The DNA Damage Sensors Ataxia-Telangiectasia Mutated Kinase and Checkpoint Kinase 2 Are Required for Hepatitis C Virus RNA Replication^▿

Cell culture.

293FT cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS). The HuH-7-derived O cells harboring a replicative genome-length HCV RNA and the HuH-7-derived sO cells harboring the subgenomic replicon RNA of HCV-O were cultured in DMEM with 10% FBS and G418 (300 μg/ml geneticin; Invitrogen) as described previously (11, 14). Oc and sOc cells, which were created by eliminating HCV RNA from O cells and sO cells by interferon (IFN) treatment (11, 14), respectively, were also cultured in DMEM with 10% FBS.

RNA interference.

Oligonucleotides with the following sense and antisense sequences were used for the cloning of short hairpin RNA (shRNA)-encoding sequences targeted to Chk2 in lentiviral vector: 5′-GATCCCCGGGGGAGAGCTGTTTGACATTCAAGAGATGTCAAACAGCTCTCCCCCTTTTTGGAAA-3′ (sense) and 5′-AGCTTTTCCAAAAAGGGGGAGAGCTGTTTGACATCTCTTGAATGTCAAACAGCTCTCCCCCGGG-3′ (antisense). The oligonucleotides above were annealed and subcloned into the BglII-HindIII site, downstream from an RNA polymerase III promoter of pSUPER (5), generating pSUPER-Chk2i. To construct pLV-Chk2i, the BamHI-SalI fragments of the pSUPER-Chk2i were subcloned into the BamHI-SalI site of pRDI292, an HIV-1-derived self-inactivating lentiviral vector containing a puromycin resistance marker allowing for the selection of transduced cells (4). pLV-ATMi, pLV-ATRi, and pLV-PARP-1i were constructed as described previously (1).

Lentiviral vector production.

The vesicular stomatitis virus G protein (VSV-G)-pseudotyped HIV-1-based vector system has been described previously (24). The lentiviral vector particles were produced by transient transfection of the second-generation packaging construct pCMV-ΔR8.91 (30) and the VSV-G envelope plasmid pMDG2 as well as the lentiviral vector into 293FT cells with FuGene6 (Roche Diagnostics, Mannheim, Germany).

Quantitative reverse transcription-PCR analysis.

Quantitative reverse transcription-PCR analysis for HCV RNA was performed by real-time LightCycler PCR as described previously (11).

Western blot analysis.

Cells were lysed in buffer containing 50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 4 mM EDTA, 1% Nonidet P-40, 0.1% sodium dodecyl sulfate (SDS), 1 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride. Supernatants from these lysates were subjected to SDS-polyacrylamide gel electrophoresis, followed by immunoblotting analysis using anti-ATM (2C1; GTX70103 [GeneTex, San Antonio, TX]), anti-ATR (GTX70133; GeneTex), anti-Chk2 (NT; ProSci, Poway, CA), anti-Chk2 (DCS-273; Medical and Biological Laboratories, Nagoya, Japan), anti-phospho-Chk2 (Thr68) (Cell Signaling, Danvers, MA), anti-PARP-1 (C-2-10; Calbiochem, Merck Biosciences, Darmstadt, Germany), anti-hemagglutinin (HA) (HA-7; Sigma, St. Louis, MO), anti-core protein (CP-9 and CP-11; Institute of Immunology, Tokyo, Japan), anti-NS3 and anti-NS5B (no. 14; a generous gift from M. Kohara, the Tokyo Metropolitan Institute of Medical Science, Japan), anti-NS5A (no. 8926; a generous gift from A. Takamizawa, The Research Foundation for Microbial Diseases of Osaka University, Japan), and anti-β-actin (Sigma) Antibodies.

Immunofluorescence and confocal microscopic analysis.

