Fingerprinting alkaloids for traceability: Semi-untargeted UHPLC-MS/MS approach in raw lupins as a case study

Lupin seeds are recognized for their nutritional value and potential health benefits, but they contain also a significant amount of alkaloids, an anti-nutritive class of compounds, which vary significantly in composition within and between species due to environmental factors. In this study, we developed a predictive multiexperiment approach using ultra-high performance liquid chromatography coupled with triple quadrupole with linear ionic trap tandem mass spectrometry (UHPLC-QqQ-LIT-MS/MS) for comprehensive alkaloid profiling and geographical classification of Lupinus albus L. samples originating from four different Italian regions. Six targeted quinolizidine alkaloids were detected and 21 other alkaloids were putatively identified. Hierarchical Cluster Analysis (HCA) and partial least squares discriminant analysis (PLS-DA) were applied to explore the data structure and successfully classify samples according to their geographical origin. The data demonstrate the efficacy of the developed approach in providing valuable insights in alkaloid profiles of lupin seeds and their potential as markers for geographical traceability.