N pro of Classical Swine Fever Virus Prevents Type I Interferon-Mediated Priming of Conventional Dendritic Cells for Enhanced Interferon-α Response

Abstract

A hallmark of acute classical swine fever is the high interferon (IFN)-α levels found in the serum early after infection, followed by an inflammatory cytokine storm. Plasmacytoid dendritic cells (pDCs) represent the only known cell type that produces IFN-α upon classical swine fever virus (CSFV) infection in vitro. In primary target cells of the virus the viral protein N^pro inhibits the induction of type I IFN via the degradation of IRF3. We hypothesized that the early systemic pDC-derived IFN-α response sensitizes immune cells for enhanced responsiveness and augment cytokine responses after CSFV infection through the upregulation of IRF7. Therefore, bone marrow-derived granulocyte macrophage-colony stimulating factor (GM-CSF)-induced DCs, were pretreated with IFN-β or conditioned medium from CSFV-activated enriched pDC, and expression of the pro-inflammatory cytokines interleukin (IL)-1β, IL-6, and IFN-α was assessed after infection with wild-type CSFV and with an N^pro mutant [N^pro(D₁₃₆N)] unable to interact with IRF3 and IRF7. While type I IFN treatment sensitized the DCs for enhanced IFN and cytokine responses after stimulation with influenza virus, lipopolysaccharide or poly(I):poly(C), this was not observed for CSFV. In contrast, the N^pro(D₁₃₆N) mutant CSFV induced elevated IFN-α responses in type I IFN-pretreated GM-CSF DCs. These results indicate that CSFV has evolved to prevent type I IFN sensitization in infected cells through the action of the N^pro.

Introduction

C lassical swine fever (CSF) is a highly contagious hemorrhagic disease of pigs with high mortality rates. The etiologic agent is the classical swine fever virus (CSFV), which is a member of the genus Pestivirus within the Flaviviridae family. Depending on the virulence of the strain, varying degrees of disease severity can be observed, ranging from acute to subclinical or chronic forms (van Oirschot 1988; van Oirschot 2004). The clinical picture and the pathogenesis of acute CSF, the most severe form of the disease, resemble those of viral hemorrhagic fever (VHF). Acute CSF is associated with high fever, central nervous disorders, and hemorrhages seen in various organs (van Oirschot 2004). Immunopathological characteristics comprise coagulation disorders, generalized leukopenia, thrombocytopenia, and primary lymphoid organ atrophy leading to severe immunosuppression. This often fatal immunological imbalance is believed to be caused by virus-induced aberrant levels of type I interferon (IFN) and pro-inflammatory mediators causing a so-called cytokine storm, which is commonly observed in acute CSF (Knoetig and others 1999; Sanchez-Cordon and others 2005; Sanchez-Cordon and others 2002; Summerfield and others 2006; Jamin and others 2008) and in human VHFs (Geisbert and Jahrling 2004).

Nevertheless, the underlying mechanism that leads to the observed immunopathological events or the cellular sources of these cytokines are not clear. Particularly puzzling is the discrepancy between in vitro and in vivo observations. In vitro, primary target cells of CSFV such as macrophages, myeloid dendritic cells, and endothelial cells show only weak or no pro-inflammatory cytokine response and no type I IFN expression after CSFV infection (Knoetig and others 2002; Carrasco and others 2004; Ruggli and others 2005). Also, CSFV-infected monocyte- or bone marrow-derived DCs show no signs of maturation or activation in terms of MHC or CD80/86 up-regulation or cytokine expression in vitro (Carrasco and others 2004). The lack of innate immune system activation by CSFV in vitro is, at least in part, caused by the ability of N^pro to prevent IRF3-mediated activation, through induction of proteosomal degradation of the transcription factor (Ruggli and others 2003; Bauhofer and others 2005; Bauhofer and others 2007). In apparent contradiction, during acute CSF high IFN-α, tumor necrosis factor (TNF)-α, interleukin (IL)-1β, and IL-6 levels are found in the serum and cytokine-expressing cells are found in lymphoid tissue (Sanchez-Cordon and others 2002; Summerfield and others 2006; Jamin and others 2008). These responses appear to relate to the virulence of the CSFV strain as well as to the severity of the disease and were suggested to contribute more to the disease development than to protection with virulent strains of CSFV (Ruggli and others 2009).

In vitro, the only cell type that was shown to produce IFN-α and TNF-α in response to CSFV infection are plasmacytoid dendritic cells (pDCs) (Balmelli and others 2005), a subset of DCs specialized in the production of type I IFN (Liu 2005). It was proposed that pDCs can respond to CSFV due to their unique constitutive expression of IRF7 and independence on IRF3 (McCullough and others 2009). However, the fact that pDCs are a very rare cell type questioned whether they are solely responsible for the massive production of IFN-α and inflammatory cytokines during acute CSF. For influenza virus infections a comparable discrepancy between nonresponsiveness of target cells in vitro and effective innate immune responses induced in vivo are observed. Nevertheless, while the NS1 of influenza virus efficiently blocks activation of conventional DCs, this can be overcome by exposure of the DCs to low doses of IFN-β prior to Influenza A virus (IAV) infection (Osterlund and others 2005; Phipps-Yonas and others 2008). This priming of DCs with type I IFN resulted in influenza virus-induced IFN-α and TNF-α secretion as well as DC activation, and is possibly the basis for a wider distribution of cell types involved in innate immune responses during infection.

