Intracellular nucleotide-binding and leucine-rich repeat receptors (NLRs) represent the largest family of plant immune receptors and are key to crop breeding for disease resistance (Kourelis and Hoorn, 2018). NLRs perceive pathogen effectors, which are initially evolved to promote virulence by perturbing host processes inside the plant cell (Dou and Zhou, 2012). The activation of NLRs activate effector-triggered immunity (ETI) that is often associated with hypersensitive response (HR), a form of regulated cell death thought to restrict pathogen progression (Cui et al., 2015).

NLR proteins are broadly classified according to their N-terminal signaling domains. NLRs carrying a COILED-COIL (CC) domain, TOLL INTERLUKIN 1 RECEPTOR/RESISTANCE PROTEIN (TIR) domain, and RPW8-like CC domain are called CNLs, TNLs, and RNLs, respectively (Duxbury et al., 2021). Some of the CNLs and TNLs are sensor NLRs because they function by recognizing effector proteins, either directly or indirectly (Cui et al., 2015). Some NLRs act as “helpers” because they are not required for effector-recognition per se but are required for signaling from sensor NLRs (Jubic et al., 2019). RNLs are common helpers that function downstream of all TNLs and some CNLs. Increasing evidence show that activated NLRs form oligomeric resistosomes to trigger immune signaling (Wang et al., 2019; Li et al., 2020; Ma et al., 2020; Martin et al., 2021). Among these, the resistosome formed by the CNL protein HOPZ-ACTIVATED RESISTANCE 1 (ZAR1) is a plasma membrane-localized channel that conduct calcium ion influx to initiate downstream signaling and cell death (Bi et al., 2021), and this mechanism is shared by RNLs (Jacob et al., 2021). For sensor TNLs, the oligomerization of the TIR domain triggers its NADases enzymatic function, and some of the yet to be identified products likely act as signal to regulate the ENHANCED DISEASE SUSCEPTIBILITY 1 (EDS1) family proteins, which then trigger RNL-dependent immune signaling and cell death (Horsefield et al., 2019; Wan et al., 2019; Lapin et al., 2020).

ZAR1 is a unique CNL that recognizes multiple effectors from various bacterial pathogens (Lewis et al., 2008; Lewis et al., 2013; Wang et al., 2015; Seto et al., 2017; Schultink et al., 2019; Martel et al., 2020). It interacts with a class of closely related pseudokinases called ZED1-RELATED KINASEs (ZRKs), named after the founding member HOPZ-EFFECTOR-TRIGGERRED IMMUNITYDEFICIENT 1 (ZED1), in the resting state, giving rise to assorted pre-formed receptor complexes, each recognizing one or more distinct effector proteins from phytopathogenic bacteria. We and others have recently shown that ZAR1 and ZRKs coevolved 140–200 million years ago (Adachi et al., 2022; Gong et al., 2022). The ZAR1-ZRK complex further interacts with members of the RECEPTOR-LIKE CYTOPLASMIC KINASES (RLCKs) family VII, such as PBS1-LIKE 2 (PBL2), when the latter are modified by effectors (Wang et al., 2015; Martel et al., 2020). The combination of ZRKs and PBLs in tertiary protein complexes further expands the recognition spectrum of the ZAR1 protein. A study in Arabidopsis showed multiple effector alleles that trigger ZAR1-specified ETI exist in ∼40% of P. syringae isolates collected around the world (Laflamme et al., 2020), highlighting great potentials in crop protection.

P. syringae pv. actinidiae (Psa) causes a devastating bacterial canker disease on fruit trees of the Actinidia (Ac) genus and has been a global threat to the fruit industry (Butler et al., 2013). To date, no bacterial canker resistance genes have been identified from Ac. Like many other Gram-negative bacterial phytopathogens, Psa delivers effectors into plant cells via the type III secretion system (T3SS) to promote virulence (Jayaraman et al., 2020). These effectors also offer molecular probes for the identification of resistance genes that are potentially useful for improving disease resistance in Ac. Indeed, the Psa effector HopZ5, an acetyltransferase belonging to the YopJ family effectors from both animal and plant bacterial pathogens, is indirectly recognized by the Arabidopsis CNL protein RESISTANCE TO PSEUDOMONAS SYRINGAE MACULICOLA 1 (RPM1) through RPM1-INTERACTING PROTEIN 4 (RIN4) (Jayaraman et al., 2017). Acetylation of RIN4 by HopZ5 leads to the activation of RPM1-mediated ETI. RPM1 and RIN4 also recognize another YopJ family effector AvrBsT from Xanthomonas campestris (Kirik and Mudgett, 2009). Interestingly, the AvrBsT- and HopZ5-triggered immunity is suppressed by the Arabidopsis SUPPRESSOR OF AVRBST ELICITED RESISTANCE 1 (SOBER1) protein, which has been shown to display both phospholipase and deacetylase activities (Kirik and Mudgett, 2009; Bürger et al., 2017; Choi et al., 2021). SOBER1 deacteylates RIN4, and this has been shown to suppress the RPM1 resistance to both HopZ5 and AvrBsT (Choi et al., 2021).

Here, we show that HopZ5 is additionally recognized by ZAR1 in Nicotiana benthamiana (Nb) and Arabidopsis, and this requires the Nb ZRK protein XOPJ4 IMMUNITY 2 (JIM2) and the Arabidopsis ZRK protein ZED1, respectively. Unexpectedly, Arabidopsis ZAR1 and ZED1 are required for disease resistance to the P. s. maculicola ES4326 (Psm) carrying hopZ5, but are largely dispensable for resistance to P. s. tomato DC3000 (Pst) carrying hopZ5. In contrast, RPM1 in the same ecotype is required for resistance to Pst carrying hopZ5, but completely dispensable for resistance to Psm hopZ5. Contrary to the RPM1-specified resistance to Pst hopZ5, which is suppressed by SOBER1, the ZAR1-specified resistance to Psm hopZ5 is completely insensitive to SOBER1. Thus Arabidopsis employs at least two distinct CNLs to recognize a single Psa effector to confer disease resistance to bacteria carrying hopZ5 in a strain-specific manner. Furthermore, we show that hopZ5 contributes to virulence of both Pst and Psm only in the absence of ETI and SOBER1, indicating a role of SOBER1 in the dampening of effector virulence. Together the data uncover previously unknown complexity of effector-triggered immunity and susceptibility.