The ubiquitin-proteasome system (UPS) maintains homeostatic levels of proteins in normal cells and controls the levels of oncogenes and tumour suppressors by tagging proteins with ubiquitin for proteasomal degradation. The UPS is regulated by sequential action of three enzymes: E1 – ubiquitin activating enzyme, E2 - ubiquitin conjugating enzyme, and E3 - ubiquitin ligase. The SCF-Skp2 complex is one of 600 E3s in the human genome, and Skp2 serves as a substrate recognition subunit. Skp2 is an oncoprotein that exerts its oncogenic functions through degradation of specific substrates. A major target of SCF-Skp2 is the cyclin-dependent kinase inhibitor p27 which positively regulates cell cycle progression. Elevated levels of Skp2 and reduced levels of p27 are common in a variety of cancers, including lymphomas and breast and prostate carcinomas. A lack of suitable binding pockets in Skp2 and the intrinsically disordered nature of p27 make this protein-protein interaction (PPI) challenging for conventional small molecule approaches. We aim to develop instead a macrocyclic peptide inhibitor for this PPI.

We have designed and synthesised p27 peptides containing unnatural amino acids and successfully constrained them using ‘click’ chemistry. The dissociation constants show that the constrained peptides (CPs) have dramatically higher affinities for the Cks1-Skp2- Skp1 complex compared with the linear (unconstrained) p27 peptide. The 30 nM affinity of the tightest binding peptide is almost two orders of magnitude higher than that of the linear peptide (3 μM). We suggest that this large enhancement of affinity arises because the binding-competent form of the peptide has a tight turn-like conformation, which is very effectively constrained by the macrocyclic linker. The CPs were also shown to inhibit p27 ubiquitination in vitro. Interestingly, the CPs also inhibited the ubiquitination of two other Skp2 substrates, p21 and N-myc, to varying degrees. A number of different approaches were taken to deliver the CPs into cells and investigate their effect on p27 protein levels and on cellular proliferation. CPs were able to restore p27 levels associated with Skp2 over-expression as well as reduce proliferation of breast cancer cell line MCF-7.

We additionally investigated a different route to constrain the p27 sequence by grafting it onto a loop of a small, stable protein scaffold. These grafted proteins were also able to inhibit p27 ubiquitination. Lastly, we constructed novel proteolysis-targeting proteins (polyproxins) that hijack SCF-Skp2 and direct it to drive the destruction of disease-causing proteins. We generated a library of polyproxins, combining a module that binds a cancer-associated protein, β-catenin, with a module that binds Skp2. β-catenin ubiquitination and degradation in cells was successfully demonstrated using these polyproxins.