The OsCPK17-OsPUB12-OsRLCK176 module regulates immune homeostasis in rice

Abstract

Plant immunity is fine-tuned to balance growth and defense. However, little is yet known about molecular mechanisms underlying immune homeostasis in rice (Oryza sativa). In this study, we reveal that a rice calcium-dependent protein kinase (CDPK), OsCPK17, interacts with and stabilizes the receptor-like cytoplasmic kinase (RLCK) OsRLCK176, a close homolog of Arabidopsis thaliana BOTRYTIS-INDUCED KINASE 1 (AtBIK1). Oxidative burst and pathogenesis-related gene expression triggered by pathogen-associated molecular patterns are significantly attenuated in the oscpk17 mutant. The oscpk17 mutant and OsCPK17-silenced lines are more susceptible to bacterial diseases than the wild-type plants, indicating that OsCPK17 positively regulates rice immunity. Furthermore, the plant U-box (PUB) protein OsPUB12 ubiquitinates and degrades OsRLCK176. OsCPK17 phosphorylates OsRLCK176 at Ser83, which prevents the ubiquitination of OsRLCK176 by OsPUB12 and thereby enhances the stability and immune function of OsRLCK176. The phenotypes of the ospub12 mutant in defense responses and disease resistance show that OsPUB12 negatively regulates rice immunity. Therefore, OsCPK17 and OsPUB12 reciprocally maintain OsRLCK176 homeostasis and function as positive and negative immune regulators, respectively. This study uncovers positive cross talk between CDPK- and RLCK-mediated immune signaling in plants and reveals that OsCPK17, OsPUB12, and OsRLCK176 maintain rice immune homeostasis.

After stimulation, OsCPK17 rapidly phosphorylates OsRLCK176; phosphorylation attenuates OsPUB12-mediated OsRLCK176 ubiquitination and degradation, which enhances rice immunity.

IN A NUTSHELL.

Background: Plants need to keep a balance between growth and defense against diseases. How plants fine-tune this trade-off remains largely a mystery. We identified 2 protein components in rice disease resistance, a calcium-dependent protein kinase OsCPK17 and an E3 ubiquitin ligase OsPUB12. OsCPK17 promotes accumulation of another immune protein, OsRLCK176, and helps rice plants fight off various diseases. On the other hand, OsPUB12 breaks down OsRLCK176 to prevent excessive immunity. We wanted to understand how the 2 immune regulators, OsCPK17 and OsPUB12, work together to keep rice plants robust and healthy.

Question: We first aimed to answer the following question: How does the calcium-dependent protein kinase OsCPK17 regulate rice defenses against pathogen attacks? To answer this question, we identified 2 other proteins, OsRLCK176 and OsPUB12, through protein–protein interaction screening. Next, we asked: How do these proteins work together to control disease resistance in rice?

Findings: In rice, OsRLCK176 is a central player in disease resistance. To balance growth and defense in rice, OsRLCK176 is under tight control by different immune components. We revealed that a calcium-dependent protein kinase OsCPK17 functions as a “boost” button in the rice immune system. When a pathogen threat appears, OsCPK17 is activated to promote OsRLCK176 accumulation and thereby enhance basal defense in rice. However, without danger signals, the E3 ubiquitin ligase OsPUB12 targets OsRLCK176 for recycling, thus keeping basal defense low. We reveal that OsCPK17, OsPUB12, and OsRLCK176 function as a team to fine-tune trade-offs between rice growth and defense.

Next steps: The calcium-dependent protein kinases OsCPK17 and OsCPK4 function as positive and negative regulators through interacting with OsRLCK176, respectively. Future research should examine the molecular mechanisms by which the 2 calcium sensors function in seemingly opposite ways in rice resistance and aim to create disease-resistant rice plants based on the uncovered immune mechanisms.

Introduction

Plants have evolved a complex immune system to defend against pathogen attacks during long-term coevolutionary interactions (Bredow and Monaghan 2019). Pattern recognition receptors (PRRs) constitute the surveillance system that recognizes “nonself” molecules from attacking microbes or damage signals to trigger defense responses termed pattern-triggered immunity (PTI). In plants, PTI responses include reactive oxygen species (ROS) production, pathogenesis-related (PR) gene expression, and activation of mitogen-associated protein kinases (MAPKs) and calcium-dependent protein kinases (CDPKs) (Jones and Dangl 2006; Boller and Felix 2009; Boller and He 2009).

In plants, Ca²⁺ influx often occurs in response to biotic and abiotic stresses, and elevated Ca²⁺ signals are sensed and decoded by Ca²⁺ sensor proteins, such as CDPKs (or CPKs) (Romeis and Herde 2014; Shi et al. 2018a). Unlike other Ca²⁺ sensors, CDPKs directly sense, respond to, and translate Ca²⁺ signals into downstream signal transduction events (Poovaiah et al. 2013). Protein phosphorylation and dephosphorylation is a key mechanism for Ca²⁺ signaling. Phosphorylation may control the activation of multifunctional kinases in different pathways by allowing CDPKs to sense specific Ca²⁺ signals (Bredow et al. 2021).

Multiple CDPK genes in plants are differentially regulated after pathogen infection and are genetically required for defense responses, suggesting that CDPKs have broad roles in plant immune signaling. For instance, in Arabidopsis thaliana, AtCPK5 rapidly regulates RESPIRATORY BURST OXIDASE HOMOLOG D (RBOHD)-mediated H₂O₂ production and enhances long-lasting defense, even systemic defense (Dubiella et al. 2013; Liu et al. 2017). Meanwhile, other CDPKs negatively regulate plant immunity. OsCPK4 functions as a negative regulator of rice (Oryza sativa) defenses by promoting OsRLCK176 degradation (Wang et al. 2018b). AtCPK28 responds to pathogen-associated molecular pattern (PAMP)-induced Ca²⁺ burst and functions as a negative regulator of BOTRYTIS-INDUCED KINASE 1 (BIK1)-mediated immunity (Monaghan et al. 2015). Phosphorylation of Ser318 leads to a conformational change in AtCPK28 so that the kinase can be activated at a low Ca²⁺ concentration and thus rapidly responds to immune signals (Bredow et al. 2021). The function of CDPKs as a Ca²⁺ sensor and responder makes them unique enzymes in plant immunity.

OsCPK17 encodes a member of the CDPK I subfamily and is downregulated by cold, drought, and salt stresses in rice (Wan et al. 2007; Almadanim et al. 2017). Through a comparative phosphoproteomic approach, OsCPK17 was found to phosphorylate some putative targets and play a crucial role in cold tolerance. The sucrose–phosphate synthase OsSPS4 and aquaporin OsPIP2;1/OsPIP2;6 are phosphorylated by OsCPK17 in a calcium-dependent manner (Almadanim et al. 2017). After being activated by calcium, OsCPK17 functions in signal transduction by autophosphorylation or possibly by phosphorylation of unknown kinases (Raorane et al. 2013; Almadanim et al. 2018). However, it is yet unknown whether OsCPK17 is involved in regulating plant immunity.

Some receptor-like cytoplasmic kinases (RLCKs) are phosphorylated by membrane-bound immune receptors, and in turn, RLCKs directly phosphorylate and regulate target proteins to activate PTI (Bi et al. 2018; DeFalco and Zipfel 2021; Köster et al. 2022). For instance, AtBIK1 and other members in A. thaliana RLCK VII subfamily are direct substrates of multiple PRR complexes. Activated AtBIK1 phosphorylates the NADPH oxidase RBOHD to promote ROS production (Kadota et al. 2014; Li et al. 2014; Gao et al. 2021). In addition, OsRLCK176, OsRLCK185, and OsRLCK107, the close homologs of AtBIK1 in rice, interact with CHITIN ELICITOR RECEPTOR KINASE 1 (OsCERK1) and are required for chitin and peptidoglycan signaling pathways (Couto and Zipfel 2016; Li et al. 2017). On the other hand, the homeostasis of RLCKs is negatively regulated by certain CDPKs in plants. AtCPK28 and OsCPK4 negatively regulate immune signaling by promoting the degradation of AtBIK1 and OsRLCK176, respectively (Wang et al. 2018a, 2018b). Both AtBIK1 and OsRLCK176 are degraded by the ubiquitin proteasome system, providing protection against inappropriately high immune signals (Monaghan 2018). However, it is largely unknown how plants stabilize these positive immune regulators and maintain their homeostasis.

