The Grapevine R2R3-MYB Transcription Factor VvMYBF1 Regulates Flavonol Synthesis in Developing Grape Berries^,,

Abstract

Flavonols are important ultraviolet light protectants in many plants and contribute substantially to the quality and health-promoting effects of fruits and derived plant products. To study the regulation of flavonol synthesis in fruit, we isolated and characterized the grapevine (Vitis vinifera ‘Shiraz’) R2R3-MYB transcription factor VvMYBF1. Transient reporter assays established VvMYBF1 to be a specific activator of flavonol synthase1 (VvFLS1) and several other promoters of grapevine and Arabidopsis (Arabidopsis thaliana) genes involved in flavonol synthesis. Expression of VvMYBF1 in the Arabidopsis mutant myb12 resulted in complementation of its flavonol-deficient phenotype and confirmed the function of VvMYBF1 as a transcriptional regulator of flavonol synthesis. Transcript analysis of VvMYBF1 throughout grape berry development revealed its expression during flowering and in skins of ripening berries, which correlates with the accumulation of flavonols and expression of VvFLS1. In addition to its developmental regulation, VvMYBF1 expression was light inducible, implicating VvMYBF1 in the control of VvFLS1 transcription. Sequence analysis of VvMYBF1 and VvFLS1 indicated conserved putative light regulatory units in promoters of both genes from different cultivars. By analysis of the VvMYBF1 amino acid sequence, we identified the previously described SG7 domain and an additional sequence motif conserved in several plant MYB factors. The described motifs have been used to identify MYB transcription factors from other plant species putatively involved in the regulation of flavonol biosynthesis. To our knowledge, this is the first functional characterization of a light-inducible MYB transcription factor controlling flavonol synthesis in fruit.

In plants, flavonoids represent a highly diverse class of low M_r polyphenolic compounds, synthesized via the general phenylpropanoid pathway, giving rise to three major classes: anthocyanins, proanthocyanidins (PAs), and flavonols (Fig. 1). Flavonols play pivotal roles in fresh fruit and fruit products, including grape berries and wine, and contribute substantially to their taste, quality, and nutrition. They exhibit anti-inflammatory, antioxidant, and antiproliferative capacities and thereby account for the health-promoting effects of the consumption of grapes and many other fruits (Havsteen, 2002; Butelli et al., 2008). Grapevine (Vitis vinifera) synthesizes derivatives of the flavonols quercetin, kaempferol, isorhamnetin, and myricetin, which are mainly found as glucosides and glucuronides in grape berries (Cheynier and Rigaud, 1986; Price et al., 1995). These flavonol glycosides have been shown to affect anthocyanin color in wine and aqueous solution by copigmentation (Dimitrić-Marković et al., 2005). Flavonols also protect plants against UV light. In this regard, sunlight and UV-B light were shown to increase concentration of quercetin glycosides in grapevine berries (Price et al., 1995; Downey et al., 2007), petunia (Petunia hybrida; Ryan et al., 2002), mustard (Sinapis alba; Beggs et al., 1987), and soybean (Glycine max; Middleton and Teramura, 1993), while a flavonol synthase (FLS) mutant was shown to be incapable of flavonol production in response to UV radiation in Arabidopsis (Arabidopsis thaliana; Wisman et al., 1998). Interestingly, flavonol synthesis can also be induced by exposure to sunlight at stages during grape berry development when flavonols are normally not accumulating (Downey et al., 2004; Matus et al., 2009). Besides their important function as UV protection shields, flavonols are thought to be involved in the regulation of auxin transport (Buer and Muday, 2004; Peer et al., 2004; Besseau et al., 2007; Santelia et al., 2008). In plant sexual reproduction, flavonols were identified as active compounds necessary and sufficient to allow the formation of functional pollen tubes, thus complementing chalcone synthase (CHS)-deficient petunia (Mo et al., 1992), and to promote tobacco (Nicotiana tabacum) pollen germination and pollen tube growth in vitro (Ylstra et al., 1992).

Figure 1.

Simplified representation of the flavonoid biosynthetic pathway of grapevine leading to the three major classes of end products: flavonols, PAs, and anthocyanins. Note that the activity of FLS is required for the synthesis of flavonols, the activity of UFGT is required for the synthesis of anthocyanins, while the activities of ANR and LAR are specifically required for the synthesis of the precursors catechin and epicatechin of PA formation. ANR, Anthocyanidin reductase; 4CL, 4-coumaroyl-CoA synthase; C4H, cinnamate-4-hydroxylase; CHI, chalcone isomerase; CHS, chalcone synthase; DFR, dihydroflavonol 4-reductase; F3H, flavanone-3β-hydroxylase; F3′H, flavonoid-3′-hydroxylase; F3′5′H, flavonoid-3′,5′-hydroxylase; FLS, flavonol synthase; LAR, leucoanthocyanidin reductase; LDOX, leucoanthocyanidin dioxygenase; OMT, O-methyltransferase; PAL, Phe ammonia lyase; RT, rhamnosyl transferase; UFGT, UDP-Glc:flavonoid-3-O-glucosyltransferase; UGT, UDP-glycosyltransferase. Note that for simplification, flavonoids catalyzed by the enzymes DFR, LDOX, LAR, and ANR have been omitted; therefore, only the final products anthocyanins and PAs are shown.

The biosynthetic pathway forming the basis of accumulation of flavonoids has been elucidated using genetic and biochemical information from many plant species, initially in parsley (Petroselinum hortense; Kreuzaler et al., 1983), subsequently also in maize (Zea mays), petunia, Antirrhinum species (Holton and Cornish, 1995), and Arabidopsis (Shirley et al., 1995), and more recently in fruit crops including apple (Malus domestica; Takos et al., 2006b; Espley et al., 2007), bilberry (Vaccinium myrtillus; Jaakola et al., 2002), and grapevine (Boss et al., 1996; Bogs et al., 2005, 2006). It is notable that flavonoid composition among plant species can be remarkably different. In Arabidopsis, for example, 5′-hydroxylated flavonoids and catechin-based PAs are not present because the Arabidopsis genome does not contain genes encoding the enzymes flavonoid 3′,5′-hydroxylase and leucoanthocyanidin reductase, which have been characterized from grapevine (Bogs et al., 2005, 2006) and other plant species. The synthesis of flavonol aglycones is catalyzed by FLS1, which employs dihydroflavonols as substrates (Fig. 1). Among other enzymes that use the same substrate are flavonoid 3′-hydroxylase and flavonoid 3′,5′-hydroxylase, which mediate the addition of hydroxyl groups to the B-ring of flavanones, flavones, and dihydroflavonols (Hagmann et al., 1983; Kaltenbach et al., 1999), and dihydroflavonol 4-reductase (DFR), which directly competes with FLS1 for the same substrate, dihydrokaempferol (Martens et al., 2002). VvFLS1 is expressed in leaves, tendrils, buds, inflorescences, and developing grape berries. In the skin of Shiraz berries, VvFLS1 expression is highest between flowering and fruit set, then declines to veraison, which is the onset of ripening, and increases again during ripening coincident with the accumulation of flavonols per berry (Downey et al., 2003b). Consistent with their role as important UV light screens, flavonols accumulate mainly in flowers and skins of grape berries, but they have also been reported in leaves and stems. No detectable amounts of flavonols have been identified in pulp and seeds of cv Chardonnay and Shiraz berries, and expression of VvFLS1 was also not detected in these tissues (Souquet et al., 2000; Downey et al., 2003b).

The flavonoid biosynthetic pathway genes are predominantly regulated at the level of transcription, both developmentally and/or in response to various biotic and abiotic stress factors. Transcriptional regulators for anthocyanin and PA synthetic pathways are mainly members of protein families containing R2R3-MYB domains, basic helix-loop-helix (bHLH) domains (also referred to as MYC proteins), and conserved WD repeats (WDR; Weisshaar and Jenkins, 1998; Stracke et al., 2001; Marles et al., 2003). In various plant species, the combination and interaction of R2R3-MYB, bHLH, and WDR factors determine the set of genes expressed (Quattrocchio et al., 1998; Walker et al., 1999; Baudry et al., 2004). Two of them, the MYB and bHLH proteins, are conserved in regulation of anthocyanin and PA pathways in all species analyzed to date (Koes et al., 2005). In vertebrates, the MYB gene family includes C-MYB, A-MYB, and B-MYB, whereas in plants, the MYB family is much more extensive, encompassing 126 R2R3-MYB genes in the Arabidopsis genome that are involved in a wide range of developmental and defense processes. All R2R3-MYB family members share two highly conserved imperfect repeats (R2 and R3) at the N terminus, while the C terminus is highly diverse (Stracke et al., 2001). The bHLH proteins may have overlapping regulatory targets (Zimmermann et al., 2004), but MYB proteins are believed to be key components by allocation of specific gene expression patterns (Stracke et al., 2001). In grape berries, the MYB proteins MYBA1 and MYBA2 (Kobayashi et al., 2002; Walker et al., 2007), MYB5a (Deluc et al., 2006), MYB5b (Deluc et al., 2008), and MYBPA1 (Bogs et al., 2007) act together with bHLH proteins and probably WDR proteins to control anthocyanin and/or PA synthesis. Most R2R3-MYB factors regulating flavonoid biosynthesis (except flavonol synthesis) depend on cofactors, including WDR proteins but, most importantly, a small subgroup of bHLH proteins that share a common motif in their N termini that interacts with a signature motif in the R3 repeat of the N-terminal R2R3 domain of MYB factors (Grotewold et al., 2000). The first MYB factor that was shown to be active without binding a bHLH protein was ZmP from maize, activating a subset of genes for 3-deoxyflavonoids and phlobaphene biosynthesis (Grotewold et al., 1994). Recently, the first R2R3-MYB factors (MYB12/PFG1, MYB11/PFG2, and MYB111/PFG3) regulating flavonol synthesis have been described in Arabidopsis (Mehrtens et al., 2005; Stracke et al., 2007). They are also cofactor independent and individually regulate flavonol accumulation in different organs of developing seedlings (Stracke et al., 2007). MYB12 mainly controls flavonol biosynthesis in roots, while MYB111 predominantly regulates biosynthesis in cotyledons. Flavonol accumulation requires these three factors because triple mutant myb11-12-111 seedlings cannot produce flavonol glycosides, while synthesis of anthocyanins is not affected. Targets of these MYB factors are CHS, chalcone isomerase (CHI), flavanone 3-hydroxylase, FLS, and UDP glycosyltransferases family 1 genes (Mehrtens et al., 2005; Stracke et al., 2007).

