Science of the Total Environment 833 (2022) 155126 Contents lists available at ScienceDirect Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv Genetic population structure defines wild boar as an urban exploiter species in Barcelona, Spain Justus Hagemann a,b, Carles Conejero c, Milena Stillfried a, Gregorio Mentaberre c,d, Raquel Castillo-Contreras c, ⁎ Jörns Fickel a,e,1, Jorge Ramón López-Olvera c, ,1 a Leibniz-Institute for Zoo and Wildlife Research, Alfred-Kowalke-Str. 17, D-10315 Berlin, Germany b University of Potsdam, Evolutionary Adaptive Genomics, Institute for Biochemistry and Biology, Karl-Liebknecht-Str. 24-25, 14476 Potsdam, Germany c Wildlife Ecology & Health group (WE&H) and Servei d'Ecopatologia de Fauna Salvatge (SEFaS), Departament de Medicina i Cirurgia Animals, Universitat Autònoma de Barcelona (UAB), 08193 Bellaterra, Barcelona, Spain d Departament de Ciència Animal, Escola Tècnica Superior d'Enginyeria Agraria (ETSEA), Universitat de Lleida (UdL), 25098 Lleida, Spain. e University of Potsdam, Molecular Ecology and Evolution, Institute for Biochemistry and Biology, Karl-Liebknecht-Str. 24-25, 14476 Potsdam, Germany H I G H L I G H T S G R A P H I C A L A B S T R A C T • Synurbic wild boar populations are genet- ically different from rural wild boars. • Conflictive synurbic wild boars establish as urban island populations. • We define wild boar for the first time as an urban exploiter species. • Whether synurbic phenotypic changes are genetic or plastic remains unknown. • Management of urban wild boars should prevent adaption to urban environments. A R T I C L E I N F O A B S T R A C T Editor: Rafael Mateo Soria Urban wildlife ecology is gaining relevance as metropolitan areas grow throughout the world, reducing natural habi- tats and creating new ecological niches. However, knowledge is still scarce about the colonisation processes of such Keywords: urban niches, the establishment of new communities, populations and/or species, and the related changes in behaviour Gene flow and life histories of urban wildlife. Wild boar (Sus scrofa) has successfully colonised urban niches throughout Europe. Island population The aim of this study is to unveil the processes driving the establishment and maintenance of an urban wild boar pop- Population genetics Sus scrofa ulation by analysing its genetic structure. A set of 19 microsatellite loci was used to test whether urban wild boars in Synurbisation Barcelona, Spain, are an isolated population or if gene flow prevents genetic differentiation between rural and urban Urban ecology wild boars. This knowledge will contribute to the understanding of the effects of synurbisation and the associated man- agement measures on the genetic change of large mammals in urban ecosystems. Despite the unidirectional gene flow from rural to urban areas, the urban wild boars in Barcelona form an island pop- ulation genotypically differentiated from the surrounding rural ones. The comparison with previous genetic studies of urban wild boar populations suggests that forest patches act as suitable islands for wild boar genetic differentiation. Previous results and the genetic structure of the urban wild boar population in Barcelona classify wild boar as an urban exploiter species. These wild boar peri-urban island populations are responsible for conflict with humans and thus should be managed by reducing the attractiveness of urban areas. The management of peri-urban wild boar populations should aim at reduc- ing migration into urban areas and preventing phenotypic changes (either genetic or plastic) causing habituation of wild boars to humans and urban environments. ⁎ Corresponding author at: Wildlife Ecology & Health group (WE&H) and Servei d'Ecopatologia de Fauna Salvatge (SEFaS), Departament de Medicina i Cirurgia Animals, Facultat de Veterinària (edifici V), Travessera dels Turons s/n, Universitat Autònoma de Barcelona (UAB), E-08193, Bellaterra, Barcelona, Spain. E-mail address: Jordi.Lopez.Olvera@uab.cat (J.R. López-Olvera). 1 These authors share senior authorship. http://dx.doi.org/10.1016/j.scitotenv.2022.155126 Received 28 January 2022; Received in revised form 27 March 2022; Accepted 5 April 2022 Available online 8 April 2022 0048-9697/© 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). J. Hagemann et al. Science of the Total Environment 833 (2022) 155126 1. Introduction Delibes et al., 2001), as defined by the ‘Population Pressure Hypothesis’ (PPH). Alternatively, the ‘Urban Island Hypothesis’ (UIH) predicts a spread Urbanisation, defined as the shift from rural to urban human residency, of urban founder populations after an initial bottleneck, assuming gene- is a global and exponentially increasing process (United Nations, 2017). flow barriers between urban and rural populations leading to genetic Synurbic wildlife can thrive in urban environments benefiting from anthro- differentiation over time (Gloor, 2002; Wandeler et al., 2003). Thus, pogenic and natural resources (Castillo-Contreras et al., 2021; Fischer et al., knowledge about the genetic composition of populations provides a much 2015; Kotulski and Koenig, 2008; Stillfried et al., 2017a), reaching even better measure to understand the colonisation process and allows the higher population densities than the surrounding countryside (Šálek ecological classification of urban wildlife and particularly wild boar et al., 2015) and increasing human-wildlife encounters (Soulsbury and populations. White, 2015). Cities may represent ecological niches for wildlife with low Similar to the overall European trend, wild boar populations in (if any) human hunting pressure and predation, high habitat diversity, Catalonia (northeastern Spain) are consistently growing and they will con- and anthropogenic food sources (Shochat et al., 2006). Thus, these niches tinue increasing in the future (Cahill and Llimona, 2004; González-Crespo may increase fitness for urban dwellers, compared with their rural conspe- et al., 2018), both in rural areas such as the Natural Park of Sant Llorenç cifics (Kotulski and Koenig, 2008). del Munt i l'Obac (SLMO) as well as in peri-urban locations, such as the Moreover, human-induced habitat changes and novel biotic interac- Collserola Natural Park (CNP) (Minuartia, 2020), which is located within tions produce divergent landscapes that promote city-specific phenotypic the Metropolitan Area of Barcelona (MAB), bordering the city of Barcelona traits (Alberti et al., 2017; Ouyang et al., 2018). Management techniques, (BCN; Fig. 1). Concurrently, the presence of wild boar in the urban area of mostly removal of individuals to solve conflict and control population Barcelona has also risen, leading to an increased number of human-wildlife size, may also act as a selective force in both urban and nonurban systems. conflicts (Castillo-Contreras et al., 2018 and 2021; Fernández-Aguilar et al., Conflict frequency and severity determine the intensity of such manage- 2018). ment actions, ranging from no intervention in low-conflict situations to ex- This study aims at investigating the effects of urbanisation and associ- tirpation in high-conflict scenarios. In turn, management drives the ated management measures as drivers of change in genetic diversity in phenotypic population changes, which are therefore more likely to appear urban wild boar populations, by investigating the genetic structure of in moderately conflictive urban wildlife. These phenotypic changes can be both the urban and peri-urban wild boar populations of the MAB and its either responses to urbanisation itself or responses to management mea- neighbouring rural areas. The results will contribute to define wild boar sures (Schell et al., 2020). The removal of individuals to control population as being an urban tolerant or an urban exploiter species, as previously de- size may also exacerbate genetic drift and decrease genetic diversity of fined (Rodewald and Gehrt, 2014). urban populations after the initial bottleneck related to the founder effect (Combs et al., 2018; Edelhoff et al., 2020; Miles et al., 2019; Nei et al., 2. Material and methods 1975). Wild boar (Sus scrofa) is an omnivorous generalist species occupying 2.1. Sampling areas ecosystems from semi-arid deserts to tropical rain forests, and successfully