Anticancer agents of kaempferol-3-O-methylether and kaempferol-3-O-(2,4-O-diacetyl-alpha-L-rhamnopyranoside) in leaf of plants Zingiber zerumbet SM using 2D, 3D descriptors

Abstract

The plants are an alternative resource for studying anti-cancer drugs and the natural products isolated from them can point out a direct impact on eliminating cancer cells and also reduce cancer side effects. Recently, we isolated two kaempferols in flavonoid group from leaves of Zingiber zerumbet Sm in Vietnam, with stronger cytotoxin relatively for Hela cancer cells. In this QSAR study on cancer Hela cell line, the techniques of multiple linear regression (MLR) and the artificial neural network (ANN) were integrated with artificial neural network to construct the different QSAR models such as the QSAR_MLR and QSAR_ANN models. The best linear model QSAR_MLR with 9 independent variables and values R2train of 0.955 and R2pred of 0.745 was also found by using a multiple linear regression technique. The artificial neural network QSAR_ANN with structural style I(9)-HL(5)-O(1) exhibited a better quality in statistical values R2train of 0.8963 and R2pred of 0.8883. The anti-cancer activities of the flavonoids in the test group and kaempferol-3-O-methylether and kaempferol-3-O-(2,4-O-diacetyl-alpha-L-rhamnopyranoside) from leaves of Zingiber zerumbet Sm resulting from those models turned out to be agreement with experimental data and those from literature.