Mixture-Based Peptide Libraries for Identifying Protease Cleavage Motifs
All proteases and peptidases are to some extent sequence-specific, in that one or more residues are preferred at particular positions surrounding the cleavage site in substrates. I describe here a general protocol for determining protease cleavage site preferences using mixture-based peptide libraries. Initially a completely random, amino-terminally capped peptide mixture is digested with the protease of interest, and the cleavage products are analyzed by automated Edman sequencing. The distribution of amino acids found in each sequencing cycle indicates which residues are preferred by the protease at positions downstream of the cleavage site. On the basis of these results, a second peptide library is designed that is partially degenerate and partially fixed sequence. Edman sequencing analysis of the cleavage products of this peptide mixture provides preferences amino-terminal to the scissile bond. As necessary, the process is reiterated until the full cleavage motif of the protease is known. Cleavage specificity data obtained with this method have been used to generate specific and efficient peptide substrates, to design potent and specific inhibitors, and to identify novel protease substrates.
Key wordsPeptide libraries Proteases Enzyme specificity High-throughput screening Edman sequencing
Synthesis of all peptide libraries described in the text was done at the Tufts University Medical School Core Facility by Michael Berne. I gratefully acknowledge Beth Piro for peptide sequencing and Lewis Cantley for mentoring and support during development of the methods described here.