exploits urban habitats (Massei et al., 1996). Across Europe, wild boar pop- Wild boar samples were collected in three focal areas and in three addi- ulations have increased and sprawled, even colonizing urban habitats tional areas connected to the focal areas (Fig. 1). The focal areas were: i) the (Geisser and Reyer, 2005; Snow et al., 2017; Castillo-Contreras et al., peri-urban CNP (41°25′52″N, 2°4′45″E) as part of the MAB; ii) the urban 2018; Zsolnai et al., 2022). Compared with rural wild boar populations, area of BCN (41°23′N, 2°11′E); and iii) the rural SLMO outside the MAB urban and peri-urban areas of European cities such as Barcelona, Berlin, Bu- (41°38′ 29″N, 02°01′05″E). dapest, Genova, Warsaw and others host populations with larger sizes and CNP is part of the Catalan Coastal Mountain Range and is located be- higher densities (Cahill et al., 2012; Castillo-Contreras et al., 2018; Frantz tween the rivers Llobregat in the West and Besòs in the East. Being part of et al., 2012; Podgórski et al., 2013; Stillfried et al., 2017b). The presence the MAB, it is a ‘Natura 2000’ site, surrounded by urban infrastructures of wild boars in urban habitats generates new conflicts, such as general nui- (Cahill and Llimona, 2004). CNP location, natural heritage, vicinity to sance, damage to private and public property (gardens, cemeteries, sport Barcelona and human disturbance of annually up to three million visitors pitches, public parks, etc.), higher risk of disease transmission, increased make CNP a peri-urban habitat (Moll et al., 2019; Parc de Collserola, risk of traffic accidents, and increased public fear of attacks on pets and 2020). The south-eastern boundary of CNP borders the second focal area, even on humans (Conejero et al., 2019; Fernández-Aguilar et al., 2018; the urban area of BCN, a city inhabited by ~1.6 million people with a den- Wang et al., 2019; Castillo-Contreras et al., 2022). sity of 16.149 inhabitants/km2 (Statistical Institute of Catalonia, 2019). The recent colonisation of the urban environment by wild boar and the The third focal area, SLMO, is a ~13,700 ha large rocky mountain massif heterogeneity of the urban environments colonised have so far precluded located ~40 km north-west of BCN and CNP (Fig. 1). the ecological classification of this species as either urban exploiter (able The three additional sampling areas were i) the area between SLMO and to exploit but not dependent on anthropogenic resources in urban areas) BCN/CNP (henceforth ‘Central’), ii) the area south-west of river Llobregat or urban tolerant (able to use anthropogenic resources and residing in (henceforth ‘West’) and iii) the area north-east of river Besòs (henceforth urban landscapes but not generally exploiting urban areas to reach greater ‘East’). densities) (Rodewald and Gehrt, 2014). Although urban wildlife popula- tions can reach higher densities than the rural ones (Luniak, 2004), density 2.2. Sampling may be a misleading measure for this classification. Urban populations may be sinks, with relatively high mortality and/or low reproduction and receiv- Muscle tissue (N = 218), blood (N = 127) and spleen tissue (N = 62) ing a constant gene flow from the rural source population (Pulliam, 1988; from 407 wild boars were collected from 2005 to 2018 in the six sampling Fig. 1. Map of sampling areas and sample locations. Focal areas are encircled by dotted colour lines, brown = Natural Park of Sant Llorenç del Munt i l'Obac (SLMO), black = Collserola (CNP), blue = Barcelona (BCN). The three smaller peripheral sampling areas are labelled ‘West’ (west of river Llobregat), ‘Central’ (between SLMO and CNP), and ‘East’ (east of river Besòs). Bold turquoise line = river Llobregat; bold purple line = river Besòs. The groups of sampled wild boars are presented as pie charts within their sampling location. Pie chart size indicates sample size (SMLO N = 118; CNP N = 112; BCN N = 98; West N = 35; Central N = 37; East N = 7). The colour of the pie chart pieces indicates cluster assignment by STRUCTURE (yellow = urban, dark yellow = urban admixed, dark blue = rural, light blue = rural admixed). The arrows display genetic exchange between neighbouring sampling areas (boldness of arrow reflects number of migrants per generation (NmFST)); numbers beside arrows indicate population pairwise distance (upper = FST; lower = RST)). The barred line between CNP and BCN indicates nearly unhindered gene flow (see also Tables 3 and 4). 2 J. Hagemann et al. Science of the Total Environment 833 (2022) 155126 3 J. Hagemann et al. Science of the Total Environment 833 (2022) 155126 areas (BCN N = 98, CNP N = 112, SLMO N = 118, area ‘Central’ N = 35, A Bayesian genetic clustering method implemented in the software area ‘West’ N = 37, and area ‘East’ N = 7). The dataset contained 18 (4.4%) STRUCTURE v.2.2.4 (Pritchard et al., 2000; Falush et al., 2003, 2007; piglets (wild boar younger than six months). In all the areas but BCN, sam- Hubisz et al., 2009) was applied to assess whether the wild boar meta- ples were collected from hunted wild boars after drive hunts carried out population was at a genotypic equilibrium or descended from more than during the regular hunting season. In BCN, the urban wild boars were cap- one ancestral population. The ‘admixture model’ was used as prior for the tured using teleanaesthesia, drop-nets or corral-traps (Torres-Blas et al., ancestry model with α inferred from the data (αinitial = 1.0) for all the pop- 2020), and euthanized as part of the management plan of the species to re- ulations. The allele frequency model was the ‘correlated allele frequencies duce human-wildlife conflicts in the urban area of Barcelona (Castillo- model’ with the following priors: different FST values for different popula- Contreras et al., 2016, authorization AC/190 from the Generalitat de tions, prior mean FST = 0.01 (sd = 0.05), allele frequency parameter Catalunya). After collection, the samples were stored at −20 °C until anal- λ = 1.0. Ten independent runs were carried out for each K from K = 1 ysis. UTM coordinates were recorded for each sampled individual in BCN, to K = 7 with a final run length of 300,000 Markov-Chain-Monte-Carlo while the centroid of the drive hunt area was assigned as position for the (MCMC) iterations after a burn-in period of 300,000 iterations. Applying wild boar samples collected at drive hunts outside BCN (Fig. 1). the web-tool STRUCTURE HARVESTER (Earl and von Holdt, 2012), the best fitting K was estimated using the ΔK-method (Evanno et al., 2005), based 2.3. Molecular analyses on the log-likelihood values and their convergence associated with each K. After deciding on the most probable K, the average proportional mem- DNA from tissue samples (muscle and spleen) was extracted using the bership (q) in percent per cluster was calculated for each individual by First-DNA all tissue kit (GEN-IAL GmbH, Troisdorf, Germany), following averaging q over all ten runs. The program was applied to four scenarios the manufacturer's extraction protocol with cell lysis carried out overnight (Sc): Sc1) including all individuals (N = 407); Sc2) including all individuals at 65 °C. DNA from blood was isolated using the Quick-gDNA™ Blood but piglets (N = 389); Sc3) as a sequential run to Sc1: if ≥2 clusters will be MiniPrep Kit (Zymo Research Corp, Irvine, CA, USA), according to the man- identified under Sc1, each of these clusters will then be analysed separately ufacturer's protocol. Once extracted, DNA was eluted with ddH2O and its to test for the presence of an underlying population substructure not de- concentration was measured with a Nanodrop ND1000 (PEQLAB tected under Sc1; Sc4) for each of the populations from the three focal sam- Biotechnologie GmbH, Erlangen, Germany). Samples with DNA concentra- pling areas. The individuals were assigned to a cluster based on their tions >100 ng/μl were diluted to 100 ng/μl. Seventeen unlinked polymor- proportional membership (q) for one cluster using a threshold of q ≥ phic microsatellite loci (Rohrer et al., 1994; Stillfried et al., 2017b) were 90%. Following the same strict rule, admixed individuals were defined as amplified in six different multiplex PCRs (Table SM1) following the same individuals with a proportional membership in one cluster of 50% < q < protocol