluded. The PCoA scores of the samples are shown as country (Fig. 2). In turn, alleles typical for the Dinaric Alps were relatively abbreviations. Colors, lines and dots indicate geographic origin of abundant in the area south of the Carpathians, present at low samples. The shorter the lines, the stronger the match between the frequency in northeastern Europe and absent, or nearly so, in PCoA biplot and geography. Filled circles with country codes are the western Europe (Fig. 2E). Lastly, alleles typical for Far East Asia mean PCoA scores per country. rIBD is the Mantel r correlation appeared to be relatively abundant in eastern Europe, and in coefficient of genetic vs. geographic distance. ‘rlong’ and ‘rlat’ are the Spearman rank correlation coefficients of the PCoA first and particular in southeastern Europe. There was a clear divide in the second axes scores with longitude and latitude, respectively, allele frequency; to the west of the line extending from the Alps to calculated at population level to reduce autocorrelation. the Baltic states, including the Italian peninsula, the frequency of alleles typical for wild boar of Far Eastern Asia was markedly lower than in eastern Europe (Fig. 2F). peninsula, respectively; both analyses: p < 0.001, see Fig. 3). The spatial map of the Multilocus Heterozygosity (MLH) Nevertheless, the clusters detected by PCoA (and supported by showed an overall west-east gradient (Fig. 4). We detected a ADMIXTURE, Fig. S7) showed deviations from an isolation by positive correlation between MLH and longitude but no correla- distance (IBD) scenario. The first divide detected was the one tion with latitude (general linear model, F = 288.7, r2adj = 0.78, between Iberian and non-Iberian wild boar (Fig. 3A), but Fig. 3B d.f. = 324; coefficient estimate longitude = 0.29; t-value estimate showed also a major genetic discontinuity in central Europe. longitude = 7.64). Additionally, with the same model, we Whereas wild boar from Slovenia, Hungary and Slovakia were observed, as expected, a positive correlation between MLH and projected on the right of the PCoA axis 1 together with the Fhybrid (coefficient estimate = 0.22, t-value estimate = 12.1) and a Heredity (2023) 130:135 – 144 J.F. de Jong et al. 140 strongly negative correlation with FROH (coefficient estimate = (next to natural barriers effect, such of the Pyrenees mountains, −0.56, t-value estimate = −31.67). All general linear models had to be discussed below). p values < 0.001. The presence of local genetic discontinuities in combination with many, long ROH segments is likely the genetic legacy of size reductions and fragmentation of wild boar populations. Earlier DISCUSSION work showed genetic fragmentation in specific regions (e.g., Using genome-wide data and a continent-wide coverage of Scandura et al. 2008; Ferreira et al. 2009; Goedbloed et al. 2013b; populations across Europe, we examined recent anthropogenic Kusza et al. 2014); the present study indicates that fragmentation impact on, and revealed the spatial structure of, the genetic is a continent-wide phenomenon that appears to be most variation of European wild boar. We found considerable and eminent in areas of high human population density. We widespread anthropogenic impact on wild boar in the form of observed genetic discontinuities in western and central Europe inbreeding, fragmentation and pig introgression, though not and among the western, northeastern, Carpathian, Dinaric and translocation. Nevertheless, continent-wide PCoA analysis, ADMIX- southeastern clusters. The separation of the northeastern cluster TURE analysis and spatial maps of allele frequencies and coincides with the near absence of wild boar in northwestern heterozygosity showed that the genetic composition of wild boar Europe (northern Netherlands and northwestern Germany; in Europe takes the form of an arch-shaped cline that extends Goedbloed et al. 2013b) and southeastern Poland at the start from southeastern Europe, via the Baltic region, to western Europe of the 20th century (Apollonio et al. 2010). In eastern and and, finally, the Iberian peninsula – the latter being much southeastern Europe, genetic discontinuities along the Galicia, differentiated from the rest of the continent. Lastly, we found an Pannonian and Dacian basins are associated with human increasing gradient of heterozygosity from west to east, with the presence. These fertile river floodplains are, and have been, highest levels of heterozygosity found in a line extending from the densely inhabited and