Ubiquitination-mediated protein degradation is an important mechanism in regulating plant immunity (Zeng et al. 2008). Covalent modification of ubiquitin with substrate proteins requires ubiquitin-activating enzyme E1, ubiquitin-conjugating enzyme E2, and ubiquitin E3 ligase (Hua and Vierstra 2011; Callis 2014; Ryu et al. 2019). According to functional domains, E3 ligases can be classified into 3 major groups, among which plant U-box (PUB) E3 ligases are a group of U-box containing proteins (Morreale and Walden 2016; Ryu et al. 2019). A total of 63 and 77 PUB-type E3 ligases have been identified in Arabidopsis and in rice, respectively (Wiborg et al. 2008; Zeng et al. 2008; Yee and Goring 2009). Multiple PUB E3 ubiquitin ligases function as negative immune regulators. For instance, AtPUB25 and AtPUB26 polyubiquitinate and degrade AtBIK1 and thereby attenuate plant immunity (Monaghan et al. 2014; Wang et al. 2018a). The homologous proteins AtPUB22/23/24 in Arabidopsis also act as negative regulators of PTI in response to various PAMPs (Trujillo et al. 2008). Besides, the Arabidopsis immune receptor FLAGELLIN-SENSITIVE 2 (AtFLS2) is a substrate for AtPUB12 and AtPUB13 E3 ligases that inhibit flagellin-induced immune responses (Lu et al. 2011). Likewise, OsPUB11/SPL11, one of AtPUB13 orthologs in rice, ubiquitinates and degrades SPIN6 and functions as a negative regulator of cell death and defense responses in plants (Zeng et al. 2004; Liu et al. 2015). In addition, OsPUB41 interacts with OsUBC25 and OsCLC6 and functions as a negative regulator in drought stress tolerance (Seo et al. 2021). By contrast, OsPUB44 might positively regulate immune responses, because silencing of OsPUB44 attenuates peptidoglycan- and chitin-triggered immune responses and resistance against Xanthomonas oryzae infection (Ishikawa et al. 2014). Despite the broad involvement of plant PUBs in response to external biotic stresses, molecular mechanisms of these PUBs in regulating disease resistance remain largely unknown.

It has been previously demonstrated that OsCPK17 is required for cold stress responses in rice and affects the activity of membrane channels and sugar metabolism (Almadanim et al. 2017). Here, we show that OsCPK17 positively regulates plant immunity through phosphorylating OsRLCK176 and promoting its stability. We further reveal that OsPUB12 ubiquitinates OsRLCK176 for degradation. OsCPK17-mediated phosphorylation of OsRLCK176 at Ser83 inhibits ubiquitination and degradation of OsRLCK176 mediated by OsPUB12. Therefore, OsCPK17 and OsPUB12 function as positive and negative immune regulators, respectively, to fine-tune defense responses through maintaining OsRLCK176 homeostasis in rice.

Results

OsCPK17 positively modulates rice resistance to bacterial blight and leaf streak

To investigate whether OsCPK17 regulates rice immunity, a mutant line (4A-50582, hereinafter called oscpk17-1) with a T-DNA insertion in the first intron of OsCPK17 was identified from Rice Mutant Database (http://signal.salk.edu/cgi-bin/RiceGE) (Fig. 1A). OsCPK17 expression was not detectable in the oscpk17-1 mutant by reverse transcription PCR (RT-PCR), indicating that oscpk17-1 is a null mutant (Fig. 1B). Furthermore, we generated and identified an OsCPK17 RNAi line with a significantly lower OsCPK17 expression (Fig. 1C). Subsequently, the wild-type, oscpk17-1, and RNAi lines were challenged with the bacterial blight pathogen X. oryzae pv. oryzae (Xoo) PXO99 and the bacterial leaf streak pathogen X. oryzae pv. oryzicola (Xoc) RS105. The oscpk17-1 and RNAi lines exhibited significantly longer disease lesions on the PXO99-inoculated leaves than did the wild-type plants (Fig. 1, D and E; Supplemental Fig. S1A). Consistently, the Xoo population size in the oscpk17-1 line was significantly greater than that in the wild-type plants at 9-d post-inoculation (dpi) and thereafter (Fig. 1F). The wild-type and oscpk17-1 leaves exhibited no significant difference in bacterial population and the amount of 16S rDNA at the early PXO99-infection stages (up to 6 dpi) (Fig. 1F; Supplemental Fig. S1B). Similarly, the oscpk17-1 and RNAi lines showed significant longer disease lesions on the RS105-inoculated leaves than did the wild-type plants (Fig. 1, G and H; Supplemental Fig. S1C). Taken together, these results indicate that OsCPK17 positively modulates rice resistance to bacterial blight and bacterial leaf streak diseases.

Figure 1.

OsCPK17 positively modulates rice resistance to bacterial blight and bacterial leaf streak. A) A schematic diagram to show a T-DNA insertion in OsCPK17 (locus Os07g06740) in the oscpk17-1 mutant line 4A-50582. A T-DNA insertion in the first intron of OsCPK17 is indicated by a triangle. Exons and introns, adapted from MSU Rice Genome Annotation Project (RGAP), are represented by black boxes and lines, respectively. The start codon ATG and stop codon TAA are also labeled. B) The OsCPK17 expression level was detected in the oscpk17-1 line by RT-PCR assay. OsACTIN was used as an internal reference gene. WT, the wild-type plant. C) An OsCPK17 RNAi line with a significantly lower expression of OsCPK17 was identified by quantitative RT-PCR. ** indicates a significant difference in the OsCPK17 transcript level between the wild-type and RNAi lines (Student's t-test; **P < 0.01). D, E) Length of disease lesions caused by X. oryzae pv. oryzae (Xoo) PXO99 in the wild-type, oscpk17-1 D) and RNAi E) lines. The length of disease lesions on inoculated leaves was measured at 14 dpi. Representative data from 3 independent experiments are shown as mean ± Sd (in D, n = 31; in E, n = 36). Asterisks indicate significant differences in the lesion length on the inoculated leaves between the wild-type and oscpk17-1 or RNAi lines (Student's t-test; ***P < 0.001). F) The Xoo population size in the inoculated oscpk17-1 leaves was significantly greater than that in the wild-type leaves at 9 dpi and thereafter (Student's t-test; *P < 0.05). G, H) The average length of disease lesions on the wild-type, oscpk17-1 G) and RNAi H) leaves after inoculation with X. oryzae pv. oryzicola RS105 by pressure infiltration. The length of disease lesions on inoculated leaves was measured at 14 dpi. Representative data from 3 independent experiments are shown as mean ± Sd (n = 12). The oscpk17-1 and RNAi lines exhibited significantly longer disease lesions on the inoculated leaves than did the wild-type plants (Student's t-test; ***P < 0.001).

OsCPK17 positively regulates ROS generation and immune marker gene expression

To further determine whether OsCPK17 functions in plant immunity, PAMP-induced defense responses were detected in the wild-type and oscpk17-1 lines. Oxidative burst induced by chitin and flg22 in the oscpk17-1 leaves was significantly reduced compared with that in the wild-type leaves (Fig. 2A). Furthermore, RT-quantitative PCR (RT-qPCR) assays showed that the expression levels of the immune marker genes including OsPR1a, OsPR5, and OsWRKY45 were significantly lower in the oscpk17-1 line than those in the wild-type plants after undergoing the flg22 and chitin treatments (Fig. 2B). Interestingly, chitin- or flg22-induced MAPK phosphorylation was not significantly altered between the wild-type and oscpk17-1 mutant lines (Supplemental Fig. S2). Altogether, these results indicate that OsCPK17 positively regulates PAMP-induced ROS burst and defense marker gene expression but does not affect MAPK activation in rice.

Figure 2.

OsCPK17 positively regulates ROS generation and immune marker gene expression in rice. A) Chemiluminescence assays to show ROS burst induced by PAMPs in the wild-type and oscpk17-1 rice leaves. The wild-type (DJ) and oscpk17-1 leaves were treated with flg22 (1 µM, left panel) or chitin (10 µg mL⁻¹, right panel). Chemiluminescence was detected at the indicated time points. B) The PAMP-induced expression levels of the defense marker genes OsPR5, OsPR1a, and OsWRKY45 in the wild-type and oscpk17-1 seedlings. OsACTIN expression was detected as an internal reference. Representative data from 3 independent assays are shown as mean ± Sd (n = 3). Asterisks indicate significant differences in the transcript levels of defense marker genes between the wild-type and oscpk17-1 rice seedlings (Student's t-test; *P < 0.05; **P < 0.01; ***P < 0.001).

OsCPK17 expression is rapidly induced by fungal and bacterial elicitors

Next, RT-qPCR was performed to detect the OsCPK17 transcript levels in rice seedlings with or without PAMP stimulation. OsCPK17 expression was rapidly induced by the bacterial elicitor flg22 and peaked at 3 h after treatment. In addition, OsCPK17 expression was significantly upregulated at 6 to 9 h after chitin treatment. Meanwhile, expression of the defense marker gene OsPAL1 was detected to confirm the effect of flg22 and chitin treatments (Supplemental Fig. S3, A and B). The PAMP-induced expression level of OsCPK17 culminated earlier than that of OsPAL1, suggesting that OsCPK17 is an early PAMP-responsive gene and is involved in plant immunity.

Furthermore, we investigated whether bacterial and fungal elicitors activate OsCPK17 phosphorylation. Immunoblot analyses demonstrated that OsCPK17-FLAG expressed in rice protoplasts exhibited a band shift at 10 to 15 min after the protoplasts were treated with flg22 (Supplemental Fig. S4, A and B). The band shift of OsCPK17-FLAG disappeared after dephosphorylation by calf intestinal alkaline phosphatase (Supplemental Fig. S4A). Meanwhile, a band migration was also observed for OsCPK17-FLAG at 4 to 10 min but was gradually vanished at 12 min after the protoplasts were treated with chitin (Supplemental Fig. S4C). The results indicate that OsCPK17 phosphorylation is activated in response to PAMP treatments.

