First genetic assessment of brackish water polychaete Tylorrhynchus heterochaetus: mitochondrial COI sequences reveal strong genetic differentiation and population expansion in samples collected from southeast China and north Vietnam.

Tylorrhynchus heterochaetus is a widespread benthic polychaete worm found in coastal brackish waters of the west Pacific. It has high ecological and economic value as a biomarker of water quality and as a high-quality feed in aquaculture and fisheries and is considered a delicacy in some areas of Asia. However, it has experienced a marked reduction in recent years due to overexploitation as well as changes in the environment and climate. Here, to comprehensively understand its genetic background and thus provide insights for better conservation and utilization of this species, we assessed the genetic variability and demographic history of T. heterochaetus individuals sampled from eight locations along the coasts of southeast China and north Vietnam based on mitochondrial cytochrome c oxidase I ( COI) sequences. We observed high haplotype diversity ( Hd), with an average of 0.926, but relatively low nucleotide diversity ( π), with a mean of 0.032 across all samples. A total of 94 polymorphic sites and 85 haplotypes were identified among 320 individuals. The pairwise genetic distances among haplotypes ranged from 0.001 to 0.067, with the high intraspecific divergence possibly reflecting geographic isolation and gene pool fragmentation. Significant genetic structures were revealed among the studied locations; specifically, the eight locations could be treated as six genetically different populations based on pairwise Φ ST results (0.026-0.951, P<0.01). A significant pattern of isolation-by-distance was detected between the genetic and geographic distances ( r=0.873, P=0.001). Three geographic lineages were defined based on phylogenetic tree and network analyses of COI haplotypes. AMOVA results indicated that genetic variations mainly occurred among the three lineages (89.96%). Tests of neutrality and mismatch distribution suggested that T. heterochaetus underwent recent population expansion. These results provide the first report on the genetic status of T. heterochaetus and will be valuable for the management of genetic resources and better understanding of the ecology and evolution in this species.

INTRODUCTION

Tylorrhynchus heterochaetus, a member of the polychaete family Nereididae (Annelida: Phyllodocida), is a widespread benthic invertebrate found in brackish waters along the coasts of China, Japan, and Southeast Asia (Tuan, 2018). Due to its high sensitivity to water quality, the species is widely used as biomarker of marine environmental conditions (Dean, 2008). Moreover, it has great potential in both aquaculture and recreational fisheries as a high-quality feed (Costa et al., 2006) and is also a favored and relatively expensive delicacy in some areas of Asia, such as Vietnam and southern China (Glasby & Timm, 2008).

During the breeding season, T. heterochaetus adults tend to swim to higher salinity waters and aggregate for spawning, and then die after releasing gametes. The salinity levels suitable for reproduction range from 10 to 13 ppt, and the hatching rate can be significantly affected by different salinities (Duan et al., 2017). For further development, however, early setiger larvae (3–5 d after hatching) prefer a low salinity environment, where they settle into mud until they reach sexual maturity in the following year. Thus, T. heterochaetus individuals are usually confined to the muddy bottom in brackish water estuaries and are thus extremely vulnerable to the impact of sudden changes in the external environment. In recent years, the increasing demand for commercial utilization, together with changes in both climate and environment, such as water pollution, ocean salinization, and habitat fragmentation, has led to a marked reduction in the natural resources of T. heterochaetus, even in historically high-yield habitats (Tuan, 2018).

Based on the above threats, a comprehensive understanding of the genetic background of T. heterochaetus could greatly facilitate its conservation and management and thus better utilization of its genetic resources. To date, however, studies on T. heterochaetus have mainly focused on reproductive physiology (Kaoawa, 1954; Okada, 1952; Sato & Osanai, 1990; Tuan, 2018) and the function of individual genes (Green et al., 1995; Suzuki & Gotoh, 1986; Suzuki et al., 1990). As such, the genetic diversity and structure of natural T. heterochaetus populations remain unclear.

Mitochondrial DNA (mtDNA) is an effective tool for molecular phylogenesis and population genetics analysis of polychaetes, such as Marenzelleria (Blank & Bastrop, 2009), Phascolosoma esculenta (Gao et al., 2018), Perinereis aibuhitensis (Liu et al., 2012), Pygospio elegans (Kesäniemi et al., 2012), Aglaophamus australiensis, and Nephtys longipes (Smith et al., 2015), due to the advantages of maternal inheritance, relatively fast mutation rate, and non-recombination (Birky et al., 1989). In the present study, we used the cytochrome c oxidase subunit I (COI) gene from mtDNA to investigate the genetic diversity, population structure, and demographic history of T. heterochaetus along the coasts of southeast China and north Vietnam. This genetic survey will provide useful information for the development of effective conservation and utilization strategies of this species.

