Deciphering Plant-Induced Responses toward Botrytis cinerea and Plasmopara viticola Attacks in Two Grapevine Cultivars Colonized by the Root Biocontrol Oomycete, Pythium oligandrum

Two major diseases that affect grapevine leaves and berries are controlled by the oomycete Pythium oligandrum. As the efficacy of biocontrol agents strongly depends on factors such as the trophic behaviors of pathogens and cultivar susceptibility, a two-disease approach was implemented to evaluate the activity of P. oligandrum against Botrytis cinerea (the necrotrophic fungus of gray mold) and Plasmopara viticola (the biotrophic oomycete of downy mildew) on two grapevine cultivars with different susceptibilities to these two pathogens. The results show that grapevine root inoculation with P. oligandrum significantly reduced P. viticola and B. cinerea infection on the leaves of the two cultivars, but with differences. This was observed when the relative expression of 10 genes was measured in response to each pathogen, and could be attributed to their lifestyles, i.e., biotrophic or necrotrophic, which are related to the activation of specific metabolic pathways of the plant. In response to P. viticola infection, genes from the jasmonate and ethylene pathways were mainly induced, whereas for B. cinerea, the genes induced were those of the ethylene–jasmonate pathway. The different levels of defense against B. cinerea and P. viticola could also explain the difference in cultivar susceptibility to these pathogens.


Introduction
Gray mold and downy mildew are major diseases that affect the berries and leaves of grapevines, causing serious economic losses in yield. Gray mold is caused by the necrotrophic plant pathogen Botrytis cinerea (Pers.), which kills grapevine cells before the colonization of plant tissues, while downy mildew is caused by the biotrophic oomycete Plasmopara viticola (Berk. & M. A. Curtis) Berl. & de Toni., which colonizes living plant cells [1]. Management of these diseases in vineyards mainly relies on preventive fungicide application, which may favor the development of pathogen-resistant strains [2], as well as the persistence of chemical residue in grapevine berries and wine [3]. In addition, due to the human health and environmental issues associated with the use of fungicides in vineyards, biocontrol, which is a presumed healthier method of controlling grapevine pathogens, has been developed. In this context, numerous bacterial and fungal potential biocontrol agents (BCAs) have been tested to control gray mold and downy mildew. As for bacteria, it has been reported that various strains of Bacillus, Pseudomonas, and Streptomyces significantly inhibit B. cinerea and P. viticola, both in vitro and in planta [4,5], by producing a wide variety of antifungal substances from their secondary metabolism, forming a biofilm that prevents the establishment of pathogenic fungi and induces plant defense responses [4,5]. mold), while the Cabernet Sauvignon cultivar is less susceptible to these pathogens. Onenode cuttings were placed in 350 mL pots containing Gramoflor Special soil (Gramoflor GmbH & Co., KG, Vechta, Germany) and grown in a greenhouse at a temperature of 24 • C/day and 20 • C/night, with 60% humidity and a photoperiod of 16 h of light. They were watered twice a day.

Microorganisms
The potential oomycete biocontrol agent P. oligandrum, strain Po37 [26], was isolated from grapevine roots. The oomycete inoculum was prepared by GreenCell (Saint Etienne Chomeil, France) according to the method previously described by Le Floch et al. [27]. The P. viticola isolate PV221 [28] used in this study was selected from the INRAE-UMR 1065 SAVE collection, Bordeaux. This strain was originally collected on V. vinifera in a commercial vineyard (Les Levès, France) in 2014. In order to multiply the P. viticola isolate, droplets (15 µL) of a spore suspension (10,000 sporangia/mL) were deposited onto the abaxial face of the Cabernet Sauvignon leaves. After seven days of incubation under controlled conditions (22 • C, RH > 90%, with a 16 h light photoperiod), an inoculum of P. viticola sporangia was prepared by washing the freshly sporulating lesions with sterile distilled water. The sporangial suspension was then adjusted to a concentration of 10,000 sporangia/mL using a Malassez hemocytometer.
The B. cinerea strain CCB16 used in this study was selected from the INRAE-UMR 1065 SAVE collection, Bordeaux. It was originally collected in 2016 from naturally infected grapevine berries from the Chateau Couhins vineyard, France. The strain was maintained on malt-agar medium (MA, 20 g/L of malt from Biokar Diagnostics and 15 g/L of agar from Biokar Diagnostics, Allonne, France) at 4 • C. In order to prepare the inoculum, the B. cinerea strain CCB16 was grown on MA medium, incubated at 22 • C and exposed to a 12 h photoperiod using white and near-UV (UV-A at 370 nm) light (Black Light UV-A, L18 w/73, OSRAM, Munich, Germany) for 15 days [29]. This B. cinerea strain was incubated for 15 days at 22 • C under ultraviolet (UV) light to stimulate the sporulation. Conidia were collected with 10 mL of sterile water as described previously by Aziz et al. [30]. The suspension obtained was filtered through sterile filter paper (11 µm) to remove the mycelium. The inoculum concentration was determined by using a Malassez hemocytometer and adjusted with sterile water to 5 × 10 5 conidia/mL.

