Principal component analysis and phytochemical fingerprinting as tools for genotype traceability in Castanea spp. fruits

The phytochemical characterisation of fruits from Castanea spp., one of the most important Italian and European fruit and forest species, is very important for the development of effective strategies for the biodiversity conservation and traceability valorisation of chestnuts. In this study, several cultivars of C. sativa (‘marrone’- and ‘chestnut’-types), C. crenata, C. pumila, and hybrids of C. sativa × C. crenata were characterised by spectrophotometric (Folin-Ciocalteu assay for the total polyphenolic content and ferric reducing antioxidant power test for the antioxidant capacity) and chromatographic (high-performance liquid chromatography coupled to a diode array UV-Vis detector) protocols to define their phytochemical composition and nutraceutical properties. Phytochemical results were then used to build a multivariate statistical model (by principal component analysis) and obtain an effective and rapid tool to discriminate unknown cultivars (i.e., no information about their origin) belonging to different species. The multivariate approach showed that the genotype was a significantly discriminating variable (p<0.05) for phytochemical composition. Polyphenols (in particular, phenolic acids and tannins) have been identified as the main bioactive classes with the highest discriminating power among the different genotypes. The principal component analysis applied to phytochemical variables has proved to be an excellent and effective tool for genotype differentiation to be used as a preliminary method for identifying the species of Castanea spp. fruits with unknown origin. The present study showed that a multivariate approach, based on phytochemical data and preliminary to genetic analysis, may represent a rapid, effective, and low-cost tool for the traceability and quality evaluation of chestnuts from different species and hybrids with no information on their origin.