Characterization of resistance mechanisms activated by Trichoderma harzianum T39 and benzothiadiazole to downy mildew in different grapevine cultivars

Authors

  • H. Banani,

    1. Department of Sustainable Agro-ecosystems and Bioresources, Research and Innovation Centre, Fondazione Edmund Mach (FEM), San Michele all'Adige, Italy, Italy
    2. Mediterranean Agronomic Institute of Bari, Valenzano, Italy
    Current affiliation:
    1. Department of Agricultural Forestry and Food Sciences, University of Torino, Grugliasco, Italy
    Search for more papers by this author
    • These authors contributed equally to this work.
  • B. Roatti,

    1. Department of Sustainable Agro-ecosystems and Bioresources, Research and Innovation Centre, Fondazione Edmund Mach (FEM), San Michele all'Adige, Italy, Italy
    2. Institute of Integrative Biology, ETH Zürich, Zürich, Switzerland
    Search for more papers by this author
    • These authors contributed equally to this work.
  • B. Ezzahi,

    1. Department of Sustainable Agro-ecosystems and Bioresources, Research and Innovation Centre, Fondazione Edmund Mach (FEM), San Michele all'Adige, Italy, Italy
    2. Mediterranean Agronomic Institute of Bari, Valenzano, Italy
    Search for more papers by this author
  • O. Giovannini,

    1. Department of Sustainable Agro-ecosystems and Bioresources, Research and Innovation Centre, Fondazione Edmund Mach (FEM), San Michele all'Adige, Italy, Italy
    Search for more papers by this author
  • G. Gessler,

    1. Institute of Integrative Biology, ETH Zürich, Zürich, Switzerland
    Search for more papers by this author
  • I. Pertot,

    1. Department of Sustainable Agro-ecosystems and Bioresources, Research and Innovation Centre, Fondazione Edmund Mach (FEM), San Michele all'Adige, Italy, Italy
    Search for more papers by this author
  • M. Perazzolli

    Corresponding author
    1. Department of Sustainable Agro-ecosystems and Bioresources, Research and Innovation Centre, Fondazione Edmund Mach (FEM), San Michele all'Adige, Italy, Italy
    Search for more papers by this author

Abstract

Downy mildew, caused by Plasmopara viticola, is one of the most destructive diseases of grapevine and is controlled with intense application of chemical fungicides. Treatment with Trichoderma harzianum T39 (T39) or benzothiadiazole-7-carbothioic acid S-methyl ester (BTH) has been previously shown to activate grapevine resistance to downy mildew and reduce disease symptoms in the Pinot noir cultivar. However, enhancement of plant resistance can be affected by several factors, including plant genotype. In order to further extend the use of resistance inducers against downy mildew, the physiological and molecular properties of T39- and BTH-activated resistance in different cultivars of table and wine grapes were characterized under greenhouse conditions. T39 treatment reduced downy mildew symptoms, but the degree of efficacy differed significantly among grapevine cultivars. However, efficacy of BTH-activated resistance was consistently high in the different cultivars. Expression profiles of defence-related genes differed among cultivars in response to resistance inducers and to pathogen inoculation. T39 treatment enhanced the expression of defence-related genes in the responsive cultivars, before and after P. viticola inoculation. A positive correlation between the efficacy of T39 and the expression level of defence-related genes was found in Primitivo and Pinot noir plants, while different genes or more complex processes were probably activated in Sugraone and Negroamaro. The data reported here suggest that the use of a responsive cultivar is particularly important to maximize the efficacy of resistance inducers and new natural inducers should be explored for the less responsive cultivars.

Introduction

Grapevine (Vitis vinifera) is one of the world's major fruit crops, but most of the commercial cultivars used for table grape or wine production are susceptible to downy mildew. Grapevine downy mildew, caused by the oomycete Plasmopara viticola, is a devastating disease, particularly in warm and wet climates (Gessler et al., 2011). Plasmopara viticola attacks leaves and young berries and is controlled with frequent applications of chemical fungicides to avoid yield and quality losses. Concerns about the environmental impact of the overuse of pesticides have sparked interest in developing alternative methods to chemical treatments (Gessler et al., 2011). Therefore, a large number of researchers are currently engaged in identifying efficient biocontrol agents to limit downy mildew infection in susceptible cultivars (Harm et al., 2011).

