WikiJournal Preprints/The Efficacy of Paxlovid against COVID-19 is the Result of the Tight Molecular Docking between Mpro and Antiviral Drugs (Nirmatrelvir and Ritonavir)

From Wikiversity
Jump to navigation Jump to search

WikiJournal Preprints
Open access • Publication charge free • Public peer review

WikiJournal User Group is a publishing group of open-access, free-to-publish, Wikipedia-integrated academic journals. <seo title=" Wikiversity Journal User Group, WikiJournal Free to publish, Open access, Open-access, Non-profit, online journal, Public peer review "/>

<meta name='citation_doi' value=>

Article information

Author: Ali Adel Dawood[a]

See author information ▼
  1. University of Mosul


Many anti-COVID-19 medicines are now being tried in clinical trials, however credible clinical studies are becoming increasingly difficult to come by. Paxlovid is a ritonavir-boosted nirmatrelvir drug that the FDA authorized it for the treatment of COVID-19. This study aimed to demonstrate the interaction of nirmatrelvir and ritonavir on the active site of SARS-CoV-2 Mpro. According to the docking data, nirmatrelvir has a stronger interaction with Mpro than ritonavir, which has more multiple bonds. The molecular docking of the antiviral drugs has a substantial influence on whether paxlovid has to be utilized as a COVID-19 treatment. According to this study, paxlovid may work on new strains, including Omicron, because the mutation P132H in the Omicron variant's Mpro has no direct effect on the protein.


[edit | edit source]

COVID-19 drew worldwide attention, touching every corner of the globe and altering the world's social and economic conditions. Even though the virus mostly causes a mild respiratory infection, a considerable number of people acquire acute illness and die as a result of it. Furthermore, there are many asymptomatic illnesses that might transfer the virus to others. Patients who have underlying diseases are at a higher risk of getting a severe condition.[1][2]

Several antiviral drugs are now being tested in clinical trials, but credible clinical studies are reportedly getting more difficult to conduct as the public's appetite for readily available therapies develops. Remdesivir is an antiviral that is now being evaluated in clinical trials to treat COVID-19. It is hypothesized to reduce RNA synthesis by targeting RdRP. [3][4]

Because it cleans proproteins after they are translated into the host cell's cytoplasm, the protease enzyme is vital for viral protein maturation. Proteases are thus commonly used as therapeutic candidates. The suppression of viral protease can prevent mature viral particles from forming.[5] The main ORF1ab encodes the Mpro (nsp5), which cleaves two overlapping proteins into 16 non-structural proteins that are critical for viral replication and maturation. During viral replication, Mpro involves in the maturation cleavage of polyproteins. It is a key player in virus entry into the host cell. The Mpro is a three-domain homodimer containing two protomers.[6][7]

Many antiviral medicines targeting proteases have been created to combat viral infections. Lots of antiviral drugs are approved by FDA to treat HIV and hepatitis C virus. Paxlovid is the first oral antiviral drug licensed by the FDA COVID-19.[8][9] Paxlovid is a ritonavir-boosted nirmatrelvir drug for individuals aged 12 and above against COVID-19, which might lead to hospitalization or death. Pfizer's Nirmatrelvir is a primary protease inhibitor that inhibits Mpro enzyme activity. Remdesivir is now being tested in clinical trials to treat COVID-19.[10][11]

Computational modeling is a new field of study that is helping to improve the success of drug development efforts. It is influenced by a variety of factors such as protein-ligand geometry, chemical interactions, and a variety of other constraints.[12] The ability to crystallize viral proteases in the presence of possible inhibitors is crucial because it enables design based on the structural dynamics of the active site in target enzymes (monomer or dimer).[13][14]

In this work, we repurposed nirmatrelvir and ritonavir, two current HIV-protease combo inhibitors, to target SARS-CoV-2 major protease (Mpro) and investigate their likely method of inhibition.


