|Home | About | Journals | Submit | Contact Us | Français|
Hepatitis C virus (HCV) proteins are known to interfere at several levels with both innate and adaptive responses of the host. A key target in these effects is the interferon (IFN) signaling pathway. While the effects of nonstructural proteins are well established, the role of structural proteins remains controversial. We investigated the effect of HCV structural proteins on the expression of interferon regulatory factor 1 (IRF-1), a secondary transcription factor of the IFN system responsible for inducing several key antiviral and immunomodulatory genes. We found substantial inhibition of IRF-1 expression in cells expressing the entire HCV replicon. Suppression of IRF-1 synthesis was mainly mediated by the core structural protein and occurred at the transcriptional level. The core protein in turn exerted a transcriptional repression of several interferon-stimulated genes, targets of IRF-1, including interleukin-15 (IL-15), IL-12, and low-molecular-mass polypeptide 2. These data recapitulate in a unifying mechanism, i.e., repression of IRF-1 expression, many previously described pathogenetic effects of HCV core protein and suggest that HCV core-induced IRF-1 repression may play a pivotal role in establishing persistent infection by dampening an effective immune response.
Infection with hepatitis C virus (HCV) represents the major cause of liver disease, affecting more than 170 million individuals worldwide (26). After a subclinical phase, more than 80% of patients progress to persistent HCV infection, which is the leading cause of chronic liver disease associated with cirrhosis and hepatocellular carcinoma (13). The persistence of the virus in the majority of infected individuals is linked to the ability of HCV to evade and/or antagonize the host immune response at both the local and systemic levels. Accordingly, although hepatocytes are a major target of HCV infection, the virus can also replicate in immune cells, including effector cells (1, 11). In this respect, resistance to interferon (IFN) therapy is a hallmark of evolution in persistence, indicating that knocking down the antiviral and immunomodulating effects of IFN is a successful strategy for evading the host immune surveillance (21). The production and secretion of IFN type I is pivotal in inducing a global antiviral state through paracrine IFN production and the subsequent activation of interferon-stimulated genes (ISGs) within the infected cells and in the surrounding tissues (70). The role of IFN in HCV infection is thus crucial (21). Functional genomic analyses from cohorts of human subjects with chronic infection have shown that infection is associated with a gene expression profile marked by ISGs whose level of expression is related to different degrees of liver fibrosis and cirrhosis (67). Similarly, gene expression profiling has demonstrated that acute resolving infections in chimpanzees are associated with high levels of hepatic ISG expression (4).
The single-stranded RNA genome of HCV is translated into a polypeptide precursor of 3,010 amino acids (aa) that is cleaved by cellular and viral proteases into three structural proteins (core, E1, and E2), p7, and at least six nonstructural proteins (NS2 to NS5B) (39). Several HCV proteins have been shown to interfere with the IFN-induced intracellular signal transduction pathway, thereby inhibiting the induction of a number of effector proteins. The structural protein E2 (72) and the nonstructural protein NS5A (22, 23) modulate interferon responses by inhibiting the interferon-inducible double-stranded RNA-dependent protein kinase R (PKR). The NS3/4A protease functions as an antagonist of virus-induced interferon regulatory factor 3 (IRF-3) and NF-κB activation and IFN-β expression by blocking the Toll-like receptor 3 and retinoic acid-inducible gene I signaling pathways (17, 18, 37). This signaling block not only impairs IFN production in hepatocytes but also breaks the IFN amplification loop (8, 18), thereby inhibiting the expression of several ISGs, including those involved in the adaptive immune response (for a review, see reference 21).
HCV core protein (21 kDa) is the first protein to be produced upon virus infection and possesses multiple functions affecting both the virus and the host. In addition to forming the viral nucleocapsids, core protein affects a whole array of host cell functions, including apoptosis, cell transformation, signal transduction, and transcriptional regulation (reviewed in references 73 and 78). In addition to cytoplasmic and nuclear localization, the HCV core protein is also secreted from stably transfected cells and has been found in the bloodstream of infected individuals, where it may provide an infection-independent mechanism of immune dysregulation (31, 62). Core-induced immune dysregulation includes suppression of Th1 polarization, inhibition of IFN-γ-mediated cytotoxic T-lymphocyte (CTL) activation, and decreased interleukin-2 (IL-2) and IL-12 production (for a review, see reference 61).
The role of HCV core protein in modulating ISG expression is more controversial. Data from different groups indicate that core protein modulates the IFN-induced Jak/STAT signaling pathway but does not affect the activation of some ISGs. In contrast, other reports showed a downregulation of interferon-induced antiviral genes (3, 12, 54).
