INTRODUCTION
Southeast Asia, one of the world’s major marine biodiversity hotspots, has been frequently documented to harbor an extraordinary species richness and endemism (de Bruyn et al. 2014). Such a biodiversity characteristic predominantly stems from complex marine landscape reconfigurations and repeated sea-level fluctuations since the Plio-Pleistocene (Hoeksema 2007). The dropped sea-level during periods of glaciation caused the continental shelf between the Thai-Malay Peninsula (TMP), Sumatra, Java, and Borneo to emerge and form the Sundaland (Metcalfe 2017), inhibiting range-wide migration and exchange of coastal marine species. These dynamic geological processes interacted with other evolutionary forces to promote speciation and differentiation of tropical marine organisms in the region (de Bruyn et al. 2014). For example, phylogeographic evidence revealed strikingly high population genetic homogeneity in the brown macroalga Sargassum polycystum across 27 sites in Southeast Asia and the western Pacific, in sharp contrast to that observed in most animal studies (Chan et al. 2013, Hu et al. 2018).
The TMP lies at the intersection of the Indo-West Pacific transition zone. The east coast of the TMP borders the Gulf of Thailand to the north and the South China Sea to the south, while the west coast borders the Andaman Sea to the north and the Malacca Strait to the south (Fig. 1). The TMP thus forms a natural geographical barrier between the Andaman Sea and the Gulf of Thailand, which is hypothesized to influence the distributional range and genetic diversity of marine species (Pongparadon et al. 2015). Shifts in oceanic currents and environmental variables led to significant differences in species composition and genetic variation on both sides of the TMP, with the Andaman Sea having higher species richness than the Gulf of Thailand (Pongparadon et al. 2015).
Halimeda macroloba Decaisne is a highly calcified green macroalga distributed in tropical and subtropical regions of the Indo-Pacific. This species grows in sandy and argillaceous substrates, anchored by a bulbous holdfast, or on rocks and dead coral, secured by numerous unorganized siphons (Kojima et al. 2015). Halimeda species are ecologically important primary producers providing habitat for marine organisms and stabilization of sediments (van Tussenbroek and van Dijk 2007). Halimeda macroloba reproduces in two ways: sexually and asexually. Sexual reproduction involves the conversion of the entire cytoplasm into reproductive cells, a phenomenon termed holocarpic reproduction, leaving only a calcified and empty husk after gamete release (Clifton and Clifton 1999). Asexual reproduction is usually attributed to the development of new thalli from either the elongating subterranean rhizoids or small fragments detached by physical or biological disturbance (Walters and Smith 1994).
Halimeda macroloba can grow fast in a branching and segmented way and is characterized by a high CaCO3 accumulation rate (Mayakun and Prathep 2019), playing an important role as a carbon sink and a source of organic matter in the sediment (Tuntiprapas et al. 2019). It was estimated that the numerous calcareous segments grow and fall off in a leaf-like form resulting in Halimeda species to account for about 8% of the global carbonate production in tropical shallow waters (van Tussenbroek and van Dijk 2007). In addition, the calcification of H. macroloba is crucial for the formation and maintenance of coral reef ecosystems (Ma et al. 2021). However, H. macroloba needs sufficient calcium to grow healthy and store calcium carbonate (Buapet and Sinutok 2021). The ongoing and predicted ocean acidification and warming have been proposed to impose substantial stresses on its growth, survival, and reproduction (Hofmann et al. 2015). Four additional Halimeda species have been reported from the east coast of Peninsular Malaysia and at least eight species occur on the west coast of the TMP (Malacca Strait and Andaman Sea). However, H. macroloba is the single representative of its genus in the Gulf of Thailand (Arina et al. 2019). In this context, phylogeographic studies can help to understand evolutionary processes and drivers associated with population structuring of H. macroloba, providing insights into managing and conserving Halimeda diversity in Southeast Asia under global change. In this study, we used the chloroplast-encoded tufA (elongation factor Tu gene) and the rpl2–rpl16 region (an amplicon spanning four ribosomal protein genes) to investigate population genetic structuring and phylogeographic diversity of H. macroloba in Southeast Asia, particularly around the TMP.