[edit | edit source]

Preparation of the structure and dynamic investigations:

[edit | edit source]

SARS-CoV-2 Mpro's initial structure was retrieved from the PDB under the ID 6Y2E. To compare with wild type, Omicron (BA.1.1.7) was acquired from GISAID. The Omicron variant has only one mutation in the nsp5. Proline is replaced with histidine in residue 132. Using PyMOL, P132H was loaded onto the wild strain. The 2D structures of nimatrelvir and ritonavir were downloaded from PubChem as SDF files, then converted to 3D structures and optimized using PuMOL software. The BIOVIA software was used to remove the hetatm and water molecules from the Mpro and add the polar hydrogen. AutoDock Vina in PyRx were used to perform molecular docking of the proteases and inhibitors. The iInterEvDock2 and SeamDock servers were used to accomplish the double docking of (nirmatrelvir and ritonavir) with Mpro. The most significant technique to judge the validity of the interaction procedure is to compare how well the minimize energy postures predicted by the object scoring function match up. The outcomes of the docking were validated by redocking many times. The optimal stance was determined by employing the MM-GBSA force field.

Estimate docking simulation

[edit | edit source]

The protein-ligand interaction complex structure was assessed using molecular docking simulation. GROMACS and VMD software were used for all of the trajectory simulations and analysis. Initially, 2000 steepest-descent energy was used to minimize energy. The interaction complex was solvated with about 18,000 TIP3P water molecules to analyze the docking postures. This occurred following energy conservation. 0.1ns constraints of heavy protein and ligand atoms were used to achieve mathematical equilibration. The restrictions were released, and the molecular simulation was run for 100 ns. The RMSD from the starting structure was calculated to explore the dynamic stability of poses and to confirm the sampling process's reasoning.

FUpred was used to determine domain borders based on map contact. It has a much superior capacity to anticipate domain boundaries than threading-based approaches and machine learning-based methods, according to benchmark studies. The FUscore with a continuous domain was recorded in the FU-curve and FU-map topologies after analysis.

Illustrate of stereochemical parameter, and Z score-RMS

[edit | edit source]

To calculate the stereochemical parameter, Z score, and RMS, several servers, mathematical equations, and computational programs were employed. Programs were employed to confirm and achieve the intensity of the double docking of Paxlovid's two antiviral drugs with the Mpro.


[edit | edit source]

The docking procedure and the interaction between the protein and the antiviral drugs are carried out. As indicated in Table 1, the autodock program determines the optimal degrees of docking for the two ligands. At a given energy, there are nine alternative models of interaction. When the degree of RMSD/Ib,ub is zero, the optimal model for docking is the first model for both nirmatrelvir and ritonavir. The Mpro active site interaction region is similar to those of the two antiviral drugs, with minor differences. The Gln-110 and Thr-111 residues of Mpro interact with nirmilvir in two conventional hydrogen bonds, whereas the Pro-108 and Gly-109 residues interact with ritonavir. Nonetheless, ritonavir interacts with the two preceding residues (Gln-110 and Thr-111) through two carbon-hydrogen interactions. Figure 1 shows that the two antiviral medicines have additional convergent residuals.

Table 1: Energy and RMSD of 9 models of molecular docking (6Y2E- Nirmatrelvir and 6Y2E- Ritonavir).
Mpro – Ligand interaction Binding Affinity RMSD/ub RMSD/lb
6Y2E- Nirmatrelvir Model 1 -8.1 0 0
6Y2E- Nirmatrelvir Model 2 -7.3 30.262 26.506
6Y2E- Nirmatrelvir Model 3 -7.2 31.754 29.604
6Y2E- Nirmatrelvir Model 4 -7.1 3.177 2.232
6Y2E- Nirmatrelvir Model 5 -7.1 32.018 28.37
6Y2E- Nirmatrelvir Model 6 -7.1 18.233 15.128
6Y2E- Nirmatrelvir Model 7 -7 31.688 29.968
6Y2E- Nirmatrelvir Model 8 -7 33.158 29.419
6Y2E- Nirmatrelvir Model 9 -6.8 18.343 15.526
6Y2E- Ritonavir Model 1 -6.9 0 0
6Y2E- Ritonavir Model 2 -6.4 23.226 19.276
6Y2E- Ritonavir Model 3 -6.4 25.013 19.654
6Y2E- Ritonavir Model 4 -6.4 2.084 1.434
6Y2E- Ritonavir Model 5 -6.3 6.531 2.747
6Y2E- Ritonavir Model 6 -6.3 22.59 18.955
6Y2E- Ritonavir Model 7 -6.3 33.699 26.307
6Y2E- Ritonavir Model 8 -6.3 15.723 9.352
6Y2E- Ritonavir Model 9 -6.3 33.135 25.987
Top: Molecular docking of Mpro with nirmatrelvir (yellow) and magnification of interaction residues shows types of bonds. Down: Molecular docking of Mpro with ritonavir (dark blue) and magnification of interaction residues.