Luo et al. (2008) showed that AtMYB12 can be used to produce transgenic tomato (Solanum lycopersicum) containing high amounts of flavonols in fruits. However, to improve flavonol content and composition of crops via agricultural management strategies or marker-assisted breeding, the endogenous genes and regulatory mechanisms of flavonol synthesis have to be examined. Recently, the genome sequence of grapevine became available (Jaillon et al., 2007) and has provided a genomic platform for Matus et al. (2008) to classify 108 members of the grape R2R3-MYB gene subfamily in terms of their genomic gene structures and similarity to their putative Arabidopsis orthologs. A partial cDNA clone (accession no. FJ418175) of an AtMYB12-like gene was identified (Matus et al., 2008, 2009), which is identical to the 3′ coding region of the flavonol regulator gene VvMYBF1.

Grapevine is one of the studied crop plants in terms of the regulation of flavonoid synthesis by transcription factors in fruit, and regulators of flavonoid biosynthesis have also been found in other fruit-producing plants, including strawberry (Fragaria species; Aharoni et al., 2001) and apple (Takos et al., 2006a; Espley et al., 2007). However, a transcription factor that regulates flavonol synthesis in fruit has not been functionally characterized so far in grapevine or any other plant species. To elucidate the regulation of flavonol synthesis in fruit, the flavonol-specific MYB transcription factor VvMYBF1 (accession no. FJ948477) from Shiraz grape was characterized. The expression pattern of VvMYBF1 directly correlated with that of VvFLS1 and subsequent flavonol accumulation during grape berry development. Furthermore, expression of both VvFLS1 and VvMYBF1 was strongly induced in grapevine cell culture after exposure to light. VvMYBF1 specifically induced promoters of VvFLS1 and VvCHI, which are involved in flavonol synthesis, whereas promoters of genes involved in anthocyanin or PA synthesis were not controlled by VvMYBF1. The functionality of VvMYBF1 was shown by complementation of the flavonol-deficient phenotype of the Arabidopsis mutant myb12. Using amino acid motifs found to be conserved in flavonol regulators from Arabidopsis and grapevine, several other putative flavonol regulators have been identified, providing a tool to screen plant databases for putative flavonol regulators from other fruit-producing crops.

RESULTS

Isolation and Sequence Analysis of the MYB Transcription Factor VvMYBF1

Using the conserved R2R3 repeat region and the SG7 motif encoded by the AtMYB12 gene (accession no. CAB09172; Stracke et al., 2001), a similar sequence was detected in the grapevine genomic sequence (highly homozygous genotype PN40024; Jaillon et al., 2007) provided by the French National Sequencing Center Genoscope. Primers were designed to the 5′ and 3′ untranslated regions (UTRs) of the predicted gene to amplify a fragment of 1,125 bp including the open reading frame (ORF), which was named VvMYBF1. The VvMYBF1 ORF was amplified from Shiraz and Chardonnay cDNA, and sequence analysis showed a difference in two amino acids at positions 213 and 359 of the encoded protein sequences. The VvMYBF1 ORF isolated from Chardonnay cDNA was submitted to GenBank (accession no. GQ423422). The transcriptional start of the VvMYBF1 gene from Shiraz was determined by 5′ RACE, and the complete 5′ UTR and the ORF of VvMYBF1 were supplied to GenBank with the accession number FJ948477. The ORF encodes a predicted protein sequence of 367 amino acid residues with a molecular mass of 40.9 kD and a pI of 4.98. Analysis of the deduced amino acid sequence revealed that VvMYBF1 contains an N-terminal R2R3 repeat that corresponds to the DNA-binding domain of plant MYB-type proteins (Fig. 2A). Although sequence similarity between MYB proteins is generally restricted to the N terminus, the SG7 motif (GRTxRSxMK; Stracke et al., 2001) characteristic of flavonol regulators of Arabidopsis was also found to be present in the C terminus of VvMYBF1 with one amino acid substitution (GRTSRWAMK). Additionally, one previously unidentified motif designated SG7-2 ([W/x][L/x]LS) was detected at the C-terminal ends of AtMYB12, AtMYB11, AtMYB111, and VvMYBF1. The conserved R2R3 repeat and the two SG7 motifs have been used as a tool to identify further putative flavonol-specific transcriptional regulators of other plant species, including Sorghum bicolor, Lotus japonicus, rice (Oryza sativa), and Gerbera hybrida (Fig. 2A). We could also detect another R2R3-MYB transcription factor from grapevine on chromosome 5 (accession no. GSVIVT00033103001) that we named VvMYBF2, with both the SG7 and SG7-2 motifs differing significantly from the other flavonol MYB-type transcription factors. The homology within the SG7 domain in the other putative flavonol regulatory proteins allowed us to expand and redefine the motif introduced by Stracke et al. (2001) to [K/R][R/x][R/K]xGRT[S/x][R/G]xx[M/x]K. Except in VvMYBF2, the SG7 motif is not conserved in SlMYB12 from tomato, while the SG7-2 domain is less conserved in ZmP and apple MdMYB22 (Fig. 2A). Phylogenetic analysis revealed the similarity of VvMYBF1 to the flavonol regulators from Arabidopsis, AtMYB12, AtMYB11, and AtMYB111, and other plant MYB proteins (Fig. 2B). The predicted protein sequence of VvMYBF1 was most closely related to MdMYB22, with 92% identity in the R2R3 domain, while the complete protein sequence showed only 39% identity. The similarity between the R2R3 domains of VvMYBF1 and AtMYB12, AtMYB11, and AtMYB111 (Fig. 2A), which have been shown to regulate flavonol synthesis in young seedlings of Arabidopsis (Stracke et al., 2007), was 83%, 81%, and 85% amino acid identity, respectively. The complete protein sequence of VvMYBF1 showed 41% amino acid identity to AtMYB12, 45% identity to AtMYB11, and 38% identity to AtMYB111. Additionally, the R2R3 repeat region of VvMYBF1 does not contain the motif [D/E]Lx₂[R/K]x₃Lx₆Lx₃R for interaction with bHLH proteins (Zimmermann et al., 2004).

Figure 2.

A, Alignment of R2R3-MYB-type transcriptional regulators of flavonol synthesis from Arabidopsis and grapevine and presumptive flavonol regulators from other plant species and the maize phlobaphene regulator ZmP. The position of the R2R3-type MYB domain is indicated below the alignment, while the SG7 and SG7-2 domains are highlighted by black boxes. Amino acids identical in all sequences are shaded black, and amino acids found in more than eight or nine sequences are highlighted in light gray and dark gray, respectively. B, Phylogenetic tree showing selected plant MYB transcription factors from the GenBank or EMBL database. GenBank accession numbers of MYB factors are listed in “Materials and Methods.” Functions of most of the proteins are given in boldface. The scale bar represents 0.1 substitutions per site, and the numbers next to the nodes are bootstrap values from 1,000 replicates.

Expression of VvMYBF1 Correlates with Flavonol Accumulation during Fruit Development

To determine if VvMYBF1 expression during grape berry development correlates with flavonol accumulation, we analyzed Shiraz grape berries collected during the season 2000 to 2001 by quantitative PCR (qPCR). The identical Shiraz developmental series was previously used by Downey et al. (2003b, 2004) to analyze VvFLS1 expression and flavonol accumulation. The pattern of VvMYBF1 expression during early grape development, in skins, and in seeds is shown in Figure 3. VvMYBF1 expression was highest around flowering from 9 to 7 weeks prior to veraison (Fig. 3A). This expression of VvMYBF1 in developing flowers and grape berries with highest expression throughout flowering correlates with the expression of VvFLS1 and flavonol accumulation in developing Shiraz berries (Downey et al., 2003b, 2004). Beginning at 4 weeks before veraison, VvMYBF1 was expressed in skins of developing berries with small peaks at 4 weeks prior to veraison, at veraison, and 6 weeks after veraison (Fig. 3B). Concomitantly, VvFLS1 expression and flavonol accumulation (measured as mg per berry) also peaked 6 weeks after veraison in Shiraz berry skins (Downey et al., 2003b). In seeds, VvMYBF1 was expressed only at very low levels during the entire grape berry development (Fig. 3C), coinciding with no detectable flavonol accumulation and no detectable expression of VvFLS1 in seeds of grape berries (Downey et al., 2003b).