One of the ISGs, IRF-1, a member of the interferon regulatory factor family, was originally identified as a regulator of the IFN-α/β promoter but later recognized as also able to regulate several ISGs by binding the IFN-stimulated response element (ISRE), also known as IRF-E, in the IFN-α/β-stimulated gene promoters and thus amplify the IFN response (20, 60). Intensive functional analyses carried out on this transcription factor have revealed a remarkable functional diversity in the regulation of cellular responses through the modulation of different sets of genes, depending on the cell type, the state of the cell, and/or the nature of the stimuli (49, 70). IRF-1 is a regulator not only of cellular antiviral responses through the induction of antiviral ISGs, including 2-5A synthetase and PKR, but also of cellular apoptosis and transformation and immune responses through the regulation of a number of specific target genes. Studies in knockout mice have implicated IRF-1 in the regulation of various immune processes, such as T-cell selection and maturation, leukemogenic development, and autoimmunity. Impairment of CD8+ cell maturation, defective Th1 responses, exclusive Th2 differentiation, impaired macrophage production of IL-12, and maturation of NK cells have all been observed in immune cells from IRF-1−/− mice (for a review, see reference 70).
Recently, IRF-1 has been implicated in HCV infection as a negative regulator of HCV subgenomic replicon (30). Using different experimental approaches, including cells expressing the entire HCV genomic replicon or cells conditionally or transiently expressing only the structural proteins, we show here that the expression of viral proteins, in particular of the HCV core protein, results in IRF-1 repression, which is reversed in IFN-cured cells. IRF-1 expression was inhibited at the transcriptional level, and this inhibition resulted in the transcriptional repression of several ISGs, including genes involved in antiviral as well as immunomodulatory activities, such as 2-5A synthetase, PKR, IL-15, IL-12, and the low-molecular-mass polypeptide 2 (LMP2). Interestingly, these effects occurred both in hepatic cells and in T cells. Our data identify a new mechanism through which the HCV core protein contributes to HCV-induced persistence and point to IRF-1 as a unique target to boost a protective innate and adaptive immune response against HCV.
Jurkat cells were grown in RPMI 1640 medium containing 10% fetal calf serum (FCS) and antibiotics. The Huh-7 cell line carrying the Sfl HCV full-length replicon (genotype 1b) was obtained from R. Bartenschlager. The cell clones that stably replicate the HCV replicon were the 21-5, 22-6, and 20-1 clones, passaged as described elsewhere (19, 58). Cured HCV replicon cells were treated with 100 U IFN-α/ml for 14 days to eliminate self-replicating full-length HCV replicon RNA (71). Clearance of replicon RNA was confirmed by reverse transcription-PCR (RT-PCR) and by the loss of resistance to G418. The U2-OS human osteosarcoma-derived, tetracycline-regulated cell line UTHCNS3-43, inducibly expressing the HCV structural region, was obtained from D. Moradpour (52). Cells were maintained in complete Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FCS, antibiotics, and 1 μg/ml puromycin, 1 μg/ml tetracycline, and 500 μg/ml G418. To induce HCV protein expression, 1 × 105/ml cells were seeded in 100-mm dishes. After 24 h, the cells were washed five times over a period of 1 h with DMEM. Finally, cells were placed in DMEM supplemented with 2% FCS with or without 1 μg/ml tetracycline.
Where indicated, cells were treated with 100 ng/ml of recombinant IFN-γ (rIFN-γ; PeproTech EC Ltd., London, United Kingdom) or 100 ng/ml tumor necrosis factor alpha (TNF-α; PeproTech) for 16 h.
Mammalian expression vectors containing the core (aa 1 to 191) and the E1E2p7 (aa 155 to 809) regions of the HCV genome obtained by RT-PCR from HCV-positive human serum were cloned into the pRc/CMV mammalian expression vector. The nucleotide sequence of the clones was determined and classified the HCV isolate as type 1a (9). In the E1E2p7 construct, the signal sequence of E1 protein (aa 155 to 191) was included to allow proteins to be transported into the endoplasmic reticulum membrane. The sequence of each recombinant plasmid was confirmed by the modified dideoxynucleotide method and an ABI 373A automatic sequencer. The IRF-1 expression vector was a generous gift of J. Hiscott and is described in reference 38.