Several molecules have been shown to increase resistance to downy mildew in susceptible grapevines, such as chitosan (Aziz et al., 2006), laminarin (Aziz et al., 2003), sulphated laminarin (Trouvelot et al., 2008; Steimetz et al., 2012), oligogalacturonide (Allègre et al., 2009), β-aminobutyric acid (BABA) (Hamiduzzaman et al., 2005), fosetyl-aluminium (Dercks & Creasy, 1989), thiamine (Boubakri et al., 2012) and benzothiadiazole-7-carbothioic acid S-methyl ester (BTH) (Perazzolli et al., 2008), as well as combinations of different resistance inducers (Reuveni et al., 2001; Pinto et al., 2012). Application of plant extracts or microbial agents can also induce resistance to downy mildew in grapevine, for example Rheum palmatum (Godard et al., 2009) and Solidago canadensis (Harm et al., 2011) plant extracts, organic amendments (Thuerig et al., 2011), the beneficial microorganisms Aureobasidium pullulans (Harm et al., 2011) and Trichoderma harzianum T39 (T39) (Perazzolli et al., 2008).

Enhancement of plant resistance exploits mechanisms of the plant immune system and two phenotypically similar forms of systemic immunity have so far been identified in plants: systemic acquired resistance (SAR) and induced systemic resistance (ISR) (Pieterse et al., 2009). SAR is controlled by salicylic acid (SA)-dependent signalling pathways and is activated systemically following pathogen recognition or treatment with some chemicals (Pieterse et al., 2009). ISR can be activated by specific strains of beneficial microorganisms and is usually regulated by jasmonic acid (JA) and ethylene (ET) signals (Pieterse et al., 2009). ISR is characterized by broad-spectrum activity against various types of pathogens and abiotic stresses and usually involves the activation of a priming state (Van der Ent et al., 2009). Primed plants display faster and/or more intense activation of the defence responses after pathogen inoculation, and this mechanism provides advantages in terms of energy costs under pathogen-free conditions, as defences are activated only when they are really needed (Conrath et al., 2006).

In grapevine, T39 and BTH significantly reduce downy mildew symptoms, both locally and systemically, without any direct toxic effects on P. viticola sporangia, indicating that the control mechanisms are mainly related to the activation of plant resistance (Perazzolli et al., 2008). Different mechanisms are activated by BTH and T39, making these resistance inducers useful tools for comparative characterization of resistance activation in grapevine (Perazzolli et al., 2011; Palmieri et al., 2012). BTH treatment directly induces the expression of SA-regulated pathogenesis-related (PR) genes (Perazzolli et al., 2011; Dufour et al., 2013), causing direct accumulation of reactive oxygen species (ROS) (Palmieri et al., 2012) and antimicrobial stilbenes (Dufour et al., 2013). T39 treatment produces a dual effect: it directly activates the microbial recognition machinery and enhances the expression of defence-related processes after downy mildew inoculation in the Pinot noir cultivar (Palmieri et al., 2012; Perazzolli et al., 2012). T39 treatment does not entail any apparent energy costs (Perazzolli et al., 2011), suggesting low risk to grape production. Activation of plant defences appears to be a promising low-impact tool for controlling crop diseases (Vallad & Goodman, 2004; Walters et al., 2012). However, this method is currently far from being used widely because resistance inducers rarely provide complete disease control and their effects are often inconsistent in field conditions (Harm et al., 2011; Walters et al., 2012, 2013). Because induced resistance is a host response, it is likely to be affected by the environment in field conditions (Thuerig et al., 2011; Walters et al., 2013) and by plant genotype (Reuveni et al., 2001; Sharma et al., 2010; Tucci et al., 2011; Walters et al., 2011). Greater understanding of these complex interactions is therefore required in order to maximize the efficacy of resistance inducers in the field (Walters et al., 2012, 2013).

In this study, the physiological and molecular mechanisms activated by T39 and BTH against downy mildew were investigated in different grapevine cultivars under controlled greenhouse conditions.

Materials and methods

Plant material

Rooted cuttings were obtained from seven wine grape cultivars: Pinot noir (PNR), Pinot gris (PGR), Chardonnay (CHR), Merlot (MER), Negroamaro (NEG), Primitivo (PRI) and Uva di Troia (UVA); and from seven table grape cultivars: Italia (ITA), Crimson seedless (CRM), Black magic (BLM), Sugraone (SUG), Victoria (VIC), Red Globe (RED) and Michele Palieri (MIC; Table 1). Cuttings were planted in 2·5 L pots (three cuttings in each pot) containing a mixture of peat and pumice (3:1) and grown for 2 months under greenhouse conditions at 25 ± 1°C with a photoperiod of 16 h of light and a relative humidity (RH) of 60 ± 10%, until each plant had one shoot with 5–10 leaves.

Table 1. Grapevine cultivars used in this study
Grapevine cultivaraCultivar abbreviationColour of berry skin
  1. a

    According to the Vitis International Variety Catalogue (www.vivc.de).

