Sugarcane cultivar ROC22 was cultured in the sugarcane clonal germplasm repository at Guangxi University (Nanning, China) and grown at 30°C in a greenhouse with a 13 h light/11 h dark cycle. Arabidopsis (Columbia, Col-0) plants were grown at 22°C under long-day conditions (16 h light/8 h dark), which provided the genetic background of plants overexpressing ScRIPK, whose coding region was cloned into the pBWA(V)HS vector, driven by 35s promoters. T₂ seedlings were used for analysis. Nicotiana benthamiana (N. benthamiana) plants were grown in a growth chamber under long day conditions (16 h light/8 h dark) at 26°C, with a light intensity of 150 µmol m^–2 s^–1 and 70% relative humidity. Four-week-old N. benthamiana was used for bimolecular fluorescence complementation analysis and luciferase complementation imaging assays.

The 1-month-old (5–7 leaf stage) sugarcane plants with uniform growth were irrigated once with 500 mL 250 mM NaCl and 20% (p/v) Polyethylene glycol (PEG) 6000 in the greenhouse, respectively (Nogueira et al., 2005; Chu et al., 2022). All leaves were then sampled at 0, 1, 3, 6, 12, 24, and 72 h after treatment. Ten sugarcane seedlings of per replicate were inoculated with Fusarium sacchari (1 × 10⁵ conidia ml⁻¹) at the 5-leaf stage and +1 leaves were sampled after at 0, 12, 24, 72, 120, and 168 h post-inoculation, which performed as described (Huang et al., 2022; Paul et al., 2022). Three biological replicates were performed. All samples were collected and immediately frozen in liquid nitrogen, and stored at −80°C.

For the Arabidopsis experiments, transgenic and wild-type seeds were germinated on 1/2 Murashige and Skoog medium (1/2 MS) for 10 days under the same conditions as described in the section on plant materials and growth conditions. For the drought treatment, seedlings were transferred from 1/2 MS to soil, and then 4-week-old seedlings were not watered for 10 days. After recovery for 3 days, the relative water content (RWC) was measured as previously described (Zhang et al., 2019). For the Pseudomonas syringae pv. tomato (Pst) DC3000 infection experiment, Pst DC3000 was grown in King’s B broth (Coolaber, Beijing, China), and then diluted with 10 mM MgCl₂ solution to OD₆₀₀ 0.001. Leaves of 4-week-old transgenic plants and wild-type plants were infiltrated with Pst DC3000 using a syringe without a needle for 3 days.

The coding sequence of ScRIPK was amplified, and then inserted into pEarleyGate 103 vector, which performed as described previously (Fang et al., 2021; Chai et al., 2022). The pEarleyGate-103-ScRIPK vector was transiently expressed in sugarcane protoplasts with the plasma membrane marker vector. After 16–20 h, fluorescence was observed and imaged under a confocal scanning microscope (Leica-TCS-SP8MP; Leica Microsystems, USA). Three biological replicates were performed.

Total RNA from sugarcane leaves and whole seedlings of Arabidopsis were extracted and purified using the Eastep Super Total RNA Extraction Kit (Promega, USA). The cDNA was synthesized with a PrimeScript RT reagent kit with gDNA Eraser (Takara, USA), and the cDNA was used for RT-qPCR assays, which performed as described previously (Fang et al., 2021; Chai et al., 2022). RT-qPCRs were performed with a procedure as follows: 95°C 30s for initial denaturation; 95°C 5s, 60°C 30s for denaturation, annealing and extension with 40 cycles; and 95°C 5s, 60°C 1min, 95°C 1s for melt curve analysis. The relative expression levels were calculated following the 2^-ΔΔCt method. GADPH (Coelho et al., 2014) and Atactin2 (Shahnejat-Bushehri et al., 2016) were used as references to determine relative expression in sugarcane and Arabidopsis, respectively. Primers used in this study are provided in Supplementary Table 1 .

