Integrin α 6 as an invasiveness marker for hepatitis B viral X-driven hepatocellular carcinoma

Abstract

BACKGROUND:

Hepatitis B virus (HBV) accounts for more than 60% of hepatocellular carcinoma (HCC) cases. However, there is limited information about the features of HBV-driven HCC that differentiate it from other types of HCC.

OBJECTIVE:

The aim of this study is to find a gene specific to HBV-driven HCC and understand its role during tumorigenesis.

METHODS:

The differences in gene expression patterns were analyzed among patients with hepatitis virus-unrelated liver cirrhosis, and hepatitis C virus- and HBV-driven HCC. Genes expressed only in HBV patients were compared to genes of transgenic mice expressing hepatitis B viral X gene.

RESULTS:

Integrin $\alpha$ 6 was commonly overexpressed in both HBV-driven HCC patients and transgenic mice expressing viral X. This gene’s activation induced overexpression of integrin $\alpha$ 6, as well as formation of integrins $\alpha$ 6 $\beta$ 1 and $\alpha$ 6 $\beta$ 4, without changing the expression of non-integrin laminin receptors. Suppression of integrin $\alpha$ 6 caused significant inhibition of tumor migration in vitro.

CONCLUSIONS:

This study found a significant association between HBV and integrin $\alpha$ 6, which may be responsible for early migration and invasion of HCC. Thus, integrin $\alpha$ 6 is a predictive marker for tumor recurrence and invasiveness of HBV-driven HCC.

1. Introduction

Hepatocellular carcinoma (HCC) is the sixth most common cancer and the third leading cause of cancer death in the world [1]. More than 70% of newly diagnosed HCC cases are in Asia due to poor hygiene, many carriers of chronic hepatitis B and C virus infection, and vertical transmission [2]. Although the use of alcohol, metabolic diseases, and non-alcoholic fatty liver disease are increasingly found to be causes of HCC, chronic hepatitis is still responsible for the majority of HCC cases [3]. Hepatitis B virus (HBV)-driven HCC accounts for more than 60% of all HCC cases [4].

Due to biological differences between HBV, a DNA virus, and hepatitis C virus (HCV), an RNA virus, different genomic factors affect their impact on HCC formation and progression [4]. Also, HCCs induced by non-viral causes have different genomic expression than hepatitis virus-driven HCC. However, diverse biomarkers for hepatocellular carcinoma, such as membrane-type 1 matrix metalloproteinase, hepatocyte growth factor, c-met, proline-rich tyrosine kinase 2, and transforming growth factor- $\beta$ 1 have been utilized to predict the prognosis of HCC patients regardless of the cause [5, 6, 7, 8]. For this reason, it is necessary to develop more specific prognostic biomarkers based on the mechanism of HBV-driven HCC, the cause of the largest proportion of HCC cases.

Hepatitis B viral X protein (HBx) is a key viral regulatory protein and is associated with carcinogenesis among chronic hepatitis B carriers [9, 10, 11]. HBx interacts with p53 to suppress its tumor suppressive role, and augments anti-apoptotic signaling, such as the Ras pathway, and pro-survival transcription factors [12, 13, 14]. Thus, animals who express HBx more susceptible to hepatocellular carcinoma [15, 16]. As the HBx gene is prone to mutations which cause functional change, its cellular roles are likely to be more variable than what has been verified so far [17, 18]. Since HBx plays a critical role in HBV-driven HCC, we focused on finding a novel target related to HBx. Additionally, we validated the potential of the novel target for predicting invasiveness and prognosis of HBV-driven HCC specifically. We also validated the effectiveness of the novel biomarker in laboratory animals and HCC cells.

2. Materials and methods

2.1 Cell culture

Two HCC cell lines, HepG2 and Hep3B, were purchased from Korea Cell Bank (Seoul, Korea). The cells were cultured in RPMI 1640 (Hyclone, Logan, UT, USA) with 10% fetal bovine serum (Hyclone, Logan, UT, USA), 200 $\mu$ g/mL penicillin, and 100 $\mu$ g/mL streptomycin at 37 ${}^{\circ }$ C and 5% CO ${}_2$ .

