Biological control of the grapevine diseases grey mold and powdery mildew by Bacillus B27 and B29 strains

Transcription

1 Indian Journal of Experimental Biology Vol. 53, Febuary 2015, pp Biological control of the grapevine diseases grey mold and powdery mildew by Bacillus B27 and B29 strains Ben Maachia S 1, Errakhi Rafik 2, Chérif M 3, Preeti Nandal 4, Trupti Mohapatra 5 & Paul Bernard 6 * 1 Laboratoire d Horticulture, Centre Régional de recherche en Agriculture Oasienne de Degache, 2260, Dégache, Tozeur, Tunisia. 2 Equipe de Génomique Cellulaire et Techniques Moléculaires d'investigations, Centre d'innovation et de Transfert, Université Moulay Ismail, Marjane 2, Meknès. Marocco. 3 Laboratoire de Phytopathologie, Institut National Agronomique de Tunisie, 1082, Tunis, Tunisie. 4 Lignocellulose Biotechnology Laboratory, Department of Microbiology, University of Delhi South Campus (UDSC), Benito Juarez Road, New Delhi , India. 5 Proteomics and Molecular Biology of Abiotic Stress Response and Tolerance Laboratory, Institute of Life sciences, Bhubaneshwar, , Odisha, India. 6 Mycologie et de Phytopathologie, Institut Jules Guyot, Université de Bourgogne, B.P , Dijon, France. Received 02 October 2013; revised 11 June 2014 Uncinula necator and Botrytis cinerea are the most destructive pathogens of the grapevine in Tunisia and elsewhere. We used two strains of Bacillus subtilis group, B27 and B29 to control powdery mildew and the grey mold disease of the grapevine. Green house experiments showed that B29 and B27 strains of the bacteria efficiently reduced the severity of powdery mildew up to 50% and 60%, respectively. Further, they decreased Botrytis cinerea development on grape leaf by 77% and 99%, respectively. The mode of action has been shown to be chitinolytic. These two bacteria showed significant production of total proteins discharged into the culture medium. Determination of some chitinolytic enzymes revealed the involvement of N-acetyl glucosaminidase (Nagase), the chitin-1,4-chitobiosidase (Biase) and endochitinase in degrading the mycelium of B. cinerea. Keywords: Biase, Botrytis cinerea, Chitinolytic activity, Endochitinase, Lytic enzymes, Nagase, Uncinula necator, Wine diseases. Powdery mildew caused by Uncinula necator and grey mold caused by Botrytis cinerea are economically important diseases in most of the world s grape growing regions. These two pathogens cause severe yield losses. Current disease control methods are primarily based on the widespread use of chemical fungicides. However, exclusive use of fungicides causes poor disease management including pollution. Biological control using microorganisms or their secretions to prevent plant diseases, offers a green alternative or supplement to pesticides and genetic resistance for management of plant diseases. It facilitates sustainable agriculture 1. Bacillus are aerobic Gram Positive and saprophytic bacteria, forming endospores. They grow well in simple medium and produce hydrolytic enzymes, such as proteases, glucanases and chitinases, depending upon the culture medium 2. They are potential source *Correspondence: Tel: ; Fax: bernard.paul@u-bourgogne.fr of antimicrobial compounds, including peptide antibiotics and hydrolytic enzymes used as biocontrol agents 3. Bacillus subtilis is a safe and friendly microorganism used to suppress a wide range of plant pathogens. Many strains of Bacillus display antagonistic activity against a number of plant pathogens on several different crops 4. In this study, we investigated two soil borne strains of Bacillus subtilis, B27 and B29, for their antifungal activity against U. necator and B. cinerea, and also the production of some hydrolytic enzymes by these bacterial strains. Materials and Methods Microorganisms The microorganisms used (Uncinula necator and Botrytis cinerea) were isolated from naturally infected grapevine plants in the field. The isolate of B. cinerea was grown on potato dextrose agar (PDA) plates at room temperature (20 C). The pure fungal isolate of U. necator was maintained on grapevine grown in greenhouse within controlled conditions.

