OsMKK6 Regulates Disease Resistance in Rice

Mitogen-activated protein kinase cascades play important roles in various biological programs in plants, including immune responses, but the underlying mechanisms remain elusive. Here, we identified the lesion mimic mutant rsr25 (rust spots rice 25) and determined that the mutant harbored a loss-of-function allele for OsMKK6 (MITOGEN-ACTIVATED KINASE KINASE 6). rsr25 developed reddish-brown spots on its leaves at the heading stage, as well as on husks. Compared to the wild type, the rsr25 mutant exhibited enhanced resistance to the fungal pathogen Magnaporthe oryzae (M. oryzae) and to the bacterial pathogen Xanthomonas oryzae pv. oryzae (Xoo). OsMKK6 interacted with OsMPK4 (MITOGEN-ACTIVATED KINASE 4) in vivo, and OsMKK6 phosphorylated OsMPK4 in vitro. The Osmpk4 mutant is also a lesion mimic mutant, with reddish-brown spots on its leaves and husks. Pathogen-related genes were significantly upregulated in Osmpk4, and this mutant exhibited enhanced resistance to M. oryzae compared to the wild type. Our results indicate that OsMKK6 and OsMPK4 form a cascade that regulates immune responses in rice.


Introduction
Plants lack the ability to flee from pathogen attack and have, therefore, evolved a twolayered immune system to defend themselves against pathogens. The first layer of their immune system is known as pathogen-associated molecular pattern-triggered immunity (PAMP-triggered immunity, PTI). Pattern recognition receptors (PRRs) located on the cell membrane recognize pathogen-associated molecular patterns (PAMPs) and activate a series of immune responses [1]. In turn, pathogens have evolved a class of effectors, which are secreted by pathogens into plant cells and inhibit the plant PTI immune response by attacking and suppressing the PTI signaling pathway. In response, R (Resistance) proteins in plants directly or indirectly monitor effectors and trigger immune responses called effector-triggered immunity (ETI) [2]. ETI is more rapid and induces a stronger immune response than PTI, with cell death often occurring at the site of infection, a phenomenon known as the hypersensitive response (HR) [3]. The HR is an effective way for plants to defend themselves against pathogens and is accompanied by programmed cell death (PCD), which limits the spread of pathogens to non-infected tissues [4].
Lesion mimic mutants (LMMs) spontaneously produce HR-like lesions accompanied by PCD [5]. Therefore, LMMs are ideal materials to elucidate the molecular mechanism of PCD and defense responses in plants. Many LMMs have been identified in various plant species, such as Arabidopsis (Arabidopsis thaliana) [6,7], maize (Zea mays) [8], wheat (Triticum aestivum) [9], barley (Hordeum vulgare) [10], and rice (Oryza sativa) [11][12][13]. Rice is a primary calorie provider for approximately one-half of people across the world [14]. Different fungal, bacterial, and viral pathogens cause large reductions in rice grain production. Magnaporthe oryzae (M. oryzae), a filamentous ascomycete fungus, is the causal agent of rice blast disease, the most destructive disease of rice, which can cause a 10-30% reduction in rice yields [15] Xanthomonas oryzae pv. oryzae (Xoo) is a bacterial disease that with OsMPK4. The Osmpk4 mutant exhibited an LMM phenotype and enhanced resistance to M. oryzae compared to wild type plants. Therefore, we identified the phenotype of a loss-of-function mutant of OsMKK6 and demonstrated that the OsMKK6-OsMPK4 cascade is involved in regulating the resistance of rice to the fungal pathogen M. oryzae. Our finding that the OsMKK6-OsMPK4 cascade regulates resistance to both fungal and bacterial pathogens underscores its importance in broad-spectrum disease resistance.

Phenotypic Characterization of the rsr25 Mutant
We identified the rsr25 mutant from an EMS mutant library in the Xiushui134 (XS134) background. Compared to the wild type (WT), rsr25 leaves developed reddish-brown spots at the heading stage in the absence of pathogen attack. The spots spread from the leaf tip to the leaf base and from old leaves to new leaves until they were distributed throughout the plant at the mature stage ( Figure 1A,B). Reddish-brown spots also appeared on the husks of the rsr25 mutant ( Figure 1C). We measured the agronomic traits of WT and rsr25 plants grown in soil, finding that rsr25 had a lower plant height and seedsetting rate, lower 1000-grain weight, shorter panicles, and fewer panicles than WT plants ( Figure 1D-H). These results suggest that the rsr25 mutant exhibits an LMM phenotype and that the formation of the lesion spots seriously affects the growth and development of rsr25 compared to the WT.
Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 3 of with OsMPK4. The Osmpk4 mutant exhibited an LMM phenotype and enhanced r sistance to M. oryzae compared to wild type plants. Therefore, we identified the phenotyp of a loss-of-function mutant of OsMKK6 and demonstrated that the OsMKK6-OsMPK cascade is involved in regulating the resistance of rice to the fungal pathogen M. oryza Our finding that the OsMKK6-OsMPK4 cascade regulates resistance to both fungal an bacterial pathogens underscores its importance in broad-spectrum disease resistance.

