The calcium-dependent protein kinase CPK16 regulates hypoxia-induced ROS production by phosphorylating the NADPH oxidase RBOHD in Arabidopsis

Abstract

Reactive oxygen species (ROS) production is a key event in modulating plant responses to hypoxia and post-hypoxia reoxygenation. However, the molecular mechanism by which hypoxia-associated ROS homeostasis is controlled remains largely unknown. Here, we showed that the calcium-dependent protein kinase CPK16 regulates plant hypoxia tolerance by phosphorylating the plasma membrane-anchored NADPH oxidase respiratory burst oxidase homolog D (RBOHD) to regulate ROS production in Arabidopsis (Arabidopsis thaliana). In response to hypoxia or reoxygenation, CPK16 was activated through phosphorylation of its Ser274 residue. The cpk16 knockout mutant displayed enhanced hypoxia tolerance, whereas CPK16-overexpressing (CPK16-OE) lines showed increased sensitivity to hypoxic stress. In agreement with these observations, hypoxia and reoxygenation both induced ROS accumulation in the rosettes of CPK16-OEs more strongly than in the rosettes of the cpk16-1 mutant or the wild type. Moreover, CPK16 interacted with and phosphorylated the N-terminus of RBOHD at 4 serine residues (Ser133, Ser148, Ser163, and Ser347) that were necessary for hypoxia- and reoxygenation-induced ROS accumulation. Furthermore, the hypoxia-tolerant phenotype of cpk16-1 was fully abolished in the cpk16 rbohd double mutant. Thus, we have uncovered a regulatory mechanism by which the CPK16-RBOHD module shapes the ROS production during hypoxia and reoxygenation in Arabidopsis.

Figure 1.

CPK16 contributes to the hypoxia–reoxygenation response in Arabidopsis. A) Seven-day-old ProUBQ10:CPK16-FLAG, ProUBQ10:CPK18-FLAG, or ProUBQ10:CPK28-FLAG transgenic seedlings were subjected to hypoxia (0.1% O₂), followed by reoxygenation (R) for the indicated times. CPK16-FLAG, CPK18-FLAG, or CPK28-FLAG protein was enriched with anti-FLAG beads. Anti-FLAG and anti-pS/T antibodies were used to detect total CPKs (CPK16, CPK18, or CPK28) and phosphorylated CPKs (pCPK16, pCPK18, or pCPK28), respectively. The relative intensity of each band normalized to total CPKs is shown below. Numbers on the left indicate the molecular mass (kD) of each band; hpt, hours post-treatment. B) Four-week-old ProUBQ10:CPK16-FLAG transgenic plants were treated with submergence, followed by recovery (R) for the indicated times. The CPK16-FLAG protein was enriched with anti-FLAG beads. Anti-FLAG and anti-pS/T antibodies were used to detect total CPK16 and phosphorylated CPK16 (pCPK16), respectively. The relative intensity of each band normalized to CPK16 is shown below. Numbers on the left indicate molecular mass (kD) of each band; hpt, hours post-treatment. C) Seven-day-old ProUBQ10:CPK16-FLAG transgenic seedlings were incubated on MS medium (MS) or pretreated with 25 mm EGTA (MS + EGTA) for 4 h, followed by hypoxia exposure to test the inhibition of CPK16 phosphorylation by EGTA. Total proteins were extracted, and CPK16-FLAG protein was enriched with FLAG beads. Anti-FLAG and anti-pS/T antibodies were used to detect total CPK16 and phosphorylated CPK16 (pCPK16), respectively. The relative intensity of each band normalized to CPK16 is shown below. Numbers on the left indicate the molecular mass (kD) of each band; MS, half-strength Murashige and Skoog (MS) liquid medium; hpt, hours post-treatment. D) In vitro phosphorylation assay showing the effect of Ser-274 mutation (S274A) of CPK16 on its autophosphorylation. Recombinant His-CPK16 and His-CPK16^S274A proteins were incubated in protein kinase buffer containing ATPγS and p-nitrobenzyl mesylate (PNBM). Phosphorylated His-CPK16 and His-CPK16^S274A were detected with an anti-thiophosphate ester rabbit monoclonal antibody (α-hapten) after gel electrophoresis (top panel). Recombinant His-CPK16 and His-CPK16^S274A proteins were detected with anti-His antibody (bottom panel).

Figure 2.