Figure 3.

Expression of VvMYBF1 during grape berry development. A, Early development. B, Skin. C, Seeds. Transcript levels of VvMYBF1 were determined by qPCR using gene-specific primers and corrected to VvUbiquitin1 (TC32075) gene expression. Data points are given as weeks from onset of ripening (veraison; labeled with the arrow), with expression values as means of three replicate PCRs and error bars indicating se. Note that from 8 weeks before veraison (−8), berry skin has been separated from seeds. Flowering occurred 8 weeks before veraison and is labeled with the star.

Induction of VvMYBF1 Expression by Light in Cell Cultures Correlates with Flavonol Accumulation

Previous studies by Downey et al. (2004) demonstrated light induction of flavonol biosynthesis in Shiraz berries. To determine if light-induced synthesis of flavonols could be the result of an increase in VvMYBF1 gene expression, dark-grown Chardonnay petiole cell cultures were exposed to light and VvMYBF1 expression was monitored over an 8-d time period and correlated with flavonol accumulation and VvFLS1 expression data. After exposure to light, the transcript levels of VvMYBF1 increased by approximately 24-fold over the first 12 h compared with expression in the dark after 12 h. Transcript levels of VvMYBF1 continued to increase with light exposure, leading to an approximately 130-fold change in its expression by 24 h (Fig. 4A). Concomitantly, the transcript level of VvFLS1 was also steadily increasing upon light exposure during the first 24 h. VvFLS1 transcript levels were highest after 24 h of light induction, with an approximately 12,300-fold increase compared with expression of VvFLS1 after 24 h in the dark (Fig. 4B). Both transcripts rapidly decreased with 36 h of light induction, with almost no detectable expression of VvMYBF1 and low expression of VvFLS1 after 2 d of light induction (Fig. 4, A and B). Corresponding with the increase in VvMYBF1 and VvFLS1 transcript levels, flavonols were detected after 2 d of light exposure, with the highest levels of flavonols observed at day 6 of the light induction experiment (Fig. 4C; Supplemental Fig. S2, A and B). The early induction of VvMYBF1 gene expression upon light exposure in Chardonnay cells correlates well with a role for this gene in regulating the expression pattern of VvFLS1. The subsequent increase in flavonol levels in Chardonnay cells confirms the inducibility of flavonol biosynthesis, providing evidence that this response is likely to be mediated by VvMYBF1 in response to light.

Figure 4.

Light induces expression of VvMYBF1 and VvFLS1 and subsequent flavonol synthesis in Chardonnay cells. Transcript levels were determined by qPCR and are shown relative to VvUbiquitin1 (TC32075) expression. Each data point represents the mean of three replicate PCRs with error bars indicating se. A, VvMYBF1 expression. B, VvFLS1 transcript levels in Chardonnay cell samples exposed to light. C, Accumulation of total flavonols in Chardonnay cell samples exposed to light (gray bars) and kept in the dark (black bars) determined by HPLC analysis. D, Positions of putative cis-regulatory elements in the promoters of VvMYBF1 and VvFLS1 from Shiraz, Pinot Noir, and Chardonnay relative to the transcriptional start codon. The putative transcriptional start sites are referred to as positions +1, while the translational start codons are marked by the bent arrows. Numbers above the boxes indicate the distance from the transcriptional start. ACS, ACGT-containing sequence similar to ACE in the Arabidopsis CHS promoter (accession no. S000355); MRS, MYB recognition sequence similar to MRE in the Arabidopsis CHS promoter (S000356); RRS, R response sequence similar to RRE in the Arabidopsis CHS promoter (S000407); IBOXCORE element (S000199), LREBOXIPCCHS1 consensus sequence (S000302), MYBCORE (S000176).

After demonstrating that VvMYBF1 and VvFLS1 are light-inducible genes, the promoter sequences of VvMYBF1 and VvFLS1 from Shiraz, cv Pinot Noir, and Chardonnay grapevine were determined and bioinformatic analysis was performed (for accession numbers, see “Materials and Methods”). Sequence analysis revealed that the VvMYBF1 and VvFLS1 promoter sequences of these three grapevine cultivars are very similar and contain several putative light regulatory elements conserved between the cultivars and their respective promoters (Fig. 4D; Supplemental Fig. S1). Hartmann et al. (2005) presented evidence that light responsiveness of the Arabidopsis CHS promoter is mediated by a light regulatory unit (LRU) necessarily containing MYB recognition (MRE) and bZIP recognition (ACE) sequence elements. This LRU is also present in the promoters of the VvMYBF1 and VvFLS1 genes with elements similar to MRE and ACE, which were designated MRS and ACS (Fig. 4D; Supplemental Fig. S1). By comparing the promoter sequences with the PLACE database (Higo et al., 1999), additional putative light regulatory elements were identified in the VvFLS1 promoter (Fig. 4D; Supplemental Fig. S1). These elements are present upstream of the LRU and include an IBOXCORE element involved in binding of MYB factors of light-regulated genes in tomato (Rose et al., 1999) and a LREBOXIPCCHS1 consensus sequence conferring light responsiveness in the parsley CHS1 promoter (Schulze-Lefert et al., 1989). In addition, two MYBCORE elements, which are similar to binding sites for MYB factors in Arabidopsis (Urao et al., 1993) and petunia PhMYB3 (Solano et al., 1995), were identified in the VvFLS1 promoter, while the VvMYBF1 promoter shows one additional MRS located approximately 350 bp upstream to the transcriptional start codon (Fig. 4D; Supplemental Fig. S1B). Except for the LRU, none of the light regulatory elements are present in the VvMYBF1 promoter. The VvMYBF1 promoter also contains elements similar to a R response element that include the bHLH factor consensus binding site (Hartmann et al., 2005), indicating another level of regulation independent of light that is absent in the first 1,000 bp proximal to the transcriptional start of the VvFLS1 gene (Fig. 4D; Supplemental Fig. S1). The similarity of the described promoter sequences for Shiraz, Pinot Noir, and Chardonnay and the conservation of putative light regulatory elements indicates that light regulation of flavonol synthesis in red and white grapevine cultivars is very similar at the molecular level.

VvMYBF1 Activates Promoters of the Flavonoid Pathway Genes Required for Flavonol Synthesis in Fruits

To identify the structural genes of the flavonoid pathway activated by VvMYBF1, transient expression experiments using Chardonnay grape suspension culture cells and the dual-luciferase assay system were conducted (Horstmann et al., 2004; Bogs et al., 2007). VvMYBF1 was introduced into pART7, a vector allowing the expression of genes under the control of the constitutive cauliflower mosaic virus 35S promoter. The VvCHI promoter was chosen as an example for a general flavonoid pathway gene involved in the synthesis of flavonols, anthocyanins, and PAs (Fig. 1). In contrast, VvUFGT and VvANR are specifically required for anthocyanin and PA synthesis, respectively, whereas VvFLS1 catalyzes the final step leading to the synthesis of flavonols (Fig. 1). The grapevine promoters of the genes encoding these proteins were then tested as potential targets of VvMYBF1. Figure 5A summarizes the results of the transient expression experiments carried out to compare the transactivation potential of VvMYBF1, AtMYB12, VvMYBA2, and VvMYBPA1 on the VvFLS1 promoter. VvMYBF1 activated the promoter of the gene VvFLS1 approximately 10-fold and AtMYB12 activated the promoter approximately 3-fold compared with the corresponding control, consisting of promoter without MYB factor, both showing their ability to induce the flavonol-specific branch point gene of grapevine. As negative controls, VvMYBPA1 and VvMYBA2 did not activate the VvFLS1 promoter, confirming their specificity for PA and anthocyanin synthesis, respectively. The addition of AtEGL3, a bHLH factor needed as a cofactor during transcription initiation by the transcription factors VvMYBPA1 and VvMYBA2 in this assay, decreased the activation potential of VvMYBF1 on the VvFLS1 promoter (Fig. 5A). VvMYBF1 also induced the promoters of the general flavonoid pathway gene VvCHI (approximately 12-fold; Fig. 5B), suggesting that it can activate the general pathway providing substrates for flavonol synthesis. The Arabidopsis homolog AtMYB12 activated the VvCHI promoter approximately half as strongly as the grapevine factor, indicating AtMYB12 binding and activation of the VvFLS1 promoter in a heterologous system. As already published, VvMYBPA1 activates the VvCHI promoter (Bogs et al., 2007) to the same extent as VvMYBF1, showing the necessity of the general pathway gene VvCHI for both PA and flavonol synthesis (Fig. 5B). VvMYBA2 does not activate the VvCHI promoter, confirming previous results indicating that VvMYBA2 specifically controls VvUFGT (Fig. 5E; Walker et al., 2007; Deluc et al., 2008). VvMYBF1 also displayed a potential to activate the leucoanthocyanidin dioxygenase (LDOX) promoter, with an approximately 9-fold induction compared with the negative control (Fig. 5C). VvMYBPA1 activates the VvANR promoter approximately 280-fold, showing its specific transactivation potential for the PA branch of the pathway, whereas VvMYBA2 activates the VvANR promoter only 0.6-fold, VvMYBF1 approximately 9-fold, and AtMYB12 approximately 2-fold (Fig. 5D). The anthocyanin-specific promoter of VvUFGT was strongly induced by VvMYBA2 (approximately 133-fold), whereas VvMYBF1, VvMYBPA1, and AtMYB12 did not activate the VvUFGT promoter substantially, therefore confirming the specificity of VvMYBA2 to the VvUFGT promoter and of VvMYBF1 to the VvFLS1 promoter (Fig. 5E). Finally, basal levels of the promoters without effectors (referred to as controls in Fig. 5) indicate that VvFLS1 and VvCHI promoters possess higher background activity compared with the promoters of the VvUFGT, VvANR, and VvLDOX genes, suggesting endogenous transcription factors and/or environmental stress conditions that might induce these promoters during the transient assay (Fig. 5F). To confirm that VvMYBF1 can also induce pathway genes in the heterologous system of Arabidopsis, At7 cells were cotransfected with VvMYBF1 and different Arabidopsis flavonoid pathway gene promoter constructs. Figure 5G summarizes the results and shows that VvMYBF1 is able to induce AtCHS, the initial enzyme in the flavonoid pathway and to a lower but still significant extent AtFLS but not AtDFR, whose gene product directly competes with FLS for the same substrate (dihydroflavonols). In summary, these results suggest VvMYBF1 to be a specific regulator of flavonol synthesis, potentially regulating the general flavonoid pathway (CHS and CHI) and the FLS1 gene specifically directing flavonoid precursors to flavonol formation in grapevine.