and using the Type-it Microsatellite PCR Kit (QIAGEN GmbH, 90%. Hilden, Germany). Each multiplex PCR was optimized in terms of combina- A Fisher's exact test using the fisher.test function within the R software tions of loci, primer sequences, and fluorescence dyes (Table SM1). Prior to environment (v.3.5.3.2; R Core Team, 2020) was performed to assess differ- the genotyping PCRs. amplification conditions were optimized by perform- ences in the proportion of admixed wild boars and in the proportion of wild ing gradient PCRs on a peqSTAR96 universal gradient thermocycler boars sampled from the area corresponding to their genetic cluster. These (PEQLAB Biotechnologie GmbH, Erlangen, Germany). Initial denaturation values were compared between the four rural populations (SLMO and all was then set at 95 °C for 5 min, followed by 35 cycles of denaturation at three peripheral areas) vs. the two peri-urban and urban populations 94 °C for 30s, annealing at 63 °C–55 °C (1× 63 °C; 1× 61 °C; 1× 59 °C; (BCN and CNP), as well as between the peri-urban and urban populations 1× 57 °C; 31× 55 °C) for 30s and elongation at 72 °C for 30s. Final elonga- (BCN vs. CNP) (Table 1). tion was carried out for 30 min at 60 °C. The PCR products were then sep- arated by electrophoresis on a capillary automated sequencer (A3130xl; 3. Results ABI Applied Biosystems, Foster, CA, USA). Allele sizes were determined using GENEMAPPER v.3.7 based on the GeneScan™ 500 ROX™ size standard Out of all 407 wild boars, 401 were genotyped at all 17 loci, while six (both ABI). To detect potential genotyping errors in the dataset, the soft- had missing data at one locus. All the genotypes were unique. Thus, all ware program MICRO-CHECKER v.2.2.3 (Van Oosterhout et al., 2004) was ap- the wild boars were submitted to subsequent analyses. For the full data plied, using a significance threshold of α = 0.05. The software package set (N = 407), potential presence of null-alleles was detected at one locus CERVUS v.3.0.7 (Marshall et al., 1998; Kalinowski et al., 2007, 2010) was only (Sw936). When applied to the six sampling populations separately, po- used to determine number of alleles per locus (NA) and total number of al- tential presence of null-alleles was detected at several loci: at Sw857 and leles per population (NA*), as well as to test for the presence of Hardy- Sw936 in CNP, at Sw856 in SLMO, at Sw936 in BCN, at Sw155 in ‘West’, Weinberg-equilibrium (HWE) across loci and for the presence of matching at Sw911 and Sw66 in ‘Central’, and at Sw936 in ‘East’ (Table SM1). The genotypes among the 407 wild boars genotyped. Allelic richness AR and in- LRT indicated potential LDs for four pairs of loci in CNP, four in SLMO, breeding coefficient FIS were estimated using the software package FSTAT six in BCN, and none in the three smaller areas (Table SM2). As no apparent v.2.9.4 (Goudet, 1995, 2001). The potential presence of linkage disequilib- distribution pattern emerged for presence of null-alleles or LD between rium (LD) among pairs of loci was determined using the likelihood ratio test pairs of loci, all the 17 loci were kept for downstream analyses. (LRT) implemented in the software package ARLEQUIN v.3.5.2.2 (Excoffier The number of alleles per locus ranged from NA = 3 (Sw72) to NA = 16 and Lischer, 2010), applying a Bonferroni correction for multiple compari- (Sw856), with a mean number of NA(mean) = 7.06 (Table SM3). The num- sons (Bonferroni, 1936). ARLEQUIN was also used to estimate the number of ber of loci deviating from HWE in each sampled population ranged from within population differences (PiX), pairwise population genetic distances, none to three (Table 2), accounting for ten out of the 17 loci deviating based both on different alleles (FST) and on the sum of squared size differ- from HWE for the whole of the sample (N = 407) (Table SM3). This indi- ences of alleles (RST), and to determine the number of migrants (Nm) cated a Wahlund effect, with a population substructure preventing free mat- among populations (Wright, 1950; Reynolds et al., 1983; Weir and ing and leading to a reduction of heterozygosity (Wahlund, 1928), Cockerham, 1984; Slatkin, 1995). suggesting that the whole population (N = 407) was a structured meta- population. Inbreeding was not detected in any of the six populations sam- 2.4. Analysis of population differentiation and structure pled (Table 2). Averaged over all the 17 loci, all six wild boar populations sampled To assess potential gene flow among the wild boar populations from dif- were at HWE (Fig. 2). The 407 wild boars shared NA* = 120 alleles across ferent sampling areas, all the individuals were grouped according to their all the loci. Out of those 120 alleles, the largest sampling population sampling area (henceforth called ‘populations’) and pairwise population ge- (SLMO, N = 118) had also the highest total number of alleles (NA* = netic distances were estimated using FST and RST statistics. 103), while the smallest sampling population (‘East’, N = 7) had the lowest 4 J. Hagemann et al. Science of the Total Environment 833 (2022) 155126 Table 1 Admixed wild boars and assignment for each genetic cluster defined by the STRUCTURE results according to the sampling location. Admixed individuals are those with av- erage membership (q) of the other cluster 50% < q < 90%. Location Admixed/Sampled (%) Urban cluster Rural cluster Total (%) 95% CI Admixed (%) Total 95% CI Admixed (%) BCN 9/98 (9.2)x 92/98 (93.9)x 89.1–98.6 5/92 (5.4) 6/98 (6.1)x 1.4–10.9 4/6 (66.7) CNP 23/112 (20.5)y 93/112 (83.0)y 76.1–90.0 10/93 (10.8) 19/112 (17.0)y 10.0–23.9 13/19 (68.4) Subtotal urban locations 32/210 (15.2)a 185/210 (88.1)a 83.7–92.5 15/185 (8.1) 25/210 (11.9)a,⁎ 7.5–16.3 17/25 (68.0) SLMO 7/118 (5.9) 2/118 (1.7) 0.0–4.0 2/2 (100.0) 116/118 (98.3) 96.0–100.0 5/116 (4.3) West 1/35 (2.9) 0/35 (0.0) – 0/0 35/35 (100.0) – 1/35 (2.9) Central 4/37 (10.8) 1/37 (2.7) 0.0–7.9 1/1 (100.0) 36/37 (97.3) 92.1–100.0 3/36 (8.3) East 1/7 (14.3) 0/7 (0.0) – 0/0 7/7 (100.0) – 1/7 (14.3) Subtotal rural locations 13/197 (6.6)b 3/197 (1.5)b,⁎ 0.0–3.2 3/3 (100.0) 194/197 (98.5)b 96.8–100.0 10/194 (5.2) Total 45/407 (11.1) 188/407 (46.2) 41.3–51.0 18/188 (9.6) 219/407 (53.8) 49.0–58.7 27/219 (12.3) BCN = City of Barcelona; CNP = Collserola Natural Park; SLMO = Natural Park of Sant Llorenç del Munt i l'Obac; West = additional sampling area southwest of Llobregat river; Central = additional sampling area between SLMO and BCN/CNP; East = additional sampling area Northeast of Besòs river; CI = confidence interval. a,b: values with different superscript were significantly (p < 0.05) different between urban and rural locations. x,y: values with different superscript were significantly (p < 0.05) different between both urban locations (Barcelona and Collserola). ⁎ The proportion of wild boars from the rural cluster in urban locations was significantly (p < 0.05) higher than the proportion of wild boars from the urban cluster in rural locations. number (NA* = 54, Table 2). However, when considering the ratio R differentiation was detected between BCN and CNP (FST/RST = 0.006/ between the total number of alleles and the number of individuals [R = 0.003; Fig. 1, Table S3). This low differentiation indicated strong gene NA*/N] as a measure of genetic diversity (Fickel et al., 2005), ‘East’ had flow, with the highest number of migrants per generation (Nm) exchanged the highest value (R = 7.714), whereas CNP had the lowest value (R = between both subpopulations (Nm(FST/RST) ~ 41/82; Table SM5, Fig. 1). 