MATERIALS AND METHODS

Sample collection and DNA extraction

A total of 320 individuals from eight locations were collected during 2016–2018 from the coastal waters of southeast China and north Vietnam. Two locations were along the coast of Vietnam (Hai Phong (HP), Nam Dịnh (ND)) and six locations were along the coast of China (Wenzhou (WZ), Fu’ an (FA), Fuzhou (FZ), Yangjiang (YJ), Zhongshan (ZS), Qinzhou (QZ)). Detailed geographic locations and sampling information are shown in Figure 1 and Table 1. The samples were stored in 95% ethyl alcohol at –20 °C until DNA extraction. Muscle tissues (~10 mg for each individual) were dissected and genomic DNA was extracted using a Genomic DNA Extraction Kit (Tiangen Biotech, DP304, Beijing, China) according to the manufacturer’ s protocols. The extracted DNA was stored at −80 °C before use.

1

Sampling locations of T. heterochaetus used in this study (indicated by solid cycles)

1

Sampling information on T. heterochaetus in this study

Location code	Locality	Country	Sample size (n)	Longitude (E)	Latitude (N)
WZ	Wenzhou, Zhejiang	China	40	121°18’	28°38’
FA	Fu’an, Fujian	China	40	119°40’	26°59’
FZ	Fuzhou, Fujian	China	40	119°18’	25°58’
YJ	Yangjiang, Guangdong	China	40	112°02’	21°51’
ZS	Zhongshan, Guangdong	China	40	113°21’	22°15’
QZ	Qinzhou, Guangxi	China	40	108°29’	21°56’
HP	Hai Phong	Vietnam	40	106°35’	20°43’
ND	Nam Dịnh	Vietnam	40	106°22’	20°20’

Corresponding location abbreviations are as in Table 1.

Bootstrap values >60% are shown near nodes (1 000 replicates), number in parentheses after location code is frequency of haplotypes. Perinereis aibuhitensis (GenBank accession No.: NC023943.1) was used as an outgroup. WZ: Wenzhou; FA: Fu’ an; FZ: Fuzhou; ZS: Zhongshan; YJ: Yangjiang; QZ: Qinzhou; HP: Hai Phong; ND: Nam Dịnh.

Gene amplification and sequencing

Partial sequences of mtDNA COI were amplified and sequenced using universal DNA primers COI-LCO1490 and HCO2198 (Folmer et al., 1994). Polymerase chain reaction (PCR) amplification was conducted in a 50 μL volume containing 0.5 µmol/L of each primer, 0.2 mmol/L of each dNTP, 30 ng of template DNA, 1.5 mmol/L MgCl2, 1×PCR buffer, and 1 unit of Taq DNA polymerase (Fermentas, Thermo Scientific, USA). PCR was conducted with an initial denaturation at 94 °C for 5 min, followed by 35 cycles of 94 °C for 30 s, 55 °C for 30 s, and 72 °C for 45 s, with a final extension at 72 °C for 3 min. The amplified products were then purified and sequenced on an ABI Prism 3730 DNA sequencer (Applied Biosystems, USA) using both forward and reverse primers individually.

Data analysis

Using the COI gene from the complete mitochondrial genome sequence of T. heterochaetus as a reference sequence (Chen et al., 2016), sequence data were aligned using Cluster X 2.0 (Thompson et al., 1997). DNA sequence polymorphisms, including number of polymorphic sites (S), number of haplotypes (H), haplotype diversity (Hd), nucleotide diversity (π), and average number of nucleotide differences (K), were estimated using DnaSP 5.10 (Librado & Rozas, 2009). To examine the genealogical relationships among mtDNA haplotypes, a haplotype network was constructed based on the median-joining algorithm in Network 5.0 (Bandelt et al., 1999). A neighbor-joining phylogenetic tree of COI haplotypes was constructed in MEGA 7.0 (Kumar et al., 2016) with the Kimura-2-parameter model (Saitou & Nei, 1987) and 1 000 bootstrap replicates. Perinereis aibuhitensis, a sea worm from the family Nereididae, was used as the outgroup (GenBank accession No.: NC023943.1). Pairwise genetic distances among haplotypes or different locations of T. heterochaetus were also calculated using MEGA 7.0 based on the Kimura-2-parameter model.