Figure 5.

VvMYBF1 specifically activates grapevine and Arabidopsis promoters of flavonoid pathway genes involved in flavonol synthesis in a transient reporter assay system. A to F, The MYB- and bHLH-type transcription factors and promoters used for transient expression in Chardonnay grape cell cultures by particle bombardment. Each transfection contained the renilla luciferase plasmid pRluc for normalization. Columns in A to E represent fold induction of the corresponding promoter plus MYB factor to the respective control (without MYB factor). Each column represents the mean value of at least six independent experiments with error bars indicating se. For F, basal activities of grapevine promoters are relative to renilla luciferase activity. G, A transient Arabidopsis At7 protoplast reporter gene assay system was used to analyze the activation potential of VvMYBF1 on promoters of Arabidopsis flavonoid biosynthesis enzymes. Each column represents the mean of six independent cotransfections with error bars indicating se. ANR, Anthocyanidin reductase; CHI, chalcone isomerase; CHS, chalcone synthase; DFR, dihydroflavonol 4-reductase; FLS, flavonol synthase; LDOX, leucoanthocyanidin dioxygenase; UFGT, UDP-Glc:flavonoid-3-O-glucosyltransferase.

Complementation of the Flavonol-Deficient Phenotype of the Arabidopsis Mutant myb12 by Heterologous Expression of VvMYBF1

The MYB transcription factor AtMYB12 has been shown to control flavonol synthesis in the roots of young Arabidopsis seedlings, while AtMYB111 controls flavonol biosynthesis primarily in the cotyledons (Stracke et al., 2007). To confirm the function of VvMYBF1 as a regulator of flavonol synthesis, VvMYBF1 was expressed under the control of the root-specific AtMYB12 promoter in Arabidopsis ecotype Columbia (Col-0) wild-type and myb12 mutant plants. Whereas the myb12 mutant showed a flavonol-deficient phenotype in roots, all BASTA-resistant independent transgenic Col-0 lines (Col-0/proAtMYB12∷VvMYBF1 Col-0/VvMYBF1 01–10 [10 lines]) and transgenic myb12 lines (myb12/proAtMYB12:VvMYBF1 010, -012, -013, -017, and -018 [five lines]) transformed with VvMYBF1 showed flavonol accumulation and therefore normal wild-type phenotype in roots. The accumulation patterns of flavonols in Col-0, myb12, and complemented mutants have been visualized with in situ diphenylboric acid-2-aminoethylester (DPBA) staining of flavonols. This is a powerful method for detection of flavonols in plant tissues that can be used to differentiate in vivo between quercetin (orange fluorescence) and kaempferol (greenish fluorescence; Peer et al., 2001). In line with previous results of Stracke et al. (2007), the Col-0 seedlings showed intense orange fluorescence 5 d after germination, indicating massive accumulation of quercetin (Fig. 6A). Root-specific expression of VvMYBF1 in Col-0 (Col-0/proAtMYB12∷VvMYBF1) did not alter flavonol-specific staining in roots and cotyledons as well as the hypocotyl root transition zone compared with untransformed Col-0 (Fig. 6B). The myb12 mutant seedlings displayed drastically reduced flavonol levels in the root and the hypocotyl root transition zone, while flavonol accumulation in cotyledons and the distal elongation zone in the root was unaffected (Fig. 6C). Complementation of the myb12 phenotype was obtained in roots of the transgenic lines myb12/proAtMYB12∷VvMYBF1 010, -012, -013, -017, and -018, which all expressed VvMYBF1 under a root-specific promoter in myb12 Arabidopsis seedlings and showed orange staining indicative for quercetin accumulation in the root (Fig. 6D). To confirm the in vivo staining results, the total flavonol accumulation profiles of the complemented lines were analyzed by HPLC and compared with flavonol profiles of Col-0, myb12 (Supplemental Fig. S2C), and myb11-12-111 mutant lines. The triple mutant consistently displayed a dramatic reduction in flavonol content: no flavonols were detected in this mutant compared with Col-0 (Stracke et al., 2007; Fig. 6E), while the single mutant myb12 showed a 5-fold decrease in accumulation of flavonols relative to Col-0 (Fig. 6E). Complemented lines always showed significantly elevated levels of flavonols compared with the myb12 mutant, with lines myb12/proAtMYB12∷VvMYBF1 010, -012, and -013 showing 7-, 9-, and 6-fold increased values compared with myb12, respectively. These lines even showed augmented levels of total flavonol content compared with Col-0. Lines myb12/proAtMYB12∷VvMYBF1 017 and -018 displayed lower levels of flavonol accumulation (Fig. 6E), showing 4- and 3-fold increased values compared with myb12, although these lines exhibited the strongest staining phenotypes (Fig. 6D). These results clearly support the in vivo staining data and confirm that VvMYBF1 is able to complement the Arabidopsis myb12 mutant deficient in flavonol synthesis. Therefore, VvMYBF1 from grapevine is a functional R2R3-MYB transcription factor involved in the regulation of flavonol biosynthesis.

Figure 6.

Flavonol accumulation in Arabidopsis seedlings. A to D, Seedlings were stained with the flavonol-specific dye DPBA. Flavonol staining in Arabidopsis seedlings was visualized using epifluorescence microscopy. Images of representative seedlings of Col-0 (A), Col-0 expressing VvMYBF1 (Col-0/proAtMYB12∷VvMYBF1; B), myb12 mutants (C), and one representative complemented line expressing VvMYBF1 (myb12/proAtMYB12∷VvMYBF1; D) are shown. Bars = 0.1 cm. E, Comparison of the relative flavonol content of methanolic extracts of Arabidopsis seedlings by HPLC analysis. The results are from one representative experiment that was repeated with similar relative values. FW, Fresh weight.

DISCUSSION

Amino Acid Sequence Features of VvMYBF1 and the Flavonol Clade of MYB Transcription Factors

Analysis of the amino acid sequence of VvMYBF1 revealed the presence of the highly conserved N-terminal R2R3 domain characteristic of MYB-related proteins. Similar to the MYB12 factor from Arabidopsis and the MYB protein P from maize, which function independently of a bHLH cofactor (Grotewold et al., 2000; Mehrtens et al., 2005), the R2R3 domain of VvMYBF1 does not contain the bHLH binding site for interaction with R/B-like bHLH proteins (Zimmermann et al., 2004). This cofactor independency seems to be specific for the MYB factors of the flavonol/phlobaphene clade, whereas other MYB transcription factors require the interaction with the bHLH proteins to control gene expression (Fig. 2B). Similar to the other 126 members of the R2R3-MYB protein family in Arabidopsis, the R2R3 repeat region of VvMYBF1 is highly conserved, whereas the C-terminal region is highly diverse (Stracke et al., 2001; Fig. 2A). Albeit this C-terminal diversity, the MYB factor families in Arabidopsis and rice have been categorized into subgroups on the basis of conserved amino acid sequence motifs detected in the C terminus of the MYB proteins (Kranz et al., 1998; Jiang et al., 2004). The C-terminal SG7 motif described by Stracke et al. (2001) was also found in VvMYBF1, suggesting similarities in function to the flavonol regulators of Arabidopsis. An additional previously unidentified motif, SG7-2, was detected by examining the protein alignment of VvMYBF1, AtMYB12, AtMYB11, and AtMYB111 at the very C-terminal end of these proteins (Fig. 2A). Due to the high sequence similarity of the R2R3 domain of VvMYBF1 to the members of the MYB subgroup 7 and the presence of the SG7 motif, VvMYBF1 was classified as a putative flavonol regulator of MYB subgroup 7. The R2R3 domain and the putative flavonol regulator-specific SG7 and SG7-2 motifs were used to screen EST databases for other putative flavonol regulators, elucidating that both motifs are also present in MYB factors of other plant species (Fig. 2A). Consensus sequences for both motifs have been redefined, expanding previous results obtained by Stracke et al. (2001) for the SG7 motif (Fig. 2A). Both motifs were detected in the maize transcription factor ZmP, in MYB-type transcription factors from S. bicolor, L. japonicus, and rice, and in MYB1 from G. hybrida (Fig. 2A). Single motifs have been detected in SlMYB12 from tomato and in MdMYB22 from apple, which has been shown to regulate the CHS promoters of different plant species (Hellens et al., 2005), while neither motif could be detected in VvMYBF2, another R2R3-MYB transcription factor from grapevine (Fig. 2A).Therefore, the combination of the SG7 and SG7-2 motifs with the R2R3 domain will be a useful tool for the initial identification of R2R3-MYB proteins that might regulate flavonol synthesis. Both motifs may be part of a specific functional domain outside the DNA-binding region of flavonol regulators. Evidence for the functional importance of C-terminal motifs as activation domains of MYB proteins has been found for the MYB factors MYB2 and WEREWOLF of Arabidopsis (Urao et al., 1996; Lee and Schiefelbein, 2001).

