Genetic diversity and population structure of Kawakawa Euthynnus affinis (Cantor, 1849) in Malaysia and its surrounding waters inferred by mitochondrial DNA cytochrome oxidase I and cytochrome b genes

1 Introduction

The assessment of animal population distribution and genetic structure is of utmost importance in the formulation of effective conservation strategies, especially for species that have been subjected to excessive exploitation. This is crucial for maintaining the biodiversity of species and populations and safeguarding them from commercial exploitation (Turan et al., 2005; Ha et al., 2020). The concept of a “fish stock” refers to a group of individuals within a species that are randomly mated and exhibit temporal or geographic integrity. This concept is of particular interest in the context of species conservation, as highlighted by Ihssen et al. (1981). The stock structure of a species is defined by its geographic distribution and genetic composition. The recognition of species stock structure holds significant importance for a multitude of reasons. For example, this technology facilitates the discernment of multiple populations or subpopulations that may possess diverse ecological, genetic, or management needs, and holds significant value in the development of customised conservation and management strategies for distinct populations (Khan et al., 2012). Moreover, comprehending the stock structure of a species can indicate areas characterised by substantial genetic diversity, a crucial factor in the conservation of a species' long-term well-being and adaptability. The determination of the degree of relatedness and gene flow among different populations can be facilitated through the utilisation of stock structure data, in which this information provides a comprehensive assessment of the genetic health of a species (Cronin-Fine et al., 2013; El Mghazli et al., 2021).

Researchers have proposed various techniques to ascertain the presence of marine and freshwater fish stocks. The earlier mentioned techniques encompass morphometric techniques (Griffiths et al., 2017), geometric morphometrics (Mohammadi-sarpiri et al., 2021; Binashikhbubkr et al., 2022), meristic analysis (Gain et al., 2017), otolith shape analysis (Wujdi et al., 2022), and DNA analysis techniques, specifically focusing on mitochondrial genes (Johnson et al., 2016; Kasim et al., 2020; Jamaludin et al., 2022). The utilization of DNA coding has been widely employed in various fields such as forensics, population ecology, phylogenetics, and biodiversity research since the early 2000s (Hürkan, 2020). Specifically, the adoption of DNA coding techniques that are dependent on the COI and Cyt b genes present in mitochondrial DNA is crucial in the process of species identification, with a particular emphasis on fish species (Hebert et al., 2003). Furthermore, the conservation of the COI and Cyt b genes within the mitochondrial genome renders them indispensable for the classification and identification of diverse animal species (Hebert et al., 2010). Recent studies conducted on marine fish have provided evidence that the utilization of the COI and Cyt b genes can serve as an effective tools for species identification, population genetics analysis, and fisheries management (Kunal and Kumar, 2013; Halim et al., 2022).

Kawakawa, scientifically referred to as Euthynnus affinis, is a member of the Scombridae family and is widely recognized as a species of tuna. The species was initially documented by Cantor (1849), and it exhibits a wide distribution throughout the Indo-Pacific region, specifically in subtropical and tropical seas. Kawakawa holds considerable commercial importance within Malaysia and the adjacent seas. This species inhabits continental shelf seas and is renowned for its superior flesh, which is frequently employed in various dishes such as fresh cuisine, sashimi, and canned products (Masazurah et al., 2012). The Kawakawa populations in Malaysian waters, Indian Ocean, and Java Sea are subjected to pressure from overfishing and illegal, unreported, and unregulated fishing activities (Effarina et al., 2020; Mardlijah et al., 2022). Therefore, it is of utmost importance to implement effective surveillance and management strategies for these species to promote sustainable utilization of resources. Despite these challenges, Kawakawa continues to hold significant importance for the fishing industry and the local economy in Malaysia.

