A Computational Study of Histamine H1-Receptor Agonist Activity Using QSPR and Molecular Surface Electrostatic Potential

Abstract

The activity of 12 human and guinea pig histamine H1-receptor agonists are investigated computationally. The mapped electrostatic potentials on the molecular surfaces of each of the structures are analyzed at HF/3-21G level of theory to identify any common features that possibly relate to their subsequent agonistic activities. Statistically derived quantities including potential’s minima VS, min, maxima VS, max, molecular volume, surface area, εHOMO, and εLUMO energies were computed. Analysis concluded that both Vmax, and εHOMO are of prime significance in the agonist activity. For the same set of agonists, hundreds of calculated descriptors at HF/3-21G* level of theory using CODESSA package followed by analysis of the multiple linear regression techniques, where several models of three to five parameters were produced. All QSPR models were investigated using full cross-validated R2 cv and leave one-out cross-validation. Among the attained models of correlation is a four-and five-parameter equations with R2= 0.995 and 0.999, and 0.974 and 0.993, R2 cv= 0.974 and 0.993, and 0.933 and 0.986 for human and guinea pig histamine H1-receptor agonists respectively. The CODESSA-based models can be useful for the prediction of the activity of the histamine H1-agonists.