VITIS VINIFERA CANES, A SOURCE OF STILBENOIDS AGAINST DOWNY MILDEW

Methods and results: Experiments on grape plants were carried out from 2007 to 2008 at commercial vineyards in « Le Tourne » (southwest region of France) on Vitis vinifera L. cv Merlot plants and in a greenhouse, where 45-day-old plants (Merlot) were used. Grapevine plants were treated with stilbene extract (STE) or « Bordeaux mixture » (BM) as positive control. STE, prepared from Vitis vinifera canes, contained 33% of total stilbenes ranging from monomers to tetramers and was used at 5g L-1. The degree of protection for downy mildew was determined as the percentage of infected leaf surface area and pathogen attack frequency. On young plants in greenhouse, P. viticola inoculation and three treatments with STE were carried out. After the treatments, the disease reduction was from 59 to 69% for pathogen attack frequency and from 83 to 88% for infected leaf surface for the two modes of infestation (artificial and secondary). STE triggered a similar intense decrease in downy mildew infection as recorded for BM. In the vineyard assay, downy mildew infection was allowed to occur naturally. Field grown plants were treated every 7th day for eight weeks or every 14th day for eight weeks during June and July. We observed that STE was also able to prevent the infection of P. viticola in vineyard but to a lesser extent. Treatment with STE every 7th day provided better results in terms of infection frequency and leaf area than that every 14th day. STE treatment of plants every 7th day in vineyard reduced the frequency of attack by 16 to 39% and leaf surface infection by 57 to 61%, which was lower than that recorded for BM. STE also displayed in vitro inhibitory effects on the release of zoospores from sporangia and the germination of zoospores of P. viticola. Zoospore germination was completely inhibited at a concentration of 100 mg L-1 and zoospore release was inhibited at 330 mg L-1.


INTRODUCTION
Downy mildew caused by an oomycete, Plasmopara viticola, is one of the most destructive diseases in vineyards. Protection against it involves several applications of fungicides between budburst/sprout and fruit-ripening depending on the disease severity and the climatic conditions of the year (Schnee et al., 2013). These pesticides have a harmful impact on the environment as well as on human health. Even copper, a natural micronutrient used in organic farming against downy mildew, may have long-term harmful consequences owing to its accumulation in the soil (La Torre et al., 2011). Thus, developing alternative strategies such as the use of natural products from plants that are biodegradable and usually less toxic has become a pressing need. Indeed, many secondary metabolites synthesized by plants to defend themselves against pathogens have antimicrobial properties. The alternative approach is to find elicitor compounds for grapevine defense stimulation, but only a few commercial products with low efficacy have been approved for this purpose (Aziz et al., 2003;Saigne-Soulard et al., 2015).
Stilbenes are a class of phenolic compounds that are abundant in Vitaceae, particularly in grapevines. Over 60 stilbenes have been identified in Vitis vinifera to date, presenting a wide structural diversity from monomers to hexamers, with resveratrol as the predominant monomer subunit Rivière et al., 2012). Several studies have shown the major involvement of these compounds in grapevine defense by acting as phytoalexins, mainly in leaves and berries (Douillet-Breuil et al., 1999;Adrian et al., 2000;Pezet et al., 2003). They can also be constitutively accumulated in stems and roots . For example, a positive correlation has been shown between the level of stilbenes such as resveratrol, pterostilbene and viniferins in grapevine leaves and berries and resistance to diseases such as gray mold, powdery mildew and downy mildew (Langcake, 1981;Bavaresco et al., 1997;Aziz et al., 2003;Pezet et al., 2004;Belhadj et al., 2006;Dufour et al., 2013). Alonso-Villaverde et al. (2011) showed that resistant grapevines react quickly to P. viticola infection by producing large concentrations of stilbenes at the exact sites of infection. Moreover, for the same cultivar, V. vinifera Cabernet Franc, some clones with a lower susceptibility for downy mildew produce a higher level of stilbenes in leaves (Van Leeuwen et al., 2013). Additionally, the antimicrobial activity of various stilbenes toward zoospore mobility and P. viticola sporulation has been assayed in vitro. The stilbenes exhibiting the most toxic activity were dand e-viniferins (dimers), vitisin B and hopeaphenol (tetramers), and methylated derivatives of resveratrol (pterostilbene and analogues) (Pezet et al., 2004;Schnee et al., 2013;Chalal et al., 2014).
Stilbenes are constitutively accumulated in grape canes at the high level of about 10 g kgDW -1 (Houillé et al., 2015). Pruning results in the production each year of a very large quantity of grape canes. For example, in France, 1.4 million tons of dry weight are available per year (FranceAgrimer, 2012). Therefore, this strategy could be used to control grapevine diseases by using stilbenes extracted from these grape canes.
The aim of this study was to investigate the antifungal activities of stilbene-rich grape cane extracts against P. viticola by conducting two in vitro tests (zoospore germination and sporulation) and two in vivo experiments on plants (in greenhouse and in vineyard).

