Oncogenic circuit constituted by Ser31-HBx and Akt increases risks of chronic hepatitis and hepatocellular carcinoma

Biochimica et Biophysica Acta 1862 (2016) 837–849

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbadis

Oncogenic circuit constituted by Ser31-HBx and Akt increases risks of chronic hepatitis and hepatocellular carcinoma Wei-Ping Lee a,f,⁎, Keng-Hsin Lan b,c,d, Chung-Pin Li c,d, Yee Chao c,e, Han-Chieh Lin c,d, Shou-Dong Lee c,d a

Institute of Biochemistry and Molecular Biology, School of Life Sciences, National Yang-Ming University, Taiwan Department of Pharmacology, National Yang-Ming University, Taipei, Taiwan School of Medicine, National Yang-Ming University, Taipei, Taiwan d Division of Gastroenterology, Department of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan e Cancer Center, Taipei Veterans General Hospital, Taipei, Taiwan f Department of Medical Research, Taipei Veterans General Hospital, Taipei, Taiwan b c

a r t i c l e

i n f o

Article history: Received 27 July 2015 Received in revised form 29 November 2015 Accepted 16 December 2015 Available online 12 January 2016 Keywords: HBx Akt Phosphorylation Hepatocellular carcinoma

a b s t r a c t The X protein of hepatitis B virus (HBx) has been specifically implicated in the development of hepatocellular carcinoma (HCC). Clinical associations of HBx isoforms with chronic hepatitis and HCC have not been well studied. HBx has two roles in liver cells, namely pro-apoptotic and anti-apoptotic. In this report, we examined the role of Ser31-HBx in HCC and chronic hepatitis. Using the case–control study, we determined risks of chronic hepatitis and HCC conferred by hepatitis B virus (HBV) containing Ser31-HBx that was phosphorylated by Akt. Ser31-HBx isoforms conferred 3.23-fold risk of HCC in male and 3.36-fold risk in female. Ser31 isoforms were associated with 3.12-fold risk of chronic hepatitis and 3.43-fold risk of cirrhosis and also associated with higher HBV viral load and replication efficiency and lower rate of HBe loss. To determine the mechanism, we found that Ser31-HBx constituted an oncogenic circuit with Akt and cooperated with ras to transform NIH3T3 cells in contrast to non-Ser31-HBxs that did not transduce oncogenic signals. Our results give a clue to account for an underlying cause of HBx-mediated hepatocarcinogenesis. It appears that Ser31 phosphorylation of HBx by Akt plays an important role. The current study provides an example of association of HBV genome variations with risks of HCC and chronic hepatitis © 2015 Published by Elsevier B.V.

1. Introduction Human hepatitis B virus (HBV) causes acute and chronic inflammation of the liver and is a causative organism in the development of hepatocellular carcinoma [1]. The HBV genome is a partially doublestranded DNA. Among the four proteins i.e. surface, core, polymerase, and X proteins produced from the HBV genome, the X protein (HBx), 154 amino acids and 17 kDa, is a multifunctional regulatory protein that plays versatile roles in transactivation of hepatitis B viral and host cellular genes [2] and has been associated with hepatocellular carcinogenesis [3,4]. Two major factors are associated with HCC: firstly, recurrent immune-mediated death of HBV-infected hepatocytes with compensatory regeneration of the liver [5], and secondly, sustained activities of HBV proteins among which HBx has been proven to be the major carcinogenic factor by transgenic models [3,6]. Although HBx does not directly bind to DNA, it modulates transcriptional activation by interaction with nuclear transcription factors [2,7,8]. Physical interaction between HBx and p53 inhibits normal function of ⁎ Corresponding author at: Department of Medical Research and Education, Taipei Veterans General Hospital, 201 Shi-Pai Rd Sec. 2, Taipei, Taiwan. E-mail address: [email protected] (W.-P. Lee).

http://dx.doi.org/10.1016/j.bbadis.2015.12.020 0925-4439/© 2015 Published by Elsevier B.V.

p53 by interrupting p53 sequence-specific DNA binding and transcriptional activity [9]. HBx has been demonstrated to interfere with DNA repair [10,11] by which it may induce mutations in hepatocytes. In addition, HBx activates oncogenic signal transductions, including the Ras/MAP kinase [12,13], PI-3 kinase/Akt [14,15], NF-κB [16,17], and Wnt-β-catenin [18,19] pathways. Despite current molecular evidence suggesting that HBx is an oncoprotein, the detailed mechanism of HBx-mediated hepatocarcinogenesis remains to be determined. Paradoxically, pro-apoptotic activity of HBx has also been recognized including sensitization of hepatoma cells to UV light [11], TNF-α [20, 21], and serum deprivation [22]. Although the molecular mechanism by which HBV causes HCC is not well understood, the correlation between them has been established [23]. Because HCC develops 20–30 years after viral infection, HBV appears not to be an active transforming virus. Multistep carcinogenesis may proceed in HBxexpressing hepatocytes [24,25]. It has been shown that HBx transgenic mice develop hepatic adenoma and subsequently HCC [3], but some HBx transgenic models require cytotoxic agents to promote HCC development [26,27]. It appears that there exist various HBx isoforms, which might give opposite experimental results in the same study. During the course of chronic HBV infection, mutations may occur throughout the genome probably due to lack of proof reading of the

838

W.-P. Lee et al. / Biochimica et Biophysica Acta 1862 (2016) 837–849

viral polymerase to synthesize its DNA or RNA [28]. We searched the NCBI gene bank and identified Akt phosphorylation sites (RXRXXS/T; R, arginine; X, any amino acid; T, threonine; S, serine) [29–34] on the 31st amino acid of some HBx isoforms. The motifs are most common as RXRPVS and RXRPLS and less common as RGRPFS and RGRSLS. Phosphorylation occurs on Ser31 [35]. HBx with Ser31 possesses higher rate of transactivation of the HBV genome than Ala31 mutant, and mutation of Ser31 to Ala31 is prevalent in HCC patients of Taiwan [36]. Thus, Ser31 phosphorylation of HBx may be associated with HCC development. Most HBV carriers are infected from their childhood in Taiwan [37]. The virus inhabits the host for a long period of time facilitating the development of mutant viruses. Thus, an infected individual may have more than one isoform of HBx. In the current study, we show that the HBx isoforms containing Ser31 are associated with increased risks of chronic hepatitis and HCC. 2. Materials and methods

between nucleotide position 120 and 604 was amplified by seminested PCR. The PCR products were treated with restriction enzymes, resolved in a 4% agarose gel electrophoresis, and stained by ethidium bromide. To confirm correct genotypes, the PCR products were subjected to direct sequencing. The genotypes of viral genomes used in the study are the following: HBV-1, genotype B; HBV-2, genotype C; HBV-3, genotype C; and HBV-4, genotype B. 2.5. HBV DNA isolation Blood samples were obtained from patients with chronic infection of hepatitis B virus. Serum was collected by centrifugation at 1000 ×g. To 0.5 ml serum was added proteinase K to final 10 μg/ml and SDS to final 1%. The serum was incubated at 55 °C overnight and then processed for phenol/chloroform extraction. DNA was precipitated with 0.1 volume of 3 M sodium acetate, pH 5.2 and one volume of isopropanol, and the pellet was washed with 70% ethanol and then air dried. The HBV DNA pellet was dissolved in 20 μl of TE (20 mM Tris, 1 mM EDTA, pH 8.0) and stored until use.

2.1. Subjects of the study 2.6. PCR amplification of HBx DNA and HBV enhancer I-X promoter A hospital-based case–control study was conducted to evaluate the association of HBx containing Ser31 with incidence of hepatocellular carcinoma (HCC) in Taipei Veterans General Hospital, Taiwan. Controls were patients with hepatitis B virus infection including inactive HBV carrier with normal ALT (alanine aminotransferase), chronic hepatitis (persistent ALT N40 U/L or some values N80 U/L over a period of at least six months), and cirrhosis (a nodular liver surface, decreased right lobe–caudate lobe ratio, or indirect evidence of portal hypertension under ultrasound). Cases were patients with new diagnosis of HCC attending the hospital from July 2007 to July 2009. All cases had received dynamic computer tomography or histological confirmation of their tumor diagnosis. Patients of cases and controls were not co-infected with hepatitis C virus. Cases and controls with other liver diseases such as primary biliary cirrhosis and autoimmune hepatitis were also excluded. Cases and controls were not treated with antiviral agents prior to the study. Individuals participating in this study signed an agreement approved by the institutional review board of human clinical study of the hospital. Each subject was scheduled for an interview after informed consent was obtained, and a structured questionnaire was administered by interviewers to collect information of demographic data. This study was implemented in accordance with Good Clinical Practice (GCP) guidelines and the Declaration of Helsinki. 2.2. Statistical analysis for case–control study The HCC, chronic hepatitis and cirrhosis risks of HBx containing Ser31 (phosphorylated by Akt) was evaluated by a case–control study. Comparison of the risks between cases and controls was assessed by the χ2 test. Odds ratios (ORs) with 95% confidence intervals (CIs) and unconditional logistic regression models were computed using the SPSS software program (SPSS Science, Chicago, IL). Differences were considered to be significant at P b 0.05.

