Cloning of a grapevine cDNA encoding a group IIc WRKY transcription factor
Degenerate oligonucleotide primers corresponding to the conserved core sequence of WRKY proteins (WRKYGQK) were used to screen a cDNA library prepared from grape berries harvested at veraison stage (V. vinifera L. cv. Cabernet Sauvignon). One of the overlapping sequences obtained after cloning and sequencing of the amplified products corresponded to a full-length cDNA of 849 bp and was designated VvWRKY1. It contains a 68 bp 5'-untranslated region (UTR) and a 298 bp 3' UTR with a poly(A) tail (Fig. 1). The predicted open reading frame encodes a protein of 151 residues with a predicted molecular mass of 17.7 kDa. Moreover, analysis of the VvWRKY1 protein sequence using the PSORT program (Nakai and Kanehisa, 1992, http://psort.nibb.ac.jp) revealed the presence of a putative nuclear localization signal, RKHR, between positions 56 and 59 of the amino acid sequence (Fig. 1).
WRKY proteins are divided into three main classes based on sequence similarity. Database searches showed that the predicted VvWRKY1 polypeptide is highly similar to group II WRKY proteins, containing only one WRKY domain at its N-terminal end followed by a Cys2/His2-type zinc-finger motif. Additional structural motifs allowed the distinction of different subgroups (Eulgem et al., 2000), and VvWRKY1 shows a high degree of similarity to the subgroup IIc (Fig. 2). VvWRKY1 is closely related to Arabidopsis thaliana AtWRKY75 (Eulgem et al., 2000) and Solanum tuberosum StWRKY1 (Dellagi et al., 2000) which share 59% and 53.9% amino acid sequence identities, respectively.
Otherwise, comparison of the VvWRKY1 sequence with expressed sequence tags (ESTs) published in the Grape TIGR database indicates that it corresponds to the TC50592 sequence identified from an abiotic stressed grapeberry library (V. vinifera L. cv. Chardonnay). The other two related TC sequences also come from abiotic stressed berry or leaf libraries (TC41321 and TC46341) and exhibit a very high sequence identity with VvWRKY1. The TIGR database does not contain any ESTs from pathogen-infected grapevine libraries.
VvWRKY1 binds to W-box elements
WRKY proteins have been identified as transcription factors that bind DNA sequences containing W-boxes (TGAC). In order to test VvWRKY1 protein–W-box DNA interactions, electrophoretic mobility shift assays (EMSAs) were performed. Production of recombinant VvWRKY1 in bacterial (Escherichia coli) and yeast (Saccharomyces cerevisiae) systems proved unsuccessful, most probably due to toxicity effects (data not shown). Therefore, the protein was synthesized by an in vitro transcription and translation assay. The biotinylated protein band of VvWRKY1 was detected by a chemiluminescent method (Fig. 3A).
In the absence of known grape target genes, EMSA was performed with three oligonucleotide probes previously shown to be bound by WRKY proteins in other plant species. These sequences contain W-boxes in different arrangements and with different surrounding sequences. Two sequences, W1 and NPR1, were used. The first one, W1, derived from the parsley PR1 promoter contains a single W-box TTGAC (
Fig. 3B;
Rushton et al., 1996). The second sequence is derived from the
A. thaliana NPR1 gene promoter (
Yu et al., 2001) and is composed of three W-box sequences (TTGAC) within 28 bp (two W-boxes in tandem and a third W-box in the reverse orientation;
Fig. 3B).
A shift band was observed with each probe incubated with the in vitro transcribed and translated protein (Fig. 3C). To verify VvWRKY1 binding specificity, competition experiments were conducted by adding unlabelled competitors in 200-fold molar excess to the binding assays. Competition with the unlabelled specific probe strongly reduced the binding signal, whereas addition of an unlabelled W-box-mutated probe (changing TTGAC to TTGAA) to the reaction volume had no effect. Similar binding specificities on the TTGAC sequence were seen with the promoter probes W1 and NPR1. These results demonstrate that the TTGAC core sequence is necessary for binding of VvWRKY1 protein to these DNA probes.
VvWRKY1 expression is developmentally regulated
Total RNA was isolated from various V. vinifera L. cv. Cabernet Sauvignon tissues and the expression of VvWRKY1 was investigated by semi-quantitative RT-PCR using primers designed within the 3' UTR (Fig. 4). VvWRKY1 was not expressed in roots but is expressed in fruit and leaves. In both of these latter organs, RT-PCR analysis revealed that expression of this transcription factor gene was developmentally controlled. The VvWRKY1 RNA level was very low in berries before veraison (about 60 daf), but it increased significantly after this stage and kept on accumulating throughout ripening (Fig. 4A). This increase appeared to be greater in flesh/skin than in seeds. In leaves, accumulation of VvWRKY1 transcripts was high in very young leaves and apices and in well-developed leaves (Fig. 4B). In contrast, in young leaves and in mature leaves, a lower signal was observed.
These results indicate that expression of
VvWRKY1 is not fruit specific and is strongly regulated during development in fruits and leaves.