Mitochondrial DNA (mtDNA) has been employed in various prior studies conducted in Malaysia and adjacent seas to underscore the stock structure and genetic diversity of E. affinis. In a study conducted by Santos et al. (2010), DNA control region markers, specifically the D-loop, were employed to examine the genetic diversity of 35 E. affinis samples collected from the Philippines, as well as 13 samples from Peninsular Malaysia, specifically Pangkor Island and Penang Island. Similarly, the population structure of E. affinis in two states in Peninsular Malaysia (Perlis and Penang) was examined by Masazurah et al. (2012) using cytochrome b (cyt b) markers. The findings derived from the investigation indicate a dearth of comprehension regarding the stock structure of E. affinis, genetic diversity, and phylogenetic connections of the species in Malaysia and the adjacent waters. Thus, the results of this research hold significant implications for fisheries managers in devising suitable policies at both regional and international levels to effectively address the irreversible decline of the resource. Moreover, the results obtained from this study have the potential to contribute to the effective management and conservation strategies for the E. affinis population within Malaysian waters.

The objective of this study is to employ mitochondrial cytochrome oxidase I (COI) and Cyt b markers to examine the genetic diversity and population structure among 20 populations of E. affinis originating from Malaysia and the adjacent waters. This study represents the inaugural investigation employing the COI and Cyt b genes, thereby offering valuable insights into the genetic diversity and population structure of E. affinis within Malaysia and its adjacent waters.

2 Materials and methods

2.2

2.2 Sampling and study area

The collection of specimens belonging to the species Euthynnus affinis was undertaken between November 2019 to September 2021. The collection encompassed a total of 20 distinct commercial fish landing sites. The samples were gathered from the landing sites pinpointed in the annual report by the (DoFM, 2021), which furnished details on the availability of fish at these areas. The aforementioned sites comprised a total of 11 landing sites located in Peninsular Malaysia, eight landing sites situated in Malaysian Borneo (specifically Sabah and Sarawak), and one landing site located in the Arabian Sea (Mukalla, Yemen). Samples were collected from five main geographical regions, specifically the Straits of Malacca, the South China Sea, the Sulu Sea, the Celebes Sea, and the Arabian Sea. In this study, samples of E. affinis were obtained from five distinct landing sites situated in the Straits of Malacca: Batu Maung (Penang), Lumut (Perak), Sungai Besar and Kuala Selangor (Selangor), and Kukup (Johor). In addition, a total of 10 landing sites have been designated in the South China Sea, namely Pasir Puteh and Tok Bali (Kelantan), Pulau Tenggol and Pantai Kijal (Terengganu), Kuantan (Pahang), Endau (Johor), Bintawa, Pulau Bruit, and Mukah (Sarawak), as well as Kota Kinabalu (Sabah), two landing sites from the Sulu Sea: Kudat and Sandakan (Sabah), two landing sites from the Celebes Sea: Lahad Datu and Tawau (Sabah). Inclusion of a landing site from the Arabian Sea (Mukalla - Yemen) in the sample collection was ultimately accomplished (Table 1, Fig. 1). All samples were morphologically identified and validated based on Collette and Nauen (1983). Distinguishing this particular species from other Thunnus species can be achieved through various identifying characteristics, including size, gill rakers, gill teeth, fins, vertebrae, and colouration (Collette and Nauen, 1983). The specimens collected in Malaysia were subsequently relocated to the Molecular Ecology Laboratory at Universiti Sains Malaysia's School of Biological Sciences. In order to enhance visibility, the fish samples underwent a process of rinsing with running water, gentle tapping, and subsequent placement on the left side of a flat surface featuring a black background. The examination and identification of samples collected from Mukalla was conducted in the Biological Laboratory at the Faculty of Science, Hadhramout University, Yemen. For each specimen obtained, a segment measuring approximately 1–2 cm from the right pectoral fin was excised and stored in 95 % ethanol for the purpose of DNA extraction.

Table 1 Sampling locations, coordinates, and sample size (N) of 20 populations of E. affinis based on COI and Cyt b genes.

