Résumé
Because of their ability to fix nitrogen in symbiosis with soil bacteria, legumes such as peas have an important role to play in the development of a sustainable agriculture. However, in the field, biological nitrogen fixation (BNF) could be suboptimal as natural Rhizobium leguminosarum viciae (Rlv) populations are quantitatively and qualitatively heterogeneous, with strains varying in their competitiveness and efficiency of BNF [1,2]. There is a general agreement concerning the interest of Rlv inoculation for improving BNF and thus pea yield [3]. However, even when pea seeds are inoculated by efficient Rlv strains these can be outcompeted by indigenous rhizobia [4]. Previous work has shown the role of signal exchanges between pea and Rlv in the specificity of interaction between them but the knowledge of genetic determinants (Rlv or plant) of partner choice is limited when a pea line is in presence of a mixture of different Rlv strains. In the GRaSP project [5] two complementary approaches are used to address this issue. First, we study the genetic architecture of the trait of partner choice and look for loci that specifically control this trait through genome-wide association genetics. In parallel, we analyze the natural variation of several pea and Rlv genes known to control the symbiotic specificity and study the role of this variation in partner choice. Preferential associations of pea genotype/Rlv strain will be identified and checked for their BNF efficiency. The most promising associations in terms of competitiveness and efficiency will be tested in the field.