Predicting pharmacokinetics remains a critical challenge to successful drug development. One of the key factors involved in determining the absorption and distribution of a compound is the strength of its binding to human serum albumin (HSA). Aiming to improve our understanding of the complex relationship between compound structure, duration of action, and ADME profile, we describe a novel structure-based model of albumin binding. Our approach incorporates the protein and ligand flexibility as well as the multi-ligand transportation inherent to HSA. The docking results suggest sensitivity to induced fit effects and structural analysis illustrates the importance of estimating ternary binding complexes. Unlike traditional methods for estimating HSA binding, such as QSPR, our technique captures the structural implications of cooperative or competitive ligand binding to HSA. We successfully utilize our model to predict serum binding (AUC = 0.85) for 80 diverse drugs and drug-like compounds with experimental data for serum binding.

Structure-based model for predicting serum albumin binding

244th National Meeting (2012)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.