Plastid genome insights into the underutilized legume Vigna minima (Fabaceae)
Article information
Abstract
Vigna minima, a wild legume belonging to the genus Vigna within Fabaceae, is broadly distributed across East Asia, Southeast Asia, and Oceania and is phylogenetically close to cultivated species such as V. angularis var. angularis (adzuki bean) and V. umbellata (rice bean). Although closely related to cultivated species, V. minima remains comparatively poorly studied. Here, we assembled and analyzed the complete plastid genome (plastome) of V. minima to establish genomic resources for the species and to clarify its phylogenetic position within Vigna. The plastome was 151,844 bp in length, exhibited a typical quadripartite structure, and had a guanine-cytosine (GC) content of 35.2%. It has 127 genes, consisting of 82 protein-coding genes, 37 tRNA genes, and eight rRNA genes. In terms of its genome size, GC content, and gene composition, the plastome was very similar to those of other species within the genus Vigna. Despite this overall structural conservation, notable sequence variations were detected in the regions accD and ycf1, which are known plastome variation hotspots in legumes. A phylogenetic analysis based on the complete plastome revealed that V. minima and V. umbellata form a strongly supported monophyletic group, which is a sister to V. angularis var. angularis. Given its close phylogenetic relationship to cultivated Vigna species and the conserved plastome structure observed across the genus, V. minima represents a potential future crop, and this study provides new plastome resource for the species.
INTRODUCTION
Fabaceae Lindl., one of the largest families of angiosperms, comprises approximately 700 genera and about 22,000 species. It is divided into six subfamilies (Caesalpinioideae, Cercidoideae, Detarioideae, Dialioideae, Duparquetioideae, and Papilionoideae) (Lewis et al., 2005; Legume Phylogeny Working Group, 2017). The family includes numerous species of high agricultural, ecological, and ethnobotanical value (Graham and Vance, 2003; Legume Phylogeny Working Group, 2017). These include protein-rich food crops, forage and pasture plants, and green manure species. Many legumes also enhance soil fertility through symbiotic nitrogen fixation with rhizobia (Sprent et al., 2017). Given these combined functions, legumes play a vital role in food security, sustainable agriculture, and ecosystem nutrient cycling (Hasanuzzaman et al., 2020).
The genus Vigna Savi., in the subfamily Papilionoideae, comprises about 100 species distributed across tropical and subtropical regions of Africa, Asia, and the Americas (Tomooka et al., 2012; Plants of the World Online, 2025). The genus includes several major crop species, such as V. angularis (Willd.) Ohwi & H. Ohashi var. angularis (adzuki bean), V. radiata (L.) R. Wilczek var. radiata (mung bean), and V. unguiculata (L.) Walp. (cowpea), which represent agriculturally and ethnobotanically important legumes within Papilionoideae (Cho, 1975; Gopinathan et al., 1987; Ehlers and Hall, 1997; Tomooka et al., 2014; Boukar et al., 2019). These crops are particularly valued for their high protein content and broad environmental adaptability (e.g., drought and heat tolerance), and have therefore been the focus of extensive agronomic and genetic improvement research (Tomooka et al., 2010; Tomooka et al., 2014).
In South Korea, seven Vigna taxa—V. angularis var. nipponensis (Ohwi) Ohwi & H. Ohashi, V. angularis var. angularis, V. minima (Roxb.) Ohwi & H. Ohashi, V. umbellata (Thunb.) Ohwi & Ohashi, V. unguiculata, V. vexillata var. tsusimensis, and V. radiata var. radiata—have been recorded (National Institute of Biological Resources, 2025). Among these, V. minima has a broad distribution across East Asia, Southeast Asia, and Oceania, and is widespread throughout South Korea, occurring in diverse native habitats (Lee et al., 2006; Tomooka et al., 2012; Plants of the World Online, 2025). Although V. minima is often confused with V. angularis var. nipponensis because of similarities in floral morphology and leaf variation, the two taxa can be distinguished by the relative length of the bracteole. In V. angularis var. nipponensis, the bracteole is nearly twice as long as the calyx, whereas in V. minima it is approximately equal to or shorter than the calyx (Lee et al., 2006). Phylogenetic analyses based on nuclear ITS and partial plastid genome sequences indicate that V. minima is closely related to other species of the genus Vigna, particularly those within the subgenus Ceratotropis ( Piper) Verdc. (Doi et al., 2002; Horton et al., 2024). Although V. minima is genetically related to other species of the subgenus Ceratotropis, including several domesticated species, research on its agricultural and ecological potential remains limited, and, importantly, its complete plastid genome has not yet been reported. Moreover, given that wild relatives of crop species are recognized as important genetic resources for crop improvement, V. minima, as a wild congener of cultivated adzuki bean, can be regarded as a potential crop wild relative (Maxted et al., 2006).
