**3. Results**

#### *3.1. Inhibition of GA against SARS-CoV-2 Infection against S Protein-Pseudotyped Virus*

We used an S protein-pseudotyped lentivirus to determine whether GA had an antiviral effect against SARS-CoV-2. The lentivirus system is easy to construct and has been widely used for virus binding and infection assay. To this end, cells were infected with varying amounts of Lenti-S in the absence or presence of varying amounts of GA (Figure 1A). We tested GA at concentrations of 0.5–5 mM since previous studies showed that GA was previously reported active against a wide range of enveloped viruses at concentrations of 1–8 mM [10,12]. Vero E6 cells in 96-well plates were untreated or treated with GA at 0.5, 1, 2.5, and 5 mM 30 min prior to Lenti-S infection, and we found that GA treatment resulted in a reduction in luciferase activity (Figure 1B). At 2.5 and 5 mM, GA treatment reduced Lenti-S-mediated luciferase gene delivery by approximately 77% and 92%, respectively. The effect of GA on Lenti-S infection was specific since treatment of 293T cells transfected with pLenti-CMV-luc did not affect luciferase expression (Figure 1C).

**Figure 1.** *Cont*.

**Figure 1.** GA effect on Lenti-S infection. (**A**) Molecular structure of glycyrrhizic acid. (**B**) Effect of GA on Lenti-S infection. Vero E6 cells were infected with Lenti-S pseudovirus in the absence or presence of GA at indicated concentrations for 24 h. Luciferase activity was determined. Data are expressed as a percentage of untreated controls (Lenti-S). The experiment was performed twice, and data are mean ± SD of triplicate wells. Independent two-sample comparisons between non-GA treated sample and sample treated with GA at different concentrations, respectively, were determined by Student's t test. ns: no significance; \*, *p* < 0.05; \*\*, *p* < 0.01. (**C**) Effect of GA on luciferase expression in pCMV-luc-transfected cells. Monolayers of 293T cells were transfected with pCMV-luc for 16 h. The cells were then treated with GA at indicated concentrations. Luciferase activity was determined after 24 h incubation. GA treatment did not affect luciferase expression delivered by an encoding plasmid. The experiment was performed twice. The readings from untreated samples were used as a control for the calculation of relative luciferase activity. Data are mean ± SD of duplicate wells from 2 independent experiments. (**D**) Time of GA addition on Lenti-S-mediated luciferase gene delivery. Vero E6 cells were untreated or treated with 3 mM GA at 2 h prior to (−2 h), during (0), or at 2 and 4 h post Lenti-S infection. Luciferase activity was determined 24 h PI. The experiment was performed 2 times. Data from the untreated controls were used for the calculation of relative luciferase activity. \*\*, *p* < 0.01.

The time effect of GA addition was tested by treating Vero E6 cells with 3 mM GA at 2 h prior to (−2 h), during (0 h), or at 2 h and 4 h post Lenti-S infection. Luciferase expression was determined at 24 h PI. As shown in Figure 1D, GA addition prior to or during Lenti-S inoculation significantly blocked Lenti-S infection. For comparison, addition of GA at 2 and 4 h post Lenti-S inoculation showed diminished effect against Lenti-S infection. This result suggests that GA likely targeted the early stages of Lenti-S infection.