0.678; Table 2). Allelic richness across all 17 loci was highest in ‘West’ These urban BCN and peri-urban CNP sampling populations were also ge- (AR = 3.78; based on a mean number of alleles per locus of NA/L = netically different from rural ‘West’ (separated by the Llobregat river; FST/ 5.18) and lowest in BCN (AR = 2.65; based on NA/L = 4.06). Even the RST = 0.169/0.147 and 0.133/0.110, respectively; Fig. 1, Table 3). Differ- smallest sampling population (‘East’) had a higher allelic richness (AR = entiation between CNP and ‘Central’ (FST/RST = 0.098/0.074) and be- 3.29) than BCN. As this was based on NA/L = 3.6 only, alleles were more tween CNP and SLMO (0.108/0.090) were similar to the differentiation equally distributed in ‘East’ than in BCN (Table 2, Fig. 2). Regarding the of CNP with ‘West’ (FST/RST = 0.133/0.110) and between the lower differ- differentiation among individuals within a population, we observed the entiation values of CNP to BCN (FST/RST = 0.006/0.003) and the higher highest number of within-population differences (PiX; Table SM4) in differentiation of CNP to ‘East’ (FST/RST = 0.256/0.227). Both CNP and ‘Central’ (PiX(FST/RST) = 9.97/1700.1), while they were lowest in BCN BCN differed from all the rural sampling populations, including ‘Central’, (PiX(FST/RST) = 7.41/1072.7). more than urban and rural sampling populations differed among them. Pairwise population differentiations among rural sampling populations 3.1. Population differentiation and structure were generally low (Fig. 1, Table 3). In STRUCTURE run scenario Sc1 (N = 407), the ΔK estimation approach Genetic population pairwise distances (FST and RST; Table 3) (Earl and von Holdt, 2012) indicated K = 2 (ΔK = 992.3) as the most likely indicated the highest differentiation between urban BCN and rural ‘East’ number of ancestral genotypic clusters for the sampled wild boars (FST/RST = 0.256/0.227), which also had the lowest number of migrants (Table SM6). One of the two clusters (‘rural’) consisted mainly of wild per generation (Nm) exchanged between any two sampling populations boars sampled in the four rural areas (N = 219, including 27 admixed indi- (Nm(FST/RST) = 0.72/0.85; Table SM5, Fig. 1). The second highest differen- viduals), while the other one (‘urban’) was formed mostly by wild boars tiation was observed between peri-urban CNP and ‘East’ (FST/RST = 0.202/ from the two urban/peri-urban locations (BCN and CNP N = 188, includ- 0.16). Both findings indicated that the Besòs river difficulted gene flow be- ing 18 admixed individuals; Fig. 1, Tables 1 and 4, Fig. SM1). The analysis tween the rural ‘East’ and the urban/peri-urban BCN and CNP. The smallest without piglets (STRUCTURE run scenario Sc2) did not change the outcome (Table SM6), with the same two ancestral genotypic clusters (K = 2, ΔK = 645.02) and the remaining 389 wild boars still assigned to the same cluster as in scenario Sc1. Table 2 The mean values for the genetic diversity indices AR, Hobs, and Hexp Population indices across 17 loci and sampling areas analysed in 407 wild boars (6.82/ 0.56/0.59) and the mean number of alleles per locus NA(mean) from the Metropolitan area of Barcelona (MAB) and its rural surroundings. (6.82) were higher in the rural cluster than in the urban one (AR = 3.65; Pop N NA* R [NA*/N] NA/L FIS Hobs = 0.451; Hexp = 0.451; NA(mean) = 3.56; Table 4, Fig. 3. Out of the CNP 112 76 0.678 4.471 0.004 120 alleles detected among all the wild boars (N = 407; Table 2), 116 SLMO 118 103 0.873 6.059 0.024 were detected in the rural cluster (including admixed individuals: NA = BCN 98 69 0.704 4.059 −0.001 119) and 61 in the urban one (including admixed individuals: NA = 67; ‘West’ 35 88 2.514 5.176 0.020 Table 4). ‘Central’ 37 92 2.486 5. 412 0.062 ‘East’ 7 54 7.714 3.6 0.025 all 407 120 0.295 7.059 0.022 3.2. Wild boar origin and admixture Pop: population, N: population size, NA*: total number of alleles across all loci, R: Most of the wild boars belonged to the cluster assigned to their areas, ratio between total number of alleles and number of individuals, NA/L: mean num- ber of alleles per locus, FIS: inbreeding coefficient (none was significant). CNP = i.e., 88.1% (185/210) of the wild boars sampled in BCN and CNP belonged Collserola Natural Park; SLMO = Natural Park of Sant Llorenç del Munt i l'Obac; to the urban cluster and 98.5% (194/197) of the wild boars sampled in BCN = Barcelona; ‘West’ = additional sampling area southwest of Llobregat rural locations were assigned to the rural cluster (both including admixed river; ‘Central’ = additional sampling area between SLMO and BCN/CNP; individuals). The percentage was significantly (p < 0.05) higher for rural lo- ‘East’ = additional sampling area Northeast of Besòs river. cations, meaning that detecting rural wild boars in urban locations (11.9%) 5 J. Hagemann et al. Science of the Total Environment 833 (2022) 155126 Fig. 2. Population indices across 17 loci and sampling areas analysed in 407 wild boars from the Metropolitan area of Barcelona (MAB) and its rural surroundings. Allelic richness across loci based on populations size of area, observed heterozygosity across all loci, expected heterozygosity across all loci.Hardy-Weinberg equilibrium + in front of names: population is at HWE, −: population is not at HWE, p values averaged over 17 loci were: p(CNP) = 0.398, p(SLMO) = 0.354, p(BCN) = 0.555, p(West) = 0.346, p(Central) = 0.509, p(East) = 0.662, p(Total) = 0.0046. CNP = Collserola Natural Park; SLMO = Natural Park of Sant Llorenç del Munt i l'Obac; BCN = Barcelona; ‘West’ = additional sampling area southwest of Llobregat river; ‘Central’ = additional sampling area between SLMO and BCN/CNP; ‘East’ = additional sampling area Northeast of Besòs river.; “Total” = all samples combined. was more likely than finding urban wild boars in rural locations (1.5%) (23/112, 20.5%) than in BCN (9/98, 9.2%; Table 1). The STRUCTURE analy- (Table 1). The numbers of admixed individuals (0.5 < q < 0.9) were signif- ses for the urban and rural clusters (STRUCTURE run scenario S3), as well as icantly higher in the two urban locations (32/210 = 15.2%) than in the for the three main sampling populations (CNP, BCN, and SLMO) (STRUCTURE four rural ones (13/197 = 6.6%; Table 1). Moreover, the proportion of run scenario S4), revealed weak indication (ΔK = 95.44) for an underlying admixed wild boars was ten-fold and significantly (p < 0.05) higher in structure only in CNP (N = 112). This further supports the previously de- wild boars sampled outside of their corresponding area (dispersers, 18/ tected structure separating wild boars in the rural and urban clusters 28, 64.3%) than in those wild boars where sampling location corresponded (Table SM6). to cluster assignment (residents, 25/379, 6.6%). All urban dispersers in rural locations were admixed (3/3, 100%), whereas in the urban locations 4. Discussion (BCN and CNP) 32% of the dispersers (8/25) belonged to the rural cluster with q ≥ 0.9 (Table 1). Within the urban locations, the proportion of This study sheds light on the colonisation process of the urban environ- wild boars assigned to the urban cluster was significantly (p < 0.05) higher ment by wild boar and the effects of both synurbisation and the associated in BCN (92/98, 93.9%) than in CNP (93/112, 83.0%). Conversely, the management measures on the genetic composition of urban wild boar pop- admixed wild boars were significantly (p < 0.05) more frequent in CNP ulations, defining the ecological relationship of wild boar with urban areas. The urban genetic cluster originated from dispersers from the neighbouring rural areas, as evidenced by the unidirectional gene flow from rural to Table 3 urban areas, the lower genetic diversity of the urban populations, the Population pairwise distances (FST, RST) among the six wild boar populations sam- higher number of rural dispersers in urban areas than urban dispersers in pled in the Metropolitan area of Barcelona (MAB) and its rural surroundings. rural areas, and the higher proportion of admixed individuals in urban Area CNP SLMO BCN ‘West’ ‘Central’ ‘East’ than in rural areas (Table 1). Although the admixed individuals in CNP CNP – 0.108 0.006 0.133 0.098 0.201 and BCN indicated that hybridization and gene flow between both clusters SLMO 0.09 – 0.145 0.043 0.016 0.086 occurs, such gene flow was not enough to prevent genetic differentiation BCN 0.003 0.124 – 0.169 0.138 0.256 between the urban and rural populations. Fragmentation of natural envi- ‘West’ 0.11 0.021 0.147 – 0.038 0.089 ‘Central’ 0.074 0.008 0.106 0.026 – 0.078 ronments is a recognised cause of genetic differentiation. However, founder ‘East’ 0.16 0.042 0.227 0.031 0.041 – effect, source-sink dynamics, and/or lack of suitable habitat for dispersers related to urbanisation make synurbisation a more powerful driver of ge- Above diagonal: FST (based on number of different alleles), below diagonal: RST netic differentiation than habitat fragmentation (Wood and Pullin, 2002; (based on sum of squared size differences of alleles). Underlined values