Pairwise Φ, analysis of molecular variance (AMOVA), as well as correlation between genetic and geographic distance (coastline distance between sampling sites measured by Google Earth) estimated with the Mantel test, were all calculated in Arlequin v3.5 with 10 000 permutations (Excoffier et al., 2005). For examining demographic history of T. heterochaetus, Tajima’ s D (Tajima, 1989) and Fu’ s Fs (Fu, 1997) tests were used to examine the neutrality of coding sequences with Arlequin v3.5. Additionally, mismatch distribution analysis was also performed to investigate the hypothesis of population expansion. The fitness between the observed and simulated distributions was tested using the sum of squared deviations (SSD) as well as Harpending’ s raggedness index (HRI) (Harpending, 1994). Time of population expansion was calculated with the formulas Tau (τ)=2ut and u=2µk, where t is the time since expansion, u is the mutation rate for the COI gene (Rogers & Harpending, 1992), k is the number of nucleotides, and µ is the mutation rate. In this study, a mutation rate of 2% per million years was used as suggested by Olson et al. (2009) for another polychaete Hobsonia florida.

RESULTS

Genetic diversity

A 709 bp fragment of the COI gene was obtained after alignment. The average nucleotide frequency was: T=29.7%, C=24.4%, A=29.1%, and G=16.9%. Of the 320 sequences from the eight locations, a total of 94 variable sites were observed, including 32 singleton variable sites and 62 parsimony informative sites. Genetic diversity indices are presented in Table 2. A total of 85 haplotypes (GenBank accession No.: MK614603–MK614686) were identified, most of which (68 out of 85) were represented by a single individual, and pairwise genetic distances among haplotypes varied from 0.001 to 0.067. The number of haplotypes at a location ranged from 8 to 23. Overall, most locations showed moderate to high haplotype diversity (0.237–0.949) due to the large number of rare haplotypes. However, nucleotide diversity was relatively low, ranging from 0.000 4 to 0.009 12, across all studied locations. The WZ location exhibited the highest haplotype diversity (Hd=0.949) and nucleotide diversity (π=0.009).

2

Genetic diversity of eight T. heterochaetus locations based on COI gene sequences

Location (lineage)	n	S	H	Hd	π	k
n: Sample size; H: Number of haplotypes; S: Number of polymorphic sites; Hd: Haplotype diversity; π: Nucleotide diversity; k: Mean number of pairwise differences. WZ: Wenzhou; FA: Fu’ an; FZ: Fuzhou; ZS: Zhongshan; YJ: Yangjiang; QZ: Qinzhou; HP: Hai Phong; ND: Nam Dịnh. Lineage A=WZ+FA+FZ, Lineage B=ZS+YJ, Lineage C=QZ+HP+ND.
WZ	40	32	23	0.949	0.009 12	6.465
FA	40	13	10	0.776	0.004 74	3.362
FZ	40	16	13	0.783	0.004 70	3.329
ZS	40	18	6	0.237	0.001 27	0.900
YJ	40	12	13	0.654	0.001 40	0.994
QZ	40	11	10	0.441	0.001 04	0.735
HP	40	17	16	0.645	0.001 59	1.124
ND	40	7	8	0.618	0.001 05	0.744
Lineage A	120	42	39	0.861	0.007 11	5.043
Lineage B	80	29	18	0.660	0.001 75	1.239
Lineage C	120	29	29	0.767	0.002 74	1.942
Overall	320	94	85	0.926	0.032 15	22.794

Population structure and phylogenetic analysis

Pairwise Φ values ranged from –0.013 to 0.951 for all locations and were highly significant (P<0.01), except for two close location pairs (FA-FZ and HP-ND), whereas pairwise genetic distance varied from 0.001 to 0.063 (Table 4). Thus, all eight locations sampled could be treated as six genetically different populations (WZ, FA+FZ, ZS, YJ, QZ, HP+ND, Table 4). The topologies produced from both the haplotype phylogenetic tree (Figure 2) and haplotype median-joining network (Figure 3) showed a consistent structure. All eight locations could be characterized into three geographically distinguishable lineages (i.e., lineage A: WZ, FA+FZ; lineage B: ZS, YJ; lineage C: QZ, HP+ND, Figures 1–3). The AMOVA results based on the three lineages revealed that genetic variation mainly occurred among lineages (89.86%). Only 6.67% and 3.47% of variation occurred within populations and among populations within lineages, respectively (Table 3), suggesting strong genetic divergence among the different regions. Additionally, a significant pattern of isolation-by-distance was detected across all studied locations using the Mantel test (r=0.873 1, P=0.001).