HBx DNA was amplified by PCR (polymerase chain reaction) using HBV DNA as a template. The primers for PCR were forward 5′-GCTGCT CGGGTGTGCTGCCAA and reverse 5′-GGCAGAGGTGAAAAAGTTGCA. The conditions for the reaction were 95 °C 1 min, 55 °C 1 min, and 72 °C 1 min; 25 cycles. The HBV enhancer I-X promoter was amplified using the primers, forward 5′-TGGATATCCTGCCTTAATGC and reverse 5′-GATGTATATTTCCGCGAGAG. The reaction condition was the same as that for HBx DNA. 2.7. DNA cloning To generate FLAG-tagged HBx, the HBx DNA synthesized by PCR was cloned in the expression vector p3XFLAG-myc-CMV-27 (abbreviated p3XFLAG, Sigma-Aldrich) via Hind III and Kpn I sites. To produce GSTHBx fusion protein, the HBx DNA was cloned in the pGSTag plasmid. To determine HBx-mediated transcriptional activation, HBV enhancer I-X promoter was amplified by PCR and cloned in the pGL3-basic plasmid encoding firefly luciferase to obtain the reporter plasmid pEN1P-Luc. 2.8. Site-directed mutations of HBx on Akt phosphorylation sites Ser31 of HBx-1 and HBx-2 were mutated to alanine, designated HBx-1A and HBx-2A, respectively with the QuickChange site-directed mutagenesis kit (Stratagene Corp.). Briefly, PCR was done with two complementary primers covering the sites for mutation and p3XFLAGHBx plasmid as a template. The PCR product was treated with the restriction enzyme Dpn I which recognized GATC with methylated adenine. Adenine methylation occurs in plasmid replication of Escherichia coli but not in DNA replication of PCR. Thus, the plasmid template was digested by Dpn I. Then, the intact plasmid i.e. the PCR product was used to transform XL-1 blue competent cells.

2.3. Viral serology and DNA tests 2.9. Cell lines and cell culture The serum samples were tested for HBeAg and anti-HBe antibody using radio-immunoassay (Abott Laboratories, North Chicago, IL). HBV DNA level was determined by Roche Cobas Tagman HBV DNA assay (Roche Diagnostics, Switzerland).

HEK293T and human hepatoma cell line HepG2 were cultured in Dulbecco's modified Eagle medium (GIBCO Invitrogen Corp) supplemented with 10% fetal bovine serum (GIBCO-BRL) and penicillin/ streptomycin (100 U/100 μg/mL, Sigma-Aldrich) at 37 °C and 5% CO2.

2.4. HBV genotyping 2.10. Replication efficiency assay of HBV DNA in HepG2 cells Genotypes of HBV were determined by PCR restriction fragment length polymorphism (PCR-RFLP) of the HBs gene [38]. Briefly, DNA was extracted from serum, and the fragment of the HBV genome

Six plasmids, pCMV-HBV-1, pCMV-HBV-2, pCMV-HBV-1A, pCMVHBV-2A, pCMV-HBV-3 and pCMV-HBV-4, encoded a greater-than-

W.-P. Lee et al. / Biochimica et Biophysica Acta 1862 (2016) 837–849

unit-length HBV genome and contained HBx-1, HBx-2, HBx-1A, HBx-2A, HBx-3, and HBx-4, respectively. These plasmids were transfected into HepG2 cells as described by Yeh et al. [36]. Transfection efficiency was monitored by cotransfecting a luciferase expression plasmid, pGL2. 48 h after transfection, cells were trypsinized, washed with phosphate-buffered saline and lysed with 0.5% Nonidet P-40 in Trisbuffered saline (10 mM Tris–HCl pH 7.2, 50 mM NaCl). After centrifugation at 1500 g for 30 s, the supernatant was adjusted to 10 mM of MgCl2 and subjected to digestion with 100 μg/ml of DNase I for 30 min at 37 °C. EDTA was added to a final concentration of 10 mM to stop the reaction. Then proteinase K and SDS was added to final 100 μg/ml and 1%, respectively, and the mixture was incubated for 3 h at 55 °C, followed by phenol/chloroform extraction and ethanol precipitation. To detect HBV DNA, the samples were dissolved and loaded onto a 1% agarose gel for electrophoresis and then blotted onto a Hybon-N membrane. HBV DNA was detected by hybridization with a digoxigenin-labeled HBV DNA probe (DIG DNA labeling kit, Roche Applied Science).

839

pellet was resuspended in 100 μl of cold buffer C (20 mM HEPES-KOH pH 7.9, 25% glycerol, 420 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 0.5 mM dithiothreitol, 0.2 mM PMSF) and incubated on ice for 20 min for high salt extraction. Cell debris was removed by centrifugation for 2 min at 4 °C, and the supernatant fraction (nuclear extract) containing DNA-binding proteins was stored at − 70 °C until use. The nuclear extract (5 μg of protein) was incubated with 1 μg of poly (dI-dC) (Roche Applied Science) on ice for 20 min and then with the 32Plabeled ds oligonucleotides (abbreviated oligos) 5′-CCCTTTGATCTTACC or a mutant oligos 5-CCCTTTGGCCTTACC. The underlined nucleotides represent the altered nucleotides which destroy the TCF-4 recognition sequence. Binding of the radioactive ds oligos to nuclear DNA binding proteins was carried out at room temperature for 20 min. The resulting ds oligos–protein complexes were resolved with 4% nondenaturing polyacrylamide gel electrophoresis, followed by X-ray film autoradiography. 2.13. Reporter assay

2.11. Western blot analysis Cells with plasmid transfection and drug treatments were lysed in a buffer containing 50 mM Tris HCl, pH 7.4, 150 mM NaCl, 1% Triton-X 100, 5 mM EDTA, and SigmaFAST Protease Inhibitor Cocktail. Cell lysates were resolved by SDS-PAGE and then transferred to a PVDF membrane. The membrane was treated with a primary antibody, followed by incubation with a peroxidase-conjugated secondary antibody and detection with the enhanced chemiluminescence method. Primary antibodies included anti-FLAG (Sigma-Aldrich), anti-Akt1/2 (Santa Cruz), anti-PSer473-Akt (cell signaling), anti-P-Thr308-Akt (cell signaling), anti-PAkt substrate (cell signaling), anti-GSK-3β (Santa Cruz), anti-P-Ser9GSK-3β (Santa Cruz), anti-HBx (abcam), anti-Ha-ras (Santa Cruz), and anti-β-catenin (Santa Cruz) antibodies.

To determine HBx-mediated transcriptional activation, we cloned HBV enhancer I-X promoter in pGL3-basic plasmid encoding firefly luciferase to obtain the reporter plasmid pEN1P-Luc. The pEN1P-Luc plasmid was co-transfected with a p3XFAG-HBx isoform and a control luciferase pTK-Ranilla into HepG2 cells. Two days after transfection, cells were harvested and assayed by the Dual-Luciferase Reporter Assay System (Promega). Το determine β-catenin/TCF-4 activity, we used the reporter plasmids containing multimerized TCF-4 binding sites (pTOPFLASH and its mutant version pFOPFLASH) linked to the firefly luciferase gene. HepG2 cells were transfected with p3XFLAG-HBx, pTOPFLASH (or pFOPFLASH) and pTK-Ranilla plasmids. The Ranilla luciferase was used as an internal control. Two days after transfection, cells were harvested and assayed by the Dual-Luciferase Reporter Assay System.