OsCPK17 interacts with OsRLCK176 and positively regulates the stability of OsRLCK176

It has been reported that the RLCK VII subfamily member AtBIK1 is phosphorylated by AtCPK28 to maintain immune homeostasis in Arabidopsis (Wang et al. 2018a; Bredow et al. 2021). In rice, OsRLCK57, OsRLCK107, OsRLCK118, and OsRLCK176 are the closest homologs to AtBIK1 (Wang et al. 2018b). To explore how OsCPK17 mediates plant immunity, we investigated whether OsCPK17 interacts with these RLCK VII subfamily members. Luciferase complementation imaging (LCI) assays showed that a strong luminescence signal was recorded in the agro-infiltrated area where OsCPK17-NLuc was coexpressed with CLuc-OsRLCK176 in Nicotiana benthamiana leaves via Agrobacterium-mediated transient expression (Fig. 3A). By contrast, no luminescence was observed when OsCPK17-NLuc was individually coexpressed with CLuc-OsRLCK57, CLuc-OsRLCK107, and CLuc-OsRLCK118 or empty vector in N. benthamiana leaves although both NLuc- and CLuc-tagged proteins were well expressed in N. benthamiana leaves (Supplemental Fig. S5A).

Figure 3.

OsCPK17 interacts with OsRLCK176 and positively regulates the stability of OsRLCK176. A) LCI assays to detect the interaction between OsCPK17 and OsRLCK176 in N. benthamiana. Strong luminescence was observed in N. benthamiana leaves coexpressing OsCPK17-NLuc and CLuc-OsRLCK176. NLuc and CLuc, the N-terminal and C-terminal firefly LUC fragment, respectively. Expression of the NLuc- and CLuc-tagged proteins was detected with anti-FLAG (α-FLAG) and anti-CLuc (α-CLuc) antibodies, respectively. B) Co-IP assays to detect the interaction between OsCPK17 and OsRLCK176 in rice protoplasts. OsRLCK176-FLAG was transiently expressed alone or together with OsCPK17-GFP or GFP in rice protoplasts. Co-IP assay was performed with anti-FLAG M2 affinity beads, and the immuno-precipitated complex was detected with anti-GFP (α-GFP) and anti-FLAG (α-FLAG) antibodies. IP, immunoprecipitation. C) The abundance of OsRLCK176-HA and OsRLCK57-HA was detected by immunoblotting when these proteins were individually expressed or together with OsCPK17-FLAG in N. benthamiana after agroinfiltration. Protein loading is indicated by Ponceau S staining. D) The abundance of OsRLCK176-HA expressed in the wild-type and oscpk17-1 rice protoplasts. OsRLCK176-HA was transiently expressed in the wild-type and oscpk17-1 rice protoplasts followed by the treatments of the protein synthesis inhibitor CHX and/or the proteasome inhibitor MG132. GFP was transiently coexpressed as an internal control. E) Cell-free degradation assays to detect the degradation of GST-OsRLCK176 after its incubation with total protein extracts of the wild-type or oscpk17-1 seedlings. In vitro-purified GST-OsRLCK176 was incubated with total protein extracts in the presence of 10 mM ATP in cell-free degradation assay. Protein loading is indicated by Ponceau S staining. Different proteins were detected via immunoblotting with anti-HA (α-HA), anti-FLAG (α-FLAG), anti-GFP (α-GFP), and anti-GST (α-GST) antibodies. The data in the charts in C) to E) are shown as mean ± Sd from 3 independent experiments (n = 3). Relative intensity was quantified by densitometry using ImageJ and normalized to protein loading. Different letters (a to d) indicate significant differences in the protein level among different treatments according to one-way ANOVA followed by Duncan's multiple range tests (at α = 0.05).

Subsequently, coimmunoprecipitation (co-IP) assays were performed to validate the interaction between OsCPK17 and OsRLCK176. The recombinant OsCPK17-GFP and OsRLCK176-FLAG proteins were coexpressed in rice protoplasts. When OsRLCK176-FLAG was immunoprecipitated from total protein extracts of the transfected protoplasts using anti-FLAG M2 affinity beads, OsCPK17-GFP was detected in the immunocomplex by immunoblotting. In addition, the association of OsCPK17 with OsRLCK176 was promoted by flg22 treatment (Supplemental Fig. S5B). By contrast, no OsCPK17-GFP was detected in the immunoprecipitates when OsCPK17-GFP was transiently expressed alone or was coexpressed with OsRLCK57-FLAG, OsRLCK107-FLAG, or OsRLCK118-FLAG in rice protoplasts (Fig. 3B; Supplemental Fig. S5C). Collectively, these results demonstrated that OsCPK17 specifically interacts with OsRLCK176 in vivo.

To elucidate biological significance of the OsCPK17-OsRLCK176 interaction, we found that OsRLCK176-HA abundance was much higher when it was coexpressed with OsCPK17-FLAG than its expression alone in N. benthamiana leaves (Fig. 3C). By contrast, abundance of OsRLCK57-HA was unaltered when it was expressed with or without OsCPK17-FLAG. Furthermore, transiently expressed OsRLCK176-HA in the wild-type protoplasts was more stable than that in the oscpk17-1 protoplasts in the presence of the protein synthesis inhibitor cycloheximide (CHX) (Fig. 3D). More convincingly, the oscpk17-1 mutant exhibited a greatly reduced endogenous OsRLCK176 level compared with the wild-type plants as detected by immunoblotting with a commercially available anti-OsRLCK176 antibody (Supplemental Fig. S5D). Immunoblot analysis using this antibody detected a specific band with an apparent molecular weight of ∼55 kD in the cultivars Dongjin and Nipponbare, but not in the osrlck176 mutant (Supplemental Fig. S5E). Although OsRLCK176 abundance in the wild-type plants was evidently elevated after flg22 and chitin treatments, its accumulation was greatly reduced in the oscpk17-1 mutant even with PAMP stimulation (Supplemental Fig. S5D). Additionally, OsRLCK176 degradation was greatly inhibited when the transfected oscpk17-1 protoplasts were treated with the proteasome inhibitor MG132 (Fig. 3D). The results indicate that the degradation of OsRLCK176 depends on the ubiquitin proteasome system and is prevented by OsCPK17. Furthermore, cell-free degradation assays revealed that in vitro-purified GST-OsRLCK176 was more stable when it was incubated with the total protein extract from the wild-type seedlings than its incubation with the oscpk17-1 extract (Fig. 3E). Altogether, these results indicate that OsCPK17 positively regulates the homeostasis of OsRLCK176.

OsCPK17 stabilizes OsRLCK176 through phosphorylating OsRLCK176 at Ser83

To investigate whether OsRLCK176 and OsCPK17 are mutually phosphorylated, we expressed and purified maltose-binding protein (MBP)-tagged OsRLCK176 and OsRLCK176^KM, a kinase-inactive mutant in which the Lys108 and Lys109 residues were replaced with Ala (Wang et al. 2018b), and MBP-tagged OsCPK17 and OsCPK17^KM, a kinase-dead mutant in which the conserved Asp230 residue was mutated into Ala. In vitro kinase assays showed that MBP-OsRLCK176^KM was phosphorylated by MBP-OsCPK17, whereas MBP-OsRLCK176 did not phosphorylate MBP-OsCPK17^KM (Fig. 4A; Supplemental Fig. S6A). Next, we compared the phosphorylation level of OsRLCK176 in the wild-type and oscpk17-1 mutant lines. OsRLCK176-HA was transiently expressed in the wild-type Dongjin and oscpk17-1 protoplasts. Immunoblot analyses showed that an evident band shift was observed for OsRLCK176-HA after the transfected wild-type protoplasts were treated with flg22, whereas the band shift of OsRLCK176-HA almost disappeared in the oscpk17-1 protoplasts (Fig. 4B). The results indicate that OsRLCK176 is strongly phosphorylated by OsCPK17 in vivo after flg22 treatment. Consistently, OsRLCK176 abundance is much lower in the oscpk17-1 protoplasts than that in the wild-type protoplasts regardless of flg22 treatment (Fig. 4B).

Figure 4.

OsCPK17 enhances OsRLCK176 stability through phosphorylating OsRLCK176 at Ser83. A) In vitro kinase assay to show that kinase-defective MBP-OsRLCK176^KM was phosphorylated by MBP-OsCPK17. Protein phosphorylation was detected by autoradiography (upper panel), and protein loading is indicated by Coomassie Brilliant Blue (CBB) staining (lower panel). B) The PAMP-induced phosphorylation levels of OsRLCK176 in the wild-type and oscpk17-1 protoplasts. OsRLCK176-HA was transiently expressed in the wild-type and oscpk17-1 protoplasts. The transfected protoplasts were treated with flg22 for 30 min or 1 h before protein extraction. GFP was transiently coexpressed as an internal control. Total protein extracts were detected by immunoblotting with anti-HA (α-HA) and anti-GFP (α-GFP) antibodies. DJ, Dongjin. C) In vitro kinase assays to show the phosphorylation levels of different OsRLCK176^KM variants when these mutant proteins were incubated with His-OsCPK17. Protein phosphorylation was detected by autoradiography (upper panel), and protein loading is indicated by CBB staining (lower panel). D) The stability of OsRLCK176-HA and OsRLCK176^S83D-HA expressed in the oscpk17-1 rice protoplasts. Abundance of OsRLCK176-HA and OsRLCK176^S83D-HA was detected in rice protoplasts after CHX treatment for 2 h. GFP was transiently coexpressed as an internal reference in rice protoplasts. As a control, OsRLCK176-HA was expressed in the wild-type protoplasts. Different proteins were detected via immunoblotting with anti-HA (α-HA) and anti-GFP (α-GFP) antibodies. E) The phosphorylation levels of OsRLCK176 at Ser83 in the wild-type (DJ), oscpk17-1 and osrlck176-OsRLCK176^S83A lines after flg22 treatment. The seedlings of different rice lines were treated with flg22 for 15 min and were then collected for protein extraction. OsRLCK176 phosphorylation at Ser83 was detected with an anti-pSer83 antibody. Protein loading is indicated by immunoblotting with an anti-β-actin antibody.

