Human embryonic kidney 293T cell line was provided by ATCC (UK). The human hepatoma cell line HepG2 cell was purchased from Shanghai Cell Bank (Shanghai, China). Cells were cultured in Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum and 1% penicillin-streptomycin solution at 37°C in a 5% CO₂ incubator.

Anti-DYKDDDDK-Tag (anti-flag) antibody was purchased from Abmart (Shanghai, China). Anti-COX-2 antibody was purchased from Abcam (Co., USA). Anti-cytochrome c oxidase subunit III (anti-COXIII) (N-20) antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-β-actin antibodies were purchased from ZSGB-BIO (Beijing, China). CFTM488-conjugated donkey anti-mouse IgG (H+L), CFTM350-conjugated donkey anti-goat IgG (H+L), N-acetylcysteine (NAC) and puromycin were obtained from Sigma (St. Louis, MO). Mito Tracker Red was obtained from Molecular Probes (Eugene, OR). NS-398 was purchased from Beyotime (Shanghai, China). Cell counting kit-8 was from Dojindo Laboratories (Kumamoto, Japan). Primers were synthesized by Biosune (Shanghai, China).

The HBx expression plasmid PcDNA3.1-x (HBV subtype ayw) was a gift from Professor Michael J. Bouchard (Drexel University College of Medicine, Philadelphia, PA). The lentivirus packaging system: pLOV.CMV.eGFP.2A.EF1a.PuroR (pLOV), psPAX2 and pMD2.G were provided by Neuron Biotech Co., Ltd. (Shanghai, China). The lentivirus vector pLOV contains the gene coding for green fluorescent protein (GFP), puromycin and the peptide DYKDDDDK (flag). Full-length HBx (465 bp) was amplified from PcDNA3.1-x by polymerase chain reaction (PCR) and then infused with flag epitope at the N-terminus and next cloned into pLOV. The recombinant lentiviral particles were generated by transient co-transfection with 293T cells by the three-plasmid expression system, and harvested by filtration through a 0.45-μm filter and ultracentrifugation at 100,000 g for 2 h at 4°C. HepG2 cells at 60% confluency were transfected with the recombinant lentivirus. Then the cell clones were treated with 0.2 μg/ml puromycin for 10 days. Positive clones expressing GFP and resistant to puromycin were screened and named HepG2-HBx and HepG2-mock, respectively.

To eliminate the impact of green fluorescence caused by GFP in HepG2-HBx cells, we used 100% methanol as a fixative. As a result, green fluorescence disappeared (data not shown), consistent with previous research (33). Cells were stained for 30 min with 150 nM Mito Tracker Red and fixed with a 100% ice methanol solution and pre-incubated in blocking solution (5% donkey serum albumin in PBS) and next incubated with primary antibodies (anti-flag, anti-COXIII) at 4°C overnight. Then, the fluorescence-labeled second antibodies (CFTM488-conjugated donkey anti-mouse, CFTM350-conjugated donkey anti-goat) were added and incubated for 60 min at 37°C in the dark, and the sections were mounted using glycerol. Fluorescence images were obtained by confocal laser scanning microscopy (Leica SP5, Solms, Germany).

Mitochondria were isolated using the mitochondrial isolation kit for mammalian cells (Thermo Scientific, MA, USA) according to the manufacturer’s instructions. Briefly, following lysis of approximately 2×10⁷ cells, cell debris and nuclei were pelleted at 700 × g for 10 min at 4°C, followed by centrifugation at 3,000 × g for 15 min at 4°C to pellet a mitochondria-enriched fraction, and then 12,000 × g for 5 min at 4°C to pellet the isolated mitochondria. Protein concentrations were measured using the bicinchoninic acid (BCA) assay and then adjusted to 1 μg/μl. COX activity was determinated using the Cytochrome c Oxidase Assay kit (GenMed, Shanghai, China) according to the manufacturer’s instructions by an enzyme mark instrument (Bio-Tek ELX-800, USA). Isolated mitochondria (10 μl) were combined with 10 μl of lysis buffer, then mixed with 205 μl of buffer solution in 96-well plates. The reaction was initiated by the addition of 25 μl of reaction liquid, and the decrease in absorbance at 550 nm was measured for 1 min. Activity was calculated based on the following equation: Units/ml = [(ΔAbs550/min for the sample - ΔAbs550/min for the blank) × dilution factor × total reaction volume]/[mitochondria isolate volume x the difference in extinction coefficients between ferro- and ferri-cytochrome c at 550 nm (21.84)]. One unit is the amount that oxidizes 1 μmol reduced cytochrome c/min at pH 7.0 and 25°C.

