Transcription factor VvWRKY70 inhibits both norisoprenoid and flavonol biosynthesis in grape

Abstract

Norisoprenoids and flavonols are important secondary metabolites in grape berries (Vitis vinifera L.). The former is a class of ubiquitous flavor and fragrance compounds produced by the cleavage of carotenoids, and the latter, which is derived from the flavonoid metabolic pathway, has been proposed as a general quality marker for red grapes. However, the transcriptional regulatory mechanisms underlying norisoprenoid and flavonol production are still not fully understood. In this study, we characterized a transcription factor, VvWRKY70, as a repressor of both norisoprenoid and flavonol biosynthesis in grape berries, and its expression was downregulated by light and high-temperature treatment. Overexpressing VvWRKY70 in grape calli reduced norisoprenoid and flavonol production, particularly under light exposure or at high temperature, by repressing the expression of several related genes in the isoprenoid and flavonoid metabolic pathways. VvWRKY70 downregulated β-CAROTENE HYDROXYLASE 2 (VvBCH2) and CHALCONE SYNTHASE 3 (VvCHS3) expression based on yeast 1-hybrid analysis combined with electrophoretic mobility shift assay and chromatin immunoprecipitation-quantitative PCR. We discuss the role of VvWRKY70 in the coordinated regulatory network of isoprenoid and flavonoid metabolism. These findings provide a theoretical basis to improve flavor, color, and other comprehensive qualities of fruit crops and their processing products.

Introduction

Norisoprenoids are volatile apocarotenoids that participate in grape (Vitis vinifera L.) and wine aroma by contributing to the floral or fruity aroma, particularly in neutral grape varieties such as ‘Chardonnay’, ‘Cabernet Sauvignon’, and their mainstream dry-styled wines (Mendes-Pinto 2009). Some norisoprenoid compounds serve as bioactive molecules that are related to photooxidative stress response and signaling (Moreno et al. 2021). Norisoprenoids can be biosynthesized from carotenoids via the isoprenoid pathway by CAROTENOID CLEAVAGE DIOXYGENASE (CCD). For example, phytoene and lycopene are the precursors of geranylacetone and 6-methyl-5-hepton-2-one (MHO), β-cyclocitral and β-ionone are derived from β-carotene (Lashbrooke et al. 2013), and the neoxanthin cleavage product, grasshopper ketone, is proposed as the precursor of β-damascenone (Mendes-Pinto 2009). 1-DEOXY-D-XYLULOSE 5-PHOSPHATE SYNTHASE (DXS) is the first key enzyme in the supply of precursors for isoprenoid biosynthesis. PHYTOENE SYNTHASE (PSY) is the main flux control step and its activity is related to carotenoid content in many plants (Sun and Li 2020). β-CAROTENE HYDROXYLASE (BCH) is a key enzyme in the biosynthesis of zeaxanthin from β-carotene (Zhu et al. 2017). Carotenoid accumulation is modulated by several transcription factors (TFs) by directly regulating target genes in the upstream 2-C-methyl-D-erythritol 4-phosphate (MEP) pathway or carotenoid biosynthesis (Sun and Li 2020; Yuan et al. 2022). However, reports on the TFs that regulate norisoprenoid accumulation are limited. OfWRKY3 and ETHYLENE RESPONSIVE FACTOR 61 (OfERF61) positively regulate the expression of OfCCD4 and β-ionone accumulation in sweet osmanthus (Osmanthus fragrans) (Han et al. 2016; Han et al. 2019), and SlMYB75 regulates production of β-cyclocitral in tomato (Solanum lycopersicum) (Jian et al. 2019). Our previous study demonstrated that VvMADS4 directly downregulates the expression of VvCCD4b in grape (Meng et al. 2020). Owing to the difficulty of transforming fruit crop and long juvenility, current knowledge about transcriptional regulation of fruit flavor is still limited, particularly in grape berries.

Flavonoids, as important secondary metabolites in grape berries, are mainly composed of flavonols, anthocyanins, and proanthocyanidins (PAs). Anthocyanins and PAs contribute to color, bitterness, astringency, and antioxidant activity, while flavonols contribute to color stability by copigmentation with anthocyanin (Zhao et al. 2022). Flavonols protect plants against UV light and play a role in auxin transport (Czemmel et al. 2009). Compared with anthocyanins and PAs, there have been relatively few reports on TFs that specifically modulate flavonol metabolism. In Arabidopsis (Arabidopsis thaliana), 3 R2R3-MYB proteins MYB11, MYB12, and MYB111 activated CHALCONE SYNTHASE (CHS), CHALCONE ISOMERASE (CHI), FLAVANONE-3-HYDROXYLASE (F3H), and FLAVONOL SYNTHASE 1 (FLS1) and modulate flavonol content (Stracke et al. 2010). The flavonol-specific TFs reported in fruits, such as apple (Malus domestica) (Wang, Xu, et al. 2017) and peach (Prunus persica) (Cao et al. 2019), are mainly FLAVONOL-SPECIFIC MYB TRANSCRIPTION FACTOR 1 (MYBF1) and its homologous genes. In apples, except for R2R3-MYBs (Wang, Xu, et al. 2017), other TFs such as ELONGATED HYPOCOTYL 5 (MdHY5), NAC (NAM, ATAF1/2, and CUC2) TRANSCRIPTION FACTOR 9 (MdNAC9), and SCARECROW-LIKE PROTEIN 8 (MdSCL8) regulate flavonol accumulation by regulating the expression of MdFLS1 (Zhang et al. 2022). In grape, VvMYBF1 was the first specific regulator of flavonol synthesis to be reported, which targeted VvCHS, VvCHI, and VvFLS1 (Czemmel et al. 2009). Overexpressing VvHY5 upregulated a set of flavonol-related genes including VvFLS1 and VvMYBF1 and increased flavonol accumulation (Loyola et al. 2016). However, additional TFs especially repressors regulating fruit flavonol biosynthesis require further investigation.

Isoprenoid and flavonoid derivatives are important components contributing to the aroma, coloration, taste, and copigmentation of wine; therefore, in viticulture research, changes in both are often analyzed simultaneously. In our previous studies, climate change or vineyard management often simultaneously increased or decreased grape aroma and flavonoid profiles (Tian et al. 2023). In plants, norisoprenoids and flavonols are involved in regulating ROS homeostasis during light or temperature stress (Xing et al. 2021; Imtiaz et al. 2023). It indicates that there may be certain connections between norisoprenoid and flavonol biosynthesis. Several reports have recently suggested that isoprenoid and flavonoid pathways can be regulated by a common TF (Jian et al. 2019). In tomato fruits, SlMYB72 modulated the metabolism of chlorophylls (Chls), carotenoids, and flavonoids by directly targeting related pathway genes (Wu et al. 2020).

