ABSTRACT
Objective: To explore the change rules of active ingredients in Phyllanthi Fructus of different storage years,in order to provide theory basis for storage. Method: Seven Phyllanthi Fruatus samples of different storage years were collected. HPLC-UV detection method was established to determine the contents of gallic acid,corilagin,chebulagic acid,ellagic acid and quercetin. Samples were fingerprinted by FT-NIR and identified by PLS-DA model. Result: Gallic acid,which was the bioactive marker in Chinese Pharmacopoeia,had the highest content. It was followed by ellagic acid and chebulagic acid,and corilagin and quercetin had the least content. The components had significant differences between samples of different storage years (P-1 respectively. The contents of chebulagic acid,corilagin and ellagic acid reached a maximum at 4 years of storage,which were 18.85,7.97,21.46 mg·g-1,respectively. FT-NIR data was optimized by MSC+SG (second derivative, the window parameter as 11,and the polynomial order as 3). The classification accuracy was 84.5%. Spectral data reduced to several important potential variables,and was fused with 5 active components based on minimum cross-validation root mean square error,and the classification accuracy increased to 98.8%. Conclusion: The analysis of PLS-DA by HPLC-UV and FT-NIR could effectively explain the accumulation characteristics of active components in Phyllanthi Fruatus. According to the data fusion strategy,PLS-DA model could distinguish samples of different qualities. The results provide a scientific basis for the quality evaluation and identification of Phyllanthi Fruatus.