SB203580 (p38 MAPK inhibitor) was obtained from Sigma (St. Louis, MO, USA). Mouse anti-p38 and phospho-p38 antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Antibodies specific for syntenin (sc-100336) were purchased from Santa Cruz Biotechnology. Mouse anti-Ki67 monoclonal antibodies were from Sigma. The corresponding horseradish peroxidase-conjugated secondary antibodies were obtained from Calbiochem (La Jolla, CA, USA).

The human liver cancer cell lines HepG2, MHCC97-L, MHCC97-H and HCCLM3 were obtained from the American Type Culture Collection (ATCC; Rockville, MD, USA). Cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum, 100 μg/ml streptomycin and penicillin. Immortalized normal liver epithelial cells THLE3 were also purchased from ATCC and were maintained in bronchial epithelial growth medium, supplemented with 5 ng/ml epithelial growth factor, 70 ng/ml phosphoethanolamine and 10% fetal bovine serum. All cells were cultured in a humidified atmosphere at 37°C with 5% CO₂.

Total RNA was isolated from human hepatoma cells HepG2 with the RNAiso Plus kit (Roche Diagnostics, Mannheim, Germany), and was then reverse-transcribed to synthesize first strand cDNA using the cDNA Synthesis kit (Fermentas). Then, ~4 μl cDNA was subjected to real-time PCR with specific primers for syntenin (sense, 5′-GAA TCC TGC AAA AAT GTC TC-3′ and antisense, 5′-GCC ATG GTG CCG TGA ATT TTA-3′); EGFR (sense, 5′-TCG GGG AGC AGC GAT GCG AC-3′ and antisense, 5′-TCA TGT GAT AAT TCA GCT C-3′); and MMP-2 (sense, 5′-CAG GGA GCG CTA CGA TGG A-3′ and antisense, 5′-GAG CCA GGG CCA GCT CAG C-3′). Each reaction consisted of 10 μl SYBR^® Premix Ex Taq™ II, 10 μmol/l specific primers, 4 μl of DNA and H₂O to a final volume of 20 μl. The reaction conditions were introduced according to the instructions of SYBR Premix Ex Taq™ II kit (Takara Bio Inc., Otsu, Japan). For normalization, β-actin mRNA was used. All samples were performed in triplicate, and results were calculated according to the 2^−ΔΔCt method.

To obtain the syntenin expression vectors, the syntenin cDNA was digested with BamHI and XhoI restriction enzymes, and was then ligated into the BamHI and XhoI cloning site of the pcDNA3.1(+) vector (Invitrogen) to induce syntenin expression. All sequences were confirmed by DNA sequencing. Transfection was performed using 8 μl of Lipofectamine 2000 (Invitrogen) and 15 μg of pcDNA3.1-syntenin or pcDNA3.1 empty vector in HCCLM3 cells. Forty-eight hours after transfection, 400 μg/ml of G418 was used to select the stable transfectants. The empty vector was packaged as a negative control. A limiting dilution was introduced to produce several stable syntenin-expressing clones.

To knock down EGFR expression in HCCLM3 cells, the specific siRNA fragments of EGFR and control siRNA were designed as previously described (15). The siRNA strands were synthesized by Shanghai Sangon Co., Ltd. (Shanghai, China). For transfection, HCCLM3 cells were seeded in 24-well microplates to reach 40–50% confluence. Then, cells were transfected with 2 μg/ml EGFR siRNA and 1 ml Lipofectamine™ RNAi-MAX using the GeneSilencer^® siRNA Transfection Reagent (GeneTherapy System, San Diego, CA, USA). Approximately 24 h later, the transfection efficiency was analyzed by western blotting as described above. Cell viability was evaluated by 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay.

Cell proliferation was assessed by MTT assay. Briefly, cells were seeded into 96-well culture plates at a density of 1×10⁵ cells/well. After incubation for 24 h, the culture medium was replaced with fresh medium containing 500 μg/ml MTT (Sigma) for an additional 5 h at 37°C and 5% CO₂. Then, the supernatant was replaced with 200 μl isopropanol to dissolve formazan production and the absorbance was measured at 570 nm on a microplate reader (Bio-Rad, Hercules, CA, USA).

HCCLM3 cells transfected with syntenin were plated in 0.35% agar mixed culture medium in 12-well plates. After 12 days, cells were washed with phosphate-buffered saline (PBS) three times, and were then fixed with methanol. Cell colony formation in soft agar was analyzed by counting the number of colonies under a stereoscopic microscope in triplicate of a 12-well plate. The results are expressed as percentage control of colonies.

The ability of cells to invade through Matrigel-coated filters was measured by a modified Boyden chamber (BD Biosciences, Bedford, MA, USA). Briefly, syntenin-over-expressed HCCLM3 cells were pre-treated with anti-MMP-2 antibody for 4 h, and were then seeded at a density of 3.0×10⁴ cells in the upper compartment. DMEM medium containing 10% fetal bovine serum was added to the lower compartment. After incubation for 48 h at 37°C, cells were stained with hematoxylin and eosin (Sigma), and counted in six randomly selected high-powered fields in the center of each well. Each experiment was repeated three times.

Following rinses with PBS three times, cells were homogenized and lysed with RIPA lysis buffer (100 mM NaCl, 50 mM Tris-HCl pH 7.5, 1% Triton X-100, 1 mM EDTA, 10 mM b-glycerophosphate, 2 mM sodium vanadate and protease inhibitor) to extract the total protein, and then a BCA assay (Pierce, Rockford, IL, USA) was introduced to quantify the protein concentrations. Following electrophoresis by SDS-PAGE, the obtained protein was transferred onto a polyvinylidene difluoride (PVDF) membrane in a semi-dry transblot apparatus. To perform the western blot assay, non-specific binding was blocked by incubating with 5% non-fat milk in TBST buffer at room temperature at 4°C overnight. Then, immune-detection of syntenin, EGFR, Ki67, MMP-2, p-p38 MAPK and p38 MAPK was performed using specific antibodies against them for 1 h at 37°C, followed by incubation with HRP-conjugated secondary antibodies for 1 h. To visualize the bound antibodies, the LumiGLO reagent (KPL, Gaithersburg, MD, USA) was used. The protein expression levels were normalized by β-actin.