4

Pairwise Ф (below diagonal) and genetic distance (above diagonal) among different T. heterochaetus locations

Location code	WZ	FA	FZ	ZS	YJ	QZ	HP	ND
ФST values in bold type indicate statistical significance (P<0.01). WZ: Wenzhou; FA: Fu’ an; FZ: Fuzhou; ZS: Zhongshan; YJ: Yangjiang; QZ: Qinzhou; HP: Hai Phong; ND: Nam Dịnh.
WZ	−	0.009	0.009	0.058	0.057	0.061	0.061	0.060
FA	0.226	−	0.005	0.059	0.058	0.062	0.063	0.063
FZ	0.238	−0.013	−	0.059	0.058	0.062	0.063	0.063
ZS	0.906	0.946	0.946	−	0.002	0.020	0.019	0.019
YJ	0.903	0.944	0.944	0.378	−	0.022	0.021	0.021
QZ	0.912	0.950	0.951	0.941	0.942	−	0.005	0.005
HP	0.907	0.944	0.947	0.922	0.926	0.711	−	0.001
ND	0.911	0.951	0.951	0.936	0.938	0.771	0.017	−

2

Neighbor-joining phylogenetic tree of T. heterochaetus based on COI haplotypes

3

Median-joining network of T. heterochaetus based on COI haplotypes

3

AMOVA results of T. heterochaetus based on mtDNA

Source of variation	d. f.	Sum of squares	Variance component	Percentage of variation (%)	Fixation indices
*: P<0.05;d. f.: Degree of freedom; Va: Variance component due to differences among lineages, Vb: Variance component due to differences among populations within lineages, Vc: Variance omponent due to differences among individuals within populations; FSC=Vb/(Vb+Vc), FST=(Va+Vb)/(Va+Vb+Vc), FCT =Va/(Va+Vb+Vc).
Among lineages	2	3171.121	14.871 Va	89.86	FSC =0.342*
Among populations within lineages	5	120.279	0.573 Vb	3.47	FST =0.933*
Within populations	312	344.225	1.103 Vc	6.67	FCT =0.818*
Total	319	3635.625	16.548

Each circle represents a haplotype and area is proportional to its frequency. Colors of circles denote geographic origin, white dots represent hypothetical intermediate haplotypes. WZ: Wenzhou; FA: Fu’ an; FZ: Fuzhou; ZS: Zhongshan; YJ: Yangjiang; QZ: Qinzhou; HP: Hai Phong; ND: Nam Dịnh.

Historical demography

As genetic differences were not significant in the two geographically close location pairs FA-FZ and HP-ND, we treated them as a whole when analyzing demographic history. Both Tajima’ s D and Fu’ s F tests showed negative values for all genetically differentiated populations (Table 5), indicating departure from mutation-drift equilibrium and possible population demographic expansion. Furthermore, mismatch distribution did not differ significantly from the model of sudden expansion when using either SSD or HRI for goodness-of-fit (Table 5), further supporting the hypothesis of population expansion in T. heterochaetus. The τ value across populations varied from 0.969 to 13.936. Using the mutation rate of 2% per million years, it was estimated that the T. heterochaetus population expansion occurred about 17 000–246 000 years ago in the middle to late Pleistocene.

5

Parameters of neutrality test and mismatch distribution analysis for six T. heterochaetus populations

Population	Neutrality test			Mismatch distribution analysis				Expansion time
	Tajima’s D	Fu’s Fs		SSD	HRI	Tau (τ)		t (Ma)
*: P<0.05, **:P<0.01. WZ: Wenzhou; FA: Fu’ an; FZ: Fuzhou; ZS: Zhongshan; YJ: Yangjiang; QZ: Qinzhou; HP: Hai Phong; ND: Nam Dịnh.
WZ	–0.486	–7.786**		0.023	0.024	6.425		0.113
FA+FZ	–0.395	–5.076*		0.052	0.123	4.834		0.085
ZS	–2.571**	–1.607		0.002	0.408	13.936		0.246
YJ	–1.995**	–11.090**		0.001	0.06	1.062		0.019
QZ	–2.172**	–7.960**		0.000 1	0.124	1.103		0.019
HP+ND	–2.333**	–24.297**		0.003	0.081	0.969		0.017