Phenotypic Characterization of the rsr25 Mutant
We identified the rsr25 mutant from an EMS mutant library in the Xiushui134 (XS13 background. Compared to the wild type (WT), rsr25 leaves developed reddish-brow spots at the heading stage in the absence of pathogen attack. The spots spread from th leaf tip to the leaf base and from old leaves to new leaves until they were distribute throughout the plant at the mature stage ( Figure 1A,B). Reddish-brown spots also a peared on the husks of the rsr25 mutant ( Figure 1C). We measured the agronomic traits WT and rsr25 plants grown in soil, finding that rsr25 had a lower plant height and see setting rate, lower 1000-grain weight, shorter panicles, and fewer panicles than WT plan ( Figure 1D-H). These results suggest that the rsr25 mutant exhibits an LMM phenotyp and that the formation of the lesion spots seriously affects the growth and developme of rsr25 compared to the WT.

OsRSR25 Encodes OsMKK6, a MAP Kinase Kinase
To identify the gene responsible for the phenotypes of the rsr25 mutant, we crosse rsr25 with XS134 plants to generate a segregating F2 population for MutMap sequencin The F1 progeny exhibited the same phenotype as the WT XS134, and the segregation rat

OsRSR25 Encodes OsMKK6, a MAP Kinase Kinase
To identify the gene responsible for the phenotypes of the rsr25 mutant, we crossed rsr25 with XS134 plants to generate a segregating F 2 population for MutMap sequencing. The F 1 progeny exhibited the same phenotype as the WT XS134, and the segregation ratio of normal to lesion mimic individuals in the F 2 population was approximately 3:1 (Table S1), suggesting that a single recessive nuclear gene is responsible for the rsr25 phenotype. MutMap sequencing of F 2 individuals with WT and mutant phenotypes revealed a missense mutation (G to A, 205 bp downstream from the start codon of the coding sequence) in the third exon of OsMKK6 (Os01g0510100) (Figure 2A-C), which caused an aspartate (Asp, D) to asparagine (Asn, N) change at amino acid residue 69 of OsMKK6 in rsr25 ( Figure S1). Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 4 of normal to lesion mimic individuals in the F2 population was approximately 3:1 (T S1), suggesting that a single recessive nuclear gene is responsible for the rsr25 phenot MutMap sequencing of F2 individuals with WT and mutant phenotypes revealed a sense mutation (G to A, 205 bp downstream from the start codon of the coding seque in the third exon of OsMKK6 (Os01g0510100) (Figure 2A-C), which caused an aspa (Asp, D) to asparagine (Asn, N) change at amino acid residue 69 of OsMKK6 in rsr25 ( ure S1). To further test whether the mutation in OsMKK6 is responsible for the rsr25 mu phenotype, we constructed the complementation (COM) plasmid proOsMKK6 MKK6CDS-GFP, which includes the full-length coding sequence of OsMKK6 driven b native promoter. We obtained 15 COM transgenic lines in the rsr25 mutant backgrou The phenotype of all COM plants was identical to that of WT plants, with no lesion formation on their leaves or husks ( Figure 2D-F). These results demonstrate that the tation in OsMKK6 is responsible for the rsr25 phenotype. To further test whether the mutation in OsMKK6 is responsible for the rsr25 mutant phenotype, we constructed the complementation (COM) plasmid proOsMKK6:OsMKK6CDS-GFP, which includes the full-length coding sequence of OsMKK6 driven by its native promoter. We obtained 15 COM transgenic lines in the rsr25 mutant background. The phenotype of all COM plants was identical to that of WT plants, with no lesion spot formation on their leaves or husks ( Figure 2D-F). These results demonstrate that the mutation in OsMKK6 is responsible for the rsr25 phenotype.

OsMKK6 Regulates Immunity against Magnaporthe oryzae and Xanthomonas oryzae pv. oryzae
To test the role of OsMKK6 in Xanthomonas oryzae pv. oryzae (Xoo) resistance, we inoculated WT and rsr25 plants at 30 dps (days post-sowing) with Xoo isolate P6. At 14 days post-inoculation (dpi), leaf chlorosis was less severe in rsr25 than in WT plants, and rsr25 leaves had produced much shorter lesions than WT leaves ( Figure 3A-C). We also tested the possible role of OsMKK6 in immunity against Magnaporthe oryzae (M. oryzae). Accordingly, we inoculated leaves of WT and rsr25 plants at 30 dps with M. oryzae isolate RB22 via punch inoculation, finding that rsr25 contained shorter lesions than the WT ( Figure 3D,E). These results suggest that OsMKK6 participates in the regulation of plant immunity against both Xoo and M. oryzae.

OsMKK6 Regulates Immunity against Magnaporthe oryzae and Xanthomonas oryzae pv. oryzae
To test the role of OsMKK6 in Xanthomonas oryzae pv. oryzae (Xoo) resistance, we inoculated WT and rsr25 plants at 30 dps (days post-sowing) with Xoo isolate P6. At 14 days post-inoculation (dpi), leaf chlorosis was less severe in rsr25 than in WT plants, and rsr25 leaves had produced much shorter lesions than WT leaves ( Figure 3A-C). We also tested the possible role of OsMKK6 in immunity against Magnaporthe oryzae (M. oryzae). Accordingly, we inoculated leaves of WT and rsr25 plants at 30 dps with M. oryzae isolate RB22 via punch inoculation, finding that rsr25 contained shorter lesions than the WT ( Figure  3D,E). These results suggest that OsMKK6 participates in the regulation of plant immunity against both Xoo and M. oryzae.

