Authentication of retail cheeses based on fatty acid composition and multivariate data analysis

Abstract

A methodology to discriminate retail cheeses by principal component analysis (PCA) and consequent orthogonal partial least squares discrimination analysis (OPLS-DA) using fatty acid (FA) profile differences is reported. Multivariate analysis included retail cheeses from 3 different varieties (Gouda, Chanco and Mantecoso) and 2 distinct scales of production. PCA was useful in discriminating cheeses according to their variety, but it did not allow differentiation according to the scale of production. Gouda and Chanco cheeses were differentiated by saturated FAs (C6:0, C8:0, C10:0, C11:0, C12:0, C14:0, C16:0, and C18:0) whereas Mantecoso cheese was discriminated by specific (C4:0, C14:1, C16:1, C17:0, and C18:1) FAs. OPLS-DA differentiated cheeses based on the scale of production, which would be related to the feeding regime of the dairy cattle. C16:1c9 showed the strongest association with large-scale production cheeses and intensive systems, while C15:0c9, C17:0, C20:1n9, C20:4n6, and C22:2 were characteristic of artisanal cheeses and extensive feeding regimes.