Ultrasound-assisted aqueous two-phase extraction of polyphenols from Terminalia chebula fruit: ANN-GA modeling, chemical characterization, and antioxidant evaluation
摘要
This study developed an ultrasound-assisted aqueous two-phase extraction (UAE-ATPE) system coupled with AB-8 macroporous resin purification for the recovery and characterization of polyphenols from Terminalia chebula fruit (TCP). The extraction process was modeled and optimized using response surface methodology (RSM) and artificial neural network-genetic algorithm (ANN-GA) approaches, with the ANN-GA model demonstrating superior predictive performance and fitting accuracy. Under ANN-optimized conditions, the TCP yield reached 40.39 ± 0.25%, exceeding that obtained with conventional 60% ethanol extraction (34.39 ± 0.38%). Purification using AB-8 macroporous resin increased the polyphenol purity from 39.69 ± 0.53% to 84.04 ± 0.49%. UPLC-MS/MS and HPLC analyses enabled polyphenol characterization and profiling, identifying and quantifying sixteen polyphenolic compounds, with chebulinic acid (88.92 mg/g) and chebulagic acid (56.10 mg/g) as the predominant constituents. The purified extract exhibited strong antioxidant properties, including reducing power, inhibition of lipid peroxidation (IC50 = 21.40 ± 0.45 µg/mL), and high oxygen radical absorbance capacity (ORAC = 4.78 ± 0.16 µmol TE/mg). SEM analysis revealed that ultrasonic treatment induced substantial disruption of the plant microstructure, facilitating polyphenol release and contributing to the enhanced antioxidant functionality of the purified extract. These findings demonstrate that the integrated UAE-ATPE system is an effective and sustainable strategy for the recovery, purification, and characterization of antioxidant polyphenols from T. chebula fruit.