A novel compound active against SARS-CoV-2 targeting uridylate-specific endoribonuclease (NendoU/NSP15): in silico and in vitro investigations

Abstract

NendoU (NSP15) is an Mn(2+)-dependent, uridylate-specific enzyme, which leaves 2'-3'-cyclic phosphates 5' to the cleaved bond. Our in-house library was subjected to high throughput virtual screening (HTVS) to identify compounds with potential to inhibit NendoU enzyme, high-rank compounds (those that bound to multiple target structures) were further subjected to 100 nanoseconds MD simulations. Among these, one was found to be bound highly stable within the active site of the NendoU protein structure. Here, we are reporting a derivative of piperazine based '(2S,3S)-3-amino-1-(4-(4-(tert-butyl)benzyl)piperazin-1-yl)-4-phenylbutan-2-ol' (IV) from our in-house libraries having potential efficacy against SARS-CoV-2 in in vitro assays. This compound demonstrated inhibition of viral replication at the same level as Ivermectin, a known SARS-CoV-2 inhibitor, which is not used due to its toxicity at a higher than the currently approved dosage. Compound IV was not toxic to the cell lines up to a 50 μM concentration and exhibited IC_50s of 4.97 μM and 8.46 μM in viral entry and spread assay, respectively. Therefore, this novel class of NendoU inhibitor could provide new insights for the development of treatment options for COVID-19.

Fig. 1. Schematic representation of 2D interaction maps against NendoU enzyme of SARS-CoV-2. A) Compound IV (identified hit analogue of virtually screened library); and B) Tipiracil (active site bound known inhibitor of NendoU enzyme, PDB ID: 6X1B).

Fig. 2. Results of a 100 nanoseconds (ns) MD simulation of compound IV. A) Schematic 2D representation of bound ligand interactions throughout the simulation; B) critical protein–ligand contacts of amino acid side chain residues with the interaction properties; C) root mean square fluctuation (RMSF) between the binding site of the target protein and interacting ligand; and D) root mean square deviations (RMSD) difference between the NendoU enzyme (NSP15) and bound ligand IV (<4 Å). The graph was obtained for the RMSD value of ligand (purple line) from the protein backbone (green line). The docked complex quickly stabilized to a very low energy state (within 20 ns) and was highly stable throughout the simulation.

Fig. 3. Frame-by-frame analysis of HEA-based compound IV with the targeted NendoU enzyme (NSP15) of SARS-CoV-2. A) for 0 ns frame in which the compound exhibited seven interactions with residues (Glu_264, Glu_266, and Tyr_278) of NendoU enzyme and three water interactions; B) for 20 ns, frame in which the compound did not show any interaction with the enzyme, it is showing five water interactions only; C) for 40 ns frame in which the compound showed one interaction with residue Asp_267 of NendoU enzyme and five water interactions; D) for 60 ns frame, where the compound interacted with the residues (Asp_267, Leu_265, and Lys_276) and four water molecules; E) for 80 ns frame, where the compound interacted with residues (Asp_267 and Lys_276) and five water molecules; and F) for 100 ns frame, where the compound interacted with the residues (Asp_267, Leu_265, and Lys_276) and three water molecules.

Scheme 1. Synthesis of the compound IV.

Fig. 4. Synchronized infections were conducted for a viral spread as described in Materials and Methods. Each curve shows a dose–response to the indicated 2 tested compounds (color-coded; key inset). The results are presented as the PFU formed in the presence of the drug as log(inhibitor) vs. normalized response (percent inhibition) -- variable slope. Standard deviations for each observed concentration in triplicate are represented as error bars. A) SARS CoV-2 anti-viral entry assay. Synchronized infections were conducted for viral entry as described in Materials and Methods. Each curve shows a dose–response to the indicated 2 tested compounds (color-coded; key inset). At 95% CI; ivermectin showed log IC₅₀ = 0.4317 and Std. error for log IC₅₀ 0.04589. Compound IV showed log IC₅₀ = 0.2979 and Std. error for log IC₅₀ 0.02746; and B) SARS-CoV-2 anti-viral spread assay. At 95% CI; ivermectin showed log IC₅₀ = 0.7792 and Std. error for log IC₅₀ 0.01854. Compound IV showed log IC₅₀ = 0.7961 and Std. error for log IC₅₀ 0.02272.