2.2 DNA, RNA, and siRNA information

HBx plasmids were provided by Dr. Eun Young Cho (Wonkwang University, Korea). A 269-bp fragment of the X gene was amplified using the primer set 5’-GGAGTTGGGGGAGGAGAT-3’ (forward) and 5’-TCCCCTTCTTCATCTCCC-3’ (reverse). siRNA specific for ITGA6 was made using the primers: 5’-AGCCTCTTCGGCTTCTCG-3’ (forward), 5’-TTG GCTTCTGCAGTGGAA-3’ (reverse). Real-time quantitative PCR was performed using TaqMan 7500 (Applied Biosystems, Foster City, CA, USA). mRNA levels were measured using the SYBR Green I assay with SYBR Green Universal PCR Master Mix, according to the manufacturer’s instructions (Applied Biosystems). Human GAPDH was used as the housekeeping gene and each reaction was run in triplicate. The expression levels of transcripts were calculated by the relative quantification ( $\mathrm{\Delta }$ Ct) study method by using 7500 System SDS software (Applied Biosystems).

2.3 Gene expression analysis of HCC

Differentially expressed gene (DEG) profiles were retrieved from the Gene Expression Omnibus (GEO) database at the National Center for Biotechnology Information (NCBI). The gene profiles of non-HBV cirrhosis, HCV-driven HCC, and HBV-driven HCC were taken from entries GSE25097, GSE44074, and GSE47197 in the GEO database, respectively. Microarray data on the HBx mouse group was provided by Dr. Dae Ho Lee (Gachon University, Korea) [19]. Genes whose expression increased more than two-fold were selected. The gene network was analyzed in Cytoscape (version 3.0.0), which is connected to the databases mantha, I2D, and IntAct. For enrichment of the transcriptome, we used a transcription factor list from Uhlén et al. [20] and for membrane protein enrichment, database FANTOM5.

2.4 Western blot

Cells were lysed at 4 ${}^{\circ }$ C in RIPA buffer (20 mM Tris-HCl pH 7.5, 150 mM NaCl, 1 mM Na ${}_2$ EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM b-glycerophosphate, 1 mM Na ${}_3$ VO ${}_4$ , 1 $\mu$ g/mL leupeptin, 1 mM PMSF) (Cell Signaling Technology, Danvers, MA, USA). The samples were separated by SDS-PAGE. After transfer of the proteins onto polyvinylidene difluoride (PVDF) membranes, the membranes were probed with anti-integrin $\alpha$ 6 (Cat # sc-374057, Santa Cruz Biotechnology, Dallas, TX, USA), anti-integrin $\beta$ 4 (Cat # sc-6628, Santa Cruz Biotechnology) and $\beta$ 1 (Cat # sc-374429, Santa Cruz Biotechnology), anti-67 kDa laminin receptor (Cat # ab99484, Abcam, Cambridge, MA, USA) and anti- $\beta$ -actin antibodies (Santa Cruz Biotechnology).

2.5 Immunoprecipitation

Cells were disrupted in a lysis buffer of 0.5% Triton X-100, 0.1% SDS, and a protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO, USA). Then, the lysates were centrifuged at 15,000 $\times$ g for 30 min. To examine the interaction of integrins $\beta$ 1 and $\beta$ 4 with integrin $\alpha$ 6, the protein extracts were incubated with normal IgG and protein G agarose for 2 h, and then centrifuged to remove non-specific IgG-binding proteins. The supernatant was mixed with the anti-integrin $\alpha$ 6 antibody (Santa Cruz Biotechnology) and incubated for 2 h at 4 ${}^{\circ }$ C with agitation before protein G agarose was added. After three washes with ice cold lysis buffer, the precipitates were dissolved in SDS sample buffer and separated by SDS-PAGE. After the proteins were transferred to PVDF membranes, co-immunoprecipitation related to integrin $\alpha$ 6 was analyzed.

Figure 1.

Schematic of meta-data analysis. (A) Public human cohort data were collected and meta-analyzed. (B) The data were compared with animal data. (C) The enriched gene sets predict the role of genes in the disease. (D) Subsequent in vitro and in vivo study validate in silico data.

Figure 2.

Differentially expressed gene (DEG) patterns. (A) Venn diagrams showing the DEG patterns among hepatitis virus-unrelated liver cirrhosis, hepatitis C virus-driven hepatocellular carcinoma, and hepatitis B virus-driven hepatocellular carcinoma patients. Three genes were commonly upregulated and 7 genes were commonly downregulated in three patient groups. DEGs were identified using a cutoff of a two-fold change. (B) Venn diagrams showing the DEG patterns between hepatitis B virus-driven hepatocellular patients and transgenic mice expressing the hepatitis B viral X gene. Thirteen upregulated genes and 28 downregulated genes were shared between HBV-driven HCC patients and HBX gene expressing transgenic mice. DEGs were identified using a cutoff of a two-fold change.