To identify the residues that are phosphorylated by OsCPK17 in OsRLCK176, the tryptic peptides of MBP-OsRLCK176^KM were analyzed by liquid chromatography-tandem MS after MBP-OsRLCK176^KM was incubated with MBP-OsCPK17 in an in vitro phosphorylation assay. Four Ser residues including Ser32, Ser35, Ser83, and Ser209 in OsRLCK176 were identified to be phosphorylation sites (Supplemental Fig. S6B). Subsequently, His-tagged OsRLCK176^KM(S32A), OsRLCK176^KM(S35A), OsRLCK176^KM(S83A), and OsRLCK176^KM(S209A) mutant proteins were purified for in vitro kinase assays. Compared with OsRLCK176^KM and OsRLCK176^KM(S35A), OsRLCK176^KM(S32A), OsRLCK176^KM(S83A), and OsRLCK176^KM(S209A) exhibited reduced phosphorylation signals, implying that Ser32, Ser83, and Ser209 in OsRLCK176 are the major residues phosphorylated by OsCPK17 (Fig. 4C).

Because OsCPK17 stabilizes and phosphorylates OsRLCK176, we speculate that OsCPK17-mediated OsRLCK176 phosphorylation regulates the homeostasis of OsRLCK176. To confirm this hypothesis, the phosphomimetic mutants OsRLCK176^S32D-HA, OsRLCK176^S83D-HA, and OsRLCK176^S209D-HA were transiently expressed in the wild-type and oscpk17-1 protoplasts. Total proteins extracted from the CHX-treated transfected protoplasts were detected via immunoblotting. Consistently, OsRLCK176-HA in the oscpk17-1 protoplasts is prone to be degraded compared with that in the wild-type protoplasts. We also found that OsRLCK176^S83D-HA was more stable than OsRLCK176-HA when these proteins were individually expressed in the oscpk17-1 protoplasts, indicating that the S83D phosphomimetic mutation enhances the stability of OsRLCK176-HA (Fig. 4D). In addition, OsRLCK176^S83A was less stable than OsRLCK176 and OsRLCK176^S83D when these proteins were transiently expressed in the wild-type protoplasts (Supplemental Fig. S7A). By contrast, the degradation rates of OsRLCK176^S32D-HA and OsRLCK176^S209D-HA were similar to that of OsRLCK176-HA in the oscpk17-1 protoplasts (Supplemental Fig. S7A).

To further confirm that Ser83 in OsRLCK176 is an important phosphorylated residue mediated by OsCPK17, a site-specific antibody against OsRLCK176^pSer83 was developed. Immunoblot analysis showed that Ser83 phosphorylation in OsRLCK176 was only detected in the wild-type plants but not in the oscpk17-1 mutant. Besides, the phosphorylation level of OsRLCK176 at Ser83 was increased by flg22 (Fig. 4E). However, co-IP assays showed that the association of OsRLCK176 with OsCPK17 was not altered by the S83D mutation (Supplemental Fig. S7B). The results indicate that OsCPK17-mediated phosphorylation of OsRLCK176 at Ser83 promotes the stability of OsRLCK176.

Ser83 phosphorylation in OsRLCK176 contributes to bacterial disease resistance in rice

To determine the importance of Ser83 phosphorylation for OsRLCK176 immune function, we generated the transgenic plants to express OsRLCK176 and OsRLCK176^S83D in the osrlck176 mutant. After being challenged with Xoo and Xoc, the OsRLCK176^S83D-expressing transgenic plants exhibited significantly elevated resistance to bacterial blight and bacterial leaf streak than did the OsRLCK176-expressing transgenic plants (Fig. 5, A and B; Supplemental Fig. S7C). Besides, OsPR1a was transcriptionally upregulated in both osrlck176-OsRLCK176 and osrlck176-OsRLCK176^S83D lines after Xoc and Xoo infection. Importantly, the induced expression level of OsPR1a in the osrlck176-OsRLCK176^S83D line was significantly higher than that in the osrlck176-OsRLCK176 plants after Xoc and Xoo infection (Supplemental Fig. S7D). To investigate whether OsCPK17 performs its immune function through phosphorylating OsRLCK176, we also generated the transgenic plants to express OsRLCK176^S83D in the oscpk17-1 mutant. Bacterial inoculation assays showed that the expression of OsRLCK176^S83D conferred the oscpk17-1 mutant an enhanced resistance to bacterial blight and bacterial leaf streak, indicating that the phosphomimetic variant OsRLCK176^S83D restores bacterial disease resistance in rice even at the absence of OsCPK17 (Fig. 5, C and D; Supplemental Fig. S7C). Similarly, OsPR1a was significantly upregulated in the oscpk17-1 and oscpk17-1-OsRLCK176^S83D lines after bacterial infection. The induced expression level of OsPR1a in the oscpk17-1-OsRLCK176^S83D line is also significantly elevated compared with that in the oscpk17-1 plant after Xoc and Xoo infection (Supplemental Fig. S7E). Phylogenetic and sequence alignment analyses revealed that Ser83 is conserved in rice homologs of AtBIK1 including OsRLCK176, OsRLCK57, OsRLCK107, and OsRLCK118 but is absent from RLCK VII subfamily members in other plant species (Supplemental Fig. S7F and File S1). These results indicate that OsRLCK176 phosphorylation at Ser83 mediated by OsCPK17 enhances its immune function in rice.

Figure 5.

OsRLCK176 phosphorylation at Ser83 mediated by OsCPK17 contributes to bacterial disease resistance in rice. A, B) The representative disease symptoms (left panels) and the disease lesion length (right panels) on the inoculated leaves of the osrlck176-OsRLCK176 and osrlck176-OsRLCK176^S83D transgenic lines after XocA) and XooB) infection. C, D) The representative disease symptoms (left panels) and the disease lesion length (right panels) on the inoculated leaves of the oscpk17-1 and oscpk17-1-OsRLCK176^S83D transgenic lines after XocC) and XooD) infection. The images were captured and the lengths of disease lesions were measured at 10 dpi. Representative data from 3 independent experiments are shown as mean ± Sd (n = 9 in A, B; n = 10 in C, D). Asterisks indicate significant differences in the lesion length on the inoculated leaves between different transgenic lines (Student's t-test; ***P < 0.001).

OsRLCK176 is a substrate of OsPUB12

To explore the degradation mechanism of OsRLCK176, we constructed a rice PUB gene mini-library and performed a yeast two-hybrid screen. The variant of a PUB E3 ligase OsPUB12^C234Y with a point mutation in the putative catalytic site was revealed to interact with OsRLCK176 (Supplemental Fig. S8A). Subsequently, co-IP assays demonstrated that OsPUB12 interacted with OsRLCK176, but not with OsRLCK57, OsRLCK107, or OsRLCK118 (Fig. 6A; Supplemental Fig. S8B). However, co-IP assays showed that flg22 or chitin treatment did not alter the interaction between OsRLCK176 and OsPUB12 (Supplemental Fig. S8C). In GST pull-down assays, in vitro-purified GST-OsRLCK176 and GST proteins were individually incubated with His-OsPUB12 in the presence of GST beads. We demonstrated that His-OsPUB12 was copulled down by GST-OsRLCK176, but not by GST (Fig. 6B). Altogether, these results indicate that OsPUB12 specifically interacts with OsRLCK176.

Figure 6.

OsRLCK176 phosphorylation at Ser83 inhibits its ubiquitination by the plant U-box E3 ligase OsPUB12. A) Co-IP assay to detect the interaction between OsRLCK176 and OsPUB12 in rice protoplasts. OsPUB12-HA and OsRLCK176-FLAG were individually expressed or were coexpressed in rice protoplasts. Total proteins extracted from transfected protoplasts were incubated with anti-FLAG M2 affinity beads. The input proteins and immunoprecipitates were detected with anti-FLAG (α-FLAG) and anti-HA (α-HA) antibodies. B) In vitro GST pull-down assay to detect the interaction between OsRLCK176 and OsPUB12. His-OsPUB12 was incubated with GST-OsRLCK176 or GST in the presence of glutathione agarose beads. The input and pull-down proteins were detected with anti-His (α-His) and anti-GST (α-GST) antibodies. C) The abundance of OsRLCK176-FLAG and OsRLCK176^S83D-FLAG was detected through immunoblotting when these proteins were transiently coexpressed with OsPUB12-HA or with OsPUB12^C234Y-HA in rice protoplasts. Protein abundance was detected in rice protoplasts at 2 h after CHX treatment. GFP was transiently expressed in rice protoplasts as an internal reference. D) In vitro ubiquitination assay to show the ubiquitination levels of OsRLCK176 and OsRLCK176^S83D mediated by OsPUB12. Recombinant proteins purified from E. coli were incubated with E1, E2, and ubiquitin at 30 °C for 4 h in an in vitro ubiquitination assay. The ubiquitination of OsRLCK176 was detected by immunoblotting with an anti-MBP antibody (α-MBP). Total protein ubiquitination was detected by immunoblotting with anti-HA antibody (α-HA). E) In vivo ubiquitination assay to show the ubiquitination levels of OsRLCK176 and OsRLCK176^S83D after coexpression with OsPUB12 in rice protoplasts. Ubiquitin-HA was transiently coexpressed in rice protoplasts with OsRLCK176-FLAG/OsRLCK176^S83D-FLAG and OsPUB12-HA/OsPUB12^C234Y-HA as indicated in the presence of 5 µM MG132. Total proteins were extracted and were then subject to immunoprecipitation with anti-FLAG M2 affinity beads. The poly-ubiquitination of OsRLCK176-FLAG was detected by immunoblotting with an anti-FLAG (α-FLAG) antibody. OsPUB12 and OsPUB12^C234Y were detected by immunoblotting with an anti-HA antibody (α-HA). β-Actin was detected with an anti-β-Actin antibody (α-β-actin) as an internal control.