Plastid genomes (plastomes) are generally maternally inherited and retain conserved genomic structure and gene content, providing important insights into plant evolution and genetic diversity. Comparative plastome analyses enable the detection of structural and sequence-level variation and are therefore widely used to infer phylogenetic relationships among closely related species (Jansen et al., 2007; Ravi et al., 2008; Jansen and Ruhlman, 2012; Choi et al., 2022). Although plastome sequences have been reported for several legume species native to Korea (Park et al., 2024, 2025), no complete plastome has yet been generated for V. minima (Choi and Choi, 2017; Jin et al., 2019; Choi et al., 2022; Kim et al., 2024; Choi, 2025; Seol et al., 2025). Within Vigna, plastomes are particularly valuable for assessing genetic diversity, resolving taxonomic ambiguity, and identifying lineage-specific characteristics in morphologically similar or underutilized species such as V. minima.
Therefore, this study focuses on assembling, characterizing, and comparatively analyzing the complete plastome structure of V. minima to elucidate its evolutionary position within Vigna and to provide genomic insights that support its potential utilization.
MATERIALS AND METHODS
Plant material, DNA extraction and next-generation sequencing
A dried leaf sample of V. minima (NIBRGR0000169324) was provided by the National Wildlife Germplasm Bank of the National Institute of Biological Resources (NIBR), Korea. The specimen was collected on 20 Aug 2014 at Mongunsa, Jinchon-ri, Baengnyeong-myeon, Ongjin-gun, Incheon, South Korea and is deposited in the NIBR. Genomic DNA was extracted using the cetyltrimethylammonium bromide method following the protocol of the Soltis Laboratory, Florida Museum of Natural History (Doyle and Doyle, 1987; Cullings, 1992). Next-generation sequencing was performed by Novogene (Hong Kong) using the Illumina NovaSeq platform, generating 150 bp paired-end reads.
Plastid genome assembly and annotation
The raw data were assembled using GetOrganelle v1.8.0.0 with default parameters (Jin et al., 2020). The resulting plastome was annotated with PlastidHub (Zhang et al., 2025), and the annotations were manually refined in Geneious Prime v2025.0.1 by comparison with V. mungo (L.) Hepper (NC_050260) and V. unguiculata (KJ468104) as reference genomes. In addition, tRNA genes were further corrected using tRNAscan-SE (Chan et al., 2021). The manually curated plastome was visualized with Organellar Genome DRAW (OGDRAW) v1.3.1 (Greiner et al., 2019). The complete plastome of V. minima was assembled and characterized, and the sequence has been submitted to GenBank with the accession number PX514772.
Codon usage analysis
Codon usage bias analysis was conducted on the protein-coding genes (PCGs) of the V. minima plastome. Genes duplicated in the inverted repeat (IR) regions were excluded, and a total of 75 PCGs were extracted and concatenated. These PCGs included genes starting with both the standard initiation codon (ATG) and alternative start codons (ACG and GTG). Relative Synonymous Codon Usage (RSCU) values were calculated using CodonW v1.4.4 (https://sourceforge.net/projects/codonw/) with default parameters.
Comparative plastid genome analyses among Vigna species
Plastome structures of eight Vigna taxa—V. subterranea (L.) Verdc. (NC_057597), V. unguiculata (KJ468104), V. angularis var. angularis (NC_021091), V. minima (PX514772), V. umbellata (NC_062586), V. mungo (NC_050260), V. radiata var. radiata (NC_013843), V. radiata var. sublobata (Roxb.) Verdc. (AP014692)—were compared using the plastome of Phaseolus vulgaris L. (NC_009259) as a reference. For the analysis, entire plastome sequences were used with one copy of the IR removed. The intact ycf1 gene, originally spanning the removed IR region, was retained to ensure accurate representation. Plastome structure analyses and visualizations were performed with the mVISTA program (https://genome.lbl.gov/vista/opublications.shtml), applying the Shuffle-LAGAN algorithm (Mayor et al., 2000).
