New Antiviral Pathway That Mediates Hepatitis C Vi
http://www.100md.com
病菌学杂志 2005年第10期
Laboratory of Hepatitis Viruses, Center for Biologics Evaluation and Research, Food and Drug Administration, Bethesda, Maryland 20892
ABSTRACT
While many clinical hepatitis C virus (HCV) infections are resistant to alpha interferon (IFN-) therapy, subgenomic in vitro self-replicating HCV RNAs (HCV replicons) are characterized by marked IFN- sensitivity. IFN- treatment of replicon-containing cells results in a rapid loss of viral RNA via translation inhibition through double-stranded RNA-activated protein kinase (PKR) and also through a new pathway involving RNA editing by an adenosine deaminase that acts on double-stranded RNA (ADAR1). More than 200 genes are induced by IFN-, and yet only a few are attributed with an antiviral role. We show that inhibition of both PKR and ADAR1 by the addition of adenovirus-associated RNA stimulates replicon expression and reduces the amount of inosine recovered from RNA in replicon cells. Small inhibitory RNA, specific for ADAR1, stimulated the replicon 40-fold, indicating that ADAR1 has a role in limiting replication of the viral RNA. This is the first report of ADAR's involvement in a potent antiviral pathway and its action to specifically eliminate HCV RNA through adenosine to inosine editing. These results may explain successful HCV replicon clearance by IFN- in vitro and may provide a promising new therapeutic strategy for HCV as well as other viral infections.
INTRODUCTION
Hepatitis C virus (HCV) infects approximately 170 million individuals worldwide and nearly 3 million in the United States alone. Most cases of HCV infection become persistent and may result in chronic liver disease, cirrhosis, and hepatocellular carcinoma. The current combination antiviral therapy of pegylated alpha interferon (IFN-) with ribavirin is effective in approximately 50% of individuals treated, while monotherapy with IFN- alone is successful in less than 20% of patients (13). IFN- allows cells to become innately primed for defense against eventual virus attack by inducing the transcription of many genes, some of which are activated during virus infection. Only a few genes have been identified and characterized as mediators of the IFN--induced antiviral response, including the Mx proteins, major histocompatibility complex proteins, 2',5' oligoadenylate synthetase, and the double-stranded RNA (dsRNA)-activated protein kinase (PKR) (21). PKR is activated during viral infection, which results in the phosphorylation of the subunit of eukaryotic translation initiation factor 2 (eIF-2) and subsequent translational shutoff. An adenosine deaminase that acts on dsRNA (ADAR1) is also IFN- induced and catalyzes the deamination of adenosine residues in dsRNA (for a review, see reference 17), resulting in inosine substitution. Inosine residues are not abundantly found in cellular mRNAs, but when inosine residues are present, they are transcribed and translated as guanosine residues, which may lead to mutations (1, 24). An RNase that specifically degrades inosine-containing RNA has been described and was proposed to be part of a putative antiviral pathway (18, 19). Although antiviral activity has not been attributed to ADAR1, hepatitis delta virus utilizes ADAR1 editing to promote its viral life cycle (15).
Typically, dsRNA is found only in cells that are virus infected, and both DNA and RNA viruses may present dsRNA in the cell in the form of replicative intermediates (12). ADAR1 contains three copies of a conserved dsRNA-binding motif also found in PKR (25). Most viruses have developed strategies to evade the effects of IFN-. For instance, a single-stranded virus-encoded RNA with partially dsRNA features, adenovirus-associated (VA) RNAI, enhances translation and counteracts the effects of IFN- in adenovirus-infected cells by inhibiting PKR (14). VA RNA has also been shown to bind and inhibit ADAR1 (10). This is the first report implicating RNA editing by ADAR1 in the control of viral replication and may provide a potent strategy for an effective treatment against HCV based on ADAR1 activation.
MATERIALS AND METHODS
Expression vectors and HCV replicon. The BB7 replicon was a gift of C. M. Rice and K. Blight and was previously described (2). Briefly, the replicon expresses HCV NS3 through NS5B nonstructural genes under the encephalomyocarditis virus (EMCV) internal ribosome entry site (IRES) and neomycin resistance under the HCV IRES. The HCV 5' and 3' nontranslated regions are also present. Bicistronic reporter plasmids were constructed by substitution of the simian virus 40 promoter in pRL-SV40 (Promega) with the herpes simplex virus type 1 alpha 27 promoter (gift of N. Martin). The mRNA expresses Renilla luciferase (Luc) through a cap-dependent translation mechanism and firefly Luc under the EMCV IRES. PCR of the BB7 plasmid (using primers that contained a SalI restriction site at the 5' end and a SacI restriction site at the 3' end, covering nucleotides 1 to 386 in pHCVrep1bBB7) (16) was used to generate an HCV IRES cassette. The EMCV IRES was removed by digestion with SalI and SacI, and the HCV-IRES was inserted. pcDNA3 (Invitrogen) plasmids expressing wild-type (WT) PKR, PKR K296R, E2, and NS5A were described previously (20). eIF-2-pRc/CMV (eIF-2 S51A) was a gift of O. Donze and was described previously (4). VA RNAI-pUC119, expressed under the polymerase III intragenic promoter was described previously (7).
Cell culture. Huh7 and Huh.BB7 cells were grown at 37°C in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum, 2 mM L-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin. Huh7 cells were transfected with RNA transcribed from linearized pHCVrep1bBB7 plasmid. Stable transfectants were selected in Geneticin (G418; Gibco) and grown in G418 at 75 μg/ml, which was removed during experiments. Cellular proteins were metabolically labeled in cell culture by incubating the cells for 1 h at 37°C in DMEM minus methionine and then supplementing the medium for 1 h with [35S]methionine. RNA was metabolically labeled in cell culture by incubating the cells for 1 h at 37°C in DMEM minus phosphates and then incubating the cells for 16 h in medium supplemented with [-32P]ATP. Cell growth was monitored by counting the number of viable cells with trypan blue staining or with a Coulter Counter (Beckman Coulter, Inc.). Cells were treated with recombinant human IFN-2A (100 IU/ml in DMEM) for 18 h at 37°C, unless stated otherwise.
Analyses of protein and RNA. Cells were washed with phosphate-buffered saline (PBS), and the attached monolayer was incubated with trypsin EDTA for 5 minutes at 37°C. Cells were collected, concentrated in 0.1% NP-40 and 10% PBS, and lysed by three repeated freeze-thaw cycles. Nuclear and cellular membranes were removed by centrifugation, and cytoplasmic extracts were quantitated for protein concentration with a colorimetric absorbance protein assay (Bio-Rad). Expression of HCV NS3 protein in lysates from Huh7 or Huh.BB7 cells was monitored by resolving 25 μg of protein extract in sodium dodecyl sulfate-polyacrylamide gels and immunoblotting with polyclonal antibody to NS3 (chimp 1536 serum [16]). Polyclonal anti-PKR antiserum was made by inoculating New Zealand White rabbits with a keyhole limpet hemocyanin-conjugated peptide to the spacer region between the two dsRNA-binding motifs of PKR (peptide P1) (22). E2 expression was monitored with mouse monoclonal antibody A11 using cell extracts that were treated with endoglycosidase H. Actin expression was detected using polyclonal goat antiactin antibodies (Santa Cruz Biotechnology). Monoclonal antibodies (Cell Signaling Technology) were used to detect phosphorylated eIF-2 and total eIF-2 proteins from 20 μg of protein from the cytoplasmic fraction (as determined by Bio-Rad protein assay) of cells that were harvested as described above with the addition of phosphatase inhibitors (90 mM sodium fluoride, 17.5 mM sodium molybdate, 17.5 mM ?-glycerophosphate).
Cellular RNA was isolated using the Trizol method (Gibco) and quantitated by UV absorbance at 260 nm. HCV replicon RNA from 0.5 x 106 cells was monitored by real-time reverse transcription-PCR (RT-PCR) (Taqman; Applied Biosystems) using primers located on the HCV 5' end and quantified on the basis of HCV RNA standards and relative amounts of cellular glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA as described previously (26).
RNase protection analysis was performed using the HybSpeed RPA kit (Ambion) to examine the transfection efficiency of luciferase reporters. Cellular RNA was isolated and incubated with an antisense RNA probe to 125 nucleotides of Renilla luciferase, synthesized with T7 RNA polymerase and [-32P]dCTP. The assay was performed per the manufacturer's suggestions. RNA was quantitated using known RNA concentrations of the sense strand of the probe, synthesized with SP6 RNA polymerase.
Luc assays. Luciferase assays were performed with a Dual-Luciferase reporter system (Promega) (per the manufacturer's directions), and Luc activity was measured with a luminometer (Turner Designs). Lysates were prepared from Huh7 or Huh.BB7 cells 48 h after transfection with a bicistronic reporter plasmid (2 μg DNA/transfection) or bicistronic Luc reporter RNA (synthesized in vitro with T7 RNA polymerase after linearization of the plasmid). Luc assay results shown are representative of four or more experiments with transfections performed in duplicate and samples tested in the Luc assay in triplicate. Similar results were obtained in all experiments.
