Two key plant adaptations for phosphorus (P) acquisition are carboxylate exudation into the rhizosphere and mycorrhizal symbioses. These target different soil P resources, presumably with different plant carbon costs. We examined the effect of inoculation with arbuscular mycorrhizal fungi (AMF) on amount of rhizosphere carboxylates and plant P uptake for 10 species of low-P adapted Kennedia grown for 23 weeks in low-P sand. Inoculation decreased carboxylates in some species (up to 50%), decreased plant dry weight (21%) and increased plant P content (23%). There was a positive logarithmic relationship between plant P content and the amount of rhizosphere citric acid for inoculated and uninoculated plants. Causality was indicated by experiments using sand where little citric acid was lost from the soil solution over 2 h and citric acid at low concentrations desorbed P into the soil solution. Senesced leaf P concentration was often low and P-resorption efficiencies reached >90%. In conclusion, we propose that mycorrhizally mediated resource partitioning occurred because inoculation reduced rhizosphere carboxylates, but increased plant P uptake. Hence, presumably, the proportion of plant P acquired from strongly sorbed sources decreased with inoculation, while the proportion from labile inorganic P increased. Implications for plant fitness under field conditions now require investigation.

Arbuscular mycorrhizal (AM) symbiosis in vascular plant roots is an ancient mutualistic interaction that evolved with land plants. More recently evolved root mutualisms have recruited components of the AM signalling pathway as identified with molecular approaches in model legume research. Earlier we reported that the reduced mycorrhizal colonisation (rmc) mutation of tomato mapped to chromosome 8. Here we report additional functional characterisation of the rmc mutation using genotype grafts and proteomic and transcriptomic analyses. Our results led to identification of the precise genome location of the Rmc locus from which we identified the mutation by sequencing. The rmc phenotype results from a deletion that disrupts five predicted gene sequences, one of which has close sequence match to the CYCLOPS/IPD3 gene identified in legumes as an essential intracellular regulator of both AM and rhizobial symbioses. Identification of two other genes not located at the rmc locus but with altered expression in the rmc genotype is also described. Possible roles of the other four disrupted genes in the deleted region are discussed. Our results support the identification of CYCLOPS/IPD3 in legumes and rice as a key gene required for AM symbiosis. The extensive characterisation of rmc in comparison with its 'parent' 76R, which has a normal mycorrhizal phenotype, has validated these lines as an important comparative model for glasshouse and field studies of AM and non-mycorrhizal plants with respect to plant competition and microbial interactions with vascular plant roots.

No abstract