Department of Forest Sciences, and ‡Department of Chemistry, Federal University of Lavras , 37200-000 Lavras, Minas Gerais (MG), Brazil.
J Agric Food Chem. 2013 Sep 11;61(36):8499-503. doi: 10.1021/jf4024257. Epub 2013 Aug 26.
Two of major weeds affecting cereal crops worldwide are Avena fatua L. (wild oat) and Lolium rigidum Gaud. (rigid ryegrass). Thus, development of new herbicides against these weeds is required; in line with this, benzoxazinones, their degradation products, and analogues have been shown to be important allelochemicals and natural herbicides. Despite earlier structure-activity studies demonstrating that hydrophobicity (log P) of aminophenoxazines correlates to phytotoxicity, our findings for a series of benzoxazinone derivatives do not show any relationship between phytotoxicity and log P nor with other two usual molecular descriptors. On the other hand, a quantitative structure-activity relationship (QSAR) analysis based on molecular graphs representing structural shape, atomic sizes, and colors to encode other atomic properties performed very accurately for the prediction of phytotoxicities of these compounds against wild oat and rigid ryegrass. Therefore, these QSAR models can be used to estimate the phytotoxicity of new congeners of benzoxazinone herbicides toward A. fatua L. and L. rigidum Gaud.
两种主要影响谷类作物的杂草是野燕麦（Avena fatua L.）和硬草（Lolium rigidum Gaud.）。因此，需要开发针对这些杂草的新型除草剂；在这方面，苯并恶嗪酮、它们的降解产物和类似物已被证明是重要的化感物质和天然除草剂。尽管早期的构效关系研究表明，氨基苯并恶嗪酮的疏水性（log P）与生物毒性相关，但我们对一系列苯并恶嗪酮衍生物的研究结果表明，生物毒性与 log P 之间没有关系，也与其他两个常用的分子描述符没有关系。另一方面，基于分子图的定量构效关系（QSAR）分析，这些分子图代表结构形状、原子大小和颜色，用于编码其他原子特性，对这些化合物对野燕麦和硬草的生物毒性进行了非常准确的预测。因此，这些 QSAR 模型可用于估计苯并恶嗪酮类除草剂对野燕麦和硬草的新同系物的生物毒性。