OsMKK6 Regulates Pathogen-Induced Defense Responses
To explore the role of OsMKK6 in plant immunity, we analyzed the expression levels of pathogenesis-related (PR) genes PATHOGENESIS-RELATED PROTEIN 1a (OsPR1a) and OsPR1b in WT and rsr25 plants with or without Xoo and M. oryzae inoculation. In the absence of pathogen inoculation, the relative expression levels of OsPR1a and OsPR1b were significantly higher in rsr25 than in WT plants ( Figure 4A,B). At 48 h post-inoculation (hpi) with Xoo, the relative expression level of OsPR1a increased 40-to 80-fold in WT plants compared to uninoculated WT plants. By contrast, the relative expression level of OsPR1a increased 100-fold in rsr25 compared to the untreated rsr25 mutant. The relative expression level of OsPR1a continued to increase at 96 hpi and 120 hpi with Xoo and was

OsMKK6 Regulates Pathogen-Induced Defense Responses
To explore the role of OsMKK6 in plant immunity, we analyzed the expression levels of pathogenesis-related (PR) genes PATHOGENESIS-RELATED PROTEIN 1a (OsPR1a) and OsPR1b in WT and rsr25 plants with or without Xoo and M. oryzae inoculation. In the absence of pathogen inoculation, the relative expression levels of OsPR1a and OsPR1b were significantly higher in rsr25 than in WT plants ( Figure 4A,B). At 48 h post-inoculation (hpi) with Xoo, the relative expression level of OsPR1a increased 40-to 80-fold in WT plants compared to uninoculated WT plants. By contrast, the relative expression level of OsPR1a increased 100-fold in rsr25 compared to the untreated rsr25 mutant. The relative expression level of OsPR1a continued to increase at 96 hpi and 120 hpi with Xoo and was significantly higher in rsr25 than in WT plants ( Figure 4A). OsPR1b was expressed at significantly higher levels in rsr25 than the WT both before and after inoculation with Xoo, with a lower Xoo-induced upregulation in rsr25 than in WT ( Figure 4A), likely due to the higher basal expression level in rsr25. Consistent with the response to Xoo inoculation, OsPR1a and OsPR1b were significantly upregulated in rsr25 compared to WT plants at 24 hpi with M. oryzae ( Figure 4B). significantly higher in rsr25 than in WT plants ( Figure 4A). OsPR1b was expressed at sig nificantly higher levels in rsr25 than the WT both before and after inoculation with Xoo with a lower Xoo-induced upregulation in rsr25 than in WT ( Figure 4A), likely due to the higher basal expression level in rsr25. Consistent with the response to Xoo inoculation OsPR1a and OsPR1b were significantly upregulated in rsr25 compared to WT plants at 24 hpi with M. oryzae ( Figure 4B). OsMPK3 and OsMPK6 regulate the immune responses of rice to pathogens [18]. Therefore, we examined the phosphorylation levels of OsMPK3 and OsMPK6 in WT and rsr25 plants before and after inoculation with Xoo. While there was no significant difference in the phosphorylation level of OsMPK3 before or after inoculation with Xoo, the phosphorylation level of OsMPK6 was significantly higher in rsr25 than in the WT at 96 hpi and 120 hpi with Xoo ( Figure 4C). These results indicate that the pathogen-induced immune response is stronger in rsr25 than the WT, suggesting that OsMKK6 plays an important role in regulating the immune response in rice.
To investigate the role of OsMKK6 in PTI, we analyzed flg22-induced ROS bursts in WT and rsr25 plants. After flg22 treatment, the peak ROS level was nearly 2.7 times higher in rsr25 than in the WT. In the control (treated with H 2 O), the basal ROS level was 2.1 times higher in rsr25 than in the WT ( Figure 4D). These results indicate that OsMKK6 regulates flg22-induced PTI in rice.