Based on this evidence and on the fact that the early systemic IFN-α could sensitize non-pDC cells for enhanced IFN-α/β and inflammatory cytokine production upon CSFV infection, we hypothesized that similar mechanisms for innate immune cell activation could play a role during CSFV infections in vivo. Considering that pDCs would provide a source of type I IFN, we addressed whether CSFV-induced cytokines and type I IFN in conventional dendritic cells or macrophages could be promoted by type I IFN priming through the induction of IRF7 expression.

Material and Methods

Generation of bone marrow hematopoietic cells-derived DCs using granulocyte macrophage-colony stimulating factor

Generation of granulocyte macrophage-colony stimulating factor (GM-CSF)-induced, bone marrow hematopoietic cells (BMHC)-derived porcine DCs was preformed as described previously (Carrasco and others 2001). Briefly, BMHCs collected from a sternum were seeded in T150 flasks at a density of 1×10⁶ cells/mL in 40 mL RBMI 1,640 medium (Invitrogen) supplemented with 10% fetal bovine serum (FBS; Biowest), 100 U/mL penicillin, 100 μg/mL streptomycin (Invitrogen), GlutaMAX^™-I (Invitrogen), and 100 U/mL recombinant porcine GM-CSF (Summerfield and others 2003). Cells were fed every 3 to 4 days with GM-CSF. After 14 days of culture, the nonadherent cells, the majority of them displaying DC morphology, were collected.

Isolation and culture of enriched pDCs

Peripheral blood mononuclear cells (PBMCs) were isolated from the blood of specific pathogen-free pigs using Ficoll-Paque (1.077g/L, Amersham Pharmacia Biotech AG) density-gradient centrifugation. pDCs were enriched by CD172a⁺ sorting using the Miltenyi magnetic-activated cell sorting separation system with LD columns (Miltenyi Biotec GmBH, Bergish Gladbach, Germany) and the anti-CD172a mAb clone 74-22-15A as described previously (Guzylack-Piriou and others 2004). CD172a⁺ PBMC, composed of monocytes, conventional blood DC, and pDCs (Summerfield and McCullough 2009), were plated at 4×10⁵ cells/well in a U-bottom 96-well plates in Dulbecco's modified Eagle's medium+GlutaMAX I (DMEM; Invitrogen) supplemented with FBS and 0.02 mM β-mercaptoethanol (Invitrogen).

Viruses

CSFV strain vA187-1 wild-type (WT) (Ruggli and others 1996) and the N^pro mutant CSFV vA187-D₁₃₆N [N^pro(D₁₃₆N)] carrying a D-to-N substitution at position 136 of the polyprotein (Ruggli and others 2009) were grown in the porcine aortic endothelial cell line PEDSV.15 (kindly obtained from Dr. Jörg Seebach, Division of Immunology and Allergology, Department of Internal Medicine, University Hospital Geneva, Switzerland). Virus titers were determined on PK-15 cells by endpoint dilution and expressed as 50% tissue culture infectious doses (TCID₅₀)/mL. The IAV NIBRG23 carrying the HA (without polybasic cleavage site) and NA from H5N1 A/turkey/Turkey/1/2005 and the remaining genes from H1N1 A/PR/8/34 (kindly obtained from Jim Robertson, National Institute for Biological Standards and Control, UK) was rescued by reverse genetics. The virus was grown in embryonated 10-day-old chicken eggs as described (Ocana-Macchi and others 2009). The IAV titers were determined by end-point titration on MDCK cells in the presence of 1 μg/mL L-1-tosylamid-2-phenylethyl chloromethyl ketone trypsin (Sigma-Alderich). About 72 h postinfection (p.i.) the cells were washed and stained with crystal violet. The TCID₅₀/mL was calculated according to the Reed-Muench formula (WHO Manual on Animal Influenza Diagnosis and Surveillance).

Infection and treatment of DCs

In some cultures, GM-CSF DCs were pretreated with different doses of recombinant porcine IFN-β (produced in house as described) (Husser and others 2011) in suspension at a density of 2×10⁶ cells/mL in RPMI medium without FBS. Thereafter, the IFN-β-containing medium was removed, fresh medium was added, and the cells were seeded in 48-well plates at a density of 5×10⁵ cells/well in 500 μL. For stimulation the cells were infected at the indicated multiplicity of infection (MOI) with WT or N^pro(D₁₃₆N) mutant CSFV or with IAV, or treated with 10 μg/mL polyinosinic polycytidylic acid (pIC) (Sigma Chemicals), 10 μg/mL LPS (Sigma), or 10μg/mL CpG DNA oligonucleotide D32 sequence (Guzylack-Piriou and others 2004) (Biosource). After 2 h of virus infection or stimulation, FBS was added to a final concentration of 10% (v/v) if not indicated otherwise.