The kinase domain (72–378 aa) of ScRIPK and mutants (ScRIPK K124R and ScRIPK S253A|T254A) were amplified in 25 µL reaction mixture containing 1 µL of cDNA template, 2.5 µL of specific forward and reverse primers (2 µM), 2.5 mM dNTP mixture (2.5 mM), 12.5 µL of 2 × GC buffer II, 0.5 µL of Takara LA Taq (Takara, USA). Steps were as follows: after initial denaturation at 94°C for 4 min, 35 cycles were run at 94°C for 30 s, 60°C for 30 s, and 72°C for 1 min 30 s, then extension at 72°C for 10 min and stored at 4°C. The two products were separately cloned into the PRSF-duet vector with the In-Fusion HD Cloning Kit (50°C for 10 min, and then placed on ice, Takara, USA) and verified by sequencing (Sangon, China). These vectors were transformed into Escherichia coli strain BL21 (DE3, TransGen Biotech, China) for protein expression. Single clones were then picked and cultured in the Luria-Bertani (LB) medium at 37°C until OD₆₀₀ reached 0.6–1.0. Then, 0.5 mM isopropyl-beta-D-thiogalactopyranoside (IPTG) was added to induce protein expression at 16°C for 18–20 h. Cells were centrifuged and then lysed with PBS (2.7 mM KCl, 10 mM Na₂HPO₄, 1.8 mM KH₂PO₄, and 10 mM MgCl₂) buffer containing 500 mM NaCl, 20 mM imidazole, and 0.01% Tween 20, and centrifuged at 12,000 rpm for 1 h at 4°C. Soluble protein was then bound to the Ni-NTA column (affinity chromatography, Smart-Lifesciences, China) for preliminary purification and the target protein was eluted by the PBS buffer containing 300 mM imidazole, and 0.01% Tween 20, which was further removed non-target proteins by Superdex^TM75 (gel filtration chromatography, GE Healthcare, USA) at 16°C with the buffer containing 20 mM Tris PH 7.4, 300 mM NaCl, 4 mM MgCl₂, and 0.01% Tween 20.

The ScRIPK-ATP, ScRIPK K124R-ATP, and ScRIPK S253A|T254A-ADP complex were Crystallized using the hanging-drop method by mixing 1 µl protein with equal volume of reservoir solution in 24-well plates at 291 K (Hampton Research, Aliso Viejo, CA). The ScRIPK-ATP protein complex crystal was successfully obtained in the reservoir solution containing 0.1 M HEPES pH 7.5, 40% v/v PEG 400. ScRIPK K124R-ATP and ScRIPK S253A|T254A-ADP protein complex crystals were successfully obtained in the reservoir solution containing 0.1 M HEPES pH 7.5, 4.3 M Sodium chloride. These crystal data were collected on beamline BL17U1 at the Shanghai Synchrotron Radiation Facility (SSRF) and processed by XDS. The three structures were determined by molecular replacement, using the structure of BIK1 (PDB ID: 5TOS) as the initial search model. Model building and structural refinement were analyzed using Coot and PHENIX, respectively. The refinement statistics are summarized in Supplementary Table 1 . The atomic coordinates and structure data (ScRIPK-ATP, PDB ID: 8HO6; ScRIPK K124R-ATP, PDB ID: 8HOA; ScRIPK S253A|T254A-ADP, PDB ID: 8HOD) have been deposited in the Protein Data Bank. All structure figures were prepared using PyMOL.

For the kinase assays, 2 µg recombinant protein was run in the buffer (25 mM Tris-HCl PH 7.4, 10 mM MgCl₂, 1 mM ATP, and 1 mM DTT) in a total volume of 25 µL at room temperature for 30 min. Then the reaction was terminated by adding 5× protein loading buffer and boiling for 5 min at 100°C. Proteins were then separated on a 12% SDS-PAGE gel, stained with Coomassie blue G250, or transferred to a nitrocellulose membrane (PALL, USA) for protein immunoblotting using anti-Phospho-(Ser/Thr) antibody (Abcam, UK).

For BiFC analysis, the full-length sequence of ScRIPK was constructed into pEarleyGate201-YN (YN) and pEarleyGate201-YC (YC) and the full-length sequence of ScRIN4 was also constructed into YN and YC. The YN-ScRIPK and YC-ScRIN4 plasmids, or YN-ScRIPK and YC-ScRIN4 plasmids, were then transformed into N. benthamiana leaves. The positive controls (YN-AtRIPK + YC-AtRIN4 and YN-AtRIN4 + YC-AtRIPK) and the negative controls (YN-ScRIPK + YC, YN + YC-ScRIPK, YN-ScRIN4 + YC, YN + YC-ScRIN4, YN-AtRIN4 + YC, YN + YC-AtRIN4, YN-AtRIPK + YC, YN + YC-AtRIPK, and YN + YC) were also transformed into N. benthamiana leaves. After 2 days, images were captured with a confocal scanning microscope (Leica-TCS-SP8MP, Leica Microsystems, USA). Three biological replicates were performed.

