717

ARTICLE Genetic diversity of the rice bean (Vigna umbellata) genepool as assessed by SSR markers Genome Downloaded from www.nrcresearchpress.com by UNIVERSITY OF ADELAIDE on 11/17/14 For personal use only.

J. Tian, T. Isemura, A. Kaga, D.A. Vaughan, and N. Tomooka

Abstract: The genetic diversity of 472 rice bean accessions (388 cultivated and 84 wild) from 16 Asian countries was evaluated by 13 simple sequence repeat (SSR) markers. In total, 168 alleles were detected, and the numbers of alleles in cultivated and wild accessions were 129 and 132, respectively. The gene diversity in cultivated populations (0.565) was about 83% of that for wild (0.678) populations. Cultivated populations from Vietnam, Myanmar, Nepal, and India had the highest gene diversity (>0.5). East Asian accessions formed a distinct genepool. Indonesian cultivated accessions showed high genetic divergence from other cultivated populations and had the most similar genetic structure to wild accessions. In Nepalese cultivated accessions, many accessions from western regions were quite distinct from others and formed a specific group. These Nepalese accessions could be considered a unique gene source for rice bean breeding. In contrast, eastern Nepalese accessions showed an SSR profile similar to that of Southeast Asian rice beans. The present study represents the first comprehensive SSR analysis in cultivated and wild rice bean germplasm and clarifies geographical distribution of genetic profile that might be used to broaden the genetic base of currently grown rice bean cultivars. Key words: rice bean, Vigna umbellata, SSR, genetic diversity, domestication. Résumé : La diversité génétique au sein d’une collection de 472 accessions du haricot riz (388 accessions cultivées et 84 sauvages) provenant de 16 pays asiatiques a été évaluée a` l’aide de 13 marqueurs microsatellites (SSR). Au total, 168 allèles ont été détectés; 129 et 132 de ceux-ci ont été observés respectivement chez les accessions cultivées et sauvages. La diversité génique chez les populations cultivées (0,565) s’établissait a` 83 % de la valeur observée chez les populations sauvages (0,678). Les populations cultivées du Vietnam, du Myanmar, du Népal et de l’Inde présentaient la plus grande diversité génique (>0,5). Les accessions de l’Asie de l’Est formaient un group distinct. Les accessions indonésiennes cultivées affichaient la plus grande divergence par rapport aux autres populations cultivées et la structure génétique la plus semblable a` celle des accessions sauvages. Au sein des accessions cultivées du Népal, plusieurs accessions provenant des régions occidentales du pays étaient très distinctes des autres et formaient un groupe a` part. Il serait possible de considérer que les accessions népalaises constituent une source unique de gènes pour l’amélioration génétique du haricot riz. Par contre, les accessions du Népal oriental présentaient des profils SSR semblables a` plusieurs accessions du Sud-Est asiatique. Cette étude constitue la première analyse SSR détaillée chez les ressources génétiques cultivées et sauvages du haricot riz, et il apporte un éclairage nouveau sur la distribution géographique des profils génétiques qui pourrait s’avérer utile pour élargir l’assise génétique des cultivars de haricot riz qui sont cultivés présentement. [Traduit par la Rédaction] Mots-clés : haricot riz, Vigna umbellata, SSR, diversité génétique, domestication.

Introduction Rice bean (Vigna umbellata (Thunb.) Ohwi & Ohashi) is cultivated mainly in a belt from Nepal, Bhutan, and northeast India through Myanmar, southern China, northern Thailand, Laos, and Vietnam to Indonesia and East Timor (Tomooka et al. 2002, 2003, 2005, 2006b, 2007; Tomooka 2009). It is widely grown under shifting cultivation systems and is particularly important for ethnic groups in these areas. In recent years, rice bean has been exported from Thailand to Japan where it is sometimes substituted for azuki bean (Vigna angularis (Willd.) Ohwi & Ohashi) in the making of confectionary (Tomooka et al. 2006a). In parts of Southeast Asia, rice bean exists as a complex with its wild form from which it is not taxonomically distinguished and

with which it can form fertile hybrids (Seehalak et al. 2006). On the basis of its prominent yellow flowers raised above the leaf canopy, rice bean is well suited to outcrossing. In the tropics, rice bean is thought to be perennial based on its very thick stem and roots (Tomooka et al. 1997). This minor pulse has been little studied, but its prolific growth and abundant pods suggest that it has a high yield potential as a vegetable (green pods), grain, and forage crop (Smartt 1991). Most research conducted on rice bean has focused on its high level of resistance to the major storage pest bruchid beetles (Callosobruchus spp.) (Tomooka et al. 2000; Kashiwaba et al. 2003; Somta et al. 2006). Arora et al. (1980) collected and evaluated 300 rice bean accessions from eastern and northeastern India and reported that this crop is free from diseases such as yellow mosaic virus,

