Jasmonates and Na-orthovanadate promote resveratrol production in Vitis vinifera cv. Barbera cell cultures

Summary

• •

  Here the effect of jasmonic acid, methyljasmonate and Na-orthovanadate on the production of resveratrol was studied in Vitis vinifera cv. Barbera cell suspension cultures.

• •

  Na-orthovanadate at 0.1 mm and 1 mm concentration was efficient in promoting the production and/or accumulation and release in the culture medium of cis-resveratrol while trans-resveratrol levels were not affected by this treatment.

• •

  Methyljasmonate was highly effective in stimulating both trans- and cis-resveratrol endogenous accumulation, as well as their release into the culture medium. Cis-resveratrol was absent or detected in very low amounts in the controls. Jasmonic acid was less efficient than methyljasmonate in promoting endogenous resveratrol accumulation, but it stimulated the release in the culture medium especially of cis-resveratrol.

• •

  Gel analysis was performed on control and 10 µm MeJA treated cell suspensions. Results showed an up-regulation of the stilbene synthase demonstrating that MeJA stimulated the synthesis ex-novo of this protein.

Introduction

Resveratrol (trans-3,5,4′-trihydroxystilbene) is a low molecular weight, constitutive and inducible phytoalexin belonging to the stilbene family (Langcake & Pryce, 1977). Stilbenes play an important role in protecting plants against fungal infections (Dixon & Harrison, 1990; Hain et al., 1990) and constitute the main group of phytoalexins within the Vitaceae (Jeandet et al., 2002).

Although resveratrol synthesis has been reported in several plants such as peanut, lily, mulberries, eucalyptus, spruce and pine (Lanz et al., 1990; Fliegmann et al., 1992; Kodan et al., 2001), grapevine is the main source of this compound. In Vitis vinifera L., resveratrol is produced in leaves and berries in response to both biotic and abiotic stresses like UV-irradiation, fungal infections (Langcake & Pryce, 1976), heavy metal exposure (Adrian et al., 1996) and ozone treatment (Schubert et al., 1997).

In recent years, much attention has been devoted to the effect of stilbenes on human health. These molecules possess antioxidant properties that could provide protection against arteriosclerosis (Lin & Tsai, 1999) as well as an interesting cancer chemopreventive activity (Jang et al., 1997).

Due to its high levels in the grape skin, variable amounts of resveratrol are present in the commercial wines, especially the red ones. This presence has been claimed as a possible explanation for the so-called ‘French paradox’, that is the apparent ability of moderate consumption of red wine to reduce the risk of cardiovascular diseases (Kopp, 1998). The importance of grapevine as a crop plant make V. vinifera a good model system for studies on the improvement of the nutriceutical properties of food products.

Some authors have undertaken studies on V. vinifera cell cultures to investigate the factors that are able to induce and/or modify stilbenes biosynthesis, regulation and metabolism (Waffo-Teguo et al., 1998; Krisa et al., 1999; Decendit et al., 2002; Commun et al., 2003). One of the main targets is the optimisation of the in vitro production of stilbenes. Jasmonic acid (JA) and its more active derivative methyljasmonate (MeJA) (Staswick, 1995) are plant stress compounds that act as regulators of defence responses (Reymond & Farmer, 1998; Chung et al., 2003). The induction of secondary metabolite accumulation is an important stress response that depends on jasmonates as a regulatory signal (Gundlach et al., 1992; Blechert et al., 1995). Jasmonates, in fact, act through the coordinate activation of the expression of multiple biosynthetic genes (Memelink et al., 2001). In Vitis vinifera, Krisa et al. (1999) showed that cell cultures respond to MeJA with an increase in piceid (3,5,4′-trihydroxystilbene-3-O-β-glucoside) production when the elicitor is added at the beginning of the exponential growth phase.

Information about the effect of Na-orthovanadate in elicitating secondary metabolism in plant tissues or cell suspensions cultures is sparse. Na-orthovanadate, a well-known inhibitor of plasmalemma H ± ATPase, was shown to inhibit hypersensitive response necrosis elicited by bacteria and Erwinia amylovora and Pseudomonas syringae produced harpins (Gopalan et al., 1996). In the present study we used two biotic elicitors, JA and MeJA, and one abiotic elicitor, Na-orthovanadate, to promote the production of trans- and cis-resveratrol and their release in the culture medium in cell suspensions obtained from V. vinifera cv. Barbera petioles. Barbera cultivar was selected for its importance among red grapes in northwest Italy to be used for wine making.