Sampling locations	Coordinates	Marine region	N
			COI	Cytb
Batu Maung, Penang (BM)	5° 16′ 60.00″ N; 100° 16′ 60.00″ E	M	24	23
Lumut, Perak (LUM)	4° 13′ 56.28″ N; 100° 37′ 47.28″ E	M	24	24
Sungai Besar, Selangor (SB)	3°39′50.4″N; 100059′16.6″E	M	14	14
Kuala Selangor, Selangor (KS)	3° 21′ 0.00″ N; 101° 15′ 0.00″ E	M	24	24
Kukup, Johor (KUK)	1° 18′ 60.00″ N; 103° 26′ 59.99″ E	M	12	15
Pasir Puteh, Kelantan (PP)	5° 49′ 58.28″N; 102° 24′ 19.02″E	SCS	18	18
Tok Bali, Kelantan (TB)	5°52′35.5″N; 102027′29.9″E	SCS	14	14
Pulau Tenggol,Terengganu (PT)	4° 47′ 59.99″ N; 103° 40′ 59.99″ E	SCS	21	21
Pantai Kijal, Terengganu (PK)	4°18′20.2″N; 103°28′57.2″E	SCS	6	8
Kuantan, Pahang (KUA)	3° 48′ 27.72″ N; 103° 19′ 33.60″ E	SCS	24	24
Endau, Johor (END)	2° 38′ 59.99″ N; 103° 36′ 59.99″ E	SCS	24	24
Bintawa, Sarawak (BIN)	1°33′50.96″N; 110°23′15.6″E	SCS	13	24
Pulau Bruit, Sarawak (PB)	2° 30′ 59.99″ N; 111° 25′ 59.99″ E	SCS	14	23
Mukah, Sarawak (MUK)	2° 53′ 45.5784′' N; 112° 6′ 13.4820′' E	SCS	24	24
Kota Kinabalu, Sabah (KK)	5° 58′ 29.64″ N; 116° 04′ 20.64″ E	SCS	24	24
Kudat, Sabah (KU)	6° 53′ 14.35″ N; 116° 49′ 25.10″ E	SS	24	24
Sandakan, Sabah (SAN)	5° 50′ 24.72″ N; 118° 07′ 4.44″ E	SS	20	20
Lahad Datu, Sabah (LD)	5° 01′ 36.48″ N; 118° 19′ 37.20″ E	CS	24	24
Tawau, Sabah (TA)	4° 14′ 41.35″ N; 117° 53′ 28.14″ E	CS	24	24
Mukalla, Yemen (M)	14.20–14.45ْ N; 48.50–49.40ْ E	AS	18	18
Total			390	414

Note: Straits of Malacca (M), South China Sea (SCS), Sulu Sea (SS), Celebes Sea (CS), and Arabian Sea (AS).

Close

Fig. 1

Sampling locations of E. affinis specimens collected from the Straits of Malacca (M), South China Sea (SCS), Sulu Sea (SS), Celebes Sea (CS), and Arabian Sea (AS).

Export to PPT

3 Results

3.3

3.3 Phylogenetic analysis and minimum spanning network (MSN)

The analysis of the neighbor-joining (NJ) tree and the minimum spanning network (MSN), using both COI and Cyt b genes revealed the formation of three distinct clades among the E. affinis specimens collected from Malaysia and its surrounding waters (Figs. 2, 3 and 4). The first clade for the COI gene consisted of individuals from the following locations: END, LUM, KUA, SB, KUK, PT, PK, TB, PP, SAN, LD, KS, and BM (Hap 1, Hap 2, Hap 19, and Hap 20). The second clade comprised samples from PK, PT, KUK, SAN, LD, M, LUM, and KUA (Hap 3, Hap 8 to Hap 14, Hap 16 to Hap 18), whereas the third clade encompassed individuals from LD, TA, KU, BIN, KK, PULAU, and MUK (Hap 4 to Hap 6, and Hap 15) (Figs. 2 and 4A). Conversely, the first clade for the Cyt b gene comprised specimens originating from KUA, MU, PULAU, BIN, KK, KU, and TA (Hap 8 to Hap 16). The second clade consisted of samples obtained from KK, LD, SAN, and KUK (Hap 3 to Hap 5), whereas the third clade encompassed individuals from PP, PK, PT, TB, M, SB, BM, LUM, and END (Hap 1, Hap 2, Hap 6, Hap 7, Hap 17 to Hap 20) (Figs. 3 and 4B). The NJ tree analysis also demonstrated that the sequence of E. affinis retrieved from GenBank exhibited a closer relationship with the E. affinis specimens in this current study, as opposed to the outgroup species (Lutjanus erythropterus). This finding provides further confirmation that the collected individuals indeed belong to the E. affinis species. The MSN analysis for both genes (Fig. 4) indicated that Hap 2 and Hap 1 exhibited the highest frequency among the haplotypes. The remaining haplotypes exhibited either limited distribution across a few populations or were exclusive to a specific population. The presence of three distinct clades (clades 1, 2, and 3) was further supported by the MSN analysis, revealing a nucleotide divergence of over five differences between them (Fig. 4).

