Genetic Diversity of Siamese Rosewood Dalbergia cochinchinensis Pierre in Thailand: Insights from Chloroplast and Nuclear DNA Analysis

1. Introduction

Siamese rosewood (Dalbergia cochinchinensis Pierre) is one of approximately 250 species in the genus Dalbergia (family Fabaceae) found across pan-tropical regions [1]. While its genus has a broad distribution, D. cochinchinensis is endemic to Southeast Asia, specifically Thailand, Myanmar, Cambodia, Lao PDR, and Vietnam. In Thailand, where 26 Dalbergia species have been documented [2], Siamese rosewood populations primarily occur in the northeastern, eastern, and central regions. The species' highly prized heartwood makes it one of Thailand's most economically valuable plants, with strong demand in both domestic and international markets. However, this commercial value has led to extensive exploitation, threatening its survival and natural distribution patterns across Thailand.

The threat to Siamese rosewood has intensified significantly in recent years. In 2018 alone, Thai authorities recorded 576 cases of illegal harvesting in natural forests, predominantly in the eastern and northeastern regions, with damages valued at 103 million baht [3]. The species' critical status has prompted multiple protective measures: it was classified as restricted timber under the Royal Decree BE 2530 and the Forest Act BE 2484, and in 2013, Thailand successfully advocated for its inclusion in CITES List 2 for controlled international trade [4]. Further protection came through National Announcement No. 106/2014 [5], and the species is currently listed as critically endangered on the IUCN Red List.

Conservation of genetic resources is crucial for preserving endangered species with economic significance and maintaining genetic diversity for future breeding programs. Two main conservation approaches are employed: in situ conservation (preservation in natural forest conditions) and ex situ conservation (establishment of plantations). Genetic studies using DNA analysis have revealed that Siamese rosewood is an outcrossing species with high genetic diversity both within and between populations. Research has shown genetic differences between populations of approximately 0.127 [6,7]. Population structure studies have employed various genetic markers, including RAPD and ISSR [8,9], microsatellite [10], chloroplast maturase K (matK), and nuclear internal transcribed spacer (ITS) regions [11]. Among these markers, the matK and ITS regions have emerged as particularly valuable tools for genetic analysis, despite some limitations in discrimination power. These markers are part of the plant DNA barcode system and offer complementary insights: matK is maternally inherited and helps track seed dispersal patterns, while ITS provides information about both parental lineages and can reveal recent evolutionary events [11].

While both government agencies and farmers have begun establishing economic forest plantations for Siamese rosewood, a significant challenge remains: the shortage of high-quality seedlings from superior mother plants. To address this issue, molecular approaches using chloroplast and nuclear genome markers are essential for studying genetic diversity. The present research aims to assess the genetic diversity of D. cochinchinensis using matK and ITS regions as genetic markers. This genetic database will support future propagation efforts, breeding development, and conservation programs for sustainable utilization.

2. Materials and Methods

2.1. Sample Collection

The Siamese rosewood specimens in this study were initially identified based on their distinctive morphological characteristics (Figure 1) using the guideline described by Niyomdham [2]. Voucher specimens from this study were deposited in the Forest Herbarium-BKF under the following codes: "Prompen, B. No. 1 BKF NO. 224371" and "Prompen, B. No. 2 BKF NO. 224372". To preserve and propagate the genetic material, grafting branches from the original trees were collected from natural forests [12] and established at the Royal Forest Department's collection within the Northeastern Forest Tree Seed Center, Khon Kaen Province, Thailand.

For molecular analysis, this study examined 72 D. cochinchinensis samples collected from 17 distinct geographical locations (Table 1 and Figure 2). Young leaves were harvested using sterile scissors, preserved in silica gel, and transported to the Walai Rukhavej Botanical Research Institute laboratory for further analysis.

2.2. Molecular Analysis

Genomic DNA was extracted from young leaves using the DNeasy® Plant Mini Kit (Qiagen, Germany) following the manufacturer's protocol, with samples stored at –20 °C until analysis. Two genomic regions were targeted for amplification: the chloroplast maturase K (matK) and the nuclear internal transcribed spacer (ITS). For matK amplification, the forward primer matK-3F_IH (5'-CGT ATA GTA CTC TTG TGT TTA CGA G-3') and reverse primer matK-3R (5'-ATC CTA TTC ATC TGG AAA TCT TGG TTC-3') were used [13], while ITS amplification employed forward primer ITS-F (5'-GGA AGG AGA AGT CGT AAC AAG G-3') and reverse primer ITS5-R (5'-TCC TCC GCT TAT TGA TAT GC-3') [14]. PCR amplification was performed using the HotStarTaq Master Mix Kit (Qiagen, Germany) in 25 μl reaction volumes. Both genomic regions were amplified using a thermal cycler (Applied Biosystems, USA) under identical conditions consisting of initial denaturation at 95 °C for 5 min, followed by 35 cycles of denaturation at 95 °C for 30 sec, annealing at 55 °C for 1 min, and extension at 72 °C for 1 min, with a final extension at 72 °C for 5 min. The amplified products were visualized on 1% agarose gels in 0.5x TBE buffer using GelRed™ Nucleic Acid Gel Stain (Biotium, Inc., Hayward, CA), after which PCR products were excised, purified using the E.Z.N.A.® Gel Extraction Kit (Omega Biotek, USA), and sequenced using Sanger sequencing (ATGC Co., Ltd., Thailand).