Received 7 July 2013. Accepted 4 November 2013. Corresponding Editor: C.L. McIntyre. J. Tian.* Institute of Cereal and Oil Crops, Hebei Academy of Agricultural and Forestry Sciences, No. 162, Heng Shan Street, Shi Jia Zhuang City, Hebei Province 050031, China. T. Isemura,* A. Kaga,* D.A. Vaughan, and N. Tomooka. National Institute of Agrobiological Sciences, Kannondai 2-1-2, Tsukuba, Ibaraki 305-8602, Japan. Corresponding author: N. Tomooka (e-mail: [email protected]). *These authors contributed equally to this study. Genome 56: 717–727 (2013) dx.doi.org/10.1139/gen-2013-0118

Published at www.nrcresearchpress.com/gen on 6 November 2013.

718

Genome Vol. 56, 2013

Genome Downloaded from www.nrcresearchpress.com by UNIVERSITY OF ADELAIDE on 11/17/14 For personal use only.

Table 1. Origin and number of wild and cultivated accessions from different countries used in this study, gene diversity, observed heterozygosity, and estimated outcrossing rate. Groups

No. of accessions

No. of alleles

Gene diversity

Observed heterozygosity

Outcrossing rate (%)

Cultivated East Asia Japan (CJ) North Korea (CKn) South Korea (CKs) China (CC) Taiwan (CCt) Southeast Asia Philippines (CP) Indonesia (CIo) East Timor (CE) Vietnam (CV) Laos (CL) Thailand (CT) Myanmar (CMy) Malaysia (CMa) South Asia Nepal (CN) India (CIi) Sri Lanka (CS) Wild Thailand A (WTA) Thailand B (WTB) Thailand C (WTC) Myanmar A (WMA) East Timor A (WEA) East Timor B (WEB) Total/mean

388 62 18 6 3 32 3 151 2 6 18 17 28 17 61 2 175 145 23 7 84 37 28 8 3 4 4 472

132 66 34 18 17 56 19 106 16 31 48 50 54 51 70 17 90 82 59 28 129 107 84 48 25 28 30 168

0.565 0.360 0.301 0.195 0.167 0.369 0.224 0.554 0.154 0.356 0.460 0.561 0.484 0.487 0.531 0.154 0.526 0.510 0.536 0.450 0.678 0.668 0.621 0.618 0.539 0.538 0.551 0.608

0.134 0.061 0.039 0.026 0.077 0.072 0.128 0.107 0.077 0.269 0.218 0.086 0.093 0.077 0.079 0.115 0.183 0.188 0.154 0.165 0.232 0.264 0.319 0.087 0.026 0.000 0.000 0.151

13.4 9.2 6.8 7.1 30.0 10.8 40.0 10.7 33.3 60.6 31.1 8.3 10.7 8.6 8.1 60.0 21.0 22.6 16.8 22.4 20.6 24.6 34.5 7.5 2.5 0.0 0.0 —

Cercospora, and bacterial leaf spot, from which all of the other species of this genus suffer greatly. Pandiyan et al. (2008) also reported the highest level of mungbean yellow mosaic virus resistance in rice bean among species of Vigna. Recently, with the aim of using these genetic resources efficiently, genetic linkage maps were constructed using the azuki bean SSR markers (Somta et al. 2006; Isemura et al. 2010). QTLs for bruchid resistance (Somta et al. 2006) and QTLs for seed weight, pod dehiscence, seed dormancy, and other morphological traits (Isemura et al. 2010) were identified using these maps. In addition, genetic diversity has been studied using a limited number of cultivated and wild rice bean accessions from Thailand, India, and Nepal using molecular markers (Seehalak et al. 2006; Muthusamy et al. 2008; Bajracharya et al. 2010). However, studies of genetic diversity using many cultivated and wild accessions from many Asian countries have not been conducted in rice bean, and the level and geographic cline of genetic diversity of rice bean is unknown. Since rice bean appears to have great unexploited potential, this study was undertaken on a comprehensive set of widely distributed cultivated and wild rice bean germplasm with the aim of developing a core collection that may be thoroughly evaluated for many traits. The objectives of this study were to measure the genetic diversity of cultivated and wild rice bean using SSR primers and to determine their relationships.

