Efficient activation of the severe acute respiratory syndrome coronavirus spike protein by the transmembrane protease TMPRSS2

Abstract

The distribution of the severe acute respiratory syndrome coronavirus (SARS-CoV) receptor, an angiotensin-converting enzyme 2 (ACE2), does not strictly correlate with SARS-CoV cell tropism in lungs; therefore, other cellular factors have been predicted to be required for activation of virus infection. In the present study, we identified transmembrane protease serine 2 (TMPRSS2), whose expression does correlate with SARS-CoV infection in the upper lobe of the lung. In Vero cells expressing TMPRSS2, large syncytia were induced by SARS-CoV infection. Further, the lysosome-tropic reagents failed to inhibit, whereas the heptad repeat peptide efficiently inhibited viral entry into cells, suggesting that TMPRSS2 affects the S protein at the cell surface and induces virus-plasma membrane fusion. On the other hand, production of virus in TMPRSS2-expressing cells did not result in S-protein cleavage or increased infectivity of the resulting virus. Thus, TMPRSS2 affects the entry of virus but not other phases of virus replication. We hypothesized that the spatial orientation of TMPRSS2 vis-a-vis S protein is a key mechanism underling this phenomenon. To test this, the TMPRSS2 and S proteins were expressed in cells labeled with fluorescent probes of different colors, and the cell-cell fusion between these cells was tested. Results indicate that TMPRSS2 needs to be expressed in the opposing (target) cell membrane to activate S protein rather than in the producer cell, as found for influenza A virus and metapneumoviruses. This is the first report of TMPRSS2 being required in the target cell for activation of a viral fusion protein but not for the S protein synthesized in and transported to the surface of cells. Our findings suggest that the TMPRSS2 expressed in lung tissues may be a determinant of viral tropism and pathogenicity at the initial site of SARS-CoV infection.

FIG. 1.

Distribution of TMPRSS2 and ACE2 and histopathology of SARS-CoV-infected cynomolgus lung. The distribution of TMPRSS2 (A, arrows indicating round shapes in middle panel) and ACE2 (C, arrows indicating thin shapes in middle panel) in healthy cynomolgus lung sections stained with antibodies against SARS-CoV, TMPRSS2, or DAPI (upper panels) was detected by an immunofluorescence staining method (see Materials and Methods section). Samples stained with secondary antibodies were used as controls (lower panels). The distribution of TMPRSS2 (B) and ACE2 (D) was also examined in mild (upper row) and severe (lower row) lesions of SARS-CoV-infected lungs stained with antibodies against SARS-CoV, TMPRSS2, or DAPI (nuclear staining).

FIG. 2.

Cytopathic changes on SARS-CoV S-protein-expressing cells. (A) Vero cells or Vero-TMPRSS2 cells were infected with SARS-CoV at an MOI of 0.1 and incubated at 37°C for 36 h. Cells were stained with crystal violet. (B) Vero cells or Vero-TMPRSS2 cells were infected with Dis/SARS-S at an MOI of 0.1 and incubated at 37°C for 20 h, after which the cells were treated with 10 μg/ml trypsin at 37°C for 30 min and then incubated for another 3 h. (C) Vero-TMPRSS2 cells were infected with Dis/SARS-S at an MOI of 0.1 and incubated at 37°C for 20 h in the absence (a) or presence of 500 μM (b) or 5 mM (c) leupeptin. Cells not infected with Dis/SARS-S were used as a control (d). (D) The size of syncytia in the absence and presence of 5 μM, 50 μM, 500 μM, and 5 mM leupeptin was quantified by counting the number of nuclei in the fused cells. The error bars are standard deviations.

FIG. 3.

Effect of TMPRSS2 on virus entry into cells. Vero or Vero-TMPRSS2 cells were infected with SARS-CoV at an MOI of 0.1 in the presence of 5 μM HR2 peptide, 50 μM EST, or 500 μM bafilomycin A1 and then cultured for 5 h. The amount of viral mRNA9 was measured by real-time PCR. Cells not treated with reagents were used as controls. The error bars are standard deviations of at least six independent measurements.

FIG. 4.

TMPRSS2 does not affect virus production. (A) SDS-PAGE and Western blot analysis were performed to detect S protein in cell lysates (cell) and culture medium (med) of Vero or Vero-TMPRSS2 (Vero-TM2) cells at 20 h after SARS-CoV infection (left panel). Cells treated with 10 μg/ml trypsin (tryp) at 37°C for 5 min were used as a cleaved-S control. Influenza virus (Flu)-HA produced by plasmid transfection was also used as a cleavage control (right panel). (B) Infectivity of SARS-CoV produced in Vero or Vero-TMPRSS2 cells. Vero-E6 cells were infected with SARS-CoV produced in either Vero or Vero-TMPRSS2 cells in the presence of 5 μM HR2 peptide or 500 nM bafilomycin A1 and then cultured for 5 h. The amount of viral mRNA9 was measured by real-time PCR. Cells not treated with reagents were used as controls. The error bars are standard deviations of at least six independent measurements.

FIG. 5.

TMPRSS2 dependence on spatial orientation for the activation of SARS-CoV S protein. (A) Schematic diagrams of S-expressing effector cells (green) and acceptor cells (orange) shown in panel B. (B) To detect cell-cell fusion of S-expressing cells, Dis/SARS-S-infected or pKS/SARS-S-transfected Vero or Vero-TMPRSS2 cells (effector cells) were collected by nonenzymatic cell dissociation solution and then overlaid onto the orange target Vero or Vero-TMPRSS2 cells, respectively. After 20 h of incubation, cells were fixed with 4% formaldehyde and observed by fluorescence microscopy. White arrows indicate fused cells. (C) The sizes of syncytia indicated in the upper row of panel B were quantified by counting the number of nuclei in the fused cells. The error bars are standard deviations.