CPK16 negatively regulates plant tolerance of hypoxia and submergence stress. A) Phenotypes of 7-d-old wild-type (WT), cpk16-1, and ProUBQ10:CPK16-OE (OE1 and OE2) seedlings subjected to normoxia (Air) or hypoxia (≤0.1% O₂) for 6 h, followed by recovery for 5 d under normoxia. Scale bar in all images, 1 cm. B and C) Survival rates (B) and relative chlorophyll contents (C) for the genotypes shown in (A). D) Phenotype of 4-wk-old WT, cpk16-1, and ProUBQ10:CPK16-OE (OE1 and OE2) plants before submergence (Air) and after submergence treatment for 7 or 8 d (Sub 7 d and Sub 8 d), followed by recovery for 7 d under normal growth conditions. Scale bar in all images, 1 cm. E and F) Survival rates (E) and dry weights (F) for the genotypes shown in (D). For each experiment, 24 seedlings were used per genotype for hypoxia treatment (A), or 14 entire plants were treated per genotype for submergence (B). All experiments were performed as 3 biological replicates with similar results. The representative data from 1 replicate are shown in (C), and data are means + SD (n = 3 technical replicates). In (B), (E), and (F), data are means + SD (n = 3 biological replicates). Different letters represent significant differences at P < 0.05 (1-way ANOVA with Tukey's HSD test).

Figure 3.

CPK16 modulates H₂O₂ accumulation in response to hypoxia reoxygenation. A) DAB staining showing H₂O₂ accumulation in the rosettes of 4-wk-old wild-type (WT), cpk16-1, and ProUBQ10:CPK16-OE (OE1 and OE2) plants before submergence (Air) and after submergence treatment for 3 d (Sub), followed by recovery for 3 h (Recovery) under normal growth conditions. DAB, diaminobenzidine; scale bar, 1 mm. B) H₂O₂ contents in leaves of 4-wk-old WT, cpk16-1, and ProUBQ10:CPK16-OE (OE1 and OE2) plants before submergence (Air) and after submergence treatment for 3 d (Sub), followed by recovery for 3 h (Recovery) under normal growth conditions. C and D) Percentages of ion leakage (C) and water loss (D) in 4-wk-old WT, cpk16-1, and ProUBQ10:CPK16-OE2 plants under submergence treatment or after reoxygenation for the indicated times. For each experiment, 10 rosettes were used per genotype for DAB staining (A) or 3 entire plants were treated per genotype for measurements of H₂O₂ contents (B), ion leakage (C) and water loss (D). All experiments were performed as 3 biological replicates with similar results. Data are means ± SD (n = 4 or 5 biological replicates). Asterisks represent significant differences from WT (*P < 0.05, **P < 0.01 by Student's t-test). In (D), different letters represent significant differences at P < 0.05 (1-way ANOVA with Tukey's HSD test).

Figure 4.

