Genetic Diversity and Population Structure of Coilia mystus in Yangtze Estuary and Its Adjacent Waters Based on Mitochondrial COI Gene 

Coilia mystus is a well-known estuarine migratory species of the family Engraulidae and under the order of Clupeiformes (He et al., 2008; Chen and Cheng, 2012). It is a short-distance anadromous fish, which migrates to estuarine brackish water to spawn in spring. After that, most of the adult populations and the spawning products descend to the sea (Cheng et al., 2008). C. mystus is a commercially important species in Yangtze Estuary, which contributes approximately 48.6% of the total fish and shrimp caught in the estuarine area (Cui et al., 2015). To date, numerous studies were focusing on its reproductive biology (He et al., 2011), nutritional composition (Song et al., 2018; Song et al., 2020), feeding habit (Cui et al., 2015) and migration (Yang et al., 2006). Cheng et al., (2005) have found differences in morphology between the ChangJiang and the ZhuJiang populations of C. mystus. Some studies also hypothesized that the populations diverged at the subspecies of ChangJiang and ZhuJiang populations (Cheng et al., 2008). Previous studies have made a comparison of C. mystus populations that are geographically far apart (Cheng et al., 2005; Cheng et al., 2008). However, the study on the population diversity of C. mystus in the waters close to each other has not been reported. In order to thoroughly elucidate the genetic diversity and evolutionary divergence of different C. mystus populations in the adjacent waters, it is necessary to use more stable molecular markers and more samples.

At present, many molecular markers have been used in elucidating genetic diversity (Xu et al., 2012; Devi et al., 2018; Ni et al., 2018) and resolving molecular phylogeography and population history (Mao et al., 2011; Singh and Kaur, 2020). Mitochondrial cytochrome oxidase subunit I (COI) is well characterized molecular marker and frequently used for population and phylogenetic studies (Sun et al., 2012; Xu et al., 2012; Singh and Kaur, 2020). But in terms of genetic diversity of C. mystus, only few studies about the mitochondrial (Cyt b and 16S rRNA) and microsatellite DNA diversity (Cheng et al., 2008; Chen and Cheng, 2012), while the population genetic structure analysis based on COI sequences have not yet been seen. In the present study, genetic divergences among four C. mystus populations from close geographical areas were assessed by COI sequence. Understanding the population genetic structure of C. mystus will conducive to the establishment of management unit for better protection and utilization of this species.

Materials

Samples of 144 C. mystus were obtained from 4 geographical localities in Yangtze Estuary and its adjacent waters (LS, Lvsi; CM, Chongming; ZS, Zhoushan; WZ, Wenzhou) from September to October in 2018. Detailed information concerning sample collection is shown in Fig 1. The back muscle of fresh fish was collected in the field and stored in 95% ethanol immediately and then transported to East China Sea Fisheries Research Institute for DNA extraction and other follow-up processes.


 
DNA extraction, amplification and sequencing
 
Total genomic DNA was isolated from muscle tissue using DNA extraction kits (Tiangen, Beijing, China). Primers were designed according to Ward et al., (2005). The common pairs of primers F₁ and R₁ were synthesized by Sangon Biological Engineering (Shanghai) Co., LTD, China. The primers F₁: 5'-TCA ACC AAC CAC AAA GAC ATT GGC AC-3', R1: 5'-TAG ACT TCT GGG TGG CCA AAG AAT CA-3'; were used for amplification of a target regions. Amplifications were performed with 50 µl of reaction mixture contained 25 μl Premix Taq (1.25 U, Takara), 2 μl forward and reverse primers F₁ and R₁. An aliquot of DNA template was added according to DNA purity and quality, at last, ddH2O was supplemented to 50 μl of the total volume. The PCR amplification condition as follows: Initial denaturation at 94oC for 5 min followed by 30 cycles of denaturation at 94oC for 30 s, annealing at 55oC for 30 s, elongation at 72oC for 1 min and final elongation at 72oC for 10 min. After visualization in the Gel Imaging System (BIO-RAD, GelDoc XR+, USA) by electrophoresis on 1% agarose gel, the appropriate PCR amplification product was separated and purified using the DNA Gel Extraction Kit (Shanghai Biotech, China). Gene sequencing was performed on a Beckman Coulter CEQ 8000 automatic sequencer (USA).
 
Data analyses
 
Sequences were edited and aligned by visual inspection using Clustal X (Jeanmougin et al., 1998). Nucleotide composition and variable sites were performed with MEGA 5.1 (Tamura et al., 2011). Haplotype diversity (h) and nucleotide diversity (p) were calculated in Dnasp 5.1 (Librado and Rozas, 2009). Genetic relationships among individuals were constructed based on the neighbor-joining (NJ) method (Saitou and Nei, 1987). The phylogenetic and geographical relationships of the haplotype sequences were showed with the median-joining method in Network 4.1 (Rohl, 2003). The geographical structure of C. mystus was investigated using the hierarchical analysis of molecular variance (AMOVA). The significance of the fixation index was tested by 1000 permutations of the data set. The population genetic structure within the four localities was revealed by pairwise fixation index (FST) statistics in ARLEQUIN version 3.0 (Excoffier et al., 2005).

