High-throughput approaches for peptide discovery have been dramatically boosted by the development of genetically-encoded methods. In these, the DNA encoding a molecule is connected to the molecule itself, as illustrated above (blue/black lines are RNA/DNA, and spheres are amino acids). This allows binders to a given protein to be very efficiently separated from non-binders, in principle isolating and identifying from a single initial molecule. This is a very powerful concept, but linear sequences of the 20 canonical amino acids are often of limited use in biological systems. Adding chemical complexity, still in a genetically encoded way, can address this issue. Two common approaches for this in the context of mRNA display are to reassign codons to non-canonical building blocks (genetic code reprogramming) and to add new functionality after translation through chemical means (chemoselective transformations). We work on development of new methods in both areas (and look at applications of these, described further below).

• Macrocyclisation of a peptide adds conformational constraints. This single modification imroves potency by removing some of the entropic penalty; it improves selectivity by preventing a peptide from adopting conformations that can bind to other targets; it improves enzymatic stability by holding backbone amide bonds in restricted angles not accessible to protease catalytic residues; and it also can improve membrane permeability by increasing intramolecular hydrogen bonding. However, it can be difficult to introduce macrocyclisation after discovery of a linear peptide. We therefore develop and apply macrocyclisation approaches that can be integrated with the discovery approach.

• Glycosylation of proteins is a very common post-translational modification, which can improve solubility, protease resistance, and uptake. For proteins that bind sugars, this interaction is often weak, and so can be boosted through conjugation to a peptide. We develop methods that allow construction of glycopeptide libraries, with a particular emphasis on combining this with new macrocyclisation approaches above.

• Diversity in peptides is limited by the number of building blocks, which in turn limits the chemical space that can be explored in mRNA display experiments. For longer sequences this is not an issue, as numerous contacts working together can still give a potent hit. However, these larger peptides are very difficult to make cell permeable, and so there are many relevant targets that they cannot access. To access cell permeable macrocyclic peptides, we need to make these molecules smaller, but we do not want to sacrifice target engagement. An increase in the chemical diversity of the building blocks may be able to solve this. For this reason we are investigating approaches to combine chemical late stage diversification reactions in a DNA-encoded setting with mRNA display to build hybrid libraries containing a peptide scaffold core and small fragment attachments.

Macrocyclic peptides are receiving increasing attention as a drug modality to can tackle challenges that are resistant to traditional small molecules. Similar to antibodies, the large contact surface allows for highly precise and strong binding to a drug target. Unlike antibodies, however, macrocyclic peptides are defined chemical entities that can be optimised, scaled up, stored, and in some cases even cross biological membranes. This makes them very powerful starting points for drug development.

We apply new and existing methodologies for macrocyclic peptide discovery by mRNA display, incorporating genetic code reprogramming and/or chemical modification to introduce valuable additions. General areas of interest include immunology, anti-infective agents, and anti-cancer agents.

Specific targets we are currently or have previously worked on include

• Dendritic Cell-Specific Intercellular adhesion molecule-3-Grabbing Non-integrin; DC-SIGN (dendritic cells)

• Macrophage galactose-type lectin; MGL (dendritic cells, macrophages)

• Nicotinamide N-methyl transferase; NNMT (cancer and xenobiotic metabolism)

• Cofactor associated arginine methyltransferase; CARM1 (histone methylation)

• Haemagglutinin; HA (influenza)

• Spike protein (SARS-CoV-2)