CYSTEINE-RICH RECEPTOR-LIKE KINASE5 (CRK5) and CRK22 regulate the response to Verticillium dahliae toxins

Abstract

Cysteine-rich receptor-like kinases (CRKs) play critical roles in responses to biotic and abiotic stresses. However, the molecular mechanisms of CRKs in plant defense responses remain unknown. Here, we demonstrated that two CRKs, CRK5 and CRK22, are involved in regulating defense responses to Verticillium dahliae toxins (Vd-toxins) in Arabidopsis (Arabidopsis thaliana). Biochemical and genetic analyses showed that CRK5 and CRK22 may act upstream of MITOGEN-ACTIVATED PROTEIN KINASE3 (MPK3) and MPK6 to regulate the salicylic acid (SA)-signaling pathway in response to Vd-toxins. In addition, MPK3 and MPK6 interact with the transcription factor WRKY70 to modulate defense responses to Vd-toxins. WRKY70 directly binds the promoter domains of the SA-signaling-related transcription factor genes TGACG SEQUENCE-SPECIFIC BINDING PROTEIN (TGA2) and TGA6 to regulate their expression in response to Vd-toxins. Thus, our study reveals a mechanism by which CRK5 and CRK22 regulate SA signaling through the MPK3/6–WRKY70–TGA2/6 pathway in response to Vd-toxins.

Cysteine-rich receptor-like kinases play critical roles in plant defense responses.

Introduction

Plants possess two complex and efficient branches of innate immunity to defend against invasion by a broad spectrum of pathogens: pattern-triggered immunity (PTI) and effector-triggered immunity (ETI; Jones and Dangl, 2006; Tsuda and Katagiri, 2010; Spoel and Dong, 2012). The triggering of PTI relies on the recognition of microbe-associated molecular patterns by plasma membrane-localized pattern recognition receptors (PRRs), such as receptor-like kinases (RLKs). PTI responses include calcium ion fluxes, activation of mitogen-activated protein kinase (MAPK) cascades, reactive oxygen species (ROS) bursts, salicylic acid-mediated signaling, and callose deposition (Boller and Felix, 2009; Nicaise et al., 2009). ETI is induced through the sensing of pathogen effectors by resistance proteins (Wu et al., 2014). ETI is a rapid immune response associated with hypersensitive response (HR)-like localized programmed cell death (Coll et al., 2011).

As a type of plasma membrane-localized protein, RLKs mediate signal transduction in the cell and induce PTI. Arabidopsis (Arabidopsis thaliana) possesses more than 600 RLKs, which have a typical architecture that includes extracellular, single-pass transmembrane, and intracellular Ser/Thr protein kinase domains (Shiu and Bleecker, 2001). Cysteine-rich receptor-like kinases (CRKs) form a large subfamily of Arabidopsis RLKs. They contain two domain 26 of unknown function (DUF26) having C-X8-C-X2-C motifs that include four conserved Cys residues in the extracellular domain (Ohtake et al., 2000; Chen, 2001). Arabidopsis CRKs play roles in response to biotic stress and cell death (Bourdais et al. 2015). The expression of CRK5 and CRK22 were altered in response biotic stress (Wrzaczek et al., 2010). Moreover, CRK4, CRK5, CRK6, CRK13, CRK20, CRK28, CRK36, and CRK45 were shown to positively regulates resistance to Pseudomonas syringae and HR-like cell death (Chen et al., 2003, 2004; Acharya et al., 2007; Ederli et al., 2011; Zhang et al., 2013; Yeh et al., 2015; Lee et al., 2017; Yadeta et al., 2017). In addition, several CRKs were reported to be involved in the signaling pathways of plant defense responses. Constitutive expression of CRK13 led to increase resistance to P.syringae, which corresponded with accumulation of SA and the activation of defense marker genes, PATHOGENESIS-RELATED GENE (PR1), PR5, and ISOCHORISMATE SYNTHATE (ICS1; Acharya et al., 2007). CRK45 phosphorylated and activated its target proteins, which could regulate the expression of WRKYs and SA-related genes and SA accumulation (Zhang et al., 2013). Furthermore, CRK2 modulated ROS bursts and MAPK activation to regulate plant innate immunity (Bourdais et al. 2015; Kimura et al., 2020). Therefore, CRKs may be an important element in triggering the signaling pathways of plant defense responses. However, the molecular mechanisms of CRK5 and CRK22 in defense response to Verticillium dahliae toxins (Vd-toxins) are not fully understood.

Receptor protein-discerning pathogens can trigger the activation of SA signaling and MAPK cascades involved in defense responses (Boller and Felix, 2009; Nicaise et al., 2009). In general, SA is required for plant defenses against biotrophic and hemi-biotrophic pathogens (Vlot et al., 2009), and it functions as a common signal component of PTI and ETI, increasing upon pathogen infection. NONEXPRESSOR OF PATHOGENESIS-RELATED GENES1 (NPR1) functions as a central receptor/regulator of SA, and it contributes to the expression of SA-triggered genes and to resistance against pathogens (Cao et al., 1994; Ding et al., 2018). NPR1 often works with the transcription factors TGA2/6 to promote the expression of SA-related defense genes (Delaney et al., 1994; Zhang et al., 1999; Ding et al., 2018). WRKY transcription factors act downstream of NPR1 to regulate the expression of PR genes (Li et al., 2004; Wang et al., 2006). WRKY46, WRKY70, and WRKY53 cooperatively positively regulate PR1 expression to enhance resistance to P. syringae (Li et al., 2004; Ren et al., 2008; Hu et al., 2012). WRKY70 and its homolog WRKY54 positively regulate SA-responsive gene expression levels and inhibit SA biosynthesis (Wang et al., 2006; Li et al., 2017). In addition, SA-induced WRKY transcription factors also positively regulate NPR1, suggesting a SA–NPR1–WRKY amplification loop that contributes to the expression of SA-responsive genes (Yu et al., 2001; Li et al., 2004). However, the molecular mechanisms of WRKY transcription factors in regulating SA signaling remain unknown.

MAPK cascades function as signal transduction components of defense responses to amplify the signal (Ichimura et al., 2002). Two MAPKs, MPK3 and MPK6, regulate defense responses to pathogens in Arabidopsis (Asai et al., 2002; Ichimura et al., 2002). For instance, MPK3 and MPK6 mediate stomatal immunity against flg22 and P.syringae, suggesting that they have collaborative functions in regulating stomatal closure (Su et al., 2017). MPK3 and MPK6 act upon the same regulatory substrates, which indicates that they may have similar activation dynamics in mediating resistance to pathogens (Mao et al., 2011; Li et al., 2012; Bigeard et al., 2015; Genot et al., 2017). Nevertheless, MPK3 and MPK6 show distinct functions rather than complete functional redundancy (Genot et al., 2017). For example, MPK3 contributes to the basal resistance against Botrytis cinerea, while MPK6 plays a major role in the elicitor-induced resistance against this fungus (Galletti et al., 2011). Additionally, MPK3 and MPK6 regulate SA signaling involved in resistance against pathogens. Constitutively active MPK3 enhances SA biosynthesis and the expression of SA-responsive genes, and the sustained activation of MPK3 and MPK6 triggers the expression of SA-responsive genes during ETI (Tsuda et al., 2013; Genot et al., 2017). However, by what means MPK3 and MPK6 mediate SA signaling remains largely unknown.