Close

Fig. 2

Neighbor-Joining (NJ) tree showing relationships between COI haplotypes of E. affinis collected from Malaysia and its surrounding waters. Only bootstrap values >50 % are shown.

Export to PPT

Close

Fig. 3

Neighbor-Joining (NJ) tree showing relationships between Cyt b haplotypes of E. affinis collected from Malaysia and its surrounding waters. Only bootstrap values >50 % are shown.

Export to PPT

Close

Fig. 4

Minimum spanning network among 20 haplotypes (A) COI (B) Cyt b of E. affinis collected from 20 populations in Malaysia and its surrounding waters.

Export to PPT

3.6

3.6 Neutrality test and mismatch distribution

The neutrality assessment of the COI gene revealed a statistically significant negative Tajima's D statistic (−2.2553, P < 0.05) across all populations. Similarly, the Fs statistic calculated by Fu yielded a negative value, although it did not reach statistical significance (−11.7200, P > 0.05). In relation to the Cyt b gene, the neutrality test revealed that both Tajima's D and Fu's Fs statistics for all populations were negative, albeit not statistically significant (−1.3609 and −0.3670, respectively) (Table 4). Mismatch distribution analysis for both genes exhibited a unimodal pattern (Fig. 5).

Close

Fig. 5

Mismatch distributions (pairwise number of differences) for the mtDNA genes (A) COI (B) Cyt b of E. affinis showing the expected and observed pairwise differences between sequences with the respective frequencies.

Export to PPT

4 Discussion

The presence of skilled taxonomists is crucial in ensuring the accurate identification of species using morphological identification techniques or traditional methods, as there is a significant risk of misidentification resulting from phenotypic plasticity. Therefore, DNA barcoding has proven to be a very useful method for identifying species, particularly for specimens that are imperfect, damaged, or at distinct morphological stages. Nevertheless, it is important to acknowledge that DNA barcoding does possess certain limitations. One such limitation is that closely related species may exhibit comparable sequences, thereby rendering DNA barcodes less effective in their ability to differentiate between species. Hence, it is highly recommended to employ diverse genes or multiple genes in order to ensure precise and accurate outcomes in DNA barcoding (Wang et al., 2018). The current study successfully identified over 99 % of E. affinis specimens obtained using the Basic Local Alignment Search Tool (BLAST) from NCBI database utilising the COI and Cyt b genes. This study demonstrates the efficacy of both COI and Cyt b gene markers in distinguishing E. affinis specimens and assessing the genetic diversity and population structure of E. affinis collected from five important marine regions in Malaysia and its adjacent waters.

The genetic distance within and between 20 populations of E. affinis, as determined by analysing the COI gene, exhibited a range from low to high values (0.000–0.082) (Table 2). The findings of this study indicate that there exists a genetic differentiation among a total of 20 populations of E. affinis. The assessment of genetic distance is frequently employed to determine the taxonomic classification in closely related species, as well as for the purposes of evolutionary investigation. According to Hebert et al. (2003), a higher genetic distance indicates a substantial decrease in relatedness between the samples, accompanied by an extensive number of variations in their nucleotide bases. Marine species commonly exhibit characteristics such as a substantial population size, the capacity to disperse extensively during their pelagic larval stages, and a wide geographical distribution (Palumbi, 1992). For example, the limited genetic diversity observed among tuna populations across different oceans can be attributed to the presence of a consistent and extensive pelagic habitat and abundant spawning areas (Durand et al., 2005). The geographical conditions in Malaysia have a substantial impact on the proliferation and breeding of aquatic life. The warm and shallow waters found in this area provide an optimal setting for the thriving of fish, coral, seaweed, and various other marine resources (Azahar et al., 2022). The vast pelagic environment and plentiful spawning grounds in Malaysia designate it as a critical region for highly migratory species, such as tuna (Prince et al., 2023). On the other hand, the physical obstacles in the marine environment, such as ocean currents, reefs, and other land formations, are thought to be the reason why marine fish display genetic diversity and population structure. The presence of these barriers has the potential to impede the movement of genes between populations, resulting in genetic differentiation (Ward et al., 1994).

