A phospho-switch constrains BTL2-mediated phytocytokine signaling in plant immunity

Abstract

Enabling and constraining immune activation is of fundamental importance in maintaining cellular homeostasis. Depleting BAK1 and SERK4, the co-receptors of multiple pattern recognition receptors (PRRs), abolishes pattern-triggered immunity but triggers intracellular NOD-like receptor (NLR)-mediated autoimmunity with an elusive mechanism. By deploying RNAi-based genetic screens in Arabidopsis, we identified BAK-TO-LIFE 2 (BTL2), an uncharacterized receptor kinase, sensing BAK1/SERK4 integrity. BTL2 induces autoimmunity through activating Ca²⁺ channel CNGC20 in a kinase-dependent manner when BAK1/SERK4 are perturbed. To compensate for BAK1 deficiency, BTL2 complexes with multiple phytocytokine receptors, leading to potent phytocytokine responses mediated by helper NLR ADR1 family immune receptors, suggesting phytocytokine signaling as a molecular link connecting PRR- and NLR-mediated immunity. Remarkably, BAK1 constrains BTL2 activation via specific phosphorylation to maintain cellular integrity. Thus, BTL2 serves as a surveillance rheostat sensing the perturbation of BAK1/SERK4 immune co-receptors in promoting NLR-mediated phytocytokine signaling to ensure plant immunity.

Keywords: NLR; NOD-like receptor; PRR; co-receptor; immune homeostasis; leucine-rich repeat receptor kinase; pattern recognition receptor; phosphorylation; phytocytokine.

Figure 1.. The btl2 mutants suppress BAK1/SERK4-regulated autoimmunity.

(A) The btl2 mutants suppress growth defects triggered by RNAi-BAK1/SERK4. Plant phenotypes are shown two weeks after VIGS of BAK1/SERK4 or the empty vector (Ctrl). Scale bars, 1 cm. (B) The btl2 mutants suppress cell death and H₂O₂ production triggered by RNAi-BAK1/SERK4. Plant leaves were stained with trypan blue for cell death (top) and DAB for H₂O₂ accumulation (bottom). Scale bars, 5 mm. (C) The btl2 mutants suppress PR1 and PR2 expression triggered by RNAi-BAK1/SERK4. The expression of PR1 and PR2 was normalized to that of UBQ10. The data are shown as mean ± SD (n = 4). Different letters denote statistically significant differences according to one-way ANOVA followed by the Tukey test (P < 0.05). (D-F) Complementation of btl2 with pBTL2::gBTL2-HA restores growth defects (D), cell death and H₂O₂ production (E), PR1 and PR2 expression (F) triggered by RNAi-BAK1/SERK4. CL1 and CL2 are two lines with protein shown by an α-HA immunoblot (IB) (D). Protein loading is shown by Ponceau S staining (Ponc.) for RuBisCo (RBC). The experiments and data analysis were performed as in A-C, respectively. (G-H) The btl2–1 mutant rescues the seedling lethality of bak1–4/serk4–1. Soil-grown seedlings are shown at 21 days post-germination (dpg) (G), 35 dpg (H, left), and 49 dpg (H, right). bak1–4/serk4–1 was at 12 dpg (G). Scale bars, 1 cm (G, top). Cell death and H₂O₂ accumulation are shown on the middle and bottom panels (G), respectively. Scale bars, 1 mm. (I) The btl2 mutant does not suppress cell death triggered by RNAi-BIR1 or RNAi-MEKK1. Scale bars, 1 cm. The experiments were repeated three times with similar results. See also Figure S1 and Table S1.

Figure 2.. Increased expression of BTL2 induces autoimmunity.