Materials and Methods

Establishment of callus cultures

Callus tissues were obtained from leaf petioles of 10-yr-old Vitis vinifera L. cv. Barbera plants at the phenological stage of berries pea size (Vitis germplasm collection of the Department of Plant Production, University of Milan, Prof Attilio Scienza, located at Torrazza Coste, Lombardy, Pavia, Italy).

The petioles were surface sterilised with ethanol 70% (v/v) for 2 min and 10% (v/v) commercial bleach for 10 min, and finally rinsed three times with sterile distilled water. The petioles were then cut into 1-cm pieces and placed onto a solid Murashige & Skoog culture medium (MS medium; Murashige & Skoog, 1962) with 10 g l⁻¹ sucrose supplemented with 1 mg l⁻¹ benzylaminopurine (BA) and 0.1 mg l⁻¹ 2,4 dichlorophenoxyacetic acid (2,4-D) and adjusted at pH 5.6. After 3–4 wk, the different callus tissues were separated from the explants and cultured separately until used for the establishment of cell suspensions. Both explants and calli were maintained at 23°C under a 16/8 h light/dark cycle (irradiance 25 W m⁻²).

Establishment of cell suspension cultures

Callus pieces (about 1 g f. wt) were transferred into 100-ml flasks containing 10 ml liquid MS medium supplemented with 1 mg l⁻¹ BA, 0.1 mg l⁻¹ 2,4-D, capped with Magenta B-Cap (SIGMA, St. Louis, MO, USA) and placed in a rotary shaker (110 rpm) at 23°C under a 16/8 h light/dark cycle (irradiance 25 W m⁻²) and subcultured every 2 wk.

Elicitors treatment

At day 0, cell cultures were supplied with 0.1 mm and 1 mm Na-orthovanadate dissolved in water, compared with a control culture performed without any exogenous addition. When cell cultures were supplemented with 10 µm jasmonic acid or 10 µm methyljasmonate (dissolved in 100% ethanol), ethanol controls were obtained by adding 0.1% ethanol final concentration, the same ethanol amount present in the media containing the elicitors. Cells were harvested every 2 d by filtration through a miracloth, rapidly washed, weighed, frozen by liquid nitrogen and stored at −80°C until analysis. The culture medium was collected, then centrifuged at 3000 rpm for 10 min to remove cell residues. The pH value of the culture medium was measured and it was stored at −20°C until further use.

Determination of cell growth and viability

A cell growth time course, measured as cell number and grams of cell f. wt, was performed both for treated and control cells. A small aliquot of each sample was used for cell counting performed by a Nageotte chamber. Cell viability assay was performed by incubating the cells for 1 min in a culture medium containing 75 µg ml⁻¹ fluorescein diacetate for the selective labelling of living cells, as described by Darzynkiewicz et al. (1994). Fluorescence was observed with a fluorescence microscope (Nikon Eclipse 600, Nikon Instruments, S.p.A., Florence, Italy) using a specific filter (Nikon B-2 A) and cell viability was expressed as percentage of living cells. About 1 g f. wt of filtered cells was placed at 100°C for 48 h and weighed to determine the percentage of dry weight (g d. wt), which was about 5% of the f. wt.

Trans- and cis-resveratrol quantification

Stilbenes were extracted from the cells collected from three different 100-ml flasks. Cells (about 1.5 g f. wt) were homogenised with 5 ml of 95% (v/v) methanol. The homogenate was incubated for 1 h in the dark at room temperature in a rotatory shaker (about 100 rpm) and filtered through Whatman GF/B filters placed on a Büchner funnel connected to a vacuum pump. The extract was concentrated to an aqueous phase in a speed vacuum for 15 min at 40°C (Speed Vac PD1, Savant Instruments, Holdboork, NY, USA). Stilbenes were extracted from the aqueous phase by adding of 5 ml of 5% (w/v) sodium bicarbonate, 10 ml of ethyl acetate and energic vortexing for 1 min. The ethyl acetate phase was completely dried with a rotavapor (R-205, BÜCHI Labortechnik, Konstanz, Germany).