The WRKY family is an important TF superfamily in plants, widely involved in the regulation of secondary metabolism, growth, and development (Schluttenhofer and Yuan 2015; Wang, Chen, et al. 2023). WRKY TFs contain conserved DNA-binding domains (WRKYGQK) and zinc finger motifs and specifically bind the W-box cis-element in the promoters (Gong et al. 2015). In addition to the OfWRKY3 mentioned above, several WRKYs have been described to reprogram the MEP pathway or regulate the carotenoid biosynthesis-related genes (Wang, Zhang, et al. 2017; Yuan et al. 2022). In the grape genome, a total of 59 putative VvWRKYs have been identified (Wang, Vannozzi, et al. 2014). VvWRKY24, VvWRKY3, and VvWRKY8 are known to directly or indirectly regulate STILBENE SYNTHASE (VvSTS) promoter activation (Vannozzi et al. 2018; Jiang et al. 2019) while Chinese wild grape (Vitis quinquangularis) WRKY TF VqWRKY56 is specific for PA biosynthesis (Wang, Wang, et al. 2023). WRKYs may have great potential in modulating the production of isoprenoid and flavonoid pathways such as norisoprenoids and flavonols.

In this study, using a yeast 1-hybrid (Y1H) screening, we identified VvWRKY70 from grapes, which modulates the biosynthesis of both norisoprenoids and flavonols under light and high-temperature condition by targeting VvBCH2 and VvCHS3. A possible regulatory network between the isoprenoid and flavonoid metabolism pathways is further explored.

Results

Identification and sequence analysis of candidate WRKYs

To explore the possible role of WRKY TFs in the regulation of norisoprenoid metabolism in grape berries, we performed a Y1H screening by focusing on the W-box. The tandem repeats of W-box (TTGAC) were constructed as a bait to screen a ‘Cabernet Sauvignon’ grape berry cDNA library. After 1-on-1 Y1H verification, 7 WRKYs were screened out and identified (Fig. 1A; Supplemental Table S1), 2 of which (VvWRKY40-like) have been verified regulating glycosylated monoterpene and anthocyanin accumulation respectively (Li et al. 2020, 2021). Further, dual-luciferase (dual-LUC) assays showed the effects of the candidate WRKYs on the promoter activities of the genes related to norisoprenoid biosynthesis. Among these candidate TFs, there was 1 WRKY that significantly inhibited the promoter activities of VvDXS1, VvPSY1, VvBCH2, and VvCCD1 (Fig. 1B). This TF was targeted for further study and was predicted to be VvWRKY70 in the National Center for Biotechnology Information (NCBI) Reference Sequence (RefSeq) database. The nucleic acid sequence of VvWRKY70 was 942 bp and encoded 313 amino acids. According to the amino acid sequences of VvWRKY70 and several reported WRKY TFs of grape, tomato, soybean (Glycine max), orange (Citrus sinensis), cotton (Gossypium hirsutum), Arabidopsis, and rice (Oryza sativa), a phylogenetic tree (Fig. 2A) was built. This protein was predicted to be 35.3 kD with 1 WRKY domain and 1 zinc finger motif and had a high degree of sequence identity with GmWRKY76, SlWRKY80, and AtWRKY70, which belonged to a group III subfamily of WRKYs according to the multiple alignment (Eulgem et al. 2000) (Fig. 2, A and B).

Figure 1.

Candidate WRKY TFs for regulating norisoprenoid biosynthesis. A) Y1H assay showed that 7 WRKYs screened from grape berries bound to the tandem W-box cis-elements. AD, pGADT7 used as a negative control; AD-VvWRKY70, prey vector containing VvWRKY70; SD/-Leu, SD medium without Leu; SD/-Leu/AbA, SD medium without Leu supplemented with AbA (ng/mL). B) Dual-LUC assay showed the effects of the WRKYs on the promoter activities. Error bars indicate ± SD from 6 replicates. Asterisks indicate significant differences relative to pGADT7 by multiple t-tests (*P < 0.05; **P < 0.01).

Figure 2.

Sequence analysis, expression profile, and subcellular localization of VvWRKY70. A) Phylogenetic tree of VvWRKY70. The triangle indicates VvWRKY70. Accession numbers of the proteins are listed in Supplemental Table S4. The scale bar represents 0.1 substitutions per site. B) Conserved domain analysis of VvWRKY70. The rectangles mark conserved domains. C) Subcellular localization of VvWRKY70 in onion epidermal cells. DAPI was used as a nuclear marker. The scale bars represented 100 µm. D)VvWRKY70 expression profile in different tissues and E-L stages of grape berries. E) Relative expression level of VvWRKY70 in the leaves of tissue culture plantlets exposed to light, compared with the expression level at 0 h. F) Relative expression level of VvWRKY70 in the leaves of tissue culture plantlets under high temperature without light. The expression levels at 25 °C without light at the same time points were used as controls. Error bars represent ± SEM from 3 replicates. Asterisks indicate significant differences by multiple t-tests (*P < 0.05; **P < 0.01).

VvWRKY70 subcellular localization and spatiotemporal expression profiling

To study the intracellular localization of VvWRKY70, a VvWRKY70-GFP fusion protein was transiently transformed into epidermal cells of onion (Allium cepa). DAPI was used as a nuclear marker. Fluorescence of the VvWRKY70 protein was found in the nucleus (Fig. 2C), confirming the role of VvWRKY70 as a TF. The expression profile showed that the transcript of VvWRKY70 can be detected in all tissues by reverse transcription quantitative PCR (RT-qPCR), and it mainly expressed in mature leaves and young berries (E-L 29 stage), whereas the expression remained at relatively low levels during berry ripening (Fig. 2D).

VvWRKY70 expression is inhibited by light and high temperature

The high expression of VvWRKY70 in mature leaves with photosynthetic capacity suggests that it is involved in the response to light and heat. Eight-week-old tissue culture plantlets were used to study the expression pattern of VvWRKY70 in response to light or high-temperature treatment. In the leaves of plantlets pretreated for 24 h in darkness, VvWRKY70 expression was significantly decreased after exposure for 3, 9, and 12 h and recovered after 24-h treatment (Fig. 2E). The expression pattern in response to high temperature (40 °C) was similar to that in response to light. The expression of VvWRKY70 was obviously reduced after treatment for 3, 6, and 12 h, and the reduction was no longer significant after treatment for 24 h (Fig. 2F). Given that the light treatment was maintained at 25 °C while the high-temperature treatment was carried out in the dark, it is likely that the VvWRKY70 expression is inhibited by short-term light and high temperature and that the responses to the 2 signals are independent.

