Two LysM receptor molecules, CEBiP and OsCERK1, cooperatively regulate chitin elicitor signaling in rice

Abstract

Chitin is a major molecular pattern for various fungi, and its fragments, chitin oligosaccharides, are known to induce various defense responses in plant cells. A plasma membrane glycoprotein, CEBiP (chitin elicitor binding protein) and a receptor kinase, CERK1 (chitin elicitor receptor kinase) (also known as LysM-RLK1), were identified as critical components for chitin signaling in rice and Arabidopsis, respectively. However, it is not known whether each plant species requires both of these two types of molecules for chitin signaling, nor the relationships between these molecules in membrane signaling. We report here that rice cells require a LysM receptor-like kinase, OsCERK1, in addition to CEBiP, for chitin signaling. Knockdown of OsCERK1 resulted in marked suppression of the defense responses induced by chitin oligosaccharides, indicating that OsCERK1 is essential for chitin signaling in rice. The results of a yeast two-hybrid assay indicated that both CEBiP and OsCERK1 have the potential to form hetero- or homo-oligomers. Immunoprecipitation using a membrane preparation from rice cells treated with chitin oligosaccharides suggested the ligand-induced formation of a receptor complex containing both CEBiP and OsCERK1. Blue native PAGE and chemical cross-linking experiments also suggested that a major portion of CEBiP exists as homo-oligomers even in the absence of chitin oligosaccharides.

Figure 1

Characterization of OsCERK1 and other OsLysM-RLK genes.(a) Expression of LysM receptor-like kinase genes (OsLysM-RLK genes) in rice cells. The primers for RT-PCR were designed to amplify a specific region of each gene.(b) Predicted structure of OsCERK1 (rice) and CERK1 (Arabidopsis). SP, signal peptide; LysM, LysM motif; TM, transmembrane domain.(c) Amino acid sequence of OsCERK1 predicted from the cDNA. Underlining indicates the sequence corresponding to the LysM motif. EC, extracellular domain; IC, intracellular domain; SP, signal peptide; TM, transmembrane domain.(d) Alignment of the three LysM motifs of CERK1 to the consensus sequence or corresponding regions of OsCERK1. The conserved amino acid residues are indicated by vertical bars.(e) Expression patterns of the OsCERK1 and CEBiP genes in each part of the rice plant. OsUBQ, rice ubiquitin; Sh, shoot; R, root; PS, proximal shoot; St, stem; F, flower.

Figure 2

Gene-specific knockdown of OsCERK1 resulted in the loss of chitin-induced ROS generation.(a) Establishment of OsCERK1-RNAi cell lines. The bar above the diagram of OsCERK1 indicates the region used for construction of the RNAi vector. The amounts of OsCERK1 mRNA normalized for the internal control were compared for the RNAi lines and the vector control (VC). All data are the mean of three independent experiments, and error bars indicate standard deviation.(b) Expression pattern of OsLysM-RLK genes in the OsCERK1-RNAi and VC cell lines.(c) Suppression of chitin oligosaccharide-induced ROS generation in the OsCERK1-RNAi cell lines. Suspension-cultured rice cells (40 mg) were incubated with (+) or without (−) 160 ng/ml of (GlcNAc)₈ for 30 and 120 min at 25°C, and analyzed for ROS generation. ROS was determined by the luminal-dependent chemiluminescence assay. The gray and white bars indicate the values for the elicitor treatment and control experiments, respectively. Error bars indicate standard deviation.(d) Specificity of the suppression of elicitor-induced ROS generation in the OsCERK1-RNAi cell lines. Suspension-cultured rice cells (40 mg) were incubated with medium containing (GlcNAc)₈ (filled circles, 160 ng/ml), Pseudomonas aeruginosa lipopolysaccharides (filled triangles, 50 μg/ml) and sterile H₂O (filled squares) for 0, 30, 60, 90 and 120 min. The y axis scales for OsCERK1-RNAi #2 and #13 are the same as those of VC and OsCERK1-RNAi #6, respectively. All the data are the means of three independent experiments, and error bars indicate standard deviation.

