The genetics of domestication in common bean

Wild common bean and its cultivated descendant display a striking contrast at the phenotypic level. These major differences represent an interesting opportunity for the study of evolution and a worthwile challenge for bean breeding. Traditional neo-Darwinian theory holds that evolution proceeds gradually by small mutations. The differences between wild crop ancestors and cultivated descendants allow us to test this tenet. In addition, analyses with biochemical and molecular markers (seed proteins, RFLPs) have shown that domestication has induced a reduction in genetic divesrity. These observations and the identification of arcelin, a seed protein with insecticidal activity, among wild beans suggest that these could represent a source of additional genetic diversity for bean breeding. Our presentation will be articulated in two major parts. In a first part, we will present our most recent data on genetic diversity at the molecular level in common bean using RFLPs for M13-related sequences. Our data confirm two of the essential observations made earlier, namely that of a pronounced divergence between the Middle American and Andean gene pools and of a reduction in genetic diversity during the various stages of domestication, including the latest step effected by breeding programs. In a second part, we will discuss the genetic control of the domestication syndrome in common bean. Our results indicate that the genetic control of many of the traits distinguishing wild and cultivated common bean involves at least one gene with major effect (>30% of the total phenotypic variation). In addition, a limited number of genomic regions appear to be involved. Overall, our results suggest that the genetic differences between wild and cultivated beans are smaller than what could be predicted based on their extensive phenotypic differences. They suggest that evolution can proceed by major mutation and that introgression by quantitative traits from wild into cultivated germplasm should be possible barring any undesirable linkage between genes controlling a trait of interest and genes controlling the domestication syndrome.