Cyclosporin A inhibits the replication of diverse coronaviruses

Abstract

Low micromolar, non-cytotoxic concentrations of cyclosporin A (CsA) strongly affected the replication of severe acute respiratory syndrome coronavirus (SARS-CoV), human coronavirus 229E and mouse hepatitis virus in cell culture, as was evident from the strong inhibition of GFP reporter gene expression and a reduction of up to 4 logs in progeny titres. Upon high-multiplicity infection, CsA treatment rendered SARS-CoV RNA and protein synthesis almost undetectable, suggesting an early block in replication. siRNA-mediated knockdown of the expression of the prominent CsA targets cyclophilin A and B did not affect SARS-CoV replication, suggesting either that these specific cyclophilin family members are dispensable or that the reduced expression levels suffice to support replication.

Fig. 1.

CsA inhibits the replication of GFP-expressing recombinant coronaviruses. Vero E6 cells (a) or 293/ACE2 cells (b) were infected with SARS-CoV–GFP at an m.o.i. of 10 and at 1 h p.i. the inoculum was replaced by medium containing different CsA concentrations. Cells were fixed at 18 h p.i. and GFP reporter expression was measured and normalized to the signal in control cells (100 %), which were treated with DMSO, the solvent used for CsA (upper panels, grey bars). Huh7 cells infected with HCoV-229E–GFP were treated with CsA from 1 h p.i. onwards and were fixed for GFP measurements at 24 h p.i. (c, upper panel). 17CL1 cells were infected with MHV–GFP, treated with CsA from 1 to 18 h p.i., and GFP fluorescence was quantified (d, upper panel). The effect of CsA treatment on the viability of the various cell lines used, compared with untreated control cells (a–d, lower panels) was determined by using a CellTiter 96 AQ_ueous MTS assay (Promega). Graphs show the results (mean and sd) of a representative quadruplicate experiment. All experiments were repeated at least twice.

Fig. 2.

CsA treatment inhibits coronavirus protein and RNA synthesis, and the production of infectious progeny. Vero E6 cells were infected with SARS-CoV–GFP (a) or wt SARS-CoV (b) and treated with CsA from 1 to 10 h p.i. Viral protein expression was analysed by Western blotting using polyclonal rabbit antisera against nsp8 (van Hemert et al., 2008), the N protein (Knoops et al., 2010) and GFP, as indicated next to the panels. β-Actin, detected with a rabbit antiserum (Sigma), was used as loading control. (c) Immunofluorescence analysis of Vero E6 cells infected with SARS-CoV (m.o.i. 10) and treated from 1 to 10 h p.i. with the CsA concentration indicated below each panel. Cells were stained with an anti-SARS-CoV nsp4 rabbit antiserum (upper panel; van Hemert et al., 2008) or an anti-dsRNA mAb (lower panel; Knoops et al., (2008)). Bar, 50 µm. (d) Vero E6 cells infected with SARS-CoV–GFP (grey bars) or wt SARS-CoV (white bars) were treated with various concentrations of CsA from 1 h p.i. onwards, and virus titres in the culture supernatant were determined at 16 h p.i. by plaque assay on Vero E6 cells. Huh7 cells infected with HCoV-229E–GFP (e) or 17CL1 cells infected with MHV-A59 (f) were treated with CsA from 1 h p.i. onwards, and infectious-progeny titres were determined at 30 h p.i. and 8 h p.i., respectively. The graphs show the mean of two independent duplicate experiments.

Fig. 3.

SARS-CoV–GFP replication in Cyclophilin A- or B-depleted cells. By using DharmaFECT1 (Dharmacon), 293/ACE2 cells were transfected with siRNAs (ON-Target PLUS pools; Dharmacon) targeting CypA and CypB mRNAs. Non-targeting siRNA and siRNA targeting GAPDH expression were used as negative and positive controls for transfection and depletion efficiency, respectively. (a) Expression levels of CypA (upper panel) and CypB (middle panel), in cells transfected with the siRNA pools indicated below the lanes, were analysed by Western blotting by using specific antisera (Abcam). β-Actin, detected with a rabbit antiserum (Sigma), was used as a loading control. (b) The viability of cells transfected with the various siRNAs was monitored by using a CellTiter 96 AQ_ueous MTS assay (Promega). Data were normalized to the mean MTS assay value of cells transfected with non-targeting control siRNAs (100 %). (c) Forty-eight hours after siRNA transfection, cells were infected with SARS-CoV–GFP and 24 h later cells were fixed and GFP fluorescence was quantified. The level of GFP expression was normalized to that in infected cells transfected with non-targeting siRNA.