Transient-transfection experiments were performed using the FuGENE 6 transfection reagent (Roche Diagnostics, Mannheim, Germany) for Jurkat cells and MBS mammalian transfection kit (Stratagene, La Jolla, CA) for Huh-7 and Vero cells, according to the manufacturers' protocols. The construct encoding a portion of the wild-type LMP2 promoter and a mutated version in the IRF-E consensus sequence (LMP2 mt) cloned upstream of the luciferase reporter gene were a generous gift of K. L. Wright and are described in reference 79. The constructs encoding the IFN-β, β-casein, and IL-4 promoters upstream of the luciferase reporter gene were generous gifts of J. Hiscott (McGill University, Montreal, Canada), T. Kitamura (University of Tokyo, Tokyo, Japan), and M. Li-Weber (Tumorimmunology Program, German Cancer Research Center, Heidelberg, Germany), respectively, and are described in references 16, 38, and 55.
The constructs for p3500 (encoding the entire IRF-1 promoter from bp −3400 to +168) and the two portions Gas/κB (corresponding to the −199/−89 region of the IRF-1 promoter) and NF-κB (corresponding to the −89/−16 region of the IRF-1 promoter), cloned upstream of the luciferase reporter gene, were a generous gift of R. Pine and are described in reference 59. The IL-12 p40 construct (corresponding to the −413/+12 region of the IL-12 p40 promoter) cloned upstream of the luciferase reporter gene was a generous gift of K. Ozato and is described in reference 44.
Huh-7 (5 × 105), Huh-21-5 (5 × 105), Jurkat (2 × 106), and Vero (1 × 106) cells were transfected with 500 ng of luciferase reporter vectors and 500 ng of expression vectors. The amount of transfected DNA was adjusted with the empty vector RcCMV. One hundred ng of pAct-Renilla plasmid was cotransfected and used as a control for transfection efficiency. Reagents from Promega were used to assay extracts for luciferase activity in a Lumat LB9501 luminometer (E&G Berthold, Bad Wildbad, Germany).
For NS5B and core analysis, cells were collected by centrifugation at 3,000 × g for 4 min at 4°C, washed twice in 1× phosphate-buffered saline, and resuspended in a volume of lysis buffer (12.5 mM Tris-HCl [pH 6.7], 2% glycerol, 0.4% sodium dodecyl sulfate [SDS], 1% mercaptoethanol, 0.1% bromophenol blue). For IRF-1 analysis, cells were washed twice in ice-cold phosphate-buffered saline and lysed in cold lysis buffer (20 mM HEPES pH 7.4, 50 mM NaCl, 10 mM EDTA pH 8.0, 2 mM EGTA, 0.5% NP-40, 0.5 mM dithiothreitol, 10 mM NaMO, 10 mM NaVO3, 100 mM NaF, 0.05 M β-glycerophosphate, 100 μg/ml leupeptin, 0.5 mM phenylmethylsulfonyl fluoride) for 20 min at 4°C. Lysates were centrifuged at 10,000 × g for 10 min at 4°C. Fifty μg of cell extracts was separated on 10% or 12% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a nitrocellulose membrane. Blots were incubated with anti-IRF-1 (1:200; sc-497; Santa Cruz Biotechnology Inc., Santa Cruz, CA), anti-IRF-2 (1:200; sc-498; Santa Cruz), anti-IRF-3 (1:200; sc-9082; Santa Cruz), anti-PKR (1:200; sc-707; Santa Cruz), anti-NS5B (1:200; sc-17532; Santa Cruz), anti-core (1:1,000; MO-I40015B; Anogen, Mississauga, Ontario, Canada), anti-E1 and anti-E2 (1:500; C65198M and B65581G; BIODESIGN Int., Maine), and anti-actin (1:200; sc-1616; Santa Cruz) antibodies and then with anti-rabbit-, anti-mouse-, or anti-goat-horseradish peroxidase-coupled secondary antibody (Calbiochem, San Diego, CA). Immune complexes were identified using an enhanced chemiluminescence system (Amersham, Buckinghamshire, United Kingdom).
Total RNA was extracted from 1 × 106 cells using RNeasy kits (QIAGEN) as described by the manufacturer and quantified by optical density. One hundred nanograms of total RNA was reverse transcribed using the high-capacity cDNA archive kit (Applied Biosystems) and random hexamer primers in an ABI Prism 7000 sequence detector system (Applied Biosystems) using the following thermal profile: 25°C for 10 min, 42°C for 1 h, and 95°C for 5 min. PCRs were performed in triplicate on the ABI Prism 7000 sequence detector system (Applied Biosystems) using TaqMan chemistry with primer and probe sets from the Assay-on-Demand list (Applied Biosystems). For each gene, the standard curve was compared with the standard curve of the reference gene and calculation of the slope of the log(ng RNA) versus ΔCt was always <0.1. Fold induction was then calculated by the ΔΔCt method (39a) using the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA level to normalize values and the mRNA level of the Huh-7 parental cell line or of the U2-OS (Tet+) cells as a calibrator.