Verticillium dahliae is a soil-borne pathogen that causes Verticillium wilt in a range of important plant species worldwide (Bhat and Subbarao, 1999). The overexpression of RLK improves cotton resistance to Verticillium wilt (Jun et al., 2015), and two lysin-motif receptor kinases LYK1 and LYK2 also enhance resistance to Verticillium wilt in cotton (Gu et al., 2017). The presence of V. dahliae induces the expression of CRKs (Xu et al., 2014), suggesting that CRKs may be involved in regulating resistance against Verticillium wilt. We have demonstrated previously that Histone H2B monoubiquitination (H2Bub1) plays an important role in defense responses to Vd-toxins, and H2Bub1 regulates the expression of CRK41 (Hu et al., 2014; Hu, 2016; Zhao et al., 2020). However, the molecular mechanisms and regulatory pathways of CRKs involved in defense responses to Verticillium remain unknown.

In our study, we showed that two receptor-like kinases, CRK5 and CRK22, positively regulate defense responses to Vd-toxins in Arabidopsis. CRK5 and CRK22 regulate the SA-signaling pathway through the activation of MPK3 and MPK6. Then, MPK3 and MPK6 interact with WRKY70, and WRKY70 directly regulates the expression of TGA2/6. Our study, therefore, reveals a mechanism used by CRK5 and CRK22 in Arabidopsis to regulate SA signaling through the MPK3/6–WRKY70–TGA2/6 pathway in defense responses to Vd-toxins.

Results

CRK5 and CRK22 positively regulate defense responses to Vd-toxins

CRK5 and CRK22 have a typical architecture that includes extracellular (possess two copies of DUF26 with conserved C-X8-C-X2-C)), single-pass transmembrane, and intracellular Ser/Thr protein kinase domain. Domain analysis using the National Center for Biotechnology Information (NCBI) conserved domain search indicated that there were two salt stress response/antifungal domain in extracellular region of CRK5 and CRK22, suggesting that CRK5 and CRK22 may have a role in salt stress response and have antifungal activity (Supplemental Figure S1A). To determine whether CRK5 and CRK22 are involved in defense responses to the fungus V. dahliae, we first monitored the relative expression levels of CRK5 and CRK22 using reverse transcription-quantitative polymerase chain reaction (RT–qPCR) after exposing wild-type Arabidopsis Columbia-0 (Col-0) to Vd-toxins. Interestingly, the CRK5 and CRK22 expression levels were induced and increased along with the Vd-toxins treatment time (Figure 1A). In addition, the protein levels of CRK5 and CRK22 were induced slightly (Supplemental Figure S1B). The results indicated that CRK5 and CRK22 were induced by Vd-toxins.

Figure 1.

CRK5 and CRK22 positively regulate defense responses to Vd-toxins in Arabidopsis. A, Relative expression levels of CRK5 and CRK22 after treating 7-day-old wild-type (Col-0) seedlings with 200 µg·mL⁻¹ Vd-toxins. Total RNA was extracted after various treatment times and assessed by RT–qPCR. Error bars indicate the sd of three replicates. B, Cell death induced by Vd-toxins in cotyledons of Arabidopsis Col-0, crk5 and crk22 mutants, CRK5/crk5 and CRK22/crk22 complementation lines, and 35S:CRK5-1 and 35S:CRK22-1 overexpression lines. Cotyledons of 7-day-old plants were treated with 200 µg·mL⁻¹ Vd-toxins for 18 h and stained with trypan blue as described in the Experimental procedures. Cotyledons treated with double distilled water (ddH₂O) were used as controls. Bars = 200 μm. C, H₂O₂ accumulations induced by Vd-toxins in rosette leaves. Rosette leaves of 4-week-old plants were treated with 200 µg·mL⁻¹ Vd-toxins for 24 h and stained with DAB at 12 h, as described in the Experimental procedures. Leaves treated with ddH₂O were used as controls. Bars = 100 µm or 2 mm. D, Callose depositions induced by Vd-toxin treatments in rosette leaves. The rosette leaves of the 4-week-old plants were treated with 200 µg·mL⁻¹ Vd-toxins for 24 h and stained with aniline blue as described in the Experimental procedures. Leaves treated with ddH₂O were used as controls. Bars = 100 μm. E, Quantification of the callose deposition in (D). Data indicate the numbers of callose deposits per mm². Leaves treated with ddH₂O were used as controls. The bars represent sd for three independent experiments (n = 8). F, Ion leakage in Arabidopsis Col-0, crk5 and crk22 mutants, CRK5/crk5/and CRK22/crk22 complementation lines, and 35S:CRK5-1 and 35S:CRK22-1 overexpression lines. Experimental details are provided in methods. Error bars indicate sd of three replicates. Different letters represent significant differences at P < 0.05 as assessed by one-way ANOVAs with Tukey’s honest significant difference post hoc tests. All the experiments were repeated three times.

The core characteristics of plant defense responses are HR-like cell death, ROS production and callose deposition (Greenberg and Yao, 2004; Torres and Dangl, 2005; Zipfel and Robatzek, 2010). Therefore, we analyzed rapid cell death, H₂O₂ generation and callose deposition after Vd-toxins treatment in the wild-type, crk5, and crk22 knockout mutants, crk5/CRK5 and crk22/CRK22 complementation lines, and 35S:CRK5 and 35S:CRK22 transgenic overexpression lines (Supplemental Figure S1, C and D). Rapid cell death was detected using the trypan blue staining method. The massive cell death rapidly decreased in the leaves of crk5 and crk22 seedlings after Vd-toxins treatment compared with in the wild-type, while it increased in the leaves of the transgenic overexpression lines (Figure 1B). Next, we used histochemical staining with 3,3ʹ-diaminobenzidine (DAB) to determine the H₂O₂ accumulation. The levels of reddish-brown oxidized DAB precipitate in mutants were substantially less than those in wild-type under Vd-toxins conditions, and transgenic overexpression lines showed greater quantities of reddish-brown precipitate (Figure 1C). Furthermore, aniline blue staining and image analyses were used to measure callose deposition. Lower levels of callose deposition occurred in mutants than in wild-type leaves after Vd-toxins treatment, but higher levels of callose deposition occurred in transgenic overexpression lines (Figure 1, D and E). To determine whether Vd-toxins caused differential damage to plasma membrane of cell, ion leakage was measured in the wild-type and mutants after treatment with Vd-toxins. The results indicated that the plasma membrane in the cells of the mutants was more severely damaged than in the wild-type (Figure 1F). To confirm that disruptions of the CRK5 and CRK22 genes were responsible for the Vd-toxin-sensitive phenotype, complemented lines CRK5/crk5 and CRK22/crk22 were used. The phenotypes of the complementation lines were restored to that of the wild-type (Figure 1, B–F; Supplemental Figure S1, E–G). In addition, we used the clustered regularly interspaced short palindromic repeats (CRISPR/Cas9) technique to generate crk mutants (crk5#5, crk22#2), which showed reductions in Vd-toxin-induced cell death, H₂O₂ accumulation and callose deposition compared with the wild-type (Supplemental Figure S1, E–G).

Next, the relative expression levels of CRK5 and CRK22 in crk5 and 35S:CRK5-1, crk22 and 35S:CRK22-1, were monitored independently. The expression of CRK5 showed no significant difference among crk22, 35S:CRK22-1 and the wild-type, and the expression of CRK22 was not obvious in crk5, 35S:CRK5-1 and the wild-type (Supplemental Figure S1H).