(A) Transcripts of BTL2 are up-regulated in bak1–4/serk4–1. Gene expression levels of BTL2 in WT and bak1–4/serk4–1 are shown as mean ± SD from two independent repeats of RNA-seq data. The number indicates the fold change. (B-D) Complementation of bak1–4/serk4–1/btl2–1 with pBTL2::gBTL2-HA restores growth defects (B), cell death and H₂O₂ production (C), and PR expression (D). Transgenic plants L1, L3, and L7 representing three categories of growth defects, with the percentages of each category and protein expression, are shown (B). Scale bars, 1 cm. Plant leaves were stained with trypan blue for cell death and DAB for H₂O₂ accumulation (C). Scale bars, 5 mm. The expression of PRs is shown as mean ± SD (n = 4) (D). (E-G) Increased expression of gBTL2 in WT triggers growth defects (E), elevated cell death and H₂O₂ production (F), and PR expression (G). Four categories of pBTL2::gBTL2-HA transgenic plants with BTL2-HA expression are shown (E). Scale bars, 1 cm. Plant cotyledons were stained with trypan blue for cell death and DAB for H₂O₂ accumulation (F). Scale bars, 1 mm. The PR expression (G) was performed as (D). (H) BTL2 but not the kinase mutant (BTL2^KM), activates the PR1 promoter. The pPR1::LUC was co-transfected with BTL2-HA, BTL2^KM-HA, or the empty vector (Ctrl) with pUBQ::GUS as an internal transfection control in protoplasts. The relative luciferase activity was normalized with GUS activity. The data are shown as mean ± SD (n = 3). Different letters in (D), (G), and (H) denote statistically significant differences according to one-way ANOVA followed by the Tukey test (P < 0.05). The experiments in B-H were repeated three times with similar results.

Figure 3.. BTL2 kinase activity is required for BAK1/SERK4-regulated autoimmunity.

(A) Schematic diagram of BTL2 domains. Signal peptide (SP), LRR N-terminal domain (LRRNT), LRR, transmembrane (TM), juxtamembrane (JM), and kinase domains with autophosphorylation T669 are shown. (B) BTL2-GFP colocalized with FM4–64 on the plasma membrane of N. benthamiana leaves. Scale bars, 50 μm (top) and 10 μm (bottom). (C) BTL2-GFP is localized in the leaf and root cell periphery of pBTL2::gBTL2-GFP/btl2–1 transgenic plants. Scale bars, 20 μm. (D) BTL2^JK but not BTL2^JK-KM undergoes autophosphorylation. Phosphorylation was analyzed by autoradiography (top) with protein loading shown by Coomassie blue staining (CBB) (bottom). (E) Complementation of btl2–1 with pBTL2::gBTL2^KM-HA cannot restore growth defects triggered by RNAi-BAK1/SERK. Two gBTL2^KM complementation lines (L1 & L9) and one gBTL2 complementation line (L1) are shown. Scale bars, 1 cm. (F) T669 of BTL2^JK is autophosphorylated in LC-MS/MS analysis. (G)BTL2 T669 is an essential autophosphorylation site in vitro. (H-I) Complementation of bak1–4/serk4–1/btl2–1 with pBTL2::gBTL2^T669A-HA cannot restore growth defects (H), cell death and H₂O₂ production (I). Scale bars, 5 mm. (J) BTL2^T669A is unable to activate the PR1 promoter. The data are shown as mean ± SD (n = 3). Different letters denote statistically significant differences according to one-way ANOVA followed by the Tukey test (P < 0.05). The experiments in B-E and G-J were repeated three times with similar results. See also Figures S2, Data S1.1, and S1.2.

Figure 4.. BAK1 phosphorylates BTL2 at Ser676 to suppress BTL2-induced cell death.

(A) BTL2 associates with BAK1. T₁ plants of pBTL2::gBTL2-FLAG/WT with normal (N) or cell death (D) phenotypes were harvested for immunoprecipitation (IP) with α-FLAG agarose and immunoblotting (IB) with α-BAK1 or α-FLAG (top two panels). Protein inputs are shown (bottom two panels). (B) Interaction between BTL2 and BAK1 by the BiFC assay. BAK1-cYFP and BTL2-nYFP or empty vector (EV) were co-expressed in protoplasts for detecting YFP and chloroplast (Chloro.) signals under confocal microscopy. Scale bars, 10 μm. (C) BAK1^JK interacts with BTL2^JK in the pull-down (PD) assay. GST, GST-BTL2^JK, or GST-BTL2^K immobilized on glutathione sepharose was incubated with MBP, MBP-BAK1^JK-HA, or MBP-BAK1^JK-KM-HA, and pelleted for immunoblotting with α-HA antibody (top) with CBB for input proteins (bottom). (D) BAK1 phosphorylates BTL2. The kinase assay using MBP-BAK1^JK-HA as kinases and GST-BTL2^JK variants as substrates is shown. (E) Ser676 of BTL2^JK is phosphorylated by BAK1^JK in LC-MS/MS analysis. (F) BTL2^S676D is compromised in activating the PR1 promoter. The data are shown as mean ± SD (n = 3). (G) S676 of BTL2 is an important phosphorylation site by BAK1. (H-J) Complementation of bak1–4/serk4–1/btl2–1 by pBTL2::gBTL2^S676D-HA cannot restore growth defects (H), cell death and H₂O₂ production (I), and the PR expression (J). Scale bars, 5 mm. The data in (J) are shown as mean ± SD (n = 4). Different letters in (F) and (J) denote statistically significant differences according to one-way ANOVA followed by the Tukey test (P < 0.05). The experiments except E were repeated three times with similar results. See also Figure S2 and Data S1.3.

