A phosphorylation-dependent ubiquitination switch orchestrates nuclear immune reprogramming upon chitin perception

Abstract

The 14-3-3 proteins, a highly conserved class in all eukaryotes, are widely associated with plant growth and stress responses. However, their role in plant immunity and its regulatory mechanisms remains elusive. Here, we show that two homologous rice 14-3-3 proteins, OsGF14f and OsGF14c, function redundantly to enhance rice resistance against Magnaporthe oryzae. The E3 ligase OsPUB20 targets OsGF14f and OsGF14c for ubiquitination and 26S proteasome-mediated degradation, thereby negatively regulating rice immunity. Remarkably, chitin perception activates the receptor-like cytoplasmic kinase OsRLCK185 that phosphorylates OsPUB20 at Thr153, which stabilizes OsGF14f and enhances rice blast resistance. Furthermore, during M. oryzae infection, OsGF14f translocates into the nucleus, where it facilitates the degradation of OsWRKY42, a transcription factor that negatively regulates defense responses. Collectively, our findings reveal a phosphorylation-dependent ubiquitination switch that links cell surface chitin perception to nuclear immune reprogramming during M. oryzae invasion.

14-3-3 proteins play roles in plant growth and stress responses. Here the authors describe a phosphorylation-dependent ubiquitination switch that governs the activity of a rice 14-3-3 protein and immune reprogramming during M. oryzae invasion.

Introduction

To combat microbial infection, plants have developed a sophisticated innate immune system that recognizes pathogen-associated molecular patterns (PAMPs) and effector proteins, triggering a cascade of defense responses^1,2. PAMPs are often mediated by pattern recognition receptors (PRRs) located at the cell surface³. PRRs directly interact with members of receptor-like cytoplasmic kinase subfamily VII (RLCK VII), which lack an extracellular domain, to initiate multiple downstream signaling pathways⁴. The RLCK family has emerged as a central component involved in the transduction of PRR-mediated signaling to mediate early immune events in plants^5,6. For example, the Arabidopsis RLCK member AtBIK1 transmits signals from the PRR FLS2 by phosphorylating RbohD, which contributes to the AtBIK1-mediated stomatal defense to bacterial infection⁷. In rice, OsRLCK185 phosphorylates the cyclic nucleotide-gated channel OsCNGC9 to enhance Ca²⁺ influx and activate its channel activity, thereby enhancing blast resistance⁸. These results suggest that an RLCK-mediated signaling cascade, which connects PAMP recognition to ROS burst and OsCNGC9-mediated Ca²⁺ influx, primarily occurs at the plasma membrane. Previous studies have shown that OsRLCK185 and its ortholog AtPBL27 in Arabidopsis are directly phosphorylated by their co-PRR CERK1 in response to chitin treatments^9,10. MAPK activation is also compromised in OsRLCK185-silenced cells and Atpbl27 mutants. Thus, OsRLCK185 and AtPBL27 link CERK1-mediated chitin perception to MAPK cascade pathway activation^11,12. However, it remains unknown how the immune signal in the nucleus is activated by RLCKs following chitin perception.

The 14-3-3 proteins are highly conserved across all eukaryotes, and increasing evidence indicates that they play a significant role in plant defense^13,14. The binding of 14-3-3 proteins to their targets typically influences protein stability, intrinsic enzymatic activity, affinity, or subcellular localization^15–18. For instance, in Arabidopsis, 14-3-3 proteins GRF6 and GRF8 activate MAP kinase cascades, thereby enhancing resistance to bacterial and fungal pathogens¹⁹. Additionally, GRF6 acts as a positive regulator of RPW8-mediated resistance to powdery mildew²⁰. In tobacco, Nt14-3-3 interacts with NtrbohD to promote reactive oxygen species (ROS) production, and associates with the resistance protein N to enhance defense against the Tobacco mosaic virus^21,22. In rice, the 14-3-3 family comprises eight members, some of which are implicated in regulating defense response. For instance, G-BOX FACTOR 14-3-3b (OsGF14b) and OsGF14e were demonstrated to positively regulate panicle blast resistance^23,24. Overexpression of OsGF14f enhances leaf blast and bacterial blight resistance²⁵. Furthermore, GF14c boosts blast resistance by stabilizing the SCARECROW LIKE 7 (OsSCL7) protein in rice²⁶. Despite these findings, the precise mechanisms by which 14-3-3 proteins are regulated in plant immunity, particularly at the post-translational level, remain largely unknown.

Ubiquitination is a prevalent post-translational modification involved in immune signaling transduction^27–29. The ubiquitination process mainly involves three types of enzymes: a ubiquitin-activating enzyme (E1), a ubiquitin-conjugating enzyme (E2), and a ubiquitin ligase (E3)³⁰. Among them, E3 ligases are crucial for selecting substrate proteins with high specificity³¹. Numerous studies have demonstrated that E3 ubiquitin ligases play vital roles in plant immune responses to pathogens³². For example, in rice, the F-box E3 ligase OsFBK16 targets phenylalanine ammonia-lyase (PAL) family proteins OsPAL1, OsPAL5, and OsPAL6 for degradation, thereby negatively regulating blast resistance³³. The RING-type E3 ubiquitin ligase IPA1-interaction protein 1 (IPI1) targets OsELF3-1 and OsELF3-2, homologs of EARLY FLOWERING3 (ELF3), for degradation to positively regulate rice immunity against M. oryzae³⁴. These findings indicate that homologous protein families can be modulated by the same E3 ubiquitin ligase. Previous studies have shown that 14-3-3 homologous proteins are also regulated by ubiquitination. For example, in Arabidopsis, the RING-type ubiquitin ligases ATL31 and ATL6 ubiquitinate and degrade 14-3-3λ and 14-3-3χ proteins to modulate the carbon/nitrogen nutrient response^35–37. In rice, OsATL38 mono-ubiquitinates OsGF14d, negatively regulating cold stress response³⁸. However, whether the homeostasis of 14-3-3 homologous proteins is modulated by ubiquitination in plant immunity remains to be investigated.

In this work, we demonstrated that the rice homologous 14-3-3 proteins OsGF14f and OsGF14c play redundant roles in positively modulating rice immunity against M. oryzae. OsGF14f and OsGF14c are ubiquitinated and degraded by the E3 ligase OsPUB20 via the 26S proteasome pathway. Interestingly, OsGF14c also interferes with E3 ligase activity of the OsPUB20, impairing the OsPUB20-mediated degradation of OsGF14f. In contrast to OsGF14f and OsGF14c, OsPUB20 acts as a negative regulator of rice immunity. We further reveal that OsRLCK185-mediated phosphorylation of OsPUB20 at Thr153 inhibits ubiquitination and degradation of OsGF14f by OsPUB20. Notably, OsGF14f traffics into the nucleus and facilitates OsWRKY42 degradation under M. oryzae infection. Collectively, our findings establish an OsRLCK185-OsPUB20-OsGF14f signaling cascade that transduces immune signals from cytoplasm to nucleus, directly linking phosphorylation mediated by RLCK to the transcriptional reprogramming of plant defense.

Results

OsGF14f and OsGF14c positively regulate rice immunity against M. oryzae

In rice, eight 14-3-3 gene members, designated as OsGF14a-OsGF14h, have been identified on rice chromosomes³⁹. Our previous RNA sequencing (RNA-seq) data from NPB plants, collected 24 h and 48 h after inoculation with M. oryzae⁴⁰, showed a significant induction of OsGF14f post-inoculation (Supplementary Fig. 1a), indicating a crucial role for OsGF14f in rice defense response against M. oryzae. To validate these findings, reverse-transcriptase quantitative PCR (RT-qPCR) analysis confirmed a rapid induction of OsGF14f at 24 h and 48 h post inoculation (Fig. 1a). Furthermore, immunoblotting analyses using an OsGF14f-specific antibody demonstrated that increased OsGF14f protein levels in NPB plants after 24 h and 48 h inoculation (Fig. 1b). To provide insight into the function of OsGF14f in blast resistance, we generated osgf14f mutants by CRISPR/Cas9 approach for blast inoculation (Supplementary Fig. 1b). The osgf14f mutants developed larger disease lesions and greater fungal biomass compared to NPB plants (Fig. 1c, d). Corresponding to reduced resistance, the osgf14f mutants showed lower expression levels of defense-related genes WRKY45 and OsPBZ1, and a diminished chitin-triggered ROS burst (Fig. 1e, f). The OsGF14f-overexpression (OsGF14f-cMyc) lines were then generated, and identified by RT-qPCR and immunoblot analyses (Supplementary Fig. 1c, d). Two independent overexpression lines showed smaller lesion and lower fungal biomass than that in the NPB when inoculated with the compatible M. oryzae isolate (Fig. 1g, h), as well as the upregulation of two defense-related genes WRKY45 and OsPBZ1 and enhanced ROS burst (Fig. 1i–k).