2.12. Immunoprecipitation-Western blotting assay 2.14. Focus-forming assay Six isoforms of p3XFLAG-HBx and control plasmids (1 μg each) were individually transfected into 1 × 106 HEK293T cells. 36 h after transfection, cells were lysed in a lysis buffer containing 50 mM of Tris HCl, pH 7.4, 150 mM of NaCl, 1% Triton-X 100, 5 mM of EDTA, and SigmaFAST Protease Inhibitor Cocktail. The cell lysate was cleared by centrifugation at 10,000 ×g. The supernatant was added with 2 μg of anti-FLAG antibody and left at 4 °C overnight. The immune complexes were collected on protein A sepharose CL-4B (Amersham Biosciences) and then resolved with SDS-polyacrylamide gel electrophoresis. The proteins in the gel were transferred to a PVDF membrane in which phospho-HBx was detected with anti-P-Akt substrate antibody. 2.12.1. In vitro kinase assay Six isoforms of HBx DNAs were cloned into the pGSTag plasmid. GST-HBx was induced in E coli by IPTG (Isopropyl β-D-1-thiogalactopyranoside) and purified by the GSH-sepharose. HBx was separated from GST by thrombin treatment. In vitro kinase reaction was performed in 20 μl of kinase buffer containing 3 μg of purified HBx with or without 100 ng of activated Akt1 (Upstate Biotechnology), 200 μM of ATP, and 10 μCi [γ-32P]ATP (PerkinElmer Life Sciences) at 30 °C for 40 min. The reaction mixtures were subjected to 10% SDS-PAGE. Phosphorylation of the HBx proteins was detected by autoradiography. 2.12.2. Electrophoretic mobility shift assay 1 × 106 HepG2 cells were transfected with 6 isoforms of p3XFLAGHBx plasmids. 48 h after transfection, cells were washed twice with PBS, scrapped into 1.5 ml of cold PBS, pelleted for 10 s, and then resuspended in 200 μl of cold buffer A (10 mM HEPES-KOH pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM dithiothreitol, 0.2 mM PMSF). Cells were allowed to swell on ice for 10 min and then vortexed for 10 s. Samples were centrifuged for 10 s, and the supernatants were discarded. The

1 × 106 NIH3T3 cells were transfected with 4 μg of activated Ha-ras (pEJ6.6) and 4 μg of p3XFLAG-HBx plasmids. 24 h after transfection, the cells were split at a 1:4 ratio and cultured in the presence of 200 μg per ml of G418. After 3 weeks of selection, the cells were fixed in methanol:acetone (1:1) solution for 10 min and then stained with 0.1% Giemsa in 6% methanol for 10 min. The number of colonies (N20 mm) was counted. 2.15. Tumorigenesis assay of HBx and Ras cotransfection NIH3T3 cells were transfected with plasmids (Ha-ras, HBx-1 + Haras, or HBx-1 A + Ha-ras), selected with G418 for one week, and then harvested by trypsinization. Cells were washed twice with sterile phosphate-buffered saline, and resuspended at 2 × 107 cells per ml. Aliquots (0.1 ml) were injected subcutaneously into 6- to 8-week-old Balb/c athymic nude mice. Mice were observed, and tumors were palpated as fixed firm masses and measured with a standard caliper at periodic intervals by the hemi-ellipsoid model, typical as tumor volume = (π / 6) × (length) × (width) × (height) [39]. 3. Results 3.1. Ser31 isoforms of HBx We and another group have shown that some HBxs can be phosphorylated by Akt and have anti-apoptotic property [35,40]. In this study, we examined multiple HBxs and chose two HBx isoforms containing Akt phosphorylation motif, RGRPVS (HBx-1) and RGRPLS (HBx-2). We mutated Ser31 of the two HBxs to alanine, designated HBx-1A and HBx-2A, respectively (Fig. 1A). Two HBxs without Akt

840

W.-P. Lee et al. / Biochimica et Biophysica Acta 1862 (2016) 837–849

phosphorylation motif were RGRPVL (HBx-3) and RGRPVP (HBx-4) (Fig. 1A). HBx DNAs were cloned in p3XFLAG-myc-CMV-27 plasmid (designated p3XFLAG-HBx). Fig. 1B showed complete sequences of HBx isoforms in which the sequences of bold-typed and underlined amino acids are Akt phosphorylation motifs. Other variations were also underlined. Variation hot spots were located between 30th and 47th amino acids. Compared with other non-Ser31-HBx, HBx-1 and HBx-2 had a little longer half-life (15 h relative to 10 h in non-Ser31 isoforms, Supplementary Fig. 1). To test phosphorylation of HBx-1 and HBx-2 by Akt at Ser31, we performed in vitro kinase assay. HBx DNAs were cloned in GSTag plasmid, and GST-HBxs were generated by E. coli and purified with GSHsepharose. HBx was separated from GST by thrombin treatment and subjected to in vitro kinase assay. As shown in the left panel of Fig. 1C, HBx-1 and HBx-2 were phosphorylated by Akt, however, those without Ser31 (HBx-1A, HBx-2A, HBx-3, and HBx-4) failed to be phosphorylated by the kinase. The in vitro phosphorylation of HBx-2 was inhibited by

the Akt inhibitor MK-2206 (right panel of Fig. 1C). Coomassie brilliant blue-stained GST and HBx were shown in the lower panels of the figure. To prove that Akt phosphorylates HBx-1 and HBx-2 within cells, we transfected HEK293T with p3XFLAG-HBx plasmid and then performed immunoprecipitation (IP) with anti-FLAG antibody, followed by immunoblotting blot (IB) with anti-P-Akt substrate (RXRXXS/T) antibody. As shown in Fig. 1D, anti-P-Akt substrate antibody recognized HBx-1 and HBx-2 but not HBx-1A, HBx-2A, HBx-3 and HBx-4. MK-2206 was used to inhibit kinase activity of Akt and abrogated phosphorylation of HBx-1 and HBx-2 by Akt. These results indicate that HBx with Ser31 can be phosphorylated by Akt. 3.2. Association of HBx containing Akt phosphorylation site Ser31 with increased risk of HCC Subjects in controls (patients infected with HBV virus) and HBVassociated HCC cases came from north Taiwan. Because male and female

Fig. 1. Phosphorylation of HBx by Akt. A. HBx isoforms with (HBX-1 and HBx-2) or without (HBx-3 and HBx-4) Akt phosphorylation motif, RXRXXS. HBx-1A is HBx-1 with mutation of Ser31 to alanine, and HBx-2A is HBx-2 with mutation of Ser31 to alanine. B. Amino acid alignment of HBx-1, HBx-2, HBx-3, and HBx-4. Bold-typed and underlined sequences are predicted Akt phosphorylation motifs. Ser31 is the putative Akt phosphorylation site in HBx-1 and HBx-2. Underlined amino acids are variations of the four HBx isoforms. C. In vitro phosphorylation of HBx by Akt. HBx was tagged by GST. GST-HBx fusion protein was purified, and HBx was separated from GST by thrombin treatment. In vitro kinase assay was performed by phosphorylating Akt with activated Akt1 and [γ-32P]ATP. The reaction mixtures were subjected to 10% SDS-PAGE, followed by autoradiography (left panel). 0.5 μg thrombin-cut GST and HBx proteins were resolved by SDS-PAGE and stained with Coomassie brilliant blue (bottom of left panel). In vitro phosphorylation of HBx-2 was inhibited by the Akt inhibitor MK-2206 (MK, 1 μM) (Sigma-Aldrich) shown in the right panel. kDa, kiloDalton. D. Phosphorylation of HBx within cells by Akt. FLAG-tagged HBx plasmids were transfected into HEK293T cells for 48 h. Cell lysates were immunoprecipitated with anti-FLAG antibody, followed by immunoblotting with anti-P-Akt substrate RXRXXS/T antibody. To inhibit kinase activity of Akt, we treated HBx-1- and HBx-2-transfected cells with 1 μM of MK-2206 (MK) for 12 h prior to cell harvest. Lower panel showed expressions of input FLAG-HBx, P-Ser473-Akt, and Akt by Western blot (WB).

W.-P. Lee et al. / Biochimica et Biophysica Acta 1862 (2016) 837–849

841

Fig. 1 (continued).

have different incidence of HCC, we divided the case–control study into two categories, i.e. male and female. Patients co-infected with hepatitis C virus were excluded. Table 1 showed demographic characteristics of the male and female. The patients were limited between 30 and 65 years of age. In male, the mean age was 46.83 for HCC cases and 38.24 for HBV controls (p b 0.001). In female, the mean age was 47.63 for cases and 39.21 for controls (p b 0.001). There was no significant difference in the status of the liver (cirrhosis, chronic hepatitis, and healthy carrier) between HBV-infected controls and HCC cases. The role of HBx in hepatic carcinogenesis is controversial. Some investigators support its pro-apoptotic effect, but others propose its

survival activity. Since some HBx isoforms contain Akt phosphorylation motif, we hypothesized that HBx isoforms with Ser31 conferred a risk in HCC development. HCC frequencies between HCC cases and non-HCC HBV-infected controls were conducted by the χ2 test. Compared with HBx without Akt phosphorylation sites, those with Ser31 had 3.23fold risk of HCC in male group (95% CI: 2.07–5.06, χ2 = 27.41, p b 0.001; Table 2, upper panel) and 3.36-fold risk in female group (95% CI: 1.99–5.67, χ2 = 21.24, p b 0.001; Table 2, lower panel). These results suggest that Ser31-phosphorylated HBxs may function more active as oncogenic proteins than those without the Akt phosphorylation site.

