Abstract
Seed coat colour is an important agronomic trait in adzuki bean (Vigna angularis). Here, we identified four genes controlling seed coat colour in this species. We conducted a quantitative trait locus (QTL) study of 133 recombinant inbred lines (F6) derived from a cross between V. angularis Shumari (red) and landrace Acc2265 (olive buff), by using simple sequence repeat (SSR) markers. We identified a strong QTL, designated OLB1, which explained 54, 43, and 56 % of the total variance in the L* (lightness), a* (redness), and b* (yellowness) values, respectively. In addition, we identified a minor QTL, designated OLB2, which explained 6 % of the total variance in redness. OLB1 and OLB2 were located on linkage group (LG) 1, >80 cM apart. The F1 phenotype was olive buff, implying that the red colour trait was recessive. Next, we crossed two spontaneous mutants, ‘Shiro-shouzu’ (ivory yellow) and ‘TA230002’ (pale olive buff), to the red-seeded ‘Beni-shouzu’. Genetic analysis of the F1, F2, and F3 progenies derived from each parental cross revealed that the ivory yellow and pale olive buff traits were recessive to the red trait. Molecular mapping studies indicated that the ivory yellow trait was controlled by a single Mendelian gene, designated IVY, located on LG 8. The pale olive buff trait was also controlled by a single Mendelian gene, designated POB, located on LG 10. Our findings will facilitate the development of novel red seed cultivars and the introgression of ivory yellow genes into desirable cultivars, via marker-assisted selection.
Similar content being viewed by others
Introduction
Adzuki bean [Vigna angularis (Willd.) Ohwi & Ohashi] has been widely cultivated in Japan, China, South Korea, and Taiwan for many years, and it is one of the most important crops in these countries (Lumpkin and McClary 1994). The annual production of this crop in Japan and China is estimated to be 800,000 metric tons (Vaughan et al. 2005). In Japan, adzuki bean is the second most economically important grain legume after soybean. The purplish-red or dark purple paste prepared from boiled red adzuki beans is called ‘an’ and is an important ingredient in Japanese and Chinese sweets. In Japan, red adzuki beans are used to prepare the sweet ‘wagashi’ and the sugar-glazed boiled bean dish ‘amanatto’. Together, these three uses account for more than 80 % of adzuki bean consumption in Japan. Furthermore, adzuki beans are cooked with glutinous rice to prepare the purplish-red rice that is traditionally consumed at Japanese celebrations such as birthdays and weddings. Some Japanese sweets also use ivory yellow (IVY)‘an’ paste prepared from adzuki cultivars having IVY seeds. Thus, adzuki bean seed coat colour is an important factor for determining quality, as well as taste. Anthocyanins and vignacyanidins are pigments involved in the production of the red colour in adzuki beans (Sasanuma et al. 1996; Yoshida et al. 1996; Takahama et al. 2013). Anthocyanins are degraded during thermal processing, and may therefore be broken down during cooking (Buckow et al. 2010; Patras et al. 2010). In contrast, the presence of vignacyanidin and its isomer in boiled adzuki beans and in the cooking water suggest that this pigment contributes to the coloration of adzuki bean paste and glutinous rice (Takahama et al. 2013). In addition to contributing to seed colour, natural compounds such as anthocyanins and vignacyanidins are currently being investigated for their medicinal and nutritional values, which are derived from their antioxidant properties (Ariga et al. 1988; Ariga and Hamano 1990; Sato et al. 2005).
Vigna angularis is a diploid species with a haploid chromosome number of 11 and a small genome size, estimated at 539 Mbp (Parida et al. 1990). Han et al. (2005) constructed a molecular linkage map for 11 linkage groups from a backcross population of (V. nepalensis × V. angularis) × V. angularis, by using markers for simple sequence repeats (SSR), amplified fragment length polymorphisms, and restriction fragment length polymorphisms. Kaga et al. (2008) subsequently identified 162 quantitative trait loci (QTLs) for 46 domestication-related traits in the F2 population derived from a cross between V. angularis var. nipponensis and V. angularis var. angularis. Among Vigna species, V. angularis possesses the most well-developed transformation systems (Yamada et al. 2001). Mutant lines represent useful alternative means of genetic analysis, gene mapping, and gene cloning, and they facilitate functional genomics research. These mutants will constitute valuable tools for the improvement of agronomically important traits in adzuki bean.
