Comparative evaluation of chemical profiles of three representative 'snow lotus' herbs by UPLC-DAD-QTOF-MS combined with principal component and hierarchical cluster analyses.

Herbal healthcare products are used worldwide as relatively safe and effective alternatives to allopathic drugs. Saussurea laniceps Hand.-Mazz. (SL), S. medusa Maxim. (SM) and S. involucrata (Kar. et Kir.) Sch.Bip. (SI) are three sources of the renowned 'snow lotus', Chinese materia medica for treating inflammatory diseases. The three species have different therapeutic effects, among which SL has been proved to be the most potent, but they are frequently confused on the market and in the academic community. An ultra-high performance liquid chromatography-diode array detector-quadrupole time of flight-mass spectrometry (UPLC-DAD-QTOF-MS) method was developed and used to analyze 49 herbal samples for species analysis and overall quality evaluation. With 25 simultaneously identified constituents, of which 12 were quantified, the three herbs showed different chemical profiles. Four-dimensional principle component analysis (4D-PCA) and orthogonal hierarchical cluster analysis (2D-HCA) results illustrated that SL should be grouped away from SM and SI, contradicting the botanical record in Flora of China. The present chemical determination and pattern recognition results directly explain the therapeutic potency of SL and distinguish the three confused snow lotus herbs. Furthermore, the findings suggest a possible extensive quality evaluation model for multi-origin medicinal plants and help monitor falsification of snow lotus herbal products on the market, contributing to a more regulated pharmaceutical industry. Copyright © 2016 John Wiley & Sons, Ltd.