Jang Mi Han (Inje University)

Miran Jang (Inje University)

Young Beom Kwak (Inje University)

Jang Mi Han (Inje University)

Opeyemi. O. Deji-Oloruntoba (Inje University)

Miran Jang (Inje University)

Young Beom Kwak (Inje University)

Recently, natural compounds with antioxidant and lipid-lowering properties have attracted considerable attention as functional ingredients for preventing oxidative stress-included obesity and metabolic disorders. In this study, 8 underutilized medicinal plants were initially screened using TPC, DPPH, ABTS, and FRAP assays, and the three most active extracts, Senna siamea (SS), Amylotheca duthieana (AD), and Ternstroemia wallichiana (TW), were selected for further investigation. Untargeted phytochemical profiling was performed by GC-MS in EI scan mode, followed by spectral deconvolution using AMDIS. Candidate compounds were tentatively identified with a minimum spectral match of 80 % against the MoNA-export-HMDB and FiehnRI libraries. Major phenolic compounds detected included 2,4-di-tert-butylphenol, 3,3’,4’5-tetrahydroxystilbene, and catechin 3-O-gallate in SS; qunic acid, apigenin-7-O-glucoside, myricetin, and gallic acid in AD; and protocatechuic acid, 4-hydroxyphenylacetic acid, and catechin 3-O-gallate in TW. These phytochemical profiles suggested the presence of bioactive metabolites associated with antioxidant activity. In vivo evaluation using Caenorhabditis elegans further demonstrated that all three extracts significantly reduced ROS production and lipid accumulation under high-glucose conditions. Mutant studies confirmed that these protective effects were mediated through the conserved skn-1/Nrf2 and daf-16/FOXO signaling pathways. Overall, this study demonstrates that the integrated GC-MS/AMDIS platform is an effective strategy for characterizing bioactive metabolites in complex plant extracts and provides valuable chemical markers for the development of functional botanical therapeutics targeting metabolic disorders.

​