Abstract
Over the past decade, 1H NMR metabolomic studies have become increasingly popular in wine research since they can provide a chemical fingerprint with information regarding authenticity, geographical origin, vintage, and wine composition. Comparatively few 1H NMR-based studies have evaluated sparkling wines. In the present study, quantitative 1H NMR metabolomics was applied to assess the composition of finished traditional method (bottle-fermented) sparkling wines aged for 18-months in cellar conditions (14 °C, 70% relative humidity) following liqueur d’expedition (dosage) additions with six sugar types (glucose, fructose, sucrose derived from sugar beets and sugar cane, maltose, and MCR Sucraisin®) and a zero-dosage control. No systematic trends were observed when comparing the metabolite profile across sugar types, nor between the zero-dosage control and treatments with sugar additions. Multivariate statistical approaches using principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) did not effectively discriminate between dosage treatments nor wine age, while orthogonal projections to latent structures discriminant analysis (OPLS-DA) successfully discriminated between wines by age and was validated through permutation tests. This study presents the first application of 1H NMR metabolomics coupled with multivariate analysis to monitor the aging of sparkling wines.