Figure 5.. BTL2 activates CNGC20 to induce cell death.

(A) Co-expression of BTL2 and CNGC20 induce cell death, which is suppressed by BAK1. Combinations of BTL2-HA, CNGC20-FLAG, BAK1-GFP, or a GFP vector (Ctrl) were expressed in N. benthamiana. Cell death was visualized as autofluorescence under UV light (top) and H₂O₂ was stained by DAB (middle). The stacked bars show the percentage of different categories (0–3) of cell death severity (bottom). (B) Co-expression of BTL2 and CNGC20 activates the PR1 promoter, which is suppressed by BAK1 in protoplasts. The data are shown as mean ± SD (n = 3). Different letters denote statistically significant differences according to one-way ANOVA followed by the Tukey test (P < 0.05). (C) S676 is important for BTL2-induced cell death in N. benthamiana. (D) The pBTL2::gBTL2-HA transgenic plants do not show severe growth defects in cngc20–1. Scale bar, 1 cm. (E) BTL2 associates with CNGC20. Protoplasts expressing BTL2-HA and CNGC20-FLAG or an empty vector (Ctrl) were subjected to IP with α-FLAG and IB with α-HA or α-FLAG (top two panels) with input proteins shown (bottom two panels). (F) BTL2, but not BTL2^KM, promotes the CNGC20 channel activity. The current-voltage relationship was recorded in Xenopus oocytes injected with water (Ctrl, n = 7), CNGC20-YFP (n = 16), CNGC20-YFP+BTL2-CFP (n = 9), or CNGC20-YFP+BTL2^KM-CFP (n = 4) in the presence of 30 mM CaCl₂. (G) Tunicamycin treatment affects BTL2 migration. Protoplasts from WT or stt3a-2 expressing BTL2-HA treated without or with tunicamycin were subjected to an α-HA immunoblotting. (H) BTL2^{N240/477/545Q} induces weaker cell death than BTL2 when co-expressing with CNGC20 in N. benthamiana. (I) Localization of CNGC20-GFP and CNGC20^{N430/452/455Q}-GFP in WT and stt3a-2. Protoplasts from WT or stt3a-2 were transfected with CNGC20-GFP or CNGC20^{N430/452/455Q}-GFP. Scale bar, 10 μm. The experiments were repeated three times with similar results. See also Figures S3, S4, and Data S1.3.

