SummaryDomestication provides an important model for the study of evolution, and information learned from domestication research aids in the continued improvement of crop species. Recent progress in de novo assembly and whole-genome resequencing of wild and cultivated soybean genomes, in addition to new archeological discoveries, sheds light on the origin of this important crop and provides a clearer view on the modes of artificial selection that drove soybean domestication and diversification. This novel genomic information enables the search for polymorphisms that underlie variation in agronomic traits and highlights genes that exhibit a signature of selection, leading to the identification of a number of candidate genes that may have played important roles in soybean domestication, diversification and improvement. These discoveries provide a novel point of comparison on the evolutionary bases of important agronomic traits among different crop species.

SUMMARYTo clarify the molecular bases of flowering time evolution in crop domestication, here we investigate the evolutionary fates of a set of four recently duplicated genes in soybean: FT2a, FT2b, FT2c and FT2d that are homologues of the floral inducer FLOWERING LOCUS T (FT). While FT2a maintained the flowering inducer function, other genes went through contrasting evolutionary paths. FT2b evolved attenuated expression potentially associated with a transposon insertion in the upstream intergenic region, while FT2c and FT2d obtained a transposon insertion and structural rearrangement, respectively. In contrast to FT2b and FT2d whose mutational events occurred before the separation of G. max and G. soja, the evolution of FT2c is a G. max lineage specific event. The FT2c allele carrying a transposon insertion is nearly fixed in soybean landraces and differentiates domesticated soybean from wild soybean, indicating that this allele spread at the early stage of soybean domestication. The domesticated allele causes later flowering than the wild allele under short day and exhibits a signature of selection. These findings suggest that FT2c may have underpinned the evolution of photoperiodic flowering regulation in soybean domestication and highlight the evolutionary dynamics of this agronomically important gene family.

Pleiotropy is considered to have a significant impact on multi-trait evolution, but its roles in the evolution of domestication-related traits in crop species have been unclear. In soybean, several known quantitative trait loci (QTL) controlling maturity, called the maturity loci, are known to have major effects on both flowering and maturity in a highly correlated pleiotropic manner. Aiming at the identification of non-pleiotropic QTLs that independently control flowering and maturity and dissecting the effects of pleiotropy in these important agronomic traits, we conducted a QTL mapping experiment by creating a population from a cross between domesticated soybean G. max and its wild ancestor G. soja that underwent stringent selection for non-pleiotropy in flowering and maturity. Our QTL mapping analyses using the experimental population revealed novel loci that acted in a non-pleiotropic manner: R1-1 controlled primarily flowering and R8-1 and R8-2 controlled maturity, while R1-1 overlapped with QTL, affecting other agronomic traits. Our results suggest that pleiotropy in flowering and maturity can be genetically separated, while artificial selection during soybean domestication and diversification may have favored pleiotropic loci such as E loci that control both flowering and maturity. The non-pleiotropic loci identified in this study will help to identify valuable novel genes to optimize soybean’s life history traits and to improve soybean’s yield potential under diverse environments and cultivation schemes.