Materials and methods

Institute of Agrobiological Sciences (NIAS) genebank, Tsukuba, Japan. Cultivated germplasm came from South Asia (India, Sri Lanka, and Nepal), Southeast Asia (Myanmar, Thailand, Laos, Indonesia, Malaysia, the Philippines, Vietnam, and East Timor), and East Asia (Japan, China, Taiwan, and North and South Korea). Wild germplasm came mainly from Thailand, with some accessions from Myanmar and East Timor. In the case of wild accessions, these were grouped on the basis of their countries of origins and also whether they had typical characteristics of wild species or atypical characteristics based on seed size and seed color. Thai wild accessions were divided into the following three groups: (A) typical small-seeded, black, wild type accessions, (B) typical smallseeded accessions with atypical seed coat color, and (C) largerseeded accessions with atypical seed coat color (Fig. S2). Wild accessions from East Timor consisted of groups A and B. All the plants (single plant per accession) were sown in pots in a greenhouse on 26 August 2005 at Tsukuba (36°01.5=N, 140°06.0=E, alt. 25 m), Japan. Day length at the sowing date was about 14.5 h and gradually decreased to about 11 h (shortest) in late December (about 120 days after sowing). Temperature in the green house was maintained above 20 °C using a heater during cool season. The number of pod-bearing lateral branches and days to first flower opening were recorded. After the seeds were harvested, 100-seed weight and seed color (Fig. S2; color type 1–6: 1, red; 2, pale brown; 3, black; 4, green; 5, red with black mottle; and 6, pale brown with black mottle) were investigated using harvested seeds.

Plant materials Three hundred and eighty eight accessions of cultivated rice bean and 84 accessions of wild rice bean were used in the experiment (Table 1, Table S1, and Fig. S1)1. They are conserved in the National

DNA extraction Total DNA was isolated from 200 to 300 mg of young leaves of a single plant per accession. DNA extraction was performed using QIAGEN DNA extraction kit (the Netherlands) following

1

Supplementary data are available with the article through the journal Web site at http://nrcresearchpress.com/doi/suppl/10.1139/gen-2013-0118. Published by NRC Research Press

Tian et al.

719

Table 2. Information on the SSR primers used in this study.

Genome Downloaded from www.nrcresearchpress.com by UNIVERSITY OF ADELAIDE on 11/17/14 For personal use only.

Primer sequence Primer

Forward

Reverse

Product size (bp)*

Motif

Annealing temperature (oC)

Linkage group†

CEDC016 CEDG003 CEDG008 CEDG011 CEDG015 CEDG021 CEDG024 CEDG026 CEDG029 CEDG041 CEDG043 CEDG044 CEDG073

ACTCTTGTCAATTGTCCAGG CCACTTTCTCTTGACTTTGC AGGCGAGGTTTCGTTTCAAG CCCAACCAAAGCGTTTTG CCCGATGAACGCTAATGCTG GCAGAATTTTAGCCACCGAG CATCTTCCTCACCTGCATTC TCAGCAATCACTCATGTGGG GATTGCTTTTAGCAGAGGGC GCTGCATCTCTATTCTCTGG AGGATTGTGGTTGGTGCATG TCAGCAACCTTGCATTGCAG CCCCGAAATTCCCCTACAC

TAACTTGTCACTGGAAAGGC GACCAAAGTGAAGCCAAGAG GCCCATATTTTTACGCCCAC CTTCTAGACTCTGAGCACTG CGCCAAAGGAAACGCAGAAC AAAGGATGCGAGAGTGTAGC TTTGGTGAAGATGACAGCCC TGGGACAAACCTCATGGTTG GAAGAAACCCATCTCGATCC GCCAACTAGCCTAATCAG ACTATTTCCAACCTGCTGGG TTTCCCGTCACTCTTCTAGG AACACCCGCCTCTTTCTCC

135 235 123 154 213 171 146 166 189 117 158 152 177

(AT)5(AC)9 (AG)9…(AG)13 (AG)26 (AG)16AA(AG)6 (AG)27 (AG)26 (AG)18 (AG)26 (AG)8 (AG)21 (AG)14 (GT)10AT(AG)18 (AG)24

55 60 60 60 60 60 60 60 60 60 60 60 60

8 1 5 4 6 10 9 2 2 7 3 11 8

*The product size is estimated in azuki bean (Vigna angularis ‘Erimoshouzu’). †The linkage groups were identified in rice bean maps (Somta et al. 2006 and Isemura et al. 2010).

