Researchers at the University of Alberta are preparing to launch clinical trials of a drug used to cure a deadly disease caused by a coronavirus in cats that they expect will also be effective as a treatment for humans against COVID-19.
“In just two months, our results have shown that the drug is effective at inhibiting viral replication in cells with SARS-CoV-2,” said Joanne Lemieux, a professor of biochemistry in the Faculty of Medicine & Dentistry.
“This drug is very likely to work in humans, so we’re encouraged that it will be an effective antiviral treatment for COVID-19 patients.”
The drug is a protease inhibitor that interferes with the virus’s ability to replicate, thus ending an infection. Proteases are key to many body functions and are common targets for drugs to treat everything from high blood pressure to cancer and HIV.
First studied by U of A chemist John Vederas and biochemist Michael James following the 2003 outbreak of severe acute respiratory syndrome (SARS), the protease inhibitor was further developed by veterinary researchers who showed it cures a disease that is fatal in cats.
The work to test the drug against the coronavirus that causes COVID-19 was a co-operative effort between four U of A laboratories, run by Lemieux, Vederas, biochemistry professor Howard Young and the founding director of the Li Ka Shing Institute of Virology, Lorne Tyrrell.
Some of the experiments were carried out by the Stanford Synchrotron Radiation Lightsource Structural Molecular Biology program.
Their findings were published today in the peer-reviewed journal Nature Communications after first being posted on BioRxIV, a research website.
“There’s a rule with COVID research that all results need to be made public immediately,” Lemieux said, which is why they were posted before being peer-reviewed.
She said interest in the work is high, with the paper being accessed thousands of times as soon as it was posted.
Lemieux explained that Vederas synthesized the compounds, and Tyrrell tested them against the SARS-CoV-2 virus in test tubes and in human cell lines. The Young and Lemieux groups then revealed the crystal structure of the drug as it binds with the protein.
“We determined the three-dimensional shape of the protease with the drug in the active site pocket, showing the mechanism of inhibition,” she said. “This will allow us to develop even more effective drugs.”
Lemieux said she will continue to test modifications of the inhibitor to make it an even better fit inside the virus.
But she said the current drug shows enough antiviral action against SARS-CoV-2 to proceed immediately to clinical trials.
“Typically for a drug to go into clinical trials, it has to be confirmed in the lab and then tested in animal models,” Lemieux said. “Because this drug has already been used to treat cats with coronavirus, and it’s effective with little to no toxicity, it’s already passed those stages and this allows us to move forward.”
“Because of the strong data that we and others have gathered we’re pursuing clinical trials for this drug as an antiviral for COVID-19.”
The researchers have established a collaboration with Anivive Life Sciences, a veterinary medicine company that is developing the drug for cats, to produce the quality and quantity of drug needed for human clinical trials.
Lemieux said it will likely be tested in Alberta in combination with other promising antivirals such as remdesivir, the first treatment approved for conditional use in some countries including the United States and Canada.
The COVID-19 outbreak evolved into a pandemic due to the virulent nature of SARS-CoV-2, reaching over 3 million cases worldwide by end of April 2020, with the number of infected growing rapidly worldwide1.
This current scenario contrasts the less virulent SARS outbreak in 2002-03, which had only 8000 cases and 774 deaths (WHO, 2004) 2. There is an urgent need for antiviral therapies for acute COVID-19 infections, especially until an efficacious vaccine is developed.
The main protease of the coronavirus is a strong drug target due to its essential nature for virus maturation and subsequent infection3. Coronaviruses are RNA viruses that hijack the host’s translational machinery to generate viral proteins. The viral RNA encodes two overlapping polyproteins: pp1a and pp1ab, which are 450 kD and 750 kD, respectively.
The polyproteins need to be cleaved in order to release individual functional proteins for viral replication and transcription. Viral encoded proteases include the main protease (Mpro), also called 3CLpro, and a papain-like protease (PLpro). Mpro cleaves the polyproteins at 11 positions primarily at conserved Leu Gln | Ser Ala Gly sequences, which allows for virus assembly. Given its crucial role in virus replication, the SARS-CoV-2 Mpro is a prominent drug target for COVID- 19 antiviral therapy.
The coronavirus Mpro is a cysteine protease for which many different inhibitor classes exist4. Protease inhibitors are common drug candidates if they meet the requirements of low toxicity, solubility, and reversibility5. Several proteases have been identified as molecular targets and used for the development of novel classes of drugs5 including Tipranavir for the treatment of HIV6.