Various seed coat colour mutants have been isolated in adzuki bean, and natural variations in landraces have been reported. Classical genetic studies have identified four genetic loci, contributing to red colour (R), green colour (G), and brown colour (F), and inhibiting red pigmentation (H); however, the genetic stocks used in these previous studies no longer exist (Takahashi and Fukuyama 1917). A recent molecular mapping study showed that red seed coat colour was selected during domestication from a wild ancestor with ivory seeds (Kaga et al. 2008). In V. angularis, the recessive allele sdc3.1.1 located on linkage group (LG) 1 contributes to red seed coat colour (Kaga et al. 2008). In V. nepalensis, seed coat colour (tan vs. red) is controlled by the gene Sdc, located in a similar chromosomal position to sdc3.1.1 (Isemura et al. 2007). In the present study, we clarified the genetic control of seed coat colour and identified the molecular map positions of four genes implicated in the olive buff, IVY, and pale olive buff seed coat colour traits of three V. angularis varieties or cultivars. Our results will facilitate the development of novel red seed coat colour cultivars and the introgression of IVY genes into desirable cultivars, via marker-assisted selection.
Materials and methods
Plant materials and growth conditions
The parental lines (two with red seeds and three with non-red seeds) used for the crosses in this study are shown in Fig. 1. The red seed coat colour parental lines were V. angularis var. angularis Shumari and Beni-shouzu. The parental strains with non-red seed coats were V. angularis accession Acc2265 (Tokachi Agricultural Experiment Station accession number), an olive buff seed coat colour landrace; ‘Shiro-shouzu-Nougyou daigakukou’ (‘Shiro-shouzu’), an IVY seed coat colour mutant originally found among the field-grown ‘Takara-shouzu’ (red seed coat) in 1986; and ‘TA230002’, a pale olive buff seed coat colour mutant originally found among the field-grown ‘Erimo-shouzu’ (red seed coat) in 1994.
All crosses and plant cultivations occurred at the Tokachi Agricultural Experiment Station, unless otherwise noted (Table 1). For the first parental cross (#0626), ‘Shumari’ (female parent) was crossed with Acc2265 (male parent). All F1 plants (n = 10) were grown during the winter of 2010. The F2 population (#0626-F2), comprising 173 individuals, was planted during the summer of 2011. Recombinant inbred lines (RILs), comprising 133 lines of population #0626-F5, were planted, as a single plant per line, in a glasshouse during the summer of 2013.
In additional parental crosses, ‘Beni-shouzu’ (male parent) was crossed with ‘Shiro-shouzu’ (cross #AZ12-001) and with ‘TA230002’ (cross #AZ12-003). All F1 plants from these crosses were grown at Obihiro University during the winter of 2012. Two F2 populations were grown in the experimental field at Tokachi Agricultural Experiment Station during the summer of 2013. For the F3 progeny test, 10 or more F3 plants derived from red seed coat colour F2 plants were grown in the glasshouse at Obihiro University during the summer of 2014.
Phenotypic evaluation of seed coat colour
We harvested self-pollinated seeds of individual plants at maturity and obtained single-seed spectra from 360 to 740 nm by using a spectrophotometer (CM-5; Konica Minolta, Inc.). For #0626-F2 plants, we conducted colour measurements for three self-pollinated seeds of each F2 individual and for 15 F6 seeds of each RIL. Colour measurements consisted of three parameters as follows: L*, indicating lightness; a*, indicating redness from green (negative values) to red (positive values); and b*, indicating yellowness from yellow (positive values) to blue (negative values). We used the average values from each sample in the QTL analysis.
