Scaffold-hopping for molecular glues targeting the 14-3-3/ERα complex

Abstract

Molecular glues, small molecules that bind cooperatively at a protein-protein interface, have emerged as powerful modalities for the modulation of protein-protein interactions (PPIs) and "undruggable" targets. The systematic identification of new chemical matter with a molecular glue mechanism of action remains a significant challenge in drug discovery. Here, we present a scaffold hopping approach, using as a starting point our previously developed molecular glues for the native 14-3-3/estrogen receptor alpha (ERα) complex. The novel, computationally designed scaffold is based on the Groebke-Blackburn-Bienaymé multi-component reaction (MCR), leading to drug-like analogs with multiple points of variation, thus enabling the rapid derivatization and optimization of the scaffold. Structure-activity relationships (SAR) are developed using orthogonal biophysical assays, such as intact mass spectrometry, TR-FRET and SPR. Rational structure-guided optimization is facilitated by multiple crystal structures of ternary complexes with the glues, 14-3-3 and phospho-peptides mimicking the highly disordered C-terminus of ERα. Cellular stabilization of 14-3-3/ERα for the most potent analogs is confirmed using a NanoBRET assay with full-length proteins in live cells. Our approach highlights the potential of MCR chemistry, combined with scaffold hopping, to drive the development and optimization of unprecedented molecular glue scaffolds.

Fig. 1. Overview of the scaffold hopping approach. SAR and crystal structures of selected benzyl and aniline analogs.

a Crystal structure of compound 127 (yellow sticks) with 14-3-3σ (grey surface) and phospho-ERα peptide (orange sticks). Interacting aminoacids are shown as sticks and water molecules as red spheres (PDB 8ALW). b Chemical structure of compound 127 and ligand overlay of compound 127 and docking pose of the new MCR scaffold. c General multi-component reaction scaffold (MCR) based on the Groebke-Blackburn-Bienaymé (GBB) reaction and main points of variation. d Overview of general synthetic routes. Detailed experimental conditions are described in the SI. e Mass Spectrometry (MS) bar graphs at 1 μΜ. For each compound, time course experiments were performed with measurements at 1 h, 8 h, 16 h and 24 h. ERα data are shown with different colors for each compound, and apo data in grey. f Crystal structure overlay for compounds 1 (cyan sticks) (PDB 9I6Y) and 127 (yellow sticks) (PDB 8ALW) bound to 14-3-3σ (grey surface) /ERα (orange sticks). g Crystal structure of compound 1 (cyan sticks) (PDB 9I6Y) with 14-3-3σ/ERα. Interacting aminoacid residues are shown as sticks and interacting water molecules as red spheres. h, i Structural overlays of compounds 2 (brown sticks) (PDB 9I6Z), and 10 (dark red sticks) (PDB 9I72) with compound 1 (cyan sticks) (PDB 9I6Y).

Fig. 2. SAR and crystal structures of selected double-ortho substituted analogs.

a MS bar graphs at 1 μΜ. For each compound, time course experiments were performed with measurements at 1 h, 8 h, 16 h, and 24 h. ERα data are shown with different colors for each compound, and apo data in grey. b–e Crystal structures of 14-3-3σ/ERα with compounds 17 (dark pink sticks) (PDB 9I70), 19 (teal sticks) (PDB 9I71) and overlays of crystal structures for compounds 17 (dark pink sticks) (PDB 9I70), 20 (pale green sticks) (PDB 9I73), 21 (pale purple sticks) (PDB 9I74) and 25 (bright yellow sticks) (PDB 9I75).

Fig. 3. SAR and crystal structures of analogs substituted in positions X and Y.

a MS bar graphs at 1 μΜ. For each compound, time course experiments were performed with measurements at 1 h, 8 h, 16 h and 24 h. ERα data are shown with different colors for each compound, and apo data in grey. b–e Crystal structure of 14-3-3σ/ERα with compound 28 (dark yellow sticks) (PDB 9I6S), and overlays of crystal structures for compounds 28 (dark yellow sticks) (PDB 9I6S), 32 (pink sticks) (PDB 9I6T) and 33 (emerald green sticks) (PDB 9I6U) with 21 (pale purple sticks) (PDB 9I74).

Fig. 4. SAR and crystal structures of 2,6-di-Me analogs substituted in positions X and W.

Biophysical data (MS, TR-FRET, SPR) and cell data (NanoBRET). a MS bar graphs at 1 μΜ. For each compound, time course experiments were performed with measurements at 1 h, 8 h, 16 h, and 24 h. ERα data are shown with different colors for each compound, and apo data in grey. b, c Overlays of crystal structures of 14-3-3σ/ERα ternary complexes with compounds 40 (orange sticks) (PDB 9I6V), 41 (red sticks) (PDB 9I6W), and 42 (blue sticks) (PDB 9I6X). d Time-resolved fluorescence energy transfer (TR-FRET) schematic and protein titration data for representative compounds at 100 μM compound or DMSO. For each compound, three independent experiments were performed (n = 3). (TR-FRET schematic was created in BioRender. Konstantinidou, M. (2025) https://BioRender.com/yp73j3h). e Chemical structures of compounds 181, 17, and 41 used in SPR and compound 85, used in NanoBRET as a negative control. f Surface Plasmon Resonance (SPR) data for the binary 14-3-3σ/ERα interaction and ternary interactions with 181, 17, and 41 (mean +/− SD, n = 2). g–h 14-3-3σ-HaloTag/NanoLuc-ERα NanoLuciferase bioluminescence resonance energy transfer (NanoBRET) assay in HEK293T cells with compound titrations (1:2 dilution, starting at 40 μΜ). Data points excluded where compound dosage was toxic to the cells. MCR compounds compared to the previously described stabilizer 181 and 85, an inactive compound as the negative control. The natural product FC-A was used as a control, dosed at 30 μM (black dashed line). Bar graphs quantifying pEC₅₀ values. Data points presented are mean values +/− SD (n = 3 technical) and representative of 4 biological replicates.