Hepatitis C Virus Nonstructural Protein 5A Inhibits MG132-Induced Apoptosis of Hepatocytes in Line with NF-κB-Nuclear Translocation

Background Hepatitis C virus (HCV) infection is one of the major causes of cirrhosis and hepatocellular carcinoma. HCV nonstructural protein 5A (NS5A) is an attractive antiviral target and plays an important role in HCV replication as well as hepatocarcinogenesis. The aim of this study was to assess the effect of HCV NS5A protein in the abrogation of apoptotic cell death induced by the proteasome inhibitor MG132. Methods Apoptotic responses to MG132 and the expression of molecules involved in NF-κB signaling pathways in human hepatocytes were investigated with or without the expression of HCV NS5A. Results HCV NS5A protected HepG2 cells against MG132-induced apoptosis, in line with NF-κB-nuclear translocation. A similar NF-κB-nuclear translocation was observed in Huh7 cells infected with HCV JFH1. In agreement with this, after treatment with MG132, HCV NS5A could elevate the transcription of several NF-κB target genes such as BCL2 and BCLXL to inhibit MG132-induced apoptosis in hepatocytes. HCV HCV NS5A also enhanced phosphorylation of IκBα. Consistent with a conferred prosurvival advantage, HCV NS5A reduced MG132-induced poly(adenosine diphosphate-ribose) polymerase cleavage. Conclusions HCV NS5A expression enhances phosphorylation of IκBα, liberates NF-κB for nuclear translocation and downregulates MG132-induced apoptotic pathways in human hepatocytes. It is possible that the disruption of proteasome-associated apoptosis plays a role in the pathogenesis of HCV infection.


Introduction
Hepatitis C virus (HCV) is one of the major risk factors for hepatocellular carcinoma (HCC) [1,2]. HCV is a global health issue, with more than 185 million individuals chronically infected worldwide [3]. Although HCV drug treatment regimens have depended on HCV genotypes [4], these regimens have been improved, and multiple direct-acting antivirals (DAAs) therapies will soon become globally realistic options [3].
The proteasome inhibitor MG132 has been shown to induce apoptosis in HepG2 cells [15,16]. MG132 activates c-Jun N-terminal kinase (JNK), which initiates apoptosis and also inhibits NF-κB activation [17]. MG132 dramatically sensitizes HDAC-inhibitor-mediated apoptosis, at least partly, through the ER stress response [18]. HCV NS5A modulates JNK and activates NF-κB [19,20]. Alterations in cell survival contribute to the pathogenesis of human liver diseases and viral carcinogenesis [21,22]. In the present study, we examined the effect of HCV NS5A protein in abrogating apoptotic cell death induced by MG132. Our results demonstrate that HCV NS5A protein suppresses MG132-mediated apoptosis.

RNA Purification, Real-time RT-PCR and Human NF-κB Signaling Targets PCR Array
Cellular RNA was extracted using the RNeasy Mini Kit (Qiagen). One microgram of RNA was reverse-transcribed with a PrimeScript RT 2 First Strand Kit (Qiagen). PCR was performed on cDNA templates using primers specific for B-cell CLL/lymphoma (BCL2), BCL2-like 1 (BCL2L1/BCL-X/BCLXL) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which were purchased from Qiagen. A Human NF-κB Signaling Targets Real-time RT-PCR Array was performed according to the manufacturer's protocol. The data were analyzed by PCR Array Data Analysis Software (http://www.sabiosciences.com/pcrarraydataanalysis.php).