2 110 INDIAN J EXP BIOL, FEBRUARY 2015 Antagonistic bacteria The Bacillus strains, B27 and B29, from the phytopathology laboratory collection of Institut National Agronomique de Tunisie (Tunis), were used for the study. These strains are stored on nutrient agar medium at 4 C. The Bacillus strains were grown on nutrient agar medium at 28 C for 24 h, and single colonies were transferred to nutrient growth medium and incubated at 28 C for 24 h in an orbital shaker. Bacterial cultures were centrifuged at 3000 g for 10 min and the pellet was re-suspended in sterile distilled water to reach approximately 10 6 CFU/ml for further treatments. Evaluation of the Bacillus strains against U. necator on rooted plants of grapevine The two Bacillus strains were tested in the greenhouse for their ability to inhibit U. necator leaf colonization. Two controls have been considered, plants inoculated and non-treated; and plants inoculated and treated with the fungicide trifloxystrobin. The antifungal activities of these treatments against powdery mildew development were examined qualitatively and quantitatively. Cuttings were obtained from grapevine (Vitis vinifera L. cv. Carignan). Rooted plants were placed in greenhouse until they developed 2-4 leaves. These plants were placed in 10 liter pots (containing salt and soil mixture in 2:1 ratio) and four vines were placed in each pot. Plants were supplemented twice weekly with a 0.1%, 20/20/20 (N/P/K) fertilizer solution. The cultivar used in the experiment was selected for its susceptibility to the powdery mildew. The foliage of potted Carignan grapevine plants were treated with a suspension of each strain of bacteria. The concentration of inoculum for each isolate tested consisted of a propagules suspension prepared from pure cultures. The suspension was adjusted to a concentration of 10 8 propagules/ml in distilled water. All plants were sprayed to run-off using an atomizer to deliver the inoculum uniformly. A second treatment was applied 15 days later (8-10 leaf stage) with the same bacteria. Plants used as non treated control were sprayed with distilled water. Treated plants were allowed to air dry for approximately 1 h before being moved to an isolated compartment in the greenhouse for 24 h to allow establishment of the test microorganisms. The grapevine plants were inoculated by dusting conidia from U. necator colonies obtained from infected leaves. After inoculation, the pots were placed in a greenhouse at 25 C to allow establishment of fungus on the grapevine plants. The incidence and severity of the fungal infection were evaluated on foliage. The severity was estimated as the mass disease index (MDI) MDI= Xi*I/5 A scale was performed with 6 classes: i=0, 0%; i=1, 0-5%; i=2, 5-25%; i=3, 25-50%; i=4, 50-75%; and i=5, % area colonized by U. Necator, respectively. Xi was the number of leaves that have the index i. U. necator incidence was reported as the percentage of leaves infected. Evaluation of Bacillus strains against B. cinerea on the detached leaves of grapevine A leaf discs bioassay was performed for this purpose using the fully expanded, 6 day old leaves of Vitis vinifera L. cv. Cot Noir cuttings instead of seedlings. Detached grapevine leaves were surface sterilized by 30 s immersion in a 5% sodium hypochlorite solution and rinsed several times in sterile distilled water. Leaves were immersed for 30 min in an aqueous bacterial solution and drained for excess liquid before being placed in Petri dishes lined with moistened filter paper. Infection by the fungus (B. cinerea) was performed 48 h after treatment with the bacterial spore suspensions. The leaves were placed in controlled temperature (20 C) and photoperiod (16 h). The percentage (%) of necrotic leaf area was evaluated after 10 days of inoculation. Antagonism assay The dual culture technique was conducted to test antagonistic activity of the Bacillus sp. strains on the growth of B. cinerea. A 0.1 cm agar plug, from the margin of a growing fungal culture on a PDA plate, was incubated centrally on a fresh PDA plate (9 cm) for 72 h at 24 C. The bacterial test cultures were grown with potato dextrose broth (PDB), shaken at 150 g for 24 h at 30 C, centrifuged at 8000 g for 15 min and the pellet was streaked on the PDA plate 1 cm away from the fungal plug. Observations of fungal reactions were recorded after 2 days and the radius of the zones of inhibition were measured in two perpendicular directions by a vernier caliper. The percentage of hyphal growth inhibition was calculated using the formula: 100-[(radius of treatment2/radius of control2) 100] 5. Four replicates were used for each Bacillus isolate tested. Preparation of dried mycelium of B. cinerea Erlenmeyer flasks (250 ml) containing 100 ml of potato dextrose broth (PDB) were incubated with 1 cm 2 discs of potato dextrose agar (PDA) of actively

