A LysM receptor-like kinase plays a critical role in chitin signaling and fungal resistance in Arabidopsis

Abstract

Chitin, a polymer of N-acetyl-d-glucosamine, is found in fungal cell walls but not in plants. Plant cells can perceive chitin fragments (chitooligosaccharides) leading to gene induction and defense responses. We identified a LysM receptor-like protein (LysM RLK1) required for chitin signaling in Arabidopsis thaliana. The mutation in this gene blocked the induction of almost all chitooligosaccharide-responsive genes and led to more susceptibility to fungal pathogens but had no effect on infection by a bacterial pathogen. Additionally, exogenously applied chitooligosaccharides enhanced resistance against both fungal and bacterial pathogens in the wild-type plants but not in the mutant. Together, our data indicate that LysM RLK1 is essential for chitin signaling in plants (likely as part of the receptor complex) and is involved in chitin-mediated plant innate immunity. The LysM RLK1-mediated chitin signaling pathway is unique, but it may share a conserved downstream pathway with the FLS2/flagellin- and EFR/EF-Tu-mediated signaling pathways. Additionally, our work suggests a possible evolutionary relationship between the chitin and Nod factor perception mechanisms due to the similarities between their potential receptors and between the signal molecules perceived by them.

Figure 1.

The Knockout of the LysM RLK1 Gene Blocked the Induction of the Selected CRGs by Chitooctaose. (A) Analysis of the expression of the selected CRGs in both the mutant and wild-type plants using RT-PCR. Both the mutant (Mu) and wild-type plants were treated with purified chitooctaose or water (as a control) for 30 min, and gene expression of the selected CRGs was then detected using RT-PCR. Actin-2 was used as an internal control, and the amplification of both actin-2 and a CRG was conducted in the same PCR reaction tube with 25 cycles. The experiment was repeated at least three times with similar results. (B) The gene structure of LysM RLK1 (At3g21630) (not drawn to scale). Square boxes are exons. Solid lines between them are introns. The start codon (ATG) and stop codon (TAG) were included in the first and last exon, respectively. The two T-DNA insertions (T-DNA1 and 2) inserted in the 10th intron in the LysM RLK1 mutant were indicated above the gene. LB, left border; RB, right border. (C) The predicted domain structure of LysM RLK1. S, signal peptide; LysM, LysM domain; TM, transmembrane domain; WT, wild-type Col-0; Mu, LysM RLK1 mutant. (D) Analysis of the LysM RLK1 transcript in both the mutant and wild-type plants using RT-PCR. The gene-specific primers were derived from the exons on each side of the insertions. (E) Analysis of the LysM RLK1 transcript in both the mutant and wild-type plants using RT-PCR. The gene-specific primer pairs were derived from the exons before the insertions. WT, wild-type Col-0; Mu, LysM RLK1 mutant. Actin-2 was used as an internal control. (F) Restoration of CRG induction in the LysM RLK1 mutant by the ectopic expression of the LysM RLK1 cDNA. The LysM RLK1 cDNA driven by a cauliflower mosaic virus 35S promoter was introduced into the homozygous LysM RLK1 mutant using a modified pCAMBIA1200 binary vector. The complemented plants (homozygous T3) were treated with chitooctaose at a final concentration of 1 μM or water (as a control) for 30 min. RT-PCR analysis showed that the selected CRGs were induced to a similar level in the selected complemented plants (Com-1 and Com-2) to that in the wild type. Com-1 and Com-2, two independent complemented lines; WT, wild-type Col-0. Actin-2 is used as an internal control.

Figure 2.

The Mutation in the LysM RLK1 Gene Blocked the Induction of Virtually All CRGs. (A) Upregulated genes in the wild-type and mutant (Mu) plants in response to chitooctaose, revealed by the Affymetrix Arabidopsis Whole Genome Array ATH1. Upregulated genes: 1.5-fold and P < 0.05. (B) Downregulated genes in the wild-type and mutant (Mu) plants in response to chitooctaose. Downregulated genes: 1.5-fold and P < 0.05.

Figure 3.