OsMKK6 Interacts with OsMPK4
Arabidopsis MKK6 and MPK4 function in a MAPK cascade to prevent the autoactivation of immunity [30]. In our phylogenetic tree, OsMKK6 is in the same cluster as AtMKK6, while OsMPK4 is in the same cluster as AtMPK4 ( Figure S2). To determine whether OsMKK6 and OsMPK4 function in a MAPK cascade, we tested their physical interaction by bimolecular fluorescence complementation (BiFC) assays in Nicotiana benthamiana leaf epidermal cells. Cells co-expressing YN-OsMPK4 and YC-OsMPK6 showed strong yellow fluorescent protein (YFP) signals, which overlapped with the nucleus-localized H2B-mCherry signal ( Figure 5A). To further test the interaction between OsMKK6 and OsMPK4, we performed a co-immunoprecipitation (Co-IP) assay. GFP-tagged OsMKK6, but not GFP, immunoprecipitated FLAG-tagged OsMPK4 when we infiltrated the encoding constructs in N. benthamiana leaves ( Figure 5B). We also performed firefly luciferase complementation imaging (LCI) assays in N. benthamiana. We detected strong luminescence in N. benthamiana leaves co-expressing Nluc-OsMKK6 and Cluc-OsMPK4 ( Figure 5C). These results indicate that OsMKK6 interacts with OsMPK4 in vivo.
To investigate whether the mutation of OsMKK6 in rsr25 affects its interaction with OsMPK4 and whether replacing the 69th amino acid residue with other types of amino acid residues would affect the interaction of OsMKK6 with OsMPK4, we performed an LCI assay to assess the interaction between Cluc-OsMPK4 and Nluc-OsMKK6X (OsMKK6m [D69N], OsMKK6n [D69Q], and OsMKK6u [D69E]). OsMPK4 interacted only with the non-mutated form of OsMKK6, whereas OsMKK6 failed to interact with OsMPK4 after changing the D69 amino acid residue to N, Q, or E ( Figure 6A). OsMKK6 contains a protein kinase domain ( Figure 6B). To delineate the interaction interface between OsMKK6 and OsMPK4, we truncated OsMKK6 into OsMKK6-N (amino acids [aa] 1-70) and OsMKK6-C (aa 71-355) based on the position of its protein kinase domain ( Figure 6B) and performed LCI assays to examine the interaction between Cluc-OsMPK4 and Nluc-OsMKK6-N or Nluc-OsMKK6-C. We detected a strong luminescent signal in N. benthamiana leaves coexpressing Cluc-OsMPK4 and Nluc-OsMKK6-N, but we observed no luminescent signal in leaves co-expressing Cluc-OsMPK4 and Nluc-OsMKK6-C ( Figure 6C). These results suggest that OsMKK6 interacts with OsMPK4 through its N terminus (aa 1-70) and that the amino acid residue D69 plays an important role in this interaction. To investigate whether the mutation of OsMKK6 in rsr25 affects its interaction with OsMPK4 and whether replacing the 69th amino acid residue with other types of amino acid residues would affect the interaction of OsMKK6 with OsMPK4, we performed an LCI assay to assess the interaction between Cluc-OsMPK4 and Nluc-OsMKK6X (Os-MKK6m [D69N], OsMKK6n [D69Q], and OsMKK6u [D69E]). OsMPK4 interacted only with the non-mutated form of OsMKK6, whereas OsMKK6 failed to interact with Os-MPK4 after changing the D69 amino acid residue to N, Q, or E ( Figure 6A). OsMKK6 contains a protein kinase domain ( Figure 6B). To delineate the interaction interface between OsMKK6 and OsMPK4, we truncated OsMKK6 into OsMKK6-N (amino acids [aa] 1-70) and OsMKK6-C (aa 71-355) based on the position of its protein kinase domain ( Figure 6B) The OsMKK6 interacts with OsMPK4 in Co-IP assay. GFP-tagged OsMKK6 or GFP was transiently co-expressed with FLAG-tagged OsMPK4 in N. benthamiana. Immunoprecipitation was carried out using anti-GFP beads. Total proteins and immunoprecipitated proteins were analyzed using anti-FLAG or anti-GFP antibodies. (C) OsMKK6 interacts with OsMPK4 as indicated through LCI assay. Nluc-OsMKK6 and CLuc-OsMPK4 were transiently expressed in N. benthamiana by coinfiltration; Nluc and Cluc were the negative controls. Luminescence was monitored at 48 h after Agrobacterium tumefaciens infection. Scale bar = 2 cm. and performed LCI assays to examine the interaction between Cluc-OsMPK4 and Nluc-OsMKK6-N or Nluc-OsMKK6-C. We detected a strong luminescent signal in N. benthamiana leaves co-expressing Cluc-OsMPK4 and Nluc-OsMKK6-N, but we observed no luminescent signal in leaves co-expressing Cluc-OsMPK4 and Nluc-OsMKK6-C ( Figure 6C). These results suggest that OsMKK6 interacts with OsMPK4 through its N terminus (aa 1-70) and that the amino acid residue D69 plays an important role in this interaction.

OsMKK6 Phosphorylates OsMPK4
We performed in vitro phosphorylation assays using non-radioactive ATP to determine whether OsMKK6 can phosphorylate OsMPK4. We established that recombinant purified OsMKK6 and OsMPK4 are autophosphorylated in the presence of ATPγS and pnitrobenzyl mesylate (PNBM) in vitro ( Figure 7A). To avoid interference from OsMPK4 autophosphorylation, we purified MBP-OsMPK4m (harboring the D187A mutation) with a mutation in the kinase active site. MBP-OsMPK4m was not autophosphorylated, but GST-OsMKK6 phosphorylated this protein in vitro ( Figure 7A). In addition, we purified a constitutively active form of OsMKK6, GST-OsMKK6 DD (S221D, T227D). Recombinant purified GST-OsMKK6 DD had a stronger ability to phosphorylate MBP-OsMPK4m compared to GST-OsMKK6 ( Figure 7B). These results indicate that OsMKK6 can phosphorylate Os-MPK4 in vitro.

OsMKK6 Phosphorylates OsMPK4
We performed in vitro phosphorylation assays using non-radioactive ATP to determine whether OsMKK6 can phosphorylate OsMPK4. We established that recombinant purified OsMKK6 and OsMPK4 are autophosphorylated in the presence of ATPγS and p-nitrobenzyl mesylate (PNBM) in vitro ( Figure 7A). To avoid interference from OsMPK4 autophosphorylation, we purified MBP-OsMPK4m (harboring the D187A mutation) with a mutation in the kinase active site. MBP-OsMPK4m was not autophosphorylated, but GST-OsMKK6 phosphorylated this protein in vitro ( Figure 7A). In addition, we purified a constitutively active form of OsMKK6, GST-OsMKK6 DD (S221D, T227D). Recombinant purified GST-OsMKK6 DD had a stronger ability to phosphorylate MBP-OsMPK4m compared to GST-OsMKK6 ( Figure 7B). These results indicate that OsMKK6 can phosphorylate OsMPK4 in vitro.