CPK16 interacts with and phosphorylates RBOHD. A) In vivo co-immunoprecipitation (Co-IP) assay showing the interaction between CPK16 and RBOHD. Constructs encoding FLAG-tagged CPK16 and HA-tagged RBOHD were transiently co-transfected into wild-type (WT) Arabidopsis protoplasts, and anti-FLAG affinity magnetic beads were used for immunoprecipitation. B) Bimolecular fluorescence complementation (BiFC) assay showing the interaction between CPK16 and RBOHD in vivo. Constructs encoding nYFP-fused to CPK16 and cYFP-fused to RBOHD were co-transfected into WT Arabidopsis protoplasts and incubated for 16 h under normal light–dark conditions. The constructs encoding the CPK16-nYFP + cYFP and nYFP + RBOHD-cYFP pairs were similarly co-transfected into Arabidopsis protoplasts as negative controls. Confocal images for yellow fluorescent protein (YFP), chlorophyll autofluorescence, and bright field are shown. Scale bars in all images, 10 μm. C) In vivo kinase assay showing the phosphorylation of RBOHD and RBOHD^S4A by CPK16. Constructs encoding FLAG-tagged RBOHD, RBOHD^S4A, and HA-tagged CPK16 kinase domain (CPK16AC) were transiently co-transfected into Arabidopsis cpk16-1 protoplasts, and anti-FLAG affinity magnetic beads were used for immunoprecipitation. Anti-FLAG and anti-pS/T antibodies were used to detect RBOHD and phosphorylated RBOHD (pRBOHD), respectively. The intensity of the pRBOHD/RBOHD protein band is shown below. Numbers on the left indicate molecular mass (kD) of each band. D) In vitro kinase assay showing the phosphorylation of RBOHD-N by CPK16. Recombinant His-CPK16 and MBP-RBOHD-N-His proteins were incubated in protein kinase buffer containing ATP. Phosphorylated His-CPK16 and MBP-RBOHD-N-His were detected with anti-pS/T after gel electrophoresis (top panel). Recombinant His-CPK16 and MBP-RBOHD-N-His proteins were detected with an anti-His antibody (bottom panel). E) In vitro kinase assay showing the inhibition of CPK16 phosphorylation by EGTA. Recombinant His-CPK16 and MBP-RBOHD-N-His proteins were incubated in protein kinase buffer with or without 10 mm EGTA and 0.1 mm CaCl₂. Phosphorylated His-CPK16 and MBP-RBOHD-N-His were detected with anti-pS/T (top panel). Recombinant His-CPK16 and MBP-RBOHD-N-His proteins were detected with anti-His (bottom panel). The intensity of the pRBOHD/RBOHD protein band is shown below. Numbers on the left indicate the molecular mass (kD) of each band. F) Diagram of the RBOHD N-terminus showing the positions of phosphorylation sites recognized by CPK16. EF, EF-hand motif; EFL, EF-hand-like motif. G) In vitro kinase assay showing the phosphorylation of RBOHD-N and RBOHD-N^S4A (mutant form with the 4 Ser residues replaced with Ala residues) by CPK16. Recombinant His-CPK16, MBP-RBOHD-N-His, and MBP-RBOHD-N^S4A-His proteins were incubated in protein kinase buffer containing ATP. Phosphorylated His-CPK16, MBP-RBOHD-N-His, and MBP-RBOHD-N^S4A-His were detected with anti-pS/T after gel electrophoresis (top panel). Recombinant His-CPK16, MBP-RBOHD-N-His, and MBP-RBOHD-N^S4A-His proteins were stained with CBB (Coomassie Brilliant Blue, bottom panel). The intensity of the pRBOHD-N/RBOHD-N protein band is shown below. Numbers on the left indicate the molecular mass (kD) of each band. H) Phosphorylation of RBOHD in WT and the cpk16-1 mutant. Seven-day-old ProUBQ10:RBOHD-HA and cpk16-1 ProUBQ10:RBOHD-HA lines were subjected to hypoxia (0.1% O₂), followed by recovery (R) for the indicated times. RBOHD-HA was tagged with GFP-HA (ProUBQ10:RBOHD-HA). RBOHD was enriched with HA beads, and anti-HA and anti-pS/T antibodies were used to detect RBOHD and phosphorylated RBOHD (pRBOHD), respectively. The intensity of the pRBOHD band is shown below. Numbers on the left indicate the molecular mass (kD) of each band. hpt, hours post-treatment.

Figure 5.

CPK16 promotes RBOHD stability in response to hypoxia. A) RBOHD protein abundance in the presence or absence of CPK16. Constructs encoding FLAG-tagged CPK16AC (0, 5, 10, or 20 μg plasmid DNA) and FLAG-tagged RBOHD were transiently co-transfected into wild-type (WT) Arabidopsis protoplasts. FLAG-RFP was co-transfected as a control for transfection efficiency. B) Effect of the S4A mutations on RBOHD protein abundance. Constructs encoding FLAG-tagged CPK16AC (CPK16 kinase domain) and FLAG-tagged RBOHD or FLAG-tagged RBOHD^S4A were transiently co-transfected into WT Arabidopsis protoplasts. FLAG-RFP was co-transfected as a control for transfection efficiency. C) RBOHD protein abundance under hypoxia treatment. Seven-day-old ProUBQ10:RBOHD-GFP-HA transgenic seedlings were subjected to hypoxia (0.1% O₂) for 0, 1, 3, 9, or 24 h, followed by reoxygenation (R) for 1, 3, or 6 h. Anti-HA antibodies were used for immunoblotting. D) Seven-day-old WT, cpk16-1, and ProUBQ10:CPK16-OE1 seedlings were subjected to hypoxia (0.1% O₂) for 0, 1, and 3 h, followed by reoxygenation (R) for 1 and 3 h. CHX (cycloheximide, 500 μm) was added into medium for the indicated times. Total protein was extracted with protein extraction buffer, and anti-RBOHD antibodies were used for immunoblotting. The relative intensity of each band normalized to the loading control is shown below. Numbers on the left indicate the molecular mass (kD) of each band. hpt, hours post-treatment.

