Abstract: The purpose of this experiment was to reveal the variety source and population genetic structure of horse population in Kangxi grassland. The blood samples of 60 local horses with good health, moderate weight and indistinguishable male and female between the ages of 3 and 6 were collected. Four foreign horse breeds such as thoroughbred horses and one local horse breed in China were used as controls. The alleles of 12 microsatellite loci on chromosomes were detected by designing primers and extracting genomic DNA combined with PCR-STR. The genetic diversity parameters, population differentiation, bottleneck effect and population structure were analyzed. The experiment was divided into 6 groups according to the number of populations, with one repetition in each group. Each repetition was based on the sample size of this group. After analyzing the genetic parameters of 6 horse populations, the results showed that 221 alleles were detected, and the average Shannon index was 1.691; The number of effective alleles (Ne) ranged from 3.649 to 5.397; The expected heterozygosity (He) was 0.681-0.798; The observed heterozygosity (Ho) ranged from 0.632 to 0.780; The average polymorphism information content (PIC) ranged from 0.643 to 0.772. These results showed that the 6 populations had high genetic diversity. Microsatellite DNA locus linkage disequilibrium and Hardy-Weinberg equilibrium analysis showed that the P value of most loci was greater than the correction value (P>0.006 5), indicating that many loci were in Hardy-Weinberg and linkage equilibrium. The population differentiation coefficient (Fst) among populations showed that the degree of differentiation among populations was low (Fst<0.15). Based on the Mantel correlation test between Nei's genetic distance and geographical distance, there was a positive correlation between them (R=0.502), but it did not reach a significant level (P=0.310). The bottleneck effect analysis of horse populations showed that the horse populations in this area were affected by the bottleneck effect, and the P value was very significant (P_(IAM)<0.01), indicating a large-scale reduction in the number of the horse population in history. Through FCA factor analysis, UPGMA tree based on Nei's standard genetic distance and Structure Bayesian cluster analysis, it is found that the horse population in this area was clustered with thoroughbred horses and warm blood horses, indicating that the population was most likely to be mainly from warm blood horses and thoroughbred horses. These results reveal that horses in Kangxi grassland have high genetic diversity, and the lineage of the local horse population is similar to that of thoroughbred horses and warm blood horses, which lays a solid foundation for the evaluation, development and utilization of local horse genetic resources.

Key words: Kangxi grassland horse population, microsatellite DNA marker, population genetic structure, genetic diversity, variety source