2.6 HBx transgenic mice and immunohistochemistry

Formalin-fixed paraffin-embedded liver blocks from HBx homozygous transgenic (HBxTg) mice that have one copy of HBx transgene on chromosome 2 were given by Drs Cho (Wonkwang University, Korea) and Lee (Gachon University, Korea). The details about the development and the characteristics of HBxTg mice has been described previously [21]. The strain of HBxTg mice was fixed to C57BL/6 by backcrossing with the C57BL/6 strain for more than 20 generations. HBx ${}^{+/+}$ transgenic mice were verified by polymerase chain reaction (PCR) analysis. The PCR primers were as follows: one set was 5’-TTCTCATCTGCCGGTCC GTG-3’ (forward) and 5’-GGGTCAATGTCCATGCC CCA-3’ (reverse), and another set was 5’-GAAAACA CACTCACTGTTCAGAG-3’ (forward) and 5’-GTAA GCCGCTTTCTCTTATGCAG-3’ (reverse).

The embedded tissues (5 mm $\times$ 5 mm) were sectioned with a microtome (Leica, Deerfield, IL, USA). The tissues were deparaffinized, rehydrated, and equilibrated with phosphate buffered saline (PBS). Endogenous peroxidase was inactivated with 1% hydrogen peroxide for 5 min and the slides were cooked with antigen retrieval buffer in a microwave oven 3 times for 5 min. The slides were blocked with 4% bovine serum albumin (BSA) for 30 min and washed with PBS containing 0.1% Tween 20 (PBST). Primary antibodies against integrin $\alpha$ 6 were added to the slides and incubated for 90 min. After three washes with PBST, slides were incubated with a horseradish peroxidase (HRP)-conjugated anti-rabbit or anti-mouse antibody for 30 min. The slides were washed twice with PBST and developed with 3,3’-diaminobenzidine (DAB) (Agilent Technologies, Inc., Dako, Santa Clara, CA, USA) for 7 min. DAB development was terminated by soaking the slides in distilled water for 10 min. Nuclei were counterstained with Mayer’s hematoxylin (Sigma-Aldrich) for 7 min. After washing with distilled water for 10 min, slides were mounted and observed under a microscope (Nikon Eclipse 80, Nikon, Toyko, Japan). Analysis of the Real-Time PCR for wild type and HBxTg mice was performed using Student’s t-test.

2.7 Cell migration assay

Cell migration was analyzed using 24-well transwell chambers with polycarbonate membranes (8.0- $\mu$ m pore) (Corning, Costar, Tewksbury, MA, USA). Hep3B and HepG2 cells were suspended in serum-free RPMI 1640 and added into the upper chamber at 10 ${}^5$ cells per well. Laminin (10 $\mu$ g/mL) was added to the lower well and the cells were allowed to migrate for 8 h at 37 ${}^{\circ }$ C in a CO ${}_2$ incubator. The cells were fixed with 70% methyl alcohol in PBS for 30 min and washed with PBS three times. Next, the cells were stained with hematoxylin (Sigma-Aldrich) for 10 min and washed with distilled water. The non-migrant cells were removed from the upper surface of the membrane with a cotton swab. The membranes were excised from the chamber and mounted with Gel Mount (Biomeda, Foster City, CA, USA). The migrant cells (those attached to the lower face of the membrane) were counted in four randomly selected high-power fields of view (20 $\times$ ). The difference in the degree of cell migration between two groups was compared using Student’s t-test.

Figure 3.

Enrichment of differentially expressed genes (DEGs) with transcription factors or membrane proteins. Venn diagram of DEGs enriched with (A) transcription factors or (B) membrane proteins. There was no overlap of transcription factors between HBx transgenic mice and HBV-driven HCC (red dotted circle). Three membrane protein genes, which were closely related to cancer invasion and initiation of cellular signaling, were shared between HBx transgenic mice and HBV-driven HCC (red dotted circle).

