Apical entry and release of severe acute respiratory syndrome-associated coronavirus in polarized Calu-3 lung epithelial cells

Abstract

Severe acute respiratory syndrome (SARS), caused by a novel coronavirus (CoV) known as SARS-CoV, is a contagious and life-threatening respiratory illness with pneumocytes as its main target. A full understanding of how SARS-CoV would interact with lung epithelial cells will be vital for advancing our knowledge of SARS pathogenesis. However, an in vitro model of SARS-CoV infection using relevant lung epithelial cells is not yet available, making it difficult to dissect the pathogenesis of SARS-CoV in the lungs. Here, we report that SARS-CoV can productively infect human bronchial epithelial Calu-3 cells, causing cytopathic effects, a process reflective of its natural course of infection in the lungs. Indirect immunofluorescence studies revealed a preferential expression of angiotensin-converting enzyme 2 (ACE-2), the functional receptor of SARS-CoV, on the apical surface. Importantly, both ACE-2 and viral antigen appeared to preferentially colocalize at the apical domain of infected cells. In highly polarized Calu-3 cells grown on the membrane inserts, we found that cells exposed to virus through the apical rather than the basolateral surface showed high levels of viral replication. Progeny virus was released into the apical chamber at titers up to 5 logs higher than those recovered from the basolateral chambers of polarized cultures. Taken together, these results indicate that SARS-CoV almost exclusively entered and was released from the apical domain of polarized Calu-3 cells, which might provide important insight into the mechanism of transmission and pathogenesis of SARS-CoV.

FIG. 1.

Kinetics of SARS-CoV replication in infected human lung epithelial Calu-3 cells. Calu-3 cells were infected with SARS-CoV at an MOI of 1 for 1 h at 37°C. After washing (twice) to remove unbound viral particles, infected Calu-3 cells were cultured in medium containing 20% (○) or 2% (□) FCS and observed daily for CPE. One milliliter of cell-free supernatant was harvested at the indicated time points after infection, accompanied by replenishment with 1 ml of fresh medium. Supernatants were kept frozen at −80°C until they were needed for assessing infectious viral titers by a standard TCID₅₀ assay with permissive Vero E6 cells, as described in Materials and Methods. Asterisks (*) indicate the time when CPE was first observed. The growth curve of SARS-CoV in Calu-3 cells was representative of two independently conducted experiments.

FIG. 2.

Assessment of SARS-CoV replication by quantitative RT-PCR. Calu-3 cells were harvested at 1, 18, 24, and 48 h after infection with γ-inactivated or live SARS-CoV at an MOI of 1. Total RNA was extracted and subjected to quantitative RT-PCR for amplification of virus-specific subgenomic mRNA 5 (M protein) species. The intensity of mRNA 5 expression was normalized to 18S RNA. The average of mRNA signals in duplicated samples is depicted.

FIG. 3.

Indirect immunofluorescence detection of SARS-CoV antigens in infected Calu-3 cells. Calu-3 monolayers grown on the chamber slides were mock infected or infected with SARS-CoV (MOI = 2). At 48 h after infection, nonpermeabilized Calu-3 cells were fixed with 2% paraformaldehyde and immunostained with a convalescent-phase serum from a SARS patient followed by a secondary FITC-conjugated goat anti-human IgG. Samples were examined by confocal microscopy. (A) Mock-infected monolayers showed only nuclear counterstain with DAPI. (B) SARS-CoV-specific antigens were readily detectable in clusters of infected Calu-3 monolayer.

FIG. 4.

SARS-CoV induces CPE on infected Calu-3 cells. Confluent Calu-3 cells were mock infected (A and C) or infected with SARS-CoV at an MOI of 1 (B and D) and cultured with DMEM medium containing 20% FCS. SARS-CoV-induced CPE was initially observed at day 8 after infection (not shown). The CPE at day 14 (B) and day 28 (D) after SARS-CoV infection were shown.

FIG. 5.

