Discovering high-affinity ligands for proteins.

Drug discovery is a difficult and time-consuming process that involves the identification of potent therapeutic | TECH VI EW I agents with good bioavailability, metabolic stability, and low toxicity. An important first step in this process is the identification of molecules that bind with high affinity and selectivity to the biological target. High-affinity ligands might be discovered by screening natural products or large chemical libraries. However, molecules may not exist that have functional groups in the correct spatial orientation to interact optimally with the binding site. To increase the chances of finding such a molecule, many compounds could be synthesized using recently developed approaches in combinatorial chemistry and parallel synthesis (1). Yet it is often difficult to develop the synthetic procedures required to assemble many different fragments with a variety of linkers. This may ultimately limit the diversity of the compounds that can be synthesized. Furthermore, the synthesis and testing of large numbers of molecules is both costly and time-consuming. If small molecules that bind to the different pockets of the protein could be identified and one could determine how these fragments should be linked, fewer compounds would need to be synthesized to create highaffinity ligands. Although the molecular fragments may only bind weakly to the protein (as expected because of their small size), the free energy of binding of the linked compound is, in principle, the sum of the free energies of each fragment plus a term due to linking (2). Thus, linked compounds with submicromolar affinities could be obtained by linking two fragments that bind in the millimolar range. A compound derived from molecular fragments could be designed rationally with molecular modeling based on the three-dimensional (3D) structure of the biological target. However, this still remains a challenging task because of possible conformational changes of the target or ligand upon complexation, the uncertainties in treating