Thus, the results indicated that CRK5 and CRK22 have important regulatory roles in defense responses against Vd-toxins.

Subcellular localizations of CRK5 and CRK22

To further understand the functions of CRK5 and CRK22, we examined the gene expression patterns and the subcellular localizations of CRK5 and CRK22. We used RT–qPCR to analyze the spatial expression pattern of CRK5 and CRK22. The results showed that CRK5 and CRK22 were constitutively expressed, and had higher expression levels in leaves compared with other tissues (Figure 2A).

Figure 2.

Characterization of CRK5 and CRK22. A, Gene expression of CRK5 and CRK22. Relative expression levels in Col-0 (7-day-old) were determined by RT–qPCR. Error bars indicate sd of three replicates. Different letters represent significant differences at P < 0.05, as assessed by one-way ANOVAs with Tukey’s honest significant difference post hoc tests. B, Localization of CRK5 and CRK22 in Arabidopsis protoplasts. CRK5 or CRK22-GFP and CLB1-OFP (marker for plasma membrane) were co-transformed into Arabidopsis protoplasts. The GFP and OFP signals were visualized under confocal microscopy after 16 h incubation. GFP was used as a control. Bar = 5 μm. C, Localizations of CRK5 and CRK22 in the root of CRK5-GFP and CRK22-GFP transgenic plants. The GFP and FM4–64 signals were visualized using confocal microscopy. Bar = 10 μm.

To determine the subcellular localization of CRK5 and CRK22, we generated the 35S:CRK5-GFP and 35S:CRK22-GFP construct and 35S:CBL1-OFP as a plasma membrane marker (Förster et al., 2019), and we co-transformed them into wild-type protoplasts. Under microscopic examination, the CRK5-GFP or CRK22-GFP was co-localized with CBL1-OFP at the plasma membrane (Figure 2B). Next, we used homozygous transgenic plants to investigate the subcellular localizations of CRK5 and CRK22. The FM4–64 dye, which stains the plasma membrane, served as a plasma membrane marker. Using confocal imaging analyses, the green fluorescence of the CRK5-GFP or CRK22-GFP fusion protein co-localized with the FM4–64 fluorescence in the plasma membranes of the root cells of the transgenic plants under normal growth conditions. As a control, the GFP protein alone was expressed in transgenic plants and found to localize throughout the cells, in the nucleus, cytoplasm, and membranes (Figure 2C). To further validate the subcellular localizations of CRK5 and CRK22, we used mannitol to treat leaves of homozygous transgenic plants. The green fluorescence patterns of CRK5-GFP and CRK22-GFP fusion proteins moved with the cell membrane (Supplemental Figure S2A). We also used membrane fractionation to confirm the subcellular localizations of CRK5 and CRK22. The results consistently indicated that CRK5-GFP and CRK22-GFP were mainly detected in the precipitation of the membrane fraction (Figure S2B).

The results suggest that CRK5 and CRK22 are plasma membrane-localized proteins.

CRK5 and CRK22 actively regulate the SA-signaling pathway in defense responses to Vd-toxins

Previous research has shown that SA is a common signaling component of PTI and ETI (Tsuda and Katagiri, 2010), and SA is an important signal molecule in the activation of cotton’s defense responses against Vd-toxins (Zhen and Li, 2004). To determine whether CRK5 and CRK22 regulate the SA-signaling pathway in response to Vd-toxins, we examined the free SA contents and the expression levels of the SA-related gene, SID2, an SA synthesis gene, NPR1, a central SA receptor/regulator, PR1, an SA-signal marker gene, and WRKY53 and WRKY70, SA-signal transcription factors, in wild-type, crk5 and crk22 mutants, and 35S:CRK5-1 and 35S:CRK22-1 lines after the Vd-toxins treatment. The free SA accumulation and the expression of SA-related genes were significantly increased in the transgenic overexpression lines and decreased in the mutants after the Vd-toxins treatment compared with the wild-type (Figure 3). To provide additional support, we generated sid2/35S:CRK5-1 and sid2/35S:CRK22-1 plants by crossing sid2 with 35S:CRK5-1 and 35S:CRK22-1, and tested the cell death and ion leakage after Vd-toxin treatment in sid2, sid2/35S:CRK5-1, sid2/35S:CRK22-1, 35S:CRK5-1, and 35S:CRK22-1 plants. The sid2/35S:CRK5-1 and sid2/35S:CRK22-1 plants displayed reduced cell death and ion leakage levels after the Vd-toxin application compared with 35S:CRK5-1 and 35S:CRK22-1 lines, respectively. The results indicated that the enhanced cell death in the 35S:CRK5-1 and 35S:CRK22-1 lines was related to SA signaling (Supplemental Figure S3).

Figure 3.

CRK5 and CRK22 regulate the SA-signaling pathway involved in defense responses to Vd-toxins. A, Free SA accumulation induced by treating 7-day-old seedlings of Col-0, crk5 and crk22 mutants, and 35S:CRK5-1and 35S:CRK22-1 overexpression lines with 200 µg·mL⁻¹ Vd-toxins for 24 h. Leaves treated with ddH₂O were used as controls. Error bars indicate SD of three replicates. B–F, RT–qPCR analyses of the relative expression levels of SID2, NPR1, PR1, WRKY53, and WRKY70 in 7-day-old seedlings of Col-0, crk5 and crk22 mutants, and 35S:CRK5-1 and 35S:CRK22-1 overexpression lines after treating with 200 µg·mL⁻¹ Vd-toxins for 24 h. Leaves treated with ddH₂O were used as controls. Error bars indicate SD of three replicates. Different letters represent significant differences at P < 0.05, as assessed by one-way ANOVAs with Tukey’s honest significant difference post hoc tests.

Thus, CRK5 and CRK22 actively regulate the SA-signaling pathway involved in defense responses to Vd-toxins.

CRK5 and CRK22 regulate the activation of MPK3 and MPK6 in response to Vd-toxins

MAPK cascades play pivotal roles in the defense-related signaling pathways of plants (Pedley and Martin, 2005), and MAPKs are involved in Vd-toxin-induced responses (Zhao et al., 2020). To further investigate how CRK5 and CRK22 regulate defense responses to Vd-toxins, we first examined the transcript and protein levels, as well as the activation, of MPK3 and MPK6 in wild-type and mutant seedlings.

The transcript levels of MPK3 and MPK6 were not obviously changed after exposure to Vd-toxins, and the changes in wild-type and mutant plants were comparable (Supplemental Figure S4A).

Next, we analyzed the Vd-toxin-induced MAPK activities using anti-phospho-p44/42MAP kinase antibodies that recognize dual phosphorylated activation loops of MAPKs. The MPK3 and MPK6 activities of the wild-type significantly increased within 30 min of Vd-toxin exposure, while their activities were partially elevated in crk5 and crk22 mutants, especially after treating with Vd-toxins for 20 min. However, the MPK3 and MPK6 protein levels did not change in both wild-type and the mutant plants after Vd-toxins exposure (Figure 4, A–D; Supplemental Figure S4B). Furthermore, MPK3 and MPK6 activities increased in 35S:CRK5-1 and 35S:CRK22-1 plant lines after the Vd-toxins treatment compared with the wild-type (Supplemental Figure S4, C and D).

Figure 4.

