Cancer Cell Biology
The hepatitis B virus X protein promotes hepatocellular carcinoma metastasis by upregulation of matrix metalloproteinases
Article first published online: 22 DEC 2006
Copyright © 2006 Wiley-Liss, Inc.
International Journal of Cancer
Volume 120, Issue 6, pages 1208–1214, 15 March 2007
How to Cite
Ou, D.-P., Tao, Y.-M., Tang, F.-Q. and Yang, L.-Y. (2007), The hepatitis B virus X protein promotes hepatocellular carcinoma metastasis by upregulation of matrix metalloproteinases. Int. J. Cancer, 120: 1208–1214. doi: 10.1002/ijc.22452
- Issue published online: 30 JAN 2007
- Article first published online: 22 DEC 2006
- Manuscript Accepted: 10 OCT 2006
- Manuscript Received: 18 JUL 2006
- National Natural Science Foundation of China. Grant Number: 30371595
- National Key Technologies R&D Program. Grant Number: 2001BA703B04
- China Basic Research Program. Grant Number: 2004CB720303
- Specialized Research Fund for Doctoral Program of Higher Education. Grant Number: 20030533037
- invasion and metastasis;
- signal transduction pathway
The hepatitis B virus (HBV) is a major cause of human hepatocellular carcinoma (HCC) which has a very high mortality rate due to high incidence of metastasis. It is unknown whether HBV contributes to HCC metastasis. In this report, we present clinical data obtained from HCC patients indicating that the expression of hepatitis B virus X protein (HBx) in HCC is associated with an increased expression of membrane-type 1 matrix metalloproteinase (MT1-MMP), and matrix metalloproteinase-2(MMP-2), which correlates with a poor prognosis. We further demonstrate experimentally that HBx upregulates MT1-MMP, which in turn induces MMP-2. Significantly, HBx-mediated MMP activation is associated with a marked increase of cell migration, as revealed by both wound-healing and transwell migration assays, suggesting that HBx may facilitate tumor cell invasion by upregulation of MMPs and subsequent destruction of the extracellular matrix. Together, our results support a model in which HBx contributes to HCC metastasis by upregulation of MMPs. © 2006 Wiley-Liss, Inc.
Hepatocellular carcinoma (HCC), is one of the most common malignant tumors in Asia, and is frequently a terminal complication of chronic inflammatory and fibrotic liver disease.1 The mortality rate for HCC is high because of its high rate of metastasis. HBx, a small regulatory protein required for viral infection,2 contributes to the development of HCC.3, 4 However, the molecular mechanisms for HBx-mediated carcinogenesis remain obscure. HBx has been shown to have a variety of biological functions, including transcriptional activation of a variety of viral and cellular promoters, its interaction with p53, interference with host DNA repair, repression of physiological proteolysis, modulation of cell proliferation and apoptosis, and induction of malignant cell migration.5, 6, 7, 8, 9 These functions may contribute to the initiation and development of HCC associated with a hepatitis B virus (HBV) infection.
The invasion and metastasis of cancer cells is a multi-step process.10 HCC frequently invades blood vessels early and metastasize intra-hepatically and later extra-hepatically,11 during which, the destruction of the extracellular matrix (ECM), including the basement membrane, is an essential initial step because the ECM surrounding tumor tissues and the basement membrane normally is a barrier against cancer invasion.12 Previous studies have shown that MMP-2 (a gelatinase A/72-kD type IV collagenase) might play an important role in HCC invasion and metastasis.13, 14
We report here a significant association between HBx expression and the induction of MT1-MMP and MMP-2 in HCC patients, which correlate with poor prognosis. We also provide experimental evidence that HBx upregulates the expression of MT1-MMP, which subsequently activates MMP-2. We further show that HBx-induced MMP activation is associated with enhanced cell migration, implicating a role for HBx in HCC metastasis.
Material and methods
Patients and materials
The population consisted of 136 patients with HCC who had undergone initial and curative hepatic resection at Central South University Xiangya Hospital, in Changsha, China, between 1992 and 2002. The indication of surgical resection and operative procedure was determined on the basis of decision criteria based on the liver function test, including level of ascites fluid, serum bilirubin level and imaging data. Systematic resection of Couinaud's segment was the preferred operative procedure. By retrospectively reviewing the findings from 136 patients, HCC was grossly classified as solitary large HCC (SLHCC) or nodular HCC (NHCC) according to the clinicopathological characteristics.15
Sections, 4 mm thick, were incubated overnight at 4°C with a monoclonal antibody against MT1-MMP, MMP-2. The second antibody was applied for 45 min at 37°C. The streptavidin-biotin-peroxidase complex tertiary system (Boster, Wuhan, China) was used according to the manufacturer's instructions. The tissues were visualized by applying 3,3-diaminobenzidine tetrahydrochloride for 3 min. Sections were counterstained using hematoxylin, dehydrated through gradient alcohols, and mounted for viewing. For control stains, phosphate buffered saline was used instead of the primary antibody.
