Coordinate induction of IFN-alpha and -gamma by SARS-CoV also in the absence of virus replication

Abstract

Background: Severe acute respiratory syndrome (SARS) is an emerging infection caused by a novel coronavirus known as SARS-CoV, characterized by an over-exuberant immune response with lung lymphomononuclear cells infiltration and proliferation that may account for tissue damage more than the direct effect of viral replication. This study is aimed at investigating the capability of SARS-CoV to activate IFN-alpha and -gamma expression in lymphomonocytes (PBMC) from healthy donors, evaluating whether viral replication is necessary for this activation.

Results: SARS-CoV virus is able to induce both IFN-alpha and -gamma mRNA accumulation and protein release in a dose-dependent manner, MOI 10 being the most effective. The time course curve indicated that IFN-alpha mRNA induction peaked at 24 h.p.i,. whereas IFN-gamma mRNA was still increasing at 48 h.p.i. Released IFN (both types) reached a plateau after 24-48 h.p.i. and remained rather stable over a 5-day period. A transient peak of negative strand viral RNA was detected after 1-2 days of infection, but neither infectious virus progeny yield nor newly produced viral genomic RNA could be evidenced in infected cultures, even after prolonged observation time (up to 13 days). Cocultivation of PBMC with fixed SARS-CoV-infected Vero cells was even more efficient than exposure to live virus in eliciting IFN-alpha and -gamma induction. A combination of IFN-alpha and -gamma strongly inhibited SARS-CoV replication in Vero cells, while the single cytokines were much less effective.

Conclusions: This study provides evidence that SARS-CoV is able to induce in normal PBMC a coordinate induction of IFN-alpha and -gamma gene expression. Virus replication is not necessary for IFN induction since efficient IFN expression could be obtained also by the cocultivation of normal PBMC with fixed SARS-CoV-infected cells. Concomitant activation of IFN-alpha and -gamma gene expression by SARS-CoV in vivo may be relevant for the pathogenesis of the disease, both for the possible involvement in immunomediated damage of the tissues and for the strong inhibition of SARS-CoV replication as a result of combined cytokine action.

Fig. 1

Induction of mRNA for IFN-α and -γ and of immunoreactive cytokines by live SARS-CoV. (a and b) Representative dose-dependence experiment: PBMC were infected with SARS-CoV at different MOI (0.1, 1, and 10). After overnight incubation, mRNA levels specific for IFN-α (–▲–) and -γ (–■–) were measured by limiting dilution RT-PCR and expressed as ratio to β-actin mRNA (×10⁻³), as described in the Methods section (a). Released cytokines were detected by ELISA (b) and expressed as pg/ml. (c and d) Representative time course experiment: PBMC were exposed for the indicated times to SARS-CoV at MOI 10. Levels of mRNA for IFN-α (–▲–) and -γ (–■–) and of released cytokines were determined as in panels a and b, respectively. (e and g) Peak levels of mRNA for IFN-α (24 h.p.i.) and IFN-γ (48 h.p.i) in PBMC from 4 different donors infected with SARS-CoV at MOI 10. Results expressed as ratio to the levels in the unstimulated cultures. (f and h) Levels of released IFN-α (48 h.p.i.) and IFN-γ (72 h.p.i) by the same PBMC cultures shown in panels e and g. Results are expressed as pg/ml.

Fig. 2

Replication kinetic of SARS-CoV on PBMC. (a) PBMC were infected at MOI 0.1 (–●–), 1 (–■–), and 10 (–▲–) with SARS-CoV. Then, the cells were extensively washed, and fresh medium was added (time 0). Sampling of the cultures was performed at the indicated time points, up to day 13. One representative experiment is shown. Results are expressed as TCID₅₀/ml. (b and c) Infectivity (b) and viral RNA (positive and negative strand), (c) in PBMC infected at MOI 10, treated with trypsin after the adsorbtion phase. One representative experiment is shown. Infectivity is expressed as TCID₅₀/ml, viral genomic RNA as Log copies/10⁶ PBMC, and negative strand viral RNA as Log ratio to β-actin ×10⁻³.

Fig. 3

Induction of IFN by SARS-CoV-fixed-infected cells. PBMC were cocultivated with fixed SARS-CoV-Vero-infected cells at different ratios, 6:1 (–■–), 20:1 (–▲–), 60:1 (–●–), 200:1 (–♦–). At the indicated time points, supernatants were collected, and immunoreactive IFN-α (a) and -γ (b) were detected by ELISA. IFN-α and -γ levels in supernatants of cocultures of PBMC with uninfected fixed Vero cells were lower than the detection limit (5 pg/ml for IFN-α and >2 pg/ml for IFN-γ, respectively) (not shown).

Fig. 4

Reduction of infectious virus yield in Vero cells by IFN-α and -γ used singularly or in combination. Vero cells were treated with either IFN-α alone (5000 IU/ml), IFN-γ alone (1000 IU/ml), or with a combination of both and then infected with SARS-CoV at MOI 0.01. After 3 days, progeny virus was harvested and titrated. Results are expressed as mean virus yield (Log TCID₅₀) over 3 experiments. Bars indicate standard error over the mean. Statistical evaluation of the reduction versus control cultures: IFN-α P = 0.072; IFN-γ P = 0.451; IFN-α + IFN-γ P < 0.01.