Nucleotide diversity (Pi) was analyzed using the complete plastome sequences of eight Vigna taxa. Sequence alignment was performed in Geneious Prime v2025.0.1 with the MAFFT v7.490 algorithm under default settings (Katoh and Standley, 2013). The aligned plastome sequences were analyzed in DnaSP v6.12.03 using a sliding window of 600 bp with a step size 200 bp (Rozas et al., 2017). The resulting Pi values were visualized with the ggplot2 package (Wickham, 2016) in RStudio v2025.05.1 (Posit Software, PBC, Boston, MA, USA).
The boundaries of the IR regions among eight Vigna plastomes were compared using IRplus (https://irscope.shinyapps.io/IRplus/) (Menéndez et al., 2023). The annotated GenBank files of each taxon were uploaded to IRplus to visualize the junctions between the IR and single-copy regions.
Phylogenetic analysis
To clarify the phylogenetic position of V. minima, a phylogenetic analysis was conducted using 16 plastomes, comprising 13 ingroup plastomes—eight Vigna taxa and five related species [Lablab purpureus (L.) Sweet (MN966634), Macroptilium erythroloma (Mart. ex Benth.) Urb. (NC_067541), Oryxis monticola (Mart. ex Benth.) A. Delgado & G. P. Lewis (NC_067542), P. acutifolius A. Gray (NC_067543), and P. vulgaris (NC_009259)]—and three outgroup plastomes: Strongylodon macrobotrys A. Gray (NC_067540), Macrotyloma uniflorum (Lam.) Verdc. (MN966638), and Sphenostylis erecta (Baker f.) Hutch. ex Baker f. (MN966645) (Online Supplemental Data Table S1).
Sequence alignment was performed in Geneious Prime v2025.0.1, and one copy of the IR region was removed to reduce redundancy in the dataset. Ambiguously aligned and low-quality regions were trimmed using Gblocks v0.91b to improve alignment reliability (Castresana, 2000).
A maximum likelihood (ML) tree was inferred in IQ-TREE v2.2.2.6 with 100,000 bootstrap replicates (Minh et al., 2020). The best-fitting nucleotide substitution model was selected according to the Akaike Information Criterion, and GTR + F + I + R4 was identified as the optimal model. The resulting phylogenetic tree was visualized using the Interactive Tree Of Life web tool (https://itol.embl.de/) (Letunic and Bork, 2021).
RESULTS
Plastid genome characteristics of Vigna minima and comparison with related Vigna species
A total of 74,893,042 raw reads were obtained from the Illumina NovaSeq platform, of which 7,842,086 reads were mapped to the plastome of V. minima, providing a coverage depth of 7,733×. The plastome exhibited the typical quadripartite structure with a total length of 151,844 bp and a guanine-cytosine (GC) content of 35.2% (Fig. 1). It consisted of a large single-copy (LSC) region of 81,522 bp, a small single-copy (SSC) region of 17,432 bp, and a pair of IR regions, each 26,445 bp in length. The plastome comprised 127 genes, consisting of 82 PCGs, 37 tRNA genes, and eight rRNA genes. Of these, 109 genes were identified as unique, whereas the remaining 18 genes were duplicated within the IR regions.
The complete plastid genome map of Vigna minima. Genes located on the outside are transcribed clockwise, while those on the inside are transcribed counterclockwise. Each gene is color-coded according to its functional category, as indicated in the legend at the lower left. In the inner circle, the dark gray area represents the guanine–cytosine content, and the light gray area represents the adenine–thymine content.
The plastome of V. minima was comparable in genome size, GC content, and gene composition to those of other Vigna species (Table 1). The total plastome lengths among the eight taxa ranged from 151,271 to 151,866 bp, and the GC content ranged from 35.0% to 35.2%. Each plastome contained 82 PCGs, 37 tRNA genes, and eight rRNA genes, and the number of genes was identical across all taxa.
Characteristics of plastid genomes from eight Vigna taxa.
Codon usage patterns and bias in Vigna minima
Analysis of codon usage in 75 PCGs of V. minima revealed the utilization of all 64 codon types, including 61 codons encoding 20 amino acids and three stop codons (UAA, UAG, and UGA). A total of 22,295 codons were identified, consisting of 22,220 amino acid-coding codons and 75 stop codons. Among the amino acids, leucine (Leu) was encoded by the largest number of codons (2,351 occurrences; UUA, UUG, CUU, CUC, CUA, and CUG), whereas cysteine (Cys) was represented by the fewest (247 occurrences; UGU and UGC) (Fig. 2).