The position of the P132H mutation of the Omicron variant does not effect on the interaction region of the two antiviral drugs, according to this study. The distance between proline or histidine 132 and the carbon atom 6 of nirmatrelvir is 16.6 Å, while the distance was 10.1 Å for ritonavir, Figure 2.

Molecular docking residues and measurements. A: Wild type Mpro with nirmatrelvir. B: Omicron variant Mpro with nirmatrelvir. C: Wild type Mpro with ritonavir. D: Omicron variant Mpro with ritonavir.
Ramachandran plot of the double docking between Mpro and (nirmatrelvir and ritonavir)
Table 2: Distribution number of residues of double docking of Mpro and antiviral drugs.
Position of Residues Residue No.
Favored regions [A,B,L] 244 92.1%
Additional allowed regions [a,b,l,p] 18 6.8%
Generous allowed regions [~a,~b,~l,~p] 2 0.8%
Disallowed regions 1 0.4%
No. of non-glycine and non-proline residues -- --
No. of end-residues (excl. GLY & PRO) 265 100%
No. of glycine residues (shown as triangles) 4
No. of proline residues 13
Total No. 308

The interaction zone may be seen in the active site because ritonavir is a nirmatrelvir booster and the residues of their fusion with Mpro are near together. P132H can be detected at an identical distance between the two antiviral drugs for the Omicron variant without influencing the protein's conformational change, Figure 4.

Molecular docking of Mpro with 2 antiviral drugs (nirmatrelvir and ritonavir).

During the simulation run, we looked at the interaction trend of two critical rings of each medication to see whether there was a probable mechanism of action between nirmatrelvir and ritonavir on SARS-CoV-2 Mpro. These two rings created a significant contact with residues within the active site, which was surprising. These interactions pushed the active site further into the background. The residual sites for the Mpro were analyzed, and it was discovered that all of the sites are graded within the general site average. The arrangement of the residues by 85% at the heat map's median axis verifies this, Figure 5.

Residues index and FU score of Mpro in a stable balance. A: FU-score curve. B: Heat map of FU score. C: Predicted residues index.

Figure (6) depicts a study of ERRAT2 total double docking based on contact atoms for fusion poses and for all protein residues. Two lines are put on the error axis at the top of the figure to illustrate the degree of certainty with which regions exceed error value. The most error bars are in red. Percentage of the Mpro expression for computed error value is less than the rejection level of 95%. The high resolution is produced values of 95% or higher. The typical all quality factors for lower resolutions (2.5 to 3A) are around 91%. Figure 7 depicts the predominance of double docked residues in Mpro and (nirmatrelvir and ritonavir), revealing a convergent distribution of residues with regard to the cutoff line (0.4).

Analysis of double dock of Mpro (chain A) using ERRAT2. The error value is above 95% (red and yellow bars). The overall bets value is under the 95% which is 95.89 quality factor.
Distribution of residues in chain A of double dock.

Various criteria have been used to evaluate the stereochemical quality of the Mpro on localized atoms. The distribution of phi, psi, chi1, and chi2 torsion angles, as well as hydrogen bond energies, are examples of general parameters. There are significant relationships between these characteristics and improved resolution. The parameter distribution becomes increasingly concentrated. Table 3 shows the main measurements of stereochemical quality that provide a direct indicator to a structure's dependability. The side-chain parameters are illustrated in Figure (8).

Table 3: Stereochemical parameters quality of Mpro structure of chi 1 and 2
Stereochemical parameter No. of data pts Parameter value Comparison value No. of band widths from mean
Typical value Band width
a. Chi-1 gauche minus std. deviation 48 128 13.6 6.5 -0.1
b. Chi-1 trans std. deviation 79 18 15.3 5.3 0.5
c. Chi-1 gauche plus std. deviation 123 11.3 13.7 4.9 -0.5
d. Chi-1 pooled std. deviation 250 13.8 14.3 4.8 -0.1
e. Chi-2 trans std. deviation 65 13.9 17.7 5 -0.8
Stereochemical parameters of Mpro shows the side chain.