Wine grape
Pinot noirPNRBlack
Pinot grisPGRGrey
ChardonnayCHRWhite
MerlotMERBlack
NegroamaroNEGBlack
PrimitivoPRIBlack
Uva di TroiaUVABlack
Table grape
ItaliaITAWhite
Crimson seedlessCRMRed
Black magicBLMBlack
SugraoneSUGWhite
VictoriaVICWhite
Red GlobeREDRed
Michele PalieriMICBlack

Resistance induction in grapevine plants

Resistance activation was obtained using conidia of T39 and the commercial formulation of BTH (Bion, 50WG Syngenta Crop Protection). To obtain the conidia, T39 mycelium was grown for 2 weeks at 25°C in the dark on malt extract agar (Oxoid). Five small pieces (3 mm diameter) of fresh mycelium were used to inoculate twice-sterilized rice grains (prepared in Erlenmeyer flasks, containing 20 g rice and 55 mL water), which were then incubated at 25°C for 3 weeks to maximize the yield of viable conidia (Longa et al., 2009). T39 conidia were collected by washing the inoculated rice in cold (4°C) distilled water then filtering the conidial suspension with a fine net. The concentration of the conidial suspension was measured by counting with a haemocytometer under a light microscope. Aqueous suspensions of 1 × 107 T39 conidia per mL and 0·5 g L−1 BTH were used to induce resistance to P. viticola (Perazzolli et al., 2011). Control plants were treated with water (H2O-treated). The abaxial and adaxial surfaces of grapevine leaves were treated three times (1, 2 and 3 days before pathogen inoculation) in order to induce the greatest resistance activation (Perazzolli et al., 2008). For each treatment, about 2 mL per leaf were applied using a compressed-air hand sprayer.

Pathogen inoculation and assessment of disease severity

A P. viticola population was collected from an untreated vineyard in northern Italy (Trentino region) and maintained by subsequent inoculations on V. vinifera cv. Pinot noir plants under greenhouse conditions. To obtain fresh sporangia, infected plants with oil spot symptoms were incubated overnight in the dark at 99–100% RH and 25 ± 1°C to promote pathogen sporulation. Plasmopara viticola sporangia were then collected by washing the abaxial surfaces bearing freshly sporulating lesions with cold (4°C) distilled water. The concentration of the inoculum suspension was adjusted to 1 × 105 sporangia per mL by counting with a haemocytometer under a light microscope. One day after the last treatment with the resistance inducers, about 2 mL of the inoculum suspension were applied to the abaxial surface of each leaf using a compressed-air hand sprayer. Inoculated plants were incubated overnight in the dark at 99–100% RH and 25 ± 1°C and then kept under controlled greenhouse conditions. Six days after inoculation, the plants were incubated overnight in the dark at 25 ± 1°C with 99–100% RH to promote downy mildew sporulation. Disease severity was visually assessed as the percentage of abaxial leaf area covered by sporulation, according to standard guidelines of the European and Mediterranean Plant Protection Organization (EPPO, 2001). Treatment efficacy was calculated according to the following formula: (disease severity in control plants – disease severity in treated plants)/disease severity in control plants × 100. Four replicates (pots containing three plants each) per treatment were analysed in a randomized complete block design and two independent repetitions of the experiment were carried out.

Sample collection

Leaf samples were collected immediately before inoculation (uninoculated) and 24 h after P. viticola inoculation (inoculated) from H2O-, T39- and BTH-treated plants. This time point was chosen because it is associated with up-regulation of defence-related genes in Pinot noir plants (Perazzolli et al., 2011, 2012). For each treatment at each time point, leaf samples were collected from three different plants of each cultivar. A total of six plants per treatment were sampled (three before inoculation and three after inoculation) in order to avoid the effects of wounding stress. Each sample comprised two half leaves taken from the same plant; only leaves of the third and fourth node from the top of the shoot were collected in order to select responsive leaves (Steimetz et al., 2012) and to avoid effects of ontogenic resistance (Perazzolli et al., 2008; Thuerig et al., 2011). Samples were immediately frozen in liquid N2 and stored at −80°C.

Gene expression analysis by quantitative real-time RT-PCR

Four cultivars were selected (Pinot noir, Primitivo, Sugraone and Negroamaro) on the basis of the phenotypic analysis, and gene expression analysis was carried out according to Perazzolli et al. (2011). Briefly, total RNA was extracted using the Spectrum Plant Total RNA kit (Sigma-Aldrich) and quantified using the NanoDop 8000 (Thermo Fisher Scientific). Total RNA was treated with DNase I (Invitrogen) and the first strand cDNA was synthesized from 1·0 μg of total RNA using Superscript III (Invitrogen) and oligo-dT. Reactions were carried out with Platinum SYBR Green qPCR SuperMix-UDG (Invitrogen) and specific primers (Table 2) using the Light Cycler 480 (Roche Diagnostics). The PCR conditions were: 50°C for 2 min and 95°C for 2 min as initial steps, followed by 50 cycles of 95°C for 15 s and 60°C for 1 min. Each sample was examined in three technical replicates and dissociation curves were analysed to verify the specificity of each amplification reaction. To validate amplification of the target genes, real-time PCR products were sequenced on both strands using the Sanger method and amplified sequences of the different cultivars were aligned using clustalW2 (www.ebi.ac.uk/Tools/msa/clustalw2/) and compared with the corresponding grapevine cv. Pinot noir predicted gene (Velasco et al., 2007) of Release 3 (http://genomics.research.iasma.it/).