the landscape has been altered into Alps to the Baltic region. agricultural land, with only small patches of forests. This change These observations stimulate the ongoing discussion on what in land use, combined with overhunting, led to low wild boar shapes the genetic structure of European wild boar, and northern densities during the 19th and (first half of the) 20th centuries. Our hemisphere wildlife in general. They are partially concordant with results appear to still reflect this situation, despite a recent previous studies based on mtDNA (Scandura et al. 2011a; comeback of the species in those areas (Apollonio et al. 2010; Alexandri et al. 2012; Kusza et al. 2014; Vilaça et al. 2014; Deinet et al. 2013). The erratic spatial pattern in inbreeding (as Veličković et al. 2015; Maselli et al. 2016) and microsatellites determined through FROH) indicates that the severity of (Vernesi et al. 2003; Scandura et al. 2008; Veličković et al. 2016), fragmentation and associated population size reductions vary that reported a gradient in genetic variability and different from region to region. The populations with the highest levels of contributions from glacial refugia. However, our findings addi- inbreeding (e.g., Collserola, Northeast Spain; Meinweg, Southeast tionally show unreported patterns highlighting the relative Netherlands) share a history of strong isolation and severe importance of (i) anthropogenic influences versus natural bottlenecks. For all these populations, the bottlenecks mostly processes, and (ii) the LGM versus other natural processes. We occurred approximately one or half a century ago. Although therefore argue that the genetic structure of wild boar can only be these populations are currently increasing and might keep doing understood from a complex, multi-faceted perspective, that so in the future, coancestry and hence inbreeding may continue integrates many different and sometimes opposing demographic to accumulate (see e.g., Kardos et al. 2018). processes. To aid understanding, our inferences—summarized in What are the potential phenotypic and ecological conse- Table S1 and introduced and explained below—are spatially quences of these human-caused alterations of genetic variation? conceptualized in Fig. S8. Apart from potentially lowering adaptive potential, the observed levels of inbreeding have in other wildlife species been shown to Anthropogenic influences be associated with body weight and juvenile survival (Bérénos Frequent and strong signatures of inbreeding and pig hybridi- et al. 2016; Huisman et al. 2016). For wild boar specifically, there zation indicate a recent anthropogenic impact on the genetic are however, to our best knowledge, no reported cases of variation of European wild boar. Earlier, inter alia Goedbloed inbreeding depression. In contrast, pig hybridization is in the field et al. (2013a) provided evidence for the occurrence of pig often inferred by aberrant phenotypic variation, in particular hybridization at regional scale, which leads to a genetic multicolored coats. Genetically, pig hybridization might cause signature of aberrant genomic segments with unusual alleles introgression of selected haplotypes in small regions of the that are indicative of introgressed haplotypes (Goedbloed et al. genome (Ai et al. 2015) over time. Although limited to few and 2013a). Our findings identified that recent hybridization, small regions of the genome, these haplotypes might cause long- although geographically confined, occurred in several countries. term phenotypic consequences (Canu et al. 2016; Fulgione et al. However, introgressed pig haplotypes can only be detected if 2016; Iacolina et al. 2019), including vital phenotypic traits such as the wild and domestic populations are two separate gene pools. litter size (Fulgione et al. 2016). The high observed genetic similarity of Iberian and Italian wild boar with domestic pigs (see Fig. S2) was difficult to ascertain Spatial genetic structure: Anthropogenic influences vs. natural through detection of introgressed haplotypes. Either the signal processes is a type-I error and pig hybridization is infrequent (in line with Whilst frequent and strong, the anthropogenic-caused alterations findings for Italian wild boar by Scandura et al. 2022) or pig of genetic variation do not appear to have erased the continent- hybridization has been so frequent and long-lasting that wide spatial genetic patterns of natural origin. The two main introgressed segments are too short to detect with medium continent-wide genetic