Volume Z-scores, which are based on atomic volume deviations from standard values, are computed the quality of protein crystal structures. The volume of variance (Z-score-RMS) was calculated. It calculates the average magnitude of the structure's volume abnormalities. According to the current study, the Z-score RMS decreases, which is consistent with the observation that these advancements typically correspond to more accurate models. In structures with a particular resolution or R-factor, the Z-score-RMS distribution is used. Outliers are structures whose Z-score-RMS surpasses specified limitations. The Z-scores have a high correlation with the atomic B-factors. Atoms with absolute Z-scores greater than 3 exist in or near parts of the model where algorithms like PROCHECK detect anomalous stereochemistry. The Z-score analysis of Mpro is measured in Figure (9).

Analysis entire structure of Mpro and distribution of atomic Z-score.


[edit | edit source]

Mpro of SARS-CoV-2 has sparked a lot of interest in therapeutic research to combat the continuing COVID-19 epidemic.[15] The effectiveness of novel and existing antiviral drugs in binding to Mpro's active site has been investigated. According to the molecular docking research, HCV NS3/ 4A protease inhibitors (danoprevir, sovaprevir, glecaprevir, and grazoprevir) bind to SARS-CoV-2 Mpro effectively.[16][17] In another study, the FDA-approved antiviral drugs tipranavir, lopinavir-ritonavir, and raltegravir demonstrated a robust, stable, and flexible binding to Mpro's active site.[18][19] Several studies concentrate on the in silico design of effective medications targeting SARS-CoV-2 Mpro. The therapeutic application of treatments is uncertain, owing to the potential limits of passing clinical trials.

The interaction of nirmatrelvir with Mpro is greater than that of ritonavir, which has several more multiple bonds, according to the docking data. Although ritonavir is a booster, it has certain unfavorable interactions. The interaction of the two antiviral drugs on the active site of the Mpro is crucial, and this demonstrates that this relationship has a significant impact on the decision to use paxlovid as a COVID-19 therapy.[20] The double docked complex of Mpro and (nirmatrelvir and ritonavir) was more stable than the single docked complex of each antiviral medication, since the inhibitors were exposed to bulk water significantly more in the double docked state. These findings suggested that molecular simulation on a long time scale using the all-atom model should be more trustworthy. The RMSD value of the crystal structure and anticipated conformation has been seen to be extensively used as a measure of whether or not the software acquired the correct docking position.

By tracking the percentage occurrence of expected hydrogen bonds across the simulation time, the stability of the hydrogen bonding network predicted by the fusion interaction approach was studied. The investigations of the molecular dynamics trajectories of representative antiviral medications reveal the presence of numerous hydrogen bonds with moderate to high frequency between the Mpro and (nirmatrelvir and ritonavir).

Because the mutation P132H in the Omicron variant's Mpro has no direct influence on the protein, paxlovid may work on new strains, including Omicron, according to this study.

Vaccines and antibody-based treatments are expected to be identified before small molecules. Vaccines, on the other hand, may not be effective, and antibodies may have immunopathological implications. As a result, it's critical to hunt for potential medications that target SARS-CoV-2 Mpro.[21][22] Using existing authorized broad-spectrum medications with the suitable design and potency modifications might be an alternate answer in times of need. Despite enormous efforts in the search for effective inhibitors of SARSCoV-2 Mpro, longitudinal investigations on the treatment safety and efficacy of potential medications are either limited, continuing, or have not yet been disseminated.

Although nirmatrelvir has been licensed for clinical use as an oral antiviral drug, there are still questions about its efficacy and possible hazards.[23] It is important to conduct extensive clinical studies of the new paxlovid to support our conclusion.


[edit | edit source]

Antiviral medicines, both new and old, have been studied for their ability for binding to Mpro's active site. The interaction of the nirmatrelvir and ritonavir in the active site of the Mpro is critical, and this shows that this connection has a substantial influence on whether paxlovid has to be used as a COVID-19 treatment. According to this study, paxlovid may work on new strains, including Omicron, because the mutation P132H in the Omicron variant's Mpro has no direct effect on the protein.