Expression of the full-length HCV poly-protein in the context of replicating RNA provides a useful system to evaluate better some aspects of HCV-cell interactions. As it has been reported that IRF-1 is not directly affected by nonstructural proteins but negatively regulates HCV replication (30), we investigated IRF-1 regulation in cell lines harboring the autonomously replicating full-length HCV genome, specifically, the 21-5, 22-6, and 20-1 clones (19, 58).
IRF-1 mRNA expression was monitored by real time RT-PCR analysis in naïve Huh-7 cells and in the clones. This analysis (Fig. (Fig.1A)1A) showed that IRF-1 levels were significantly and comparably decreased (between 50% and 60% [± 8%, standard deviation]) in all the clones examined compared with the Huh-7 parental cell line. For further analyses we chose the 21-5 clone, in which IRF-1 mRNA expression was 46% compared with the Huh-7 parental cell line (Fig. (Fig.1B).1B). The IRF-1 mRNA decrease in Huh-21-5 cells resulted in an almost complete reduction in IRF-1 protein expression as assessed by Western blot analysis (Fig. (Fig.1C,1C, lane 2).
To determine whether the inhibition of IRF-1 expression was specifically mediated by negative regulatory effects of HCV protein expression, we cured Huh-21-5 cells by culturing them in the presence of 100 IU/ml of natural IFN-α for 2 weeks, a treatment which has been shown to clear the virus (71). As shown in Fig. 1B and C, in cured Huh-21-5 cells the levels of IRF-1 were comparable to those present in the Huh-7 parental cell line. In Fig. Fig.1D1D the expression of NS5B confirms the expression of viral proteins in Huh-21-5 cells but not in cells treated with IFN-α.
These results suggest that HCV replication and viral protein expression specifically downregulate IRF-1 expression in Huh-21-5 cells.
It has been reported that in cells harboring the subgenomic replicon the double-stranded RNA-stimulated IRF-1 is specifically inhibited by NS5A expression (57). In order to distinguish between the effects of nonstructural and structural proteins, we first expressed the structural proteins of HCV in a tetracycline-regulated system (52). As shown in Fig. Fig.2B,2B, when we performed Western blotting with an anti-core antibody, only cells cultured in the absence of tetracycline expressed viral proteins on a kinetic base, whereas in the presence of tetracycline, viral protein synthesis was inhibited. To analyze the kinetics of gene regulation in cells expressing the HCV structural proteins, we performed a time course experiment. Cells that inducibly expressed viral proteins were cultured for 12, 24, and 48 h in the presence or in the absence of tetracycline, and IRF-1 expression was monitored by real-time PCR and Western blotting (Fig. 2A and B). Real-time PCR analysis indicated that at time zero there were no significant differences between levels of IRF-1 in cells expressing the structural proteins and in control cells (data not shown). Similarly, no variations in the IRF-1 mRNA amount were found in cells growing in the presence of tetracycline and not expressing viral proteins from 12 h onward (Fig. (Fig.2A).2A). Conversely, withdrawal of the antibiotic from the medium resulted in the expression of HCV proteins and in a significant decrease in the IRF-1 mRNA levels at all time points. The decrease was maximum after 48 h of culture, when IRF-1 mRNA levels were 50% ± 10% of the control (Fig. (Fig.2A).2A). Western blot analysis (Fig. (Fig.2B)2B) indicated that the inhibition of IRF-1 protein expression was even more marked, and the protein was barely detectable after 48 h. Interestingly, at this time synthesis of the core protein in this system reached its maximum.
These results indicate that the downregulation of IRF-1 expression observed in cells expressing the entire HCV replicon (Fig. (Fig.1)1) could be due mainly to the expression of HCV structural proteins.
To identify the HCV-encoded structural protein(s) that influences IRF-1 gene expression, we transiently expressed individual fragments from the structural region of the HCV genome coding for the core or E1 or E2 proteins in Huh-7 hepatoma cells. We decided to use transient expression, even though the effects may be underestimated if only a small fraction of the cell population is transfected, because stable transfection, especially in the case of the core protein, may be susceptible to clonal selection. The ability of HCV-expressing plasmids to code for the core and for the E1-E2 proteins was confirmed by Western blot analysis (Fig. (Fig.3B).3B). After transfection, total mRNA was isolated, and the amount of IRF-1 mRNA was evaluated by real-time PCR. As shown in Fig. Fig.3A,3A, the IRF-1 mRNA level was substantially lower (more than 60% ± 10%) in cells transfected with the plasmid expressing the HCV core protein than in cells transfected with the empty vector. In contrast, cells expressing the envelope proteins (E1 and E2) showed an increase in the amount of IRF-1 mRNA, although this increase was not statistically significant. Analysis of IRF-1 protein expression by Western blotting (Fig. (Fig.3B)3B) indicated that IRF-1 expression was reduced to 30% in cells expressing the core protein but was not affected by E1-E2 expression (lane 3). To assess the specificity of the inhibitory effect of the core protein on IRF-1 expression, dose-response experiments were performed. As shown in Fig. Fig.3C,3C, increasing doses of HCV core protein resulted in a correspondingly dose-dependent inhibition of IRF-1 protein expression, whereas expression of other IRFs, e.g., IRF-2 and IRF-3, was not affected. The core protein was detectable by Western blot analysis starting from a dose of 1 to 5 μg, although doses as low as 0.2 μg were equally able to decrease IRF-1 expression.