Experimental Design and Plant Inoculations
For each cultivar, 60 plants were used per experiment. The plants were divided into three batches of 10 plants for the control and three batches of 10 plants for the biocontrol agent treatments. Three series of independent experiments were conducted. At the fiveto-six leaf stage, the grapevine plants were inoculated with the biocontrol agent using the following protocol: 50 mL of the oomycete inoculum was applied at the collar level of each plant. The inoculum concentration was adjusted to 2 × 10 4 oospores/mL. Fourteen days after P. oligandrum inoculation (14 dpi), the third or fourth leaves from the apex were collected (one leaf per plant).
The grapevine leaves were dried with filter paper, and then foliar discs of 2 cm in diameter were excised using a cork borer, as described previously by Dufour et al. [31], and deposited in Petri dishes containing Whatman paper moistened with 3 mL of sterile water. Eight discs were deposited per dish. For each cultivar, the experimental design consisted of foliar discs (i) infected with the pathogen, (ii) elicited with P. oligandrum and infected with the pathogen, (iii) elicited with P. oligandrum and non-infected with the pathogen, and (iv) non-elicited with P. oligandrum and non-infected with the pathogen. For the two last treatments, water droplets were deposited on the foliar discs.
On the same day (14 dpi), the plants roots were sampled to evaluate their colonization with P. oligandrum. For this, a plate-counting method was used, as described by Yacoub et al. [18] and Le Floch et al. [27]. Briefly, 20 root fragments per plant were sampled and deposited on the selective medium CMA-PARP (corn meal agar (a corn meal infusion of 2.0 g made with 7.0 mL of Tween 80 and 15.0 gm of Agar) with Pimaricin, Ampicillin, Rifampicin, and Pentachloronitrobenzene added) [32] and incubated at 25 • C in the dark for 10 days [18]. For each root fragment, the presence of typical P. oligandrum echinulated oospores was measured via an optical microscope. These echinulated oospores were identified based on Plaats-Niterink's (1981) book [33] containing the morphological descriptions of Pythium species. The average frequency of P. oligandrum oospores was determined for each treatment.
For foliar disc P. viticola infection, three droplets (15 µL) of the prepared sporangia suspension (10,000 sporangia/mL) were deposited on the abaxial side. Two Petri dishes were used per treatment. The foliar discs were incubated overnight, and then the residual water was removed. Finally, the discs were incubated for seven days under controlled conditions (22 • C, RH > 90% with a 16 h light photoperiod). For the two cultivars, downy mildew development was measured according to the density of the mycelium and sporulation [34,35] at five, six, and seven days after pathogen infection. At 7 dpi, the pathogen sporangia were automatically counted in the P. viticola and P. oligandrum + P. viticola treatments using a particle counter, as described by Delmas et al. [36].
Regarding B. cinerea infection, the foliar discs were wounded on the abaxial side using a needle before pathogen infection. Then, a 5 µL drop of the prepared conidia suspension of B. cinerea (5 × 10 5 conidia/mL) was deposited on the wound. For the control foliar discs, a water drop was used and five Petri dishes were used per treatment. The inoculated leaf discs were placed in a growth chamber for nine days at 25 • C with a 16 h light/8 h dark photoperiod. Gray mold development was evaluated by measuring the necrosis length three, five, six, seven, eight, and nine days after the pathogen infection [30,37]. As the overall leaf disc surface was necrosed, disease evaluation was stopped seven days after pathogen infection for the most sensible cultivar (Chardonnay).

