The human liver cancer cell line HepG2 was purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA) and cultured in Dulbecco's modified Eagle's medium containing 10% FBS (HyClone Laboratories; GE Healthcare Life Sciences, Logan, UT, USA) and 1% penicillin/streptomycin. Experiments were conducted with cells at <25 passages. GA (C15:1; C₂₂H₃₄O₃) was obtained from Sigma-Aldrich; Merck KGaA (Darmstadt, Germany) and initially dissolved in pure methanol as a stock solution of 1 mM. Working dilutions of GA were performed with culture medium immediately before use. Recombinant human HGF was purchased from R&D Systems (Minneapolis, MN, USA). Primary antibodies against E-cadherin (cat. no. 3195, rabbit), N-cadherin (cat. no. 13116, rabbit), ZO-1 (cat. no. 13663, rabbit), vimentin (cat. no. 5741, rabbit), HGF (cat. no. 52445, rabbit), c-Met (cat. no. 8198, rabbit) and phosphorylated c-Met (p-c-Met) (cat. no. 3077, rabbit) were purchased from Cell Signaling Technology (Danvers, MA, USA); mouse anti-MMP-2 (cat. no. ab86607), mouse anti-MMP-9 (cat. no. ab58803) and mouse anti-β-actin (cat. no. ab8226) were obtained from Abcam (Cambridge, MA, USA); and secondary antibodies (goat anti-rabbit IgG-HRP; cat. no. sc-2004; goat anti-mouse IgG-HRP; cat. no. 2005) were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA).

The effect of GA on HepG2 cell proliferation was assessed by an MTT assay. HepG2 cells were seeded in 96-well plates (3,000 cells/well) and incubated for 12 h. After treatment with GA at different concentrations (0, 5, 10, 25, 50 and 100 µM) at the indicated time-points (12, 24, 36 and 48 h), 20 µl of 5 mg/ml 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich, Merck KGaA) was added to each well and cells were incubated for an additional 4 h at 37°C (19). Then 150 µl DMSO (Sigma-Aldrich; Merck KGaA) per well was added to dissolve the crystals. Cell proliferation was assessed by measuring the absorbance at 490 nm using a 96-well plate spectrophotometer (Thermo Fisher Scientific, Inc., Waltham, MA, USA).

Wound healing migration assays were conducted in order to assess the migration ability of HepG2 cells. HepG2 cells were seeded in 6-well plates and grown to 80% confluence in complete medium. After serum-starvation for 12 h, the cells were then treated with GA (0, 25 and 50 µM) for 24 h. Subsequently, a sterile 200-ml pipette tip was used to make a wound across the cell culture monolayer. Discarding the medium and washing three times with PBS removed floating cells. Multiple images of the matched-pair wound regions were obtained immediately after wounding (0 h) and after 36 h using a light microscope (Nikon Corp., Tokyo, Japan) at a magnification of ×100. The migration distance was determined by calculating the area of the cell gap at the indicated time-points (0 and 36 h).

Cell invasion assays were performed using Matrigel-coated Transwell inserts (8-µm pore size; BD Biosciences, Franklin Lakes, NJ, USA) according to the manufacturer's recommendation. Cells (5×10⁴) were added to the upper chamber and incubated in the presence of GA (0, 25 and 50 µM) or GA plus HGF (5 ng/ml) in serum-free medium. Complete medium (500 µl) was added to the lower chamber. After 48 h in culture, cells in the upper chamber were carefully removed with a cotton-tipped swab. The invaded cells were fixed and stained with 0.1% crystal violet solution. Quantification of invasion was calculated by counting stained invaded cells in at least 10 randomly selected fields using a light microscope (Nikon Corp.) at a magnification of ×200. Each experiment was performed in triplicate to confirm the reproducibility of the data.

Tissues or cells were lysed in RIPA Lysis Buffer (Beyotime Institute of Biotechnology, Guangzhou, China). Following protein quantification with BCA (Pierce; Thermo Fisher Scientific, Inc.), samples were subjected to 10–12.5% SDS-PAGE and transferred to PVDF membranes. After blocking with 5% BSA, the membranes were incubated with the primary antibodies (dilutions for antibodies: mmp-2, 1:800; mmp-9, 1:750; E-cadherin, 1:1,500; ZO-1, 1:1,000; vimentin, 1:750; N-cadherin, 1:1,000; β-actin, 1:1,000; c-Met, 1:800; p-c-Met, 1:1,000) and then with species-specific secondary antibodies (1:5,000). β-actin was used as an internal loading control. Protein bands were detected on a ChemiDoc XRS imaging system (Bio-Rad Laboratories, Inc., Hercules, CA, USA) using enhanced chemiluminescence (EMD Millipore, Billerica, MA, USA).

