Hepatocellular carcinoma (HCC) is one of the most common cancers worldwide and chronic hepatitis B virus (HBV) infection is the most common risk factor for HCC. The HBV proteins can induce oncogenic or synergy effects with a hyperproliferative response on transformation into HCC. CREBH (cAMP-responsive, element-binding protein H), activated by stress in the endoplasmic reticulum (ER), is an ER-resident transmembrane bZIP (basic leucine zipper) transcription factor that is specifically expressed in the liver. In the present study, we address the role played by CREBH activated by ER stress in HBV-induced hepatic cell proliferation. We confirmed CREBH activation by ER stress and showed that it occurred as a result of/via hepatitis B virus X (HBx)-induced ER stress. CREBH activated by HBx increased the expression of AP-1 target genes through c-Jun induction. Under pathological conditions such as liver damage or liver regeneration, activated CREBH may have an important role to play in hepatic inflammation and cell proliferation, as an insulin receptor with dual functions under these conditions. We showed that CREBH activated by HBx interacted with HBx protein, leading to a synergistic effect on the expression of AP-1 target genes and the proliferation of HCC cells and mouse primary hepatocytes. In conclusion, in HBV-infected hepatic cells or patients with chronic HBV, CREBH may induce proliferation of hepatic cells in co-operation with HBx, resulting in HCC.
Chronic infection of hepatitis B virus (HBV) is one of the major causes of liver diseases such as hepatitis, cirrhosis and hepatocellular carcinoma (HCC) . The HBV genome is a 3.2-kb, circular, partially double-stranded DNA encoding four overlapping genes: S/preS, C/preC, P and X . The X gene encodes a 17-kDa protein termed ‘hepatitis B virus X’ (HBx) which functions as a multifunctional regulator . Its continuous expression in hepatocytes has a role in the development of various liver diseases and an important role in neoplastic transformation of hepatocytes in HBV-infected liver, in particular [3,4].
OASIS (old astrocyte, specifically induced substance) family members are endoplasmic reticulum (ER)-resident transmembrane bZIP (basic leucine zipper) transcription factors that are specifically expressed in certain tissues; they include Luman, OASIS, the BBF2 human homologue on chromosome 7 (BBF2H7), cAMP-responsive, element-binding protein H (CREBH) and CREB4, all of which are expressed in dendritic cells, osteoblasts, chondrocytes, liver cells and the prostate, respectively . The protein sequences and activated mechanisms are similar to those of activating transcription factor 6 (ATF6), an ER stress marker. They have a transmembrane domain, a bZIP domain and a transcription activation domain. In addition, about 30 amino acids are conserved in the OASIS family, but not in ATF6 . According to previous studies, lipid accumulation, HBV infection, cytokine and hypoxia induce ER stress [6–9], leading to CREBH activation . Activated CREBH can regulate iron metabolism, gluconeogenesis, lipid metabolism and inflammation as a master gene [10–13].
Activator protein 1 (AP-1) is a transcription factor; it recognizes the PMA-responsive element (PRE) due to strong induction by the tumour promoter PMA . AP-1 is known as a bZIP protein and a dimeric complex that consists of the Jun and Fos ATF complexes and the musculoaponeurotic fibrosarcoma (MAF) protein . AP-1 can regulate the expression of genes relating to tumour development, such as cyclin D1, vascular endothelial growth factor (VEGF) and matrix metallopeptidase 9 (MMP9) [16–18]. In mammalian cells, the main AP-1 proteins are Fos and Jun, and c-Jun functions as a positive regulator of cell proliferation, leading to liver cancer . In the present study, we addressed the role of activated CREBH in HBV-induced hepatic cell proliferation.
Plasmid constructs and reagents
The following have been described previously : pT-HBV1.2mer, pcDNA3-HBx, pcDNA3-HA-HBx and pcDNA3-GST-HBx vectors. Recombinant adenovirus Ad-HBx and Ad-Con were also described previously . AP-1 luc (luciferase) and Gal4-TK (thymidine kinase) luc were provided by J.W. Lee and pGL2-cyclin D1, pGL2-VEGF-luc and pGL2-MMP9 vectors by K.H. Kim. pcDNA3-HA-CREBH(n) and pM-CREBH(n) vectors were cloned by reverse transcriptase (RT)-PCR using human primary mRNA and primers: forward 5′-CCGGAATTCATGAATACGGATTTAGCT-3′; reverse 5′-ACTCTAGATCATGTCTGGGCTGACTTGCT-3′. The PCR products were digested with EcoRI and XbaI and cloned in pcDNA3-HA and pM vectors. All the tags are located at the N-terminus of CREBH(n). Negative control siRNA (no. SN-1002) and small interfering (si)CREBH (no. 84699) were purchased from Bioneer, DMSO, 4-phenylbutyric acid (4-PBA), thapsigargin and cycloheximide (CHX) from Sigma, and the transfection reagent jetPEI and jetPRIME from PolyPlus Transfection.
