Fig 2: IFN-γ-induced control of SARS-CoV-2 infection in human lung epithelial cells is mediated by nitric oxide. (A, B) Representative live-cell imaging analysis of A549-ACE2 cells pretreated with IFN-γ (100 U/mL) with or without IL-1β (2.5 ng/mL) and iNOS inhibitors (1 mM L-NIL, 2 mM L-NAME, and 2 mM D-NAME) for 48 h, and infected with SARS-CoV-2-mNG (green; MOI = 1). Nuclei (blue) were counter-stained with NucBlue. Scale bars, 100 µm. The MFI of SARS-CoV-2-mNG in A549-ACE2 cells was quantified with ImageJ software. Data are means ± SEM (n = 3). Data were analyzed by one-way ANOVA followed by Tukey’s post hoc test (**P < 0.01, ***P < 0.001, and ****P < 0.0001; ns, not statistically significant). (C, D) Representative confocal images of SARS-CoV-2 infected (MOI = 1) A549-ACE2 cells treated with IFN-γ as described above and stained with anti-SARS-CoV-2 nucleocapsid Ab (N; magenta) and anti-iNOS Ab (yellow), or matching isotype controls (C). Nuclei (cyan) were counter-stained with NucBlue. Scale bars, 10 µm. The MFI of SARS-CoV-2 nucleocapsid protein, iNOS, and the two-color colocalization in infected A549-ACE2 cells was quantified with ImageJ software (D). Data are means ± SEM (n = 3). Data were analyzed by one-way ANOVA followed by Tukey’s post hoc test (**P < 0.01, ***P < 0.001, and ****P < 0.0001; ns, not statistically significant).

Fig 3: Schematic overview of the immunological induction of nitric oxide by IFN-γ to control SARS-CoV-2 infection in human pulmonary epithelial cells. Graphical summary of nitric oxide-mediated, cytokine-triggered anti-SARS-CoV-2 viral activity in lung epithelial cells. Our data showed that IFN-γ, which is mainly produced by CD8+ T cells in response to SARS-CoV-2 infection, interfered with viral activity through the induction of iNOS/NOS2 and subsequent production of nitric oxide. The inhibition of SARS-CoV-2 replication cycle induced by IFN-γ was prevented by the treatment with iNOS/NOS2 inhibitors, demonstrating the requirement for nitric oxide in IFN-γ-mediated antiviral responses in human cells.

Fig 4: IFN-γ-triggered killing of SARS-CoV-2 in Vero E6 cells is nitric oxide dependent. (A, B) Representative live-cell imaging analysis of Vero E6 cells pretreated with IFN-γ (100 U/mL) with or without IL-1β (2.5 ng/mL) and iNOS inhibitors (1 mM L-NIL, 2 mM L-NAME, and 2 mM D-NAME) for 48 h, and infected with SARS-CoV-2-mNG (MOI = 0.1). Scale bars, 100 µm. The MFI of SARS-CoV-2-mNG (green) in Vero E6 cells was quantified with ImageJ software. Data are means ± SEM (n = 3). Data were analyzed by one-way ANOVA followed by Tukey’s post hoc test (*P < 0.05 and **P < 0.01; ns, not statistically significant). (C–F) Flow cytometry analysis of SARS-CoV-2-mNG infected Vero E6 cells treated as described above. FlowJo software was used to determine the ▵MFI of SARS-CoV-2-mNG in Vero E6 cells (▵MFI = MFIinfected – MFImock infected). Data are means ± SEM (n = 3). Data were analyzed by one-way ANOVA followed by Tukey’s post hoc test (*P < 0.05, **P < 0.01, and ***P < 0.001; ns, not statistically significant). (G–I) Flow cytometry analysis of SARS-CoV-2-mNG infected Vero E6 cells treated as described above and stained with anti-iNOS Ab conjugated to AF594. (H) FlowJo software was used to determine the % of SARS-CoV-2-mNG+ (green bars), iNOS+ (red bars), and double-positive Vero E6 cells (orange bars). Data are means ± SEM (n = 3). Data were analyzed by one-way ANOVA followed by Tukey’s post hoc test (*P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001; ns, not statistically significant). (I) Scatter plot showing the inverse relationship between SARS-CoV-2-mNG infection and iNOS expression in cytokine-treated Vero E6 cells. Pearson correlation coefficient (r) and P value are indicated in the graph.

Fig 5: IFN-γ induces nitric oxide in SARS-CoV-2-infected Vero E6 cells. (A, B) Representative live-cell imaging analysis of Vero E6 cells pretreated with increasing concentrations of IFN-γ (50, 100, 200, 400 U/mL), IL-1β (1.25, 2.5, 5, or 10 ng/mL), or IFN-γ in combination with IL-1β (50/1.25, 100/2.5, 200/5, or 400/10, respectively) for 48 h, infected with SARS-CoV-2 (MOI = 0.1), and labeled with nitric oxide indicator DAF-FM (green). Displayed images represent cytokines at the highest concentrations (A). Scale bars, 50 µm. (B) The MFI of DAF-FM in Vero E6 cells was quantified with ImageJ software. M, media. Data are means ± SEM (n = 6). Data were analyzed by one-way ANOVA followed by Tukey’s post hoc test (*P < 0.05 and **P < 0.01). All cytokine concentrations were significantly different from the media control. (C–E) Flow cytometry analysis of DAF-FM. Infected Vero E6 cells were treated and stained as described in (A). (D) FlowJo software was used to determine the % of infected Vero E6 cells positive for DAF-FM. M, media. Data are means ± SEM (n = 6). Data were analyzed by one-way ANOVA followed by Tukey’s post hoc test (**P < 0.01 and ***P < 0.001). All cytokine concentrations were significantly different from the media control. (E) Dose-response curves of DAF-FM signal induced by cytokines treatment. Data are means ± SEM (n = 6). Data were analyzed by two-way ANOVA followed by Tukey’s post hoc test (***P < 0.001 and #### P < 0.0001; ns, not statistically significant). (F, G) Vero E6 cells were treated with IFN-γ (400 U/ml), IL-1β (10 ng/mL), or IFN-γ in combination with IL-1β (400/10, respectively) for 48 h, mock infected or infected with SARS-CoV-2-mNG (green; MOI = 0.1), and then stained with anti-iNOS (red) Ab or isotype control and observed by fluorescent confocal microscopy (F). Yellow denotes colocalization between green and red channels. Scale bars, 10 µm. The MFI of SARS-CoV-2-mNG (left), iNOS (middle), and the two-color colocalization (right) in infected Vero E6 cells was quantified with ImageJ software (G). Data are means ± SEM (n = 6). Data were analyzed by one-way ANOVA followed by Tukey’s post hoc test (***P < 0.001 and ****P < 0.0001).