indicate sig- nificant differentiation (p < 0.05). Bold: highest and lowest value in the correspond- Wandeler et al., 2003; Schell et al., 2020). The genetic differentiation of ing comparison. CNP = Collserola Natural Park; SLMO = Natural Park of Sant an urban and a rural cluster, the lower genetic diversity of the urban cluster, Llorenç del Munt i l'Obac; BCN = City of Barcelona; ‘West’ = additional sampling and the lower proportion of admixed wild boars and wild boars from the area southwest of Llobregat river; ‘Central’ = additional sampling area between rural cluster in BCN indicate gene-flow barriers and genetic isolation. Alto- SLMO and BCN/CNP; ‘East’ = additional sampling area Northeast of Besòs river. gether, they suggest that the urban wild boar population in the MAB was 6 J. Hagemann et al. Science of the Total Environment 833 (2022) 155126 Table 4 Little (FST = 0.006, p < 0.05; number of migrants based on FST ~ 41) to Molecular indices per cluster (rural, urban) with and without admixed individuals no genetic differentiation (RST = 0.003, p > 0.05; number of migrants for the wild boar populations sampled in the Metropolitan area of Barcelona based on RST ~ 82), as previously defined (Hartl and Clark, 1997), was (MAB) and its rural surroundings. found between the urban wild boars in CNP and BCN (Fig. 1, Tables 3 Cluster N NA* NA AR Hobs (SD) Hexp (SD) HWE (p) FIS and SM5), although wild boars from BCN have been reported to grow faster Rural 192 116 6.82 6.82 0.560 0.590 +(0.225) 0.051 and to be heavier and bigger than their conspecifics from CNP (Castillo- (0.202) (0.203) Contreras et al., 2021). Rural (with 219 119 7.00 6.99 0.564 0.590 +(0.248) 0.043 This phenotypic changes could be the result of transgenerational plastic- admixed†) (0.200) (0.200) ity (TGP), whereby the phenotype of a generation (in this case, bolder be- Urban 170 61 3.81 3.65 0.451 0.451 +(0.421) −0.001 (0.179) (0.177) haviour, consumption of anthropogenic food resources, and faster growth Urban (with 188 67 4.12 3.99 0.462 0.459 +(0.418) −0.006 and weight gain) is influenced by the environment experienced by former admixed†) (0.175) (0.172) generations (Bell and Hellmann, 2019). Although a genetic basis has been † : admixed individuals with proportional membership in the indicated cluster of proposed for the changes observed in urban wild boar populations 0.90 > q > 0.50; N: number of individuals; NA*: total number of alleles; NA: average (Zsolnai et al., 2022), plasticity can overcome the constraints of adaptive number of alleles across loci; AR: allelic richness; Hobs: observed heterozygosity; genetic change by decoupling the genotype from phenotype Hexp: expected heterozygosity; SD: standard deviation, HWE: Hardy-Weinberg equi- (Bonduriansky and Day, 2009), and TGP can promote population persis- librium; +: individuals in cluster are at HWE; p: values were calculated across all 17 tence over generations in changing environments (Jablonka et al., 1995; loci. FIS: inbreeding coefficient. Pal, 1998; West-Eberhard, 2003; Weyrich et al., 2019). Therefore, disentangling whether changes in phenotypic traits are due to phenotypic plasticity or genetic changes is essential in urban evolutionary ecology established through a founder effect, supporting the UIH (Gloor, 2002; (Alberti et al., 2017; Ouyang et al., 2018; Schell, 2018) and to devise the Wandeler et al., 2003), as previously reported for other urban wild boar most effective management strategy (Lambert and Donihue, 2020). If populations (Stillfried et al., 2017b; Zsolnai et al., 2022). these urban phenotypic traits have a genetic basis, the management actions Once established in the urban and peri-urban areas, both urbanisation against them will be more efficient than if they are explained by TGP. The and the associated management measures shape the phenotypic changes genetic structure of peri-urban and urban wild boar populations in this and genetic structure of synurbic populations (Rodewald and Gehrt, study and the phenotypic changes previously reported (Castillo-Contreras 2014; Schell et al., 2020). The higher proportions of admixed (20.5%) et al., 2021), do not only allow to understand the synurbisation process of and rural (17.0%) wild boars in CNP as compared to BCN (9.2% and wild boar, but emphasize the need of future epigenetic studies to further 6.1%, respectively; Table 1) suggest that admixture between clusters likely clarify the relationship between phenotypic changes, genetic changes and occurs in CNP rather than in BCN, with bolder individuals from the peri- potential TGP. urban CNP daring to explore the urban area of BCN (Castillo-Contreras The phenotypic changes observed in the urban wild boars from BCN et al., 2018), as reported for urban wild boar and deer (Stillfried et al., (Castillo-Contreras et al., 2021) without genetic differentiation from adja- 2017c; Honda et al., 2018; Zsolnai et al., 2022). This corresponds to a cent populations correspond to a situation of frequent conflict with moder- source-to-sink dynamics from peri-urban CNP to urban BCN as described ate management measures, which exert a selection pressure on bolder by the PPH (Pulliam, 1988; Delibes et al., 2001). phenotypes but do not eradicate the species from the area (Stillfried et al., Fig. 3. Allelic richness, expected heterozygosity and observed heterozygosity of the wild boars from the rural and urban clusters sampled in the Metropolitan Area of Barcelona and its rural surroundings. 7 J. Hagemann et al. Science of the Total Environment 833 (2022) 155126 2017c; Honda et al., 2018; Schell et al., 2020). In Barcelona, conflictive Urban environment and population management act as phenotypic and urban wild boars are individually removed (Castillo-Contreras et al., genetic evolution drivers on urban wild boar populations, which have re- 2016), which may act as a management selection driver (Honda et al., duced genetic variability due to the bottleneck related to the founder effect. 2018; Schell et al., 2020). However, culled individuals might be replaced As a consequence, urban wild boar populations differentiate quickly both by the next boldest individuals from the CNP population (Found and St. phenotypically and genetically from rural populations, with lower genetic Clair, 2019), where removal pressure is lower than in all other rural and variability and reduced allelic richness. The intensity of such differentiation urban areas due to legal and practical limitations, respectively (Minuartia, depends on the strength of genetic drift combined with the strength of se- 2020). Nevertheless, if the bolder behaviour of BCN wild boars is lection pressure posed by the urban environment and the associated man- a phenotypic traits derived from TGP (Bell and Hellmann, 2019), the re- agement measures. The relation between the species phenotypic plasticity moval of conflictive individuals could break up such TGP, taking some (including TGP) and the selection pressure exerted by the urban environ- time for the plastic behaviour to reappear and spread among the wild ment and the management measures will determine the responses of the boar population. urban wildlife population, ranging from the establishment of urban popula- Altogether, the assignment of peri-urban and urban wild boars in a tions phenotypically and/or genetically different from rural populations to common genetic cluster with a source-to-sink dynamics between both the eradication of the population through high management pressure. locations, and the behavioural adaptions of urban wild boars to the urban Supplementary data to this article can be found online at https://doi. environment, selecting for anthropogenic food resources (Castillo- org/10.1016/j.scitotenv.2022.155126. Contreras et al., 2021) and with a bolder behaviour (Stillfried et al., 2017c), classify wild boar as an urban exploiter species (Rodewald and CRediT authorship contribution statement Gehrt, 2014). Wildlife responses to urbanisation are associated to landscape-scale dif- Justus Hagemann: Formal analysis, Investigation, Methodology, ferences among cities, causing within-city and inter-city differences. Thus, Software, Writing – original draft. city structure (including the density of human households and the propor- Carles Conejero: Data