Genetic code expansion to improve the drug-likeness of hits from mRNA display - Project Summary. Many important drug targets including protein-protein interactions (PPIs) are still intractable with modern drug discovery methods. In vitro selection technologies, in particular mRNA display, have proven to be powerful methods for discovering macrocyclic peptides that target PPIs. The extreme diversity of mRNA-displayed macrocyclic peptides (10s of trillions of individual peptides) allows rapid generation of a high-affinity hit where other technologies fail. For this reason, mRNA display has been increasingly employed as a tool for lead development in pharmaceutical companies worldwide. Unfortunately, while these hits have high affinity for their targets, they typically greatly exceed even the most generous beyond rule of 5 (bRo5) parameters for drug-likeness. This is because mRNA display relies on in vitro translation to create the diverse libraries and is constrained in monomer diversity by the genetic code. Therefore, accessing the high diversities required to identify hits built from 20 monomers requires random regions of at least 10 residues (20^10 = 10 trillion). To get into the bRo5 space one needs to create shorter peptides of <6 monomers, which are significantly less diverse (20^6 = 64 million). This proposal leverages our expertise in in vitro translation to expand the genetic code in order to increase monomer set available for mRNA display from 20 to >40, dramatically enhancing the diversity of short macrocyclic peptides within the libraries (40^6 = 4 billion) . Our prior research shows that by combining hyperaccurate ribosomes and in vitro transcribed tRNAs one can dramatically break apart the degeneracy of the code. Our first research goal is to leverage these resources and tRNA engineering move as close to a one codon-one amino acid paradigm as possible. A second research area also aims to build monomer diversity through the addition of unnatural base pairs (UBPs). The goal is to show that at least 10 UBP codons are amenable to mRNA display, further enhancing the potential monomer diversity. The third direction focuses on the creation of a strategy for prioritization of monomers to use with a given mRNA-displayed peptide library. This will be achieved by first screening for small molecule fragments that bind followed by appending these fragments onto amino acid side chains within the library. The inclusion of known binding fragments as side chains will improve the affinity and specificity of the downstream hits. Each of the three research goals will utilize model protein targets to validate they are useful within the context of mRNA display. This research is expected to give new, valuable insights into the malleability of the genetic code and the translation apparatus towards engineering. The research will also enable the creation of diverse, bRo5 compliant libraries with monomers tailored to a given target, enabling the downstream discovery of many new drug-like macrocyclic peptide leads to currently intractable targets.