patterns observed were an arch shaped density SNP data. Both Iberia and Italy share a long tradition of genetic cline (extending from western to northeastern and extensive pig herding for regional ham production, up to the southeastern Europe), and a complex heterozygosity gradient present, that possibly has led to a prolonged gene-flow (low in the west, high in the east, the highest in the line from the between the wild boar and domestic pigs (Herrero-Medrano Alps to the Baltic region). Assuming that anthropogenic influences et al. 2013; Canu et al. 2014; Iacolina et al. 2016; Maselli et al. typically produce erratic patterns (see e.g., De Jong et al. 2020), we 2016). Pig hybridization may perhaps even partially explain suspect that the regularity of these patterns imply underlying the observed genetic differentiation of the southern peninsulas, forces of natural origin. This regularity is highlighted by the fact in particular Iberia, from the rest of the European continent that IBD proved to be a strong predictor in our results. Heredity (2023) 130:135 – 144 J.F. de Jong et al. 141 Several apparently contradictory patterns emerge from our other species as well (roe deer, Lorenzini et al. 2003; red deer, results: (i) multiple local genetic discontinuities despite the Carranza et al. 2016). existence of a strong pattern of isolation by distance; (ii) severe In the east, northward colonization likely took place from the inbreeding occurring locally, despite a continental-wide het- Balkans with multiple refugia (i.e., Dinaric Alps, the Carpathians erozygosity gradient; (iii) differentiation patterns correlated and Slavyanka mountains), as previously suggested (Alexandri with longitude and latitude, despite substantial pig hybridiza- et al. 2012, 2017; Veličković et al. 2015, 2016) and in agreement tion. These seemingly contradictory results are likely explained with environmental simulations (Vilaça et al. 2014). However, the by scale and resolution as large-scale spatial genetic structure spatial distribution of allelic frequencies showed a non- may be insensitive to local alterations of genetic variation. neglectable contribution of eastern lineages to eastern European Hence, the question on to what extent human activities alter populations, contrary to what has previously been reported genetic variation of wildlife, is much dependent on spatial scale (Manunza et al. 2013), suggesting a possible colonization route (see, for example, the continental-wide study of Scandura et al. from the Caucasus (Schmitt and Varga 2012). This eastern (2008) and the regional study of Ferreira et al. (2009)). component might contribute to explaining why northeastern Furthermore, even when genetic alteration may be negligible Europe is genetically more similar to southeastern Europe than to in a whole genome perspective, the phenotypic and ecological the nearby Carpathians and the higher variability in eastern impact may be considerable (such as e.g., an increase in litter Europe compared to the southern populations. size; Fulgione et al. (2016)). Human alteration of genetic The formation of the suture zone on the northeastward line variation of wildlife should thus perhaps be evaluated at the Alps-Baltics rather than a northward line, might have been phenotypic scale. caused by an earlier, or faster, expansion by wild boar from southern France, that consequently recolonized most of north- Spatial genetic structure: LGM vs. other natural processes ern Europe. A similar pattern is suggested for red deer by fossil The existence of refugia during the LGM, and the southwards records and microsatellite variation (Sommer et al. 2008; Zachos retraction and northwards recolonization from these refugia, is et al. 2016). An explanation may be that at the start of the generally considered to be the most dominant natural force Holocene (in particular, around 9000 years ago) the region south shaping genetic variation of European wildlife (Hewitt 1999). In of the Tatras Mountains was still relatively cold and dominated two ways, our findings may give understanding of the genetic by coniferous trees, while in western Europe temperatures were legacy of the LGM: (i) the location of a suture zone (i.e., area of milder and broadleaved forests dominated the vegetation secondary contact), and (ii) the source populations for the post- (Brewer et al. 2002; Cheddadi and Bar-Hen 2009). The absence LGM northward recolonization. High levels of heterozygosity of autumn mast and occurrence of severe winters likely limited along the line