The Osmpk4 Mutant Exhibits an LMM Phenotype and Enhanced Resistance to M. oryzae
To investigate the role of OsMPK4 in the immune response of rice, we examined the phenotype of the OsMPK4 loss-of-function mutant Osmpk4 (R89K mutation in OsMPK4 protein) [32]. This mutant developed reddish-brown lesions on its leaves and husks, as previously reported ( Figure 8A,B). Suppressing or knocking out of OsMPK4 enhances resistance to Xoo in rice [35,36]. Our findings show Osmpk4 also exhibits enhanced resistance to M. oryzae compared to WT ( Figure 8C,D). To investigate the mechanism underlying the enhanced resistance of Osmpk4, we examined the transcript levels of PR genes in the mutant via RT-qPCR. OsPR1a, OsPR1b, PROBENAZOLE-INDUCIBLE1 (OsPBZ1), ALLENE OXIDE SYNTHASE 2 (OsAOS2), and PHENYLALANINE AMMONIA-LYASE 4 (OsPAL4) were expressed at significantly higher levels in rsr25 than in WT plants ( Figure 8E). These results suggest that, the same as OsMKK6, OsMPK4 regulates immunity against both Xoo and M. oryzae in rice. Taken together, our findings indicate that OsMKK6 and OsMPK4 form a cascade that regulates immune responses in rice.

The Osmpk4 Mutant Exhibits an LMM Phenotype and Enhanced Resistance to M. or
To investigate the role of OsMPK4 in the immune response of rice, we examin phenotype of the OsMPK4 loss-of-function mutant Osmpk4 (R89K mutation in O protein) [32]. This mutant developed reddish-brown lesions on its leaves and hu previously reported ( Figure 8A,B). Suppressing or knocking out of OsMPK4 enhan sistance to Xoo in rice [35,36]. Our findings show Osmpk4 also exhibits enhanced res to M. oryzae compared to WT ( Figure 8C,D). To investigate the mechanism underly enhanced resistance of Osmpk4, we examined the transcript levels of PR genes in t tant via RT-qPCR. OsPR1a, OsPR1b, PROBENAZOLE-INDUCIBLE1 (OsPBZ1), A OXIDE SYNTHASE 2 (OsAOS2), and PHENYLALANINE AMMONIA-LYASE 4 (OsPAL4) were expressed at significantly higher levels in rsr25 than in WT plants ( Figure 8E). These results suggest that, the same as OsMKK6, OsMPK4 regulates immunity against both Xoo and M. oryzae in rice. Taken together, our findings indicate that OsMKK6 and OsMPK4 form a cascade that regulates immune responses in rice.

Discussion
Plant LMMs spontaneously produce HR-like lesions in the absence of pathogen attack; most of these mutants have improved resistance to pathogens [5]. Many LMM genes have been cloned in plants and shown to encode proteins involved in various aspects of the immune response, such as the ROS burst and signaling pathways of the phytohormones salicylic acid, jasmonate, and ethylene or components directly involved in the immune response [37]. For example, rice lesion-mimic mutant genes SPL11 and SDS2 are involved in the regulation of Rac1-mediated ROS bursts process [22,23]. Plants defend themselves against pathogen invasion through the PTI and ETI systems. ETI evolved in plants to detect pathogen effectors and initiate defense response. ETI is mediated by the host resistance (R) genes and recognition of pathogen effectors by R proteins can be either direct or indirect [38]. In Arabidopsis, R proteins are encoded by approximately 150 genes, and are categorized according to the structural domains they contain. R proteins generally have a very conserved protein structural domain, the leucine repeat sequence domain (LRR structural domain) [39], and many LMM are deficient for R proteins. The suppressor of SA insensitivity 4 (ssi4) LMM is a gain-of-function mutant affected in a TIR-NBS-LRR factor, and constitutive activation of the SSI4 protein activates SA signaling pathways and