2.3. Data Analysis

Sequence alignment and analysis were performed using the ClustalW program [15], with manual editing conducted in BioEdit [16]. Molecular diversity indices and haplotype data were generated using DnaSp v5 [17]. Genetic differentiation between populations was assessed using both p-distance and Kimura 2-parameters (K2P) [18] in MEGA XI [19]. To visualize haplotype relationships, a minimum-spanning network was constructed using Network version 10.2, employing the median-joining algorithm [20].

Phylogenetic analyses were conducted separately for matK and ITS sequences using haplotypes identified from all D. cochinchinensis populations in this study. Additional Dalbergia species sequences were retrieved from GenBank for comparative analysis. Tectona grandis and Pterocarpus indicus were selected as outgroups for matK and ITS trees, respectively. Phylogenetic trees were constructed using two methods in MEGA XI [21]: Maximum likelihood (ML) with the general time reversible model incorporating gamma distribution and invariant sites (GTR+G+I) [22], and neighbor-joining [23]. Node support was assessed using 1,000 bootstrap replicates.

3. Results

3.1. Genetic Diversity of D. cochinchinensis

We analyzed genetic variation in D. cochinchinensis populations using two genetic markers: matK and ITS sequences. For the matK gene, we successfully amplified and sequenced an 821 bp fragment from 48 samples collected across 12 localities. These sequences were deposited in GenBank (accession numbers PQ317478 – PQ317748). Analysis revealed 66 variable nucleotide sites (8% of total sites), comprising 5 singletons and 61 parsimony informative sites. We identified 34 distinct haplotypes (DcM1 – DcM34), each exclusively found in a specific geographical area (Figure 3). The matK sequences showed high genetic diversity, with haplotype diversity of 0.986±0.007 and nucleotide diversity of 0.0161±0.0022 (Table 2).

For the ITS region, we obtained sequences from 65 samples representing 17 localities (GenBank accession numbers PQ317478 – PQ317748). The analysis identified 19 variable sites, including 1 singleton and 18 parsimony informative sites. These variations defined 31 haplotypes (DcI1 – DcI34), with 18 haplotypes being location-specific and 13 shared across multiple localities (Figure 3). The ITS sequences also showed substantial genetic diversity, with haplotype diversity of 0.968±0.008 and nucleotide diversity of 0.0069±0.0003 (Table 2).

Genetic differentiation among populations was assessed using both p-distance and K2P methods. For matK, p-distance values ranged from 0.0061 to 0.0572, while K2P distances ranged from 0.0061 to 0.0597 (Table 3). The ITS sequences showed lower genetic differentiation, with p-distance ranging from 0.0010 to 0.0127 and K2P from 0.0010 to 0.0129 (Table 4). Notably, populations from SRI showed remarkably high genetic differentiation based on matK analysis, with p-distance (0.0398-0.0572) and K2P values (0.0411-0.0597) consistently higher than other populations.

3.2. Haplotype Network

Haplotype networks were constructed separately for 34 matK (DcM1 – DcM34) and 31 ITS (DcI1 – DcI34) sequences. The matK analysis revealed distinct geographical patterns with no shared haplotypes between localities. Three haplotypes showed notable genetic distances from the others: DcM10 from Muak Lek District, Saraburi Province (25 mutational steps), DcM32 from Wiang Kao District, Khon Kaen Province (17 mutational steps), and DcM7 from Kantaralak District, Sisaket Province (11 mutational steps) (Figure 3).

In contrast, the ITS sequence analysis showed minimal genetic differentiation, with a maximum of two mutational steps between haplotypes. Unlike matK, the ITS analysis identified 14 shared haplotypes across different localities: DcI1, DcI5, DcI11, DcI12, DcI13, DcI14, DcI15, DcI16, DcI17, DcI18, DcI19, DcI21, DcI22, and DcI24 (Figure 3).

3.3. Phylogenetic Tree

Phylogenetic analyses were conducted separately for matK and ITS sequences, incorporating both our sequences and previously published sequences of D. cochinchinensis and other Dalbergia species from GenBank. The ITS-based phylogenetic tree demonstrated that all D. cochinchinensis samples formed a well-supported monophyletic group, distinct from other Dalbergia species. Within the genus, D. cochinchinensis showed the closest phylogenetic affinity to D. sericea from Bhutan (Figure 4).