For the #AZ12-001 and #AZ12-003 populations, we visually scored self-pollinated seeds of individual plants for coat colour. We used segregation for red/IVY or red/pale olive buff colour in the F3 progeny of the #AZ12-001 and #AZ12-003 populations, respectively, to determine the genotype of each red seed coat colour F2 plant.
Analysis of simple sequence repeats
For the SSR analysis of population #0626 and its parental lines, DNA was extracted from the fresh young leaves of parental line plants and an F5 individual, via a modified cetyltrimethylammonium bromide (CTAB) method (Suzuki et al. 2012). SSR analysis was performed according to Suzuki et al. (2012), with the following modifications: M13 primer (5′-CACGACGTTGTAAAACGAC-3′; Applied Biosystems), fluorescently labelled with either 6-FAM, VIC, NED, or PET, was added to a final concentration of 0.1 μM; the final concentrations of forward and reverse primers were 0.1 and 0.2 μM, respectively. A 19-nucleotide 5′ M13 tail (as described above) was added to each forward primer (Schuelke 2000). Polymerase chain reaction (PCR) products were analysed by using an ABI Prism 3500 Genetic Analyzer, according to the manufacturer’s instructions (Applied Biosystems), with GeneScan software and selection of GeneScan-600 LIZ as the size standard.
For the SSR analysis of population #AZ12-001, population #AZ12-003, and their parental lines, DNA was extracted from fresh young leaves, and PCR was performed according to the method described by Han et al. (2005). PCR products were separated on 4 % agarose gel or on 10.0–12.5 % acrylamide gel and were then stained with ethidium bromide.
In each of the above-mentioned procedures, 196 SSR primer pairs specific to adzuki bean (Han et al. 2005) were used to detect polymorphisms between parental lines. The primer sequences of the SSR markers are available at http://www.gene.affrc.go.jp/databases-marker_information.php (Han et al. 2005).
Linkage and QTL analyses
We constructed a linkage map by using JoinMap version 4.1 (available at http://www.kyazma.nl/), with a LOD threshold of 3.0. We used the Kosambi mapping function to convert recombination frequencies for mapping distances (Kosambi 1944). We performed interval mapping (IM) and multiple-QTL model (MQM) mapping to identify putative QTLs, by using the established linkage map and the observed phenotypic traits. This method was run with MapQTL® version 6 (Van Ooijen 2009); a P < 0.05 LOD score significance threshold was calculated by creating a group-wide distribution of the data based on a 1000 permutation test. We used LOD peaks to estimate the position of QTLs on the map. In addition, we estimated genetic parameters, i.e. additive effects and variation explained by each QTL. In the QTL analysis of the F6 population, we considered heterozygotes as missing data.
Results
Inheritance and QTL mapping of olive buff seed coat colour in Acc2265 (cross #0626)
All the self-pollinated seeds produced by the F1 progeny derived from cross #0626, between ‘Shumari’ (L* = 26.02, a* = 19.55, b* = 8.85) and Acc2265 (L* = 40.33, a* = 6.84, b* = 19.42), were olive buff in colour (L* = 40.26, a* = 7.06, b* = 20.62). The similarity to the Acc2265 phenotype indicated dominance of the olive buff seed colour trait over the red seed colour trait. According to our unaided visual observations, seed coat colour in the F3 progeny seemed to vary continuously, (Fig. S1). Therefore, we obtained single-seed spectra from 360 to 740 nm, by using a spectrophotometer. The distributions of the L*, a*, and b* values in F2 individuals ranged broadly over the parental values, with two peaks (Fig. 2a, b, c). In the F6 population, the distributions of each value were bimodal and non-overlapping (Fig. 2d, e, f). Scatter plots of each trait showed that the F6 lines fell into one of two groups—a ‘Shumari’ type or an Acc2265 type (Fig. 2g, h, i). The Acc2265 type lines exhibited greater variation in the L*, a* and b* values than did the ‘Shumari’ type lines.