Additionally, sequence analysis revealed a highly conserved sequence N terminal to the R2R3 domain (MGR[A/T]PCC[E/D]K[V/I]G) in the presumptive MYB-type flavonol regulators analyzed during this study (Fig. 2A). This motif is also present in VvMYBF2 and in the PA-specific regulator MYBPA1 but not in the anthocyanin regulator MYBA2 (data not shown), indicating that, based on sequence similarities, flavonol regulators seem to be more closely related to PA than to anthocyanin regulators (Fig. 2B). This hypothesis is intriguing, because the flavonoid biosynthetic pathway leading to flavonols and PAs first concomitantly appeared in the group Filicopsida (ferns), while anthocyanins finally made their appearance in the gymnosperms later during development (Markham, 1988). It is worth considering if this motif did belong to the R2R3 domain in the common ancestors of flavonoid regulators and was lost during the speciation process leading to the formation of anthocyanin regulators, but only further investigations regarding evolutionary relationships between MYB-type transcription factors in plants can strengthen this hypothesis.

In summary, our analyses clearly support the categorization of VvMYBF1 and the other putative flavonol MYB factors from different plant species as members of the flavonol/phlobaphene clade of R2R3-MYB transcription factors (Fig. 2).

Spatiotemporal and Light-Dependent Expression of VvMYBF1

During grape berry development, the synthesis of flavonols, anthocyanins, and PAs and the expression of the respective flavonoid pathway genes are temporally separated, presumably to avoid competition for common substrates during synthesis of the different flavonoid compounds (Robinson and Davies, 2000; Downey et al., 2003a, 2003b, 2004; Bogs et al., 2005). Flavonols accumulate before anthesis and increase during flowering, probably to protect the inflorescence and pollen against UV light. After bloom, accumulation of flavonols and expression of FLS1 decrease and PA synthesis is induced (Downey et al., 2003a, 2003b, 2004; Bogs et al., 2005). VvMYBF1 is highly expressed in buds and flowers early in berry development, with a decrease of expression 2 weeks after flowering (Fig. 3). This early expression of VvMYBF1 correlates with the accumulation of flavonols and the expression of VvFLS1 during flowering, although VvMYBF1 expression does not precede the highest expression of VvFLS1 and accumulation of flavonols. It was shown that expression of transcription factors normally precedes the expression of the corresponding regulated genes (Bogs et al., 2005, 2007). This consecutive induction of VvMYBF1 and VvFLS1 expression might be present in flower buds between 9 and 10 weeks before veraison, which have not been analyzed. Additionally, the decrease of VvMYBF1 expression after flowering seems to be delayed when compared with VvFLS1 expression, which may be due to an additional function of VvMYBF1 or the repression of VvFLS1 by another factor. In developing grape berries, flavonols seem to be present only in skins, with increases of synthesis and VvFLS1 expression with ripening of the berries (Downey et al., 2003b). VvMYBF1 is expressed in skins of berries and could regulate flavonol synthesis; however, expression is not increasing when berry ripening proceeds (Fig. 3). This increase of flavonol synthesis before harvest of the berries could be regulated by VvMYBF2, a MYB factor also belonging to the flavonol MYB factor clade (Fig. 2B). It will be interesting to analyze the expression pattern of VvMYBF2 and its function in the regulation of flavonol synthesis. The transcription factor VvMYB5b does not belong to the SG7 flavonol subgroup of R2R3-MYB factors (Fig. 2), but it is also expressed early and late in berry development when flavonols are synthesized (Deluc et al., 2008). Ectopic VvMYB5b expression in tobacco increased expression of a number of structural genes in the phenylpropanoid pathway and modified the levels of various flavonoid compounds in flowers, including flavonols. The authors suggest that VvMYB5b is involved in a general up-regulation of the flavonoid pathway and acts together with the specific factors VvMYBA and VvMYBPA1 to control anthocyanin and PA synthesis, respectively (Deluc et al., 2008). In this respect, VvMYB5b could also play a role in the regulation of flavonol synthesis during grape berry development. In seeds, VvMYBF1 is expressed only at very low levels during grape berry development (Fig. 3C), coinciding with the absence of detectable flavonol accumulation and low expression of VvFLS1 in seeds (Downey et al., 2003b). Comparing expression data of VvMYBF1 with VvFLS1 expression and accumulation of flavonols in developing Shiraz berries (Downey et al., 2003b, 2004), we postulate that VvMYBF1 is responsible for the induction of flavonol synthesis in buds, flowers, and skins of young developing berries (Fig. 3).

Flavonol biosynthesis can be induced by light in Shiraz (Downey et al., 2004) and cv Cabernet Sauvignon (Matus et al., 2009) berries even during developmental stages when flavonols are normally not synthesized. In the latter study, Matus et al. (2009) were able to induce the expression of VvFLS1 (termed VvFLS4) and a partial cDNA (accession no. FJ418175) corresponding to the 3′ end of VvMYBF1 by exposure to sunlight in post-veraison grape berry skins. Our expression analysis in grape suspension culture cells confirmed these results: expression levels of VvMYBF1 and VvFLS1 were significantly increased after 6 h of exposure of Chardonnay cell culture to light, leading to a concomitant increase in total flavonol content (Fig. 4, A–C; Supplemental Fig. S2A). The early induction of VvMYBF1 gene expression upon light exposure in the Chardonnay cells after 6 h (Fig. 4A) correlates well with a role for this gene in regulating the expression of VvFLS1 (Fig. 4B). The coinduction of VvFLS1 and VvMYBF1 implies that VvFLS1 light induction is not exclusively mediated via VvMYBF1, because induction of VvMYBF1 should then precede the expression of the VvFLS1 gene. However, as we have not measured gene expression before 6 h after light exposure, early expression of VvMYBF1, which precedes the induction of VvFLS1, cannot be excluded. The delay of about 12 h for the increase of flavonol content of the Chardonnay cells compared with the onset of VvFLS1 and VvMYBF1 expression (Fig. 4) might reflect the period of time needed to subsequently translate VvFLS1 and synthesize flavonols. This increase in flavonol levels of the Chardonnay cells eventually confirms the light induction of flavonol biosynthesis and provides evidence that VvMYBF1 is involved in mediating this light induction response in Chardonnay (Fig. 4C; Supplemental Fig. S2A).

In Arabidopsis, LRUs have been described for the AtCHS and AtFLS promoters and were characterized to contain an MRE and an ACE cis-element, which are recognized by MYB and bZIP factors, respectively, to synergistically confer light responsiveness (Hartmann et al., 2005). Analysis of VvFLS1 and VvMYBF1 promoter regions from Pinot Noir, Chardonnay, and Shiraz revealed the presence of LRUs in promoters of both genes (Fig. 4D; Supplemental Fig. S1), suggesting that VvMYBF1 and VvFLS1 induction by light in red- and white-berried cultivars may be mediated via MYB and bZIP transcription factors. The Arabidopsis HY5 encodes a bZIP factor that is a key positive regulator of light signaling during plant development and that regulates numerous genes during photomorphogenesis, including AtCHS, AtFLS, and AtMYB12 (Lee et al., 2007). The presence of the LRUs in VvFLS1 and VvMYBF1 promoters and their light responsiveness indicate that they are targets of a grapevine HY5 homolog. Comparison of the LRUs from VvFLS1 promoters with the Arabidopsis FLS promoter (Hartmann et al., 2005) revealed that the relative positions of the ACE and MRE in the AtFLS promoter as well as the ACS and MRS in the VvFLS1 promoter were very similar (Fig. 4D; Supplemental Fig. S1C). Our sequence analysis indicates that besides the LRUs, additional putative light regulatory cis-elements are present in both red and white grapevine cultivars, which could also play a role in the regulation of flavonol synthesis in response to light (Fig. 4D; Supplemental Fig. S1). Additional MYB factor, bZIP, and bHLH binding sites have been found in grapevine promoters (Fig. 4D; Supplemental Fig. S1), possibly regulating light responsiveness and/or developmental regulation of the promoters apart from the LRUs. The mechanism of derepression in response to light as proposed by Hartmann et al. (2005) is unlikely for the grapevine FLS1 gene, as an additional ACS motif serving as a binding site for an inhibitor protein was not found next to the LRU.