TLC. Radioactive monophosphates were resolved by thin-layer chromatography (TLC) on polyethylenimine cellulose was developed in ammonium acetate buffer as described previously (19). Briefly, RNA was digested with nuclease P1 (Roche) after Trizol isolation from Huh.BB7 cells that were grown in the presence of [-32P]ATP. Nonradioactive AMP and IMP were used as migration standards and visualized with UV light. Radioactivity was visualized by autoradiography and quantitated by PhosphorImager analysis.
siRNA. Huh.BB7 cells were plated at 1 x 106 cells/ml. Huh.BB7 cells were transfected with ADAR1 or ADAR2 small interfering RNA (siRNA) (final concentration of 100 nM) (a gift of John Casey) as described previously (23) with DMRIE-C transfection reagent (Invitrogen). Cells were transfected with ADAR siRNA once or transfected twice (24 h later) and harvested 7 days after the initial transfection.
Isolation and sequencing of replicon RNA. Huh.BB7 cells were treated with IFN- (100 IU/ml, 18 h), and cytoplasmic RNA was extracted by disrupting cells in 0.1% NP-40 and 0.1x (10%) PBS as described above. RT was performed using First strand cDNA kit (Amersham) with a primer in the antisense direction of the NS5 region (5'-CAACCGTCCTCTTCCTCCG-3'), and PCR was performed using Expand high fidelity PCR system (Roche) with primers directed towards the HCV IRES region (5'-GCATGCGTCGACGCCAGCCCCGGATTGGGG-3' and 5'-AGGTCGAGCTCGGCGCGCCCTTTGGTTTTTC-3'). PCR products were inserted into pCR II-Topo with a Topo TA cloning kit (Invitrogen), white clones were selected, and DNA was purified and sequenced with the forward primer included in the kit.
RESULTS
IFN- sensitivity in cell culture is replicon specific. The mechanisms underlying IFN- resistance by HCV have been examined with the replicon-based system (11). Efficient replication of HCV RNA was observed in Huh7 cells that had been stably transfected with the HCV replicon (BB7) (2, 3, 11). While the replicons were derived from a genotype 1b strain, which is usually the most IFN resistant, these replicons were highly sensitive to IFN- (2). It was even possible to completely cure the cells of the replicon with IFN- treatment (3). In addition, sequence adaptations that conferred robust growth in cell culture did not reduce IFN- sensitivity (2).
We found that expression of the HCV replicon proteins was highly sensitive to IFN- (Fig. 1A). Total protein synthesis in Huh7 cells or replicon-containing cells (Huh.BB7) was also affected, as demonstrated by the decrease in total cellular proteins synthesized after IFN- treatment (Fig. 1B). IFN- treatment caused relatively little cytotoxicity in cells that contained the replicon or in the parental cells (Huh7) (data not shown). We performed Northern blot analysis to detect replicon RNA in IFN--treated cells. With increasing levels of IFN-, a decrease in replicon RNA was observed (data not shown). As previously reported by Blight et al. (2), we observed a precipitous decrease in replicon RNA after IFN- treatment as measured by Taqman analysis, while total RNA in the cells was largely unaffected (Fig. 1C), demonstrating that the primary effects of IFN- are replicon specific.
The HCV IRES is IFN- resistant in replicon-containing cells. Translation of HCV is initiated at the highly structured 5' untranslated region of the viral RNA containing an IRES. To determine how replicon-specific protein synthesis was affected by IFN-, we engineered a bicistronic DNA reporter that expresses Renilla luciferase under a 5' cap-dependent translational mechanism and firefly Luc under a cap-independent, IRES-directed mechanism. We examined the effects of IFN- on expression from both the EMCV IRES and the HCV IRES (Fig. 2A), because both are required for replicon expression. Both cap-dependent and EMCV IRES-dependent Luc expression was inhibited by IFN- in both Huh7.BB7 and Huh7 cells (Fig. 2B and C). Complete inhibition of Luc activity was not observed, because Luc protein accumulated for 30 h prior to the addition of IFN-. HCV IRES-dependent Luc expression was also inhibited by IFN- (Fig. 2C), suggesting that there is a global effect of IFN- at the posttranscriptional stage in transfected cells. This is consistent with an inhibition of translation by PKR, as PKR activation would result in a decrease in both cap- and IRES-dependent translation.
To exclude the possibility that transcription of the Luc reporter was inhibited by IFN-, we compared Luc expression using transfected DNA to Luc expression from transfected capped RNA synthesized in vitro from the same reporter and observed radically different responses to IFN- treatment. When DNA was transfected into Huh.BB7 cells, the EMCV-IRES-dependent expression, HCV-IRES-dependent expression, and cap-dependent expression of Luc were all inhibited by IFN- (Fig. 2B). The parental Huh7 cells lacking the replicon behaved similarly (Fig. 2C). However, when we transfected RNA encoding the Luc reporter, only HCV IRES-dependent Luc expression was resistant to IFN- in those cells that contained the replicon (Fig. 2D) and was sensitive to IFN- in Huh7 cells (Fig. 2E). These results were repeated with similar reporters described previously and are consistent with those reported by Koev et al. who proposed that a PKR-independent mechanism is responsible for IFN- activity on the replicon (9). Because PKR activation results in inhibition of both cap-dependent and -independent translation, if PKR were solely responsible for IFN- sensitivity of the cap-dependent Luc expression, then IRES-dependent Luc expression should be affected as well. These results show that in transfected cells, in the absence of the replicon, both cap-dependent translation and cap-independent translation were inhibited (Fig. 2C and E), which is consistent with inhibition due to activated PKR, but this does not exclude other mechanisms of inhibition. These results contradict results demonstrating that the replicon is sensitive to G418 in the presence of IFN- (data not shown). Therefore, the HCV IRES, in the context of the replicon, is sensitive to IFN-. These data may suggest that the RNA of the HCV IRES-containing reporter in the presence of the replicon may be inhibiting or blocking the main effects of IFN- on IRES translation in a manner that cannot be fulfilled by the DNA reporter or in cells that lack the replicon (Fig. 2F). Because PKR is activated by dsRNA, which is probably present in replicon-containing cells, translation inhibition through PKR is expected. However, PKR did not shut down translation of the HCV IRES reporter in replicon cells. Taken together, it appears that while PKR may be activated, this cannot fully explain the IFN- sensitivity of the replicon.
E2 and NS5A stimulate translation but did not rescue the replicon from IFN- sensitivity. To examine the importance of PKR-dependent translation inhibition on the replicon, we used PKR inhibitors to counteract the effect of IFN- in the Luc assay. Increasing amounts of transfected E2 stimulated both cap-dependent and IRES-dependent translation with and without IFN- (Fig. 3A). Stimulation of translation corresponded with increasing levels of E2 from cells cotransfected with the Luc reporter. Equal amounts of Luc reporter were transfected in these cells, as demonstrated by the RNase protection assay (Fig. 3A, RPA). Both E2 and NS5A have been shown to inhibit PKR (17, 18), and expression of each of these HCV genes enhanced Luc expression in the presence and absence of exogenous IFN- (Fig. 3B). We used a high level of IFN- to show the enhanced effect of E2 and NS5A. No differences in reporter expression were seen in Huh7 cells transfected with vector DNA. Interestingly, the combination of both E2 and NS5A enhanced Luc expression in Huh.BB7 cells in an additive manner (Fig. 3B), suggesting that E2 and NS5A counteract the effects of IFN- in different and complementary ways. Both E2 and NS5A were capable of rescuing the replication of HCV replicon RNA during IFN- treatment. Complete rescue was not observed, as the amount of HCV RNA declined after 24 h of IFN- treatment (Fig. 3C).