curation, Investigation, Methodology, Writing – tion and distribution of green areas within the urban landscape) conditions review & editing. both wildlife occupancy and population demography (Fidino et al., 2021). Milena Stillfried: Conceptualization, Formal analysis, Methodology, Forested areas (e.g. CNP) allow the establishment of island populations Writing – review & editing. (UIH) as a transition from rural habitats, whereas purely built-up areas Gregorio Mentaberre: Data curation, Funding acquisition, Investiga- act as an attractive sink, mainly because of anthropogenic food resources tion, Methodology, Writing – review & editing. (Castillo-Contreras et al., 2018 and 2021; Delibes et al., 2001; Stillfried Raquel Castillo-Contreras; Data curation, Investigation, Methodol- et al., 2017b) and probably a higher mortality rate. Connectivity among for- ogy, Writing – review & editing. ested areas determines gene flow and genetic structure (Munshi-South, Jörns Fickel: Conceptualization, Formal analysis, Funding acquisition, 2012). Thus, the distribution and connectivity of green areas in cities deter- Investigation, Methodology, Resources, Supervision, Validation, Visualiza- mine the existence of different genetic clusters in urban wild boar popula- tion, Writing – review & editing. tions, which evolve by spatial isolation and genetic drift (Stillfried et al., Jorge Ramón López-Olvera; Conceptualization, Data curation, 2017b; Wandeler et al., 2003; Zsolnai et al., 2022). In BCN the single Funding acquisition, Investigation, Project administration, Resources, Su- urban wild boar cluster corresponds to the single green area concentrated pervision, Validation, Visualization, Roles/Writing – original draft. in CNP. The connectivity among the rural areas surrounding the MAB, as in- dicated by the little genetic differentiation among them (Fig. 1), suggests Declaration of competing interest that rivers do not pose a barrier for gene flow among wild boar populations, agreeing with previous reports (Frantz et al., 2012; Leaper et al., 1999). The authors declare that they have no known competing financial inter- However, the urban structures around both rivers could limit gene flow be- ests or personal relationships that could have appeared to influence the tween the rural areas beyond them and the urban locations in the MAB, work reported in this paper. closer to the city (Fig. 1, Tables 3 and SM5), as suggested by the genetic differentiation between the areas beyond the rivers and the urban areas, Acknowledgements as previously reported (Zsolnai et al., 2022). When disentangling the effects of urban areas and rivers in preventing connectivity among urban and peri- The authors are grateful to the Ajuntament de Barcelona, who provided urban wild boar populations, the type of waterflow (i.e., Mediterranean funding for the field sampling through the contracts 13/051, 15/0174, stream or Central European river) and the urban structure (presence of for- 16/0243 and 16/0243-00-PR/01. This study benefitted from the Research ested patches and connectivity among them) must be considered. Further Grant PID2020-115046GB-I00 Ecología, salud pública y gestión del jabalí monitoring of the wild boar populations in the MAB (even non-invasively; urbano, funded by the Spanish Ministerio de Ciencia e Innovación. The Fickel and Hohmann, 2006) should contribute to monitor the colonisation Secretaria d'Universitats i Recerca del Departament d'Economia i Coneixement of the urban area and potential genetic drift or clustering in other forested de la Generalitat de Catalunya and the European Social Fund funded Raquel patches. Castillo-Contreras through the grant numbers 2016FI_B 00425, 2017FI_B1 00040 and 2018FI_B2_00030. The authors are also indebted to Marià Martí, 5. Conclusion Lluís Cabañeros and the rest of the Consorci del Parc Natural de la Serra de Collserola members for their logistic support. Sample collection and the The wild boar population of Barcelona originated from founder individ- related information on location would not have been possible without the uals from the surrounding rural areas. The limited gene flow between the essential collaboration of the local hunters from Collserola, especially Jesús founders and their descendants, on the one hand, and the source popula- Escarrà, Pere Martín, Joan Carles Montagut, Àngel Obiols and Jorge tion, on the other, led to the differentiation of the wild boar population in Sánchez, and the hunter associations of the MAB. We also wish to thank two genetically distinct clusters, a rural and an urban one. As the degree Josep Maria López Martín and Joan Roldán, from the Departament of admixture between both clusters was higher in the peri-urban area d'Agricultura, Ramaderia, Pesca i Alimentació (Generalitat de Catalunya), for than in the urban area, wild boar can be classified as an urban exploiter spe- serving as a liaison among public administration, management, hunters cies. The presence of suitable habitat patches in the urban area and the con- and research. We are also thankful to the many collaborators and colleagues nectivity both among them and with the source rural population of the Servei d'Ecopatologia de Fauna Salvatge (SEFaS) who helped in the wild determines the gene flow among urban and rural populations and the fu- boar data and sample collection. Finally, we thank A. Schmidt for her help ture genetic differentiation in one or more urban wild boar clusters. with the laboratory analyses. The authors declare no conflict of interest. 8 J. Hagemann et al. Science of the Total Environment 833 (2022) 155126 References among cities alter common species’ responses to urbanization. Ecol. Appl. 31 (2), e02253. https://doi.org/10.1002/eap.2253. Fischer, J.D., Schneider, S.C., Ahlers, A.A., Miller, J.R., 2015. Categorizing wildlife responses Alberti, M., Correa, C., Marzluff, J.M., Hendry, A.P., Palkovacs, E.P., Gotanda, K.M., Hunt, to urbanization and conservation implications of terminology. Conserv. Biol. 29 (4), V.M., Apgar, T.M., Zhou, Y., 2017. Global urban signatures of phenotypic change in ani- 1246–1248. https://doi.org/10.1111/cobi.12451. mal and plant populations. Proc. Natl. Acad. Sci. U. S. A. 114 (34), 8951–8956. https:// Found, R., St. Clair, C.C., 2019. Influences of personality on ungulate migration and manage- doi.org/10.1073/pnas.16060 34114. ment. Front. Ecol. Evol. 7 (November), 1–11. https://doi.org/10.3389/fevo.2019.00438. Bell, A.M., Hellmann, J.K., 2019. An integrative framework for understanding the mecha- Frantz, A.C., Bertouille, S., Eloy, M.C., Licoppe, A., Chaumont, F., Flamand, M.C., 2012. Com- nisms and multigenerational consequences of transgenerational plasticity. Annu. Rev. parative landscape genetic analyses show a belgian motorway to be a gene flow barrier Ecol. Evol. Syst. 50, 97–118. https://doi.org/10.1146/annurev-ecolsys-110218-024613. for red deer (Cervus elaphus), but not wild boars (Sus scrofa). Mol. Ecol. 21 (14), Bonduriansky, R., Day, T., 2009. Nongenetic inheritance and its evolutionary implications. 3445–3457. https://doi.org/10.1111/j.1365-294X.2012.05623.x. Annu. Rev. Ecol. Evol. Syst. 40, 103–125. https://doi.org/10.1146/annurev.ecolsys.39. Geisser, H., Reyer, H.U., 2005. The influence of food and temperature on population density 110707.173441. of wild boar sus scrofa in the Thurgau (Switzerland). J. Zool. 267 (1), 89–96. https://doi. Bonferroni, C.E., 1936. Teoria statistica delle classi e calcolo delle probabilità (vol. 8, pp). org/10.1017/S095283690500734X. Libreria internazionale Seeber. https://books.google.de/books?id=3CY-HQAACAAJ. Gloor, S., 2002. The rise of urban foxes (Vulpes vulpes) in Switzerland and ecological and par- Cahill, S., Llimona, F., 2004. Demographics of a wild boar sus scrofa linnaeus, 1758 popula- asitological aspects of a fox population in the recently colonised city of Zürich. PhD tion in a metropolitan park in Barcelona. Galemys 16 (1), 37–52. https://dialnet. Dissertation, 116. http://rokabaratklub.hu/system/files/attachment/file/document/ unirioja.es/servlet/articulo?codigo=1235997&info=resumen&idioma=SPA. phd_gloor2002.pdf%5Cnhttp://www.swild.ch/gloor/phd/PhD_Gloor2002.pdf. Cahill, S., Llimona, F., Cabañeros, L., Calomardo, F., 2012. Characteristics of wild boar (Sus González-Crespo, C., Serrano, E., Cahill, S., Castillo-Contreras, R., Cabañeros, L., López- scrofa) habituation to urban areas in the collserola Natural Park (Barcelona) and compar- Martín, J.M., Roldán, J., Lavín, S., López-Olvera, J.R., 2018. Stochastic assessment of ison with other locations. Anim. Biodivers. Conserv. 