Cross-Crystal-Form Fragment Hit Validation: Only 2 of 96 F2X-Entry Screen Compounds Reproduce Binding in Both Monoclinic and Orthorhombic Mpro Crystals
In a head-to-head comparison across two independent CFS campaigns using the identical 96-compound F2X-Entry Screen library, fragment D08 (this compound) was one of only two fragments—alongside B08—that produced binding events in both the monoclinic C2 and orthorhombic P2₁2₁2₁ crystal forms of SARS-CoV-2 Mpro [1]. The remaining 94 screening compounds either failed to bind entirely or bound in only one crystal form. The monoclinic campaign (C2) yielded a 3% hit rate (3 fragments, 4 events), while the orthorhombic campaign (P2₁2₁2₁) yielded a 10–16% hit rate; only B08 and D08 overlapped across both [1]. This mutual cross-validation provides orthogonal structural evidence for target engagement that is absent for the 94 non-overlapping fragments. In the monoclinic structure (7HUE, 1.70 Å), D08 (as the (3R)-enantiomer, ligand ID A1BNP) exhibited an average occupancy of 0.67 with a real-space correlation coefficient (RSCC) of 0.835 [2]. In the orthorhombic structure (7I19, 1.93 Å), D08 (as the (3S)-enantiomer, ligand ID A1BVW) exhibited occupancies of 0.60 and 0.54 across two chains, with RSCC values of 0.88 and 0.79 respectively [3].
| Evidence Dimension | Cross-crystal-form fragment hit reproducibility (binary validated hit / total library) |
|---|---|
| Target Compound Data | Fragment D08: hit in both monoclinic C2 (7HUE, 1.70 Å, occupancy 0.67, RSCC 0.835) and orthorhombic P2₁2₁2₁ (7I19, 1.93 Å, occupancy 0.60/0.54, RSCC 0.88/0.79) [2][3] |
| Comparator Or Baseline | 94 of 96 F2X-Entry Screen fragments: did NOT reproduce across both crystal forms (only B08 shared this property) [1] |
| Quantified Difference | 2/96 (2.1%) cross-validated; D08 is in the top 2% of the library for binding reproducibility across crystal systems, providing >47-fold enrichment vs. non-cross-validating fragments [1] |
| Conditions | PanDDA crystallographic fragment screening; SARS-CoV-2 Mpro (3CLpro, nsp5); F2X-Entry Screen (96 fragments); monoclinic C2 (median 2.19 Å) and orthorhombic P2₁2₁2₁ (median 1.94 Å) crystal forms; X-ray diffraction [1] |
Why This Matters
Cross-crystal-form validation reduces the probability that the observed binding is a crystallographic artifact, enabling procurement of a fragment hit with structurally confirmed target engagement for downstream hit-to-lead chemistry.
- [1] Barthel, T.; Wollenhaupt, J.; Benz, L.S.; Reincke, P.; Zhang, L.; et al. Orthorhombic SARS-CoV-2 main protease crystals provide higher success rate in fragment screening. bioRxiv 2025, doi:10.1101/2025.08.20.671191. View Source
- [2] RCSB PDB Ligand Validation for 7HUE: A1BNP (N-[(3R)-1,1-dioxo-1λ⁶-thiolan-3-yl]-N,N',N'-trimethylurea). Occupancy: 0.67; RSCC: 0.835; RSR: 0.166; Resolution: 1.70 Å. View Source
- [3] RCSB PDB Ligand Validation for 7I19: A1BVW (N-[(3S)-1,1-dioxo-1λ⁶-thiolan-3-yl]-N,N',N'-trimethylurea). Occupancy: 0.60 (chain B) / 0.54 (chain A); RSCC: 0.88 / 0.79; Resolution: 1.93 Å. View Source