Cells were fixed in 3.5% formaldehyde in phosphate-buffered saline (PBS) and permeabilized in 0.1% NP-40 in PBS at room temperature. Cells were incubated with anti-ATM antibody (5C2; GTX70107 [GeneTex] or PM026 [MBL]), anti-HA antibody (3F10), anti-NS5B antibody and/or anti-NS3 antibody at a 1:300 dilution in PBS containing 3% bovine serum albumin at 37°C for 30 min. Cells were then stained with fluorescein isothiocyanate (FITC)-conjugated anti-rabbit antibody (Jackson ImmunoResearch, West Grove, PA) or anti-Cy3-conjugated anti-mouse antibody (Jackson ImmunoResearch) at a 1:300 dilution in PBS containing bovine serum albumin at 37°C for 30 min. Following extensive washing in PBS, cells were mounted on slides using a mounting medium of 90% glycerin-10% PBS with 0.01% p-phenylenediamine added to reduce fading. Samples were viewed under a confocal laser-scanning microscope (LSM510; Zeiss, Jena, Germany).

Immunoprecipitation.

Cells were lysed in buffer containing 10 mM Tris-HCl (pH 8.0), 150 mM NaCl, 4 mM EDTA, 0.5% NP-40, 10 mM NaF, 1 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride. Lysates were precleared with 30 μl of protein G-Sepharose (GE Healthcare Biosciences, Uppsala, Sweden). Precleared supernatants were incubated with 5 μg of anti-HA antibody (3F10; Roche), 10 μl of anti-NS5B antibody, 5 μg of anti-Chk2 antibody (DCS-273; MBL), 5 μg of anti-FLAG antibody (M2; Sigma), or 5 μg of anti-ATM antibody (2C1) (GTX70103; GeneTex) at 4°C for 1 h. Following absorption of the precipitates on 30 μl of protein G-Sepharose resin for 1 h, the resin was washed four times with 700 μl of lysis buffer. Proteins were eluted by boiling the resin for 5 min in 2× Laemmli sample buffer. The proteins were then subjected to SDS-polyacrylamide gel electrophoresis, followed by immunoblotting analysis using anti-ATM, anti-Chk2, anti-HCV core protein (CP-9 and CP-11 mixture), anti-NS5A, anti-NS5B, anti-HA (HA-7; Sigma), or anti-NS3 antibody.

ATM and Chk2 are required for HCV RNA replication.

To determine the potential role of DNA damage sensors in HCV replication, we first used lentiviral vector-mediated RNA interference to stably knockdown ATM, ATR, PARP-1 (1), or Chk2 in the following human hepatoma HuH-7-derived cell lines: O cells harboring a replicative genome-length HCV RNA (HCV-O, genotype 1b) (11), Oc cells derived from O cells (created by eliminating genome-length HCV RNA from O cells by IFN treatment) (11), sO cells harboring the subgenomic replicon of HCV-O (14), or RSc cells that cell culture-generated HCV (HCVcc) (JFH1, genotype 2a) (29) could infect and effectively replicate (3). To express shRNAs targeted to ATM, ATR, PARP-1 (1), or Chk2, we used a VSV-G-pseudotyped HIV-1-based vector system (24). We used puromycin-resistant pooled cells 10 days after the lentiviral transduction in all experiments. Western blot analysis of the lysates demonstrated very effective knockdown of ATM, ATR, Chk2, and PARP-1 in Oc cells (Fig. 1A). The effective knockdown of ATM, ATR, Chk2, or PARP-1 in O cells or sO cells was also confirmed by Western blot analysis (data not shown). In this context, the efficiency of colony formation (ECF) in ATM- or Chk2-, but not ATR- or PARP-1-, knockdown Oc cells transfected with the genome-length HCV-O RNA with an adapted mutation at amino acid position 1609 in the NS3 helicase region (ON/C-5B K1609E RNA) (11) was notably reduced compared with the control cells (Fig. 1B) even though Chk2-knockdown cells had a slightly faster growth rate than the control cells (Fig. 1C), suggesting that both ATM and Chk2 are crucial for HCV RNA replication. To further confirm this observation, we quantitatively examined the level of HCV RNA in the O cell- or sO cell-derived knockdown cells. Consequently, we found that replication of both genome-length HCV RNA (HCV-O) and its subgenomic replicon RNA (sO) were notably suppressed in ATM- or Chk2-knockdown cells but not in ATR- or PARP-1-knockdown cells (Fig. 1D and E). Consistent with this finding, the expression levels of core and NS5B proteins were also significantly decreased in the cell lysates of ATM- or Chk2-knockdown O cells (Fig. 1F). We next examined the replication level of HCV-JFH1 in ATM-, ATR-, Chk2-, or PARP-1-knockdown RSc cells (Fig. 2A). The results revealed that RNA replication of HCV-JFH1 and release of core protein into the culture supernatants were suppressed in only ATM- or Chk2-knockdown RSc cells after inoculation with HCVcc (Fig. 2B and C). Interestingly, the release of core protein into the culture supernatant was also significantly suppressed in ATR-knockdown RSc cells, while HCV RNA replication was slightly suppressed in these cells (Fig. 2B and C), suggesting that ATR participates in the production of HCV virion.

