An integrative structural study of RAD52, an emerging target in anticancer drug discovery

Human RAD52 plays a key role in the DNA damage response through Homologous Recombination (HR) and single-strand annealing (SSA). Although nonessential for the viability of normal cells, RAD52 becomes critical in cancer cells harboring mutations in other DNA repair proteins, such as BRCA1/2. This findings identify RAD52 as a promising target for the development of innovative Synthetic Lethality (SL)-based therapies. In this context, the RAD52 inhibition is proposed as a compelling alternative to PARP inhibition (the only SL-based strategy currently approved for clinical application) offering the potential not only to overcome but also to prevent the increasingly common issue of PARPi resistance. Despite growing interest in RAD52 as a drug discovery, no inhibitors have yet progressed to clinical trials. Its intrinsic tendency to form ring-shaped undecameric structures makes RAD52 both a fascinating and challenging target to study in this context. RAD52 oligomerization is mediated by its N-terminal domain, which is also responsible for DNA binding, while the C-terminus serves as a hub for interactions with key DNA-repair factors such as RAD51 and RPA. The structures of the truncated N-terminal portion of RAD52 have already been resolved by X-Ray Crystallography, however structural information on the protein C-terminus are still missing. This gap limits our understanding of RAD52’s interactions with known partners and hinders the identification of novel hotspots for drug discovery efforts. To address this gap, we performed Cryo-EM studies on full-length RAD52. Our results remarkably improved the resolution of the N-terminal domain, achieving the highest resolution reported to date (2.16 Å), thus providing a more accurate structure for in silico drug design. Additionally, the data enabled a detailed description of the protein’s hydration shell, which can be strategically exploited to predict and enhance the binding affinity of candidate ligand molecules. Nevertheless, the C-terminal domain was poorly resolved and largely absent, highlighting that this region is intrinsically disordered. To better characterize its flexibility and obtain additional structural insights, we combined AlphaFold2 predictions and SEC-SAXS experiments. These analyses confirmed the high degree of flexibility of the RAD52 C-terminal domain, which can be described as an intrinsically disordered region. This finding further underscores the potential significance of this protein domain, which may adopt a defined structure only upon partner binding. These results will significantly inform future investigations into RAD52’s mechanism of action and the development of inhibitors, particularly in the context of emerging synthetic lethality strategies.