DISCUSSION

This study examined the population genetic variability of T. heterochaetus using mtDNA for the first time. Compared with other polychaetes, the northern location (FA, FZ, WZ) in lineage A showed high haplotype and nucleotide diversities (Hd=0.861, π=0.007 11; Figure 1, Table 1), differing from several other species, such as Branchipolynoe symmytilida (Hd=0.970, π=0.007; Plouviez et al., 2009), Owenia fusiformis clade 1 and 2 (Hd=0.924–0.978, π=0.007 2–0.007 7; Jolly et al., 2006), and Aglaophamus australiensis (Hd=0.78–0.99; Smith et al., 2015). However, remarkably lower nucleotide diversity was detected in the southern locations (YJ, ZS, QZ, HP, and ND; Figure 1, Table 1), which may be due to the commercial overexploitation of wild stocks as T. heterochaetus is a very popular specialty food in this region. Although small-scale artificial breeding of T. heterochaetus has been conducted in southern China and Vietnam, production still comes primarily from natural exploitation.

Previous meta-analysis revealed that overharvesting can drive the decay of genetic diversity in many highly abundant marine species (Pinsky & Palumbi, 2014). Genetic diversity is closely related to the long-term adaptability and survivability of populations, especially in suddenly and drastically changing marine environments (Barrett & Schluter, 2008). Thus, the strikingly varied genetic diversity among different T. heterochaetus locations, as shown in this study, should be an important consideration in relation to conservation strategies or artificial breeding programs for future aquaculture enhancement.

According to the pairwise Φ values (Table 4), we identified six genetically different populations (WZ, FA+FZ, ZS, YJ, QZ, HP+ND) among the eight locations in this study. Given the geographic positions and significant divergences among the populations, Leizhou Peninsula and Hainan Island appear to be remarkable oceanographic barriers that block gene flow among the different populations inside the Beibu Gulf and others (Figure 1), as also demonstrated in Perinereis aibuhitensis populations from the southern coastal zone of China (Liu et al., 2014). Over broader geographic scales, long distance/isolation between populations may play an important role in the high level of genetic structuring. In fact, an evident isolation by distance pattern was detected using the Mantel test (r=0.873 1, P=0.001). As for the genetic homogeneity in the two location pairs FA-FZ and HP-ND, the short distance along the coastline (<60 km, measured by Google Earth) may not be strong enough to prevent effective gene flow between the two adjacent estuaries.

Geographically, the southern populations (YJ, ZS, QZ, HP+ ND) are isolated from the northern populations (FA+FZ, WZ) by several main water systems, such as the Pearl River, Hanjiang River, and Minjiang River. Such isolation could be a strong barrier to gene flow. In addition, according to our resource investigations over the last five years, many unsuitable habitat patches exist along the coastline among the sampling sites. This habitat discontinuity could also lead to the significant genetic divergences observed in different populations.

Apart from geographic factors, its narrow habitat niche is also suspected to be partly responsible for the high structuring among T. heterochaetus populations. The pairwise Φ results among sampling sites suggest striking genetic divergence, except for the FA-FZ and HP-ND pairs, which exhibit close geographical positions (Figure 1, Table 4), implying that effective dispersal of T. heterochaetus over large distances is limited. In marine invertebrates with sedentary adults, a longer planktonic larval stage is correlated with increased dispersal capacity and higher connectivity among populations (Cowen & Sponaugle, 2009; Kyle & Boulding, 2000). However, pelagic larva duration may not always be a good predictor of gene flow and population structure (Weersing & Toonen, 2009). Although it has a relatively long planktonic larval stage (~20 d), wild T. heterochaetus individuals favor the muddy bottoms of brackish environments with low salinity, such as estuaries (Tuan, 2018). This specialized habitat and narrow salinity requirement may limit colonization potential of the dispersed larvae. A high level of genetic divergence among invertebrates inhabiting estuarine systems is not uncommon (e. g., Darling et al., 2004; Olson et al., 2009; Virgilio et al., 2006), as estuaries represent spatially discrete habitats that tend to restrict gene flow and lead to different levels of isolation (Bilton et al., 2002).

