Population Genetic Diversity and Differentiation of Mitten Crab, Genus Eriocheir , Based on Microsatellite Markers

: Mitten crab, Eriocheir sensu stricto , is an important indigenous aquatic species and food source in Eastern Asia. Genetic diversity is an important prerequisite for the conservation and utilization of germplasm resources. However, the genetic diversity and differentiation of Eriocheir s. s. remains unclear. This study evaluated population genetic diversity and genetic differentiation of three Eriocheir s. s. populations from Suifenhe (SFH), Liaohe (LH), and Nanliujiang (NLJ), China, based on 19 microsatellite markers. The mean observed number of alleles ( N a ) was 22.84 alleles, and the mean polymorphism information content ( PIC ) was 0.86, which demonstrated high genetic diversity. The allele frequency distribution showed an “L” shape, ranging from 0.01 to 0.74. The genetic diversity parameter values of the LH population were higher than those of the other two populations. All pairwise F ST values showed signiﬁcant differences among the three Eriocheir s. s. populations ( p < 0.01). The value of Nei’s genetic distance ( D S ) varied from 0.31 (between SFH and NLJ populations) to 0.33 (between SFH and LH populations). Overall, The results illustrate that the three wild Eriocheir s. s. populations have high genetic diversity and high genetic differentiation. This study provides the basis for Eriocheir s. s. utilization in China. Author Contributions: Conceptualization, S.W., W.X. and Z.Z.; methodology, S.W.; software, S.W.; validation, W.X. and Z.Z.; formal analysis, L.L.; investigation, R.Z.; resources, K.G.; data curation, S.W. and K.G.; writing—original draft preparation, S.W.; writing—review and editing, S.W. and Z.Z.; visualization, S.W.; supervision, W.X. and Z.Z.; project administration,


Introduction
The mitten crab, Eriocheir sensu stricto, belongs to the Class Crustacea, Order Decapoda, and Family Syringidae, and inhabits the coastal rivers and estuaries from Heilongjiang Province to Guangxi Zhuang Autonomous Region along the coast of China [1,2]. The Chinese mitten crab (Eriocheir sinensis), as one of the most widely developed and utilized mitten crabs, is an economically important species for freshwater aquaculture in China because of its delicate flavor and high nutritional value [3,4]. The aquaculture yield of E. sinensis has increased steadily during the past two decades and stabilized at an annual yield between 750,000 and 800,000 tons in the past four years [5]. In the wild, the major natural habits for E. sinensis are in the Liaohe (LH) and Yangtze (YZ) rivers in China [6], forming the LH and YZ populations of E. sinensis [7]. The LH population of E. sinensis is mainly cultured in paddies, lakes, and a few ponds in northern China. However, YZ populations of E. sinensis are mainly cultured in earth ponds, lakes, and a few paddies in southern China [8]. Of these two populations, the YZ population of E. sinensis is considered more popular for aquaculture practitioners because of larger adult size and good flavor [9]. However, the shortage of wild juveniles from the YZ river basin caused by over-exploitation drove frequent stock transfer among the different geographical populations (mainly transport from the Rivers LH and Oujiang (OJ) to the YZ river basin) in the 1990s [6]. This transport has caused the YZ population of E. sinensis to become a mixed gene pool of LH and OJ E. sinensis. Little is known about transport from the YZ or OJ populations to the LH delta for culture. Meanwhile, combined with the previous studies [10], the LH population of E. sinensis is a relatively pure gene pool of Eriocheir s. s. compared with the YZ population in the Yangtze delta.
Currently, the majority of Eriocheir s. s. derived from the respective river basins have definite species names. The Eriocheir s. s. derived from river basin Tumenjiang (TMJ) in the northeast of China comprises overlapping distributions or a genetically mixed population of E. sinensis and E. japonicus based on morphometrics analysis [11]. Eriocheir s. s. derived from the river basins LH, Yellow (YL), and YZ are treated as E. sinensis based on mitochondrial and microsatellite markers [12,13], because only slight differences in morphological and molecular characteristics were previously observed [14,15]. Eriocheir s. s. derived from the river basins OJ and Minjiang (MJ) are classified into a mixed population of E. sinensis and E. hepuensis based on mitochondrial and microsatellite markers [12,16]. The Eriocheir s. s. derived from the river basin Nanliujiang (NLJ) is classified as E. hepuensis based on morphometrics and microsatellite marker analysis [12,17].
The Eriocheir s. s. derived from river basin SFH is indigenous and distributed in the southeast of Heilongjiang Province, China. According to previous reports [18], the SFH population of Eriocheir s. s. is considered as E. japonicus because Eriocheir s. s. is a migratory aquatic animal, and the megalopa migrate upstream to the river basin SFH from the Sea of Japan. However, based on landmark-based morphometric analysis, the Eriocheir s. s. in the SFH river basin, has closer morphological characteristics that more closely resemble E. sinensis [19]. Currently, there are no studies in the literature that report the genetic diversity and differentiation of Eriocheir s. s. in the SFH river basin from the perspective of genetics. Hence, we used microsatellite loci to explore the genetic diversity and differentiation of Eriocheir s. s. in the SFH river basin.
Previous studies have demonstrated the genetic diversity and differentiation of E. sinensis and E. hepuensis [6,10,12,17,20]. Due to the relatively pure germplasm of Eriocheir s. s. in the LH river basin, we selected the LH population of Eriocheir s. s. as the representative E. sinensis in this study. Likewise, the NLJ population is a representative population of E. hepuensis. This study was conducted using classical population genetics to assess the genetic diversity and differentiation of these three wild Eriocheir s. s. populations (especially the SFH population) and to provide appropriate guidelines for the conservation and utilization of germplasm resources.