In contrast, highly efficient knockdown of PARP-1 had no observable effects on the ECF (Fig. 1B), HCV RNA replication (Fig. 1D and E and 2B), or core protein expression in the cell lysate or in the supernatant (Fig. 1F and 2C), suggesting that our finding was not due to a nonspecific event. Thus, we have demonstrated for the first time that DNA damage sensors, ATM and Chk2, are required for HCV RNA replication.

ATM kinase activity in HCV RNA-replicating cells.

Although it has been proposed that HCV causes dsDNA breaks (18, 19), little is known about whether HCV activates or inhibits the ATM-dependent damage response pathway. In this regard, it is worth noting that we observed weak but significant Chk2 phosphorylation at threonine 68, the specific marker for ATM activation (20, 21), in the HCV RNA-replicating cells (O and sO cells) but not in the HCV-negative Oc and sOc cells (created by eliminating replicon RNA from sO cells by IFN treatment) (Fig. 3A), suggesting that the persistent HCV RNA replication stimulated the ATM-dependent DNA damage response. Furthermore, a 2-h treatment with 100 nM adriamycin, a dsDNA break inducer, markedly induced Chk2 phosphorylation in Oc, O, and sO cells (Fig. 3A). Importantly, Chk2 phosphorylation was not inhibited even in the HCV RNA-replicating cells (O and sO cells) (Fig. 3A), suggesting that the persistent HCV RNA replication and the HCV proteins are not able to suppress the ATM-dependent DNA damage response. To examine whether such a DNA damage response activates HCV RNA replication, we quantified the level of HCV RNA in the O cells treated with 100 nM adriamycin for 24 h. The results show that HCV RNA replication was increased (approximately 1.3-fold) after treatment with adriamycin (Fig. 3B), suggesting that the DNA damage response activates HCV RNA replication.

Suppression of HCV RNA replication by a small-molecule inhibitor of the ATM kinase.

We next examined the effect of a specific small-molecule inhibitor of the ATM kinase (2-morpholin-4-yl-6-thianthren-1-yl-pyran-4-one [KU-55933]) (16) on HCV RNA replication. As expected, the ATM kinase inhibitor effectively inhibited Chk2 phosphorylation after adriamycin treatment in both sO and O cells (Fig. 3C). In this context, the ATM kinase inhibitor could efficiently suppress genome-length HCV RNA replication with an in vitro 50% effective concentration (EC₅₀) of approximately 2 μM at 72 h after treatment with adriamycin (Fig. 4A). Although this ATM kinase inhibitor did not affect cell viability at 24 h after the treatment, there was a slight decrease in the cell viability at 72 h after treatment (Fig. 4B). Thus, this or other ATM kinase inhibitors may be useful for the clinical treatment of patients with chronic hepatitis C.