Stilbenes were extracted from 20 ml of tissue culture medium after addition of 10 ml 5% (w/v) sodium bicarbonate, 20 ml of ethyl acetate and energic vortexing for 1 min. The ethyl acetate phase was completely dried. The ethyl acetate residue in the cells and medium extractions was resuspended in 200 µl acetonitrile before being directly injected into the HPLC.

Cis- and trans-resveratrol were analysed by reverse-phase HPLC separation (Soleas et al., 1997) (Jasco, Großumstad, Germany; column Phenomex Luna C18(2), 5 µm particles 250 × 4.6 mm, Phenomenex, Torrence CA, USA; precolumn SecurityGuard Ea, Phenomenex) equipped with a UV detector. The solvent gradient was as follows: 0 min acetonitrile/0.02 m KH₂PO₄ (pH 3.0) (10/90 v/v); 5 min acetonitrile/0.02 m KH₂PO₄ (10/90 v/v); 8 min acetonitrile/0.02 m KH₂PO₄ (20/80 v/v); 10 min acetonitrile/0.02 m KH₂PO₄ (20/80 v/v); 13 min acetonitrile/0.02 m KH₂PO₄ (40/60 v/v); 25 min acetonitrile/0.02 m KH₂PO₄ (40/60 v/v); 27 min acetonitrile/0.02 m KH₂PO₄ (10/90 v/v). Trans- and cis-resveratrol were determined, respectively, at 306 and 285 nm. Cis-resveratrol standard was obtained after irradiation for 10 min with a 254 nm UV light of trans-resveratrol standard (SIGMA) with about 96% of conversion between the two isomers (Goldberg et al., 1995a). Experiments were repeated twice with similar results. All measurements were made in triplicate.

Protein extraction for 2D polyacrylamide gel electrophoresis separation and mass spectrometric analysis

Cells were collected at day 2 of culture, from 10 µm MeJA treated and ethanol control suspensions, ground in liquid nitrogen and precipitated with the same volume of 50% trichloroactetic acid (w/v) on ice for 1 h. After centrifugation for 45 min at 4°C at 23 700g (Beckman J2-HS centrifuge, rotor JA 20) the pellets were washed with 3 volumes of 100% acetone three times as above. After washing in 2 volumes of ethyl ether and centrifugation for 10 min at 4°C at 1000g (Beckman J2-HS centrifuge, rotor JA 20) the powder was air dried for 30 min and 30 mg of powder was resuspended in 1.3 ml of liquid phenol (pH 7.9) added with 1% (v/v) 2-mercaptoethanol. Liquid extracts were recovered by centrifugation at 20 000g at 4°C for 30 min, extracted twice with 50 mm Tris-HCl pH 8, 1% (w/v) SDS, and then precipitated with 5 volumes of acetone on ice for 2 h. Samples were centrifuged 45 min at 6°C at 20 000g, the pellet washed twice with 1.3 ml of acetone and once with 1 ml of diethyl ether, air dried for 10 min and resuspended in 400 µl of rehydration buffer containing 2% (w/v) 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS), 7 m urea, 2 m thiourea, 1% (v/v) IPG buffer pH 3–10 (Amersham Bioscience, Chalfont St. Giles, UK). After centrifugation samples were loaded onto 18 cm Immobiline DryStrip pH 3–10 NL (nonlinear) (Amersham Bioscience, Chalfont St. Giles, UK), overnight at 18°C, then focused using a 18 cm pHaser with chiller block (Genomic Solutions, Huntingdon, UK) at a total of 85 KVh over a period of 23–24 h at 20°C. Strips were then equilibrated in equilibration buffer (0.063 m Tris-acetate, 3.3% (w/v) SDS, 6 m Urea, 30% (v/v) glycerol, traces bromophenol blue) added with 0.8% DTT (w/v) for 30 min and then in equilibration buffer added with 2.5% (w/v) iodoacetamide for 30 min. Second dimension was run on 10% Duracryl gels in a Tris-tricine buffer (top running buffer: 0.2 m Tris base, 0.2 m tricine, 0.4% (w/v) SDS; bottom running buffer: 25 mm Tris-acetate), in an Investigator 2-D large format 5-gel electrophoresis tank (Genomic Solutions, Huntingdon, UK) at 20 W per gel, at 20°C, for 4–5 h, until the bromophenol blue dye front was within 2 cm from the bottom of the gel. 2-D gels were fixed in 40% (v/v) methanol, 10% (v/v) acetic acid 1 h, stained with Sypro Ruby fluorescent dye (Bio-Rad Laboratories, Hercules, CA, USA) for 12 h, and washed in 10% (v/v) methanol, 6% (v/v) acetic acid for 1 h. Gel images were recorded using a ProExpress multiwavelength fluoroimager (Perkin Elmer Lifesciences, Cambridge, UK) and analysed using the ProteomeWeaver 2.1 software from Definiens AG (Munich, Germany). Spots were picked with an ‘Investigator’ ProPic spot picker and trypsin digested with an ‘Investigator’ ProGest in gel digestion unit (both from Genomic Solutions, Huntingdon, UK) as described by Speicher et al. (2000).