Furthermore, a phylogenetic tree was constructed to reveal the phylogenetic relationship among some OsPUB12-related PUB-type E3 ligases in rice and Arabidopsis. The tree showed that OsPUB12 is the most closely related to OsPUB11 but has a lower similarity with AtPUB25 and AtPUB26, which form a separate clade with OsPUB38 and OsPUB41-44 (Supplemental Fig. S8D and File S1). In addition, LCI assays demonstrated that OsRLCK176 did not interact with OsPUB41 (Supplemental Fig. S8E). To investigate whether OsRLCK176 is a substrate of OsPUB12, OsRLCK176-FLAG was expressed alone or was coexpressed with OsPUB12-HA and OsPUB12^C234Y-HA in rice protoplasts. Total proteins were isolated and subject to immunoblot analyses after the transfected protoplasts were incubated with CHX for 2 h. The results showed that OsRLCK176-FLAG was rapidly degraded after CHX treatment when it was coexpressed with OsPUB12, whereas the protein remained relatively stable when it was expressed alone or was coexpressed with OsPUB12^C234Y (Fig. 6C). Interestingly, the abundance of OsRLCK176^S83D-FLAG was almost unaltered when the mutant protein was coexpressed with OsPUB12 in rice protoplasts after undergoing the CHX treatment (Fig. 6C). Consistently, immunoblot analysis showed the endogenous level of OsRLCK176 in the ospub12-1 mutant was much higher than that in the wild-type plant in the absence of elicitors. Interestingly, OsRLCK176 abundance in the wild-type plants was elevated to a similar level in the ospub12-1 mutant after flg22 treatment (Supplemental Fig. S8F).

Next, in vitro ubiquitination assays were performed to investigate whether OsRLCK176 is ubiquitinated by OsPUB12. In the presence of E1, E2, His-OsPUB12, and ubiquitin-HA, auto-ubiquitination of His-OsPUB12 was detected, indicating that OsPUB12 has an E3 ubiquitin ligase activity. Furthermore, we found that OsRLCK176 was ubiquitinated by OsPUB12 when His-OsPUB12 was incubated with MBP-OsRLCK176 in an in vitro ubiquitination assay. Interestingly, OsRLCK176^S83A was ubiquitinated by OsPUB12 as strongly as OsRLCK176, whereas OsRLCK176^S83D ubiquitination by OsPUB12 was much weaker than the wild-type protein (Fig. 6D). These results indicate that OsRLCK176 is a direct degradation target of OsPUB12. Furthermore, we tested whether OsRLCK176 is ubiquitinated by OsPUB12 in vivo. Ubiquitin-HA and OsRLCK176-FLAG were transiently coexpressed with OsPUB12-HA or OsPUB12^C234Y-HA in rice protoplasts in the presence of MG132. After immunoprecipitation with anti-FLAG affinity beads, an anti-FLAG antibody was used to detect poly-ubiquitinated OsRLCK176. We found that OsRLCK176 ubiquitination was dramatically increased after OsPUB12-HA coexpression but was almost unaltered after OsPUB12^C234Y-HA coexpression. In addition, OsRLCK176^S83D was much more weakly ubiquitinated by OsPUB12 than OsRLCK176 (Fig. 6E). These results demonstrated that OsRLCK176 was targeted by OsPUB12 for poly-ubiquitination, but the S83D mutation in OsRLCK176 prevented its ubiquitination by OsPUB12.

OsPUB12 negatively regulates rice immunity

To determine the role of OsPUB12 in plant immunity, we generated two ospub12 mutant lines through the CRISPR/Cas9 system. In the ospub12-1 homozygous line, a base pair was inserted after the nucleotide 666 in OsPUB12 and therefore generated a premature stop codon. In the ospub12-2 line, a 2-bp deletion also caused a premature stop codon in OsPUB12 (Fig. 7A). Subsequently, PAMP-induced defense marker gene expression and ROS production were examined in the wild-type and ospub12 mutant seedlings. Compared with the wild-type seedlings, both of ospub12 mutant lines exhibited significantly elevated expression of OsPR1a, OsPR10a, and OsWRKY45 after the flg22 and chitin treatments (Fig. 7B). The flg22- and chitin-induced ROS burst was much higher in the ospub12 mutant lines than that in the wild-type plants (Fig. 7C). Furthermore, the ospub12 mutant lines exhibited significantly shorter disease lesions on the Xoo- and Xoc-inoculated leaves than did the wild-type plants (Fig. 7, D and E). In contrast, flg22- and chitin-triggered MAPK activation was not significantly changed between the wild-type and ospub12-1 mutant lines (Supplemental Fig. S9). The results indicate that OsPUB12 negatively regulates rice immunity.

Figure 7.

OsPUB12 negatively regulates rice immunity against bacterial diseases. A) A schematic diagram to show a base pair (bp) insertion and a 2-bp deletion in ospub12-1 and ospub12-2 knockout lines, respectively. Both insertion and deletion result in reading frameshifts and therefore premature stop codons in ospub12-1 and ospub12-2 lines. The premature stop codons are highlighted in the red boxes. B) The transcript levels of the defense marker genes OsPR1a, OsPR10a, and OsWRKY45 in the wild-type and ospub12 seedlings undergoing the treatments with flg22 and chitin. OsACTIN expression was detected as an internal reference. Asterisks indicate significant differences in the transcript levels of defense marker genes between the wild-type and ospub12 rice seedlings (Student's t-test; * P < 0.05; **P < 0.01; ***P < 0.001). C) ROS burst induced by PAMPs in the wild-type (NIP) and ospub12 leaves. The NIP, ospub12-1, and ospub12-2 leaves were treated with flg22 (1 µM, left panel) or chitin (10 µg/mL, right panel). D, E) The representative disease symptoms (left panels) and the disease lesion length (right panels) on the inoculated leaves of the wild-type (NIP), ospub12-1 and ospub12-2 lines caused by Xoc RS105 D) and Xoo PXO99 E) infection. The images were captured, and the lengths of disease lesions were measured at 10 dpi. Representative data from 3 independent experiments are shown as mean ± Sd (n = 10). Asterisks indicate significant differences in the disease lesion length between the wild-type and ospub12 leaves (Student's t-test; **P < 0.01; ***P < 0.001).

Discussion

In plants, CDPKs not only play important roles in tolerance to abiotic stresses but also help fight against pathogen attacks (Hyodo et al. 2017). In this study, we demonstrated that OsCPK17 is an important immune regulator. OsCPK17 interacts with and phosphorylates OsRLCK176 at Ser83 and thereby promotes OsRLCK176 stability. As an E3 ubiquitin ligase, OsPUB12 ubiquitinates and degrades OsRLCK176. Therefore, the OsCPK17-OsPUB12-OsRLCK176 module delicately regulates rice immunity through maintaining the homeostasis of OsRLCK176.

Distinct CDPKs in plants regulate innate immunity in different ways. OsCPK4, OsCPK18, and AtCPK28 in the CDPK IV subgroup all function as negative regulators of plant immunity (Monaghan et al. 2014; Xie et al. 2014; Wang et al. 2018a). Here, we demonstrated that OsCPK17 positively regulates immune responses and resistance against bacterial diseases in rice. First, OsCPK17 is transcriptionally induced, and its encoded kinase is rapidly activated upon PAMP stimulation (Supplemental Figs. S3 and S4). Second, PAMP-induced ROS burst and PR gene expression are significantly attenuated in OsCPK17-knockout rice plants (Fig. 2, A and B). More convincingly, inoculation assays demonstrated that the oscpk17-1 mutant and OsCPK17 RNAi lines are more susceptible to bacterial blight and leaf streak diseases than the wild-type plants (Fig. 1; Supplemental Fig. S1). Similarly, AtCPK5 in Arabidopsis, a close homolog of OsCPK17 in the CDPK I subgroup, plays an important role in plant immunity through phosphorylating AtBIK1 and AtRBOHD (Dubiella et al. 2013). In addition, the members in the Arabidopsis CDPK I subgroup, including AtCPK4, AtCPK5, AtCPK6, and AtCPK11, are activated in response to flg22 and are thought to be early transcriptional regulators of MAMP signaling (Boudsocq et al. 2010). AtCPK1 overexpression leads to the accumulation of salicylic acid (SA) and further induces expression of SA-regulated defense genes, thus providing broad-spectrum protection against fungal and bacterial pathogens (Coca and San 2010; Singh et al. 2017). Our results showed that OsCPK17 is essential for induced expression of SA-dependent immune marker genes such as OsPR1a and OsPR5, suggesting that OsCPK17 might also be involved in SA-dependent immunity (Fig. 2B). Previous studies revealed that OsCPK17 is downregulated by abiotic stresses like cold, drought, and salt stresses (Wan et al. 2007; Almadanim et al. 2017). Upon cold stress, OsCPK17 phosphorylates sucrose phosphate synthases, the rate-limiting enzymes for sucrose synthesis, thus leading to a decrease in enzymatic activity. Xoo targets sucrose transporters through the type III effectors and the deficiency of sucrose transporters restricts a broad range of pathogens (Oliva et al. 2019). Therefore, it will be interesting to determine whether and how OsCPK17 is involved in the cross talk between biotic and abiotic stress signaling.

