Discovery of benzyl carbamate inhibitors of coronavirus Mpro enzymes from a legacy collection of cysteine protease inhibitors - PubMed (original) (raw)

. 2025 Dec;40(1):2585619.

doi: 10.1080/14756366.2025.2585619. Epub 2025 Nov 17.

Mateus Sá Magalhães Serafim 1 2 3, Karol R Francisco 1, Erik Vinicius de Sousa Reis 2, Ellen Gonçalves de Oliveira 1 2, Fernanda Kelly Marcelino E Oliveira 2 3, Isadora Serraglio Fortes 1 6 7, Thaís Helena Maciel Fernandes 1 8, Elany Barbosa da Silva 1, Pavla Fajtova 1, Danielle E Skinner 1, Rafay O Syed 1, Jair Lage de Siqueira-Neto 1, Antti Poso 5 9, Bruno Eduardo Fernandes Mota 3, Jordana Grazziela Alves Coelho-Dos-Reis 2, Jônatas Santos Abrahão 2, Vinícius Gonçalves Maltarollo 10, Anthony J O'Donoghue 1, Conor R Caffrey 1

Affiliations

Discovery of benzyl carbamate inhibitors of coronavirus Mpro enzymes from a legacy collection of cysteine protease inhibitors

Mateus Sá Magalhães Serafim et al. J Enzyme Inhib Med Chem. 2025 Dec.

Abstract

The constant emergence of SARS-CoV-2 resistance drives the search for new antivirals. We screened the SARS-CoV-2 cysteine proteases, the main protease (Mpro) and papain-like protease (PLpro), with 141 peptidyl and peptidomimetic inhibitors designed to target a trypanosome cysteine protease. Five compounds (1a-5a) inhibited Mpro (IC50 of 0.1601-16.42 µM), whereas none inhibited PLpro. Compounds 1a-4a inhibited human cathepsin L (hCatL; 0.184-10.74 µM), which is important for viral entry into human cells. Compounds 1a and 5a, and its synthesised (R,S) enantiomer, 5b, which share a benzyl carbamate moiety, inhibited the Mpro of SARS-CoV/MERS-CoV (0.0732-0.8295 µM). The three compounds were biochemically characterised as covalent reversible inhibitors. Compounds 5a and 5b, which contain vinyl ketone warheads, were specific for Mpro, and this behaviour was supported by covalent and noncovalent computational simulations. This study highlights the importance of revisiting legacy assets to identify starting points for new antiviral drugs.

Keywords: COVID-19; Mpro; SARS-CoV-2; cysteine protease; human cathepsin L.

PubMed Disclaimer

Conflict of interest statement

C.R.C., A.J.O. and M.S.M.S. are inventors on a patent related to the technology described in this study. Patent WO 2024/263680 corresponds to the description of the methods and mechanisms associated with the enzymatic inhibitory and antiviral activity of compounds 1a, 5a and 5b, available at the World Intellectual Property Organisation (WIPO). The authors have no other relevant affiliations or financial involvement with any organisation or entity that may have a financial interest in. The authors also have no financial conflict with the subject matter or materials discussed in the manuscript (apart from those disclosed).

Figures

Scheme 1.

Scheme 1.

Reactions and conditions. (a) _N-_Cbz-_L_-phenylalanine i (1.00 eq), NHS (1.00 eq), DCC (1.10 eq), CHCl3 (0.05 M), room temperature (rt), 1 h. (b) _D_-alanine (1.00 eq), NaHCO3 (2.00 eq), H2O/acetone (1:2, 0.05 M), rt, 16 h (89% over two steps). (c) _N-_Cbz-_L_-phenylalanyl-_D_-alanine ii (1.00 eq), N,_O_-dimethyl-hydroxylamine hydrochloride (1.10 eq), ethyl chloroformate (1.00 eq), _N-_methylmorpholine (1.00 eq), freshly prepared diazomethane (6.00 eq), THF (0.1 M), −78 °C to rt, 16 h (37%). (d) Diazoketone iii (1.00 eq), freshly prepared DMDO (3.00 eq), acetone (0.02 M), −78 °C to rt, 16 h (99%). (e) Glyoxal iv (1.00 eq), methyl 2-(dimethoxyphosphoryl)acetate (1.10 eq), NaH (1.10 eq), CH3CN (0.05 M), 0 °C to rt, 16 h (12%).

Scheme 2.

Scheme 2.

Reactions and conditions. (a) _N-_Cbz-_D_-phenylalanine v (1.00 eq), NHS (1.00 eq), DCC (1.10 eq), CHCl3 (0.05 M), rt, 1 h. (b) _D_-alanine (1.00 eq), NaHCO3 (2.00 eq), H2O/acetone (1:2, 0.05 M), rt, 16 h (91% over two steps). (c) _N-_Cbz-_D_-phenylalanyl-_D_-alanine vi (1.00 eq), _N,O-_dimethyl-hydroxylamine hydrochloride (1.10 eq), HATU (1.10 eq), DIPEA (2.50 eq), DMF (0.1 M), rt, 16 h (88%). (d) Weinreb amide vii (1.00 eq), vinyl magnesium bromide (3.00 eq), THF (0.1 M), 0 °C to rt, 3 h (58%).

Figure 1.

Figure 1.

Compounds that inhibit SARS-CoV-2 Mpro activity at 10 μM. aPercent inhibition relative to DMSO control is reported as the mean ± SE calculated from two independent assays in triplicate (n = 6). Errors are given by the ratio of the standard deviation to the square root of the number of measurements. MW corresponds to molecular weight. bCompound 5b was synthesised based on the activity observed for 5a. cThe Mpro inhibitor, nirmatrelvir, used as a positive control, was tested at 100 nM.