Samples were submitted to the Institute of Food Research and John Innes Center Joint Proteomics Facility (Norwich, UK). The acidified digests were spotted directly onto a thin layer of matrix on a stainless steel target made by mixing three parts of a saturated solution of α-cyano-4-hydroxycinnamic acid (CCA) in acetone with one part of a 1 : 1 mixture of acetone : isopropanol containing 10 mg ml⁻¹ nitrocellulose. Digests were externally calibrated to yield data of better than 50 ppm mass accuracy. Analysis of peptide digests was carried out on a Reflex III MALDI-ToF (Bruker Ltd, Coventry, UK) with Scout 384 ion source using a nitrogen laser (λ = 337 nm) to desorb/ionise the matrix/analyte material from the sample substrate. Ions generated in this way were allowed to drift for the short delayed extraction time setting, before being accelerated by a potential of +25 kV. Spectra were acquired in reflectron mode. Peptide fingerprints were searched against the National Center for Biotechnology Information-non-redundant database, taxonomy Viridiplantae, using the searching algorithm Mascot (http://www.matrixscience.com).

Peptides generated from tryptic digestion were loaded at high flow rate onto a reverse-phase trapping column (0.3 mm i.d. × 1 mm, containing 5 µm C18 100 Å PepMap packing, LC Packings, the Netherlands) and eluted through a reverse-phase capillary column (75 µm i.d. × 150 mm column, containing Symmetry C18 300 Å packing, Waters Ltd, Elstree, UK) directly into the nano-electrospray ion source of a quadrupole time-of-flight mass spectrometer (Q-ToF2, Micromass UK Ltd, Manchester, UK).

Fragment ion spectra were searched using the MASCOT search tool (Matrix Science Ltd, London, UK) against a weekly updated copy of the SPTrEMBL (UNIPROT) database using appropriate parameters. Peptides that were not identified by database search were sequenced de novo using the PepSeq software (Micromass, Manchester, UK).

Results

Jasmonic acid and methyljasmonate treatment

The ability of jasmonates to induce many secondary metabolic processes makes these molecules useful elicitors for enhancing the in vitro production of several compounds of interest. However, their mechanism of action can lead to undesired side-effects, such as a general reduction of cell growth-rate.

V. vinifera cv. Barbera cell culture was supplied (day 0) with 10 µm jasmonic acid (JA) and 10 µm methyljasmonate (MeJA). As shown in Fig. 1(a), control cells displayed a linear growth phase between day 7 and day 11, which was followed by a stationary growth phase. A similar trend was found in the culture treated with 10 µm JA even though with lower g f. wt than the control. This pattern was confirmed by cell numbers (data not shown). When MeJA was supplied, cell growth was highly reduced and cells did not present any linear growth phase, although according to cell viability (Fig. 1b) there was not a massive cell lysis. In fact a decrease in cell viability could be detected only in the second part of the culture period, in which MeJA treated cells showed 25% less viability than the control. Treatment with JA did not appear to affect cell viability (Fig. 1b). The pH of the culture medium measured every 2 d was almost constant at pH 5.3 in the control, but decreased from pH 5.3 (days 0–2) to 4.7 (days 4–14) in both treated cultures (data not shown). At least for MeJA treated cultures, the decreasing pH value seemed to reflect the reduction of g f. wt and cell viability (Fig. 1).