Alps-Baltics (Fig. 4), and in addition a sharp the survival of wild boar in southeastern Europe during the early transition of allele frequencies over this line (Fig. 2) may indicate Holocene, similar to boreal forests of northern Europe today that this is a suture zone, i.e., an area of secondary contact (Melis et al. 2006; Apollonio et al. 2010). between two divergent source populations. This observation is Next to assumed LGM signatures, we also observed spatial further supported by the correspondence with the known genetic patterns not easily reconcilable with the LGM impact. transition of the western karyotype of 2n = 36 to the ancestral First and foremost, the PCoA (Fig. 3) showed an ordination that is eastern karyotype of 2n = 38 (Rejduch et al. 2003; Fang et al. atypical for a postglacial expansion. Due to allele surfing (Braga 2006). However, we could not exclude that the genetic et al. 2019) and the consequential genetic drift, a PCoA on discontinuity in central Europe is a consequence of local expanding populations typically opposes individuals of recolo- eradications and subsequent recovery (or reintroductions) of nized areas, not refugial areas (Franois et al. 2010). The fact that wild boar since the second world war, as discussed above (Krže our PCoA found the largest genetic differences among southern 1982; Deinet et al. 2013; Bragina et al. 2015). populations, rather than northern ones, shows that the signature Accepting the idea of a western and an eastern clade, based of the LGM has become relatively weak in the autosomal on our results we infer a pivotal role of southern France as a genome. Second, except for a circumstantial bottleneck in the western post-glacial expansion source, concordantly with western refugium, there is no known LGM mechanism that could previous genetic and archeological studies (Sommer and have caused low heterozygosity in western Europe. Instead, the Nadachowski 2006; Sommer and Zachos 2009; Vilaça et al. central–marginal hypothesis may apply, which holds that in the 2014), whereas contributions from the Iberian and Italian core of the distribution, which for wild boar is Asia, there is a peninsulas are unlikely. Although, in agreement with previous higher genetic diversity than in the periphery (Eckert et al. 2008). studies (Scandura et al. 2011a; Alexandri et al. 2012; Kusza et al. Arguably, the immigration history from Asia towards Europe 2014; Vilaça et al. 2014; Veličković et al. 2015; Maselli et al. 2016), (Azzaroli et al. 1988) and particularly western Europe, the the Iberian and Italian peninsulas showed signs of differentia- periphery, is a more likely explanation of the observed tion compatible with glacial refugia, our results differ in terms of heterozygosity cline. Such immigration has commenced in the their contribution to post-glacial recolonization patterns. The Middle Pleistocene and might have occurred repeatedly heterogeneous but still distinct genetic characteristics of the throughout the Lower Pleistocene and Holocene (Palombo and Italian population suggests a minor role in post-glacial Romana 2003; Magri 2013). Ultimately, this implies that the LGM colonization patterns, which is in contrast with previous studies as dominant, leave alone, sole natural force shaping the genetic (Alexandri et al. 2012; Vilaça et al. 2014; Veličković et al. structure of wild boar–and other northern hemisphere wildlife – 2015, 2016, but see Hewitt 1999). While we may possibly have may be too simple a picture. failed to detect a contribution of Italian wild boar in post-glacial recolonization of Europe, due to the relatively small sample size, a negligible contribution of the Italian peninsula matches CONCLUSIONS patterns inferred in other European mammals (Taberlet et al. We provided evidence that the spatial genetic variation of 1998; Hewitt 1999), including roe (Capreolus capreolus; Lorenzini European wild boar is the outcome of a complex interplay of et al. 2002) and red deer (Sommer et al. 2009). Additionally, the multiple processes of both anthropogenic and natural origin. We Iberian peninsula - with its marked genetic differentiation from observed many and strong signatures that we attribute to human the rest of the continent – also appears to have played a impacts, including loss of genetic variation due to inbreeding, negligible role during the recolonization process, which is increased genetic variation due to pig introgression, and the concordant with results from Vilaça et al. (2014) and observed in existence of genetic discontinuities