The HepG2 cells were maintained in Dulbeco's modified Eagle's medium (DMEM) containing 10% FBS and 1% penicillin/streptomycin (PS) at 37°C in a humid atmosphere of 5% CO2. HepG2-HA (hepatitis A), HepG2-HA-HBx and HepG2-HA-CREBH(n) stable cells were maintained in a medium containing 500 μg/ml of G418. HepG2 cells were transfected with 2 μg of human influenza hemagglutinin (HA), HA-HBx or HA-CREBH(n) using JetPEI reagent. After 48 h, the cells were trypsinized and plated in a medium containing 800 μg/ml of G418. After selection for 2 weeks, total populations of G418-resistant cells were pooled and single-cell sorted into 96-well plates with a growth medium containing 800 μg/ml of G418. Sorted single cells were grown under selection for an additional 2 weeks and expanded into stable cell lines. The candidate clones were analysed by Western blot analysis using a specific HA antibody. Mouse primary hepatocytes were isolated through a collagenase perfusion method and plated in six-well plates with M199 medium containing 10% FBS, 1% PS and 10 nM dexamethasone.
Cells were seeded in 24-well culture plates and transfected with reporter vector and a plasmid expressing β-galactosidase, together with each indicated expression plasmid using JetPEI. Total amounts of expression vectors were kept constant using a pcDNA3 vector (Invitrogen). After 24 h of transfection, the cells were lysed in a Luciferase Cell Culture Lysis 5X Reagent [25 mM Tris-phosphate (pH 7.8), 2 mM DTT, 2 mM 1,2-diaminocyclohexane-N,N,N′,N′-tetraacetic acid, 10% glycerol, and 1% Triton X-100] (Promega). Luciferase activity was determined using a GloMax 96 Microplate Luminometer (Promega) according to the manufacturer's instructions. It was normalized for transfection efficiency using the corresponding β-galactosidase activity. All assays were performed at least in triplicate.
RNA isolation, RT-PCR and real-time PCR analysis
Total RNA was prepared from cell lines or tissues using Trizol reagent (Invitrogen) following the manufacturer's instructions. Total RNA (1 μg) was converted into single-strand cDNA by MMLV reverse transcriptase (Takara) with oligo-dT. A one-tenth aliquot of the cDNA was subjected to PCR amplification using gene-specific primers. The PCR products were examined by electrophoresis on a 1.2% agarose gel. Real-time PCR was performed with an SYBR Green I LightCycler-based, real-time PCR assay (Roche). The reaction mixtures were prepared using LightCycler Fast DNA master mixture for SYBR Green I, 0.5 μM of each primer and 4 mM MgCl2. PCR was carried out using the following forward and reverse primers. CREBH: forward 5′-GTCTTGCATCTCGAGAAGCAAAAC-3′ and reverse 5′-CA-GCATCGTTGTGCAAAG TTCTG-3′; c-Jun: forward 5′-ATG-GAGTCCCAGGAGCGGATCA-3′ and reverse 5′-GCACCCAC-TGTTAACGTGGTTCA-3′; cyclin D1: forward 5′-CGTCCA-TGCGGAAGATCGTC-3′ and reverse 5′-GAAATCGTGCG-GGGTCATTG-3′; β-actin: forward 5′-GACTACCTCA-TGAAGATC-3′ and reverse 5′-GATCCACATCTGCTGGAA-3′; VEGF: forward 5′-CTTGCCTTGCTGCTCTACC-3′ and rev-erse 5′-GGCTTGAAGATGTACTCGATC-3′; MMP9: forward 5′-TGGAGAGTCGAAATCTCTGG-3′ and reverse 5′-ATCC-AATAGGTGATGTTGTGG-3′.
