Ultra-high-performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry (MS) and an alternative technology represented by direct analysis in real time coupled with quadrupole time-of-flight MS were investigated for metabolic fingerprinting of 343 red and white wine samples. Direct injection of pure wine and an extraction procedure optimized for isolation of polyphenols were used to compare different analytical and data handling strategies. After data processing and data pretreatment, principal component analysis was initially used to explore the data structure. Initially, the unsupervised models revealed a notable clustering according to the grape varieties, and therefore supervised orthogonal partial least squares discriminant analysis models were created and validated for separation of red and white wines according to the grape variety. The validated orthogonal partial least squares discriminant analysis models based on data (ions) recorded in positive ionization mode were able to classify correctly 95% of samples. In parallel, authentication parameters, such as origin and vintage, were evaluated, and they are discussed. A tentative identification of markers was performed using accurate mass measurement of MS and MS/MS spectra, different software packages and different online libraries. In this way, different flavonol glucosides and polyphenols were identified as wine markers according to the grape varieties.

Department of Food Analysis and Nutrition Faculty of Food and Biochemical Technology Institute of Chemical Technology Prague Technicka 3 166 28 Prague 6 Czech Republic

Citace poskytuje Crossref.org

Characterization and Discrimination of Ancient Grains: A Metabolomics Approach