Expression of the SARS-CoV receptor (ACE-2) on the surface of Calu-3 cells. Nonpermeabilized Calu-3 cells grown on the chamber slides were fixed with 2% paraformaldehyde and stained with goat anti-human ACE-2 plus FITC-conjugated rabbit anti-goat IgG and counterstained with DAPI for nucleus (blue). Stained samples were analyzed by both conventional fluorescence microscopy and confocal microscopy. (A to C) ACE-2 is expressed on the surface of a fraction of Calu-3 cells, as revealed by conventional microscopy. (E and F) Confocal images of ACE-2 expression on the surface of Calu-3 cells. When the primary antibody was substituted with normal goat serum, no signal of ACE-2 expression was detected (D). Data shown are representative of five independent experiments.

FIG. 6.

Colocalization of ACE-2 and viral antigen in infected Calu-3 cells. Nonpermeabilized, mock-infected, and infected Calu-3 cells (MOI = 1) grown on the chamber slides were fixed with 2% paraformaldehyde at 48 h after infection. Uninfected and infected cells were stained with both goat anti-human ACE-2 antibody plus FITC-conjugated rabbit anti-goat IgG and rabbit anti-SARS-CoV NP antibody plus Texas red-conjugated donkey anti-rabbit IgG to detect the expression of ACE-2 (green) and SARS-CoV antigen (red). Samples were subjected to analysis with confocal microscopy. (A) Only ACE-2 signals could be detected in mock-infected cells; (B) both ACE-2 (green) and viral antigen (red) could be detected in infected cells. Importantly, both ACE-2 and viral antigen appeared to colocalize in infected cells (yellowish). Data shown are representative of four independent experiments.

FIG. 7.

Anti-ACE-2 antibody blocks the SARS-CoV replication in Calu-3 cells in a dose-dependent manner. Confluent Calu-3 cells grown in 96-well microtiter plates were treated with control goat antibody or goat anti-human ACE-2 antibody before infection with SARS-CoV at an MOI of 0.01. The resulting cell-free supernatants harvested at day 3 postinfection were subjected to the TCID₅₀ assays for quantifying the titers of infectious viruses. Data shown are representative of two independent experiments.

FIG. 8.

Preferential colocalization of SARS-CoV antigens and ACE-2 at the apical surface of infected Calu-3 cells. Nonpermeabilized infected Calu-3 cells grown on the chamber slides were fixed with 2% paraformaldehyde at 48 h after infection and stained with a SARS convalescent-phase serum visualized by FITC-conjugated goat anti-human IgG alone (green [A]) or with both goat anti-human ACE-2 antibody plus FITC-conjugated rabbit anti-goat IgG (green [B]) and rabbit anti-SARS-CoV NP antibody plus Texas red-conjugated donkey anti-rabbit IgG (red [B]). Samples were examined by confocal microscopy. (A) SARS-CoV antigens (green) were detected in clusters of infected Calu-3 cells. Cross-Z section of the confocal imaging revealed a preferential viral antigen expression at the apical surface. (B) Confocal image (cross-Z section) of two-color staining showing colocalization of ACE-2 (Green) and SARS-CoV-specific NP antigen (red) at the apical surface of infected Calu-3 cells. Data shown are representative of two independent experiments.

FIG. 9.

Polarity of SARS-CoV entry and budding in Calu-3 cells. Polarized Calu-3 cells grown on filter inserts were inoculated with SARS-CoV (MOI = 1) through either the apical or basolateral surface. At 48 h after infection, the culture supernatants collected from the apical and basolateral chambers were assessed for the contents of infectious viral particles by the TCID₅₀ assay. The results are derived from a representative of two independently conducted experiments.

FIG. 10.

Transmission electron microscopy of polarized Calu-3 cells infected with SARS-CoV. (A) Vesicles containing multiple viral particles (arrowheads) were frequently detected underneath the apical surface of polarized Calu-3 cells, but individual virions (arrow) were also detected at the apical junction of two polarized Calu-3 cells. (B) Release of SARS-CoV virions from the apical surface of polarized Calu-3 cells.