After the designated treatment, total cell RNA was isolated using TRIzol reagent (Invitrogen, Thermo Fisher Scientific, Inc.) and cDNA was synthesized using a PrimeScript RT reagent kit (Takara Biotechnology Co., Ltd., Dalian, China) according to the manufacturer's instructions. Quantitative PCR was then performed with an iQ5 Multicolor Real-Time PCR Detection System (Bio-Rad Laboratories, Inc.) using a SYBR Green reagent (Takara Biotechnology Co., Ltd.). The cycling conditions were as follows: Denaturing at 95°C for 10 min followed by 35 cycles of denaturing at 95°C for 10 min, 38 cycles of denaturing at 95°C for 20 sec, annealing and extension at 60°C for 1 min. The primer sequences used in this study are listed in Table I. Relative expression of the sample genes was calculated using the ΔΔCq method (20) with β-actin as the endogenous control.

After the designated treatment, cells were fixed with 4% ice-cold methanol and permeabilized with 0.5% Triton X-100. Cells were then blocked with 1% bovine serum albumin (BSA) followed by incubation with a primary antibody against HGF (dilution 1:200) at 4°C overnight. After washing with PBS, cells were incubated with a goat anti-rabbit FITC (green) IgG antibody (cat. no. ZF-0311; ZSGB-BIO Inc., Beijing, China) at 1:200 dilutions for 1 h at room temperature and then washed with PBS again. Subsequently, the cells were stained with DAPI in order to visualize the nuclei. Images were captured with a fluorescence microscope (Nikon Eclipse Ti-s; Nikon Corp.) using the appropriate excitation wavelength at a magnification of ×200.

Animal experimental protocols were approved by the Ethics Committee of The Second Affiliated Hospital of Medical College, Xi'an Jiaotong University (Xi'an, Shaanxi, China). Twenty 6-week-old male BALB/c nude mice were supplied by and housed in the Animal Center at the Medical College, Xi'an Jiaotong University. The mice were housed in a ventilated, temperature-controlled (22–24°C) and standardized sterile animal room with a 12-h light/dark cycle, and free access to food and water. A single-cell suspension (30 µl) of HepG2 cells (suspended in HBSS, containing 1×10⁶ cells) was inoculated subcutaneously into the back of the BALb/c nude mice. One week after inoculation, mice were randomly divided into two cohorts with 10 mice in each group. One cohort received the vehicle (100 µl saline) by oral gavage and the other was orally administrated GA (suspended in saline, 50 mg/kg) daily for 5 weeks. The tumor volume was monitored throughout the experiment and calculated using the following formula: V (tumor volume)=S (shorter diameter)² × L (longer diameter) ×0.5. At the end of the experiment, the mice were euthanized by CO₂ asphyxiation (the CO₂ flow rate used for euthanasia was 30% of the chamber's volume/min) and the tumor samples were harvested. Tissue protein was prepared and subjected to western blot assays as previously described.

Data are presented as the mean ± standard deviation. Each experiment was performed at least three times. All quantitative data were analyzed using SPSS (version 15.0; SPSS, Inc., Chicago, IL, USA). Student's t-test was performed to assess the difference between two groups. Two-way analysis of variance was used to analyze data between groups and post-hoc Tukey's test was used for multiple comparisons, and P<0.05 was considered to indicate a statistically significant difference.

Firstly, the effect of GA on the migration ability of HepG2 cells was investigated. The concentration of GA used was determined according to a previous study (15) and by cell proliferation experiments. The proliferation of HepG2 cells was suppressed by GA treatment at a concentration of more than 25 µM (Fig. 1). HepG2 cells were pre-treated with GA (25 and 50 µM) for 24 h and then wound healing migration assays were performed. As shown in Fig. 2A, the migration capacity of HepG2 cells was impaired by 25 and 50 µM GA intervention compared to that of the control cells (0 µM GA), as determined by the cell migration distance. To further assess the effect of GA on the invasion capacity of HepG2 cells, a Matrigel invasion assay was conducted. The results revealed that there was a significant reduction in the number of invaded cells after the HepG2 cells were treated with 25 or 50 µM GA (Fig. 2B). Altogether, the data demonstrated that GA inhibited the migration and invasion of HepG2 cells in vitro.