VvWRKY70 overexpression inhibits light-induced norisoprenoid accumulation

To investigate the function of VvWRKY70 in norisoprenoid biosynthesis, the norisoprenoid content in mature V. quinquangularis leaves with transient overexpression of VvWRKY70 was determined using gas chromatography-MS (GC-MS). Five norisoprenoid components were identified, including MHO, geranylacetone, β-cyclocitral, β-ionone, and (E)-β-damascenone. Among them, β-ionone and (E)-β-damascenone were significantly decreased in the VvWRKY70-overexpressing leaves (Fig. 3, A and B). The negative regulation of VvWRKY70 on norisoprenoid production was further confirmed in stable overexpressed grape calli. Two positive transformants (OE-9 and OE-11) with high levels of the VvWRKY70 transcript were selected from several independent lines (Fig. 3, C and D). There was no obvious phenotypic difference between the overexpressing lines and the wild type (WT), either in the dark or after light treatment. However, both showed yellowing and partial reddening after 7.5 d of light treatment (Fig. 3C). Four norisoprenoid components (MHO, geranylacetone, β-cyclocitral, and β-ionone) were detected in grape calli. In the dark, the norisoprenoid content in the transgenic lines showed no difference or a slight increase compared to that in the WT. However, after exposure to light, the levels of 4 norisoprenoids were increased in both WT and overexpressing calli. Notably, the difference in norisoprenoid content between the 2 independent transgenic lines and the WT became more significant as light treatment time increased (Fig. 3E), indicating that overexpression of VvWRKY70 inhibited light-induced norisoprenoid accumulation.

Figure 3.

The effects of VvWRKY70 on norisoprenoid and carotenoid accumulation. A) Norisoprenoid contents and B)VvWRKY70 expression levels in VvWRKY70 transient overexpression (W70-GFP) leaves of V. quinquangularis. Empty vector (GFP) transformed leaves were used as controls. C) Phenotypes of transgenic grape calli in the dark and light conditions. OE, VvWRKY70-overexpressing calli. D)VvWRKY70 expression levels in WT and transgenic grape calli. E) Norisoprenoid contents in WT and transgenic grape calli in dark and exposed to light. F) Carotenoid contents in WT and transgenic grape calli at 7.5 d of light treatment. G) Carotenoid extraction from WT and transgenic grape calli at 7.5 d of light treatment. Error bars in A), E), and F) indicate ± SD from 3 replicates. Error bars in B) and D) represent ± SEM from 3 replicates. Asterisks indicate significant differences by multiple t-tests (*P < 0.05; **P < 0.01).

VvWRKY70 overexpression inhibits carotenoid accumulation

Carotenoids are the precursors of norisoprenoid biosynthesis. We also analyzed carotenoid content in transgenic calli after 7.5 d of light treatment using ultrahigh-performance liquid chromatography-triple quadrupole-MS (UHPLC-QqQ-MS). Seven carotenoid components were identified in the grape calli. Compared to the WT, all carotenoids were significantly reduced in the transgenic lines (Fig. 3, F and G). The results indicate that VvWRKY70 suppresses norisoprenoid production, which is caused mainly by reducing carotenoids in grape calli.

VvWRKY70 overexpression alters transcriptome profiling of grape calli

To dissect the mechanism by which VvWRKY70 regulates norisoprenoid biosynthetic metabolism, we performed an RNA-sequencing (RNA-seq) analysis on WT and 3 overexpressing lines (OE-1, OE-9, and OE-11, marked as W70) cultured without light or in 16-h/8-h light/dark for 7 d, respectively. Transcript abundance was evaluated by fragments per kilobase of transcript per million fragments mapped (FPKM). In the W70 versus WT transcriptomes, 2,334 differentially expressed genes (DEGs; |log2FoldChange| ≥ 1.0 and false discovery rate < 0.05) were found in darkness (Supplemental Data Set 1) and 1,360 DEGs were found in light (Supplemental Data Set 2). Among these, 658 DEGs were present in both data sets (Supplemental Fig. S1). The DEGs of the calli in darkness were found to be mainly involved in metabolism and plant hormone signal transduction, whereas those of the calli in light were mainly enriched in photosynthesis, phytohormone signaling, and flavonoid biosynthesis pathways based on Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) enrichment (Supplemental Fig. S2 and Table S2). Gene transcript levels were verified by RT-qPCR, and the results were highly consistent with the RNA-seq analysis (R² = 0.9467; Supplemental Fig. S3). The expression levels of VvDXS1, VvPSY1, and VvBCH2 were downregulated by VvWRKY70 in the light condition, which was consistent with the results of dual-LUC described above (Fig. 1B). However, the transcript level of VvCCD1 showed no difference between transgenic calli and WT. Furthermore, VvWRKY70 was also found to downregulate some genes involved in flavonoid metabolism (Fig. 4, A and B).

Figure 4.

Transcriptome analysis of VvWRKY70 overexpression in grape calli. A) DEGs of isoprenoid metabolism pathway. B) DEGs of flavonoid metabolism pathway. L-W70 and L-WT represent transgenic calli and WT in light; D-W70 and D-WT represent transgenic grape calli and WT in darkness. The boxes represent the intensity of log2FoldChange. DEGs (|log2FoldChange| ≥ 1.0 and false discovery rate < 0.05) are indicated by bold margins. 4CL, 4-COUMARATE:COENZYME A LIGASE; ANR, ANTHOCYANIDIN REDUCTASE; ANS, ANTHOCYANIN SYNTHASE; BCH, β-CAROTENE HYDROXYLASE; C4H, CINNAMATE 4-HYDROXYLASE; CCD, CAROTENOID CLEAVAGE DIOXYGENASE; CHI, CHALCONE ISOMERASE; CHS, CHALCONE SYNTHASE; CRTISO, PROLYCOPENE ISOMERASE; DFR, DIHYDROFLAVONOL REDUCTASE; DXS, 1-DEOXY-D-XYLULOSE 5-PHOSPHATE SYNTHASE; F3H/F3′H/F3′5′H, FLAVANONE-3/3′/3′5′-HYDROXYLASE; FLS, FLAVONOL SYNTHASE; GA3P, glyceraldehyde triphosphate; GGPP, geranylgeranyl diphosphate; GGPPS, GERANYLGERANYL DIPHOSPHATE SYNTHASE; LAR, LEUCOANTHOCYANIDIN REDUCTASE; LCYb, LYCOPENE β-CYCLASE; LCYe, LYCOPENE ε-CYCLASE; LUT, β-RING HYDROXYLASE; NCED, 9-cis-EPOXYCAROTENOID DIOXYGENASE; NSY, NEOXANTHIN SYNTHASE; PA, proanthocyanidin; PAL, PHENYLALANINE AMMONIA-LYASE; PDS, PHYTOENE DESATURASE; PSY, PHYTOENE SYNTHASE; UFGT, UDP-GLUCOSE: FLAVONOID-3-O-GLUCOSYLTRANSFERASE; VDE, VIOLAXANTHIN DE-EPOXIDASE; ZDS, ζ-CAROTENE DESATURASE; ZEP, ZEAXANTHIN EPOXIDASE; ZISO, ζ-CAROTENE ISOMERASE. Solid or dashed arrows indicate direct or multiple-step reaction in the pathway, respectively.

