Kusaka et al.: Genetic population structure of Dipturus chinensis around Japan 
101 
Ex Taq DNA polymerase. The reaction 
conditions were as follows: an initial 
denaturation at 94°C for 3 min; 30 cycles 
of denaturation at 94°C for 30 s, anneal- 
ing at 53°C (LERI26, LERI33, LERIS50, 
and LERI63) or 57°C (LERI21, LERI34, 
and LERI44) for 30 s, and extension at 
72°C for 30 s; and a final extension step 
of 72°C for 10 min. Forward primers 
were labelled with the fluorescent dyes 
FAM, VIC, NED, or PET (Thermo Fisher 
Scientific Inc.). Fragments were sized 
on an automated DNA sequencer, the 
ABI Prism 3130 XL Genetic Analyzer, 
by using dye-labeled Applied Biosys- 
tems GeneScan 600 LIZ Size Standard 
(Thermo Fisher Scientific Inc.), and 
scored by using the Applied Biosystems 
software Peak Scanner, vers. 1.0 (Thermo 
Fisher Scientific Inc.). 
Micro-Checker (vers. 2.2.3; Van Oos- 
terhout et al., 2004) was used for iden- 
tifying possible genotyping errors (i.e., 
stuttering, large allele dropout, and null 
alleles) within the SSR data set by using 
1000 randomizations. Linkage disequi- 
librium for all loci was calculated by using GENEPOP, 
vers. 4.2 (Raymond and Rousset, 1995; Rousset, 2008). 
The observed (Hy) and expected (H;) heterozygosities 
were calculated by using Arlequin. Deviations of geno- 
typic distributions from the Hardy—Weinberg equilib- 
rium (exact test) (Guo and Thompson, 1992) within each 
sampling location, by locus, were tested by using Arle- 
quin along with a sequential Bonferroni correction (Rice, 
1989), to adjust the significance level. Allelic diversity 
assessed as allelic richness (Ag) was calculated by using 
FSTAT, vers. 2.9.3.2 (Goudet, 2002). The levels of genetic 
diversity within and among the geographic populations 
were tested by hierarchical AMOVA, as implemented in 
Arlequin. In these microsatellite analyses, we assumed 
different groupings of locations to find the optimal 
grouping of locations with the highest value of molecular 
genetic diversity among geographic groups (Ro), a value 
analogous to ®op in the mtDNA analyses. The pairwise 
values of molecular genetic diversity among populations 
(Rgp), values analogous to ®gp in the mtDNA analyses, 
were calculated, and their significance was tested with 
10,000 permutations and adjusted with a sequential 
Bonferroni correction (Rice, 1989). 
Locality 
Clade A 
Taiwan 
Clade B1 
Clade B2 
Results 
Mitochondrial DNA analyses 
A 931-base-pair fragment of the mt cyt b gene was obtained 
from 214 polkadot skate from 9 locations (Table 1, Fig. 1). 
There were 53 polymorphic sites, and transitions and trans- 
versions were observed at 49 and 4 of these sites, respectively. 
Danjo Islands 
Goto Islands 
Kyoto Prefecture 
Niigata Prefecture 
Koshiki-jima Islands — 
Kochi Prefecture 
Aomori Prefecture 
Table 2 
Number of samples (7), number of haplotypes (H), haplotype diversity (h), 
and nucleotide diversity (m) estimated from mitochondrial cytochrome 6 gene 
(931 base pairs) for each population of the polkadot skate (Dipturus chinensis) 
sampled between 2010 and 2017 at 7 locations around Japan. Information is 
also provided for populations of Taiwan and the Korean Peninsula for which 
sequences were obtained through the International Nucleotide Sequence 
Database Collaboration. The mean values of h and n are given with standard 
deviations (SDs). 
H h (SD) mt (SD) 
0.884 (0.028) 
0.817 (0.073) 
0.617 (0.072) 
0.521 (0.067) 
0.0031 (0.002) 
0.0023 (0.001) 
0.0023 (0.001) 
0.0019 (0.001) 
Korean Peninsula - 
0.245 (0.113) 0.0005 (0.000) 
0.726 (0.069) 0.0014 (0.001) 
Genetic variation within populations Within 6 of the 
9 geographic populations, excluding those at the Koshiki- 
jima Islands, Taiwan, and the Korean Peninsula (each 
with a sample size of only 1 individual), the number of 
haplotypes ranged from 3 to 13 haplotypes (Table 2). The 
number of haplotypes within each geographic population 
in the Sea of Japan (6 haplotypes in the populations in 
both Kyoto Prefecture and Niigata Prefecture) was lower 
than the number in the East China Sea (13 and 8 haplo- 
types in the populations at the Danjo Islands and Goto 
Islands, respectively). Additionally, the numbers in the 
Pacific Ocean (3 and 7 haplotypes in the populations in 
Kochi and Aomori Prefectures, respectively) were lower 
than those in the East China Sea. The h values were high 
in all the geographic populations (0.521—0.884), except 
for in the population in Kochi Prefecture (0.245). The 
m values were low in all geographic populations (0.0005— 
0.0031), with the values in the Pacific Ocean (0.0005— 
0.0014) lower than those in the East China Sea and the 
Sea of Japan (0.0019—0.0031). The h and zx values in the 
Hast China Sea and the Sea of Japan gradually decrease 
among populations from the Danjo Islands northward to 
Niigata Prefecture. 
Phylogenetic relationships among haplotypes A total of 
34 haplotypes were defined. Among them, 8 haplotypes 
were shared between populations in at least 2 differ- 
ent locations, accounting for 65.0% of all observations; 
12 haplotypes were found in more than 1 specimen from 
1 location; and the remaining 14 haplotypes were single- 
tons (Figs. 2 and 3). 
The creation of a maximum likelihood tree of haplo- 
types (Fig. 2) revealed 2 major lineages, clades A and 
