Previous Article | Next Article ![]()
Journal of Virology, March 2005, p. 3174-3178, Vol. 79, No. 5
0022-538X/05/$08.00+0 doi:10.1128/JVI.79.5.3174-3178.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Department of Biological Sciences, Boehringer Ingelheim Canada Ltd., Research and Development, Laval, Quebec, Canada
Received 20 July 2004/ Accepted 19 October 2004
|
|
|---|
|
|
|---|
dsRNA, a by-product of virus replication, is believed to activate the innate immune system and mediate the production of alpha/beta interferon (IFN-
/ß) and other cytokines (10). dsRNA and poly(I · C) are candidate ligands for Toll-like receptor 3 (TLR-3) (1, 10, 21). Engagement of poly(I · C) with the receptor is proposed to activate a signaling cascade that is MyD88 independent and involves the recruitment of Toll or interleukin 1 receptor domain-containing, adaptor-inducing IFN-ß (TRIF) (22, 26, 27), which in turn is postulated to activate various signaling mediators, including I
B kinase
and TANK-binding kinase 1. These kinases phosphorylate and activate IFN-regulatory factor 3 (IRF-3) (8, 22, 24), which has emerged as a critical mediator of antiviral responses (23). The role of IRF-3 in the induction of antiviral responses by dsRNA remains to be fully elucidated. HEK 293 cells do not normally express TLR-3, and dsRNA can activate IRF-3 in these cells when TLR-3 is ectopically expressed (1, 14). HEK 293 cells harboring the HCV replicon that we had generated are sensitive to dsRNA (2), and they provided a unique tool to examine the role of IRF-3 in mediating the inhibition of HCV RNA replication.
dsRNA does not induce the nuclear localization of IRF-3. The nuclear localization of IRF-3 was examined with a green fluorescent protein (GFP)-IRF-3 construct that fused the GFP coding sequence to the N-terminal portion of IRF-3 in a mammalian expression vector. The GFP-IRF-3 construct was transfected into 293 and Huh-7 cells using FuGene-6 (Roche). The cells were then treated either with 20 hemagglutination units (HAU) of Sendai virus (SenV) per ml (17) or with a dose of two dsRNA mimics, poly(I · C) and poly(U · A), ranging from 50 to 500 µg per ml and then fixed, permeabilized, and stained with DAPI (4',6'-diamidino-2-phenylindole). Transfection of dsRNA into the cells was unnecessary, as adding poly(I · C) to cell culture media strongly inhibits HCV RNA replication in 293 cells (2). In untreated cells, IRF-3 localized primarily to the cytoplasm. As expected, treating S22.3 and LIP3 cells (Huh-7 and 293 cells that harbor the subgenomic replicon, respectively) with SenV and/or poly(I · C) failed to induce the nuclear localization of IRF-3 in either cell line, consistent with the known suppression of IRF-3 activation by the NS3/NS4A protease (9) (Fig. 1A). Treating naïve 293 and Huh-7 cells with SenV induced clear nuclear localization of IRF-3, suggesting that the activation of IRF-3 is not defective in either cell type, even in Huh-7 cells that do not respond to treatment with dsRNA in inhibiting the replicon RNA (Fig. 1B and C). In contrast to the results of treatment with SenV, naïve cells treated with 50 to 500 µg of dsRNA per ml showed no effect on IRF-3 localization; extending the treatment with 250 µg of dsRNA per ml for up to 72 h also had no effect on the nuclear localization of IRF-3 [Fig. 1B and C depict representative poly(I · C) concentrations and time points]. The activation of endogenous IRF-3 was also examined using indirect immunofluorescence staining and was found to correlate to the activation pattern of the GFP-IRF-3 reporter. Only SenV, and not poly(I · C), treatment induced the nuclear localization of endogenous IRF-3 (data not shown). Therefore, it appears that dsRNA does not induce the activation of IRF-3 in either naïve or replicon 293 and Huh-7 cell lines.
