Premium
Object‐oriented approach to drug design enabled by NMR SOLVE: First real‐time structural tool for characterizing protein–ligand interactions
Author(s) -
Sem Daniel S.,
Yu Lin,
Coutts Stephen M.,
Jack Rick
Publication year - 2001
Publication title -
journal of cellular biochemistry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.028
H-Index - 165
eISSN - 1097-4644
pISSN - 0730-2312
DOI - 10.1002/jcb.10070
Subject(s) - structural genomics , computational biology , genomics , computer science , drug discovery , protein data bank , functional genomics , protein design , protein structure , bioinformatics , biology , gene , genetics , biochemistry , genome
As a result of genomics efforts, the number of protein drug targets is expected to increase by an order of magnitude. Functional genomics efforts are identifying these targets, while structural genomics efforts are determining structures for many of them. However, there is a significant gap in going from structural information for a protein target to a high affinity (K d < 100 nM) inhibitor, and the problem is multiplied by the sheer number of new targets now available. nature frequently designs proteins in classes that are related by the reuse, through gene duplication events, of cofactor binding domains. This reuse of functional domains is an efficient way to build related proteins in that it is object‐oriented. There is a growing realization that the most efficient drug design strategies for attacking the mass of targets coming from genomics efforts will be systems‐based approaches that attack groups of related proteins in parallel. We propose that the most effective drug design strategy will be one that parallels the object‐oriented manner by which nature designed the gene families themselves. IOPE (Integrated Object‐Oriented PharmacoEngineering) is such an approach. It is a three‐step technology to build focused combinatorial libraries of potential inhibitors for major families and sub‐families of enzymes, using cogent NMR data derived from representatives of these protein families. The NMR SOLVE (Structurally Oriented Library Valency Engineering) data used to design these libraries are gathered in days, and data can be obtained for large proteins (> 170 kDa). Furthermore, the process is fully object‐oriented in that once a given bi‐ligand is identified for a target, potency is retained if different cofactor mimics are swapped. This gives the drug design process maximum flexibility, allowing for the more facile transition from in vitro potency to in vivo efficacy. J. Cell. Biochem. Suppl. 37: 99–105, 2001. © 2002 Wiley‐Liss, Inc.