the manufacture’s instructions. DNA concentration was adjusted to 25 ng/␮L as determined by spectrophotometer (DU-7500, Beckman, USA). PCR amplification Thirteen primers developed for azuki bean that are distributed across its 11 chromosomes were used to assess the genetic diversity of rice bean, and the information on these primers is shown (Table 2; Wang et al. 2004). In the rice bean genome, the marker loci (13 loci) from these primers were previously identified in the two genetic linkage maps of rice bean (Somta et al. 2006; Isemura et al. 2010). Primer combinations for multiplex analysis were screened to determine their compatibility. The forward primer of each set was labeled with 5-FAM, VIC, NED, or PET fluorescent dye (Applied Biosystems, USA). Four sets of multiplex PCR reactions per sample were set up for the amplification process. DNA amplification was carried out on the GeneAmp PCR system 9700 (Applied Biosystems, USA). PCR reaction was performed in a 10 ␮L volume containing 25 ng of genomic DNA, 1× KOD-plus PCR buffer, 1.5 mmol/L MgSO4, 0.2 mmol/L dNTPs, 1 U KOD-plus (Thermococcus kodakaraensis, KOD strain) DNA polymerase (TOYOBO, Japan), and 5 pmol of forward and reverse primers. The temperature cycles were programmed as follows: 2 min at 94 °C, followed by 35 cycles of 15 s at 94 °C, 15 s at 55 °C or 60 °C, and 15 s at 68 °C. Genotyping One microliter of PCR product or 1/40 diluted PCR product was mixed with 9 ␮L of Hi-Di formamide containing 0.25 ␮L GeneScan 500 LIZ size standard (Applied Biosystems, USA). The mixtures were denatured at 95 °C for 5 min, and then placed immediately on ice. The heat-denatured products were run on an automated capillary DNA sequencer (ABI Prism 3100, Genetic Analyzer). SSR fragment size was determined using GENEMAPPER ver. 3.0 (Applied Biosystems, USA) with minor modifications. Azuki bean (Vigna angularis ‘Erimoshouzu’) was used as a control in each run to create bin set and to ensure size accuracy. Data analysis Numbers of alleles per locus, observed heterozygosity per population, and allele frequency per locus were calculated using Microsatellite tools for Excel software. To measure the informativeness of each SSR marker, the polymorphism information content (PIC) value was determined for each locus according to the formula of Anderson et al. (1993): PIC = 1 – ⌺P2ij, where Pij is the frequency of the jth alleles for ith loci. Gene diversity, allelic richness, and fixation index were calculated using FSTAT ver. 2.9.3.2

software (Goudet 2002). Outcrossing rate (t) was calculated from the fixation index (FIS) using the equation t = (1 – FIS)/(1 + FIS) (Weir 1996). Nei’s Da genetic distance (Nei et al. 1983) was computed using PowerMarker ver. 3.25 (Liu and Muse 2005). Principle coordinate analysis (PCoA) was performed using NTSYSpc 2.1 (Rohlf 2001) based on Da genetic distance. The relationship of individuals based on PCoA analysis, geographic origin, and status of the germplasm (wild or cultivated) enabled 22 groups to be determined (Table 1). An unrooted dendrogram showing relationships between 472 accessions based on Da genetic distance was constructed by neighbor-joining method using PowerMarker ver. 3.25 (Liu and Muse 2005). The tree was visualized with MEGA ver. 5 (Tamura et al. 2011). To visualize the population structure and to assess the genetic relationships among 472 accessions, a model-based Bayesian clustering method, based on multilocus genotype data, was applied using software STRUCTURE ver. 2.3.4 (Pritchard et al. 2000, 2007; Rosenberg et al. 2002). The program was run 10 times for each K value using the admixture model with a burn-in period of 10 000 steps and then 10 000 times run of a Markov chain Monte Carlo from K equals 2–15. The average likelihood value, L(K), across all runs was calculated for each K. The model choice criterion to detect the most probable value of K was ⌬K, which is an ad hoc quantity related to the second-order change in the log probability of data with respect to the number of clusters inferred by STRUCTURE (Evanno et al. 2005). Core collection selection For core collection selection, we used the simulated annealing algorithm of Power Marker ver. 3.25, based on SSR allele data (Liu and Muse 2005). The total allele number was used as the main selection criteria to maximize genetic diversity in the core collection combined with information on four morphological and growth characteristics investigated in this study.

Results Morphological diversity and their geographical distribution Seed size (100-seed weight), number of lateral branches, days to flowering, and seed color were recorded (Table 3 and Table S1), and frequency distributions of each character are shown for each population (Fig. 1). A clear geographic cline was detected as follows. Cultivated accessions in East Asia In Japan and Korea, small (