in areas without natural Heredity (2023) 130:135 – 144 J.F. de Jong et al. 142 barriers on the background of a continent-wide pattern dating Deinet S, Ieronymidou C, McRae L, Burfield IJ, Foppen RP, Collen B, et al. (2013) back to the LGM, or possibly even earlier. Future research, Wildlife comeback in Europe. The recovery of selected mammal and bird species. involving historical samples and sequence data should study the London, UK origin of haplotypes to shed light on the diverse ancestry of wild Eckert CG, Samis KE, Lougheed SC (2008) Genetic variation across species’ geo- graphical ranges: the central–marginal hypothesis and beyond. Mol Ecol boar, and other species, of the European continent, while modern 17:1170–1188 samples should be investigated to identify how anthropogenic Fang M, Berg F, Ducos A, Andersson L (2006) Mitochondrial haplotypes of European influences are affecting the survival and adaptability of the wild boars with 2n = 36 are closely related to those of European domestic pigs species, also considering the need of science-based management with 2n = 38. Anim Genet 37:459–464 practices for a species that is increasingly considered a source of Ferenčaković M, Sölkner J, Curik I (2013) Estimating autozygosity from high- human-wildlife conflict. throughput information: Effects of SNP density and genotyping errors. Genet Sel Evol 45:42 Ferreira E, Souto L, Soares AMVM, Fonseca C (2009) Genetic structure of the wild boar DATA AVAILABILITY population in Portugal: Evidence of a recent bottleneck. Mamm Biol 74:274–285 Genotypes analysed in this article are available on Dryad https://doi.org/10.5061/ Franois O, Currat M, Ray N, Han E, Excoffier L, Novembre J (2010) Principal component dryad.1rn8pk0z6. analysis under population genetic models of range expansion and admixture. Mol Biol Evol 27:1257–1268 Frantz AC, Bertouille S, Eloy MC, Licoppe A, Chaumont F, Flamand MC (2012) Com- parative landscape genetic analyses show a Belgian motorway to be a gene REFERENCES flow barrier for red deer (Cervus elaphus), but not wild boars (Sus scrofa). Mol Ai H, Fang X, Yang B, Huang Z, Chen H, Mao L et al. (2015) Adaptation and possible Ecol 21:3445–3457 ancient interspecies introgression in pigs identified by whole-genome Fulgione D, Rippa D, Buglione M, Trapanese M, Petrelli S, Maselli V (2016) Unexpected sequencing. Nat Genet 47:217–225 but welcome. Artificially selected traits may increase fitness in wild boar. Evol Alexander DH, Novembre J, Lange K (2009) Fast model-based estimation of ancestry Appl 9:769–776 in unrelated individuals. Genome Res 19:1655–1664 Goedbloed DJ, Megens HJ, van Hooft P, Herrero-Medrano JM, Lutz W, Alexandri P Alexandri P, Megens HJ, Crooijmans RPMA, Groenen MAM, Goedbloed DJ, Herrero- et al. (2013a) Genome-wide single nucleotide polymorphism analysis reveals Medrano JM et al. (2017) Distinguishing migration events of different timing for recent genetic introgression from domestic pigs into Northwest European wild wild boar in the Balkans. J Biogeogr 44:259–270 boar populations. Mol Ecol 22:856–866 Alexandri P, Triantafyllidis A, Papakostas S, Chatzinikos E, Platis P, Papageorgiou N Goedbloed DJ, van Hooft P, Megens HJ, Langenbeck K, Lutz W, Crooijmans RPMA et al. (2012) The Balkans and the colonization of Europe: the post-glacial range et al. (2013b) Reintroductions and genetic introgression from domestic pigs expansion of the wild boar, Sus scrofa. J Biogeogr 39:713–723 have shaped the genetic population structure of Northwest European wild Alves PC, Pinheiro I, Godinho R, Vicente JJ, Gortázar C, Scandura M et al. (2010) boar. BMC Genet 14:2–10 Genetic diversity of wild boar populations and domestic pig breeds (Sus scrofa) Groenen MAM, Archibald AL, Uenishi H, Tuggle CK, Takeuchi Y, Rothschild MF et al. in South-western Europe. Biol J Linn Soc 101:797–822 (2012) Analyses of pig genomes provide insight into porcine demography and Apollonio M, Andersen R, Putman R (2010) European ungulates and their management evolution. Nature 491:393–398 in the 21st century (M Apollonio, R Andersen, and R Putman, Eds.) Cambridge Herrero-Medrano JM, Megens H-J, Groenen MAM, Ramis G, Bosse M, Pérez-Enciso M University Press: Cambridge, UK et al. (2013) Conservation genomic analysis of domestic and wild pig popula- Azzaroli A, De Giuli C, Ficcarelli G, Torre D (1988) Late pliocene to early mid- tions from the Iberian Peninsula. BMC Genet 14:1–13 pleistocene mammals in Eurasia: Faunal succession and dispersal events. Hewitt GM (1999) Post-glacial re-colonization of European