Table 2. Primer sequences for quantitative real-time RT-PCR expression analysis of grapevine genes
Gene nameGene codeaPrimer sequence (5′–3′)b 
  1. a

    The corresponding Pinot noir genes used for primer design and sequences alignments were identified on the Vitis vinifera cv. Pinot noir predicted genes (Velasco et al., 2007) of Release 3 (http://genomics.research.iasma.it/).

  2. b

    Primer pairs PR-1, PR-2, PR-4 and ACT: Polesani et al., 2010; Perazzolli et al., 2011, 2012.

Pathogenesis-related protein 1 (PR-1)fgenesh.VV78X169899.6_1

PR-1 forward

PR-1 reverse

ACTTGTGGGTGGGGGAGAA

TGTTGCATTGAACCCTAGCG

Pathogenesis-related protein 2 (PR-2)twinscan.VV78X005385.7_1

PR-2 forward

PR-2 reverse

GTTATTTCAGAGAGTGGTTGGC

AACATGGCAAACACGTAAGTCT

Pathogenesis-related protein 4 (PR-4)glimmer.VV78X053121.6_1

PR-4 forward

PR-4 reverse

CAGGCAACGGTGAGAATAGT

ACCACAGTCCACAAACTCGTA

Osmotin 1 (OSM-1)glimmer.VV78X132476.3_2

OSM-1 forward

OSM-1 reverse

CGCTGCGCTAAAGACTACC

AAAAACCTTGAGTAATCTGTAGCA

Actin glimmer.VV78X114914.6_2

ACT forward

ACT reverse

ATTCCTCACCATCATCAGCA

GACCCCCTCCTACTAAAACT

Cycle threshold (Ct) values were obtained by second derivative calculation using light cycler 480 SV 1.5.0 software (Roche), and reaction efficiencies were calculated with LinRegPCR software (Ruijter et al., 2009). The Actin gene was used as the constitutive gene for normalization because its expression was not significantly affected by the treatments (Perazzolli et al., 2012). The relative expression of each gene was calculated according to the Pfaffl equation (Pfaffl, 2001) using uninoculated H2O-treated Pinot noir plants as the calibrator. The Pinot noir cultivar was selected as the reference for data calibration, based on previous gene expression analysis (Perazzolli et al., 2011, 2012) and physiological characterization of T39- and BTH- activated resistance (Perazzolli et al., 2008; Palmieri et al., 2012). The priming effect was calculated as the ratio between the expression level in inoculated leaves of plants treated with the resistance inducer and the expression level in inoculated leaves of H2O-treated plants, using a threshold of 1·5-fold to identify augmented expressions (Verhagen et al., 2004). Mean expressions and standard errors of three biological replicates (plants) were calculated for each sample, and two independent repetitions of the experiment were analysed.

Statistical analysis

The results of the two independent experiments were analysed. Data were analysed using statistica v. 9 software (StatSoft). F-test revealed non-significant treatment × experiment interactions (> 0·05) so the data from the two experiments were pooled. Disease severity and efficacy scores were normalized (K–S test, > 0·05) by root square transformation, while fold-change values of gene expression analysis were transformed using the equation = log10 (1 + x) (Casagrande et al., 2011). Following validation of variance homogeneity (Levene's test, P > 0·05), an analysis of variance (anova) was carried out using Fisher's test to detect significant differences (< 0·05) among cultivars.

Results

Efficacy of induced resistance in the grapevine cultivars

The degree of downy mildew severity in 10 different V. vinifera cultivars was compared to the susceptible cv. Pinot noir (Fig. 1a), which was used as the reference for ISR characterization, based on previous results (Perazzolli et al., 2008, 2011, 2012; Palmieri et al., 2012). The disease severity was lower in Negroamaro and Crimson seedless plants than in Pinot noir and the highly susceptible Primitivo and Michele Palieri cultivars. Moreover, disease severity was lower in cv. Black magic than in the other V. vinifera cultivars, suggesting a higher level of basal resistance.

Figure 1.

