On Separation And Analysis Of Flavonoids In Chinese Traditional And Herbal Drugs

Extraction off flavonoids from Chinese traditional and herbal drugs was of important significance to cure, treat and prevent tumour, senescence and cardiovascular diseases, which had fairly good market prospect at present. Flavonoids that are important and natural organic compounds were produced during plants metabolization, which has been the potential source of the search of lead biologically active compounds and has also been the emphasis of the research, development and ultilization at home and aboard in the recent 30 years.In this thesis, flavonoids were extracted from Glycyrrhiza , Pueraria lobata (Willd.) Ohwi and Ginkgo biloba L by several conventional extraction ways and microwave-assisted extraction(MAE) through experimental design and optimization and the flavonoid extracts were further separated and analyzed by chromatography and spectrophotometry. On the other hand, the chemical shift of carbon-13 nuclear magnetic resonance of flavonoids was estimated and predicted through moleculat modeling of quantitative structure-spectrum relationship (QSSR) by the novel atomic electronegative distance vector (AEDV); and molecular modeling for 104 flavonoid derivatives as p56lck protein tyrosine kinase (PTK) inhibitors were quantitatively performed by the new molecular electronegative distance vector (MEDV). The main contents and some conclusions are provided as follows:1.Glycyrrliza flavonoids, Pueraria lobata (Willd.) Ohwi flavonoids and Ginkgo biloba L flavonoids were extracted by the microwave technique through homogeous / uniform design, the orthogonal experimental design or central composite design. The results indicated that the extraction rate of MAE under the optimal condition obtained through the experiments design was remarkably higher than the one of traditional extraction, the extraction time of MAE was greatly shorten and the purity of flavonoids was obviously enhanced as well..2.Various chemical shifts of 13CNMR for different chemically equivalent carbon atoms in nineteen natural flavonols and twenty-one natural flavones were estimated and /or predicted by using the novel atomic electronegative distance vector (VAED) with 4 elements parameter: The sample numbers n, correlation coefficients R, standard deviation SD, F-statistic value F, regression quadratic sum and residual quadratic sum.being n= 285, R=0.796, SD=14.734, F=162.315, U=105705.359, Q=60999.156; n= 315, R=0.884,SD=12.381, F=369.114, U=169745.141, Q=47673.375; N= 600,R=0.831, SD=14.122, F=443.449, U=265310.563 and Q=118859.906, for 19 flavonols, 21 flavones, and all 40 flavonoids, respectively. Finally, cross validation with the leave-one-out (LOO) procedure was made by the QSSR equations for all carbon atoms, with the good results obtained. A relatively reliable correlation model has been developed by the atomic electronegative distance vector, as adjoin structural descriptors. The modeling results were fairly good, but this is preliminary approach to estimation and prediction of chemical shift of carbon-13 nuclear magnetic resonance spectroscopy in flavonoids.3.Quantitative structure-activity relationship (QSAR) studies on 104 flavonoid derivatives as the p56lck protein tyrosine kinase (PTK) inhibitors were performed, using the newly developed molecular electronegative distance vector(VMED). Multiple linear regression(MLR) and leave-one-out cross validation were applied to generate models to predict of biological activities for binding flavonoids to PTK. The result obtained was compared with modeling results of 504 descriptors that were divided into five groups: constitutional, topological, geometric, electrostatic and quantum-chemical, which was not the best but fairly good. It illustrated that the molecular electronegative distance vector(VMED) could be widely applied, especially.