Detection of intracellular viral proteins by flow cytometry

For labeling of intracellular viral proteins, cells were fixed and permeamilized with Fix & Perm permeabilization kit (ADG Bio Research GmbH) according to the manufacturer's instructions. The CSFV protein NS3 was stained with the monoclonal antibody (mAb) C16 (kindly provided by Dr. Irene Greiser-Wilke, Veterinary School Hannover, Germany), and the NP protein of NIBRG23 was detected with the mAb clone HB65 (ATCC). Binding of the primary antibodies were revealed using phycoerythrin-conjugated anti-mouse IgG F(ab’) 2 fragments (Dako).

mRNA isolation and real-time reverse transcription–polymerase chain reaction

Samples of 5×10⁵ DCs treated as indicated were pelleted, and total RNA was isolated using TRIZOL (Invitrogen) according to manufacturer's protocol and treated with RNase-free DNase I (Ambion). The real-time reverse transcription–polymerase chain reaction (real-time RT PCR) was performed with the SuperScript^™ III Platinum^®One-Step qRT-PCR System (Invitrogen) using the ABI PRISM 7,700 sequence Detector system (Applied Biosystems). The relative expression of IRF3 and IRF7 mRNA was calculated by the ΔC_t method and the amount of IRF-3 or IRF-7 mRNA relative to 18S mRNA was expressed as 2^−Ct (Livak and Schmittgen 2001). The following primers and probes were used: 18S forward primer 5′-CGC CGC TAG AGG TGA AAT TC-3′; 18S reverse primer 5′-GGC AAA TGC TTT CGC TCT G-3′; 18S probe 5′-TGG ACC GGC GCA AGA CGG A-3′; IRF-3 forward primer 5′-GGACCTGCACATTTCCAACAG-3′; IRF-3 reverse primer 5′-AGCGGCCTAGAAATCCATGTC-3′; IRF-3 probe 5′-CCTCACCTCCGACCAGTACAAGGCCT-3′; IRF-7 forward primer 5′-CTGCGATGGCTGGATGAA-3′; IRF-7 reverse primer 5′-TAAAGATGCGCGAGTCGGA-3′; IRF-7 probe 5′-CCGCGTGCCCTGGAAGCACTT-3′ (Alves and others 2009).

Detection of cytokines in the cell culture supernatant

Secreted IFN-α was quantified by enzyme-linked immunosorbent assay (ELISA) using the mAbs K9 and F17 (kindly provided by Dr. B. Charley, INRA, Jouy-en-Josas, France) as described previously (Guzylack-Piriou and others 2006). IL-6 and IL-1β in the cell culture supernatants were quantified using the porcine IL-6 DuoSet^® ELISA and the porcine IL-1β/IL-1F2 DuoSet^® ELISA, respectively (both from R&D Systems).

Statistical analysis

P values were calculated by unpaired t-test in SigmaPlot^® for normally distributed data. When normality test failed, a Mann–Whitney Rank Sum test was performed. Differences were considered significant at a P value of <0.05. Error bars represent the standard deviations of triplicates in one experiment and were calculated with Excel (Microsoft).

Results

Pretreatment of GM-CSF DCs with various doses of IFN-β does not lead to enhanced IFN-α in response to CSFV

We first explored the dose and time dependency of IRF7 and IRF3 induction after IFN-β treatment of GM-CSF DCs. Western blot was not sensitive enough to detect endogenous IRF7 protein in porcine GM-CSF DCs (data not shown). Therefore we employed real-time RT-PCR to assess IRF7 and IRF3 mRNA induction in GM-CSF DCs that were pretreated with 50, 100 and 1,000 U/mL IFN-β for 2, 6 and 16 h (Fig. 1A). Two hours after IFN-β treatment IRF7 mRNA was strongly up-regulated at all 3 applied IFN-β concentrations, and the expression increased only slightly with longer incubation or higher IFN-β doses. As expected, IRF3 mRNA levels remained unchanged.

FIG. 1.

Effect of pretreatment with IFN-β on IFR7 and IRF3 mRNA up-regulation, virus-induced IFN-α response, and viral replication. GM-CSF DCs were left untreated or stimulated with 50, 100, or 1,000 U/mL IFN-β for 2 h if not otherwise indicated. (A) mRNA levels of porcine IRF7 and IRF3 were measured by real-time reverse transcription–polymerase chain reaction after 2, 6, and 16 h of IFN-β treatment. After normalization to 18S mRNA, mean expression levels relative to untreated cells were calculated. Samples were done in triplicate. The error bars represent the minimum and maximum values. n.d.: not done (B) DCs were stimulated with 10 μg/mL of pIC, CpG, or LPS, or infected with IAV at an MOI of 0.5 TCID₅₀/cell or with CSFV at an MOI of 0.5, 1, or 3 TCID₅₀/cell. After 24 h IFN-α was measured by ELISA. (C) Viral protein expression in cells from (B) at 24 h postinfection. The percentage of CSFV- and IAV-infected cells was determined by flow cytometry using intracellular staining of NS3 and NP, respectively. (D) IFN-α response of IFN-β-treated GM-CSF DCs. DCs were pretreated for 2 h with the indicated doses of IFN-β before infection with CSFV at an MOI of 3 TCID₅₀/cell or IAV at an MOI of 0.5 TCID₅₀/cell. After 24 h IFN-α was measured by ELISA. (E) Expression of CSFV NS3 and IAV NP in the cells from (D). The percentage of infected cells was determined as described for panel (C). In panels (B–E), allantoic fluid from uninfected embryonated chicken eggs (Mock IAV) and PEDSV.15 cell lysate (Mock CSFV) served as mock controls for the IAV and CSFV infections, respectively. The bars represent the mean values of 3 replica, and the error bars show the standard deviation. Statistical significant differences between untreated and IFN-β-primed cells are indicated with an asterisk (P<0.05). The data are representative of at least 3 independent experiments for panels (B) and (C) and of at least 2 independent experiments for panels (A), (D), and (E). DC, dendritic cell; IAV, Influenza A virus; MOI, multiplicity of infection; CSFV, classical swine fever virus; ELISA, enzyme-linked immunosorbent assay; IFN, interferon.

