All-Trans Retinoic Acid Exhibits Antiviral Effect against SARS-CoV-2 by Inhibiting 3CLpro Activity

Abstract

The pandemic of COVID-19 caused by SARS-CoV-2 continues to spread despite the global efforts taken to control it. The 3C-like protease (3CLpro), the major protease of SARS-CoV-2, is one of the most interesting targets for antiviral drug development because it is highly conserved among SARS-CoVs and plays an important role in viral replication. Herein, we developed high throughput screening for SARS-CoV-2 3CLpro inhibitor based on AlphaScreen. We screened 91 natural product compounds and found that all-trans retinoic acid (ATRA), an FDA-approved drug, inhibited 3CLpro activity. The 3CLpro inhibitory effect of ATRA was confirmed in vitro by both immunoblotting and AlphaScreen with a 50% inhibition concentration (IC₅₀) of 24.7 ± 1.65 µM. ATRA inhibited the replication of SARS-CoV-2 in VeroE6/TMPRSS2 and Calu-3 cells, with IC₅₀ = 2.69 ± 0.09 µM in the former and 0.82 ± 0.01 µM in the latter. Further, we showed the anti-SARS-CoV-2 effect of ATRA on the currently circulating variants of concern (VOC); alpha, beta, gamma, and delta. These results suggest that ATRA may be considered as a potential therapeutic agent against SARS-CoV-2.

Keywords: 3CL protease; FDA-approved drug; SARS-CoV-2; all-trans retinoic acid; antiviral efficacy; high throughput screening.

Figure 1

Development of SARS-CoV-2 3CLpro inhibitor screening assay using AlphaScreen. (A) Schematic representation of AlphaScreen-based assay for SARS-CoV-2 3CLpro enzymatic activity. A high signal is detected when the substrate is not cleaved by 3CLpro, and the signal becomes weaker when the substrate is cleaved. (B) Results of an enzymatic assay using wild-type SARS-CoV-2 3CLpro and its inactive mutant, C145A. 3CLpro decreases the signal when the substrate is cleaved. Substrate cleavage was also detected by immunoblots. Error bars obtained from duplicate testing. p-value < 0.005 (***). WT-Wild type. (C) Optimization of the enzyme assay. A concentration gradient of 3CLpro was reacted with 100 nM of the recombinant substrate and the signal was detected. Error bars obtained from duplicate testing. (D) The inhibitory effect of 3CLpro using the known SARS-CoV-2 3CLpro inhibitor GC376. Error bars obtained from duplicate testing. (E) Screening of 91 compounds using the 3CLpro enzyme assay. The substrate and enzyme concentration was 100 nM in this experiment. GC376 was used as a positive control. ATRA showed in the red bar. Error bars obtained from duplicate testing.

Figure 2

ATRA as a potent SARS-CoV-2 3CLpro inhibitor (A) The structure of retinoids used in this study. (B) The results of enzyme assay for different retinoids (ATRA; colored red). Error bars obtained from duplicate testing. (C) In vitro cleavage assay, the recombinant substrate was cleaved by SARS-CoV-2 3CLpro, which was confirmed by immunoblot. A cleaved substrate formed a band right below the substrate band. Substrate cleavage by SARS-CoV-2 3CLpro was inhibited by GC376 (100 µM) and ATRA (0.1–100 µM). (D) Results of the enzyme assay for ATRA in a concentration gradient (0–100 µM, colored red). Error bars obtained from duplicate testing. (E,F) Docked poses of ATRA and SARS-CoV-2 3CLpro showed as surface (E) and ribbon (F). The catalytic residues are His ⁴¹ (colored green) and Cys ¹⁴⁵ (colored red). The substrate-binding region is 163–167 and 178–192 (colored blue).

Figure 3

ATRA inhibits SARS-CoV-2 replication (A) Assessing the antiviral effect of ATRA in VeroE6/TMPRSS2 cells. VeroE6/TMPRSS2 cells treated with ATRA (0.1–100 µM) for 3 h prior to infection were infected with SARS-CoV-2 (MOI = 0.05) and washed off the unbound virus at 2 h of infection. Cell viability assay and RT-PCR were performed on the attached cells and supernatant culture medium respectively, 48 h after infection. (B) The results of the cell viability assay. IC₅₀ was calculated for the inhibition of SARS-CoV-2 induced cell death. Error bars obtained from triplicate testing. p-value < 0.005 (***) (C) Viral quantification by RT-PCR. Error bars obtained from duplicate testing. p-value < 0.005 (***) (D) Immunostaining of SARS-CoV-2 N protein in infected cells. VeroE6/TMPRSS2 cells treated with ATRA (1, 10, 25 µM) 3 h before infection were infected with SARS-CoV-2 (MOI = 0.05). After 24 h of infection, cells were fixed with 4% paraformaldehyde and immunostained. Red stained cells represent SARS-CoV-2 N protein, blue-stained nuclei with DAPI. (E) Schematic representation of the time of addition assay of ATRA. (F-H) The results of RT-PCR for ATRA (F), Remdesivir (G) and Camostat (H). The result of full-time colored grey, entry colored blue, and post-entry colored red. Error bars obtained from duplicate testing. p-value < 0.005 (***).

Figure 4

ATRA is effective against the SARS-CoV-2 variants of concern in the human lung cell line (A) Calu-3 cells, pre-incubated with ATRA at 37 °C for 3 h, were infected with SARS-CoV-2; A, alpha, beta and gamma (MOI = 0.5) for 2 h. Following this, the non-adsorbed virus was washed, and the infected cells were incubated in a fresh medium containing ATRA for 72 h. (B–F) Inhibition of different VOC by ATRA. The results of RT-PCR for lineage A (B), alpha strain (C), beta strain (D), gamma strain (E), and delta strain (F). Error bars obtained from duplicate testing. p-value < 0.005 (***).