Intracellular expression of IRF9 Stat fusion protein overcomes the defective Jak-Stat signaling and inhibits HCV RNA replication

IFN-α resistant cells containing sub-genomic HCV genotype 1b RNA were generated in our laboratory by prolonged treatment of the low inducer replicon cell line Con-15-3 with IFN-α as described previously [12]. IFN sensitive S9-13 cells containing sub-genomic HCV genotype 1b HCV RNA were kindly provided by Ralf Bartenschlager, University of Heidelberg, Heidelberg, Germany. Stable cell lines, S3-GFP (IFN sensitive) and R4-GFP (IFN resistant) containing HCV genotype 2a subgenomic RNA with a green fluorescence protein (GFP) fusion in frame with HCV NS5A were prepared in our laboratory as previously described [13]. All cell lines were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 2 mM L-glutamine, nonessential amino acids, 100 U/ml of penicillin, 100 μg/ml of streptomycin, 500 μg/ml of G418, and 5% fetal bovine serum.

Three different expression plasmid constructs (Fig. 1) were used in this study and assessed for their ability to overcome defective Jak-Stat signaling and to induce an antiviral state in R15-3 cells. The first plasmid, pcDNA3-IRF9 contains the open reading frame of the human IRF9 gene under the control of a cytomegalovirus (CMV) promoter. The second plasmid, IRF9-S1C contains the 38 C-terminal amino acids of Stat1 protein fused in frame with the human IRF9. The third plasmid construct, IRF9-S2C contains the 104 C-terminal amino acids of Stat2 fused in frame with the human IRF9 coding sequence. All three recombinant plasmid constructs were obtained as a gift from Curt Horvath, Feinberg School of Medicine, Northwestern University, Evanston, Illinois [14]. The ISRE-Luciferase plasmid was obtained from Steve Goodbourn, St.George's Hospital, University of London, London, UK [15]. The pRL-TK plasmid was obtained from Promega Corporation, Madison, WI. The Stat1-GFP and IRF9-GFP constructs were obtained from Addgene Inc, Cambridge, MA [16]. The Stat2-GFP construct was obtained from Dr. Mario Koster Braunschweig, Germany [17]. DNA transfections were performed using Fugene-6 transfection reagent following the manufacturer's recommendations (Roche Molecular Diagnostics, Indianapolis, IN).

The R15-3 cell line was transfected using either IRF9, IRF9-S1C, or IRF9-S1C plus IRF9-S2C plasmid, ISRE firefly-luciferase (FL) plasmid, pRL-TK renilla-luciferase (RL) plasmid and treated with or without IFN-α (Schering, Kenilworth, NJ). At 24 hours post-transfection FL and RL activity was measured by integrating the total light emission over ten seconds with a luminometer (Luman LB9507, EG&G Bethold, Berlin, Germany). The results were normalized by dividing the FL value with the RL value for the same sample.

R15-3 and 9-13 cells were were plated in a two well Lab-Tek chamber slide (Electron Microcopy Sciences, Hatfield, PA) at a density of 5 × 10⁴ cells per ml. Twenty four hours later, cells were transfected with 1 μg of the respective Stat1-GFP, Stat2-GFP or IRF9-GFP plasmid DNA using Fugene-6 transfection reagent. At 48 hours post- transfection, To-Pro3 nuclear marker (Invitrogen, Molecular Probes, Oregon) was added to the samples at 1 μg/ml, and incubated for five minutes in PBS. IFN-α (1000 IU/ml) was then added to the appropriate groups. Confocal microscopy was performed using a Leica TCS SP2 confocal microscope equipped with three lasers (Leica Microsystems, Exton, PA). Optical slices were collected at 512 × 512 pixel resolution. NIH Image version 1.62 and Adobe Photoshop version 7.0 were used to assign correct colors of channels collected, including the Green Fluorescent Protein (green), To-Pro3 633 (far red), and the differential interference contrast image (DIC) (gray scale).