Cells in the three groups were harvested with trypsin, and resuspended in PBS (1×10⁶ cells/ml) for analyzing ROS and mitochondrial membrane potential by flow cytometry (Becton-Dickson Co., C6, San Jose, CA). i) The mitochondrial membrane potential was evaluated using the cationic fluorescent dye TMRM (Sigma). Cells incubated in 100 nM TMRM solution, 20 min at 37°C in the dark, then analysed by a flow cytometer. ii) Intracellular ROS levels were detected by staining cells with 50 μM DihyDroeThidium (DHE) fluorescence probe (Vigorous Biotechnology, Beijing, China) for 30 min in the dark, then analysed by a flow cytometer.

The morphological changes of cell and mitochondria were observed by TEM. A total of 1×10⁶ of cells were collected and subjected to fixation with 3% fresh glutaraldehyde and 1.5% paraformaldehyde solution at 4°C for 1 h, next to post-fixation with 1% osmium tetroxide and 1.5% potassium ferrocyanide solution at 4°C for 1.5 h, then dehydration was carried out with a graded series of ethanol solution, and finally the cells were to embedded in Epon-618. Ultrathin sections were cut to stain with uranyl acetate and lead citrate, and observed under TEM (Phillips EM208, WI, USA).

i) RT-PCR: total RNA was extracted using TRIzol reagent. PCR amplification reactions were set up according to the manufacturer’s protocol (Thermo Scientific). The primer sequences of each gene and the PCR conditions are listed in Table I. PCR was carried out in a GeneAmp PCR System (PE2400, USA). The results of electrophoresis were scanned by a UVP scanner with Grab-IT software (Gel Doc 100, Bio-Rad, USA) and gray scale of each band was calculated using Image J software. ii) Western blotting: cells were harvested and cell lysate was prepared. Protein concentration was measured using BCA Protein Assay kit (Beyotime). Immunoblotting was performed using specific primary antibodies then incubated with the appropriate horseradish peroxidase-conjugated secondary antibodies and detected using an ECL kit (ZSGB-BIO). Densitometry was performed on the western blot images by using Image-J software and expressed as arbitrary densitometric units normalised to β-actin expression.

Cell proliferation was analyzed by the cell-counting kit-8. Cells (3×10³) were seeded in 96-well plates and each well contained 100 μl of complete DMEM medium (6 replica wells for each cell line). After 1, 2 and 3 days of incubation, the media was removed and 100 μl of fresh complete DMDM medium containing premixed CCK-8 (110 dilution in complete DMEM medium) was added to each well. The plates were incubated at 37°C with 5% CO₂ for 3 h, followed by shaking thoroughly for 1 min on a shaker. The absorbance at 450 nm was measured using an ELISA reader, each day of the 3-day experiment to calculate the growth curve.

Cell suspensions were transferred into 6-well plates (500 cells/well), and each group contained 3 wells. After incubation for 14 days, the cells were rinsed in PBS for twice and stained with Crystal Violet Staining Solution. The experiments were performed in triplicate and the number of colonies containing >50 cells were microscopically counted to calculate the colony formation rate as number of colonies/number of cells × 100%.

Each set of experiments was repeated at least three times with similar results. Results are expressed as means with standard deviations (SD). Statistical evaluation was carried out by one-way analysis of variance (ANOVA). P-values of <0.05 were considered statistically significant.

Lentiviral vectors have become one of the most powerful gene transfer vehicles, they provide several advantages that include being able to transduce a wide range of cell types irrespective of their division status, leading to prolonged transgene expression, lack of pre-existing immunity and less genotoxic (34,35). Our study used lentiviral vectors to establish a HepG2 cell line stably expressing HBx gene in vitro. The GFP positive cells were 95–98% by fluorescence microscope (Fig. 1A). Stable expression the HBx mRNA and protein was verified by RT-PCR and western blot, respectively (Fig. 1B). A HepG2 cell line stably transduced with a lentivirus expressing the HBx gene was successfully constructed.