VvWRKY1 expression is regulated by defence signals
The effect of wounding and some of the signal molecules known to trigger plant defence gene expression were investigated on the expression of VvWRKY1. Leaves of grapevine cuttings were wounded using scissors or sprayed with solutions containing SA, ethephon, or H2O2, and were collected for RNA isolation at the time points indicated. VvWRKY1 expression was analysed by semi-quantitative RT-PCR. The expression of ß-Glu encoding a pathogenesis-related protein (PR2) was monitored as a plant defence control. Two independent experiments were performed and gave similar results.
After wounding, the level of VvWRKY1 transcripts increased progressively from 2 h and reached a maximum at 8 h (Fig. 5A). This activation persisted 24 h post-treatment (hpt). The same pattern is observed for ß-Glu gene expression.
Furthermore, it was observed that all signal molecules used induced an increase of
VvWRKY1 and
ß-Glu expression from 2 hpt. The expression of
VvWRKY1 and
ß-Glu was weakly affected by water (
Fig. 5B), indicating that spraying itself did not affect gene expression. On the other hand, the strong transient induction of
ß-Glu expression observed 8 hpt for all compounds used confirms the efficiency of the treatments. The strongest induction of
VvWRKY1 expression was detected 8 h after spraying the plants with ethephon, an ethylene-releasing compound. After SA or H
2O
2 treatments, transcript accumulation commenced at 2 hpt and reached a maximum at 4 hpt and 8 hpt before decreasing at 24 hpt. In each case, the activation of
VvWRKY1 expression was transient, with
VvWRKY1 mRNA levels decreasing at 24 hpt. It is also interesting to note that
VvWRKY1 and
ß-Glu expression patterns appeared quite similar in response to the treatments performed in this study.
These results show that VvWRKY1 expression is positively affected by several defence signalling compounds and by wounding.
Ectopic expression of VvWRKY1 in tobacco results in reduced fungal susceptibility
The putative role of VvWRKY1 in plant defence was also addressed by overexpressing the cDNA under control of the 35S promoter of cauliflower mosaic virus in N. tabacum cv. Xanthi plants, since grapevine transformation still presents several difficulties. None of the generated transgenic plants showed any phenotypical changes compared with control plants. The expression level of the transgene was estimated for 10 independent lines by semi-quantitative RT-PCR using specific primers. Three lines were chosen on the basis of different accumulation levels of VvWRKY1 transcripts, two lines sharing a high level (lines 10 and 18) and one line (17) having a low level of transgene expression (Fig. 6A). Southern blot analysis revealed one copy of the transgene for line 10 and at least three copies for lines 17 and 18 (data not shown).
Because many WRKY transcription factors have been shown to be involved in the activation of defence gene transcription, and particularly of
PR genes (
Yu et al., 2001),
PR gene expression was investigated in the
VvWRKY1-overexpressing plants under normal growth conditions. Despite the study of three generations of transgenic plants, no clear correlation between the transgene expression level and the amount of transcripts encoding PR proteins (five classes of PR proteins have been tested: PR1, PR2, PR3, PR4, and PR8) could be established. Molecular analysis by semi-quantitative RT-PCR showed that expression of some
PR genes was slightly increased in only some lines compared with control plants (data not shown). These results suggest that VvWRKY1 ectopic expression does not act directly on the regulation of
PR gene expression.
Susceptibility tests were performed on T2 progeny of these three transgenic lines using several types of fungi and one virus, PVYN. Pathogenic fungi with particular agronomic interest for tobacco were chosen: two oomycetes, Pythium (causing plantlets damping off) and P. tabacina (the tobacco downy mildew agent), and an ascomycete, E. cichoracearum (the tobacco powdery mildew agent). Susceptibility was estimated for each disease at different time points post-infection (Figs 6B, 7). Results were submitted for statistical analyses (ANOVA, P-value their significance.
For tolerance to
Pythium, disease symptoms were estimated by visual scoring according to a severity scale at 7, 10, and 15 dpi. All transgenic lines showed reduced symptoms, with a 3-fold reduction of necrotic root surface in
VvWRKY1 plants compared with wild-type plants. Indeed, at 10 dpi, transgenic lines developed necrosis on 20% of total root surfaces (disease index 2) in comparison with 60% for wild-type plants (disease index 4). The three transgenic lines were also more tolerant to powdery mildew, the percentage of leaf surface covered by mycelium 10 and 13 dpi being reduced from 1.5-fold to 4-fold (line 18 at 13 dpi and line 17 at 10 dpi, respectively) compared with the wild type (
Fig. 6B). Concerning downy mildew infection experiments, two types of symptoms are shown in
Fig. 6B: the diameter of chlorotic surfaces at 6 dpi and the sporulation rate at 9 dpi. For each observation, the three transgenic lines were slightly but significantly less affected by this pathogen than the control tobacco. Necrotic symptoms were also observed at infection sites at a significantly higher level than in control plants (data not shown). As a consequence of the limited fungal growth and the presence of necrotic sites, fungal sporulation 9 dpi was significantly reduced in the transgenic plants. Interestingly, it can be noted that a weak level of
VvWRKY1 expression (line 17) was sufficient to induce a significantly better tolerance to theses fungal pathogens.
Disease resistance of the transformed tobacco plants against a virus, PVYN, was also studied. As shown in Fig. 7, no difference was observed between wild-type and transgenic plants.
Taken together, these data suggest the involvement of the VvWRKY1 transcription factor in activating defence mechanisms toward fungal pathogens in tobacco.
Answers to all your Biology Questions
