Chinese wild V. quinquangularis clone Shang-24 seedlings were grown in the grape germplasm resources orchard at the Northwest A&F University, Yangling, Shaanxi, China. Wild type (WT) A. thaliana (ecotype type, Columbia- 0), the pad4 mutant and N. benthamiana were preserved in our lab. A. thaliana plants were grown under the following conditions: 21°C, 50% relative humidity and a long-day photoperiod (16 h- light/ 8 h- dark). N. benthamiana was grown in a growth chamber under the following conditions: 26°C, 50% relative humidity and a long-day photoperiod (16 h- light/ 8 h- dark). G. cichoracearum was cultured on A. thaliana pad4 mutant plants at 21°C and a photoperiod of 16 h light/8 h dark. PstDC3000 was preserved at -80°C. B. cinerea was maintained at 22°C on Potato Glucose Agar as described by Wan et al. (2015).

Young leaves of 2-year-old grapes were sprayed with 100 μM SA or 50 μM MeJA (Repka et al., 2004; Wang and Li, 2006). Sterile distilled water was used as a mock control. Samples were collected at 1, 12, 24, and 48 h post treatment (hpt) and frozen at -80°C.

Quantitative real-time PCR analysis was performed as previously described (Tu et al., 2016). The E.Z.N.A.^® Plant RNA Kit (Omega Bio-tek, USA, R6827-01) was used to extract grapevine total RNA and the RNA prep plant kit (Tiangen Biotech., China) was used to extract A. thaliana RNA. Prime Script TMR Tase (TaKaRa Biotechnology, Dalian, China) was used to synthesize first-strand cDNA. We used SYBR green (TaKaRa Biotechnology) and an IQ5 real-time PCR instrument (Bio-Rad, Hercules, CA, USA) to conduct quantitative real-time PCR (qRT-PCR) analysis. All of the above procedures were carried out according to the manufacturers’ instructions. VvActin1 or AtActin2 were used as references genes. Primers used for the qRT-PCR analyses are listed in Supplementary Table S1. Three biological replicates were analyzed for each experiment and three technical replicates for each biological replicate. Relative expression levels were analyzed with the IQ5 software using the Normalized Expression Method.

Total RNA extractions from leaves of V. quinquangularis clone Shang-24 and first-strand cDNA synthesis were performed as described above. The open reading frame (ORF) of VqWRKY52 was amplified by PCR using the specific primers F1 (5′-CGGGATCCATGGAGAACATGGGAAGTTGGGAAC-3′) and R1 (5′-GGGGTACCTTAAAAGAATCCCAGGTGGTCGAAGTTA-3′). The PCR product was cloned into the pGEM-T easy vector (Promega, Madison, WI, USA), to give the construct pGEM-Teasy-VqWRKY52, which was then sequenced. To obtain the over-expression vector, the VqWRKY52 ORF from pGEM-Teasy-VqWRKY52 was inserted immediately downstream of the CaMV35S promoter in the plant over-expression vector, pCambia 2300 (Cambia, Brisbane, QLD, Australia) using the BamHI and KpnI restriction endonucleases.

Grapevine DNA was extracted from leaves of V. quinquangularis clone Shang-24 as previously described (Tu et al., 2016). A 2107 bp VqWRKY52 promoter fragment was amplified by PCR from genomic DNA using the gene specific primers F2 (5′-CCCAAGCTTCGGAATTCGCGTGATCAAAGTAATTGAGG-3′) and R2 (5′-TCCCCCGGGTTTTAAACCACCCAAAGAAGAAGAA-3′), and inserted into the binary vector pBI121 (Clontech, Palo Alto, CA, USA), to replace the CaMV 35S promoter, upstream from the β-glucuronidase (GUS) reporter gene, using the HindIII and SmaI restriction endonucleases. The resulting vector was named pro_{V qWRKY 52}:GUS. pBI121 was used as a positive control and renamed pro_35S:GUS.

Each of the above constructs was introduced into Agrobacterium tumefaciens strain GV3101, and these resulting A. tumefaciens were used to transform A. thaliana, using the floral dip method (Clough and Bent, 1998). Transient expression in N. benthamiana was performed by infiltration as previously described (Guan et al., 2014). The infiltrated plants were maintained for an additional 3 days under the same conditions and the hormone treatments were performed as described above. For the A. thaliana transformation, the three T3 homozygous lines with the strongest resistance to powdery mildew (#28, #30, and #33) were selected and used for all subsequent experiments. To assess GUS activity in the transgenic pro_{V qWRKY 52}:GUS plants, three independent transgenic T3 lines were analyzed.