Relative synonymous codon usage (RSCU) analysis of 75 protein-coding genes in the plastid genome of Vigna minima. Amino acids are shown along the X-axis. RSCU values are plotted as bars on the primary Y-axis, and the secondary Y-axis illustrates the number of codons used for each amino acid with a line plot.
Based on RSCU values calculated from 61 codons encoding 20 amino acids, 30 codons were overrepresented (RSCU > 1) and 29 were underrepresented (RSCU < 1), while methionine (AUG) and tryptophan (UGG) showed no bias (RSCU = 1). The codon UUA (Leu) showed the highest usage bias (RSCU = 2.14), whereas CUG (Leu) exhibited the lowest (RSCU = 0.29) (Fig. 2, Online Supplemental Data Table S2).
Comparative analysis of plastid genome sequences in Vigna taxa
Comparative analysis of the plastomes from eight Vigna taxa (with one copy of the IR removed) revealed a high level of sequence conservation across most protein-coding regions, whereas higher sequence variation was observed in non-coding regions as well as in specific PCGs and tRNA genes (Fig. 3). The PCGs exhibited 70–100% sequence identity, while the rpoC2 (43–47 kb), psbN (73–74 kb), and ycf1 (119–124 kb) genes showed relatively lower similarity. In the rpoC2, V. angularis var. angularis (VIAN), V. minima (VIMI), and V. umbellata (VIUM) displayed distinct sequence patterns compared with the other taxa, whereas V. subterranea (VISU) exhibited pronounced sequence variation in the psbN. The ycf1 displayed considerable variation across all taxa. Among the tRNA genes, V. angularis var. angularis (VIAN) and V. unguiculata (VIUN) showed notable sequence differences in trnS-GCU and trnV-GAC, respectively. Overall, most tRNA gene regions exhibited 80–100% sequence identity, and the rRNA genes (rrn16, rrn23, rrn4.5, and rrn5) located within the IR region (83–106 kb) were identified as the most conserved among the eight Vigna taxa.
Visualization of plastid genome alignment among eight Vigna taxa using Phaseolus vulgaris as a reference. Gray arrows indicate the positions and transcriptional orientations of genes. Red regions represent intergenic spacers, blue regions correspond to protein-coding genes, and sky-blue regions indicate RNA genes. The vertical scale denotes sequence identity ranging from 50% to 100%. PHVU, P. vulgaris; VISU, V. subterranea; VIUN, V. unguiculata; VIAN, V. angularis var. angularis; VIMI, V. minima; VIUM, V. umbellata; VIMU, V. mungo; VIRR, V. radiata var. radiata; VIRS, V. radiata var. sublobata.
Identification of nucleotide diversity hotspots in the plastid genome of Vigna taxa
Across the complete plastomes of eight Vigna taxa, Pi values ranged from 0 to 0.05768, with an overall average of 0.01156. The Pi values exhibited greater variability in the SSC region than in the LSC and IR regions, whereas the IR regions were the most conserved. Ten highly variable loci were identified— five loci [trnfM–psbZ (0.039606), psbD–trnT (0.03667), atpI– atpH (0.03411), trnQ–accD (0.05768), and psbE–petL (0.03405)] in the LSC and five loci [ndhF (0.03577), ndhF–trnL (0.04155), and three ycf1 regions (0.03976, 0.04702, and 0.03625)] in the SSC. Among these, trnQ–accD exhibited the highest nucleotide diversity (Fig. 4, Online Supplemental Data Table S3).
Nucleotide diversity (Pi) across the complete plastid genomes of eight Vigna taxa. Genome positions (bp) are shown on the X-axis, and Pi values are plotted on the Y-axis. The red dashed line indicates the threshold used to define highly variable regions, and red dots denote loci exceeding this threshold (hotspots).
IR boundary variation among Vigna taxa
A comparative analysis of the IR boundary regions among the eight Vigna taxa revealed that the same genes were consistently positioned around the junctions, with only minor differences in their relative positions (Fig. 5). The rps8 gene was located within the LSC region, 52–60 bp away from the JLB (LSC/IRb) junction. In the JSB (IRb/SSC) region, the ndhF gene was found in the SSC region and extended 19–25 bp into the IRb region. The ycf1 gene was located in the SSC region and extended 476–492 bp into the IRa region across the JSA (SSC/IRa) junction. The rps3 gene was positioned within the LSC region, 16–34 bp from the JLA (IRa/LSC) junction, and the rps19 gene occurred in both IRb and IRa regions across all taxa, positioned near the JLB and JLA junctions, respectively.