R15-3 and 9-13 cells were transfected with IRF9, IRF9-S1C, IRF9-S2C, or IRF9-S1C plus IRF9-S2C plasmid. At 48 hours post-transfection the cultured cells were suspended in 100 μl of PBS and 20 μl of phycoerythrin conjugated mouse anti-human HLA-ABC antibody (Pharmingen, San Jose, CA) and incubated for 15 minutes at 4°C. The cells were then resuspended in 500 μl of PBS and analyzed by a BD LSR-II flow cytometer (Becton-Dickinson, Franklin Lakes, New Jersey) using BD FACS Diva software.

The antiviral properties of each IRF9 construct was evaluated by RPA detection of the HCV RNA negative sense strand utilizing a previously described method [15]. R15-3 cells were transfected with IRF9, IRF9-S1C, IRF9-S2C, or IRF9-S1C plus IRF9-S2C plasmid in the presence or absence of IFN-α (1,000 IU/ml). At 72 hours total RNA was isolated, hybridized, and analyzed by RPA.

R15-3 cells were transfected with IRF9, IRF9-S1C, IRF9-S2C or both IRF9-S1C, and IRF9-S2C, and were treated with or without IFN-α (1,000 IU/ml). At 48 hours the cells were mounted onto a glass slide and stained for HCV NS3 utilizing a previously described method [12].

R15-3 and S9-13 cells were transfected with each of the plasmids listed in Fig. 1A. At 72 hours the cells were treated with or without IFN-α (1000 IU/ml). Forty-five minutes after IFN-α addition the cells were lysed by the addition of RIPA buffer with proteinase and phosphatase inhibitors (1 × PBS, 1% NP-40, 0.5% Deoxycholate, 0.1% SDS, 50 μg/ml PMSF, 5 μg/ml aprotinin, 5 μg/ml leupeptin, 1 μg/ml pepstatin). The GFP primary antibody (Santa Cruz Biotechnology, Santa Cruz, CA) was added to the lysate and rotated at 4°C overnight. The next morning protein A/G plus-agarose (Santa Cruz Biotechnology, Santa Cruz, CA) was added to each sample and rotated at 4°C for three hours. Western blotting was then performed using a Phospho-Stat1 Y701 (p-Stat1) primary antibody (Cell Signaling Technology, Danvers, MA) diluted (1:1,000) in blocking reagent, (0.1% tween 20, Tris buffered saline, and 5% NFDM) as described below.

Cell lysates were prepared from IRF9, IRF9-S1C, IRF9-S2C, or both IRF9-S1C, and IRF9-S2C transfected R15-3 cells at 72 hours post-transfection. Western blot analysis was then performed using HCV NS5B (Abcam, Cambridge, MA), p-eIF-2α (Cell Signalling Technologies, Danvers, MA) and β-actin (Cell Signalling Technologies, Danvers, MA) primary antibodies.

The viability of IRF9 fusion construct transfected R15-3 cells was evaluated by the MTT assay. At 72 hours post-transfection 100 μl of the MTT solution (Sigma, St. Louis, MO) was added to the media in each well. After three hours of incubation at 37°C the media was aspirated and 1 ml of solubilization buffer (0.1 N HCL in absolute isopropanol) was added to each well. The absorbance of each sample was then measured at 570 nm utilizing a Beckman DU Spectrophotometer after blanking. The viability was then calculated by the formula dividing the OD₅₇₀ of each sample by the mean OD₅₇₀ of the controls.