Western Blot Analysis
Cells were harvested with sodium dodecyl sulfate sample buffer. After sonication, cell lysates were subjected to electrophoresis on 5-20% polyacrylamide gels and transferred onto polyvinylidene difluoride membranes (ATTO). Membranes were probed with specific antibodies for poly(adenosine diphosphate-ribose) polymerase (PARP), nuclear factor of kappa light polypeptide gene enhancer in B cells inhibitor, alpha (IκBα), phosphorylated IκBα (Ser32) (Cell Signaling Technology, Danvers, MA, USA) and GAPDH (Santa Cruz Biotechnology, Santa Cruz, CA, USA). After washing, membranes were incubated with secondary horse-radish peroxidase-conjugated antibodies. Signals were detected by means of enhanced chemiluminescence (GE Healthcare Japan, Tokyo, Japan) and scanned with an image analyzer LAS-4000 and Image Gauge (version 3.1) (Fuji Film, Tokyo, Japan). Band intensities were determined by Ima-geJ software [26].

Transfection and Apoptosis Assay
Approximately 1 x 10 5 HepG2 cells were placed on 6-well tissue culture plates (Iwaki Glass, Tokyo, Japan) 24 hours prior to transfection. Cells were transfected with 0.3 μg of pCXN2 or pCXN2-NS5A vector using Effectene transfection reagents (Qiagen) according to the manufacturer's protocol. After 48 hours of transfection, cells were treated with 0-10 μM MG132 (Sigma-Aldrich) for 24 hours, and APOPercentage Apoptosis Assay (Biocolor, Belfast, Northern Ireland) was used to quantify apoptosis according to the manufacturer's instructions. The transfer and exposure of phosphatidylserine to the exterior surface of the membrane has been linked to the onset of apoptosis. Phosphatidylserine transmembrane movement results in the uptake of APOPercentage dye by apoptotic cells. Purple-red stained cells were identified as apoptotic cells by light microscopy. The number of purple-red cells per 300 cells was counted as previously described [14].

Immunofluorescence Study
Cells were washed and fixed with 3.7% formaldehyde, followed by blocking with 3% horse serum albumin. Cells were incubated with an NF-κB P65 (D14E12) antibody (Cell Signaling) for 16 hours at 4°C. Cells were washed and incubated with anti-rabbit immunoglobulin secondary antibody conjugated with Alexa Fluor 555 (Cell Signaling) for 1 hour at room temperature. Nuclear staining was performed with Hoechst 33342, trihydrochloride, trihydrate (Molecular Probes, Eugene, OR, USA). Finally, cells were washed and mounted for confocal microscopy (ECLIPSE TE 2000-U, Nikon, Tokyo, Japan), and the images were superimposed digitally to allow for fine comparisons.

Statistical Analysis
Results were expressed as mean ± standard deviation (SD). Statistical analyses were performed using Student's t-test. A P-value of < 0.05 was considered statistically significant. All statistical analyses were performed using DA Stats software (O. Nagata, Nifty Serve: PAF01644).

MG132 reduced cell viabilities in HepG2 control cells more than in HepG2-NS5A cells
HepG2 cells stably transfected with HCV NS5A (HepG2-NS5A) were used to examine its effects on MG132-mediated cell death. We treated HepG2-NS5A and HepG2 control cells with MG132 at various concentrations and analyzed cell death 48 hours later (Fig 1A and 1B). Compared with HepG2-NS5A cells, a much higher level of cell death was observed in HepG2 control cells treated with 10 μM MG132.

