Comparative Evaluation of Pressurized Liquid and Conventional Solvent Extraction of Agro-Industrial By-Products: Phenolic and Flavonoid Contents, Antioxidant Capacity and Cytotoxicity Assessment

Abstract

In the present study, extracts obtained from four agro-industrial by-products, namely pomegranate peels (Punica granatum L.), turnip tops (Brassica rapa L.) residues, prickly pear (Opuntia ficus-indica L.) peels, and artichoke (Cynara cardunculus var. scolymus) stems and leaves, were produced by conventional solvent extraction (acidified methanol–water, 80:20, v/v) and water-based pressurized liquid extraction (PLE). The extracts were comparatively evaluated for total phenolic and flavonoid contents, chemical antioxidant capacity by ABTS and FRAP assays, and cytotoxicity toward Caco-2 cells. Conventional solvent extraction generally resulted in higher Folin–Ciocalteu-estimated total phenolic contents and ABTS activity, whereas flavonoid contents were comparable between the two extraction procedures. Pomegranate peel exhibited the highest values among the investigated matrices, with total phenolic contents of 141 and 193 mg GAE/g dry matter for PLE and conventional extraction, respectively and flavonoid content of approximately 100 mg QE/g dry matter for both procedures. Consistently, pomegranate peel showed the highest ABTS and FRAP responses, reaching 492 mg Trolox/g dry matter and 6032 μmoL Fe (II)/g dry matter for PLE and 618 mg Trolox/g dry matter and 8015 μmoL Fe (II)/g dry matter for conventional extraction, respectively. In contrast, PLE resulted in higher FRAP responses for all three remaining matrices, namely turnip tops, prickly pear, and artichoke by-products, with values approaching twice those obtained by conventional extraction for turnip tops and artichoke. Among these matrices, turnip tops showed the highest FRAP response, followed by prickly pear and artichoke, whereas prickly pear exhibited higher ABTS activity than turnip tops and artichoke. Overall, the non-pomegranate matrices showed substantially lower phenolic and flavonoid contents and chemical antioxidant capacity than pomegranate peel. Extracts generally showed low cytotoxic toward Caco-2 cells under the experimental conditions tested, with cell viability remaining close to or above 80% across the tested concentrations, although some matrix- and concentration-dependent effects were observed. These results demonstrate that the relative performance of conventional solvent extraction and PLE depended on both the plant matrix and the analytical endpoint: conventional extraction produced higher phenolic and ABTS responses for pomegranate peel, whereas water-based PLE yielded higher FRAP responses for the other matrices while providing comparable flavonoid contents. Thus, the findings highlight the importance of considering both matrix characteristics and the analytical endpoint when comparing extraction procedures for agro-industrial by-product valorization. The study provides a preliminary in vitro characterization of the investigated extracts, while further phytochemical and biological investigations are required to establish their potential for food-related applications.