3 MAACHIA et al: BIOLOGICAL CONTROL OF GRAPEVINE DISEASES GREY MOLD & POWDERY MILDEW 111 growing mycelium of B. cinerea. The inoculated flasks were incubated at 20 C for 5 days. The mycelium was then collected by filtration through Whatman No.1 filter paper, washed with distilled water and used as C-source. Screening of Bacillus isolates for enzyme production Spore suspension of the two Bacillus strains ( spore/ml of culture medium) were used and inoculated into duplicate 100 ml flasks containing 20 ml of buffered mineral synthetic medium (MSM) supplemented with 1 g/l of the appropriate carbon source (dried mycelium, sucrose). The cultures were grown at 28 C for 5 days with 120 rpm. Culture filtrates were centrifuged at 4 C for 10 min at 5000 g and the clear supernatants were either immediately tested for enzyme activity or stored at 20 C until assayed. Enzyme assays The nomenclature for chitinolytic enzymes proposed by Harman et al. 6 were used for detection and assay of enzymes. N-acetyl-β- D-glucosaminidase (hereafter designated glucosaminidase) and chitin 1,4-β-chitobiosidase (hereafter designated chitobiosidase) activities were quantified by detecting the release of pigmented nitrophenol from p-nitophenyl-β-d-nacetylglucosaminide and p-nitrophenyl-β-d-n, N -diacetylchitobiose (Sigma Chemical Co., St. Louis, MO), respectively. Endochitinase activity was measured by the reduction of turbidity of a suspension of colloidal chitin. Glucan 1,3-β- Glucosidase activity was detected by measuring the release of reducing groups from a 0.1% solution of laminarin in 50 mm phosphate buffer, ph 6.7 (Sigma), as indicated by the reducing sugar assay described by Ashwell 7. Identification of bacteria DNA extraction Single isolated colonies of the two bacterial cultures were taken from agar plates and suspended in 500 µl of lysis buffer (400 mm Tris-HCl [ph 8.0], 60 mm EDTA [ph 8.0], 150 mm NaCl, 1% w/v sodium dodecyl sulfate). The tube kept at room temperature for 10 min was added with 150 µl of potassium acetate (ph 4.8), vortexed briefly and centrifuged at >10000 g for 1 min. After transferring the supernatant to a new Eppendorf tube, an equal volume of isopropyl alcohol was added, mixed by inversion briefly, centrifuged at >10000 g for 2 min, and the supernatant discarded. The resultant DNA pellet was washed in 300 µl of 70% ethanol. After spinning the pellet at rpm for 1 min, the supernatant was discarded. The DNA pellet was air dried and dissolved in 50 µl of sterile ultrapure water 8. PCR amplification PCR amplification of almost full length 16S rrna gene was carried out with specific primers 27F (5 -AGAGTTTGATCCT GGCTCAG-3 ) and 1492R (5 -GGTTACCTTGTTA CGACTT-3 ). A 50 µl reaction volume PCR was performed using approx. 100 ng of genomic DNA, 10X Reaction buffer, 10 mm (each) deoxynucleoside triphosphates, 25 mm MgCl 2 and 0.5 U of DNA polymerase. The PCR was performed in an automated Gene Amp PCR System thermal cycler under the following conditions. The amplification conditions were as follows: 98 C for 4 min (denaturation), 55 C for 1 min (annealing), 72 C for 3 min (elongation) and 72 C for 10 min final elongation. Expected PCR product of around 1.5 Kb was checked by electrophoresis of 5 µl of the PCR product on 1% agarose gel in 1X TBE buffer and stained with Ethidium bromide 0.5 µg/ml. The PCR products obtained were submitted to Genome Express (Grenoble, France) for sequence determination. The analysis of sequences was done at National Center For Biotechnology Information server ( All the analyzed sequences were deposited in Genbank. Results Effectiveness of Bacillus strains against U. necator on rooted plants of grapevine Experiments were conducted in the greenhouse to evaluate disease control efficacy of foliar biological control against the powdery mildew. In two independent trials, the Bacillus bacteria strains (B27 and B29) provided significant levels of control (60% and 40%) compared to the pathogen only control. Application of the foliar biological control agents enhanced the disease suppression significantly (Fig. 1A). B27 provided significantly higher disease reduction (50%) than that of B29 (40%; Fig. 1B). Effectiveness of Bacillus strains against B. cinerea on detached leaves of grapevine Effectiveness of antagonistic bacteria against grey mold, as examined by foliar symptoms, were more pronounced on the leaves inoculated with B. cinerea only. Very few necrotic symptoms were observed when the leaves were inoculated with the pathogen and the bacteria. A reduction in the percent infection, ranging from 95% to over 98% relative to the control, was recorded with B27 and B29 bacteria, respectively (Fig. 1C).