PKR is not induced but is activated by dsRNA present in replicon-containing cells. PKR was fully induced by IFN- treatment in replicon-expressing cells (Fig. 3D), but PKR was not endogenously induced in the absence of IFN- as indicated by the low level of PKR expression seen in untreated cells and the induction of PKR expression after IFN treatment. Endogenous IFN- or IFN-? was, therefore, not being secreted by untreated Huh.BB7 cells. PKR activation was monitored by observing phosphorylation levels of PKR substrate eIF-2 (Fig. 3E) compared to the total amount of eIF-2 (Fig. 3E). In Huh7 cells, eIF-2 was phosphorylated with the addition of IFN- and dsRNA, but very little eIF-2 was phosphorylated in the absence of treatment (Fig. 3E, lanes 1 and 2). This is consistent with the dsRNA-dependent activation of PKR. In Huh.BB7 cells, however, a high level of eIF-2 phosphorylation was observed in the absence of IFN-, and no additional phosphorylation was observed with the addition of dsRNA or IFN- (Fig. 3E, lanes 3 and 4). This indicates that although PKR was at a low basal level of expression in untreated Huh.BB7 cells (Fig. 3D), it was already activated in replicon-containing cells (Fig. 3E, lanes 3 and 4). Because no additional eIF-2 phosphorylation was observed after IFN- and dsRNA treatment (Fig. 3E, compare lanes 3 to 4), eIF-2 phosphorylation may have already saturated. Alternatively, the assay itself may be saturated, allowing for no additional detection of eIF-2 phosphorylation. To test this possibility and to evaluate the importance of PKR activation, we transfected the cells with the known PKR inhibitors HCV E2 (20) or NS5A (6) or both E2 and NS5A. All samples containing transfected E2 or NS5A demonstrated a decrease in the level of phosphorylated eIF-2 (Fig. 3E, compare lanes 5 to 10 to lanes 3 and 4), confirming that these two HCV genes inhibit PKR. Still, there was no additional phosphorylation of eIF-2 in the presence of IFN- (Fig. 3E), even though translation decreased with IFN- (Fig. 3A) and PKR protein expression increased in these cells (Fig. 3D), confirming that in these samples the assay was not saturated. These results also suggest that IFN- treatment of replicon-containing cells does not lead to additional activation of PKR. This supports evidence demonstrated here and by others (9) suggesting that not only PKR but other pathways modulate IFN- sensitivity of the replicon. It is consistent with the results showing that the PKR inhibitors can stimulate the replicon but cannot overcome all the negative effects of IFN- treatment (Fig. 3B, C, and E).
Because PKR was not fully induced (Fig. 3D) yet was already activated in the cells expressing the replicon (Fig. 3E), dsRNA was most likely present, presumably from RNA replication of both positive- and negative-strand RNA. Taken together, these data are consistent with a stimulation of replicon expression by E2 and NS5A due to inhibition of PKR, which is activated by the replicon. Because IFN- treatment was not required for activation of PKR (or its downstream effects) in replicon-containing cells (Fig. 3E) and yet, IFN- is a potent antagonist of the replicon, a second IFN--induced antiviral pathway may be involved. This is also consistent with the results obtained in Fig. 2, where HCV IRES translation was insensitive to IFN-. As shown previously by Koev et al. (9), PKR activation should result in the inhibition of both IRES-mediated and cap-mediated translation. Therefore, inhibition of the replicon by IFN may proceed through a secondary mechanism to PKR activation. These data may suggest that the low level of PKR expression was sufficiently activated to enable phosphorylation of maximum levels of eIF-2.
Activation of PKR probably occurred as a result of dsRNA present in replicon-containing cells but not in Huh7 cells alone. This is consistent with other findings showing that IFN- regulatory factor 3 (IRF-3), which is important for the initial induction of IFN-? in response to dsRNA or virus infection of cells, was inhibited in the presence of the replicon (5). The activation of IRF-3 usually leads to the induction of IFN-? and subsequently of IFN-/?-induced genes, such as PKR. We found that while dsRNA was present in replicon-containing cells, PKR expression was not induced, suggesting that the dsRNA induction of IFN-? was blocked, which is consistent with an inhibition of IRF-3.
VA RNAI confers IFN- resistance to the replicon. To examine the importance of eIF-2 phosphorylation in cellular and viral translation, we tested IFN- sensitivity of a dual-Luc reporter. We used WT PKR and well-characterized PKR inhibitors to measure stimulation of translation, which was measured as an output of Luc expression. WT PKR, a catalytically inactive dominant-negative mutant PKR (PKR K296R), an eIF-2 phosphorylation site mutant (eIF-2 S51A) that is nonphosphorylatable, and VA RNAI were cotransfected with the HCV IRES Luc reporter into Huh.BB7 cells. Overexpression of PKR (WT) inhibited expression of the Luc reporter, as expected (Fig. 4A), consistent with the behavior of transfected, active PKR (22). Both PKR K296R and eIF-2 S51A stimulated Luc expression, even in the presence of IFN (Fig. 4A), suggesting that eIF-2 phosphorylation was involved in the inhibition of both IRES-dependent and cap-dependent translation by IFN-. Cotransfection with a DNA plasmid encoding VA RNAI resulted in 25-fold stimulation of IRES-directed Luc expression in the absence of IFN and nearly 20-fold stimulation in the presence of IFN (Fig. 4A). Cap-dependent translation was also stimulated by VA and to a higher degree than demonstrated by any of the other PKR inhibitors. While VA RNA is a potent inhibitor of PKR, this finding may suggest that VA RNAI can stimulate translation in replicon-containing cells by conferring IFN resistance through a pathway in addition to PKR. To examine the effects of these inhibitors on replicon RNA expression, we measured HCV replicon RNA in cells that were treated with IFN- and transfected with the PKR inhibitors. Replicon expression was stimulated by all of the inhibitors but was stimulated most strongly in the presence of VA RNAI and eIF-2 S51A (Fig. 4B). When these cells were treated with IFN-, all inhibitors stimulated replicon expression over vector alone or the catalytically inactive PKR K296R mutant. VA and NS5A, however, showed very efficient rescue of replicon RNA expression (Fig. 4B). Because the PKR inhibitors eIF-2 S51A and PKR K296R stimulated the replicon weakly (Fig. 4B), it is evident that PKR is involved in the limitation of the replicon in these cells. However, the robust stimulation of translation (Fig. 4A) and replicon RNA by VA RNAI (Fig. 4B) may suggest that a second potent antiviral pathway may be involved in the IFN--induced inhibition of the HCV replicon. Because VA RNAI binds and inhibits ADAR1 in vitro (10), we looked for evidence of ADAR1 activity in replicon-containing cells.
IFN- results in A-to-I mutations in replicon RNA. To examine evidence for editing in IFN--treated replicon cells, we sequenced RT-PCR products from cytoplasmic fractions of IFN--treated and untreated Huh.BB7 cells. If adenosine-to-inosine editing events occurred, then adenosine residues will read as guanosine residues. None of the clones from untreated cells, sequenced from three distinct regions of the replicon (data not shown), diverged from the WT replicon sequence. However, one clone obtained from the IFN--treated cells contained mutations in adenosine residues resulting in guanosine (Fig. 5), suggesting that the replicon RNA was directly edited. Because we had difficulty in obtaining PCR products that contained mutations, this suggests that once sequences are edited, they may not be replicated or they may be degraded.
VA RNAI impairs A-to-I editing in replicon cells. In addition to its inhibition of PKR, VA RNAI is also known to inhibit ADAR1 (10). ADAR1 acts specifically on dsRNA, is IFN- inducible, and causes the conversion of adenosine in dsRNA to inosine by deamination. To test the hypothesis that RNA editing occurs in replicon-containing cells, we examined the effects of IFN- treatment on conversion of radiolabeled AMP to IMP. We isolated RNA that was metabolically labeled with [-32P]ATP from the cytoplasmic fraction of IFN--treated Huh.BB7 cells. Because VA RNAI has been shown to specifically inhibit ADAR1 (10) and can also rescue IFN--inhibited replicon expression very efficiently (Fig. 4B), we wanted to know whether VA was abrogating IFN- sensitivity by inhibiting ADAR in replicon-containing cells. We measured radiolabeled IMP production in cytoplasmic RNA from Huh.BB7 cells transfected with VA RNAI by PhosphorImager quantitation (Fig. 6A). Radiolabeled IMP was efficiently produced (16% of total adenosine products) in IFN-treated replicon cells. When we transfected the plasmid encoding VA RNAI, the generation of radiolabeled IMP was inhibited to less than 1%, consistent with a role for VA in the inhibition of A-to-I editing by ADAR1 in IFN--treated replicon-containing cells. Previously, ADAR has been shown to convert approximately 40% of adenosines to inosines in dsRNA (19).
In VA RNA-transfected cells, the replicon was stimulated in the presence and absence of IFN- (Fig. 6B), and yet ADAR1 was upregulated only when the cells were treated with IFN- (Fig. 6C). This is consistent with previous observations whereby replicon expression and Luc expression were stimulated by VA RNA in the absence of IFN- in replicon cells (Fig. 4) and suggests that ADAR1 may already be active (as with PKR) due to dsRNA present in replicon-containing cells.
siRNA knockdown of ADAR1 stimulates replicon expression. To confirm that the RNA editing occurring in replicon-containing cells was attributable to ADAR1, we used an RNA interference assay utilizing siRNA specifically directed to knock down the expression of ADAR1 (8). We used a siRNA that has been shown upon transfection to markedly decrease ADAR1 expression specifically (8). Cells were transfected once, and after 24 h, some cells were transfected for a second time. HCV replicon RNA increased with transfection of ADAR1 siRNA (Fig. 7A). At day 7 posttransfection, the HCV replicon RNA increased by 41-fold for one transfection and fivefold for two transfections of ADAR1 siRNA (Fig. 7A). Two transfections of ADAR1 siRNA were slightly toxic to the cells (data not shown), which may be caused by the loss of activity of the siRNA. The results observed directly correlate with the expression level of the p150, cytoplasmic, IFN--induced form of ADAR1 in cells (Fig. 7B). siRNA directed to knockdown ADAR2 (Fig. 7A) showed no stimulation of HCV replicon RNA, suggesting that ADAR1 and not ADAR2 is responsible for instability of HCV replicon RNA.