Downy mildew severity in water-treated control plants (a), efficacy of Trichoderma harzianum T39 (b), and efficacy of benzothiadiazole (BTH) (c) on Pinot noir (PNR), Pinot gris (PGR), Chardonnay (CHR), Merlot (MER), Negroamaro (NEG), Primitivo (PRI), Uva di Troia (UVA), Italia (ITA), Crimson seedless (CRM), Black magic (BLM), Sugraone (SUG), Victoria (VIC), Red Globe (RED) and Michele Palieri (MIC) cultivars. Disease severity was assessed as percentage of abaxial leaf area covered by sporulation. Efficacy of T39 and BTH were calculated for each cultivar accordingly to the formula: (disease severity in control plants – disease severity in treated plants)/disease severity in control plants × 100. F-test revealed non-significant treatment × experiment interactions (> 0·05) and data from two repetitions of the experiment were pooled. The mean scores and standard errors of eight replicates (eight pots each containing three plants) pooled from the two experiments are presented for each cultivar. For each panel, different letters indicate significant differences among cultivars according to Fisher's test (< 0·05).

Treatment with T39 conidia significantly reduced downy mildew symptoms in cv. Pinot noir (Fig. 1b), in agreement with previous findings obtained with a commercial formulation of T39 (Perazzolli et al., 2008, 2011, 2012). Although T39-induced resistance reduced downy mildew symptoms in all cultivars tested, there were significant differences in the degree of efficacy. In particular, T39 efficacy was higher in Negroamaro and Black magic than in the reference cv. Pinot noir, but it was lower in Primitivo than in Pinot noir, suggesting that this mechanism is affected by the plant genotype. On the other hand, BTH treatment activated a consistently high level of resistance to downy mildew with no differences among the cultivars tested (Fig. 1c).

Molecular mechanisms of induced resistance in the grapevine cultivars

In order to analyse the molecular mechanisms involved in T39-induced resistance in grapevine cultivars, gene expression analysis was carried out on plants exhibiting lowest (Primitivo) and highest (Negroamaro) levels of T39 efficacy. Gene expression profiles were also analysed in a table grape cultivar (Sugraone) with a level of T39-efficacy comparable to the wine grape cultivar (Pinot noir) used as reference for ISR characterization (Perazzolli et al., 2011, 2012). The expression of four defence-related genes (Table 2) was analysed in the selected cultivars before and 24 h after P. viticola inoculation. Gene expression profiles were consistent in two independent repetitions of the experiment (F-test, > 0·05) and data were pooled. Analysis of sequences obtained from real-time RT-PCR products confirmed that the target genes were correctly amplified in each cultivar, with the exception of PR-4 in Sugraone. Alignments of primer sequences to the Pinot noir predicted genes suggested that all isoforms of each gene were amplified in real-time PCR.

PR-1 was induced by P. viticola inoculation in H2O-treated Pinot noir plants (Fig. 2a). PR-1 was directly induced by BTH treatment before inoculation, and its expression remained high at 24 h after P. viticola inoculation in BTH-treated plants. PR-1 was not induced in T39-treated plants, suggesting that T39-induced resistance could be mediated by SA-independent pathways (Perazzolli et al., 2011). In Primitivo, PR-1 expression was induced by P. viticola in H2O-treated plants and the P. viticola-dependent up-regulation of PR-1 was inhibited in T39-treated plants (Fig. 2b). As in Pinot noir, BTH treatment directly induced PR-1, which was not further affected by P. viticola inoculation. No significant up-regulation of PR-1 expression was observed in Sugraone (Fig. 2c) and Negroamaro (Fig. 2d) plants, with the exception of a P. viticola-dependent up-regulation in BTH-treated Sugraone. The level of PR-1 expression in uninoculated H2O-treated plants was fourfold greater in Negroamaro than in Pinot noir, suggesting a higher level of basal resistance.

Figure 2.

Expression of the defence gene encoding the pathogenesis-related protein 1 (PR-1) was analysed in leaf samples collected immediately before (white) and 24 h after (black) Plasmopara viticola inoculation of leaves of Pinot noir (a), Primitivo (b), Sugraone (c) and Negroamaro (d) plants treated with water (H2O), benzothiadiazole (BTH) or Trichoderma harzianum T39 (T39). Relative expression levels (fold-changes) were calculated using Actin as constitutive gene for normalization and data were calibrated on H2O-treated Pinot noir plants before inoculation. F-test revealed non-significant treatment × experiment interactions (> 0·05) and data from two repetitions of the experiment were pooled. Mean levels of relative expression and standard errors from six replicates (plants) pooled from the two experiments are presented for each treatment. Different letters indicate significant differences among treatments and cultivars, according to Fisher's test (< 0·05) on fold-change values transformed according to the equation = log10 (1 + x) (Casagrande et al., 2011).

