A Non-Targeted Metabolomic Approach for the Characterization of Chemical Profile of Sparkling Wines Produced Using Autochthonous Yeast Strains

Abstract

The Champenoise method for the production of sparkling wines consists of an “in bottle” re-fermentation induced by a low number of commercial Saccharomyces cerevisiae strains. However, the limited use of these strains can induce flattening of the sensorial properties of the produced wine. Indeed, the volatile compounds produced by the yeasts during the fermentative process play an essential role in the constitution of the Bouchet sparkling wines prepared using indigenous grape varieties and, in particular, deriving from Apulia (Southern Italy), whose economic significance is constantly increasing [1]. To the best of our knowledge, only two investigations [2,3] have assessed the influence of autochthonous starter strains on the volatile profiles of sparkling wine. Moreover, no information exists on the oenological potential of the indigenous grape cv Maresco for the production of typical sparkling wine. This investigation evaluated the impact of the selected autochthonous yeast strains on the chemical properties of Maresco sparkling wines, applying a non-targeted metabolomics approach by the application of HPLC–HRMS and GC–MS techniques. Saccharomyces cerevisiae DV10 (Lallemand, Petaluma, CA, USA) was used as a control. The obtained findings indicated that previously selected native yeast strains influenced the amount of several fermentative compounds, i.e., polysaccharides, organic acids, phenolic acids, and lipids [4,5]. In order to reveal any grouping of wines based on the composition of volatile compounds and chemical data, as well as to identify the main components contained within each group, the data were subjected to principal component analysis (PCA). Additionally, the correlation matrix on HPLC–HRMS/GC–MS data was performed and a map was obtained, which represents a useful tool to monitor the different patterns of aroma release. The contribution supplies information that will support the modulation of the sensorial properties of regional sparkling wines obtained from typical autochthonous grape varieties using selected autochthonous microbial resources.

Keywords

• sparkling wine
• volatile fraction
• chemical profile
• multivariate statistical analysis
• autochthonous yeasts