Immunodiversity of the Arabidopsis ZAR1 NLR Is Conveyed by Receptor-Like Cytoplasmic Kinase Sensors

Abstract

The Arabidopsis nucleotide-binding leucine-rich repeat protein ZAR1 can recognize at least six distinct families of pathogenic effector proteins to mount an effector-triggered immune response. This remarkable immunodiversity appears to be conveyed by receptor-like cytoplasmic kinase (RLCK) complexes, which associate with ZAR1 to sense several effector-induced kinase perturbations. Here we show that the recently identified ZAR1-mediated immune responses against the HopX1, HopO1, and HopBA1 effector families of Pseudomonas syringae rely on an expanded diversity of RLCK sensors. We show that individual sensors can recognize distinct effector families, thereby contributing to the expanded surveillance potential of ZAR1 and supporting its role as a guardian of the plant kinome.

Keywords: Arabidopsis; Pseudomonas syringae; ZAR1; ZED/ZRK; effector-triggered immunity; immunodiversity; receptor-like cytoplasmic kinase (RLCK).

Figure 1

Expanded ZAR1 immunodiversity is conferred by ZED1-related kinases (ZRKs). (A–C) Arabidopsis Col-0 and two independent knockout lines of zed1 (A), zrk3 (B), and zrk2 (C) were spray inoculated with PtoDC3000 carrying an empty vector (EV) or expressing HopX1i (A), HopO1c (B), or HopBA1a (C). Images were taken 10 days post infection. (D–F) Bacterial growth assays 3 days post infection of PtoDC3000 EV or expressing HopX1i (D), HopO1c (E), or HopBA1a (F) on Arabidopsis Col-0 and zed1-2 (D), zrk3-1 (E), or zrk2-1 (F). Boxplots represent data from seven to eight samples. Letters represent statistically significant differences (Tukey’s HSD, P < 0.05). Experiments were replicated three times with similar results.

Figure 2

HopX1 ETI requires SZE1 but not SZE2. (A) Arabidopsis Col-0, sze1-3, and sze2-1 were spray inoculated with PtoDC3000 EV, HopX1i, or HopZ1a. Images were taken 10 days post infection. (B, C) Bacterial growth assays 3 days post infection of PtoDC3000 EV or expressing HopX1i on Arabidopsis Col-0 and sze1-3 (B), or sze2-1 (C). Boxplots represent data from seven to eight samples. Letters represent statistically significant differences (Tukey’s HSD, P < 0.05). Experiments were replicated three times with similar results.

Figure 3

HopX1i strengthens the ZED1/SZE1 protein-protein interaction. (A) Yeast two-hybrid assay assessing interactions between ZED1 (in pEG202) and PBL15, SZE1, or SZE2 (in pJG4-5) on X-gal reporter plates. To determine protein-protein interaction in the presence of a T3SE, HopX1i, or HopZ1a was integrated into the yeast HO locus (see Methods). (B) Quantitative yeast interaction assays were performed between ZED1 and SZE1 in the presence or absence of the T3SEs HopZ1a (Z1a), HopX1i (X1i), or the catalytic mutant of HopX1i (C198A; X1i C/A). Letters represent statistically significant differences (Tukey’s HSD, P < 0.05). Experiments were replicated three times with similar results. Plasmids or genome integrations expressing each component for every strain used are listed below the bar graph. Strains 5 and 7 were independently generated, but harbor the same combinations of constructs (pBA350V::HopX1i, pEG202::ZED1, and pJG4-5::SZE1). Strains 6 and 8 were independently generated, but harbor the same combinations of constructs (pBA350V::HopX1i C182A, pEG202::ZED1, and pJG4-5::SZE1).

Figure 4

ZAR1 uses a diversity of kinase sensors to recognize P. syringae. Specific ZAR1/ZRK modules confer putative resistance to predominantly independent P. syringae clades. The generation of this P. syringae core genome phylogeny, with associated phylogroup designations (P) is described in Laflamme et al., 2020. Colored bars above the phylogeny represent strains that harbor an ETI eliciting allele that requires ZAR1 (black) and the specific ZRK required for each ZAR1 ETI (blue): ZED1, ZRK3 and ZRK2. Numbers indicate the total number of strains that carry a T3SE whose ETI requires the associated genetic component. Putatively truncated sequences (less than 75% the length of the representative allele) are not displayed (Laflamme et al., 2020).

Figure 5

ZAR1 uses independent combinations of RLCKs to sense T3SE activity. A summary of all known genetic components involved in ZAR1-mediated ETI in A. thaliana and N. benthamiana to T3SEs from P. syringae, X. campestris, and X. perforans. X. campestris T3SE AvrAC uridylylation of PBL2 alters the conformation of the PBL2/RKS1/ZAR1 complex, resulting in ADP/ATP exchange in ZAR1 and the formation of a resistosome structure (Wang et al., 2019a; Wang et al., 2019b). HopZ1 activity modulates PBL/ZED1 interactions and its ETI was shown to redundantly require SZE1 and SZE2 (Bastedo et al., 2019; Liu et al., 2019). HopX1 ETI requires SZE1 and ZED1 (this study). HopF1 and HopO1 ETIs require ZRK3, but associated PBLs have not been identified (Seto et al., 2017) (this study). HopBA1 ETI requires ZRK2, but no associated PBLs have been identified (this study). Note that in the Arabidopsis Ag-0 ecotype, HopBA1 elicits ETI via the TIR-only NLR RBA1 (Nishimura et al., 2017). The X. perforans T3SE XopJ4 requires JIM2, an RLCK XII homologous to ZRKs, and NbZAR1 for ETI (Schultink et al., 2019).