The stimulation of the replicon by ADAR1 knockdown was seen in the absence of IFN- (Fig. 7A and B). We, therefore, tested the effects of siRNA directed to ADAR1 in the presence of IFN-. In the absence of siRNA transfection, IFN- treatment resulted in a decrease in replicon RNA (Fig. 7C) corresponding with an increase in ADAR1 expression (Fig. 7D). Transfection with siRNA directed to ADAR1 resulted in loss of replicon RNA during IFN- treatment as well (Fig. 7C). However, compared to no-siRNA controls, the samples with siRNA yielded more replicon RNA (Fig. 7C), supporting the conclusion that IFN- sensitivity of the replicon is mediated through ADAR1. The ADAR1 immunoblot (Fig. 7D) demonstrates that in the presence of IFN-, siRNA does not completely knock down ADAR1 expression, consistent with the observed replicon sensitivity in these samples.
DISCUSSION
Our findings point to a new IFN--induced antiviral pathway important to the modulation of HCV replicon replication in cell culture. This pathway involves dsRNA-specific editing of adenosine residues by ADAR1 in HCV replicon-containing cells, which then leads to the loss of HCV replicon RNA. Viral RNA clearance may be attributable to one or several factors. (i) The viral RNA may be edited and degraded, possibly by an inosine-specific RNase (18, 19). (ii) Inefficient replication and genome instability may result from mutation of important viral sequences. (iii) A cellular message that is required for viral replication may be edited and inactivated. Taken together, our results explain why IFN- action is specific for HCV replicon RNA and not cellular RNA and how VA RNAI can rescue replicon expression so effectively. We are currently testing the HCV IRES RNA for its ability to bind and inhibit ADAR1 function. Why the addition of HCV IRES-containing RNA confers IFN- resistance to IRES-dependent translation and not cap-dependent translation is not easily answered, although it may indicate that the HCV IRES itself is protected from an IFN--induced mechanism (23). These results confirm the findings that IFN- action on the replicon is a result of both PKR and ADAR1 activation.
At this time we have not identified a specific editing site in HCV RNA, but random hyperedited viral RNA may be the result of ADAR1 action on the replicon. Because of the high level of radioactive inosine obtained, the HCV replicon may be hyperedited as a result of dsRNA produced by the replicase. At this time it has been difficult to obtain RT-PCR products containing mutations in the replicon that represent edited adenosines, and RT-PCR products from IFN--treated replicon cells have not yielded an abundance of clones. We believe that this may be due to the inability of the mutated replicon to replicate, the inability of PCR primers to recognize mutated sequences, and/or the HCV replicon RNA may be quickly degraded after IFN- treatment.
To what extent ADAR is involved in IFN- sensitivity in patients is also not understood at this time. The presence and genetic variability of viral genes that interact with ADAR1 may play a role in the outcome of IFN- therapy. Finally, the discovery that RNA editing negatively affects HCV RNA replication may lead to new approaches in the development of therapies that target this new IFN--induced antiviral pathway important for the clearance of HCV and potentially other viruses.
ACKNOWLEDGMENTS
We thank John Casey for ADAR siRNAs and ADAR1 antibody, Charles Rice and Keril Blight for replicon cell lines and DNA, Gennadiy Koev and Michael Lai for Luc plasmids and Natalia Martin for EMCV Luc plasmid, Olivier Donze for the mutant eIF-2 plasmid, and Nahum Sonenberg, Barry Falgout and Geetha Jayan for comments on the manuscript. IFN- was supplied by the NIH Research and Reference Reagent Program of NIAID.
REFERENCES
Bass, B. L. 1997. RNA editing and hypermutation by adenosine deamination. Trends Biochem. Sci. 22:157-162.
Blight, K. J., A. A. Kolykhalov, and C. M. Rice. 2000. Efficient initiation of HCV RNA replication in cell culture. Science 290:1972-1974.
Blight, K. J., J. A. McKeating, and C. M. Rice. 2002. Highly permissive cell lines for subgenomic and genomic hepatitis C virus RNA replication. J. Virol. 76:13001-13014.
Donze, O., R. Jagus, A. E. Koromilas, J. W. Hershey, and N. Sonenberg. 1995. Abrogation of translation initiation factor eIF-2 phosphorylation causes malignant transformation of NIH 3T3 cells. EMBO J. 14:3828-3834.
Foy, E., K. Li, C. Wang, R. Sumpter, Jr., M. Ikeda, S. M. Lemon, and M. J. Gale. 2003. Regulation of interferon regulatory factor-3 by the hepatitis C virus serine protease. Science 300:1145-1148.
Gale, M. J., M. J. Korth, N. M. Tang, S.-L. Tan, D. A. Hopkins, T. E. Dever, S. J. Polyak, D. R. Gretch, and M. G. Katze. 1997. Evidence that hepatitis C virus resistance to interferon is mediated through repression of the PKR protein kinase by the nonstructural 5A protein. Virology 230:217-227.
Gunnery, S., and M. B. Mathews. 1995. Functional mRNA can be generated by RNA polymerase III. Mol. Cell. Biol. 15:3597-3607.
Jayan, G. C., and J. L. Casey. 2002. Inhibition of hepatitis delta virus RNA editing by short inhibitory RNA-mediated knockdown of ADAR1 but not ADAR2 expression. J. Virol. 76:12399-12404.
Koev, G., R. F. Duncan, and M. M. C. Lai. 2002. Hepatitis C virus IRES-dependent translation is insensitive to an eIF2-independent mechanism of inhibition by interferon in hepatocyte cell lines. Virology 297:195-202.
Lei, M., Y. Liu, and C. E. Samuel. 1998. Adenovirus VAI RNA antagonizes the RNA-editing activity of the ADAR adenosine deaminase. Virology 245:188-196.
Lohmann, V., F. Korner, J.-O. Koch, U. Herian, L. Theilmann, and R. Bartenschlager. 1999. Replication of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science 285:110-113.
Maran, A., and M. B. Mathews. 1988. Characterization of the dsRNA implicated in the inhibition of protein synthesis in cells infected with a mutant adenovirus defective for VA RNAI. Virology 164:106-113.
McHutchison, J. G., S. C. Gordon, and E. R. Schiff. 1998. Interferon alfa-2b alone or in combination with ribavirin as initial treatment for chronic hepatitis C. N. Engl. J. Med. 339:1485-1492.
O'Malley, R. P., T. M. Mariano, J. Siekierka, and M. B. Mathews. 1986. A mechanism for the control of protein synthesis by adenovirus VA RNAI. Cell 44:391-400.
Polson, A. G., B. L. Bass, and J. L. Casey. 1996. RNA editing of hepatitis delta virus antigenome by dsRNA-adenosine deaminase. Nature 380:454-456.
Puig, M., M. E. Major, K. Mihalik, and S. M. Feinstone. 2004. Immunization of chimpanzees with an envelope protein-based vaccine enhances specific humoral and cellular immune responses that delay hepatitis C virus infection. Vaccine 22:991-1000.
Saunders, L. R., and G. N. Barber. 2003. The dsRNA binding protein family: critical roles, diverse cellular functions. FASEB J. 17:961-983.
Scadden, A. D. J., and C. W. J. Smith. 1997. A ribonuclease specific for inosine-containing RNA: a potential role in antiviral defense? EMBO J. 16:2140-2149.
Scadden, A. D. J., and C. W. J. Smith. 2001. Specific cleavage of hyper-edited dsRNAs. EMBO J. 20:4243-4252.
Taylor, D. R., S. T. Shi, P. R. Romano, G. N. Barber, and M. M. C. Lai. 1999. Inhibition of the interferon-inducible protein kinase PKR by HCV E2 protein. Science 285:107-110.
Taylor, D. R. 2000. Hepatitis C virus: evasion of the interferon-induced antiviral response. J. Mol. Med. 78:182-190.
Taylor, D. R., B. Tian, P. R. Romano, A. G. Hinnebusch, M. M. C. Lai, and M. B. Mathews. 2001. HCV E2 envelope protein does not inhibit PKR by simple competition with autophosphorylation sites in the RNA-binding domain. J. Virol. 75:1265-1273.
Vyas, J., A. Elia, and M. J. Clemens. 2003. Inhibition of the protein kinase PKR by the internal ribosome entry site of hepatitis C virus genomic RNA. RNA 9:858-870.
Wang, Q., J. Khillan, P. Gadue, and K. Nishikura. 2000. Requirement of the RNA editing deaminase ADAR1 gene for embryonic erythropoiesis. Science 290:1765-1768.
Wong, S. K., and D. W. Lazinski. 2002. Replicating hepatitis delta virus RNA is edited in the nucleus by the small form of ADAR1. Proc. Natl. Acad. Sci. USA 99:15118-15123.