Figure 3

Chitin-induced defense responses were mostly diminished in the OsCERK1-RNAi cell lines.(a) Effect of OsCERK1 knockdown on chitin-induced phytoalexin biosynthesis. OsCERK1-RNAi and vector control (VC) cell lines were treated with 160 ng/ml of (GlcNAc)₈ for 48 h, and the culture medium was analyzed for phytoalexins by HPLC-ESI-MS/MS. M, momilactone; P, phytocassane.(b) Microarray analysis of elicitor-responsive genes in the OsCERK1-RNAi and VC cell lines. Total RNAs were extracted from rice cells treated with or without 160 ng/ml (GlcNAc)₈ for 2 h, and used for microarray analysis. Open box, genes that responded to the elicitor only in VC; dark gray box, genes with significantly decreased responsiveness to the elicitor (ratio of VC/OsCERK1-RNAi >2) in the OsCERK1-RNAi line; black box, genes equally responsive to the elicitor in both the OsCERK1-RNAi line and VC; light gray box, genes that responded to the elicitor only in the OsCERK1-RNAi line.(c) Suppression of chitin-induced expression of defense genes in the OsCERK1-RNAi cell line. The examined genes were selected from microarray data (b). Total RNAs were extracted from rice cells treated with or without 160 ng/ml (GlcNAc)₈ for 2 h, and used for quantitative RT-PCR analysis using gene-specific probes to determine the amount of mRNAs. The amount of mRNA was normalized against the internal control, and relative values compared with non-treated samples are shown. VC, vector control cell line; Ri, OsCERK1-RNAi cell line; PAL, phenylalanine anmonia lyase; β-Glu, β-glucanase; HIP, harpin-induced 1 domain-containing protein; OsKS4, OsKS4 encoding gene.

Figure 4

Interaction of LysM-containing receptor molecules.(a) Possible interactions between LysM-containing receptor molecules. A yeast two-hybrid assay was performed using the extracellular domains of CEBiP (CEBiP_ex) and OsCERK1(OsCERK1_ex) as bait or prey. Growth on selective medium (SD/-LWHA) and blue colony formation in the X-gal assay indicated a positive interaction. EV, empty vector.(b) Localization of OsCERK1 in rice cells. Cytosol, plasma membrane (PM) and microsome (MF) fractions were prepared from rice cells expressing myc-tagged OsCERK1 (OsCERK1:myc), and analyzed for the presence of OsCERK1:myc by Western blotting with anti-myc antibody. The arrowhead indicates the denatured OsCERK1:myc protein.(c) Analysis of protein complex in the plasma membrane by BN-PAGE. A plasma membrane preparation from rice cells expressing OsCERK1:myc was solubilized with 0.5%n-dodecyl-β-d-maltoside and subjected to BN-PAGE. CEBiP and OsCERK1:myc were detected with anti-CEBiP antiserum (left) and anti-myc antibody (right). Microsome proteins from the same cell line were completely denatured by boiling with SDS–PAGE sample buffer and used to show the positions of the CEBiP or OsCERK1:myc monomers. S, solubilized plasma membrane proteins; D, denatured microsome proteins.(d) In vitro chemical cross-linking for the detection of protein complexes containing CEBiP or OsCERK1:myc. Plasma membranes from rice cells expressing OsCERK1:myc were treated with 3 mg/ml of DTSSP, followed by SDS–PAGE. Immunodetection was performed with anti-CEBiP antiserum (left) and anti-myc antibody (right).(e) Analysis of the CEBiP–OsCERK1 interaction by immunoprecipitation. Microsomes from rice cells expressing OsCERK1:myc pre-treated with or without (GlcNAc)₈ were solubilized with 0.5% Triton X-100 and immunoprecipitated with anti-CEBiP or anti-myc antibody. The immunoprecipitates were recovered using protein A beads and eluted with SDS–PAGE sample buffer. Aliquot (5 or 10 μg) of the total eluate (60 μl) were subjected to SDS–PAGE and detected with the corresponding antibodies. The presence of CEBiP and OsCERK1:myc in the immunoprecipitates was determined using the corresponding antibodies.

Figure 5

Phylogenetic tree of OsCERK1 and related plant LysM receptor-like kinases.Phylogenies were analyzed by multiple sequence alignment of the protein sequences of the LysM receptor-like kinases. The names of proteins described previously (Zhang et al., 2007) are shown in parentheses for ease of comparison. The scale indicates the base substitution rate, and numbers at the nodes represent bootstrap values with 1000 replicates.