Genetic structure of the small yellow croaker (Larimichthys polyactis) across the Yellow Sea and the East China Sea by microsatellite DNA variation: implications for the division of management units

Introduction

Marine fish have complex genetic structures because of their special geographical environment (Hanne et al., 2010). Relatively independent dynamic trends are usually detected in different geographic populations, yet there is also large-scale migration among them (Gyllensten, 1985). Therefore, genetic variation between geographical populations cannot be ignored in the assessment and management of marine fish resources. The genetic theory of adaptive evolution predicts that species rarely exist as a single breeding population due to genetic differentiation among populations (Husband & Barrett, 1995). Population genetic differentiation within species can generate population structure which plays an important role in evolution (Wu, Qin & Gu, 1992). Compared with terrestrial biota, higher genetic connectivity is usually generated in marine fishes as high migration rate and lack of physical barriers, resulting in reduced genetic differentiation (Hewitt, 2000). In order to protect the genetic resources of species more reasonably, devising management strategies based on the genetic structure of the species is important. Populations with specific genetic structure are usually managed as separate conservation units (Avise, 1992; Spielman et al., 2004; Zhang et al., 2020a). Thus, the study of genetic diversity and genetic structure is a prerequisite for making practical conservation policy. Direct observation studies of the movements in marine organism to assess population structure in the marine environment are impractical (Lowe & Allendorf, 2010), and precise characterization of fish stock structure using simpler and more effective analytical approaches has long been an aim for sustainable fisheries management.

Larimichthys polyactis is a famous seafood product in China, and its population structure is the cause of major concern (Zhang, Xue & Wang, 2015). L. polyactis from offshore areas of China used to be divided into three populations according to the migratory route, spawning ground and morphological differences (Zhang & Liu, 1959; Ikeda, 1964; Hu, 1998). This division was supported by the RAPD and AFLP markers (Meng et al., 2003; Han, Lin & Shui, 2009). There are also two (Xu & Chen, 2009) or four management populations (Lin, 1987) were reported based on different methods, such as migratory route, morphological differences or fishery resources surveys (Table S1). However, mitochondrial DNA and microsatellite markers detected no significant differentiation among L. polyactis populations and restriction-site associated DNA sequencing verified these results (Xiao et al., 2009; Kim et al., 2012; Li et al., 2013; Zhang, Xue & Wang, 2015; Zhang et al., 2020b) (Table S1). The trend of younger age and miniaturization of L. polyactis is on the rise (Xu & Chen, 2009; Lin, Liu & Jiang, 2011; Zhang et al., 2020a). Therefore, more studies are needed to detect the population genetic structure of this species, thus developing more reasonable resource conservation policies.

The Yellow Sea and the East China Sea are the main habitats of L. polyactis where intricate hydrology and unique tectonic features have been detected in previous studies (Lin et al., 2009; Ni et al., 2014). Both freshwater outflow and ocean currents play important role in the phylogeographical patterns of L. polyactis. For example, the Yangtze River pours 900 billion m³ freshwater into the East China Sea, which can be a barrier to block the gene flow of some marine organisms (Su & Yuan, 2005; Ni et al., 2014). Besides, the seawater with low salinity from the Yellow Sea was carries to the East China Sea by the Subei Coastal Current in summer (Su & Yuan, 2005). Meanwhile, the China Coastal Currents could carry the warm water from the South China Sea into the East China Sea (Su & Yuan, 2005), which may lead to complex habitat environment for L. polyactis. The generation time of L. polyactis is about 2 years based on previous studies (Lin & Cheng, 2004). However, the most recent study to focus on the genetic structure of L. polyactis used samples collected in 2014 (Zhang, Xue & Wang, 2015; Zhang et al., 2020b). Therefore, it is necessary to collect up to date samples to accurately study the current genetic structure of L. polyactis under complex geographic environments.

