Dear Editor,
The outbreak of coronavirus disease 2019 (COVID-19) has become a severe threat to global public health. Although many drug repurposing researches have been carried out, no effective drugs have been found in clinical studies.1 Among the viral proteins of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) that causes COVID-19, the 3C-like protease (3CLpro), the main protease responsible for viral polyprotein processing, is highly conserved among coronaviruses and serves as a promising target for broad-spectrum anti-CoV therapy. A number of SARS-CoV 3CLpro inhibitors have been reported before. Recently, several SARS-CoV-2 3CLpro inhibitors were discovered by structural-based drug design and high-throughput screening.2 Though these compounds showed encouraging antiviral activity in vitro, their in vivo efficacy, safety and metabolism need further investigation. There is a continuous and urgent need to discover new inhibitors with diverse chemical structures and novel mode of action. Many currently known coronavirus 3CLpro inhibitors act through covalent binding.2 Covalent inhibitors are especially advantageous with enhanced therapeutic efficacy and minimized side effects, as exemplified by several approved anti-tumor covalent drugs.3 Thus, the screening of covalent 3CLpro inhibitors, especially those with primary in vivo safety evaluation, may facilitate the discovery of antiviral agents.
We experimentally screened 315 compounds with electrophilic moieties that may covalently bind to the active site cysteine of SARS-CoV-2 3CLpro. Among them, 182 have been approved for clinical use and 51 have been processed to clinical trials. We found 15 compounds that inhibit SARS-CoV-2 3CLpro activity with IC50’s of <30 μM (Fig. 1a, Supplementary Fig. S1a and Supplementary Table S1). We focused on the two compounds, bardoxolone methyl and bardoxolone, that have been processed to clinical trials and their anti-SARS-CoV-2 activity has not been reported before. The safety and pharmacokinetics of these two compounds are well characterized. We further tested the anti-SARS-CoV-2 activity of bardoxolone and bardoxolone methyl. With full-time incubation, bardoxolone methyl and bardoxolone inhibit SARS-CoV-2 replication in Vero cells with EC50 values of 0.29 μM (SI = 23.9) and 0.43 μM (SI = 56.6), respectively (Fig. 1b). Both compounds also inhibit the SARS-CoV-2 viral replication in human Calu-3 cells with EC50 values of 0.20 μM (SI = 5.8) and 0.42 μM (SI = 28.2), respectively. Efficacies were also confirmed with visualization of virus nucleoprotein (NP) expression using immunofluorescence microscopy at 48 h post infection (Fig. 1c).
To elucidate the mode of action of hit compounds against SARS-CoV-2 3CLpro, we performed enzyme kinetic studies with different concentrations of bardoxolone and bardoxolone methyl. In the absence of inhibitors, substrate cleavage increased with time. In contrast, in the presence of inhibitors, enzyme activity rapidly approached a plateau that is typical of covalent inhibition (Supplementary Fig. S2). We calculated the equilibrium dissociation constant (Ki) and the inactivation rate constant (kinact) for these inhibitors. The kinact values of bardoxolone and bardoxolone methyl are 0.00792 ± 0.00419 and 0.00218 ± 0.00018 s−1, respectively. Moreover, prolonged incubation of SARS-CoV-2 3CLpro with the two compounds exhibited a time-dependent increase of inhibition activity (Supplementary Table S2 and Supplementary Fig. S3). Further liquid chromatography-tandem mass spectrometry (LC–MS/MS) analysis confirmed that bardoxolone binds SARS-CoV-2 3CLpro in a reversible covalent manner (Supplementary Fig. S4). Molecular docking indicated that both bardoxolone methyl and bardoxolone bind to a pocket between domain I and domain II and form hydrogen bonds with Arg40 as well as hydrophobic interaction with Phe181 and Val186 (Supplementary Fig. S5). The sulfur atom of Cys85 is close to the reactive carbon atoms of bardoxolone methyl and bardoxolone with ~5 Å distance, suggesting the potential for covalent bond formation. The methyl group of bardoxolone methyl neutralizes the negative charge of bardoxolone, thus reduces its electrostatic repulsion with Glu55 and enhances its binding affinity.