RNA Extraction and RT-qPCR
In order to evaluate the grapevines' defenses induced by P. oligandrum, foliar discs were collected 48 h after pathogen inoculation. For each treatment, 15 discs were collected and regrouped on three independent biological replicates. The samples were stored at -8 • C for the RT-qPCR analyses.
The RNA extraction protocol was performed according to Reid et al. [38]. After being ground in liquid nitrogen, 1 mL of an extraction buffer (300 mM Tris HCl, pH 8.0; 25 mM EDTA, 2 mM NaCl, 2% CTAB, 2% polyvinylpolypyrrolidone (PVPP), 0.05% spermidine trihydrochloride, and 2% β-mercaptoethanol added extemporaneously) that had been preheated to 65 • C was added to 200 mg of leaf powder. The mixture was stirred vigorously and incubated in a water bath at 65 • C for 10 min with regular stirring. An equal volume of chloroform/isoamyl alcohol (24:1, v/v) was added and the mixture was centrifuged at 3500× g for 15 min (at 4 • C). The following RNA extraction steps were conducted using a Spectrum™ Plant Total RNA Kit, according to the manufacturer's instructions: The obtained RNA was reverse-transcribed to cDNA using M-MLV reverse transcriptase (Promega, Madison, WI, USA) following the manufacturer's instructions.

Quantitative Polymerase Chain Reaction
We evaluated the expression of 10 defense-related genes using RT-qPCR. These genes encode PR proteins (PR2, GLU, and PR5) or are involved in the jasmonic acid (AOS and JAR), ethylene biosynthetic (ACC and EIN3) and oxylipin (LOX2) pathways, as well as in phenylpropanoid-related genes (PAL and STS) [31,35]. Tubuline alpha (TUA) and TIP41like protein (TIP41) were used as the reference genes to calculate the transcripts' relative gene expression. PCR reactions were realized in a Stratagene Mx3005P qPCR thermocycler (Agilent Technologies, Santa Clara, CA, USA) with SYBR Green dsDNA binding die, 2x MESA BLUE qPCR MasterMix Plus for SYBR ® Assay Low ROX (Eurogentec, Seraing, Belgium). The data were analyzed using MxPro qPCR software (Agilent Technologies) to obtain the cycle quantification number (Cq) corresponding to the fluorescence signal of the amplified DNA intersected with the background noise for each sample/gene couple. The relative expression (RE) of genes was calculated using the 2 −∆∆Cq method, where ∆∆Cq was the ∆Cq between one sample and the control. The geometric mean of the reference genes was used as an accurate normalization factor to calculate the RE levels of the genes [39].

Statistical Analyses
Statistical analyses were performed using the R statistical software (version R 3.1.2). For the disease evaluation experiment, data were compared at each sampling time point using the non-parametric Kruskal-Wallis test (p < 0.05). For the RT-qPCR experiments, principal component analysis (PCA) was carried out using the RCMD package (version 2.6-2) and the plug-in FactoMiner (version 1.7) of R statistical software. For each gene, differential gene expression was subjected to statistical analyses using non-parametric multiple comparisons with the "nparcomp" package in R software, and significant differences compared to the untreated control were determined via Tukey's test (p < 0.05).

Assessment of Grapevine Root Colonization by P. oligandrum
As shown in Figure 1, overall, the root systems of the Cabernet Sauvignon and Chardonnay plants were well colonized with P. oligandrum 14 days after inoculation. The percentage of P. oligandrum root colonization tended to be higher in the Cabernet Sauvignon plants (62%) than in the Chardonnay plants (58%). P. oligandrum was not detected on the roots of the control plants of the two cultivars.
like protein (TIP41) were used as the reference genes to calculate the transcripts' relative gene expression. PCR reactions were realized in a Stratagene Mx3005P qPCR thermocycler (Agilent Technologies, Santa Clara, CA, USA) with SYBR Green dsDNA binding die, 2x MESA BLUE qPCR MasterMix Plus for SYBR ® Assay Low ROX (Eurogentec, Seraing, Belgium). The data were analyzed using MxPro qPCR software (Agilent Technologies) to obtain the cycle quantification number (Cq) corresponding to the fluorescence signal of the amplified DNA intersected with the background noise for each sample/gene couple. The relative expression (RE) of genes was calculated using the 2 −ΔΔCq method, where ΔΔCq was the ΔCq between one sample and the control. The geometric mean of the reference genes was used as an accurate normalization factor to calculate the RE levels of the genes [39].

Statistical Analyses
Statistical analyses were performed using the R statistical software (version R 3.1.2). For the disease evaluation experiment, data were compared at each sampling time point using the non-parametric Kruskal-Wallis test (p < 0.05). For the RT-qPCR experiments, principal component analysis (PCA) was carried out using the RCMD package (version 2.6-2) and the plug-in FactoMiner (version 1.7) of R statistical software. For each gene, differential gene expression was subjected to statistical analyses using non-parametric multiple comparisons with the "nparcomp" package in R software, and significant differences compared to the untreated control were determined via Tukey's test (p < 0.05).