35 (2), 221–233. management strategies for a Mediterranean peri-urban wild boar population. PLoS Castillo-Contreras, R., Fernández-Aguilar, X., Mentaberre, G., Colom-Cadena, A., Lavín, ONE 13 (8), 1–19. https://doi.org/10.1371/journal.pone.0202289. S., López-Olvera, J.R., 2016. Operating protocol for urban wild boar incidences in Goudet, J., 1995. FSTAT (version 1.2): a computer program to calculate F-statistics. J. Hered. the urban area of Barcelona. ISBN 978-3-9815637-3-3In: Schumann, A., Wibbelt, 86 (6), 485–486. https://doi.org/10.1093/oxfordjournals.jhered.a111627. G., Greenwood, A.D., Hofer, H. (Eds.), Contributions to the 12th Conference of the Goudet, J., 2001. FSTAT, version 2.9.3.2: a computer program to calculate F-Statistics. Re- European Wildlife Disease Association (EWDA), August 27th – 31st, 2016, Berlin, trieved from http://www2.unil.ch/popgen/softwares/fstat.htm. p. 107. http://www.izw-berlin.de/tl_files/downloads/EWDA_2016/EWDA2016_P Hartl, D.L., Clark, A.G., 1997. Principles of Population Genetics. 3rd edn. Sinauer Associates, roceedings.pdf. Inc, Sunderland, MA, USA. Castillo-Contreras, R., Carvalho, J., Serrano, E., Mentaberre, G., Fernández-Aguilar, X., Colom, Honda, T., Iijima, H., Tsuboi, J., Uchida, K., 2018. A review of urban wildlife management A., González-Crespo, C., Lavín, S., López-Olvera, J.R., 2018. Urban wild boars prefer from the animal personality perspective: the case of urban deer. Sci. Total Environ. fragmented areas with food resources near natural corridors. Sci. Total Environ. 615, 644, 576–582. https://doi.org/10.1016/j.scito tenv.2018.06.335. 282–288. https://doi.org/10.1016/j.scitotenv.2017.09.277. Hubisz, M.J., Falush, D., Stephens, M., Pritchard, J.K., 2009. Inferring weak population struc- Castillo-Contreras, R., Mentaberre, G., Fernández, Aguilar X., Conejero, C., Colom-Cadena, A., ture with the assistance of sample group information. Mol. Ecol. Resour. 9 (5), Ráez-Bravo, A., González-Crespo, C., Espunyes, J., Lavín, S., López-Olvera, J.R., 2021. 1322–1332. https://doi.org/10.1111/j.1755-0998.2009.02591.x. Wild boar in the city: phenotypic responses to urbanisation. Sci. Total Environ. 773, Jablonka, E., Oborny, B., Molnár, I., Kisdi, É., Hofbauer, J., Czáránet, T., 1995. The adaptive 145593. https://doi.org/10.1016/j.scitotenv.2021.145593. advantage of phenotypic memory in changing environments. Philos. Trans. R. Soc., B Castillo-Contreras, R., Marín, M., López-Olvera, J.R., Ayats, T., Fernández Aguilar, X., Lavín, 350, 133–141. https://doi.org/10.1098/rstb.1995.0147. S., Mentaberre, G., Cerdà-Cuéllar, M., 2022. Zoonotic campylobacter spp. and Salmonella Kalinowski, S.T., Taper, M.L., Marshall, T.C., 2007. Revising how the computer program spp. carried by wild boars in a metropolitan area: occurrence, antimicrobial susceptibility CERVUS accommodates genotyping error increases success in paternity assignment. and public health relevance. Sci. Total Environ. 822, 153444. https://doi.org/10.1016/j. Mol. Ecol. 16 (5), 1099–1106. https://doi.org/10.1111/j.1365-294X.2007.03089.x. scitotenv.2022.153444. Kalinowski, S.T., Taper, M.L., Marshall, T.C., 2010. Erratum: revising how the computer pro- Combs, M., Byers, K.A., Ghersi, B.M., Blum, M.J., Caccone, A., Costa, F., Munshi-South, J., gram CERVUS accommodates genotyping error increases success in paternity assignment 2018. Urban rat races: spatial population genomics of brown rats (Rattus norvegicus) (Molecular ecology (2007) 16 (1099–1106)). Mol. Ecol. 19 (7), 1512. https://doi.org/10. compared across multiple cities. Proc. R. Soc. B Biol. Sci. 285 (1880), 20180245. 1111/j.1365-294X.2010.04544.x. https://doi.org/10.1098/rspb.2018.0245. Kotulski, Y., Koenig, A., 2008. Conflicts, crises and challenges: wild boar in the Berlin City - a Conejero, C., Castillo-Contreras, R., González-Crespo, C., Serrano, E., Mentaberre, G., Lavín, social empirical and statistical survey. Nat. Croat. 17 (4), 233–246. https://doi.org/york. S., López-Olvera, J.R., 2019. Past experiences drive citizen perception of wild boar in kotulski@gmx.de. urban areas. Mamm. Biol. 96 (1), 68–72. https://doi.org/10.1016/j.mambio.2019.04. Lambert, M.R., Donihue, C.M., 2020. Urban biodiversity management using evolutionary 002. tools. Nat. Ecol. Evol. 4 (7), 903–910. https://doi.org/10.1038/s4155 9-020-1193-7. Delibes, M., Ferreras, P., Gaona, P., 2001. Attractive sinks, or how individual behavioural de- Leaper, R., Massei, G., Gorman, M.L., Aspinall, R., 1999. The feasibility of reintroducing wild cisions determine source-sink dynamics. Ecol. Lett. 4 (5), 401–403. https://doi.org/10. boar (Sus scrofa) to Scotland. Mammal Rev. 29 (4), 239–258. https://doi.org/10.1046/j. 1046/j.1461-0248.2001.00254.x. 1365-2907.1999.2940239.x. Earl, D.A., von Holdt, B.M., 2012. STRUCTURE HARVESTER: a website and program for visu- Luniak, M., 2004. Synurbisation – adaption of animal wildlife to urban development. In: alizing STRUCTURE output and implementing the evanno method. Conserv. Genet. Shaw, K.K., Harris, K.L., Van Druff, L. (Eds.), Proceedings of the 4th International Resour. 4 (2), 359–361. https://doi.org/10.1007/s12686-011-9548-7. Urban Wildlife Symposium. University of Arizona, Tucson, Arizona, USA. Edelhoff, H., Zachos, F.E., Fickel, J., Epps, C.W., Balkenhol, N., 2020. Genetic analysis of red Marshall, T.C., Slate, J., Kruuk, L.E.B., Pemberton, J.M., 1998. Statistical confidence for deer (Cervus elaphus) administrative management units in a human-dominated land- likelihood-based paternity inference in natural populations. Mol. Ecol. 7 (5), 639–655. scape. Conserv. Genet. 21 (2), 261–276. https://doi.org/10.1007/s10592-020-01248-8. https://doi.org/10.1046/j.1365-294x.1998.00374.x. Evanno, G., Regnaut, S., Goudet, J., 2005. Detecting the number of clusters of individuals Massei, G., Genov, P.V., Staines, B.W., 1996. Diet, food availability and reproduction of wild using the software STRUCTURE: a simulation study. Mol. Ecol. 14 (8), 2611–2620. boar in a Mediterranean coastal area. Acta Theriol. 41, 307–320. https://doi.org/10. https://doi.org/10.1111/j.1365-294X.2005.02553.x. 4098/AT.arch.96-29. Excoffier, L., Lischer, H.E.L., 2010. Arlequin suite ver 3.5: a new series of programs to perform Miles, L.S., Rivkin, L.R., Johnson, M.T.J., Munshi-South, J., Verrelli, B.C., 2019. Gene flow population genetics analyses under linux and windows. Mol. Ecol. Resour. 10 (3), and genetic drift in urban environments. Mol. Ecol. 28 (18), 4138–4151. https://doi. 564–567. https://doi.org/10.1111/j.1755-0998.2010.02847.x. org/10.1111/mec.15221. Falush, D., Stephens, M., Pritchard, J.K., 2003. Inference of population structure using Minuartia, 2020. Programa de seguiment de les poblacions de senglar a Catalunya. multilocus genotype data: linked loci and correlated allele frequencies. Genetics 164 Temporada 2019-2020. Retrieved from http://agricultura.gencat.cat/web/.content/06- (4), 1567–1587. https://doi.org/10.1111/j.1471-8286.2007.01758.x. medi-natural/caca/enllacos-documents/informes-tecnics/programa-seguiment- Falush, D., Stephens, M., Pritchard, J.K., 2007. Inference of population structure using poblacions-senglar-sus-scrofa/fitxers-binaris/seguiment_senglar_cat_2019-20-WEB.pdf. multilocus genotype data: dominant markers and null alleles. Mol. Ecol. Notes 7 (4), Moll, R.J., Cepek, J.D., Lorch, P.D., Dennis, P.M., Tans, E., Robison, T., Millspaugh, J.J., 574–578. https://doi.org/10.1111/j.1471-8286.2007.01758.x. Montgomery, R.A., 2019. What does urbanization actually mean? A framework for Fernández-Aguilar, X., Gottschalk, M., Aragon, V., Càmara, J., Ardanuy, C., Velarde, R., urban metrics in wildlife research. J. Appl. Ecol. 56 (5), 1289–1300. https://doi.org/ Galofré-Milà, N., Castillo-Contreras, R., López-Olvera, J.R., Mentaberre, G., Colom- 10.1111/1365-2664.13358. Cadena, A., Lavín, S., Cabezón, O., 2018. Urban wild boars and risk for zoonotic Strepto- Munshi-South, J., 2012. Urban landscape genetics: canopy cover predicts gene flow between coccus suis, Spain. Emerg. Infect. Dis. 24 (6), 1083–1086. https://doi.org/10.3201/ white-footed mouse (Peromyscus leucopus) populations in New York City. Mol. Ecol. 21, eid2406.171271. 