From our data, VvMYBF1 could contribute to both the developmental and light regulation of flavonol synthesis in both white and red grapevine cultivars; however, this has yet to be proven experimentally. The comparison of Shiraz, Pinot Noir, and Chardonnay promoters of VvFLS1 and VvMYBF1 showed that the respective sequences are very similar and contain the identical putative light regulatory cis-elements (Fig. 4D; Supplemental Fig. S1). This indicates that also at the molecular level, light regulation in red and white grapevine cultivars is mediated by a conserved mechanism. Future work should consider mutational analysis of the putative MYB binding sites in the VvMYBF1 and VvFLS1 promoters and light induction experiments to show whether VvMYBF1 binds to these elements and if they are involved in light responsiveness.

VvMYBF1 Specifically Activates Select Flavonol Pathway Genes

Transient promoter assays of VvMYBF1 in grape suspension culture cells revealed substantial functional similarities between the R2R3-MYB transcription factors VvMYBF1 and AtMYB12, as predicted by structural similarities (Fig. 2). AtMYB12 was previously shown to activate a subset of flavonoid pathway genes without the need of a bHLH cofactor (Mehrtens et al., 2005). VvMYBF1 is also missing the bHLH interaction motif in the R2R3 domain (Fig. 2A), thereby activating a subset of flavonoid pathway genes without the need of a cofactor (Fig. 5, A–C), confirming that flavonol synthesis does not depend on bHLH cofactors in Arabidopsis and grapevine. This was supported by the transient expression assays, where inclusion of a bHLH was not required for activation by VvMYBF1 and the presence of the bHLH actually diminished activation. This decrease cannot be explained by the fact that EGL3 binds to the FLS1 promoter, thereby blocking the nearby MRS for binding of VvMYBF1, because a significant activation or repression in the presence of only EGL3 could not be observed (Fig. 5F) and a bHLH binding motif is not located in the FLS1 promoter. The inhibitory effect on the VvFLS1 promoter activity might be explained by a recruitment of unknown interacting factors to the bHLH proteins that bind to DNA (as suggested by Hernandez et al., 2004), thereby influencing the capability of the MYBF1 transcription factor to activate the VvFLS1 promoter. Our results of the transient promoter assays displayed the ability of VvMYBF1 to activate the flavonol-specific gene VvFLS1 and the promoters of VvCHS and VvCHI, which encode enzymes providing the precursors for flavonol, anthocyanin, and PA synthesis (Fig. 5). These results reveal the capacity of VvMYBF1 to control the entire pathway leading to flavonol synthesis in grapevine. Similar results were obtained for the control of flavonol synthesis by MYB12, MYB11, and MYB111 of Arabidopsis (Stracke et al., 2007). We have shown that MYB12 and VvMYBF1 have the ability to regulate the CHS and FLS1 promoters of Arabidopsis and grapevine (Fig. 5G), suggesting conservation of the MYB binding sites in the respective heterologous promoters. The basal promoter activities of VvFLS1 and VvCHI in grapevine suspension cells are about 100-fold higher compared with the activities of the VvUFGT and VvANR promoters (Fig. 5F), suggesting that either these promoters still drive a low level of mRNA production when not stimulated or endogenous factors specifically induce these promoters. This high basal activity might also be triggered by the cell culture medium containing hormones (e.g. auxin) or various environmental factors that were shown to influence flavonol synthesis (for review, see Winkel-Shirley, 2002).

In grapevine, anthocyanin synthesis is specifically controlled by the transcription factors VvMYBA1 and VvMYBA2 (Kobayashi et al., 2002; Walker et al., 2007), whereas PA synthesis is regulated by VvMYBPA1 and VvMYBPA2 (Bogs et al., 2007; Terrier et al., 2009). The results of our promoter assays confirm previous data of Bogs et al. (2007) showing that in grape, the transcription factors VvMYBA1/VvMYBA2 and VvMYBPA1 control whether the synthesis of anthocyanins or PAs is induced by regulating the VvUFGT or VvANR and VvLDOX promoters, respectively (Fig. 5, D and E). In line with these results, our promoter assays show that VvMYBF1 controls the expression of the flavonol-specific gene VvFLS1 but not of the VvUFGT or VvANR gene, whereas VvMYBA2 and VvMYBPA1 were not able to induce the VvFLS1 promoter substantially (Fig. 5, A, D, and E). Surprisingly, the promoter of the VvLDOX gene, which encodes an enzyme that was previously shown to produce anthocyanidins as substrates for anthocyanin and PA production (Abrahams et al., 2002), was activated moderately by VvMYBF1 (Fig. 5C). This is in line with previous results, suggesting an alternative route to produce flavonols in plants by takeover of DFR activity (Wilmouth et al., 2002; Wellmann et al., 2006; Stracke et al., 2009).

Taken together, our results suggest that the tissue- and development-specific regulation of flavonoid biosynthesis is orchestrated by the grapevine transcription factors VvMYBF1, VvMYBA1/VvMYBA2, and VvMYBPA1 and is mediated by their ability to specifically induce VvFLS1, VvUFGT, and VvANR gene expression, respectively. This network of transcriptional control could include VvMYBPA2 (Terrier et al., 2009), VvMYBF2 (Fig. 2A), and other transcription factors that coordinate regulation during development but also in response to various environmental factors including light and temperature (for review, see Downey et al., 2006).

VvMYBF1 Is a Functional Regulator of Flavonol Synthesis

In Arabidopsis seedlings, flavonols appear to be present in the entire seedling, but accumulation was preferentially observed in the cotyledons, the hypocotyl root transition zone, the distal root elongation zone, and the root cap (Sheahan and Rechnitz, 1992; Murphy et al., 2000; Peer et al., 2001; Fig. 6A).While the root cap often stained slightly green, indicative for kaempferols in the area between the root cap and the distal elongation zone, the meristematic region showed no staining. This accumulation pattern results from differential expression of AtMYB12, AtMYB11, and AtMYB111, which encode the MYB proteins regulating flavonol synthesis in Arabidopsis. AtMYB12 is preferentially expressed in the root and the hypocotyl root transition zone of Arabidopsis seedlings, while AtMYB111 expression is mainly detected in cotyledons (Stracke et al., 2007). Therefore, the myb12 mutant still accumulates flavonols in cotyledons, whereas flavonols are absent in its roots. This phenotype was confirmed by our in situ DPBA staining of myb12 (Fig. 6C). Comparing in situ staining and HPLC profiles of flavonols in myb12 mutants and mutants expressing VvMYBF1 (Fig. 6; Supplemental Fig. S2C), it is obvious that VvMYBF1 complements the flavonol-deficient phenotype of myb12. However, the orange/yellow DPBA staining in roots of the complemented mutant lines, indicative for quercetin derivatives, does not reach the level of Col-0 roots, whereas the total flavonol content of the complemented mutants is even higher than for Col-0 plants (Fig. 6). This could be due to the low but still significant activity of the MYB12 promoter in cotyledons (Stracke et al., 2007), inducing VvMYBF1 expression in this tissue and therefore increasing the total flavonol amount of the complemented mutants.

It is of great interest to engineer flavonol content of fruit to take advantage of the benefits of flavonols for human health. This was successfully achieved in tomato by ectopic expression of the Arabidopsis MYB12 gene (Luo et al., 2008). The results of Stracke et al. (2007) suggest that the transactivation capacity of the flavonol MYB regulators (MYB12/MYB11/MYB111) can vary for specific target genes, which possibly leads to the different flavonol levels and composition in Arabidopsis. Therefore, it is possible that ectopic expression of VvMYBF1 in fruits like tomato or grape can lead to different flavonol levels and composition as compared with the expression of the Arabidopsis MYB12 factor, considering on the one hand the genetic differences of the flavonoid pathway genes in Arabidopsis and grapevine and on the other hand the different target specificities of transcriptional regulators like VvMYBF1 and AtMYB12. In various grape cultivars, differential expression of VvMYBF1 might lead to different flavonol contents, similar to the example of VvMYBA expression controlling the anthocyanin/fruit color content (Castellarin and Di Gaspero, 2007). Finding the genetic differences of these MYB factors in different grapevine cultivars might lead to molecular markers for breeding of grapevine with optimized flavonoid levels and composition.

