Fig 2: The IFN-β and sACE2 domains of IFNβ-ACE2 exhibited predicted bioactivities. (A) To assay the IFN-β domain, TF-1 cells were incubated with GM-CSF and either IFN-β, sACE2(18-611), or IFNβ-ACE2 then pulsed with [3H]thymidine during the last 24 h of a 3-day culture. The y-axis represents counts per minute (CPM), and error bars represent the SD. Statistical significance was analyzed by use of two-way ANOVA with Tukey’s multiple comparisons test comparing the three control groups to the IFNβ-ACE2 treatment group at each concentration (ns nonsignificant, * p < 0.05, **** p < 0.0001). (B–E) To assay the ACE2 domain, HEK-Spike or HEK-control cells were incubated with designated concentrations of either sACE2(18-611) or IFNβ-ACE2 for 1 h at 4 °C. After washing, cells were stained with AF647-conjugated anti-human ACE2 antibody for 1 h at 4 °C. Cells were analyzed for ACE2 binding by flow cytometry. (B) Viable, single, live, and GFP+ stably transfected cells (parental gate representing all cells in plot) were subgated to show the ACE2+ subset. Representative dot plots show binding of either sACE2(18-611) or IFNβ-ACE2 (2 μM each) to HEK-Spike or HEK-control cells. (C) Shown are percentages of ACE2+ HEK-Spike cells (ACE2+ gate/parental gate). (D) The MFIs of anti-ACE2 fluorescence are shown for the parental gate. (E) Bar graphs show mean percentages of HEK-Spike or HEK-control cells bound to ACE2 (ACE2+ gate/parental gate). Each data point represents the mean value (n = 2), and error bars represent SD. These data are representative of three independent experiments.

Fig 3: In a non-washed in vitro infection system, IFNβ-ACE2 exhibited enhanced antiviral activity compared to IFN-β alone, ACE2 alone, or the unlinked combination. NL63 was incubated for 1 h at 4 °C with either IFNβ-ACE2, sACE2(18-611), sACE2(18-740), recombinant IFN-β, IFN-β (Peprotech), or the unlinked combination of sACE2(18-611) and IFN-β. In contrast to experiments shown in Figure 3 and Figure 4, we omitted the virus-washing step. The NL63 + protein mixtures were added to Vero E6-TMPRSS2-T2A-ACE2 cells in a 96-well plate. The cells were harvested after a 2-day incubation at 33 °C and stained with LIVE/DEAD Fixable Blue Dead Cell Stain. Cells were surface-stained with FITC-conjugated mouse anti-human ACE2, were fixed and permeabilized, and then were intracellularly stained with AF647-conjugated rabbit anti-NL63 nucleocapsid antibody. Cells were then analyzed for viral infection by flow cytometry. Cells gated as viable, single, and live cells (parent gate) were subgated to define nucleocapsid+ and ACE2high subsets. Shown (A,D) are representative dot plots showing percentages of the nucleocapsid+ subset at the 1 pM concentration. Shown (B,F) are the percentages of the nucleocapsid+ subset for each group over concentrations ranging from 100 fM to 1 μM. Bar graph (C) shows mean percentage values of nucleocapsid+ cells at the 1 pM concentration. Shown (E) are representative dot plots showing percentages of the ACE2high subset at the 1 pM concentration. Shown (G) are the percentages of ACE2high subset for each group over concentrations ranging from 100 fM to 1 μM. Each data point represents the mean value (n = 2), and error bars represent SD. Statistical significance was analyzed by (C) one-way ANOVA with the Dunnett multiple comparisons test or (F,G) two-way ANOVA with Tukey’s multiple comparisons test comparing the unlinked combination of IFN-β and ACE2 treatment groups to the IFNβ-ACE2 treatment group at each concentration (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001). These data are representative of three independent experiments.

