Multivariate image analysis applied to quantitative structure-activity relationships and docking studies of recent hydroxyphenylpyruvate deoxygenase inhibitors

Abstract

Background: Mesotrione is a triketone widely used as an inhibitor of the hydroxyphenylpyruvate deoxygenase (HPPD) enzyme.However, new agrochemicals should be developed continuously to tackle the problem of herbicide resistance. Two sets ofmesotrione analogs have been synthesized recently and they have demonstrated successful phytotoxicity against weeds. Inthis study, these compounds were joined to form a single data set and the HPPD inhibition of this enlarged library of triketoneswas modeled using multivariate image analysis applied to quantitative structure–activity relationships (MIA-QSAR). Dockingstudies were also carried out to validate the MIA-QSARfindings and to aid the interpretation of ligand-enzyme interactionsresponsible for the bioactivity (pIC50).Results: The MIA-QSAR models based on van der Waals radii (rvdW), electronegativity (ε), and thervdW/εratio as moleculardescriptors were both predictive to an acceptable degree (r2≥0.80,q2≥0.68 andr2pred≥0.68). Subsequently, partial leastsquares (PLS) regression parameters were applied to predict the pIC50values of newly proposed derivatives, yielding a fewpromising agrochemical candidates. The calculated logPfor most of these derivatives was found to be higher than that ofmesotrione and the library compounds, indicating that they should be less prone to leach out and contaminate groundwater.Conclusion: Multivariate image analysis descriptors corroborated by docking studies were capable of modeling the herbicidalactivities of 68 triketones reliably. Due to the substituent effects at the triketone framework, particularly of a nitro group inR3,promising analogs could be designed. TheP9proposal demonstrated higher calculated activity and logPthan commercialmesotrione.