MATERIALS AND METHODS

Bioinformatics

Oligonucleotides were designed using Primer3 (http://biotools.umassmed.edu/bioapps/primer3_www.cgi) on grape sequences provided by Genoscope (http://www.cns.fr/externe/English/corps_anglais.html, France) and purchased from Biomers. The sequences of all clones were confirmed by sequencing carried out by Starseq. DNA promoter sequences were screened for plant cis-acting regulatory DNA elements manually and by the use of PLACE (http://www.dna.affrc.go.jp/PLACE/; Higo et al., 1999). Sequence alignments were assembled using the ClustalW program (Ramu et al., 2003), and the edition of aligned sequences was performed using GeneDoc software (version 1.6). Evolutionary trees were built up using the following GenBank accession numbers and MEGA 3.1 software (Kumar et al., 2004): P27898 (ZmP), AAX44239 (yellow seed1 [Sorghum bicolor]), EAY89678 (hypothetical protein [Oryza sativa]), CAD87007 (MYB1 protein [Gerbera hybrida]), ACB46530 (SlMYB12 [Solanum lycopersicum]), CAB09172 (AtMYB12), NP_191820 (AtMYB11), AAK97396 (AtMYB111), BAF74782 (MYB-related protein [Lotus japonicus]), DQ074470 (MdMYB22 [Malus domestica]), FJ948477 (VvMYBF1), GSVIVT00033103001 (VvMYBF2), U26935 (AtMYB5), AAS68190 (VvMYB5a), Q58QD0 (VvMYB5b), AM259485 (VvMYBPA1), AAA82943 (PmMYBF1), AAF18515 (AtMYB3), NP_179263 (AtMYB32), NP_179263 (AtMYB7), AAC83582 (AtMYB4), Q9FJA2 (AtMYB123/TT2), D88619 (OsMYB3), AAA33482 (ZmC1), AAG42001 (AtMYB75/PAP1), NP_176811 (AtMYB113), AAG42002 (AtMYB90/PAP2), BAD18977 (VvMYBA1), BAD18978 (VvMYBA2), AAF66727 (PhAN2), AAQ55181 (LeANT1), Z13996 (PhMYB1), CAA55725 (AmMIXTA), NP_177548 (AtMYB122), NP_200950 (AtMYB28), NP_196386 (AtMYB29), P27900 (AtMYB0/GL1), CAC01874 (AtMYB66/WER), NP_180805 (AtMYB101), NP_187751 (AtMYB65), NP_850779 (AtMYB33), NM_123397 (AtMYB23), and NP_176812 (AtMY114).

Plant Material

Grapevine (Vitis vinifera ‘Shiraz’) tissues were collected from a commercial vineyard (Slate Creek, Willunga, Australia) during the 2000 to 2001 season. The sampling procedure and the dynamic of ripening (berry growth, degrees Brix values, and chlorophyll content) of the used developmental series are described in detail by Downey et al. (2003b, 2004). Exposure of the berry clusters to the light was controlled as described for “exposed” clusters (Downey et al., 2004). All samples were frozen in liquid nitrogen upon collection in the field and stored at −80°C until analyzed. Weekly subcultures of suspension cells established from cv Chardonnay petiole callus culture were grown until log phase, filtered, and inoculated at a cell density of 10% (v/v) in liquid grape Cormier medium (adapted from Do and Cormier, 1991). The suspension culture was grown at 25°C in the dark with continuous shaking at 90 rpm.

The myb12-1f mutation underlying the Arabidopsis (Arabidopsis thaliana) myb12 mutant was described by Mehrtens et al. (2005), and the triple mutant generation (myb11-12–111) is depicted by Stracke et al. (2007). For growing Arabidopsis seedlings on plates, seeds were surface sterilized and sown on 0.5 Murashige and Skoog/1.5% (w/v) Suc/0.8% (w/v) plant agar plates, kept for 2 d at 4°C in the dark, and then transferred to a phytochamber with 14 h of light illumination with 160 μmol m⁻² s⁻¹ per day at 24°C. For HPLC analysis, seedlings were harvested 5 d after germination, frozen in liquid nitrogen, and stored at −80°C until use.

Preparation of cDNA

Total grapevine RNA was isolated from the various plant tissues and developmental stages according to Downey et al. (2003a). RNA of Arabidopsis seedlings was isolated with the RNeasy Kit (Qiagen) following the supplier's protocol. RNA quality was verified by agarose gel electrophoresis in addition to the absorbance ratios (A₂₆₀/A₂₈₀) of 1.8 to 2.0. For cDNA synthesis, 3 μg of grapevine or 1 μg of Arabidopsis total RNA was reverse transcribed in 20 μL for 1 h at 42°C using oligo(dT)₁₈ primer, RnaseOut, and SuperScript III reverse transcriptase (Invitrogen Life Technologies).

RACE

To identify the putative transcriptional start site of the VvMYBF1 gene, a RACE library from Shiraz RNA was generated according to the GeneRacer Kit (Invitrogen). Outer and nested primers MYBF1 3′ RACE (5′-CCATTCCCATCATCATCGCCTTTGG-3′) and MYBF1 3′ RACE nested (5′-GCGCTTGTGGGCTGTGGTTACCTTT-3′) were used for 5′ RACE following the supplier's protocol. The expected RACE PCR product was gel extracted and inserted into pDRIVE (Qiagen) to be analyzed by sequencing.

Expression Analysis of VvMYBF1

qPCR was carried out using the SYBR Green method for detection of double-stranded PCR products on an iCycler optical module real-time cycler (Bio-Rad). VvUbiquitin1 (TC32075; The Institute for Genomic Research database) was used for normalization of transcript levels of VvMYBF1 in grapevine during grape berry development by detection of a 182-bp amplicon with the primers VvUbiquitin Forward (5′-GTGGTATTATTGAGCCATCCTT-3′) and VvUbiquitin Reverse (5′-AACCTCCAATCCAGTCATCTAC-3′). The primers MYBintF (5′-GGAGGTTGAGGGGTTGTG-3′) and MYBintR2 (5′-AAGTTGGGGAAGAGCAGGAG-3′) were used to detect the transcript level of VvMYBF1 in grapevine by amplification of a 214-bp PCR fragment from the 3′ region of the gene. PCR was carried out using 0.5 μL of 10 μm primer (each), cDNA (diluted 1:20), 7.5 μL of 2× ABsolute QPCR SYBR Green Fluorescein Mix (ABgene), and water in a final volume of 15 μL. The efficiency of the primers was tested in preliminary experiments with dilutions of the pART7MYBF1 plasmid and purified PCR products maintaining a value of r² ≥ 0.95. Data points in qPCR time courses are reported as means ± sd of three technical replicates with thermal cycling conditions identical for all primer pairs: 95°C for 15 min, followed by 95°C for 30 s, 58°C for 30 s, and 72°C for 30 s for 35 cycles, followed by a melt cycle from 50°C to 96°C. The difference between the cycle threshold (Ct) of the target gene and the Ct of ubiquitin, Ct = Ct_Target − Ct_Ubiquitin, was used to obtain the normalized expression of target genes, which corresponds to 2^−Ct. Q-Gene software (Muller et al., 2002) was used to calculate the mean normalized expression of VvMYBF1.With all cDNAs used, the primer set gave a single PCR product that was verified by agarose gel electrophoresis and by determination of the melt curves for the product at the end of each run.

Light Induction Experiments with Chardonnay Suspension Cell Cultures and HPLC Analysis of Flavonols

For light induction experiments, Chardonnay grape suspension culture cells were deposited onto sterile filter paper discs and placed on solid grape Cormier medium as described by Torregrosa et al. (2002). The plates containing the Chardonnay cells were cultured under the following conditions: temperature, 22°C; light, 240 μmol m⁻² s⁻¹ (cool white); 16-h light cycle. Control plates were kept in the dark and incubated under the same conditions. Cells were harvested at 6 and 12 h after the onset of light exposure followed by daily sampling on days 1, 2, 3, 6, and 8. Cells were harvested at the same time on each day by collecting them with a spatula, immediately freezing them in liquid N₂, and grinding to a fine powder. For HPLC analysis, three separate 0.1-g aliquots of frozen samples were extracted with 1 mL of acidified methanol by rotating on a spin wheel (30 rpm) for 1 h in darkness at room temperature. The resulting mixture was centrifuged for 15 min at 13,000g, and the supernatant was used for flavonol determination using the HPLC separation method as described by Downey et al. (2007).

Cloning of Promoter and Transcription Factor Constructs

For bioinformatic studies, promoter regions of VvMYBF1 from Shiraz and Chardonnay grapevine were isolated from genomic DNA using the same set of primers and PCR conditions. The forward primer VvMYBF1ProF2 (5′-TATGAGCTCAAAGGAAGGTAATGTTTAACCAAA-3′) and reverse primer VvMYBF1ProR1 (5′-TATCTCGAGTTCCAAACCCAGCAGACA-3′), with restriction sites in boldface, were used to produce PCR products of the size of 1,258 bp using Phusion DNA Polymerase (Finnzymes). The VvFLS1 promoter from Chardonnay was isolated from genomic DNA using the primers FLS1for1 (5′-CGATTCTTTCCATCATTTCAC-3′) and FLS1rev1 (5′-TTCTTCTGTGTTGTCTTGGGTTT-3′), producing a 952-bp fragment with the use of Phusion DNA Polymerase (Finnzymes). The VvMYBF1 and VvFLS1 promoter fragments were inserted into pDRIVE (Qiagen) following the instructions in the manual. The cloned promoter sequences of Chardonnay (VvMYBF1 [accession no. GQ423421] and VvFLS1 [accession no. GQ423420]) and Shiraz (VvMYBF1 [accession no. GQ423423]) were determined and submitted to GenBank.

