Regulation of IRF-3-dependent innate immunity by the papain-like protease domain of the severe acute respiratory syndrome coronavirus

Abstract

Severe acute respiratory syndrome coronavirus (SARS-CoV) is a novel coronavirus that causes a highly contagious respiratory disease, SARS, with significant mortality. Although factors contributing to the highly pathogenic nature of SARS-CoV remain poorly understood, it has been reported that SARS-CoV infection does not induce type I interferons (IFNs) in cell culture. However, it is uncertain whether SARS-CoV evades host detection or has evolved mechanisms to counteract innate host defenses. We show here that infection of SARS-CoV triggers a weak IFN response in cultured human lung/bronchial epithelial cells without inducing the phosphorylation of IFN-regulatory factor 3 (IRF-3), a latent cellular transcription factor that is pivotal for type I IFN synthesis. Furthermore, SARS-CoV infection blocked the induction of IFN antiviral activity and the up-regulation of protein expression of a subset of IFN-stimulated genes triggered by double-stranded RNA or an unrelated paramyxovirus. In searching for a SARS-CoV protein capable of counteracting innate immunity, we identified the papain-like protease (PLpro) domain as a potent IFN antagonist. The inhibition of the IFN response does not require the protease activity of PLpro. Rather, PLpro interacts with IRF-3 and inhibits the phosphorylation and nuclear translocation of IRF-3, thereby disrupting the activation of type I IFN responses through either Toll-like receptor 3 or retinoic acid-inducible gene I/melanoma differentiation-associated gene 5 pathways. Our data suggest that regulation of IRF-3-dependent innate antiviral defenses by PLpro may contribute to the establishment of SARS-CoV infection.

FIGURE 9

The IRF-3 pathway regulates SARS-CoV replication in cell culture.A, Huh7 DN-18 cells with tet-regulated, conditional expression of a dominant negative form of IRF-3 (DN IRF-3) were cultured in the presence or absence of tet for 3 days, to repress or induce DN IRF-3 expression, respectively, and were subsequently mock-infected or challenged with SeV for 12 h. Expression of IRF-3, ISG56, SeV, and actin proteins were detected in cell lysates by immunoblot analysis. The endogenous IRF-3 (FL) and the DN IRF-3 (DN) bands were marked as filled and empty circles, respectively. The asterisk in the ISG56 panel denotes a nonspecific band. B, Huh7 DN-18 cells repressed or induced for DN IRF-3 expression were infected with SARS-CoV at an m.o.i. of 0.01 at 37 °C for 1 h. The virus inoculum was then removed, and cells were washed extensively and refed with complete medium. Cell-free culture supernatants were collected at the indicated times postinfection and stored at -80 °C until they were subjected to assessment of infectious viral titers using a standard TCID₅₀ assay in Vero E6 cells, as described under “Experimental Procedures.” The experiments were terminated at day 3 postinfection as CPE was observed in ∼30 and ∼70% of infected cells without and with DN IRF-3 expression, respectively. The growth curve of SARS-CoV shown was representative of three independently conducted experiments.

FIGURE 1

SARS-CoV weakly activates an IFN response without inducing the hyperphosphorylation of IRF-3.A, human lung/bronchial epithelial Calu-3 cells were mock-infected or infected with 100 HAU/ml SeV (left panels) or SARS-CoV (right panels) at an m.o.i. of 5 for the indicated times prior to cell lysis and immunoblot analysis of IRF-3, ISG56, and SARS-CoV nucleocapsid (NC) protein. B, left panels show African green monkey kidney MA104 cells that were mock-infected or infected with SeV (100 HAU/ml) or with SARS-CoV (m.o.i. = 10) for the indicated times prior to cell lysis and immunoblot analysis of phospho-Ser³⁹⁶ IRF-3, total IRF-3, MxA, ISG56, SARS-CoV NC, and SeV proteins. A nonspecific band (*) detected by the anti-ISG56 antibody indicates equal loading. The arrowhead and hatch marks in the total IRF-3 panel denote phosphorylated (virus-activated) and inactive IRF-3 forms, respectively. In the right panel, MA104 cells grown in 35-mm dishes were mock-treated, infected with SARS-CoV (m.o.i. = 10) for 20 or 30 h, or transfected with 5 μg of poly(I-C) for 10 h prior to total RNA extraction and real time RT-PCR analysis of IFN-β mRNA abundance, which was subsequently normalized to cellular 18 S ribosomal RNA. Bars show fold change in IFN-β mRNA levels (relative to mock-treated cells).