Taken together, these results indicate that among structural HCV proteins only the core is able to down-regulate significantly IRF-1 expression in HCV-infected cells.
IRF-1 expression is mainly regulated at the transcriptional level by several inducers (70). To determine whether the HCV core protein was able to inhibit IRF-1 promoter activity, Huh-7 cells were transiently transfected with a 3,500-bp IRF-1 promoter construct upstream of the luciferase reporter gene in the presence of an expression vector coding for the HCV core protein. As shown in Fig. Fig.4A,4A, basal IRF-1 promoter activity was substantially lower (80% ± 5%) in cells expressing the core protein than in cells expressing the empty vector (lane 2 versus lane 1). Moreover, since it is known that inflammatory cytokines such as IFN-γ and TNF-α substantially induce the expression of IRF-1 (59), we checked the ability of HCV core protein to affect cytokine-induced IRF-1 transcription. Interestingly, as shown in Fig. Fig.4A4A (lanes 3 to 6), the treatment with IFN-γ and TNF-α did not counteract the inhibitory effect exerted by the core protein, suggesting that cells expressing the core protein do not respond to inflammatory stimuli by inducing IRF-1. As shown in Fig. Fig.4B,4B, HCV core protein expression also impacted the basal and cytokine-induced IRF-1 transcriptional activity in cells of the immune system, specifically Jurkat T cells. This may be of physiological importance, since cells of the immune system not only represent another target of HCV infection (11) but can also be affected in a bystander manner by core protein circulating in the bloodstream (31).
To determine which sequences on the IRF-1 promoter were targets of the inhibitory effect of the HCV core protein, Huh-7 and Jurkat cells were transfected with portions of the IRF-1 promoter containing the consensus binding sites for STAT1 (GAS/NF-κB) or NF-κB transcription factors together with an expression vector coding for the core protein. Eight hours after transfection, cells were mock treated or stimulated with IFN-γ or TNF-α for a further 16 h, after which the promoter transcriptional activity was measured. As shown in Fig. 5A to D, both basal and cytokine-induced activation of constructs bearing the GAS or the NF-κB elements, respectively, was significantly reduced in Huh-7 and Jurkat cells expressing the HCV core protein.
Finally, IRF-1 expression following core expression was determined in IFN-deficient Vero cells to exclude an involvement of IFN blockade in the core-mediated IRF-1 repression. As shown in Fig. Fig.5E,5E, the amount of IRF-1 mRNA, evaluated by real-time PCR, was reduced in core-expressing Vero cells at levels comparable to those observed in core-expressing Huh-7 cells (45% ± 8%). Accordingly, basal IRF-1 promoter activity was significantly repressed in Vero cells expressing the core protein compared to cells transfected with the empty vector (Fig. (Fig.5F5F).
To determine whether the regulation of IRF-1 by HCV affected host cell gene expression during HCV RNA replication, we examined the expression of different ISGs, targets of IRF-1 and endowed with both antiviral and immunomodulatory activities, in cells either expressing the entire replicon (Fig. (Fig.6A)6A) or conditionally expressing only the structural proteins (Fig. (Fig.6B).6B). The ISGs examined were PKR and 2-5A synthetase, IL-15, IL-12, and IFN-β. Real-time PCR analysis indicated that expression of the entire replicon (Fig. (Fig.6A)6A) resulted in a significant decrease in mRNA accumulation in all the genes examined. The expression of structural proteins (Fig. (Fig.6B)6B) determined a comparable inhibition of PKR, IL-15, and IL-12 but not of IFN-β. Interestingly, data in the literature report that the signaling pathway leading to IFN type I production is mainly affected by the HCV nonstructural protein (21). We further determined the levels of PKR protein expression by Western blotting (Fig. (Fig.6C),6C), confirming the data obtained by RT-PCR analysis.
