Generating macrocyclic inhibitors of protein-protein interactions

Protein-protein interactions (PPIs) play important roles in many diseases and their modulation is an attractive strategy for developing new therapeutics. However, the relatively large and often flat binding surfaces of PPIs makes the development of inhibitors based on classical small molecule drugs rather challenging. Macrocycles, a class of ring-shaped chemical structures, are able to bind to challenging targets such as PPIs, while having the potential to be cell-permeable or even orally available. The potential of targeting intracellular PPIs by macrocyclic compounds has triggered a great deal of interest in the pharmaceutical industry. A challenge with developing macrocycle-based drugs to new targets, however, is the identification of ligands, the reason for this mainly being the lack of large libraries of macrocyclic compounds that could be screened.The goal of my Ph.D. thesis project was to develop macrocyclic inhibitors for several protein-protein interactions. Towards this end, I contributed to establishing novel strategies for synthesizing and screening large macrocyclic compound libraries. Additionally, I established reported or developed new bioassays suited for the high-throughput screening of compound libraries, and I applied them for screening macrocycle libraries generated with the new methods.In my first project, I applied acoustic droplet ejection technology (ADE) for the synthesis and screening of large macrocyclic compound libraries at a picomole scale. Towards this aim, I used a synthetic strategy based on the combinatorial diversification of m bromoacetamide linear peptides with n primary amines followed by cyclization with o bis-electrophile linkers yielding m x n x o macrocycles. I used ADE to perform all liquid transfers, which proved efficient and yielded macrocycles at picomole scale. I performed the screening by directly adding assay reagents to the synthesis plates containing crude macrocycle products. As a proof-of-concept, I assembled a 2,700-member target-focused macrocycle library and screened it against the MDM2:p53 protein-protein interaction, and was able to identify macrocyclic inhibitors with micromolar to sub-micromolar affinity. Most importantly, this work validated the feasibility to synthesize combinatorial macrocycle libraries using ADE at the picomole scale, which was subsequently applied to several other approaches later developed in our lab.In my second project, I aimed at finding macrocyclic inhibitors of the PPI IL-23:IL-23R, which is an important target of several autoimmune disorders. To that end, I developed a high-throughput screening assay for identifying inhibitors of the interaction. I initially developed a screening assay based on fluorescence polarization, which turned out to be prone to artifacts. In a second attempt, I established an assay based on TR-FRET that proved reliable for the identification of IL-23:IL-23R inhibitors, and in particular also for the screening of crude products of macrocyclization reactions. I subsequently synthesized a pilot-scale library by combinatorially cyclizing 384 linear peptides with 8 cyclization linkers to generate 3,072 target-focused macrocyclic compounds. Screening the library using the TR-FRET-based assay identified single-digit micromolar inhibitors of the IL-23:IL-23R PPI, that however, were based on linear side-products of the macrocyclization reactions rather than macrocyclic structures. While no potent ma