3. Results

3.1 Collection of HCC-related human and animal data

Many sets of gene expression data for particular diseases have accumulated in the last 20 years. As surgical resection and biopsy are performed for most types of cancer, tissue samples are collected and their related gene expression data sets provide statistically reliable information. The analysis of expression patterns is effective for finding a gene marker for a disease (Fig. 1A). It can be compared with animal data of same disease (Fig. 1B). This comparison may indicate a mode of action of the target gene. Additionally, the gene can be categorized based on ontology, so the gene-associated signaling pathway, cellular location, and the other characteristics can provide clues about the gene function (Fig. 1C). Subsequent in vitro or in vivo validation could logically explain the role of a target gene (Fig. 1D). The series of this study supports cost and time-effective marker study.

3.2 DEG profiles of non-HBV liver cirrhosis, HCV-driven HCC, and HBV-driven HCC patients, and HBx mice

We searched for gene expression patterns of non-HBV cirrhosis, HCV-driven HCC, and HBV-driven HCC groups in the GEO database (Fig. 2A). As tumor growth-related factors of HBV-driven HCC are expected to show differences from HCV-related factors, we focused our analysis on the HBV-driven HCC group. Though the HBV-driven HCC group shared 4 and 41 overexpressed genes with the HCV-driven HCC group and non-HBV cirrhosis group, respectively, 88 genes were overexpressed only in the HBV-driven HCC group. Because HBx augments tumor malignancy, the relationship between HBx and HBV-driven HCC was revealed by comparing the gene profiles of the HBV-driven HCC group in the GEO database and HBx transgenic mice (Fig. 2B). DEGs were identified using a cutoff two-fold change. Integrin $\alpha$ 6 (ITGA6) was overexpressed only in HBV-driven HCC, not in non-HBV cirrhosis-related or HCV-driven HCC groups. Furthermore, ITGA6 was commonly overexpressed in both the HBx transgenic mouse group and HBV-driven HCC group. Commonly upregulated and downregulated genes among HBx transgenic mice, and patients with non-HBV liver cirrhosis, HBV-driven HCC, and HCV-driven HCC are listed in Table S1.

3.3 Ontology-based marker study

The HBV induced network in HCC was significantly different from the network in other types of HCC (Fig. S1). There were many more nodes between genes compared with HCV-driven HCC. Transcription factors and membrane proteins or receptors are generally considered upstream driver genes rather than passenger genes in a disease. The DEG lists were enriched with transcription factors and membrane proteins separately. In the transcription factor enrichment analysis, a large portion showed an independent pattern with little or no overlap among the two HCC, one HBx-transgenic liver, and one cirrhosis group. There is no overlap in transcription factors between HBx transgenic and HBV related cancer groups. On the other hand, genes of membrane proteins overlapped among the four groups (Fig. 3). Membrane proteins are important for interaction with the extracellular matrix during cancer progression and initiate outside-in cellular signaling.

Figure 4.

Association between hepatitis B viral X and integrin $\alpha$ 6, and the binding affinity of integrin $\beta$ 1 and integrin $\beta$ 4 with integrin $\alpha$ 6.(A) mRNA and (B) protein expression changes of integrin $\alpha$ 6 and 67 kDa laminin receptor as a function of time after HBx transfection into Hep3B cells. LR: laminin receptor. (C) mRNA and (D) protein expression changes of integrin $\alpha$ 6 and 67 kDa laminin receptor as a function of time after HBx transfection into HepG2 cells. LR: laminin receptor; HBx: hepatitis B viral X. (E) Integrin $\beta$ 1 and integrin $\beta$ 4 showed increased binding affinity to the anti-ITGA6 antibody without a change in the amount of integrin in the whole-cell lysate. ITGB1: integrin $\beta$ 1; ITGB4: integrin $\beta$ 4; ITGA6: integrin $\alpha$ 6; EV: empty vehicle; WCL: whole-cell lysate; HBx: hepatitis B viral X.

Figure 5.

Increased expression of ITGA6 in HBxTg mouse liver tissue. (A) The mRNA level of ITGA6 in WT and HBxTg mice was validated by RT-PCR. (B) Real-Time PCR was performed to compare wild type and HBxTg mice. * $p<$ 0.05.