[edit | edit source]
  1. Mahdi, Mohamed; Mótyán, János András; Szojka, Zsófia Ilona; Golda, Mária; Miczi, Márió; Tőzsér, József (2020-12). "Analysis of the efficacy of HIV protease inhibitors against SARS-CoV-2′s main protease". Virology Journal 17 (1): 190. doi:10.1186/s12985-020-01457-0. ISSN 1743-422X. 
  2. Dawood, Ali A. (2022-07-13). "Influence of SARS-CoV-2 variants’ spike glycoprotein and RNA-dependent RNA polymerase (nsp12) mutations on remdesivir docking residues". Medical Immunology (Russia) 24 (3): 617–628. doi:10.15789/1563-0625-IOS-2486. ISSN 2313-741X. 
  3. Soremekun, Opeyemi S; Omolabi, Kehinde F; Adewumi, Adeniyi T; Soliman, Mahmoud ES (2021-01). "Exploring the effect of ritonavir and TMC-310911 on SARS-CoV-2 and SARS-CoV main proteases: potential from a molecular perspective". Future Science OA 7 (1): FSO640. doi:10.2144/fsoa-2020-0079. ISSN 2056-5623. PMID 33432269. PMC PMC7651988. 
  4. Rut, Wioletta; Groborz, Katarzyna; Zhang, Linlin; Sun, Xinyuanyuan; Zmudzinski, Mikolaj; Pawlik, Bartlomiej; Wang, Xinyu; Jochmans, Dirk et al. (2021-02). "SARS-CoV-2 Mpro inhibitors and activity-based probes for patient-sample imaging". Nature Chemical Biology 17 (2): 222–228. doi:10.1038/s41589-020-00689-z. ISSN 1552-4450. 
  5. Dawood, Ali Adel (2022-03). "Increasing the frequency of omicron variant mutations boosts the immune response and may reduce the virus virulence". Microbial Pathogenesis 164: 105400. doi:10.1016/j.micpath.2022.105400. PMID 35077833. PMC PMC8783435. 
  6. Mengist, Hylemariam Mihiretie; Dilnessa, Tebelay; Jin, Tengchuan (2021-03-12). "Structural Basis of Potential Inhibitors Targeting SARS-CoV-2 Main Protease". Frontiers in Chemistry 9: 622898. doi:10.3389/fchem.2021.622898. ISSN 2296-2646. PMID 33889562. PMC PMC8056153. 
  7. Cao, Yunlong; Wang, Jing; Jian, Fanchong; Xiao, Tianhe; Song, Weiliang; Yisimayi, Ayijiang; Huang, Weijin; Li, Qianqian et al. (2022-02-24). "Omicron escapes the majority of existing SARS-CoV-2 neutralizing antibodies". Nature 602 (7898): 657–663. doi:10.1038/s41586-021-04385-3. ISSN 0028-0836. PMID 35016194. PMC PMC8866119. 
  8. Tay, Matthew Zirui; Poh, Chek Meng; Rénia, Laurent; MacAry, Paul A.; Ng, Lisa F. P. (2020-06). "The trinity of COVID-19: immunity, inflammation and intervention". Nature Reviews Immunology 20 (6): 363–374. doi:10.1038/s41577-020-0311-8. ISSN 1474-1733. PMID 32346093. PMC PMC7187672. 
  9. Zivcec, Marko; Safronetz, David; Haddock, Elaine; Feldmann, Heinz; Ebihara, Hideki (2011-05). "Validation of assays to monitor immune responses in the Syrian golden hamster (Mesocricetus auratus)". Journal of Immunological Methods 368 (1-2): 24–35. doi:10.1016/j.jim.2011.02.004. PMID 21334343. PMC PMC3085612. 
  10. University of Mosul; Dawood, A.A.; Altobje, M.A.A.; University of Mosul; Al-Rrassam, Z.T.; University of Mosul (2021-04-17). "Molecular Docking of SARS-CoV-2 Nucleocapsid Protein with Angiotensin-Converting Enzyme II". Mikrobiolohichnyi Zhurnal 83 (2): 82–92. doi:10.15407/microbiolj83.02.082. 
  11. Harvey, William T.; Carabelli, Alessandro M.; Jackson, Ben; Gupta, Ravindra K.; Thomson, Emma C.; Harrison, Ewan M.; Ludden, Catherine; Reeve, Richard et al. (2021-07). "SARS-CoV-2 variants, spike mutations and immune escape". Nature Reviews Microbiology 19 (7): 409–424. doi:10.1038/s41579-021-00573-0. ISSN 1740-1526. PMID 34075212. PMC PMC8167834. 