![]() View larger version (36K): [in a new window] |
FIG. 1. Induction of IRF-3 nuclear localization by dsRNA. (A) Huh-7 or 293 cells harboring the HCV replicon, i.e., S22.3 and LIP3 cells, respectively, were transfected with a GFP-IRF-3 expression vector and then treated with 20 HAU of SenV per ml or 250 µg of poly(I · C) per ml for 24 h, and the localization of IRF-3 was visualized by fluorescence microscopy. (B) The GFP-IRF-3 construct was transfected into Huh-7 cells. The cells were then treated with 20 HAU of SenV per ml, with the indicated amounts of the dsRNA mimic poly(I · C) or poly(A · U) for 24 h (Dose), or with 250 µg of dsRNA per ml for the indicated durations (Time). (C) 293 cells were treated like the Huh-7 cells described above. DAPI staining is presented to the right sides of selected panels.
|
![]() View larger version (16K): [in a new window] |
FIG. 2. Induction of IRF-3-responsive reporter genes by dsRNA. Luciferase reporter genes regulated by ISRE (A) or the RANTES promoter (B) were transfected into 293 cells. Cells were treated with 0, 5, 20, and 50 HAU (first to last bars, respectively) of SenV per ml or 0, 50, 250, and 500 µg (first to last bars, respectively) of poly(I · C) and poly(A · U) (A) or poly(I · C) and poly(G · C) per ml (B). The averages of results of three independent experiments are presented.
|
N, which we used to examine the role of IRF-3 in mediating the dsRNA-induced inhibition of HCV RNA replication. Flag-tagged IRF-3
N was constructed and transfected (Lipofectamine 2000; Invitrogen) into 293/FRT cells (Flp-In system; Invitrogen). A stable cell line, 2IN, was generated by selecting cells with 0.2 mg of hygromycin per ml. The stable expression of truncated IRF-3 was confirmed in 2IN cells by using IRF-3- and Flag-specific Western blotting (Fig. 3A). The level of expression of IRF-3
N in these cells was significantly higher than that of the endogenous protein (Fig. 3A), a prerequisite to eliciting the dominant negative effect (compare lanes 293 and 2IN-4).
![]() View larger version (30K): [in a new window] |
FIG. 3. Establishment of dominant negative IRF-3 N 293 cell lines. (A) Stable expression levels of Flag-IRF-3 N ( N) were examined using Western blotting of 2IN cells, 293 cells transiently overexpressing IRF-3 (Transient), stable cells lines generated by transfecting an empty vector (2pd), or stable cells lines expressing wild-type (WT) IRF-3 (2IW-5 and 2IW-2). Proteins were detected with an IRF-3-specific polyclonal antibody (upper panel) or a Flag tag-specific monoclonal antibody ( -Flag; lower panel). Molecular size markers (in kilodaltons) are noted at the left. (B) 293, LIP3, or 2IN cells were transfected with the ISRE or RANTES luciferase reporter genes and were treated with SenV as described in the legend for Fig. 2.
