Transmissible Gastroenteritis Coronavirus Induces Programmed Cell Death in Infected Cells through a Caspase-Dependent Pathway

The American high cell passage Purdue-115 strain of TGEV and French isolate RM4 of porcine respiratory coronavirus (PRCV) were used as virus sources and propagated on ST and swine kidney (strains PDH and RPTG [rein de porc Thiverval-Grignon]) cells as described previously (19). For TGEV UV inactivation, 25 ml of a virus inoculum was placed into a petri dish (10-cm diameter) on ice and exposed to a 254-nm wavelength lamp (1.368 mW/cm²) for 6 min. The thiol reducing agent PDTC was added to the medium immediately after infection to a final concentration of 100 μM. Because PDTC is expected to rapidly lose its activity during incubation (32), the concentration was readjusted every 8 h. For caspase inhibition experiments, the following drugs (from Bachem, Bubendorf, Switzerland) were employed: the interleukin 1β-converting enzyme inhibitor tetrapeptide acetyl-Tyr-Val-Ala-Asp-chloromethylketone (Ac-YVAD-cmk), the apopain inhibitor acetyl-Asp-Glu-Val-l-aspartic acid aldehyde (Ac-DEVD-CHO), and the ICE-like protease inhibitor tripeptide Z-VAD.fmk. Ac-DEVD-CHO was dissolved in H₂O; the two other caspase inhibitors were dissolved in dimethyl sulfoxide with a final concentration in cultures of <0.1% (vol/vol). Two hours before infection, these synthetic modified peptides were added to the medium at a concentration of 100 μM and maintained at that concentration after infection. Control cells received 0.1% dimethyl sulfoxide, which had no effect on cell proliferation or viability.

Low-molecular-weight DNA was extracted from approximately 10⁶ cells as described by Hinshaw et al. (14), with slight modifications. Briefly, infected or uninfected cells detaching from the plate were centrifuged at 200 × g and pooled with the trypsinized adhering cells for each of the time point tested. Cells were washed in phosphate-buffered saline (PBS), resuspended in 500 μl of ice-cold lysis buffer (10 mM Tris [pH 7.5], 10 mM EDTA, 0.2% Triton X-100), and incubated on ice for 30 min. Lysates were centrifuged at 10,000 × g at 4°C for 10 min, and supernatants were extracted once with buffered phenol, once with buffered phenol-chloroform, and once with chloroform-isoamyl alcohol (24:1, vol/vol). DNA was ethanol precipitated with 500 mM NaCl. DNA samples were resuspended in 15 μl of sterile water and treated for 15 min at 37°C with RNase A at a final concentration of 0.5 μg/μl, and half of the sample was run on a 2% agarose gel in 1× Tris-borate-EDTA buffer and stained with ethidium bromide.

For each plate, detached cells and trypsinized adherent cells were pooled, centrifuged, fixed in 70% ethanol for 1 h, washed in PBS, incubated for 15 min at 37°C with 100 μM RNase A, and stained with propidium iodide as described by Nicoletti et al. (30).

Cells were trypsinized, washed twice in PBS, and incubated with 40 nM 3,3′-dihexyloxacarbocyanine iodide [DIOC₆(3); Molecular Probes] in PBS for 15 min at 37°C. Cells treated with the mitochondrial transmembrane potential (ΔΨ_m)-disrupting protonophore carbonyl cyanidem-chlorophenylhydrazone (100 μM, 15 min) served as a negative control. Flow cytometry was done with a Becton Dickinson cytofluorometer. Data analysis was performed with Lysis II software (Becton Dickinson).

Apoptotic nuclei were visualized by using cells fixed and stained as described above. Observations were performed on a UV light microscope. For electron microscopy, ST cells were pelleted by centrifugation, washed twice with PBS, fixed with 1.25% glutaraldehyde buffered with sodium cacodylate (0.1 M, pH 7.3), and then postfixed in 2% osmium tetroxide. After dehydration, samples were embedded in Epon araldite. Thin (70-nm) sections were stained with lead citrate and uranyl acetate and then examined with a Philips EM12 electron microscope operated at 80 kV.