Assessment of Grapevine Root Colonization by P. oligandrum
As shown in Figure 1, overall, the root systems of the Cabernet Sauvignon and Chardonnay plants were well colonized with P. oligandrum 14 days after inoculation. The percentage of P. oligandrum root colonization tended to be higher in the Cabernet Sauvignon plants (62%) than in the Chardonnay plants (58%). P. oligandrum was not detected on the roots of the control plants of the two cultivars.    Figure 2 shows that the growth of P. viticola in the plants pre-treated with P. oligandrum before pathogen inoculation was significantly reduced on the leaves of the Cabernet Sauvignon and Chardonnay plants, at all different sampling time points. Regarding the Chardonnay plants, P. oligandrum significantly reduced the pathogen growth percentage when compared with the corresponding controls at 5, 6, and 7 dpi, with inhibition levels ranging from 13% to 25%. The inhibition growth of P. viticola stayed at the same level over time ( Figure 2B). Figure 2 shows that the growth of P. viticola in the plants pre-treated with P. oligandrum before pathogen inoculation was significantly reduced on the leaves of the Cabernet Sauvignon and Chardonnay plants, at all different sampling time points. Regarding the Chardonnay plants, P. oligandrum significantly reduced the pathogen growth percentage when compared with the corresponding controls at 5, 6, and 7 dpi, with inhibition levels ranging from 13% to 25%. The inhibition growth of P. viticola stayed at the same level over time ( Figure 2B). On the Cabernet Sauvignon plants, the percentage of P. viticola growth leaf discs in the controls inoculated only with P. viticola reached 48%, 56%, and 72% at 5, 6, and 7 dpi, respectively. These values were significantly reduced to 30%, 36%, and 48% at 5, 6, and 7 dpi, respectively, on the leaf discs collected from the plants inoculated with P. oligandrum at the root level, before P. viticola inoculation (Figure 2A). The leaf discs exhibited no symptoms in the control or P. oligandrum-inoculated plants.
The quantification of P. viticola sporangia was evaluated at 7 dpi using a particle counter, and for the two cultivars, the sporulation of P. viticola was significantly reduced on leaf discs collected from plants pre-treated with P. oligandrum and then infected with P. viticola, in comparison to those of the controls, which were inoculated only with the pathogen (Figure 3). The observed reduction was more important on the Chardonnay leaf discs, which was approximately 40% on the Chardonnay leaves, while it was approximately 25% on the Cabernet Sauvignon leaves.  On the Cabernet Sauvignon plants, the percentage of P. viticola growth leaf discs in the controls inoculated only with P. viticola reached 48%, 56%, and 72% at 5, 6, and 7 dpi, respectively. These values were significantly reduced to 30%, 36%, and 48% at 5, 6, and 7 dpi, respectively, on the leaf discs collected from the plants inoculated with P. oligandrum at the root level, before P. viticola inoculation (Figure 2A). The leaf discs exhibited no symptoms in the control or P. oligandrum-inoculated plants.
The quantification of P. viticola sporangia was evaluated at 7 dpi using a particle counter, and for the two cultivars, the sporulation of P. viticola was significantly reduced on leaf discs collected from plants pre-treated with P. oligandrum and then infected with P. viticola, in comparison to those of the controls, which were inoculated only with the pathogen (Figure 3). The observed reduction was more important on the Chardonnay leaf discs, which was approximately 40% on the Chardonnay leaves, while it was approximately 25% on the Cabernet Sauvignon leaves.   Figure 4 shows the measurement of the necroses due to B. cinerea on the surface of the leaf discs of Cabernet Sauvignon and Chardonnay at different sampling time points. Figure 3. P. oligandrum reduced P. viticola sporulation in the leaf discs of the two grapevine cultivars: Cabernet Sauvignon (A) and Chardonnay (B) at 7 days post-inoculation by the pathogen. Pv and PoPv correspond to the P. viticola and P. oligandrum + P. viticola treatments, respectively. * Significantly different between the two treatments at p < 0.05 (Kruskal-Wallis test). Figure 4 shows the measurement of the necroses due to B. cinerea on the surface of the leaf discs of Cabernet Sauvignon and Chardonnay at different sampling time points. Regarding the controls inoculated only with B. cinerea, the results indicate that the severity of the pathogen attack depends on the cultivar. While the necrosis lesions of B. cinerea reached 1 cm at 9 dpi on the Cabernet Sauvignon leaves, these lesions were higher than 1 cm after only 7 dpi on the leaves of the Chardonnay. While the average fungal growth rate was 0.63 cm per day on the leaf discs of the Chardonnay controls inoculated with B. cinerea only, it was 0.47 cm on the same control of Cabernet Sauvignon. Thus, the inhibitory effect of P. oligandrum against B. cinerea depends on the grapevine cultivar. Regarding the controls inoculated only with B. cinerea, the results indicate that the severity of the pathogen attack depends on the cultivar. While the necrosis lesions of B. cinerea reached 1 cm at 9 dpi on the Cabernet Sauvignon leaves, these lesions were higher than 1 cm after only 7 dpi on the leaves of the Chardonnay. While the average fungal growth rate was 0.63 cm per day on the leaf discs of the Chardonnay controls inoculated with B. cinerea only, it was 0.47 cm on the same control of Cabernet Sauvignon. Thus, the inhibitory effect of P. oligandrum against B. cinerea depends on the grapevine cultivar.

