Furin Inhibitors Block SARS-CoV-2 Spike Protein Cleavage to Suppress Virus Production and Cytopathic Effects

Abstract

Development of specific antiviral agents is an urgent unmet need for SARS-coronavirus 2 (SARS-CoV-2) infection. This study focuses on host proteases that proteolytically activate the SARS-CoV-2 spike protein, critical for its fusion after binding to angiotensin-converting enzyme 2 (ACE2), as antiviral targets. We first validate cleavage at a putative furin substrate motif at SARS-CoV-2 spikes by expressing it in VeroE6 cells and find prominent syncytium formation. Cleavage and the syncytium are abolished by treatment with the furin inhibitors decanoyl-RVKR-chloromethylketone (CMK) and naphthofluorescein, but not by the transmembrane protease serine 2 (TMPRSS2) inhibitor camostat. CMK and naphthofluorescein show antiviral effects on SARS-CoV-2-infected cells by decreasing virus production and cytopathic effects. Further analysis reveals that, similar to camostat, CMK blocks virus entry, but it further suppresses cleavage of spikes and the syncytium. Naphthofluorescein acts primarily by suppressing viral RNA transcription. Therefore, furin inhibitors may be promising antiviral agents for prevention and treatment of SARS-CoV-2 infection.

Keywords: SARS-CoV-2; cytopathic effect; furin; spike; syncytium.

Graphical abstract

Figure 1

Cleavage of the SARS-CoV-2 S Protein at the Putative Furin Substrate Site Is Critical for Syncytium Formation in S-Only-Expressing VeroE6 Cells (A) Top panel: schematic illustration of the SARS-CoV-2 S protein, including the N-terminal domain (NTD), receptor binding domain (RBD), and transmembrane (TM) domain. The putative furin cleavage site at the S1/S2 boundary is indicated by a red arrowhead. Bottom panel: sequence of amino acids (aa) 676–688 in the WT and mutant S-R682A proteins. The SARS-CoV-2-specific RRAR sequence is marked in red, and the mutation site is underlined. (B) Immunoblot analysis of lysates from WT S- and R682A mutant S-expressing VeroE6 cells. Immunoblots were probed with anti-S or anti-HA tag Abs. The full-length (FL) S protein and cleaved S fragments are marked as indicated; glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was included as the loading control. (C and D) Microscopic observation and (D) immunofluorescence staining with 4′,6-diamidino-2-phenylindole (DAPI) (blue) or an anti-S Ab (green) of VeroE6 cells transfected with vector (control), WT S, or mutant S-R682A plasmids 24 h after transfection. The syncytial phenotype is indicated by the multinucleated giant cells. Scale bars, 100 μm (C) and 200 μm (D).

Figure 2

The Furin/PC Inhibitors, but Not the TMPRSS2 Inhibitor, Suppress SARS-CoV-2 S Protein Cleavage and Development of the Syncytial Phenotype in S-Only-Expressing VeroE6 Cells (A) Immunoblot of lysates from VeroE6 cells transfected with the SARS-CoV-2 S (WT) expression construct and treated with different inhibitors as indicated (the furin/PC inhibitors CMK [50 μM], D6R [50 μM], SSM3 [25 μM], and naphthofluorescein [20 μM] and the TMPRSS2 inhibitor camostat [500 μM]) and harvested 24 h after treatment. The immunoblot was probed with anti-S and anti-HA Abs, and the FL and cleaved S proteins are marked as indicated. The furin substrate in VeroE6 cells, integrin αV, was included as an indicator of the inhibitory efficacy of the furin inhibitors, with pro-integrin αV and the cleaved integrin αV light-chain fragment as indicated; GAPDH was included as the loading control. (B and C) Microscopic observation and (C) immunofluorescence staining with DAPI (blue) or an anti-S Ab (green) of VeroE6 cells transfected with plasmids and subjected to inhibitor treatment as described in (A). Cells were harvested 24 h after treatment with the indicated specific inhibitors. The syncytial phenotype is indicated by the multinucleated giant cells. Scale bars, 100 μm (B) and 200 μm (C).