However, inhibition of cysteine proteases by thiol reactive species is often untenable for human drugs unless the inhibitor is reversible. Michael acceptor drugs that are irreversible in vivo, such as Rupintrivir, have failed in clinical trials due to low
bioavailabilty3. Undesired irreversible reaction occurs with numerous mammalian thiols to destroy the inhibitor. Reaction with host protein thiols could also potentially lead to acute toxicity or immune reaction.
In this regard, the reversible reaction of thiols with aldehyde inhibitors to make hemithioacetals presents a unique opportunity for effective cysteine protease inhibition, as they can potentially bind more effectively in the active site of their target protein than with other thiols7.
Water soluble aldehyde bisulphite adducts are readily made, reversibly from the parent aldehyde under physiological conditions, and can be ideal prodrugs for cysteine protease inhibition as described below.
In early studies we developed peptide-based inhibitors, including aldehydes, against viral cysteine proteases7 that were subsequently studied with Mpro during the SARS coronavirus (SARS-CoV) outbreak in 20038.
Peptide aldehydes and their bisulphite derivatives were later used to inhibit the main protease of the Feline Coronavirus FCoV9. FCoV generally causes mild symptoms, but it can lead to feline infectious peritonitis (FIP), which is usually fatal in cats.
The bisulphite adduct GC376, which converts readily to peptide aldehyde GC373, was well tolerated and able to reverse the infection in cats10. This, along with other studies that included ferret and mink coronavirus Mpro, demonstrated the broad specificity of this protease inhibitor11.
A crystal structure of GC376 was solved with the homologous MERS Mpro and demonstrated a covalent interaction with the catalytic cysteine of the Mpro 12. Recently, structures of the SARS-CoV-2 Mpro protease were solved with a peptide-based ketoamide inhibitor13 and various re-purposed drugs such as anti-cancer agents14.
However, these Mpro inhibitors have not been tested in animal models of coronavirus infection nor have they been reported in human or animal trials for SARS. A number are known to have severe side effects and human cell toxicity, especially those that are anticancer agents. In this study, we examine whether GC373 and GC376, established as effective
drugs in cats, inhibit SARS-CoV-2 Mpro reversibly and have potential for use as antiviral therapy in humans.
GC373 and GC376 inhibit SARS-CoV-2 Mpro
We synthesized the key dipeptidyl compounds, aldehyde GC373 and bisulphite adduct GC376 (Figure 1A)9 (Extended data, Figure S1), to test whether these FIP inhibitors are efficacious towards the Mpro of the SARS-CoV-2 and Mpro from SARS-CoV (associated with the 2002 outbreak).
This compound consists of a glutamine surrogate in the substrate, P1 position, a Leu in P2 position and benzene ring in the P3 position, which reflects the known specificity for the SARS-CoV-2 Mpro. The SARS-CoV-2 Mpro was cloned as a SUMO-tag fusion, which allowed for high-yield expression, enhanced stability, and generation of native N- and C-termini (Extended data, Figure S2 and S3).
Similarly, SARS-CoV Mpro was expressed and purified to obtain native N- and C-termini according to previous methods8. Kinetic parameters for both the SARS-CoV Mpro and SARS-CoV-2-Mpro were determined using a synthetic peptide FRET- substrate with an anthranilate-nitrotyrosine donor-acceptor pair (Abz-SVTLQSG-TyrNO2R – Extended data, Figure S4) as it displays over 10-fold more sensitivity compared to the equivalent EDANS-Dabcyl system15.
Both SARS-CoV-2 Mpro and SARS-CoV Mpro exhibited cooperative substrate binding of the FRET-substrate (Extended data, Figure S5)
IC50 measurements revealed that both GC373 and GC376 inhibit the SARS-CoV Mpro
and the SARS-CoV-2 Mpro in vitro at nanomolar concentrations (Figure 1B and C). For the SARS-CoV-2 Mpro, IC50 for GC373 and GC376 are 0.40±0.05 mM and 0.19±0.04 mM,
respectively. This is in agreement with studies of these compounds with Mpro from related viruses. For FCoV Mpro the IC50 for GC376 was 0.04 ±0.04 mM and for GC373 was 0.02 ±0.01
mM9. For SARS-CoV Mpro we observed an enhanced IC50, demonstrating the broad inhibition by both compounds, with GC373 and GC376 being 0.070 ± 0.02 mM and 0.05 ± 0.01 mM, respectively.