Reduction of B. cinerea's Necroses on P. oligandrum-Inoculated Plants
The leaf discs of the Chardonnay plants pre-inoculated with P. oligandrum before inoculation with B. cinerea displayed significantly lower (p < 0.05) necrotic lesions than in those of the controls. This inhibition was approximately 35% at all sampling time points ( Figure 4A). As for the leaf discs of the Cabernet Sauvignon, when the plants were preinoculated with P. oligandrum, a significant reduction in necrotic lesions due to B. cinerea was observed only at the late sampling time points, i.e., 8 and 9 dpi. The most important reduction was approximately 35% at 9 dpi ( Figure 4B).

Evaluation, in the Two Cultivars, of Specific Foliar Grapevine Gene Defense Induction by P. oligandrum in Response to P. viticola and B. cinerea Infection
The P. oligandrum root inoculation on the grapevines' foliar gene defenses was assessed in response, or not, to P. viticola or B. cinerea infection on the two cultivars. In the Cabernet Sauvignon and Chardonnay cultivars, the expression levels of 10 defense-related grapevine genes were evaluated two days post-inoculation with the pathogens. The gene expression levels in the control plants, not inoculated with P. oligandrum or any other pathogen, were used as references.
3.3.1. P. oligandrum Induction of Grapevine Gene Defenses against P. viticola Attack PCA was performed to evaluate possible separation between plant responses to the different treatments ( Figures 5A and 6A), and the correlation circles were exam-ined to characterize the effect of each treatment on grapevine gene defense responses ( Figures 5B and 6B). The results obtained with the Cabernet Sauvignon cultivar are shown in Figure 5A,B. The PCA eigenvalues indicate that the first two principal components explained 80% of the total data variance. As the ellipses overlapped at certain points, the grapevines' responses for each treatment were not significantly different ( Figure 5A). The correlation circles revealed that except for the gene AOS, all of the studied genes were well represented ( Figure 5B). The position of the different genes on the correlation circle shows that all of them were associated with the P. viticola and P. oligandrum + P. viticola treatments. Most of these genes (PR1, GLU, PR5, LOX2, JAR, PAL, and STS) were expressed slightly more in those leaf discs elicited with P. oligandrum, followed by P. viticola, than in those inoculated with the pathogen only.

