Predicting the mutagenic potential of chemicals in tobacco products using in silico toxicology tools

Abstract

Tobacco products contain thousands of chemicals, including addictive and toxic chemicals. We utilized in silico toxicology tools to predict in a validation test and in a separate screening test, the mutagenic potential of chemicals reported in tobacco products and tobacco smoke. Different publicly available (quantitative) structure-activity relationship (Q)SAR software platforms were used in this study. The models were validated against 900 chemicals relevant to tobacco for which experimental Ames mutagenicity data are available from public sources. The predictive performance of the individual and combined (Q)SAR models was evaluated using various performance metrics. All the (Q)SAR models represented >95% of the tobacco chemical space indicating a high potential for screening tobacco products. All the models performed well and predicted mutagens and nonmutagens with 75-95% accuracy, 66-94% sensitivity and 73-97% specificity. Subsequently, in a screening test, a combination of complementary SAR-based and QSAR-based models was used to predict the mutagenicity of 6820 chemicals catalogued in tobacco products and/or tobacco smoke. More than 1200 chemicals identified in tobacco products are predicted to have mutagenic potential, with 900 potential mutagens in tobacco smoke. This research demonstrates the validity of in silico (Q)SAR tools to make mutagenicity predictions for chemicals in tobacco products and/or tobacco smoke, and suggest they hold utility as screening tools for hazard identification to inform tobacco regulatory science.

Keywords: In silico toxicology; QSAR; SAR; mutagenicity; tobacco products.