For luciferase complementation imaging assays (LCI), the full-length coding sequence of ScRIPK/AtRIPK was cloned into Nluc, and the full-length coding sequence of ScRIN4/AtRIN4 was cloned into Cluc. The Nluc-ScRIPK, Cluc-ScRIN4, Nluc-AtRIPK, Cluc-AtRIN4, and the empty vectors Nluc and Cluc were transformed into Agrobacterium strain GV3101, respectively. Then, Nluc-ScRIPK was mixed with Cluc-ScRIN4 for the transient infiltration of N. benthamiana leaves. The same was done for Nluc-AtRIPK and Cluc-AtRIN4 as the positive control. Nluc-ScRIPK + Cluc, Nluc + Cluc-ScRIN4, Nluc-AtRIPK+ Cluc, Nluc + Cluc-AtRIN4, and Nluc + Cluc were used as the controls. After 2 days, LUC activity was observed and imaged using a CDD imaging apparatus (IVIS Lumina LT, PerkinElmer, USA).

Student’s t-test and Duncan’s multiple range test were performed in our study to determine their significance using the SPSS software (IBM Analytics, USA).

To explore the function of the AtRIPK and OsGUDK homologous genes in sugarcane, we cloned and sequenced a sequence from the sugarcane cultivar ROC22. The deduced coding sequence contains a 1278 bp open reading frame, encoding a 425 amino acid protein with a predicted molecular weight of 46.36 kDa. BLAST analysis revealed that the deduced protein shares high similarity to Zea mays ZmRIPK, Setaria italica SiRIPK and Oryza sativa Japonica OsGUDK (88.02%, 86.14% and 82.91% amino acid identity, respectively), which was in the same clade as RLCK class VII (RLCK VII, Figure 1A ). Therefore, this gene was designated ScRIPK. Like other kinases, ScRIPK protein kinase domain (residues 95 to 369) has 11 major conserved subdomains, I to XI ( Figure 1B ). These conserved subdomains are necessary for the catalytic function.

AtRIPK and related RLCKs contain an N-terminal palmitoylation/myristoylation motif that should serve as a targeting signal for plasma membrane localization (Veronese et al., 2006; Liu et al., 2011). To confirm the predicted ScRIPK localization, we conducted subcellular localization experiments using sugarcane protoplasts, which indicated that ScRIPK localized in the plasma membrane ( Figure 1C ). Thus, ScRIPK is a member of the RLCK VII subfamily localized in the plasma membrane.

To test whether ScRIPK expression responds to abiotic and/or biotic stresses, we analyzed the transcript levels of ScRIPK under polyethylene glycol (PEG), NaCl, and pathogen infection conditions ( Figure 2 ). ScRIPK transcription increased as early as 1 h after PEG treatment and exhibited an approximately 2-fold peak response at 6 h ( Figure 2A ). ScRIPK transcription levels had no noticeable change under NaCl treatment conditions ( Figure 2B ), but did respond to biotic stresses. ScRIPK transcription revealed a slight increase as early as 12 h post-fungus Fusarium sacchari treatment inoculation, followed by a downward trend until 168 h post-inoculation, with a total decrease of approximately 0.5-fold ( Figure 2C ). These data suggest that ScRIPK may play a role in responding to biotic and abiotic stresses.

To investigate the function of ScRIPK, we examined transgenic Arabidopsis plants over-expressing ScRIPK, driven by the constitutive 35s CaMV promoter. RT-qPCR was used to measure the expression levels of transgenic lines and three independent T₂ lines (OE-2, OE-10, and OE-13) were subsequently selected for further analysis ( Figure 3A ). We performed a drought tolerance test in which 4-week-old seedlings of both wild-type lines and transgenic lines were not watered for 10 days and then rewatered for 3 days. After 10 days of withholding water, most of the leaves in transgenic plants were green in color and showed signs of slight water loss. In contrast, most wild-type plants’ leaves were yellow in color, rolled, and showed signs of severe water loss. Three days after rewatering, most of the leaves of the transgenic plants were completely restored and fully expanded. The leaves of wild-type plants did not recover and most of them were white ( Figures 3B, C ). To analyze the function of ScRIPK in plant biotic stress, wild-type plants and transgenic plants were inoculated with Pseudomonas syringae pv. tomato (Pst) DC3000 using syringe infiltration, and the progress of the infection was observed over 3 days. As shown in Figure 3D , wild-type plants did not exhibit macroscopic visible signs of infection, while transgenic plants displayed chlorotic symptoms. Furthermore, bacterial growth was measured to determine the bacterial proliferation inside plants, showing that transgenic plants exhibited significantly higher bacterial proliferation than that in wild-type plants at 3 days ( Figure 3E ). These results suggest that ScRIPK plays role in regulating drought tolerance and disease defense in Arabidopsis.