Dim2 (21.55%)
Regarding the Chardonnay cultivar, the plant responses for each treatment, i.e., the expression level of the 10 genes involved in the grapevine defense system, are shown in Figure 5C,D. The PCA eigenvalues indicate that 48.89% and 21.55% of the total data variance were explained by the two first principal components Dim1 and Dim2, respectively ( Figure 5C). The results show that P. oligandrum significantly modulated the grapevines' responses to P. viticola attack. As for the plants inoculated with P. oligandrum, the responses were not significantly different in those grapevines infected with P. oligandrum alone and in those with the pathogen only. The corresponding correlation circle was examined ( Figure 5D) and only well-represented genes were presented. Most of these genes correlated with the grapevines' responses to P. oligandrum treatment, in the presence or absence of the pathogen. In order to better understand the effect of P. oligandrum on the relative expression of the studied genes, two days after P. viticola inoculation, Table 1A,B shows the expression level of each gene according to the individual treatments for the Cabernet Sauvignon and Chardonnay cultivars, respectively. Table 1. P. oligandrum primed the differential expression of defense-related genes in the leaf discs of Cabernet Sauvignon (A,C) and Chardonnay (B,D) after P. viticola (A,B) and B. cinerea (C,D) inoculation. The gene expression of the control plants was used as the reference to calculate the relative expression. Columns represent the treatments (Po = P. oligandrum; Pv: P. viticola; PoPv: P. oligandrum + P. viticola; Bc: B. cinerea; PoBc: P. oligandrum + B. cinerea), and each line corresponds to one gene, represented by a single row of boxes. The color scale bars represent the ratio values corresponding to the mean of three independent samples. Upregulated genes are shown in shades of red, with relative expression levels greater than three in bright red. Downregulated genes are shown in shades of grey, with relative expression levels less than 0.1 in dark grey. Numbers in boxes represent significant changes (p < 0.05; Kurskal-Wallis test) in gene expression compared to the control. absence of the pathogen. In order to better understand the effect of P. oligandrum on the relative expression of the studied genes, two days after P. viticola inoculation, Table 1A,B shows the expression level of each gene according to the individual treatments for the Cabernet Sauvignon and Chardonnay cultivars, respectively. Table 1. P. oligandrum primed the differential expression of defense-related genes in the leaf discs of Cabernet Sauvignon (A,C) and Chardonnay (B,D) after P. viticola (A,B) and B. cinerea (C,D) inoculation. The gene expression of the control plants was used as the reference to calculate the relative expression. Columns represent the treatments (Po = P. oligandrum; Pv: P. viticola; PoPv: P. oligandrum + P. viticola; Bc: B. cinerea; PoBc: P. oligandrum + B. cinerea), and each line corresponds to one gene, represented by a single row of boxes. The color scale bars represent the ratio values corresponding to the mean of three independent samples. Upregulated genes are shown in shades of red, with relative expression levels greater than three in bright red. Downregulated genes are shown in shades of grey, with relative expression levels less than 0.1 in dark grey. Numbers in boxes represent significant changes (p < 0.05; Kurskal-Wallis test) in gene expression compared to the control. Regarding the Cabernet Sauvignon cultivar, the results show that all of the studied genes, except PR1, were not induced in response to P. oligandrum treatment (Table 1A). Compared to the control plants, only the PR1 expression level was significantly overexpressed 16 days after P. oligandrum inoculation. Leaf disc inoculation with the pathogen significantly induced the expression of 40% and 80% of the studied genes in the P. viticola and P. oligandrum + P. viticola treatments, respectively. In the leaf discs collected from the P. oligandrum-treated plants that were then infected with the pathogen, only two genes involved in the signaling pathways (AOS and EIN3) were not significantly overexpressed.

A
As for the Chardonnay cultivar, the results are presented in Table 1B. As observed with the Cabernet Sauvignon cultivar, most of the studied genes were not significantly expressed 16 days after P. oligandrum inoculation, except for those genes involved in secondary metabolite biosynthesis (PAL and STS), which were slightly repressed. Following the pathogen inoculation, the expression levels of all of the studied genes (except LOX2) were repressed in all of the treatments, but the downregulation of these genes varied according to the treatment. The highest repression levels were observed in those leaf discs inoculated with the pathogen only, especially PR5 (ER = 0.04), AOS (ER = 0.1), and STS (ER = 0.09).