Chromatin immunoprecipitation assay
Chromatin immunoprecipitation (ChIP) assays were performed as described by the manufacturer (Upstate Biotechnology) with some modifications. Chromatin solutions were sonicated and incubated with anti-HA or control IgG, and rotated overnight at 4°C. Chromatin DNA was purified and subjected to PCR analysis. To amplify the human cyclin D1 promoter region containing the AP-1-binding site and the c-Jun promoter region containing the CREBH-binding site, the following primer sets were used. AP-1 (PRE) ChIP: forward 5′-CCAATTAGGAACCTTCGGTGGTC-3′ and reverse 5′-GGTGGCCAGCATTTCCTTCAT C-3′; CREBH ChIP: forward 5′- CAGCGGAGCATTACCTCATCCC-3′and reverse 5′-TGTCTGCCTGACTCCGCGCAC-3′. As a negative control, a human c-Jun promoter region without a CREBH-binding site was used together with the following primer set. Negative control ChIP: forward 5′-GGCTCTGCGAGGATGGAAACT-3′ and reverse 5′-CCTCTAGGAAGCGCTAGACACATG-3′. After amplification, PCR products were resolved on a 1.5% agarose gel and visualized using ethidium bromide staining.
Western blot analysis
Cells were prepared by washing with cold PBS and lysed. The protein concentration was determined using BSA as the standard and Bradford reagent (Bio-Rad). Equal amounts of proteins were loaded and separated by SDS/PAGE, and the gels were transferred to polyvinylidene fluoride membranes (Millipore). For Western blotting, the membranes were incubated in TBS with Tween (TBST), supplemented by 3% non-fat dry-skim milk overnight at 4°C, with: anti-HBx (Chemicon); anti-CREBH, anti-c-Jun, anti-cyclin D1 and anti-GST (all from Santa Cruz Biotechnology); anti-HA (Roche); and anti-actin (Sigma). After washing three times with cold TBST, the blotted membranes were incubated with secondary antibody goat anti-rabbit or mouse IgG-HRP (Santa Cruz Biotechnology) for 30 min at room temperature. After washing three times with cold TBST, the proteins were visualized using an Amersham ECL Select Western Blotting Detection Reagent (Amersham Pharmacia Biotech).
In vivo GST pull-down assay
To evaluate the co-immunoprecipitation of HA-tagged CREBH(n) with GST-tagged HBx proteins, cell extracts were mixed with GST antibody immobilized on protein G/Sepharose (Invitrogen), and incubated for 12 h in a cold room. After washing with lysis buffer, the immune-complex immunoprecipitates were collected by centrifugation and dissolved in Laemmli sample buffer. The sample was boiled and subjected to 10% SDS/PAGE followed by immunoblot analysis.
Cell proliferation was assessed using a colorimetric BrdU Cell Proliferation Kit (Millipore). After 24 h of siRNA transfection or adenovirus infection, cells were cultured in fresh 5-bromo-2′-deoxyuridine (BrdU)-containing medium for 24 h. The colorimetric ELISA for BrdU quantification was performed following the manufacturer's instruction.
Statistical analyses were carried out by an unpaired or paired Student's t-test as appropriate. All data are reported as means±S.D. A P value of <0.05 was considered significant.
HBx induces CREBH expression and activation via ER stress
We reported that HBx induces ER stress in a previous study  and it was well known that CREBH is activated through ER stress . To investigate whether HBx can induce CREBH activation, Western blotting was performed. In Figure 1(A), HBV or HBx increases the expression of full-length CREBH [CREBH(f)] as well as of CREBH(n), the activated CREBH form. HBV constructs used in this study contain the HBV genome with the endogenous HBx promoter . To confirm whether expression of CREBH(f) and CREBH(n) is increased by CREBH mRNA up-regulation, real-time PCR was performed. As a result, the CREBH mRNA level was increased about 1.3- or 1.4-fold by HBV or HBx compared with Mock, respectively (Figure 1B). As a positive control, thapsigargin, an ER Ca2+ pump inhibitor and stress inducer, was used, resulting in a significant increase in protein and mRNA levels of CREBH (Figures 1C and 1D). To confirm that HBV- or HBx-induced CREBH up-regulation and activation are mediated by ER stress, 4-PBA, a chemical chaperone, was used to treat HBV- or HBx-transfected cells. As shown in Figures 1(E) and 1(F), HBV- or HBx-induced CREBH up-regulation and activation were attenuated by treatment with 4-PBA. These results suggest that HBx may induce the expression and activation of CREBH via ER stress.
