The pike-perch (Sander lucioperca) is a native fish species in European and Asian river basins. Owing to its fast growth rate, strong adaptability, and high nutritional value, it has become a promising candidate species for aquaculture. In China, artificial reproduction of pike-perch has been realized, and most of its parents are caught from wild populations. However, the genetic background and diversity information of wild populations are still limited, which is not conducive to the establishment of genetic breeding management strategies. In order to investigate the genetic structure of S. lucioperca wild populations, we analyzed the genetic differences of six populations in China and two populations in Central Asia using partial mitochondrial COⅠ gene sequences, and compared them with the haplotype sequences of European populations. As a result, 5 variable sites were detected in the 640 bp COⅠ gene sequence, and 7 haplotypes were defined, among which Hap1 was the shared haplotype between eight populations, and was identical to HapA of European groups. The proportion of Hap1 in Chinese populations (93.36%) was higher than that in Central Asian populations (72.58%) and European population (55.56%). Hap2 and Hap3 were specific haplotypes of Chinese populations, while Hap4, Hap5, Hap6, and Hap7 were specific haplotypes of Central Asian populations. The cluster and network analysis of haplotype sequences showed that Hap1/A was the ancestral haplotype in Sander lucioperca populations. The specific haplotype of Chinese and Central Asian populations belonged to Hap1/A subtype with only 1-2 loci variation compared with the ancestral haplotype, which was quite different from the specific haplotype of European populations. The number of haplotypes per population ranged from one to four, with the largest number of haplotypes in the Zaysan Lake (ZS) population, while there was only one haplotype (Hap1) in Tengger Lake (NX), Xingkai Lake (XK) and Yalu River (YJ) populations in China. The haplotype diversity (H_d) and nucleotide diversity (π) of TS population were the highest (H_d=0.514±0.069; P_i=0.000 80±0.000 11), followed by ZS population, while the diversity parameters of six populations in China were lower. The AMOVA analysis showed that the genetic variation among S. lucioperca populations accounted for 20.74%, and the degree of genetic differentiation (F_st) among populations was relatively high (0.15≤F_st=0.207 36<0.25). The genetic differentiation of the TS and the ZS, the TS and the six Chinese populations all were great (F_st>0.25). Among Chinese populations, Heihe (HH) population had great genetic differentiation against other populations, while there was no genetic differentiation between the other five populations. The phylogenetic tree based on genetic distance showed that six populations from China and ZS population from Kazakhstan were clustered into one branch, while TS population from Uzbekistan was a separate branch. The results of this study provide a reference for the breeding and releasing management of S. lucioperca populations.