Table 1 Demographic data of patients included in this study. Male

Mean age Cirrhosis (%) Chronic hepatitis (%) Inactive HBV carrier (%)

Female

HBV positive (n = 181)

HCC (n = 172)

P value

HBV positive (n = 127)

HCC (n = 130)

P value

38.24 ± 7.33 22 (12.3%) 112 (61.9%) 47 (25.8%)

46.83 ± 8.14 24 (13.9%) 108 (62.8%) 40 (23.3%)

b0.001 0.754 1.000 0.720

39.21 ± 8.47 23 (18.1%) 82 (64.6%) 22 (17.3%)

47.63 ± 9.22 24 (18.5%) 85 (65.4%) 21 (16.1%)

b0.001 1.000 1.000 1.000

HBV: hepatitis B virus; HBV positive: subjects with HBV infection; HCC: HBV-infected subjects with hepatocellular carcinoma; Chronic hepatitis: persistent ALT N40 U/L or some values N80 U/L over a period of at least six months with exclusion of fatty liver and other liver diseases. Inactive HBV carrier: HBV infection but with normal ALT.

842

W.-P. Lee et al. / Biochimica et Biophysica Acta 1862 (2016) 837–849

Table 2 Increased HCC risk in patients infected by HBV with Ser31-HBx. HBV positive

HCC

Odds ratio (95% CI)

No of male patients HBx without Ser31 HBx with Ser31 Total No of subjects

98 83 181

46 126 172

1.00 3.23 (2.07–5.06)

No of female patients HBx without Ser31 HBx with Ser31 Total no. of subjects

69 58 127

34 96 130

1.00 3.36 (1.99–5.67)

Table 4 Ser31-HBx is associated with the development of cirrhosis. P

b0.001

b0.001

3.3. Ser31-HBx is associated with increased risk of chronic hepatitis and cirrhosis As shown in Table 2, Ser31-HBx is related to the development of HCC. Since chronic inflammation is a causative factor of HCC, we hypothesized that patients infected with HBV of Ser31-HBx had increased risk to develop chronic hepatitis and cirrhosis. We summed up male and female patients of healthy carrier, chronic hepatitis, and cirrhosis in Tables 1 and 2 to determine whether Ser31-HBx is associated with increased risk of chronic hepatitis and cirrhosis. As shown in Tables 3 and 4, Ser31-HBx had 3.12-fold risk of developing chronic hepatitis (95% CI: 1.74–5.61, χ2 = 15.17, p b 0.001) and 3.43-fold risk of developing cirrhosis (95% CI: 1.56–7.53, χ2 = 9.77, p = 0.002) in comparison with non-Ser31-HBx.

3.4. Ser31-HBx is associated with higher HBV viral load High HBV viral load is associated with HCC risk [41,42]. To determine whether Ser31-HBx is associated with higher HBV viral load than nonSer31 isoforms, we examined HBV DNA from 211 patients with Ser31HBx and 206 patients with non-Ser31-HBx. These patients were not treated with anti-viral drugs. As shown in Fig. 2A, Ser31-HBx showed higher HBV viral load than non-Ser31-HBx (6.32 ± 1.33 vs 4.71 ± 0.84 log10 copies/ml, p b 0.001). In the Ser-31-HBx group, there was no significant difference between HCC cases and HBV controls (Fig. 2B). However, in the non-Ser31 HBx group, the subgroup of HCC showed significantly higher viral load than that of HBV control (Fig. 2B, 4.81 ± 0.92 vs 3.24 ± 0.82 log10 copies/ml, p = 0.007). This result suggests that there exist other factors affecting viral load in HBV of non-Ser31-HBx and contributing to HCC development. As described above, Ser31-HBx is associated with higher HBV viral load. Next we made an attempt to determine the underlying mechanism. We cloned HBV enhancer I-X promoter (EN1P) in pGL3-basic plasmid encoding firefly luciferase to obtain the reporter plasmid pEN1P-Luc. The pEN1P-Luc plasmid was co-transfected with a p3XFAGHBx isoform and a control luciferase plasmid pTK-Renilla into HepG2 cells. Ser31-HBx (HBx-1 and HBx-2) showed higher efficiency of transcriptional activation upon HBV enhancer pEN1P-Luc (Fig. 2C, the right panel showed expressions of FLAG-HBx, P-Ser473-Akt, and Akt), accounting for the association of Ser31-HBx with higher HBV viral

Table 3 Ser31-HBx is associated with the development of chronic hepatitis. No of patients

HBx without Ser31 HBx with Ser31 Total No of subjects

HBV carrier

Chr. hepatitis

48 21 69

82 112 194

Odds ratio (95% CI)

P

1.00 3.12 (1.74–5.61)

b0.001

Subjects of inactive HBV carrier and chronic hepatitis were from male and female patients in Table 1.

No of patients

HBx without Ser31 HBx with Ser31 Total No of subjects

HBV carrier

Cirrhosis

Odds ratio (95% CI)

P

48 21 69

18 27 45

1.00 3.43 (1.56–7.53)

0.002

Subjects of inactive HBV carrier and cirrhosis were from male and female patients in Table 1.

load. In Fig. 2D, the same plasmid transfection as shown in Fig. 2C was performed after siRNA transfection and diagrammed using HBx-2siCtr as 100% efficiency of transcriptional activation that was suppressed by siRNA of HBx (left panel of Fig. 2D). HBx-2-mediated transcriptional activation was suppressed by the Akt inhibitor MK2206 (right panel of Fig.2D). Lower panels showed expressions of FLAG-HBx, P-Ser473-Akt, and Akt.

W.-P. Lee et al. / Biochimica et Biophysica Acta 1862 (2016) 837–849

843

Fig. 2. Ser31-HBx is associated with increased HBV viral load.A. HBV viral loads in HBV-infected patients with Ser31-HBx or non-Ser31-HBx. HBV DNA was presented as log10 copies/ml. 211 patients with Ser31-HBx and 206 patients with non-Ser31-HBx were included.B. Comparison of HBV DNA between HCC and HBV subgroups in Ser31-HBx or non-Ser31-HBx.C. Ser31HBx isoforms (HBx-1 and HBx-2) showed higher efficiency of transcriptional activation upon HBV promoter than non-Ser31 isoforms. HBV enhancer I-X promoter reporter plasmid pEN1P-Luc was co-transfected with a p3XFAG-HBx isoform and a control luciferase plasmid pTK-Ranilla into HepG2 cells. Two days after transfection, cells were subjected to the DualLuciferase Reporter Assay System. Each bar was the mean of triplicate assays. The right panel showed expressions of FLAG-HBx, P-Ser473-Akt, and Akt.D. HBx-2-mediated transcriptional activation was suppressed by siRNA of HBx and the Akt inhibitor MK-2206. Before plasmid transfection as shown in panel C, HepG2 cells were transfected with 150 nM of siRNA of HBx (HBx-2-siHBx) or control siRNA (HBx-2-siCtr) (left panel). 1 μM of MK-2206 was used to inhibit Akt-mediated phosphorylation of HBx-2 (HBx-2-MK) (right panel). Each bar was the mean triplicate assays. Expressions of FLAG-HBx, P-Ser473-Akt, and Akt were shown in lower panels by Western blot (WB). The HBx-siRNA targeting sequence used was 5′-AAUGUCAACAACCG ACCUUGA.E. HBVs containing Ser31-HBx isoforms (HBx-1 and HBx-2) showed higher replication efficiency than those with non-Ser31-HBx isoforms. Six pCMV-HBVs were transfected into HepG2 cells. HBV-1, HBV-2, HBV-1A, HBV-2A, HBV-3, and HBV-4 contained HBx-1, HBx-2, HBx-1A, HBx-2A, HBx-3, and HBx-4, respectively. HBV DNAs produced in HepG2 cells were isolated and subjected to Southern blotting with a digoxigenin-labeled HBV DNA probe (left panel; RC, relaxed circular; SS, single-stranded). In the right panel, replication efficiency of HBV-2 was inhibited by siRNA of HBx (HBx-2-siHBx) or 1 μM MK-2206 (HBx-2-MK). Expressions of HBx, P-Ser473-Akt, and Akt were shown in the lower panels by Western blot (WB).F. HBV-1 and HBV-2 had higher viral secretion in mixed stable HBV transfections. 5 × 105 HepG2 cells were transfected with 9 μg of a pCMV-HBV isoform and 1 μg of pSV-neo plasmids in 6-cm tissue culture dishes. 24 h after transfection, cells were trypsinized and transferred to 10-cm tissue culture dishes and selected with 500 μg/ml G418 for two weeks. HBV DNAs were determined from supernatants by Roche Cobas Tagman HBV DNA assay two days after 10 ml of new medium was added with or without 1 μM of MK-2206. Each bar of the left panel was the mean of triplicate assays. The right panel showed expressions of HBx, P-Ser473-Akt, and Akt by Western blot.