P. oligandrum Induction of Grapevine Defenses against B. cinerea
As per the previous study on grapevine foliar defenses after P. viticola inoculation, PCA and correlation circles were carried out in the same way with B. cinerea treatment.
For the Cabernet Sauvignon cultivar, the PCA eigenvalues indicate that the first two principal components explained 75% of the total data variance. The results show that the grapevines' responses differed significantly according to the treatment ( Figure 6A). The correlation circles of the studied genes reveal that, except for LOX2, all of the genes were induced in response to the pathogen attack ( Figure 6B). The level of this gene expression was higher in leaf discs collected from P. oligandrum-inoculated plants and then infected by B. cinerea, than in those inoculated with the pathogen only.
Regarding the Chardonnay cultivar, the PCA eigenvalues indicate that the first two principal components, Dim1 and Dim2, explained 73% of the total data variance. Similarly to the Cabernet Sauvignon gene responses, the Chardonnay responses to P. oligandrum differed significantly from those of the B. cinerea and P. oligandrum + B. cinerea treatments ( Figure 6C). Dim1, which represented 49.04%, separated specific grapevine responses into two distinct groups depending on the presence or absence of the pathogen. The correlation circles show that three of the 10 studied genes (EIN3, LOX2, and PR1) were associated with the plants' responses to P. oligandrum ( Figure 6D). However, most of the studied genes were induced in the presence of B. cinerea (JAR, STS, PAL, GLU, ACC, and AOS). When the plants were pretreated with P. oligandrum, the expression of these genes in response to B. cinerea infection was higher than those observed in leaves infected with the pathogen only.
In order to go further with the analyses of the grapevines' gene expression in response to B. cinerea and P. oligandrum, Table 1C,D shows the expression level of each gene for each treatment. For the Cabernet Sauvignon cultivar, the majority of the 10 studied genes were not highly modulated in the leaf discs of those plants inoculated with P. oligandrum alone, except for PR5, which was four times more induced. Following inoculation with B. cinerea, all of these genes were induced more, except LOX2, which was significantly downregulated. The gene that showed the highest level of expression in the presence of the pathogen (100 times more expressed than the control) was the one involved in the jasmonate synthesis pathway, AOS. Overall, all of the studied genes were induced more in the leaf discs of those plants inoculated with P. oligandrum and B. cinerea than in those of the plants inoculated with B. cinerea only. Thus, inoculation of the plant roots with P. oligandrum strongly induced the expression of the studied genes, especially PR5 and AOS, for which the level of expression was multiplied by five in the leaves inoculated with P. oligandrum and then infected with B. cinerea compared to those inoculated with the pathogen only. As for the genes involved in the synthesis of secondary metabolites (i.e., PAL and STS), their relative expression levels were twice as high in the presence of P. oligandrum and B. cinerea.
Regarding the Chardonnay cultivar, the results show that in presence of the pathogen, most of the 10 studied genes were strongly induced (Table 1D). As an example, the genes involved in secondary metabolite synthesis (i.e., PAL and STS), signaling pathways (i.e., AOS, ACC), and PR proteins (i.e., PR1 and GLU) were at least three times more overexpressed in plants inoculated with B. cinerea than in the controls. In the Chardonnay plants, the gene that showed the highest expression level (5890 times more than in the control plants), in response to pathogen infection, was the AOS gene involved in jasmonate synthesis. The expression of the PR1, PR5, AOS, and LOX2 genes were induced more in the leaf discs of those plants inoculated with P. oligandrum and B. cinerea than in those plants inoculated with the pathogen only. Similarly to what was observed for the Cabernet Sauvignon cultivar, the level of expression of the AOS gene was more strongly induced in those plants inoculated with P. oligandrum and B. cinerea (relative expression = 8009) than in those infected by the pathogen only (relative expression = 5890).

