Hepatitis C Virus Infection Impairs IRF-7 Translocation and Alpha Interferon Synthesis in Immortalized Human Hepatocytes

Abstract

Hepatitis C virus (HCV) establishes chronic infection in a significant number of infected humans, although the mechanisms for chronicity remain largely unknown. We have previously shown that HCV infection in immortalized human hepatocytes (IHH) induces beta interferon (IFN-β) expression (T. Kanda, R. Steele, R. Ray, and R. B. Ray, J. Virol. 81:12375-12381, 2007). However, the regulation of the downstream signaling pathway for IFN-α production by HCV is not clearly understood. In this study, the regulation of the IFN signaling pathway following HCV genotype 1a (clone H77) or genotype 2a (clone JFH1) infection of IHH was examined. HCV infection upregulated expression of total STAT1 but failed to induce phosphorylation and efficient nuclear translocation. Subsequent study revealed that HCV infection induces IFN-stimulated response element activation, as evidenced by upregulation of 2′,5′-oligoadenylate synthetase 1. However, nuclear translocation of IRF-7 was impaired following HCV infection. In HCV-infected IHH, IFN-α expression initially increased (up to 24 h) and then decreased at later time points, and IFN-α-inducible protein 27 was not induced. Interestingly, HCV infection blocked IRF-7 nuclear translocation upon poly(I-C) or IFN-α treatment of IHH. Together, our data suggest that HCV infection enhances STAT1 expression but impairs nuclear translocation of IRF-7 and its downstream molecules. These impairments in the IFN-α signaling pathway may, in part, be responsible for establishment of chronic HCV infection.

Hepatitis C virus (HCV) infection affects approximately 3.2 million people in the United States (1). The approved treatment for HCV infection is pegylated alpha interferon (IFN-α) alone or in combination with ribavirin. This leads to clearance of HCV in ∼50% and ∼80% of the cases of HCV genotype 1 and 2 infection, respectively. Type I IFNs are crucial components of the innate immune response to virus attack. The host response is triggered when a pathogen-associated molecular pattern (PAMP) presented by the infecting virus is recognized and engaged by specific PAMP receptor factors expressed in the host cell, initiating signals that ultimately induce the expression of antiviral effector genes. IFN-α and IFN-β are rapidly synthesized after virus infection and trigger intracellular signaling events. The subsequent expression of IFN-stimulated genes (ISGs) is central to these antiviral responses. IFN-stimulated gene factor 3 (ISGF3) assembles and translocates from the cytoplasm to the nucleus upon IFN stimulation. ISGF3 is a multisubunit transcription factor that interacts with the IFN-stimulated response element (ISRE) present in the promoters of ISGs (25). ISGF3 consists of hetero-oligomers of signal transducers and activators of transcription 1 and 2 (STAT1 and STAT2) and IFN regulatory factor 9 (IRF-9). Homodimers of STAT1-α and heterodimers of STAT1 and STAT2 are also activated, and IRF-9 is indispensable for their formation, by binding to inverted repeat elements in the promoters of ISGs to induce transcription (29).

Interferon and ISGs are amplified during chronic HCV infection (2, 22, 27) but fail to eliminate the virus from the liver in a large number of HCV-infected patients. IFN-induced genes are also stimulated during HCV RNA replication within the liver of acutely infected chimpanzees (3). The changes due to endogenous antiviral responses in liver (e.g., induction of type I IFN-induced genes) occur by intrahepatic gene expression as soon as HCV RNA is detectable in the serum of chimpanzees (31). We have shown previously that in vitro HCV infection of IHH results in nuclear localization of IRF-3 and enhances IFN-β expression (13). However, modulatory effects of the downstream IFN signaling pathway in cells infected with HCV are unknown. In this study, we investigated the molecular determinants of the interferon signaling pathway following HCV infection in hepatocytes, and we identified specific sites responsible for impairment of the downstream intracellular IFN signaling pathway.

MATERIALS AND METHODS

Cell culture, transfection, and HCV infection.

