Skip to main content
ARS Home » Southeast Area » Dawson, Georgia » National Peanut Research Laboratory » Research » Publications at this Location » Publication #223393
Title: Quantitative Trait Loci Mapping of Seed Hardness in Soybean

Author
item ZHANG, BO - University Of Arkansas
item CHEN, PENGYIN - University Of Arkansas
item Chen, Charles
item WANG, D - Michigan State University
item SHI, AINONG - University Of Arkansas
item HOU, ANFU - University Of Arkansas
item ISHIBASHI, TESUAKI - University Of Arkansas

Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/11/2008
Publication Date: 7/1/2008
Citation: Zhang, B., Chen, P., Chen, C.Y., Wang, D., Shi, A., Hou, A., Ishibashi, T. 2008. Quantitative Trait Loci Mapping of Seed Hardness in Soybean. Crop Science. Vol 48: 1341-1349.

Interpretive Summary: Soybean seeds with undesirable texture such as seed hardness will cause a difficulty on processing of soyfood production. To develop food-grade soybean variety with less seed hardness is one of the economic and efficient solutions for the problem. Evaluation of seed hardness in soybeans is time consumed and cost. Identifying quantitative trait loci (QTL) associated with seed hardness in soybeans will allow DNA marker assisted selection applicable to soybean breeders to select less seed hardness in a more efficient fashion. Therefore, the objective of this study was to identify QTL associated with seed hardness in soybean. Three generations of the mapping population from a hard (‘Camp’) × soft (‘SS-516’) soybean cross were grown in a replicated test in Fayetteville, AR, in 2003, 2004, and 2005. We used a texture analyzer to determine pressure-cooked samples from each line for seed hardness. A total of 874 simple sequences repeat (SSR) markers were used to screen the parents, and 177 out of 236 markers showing differences between the parents. A linkage map based on the population was established with 148 SSR markers, 15 of which were new and added to the current public soybean genetic linkage map. All identified markers were placed on 19 linkage groups (LGs). Broad sense heritability was estimated to be 0.56 for seed hardness. Two stable QTL, Ha1 and Ha2, acorss environments were identified near SSR markers “Satt229” on LG L and “Satt531” on LG D1a, respectively, for the average seed-hardness over three years. Ha1 explained 12.7 %, of the phenotypic variance and Ha2 explained 36.1 % of the phenotypic variance, respectively. The markers, Satt229 and Satt531, could be used in breeding program as indications of selection for seed hardness.

Technical Abstract: Soybean seeds with undesirable texture cause processing complications in soyfood production. Seed hardness is an important quality attribute for food-grade soybeans. The objective of this study was to identify QTL associated with seed hardness in soybean. Three generations of F2-derived lines (159 F2:3, F2:4, and F2:5 lines) from a hard (‘Camp’) × soft (‘SS-516’) soybean cross were grown in a replicated test in Fayetteville, AR, in 2003, 2004, and 2005. Pressure-cooked samples from each line were tested for seed hardness using a texture analyzer. A total of 874 simple sequences repeat (SSR) markers were used to screen the parents, and 177 out of 236 bimorphic markers between the parents showed polymorphism in the F2:3 lines. A linkage map for seed hardness was established using 148 SSR markers, 15 of which were new and added to the current public soybean genetic linkage map. All identified markers were placed on 19 linkage groups (LGs) and covered 1363.7 cM of the soybean genome with an average distance of 9.6 cM between markers. Broad sense heritability was estimated to be 0.56 for seed hardness. Two stable QTL acorss environments (Ha1 and Ha2, p < 0.00001) were identified near Satt229 on LG L and Satt531 on LG D1a, respectively, for the average seed-hardness over three years. Ha1 had an LOD score of 6.17 with R2 = 12.7 %, and Ha2 had an LOD score of 5.08 with R2 = 36.1 %. A dominance-by-dominance interaction was detected between Ha1 and Ha2, which explained 7.9 % of the phenotypic variance.