The bisulphite adduct GC376 shows slightly higher potency for both enzymes compared to the free aldehyde. Our in vitro IC50 values for GC373 and GC376 reflect tight binding for the SARS-CoV-2 Mpro compared to other inhibitors tested in vitro, e.g. ebselen (IC50
0.67 µM)14, tideglusib (IC50 1.55 µM)14, carmofur (IC50 1.82 µM)14, disulfiram (IC50 9.35 µM)14, shikonin (IC50 15.75 µM)14, PX-12 (IC50 21.39 µM)14. Both GC373 and GC376 are also more potent than recently reported ketoamide inhibitors (IC50 of the lead compound 0.67 µM)13.
Recently, a related peptidyl inhibitor was reported with a similar warhead to our compound, but with an indole group at the P3 position and an IC50 of 0.05±0.005 µM16. However, that compound has not been demonstrated to be efficacious in animals, as is the case for GC376.
Crystal structure of SARS-CoV-2 Mpro in complex with GC373 and GC376
To gain insight into the mechanism of inhibition, the SARS-CoV-2 Mpro crystal structures with inhibitors GC373 and GC376 were determined at 2.0 Angstroms (Figure 2 and Extended data, Table S1). The three-dimensional structure of the SARS-CoV-2 Mpro (PDB Code 6WTM) is highly similar to the recently solved structures with an RMSD of 0.38 Å2 (PDB Code 6LU7) 13,14.
SARS-CoV-2 Mpro crystallized as a dimer facilitated by an N-finger of protomer A (residues 1 to 7) that fits into a pocket in protomer B. Each promoter displayed a two-lobe structure with one lobe composed of two-antiparallel b-barrels (Domains I and II), which form a chymotrypsin and 3C-like peptidase fold, with the active site comprised of a Cys-144 and His-41 dyad located at the domain interface.
The oxyanion hole, influenced by dimerization17, is formed from the main chain residues Gly143, Ser144, and Cys145. The C-terminal domain III, is involved in domain swapping and facilitates dimer formation18. Molecular replacement with structure 6Y7M.PDB revealed electron density in the Fo-Fc map at the catalytic cysteine for both inhibitors GC373 (PDB Code 6WTK) and GC376 (PDB Code 6WTJ).
In both structures the peptidyl inhibitor is covalently attached to Cys-145 as a hemithioacetal, showing that as expected the bisulphite group leaves GC376 (Extended data, Figure S6).
In contrast to the MERS Mpro-GC376 structure, the SARS-CoV-2 Mpro electron density indicated the formation of only one enantiomer for this inhibitor12. A strong hydrogen bond network is established from side chain of His163 and backbone amide of His164, and Glu166, with backbone contributions from Gly143, Ser144 and Cys145 defining the oxyanion hole (Figure 3).
Together this provides strong binding and a low IC50 for the inhibitor. The glutamine surrogate in the substrate P1 position interacts with the side chain of His163, while the Leu in P2 inserts into a hydrophobic pocket, representing the S2 subsite of the enzyme.
Similar to what was observed in the MERS- Mpro-GC376 structure12, the benzyl ring and the b-lactam of the Gln surrogate forms a stacked hydrophobic interaction, which stabilizes the inhibitor in the active site of the protease. A close examination of the subsite for SARS-CoV-2 Mpro reveals regions to allow for future inhibitor development (Extended data, Figure S6).
To confirm the formation of a covalent hemithioacetal as a single enantiomer in the active site, GC373 was prepared with 13C label (>99%) at the aldehyde carbon and mixed in 7.8- fold excess with the Mpro protease from SARS-CoV-2 in deuterated buffer. HSQC NMR analysis (700 MHz) showed appearance of a single crosspeak signal (one isomer only) for the hemithioacetal carbon at 76 ppm (13C) and 5.65 ppm (1H) in accordance with previous chemical shift reports for hemithioacetals7 (Extended data, Figure S7).
SARS-CoV-2 Mpro has enhanced catalytic activity compared to SARS-CoV Mpro
Recent crystal structural analysis reported differences in the residues residing between the dimer interface of SARS-CoV-2 Mpro when compared with the SARS-CoV Mpro13. Previous mutagenesis studies, which altered residues at the dimer interface of SARS-CoV Mpro, enhanced catalytic activity 3.6 fold19.
