Metabolomics Insight into the Variety-Mediated Responses to Aspergillus carbonarius Infection in Grapevine Berries

Abstract

Limited knowledge regarding the susceptibility of grape varieties to ochratoxin A (OTA)-producing fungi is available to date. This study aimed to investigate the susceptibility of different grape varieties to Aspergillus carbonarius concerning OTA contamination and modulation at the metabolome level. Six grape varieties were selected, sampled at early veraison and ripening, artificially inoculated with A. carbonarius, and incubated at two temperature regimes. Significant differences were observed across cultivars, with Barbera showing the highest incidence of moldy berries (around 30%), while Malvasia and Ortrugo showed the lowest incidence (about 2%). OTA contamination was the lowest in Ortrugo and Malvasia, and the highest in Croatina, although it was not significantly different from Barbera, Merlot, and Sauvignon Blanc. Fungal development and mycotoxin production changed with grape variety; the sugar content in berries could also have played a role. Unsupervised multivariate statistical analysis from metabolomic fingerprints highlighted cultivar-specific responses, although a more generalized response was observed by supervised OPLS-DA modeling. An accumulation of nitrogen-containing compounds (alkaloids and glucosinolates), phenylpropanoids, and terpenoids, in addition to phytoalexins, was observed in all samples. A broader modulation of the metabolome was observed in white grapes, which were less contaminated by OTA. Jasmonates and oxylipins were identified as critical upstream modulators in metabolomic profiles. A direct correlation between the plant defense machinery and OTA was not observed, but the information was acquired and can contribute to optimizing preventive actions.

Figure 1

Mean incidence of A. carbonarius (%) and ochratoxin A content in untreated and artificially inoculated grape berries of white varieties (A,B) and red varieties (C,D) at the growth stage of fully ripening. Bar plots represent the average number obtained from the 3 experimental replicates considered in the study. The error standard is reported on each bar.

Figure 2

Multivariate statistical elaboration of metabolomic profiles of grape juices following bunch inoculation with A. carbonarius (ITEM, as a pool of two strains and two different temperatures), compared to non-inoculated samples (TEST). Unsupervised hierarchical cluster analysis (HCA, Euclidean distance) obtained from the FC heatmap (A) and supervised orthogonal projection to latent structures discriminant analysis (OPLS-DA) for white (B) and red grape varieties (C).

Figure 3

Pathway analysis of white grape varieties (Malvasia, Ortrugo, and Sauvignon Blanc) conducted on the significantly and differentially modulated metabolites (ANOVA p-value < 0.05; FC ≥ 1.2) following inoculation with A. carbonarius (ITEM, as a pool of two strains and two different temperatures), compared with non-inoculated samples (TEST). Differential metabolites were interpreted in terms of biosynthesis pathways (A), secondary metabolites (B), and hormone biosynthesis (C). The bars represent the sum of the log FC values of each metabolite involved in the biosynthetic pathway. The different dots within each vertical bar indicate the log FC value for each metabolite belonging to the biosynthetic pathway, while the larger dots indicate the median value. Abbreviations: AA: amino acids; FA/Lip: fatty acids and lipid; Sec Metab: secondary metabolites; Cell-Struct: cell-structure; Phenylprop Derivs: phenylpropanoid and derivatives; syn: synthesis.

Figure 4

Pathway analysis of red grape varieties (Barbera, Croatina, and Merlot) conducted on the significantly and differentially modulated metabolites (ANOVA p-value < 0.05; FC ≥ 1.2) following inoculation with A. carbonarius (ITEM, as a pool of two strains and two different temperatures), compared to non-inoculated samples (TEST). Differential metabolites were interpreted in terms of biosynthesis pathways (A), secondary metabolites (B), and hormone biosynthesis (C). The bars represent the sum of the log FC values of each metabolite involved in the biosynthetic pathway. The different dots within each vertical bar indicate the log FC value for each metabolite belonging to the biosynthetic pathway, while the larger dots indicate the median value. Abbreviations: FA/Lip: fatty acids and lipids; Carbo: carbohydrates; Sec Metab: secondary metabolites; Cell-Struct: cell-structure; Phenylprop Derivs: phenylpropanoid and derivatives; syn: synthesis.

Figure 5

Venn analysis conducted from the lists of differential metabolites that passed ANOVA and FC analysis (p-value < 0.05, FC ≥ 1.2) following inoculation with A. carbonarius for (A,B) white grape varieties (Malvasia, Ortrugo, Sauvignon Blanc) and (C,D) red grape varieties (Barbera, Croatina, and Merlot) as up-accumulated metabolites and down-accumulated metabolites, respectively.