Comparative assessment of favipiravir and remdesivir against human coronavirus NL63 in molecular docking and cell culture models

Abstract

Human coronavirus NL63 (HCoV-NL63) mainly affects young children and immunocompromised patients, causing morbidity and mortality in a subset of patients. Since no specific treatment is available, this study aims to explore the anti-SARS-CoV-2 agents including favipiravir and remdesivir for treating HCoV-NL63 infection. We first successfully modelled the 3D structure of HCoV-NL63 RNA-dependent RNA polymerase (RdRp) based on the experimentally solved SARS-CoV-2 RdRp structure. Molecular docking indicated that favipiravir has similar binding affinities to SARS-CoV-2 and HCoV-NL63 RdRp with LibDock scores of 75 and 74, respectively. The LibDock scores of remdesivir to SARS-CoV-2 and HCoV-NL63 were 135 and 151, suggesting that remdesivir may have a higher affinity to HCoV-NL63 compared to SARS-CoV-2 RdRp. In cell culture models infected with HCoV-NL63, both favipiravir and remdesivir significantly inhibited viral replication and production of infectious viruses. Overall, remdesivir compared to favipiravir is more potent in inhibiting HCoV-NL63 in cell culture. Importantly, there is no evidence of resistance development upon long-term exposure to remdesivir. Furthermore, combining favipiravir or remdesivir with the clinically used antiviral cytokine interferon-alpha resulted in synergistic effects. These findings provided a proof-of-concept that anti-SARS-CoV-2 drugs, in particular remdesivir, have the potential to be repurposed for treating HCoV-NL63 infection.

Figure 1

Binding mode of favipiravir to the SARS-CoV-2 and HCoV-NL63 RdRp. (A) Favipiravir, bound to the SARS-CoV-2 RdRp (atom color ribbons) binding site, was depicted as surface representation with H-bond donor (purple) and acceptor (green). (B) Molecular interactions of favipiravir (atom color sticks) with the SARS-CoV-2 RdRp residues (yellow sticks). (C) 2D diagram of interactions between SARS-CoV-2 RdRp and favipiravir. (D) Favipiravir, bound to the HCoV-NL63 RdRp (atom color ribbons) binding site, was depicted as surface representation with H-bond donor (purple) and acceptor (green). (E) Molecular interactions of favipiravir (atom color sticks) with the HCoV-NL63 RdRp residues (yellow sticks). (F) 2D diagram of interactions between HCoV-NL63 RdRp and favipiravir.

Figure 2

Binding mode of remdesivir to the SARS-CoV-2 and HCoV-NL63 RdRp. (A) Remdesivir, bound to the SARS-CoV-2 RdRp (atom color ribbons) binding site, was depicted as surface representation with H-bond donor (purple) and acceptor (green). (B) Molecular interactions of remdesivir (atom color sticks) and the SARS-CoV-2 RdRp residues (yellow sticks). (C) 2D diagram of interactions between SARS-CoV-2 RdRp and remdesivir. (D) Remdesivir, bound to the HCoV-NL63 RdRp (atom color ribbons) binding site, was depicted as surface representation with H-bond donor (purple) and acceptor (green). (E) Molecular interactions of remdesivir (atom color sticks) and the HCoV-NL63 RdRp residues (yellow sticks). (F) 2D diagram of interactions between HCoV-NL63 RdRp and remdesivir.

Figure 3

The root-mean-square deviations (RMSDs) of the backbone atoms and its ligands complex relative to their coordinates as a function of simulation time. (A) The average RMSDs of SARS-CoV-2 RdRp backbone atoms (Blue line) and its complexes with favipiravir (Green Line) or remdesivir (Red line) were plotted against the total simulation time. (B) The average RMSDs of HCoV-NL63 RdRp backbone atoms (Blue line) and its complexes with favipiravir (Green Line) or remdesivir (Red line) were plotted against the total simulation time.