Among the 196 SSR markers, 81 (41.3 %) were polymorphic between the parental lines, ‘Shumari’ and Acc2265. We used these 81 SSR markers to construct a molecular linkage map of adzuki bean. The results of QTL analysis demonstrated that the L*, a*, and b* values were determined by a single 20.0-cM region located between SSR markers CEDG141 and CEDG001 on LG 1 based on a LOD threshold (1000 permutation test, P < 0.05) (Fig. 3; Table 2); we tentatively designated this region as the olive buff1 (OLB1) gene (Fig. 4). This gene explained 54, 43, and 56 % of the total phenotypic variances in lightness, redness, and yellowness, respectively. The Acc2265 allele at OLB1 increased lightness and yellowness, but decreased redness (Table 2). In addition, the a* value was determined by the second QTL, located at a distance of >80 cM from OLB1 and on SSR marker CEDG214 based on a LOD threshold (Figs. 3, 4; Table 2); we tentatively designated this region as the olive buff2 (OLB2) gene. This gene explained 6 % of the total phenotypic variance in redness. The Acc2265 allele at OLB2 decreased redness (Table 2).
Genetic study and molecular mapping of an ivory yellow seed colour mutant (cross #AZ12-001)
All the self-pollinated seeds produced by the F1 progeny derived from cross #AZ12-001, between ‘Shiro-shouzu’ (IVY seeds) and ‘Beni-shouzu’ (red seeds), were red in colour. In the F2 progeny, the ratio of red to IVY seed coat plants was 73:19, or approximately 3:1 (\(\chi_{3:1}^{2}\) = 0.93, P = 0.36). Analysis of the F3 progeny derived from self-pollination of red seed coat F2 plants, constituting a test for the F2 genotype, revealed that the ratio of homozygous to heterozygous red-seeded plants was 27:41, or approximately 1:2 (\(\chi_{1:2}^{2}\) = 1.24, P = 0.26). Our results indicate that the IVY colour trait is determined by the recessive allele of a single Mendelian gene, which we tentatively designated the ivory yellow gene.
Among the 196 SSR markers, 63 (32.1 %) were polymorphic between the parental lines, ‘Shiro-shouzu’ and ‘Beni-shouzu’. On the basis of the LOD scores generated by using JoinMap version 4.1, IVY was linked to SSR markers CEDG016, CEDG030, CEDG059, CEDG092, CEDG112, and CEDG286 on LG 8. The most likely linkage order was [CEDG016, CEDG092]-IVY-CEDG112-CEDG030-CEDG286-CEDG059 (Fig. 4c, d).
Genetic study and molecular mapping of a pale olive buff colour mutant (cross #AZ12-003)
All the self-pollinated seeds produced by the F1 progeny derived from cross #AZ12-003, between ‘TA230002’ and ‘Beni-shouzu’, were red in colour. Among self-pollinating F2 individuals, red and pale olive buff seed coat colour traits occurred in a ratio of 60:16, or approximately 3:1 (\(\chi_{3:1}^{2}\) = 0.63, P = 0.43). Analysis of the F3 progeny derived from self-pollination of red seed coat F2 plants, constituting a test for the F2 genotype, revealed that the ratio of homozygous to heterozygous red-seeded plants was 20:38, or approximately 1:2 (\(\chi_{1:2}^{2}\) = 0.03, P = 0.85). Our results indicate that the pale olive buff colour trait is determined by the recessive allele of a single Mendelian gene, which we tentatively designated the pale olive buff (POB) gene.
Among the 196 SSR markers, 53 (27.0 %) were polymorphic between ‘TA230002’ and ‘Beni-shouzu’. On the basis of LOD scores generated by using JoinMap ver4.1, POB was linked to two SSR markers on LG 10, namely, CEDG081, and CEDG116, at distances of 36.1 cM and 34.9 cM (LOD ≥ 3.0), respectively (Fig. 4e, f). The most likely linkage order was CEDG116-CEDG081-POB (Fig. 4f).