4 112 INDIAN J EXP BIOL, FEBRUARY 2015 Fig. 1 Effect of biological treatments: (A) on the percentage of attacked leaves of grapevine inoculated by Uncinula necator; (B) on MDI on leaves of vine inoculated by Uncinula necator; (C) on the percent leaf infestation 10 days after the inoculation by Botrytis cinerea. Bars are standard errors of means from five replicates. Fig. 2 Time course of protein production in different carbon source medium by B27 (A) and B29 (B) bacteria. Bars are standard errors of means from five replicates. Antagonism assay The two bacterial strains, B27 and B29, exhibited antifungal activity towards the polyphagous fungus B. cinerea. They produced clear inhibition zones with > 40 % inhibition of mycelial growth of the fungus on PDA plates. These Bacillus spp. produced antifungal substances. Enzyme assays The effect of culture medium on total protein production during cultivation of B27 and B29 are shown in Fig. 2. Total protein amount increased proportionally with the length of the incubation period. The highest value of proteins was achieved after 120 h for B27 (Figure 2A) and after 24 h for B29 (Figure 2B). Initially glucosidase production was significantly low. However, after 72 h for B27 (Fig. 3A) and 120 h for B29 (Fig. 3B) of growth, 1,3-β-glucosidase was produced at higher levels in sucrose containing medium than on B. cinerea cell wall containing medium. Glucosidase reached a maximum of 70 µg/ml and 100 µg/ml for B27 and B29, respectively, in 5 days of fermentation (P <0.05). Little activity was detected in media containing mycelium of B. cinerea as the sole carbon source. The type of carbon source used in the growth medium had a significant effect on the glucosaminidase production. The maximum activity Fig. 3 Time course of 1,3-β-Glucosidase produced by B27 (A); B29 (B) in different carbon source medium. Bars are standard errors of means from five replicates. Fig. 4 Time course of glucosaminidase produced by B27 and B29 in different carbon source medium. Bars are standard errors of means from five replicates. was achieved with mycelium of B. cinerea as carbon source after 120 h for both bacteria (180 µg for B27 and 80 µg for B29) enzyme production in sucrose was low but not totally repressed. Increased activity was observed in 96 h of fermentation for both bacteria (Fig. 4). A chitobiosidase activity was induced by the addition of mycelium of B. cinerea into the growth medium for B27. Synthesis of chitobiosidase increased rapidly after 6 h and remained till the end of the assay. Strain B29 incubated with the mycelium of Botrytis showed more than threefold higher levels of chitinase than that of B27 (Fig. 5).

5 MAACHIA et al: BIOLOGICAL CONTROL OF GRAPEVINE DISEASES GREY MOLD & POWDERY MILDEW 113 In the presence of sucrose as sole carbon source, the enzyme production by bacteria was low. It reached the maximum of 140 and 130 µg/ml respectively, for B27 and B29 after 120 h of growth. Fig. 5 Time course of chitobiosidase produced by B27 and B29 in different carbon source medium. Bars are standard errors of means from five replicates. The release of endochitinase by the two bacteria was determined by the measure of the degradation of turbidity of the medium containing colloidal chitin. The results showed a decrease in the medium color intensity. However, the rate of reduction was much higher when B29 was incubated in this medium. The color turbidity reached 2.5% and 3.5% for B27 and B29, respectively (Fig. 6). Molecular identification and phylogenetic analysis of bacteria Based on the sequence homology and phylogenic analysis, organism B27 was found to be Bacillus ( ), closely related to Bacillus sp. Z19 (EU ). For the Bacillus B29 strain ( ), it was closely related to Bacillus amyloliquefaciens strain EXWB3-03 (EU ) (Fig. 7). The partial sequence analysis of the 16S rdna gene and BLAST sequence comparison in the GenBank database showed an alignment of 1414/1415 bases (99% similarity) of the strain B27 (accession number ) with the sequence Fig. 6 Time course of turbidity degradation of medium containing colloidal chitin and B27 (A); and B29 (B). Bars are standard errors of means from three replicates. Fig. 7 Phylogenetic tree derived from parsimony analysis of 16S rdna gene from Bacillus sp. (A) Strain B27 (accession number ); (B) Strain B29 (accession number ). The 1415 basis determined for B27 were aligned to the corresponding regions from bacteria. The numbers indicate the levels of the bootstrap support based on neighbor-joining analyses of 1000 re-sampled data sets.

