Introduction: The Salvia genus contains numerous economically important plants that have horticultural, culinary and nutraceutical uses. They are often similar in appearance, making species determination difficult. Species identification of dried Salvia products is also challenging since distinguishing plant morphological features are no longer present.
Objective: The development of a simple high-throughput method of analysis of fresh and dried Salvia leaves that would permit rapid species-level identification and detection of diagnostic biomarkers.
Methodology: Plant leaves were analysed in their native form by DART-MS without the need for any sample preparation steps. This furnished chemical fingerprints characteristic of each species. In the same experiment, in-source collision-induced dissociation was used to identify biomarkers. Biomarker presence was also independently confirmed by GC-MS. Chemometric processing of DART-MS profiles was performed by kernel discriminant analysis (KDA) and soft independent modelling of class analogy (SIMCA) to classify the fingerprints according to species.
Results: The approach was successful despite the occurrence of diurnal cycle and plant-age related chemical profile variations within species. In a single rapid experiment, the presence of essential oil biomarkers such as 3-carene, α-pinene, β-pinene, β-thujone, β-caryophyllene, camphor and borneol could be confirmed. The method was applied to rapid identification and differentiation of Salvia apiana, S. dominica, S. elegans, S. officinalis, S. farinacea and S. patens.
Conclusion: Species-level identification of Salvia plant material could be accomplished by chemometric processing of DART-HRMS-derived chemical profiles of both fresh and dried Salvia material. Copyright © 2016 John Wiley & Sons, Ltd.
Keywords: DART-MS; KDA; SIMCA; Salvia; ambient ionisation; essential oils; multivariate statistical analysis; sage; triterpenoids.
Copyright © 2016 John Wiley & Sons, Ltd.