Fig. 1. OsGF14f and OsGF14c positively regulate rice immunity against M. oryzae.

a RT-qPCR verification of the relative transcript levels of OsGF14f identified in RNA-seq analysis. Total RNA was extracted at 0 h, 24 h and 48 h post M. oryzae infection and subjected to RT-qPCR analysis. b Immunoblot analysis of OsGF14f protein levels in NPB plants at 0 h, 24 h and 48 h post M. oryzae infection with anti-OsGF14f antibody. HSP was used as a loading control. Experiments were performed three times with similar results. c, d Disease symptoms (c) and relative fungal biomass (d) of osgf14f, osgf14c, osgf14f/14c and NPB plants. e Relative transcript levels of defense-related genes of WRKY45 and OsPBZ1 in osgf14f, osgf14c, osgf14f/14c and NPB plants. f Chitin-induced ROS accumulation dynamics in osgf14f, osgf14c, osgf14f/14c and NPB plants. Water treatment (Mock) was the negative control. g, h Disease symptoms (g) and relative fungal biomass (h) of OsGF14f-cMyc and NPB plants. i, j Relative transcript levels of defense-related genes WRKY45 (i) and OsPBZ1 (j) in OsGF14f-cMyc and NPB plants. k Chitin-induced ROS accumulation dynamics in OsGF14f-cMyc and NPB plants. Water treatment (Mock) was the negative control. l, m Disease symptoms (l) and relative fungal biomass (m) of OsGF14c-cMyc and NPB plants. n, o Relative transcript levels of defense-related genes WRKY45 (n) and OsPBZ1 (o) in OsGF14c-cMyc and NPB plants. p Chitin-induced ROS accumulation dynamics in OsGF14c-cMyc and NPB plants. Water treatment (Mock) was the negative control. All the statistical analysis data are shown as means ± SD, and significance was determined at *p < 0.05 and **p < 0.01 (n = 3, biological repeats) with a two-tailed Student’s t test. Different lowercase letters in d, e indicate a significant difference (p < 0.05, Dunnett’s multiple comparison test), and the adjusted p values were shown in figures. Data in b, c, g, l are representatives of three independent experiments with similar results.

Phylogenetic analysis revealed that OsGF14c is closely related to OsGF14f (Supplementary Fig. 1e). To assess the role of OsGF14c in blast resistance, osgf14c mutants and osgf14f/14c double mutants were then created (Supplementary Fig. 1f, g). Compared with NPB, osgf14c mutants displayed a similar susceptible phenotype to M. oryzae isolate as osgf14f (Fig. 1c, d). The osgf14f/14c double mutants were more susceptible, with greater fungal biomass, than either single mutant (Fig. 1c, d). Moreover, the expressions of WRKY45 and OsPBZ1, as well as ROS accumulation, were lower in the osgf14f/14c mutants than in either single mutant (Fig. 1e, f). We then generated OsGF14c-overexpression (OsGF14c-cMyc) lines and identified them by RT-qPCR and immunoblot analyses (Supplementary Fig. 1h, i). Two overexpression lines displayed smaller disease lesions and less fungal biomass than that in NPB (Fig. 1l, m), as well as the upregulation of WRKY45 and OsPBZ1, and the increased ROS burst (Fig. 1n–p). Taken together, these results suggest that OsGF14f and OsGF14c have redundant roles in positively modulating rice immunity against M. oryzae.

OsPUB20 mediates the ubiquitination and degradation of OsGF14f and OsGF14c

Our previous proteomic analysis of ubiquitinated proteins identified the 14-3-3 family proteins, OsGF14f and OsGF14c, as targets of ubiquitination in rice immune responses⁴¹. To verify this, we transfected OsGF14f-GFP and OsGF14c-GFP plasmids into rice protoplast individually and demonstrated that both OsGF14f and OsGF14c were ubiquitinated in planta (Supplementary Fig. 2a, b). Furthermore, OsGF14f and OsGF14c protein levels were decreased rapidly after treatment with cycloheximide (CHX, an inhibitor of protein synthesis), but stabilized after treatment with MG132 (a 26S proteasome inhibitor) (Fig. 2a, b), indicating that OsGF14f and OsGF14c are degraded by the 26S proteasome pathway. Using OsGF14f and OsGF14c as bait to screen the rice UbE3-ORFeome, a complete E3 ligase library in rice³³, we identified a U-box domain-containing protein, OsPUB20, as the simultaneous interacting proteins of OsGF14f and OsGF14c. The yeast two-hybrid (Y2H) assay confirmed that OsPUB20 interacts with OsGF14f and OsGF14c (Fig. 2c). We further validated these interactions through co-immunoprecipitation (co-IP) assays in rice protoplasts (Fig. 2d, e). Moreover, bimolecular fluorescence complementation (BiFC) assays showed that these interactions predominantly occur in the cytoplasm (Supplementary Fig. 2c). These results indicate that OsPUB20 interacts with OsGF14f and OsGF14c in planta.

Fig. 2. OsPUB20 mediates the ubiquitination and degradation of OsGF14f and OsGF14c.

a, b OsGF14f-GFP (a) and OsGF14c-GFP (b) abundance were detected in transfected protoplasts after treatment with DMSO or MG132. OsGF14f-GFP and OsOsGF14c-GFP were then detected by immunoblotting with anti-GFP antibody. Band intensities were determined by ImageJ. c OsGF14f and OsGF14c interact with OsPUB20 in yeast. Yeasts were grown on DDO (SD-Leu-Trp) or QDO (SD-Leu-Trp-His-Ade) medium. d, e Co-IP assays detect the interaction OsGF14f (d) and OsGF14c (e) with OsPUB20, and OsPUB20^C37A in rice protoplasts. Experiments were performed two times with similar results. f In vitro ubiquitination assay of OsGF14f and OsGF14c by OsPUB20. Mutant OsPUB20^C37A was used as a negative control. Immunoblot analysis was performed using anti-Ub, anti-MBP, and anti-GST antibodies. OsGF14f/14c-Ubn denotes ubiquitinated OsGF14f and OsGF14c protein band. g OsPUB20 promotes OsGF14f and OsGF14c degradation via 26S proteasome. OsGF14f-HA and OsGF14c-HA were co-expressed with OsPUB20-GFP and OsPUB20^C37A-GFP in rice protoplasts, respectively. Similar results were observed in three independent experiments. h OsGF14f does not alter the E3 ligase activity of OsPUB20. GST-OsGF14f and GST proteins were added into the reaction mixtures with E1, E2, Ub and MBP-OsPUB20. Immunoblot analysis with anti-Ub, anti-GST and anti-MBP antibodies. i OsGF14c inhibits the E3 ligase activity of OsPUB20. GST-OsGF14c and GST proteins were added into the reaction mixtures with E1, E2, Ub and MBP-OsPUB20. Immunoblot analysis with anti-Ub, anti-GST and anti-MBP. j Degradation of OsGF14f by OsPUB20 in NPB and osgf14c rice protoplasts. OsGF14f-GFP was co-transfected with HA-nLuc or HA-OsPUB20 into NPB and osgf14c rice protoplasts. Actin was served as the internal control. OsGF14f protein levels were determined by immunoblotting. k Degradation of OsGF14c by OsPUB20 in NPB and osgf14f rice protoplasts. OsGF14c-GFP was co-transfected with HA-nLuc or HA-OsPUB20 into NPB and osgf14f rice protoplasts. Actin was served as the internal control. OsGF14c protein levels were determined by immunoblotting. Data in a, b, f–k are representatives of three independent experiments with similar results.