We proceeded with fluorescent staining experiments by a combination of anti-flag antibody, anti-COXIII antibody and a mitochondria-specific staining agent, Mito Tracker Red, then observed cells by confocal microscopy. As indicated in Fig. 2, the staining of HBx protein (green color) together with COXIII protein (blue color) in mitochondria (red color) was observed (white) in HepG2-HBx cells, which suggested that HBx protein co-localized with inner mitochondrial membrane protein COXIII in HepG2 cells, further verifying our previous studies (12,13). Moreover, the localization of HBx was both in the nucleus and the cytoplasm, with a fraction of cytosolic HBx in the mitochondria.

Since COXIII is an essential component of the mitochondrial respiratory chain, the co-localization of HBx with COXIII would induce an alteration in the normal mitochondrial function and morphology in HepG2-HBx cells. We found that the protein expression of COXIII was increased in HepG2-HBx cells, while the RNA level did not change as shown in Fig. 3A and B. We then examined the COX activities. Correlating with their upregulated proteins level, they were significantly increased in HepG2-HBx cells compared with HepG2-control and HepG2-mock cells (Fig. 3C). Next, we measured mitochondrial membrane potencial (ΔΨm) of cells in the three groups using the fluorochrome TMRM, HepG2-HBx cells displayed the highest ΔΨm (Fig. 3D). Increased COXIII protein expression level, upregulation of COX activities as well as the highest ΔΨm demonstrated that HBx promoted mitochondrial function in HepG2 cells. To assess the changes of mitochondrial morphology accompanied by the alteration of mitochondrial biogenesis in HepG2-HBx cells, we observed the mitochondria, nuclei and other intracellular membrane structures by TEM. Results showed that only a detectable slight swelling of mitochondria in HepG2-HBx cells was found compared with the other two groups (Fig. 3E). Thus, our studies demonstrated that HBx co-localized with inner mitochondrial protein COXIII in HepG2 cells, leading to the changes of mitochondrial biogenesis and morphology.

Mitochondria are major ROS generators, thus changes in mitochondrial function and morphology by HBx would be expected to cause ROS generation in cells (36). We evaluated the ROS level by flow cytometry with the DihyDroeThidium (DHE) probe. As shown in Fig. 4A, the ROS generation was the strongest in HepG2-HBx cells compared to the others. Recent studies have shown that ROS from mitochondria is necessary for COX-2 induction (28,32). We then analyzed whether COX-2 gene expression increased with the ROS generation. Fig. 4B and C show that expression of both COX-2 mRNA and protein was increased in HepG2-HBx cells. To further confirm the notion that ROS from mitochondria induced cyclooxygenase-2 gene expression in HepG2-HBx cells, we treated cells with the antioxidant, N-acetyl cysteine (NAC) (10 mM) for 12 h. As can be seen from Fig. 4D, the induction of COX-2 protein expression by HBx was decreased in HepG2-HBx cells with NAC, but still higher than the other two groups. Therefore, it was clear that ROS from mitochondria induce COX-2 gene expression in HepG2-HBx cells, resulting in inflammatory injury.

Substantial research in recent years has demonstrated that HBx promoted cell proliferation associating with HCC in different models and conditions (37,38). To observe the effects of HBx on cell proliferation, we performed the cell viability assay and plate colony formation assay. As shown in Fig. 5A and B, HepG2-HBx cells increased the proliferation rate and formed more colonies, indicating that HBx promoted the proliferation of HepG2 cells. Previous studies have pointed out that COX-2 stimulates proliferation, angiogenesis, invasiveness and inhibits apoptosis to promote the growth of HCC cells (39). We considered that the upregulation of COX-2 contributed to the proliferation mediated by HBx. Further studies were carried out by treating cells with the selective COX-2 inhibitor NS-398 (50 μmol/l) for 72 h. The proliferation of HepG2-HBx cells were significantly suppressed following the inhibition of COX-2 protein expression (Fig. 5C and D). Of note, our data strongly suggest that HBx induced cell proliferation through inflammation.