MAPK cascades are involved in regulating ROS burst, stomatal development, hormone signaling, disease resistance, and abiotic stress tolerance (Ichimura et al. 2002; Pitzschke et al. 2009; Meng and Zhang 2013; Xie et al. 2014). Rice OsCPK18, a homolog of OsCPK4 in the CDPK IV subgroup, initiates a downstream negative regulator OsMPK5 (Monaghan et al. 2014; Xie et al. 2014). The circuits between CDPKs and MAPKs, as in the case of OsCPK18 and OsMPK5, may be the common mechanisms in regulating cross talk between different cellular signaling pathways (Li et al. 2022). We demonstrated here that PAMP-triggered MAPK phosphorylation is not significantly altered in the wild-type and oscpk17-1 seedlings (Supplemental Fig. S2), suggesting that other CPKs might function redundantly with OsCPK17 in activation of MAPK pathways. Otherwise, OsCPK17 may function in rice immunity in a MAPK-independent manner.

RLCKs, including AtBIK1 and PBS1-like kinases, are converging substrates of multiple PRR-mediated signaling pathways in A. thaliana and O. sativa (Bi et al. 2018). Upon PAMP stimulation, PRRs or PRR complexes phosphorylate and activate RLCKs (Lu et al. 2010; Liu et al. 2013), allowing subsequent phosphorylation of the substrates, such as the transmembrane NADPH-oxidase RBOHD (Kadota et al. 2014; Li et al. 2014). OsRLCK176, one of the closest homologs of AtBIK1, is an essential downstream component of OsCERK1-mediated signaling induced by peptidoglycan and chitin (Ao et al. 2014). OsRLCK176 homeostasis plays an important role in fine-tuning rice immunity. In our previous study, we showed that OsCPK4 promotes OsRLCK176 degradation at the resting phase (Wang et al. 2018b). In contrast, we revealed here that OsCPK17 enhances OsRLCK176 stability, especially in the presence of elicitors (Fig. 3, C to E; Supplemental Fig. S5D).

Although OsCPK4 and OsCPK17 function as negative and positive immune regulators, respectively, both CPKs can phosphorylate OsRLCK176 (Wang et al. 2018b) (Fig. 4A). OsCPK4 phosphorylates OsRLCK176 at Ser74, Ser209, Ser267, and Ser336 residues (Wang et al. 2018b), whereas OsCPK17 phosphorylates OsRLCK176 at Ser32, Ser83, and Ser209. Besides, in vivo phosphorylation of OsRLCK176 at Ser83 is induced by flg22 and is completely dependent on OsCPK17 (Fig. 4E). Interestingly, the phosphorylation of Ser83 specifically promotes the stability of OsRLCK176 (Fig. 4). An increasing number of studies showed that the phosphorylation is involved in protein ubiquitination and degradation (Wang et al. 2018a; Bredow et al. 2021; Zheng et al. 2022; Liu et al. 2023). For instance, AtPUB4 in Arabidopsis promotes the ubiquitination and degradation of nonactivated AtBIK1, whereas AtPUB4 contributes to the accumulation of PAMP-activated AtBIK1 (Yu et al. 2022). Nonactivated AtBIK1 is also targeted for proteasomal degradation by the E3 ubiquitin ligases AtPUB25 and AtPUB26. The activity of both PUB E3 ligases is promoted by AtCPK28-mediated phosphorylation, whereas heterotrimeric G proteins directly inhibit AtPUB25/26 E3 ligase activities and negatively regulate AtPUB25/26-mediated ubiquitination of AtBIK1 (Wang et al. 2018a). It is interesting to note that AtPUB22 is phosphorylated at Thr62 and Thr88 residues by AtMPK3, which inhibits AtPUB22 auto-ubiquitination and degradation, thus suppressing immune responses (Furlan et al. 2017).

Although 77 putative U-box E3 ligases are encoded in rice, very few of them have been functionally characterized (Kim et al. 2021). Rice SPL11 (OsPUB11) has been shown to negatively regulate innate immunity (Zeng et al. 2004; Liu et al. 2015). In contrast, OsPUB15 interacts with the receptor-like kinase OsPID2 to positively regulate cell death and defenses against fungal blast (Wang et al. 2015). In this study, we demonstrated that the E3 ubiquitin ligase OsPUB12 interacts with OsRLCK176 in rice (Fig. 6, A and B). OsPUB12 ubiquitinates OsRLCK176 and promotes its degradation. OsPUB12 functions as a negative regulator of rice immunity because the ospub12 mutant lines exhibited enhanced PAMP-induced defense responses and elevated resistance to bacterial diseases (Fig. 7). Upon PAMP stimulation, OsCPK17 is activated and in turn phosphorylates OsRLCK176 at Ser83. The phosphorylation of Ser83 in OsRLCK176 prevents its ubiquitination by OsPUB12 and subsequent degradation (Fig. 8). More importantly, expression of the phosphomimetic variant OsRLCK176^S83D confers enhanced bacterial disease resistance in the osrlck176 mutant compared with OsRLCK176 expression, indicating that phosphorylation of OsRLCK176 at Ser83 enhances its immune function. Furthermore, OsRLCK176^S83D expression can restore bacterial disease resistance of the oscpk17-1 mutant, indicating that the S83D phosphomimetic mutant circumvents OsCPK17 action to enhance rice immunity (Fig. 5). This proposed model illustrates a new molecular mechanism how rice plants maintain immune homeostasis through fine-tuning OsRLCK176 abundance. Interestingly, protein sequence alignment revealed that the phosphorylation residue Ser83 in OsRLCK176 is conserved in rice RLCK VII subfamily members but is absent in Arabidopsis AtBIK1 and homologs (Supplemental Fig. S7F). In addition, AtBIK1 is targeted by AtPUB25/26 for degradation. AtPUB25/26 and OsPUB12 are separated in distinct clades in the phylogenetic tree of PUB homologs (Supplemental Fig. S8D). Altogether, these findings suggest that the homeostasis of RLCK VII subfamily members in rice and Arabidopsis might be regulated by different mechanisms.

Figure 8.

A working model for the OsCPK17-OsPUB12-OsRLCK176 module in regulating immune homeostasis. Without pathogen attacks, OsRLCK176 is degraded after ubiquitination by OsPUB12, and therefore rice immunity is kept in a relatively low level. Upon pathogen infection and PAMP stimulation, OsCPK17 is rapidly activated and thereby phosphorylates OsRLCK176 at Ser83. This phosphorylation greatly attenuates OsRLCK176 ubiquitination and degradation, which enhances rice immunity. Created with BioRender (www.BioRender.com).

Both CDPKs and RLCKs are important immune components. The OsCPK17-OsPUB12-OsRLCK176 module plays an important role in immune homeostasis. We uncover a module regulating rice immunity, which facilitates the development of molecular strategy to create genetic germplasms with disease resistance.

Materials and methods

Unless noted, all experiments were repeated at least 3 times with similar results. All gene constructs were confirmed by sequencing.

Plant materials and growth conditions

The T-DNA insertion line 4A-50582 generated from O. sativa ssp. japonica cv. Dongjin (DJ) was obtained from the POSTECH rice insertion mutant library (Korea). Six-week-old rice plants grown in the greenhouse were used for inoculation assays with X. oryzae pvs. oryzae and oryzicola. N. benthamiana plants were grown under a 16-h/8-h photoperiod at 25 °C in environmentally controlled growth chambers with the white fluorescent light intensity of 200 µmol m⁻² s⁻¹. Agrobacterium strains were cultured at 28 °C in Luria Bertani (LB) medium (10-g tryptone, 10-g NaCl, and 5-g yeast extract per liter) containing 25 µg mL⁻¹ rifampicin and 50 µg mL⁻¹ kanamycin.

Generation of miRNA-based OsCPK17-knockdown transgenic rice lines

The artificial microRNA (amiRNA) oligo sequences used for OsCPK17 silencing were designed using a customized miRNA design tool Web MicroRNA Designer (WMD) platform (http://wmd3.weigelworld.org) as previously described (Warthmann et al. 2008). The primary amiRNA construct was engineered from pNW55 by replacing the natural osa-MIR528 miRNA by amiRNA sequences for OsCPK17 with overlapping PCR using the primers I to IV, G-4368 and G-4369 (Supplemental Data Set S1) as described (Warthmann et al. 2008). The fusion product was subcloned into the multiple cloning site of the binary vector pUN1301 for subsequent Agrobacterium-mediated transformation.