Experimental sites and plant material
Experiments on grape plants were carried out in 2007 and 2008. Field trials were performed in commercial vineyards in « Le Tourne » (southwest region of France) on healthy Vitis vinifera L. cv Merlot plants. The grapevines (planted in 1970) were trained in a double Guyot system (two canes in opposite direction and two spurs) with a planting density of 2777 plants ha -1 . Greenhouse studies (SciencesAgro Atlantique, Loupes, France) were conducted on 45-day-old plants (Merlot) with 5 to 7 leaves.

Treatments in greenhouse
Six plants per treatment were sprayed entirely with the test material and each experiment was replicated in three sets (T1 = 17/10/2008). Lower surfaces of leaves of two plants were inoculated three days after the first treatment by spraying freshly prepared sporangia of P. viticola (40 000 mL -1 ). The second treatment was given 14 days later (T2 = 31/10/2008) followed by the third 12 days later (T3 = 12/11/2008). For the four plants without artificial infection, natural secondary infestation was obtained by overhead irrigation three days after the second and third treatments (T2, T3).
Disease intensity was evaluated visually by estimating percentage of infected leaf surface and pathogen attack frequency on the following dates: 04/11 for the two plants artificially infected; 14/11, 21/11, and 26/11/2008 for the four plants with secondary infestation. The application volume was about 600 L ha -1 . The control plants remained untreated.

Treatments in vineyard
In this experiment, downy mildew infestation was natural. Three plants per treatment were sprayed entirely and each experiment was repeated in quadruplicate (T'1 = 07/06/2007). Afterwards, seven other treatments were performed every 7 days or three other treatments every 14 days. Disease intensity was evaluated visually by estimating the percentage of infected leaf surface and pathogen attack frequency on the following dates: 16/07, 23/07, and 01/08/2007. The application volume was about 600 L ha -1 . The control plants remained untreated with reagents.
Before the experiment, all plants were protected by using Dithane Neotec

Stilbenoid extract activity on release of zoospores from sporangia and germination of P. viticola zoospores
STE was tested on sporangia (50 000 mL -1 ) using microplates at 21°C in the dark. After six hours, cotton blue was added to the wells and the numbers of germinated zoospores (viable) and empty sporangia (representing the release of zoospores) were determined using a microscope. STE was tested at different final concentrations, N (1 g L -1 ), N/3, N/10, N/30 and N/100, using acetone to ensure dissolution at a concentration of 1% (v/v). The experiment was repeated three times.

Statistical analysis
Statistical analyses were performed using the Newman-Keuls test and the Student t-test at P ≤0.05.

Protection of grapevine plants by stilbenoid extract (STE) against downy mildew
On young plants in greenhouse. Grapevine plants pretreated with STE or BM as positive control three days before P. viticola inoculation, followed by two subsequent treatments spaced 12-14 days apart, showed no phytotoxic effect. The degree of protection against downy mildew was recorded as the percentage of infected leaf surface area and pathogen attack frequency. The first observation (Nov 4) of two plants 15 days after inoculation (artificial infestation), i.e. 18 days after the first treatment and four days after the second treatment ( Fig. 2A and  2B), indicated a 38% reduction in pathogen attack frequency and a 79% reduction in infected leaf surface area. After the third treatment (Nov 12), three observations (Nov 14, 21 and 26) of four plants with secondary infestation showed a 59 to 69% reduction in pathogen attack frequency and a 83 to 88% reduction in infected leaf surface area. On young plants in greenhouse, STE triggered a similar intense decrease in downy mildew to that observed for BM ( Fig. 2A and 2B).
In vineyard. In vineyard assays, downy mildew infection was allowed to occur naturally. Grapevine plants treated with STE or BM from June 7 to July 26 (treatments every 7 th day) or to July 19 (treatments every 14 th day) showed no phytotoxic effect. The pathogen attack frequency was similar in control plants (~75%) throughout the observation period (Fig. 3A). Concerning the treatment with STE at the 7 th day, a low reduction of 39% was observed in attack frequency for the observation of Jul 16, 34% on Jul 23 and 16% on Aug 1. No significant decrease was observed for treatment with STE at the 14 th day.  STE or Bordeaux mixture (BM) was sprayed on plants at a concentration of 5 g L -1 and 7.5 g L -1 respectively on the following dates: 17/10, 31/10 and 12/11/2008. Observation Nov 4 concerns two plants with artificial infestation three days after the first treatment; observations Nov 14, Nov 21 and Nov 26 concern four plants with secondary infestation three days after the second and third treatments. Letters above columns indicate significant differences at P ≤0.05.

