Arabidopsis WRKY33 is a key transcriptional regulator of hormonal and metabolic responses toward Botrytis cinerea infection

Abstract

The Arabidopsis (Arabidopsis thaliana) transcription factor WRKY33 is essential for defense toward the necrotrophic fungus Botrytis cinerea. Here, we aimed at identifying early transcriptional responses mediated by WRKY33. Global expression profiling on susceptible wrky33 and resistant wild-type plants uncovered massive differential transcriptional reprogramming upon B. cinerea infection. Subsequent detailed kinetic analyses revealed that loss of WRKY33 function results in inappropriate activation of the salicylic acid (SA)-related host response and elevated SA levels post infection and in the down-regulation of jasmonic acid (JA)-associated responses at later stages. This down-regulation appears to involve direct activation of several jasmonate ZIM-domain genes, encoding repressors of the JA-response pathway, by loss of WRKY33 function and by additional SA-dependent WRKY factors. Moreover, genes involved in redox homeostasis, SA signaling, ethylene-JA-mediated cross-communication, and camalexin biosynthesis were identified as direct targets of WRKY33. Genetic studies indicate that although SA-mediated repression of the JA pathway may contribute to the susceptibility of wrky33 plants to B. cinerea, it is insufficient for WRKY33-mediated resistance. Thus, WRKY33 apparently directly targets other still unidentified components that are also critical for establishing full resistance toward this necrotroph.

Figure 1.

Fungal development and ROS production on wild-type (wt) and wrky33 plants. Leaves from 4-week-old wild-type and wrky33 plants were spray inoculated with B. cinerea and analyzed for fungal growth and the development of cell death by staining with trypan blue at 24 hpi (A and B), 40 hpi (C and D), and 64 hpi (E and F). To visualize H₂O₂, leaves were stained with DAB at 16 hpi (G and H). Bars = 100 μm in A, B, G, and H and 500 μm in C to F.

Figure 2.

Kinetics of WRKY33 expression upon B. cinerea infection. A, Wild-type (wt) and wrky33 plants were spray inoculated with B. cinerea, and leaf samples were taken at the indicated time points to quantify the transcript levels of WRKY33 and other genes (subsequent figures) by qRT-PCR. Data were normalized to the expression of At4g26410 and are expressed relative to mock-treated (mo) wild-type plants set to 1. Error bars represent sd of three biological replicates (n = 3). B, WRKY33 protein accumulation during infection was monitored by using the WRKY33-HA complementation line. Plant treatment and sample collection were as in A. Protein levels were visualized by western blotting using an anti-HA antibody. As a loading control, a section of the filter stained with Ponceau S is shown.

Figure 3.

Transcriptional reprogramming in wild-type (wt) and wrky33 plants upon B. cinerea infection. A, Numbers of differentially expressed genes (2-fold or greater; P ≤ 0.05) between the wild type and wrky33 at 14 h after mock treatment (mo) or B. cinerea spray inoculation (Bc). Indicated are total numbers (boxes) and numbers of up- and down-regulated genes (arrows) between treatments and genotypes. B, Venn diagram illustrating total numbers and overlap of genes affected in wild-type and wrky33 plants at 14 h post B. cinerea inoculation.

Figure 4.

Expression levels of JA pathway-associated genes and JA levels during B. cinerea infection. A, Transcript levels of selected JA pathway-associated genes were determined by qPCR using the cDNAs described in Figure 2A. Expression of At4g26410 was used for normalization, and fold induction values of all genes were calculated relative to the expression level of mock-treated (mo) wild-type (wt) plants set to 1. Error bars represent sd (n = 3). B, JA and JA-Ile concentrations measured at 5, 12, and 24 hpi in wild-type and wrky33 plants spray inoculated with B. cinerea or mock treated. Error bars represent sd (n = 3). Different letters indicate significant differences among time points within each genotype (ANOVA; P < 0.0001 for JA, P < 0.05 for JA-Ile). Fw, Fresh weight.

Figure 5.

Differential up-regulation of JAZ genes upon B. cinerea infection in wrky33. The cDNAs described in Figure 2A were used to analyze by qPCR the expression kinetics of selected JAZ genes from B. cinerea-inoculated or mock-treated (mo) wild-type (wt) and wrky33 plants. After normalization to gene At4g26410, fold induction values were calculated relative to mock-treated wild-type plants set to 1. Error bars represent sd (n = 3).

Figure 6.

Direct binding of WRKY factors to JAZ gene promoters. ChIP qPCR was used to demonstrate WRKY factor binding to the promoters of selected JAZ genes. On top for each gene locus, a schematic drawing including the promoter region is shown (A, JAZ1; B, JAZ5; C, JAZ8; D, JAZ10). W-boxes (T/CTGACT/C) at the indicated positions are depicted by black triangles, and W-box-like (TGACT/C) sequences are depicted by white triangles. Open reading frames are represented by large arrows in the direction of translation. The scale is in bp relative to the translation start of the respective gene. For ChIP, leaves from wrky33 (w33), wild-type (WT), and WRKY33-HA (33HA) plants were spray inoculated with B. cinerea (Bc) or mock treated (mo) for 24 h. Input DNA before immunoprecipitation (IN) and coimmunoprecipitated DNA using an anti-HA or the anti-all WRKY antibody (IP) were analyzed by qPCR employing gene-specific primer pairs (Pr1, Pr2, Pr3, gene) and are expressed as fold enrichment relative to a DNA fragment from At4g26410. As a control for primer efficiency, purified genomic DNA was included in the analysis. Each ChIP experiment was repeated at least twice with the same result.

