Genome-wide SNP assessment of contemporary European red deer genetic structure highlights the distinction between peripheral populations and the main admixture zones in Europe

Genome-wide technologies open up new possibilities to clarify questions on genetic structure and phylogeographic history of taxa previously studied with microsatellite loci and mitochondrial sequences. Here, we used 736 individual red deer (Cervus elaphus) samples genotyped at 35,701 single nucleotide polymorphism loci (SNPs) to assess the population structure of the species throughout Europe. The results identified 28 populations, with higher degrees of genetic distinction in peripheral compared to mainland populations. Iberian red deer show high genetic differentiation, with lineages in Western and Central Iberia maintaining their distinctiveness, which supports separate refugial ranges within Iberia along with little recent connection between Iberian and the remaining Western European populations. The Norwegian population exhibited the lowest variability and the largest allele frequency differences from mainland European populations, compatible with a history of bottlenecks and drift during postglacial colonization from southern refugia. Scottish populations showed a high genetic distance from the mainland but high levels of diversity. Hybrid zones were found between Eastern and Western European lineages in Central Europe as well as in the Pyrenees, where red deer from France are in close contact with Iberian red deer. Anthropogenic restocking has promoted the Pyrenean contact zone, admixture events in populations on the Isle of Rum and the Netherlands, and at least partly the admixture of the two main lineages in central-eastern Europe. Our analysis enabled detailed resolution of the population structure of a large mammal widely distributed throughout Europe and contributes to resolving the evolutionary history, which can also inform conservation and management policies.

Methods
We collected tissue samples (e.g., fragments of ears, internal organs, or muscles) from culled individuals, immediately kept at 4°C and then stored at -20°C until use. Genomic DNA was isolated from pieces of tissue using the commercial DNA purification kits DNeasy Blood and Tissue Kit (Qiagen, Germantown, MD, USA) according to the manufacturer’s protocol. Due to the number of analyses that had to be carried out on each sample and the minimum requirements needed, it was necessary to obtain high-quality DNA (in terms of high molecular weight, purity, and yield) and carry out quantity and quality controls. Extracted DNA was visualized for integrity and quantified in a 0.8% agarose gel: a 3 ml aliquot of each sample was loaded with 2 ml of Loading Dye. The integrity of DNA samples was verified by comparing them against a DNA Ladder Sibenzyme molecular weight marker (1 Kb). After that, two μl of genomic DNA were measured by NanoDropTM One (Thermo Fisher Scientific Inc, Waltham, Massachusetts, USA). If the purified DNA did not have the required degree of purity (Absorbance ratios of A260/A280 and A260/230 of 1.8 to 2), a second extraction was carried out. The individuals were genotyped using the cervine 50K Illumina Infinium iSelect HD Custom BeadChip that includes 50,841 SNP markers.