A mathematical model for quality evaluation of total saponins of panax japonicas based on hypolipidemic activity

Abstract Objective The chemical finger printing-based methods for evaluating TCMs quality can report partial of TCMs quality without linking to effective constituents. In this study, a mathematical model was established for the quality evaluation of total saponins of Panax japonicus (TSPJ), a folk medicine in China and Japan for treating diseases, through coupling the dynamic changes of chemical constitutions with corresponding activities. Methods High-performance liquid chromatography (HPLC) fingerprints were applied to establish the chromatographic database of TSPJ. The associated hypolipidemic activity database was determined by TG assay using HepG2 cell model. Correlation analyses of two databases were performed by partial least squares (PLS) for calculating regression coefficients, and the interval value of YZL value (the ratio of positive and negative peak-to-peak area coefficient) closely related to hypolipidemic activity was refined by the formula of Norminv function to value the quality of TSPJ. Results In this study, the chromatographic data of 16 common peaks were obtained from 20 batches of TSPJ. After the estimate by this mathematical evaluation model, seven peaks were positively correlated with hypolipidemic activity, and nine peaks were negatively correlated with hypolipidemic activity. When the YZL value was less than 0.7861, the quality of sample was inferior, while YZL value was more than 6.6992, and the quality of samples was superior. The quality of another ten batches of TSPJ was further assessed to verify this method. Conclusion These results indicated that the established model could be usefully applied to evaluate the quality of TSPJ in the hypolipidemic activity.