The complex geographic environments and the biological characteristics of migration and larval dispersal may contribute to the unique population structure of L. polyactis in offshore China (Lin & Cheng, 2004; Lin, Liu & Jiang, 2011). Moreover, the genetic structure of L. polyactis is dynamic due to the resource change caused by environment and fishing pressure. Therefore, the current genetic structure of L. polyactis is the basis for the formulation and implementation of management policy. In this study, we analyzed the variation in genetic diversity and genetic structure of L. polyactis based on 12 polymorphic microsatellite markers developed by Liu, Gao & Liu (2014). Detection of genetic diversity and current genetic structure of L. polyactis can help in reflecting the evolutionary potential and providing the foundation for the division of management units. Our results can be used as basis for developing more reasonable management strategies.

Material and Methods

Primers selection and genotyping

A total of 12 microsatellite loci for L. polyactis developed by Liu, Gao & Liu (2014) were selected to study the population genetics of L. polyactis (Table S2). Forward primers were 5′ -labelled with a fluorescent dye (HEX, FAM or TAMRA). The PCR was performed in an A300 Fast Thermal Cycler (LongGene Scientific Instruments, Co. Ltd., Hangzhou, China). The reaction system and amplification conditions were carried out according to the methods of Song et al. (2010). The annealing temperature (Ta) of each locus is showed in Table S2. The experiment was performed away from light in order to avoid fluorescence quenching. Amplicons with different fluorescent labels or different sizes were pooled and analyzed on an ABI3730 DNA sequencer (Tsingke Biotech Co., Ltd., Qingdao, China) for the genotyping of microsatellites DNA. Fragment sizes were determined with the ROX-500 standard using GeneMapper.

Table 1:

Sampling information and genetic diversity parameters of L. polyactis localities.

ID	Locality	Coordinates	Sampling date	Sampling size	HO	HE	PIC	AR	uHE	FIS
YT	Yantai	37°84′N, 121°22′E	2019.12	24	0.976	0.920	0.889	11.950	0.917	0.076
RS	Rushan	36°82′N, 121°99′E	2019.04	24	0.977	0.916	0.889	12.423	0.918	0.086
QD	Qingdao	35°99′N, 120°42′E	2019.08	24	0.971	0.918	0.894	12.917	0.919	0.084
LYG	Lianyungang	34°52′N, 120°16′E	2019.08	24	0.981	0.919	0.891	11.882	0.925	0.072
YC	Yancheng	33°78′N, 120°86′E	2020.10	24	0.978	0.919	0.894	13.325	0.915	0.079
ZS	Zhoushan	29°78′N, 122°36′E	2019.08	24	0.980	0.924	0.903	13.083	0.910	0.049
WZ	Wenzhou	27°84′N, 120°98′E	2020.11	24	0.970	0.923	0.897	14.000	0.915	0.033

DOI: 10.7717/peerj.13789/table-1

Notes:

H_O

observed heterozygosity

H_E

expected heterozygosity

PIC

polymorphic information content

A_R

allelic richness

uH_E

unbiased expected heterozygosity

F_IS

inbreeding coefficient

Result

Genetic structure and differentiation

The genetic structure of L. polyactis localities was estimated based on pairwise F-statistics (F_ST). The F_ST among localities ranged from −0.0073 (LYG vs ZS) to 0.0153 (QD vs WZ) (Table 2). Besides, the range of R_ST was −0.0083 (RS vs WZ) to 0.0145 (QD vs ZS) (Table S5). Significant genetic differentiation was detected only between Zhoushan and a few other localities. No clear phylogeographic signal was identified due to similar value of F_ST and R_ST, revealing low level of differentiation among L. polyactis localities. This result also showed that migration and genetic drift were more important relative to mutation at this sampling scale. The Mantel test indicated no significant relationship between pairwise estimates of F_ST/(1- F_ST) and geographic distance (r = 0.5663; p = 0.142). The result of 3D-FCA showed that the contribution rates of three principal components were 46.37%, 21.79% and 17.21%, respectively, and no significant population structure existed throughout the examined range of L. polyactis (Fig. S1). This result was also supported by a DAPC analysis (Fig. 2).