Discussion
Plant LMMs spontaneously produce HR-like lesions in the absence of pathogen attack; most of these mutants have improved resistance to pathogens [5]. Many LMM genes have been cloned in plants and shown to encode proteins involved in various aspects of the immune response, such as the ROS burst and signaling pathways of the phytohormones salicylic acid, jasmonate, and ethylene or components directly involved in the immune response [37]. For example, rice lesion-mimic mutant genes SPL11 and SDS2 are involved in the regulation of Rac1-mediated ROS bursts process [22,23]. Plants defend themselves against pathogen invasion through the PTI and ETI systems. ETI evolved in plants to detect pathogen effectors and initiate defense response. ETI is mediated by the host resistance (R) genes and recognition of pathogen effectors by R proteins can be either direct or indirect [38]. In Arabidopsis, R proteins are encoded by approximately 150 genes, and are categorized according to the structural domains they contain. R proteins generally have a very conserved protein structural domain, the leucine repeat sequence domain (LRR structural domain) [39], and many LMM are deficient for R proteins. The suppressor of SA insensitivity 4 (ssi4) LMM is a gain-of-function mutant affected in a TIR-NBS-LRR factor, and constitutive activation of the SSI4 protein activates SA signaling pathways and induces the formation of chlorotic lesions [40,41]. Three R proteins of the TIR-NBS-LRR family, ACTIVATED DISEASE RESISTANCE 1 (ADR1), ADR1-LIKE 1 (ADR1-L1), and ADR1-LIKE 2 (ADR1-L2) are activators of defense responses and cell death. Loss of function of all three genes inhibits lesion formation of lesion simulating disease1 (lsd1) [41,42]. NahG transgenic plants that fail to accumulate salicylic acid were crossed with multiple LMMs to determine the role of SA in immune signaling. In the presence of NahG, spontaneous lesion formation was suppressed in the lesion mimic mutants lesion simulating disease 6 (lsd6), lsd7, accelerated cell death 5 (acd5), acd6, acd11, constitutive expressor of PR genes 22 (cpr22), ssi1, and disease-like lesions 1 (dll1), suggesting that SA plays an important role in spot-like formation [43][44][45][46][47][48][49]. Therefore, LMMs are ideal materials for studying the immune responses of plants. In this study, we identified the LMM rsr25, which displays reddish-brown lesions on its leaves and husks ( Figure 1A-C). The rsr25 mutant exhibited poor agronomic traits, including shorter plants and panicles, and lower thousand-grain weight, seed-setting rate, and panicle number compared to WT plants ( Figure 1D-H). MutMap sequencing and complementation assays identified the responsible mutated gene as being OsMKK6 (Os01g0510100) (Figure 2). The mutation of OsMKK6 caused the LMM phenotype of rsr25, suggesting that OsMKK6 might regulate the immune response in rice.
During PTI, the activation of PRRs located on the cell membrane triggers a series of immune signal transduction steps including the rapid phosphorylation of RLCKs, a burst of ROS, and activation of MAPK cascades [50]. MAPK cascades transduce immune signals to a wide range of downstream immune receptors [51]. Defects in MAPK cascade components can lead to altered immune responses in plants. In Arabidopsis, the MEKK1-MKK1/2-MPK4 cascade is activated during PTI [26,27], and defects in this cascade result in autoimmunity [28,29]. In the current study, rsr25, a loss-of-function mutant of OsMKK6, exhibited an LMM phenotype, along with increased resistance to M. oryzae and Xoo compared to the WT (Figure 3). The upregulated expression of PR genes and flg22-induced ROS production was significantly higher in rsr25 than in WT plants in response to pathogen infection (Figure 4), suggesting that OsMKK6 might play a key role in PTI. During plant immunity, effectors secreted by pathogens can increase plant susceptibility by inhibiting PTI signaling. OsMPK4 can be dephosphorylated by SCRE6, a phosphatase effector secreted by Ustilaginoidea virens, which enhances OsMPK4 stability and, thus, inhibits the immune response in rice [33]. The OsMKK6-OsMPK4 cascade, therefore, plays a vital role in PTI. Whether the OsMKK6-OsMPK4 cascade interacts with other pathogen effectors should be explored to determine whether this cascade is involved in broad-spectrum disease resistance in rice.
The AtMKK6-AtMPK4 cascade negatively regulates the immune response in Arabidopsis [30]. OsMKK6 is a homolog of AtMKK6, and OsMPK4 is a homolog of AtMPK4 ( Figures S1 and S2). OsMKK6 interacts with OsMPK4 in vivo and phosphorylates OsMPK4 in vitro ( Figures 5 and 7), suggesting that the regulation of immunity by this MAPK cascade is conserved between rice and Arabidopsis. MPKs regulate plant immunity by phosphorylating a wide range of target proteins [36]. In Arabidopsis, AtMKK6 functions simultaneously with AtMKK1 and AtMKK2 to form a MAPK cascade with AtMPK4 to prevent the activation of SUMM2 (SUPPRESSOR OF MKK1 MKK2)-mediated immunity. AtMKK6 also functions with ANP2, ANP3, and MPK4 in a separate MAPK cascade to inhibit the PHYTOALEXIN DEFICIENT 4 (PAD4)-dependent defense response, suggesting that MPK4 regulates plant immunity by targeting different substrate proteins [30].