The present study used 576 bp of COI sequence to investigate the genetic diversity and population structure of 144 C. mystus individuals from 4 localities. The COI sequence is polymorphic and 58 variable sites and 32 haplotypes were detected. The C. mystus populations from the north localities of LS, CM and ZS exhibited high haplotype diversity (h>0.5) and low nucleotide diversity (p<0.005) (Table 1). This result is consistent with previous findings, which also reported high haplotype diversity and low nucleotide diversity among the same species distributed in 3 Chinese estuaries (Cheng et al., 2008). While the C. mystus population from the south locality of WZ exhibited low haplotype diversity (h<0.5) and low nucleotide diversity (p<0.005). Table 1 showed that the p values for C. mystus in this study were lower than that reported for many other marine fishes. (Sun et al., 2012; 2013).



Phylogenetic analysis with NJ tree showed that the individuals grouped into two distinct clusters, basically in accord with the geographical sources of the samples, as the individuals from LS, CM and ZS grouped into one cluster and the individuals from WZ grouped into the other cluster (Fig 2). The resultant network exhibited two star-like patterns, one of which surrounding haplotype H_1, with the haplotypes from LS, CM, ZS populations and the other of which surrounding haplotype H_16, with the haplotypes only from the WZ population (Fig 3). The NJ tree of the COI sequence from all individuals and the median-joining networks of all haplotypes suggested that there is a distinct phylogeo--graphic structure across the 4 populations. Table 2 showed the AMOVA analysis results for COI, which exhibited that among group variation (95.888%) accounts for significantly more of the total genetic variation than variation within groups (0.022%) and that within populations (3.990%). The results of the AMOVA revealed that there were two groups, the populations from LS, CM and ZS as the north group and WZ as the south group. In this study, FST in the three north populations showed no significant difference (P>0.05), while compared with the southern population, which exhibited significant divergence (P<0.001) (Table 2). These results were similar with many marine fishes including Coilia ectenes (Ma et al., 2010), Pampus argenteus (Zhao et al., 2011), Mugil cephalus (Sun et al., 2012) and Nibea albiflora (Xu et al., 2012), which exhibited significant genetic divergence.







Genetic diversity is influenced by many factors, including historical factors, anthropogenic activity, habitation and a low rate of mitochondrial evolution (Grant et al., 2006). Historical factors may play an important role in determining the patterns of genetic variability (Xiao et al., 2009; Yamaguchi et al., 2010). Populations of fishes that experienced rapid expansion following a period of low effective population size often display high haplotype but medium to low nucleotide diversities (Grant and Bowen, 1998). Moreover, the haplotype network was characterized by two star-like phylogenesis, a typical signature of a past recent population expansion following a population bottleneck (Avise, 2004). Another factor likely to be responsible for the low genetic diversity in C. mystus is overexploitation (Cheng et al., 2008; He et al., 2011), which is known to be one of the main causes of extinction of marine species (Rodrigues et al., 2008). Also, the genetic divergence for different populations often attributed to the presence of geographic barriers or temporal reproductive isolation. The WZ samples were collected at the inner edge of the Gulf of Wenzhou (Fig 1). The Gulf and islands around the region of WZ seem to act as barriers to gene flow, which could result in high genetic and morphological divergence between WZ and the northern populations (Yang et al., 2019). We speculated that the physical barriers restricted the gene flow in different group populations. In this study, the gene flow values were very low between the north and south groups (Table 3). This suggests that the different groups have isolated genetic structure and restricted gene flow. Ecological habits, limited long-distance migration and long-term separation have likely played an important role in producing the current genetic structure.

The results obtained in this study identified low genetic diversity and significant genetic differentiation in C. mystus between different groups from Yangtze Estuary and its adjacent waters. The low genetic diversity raised concerns over the conservation status of C. mystus. Therefore, fisheries management strategies should be undertaken to protect this species and prevent the loss of genetic variation. This study also showed that subdivisions exist between different groups, so it should be considered as different management unit in the further conservation and management work. C. mystus is an important commercial and ecologically species and the evaluation of its genetic diversity and population structure is therefore very important. Genetic data on wild C. mystus from Chinese coastal waters has broad implications for genetic assessment, fisheries management and conservation.

This work was supported by the Shanghai Nature Fund Project (19ZR1470200), the special research fund for national non-profit institutes (East China Sea Fisheries Research Institute) (2019T05), the National Key Research and Development Programme of China (2019YFD0901201) and the Key Project of Promoting Agriculture by Science and Technology in Shanghai (2017-02-08-00-07-F00075).