MATERIALS AND METHODS
Halimeda macroloba samples were newly collected from nine sites along the east and west coasts of the TMP and two sites from Mindoro, Philippines (Table 1). Despite search efforts, H. macroloba was not found along the Andaman coast of Thailand, i.e., north of Phuket Island (Pongparadon et al. 2015, 2017, this study). All samples were cleaned, dried, and preserved in silica gel. Total genomic DNA was extracted following the method described by Li et al. (2017). In green algae, the tufA and the rpl2–rpl16 markers are thought to inherit maternally without recombination (Miyamura 2010). These markers have been used in phylogeographic studies of Halimeda species (Rindi et al. 2020, Nguyen et al. 2022). In the present study, tufA and rpl2-rpl16 were thus used to investigate population genetic diversity and structure of H. macroloba.
Polymerase chain reaction (PCR) amplification was performed in 50 μL of reaction volume (Rindi et al. 2020), containing 25 μL of 2× Taq plus Master Mix II (Vazyme Biotech Co., Ltd., Nanjing, China), 2 μL of forward primer (10 μmoL L−1), 2 μL of reverse primer (10 μmoL L−1), 1 μL of template DNA, and 20 μL of RNase-free water. Primer pairs tufA-F (5′-TGAAAGAAMAWCGTC-3′)/tufA-R (5′-CCTTCNCGAATMGCRAAWCGC-3′) and rpl2F (5′-CWAAAAAYCCAGTRGACCATCC-3′)/rpl14R (5′-CAG CAACATTWACAYAACTTTCAG-3′) were used to amplify the tufA gene and the rpl2–rpl16 region, respectively (Rindi et al. 2020). PCR reactions used a profile of an initial denaturation step at 95°C for 3 min, followed by 35 cycles of denaturation (94°C for 15 s), primer annealing (tufA: 53°C for 20 s; rpl2–rpl16: 49°C for 20 s), and elongation (72°C for 1 min), and a final elongation step at 72°C for 4 min.
PCR products were cleaned and sequenced commercially (Sangon Biotech, Shanghai, China) in both directions using amplification primers. The amplified tufA and rpl2–rpl16 sequences were manually trimmed in BioEdit v.7.1.9.0 (Hall 1999) and aligned in MEGA v.10.1.8 (Kumar et al. 2016). The tufA alignment was supplemented with 15 tufA sequences from Vietnam (Nguyen et al. 2022). Genetic diversity indices including the number of haplotypes (Nh), number of polymorphic sites (S), haplotype diversity (h), and nucleotide diversity (π), were estimated using DnaSP v.5.10.01 (Librado and Rozas 2009). Considering that only three tufA sequences were obtained from each of five sites in southern Vietnam (Table 1), we pooled the 15 sequences together and treated it as a single population for genetic analysis. Pairwise population genetic differentiation (FST) was estimated using Arlequin v.3.5 (Excoffier and Lischer 2010). This program was also used for hierarchical analysis of molecular variance (AMOVA) to detect the proportion of genetic differentiation among regions and within and among populations. To evaluate the evolutionary relationships among tufA and rpl2–rpl16 haplotypes, parsimony median-joining networks were constructed using Network v.10.2.0.0 (Bandelt et al. 1999).
RESULTS
A total of 260 tufA sequences were newly generated with a length of 843 base pairs (bp). The 275 sequence alignment contained 3 polymorphic sites and yielded 4 haplotypes (T1–T4), with 1–3 base mutations between each haplotype (Fig. 1A). T1 was the most widely distributed haplotype on both sides of the TMP, accounting for 71% of all individuals in 12 populations. T2 was only found in the Philippines and Vietnam. T3 was only found in the Malacca Strait (at two sites), whereas T4 was restricted to a single site in Vietnam (Fig. 1A). Each site was represented by a single haplotype, except PB (Pulau Besar, Malacca) in the Malacca Strait which had extremely low genetic diversity (h = 0.067, π = 0.008).