Figure 7.

ORA59 and GLIP1 are direct targets of WRKY33. A and B, Expression levels of ORA59 (A) and GLIP1 (B) from B. cinerea-inoculated or mock-treated (mo) wild-type (wt) and wrky33 plants were analyzed by qRT-PCR. Fold induction was calculated relative to mock-treated wild-type plants set to 1 after normalization to gene At4g26410. Error bars represent sd (n = 3). C and D, Binding of WRKY33 to the ORA59 (C) and GLIP1 (D) promoters. A schematic drawing for each gene locus including the promoter region is shown at the top (described in Fig. 6). For ChIP qPCR, leaves from wrky33 (w33) and WRKY33-HA (33HA) were spray inoculated with B. cinerea (Bc) or mock treated for 24 h. Input DNA before immunoprecipitation (IN) and coimmunoprecipitated DNA using an anti-HA antibody (IP) were analyzed by qPCR employing gene-specific primer pairs (Pr1, Pr2, Pr3, gene). Values are fold enrichment relative to a DNA fragment from At4g26410. Each ChIP experiment was repeated at least twice with the same result.

Figure 8.

Expression levels of SA pathway-associated genes and SA levels after B. cinerea inoculation. A, Transcript levels of selected SA-associated genes were determined by qPCR using as templates the cDNAs described in Figure 2A. Fold induction values of all genes except for PR1 were calculated relative to mock-treated (mo) wild-type (wt) plants set to 1. Due to very low expression in the mock-infected wild type, PR1 is shown as fold change relative to the expression level of gene At4g26410, which was used in all qPCRs for normalization. Error bars represent sd (n = 3). B, Total SA and free SA concentrations at the indicated time points were determined in wild-type and wrky33 leaves after mock treatment or spray inoculation with B. cinerea (Bc). Error bars represent sd (n = 3). Different letters indicate significant differences among time points within each genotype (ANOVA; P < 0.05 for free SA, P < 0.0001 for total SA). Fw, Fresh weight.

Figure 9.

Direct regulation of camalexin biosynthesis by WRKY factors upon B. cinerea infection. A, Accumulation of camalexin in B. cinerea-inoculated wild-type (wt), wrky33, and pad3 plants. At the indicated time points, camalexin concentrations were determined in mock-treated and inoculated wild-type, wrky33, and pad3 leaves. Error bars represent sd (n = 3). Fw, Fresh weight. B, Transcript levels of CYP71A13 and PAD3. Transcript levels of CYP71A13 and PAD3 at the indicated time points after inoculation were determined by qRT-PCR. Fold change was calculated relative to the expression level of gene At4g26410. C, Direct binding of WRKY factors to the CYP71A13 and PAD3 promoters. A schematic drawing for each gene locus including the promoter region is shown at the top (described in Fig. 6). For ChIP, leaves from wild-type, wrky33 (w33), and WRKY33-HA (33HA) plants were spray inoculated with B. cinerea (Bc) or mock treated (mo) for 24 h. Input DNA before immunoprecipitation (IN) and coimmunoprecipitated DNA (IP) using an anti-HA or the anti-all WRKY antibody were analyzed by qPCR employing gene-specific primer pairs (Pr1, Pr2, Pr3, Pr4, gene). Fold enrichment is expressed relative to a DNA fragment from At4g26410. Each ChIP experiment was repeated twice with the same result.

Figure 10.

WRKY33 modulates the expression of redox-related genes and directly targets TRX-H5. A, Differentially expressed redox-related genes after B. cinerea infection. Transcript levels of selected redox genes at the indicated time points of B. cinerea-inoculated wild-type (wt) and wrky33 leaves were quantified by qRT-PCR. After normalization to gene At4g26410, fold induction values were calculated relative to mock-treated (mo) wild-type plants set to 1. Error bars represent sd (n = 3). B, Direct binding of WRKY33 to the TXR-H5 promoter upon B. cinerea infection. A schematic drawing of the gene locus including the promoter region is shown at the top (described in Fig. 6). ChIP was performed using leaves from spray-inoculated (Bc) and mock-treated wrky33 (w33) and WRKY33-HA (33HA) plants at 14 hpi. DNA before immunoprecipitation (IN) and after immunoprecipitation using an anti-HA antibody (IP) were quantified by qPCR employing the primer pairs Pr1, Pr2, and gene. Values are fold enrichment relative to a fragment of gene At4g26410. The ChIP experiment was repeated twice with the same result.

Figure 11.

B. cinerea growth quantification on wrky33 double mutants. The indicated Arabidopsis lines were inoculated with two droplets of B. cinerea spores per leaf, and after 3 d the relative abundance of B. cinerea and Arabidopsis DNA was determined by qPCR employing specific primers for BcCutinase A and AtSKII, respectively. Error bars represent sd (n = 3). wt, Wild type.