Figure 1

Time-course of treated V. vinifera cv. Barbera cell suspension cultures. Data are expressed as gf. wt (a) and percent of viable cells (b). Cell viability was measured by selective labelling of living/dead cells with fluorescein diacetate and expressed as percent of living cells. Experiments were repeated twice with similar results. Data are the mean ± SD of one experiment with three replicates. Open square, ethanol control cells; solid circle, MeJA treated cells; solid triangle, JA treated cells.

Resveratrol is present in two isomeric forms, trans-resveratrol and cis-resveratrol, and their ratio both in grapevine and in wines can change (Goldberg et al., 1995b).

The endogenous accumulation and the release in the culture medium of both trans- and cis-resveratrol were measured in cell suspensions of V. vinifera cv. Barbera treated with 10 µm JA or MeJA.

Although the trans-resveratrol content in the treated cells was almost always higher than in the control, its production followed an irregular pattern. In fact two peaks of accumulation were found in the culture supplied with MeJA, respectively, at day 2 (3-fold increase with respect to control culture) and at day 7 (20-fold increase), while a single peak at day 7 (16-fold increase) was found in the JA-treated culture (Fig. 2a). The decrease in the trans-resveratrol synthesis and/or accumulation at the beginning of the exponential cell growth phase (Fig. 1, day 4 of treatment) agreed with the findings of Waffo-Teguo et al. (1996a), who hypothesised a coupling of stilbene production and cell growth, as already demonstrated for anthocyanins and condensed tannins (Decendit & Mérillion, 1996).

Figure 2

Trans-resveratrol endogenous accumulation (a) and release (b) in V. vinifera cv. Barbera cell suspension cultures treated with jasmonic acid and methyljasmonate. Experiments were repeated twice with similar results. Data are the mean ± SD of one experiment with three replicates. Open square, ethanol control cells; solid circle, MeJA treated cells; solid triangle, JA treated cells.

A similar pattern was found in the release of trans-resveratrol in the culture medium in response to the treatments (Fig. 2b). Starting from day 4, MeJA stimulated trans-resveratrol release, which remained up to 5-fold higher than in the control during all the culture period. At day 2, the release into the culture medium of total trans-resveratrol was about 60% in the control, 70% in the JA-treated and 30% in the MeJA-treated cultures. At day 7 of culture, this percentage remained constant for the control and MeJA-treated cells, while in the JA-supplied culture it decreased to 30%.

Cis-resveratrol was detected only in very low amounts in the cells of Barbera control culture (Fig. 3a). Nevertheless, when cells were treated with JA or MeJA, the cis-isomer became predominant, its average content being 1.5-fold higher than trans-resveratrol in treated cultures (2, 3). Two peaks of production were found in MeJA-treated cells (day 2 and day 7) and one peak in JA-treated cells (day 7), and in both cases the major production seemed to occur at day 7 of treatment, by contrast to trans-resveratrol.

Figure 3

Ci-resveratrol endogenous accumulation (a) and release (b) in V. vinifera cv. Barbera cell suspension cultures treated with jasmonic acid and methyljasmonate. Experiments were repeated twice with similar results. Data are the mean ± SD of one experiment with three replicates. Open square, ethanol control; cells; solid circle, MeJA treated cells; solid triangle, JA treated cells.

Jasmonate treatment also affected cis-resveratrol release (Fig. 3b). Although the level of cis-resveratrol was negligible in the control culture medium, jasmonates strongly increased the release after day 2 of culture. JA had a stronger effect than MeJA in stimulating the release of cis-resveratrol, with two peaks at day 4 and day 9 of treatment in which 65% and 54%, respectively, of the total cis-isomer was released in the medium (Fig. 3b).