A total of 225 rpl2–rpl16 sequences were obtained with a length of 1,659 bp. The sequences contained 5 polymorphic sites, yielding 4 haplotypes (R1–R4), with 1–5 base mutations between them (Fig. 1B). R1 was only found at two sites in the Malacca Strait, while R2 is endemic to Paniquian Island in the Philippines. R3 is the most widely distributed haplotype on the east and west coasts of the TMP, accounting for 65% of all 225 individuals. R4 was only found at a single site (Suratthani [KE]) in the Gulf of Thailand (Fig. 1B). This was the only site that had more than one rpl2–rpl16 haplotype, yet genetic diversity was extremely low (h = 0.118 ± 0.101, π = 0.010 ± 0.019).
The concatenated tufA and rpl2–rpl16 sequences defined 5 haplotypes (Supplementary Fig. S1). A single haplotype (T2R2) was found in the Philippines which was not found anywhere else. However, rpl2–rpl16 was not determined for Vietnamese specimens, which mostly exhibited the tufA haplotype T2. All populations in the Gulf of Thailand exhibited a single tufA/rpl2–rpl16 haplotype (T1R3), except for a single T1R4 individual at KE. Three haplotypes (T1R1, T1R3, and T3R3) were found in the Malacca Strait, where only one population (PB) had more than one haplotype.
Halimeda macroloba populations in this study were divided into three groups according to geographic proximity: the Malacca Strait, the Gulf of Thailand, and the South China Sea (Philippines + Vietnam) (Table 2). tufA-based AMOVA showed 62.18% of the total genetic variation occurred among groups (p < 0.01), 34.17% among populations (p < 0.001), and only 3.65% within populations (p < 0.001) (Table 2). The rpl2–rpl16 results (without Vietnamese samples) showed 82.20% of the total genetic variation occurred among group (p < 0.01), and only 0.44% observed within populations (p < 0.001) (Table 2).
DISCUSSION
Phylogeographic pattern
Rindi et al. (2020) applied the same two markers to investigate genetic diversity of H. tuna (J. Ellis & Solander) J.V.Lamouroux in the Mediterranean Sea. They also found very little variation in the tufA gene with a dominant and widespread haplotype, whereas rpl2–rpl16 (as rpl2–rpl14 in Rindi et al. 2020) was more variable and structured than tufA. In particular, rpl2–rpl16 haplotypes in the Western Mediterranean basin featured a West-East gradient. Geographically, the Mediterranean Sea was configured through the closing of the Tethyan Seaway in the Middle Miocene (13 million years ago [Ma]), followed by the temporary closing of the Strait of Gibraltar during the Messinian Salinity Crisis (5.3–5.9 Ma) and cryptic speciation during the Pleistocene glaciations (Rindi et al. 2020). Consequently, the vicariance-caused biogeographic discontinuity between the western and eastern Mediterranean basins was maintained by a low-level gene flow since the Last Glacial Maximum (LGM, 1.8–2.0 thousand years ago). In the present study, H. macroloba in Southeast Asia also exhibited a West-East (the western TMP vs. the eastern Philippines + Vietnam) divergence of endemic haplotypes (Fig. 1, Supplementary Fig. S1). Post-LGM recolonization of coastlines from different glacial refugia may explain the disjunct diversity pattern in H. macroloba. However, since it is most likely that the Gulf of Thailand was recolonized from the east, driven by ocean currents (see Discussion below, Fig. 1A), it is surprising that the dominant haplotype in the Gulf of Thailand was not found in the South China Sea. Unfortunately, our study failed to include samples from the Andaman Sea, which is expected to be the donor region for the Malacca Strait (Wee et al. 2020).
