Researchers have reported a powerful new screen for identifying weak binders to specific proteins. The new technique, described by researchers at Sunesis Pharmaceuticals (Redwood City, CA) and the University of California (San Francisco, CA), should be especially useful for identifying starting fragments for rational drug design. Standard screening approaches for small-molecule inhibitors are limited to identifying compounds with relatively high affinities for the target protein, and do not allow selection for specific target sites. In the new procedure (PNAS 97, 9367–9372, 2000), the disulfide-containing compounds being screened can form a covalent linkage with the protein. A cysteine residue in the protein will “tether” any compound that has even a moderate affinity for the nearby active site, allowing the compound to be identified by mass spectrometry. The team demonstrated the concept by identifying an inhibitor of thymidylate synthase. After identifying an inhibitor from the initial library, the scientists improved its affinity by several orders of magnitude through rational design. James Wells, senior author on the paper, says extending the approach to a wide range of targets should be straightforward: “Enzyme targets are very exciting, and we're also going after protein–protein targets, which have been virtually intractable to drug screening.”