After having shown that IRF7, but not IRF3, was strongly induced in IFN-β-treated GM-CSF DCs, we examined the virus-induced IFN-α responses in GM-CSF DCs pretreated with 50 U/mL IFN-β for 2 h. IAV for which it was reported that type I IFN priming enhances virus-induced cytokine expression (Phipps-Yonas and others 2008) served as a positive control. CpG treatment was employed to confirm the absence of pDCs that are the only porcine cells known to respond to both CpG and CSFV by IFN-α secretion (Summerfield and McCullough 2009). While unprimed cells secreted no or low levels of IFN-α after IAV infection, IFN-β pretreatment for 2 h rendered the GM-CSF DCs clearly responsive to IAV in terms of IFN-α secretion (Fig. 1B). Furthermore, IFN-β pretreatment primed GM-CSF DCs for enhanced IFN-α expression after stimulation with pIC. In contrast, CSFV WT did not trigger any IFN-α response in GM-CSF DCs regardless of the viral dose employed and whether the cells were pretreated with IFN-β or not (Fig. 1B).

In order to determine a potential impact of the virus sensitivity to type I IFN, we monitored in parallel the capacities of CSFV and IAV to infect IFN-β-primed GM-CSF DCs as compared to untreated cells. While IAV infection was not affected by preexposure to IFN-β, the infection levels of CSFV were clearly reduced (Fig. 1C).

Next we increased the IFN-β doses to examine whether a higher concentration of IFN-β would trigger a CSFV-induced IFN-α response in GM-CSF DCs. Relating to the IRF7 levels shown in Fig. 1A, all doses of IFN-β had a similar biological effect on IAV-induced IFN-α expression. Importantly, priming with up to 1,000 U/mL IFN-β for 2 h did not lead to IFN-α secretion upon CSFV infection (Fig. 1D). Interestingly, also the antiviral effect measured in terms of the ability to prevent CSFV NS3 or IAV NP expression was similar with all doses. While CSFV infection was reduced by ∼60%, IAV infection was remarkably resistant to IFN-β even with a dose of 1,000 U IFN-β/mL (Fig. 1E).

Impact of IFN-β priming duration

Priming with different doses of IFN-β for 2 h did not result in augmented IFN-α response to CSFV infection. Therefore, we continued our investigations by modulating the duration of exposure to IFN-β. GM-CSF DCs were cultured in the presence of IFN-β for 6 or 16 h, and subsequently stimulated with pIC or CSFV. Significantly enhanced IFN-α expression was still observed when GM-CSF DCs were primed for 6 h but not for 16 h before stimulation with pIC (Fig. 2 A). Also, under these conditions CSFV was not able to induce consistent and significant amounts of IFN-α. It is important to note that we were also unable to prime other cell types such as PK-15 cells, monocyte-derived DCs, and monocyte-derived macrophages for type IFN responses after infection with CSFV (data not shown).

FIG. 2.

IFN-β dose- and time-dependent IFN-α, IL-6, and IL-1β responses after virus infection or PRR stimulation. GM-CSF DCs were incubated with 50, 100, or 1,000 U/mL of IFN-β for 2, 6, and 16 h. Thereafter, the cells were washed and stimulated with LPS (10 μg/mL), pIC (10 μg/mL), CSFV (MOI 3 TCID₅₀/cell), or IAV (MOI of 0.5 TCID₅₀/cell). The mock controls were as described in Figure 1. After 24 h, IFN-α (A), interleukin −6 (B), and IL-1β (C) in supernatants were quantified by ELISA. The bars represent the mean values of 3 replica, and the error bars show the standard deviation. Statistical significant differences between untreated and IFN-β-primed cells are indicated with an asterisk (P<0.05). The data are representative of at least 3 independent experiments. IL, interleukin.

IFN-β priming and proinflammatory cytokines

In acute CSF, besides IFN-α, also elevated levels of pro-inflammatory cytokines are detected in the serum of CSFV-infected pigs (Sanchez-Cordon and others 2002; Summerfield and others 2006; Jamin and others 2008). Therefore, we analyzed whether IFN-β pretreatment of GM-CSF DCs followed by CSFV infection would lead to enhanced expression of the IL-1β and IL-6 cytokines, as models for the pro-inflammatory response. Neither IL-1β nor IL-6 secretion was enhanced after exposure of GM-CSF DCs to various concentrations of IFN-β (50, 100, and 1,000 U/mL) for different time periods (2, 6, and 16 h) before CSFV infection (Fig. 2B,C). Yet, a priming effect of both cytokines was observed with LPS and pIC treatment confirming that GM-CSF DCs are able to display enhanced pro-inflammatory cytokine responses after IFN-β treatment. Compared with unprimed GM-CSF DCs, LPS- and pIC-induced IL-6 responses were clearly augmented, in particular with short priming periods (Fig. 2B). The strongest enhancement of IL-1β expression was observed after LPS treatment and only when priming was limited to 2 h of pretreatment (Fig. 2C). IAV did induce neither IL-6 nor IL-1β (data not shown).