Whether the OsMKK6-OsMPK4 cascade positively or negatively regulates resistance to pathogens in rice is controversial. OsMPK4 was first reported to negatively regulate resistance to Xoo in rice, as transgenic lines with a knockout or RNAi interference (RNAi) for OsMPK4 exhibited enhanced resistance to Xoo compared to WT plants [36]. The R89K mutation of OsMPK4 improved plant resistance to Xoo, and transgenic lines overexpressing OsMPK4 were more susceptible to Xoo than the WT [32]. Furthermore, Ustilaginoidea virens secretes the phosphatase SCRE6, which stabilizes OsMPK4 to suppress plant immunity [33]. These observations suggest that OsMPK4 negatively regulates rice resistance to pathogens. Notably, other studies have suggested that OsMPK4 functions in two layers of the rice-Xoo interaction. Both OsMPK4 knockout and OsMPK4-overexpressing plants showed enhanced resistance to Xoo [35,36]. During the course of our study, Li et al. (2021) independently demonstrated that the OsMKK6-OsMPK4 cascade positively regulated the resistance of rice to Xoo by targeting OsVQ14 and OsVQ32: OsMKK6-RNAi plants showed no difference in resistance to Xoo from the WT, whereas OsMPKK6-overexpressing transgenic plants showed improved resistance to Xoo [34]. The difference in resistance to Xoo between rsr25 and OsMKK6-RNAi transgenic plants is likely due to the presence of the D69N mutation in MKK6 in rsr25, which has a more severe effect on the function of OsMKK6 than the mutation employed by Li et al. (2021). We also determined that the loss of function in OsMKK6 or OsMPK4 increased the resistance of rice to M. oryzae ( Figure 3D,E and Figure 8C,D), suggesting that the OsMKK6-OsMPK4 cascade regulates the resistance of rice to the fungal pathogen M. oryzae. Based on our current findings, it is not possible to determine whether OsMKK6-OsMPK4 positively or negatively regulates the immune response in rice. ENHANCED DISEASE RESISTANCE 1 (OsEDR1) negatively regulates the immune response in rice by suppressing the OsMPKK10.2-OsMPK6 cascade. The Ossedr1 mutant shows an LMM phenotype, accumulates large amounts of H 2 O 2 in its leaves, and shows increased resistance to Xanthomonas oryzae pv. oryzicola (Xoc) compared to the WT. The Osedr1 Osmpk6 double mutant has no lesion spots on its leaves and is more susceptible to Xoc, suggesting that OsMPK6 positively regulates the resistance of rice to Xoc [21]. OsMPK6 positively regulates rice immunity by targeting OsWRKY45 [52]. In the current study, the phosphorylation level of OsMPK6 was higher in rsr25 than in WT plants, and the Xoo-induced increase in OsMPK6 phosphorylation was significantly higher in rsr25 than in WT plants at 48 h and 96 h post-inoculation ( Figure 4C). These results indicate that MPK6 is activated when the OsMKK6-OsMPK4 cascade is defective. There are 8 putative MKK and 15 putative MPK genes in the rice genome [38], suggesting that MKKs likely activate multiple MPKs; the crosstalk between various immune signaling pathways in rice might occur at this level. After receiving external signals, plant MAPKKKs mostly phosphorylate the two conserved serine (S) and threonine (T) residues in the S/T-X5-S/T (X is any amino acid) motif of MKKs and activate MKKs. The activated MKKs phosphorylate both the threonine (T) and the tyrosine (Y) in the T-D-Y or T-E-Y motif of MPKs and activate MPKs [53]. The genetic compensation response, which was first reported in zebrafish (Danio rerio), describes how the knockout of one gene has no phenotypic effect due to genetic compensation by other homologous genes whose expression is stimulated [54,55]. We propose that, if the OsMKK6-OsMPK4 cascade is defective, an enhanced genetic compensation response may occur in rice as well, causing another cascade to be activated. Whether the formation of lesion spots in the rsr25 mutant results from the activation of OsMPK6 requires further study, such as examining whether the loss of function of OsMPK6 in the rsr25 mutant can suppress its phenotype. The next steps are to verify whether OsMPK6 is also activated in osmpk4 mutant by in vivo phosphorylation assay and to develop Osmkk6 Osmpk6 and Osmpk4 Osmpk6 double mutant materials to further investigate the relationship between the OsMKK6-OsMPK4 cascade and OsMPK6 in regulating the immune response of rice through phenotypic observation.
In conclusion, we examined the phenotype of an OsMKK6 loss-of-function mutant. We demonstrated that OsMKK6 plays a vital role in the resistance of rice to pathogens. We also demonstrated that the OsMKK6-OsMPK4 cascade is involved in regulating the resistance of rice to the fungal pathogen M. oryzae, providing new clues for further studying the roles of MAPK cascades in rice immunity.