FIGURE 2

SARS-CoV PLpro domain inhibits activation of IFN-β promoter following engagement of TLR3 or RIG-I pathways independent of its protease activity.A, HEK293-TLR3 cells were co-transfected with pIFN-β-Luc and pCMVβgal plasmids and a vector (Vect) encoding SARS PLpro-Sol (Sol) or SARS PLpro-TM (TM), bovine viral diarrhea virus Npro, or A20, or an empty vector. Twenty four hours later, cells were either mock-treated (empty bars) or incubated with 50 μg/ml poly(I-C) in culture medium for 6 h (hatched bars, left panel), or infected with SeV at 100 HAU/ml for 16 h (solid bars, right panel) prior to cell lysis for both luciferase and β-galactosidase assays. Bars show relative luciferase activity normalized to β-galactosidase activity, i.e. IFN-β promoter activity. B and C, IFN-β promoter activity in cells transfected with increasing amounts of WT or C1651A mutant PLpro-TM-expressing plasmid (B) or equivalent amounts of WT or D1826A mutant PLpro-TM plasmid (C). D and E, IFN-β promoter activity in cells transfected with increasing amounts of WT or C1651A mutant PLpro-Sol-expressing plasmid (D) or equivalent amounts of WT, C1651A, or D1826A mutant PLpro Sol plasmids (E).

FIGURE 3

SARS-CoV PLpro inhibits activation of IRF-3-dependent promoters by acting at a level downstream of the IRF-3 kinases and proximal to IRF-3.A, activation of p55C1B, an IRF-3-dependent synthetic promoter, by medium poly(I-C) (hatched bars) or SeV infection (solid bars) in HEK293-TLR3 cells with expression of SARS-CoV PLpro-TM (TM) or a control vector (Vect). B and C, activation of IFN-β (B) and p55C1B (C) promoters by ectopic expression of various signaling molecules within TLR3 and RIG-I/MDA5 pathways at or above the level of IRF-3 in HEK293 cells with (solid bars) or without (empty bars) expression of SARS-CoV PLpro-TM. N-RIG and N-MDA5 denote the caspase recruitment domain of RIG-I and MDA5, respectively.

FIGURE 4

Conditional expression of SARS-CoV PLpro-TM disrupts endogenous, IRF-3-dependent gene expression following dsRNA stimulation or SeV infection.A, immunofluorescence staining of PLpro-TM in HeLa PLpro-4 cells cultured in the presence (left) or absence (right) of 2 μg/ml tetracycline for 3 days followed by confocal microscopy. PLpro-TM showed green fluorescence staining (detected with a V5 tag antibody), whereas nuclei were counterstained blue with DAPI. B, left panels show HeLa PLpro-4 and PLpro-10 cells that were cultured with and without tetracycline for 3 days and subsequently mock-infected or challenged with SeV for 16 h prior to cell lysis and immunoblot blot analysis of PLpro (using a V5 tag antibody), ISG56, actin, and SeV. The right panels show HeLa PLpro-4 cells that were cultured in the indicated concentrations of tetracycline for 3 days prior to mock infection (lanes 9-13) or infection of SeV (lanes 14-17) for 16 h. Expression of PLpro-TM, ISG56, and actin were determined by immunoblot analysis. C, real time RT-PCR analysis of IFN-β (left), A20 (middle), and IL-6 (right) mRNA transcripts in HeLa PLpro-4 cells repressed or induced for PLpro-TM expression, and mock-treated (empty bars), treated with 50 μg/ml poly(I-C) in culture medium (hatched bars), or infected with SeV (solid bars) for 16 h. mRNA abundance was normalized to cellular 18 S ribosomal RNA. Fold changes were calculated by dividing the normalized mRNA abundance under various treatment conditions by that of the mock-treated cells without PLpro expression.

