What determines the efficiency of N(2)-fixing Rhizobium-legume symbioses?

Biological nitrogen fixation is vital to nutrient cycling in the biosphere and is the major route by which atmospheric dinitrogen (N(2)) is reduced to ammonia. The largest single contribution to biological N(2) fixation is carried out by rhizobia, which include a large group of both alpha and beta-proteobacteria, almost exclusively in association with legumes. Rhizobia must compete to infect roots of legumes and initiate a signaling dialog with host plants that leads to nodule formation. The most common form of infection involves the growth of rhizobia down infection threads which are laid down by the host plant. Legumes form either indeterminate or determinate types of nodules, with these groups differing widely in nodule morphology and often in the developmental program by which rhizobia form N(2) fixing bacteroids. In particular, indeterminate legumes from the inverted repeat-lacking clade (IRLC) (e.g., peas, vetch, alfalfa, medics) produce a cocktail of antimicrobial peptides which cause endoreduplication of the bacterial genome and force rhizobia into a nongrowing state. Bacteroids often become dependent on the plant for provision of key cofactors, such as homocitrate needed for nitrogenase activity or for branched chain amino acids. This has led to the suggestion that bacteroids at least from the IRLC can be considered as ammoniaplasts, where they are effectively facultative plant organelles. A low O(2) tension is critical both to induction of genes needed for N(2) fixation and to the subsequent exchange of nutrient between plants and bacteroids. To achieve high rates of N(2) fixation, the legume host and Rhizobium must be closely matched not only for infection, but for optimum development, nutrient exchange, and N(2) fixation. In this review, we consider the multiple steps of selection and bacteroid development and how these alter the overall efficiency of N(2) fixation.