CRK5 and CRK22 regulate the activation of MPK3 and MPK6 in response to Vd-toxins. A, C, MPK3 and MPK6 kinase activities (p-MPK6 and p-MPK3) were assessed by immunoblotting using anti-phospho-p44/42MAP kinase antibodies. The 7-day-old seedlings of Col-0 and the crk5 and crk22 mutants were treated with 200 µg·mL⁻¹ Vd-toxins for 30 min, and then total protein was extracted after different times for the immunoblot analyses. The MPK3 and MPK6 proteins were assessed by immunoblotting using anti-MPK3 and anti-MPK6 antibodies, respectively. β-Actin was used as the loading control. B, D, Quantification of p-MPK6 and p-MPK3 in (A) and (C), respectively, after a Vd-toxin treatment of 20 min using Fusion software. Error bars indicate sd of three replicates. E, Relative expression levels of FRK1 and NHL10 in the 7-day-old seedlings after treatment with 200 µg·mL⁻¹ Vd-toxins for 24 h as assessed by RT–qPCR. Leaves treated with ddH₂O were used as controls. Total RNA was extracted for the RT–qPCR analysis. Error bars indicate sd of three replicates. F, Cell death induced by Vd-toxins. Cotyledons of 7-day-old plants were treated with 15-μM DEX, and a mock treatment served as the control. After 6 h, seedlings were treated with ddH₂O and 200 µg·mL⁻¹ Vd-toxins for18 h and stained with trypan blue (n = 8). Bars = 200 μm. G, Callose depositions induced by Vd-toxins. Rosette leaves of 4-week-old plants were treated with 15-μM DEX, and a mock treatment served as the control. After 6 h, leaves were treated with 200 µg·mL⁻¹ Vd-toxins for 18 h and stained with aniline blue. Leaves treated with ddH₂O were used as controls. Bars = 100 μm. H, Quantification of the callose deposition in (G). Data represent the numbers of callose deposits per mm². Leaves treated with ddH₂O were used as controls. The bars represent sd for three independent experiments (n = 8). I, K, MPK3 and MPK6 kinase activities in Col-0 and mpk3 and mpk6 mutants. J, L, Quantification of p-MPK6 and p-MPK3 in (I) and (K), respectively, after a Vd-toxin treatment of 30 min using Fusion software. Error bars indicate sd of three replicates (n=3). M, Cell death induced by Vd-toxins in cotyledons of Arabidopsis Col-0, mpk3 and mpk6 mutants, and transgenic MKK5^DD plants. Cotyledons of 7-day-old plants were treated with 15-μM DEX, and a mock treatment served as the control. After 6 h, seedlings were treated with ddH₂O and 200 µg·mL⁻¹ Vd-toxins for18 h and stained with trypan blue as described in the Experimental procedures (n = 8). Bars = 200 μm. Different letters represent significant differences at P < 0.05, as assessed by one-way ANOVAs with Tukey’s honest significant difference post hoc tests.

Because FLG22-INDUCED RECEPTOR KINASE (FRK) 1 and NDR1/HIN1-LIKE (NHL) 10 are MAPK-specific target genes in response to patterns (Bi et al., 2018), we examined the expression levels of FRK1 and NHL10 in the wild-type, crk5, and crk22 mutants, and 35S:CRK5-1 and 35S:CRK22-1 lines after Vd-toxins treatment. The FRK1 and NHL10 expression levels in the mutants decreased by 30%–50% compared with the wild-type. However, the transcript levels of the two genes in transgenic overexpression lines were higher than those in the wild-type (Figure 4E). The results indicated that CRK5 and CRK22 regulate the activation of MPK3 and MPK6 in response to Vd-toxins.

To further dissect the functions of MPK3 and MPK6 in defense responses to Vd-toxins, we used wild-type, mpk3 and mpk6 mutants, and plants harboring MKK5^DD, a constitutively active MKK5 kinase line under the control of a DEX-inducible promoter, which leads to the sustained activation of MPK3 and MPK6 (Supplemental Figure S5A; Zhou et al., 2017), to examine the effects of MPK3/MPK6 activation. Trypan blue and aniline blue staining were employed to determine whether MPK3 and MPK6 affected the HR-like cell death and callose deposition-related phenotypes. The trypan blue staining assay showed that mpk3 and mpk6 mutants had reduced cell death levels after the Vd-toxin application compared with wild-type. Nevertheless, transgenic MKK5^DD plants underwent extensive cell death (Figure 4F). Furthermore, there were lower levels of callose deposition in leaves of mutants compared with wild-type after the Vd-toxins treatment, but there was a higher level of callose deposition in MKK5^DD plants (Figure 4, G and H). Thus, MPK3 and MPK6 appear to be positive regulators of the defense responses to Vd-toxins in Arabidopsis.

The complex relationships between the MAPKs revealed by a transcriptome analysis suggests that the absence of one MAPK influences the functions of the other MAPKs (Frey et al., 2014). We therefore analyzed the MPK3 and MPK6 activities in mpk3 and mpk6 mutants after the Vd-toxins treatment. Indeed, mpk3 showed a higher and longer activation of MPK6 in response to Vd-toxins within 30 min, and mpk6 also displayed a higher and longer activation of MPK3, especially after the Vd-toxins treatment of 30 min. Interestingly, MPK3 and MPK6 protein accumulations remained unaltered in the respective genetic backgrounds (Figure 4, I–L). These data indicate that MPK3 and MPK6 influence each other’s activities regardless of their protein levels in response to Vd-toxins.

Next, we dissected the genetic interactions of CRK5/CRK22 and MPK3/MPK6 by crossing the MKK5^DD line independently with crk5 and crk22 mutants to generated crk5/MKK5^DD and crk22/MKK5^DD double mutants, respectively. The crk5/MKK5^DD and crk22/MKK5^DD lines displayed increased HR-like cell death and MPK3/MPK6 activities compared with crk5 and crk22 mutants, respectively. HR-like cell death and MPK3/MPK6 activities were no different in crk5/MKK^5DD, crk22/MKK^5DD, and MKK^5DD (Figure 4M; Supplemental Figure S5B). Furthermore, the activation of MPK3 and MPK6 compromised the CRK5 or CRK22 loss of function in the SA-signaling pathway in response to Vd-toxins (Supplemental Figure S6). Thus, CRK5 and CRK22 may act upstream of MPK3 and MPK6 to regulate defense responses to Vd-toxins.

MPK3 and MPK6 interact with WRKY70, which is involved in regulating defense responses to Vd-toxins

MPK3 and MPK6 have the same regulatory substrates WRKY22/28/33 (Bigeard et al., 2015). Rice OsWRKY53 directly interacts with OsMPK3/MPK6 in vitro (Hu et al., 2015), and WRKY70 positively regulates SA-responsive genes expression (Wang et al., 2006; Li et al., 2017). To examine whether MPK3 and MPK6 interact with WRKY70 to regulate the SA-signaling pathway in defense responses to Vd-toxins, we used a luciferase complementation imaging (LCI) assay. MPK3 and MPK6 interacted with WRKY70 in Nicotiana benthamiana leaves (Figure 5, A–D). To further determine the interactions between MPK3/6 and WRKY70 in plants, a co-immunoprecipitation (co-IP) assay was carried out in the 35S:WRKY70#1 line. The result displayed MPK3 and MPK6 interacted with WRKY70 (Figure 5E).

Figure 5.