VvWRKY70 targets repression of VvBCH2 expression

Based on the results of the Y1H screening, dual-LUC, and RNA-seq described above (Fig. 1, A and B), the promoter fragments of 3 candidate target genes (VvDXS1, VvPSY1, and VvBCH2) containing at least 1 W-box element were chosen for further Y1H assays. The promoter fragments were integrated into the genome of the Y1HGold yeast strain. The result indicated that VvWRKY70 only directly binds to the promoters of VvPSY1 and VvBCH2 in the yeast system (Fig. 5A). Electrophoretic mobility shift assay (EMSA) was also performed. Recombinant MBP-VvWRKY70 fusion protein was obtained using a prokaryotic expression system and then purified. The DNA fragments containing the W-box of promoters were synthesized as probes (Fig. 5B). Recombinant MBP-VvWRKY70 shifted the band with a labeled DNA probe from VvBCH2 promoters, indicating the formation of a DNA-protein complex. When the excess unlabeled probe was added as a competitor, the shifted band became unobservable. When the competitor was changed to excess mutant unlabeled probe, a band corresponding to the labeled DNA probe was still observed. There was no DNA-protein complex formation between the VvWRKY70 protein and the mutant labeled probe (Fig. 5C). Chromatin immunoprecipitation (ChIP)-qPCR was performed and the VvBCH2 promoter showed significant enrichment (Fig. 5D). Combined with the dual-LUC results described above (Fig. 1B), VvWRKY70 protein could inhibit the promoter activity of VvBCH2 and specifically bind to the W-box on the promoter both in vitro and in vivo.

Figure 5.

Interaction between VvWRKY70 and the promoters of target gene. A) Y1H assay showed the binding of VvWRKY70 to promoter fragments. AD, pGADT7 used as a negative control; AD-VvWRKY70, prey vector containing VvWRKY70; SD/-Leu, SD medium without Leu; SD/-Leu/AbA, SD medium without Leu supplemented with AbA (ng/mL). B) The schematic diagram of the VvBCH2 promoter. Rectangle indicates the fragment used for Y1H. The probe with W-box (TTGAC) was used for the EMSA. C) The EMSA showed the binding of VvWRKY70 to the W-box in the promoters. D) ChIP-qPCR showed the binding of VvWRKY70 to the VvBCH2 promoter. Error bars indicate ± SD from 3 replicates. The asterisk indicates significant difference relative to lgG by t-tests (*P < 0.05).

Overexpression of VvWRKY70 inhibits high-temperature–induced norisoprenoid accumulation

A previous study suggested that the norisoprenoid content in grapes may be affected by temperature (Tian et al. 2023). This result was confirmed in this study using leaves of the heat-tolerant V. quinquangularis under high-temperature stress. After prolonged treatment at 37 °C for 5 d, the levels of MHO, geranylacetone, β-cyclocotral, and (E)-β-damascenone were significantly higher than that of the control (25 °C) (Fig. 6A). In another experiment using tissue culture plantlets, it was observed that the expression of VvBCH2, the target gene of VvWRKY70, was upregulated by high temperature (Fig. 6B), suggesting that VvWRKY70 may be involved in the regulation of norisoprenoid biosynthesis in response to high temperature. Based on the suppression of VvWRKY70 expression under light and high temperature (Fig. 2F), we cultured 2 VvWRKY70-overexpressing lines (OE-2 and OE-3) and WT in the dark and subjected them to the 35 °C/25 °C period for 5 d. As expected, the expression of VvBCH2 was repressed by VvWRKY70 overexpression after high-temperature treatment (Fig. 6C). Compared to WT calli cultured at 25 °C, high-temperature treatment increased the content of 4 norisoprenoids (Fig. 6D), which is consistent with the results in V. quinquangularis leaves. However, after high-temperature treatment, 4 norisoprenoids in the transgenic lines (designed as 35 °C OE) were all significantly reduced compared to the 35 °C WT (Fig. 6D), which was similar to the results of light treatment (Fig. 3E). Taken together, VvWRKY70 overexpression also inhibits high-temperature–induced norisoprenoid accumulation.

Figure 6.

The effects of VvWRKY70 on high-temperature–mediated norisoprenoid biosynthesis. A) Norisoprenoid accumulation was induced by high temperature in leaves of V. quinquangularis. Error bars indicate ± SD from 3 replicates. B) Relative expression level of VvBCH2 was upregulated by high temperature in the leaves of tissue culture plantlets. C) Relative expression level of VvWRKY70 and VvBCH2 in WT and transgenic grape calli at 5 d of high-temperature treatment (35 °C). Error bars in B) and C) indicate ± SEM from 3 replicates. Asterisks in A), B), and C) indicate significant differences by multiple t-tests (*P < 0.05; **P < 0.01). D) Norisoprenoid contents in WT and transgenic grape calli at 5 d of high-temperature treatment. Error bars indicate ± SD from 3 replicates. Different letters above the columns indicate significant differences (P < 0.05, ANOVA, Tukey test).

VvWRKY70 negatively regulates flavonol production in grape calli

Figure 4 illustrated that the expression of some genes in flavonoid metabolism pathway was changed in the VvWRKY70-overexpressing calli lines. This pathway can generate a series of important components determining grape and wine quality, such as flavonols, PAs, and anthocyanins. Combined with the observed phenotype, we detected the content of flavonols in transgenic lines and WT in light using UHPLC-QqQ-MS. Five flavonol components were identified in the grape calli. Compared with that in the WT line, the quercetin-glucuronide, quercetin-glucoside, isohamnetin-glucoside, and kaempferol-galactoside contents in transgenic lines decreased significantly, except for no change in the myricetin-glucoside content (Fig. 7A). The promoters of the DEGs involved in flavonol biosynthesis were cloned and dual-LUC was performed to verify whether these genes were the targets of VvWRKY70. The results showed that VvWRKY70 inhibited the promoter activities of VvCHS2, VvCHS3, and VvFLS4 (Fig. 7B). Furthermore, Y1H assay and ChIP-qPCR confirmed that VvCHS3 can be directly targeted by VvWRKY70 in vitro and in vivo (Fig. 7, C and D).

Figure 7.

The effects of VvWRKY70 on light-mediated and high-temperature–mediated flavonol biosynthesis. A) Flavonol contents in WT and transgenic grape calli at 7.5 d of light treatment. Error bars indicate ± SD from 3 replicates. B) Dual-LUC assay showed that VvWRKY70 inhibited the promoter activities of genes involved in flavonol biosynthesis. Error bars in B) indicate ± SD from 6 replicates. C) Y1H and D) ChIP-qPCR showed the binding of VvWRKY70 to VvCHS3 promoter. Error bars indicate ± SD from 3 replicates. E) Flavonol contents in WT and transgenic grape calli at 5 d of high-temperature treatment. Error bars indicate ± SD from 3 replicates. Different letters above the columns indicate significant differences (P < 0.05, ANOVA, Tukey test). F) Relative expression level of VvCHS3 in WT and transgenic grape calli at 5 d of high-temperature treatment. G) Relative expression level of VvCHS3 was upregulated by high temperature in the leaves of tissue culture plantlets. Error bars in F) and G) indicate ± SEM from 3 replicates. Asterisks in A), B), D), F), and G) indicate significant differences relative to empty vector control by multiple t-tests (*P < 0.05; **P < 0.01).

