Genetic diversity and population structure of Lethrinus lentjan (Lethrinidae) in Vietnam based on SNP markers

Abstract

The pink-ear emperor, Lethrinus lentjan (Lacepède, 1802) is an important commercial species for marine fisheries in subtropical and tropical regions of the Indian and Indo-Pacific oceans, including Vietnam coastal waters [1]. Like the majority of lethrinids, this species occurs in various habitats throughout its lifespan, young settle and develop in shallow coastal waters, most often over sand flats and in mangrove swamps, before migrating to coral reefs and surrounding habitats [2], and has a pelagic larval development period of approximately 30 days [3]. Complex seasonal induced surface currents and Mekong River outflow are supposed to be the genetic barriers that may lead to the population dis-connectivity of Lethrinus lentjan. Although heavily exploited, their genetic diversity and population structuring along the Vietnam coastline are unclear. This study investigates the genetic population structure of the pink-ear emperor, Lethrinus lentjan in Vietnam using a panel of Single Nucleotide Polymorphisms (SNPs) generated from restriction site-associated DNA sequencing (RADseq).  

A total of 85 individuals were collected from four locations representing the Central (Da Nang and Nha Trang), Northeast (Vung Tau), Southwest (Phu Quoc), Vietnam. The sampling location is shown in Fig. 1A. Initially, 277,586 raw SNPs were detected. Following filtering steps, 62 individuals were successfully genotyped at 1,911 valid SNPs. After removing 405 outlier loci detected by LD network, BayeScan and Lositan analyses, 1,506 putatively neutral SNPs from 62 individuals were used to assess the genetic diversity and population structure of L. lentjan. The mean observed heterozygosity (Ho) and expected heterozygosity (He) per locus were 0.180 and 0.209, respectively. Observed/ expected heterozygosity (Ho/He) within sites ranged from 0.170/0.202 (Southeast) to 0.188/0.213 (Central). Rates of inbreeding were highest in the central (Gis = 0.117), and lowest in the southeast population (Gis = 0.158) (Table 1). Pairwise genetic differentiation between populations were not significant (P > 0.05) (Table 2). High levels of relatedness were found, including 21 putatively related pairs (1 identicals pair, 1 twin pair, 5 sib-plus pairs, and 14 half-sibling pairs). For that, 14 individuals were removed to the final 48 individuals retained. Due to the imbalanced numbers of related pairs among populations leading to reduced sample size and avoiding positive bias in estimates due to underestimating relatedness in the overall population [4], further analyses were run with two datasets, one containing all individuals (with related pairs) and one with one putative individual removed per related pair (related individuals removed). AMOVA results (data not showed) of two data set showed the largest component of total genetic variation was observed within individuals (86.77% with related pairs and 86.21% when related individuals removed) and highly significant (F_IT=0.312 and 0.318, respectively, P<0.001). With two datasets, among populations (F_ST) differentiation was not significant. Bayesian model-based clustering (STRUCTURE) and Discriminant Analysis of Principal Components (DAPC) analysis both indicate the lack of population structure at the geographic locations surveyed, suggesting the presence of a single panmictic population in both putative neutral (Fig. 1B,C), and outlier loci from two data set (Fig. 1D). These results suggested highly gene flow of L. lentjan along the Vietnamese coastline, larval dispersal may be facilitated by ocean currents, and not to be restricted by Mekong river outflow. More research needs to be performed to extend the spatial and temporal sampling implemented for fisheries and conservation management.

Fig.1. Sampling map and population structure of Lethrinus lentjan in Vietnam with related pairs (left) and related individuals removed (right). (A) Sampling map of Lethrinus lentjan and surface currents following northeast (bold line) and southwest (dash line) monsoons, INSET: Mekong River (black box) in Mekong delta, Vietnam; (B) The bar plot showing individual assignments to inferred clusters (optimal K = 2) in the program STRUCTURE using the neutral SNP panels. Each genotype is represented by a single vertical bar; Scatter plot from DAPC following neutral SNP panels (C) and outlier loci (D), the percentage of variability explained by each coordinate is shown in brackets

Table 1. Lethrinus lentjan sample site information and genetic diversity. Number of alleles (Na), effective number of alleles (Ne), observed (Ho) and expected (He) heterozygosity, percentage of polymorphic loci (%P), and inbreeding coefficient (Gis) using 1,506 neural SNP panels

Pop ID

N

Na

Ne

Ho

He

%P

Gis

Central

26

1.966

1.321

0.188

0.213

96.61

0.117

Southeast

11

1.792

1.312

0.170

0.202

79.15

0.158

Southwest

25

1.954

1.320

0.182

0.211

95.35

0.137

Total/ Mean

62

1.904

1.318

0.180

0.209

90.37

0.139

Table 2. Pairwise values of F_ST (above the diagonal) and their respective P-values (below the diagonal)

With related pairs

With related individuals removed

Pop ID

Central

Southeast

Southwest

Central

Southeast

Southwest

Central

--

0.063

0.180

--

0.297

0.180

Southeast

0.0084

--

0.180

0.0089

--

0.667

Southwest

0.0044

0.0068

--

0.0048

0.0048

--