| Summary: | <p>Epigenetic modifications, including chemical modification of DNA and post-translational modifications (PTMs) to histone proteins, are important in regulating gene expression. Patterns of PTMs recruit specific recognition modules, including bromodomains, which recognise acetylated lysine residues. Bromodomains play a central role in epigenetic gene regulation and, as aberrant access to genetic information is responsible for many diseases, they are considered attractive therapeutic targets.</p> <p>TRIM33 belongs to the C-VI subfamily of TRIM proteins, possessing a C-terminal chromatin binding unit, which comprises a bromodomain adjacent to a PHD finger. In the PARP-dependent DNA damage repair pathway, TRIM33 is responsible for the timely dissociation of chromatin-remodelling enzyme ALC1 from chromatin. Regulation of ALC1 at chromatin is important as prolonged chromatin relaxation leaves DNA susceptible to further damage. It is hypothesised that TRIM33 bromodomain ligands will prevent TRIM33 binding to chromatin, resulting in the prolonged association of ALC1 at chromatin, inefficient repair of single-strand breaks, and further DNA damage. <p>This dissertation describes the development of chemical tools to validate TRIM33 as a therapeutic target. Bromodomain ligands and PROTACs, which enable the role of the bromodomain to be disconnected from the function of the whole protein, were designed and synthesised. <p>Previous work in the group had identified benzimidazolone 23 as a TRIM33 bromodomain ligand (TRIM33β IC<sub>50</sub> = 11.9 μM) (Figure 1). The optimisation of the selectivity and affinity of 23 through iterative cycles of design, synthesis and testing is detailed. Compounds were first evaluated using an AlphaScreen™ assay, and hits were validated in waterLOGSY and ITC experiments. Initial work demonstrated the importance of the benzylic amine substituent for imparting selectivity over the closely related TRIM24 bromodomain. The selectivity results from interactions with a glutamic acid residue (E981) in TRIM33, which is switched to an alanine residue (A923) in TRIM24. Compound 90 (TRIM33β IC50 = 13.7 μM) was developed as a selective ligand, which retains affinity for TRIM33, and was subsequently used as the basis for ligand design (Figure 1). The region ‘above’ the acetyl-lysine binding pocket was explored by extending off the headgroup of 90. The amine chain and methylation state of the benzimidazolone core were also investigated. Despite an extensive investigation focused on benzimidazolone analogues, a relatively flat SAR was observed, with most ligands displaying low micromolar affinity. A structural water molecule was identified as preventing the amine chain residing in the ZA channel. Subsequent ligand design focused on a bidentate interaction of the amine chain with E981 which avoided displacing the tightly bound ZA channel water molecule. This approach resulted in 161 (TRIM33β IC<sub>50</sub> = 4.83 μM), which has increased affinity for the TRIM33 bromodomain, potentially due to the phenyl ring enforcing a favourable conformation for a bidentate interaction with E981 (Figure 1). <p>The knowledge gained from ligand optimisation was used to facilitate PROTAC design. The core structure of the bromodomain ligands was incorporated as the TRIM33β bromodomain binding moiety, and tethered via an appropriate attachment point to a ligand for the E3 ligase cereblon. A range of linker lengths and linking strategies were used to maximise the chance of achieving potent degradation of TRIM33. Some of the potential TRIM33 PROTACs were found to interfere with the AlphaScreen™ assay therefore an alternative assay is required to evaluate their binding to TRIM33.</p> <p>Together the bromodomain ligands and PROTACs can be used to gain a deeper understanding of the physiological and pathological roles of TRIM33 and its bromodomain.</p></p></p></p>
|
|---|