To determine the kinase activity of ScRIPK, an in vitro assay was performed with expressed ScRIPK kinase domain (72–378 aa, ScRIPK-KD) in Escherichia coli. Western blot analysis using a phosphor-Ser/Thr antibody confirmed the autophosphorylation of ScRIPK ( Figure 4A ). Most protein kinases phosphorylate and adopt active conformations. To explore the conformation and activation mechanism of ScRIPK, we determined the crystal structure of ScRIPK-KD using the molecular replacement method ( Table 1 ). The solution compromises one ScRIPK-KD kinase domain per asymmetric unit. The ScRIPK-KD has a canonical bilobed structure architecture ( Figure 4B ). Like the typical structure of protein kinases, ScRIPK-KD contains a small N-lobe, composed of five β-sheets and an α-helix, and a large C-lobe, composed of six helices, an activation loop, and a catalytic loop ( Figures 4B, C ; Supplementary Figure 1 ). We performed three-dimensional homology searches using DALI (Holm et al., 2008), which indicated that ScRIPK-KD was closely related to the human interleukin receptor-associated kinase 4 (IRAK-4). A well-positioned ATP is one of the crucial features of a kinase catalytic machinery. In the ScRIPK-KD-ATP complex structure, Y169 and E170–M172 formed a deep cavity to clamp the adenine group of ATP and a magnesium ion bond with the catalytic base D237 and the β- and γ-phosphates of the nucleotide. Other residues, including G99–G101, K124, Q132, S176, K179, K221, and S223, may contribute to the stabilization of the nucleotide through hydrogen bonding ( Figure 4D ). The catalytic spine also facilitates the binding of ATP. The catalytic spine is completed by the adenine base of ATP, which binds to the V103 (from the beginning of the β2-strand), A122 (from the conserved Ala-Xxx-Lys of the β3-strand), and L226 (from the middle of the C-lobe β2-strand). I225 and L227 binds to L177 at the beginning of the αD-helix, and L177 binds to V285 and L289 in the αF-helix ( Figure 4E ). We also analyzed the ScRIPK-KD activation loopresidues that mediated the kinase functions. This activation loop contains three phosphorylation sites, namely, T250, S253, and T254, as indicated by the electron density ( Supplementary Figure 2A ).

The ScRIPK kinase domain adopted an active conformation with a salt-bridge formed between K124 and E140, similar to the active conformation of other kinase proteins, such as IRAK-4, BAK1, and BRI1 ( Figure 4F ). In addition, we determined the location of the DFG-Phe residue and calculated its distance from two conserved residues (K124 and L144, labeled as D2 and D1, respectively) in the N-terminal domain ( Figure 4G ; Supplementary Table 2 ), revealing that it belongs to the DFGin group. By analyzing the Ramachandran region annotation (A, B, L, and E) for the X, D, and F residues of the ScRIPK X-DFG motif and the DFG-Phe χ₁ rotamer, we found that the ScRIPK-KD structure was in the BLAminus cluster ( Supplementary Table 3 ). Furthermore, the regulatory spine (R-spine), which determines the positioning of the protein substrates so that catalysis occurs, was assembled in the ScRIPK-KD structure ( Figure 4F ). The R-spine contained residues from the activation segment and the αC-helix, which assembled by the L155 (from the beginning of the β4-strand), L144 (from the C-terminal end of the αC-helix), F238 (from the activation segment conserved DFG motif), along with H/YRD-Tyr217 of the catalytic loop. The backbone of Y217 was anchored to the αF-helix by a hydrogen bond to a conserved aspartate residue (D278). These findings show that ScRIPK is an autophosphorylating protein and adopts an active conformation.