Proteomic analysis of methyljasmonate-treated cell suspensions

Cell suspensions treated with 10 µm MeJA and relative ethanol control were collected at day 2 of culture, corresponding to the maximum resveratrol production (Fig. 2), and the total protein extract analysed on pH 3–10-nonlinear gels (Fig. 4); this procedure was repeated three times. On the whole, the intensity of 34 spots was consistently changed in all the three treated samples, compared with the relative controls (data not shown). These spots were picked and analysed by matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-ToF MS). Peptide mass fingerprints were searched against the National Center for Biotechnology Information nonredundant database, using the searching algorithm Mascot. Only 10 proteins were identified (Mowse score ≥ of 64) and are listed in Table 1. Among them only six were identified as Vitis vinifera proteins and four correspond to stilbene synthase 1. These four spots are among the most highly induced by treatment with MeJA (from 2.4 to 3.1 fold increase). Only two other spots are induced to higher level by the treatment (3.6 and 4.0 fold increase), but they have not been identified (data not shown). Three other spots matched proteins belonging to the stilbene or to the chalcone pathways, but with a low significance score (< 64) and were subsequently further analysed by quadrupole time-of flight mass spectrometry (Q-ToF MS), which provided amino acid sequences of peptides (Table 1). No peptide sequence was successfully obtained by Q-ToF MS.

Figure 4

Two dimensional pH 3–10-nonlinear gels of total protein extract from ethanol control and 10 µm MeJA treated Barbera cell suspensions harvested at day 2 of culture. The analyses were performed on triplicate gels. Spots 1, 2, 3, 4 Stilbene synthase 1 (Acc. N gi|1729956); 5 Glutamine synthetase cytosolic isozyme 2 (Acc. N gi|1707959); 6 Pathogenesis-related protein 10 (Acc. N gi|11182126); 7 Auxin-induced protein PCNT115 (Acc. N gi|728744); 8 Phosphoenolpyruvate carboxylase (Acc. N gi|18073820); 9 Dihydroflavonol-4-reductase (Acc. N gi|1706369); 10 ATP synthase alpha chain (Acc. N gi|5305369); 11 Phenylalanine ammonia lyase (Acc. N gi|170792); 12 Chalcone isomerase (Acc. N gi|28460789); 13 Cinnamoyl CoA reductase-like protein (Acc. N gi|21404418). MW, molecular weight markers; pI, isofocusing pH.

Table 1. MALDI-ToF identification of protein from control and 10 µm MeJA treated Barbera cell suspensions harvested at day 2 of culture

Spot number	Mowse score	Protein	Organism	Acc.N	Fold change of protein amount treated/control
1	69	Stilbene synthase 1 (EC 2.3.1.95)	Vitis vinifera	gi|1729956	+3.1 ± 0.4
2	71	Stilbene synthase 1 (EC 2.3.1.95)	Vitis vinifera	gi|1729956	+3.0 ± 0.8
3	75	Stilbene synthase 1 (EC 2.3.1.95)	Vitis vinifera	gi|1729956	+2.9 ± 0.4
4	70	Stilbene synthase 1 (EC 2.3.1.95)	Vitis vinifera	gi|1729956	+2.4 ± 0.3
5	116	Glutamine synthetase cytosolic isozyme 2 (EC 6.3.1.2)	Vitis vinifera	gi|1707959	+2.2 ± 0.1
6	202	Pathogenesis-related protein 10	Vitis vinifera	gi|11182126	+2.1 ± 0.2
7	66	Auxin-induced protein PCNT115.	N. tabacum	gi|728744	+2.0 ± 0.1
8	68	Phosphoenolpyruvate carboxylase (EC 4.1.1.31) (Fragment).	Ginkgo biloba	gi|18073820	−2.0 ± 0.7
9	68	Dihydroflavonol-4-reductase (EC 1.1.1.219)	C. chinensis	gi|1706369	−5.5 ± 0.1
10	121	ATP synthase alpha chain.	Vigna radiata	gi|5305369	−6.4 ± 0.2
11	*55	Phenylalanine ammonia lyase (Fragment) (EC 4.3.1.5)	T. aestivum	gi|170792	+2.0 ± 0.1
12	*47	Chalcone isomerase (EC 5.5.1.6)	Oryza sativa	gi|28460789	−2.2 ± 0.1
13	*50	Cinnamoyl CoA reductase-like protein (EC 1.2.1.44)	A. thaliana	gi|21404418	Not detectable in treated cells

• Three spots (*) gave as top hit a protein taking part to important steps of resveratrol and/or chalcone biosynthesis, but having a score lower that significance were further analysed by Q-ToF. It was not possible to obtain a peptide sequence from any of them. NCBI-nonredundant database searched, Taxonomy Viridiplantae, a Mowse score higher than 65 is significant. The fold change represent the mean ± SD from three different gels both for control and treated cells. Positive values (+) indicate a protein more abundant in the treated cells, negative values (–) indicate a protein less abundant in treated cells.