For electrophoretic mobility shift assays (EMSAs), ST cells mock infected or infected with TGEV at a multiplicity of infection (MOI) of 5 were lysed in Totex buffer (31). A ³²P-labeled oligonucleotide (35 fmol) with a κB consensus sequence (5′-AGTTGAGGGGACTTTCCCAGGC-3′) was used as a probe and incubated for 20 min at room temperature with a total extract containing 4 μg of protein. Competition was performed by using an excess of unlabeled NF-κB or an unrelated SP1 probe (5′-ATTCGATCGGGGCGGGGCGAGC-3′). DNA-protein complexes were run on a 4% polyacrylamide gel. Supershifts were performed with antibodies to NF-κB subunits p65 (murine) and p50 (human), kindly provided by Alain Israel, Institut Pasteur, Paris, France. The antibody (1 μl) was added to the binding mixture 20 min after the radiolabeled NF-κB probe. The reaction mixture was incubated for 20 min, and the complexes were resolved as described above.

Phosphorothioated oligonucleotides were kindly provided by P. Marianneau and P. Després, Institut Pasteur, Paris, France. These modified NF-κB oligonucleotides, used as transcription factor decoys (TFDs), contained three copies of the κB consensus sequence (26). TFDs were added to the cell medium to a concentration of 3 μM 24 h before infection and maintained at that concentration in the medium until the cells were lysed.

To investigate whether apoptotic cell death was triggered by TGEV, virus-infected cells were first examined for the presence of a DNA ladder. For each cell line, an extensive cytopathic effect (CPE) was observed at around 16 h postinfection (p.i.) when an MOI of 5 was used, i.e., the cells rounded and detached from the plate. Internucleosomal DNA cleavage was analyzed at 18 h p.i. in three porcine cell lines, ST, PDH, and RPTG, and in MDCK-APN, a clone of a canine kidney cell line expressing porcine APN, which is a receptor for TGEV (7). A DNA ladder was observed for the four cell lines after infection (Fig. 1A). This was not the case for MDCK cells not expressing porcine APN, which are fully refractory to TGEV infection. Neither a CPE nor DNA ladder formation was observed when a UV-inactivated inoculum was used as a control. PRCV, a respiratory variant of TGEV (reviewed in reference20), was also capable of inducing DNA fragmentation (Fig. 1B), with delayed kinetics, consistent with a replication rate slower than that of TGEV.

TGEV-infected cells were examined by optical microscopy after fixation with ethanol and staining with propidium iodide, a fluorescent DNA intercalator (30). Most of the infected ST cells showed a marked nuclear diameter reduction and obvious chromatin condensation 18 h after infection (Fig.2B). TGEV-infected and mock-infected cells fixed at 12 and 24 h p.i. were used for analysis of ultrastructural alterations. Morphological changes typical of the late stage of apoptosis were only observed with cells fixed at 24 h p.i. At that stage, only nucleolar remnants could be found in the convoluted nuclei, which contained small masses of condensed chromatin (Fig. 2D). Mitochondria appeared swollen, indicating that they were no longer functional.

Further experiments were carried out to establish the time course of nuclear DNA degradation relative to that of cell death. As shown in Fig.3A, internucleosomal DNA cleavage was detected by 12 h p.i. Nuclear DNA degradation was also monitored by flow cytometry analysis of cells fixed and stained with propidium iodide. With this technique, a lower fluorescence signal is emitted by apoptotic cells compared to normal cells, corresponding to a lower nuclear DNA content (see references 30 and35). Results from a typical experiment are shown in Fig. 3B. Apoptotic nuclei appeared as a broad, subdiploid DNA (A₀) peak easily distinguished from the narrow peak of cells with normal (diploid) DNA content in the red fluorescence. The proportions of subdiploid DNA-containing nuclei in samples were 20, 60, and 70% at 12, 18, and 24 h p.i., respectively. This result shows that the majority (>70%) of cells entered apoptosis following TGEV infection. A usual feature of apoptotic cells is that loss of membrane integrity occurs in the final stages of the process. When ST cells were tested for the ability to exclude the vital dye trypan blue, the percentages of blue cells were only 6, 25, and 40% ± 5% at 18, 24, and 48 h p.i., respectively. Taken together, these results supported the view that PCD is an important component of TGEV-induced cell death.

Accumulating data indicate a breakdown of mitochondrial function during apoptosis (reviewed in references 34 and39). Cytofluorometric analysis of ΔΨ_m was performed for TGEV-infected ST cells with the potential-sensitive dye DIOC₆(3). As shown in Fig.4, a ΔΨ_m transition, indicating a loss of mitochondrial function, occurred between 10 and 12 h p.i. and progressed until 24 h p.i.