In agreement with this, our analysis shows that the catalytic turnover rate for SARS-CoV-2-Mpro (135±6 min-1) is almost 5 times faster than SARS-CoV Mpro (30±2 min-1) with our substrate Abz-SVTLQSG-TyrNO2R (Table 1). With this FRET substrate, we demonstrate a higher catalytic efficiency with SARS-CoV-2 Mpro (1.8±0.4 min-1µM-1) compared to SARS-CoV Mpro (0.6±0.2 min-1µM-1).
This finding is in contrast to recent reports where no differences were observed in the catalytic efficiency between SARS-CoV Mpro and SARS-CoV-2 Mpro using a different substrate: 3011±294 s-1M-1 (0.18±0.02 min-1µM-1) and 3426±416.9 s-1M-1 (0.2±0.03 min-1µM-1), respectively)13. It remains to be determined whether this influences the virulence of SARS-CoV-2.
GC373 and GC376 are potent inhibitors of SARS-CoV-2 in cell culture
To test the ability of GC373 and GC376 to inhibit SARS-CoV-2, plaque reduction assays were performed on infected Vero E6 cells in the absence or the presence of increasing concentrations of either GC373 (Figure 4A) or GC376 (Figure 4B) for 48 hours. The results were plotted as a percent inhibition of the number of plaque forming units per well. The EC50 for GC373 was 1.5
mM while the EC50 for GC376 was 0.92 mM. To examine cell cytotoxicity, ATP production was measured using the CellTiter-Glo assay on either Vero E6 cells or A549 cells incubated in the presence of the inhibitors for 24 hours. The CC50s of both GC373 and GC376 were greater than 200 mM. To further examine their antiviral activities, quantitative RT-PCR was performed on
supernatants from cells untreated, and GC373 (Figure 4C) or GC376 (Figure 4D) treated cell cultures. It was observed that both GC373 and GC376 are potent inhibitors of SARS-CoV-2 decreasing viral titers 3 logs, compared with a 2 log decrease in recently published results using other aldehyde compounds16. These results indicate that both GC373 and GC376 are potent inhibitors of SARS-CoV-2 with a therapeutic index of >200.
Numerous drugs were designed originally to inhibit the SARS-CoV Mpro 3. However, the SARS outbreak of 2002 was controlled by public health measures and these compounds were never licensed. GC376 has been used to cure FIP in cats as the Mpro of FIP is effectively inhibited by its breakdown product GC373.
Analogs of these drugs also inhibited the MERS CoV Mpro and blocked viral replication in cells at an EC50 of 0.5 µM12. Our studies show GC376 and GC373 to be effective inhibitors of SARS-CoV-2 Mpro. Clearly these drugs need to be advanced quickly into human trials for COVID-19. SARS-CoV-2 is the cause of COVID-19 and is a virus with a significant mutation rate20.
Also, in some patients the virus has persisted longer than 2 months with some possibility of re-infection21 . Vaccines are critically important, but still likely a year or more away as this virus will likely present vaccine challenges.
There are many clinical trials testing drugs repurposed from their original indications. Remdesivir, a polymerase inhibitor developed as a treatment for Ebola virus22 is showing very promising early results and will likely be confirmed in clinical trials23. Another drug designed to inhibit RNA dependent RNA polymerase, including in coronaviruses, is EIDD 2801 (a N4- hydroxycytidine triphosphate that is incorporated into viral RNA to promote errors in progeny RNA).
These examples of direct acting antivirals (DAAs) for COVID-19 are critically important24. Both GC373 and GC376 compounds are also DAAs designed specifically for coronaviruses. It is likely that several very potent drugs will be required to treat SARS-CoV-2
and to prevent the evolution of resistance they may need to be used in combination. We believe that GC373 and GC376 are candidate antivirals that should be accelerated into clinical trials for COVID-19.