The TMP has often been hypothesized as a geographic barrier to gene flow between the water bodies on either side (e.g., Reid et al. 2006), yet many studies failed to demonstrate it (Wichachucherd et al. 2014, Pongparadon et al. 2017). In this study, both markers revealed remarkable low phylogeographic diversity in H. macroloba along the TMP (Fig. 1, Supplementary Fig. S1), with no indication that gene flow is hindered along its southern end (group-level) (Table 2).
Low genetic diversity
There is no obvious population structuring in either the Malacca Strait or the Gulf of Thailand (Fig. 1, Supplementary Table S1). Such a pattern is in accordance with previously reported tufA haplotypes of H. macroloba in the Gulf of Thailand (Pongparadon et al. 2017). A potential explanation may be the short time frame since the inundation of the Gulf of Thailand after the LGM (Sathiamurthy and Voris 2006), in which H. macroloba was apparently the only Halimeda species to have colonized the Gulf of Thailand with a long-term founder effect. When the Sundaland formed a landmass during the LGM, the Gulf of Thailand was dry land due to the ~120 m lower sea level compared to the present-day. This resulted in intertidal and subtidal habitat to become lost and marine populations were confined to the eastern (South China Sea) and western (the Andaman Sea) coasts of Sundaland (Woodruff 2010). As the ice margin retreated ca. 11 thousand years ago, the sea level rose and the Gulf of Thailand and Malacca Strait became submerged again (Guo et al. 2020). The geological history of ca. 11 thousand years cannot account for marine speciation in the Gulf of Thailand, and the currently inhabiting marine species must have migrated from areas that remained submerged during the LGM such as the central South China Sea and the Andaman Sea (Guo et al. 2020, Bulan et al. 2022).
The present study sampled four sites (i.e., Thang Khen Bay, Phuket [TB], LiDi Island, Satun [KL], Tean Island (East), KE, and SamaeSarn village, ChonBuri [SS]) that were also sampled by Pongparadon et al. (2017). However, we sampled TB (n = 30) in July 2019 and found a different haplotype (T3) than Pongparadon et al. (2017), who only found T1 at TB (n = 5) prior to 2009 (collection season not reported). The dominant haplotype in a population may thus shift over time. In this respect, it is also worth noting that the Malacca Strait populations sampled in 2019 (i.e., TB, KL, and PB) all consisted exclusively of rpl2–rpl16 haplotype R3, whereas those sampled in 2021 (i.e., Phi Phi Don Island, Krabi [HH] and LiPe Island, Satun [SB]) consisted exclusively of R1 (Fig. 1B). Therefore, more spatial and temporal sampling and genotyping are required in the future to better understand phylogeographic diversity of H. macroloba in Southeast Asia.
The Gulf of Thailand is part of the Sunda Shelf which borders the center of marine biodiversity (Hoeksema 2007, Carpenter et al. 2011). However, all but one specimen sampled in the Gulf of Thailand exhibited one and the same tufA (T1) and rpl2–rpl16 (R3) haplotype (Fig. 1), demonstrating lower genetic variation and haplotype diversity than the Malacca Strait (Table 1, Fig. 1). Such a phylogeographic pattern is in accordance with the findings in the red algae Bostrychia tenella (J. V. Lamouroux) J. Agardh (Bulan et al. 2022), Halymenia malaysiana P.-L. Tan, P.-E. Lim, S.-M. Lin & S.-M. Phang (Nguyen et al. 2023), and Gracilaria salicornia (C. Agardh) E. Y. Dawson (Muangmai et al. 2023) and the brown alga Sargassum polycystum C. Agardh (Chan et al. 2013, Hu et al. 2018) across the TMP, suggesting that population genetic homogeneity around the TMP is probably a common phenomenon in a broad range of marine macroalgae.