For transient expression assays, a 3,032-bp fragment of the VvFLS1 promoter (accession no FJ948478) was isolated from Shiraz genomic DNA by specific primers using Phusion DNA Polymerase (Finnzymes) to clone the isolated promoter region as a BamHI/XhoI fragment into the luciferase vector pLuc (Horstmann et al., 2004). The DNA sequences of the primers specific for VvFLS1 promoter were as follows, with restriction sites in boldface: FLSfor2 (5′-ATAGGATCCTAGGAGCCTTGTACACCCTTG-3′) and FLSrev1 (5′-ATACTCGAGACTGCTTCCCTCTCTCTCTTACT-3′). The design of the effector constructs of VvMYBPA1 (accession no. AM259485), VvMYBA2 (BAD18978,), and AtEGL3 (NM20235) and the cloning of the promoter fragments of VvCHI (X75963), VvANR (CAD91911), VvUFGT (AY955269), and VvLDOX (AF290432) into pLuc were described by Bogs et al. (2007). Full-length AtMYB12 (AT2G47460) was amplified using cDNA from Arabidopsis leaves and cloned in pDRIVE using the primers MYB12F (5′-CTACTCGAGATTGAGGCGCATGT-3′) and MYB12R (5′-GTCTAGATCATGACAGAAGCCAAGC-3′), with restriction sites labeled in boldface, and subcloned as a XhoI/XbaI fragment into pART7 Gleave, 1992). The VvMYBF1 ORFs from Shiraz and Chardonnay were amplified as a 1,125-bp PCR fragment from a mixture of cDNAs covering the whole grape berry development of Shiraz and Chardonnay. The primers that were designed to bind to the putative 5′ and 3′ UTRs of the predicted gene were VvMYB12F (5′-ATACTCGAGGGTTTGGAAATGGGGAGAG-3′) and VvMYB12R (5′-CTAGGTACCGAGCAGAAGAAAATCAAGAAAGAAG-3′), with restriction sites labeled in boldface, to give a XhoI/KpnI fragment by the use of Phusion Polymerase (Finnzymes). The VvMYBF1 ORF from Shiraz (accession no. FJ948477) was cloned and sequenced in the vector pART7 (Gleave, 1992) to give pART7MYBF1, which was used for transient expression assays. The VvMYBF1 ORF from Chardonnay was introduced into pDRIVE (Qiagen) and sequenced (accession no. GQ423422).

Transient Expression Experiments Using Particle Gun Bombardment

A transient expression system using Chardonnay suspension culture cells was established as described previously (Bogs et al., 2007; Walker et al., 2007). Cells were bombarded with 1.6-μm gold particles (Bio-Rad) using the model PDS-1000/He Biolistic Particle Delivery System from Bio-Rad with 4,481-kPa helium pressure, a vacuum of 86 kPa, and a distance of 9.5 cm. Gold particles were coated with 500 ng of each respective plasmid, giving a total amount of 2 μg of plasmid per transformation for reporter/effector bombardments. Transformations containing lesser plasmid amounts than 2 μg were filled up with empty pART7 plasmid to 2 μg to ensure equal coating of gold particles for each experiment. Additionally, each bombardment contained a positive control of 100 ng of the renilla luciferase plasmid pRluc (Horstmann et al., 2004). Bombarded cells were harvested and ground 48 h after transfection on ice in 300 μL of 1.5× Passive Lysis buffer (Promega). After centrifugation of the lysates for 1 min at 10,000 rpm, measurement of the luciferase activities was performed according to the dual-luciferase reporter assay system (Promega) by sequential addition of 25 μL of LARII and Stop & Glo to 10 μL of the lysate supernatant. Light emission was measured with a Lumat LB 9507 Luminometer (Berthold Technologies), and the relative luciferase activity was calculated as the ratio between the firefly and the renilla (control) luciferase activities after subtraction of the cell background (ground cells that were not bombarded). All transfection experiments were carried out in triplicate, and each set of promoter experiments was repeated at least two times, with each column representing the mean value and error bars indicating ± se.

Transfection Experiments of At7 Protoplasts

Protoplast isolation, transfection experiments, and the creation of the Arabidopsis flavonoid biosynthetic enzyme reporter constructs have been described previously (Mehrtens et al., 2005; Stracke et al., 2007). In the experiments, 10 μg of reporter plasmid, 1 μg of effector plasmid, and 5 μg of the luciferase standardization plasmid were premixed and cotransfected, giving a total amount of 25 μg. Protoplasts were incubated for 20 h at 26°C in the dark before luciferase and GUS enzyme activities were measured. Specific GUS activity is given in pmol 4-methylumbelliferone per mg protein per min. Standardized specific GUS activity was calculated by multiplication of the specific GUS activity value with a correction factor derived from the ratio of the specific luciferase activity in the given sample to the mean specific luciferase activity (describing the transformation efficiency) of a set of six complete experiments.

Transformation of Arabidopsis with VvMYBF1

For plant transformation, the complete coding sequence of VvMYBF1 from Shiraz was amplified with a specific forward primer designed to introduce an NcoI restriction site (MYB12 bastaF [5′-ATACCATGGGGTTTGGAAATGGGGAGAG-3′], with restriction site labeled in boldface) and a reverse primer designed to introduce a SpeI restriction site (MYB12 bastaR [5′-CTAACTAGTGAGCAGAAGAAAATCAAGAAAGAAG-3′], with restriction site labeled in boldface) using pART7MYBF1 as template. The NcoI/SpeI fragment was ligated to the binary vector pNR007 to give pNR007MYBF1, which allows control of VvMYBF1 by the root-specific AtMYB12 promoter. Arabidopsis Col-0 and myb12 mutant plants were then transformed with pNR007MYBF1 binary vector using the floral dip method (Clough and Bent, 1998). T1 transgenic plants were selected on soil by 0.02% (v/v) BASTA (glufosinate) spraying. BASTA-resistant T1 seedlings were transferred again to soil and grown under short-day conditions in the greenhouse until stable rosettes were generated. Plants were then transferred to long-day conditions in the greenhouse, and eventually seeds were harvested.

DPBA Staining of Flavonoids

The technique of staining flavonoids in Arabidopsis seedlings is adapted from Sheahan and Rechnitz (1992) and provides a method to observe flavonoid accumulation in plant tissues. Immature Arabidopsis seedlings were surface sterilized and kept at 4°C for 2 d on a filter paper soaked with 3 μg mL⁻¹ of the herbicide norflurazon (Supelco), which blocks the synthesis of carotenoids, thereby bleaching the seedlings (Buer et al., 2007). After 2 d at 4°C, the seeds were transferred to a phytochamber with 14 h of light illumination per day and 160 μmol m⁻² s⁻¹ at 24°C. Five days after germination, the seedlings were transferred in a freshly prepared solution of 0.25% (w/v) diphenylboric acid 2-amino-ethyl ester (Naturstoffreagenz A; Roth) and 0.00375% (v/v) Triton X-100. Fluorescence was visualized with either an inverted epifluorescence microscope (Leica DM IRB) or a stereomicroscope (Leica MZ FL III) with a color camera (Leica DFC 320) and a 4′,6-diamino-phenylindole fluorescence filter with an excitation wavelength of 340 to 380 nm and an emission wavelength of 425 nm. Images were captured using the Leica Image Manager 50 software and handled with Adobe Photoshop version 8.0.1 without changing the color parameters.

Flavonol Analysis by HPLC

HPLC analysis was performed on total methanolic extracts of Col-0, myb12 and myb11-12-111 mutant lines, and complemented lines on a reverse-phase HPLC device (Kontron Instruments; 322 pump system/360 autosampler/335 HPLC detector) with a Symmetry C18 column (3.5 μm, 4.6 × 150 mm [WAT200632]; Waters) protected by a guard column. Total extracts were sampled 5 d after germination and frozen in liquid nitrogen, and using aliquots of 20 mg, flavonoids were extracted by adding 200 μL of 50% (v/v) methanol (HPLC grade) in water, sonification for 20 min in an ice-water bath, and centrifugation for 10 min at 13,000 rpm. A total of 200 μL of the supernatant was loaded for HPLC analysis. Separation was carried out with a binary gradient of solvent A (10% [v/v] formic acid in water) to solvent B (100% [v/v] methanol; HPLC grade). The gradient conditions were 0 min, 17% solvent B; 15 min, 35% solvent B; 40 min, 37% solvent B; 42 min, 100% solvent B; 50 min, 100% solvent B; 51 min, 17% solvent B; 58 min, 17% solvent B. The column was maintained at 40°C, and the flow rate was 1.0 mL min⁻¹. Data acquisition and processing were performed by Kroma System 2000 software (Kontron). For the analysis of flavonols from Arabidopsis, initially peaks were identified as flavonol peaks by comparison of chromatograms of Col-0, myb12, and myb11-12-111. Peaks found in myb11-12-111 were classified as sinapate glycosides or other derivatives according to Stracke et al. (2007), while peaks vanishing or decreasing in both triple and single mutants were classified as flavonols belonging to kaempferol and quercetin derivatives. Concentrations were calculated from calibration curves prepared from commercial standards and expressed as quercetin-3-O-glucoside equivalents for flavonols. All HPLC separation experiments were performed in triplicate and exhibited similar relative ratios to the respective untransformed controls.

Grapevine flavonols were identified by comparison of spectra and retention times with commercial standards and with known flavonol glycosides from the literature. Unknown flavonols were included in the quantitation for total flavonols but were calculated and expressed as quercetin-3-glucoside equivalents (Supplemental Fig. S2).

Sequence data from this article can be found in the GenBank/EMBL data libraries under accession numbers FJ948477, FJ948478, GQ423420, GQ423421, GQ423422, and GQ423423.

Supplemental Data

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

• Supplemental Figure S1. Sequence analysis of the VvFLS1 and VvMYBF1 promoters.

• Supplemental Figure S2. HPLC chromatograms of flavonol analysis.