Abstract
ABSTRACT Plant pathogenic fungi secrete a wide variety of small proteins, named effectors. Magnaporthe AVRs and ToxB-like (MAX) effectors constitute a superfamily of secreted proteins widely distributed in Pyricularia (syn. Magnaporthe) oryzae , a devastating fungus responsible for blast disease in cereals such as rice. In spite of high evolutionary sequence divergence, MAX effectors share a common fold characterized by a ß-sandwich core often stabilized by a conserved disulfide bond. In this study, we investigated the structural landscape and diversity within this effector family based on a previous phylogenetic analysis of P. oryzae protein sequences that identified 94 ortholog groups (OG) of putative MAX effectors. Combining protein structure modeling approaches and experimental structure determination, we validated the prediction of the conserved MAX core domain for 77 OG clusters. Four novel MAX effector structures determined by NMR were in remarkably good agreement with AlphaFold2 (AF) predictions. Based on the comparison of the AF-generated 3D models we propose an updated classification of the MAX effectors superfamily in 20 structural groups that highlight variation observed in the canonical MAX fold, disulfide bond patterns and decorating secondary structures in N- and C-terminal extensions. About one-third of the MAX family members remain single, showing no obvious structural relationship with other MAX effectors. Analysis of the surface properties of the AF MAX models also highlights the very high variability remaining within the MAX family when examined at the structural level, probably reflecting the wide diversity of their virulence functions and host targets. Author summary MAX effectors are a family of virulence proteins from the plant pathogenic fungus Pyricularia (syn. Magnaporthe) oryzae that share a similar 3D structure despite very low amino-acid sequence identity. Characterizing the function and evolution of these proteins requires a detailed understanding of their structural diversity. We used a combination of experimental structure determination and structural modeling to characterize in detail the MAX effector repertoire of P. oryzae . A prediction pipeline based on similarity searches and structural modeling using the AlphaFold2 (AF) software were used to predict MAX effectors in a collection of 120 P. oryzae genomes. We then compared AF models with experimentally validated NMR structures. The resulting models and experimental structures revealed that the preserved MAX core coexists with extensive structural variability in terms of structured N- or C-terminal extensions. For each of the AF models, we also analyzed the surfaces of the canonical fold that may be involved in protein-protein interactions. This work constitutes a major step in mapping the functional network of MAX effectors through their structure by identifying possible recognition sites that may help focusing studies of their putative targets in infected plant hosts.