Aug 2026· ChemMedChem· Vol 21· 0 citations· 13 references
Medicine
TL;DR
Optimization of UM171, a molecular glue initiating the degradation of neosubstrate HDAC1/2–CoREST–LSD1 through a multiprotein complex formed with KBTBD4, resulted in a 20‐fold improvement in glue activity, demonstrated by the stability of its ternary complex with KBTBD4‐HDAC2.
Abstract
Optimization of UM171, a molecular glue initiating the degradation of neosubstrate HDAC1/2–CoREST–LSD1 through a multiprotein complex formed with KBTBD4, was carried out using the cryo‐EM structure of its KBTBD4‐HDAC2 complex. Structural modifications resulted in a 20‐fold improvement in glue activity, demonstrated by the stability of its ternary complex with KBTBD4‐HDAC2. These improvements were also accompanied by a robust degradation of LSD1 and the CoREST1 protein. Our results, although promising, also highlight the complexity of structure‐guided glue design. The effect of the developed glues on the viability of HepG2 cells revealed a varying level of toxicity, which was disconnected from the glue activity. These observations highlight the potential impact of off‐target effects on the biological activity of these glues.
This work reports on the development of selective histone deacetylase 6 (HDAC6) degraders based on a peptoid scaffold. Structure-based design identified the isocyanide-derived cap region as suitable exit vector for linker attachment, enabling rapid generation of CRBN-recruiting PROTACs via the Ugi four-component reaction. A focused library of 12 degraders revealed a strong dependence of activity on linker composition, with octyl-linked compounds (9e–h) showing the strongest HDAC6 degradation with half-maximal degradation values of 17–36 nM in kinetic HDAC6 degradation assays. The lead compounds selectively degraded HDAC6 in MM.1S and MV4–11 cells without affecting class I HDACs and demonstrated clean proteomic profiles. Functionally, compounds 9e and 9f displayed submicromolar antiproliferative activity against FLT3-ITD-mutated acute myeloid leukemia cells and suppressed proinflammatory signaling in immune cells. Both effects were associated in part with residual class I HDAC inhibition. Overall, this study establishes an efficient multicomponent strategy for PROTAC synthesis and highlights key structure-degradation relationships.
Mikhail Tsymliakov, M. Hanl, Janae Enns et al.· Journal of Medicinal Chemist...· 0 citations
The interaction between c-Myb and the CBP/p300 KIX domain is a critical transcriptional regulatory event and an attractive target for the development of candidate disruptors of the recombinant c-Myb-KIX interaction. In this study, we used an integrated computational and experimental strategy to identify new small molecules capable of disrupting this protein-protein interaction. A focused Umbelliferyl phosphate scaffold library was subjected to stepwise virtual screening via drug-likeness assessment and docking to the c-Myb-binding region of the KIX domain and short molecular dynamics refinement. Selected compounds were then evaluated by 500 ns molecular dynamics simulations, MM/PBSA analysis, free energy landscape (FEL) mapping, and finally by microscale thermophoresis (MST) assay. Computational analyses showed that stable ligand binding did not necessarily translate into disruption of the c-Myb-KIX interface, allowing separation of compounds that stabilized the complex from those predicted to weaken it. Consistent with this distinction, ΔΔGPPIanalysis identified only MUP and Naphthol AS-BI phosphate as protein-protein interaction-weakening ligands, with Naphthol AS-BI phosphate showing the strongest predicted disruptive effect (ΔΔGPPI=+3.25 kcal/mol), whereas DiFMUP and Naphthol AS-D phosphate were predicted to stabilize the complex. Among the tested molecules, Naphthol AS-BI phosphate showed the clearest disruption-like behavior in silico and was the most potent inhibitor in vitro, with an IC₅₀ of 18.9 ± 0.6 μM. Importantly, MUP emerged as the most promising umbelliferyl phosphate-derived hit, displaying measurable inhibitory activity (IC₅₀ = 33.5 ± 0.3 μM) comparable to the reference Naphthol AS-E phosphate (IC₅₀ = 31.2 ± 1.3 μM) and a more favorable predicted ADMET profile. Overall, this work identifies new chemical starting points for targeting the c-Myb-CBP/p300 KIX interaction and supports MUP as an attractive scaffold for further optimization.