Wong, S. K., S. Sato, and D. W. Lazinski. 2003. Elevated activity of the large form of ADAR1 in vivo: very efficient RNA editing occurs in the cytoplasm. RNA 9:586-598.(Deborah R. Taylor, Montse)
ABSTRACT
While many clinical hepatitis C virus (HCV) infections are resistant to alpha interferon (IFN-) therapy, subgenomic in vitro self-replicating HCV RNAs (HCV replicons) are characterized by marked IFN- sensitivity. IFN- treatment of replicon-containing cells results in a rapid loss of viral RNA via translation inhibition through double-stranded RNA-activated protein kinase (PKR) and also through a new pathway involving RNA editing by an adenosine deaminase that acts on double-stranded RNA (ADAR1). More than 200 genes are induced by IFN-, and yet only a few are attributed with an antiviral role. We show that inhibition of both PKR and ADAR1 by the addition of adenovirus-associated RNA stimulates replicon expression and reduces the amount of inosine recovered from RNA in replicon cells. Small inhibitory RNA, specific for ADAR1, stimulated the replicon 40-fold, indicating that ADAR1 has a role in limiting replication of the viral RNA. This is the first report of ADAR's involvement in a potent antiviral pathway and its action to specifically eliminate HCV RNA through adenosine to inosine editing. These results may explain successful HCV replicon clearance by IFN- in vitro and may provide a promising new therapeutic strategy for HCV as well as other viral infections.
INTRODUCTION
Hepatitis C virus (HCV) infects approximately 170 million individuals worldwide and nearly 3 million in the United States alone. Most cases of HCV infection become persistent and may result in chronic liver disease, cirrhosis, and hepatocellular carcinoma. The current combination antiviral therapy of pegylated alpha interferon (IFN-) with ribavirin is effective in approximately 50% of individuals treated, while monotherapy with IFN- alone is successful in less than 20% of patients (13). IFN- allows cells to become innately primed for defense against eventual virus attack by inducing the transcription of many genes, some of which are activated during virus infection. Only a few genes have been identified and characterized as mediators of the IFN--induced antiviral response, including the Mx proteins, major histocompatibility complex proteins, 2',5' oligoadenylate synthetase, and the double-stranded RNA (dsRNA)-activated protein kinase (PKR) (21). PKR is activated during viral infection, which results in the phosphorylation of the subunit of eukaryotic translation initiation factor 2 (eIF-2) and subsequent translational shutoff. An adenosine deaminase that acts on dsRNA (ADAR1) is also IFN- induced and catalyzes the deamination of adenosine residues in dsRNA (for a review, see reference 17), resulting in inosine substitution. Inosine residues are not abundantly found in cellular mRNAs, but when inosine residues are present, they are transcribed and translated as guanosine residues, which may lead to mutations (1, 24). An RNase that specifically degrades inosine-containing RNA has been described and was proposed to be part of a putative antiviral pathway (18, 19). Although antiviral activity has not been attributed to ADAR1, hepatitis delta virus utilizes ADAR1 editing to promote its viral life cycle (15).
Typically, dsRNA is found only in cells that are virus infected, and both DNA and RNA viruses may present dsRNA in the cell in the form of replicative intermediates (12). ADAR1 contains three copies of a conserved dsRNA-binding motif also found in PKR (25). Most viruses have developed strategies to evade the effects of IFN-. For instance, a single-stranded virus-encoded RNA with partially dsRNA features, adenovirus-associated (VA) RNAI, enhances translation and counteracts the effects of IFN- in adenovirus-infected cells by inhibiting PKR (14). VA RNA has also been shown to bind and inhibit ADAR1 (10). This is the first report implicating RNA editing by ADAR1 in the control of viral replication and may provide a potent strategy for an effective treatment against HCV based on ADAR1 activation.
MATERIALS AND METHODS
Expression vectors and HCV replicon. The BB7 replicon was a gift of C. M. Rice and K. Blight and was previously described (2). Briefly, the replicon expresses HCV NS3 through NS5B nonstructural genes under the encephalomyocarditis virus (EMCV) internal ribosome entry site (IRES) and neomycin resistance under the HCV IRES. The HCV 5' and 3' nontranslated regions are also present. Bicistronic reporter plasmids were constructed by substitution of the simian virus 40 promoter in pRL-SV40 (Promega) with the herpes simplex virus type 1 alpha 27 promoter (gift of N. Martin). The mRNA expresses Renilla luciferase (Luc) through a cap-dependent translation mechanism and firefly Luc under the EMCV IRES. PCR of the BB7 plasmid (using primers that contained a SalI restriction site at the 5' end and a SacI restriction site at the 3' end, covering nucleotides 1 to 386 in pHCVrep1bBB7) (16) was used to generate an HCV IRES cassette. The EMCV IRES was removed by digestion with SalI and SacI, and the HCV-IRES was inserted. pcDNA3 (Invitrogen) plasmids expressing wild-type (WT) PKR, PKR K296R, E2, and NS5A were described previously (20). eIF-2-pRc/CMV (eIF-2 S51A) was a gift of O. Donze and was described previously (4). VA RNAI-pUC119, expressed under the polymerase III intragenic promoter was described previously (7).
Cell culture. Huh7 and Huh.BB7 cells were grown at 37°C in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum, 2 mM L-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin. Huh7 cells were transfected with RNA transcribed from linearized pHCVrep1bBB7 plasmid. Stable transfectants were selected in Geneticin (G418; Gibco) and grown in G418 at 75 μg/ml, which was removed during experiments. Cellular proteins were metabolically labeled in cell culture by incubating the cells for 1 h at 37°C in DMEM minus methionine and then supplementing the medium for 1 h with [35S]methionine. RNA was metabolically labeled in cell culture by incubating the cells for 1 h at 37°C in DMEM minus phosphates and then incubating the cells for 16 h in medium supplemented with [-32P]ATP. Cell growth was monitored by counting the number of viable cells with trypan blue staining or with a Coulter Counter (Beckman Coulter, Inc.). Cells were treated with recombinant human IFN-2A (100 IU/ml in DMEM) for 18 h at 37°C, unless stated otherwise.
Analyses of protein and RNA. Cells were washed with phosphate-buffered saline (PBS), and the attached monolayer was incubated with trypsin EDTA for 5 minutes at 37°C. Cells were collected, concentrated in 0.1% NP-40 and 10% PBS, and lysed by three repeated freeze-thaw cycles. Nuclear and cellular membranes were removed by centrifugation, and cytoplasmic extracts were quantitated for protein concentration with a colorimetric absorbance protein assay (Bio-Rad). Expression of HCV NS3 protein in lysates from Huh7 or Huh.BB7 cells was monitored by resolving 25 μg of protein extract in sodium dodecyl sulfate-polyacrylamide gels and immunoblotting with polyclonal antibody to NS3 (chimp 1536 serum [16]). Polyclonal anti-PKR antiserum was made by inoculating New Zealand White rabbits with a keyhole limpet hemocyanin-conjugated peptide to the spacer region between the two dsRNA-binding motifs of PKR (peptide P1) (22). E2 expression was monitored with mouse monoclonal antibody A11 using cell extracts that were treated with endoglycosidase H. Actin expression was detected using polyclonal goat antiactin antibodies (Santa Cruz Biotechnology). Monoclonal antibodies (Cell Signaling Technology) were used to detect phosphorylated eIF-2 and total eIF-2 proteins from 20 μg of protein from the cytoplasmic fraction (as determined by Bio-Rad protein assay) of cells that were harvested as described above with the addition of phosphatase inhibitors (90 mM sodium fluoride, 17.5 mM sodium molybdate, 17.5 mM ?-glycerophosphate).
Cellular RNA was isolated using the Trizol method (Gibco) and quantitated by UV absorbance at 260 nm. HCV replicon RNA from 0.5 x 106 cells was monitored by real-time reverse transcription-PCR (RT-PCR) (Taqman; Applied Biosystems) using primers located on the HCV 5' end and quantified on the basis of HCV RNA standards and relative amounts of cellular glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA as described previously (26).
RNase protection analysis was performed using the HybSpeed RPA kit (Ambion) to examine the transfection efficiency of luciferase reporters. Cellular RNA was isolated and incubated with an antisense RNA probe to 125 nucleotides of Renilla luciferase, synthesized with T7 RNA polymerase and [-32P]dCTP. The assay was performed per the manufacturer's suggestions. RNA was quantitated using known RNA concentrations of the sense strand of the probe, synthesized with SP6 RNA polymerase.
Luc assays. Luciferase assays were performed with a Dual-Luciferase reporter system (Promega) (per the manufacturer's directions), and Luc activity was measured with a luminometer (Turner Designs). Lysates were prepared from Huh7 or Huh.BB7 cells 48 h after transfection with a bicistronic reporter plasmid (2 μg DNA/transfection) or bicistronic Luc reporter RNA (synthesized in vitro with T7 RNA polymerase after linearization of the plasmid). Luc assay results shown are representative of four or more experiments with transfections performed in duplicate and samples tested in the Luc assay in triplicate. Similar results were obtained in all experiments.