  12. Zhang, Jinyong; Zeng, Hao; Gu, Jiang; Li, Haibo; Zheng, Lixin; Zou, Quanming (2020-03-29). "Progress and Prospects on Vaccine Development against SARS-CoV-2". Vaccines 8 (2): 153. doi:10.3390/vaccines8020153. ISSN 2076-393X. PMID 32235387. PMC PMC7349596. 
  13. Dawood, Ali A.; Jasim, B. I.; Riadh Al-Jalily, O. (2022-07-13). "Identification of Surface Glycoprotein Mutations of SARS-CoV-2 in Isolated Strains from Iraq". Medical Immunology (Russia) 24 (4): 729–740. doi:10.15789/1563-0625-IOS-2455. ISSN 2313-741X. 
  14. Subbarao, Kanta (2021-07). "The success of SARS-CoV-2 vaccines and challenges ahead". Cell Host & Microbe 29 (7): 1111–1123. doi:10.1016/j.chom.2021.06.016. PMID 34265245. PMC PMC8279572. 
  15. Ullrich, Sven; Ekanayake, Kasuni B.; Otting, Gottfried; Nitsche, Christoph (2022-04). "Main protease mutants of SARS-CoV-2 variants remain susceptible to nirmatrelvir". Bioorganic & Medicinal Chemistry Letters 62: 128629. doi:10.1016/j.bmcl.2022.128629. PMID 35182772. PMC PMC8856729. 
  16. Hoffmann, Markus; Krüger, Nadine; Schulz, Sebastian; Cossmann, Anne; Rocha, Cheila; Kempf, Amy; Nehlmeier, Inga; Graichen, Luise; Moldenhauer, Anna-Sophie (2021-12-13). "The Omicron variant is highly resistant against antibody-mediated neutralization – implications for control of the COVID-19 pandemic". Retrieved 2023-02-07.
  17. Painter, George R; Natchus, Michael G; Cohen, Oren; Holman, Wendy; Painter, Wendy P (2021-10). "Developing a direct acting, orally available antiviral agent in a pandemic: the evolution of molnupiravir as a potential treatment for COVID-19". Current Opinion in Virology 50: 17–22. doi:10.1016/j.coviro.2021.06.003. PMID 34271264. PMC PMC8277160. 
  18. Lim, Liangzhong; Shi, Jiahai; Mu, Yuguang; Song, Jianxing (2014-07-18). Fraternali, Franca. ed. "Dynamically-Driven Enhancement of the Catalytic Machinery of the SARS 3C-Like Protease by the S284-T285-I286/A Mutations on the Extra Domain". PLoS ONE 9 (7): e101941. doi:10.1371/journal.pone.0101941. ISSN 1932-6203. PMID 25036652. PMC PMC4103764. 
  19. Shi, Jiahai; Song, Jianxing (2006-03). "The catalysis of the SARS 3C-like protease is under extensive regulation by its extra domain". FEBS Journal 273 (5): 1035–1045. doi:10.1111/j.1742-4658.2006.05130.x. ISSN 1742-464X. 
  20. Cheng, Shu-Chun; Chang, Gu-Gang; Chou, Chi-Yuan (2010-04). "Mutation of Glu-166 Blocks the Substrate-Induced Dimerization of SARS Coronavirus Main Protease". Biophysical Journal 98 (7): 1327–1336. doi:10.1016/j.bpj.2009.12.4272. PMID 20371333. PMC PMC2849084. 
  21. Spratt, Austin N.; Gallazzi, Fabio; Quinn, Thomas P.; Lorson, Christian L.; Sönnerborg, Anders; Singh, Kamal (2021-04-03). "Coronavirus helicases: attractive and unique targets of antiviral drug-development and therapeutic patents". Expert Opinion on Therapeutic Patents 31 (4): 339–350. doi:10.1080/13543776.2021.1884224. ISSN 1354-3776. PMID 33593200. PMC PMC8074651. 
  22. Owen, Dafydd R.; Allerton, Charlotte M. N.; Anderson, Annaliesa S.; Aschenbrenner, Lisa; Avery, Melissa; Berritt, Simon; Boras, Britton; Cardin, Rhonda D. et al. (2021-12-24). "An oral SARS-CoV-2 M pro inhibitor clinical candidate for the treatment of COVID-19". Science 374 (6575): 1586–1593. doi:10.1126/science.abl4784. ISSN 0036-8075. 
  23. Sasaki, Michihito; Tabata, Koshiro; Kishimoto, Mai; Itakura, Yukari; Kobayashi, Hiroko; Ariizumi, Takuma; Uemura, Kentaro; Toba, Shinsuke et al. (2022-02-15). Oral administration of S-217622, a SARS-CoV-2 main protease inhibitor, decreases viral load and accelerates recovery from clinical aspects of COVID-19 (in en). doi:10.1101/2022.02.14.480338.