Selective covalent inhibitors should specifically bind the targets in addition to covalent bond formation. We further used isothermal titration calorimetry to measure the thermodynamic binding parameters of bardoxolone and bardoxolone methyl with SARS-CoV-2 3CLpro (Supplementary Fig. S6). All two compounds showed specific binding to SARS-CoV-2 3CLpro with dissociation constants of 25.90 μM and 2.72 μM, respectively, which are in good agreement with their enzyme inhibition activity.
Bardoxolone and bardoxolone methyl are oleanolic acid-derived semi-synthetic triterpenoids that activate the Nrf2 pathway and inhibit the NF-κB pathway. Clinical trials are ongoing to explore the potential of bardoxolone methyl to treat chronic kidney diseases. The steady-state plasma concentration of bardoxolone by intravenous infusion exceeds 1 μM at doses below the maximum tolerated doses,4 which is well above EC50 value (0.43 μM) against SARS-CoV-2. It was reported that the Nrf2 pathway was suppressed in lung biopsies from COVID-19 patients and the induction of Nrf2 by 4-octyl-itaconate and dimethyl fumarate limited the host inflammatory response and inhibited the replication of SARS-CoV-2.5 Thus, the Nrf2 activators bardoxolone and bardoxolone methyl can be developed as a multifaceted antiviral treatment strategy by inhibiting viral replication, promoting resolution of inflammation, providing robust cyto-protection, and facilitating tissue repair. We recommend testing their activities as anti-COVID-19 agents.
Supplementary information
Acknowledgements
This work was supported in part by the Ministry of Science and Technology of China (2016YFA0502303, 2016YFD0500301), the National Natural Science Foundation of China (21633001), and the Fundamental Research Funds for the Central Universities of China.
Author contributions
Conceptualization: L.L. and Q.S.; experimental studies: Q.S., F.Y., H.L., C.L., R.L., B.H., and L.Z.; docking studies: H.L.; writing: Q.S., H.L., F.Y., and L.L.; supervision: L.L. and W.T.
Data availability
The data used and analyzed in this study are available in the main text and the Supplementary Materials.
Competing interests
The authors declare no competing interests.
Footnotes
These authors contributed equally: Qi Sun, Fei Ye, Hao Liang, Hongbo Liu
Contributor Information
Wenjie Tan, Email: tanwj28@163.com.
Luhua Lai, Email: lhlai@pku.edu.cn.
Supplementary information
The online version contains supplementary material available at 10.1038/s41392-021-00628-x.
References
- 1.Parvathaneni V, Gupta V. Utilizing drug repurposing against COVID-19—efficacy, limitations, and challenges. Life Sci. 2020;259:118275. doi: 10.1016/j.lfs.2020.118275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Cannalire, R., Cerchia, C., Beccari, A. R., Di Leva, F. S. & Summa, V. Targeting SARS-CoV-2 proteases and polymerase for COVID-19 treatment: state of the art and future opportunities. J. Med. Chem. 10.1021/acs.jmedchem.0c01140 (2020). [DOI] [PMC free article] [PubMed]
- 3.Chaikuad A, Koch P, Laufer SA, Knapp S. The cysteinome of protein kinases as a target in drug development. Angew. Chem. Int. Ed. Engl. 2018;57:4372–4385. doi: 10.1002/anie.201707875. [DOI] [PubMed] [Google Scholar]
- 4.Speranza G, et al. Phase I study of the synthetic triterpenoid, 2-cyano-3, 12-dioxoolean-1, 9-dien-28-oic acid (CDDO), in advanced solid tumors. Cancer Chemother. Pharm. 2012;69:431–438. doi: 10.1007/s00280-011-1712-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cuadrado A, et al. Can activation of NRF2 be a strategy against COVID-19? Trends Pharm. Sci. 2020;41:598–610. doi: 10.1016/j.tips.2020.07.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data used and analyzed in this study are available in the main text and the Supplementary Materials.