Plant Materials and Growth Conditions
The rsr25 mutant was isolated from an ethyl methanesulfonate (EMS) mutant library of the rice (Oryza sativa) cultivar Xiushui134 (XS134, wild type [WT]). The rsr25 mutant was crossed with XS134 to obtain an F 1 population. The F 2 population was obtained by selfing F 1 plants and was subjected to MutMap sequencing. The Osmpk4 mutant and its corresponding wild type (Chang Geng 3, CG3) were a gift from the China National Rice Research Institute; the loss-of-function Osmpk4 mutant was described previously [29]. Plants from the F 2 population were grown in Hainan, China (N:18 • , E:109 • ). The agronomic traits of wild type and rsr25 plants were measured in Changxing (N:30 • , E:119 • ). Plants used for phenotypic analysis and molecular biology experiments were grown in a growth chamber at 30 • C/25 • C (day/night) and 60% to 70% relative humidity, with light at a photon density of 300 µmol m −2 s −1 supplied by light bulbs and a photoperiod of 12 h light/12 h dark. The plants were grown in Kimura nutrient solution [56], which was changed once per week.

Vector Construction
To construct the complementation vector proOsMKK6:OsMKK6CDS-GFP, the promoter sequence of OsMKK6 (3000 bp upstream of ATG) was amplified from XS134 genome DNA, and the CDS (Coding sequence) of OsMKK6 (with the stop codon removed) was amplified from the XS134 cDNA. Those two sequences were amplified together using overlapping extension PCR and ligated into the pCAMBIAI1300-GFP vector. The proOsMKK6:OsMKK6CDS-GFP vector was introduced into rsr25 mutant by Agrobacterium tumefaciens-mediated transformation.
To construct the 35S:OsMKK6-GFP vector, the CDS of OsMKK6 was ligated into the pCAMBIAI1300-35S-GFP vector.
To construct the GST-OsMKK6 vector, the CDS of OsMKK6 was ligated into the pGEX-4T-1 vector. The point mutation in OsMKK6 (the 221st serine residue and 227th threonine residue were replaced with aspartate residue) was introduced using an overlap extension PCR, and then the mutated OsMKK6 CDS sequence was ligated into pGEX-4T-1 vector to generate GST-OsMKK6 DD vector. To generate MBP-OsMPK4 and MBP-OsMPK4m vectors, the CDS of OsMPK4 and mutated OsMPK4 CDS sequence were ligated into the MBP-pET-28a vector. The point mutation in OsMPK4 (the 187th aspartate residue was replaced with alanine residue) was introduced using an overlap extension PCR. The CDS of OsMPK4 was ligated into the PTCK303-FLAG vector to obtain the Ubi:OsMPK4-FLAG vector.
The CDS of OsMKK6 was ligated into the p35Spro-YFP C vector to obtain the YC-OsMKK6 vector. To generate the YN-OsMPK4 construct, the CDS of OsMPK4 was ligated into the p35Spro-YFP N vector.
To detect MAPK activation, WT and rsr25 plants were inoculated with Xoo, and total proteins were extracted from the plants at 0 h, 48 h, 96 h, and 120 h after inoculation. MPK activation was detected by immunoblotting with an anti-p44/42 ERK antibody (Cell Signaling, Danvers, MA, USA; 4370S).

Pathogen Infection
Detached leaves from one-month-old plants were subjected to punch inoculation with M. oryzae. Then, 5 µL of a spore suspension of M. oryzae isolate RB22 (5 × 10 5 spores/mL in 0.05% [v/v] Tween-20) was added to holes in the leaves that were created by gentle punching with a pipette tip. Following inoculation, the leaves were incubated at 25 • C for 24 h in the dark and switched to a 12 h light/12 h dark photoperiod at 25 • C. Lesion length was measured at seven days post-inoculation (dpi).
Three-week-old seedlings were sprayed with a spore suspension of M. oryzae isolate RB22 (5 × 10 5 spores/mL in 0.05% [v/v] Tween-20) or 0.05% (v/v) Tween-20 as a control. The inoculated seedlings were grown in a growth chamber under the same growth conditions, except that the relative humidity was adjusted to 95%. Total RNA was extracted from the seedlings at 24 h post-inoculation (hpi).
For the Xoo incubation assay, the concentration of Xoo strain P6 in the liquid medium (20 g/L sucrose; 5 g/L peptone; 0.25 g/L MgSO 4 -7H 2 O; 0.5 g/L K 2 HPO 4 ; pH 7.2~7.5) was adjusted to 10 9 cells/mL (OD 600 = 1.0). For inoculation, scissors were dipped into the bacterial suspension and used to remove leaf tips (5 cm). The inoculated plants were grown in a growth chamber under the same conditions described above, except that the relative humidity was adjusted to 95%. Lesion length was measured at 14 dpi.

Measurement of ROS
ROS was measured in flg22-treated seedlings as previously described [17]. Briefly, leaf discs were collected from the leaves of 3-week-old wild type and rsr25 seedlings using a 1.5 mm diameter Miltex puncher. The leaf discs were immersed in ddH 2 O for 12 h to eliminate the effects of physical damage. Two leaf discs per sample were placed into the bottom of a 1.5 mL centrifuge tube with 100 µL luminol (Bio-Rad Immun-Star Horseradish Peroxidase Substrate, Hercules, CA, USA, 170-5040), 1 µL horseradish peroxidase, and 100 nM flg22 or ddH 2 O as a control. Luminescence was measured immediately in a Glomax 20/20 photometer (Promega, Madison, WI, USA) at 30 s intervals over a 20 min period, with each seedling subjected to three biological replicates.

Bimolecular Fluorescence Complementation (BiFC) Assays
The coding sequence of OsMKK6 was ligated into the p35Spro-YFP C vector, and the coding sequence of OsMPK4 was ligated into the p35Spro-YFP N vector. The constructs were transiently expressed in Nicotiana benthamiana leaf epidermal cells by Agrobacterium (Agrobacterium tumefaciens)-mediated infiltration and co-expressed with H2B-mCherry (encoding a fusion between histone H2B and mCherry) as a nuclear localization marker. p35Spro-YFP C (YC) and p35Spro-YFP N (YN) were used as negative controls. The YFP and mCherry signals were imaged under a Zeiss 595 LSM710NLO confocal laser (Oberkohen, Battenwürburg, Germany) scanning microscope at 96 h after infiltration.

RNA Extraction and RT-qPCR
Total RNA was extracted from WT, rsr25, CG3, and Osmpk4 leaves using a TaKaRa MiniBEST Plant RNA Extraction Kit (TaKaRa, Kusatsu, Shiga, Japan; 9769S). First-strand complementary DNA (cDNA) was obtained using a PrimeScript ™ RT reagent Kit with gDNA Eraser (TaKaRa, RR047A). qPCR analysis was performed using TB Green Fast qPCR Mix (TaKaRa, RR430s) in a LightCycler 480 Real-Time PCR System (Roche, Basel, Switzerland). The 2 −∆∆ CT method was used to analyze the RT-qPCR data, and the housekeeping gene OsACTIN (LOC_Os03g50885) was used as an internal control. The primers used for RT-qPCR analysis are listed in Table S3.