To further determine whether HBV containing Ser-31-HBx has higher DNA replication efficiency, we transfected pCMV-HBV plasmids into HepG2 cells and extracted HBV DNA for Southern blotting. As shown in the left panel of Fig. 2E, HBVs containing Ser31-HBx (HBx-1 and

HBx-2) replicated more efficiently than those containing non-Ser31HBxs, correlated with higher viral loads in patients with HBV of Ser31HBx. The same plasmid transfection was performed using pCMV-HBV containing HBx-2 to examine Ser31-HBx-enhanced HBV replication

844

W.-P. Lee et al. / Biochimica et Biophysica Acta 1862 (2016) 837–849

Fig. 2 (continued).

that was suppressed by siRNA of HBx or MK-2206 (right panel of Fig. 2E). Lower panels showed expressions of HBx, P-Ser473-Akt, and Akt. To determine HBV secretion in the system, pCMV-HBV was cotransfected with pSV-neo plasmid into HepG2 cells and then selected with G418 for two weeks. Supernatants of the cultures were collected to determine HBV DNA two days after new medium was added. As shown in left panel of Fig. 2F, HBV-1 and HBV-2 had higher viral secretion than other HBV genomes. MK-2206 suppressed viral secretion of HBV-1 and HBV-2. The right panel of Fig. 2F showed expressions of HBx, P-Ser473-Akt, and Akt in mixed stable transfections.

3.6. Ser31-HBx forms an oncogenic circuit with Akt HBx has been shown to activate various oncogenic signaling pathways. In this study, we included MAP kinase, Akt, and β-catenin to examine oncogenic activity of HBxs with Ser31. HepG2 cells were individually transfected with six isoforms of p3XFLAG-HBx plasmids and then subjected to Western blot analysis with anti-P-Erk and antiErk antibodies. HBx-1 and HBx-2, both of which contain Ser31 promoted Erk phosphorylation stronger than other isoforms, indicating that HBx with Ser31 stimulates MAP kinase pathway (Fig.3A, P-Erk band intensity

3.5. Non-Ser31 HBx is associated with increased rate of HBe loss HBeAg is translated from genomic length mRNA transcripts that are similar to those for synthesis of HBV DNA. As shown in Fig. 2, Ser31-HBx is associated with increased HBV viral load. Since HBe loss is closely related to decreased viral load, we made an attempt to determine the association of non-Ser31 HBx with HBe loss. Table 5 shows that nonSer31 HBx has 2.16-fold higher in rate of HBe loss than Ser31 HBx. Genotypes B and C are the major genotypes of HBV in Taiwan. Table 6 shows that the Ser31-HBx isoform is not associated with genotypes B and C.

Table 5 Increased HBe loss in HBV with non-Ser31-HBx. No of patients

HBx with Ser31 HBx without Ser31 Total No of subjects

HBe (+)

HBe (−)

Odds ratio (95% CI)

P

146 61 207

112 101 213

1.00 2.16 (1.44–3.23)

b0.001

HBe (+): HBe positive. HBe (−): HBe negative, either HBe loss only or HBe seroconversion.

W.-P. Lee et al. / Biochimica et Biophysica Acta 1862 (2016) 837–849

that HBxs with Ser31 may determine the property of HBx, namely anti-apoptotic activity or pro-apoptotic activity driven by HBx with or without Ser31, respectively.

Table 6 Ser31-HBx is not associated with HBV genotypes B and C. Genotypes No. of patients

HBx with Ser31 HBx without Ser31 Total No of subjects

845

B

C

Odds ratio (95% CI)

P

131 74 205

127 88 215

1.00 1.23 (0.83–1.82)

0.309

Genotypes of HBV were determined by PCR restriction fragment length polymorphism (PCR-RFLP) of the HBs gene.

was measured by NIH Image J, and Erk phosphorylation was adjusted by the intensity of Erk input.). The same lysates were subjected to Western blotting with anti-PThr308-Akt, anti-P-Ser473-Akt, anti-Akt antibodies (Fig. 3B), anti-PSer9-GSK-3β, and anti-GSK-3β antibodies (Fig. 3C). HBx-1 and HBx-2 activated Akt that phosphorylated and inactivated GSK-3β, and subsequently β-catenin translocated into the nucleus (upper panel of Fig. 3D) to bind to its consensus oligonucleotides (lower panel of Fig. 3D, EMSA). Non-Ser31-HBx isoforms showed weaker Akt activation and less β-catenin translocation into the nucleus to bind with its consensus oligonucleotides. All tested bands were quantified by NIH Image J and compared with HBx-1 that was set as 1.00. In addition, we used reporter constructs that contained three repeats of wild-type (TOP) or mutant (FOP) β-catenin/TCF4-binding sites [43] to determine the transcriptional activity of endogenous β-catenin/ TCF4 in HBx-transfected HepG2 cells. Higher ratios of the two reporter activities (TOP/FOP) indicate higher transcriptional activity of βcatenin/TCF4. Relative to vector and other non-Ser31-HBxs, HBx-1 and HBx-2 resulted in higher β-catenin/TCF4 activity (Fig. 3E, Student's t-test, HBx-1 vs HBx-1A, HBx-1B, HBx-3, or HBx-4, P b 0.01; vector vs HBx-1 or HBx-2, P b 0.01). These data suggest that Akt activates Ser31-HBxs which then activate Akt and escalate oncogenic potential, that is, Ser31-HBx constitutes an oncogenic circuit with Akt. However, non-Ser31-HBx cannot promote oncogenic signal transduction. 3.7. Cooperation of HBx-Ser31 with activated ras to transform NIH3T3 cells Although Kim et al. showed that HBx induced HCC in transgenic mice [3], Lee et al. reported that HBx is not tumorigenic in transgenic mice [44]. We tested two isoforms HBx-1 (Ser31-HBx) and HBx-3 (Leu31-HBx) and found that the two isoforms did not induce tumor in transgenic mice. In addition, HBx-1 and HBx-3 did not transform mouse hepatocytes in primary culture even in cooperation with ras oncogene. Therefore, we adopted NIH3T3 cell line which has been widely exploited to study oncogene-mediated transformation. Although HBx may exert pro-apoptotic effect on hepatoma cells [11, 20–22], it has been shown to cooperate with activated ras to transform NIH3T3 cells [45]. To examine whether Ser31-HBx (HBx-1 and HBx-2) have higher oncogenic potential, we performed focus-forming assay by co-transfecting HBx with Ha-ras. As expected, co-transfection of HBx-1 and HBx-2 with Ha-ras resulted in higher number of colonies than non-Ser31 HBxs co-transfected with Ha-ras (Table 7, Student's ttest: HBx-1 + Ha-ras vs Ha-ras, P b 0.01; HBx-2 + Ha-ras vs Ha-ras. P b 0.01). HBxs without Ser31 appeared to suppress Ha-ras-mediated transformation (Table 7, Student's t-test: Ha-ras + HBx-1A, HBx-2A, HBx-3, or HBx-4 relative to Ha-ras alone, P b 0.01). Tumorigenesis assay (Table 8) showed that co-transfection of HBx-1 or HBx-2 with Ha-ras resulted in higher number of tumors in nude mice than nonSer31 HBxs co-transfected with Ha-ras. We compared tumor growth rates among Ha-ras, HBx-1 + Ha-ras, and HBx-1 A + Ha-ras and found that HBx-1 enhanced but HBx-1A suppressed Ha-ras-mediated tumor growth (Fig. 4A), statistically compared by analysis of variance (ANOVA, P b 0.01). The same results were obtained among Ha-ras, HBx-2 + Ha-ras, and HBx-2A + Ha-ras (Fig. 4B). These results suggest