biota. Biol J Linn Soc Palaeogeogr Palaeoclimatol Palaeoecol 66:77–100 68:87–112 Bérénos C, Ellis PA, Pilkington JG, Pemberton JM (2016) Genomic analysis reveals Hewitt GM (2004) Genetic consequences of climatic oscillations in the Quaternary. depression due to both individual and maternal inbreeding in a free‐living Philos Trans R Soc Lond Ser B Biol Sci 359:183–195 mammal population. Mol Ecol 25:3152–3168 Hiemstra PH, Pebesma EJ, Twenhöfel CJW, Heuvelink GBM (2009) Real-time auto- Braga RT, Rodrigues JFM, Diniz-Filho JAF, Rangel TF (2019) Genetic population matic interpolation of ambient gamma dose rates from the Dutch radioactivity structure and allele surfing during range expansion in dynamic habitats. An da monitoring network. Comput Geosci 35:1711–1721 Academia Brasileira de Ciências 91:e20180179 Howrigan DP, Simonson MA, Keller MC (2011) Detecting autozygosity through runs of Bragina EV, Ives AR, Pidgeon AM, Kuemmerle T, Baskin LM, Gubar YP, Piquer- homozygosity: a comparison of three autozygosity detection algorithms. BMC Rodríguez M, Keuler NS, Petrosyan VG, Radeloff VC (2015) Rapid Declines of Genomics 12:460 Large Mammal Populations after the Collapse of the Soviet Union. Cons Biol Huisman J, Kruuk LEB, Ellis PA, Clutton-Brock T, Pemberton JM (2016) Inbreeding 29:844–853 depression across the lifespan in a wild mammal population. Proc Natl Acad Sci Brewer S, Cheddadi R, de Beaulieu JL, Reille M, Allen J, Almqvist-Jacobson H et al. 113:3585–3590 (2002) The spread of deciduous Quercus throughout Europe since the last Iacolina L, Corlatti L, Buzan E, Safner T, Šprem N (2019) Hybridisation in European glacial period. For Ecol Manag 156:27–48 ungulates: an overview of the current status, causes, and consequences. Mamm Cahill S, Llimona F, Cabañeros L, Calomardo F (2012) Characteristics of wild boar (Sus Rev 49:45–59 scrofa) habituation to urban areas in the Collserola Natural Park (Barcelona) and Iacolina L, Pertoldi C, Amills M, Kusza S, Megens H-J, Bâlteanu VA et al. (2018) Hot- comparison with other locations. Anim Biodivers Conserv 35:221–233 spots of recent hybridization between pigs and wild boars in Europe. Sci Rep. Canu A, Costa S, Iacolina L, Piatti P, Apollonio M, Scandura M (2014) Are captive wild 8:17372 boar more introgressed than free-ranging wild boar? Two case studies in Italy. Iacolina L, Scandura M, Goedbloed DJ, Alexandri P, Crooijmans RPMA, Larson G et al. Eur J Wildl Res 60:459–467 (2016) Genomic diversity and differentiation of a managed island wild boar Canu A, Vilaça STT, Iacolina L, Apollonio M, Bertorelle G, Scandura M (2016) Lack of population. Heredity 116:60–67 polymorphism at the MC1R wild-type allele and evidence of domestic allele Jombart T, Ahmed I (2011) adegenet 1.3-1: new tools for the analysis of genome-wide introgression across European wild boar populations. Mamm Biol 81:477–479 SNP data. Bioinformatics 27:1–2 Carranza J, Salinas M, de Andrés D, Pérez-González J (2016) Iberian red deer: para- de Jong JF, Hooft van P, Megens HJ, Crooijmans RPMA, Groot de GA, Pemberton phyletic nature at mtDNA but nuclear markers support its genetic identity. Ecol JM, Huisman J et al. (2020) Fragmentation and translocation distort the Evol 6:905–922 genetic landscape of ungulates: red deer in the Netherlands. Front Ecol Evol Chang CC, Chow CC, Tellier LCAM, Vattikuti S, Purcell SM, Lee JJ (2015) Second- 8:535715 generation PLINK: Rising to the challenge of larger and richer datasets. Giga- Kamvar ZN, Tabima JF, Grünwald NJ (2014) Poppr: an R package for genetic analysis science 4:1–16 of populations with clonal, partially clonal, and/or sexual reproduction. Cheddadi R, Bar-Hen A (2009) Spatial gradient of temperature and potential vege- PeerJ 2:e281 tation feedback across Europe during the late Quaternary. Clim Dyn 32:371–379 Kardos M, Åkesson M, Fountain T, Flagstad Ø, Liberg O, Olason P et al. (2018) Clark PU, Dyke AS, Shakun JD, Carlson AE, Clark J, Wohlfarth B et al. (2009) The Last Genomic consequences of intensive inbreeding in an isolated wolf population. Glacial Maximum. Science 325:710–714 Nat Ecol Evol 2:124–131 DeGiorgio M, Rosenberg NA (2013) Geographic sampling scheme as a determinant of Kaplan JO, Krumhardt KM, Zimmermann N (2009) The prehistoric and preindustrial the major axis of genetic variation in principal components analysis. Mol Biol deforestation of Europe. Quat Sci Rev 28:3016–3034. https://doi.org/10.1016/ Evol 30:480–488 j.quascirev.2009.09.028 Heredity (2023) 130:135 – 144 J.F. de Jong et al. 143 Koemle D, Zinngrebe Y, Yu X (2018) Highway