Genetic differentiation of pink salmon Oncorhynchus gorbuscha Walbaum in the Asian part of the range

Genetic variation at 19 enzyme (including 11 polymorphic) and 10 microsatellite loci was examined in the population samples of odd-and even-broodline pink salmon from the southern part of Sakhalin Island, Southern Kuril Islands, and the northern coast of the Sea of Okhotsk. The estimates of relative interpopulation component of genetic variation for the allozyme loci, per broodline, were on average 0.43% (G (ST)), while over the microsatellite loci it was 0.26% (the theta(ST) coefficient, F-statistics based on the allele frequency variance), and 0.90% (the rho(ST) coefficient, R-statistics based on the allele size variance). The values of interlinear component constituted 2.34, 0.31, and 1.05% of the total variation, respectively. Using the allozyme loci, statistically significant intralinear heterogeneity was demonstrated among the regions, as well as among the populations of southern Sakhalin. Multidimensional scaling based on the allozyme data demonstrated regional clustering of the sample groups, representing certain populations during the spawning run or in different years. Most of the microsatellite loci examined were found to be highly polymorphic (mean heterozygosity > 0.880). The estimates of interlinear, interregional, and interpopulation variation over these loci in terms of theta(ST) values were substantially lower than in terms of rho(ST) values. Regional genetic differentiation, mostly expressed at the allozyme loci between the populations from the northern Sea of Okhotsk and the Sakhalin and Kuril group of populations, was less expressed at the microsatellite loci. The differentiation between these regions observed can be considered as the evidence in favor of a large-scale isolation by distance characterizing Asian pink salmon. It is suggested that in pink salmon, low genetic differentiation at neutral microsatellite loci can be explained by extremely high heterozygosity of the loci themselves, as well as by the migration gene exchange among the populations (the estimate of the gene migration coefficient inferred from the "private" allele data constituted 2.6 to 3.4%), specifically, by the ancient migration exchange, which occurred during postglacial colonization of the range.