CSFV-induced cytokine responses in GM-CSF DCs pretreated with conditioned medium derived from CSFV-stimulated pDCs

According to our hypothesis, pDC-derived IFN-α would act on cDCs to prime enhanced cytokine responses in vivo. Yet, in our in vitro studies we employed recombinant IFN-β to mimic a type I IFN-containing environment. In order to investigate potential effects of other type I IFN or other cytokines produced by pDC could have on GM-CSF DCs, we compared cytokine production of GM-CSF DCs that had been pretreated with either recombinant IFN-β or with conditioned medium derived from CSFV-activated enriched pDCs. In both cases, primed DCs responded with significantly higher IFN-α (Fig. 3A) or IL-6 expression (Fig. 3B) compared to unprimed DC when stimulated with IAV or LPS, respectively. In contrast, neither the conditioned medium containing IFN-α derived from CD172a⁺ PBMC composed of monocytes, conventional blood DC, and pDC nor recombinant IFN-β was able to prime the GM-CSF DC for CSFV-induced activation in terms of type I IFN or IL-6 responses.

FIG. 3.

Priming effect of conditioned medium derived from CSFV-infected enriched pDCs on IFN-α and IL-6 responses in GM-CSF DCs CD172a⁺ peripheral blood mononuclear cells (enriched pDCs) were mock (CSFV Mock) treated or infected with wild-type CSFV (CSFV WT) at an MOI of 2 TCID₅₀/cell and 24 h postinfection (p.i.) cell culture supernatant were harvested and treated with 75 mJ UV to inactivate the virus. GM-CSF DCs were then incubated for 3 h with either pDC-derived medium diluted 1:2, 1:4, or 1:40, or with medium conditioned with 50 U/mL recombinant IFN-β, which was added after (IFN-β) or before UV treatment (UV IFN-β) or with medium only (No IFN). Thereafter, GM-CSF DCs were stimulated with LPS (10μg/mL), CSFV (MOI of 2 TCID₅₀/cell), or IAV (MOI of 0,5 TCID₅₀/cell), and 24 h p.i. IFN-α (A) and IL-6 (B) were measured by ELISA. The mock controls were as described in Figure 1. The bars represent the mean value of 3 replica, and the error bars show the standard deviation. Statistical significant differences between untreated and type I IFN-primed cells are indicated with an asterisk (P<0.05). The data are representative of 3 independent experiments.

IFN-α responses to N^pro(D₁₃₆N) mutant CSFV can be primed by IFN-β

A recent report published by our laboratory showed that the CSFV-encoded IFN antagonist N^pro interacts not only with IRF3 but also with IRF7 (Fiebach and others 2011). Therefore, we asked whether the inability of CSFV to induce type I IFN or cytokine responses in IFN-β-primed GM-CSF DCs expressing IRF7 was related to the function of N^pro. In order to address this, we employed the N^pro(D₁₃₆N) mutant CSFV that is unable to interact with IRF3 and IRF7 (Ruggli and others 2009; Fiebach and others 2011). Interestingly, the N^pro(D₁₃₆N) mutant virus was more sensitive to IFN-β pretreatment than the WT virus in terms of its ability to infect GM-CSF DCs, irrespectively of the MOI employed (Fig. 4A). Importantly, even though IFN-β priming reduced the infection of the N^pro(D₁₃₆N) mutant virus by at least 70%, the IFN-α levels were comparable in primed and nonprimed cells (Fig. 4B). To take the reduced infection caused by the IFN-β priming into account, we calculated the IFN-α levels per NS3-positive cell after infection with the N^pro(D₁₃₆N) mutant virus (Fig. 4C). The results show clearly that IFN-β priming promotes an elevated IFN-α secretion in NS3⁺ GM-CSF DCs only when the cells are infected with a CSFV mutant devoid of IRF7 binding. This was also confirmed by simultaneous infection and IFN-β treatment (Fig. 4D). Under these conditions, IFN-β enhanced also the absolute IFN-α level induced by the N^pro(D₁₃₆N) mutant virus. Again, the WT CSFV remained unable to induce IFN-α. Also, the simultaneous addition of IFN-β with IAV virus primed the GM-CSF DCs to respond by production of IFN-α, contrasting with the pIC-induced IFN-α secretion that required IFN-β pretreatment. Under these priming conditions, none of the viruses elicited an IL-6 or IL-1β response (data not shown).

FIG. 4.

Role of N^pro in viral escape from the IFN-β-mediated IFN-α amplification loop. In (A–C), GM-CSF DCs were incubated with 50 U IFN-β/mL for 2 h, washed, and incubated with wild-type (WT) or N^pro(D₁₃₆N) mutant CSFV at an MOI of 0.5, 1, or 3 TCID₅₀/cell. About 24 h after infection, the percentage of NS3-expressing cells was determined by flow cytometry (A), IFN-α was quantified by ELISA (B), and the IFN-α response (A) per NS3-positive DC (B) was calculated (C). In (D), GM-CSF DCs were incubated simultaneously with 50 U IFN-β/mL and with WT or N^pro(D₁₃₆N) mutant CSFV at an MOI 3 of TCID₅₀/cell, IAV at an MOI of 0.5 TCID₅₀/cell, pIC (10 μg/mL), medium or mock controls as described in Figure 1. After 2 h the cells were washed and cultured for 24 h or 48 h in medium containing fetal bovine serum before quantification of IFN-α in the supernatants by ELISA. The bars represent the mean values of 3 replica, and the error bars show the standard deviation. Statistical significant differences between untreated and IFN-β-primed cells are indicated with an asterisk (P<0.05). The data in (A, B) are representative of at least 3 independent experiments.