PR-2 expression was induced by P. viticola inoculation in H2O-treated Pinot noir plants (Fig. 3a) and it was directly induced by BTH and T39 treatments before pathogen inoculation. PR-2 expression was enhanced after P. viticola inoculation in T39-treated plants, and the resultant level was 2·1-fold greater in T39-treated than in H2O-treated plants 24 h after pathogen inoculation (priming effect). No significant changes in PR-2 expression were observed in H2O- and BTH-treated Primitivo plants (Fig. 3b). The expression level of PR-2 was directly induced by T39, and it was threefold greater in T39-treated than in H2O-treated Primitivo plants 24 h after P. viticola inoculation. PR-2 was not affected by any of the treatments in Sugraone (Fig. 3c) and in Negroamaro (Fig. 3d) plants. The basal level of PR-2 expression in uninoculated H2O-treated plants was fourfold lower in Sugraone than in the other cultivars.

Figure 3.

Expression of the defence gene encoding the pathogenesis-related protein 2 (PR-2) was analysed in leaf samples collected immediately before (white) and 24 h after (black) Plasmopara viticola inoculation of leaves of Pinot noir (a), Primitivo (b), Sugraone (c) and Negroamaro (d) plants treated with water (H2O), benzothiadiazole (BTH) or Trichoderma harzianum T39 (T39). Relative expression levels (fold-changes) were calculated using Actin as constitutive gene for normalization and data were calibrated on H2O-treated Pinot noir plants before inoculation. F-test revealed non-significant treatment × experiment interactions (> 0·05) and data from two repetitions of the experiment were pooled. Mean levels of relative expression and standard errors from six replicates (plants) pooled from the two experiments are presented for each treatment. Different letters indicate significant differences among treatments and cultivars, according to Fisher's test (< 0·05) on fold-change values transformed according to the equation = log10 (1 + x) (Casagrande et al., 2011).

PR-4 was induced more than fourfold by P. viticola in H2O-treated Pinot noir plants (Fig. 4a). No direct induction and no priming effect of PR-4 were observed in BTH-treated plants. T39 treatment caused direct induction (2·5-fold) and priming effect (1·8-fold) of PR-4 expression, in agreement with previous results (Perazzolli et al., 2011). The basal level of PR-4 expression in uninoculated H2O-treated plants was sixfold lower in Primitivo (Fig. 4b) and 5000-fold lower in Negroamaro (Fig. 4c) than in Pinot noir plants. PR-4 was induced by P. viticola in H2O-treated Primitivo plants (inset of Fig. 4b). BTH treatment induced PR-4 before pathogen inoculation and similar expression levels of PR-4 were observed in inoculated BTH- and H2O-treated Primitivo plants. T39 treatment caused direct induction and priming effect of PR-4 in Primitivo plants. PR-4 was induced by P. viticola inoculation in H2O- and BTH-treated Negroamaro plants (inset of Fig. 4c). T39 treatment caused direct induction and no significant priming effect of PR-4 in Negroamaro plants. PR-4 expression was not detectable in Sugraone with the primer pairs designed on the Pinot noir sequence, probably as a result of the presence of polymorphisms in the gene sequences.

Figure 4.

Expression of the defence gene encoding the pathogenesis-related protein 4 (PR-4) was analysed in leaf samples collected immediately before (white) and 24 h after (black) Plasmopara viticola inoculation of leaves of Pinot noir (a), Primitivo (b) and Negroamaro (c) plants treated with water (H2O), benzothiadiazole (BTH) or Trichoderma harzianum T39 (T39). Expression of PR-4 was not detected in Sugraone. Relative expression levels (fold-changes) were calculated using Actin as constitutive gene for normalization and data were calibrated on H2O-treated Pinot noir plants before inoculation. F-test revealed non-significant treatment × experiment interactions (> 0·05) and data from two repetitions of the experiment were pooled. Mean levels of relative expression and standard errors from six replicates (plants) pooled from the two experiments are presented for each treatment. Different lower case letters indicate significant differences among treatments and cultivars, according to Fisher's test (< 0·05) on fold-change values transformed according to the equation = log10 (1 + x) (Casagrande et al., 2011). The insets represent fold-changes calibrated on the expression in leaves of H2O-treated Primitivo (b) and Negroamaro (c) plants before inoculation, respectively. For each inset, different upper case letters indicate significant differences among treatments within each cultivar, according to Fisher's test (< 0·05).