After high-temperature treatment, the contents of quercetin-glucoside and myricetin-glucoside in WT calli were increased compared with the control (25 °C WT). Further, in the high-temperature treatment calli, VvWRKY70 overexpression significantly decreased the contents of quercetin-glucuronide, quercetin-glucoside, myricetin-glucoside, and isohamnetin-glucoside and inhibited the expression of VvCHS3 (Fig. 7, E and F), which was similar with the results in light treatment assay. RNA-seq showed that VvCHS3 can be induced by light (Fig. 4B) and the study on tissue culture plantlets suggested that VvCHS3 can also be induced by high temperature (Fig. 7G). It is thus suggested that VvWRKY70 is also involved in the regulation of flavonol biosynthesis in response to light and high temperature by targeting the expression of VvCHS3.

Discussion

VvWRKY70 is involved in light-induced and high-temperature–induced norisoprenoid accumulation

WRKY TFs play important roles in biotic and abiotic stress response, but studies on their role in norisoprenoid metabolism in grapes are still very limited. In grapes, some of the WRKYs have been confirmed to be involved in the response to light or temperature signals. VvWRKY8 contributes to the regulatory loop of UV-C-mediated resveratrol biosynthesis (Jiang et al. 2019). VaWRKY12 was identified from Vitis amurensis with excellent cold tolerance, and this TF is induced by low temperature and in turn enhances the cold tolerance of grapevine (Zhang et al. 2019). The genome-wide analysis of WRKY response to cold stress in V. vinifera has shown that the expression of VvWRKY45, which is homologous to VvWRKY70 in this study, can be upregulated after 48 h of cold treatment, but remains unchanged at 8 and 24 h (Wang, Zhu, et al. 2014). In this study, we found that the VvWRKY70 expression was downregulated by light and high temperature. The expression level was significantly decreased after 3- to 12-h treatment and recovered after 24-h treatment (Fig. 2, E and F). This suggested that the response of VvWRKY70 to light and high temperature was short-lived and that there may be a feedback loop regulation.

The biosynthetic pathway of norisoprenoids has been described, but the regulatory mechanism is not fully understood. In field experiments, light induced the production of carotenoids and norisoprenoids in grape berries (Young et al. 2016), but opposite results have been reported (He et al. 2020). In this study, the norisoprenoid content in WT calli continued to accumulate with the 16-h/8-h photoperiod treatment, compared to the calli grown in darkness (Fig. 3E). Under field conditions, light and heat variations appear to be synchronized, and cultural practices often affect both light and heat factors in the microclimate around the cluster. However, sometimes these effects do not operate simultaneously throughout the day and night. For example, covering with gravel (Tian et al. 2023) or inter-row mulching (Wang, Gao, et al. 2021) affects not only the light reflectance during the day but also the temperature at night. Our study on mature leaves and WT calli shows that the norisoprenoid accumulation can be induced by short-lived high temperature. The evidence in this study suggests that VvWRKY70 is involved in the norisoprenoid accumulation by responding to light and high temperature independently (Figs. 3 and 6). This complex response mechanism may explain the phenomenon of inconsistent changes in norisoprenoid content in different cultivation experiments of grape berries.

VvWRKY70 inhibits carotenoid and norisoprenoid biosynthesis by directly targeting VvBCH2

The present study provides evidence that VvWRKY70 inhibits the norisoprenoid accumulation induced by light or high temperature, and VvWRKY70 targets the transcriptional repression of VvBCH2 by binding to the W-box (TTGAC) on the promoter according to Y1H, EMSA, and ChIP-qPCR assays (Fig. 5). The expression of VvBCH2 was upregulated by light and high temperature (Figs. 4A and 6B), suggesting that the norisoprenoid content can be influenced by carotenoid metabolism. Focusing on the transgenic calli after light or high-temperature treatment, similar responses were observed in that the 4 norisoprenoid compounds derived from different carotenoids were all reduced. With the exception of β-carotene, which is the substrate of BCH, the carotenoids detected in the calli are all mainly downstream products of BCH2. Therefore, it is very likely that the downregulation of VvBCH2 expression directly leads to a decrease in the precursors, which ultimately results in a decrease in norisoprenoid accumulation.

Although only a limited number of WRKY TFs have been shown to be involved in norisoprenoid metabolism to date (Han et al. 2016), the upstream involvement of several WRKYs has been described. SlWRKY35 has been defined as a positive regulator of carotenoid biosynthesis in tomato fruit and it is known to target SlDXS1 and reprogram the MEP pathway (Yuan et al. 2022). In addition, 8 WRKYs have been found to respond to ethylene and activate the promoters of SlPSY and SlPDS (Wang, Zhang, et al. 2017). According to the preliminary results obtained in this study, VvWRKY70 also downregulates the transcript levels of VvDXS1 and VvPSY1 in isoprenoid metabolism (Figs. 1B and 4A). DXS and PSY control the upstream flux of the isoprenoid pathway, and inhibition of their expression affects the size of the carotenoid pool, which is the precursor of norisoprenoids in grape berries. VvWRKY70 did not bind to the VvDXS1 and VvPSY1 promoters but inhibited the promoter activity (Fig. 1B). Compared to WT calli, VvWRKY70-overexpressing calli at 25 °C in the dark showed no difference or even slightly higher than that of WT (Fig. 3E). Given that there was no transcriptional activation domain in VvWRKY70 (Supplemental Fig. S4), this slight upregulation may not be caused by the direct regulation of VvWRKY70 on the biosynthetic metabolism. Since β-cyclocitral has been reported to affect carotenoid content in Arabidopsis and citrus (Zheng et al. 2020), there may be some other effectors or a loop regulation mechanism. Moreover, other potential mechanisms such as protein-protein interaction or protein modification require further study (Sun et al. 2022).

VvWRKY70 inhibits flavonol biosynthesis in response to light and high temperature by targeting VvCHS3

CHS is known as the gatekeeper of flavonoid biosynthesis, and both CHS and FLS are involved in flavonol accumulation in fruits (Liu et al. 2016; Cao et al. 2019). The expression of VvCHS and VvFLS4 was light inducible according to the transcriptome data, while flavonols were reported to be induced by light in grape berries (Liu et al. 2015). In our study, VvWRKY70 downregulated the expression of VvCHS2, VvCHS3, and VvFLS4 (Fig. 7B), which associated with the reduction of flavonols in transgenic grape calli under light treatment (Fig. 7A). It was further confirmed that VvWRKY70 can downregulate flavonol biosynthesis by directly targeting VvCHS3 (Fig. 7, C and D).