TLC. Radioactive monophosphates were resolved by thin-layer chromatography (TLC) on polyethylenimine cellulose was developed in ammonium acetate buffer as described previously (19). Briefly, RNA was digested with nuclease P1 (Roche) after Trizol isolation from Huh.BB7 cells that were grown in the presence of [-32P]ATP. Nonradioactive AMP and IMP were used as migration standards and visualized with UV light. Radioactivity was visualized by autoradiography and quantitated by PhosphorImager analysis.
siRNA. Huh.BB7 cells were plated at 1 x 106 cells/ml. Huh.BB7 cells were transfected with ADAR1 or ADAR2 small interfering RNA (siRNA) (final concentration of 100 nM) (a gift of John Casey) as described previously (23) with DMRIE-C transfection reagent (Invitrogen). Cells were transfected with ADAR siRNA once or transfected twice (24 h later) and harvested 7 days after the initial transfection.
Isolation and sequencing of replicon RNA. Huh.BB7 cells were treated with IFN- (100 IU/ml, 18 h), and cytoplasmic RNA was extracted by disrupting cells in 0.1% NP-40 and 0.1x (10%) PBS as described above. RT was performed using First strand cDNA kit (Amersham) with a primer in the antisense direction of the NS5 region (5'-CAACCGTCCTCTTCCTCCG-3'), and PCR was performed using Expand high fidelity PCR system (Roche) with primers directed towards the HCV IRES region (5'-GCATGCGTCGACGCCAGCCCCGGATTGGGG-3' and 5'-AGGTCGAGCTCGGCGCGCCCTTTGGTTTTTC-3'). PCR products were inserted into pCR II-Topo with a Topo TA cloning kit (Invitrogen), white clones were selected, and DNA was purified and sequenced with the forward primer included in the kit.
RESULTS
IFN- sensitivity in cell culture is replicon specific. The mechanisms underlying IFN- resistance by HCV have been examined with the replicon-based system (11). Efficient replication of HCV RNA was observed in Huh7 cells that had been stably transfected with the HCV replicon (BB7) (2, 3, 11). While the replicons were derived from a genotype 1b strain, which is usually the most IFN resistant, these replicons were highly sensitive to IFN- (2). It was even possible to completely cure the cells of the replicon with IFN- treatment (3). In addition, sequence adaptations that conferred robust growth in cell culture did not reduce IFN- sensitivity (2).
We found that expression of the HCV replicon proteins was highly sensitive to IFN- (Fig. 1A). Total protein synthesis in Huh7 cells or replicon-containing cells (Huh.BB7) was also affected, as demonstrated by the decrease in total cellular proteins synthesized after IFN- treatment (Fig. 1B). IFN- treatment caused relatively little cytotoxicity in cells that contained the replicon or in the parental cells (Huh7) (data not shown). We performed Northern blot analysis to detect replicon RNA in IFN--treated cells. With increasing levels of IFN-, a decrease in replicon RNA was observed (data not shown). As previously reported by Blight et al. (2), we observed a precipitous decrease in replicon RNA after IFN- treatment as measured by Taqman analysis, while total RNA in the cells was largely unaffected (Fig. 1C), demonstrating that the primary effects of IFN- are replicon specific.
The HCV IRES is IFN- resistant in replicon-containing cells. Translation of HCV is initiated at the highly structured 5' untranslated region of the viral RNA containing an IRES. To determine how replicon-specific protein synthesis was affected by IFN-, we engineered a bicistronic DNA reporter that expresses Renilla luciferase under a 5' cap-dependent translational mechanism and firefly Luc under a cap-independent, IRES-directed mechanism. We examined the effects of IFN- on expression from both the EMCV IRES and the HCV IRES (Fig. 2A), because both are required for replicon expression. Both cap-dependent and EMCV IRES-dependent Luc expression was inhibited by IFN- in both Huh7.BB7 and Huh7 cells (Fig. 2B and C). Complete inhibition of Luc activity was not observed, because Luc protein accumulated for 30 h prior to the addition of IFN-. HCV IRES-dependent Luc expression was also inhibited by IFN- (Fig. 2C), suggesting that there is a global effect of IFN- at the posttranscriptional stage in transfected cells. This is consistent with an inhibition of translation by PKR, as PKR activation would result in a decrease in both cap- and IRES-dependent translation.
To exclude the possibility that transcription of the Luc reporter was inhibited by IFN-, we compared Luc expression using transfected DNA to Luc expression from transfected capped RNA synthesized in vitro from the same reporter and observed radically different responses to IFN- treatment. When DNA was transfected into Huh.BB7 cells, the EMCV-IRES-dependent expression, HCV-IRES-dependent expression, and cap-dependent expression of Luc were all inhibited by IFN- (Fig. 2B). The parental Huh7 cells lacking the replicon behaved similarly (Fig. 2C). However, when we transfected RNA encoding the Luc reporter, only HCV IRES-dependent Luc expression was resistant to IFN- in those cells that contained the replicon (Fig. 2D) and was sensitive to IFN- in Huh7 cells (Fig. 2E). These results were repeated with similar reporters described previously and are consistent with those reported by Koev et al. who proposed that a PKR-independent mechanism is responsible for IFN- activity on the replicon (9). Because PKR activation results in inhibition of both cap-dependent and -independent translation, if PKR were solely responsible for IFN- sensitivity of the cap-dependent Luc expression, then IRES-dependent Luc expression should be affected as well. These results show that in transfected cells, in the absence of the replicon, both cap-dependent translation and cap-independent translation were inhibited (Fig. 2C and E), which is consistent with inhibition due to activated PKR, but this does not exclude other mechanisms of inhibition. These results contradict results demonstrating that the replicon is sensitive to G418 in the presence of IFN- (data not shown). Therefore, the HCV IRES, in the context of the replicon, is sensitive to IFN-. These data may suggest that the RNA of the HCV IRES-containing reporter in the presence of the replicon may be inhibiting or blocking the main effects of IFN- on IRES translation in a manner that cannot be fulfilled by the DNA reporter or in cells that lack the replicon (Fig. 2F). Because PKR is activated by dsRNA, which is probably present in replicon-containing cells, translation inhibition through PKR is expected. However, PKR did not shut down translation of the HCV IRES reporter in replicon cells. Taken together, it appears that while PKR may be activated, this cannot fully explain the IFN- sensitivity of the replicon.
E2 and NS5A stimulate translation but did not rescue the replicon from IFN- sensitivity. To examine the importance of PKR-dependent translation inhibition on the replicon, we used PKR inhibitors to counteract the effect of IFN- in the Luc assay. Increasing amounts of transfected E2 stimulated both cap-dependent and IRES-dependent translation with and without IFN- (Fig. 3A). Stimulation of translation corresponded with increasing levels of E2 from cells cotransfected with the Luc reporter. Equal amounts of Luc reporter were transfected in these cells, as demonstrated by the RNase protection assay (Fig. 3A, RPA). Both E2 and NS5A have been shown to inhibit PKR (17, 18), and expression of each of these HCV genes enhanced Luc expression in the presence and absence of exogenous IFN- (Fig. 3B). We used a high level of IFN- to show the enhanced effect of E2 and NS5A. No differences in reporter expression were seen in Huh7 cells transfected with vector DNA. Interestingly, the combination of both E2 and NS5A enhanced Luc expression in Huh.BB7 cells in an additive manner (Fig. 3B), suggesting that E2 and NS5A counteract the effects of IFN- in different and complementary ways. Both E2 and NS5A were capable of rescuing the replication of HCV replicon RNA during IFN- treatment. Complete rescue was not observed, as the amount of HCV RNA declined after 24 h of IFN- treatment (Fig. 3C).
PKR is not induced but is activated by dsRNA present in replicon-containing cells. PKR was fully induced by IFN- treatment in replicon-expressing cells (Fig. 3D), but PKR was not endogenously induced in the absence of IFN- as indicated by the low level of PKR expression seen in untreated cells and the induction of PKR expression after IFN treatment. Endogenous IFN- or IFN-? was, therefore, not being secreted by untreated Huh.BB7 cells. PKR activation was monitored by observing phosphorylation levels of PKR substrate eIF-2 (Fig. 3E) compared to the total amount of eIF-2 (Fig. 3E). In Huh7 cells, eIF-2 was phosphorylated with the addition of IFN- and dsRNA, but very little eIF-2 was phosphorylated in the absence of treatment (Fig. 3E, lanes 1 and 2). This is consistent with the dsRNA-dependent activation of PKR. In Huh.BB7 cells, however, a high level of eIF-2 phosphorylation was observed in the absence of IFN-, and no additional phosphorylation was observed with the addition of dsRNA or IFN- (Fig. 3E, lanes 3 and 4). This indicates that although PKR was at a low basal level of expression in untreated Huh.BB7 cells (Fig. 3D), it was already activated in replicon-containing cells (Fig. 3E, lanes 3 and 4). Because no additional eIF-2 phosphorylation was observed after IFN- and dsRNA treatment (Fig. 3E, compare lanes 3 to 4), eIF-2 phosphorylation may have already saturated. Alternatively, the assay itself may be saturated, allowing for no additional detection of eIF-2 phosphorylation. To test this possibility and to evaluate the importance of PKR activation, we transfected the cells with the known PKR inhibitors HCV E2 (20) or NS5A (6) or both E2 and NS5A. All samples containing transfected E2 or NS5A demonstrated a decrease in the level of phosphorylated eIF-2 (Fig. 3E, compare lanes 5 to 10 to lanes 3 and 4), confirming that these two HCV genes inhibit PKR. Still, there was no additional phosphorylation of eIF-2 in the presence of IFN- (Fig. 3E), even though translation decreased with IFN- (Fig. 3A) and PKR protein expression increased in these cells (Fig. 3D), confirming that in these samples the assay was not saturated. These results also suggest that IFN- treatment of replicon-containing cells does not lead to additional activation of PKR. This supports evidence demonstrated here and by others (9) suggesting that not only PKR but other pathways modulate IFN- sensitivity of the replicon. It is consistent with the results showing that the PKR inhibitors can stimulate the replicon but cannot overcome all the negative effects of IFN- treatment (Fig. 3B, C, and E).
