UHPLC-qMS spectrum-effect relationships for Rhizoma Paridis extracts

Rhizoma Paridis (RP) with significant anti-tumor and haemostatic effects, has been used as the raw material of many Traditional Chinese preparations. However, its active ingredients are still unclear. The present study aimed to discover bioactive ingredients from RP based on spectrum-relationship and chemometric methods. Firstly, the saponins extract was prepared by phytochemical methods. Furthermore, UHPLC-QTOF-MS and UHPLC-qMS were incorporated to establish an efficient and sensitive method for obtaining the chemical profiles of RP. A total of 34 saponins were characterized in RP and 13 of them were assigned as common peaks in 25 batches of samples. After evaluation of the anti-tumor and haemostatic activities of samples, spectrum-effect relationships were investigated by the grey relational analysis (GRA), orthogonal projections to latent structures (OPLS) and back propagation artificial neural network (BP-ANN). These analyses showed that polyphyllin VII (P27), polyphyllin II (P30), dioscin (P31) and polyphyllin I (P33) play a role in the haemostatic effects of RP whereas polyphyllin VII (P27), dioscin (P31), polyphyllin I (P33), progenin III (P34) were assigned as candidate ingredients accounting for the anti-tumor activity of RP. The anti-tumor and haemostatic activities of these screened ingredients were subsequently verified in vitro. Collectively, the present study established the spectrum-effect relationship mode of RP and discovered the bioactive compounds of RP, which could be also used for exploration of bioactive compounds in herbal medicines, especially for trace compounds.