|
N was confirmed by using the ISRE and RANTES luciferase reporters in naïve 293, LIP3, and in 2IN cells. Transfecting the three cell lines as described above and treating them with 5 to 50 HAU of SenV per ml revealed that, relative to the induction of luciferase activity in naïve 293 cells, IRF-3 activation was suppressed in LIP3 replicon cells due to the presence of the NS3/NS4A protease (9) and was diminished in 2IN cells due to the expression of IRF-3
N (Fig. 3B). This result confirms that the dominant negative form interferes with the function of endogenous IRF-3, as previously reported (12, 17). Basal luciferase activities were comparable in luciferase reporter-transfected cells expressing IRF-3
N (2,776 ± 95 cps) and wild-type IRF-3 (1,895 ± 274 cps), indicating that the transfection procedure did not induce the activation of IRF-3. Dominant negative IRF-3 does not affect HCV RNA replication. Transfection into naïve 293 cells of total RNA obtained from 293 cells harboring the replicon generates approximately 7.5 x 103 neomycin-resistant colonies per µg of replicon RNA (2). In order to assess the effect of the stable expression of dominant negative IRF-3 on the efficiency of colony formation, 20 µg of total RNA isolated from 293Rep17 cells containing an estimated 107 replicon copies per µg was transfected as described previously (2) into 2PD cells (equivalent to naïve 293 cells) (stable cells were generated by transfecting the empty Flp-In vector), 2IW cells (293 cells stably overexpressing wild-type IRF-3), or 2IN cells. Selecting the cells with 0.6 µg of neomycin per ml, staining the cells with crystal violet, and quantifying the resulting colonies indicated that there was no measurable effect on the number of replicon colonies generated with 2IN cells compared to the number of 293 or 2IW cells in repeated experiments (Fig. 4A). Therefore, IRF-3 does not appear to mediate an antiviral response expected to be activated by the presence of replicon RNA in the cells.
![]() View larger version (30K): [in a new window] |
FIG. 4. Dominant negative IRF-3 N has no effect on stable replicon establishment or inhibition by dsRNA. (A) Stable 293 cells expressing dominant negative IRF-3 (2IN) or wild-type IRF-3 (2IW) or stable cells that were generated by introducing an empty vector (2PD) were transfected with either total RNA isolated from replicon-containing 293Rep17 cells, an in vitro-transcribed R3 replicon, or replication-incompetent mutant RNA (GDD). Cells were selected for 3 weeks, and the numbers of colonies generated were quantified (shown in the lower right corners). The panel is representative of two separately performed experiments. (B) Inhibition of HCV RNA replication by dsRNA (3.9 to 1,000 µg/ml) measured by TaqMan real-time reverse transcription-PCR of LIP3 cells or stable 293 cells expressing dominant negative IRF-3 and harboring the replicon (2NR cells). Results are the averages of three independent determinations.
|
N phenotype in these expanded cell lines was confirmed (data not shown).
IRF-3 does not modulate HCV RNA inhibition by dsRNA.
dsRNA potently inhibits HCV RNA replication in 293 cells harboring the replicon (IC50, 77 µg/ml) (2). Treating naïve 293, LIP3, and 2NR cells (replicon 293 cells stably expressing IRF-3
N) with a 3.9- to 1,000-µg/ml dose of poly(I · C) for 72 h revealed that the stable overexpression of dominant negative IRF-3
N did not desensitize the cells and shift the IC50 of dsRNA. The levels of inhibition of HCV RNA replication were comparable in LIP3 and 2NR cells, yielding IC50s of 63.5 ± 8.9 µg/ml for LIP3 cells and 44.7 ± 3.1 µg/ml for 2NR cells, as measured by TaqMan real-time reverse transcription-PCR (2) (Fig. 4 B). Therefore, IRF-3 does not appear to mediate an essential role in the inhibition of HCV replicons by dsRNA.
Our combined data examining IRF-3 subcellular localization, transactivation, and dominant negative suppression indicate that IRF-3 is not activated by topical dsRNA treatment and that the inhibitory response to dsRNA on HCV RNA replication observed in 293 cells is IRF-3 independent.
A recent report by Edelmann et al. (7) questioned the role of TLR-3 as the sole recognition element mediating antiviral responses to dsRNA, and a study by Hoebe et al. (13) reported TLR-3/TRIF axis-independent antiviral responses to dsRNA. Therefore, it appears that in addition to the reported TLR-3/TRIF/IRF-3 antiviral pathway (1, 17, 22, 26), there may be other independent mechanisms that warrant a detailed examination of alternative routes that might mediate the recognition of and response to dsRNA, particularly with regard to HCV RNA replication.
S.A. is a recipient of the Canadian Institutes of Health Research Rx&D postdoctoral research award.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»