The kinase-dead mutation of FER K565R eliminates in vitro phosphorylation activity and causes reduced responses of Arabidopsis root to RALF1(Haruta et al., 2018). K124 forms a salt bridge with conserved E233, further stabilizing the nucleotide. In this study, the catalytic base K124–Arginine (K124R) mutation was performed and the mutant protein ScRIPK-KD K124R was purified and crystallized with ATP. The alignment of overall structure between wild type and K124R mutant protein showed no significant conformation changes (root mean square deviation (RMSD) is 1.090 Å comparing 220 corresponding C_α atoms) except for the C-helix and activation loop ( Figure 5A ). The C-helix was outward, away from the active center, and the activation loop had no electron density map. In addition, the position of ATP in protein structure of wild type and K124R mutant, and the crucial residues of ATP binding pocket were partially different ( Figure 5B ). In the ScRIPK-KD K124R mutant structure, Y169 and E170–M172 formed a deep cavity to clamp the adenine group of ATP. Other residues, including G98, G99, and K124, may contribute to the stabilization of the nucleotide through hydrogen bonds ( Figure 5B ). The catalytic spine was also completed by the adenine base of ATP like the wild type ( Figure 5C ). To explore the conformation of this mutant protein, we analyzed whether there was the salt bridge between R124 and E140 and the results showed that there was no salt bridge between them ( Figure 5D ). We also measured the distance of D1 and D2, and dihedral angles of the X-D-F residues, which revealed that the ScRIPK-KD K124R mutant was classified into DFGin group and BLAplus cluster ( Supplementary Tables 2 , 3 ). Furthermore, the regulation spine was analyzed, which was not assembled ( Figure 5C ).

BRI1 T1049A and BAK1 T455A mutants lost nearly all kinase activity (Wang et al., 2005; Wang et al., 2008). The biological effect of mutating T446, T449 and T450 is less pronounced. However, T450 substitution restores growth while retaining complete flg22 insensitivity (Wang et al., 2008). We analyzed the electron density of the activation loop and found three phosphorylation sites, T250, S253, and T254 ( Supplementary Figure 2A ). According to the sequence alignment of ScRIPK, BRI1, and BAK1 ( Supplementary Figure 1 ), we simultaneously mutated S253 and T254 in the ScRIPK protein, purified and crystallized the mutant protein, and analyzed its conformation. The C-helix of the ScRIPK-KD S253A|T254A with ADP complex was outward, away from the active center, similar to the ScRIPK-KD K124R mutant protein ( Figure 5A ). In the ScRIPK-KD S253A|T254A mutant structure, Y169 and E170–M172 formed a deep cavity to clamp the adenine group of ADP. Residues G99 and K124 may contribute to the stabilization of the nucleotides through hydrogen bonding ( Figure 5B ), and the C-spine was completed by the adenine base of ADP to stabilize the ADP ( Figure 5E ). To explore the conformation of this mutant protein, we analyzed the salt bridge between K124 and E140, and measured the distance of D1 and D2 and dihedral angles of the X-D-F residues. This revealed that there was no salt bridge between K124 and E140, and this mutant was classified into DFGin group and BLAplus cluster ( Figure 5C ; Supplementary Figure 2C ; Supplementary Tables 2 , 3 ). In addition, the regulation spine was analyzed, which was also not assembled ( Figure 5E ). Therefore, the two mutant proteins were shown to adopt inactive conformation, suggesting that K124, S253, and T254 function in nucleotide binding and catalysis occurring.

The receptor-like cytoplasmic kinase AtRIPK interacts with AtRIN4 and phosphorylates it, activating plant’s innate immune receptor (Liu et al., 2011). In the present study, we identified and cloned a protein in sugarcane (cv. ROC22) through homologous cloning. The deduced coding sequence contained a 702 bp open reading frame encoding a protein with 233 amino acid residues. Phylogenetic analysis revealed that the deduced protein was related to SbRIN4 ( Supplementary Figure 3A ). Therefore, we named this protein ScRIN4. Then we aligned the sequences of RIN4s from monocots and dicots to analyze the F/YTxxFxK motif surrounding the phosphorylation sites, which showed that ScRIN4 had a conserved motif similar to other RIN4 orthologs ( Supplementary Figure 3B ). To explore whether ScRIPK interacts with ScRIN4, as the homologs do in Arabidopsis, we used bimolecular fluorescence complementation (BiFC) and luciferase complementation imaging (LCI) assays. The results showed that ScRIPK did interact with ScRIN4 ( Figures 6A, B ; Supplementary Figure 4 ) and the results of the interaction between AtRIPK and AtRIN4 was used as the positive control.