Discussion
Seed coat colour is an important agronomic trait in adzuki bean. The red seed coat colour is preferred by consumers in East Asia, and the IVY seed coat colour is popular in Japan. In addition to red and IVY, various seed coat colours such as black, speckled purple, brown, and green are known to occur. In the present study, we identified four seed coat colour genes, which we tentatively designated as OLB1, OLB2, IVY, and POB. We found that OLB1 and OLB2 were located on LG 1, IVY was located on LG 8, and POB was located on LG 10 (Fig. 4b, d, f). On the basis of our findings, we propose a model for the control of seed coat colour by these four genes. At the OLB1 and OLB2 loci, the red colour trait was recessive to the olive buff colour trait, suggesting that OLB1 and OLB2 play a role in the inhibition of red pigment production. In contrast, the red colour phenotype was dominant at the IVY and POB loci, suggesting that these two genes act as positive factors for red pigmentation and that their effects are manifested in a genetic background of olb1 and olb2. To date, two genes have been reported to control seed coat colour at a chromosomal region similar to the OLB1 locus. Polymorphism in Sdc explains seed colour variation between wild-type V. nepalensis (tan) and a red seed coat colour landrace, JP81481; at this locus, red colour is recessive to tan colour (Isemura et al. 2007). The second gene, Sdc3.1.1, explains the genetic variation in seed coat colour between the ivory-seeded wild adzuki bean JP110658 and the red-seeded Kyoto-Dainagon (Kaga et al. 2008). Interestingly, these non-red seed colour lines (Acc2265, V. nepalensis, and wild adzuki bean) constitute different genetic sources, despite their similarities in seed coat colour inheritance. The multiple allelic relationships between OLB1, Sdc, and Sdc3.1.1 remain to be clarified.
Seed coat colour is typically controlled by the maternal genotype, which delays phenotypic expression of the trait for a single generation. IVY adzuki beans are the second most commonly produced type of adzuki bean in Japan, after red adzuki beans. In the case of the IVY gene, seed coat colour is determined by the recessive allele of a single Mendelian gene; thus, selection for the IVY trait in the segregating population is delayed for at least two generations. The development of co-dominant molecular makers enables us readily to identify desirable plants. In the present study, the markers CEDG016, CEDG092, and CEDG112, which were linked to IVY by <2 cM, were effective for marker-assisted selection to determine the presence and inheritance of IVY alleles. By eliminating unfavourable seed colours via marker-assisted selection, our present findings will facilitate the use of a broader range of genetic stocks with variable seed coat colours, in breeding programs to improve other agronomic traits such as disease resistance and cold tolerance. In addition to their role in seed appearance, the natural compounds involved in seed coat colour are currently being investigated for their potential medicinal and nutritional values, which are derived from their antioxidant properties (Ariga et al. 1988; Ariga and Hamano 1990; Sato et al. 2005). Hence, the results of our present study will facilitate artificial selection, not only for appearance, but also for medicinal and nutritional properties of adzuki bean.
References
Ariga T, Hamano M (1990) Radical scavenging action and its mode in procyanidins B-1, and B-3 from azuki beans to peroxyl radicals. Agric Biol Chem 54:2499–2504. doi:10.1271/bbb1961.54.2499
Ariga T, Koshiyama I, Fukushima D (1988) Antioxidative properties of procyanidins B-1 and B-3 from azuki beans in aqueous system. Agric Biol Chem 52:2717–2722. doi:10.1271/bbb1961.52.2717
Buckow R, Kastell A, Terefe NS, Versteeg C (2010) Pressure and temperature effects on degradation kinetics and storage stability of total anthocyanins in blueberry juice. J Agric Food Chem 58:10076–10084. doi:10.1021/jf1015347