β-Glucuronidase activity assays were performed as previously described (Jefferson et al., 1987). A vacuum was applied to the samples for 30 min prior to incubation at 37°C for 24–48 h in the staining solution [1 mM 5-bromo-4-chloro-3-indolyl-β-D-glucuronide (X-gluc; Biosynth AG), 100 mM sodium phosphate (pH 7.0), 0.5 mM K₃Fe(CN)₆, 0.5 mM K₄Fe(CN)₆, 0.1% Triton X-100 and 0.1 mM EDTA]. Chlorophyll was then cleared from the samples with 70% ethanol, and the samples were viewed under a light microscope (BX53, Olympus, Japan).

Four-week-old T3 transgenic and WT plants were inoculated with G. cichoracearum, and the number of conidiophores per colony was counted at 7 days post inoculation (dpi), as previously described (Wang et al., 2011). Samples for the expression profile analysis of defense related genes and the accumulation of O₂^- were collected at 0, 24, 48, and 72 h post inoculation (hpi), samples used for analysis of fungal structures and monitoring cell death were collected at 7 dpi.

Four-week-old plants were inoculated with PstDC3000 by dipping whole rosettes in PstDC3000 solutions (10⁸ cfu/mL, in 10 mM Mg₂SO₄ supplemented with 0.025% Silwet77) as previously described (Wen et al., 2015; Guo et al., 2016). The inoculated plants were maintained under 90% relative humidity for 24 h before being moved to normal growth conditions. Samples used for morphological observation were taken at 5 dpi. Samples used for the expression analysis of defense related genes were collected at 0, 6, 12, and 24 hpi. Leaves inoculated with PstDC3000 by infiltration (Varet et al., 2003) were used to monitor the bacterial growth at 3 dpi, and the detection of cell death was performed at 0, 24, 48, and 72 hpi, while accumulation of O₂^- and H₂O₂ was measured at 0 and 72 hpi.

The B. cinerea conidial suspension (1.5 × 10⁶ conidia/ml) used for inoculation was prepared as previously described (Wan et al., 2015). Detached leaves were used for morphological observation and the lesion diameter analysis, which were performed by droplet inoculating with 10 μL of the conidial suspension, as previously described (Guo et al., 2016). Samples used for morphological observation and lesion diameters analysis were photographed and measured at 3 dpi. Adult plants were inoculated by spraying, as previously described (Wan et al., 2015), and were then used for analyzing the expression of defense related genes at 0, 12, 24, and 48 hpi. Cell death was measured at 0, 24, 48, and 72 hpi and the accumulation of O₂^- and H₂O₂ was measured at 0 and 48 hpi.

Cell death and the fungal structures were visualized by staining with trypan blue as previously described (Vogel and Somerville, 2000). Diaminobenzidine (DAB) staining was used to detect the accumulation of H₂O₂ (Fryer, 2002), and nitro blue tetrazolium (NBT) staining was used to detect the accumulation of O₂^- (Kim et al., 2011), as previously described. All samples were imaged with a light microscope (Olympus, Japan). At least six leaves were used for each independent experiment and three biological replicates were analyzed.

Data analysis and plotting were performed using Microsoft Excel (Microsoft Corporation, USA) and Sigma plot (v. 10.0, Systat, Inc., Point Richmond, CA, USA). Significant differences were assessed through paired t-test using the SPSS Statistics software (IBM China Company, Ltd, Beijing, China) as previously described (Tu et al., 2016). All experiments were performed using three biological replicates, with each biological replicate having three technical replicates.

The accession numbers of the genes used in this paper are found in The Arabidopsis Information Resource¹ and the grape genome Sequence²: AtActin2 (AT3G18780), AtEDS1 (AT3G48090), AtPR1 (AT2G14610), AtPR2 (AT3G57260), AtPR5 (AT1G75040), AtPDF1.2 (At5g44420), AtICS1 (At1g74710), VvActin1 (AY680701), VqWRKY52 (KY411919).

