Yoshinori Fujimura; Kana Kurihara; Megumi Ida; Reia Kosaka; Daisuke Miura; Hiroyuki Wariishi; Mari Maeda-Yamamoto; Atsushi Nesumi; Takeshi Saito; Tomomasa Kanda; Koji Yamada; Hirofumi Tachibana

“ In this study, we created a bioactivity-predictive PLS regression model using 43 Japanese green tea cultivars (Fig. 6) . These analyses yielded unique lists of bioactivity-correlated constituents (46 peaks) in tea extracts. In future, such information may be useful for the development of markers to produce new cultivars with greater bioactivity, and to screen for bioactive tea cultivars. Furthermore, we can predict the potential bioactivity of numerous tea cultivars by analyzing their metabolomic data with our regression model, without the requirement for additional bioactivity assays. ”

“ To provide comparative interpretations and to visualize metabolic differences among cultivars in relation to their bioactivity, we analyzed the LC-MS spectra datasets using several multivariate analyses (see detailed information about multivariate statistical analyses in Materials and methods) .
Heat map analysis provides an overview of all observations or samples in a dataset by highlighting holistic differences in the complex metabolic data. This method can be used to visualize simultaneously the metabolic profiles of many cultivars. ”

“ LC-MS can be adapted to a wide range of molecules, such as secondary metabolites [6] . Metabolomic studies coupled with chemometric methods including principal component analysis (PCA) and partial least-squares (PLS) regression analysis have been used to explore the relationships between the metabolome of diverse plant species and their genotype, origin, vintage, quality, or other specific attributes [7] – [8] . Metabolic profiling techniques are often used to evaluate the nutraceutical (nutritional or physiological) value of a single plant cultivar for quality control and breading. ”

“ Interestingly, a new Japanese green tea cultivar, Sunrouge (SR) , showed some potential to improve endothelial dysfunction by suppressing thrombin-induced MRLC phosphorylation. Furthermore, this metabolomic approach allowed us to identify several bioactive factors that are enriched in SR. If introduced into non-bioactive lines, such factors may be capable of transforming non-bioactive cultivars into bioactive ones. ”

“ Application of a metabolomic approach to evaluate the bioactivity of plants, plant-derived functional foods, or botanical multicomponent drugs may be useful for unraveling their complex mechanisms of action, and/or for exploring the complex interactions between foods/drugs and human health. ”

“ An understanding of the contribution of each original variable to the synthetic variables (loading plot) leads to the identification of key variables (compounds) that contribute to the relationships (similarity or difference) among samples. In this study, we carried out multivariate data analyses (PCA and OPLS-DA) [14] using SIMCA-P+ version 12.0 (Umetrics, Umea, Sweden) . PCA models are depicted as score plots and consist of two synthetic variables: principal component (PC) 1 (the greatest variance of data) and PC2 (the second greatest variance of data, orthogonal with PC1) . ”

“ LC-MS-based metabolic profiling to evaluate bioactivity of Japanese green tea
Aqueous crude extracts of tea leaves from the 43 cultivars were subjected to LC-MS to investigate differences in their compositions. In analyses of complex mixtures such as crude extracts, two or more compounds can be co-eluted. The obtained complex spectral data are usually processed to extract and align peaks. We extracted 541 peaks from a complex chromatogram and used multivariate statistical analysis to decrease the complexity of the spectra datasets. ”

“ In addition to positive regulators in bioactive plants and foods, elucidation of negative regulators and the unique concomitant factors would be useful for the design of nutraceutical products for use in effective and safe functional foods. ”

“ When all 43 samples were ranked according to bioactivity, these samples corresponded to every eighth sample, and were used to verify the model's predictive ability. They were not included in the regression model. The PLS or OPLS relationship between measured and predicted inhibitory activities of green tea cultivars is shown in Fig. 6. The quality of the regression model can be verified by the correlation coefficient R2 and the cross-validated correlation coefficient Q2, as well as the validation errors of estimation and that between measured and predicted values ”

“ Although successive OPLS-DA (Fig. 4) and regression analysis (Fig. 6) revealed several tens of important metabolite peaks, many of these peaks remained unassigned. Elucidating the bioactivity of each identified metabolite in the presence of numerous known and unknown constituents in tea extracts is an important approach for understanding the bioactivity of green tea and its effective nutraceutical applications. ”

“ The membranes were blocked with 2.5% bovine serum albumin and incubated with anti-phosphorylated MRLC (Thr-18/Ser-19) or anti-MLC2 antibody to evaluate phosphorylated or whole MRLC, respectively, followed by incubation with HRP-conjugated anti-rabbit IgG secondary antibodies. Epitopes on proteins specifically recognized by the antibody were visualized using the ECL Advance kit (GE Healthcare UK Ltd., Buckinghamshire, England) . Band intensities were quantified using NIH Image-J software (Bethesda, MD, USA) . Samples were analyzed in triplicate, and representative data are shown in figures. ”