Han OK, Kaga A, Isemura T, Wang XW, Tomooka N, Vaughan DA (2005) A genetic linkage map for azuki bean [Vigna angularis (Willd.) Ohwi & Ohashi]. Theor Appl Genet 111:1278–1287
Isemura T, Kaga A, Konishi S, Ando T, Tomooka N, Han OK, Vaughan DA (2007) Genome dissection of traits related to domestication in azuki bean (Vigna angularis) and comparison with other warm-season legumes. Ann Bot 100:1053–1071. doi:10.1093/aob/mcm155
Kaga A, Isemura T, Tomooka N, Vaughan DA (2008) The genetics of domestication of the azuki bean (Vigna angularis). Genetics 178:1013–1036. doi:10.1534/genetics.107.078451
Kosambi DD (1944) The estimation of map distance from recombination values. Ann Eugen 12:172–175
Lumpkin TA, McClary DC (1994) Azuki bean: botany, production and uses. CAB International, Wallingford
Parida A, Raina SN, Narayan RKJ (1990) Quantitative DNA variation between and within chromosome complements of Vigna species (Fabaceae). Genetica 82:125–133
Patras A, Brunton NP, O’Donnell C, Tiwari BK (2010) Effect of thermal processing on anthocyanin stability in foods; mechanisms and kinetics of degradation. Trends Food Sci Technol 21:3–11
Sasanuma S, Takeda K, Hayashi K (1996) Black red pigment of “adzuki bean” studies on anthocyanins LV. Bot Mag Tokyo 79:807–810
Sato S, Yamate J, Hori Y, Hatai A, Nozawa M, Sagai M (2005) Protective effect of polyphenol-containing azuki bean (Vigna angularis) seed coats on the renal cortex in streptozotocin-induced diabetic rats. J Nutr Bioch 16:547–553. doi:10.1016/j.jnutbio.2005.02.003
Schuelke M (2000) An economic method for fluorescent labeling of PCR fragments. Nat Biotech 18:233–234. doi:10.1038/72708
Suzuki T, Sato M, Takeuchi T (2012) Evaluation of the effects of five QTL regions on Fusarium head blight resistance and agronomic traits in spring wheat (Triticum aestivum L.). Breed Sci 62:11–17. doi:10.1270/jsbbs.62.11
Takahama U, Yamauchi R, Hirota S (2013) Isolation and characterization of a cyanidin-catechin pigment from adzuki bean (Vigna angularis). Food Chem 141:282–288. doi:10.1016/j.foodchem.2013.02.113
Takahashi Y, Fukuyama J (1917) Morphological and genetic studies on the Adzuki-bean. Hokkido Agricultural Experiment Station report no. 7. Hokkaido Agricultural Experiment Station, Sapporo, Japan
Van Ooijen JW (2009) MapQTL® 4, software for the mapping of quantitative trait loci in experimental populations of diploid species. Kyazma BV, Wageningen
Vaughan DA, Tomooka N, Kaga A (2005) Azuki bean [Vigna angularis (L.) Ohwi and Ohashi]. In: Singh RJ, Jauhar PP (eds) Genetic resources, chromosome engineering, and crop improvement. Grain legumes, vol 1. CRC Press, Boca Roca, pp 341–353
Yamada T, Teraishi M, Hattori K, Ishimoto M (2001) Transformation of azuki bean by Agrobacterium tumefaciens. Plant Cell Tissue Organ Cult 64:47–54
Yoshida K, Sato Y, Okuno R, Kameda K, Isobe M, Kondo T (1996) Structural analysis and measurement of anthocyanins from colored seed coats of Vigna, Phaseolus, and Glycine legumes. Biosci Biotechnol Biosci 60:589–593
Acknowledgments
We thank Tokachi Agricultural Cooperative Associations and the Hokkaido College of Agriculture for providing the original seeds of ‘TA230002’ and ‘Shiro-shouzu’, respectively.
Conflict of interest
The authors declare that they have no conflict of interest.
Funding
This study was funded by the Japan Beans Fund Association and the Obihiro University of Agriculture and Veterinary Medicine (10089).
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
10681_2015_1461_MOESM1_ESM.pptx
Supplementary material 1 (PPTX 1042 kb). F3 seeds of a cross between adzuki bean (Vigna angularis) cultivars ‘Shumari’ (red) and Acc2265 (olive buff)
Rights and permissions
About this article
Cite this article
Horiuchi, Y., Yamamoto, H., Ogura, R. et al. Genetic analysis and molecular mapping of genes controlling seed coat colour in adzuki bean (Vigna angularis). Euphytica 206, 609–617 (2015). https://doi.org/10.1007/s10681-015-1461-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10681-015-1461-9