In previous studies, a transcriptome analysis of Chinese wild V. quinquangularis clone Shang-24 at different time points after inoculation with E. necator indicated that VqWRKY52 was highly induced by this treatment (Jiao et al., 2015). To identify which hormones could affect the expression of VqWRKY52, we measured its expression levels in V. quinquangularis clone Shang-24 at 1, 12, 24, and 48 h post treatment with SA or MeJA. We observed that the expression of VqWRKY52 was strongly induced by SA treatment, but not by MeJA, at 12 h post treatment, compared to the mock treatment, followed by a decrease at 24 h compared with the expression levels at 1 and 12 h (Figure 1A).

To validate this result, a 2107 bp VqWRKY52 promoter fragment was fused to the GUS reporter gene and transiently expressed in N. benthamiana, using a Pro_35S:GUS construct as a positive control. The transiently expressing leaves were treated with SA, MeJA or a negative ddH₂O control and then subjected to GUS staining. Leaves expressing the Pro_35S:GUS construct showed strong GUS activity, while little activity was detected in the leaves transiently expressing the Pro_{V qWRKY 52}:GUS construct. However, when Pro_{V qWRKY 52}:GUS expressing leaves were treated with SA, high GUS activity levels were observed, while only low levels of activity were detected in MeJA treated leaves (Figure 1B). This was consistent with the expression results presented in Figure 1A.

Gene specific primers were designed according to the VvWRKY52 (GSVIVT01028718001) cDNA sequence from the Grape Genome Database (12ײ) and used to amplify the VqWRKY52 ORF. The VqWRKY52 coding sequence (CDS) is 1092 bp, encoding a 364 amino acid protein, and the nucleotide sequence had 99.27% identity to the V. vinifera homolog, with only eight single nucleotide polymorphisms (SNPs) found between the CDS from the two grape genotypes (Supplementary Figure S1). The corresponding deduced amino acid sequences shared 99.18% identity (Supplementary Figure S2).

To further understand the temporal and spatial expression profile of VqWRKY52, transgenic A. thaliana lines constitutively expressing the Pro_{V qWRKY 52}:GUS construct were generated. Plants from the T3 generation were stained for GUS activity and while not activity was detected at the germination stage (Figure 2A), during early seedling growth, low GUS activity was observed in the cotyledon tips and roots (Figure 2B). Two-week-old plants grown on Murashige-Skoog (MS) basal medium showed strong GUS activity in all organs, but especially in leaves (Figure 2C). Aging leaves of 3-week-old plants in soil showed more GUS activity than the young leaves (Figure 2D), and strong GUS activity was observed in flowers and siliques, but not in the seeds and anthers (Figures 2E–H).

To further investigate the putative function of VqWRKY52 in defense process, three T3 generation transgenic A. thaliana lines expressing VqWRKY52 (Supplementary Figure S3), together with WT, and the A. thaliana pad4 mutant, which is susceptible to powdery mildew, were inoculated with G. cichoracearum. Over-expressing lines showed enhanced resistance to G. cichoracearum at 7 dpi (Figure 3A) and showed a large number of dead cells, while minimal cell death was observed in the WT plants and no obvious cell death occurred in the pad4 mutant (Figure 3B). The G. cichoracearum colonies growing on the pad4 mutant were the largest, followed by those on the WT plant, while the three over-expressing lines had the smallest colonies. This correlated with the extent of the cell death triggered by G. cichoracearum that was observed at the infection site, which is consistent with cell death restricting fungal growth (Figure 3B). We also counted the number of conidiophores per colony from the five different genotypes (Figure 3C) and determined that the three transgenic lines had a significantly fewer than the WT plants.

Since the accumulation of the superoxide anion (O₂^-) is associated with cell death (Yi et al., 2011), we measured O₂^- levels in the different genotypes at 0, 24, 48, and 72 hpi. A significant difference was observed between WT, pad4, and the three transgenic lines at 48, 72 hpi, with high levels in the latter. Interestingly, almost all of the leaves from the tested lines accumulated O₂^- at 24 hpi, while the mottled leaves with spots of O₂^- accumulation were appeared at 48 and 72 hpi (Figure 3D).

