| As one of the most important economic crops Tea has a long history. In traditional famous quality green tea, longjing tea, especially the status of the west lake longjing. At present, in addition to the west lake longjing, like Dafo longjing tea, "regional brand" has momentum of rapid development, and formed unique longjing tea tea industry "brand plate", also caused the industry high attention. At the same time, in the west lake longjing, for example some of the famous tea market still exist the phenomenon of "genuine", in order to give consumers a clear judgment standard and complete scientific basis, this study on judgment oblate tea origin for the purpose to study metabolite fingerprinting classification and discrimination.As Longjing tea quality is assessed from attributes of color, taste and aroma, more comprehensive understanding on chemical compositions associating with these features were essential. Unbiased non-targeted analyses based on high-throughput analytical tools provide more comprehensive picture of the chemical composition in teas therefore may be used to assess the tea quality objectively and reliably. In the present experiment, 50 samples of green teas with flat appearance processed with standardized and uniform procedure collected from 4 provinces in China were subjected to metabolomics analysis of UPLC-Q-TOF/MS based on water or methanol extracts. In the meantime, sensory evaluation was blindly performed to tea samples by officially certified tasters. The objectives were to identify potential marker metabolites elucidating the relationship with green tea quality and to establish a model for quality prediction.This experiment mainly the results of the study are as follows:1 The samples were well separated into different groups by partial least-squares discriminant analysis(PLS-DA) include Zhejiang, Sichuan, Guizhou and Shandong(R~2X=0.396,R~2Y=0.874,Q~2X=0.898) and also separated by principal component analysis(PCA) well r(R~2X=0.643, Q~2X=0.363). Cysteinyl-Glycine 、 3-Nitrotyrosine 、 Mevalonic acid-5P 、N-Carbamoyl-2-amino-2-(4-hydroxyphenyl)acetic acid 、 3-O-p-Coumaroylquinic acid 、 Gallic acid 4-O-(6-galloylglucoside) were significantly in different tea product province.2 The samples were well separated into different groups by PLS-DA include Xihu, Qiantang, Yuezhou and Jinyun which were all in Zhejiang province(R~2X=0.437,R~2Y=0.784,Q~2X=0.845), but wasn’t separated well by PCA(R~2X=0.730, Q~2X=0.388). 5-Amino-6-(5’-phosphoribitylamino)uracil 、4-Methyl-2-phenyl-2-pentenal 、 Ethyl aconitate 、1-(2-Hydroxyphenylamino)-1-deoxy-beta-D-gentiobioside 1,2-carbamate、Imidazoleacetic acid ribotide were significantly in different tea product area.3 The samples were well separated into different groups by PLS-DA include three different area in West Lake District(R~2X=0.453,R~2Y=0.848,Q~2X=0.895) and also separated well by PCA(R~2X=0.786,Q~2X=0.464). 4-O-Methylgallic acid 、 3,4,5-Trimethoxycinnamic acid 、 Ethyl 2-phenyl-3-furancarboxylate 、 3-(3,4-Dihydroxyphenyl)lactic acid 、 Dimethyl tetrasulfide were significantly in different Tea product countries.4 The samples were well separated into different groups by PCA, only those based on water extracts well corresponded to the result of sensory evaluation. Key components differentiating groups of high and low quality scores were screened by PLS-DA. Flavonoids epigallocatechin, catechin-(4α'8)-epigallocatechin, myricetin 3,3’-digalactoside, trigalloyl-b-D-glucopyranose, kaempferol, oxidation products neotheaflavin and neotheaflavin 3-O-gallate were significantly higher in the lowest than in the highest quality group. By contrast, theogallin and gallocatechin were greater in high than in low ranks. The sensory score of green teas was well predicted by a partial least-squares(PLS) regression based on metabolic fingerprinting of water extracts with excellent power(R~2 = 0.87 and Q~2 = 0.82) and accuracy(RMSEP=1.36). |
PDF Full Text Request