Similarly, 2 flavonol components, quercetin-glucoside and myricetin-glucoside, were induced by high temperature in WT calli and most of the flavonol components were decreased by VvWRKY70 overexpressing compared to the WT (Fig. 7, A and E). At the same time, the expression of VvCHS3 was upregulated by high temperature and was strongly inhibited by VvWRKY70 overexpression under high temperature. A previous study showed that the accumulation of flavonoids is decreased under high temperature, but mainly anthocyanins (Tian et al. 2023). It should be noted that the types of flavonols inhibited by VvWRKY70 under high-temperature conditions are not completely consistent with those under light conditions, indicating that there is a refined and complex regulatory network for the regulation of different flavonol compounds in response to light and high temperature. AtWRKY23 is known to maintain root development in Arabidopsis by stimulating local flavonol biosynthesis (Grunewald et al. 2012), and NtWRKY11b induces the transcription of NtFLS and promotes flavonol biosynthesis in tobacco (Nicotiana tabacum) leaves (Wang, Luo, et al. 2021). In this study, we show that a WRKY TF also regulates flavonol synthesis in fruits.

VvWRKY70 is involved in the hormone-mediated regulatory network between isoprenoid and flavonoid metabolism

In addition to the findings mentioned above, there are some other transcriptome data worth noting. Chl biosynthesis is a branch of the isoprenoid metabolic pathway, and carotenoid accumulation occurs concomitantly with Chl degradation during fruit ripening. According to the RNA-seq data in this study, Chl metabolism-related genes such as STAY GREEN 1 (VvSGR1), 7-HYDROXYMETHY CHLOROPHYLL A REDUCTASE (VvHCAR), CHLOROPHYLL A OXYGENASE (VvCAO), PROTOCHLOROPHYLLIDE REDUCTASE (VvPOR), and Mg-CHELATASE H SUBUNIT (VvCHLH) were substantially downregulated by VvWRKY70 (Supplemental Table S2). Moreover, VvWRKY70 directly inhibited the promoter activities of VvPOR and VvCHLH (Supplemental Fig. S5). Although it was difficult to investigate the change in Chl content in grape calli due to its extremely low content, it would be interesting to further explore the role of VvWRKY70 in fruit Chl metabolism.

In this study, the upstream flux of ABA biosynthesis (carotenoid biosynthesis) was inhibited by VvWRKY70, and the key ABA catabolic gene ABA 8′hydroxylase 2 (VvHYD2) was upregulated. The putative ABA receptor genes, VvCHLH and PYRABACTINRESISTANCE 1-LIKE 8-LIKE (VvPYL8-like), were downregulated by VvWRKY70 (Supplemental Table S2). These changes suggest that VvWRKY70 inhibits ABA signaling, which may affect isoprenoid (Meng et al. 2020) and flavonoid (Zhang et al. 2020) metabolism. ABA and auxin are important for grape ripening (Fortes et al. 2015). In transgenic grape calli, several auxin signal reduction genes, such as AUXIN RESPONSE FACTOR 8 (VvARF8), SMALL AUXIN UPREGULATED GENES (VvSAURs), and INDOLE-3-ACETIC ACID-AMIDO SYNTHETASE (VvGH3.1), were identified as DEGs (Supplemental Table S2). Flavonols act as endogenous regulators of auxin transport, and WRKY23 is part of the transcriptional feedback loop of auxin transport through regulation of flavonol biosynthesis (Grunewald et al. 2012). Taken together, the results presented here suggest that VvWRKY70 may play a role in the hormone-mediated regulatory network of biosynthetic pathway genes. However, due to the limitations of grape calli, some hypotheses need to be verified.

Conclusion

Based on the results of this study, a model was proposed to illustrate the contribution of VvWRKY70 to light-induced and high-temperature–induced norisoprenoid and flavonol biosynthesis in grape berry: Under dark and room temperature conditions, VvWRKY70 inhibits the expression of VvBCH2 in isoprenoid metabolism and VvCHS3 in flavonoid metabolism and keeps norisoprenoids and flavonols at low levels; when transferred to light or high-temperature environment, VvWRKY70 expression is downregulated, and its transcriptional inhibition on VvBCH2 and VvCHS3 is relieved, resulting in the accumulation of norisoprenoids and flavonols (Fig. 8). These findings enrich our understanding of the regulatory network of light-mediated and high-temperature–mediated norisoprenoid and flavonol metabolism in grapes and provide a perspective for the coordinated regulation of multiple qualities of grape berries.

Figure 8.

Proposed model for illustrating the contribution of VvWRKY70 in light-mediated and high-temperature-mediated transcriptional regulation of norisoprenoid and flavonol metabolism in grape berries. Light and high temperature can respectively downregulate the expression of VvWRKY70 and relieve the transcriptional inhibition of VvWRKY70 on VvBCH2 in the norisoprenoid biosynthesis pathway and VvCHS3 in flavonol biosynthesis pathway, thereby promoting the accumulation of norisoprenoids and flavonols.

Materials and methods

Plant material and treatment conditions

Roots, shoots, tendrils, young leaves (10-d-old), mature leaves (30-d-old), flowers, and berries at 9 developmental stages following the E-L system (Coombe 1995) were collected from own-rooted grapevines (V. vinifera L. cv Cabernet Sauvignon) in 2019, China Agricultural University Shangzhuang Experimental Station (Beijing, China, 40°14′N, 116°20′E). All samples were divided into 3 biological replicates collected and from 9 grapevines and stored at −80 °C after quick-freezing. Eight-week-old tissue culture grape plantlets of ‘Cabernet Sauvignon’ and Nicotiana benthamiana plants were cultured in a growth chamber at 25 °C in 16-h/8-h light/dark conditions. The grape calli were induced from ‘Cabernet Sauvignon’ grape berries and maintained on B5 medium and subcultured every 25 d in darkness at 25 °C (Meng et al. 2020).

For the light treatment, the calli were subcultured and maintained on B5 medium for 18 d in darkness. They were then cultured in the climate chamber with a 16-h/8-h light/dark photoperiod for 7.5 d while the control was cultured in darkness. The samples were collected on the 0, 2.5, 4.5, and 7.5 d after treatment. Eight-week-old tissue culture grape plantlets of ‘Cabernet Sauvignon’ pretreated in darkness for 24 h were exposed to light and the leaves were collected after 0, 3, 6, 9, 12, and 24 h. The light intensity was 250 µmol photons m⁻² s⁻¹.

For high-temperature treatment, the 18-d-old calli were subjected to a 16-h/8-h 35 °C/25 °C period while the control was cultured at 25 °C. Both groups were kept without light for 5 d. Six Chinese wild grape (V. quinquangularis) mature leaves were incubated in vitro at 37 °C for 5 d without light, covered by wet gauze and divided into 3 replicates for further analysis. The control was cultured at 25 °C at the same dark condition. Eight-week-old tissue culture grape plantlets of ‘Cabernet Sauvignon’ were treated with 40 °C in dark and the leaves were collected after 0, 3, 6, 12, and 24 h. The plantlets at 25 °C in dark at the same points of time were collected as controls. Each group contained at least 3 plantlets.