To determine whether OsPUB20 possesses E3 ubiquitin ligase activity, an in vitro E3 ubiquitin ligase assay was performed that included a series of negative controls lacking E1, or E2. Immunoblot analysis showed that auto-ubiquitination of MBP-OsPUB20 was detected in the presence of all the components, but not MBP-OsPUB20^C37A with a Cys-to-Ala mutation in the U-box domain (Supplementary Fig. 2d), suggesting that OsPUB20 is an active E3 ligase and an intact U-box domain is required for the E3 ligase activity. To investigate whether OsPUB20 ubiquitinates OsGF14f and OsGF14c, we included OsGF14f and OsGF14c in the in vitro ubiquitination assay and found that a high-molecular weight ubiquitination band of GST-OsGF14f and GST-OsGF14c were detected when incubated with MBP-OsPUB20, E1, E2, and ubiquitin, but not with MBP-OsPUB20^C37A (Fig. 2f), which also interacted with OsGF14f and OsGF14c (Fig. 2d, e), suggesting that OsGF14f and OsGF14c were ubiquitinated by OsPUB20 in vitro. To further analyze the OsPUB20-mediated degradation of OsGF14f and OsGF14c in vivo, OsGF14f-HA or OsGF14c-HA and OsPUB20-GFP, OsPUB20^C37A-GFP were transiently co-expressed in rice protoplasts. Immunoblot analysis revealed that the protein abundance of OsGF14f-HA and OsGF14c-HA was lower when co-expressed with OsPUB20-GFP than 2×GFP or OsPUB20^C37A-GFP (Fig. 2g). Moreover, this degradation is inhibited when treated with MG132 (Fig. 2g). These results suggest that OsPUB20 promotes the degradation of OsGF14f and OsGF14c in vivo via the 26S proteasome pathway.

To investigate whether OsGF14f and OsGF14c affect OsPUB20 E3 ligase activity, in vitro ubiquitination assays showed that the intensity of OsPUB20 polyubiquitinated bands was similar when incubated with GST-OsGF14f compared to GST (Fig. 2h). However, the polyubiquitinated bands of OsPUB20 were noticeably weaker when incubated with GST-OsGF14c compared to GST (Fig. 2i), indicating that OsGF14c, but not OsGF14f, inhibited E3 ligase activity of OsPUB20. To further analyze the impact of OsGF14c on the OsPUB20-mediated degradation of OsGF14f, OsGF14f-GFP and HA-OsPUB20 were transiently co-expressed in NPB and osgf14c rice protoplasts. Immunoblot analysis revealed lower levels of OsGF14f-GFP protein in osgf14c than NPB (Fig. 2j). Conversely, OsGF14c-GFP protein levels were similar in both osgf14f and NPB (Fig. 2k). These results indicate that OsGF14c interferes with OsPUB20 E3 ligase activity, impairing the OsPUB20-mediated degradation of OsGF14f. The interactions between OsPUB20, OsGF14f, and OsGF14c demonstrate a negative feedback regulatory mechanism that prevents the excessive degradation of OsGF14f and OsGF14c, thereby maintaining the rice immune homeostasis.

OsPUB20 acts upstream of OsGF14f and OsGF14c to negatively regulate rice blast resistance

To investigate the role of OsPUB20 in disease resistance against M. oryzae, we generated OsPUB20-overexpression lines (OsPUB20-OX) in NPB background (Supplementary Fig. 3a). OsPUB20-OX lines exhibited larger disease lesions and a greater fungal biomass thanNPB when inoculated with the compatible M. oryzae isolate (Fig. 3a, b), as well as the downregulation of WRKY45 and OsPBZ1 (Fig. 3c, d), and the reduced ROS burst (Fig. 3e). Concurrently, ospub20 mutants were generated using the CRISPR/Cas9 approach in NPB background (Supplementary Fig. 3b). Punch inoculations with M. oryzae isolate resulted in smaller disease lesions and less fungal biomass in ospub20 mutants than that in NPB (Fig. 3f). Phylogenetic analysis showed that OsPUB20 is closely related to OsPUB19⁴², which promotes us to generate ospub19/20 double mutants for further blast inoculation (Supplementary Fig. 3c). The smaller disease lesions and less fungal biomass on the ospub19/20 lines than those on the ospub20 mutants (Fig. 3f). Consistent with these phenotypes, the transcript levels of defense-related genes WRKY45 and OsPBZ1, along with the ROS burst, were upregulated in the ospub19/20 mutants compared to ospub20 (Fig. 3g–i). Together, these results demonstrated that OsPUB19 and OsPUB20 are redundantly negative regulators of rice blast resistance.

Fig. 3. OsPUB20 acts upstream of OsGF14f and OsGF14c to negatively regulate rice blast resistance.

a, b Disease symptoms (a) and relative fungal biomass (b) of representative leaves of OsPUB20-OX and NPB plants. c, d Relative transcript levels of defense-related genes WRKY45 (c) and OsPBZ1 (d) in OsPUB20-OX and NPB plants. e Chitin-induced ROS accumulation dynamics in OsPUB20-OX and NPB plants. Water treatment (Mock) was the negative control. f Disease symptoms and relative fungal biomass of representative leaves of ospub20, ospub19/20 and NPB plants. g, h Relative transcript levels of WRKY45 (g) and OsPBZ1 (h) in ospub20, ospub19/20 and NPB plants. i Chitin-induced ROS accumulation dynamics in ospub20, ospub19/20 and NPB plants. Water treatment (Mock) was the negative control. j Immunoblot analysis of OsGF14f and OsGF14c protein levels in ospub20, ospub19/20 and NPB plants with anti-OsGF14f and anti-OsGF14c antibodies. HSP was used as a loading control. Similar results were observed in four independent experiments. k, l M. oryzae impairs the ubiquitination levels of OsGF14f (k) and OsGF14c (l). OsGF14f-cMyc, ospub19/20/OsGF14f-cMyc and OsGF14c-cMyc, ospub19/20/OsGF14c-cMyc transgenic seedlings were inoculated with M. oryzae. Total proteins were extracted and enriched with anti-cMyc antibody. OsGF14f and OsGF14c were detected with anti-cMyc antibody. The poly-ubiquitination of OsGF14f and OsGF14c were detected with anti-Ub antibody. Similar results were observed in at least three independent experiments. m Disease symptoms and relative fungal biomass of OsGF14f-cMyc, ospub19/20/OsGF14f-cMyc and NPB plants. n Disease symptoms and relative fungal biomass of OsGF14c-cMyc, ospub19/20/OsGF14c-cMyc and NPB plants. o Disease symptoms and relative fungal biomass of osgf14f/14c, ospub19/20, ospub19/20osgf14f/14c and NPB plants. p Relative transcript levels of defense-related genes WRKY45 and OsPBZ1 in osgf14f/14c, ospub19/20, ospub19/20osgf14f/14c and NPB plants. All the statistical analysis data are shown as means ± SD, and significance was determined at *p < 0.05 and **p < 0.01 (n = 3, biological repeats) with a two-tailed Student’s t test. Different lowercase letters in f–h, m–p indicate a significant difference (p < 0.05, Dunnett’s multiple comparison test). Data in (a, f, and m–o) are representatives of three independent experiments with similar results.

To examine the degradation of OsGF14f and OsGF14c by OsPUB20 in planta, we compared OsGF14f and OsGF14c protein abundance in ospub20 and ospub19/20 mutants. Immunoblotting analysis demonstrated that both OsGF14f and OsGF14c protein accumulation were higher in ospub20 plants compare with NPB (Fig. 3j). Notably, a much higher accumulation of OsGF14f and OsGF14c in ospub19/20 plants than that of ospub20 plants (Fig. 3j), while the transcript levels of OsGF14f and OsGF14c in NPB were comparable to those in the mutant plants (Supplementary Fig. 3d, e). These results strongly support the notion that OsPUB20 promotes the degradation of OsGF14f and OsGF14c in planta.

