Identification of Active Sequences

The beads from the library that demonstrated activity were removed using tweezers. Each bead contains approximately 30 pmoles of peptide, which is sufficient for micro-analytical methods to determine the sequence of the peptide without ambiguity. Active sequences were read by automated Edman analysis. All the active sequences were found to show a strong consensus, and multiple copies were independently found for almost all sequences. The consensus for chymotrypsin inhibition at the positions varied was found to be X=Threonine, Y=Phenyl Alanine, Z=Isoleucine; and this particular sequence was found on no less than 5 different individual beads. Re-synthesis of this most active sequence was found to provide an extremely potent inhibitor of the target enzyme, with a K_i value of 20 nM.

Applications and Future Directions

This work was the first time that randomisations within a specific binding motif were used to redirect specificity using a synthetic peptide library. The advantage of a template-based library, rather than a completely random one, is that it is much more likely to produce potent inhibitors. In conjunction with support from GlaxoSmithKline, we are currently extending this work to generate inhibitors of therapeutically interesting proteases, and are examining the molecular details of the interaction in much greater detail.