Interaction of HCV NS3-4A with ATM.

Since HCV NS3 has been proposed to be a viral factor involved in the induction of dsDNA breaks (18, 19), we first examined the subcellular localization of NS3-NS4A ([NS3-4A] 1B-1 or HCV-O strain) and ATM by confocal laser scanning microscopy. In most of the observed cells, ATM partially colocalized with NS3-4A in the perinuclear region and in dispersed points throughout the cytoplasm (Fig. 5A). In particular, we observed prominent colocalization of ATM with NS3-4A in some cells (Fig. 5A). Next, using anti-FLAG and anti-ATM antibodies, we immunoprecipitated lysates from 293FT cells in which FLAG-tagged ATM and either NS3-4A (HCV-O) or NS3 (HCV-O) were overexpressed and then performed immunoblotting analysis using either anti-ATM or anti-NS3 antibody to determine whether ATM binds to NS3-4A or NS3. The results revealed that ATM preferentially bound to NS3-4A over NS3 alone (Fig. 6A). Similarly, we found that ATM bound to NS3-4A using the O cell lysates (Fig. 6B), while HA-tagged Chk2 did not bind to NS3-4A in immunoprecipitation analysis using lysates from 293FT cells in which NS3-4A and HA-tagged Chk2 were overexpressed (Fig. 6C). Although NS3-4A has protease activity, ATM was not cleaved by the NS3-4A protease (Fig. 6D). Taking these results together, we conclude that ATM is able to interact with NS3-4A.

Interaction of HCV NS5B with ATM and Chk2.

We next examined the subcellular localization of ATM and/or Chk2 in HCV NS5B-expressing cells by confocal laser scanning microscopy since we previously demonstrated that HCV NS5B-expressing PH5CH8 immortalized human hepatocyte cells were susceptible to DNA damage in the form of dsDNA breaks (23). ATM partially colocalized with NS5B in dispersed points throughout the cytoplasm (Fig. 5B), similar to the subcellular localization of HCV NS3-4A and ATM. Furthermore, Chk2 also partially colocalized with NS5B in the perinuclear region and in dispersed points in the nucleus (Fig. 5C). To determine whether endogenous ATM binds to NS5B, lysates from Oc or O cells were immunoprecipitated with anti-ATM antibody, and then immunoblotting analysis using either anti-ATM or anti-NS5B antibody was performed. The results revealed that endogenous ATM bound to endogenous NS5B (Fig. 6E). Furthermore, we confirmed that ATM bound to NS5B in immunoprecipitation analysis using lysates from 293FT cells, in which NS5B (1B-1 strain) and FLAG-tagged ATM were overexpressed (Fig. 6F). Similarly, we confirmed that FLAG-tagged ATM bound to NS5B derived from O cell lysates in immunoprecipitation analysis using lysates from 293FT cells in which FLAG-tagged ATM was overexpressed (Fig. 6G). Finally, to determine which HCV protein binds to Chk2, the 293FT cell lysates of overexpressed HA-Chk2 were mixed with the O cell lysates and were immunoprecipitated with anti-HA antibody, followed by Western blot analysis using anti-HCV NS5B, anti-HCV NS5A, anti-HCV core protein, or anti-HA antibody. Consistent with the immunofluorescence result that Chk2 partially colocalized with NS5B (Fig. 5C), we observed that HA-tagged Chk2 bound to NS5B (Fig. 6H). Importantly, we found that endogenous Chk2 bound to endogenous NS5B derived from O cells (Fig. 6I). In addition, HA-tagged Chk2 bound to NS5B in immunoprecipitation analysis using lysates from 293FT cells in which NS5B (1B-1 strain) and HA-tagged Chk2 were overexpressed (Fig. 6J). Thus, Chk2 also interacted with NS5B as well as ATM. Taking these results together, we conclude that HCV targets ATM and Chk2 DNA damage sensors and that the ATM signaling pathway is required for HCV RNA replication.