To further explore the degradation of OsGF14f and OsGF14c mediated by OsPUB20 in response to M. oryzae infection, we generated transgenic plants knocking-out ospub19 and ospub20 in OsGF14f-cMyc and OsGF14c-cMyc overexpression backgrounds (ospub19/20/OsGF14f-cMyc, ospub19/20/OsGF14c-cMyc) (Supplementary Fig. 3f, g) to examine the ubiquitination levels of OsGF14f and OsGF14c in response to M. oryzae. Immunoblot analysis indicated that the poly-ubiquitinated levels of OsGF14f and OsGF14c were substantially attenuated upon M. oryzae infection (Fig. 3k, l). Moreover, the ubiquitination levels of OsGF14f and OsGF14c were impaired in the ospub19/20 mutant background (Fig. 3k, l), whereas the protein levels of OsGF14f and OsGF14c were increased in ospub19/20 mutants (Fig. 3k, l). M. oryzae infection also induced the accumulation of OsGF14f and OsGF14c (Fig. 3k, l). We then evaluated the blast resistance of OsGF14f-cMyc, ospub19/20/OsGF14f-cMyc, OsGF14c-cMyc and ospub19/20/OsGF14c-cMyc plants. Punch inoculation assays demonstrated that ospub19/20/OsGF14f-cMyc plants exhibited elevated resistance to RB22, with lower fungal biomass compared to OsGF14f-cMyc (Fig. 3m). Similarly, ospub19/20/OsGF14c-cMyc plants exhibited increased resistance compared to OsGF14c-cMyc (Fig. 3n). These data strongly support the notion that OsPUB20 is responsible for the degradation of OsGF14f and OsGF14c during M. oryzae infection.

To further elucidate the genetic relationship of OsPUB20 with OsGF14f and OsGF14c, we knocked out OsPUB19 and OsPUB20 in osgf14f/14c mutants to generate ospub19/20osgf14f/14c quadruple mutants (Supplementary Fig. 3h). Punch inoculation assay showed that the ospub19/20osgf14f/14c mutant had a similar blast resistance and fungal biomass as osgf14f/14c mutant (Fig. 3o). Consistently, the expressions of WRKY45 and OsPBZ1 were similar in ospub19/20osgf14f/14c and osgf14f/14c plants (Fig. 3p). These results suggest that OsPUB20 acts genetically upstream of OsGF14f and OsGF14c to negatively modulate blast resistance.

OsRLCK185 is upstream of OsPUB20 to positively regulate blast resistance

To explore the regulation mechanism of OsPUB20 in blast resistance, we transiently expressed OsPUB20-GFP plasmids in rice protoplasts to identify OsPUB20-interacting proteins by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Among candidate proteins identified was OsRLCK185 (Supplementary Data 1), a key immune regulator⁸. We first tested the interaction between OsPUB20 and OsRLCK185 in Y2H assay (Fig. 4a). A co-IP assay confirmed that OsPUB20 associated with OsRLCK185 in rice protoplasts (Fig. 4b). Moreover, BiFC assay demonstrated that OsPUB20 interacted with OsRLCK185 primarily at the plasma membrane, while no signal was observed in the negative control combinations (Supplementary Fig. 4a). These results indicate that OsRLCK185 interacts with OsPUB20 in planta.

Fig. 4. OsRLCK185 is upstream of OsPUB20 to positively regulate blast resistance.

a OsPUB20 interacts with OsRLCK185 in yeast. Yeast was grown on DDO (SD-Leu-Trp) or TDO (SD-Leu-Trp-His) selective medium containing 0.3 mM 3-amino-1,2,4-triazole (3AT). b Co-IP assays of interactions between OsPUB20 and OsRLCK185 in rice protoplasts. OsPUB20-GFP was co-expressed with HA-nLuc, or OsRLCK185-HA in rice protoplasts. Experiments were performed two times with similar results. c, d Disease symptoms (c) and relative fungal biomass (d) of representative leaves of osrlck185 and NPB plants. e, f Relative transcript levels of WRKY45 (e) and OsPBZ1 (f) in osrlck185 and NPB plants. g, h Disease symptoms (g) and relative fungal biomass (h) of representative leaves of OsRLCK185-OX and NPB plants. i, j Relative transcript levels of WRKY45 (i) and OsPBZ1 (j) in OsRLCK185-OX and NPB plants. k Disease symptoms and relative fungal biomass of representative leaves of osrlck185, ospub19/20, osrlck185ospub19/20 and NPB plants. Different lowercase letters indicate a significant difference (p < 0.05, Dunnett’s multiple range test). l, m Relative transcript levels of WRKY45 (l) and OsPBZ1 (m) in osrlck185, ospub19/20, osrlck185ospub19/20 and NPB plants. Different lowercase letters indicate a significant difference (p < 0.05, Dunnett’s multiple comparison test). All the statistical analysis data are shown as means ± SD, and significance was determined at *p < 0.05 and **p < 0.01 (n = 3, biological repeats) with a two-tailed Student’s t test.

OsRLCK185 was previously shown to modulate rice blast resistance⁸. To verify this, we generated osrlck185 mutants in NPB background to evaluate the blast resistance (Supplementary Fig. 4b). The osrlck185 mutants were more susceptible, with greater fungal biomass than NPB (Fig. 4c, d). Moreover, the transcript levels of defense-related genes, including WRKY45 and OsPBZ1, were significantly repressed in osrlck185 plants compared to NPB (Fig. 4e, f). The OsRLCK185 overexpressing plants (OsRLCK185-OX) were then generated and identified by RT-qPCR analysis (Supplementary Fig. 4c). Contrast with the osrlck185 mutants, OsRLCK185-OX plants showed elevated resistance and reduced fungal biomass (Fig. 4g, h), and upregulation of WRKY45 and OsPBZ1 (Fig. 4i, j), demonstrating that OsRLCK185 positively regulates rice blast resistance. To further elucidate the genetic relationship between OsPUB20 and OsRLCK185, we knocked out OsRLCK185 in the ospub19/20 mutants background to generate osrlck185ospub19/20 triple mutants (Supplementary Fig. 4d). The osrlck185ospub19/20 triple mutants exhibited greater susceptible lesion and more fungal biomass than ospub19/20 plants, but showed greater resistance than osrlck185 plants (Fig. 4k), indicating that OsPUB20-mediated blast resistance partially depends on OsRLCK185. Consistently, the transcript levels of the defense-associated genes WRKY45 and OsPBZ1 in the osrlck185ospub19/20 triple mutant was repressed compared to ospub19/20 plants, but increased compared to osrlck185 plants (Fig. 4l, m). These results suggest that OsRLCK185 functions genetically upstream of OsPUB20 to positively modulate blast resistance.

Phosphorylation of OsPUB20 by OsRLCK185 decreases the ubiquitination and stabilizes OsGF14f protein

OsRLCK185 is a functional kinase⁹, we then examined whether OsRLCK185 phosphorylates OsPUB20 using an in vitro kinase assay, OsRLCK185^K108A served as a negative control⁹. The results showed that GST-OsRLCK185, but not GST-OsRLCK185^K108A, phosphorylates OsPUB20 in vitro (Fig. 5a). Using LC-MS/MS analysis after an in vitro phosphorylation reaction with GST-OsRLCK185 and His-OsPUB20, we identified threonine 153 (T153), located between U-box domain and ARM repeats of OsPUB20 (Supplementary Fig. 2d and Supplementary Data 2), as the OsRLCK185 phosphorylation site (Supplementary Fig. 5a). To assess the importance of this site, we mutated T153 to alanine (T153A) and found that OsRLCK185 could no longer phosphorylate OsPUB20^T153A (Fig. 5b). To further determine if OsPUB20 is phosphorylated in vivo, we transiently expressed OsPUB20-GFP in NPB rice protoplasts treated with chitin. Immunoblotting analysis revealed that OsPUB20 is phosphorylated, and chitin treatment enhances its phosphorylation (Supplementary Fig. 5b). To confirm OsRLCK185-mediated OsPUB20 phosphorylation in vivo, we co-expressed cMyc-OsRLCK185 with OsPUB20-GFP and OsPUB20^T153A-GFP in rice protoplasts and examined OsPUB20 phosphorylation through immunoblotting following immunoprecipitation of OsPUB20 proteins. Immunoblotting analysis indicated that chitin treatment induced OsPUB20 phosphorylation (Fig. 5c). Moreover, the OsPUB20^T153A phosphorylation was largely inhibited in vivo (Fig. 5c). These results confirmed that OsRLCK185 phosphorylates OsPUB20 at Thr153.