3.4 Effect of HBx on ITGA6 expression

HBx was transfected into Hep3B cells, human HBV infected hepatocytes, and HepG2 cells, well-differentiated human HCC cells not infected with HBV. Post transfection, mRNA and protein levels of ITGA6 were monitored by PCR and western blot, respectively. ITGA6 was slightly downregulated 1 week after transfection. Then, it was gradually upregulated until week 4 without affecting the 67 kDa laminin receptor, a nonintegrin laminin receptor in Hep3B (Fig. 4A and B) and HepG2 cells (Fig. 4C and D). As the expression of HBx increased, the expression of ITGA6 increased. Additionally, integrin $\beta$ 1 (ITGB1) and integrin $\beta$ 4 (ITGB4), partners of ITGA6, showed increased binding affinity to the ITGA6 antibody without a change in the amount of integrin in the whole-cell lysate (Fig. 4E).

The increased ITGA6 expression was confirmed in liver tissue isolated from HBx transgenic mice. The RT-PCR experiment showed a slightly increased pattern of ITGA6 in HBxTg compared with the wild type control (Fig. 5A). Real-Time PCR showed a significant difference between two groups (Fig. 5B).

Figure 6.

The influence of laminin and hepatitis B viral X on hepatocytes. Laminin and HBx induced cellular migration in (A, B) Hep3B cells and (C, D) HepG2 cells. FOV: field of view; EV: empty vehicle; LN: laminin; HBx: hepatitis B viral X; NS: non-specific * $p<$ 0.05, ** $p<$ 0.01.

Figure 7.

Influence of hepatitis B viral X and effects of siRNA specific for ITGA6 on inhibition of proliferation. siRNA specific for ITGA6 reduced the migration ability of HBx-transfected cells to pre-HBx transfection levels in (A) Hep3B and (B) HepG2 cells. The inhibitory effect of siRNA specific for ITGA6 in (B) HEP3B cells was not significant. The inhibitory effect of siRNA specific for ITGA6 in (D) HEPG2 cells was significant. ITGA6: integrin $\alpha$ 6; FOV: field of view; HBx: hepatitis B viral X gene; CTRL: control * $p<$ 0.05, *** $p<$ 0.001.

3.5 Effects of HBx and laminin on the migration of HCC cells

Administration of laminin induced migration of Hep3B and HepG2 cells (Fig. 6). The migration was significantly different between the laminin treated and control cells. In both cell lines, transfection with HBx resulted in more pronounced migration. The rate of migration of HepG2 cells was more significant than that of Hep3B cells. In addition, laminin administration and HBx transfection had strong synergistic effects on cellular migration.

3.6 Effects of ITGA6 suppression on the migration of HCC cells

As we found, HBx increased cellular migration in both cell lines (Fig. 7A–D, middle panels). Nevertheless, siRNA specific for ITGA6 reduced the migration ability of HBx-transfected cells to pre-HBx transfection levels (Fig. 7A–D, right panels; Fig. S2). The inhibitory effects of siRNA specific for ITGA6 were more prominent in HepG2 cells. In HepG2 cells, the suppression of migration activity was 50%, while the inhibition of proliferation was only about 25% (Fig. S3). Thus, ITGA6 is more strongly associated with migration in this condition.

4. Discussion

Based on the gene expression pattern analysis, we found that higher expression of ITGA6 is a feature specific to HBV-driven HCC and not HCC induced by other causes. Although ITGA6 is found in many types of cancer, we showed that HBx activation in HBV-infected hepatocytes and non-HBV HCC cells induced overexpression of ITGA6 and formation of integrin $\alpha$ 6 $\beta$ 1 and $\alpha$ 6 $\beta$ 4 without alteration in the expression of 67 kDa laminin receptors. Both laminin and HBx play crucial roles in HCC migration. siRNA specific for ITGA6 caused significant inhibition of tumor migration in vitro.

