Abstract
Plant nucleotide-binding and leucine-rich repeat domain proteins (NLRs) are the most important class of pathogen resistance proteins in plants and have a primary role in sustainable plant health management. They act as immune sensors that recognize pathogen virulence factors named effectors that pathogens secrete during infection to promote disease. Upon effector recognition, NLRs assemble into ring-shaped multimeric complexes called resistosomes that trigger immunity. Some NLRs form heterologous pairs consisting of one sensor NLR (sNLR) that detects the pathogen and one helper NLR (hNLR) that triggers immune signaling.In my PhD, I studied the molecular function of the rice hNLR/sNLR pair RGA4/RGA5 that specifically recognizes the effectors AVR-Pia and AVR1-CO39 from the blast fungus Magnaporthe oryzae. While RGA4 contains only canonical NLR domains, RGA5 has an additional unconventional heavy metal-associated (HMA) domain. This RGA5HMA domain binds the effectors and is crucial for their recognition. Investigation of the three-dimensional structure of the AVR1-CO39/RGA5HMA complex by X-ray crystallography identified a candidate surface for effector-binding in the HMA domain and showed that the HMA domain self-interacts in the absence of effector through the same surface.The first part of my work consisted in the establishment of a high-throughput method to quantify cell death responses in Nicotiana benthamiana agroinfiltration assays, a widely used approach in molecular plant pathology. The new technique measures the reduction of leaf autofluorescence in the red spectrum upon cell death and provides quantitative data for straightforward statistical analysis.The second part of my work aimed at designing artificial NLRs with new effector recognition specificities, which is a promising perspective for sustainable, knowledge-driven crop protection. We introduced into the HMA domain of RGA5 the residues through which another rice sNLR, Pikp-1 recognizes the M. oryzae effector AVR-PikD. This approach created a high-affinity binding surface for the new effector through which engineered RGA5 variants detect the new ligand, AVR-PikD. These results provide a proof of concept for the design of new effector recognition specificities in NLRs through molecular engineering of integrated sensor domains (IDs).The third part of my work elucidated the relevance of HMA homo-dimerization for RGA5 function and the role of the RGA5HMA effector-binding and self-interaction surface in effector recognition. By analyzing point mutations in the RGA5HMA interaction surface with protein interaction studies and N. benthamiana cell-death assays, we found that HMA self-interaction does not contribute to RGA5 function. However, the effector-binding surface of RGA5HMA identified by X-ray crystallography is crucial for effector binding and recognition.As a fourth part of my work, I developed experiments aiming to deepen our understanding of the interaction between RGA4 and RGA5. They indicate that a baculovirus-insect cell system could be suitable to produce recombinant RGA4 and RGA5 for future in vitro studies. Analysis of the association between isolated domains of RGA4 and RGA5 suggest that the NB-ARC domain of RGA5 establishes intramolecular interactions with the HMA domain and interacts, in addition, with the LRR and NB-ARC domains of RGA4.Taken together my work supports the current hypothesis that non-canonical integrated domains of NLRs act as effector traps and deepens our understanding of sNLRs' function within NLR pairs. In addition, it prepared future work aiming at deciphering the functional and physical interactions between RGA4 and RGA5 at the resting and activated states.