Fig. 5. Phosphorylation of OsPUB20 by OsRLCK185 decreases the ubiquitination and stabilizes OsGF14f protein.

a In vitro kinase assay indicates that OsRLCK185 phosphorylates OsPUB20. His-OsPUB20 protein was subjected to an in vitro kinase assay using GST-OsRLCK185 as the kinase. The GST-OsRLCK185 ^K108A was used as a negative control. b OsPUB20^T153 is required for OsRLCK185-mediated phosphorylation. The threonine (T) residue around OsPUB20 was mutated to alanine (A). His-OsPUB20/OsPUB20^T153A proteins were subjected to an in vitro kinase assay. c OsRLCK185 phosphorylates OsPUB20 at Thr153 site in vivo. Rice protoplasts were co-transfected with OsRLCK185-cMyc, and OsPUB20/OsPUB20^T153A-GFP with or without 80 nM chitin treatment. After immunoprecipitation using anti-GFP, the phosphorylated OsPUB20 or OsPUB20^T153A proteins were probed with anti-pSer/Thr antibody. The input proteins were probed with anti-cMyc and anti-GFP antibodies. d, e Co-IP assays detect the interactions of OsPUB20/OsPUB20^T153D/OsPUB20^T153A-GFP with OsGF14f-HA (d) and OsGF14c-HA (e). OsPUB20-GFP, OsPUB20^T153A-GFP, and OsPUB20^T153D-GFP were individually co-expressed with OsGF14f-HA or OsGF14c-HA in rice protoplasts. The immunocomplex and input proteins were analyzed by immunoblotting using anti-HA and anti-GFP antibodies. f, g OsPUB20^T153D promotes OsGF14f stability (f), but not alter OsGF14c (g) protein level in rice protoplasts. h The ubiquitination levels of OsGF14f after co-expression with OsPUB20, OsPUB20^T153A, OsPUB20^T153D in rice protoplasts. Flag-Ub was transiently co-expressed with the indicated combinations in rice protoplasts. Total protein was extracted and then subject to immunoprecipitation with anti-Flag. The poly-ubiquitination of OsGF14f-GFP was detected by immunoblotting. i, j Disease symptoms (i) and relative fungal biomass (j) of OsPUB20-OX, OsPUB20^T153A-OX, and OsPUB20^T153D-OX transgenic plants. Different lowercase letters indicate a significant difference (p < 0.05, Dunnett’s multiple comparison test). Data are means ± SD (n = 3, biological replicates). k Immunoblot analysis of OsGF14f and OsGF14c protein levels in OsPUB20-OX, OsPUB20^T153A-OX, and OsPUB20^T153D-OX and NPB plants with anti-OsGF14f and anti-OsGF14c antibodies. All the statistical analysis data are shown as means ± SD, and significance was determined at *p < 0.05 and **p < 0.01 (n = 3, biological repeats) with a two-tailed Student’s t test. All data are representative of at least three independent experiments with similar results.

To explore whether OsPUB20^T153 phosphorylation affects its function, in vitro ubiquitination assays showed that E3 activity of OsPUB20^T153A and OsPUB20^T153D was similar with OsPUB20 (Supplementary Fig. 5c), indicating that phosphorylation of OsPUB20 at T153 is not essential for E3 ligase activity. A co-IP assay in rice protoplasts revealed that OsPUB20^T153A exhibited stronger interaction with OsGF14f and OsGF14c, while OsPUB20^T153D markedly reduced the interaction with OsGF14f, but had no effect on OsGF14c (Fig. 5d, e). To determine whether OsPUB20 T153 phosphorylation affects the degradation of OsGF14f and OsGF14c, we transiently co-expressed OsPUB20-GFP, OsPUB20^T153D-GFP, and OsPUB20^T153A-GFP with OsGF14f-HA and OsGF14c-HA in rice protoplasts. The results revealed that OsPUB20^T153A decreased OsGF14f and OsGF14c protein levels compared to OsPUB20 (Fig. 5f, g). In contrast, OsPUB20^T153D caused higher OsGF14f accumulation, but without affecting OsGF14c (Fig. 5f, g), indicating that phosphorylation of OsPUB20 at T153 stabilizes OsGF14f, but does not affect OsGF14c. We further co-expressed HA-OsPUB20 variants with OsGF14f-GFP or OsGF14c-GFP in rice protoplasts. Following immunoprecipitation, the results showed that OsPUB20^T153A enhanced OsGF14f ubiquitination, whereas OsPUB20^T153D largely decreased ubiquitination of OsGF14f (Fig. 5h). Conversely, a higher ubiquitination level was observed in OsGF14c when co-expressed with OsPUB20^T153A, whereas a similar ubiquitination level of OsGF14c when co-expressed with OsPUB20^T153D (Supplementary Fig. 5d). These results indicated that phosphorylation of OsPUB20 at T153 reduces association with OsGF14f, thereby impairing ubiquitination of OsGF14f and enhancing its stability.

To evaluate whether Thr153 phosphorylation in OsPUB20 contributes to disease resistance in rice, we generated the transgenic plants overexpressing OsPUB20^T153D (OsPUB20^T153D-OX) and OsPUB20^T153A (OsPUB20^T153A-OX) (Supplementary Fig. 5e). Punch inoculation assay showed that the OsPUB20^T153D-OX plants developed smaller lesions and less fungal biomass relative to the OsPUB20-OX (Fig. 5i, j). By contrast, OsPUB20^T153A-OX plants exhibited reduced resistance relative to OsPUB20-OX (Fig. 5i, j). Further analysis of OsGF14f and OsGF14c protein abundance showed that OsPUB20^T153D-OX plants accumulated more OsGF14f protein, but similar OsGF14c protein compared with NPB (Fig. 5k). However, there is less accumulation of OsGF14f and OsGF14c in OsPUB20^T153A-OX plants than that of OsPUB20-OX plants (Fig. 5k). Collectively, our results demonstrated that OsRLCK185-mediated phosphorylation of OsPUB20 enhances rice blast resistance.

OsGF14f translocates into the nucleus and facilitates OsWRKY42 degradation

To investigate how OsGF14f and OsGF14c function in rice blast resistance, we generated the OsGF14f-overexpression (OsGF14f-GFP) and OsGF14c-overexpression (OsGF14c-GFP) transgenic plants (Supplementary Fig. 6a, b), to examine their subcellular localization. Chitin treatment induced the translocation of OsGF14f into nucleus in root cells of OsGF14f-GFP overexpressing plants, whereas OsGF14c-GFP plants did not exhibit this nuclear localization (Supplementary Fig. 6c, d). To verify the translocation during M. oryzae infection, we inoculated OsGF14f-cMyc and OsGF14c-cMyc plants and investigated the nucleocytoplasmic distribution of OsGF14f and OsGF14c. Immunoblot analyses revealed that OsGF14f predominantly accumulated in the nucleus during M. oryzae infection, with a slight increase in the cytoplasm (Fig. 6a). In contrast, OsGF14c exhibited a slight cytoplasmic increase but remained largely unchanged in the nucleus, except for a decrease at 120 h (Fig. 6b), suggesting that OsGF14f traffics into nucleus in response to blast infection. To determine whether the nuclear entry of OsGF14f contributes to blast resistance, we overexpressed OsGF14f fusing NLS from the SV40 virus (PKKKRKV) (OsGF14f-NLS-GFP) in NPB background (Supplementary Fig. 6a). Punch inoculation assay showed that OsGF14f-NLS-GFP plants exhibited elevated resistance with reduced fungal biomass compared to OsGF14f-GFP plants (Fig. 6c, d), while OsGF14c-NLS-GFP plants displayed similar resistance and fungal biomass compared to OsGF14c-GFP plants (Fig. 6e, f and Supplementary Fig. 6b). These results indicated that nuclear localization of OsGF14f is crucial for blast resistance. Furthermore, we found that chitin treatment slightly enhances the nuclear translocation of OsGF14f in NPB and ospub20 mutants, whereas has no effect on osrlck185 mutants (Supplementary Fig. 6e). These results indicate that the nuclear translocation of OsGF14f may depends on OsRLCK185.