| Protease | K0.5 (µM) | kcat (min-1) | kcat /K0.5 (min-1µM-1) | Hill coefficient |
| SARS-CoV-2 Mpro | 70±10 | 135±6 | 1.8±0.4 | 2.2 |
| SARS-CoV Mpro | 52±17 | 30±2 | 0.6±0.2 | 1.9 |
Table 1. Catalytic parameters of SARS-CoV and SARC-CoV-2 Mpro mediated cleavage of a FRET-peptide substrate. Catalytic parameters were determined for SARS-CoV Mpro and SARS-CoV-2 Mpro with the Abz-SVTLQSG-Y(NO2)-R substrate. Experiments were conducted in duplicate with an N=3. Values are represented as mean ±SEM.
| Protein/Ligand | SARS-CoV-2 Mpro | SARS-CoV2 Mpro-GC373 | SARS-CoV2 Mpro-GC376 |
| PDB Entry | 6WTM | 6WTK | 6WTJ |
| Data collection statistics | |||
| X-ray source | SSRL 12-2 | SSRL 12-2 | SSRL 12-2 |
| Wavelength [Å] | 0.97946 | 0.97946 | 0.97946 |
| Vm [Å3/Da] | 2.0 | 1.97 | 2.04 |
| Solvent content [%] | 38.56 | 37.45 | 39.64 |
| Space group | P21 | C2 | C2 |
| Unit cell dimensions a, b, c (Å) α, β, γ (°) | 44.84 53.60 114.80 90 101.20 90. | 113.35 53.03 45.23 90 102.033 90 | 114.90 53.81 45.50 90 101.70 90 |
| Resolution range a [Å] | 3849-1.85 (1.89-1.85) | 34.04-2.0 (2.09-2.03) | 34.36-1.9 (1.95-1.90) |
| Number of observationsa | 304078 (21067) | 115236 (7722) | 141537 (9860) |
| Number of unique reflections | 88303 (6137) | 34459 (2488) | 41270 (2925) |
| Completeness [%] a | 97.8 (92.2) | 96.0 (89.7) | 97.9 (93.4) |
| Mean I/σ(I) | 6.42 (1.0) | 5.60 (1.0) | 7.18 (1.0) |
| R-meas % | 15.1 | 16.2 | 11.8 |
| CC1/2 a | 99.6 (24) | 99.2(25) | 99.6 (25) |
| Refinement statistics | |||
| Number of unique reflections used for refinement | 88030 | 33622 | 41326 |
| Rwork /Rfree [%] | 20.80/25.18 | 20.13/25.99 | 20.67/25.47 |
| r.m.s.d. in bond lengths [Å] bond angle (°) | 0.008/1.034 | 0.009/1.074 | 0.009/1.113 |
| Clashscore | 5.94 | 11.8 | 8.8 |
| Number of protein atoms | 4761 | 2367 | 2386 |
| Number of ligand atoms | N/A | 29 | 29 |
| Number of water molecules | 290 | 34 | 58 |
| Ramachandran plot | |||
| Preferred regions [%] | 95.01 | 97.70 | 95.68 |
| Allowed regions [%] | 4.49 | 1.32 | 3.32 |
| Outlier regions [%] | 0.50 | 0.99 | 1.33 |
Table S1. Diffraction data and model refinement statistics
a Highest resolution bin in parentheses
b Rmeas=Σhkl [N/(N-1)]1/2 ∑i| Ii,(hkl) – I(hkl)| / ΣhklΣi Ii(hkl)
c R-work = Σ (|Fo| – k|Fc|) / Σ |Fo|
References. – https://doi.org/10.1101/2020.05.03.073080
- WHO. <www.who.int/emergencies/diseases/novel-coronavirus-2019/situation-reports> (2020).
- WHO. <https://www.who.int/csr/sars/country/table2004_04_21/en/> (2003).
- Pillaiyar, T., Manickam, M., Namasivayam, V., Hayashi, Y. & Jung, S. H. An Overview of Severe Acute Respiratory Syndrome-Coronavirus (SARS-CoV) 3CL Protease Inhibitors: Peptidomimetics and Small Molecule Chemotherapy. Journal of medicinal chemistry 59, 6595-6628, doi:10.1021/acs.jmedchem.5b01461 (2016).
- Otto, H. H. & Schirmeister, T. Cysteine Proteases and Their Inhibitors. Chem Rev 97, 133- 172, doi:10.1021/cr950025u (1997).
- Drag, M. & Salvesen, G. S. Emerging principles in protease-based drug discovery. Nature reviews. Drug discovery 9, 690-701, doi:10.1038/nrd3053 (2010).