HCV NS5A reduced MG132-induced apoptosis
The quantification of apoptosis showed a significant 4.6-fold or 2.5-fold increase in apoptosis in HepG2 cells treated with 1 μM or 10 μM MG132 compared with HepG2-NS5A cells treated with 1μM or 10 μM MG132, respectively (Fig 1C and 1D). We also investigated whether HCV NS5A interfered with apoptosis, using western blot analysis, to detect PARP cleavage, which is thought to be one of the hallmarks of apoptosis. Whereas PARP of HepG2-NS5A cells was expressed at higher levels, cleaved PARP induced by 10 μM MG132 was observed more strongly in HepG2 control cells than in HepG2-NS5A cells (Fig 1E). Caspase-3/-7 activities measured in the presence of MG132 were increased in HepG2 control cells (~1.5-fold) compared with HepG2-NS5A cells. These results showed that HCV NS5A protected the human hepatoma cell line HepG2 from MG132-induced apoptosis.
HCV NS5A enhanced MG132-induced NF-κB-nuclear translocation MG132, a compound that can inhibit the activation of NF-κB, initiates apoptosis [17,27]. To further analyze the underlying mechanisms of this process, we compared the localization of NF-κB p65 in HepG2 control cells with that in HepG2-NS5A cells in the presence or absence of 5 μM MG132 using immunofluorescence (Fig 2A and 2B). After 24 hours, cells were stained with rabbit monoclonal NF-κB p65 antibody. NF-κB nuclear localization was more readily detected in HepG2-NS5A cells than in HepG2 control cells in the presence and absence of 5 μM MG132 (Fig 2A and 2B). In accordance with these results, NF-κB-nuclear translocation was observed in Huh7 infected with HCV JFH1 (Fig 2C). A reporter assay also revealed that NF-κB transcriptional activity measured in the presence of MG132 was increased in HepG2-NS5A cells (~4.1-fold) compared with HepG2 control cells. We treated both cell lines, HepG2 control and HepG2-NS5A with 0, 5, and 10 μM MG132 for 8 hours, collected samples and subjected them to SDS-PAGE and western blotting by antibody specific for phosphorylated IκBα, IκBα or GAPDH (Fig 2D). Band intensity rates of phosphorylated IκBα/total IκBα were 1.0-, 1.19-and 0.87-fold in HepG2 control, and they were 1.41-, 1.33-, and 1.18-fold in HepG2-NS5A cells. Together, our data demonstrated that HCV NS5A may enhance NF-κB activation, resulting in reduced MG132-induced apoptosis in hepatocytes.
Different amino acid sequences of ISDR have no impact on MG132-induced apoptosis HCV NS5A ISDR has an association with the treatment response in interferon-including regimens against chronic HCV infection [28]. As we reported previously [14], we made HCV NS5A expression vectors, having 0 (W1 and W2), 1 (I1) or 2 (I2), or 5 amino acid changes (M1) in the HCV NS5A2209-2248 region, compared with those of HCV-J (wild type). We examined MG132-induced apoptosis after transient transfection of each HCV NS5A expression plasmid into HepG2 cells. After 24 hours, cells were treated with MG132 for 24 hours and apoptosis was evaluated as previously described (Fig 3A and 3B). HCV NS5A containing different sequences in the ISDR region have a similar effect on MG132-induced apoptosis as compared to wild type sequence. It may be possible that NS5A ISDR interacts with host protein and blocks MG132-induced apoptosis.

NF-κB-associated genes upregulated in HepG2-NS5A cells
To obtain further mechanistic insights into HCV NS5A for NF-κB-signaling pathways, we used a pathway-specific gene array to identify HCV NS5A target genes in HepG2-NS5A cells by comparison with HepG2 control cells. We extracted total RNAs from both cell lines for studying the influence of HCV NS5A on NF-κB-signaling pathway-associated gene expression using real-time PCR arrays (Tables 1 and 2). Out of 84 NF-κB-signaling pathway-associated HCV NS5A enhanced MG132-induced NF-κB p65 nuclear translocation. A, HepG2 control and HepG2-NS5Acells were cultured for 6 hours with or without 5 μM MG132. Confocal microscope high-power view demonstrates that NF-κB p65 nuclear localization was detected (x200). Nuclear staining was performed with Hoechst 33342, trihydrochloride, trihydrate (blue). Localization of NF-κB p65 was detected with anti-NF-κB p65 primary antibody, and Alexa-Fluor-555 secondary antibody (red). Merge images of A and B were superimposed digitally to allow for fine comparisons. NF-κB p65 nuclear translocation was observed (pink). B, The number of NF-κB p65 nuclear-translocated cells per field was counted at low-power view (40X). C, Huh7 cells infected with or without HCV JFH1 genotype 2 strain were incubated for 6 hours with or without 5 μM MG132. Nuclear staining was performed with Hoechst 33342, trihydrochloride, trihydrate (blue). NF-κB p65 nuclear localization was detected with anti-NF-κB p65 primary antibody and anti-rabbit Alexa-Fluor-555 secondary antibody (red). HCV was detected using anti-HCV core primary antibody and an Alexa-Fluor-488 anti-mouse secondary antibody (green). D, HCV NS5A expression enhances phosphorylation of IκBα. Western blot analyses of phosphorylated IκBα (Ser32) (P-IκBα), IκBα and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression in HepG2 control or HepG2-NS5A cells at 8 hours after treatment with or without MG132.   HepG2-NS5A cells compared with HepG2 control cells (n = 3, p <0.05) ( Table 1). Of interest, chemokine genes, such as CCL5, CCL2 and CXCL10, could activate human hepatic stellate cells, which are associated with hepatic fibrosis [29][30][31], and CXCL10 is an important gene for the treatment response of interferon treatment [32]. CCL5, CCL2 and CXCL10 have anti-apoptotic functions [33][34][35]. CSF1 and SNAP25 were associated with the development and differentiation of myeloid cells and the nervous system, respectively. CFB is classified as one of the innate immune response genes.