FIGURE 5

SARS-CoV PLpro inhibits virus-induced IRF-3 phosphorylation, dimerization, and nuclear translocation.A, HeLa PLpro-4 cells were manipulated to repress or induce PLpro-TM expression for 3 days and were subsequently mock-treated or incubated with 50 μg/ml poly(I-C) in culture medium for 6 h or challenged with SeV for 16 h. Cell lysates were separated on native PAGE followed by immunoblot analysis to detect IRF-3 monomer and dimer forms (top panel, empty arrowhead indicates IRF-3 dimer), or subjected to SDS-PAGE and immunoblot analysis of IRF-3, ISG56, PLpro, SeV, and actin (lower panels). B, HeLa PLpro-4 cells were cultured to repress or induce PLpro-TM expression and subsequently mock-infected or infected with SeV for 16 h. Where indicated, cells were incubated with 0.05 μg/ml of OA for the last 4 h of SeV infection/mock infection. Cells were then harvested for immunoblot analysis of IRF-3, p396-IRF3, PLpro (anti-V5), and SeV. Asterisk denotes a nonspecific band. C, lanes 1-4, HeLa cells were mock-infected or infected with SeV for 16 h. Where indicated, OA was included in the last 4-h duration of infection/mock infection. Lanes 5-8, cell lysates of HeLa cells infected with SeV were mock-treated, treated with CIP, OA, or CIP in the presence of OA, as indicated under “Experimental Procedures.” IRF-3 and p396-IRF3 were detected by immunoblot analysis. Asterisk denotes a nonspecific band. D, immunofluorescence staining of IRF-3 subcellular localization in HeLa PLpro-4 cells repressed (upper panels) or induced (lower panels) for PLpro-TM expression and mock-infected (left) or infected (right) with SenV for 16 h. Numbers in SeV-infected cells were the averages of the percentage of cells that had IRF-3 nuclear translocation and were counted from 10 random fields (×40 magnification). E, HeLa PLpro-4 cells were cultured to repress or induce PLpro-TM expression and subsequently mock-infected or infected with SeV. Cell lysates were subjected to immunoprecipitation (IP) with a rabbit anti-CBP antibody (upper panels) or with a rabbit anti-IRF-3 antiserum (lower panels). The immunoprecipitates were separated on SDS-PAGE, followed by immunoblot (IB) analysis of IRF-3 (using an mAb anti-IRF-3) or CBP (using a rabbit anti-CBP antibody).

FIGURE 6

SARS-CoV PLpro interacts with IRF-3.A, HeLa PLpro-4 cells were cultured to repress or induce PLpro-TM expression and subsequently mock-infected or infected with SeV for 16 h. Left panels, expression of p396-IRF-3, total IRF-3, PLpro, and actin proteins in cell lysates were determined by immunoblot (IB) analysis. Right panels, cell lysates were subjected to immunoprecipitation (IP) with a rabbit anti-IRF-3 antibody (lanes 1-4) or a control rabbit serum (lanes 5-8). The immunoprecipitates were separated on SDS-PAGE, followed by immunoblot analysis of IRF-3 (using an mAb anti-IRF-3) and PLpro (using an mAb anti-V5 tag). B, HEK293 cells were transfected with WT (lanes 1-3), C1651A (lanes 4-6), or D1826A (lanes 7-9) PLpro-TM, respectively, and, where indicated, with an empty vector (lanes 1, 4, and 7), or a vector expressing FLAG-IRF-3 (lanes 2, 5, and 8) or untagged IRF-3 (lanes 3, 6, and 9). Cell lysates were immunoprecipitated with anti-V5 (PLpro-TM), followed by immunodetection of FLAG-IRF-3 (anti-FLAG) or PLpro-TM (anti-V5) (right panels). Asterisk denotes IgG heavy chain. Immunoblot analysis of input for FLAG-IRF-3 and PLpro-TM is shown in left panels. All three forms of PLpro-TM interacted with IRF-3. C, HEK293 cells were transfected with WT (lanes 1 and 2), C1651A (lanes 3 and 4), or D1826A (lanes 5 and 6) PLpro-Sol, respectively, and, where indicated, with an empty vector (lanes 1, 3, and 5) or a vector expressing FLAG-IRF-3 (lanes 2, 4, and 6). Cell lysates were immunoprecipitated with anti-V5 (PLpro-Sol), followed by immunodetection of FLAG-IRF-3 (anti-FLAG) or PLpro-Sol (anti-V5) (lower panels). Immunoblot analysis of input for FLAG-IRF-3 and PLpro-Sol is shown in upper panels. Note that WT and D1826A PLpro-Sol interacted with IRF-3, whereas C1651A PLpro-Sol did not. D, HeLa PLpro-4 cells induced for PLpro-TM expression were transfected with GFP, GFP-IRF-3, or GFP-IRF3-5D, respectively. Cell lysates were subjected to co-IP experiments using either GFP antibody or V5 antibody for immunoprecipitation, followed by immunoblot analysis of the immunoprecipitates using anti-GFP or anti-V5 antibodies. Note that both GFP-IRF3 and GFP-IRF3-5D were associated with PLpro-TM (detected by anti-V5), whereas free GFP was not.

