In December 2019, a new coronavirus disease (coronavirus disease 2019, COVID-19) emerged suddenly in China [1
]. COVID-19 spread rapidly, resulting in a pandemic [2
]. Over 34 million people were confirmed COVID-19-positive as of the end of September 2020 [2
]. The agent is classified into genus Betacoronavirus
, based on detailed genome analyses, and formally named severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) [3
]. About 10–20% of COVID-19 cases may cause fever, fatigue, cough, and pneumonia, while the infection in some cases may result in inapparent symptoms [4
]. However, some cases of COVID-19 may be complicated acute respiratory distress syndrome (ARDS) leading to death [5
]. Thus, a need exists for the early development of new therapeutic drugs and applications of existing drugs for the treatment of COVID-19. To date, some antiviral agents including ciclesonide, remdesivir, and favipiravir have been tried to treat COVID-19 [6
Inhibition of viral replication is assumed to be the mechanism for therapeutic agents for SARS-CoV-2. RNA-dependent RNA polymerase (RdRp) of SARS-CoV-2 is nonstructural protein 12 (NSP-12) and is similar to SARS-CoV RdRp [7
]. NSP-7 and NSP-8 are known cofactors of SARS-CoV-2 RdRp proteins. Further, NSP-15 protein is an endonuclease from SARS-CoV-2 that plays an important role in the proofreading of viral RNA. Thus, these proteins may be targets for antiviral drugs for treating COVID-19.
Favipiravir (6-fluoro-3-hydroxypyrazine-2-carboxamide, Avigan®
) was initially developed as an antiviral agent for the treatment of influenza. The mechanism of favipiravir is inhibition of viral RNA replication by inhibition of RdRp formed as a complex of PA protein, PB1 protein, and PB2 protein [8
]. Subsequently, favipiravir was reported to have similar activity against RdRp proteins from RNA viruses other than influenza and showed efficacy for treatment of ebolavirus disease (EVD), Lassa fever, and norovirus infections [12
]. Thus, favipiravir might also be effective against some RNA virus infections, and is currently in clinical evaluation as a treatment for COVID-19 [6
]. However, no detailed molecular interactions between favipiravir and SARS-CoV-2 proteins, such as RdRp and NSP-15 endonuclease, are exactly known. With this background, we performed an in silico study regarding the molecular interactions among favipiravir, influenza RdRp, SARS-CoV-2 RdRp, SARS-CoV RdRp, Middle East respiratory syndrome coronavirus (MERS-CoV) RdRp, and the coronaviruses’ NSP-15 endonuclease.
We showed in the present in silico study that F-RTP (the active form of favipiravir) could bind to RdRp active sites of SARS-CoV-2, SARS-CoV, and MERS-CoV in the presence of the agent and protein. Moreover, the F-RTP bound to the replicated RNA termini in the presence of the agent, magnesium ions, nucleotide triphosphate, and RdRp proteins. Conversely, F-RTP did not bind to PB1 (RdRp) active sites of influenza virus H1N1 in the presence of agent and protein. Further, F-RTP may bind to the tunnel of the PB1 protein leading to the inhibition of the replicated RNA passage. Thus, F-RTP displays distinct pharmacological effects on various coronaviruses and influenza virus subtype AH1N1.
Favipiravir is a nucleic acid analog and was developed as a therapeutic drug to be used to treat influenza [25
]. The drug was approved in Japan in 2014 and was stockpiled for use in the event of a new influenza epidemic. Favipiravir is activated by intracellular conversion to F-RTP by hypoxanthine-guanine phosphoribosyltransferase [8
]. Previous reports revealed that the actions of favipiravir/F-RTP against influenza involves the termination of genome replication, although the detailed molecular interactions between the drug and the PB1 polymerase had not been fully elucidated. Favipiravir is currently undergoing clinical evaluation for use in treating COVID-19 in some countries, including Japan. Some reports suggest that favipiravir is effective in this setting [26
], although the molecular mechanisms underlying drug efficacy against SARS-CoV-2 have not been fully explored. To date, a few antiviral agents against coronaviruses have been approved. However, our results suggest favipiravir may show antiviral activity against SARS-CoV-2, SARS-CoV, and MERS-CoV, though the present study was purely in silico. Thus, further clinical studies may be needed to demonstrate efficacy.
We examined the presence of the agent and coronavirus RdRp proteins and also the presence of the agent, magnesium ions, nucleotide triphosphate, and the viral RdRp proteins. The agent could bind to RdRp proteins and could inhibit genome replication. A previous report showed that the agent inhibits SARS-CoV-2 in vitro [28
]. This study suggests that inhibition of genome replication is termination [28
]. However, detailed molecular interactions between the agent and the viral replication systems may not currently be known. Thus, the present molecular pharmacological results may be the first observations.
Further, previous reports suggest that the antiviral effect of the agent was premature termination of genome replication [8
]. In our study, F-RTP could bind to the replicated RNA termini of influenza RdRp proteins, suggesting that the inhibition of the genome replication mechanism of influenza virus is termination. This may be compatible with earlier reports [8
]. The present in silico study suggests that F-RTP does not bind to PB1 protein active sites of influenza virus subtype AH1N1 in the presence of agent and protein, but does suggest that F-RTP may bind to the tunnel of the protein. This binding may result in inhibition of replicated RNA passage. Based on the results and speculation, we suggest that favipiravir may exhibit two mechanisms of antiviral activity. This observation may also be a first.
The half-maximal effective concentration (EC50
) of favipiravir against SARS-CoV-2 is 61.88 μΜ (9.4 μg/mL), which is comparable to Ebola virus (10.8–63 μg/mL) but higher than EC50′
s of 0.030–0.46 μg/mL for influenza virus [25
]. Thus, the clinical dose of favipiravir for COVID-19 will be higher than the dose for influenza but comparable to the dose for Ebola hemorrhagic fever [34
]. However, the underlying cause of differences in EC50
for influenza and SARS-CoV-2 is unknown. The difference in mechanisms of action of favipiravir in influenza and SARS-CoV-2 may explain this difference in EC50
Favipiravir binding to proteins other than SARS-CoV-2 RdRp has been investigated [35
]. NSP-15 protein is an endonuclease from SARS-CoV-2 which plays an important role in the replication of viral RNA. We previously reported that ciclesonide inhibits viral replication in SARS-CoV-2 by binding to active sites of NSP-15 [19
]. Hence, we also examined a docking simulation for interactions between favipiravir/F-RTP and NSP-15. However, favipiravir/F-RMP did not bind to these active sites. Thus, NSP-15 is not involved in differences in EC50
of favipiravir between influenza virus and SARS-CoV-2.
In conclusion, we found, in silico, that favipiravir/F-RTP could bind to active sites of coronavirus RdRp proteins and replicated RNA termini. We also showed that F-RTP binds near the tunnel of influenza RdRp protein. Distinct mechanisms underlying favipiravir-mediated interactions with influenza RdRp and coronavirus RdRp may help explain the need for different doses of the drug for effective clinical responses for treating influenza vs. SARS-CoV-2 infections.