IHH and Huh-7 cells were grown in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum, 100 U/ml of penicillin G, and 100 μg/ml of streptomycin at 37°C in a 5% CO₂ atmosphere. We cultured HCV genotypes 1a and 2a in IHH and Huh-7.5 cells, respectively. HCV genotype 2a also infects IHH (12). In this study, IHH were infected with HCV genotype 1a (clone H77) or genotype 2a (clone JFH1-GFP) at a multiplicity of infection (MOI) of 0.1 to 1 focus-forming units/cell in a minimum volume of serum-free medium. JFH1-GFP was generated by inserting green fluorescent protein (GFP) in frame at domain III of the NS5A region of JFH-1 (11). Cells were treated with 400 units of IFN-α for 1 to 17 h. After 8 h of adsorption of virus, DMEM supplemented with 5% heat-inactivated fetal bovine serum was added. IRF-7-GFP plasmid DNA (kindly provided by Betsy Barnes, NJMSUH Cancer Center) was transfected into IHH by using Lipofectamine reagent (Invitrogen) for subcellular localization studies at 72 h postinfection with HCV.

Immunofluorescence.

Mock- or HCV-infected IHH were washed in phosphate-buffered saline (PBS), fixed with acetone-acetic acid for 30 min at −20°C, and blocked with 3% bovine serum albumin for 1 h. Fixed cells were incubated with an HCV NS5A-specific mouse monoclonal antibody (kindly provided by Chen Liu, University of Florida) and STAT1-specific rabbit polyclonal antibody (Santa Cruz Biotechnology) for 1 h. Cells were washed and incubated with anti-mouse Ig conjugated with Alexa 488 (Molecular Probes) and anti-rabbit Ig conjugated with Alexa 594 (Molecular Probes) secondary antibodies for 1 h at room temperature. Finally, cells were washed and mounted for confocal microscopy (Olympus FV1000). Images were superimposed digitally for fine comparisons.

RNA quantitation.

Total RNA was isolated using TRIzol reagent (Invitrogen). cDNA was synthesized using a random hexamer and the ABI reverse transcriptase (RT) kit (Applied Biosystems). Real-time PCR was performed on the cDNA for RNA quantitation by using TaqMan Universal PCR master mix (Applied Biosystems) and the minor groove binding protein labeled with 6-carboxyfluorescein (FAM-MGB) probe for alpha interferon-inducible protein 27 (IFI-27; Hs00271467_m1), MxA (Hs00895598_m1), 2′,5′-oligoadenylate synthetase 1 (OAS1; Hs00973637_m1), and HCV (AI6Q1GI). FAM-MGB probe for glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Hs99999905_m1; Applied Biosystems TaqMan gene expression assay) was used as an endogenous control. IFN-α RΝΑ was measured by quantitative RT-PCR using a set of primers corresponding to several IFN-α subtypes (kindly provided by Betsy Barnes).

Luciferase assay.

IHH infected with HCV (3 or 8 days) or mock-infected cells were transfected with the plasmid ISRE-TA-luc (Clontech) containing five copies of the ISRE enhancer elements upstream of the firefly luciferase gene. At 48 h posttransfection, cells were lysed with reporter lysis buffer (Promega), and luciferase activity was determined using a luminometer (Glomax). Luciferase activity was normalized with respect to the protein concentration as described previously (13). The results are presented as means from three independent experiments.

Immunoblot analysis.

Mock- or HCV-infected cells were lysed in a sample buffer, subjected to SDS-PAGE, and transferred onto nitrocellulose membranes. The membranes were probed with a monoclonal antibody to total STAT1 (Santa Cruz), phosphorylated STAT1, or a polyclonal antibody to double-stranded RNA (dsRNA)-activated protein kinase (PKR; Cell Signaling). The membranes were reprobed with a monoclonal antibody to actin or tubulin (Santa Cruz Biotechnology) for comparison of the protein loads. Proteins were visualized using an enhanced chemiluminescence (ECL) Western blot substrate (Pierce) and subjected to densitometric scanning by using an image analyzer and Quantity One software (Bio-Rad).

RESULTS

HCV infection enhances total STAT1 but inhibits phosphorylation.