The osmotin 1 (OSM-1) gene, which belongs to the PR-5 family, was induced more than fourfold by P. viticola in H2O-treated Pinot noir plants (Fig. 5a). T39 treatment caused direct induction of OSM-1 before inoculation (5·5-fold) and priming effect (twofold) for enhanced expression after pathogen inoculation compared with inoculated H2O-treated plants. OSM-1 expression was similar in BTH-treated and T39-treated Pinot noir plants, both before and after P. viticola inoculation. OSM-1 was induced by P. viticola in H2O-, BTH- and T39-treated Primitivo plants (Fig. 5b). Unlike Pinot noir, the expression levels of OSM-1 were similar in T39-, BTH- and H2O-treated Primitivo plants after pathogen inoculation. OSM-1 was induced by P. viticola in Sugraone plants treated with BTH, while no significant changes were observed with the other treatments (Fig. 5c). OSM-1 was induced by Pviticola in H2O-treated Negroamaro plants, and it was not affected by T39 treatment (Fig. 5d). OSM-1 was directly induced (10-fold) by BTH treatment in Negroamaro plants and its expression was not further affected by P. viticola inoculation.

Figure 5.

Expression of the defence gene encoding osmotin 1 (OSM-1) was analysed in leaf samples collected immediately before (white) and 24 h after (black) Plasmopara viticola inoculation of leaves of Pinot noir (a), Primitivo (b), Sugraone (c) and Negroamaro (d) plants treated with water (H2O), benzothiadiazole (BTH) or Trichoderma harzianum T39 (T39). Relative expression levels (fold-changes) were calculated using Actin as constitutive gene for normalization and data were calibrated on H2O-treated Pinot noir plants before inoculation. F-test revealed non-significant treatment × experiment interactions (> 0·05) and data from two repetitions of the experiment were pooled. Mean levels of relative expression and standard errors from six replicates (plants) pooled from the two experiments are presented for each treatment. Different letters indicate significant differences among treatments and cultivars, according to Fisher's test (< 0·05) on fold-change values transformed according to the equation = log10 (1 + x) (Casagrande et al., 2011).

Discussion

Induced resistance is a broad-spectrum disease control method based on the plant's own defences (Vallad & Goodman, 2004; Van der Ent et al., 2009) and is a promising low-impact approach for the control of crop diseases (Vallad & Goodman, 2004). However, the full potential of induced resistance has yet to be realized, mainly because of its inconsistency in field conditions (Walters et al., 2012, 2013), suggesting a lack of knowledge on the mechanisms underlying resistance activation in plants. Besides optimal timing of treatments and integration with other protection programmes (Walters et al., 2012), the efficacy of different resistance inducers and the influence of the plant genotype should be clarified for each crop (Walters et al., 2013) in order to maximize the effects of this biocontrol method.

The biocontrol agent T39 reduces downy mildew severity in cv. Pinot noir (Perazzolli et al., 2008, 2011) by activating resistance through a complex reprogramming of the plant transcriptome (Perazzolli et al., 2012) and proteome (Palmieri et al., 2012). This study demonstrated that T39 is also effective in reducing the severity of downy mildew in other grapevine cultivars. However, different levels of T39 efficacy were observed, indicating a possible effect of the plant genotype in resistance activation. The highest levels of T39 efficacy were observed in the grapevine cultivars that were least susceptible to the disease (severity of disease in H2O-treated plants), such as Negroamaro and Black magic. Conversely, T39-induced resistance was least effective in the Primitivo cultivar, which exhibited the highest level of disease severity in H2O-treated plants. In contrast to T39, BTH activated a consistently high level of resistance with no differences among the grapevine cultivars. The different effects of T39 and BTH on grapevine cultivars could be related to the type of resistance inducer and/or the defence signals elicited in the plant. A negative correlation between the susceptibility of control plants and the efficacy of T39-induced resistance was observed, confirming that ISR is an enhancement of basal plant resistance (Pieterse et al., 2002). Plant genotype has previously been shown to influence the expression of induced resistance, and the relationship between basal and induced resistance seems to depend on the plant species and the resistance inducer. The highest level of induced resistance has been found in wild accessions of resistant bean (Córdova-Campos et al., 2012) and tobacco (Perez et al., 2003) and in partially resistant cucumber cultivars (Hijwegen & Verhaar, 1995). On the other hand, resistance activation was found to be greater in highly susceptible than in partially resistant soyabean (Dann et al., 1998) and tomato (Tucci et al., 2011) cultivars, but no relationship between levels of basal and induced resistance was observed in tomato (Sharma et al., 2010) and spring barley (Walters et al., 2011) genotypes.