6 114 INDIAN J EXP BIOL, FEBRUARY 2015 of Bacillus subtilis strain PCL 1605 (accession number DQ ). Similar results were also obtained for Bacillus mojavensis strain BCRC (accession number DQ ). For the strain B29 (accession number ), the 16S RDNA gene partial sequence aligned with the sequence of Bacillus amyloliquefaciens strain TB2 (accession number EU ), though at the lower similarity level of 99% with an alignment of 1414/1415 bases and with the Bacillus subtilis strain Pab02 (accession number EU ). The phylogenetic trees are given in Fig. 7. Discussion The limitation of traditional approaches for control of grapevine diseases necessitates the exploration of alternative or complementary control methods such as biological control. However, the limited efficacy of available biological control agents 9 indicates that integration of different disease control approaches will be essential to achieve satisfactory disease suppression. In an attempt to develop an integrated biological control approach for powdery mildew and grey mold on grapevine, foliar biological control agents were applied as foliar spray. The two tested microorganisms applied to grapevine plants were effective in reducing the incidence and severity of powdery mildew under the experimental conditions considered. In the green house trial conducted the foliar biological control agent provided significantly greater suppression of powdery mildew than that provided by negative control. Our results indicate that these microorganisms are effective also against B. cinerea disease. We have demonstrated that B27 and B29 are antifungal strains. We have also shown the involvement of metabolites in the inhibition of B. cinerea. Among the metabolites, hydrolytic enzymes, such as chitinase and glucanase, are thought to be closely related to the mycoparasitism 10. 1,3-β-Glucosidase production by B27 and B29 was significantly influenced by carbon source incorporated into the medium. The enzyme production was significantly stimulated in the presence of sucrose with yields 100 fold higher than those produced on B. cinerea mycelium. The enzyme 1,3-β-glucosidase was proved to be carbon source dependent with sucrose inducing the production of the largest amount (Fig. 3). These results are in contrast with those obtained by Elad et al., in which 1,3-β-glucosidase activity from Trichoderma harzianum, grown on media supplemented with fungal cell walls, was higher than in media containing glucose 11. However, Saligkarias et al. 12 have proved that glucose induced the production of largest amount of enzyme followed by laminarin and cell wall pathogen. This finding suggests that 1,3-β-glucosidase may constitute a small part of the total protein secreted and has limited production in our bacteria. 1,3-β-Glucosidase might not be the only mode of action for these bacteria. Cell wall lytic enzymes, such as protease, may play a significant role, since fungal cell wall skeletal components are embedded in a protein matrix 13. Cell walls represent a substantial portion of the dry weight of fungal cells. In general, the cell wall of higher fungi are composed of linear polymers of β-1,4-n-acetyl glucosamine (chitin), α-1,3-glucans and proteins with extensive cross-linking between these components 14. These observations suggest that the hydrolytic enzymes produced by some Bacillus play an important role in destruction of plant pathogens 3. In this study, we have demonstrated that the two Bacillus exhibit an extracellular enzyme with glucosaminidase, chitobiosidase and endochitinases activities in medium containing mycelium of Botrytis. Most of the chitinolytic enzyme systems, reported in the literature are inducible 15. The most probable inducers of chitinase in B. subtilis strain are soluble oligomers derived from the chitin 3. In agreement with these findings, we found high chitinase activity only in cultures supplied with B. cinerea mycelium but less important with sucrose, which is further indicative of induction. Previous researchers have shown significant activity of the two bacteria, B27 and B29, against B. cinerea and U. necator on grape vine by the production of antifungal product 16 and also by the induction of defense mechanisms in plant (induction of phenolic compounds) 17. With the results it can be concluded that the two strains, B27 and B29 of Bacillus subtilis subgroup, are promising antagonists of the powdery mildew and grey mold pathogens, and thus can be used as effective biocontrol agents as an alternative to fungicides.

