Computer-aided design and activity prediction of leucine aminopeptidase inhibitors.

Abstract

The Ligand Design (LUDI) approach has been used in order to design leucine aminopeptidase inhibitors, predict their activity and analyze their interactions with the enzyme. The investigation was based on the crystal structure of bovine lens leucine aminopeptidase (LAP) complexed with its inhibitor--the phosphonic acid analogue of leucine (LeuP). More than 50 potential leucine aminopeptidase inhibitors have been obtained, including the most potent aminophosphonic LAP inhibitors with experimentally known activity, which have been the subject of more detailed studies. A reasonable agreement between theoretical and experimental activities has been obtained for most of the studied inhibitors. Our results confirm that LUDI is a powerful tool for the design of enzyme inhibitors as well as in the prediction of their activity. In addition, for inhibitor-active site interactions dominated by the electrostatic effects it is possible to improve binding energy estimates by using a more accurate description of inhibitor charge distribution.