4. Discussion Hepatitis B virus (HBV) X protein (HBx) has been involved in the development of hepatocellular carcinoma (HCC). In this report, we show that HBx with Akt phosphorylation site is a risk factor of HCC (Table 2), chronic hepatitis (Table 3), and cirrhosis (Table 4) by using the case–control study. HBV polymerase possesses two distinct functions, namely DNA-dependent RNA polymerase and RNA-dependent DNA polymerase. Both DNA and RNA syntheses lack proof-reading activity. Thus, mutation may occur and result in diverse isoforms of HBx and other viral proteins encoded by HBV DNA. While conducting the case–control study, we are aware that there are potential limitations in any case–control study and that the use of hospital controls is not ideal. Because the controls were not selected from the same population from which the cases arose, we could not rule out the possibility of selection bias. In addition, we restricted our analysis to Chinese subjects, so it is uncertain whether these results are applicable to other populations. There are still two factors we did not take into consideration, namely, host genetics and viral variations other than HBx. HCC may be caused by persistent destruction and regeneration of the liver. Cytotoxic T cell-mediated death of hepatocytes is the leading cause of liver injury after viral infection. Cytokine polymorphisms such as TNF-α [46] and IL-10 [47] have been associated with HCC development, and increased risk was observed in TNFA-308G/A, TNFA-238G/A, TNFA-863C/A, and IL-10-592A/C polymorphisms among Asians. In addition, the 61*G polymorphism in EGF is a risk factor for hepatocarcinogenesis while the EGF 61*A allele is a protective factor [48]. However, HCC risk in variations of HBV genome has not yet been elucidated. In this study, we showed that HBV genomes containing Ser31-HBx had higher efficiencies of replication and transcription (Fig. 2C-F), correlated with higher viral load (Fig. 2A) and clinical association between high viral load and HCC [41, 42]. The current study provides an example of association of HBV genome variations with HCC risk. HBx had opposite effects on ras-mediated transformation of NIH3T3 cells. Kim et al. showed that HBx abrogated Ha-ras-induced focus formation [22], however, another group showed that HBx cooperated with Ha-ras to transform NIH3T3 cells [45]; the former used the CMV promoter to drive HBx expression, and the latter used the SV40 promoter. The driving force of the SV40 promoter is weaker than that of the CMV promoter. It was explained that low level of HBx may cooperate with Ha-ras to transform cells, but high level may abrogate the effect [45]. Our results of focus-forming assay showed that HBx-1 and HBx-2 cooperated with Ha-ras to transform NIH3T3 cells (Table 7), suggesting that phosphorylation of HBx by Akt may determine oncogenic potential of HBx. We understand that NIH3T3 is a murine fibroblast, but it is a convenient method to test oncogenic potential of HBx isoforms. HCC often occurs many years after HBV infection, and HBV itself appears unable to transform hepatocytes. Multiple stages are required for HCC to develop in decades. HBx transgenic mice developed hepatic adenoma and later HCC [3,4], however, another group showed that HBx was not tumorigenic in transgenic mice [44]. In addition, HBx in some situations induced apoptosis [11,20–22], but contradictory antiapoptotic results were observed in other study groups [12–19]. Patients infected with HBV containing Ser31-HBx had higher rate of developing HCC than those infected with HBV with non-Ser31-HBx (Table 2), indicating that HBx isoforms play different roles in HCC development. This case–control study may explain why some HBx transgenic mice failed to develop HCC in literature. HBV biology is complicated inside hepatocytes. Inhibition of Akt has been shown to enhance HBV replication and transcription [49], contrary to our observation that Ser31-HBx and Akt constitute an oncogenic

846

W.-P. Lee et al. / Biochimica et Biophysica Acta 1862 (2016) 837–849

Fig. 3. Oncogenic circuit constituted by Ser31-HBx and Akt. A. HBxs with Ser31 (HBx-1 and HBx-2) activate MAP kinase (Erk) stronger than those without Ser31. HepG2 cells were transfected with FLAG-tagged HBx and then subjected to Western blotting with anti-Erk1, anti-P-Erk, and anti-FLAG antibodies. P-Erk band intensity was measured by NIH Image J, and Erk phosphorylation relative to HBx-1 was adjusted by the intensity of Erk input. B. HBxs with Ser31 activate Akt stronger than those without the site. HepG cell lysates in panel A were subjected to Western blotting with anti-Akt, anti-P-Thr308-Akt, anti-P-Ser473-Akt, and anti-FLAG antibodies. C. HBxs with Ser31 inactivate GSK-3β by Akt-mediated Ser9 phosphorylation stronger than those without the site. HepG cell lysates in panel A were subjected to Western blotting with anti-GSK-3β, anti-P-GSK-3β, and anti-FLAG antibodies. D. HBxs with Ser31 activate β-catenin/TCF-4-mediated transcription stronger than those without the site. HepG2 cells were transfected with FLAG-tagged HBx and then subjected to nuclear extract preparation. The nuclear extracts were subjected to Western blotting with anti-β-catenin and anti-histone3 antibodies (upper panel) and electrophoretic mobility shift assay (EMSA, lower panel). Compared with the lane of HBx-1, gel-shifted band was not observed in the lane of HBx-1*, of which the nuclear extract was incubated with mutant ds oligonucleotides. E. HBxs with Ser31 activate β-catenin/TCF-4-dependent transcription. HepG2 cells were transfected with FLAG-tagged HBx, pTOPFLASH (or pFOPFLASH), and pTKrenilla luciferase plasmids. 48 h after transfection, cells were subjected to the dual luciferase assay system. β-catenin/TCF-4-mediated transcriptional activity was determined by the ratio of luciferase value of TOP/FOP. Vector transfection was set as 1. Each bar was the mean of triplicate assays.

circuit that enhances HBV replication and transcription. HBV and HepG2 cells use the same cellular machinery to produce DNA, RNA and protein. Akt is a major kinase in maintaining cell growth and survival. The mTOR pathway driven by Akt signaling involves protein synthesis and RNA transcription [50]. Inhibition of Akt by inhibitors or siAkt interferes with cellular synthesis of protein and RNA but may provide environment for viral DNA replication and RNA transcription [49]. Cell growth is inhibited, and cells then go to apoptosis under Akt inhibitor such as

MK-2206 [51,52]. Thus, abrogation of Akt kinase function with a dose of an inhibitor detrimental to cellular protein and RNA syntheses may transiently benefit viral propagation of certain HBV isoforms before the host cells undergo apoptosis [49]. On the other hand, overexpression of Akt1 in HBV-propagating HepG2 cell system does not affect HBV replication and transcription [49], both of which decrease under myr-Akt overexpression. Myr-Akt is a constitutively activated Akt, the event of which is not physiologically present in human liver and also

W.-P. Lee et al. / Biochimica et Biophysica Acta 1862 (2016) 837–849

847

Table 7 Focus-forming assay of NIH3T3 cells transfected with Ha-ras + HBx. Plasmids transfected

No. of colonies

Vector Ha-Ras HBx-1 HBx-1 + Ha-Ras HBx-2 HBx-2 + Ha-ras HBx-1A HBx-1A + Ha-ras HBx-2A HBx-2A + Ha-ras HBx-3 HBx-3 + Ha-ras HBx-4 HBx-4 + Ha-ras

5 ± 0.4 38 ± 4.6 4 ± 0.6 65 ± 7.1 6 ± 0.7 61 ± 8.2 5 ± 0.4 12 ± 1.6 6 ± 0.8 16 ± 1.9 4 ± 0.5 21 ± 2.8 7 ± 0.8 19 ± 2.1

Focus-forming assay of NIH3T3 cells transfected with Ha-ras + HBx. NIH3T3 cells were transfected with plasmids as shown in the figure, plated on 10-cm tissue culture dishes and then selected with G418 for one week. The number of colonies (N20 mm in diameter) was counted 3 weeks after plasmid transfection. No of colonies was the mean of triplicate samples.

pathologically not present in HBV-infected hepatocytes. Myr-Akt signaling benefits host cells in the syntheses of cellular DNA, RNA and protein but may prevent HBV replication and transcription [49] because HBV genome is too simple to get benefit from the signaling that may divert limited energy sources for host cellular biosynthesis. In our study, Ser31-HBx and Akt interact each other and augment endogenous kinase activity of Akt. Under such a circumstance, HBV replication and transcription are enhanced by P-Ser31-HBx. In chronic HBV infection, HCC develops in 10–20% of livers without cirrhosis or inflammation [53], suggesting that virus by itself may be one of the carcinogenic factors. Proteins encoded by HBV genome have oncogenic potential [54], but the effect of HBV proteins in HCC development is hard to be evaluated in chronically infected liver because oncogenic HBV proteins cannot be studied without the presence of immune response to the virus. HCC is associated with high serum HBsAg level [55]. Accumulation of mutant HBs proteins in endoplasmic reticulum (ER) induces ER stress that may shut down protein synthesis and drive cell into apoptosis, leading to histological feature of ground-glass hepatocytes [56]. High HBsAg level is correlated with pre-neoplastic inflammation and HCC development, and chronic inflammation increases frequency of integration of HBV genome into host hepatocyte genome [57]. These observations explain that HBx is expressed at low levels in HBV-infected liver but frequently detected Table 8 Tumorigenesis of NIH3T3 cells transfected with Ha-ras + HBx. Plasmids transfected

No. of tumors

Vector Ha-Ras HBx-1 HBx-1 + Ha-Ras HBx-2 HBx-2 + Ha-ras HBx-1A HBx-1A + Ha-ras HBx-2A HBx-2A + Ha-ras HBx-3 HBx-3 + Ha-ras HBx-4 HBx-4 + Ha-ras

0/8 8/8 0/8 8/8 0/8 7/8 0/8 3/8 0/8 3/8 0/8 4/8 0/8 3/8

Tumorigenesis of NIH3T3 cells transfected with Ha-ras + HBx. NIH3T3 cells were transfected with specified plasmids, selected with G418 for one week, and then injected to 4 nude mice at both flank fat pad. Number of tumors (N20 mm3) was counted three weeks after cell injection.

