To gain a better understanding of wild and weedy azuki population structures in relation to the cultigens we have developed simple sequence repeat (SSR) markers based on a new methodology for plant material. In the azuki bean genome, the number of (AG)(n) and (AC)(n) motif loci per haploid genome has been estimated to be 3,500 and 2,100, respectively, indicating that (AG)(n) motifs are a rich source of markers. We constructed a (AG)(n)-SSR-enriched library in azuki bean in order to obtain a comprehensive range of SSR markers efficiently. The method applied in this study resulted in a 116-fold enrichment over the non-enriched genomic library, with a high percentage (98%) of successful single-locus amplification by the primer pairs designed. Consequently, this method can be applied to construct SSR-enriched libraries suitable for large-scale sequencing. We obtained 255 unique sequences from an (AG)(n)-enriched library for azuki bean. Fifty primer pairs were designed and screened against five populations of wild azuki bean. Among these five populations, one population from Bato town, Tochigi prefecture, Japan, showed greater polymorphism using these primers than the others and was therefore chosen for the in-depth study. The genotypes of 20 individuals were investigated using eight of the SSR primers developed. The genetic relationships among individuals revealed a complex spatial pattern of population structure. Although azuki bean is considered to be a predominantly self-pollinating species, 3 of the 20 individuals tested in the population showed heterozygous genotypes, indicating outcrossing. Allele size and DNA sequence in each of the 20 individuals were compared with those of landraces and released cultivars of azuki bean. Plants in part of the population had many alleles of the same size and with the same sequence as those in cultivated azuki bean, suggesting that gene flow from the cultigen to wild plants has occurred in this population. Unintentional transgene escape from azuki could therefore occur when transgenic azuki is grown in areas where its wild and weedy relatives occur. The approach used here could be applied to biosafety monitoring of transgenic azuki bean.

To make progress in genome analysis of azuki bean (Vigna angularis) a genetic linkage map was constructed from a backcross population of (V. nepalensis x V. angularis) x V.angularis consisting of 187 individuals. A total of 486 markers-205 simple sequence repeats (SSRs), 187 amplified fragment length polymorphisms (AFLPs) and 94 restriction fragment length polymorphisms (RFLPs) -were mapped onto 11 linkage groups corresponding to the haploid chromosome number of azuki bean. This map spans a total length of 832.1 cM with an average marker distance of 1.85 cM and is the most saturated map for a Vigna species to date. In addition, RFLP markers from other legumes facilitated finding several orthologous linkage groups based on previously published RFLP linkage maps. Most SSR primers that have been developed from SSR-enriched libraries detected a single locus. The SSR loci identified are distributed throughout the azuki bean genome. This moderately dense linkage map equipped with many SSR markers will be useful for mapping a range of useful traits such as those related to domestication and stress resistance. The mapping population will be used to develop advanced backcross lines for high resolution QTL mapping of these traits.

Microsatellite discovery from genomic libraries is tedious because of the low number of clones that contain inserts and costly because of screening methodologies. A new procedure for screening clones for microsatellite DNA is described herein. Instead of colony hybridization, a polymerase chain reaction (PCR) with two vector standard primers and one synthesized repeat primer was used to directly screen colonies. PCR of colonies that produced a strong smear in gels contained the desired motif, whereas a single strong band indicated the lack of the desired motif. This simple screening method is a cost‐effective way to identify microsatellite‐containing colonies.

Both restriction fragment length polymorphism (RFLP) and amplified fragment length polymorphism (AFLP) analyses were employed to map a new source of resistance to powdery mildew in mungbean. Disease scores of an F2 population derived from the cross between a moderately resistant breeding line VC1210A and a susceptible wild relative (Vigna radiata var. sublobata, accession TC1966) showed a continuous distribution and was treated as a quantitative trait. Although no significant quantitative trait loci (QTL) that can explain the variation was detected by QTL analysis based on the reconstructed RFLP linkage map, new marker loci associated with resistance were discovered by AFLP analysis. The RFLP loci detected by two of the cloned AFLP bands are associated with resistance and constitute a new linkage group. A major resistance quantitative trait locus was found on this linkage group that accounted for 64.9% of the variation in resistance to powdery mildew. One of the probes developed in this study has the potential to assist in breeding for powdery mildew resistance in mungbean.