Fig. 6. OsGF14f translocates into the nucleus and facilitates OsWRKY42 degradation.

a, b OsGF14f-cMyc (a) and OsGF14c-cMyc (b) abundance in cytosolic- and nuclear-enriched fractions from OsGF14f-cMyc and OsGF14c-cMyc plants inoculated with isolate RB22. The leaf tissue was harvested at 0, 24, 48, 72, 96, and 120 h post-inoculation. Histone H3 served as a nuclear marker and HSP as a cytosolic marker. The experiment was repeated two biological replicates with similar results. c, d Disease symptoms (c) and relative fungal biomass (d) of representative leaves of OsGF14f-NLS-GFP and NPB plants. e, f Disease symptoms (e) and relative fungal biomass (f) of representative leaves of OsGF14c-NLS-GFP and NPB plants. g Co-IP assay shows that OsWRKY42 interacts with OsGF14f, but not OsGF14c in rice protoplasts. Experiments were performed two times with similar results. h BiFC detects the interactions of OsGF14f and OsGF14c with OsWRKY42 in N. benthamiana. Scale bars, 20 μm. Experiments were performed two times with similar results. i OsGF14f enhances the degradation of OsWRKY42 with 100 μM CHX and 80 nM chitin treatment in rice protoplasts. HA-OsWRKY42 was co-transfected with 2×GFP or OsGF14f-GFP into rice protoplasts. OsWRKY42 protein levels were determined by immunoblotting. j Disease symptoms and relative fungal biomass of oswrky42 and NPB plants. k, l Relative transcript levels of WRKY45 (k) and OsPBZ1 (l) in oswrky42 and NPB plants. m Relative transcript levels of JA-responsive genes OsAOS2 and OsJAZ8 in OsGF14f-cMyc, osgf14f, and NPB plants. All the statistical analysis data are shown as means ± SD, and significance was determined at *p < 0.05 and **p < 0.01 (n = 3, biological repeats) with a two-tailed Student’s t test. Different lowercase letters in (d, f) indicate a significant difference (p < 0.05, Dunnett’s multiple comparison test). Data in (i, j), and (k-m) are representatives of three independent experiments with similar results.

To understand how OsGF14f functions in the nucleus to modulate blast resistance, we identified OsGF14f-interacting proteins by LC-MS/MS assays with OsGF14f-GFP plants. Among the candidate proteins, a nuclear-localized protein was OsWRKY42 (Supplementary Fig. 6f and Supplementary Data 3). Co-IP assays confirmed OsWRKY42 interacts with OsGF14f, but not OsGF14c (Fig. 6g). BiFC assay showed that OsWRKY42 interacts with OsGF14f in the nucleus, but no signal with OsGF14c (Fig. 6h). To elucidate the regulation relationship between OsGF14f and OsWRKY42, we transiently co-expressed OsGF14f-GFP and HA-OsWRKY42 in rice protoplasts. Immunoblot analyses showed that OsGF14f promotes OsWRKY42 degradation, which is enhanced by chitin treatment (Fig. 6i). Treatment with MG132 markedly inhibited the degradation of OsWRKY42 by OsGF14f (Supplementary Fig. 6g). Furthermore, cell-free degradation assay revealed that MBP-OsWRKY42 degradation was slower in osgf14f mutant compared to NPB (Supplementary Fig. 6h), and this degradation was inhibited by MG132 (Supplementary Fig. 6h), suggesting that OsGF14f promotes the destabilization of OsWRKY42.

Previous study showed that OsWRKY42 negatively regulates blast resistance by suppressing the expressions of jasmonic acid (JA) signaling-related genes⁴³. To verify this, we firstly generated oswrky42 mutant using the CRISPR/Cas9 method (Supplementary Fig. 6i). Punch inoculation assays showed that oswrky42 mutants enhanced blast resistance with less fungal biomass than NPB plants (Fig. 6j). Consistently, the defense-related genes WRKY45 and OsPBZ1 were significantly upregulated in oswrky42 plants compared to NPB (Fig. 6k, l). The promotion of OsWRKY42 degradation by OsGF14f led us to investigate whether OsGF14f influences the expression of JA-mediated related genes. RT-qPCR results revealed that the expression levels of OsAOS2 and OsJAZ8 were upregulated in OsGF14f-cMyc plants and repressed in osgf14f mutants (Fig. 6m). These results indicate that OsGF14f promotes the degradation of OsWRKY42, thereby enhancing the expression of the JA-responsive genes to positively regulate blast resistance.

Discussion

Ubiquitination of 14-3-3 proteins by OsPUB20 modulates plant immunity

14-3-3 proteins play crucial roles in plant immune responses⁴⁴. In this study, we found that OsGF14f exhibited significantly higher expression and protein accumulation during M. oryzae infection (Fig. 1a, b). Genetic evidence showed that OsGF14f and OsGF14c, the most closely homolog protein, redundantly and positively modulate rice immunity (Fig. 1). However, the specific E3 ubiquitin ligase responsible for the stability of 14-3-3 proteins in plant immunity had not been identified. Here, we discovered that the U-box-type E3 ligase OsPUB20 interacts with and specifically ubiquitinates OsGF14f and OsGF14c, promoting their degradations through the 26S proteasome pathway (Fig. 2). Phenotypic analysis revealed that OsPUB20 negatively modulates rice defense against M. oryzae, which contrasts with the phenotype of OsGF14f and OsGF14c. The quadruple ospub19/20osgf14f/14c mutant displays a similar susceptible phenotype to the osgf14f/14c mutant (Fig. 3). These results suggest that 14-3-3 proteins OsGF14f and OsGF14c are ubiquitination substrates of the E3 ubiquitin ligase OsPUB20 and act downstream of OsPUB20 to regulate defense response in rice.

The 14-3-3 proteins are known to function by binding to their target proteins and affecting their activity⁴⁵. For instance, in Arabidopsis, 14-3-3λ and 14-3-3κ repress the SOS2-LIKE PROTEIN KINASE5 (PKS5) activity under salt stress⁴⁶. However, no studies have reported whether 14-3-3 proteins affect the E3 ligase activity. Here, we showed that OsGF14c, but not OsGF14f, acts as a feedback inhibition of OsPUB20 E3 ligase activity (Fig. 2h, i). OsGF14f protein abundance is primarily controlled by OsPUB20, enabling exact immune signal transduction, while OsGF14c suppresses OsGF14f degradation by inactivating OsPUB20 (Fig. 2), which maintains the dynamic homeostasis of OsGF14f. Further research is needed to understand how OsGF14c, but not OsGF14f, inhibits OsPUB20 E3 ligase activity.

Phosphorylation of OsPUB20 by OsRLCK185 enhances immune response

RLCKs are part of a large superfamily of protein kinases, divided into 13 subfamilies, phosphorylating various substrates to mediate immune signaling^5,47. For instance, in rice, OsRLCK118 interacts with and phosphorylates the NADPH oxidase OsRbohB to positively regulate immunity⁴⁸. OsRLCK176 phosphorylates the calcium-dependent protein kinase OsCPK4⁴⁹. However, the phosphorylation of E3 ubiquitin ligases by RLCKs has not been documented. Our study revealed that OsRLCK185 interacts with and phosphorylates the E3 ubiquitin ligase OsPUB20 at the Thr153 residue (Fig. 5a, b). Moreover, chitin induces the phosphorylation of OsPUB20 by OsRLCK185 (Fig. 5c). Genetic evidence indicated that the osrlck185ospub19/20 mutant exhibits impaired resistance compared to ospub19/20 mutant (Fig. 4k), indicating that OsRLCK185-mediated phosphorylation of OsPUB20 is required for OsPUB20-mediated blast resistance.