Emadeldin M. Kamel, H. Rudayni, A. A. Allam et al.· Biophysical Chemistry· 0 citations
Lens epithelium‐derived growth factor p75 (LEDGF/p75) is a chromatin reader that recognizes di‐ or trimethylated Lys36 of histone H3 (H3K36me2/3)‐modified nucleosomes and is implicated in diverse diseases, including cancer and human immunodeficiency virus (HIV) infection. Inhibiting the interaction between the Pro‐Trp‐Trp‐Pro (PWWP) domain of LEDGF and chromatin through a small‐molecule drug presents an attractive therapeutic opportunity, but the compounds developed to date bind entirely within the small H3K36me2/3 pocket and achieve only modest affinity. Here, we report de novo computational design and structural validation of proteinaceous binders that engage both this canonical pocket and adjacent DNA‐interacting surfaces of the PWWP domain. Using a hotspot‐driven workflow integrating RFdiffusion, ProteinMPNN, AlphaFold, molecular dynamics simulations and manual structural assessment, we generated four protein designs that were subjected to experimental validation. Biophysical analysis confirmed that one designed binder had a low‐micromolar affinity for the LEDGF PWWP domain. Another designed binder revealed unexpected homodimerization which apparently interfered with its binding to the target in solution. Nevertheless, this binder could be co‐crystallized with the PWWP domain. The resulting atomic structure at 2.1 Å resolution confirms correct engagement of the intended binding interface. This crystal structure enabled the construction of an expanded pharmacophore model that can instruct the design of next‐generation small‐molecule or peptide‐based inhibitors targeting the LEDGF PWWP domain and related epigenetic readers. These results demonstrate that modern in silico design pipelines can directly yield functional proteinaceous binders without the need for additional experimental screening using phage display or related technologies.
Thibault Vantieghem, Julie Delepine, Sam Noppen et al.· Protein Science· 0 citations
Background: Hypoxia-inducible factor-1 alpha (HIF-1α) drives tumor adaptation to hypoxia by promoting angiogenesis, metabolic reprogramming, and survival signaling in clear cell renal cell carcinoma (ccRCC). Transcriptional activity of HIF-1α depends on its interaction with the p300/CBP (CREB-binding protein) coactivator via the C-terminal transactivation domain (C-TAD). Direct targeting of this protein–protein interface remains limited, highlighting the need for novel inhibitors.Methods: A structure-based virtual screening of ~250,000 compounds from the ZINC database was conducted to identify candidates targeting a predicted pocket within the HIF-1α C-TAD. Top hits were evaluated using molecular docking and 100 ns molecular dynamics simulations. The lead compound, HFS764, was further characterized using in silico absorption, distribution, metabolism, excretion, and toxicity (ADMET) profiling, homogeneous time-resolved fluorescence (HTRF) assays, and differential scanning fluorimetry (DSF). Functional effects were assessed in Caki-1 cells under hypoxia using hypoxia response element (HRE)-luciferase assays, cell viability and vascular endothelial growth factor (VEGF) secretion.Results: HFS764 demonstrated stable binding within the HIF-1α pocket, with ligand root-mean-square deviation (RMSD) <0.20 nm and a predicted binding free energy of –22.38 kcal/mol. HTRF assays confirmed inhibition of the HIF-1α–p300 interaction (IC50 = 4.89 μM), while DSF showed ligand-induced stabilization (ΔTm = +2.0 °C). In Caki-1 cells, HFS764 suppressed HRE-driven transcription (IC50 = 4.27 μM), and reduced VEGF secretion under hypoxia. Selective cytotoxicity was observed (GI50: 14.70 μM in Caki-1 vs. 37.98 μM in HK-2; therapeutic index = 2.58).Conclusion: HFS764 disrupts the HIF-1α–p300 interface and suppresses hypoxia-driven signaling in ccRCC, supporting the use of transcriptional complex inhibition as a promising therapeutic strategy.
Mesfer Al Shahrani· Discover medicine· 0 citations
Findings validate the diphenylpyrazine scaffold as a promising chemotype for Skp2–Cks1 inhibition and identify C3 as a strong lead for further optimization.
Emadeldin M. Kamel, A. A. Allam, H. Rudayni et al.· Journal of Computer-Aided Mo...· 0 citations
The engineering of induced proximity has transformed drug discovery, yet the development of molecular glues remains largely serendipitous and restricted to the retrospective optimisation of accidental discoveries. Here, we present EvoBind-multimer, a deep learning framework for the de novo design of molecular glues directly from protein sequences. Unlike structure-based docking, our method generates small macrocyclic peptides that bridge user-defined protein pairs without requiring prior interface knowledge or existing ligands. We applied this framework to recruit the E3 ligase VHL to two challenging oncoproteins: KRAS and BRD4. Live-cell NanoBRET demonstrated robust design-induced proximity for both pairs. Mechanistic validation demonstrated that the generated macrocycles form functional VHL-target ternary complexes capable of driving Cullin-RING ligase-dependent proteasomal degradation and downstream signalling shutdown. Finally, evaluation in patient-derived xenograft neuroblastoma tumoroids revealed that ternary complex processing is deeply context-dependent: identical macrocycles acted as potent degraders in one patient model, yet functioned as stabilising “LOCKTACs” in another, driving VHL-dependent target sequestration without turnover. By enabling the de novo design of induced proximity from sequence alone, EvoBind-multimer provides a route towards designing new protein functions.
Andrä Brunner, Krzysztof Wierbiłowicz, Diandra Daumiller et al.· bioRxiv· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduAug 27, 2026
A new machine-learning framework aims to improve the success rate of computational protein design while moving away from results that reproduce sequences found in nature.