DNA Extraction
Total genomic DNA was extracted from leg muscle tissue using the Magnetic Animal Tissue Genomic DNA Kit produced by Tiangen Biotech (Beijing, China) Co., Ltd. (DP341-2). After extraction, the gel was dyed with ethidium bromide replacement, and captured by GeneSnap with the Syngene System (Bio-Rad, Hercules, CA, USA). Subsequently, DNA concentration was measured using a spectrophotometer (Quawell, San Jose, CA, USA). The genomic DNA samples were diluted to 50 ng/µL and stored at −20 • C before use.

DNA Extraction
Total genomic DNA was extracted from leg muscle tissue using the Magnetic Animal Tissue Genomic DNA Kit produced by Tiangen Biotech (Beijing, China) Co., Ltd. (DP341-2). After extraction, the gel was dyed with ethidium bromide replacement, and captured by GeneSnap with the Syngene System (Bio-Rad, Hercules, CA, USA). Subsequently, DNA concentration was measured using a spectrophotometer (Quawell, San Jose, CA, USA). The genomic DNA samples were diluted to 50 ng/μL and stored at −20 °C before use.

Microsatellite Markers and PCR Amplification
A total of 20 loci (Table 1) was screened with information regarding microsatellite primers obtained from the previous literature [21][22][23][24][25]. The primer sequences and allele size ranges are listed in Table 2. All forward primers (5′ end) were fluorescently labeled with either an FAM or HEX tag, and all reverse primers were ordinary primers. These primers were synthesized by Sangon Biotech Co., Ltd. (Shanghai, China).

Microsatellite Markers and PCR Amplification
A total of 20 loci (Table 1) was screened with information regarding microsatellite primers obtained from the previous literature [21][22][23][24][25]. The primer sequences and allele size ranges are listed in Table 2. All forward primers (5 end) were fluorescently labeled with either an FAM or HEX tag, and all reverse primers were ordinary primers. These primers were synthesized by Sangon Biotech Co., Ltd. (Shanghai, China). A total volume of 20 µL was used for each reaction, including 1 µL genomic DNA (50 ng/µL), 0.5 µL forward primer (10 mmol/L), 0.5 µL reverse primer (10 mmol/L), 2 × Es Taq Master Mix (CW Biotech Co., Ltd., Beijing, China), and 8 µL ddH 2 O. PCR products were obtained with a PCR instrument (Bio-Rad, Hercules, CA, USA). The PCR amplification conditions were as follows: firstly, pre-denaturation, 94 • C for 2 min; secondly, a total of 35 cycles including denaturation at 94 • C for 30 s, optimal annealing temperature (listed in Table 1) for 30 s, and 72 • C for 30 s; and finally, an extension step at 72 • C for 10 min. After all the PCR products were detected by agarose gel electrophoresis, the PCR products of each FAM and HEX primer were mixed together, and then sent to Sangon Biotech Co., Ltd. (Shanghai, China) for short tandem repeat (STR) sequencing. Formamide denaturation was used, ROX-500 was used as the internal standard, and an ABI-3730XL automatic DNA sequencer (ABI, Waltham, MA, USA) was used for detection and analysis. The peak map was scanned by Genemapper version 3.5 software (Thermo, MA, USA) and the length of allele fragment represented by the target peak was read. All data were imported into Microsoft Excel 2017 for further analysis [13,26].