This present findings are consistent with prior studies that have examined the genetic structure of E. affinis in various marine regions, including the North Indian Ocean (Kumar et al., 2012), the Straits of Malacca (Masazurah et al., 2012), and the coastal waters of Tanzania (Johnson et al., 2016). These studies have identified discrete clades of E. affinis. Similarly, the results of the current study align with the findings presented by Kasim et al. (2020), wherein a minimal genetic differentiation was observed among 11 distinct populations of Thunnus tonggol (longtail tuna) in Malaysia. In their study, Akbar et al. (2018) provided findings on the genetic distance observed in Thunnus obesus. Their research focused on two distinct populations inhabiting the North and South Moluccas Seas in Indonesia. The reported genetic distance values ranged from 0.023 to 0.027, indicating the level of genetic differentiation within and between these populations.

Both haplotype diversity (Hd) and nucleotide diversity (π) are highly responsive indicators of a population's genetic diversity and can be employed to infer population’s historical demographic patterns. The low nucleotide diversity and high haplotype diversity, as observed through analysis of the COI gene in this study provide evidence for a large population size, recent population expansion, and the presence of unique genetic variations within the population. Nevertheless, this also implies that there has been an insufficient duration for the accumulation of supplementary nucleotide substitutions across the various haplotypes (Kasim et al., 2020). In contrast, the Cyt b gene exhibited significant levels of nucleotide and haplotype diversities. Based on the demographic history classifications proposed by Grant and Bowen (1998), the results obtained from the analysis of the Cyt b gene are consistent with category 4. This classification suggests a substantial level of diversity in haplotypes (h > 0.5) and nucleotides (π > 0.5 %). The findings of this study indicate the possibility of either a lengthy evolutionary timeline within a large and stable population, or significant divergence among haplotypes resulting from a secondary encounter between distinct lineages. As stated by Kumar et al. (2012), it is widely accepted that the substantial genetic diversity observed in marine fishes can be attributed to their large population sizes. Thus, the extensive haplotype diversity observed in this study can be attributed to the wide geographic range and substantial population sizes of E. affinis. The present findings are consistent with previous studies conducted on various species of tuna, including E. affinis (Hd = 0.67, π = 0.0124) (Santos et al., 2010), longtail tuna (Thunnus tonggol) (Hd = 0.990, π = 0.0195) (Kasim et al., 2020).

The findings of the phylogenetic tree and MSN analyses for both mitochondrial genes are depicted in Figs. 2, 3, and 4. The results of these analyses demonstrate that the 20 populations of E. affinis can be classified into three distinct clades. The existence of distinct genetic clades within a species can be attributed to the occurrence of secondary contact and subsequent interbreeding between populations that have experienced prolonged geographic isolation (Durand et al., 2005). The results of the MSN analysis conducted on both genes indicate that a significant proportion of the populations exhibited shared genetic characteristics with Hap 2 and Hap 1, respectively. This finding suggests that there has been a recent exchange of genetic material between these populations. Furthermore, Horne et al. (2008) demonstrated that over a long period of time, significant genetic exchange and movement of individuals between geographically distant populations resulted in the sharing of haplotypes. Additionally, the populations of LD, TA, BIN, MUK, END, and BM exhibited unique COI haplotypes. In contrast, the populations of KS, KUK, KU, PB, MUK, LUM, and END exhibited distinct Cyt b haplotypes. The findings of this study indicate a potential decrease in the exchange of genetic material between the studied locations, most probably caused by oceanic currents (Halim et al., 2022).