Table 2:

Pairwise F_ST (below diagonal) and pairwise geographic distances (km, above diagonal) among L. polyactis localities.

Population	YT	RS	QD	LYG	YC	ZS	WZ
YT		86	230	426	696	1159	1358
RS	0.0080		181	419	654	1137	1331
QD	0.0077	0.0012		277	285	768	961
LYG	−0.0057	0.0038	0.0049		285	750	934
YC	0.0067	−0.0076	0.0032	−0.0097		572	762
ZS	0.0002*	0.0081	0.0142	−0.0073*	0.0115*		419
WZ	0.0014	−0.0070	0.0153	−0.0071	0.0053	0.0052

DOI: 10.7717/peerj.13789/table-2

Notes:

*Significant p < 0.005 after Bonferroni correction for multiple comparisons.

The cryptic population structure was detected using the software of STRUCTURE. The number of K value indicated that the model where K = 3 was the most appropriate (Fig. 3). Seven localities did not show significant clustering trends using K = 3 (Fig. 4), which supported the result of F_ST and the discriminant analysis of principal components.

Discussion

Genetic diversity is an important component of biodiversity whose variation can generally reflect adaptive evolutionary potential (Marty, Dieckmann & Ernande, 2015) and disease resistance (Zhu et al., 2013). Polymorphic information content is an important parameter to assess the discriminatory power of molecular markers in genetic population studies (Serrote et al., 2020). In this study, high PIC values were detected, showing the usefulness of the genetic markers used in the study. The genetic diversity, measured by observed heterozygosity and expected heterozygosity, was considered high. No recently genetic bottleneck was detected, which also indicated that there was high genetic diversity in L. polyactis. This was consistent with the results of previous studies based on mitochondrial DNA (Xiao et al., 2009; Kim et al., 2012). The life history characters of marine organisms, such as breeding systems and larval dispersal duration, are closely related to genetic diversity (Li et al., 2013). Higher genetic diversity means that species may have higher evolutionary potential and stronger ability to adapt to environmental changes (Giovannoni et al., 1990; Hu et al., 2021). Both Larimichthys crocea and L. polyactis are important seafood products in China. The resources of L. crocea have declined rapidly because of overfishing since the 1980s, while the number of L. polyactis at market has remained at a high level in recent years (Lin et al., 2009; Liu et al., 2016). In the 1950s, L. polyactis was in the fishery boom period, whose average annual production could reach 120,0000 tons. However, in the 1960s and 1970s, it has gradually decreased due to overfishing. With the continuous increase of fishing pressure, the resource of much marine fish has declined severely in the 1980s. Some protection policies have been developed to protect the fish stock since the 1990s (Lin, Cheng & Jiang, 2008), which playing a key role in the recovery of L. polyactis resources. It is reported that the capture production of L. polyactis was 53,000 tons in the 1990s, while this data was 300,000 in the 2000s (Li et al., 2013). These huge fluctuations can be attributed to the strong adaptability of this species and the implementation of the summer closed fishing policy (Cheng et al., 2004; Chen et al., 2020).

