Computational Assessment of CKP-25 Binding at Receptor-Recognition Interfaces Across Diverse Human Coronaviruses

Background/Objectives: The continued emergence of novel coronaviruses motivates the search for small molecules that bind conserved features of the spike receptor-recognition interface across diverse species. CKP-25 was recently reported as a small-molecule inhibitor of the SARS-CoV-2 spike receptor-binding domain. This study aimed to assess computationally whether CKP-25 can maintain stable binding at receptor-recognition interfaces across phylogenetically diverse human coronaviruses. Methods: CKP-25 was docked to the spike receptor-binding regions of Middle East respiratory syndrome coronavirus (MERS-CoV), severe acute respiratory syndrome coronavirus 1 (SARS-CoV-1), and human coronaviruses NL63, 229E, and HKU1 (HCoV-NL63, HCoV-229E, and HCoV-HKU1). SARS-CoV-2 (PDB 6M0J), for which CKP-25 has documented antiviral activity, was processed identically and included as an experimentally validated reference system. Each selected complex was subjected to three independent 100 ns all-atom molecular dynamics (MD) simulations initialized with different random velocities. Binding stability was evaluated using protein and ligand root-mean-square deviation (RMSD), residue flexibility, radius of gyration, protein-ligand contact persistence, polar interactions, and Molecular Mechanics-Generalized Born Surface Area (MM-GBSA) calculations. Results: CKP-25 was retained at the targeted receptor-recognition interface in all three replicates for four of the five test targets-MERS-CoV, HCoV-NL63, SARS-CoV-1, and HCoV-HKU1-each of which reproducibly engaged the same aromatic and hydrophobic anchor residues in every independent run. HCoV-229E was the sole exception, the ligand relocating from its docked site in two of three replicates. Ligand RMSD plateaus were generally low, mostly below 2.0 Å. Contact analyses indicated a shared dependence on shape-complementary van der Waals packing involving hydrophobic and aromatic residues, including tyrosine- and tryptophan-rich anchoring environments. MM-GBSA estimates, averaged across replicates, ranged from -24.28 to -34.57 kcal/mol. No test target differed from the SARS-CoV-2 reference (-29.15±3.46 kcal/mol) by more than the replicate-to-replicate spread, with MERS-CoV and HCoV-NL63 giving the most favorable means. Conclusions: CKP-25 remained at the receptor-recognition interface of four of the five human coronaviruses tested, and it did so in all three replicates. In the SARS-CoV-2 control, the same pipeline independently reproduced the experimentally established binding mode, which supports the reliability of these predictions. HCoV-229E was the exception: the ligand left the docked site in two of the three replicates, and no contact was reproducible across all three repetitions. Experimental binding measurements and cell-based entry assays are required to test these predictions.