In addition, the AMOVA analysis corroborated the findings derived from the analyses of the phylogenetic tree and MSN, thus emphasising the presence of a genetic structure among the populations of E. affinis. The utilization of F_ST values is a common practise in the estimation of gene flow, whereby a greater F_ST value signifies a substantial level of genetic differentiation, while a lesser F_ST value indicates a minimal level of genetic differentiation (Jose et al., 2023). AMOVA analysis based on the COI gene indicated a substantial level of F_ST values (0.5364) with a statistically significant difference (P < 0.05, 0.0000). Furthermore, the analysis of the Cyt b gene revealed an important proportion of F_ST values (0.9521) accompanied by a statistically significant difference at a significance level of P < 0.05 (0.0000) (Table 5). Moreover, the isolation by distance (IBD) or Mantel test reveals a significant correlation between genetic differentiation and geographical distance. Again, the results of this analysis provide evidence supporting the presence of a genetic framework within the populations studied. Moreover, earlier studies have reported the existence of a genetic framework within tuna species (see e.g. Kunal et al., 2013; Li et al., 2015). The genetic structure of E. affinis populations plays a vital role in their response to environmental changes and human activities. High genetic diversity within a population enhances adaptability, furnishing a reservoir of traits that can help in coping with alterations in their environment. This diversity promotes resilience to stressors and fosters evolutionary flexibility (Conover et al., 2006).

The analysis of demographic history examines fluctuations in the size of a population over time, using nucleotide sequences as a tool to infer historical variations in mutation patterns (Tajima, 1989). The present study employed an independent demographic history test to ascertain that populations of E. affinis experienced a period of population expansion (Table 4). Concerning the COI and Cyt b genes, the results of this study indicate that the effectiveness of Fu's Fs in detecting significant changes in population surpasses that of Tajima's D, thereby addressing the discrepancy observed between the two measures (Fu, 1997). Generally, the results indicate the occurrence of a population change in the past, suggesting a population expansion, based on Tajima's D (Tajima, 1989). Tajima's D is a widely employed method for assessing whether the nucleotide sequence is in a state of mutation-drift equilibrium. According to Korneliussen et al. (2013), a negative value of Tajima's D suggests the possibility of population growth, expansion, or the occurrence of purifying selection. As stated by Fu)1997), the Fs test is highly sensitive in detecting evolutionary forces resulting from an abundance of new mutations within a population. Moreover, it is considered a potent tool for identifying instances of population expansion. Furthermore, the presence of a unimodal pattern of mismatch distribution as depicted in Fig. 5 offers additional support for the hypothesis of a sudden demographic expansion in the recent history of the taxa. This findings align with prior studies conducted on pelagic species including Thunnus albacares (Kunal et al., 2013), Thunnus tonggol (Kasim et al., 2020). In general, the results of this investigation suggest that E. affinis specimens gathered from various geographical regions experienced a phase of population growth.

5 Conclusion

This study represents the inaugural documentation encompassing multiple locations in Malaysia, as well as the initial investigation conducted in Yemeni waters, pertaining to the examination of genetic diversity and population structure of E. affinis through the utilization of COI and Cyt b genes. The COI and Cyt b have been effectively utilized to assess 20 distinct populations of E. affinis in the waters of Malaysia and Yemen. This analysis has unveiled the presence of three distinct lineages within these populations, which are distributed across five marine regions: the Straits of Malacca, South China Sea, Sulu Sea, Celebes Sea, and the Arabian Sea. Obviously, the findings of this study will establish a valuable foundation of knowledge for the effective management and conservation of neritic tuna populations in the waters of Malaysia and Yemen. It is anticipated that future research on the current study will require the use of different genetic markers and a larger number of sampling sites and specimen.

CRediT authorship contribution statement

Khaled Binashikhbubkr: Formal analysis, Investigation, Methodology, Software. F. Al-Misned: Validation, Writing – review & editing. Darlina Md. Naim: Conceptualization, Project administration, Resources, Supervision, Visualization.