Construction of plasmids
The 480 bp HBx gene fragment was cloned from the full-length HBV plasmid pTTHK, which was kindly provided by Dr. Shuang-Ping Guo (The Fourth Military Medical University, Xi-an, China). The HBx gene(Genbank: AY310322) was amplified using forward (5′-ATCGGTACCATGGCTGCTAGGCTG-3′) and reverse (5′-GGAGAATTCAT GATTAGGCGAAGGTG-3′) primers. The pcDNA3.1-HBx plasmid was constructed by cloning the product of the 480 bp HBx gene fragment, which had been amplified by a polymerase chain reaction (PCR), into the EcoR I and Kpn I sites of the mammalian expression vector, pcDNA3.1 (Promega).
Cells and transfections
The CCL13 cells (Chang liver; American Type Culture Collection, Manassas) were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and incubated in 5% CO2 at 37°C. Twenty-four hours before transfection, the Chang cells were plated at 2 × 105 per well in 6-well plates. The Chang cells were then transfected with pcDNA3.1-HBx using Lipofectamine 2000 (Invitrogen, USA) according to the manufacturer's instructions. Forty-eight hours after transfection, the transfected Chang cells were selected in Dulbecco's modified Eagle's medium containing 700 μg/ml G-418 (Gibco) to obtain positive HBx-expressing stable cells.
Reverse transcription polymerase chain reaction
Expression of MMPs was determined by reverse transcription polymerase chain reaction (RT-PCR). Briefly, total RNA was extracted from the cells using TRIzol Reagent (Invitrogen), according to the manufacturer's instructions. Total RNA extract was treated with DNase I (Promega) at 37°C for 30 min, and 1.0 μg of the total RNA was reverse-transcribed by MMLV-RT (Promega) with an oligo-dT14 at 42°C for 60 min. MMLV-RT was inactivated at 75°C for 10 min. MTI-MMP, MMP-2, and glyceraldehyde 3′-phosphate dehydrogenase (GAPDH) were co-amplified using one fifth of the total cDNA. The following sequences were synthesized for use (Shenggong Bio, Shanghai, China): MT1-MMP forward 5′-TTGGGGTACTCGCTATCCAC-3′, MT1-MMP reverse 5′-CACTGCCTACGAGAGGAAGG-3′, MMP-2 forward 5′-CAGGCTCTTCTCCTTT CACAAC-3′, MMP-2 reverse 5′-AAGCCACGGCTTGGTTTTCCTC-3′, GAPDH forward 5′-GGTGAAGGTCGGAGTCAACG-3′ and GAPDH reverse 5′-CAAAGTTGT CATGG ATGACC-3′. The amplified products were identified by electrophoresis of a PCR on a 1.5% agarose gel containing ethidium bromide and ultraviolet illumination.
To determine the activity of MMP-2, 2 × 105 cells were seeded per well of a 24-well plate. After the cells were deprived of serum overnight, they were harvested and lysed in 35 μl 2X sodium dodecyl sulfate sample buffer after incubation for 30 min at 37°C. The samples were then mixed 1:1 with 50% glycerol and loaded onto a pre-cast gelatin gel. Following electrophoresis, the gels were incubated for 30 min at room temperature in 2.5% Triton X-100 and then overnight at 37°C in the developing buffer. The bands signifying gelatinolytic activity appeared after staining the gels with Coomassie brilliant blue.
