Article Preview
TopIntroduction
Agricultural weed management continues to rely heavily on chemical control methods using herbicides (Gianessi, 2013). However, numerous studies have long demonstrated that these chemicals may produce several hazardous effects due to their persistence in the environment, which results in their incorporation in the food chain at different trophic levels, through the processes of bioconcentration and/or soil sorption (Buchanan et al., 2009). Consequently, research efforts have been entrenched in the search of more ecofriendly herbicides with a reduced propensity to bioaccumulate in aquatic organisms or persist in soils or sediments (Freitas, Barigye, Daré, & Freitas, 2016; Saxena & Pandey, 2001; Varejão, Demuner, Barbosa, & Barreto, 2013). On the other hand, weed species have continuously adapted and thus developing resistance to commonly used herbicides and thus increasing the pressure on the chemical industry to develop new chemicals with strong herbicidal activity (Renton, Busi, Neve, Thornby, & Vila-Aiub, 2014).
A deep analysis of the reports in the literature shows that these two objectives have been pursued separately (Bettiol et al., 2016; Clark, 2012; Ghose & Crippen, 1987; Larif et al., 2013; Liu & Qian, 1995; Morita, Nagare, & Hayashi, 1987; Varejão et al., 2014). However, seeking to reduce the bioconcentration and/or soil sorption tendency of chemicals may result in the loss/lower herbicidal activity and vice-versa as chemical structural properties are not necessarily orthogonal. To contextualize better this point, bioconcentration and soil sorption described by the bioconcentration factor (BCF) and the soil/water–organic carbon partition coefficient (log KOC), respectively, are known to possess a strong relationship with the physicochemical parameter n-octanol/water partition coefficient (log P) and in principle compounds with lower log P values should have a lower tendency to bioconcentrate or to interact strongly with soil particles. However, low logP compounds may just as well be ineffective inhibitors due to failure to penetrate the lipophilic plant tissue. Therefore, the search of a successful herbicide candidate should be as a result of striking a compromise between various objectives, which sometimes may unfortunately compete among each other.
In this sense, the aim of the present paper is to perform a multi-objective modeling experiment of the inhibitory activity and ecotoxicological profile of the triazine derivatives, which is a family of well-known and widely used herbicides, with the goal of understanding the structural features that favor high herbicidal activity and paralleled with a favorable environmental profile. In the modeling of the herbicidal activity, it is important that a plant species independent target is considered, to allow for a broad spectrum analysis. In this sense, the Photosynthetic Electron Transport (PET) inhibitory activity is selected as the property to be maximized while the log KOC is to be minimized.
The Quantitative Structure-Property Relationships (QSPR) studies play an important role in the prediction of molecular properties and consequently in the virtual screening of chemical compound libraries for molecular entities of interest. This methodology is centered on the principle that there exists a close relationship between the structural characteristics of chemical compounds and their chemical, physicochemical and biological properties. Several parameters for characterizing chemical structures, ranging from 0D – 6D approaches have been reported in the literature (Barigye, Marrero-Ponce, Pérez-Giménez, & Bonchev, 2014; Todeschini & Consonni, 2009). While, each of these approaches presents particular advantages and drawbacks, a tradeoff between simplicity and modeling power is largely considered as beneficial. In this sense, the multivariate image analysis applied to quantitative structure-property relationships (MIA-QSPR) method appears to be an interesting approach in molecular modeling (Barigye & Freitas, 2016).