Is NMR Fragment Screening Fine-Tuned to Assess Druggability of Protein–Protein Interactions?

Modulation of protein–protein interactions (PPIs) with small molecules has been hampered by a lack of lucid methods capable of reliably identifying high-quality hits. In fragment screening, the low ligand efficiencies associated with PPI target sites pose significant challenges to fragment binding detection. Here, we investigate the requirements for ligand-based NMR techniques to detect rule-of-three compliant fragments that form part of known high-affinity inhibitors of the PPI between the von Hippel–Lindau protein and the alpha subunit of hypoxia-inducible factor 1 (pVHL:HIF-1α). Careful triaging allowed rescuing weak but specific binding of fragments that would otherwise escape detection at this PPI. Further structural information provided by saturation transfer difference (STD) group epitope mapping, protein-based NMR, competitive isothermal titration calorimetry (ITC), and X-ray crystallography confirmed the binding mode of the rescued fragments. Our findings have important implications for PPI druggability assessment by fragment screening as they reveal an accessible threshold for fragment detection and validation.


-Chemicals and synthesis
All reagents and solvents were obtained from commercial sources, and used as supplied unless otherwise indicated. Reactions requiring anhydrous conditions were conducted in heated glassware (heat gun), under an inert atmosphere (argon), and using anhydrous solvents. CH 2 Cl 2 , toluene and MeOH were distilled over CaH 2 . Et 2 O was distilled on CaH 2 /LiAlH 4 . All reactions were monitored by analytical thin-layer chromatography (TLC) using indicated solvent systems on E. Merck silica gel 60 F254 plates (0.25 mm). TLC plates were visualized using UV light (254 nm) and/or by staining in potassium permanganate followed by heating. Solvents were removed by rotary evaporator below 40˚C and the compounds further dried using high vacuum pumps. 1 H and 13 C NMR were recorded on a Bruker Advance 400 MHz and 100 MHz NMR spectrometer respectively. Chemical shifts (δ H) are quoted in ppm (parts per million) and referenced to residual solvent signals: 1

3-NMR experiments
All ligand-based NMR experiments were carried out at 278 K using Bruker Avance 500 MHz with TCI cryoprobe. Protein-based 1 H-15 N 2D NMR HSQC (correlation via double inept transfer, phase sensitive and with decoupling during acquisition) were performed at 305 K on Bruker Avance 700 MHz with TCI cryoprobe (Department of Chemistry, University of Cambridge). 1

-X-ray diffraction experiments
The X-ray data collection and statistics for VCB crystals soaked with 2 (pdb code 3zrc) and 6 (pdb code 4awj) have been described previously. 1,2 In order to collect data for the 4 and 5 bound VCB complexes, VCB crystals were soaked overnight in a 20 mM solution of 4 or 5, in Na cacodylate pH 6.0, 15% PEG3350, 0.2 M Mg acetate, 5 mM DTT. X-ray data were collected at 100 K at the Diamond synchrotron facilities and processed using XDS. The structures were solved by rigid body refinement using Buster TNT 4 using the VCB apo structure (PDB code 3zrf) 1 as a starting model. The initial structure obtained this way was further refined using Buster TNT and corrected manually using Coot. 5 Data collection statistics and refinement parameters are summarized in Table S1. (a) R cryst = ∑||F obs |-|F calc ||/∑|F obs |, F obs and F calc are observed and calculated structure factor amplitudes (b) R free as for R cryst using a random subset of the data excluded from the refinement (c) Data in brackets are for the highest resolution shell ITC traces together with the results of the data fitting are shown in the Figures S1-S13. S12 Figure S1: Binding detection using ligand-based NMR spectroscopy for 1. Panels a), b) and c) depict spectra for VCB+1 using set-ups 1 and 2 (red and black respectively), the compound alone (blue) and in competition with 100 μM 19-mer HIF-1α peptide under set-up 2 (green). d) Direct ITC titration for 1 (1 mM compound and 100 μM VCB). K a = 6.3 × 10 5 ± 3.5 × 10 4 M -1 ; ΔH = -5435 ± 25 cal/mol and ΔS = 8.3 cal/mol/degree. e) Modeled bound 1 (white carbon sticks) into VCB structure. The protein surface is shown in blue at 30% transparency. e) S13 Figure S2: a) Zoom on the 2D 1 H-15 N HSQC spectrum of 100 μM VCB (red) highlighting the residues shifting due the protein-ligand interaction with 1 mM concentration of 1 (blue). b) Site specific mapping into the VCB structure of the residues involved in the interaction (red).

g)
h) i) S18 Figure S7: Binding detection using ligand-based NMR spectroscopy for 7. Panels a), b) and c) depict spectra for VCB+7 using set-ups 1 and 2 (red and black respectively) and the compound alone (blue). Panels d), e) and f) depict spectra for VCB+7 using set-up 3 (red) and compound alone (blue). Figure S8: Binding detection using ligand-based NMR spectroscopy for 8. Panels a), b) and c) depict spectra for VCB+8 using set-ups 1 and 2 (red and black respectively), the compound alone (blue) and in competition with 100 μM 19-mer HIF-1α peptide under set-up 2 (green). Panels d), e) and f) depict spectra for VCB+8 using set-up 3 (red) and compound alone (blue). g) Competitive ITC titration between 2 and 8. Direct titration of 2 into 100 μM VCB is shown in the absence (black trace) and presence (red trace) of 3 mM of 8. h) 1 H-15 N HSQC of 0.3 mM VCB (black contours) and when titrated with 5 mM of 8 (red contours).
Figure S10: Binding detection using ligand-based NMR spectroscopy for 10. Panels a), b) and c) depict spectra for VCB+10 using set-ups 1 and 2 (red and black respectively), the compound alone (blue) and in competition with 100 μM 19-mer HIF-1α peptide under set-up 2 (green). Panels d), e) and f) depict spectra for VCB+10 using set-up 3 (red) and compound alone (blue). Figure S11: Binding detection using ligand-based NMR spectroscopy for 11. Panels a), b) and c) depict spectra for VCB+11 using set-ups 1 and 2 (red and black respectively), the compound alone (blue) and in competition with 100 μM 19-mer HIF-1α peptide under set-up 2 (green). Panels d), e) and f) depict spectra for VCB+11 using set-up 3 (red) and compound alone (blue). g) Competitive ITC titration between 2 and 11. Direct titration of 2 into 100 μM VCB is shown in the absence (black trace) and presence (red trace) of 3 mM of 11. Figure S12: Binding detection using ligand-based NMR spectroscopy for 12. Panels a), b) and c) depict spectra for VCB+12 using set-ups 1 and 2 (red and black respectively). Panels d), e) and f) depict spectra for VCB+12 using set-up 3 (red) and compound alone (blue)