Discrimination of Polygonatum species and identification of novel markers using 1H NMR- and UPLC/Q-TOF MS-based metabolite profiling

BACKGROUND Rhizomes of Polygonatum species are commonly used as herbal supplements in Asia. They have different medicinal effects by species but have been misused and mixed owing to their similar taste and smell. Therefore accurate and reliable analytical methods to discriminate between Polygonatum species are required. RESULTS In this study, global and targeted metabolite profiling using 1H nuclear magnetic resonance (1H NMR) spectroscopy and ultra‐performance liquid chromatography/quadrupole time‐of‐flight mass spectrometry (UPLC/Q‐TOF MS) was applied to discriminate between different Polygonatum species. Partial least squares discriminant analysis (PLS‐DA) models were used to classify and predict species of Polygonatum. Cross‐validation derived from PLS‐DA revealed good predictive accuracy. Polygonatum species were classified into unique patterns based on K‐means clustering analysis. 4‐Hydrobenzoic acid and trigonelline were identified as novel marker compounds and quantified accurately. CONCLUSION The results demonstrate that metabolite profiling approaches coupled with chemometric analysis can be used to classify and discriminate between different species of various herbal medicines. © 2015 Society of Chemical Industry