Mapping QTLs for Seed Weight, Starch and Protein Content in Common Bean (Phaseolus vulgaris L.)

Seed size was the principle target of selection during domestication of common bean.  An 8 to 10-fold increase in seed size can be observed between landraces and wild accession of the Andean gene pool. This observation raises the question of whether the increase was nutritionally neutral, or it was accompanied by the preferential accumulation of one of the main seed storage components (protein vs starch). This question was addressed via QTL analysis of  seed size, starch and protein content in a RI family (n = 168) generated from a cross between a landrace (G19833) and a wild accession (G23419), both from the Andean gene pool. Seed weight (SW), starch and protein content were obtained from individual seeds of the RI family using near-infrared reflectance (NIR). Seed size was significantly and positively correlated with starch content, and negatively correlated with protein content. As expected, protein and starch contents were negatively correlated. The genotyping-by-sequencing method was used to construct a linkage map with the RIL family. Composite Interval Mapping (CIM) detected four QTL for seed weight, three for starch and 5 for protein. Two of these were detected on LG-G and displayed complete overlap for seed weight and starch content. Also, a QTL on LG-F showed complete overlap for starch and protein and their additivity plots were mirror images indicating that this QTL was primarily responsible for the negative correlation between starch and protein.