Comparison of IR boundary regions among eight Vigna plastid genomes.
Phylogenetic position of Vigna minima within the genus Vigna
The ML phylogenetic analysis was conducted to clarify the phylogenetic position of V. minima within the genus Vigna. The resulting tree was well-resolved and robustly supported, with bootstrap values of 100 for all nodes (Fig. 6). Within the subgenus Ceratotropis, V. minima and V. umbellata, both belonging to section Angulares, formed a monophyletic group. This group was sister to V. angularis var. angularis, another member of section Angulares. In contrast, V. radiata (var. radiata and var. sublobata) and V. mungo, which are assigned to section Ceratotropis, formed a separate clade distinct from the Angulares lineage. Additionally, the genus Vigna formed a sister relationship with a clade comprising Phaseolus, Macroptilium, and Oryxis.
Phylogenetic tree of Vigna minima inferred from maximum-likelihood analysis based on complete plastid genome sequences. Bootstrap support values are shown at each node.
DISCUSSION
The complete plastome of V. minima analyzed in this study exhibits a typical quadripartite structure and shows high similarity in genome size, GC content, and gene composition to plastomes of other taxa in the subgenus Ceratotropis, including V. angularis var. angularis, V. umbellata, V. mungo, V. radiata var. radiata, and V. radiata var. sublobata (Figs. 3, 5, Table 1). This overall genomic resemblance indicates that plastomes within the subgenus Ceratotropis have retained a largely conserved structural organization. Such structural stability is likely the result of early large-scale rearrangement events that became fixed during the initial divergence of Papilionoideae. Early plastome rearrangements in Papilionoideae include a ~50 kb inversion and a ~78 kb rearrangement in the Phaseolinae group (including Phaseolus and Vigna), after which the plastome structure of Vigna appears to have remained stable (Tangphatsornruang et al., 2010; Choi and Choi, 2017).
Although Vigna plastomes are generally conserved in overall structure, localized sequence variation was detected in several coding and non-coding regions (Fig. 3, Table 1). In particular, pronounced variability was observed in the regions adjacent to accD and within ycf1 (Fig. 4). These regions are well-known mutation hotspots in many angiosperms, where relaxed selective constraints or lineage-specific adaptive pressures may accelerate sequence divergence. The accD gene serves essential roles in plastid fatty acid biosynthesis, leaf development, and plastid-nuclear coordination, and it has been reported to exhibit lineage-specific patterns of accelerated sequence evolution and structural modification (Kode et al., 2005; Magee et al., 2010). Likewise, ycf1 is one of the most rapidly evolving plastid genes and has been widely used as a phylogenetic and adaptive marker across angiosperms (Dong et al., 2015). Collectively, these observations suggest that, while the overall plastome architecture of Vigna remains conserved, localized adaptive evolution may be occurring in functional genes associated with physiological or ecological differentiation.
Phylogenomic inference revealed that V. minima and V. umbellata form a strongly supported monophyletic group, which is sister to V. angularis var. angularis (Fig. 6). This close relationship is consistent with previous phylogenetic studies based on ITS and partial plastid sequences, but is more robustly resolved here due to the use of complete plastome data (Doi et al., 2002; Horton et al., 2024). Notably, V. umbellata (rice bean) and V. angularis var. angularis (adzuki bean) are recognized crops with documented drought and heat tolerance, suggesting that V. minima may likewise possess latent physiological traits associated with environmental stress adaptation (Yang et al., 2015; Tayade et al., 2022; Katoch et al., 2023). Such a perspective becomes particularly relevant given the increasing climate variability and water limitation projected for current and future agricultural systems. Therefore, by establishing the phylogenetic proximity of V. minima to major Vigna crops and providing its complete plastome, this study offers a foundational genomic resource that can facilitate future assessments of its physiological traits and its potential incorporation into crop improvement programs.
On-line Supplementary Data Tables S1–S3 are available at https://doi.org/10.11110/kjpt.2026.56.1.65.
Notes
ACKNOWLEDGMENTS
This research was supported by the Regional Innovation System & Education (RISE) program through the Daejeon RISE Center, funded by the Ministry of Education (MOE) and the Daejeon Metropolitan City, Republic of Korea (2025-RISE-06-013) and the 2025 Campus Innovation Project, funded by the Ministry of Education, in the Republic of Korea.
CONFLICTS OF INTEREST
The authors declare that there are no conflicts of interest.