Figure 2.

Figure 2.

Proposed binding modes of 5a and 5b with different cysteine proteases. Noncovalent docking was performed using GOLD. Similar binding poses were predicted for both 5a (bright orange) and 5b (sky blue) with SARS-CoV-2 Mpro (grey, A and B, respectively) and SARS-CoV Mpro (aquamarine, C and D, respectively). In contrast, the predicted interactions of 5a (E) and 5b (F) with MERS-CoV Mpro (slate blue) show that the electrophilic vinyl ketone warhead faces away from the catalytic cysteine, despite having other hydrogen bond interactions in the site. The warhead for 5a (G) and 5b (H) is also predicted to face away from the catalytic Cys25 residue of hCatL (orange), which supports their absence of activity against this enzyme in vitro. Conversely, compound 1a (I, light teal) is predicted to interact with the catalytic His163 and other residues in the active site of hCatL. The catalytic residues, His41 and Cys145 (Mpro), and Cys25 and His163 (hCatL), are displayed as sticks and labelled black, along with other important residues for these enzymes. Hydrogen atoms are hidden for clarity. The predicted interactions and measured distances (vinyl ketone with Cys145) are shown as yellow dashed lines. Images were generated with PyMOL (v2.5.7).

Figure 3.

Figure 3.

Time-dependent concentration-response curves for inhibition of SARS-CoV-2 Mpro activity by compounds 1a, 5a, and 5b. Two independent assays performed in triplicate (n = 6 data points). The violet, black, and teal curves represent assay data generated after a 30-, 15-, or 0-min pre-incubation period, respectively, prior to addition of substrate.

Figure 4.

Figure 4.

Proposed covalent binding modes of 5a, 5b, and 1a with SARS-CoV-2 Mpro. Covalent docking was performed using Glide XP. Similar binding poses were predicted for 5a (A, bright orange), 5b (B, sky blue), and 1a (C, light teal) with SARS-CoV-2 Mpro. Both enantiomers are predicted to have P4 interactions with Gln189 and P3 interactions with Glu166. In addition, 5a is predicted to have a P1 contact with Gly143, while 5b interacts with His164. The benzyl carbamate, 1a, is not predicted to have P3 or P4 interactions, having only one P2 contact with His41, and two P1 contacts with Ser144 and His163. The catalytic residues, His41 and Cys145, are displayed as sticks and labelled black, along with other important residues. Hydrogen atoms are hidden for clarity. The predicted interactions and measured distances are shown as yellow dashed lines. Images were generated with PyMOL (v2.5.7).

Figure 5.

Figure 5.

Inhibition of SARS-CoV-2 Mpro by compounds 1a, 5a, and 5b is reversible. Following pre-incubation of the enzyme with excess inhibitor, the mixture was diluted, and the formation of the cleavage product was monitored for 2 h. Two independent assays in triplicate were performed (n = 6 data points). Image displays the average of three data points of product formation (i.e. cleaved substrate) represented by the relative fluorescence units (RFU) over time (h:mm:ss).

Figure 6.

Figure 6.

Proposed binding mode of compounds 5a and 5b in the active site of SARS-CoV-2 Mpro represented by the most populated cluster conformation from the MD simulations for (A) 5a (bright orange) and (B) 5b (sky blue). Residues performing polar contacts are labelled in bold and their interaction frequency in the analysed trajectory time (10 x 200 ns per ligand) are shown in olive as percentage values. (C) Violin plot displaying the variation of the calculated binding energy using MM/GBSA along the entire MD trajectory and normalised to the natural log number of heavy atoms (see Methods). Median energy values are highlighted. Differences between distributions were analysed by a one-sample Mann-Whitney multiple comparison test (****p < 0.0001). (D) Most populated clusters depict a short distance (3.6–4 Å) of the reactive vinyl ketone group of 5a to the catalytic cysteine (orange box). Likewise, 5a and 5b are likely to form covalent reactions, even though the average distance of the warhead reactive atoms from the reactive catalytic cysteine sulphur was unstable along the simulations’ trajectory (E).

References

    1. WHO . WHO coronavirus (COVID-19) dashboard. [accessed 2025 Aug 10]. https://covid19.who.int
    1. Chaudhary N, Weissman D, Whitehead KA.. mRNA vaccines for infectious diseases: principles, delivery and clinical translation. Nat Rev Drug Discov. 2021;20(11):817–838. -PMC -PubMed
    1. Li G, Hilgenfeld R, Whitley R, De Clercq E.. Therapeutic strategies for COVID-19: progress and lessons learned. Nat Rev Drug Discov. 2023;22(6):449–475. -PMC -PubMed
    1. Kronenberger T, Laufer SA, Pillaiyar T.. COVID-19 therapeutics: small-molecule drug development targeting SARS-CoV-2 main protease. Drug Discov Today. 2023;28(6):103579. -PMC -PubMed
    1. Kumari M, Lu R-M, Li M-C, Huang J-L, Hsu F-F, Ko S-H, Ke F-Y, Su S-C, Liang K-H, Yuan JP-Y, et al. A critical overview of current progress for COVID-19: development of vaccines, antiviral drugs, and therapeutic antibodies. J Biomed Sci. 2022;29(1):68. -PMC -PubMed

MeSH terms

Substances

LinkOut - more resources