We examined the expression status of total STAT1 and phosphorylated STAT1 protein in HCV-infected IHH. Cells were infected with HCV genotype 1a or 2a, and cell lysates were prepared 10 days postinfection for Western blot analysis. Increased expression of STAT1 was observed after HCV infection of IHH compared to mock-infected control hepatocytes (Fig. 1 A). However, phosphorylated STAT1 protein was not detected following HCV genotype 1a or genotype 2a infection of IHH. On the other hand, mock-infected cells treated with IFN-α displayed an enhanced level of phospho-STAT1. We also examined the phospho-STAT1 status in IFN-α-treated HCV-infected IHH. The phospho-STAT1 expression was restored in HCV-infected IFN-α-treated IHH, although the extent of STAT1 phosphorylation was much lower (∼50%) than for IHH treated with IFN-α.

FIG. 1.

HCV infection induces STAT1 expression in hepatocytes. (A) IHH were infected with HCV genotype 1a or genotype 2a and incubated for 10 days. Western blot analysis was performed for STAT1 or phospho-STAT1 protein expression with specific antibodies. The blot was reprobed with an antibody to tubulin for comparison of protein loads. Mock- and HCV-infected IHH were also treated with IFN-α (400 units) for 1 h as a positive control to evaluate the phopsho-STAT1 expression status. (B) IHH were either mock treated (a), transfected with poly(I-C) as a positive control (b), or infected with HCV genotype 1a (c to f). Mock-treated and poly(I-C)-treated cells were fixed and stained with STAT1 antibody (red) after 24 h. HCV-infected IHH were fixed at day 10 postinfection and stained for STAT1 (red) and HCV NS5A (green). Nuclei were visualized by staining with 4′,6-diamidino-2-phenylindole (blue). Merged images from panels c to e are shown in panel f. Arrowheads indicate the selected uninfected cells, and arrows point to HCV-infected cells showing green fluorescence.

Next, we examined localization of STAT1 in HCV-infected cells and in the neighboring uninfected cells. For this, IHH were infected with HCV genotype 1a (12) or GFP-tagged HCV genotype 2a (11) at an MOI of 0.05. Hepatocytes were fixed and subjected to immunofluorescence after 10 days of infection by using anti-STAT1 and anti-HCV NS5A antibodies (Fig. 1B). In mock-infected cells, a weak level of endogenous STAT1 was localized exclusively in the cytoplasm, indicative of an inactive state (Fig. 1B, panel a), while cells transfected with poly(I-C) as a positive control displayed nuclear localization of STAT1 (panel b). On the other hand, cells infected with HCV genotype 1a exhibited a predominant perinuclear localization of STAT1 (panels c to f). Occasionally, efficient nuclear localization of STAT1 was observed in HCV-infected cells. Cells stained with a secondary antibody only and used as a negative control did not display detectable fluorescence (data not shown). Similar results were obtained on days 3 and 7 postinfection, as well as from cells infected with HCV genotype 2a. Furthermore, we have observed that cells infected with HCV display enhanced STAT1 expression, but not in the neighboring uninfected cells. Together, these results suggest that HCV infection enhances total STAT1 expression and impairs STAT1 phosphorylation in infected cells.

HCV infection induces ISRE promoter activity and OAS1 gene expression.

Type I IFN induces phosphorylation of STAT1 and STAT2, which form ISGF3, a ternary complex that includes IRF-9. Accumulation of unphosphorylated STAT1 also regulates the downstream signaling pathways (7). We measured ISRE promoter activity in virus-infected hepatocytes (at 5 days postinfection) by using a synthetic promoter tagged with a luciferase construct. Our results suggested induction of ISRE promoter activity by IHH infected with either HCV genotype 1a or 2a compared to mock-infected control hepatocytes (Fig. 2 A). Similar results were obtained 10 days after HCV infection (data not shown).

FIG. 2.