Low genetic diversity within a population may be explained by the founder effect and subsequent predominant clonal propagation. Although H. macroloba can reproduce sexually, its success is impeded for multiple reasons. Firstly, in contrast to most green algae which reproduce synchronously (Clifton and Clifton 1999), gamete release is non-synchronized in H. macroloba populations in the TMP (Mayakun et al. 2012). Also, the peak seasonal reproduction occurs within only 3 months (May–July) in Thailand, Indonesia and the Philippines (Mayakun et al. 2012). The short reproductive season, non-synchronization of maturation, and population-level low reproductive rate (e.g., 2–5% month−1 at peak reproduction) (Vroom et al. 2003) limit the contribution of sexual reproduction. In the field, sexual fertile thalli were not or rarely found, neither in the Malacca Strait nor in the Gulf of Thailand (Mayakun and Prathep 2019). Instead, field observations and laboratory experiments showed that asexual reproduction (e.g., fragmentation) can span the seasons and persist from year to year for Halimeda (Vroom et al. 2003, Mayakun et al. 2012). Fragmentation is thus likely the dominant means of reproduction in H. macroloba to quickly boost plant numbers, increase, obtain space, and maintain it (Chan et al. 2013, Yñiguez et al. 2015).
The huge distributional range of H. macroloba from East Africa to the South Pacific (Nguyen et al. 2022), suggests that long-distance dispersal is not a problem for this species, which, however, is unlikely to happen by mature thalli given their negative buoyancy due to calcification. Stepping-stone dispersal is a frequent process that may well underpin rapid colonization of water bodies and long-distance dispersal between ecosystems (Coughlan et al. 2017). The juvenile uncalcified thalli are the likely vectors that facilitate stepping-stone dispersal along the TMP with sea surface currents, which are influenced by the northeast and the southwest monsoons (Masseran and Razali 2016). In particular, during the northeast and the southwest monsoon seasons, the North Indian Ocean Current flows from the southwest towards the Andaman Sea and eventually towards the Malacca Strait, while the Malacca Strait Current flows northwest towards the Andaman Sea (Fig. 1). The offshore flow and partial mixing of the two currents (Rizal et al. 2012) can potentially facilitate genetic exchange between H. macroloba populations in the Malacca Strait. In contrast, the semi-enclosed Gulf of Thailand is mainly influenced by eddy currents (Sojisuporn et al. 2010), which restricts circulation within the Gulf to exchange with open ocean, resulting in low phylogeographic diversity of the only Halimeda species in the Gulf of Thailand (Pongparadon and Prathep 2013, Pongparadon et al. 2015). The reproductive season (i.e., May–July) is likely of importance, because the currents in the South China Sea change direction with the monsoon seasons. This could explain population genetic homogeneity of H. macroloba on both sides of the TMP as well as genetic uniqueness of haplotypes between the South China Sea and the Gulf of Thailand.
Concluding perspective
As a foundation species, H. macroloba provides structural substrate and food for associated coastal communities and ecosystems. Moreover, Halimeda species can serve as biological indicator species (Wizemann et al. 2015), showing important conservation value for long-term monitoring of Halimeda-rich coastal marine ecosystem (e.g., the status of reef health). However, Halimeda is also highly susceptible to bathymetry, sea surface temperature, and phosphate concentration (Stankovic et al. 2022), particularly ocean acidification, which can negatively affect the growth rate, calcification rate, and photosynthetic efficiency. All sampled populations in the present study were in the intertidal or shallow subtidal. These will make H. macroloba populations under a high risk of loss of unique genetic variation (Gerstenmaier et al. 2016), ultimately affecting its primary productivity and altering the functioning of associated communities and ecosystem (Ellison et al. 2005). In such a circumstance, the populations from the TMP and the South China Sea should be considered as different conservation units for sustainable management. Sampling needs to be expanded across the Philippines and Indonesia to determine whether the haplotype found in Mindoro in the present study is truly unique and deserves conservation efforts.
PDF Links
PubReader
ePub Link
Full text via DOI
Download Citation 
Supplement
Print 