Fig. 4. Tumorigenesis assays showing that HBx with Ser31 promotes tumor growth in nude mice. A. Growth rate of NIH3T3 cells transfected with Ha-ras + HBx. NIH3T3 cells were transfected with plasmids (Ha-ras, HBx-1 + Ha-ras, or HBx-1 A + Ha-ras), selected with G418 for one week, and then harvested by trypsinization. Cells were washed twice with sterile phosphate-buffered saline and resuspended at 2 × 107 cells per ml. Aliquots (0.1 ml) were injected subcutaneously into 6- to 8-week-old Balb/c athymic nude mice. Tumors were measured at periodic intervals by the hemi-ellipsoid model, typical as tumor volume = (π / 6)(length) × (width) × (height). Lines of Ha-ras vs HBx-1 + Ha-ras and Ha-ras vs HBx-1 A + Ha-ras were statistically compared by analysis of variance (ANOVA). Western blot on the left upper quadrant showed expression of Ha-ras and HBx-1 in tumors. B. The same experiment was performed as described in panel A using the plasmids Ha-ras, HBx-2 + Ha-ras, and HBx-2 A + Ha-ras.

at high levels in HBV-related HCCs [58]. In this study, we selected two HBx isoforms with Akt phosphorylation site, i.e. HBx-1 and HBx-2 that can be phosphorylated by Akt and activate MAP kinase, Akt, and βcatenin/TCF4-mediated signaling pathway (Figs. 1 and 3). In addition, HBx-1 and HBx-2 also cooperated with Ha-ras to transform NIH3T3 cells (Tables 7 and 8). Ser31-HBx and Akt constitute an oncogenic circuit by which the two proteins activate each other and may lead to accumulation of HBV DNA integration into genomes of proliferative hepatocytes during chronic hepatitis. Our study shows that Ser-31-HBx is associated with chronic hepatitis as well as HCC. We searched the gene bank of NCBI and found that approximately 60% of HBx contains Akt phosphorylation site. It appears that the Ser31 phosphorylation is a major carcinogenic motif that accounts for the controversial results in literature, such as tumorigenic versus nontumorigenic in nude mice and anti-apoptotic and pro-apoptotic in response to drugs, UV, or serum deprivation. The biological effect of HBx on HCC development remains unclear. Both the host and virus play important roles. In supplementary data, we show further evidence of dual role of HBx, i.e. proapoptotic and anti-apoptotic activities (sFig.2). Ser31-HBx has tumorigenic activity, however, non-Ser31-HBx is a tumor suppressor (sFig. 3 A and 3B). In this study, we isolated and

848

W.-P. Lee et al. / Biochimica et Biophysica Acta 1862 (2016) 837–849

cloned more than 50 HBx DNAs from patients' blood samples. We observed that Ser31-HBx and non-Ser31-HBx did coexist in the serum of a single patient. Because these cases were a minor portion emerging occasionally in blood samples tested in our laboratory, it is difficult to determine how the two co-existent HBxs affect each other in clinical situations. We tested co-expression of the two HBxs in HepG2 cells and found that Ser31-HBx overcomes Leu31-HBx-mediated tumor suppression (sFig. 4A and 4B). In conclusion, the current study provides a mechanism of HBV-induced HCC in that phosphorylation of HBx by Akt increases HCC risk and oncogenic potential of HBV. Conflict of interest The authors declare no conflict of interest. Transparency document The Transparency document associated with this article can be found, in the online version. Acknowledgments This study was supported by the National Science Council, Taiwan (NSC96-2314-B- 075-072 to Wei-Ping Lee) and the Taipei Veterans General Hospital, Taiwan (V97C1–102, V97F-004, and V98C1–114 to Wei-Ping Lee). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.bbadis.2015.12.020. References [1] A.S. Lok, E.J. Heathcote, J.H. Hoofnagle, Management of hepatitis B: 2000—summary of a workshop, Gastroenterology 120 (2010) 1828–1853. [2] S. Murakami, Hepatitis B virus X protein: a multifunctional viral regulator, J. Gastroenterol. 36 (2001) 651–660. [3] C.M. Kim, K. Koike, I. Saito, T. Miyamura, G. Jay, HBx gene of hepatitis B virus induces liver cancer in transgenic mice, Nature (London) 351 (1991) 317–320. [4] Y. Wang, F. Cui, Y. Lv, C. Li, X. Xu, C. Deng, D. Wang, et al., HBsAg and HBx knocked into the p21 locus causes hepatocellular carcinoma in mice, Hepatology 39 (2004) 318–324. [5] J. Diao, R. Garces, C.D. Richardson, X protein of hepatitis B virus modulates cytokine and growth factor related signal transduction pathways during the course of viral infections and hepatocarcinogenesis, Cytokine Growth Factor Rev. 12 (2001) 189–205. [6] C. Wang, W. Yang, H.X. Yan, T. Luo, J. Zhang, L. Tang, et al., Hepatitis B virus X (HBx) induces tumorigenicity of hepatic progenitor cells in 3,5-diethoxycarbonyl-1,4dihydrocollidine-treated HBx transgenic mice, Hepatology 55 (2012) 108–120. [7] D.K. Lee, S.H. Park, Y. Yi, S.G. Choi, C. Lee, W.T. Parks, et al., The hepatitis B virus encoded oncoprotein pX amplifies TGF-beta family signaling through direct interaction with Smad4: potential mechanism of hepatitis B virus-induced liver fibrosis, Genes Dev. 15 (2001) 455–466. [8] I. Haviv, M. Shamay, G. Doitsh, Y. Shaul, Hepatitis B virus pX targets TFIIB in transcription coactivation, Mol. Cell. Biol. 18 (1998) 1562–1569. [9] X.W. Wang, K. Forrester, H. Yeh, M.A. Feitelson, J.R. Gu, C.C. Harris, Hepatitis B virus X protein inhibits p53 sequence-specific DNA binding, transcriptional activity, and association with transcription factor ERCC3, Proc. Natl. Acad. Sci. U. S. A. 91 (1994) 2230–2234. [10] S.A. Becker, T.H. Lee, J.S. Butel, B.L. Slagle, Hepatitis B virus X protein interferes with cellular DNA repair, J. Virol. 72 (1998) 266–272. [11] A.T. Lee, J. Ren, E.T. Wong, K.H. Ban, L.A. Lee, C.G. Lee, The hepatitis B virus X protein sensitizes HepG2 cells to UV light-induced DNA damage, J. Biol. Chem. 280 (2005) 33525–33535. [12] C. Shan, F. Xu, S. Zhang, J. You, X. You, L. Qiu, et al., Hepatitis B virus X protein promotes liver cell proliferation via a positive cascade loop involving arachidonic acid metabolism and p-ERK1/2, Cell Res. 20 (2010) 563–575. [13] T.W. Chung, Y.C. Lee, C.H. Kim, Hepatitis B viral HBx induces matrix metalloproteinase-9 gene expression through activation of ERK and PI-3K/AKT pathways: involvement of invasive potential, FASEB J. 18 (2004) 1123–1125. [14] Y.I. Lee, S. Kang-Park, S.I. Do, Y.I. Lee, The hepatitis B virus-X protein activates a phosphatidylinositol 3-kinase-dependent survival signaling cascade, J. Biol. Chem. 276 (2001) 16969–16977.