Statistical Analysis
According to the size and quantity of each locus gene fragment, the genotype of each sample was determined. The measured genetic diversity, genetic differentiation, and genetic structure parameters were according to the method of Wang et al. [16], which included

Genetic Diversity of Microsatellite Loci
The genetic diversity parameters of 19 microsatellite loci in three Eriocheir s. s. populations are shown in Table 2  The allele frequency distribution is shown in Figure 3. The allele frequency distribution showed an "L" shape, ranging from 0.01 to 0.74. The proportion between 0.0 and 0.1 was the highest, reaching more than 80%, followed by 0.1~0.2, accounting for approximately 10% of the total. The remaining allele frequencies accounted for lower proportions.
The mean PIC values varied from 0.81 ± 0.03 (NLJ) to 0.85 ± 0.03 (LH) (Figure 4c (Figure 4f). The mean Nm values varied from 7.56 ± 1.54 (SFH) to 11.09 ± 1.11 (LH) (Figure 4g). Overall, the parameter values for the LH population were higher than those of the remaining two populations. However, no significant differences were observed for these populations among these three wild populations (p > 0.05). The allele frequency distribution is shown in Figure 3. The allele frequency distribution showed an "L" shape, ranging from 0.01 to 0.74. The proportion between 0.0 and 0.1 was the highest, reaching more than 80%, followed by 0.1~0.2, accounting for approximately 10% of the total. The remaining allele frequencies accounted for lower proportions.  The allele frequency distribution is shown in Figure 3. The allele frequency distribution showed an "L" shape, ranging from 0.01 to 0.74. The proportion between 0.0 and 0.1 was the highest, reaching more than 80%, followed by 0.1~0.2, accounting for approximately 10% of the total. The remaining allele frequencies accounted for lower proportions.

Genetic Diversity of Eriocheir s. s. Populations
The genetic diversity parameters for Eriocheir s. s. populations are shown in Figure 4. The mean N a values varied from 15.95 ± 1.43 (SFH) to 18.53 ± 2.16 (LH) (Figure 4a). Likewise, the mean N e values ranged from 8.43 ± 0.88 (SFH) to 11.79 ± 1.72 (LH) (Figure 4b). The mean PIC values varied from 0.81 ± 0.03 (NLJ) to 0.85 ± 0.03 (LH) (Figure 4c (Figure 4f). The mean N m values varied from 7.56 ± 1.54 (SFH) to 11.09 ± 1.11 (LH) (Figure 4g). Overall, the parameter values for the LH population were higher than those of the remaining two populations. However, no significant differences were observed for these populations among these three wild populations (p > 0.05).

Genetic Differentiation of Eriocheir s. s. Populations
All pairwise FST values between populations, i.e., indices of population differentiation, were significantly different among these three Eriocheir s. s. populations (p < 0.01; Table 3 above the diagonal). The pairwise FST value ranged from 0.0352 (between SFH and LH) to 0.0409 (between SFH and NLJ). A significant pattern of differentiation was observed over all three Eriocheir s. s. populations (FST = 0.03822, p < 0.01; Table 4). Results of AMOVA showed significant differences both among and within populations (p < 0.01; Table 4), and 3.82% of the genetic variation was among populations and 96.18% within populations (Table 4). Unbiased genetic distances (Table 3 below the diagonal) over 19 microsatellite loci were estimated. The DS value varied from 0.3144 (between SFH and NLJ populations) to 0.3310 (between SFH and LH populations). The genetic relationships among Eriocheir s. s. populations were further assessed based on DS values by the reconstruction of phylogenetic trees. The neighbor-joining phylogenetic tree result is shown in Figure 5. SFH and NLJ populations were grouped together, and they then clustered together with the LH population.

Genetic Differentiation of Eriocheir s. s. Populations
All pairwise F ST values between populations, i.e., indices of population differentiation, were significantly different among these three Eriocheir s. s. populations (p < 0.01; Table 3 above the diagonal). The pairwise F ST value ranged from 0.0352 (between SFH and LH) to 0.0409 (between SFH and NLJ). A significant pattern of differentiation was observed over all three Eriocheir s. s. populations (F ST = 0.03822, p < 0.01; Table 4). Results of AMOVA showed significant differences both among and within populations (p < 0.01; Table 4), and 3.82% of the genetic variation was among populations and 96.18% within populations (Table 4). Unbiased genetic distances (Table 3 below the diagonal) over 19 microsatellite loci were estimated. The D S value varied from 0.3144 (between SFH and NLJ populations) to 0.3310 (between SFH and LH populations). The genetic relationships among Eriocheir s. s. populations were further assessed based on D S values by the reconstruction of phylogenetic trees. The neighbor-joining phylogenetic tree result is shown in Figure 5. SFH and NLJ populations were grouped together, and they then clustered together with the LH population.