Gene cloning and sequence analysis

The WRKYs CDS were cloned from the ‘Cabernet Sauvignon’ library. The primers used were shown in Supplemental Table S3. The structure of VvWRKY70 was analyzed using the online database (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). The associated WRKY amino acid sequences were obtained from NCBI under accession numbers shown in Supplemental Table S4. The phylogenetic tree was conducted by the neighbor-joining method from MEGA 6. DNAMAN 6.0 was used for multiple sequence alignments.

Y1H analysis

Clontech Matchmaker Gold Y1H System was used for yeast assay. Y1HGold yeast strain was transformed with pAbAi vectors, which contained the baits. The baits included the synthetic tandem repeats of “TTGAC” motif and fragments cloned from target promoters of ‘Cabernet Sauvignon’. The ‘Cabernet Sauvignon’ library constructed previously was used for Y1H screening (Mu et al. 2014). The WRKY coding sequences (CDS) were transformed into pGADT7 as preys. The primers used were shown in Supplemental Table S3. The detailed procedures have been previously described (Meng et al. 2020).

Subcellular localization

VvWRKY70-GFP fusion protein was integrated into the pEZS-NL plasmid. The primers used were shown in Supplemental Table S3. A particle gun was used to transient transformed the recombinant into epidermal cells of onion (A. cepa) (Meng et al. 2020). The cells were observed by an Olympus FV1000 laser scanning confocal microscope (Tokyo, Japan). GFP fluorescence signal was detected with an excitation wavelength at 488 nm, and DAPI signal was detected using an excitation wavelength of 405 nm.

Transformation of grape calli and leaves

The CDS of VvWRKY70 was cloned into the pCXSN vector by ClonExpress II One Step Cloning Kit (C112, Vazyme, China), and the recombinant plasmid was transformed into Agrobacterium tumefaciens strain GV3101. The transformation of grape calli and positive clone selection were performed as previously described (Meng et al. 2020).

Transient overexpression was performed using Chinese wild grape (V. quinquangularis) mature leaves. The VvWRKY70-GFP fusion protein was transiently overexpressed by pCAMBIA 1300-GFP. The empty vector was used as the control. The transformation method was described in a previous study (Meng et al. 2020). The primers used were shown in Supplemental Table S3.

ChIP-qPCR

For ChIP-qPCR, the vector of pCAMBIA 1300-VvWRKY70-GFP was used for calli transformation. The primers used were shown in Supplemental Table S3. The ChIP assay was conducted by Igenebook (Wuhan, China) following the method previously described (Liu et al. 2023). The enrichment levels were normalized to the input sample.

Dual-LUC

For promoter cloning, the ‘Cabernet Sauvignon’ genomic DNA was extracted by Plant Genomic DNA Rapid Extraction Kit (BioTeke, China). The candidate gene promoters were cloned by 2 × Phanta Flash Master Mix (P520, Vazyme, China) and were inserted into the pGreenII 0800 double-reporter vector. The promoter sequence data were shown in Supplemental Table S5. The CDS of VvWRKYs were cloned into the effector vector pCAMBIA 1301, replacing the gusA gene. An empty vector was set as a negative control. The primers used were shown in Supplemental Table S3. The dual-LUC was performed by N. benthamiana transient expression system (Meng et al. 2020).

EMSA

The MBP-VvWRKY70 fusion protein was expressed using the pMAL-C5X vector in Escherichia coli strain Transetta (DE3). The primers used were shown in Supplemental Table S3. Promoter fragments containing the W-box were synthesized and labeled with biotin. Unlabeled probes were used as competitors. The samples were loaded onto a 6% (w/v) native polyacrylamide gel. EMSA was conducted by following the instructions of the Chemiluminescent EMSA Kit (Beyotime, Shanghai, China). The electrophoresis was performed at 100 V in an ice bath. Gel images were captured using e-BLOT Touch Imager (e-BLOT Life Science, Shanghai, China).

RNA extraction, cDNA synthesis, and RT-qPCR analysis

The frozen samples were ground by porcelain mortar before extraction. RNA extraction was performed following the instructions of plant total RNA extraction kit (RP3302, BioTeke, China). Total RNA (1 µg) was used for the reverse transcription kit of Vazyme (R333, China). RT-qPCR was performed by SYBR master mix (Q711, Vazyme, China) on the CFX96 (Bio-Rad, USA). The UBIQUITIN was used as the reference. The primers used were shown in Supplemental Table S3. Each reaction was conducted in triplicate, and the results were analyzed by the CFX Maestro Software (Bio-Rad, USA).

Transcriptome analysis by RNA-seq

Four groups were sequenced (WT calli in darkness, D_WT; WT calli in light, L_WT; VvWRKY70-overexpressing calli in darkness, D_W70; and VvWRKY70-overexpressing calli in light, L_W70). Three independent lines (OE-1, OE-9, and OE-11) of VvWRKY70-overexpressing calli were used as replicates. RNA library construction was conducted by Novogene Bioinformatic Technology (Tianjing, China) and an Illumina HiSeq was used for sequencing. Total reads were mapped to the V. vinifera PN40024 genome 12X.2 and annotated based on the V2.1 version. FPKM was used to calculate the gene transcription levels. DEGs (|log2FoldChange| ≥ 1.0 and false discovery rate < 0.05) were defined by DESeq2 R package (1.20.0). DEGs are listed in Supplemental Data Sets S1 and S2. KEGG and GO enrichment analyses were performed by the clusterProfiler R package.

Carotenoid measurements

Carotenoids were measured in 6 replicates. The extraction method was used according to a previously described method with some modifications (Kamffer et al. 2010). The samples were ground into powder in liquid nitrogen and freeze-dried. The lyophilized powder (100 mg) was mixed with 1.0-mL Millipore water and 10-µL 8′-apo-β-carotenal (20 µg/mL) as an internal standard (IS) before extraction in a 5-mL tube. Each sample was extracted twice with 1.0 mL of diethyl ether/hexane (1:1). The extract was sonicated in ice water for 20 min in darkness and vortexed for 1 min. Then, the supernatant was collected after being centrifuged at 12,000 × g for 2 min at 4 °C. The residue was then re-extracted. The supernatant was collected and dried with nitrogen. Dried samples were dissolved in 200 µL of methanol containing 0.1% (w/v) BHT. The samples were filtered through 0.22-µm membrane filters prior to conducting the analysis based on the Agilent 1290 series UPLC in tandem with Agilent 7670B QqQ-MS. Carotenoids were measured according to a method from Agilent (https://www.agilent.com.cn/cs/library/applications) with some modifications. An Agilent ZORBAX RRHD Eclipse Plus 95Å PAH (1.8 µm, 100 mm × 2.1 mm) column was used, and the UPLC conditions were as follows: mobile phase A was 0.1% (v/v) formic acid in water, and mobile phase B contained 0.1% (v/v) formic acid in methanol; gradient program, started at 75% B (0 to 1 min), increased to 100% B (2 to 6 min), then kept 100% B (6 to 19 min), and finally ramped back to 25% B (19 to 20 min): flow rate, 0.3 mL/min; temperature, 45 °C; and injection volume, 1 µL. An Agilent jet stream electrospray ionization (AJS-SEI) source in the positive ion mode was employed. The capillary voltage was set at 4,000 V. Standards were used as qualitative reference; MS analysis was conducted in the multiple reactor monitoring (MRM) mode for identification and quantification. The related qualitative and quantitative information was shown in Supplemental Table S6.