1360–1378. https://doi.org/10.1111/j.1365-294X.2012.05476.x. Fickel, J., Hohmann, U., 2006. A methodological approach for non-invasive sampling for pop- Nei, M., Maruyama, T., Chakraborty, R., 1975. The bottleneck effect and genetic variability in ulation size estimates in wild boars (Sus scrofa). Eur. J. Wildl. Res. 52, 28–33. https://doi. populations. Evolution 29 (1), 1–10. https://doi.org/10.2307/2407137. org/10.1007/s10344-005-0003-5. Ouyang, J.Q., Isaksson, C., Schmidt, C., Hutton, P., Bonier, F., Dominoni, D., 2018. A new Fickel, J., Schmidt, A., Putze, M., Spittler, H., Ludwig, A., Streich, W.J., Pitra, C., 2005. Ge- framework for urban ecology: an integration of proximate and ultimate responses to an- netic structure of populations of european Brown hare: implications for management. thropogenic change. Integr. Comp. Biol. 58 (5), 915–928. https://doi.org/10.1093/icb/ J. Wildl. Manag. 69 (2), 760–770. https://doi.org/10.2193/0022-541x(2005)069 icy110. [0760:gsopoe]2.0.co;2. Pal, C., 1998. Plasticity, memory and the adaptive landscape of the genotype. Proc. R. Soc. B Fidino, M., Gallo, T., Lehrer, E.W., Murray, M.H., Kay, C.A.M., Sander, H.A., MacDougall, B., Biol. Sci. 265, 1319–1323. https://doi.org/10.1098/rspb.1998.0436. Salsbury, C.M., Ryan, T.J., Angstmann, J.L., Belaire, J.A., Dugelby, B., Schell, C.J., Parc de Collserola, 2020. Public use, awareness-raising and environmental education. 2016- Stankowich, T., Amaya, M., Drake, D., Hursh, S.H., Ahlers, A.A., Williamson, J., 2020 public usage strategy. Retrieved from https://www.parcnaturalcollserola.cat/en/ Hartley, L.M., Zellmer, A.J., Simon, K., Magle, S.B., 2021. Landscape-scale differences public-use-awareness-raising-and-environmental-education/. 9 J. Hagemann et al. Science of the Total Environment 833 (2022) 155126 Podgórski, T., Baś, G., Jędrzejewska, B., Sönnichsen, L., Śnieżko, S., Jędrzejewski, W., urban wild boar population structure? J. Appl. Ecol. 54 (1), 272–281. https://doi.org/ Okarma, H., 2013. Spatiotemporal behavioral plasticity of wild boar (Sus scrofa) under 10.1111/1365-2664.12756. contrasting conditions of human pressure: primeval forest and metropolitan area. Stillfried, M., Gras, P., Börner, K., Göritz, F., Painer, J., Röllig, K., Wenzler, M., Hofer, H., J. Mammal. 94 (1), 109–119. https://doi.org/10.1644/12-MAMM-A-038.1. Ortmann, S., Kramer-Schadt, S., 2017c. Secrets of success in a landscape of fear: urban Pritchard, J.K., Stephens, M., Donnelly, P., 2000. Inference of population structure using wild boar adjust risk perception and tolerate disturbance. Front. Ecol. Evol. 5, 157. multilocus genotype data. Genetics 155 (2), 945–959. https://doi.org/10.1111/j.1471- https://doi.org/10.3389/fevo.2017.00157. 8286.2007.01758.x. Torres-Blas, I., Mentaberre, G., Castillo-Contreras, R., Fernández-Aguilar, X., Conejero, C., Pulliam, H.R., 1988. Sources, sinks, and population regulation. Am. Nat. 132 (5), 652–661. Valldeperes, M., González-Crespo, C., Colom-Cadena, A., Lavín, S., López-Olvera, J.R., https://doi.org/10.1086/284880. 2020. Assessing methods to live-capture wild boars (Sus scrofa) in urban and peri- R Core Team, 2020. R: a language and environment for statistical computing. R Foundation urban environments. Vet. Rec., e85. https://doi.org/10.1136/vr.105766 E-published for Statistical Computing, Vienna, Austria. https://www.R-project.org/. ahead of print. Reynolds, J., Weir, B.S., Cockerham, C.C., 1983. Estimation of the coancestry coefficient: basis United Nations, 2017. World population prospects: the 2017 revision, key findings and advance for a short-term genetic distance. Genetics 105 (3), 767–779. tables. World Popul. Prospects 1–46. https://doi.org/10.1017/CBO9781107415324.004. Rodewald, A.D., Gehrt, S.D., 2014. Wildlife population dynamics in urban landscapes. In: Van Oosterhout, C., Hutchinson, W.F., Wills, D.P.M., Shipley, P., 2004. MICRO-CHECKER: McCleery, R.A., Moorman, C.E., Peterson, M.N. (Eds.), Urban Wildlife Conservation. The- software for identifying and correcting genotyping errors in microsatellite data. Mol. ory and Practice. Springer, New York, Heidelberg, Dordrecht, London, pp. 117–148 Ecol. Notes 4 (3), 535–538. https://doi.org/10.1111/j.1471-8286.2004.00684.x. https://doi.org/10.1007/978-1-4899-7500-3. Wahlund, S., 1928. Zusammensetzung von populationen und korrelationserscheinungen vom Rohrer, G.A., Alexander, L.J., Keele, J.W., Smith, T.P., Beattie, C.W., 1994. A microsatellite standpunkt der vererbungslehre aus betrachtet. Hereditas 11 (1), 65–106. https://doi. linkage map of the porcine genome. Genetics 136 (1), 231–245. org/10.1111/j.1601-5223.1928.tb02483.x. Šálek, M., Drahníková, L., Tkadlec, E., 2015. Changes in home range sizes and population den- Wandeler, P., Funk, S.M., Largiadèr, C.R., Gloor, S., Breitenmoser, U., 2003. The city-fox phe- sities of carnivore species along the natural to urban habitat gradient. Mammal Rev. 45 nomenon: genetic consequences of a recent colonization of urban habitat. Mol. Ecol. 12 (1), 1–14. https://doi.org/10.1111/mam.12027. (3), 647–656. https://doi.org/10.1046/j.1365-294X.2003.01768.x. Schell, C.J., 2018. Urban evolutionary ecology and the potential benefits of implementing ge- Wang, H., Castillo-Contreras, R., Saguti, F., López-Olvera, J.R., Karlsson, M., Mentaberre, G., nomics. J. Hered. 109 (2), 138–151. https://doi.org/10.1093/jhere d/esy001. Lindh, M., Serra-Cobo, J., Norder, H., 2019. Genetically similar diseases virus strains in- Schell, C.J., Stanton, L.A., Young, J.K., Angeloni, L.M., Lambert, J.E., Breck, S.W., Murray, fect both humans and wild boars in the Barcelona area, Spain, and Sweden. Transbound. M.H., 2020. The evolutionary consequences of human–wildlife conflict in cities. Evol. Emerg. Dis. 66, 978–985. https://doi.org/10.1111/tbed.13115. Appl. 14, 178–197. https://doi.org/10.1111/eva.13131. Weir, B.S., Cockerham, C.C., 1984. Estimating F-statistics for the analysis of population struc- Shochat, E., Warren, P.S., Faeth, S.H., McIntyre, N.E., Hope, D., 2006. From patterns to emerg- ture. Evolution 38 (6), 1358. https://doi.org/10.2307/2408641. ing processes in mechanistic urban ecology. Trends Ecol. Evol. 21 (4), 186–191. https:// West-Eberhard, M.J., 2003. Developmental Plasticity and Evolution. Oxford University Press, doi.org/10.1016/j.tree.2005.11.019. Oxford, UK. Slatkin, M., 1995. A measure of population subdivision based on microsatellite allele frequen- Weyrich, A., Lenz, D., Fickel, J., 2019. Environmental change-dependent inherited epigenetic cies. Genetics 139 (1), 457–462. response. Genes 10 (4). https://doi.org/10.3390/genes10010004. Snow, N.P., Jarzyna, M.A., VerCauteren, K.C., 2017. Interpreting and predicting the spread of Wood, B.C., Pullin, A.S., 2002. Persistence of species in a fragmented urban landscape: the im- invasive wild pigs. J. Appl. Ecol. 54 (6), 2022–2032. https://doi.org/10.1111/1365- portance of dispersal ability and habitat availability for grassland butterflies. Biodivers. 2664.12866. Conserv. 11, 1451–1468. Soulsbury, C.D., White, P.C.L., 2015. Human-wildlife interactions in urban areas: a review of Wright, S., 1950. Genetical structure of populations. Nature 166 (4215), 247–249. https:// conflicts, benefits and opportunities. Wildl. Res. 42 (7), 541–553. https://doi.org/10. doi.org/10.1038/166247a0. 1071/WR14229. Zsolnai, A., Csókás, A., Szabó, L., Patkó, L., Csányi, S., Márton, M., Lakatos, E.A., Anton, I., Statistical Institute of Catalonia, 2019. https://www.idescat.cat/pub/?id=aec&n=250&t= Deutsch, F., Heltai, M., 2022. Genetic adaptation to urban living: molecular DNA analyses 2019&lang=en. of wild boar populations in Budapest and surrounding area. Mamm. Biol. https://doi.org/ Stillfried, M., Gras, P., Busch, M., Borner, K., Kramer-Schadt, S., Ortmann, S., 2017a. Wild in- 10.1007/s42991-021-00212-4. side: urban wild boar select natural, not anthropogenic food resources. PLoS ONE 12 (4), 1–20. https://doi.org/10.1371/journal.pone.0175127. Stillfried, M., Fickel, J., Boerner, K., Wittstatt, U., Heddergott, M., Ortmann, S., Kramer- Schadt, S., Frantz, A.C., 2017b. Do cities represent sources, sinks or isolated islands for 10