HBV or HBx induces CREBH expression and activation via ER stress
CREBH up-regulates the expression of AP-1 target genes through c-Jun overexpression
To identify target genes of CREBH(n) activated by HBx, a luciferase assay was performed. Among the various genes, cyclin D1, VEGF and MMP9 luciferase activity were increased by CREBH(n) (Figure 2A). AP-1 luc is a vector containing artificial AP-1-binding sites and AP-1 consists, generally, of a c-Jun and c-Fos dimer. In real-time PCR, c-Jun, cyclin D1, VEGF and MMP9 transcripts were increased by CREBH(n) (Figure 2B). In Western blotting, the expression levels of c-Jun and cyclin D1 protein were increased by CREBH(n) (Figure 2C). Cyclin D1, VEGF and MMP9 are AP-1 target genes. To investigate whether CREBH(n) regulates the expression of AP-1 target genes by binding at the AP-1 sites or through c-Jun induction, a ChIP assay was performed using an AP-1-binding site (5′-TGAGTCA-3′) on the cyclin D1 promoter and a putative CREBH-binding site (5′-TGACATCA-3′) on the c-Jun promoter (Figure 2D). As shown in Figure 2(E), CREBH(n) did not bind at the AP-1 site on the cyclin D1 promoter; however, it bound at a putative CREBH site on the c-Jun promoter. Therefore, c-Jun induced by CREBH(n) increased the expression of cyclin D1, VEGF and MMP9. To confirm the results of the ChIP assay, cycloheximide (CHX), a translation inhibitor, was used. The results showed that c-Jun expression was mildly increased by CREBH(n); however, cyclin D1, MMP9 and VEGF expression was not changed (Figure 2F). Although the c-Jun mRNA level was increased by transfected CREBH(n) before CHX treatment, synthesis of the c-Jun protein was blocked after CHX treatment. Therefore, cyclin D1, VEGF and MMP9 mRNA levels were not increased by transfected CREBH(n). The results suggest that CREBH(n) can up-regulate the expression of AP-1 target genes via c-Jun.
CREBH(n) up-regulates the expression of AP-1 target genes through c-Jun over-expression
CREBH mediates the expression of HBV- or HBx-induced c-Jun and AP-1 target genes
To address the role of CREBH(n) in the expression of HBV- or HBx-induced AP-1 target genes, a luciferase assay was performed using AP-1, cyclin D1, VEGF and MMP9 luc. In Figure 3(A), HBV- or HBx-induced AP-1, cyclin D1, VEGF and MMP9 luc activity was attenuated by siCREBH, a knockdown system against the CREBH gene. At the RNA level, the expression of c-Jun, cyclin D1, VEGF and MMP9 induced by HBV or HBx was attenuated by siCREBH (Figure 3B). At the protein level, the expression of the c-Jun and cyclin D1 proteins increased by HBV or HBx was attenuated by siCREBH (Figure 3C). As AP-1 and cyclin D1 are closely related to cell proliferation, we investigated the effects of CREBH on HBx-induced cell proliferation through a BrdU assay. As shown in Figure 3(D), the proliferation of HCC cells induced by HBx was decreased by knockdown of the CREBH gene. The results suggest that CREBH mediates HBV- or HBx-induced AP-1 target gene expression and hepatic cell proliferation.
CREBH(n) mediates the expression of HBV- or HBx-induced c-Jun and AP-1 target genes
CREBH interacts with HBx
To identify genes interacting with CREBH(n), a mammalian two-hybrid assay was performed using Gal4-TK-luc. Among the various transcription factors and co-factors, CREBH(n) interacted with HBx (Figure 4A). To confirm the results of the assay, an in vivo GST pull-down assay was performed. As a result, we confirmed through Western blotting that CREBH(n) interacted with HBx (Figure 4B). These results therefore suggest that CREBH(n) may interact with HBx.
CREBH(n) interacts with HBx
The synergistic effects of CREBH and HBx on the expression of AP-1 target genes and hepatic cell proliferation
To investigate the effects of interaction between CREBH(n) and HBx, a luciferase assay was performed. HBx and CREBH(n) showed the synergistic effect on luciferase activity of AP-1 and the AP-1 target genes (Figure 5A). At the RNA and protein levels, expression of c-Jun and the AP-1 target genes was significantly increased by HBx and CREBH(n) (Figures 5B and 5C). Thus, HBx and CREBH(n) showed the synergistic effect on proliferation of HCC cells and mouse primary hepatocytes via a BrdU assay (Figures 5D and 5E). Taken together, HBx activates CREBH via ER stress, and then activated CREBH interacts with HBx, leading to expression of c-Jun and the AP-1 target genes, and resulting in proliferation of HCC cells.