HCV infection induces ISRE and modulates ISRE response genes. (A) IHH infected with HCV genotype 1a or genotype 2a (5 days postinfection) were transfected with the plasmid ISRE-TA-luc. Cell extracts were prepared after 48 h, and relative luciferase activity was measured. IHH infected with HCV exhibited enhanced ISRE promoter activity. Results presented are means with standard errors from three different experiments. (B and C) Total cellular RNA was extracted from HCV-infected IHH 10 days postinfection. Intracellular gene expression levels of 2′,5′-OAS1, MxA, and GAPDH were measured by real-time RT-PCR. The OAS-1/GAPDH and MxA/GAPDH ratios are presented as the fold induction relative to basal levels in mock-infected cells. Uninfected IHH were treated with IFN-α (400 units) for 8 h to examine MxA induction by real-time RT-PCR and compared with results with the untreated control. The results shown are means from three independent experiments. (D) Western blot analysis for PKR protein expression using specific antibody. The blot was reprobed with an antibody to actin for comparisons of protein loads. Densitometric scanning for relative protein expression (as fold differences) are shown at the bottom.

Because ISRE promoter activation in HCV-infected IHH was observed, we focused our examination on well-characterized IFN response genes. OAS1 is a major component of the antiviral pathways induced by IFNs. In the presence of dsRNA, they polymerize ATP to form 2′,5′-oligoadenylate oligomers, which in turn activate the latent RNase L, causing mRNA degradation. A significantly higher mRNA expression level of 2′,5′-OAS was observed in IHH infected with either HCV genotype 1a or 2a than in mock-infected control hepatocytes after 10 days of infection (Fig. 2B). MxA protein, a specific and sensitive marker for type I interferon production, is highly expressed in peripheral blood mononuclear cells of chronic HCV disease (21). The human MxA protein is an IFN-induced GTPase that has antiviral activity against various RNA viruses (10, 17). A significant change in mRNA expression of MxA was not observed in HCV-infected IHH compared to mock-infected control cells (Fig. 2C). To further examine whether MxA expression can be induced in IHH, MxA mRNA was measured in IFN-α-treated IHH, and there was a ∼3-fold activation compared to untreated IHH. A similar observation was reported for HCV genotype 2a-infected Huh-7 cells (9). Real-time PCR analyses also demonstrated that MxA is not induced in HCV-infected liver (23).

IFN-inducible PKR is one of a number of host ISGs (28). Nearly all mammalian cells express PKR at low levels (15). dsRNA, produced during RNA viral replication, is a potent activator of PKR (24). Activated PKR in turn induces phosphorylation of PKR and eukaryotic initiation factor 2a (eIF2a), which inhibits protein synthesis, including that of virally encoded proteins (26). Infection of IHH with HCV genotype 1a or 2a modestly enhanced (1.5- to 1.7-fold) the expression of PKR (Fig. 2D). Together, our results suggest that HCV infection activates the downstream ISRE response genes at different levels.

HCV infection impairs nuclear translocation of IRF-7 and inhibits IFN-α/IFI27 mRNA expression.

IRF-7 is a downstream IFN signaling molecule. IRF-7 undergoes phosphorylation when activated, translocates into the nucleus, and induces IFN-α transcription. We examined subcellular localization of IRF-7 in HCV-infected IHH. Cells were infected with HCV genotype 1a or 2a and incubated for 10 days. IRF-7 was detected by immunofluorescence primarily in the cytoplasm of HCV-infected cells (Fig. 3A). Staining of infected cells also displayed perinuclear localization of NS5A, as expected. Expression of NS5A and IRF-7 was mostly in separate locations, as shown with the two distinct fluorochromes. On the other hand, cells transfected with poly(I-C) and IRF-7 displayed nuclear localization of IRF-7, as expected (data not shown). Similar results were observed with HCV-infected cells examined after 3 days of incubation.

FIG. 3.