Discussion

In this study we examined IFN-α and cytokine responses of GM-CSF DCs upon CSFV infection in the context of a type I IFN containing environment. Type I IFN has broad antiviral and immunomodulatory effects. On one hand, it has beneficial effects in viral infections restricting viral dissemination and promoting adaptive antiviral effector immune responses. On the other hand, it is also known for promoting immunopathological events when released at high levels during longer periods (Theofilopoulos and others 2005). Accordingly, during CSFV infection a strong and long-lasting IFN-α response is detectable in the serum of infected pigs, and this has been suggested to relate to disease severity rather than to protective immune responses (Summerfield and others 2006; Summerfield 2011). Considering that CSFV efficiently prevents type I IFN induction in all analyzed target cells through the induction of viral N^pro-mediated proteasomal degradation of IRF3, the question arose about the cellular source of the systemic type I IFN found in diseased pigs. So far pDCs are the only cell type known to respond to WT CSFV infection by IFN-α production in vivo and in vitro. For IFN-α induction this specialized cell type relies on constitutively expressed IRF7, a unique property of pDCs. In contrast, non-pDCs only employ IRF7 after its induction during the IFN-dependent autocrine-paracrine amplification loop (Marie and others 1998). Notably, IRF7 is the most potent transcription factor for the type I IFN system (Sato and others 2000). These facts led us to the sensitization hypothesis proposing that pDC-derived IFN-α would promote systemic IRF7 expression and enable non-pDCs to bypass the N^pro-mediated IRF3 inhibition. This would result in IFN-α and/or pro-inflammatory cytokine expression in a wide range of cells. In principle, this scenario may represent a key event contributing to the cytokine storm observed in the acute phase of the disease and would also have been supported by the fact that type I IFN is an early cytokine detectable in the serum of pigs suffering from acute CSF. However, despite the increased expression of IRF7 transcripts after IFN-β exposure, preincubation of GM-CSF DCs with various concentrations of IFN-β or condition medium from CSFV-activated enriched pDC for different time periods did not result in enhanced IFN-α, Il-6, or IL-1β expression in response to CSFV WT infection. This was not due to the culture conditions, since enhanced IFN-α responses were found upon IAV infections in IFN-β pre-exposed GM-CSF DCs.

In contrast to the responses to WT CSFV, those to CSFV with a functionally defective N^pro unable to interact with IRF3 and IRF7 (Fiebach and others 2011) were enhanced by IFN-β priming. This was observed either by calculating IFN-α units per NS3⁺ GM-CSF DC or when CSFV and IFN-β were added simultaneously. In this context, it was observed that CSFV was clearly more sensitive to IFN-β pretreatment when compared to IAV that showed a remarkable resistance to IFN-β, even at high concentrations. In addition, the N^pro(D₁₃₆N) mutant was even more sensitive to IFN-β pretreatment when compared to the parent virus. The reason for this enhanced sensitivity is unknown, but it could possibly relate to the antiviral activity induced by the mutant but not by the WT. Alternatively, this observation may point toward an interference of N^pro with the type I IFN response pathway. We also observed lower percentages of NS3-expressing cells in BMHC-derived Flt3L-induced DCs, monocyte-derived DCs, and macrophages when they were treated with various doses of IFN-β before CSFV infection (data not shown). These results relate to the observation that pestiviruses are sensitive to the action of type I IFN only when the IFN is added before infection but has little or no antiviral effect is found when added after infection (Schweizer and others 2006; Suter and others 2011).

The present work emphasizes the important role of N^pro in blocking innate immune responses. This viral protein not only prevents IRF3-dependent IFN responses but also acts on the autocrine–paracrine amplification loop most probably by preventing also IRF7-dependent IFN responses. This conclusion is based on the specific and potent interaction of N^pro with IRF7, although the direct mechanism and consequences of this interaction remain to be elucidated (Fiebach and others 2011). Our conclusion is also in line with the observation that in pDCs, which are known to respond in an IRF3-independent, but IRF7-dependent manner (Honda and others 2005), WT CSFV infection leads to reduced IFN-α responses when compared to infections with N^pro(D₁₃₆N) mutant virus (Fiebach and others 2011). For the immunopathogenesis of CSF, we conclude that the aberrant type I IFN and inflammatory cytokines responses observed in vivo cannot be explained by type I IFN-induced sensitization of immune cells, as seen in other viral infections such as IAV (Osterlund and others 2005; Phipps-Yonas and others 2008). We currently have several explanations for the discrepancy of in vitro and in vivo results. First, the high tropism of CSFV for macrophages and DCs and its accumulation in lymphoid tissue results in overall high type I IFN and cytokine responses although at the cellular level the responses are low. Second, in vivo DCs and macrophages are exposed to a more complex (cytokine) environment resulting in a strong CSFV-induced activation. Despite this still puzzling situation, our data demonstrate that certain viruses have evolved that inhibit not only primary type I IFN responses mediated by IRF3 but also the IRF7-amplified secondary IFN responses in non-pDCs.

Footnotes

Acknowledgments

We are very grateful to Daniel Brechbühl for providing the sterna.