construction and wildlife populations: Sommer RS, Zachos FE (2009) Fossil evidence and phylogeography of temperate species: Evidence from Austria. Land use policy 73:447–457 ‘glacial refugia’ and post-glacial recolonization. J Biogeogr 36:2013–2020 Krže B (1982) Divji prašič: biologija, gojitev, ekologija. Lovska zveza Slovenije, Sommer RS, Zachos FE, Street M, Jöris O, Skog A, Benecke N (2008) Late Quaternary Ljubljana distribution dynamics and phylogeography of the red deer (Cervus elaphus) in Kusza S, Podgórski T, Scandura M, Borowik T, Jávor A, Sidorovich VE et al. (2014) Europe. Quat Sci Rev 27:714–733 Contemporary genetic structure, phylogeography and past demographic pro- Stillfried M, Fickel J, Börner K, Wittstatt U, Heddergott M, Ortmann S et al. (2017) Do cesses of wild boar Sus scrofa population in central and eastern Europe. PLoS cities represent sources, sinks or isolated islands for urban wild boar population One 9:e91401 structure? J Appl Ecol 54:272–281 Lorenzini R, Lovari S, Masseti M (2002) The rediscovery of the Italian roe deer: Genetic Taberlet P, Fumagalli L, Wust-Saucy AG, Cossons JF (1998) Comparative phylogeo- differentiation and management implications. Ital J Zool 69(4):367–379 graphy and post-glacial colonization routes in Europe. Mol Ecol 7:453–461. Lorenzini R, San José C, Braza F, Aragón S (2003) Genetic differentiation and phylo- Veličković N, Djan M, Ferreira E, Stergar M, Obreht D, Maletić V et al. (2015) From geography of roe deer in Spain, as suggested by mitochondrial DNA and north to south and back: the role of the Balkans and other southern peninsulas microsatellite analysis. Ital J Zool 70(1):89–99 in the recolonization of Europe by wild boar. J Biogeogr 42:716–728 Magri D (2013) Early to Middle Pleistocene dynamics of plant and mammal com- Veličković N, Ferreira E, Djan M, Ernst M, Obreht Vidaković D, Monaco A et al. (2016) munities in South West Europe. Quat Int 288:63–72 Demographic history, current expansion and future management challenges of Manunza A, Zidi A, Yeghoyan S, Balteanu VA, Carsai TC, Scherbakov O et al. (2013) A wild boar populations in the Balkans and Europe. Heredity 117:348–357 high throughput genotyping approach reveals distinctive autosomal genetic Vernesi C, Crestanello B, Pecchioli E, Tartari D, Caramelli D, Hauffe H et al. (2003) The signatures for European and Near Eastern wild boar. PLoS One 8:e55891 genetic impact of demographic decline and reintroduction in the wild boar (Sus Maselli V, Rippa D, De Luca A, Larson G, Wilkens B, Linderholm A et al. (2016) Southern scrofa): A microsatellite analysis. Mol Ecol 12:585–595 Italian wild boar population, hotspot of genetic diversity. Hystrix 27:137–144 Vilaça ST, Biosa D, Zachos F, Iacolina L, Kirschning J, Alves PC et al. (2014) Mito- McVean G (2009) A genealogical interpretation of principal components analysis. chondrial phylogeography of the European wild boar: The effect of climate PLoS Genet 5:e1000686 on genetic diversity and spatial lineage sorting across Europe. J Biogeogr Megens H-J, Crooijmans RP, Cristobal M, Hui X, Li N, Groenen MA (2008) Biodiversity 41:987–998 of pig breeds from China and Europe estimated from pooled DNA samples: Zachos FE, Frantz AC, Kuehn R, Bertouille S, Colyn M, Niedziałkowska M et al. (2016) differences in microsatellite variation between two areas of domestication. Genetic structure and effective population sizes in European red deer (Cervus Genet Sel Evol 40:103 elaphus) at a continental scale: insights from microsatellite DNA. J Hered Melis C, Szafrańska PA, Jȩdrzejewska B, Bartoń K (2006) Biogeographical variation in 107:318–326 the population density of wild boar (Sus scrofa) in western Eurasia. J Biogeogr 33:803–811 Mihalik B, Stéger V, Frank K, Szendrei L, Kusza S (2018) Barrier effect of the M3 ACKNOWLEDGEMENTS highway in Hungary on the genetic diversity of wild boar (Sus scrofa) popula- This study was financially supported by ‘Stichting De Eik’ and ‘De Koninklijke tion. Res J Biotechnol 13:32–38 Nederlandse Jagersvereniging’. The collection of samples from Russia was supported NCBI (2018) Genome Organism Overview: Sus scrofa (pig). https:// by The Russian Ministry of Science and Higher Education (Project No. AAAA-A19- www.ncbi.nlm.nih.gov/genome?term=sus%20scrofa%20%5BOrganism% 119022190168-8). LI received funding from the Horizon 2020 research and 5D&cmd=DetailsSearch&report=Overview innovation programme under the Marie Sklodowska-Curie Action (Grant Agreement Nikolov IS, Gum B, Markov G, Kuehn R (2009) Population genetic