MPK3 and MPK6 interact with WRKY70 in response to Vd-toxins. A, C, Interactions of MPK3 and MPK6 with WRKY70 revealed using an LCI assay in N. benthamiana leaves. nLuc-SGT1 + RAR1-cLuc was used as a positive control. GUS-nLuc + WRKY70-cLuc and nLuc-MPK3/MPK6 + GUS-cLuc were used as negative controls. The experiments were performed three times with similar results. B and D, RT–PCR analyses of MPK3, MPK6, and WRKY70 expression in (A) and (C). E, Co-immunoprecipitation of WRKY70 with MPK3/6 in plants. Total proteins were extracted from 7-day-old transgenic plants expressing GFP or WRKY70-GFP, and immunoprecipitated with GFP beads. Input and immunoprecipitated proteins were analyzed by immunoblotting with anti-GFP, anti-MPK3, and anti-MPK6 antibodies. Transgenic plants expressing GFP used as control. F, Cell death induced by an 18-h Vd-toxin treatment in cotyledons of Arabidopsis Col-0, wrky70-1 and wrky70-3 mutants, and 35S:WRKY70#1 and 35S:WRKY70#2 lines. Cotyledons of 7-day-old plants were treated with 200 μg·mL⁻¹ Vd-toxins for 18 h and stained with trypan blue as described in the Experimental procedures. Leaves treated with ddH₂O were used as controls (n = 8). Bars = 200 μm. G, RT–qPCR analysis of relative expression levels of PR1 in 7-day-old seedlings treated with 200 μg·mL⁻¹ Vd-toxins for 24 h. Leaves treated with ddH₂O were used as controls. Error bars indicate sd of three replicates. H, Cell death induced by Vd-toxins in cotyledons of Arabidopsis Col-0, mpk3, and mpk6 mutants, the 35S:WRKY70#1 overexpression line, and mpk3/WRKY70 and mpk6/WRKY70 lines. Cotyledons of 7-day-old plants were treated with 200 µg·mL⁻¹ Vd-toxins for 18 h and stained with trypan blue as described in the Experimental procedures. Leaves treated with ddH₂O were used as controls (n = 8). Bars = 200 μm. Different letters represent significant differences at P < 0.05 as assessed by one-way ANOVAs with Tukey’s honest significant difference post hoc tests. All the experiments were repeated three times.

We then assessed whether WRKY70 affected defense responses to Vd-toxins using wild-type, wrky70-1 and wrky70-3 mutants (described previously as Knoth et al., 2007), and the transgenic 35S:WRKY70#1 and 35S:WRKY70#2 overexpression lines (Supplemental Figure S7A). The HR-like cell death and PR1 expression were not significantly different in the wild-type and wrky70 mutants after the Vd-toxins treatment; however, their levels noticeably increased in the 35S:WRKY70#1 and 35S:WRKY70#2 lines (Figure 5, F and G; Supplemental Figure S7B). The results indicated that WRKY70 is involved in regulating defense responses to Vd-toxins.

To dissect the genetic interactions of WRKY70 with MPK3/MPK6, we generated mpk3/WRKY70 and mpk6/WRKY70 plants by overexpressing WRKY70 in mpk3 and mpk6 mutants, respectively. Interestingly, mpk3/WRKY70 and mpk6/WRKY70 displayed increased HR-like cell death and electrolyte leakage compared with mpk3 and mpk6 mutants, respectively (Figure 5H; Supplemental Figure S7C). In summary, MPK3 and MPK6 may act upstream of WRKY70 to regulate defense responses to Vd-toxins.

Next, we analyzed the MPK3 and MPK6 activities in wild-type, wrky70-1 and 35S:WRKY70#1. MPK3 and MPK6 activities were partially elevated in 35S:WRKY70#1 compared with wild-type and wrky70 within 30 min of exposure to Vd-toxins (Supplemental Figure S7, D and E). Thus, WRKY70 might repress the MPK3 and MPK6 activities in defense responses to Vd-toxins.

WRKY70 positively regulates TGA2 and TGA6 expression in response to Vd-toxins

TGA2 and TGA6, as transcription factors, positively regulate SA-related defense gene expression and pathogen resistance with functional redundancy (Zhang, 2003; Ding et al., 2018). To examine whether the expression levels of TGA2 and TGA6 are induced by Vd-toxins, their expression levels were analyzed using RT–qPCR in the wild-type, wrky70 mutants, and 35S:WRKY70#1 and 35S:WRKY70#2 lines. The expression levels of TGA2 and TGA6 were not different in the wild-type and wrky70 mutants after the Vd-toxins treatment; however, their transcription levels were significantly increased in 35S:WRKY70#1 and 35S:WRKY70#2 compared with in the wild-type (Figure 6A).

Figure 6.

WRKY70 directly bound to TGA2 and TGA6 promoters to positively regulate TGA2 and TGA6 expression in defense responses to Vd-toxins. A, RT–qPCR analysis of relative expression levels of TGA2 and TGA6 in 7-day-old seedlings of Col-0, wrky70 mutants, and 35S:WRKY70#1 and 35S:WRKY70#2 lines treated with 200 μg⸱mL⁻¹ Vd-toxins for 24 h. Leaves treated with ddH₂O were used as controls. Error bars indicate sd of three replicates. B, Schematic diagrams of plasmids used in GUS activity assays. The promoters of TGA2 and TGA6 were independently fused to the GUS as reporters, and the WRKY70 effector construct was expressed under a CaMV 35S super promoter. The empty vector was used as a control, and luciferase (LUC) was used as the loading control. C, GUS activity assays of N. benthamiana leaves after a 24-h Vd-toxin treatment. The relative GUS activity was calculated as the GUS/LUC ratio. Error bars indicate sd of three replicates. D, Schematics showing the promoter structures of TGA2 and TGA6 genes. Black triangles indicate W-box sequences. Black boxes represent the fragments amplified in the yeast one-hybrid assays and EMSAs. E, Yeast one-hybrid assays showing that WRKY70 bound directly to the promoter fragments of TGA2 and TGA6. F, EMSAs showing that WRKY70 bound directly to the promoter fragments of TGA2 and TGA6. Biotin-labeled fragments decreased as promoter constructs were incubated with His or the WRKY70-His protein. Competitor fragments were added in 100 × and 300 × excesses to analyze the binding specificity. G, Schematic diagrams of plasmids used in the GUS activity assays. H, GUS activity assays of N. benthamiana leaves after a 24-h Vd-toxin treatment. The relative GUS activity was calculated as the GUS/LUC ratio. Error bars indicate sd of three replicates. Different letters represent significant differences at P < 0.05 as assessed by one-way ANOVAs with Tukey’s honest significant difference post hoc tests. All the experiments were repeated three times.

To further determine whether WRKY70 directly regulates TGA2 and TGA6 expression, we performed transient expression experiments (GUS activity assays) in the leaves of N. benthamiana. When the WRKY70 effector construct was co-expressed with the reporter constructs in N. benthamiana leaves, the GUS activity was significantly increased compared with in the control (Figure 6, B and C). The data suggest that WRKY70 positively modulates TGA2 and TGA6 expression in response to Vd-toxins.