The LysM RLK1 Mutant Was More Susceptible to a Biotrophic Fungal Pathogen Than Wild-Type Plants. (A) Three-week-old plants were inoculated with E. cichoracearum. Pictures were taken 10 d after inoculation. (B) Trypan blue staining showing fungal hyphae and conidiophores of E. cichoracearum on Arabidopsis leaves. Arrows indicate sites where conidiophores were forming. Pictures were taken 6 d after inoculation. Only one colony was shown in the pictures. Bars = 0.1 mm. (C) The quantification of the number of conidiophores (stalks bearing asexual spores) per colony (c/c). Conidiophores were counted 6 d after inoculation. The average value and se were based on at least 15 replications. WT, wild-type Col-0; Mu, LysM RLK1 mutant; NahG, transgenic plants expressing a bacterial salicylate hydrolase.

Figure 4.

The LysM RLK1 Mutant Was More Susceptible to a Necrotrophic Fungal Pathogen Than Wild-Type Plants, and Exogenously Applied Chitooligosaccharides Enhanced Resistance in the Wild-Type Plants. (A) Four-week-old plants were inoculated with A. brassicicola. Plants were pretreated twice (24 and 4 h before pathogen inoculation) with CSC (200 μg/mL), purified chitooctaose (8mer, 5 μM), or water. Pictures were taken 3 d after inoculation. (B) The average diameter of the lesions caused by A. brassicicola on both the mutant and wild-type Arabidopsis leaves. The diameters of lesions from 30 leaves were measured 3 d after inoculation. (C) The average number of spores of A. brassicicola produced per lesion by both the mutant and wild-type plants 6 d after inoculation. The number of spores per lesion was measured from 30 leaves. WT, wild-type Col-0; Mu, LysM RLK1 mutant; CSC, crab shell chitin; 8mer, chitooctaose; water, distilled water. The asterisk indicates a significant difference between the mutant and wild-type plants under the same treatments based on a Student's t test (P < 0.05). The plus symbol indicates a significance difference between the chitin-treated plants and control (water-treated) plants of the same genotype based on a Student's t test (P < 0.05). Error bars indicate se. Each experiment was repeated at least twice with similar results.

Figure 5.

The Mutation in the LysM RLK1 Gene Did Not Affect Growth of a Bacterial Pathogen. Four-week-old plants were pretreated twice (24 and 4 h before pathogen inoculation) with CSC (200 μg/mL), purified chitooctaose (5 μM), or water before being infiltrated with P. syringae pv tomato DC3000. Leaf discs were collected 1 and 3 d after inoculation to determine bacterial growth. WT, wild-type Col-0; Mu, LysM RLK1 mutant; 8mer, chitooctaose; water, distilled water. Asterisks indicate that the mutant was statistically significantly different from the wild type based on a Student's t test (P < 0.05). Error bars indicate se. Each experiment was repeated at least twice with similar results.

Figure 6.

The LysM RLK1 Mutation Did Not Block the Induction of Flagellin-Responsive Genes. Fourteen-day-old, hydroponically grown seedlings were treated with the flagellin-derived flg22 peptide (dissolved in DMSO) at a final concentration of 10 μM or with an equivalent amount of DMSO (as a control). The selected flagellin-responsive genes were detected using RT-PCR. Actin-2 was used as an internal control, and the amplification of both actin-2 and a flagellin-responsive gene was conducted in the same PCR reaction tube with 25 cycles. The experiment was repeated twice with similar results.

Figure 7.

Comparison of the Genes Regulated by flg22, elf26, and Chitooctaose. (A) Venn diagram showing the overlap of the upregulated genes (≥2-fold) by flg22, elf26, and chitooctaose. (B) Venn diagram showing the overlap of the downregulated genes (≥2-fold) by flg22, elf26, and chitooctaose. The genes regulated by flg22 and elf26 were obtained from publications (Zipfel et al., 2004, 2006). The elf26 data (60 min after treatment) were chosen, instead of elf18 data, for comparison due to the following reasons: at 60 min, elf26 induced a comparable number of genes to that by flg22 at 30 min; at 30 min, elf26 only induced approximately half of the genes by flg22; no microarray data were available from the wild-type plants after treated with elf18; elf18 data were obtained using the fls2 mutant instead of the wild type. To be consistent with the twofold cutoff used for flg22 and efl26-regulated genes, only CRGs with a change ≥2-fold were included in the comparison.