structure of wild no. 656697). BP acknowledges the financial support from the Ministry of Agriculture, boar Sus scrofa in Bulgaria as revealed by microsatellite analysis. Acta Theriol Forestry and Food of the Republic of Slovenia and the Slovenian Research Agency (Warsz) 54:193–205 (project V4–1825; programme group P4–0107). NŠ acknowledges the financial Nykänen M, Rogan E, Foote AD, Kaschner K, Dabin W, Louis M et al. (2019) Postglacial support by the Croatian Science Foundation, project IP 2019-04-4096 “The role of colonization of northern coastal habitat by bottlenose dolphins: a marine hunting related activities in the range expansion of recently established wild leading-edge expansion? J Hered 110:662–674 ungulate populations in the Mediterranean”. CF and EF are thankful to FCT/MCTES, Palombo M, Romana AV-G (2003) Remarks on the biochronology of mammalian for financial support to CESAM (UIDP/50017/2020+UIDB/50017/2020), through faunal complexes from the Pliocene to the Middle Pleistocene in France. Geol Portuguese national funds. We are very grateful to Leonardo Bereczky, Laura Maria Rom: 145–163 Boschiroli, Luigi Faino, Harry Enzerink, Jacqueline Korte, Marcel Lehocky, Igor Paradis E, Claude J, Strimmer K (2004) APE: analysis of phylogenetics and evolution in Lysanskiy, Nacho Rodríguez, Stefan Sand, Gerrit-Jan Spek and Reinhold Weinberger R language. Bioinformatics 20:289–290 for their help in sample collection. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MAR, Bender D et al (2007) PLINK: A tool Set for whole-genome association and population-based linkage analyses. Am J Hum Genet 81:559–575. www.cog-genomics.org/plink/1.9/ Putman R, Apollonio M, Andersen R (2011) Ungulate management in Europe: pro- AUTHOR CONTRIBUTIONS blems and practices. Cambridge University Press, Cambridge, UK JFdJ and HJM designed the study. JFdJ analysed the data. JFdJ, LI and HJM wrote the first R Core Team (2018) R: A language and environment for statistical computing. Vienna, draft of the manuscript. PvH, RPMAC, SEvW, RCY and HHTP contributed to the review of Austria the study design. JVB, EB, SC, EF, CF, PMG, IJT, VMLM, AN, BP, TP, NS, RV and LI provided Rejduch B, Sota E, Ró M, Ko M (2003) Chromosome number polymorphism in a litter samples and ecological information from the sampled areas, and contributed to the of European wild boar (Sus scrofa scrofa L.). Anim Sci Pap Rep. 21:57–62 interpretation of the results. JFdJ, LI, SEvW, RCY, HHTP and HJM interpreted the results. Scandura M, Iacolina L, Apollonio M (2011a) Genetic diversity in the European wild All authors revised the manuscript and contributed to the review process. boar Sus scrofa: phylogeography, population structure and wild x domestic hybridization: Genetic variation in European wild boar. Mamm Rev 41:125–137 Scandura M, Iacolina L, Cossu A, Apollonio M (2011b) Effects of human perturbation COMPETING INTERESTS on the genetic make-up of an island population: The case of the Sardinian wild The authors declare no competing interests. boar. Heredity 106:1012–1020 Scandura M, Iacolina L, Crestanello B, Pecchioli E, Di Benedetto MF, Russo V et al. (2008) Ancient vs. recent processes as factors shaping the genetic variation of ADDITIONAL INFORMATION the European wild boar: Are the effects of the last glaciation still detectable? Supplementary information The online version contains supplementary material Mol Ecol 17:1745–1762 available at https://doi.org/10.1038/s41437-022-00587-1. Scandura M, Fabbri G, Caniglia R, Iacolina L, Mattucci F, Mengoni C, Pante G, Apol- lonio M, Mucci N (2022) Resilience to Historical Human Manipulations in the Correspondence and requests for materials should be addressed to Laura Iacolina. Genomic Variation of Italian Wild Boar Populations. Front Ecol Evol 10:833081 Schmitt T, Varga Z (2012) Extra-Mediterranean refugia: the rule and not the exception. Reprints and permission information is available at http://www.nature.com/ Front Zool 9:22 reprints Sommer RS, Fahlke JM, Schmölcke U, Benecke N, Zachos FE (2009) Quaternary history of the European roe deer Capreolus capreolus. Mamm Rev 39:1–16 Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims Sommer RS, Nadachowski A (2006) Glacial refugia of mammals in Europe: evidence in published maps and institutional affiliations. from fossil records. Mamm Rev 36:251–265 Heredity (2023) 130:135 – 144 J.F. de Jong et al. 144 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http:// creativecommons.org/licenses/by/4.0/. © The Author(s) 2023 Heredity (2023) 130:135 – 144