Since VqWRKY52 expression was induced by SA (Figures 1A,B), we hypothesized that it operates via SA mediated signaling pathways and the expression levels of marker genes involved in SA signaling pathways in Arabidopsis will be affected in three over-expressing lines. To test this, we measured the expression of four marker genes. The expression of AtICS1, which is involved in SA biosynthesis and affects SA accumulation (Strawn et al., 2007), was higher in the over-expressing lines than in the WT at all three time point post-inoculation, showing an initial increase at 24 hpi, peaking at 48 hpi and declining again at 72 hpi. AtEDS1 is involved in the SA related signaling pathway and plays essential roles upstream of SA biosynthesis (Rustérucci and Parker, 2001). The expression of this gene was similar to AtICS1 at 24 and 48 hpi; however, its expression levels were lower in the over-expressing lines than in the WT at 72 hpi. In addition, the expression of AtPR1 and AtPR5 increased at 24, 48, and 72 hpi compared to 0 hpi, with expression being higher in the three over-expression lines than in the WT (Figure 4).

To examine the association between VqWRKY52 and responses to infection by PstDC3000, three T3 generation transgenic lines over-expressing VqWRKY52 and WT plants were inoculated with PstDC3000 and examined at 5 dpi. The three transgenic lines showed increased resistance to PstDC3000, based on less severe disease symptoms and fewer diseased leaves than the WT plants (Figure 5A). We also measured the growth of PstDC3000 by counting the bacterial numbers per unit leaf area at 3 dpi. As shown in Figure 5B, we observed less growth in the three over-expressing lines than in WT plants, which suggested VqWRKY52 mediated suppression of bacterial growth. The transgenic lines showed enhanced resistance to PstDC3000 and strongly induced cell death. Specifically, no cell death was detected in the three transgenic lines and WT plants at 0 hpi, and less cell death was observed in the transgenic lines than in the WT plants at 24 hpi. Strong cell death was apparent at 48 hpi and was enhanced in the overexpression lines at 72 hpi. At this time point less cell death was detected in the WT plants to an extent that was similar to that observed in the transgenic lines at 24 hpi (Figure 5C). Strong cell death may be associated with a burst of ROS production and so we examined the accumulation of O₂^- and H₂O₂ at 72 hpi, where more cell death was detected in the transgenic lines than in the WT plants. We found a larger accumulation of O₂^- and H₂O₂ in the transgenic lines than in the WT plants (Figure 5D).

Since cell death induced by PstDC3000 first appeared at 24 hpi in the three over-expressing lines and WT plants, we analyzed the expression levels of defense related genes at earlier times points, specifically 0, 6, 12, and 24 hpi. The expression of AtPR1, AtPR5, and AtEDS1 was induced in WT plants, but inhibited in the transgenic lines to varying degrees, which is the opposite of their expression pattern following the powdery mildew infection. This was also true for the AtPDF1.2 gene, which is associated with the MeJA signaling (Xu et al., 2006) (Figure 6).

Cell death induced by pathogens is important in the resistance to biotrophic pathogens, and we also wanted to determine whether this was also true for necrotrophic pathogens, such as B. cinerea. We observed the detached leaves of three overexpression lines and WT plants that had been inoculated with B. cinerea at 3 dpi and measured the lesion diameters of infected leaves, which were larger in the overexpression lines (Figures 7A,B). Subsequently, three plants from each genotype were incubated with B. cinerea by spraying, and cell death was detected at 0, 24, 48, and 72 hpi, with ROS staining at 0 and 48 hpi. Compared with the WT plants, cell death was more extensive in the transgenic lines at 24, 48, and 72 hpi, but was not observed at 0 hpi (Figure 7C). Minimal O₂^- and H₂O₂ accumulation was observed at 0 hpi in the VqWRKY52 over-expressing lines and WT plants, while higher levels were detected in the three over-expressing lines than in WT plants at 48 hpi (Figure 7D).

To investigate whether the over-expressing lines had altered expression of defense-related genes, we measured the transcript levels of AtPR1, AtPR2, AtPR5, and AtPDF1.2 at different time points after B. cinerea inoculation. All plants showed low expression of AtPDF1.2 before inoculation and an induction of expression at 12 hpi, with the three transgenic lines having higher levels than the WT plants. At 24 hpi, the expression in WT plants was high, whereas it declined in the three transgenic lines such that the WT plants had the highest levels. Low AtPDF1.2 expression levels were detected at 48 hpi in all the plants, while the expression of AtPR2 and AtPR5 was low at 12, 24, and 48 hpi in the WT plants compared with 0 hpi, but were significantly higher in the overexpression lines than in the WT at all three time point, indicating less inhibition. Although the expression of AtPR1 was more highly induced in the overexpression lines than in the WT plants following the incubation with powdery mildew, no significant differences were observed in these plants following incubation with B. cinerea (Figure 8).