Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 15 (7), e1007900

Perturbations of the ZED1 Pseudokinase Activate Plant Immunity


Perturbations of the ZED1 Pseudokinase Activate Plant Immunity

D Patrick Bastedo et al. PLoS Pathog.


The Pseudomonas syringae acetyltransferase HopZ1a is delivered into host cells by the type III secretion system to promote bacterial growth. However, in the model plant host Arabidopsis thaliana, HopZ1a activity results in an effector-triggered immune response (ETI) that limits bacterial proliferation. HopZ1a-triggered immunity requires the nucleotide-binding, leucine-rich repeat domain (NLR) protein, ZAR1, and the pseudokinase, ZED1. Here we demonstrate that HopZ1a can acetylate members of a family of 'receptor-like cytoplasmic kinases' (RLCK family VII; also known as PBS1-like kinases, or PBLs) and promote their interaction with ZED1 and ZAR1 to form a ZAR1-ZED1-PBL ternary complex. Interactions between ZED1 and PBL kinases are determined by the pseudokinase features of ZED1, and mutants designed to restore ZED1 kinase motifs can (1) bind to PBLs, (2) recruit ZAR1, and (3) trigger ZAR1-dependent immunity in planta, all independently of HopZ1a. A ZED1 mutant that mimics acetylation by HopZ1a also triggers immunity in planta, providing evidence that effector-induced perturbations of ZED1 also activate ZAR1. Overall, our results suggest that interactions between these two RLCK families are promoted by perturbations of structural features that distinguish active from inactive kinase domain conformations. We propose that effector-induced interactions between ZED1/ZRK pseudokinases (RLCK family XII) and PBL kinases (RLCK family VII) provide a sensitive mechanism for detecting perturbations of either kinase family to activate ZAR1-mediated ETI.

Conflict of interest statement

The authors have declared that no competing interests exist.