Fig 4: The covalent linkage of IFN-β and ACE2 was required for IFN-β targeting to NL63. NL63 was incubated at 4 °C with either IFNβ-ACE2 or the unlinked combination of sACE2(18-611) and IFN-β. After a 1 h incubation, NL63 was repeatedly washed with 300kD centrifugal filters to remove proteins that lacked binding to virions. The retentates, which included virions and virion-bound proteins, were added to Vero E6-TMPRSS2-T2A-ACE2 cells in a 96-well plate. Cells were harvested after a 2-day incubation at 33 °C and stained with LIVE/DEAD Fixable Blue Dead Cell Stain. Cells were surface-stained with PE-conjugated mouse anti-human TMPRSS2 and FITC-conjugated mouse anti-human ACE2. After fixation and permeabilization, cells were stained with AF647-conjugated rabbit anti-NL63 nucleocapsid antibody. Cells were analyzed for viral infection by flow cytometry. Viable, single, and live cells in the parental gate were subgated as the nucleocapsid+ subset (A), the ACE2high subset (C), and the TMPRSS2high subset (F) as shown for the 1 nM concentration value. Shown are the percentages of nucleocapsid+, ACE2high, and TMPRSS2high subsets together with the respective MFI values (B,D,E), and (G,H), respectively. Cell percentages were calculated by dividing the events in the subset-positive/high subgate by those in the parental gate. MFI values were gated on all viable, single, and live cells (i.e., cells in the parental gate). Each data point represents the mean value (n = 2), and error bars represent SD. Statistical significance of the IFNβ-ACE2 versus the ‘IFN-β + sACE2’ treatment group at each concentration was analyzed by use of two-way ANOVA with Tukey’s multiple comparisons test (** p < 0.01, *** p < 0.001, **** p < 0.0001). Statistical significance was also assessed for treatment groups at each concentration compared to the ‘blank’ control via two-way ANOVA with Tukey’s multiple comparisons test (°° p < 0.01, °°° p < 0.001, °°°° p < 0.0001). These data are representative of three independent experiments.

Fig 5: IFNβ-ACE2 exhibited virus-specific targeting in accordance with viral host receptor specificity. The NL63 or 229E viruses were incubated at 4 °C with either IFNβ-ACE2, sACE2(18-611), IFN-β, or the unlinked combination of IFN-β and sACE2(18-611). After a 1 h incubation, the virus + protein mixtures were washed, and the retentate containing virion–protein complexes was used for infection of Vero E6-TMPRSS2-T2A-ACE2 or A549 cells, respectively, in a 96-well plate (A,B). After the washing step, IFNβ-ACE2 or the unlinked combination of IFN-β and sACE2(18-611) were directly added to designated groups (B). Alternatively, the virus + protein mixtures were not subjected to a virus-washing step and the mixtures were used for infection of the respective host cells (C–E). The cells were harvested after a 2-day incubation and stained with LIVE/DEAD Fixable Blue Dead Cell Stain. After fixation and permeabilization, Vero E6-TMPRSS2-T2A-ACE2 cells were stained with AF647-conjugated rabbit anti-NL63 nucleocapsid antibody and A549 cells were stained with AF647-conjugated rabbit anti-229E nucleocapsid antibody. Cells were then analyzed for viral infection by flow cytometry. Cells were gated on viable, single, and live cells before subgating on nucleocapsid+ cells. Shown (A) are the percentages of nucleocapsid+ cells for each treatment group normalized to the ‘no protein’ control group (1 nM concentrations). (B) Bar graph shows mean percentages of nucleocapsid+ 229E-infected cells when proteins were or were not added after the washing step (1 nM concentrations). Shown (C) are representative dot plots including percentages of nucleocapsid+ 229E-infected cells (1 nM concentrations). Shown (D) are the percentages of nucleocapsid+ cells for each treatment group normalized to the ‘no protein’ group for each virus (1 μM concentrations). Shown (E) are the percentages of nucleocapsid+ 229E-infected cells for each treatment group at designated concentrations (100 fM to 100 nM). Each data point represents the mean value (n = 2), and error bars represent SD. Statistical significance was analyzed by use of two-way ANOVA with Tukey’s multiple comparisons test comparing the IFN-β and/or ACE2 treatment groups to the IFNβ-ACE2 treatment group at each concentration (unless otherwise noted in the figure) (ns nonsignificant, **** p < 0.0001). Experiments shown are representative of three independent experiments.