Na-orthovanadate treatment

The time course performed on Barbera cell suspension treated with 0.1 mm and 1 mm Na-orthovanadate displayed a similar trend with cell number and g f. wt to that observed in Fig. 1 for treatment with jasmonates. In particular 1 mm Na-orthovanadate reduced cell growth similarly to MeJA (data not shown). Treatment with 0.1 mm Na-orthovanadate did not affect viability, while 1 mm treated suspension showed a 40% reduction in viability at day 14 of culture (data not shown). The pH of the culture medium remained around pH 5.5 in the 0.1 mm Na-orthovanadate treated culture, while in the 1 mm treated culture it decreased to pH 4.8 from day 7 of treatment in agreement with the reduction of cell number and viability (data not shown).

Na-orthovanadate was not very effective in elicitating trans-resveratrol production. However, trans-resveratrol production in 1 mm Na-orthovanadate treated cells slightly increased at day 7 of culture (Fig. 5a). By comparison with the control culture, the treatments with both Na-orthovanadate concentrations did not stimulate trans-resveratrol release (Fig. 5b).

Figure 5

Trans-resveratrol endogenous accumulation (a) and release (b) in V. vinifera cv. Barbera cell suspension cultures treated with Na-orthovanadate. Experiments were repeated twice with similar results. Data are the mean ± SD of one experiment with three replicates. Open circle, control cells; solid square, 0.1 mm Na-orthovanadate treated cells; solid diamond, 1 mm Na-orthovanadate treated cells.

When cells were treated with 1 mm Na-orthovanadate the cis-isomer was released at the same level as the trans-resveratrol with a peak at day 4, while when cells were treated with 0.1 mm Na-orthovanadate two peaks of cis-resveratrol were detected, respectively, at day 2 and day 7 (Fig. 6a). No cis-resveratrol was detectable in the control culture. The treatment with 0.1 mm Na-orthovanadate induced 25% release of the cis-resveratrol at day 2, while at day 7 no release was detected suggesting an accumulation of this isomer (Fig. 6b). The increase of vanadate concentration up to 1 mm seemed to induce the release of 75% of cis-resveratrol at day 2; this decreased to 22% at day 4 of treatment. By comparison with JA and MeJA, Na-orthovanadate was the least effective elicitor of resveratrol synthesis and accumulation and for this reason no further analyses were performed on Na-orthovanadate treated cells.

Figure 6

Ci-resveratrol endogenous accumulation (a) and release (b) in V. vinifera cv. Barbera cell suspension cultures treated with of Na-orthovanadate. Experiments were repeated twice with similar results. Data are the mean ± SD of one experiment with three replicates. Open circle, control cells; solid square, 0.1 mm Na-orthovanadate treated cells; solid diamond, 1 mm Na-orthovanadate treated cells.

Discussion

Our results show that cis-resveratrol is present in very low amounts or is absent in the controls, while it increased in all the elicitated cell suspensions. In addition, in the majority of the treatments, the cis-form is predominant with respect to the trans-isomer. Even though a positive correlation between the decrease of the trans and the increase of cis-isomer is not demonstrable, the variation of cis-resveratrol level could be explained by the presence of a conjugation process to form trans- and cis-piceid as well as the probable release of cis-resveratrol from the glucoside cis-piceid (Vrhovsek et al., 1997).

We report, for the first time, data on the release in the culture medium of trans- and cis-resveratrol. Considering both endogenous accumulation and release (2, 3, 5, 6), the export of trans- and cis-resveratrol in all the treatments seems to follow a concentration gradient, in all the growth phases. Moreover, cell lysis phenomena and decrease of viability seem not to affect resveratrol synthesis and release (Fig. 1b).

Taking into consideration the total resveratrol amount (cis and trans in the cells and media) most of the resveratrol produced at day 2 seems to be no longer present in the cell suspensions at day 4. The high decrease of trans-resveratrol shown for example in MeJA and in 0.1 mm Na-orthovanadate at day 4 of treatment, seems counterbalanced only by a slight increase of the cis form. A possible interpretation of these results could be an oxidation due to a peroxidase (Morales et al., 1997) and/or glucosilation processes (Waterhouse & Lamuela-Raventos, 1994; Waffo-Teguo et al., 1996a,b).