Fig 1
Fig 1. HopZ1a binds to PBL kinases and acetylates PBS1.
(A) Yeast two-hybrid interaction assay using wild-type or catalytically-inactive HopZ1a alleles as bait and an array of 46 PBL kinases as prey. Yeast colonies on an X-Gal reporter plate are shown on top. A grid describing the PBL array layout is shown below each half of the plate, with the colours of array positions representing the averaged colony colour for each bait-prey interaction. We used a ‘relative interaction strength’ metric derived from these averaged colony colours to discriminate strong interactions from the background (see S2 Fig; Materials and Methods); prey array positions with interaction strength ≥ 0.455 (i.e., strong interactions) were assigned white labels while other positions were assigned black labels). The interaction strength threshold is indicated in the colour-bar at right with a horizontal white line. Interaction scheme refers to S1 Fig. (B) LC-MS/MS identification of PBS1 residues acetylated by HopZ1a. The PBS1 protein sequence is indicated as horizontal bars for three experimental replicates in the presence of each HopZ1a allele (wild-type HopZ1a or the catalytically-inactive mutant, HopZ1aC216A. Black bands indicate peptides that were reliably detected by the mass-spectrometry instruments, while peptides not detected with high confidence are indicated by grey shading. (C) LC-MS/MS identification of ZED1 residues acetylated by HopZ1a. Peptide detection and acetylation sites are presented as for panel B above. Note that in addition to S84, T87 and T177, S137 was also acetylated, although this modification was not consistently observed across multiple independent experiments.
Fig 2
Fig 2. HopZ1a promotes binding between ZED1 and several PBL kinases.
Y3H assays testing interactions between ZED1 and 46 PBLs in the presence of HopZ effector alleles are contrasted with a Y2H assay in the absence of effectors. (A) HopZ1awt (left) or HopZ1aC216A (center) were expressed from the chromosome (see Materials and Methods) with ZED1 as bait and 46 individual PBLs as prey. The prey array layout shown at right represents PBL interactions with ZED1 in the presence of HopZ1awt. The colours of the labels at each array position (white or black) are determined by the relative interaction strength (see S2 Fig and Materials and Methods). Interaction scheme refers to S1 Fig. (B) ZED1-PBL interactions in the absence of chromosomally expressed effector (left) or in the presence of HopZ1b (center). The prey array layout shown at right represents PBL interactions with ZED1 in the presence of HopZ1b. Interaction schemes refer to S1 Fig.
Fig 3
Fig 3. ZED1 mutagenesis targets, phylogenetic analysis of PBL kinase domains, and a summary of ZED1-PBL binding interactions.
(A) A linear representation of the ZED1 sequence showing the positions of mutated sites (labeled a-e, according to their descriptions in the text), and predicted secondary structure elements are labeled according to homologous features present in the well-studied kinase, PKA [–91]. β-strands are coloured yellow, while α-helices are coloured dark blue, as for the three-dimensional structural models of ZED1 presented in S7 Fig, S10 Fig and S17 Fig. The activation loop sequence is highlighted with an orange bracket. (B) Left–phylogenetic tree showing relationships between the kinase domains of RLCK family VII (PBS1-like) kinases. Labels for tree leaves are coloured according to yeast colony colours from the HopZ1awt-dependent interactions presented in Fig 2A. Right—condensed graphical summary of the Y2H/Y3H interaction data presented in Fig 2, S4 Fig, S5 Fig, S8 Fig and S9 Fig. Subsets of the five ZED1 mutagenesis targets (sites a-e) that are altered in a given ZED1 allele are indicated at the top of each column. Mutant ZED1 alleles whose expression causes induction of ETI in planta are highlighted with red text. The colour-bar at lower left depicts the range of relative interactions strengths, as defined in S2 Fig (see Materials and Methods).
Fig 4
Fig 4. Isoleucine substitution of a PBL activation loop phospho-accepting residue mimics acetylation by HopZ1a.
(A) Comparison of the molecular structures of phospho-accepting residues L-threonine and L-serine, their acetylated derivatives, candidate acetyl-mimetic residues L-isoleucine and L-glutamine, and L-alanine, a presumed loss-of-function substitution that should block both phosphorylation and acetylation. (B) Mutant alleles of PBL15 were tested in both Y2H (ZED1 bait and PBL15 prey) and Y3H contexts (ZED1 bait and PBL15 prey with HopZ1a expression) to screen for substitutions that confer HopZ1a-independent ZED1 binding activity. Interaction schemes refer to S1 Fig.
Fig 5
Fig 5. ZAR1-ZED1-PBL ternary complexes are formed in the presence either of HopZ1a activity or of a mutant ZED1 allele.
(A) Schematic showing ZAR1 domain truncation boundaries. Subdomains of the central nucleotide-binding region are labeled as described by Wang et al [19,20]: NBD, nucleotide-binding domain; HD1, helical domain 1; and WHD, winged helix domain. (B) Confirmation of expression for each of the three ZAR1 constructs shown in panel A. A western blot of yeast cell lysates probed with serum raised against the LexA DNA-binding domain (top) is compared with Ponceau S staining of the total protein in each lane (bottom). (C) Interactions between ZAR1 bait constructs and 11 PBL preys were assessed in the absence and presence of HopZ1a and/or ZED1 alleles integrated at the ho locus of strains EGY48 (MAT α) or RFY206 (MAT A), corresponding to Y2H (interaction scheme A), Y3H (interaction schemes B), and Y4H (interaction scheme C) assays (see S1 Fig). The prey array layout shown at right represents PBL interactions with ZAR1ΔCC in the presence of both ZED1 and HopZ1awt. The colours of the labels at each array position (white or black) are determined by the relative interaction strength (see S2 Fig; Materials and Methods).
Fig 6
Fig 6. Dexamethasone-induced expression of ZED1‘DT’ (N173D V212T) causes whole-plant HR in Arabidopsis.
BASTA-resistant Arabidopsis transformants bearing dexamethasone-inducible ZED1 alleles (columns 4–7) were tested alongside control plants (columns 1–3; untransformed Col-0, transgenic plants with dexamethasone-inducible AvrRpt2, and untransformed zar1-1). Photographs of control plants and Arabidopsis T1 transformants are shown both before (A) and ~72 h after dexamethasone induction of transgenes (B). Panels A and B present full-colour RGB images (top) as well as a filtered representation (‘Live/Dead filter’; bottom) that converts dead/dying plant tissue and surrounding soil to grayscale pixels while leaving pixels representing healthy Arabidopsis tissue unchanged (see S21 Fig; Materials and Methods). Red numbered circles indicate those plants for which transgene expression was detected and correspond to the electrophoresis lane indices shown in panel C, below. Note that plants presented in this Figure were grown on the same flat and so experienced identical conditions with respect to watering, lighting, and dexamethasone sprays; whole-flat images were cropped to remove unrelated plants. (C) Parallel assessment of transgene expression in tissue from the same plants shown in panels A and B. HA-tagged transgene expression in T1 transformants is shown (top), and Ponceau S staining of total protein transferred to the nitrocellulose membrane (bottom) demonstrates consistent sample loading and protein transfer across all lanes.
Fig 7
Fig 7. Virus-induced gene silencing of NbZAR1 expression abolishes HopZ1a-independent HR triggered by ZED1 alleles.
Leaves from N. benthamiana plants were infected with Tobacco Rattle Virus-derived gene silencing constructs targeting GUS (left) or ZAR1 (right) two weeks prior to transformation by localized pressure-infiltration of Agrobacterium cell suspensions delivering the indicated dexamethasone-inducible transgenes. The outcomes of hypersensitive response (HR) assays for GUS-silenced plants are indicated as the number of strong (+), weak/partial (+/-), or absent (-) responses observed for each infiltration. (The hypersensitive response was absent for all infiltrations in ZAR1-silenced plants.) (A) ZED1V212T substitution mutants are contrasted with ZED1wt and ZED1N173D. Images show infiltrated leaves ~48 h post-induction by spray with dexamethasone. Note that the images showing infiltration spot #6 are from separate leaves but from the same plants as the corresponding infiltration spots #1–5; leaf images were cropped to remove additional unrelated infiltrations. Images are representative of nine replicates derived from three independent experiments. (B) A ZED1T177I substitution mutant is contrasted with ZED1wt. Images are representative of ten different leaves infiltrated in the same way (five replicates each in two independent experiments).
Fig 8
Fig 8. Alternate models for explaining activation of ZAR1-dependent immunity in Arabidopsis.
ZAR1 interacts with pseudokinases and kinases to achieve ADP/ATP nucleotide exchange and assembly of a resistosome. Active or ‘active-like’ kinase domain conformations are represented with rounded rectangles, while inactive and ‘inactive-like’ kinase domain conformations are represented with parallelograms. (A) In the decoy model, post-translationally-modified ZRKs are presumed to adopt an ‘active-like’ conformation that is required to activate ZAR1 and trigger ETI. (B) In the adaptor model proposed by Wang et al. [14], post-translational modification of PBLs results in conversion to an inactive conformation that is sensed by ZRKs to promote nucleotide exchange by ZAR1 and activation of ETI. (C) An alternative adaptor model suggested by our data implies that post-translational-modification of ZRK pseudokinases may be sufficient to recruit PBLs for ZAR1 activation by inducing an ‘active-like’ pseudokinase conformation. In other words, models B and C both allow heterodimer formation only when kinase and pseudokinase have adopted similar structural conformations–inactivated kinases bind to ‘inactive-like’ pseudokinases, but similarly, modified pseudokinases adopting ‘active-like’ conformations as a result of post-translational modifications can also bind to active (unmodified) PBLs to activate ZAR1.

