Thirty-seven samples of naturally occurring phenolic compounds were evaluated using three common in vitro assays for total antioxidant activity (TAC) testing: the Trolox Equivalent Antioxidant Capacity (TEAC), the Ferric Reducing Antioxidant Potential (FRAP) and the 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging assay, in addition to the Folin-Ciocalteu reagent reactivity (FCR). We found that antioxidant hierarchies depended on the choice of assay and applied ANOVA analyses to explore underlying structure-TAC dependencies. In addition to statistically confirming the empirically established connection between flavonoid ring-B catechol and high TEAC or FRAP, new correlations were also found. In flavonoids, (i) hydroxyl groups on ring-B had a positive effect on all four TAC assays; (ii) the presence of a 3-hydroxyl group on ring-C increased TEAC and FRAP, but had no effect on DPPH or FCR; (iii) Phenolic acids lacking a 3-hydroxyl group had significantly lower FRAP or DPPH than compounds having this structure, while TEAC or FCR were not affected. Results demonstrated that any TAC-based ranking of phenolic rich samples would very much depend on the choice of assay, and argue for use of more than one technique. As an illustration, we compared results of the above four assays using either grapevine leaf extracts or synthetic mixtures of compounds prepared according to major polyphenols identified in the leaves.
Keywords: antioxidant activity assessment; antioxidants; caftaric acid; grapevine (Vitis vinifera) leaves; polyphenols; quercetin-3-glucoside; quercetin-3-glucuronide; structure-activity relationship.