Amino acid recognition sequences that drive protein-protein interactions were initially identified via partial digestion of one of the protein-binding partners. Fragments that were determined to retain binding potency were then sequenced. Further refinement of the amino acid recognition sequence could then be explored via the preparation of synthetic peptides. This strategy is best exemplified by the work described in the 1980s on the potent "heat-stable" inhibitor of the cyclic adenosine monophosphate (cAMP)-de-pendent protein kinase (PKA) known as PKI (protein kinase inhibitor). Krebs, Walsh, and their colleagues (Scott et al. 1985a,b; Cheng et al. 1986; Scott et al. 1986; Van Patten et al. 1986; Glass et al. 1989) identified a series of peptides that serve as extraordinarily potent inhibitors (Ki<50 nM) of PKA. Protease digestion of the isolated protein furnished a 20-mer peptide that acts as a competitive inhibitor versus peptide substrate with a Ki in the subnanomolar range. These investigators demonstrated that the sequence Gly-Arg-Thr-Gly-Arg-Arg-Asn-Ala-Ile is the active site-directed component of PKI, where the Ala residue is positioned at the site normally reserved for the phosphorylatable serine. Indeed, subsequent studies demonstrated that insertion of serine in active site-directed sequences derived from PKI generates powerful peptide substrates (Mitchell et al. 1995). However, the new library-based methods introduced in the 1990s have largely supplanted the biochemical approaches for identifying amino acid sequences recognized by protein interaction domains. The new methodologies are not only significantly less labor intensive than their classical counterparts, but are also able to bypass the need for large quantities of both binding partners (for digestion and sequencing purposes).
Was this article helpful?