Figure 6.

CPK16 promotes RBOHD-mediated production of ROS in response to hypoxia reoxygenation. A) Phenotype of 7-d-old wild-type (WT), Pro35S:RBOHD-HA (RBOHD-OE1 and RBOHD-OE2), and Pro35S:RBOHD^S4A-HA (S4A-1 and S4A-2) seedlings subjected to hypoxia (≤0.1% O₂) or normoxia (Air) for 6 h, followed by reoxygenation for 5 d under normal growth conditions. Scale bar in all images, 1 cm. B) Relative chlorophyll contents for the genotypes shown in (A). C) DAB (diaminobenzidine) staining showing H₂O₂ accumulation in the rosettes of 4-wk-old WT, Pro35S:RBOHD-HA (RBOHD-OE1 and RBOHD-OE2), and Pro35S:RBOHD^S4A-HA (S4A-1 and S4A-2) before submergence (Air) and after submergence treatment for 3 d (Sub), followed by recovery for 3 h (Recovery) under normal growth conditions. Scale bar in all images, 1 mm. For each experiment, 18 seedlings were used per genotype for hypoxia treatment (A), and 10 rosettes were used per genotype for DAB staining (C). All experiments were performed as 3 biological replicates with similar results. The representative data from 1 replicate are shown, and data are means + SD (n = 3 technical replicates). Asterisks represent significant differences from WT (*P < 0.05, **P < 0.01 by Student's t-test).

Figure 7.

Loss of RBOHD function attenuates the improved tolerance of cpk16-1 mutant to hypoxic stress. A) Phenotype of 7-d-old wild-type (WT), cpk16-1, rbohd, and cpk16 rbohd seedlings subjected to hypoxia (≤0.1% O₂) or normoxia (Air) for 6 h, followed by reoxygenation for 5 d under normal growth conditions. Scale bar in all images, 1 cm. B and C) Survival rates (B) and relative chlorophyll contents (C) for the genotypes shown in (A). D) Phenotype of 4-wk-old WT, cpk16-1, rbohd, and cpk16 rbohd plants before submergence (Air) and after submergence treatment for 7 d (Submergence), followed by recovery for 7 d under normal growth conditions. Scale bar in all images, 1 cm. E and F) Survival rates (E) and relative dry weights (F) for the genotypes shown in (D). For each experiment, 24 seedlings were used per genotype for hypoxia treatment (A), or 14 entire plants were treated per genotype for submergence (D). All experiments were performed as 3 biological replicates with similar result. The representative data from 1 replicate are shown in (C), and data are means + SD (n = 3 technical replicates). In (B), (E), and (F), data are means + Sd (n = 3 biological replicates). Different letters represent significant differences at P < 0.05 (1-way ANOVA with Tukey's HSD test). G) Working model showing the CPK16–RBOHD regulatory module in regulating ROS production in plant response to hypoxia and reoxygenation. Hypoxia-induced cytosolic Ca²⁺ spike triggers the phosphorylation of CPK12 and CPK16. CPK12 translocates from the cytoplasm to the nucleus, where it interacts with and phosphorylates ERF-VII transcription factors to potentiate plant hypoxia sensing. During hypoxia and post-hypoxia reoxygenation, however, activated CPK16 interacts with and phosphorylates plasma membrane-anchored RBOHD to improve its stability and catalyze H₂O₂ production in the apoplast region. Subsequently, extracellular H₂O₂ is transported into the cytoplasm to promote plant hypoxia signaling. The continuously entering H₂O₂ leads to ROS accumulation, which may accelerate oxidative damage and destroy cellular integrity, negatively regulating plant survival from hypoxia. The solid and dashed line arrows indicate direct and indirect positive effects, respectively, and the T-type lines indicate inhibitory effects. The black lines represent interactions presented in this study. Ca²⁺, calcium ion; EFRs, group VII of the ethylene response factor; HREs, hypoxia-responsive genes; O₂, oxygen; PA, phosphatidic acid; ROS, reactive oxygen species.