Modelagem QSAR das atividades herbicidas de Benzamidas para controle de pragas em agricultura

Abstract

The need to use herbicides to control weeds in agricultural systems is incontestable. Benzamide herbicides consist of a class of photosynthetic system II (PSII) inhibitors widely used for weed control. However, the development of resistance by these weeds to the known herbicides requires an ongoing search for new agrochemicals. The study reported here starts from combining two congeneric series of (thio)benzamide herbicides into a single dataset and subsequent modeling of their herbicidal activities against PSII (pIC50) using multivariate image analysis applied to quantitative structure-activity relationship (MIA- QSAR). The models were robust and predictive (mean values of r2 = 0.91, q2 = 0.80 and r2pred = 0.81) and were used to estimate the pIC50 of new agrochemical candidates, which were proposed based on the mixture chemistry of the substructures of the most active compounds present in the dataset. The chemical features affecting the herbicidal activities were analyzed using MIA contour maps, whereas the predicted pIC50 values were compared to docking scores to support the MIA-QSAR outcomes. The MIA-QSAR models, whose molecular descriptors explain topochemical information, atomic radius, and Pauling´s electronegativity, presented high modeling performance. Accordingly, the pIC50 data for proposed agrochemical candidates were calculated using the PLS regression parameters of the models. The proposed compound possessing a thiobenzamide moiety and C-11 chain, H, NO2, OH, and OH as variable substituents was the most promising alternative.