Genetic Diversity
The higher the genetic diversity, the greater the potential response to natural selection in the population, and the stronger the ability to adapt to the environment, which is conducive to the survival and evolution of the population [27]. In this study, 19 out of 20 highly polymorphic loci were screened to assess the genetic diversity of 180 wild Eriocheir s. s. individuals from three river basins, SFH, LH, and NLJ. The numbers of samples and loci in this study are sufficient to evaluate the genetic diversity parameters, because 30 samples and 10 loci are usually enough for genetic diversity analysis, while larger numbers of samples and loci have a great impact in terms of improving the accuracy of the genetic diversity parameters Na, Ne, and Ho [28]. In this study, the Na value was relatively high at Esin18, Esin42, and Esin67 loci within individuals compared with earlier studies [6,13,16,29]; this value represented the high number of alleles in our sampled individuals. These Na values, either low or high, may be determined by the numbers of samples or electrophoretic typing techniques [16]. All the average values of the Ne and I parameters at 19 loci showed high genetic diversity within individuals, and the results are similar to those of previous studies [13,30,31]. Polymorphism information content (PIC), i.e., a metric of genetic diversity, was not lower than 0.5, which illustrated high genetic diversity per locus. Through comparison with the literature, both wild or cultured Eriocheir s. s. were found to have high genetic diversity, unlike wild Cyprinus carpio [32], Ctenopharyngodon idellus [33], Pelteobagrus fulvidraco [34], Portunus trituberculatus [35], and Litopenaeus vannamei [36]. The allele frequencies of the three wild Eriocheir s. s. populations in this study displayed an "L" shape, indicating that the wild populations did not experience a recent bottleneck effect in their allele frequency distribution [37]. The heterozygosity in a population can reflect the level of genetic variation of that population. The higher the proportion of heterozygotes in the population, the greater the genetic variation [38]. The Ho and He values in this study illustrated high genetic variation, suggesting that the population can easily adapt to the changing environment. If the individuals are in HWE, the Ho and He values are close to each other at the locus. Otherwise, the individuals exhibit an excess or deficiency of heterozygotes [39]. In this study, 13 out of 19 microsatellite loci deviated from HWE, and 11 out of 13 deviated microsatellite loci showed heterozygote deficiency.
This study showed that the genetic diversity parameters of Na, Ne, PIC, I, and Nm values in the LH Eriocheir s. s. population were slightly higher than the other two Eriocheir

Genetic Diversity
The higher the genetic diversity, the greater the potential response to natural selection in the population, and the stronger the ability to adapt to the environment, which is conducive to the survival and evolution of the population [27]. In this study, 19 out of 20 highly polymorphic loci were screened to assess the genetic diversity of 180 wild Eriocheir s. s. individuals from three river basins, SFH, LH, and NLJ. The numbers of samples and loci in this study are sufficient to evaluate the genetic diversity parameters, because 30 samples and 10 loci are usually enough for genetic diversity analysis, while larger numbers of samples and loci have a great impact in terms of improving the accuracy of the genetic diversity parameters N a , N e , and H o [28]. In this study, the N a value was relatively high at Esin18, Esin42, and Esin67 loci within individuals compared with earlier studies [6,13,16,29]; this value represented the high number of alleles in our sampled individuals. These N a values, either low or high, may be determined by the numbers of samples or electrophoretic typing techniques [16]. All the average values of the N e and I parameters at 19 loci showed high genetic diversity within individuals, and the results are similar to those of previous studies [13,30,31]. Polymorphism information content (PIC), i.e., a metric of genetic diversity, was not lower than 0.5, which illustrated high genetic diversity per locus. Through comparison with the literature, both wild or cultured Eriocheir s. s. were found to have high genetic diversity, unlike wild Cyprinus carpio [32], Ctenopharyngodon idellus [33], Pelteobagrus fulvidraco [34], Portunus trituberculatus [35], and Litopenaeus vannamei [36]. The allele frequencies of the three wild Eriocheir s. s. populations in this study displayed an "L" shape, indicating that the wild populations did not experience a recent bottleneck effect in their allele frequency distribution [37]. The heterozygosity in a population can reflect the level of genetic variation of that population. The higher the proportion of heterozygotes in the population, the greater the genetic variation [38]. The H o and H e values in this study illustrated high genetic variation, suggesting that the population can easily adapt to the changing environment. If the individuals are in HWE, the H o and H e values are close to each other at the locus. Otherwise, the individuals exhibit an excess or deficiency of heterozygotes [39]. In this study, 13 out of 19 microsatellite loci deviated from HWE, and 11 out of 13 deviated microsatellite loci showed heterozygote deficiency.
This study showed that the genetic diversity parameters of N a , N e , PIC, I, and N m values in the LH Eriocheir s. s. population were slightly higher than the other two Eriocheir s. s. populations, SFH and NLJ, indicating that the wild LH Eriocheir s. s. population had higher genetic diversity. Two reasons may explain this result. First, the LH Eriocheir s. s. population is an important species cultured in paddies and lakes in northeastern China, with a large population size, accompanied by proliferation and release. On the other hand, wild germplasm of the SFH and NLJ rivers has not been developed and utilized, and there are few river branches in its basin, which inhibits inter-population migration. Second, SFH and NLJ Eriocheir s. s. may not be the same species as LH E. sinensis. The microsatellite markers were specific primers for E. sinensis, but not for other Eriocheir s. s.; therefore, the microsatellite markers may not be absolutely precise for SFH and NLJ Eriocheir s. s. population analysis [16]. This may explain why genetic diversity parameters of the SFH and NLJ Eriocheir s. s. populations are lower than those of the LH Eriocheir s. s. population. To analyze the population genetic diversity of the SFH and NLJ Eriocheir s. s. populations, it is urgent to conduct screenings using genetic markers appropriate for evolutionary studies of this lineage.