- Flexner, C., Bate, G. & Kirkpatrick, P. Tipranavir. Nature reviews. Drug discovery 4, 955- 956, doi:10.1038/nrd1907 (2005).
- Malcolm, B. A. et al. Peptide aldehyde inhibitors of hepatitis A virus 3C proteinase.Biochemistry 34, 8172-8179, doi:10.1021/bi00025a024 (1995).
- Yin, J. et al. A mechanistic view of enzyme inhibition and peptide hydrolysis in the active site of the SARS-CoV 3C-like peptidase. J Mol Biol 371, 1060-1074, doi:10.1016/j.jmb.2007.06.001 (2007).
- Kim, Y. et al. Broad-spectrum inhibitors against 3C-like proteases of feline coronaviruses and feline caliciviruses. J Virol 89, 4942-4950, doi:10.1128/JVI.03688-14 (2015).
- Pedersen, N. C. et al. Efficacy of a 3C-like protease inhibitor in treating various forms of acquired feline infectious peritonitis. Journal of feline medicine and surgery 20, 378-392, doi:10.1177/1098612X17729626 (2018).
- Perera, K. D. et al. Protease inhibitors broadly effective against feline, ferret and mink coronaviruses. Antiviral research 160, 79-86, doi:10.1016/j.antiviral.2018.10.015 (2018).
- Galasiti Kankanamalage, A. C. et al. Structure-guided design of potent and permeable inhibitors of MERS coronavirus 3CL protease that utilize a piperidine moiety as a novel design element. European journal of medicinal chemistry 150, 334-346, doi:10.1016/j.ejmech.2018.03.004 (2018).
- Zhang, L. et al. Crystal structure of SARS-CoV-2 main protease provides a basis for design of improved alpha-ketoamide inhibitors. Science, doi:10.1126/science.abb3405 (2020).
- Jin, Z. et al. Structure of M(pro) from COVID-19 virus and discovery of its inhibitors.Nature, doi:10.1038/s41586-020-2223-y (2020).
- Blanchard, J. E. et al. High-throughput screening identifies inhibitors of the SARS coronavirus main proteinase. Chemistry & biology 11, 1445-1453, doi:10.1016/j.chembiol.2004.08.011 (2004).
- Dai, W. et al. Structure-based design of antiviral drug candidates targeting the SARS- CoV-2 main protease. Science, doi:10.1126/science.abb4489 (2020).
- Shi, J., Sivaraman, J. & Song, J. Mechanism for controlling the dimer-monomer switch and coupling dimerization to catalysis of the severe acute respiratory syndrome coronavirus 3C-like protease. Journal of virology 82, 4620-4629, doi:10.1128/JVI.02680- 07 (2008).
- Zhong, N. et al. C-terminal domain of SARS-CoV main protease can form a 3D domain- swapped dimer. Protein Sci 18, 839-844, doi:10.1002/pro.76 (2009).
- Hu, T. et al. Two adjacent mutations on the dimer interface of SARS coronavirus 3C-like protease cause different conformational changes in crystal structure. Virology 388, 324- 334, doi:10.1016/j.virol.2009.03.034 (2009).
- Wang, C. et al. The establishment of reference sequence for SARS-CoV-2 and variation analysis. Journal of medical virology, doi:10.1002/jmv.25762 (2020).
- Biswas, A. et al. Emergence of Novel Coronavirus and COVID-19: whether to stay or die out? Critical reviews in microbiology, 1-12, doi:10.1080/1040841X.2020.1739001 (2020).
- Gordon, C. J. et al. Remdesivir is a direct-acting antiviral that inhibits RNA-dependent RNA polymerase from severe acute respiratory syndrome coronavirus 2 with high potency. J Biol Chem, doi:10.1074/jbc.RA120.013679 (2020).
- Grein, J. et al. Compassionate Use of Remdesivir for Patients with Severe Covid-19. N Engl J Med, doi:10.1056/NEJMoa2007016 (2020).
- Sheahan, T. P. et al. An orally bioavailable broad-spectrum antiviral inhibits SARS-CoV-2 in human airway epithelial cell cultures and multiple coronaviruses in mice. Science translational medicine, doi:10.1126/scitranslmed.abb5883 (2020).
More information: Wayne Vuong et al, Feline coronavirus drug inhibits the main protease of SARS-CoV-2 and blocks virus replication, Nature Communications (2020). DOI: 10.1038/s41467-020-18096-2
