Red wines constitute the majority of the wines produced in Bordeaux. All along the winemaking process, many microorganisms may develop in wine. A lot of them are useful but a common defect found in wine is linked to the development of Brettanomyces bruxellensis, a yeast that produces volatile phenols. These molecules are responsible for an unwanted sensorial defect described as similar to “horse sweat”, “burnt plastic” or “leather”. It has been shown that while some wines are very permissive and easily contaminated, others are pretty resistant to Brettanomyces development. However, common parameters such as pH, alcohol or sugars composition cannot fully explain the differences observed in wine permissiveness.
In this study, we aim to explain the wine permissiveness by identifying chemical markers specifically present in permissive wines or, on the contrary, in resistant ones. To achieve this goal, we will analyze the metabolite profiles of red wines coming from different châteaux in Bordeaux and displaying different permissiveness, using targeted and untargeted metabolic profiling by UHPLC-UV-HRMS and 1H-NMR. A microbiological study measuring the growth of a couple of Brettanomyces strains will also be conducted to create and assess a permissiveness score for each wine. With the help of unsupervised statistical analyses, these results will be combined in order to draw correlations between the chemical markers and the score obtained by each wine.
Authors: Miranda Julie1, Dols-Lafargue Marguerite1 and Rouger Caroline1
1Univ. Bordeaux, ISVV, UMR Œnologie EA 4577, UMR 1366 INRAE
*corresponding author: julie.miranda@u-bordeaux.fr
Keywords: Untargeted Metabolomics, Brettanomyces bruxellensis, UHPLC-UV-HRMS, Wine
