Computational modeling of distribution coefficients and their correlations with pharmacokinetic properties of 17α-picolyl and 17(E)-picolinylidene androstane derivatives

Abstract

The present study describes the computational modeling of distribution coefficient (logD) of 17α-picolyl and 17(E)-picolinylidene androstane derivatives, as a group of compounds with significant anticancer activities. The determination of logD is practically important for estimation and prediction of pharmacokinetic and pharmacodynamic behavior of compounds in a living organism and it is related to blood-brain barrier permeability, skin permeability, gastrointestinal absorption, binding to plasma proteins, etc. These features are crucial for the determination of potential drug candidates as well. The results presented in this study include determination of pH versus logD profiles, pH versus molecular charge profiles and determination of isoelectric point of eleven 17αpicolyl androstane derivatives and thirteen 17(E)-picolinylidene androstane derivatives. Since the pH of the organism differs depending on the organ (for example, the pH of the blood is significantly different from the pH of the stomach), these profiles are significant because they indicate in what form the molecule will exist and how it will be distributed between different phases at certain pH value. The influence of tautomerization and resonance was taken into account during the modeling of logD parameters. Eventually, the correlations between logD values and specific absorption, distribution, metabolism, and excretion (ADME) properties, such as human intestinal absorption (HIA) and permeability of heterogeneous human epithelial colorectal adenocarcinoma cells (Caco-2), were determined.