The neuroprotective transcription factor nuclear receptor related 1 (Nurr1, NR4A2) is in the focus of biomedical research for its promising neuroprotective role in Parkinson’s disease, Alzheimer’s disease, and multiple sclerosis. Its activity can be controlled by ligands offering access to pharmacological Nurr1 modulation. However, the binding epitope(s) and molecular activation mechanisms of synthetic Nurr1 activators remained elusive but are essential to advance Nurr1 ligands towards new medicines. Here we characterized Nurr1 dimer dissociation and coregulator release as molecular contributions to Nurr1 activation by vidofludimus and locate its binding in an allosteric surface pocket lined by helices 1, 5, 7, and 8 by mutagenesis and molecular dynamics simulation. Structure-guided ligand design using these insights resulted in an optimized Nurr1 agonist with substantially enhanced potency and binding affinity. Our results provide a structural and molecular basis for Nurr1 activation by a synthetic agonist which was lacking for rational ligand design.
Sirtuins are involved in cellular processes in multiple ways. Therefore, the development of potent and selective Sirt2 inhibitors provides an important contribution to understanding physiological and pathophysiological mechanisms, particularly for the research and treatment of cancer and neurodegenerative diseases. Based on established SirReal-type lead inhibitors, further selective Sirt2 inhibitors were synthesized in a docking-guided rigidization approach, and the knowledge regarding requirements and properties of the Sirt2-binding pocket was expanded by means of a comprehensive SAR study. Naphthalene derivative FM69 emerged from the screening as the most potent rigidized inhibitor, which, with an IC50 value of 0.15 µM against Sirt2, represents a promising foundation for the further development of novel potent and selective Sirt2 inhibitors based on the presented rigidization strategy.
The duality of halogen-bonding-mediated ligand stabilization and the inteference of NAD + co-factor binding represents an effective strategy in the development of highly potent SIRT2 inhibitors.
Selective and potent inhibitors of the NAD+-dependent deacetylase Sirt2 represent a valuable epigenetic strategy for the treatment of currently incurable diseases such as Parkinson’s disease, Huntington’s disease, Alzheimer’s disease, and multiple sclerosis. Guided by molecular docking and MM/GBSA validation studies, a lead structure-based hybridization strategy was developed, resulting in a series of very effective Sirt2 inhibitors. With RW-93, we present a highly potent and subtype selective Sirt2 inhibitor (IC50 = 16 nM), which as a next generation SirReal-type inhibitor significantly surpasses established Sirt2 inhibitors and contributes to the extension of the current SAR profile. The structural modification strategy employed in this study proved to be highly promising, resulting in the identification of the most potent low-molecular-weight Sirt2 inhibitor reported to date, providing a promising target for further medicinal chemistry-driven SAR studies.
Selective and potent inhibitors of the NAD+-dependent deacetylase Sirt2 represent a valuable epigenetic approach for the treatment of hitherto incurable diseases such as Parkinson's disease, Huntington’s disease, Alzheimer's disease and multiple sclerosis. Guided by docking studies, a lead structure-based hybridization concept was developed, resulting in a series of very effective Sirt2 inhibitors. With RW-93 we present a highly potent and selective Sirt2 inhibitor (IC50 = 16 nM), which as a next generation SirReal-type inhibitor significantly surpasses established Sirt2 inhibitors and extends current structure-activity relationships. The structural modification strategy employed in this study proved to be a highly promising approach, resulting in the identification of the most potent low-molecular Sirt2 inhibitors reported to date.
Automating structural optimization of drug molecules for on-target potency by machine learning is an open challenge in chemistry. Here, we capitalize on the ability of chemical language models (CLMs) to learn from sequential data and design new molecules with desired properties. We establish a training strategy mimicking the learning trajectory of a drug discovery program. Incremental CLM fine-tuning with increasingly potent template molecules from a given structure-activity relationship (SAR) series successfully biases the model to design highly active analogues. Prospective application of this technique to ligand development enables the data-driven design of molecules exceeding known representatives of given bioactive chemotypes in potency without external scoring. Our results reveal an ability of CLMs to capture SAR patterns and long-range dependencies, and to exploit SAR knowledge in designing analogues with improved on-target activity de novo corroborating their applicability to structural optimization of drug molecules.