In the present study, the phylogenetic tree, haplotype network, and AMOVA results indicated that the eight geographical locations of T. heterochaetus along the coasts of southeast China and north Vietnam could be characterized into three distinguishable lineages (i.e., A, B, C; Figures 2–3), corresponding to different geographic regions (A: East China Sea; B: Eastern Leizhou Peninsula; C: Western Leizhou Peninsula). Pairwise genetic distances among the 85 haplotypes ranged from 0.001 to 0.067, some of which were higher than 0.02, which is a commonly used standard for species identification (Avise, 2000). Hebert et al. (2003) suggested that higher intraspecific divergences ordinarily occur as geographic isolates, reflecting gene pool fragmentation in the origin of species in past episodes. In the last glacial maximum of the late-Pleistocene, the mean sea level was about 120 m lower than the current level (Fairbanks, 1989). During this period, the East China Sea shrank to a long, narrow ocean trough (Okinawa trough) and was isolated with the South China Sea (Koizumi et al., 2006). In addition, Hainan Island and the Chinese mainland were connected by the Qiongzhou Strait land-bridge (Voris, 2000). These historical geographic barriers may have blocked gene flow among different coastal regions and led to the deep lineages observed in the current T. heterochaetus populations.

Population genetic analysis can provide guidelines for strategies in species conservation and germplasm resource management (Loeschcke et al., 2013; Whiteley et al., 2006). Different lineages can show varying degrees of physiological adaptations (Méndez et al., 2001). Thus, we suggest that the genetically divergent lineages confirmed in this study should be treated as independent management units regarding conservation issues. Based on the relatively low level of genetic diversity in samples from ZS and QZ, special attention should be paid to these two populations. As a typical r-selected species, T. heterochaetus is characterized by its short life span (one year) and hypersensitivity to environmental deterioration, habitat fragmentation, and overfishing (Tuan, 2018). To effectively protect the local populations of T. heterochaetus, water pollution control, overfishing reduction, and habitat preservation are all high priorities for future conservation.

Previous studies have suggested that the genetic structure of marine invertebrates can be shaped by various factors, such as currents, geographic segregation, life history characteristics, human-mediated transfer, and local selection (Selkoe & Toonen, 2011; Simon & Sato-Okoshi, 2015; Zakas & Wares, 2012; Zardi et al., 2007). Therefore, further research is still needed to identify the underlying mechanisms that may contribute to the substantial genetic heterogeneity in T. heterochaetus.

Both mismatch distribution analysis and significant negative values of neutrality tests (Tajima's D and Fu’ s Fs) indicate a pattern of recent population expansion in T. heterochaetus. This hypothesis was also supported by the star-shaped haplotype network (Figure 3), a characteristic of exponential population growth (Slatkin & Hudson, 1991). The demographic history was reflected in the genetic indices of this species, which showed low nucleotide diversity (π=0.032 15) but high haplotype diversity (Hd=0.926) (Table 2). As described by Grant & Bowen (1998), high Hd and low π can be attributed to rapid population expansion that enhances the retention of new mutations, which is consistent with the large number of unique and low-frequency haplotypes found within lineages in the present study (Figure 3).

Due to the lack of fossil and geological records, which are major obstacles for phylogeographic analysis of marine invertebrates (Provan & Bennett, 2008), the precise population expansion time based on a species-specific molecular clock for T. heterochaetus is not available. Based on a mutation rate of 2% per million years, our rough estimation suggests population expansion for T. heterochaetus in the middle to late Pleistocene, a period dominated by glaciation cycles (Imbrie et al., 1992) and periodic climatic oscillations, which may have impacted the distribution of T. heterochaetus. A similar demographic history pattern has also been found in other marine invertebrates along the Chinese coast (Gao et al., 2018; Liu et al., 2012).

CONCLUSIONS

Based on partial sequences of the COI gene, this study provides preliminary information on the genetic status of T. heterochaetus collected from the coasts of southeast China and north Vietnam. The genetic diversity in this species was highly variable and we identified six populations with significant genetic differences. These six populations could be divided into three genetically divergent lineages, corresponding to three geographic regions. We suggest that the lineages confirmed in this study should be managed separately from a conservation point of view. For future studies, the application of multiple markers with higher resolution (e. g., microsatellites and single-nucleotide polymorphisms) and greater spatial sampling, as well as a broader understanding of biological and ecological factors, should provide a more detailed assessment of the T. heterochaetus population structure.

COMPETING INTERESTS

The authors declare that they have no competing interests.

AUTHORS' CONTRIBUTIONS

Z.N.M. and S.Y. conceived and designed the experiment; X.H.C. and S.Y. performed the experiments; X.H.C. and Z.N.M. wrote the manuscript; W.Y., Y.Y.S., and R.W.X. contributed reagents/materials/analysis tools; B.F. and L. W. collected the samples and made English corrections. All authors read and approved the final version of the manuscript.

ACKNOWLEDGEMENTS

We thank Bo Li from Yangjiang Polytechnic for his assistance in field investigation.