This project was supported by the Swiss National Science Foundation (project 310000-116800).

Author Disclosure Statement

No competing financial interests exist.

References

Alves

, Guzylack-Piriou

, Juillard

, Audonnet

, Doel

, Dawson

, Golde

, Gerber

, Peduto

, McCullough

others.2009. Innate immune defenses induced by CpG do not promote vaccine-induced protection against foot-and-mouth disease virus in pigs. Clin Vaccine Immunol, 16,8:1151–1157.

Balmelli

, Vincent

, Rau

, Guzylack-Piriou

, McCullough

, Summerfield

. 2005. Fc gamma RII-dependent sensitisation of natural interferon-producing cells for viral infection and interferon-alpha responses. Eur J Immunol, 35,8:2406–2415.

Bauhofer

, Summerfield

, McCullough

, Ruggli

. 2005. Role of double-stranded RNA and Npro of classical swine fever virus in the activation of monocyte-derived dendritic cells. Virology, 343,1:93–105.

Bauhofer

, Summerfield

, Sakoda

, Tratschin

, Hofmann

, Ruggli

. 2007. Classical swine fever virus Npro interacts with interferon regulatory factor 3 and induces its proteasomal degradation. J Virol, 81,7:3087–3096.

Carrasco

, Rigden

, Vincent

, Balmelli

, Ceppi

, Bauhofer

, Tache

, Hjertner

, McNeilly

, van Gennip

others.2004. Interaction of classical swine fever virus with dendritic cells. J Gen Virol, 85,Pt 6:1633–1641.

Carrasco

, Ruiz-Villamor

, Gomez-Villamandos

, Salguero

, Bautista

, Macia

, Quezada

, Jover

. 2001. Classical swine fever: morphological and morphometrical study of pulmonary intravascular macrophages. J Comp Pathol, 125,1:1–7.

Fiebach

, Guzylack-Piriou

, Python

, Summerfield

, Ruggli

. 2011. Classical swine fever virus Npro limits type I interferon induction in pDC by interacting with interferon regulatory factor 7. J Virol, 85,16:8002–8011.

Geisbert

, Jahrling

. 2004. Exotic emerging viral diseases: progress and challenges. Nat Med, 10,Suppl. 12:S110–S121.

Guzylack-Piriou

, Balmelli

, McCullough

, Summerfield

. 2004. Type-A CpG oligonucleotides activate exclusively porcine natural interferon-producing cells to secrete interferon-alpha, tumour necrosis factor-alpha and interleukin-12. Immunology, 112,1:28–37.

10.

Guzylack-Piriou

, Bergamin

, Gerber

, McCullough

, Summerfield

. 2006. Plasmacytoid dendritic cell activation by foot-and-mouth disease virus requires immune complexes. Eur J Immunol, 36,7:1674–1683.

11.

Honda

, Yanai

, Negishi

, Asagiri

, Sato

, Mizutani

, Shimada

, Ohba

, Takaoka

, Yoshida

others.2005. IRF-7 is the master regulator of type-I interferon-dependent immune responses. Nature, 434,7034:772–777.

12.

Husser

, Alves

, Ruggli

, Summerfield

. 2011. Identification of the role of RIG-I, MDA-5 and TLR3 in sensing RNA viruses in porcine epithelial cells using lentivirus-driven RNA interference. Virus Res, 159,1:9–16.

13.

Jamin

, Gorin

, Cariolet

, Le Potier

, Kuntz-Simon

. 2008. Classical swine fever virus induces activation of plasmacytoid and conventional dendritic cells in tonsil, blood, and spleen of infected pigs. Vet Res, 39,1:7.

14.

Knoetig

, McCullough

, Summerfield

. 2002. Lipopolysaccharide-induced impairment of classical swine fever virus infection in monocytic cells is sensitive to 2-aminopurine. Antiviral Res, 53,1:75–81.

15.

Knoetig

, Summerfield

, Spagnuolo-Weaver

, McCullough

. 1999. Immunopathogenesis of classical swine fever: role of monocytic cells. Immunology, 97,2:359–366.

16.

Liu

. 2005. IPC: professional type 1 interferon-producing cells and plasmacytoid dendritic cell precursors. Annu Rev Immunol, 23:275–306.

17.

Marie

, Durbin

, Levy

. 1998. Differential viral induction of distinct interferon-alpha genes by positive feedback through interferon regulatory factor-7. EMBO J, 17,22:6660–6669.

18.

McCullough

, Ruggli

, Summerfield

. 2009. Dendritic cells—at the front-line of pathogen attack. Vet Immunol Immunopathol, 128,1–3:7–15.

19.

Ocana-Macchi

, Bel

, Guzylack-Piriou

, Ruggli

, Liniger

, McCullough

, Sakoda

, Isoda

, Matrosovich

, Summerfield

. 2009. Hemagglutinin-dependent tropism of H5N1 avian influenza virus for human endothelial cells. J Virol, 83,24:12947–12955.

20.

Osterlund

, Veckman

, Siren

, Klucher

, Hiscott

, Matikainen

, Julkunen

. 2005. Gene expression and antiviral activity of alpha/beta interferons and interleukin-29 in virus-infected human myeloid dendritic cells. J Virol, 79,15:9608–9617.

21.

