Altering specificity and auto-activity of plant immune receptors Sr33 and Sr50 via a rational engineering approach.

Abstract

Many resistance genes deployed against pathogens in crops are intracellular nucleotide-binding leucine-rich repeat receptors (NLRs). The ability to rationally engineer the specificity of NLRs will be crucial in the response to newly emerging crop diseases. Successful attempts to modify NLR recognition have been limited to untargeted approaches or depended on previously available structural information or knowledge of pathogen-effector targets. However, this information is not available for most NLR-effector pairs. Here, we demonstrate the precise prediction and subsequent transfer of residues involved in effector recognition between two closely related NLRs without their experimentally determined structure or detailed knowledge about their pathogen effector targets. By combining phylogenetics, allele diversity analysis, and structural modeling, we successfully predicted residues mediating interaction of Sr50 with its cognate effector AvrSr50 and transferred Sr50's recognition specificity to the closely related NLR Sr33. We created synthetic versions of Sr33 that contain amino acids from Sr50, including Sr33syn that gained the ability to recognize AvrSr50 with 12 amino acid substitutions. Furthermore, we discovered that sites in the leucine rich-repeat domain needed to transfer recognition specificity to Sr33 also influence auto-activity in Sr50. Structural modeling suggests these residues interact with a part of the NB-ARC domain, which we named the NB-ARC latch, to possibly maintain the inactive state of the receptor. Our approach demonstrates rational modifications of NLRs, which could be useful to enhance existing elite crop germplasm.