Natural and artificial ligands of cereblon and their therapeutic potential

Abstract

Dissertation ist gesperrt bis 31. Dezember 2024 !

The infamous teratogenic small molecule thalidomide, previously used as a sedative and to relieve morning sickness during pregnancy, caused many miscarriages and led to the birth of thousands of children with abnormalities. However, owing to their pleiotropic properties, thalidomide and its analogues, called immunomodulatory drugs (IMiDs), are indispensable today in the treatment of various forms of cancer and autoimmune diseases. Generally, IMiDs are capable of binding to cereblon (CRBN), the substrate receptor of the CRL4CRBN ligase complex. The natural substrates and functions of this ligase complex in absence of IMiDs and its natural substrates remain largely elusive. However, binding of IMiDs was found to modulate the substrate recognition interface of CRBN to recruit neo-substrates to the E3 complex, which leads to ubiquitination and proteasomal degradation of the latter. This mode of action dictates part of their efficacy in multiple diseases and has also been linked to their teratogenic effects. E3 binders, which are moieties that are capable of binding to E3 ligases, gained even more attention with the development of a directed proteolysis approach. In so-called PROTACs, an E3 binder and a warhead that binds to a protein of interest (POI) are connected via a flexible linker, which leads to the formation of a ternary complex. Close proximity within this ternary complex consisting of PROTAC, E3 ligase and POI can then lead to the ubiquitination and subsequent degradation of the latter. Current constraints of this approach include the limited availability of E3-recruiting moieties. To address this limitation, we have developed crystallographic and affinity-based assays for the characterization of CRBN ligands. Constructs of the human CRBN protein and of a single-domain bacterial homologue from Magnetospirillum gryphiswaldense were designed and used to establish affinity assays based on Microscale Thermophoresis, which showed superior sensitivity and are orthogonal to previously established assays. In these assays and in angiogenesis assays, we investigate the importance of CRBN binding on the anti-angiogenic effects of fluorinated and un-fluorinated thalidomide analogues. We have probed the chemical space of CRBN binding by using a high-throughput screen and specifically designed de novo CRBN effectors. The latter was guided by hydrolysis products of thalidomide and analogs, which we used to design more compact CRBN-binders based on amido-succinimide as a scaffold. In affinity assays and X-ray crystal structures, we show that this scaffold allows the attachment of almost arbitrary chemistry, while still retaining high affinity to CRBN. Without further chemical optimization, several of these compounds already showed neosubstrate degradation in cell culture. In the high-throughput screen, we identified novel, chemically distant binding moieties, showing that the substrate spectrum of CRBN is much bigger than previously thought. Validation in affinity assays and crystal structures showed that some of the newly identified moieties have superior affinities and revealed previously unexplored regions of the ligand space, which can be used to expand the geometrical and chemical space of CRBN binders. These results may guide the design of more effective and selective protein degraders in the future. Additionally, we characterized the binding of endogenous CRBN substrates in affinity assays and in a high-resolution crystal structure with a substrate-derived peptide. We identified a modification of this peptide, which points at a possible connection between artificial small-molecule CRBN effectors and a degron that might be recognized by CRBN in vivo.