[15] W.L. Shih, M.L. Kuo, S.E. Chuang, A.L. Cheng, S.L. Doong, Hepatitis B virus X protein activates a survival signaling by linking SRC to phosphatidylinositol 3-kinase, J. Biol. Chem. 27834 (2003) 31807–31813. [16] K.H. Lim, H.S. Choi, Y.K. Park, E.S. Park, G.C. Shin, D.H. Kim, et al., HBx-induced NF-κB signaling in liver cells is potentially mediated by the ternary complex of HBx with p22-FLIP and NEMO, PLoS One 8 (2013), e57331. [17] P. Chirillo, M. Falco, P.L. Puri, M. Artini, C. Balsano, M. Levrero, et al., Hepatitis B virus pX activates NF-κB-dependent transcription through a Raf-independent pathway, J. Virol. 70 (1996) 641–646. [18] M.Y. Cha, C.M. Kim, Y.M. Park, W.S. Ryu, Hepatitis B virus X protein is essential for the activation of Wnt/beta-catenin signaling in hepatoma cells, Hepatology 39 (2004) 1683–1693. [19] L. Shen, X. Zhang, D. Hu, T. Feng, H. Li, Y. Lu, et al., Hepatitis B virus X (HBx) plays an anti-apoptosis role in hepatic progenitor cells by activating Wnt/β-catenin pathway, Mol. Cell. Biochem. 383 (2013) 213–222. [20] F. Su, R.J. Schneider, Hepatitis B virus HBx protein sensitizes cells to apoptotic killing by tumor necrosis factor alpha, Proc. Natl. Acad. Sci. U. S. A. 94 (1997) 8744–8749. [21] H. Zhang, C. Huang, Y. Wang, Z. Lu, N. Zhuang, D. Zhao, et al., Hepatitis B virus X protein sensitizes TRAIL-induced hepatocyte apoptosis by inhibiting the E3 ubiquitin ligase A20, PLoS One 10 (2015), e0127329. [22] H. Kim, H. Lee, Y. Yun, X-gene product of hepatitis B virus induces apoptosis in liver cells, J. Biol. Chem. 273 (1998) 381–385. [23] J.D. Yang, L.R. Roberts, Hepatocellular carcinoma: a global view, Nat. Rev. Gastroenterol. Hepatol. 7 (2010) 448–458. [24] F.V. Chisari, K. Klopchin, T. Moriyama, Molecular pathogenesis of hepatocellular carcinoma in hepatitis B virus transgenic mice, Cell 59 (1989) 1145–1156. [25] H.A. Dunsford, S. Sell, F.V. Chisari, Hepatocarcinogenesis due to chronic liver cell injury in hepatitis B virus transgenic mice, Cancer Res. 50 (1990) 3400–3407. [26] H. Zhu, Y. Wang, J. Chen, G. Cheng, J. Xue, Transgenic mice expressing hepatitis B virus X protein are more susceptible to carcinogen induced hepatocarcinogenesis, Exp. Mol. Pathol. 76 (2004) 44–50. [27] C.R. Madden, M.J. Finegold, B.L. Slagle, Hepatitis B virus X protein acts as a tumor promoter in development of diethylnitrosamine-induced preneoplastic lesions, J. Virol. 75 (2001) 3851–3858. [28] W. Chotiyaputta, A.S. Lok, Hepatitis B virus variants, Nat. Rev. Gastroenterol. Hepatol. 6 (2009) 453–462. [29] M. Ashcroft, R.L. Ludwig, D.B. Woods, T.D. Copeland, H.O. Weber, E.J. MacRae, et al., Phosphorylation of HDM2 by Akt, Oncogene 21 (2002) 1955–1962. [30] B.P. Zhou, Y. Liao, W. Xia, Y. Zou, B. Spohn, M.C. Hung, HER-2/neu induces p53 ubiquitination via Akt-mediated MDM2 phosphorylation, Nat. Cell Biol. 3 (2001) 973–982. [31] D.A. Cross, D.R. Alessi, P. Cohen, M. Andjelkovich, Hemmings BA. Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B, Nature (London) 378 (1995) 785–789. [32] S.R. Datta, H. Dudek, X. Tao, S. Masters, H. Fu, Y. Gotoh, et al., Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery, Cell 91 (1997) 231–241. [33] M.H. Cardone, N. Roy, H.R. Stennicke, G.S. Salvesen, T.F. Franke, E. Stanbridge, et al., Regulation of cell death protease caspase-9 by phosphorylation, Science 282 (1998) 1318–1321. [34] O.N. Ozes, L.D. Mayo, J.A. Gustin, S.R. Pfeffer, L.M. Pfeffer, D.B. Donner, NF-kappaB activation by tumour necrosis factor requires the Akt serine–threonine kinase, Nature 401 (1999) 82–85. [35] W.P. Lee, K.H. Lan, C.P. Li, Y. Chao, H.C. Lin, S.D. Lee, Pro-apoptotic or anti-apoptotic property of X protein of hepatitis B virus is determined by phosphorylation at Ser31 by Akt, Arch. Biochem. Biophys. 528 (2012) 156–162. [36] C.T. Yeh, C.H. Shen, D.I. Tai, C.M. Chu, Y.F. Liaw, Identification and characterization of a prevalent hepatitis B virus X protein mutant in Taiwanese patients with hepatocellular carcinoma, Oncogene 19 (2000) 5213–5220. [37] C.E. Stevens, R.A. Neurath, R.P. Beasley, W. Szmuness, HBeAg and anti-HBe detection by radioimmunoassay: correlation with vertical transmission of hepatitis B virus in Taiwan, J. Med. Virol. 3 (1979) 237–241. [38] A. Bartholomeusz, S. Schaefer, Hepatitis B virus genotypes: comparison of genotyping methods, Rev. Med. Virol. 14 (2004) 3–16. [39] M.M. Tomayko, C.P. Reynolds, Determination of subcutaneous tumor size in athymic (nude) mice, Cancer Chemother. Pharmacol. 24 (1989) 148–154. [40] E. Khattar, A. Mukherji, V. Kumar, Akt augments the oncogenic potential of the HBx protein of hepatitis B virus by phosphorylation, FEBS J. 279 (2012) 1220–1230. [41] H.L. Chan, V.W. Wong, G.L. Wong, A.M. Chim, L.H. Lai, J.J. Sung, Evaluation of impact of serial hepatitis B virus DNA levels on development of hepatocellular carcinoma, J. Clin. Microbiol. 47 (2009) 1830–1836. [42] C.J. Liu, B.F. Chen, P.J. Chen, M.Y. Lai, W.L. Huang, J.H. Kao, et al., Role of hepatitis B viral load and basal core promoter mutation in hepatocellular carcinoma in hepatitis B carriers, J. Infect. Dis. 193 (2006) 1258–1265. [43] V. Korinek, N. Barker, P.J. Morin, D. van Wichen, R. de Weger, K.W. Kinzler, et al., Constitutive transcriptional activation by a beta-catenin-Tcf complex in APC−/− colon carcinoma, Science 275 (1997) 1784–1787. [44] T.H. Lee, M.J. Finegold, R.F. Shen, J.L. DeMayo, S.L. Woo, J.S. Butel, Hepatitis B virus transactivator X protein is not tumorigenic in transgenic mice, J. Virol. 64 (1990) 5939–5947. [45] Y.C. Kim, K.S. Song, G. Yoon, M.J. Nam, W.S. Ryu, Activated ras oncogene collaborates with HBx gene of hepatitis B virus to transform cells by suppressing HBx-mediated apoptosis, Oncogene 20 (2001) 16–23. [46] Y. Wei, F. Liu, B. Li, X. Chen, Y. Ma, L. Yan, et al., Polymorphisms of tumor necrosis factor-alpha and hepatocellular carcinoma risk: a HuGE systematic review and meta-analysis, Dig. Dis. Sci. 56 (2011) 2227–2236.

W.-P. Lee et al. / Biochimica et Biophysica Acta 1862 (2016) 837–849 [47] Y.G. Wei, F. Liu, B. Li, X. Chen, Y. Ma, L.N. Yan, et al., Interleukin-10 gene polymorphisms and hepatocellular carcinoma susceptibility: a meta-analysis, World J. Gastroenterol. 17 (2011) 3941–3947. [48] J.H. Zhong, X.M. You, W.F. Gong, L. Ma, Y. Zhang, Q.G. Mo, et al., Epidermal growth factor gene polymorphism and risk of hepatocellular carcinoma: a meta-analysis, PLoS One 7 (2012), e32159. [49] H. Guo, T. Zhou, D. Jiang, A. Cuconati, G.H. Xiao, T.M. Block, et al., Regulation of hepatitis B virus replication by the phosphatidylinositol 3-kinase-akt signal transduction pathway, J. Virol. 81 (2007) 10072–10080. [50] M. Shimobayashi, M.N. Hall, Making new contacts: the mTOR network in metabolism and signalling crosstalk, Nat. Rev. Mol. Cell Biol. 15 (2014) 155–162. [51] F. Ewald, N. Grabinski, A. Grottke, S. Windhorst, D. Nörz, L. Carstensen, et al., Combined targeting of AKT and mTOR using MK-2206 and RAD001 is synergistic in the treatment of cholangiocarcinoma, Int. J. Cancer 133 (2013) 2065–2076. [52] P. Jiao, Y.S. Zhou, J.X. Yang, Y.L. Zhao, Q.Q. Liu, C. Yuan, et al., MK-2206 induces cell cycle arrest and apoptosis in HepG2 cells and sensitizes TRAIL-mediated cell death, Mol. Cell. Biochem. 382 (2013) 217–224.

849

[53] G. Fattovich, T. Stroffolini, I. Zagni, F. Donato, Hepatocellular carcinoma in cirrhosis: incidence and risk factors, Gastroenterology 127 (2004) 35–50. [54] C. Xu, W. Zhou, Y. Wang, L. Qiao, Hepatitis B virus-induced hepatocellular carcinoma, Cancer Lett. 345 (2014) 216–222. [55] T.C. Tseng, C.J. Liu, H.C. Yang, T.H. Su, C.C. Wang, C.L. Chen, et al., High levels of hepatitis B surface antigen increase risk of hepatocellular carcinoma in patients with low HBV load, Gastroenterology 142 (2012) 1140–1149. [56] M. Wang, R.J. Kaufman, The impact of the endoplasmic reticulum protein-folding environment on cancer development, Nat. Rev. Cancer 14 (2014) 581–597. [57] W.K. Sung, H. Zheng, S. Li, R. Chen, X. Liu, Y. Li, et al., Genome-wide survey of recurrent HBV integration in hepatocellular carcinoma, Nat. Genet. 44 (2012) 765–769. [58] P. Paterlini, K. Poussin, M. Kew, D. Franco, C. Brechot, Selective accumulation of the X transcript of hepatitis B virus in patients negative for hepatitis B surface antigen with hepatocellular carcinoma, Hepatology 21 (1995) 313–321.