A structural-feature-based computational approach for toxicity prediction of water-soluble arsenicals

Abstract

We have used the density functional theory to make a toxicity prediction model of water-soluble arsenicals (WSA). The structures have been optimised for the minimum energy of the Schrödinger equation. In the present work, the usefulness of electrophilicity and charge transfer in predicting the toxicity of WSA, namely, monomethylarsenic acid (MMA) (III), dimethylarsenic acid (DMA) (III), arsenic acid, arsenous acid, MMA (V) and DMA (V), is assessed. It is demonstrated that the toxicity of arsenicals (both electron donors and acceptors) in gas and solution phases can be adequately explained in terms of electrophilicities. Amount of charge transfer between the WSA and the biosystem, simulated as nucleic acid (NA) bases and DNA base pairs, indicates the importance of charge transfer in experimental toxicity. Interaction of arsenicals with NA bases/selected base pairs is determined using Paar's charge-transfer formula. The experimental toxicity (LD50) of arsenicals are taken as dependent variables and the energy (E) along with DFT-based descriptors, namely, electrophilicity index (ω) and charge transfer (ΔN), are taken as independent variables. Fairly good correlation is obtained showing the significance of the selected descriptors on arsenicals that act as electron donors or acceptors in the presence of bio-molecules. Raman micro-spectroscopy also showed its potential applications in the toxicology screening of chemicals and new biomaterials, with a range of cell types.