HCV infection impairs nuclear translocation of IRF-7 and inhibits IFN-α/IFI27 expression. (A) HCV genotype 1a infection exhibits cytoplasmic localization of IRF-7. IHH were infected with HCV and transfected at day 8 postinfection with recombinant GFP-conjugated IRF-7. Cells were fixed at 48 h posttransfection and stained with NS5A antibody (c; red color). Nuclei were visualized by staining with 4′,6-diamidino-2-phenylindole (a; blue color) and IRF-7 was stained with GFP (b; green color). Merged figures from panels a to c are shown in panel d. Arrows indicate the selected HCV-infected and IRF-7-transfected cells. (B and C) HCV infection in IHH impairs expression of IFN-α and IFI27. Total cellular RNA was extracted from HCV-infected IHH after 10 days of infection. Intracellular gene expression of IFN-α (B) and downstream signaling molecule IFI27 (C) were measured by real-time RT-PCR. GAPDH was used as an internal control. The fold changes of IFN-α and IFI27 mRNA in HCV-infected IHH, relative to mock-infected cells, are presented after normalizing with GAPDH mRNA expression. The results presented are means from three independent experiments. (D) Kinetics of IFN-α and IFI27 mRNA expression following HCV genotype 1a infection of IHH. IFN-α and IFI27 expression were measured at the indicated time points by real-time RT-PCR as described above.

IFN-α is produced by most cells in response to viral infection. Because we could not detect IFN-α from HCV-infected IHH culture supernatants, the status of IFN-α synthesis and its downstream molecules was further examined at the intracellular level by quantitative RT-PCR. A significant level of downregulation (2- to 3-fold) of IFN-α expression was observed (Fig. 3B). IFI27 (also known as ISG12) is strongly induced by IFN-α. Our results demonstrated that IFI27 expression is inhibited in HCV genotype 1a- or 2a-infected IHH compared to mock-infected hepatocytes (Fig. 3C). Together, these results suggest that HCV infection of IHH impairs IFN-α and IFΙ27 synthesis.

To investigate whether HCV infection counteracts host innate immune responses at early time points after infection, the kinetics of IFN-α and IFI27 mRNA expression in IHH were examined beginning at 2 h after infection. An upregulation of IFN-α at early time points (2 h to 24 h) was observed, which decreased starting on day 2 (Fig. 3D). Interestingly, we did not observe induction of IFI27 mRNA expression following HCV infection. IFI27 expression upon IFN-α treatment of uninfected control IHH was measured and displayed >60-fold mRNA induction compared to the untreated control, suggesting that the IFI27 signaling pathway is intact in IHH. The kinetics of IFN-α and IFI27 induction in IHH infected with HCV genotype 2a were similar to those observed with HCV genotype 1a infection. It is interesting that IFI27 expression did not increase following HCV infection at 24 h, even though IFN-α was increased ∼6-fold. Indeed, further study is necessary to unravel the mechanism for inhibition of IFI27 expression in HCV-infected hepatocytes.

HCV infection inhibits poly(I-C)- or IFN-α-induced IRF-7 nuclear translocation in hepatocytes.

We asked whether poly(I-C) as an IFN inducer or IFN-α itself is able to override an HCV-specific impairment of IRF-7 translocation. For this analysis, mock-infected or HCV genotype 1a-infected IHH were transfected with IRF-7-GFP as described above. Cells were treated with poly(I-C) for 24 h or IFN-α for 4 h or 16 h. IRF-7 localization was examined by confocal microscopy. Hepatocytes treated with poly(I-C) or IFN-α displayed nuclear localization of IRF-7 (Fig. 4a and e). In contrast, IRF-7 was retained in the cytoplasm even after treatment with poly(I-C) or IFN-α (Fig. 4b to d and f to h) in HCV-infected IHH. Similar results were obtained when cells were infected with HCV genotype 2a and treated with poly(I-C). The results indicated an impaired effect of HCV infection upon translocation of IRF-7 in HCV-infected hepatocytes, and they raise questions regarding the efficient therapeutic efficacy of IFN-α in chronically infected patients. We further examined IFN-α expression in HCV-infected IHH treated with or without IFN-α (400 units) for 17 h, and we observed a similar level of mRNA expression (data not shown).

FIG. 4.