WRKY70, a known transcription factor, binds to the W-box (TTGACC/T) motif (Rushton et al., 2010). We generated the 35S:WRKY70-GFP line to examine the subcellular localization of WRKY70. WRKY70-GFP was localized to the nuclei of root cells in transgenic overexpression lines (Supplemental Figure S7F). The analysis of TGA2 and TGA6 promoter region sequences revealed that their promoter regions contain several putative W-box motifs (Figure 6D). Next, yeast one-hybrid assays were performed to determine whether WRKY70 directly binds to the promoters of TGA2 and TGA6. The respective promoter fragments (Figure 6D) were independently fused with a β-galactosidase gene (LacZ), and WRKY70 was fused with AD to activate LacZ expression. X-gal assays of β-galactosidase activity were utilized to detect reporter activation. Interestingly, WRKY70 directly bound to promoter fragments of TGA2 P1 and TGA6 P1 (Figure 6E). Furthermore, we performed electrophoresis mobility shift assays (EMSAs) to validate the WRKY70 binding activity towards promoters of TGA2 and TGA6 in vitro. WRKY70-His, the recombinant protein fused with His tags, was expressed in E. coli and purified using a His antibody. The WRKY70-His fusion protein bound to TGA2 P1 and TGA6 P1 promoter fragments, but no binding was observed in the control His protein. In addition, the binding was dramatically reduced when increasing amounts of cold competitors with the same promoter sequences were added (Figure 6F). The results indicate that WRKY70 directly binds to the promoters of TGA2 and TGA6.

In addition, we performed GUS activity assays to investigate the biological importance of independent interactions between WRKY70 and both MPK3 and MPK6. WRKY70 activated TGA2 and TGA6 expression, whereas the co-expression of MPK3/MPK6 obviously increased WRKY70-activated TGA2/GA6 expression after a 24 h Vd-toxins treatment. Thus, MPK3/MPK6 appear to work together with WRKY70 to enhance the expression of TGA2 and TGA6 (Figure 6, G and H). The interactions of MPK3/MPK6 with WRKY70 increase the transcriptional activity of WRKY70, which is involved in defense responses to Vd-toxins.

Next, we investigated whether CRK5, CRK22, MPK3, and MPK6 regulate TGA2/6 expression. We examined the expression levels of TGA2 and TGA6 after the Vd-toxins treatment. The expression of TGA2 and TGA6 were significantly increased in MKK5^DD and decreased in crk5 and crk22 mutants after the Vd-toxins treatment compared with the wild-type. Furthermore, the activation of MPK3 and MPK6 compromised the CRK5 or CRK22 loss-of-function effect on the expression of TGA2 and TGA6 (Supplemental Figure S8). Thus, CRK5 and CRK22 may act upstream of MPK3 and MPK6 to the expression of TGA2 and TGA6 in response to Vd-toxins.

Thus, MPK3 and MPK6 improve the transcriptional activity of WRKY70, which is involved in plant defense responses to Vd-toxins.

Discussion

CRK5 and CRK22 function as RLKs, which recognize extracellular stimuli and transmit the signals intracellularly during disease resistance (Shiu and Bleecker, 2001, 2003). However, whether and how CRK5 and CRK22 mediate signal transduction and function as sensor- and receptor-like molecules in defense responses against Vd-toxins remain unknown. In this study, we demonstrated that CRK5 and CRK22 positively regulate defense responses to Vd-toxins in Arabidopsis.

The SA pathway is an integral part of defense responses to biotrophic and hemi-biotrophic pathogens (Glazebrook, 2005; Vlot et al., 2009), and SA-related defense genes are involved in defense responses against V. dahliae (Meng et al., 2018). NPR1, known as a central receptor/regulator of SA, contributes to the expression of SA-related genes and to resistance against pathogens (Cao et al., 1994; Ding et al., 2018). Furthermore, NPR1 is required for the up-regulation of CRKs induced by SA (Chen et al., 2004; Wrzaczek et al., 2010). Here, CRK5 and CRK22 positively regulated SA synthesis and modulated the expression of NPR1 and SA-triggered defense genes (Figure 3). Thus, CRK5, SA, and NPR1 may constitute a regulatory loop for defense amplification in response to Vd-toxins.

The WRKY family of transcription factors mediate defense responses by controlling the transcriptional reprogramming of SA-related defense genes and are important positive and negative regulators of defense responses against pathogens (Ren et al., 2008; Chen et al., 2010; Hu et al., 2012). Massive comparative transcriptomic and reverse genetics analyses of plant genes have indicated that they were up-regulated by the V. dahliae-induced expression of WRKY (Zhang et al., 2017). HUB1-mediated H2Bub1 modulates the expression of WRKY33 involved in responses to Vd-toxins (Zhao et al., 2020). Our results indicate that WRKY70 improves resistance to Vd-toxins and regulates the expression of SA-related genes, such as PR1, TGA2, and TGA6 (Figures 5 and 6). WRKY70 directly inhibits the expression of SARD1, a regulator of SA synthesis, to affect the expression of SA-related genes in the absence of a pathogen, but how WRKY70 positively regulates SA signaling in the presence of a pathogen remains unclear (Zhou et al., 2018). TGA2/6 actively regulates the expression of SARD1 to mediate the expression of SA-related genes (Ding et al., 2018). We found that WRKY70 directly bound the promoter sequences of TGA2/6 to enhance their expression (Figure 6). Thus, WRKY70 may modulate the expression of TGA2/6 to positively regulate SA signaling in the presence of a pathogen. Furthermore, NPR1 and NPR3/4 works together with TGA2/6 to active and repress the expression of PR1 and WRKY, respectively (Delaney et al., 1994; Zhang et al., 1999; Ding et al., 2018). We surmise that TGA2/6 and WRKY70 are mutually regulated and jointly facilitate the mediation of SA signaling.

MPK3 and MPK6 function as signaling transduction components of defense responses that amplify signal transduction (Ichimura et al., 2002). In the present study, we demonstrated that CRK5 and CRK22 were required for MPK3 and MPK6 activities (Figure 4). The sustained, but not the transient, activation of MPK3 and MPK6 regulates immune responses redundantly with SA signaling (Tsuda et al., 2013; Wang et al., 2018). MAPK activation induced by Pto AvrRpt2 regulates the SA-responsive genes (such as PR1) expression (Tsuda et al., 2013). Constitutively active Arabidopsis MPK3 produces higher levels of SA and SA-responsive gene expression (Genot et al., 2017). CRK5 and CRK22 might act upstream of MPK3 and MPK6 to regulate SA signaling in defense responses to Vd-toxins. However, how CRK5 and CRK22 activate MPK3 and MPK6 is still unknown. In addition, MPK3 and MPK6 show distinct functions rather than complete functional redundancy (Genot et al., 2017). One MAPK may influence the functions of the other MAPKs (Frey et al., 2014). MPK3 and MPK6 influence each other’s activities regardless of their protein levels (Figure 4). Phosphorylated PTP1 and MKP1 interact with MPK3 and MPK6 to negatively regulate their activation in response to Vd-toxins (Zhao et al., 2020). Thus, MKP1 and PTP1 may affect the interactions between these MAPKs in response to Vd-toxins.