The downstream effect of IRF-1 inhibition in cells harboring the entire replicon or transiently expressing increasing amounts of the HCV core protein was also demonstrated by determining the transcriptional activity of reporter constructs bearing the IRF-E/ISRE sequences present on the LMP2 and IL-12 gene promoters, both of which are dependent on IRF-1 for transcription. As shown in Fig. Fig.7A,7A, the relative IRF-E-regulated luciferase activity was significantly repressed (80% ± 10%) in Huh-21-5 cells harboring the HCV replicon compared with naïve Huh-7 cells. Notably, increasing doses of the core protein (Fig. 7B and C) cotransfected with the same construct caused a dramatic dose-dependent inhibition of the wild-type IRF-E luciferase reporter construct activity in both T and hepatoma cells. Indeed, doses of 1 μg of the HCV core protein-expressing vector practically abolished the promoter transcriptional activity. Conversely, basal expression of the LMP2 promoter construct mutated in the IRF-E element (Fig. (Fig.7B,7B, mt) was lower than in the wild-type construct, but its transcriptional activity was not affected by the expression of the HCV core protein. The inhibitory effect of the core protein was directly mediated by IRF-1 repression, as demonstrated by restoring IRF-1 expression in Huh-7 cells expressing the HCV core protein. As shown in Fig. Fig.7D,7D, repression of the IRF-E-regulated luciferase activity was completely reversed by overexpression of IRF-1 in a dose-dependent fashion. To demonstrate further the specificity of the core inhibition, the activity of other promoter constructs was determined in Huh-7 cells transfected with 1 μg of the core protein. As shown in Fig. Fig.7E,7E, neither the transcriptional activity of β-casein nor that of the IL-4 gene promoter was affected by the core protein expression. Similarly, and in accord with the RT-PCR data shown in Fig. Fig.6B,6B, the IFN-β proximal promoter basal activity was maintained irrespective of the expression of the HCV core protein.
It is known that IRF-1 binds not only to IRF-E sequences but also to the overlapping ISRE, regulating the expression of several ISGs (69, 70). We therefore investigated whether the promoter activity of an ISRE-containing promoter was similarly affected by HCV core protein expression. To this end we cotransfected in Huh-7 and Jurkat cells a luciferase reporter construct under the control of the ISRE present on the IL-12 p40 gene promoter. As shown in Fig. 8A and B, a dose of 0.5 μg of a core-expressing vector significantly reduced IL-12 promoter activity in both cell types (50 to 70% [± 10%] compared with control cells).
Together, these results suggest that the block of IRF-1 expression in HCV genomic replicon-bearing cells was mainly mediated by the core protein and was sufficient to repress the expression of IRF-1-dependent genes during HCV replication.
Most HCV-infected individuals are unable to clear the virus and develop a chronic infection with inauspicious outcome. The mechanisms by which HCV resists the host antiviral defenses and induces liver injury remain poorly defined. Nevertheless, it is well established that HCV induces IFN, and the ability to circumvent the antiviral effects of this cytokine seems to be mainly responsible for the establishment of persistence. Accordingly, only 10% to 20% of patients with chronic infection respond to IFN-α therapy (51). Thus, clarification of the mechanisms involved in impaired anti-HCV actions by IFN-α is an important goal for the definition of effective therapies.
In the present study we have demonstrated that HCV core gene expression, either alone or in the context of HCV replication, inhibits IRF-1, a transcription factor involved in IFN signaling. Furthermore, IRF-1 repression attenuates ISG responses mediated by the IRF-E/ISRE. Specifically, we showed that IRF-1 target genes, endowed with both antiviral and immunomodulatory functions, are repressed and IRF-1 suppression occurs at the transcriptional level through inhibition of both basal and cytokine-induced IRF-1 promoter activity.
IRF-1 is a pleiotropic transcription factor, critical for cell defense against viral infections but also crucial for the development of both the innate and adaptive responses. Indeed, the targets of IRF-1 during the antiviral response are genes directly involved not only in virus elimination but also in differentiation, proliferation activity, and apoptosis of cells, including those of the immune system (42, 70). Accordingly, dysregulation of IRF-1 expression can lead to defective antigen presentation, NK and T-cell activity disorders, and tumorigenesis.
The different levels of IRF-1 expression seem to be one of the factors that dictate how it functions. Expression of the IRF-1 gene is rapidly induced following virus infection, and IFNs, proinflammatory cytokines, and high IRF-1 expression levels are required for the induction of a set of ISGs necessary for an efficient antiviral response (24, 33). Here we have shown that HCV RNA replication affects IRF-1 expression, down-modulating both mRNA and protein at levels no more sufficient to induce the expression of ISGs direct targets of IRF-1.