Discussion
In the present study, the MG132-induced apoptotic pathway is inhibited in HCV NS5Aexpressing HepG2 cells by activation of NF-κB. HCV NS5A influences NF-κB-targeting gene expression and blocks PARP cleavage and apoptosis induced by MG132, leading to hepatocyte survival. The present study supported the previous observation that MG132 reduced the viability of HepG2 cells in a time-and dose-dependent manner and that the effect was in tight connection with the induction of apoptosis [15]. Emanuele et al. [15] also reported that MG132 caused the degradation of PARP and that protease inhibitors could have potential as a treatment against HCC.
NF-κB activation was blocked by either adenovirus-mediated overexpression of Iκ0042α suppressor or pretreatment with MG132 in lung cancer cells [40]. NF-κB activation confers resistance to TNF-mediated apoptosis in hepatocytes, as TNF is one of the important cytokines for eradication of HCV from the liver [41]. HCV NS5A could play a role in the activation of NF-κB. In the present study, TNF expression was also suppressed in HepG2-NS5A cells, compared with HepG2 control cells (Table 2). These results are in agreement with the previous reports of HCV NS5A-transgenic animals [41]. NF-κB is also critical for the apoptosis of HCC and plays a role in the sensitivity to sorafenib [42]. Further studies will be needed for this issue.
Intracellular protein degradation is an important mechanism, which includes ubiquitinproteasome and ubiquitin-independent proteasome pathways for the modulation of certain proteins and the elimination of damaged proteins. MG132 effectively blocks the proteolytic activity of the 26S proteasome complex [43]. It is evident that several viruses are able to manipulate the ubiquitin-proteasome pathway by redirecting the cellular ubiquitin machinery to enable replication. Growing evidence suggests that ubiquitin-proteasome and ubiquitinindependent proteasome pathways are also involved in controlling the stability of HCV proteins such as core [43,44], p7 [45], NS2 [46] and NS5A [47]. Of interest, it has been reported that, in the presence of HCV, STAT1 and STAT3 proteins were ubiquitinated and that the degradation was blocked by MG132 [48,49]. HCV may inhibit interferon responses via proteasomal degradation of JAK/STAT pathway components [49]. It is possible that HCV NS5A may interact with intracellular protein degradation pathways [47]. Although proteasome inhibition elevates IκBα levels and leads to inhibition of NF-κB activity [50,51], we observed that HCV NS5A expression enhances phosphorylation of IκBα and liberates NF-κB for nuclear translocation in the present study (Fig 2D).
In conclusion, HCV NS5A enhances the phosphorylation of IκBα, which in turn enhances NF-κB-nuclear translocation and down-regulates MG132-induced apoptotic pathways in human hepatocytes. Together, the disruption of proteasome-associated apoptosis may play a role in the pathogenesis of HCV infection.