HPLC fingerprinting-based multivariate analysis of phenolic compounds in mango leaves varieties: Correlation to their antioxidant activity and in silico α-glucoidase inhibitory ability.

Mango leaves (MLs) have many important medical values owing to its high contents of phytochemical compounds. Among them, phenolic compounds existing in MLs showed multiple pharmacological activities. However, there is a little information about the quality evaluation and discrimination of different varieties of MLs. In the present study, the chemical compositions of MLs were identified by using high performance liquid chromatography coupled with electrospray ionization-quadrupole-time of flight-mass spectrometry (HPLC-ESI-qTOF-MS/MS). Then, the quality of ten MLs varieties collected from a same plantation was assessed according to integrated HPLC fingerprinting coupled with multivariate analysis. The results revealed that Cui Yu (S5) showed the highest TPC/TFC and the strongest bio-activity, followed by Tai Long (S7) and Hong Bao Shi (S3). Among different HPLC fingerprinting, twenty compounds were selected as common characteristic peaks, and the similarity was within the range of 0.792-0.995. Meanwhile, these varieties were divided into three groups: G1 (S3, S5, S7, and S10), G2 (S1 and S4) and G3 (S2, S6, S8, and S9). Two discriminant functions with the discriminant rate near 100 % were constructed. Additionally, neomangiferin, mangiferin, kaempferol-3-O-rutinoside, isoquercitrin and quercetin were found to be the key compounds in quality evaluation of MLs varieties. Pearson correlation coefficient analysis results confirmed that these key compounds directly contributed to the antioxidant activity and α-glucosidase inhibitory ability of MLs. Importantly, the possible inhibitory mechanisms of these key compounds against α-glucosidase were preliminary clarified by in silico analysis, and the analysis results provide a theoretical basis for future development and utilization of mango leaves byproducts.