7 MAACHIA et al: BIOLOGICAL CONTROL OF GRAPEVINE DISEASES GREY MOLD & POWDERY MILDEW 115 References 1 San-Lang W, Shin IL, Wang CH, Tseng KC, Chang WT, Twu YK, Ro JJ & Wang CL, Production of antifungal compounds from chitin by Bacillus subtilis, Enzyme Microb Technol, 31 (2002) Perez LM, Polanco R, Rios J, Montealegre J, Valderrama L, Herrera R & Besoain X, The increase in endochitinases and β-1,3-glucanases in the mutant Th650-NG7 of the Trichoderma harzianum Th 650, improves the biocontrol activity on Rhizoctonia solani infecting tomato. IOBC/WPRS Bulletin, 30 (6) (2007) Leelasuphakul W, Sivanunsakula P, Phongpaichit S, Purification, characterization and synergistic activity of beta-1,3-glucanase and antibiotic extract from an antagonistic Bacillus subtilis NSRS against rice blast and sheath blight, Enzyme Microb Technol, 38 (2006) Collins DP, Jaconsen BJ, Maxwell B, Spatial and temporal population dynamics of a phyllosphere colonizing Bacillus subtilis biological control agent of sugar beet Cercospora leaf spot, Biol control, 28 (2003) Gamliel A, Katan J, Cohen E, Toxicity of chloronitrobenzenes to Fusarium oxysporum and Rhizoctonia solani as related to their structure, Phytoparasitica, 17 (1989) Harman GE, Hayes CK, Lorito M, Broadway RM, Di Pietro A, Peterbauer CK, Tronsmo A, Chitinolytic enzymes of Trichoderma harzianum: purification of chitobiosidase and endochitinase, Phytopathology 83 (1993) Ashwell G, Colorimetric analysis of sugars. In: Golowick SP, Kaplan NO (ed) Methods in enzymology, VIII. Academic Press Inc., Pub, New York (1957) Liu, X., Mann, DB, Suquet C, Springer DL, Smerdon MJ, Ultraviolet damage and nucleosome folding of the 5S ribosomal RNA gene. Biochemistry 39 (2000) Byrne JM, Dianese AC, Ji P, Camp Bell HL, Cuppels DA, Louws FJ, Miller SA, Jones JB, Wilson M, Biological control of bacterial spot of tomato under field conditions at several locations in North America, Biol control, 32 (2005) Uppal AK, El Hadrami A, Adam LR, Tenuta M, Daayf F, Biological control of potato Verticillium wilt under controlled and field conditions using selected bacterial antagonists and plant extracts, Biol control, 44 (2008) Elad Y, Chet IN, Henis Y, Degradation of plant pathogenic fungi by Trichoderma harzianum, Can J microbial, 28 (1982) Saligkatias ID, Gravanis FT, Epton HAS, Biological control of Botrytis cinerea on tomato plants by the use of epiphytic yeasts Candida guilliermondii strains 101 and US7 and Candida oleophila strain I-182: II. A study on mode of action, Biol control, 25 (2002) Peberdy JF, Fungal cell walls- A review. In: Kuhn PJ, Trinci APJ, Jung MJ, Goosey MW, Copping LG (ed) Biochemistry of Cell Walls and Membranes in Fungi. Springer, Berlin, (1989) Adams DJ, Fungal cell wall chitinases and glucanases. Microbiology, 150 (2004) El Katatny MH, Abdelzaher HMA, Shoul Kamy MA, Antagonistic actions of Pythium oligandrum and Trichoderma harzianum against phytopathogenic fungi (Fusarium oxysporum and Pythium ultimum var. ultimum), Archi Phytopathol Plant Protec, 39 (4) (2006) Ben Maachia S, Errakhi R, Mathieu F, Cchérif M, Lebrihi A, Identification and partial characterization of antifungal and antibacterial activities of two Bacillus sp. strains isolated from salt soil in Tunisia, Afr J Microbiol Res, 5 (13) (2011) Ben Maachia S, Omri N, Chebil S, El Hassni M, El Hadrami I, Chérif M, Bacillus induces phenolic compounds and enhances resistance to Uncinula necator infection in grapevine leaves, Afr J Plant Sci Biotechnol, 4 (Special issue 2) (2010) 46.