The inhibition of IRF-1 expression and IRF-E/ISRE activity in cells expressing a subgenomic replicon of HCV was reported by Kanazawa and colleagues (30). However, in the system analyzed, no direct correlation was found between the viral RNA replication and IRF-1 repression.
The development of in vitro culture systems for expression of the entire replicon of HCV greatly improved our understanding of the complex virus-host interactions and recapitulated most of the previous observations of the impact of single HCV proteins on cell physiology. In contrast to observations in HCV subgenome-expressing cells, we found that in cured Huh-7 cells from which the entire replicon has been eliminated, IRF-1 expression is restored (Fig. (Fig.1),1), and we postulate that the decrease in IRF-1 expression could be due to active suppression by viral proteins. Indeed, using cells that transiently or conditionally express HCV structural proteins, we have demonstrated a specific repression of IRF-1 by the core protein.
Several in vitro studies have already defined the role of some viral structural and nonstructural proteins in inhibiting the expression and/or activity of key factors in IFN synthesis and in the IFN-induced signal transduction pathway (21). Conversely, to date conflicting results have been reported on the role of HCV core protein. Whereas recent papers agree on the interference with the Jak/STAT pathway by inhibition of ISGF3 and STAT-1 activities, there is no agreement on whether this interference results in a modification of transcription of ISGs (3, 6, 7, 12, 47, 48, 50, 54). Variations in the experimental system used, structural or genotypic differences in the protein and, possibly, the amount of protein used may all be implicated in the reported discrepancies. Using conditional and transient expression of HCV structural proteins, we have shown that the core protein was mainly responsible for the inhibition of IRF-1 expression in cells bearing the entire HCV replicon. Conversely, and in agreement with previous reports (57), other structural proteins, E1 and E2, did not impact IRF-1 expression (Fig. (Fig.22).
We have also shown that the inhibition of IRF-1 expression occurred at the transcriptional level through both the STAT-1 and NF-κB consensus sequences on the IRF-1 promoter. In agreement with these data it has recently been reported that core protein inhibits both the activation and nuclear translocation of STAT-1 in IFN-treated cells (7, 47). In addition, in core-expressing cells NF-κB nuclear translocation and activity is also suppressed (29). Data obtained in Vero cells, which are deficient in IFN production, seem also to exclude the possibility that the core-inhibited IRF-1 expression resulted from the blockade of the endogenous IFN in core-expressing cells. Given that no reports have, as yet, clarified the mechanism through which the core regulates gene transcription, our data support the conclusion that the core regulates gene transcription by means of indirect effects eventually leading to modulation of signal transduction and transcription factors.
Notably, we have also shown that the down-modulation of IRF-1 expression by the HCV replicon and by the core protein in turn affects host gene expression mediated by IRF-E/ISRE. Among the repressed genes, we found genes with antiviral functions, including 2-5A synthetase and PKR, but also immunomodulatory cytokines, such as IL-12 and IL-15, and genes that are important for antigen presentation, such as LMP2. These results suggest that inhibition of IRF-1 by the core in the context of HCV replication may trigger a cascade of events that affect a wide range of immune responses via direct or indirect mechanisms.
Numerous studies have shown that the HCV core protein is a major cause of most of the pathogenic features associated with HCV infection, and in this respect its role in the suppression of immunological functions has been established. In particular, impairment of the activation and functions of dendritic cells and of the proliferation of T cells has been reported (25, 34). In accordance with our observations, a selective suppression of IL-12 in stimulated human macrophages mediated by the suppression of AP1 has also recently been described (15, 36). Our data also indicate that the HCV core protein affects IL-12 p40 gene promoter activity in both hepatoma and T cells. In this respect it has been demonstrated that induction of the gene encoding both the p40 and p35 subunits of IL-12 is totally dependent on IRF-1 (41, 43, 68). Therefore, our results imply that in addition to the reported mechanism of IL-12 repression in macrophages mediated by AP-1 (15), the inhibition of IRF-1 is probably most responsible for IL-12 suppression by the HCV core protein.
IL-12 is known to play a pivotal role in the generation of Th1 immune responses and provides a crucial link between innate and adaptive immunity (74). IRF-1 has been defined as a “super” Th1-cell transcription factor (42) in that IRF-1−/− mice display completely defective Th1 responses due to a lack of IL-12 production by APC, accompanied by exclusive Th2 differentiation. In this respect, it is interesting that the study of IRF-1 promoter polymorphisms in relation to the response to IFN in chronic patients indicated that a lower viral load corresponded to higher IRF-1 promoter activity and to a significantly higher proportion of Th1 CD4+ cells after IFN administration (63). Interestingly, the establishment of a Th1 environment in HCV-infected individuals is thought to be important in determining the outcome of the infection. As indicated by in vivo studies, patients who demonstrate Th2 dominance tend to develop chronic infection, whereas those with a Th1 phenotype can clear the virus (2, 10, 32, 75, 76, 81). The immunosuppressive action of core protein may thus be pivotal in viral escape by down-modulating Th1 responses in favor of Th2 responses, a mechanism in which core-mediated inhibition of IRF-1 may be instrumental.