Western blot analysis
To perform SDS-PAGE and Western blot analysis, the Chang cells transfected with pcDNA3.1-HBx were scraped in 200 μl of lysis buffer [50 mmol/l Tris-HCL (pH 8.0), 150 mmol/l NaCl, 1 mM EDTA, 1% NP-40, 0.1% SDS, 0.01% PMSF protease inhibitor cocktail]. The supernatant was collected by centrifugation at 15,000 rpm at 4°C for 15 min. The supernatant protein concentration was determined using BCA Protein assay Kit. One hundred micrograms of total protein sample treated with 2X SDS load buffer were separated by stacking gel and SDS-PAGE separating gel with Tris-glycine system at 100 V for 1 hr. Blotting was performed to nitrocellulose membrane (Millipore, USA) at 100 V for 1 hr in a tank of transfer buffer (48 mM Tris-HCl; 192 mM glycine; 20% methanol, pH 8.4). The membranes were blocked in 5% nonfat dry milk in PBS containing 0.1% Tween-20 for 2 hr at room temperature. Then the membranes were incubated with primary antibodies [MAb Anti-HBx (Alexis Biochemicals, USA), MAb Anti-MT1-MMP (R&D Systems, USA), MAb anti-MMP-2 (NeoMarker, USA)] and diluted at 1:500 for 1 hr at 37°C. After washing, the membranes were incubated with a 1:3,000 dilution of horseradish peroxidase-linked mouse anti-goat antibody (KPL, Gaithersburg) for 2 hr at room temperature. Then the membranes were washed and treated with SuperSingnal West Pico chemiluminescence (Pierce, Rockford) to visualize the bands; the results were obtained on Kodak film and quantified by densitometry (Beckman, South Pasadena, Canada). β-actine (Sigma, St Louis, MO) was used as loading control.
Delivery of antibody by electroporation
To block the expression of MT1-MMP protein, 800 μl of cell suspension containing with 10 μg of anti-MT1-MMP antibodies was transferred to a cuvette with a 4 mm gap with parallel-plate aluminum electrodes (Becton-Dickinson). Three exponential decay pulses were applied at room temperature (∼22°C) at a field strength of 500 V/cm (unless otherwise specified) with a 20 msec time constant and 20 sec inter-pulse spacing. After “recovery” at 37°C for 10 min in an ice water bath, the cell suspension was transferred to a 6-well plate, and the cuvette was rinsed into the well plate with 1 ml of growth medium. An additional 1–2 ml of growth medium was added to the well, and the cells were incubated for 2–3 days in 5% CO2 at 37°C.
To assay cell migration during wound-healing, 5 × 105cells were seeded on 6-well plates coated with 10 μg/ml type I collagen. The cells were incubated for 24 hr, the monolayers were disrupted with a cell scraper (1.2 mm wide), and photographs were taken at 0 and 24 hr under a phase- contrast microscope (Olympus). Experiments were carried out in duplicate, and 4 fields of each point were recorded.
Cell migration in transwell
Matrigel-coated filter inserts (with 8 μm pores) that fit into 24-well invasion chambers were obtained from Becton-Dickinson. pcDNA3.1-HBx Chang (Chang/HBx), pcDNA3.1 Chang (Chang/pcDNA3.1) and Chang cells were detached from the tissue culture plates, washed, resuspended in conditioned medium (2 × 105cells/500 μl), and then added to the upper compartment of the invasion chamber. Conditioned medium (500 μl) was added to the lower compartment of the invasion chamber. The Matrigel invasion chambers were incubated at 37°C for 24 hr in 5% CO2. After incubation, the filter inserts were removed from the wells, and the cells on the upper side of the filter were removed using cotton swabs. The filter inserts were fixed, mounted and stained according to the manufacturer's instructions. The cells that invaded through the Matrigel and were located on the underside of the filter were counted. Three invasion chambers were used per condition, and the procedure was repeated 3 times. The values obtained were calculated by averaging the total number of cells from 9 filters.
Statistical analysis was performed using SPSS software (version 11.0; Chicago, IL). Data are expressed as the mean ± standard error of the mean. The relationships between parameters were estimated using linear regression analysis. Nominal data were compared by the χ2 test, or Fisher exact test. Postoperative survival was analyzed by the log-rank test. p-values <0.05 were considered statistically significant.
Expression of HBx, MT1-MMP and MMP-2 is associated with a poor prognosis in HCC patients
MT1-MMP has been shown to be the upstream regulator of MMP-2.16 In light of our recent finding that the invasion and metastasis of HCC was associated with elevated expression of MMP-2,17 we asked whether the expression of MT1-MMP was also increased in HCC. For this, we used immunohistochemistry analysis to examine the expression of MT1-MMP in specimens of HCC patients. Besides confirming the increased expression of MMP-2 (in 84.5% of 136 cases of HCC (Figs. 1C–1E), the results revealed increased staining of MT1-MMP in most of the specimens examined (88.2% of 136 cases of HCC) (Figs. 1A, 1B and 1E), indicative of a co-upregulation of MT1-MMP and MMP-2 in HCC.
To gain insights into the upregulation of MMPs in HCC, we determined status of HBV infection in those patients. The expression of HBx was determined by in situ hybridization of HCC tissues performed by staff at the Department of Infectious Disease of our hospital. Examination of HBV gene expression in these 136 HCC specimens recorded 80.1% positive in HBx staining (Fig. 1E), an observation consistent with the finding of a previous study.18 Together with our data showing co-upregulation of MT1-MMP and MMP-2 in the HCC patients, our clinical data suggest that HBx expression is associated with upregulation of MMPs in HCC.