Similar articles

See all similar articles


    1. Jones JDG, Dangl JL. The plant immune system. Nature. 2006;444: 323–329. 10.1038/nature05286 - DOI - PubMed
    1. Leipe DD, Koonin E V., Aravind L. STAND, a class of P-loop NTPases including animal and plant regulators of programmed cell death: multiple, complex domain architectures, unusual phyletic patterns, and evolution by horizontal gene transfer. J Mol Biol. 2004;343: 1–28. 10.1016/j.jmb.2004.08.023 - DOI - PubMed
    1. Duxbury Z, Ma Y, Furzer OJ, Huh SU, Cevik V, Jones JDG, et al. Pathogen perception by NLRs in plants and animals: parallel worlds. BioEssays. 2016;38: 769–781. 10.1002/bies.201600046 - DOI - PubMed
    1. Jones JDG, Vance RE, Dangl JL. Intracellular innate immune surveillance devices in plants and animals. Science (80-). 2016;354: aaf6395–aaf6395. 10.1126/science.aaf6395 - DOI - PubMed
    1. Meunier E, Broz P. Evolutionary convergence and divergence in NLR function and structure. Trends Immunol. Elsevier Current Trends; 2017;38: 744–757. 10.1016/ - DOI - PubMed

Publication types

MeSH terms

Grant support

This work was supported by Natural Sciences and Engineering Research Council of Canada ( Discovery Grants (D.S.G and D.D.), Natural Sciences and Engineering Research Council of Canada Postgraduate Awards (D.S. and A.M.), Canada Research Chairs ( in Comparative Genomics (D.S.G.) and Plant-Microbe Systems Biology (D.D.), and the Center for the Analysis of Genome Evolution and Function (; D.S.G. and D.D.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.