Integrative in silico and experimental identification of non-covalent UBE2N inhibitors enhancing PARP inhibitor sensitivity

Abstract

UBE2N, an E2 ubiquitin-conjugating enzyme, has emerged as a compelling therapeutic target in oncology due to its critical roles in DNA damage repair and NF-{kappa}B signalling. While covalent inhibitors have shown preclinical promise, non-covalent inhibitors offer potential advantages in terms of selectivity and reduced off-target effects. However, structural and mechanistic data for non-covalent UBE2N inhibitors remain scarce. To address this gap, we implemented a dual in silico strategy combining structure-based molecular docking and ligand-based 3D pharmacophore modelling. Screening a home library of [~]19,000 compounds targeting both the ubiquitin-binding and cofactor interfaces of UBE2N, we identified 22 candidates suitable for biological evaluation. Among these, two compounds, CERMN-2 and CERMN-16, emerged as promising non-covalent inhibitors. CERMN-16, structurally related to the natural compound Variabine B (identified through 3D pharmacophore screening), significantly reduced SKOV-3 ovarian cancer cell viability and enhanced their sensitivity to the PARP inhibitor Olaparib. CERMN-2, identified through docking, also demonstrated a synergistic effect with Olaparib and showed low toxicity in normal ovarian epithelial cells. Molecular dynamics simulations indicated distinct binding modes for each compound, consistent with their targeted binding sites. Biophysical experiments revealed weak binding of CERMN-16 to UBE2N, whereas CERMN-2 bound UBE2N in two orthogonal assays (Microscale thermophoresis and Nano differential scanning fluorimetry). CERMN-16, and more importantly CERMN-2, therefore represent promising leads for the development of selective, non-natural, non-covalent UBE2N inhibitors. These results provide new insights into UBE2N inhibition and support further investigation of their mechanisms of action and therapeutic potential in combination cancer therapies. HIGHLIGHTSO_LIDual in silico screening (docking and 3D pharmacophore) identified new non-covalent UBE2N inhibitor candidates. C_LIO_LITwo compounds, CERMN-2 and CERMN-16, displayed synergistic activity with Olaparib in ovarian cancer cells. C_LIO_LIMD simulations revealed distinct, site-specific binding modes for both compounds. C_LIO_LIBiophysical assays confirmed UBE2N binding for CERMN-2, identifying it as a promising non-natural, non-covalent lead. C_LI