On the basis of our clinical observations of the clinicopathological characteristics of HCC over many years, we routinely classify HCC as solitary large HCC (SLHCC) or nodular HCC (NHCC).15 We have shown that NHCC exhibits a higher rate of invasion and metastasis than SLHCC.19, 20 To explore a potential connection of HCC metastasis with HBx expression and its associated MMP upregulation, we analyzed the correlation of these 2 types of HCCs with HBx expression and MMP levels. The results showed that the rates of positive expression of HBx, MT1-MMP and MMP-2 were significantly higher in NHCC than in SLHCC (p < 0.05) (Table I). Shown in Figure 1F is the higher level of expression of HBx, MT1-MMP and MMP-2 in NHCC and SLHCC, and SLHCC exhibits more benign biologic behavior than NHCC. The correlation of HBx and its associated increase in MMPs with NHCC implicates an important role of HBx in HCC metastasis.
|Positive (%)||Negative (%)||Positive (%)||Negative (%)|
|HBx||68 (50.0)||23 (16.9)||41 (30.1)||4 (3.0)||5.081||0.024|
|MT1-MMP||76 (55.9)||15 (11.0)||44 (32.4)||1 (0.7)||5.899||0.021|
|MMP-2||72 (52.9)||19 (14.0)||43 (31.6)||2 (1.5)||6.229||0.012|
Upregulation of MT1-MMP expression by HBx
In light of the finding that HBx can transcriptionally regulate cellular genes,5 we asked whether the HBx protein could upregulate the expression of MT1-MMP in HCC. To test this, we created a mammalian expression vector (pcDNA3.1) containing the cDNA of HBx to functionally characterize the HBx protein. After verifying the vector by DNA sequencing (the Gene Company Boya Bio, Shanghai, China), we introduced the plasmid encoding HBx into Chang cells with Lipofectamine-mediated transfection (Invitrogen). The cells were subjected to selection 48 hr after transfection in G418 containing medium, and positive clones were selected, amplified, and tested for the expression of HBx. Figure 2 shows the expression of HBx mRNA in two representative clones (Fig. 2a, lanes 2 and 3), with the empty pcDNA3 vector expressing cells as a control (Fig. 2a, lane 1), as well as the expression of HBx protein (Fig. 2b).
Having established stable HBx expressing Chang cell lines, we tested whether HBx expression could regulate the expression of MT1-MMP (Figs. 1A and 1B) by measuring the levels of mRNA and protein. Significantly, RT-PCR analysis revealed that the expression level of MT1-MMP mRNA was markedly elevated in HBx-expressing Chang cells when compared with the vector expressing Chang cells (Fig. 3a). Consistent with the result obtained from RT-PCR analysis, Western blot with an anti-MT1-MMP antibody indicated a significant increase in the MT1-MMP protein level in HBx expressing cells over that in control cells (Fig. 3b). These findings indicate that HBx induces the expression of MT1-MMP.
HBx induces MMP-2
The increased staining of MMP-2 in HBx positive HCC patients prompted us to test the ability of the HBx protein to induce MMP-2. Again, the levels of MMP-2 mRNA and protein were determined by RT-PCR or Western blot analysis, respectively. Interestingly, no apparent difference in the mRNA level of MMP-2 was detected when HBx expressing cells were compared with the control cells. We did exactly the same number of RT-PCR cycles in these experiments and did not detect any method with more accurate quantitation that might reveal a difference, excluding the possibility of regulation at the transcriptional level (Fig. 4a). However, the MMP-2 protein level was found elevated in HBx-expressing Chang cells over that in vector control cells (Fig. 4b), suggesting that HBx upregulates MMP-2 through a mode of post-transcriptional regulation. The increased MMP-2 levels, a protease, would predict a higher proteolysis activity. We therefore performed a gelatin zymography assay to test this possibility. The MMP-2 protein usually exists in pro-MMP-2, intermediate-MMP-2 and activated forms with the corresponding molecular weights of 72 kDa, 66 kDa and 62 kDa, respectively. The gelatin zymography assays indicated that the 62 kDa activated form of MMP-2 was significantly increased in HBx expressing Chang cells when compared with that in the control cells, whereas the expression level of 72 kDa pro-MMP-2, 66 kDa intermediate-MMP-2, or 92 kDa pro-MMP-9 did not change (Fig. 4c). A similar pattern of alteration was also observed when the conditioned medium was analyzed (Fig. 4d).
