Optimization of Stropharia rugosoannulata extracts using artificial intelligence and response surface methodology and comparative evaluation of their biological activities

Abstract

Stropharia rugosoannulata Farl. ex Murrill is an edible mushroom species recognized for its nutritional value and potential functional properties. In this study, extraction conditions were optimized to maximize the biological potential of its extracts. Ultrasonically assisted extraction parameters, including temperature, extraction time, and ethanol/water ratio, were optimized using two advanced modeling approaches: Response Surface Methodology (RSM) and Artificial Neural Network–Genetic Algorithm (ANN-GA). The biological activities of the optimized extracts were evaluated through antioxidant assays (TAS, TOS, OSI, DPPH, and FRAP), cholinesterase inhibition tests (AChE and BChE), and antiproliferative activity against A549, MCF-7, and DU-145 cancer cell lines. Chemical characterization of the extracts was performed by LC-MS/MS for phenolic compounds and by GC-MS for volatile and semi-volatile constituents. The results indicated that the RSM optimized extract exhibited stronger biological performance than the ANN-GA extract, including higher antioxidant capacity, more pronounced cholinesterase inhibition, and greater antiproliferative activity. LC-MS/MS analysis revealed higher concentrations of major phenolic compounds such as gallic acid, caffeic acid, protocatechuic acid, and quercetin in the RSM extract, while GC-MS analysis showed differences in the distribution of terpenoid and fatty acid-derived compounds between the extracts. These findings demonstrate that extraction parameters significantly influence both the chemical composition and biological activity of mushroom extracts and highlight S. rugosoannulata as a promising natural source of bioactive compounds for functional food and natural product development.