Of the ISGs analyzed, IL-15, a unique target of IRF-1, was also substantially reduced in both hepatic and T cells expressing either the entire HCV replicon or the structural proteins, and to our knowledge this is the first study reporting a repression of IL-15 in cells bearing the entire HCV replicon. IL-15 is involved in the activation and homeostatic maintenance of cells of both the innate and adaptive immune systems, playing a key role in CD8+ T-cell homeostasis by promoting survival or proliferation of naive and memory phenotype CD8 T cells and in NK cell development, maturation, and survival (14, 40, 80). Notably, it has been reported that impaired IL-15 production is one of the mechanisms of the aberrant response of DC to IFN in HCV-infected patients (28). In addition, a significant reduction in serum IL-15 levels in HCV patients has also been recently reported (46).
We have also shown that the HCV core, through inhibition of IRF-1, specifically inhibits, at the transcriptional level, the LMP2 gene promoter activity (Fig. 7B to D). LMP2 is a subunit of immunoproteasomes, which are very efficient for the generation of specific CTL epitopes. It has, in fact, been shown that substitution of standard β-subunits of the proteasome with LMP2, LMP7, and MECL1 subunits improves the production of peptide antigens with the correct C termini for binding to MHC class I (53, 64, 66, 77). Since both basal and IFN-γ-induced LMP2 expression is absolutely dependent on IRF-1, it is not surprising that inhibition of IRF-1 by the core protein also results in suppression of LMP2 expression. Considering that a reduction in the population and specific activity of CTL (2, 32, 35) have been observed in chronically infected individuals, we speculate that the HCV core protein down-modulation of LMP2 may be partially responsible for these in vivo observations. An analysis of the proteasome composition in HCV patients would, therefore, be very useful.
Data reported in the present study suggest that HCV regulation of IRF-1 may also impact other IRF-regulated pathways influencing host gene expression on a more global scale. In this respect, given that IRF-1 together with NF-κB is induced and is also an effector of inflammatory cytokines, it is interesting that the HCV core protein inhibits IRF-1 promoter activity even after treatment with inflammatory cytokines (Fig. (Fig.4).4). This raises the possibility that both locally, in hepatoma cells, and systemically, by affecting PBMC, the core protein can depress the inflammatory response induced by virus infection also through repression of IRF-1. In agreement with this hypothesis, a recent report (27) indicated that cyclooxygenase 2, an enzyme that contributes to homeostasis and to inflammatory pathways (56), is substantially repressed in core-expressing cells. Notably, cyclooxygenase 2 is another gene tightly regulated by IRF-1 (5).
Finally, it has been reported that other cellular and viral genes altered by the core include p21waf1, p53, and human immunodeficiency virus type 1 LTR (45, 73). As others and we ourselves have reported (65, 70), these genes are all specific targets of IRF-1.
Taking all these results into account and considering that they well mirror clinical observations in chronically infected patients, we conclude that IRF-1 repression by the HCV core protein can be considered a unifying mechanism that recapitulates most of the data so far reported on the dysregulation of cellular processes induced by this protein and may at least partially account for its role in evading the host response at several levels: antiviral, inflammatory, and immune. Restoration of the correct expression of IRF-1 in HCV-infected cells could, therefore, represent a new avenue for therapeutic interventions.
We thank R. Bartenschlager for the Huh-7, Huh-21-5, Huh-22-6, and Huh-20-1 cell lines, D. Moradpour for the UTHCNS3-43 cell line, and J. Hiscott, K. L. Wright, T. Kitamura, and M. Li-Weber for providing luciferase constructs. We thank Sabrina Tocchio for editorial assistance and Roberto Gilardi for preparing graphs.
This work was supported by grants from the Italian AIDS Project, the Italian Ministry of Health, and ISS-NIH Scientific Cooperation agreement to A.B. and by grants from the “Integrated National Project for the Study, the Prevention and the Treatment of the Chronic Hepatology” (RF02.188), and the “Viral Hepatitis National Projects” of the Istituto Superiore di Sanità (D. leg.vo no. 502) to M.R.
Published ahead of print on 18 October 2006.