HCV infection inhibits poly(I-C)- or IFN-α-induced IRF-7 nuclear translocation in hepatocytes. IHH were infected with HCV genotype 1a at an MOI of 0.5 and transfected with IRF-7-GFP as discussed in the legend for Fig. 3. Hepatocytes were exposed to poly(I-C) for 24 h or IFN-α (400 U/ml) for 16 h. Cells were fixed and stained for NS5A (red) by using a specific antibody. Confocal microscopy suggested cytoplasmic localization of IRF-7 in HCV-infected cells (b, d, f, and h) even after treatment with poly(I-C) or IFN-α. On the other hand, mock-infected cells displayed nuclear localization of IRF-7 when exposed to poly(I-C) or IFN-α as a positive control (a and e).

HCV genome replication is enhanced in STAT1 knockdown IHH.

Primary induction of the type I interferon response results in synthesis of IFN-β, which in turn induces a large number of genes, including the effectors of the interferon response by autocrine and paracrine mechanisms. We examined whether inhibition of STAT1 modulated HCV genome replication. For this, IHH were transfected with small interfering RNA (siRNA) targeted to STAT1 (Santa Cruz Biotechnology) or with scrambled siRNA as a negative control. After verifying that total STAT1 protein content was downregulated (Fig. 5A), cells were infected with HCV at an MOI of ∼0.1. Intracellular HCV RNA content was determined, and our results suggested 10- to 18-fold higher STAT1 knockdown in IHH compared to cells transfected with the scrambled siRNA negative control (Fig. 5B). This result corroborates earlier reports indicate that when PKR was inhibited using siRNA, HCV replication was enhanced (5). Thus, the results suggested that the impairment of IFN signaling, especially an inhibition of STAT1, enhances HCV genome replication.

FIG. 5.

STAT1 downregulation enhances HCV genome replication. (A) IHH were transfected with STAT1 siRNA or control (scrambled) siRNA. Cells were lysed after 5 days posttransfection, and STAT1 protein was analyzed by Western blotting using specific antibody. (B) STAT1-downregulated cells were infected with HCV genotype 1a or genotype 2a at an MOI of 0.1. Total cellular RNA was extracted 72 h postinfection, and HCV RNA was analyzed by quantitative RT-PCR. The results shown are means from three independent experiments.

DISCUSSION

How HCV establishes chronic infection in humans is largely unknown. It is conceivable that HCV interferes with the IFN pathway at many different levels for establishment of persistent infection. However, quite paradoxically, hundreds of ISGs are induced in the liver of chimpanzees acutely or chronically infected with HCV (3, 22, 27). The results obtained with chimpanzees are difficult to extrapolate to humans, because there are important differences in the pathobiology of HCV infection between these species. Nevertheless, despite the activation of the endogenous IFN system, the virus is not cleared from chronically infected humans or chimpanzees. Further, the capacity of IFN-α production in HCV-infected patients is controversial, since both increased and reduced IFN-α production have been reported (8).

The mechanism controlling the IFN response in patients is likely to be complex and is not well understood. Our goal is to determine how HCV modulates the IFN signaling pathway in infected hepatocytes. We have previously reported that IFN-β expression is enhanced for up to 3 days following HCV infection (13). The upregulation of IFN-β in HCV-infected IHH was noted for even up to 10 days of the study period (data not shown). In this study, we investigated downstream molecules of the IFN-α pathways following HCV infection in IHH. Higher expression of STAT1 with partial nuclear localization in HCV-infected IHH, compared to mock-infected hepatocytes, was observed. However, phosphorylation of STAT1 was not observed in HCV-infected IHH. HCV core protein has been shown to inhibit IFN-α-induced STAT1 phosphorylation (4). HCV NS5A has also been implicated for suppression of STAT1 phosphorylation in hepatocytes (16). On the other hand, the HCV genome transfected in Huh-T7 cells reduces STAT1 protein expression (20). The difference in STAT1 expression in these studies may be due to the different cell lines used and/or single HCV protein expression. A recent study showed that nuclear unphosphorylated STAT1 that accumulates in response to IFNs maintains or increases the expression of a subset of IFN-induced genes independently of tyrosine-phosphorylated STAT1 (7). In response to IFNs, the phosphorylation of STAT1 can last for several hours, but unphosphorylated STAT1 newly synthesized in response to tyrosine-phosphorylated STAT1 persists for several days, raising the possibility that the increased concentration of unphosphorylated STAT1 might play an important role in IFN-dependent signaling. Interestingly, we observed that HCV-infected IHH display upregulation of STAT1 at day 3, 7, or 10 without detectable STAT1 phosphorylation, although perinuclear and occasional nuclear localization of STAT1 was observed at different time points.

