Stiftung des öffentlichen Rechts an der Martin-Luther-Universität Halle-Wittenberg

AlphaFold is an artificial intelligence approach for predicting the 3D structures of proteins with atomic accuracy. One challenge that limits the use of AlphaFold models for drug discovery is the correct prediction of folding in the absence of ligands and cofactors, which compromises their direct use. We have previously described the optimization and use of the HDAC11-AlphaFold model for the docking of selective inhibitors such as FT895 and SIS17. Based on the predicted binding mode of FT895 in the optimized HDAC11 AlphaFold model, a new scaffold for HDAC11 inhibitors was designed, and the resulting compounds were tested in vitro against various HDAC isoforms. Compound 5a proved to be the most active compound with an IC50 of 365 nM and was able to selectively inhibit HDAC11. 5a also showed promising activity with an EC50 of 3.6 µM on neuroblastoma cells. Furthermore, we supported our study by comparative docking and MD simulations.

The Ataxia telangiectasia and RAD3-related (ATR) kinase is a key regulator of DNA replication stress responses and DNA-damage checkpoints. Several potent and selective ATR inhibitors are reported and four of them are currently in clinical trials in combination with radio- or chemotherapy. Based on the idea of degrading target proteins rather than inhibiting them, we designed, synthesized and biologically characterized a library of ATR-targeted proteolysis targeting chimera (PROTACs). Among the synthesized compounds, the lenalidomide-based PROTAC 42i was the most promising. In pancreatic and cervix cancer cells cancer cells (MIA PaCa-2), it reduced ATR to 40% of the levels in untreated cells. 42i selectively degraded ATR through the proteasome, dependent on the E3 ubiquitin ligase component cereblon, and without affecting the associated kinases ATM and DNA-PKcs. 42i may be a promising candidate for further optimization and biological characterization in various cancer cells.

Proteolysis-targeting chimeras (PROTACs) offer a novel therapeutic strategy for degrading disease-causing proteins, but designing effective degraders remains challenging. PROTACs function by inducing a ternary complex between the target protein and an E3 ligase, requiring structural insights for rational design. Key factors include linker optimization, attachment points, and warhead refinement. Computational approaches, particularly protein-protein docking, are essential for modeling ternary complexes and predicting critical interactions. However, existing docking methods struggle with cereblon (CRBN)-based ternary complexes. To address this, we introduce a computational approach combining HADDOCK protein-protein docking with induced fit PROTAC docking. Validated against 26 crystal structures from the Protein Data Bank (PDB), this method demonstrated high accuracy, especially for CRBN-based complexes. Additionally, molecular dynamics (MD) simulations of CRBN-BRD4-BD1 complexes (PDB IDs 6BN7, 6BOY) provided insights into complex stability through buried surface area and radius of gyration calculations. This validated approach was then applied to five Ataxia telangiectasia and RAD3-related (ATR) kinase PROTACs, enabling modeling in the absence of experimental structures. Our method provides a robust framework for optimizing and designing novel PROTACs targeting diverse proteins.

Acute myeloid leukemia (AML) is a hematological malignancy frequently driven by mutations in the FLT3 gene, particularly internal tandem duplications (FLT3-ITD), which contribute to aberrant cell proliferation and resistance to tyrosine kinase inhibitors (FLT3i). The limitations of current FLT3i therapies, including drug resistance, off-target effects, and poor selectivity, necessitate the development of novel therapeutic strategies. Proteolysis-targeting chimeras (PROTACs) represent a promising approach to achieve selective degradation of oncogenic proteins. We developed and characterized an optimized series of FLT3-targeting PROTACs based on our previously described compound MA49, with a focus on linker modifications and pharmacophore optimization to improve degradation efficiency and pharmacokinetic properties. Among these, compounds MA190 and MA191,containing rigid cyclohexyl-piperidine/piperazine linkers, demonstrate superior degradation of FLT3-ITD in MV4-11 AML cells at nanomolar concentrations, achieving >95% reduction in FLT3-ITD levels, outperforming MA49. These compounds exhibit good solubility, high chemical and plasma stability, and improved kinase selectivity. In cellular assays, MA190 and MA191 induce potent apoptosis in FLT3-ITD+ AML cells but have minimal effects on cells with wild-type FLT3. Proteomic analysis reveal that MA191 also degrades MAPK14 (p38α), a kinase upregulated in leukemia, in addition to FLT3. Dual targeting of FLT3-ITD and MAPK14 enhances pro-apoptotic signaling. The co-inhibition using MA191 or a combination of doramapimod (a MAPK14 inhibitor) with a non-degrading FLT3 inhibitor results in greater caspase-3 activation than either treatment alone. This synergistic effect underscores the therapeutic advantage of degrading multiple oncogenic drivers simultaneously.

Using an electron paramagnetic resonance (EPR) spectroscopic strategy that has been developed for core-shell polymers, the complexity of the binding of fatty acids to human serum albumin (HSA) is characterized in detail. We unravel the internal dynamics of HSA solutions with fatty acids by applying continuous wave EPR (CW EPR) from which we derive a consistent thermodynamic interpretation about fatty acid interactions with HSA in the investigated temperature range of 5 °C – 97 °C. Additionally, data from CW EPR are corroborated by dynamic light scattering (DLS), differential scanning calorimetry (DSC) and nanoscale distance measurements using double electron-electron resonance (DEER) spectroscopy. We discuss our data in light of decades of biophysical studies on albumin and aim at drawing a complete functional and dynamic picture of HSA “at work”. This picture suggests that HSA is built from modular, rotationally-decoupled domains that resemble an entangled three-piece boleadora in solution.

The therapeutic potential of HDAC inhibitors containing a hydroxamic acid moiety as a zinc-binding group (ZBG) is limited in clinical use due to their potential mutagenicity. In addition, hydroxamic acids often exhibit off-target effects that can lead to undesirable toxicity. Therefore, the development of HDAC inhibitors with alternative ZBGs has proven to be a promising approach to overcome these drawbacks. HDAC inhibitors carrying alkyl hydrazide as ZBG have recently been published as selective inhibitors for different HDAC subtypes. In the present study, a ligand-based virtual screening workflow was developed and performed for a designed targeted chemical space. The identification of hit compounds was based on a categorical classification model of alkyl hydrazides as HDAC inhibitors. The two most promising hits of the screening were synthesized by docking against different HDACs, followed by in vitro enzyme inhibition assays. Both hits showed strong inhibition of HDAC11 with IC50 values in the nanomolar range. In addition, the compounds showed good selectivity towards HDAC11 at a concentration of 1 µM, only HDAC8 was also significantly inhibited. Finally, the binding mode of the selected candidates was further investigated through molecular dynamics simulations and metadynamics studies to provide insights for further chemical optimization.