Discovery of Antibacterial Biotin Carboxylase Inhibitors by Virtual Screening and Fragment-Based Approaches.
Mochalkin, I., Miller, J.R., Narasimhan, L.S., Thanabal, V., Erdman, P., Cox, P., Prasad, J.V., Lightle, S., Huband, M., Stover, K.(2009) ACS Chem Biol 4: 473
- PubMed: 19413326 
- DOI: https://doi.org/10.1021/cb9000102
- Primary Citation of Related Structures:  
2W6M, 2W6N, 2W6O, 2W6P, 2W6Q, 2W6Z, 2W70, 2W71 - PubMed Abstract: 
As part of our effort to inhibit bacterial fatty acid biosynthesis through the recently validated target biotin carboxylase, we employed a unique combination of two emergent lead discovery strategies. We used both de novo fragment-based drug discovery and virtual screening, which employs 3D shape and electrostatic property similarity searching. We screened a collection of unbiased low-molecular-weight molecules and identified a structurally diverse collection of weak-binding but ligand-efficient fragments as potential building blocks for biotin carboxylase ATP-competitive inhibitors. Through iterative cycles of structure-based drug design relying on successive fragment costructures, we improved the potency of the initial hits by up to 3000-fold while maintaining their ligand-efficiency and desirable physicochemical properties. In one example, hit-expansion efforts resulted in a series of amino-oxazoles with antibacterial activity. These results successfully demonstrate that virtual screening approaches can substantially augment fragment-based screening approaches to identify novel antibacterial agents.
Organizational Affiliation: 
Pfizer, Inc., Michigan Laboratories, Ann Arbor, Michigan 48105, USA. [email protected]