Phosphorylation of E3 ligases is vital for E3 ubiquitin ligase function^50,51. For instance, in Arabidopsis, BAK1-mediated phosphorylation of PUB12 and PUB13 is essential for their interaction with and flagellin-induced degradation of FLS2⁵². CPK28 has been reported to phosphorylate and enhance the E3 activity of PUB25 and PUB26, promoting the degradation of the immune-regulatory kinase BIK1⁵³. Interestingly, our findings showed that OsPUB20 phosphorylation at T153 by OsRLCK185 does not alter the E3 ligase activity of OsPUB20 (Supplementary Fig. 5c). Instead, it reduces its interaction with OsGF14f, thereby impairing the ubiquitination and degradation of OsGF14f and leading to its accumulation (Fig. 5). Accordingly, plants overexpressing OsPUB20^T153D exhibited enhance blast resistance (Fig. 5i, j). Thus, OsRLCK185 positively regulates the phosphorylation of OsPUB20 to activate immunity. However, OsPUB20^T153D does not affect the degradation of OsGF14c, suggesting that the stability of OsGF14c protein may be regulated by another pathway, which warrants further investigation.

OsGF14f translocation to the nucleus facilitates OsWRKY42 degradation and enhances blast resistance

The nuclear-cytoplasmic trafficking of 14-3-3 proteins is crucial for their role in biological process⁵⁴. For instance, In Arabidopsis, the cold-responsive protein kinase 1 (CRPK1) phosphorylates 14-3-3 proteins, promoting their translocation from cytosol to nucleus, thereby negatively affecting freezing tolerance⁵⁵. However, the role of nuclear-cytoplasmic trafficking of 14-3-3 proteins in plant immunity remains largely unexplored. Our study revealed that M. oryzae infection promotes the translocation of OsGF14f to the nucleus, which is essential for OsGF14f-mediated blast resistance (Fig. 6). In contrast, the nucleus location of OsGF14c remains unchanged upon M. oryzae challenge. The selective nuclear accumulation suggests that OsGF14f acts as a key regulator of disease resistance, with its translocation initiating immune responses in the nucleus against blast fungus infection.

Emerging evidence indicates 14-3-3 proteins interact with transcription factors (TFs) to influence downstream processes^56,57. For instance, OsGF14c interacts with BRASSINAZOLE-RESISTANT 1 (OsBZR1) and reduce its nuclear localization⁵⁸. OsGF14f interacts with OsbZIP23, enhancing its transcription activity to confer osmotic stress tolerance⁵⁹. However, no studies have reported interactions between 14-3-3 proteins and WRKY-type TF in plant immunity. Our research demonstrated that OsGF14f interacts with OsWRKY42 in the nucleus, and promotes its degradation (Fig. 6). The OsWRKY42 degradation is subject to the 26S proteasome pathway. These results imply that OsGF14f might act as a scaffold protein that promotes OsWRKY42 degradation by OsPUB20 or an unknown E3 ligase. WRKY42, a transcriptional repressor, negatively regulates rice disease resistance by suppressing JA-dependent signaling⁴³. We showed oswrky42 mutants indeed enhance the resistance to M. oryzae (Fig. 6j-l). Furthermore, the expression levels of OsAOS2 and OsJAZ8, which are negatively regulated by OsWRKY42, increased in OsGF14f overexpression plants, decreased in osgf14f mutants (Fig. 6m). These results indicate that OsGF14f confers blast resistance by activating JA signaling. OsGF14f upregulates the transcript levels of two defense-related genes, WRKY45 and OsPBZ1, both of which are key components of the salicylic acid (SA) pathway. Therefore, the SA and JA pathways may act synergistically to modulate OsGF14f-mediated rice blast resistance against M. oryzae infection. However, OsGF14c does not interact with OsWRKY42, suggesting that it regulates blast resistance through a different pathway. Therefore, further identification of OsGF14c-intaracting proteins in the cytoplasm will further elucidate its role in rice immunity.

In summary, our findings revealed a phosphorylation and ubiquitination immune pathway, initiated by chitin perception, along with a hierarchical regulatory mechanism through which OsRLCK185 regulates nucleus signaling during blast fungus invasion (Fig. 7). We revealed that the U-box E3 ligase OsPUB20 targets the positive regulators OsGF14f and OsGF14c for degradation to negatively regulate blast resistance. While, OsRLCK185 phosphorylates OsPUB20 to inhibit ubiquitination and degradation of OsGF14f, leading to the stabilization of OsGF14f. In response to infection, OsGF14f translocates into nucleus and promotes the degradation of OsWRKY42, releasing its suppression on the transcriptions of OsAOS2 and OsJAZ8. Together, our study uncovers a mechanism that links receptor-like cytoplasmic kinases to nuclear immune reprogramming via a phosphorylation-dependent ubiquitination switch.

Fig. 7. A proposed working model of OsRLCK185, OsPUB20, 14-3-3 proteins and OsWRKY42 in rice immunity.

In normal conditions, OsPUB20 ubiquitinates and degrades OsGF14f and OsGF14c to negatively regulate blast resistance. OsRLCK185 phosphorylates OsPUB20 at Thr153, decreasing the ubiquitination and stabilizing OsGF14f protein. During infection with M. oryzae and chitin stimulation, the phosphorylation of OsPUB20 by OsRLCK185 is enhanced. OsGF14f, but not OsGF14c, translocates from the cytosol to the nucleus to promote degradation of OsWRKY42, a negative regulator of blast resistance, leading to the activation of JA-dependent pathways and enhanced blast resistance.

Methods

Plant materials and growth conditions

The rice (Oryza sativa L.) cultivars NPB was used for disease evaluation in this study. All transgenic plants were generated through transformation of rice callus with Agrobacterium strain EHA105 carrying the indicated constructs. T0 progeny were screened by PCR, sequencing, or RT-qPCR. All rice plants were cultivated in the greenhouse (28 °C for 12 h in the light, 26 °C for 12 h in the dark, 70% humidity) or field for seed production and phenotyping.

Rice blast inoculation and disease resistance assay

M. oryzae isolate RB22 was grown on the oatmeal agar for about two weeks. The spore concentration in the suspension was about 3.0×10⁵/mL. The suspension was applied to 8-week-old plants through punch inoculation. At 14 days after inoculation, leaves with representative lesions were photographed, and the relative fungal biomass was determined using the threshold cycle value (C_T) of M. oryzae Pot2 against the C_T of rice genomic UBIQUITIN (gUBQ)³⁴. qPCR primers for MoPot2 and OsgUBQ can be found in Supplementary Data 4. Three biological replicates were performed for all inoculation experiments.

Yeast two-hybrid assays

The genes coding sequences were cloned into the yeast vectors pGADT7 and pGBKT7. The constructs were co-transformed into yeast strain AH109, and the corresponding transformants were incubated on SD-Leu/-Trp (DDO) and SD-Leu/-Trp/-His (TDO) or SD-Leu/-Trp/-His/-Ade (QDO) medium at 28 °C for 2-3 days. Primers used to generate the constructs are listed in Supplementary Data 4.

Co-IP assays

The desired combinations of constructs were co-transfected into rice protoplasts. Total protein was extracted with native buffer (50 mM Tris-MES, pH 8.0, 0.5 M sucrose, 1 mM MgCl₂, 10 mM EDTA, 5 mM DTT, and protease inhibitor cocktail). The supernatant was transferred to a new 1.5 mL centrifuge tube, to which 30 μL of protein G agarose beads coupled with anti-GFP antibody was added and slowly rotated at 4 °C for 4 h. The beads were collected by centrifugation at 500 g for 5 min at 4 °C, and the beads were washed with 1× PBST buffer at least three times.