Figure 3 -Protection of grapevine plants (Merlot) by stilbenoid extract (STE) treatment against P. viticola (natural infestation) in vineyard: A. Pathogen attack frequency (%) and B. Infected leaf surface (%).
STE or Bordeaux mixture (BM) was sprayed on plants at a concentration of 5 g L -1 and 7.5 g L -1 respectively. From the first treatment (07/06/2007), seven other treatments were performed every 7 days or three treatments every 14 days. Letters above columns indicate significant differences at P ≤0.05.
However, BM treatment induced a reduction in attack frequency of 64-74% (treatment every 7 th day) and 51-54% (treatment every 14 th day).
With regard to infected leaf surface area (Fig. 3B), downy mildew progressed constantly in control plants from 8 to 21% between July 16 and August 1. In contrast, in grapevines treated with STE every 7 th day, the percentage of leaf surface infection by P. viticola progressed only from 3.4 to 8.4%, which represents a significant reduction of 57-61% for the three observations with this treatment. The reduction was 21-35% for treatment with STE at the 14 th day. With BM, reductions of 85-92% (treatment every 7 th day) and 75-83% (treatment every 14 th day) were obtained. In vineyard, STE triggered a less intense decrease in downy mildew than that observed with BM ( Fig. 3A and 3B).

Inhibitory effects of stilbenoid extract (STE) on release of zoospores from sporangia and germination of P. viticola zoospores
To investigate the mode of action of the stilbenoid extract, we studied various concentrations of STE from 10 mg L -1 to 1 g L -1 on the release of zoospores from sporangia and the germination of zoospores (Table 1). STE displayed a significant activity on both modalities from the concentration of N/30, which is equivalent to 33 mg L -1 . Zoospore germination was completely inhibited at the concentration of 100 mg L -1 and zoospore release was completely inhibited at 330 mg L -1 .

DISCUSSION
Treatment of young grapevine plants in greenhouse with STE led to considerable protection against P. viticola identical to that offered by Bordeaux mixture. After the treatments with STE, the disease reduction was from 59 to 69% for pathogen attack frequency and from 83 to 88% for infected leaf surface for the two modes of infestation (artificial and secondary). STE was also able to prevent the infection of P. viticola in vineyard but to a lesser extent. Treatment with STE every 7 th day led to better results than treating every 14 th day. STE treatment of plants every 7 th day in vineyard reduced attack frequency by 16 to 39% and leaf surface infection by 57 to 61%, which was lower than the values obtained with BM.
For these experiments, we used the concentration of 5 g L -1 on the basis of preliminary results obtained after a single treatment on young plants that showed a reduction in downy mildew correlated with STE concentration from 1 g L -1 (data not shown). No disease reduction was observed at a concentration lower than this concentration, although inhibition of zoospore germination was complete at a concentration of 100 mg L -1 and that of zoospore release at 330 mg L -1 . The chemical characterization of STE showed that it contains a variety of stilbenoids including monomers (resveratrol, piceatannol), dimers (e-and w-viniferins, ampelopsin A, pallidol), a trimer (miyabenol C), and tetramers (vitisin B, isohopeaphenol, hopeaphenol). Using in vitro antifungal assays, Schnee et al. (2013) also observed the inhibition of sporulation and the mobility of P. viticola zoospores with a methanolic extract of grapevine canes at a concentration of 1 g L -1 . To identify the stilbenes responsible for the antifungal activity, they fractionated the extract by solid phase extraction and obtained four fractions. All of them exhibited an activity against P. viticola at a concentration of 1 g L -1 , which indicated that many of these compounds were involved in the antifungal activity. A few compounds appeared particularly active in in  (Pezet et al., 2004;Schnee et al., 2013). Pterostilbene has been found to be the most toxic stilbene against various fungi, in particular P. viticola (Pezet et al., 2004;Chalal et al., 2014). However, to our knowledge, this has never been shown in grapevine canes . Some of these stilbenes which are very active against P. viticola, such as pterostilbene, dand e-viniferins, were also present in higher concentrations in clones of V. vinifera Cabernet Franc showing a lower susceptibility to downy mildew and in the resistant genotype V. riparia (Boso et al., 2012;Van Leeuwen et al., 2013).

CONCLUSION
Treatment of young grapevine plants in greenhouse with STE from grapevine canes offered considerable protection against P. viticola identical to that afforded by Bordeaux mixture. STE also prevented P. viticola infection in vineyard but to a lesser extent. Treatment with STE every 7 days gave better results than treatment every 14 days in vineyard.
Taken together, these findings indicate that stilbenes from grapevine canes are good candidates as natural fungicides against downy mildew. Nevertheless, the final product will require refinement to enhance its antifungal activity in the vineyard.