MPK3 and MPK6 control the activities of key enzymes that modulate the jasmonic acid–ethylene and SA-signaling pathways (Hu et al., 2015). MPK3 and MPK6 directly activate ACS2 and ACS6 to regulate the production of ethylene (Liu and Zhang, 2004; Li et al., 2012). WRKY33, which functions as a direct substrate of MPK3 and MPK6, binds to the promoters of ACS2 and ACS6 to modulate the expression of the two genes (Li et al., 2012). OsMPK3 and OsMPK6 have been reported to phosphorylate OsWRKY53, and phosphorylation enhances OsWRKY53 transactivation activity (Chujo et al., 2014; Yoo et al., 2014). We found that MPK3 and MPK6 interact with WRKY70 and enhance its transcriptional activity during defense responses to Vd-toxins (Figures 5 and 6). Thus, MPK3 and MPK6 may enhance the transcriptional activity of key transcription factors to mediate the jasmonic acid–ethylene and SA signals. The PTP–MPK3/6–WRKY pathway regulates H₂O₂ signaling in response to Vd-toxins (Zhao et al., 2020). The MPK3/6–WRKY33–ALD1–Pipecolic acid regulatory loop contributes to defense signal amplification (Wang et al., 2018). In addition, the MPK3/6–WRKY70–TGA2/6 pathway plays a vital role in the response against Vd-toxins. However, further studies are needed to investigate whether MPK3/6 phosphorylates WRKY70 and the biological importance of this phosphorylation. Previous studies have shown that VQ protein (encode VQ motif-containing) can bridge MAPK and WRKY transcription factors to form a ternary complex, and regulate the DNA-binding activity of WRKY. AtVQ21 bridges MPK4 and WRKY33 to form a ternary complex to enhance WRKY33 binding to W-box elements (Andreasson et al.,2005; Lai et al., 2011). AtVQ4 can participate in MPK3/MPK6 and WRKY bridging, and antagonize WRKY-mediated disease resistance. However MPK3 or MPK6 phosphorylate AtVQ4 to reduced protein stability to regulated expression of WRKY-targeted genes (Pecher et al., 2014). Therefore, there is a complex and delicate triangle relationship among MAPK, VQ, and WRKY (Weyhe et al., 2014). This ternary complex provides a specific, accurate, and effective regulation mechanism to enhance the resistance of plants to complex environmental conditions.

In this study, our results revealed that CRK5 and CRK22 regulate defense responses to Vd-toxins through the MPK3/6–WRKY70–TGA2/6 pathway. CRK5 and CRK22 are involved in the induction of SA-related gene expression and MPK3/6 activity. Moreover, MPK3 and MPK6 associate with WRKY70, which directly activates TGA2/6 expression to mediate the expression of SA-related defense genes, such as PR1, thereby modulating the defense responses to Vd-toxins (Figure 7).

Figure 7.

A model of the CRK5 and CRK22 mechanism used to regulate defense responses to Vd-toxins in Arabidopsis. In this model, CRK5 and CRK22 active MAPK and SA signaling. The activated MPK3 and MPK6 work together with WRKY70 to directly induce the expression of TGA2 and TGA6. Moreover, TGA2 and TGA6 work together with NPR1 and NPR3/NPR4 to directly regulate the expression of SA-related defense genes, such as PR1. SA binds to NPR1 to promote activation of the transcription of defense genes, and interacts with NPR3/NPR4 to release the transcriptional inhibition of defense genes.

Materials and methods

Plant materials and growth conditions

The Arabidopsis (A.thaliana) single mutants crk5 (SALK_063519), crk22 (SALK_019124), mpk3 (SALK_100651), mpk6 (SALK_062471), wrky70-1 (SALK_025198), and wrky70-3 (SAIL_720_E01) were used in this study. The 35S:CRK5-1/-3 and 35S:CRK22-1/-4 overexpression lines were generated by transforming wild-type with CRK5-GFP and CRK22-GFP, respectively, driven strongly by the CaMV 35S promoter. The MKK5^DD line was described previously by Zhou et al. (2017). The crk5/MKK5^DD and crk22/MKK5^DD mutants were produced by crossing MKK5^DD with crk5 and crk22, independently. The sid2/35S:CRK5-1 and sid2/35S:CRK22-1 plants were generated by crossing. The 35S:WRKY70#1 and 35S:WRKY70#2 lines were generated by transforming wild-type plants with WRKY70-GFP driven strongly by the CaMV 35S promoter. The mpk3/WRKY70 and mpk6/WRKY70 mutants were generated by transforming mpk3 and mpk6 mutants, respectively, with WRKY70-GFP driven strongly by the CaMV 35S promoter. The seeds were sterilized with 0.2% sodium hypochlorite (v/v) for 15 min, rinsed with water, and grown on half-strength Murashige and Skoog medium (Murashige and Skoog, 1962) supplemented with 1% sucrose (w/v) and 0.8% agar (w/v) at 16°C or 22°C with a 16-h light/8-h dark cycle.

Preparation of crude Vd-toxins

Vd-toxins, produced by V. dahliae (Vd991) in secondary metabolic process. Vd-toxins is an important factor causing wilt and protein–lipid–polysaccharide complexes, whose main component is acid glycoprotein, and the contents of protein and sugar are 85.26% and 14.74%, respectively (Chen and Liu, 2012). The Verticillium spp. culture filtrate purification methods were described previously (Jia et al., 2007; Shi et al., 2009). The fungal culture was filtered and centrifuged at 5,000g for 20 min to remove the spores. The supernatant was frozen in −40°C for 24 h, lyophilized in −50°C for 36 h, and dissolved in ddH₂O to make a 1 mg·mL⁻¹ solution. The solution was dialyzed with 1-kDa dialysis membranes (MWCO) in 4°C for 24 h. The solution was then refiltered through a 0.45-mm Millipore filter. The resulting filtrate was used as crude Vd-toxins extract for further experiments. For Vd-toxin treatment assays, seedlings or leaves were soaked in 200 µg·mL⁻¹ Vd-toxins.

Cell fraction preparation

The plant material was lapped with liquid nitrogen and extracted in extraction buffer (50-mM Tris, pH 8.0, 2-mM EDTA, 20% glycerol, 1-mM DTT, 0.1% Triton X-100, and protease inhibitor cocktail). The samples were then centrifuged at 4°C, 5,000g for 5 min to remove debris. The supernatant was total protein (T). The supernatant was performed by ultracentrifugation of the homogenate for 1 h at 4°C, 100,000g to obtain a soluble fraction (S) and a pellet (P). Anti-PEPC (Agrisera, Cat#AS09458) and anti-AHA2 antibodies (Zhang et al., 2020) were used as cytosolic and membrane fraction markers, respectively.

Free SA accumulation assays

Cotyledons of 7-day-old plants were treated with 200 µg·mL⁻¹ Vd-toxins for 24 h, and proteins were extracted using an extraction buffer (isopropyl alcohol:H₂O:HCl = 2:1:0.002, v/v/v) with a known concentration of interior labeling at 900g nd 4°C for 30 min, and then, CHCl₃ was used for a second extraction at 900g and 4°C for 30 min. Further, free SA was determined using ultra-high-performance liquid chromatography and high-resolution mass spectrometry after drying under nitrogen gas and being dissolved in NaOH. The ultra-high-performance liquid chromatography and high resolution mass spectrometry protocol was described previously (Cao et al., 2016).

Trypan blue staining

Cotyledons of 7-day-old plants were treated with 200 µg·mL⁻¹ Vd-toxins for 18 h and stained with trypan blue. The trypan blue staining protocol was described previously (Bowling et al., 1997).

DAB staining

Cotyledons of 4-week-old plants were treated with 200 µg·mL⁻¹ Vd-toxins for 24 h and stained with DAB for 12 h. The DAB staining protocol was described previously (Thordal-Christensen et al., 1997).

Callose staining

Cotyledons of 4-week-old plants were treated with 200 µg·mL⁻¹ Vd-toxins for 24 h and stained with aniline blue for 2 h. The callose staining protocol was described previously (Chen et al., 2014).