Phipps-Yonas

, Seto

, Sealfon

, Moran

, Fernandez-Sesma

. 2008. Interferon-beta pretreatment of conventional and plasmacytoid human dendritic cells enhances their activation by influenza virus. PLoS Pathog, 4,10:e1000193.

22.

Ruggli

, Bird

, Liu

, Bauhofer

, Tratschin

, Hofmann

. 2005. N(pro) of classical swine fever virus is an antagonist of double-stranded RNA-mediated apoptosis and IFN-alpha/beta induction. Virology, 340,2:265–276.

23.

Ruggli

, Summerfield

, Fiebach

, Guzylack-Piriou

, Bauhofer

, Lamm

, Waltersperger

, Matsuno

, Liu

, Gerber

others.2009. Classical swine fever virus can remain virulent after specific elimination of the interferon regulatory factor 3-degrading function of Npro. J Virol, 83,2:817–829.

24.

Ruggli

, Tratschin

, Mittelholzer

, Hofmann

. 1996. Nucleotide sequence of classical swine fever virus strain Alfort/187 and transcription of infectious RNA from stably cloned full-length cDNA. J Virol, 70,6:3478–3487.

25.

Ruggli

, Tratschin

, Schweizer

, McCullough

, Hofmann

, Summerfield

. 2003. Classical swine fever virus interferes with cellular antiviral defense: evidence for a novel function of N(pro) J Virol, 77,13:7645–7654.

26.

Sanchez-Cordon

, Nunez

, Salguero

, Carrasco

, Gomez-Villamandos

. 2005. Evolution of T lymphocytes and cytokine expression in classical swine fever (CSF) virus infection. J Comp Pathol, 132,4:249–260.

27.

Sanchez-Cordon

, Romanini

, Salguero

, Nunez

, Bautista

, Jover

, Gomez-Villamos

. 2002. Apoptosis of thymocytes related to cytokine expression in experimental classical swine fever. J Comp Pathol, 127,4:239–248.

28.

Sato

, Suemori

, Hata

, Asagiri

, Ogasawara

, Nakao

, Nakaya

, Katsuki

, Noguchi

, Tanaka

others.2000. Distinct and essential roles of transcription factors IRF-3 and IRF-7 in response to viruses for IFN-alpha/beta gene induction. Immunity, 13,4:539–548.

29.

Schweizer

, Matzener

, Pfaffen

, Stalder

, Peterhans

. 2006. “Self” and “nonself” manipulation of interferon defense during persistent infection: bovine viral diarrhea virus resists alpha/beta interferon without blocking antiviral activity against unrelated viruses replicating in its host cells. J Virol, 80,14:6926–6935.

30.

Summerfield

. 2011. Viewpoint: factors involved in type I interferon responses during porcine virus infections. Vet Immunol Immunopathol[Epub ahead of print]10.1016/j.vetimm.2011.03.011

31.

Summerfield

, Alves

, Ruggli

, de Bruin

, McCullough

. 2006. High IFN-alpha responses associated with depletion of lymphocytes and natural IFN-producing cells during classical swine fever. J Interferon Cytokine Res, 26,4:248–255.

32.

Summerfield

, Guzylack-Piriou

, Schaub

, Carrasco

, Tache

, Charley

, McCullough

. 2003. Porcine peripheral blood dendritic cells and natural interferon-producing cells. Immunology, 110,4:440–449.

33.

Summerfield

, McCullough

. 2009. The porcine dendritic cell family. Dev Comp Immunol, 33,3:299–309.

34.

Suter

, Summerfield

, Thomann-Harwood

, McCullough

, Tratschin

, Ruggli

. 2011. Immunogenic and replicative properties of classical swine fever virus replicon particles modified to induce IFN-alpha/beta and carry foreign genes. Vaccine, 29,7:1491–503.

35.

Theofilopoulos

, Baccala

, Beutler

, Kono

. 2005. Type I interferons (alpha/beta) in immunity and autoimmunity. Annu Rev Immunol, 23:307–336.

36.

van Oirschot

. 1988. Description of the Virus. Liess

. Classical swine fever and related viral infections. Boston: Mantimus Nishoff Publishing, 1–25.

37.

van Oirschot

. 2004. Hog cholera. Infectioius Diseases of Livestock. 2nd. Cape Town: Oxford University Press, 975–986.

N pro of Classical Swine Fever Virus Prevents Type I Interferon-Mediated Priming of Conventional Dendritic Cells for Enhanced Interferon-α Response

Abstract

Introduction

Material and Methods

Generation of bone marrow hematopoietic cells-derived DCs using granulocyte macrophage-colony stimulating factor

Isolation and culture of enriched pDCs

Viruses

Infection and treatment of DCs

Detection of intracellular viral proteins by flow cytometry

mRNA isolation and real-time reverse transcription–polymerase chain reaction

Detection of cytokines in the cell culture supernatant

Statistical analysis

Results

Pretreatment of GM-CSF DCs with various doses of IFN-β does not lead to enhanced IFN-α in response to CSFV

Impact of IFN-β priming duration

IFN-β priming and proinflammatory cytokines

CSFV-induced cytokine responses in GM-CSF DCs pretreated with conditioned medium derived from CSFV-stimulated pDCs

IFN-α responses to Npro(D136N) mutant CSFV can be primed by IFN-β

Discussion

Footnotes

Acknowledgments

Author Disclosure Statement

References

IFN-α responses to N^pro(D₁₃₆N) mutant CSFV can be primed by IFN-β