Membrane-interaction QSAR analysis: application to the estimation of eye irritation by organic compounds.

Abstract

The purpose of this study was to explore a potential mechanism of eye irritation, and to construct a corresponding general quantitative structure-activity relationship (QSAR) model, in terms of diversity of irritant chemical structure, based on the Draize eye irritation ECETOC data set. Molecular dynamic simulation (MDS) was used to generate intermolecular membrane-solute interaction properties. These intermolecular properties were combined with intramolecular physicochemical properties and features of the solute (irritant) to construct QSAR models using multi-dimensional linear regression and the Genetic Function Approximation (GFA) algorithm. Significant QSAR models for estimating eye irritation potential were constructed in which solute aqueous solvation free energy and solute-membrane interaction energies are the principle correlation descriptors. These physicochemical descriptors were selected from a trial set of 95 descriptors for 18 structurally diverse compounds fully representative of the ECETOC set of 38 compounds. Combining intermolecular solute-membrane interaction descriptors with intramolecular solute descriptors yields statistically significant eye irritation QSAR models. The resultant QSAR models support an eye irritation mechanism of the action in which increased aqueous solubility of the irritant and its strength of binding to the membrane both increase eye irritation.