In plants, specific recognition of pathogen effector proteins by nucleotide-binding leucine-rich repeat (NLR) receptors leads to activation of immune responses. RPP1, an NLR from Arabidopsis thaliana, recognizes the effector ATR1, from the oomycete pathogen Hyaloperonospora arabidopsidis, by direct association via C-terminal leucine-rich repeats (LRRs). Two RPP1 alleles, RPP1-NdA and RPP1-WsB, have narrow and broad recognition spectra, respectively, with RPP1-NdA recognizing a subset of the ATR1 variants recognized by RPP1-WsB. In this work, we further characterized direct effector recognition through random mutagenesis of an unrecognized ATR1 allele, ATR1-Cala2, screening for gain-of-recognition phenotypes in a tobacco hypersensitive response assay. We identified ATR1 mutants that a) confirm surface-exposed residues contribute to recognition by RPP1, and b) are recognized by and activate the narrow-spectrum allele RPP1-NdA, but not RPP1-WsB, in co-immunoprecipitation and bacterial growth inhibition assays. Thus, RPP1 alleles have distinct recognition specificities, rather than simply different sensitivity to activation. Using chimeric RPP1 constructs, we showed that RPP1-NdA LRRs were sufficient for allele-specific recognition (association with ATR1), but insufficient for receptor activation in the form of HR. Additional inclusion of the RPP1-NdA ARC2 subdomain, from the central NB-ARC domain, was required for a full range of activation specificity. Thus, cooperation between recognition and activation domains seems to be essential for NLR function.

Hyaloperonospora arabidopsidis (Hpa) is an obligate biotroph oomycete pathogen of the model plant Arabidopsis thaliana and contains a large set of effector proteins that are translocated to the host to exert virulence functions or trigger immune responses. These effectors are characterized by conserved amino-terminal translocation sequences and highly divergent carboxyl-terminal functional domains. The availability of the Hpa genome sequence allowed the computational prediction of effectors and the development of effector delivery systems enabled validation of the predicted effectors in Arabidopsis. In this study, we identified a novel effector ATR39-1 by computational methods, which was found to trigger a resistance response in the Arabidopsis ecotype Weiningen (Wei-0). The allelic variant of this effector, ATR39-2, is not recognized, and two amino acid residues were identified and shown to be critical for this loss of recognition. The resistance protein responsible for recognition of the ATR39-1 effector in Arabidopsis is RPP39 and was identified by map-based cloning. RPP39 is a member of the CC-NBS-LRR family of resistance proteins and requires the signaling gene NDR1 for full activity. Recognition of ATR39-1 in Wei-0 does not inhibit growth of Hpa strains expressing the effector, suggesting complex mechanisms of pathogen evasion of recognition, and is similar to what has been shown in several other cases of plant-oomycete interactions. Identification of this resistance gene/effector pair adds to our knowledge of plant resistance mechanisms and provides the basis for further functional analyses.

Interactions between Arabidopsis thaliana and its native obligate oomycete pathogen Hyaloperonospora arabidopsidis (Hpa) represent a model system to study evolution of natural variation in a host/pathogen interaction. Both Arabidopsis and Hpa genomes are sequenced and collections of different sub-species are available. We analyzed ∼400 interactions between different Arabidopsis accessions and five strains of Hpa. We examined the pathogen's overall ability to reproduce on a given host, and performed detailed cytological staining to assay for pathogen growth and hypersensitive cell death response in the host. We demonstrate that intermediate levels of resistance are prevalent among Arabidopsis populations and correlate strongly with host developmental stage. In addition to looking at plant responses to challenge by whole pathogen inoculations, we investigated the Arabidopsis resistance attributed to recognition of the individual Hpa effectors, ATR1 and ATR13. Our results suggest that recognition of these effectors is evolutionarily dynamic and does not form a single clade in overall Arabidopsis phylogeny for either effector. Furthermore, we show that the ultimate outcome of the interactions can be modified by the pathogen, despite a defined gene-for-gene resistance in the host. These data indicate that the outcome of disease and disease resistance depends on genome-for-genome interactions between the host and its pathogen, rather than single gene pairs as thought previously.