BiFC assays

Full-length coding sequences of OsPUB20, OsGF14f, OsGF14c, OsRLCK185, and OsWRKY42 were individually cloned into the p2YN (nYFP) or p2YC (cYFP) vectors to produce the fusion to the N- or C-terminal of YFP, including OsGF14f-nYFP, OsGF14c-nYFP, OsRLCK185-nYFP, OsPUB20-cYFP, and OsWRKY42-cYFP. The desired plasmids were separately transformed into Agrobacterium strain EHA105 and then transiently infiltrated in N. benthamiana leaves. Images were captured with a Zeiss LSM 880 confocal laser microscope at 48 h after infiltration. Primers used to generate the constructs are listed in Supplementary Data 4.

In vitro E3 ligase activity and ubiquitination assay

The E3 ligase activity was done by following the method³³. Briefly, crude extract containing E1 (wheat E1), E2 (AtUBC8), 2 μg/μL ubiquitin, MBP-OsPUB20, or MBP-OsPUB20^C37A were mixed in 1×reaction buffer (50 mM Tris-HCl, pH 7.4, 10 mM MgCl₂, 5 mM ATP, and 2 mM DTT). The reactions were incubated at 30 °C for 2 h, and the ubiquitination activities of the proteins were determined using anti-Ub and anti-MBP antibodies. For the substrate ubiquitination assay, equal amounts of GST-OsGF14f and GST-OsGF14c were added to the reaction mixture, and ubiquitination was measured by immunoblotting.

Degradation assays in rice protoplasts

The rice protoplasts were transformed with the indicated combinations of plasmids, protoplasts were incubated with MG132 (50 μM), CHX (100 μM), and/or chitin (80 nM), and the total proteins were extracted using protein denaturing lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.5% NP-40, 4 M urea, and 1 mM PMSF)⁶⁰. Appropriate amounts of SDS loading buffer were added, and a western blot was performed to detect the protein levels.

Protein extraction and immunoblot analysis

The total proteins from 10-day-old rice plants were extracted in a denaturing buffer. After collecting the total protein extracts by centrifuge at 15,000 g for 10 min at 4°C, the protein extracts were mixed with 4×SDS loading buffer and separated by SDS-PAGE. Immunoblotting for OsGF14f and OsGF14c protein was performed with anti-OsGF14f (ABclonal, A20626) and anti-OsGF14c (ABclonal, A20583) antibodies according to the manufacturer’s instructions.

Nuclear-cytoplasmic fractionation assay

Nuclear-cytoplasmic fractionation assays were performed by following the method with slight modifications⁶¹. In brief, rice seedlings were harvested at the indicated time after infection and extracted in a lysis buffer (20 mM Tris-HCl, pH 7.5, 20 mM KCl, 2 mM EDTA, 2.5 mM MgCl₂, 25% glycerol, 250 mM sucrose, and 5 mM DTT) supplemented with a protease inhibitor cocktail. The crude extracts were filtered individually through double layers of Miracloth. The flow-through from each extract was spun at 1500 g for 10 min, and the supernatant (cytoplasmic fraction) was centrifuged at 1000 g for 10 min at 4 °C. The resulting pellet was rinsed four times with 5 mL of nuclear resuspension buffer NRBT (20 mM Tris-HCl, pH 7.4, 25% glycerol, 2.5 mM MgCl₂, and 0.2% Triton X-100) and then resuspended in 500 μL of NRB2 (20 mM Tris-HCl, pH 7.5, 250 mM sucrose, 10 mM MgCl₂, 0.5% Triton X-100, and 5 mM β-mercaptoethanol) supplemented with protease inhibitor cocktail. Each sample was carefully laid on top of 500 μL of NRB3 (20 mM Tris-HCl, pH 7.5,1.7 M sucrose, 10 mM MgCl₂, 0.5% Triton X-100, and 5 mM β-mercaptoethanol) supplemented with protease inhibitor cocktail and centrifuged at 16,000 g for 45 min at 4 °C. The final pellet was resuspended in 50 μL of lysis buffer. Samples with similar amounts of protein were loaded into the wells of SDS-PAGE gels for immunoblot analyses. HSP and histone H3 proteins were used as the cytoplasmic and nuclear markers, respectively.

In vitro phosphorylation assays

Recombinant GST-OsRLCK185, OsRLCK185^K108A, and His-OsPUB20, OsPUB20^T153A proteins were added to 30 μL of the kinase reaction buffer (50 mM Tris-HCl pH 7.5, 10 mM MgCl₂, 10 mM MnCl₂, 1 mM CaCl₂, 0.1 mM ATP, and 1 mM DTT) incubated at 30 °C for 1 h⁶². The reactions were stopped by adding 4×SDS loading buffer. The proteins were detected using anti-pSer/Thr and anti-His, or anti-GST antibodies.

In vivo phosphorylation assay

Rice protoplasts were co-expressed with cMyc-OsRLCK185, and OsPUB20-GFP, OsPUB20^T153A-GFP. Total proteins were extracted and were then subject to immunoprecipitation with anti-GFP. The phosphorylated OsPUB20 and OsPUB20^T153A were detected with anti-pSer/Thr. The input protein OsRLCK185 was probed with anti-cMyc. Actin was served as an internal control.

Liquid chromatography-tandem mass spectrometry (LC-MS/MS) assay

To prepare samples for identifying OsPUB20 phosphorylation sites by mass spectrometry, GST-OsRLCK185 and His-OsPUB20 proteins were added to 30 μL of the kinase reaction buffer incubated at 30 °C for 1 h. The total proteins were subjected to LC-MS/MS. To prepare samples for identifying OsGF14f-interacting proteins by mass spectrometry, OsGF14f was enriched and purified using GFP-trap magnetic agarose from the total protein extracts of OsGF14f-GFP transgenic seedlings inoculated with M. oryzae and subjected to LC-MS/MS.

LC-MS/MS analysis was performed by following the method with a slight modification⁵⁵. Raw data from the mass spectrometer were preprocessed with MaxQuant and the peak list was searched against Oryza_sativa. IRGSP-1.0 database. The search parameters: Fixed modifications: Carbamidomethyl (C); Variable modification: Oxidation (M) and Acetyl (Protein N-term); Digestion: Trypsin; First search peptide tolerance: 20 ppm; Main search peptide tolerance: 4.5 ppm; Max missed: 2.

Statistics and reproducibility

Statistical analysis was performed with GraphPad Prism 7.0. Data were shown as means ± SD and analyzed by two-tailed Student’s t test, and one-way ANOVA where appropriate. Different lowercase letters or asterisks (*p < 0.05; **p < 0.01; ns, no significance) representing significance levels are given in the figures. All experiments were repeated at least twice, and multiple biological replicates were used in each experiment. Details of statistical analyses including analytical methods, n values, and significance levels (p-values) can be found in the figures or figure legends.

Accession numbers

Sequence data from this article can be found in the Rice Genome Project database under the following accession numbers: OsRLCK185: LOC_Os05g30870, OsPUB20: LOC_Os06g13080, OsPUB19: LOC_Os06g13090, OsGF14f: LOC_Os03g50290, OsGF14c: LOC_ Os08g33370, OsWRKY42: LOC_ Os02g26430.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

C.Z. and Y.N. designed the experiments. C.Z., P.S., R.W., F.H., H.T., D.W., J.W., L.F., Z.H., X.Y., and W. L. conducted the experiments. C.Z., R.W., G.-L.W., and Y.N. wrote the manuscript.

Peer review

Peer review information

Nature Communications thanks Tsutomu Kawasaki and Jian Zhang for their contribution to the peer review of this work. A peer review file is available.

Data availability

The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium (https://proteomecentral.proteomexchange.org) via the iProX partner repository with the dataset identifier PXD073757 and PXD073756. All data supporting the findings of this work are available in the paper, Supplementary Information files, and repository platform. Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-69627-2.