Integrins are heterodimeric structures that are made of $\alpha$ and $\beta$ subunits [22]. When ITGB1 is in a complex with various members of the ITGA family, ITGB1 can act as a receptor for collagen, fibronectin, and laminin [23]. ITGB4, on the other hand, can only bind to ITGA6 and only act as a laminin receptor [23]. Integrin $\alpha$ 6 $\beta$ 4, expressed mainly in epithelial cells, is associated with invasiveness, metastasis, and aggressiveness of cancers, including breast, bladder, lung, thyroid, and pancreatic [24, 25, 26, 27, 28]. Integrin $\alpha$ 6 $\beta$ 4, stimulated by laminin, affects cancer cell survival, apoptosis, tumor invasion, proliferation, and metastasis via multiple signaling pathways, including p53, Ras, RhoA, and PI3K [27, 29]. Despite the significance of integrin $\alpha$ 6 $\beta$ 4 in tumor progression, only a few studies on integrin $\alpha$ 6 $\beta$ 4 and HCC have been published [29, 30]. These studied have shown the involvement of integrin $\alpha$ 6 $\beta$ 4 in HCC without consideration of the differences between the types of hepatitis virus. We revealed a strong correlation among HBx, laminin, integrin $\alpha$ 6 $\beta$ 4, and HBV-driven HCC. The key signal between HBx and ITGA6 is p53. HBx downregulates p53 which decreases expression of ITGA6. An siRNA against p53 increases ITGA6 expression. By suppressing p53, HBx can enhance the aggressiveness of HCC by overexpression of ITGA6. Lara-Pezzi et al. also demonstrated that HBx affects integrin-mediated adhesion and migration [31]. Though Torimura et al. detected scant integrin $\beta$ 4 expression in hepatoma cell lines, we detected overexpression of integrin $\beta$ 4 in HCC [32]. As that experiment was performed about 20 years ago, the insensitivity of the antibody used to detect integrin $\beta$ 4 may be one reason for the discrepancy. Integrin $\alpha$ 6 $\beta$ 1 also participates in HCC development. Integrin $\alpha$ 6 $\beta$ 1 helps laminin to easily attach to hepatoma cells and affects the invasiveness of HCC [32, 33, 34]. The migration suppression effects of siRNA specific for ITGA6 on HCC can be mediated by inhibition of the dimerization of integrin $\alpha$ 6 $\beta$ 1 and integrin $\alpha$ 6 $\beta$ 4, which plays a significant role in HCC development.

In our experiments, treatment of Hep3B and HepG2 cells with laminin and HBx enhanced migration. The increase in migration after treatment with HBx was more prominent in HepG2 cells than in Hep3B cells. As Hep3B cells are an HBV-carrying cell line, it is likely that the effect of HBx is weaker in this cell line. Furthermore, HepG2, a non-HBV HCC cell line, showed markedly increased cellular migration after the addition of HBx. Based on these results, it is possible that HBx closely correlates with the aggressiveness of HCC, as shown previously [35, 36, 37]. In addition, knockdown of ITGA6 in HBx-expressing cells inhibits cellular migration and proliferation to the level of HBx-untransfected cells. This means that ITGA6 may be closely related to the activity of HBx. However, the 2 to 3 week delay between administration of HBx and induction of ITGA6 suggests that increase in ITGA6 may be caused by reorganization of the intracellular network rather than a direct effect of HBx. As discussed, laminin has important functions in the growth and development of HCC [29, 32]. However, preventing and reducing laminin attachment to HCC cells is impossible in this context. The siRNA specific for ITGA6 prevented laminin from attaching to integrins $\alpha$ 6 $\beta$ 1 and $\alpha$ 6 $\beta$ 4 and significantly inhibited migration of HCC cells. Although binding between ITGA6 and ITGB1 or ITGB4 was increased after HBx transfection, the total amounts of ITGB1 and ITGB4 did not change (Fig. 4E). Since the amount of integrin did not increase until 2 weeks after HBx transfection (Fig. 4A and B), we could not detect a change in the amount of integrin in whole-cell lysate experiments performed 24 hr after HBx transfection. The increase of binding between ITGA6 and ITGB started soon after HBx transfection despite the lack of a change in the amounts ITGA6 and ITGB. Furthermore, the 67 kDa laminin receptor, a nonintegrin laminin receptor, did not affect the growth of HCC cells (Fig. 4B). The factors that play a crucial role in the growth of HCC tumors are not laminin itself, but the integrin-based laminin receptors, including $\alpha$ 6 $\beta$ 1 and $\alpha$ 6 $\beta$ 4.

5. Conclusions

We demonstrated for the first time a relationship between ITGA6, laminin, and HBx. In addition, we revealed that knockdown of ITGA6 inhibits migration of HCC cells by inhibiting integrins $\alpha$ 6 $\beta$ 1 and $\alpha$ 6 $\beta$ 4. Further studies on ITGA6 could reveal it as a predictive biomarker for invasiveness or tumor recurrence in HBV-driven HCC. Additionally, given that suppression of ITGA6 reduced migration of HCC cells, ITGA6 could be studied as a therapeutic target for the treatment of HBV-drive HCC.