Natasa P. Kalogiouri
1,2, Natalia Manousi
2,3, Antonio Ferracane
2,4, George A. Zachariadis
3, Stephanos Koundouras
5, Victoria F. Samanidou
3, Peter Q. Tranchida
4, Luigi Mondello
4,6,7, Erwin Rosenberg
21Laboratory of Analytical Chemistry, Department of Chemistry, Aristotle University of Thessaloniki, 54124 Thessaloniki, Greece
2Institute of Chemical Technology and Analytics, Vienna University of Technology, Getreidemarkt 9/164, 1060 Vienna, Austria
3Laboratory of Analytical Chemistry, Department of Chemistry, Aristotle University of Thessaloniki, 54124 Thessaloniki, Greece
4Department of Chemical, Biological, Pharmaceutical and Environmental Sciences, University of Messina, Messina, Italy
5Laboratory of Viticulture, School of Agriculture, Aristotle University of Thessaloniki, Thessaloniki, 54124, Greece
6Chromaleont s.r.l., c/o Department of Chemical, Biological, Pharmaceutical and Environmental Sciences, University of Messina, Messina, Italy
7Department of Sciences and Technologies for Human and Environment, University Campus Bio-Medico of Rome, Rome, Italy
This email address is being protected from spambots. You need JavaScript enabled to view it.
Authentication is defined as the process that unquestionably verifies that food is genuine. In a global food market, Protected Designation of Origin (PDO) wines are highly appreciated for the period of aging that allows the evolution of unique organoleptic properties, and thus, the guarantee of the vintage age is a critical authenticity issue [1]. The coupling of SPME-Arrow to GCxGC/MS analysis in combination with chemometric techniques could establish advanced analytical methods for the investigation of the volatile fingerprint of wine and revolutionize the field of wine authenticity. In this work, an SPME-Arrow-GCxGC/MS method was optimized [2], and used for the determination of volatile markers in 24 monovarietal red wine samples belonging to the PDO Xinomavro Naoussa, produced during 4 different years (1998, 2005, 2008 and 2015), in Northern Greece. Overall, 258 volatile compounds were tentatively identified in all samples. The data matrix that constituted of 24 samples and 258 features was further processed with multivariate techniques to establish mathematical models and reveal volatile markers for each vintage age. A partial least square – discriminant analysis (PLS-DA) model was developed and successfully classified all the samples to the proper class according to the vintage age with an explained total variance of 85.1%. Variant Importance in Projection (VIP) algorithm was used to calculate the VIP scores of the determined volatiles and distinguish the most important features that affect the discrimination, revealing markers [3]. The developed prediction model was validated and the analyzed samples were classified with 100% accuracy according to the vintage age, on the basis of their volatile fingerprint. The established PLS-DA model provided satisfying clustering, revealing 15 markers as most important for the classification of the wine samples.
References
[1] N.P. Kalogiouri, V.F. Samanidou, Environ. Sci. Pollut. Res. Int. 28, 42 (2021) 59150-59164.
[2] I. Šikuten, P. Štambuk, J. Karoglan Kontic, E. Maletic, I. Tomaz, D. Preiner, Molecules 26, 23 (2021) 1-16.
[3] N.P. Kalogiouri, N. Manousi, E. Rosenberg, G.A. Zachariadis, A. Paraskevopoulou, V. Samanidou, Food Chem. 363 (2021) 130331.