Exploring predictive QSAR models for hepatocyte toxicity of phenols using QTMS descriptors Kunal Roy * ,  and Paul L. A. Popelier * Manchester Interdisciplinary Biocenter (MIB), 131 Princess Street, Manchester, M1 7DN, United Kingdom Received 5 February 2008; revised 12 March 2008; accepted 12 March 2008 Available online 16 March 2008 Abstract—We construct predictive QSAR models for hepatocyte toxicity data of phenols using Quantum Topological Molecular Similarity (QTMS) descriptors along with hydrophobicity (log P) as predictor variables. The QTMS descriptors were calculated at different levels of theory including AM1, HF/3-21G(d), HF/6-31G(d), B3LYP/6-31+G(d,p), B3LYP/6-311+G(2d,p) and MP2/ 6-311+G(2d,p). The external predictability of the best models at the higher levels of theory is higher than that at the lower levels. Moreover, the best QTMS models are better in external predictability than the PLS models using pK a and Hammett r + along with log P. The current study implies the advantage of quantum chemically derived descriptors over physicochemical (experimentally derived or tabular) electronic descriptors in QSAR studies. Ó 2008 Elsevier Ltd. All rights reserved. Phenols are widely distributed in edible plants and found in tea, fruits and vegetables. 1,2 Phenols are also used in many industries as intermediates or as biocides themselves. 3 They occur as industrial wastes and being relatively soluble in water and detectable in rivers, ponds, and soil, they are direct pollutants in the environ- ment. They can also be produced by environmental deg- radation of more complex molecules. 4 Phenols are supposed to be the most toxic water pollutant as they are carcinogenic in nature. 5,6 Because of their prevalence in the environment, human exposure to phenols is ubiq- uitous, and thus there is much interest in determining their potential hazard. Although the ability of polyphe- nols to protect cell from the oxidative stress has been demonstrated, there is increasing evidence of their pro- oxidant cytotoxicity. 7 The toxicity of phenols is confus- ing as the same polyphenol compounds can behave both as antioxidants and prooxidants, depending on concen- tration and free radical source. 8–11 Exposure of mamma- lian cells to polyphenols is accompanied by an increase in intracellular reactive oxygen species levels. 12 The mechanism underlying the toxicity of phenols is mainly related to lipophilicity and electrophilic effects. 13–15 Phenols are important in nutrition and medicine given their cytotoxic potential. Second, they are prevalent in the environment and are likely to elicit often unknown ecotoxic effects. Hence, there is much interest recently in applying quantitative structure–activity relationships (QSARs) to predict the toxic potential of phenolic com- pounds. Liu et al. have explored a QSAR for toxicity of chlorophenols on L929 cells. 16 Padmanavan et al. have used group philicity and electrophilicity as possible descriptors for modelling ecotoxicity applied to chlor- ophenols. 3 Rule-based ensemble modelling was applied to develop a model with predictive capability for distin- guishing between four different modes of toxic action for a set of 220 phenols by Norinder et al. 17 A three-dimen- sional QSAR study has been reported for in-vitro toxic- ity of chlorophenols to HepG2 cells. 18 Aptula et al. have modelled toxicity of di- and trihydroxybenzenes to Tet- rahymena pyriformis using hydrophobilicity and electro- philicity indices. 15 Non-linear QSAR modelling of the toxicity of phenol derivatives to Tetrahymena pyriformis has been reported by Devillers. 19 Schu ¨u ¨rmann et al. re- ported stepwise discrimination between four modes of toxic action of phenols in the Tetrahymena pyriformis assay. 20 Determination of mechanisms of toxic action of phenols to Tetrahymena pyriformis by QSAR 0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.bmcl.2008.03.035 Keywords: QTMS; Toxicity; Ab initio; Phenols; QSAR; External val- idation; Electron density; Atoms in molecules; Quantum chemical topology. * Corresponding authors. Tel.: +91 98315 94140; fax: +91 33 2837 1078 (K.R.); tel.: +44 161 3064511; fax: +44 161 3065201 (P.L.A.P.); e-mail addresses: kroy@pharma.jdvu.ac.in; pla@manchester.ac.uk   Currently on leave from Drug Theoretics and Cheminformatics Lab, Department of Pharmaceutical Technology, Jadavpur University, Kolkata 700 032, India. Available online at www.sciencedirect.com Bioorganic & Medicinal Chemistry Letters 18 (2008) 2604–2609