Wine metabolomics and sensory profile in relation to terroir: a case study focusing on different wine-growing areas of Piacenza province (Italy)

In this work, a metabolomic approach based on ultra-high-performance liquid chromatography coupled to quadrupole time-of-flight mass spectrometry (UHPLC-ESI-QTOF) has been applied to investigate the untargeted phenolic profiles of wines from different growing areas, being Val Tidone, Val Nure and Val d'Arda, three valleys located in the province of Piacenza (Emilia Romagna, Italy). In this regard, eight "Gutturnio" DOC wines (both sparkling and still) from the same vintage (2016) were compared in order to highlight the impact of terroir on their phytochemical composition. At the same time, two white wines from the same vintage (2017) were analysed, being "Ortrugo" DOC and "Malvasia di Candia aromatica" DOC. Besides, a sensory evaluation was done in order to underline possible correlations between chemical and climatic data, as related to each growing area. Overall, interesting results were obtained by using UHPLC-ESI-QTOF mass spectrometry; white "Malvasia" and "Ortrugo" wines were very abundant in phenolic acids, being 107.36 and 98.36 mg/L, respectively, while among the "Gutturnio" wines, the highest content were recorded in still Gutturnio wines from Val Tidone and Val d'Arda (i.e., 389.92 and 388.18 mg/L, respectively). Unsupervised multivariate statistics (hierarchical clustering and principal component analysis) following metabolomics-based data allowed to clearly discriminate the different samples according to the corresponding growing-areas. Interestingly, the most discriminant compounds allowing sample grouping belonged to phenolic acids (such as isomeric forms of diferuloylquinic acid) and alkylphenols (such as 5-Heptadecylresorcinol). Besides, the venn diagram analysis revealed 7 common markers belonging to both conditions under investigation (i.e., terroir and winemaking typology). Finally, strong correlations were outlined when considering polyphenol profiles and climatic data, although further ad-hoc studies are extremely needed to confirm this trend.