Genetic mechanisms controlling autoregulation of mycorrhizal symbioses

Plants form mutualistic symbiotic relationships with a diverse array of microbes including arbuscular mycorrhizal fungi (AMF) and rhizobia (resulting in nodulation), which provide nutrients to the plants. However, the formation of these symbioses is an energetically costly process for the plant. In order to balance the energy cost and benefit gained, plants employ negative feedback loops to control the formation of these symbioses, known as autoregulation of nodulation (AON) and autoregulation of mycorrhizae (AOM). Elegant physiological studies in legumes have indicated there is at least some overlap in the genes and signals that regulate AON and AOM. While the molecular/genetic elements involved in AON are relatively well understood, the molecular/genetic basis of AOM is largely unknown. In this thesis, the genes and signals with important roles in the AON pathway of legumes were investigated for their potential role in the AOM pathway. This was examined in both a non-legume (tomato) and legume (pea) system using a mutant based approach. Studies in a non-legume system are important because an efficient autoregulation system might be needed if we want to transfer the nitrogen fixation ability to non-legumes. Briefly, the negative AON feedback loop begins with events associated with nodulation inducing a specific subset of CLE peptides, some of which are tri-arabinosylated by the RDN1 enzyme. These CLE peptides are translocated to the shoot and perceived by a receptor complex, including a CLAVATA1 (CLV1) - like receptor and CLV2. The perception of the signal activates shootderived signal(s) that are transported to the root and inhibit further nodule formation. Studies presented here indicate that the roots of tomato mutants disrupted in the FAB/CLV1, CLV2 and FIN/RDN1 genes are more heavily colonised by mycorrhizal fungi, but the mycorrhizal structures formed in the mutants are similar to the wild type plants. This suggests these genes act to suppress mycorrhizal development as part of the AOM pathway. Reciprocal grafting experiments suggest that the FAB/CLV1 gene acts locally in the root, while the CLV2 gene may act in both the root and the shoot. Strigolactones are key signals in AM and are exuded by plant roots to activate and attract AMF, although their potential role in AOM is unclear. Studies presented here found no difference in the strigolactone level in the fab/clv1 and fin/rdn1 tomato mutants under either mycorrhizal colonised or un-colonised conditions compared with wild type plants. To further test the role of strigolactones in the AOM, a genetic approach was taken using the pea nark ccd8 double mutant, which is disrupted in both the CLV1-like/NARK gene and the strigolactone biosynthesis pathway gene CCD8. Intriguingly, the nark ccd8 double mutants developed extremely low mycorrhizal colonization compared with the nark mutants and wild type plants, indicating that the strigolactones do play an important role in mycorrhizal colonization in nark mutants, but it is still unclear whether strigolactones act downstream of NARK in the AOM pathway. The role of tomato CLE peptides in AOM was also examined. Gene expression studies of the 15 tomato CLE genes were carried out by comparing the expression of these genes in mycorrhizal colonised and uncolonised roots, and in response to nitrogen (N). The expression of the CLE10 gene was significantly higher under high N conditions compared with low and medium N, while the expression of CLE14 and CLE2 were also slightly increased by high N. In contrast, no significant responses to AM were identified for any of the CLE genes. Importantly, the examination of the mycorrhizal phenotype of several cle mutants indicated a role for CLE11 in suppressing AM and identified this gene as another element in the AOM pathway. The AON signalling pathway plays a key role in how legumes inhibit nodulation in response to N. Although AM can also lead to the transfer of substantial amounts of nitrogen to the host, how N may interact with the AOM pathway is unknown. Hence, the interaction between N, mycorrhizal colonization and AOM was also explored. Results showed strong inhibition by high N of mycorrhizal colonization in wild type tomato, and under some conditions this inhibition was systemic. In contrast, this N suppression of mycorrhizae was abolished in the fin/rdn1 and fab/clv1 mutants, indicating that the FAB/CLV1 and FIN/RDN1 genes are required for this suppression. Recent studies in several species have revealed that the CLE-CLV pathway genes are also involved in mediating N induced changes to root morphology. Therefore, the root phenotype and their role in the N regulation of root morphology was also examined in tomato seedlings. Under most growth conditions, the clv2 mutants have smaller roots than the wild type plants, while the fin/rdn1 and fab/clv1 mutants did not differ from wild type plants in their root phenotypes. These three mutants had a similar response in their root morphology under different N concentrations as was observed in wild type plants, which suggests that although these genes play a role in N regulation of AM, they are not involved in mediating N regulation of root morphology in tomato. In conclusion, this study has characterised four main elements that play important roles in AOM in the non-legume system tomato. These are a CLE peptide -CLE11, a putative enzyme that modifies CLE peptides - FIN/RDN1, a receptor like kinases - FAB/CLV1 and a receptor like protein - CLV2. At least some of these elements also play a role in the N regulation of mycorrhizal colonization. This conservation of function between AON and AOM and between legume and non-legume systems provides fundamental knowledge that is required to achieve future goals such as the transfer of nitrogen fixation to non-legume crops.