The GFP fused Stat1, Stat2 and IRF9 plasmid clones illustrated in Fig. 1A were used to establish the dynamics of nuclear translocation in the S9-13 and R15-3 cells in the presence and absence of IFN-α. Using a transient transfection experiment, we demonstrate that both the Stat1-GFP and Stat2-GFP fusion proteins were expressed at high levels in the cytoplasm of both cell lines. Both the Stat1-GFP and Stat2-GFP fusion protein translocated to the nucleus of S9-13 cells 30 minutes after IFN-α treatment and returned to the cytoplasm by 24 hours (Fig. 2). However, the Stat1-GFP and Stat2-GFP proteins did not localize to the nucleus of the R15-3 cell line following IFN-α treatment. These results suggest that the nuclear translocation of the Stat1 and Stat2 proteins were impaired in the R15-3 cells. The IRF9-GFP fusion protein was expressed at intermediate levels in the S9-13 and R15-3 cell lines. The distribution of IRF9-GFP chimera protein was localized in both the cytoplasm and nucleus at all time points in both the R15-3 and S9-13 cell lines. These results suggest that the IRF9 protein diffuses freely from cytoplasm to the nucleus of both cell lines independent of IFN-α treatment. To understand the reason for differences in the nuclear translocation of Stat1 and Stat2-GFP fusion proteins in R15-3 and S9-13 cells, their phosphorylation status was examined after transfection. The co-immunoprecipitation experiments show that the Stat1-GFP and Stat2-GFP proteins were phosphorylated in the S9-13 cells but not in the R15-3 cells (Fig. 3).

Since the IRF9-GFP construct efficiently localized to the nucleus of the R15-3 cells, we wanted to test whether fusing the TAD of either the Stat1 or Stat2 protein could induce IFN promoter activation. For this reason, expression plasmid constructs of IRF9 fused with the TAD of either Stat1 or Stat2 (Fig. 1B) were tested for their ability to activate the ISRE promoter in the R15-3 cell line. Intracellular expression of IRF9 after plasmid DNA transfection did not activate the ISRE-luciferase activity in the R15-3 cell line even after IFN-α treatment (Fig. 4), whereas the IRF9-S1C and IRF9-S2C fusion constructs significantly induced the ISRE-luciferase promoter (p < .025). The combination treatment with both IRF9-S1C and IRF9-S2C also showed high ISRE-luciferase activity in the R15-3 cells. There were no significant differences (p < .05 student's t-test) in ISRE-luciferase activity between the IFN-α (+) and (-) groups suggesting that the ISRE-luciferase induction was IFN independent. These results suggest that the fusion of the TAD of either Stat1 or Stat2 to IRF9 efficiently localized to the nucleus and activated the ISRE promoter in the R15-3 cell line. The results of the ISRE promoter induction experiments were confirmed by examining the HLA-ABC (HLA-1) surface expression of IRF9-Stat fusion transfected R15-3 cells by flow cytometry at 72 hours post-transfection. HLA-1 is an important molecule induced by the ISRE promoter which is involved in the immune recognition of virally infected cells [8]. Intracellular expression of either IRF9-S1C or IRF9-S2C induced HLA-1 surface expression in the R15-3 cells (Fig. 5A). The change in surface HLA-1 expression in R15-3 cells following plasmid transfection was quantified and compared to the fold change in HLA-1 surface expression of untransfected R15-3 cells after the addition of IFN-α (Fig. 5B). Significant increases (p < .05) in HLA-1 expression were observed for IRF9-S1C, IRF9-S2C and IRF9-S1C plus IRF9-S2C transfected R15-3 cells. In the control S9-13 cells, the expression of HLA-1 was induced after IFN-α treatment.