The synergy effects of CREBH(n) and HBx on the expression of AP-1 target genes
HCC is the fifth most common cancer and the third most common cause of cancer mortality worldwide . The most common risk factor for HCC is chronic HBV infection; there are around 2 billion people infected with HBV and more than 350 million chronic carriers worldwide . Although a high HBV load and chronic hepatitis B (CHB) infection increase the risk of HCC development, its aetiology has not been fully elucidated. However, accumulating evidence shows that the viral proteins of HBV can induce oncogenic effects or synergistic effects in the transformation into HCC, with a hyperproliferative response induced by liver cell damage [22,24]. Although the role of ER stress in cancer is still not clear, it is induced in cancer cells due to the physiological environment of solid tumours such as hypoxia, low pH and low nutrient levels, and the cancer cells utilize altered ER stress responses to overcome ER-associated cell death [25,26]. In the present study, we showed that ER stress could mediate HBV-induced hepatic cell proliferation at the molecular level.
ER-resident, transmembrane, bZIP transcription factors in the OASIS family reveal a certain cell- or tissue-specific expression pattern, suggesting that the transcription factors may have specialized physiological functions depending on the cell or tissues . OASIS family members contain the consensus sequence for site-1 protease (S1P) in a luminal region such as ATF6. However, all the members are not activated by ER stress, e.g. Luman and CREB4 [5,27]. It is well known that CREBH is activated by ER stress. In the present study, we confirmed CREBH activation by ER stress (see Figure 1C) and showed CREBH activation by HBV- or HBx-induced ER stress (see Figures 1A and 1E). Interestingly, CREBH is both over-expressed and activated by ER stress. CREBH was over-expressed by HBV/HBx as well as by an ER stress inducer, and HBV- or HBx-induced CREBH over-expression was attenuated by a chemical chaperone, suggesting that CREBH expression is regulated by ER stress. It will be interesting to elucidate the mechanism by which ER stress regulates CREBH expression.
Reports dealing with the effect of CREBH on acute phase response and inflammation have been published in earlier studies [10,28]. However, those dealing with metabolism, including iron metabolism, triacylglycerol metabolism, hepatic gluconeogenesis and hepatic lipogenesis were published more recently [11–13,29]. Consideration of the papers shows that activated CREBH may play an important role in hepatic metabolism under normal, but not pathological, conditions. In the latter conditions, such as HBV infection, activated CREBH may have an important role to play in hepatic inflammation and cell proliferation. In the case of insulin receptors, those activated by insulin have a role related to glucose metabolism in the cytosol under normal conditions . However, the insulin receptor translocates to the nucleus, leading to proliferation of hepatic cells via amplification of the Ca2+ signal under conditions of liver damage or liver regeneration [31,32]. Each CREBH activated under normal or HBV-infected conditions may produce different results in hepatic cells because insulin receptors have a dual function according to the conditions. In Figures 4 and 5, CREBH activated by HBV interacted with HBx protein, leading to a synergistic effect in the expression of target genes related to cell proliferation. This effect of CREBH and HBx may induce hepatic cell proliferation and can be shown just in HBV-infected cells.
HBV-induced HCC develops in an environment of inflammation and regeneration induced by chronic liver damage . In CHB, CREBH activated by HBV or ER stress may play an important role in liver regeneration. However, deregulation of liver regeneration by HBx and CREBH may induce HCC which results from hepatic cell proliferation.
H.K. Cho produced the main experimental results, and S.Y. Kim, Y.Y Kyaw and A.A Win produced some of them. S.-H. Koo provided mouse primary hepatocytes. H.-H. Kim and J. Cheong established the experimental design. J. Cheong prepared the manuscript.
We thank Dr Jae Won Lee (Chungbuk National University, Korea) for providing the AP-1 luc (luciferase) and Gal4-TK-luc vectors and Dr Kook Hwan Kim (Sungkyunkwan University School of Medicine, Korea) for providing pGL2-cyclin D1, pGL2-VEGF-luc and pGL2-MMP9 vectors.
This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean Government (NRF-2009-0093195) and Basic Science Research Program through the NRF funded by the Ministry of Education (NRF-2013R1A1A2057634).
activator protein 1
activating transcription factor 6
basic leucine zipper
chronic hepatitis B
cAMP-responsive, element-binding protein H
human influenza hemagglutinin
hepatitis B virus
hepatitis B virus X
matrix metallopeptidase 9
old, astrocyte, specifically induced substance
reverse transcriptase PCR
small interfering CREBH
TBS with Tween
vascular endothelial growth factor