Covalent narlaprevir- and boceprevir-derived hybrid inhibitors of SARS-CoV-2 main protease

Abstract

Emerging SARS-CoV-2 variants continue to threaten the effectiveness of COVID-19 vaccines, and small-molecule antivirals can provide an important therapeutic treatment option. The viral main protease (Mpro) is critical for virus replication and thus is considered an attractive drug target. We performed the design and characterization of three covalent hybrid inhibitors BBH-1, BBH-2 and NBH-2 created by splicing components of hepatitis C protease inhibitors boceprevir and narlaprevir, and known SARS-CoV-1 protease inhibitors. A joint X-ray/neutron structure of the Mpro/BBH-1 complex demonstrates that a Cys145 thiolate reaction with the inhibitor’s keto-warhead creates a negatively charged oxyanion. Protonation states of the ionizable residues in the Mpro active site adapt to the inhibitor, which appears to be an intrinsic property of Mpro. Structural comparisons of the hybrid inhibitors with PF-07321332 reveal unconventional F···O interactions of PF-07321332 with Mpro which may explain its more favorable enthalpy of binding. BBH-1, BBH-2 and NBH-2 exhibit comparable antiviral properties in vitro relative to PF-07321332, making them good candidates for further design of improved antivirals.

Recent changes in the mutational dynamics of the SARS-CoV-2 main protease substantiate the danger of emerging resistance to antiviral drugs

“…Insert: Zoom in on positions R188-G195 and adjacent residues V186, D187, T196, and D197. The sequence position, complemented by structural properties partly obtained from PDB entry 7SI9 ( 43 ), is displayed on the x -axis.…”

Improving the preclinical and clinical success rates of LMW drugs depends on radical revisions to the status quo scientific foundations of medicinal chemistry: a case study on COVID Mproinhibition

“…Since WATMD calculations are focused on water exchanges rather than solute rearrangements, we refer to these simulations as “solvation dynamics (SD) simulations” (noting that predicted solute rearrangements are highly unreliable due to the overestimation of interatomic contact energies and underestimation of solvation free energy by force-field-based methods). Our overall method is fully described in references 5, 7, and 10 and briefly summarized here: Nirmatrelvir (PF-07321332) 17 and PF-00835231 18 were extracted from their crystallized M pro -bound structures (PDB code = 7SI9 17 and 6XHM, 18 respectively). The inhibitors were simulated in their fully restrained conformations using AMBER 20 19, 20 PMEMD CUDA (GAFF and ff99sb force-fields) for 100 nanoseconds (ns) in a box of explicit TIP3P water molecules. The fully unrestrained monomeric M pro structure (PDB code = 2QCY 21 ) was simulated in the same manner using AMBER 16. The time-averaged hydrogen (H) and oxygen (O) occupancies within a stationary three-dimensional grid comprised of 1 Å 3 voxels in which each solute structure was embedded were calculated over the last 40,000 frames (10 ns) of the trajectory using WATMD V9 3, 16, 22 (noting that the solvation fully converges within 100 ns). The high and low occupancy voxel data was normalized to the bulk-like solvating water (i.e., the mean of the distribution) on a solute-by-solute basis, so as to achieve self-consistency (WATMD calculations can be considered as first principles for this reason). The high occupancy voxels were then scaled to the largest voxel in the entire dataset. The voxels were assigned to bulk-like, H-bond enriched, and H-bond depleted solvation states (noting that the results in all cases are distributed in a Gaussian fashion, the specific properties of which vary among solutes), as follows: Bulk and bulk-like voxels reside at the mean of the distribution, where the H and O positions in the same voxel are fully uncorrelated (corresponding to no orientational preference of the occupying water molecule). H-bond depleted voxels reside in the extreme left tail of the distribution (where G bulk < G solvation ). H-bond enriched voxels reside in the extreme right tail of the distribution (where G bulk > G solvation ). Water occupancies are underestimated in voxels that are transiently occupied by mobile solute atoms that compete with water exchanges.…”

“…Nirmatrelvir (PF-07321332) 17 and PF-00835231 18 were extracted from their crystallized M pro -bound structures (PDB code = 7SI9 17 and 6XHM, 18 respectively).…”

Small molecules in the treatment of COVID-19

“… 79 Kneller et al developed three hybrid peptidomimetic inhibitors, BBH-1, BBH-2, and NBH-2, by splicing components of boceprevir and narlaprevir, and proved their antiviral properties in vitro relative to nirmatrelvir. 86 A study conducted by Xia et al also showed the in vitro broad-spectrum coronavirus antiviral effect of two rationally designed inhibitors based on the peptidomimetic compounds GC-376, telaprevir, and boceprevir. 528 To date, the development of peptidomimetics is the most used strategy in the search of anti-COVID-19 drugs.…”

“… 72 , 85 Kenller et al presented the design and characterization of three hybrid reversible covalent SARS-CoV-2 M pro inhibitors named BBH-1, BBH-2, and NBH-2 by splicing the SARS-CoV protease inhibitors boceprevir and narlaprevir. 86 By substituting the ketoamide group of boceprevir with the keto-benzothiazole moiety or introducing the nitrile warhead, they directed the warhead into the oxyanion hole. Then, they substituted the P1 group of boceprevir and narlaprevir with a Gln-mimic γ-lactam, thereby synthesizing the hybrid reversible covalent inhibitors BBH-1, BBH-2, and NBH-2.…”