Optimization of conventional extraction and multivariate metabolite profiling of soybean seeds (Glycinemax L. Merrill) for enhanced isoflavones and antioxidant properties.

This study presents a comprehensive investigation of fluctuations in isoflavone contents and antioxidant properties of soybean seeds under various conventional extraction parameters, including solvent, time, temperature, and pH. Both individual and total isoflavones varied considerably throughout different extraction conditions. The optimal method was established using 50% methanol (pH 7) at 50 °C for 30 h, yielding 3078.4 μg/g, with the following composition: glucosides (1559.1 μg/g, 50.6%) > malonylglucosides (1426.3 μg/g, 46.3%) > acetylglucosides (86.6 μg/g, 2.8%) > aglycones (6.4 μg/g, 0.2%). Moreover, multivariate analyses (PCA and heatmap) confirmed that solvent and temperature predominantly influenced isoflavone clustering. Isoflavone glucosides showed higher levels with increasing temperature (25 → 65 °C; 414.1 → 1878.6 μg/g) and pH (1 → 7; 1138.6 → 1559.1 μg/g), whereas malonylglucosides were more abundant with increasing solvent concentrations (10 → 50% methanol; 301.6 → 2117.1 μg/g) and times (6 → 30 h; 1994.8 → 2358.3 μg/g). The optimized extract exhibited potent, dose-dependent antioxidant abilities, ranked as follows: DNA protection (91.6% at 75 μg/mL) > ABTS (95.8% at 500 μg/mL) > DPPH (53.8% at 1000 μg/mL) > FRAP (0.45 OD593nm at 2000 μg/mL). Specifically, metabolite profiling of this extract using UPLC-Q-TOF-MS/MS identified 33 metabolites (isoflavone, soyasaponin, alkaloid, lipid, terpene, and flavonoid derivatives). These findings suggest that metabolite composition plays an important role in determining the bifunctional properties of soybean, highlighting the integration of multivariate analysis and UPLC-Q-TOF-MS/MS profiling under varied conditions.