Abstract
A untargeted metabolomics approach was proposed in this study based on ultra-high performance liquid chromatography quadrupole time-of-flight (UHPLC-QTOF) and rapid evaporative ionization mass spectrometry (REIMS) to discriminate lamb and mutton meat and investigate their subtle metabolic differences, considering the higher popularity of lamb meat than mutton in the market. Multivariate statistical analysis was performed for data processing in order to distinguish between the two sample types. A total of 42 potential metabolites (20 in positive and 22 in negative ion mode) were defined for UHPLC-QTOF analysis, which provided references for discriminating the two kinds of meat. Furthermore, three potential markers were tentatively identified using LC/MS data against chemical databases. In addition, 14 potential metabolites were putatively identified in negative ion mode using the LipidMaps database. Meanwhile, the data-driven soft independent modeling of class analogy (DD-SIMCA) model was established, which could rapidly differentiate non-pretreated lamb meat and mutton with 92% specificity, rendering REIMS a promising technique for meat identification.