Generation of the ospub12 mutant lines via the CRISPR/Cas9 system

Two ospub12 knockout lines were generated in Nipponbare using the CRISPR/Cas9 system as previously published (Miao et al. 2013). Briefly, 2 pairs of guide RNA primers listed in Supplemental Data Set S1 were designed to specifically target OsPUB12 according to the instructions on the website (http://crispr.hzau.edu.cn/CRISPR2/). The designed primers were annealed and inserted into the Bsa I-digested entry vector pOS-sgRNA. The single guide RNAs were then recombined into the destination expression vector pOS-Cas9 using LR Clonase II mix (Invitrogen). The constructs were transformed into rice calli, and the gene knockout lines were identified by sequencing.

Identification of the oscpk17-1 mutant line

Homozygous oscpk17-1 mutant plants were identified by PCR using genomic DNA as templates. OsCPK17 expression was detected by RT-PCR with the primer set OsCPK17-RT-F/OsCPK17-RT-R and RT-qPCR with the primer set OsCPK17-qRT-F/OsCPK17-qRT-R (Supplemental Data Set S1).

Bacterial inoculation assay

Xoo inoculation was performed by the leaf clipping method (Kauffman et al. 1973). Briefly, the Xoo strain PXO99 was cultured in nutrient broth overnight and was then resuspended with 10 mM MgCl₂ to an OD₆₀₀ of 0.8 for inoculation. The inoculated plants were kept under high humidity. The length of disease lesions on the inoculated leaves was evaluated at 12 to 14 dpi. To determine bacterial population sizes in the inoculated leaves, leaf portions (1 cm wide) around the edge of lesions were harvested at 0, 3, 6, 9, and 12 dpi, and were immediately sliced into small pieces. Leaf slices were immersed in a cefalexin solution (18 µg mL⁻¹, 5 mL) for 1 h with shaking. After the leachate was filtered through 2 layers of sterile gauze, the filtrate was serially diluted and was then cultured at 28 °C on nutrient agar plates containing cephalexin. Colonies formed on the plates were counted after 3 d of culturing.

Xoc inoculation was performed by pressure infiltration (Wang et al. 2007). The Xoc strain RS105 was cultured in NB medium for 24 h at 28 C. Bacterial cells were collected and were then adjusted to OD₆₀₀ = 0.3 with sterilized water. Xoc suspension was pressure-inoculated into 6-wk-old plant leaves with a needleless syringe. The length of disease lesions on the inoculated leaves was measured at 10 to 14 dpi.

Oxidative burst assay

PAMP-triggered oxidative burst in rice leaves of the wild-type and different mutant lines was detected as described previously (Shi et al. 2018b). Leaf punches collected from the second leaves of 4- to 6-wk-old plants were immersed in ddH₂O overnight and were then transferred into 100 µL of luminol (Bio-Rad Immun-Star horseradish peroxidase (HRP) substrate 170-5040) containing 1.0 µL of 1 µg µL⁻¹ HRP (Jackson ImmunoResearch), and elicitor (1 µM flg22, 8 µM hexaacetylchitohexaose, or water as control). Chemiluminescence signal was immediately recorded by a Glomax 20/20 Luminometer over 30 min at 1-min intervals for graphical analysis (3 to 5 technical replicates).

Reverse transcription quantitative PCR

Total RNAs were isolated from PAMP-treated rice leaves using an Ultrapure RNA Isolation Kit according to the manufacturer's instructions (CWBIO, Beijing). RT-qPCR was performed in a 20-µL reaction mixture consisting of 0.35 µL of cDNA, 0.5 µL of each gene-specific primer (10 mM), and 10 µL of 2× Maxima SYBR Green qPCR Master Mix using an ABI PRISM 7500 Sequence Detection System (Applied Biosystems, Carlsbad, CA) as described previously (Li et al. 2015). The transcript levels of defense marker genes were normalized to that of the reference gene OsACTIN (Os03g0718100). The primers used to detect the expression of OsPR5 (Os04g0689900), OsPR1a (Os07g0129200), OsPR10a (Os12g0555500), OsPAL1 (Os02g0627100), OsWRKY45 (Os05g0322900), and OsACTIN are listed in Supplemental Data Set S1.

Detection of MAPK activation

MAPK activation was detected as previously described (Zhang et al. 2020). Briefly, 7-d-old seedlings were treated with flg22 (1 µM) or chitin (10 µg mL⁻¹) solution supplemented with 0.01% Silwet L-77 (GE Healthcare, Amersham, United Kingdom) for 15 min. Total protein extracts were electrophoresed on 10% SDS-polyacrylamide gels and were then blotted onto PVDF membrane. MAPK phosphorylation was detected by immunoblotting using a phospho-specific p44/42 MAPK antibody (Cat#9101, 1:2,000 dilution, Cell Signaling Technologies, Danvers, MA).

LCI assay

LCI assays were performed as described previously (Chen et al. 2008). The coding sequences (CDSs) of OsRLCKs and OsCPK17 were amplified and subcloned into the binary vectors pCAMBIA1300-35S:CLuc and pCAMBIA1300-35S:NLuc, respectively. These constructs were transformed into the A. tumefaciens strain EHA105 through the freeze–thaw method (Weigel and Glazebrook 2006). Overnight-cultured Agrobacterium cells were collected and resuspended in induction medium (10 mM MES, pH 5.6, 10 mM MgCl₂, and 150 µM acetosyringone) to a final concentration of OD₆₀₀ = 0.5. After incubation for 2 h or more, Agrobacterium strains carrying different gene constructs were pressure-infiltrated into ∼4-wk-old N. benthamiana plant leaves. Luminescence signals were captured using a cooled low-light charge-coupled device imaging apparatus (Berthold LB985) immediately after the infiltrated leaves were sprayed with luciferin (1 mM) at 2 d post-agroinfiltration.

Co-IP assay

Co-IP assays were performed as previously described (Li et al. 2015). Briefly, the ORF of OsCPK17 was subcloned into pUC19-35S:FLAG-RBS. The ORFs of OsRLCKs were amplified and ligated into the expression vector pHBT-HA. These gene constructs were transfected into the protoplasts isolated from rice seedlings (Li et al. 2015). Total proteins were extracted from the transfected protoplasts using IP buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100) supplemented with 1× protease inhibitor cocktail (CWBIO). The protein extracts (1 mL) were incubated with 25 µL of anti-FLAG M2 affinity beads (Sigma-Aldrich, A2220) at 4 °C for 4 h. After being rinsed thoroughly with ice-cold phosphate-buffered saline (PBS) buffer, the immunoprecipitates on the beads were eluted by SDS-PAGE loading buffer and were subsequently subject to immunoblotting using HRP-conjugated anti-HA (Roche, 11667475001, 1:2,000 dilution) and anti-FLAG (Sigma-Aldrich, F1804, 1:5,000 dilution) antibodies.

Expression and purification of recombinant proteins

The CDSs of OsCPK17 and OsRLCK176^KM were subcloned into the modified pET28a, in which the MBP-coding sequence is in frame fused with a His6-coding sequence. The CDS of OsRLCK176 was subcloned into pGEX-6P-1 to express glutathione-S-transferase (GST)-tagged protein for pull-down assays. Tagged proteins were expressed in the Escherichia coli strain BL21 (DE3) after induction by 0.1 mM IPTG and were then purified with Ni-NTA His-bind Resin (Millipore Corp, Billerica, MA) or Glutathione Sepharose beads (GE Healthcare, Piscataway, NJ) according to the manufacturers' instructions. The purified proteins were quantified with a BCA protein assay kit (Thermo Fisher Scientific, Waltham, MA).

In vitro GST pull-down assay

GST pull-down assay was performed as described previously (Wang et al. 2021). GST- and His6-tagged proteins (10 µg each) were incubated with 40 µL of GST·Bind resin in IP buffer at 4 °C for 4 h. After centrifugation, the supernatant was then removed, and the resin was rinsed thoroughly with ice-cold PBS buffer. The bead-bound proteins were eluted with 100 µL of 1× SDS sample buffer and were then subject to immunoblotting with anti-His (CWBIO, CW0285, 1:5,000 dilution) and anti-GST (CWBIO, CW0084, 1:5,000 dilution) antibodies.

Protein stability assay in N. benthamiana

The ORFs of OsCPK17 and OsRLCK176 without stop codons were cloned into the pENTR/D-TOPO cloning vector (Invitrogen). These constructs were recombined into the Gateway destination vectors pGWB11 and pGWB14 using LR Clonase II mix (Invitrogen), respectively (Nakagawa et al. 2007). The Agrobacterium strain carrying the OsCPK17-FLAG construct was coinfiltrated with the strains containing OsRLCK176-HA or OsRLCK57-HA into N. benthamiana leaves. Total proteins were extracted from the infiltrated leaves at 2 dpi and were then detected by immunoblotting with anti-FLAG and anti-HA antibodies.

