Detection of goat milk adulteration in horse milk using LC-HRMS-based non-targeted metabolomics and chemometrics

For economic reasons, high-price milk, such as horse milk (HM), can be adulterated using lower-price milk by unethical producers and traders. Milk adulteration is a serious problem because it is closely associated with the product's quality and safety, which could harm consumers. The purpose of this research was to develop a non-targeted metabolomics approach using liquid chromatography-high resolution mass spectrometry (LC-HRMS) in combination with chemometrics for the detection of goat milk (GM) adulterated in horse milk (HM). Principal component analysis (PCA) successfully differentiated between authentic HM and adulterated HM with GM. In addition, discrimination and classification between authentic and adulterated HM with GM were successfully performed using partial least square-discriminant analysis (PLS-DA). The PLS-DA was confirmed for its accuracy, precision, and validity. Moreover, multivariate regression of partial least squares (PLS) and orthogonal PLS (OPLS) could predict the amount of GM added in HM with high accuracy. Analysis of the variable importance for projection (VIP) and S-line plot found that metabolites of ( ±)9(10)-EpOME, 1,2:5,6-dianhydro-3,4-dideoxy-1-dodecyl-6-[12-(5-methyl-2-oxo-2,5-dihydro-3-furanyl)dodecyl]hexitol, 1-Palmitoleoyl-2-oleoyl-sn-glycerol, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine, DG (14:0/18:3(9Z,12Z,15Z)/0:0), DG (16:1(9Z)/18:3(9Z,12Z,15Z)/0:0), ditridecanoin, DMG (dimyristoyl glycerol), ethyl palmitoleate, giganin, octadec-9-ynoic acid, and palmitoleic acid were responsible for the discrimination of HM and GM as well as for the prediction of GM in HM. It can be concluded that a non-targeted metabolomics approach using LC-HRMS combined with chemometrics is potential for milk authentication.