Because PKR was not fully induced (Fig. 3D) yet was already activated in the cells expressing the replicon (Fig. 3E), dsRNA was most likely present, presumably from RNA replication of both positive- and negative-strand RNA. Taken together, these data are consistent with a stimulation of replicon expression by E2 and NS5A due to inhibition of PKR, which is activated by the replicon. Because IFN- treatment was not required for activation of PKR (or its downstream effects) in replicon-containing cells (Fig. 3E) and yet, IFN- is a potent antagonist of the replicon, a second IFN--induced antiviral pathway may be involved. This is also consistent with the results obtained in Fig. 2, where HCV IRES translation was insensitive to IFN-. As shown previously by Koev et al. (9), PKR activation should result in the inhibition of both IRES-mediated and cap-mediated translation. Therefore, inhibition of the replicon by IFN may proceed through a secondary mechanism to PKR activation. These data may suggest that the low level of PKR expression was sufficiently activated to enable phosphorylation of maximum levels of eIF-2.
Activation of PKR probably occurred as a result of dsRNA present in replicon-containing cells but not in Huh7 cells alone. This is consistent with other findings showing that IFN- regulatory factor 3 (IRF-3), which is important for the initial induction of IFN-? in response to dsRNA or virus infection of cells, was inhibited in the presence of the replicon (5). The activation of IRF-3 usually leads to the induction of IFN-? and subsequently of IFN-/?-induced genes, such as PKR. We found that while dsRNA was present in replicon-containing cells, PKR expression was not induced, suggesting that the dsRNA induction of IFN-? was blocked, which is consistent with an inhibition of IRF-3.
VA RNAI confers IFN- resistance to the replicon. To examine the importance of eIF-2 phosphorylation in cellular and viral translation, we tested IFN- sensitivity of a dual-Luc reporter. We used WT PKR and well-characterized PKR inhibitors to measure stimulation of translation, which was measured as an output of Luc expression. WT PKR, a catalytically inactive dominant-negative mutant PKR (PKR K296R), an eIF-2 phosphorylation site mutant (eIF-2 S51A) that is nonphosphorylatable, and VA RNAI were cotransfected with the HCV IRES Luc reporter into Huh.BB7 cells. Overexpression of PKR (WT) inhibited expression of the Luc reporter, as expected (Fig. 4A), consistent with the behavior of transfected, active PKR (22). Both PKR K296R and eIF-2 S51A stimulated Luc expression, even in the presence of IFN (Fig. 4A), suggesting that eIF-2 phosphorylation was involved in the inhibition of both IRES-dependent and cap-dependent translation by IFN-. Cotransfection with a DNA plasmid encoding VA RNAI resulted in 25-fold stimulation of IRES-directed Luc expression in the absence of IFN and nearly 20-fold stimulation in the presence of IFN (Fig. 4A). Cap-dependent translation was also stimulated by VA and to a higher degree than demonstrated by any of the other PKR inhibitors. While VA RNA is a potent inhibitor of PKR, this finding may suggest that VA RNAI can stimulate translation in replicon-containing cells by conferring IFN resistance through a pathway in addition to PKR. To examine the effects of these inhibitors on replicon RNA expression, we measured HCV replicon RNA in cells that were treated with IFN- and transfected with the PKR inhibitors. Replicon expression was stimulated by all of the inhibitors but was stimulated most strongly in the presence of VA RNAI and eIF-2 S51A (Fig. 4B). When these cells were treated with IFN-, all inhibitors stimulated replicon expression over vector alone or the catalytically inactive PKR K296R mutant. VA and NS5A, however, showed very efficient rescue of replicon RNA expression (Fig. 4B). Because the PKR inhibitors eIF-2 S51A and PKR K296R stimulated the replicon weakly (Fig. 4B), it is evident that PKR is involved in the limitation of the replicon in these cells. However, the robust stimulation of translation (Fig. 4A) and replicon RNA by VA RNAI (Fig. 4B) may suggest that a second potent antiviral pathway may be involved in the IFN--induced inhibition of the HCV replicon. Because VA RNAI binds and inhibits ADAR1 in vitro (10), we looked for evidence of ADAR1 activity in replicon-containing cells.
IFN- results in A-to-I mutations in replicon RNA. To examine evidence for editing in IFN--treated replicon cells, we sequenced RT-PCR products from cytoplasmic fractions of IFN--treated and untreated Huh.BB7 cells. If adenosine-to-inosine editing events occurred, then adenosine residues will read as guanosine residues. None of the clones from untreated cells, sequenced from three distinct regions of the replicon (data not shown), diverged from the WT replicon sequence. However, one clone obtained from the IFN--treated cells contained mutations in adenosine residues resulting in guanosine (Fig. 5), suggesting that the replicon RNA was directly edited. Because we had difficulty in obtaining PCR products that contained mutations, this suggests that once sequences are edited, they may not be replicated or they may be degraded.
VA RNAI impairs A-to-I editing in replicon cells. In addition to its inhibition of PKR, VA RNAI is also known to inhibit ADAR1 (10). ADAR1 acts specifically on dsRNA, is IFN- inducible, and causes the conversion of adenosine in dsRNA to inosine by deamination. To test the hypothesis that RNA editing occurs in replicon-containing cells, we examined the effects of IFN- treatment on conversion of radiolabeled AMP to IMP. We isolated RNA that was metabolically labeled with [-32P]ATP from the cytoplasmic fraction of IFN--treated Huh.BB7 cells. Because VA RNAI has been shown to specifically inhibit ADAR1 (10) and can also rescue IFN--inhibited replicon expression very efficiently (Fig. 4B), we wanted to know whether VA was abrogating IFN- sensitivity by inhibiting ADAR in replicon-containing cells. We measured radiolabeled IMP production in cytoplasmic RNA from Huh.BB7 cells transfected with VA RNAI by PhosphorImager quantitation (Fig. 6A). Radiolabeled IMP was efficiently produced (16% of total adenosine products) in IFN-treated replicon cells. When we transfected the plasmid encoding VA RNAI, the generation of radiolabeled IMP was inhibited to less than 1%, consistent with a role for VA in the inhibition of A-to-I editing by ADAR1 in IFN--treated replicon-containing cells. Previously, ADAR has been shown to convert approximately 40% of adenosines to inosines in dsRNA (19).
In VA RNA-transfected cells, the replicon was stimulated in the presence and absence of IFN- (Fig. 6B), and yet ADAR1 was upregulated only when the cells were treated with IFN- (Fig. 6C). This is consistent with previous observations whereby replicon expression and Luc expression were stimulated by VA RNA in the absence of IFN- in replicon cells (Fig. 4) and suggests that ADAR1 may already be active (as with PKR) due to dsRNA present in replicon-containing cells.
siRNA knockdown of ADAR1 stimulates replicon expression. To confirm that the RNA editing occurring in replicon-containing cells was attributable to ADAR1, we used an RNA interference assay utilizing siRNA specifically directed to knock down the expression of ADAR1 (8). We used a siRNA that has been shown upon transfection to markedly decrease ADAR1 expression specifically (8). Cells were transfected once, and after 24 h, some cells were transfected for a second time. HCV replicon RNA increased with transfection of ADAR1 siRNA (Fig. 7A). At day 7 posttransfection, the HCV replicon RNA increased by 41-fold for one transfection and fivefold for two transfections of ADAR1 siRNA (Fig. 7A). Two transfections of ADAR1 siRNA were slightly toxic to the cells (data not shown), which may be caused by the loss of activity of the siRNA. The results observed directly correlate with the expression level of the p150, cytoplasmic, IFN--induced form of ADAR1 in cells (Fig. 7B). siRNA directed to knockdown ADAR2 (Fig. 7A) showed no stimulation of HCV replicon RNA, suggesting that ADAR1 and not ADAR2 is responsible for instability of HCV replicon RNA.