Protein stability and abundance assays

The pUC19-OsRLCK176-HA and pUC19-GFP constructs were cotransfected into rice protoplasts (2.5 × 10⁶ cells mL⁻¹) isolated from the wild-type and oscpk17-1 seedlings. The protoplasts were then incubated in the dark for 12 to 16 h followed by the treatment with the protein synthesis inhibitor CHX (50 µM). GFP was coexpressed with OsRLCK176-HA to serve as an internal reference. The abundance of OsRLCK176 and GFP was determined at different durations of time (0 and 2 h) after CHX treatment by immunoblotting with HRP-conjugated anti-HA (Roche) and anti-GFP (CWBIO, CW0086, 1:5,000 dilution) antibodies, respectively. Alternatively, the endogenous OsRLCK176 protein levels were detected by immunoblotting with an anti-OsRLCK176 antibody (Cat#AbP80608-A-SE, Beijing Protein Innovation).

Cell-free protein degradation assay

Cell-free protein degradation was detected as described (Kong et al. 2015). Briefly, total proteins were extracted from rice seedlings using the extraction buffer containing 25 mM Tris-Cl, pH 7.5, 10 mM MgCl₂, 10 mM NaCl₂, 5 mM dithiothreitol (DTT), 4 mM phenylmethanesulfonyl fluoride (PMSF), and 1 mM ATP. In vitro-purified GST-OsRLCK176 (200 ng) was incubated with rice protein extracts (500 µg) at 28 °C for different time durations. The SDS-PAGE loading buffer was used to stop reactions. The samples were then boiled and subjected to immunoblotting with an anti-GST antibody.

Site-directed mutagenesis

The constructs to express point-mutated proteins including OsCPK17^KM, OsRLCK176^KM, OsRLCK176^KM(S32A), OsRLCK176^KM(S35A), OsRLCK176^KM(S83A), and OsRLCK176^KM(S209A) were generated using a QuikChange II site-directed mutagenesis kit (Stratagene, Agilent Technologies, La Jolla, CA) according to the manufacturer's instructions. The primers used for site-directed mutagenesis are listed in Supplemental Data Set 1.

In vitro kinase assay

In vitro kinase assay was performed as previously described (Yang et al. 2022). Briefly, the test protein (1 µg) and substrate protein (2 µg) were incubated in the kinase assay buffer (25 mM Tris-HCl, pH 7.5, 10 mM MgCl₂, 1 mM DTT, 50 mM cold ATP, and 1 µCi of γ-³²P-ATP) supplemented with 10 mM MnCl₂ for RLCK kinase assay or 10 mM CaCl₂ for CPK kinase assay at 30 °C for 1 h. The samples were heated at 95 °C for 5 min and were then separated on 10% SDS polyacrylamide gels. After electrophoresis, protein phosphorylation was detected by autoradiography using the Storage Phosphor Screen with a Typhoon Trio Variable Mode Imager (GE Healthcare, Piscataway, NJ). The relative phosphorylation level was quantified using ImageJ.

Determination of phosphorylation sites in OsRLCK176

The phosphorylation residues in OsRLCK176 were determined by MS/MS as described (Wang et al. 2018b). His-OsCPK17 was incubated with MBP-OsRLCK176^KM in the kinase assay buffer at 30 °C for 2 h. The proteins were digested with trypsin at 37 °C overnight and subject to NanoLC separation with a Waters nanoAcquity nano HPLC (Waters, Milford, MA). Nanospray desorption electrospray ionization (ESI)-MS was performed with a Thermo Q Exactive high-resolution mass spectrometer (Thermo Scientific, Waltham, MA). The MS/MS spectra were analyzed with Mascot Distiller (Matrix Science, version 2.4), and the phosphorylated peptides were identified and manually inspected to assign the phosphorylation sites.

In vivo phosphorylation assay

The constructs to express OsRLCK176^KM-FLAG and GFP were transfected into the wild-type and oscpk17-1 protoplasts. Total proteins were isolated from transfected protoplasts with IP buffer containing 1× protease inhibitor cocktail and 1× phosphatase inhibitor cocktail (CWBIO). OsRLCK176^KM-FLAG was immunoprecipitated as described above and was then detected for its abundance and phosphorylation by immunoblotting with an HRP-conjugated anti-FLAG (Sigma-Aldrich, A8592, 1:5,000) and anti-phospho-Ser/Thr (Abcam, AB117253, 1:10,000) antibodies, respectively. An anti-OsRLCK176^pS83 polyclonal antibody was generated by immunizing rabbits with a phosphopeptide, Ac-C-EGGFGS(p)VFKG-NH2, to specifically detect Ser83 phosphorylation in OsRLCK176 using immunoblotting (Abmart, Shanghai, China). β-Actin was detected to indicate protein loading with an anti-β-actin antibody (CWBIO, CW0264, 1:5,000 dilution).

In vitro ubiquitination assay

In vitro ubiquitination assay was performed following the described procedure (Zhao et al. 2013). His-tagged OsPUB12, MBP-OsRLCK176, MBP-OsRLCK176^S83A, and MBP-OsRLCK176^S83D proteins were purified from E. coli. Wheat E1 (50 ng), purified E2 (50 ng), ubiquitin-HA (1 mg, Boston Biochem), and His-OsPUB12 (1 mg) were incubated with 5 mg of MBP-OsRLCK176, MBP-OsRLCK176^S83A, or MBP-OsRLCK176^S83D in 30 µL of ubiquitination reaction buffer (50 mM Tris-Cl, pH 7.5, 20 mM MgCl₂, 5 mM ATP, 1 mM DTT) at 30°C for 4 h. After the proteins were separated on SDS-PAGE gels, MBP-OsRLCK176 ubiquitination was detected with anti-MBP antibody (Abcam, AB0029, 1:5,000 dilution), and the total ubiquitination level was detected with anti-HA antibody (Roche).

In vivo ubiquitination assay

In vivo ubiquitination assay was performed as described previously (Wang et al. 2019). The wild-type protoplasts were transfected to express ubiquitin-HA and OsPUB12-HA together with OsRLCK176-FLAG or OsRLCK176^S83D-FLAG in the presence of MG132 (5 µM, Sigma). Total proteins were extracted with IP buffer and were then incubated with anti-FLAG affinity beads (Sigma) in the presence of MG132 (50 µM). After incubation at 4°C for 4 h, the beads were rinsed with 1× PBS buffer for 3 times. The bead-bound proteins were separated on SDS-PAGE gels, and ubiquitination of OsRLCK176-FLAG and OsRLCK176^S83D-FLAG was detected with anti-FLAG antibody.

Phylogenetic analysis

The phylogenetic trees were generated for OsRLCK176 and OsPUB12 homologs in rice and other plant species with 500 bootstrap replicates using MEGA 7.0 (Kumar et al. 2016). The evolutionary distances were calculated using the Poisson correction method.

Statistical analysis

Statistical differences among distinct treatments were determined by a 2-tailed Student's t-test (*P < 0.05, **P < 0.01, and ***P < 0.001) using Microsoft Excel, or by one-way ANOVA followed by Duncan's multiple range tests (at α = 0.05) using SPSS software. Statistical data are provided in Supplemental Data Set S2.

Accession numbers

OsACTIN (Os03g0718100), OsPR5 (Os04g0689900), OsPR1a (Os07g0129200), OsPR10a (Os12g0555500), OsPAL1 (Os02g0627100), OsWRKY45 (Os05g0322900), OsCPK17 (Os07g0161600), and OsRLCK176 (Os05g0110900).

Author contributions

B.M., J.W., and W.S. designed the research; B.M., G.Z., J.W., S.W., F.H., Y.N., D.L., X.Z., F.C., F.X., and S.Z. conducted the experiments and analyzed the data; B.M., J.W., and W.S. wrote the paper; all authors participated in discussion and revision of the paper.

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. Knockout and knockdown of OsCPK17 cause the transgenic plants more susceptible to infection by X. oryzae pvs. oryzae and oryzicola.

Supplemental Figure S2. OsCPK17 does not alter PAMP-induced MAPK activation in rice.

Supplemental Figure S3. OsCPK17 expression is induced by fungal and bacterial elicitors.

Supplemental Figure S4. OsCPK17 phosphorylation is activated by various PAMPs.

Supplemental Figure S5. OsCPK17 specifically interacts with OsRLCK176 in vivo and is essential for OsRLCK176 stability in rice.

Supplemental Figure S6. OsCPK17 is not phosphorylated by OsRLCK176, and OsRLCK176 is phosphorylated by OsCPK17 at Ser32, Ser35, Ser83, and Ser209 residues.

Supplemental Figure S7. The Ser83 residue in OsRLCK176 specific to rice RLCKs is a major phosphorylation site for its stability and immune function.

Supplemental Figure S8. OsPUB12 specifically interacts with OsRLCK176.

Supplemental Figure S9. OsPUB12 knockout does not alter PAMP-triggered MAPK activation in rice.

Supplemental Data Set S1. Primers used in this study.

Supplemental Data Set S2. Data for all statistical analyses performed in this study.

Supplemental File S1. Fasta file of protein alignment used for the phylogenetic analysis in Supplemental Figs. S7F and S8.

Data availability

All data presented in this study are available from the corresponding authors upon request.

Dive Curated Terms

The following phenotypic, genotypic, and functional terms are of significance to the work described in this paper:

• PTI Gramene: pattern-triggered immunity

• PTI Araport: pattern-triggered immunity