The ability of the different IRF9-Stat fusion plasmids to inhibit viral replication was examined using stable replicon cell lines containing either HCV genotype 1b or HCV 2a sub-genomic RNA. RPA detection of the HCV negative strand RNA replication intermediate was used to assess viral replication in IRF9-Stat fusion transfected R15-3 cells. The RPA results demonstrate that intracellular expression of IRF9 alone had no effect upon HCV negative strand RNA levels, whereas all of the IRF9-Stat fusion constructs showed antiviral activity in an IFN-α independent manner (Fig. 6A). Transfection with IRF9-S1C reduced HCV negative sense RNA strand levels, and IRF9-S2C alone or in combination with IRF9-S1C inhibited HCV replication below the limit of detection of the RPA assay. The HCV negative strand RNA RPA results were verified by quantifying HCV positive strand RNA by real-time RT-PCR. The results in Fig. 6B demonstrate that R15-3 cells transfected with IRF9-S1C, IRF9-S2C, or the combination experienced a significant reduction (p < .05) in log copies of HCV RNA per μg of cellular RNA compared to R15-3 cells plus IFN-α (1,000 IU/ml). IRF9 alone transfected R15-3 cells and showed a modest reduction of HCV RNA. S9-13 cells with and without IFN-α (1,000 IU/ml) were used as a positive control.

Viral protein expression was assessed by immunocytochemical staining for HCV NS3 protein. The results presented in Fig. 7A suggest that the transfection of R15-3 cells using the individual constructs IRF9-S1C, IRF9-S2C or the combination reduced HCV NS3 protein expression in an IFN-α independent manner. The IRF9 alone-transfected R15-3 cells showed no reduction in viral protein levels while the IRF9-S1C, IRF9-S2C and the combination of both transfected R15-3 cells displayed a reduction of HCV NS3 expression (Fig. 7A). The addition of IFN-α had no effect on viral protein levels in each experimental group. Western blot analysis was then performed using an antibody against HCV NS5B to verify the results of the antiviral effect of the IRF9-Stat fusions. IRF9-S2C and IRF9-S1C plus IRF9-S2C reduced HCV NS5B protein levels in R15-3 cells (Fig. 7B). R15-3 cells transfected with IRF9 alone showed no reduction in HCV NS5B, while the IRF9-S1C transfected cells showed weak reduction in viral protein expression. All constructs exerted their antiviral activity in an IFN independent manner. The antiviral properties of the IRF9-Stat fusion constructs appear more robust at the 72 hour time point as compared to the previous immunocytochemical assessment of HCV NS3 at 48 hours. This increase in antiviral efficacy between these time points correlates with our experience with the antiviral activity of IFN-α in the S9-13 cell line. The antiviral effect of IRF9-Stat fusion protein expression in the HCV 2a-GFP resistant replicon cells (R4-GFP) was examined directly via fluorescence microscopy and then quantified by flow analysis (Fig. 8). It was determined that R4-GFP cells expressing IRF9 alone had no antiviral effects whereas all the remaining constructs had potential anti-HCV activity as they all inhibited HCV-GFP expression to various degrees. The strongest antiviral activity was observed in the IRF9-S2C transfected cells. The viability of the R15-3 cell line expressing different IRF9-Stat fusion constructs was then examined to exclude the possibility that the enhanced viral clearance was not due to the toxic effect of intracellular expression of the IRF9-Stat fusion proteins. The results presented in Fig. 9 suggest that no cellular toxicity was associated with the expression of the Stat fusion proteins in the R15-3 cells. Taken together, these results suggest that the intracellular expression of IRF9-S1C and IRF9-S2C leads to effective clearance of HCV RNA replication in an IFN-α resistant replicon cell line of both HCV1b and HCV2a genotype viruses.

Next, the mechanism of inhibition of viral RNA replication in cells expressing the IRF9-Stat fusions was examined. A principal mechanism that contributes to the antiviral state and which is activated upon ISRE promoter induction is the phosphorylation of eIF2α. Phosphorylation of eIF2α leads to translational inhibition within the cell thereby preventing viral replication. Therefore, we assessed the ability of each construct to induce p-eIF2α both in the presence and absence of IFN-α. The result shown in Fig. 10 suggests that IRF9 alone was unable to induce p-eIF2α, while IRF9-S1C caused low-level induction of p-eIF2α. Transfection of IRF9-S2C alone and IRF9-S1C plus IRF9-S2C caused high-level p-eIF2α induction. The addition of IFN-α did not significantly (p < .025) alter the expression patterns of p-eIF2α.