Author/Authors :
Daniel S Sem، نويسنده , , Bonnie Bertolaet، نويسنده , , Brian Baker، نويسنده , , Edcon Chang، نويسنده , , Aurora D Costache، نويسنده , , Stephen Coutts، نويسنده , , Qing Dong، نويسنده , , Mark Hansen، نويسنده , , Victor Hong، نويسنده , , Xuemei Huang، نويسنده , , Richard M. Jack، نويسنده , , Richard Kho، نويسنده , , Henk Lang، نويسنده , , Chen-Ting Ma، نويسنده , , David Meininger، نويسنده , , Maurizio Pellecchia، نويسنده , , Fabrice Pierre، نويسنده , , Hug، نويسنده ,
Abstract :
Genomics-driven growth in the number of enzymes of unknown function has created a need for better strategies to characterize them. Since enzyme inhibitors have traditionally served this purpose, we present here an efficient systems-based inhibitor design strategy, enabled by bioinformatic and NMR structural developments. First, we parse the oxidoreductase gene family into structural subfamilies termed pharmacofamilies, which share pharmacophore features in their cofactor binding sites. Then we identify a ligand for this site and use NMR-based binding site mapping (NMR SOLVE) to determine where to extend a combinatorial library, such that diversity elements are directed into the adjacent substrate site. The cofactor mimic is reused in the library in a manner that parallels the reuse of cofactor domains in the oxidoreductase gene family. A library designed in this manner yielded specific inhibitors for multiple oxidoreductases.