The general increase in resveratrol in response to MeJA, seems to be by contrast to the observations of Krisa et al. (1999) in cell cultures of two strains of Cabernet-Sauvignon grapes. These authors did not observe any variation of resveratrol accumulation, but only of the glucoside form (piceid), and reported a negligible amount of the two isomers in the culture medium. A partial explanation of this discrepancy may be the different concentration of the elicitor, added at a later stage of the culture and the different sensibility of the cultivar used.

The evidences from the proteomic analysis of MeJA treated cells support the hypothesis of a new synthesis of resveratrol at day 2 of culture due to an increase in the amount of stilbene synthase. In fact, as reported in Table 1 the MALDI-ToF results showed an up-regulation of the stilbene synthase. In Vitis (cv. Optima), the elicitor-dependent expression of stilbene synthase was already reported by Melchior & Kindl (1991) in response to fungal cell wall treatments. Special attention has to be given to the flavonoids and quercetin pathways, which take origin from chalcone via chalcone synthase (Fig. 7). The dihydroflavonol-4-reductase was down-regulated (Table 1) in treated cells leading to hypothesis of the prevalent use of 4-hydroxy-cinnamoyl-CoA, intermediate, common to both resveratrol and chalcone pathways (Fig. 7) (Schröder, 1999), by stilbene synthase for resveratrol synthesis (Fig. 7). Furthermore, four different isoforms of stilbene synthase, probably differing for the degree of phosphorilation or other minor post-translational modifications (same MW but different pI, Fig. 4, spots 1–4), all up-regulated by MeJA treatment (Table 1), were found to be expressed in Barbera cell suspensions. Although several gene isoforms have been reported for this enzyme (Melchior & Kindl, 1991; Wiese et al., 1994), all these four isoforms correspond to stilbene synthase 1 (Acc. N gi|1729956, Table 1), which seems to be the only one induced by MeJA treatment, at least at a level detectable by 2-D gel analysis. Finally, MALDI-ToF analysis also pointed out the up-regulation of a pathogenesis-related protein type 10 (Table 1), which could be part of the defence-related proteins synthesised, together with phytolaexins (Derckel et al., 1999), during pathogen attack response in which jasmonates seem to play a central role as second messengers (Reymond & Farmer, 1998; Memelink et al., 2001). A similar role for proteins belonging to the PR10 family has been suggested by Rakwal and colleagues, who demonstrated an increased expression of mRNA coding for this proteins by jasmonate treatment in 2-wk-old rice seedlings (Rakwal et al., 2001; Jwa et al., 2001).

Figure 7

Schematic pathway of resveratrol and chalcone biosynthesis and effect of MeJA treatment. (*) protein giving a score lower that significance with MALDI-ToF analisys.

In conclusion, jasmonates, and in particular MeJA, represent suitable elicitors for enhancing the production of cis- and trans-resveratrol in Barbera cells in liquid culture, and furthermore their release in the culture medium, while Na-orthovanadate mainly induced the synthesis and/or accumulation only of the cis-form. On a future perspective, the treatment of grapevine cell cultures with these elicitors could have important biotechnological application.

Acknowledgements

This work was supported by the financed project ‘Varietal and intravarietal characterisation of Vitis vinifera cultivars and their valorisation through increments of resveratrol and its derivatives’, 40% funds from Ministry of University and Scientific and Technological Research (MURST) of Italy to NB. AT was supported by a ‘Women in Science’ grant from L’Oreal Paris, Italy. MF was supported by a 1-yr Assegno di Ricerca from Università di Bologna. We wish to thank Dr Osvaldo Failla (Department of Plant Production, University of Milan, Italy) for kindly supplying the Vitis samples and Dr Cristina Pagnucco (Department of Biology, University of Bologna, Italy) for technical support in HPLC analyses. The Joint IFR-JIC Proteomics Facility was funded in part by: BBSRC JREI grant numbers JRE10832, JE412701, JE412631 and grants from Syngenta and Unilever.