Highly-constrained peptides such as the knotted peptide natural products are promising medicinal agents because of their impressive biostability and potent activity. Yet, libraries of highly-constrained peptides are challenging to prepare. Here we present a method which utilizes two robust, orthogonal chemical steps to create highly-constrained bicyclic peptide libraries. This technology was optimized to be compatible with in vitro selections by mRNA display. We performed side-by-side monocyclic and bicyclic selections against a model protein (streptavidin). Both selections resulted in peptides with mid nM affinity, and the bicyclic selection yielded a peptide with remarkable protease resistance.

Many intracellular protein–protein interactions are mediated by the phosphorylation of serine, and phosphoserine-containing peptides can inhibit these interactions. However, hydrolysis of the phosphate by phosphatases, and the poor cell permeability associated with phosphorylated peptides has limited their utility in cellular and in vivo contexts. Compounding the problem, strategies to replace phosphoserine in peptide inhibitors with easily accessible mimetics (such as Glu or Asp) routinely fail. Here, we present an in vitro selection strategy for replacement of phosphoserine. Using mRNA display, we created a 10 trillion member structurally diverse unnatural peptide library. From this library, we found a peptide that specifically binds to the C-terminal domain (BRCT)2 of breast cancer associated protein 1 (BRCA1) with an affinity comparable to phosphorylated peptides. A crystal structure of the peptide bound reveals that the pSer-x-x-Phe motif normally found in BRCA1 (BRCT)2 binding partners is replaced by a Glu-x-x-4-fluoroPhe and that the peptide picks up additional contacts on the protein surface not observed in cognate phosphopeptide binding. Expression of the peptide in human cells led to defects in DNA repair by homologous recombination, a process BRCA1 is known to coordinate. Overall, this work validates a new in vitro selection approach for the development of inhibitors of protein–protein interactions mediated by serine phosphorylation.

Peptide macrocyclization is typically associated with the development of higher affinity and more protease stable protein ligands, and, as such, is an important tool in peptide drug discovery. Yet, within the context of a diverse library, does cyclization give inherent advantages over linear peptides? Here, we used mRNA display to create a peptide library of diverse ring sizes and topologies (monocyclic, bicyclic, and linear). Several rounds of in vitro selection against streptavidin were performed and the winning peptide sequences were analyzed for their binding affinities and overall topologies. The effect of adding a protease challenge on the enrichment of various peptides was also investigated. Taken together, the selection output yields insights about the relative abundance of binders of various topologies within a structurally diverse library.

To extend the plasma half-life of a bicyclic peptide antagonist, we chose to link it to the Fc fragment of the long-lived serum protein IgG1. Instead of chemically conjugating the entire bicyclic peptide, we recombinantly expressed its peptide moiety as a fusion protein to an Fc fragment and subsequently cyclized the peptide by chemically reacting its three cysteine residues with tris-(bromomethyl)benzene. This reaction was efficient and selective, yielding completely modified peptide fusion protein and no side products. After optimization of the linker and the Fc fragment format, the bicyclic peptide was fully functional as an inhibitor (K(i) = 76 nM) and showed an extended terminal half-life of 1.5 days in mice. The unexpectedly clean reaction makes chemical macrocyclization of peptide-Fc fusion proteins an attractive synthetic approach. Its good compatibility with the Fc fragment may lend the bromomethylbenzene-based chemistry also for the generation of antibody-drug conjugates.