Metabolite Profiling of Psychotria malayana Jack Leaf Extracts Possessing Alfa-glucosidase Inhibitory, Antiinflammatory and Antioxidant Activities Using Metabolomics and In silico Molecular Docking Analysis

Abstract

Psychotria malayana Jack leaves (family: Rubiaceae, local name: Salung) is a traditional medicinal herb that is used to treat diabetes and has been reported to possess in vitro and in vivo antidiabetic activities. According to previous research, it was found that the water extracts of P. malayana leaf exhibited noteworthy alpha-glucosidase inhibitory (AGI) activity and toxicity, with IC50 and LC50 values of 6.75 and 252.45 µg/mL, respectively. This prompts us to optimize the extraction condition of the herb to improve its bioactivity and reduce toxicity. Thus, the objectives of this study are to optimize the extraction of this plant; to evaluate alpha-glucosidase inhibitory, anti-inflammatory, and antioxidant properties of the optimised extract; to evaluate its toxicity using a zebrafish embryo/larvae model; to identify its bioactive compounds; and to develop a multivariate calibration model that could predict the AGI activity, radical scavenging activity (DPPH), and ferric reducing antioxidant power (FRAP) activity of the optimised extract. The optimisation of the extraction was performed using response surface methodology considering three factors, e.g. ethanol concentration, extraction temperature and extraction time. These extracts were analysed subsequently using metabolomics through instrumental analysis and in vitro bioactivities evaluation. The data were statistically calculated using multivariate data analysis. The putative bioactive compounds were identified through the analysis of the LC-MSMS and GC-MS fragmentation spectra of each compound. The AGI activity of each putative compound was further confirmed using in silico molecular docking. The optimised extract was further tested for in vitro anti-inflammatory (soybean lipoxygenase inhibitory-SLOXI activity) and toxicity using zebrafish embryo/larvae model. While the FTIR-based fingerprinting was utilized for the development of the multivariate calibration model. The optimised extract showed strong AGI, SLOXI, and DPPH with IC50 values of 2.02, 4.92, 13.08 µg/mL, respectively. The extract exhibited 95.44 mmol TE/mg DW of FRAP value. The optimised extract achieved an LC50 value of 224.29 µg/mL, with therapeutic index of 111.03. LC-MS based metabolomics showed putative bioactive compounds related to AGI and antioxidant activities such as isorhamnetin, moracin M-3?-O-?-D-glucopyranoside, isoaloeresin A, procyanidin B3, and stachyose. While GC-MS-based metabolomics found 1-cyclohexene-1-carboxylic acid, D-gluconic acid, propanoic acid, myo-inositol, glutaric acid, succinic acid, and ribitol as the putative bioactive compounds. Twelve metabolites show moderate to high affinities (-4.0 to -10.0 kcal/mol) to the amino acid residues ASP352, ASP215, GLH277, ARG422, ASP69, GLN182, HID112, and ARG213 at the enzymatic protein's active region in silico. The multivariate calibration model has successfully developed based on the FTIR-fingerprint with R2 value of > 0.97 for the correlation regression between predicted and actual bioactivities. This finding shows its potential for further investigation in antidiabetic study and thus has implications for diabetes therapy in the future.