Figure 6.. BTL2 activates PEP and SCOOP signaling upon depletion of BAK1

(A & B) BTL2 mediates sensitization of PEP1-induced seedling growth inhibition in bak1–4. Seedlings treated without (Ctrl) or with 1 μM PEP1 for seven days are shown (A) with quantification of root length as mean ± SD (n = 6) (B). (C) BTL2 mediates PEP1-induced disease resistance in bak1–4. Four-week-old plant leaves were infiltrated with water (Ctrl) or 100 nM PEP1 followed by hand-inoculation with Pst DC3000 at OD₆₀₀ = 5×10⁻⁴. Bacterial counting at three dpi is shown as means ± SD (n = 3). (D) BTL2 mediates PEP1-induced MAPK activation in bak1–4. Seedlings were treated with 100 nM PEP1 for immunoblotting by α-pERK with protein loading shown by CBB staining. (E) BTL2 mediates PEP1-induced ROS production in bak1–4. Four-week-old plant leaf discs were treated with 100 nM PEP1 for 40 min with data shown as means ± SE (n = 8). (F) BTL2 mediates PEP1-induced callose deposition in bak1–4. Two-week-old seedlings were stained with aniline blue solution 24 hr after 100 nM PEP1 treatment and visualized under UV light with quantification by ImageJ shown as mean ± SD (n = 4). (G) PEP1 induces BTL2 and PEPR1 association. Protoplasts expressing BTL2-HA and PEPR1-FLAG, empty vector (−), or GFP-FLAG were treated with or without 100 nM PEP1 for 15 min for IP with α-FLAG and IB with α-HA or α-FLAG (top three panels) with input proteins shown (bottom three panels). (H) The pepr1/2 mutant partially suppresses growth defects triggered by the increased BTL2 expression. pBTL2::gBTL2-HA transgenic plants in pepr1/2 were phenotypically grouped into three categories with indicated ratios. Scale bar, 1 cm. (I) BTL2 mediates SCOOP10^B-induced ROS production in bak1–4. Leave discs were treated with 1 μM SCOOP10^B for 50 min with data shown as means ± SE (n = 6). (J) BTL2 mediates SCOOP10^B-induced callose deposition in bak1–4. (K) SCOOP10^B induces BTL2 and MIK2 association. Protoplasts were treated with or without 20 μM SCOOP10^B for 15 min. (L) BTL2 mediates SCOOP10^B-induced MAPK activation in bak1–4/serk4–1. Seedlings were treated with 1 μM SCOOP10^B. Asterisks in B and C indicate a significant difference by Student’s two-tailed t-test. (*, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; n.s., no significance). The experiments were repeated three times with similar results. See also Figures S4, S5, S6, and Data S1.4.

Figure 7.. BTL2 activates ADR1-mediated SCOOP signaling and immunity in the absence of BAK1.

(A) Upregulation of peptide genes in bak1–4/serk4–1. The gene expression from RNA-seq data was subjected to log₂ transformation using GraphPad for the heat map. (B) Overexpression of p35S::SCOOPL1-HA triggers MIK2-, BTL2-, and ADR1-dependent growth defects. Scale bars, 1 cm. (C) Alignment of SCOOP and SCOOPL1 consensus sequence with the conserved SxS motif. (D) BTL2 mediates SCOOP10^B- and SCOOPL1-triggered root growth inhibition in bak1–4. Quantification of seedling root length is shown with data as mean ± SD (n = 24). The different letters denote statistically significant differences according to one-way ANOVA followed by the Tukey test (P < 0.05). (E) BTL2 mediates SCOOPL1-induced callose deposition in bak1–4. (F) The mik2–1 and adr1 triple mutants suppress BTL2-triggered growth defects. Scale bars, 1 cm. (G) ADR1s mediate PEP/SCOOP-induced disease resistance. Four-week-old plant leaves were infiltrated with water (Ctrl), 1 μM PEP1, 1 μM SCOOP10^B, or 10 μM SCOOPL1 followed by hand-inoculation with Pst DC3000 at OD₆₀₀ = 1×10⁻³. Bacterial counting at three dpi is shown as means ± SD (n = 6). Asterisks indicate a significant difference by Student’s two-tailed t-test. (**, P < 0.01; ***, P < 0.001). (H) A model of BTL2 function in phytocytokine signaling and ADR1-mediated autoimmunity. Upon pathogen infection, BAK1/SERK4 associate with PRRs perceiving MAMPs and trigger PTI responses and the expression/production of DAMPs/phytocytokines. Phytocytokines, such as PEPs/SCOOPs, are perceived by BAK1/SERK4-associated PRRs to induce ADR1-dependent DTI for a robust and balanced PTI. Meanwhile, BAK1/SERK4 phosphorylates BTL2 at S676 and suppresses BTL2-mediated signaling. Pathogen effectors perturb BAK1/SERK4, leading to the compromised PTI and derepression of BTL2. BTL2 autophosphorylates at T669 and activates CNGC19/CNGC20 Ca²⁺ channels and the production of phytocytokines. BTL2 associates with multiple phytocytokine receptors to potentiate phytocytokine signaling. Amplification of phytocytokine signaling and massive Ca²⁺ influx triggers ADR1-mediated autoimmunity. ADR1s could also facilitate Ca²⁺ influx to promote autoimmunity. The experiments were repeated three times with similar results. See also Figure S6, S7, Data S1.4, S1.5, and S1.6.