Antioxidant effects of polyphenolic compounds and structure-activity relationship predicted by multivariate regression tree

Suzana C. Santos,Gilmara A.C. Fortes,Lilian T.F.M. Camargo,Ademir J. Camargo,Pedro H. Ferri
DOI: https://doi.org/10.1016/j.lwt.2020.110366
IF: 6.056
2021-02-01
LWT
Abstract:<p>Phenolic antioxidants form a class of compounds which inhibit the oxidation of materials of biological importance, playing a preventive role in a number of degenerative diseases. Polyphenol activity depends on their chemical structure, which is influenced by factors such as number and position of phenolic hydroxyl groups, steric effects, and molecular properties. In this research study, 18 polyphenol structures (13 hydrolysable tannins, 4 flavonoids and ellagic acid) were modelled and their molecular properties (volume, area, dipole moment, polarisability, Log<em>P</em>, E<sub>HOMO</sub>, E<sub>LUMO</sub>, and E<sub>LUMO</sub>–E<sub>HOMO</sub> gap) calculated by density functional theory (DFT), using the M062X/6-311+G(d,p) level of theory. Multivariate regression tree (MRT) and canonical redundancy analysis (RDA) were applied to correlate the experimental antioxidant activities obtained by ORAC and DPPH assays and molecular and lipophilic parameters. MRT/RDA split all 18 polyphenols into four clusters with increasing antioxidant capacity. Flavonoids and galloyl esters belong to clusters C1-C3, while cluster C4 is composed mainly of monomeric and dimeric ellagitannins. The RDA1 axis shows a positive correlation between antioxidant activity and parameters polarisability, E<sub>HOMO</sub>, and Log<em>P</em>. These molecular properties proved to be the most suitable for predicting the relationship between chemical structure and antioxidant capacity for this group of phenolic compounds.</p>
What problem does this paper attempt to address?