As activation of the ISRE promoter in the presence of HCV has been noted, it is implied that STAT1 and STAT2 heterodimerization occurs in the presence of IRF-9. Activation of ISRE promoters leads to an increase of several ISGs. Our results showed an enhanced expression of 2′,5′-OAS1 and PKR in HCV-infected IHH, which are important antiviral proteins downstream of the IFN-α pathway. Clinical studies using HCV-infected human liver have reported an upregulation of OAS1 and PKR (22). Recently, upregulation of phospho-PKR was noted in HCV genotype 2a-infected Huh-7 cells (9, 14). We also observed an enhancement of several ISGs following 10 days of HCV infection, which is in agreement with the earlier report. It is possible that exogenous IFN-α enhances RNase L and/or PKR, resulting in inhibition of HCV genome replication. However, HCV infection may not produce enough RNase L and/or PKR to eliminate virus from infected cells. This result differs from other published reports, suggesting that HCV efficiently blocks double-stranded RNA signaling by NS3/4A-dependent or -independent mechanism (6, 18, 32). The difference may arise from poly(I-C) or Sendai virus and HCV-induced distant interferon signaling mechanisms.

IRF-7 undergoes phosphorylation when activated and translocates into the nucleus. IRF-7 amplifies the type I interferon response by inducing expression of IFN-α, which also acts in both autocrine and paracrine manners through the IFN-α/β receptor. We have observed that IRF-7 remains localized in the cytoplasm of HCV-infected IHH. Intracellular IFN-α or its downstream signaling molecule IFI27 was downregulated following HCV infection. However, an upregulation of IFN-α at early time points (up to 24 h) was observed. The initial burst of IFN expression may be for uncoating of the HCV genome and RNA replication. In fact, we previously observed an induction of IFN-β at early time points (13). During HCV infection, unphosphorylated STAT1 goes to the nucleus and activates the ISREs. However, this activation was not sufficient to trigger high enough antiviral responses to clear the virus, as HCV infection failed to translocate IRF-7 and inhibit IFN-α synthesis (Fig. 6). Therefore, it is possible that HCV impairs the IFN signaling pathway at multiple stages. (i) HCV NS3/4A cleaves IPS-1 and TRIF, resulting in inactivation of IRF-3 and inhibition of IFN-β expression in Huh-7 cells or derivatives (18, 30). IHH infected with HCV, however, displayed enhanced IFN-β expression (13). (ii) HCV infection fails to induce STAT1 phosphorylation. Although unphosphorylated STAT1 partially translocates into the nucleus and activates the ISREs, this activation may not be sufficient to trigger an efficient antiviral response. (iii) HCV infection impairs nuclear translocation of IRF-7 and inhibits IFN-α synthesis. Thus, our results demonstrate that HCV infection blocks IFN-α production, which may be a critical step for establishment of chronic infection.

FIG. 6.

Schematic showing HCV-mediated interference of IFN signaling pathways in hepatocytes.

In conclusion, there are several reasons it has been difficult to reach a consensus on the status of IFN-signaling molecules in HCV-infected cells. First, HCV is a weak inducer of IFN-α synthesis, so negative results are difficult to address. Second, the cell lines commonly used for HCV growth (Huh-7 or Huh-7.5) are defective in the Toll-like receptor 3 (TLR3) or RIG-I pathway (19, 30, 32). Therefore, it is difficult to correlate the results. We have used immortalized human hepatocytes, which appear to have intact TLR3 and RIG-I pathways (13). Results from our study collectively demonstrate that HCV perturbs STAT1 phosphorylation and inhibits IFN-α synthesis by retaining IRF-7 in the cytoplasm.