|TU, YI-SHU - Ministry Of Science And Technology|
|Compton, David - Dave|
Submitted to: Structural Chemistry
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/29/2020
Publication Date: 5/15/2020
Citation: Appell, M., Tu, Y.-S., Compton, D.L., Evans, K.O., Wang, L.C. 2020. Quantitative structure-activity relationship study for prediction of antifungal properties of phenolic compounds. Structural Chemistry. https://doi.org/10.1007/s11224-020-01549-1.
Interpretive Summary: Antifungal chemicals are often used to reduce agricultural commodity spoilage and the occurrence of mycotoxins. However, there is a need for safer, better antifungal agents. Phenolic compounds are potential antifungal agents and have many uses due to their consumer-friendly properties. To aid in the development of better antifungal compounds to improve food safety, we applied computational and machine learning methods to develop mathematical models that identified chemical properties of phenolic compounds to help reduce contamination of mycotoxin producing fungi. We discovered that chemical stability and the electronic properties of the phenolic compounds are important features for successful antifungal activity against toxin producing fungi. The more potent antifungal compounds studied are the safe antioxidant essential oil components thymol and carvacrol that are associated with many plants, including the popular culinary herb thyme. These models will help toxicologists, microbiologists, and chemists discover better antifungal agents to benefit the food industry.
Technical Abstract: Antifungal compounds are of interest to reduce commodity spoilage and exposure to mycotoxins. In this study, a series of Quantitative Structure- Activity Relationship (QSAR) equations based on quantum chemical and topological properties were developed to gain insight into the antifungal activities of phenolic compounds. The molecules were geometry optimized using B3LYP/6-311+G** density functional theory calculations. Analysis of the frontier orbital properties revealed that conjugated phenolic compounds possessed smaller band gap energies. More than 750 molecular descriptors were evaluated in the QSAR analysis. Genetic Function Approximation (GFA) on populations of 100 one to three descriptor models over 10,000 generations identified several models for antifungal activity against Fusarium verticillioides, Fusarium oxysporum, and Penicillium expansum. Phenolic compounds with greater antifungal activity possessed a lower electrophilicity index. Molecular descriptors associated with electrostatic and topological properties are important to describe the antifungal activities of the phenolic compounds studied.