The Student’s t-test was used to evaluate statistically significant differences between two groups in all relevant experiments. All numerical results were analyzed by SPSS 11.0 and are presented as means ± SD. A P-value <0.05 was considered to indicate a statistically significant result.

Syntenin has been confirmed to be upregulated in several cancers and is critical for the development of cancer (9,13). However, research regarding syntenin in HCC remains to be performed. To address this, we analyzed the expression of syntenin in various HCC cell lines. Compared with normal liver epithelial cells THLE3, an apparent increase in syntenin mRNA levels was observed in all four HCC cell lines including HepG2, MHCC97-L, MHCC97-H and HCCLM3 by RT-PCR analysis (Fig. 1A). Furthermore, western blot analysis demonstrated that the expression levels of syntenin were also markedly upregulated in these four HCC cell lines, compared with THLE3 cells (Fig. 1B). Moreover, the mRNA and protein levels of syntenin in the high metastatic potential cell line MHCC97-H were higher than in the low-invasive cell line MHCC97-L, indicating a potential correlation between syntenin expression and the metastatic ability of HCC cell lines. Collectively, these results corroborated that syntenin is upregulated in HCC cells, particularly in higher metastatic cells.

To clarify the role of syntenin in the development and progression of HCC, we assessed the effect of syntenin overexpression on HCCLM3 cell growth. To quantify the direct contribution of syntenin in hepatoma, the recombinant pcDNA3.1-syntenin was transfected into HCCLM3 cells, and then a stable syntenin overexpression clone termed HCCLM3-150 cell line was obtained. As shown in Fig. 2A, a striking upregulation of syntenin protein was detected in the HCCLM3-150 cell line, compared with the vector control clone. Based on the stable expression of syntenin in HCCLM3-150 cells, we evaluated syntenin function on cell growth by MTT and colony formation assays. As shown in Fig. 2B, an apparent increase in cell proliferation was observed as the gradually increased transfection times in syntenin-overexpressed group, compared with the control group. Furthermore, syntenin overexpression displayed more colonies in contrast to controls (Fig. 2C). Therefore, these results showed that syntenin transfection strongly promotes cell proliferation.

It is generally believed that EGFR is overexpressed in several cancers, including colorectal cancer, HCC and urothelial carcinoma, and can activate a cascade of multiple signaling pathways that facilitate the tumor growth process (16–18). The fact that syntenin can promote hepatoma cell proliferation was demonstrated above. However, the precise molecular mechanism of this action remains unclear. Hence, we attempted to link EGFR to syntenin-induced cell proliferation mechanistically. When cells were stably transfected with syntenin, the mRNA levels of EGFR were 2.7-fold over controls, indicating that elevated syntenin expression strongly upregulated EGFR mRNA levels (Fig. 3A). Moreover, a similar upregulation of EGFR protein levels was also ascertained by western blot assay (Fig. 3B). Further mechanistic analysis corroborated that EGFR silencing significantly reduced cell viability induced by syntenin overexpression (Fig. 3C). Consistently, the expression levels of proliferation marker Ki67 were markedly attenuated when blocking EGFR expression in HCCLM3-150 cells (Fig. 3D). Taken together, our data demonstrated that syntenin overexpression promotes hepatoma cell proliferation predominantly through the EGFR pathway.

To further assess the function of syntenin in hepatoma cells, we detected cell invasion ability in HCCLM-150, which stably overexpressed syntenin. As shown in Fig. 4, the forced syntenin expression notably upregulated the number of invaded cells from 56 to 135 compared to the control groups, indicating it is a positive regulator for HCC cell invasion. It is known that MMP-2 can degrade extracellular matrix to facilitate tumor cell invasion (19). Thus, we linked the MMP-2 to syntenin-induced cell invasion. Following transfection of syntenin in HCCLM3 cells, the expression levels of MMP-2 mRNA were clearly increased and were 3.75-fold times higher than those of the controls (Fig. 5A). Consistent with the above observation, syntenin overexpression significantly enhanced the expression of MMP-2 protein (Fig. 5B). To further analyze the underlying mechanism involved in syntenin-induced cell invasion, we silenced the expression levels of MMP-2 with its specific antibody and a notable downregulation of MMP-2 levels was observed in HCCLM3-150 cells (Fig. 5C). Also, MMP-2 silencing clearly abrogated syntenin-induced cell invasion number (Fig. 5D). Hence, these data confirmed that syntenin enhances cell invasion ability in an MMP-2-dependent manner.

To further clarify the mechanism associated with syntenin-triggered hepatoma cell invasion through MMP-2, we evaluated the activation of p38 MAPK. Only slight phosphor-p38 MAPK was observed in the control group; however, when cells overexpressed syntenin, the expression of p-p38 MAPK was significantly increased, suggesting that syntenin overexpression induced the activation of the p38 MAPK pathway (Fig. 6A). Notably, the specific inhibitor of p38 MAPK, SB203580, strikingly blocked syntenin-induced MMP-2 mRNA levels (Fig. 6B), concomitant with the reduction of MMP-2 protein levels (Fig. 6C). These results revealed that syntenin promotes cell invasion via MMP-2 expression through the activation of the p38 MAPK pathway.