Ion leakage assays

The assays were performed as previously described (Overmyer et al., 2005; Wrzaczek et al., 2015). We did some remodeling. Seedlings of wild-type and mutants, 7-day-old, were treated with 200 µg·mL⁻¹ Vd-toxins for 18 h. The tubes were shaken overnight, and the conductivity of the solution was measured as S₁ using a conductivity meter. 200 µg·mL⁻¹ Vd-toxins without seedlings was measured as S₁0. Tubes were then autoclaved, and after the tubes cooled to room temperature, conductivities of solutions were again measured as S₂ or S₂0. The percentage of electrolyte leakage was calculated as S = S₁−S₀.

RT–qPCR analysis

Total RNA of Arabidopsis seedlings was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and treated with DNaseI in accordance with the manufacturer’s specifications (Invitrogen). The specific steps of quantitative PCR were described previously in Zhou et al. (2017). The β-Actin gene was used as an internal standard for the normalization of the test gene expression levels. Each RT–qPCR analysis contained three biological replicates, with each having 60–80 individual plants.

Yeast one-hybrid assays

The yeast one-hybrid assay protocol was as described by Wang et al. (2019) and Li et al. (2010). WRKY70 and promoter fragments of TGA2/6 were cloned into pB42AD and pLacZi, respectively. Different combinations of the fusion constructs were co-transformed into Saccharomyces cerevisiae strain EGY48. The transformants were tested on synthetic dropout (SD) medium lacking uracil and leucine, but containing 20 μg·mL⁻¹ 5-bromo-4-chloro-3-indolyl-beta-d-galacto-pyranoside.

Yeast two-hybrid assays

The coding sequences of MPK3 and MPK6 were independently inserted into the pGBKT7 vector (Clontech, Mountain View, CA, USA) as the bait plasmids, while WRKY70 was inserted into the pGADT7 vector (Clontech) as the prey plasmid. All the PCR primer sequences are listed in Supplemental Table S1 Different combinations of the fusion constructs were co-transformed into AH109. Experimental operations after transformation were in accordance with those of Zhou et al. (2017).

LCI assays

MPK3 and MPK6 cDNA sequences were cloned independently into the nLuc vector, and the WRKY70 sequence was cloned into the cLuc vector. The PCR primers are listed in Supplemental Table S1. nLuc-SGT1 + RAR1-cLuc was used as a positive control. GUS-nluc and GUS-cluc were used as negative controls. Agrobacterium tumefaciens strain GV3101, containing different combinations of constructs, was injected into 5-week-old N.benthamiana. Leaves were collected 2–3 days after infiltration. The LCI assays were performed as previously described (Zhou et al., 2017).

GUS activity assays

The promoters of TGA2 and TGA6 were cloned independently into pCAMBIA1391 vector as reporters, WRKY70 was cloned into pSuper1300-GFP as an effector, and MPK3 and MPK6 were cloned independently into pSuper1300-MYC as other effectors. All the primer sequences are listed in Supplemental Table S1. Different combinations of reports and effectors were transfected into A. tumefaciens strain GV3101 and then expressed in N. benthamiana. The GUS activity assays were performed as previously described (Feng et al., 2014). The binding activity (GUS activity) levels of WRKY70 to the TGA2 and TGA6 promoters were determined using GUS/LUC ratios.

Co-IP assays

Transgenic 35S:WRKY70#1 plants were used to detect the interactions of mpk3/6 and WRKY70. The total proteins extracted from stable transgenic plants were immunoprecipitated with anti-GFP beads (Thermo Fisher Scientific, Waltham, MA, USA). The immunoprecipitates were separated on a 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) gel and detected with anti-MPK3 and anti-MPK6 (Sigma-Aldrich).

EMSAs

The WRKY domain of WRKY70 (108–213 aa) was amplified and inserted into the pET30a vector and then transformed into E. coli (BL21). The proteins were purified using His beads. Biotin-labeled or -unlabeled primers were used to obtain fragments of TGA2P1, TGA2P2, TGA6P1, and TGA6P2 by PCR amplification. All the primer sequences are listed in Supplemental Table S1. EMSAs were performed as described in the LightShift EMSA Optimization Control Kit’s instructions (Thermo, Rockford, IL, USA). An EMSA protocol was also consulted (Li et al., 2017).

MAP kinase and protein assays

At several time points, 7-day-old Arabidopsis seedlings were treated with 200 μg·mL⁻¹ Vd-toxins and lysed using a Plant Protein Extraction Kit (Beijing ComWin Biotech) with 1-mM proteinase inhibitor cocktail and phosphatase inhibitor (Real-Times, China). For immunoblotting analyses, 60 µg of total proteins per lane was subjected to 8% SDS–PAGE. After electrophoresis, the proteins were transferred onto polyvinylidene difluoride membranes. They were then incubated with anti-phospho-p44/42 MAPK (Cell Signaling Technology, Danvers, MA, USA) anti-AtMPK3 (Sigma), and anti-AtMPK6 (Sigma) primary antibodies and then the corresponding secondary antibodies. Then, the proteins were visualized using an enhanced chemiluminescent substrate (Thermo Fisher, Waltham, MA, USA). Quantification of p-MPK6 and p-MPK3 was performed using Fusion software, and β-Actin was used as a loading control.

Statistical analyses

Statistical data were analyzed using SPSS statistics software. One-way analysis of variance (ANOVA) with Tukey’s honestly significant difference post hoc tests were used to analyze data. Differences were considered significant at the P < 0.05 level.

Accession numbers

Sequence data for the genes described in this article can be found in TAIR (https://www.arabidopsis.org/) under the following accession numbers: AT4G23130 for CRK5, AT4G23300 for CRK22, AT2G35980 for NHL10, AT2G19190 for FRK1, AT1G74710 for SID2, AT1G64280 for NPR1, AT2G14610 for PR1, AT3G56400 for WRKY70, AT5G06950 for TGA2, AT3G12250 for TGA6, AT4G23810 for WRKY53, AT3G45640 for MPK3, and AT2G43790 for MPK6.

Supplemental data

The following material is available in the online version of this article.

Supplemental Figure S1. CRK5 and CRK22 positively regulate defense responses to Vd-toxins in Arabidopsis.

Supplemental Figure S2. Subcellular localizations of CRK5 and CRK22.

Supplemental Figure S3. Ion leakage in sid2, sid2/35S:CRK5-1, 35S:CRK5-1, sid2/35S:CRK22-1, and 35S:CRK22-1.

Supplemental Figure S4. Activation levels of MPK3 and MPK6 induced by Vd-toxins in transgenic Arabidopsis lines.

Supplemental Figure S5. Transgenic MKK5^DD, crk5/MKK^5DD, and crk22/MKK^5DD plants showed the sustained activation of MPK3 and MPK6.

Supplemental Figure S6. Relative expression levels of SA-related genes.

Supplemental Figure S7. Overexpression of WRKY70 increased the cell death and repressed MPK activities.

Supplemental Figure S8. CRK5 and CRK22 may act upstream of MPK3 and MPK6 to the expression of TGA2 and TGA6.

Supplemental Table S1. Primers used in this work.

 

J.Z. and Y.L. designed the research and wrote the article. J.Z., Y.S., and X.Z. performed specific experiments and analyzed the data. Y.L. revised and edited the article.

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plphys/pages/general-instructions) is: Yingzhang Li (liyingzhang@cau.edu.cn).