The matK-based phylogenetic analysis revealed two distinct clades within D. cochinchinensis (Figure 5). Clade A comprised all matK haplotypes generated in this study except DcM10, while Clade B contained haplotype DcM10 from Saraburi Province together with previously published D. cochinchinensis sequences from Thailand, Laos, and Cambodia. Notably, both clades showed close phylogenetic relationships with D. ovata sequences from Thailand and Myanmar, suggesting potential evolutionary connections between these species within the region.

4. Discussion

Our analysis revealed significant genetic diversity patterns in D. cochinchinensis populations across Thailand. The matK sequences showed high overall genetic diversity (haplotype diversity = 0.986±0.007 and nucleotide diversity = 0.0161±0.0022), with 34 distinct haplotypes exhibiting strong geographical structuring. Population-level diversity varied considerably, with the highest haplotype diversity (1.000) observed in populations from Khon Kaen (DC-KKN-WK), Nakhon Ratchasima (DC-NMA-KB), Sa Kaeo (DC-SKW-WY), and Sakon Nakhon provinces (DC-SNK-M), while the lowest diversity (0.000) was found in Saraburi (DC-SRI). Similarly, ITS sequences revealed 31 haplotypes with substantial overall diversity (haplotype diversity = 0.968±0.008 and nucleotide diversity = 0.0069±0.0003), though with less pronounced geographical structuring.

The relatively low nucleotide diversity observed in both markers aligns with previous findings. Yooyuen [7] reported low genetic diversity (haplotype diversity = 0.608 and nucleotide diversity = 0.00033) in northeastern and central Thai populations, identifying only 11 haplotypes across populations. This consistently low genetic diversity may reflect the species' historically narrow distribution and specific ecological requirements, as D. cochinchinensis occupies a more restricted ecological niche compared to related species like D. oliveri [2,24,25].

Our phylogenetic analyses revealed distinct patterns between markers. The ITS-based phylogeny showed D. cochinchinensis as a well-supported monophyletic group, while matK analysis identified two distinct clades with differential geographic distribution. These patterns suggest that landscape features have played a crucial role in shaping the species' genetic structure, consistent with recent findings by Hartvig et al. [10] and Huang et al. [26], who demonstrated that drainage systems significantly influence phylogeographic patterns of Dalbergia species in Indochina.

The genetic patterns observed in this study have important implications for conservation and breeding strategies. The selected specimens exhibited desirable traits for breeding programs, including high growth rates, straight trunks, quality wood characteristics, and disease resistance. The distinct genetic lineages identified through both markers suggest the importance of maintaining population-level diversity in conservation efforts. Moreover, the strong geographic structuring observed particularly in matK sequences indicates that conservation strategies should consider preserving populations across different regions to maintain the full range of genetic diversity.

Several limitations should be considered when interpreting our results. Our study focused primarily on Thai populations, potentially missing important genetic variations in neighboring countries. The use of only chloroplast and nuclear markers may not fully capture the genetic complexity of the species, and the lack of temporal genetic data limits our understanding of how diversity patterns have changed over time in response to anthropogenic pressures.

Future research directions should focus on expanding the geographical scope to include populations from neighboring countries, which would provide a more comprehensive understanding of regional genetic patterns. The incorporation of additional genetic markers, particularly those related to adaptive traits, would offer deeper insights into the species' evolutionary potential. Studies investigating the impact of recent habitat fragmentation on genetic diversity, the relationship between genetic diversity and phenotypic variation in important timber quality traits, and the effects of current conservation practices on genetic diversity maintenance would be valuable. Furthermore, the examination of potential hybridization patterns with closely related Dalbergia species and the development of more comprehensive landscape genetic models incorporating contemporary environmental changes would enhance our understanding of this economically important species and inform more effective conservation strategies.

5. Conclusions

This study reveals significant insights into the genetic diversity of D. cochinchinensis through analysis of matK and ITS markers. Our findings demonstrate high genetic diversity within the species, evidenced by numerous haplotypes (34 matK and 31 ITS) and high haplotype diversity indices (matK: 0.986±0.007; ITS: 0.968±0.008). The markers showed contrasting patterns: matK sequences exhibited strong geographical structuring with location-specific haplotypes, while ITS sequences showed more limited differentiation with shared haplotypes across localities. Notably, phylogenetic analyses identified two distinct evolutionary lineages, with the Saraburi Province population showing remarkable genetic divergence. This population, along with samples from neighboring countries, forms a separate clade that may represent a unique evolutionary line requiring special conservation attention.

These findings suggest that despite recent population declines, D. cochinchinensis maintains substantial genetic diversity. For conservation, this implies the need to protect multiple populations across its range, with a particular focus on preserving distinct evolutionary lineages and maintaining connectivity between populations to ensure the species' long-term survival.