Figure 3

The Furin/PC Inhibitors and the TMPRSS2 Inhibitor Block CPE and Virus Production in SARS-CoV-2-Infected VeroE6 Cells (A) Microscopic observation of CPEs in VeroE6 cells infected with SARS-CoV-2 (MOI = 1) in the absence or presence of the specific furin/PC inhibitors (CMK, 50 μM; D6R, 50 μM; SSM3, 25 μM; naphthofluorescein, 20 μM) or the TMPRSS2 inhibitor (camostat, 500 μM), which were applied 1 h before infection and maintained in the medium until cell assessment 24 h after treatment. Scale bars, 100 μm. (B) Immunoblot of lysates from VeroE6 cells infected with SARS-CoV-2 (MOI = 1) that were treated with different inhibitors as described in (A). The immunoblot was probed with anti-S and anti-N Abs. FL S proteins, cleaved S proteins, and nucleocapsid (N) proteins are marked as indicated. GAPDH was included as the loading control. (C) A plaque assay was performed to determine the plaque-forming units (PFUs) of SARS-CoV-2 virus in the supernatant of VeroE6 cells infected and subjected to inhibitor treatment as described in (A). A representative photo of plaque assay is shown in the left panel. A bar graph and statistic results of the triplicate plaque assay results are shown in the right panel (mean ± SD, ^∗∗p < 0.01, ^∗∗∗p < 0.001). See also Figures S1 and S2.

Figure 4

CMK, Naphthofluorescein, and Camostat Treatment Dose-Dependently Decreased Virus Production and CPEs in Association with Reduced Viral RNA and Protein Levels in SARS-CoV-2-Infected VeroE6 Cells (A, E, and I) Microscopic observation of CPEs in VeroE6 cells infected with SARS-CoV-2 (MOI = 1) in the absence or presence of serial doses of (A) CMK, (E) naphthofluorescein, and (I) camostat, which were applied 1 h before infection and maintained in the medium until cell assessment 24 h after treatment. Scale bars, 100 μm. (B, F, and J) The virus titers (PFUs per milliliter) in the supernatant of SARS-CoV-2-infected VeroE6 cells treated with serial doses of (B) CMK, (F) naphthofluorescein, and (J) camostat were determined by plaque assay (mean ± SD of triplicates, ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001). (C, G, and K) The selectivity index (SI) of (C) CMK, (G) naphthofluorescein, and (K) camostat was determined by CC₅₀/IC₅₀. The left y axis indicates the inhibition of virus titer (percent) relative to that of the untreated control group. The right y axis indicates the cell viability (percent) relative to that of the untreated control group. The x axis indicates the concentration of inhibitors. The CC₅₀, IC₅₀, and SI values for each inhibitor are shown above the figures. (D, H, and L) Viral protein and viral RNA in SARS-CoV-2-infected VeroE6 cells treated with serial doses of (D) CMK, (H) naphthofluorescein, or (L) camostat were detected by immunoblot (top panels) and northern blot (bottom panels) analyses. Viral RNA isolated from VeroE6 cells was quantified by qRT-PCR targeting the E gene, as indicated below the northern blot results.

Figure 5

CMK and Camostat Treatment Affects the Early Stage of the Virus Replication Cycle, but Naphthofluorescein Decreases Viral RNA Levels after Virus Entry (A and E) Schematic illustration of (A) the preinfection treatment experiments and (E) the postinfection treatment experiments. (B and F) Microscopic observation of CPEs in VeroE6 cells assessed at the end of (B) preinfection treatment and (F) postinfection treatment with specific inhibitors (CMK [50 μM], D6R [50 μM], SSM3 [25 μM], naphthofluorescein [20 μM], or camostat [500 μM]). Scale bars, 100 μm. (C and G) The virus titers (PFUs per milliliter) in the supernatant of SARS-CoV-2-infected VeroE6 cells at the end of the (C) preinfection treatment experiments and (G) postinfection treatment experiments were determined by plaque assay (mean ± SD of triplicates, ^∗∗p < 0.01, ^∗∗∗p < 0.001). (D and H) The viral protein and viral RNA in VeroE6 cells at the end of the (D) preinfection treatment experiments and (H) postinfection treatment experiments were analyzed by immunoblotting (top panels) and northern blotting (bottom panels). Viral RNA isolated from VeroE6 cells was quantified by qRT-PCR targeting the E gene, as indicated below the northern blot results. (I) Immunoblot of SARS-CoV-2 viral proteins in supernatant harvested from VeroE6 cells at the end of the postinfection treatment experiments. The immunoblot was probed with anti-S and anti-N Abs. (J) Representative results of a plaque assay to determine the PFUs of SARS-CoV-2 virus in the supernatant of VeroE6 cells at the end of the postinfection treatment experiments. #, the lower number of PFUs in the 10⁻³ dilution than in the 10⁻⁴ dilution of the supernatant could be due to the inhibitory effect of camostat on virus entry in the higher-dose 10⁻³ dilution. See also Figures S3–S5.