Figure 4

Antiviral effects of favipiravir against HCoV-NL63 in cell culture models. (A) Dose-dependent inhibition of HCoV-NL63 replication in LLC-MK2 cell line by favipiravir treatment. Intracellular viral RNA quantified by qRT-PCR was normalized to housekeeping gene GAPDH and presented relative to the control (CTR) (set as 1) (n = 6). (B) Immunofluorescence microscopy analysis of dsRNA, the intermediate of replicating HCoV-NL63 genomic RNA, upon treatment of different concentrations of favipiravir in LLC-MK2 and Caco-2 cell lines. Nuclei were visualized by DAPI (blue). (C) Dose-dependent inhibition of HCoV-NL63 replication by favipiravir in Caco-2 cell line (n = 6). (D) Caco-2 cells were infected with 0.1 MOI HCoV-NL63 and treated with different concentrations of favipiravir for 48 h. 50% effective concentration (EC50) curve was quantified by qRT-PCR, and 50% cytotoxic concentration (CC50) curve was determined by MTT assay. The left and right Y-axis of the graphs represent mean % inhibition of virus yield and cytotoxicity of the drug, respectively (n = 6–8). (E) Caco-2 cells were infected with 0.1 MOI HCoV-NL63, and then untreated or treated with 10 μM favipiravir for 5 days. Supernatant was collected every day to quantify secreted viruses by qRT-PCR, calculated as genomic copy numbers (n = 6). Standard curve for calculation of genomic copy numbers is included in Supplementary Fig. S3. (F) Caco-2 cells were infected with 0.1 MOI HCoV-NL63, and then untreated or treated with 1 or 100 μM favipiravir for 48 h. Virus titers from different groups was determined by TCID50 assay (n = 6). Data represent as mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 5

Antiviral effects of remdesivir against HCoV-NL63 in vitro. (A) Dose-dependent inhibition of HCoV-NL63 replication by remdesivir in LLC-MK2 cell line. Intracellular viral RNA quantified by qRT-PCR was normalized to housekeeping gene GAPDH and presented relative to the control (CTR) (set as 1) (n = 6–8). (B) Immunofluorescence microscopy analysis of dsRNA, the replicating HCoV-NL63 genomic RNA, upon treatment of different concentrations of remdesivir in LLC-MK2 and Caco-2 cell lines. Nuclei were visualized by DAPI (blue). (C) Dose-dependent inhibition of HCoV-NL63 replication by remdesivir on in Caco-2 cell line (n = 6). (D) Caco-2 cells were infected with 0.1 MOI HCoV-NL63 and treated with different concentrations of remdesivir for 48 h. 50% effective concentration (EC50) curve was quantified by qRT-PCR, and 50% cytotoxic concentration (CC50) curve was determined by MTT assay. The left and right Y-axis of the graphs represent mean % inhibition of virus yield and cytotoxicity of the drug, respectively (n = 6–8). (E) Caco-2 cells were infected with 0.1 MOI HCoV-NL63, and then untreated or treated with 1 μM remdesivir for 5 days. Supernatant was collected every day to quantify secreted viruses by qRT-PCR and calculated as genomic copy numbers (n = 6). Standard curve for calculation of genomic copy numbers is included in Supplementary Fig. S3. (F) Caco-2 cells were infected with HCoV-NL63 at an MOI of 0.1, then untreated or treated with 0.3 or 3 μM remdesivir for 48 h. Virus titers from different groups was determined by TCID50 assay (n = 6). Data represent as mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001. (G) Comparing the inhibitory potency of favipiravir and remdesivir in Caco-2 cells infected with HCoV-NL63. (H) HCoV-NL63 was serially passaged in Caco-2 cells exposed to no remdesivir (as control) or increasing concentrations of remdesivir for 20 passages. 1 μM remdesivir was used in passage 1–10, which was increased to 2 μM at the subsequent passages. The effect of remdesivir (1 μM) on HCoV-NL63 harvested at passage 5, 10, 15 and 20 was quantified using qRT-PCR. Data represent as mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 6

The effects of combining antivirals in Caco-2 cells infected with HCoV-NL63. The antiviral effects of favipiravir in combination with remdesivir (A), or IFN-α in combination with favipiravir (B) or remdesivir (C) respectively. The results were analyzed by the MacSynergyII model. The three-dimensional surface plot represents the differences (within 95% confidence interval) between actual experimental effects and theoretical additive effects of the combination at various concentrations (n = 4).