Study of genetic diversity and population structure of five Russian cattle breeds using whole-genome SNP analysis

Abstract

With the publication of the complete sequence of a cattle genome, it became possible to trace the history of breed origins and to evaluate genetic relationships between modern breeds, based on the results of genome-wide SNP screening. Whilst numerous studies have been undertaken to characterize the commercial breeds and some local cattle breeds of Europe, North America, Asia and Africa at whole-genome level, little is known about genetic differences, relationships and population genetic structure of the Russian native cattle breeds. The aim of our work was to study the genetic diversity and population structure of five locally-developed Russian cattle breeds, based on genome-wide single nucleotide polymorphisms (SNPs) generated using Illumina Bovine SNP50 BeadChips (Illumina, San Diego, CA, USA). In total, 116 samples (sperm or tissue) collected from five breeds were analyzed, including Bestuzhev (BEST, n = 27), Kholmogor (KHLM, n = 25), Kostromsky (KSTR, n = 20), Red Gorbatov (RGBT, n = 23) and Yaroslavl breeds (YRSL, n = 21). Samples of Holstein cattle (HLST, n = 29) were used for comparison. Quality filtering of genetic markers was performed in PLINK v 1.07. Data processing was performed using software PLINK 1.07, HP-Rare 1.1, STRUCTURE, ver. 2.3.4, Phylip, ver. 3.695, FigTree 1.4.2, Arlequin suite, ver. 3.5.2.2 and R pocket. The final set of markers passed through the quality control and selected for further analysis included 35874 SNPs. Average heterozygosity within breeds ranged from 0.378 in BEST to 0.390 in KHLM and was higher comparing to HLST (0.377). Allelic richness was ranging from 1.914±0.001 in KSTR to 1.955±0.001 in BEST. A slight heterozygote excess was detected in all breeds studied (FIs from -0.015 in BEST to -0.054 in KHLM). The multidimensional scaling (MDS) showed the presence of non-overlapping breed specific clusters, whereas the first principal component (PC1) accounted for 5.46 % and the second principal component (PC2) was responsible for 5.05 % of the genotypic variance. Phylogenetic analysis based on parsimony method grouped individuals into six clusters according to their breeds. The STRUCTURE analyses supported the assumption that the ancestry of the locally developed Russian cattle breeds is distinct from Holsteins and Holstein-related breeds. The highest AK showing the assumed number of populations was observed for k = 6. At k = 6, the genetic structure is in agreement with breed origin of individuals: Q1/6 = 0.855±0.018 for BEST, Q2/6 = 0.818±0.029 for KHLM, Q3/6 = 0.923±0.015 for KSTR, Q4/6 = 0.816±0.027 for RGBT, Q5/6 = 0.873±0.031 for YRSL and Q6/6 = 0.935±0.014 for HLST. Analysis of molecular variance (AMOVA) showed highly significant results for genetic differentiation (p < 0.001) in studied breeds. AMOVA revealed that most of the genetic variation in cattle breeds was found within populations (91.2 %), and less among populations (8.8 %). The emerging structure of the phylogenetic tree constructed on the Nei genetic distances, is in full concordance with the historical origin of breeds and confirms the MDS and STRUCTURE results. Thus, using the method of genome-wide SNP studies we were able for the first time to study the population structure and genealogical relationships among the five Russian cattle breeds. The received information is the first step towards the evaluation of the value of these breeds regarding their conservation and usage in the agricultural production of the future.