MMP-2 activation is mediated by MT1-MMP
MT1-MMP is a membrane-bound form of MMPs that has been shown to selectively activate MMP-2.21 With the finding that both MT1-MMP and MMP-2 were upregulated by the HBx protein, we asked whether the MMP-2 activation in HBx-expressing cells was mediated by MT1-MMP. For this, we introduced a monoclonal anti-MT1-MMP antibody to functionally neutralize the activity of MT1-MMP. Significantly, although the anti-MT1-MMP antibody had no effect on the MMP-2 mRNA level (Fig. 5a), the MT1-MMP antibody completely abrogated HBx-dependent upregulation of the MMP-2 protein (Fig. 5b). Moreover, the gelatin zymography assay confirmed the inhibitory effect of the MT1-MMP antibody on MMP-2 activation (Figs. 5c and 5d). Together, the results indicate that HBx-induced MMP-2 activation is mediated by MT1-MMP.
HBx promotes migration of Chang liver cells in vitro
As a result of the upregulation of MMP-2, an increased ECM degradation would foster a permissive microenvironment for tumors cells to migrate to and invade. We therefore employed 2 independent functional assays, wound-healing and transwell migration, to assess the biological consequence of HBx-induced activation of MMP-2. As shown in Figure 6, HBx-expressing Chang cells exhibited a much greater ability to repair the wound than did the control cells. The transwell migration assays also indicated an increased migration potential in HBx-expressing cells, as evidenced by 43.5 ± 3.8 migrated cells/sight in HBx-expressing cells verses 5.2 ± 1.3 migrated cells/sight in control cells. To further verify the observed effects that were indeed mediated by the action of MMP, a specific MMP inhibitor was used. Significantly, addition of the MMP inhibitor substantially suppressed the ability of HBx-expressing cells to repair wounds and migrate. Together, these results indicate that HBx expression in hepatocyte is associated with upregulation of MMPs, which then facilitate hepatocyte migradation.
Although it is well documented that a chronic HBV infection is a major risk factor for HCC,22 the mechanisms underlying HBV-mediated carcinogenesis remain poorly understood. Among commonly expressed HBV proteins, HBx is frequently detected in HBV-infected patients and has in fact been proposed as a more prevalent marker of HBV infection than the HBV surface and core antigens.23 Moreover, HBx is often found as the only HBV protein expressed in transformed hepatocytes.24 Our clinical data showed a positive HBx expression in 80.1% of specimens from 136 HCC patients, which was highly associated with the elevated levels of MT1-MMP and MMP-2. Significantly, HBx expression and high levels of MMPs were associated with a poor prognosis.
To investigate whether the HBx protein was responsible for MMP upregulation, we introduced HBx into a hepatocyte line, Chang cells. The results showed that the HBx protein indeed induces the expression of MT1-MMP at both mRNA and protein levels. We also demonstrated that HBx-mediated upregulation of MT1-MMP resulted in activation of MMP-2, a result consistent with the function of MT1-MMP as an upstream regulator of MMP-2.16
MMP activation would predict an increased ECM degradation, which could disrupt the constraint imposed by the ECM and thus promote cell spreading. Consistent with this notion are the results that HBx-expressing Chang cells exhibit a markedly increased ability of wound repair and migration. HBx does so by the activation of MMPs, as evidenced by the finding that a MMP inhibitor completely abrogated the effects of HBx.
Tumor progression is greatly dependent on a permissive microenvironment, which has a profound impact on the fate of tumor cells. The normal architecture of the ECM is very important in regulating cell growth. In order to invade, tumor cells must overcome the mechanical restrictions imposed by the ECM. Destruction of the ECM by MMPs has been shown to play an important role in the migration and spreading of tumor cells, leading to invasion and metastasis.25 The HBx-induced MMP activation implicates a role for the HBx protein in HCC metastasis. Although HBx has been reported to possess transcriptional activation activity, it remains to be determined how HBx induces the expression of MT1-MMP.
Together with the strong positive association between MMP expression and the poor prognosis observed in HCC patients, the high prevalence of HBx expression and its induced activation of MMPs strongly suggest a role for HBx in HCC invasion and metastasis.
- 17MVD and expression of integrin αv subunit and MMP-2 in solitary large hepatocellular carcinoma and their clinical significance. Chin J Hepatobiliary Surg 2004; 10: 115–17. In Chinese., , , .
- 18HBx gene facilitates the proliferation activity of hepatoma cells in vitro and in vivo. Zhonghua Wai Ke Za Zhi 2004; 42: 104–6. In Chinese., , , .