As a sensitive molecular marker, microsatellite marker may detect potential genetic differentiation among localities that cannot be found based on tradition molecular markers, such as AFLP, RAPD and mitochondrial DNA (Simbine, Viana & Hilsdorf, 2014). It has become one of the most commonly used molecular markers in population genetics due to its characteristics of codominant inheritance and a high variability (Gupta, Varshney & Sharma, 1999; Zane, Bargelloni & Patarnello, 2002; Parida et al., 2009; Askari, 2013; Song et al., 2017). For instance, in contrast to works that did not find significant genetic differentiation based on other markers, Song (2020) detected genetic differentiation in Acanthogobius ommaturus between Zhoushan and other localities using 14 microsatellite loci. Beacham et al. (2010) also detected significant genetic differentiation in Oncorhynchus keta from Pacific Coast Honshu based on 14 microsatellite loci, which was not found by mitochondrial DNA markers and allozyme loci. In this study, no significant genetic structure consistent with the distribution pattern was detected. Only some small genetic differentiation was found between ZS and a few other localities based on the conventional population statistic F_ST. Most researchers believed that L. polyactis from the southern Yellow Sea and the East China Sea should be lumped together into one single group (Lin et al., 2009). The genetic differentiation between ZS and other localities in this study may be attributed to the complex geographical environment of the Zhoushan Islands. In the 1960s, 32° N was regarded as the geographic boundary separating the South Yellow Sea group and the East China Sea group of L. polyactis based on fishery-dependent studies. However, recent studies have suggested that this geographical distribution boundary might move southwards to Zhoushan Island (31° N) (Lin et al., 2009). Lin et al. (2009) also found that there was a big spawning ground of L. polyactis in the coastal waters of Zhoushan, which could also be the reason for the genetic differentiation. The special geographical environment of the coastal waters of Zhoushan has resulted in the complex population structure of many species (Song, 2020). Also, the freshwater of Yangtze River has a wide influence on the East China Sea, including the Yangtze River Estuary fishing ground, Zhoushan fishing ground, etc (Ni et al., 2014; Zhang & Hu, 2005). Previous studies have demonstrated that the fish population near Zhoushan Island often showed different genetic characters, such as Oplegnathus fasciatus (Xiao, Ma & Dai, 2016), Sebastiscus marmoratus (Xu, Song & Zhao, 2017). Therefore, it is also vital to focus on the protection of marine organisms based on the spawning grounds and migration routes of L. polyactis.

There are two traditional patterns of population division in the evaluation and management of fishery resources, including geographical boundaries and administrative areas (Ying et al., 2011). However, these patterns are sometimes not consistent with the population structure defined by biological research, which may bring potential risks in fisheries management (Harte, Kaczynski & Schreck, 2007). Therefore, studying the population structure of marine fish based on different methods is important for defining management unit divisions. In this study, the results of STRUCTURE and 3D-FCA indicated no genetic structure consistent with the distribution pattern. Low or not significant F_ST was detected in most localities, revealing high flow among L. polyactis. This pattern was common for other marine organisms across this area (Zhang et al., 2020a; Gao et al., 2020). Zhang et al. (2020a); Zhang et al. (2020b) detected genetic variation in Engraulis japonicus in the northwestern Pacific based on restriction-site associated DNA (RAD) sequencing, suggesting high gene flow between the Bohai sea and the Yellow sea E. japonicus populations and no sign of local adaptation was found. High gene flow may be attributed to the spawning migrations and strong dispersal ability of L. polyactis (Wirgin et al., 2000).

Many studies have confirmed that it is important for the fishery management to consider the spatial structure of marine organisms (Waples, 1998; Ying et al., 2011). Species with significant genetic differentiation among populations should be managed separately, and if not, they are better managed jointly (Waples, 1998). According to the findings of the present study, L. polyactis in the coastal waters of China should be managed as a single management unit. There may be a single genetic stock of L. polyactis in different sampling sites due to long-distance migration and larval dispersal (Hewitt, 2000). Moreover, small genetic heterogeneity was detected between ZS and other sampling sites, which indicated that we should pay more attention. There were several separated spawning and feeding grounds across Zhoushan Archipelago sea area (Lin et al., 2009). Previous studies have shown that this area is the germplasm resources center of many marine organisms (Lin et al., 2009; Xiao, Ma & Dai, 2016; Xu, Song & Zhao, 2017). To protect marine living resources including L. polyactis, therefore, we suggest that the marine national nature reserve should be established in Zhoushan Archipelago sea area. Overall, effective management for important economic species should not only consider administrative division and geographical boundaries but also the biological characteristics of the species such as migration and genetic structure.

Conclusion

In summary, L. polyactis in offshore areas of China had high genetic diversity, indicating that they had large resources and population size. No genetic structure consistent with the distribution pattern was detected. However, a possible genetic differentiation existed in Zhoushan as detected in this study, which should arouse the attention of researchers. This study can provide the foundation for the division of management units and the formation of laws and regulations on fishery protection.

Supplemental Information