Norisoprenoid measurements

For calli, a frozen (100 g) mixed with 1-g PVPP and 0.5-g d-gluconic acid lactone was frozen and ground to powder. NaCl (1.5 g), citric acid/sodium citrate buffer (pH 5.0, 3 mL), sample powder (4.0 g), and [2H3]-linalool (10 µL, 5 × 10⁻³ g/L, IS) were blended in a sample vial for further analysis. For mature leaves, fresh powder ground in liquid nitrogen as described above (1.0 g), NaCl (1.0 g), citric acid/sodium citrate buffer (pH 3.0, 5 mL), and 4-methyl-2-pentanol (10 µL, 0.01 g/L, IS) were blended in a sample vial for further analysis.

The aromatic compounds were concentrated by headspace solid-phase microextraction (HS-SPME) and detected by GC-MS based on a published method (Wei et al. 2021) with some modifications. The norisoprenoid compounds were identified on the basis of retention time and mass spectra matching the standards. Notably, the norisoprenoid content of grape calli was measured using the selected ion monitoring (SIM) mode, and the full scan mode was used for mature leaf samples. The corresponding qualitative and quantitative information is presented in Supplemental Table S7.

Flavonol measurements

The lyophilized powder of grape calli (50 mg) was immersed in 1 mL of 50% (v/v) MeOH and sonicated for 20 min in ice water to avoid light. The supernatant was transferred into another tube after 12,000 × g, 5-min centrifuging at 10 °C, then the residue was re-extracted, and the pooled supernatants were collected and stored at −80 °C before analysis.

Flavonol measurements were measured according to a previously described method (Li et al. 2017) with some modifications. UPLC was performed on samples filtered through 0.22-µm membrane filters. Agilent 1290 series UPLC in tandem with Agilent 7670B QqQ-MS was used. A Poroshell 120 SB-C18 column (150 mm × 2.1 mm, 2.7 µm, Agilent) was used, and the composition of the mobile phases was as follows: mobile phase A, 0.1% (v/v) formic acid in water; mobile phase B, 0.1% (v/v) formic acid in acetonitrile. The gradient elution was the same as the short-gradient elution program described previously (Zhao et al. 2022). The injection volume was 20 µL. An AJS-ESI source in the negative ion mode was employed. The capillary voltage was set as 3,500 V. MS analysis was conducted in the MRM mode. The selected parameters were shown in Supplemental Table S8.

Statistical analysis

String diagrams and bar charts were constructed using GraphPad Prism version 9. Statistical analyses were conducted using 1-way ANOVA or multiple t-tests (*P < 0.05; **P < 0.01).

Accession numbers

Sequence data from this article can be found in the GenBank data library or the genome database of V. vinifera-PN40024 version 12x.V2 (http://www.grapegenomics.com/pages/PN40024/) under accession numbers shown in Fig. 3 and Supplemental Tables S3 and S4.

Acknowledgments

We thank Prof. Daqi Fu, Prof. Hongliang Zhu, Assoc. Prof. Xiuqin Wang, and Assoc. Prof. Guiqin Qu (China Agricultural University) for donating the pCXSN, pGreen II 0800, and pCAMBIA 1301 vectors, grape calli, and tomato seeds. We thank Assoc. Researcher Lei Sun (Institute of Forestry and Pomology, Beijing Academy of Agriculture and Forestry Sciences) for providing mature leaves of V. quinquangularis. We also thank Keji Yu, Ning Zhang, Xuechen Yao, and Mengqi Ling (China Agricultural University) for technical assistance. We would like to thank Editage (www.editage.cn) for English language editing.

Author contributions

Y.Wei performed the main experiments, interpreted the data, and wrote the manuscript. N.M. and J.C. assisted in the Y1H and dual-LUC assays and contributed to data interpretation. Y.Wang contributed to the transgenic calli experiment and the measurement of norisoprenoids. C.D. contributed to the research conception and design. Q.P. conducted a critical review of the work and its scientific and intellectual content and approved the final version of the manuscript for publication. All authors contributed to the manuscript and approved the submitted version.

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. Venn diagram analysis of DEGs in VvWRKY70-overexpressing calli compared to that in the WT in dark (D_W70vsD_WT_all) and light (L_W70vsL_WT_all) conditions.

Supplemental Figure S2. GO and KEGG pathway enrichment analyses.

Supplemental Figure S3. Correlation analysis of DEGs obtained from RNA-seq and RT-qPCR.

Supplemental Figure S4. Transcriptional activity of VvWRKY70.

Supplemental Figure S5. Dual-LUC assay of other selected gene promoters.

Supplemental Table S1. Accession numbers and descriptions of candidate TFs identified by Y1H screening.

Supplemental Table S2.VvWRKY70 and selected DEGs (|log2FoldChange| ≥ 1.0 and false discovery rate < 0.05) associated with Chl and signal transduction.

Supplemental Table S3. Primers used in this study.

Supplemental Table S4. Sequence accession numbers used in the phylogenetic analysis.

Supplemental Table S5. The promoter sequence data for dual-LUC transient expression assay.

Supplemental Table S6. Qualitative and quantitative information relating to identified carotenoids.

Supplemental Table S7. Qualitative and quantitative information relating to identified norisoprenoids.

Supplemental Table S8. Qualitative and quantitative information relating to identified flavonols.

Supplemental Data Set 1. DEGs in the VvWRKY70-overexpressing calli in darkness.

Supplemental Data Set 2. DEGs in the VvWRKY70-overexpressing calli in light.

Funding

This research was funded by the National Natural Science Foundation of China (Grant Nos. 32072513, U20A2042, and 32102314).

Data availability

Raw data of RNA-seq were uploaded in the SRA database and the accession number is PRJNA833286. Other data used in this study are available from the corresponding authors upon reasonable request.

Dive Curated Terms

The following phenotypic, genotypic, and functional terms are of significance to the work described in this paper:

• MYB111 Gramene: AT3G46130

• MYB111 Araport: AT3G46130

• MYB12 Gramene: AT2G47460

• MYB12 Araport: AT2G47460

• WRKY23 Gramene: AT2G47260

• WRKY23 Araport: AT2G47260

• VvWRKY70 Gramene: VIT_213s0067g03140

• VvWRKY70 Araport: VIT_213s0067g03140

Author notes

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 Qiuhong Pan.

Conflict of interest statement. The authors declare that they have no conflicts of interest.