The stimulation of the replicon by ADAR1 knockdown was seen in the absence of IFN- (Fig. 7A and B). We, therefore, tested the effects of siRNA directed to ADAR1 in the presence of IFN-. In the absence of siRNA transfection, IFN- treatment resulted in a decrease in replicon RNA (Fig. 7C) corresponding with an increase in ADAR1 expression (Fig. 7D). Transfection with siRNA directed to ADAR1 resulted in loss of replicon RNA during IFN- treatment as well (Fig. 7C). However, compared to no-siRNA controls, the samples with siRNA yielded more replicon RNA (Fig. 7C), supporting the conclusion that IFN- sensitivity of the replicon is mediated through ADAR1. The ADAR1 immunoblot (Fig. 7D) demonstrates that in the presence of IFN-, siRNA does not completely knock down ADAR1 expression, consistent with the observed replicon sensitivity in these samples.
DISCUSSION
Our findings point to a new IFN--induced antiviral pathway important to the modulation of HCV replicon replication in cell culture. This pathway involves dsRNA-specific editing of adenosine residues by ADAR1 in HCV replicon-containing cells, which then leads to the loss of HCV replicon RNA. Viral RNA clearance may be attributable to one or several factors. (i) The viral RNA may be edited and degraded, possibly by an inosine-specific RNase (18, 19). (ii) Inefficient replication and genome instability may result from mutation of important viral sequences. (iii) A cellular message that is required for viral replication may be edited and inactivated. Taken together, our results explain why IFN- action is specific for HCV replicon RNA and not cellular RNA and how VA RNAI can rescue replicon expression so effectively. We are currently testing the HCV IRES RNA for its ability to bind and inhibit ADAR1 function. Why the addition of HCV IRES-containing RNA confers IFN- resistance to IRES-dependent translation and not cap-dependent translation is not easily answered, although it may indicate that the HCV IRES itself is protected from an IFN--induced mechanism (23). These results confirm the findings that IFN- action on the replicon is a result of both PKR and ADAR1 activation.
At this time we have not identified a specific editing site in HCV RNA, but random hyperedited viral RNA may be the result of ADAR1 action on the replicon. Because of the high level of radioactive inosine obtained, the HCV replicon may be hyperedited as a result of dsRNA produced by the replicase. At this time it has been difficult to obtain RT-PCR products containing mutations in the replicon that represent edited adenosines, and RT-PCR products from IFN--treated replicon cells have not yielded an abundance of clones. We believe that this may be due to the inability of the mutated replicon to replicate, the inability of PCR primers to recognize mutated sequences, and/or the HCV replicon RNA may be quickly degraded after IFN- treatment.
To what extent ADAR is involved in IFN- sensitivity in patients is also not understood at this time. The presence and genetic variability of viral genes that interact with ADAR1 may play a role in the outcome of IFN- therapy. Finally, the discovery that RNA editing negatively affects HCV RNA replication may lead to new approaches in the development of therapies that target this new IFN--induced antiviral pathway important for the clearance of HCV and potentially other viruses.
ACKNOWLEDGMENTS
We thank John Casey for ADAR siRNAs and ADAR1 antibody, Charles Rice and Keril Blight for replicon cell lines and DNA, Gennadiy Koev and Michael Lai for Luc plasmids and Natalia Martin for EMCV Luc plasmid, Olivier Donze for the mutant eIF-2 plasmid, and Nahum Sonenberg, Barry Falgout and Geetha Jayan for comments on the manuscript. IFN- was supplied by the NIH Research and Reference Reagent Program of NIAID.
REFERENCES
Bass, B. L. 1997. RNA editing and hypermutation by adenosine deamination. Trends Biochem. Sci. 22:157-162.
Blight, K. J., A. A. Kolykhalov, and C. M. Rice. 2000. Efficient initiation of HCV RNA replication in cell culture. Science 290:1972-1974.
Blight, K. J., J. A. McKeating, and C. M. Rice. 2002. Highly permissive cell lines for subgenomic and genomic hepatitis C virus RNA replication. J. Virol. 76:13001-13014.
Donze, O., R. Jagus, A. E. Koromilas, J. W. Hershey, and N. Sonenberg. 1995. Abrogation of translation initiation factor eIF-2 phosphorylation causes malignant transformation of NIH 3T3 cells. EMBO J. 14:3828-3834.
Foy, E., K. Li, C. Wang, R. Sumpter, Jr., M. Ikeda, S. M. Lemon, and M. J. Gale. 2003. Regulation of interferon regulatory factor-3 by the hepatitis C virus serine protease. Science 300:1145-1148.
Gale, M. J., M. J. Korth, N. M. Tang, S.-L. Tan, D. A. Hopkins, T. E. Dever, S. J. Polyak, D. R. Gretch, and M. G. Katze. 1997. Evidence that hepatitis C virus resistance to interferon is mediated through repression of the PKR protein kinase by the nonstructural 5A protein. Virology 230:217-227.
Gunnery, S., and M. B. Mathews. 1995. Functional mRNA can be generated by RNA polymerase III. Mol. Cell. Biol. 15:3597-3607.
Jayan, G. C., and J. L. Casey. 2002. Inhibition of hepatitis delta virus RNA editing by short inhibitory RNA-mediated knockdown of ADAR1 but not ADAR2 expression. J. Virol. 76:12399-12404.
Koev, G., R. F. Duncan, and M. M. C. Lai. 2002. Hepatitis C virus IRES-dependent translation is insensitive to an eIF2-independent mechanism of inhibition by interferon in hepatocyte cell lines. Virology 297:195-202.
Lei, M., Y. Liu, and C. E. Samuel. 1998. Adenovirus VAI RNA antagonizes the RNA-editing activity of the ADAR adenosine deaminase. Virology 245:188-196.
Lohmann, V., F. Korner, J.-O. Koch, U. Herian, L. Theilmann, and R. Bartenschlager. 1999. Replication of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science 285:110-113.
Maran, A., and M. B. Mathews. 1988. Characterization of the dsRNA implicated in the inhibition of protein synthesis in cells infected with a mutant adenovirus defective for VA RNAI. Virology 164:106-113.
McHutchison, J. G., S. C. Gordon, and E. R. Schiff. 1998. Interferon alfa-2b alone or in combination with ribavirin as initial treatment for chronic hepatitis C. N. Engl. J. Med. 339:1485-1492.
O'Malley, R. P., T. M. Mariano, J. Siekierka, and M. B. Mathews. 1986. A mechanism for the control of protein synthesis by adenovirus VA RNAI. Cell 44:391-400.
Polson, A. G., B. L. Bass, and J. L. Casey. 1996. RNA editing of hepatitis delta virus antigenome by dsRNA-adenosine deaminase. Nature 380:454-456.
Puig, M., M. E. Major, K. Mihalik, and S. M. Feinstone. 2004. Immunization of chimpanzees with an envelope protein-based vaccine enhances specific humoral and cellular immune responses that delay hepatitis C virus infection. Vaccine 22:991-1000.
Saunders, L. R., and G. N. Barber. 2003. The dsRNA binding protein family: critical roles, diverse cellular functions. FASEB J. 17:961-983.
Scadden, A. D. J., and C. W. J. Smith. 1997. A ribonuclease specific for inosine-containing RNA: a potential role in antiviral defense? EMBO J. 16:2140-2149.
Scadden, A. D. J., and C. W. J. Smith. 2001. Specific cleavage of hyper-edited dsRNAs. EMBO J. 20:4243-4252.
Taylor, D. R., S. T. Shi, P. R. Romano, G. N. Barber, and M. M. C. Lai. 1999. Inhibition of the interferon-inducible protein kinase PKR by HCV E2 protein. Science 285:107-110.
Taylor, D. R. 2000. Hepatitis C virus: evasion of the interferon-induced antiviral response. J. Mol. Med. 78:182-190.
Taylor, D. R., B. Tian, P. R. Romano, A. G. Hinnebusch, M. M. C. Lai, and M. B. Mathews. 2001. HCV E2 envelope protein does not inhibit PKR by simple competition with autophosphorylation sites in the RNA-binding domain. J. Virol. 75:1265-1273.
Vyas, J., A. Elia, and M. J. Clemens. 2003. Inhibition of the protein kinase PKR by the internal ribosome entry site of hepatitis C virus genomic RNA. RNA 9:858-870.
Wang, Q., J. Khillan, P. Gadue, and K. Nishikura. 2000. Requirement of the RNA editing deaminase ADAR1 gene for embryonic erythropoiesis. Science 290:1765-1768.
Wong, S. K., and D. W. Lazinski. 2002. Replicating hepatitis delta virus RNA is edited in the nucleus by the small form of ADAR1. Proc. Natl. Acad. Sci. USA 99:15118-15123.
Wong, S. K., S. Sato, and D. W. Lazinski. 2003. Elevated activity of the large form of ADAR1 in vivo: very efficient RNA editing occurs in the cytoplasm. RNA 9:586-598.(Deborah R. Taylor, Montse)