In order to investigate the effect of the host genotype on T39-induced resistance, the transcriptional regulation of four defence marker genes in four grapevine cultivars was analysed. PR-1 and PR-4 genes were used as markers of SA- and JA-mediated responses, respectively (Hamiduzzaman et al., 2005). PR-2 and OSM-1 were used as markers of T39-induced resistance, on the basis of their direct induction and priming effect in T39-treated Pinot noir plants (Perazzolli et al., 2011, 2012). Whereas the differences in defence gene expression of resistant and susceptible Vitis genotypes have been widely investigated (Polesani et al., 2010; Casagrande et al., 2011; Figueiredo et al., 2012), this current study showed that different susceptible cultivars react differently to downy mildew inoculation. More specifically, PR-1 was induced by P. viticola in H2O-treated Pinot noir and Primitivo plants, whereas PR-2 was induced only in Pinot noir plants. PR-4 and OSM-1 were induced by P. viticola in Pinot noir, Primitivo and Negroamaro. Moreover, the expression levels of PR-2 and PR-4 in inoculated H2O-treated plants were greater in Pinot noir than in Primitivo, in agreement with the lower susceptibility to downy mildew.

Gene expression analysis showed that T39 has a dual effect on Pinot noir and Primitivo plants: it directly stimulates up-regulation of PR-2 and PR-4 and enhances expression of these genes after pathogen inoculation. Interestingly, the direct induction by T39 and the priming effect in T39-treated plants were less evident for PR-2, PR-4 and OSM-1 genes in Primitivo than in Pinot noir, in agreement with the lower efficacy of T39-induced resistance. However, different reactions to T39 were observed in the other two grapevine cultivars, Sugraone and Negroamaro, suggesting the involvement of more complex mechanisms or additional defence genes in induced resistance processes. There were less relevant changes in PR gene expression in Sugraone and Negroamaro compared with Primitivo and Pinot noir during T39-induced resistance, indicating a natural variation in the reaction to the beneficial microorganism. In particular, T39-induced resistance was most effective in Negroamaro, which did not exhibit transcriptional changes of all tested genes in T39-treated plants. Moreover, the cultivars of table (Sugraone) and wine (Pinot noir) grape showed comparable levels of T39 efficacy, but they differed significantly in the expression profiles of defence-related genes. In BTH-treated plants, the expression levels of PR-1, PR-2 and PR-4 were lower in Negroamaro and Sugraone than in Pinot noir, confirming that there are few transcriptional changes of PR genes in these cultivars. OSM-1 was strongly induced by BTH in Negroamaro and Pinot noir plants, suggesting that this gene plays a role in BTH-activated defence in both cultivars. These findings suggested that the mechanisms of induced resistance are affected by genetic factors of different grapevine cultivars. However, no absolute correlation was found between the expression of defence genes and the efficacy of induced resistance in grapevine. Similar results have been reported previously in other pathosystems, and no positive correlations between the efficacy of Trichoderma spp.-induced resistance and the expression of PR genes were observed in tomato cultivars (Tucci et al., 2011). Likewise, no direct correlations between the efficacy of BTH and the activity of a defence-related enzyme were found in spring barley (Walters et al., 2011).

In conclusion, this current study showed that induction of grapevine resistance can significantly reduce the severity of downy mildew symptoms in wine and table grape cultivars under controlled greenhouse conditions. The efficacy of T39-induced resistance differed significantly among the grapevine cultivars, suggesting the importance of genetic background in ensuring optimal results of this biocontrol method. Grapevine cultivars exhibited different molecular reactions to the pathogen and to the resistance inducers, suggesting that differences in the receptors or in some cellular components of the signalling network might influence the plant response. In general terms, no absolute relationship was seen between the efficacy of BTH or T39 and the expression levels of defence-related genes in grapevine. While PR-1, PR-2, PR-4 and OSM-1 are involved in resistance mechanisms, their expression profiles do not generally predict the efficacy of induced resistance. Analysis of PR gene expression partially helped in understanding the molecular processes that underpin the genetic control of the interactions between beneficial and pathogenic microorganisms in different grapevine cultivars. As observed in tomato lines (Tucci et al., 2011), the specificities of PR expression profiles revealed complex regulation of induced resistance mechanisms in the different varieties. Resistance is probably based on the activation of multiple genotype-dependent mechanisms in plants, and detailed knowledge of each pathosystem and each resistance inducer is needed for each crop in order to select the most responsive cultivars. Further studies are required to identify the major genetic determinants of Trichoderma spp. recognition and resistance induction in order to understand the cellular processes responsible for different reactions to the same stimulus. In practical terms, it was shown that resistance inducers, such as T39, should preferably be applied to highly responsive cultivars in order to maximize resistance activation. However, further studies are required to implement these findings in plant protection strategies under field conditions. There is also a need for new natural resistance inducers, which can activate higher levels of resistance in less responsive cultivars.

Acknowledgements

The authors dedicate this article to the memory of their colleague Vito Simeone. This research was supported by the EU project CO-FREE (theme KBBE.2011.1.2-06, grant agreement number 289497) and the Envirochange project funded by the Autonomous Province of Trento.

Ancillary