The central effector machinery of PCD is composed of cytoplasmic proteases called caspases. These are present in an inactive form and are irreversibly activated during the effector phase of apoptosis, leading to ultrastructural alterations, chromatin condensation, and nuclear fragmentation. In humans, 10 caspases have been identified and some of them can be blocked by specific synthetic peptides (reviewed in references 13 and29). To determine whether TGEV-induced apoptosis involves caspase activation, experiments were performed with the irreversible tetrapeptide ICE (caspase 1) subfamily protease inhibitor Ac-YVAD.cmk, the reversible tetrapeptide apopain (caspase 3) inhibitor Ac-DEVD-CHO, and the irreversible tripeptide pan-ICE inhibitor Z-VAD.fmk (40, 48). These inhibitors were used at a concentration of 100 μM, which has been shown to allow complete protease inhibition in cultured mammalian cells after induction of apoptosis by various molecules (8, 9, 16, 28). Incubation of cells with Z-VAD.fmk prior to infection with TGEV completely blocked the appearance of apoptotic nuclei and the onset of nuclear DNA fragmentation until at least 48 h p.i. (Fig.5A). Importantly, virus production in Z-VAD.fmk-treated cells was similar to that in untreated cells (Fig.5B). Apparently, the Z-VAD.fmk treatment had no clear effect on the maintenance of cell membrane integrity, as assessed by trypan blue staining. However, there was a dramatic change in the appearance of infected cells (Fig. 6). No protection against TGEV-induced apoptosis was observed when Ac-YVAD.cmk or Ac-DEVD-CHO was used, possibly due to a low permeability of the cells to these drugs (data not shown). Similar results have been obtained with leukemic cell death induced by interleukin-3 withdrawal, against which DEVD.fmk and Ac-YVAD.cmk did not protect, in contrast to Z-VAD.fmk (3).

Direct exposure of various cell types to oxidants such as hydrogen peroxide or lipid hydroperoxides can directly induce apoptosis, while in many experimental models, pretreatment of cells with antioxidants has been shown to protect against this form of cell death. For example, the thiol reducing agent PDTC has been shown to prevent alphavirus-induced apoptosis (24). To investigate whether such agents could prevent apoptosis in TGEV-infected cells, various concentrations of PDTC were added to the medium at the beginning of infection. A nearly complete protective effect of PDTC against TGEV-induced apoptosis at concentrations as low as 50 μM was observed at 18 h p.i., as shown by a DNA ladder assay (data not shown). Flow cytometry analysis of DNA degradation showed that TGEV-induced apoptosis was markedly inhibited by PDTC (Fig. 5B), with no change in cumulative virus production compared to that in untreated cells (Fig. 5B). PDTC also had a protective effect against the CPE (Fig. 6) and cellular membrane damage, with only 15% ± 5% of trypan blue-positive cells at 48 h p.i. These results suggest that an abnormal cellular oxidation event may occur during TGEV infection, which might account for cell injury and PCD induction.

PDTC is a very effective inhibitor of the activation of NF-κB (36), a family of dimeric transcription factors (reviewed in reference2). NF-κB activation can induce or inhibit apoptosis, and this probably depends on its subunit composition and cell type and the nature of the apoptotic stimulus. To investigate whether TGEV infection induces NF-κB activation, total cell lysates were prepared at various times p.i. and DNA-protein complexes were analyzed by EMSAs. Strong and quick NF-κB DNA-binding activity was induced during infection by TGEV, whereas only background activity was detected in noninfected cells (Fig. 7A and B). NF-κB DNA-binding activity began to increase as early as 3 h p.i. and reached a maximum at 9 h p.i. To identify the NF-κB subunits detected by EMSA in infected ST cells, subunit-specific antibodies to human p50 and murine p65 were added during in vitro DNA binding. A strong band corresponding to the complex supershifted was observed with both anti-p50 and anti-p65 antibodies (Fig. 7C). These results suggest that the activated NF-κB complex detected may involve p65 and p50 subunits, even if the presence of other NF-κB subunits cannot be excluded.

To determine whether the activation of NF-κB could be responsible for the induction of PCD in TGEV-infected ST cells, NF-κB decoy experiments were performed with a probe containing human NF-κB binding sites. Treatment of cells with TFD completely inhibited the NF-κB binding activity induced by TGEV (Fig. 7D). Fluorocytometry analysis showed that nuclear DNA degradation kinetics were slightly slower in TFD-treated cells than in untreated cells (Fig. 5A). No effect on cell viability was observed after TFD treatment of uninfected control cells until after 11 h p.i. However, virus production was also slightly retarded (virus titer fourfold lower at 12 h p.i.; Fig. 5B). From these results, it was concluded that the NF-κB activation observed in TGEV-infected ST cells plays a secondary role, if any, in the induction of PCD.