Genetic Differentiation
Gene flow (N m ) and the genetic differentiation coefficient (F ST ) are two important parameters used to analyze the degree of genetic differentiation [32]. This study demonstrated a relatively high N m value (7.31) and low F ST value (0.0330) among populations at all loci, which implied low genetic differentiation. Pairwise F ST values indicated genetic differentiation among the SFH, LH, and NLJ Eriocheir s. s. populations. Previous studies also demonstrated genetic differentiation, although the pairwise F ST value between populations LH and NLJ was 0.0324 [6], which is very close to the pairwise F ST value (0.0385) in this study. Likewise, the pairwise F ST values among SFH, LH, and NLJ in this study were higher than those in earlier studies [13,16], e.g., 0.0050 (between LH and YZ) and 0.0105 (between YZ and MJ). Although the pairwise F ST values (0.0352, 0.0385, and 0.0409) in this study were lower than 0.05, the values still suggest considerable genetic differentiation for Eriocheir s. s. Pairwise estimations of Nei's D S exhibited a similar result. The D S values (0.3144, 0.3292, and 0.3310) were higher than in previous studies [13,16,23,29,40], which showed a relatively great genetic distance among SFH, LH, and NLJ Eriocheir s. s. populations. Based on Nei's D S , the neighbor-joining phylogenetic tree was constructed among these three wild Eriocheir s. s. populations. This study illustrated that populations SFH and NLJ were closer to each other than to the LH population, which was consistent with results from a previous study [41] but contradicted those from Jiang et al.'s study [19]. In Jiang et al.'s study, the SFH population was clustered with the YZ population (E. sinensis) rather than the NLJ population. However, the NLJ population clustered with E. japonicus was sampled in Hyogo, Japan. Two reasons may explain the difference. First, microsatellite analysis is more accurate from the perspective of genetics, while a morphological method was used in Jiang et al.'s work. Second, E. japonicus was sampled in Hyogo, Japan (35.62 • N, 134.48 • E), which is far from the SFH population. Therefore, the authors maintained that E. japonicus should have many different geographical populations, just like E. sinensis, including the LH, YL, and YZ populations. The Hyogo population of E. japonicus is located in the south of Japan's islands.
Results of AMOVA analysis revealed that total variation among populations was 3.82%, whereas 96.18% of variation was observed within populations. Low variation among populations may be caused by the Eriocheir s. s. lifecycle. Eriocheir s. s. has a catadromous lifecycle and travels through many different habitats during migration. Therefore, it was interpreted that Eriocheir s. s. probably loses local environmental adaptability [16].

Conclusions
Three wild Eriocheir s. s. populations (SFH, LH, and NLJ) had high genetic diversity and significant genetic differentiation. The study will provide important support for the conversation and management of Eriocheir s. s. populations.

Institutional Review Board Statement:
The studies on Eriocheir sensu stricto were reviewed and approved by the Committee for the Welfare and Ethics of Laboratory Animals of Heilongjiang River Fisheries Research Institute (Approval numbers: 20210901-001 (approved on 1 September 2021)).

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.