In silico prediction of human serum albumin binding for drug leads

Abstract

Introduction: Binding of drugs to human serum albumin (HSA) strongly influences their pharmacokinetic behavior and is associated with drug safety issues, low clearance, low brain penetration, as well as drug-drug interactions. Thus, in silico prediction of HSA binding contributes significantly to the discovery of new drug candidates.

Areas covered: The authors provide a short overview on the principles of HSA binding and the crystal structure of HSA, as well as discussing and analyzing the recent structure- and ligand-based HSA binding models. The authors also present the advantages and limitations of each methodology to construct efficient local or global models and outline the critical structural features contributing to HSA.

Expert opinion: The in silico estimation of drug binding to HSA in early drug discovery contributes to the lead optimization process. Local models are useful for the design of new compounds with reduced HSA binding for a particular target receptor, while real-time quantitative structure-activity relationships or global models combining structure- and ligand-based approaches serve for compound libraries screening. However, research efforts on other important plasma proteins should be strengthened in the perspective to enable predictions of total plasma protein binding for clinical candidates.

Keywords::

• HSA binding
• HSA crystal structures
• ligand-based modeling
• property-based drug design
• QSAR methodology
• structure-based modeling

Notes

This box summarizes key points contained in the article.