Structure-based focusing using pharmacophores derived from the active site of 17beta-hydroxysteroid dehydrogenase.

Structure-based focusing constitutes a powerful approach to design libraries of compounds with a given biological profile. Computers with special software can be used to analyse the large amount of data usually available for the compounds. Pharmacophores can be used to identify new compounds that present a specific arrangement of features responsible for a certain type of activity. When available, information about the 3D structure of a biological target can also be included in the building of pharmacophore models. These pharmacophores can then be used as queries to search and/or focus large compound libraries. Multiple pharmacophores were generated from the 3D structure 17beta-hydroxystreoid dehydrogenase type1 complexed with different inhibitors. The validity of these pharmacophores was assessed against a test database containing known active and inactive 17beta-hydroxystreoid dehydrogenase type1 inhibitors. The most selective models were then used to search commercial databases for new structural lead molecules. This approach has allowed us to identify a few new compounds possessing structural features common to flavonoids, a structural class of compounds known to contain good inhibitors of 17beta-hydroxystreoid dehydrogenase type1 enzyme. A structure-based focusing approach is demonstrated to be a meaningful and powerful technique for identifying new lead candidates, which can be taken into the lead optimization process.