Discussion
Based on the hypothesis that P. oligandrum enhances grapevine resistance, we implemented a two-disease approach to decipher the induced plant responses when two major grapevine leaf and berry pathogens attack young vines. We selected two pathogens with different trophic behaviors, i.e., P. viticola (a necrotrophic oomycete) and B. cinerea (a biotrophic fungus), and we used two cultivars, i.e., Cabernet Sauvignon and Chardonnay, with different susceptibilities to both diseases.
With root colonization being an absolute pre-requisite for a biocontrol agent to protect plants against pathogenic attacks [40,41], we first checked this point. Under our experimental conditions, we confirmed that P. oligandrum effectively colonizes the roots of Cabernet Sauvignon and Chardonnay, two grapevine cultivars among the most widely planted in the world. Then, before deciphering the induced resistance, we estimated the level of biocontrol protection. Unlike Haidar et al. [25], who observed in another two-pathogen approach that certain bacteria used as biocontrol agents inhibit B. cinerea infection but favor that of another pathogen, namely Neofusiccocum parvum, in our experiment, both pathogenic infections were reduced when P. oligandrum colonized the roots of the vines. For P. viticola, the pathogen growth was reduced by 35% regardless of the cultivar used, while there was a 30-35% reduction in the development of B. cinerea necrosis on the leaves of the Chardonnay and Cabernet Sauvignon.
With respect to P. viticola, this result is consistent with what has previously been documented by Taibi et al. [24] on the ability of the biocontrol product Polyversum (based on P. oligandrum) to control P. viticola on the grapevine cultivar Barbera. As for B. cinerea, the protection we obtained was 30-35%, and the necroses of B. cinerea in the Chardonnay leaf discs were higher than those measured on the Cabernet Sauvignon leaf discs. This result is in accordance with [42,43], who showed that the Chardonnay cultivar is more susceptible to gray mold than Cabernet Sauvignon. In a previous experiment with P. oligandrum, Mohamed et al. [21] obtained 75% protection on the Pinot Noir cultivar, known for its high susceptibility to B. cinerea. The differences in protection (35% vs. 75%) between our study and that of Mohamed et al. [21] are certainly linked to the different experimental conditions used.
In the literature, further to plant inoculation with potential BCAs, the induction of resistance was reported against P. viticola and B. cinerea attack. For instance, Lakkis et al. [37] indicated that the bacterial strain Pseudomonas fluorescens PTA-CT2 induced systemic resistance (ISR) against P. viticola and B. cinerea by priming various defensive pathways. Among the fungi, various Trhichoderma species were able to induce a priming-type systemic defensive response against B. cinerea through the inhibition of ROS production [44]. Similarly for P. viticola, several fungal species with biocontrol potential induced plant resistance against this pathogen [5]. Therefore, the grapevine resistance induced by the oomycete P. oligandrum observed in this study is consistent with this literature [4,5]. However, our results also showed that the relative expression levels of the genes depended on the two cultivars and the two pathogens.
Regarding the plant interaction with the microorganisms we used, it was first shown that P. oligandrum inoculation alone did not induce any consistent change in the basal expression of the defense genes in the Chardonnay and Cabernet Sauvignon cultivars. Therefore, P. oligandrum presumably primed the plants to enhance their resistance when a subsequent pathogenic attack occurred. This result is in agreement with previous studies on P. oligandrum and other BCA plant inoculations alone, showing that they were not associated with a significant induction of plant defenses [4,5].
When the leaf discs were challenged with the pathogens, the grapevines' defenses were activated, though they varied depending on each pathogen/cultivar combination. In another study, Solanum lycopersicum L. cv. Micro-Tom seeds were coated with P. oligandrum strains, and then the seedlings were exposed to fungal pathogens, i.e., Alternaria brassicicola or Verticillium albo-atrum [45]. Their results depended on the plant species, but also, as is the case in our study, on the cultivars and type of pathogens.
Thus, a difference in the relative expression of the studied genes in response to each pathogen (B. cinerea and P. viticola) was observed in our study. This could be attributed to the lifestyles of the two pathogens, i.e., biotrophic or necrotrophic, which are linked to the activation of the specific plant metabolic pathways. Indeed, the mechanisms of plant defense against biotrophic pathogens described in the literature are associated with responses regulated by the salicylic-acid-dependent pathway, whereas for the necrotrophic pathogens, there was a different set of defense responses activated by jasmonic acid and ethylene signaling [46,47]. Here, in response to P. viticola infection, in Cabernet Sauvignon, the genes from the jasmonate and ethylene pathways were induced more than in the more susceptible Chardonnay cultivar. This result is in accordance with several authors [46][47][48], who reported that P. viticola infection results in an induction of the genes involved in jasmonate and ethylene biosynthesis, which is more important in resistant than susceptible cultivars.
A different result was observed with B. cinerea; the relative expression levels of the studied genes involved in PR protein, ethylene/jasmonate, and secondary metabolite biosynthesis were higher in the susceptible Chardonnay cultivar than in the less-susceptible Cabernet Sauvignon cultivar. In another experiment with a different plant/pathogen system, the JA-dependent signaling pathway was reported to be required for the induction of tomato resistance by P. oligandrum against Ralstonia solanacearum [49].
Regarding the studied genes, we showed that the AOS gene involved in the jasmonate pathway was strongly induced in response to B. cinerea compared to P. viticola infection in leaf discs primed, or not, with P. oligandrum. The same kind of result was observed with the AOS gene in a previous study involving P. oligandrum control of the GTD pathogen P. chlamydospora [20]. Consequently, the AOS gene could be an interesting marker for the grapevine protection induced by P. oligandrum against certain grapevine pathogens.

Conclusions
The two-pathogen approach used in this paper provided evidence that the strain of P. oligandrum we used has real potential to control attacks of both P. viticola and B. cinerea on grapevines. The different levels of defense against B. cinerea and P. viticola could explain the difference in susceptibility of the two grapevine cultivars (Chardonnay and Cabernet Sauvignon) to these two pathogens. Moreover, as the efficacy of BCA products based on P. oligandrum, e.g., Polyversum ® , against P. viticola and B. cinerea is variable in vineyards, a solution to improve and secure the protection offered by P. oligandrum could be the use of mixtures of P. oligandrum strains, such as the one used in this paper. This mixture of P. oligandrum strains could then be integrated into agroecological strategies for grapevine protection. Further studies, but under viticultural conditions, should be carried out.