COMBINATORIAL PEPTIDE LIBRARIES

Small peptides that mimic hormones, cytokines or other large proteins have been identified from random peptide libraries but typically have orders of magnitude lower affinities than their larger polypeptide counterparts. Current random peptide libraries, with approximately 10 9 members, have limited diversity and we predict that expanding the size of peptide libraries to greater than or equal to 10 12 members will greatly enhance the probability of identifying high affinity mimic peptides. We propose to create large scale combinatorial peptide libraries by displaying multiple peptides on the same M13 phage. In this system the genotype and the phenotype are physically linked and binding clones could be identified by enriching the library by biopanning. To test the feasibility of combinatorial peptide phage display our Phase I goal will be to show that two different peptides can be simultaneously displayed on a single phage. In Phase ll we will create large combinatorial peptide libraries with 1012-to- 1014 total phenotypes. We expect our combinatorial peptide libraries to be useful for the identification of high affinity peptides and peptides that mimic heterodimeric proteins or proteins that bind to heterodimeric receptOrs.