Ultrasound-assisted extraction of naringin from Exocarpium Citri Grandis using a novel ternary natural deep eutectic solvent based on glycerol: Process optimization using ANN-GA, extraction mechanism and biological activity.

This study aimed to develop an efficient green ultrasound-assisted extraction (UAE) method for naringin (Nar) from Exocarpium Citri Grandis (ECG) using a glycerol-based ternary natural deep eutectic solvent (NADES) and explore its biofunctional relevance. After screening and single-factor optimization, the optimal NADES was identified as glycerol:malic acid:propanedioic acid (1:1:2 M ratio, 30 % water content). Extraction conditions (liquid-solid ratio, temperature, time) were optimized via response surface methodology (RSM) and an artificial neural network-genetic algorithm (ANN-GA), with ANN-GA demonstrating superior predictive capability. Optimal parameters (57 °C, 25.9 mL/g, 24.4 min) yielded a high Nar extraction yield of 175.30 ± 2.94 mg/g. Scanning electron microscopy (SEM) revealed structural changes in ECG, while extraction kinetics indicated an activation energy of 40.45 kJ/mol and molecular dynamics simulations elucidated improved solvation mechanisms. NADES was reusable for at least three cycles before efficiency decreased significantly at the fourth reuse. Furthermore, the extracted Nar showed significant inhibitory activity against key enzymes, with IC values of 19.6 μg/mL (α-glucosidase), 2.73 mg/mL (pancreatic lipase), and 1,086 μg/mL (acetylcholinesterase). These results indicate its potential for hypoglycemic, hypolipidemic, and anti-Alzheimer's therapeutic effects. This work establishes a sustainable strategy for efficient Nar extraction, integrating green solvent technology with bioactivity assessment to advance its therapeutic potential.