Department of Pharmacy, Federal University of Juiz de Fora (UFJF), Campus Governador Valadares, MG, Brazil.
Department of Pharmaceutical Sciences, Federal University of Santa Catarina (UFSC), Florianópolis, SC, Brazil.
J Mol Graph Model. 2014 Mar;48:70-9. doi: 10.1016/j.jmgm.2013.12.004. Epub 2013 Dec 12.
This article describes structure-activity relationship (SAR/QSAR) studies on the cytotoxic activity in a human lung adenocarcinoma cell line (A549) of 43 cucurbitacin derivatives. Modeling was performed using the methods partial least squares with discriminant analysis (PLS-DA) and PLS. For both studies, the variables were selected using the ordered predictor selection (OPS) algorithm. The SAR study demonstrated that the presence or absence of cytotoxic activity of the cucurbitacins could be described using information derived from their chemical structures. The QSAR study displayed suitable internal and external predictivity, and the selected descriptors indicated that the observed activity might be related to electrophilic attack on cellular structures or genetic material. This study provides improves the understanding of the cytotoxic activity of cucurbitacins and could be used to propose new cytotoxic agents.
本文描述了 43 种葫芦素衍生物对人肺腺癌细胞系（A549）细胞毒性活性的构效关系（SAR/QSAR）研究。采用偏最小二乘判别分析（PLS-DA）和偏最小二乘（PLS）方法进行建模。对于这两项研究，变量均使用有序预测器选择（OPS）算法进行选择。SAR 研究表明，可以使用源自葫芦素化学结构的信息来描述葫芦素的存在或不存在细胞毒性活性。QSAR 研究显示出了良好的内部和外部预测能力，所选描述符表明观察到的活性可能与细胞结构或遗传物质的亲电攻击有关。这项研究提高了对葫芦素细胞毒性活性的理解，并可用于提出新的细胞毒性剂。