FIGURE 7

SARS-CoV nsp3 is associated with IRF-3 in cells infected with SARS-CoV. MA104 cells were mock-infected (lane 1), infected with SeV for 10 h (lane 2), infected with SARS-CoV (m.o.i. = 10) for 18 and 28 h (lanes 3 and 4), respectively, or infected with SARS-CoV for 18 h and then superinfected with SeV for an additional 10 h (lane 5). Expression of IRF-3, RIG-I, MxA, ISG56, SARS-CoV NC, and nsp3, and SeV proteins in cell lysates was determined by immunoblot (IB) analysis (upper panels). A nonspecific band detected by the rabbit anti-MxA antiserum (bottom panel) indicates equal loading. Data shown are representative of three independently conducted experiments. In lower panels, the cell lysates were subjected to immunoprecipitation (IP) with a rabbit anti-IRF3 antiserum, followed by immunoblot analysis of SARS-CoV nsp3, and IRF-3 (using an mAb anti-IRF3 antibody). Data shown are representative of two independently conducted experiments.

FIGURE 8

Regulation of IFN responses in cells infected with SARS-CoV.A, MA104 cells were mock-infected (lanes 1-3) or infected with SARS-CoV (m.o.i. = 10) for 18 h (lanes 4-6), followed by mock treatment (lanes 2 and 4), transfection of poly(I-C) (lanes 1 and 6), or superinfection of SeV (lanes 3 and 5) for 10 h, respectively (a total of 28 h after mock or SARS-CoV infection under these conditions). Immunoblot analysis of cell lysates for expression of ISG56, IRF-3, SeV, and actin is shown. Asterisks in ISG56 panels denote a nonspecific band. B, MA104 cells were mock-infected, infected with SeV for 10 h, infected with SARS-CoV (m.o.i. = 10) for 18 or 28 h, or transfected with poly(I-C) for 10 h, respectively, or pre-infected with SARS-CoV for 18 h and then superinfected with SeV or transfected with poly(I-C) for additional 10 h (a total of 28 h after SARS-CoV infection under these conditions). Cell-free culture supernatants were collected, irradiated, and subjected to IFN bioactivity assay against vesicular stomatitis virus as described under “Experimental Procedures.” Bars indicate IFN bioactivity (units/ml) in culture supernatants under the indicated conditions. C, MA104 cells were mock-infected, infected with SeV for 8 h, infected with SARS-CoV (m.o.i. = 10) for 16 or 24 h, or pre-infected with SARS-CoV for 16 h and then superinfected with SeV for additional 8 h. Cells were harvested for total RNA extraction, cDNA synthesis, and subsequent real time RT-PCR detection of IFN-β mRNA. Data shown are representative of two independently conducted experiments.