Epithelial-mesenchymal transition (EMT), which endows tumor cells with enhanced motility and invasion capacities, is a prerequisite for tumor infiltration and metastasis (21). In order to investigate the effect of GA on the expression of invasion-related genes (MMP-2 and MMP-9) and EMT-related genes (E-cadherin, ZO-1, vimentin and N-cadherin), HepG2 cells were treated with GA (25 and 50 µM) for 24 h and then the total cell RNA was extracted and subjected to RT-qPCR analysis. As revealed in Fig. 3A, the mRNA expression levels of MMP-2 and MMP-9 were significantly reduced in the HepG2 cells after treatment with GA. In addition, increased expression levels of epithelial markers (E-cadherin and ZO-1) and decreased expression levels of mesenchymal markers (vimentin and N-cadherin) were detected in the HepG2 cells with GA intervention. These observations were confirmed at the protein level by the results of the western blot assays (Fig. 3B). Collectively, these results indicated that GA inhibited the expression of invasion-related molecules and prevented the EMT process in HepG2 cells.

A previous study has demonstrated that HGF/c-Met signaling plays critical roles in the promotion of cell invasion and the EMT process (8). To determine whether GA has an influence on HGF/c-Met signaling activation in liver cancer cells, HepG2 cells were treated with GA (25 and 50 µM) for 48 h, and then the phosphorylation level of c-Met (p-c-Met) was determined by immunoblotting. Exogenous recombinant HGF (5 ng/ml) was used as a positive control. As observed in Fig. 4A, the total c-Met (t-c-Met) remained unchanged in HepG2 cells after GA treatment. However, treatment with GA resulted in a dose-dependent decrease of p-c-Met expression in HepG2 cells. The binding of HGF to its corresponding RTK c-Met is necessary for c-Met phosphorylation and triggering of downstream events. To further confirm that GA-prevented c-Met phosphorylation is mediated by a reduction in HGF production, the expression of HGF after GA intervention was detected. As revealed in Fig. 4B, the RT-qPCR results demonstrated that the mRNA expression of HGF was suppressed by GA treatment. This was further confirmed by immunofluorescence against HGF (Fig. 4C). These findings indicated that GA suppressed the activity of HGF/c-Met in HepG2 cells via reduction of HGF production.

To further confirm that the GA-mediated migratory/invasive response in HepG2 cells was due to HGF suppression, recombinant HGF was used to treat HepG2 cells. As revealed in Fig. 5A, consistent with the previous data, 50 µM GA treatment resulted in reduced expression of p-c-Met, HGF, MMP-2, MMP-9, vimentin and N-cadherin and increased expression of E-cadherin and ZO-1 in HepG2 cells. However, exogenous HGF (5 ng/ml) supplementation reversed these effects that were mediated by GA. The inhibited expression of p-c-Met, MMP-2, MMP-9, vimentin and N-cadherin by GA was improved in the presence of HGF. Furthermore, the enhanced expression of E-cadherin and ZO-1 by GA was markedly suppressed by HGF supplementation. In addition, the Matrigel invasion assay results revealed that exogenous HGF supplementation restored the inhibitory effect of GA on the invasiveness of HepG2 cells (Fig. 5B). These results indicated that reduced HGF expression was responsible for the GA-suppressed invasion response in HepG2 cells.

Animal experiments were conducted in order to assess the effect of GA on the growth of HepG2 cells in vivo. A single-cell suspension (30 µl) of HepG2 cells (containing 1×10⁶ cells) was inoculated subcutaneously into the back of BALb/c nude mice. One week after inoculation, mice were randomly divided into two cohorts, one of which received the vehicle and the other was orally administrated with GA. Through monitoring the tumor volume, tumor growth was found to be significantly suppressed by GA intervention as compared to the mice in the vehicle group (Fig. 6A). The volumes of the tumors from the GA-treated mice were reduced by >50% compared with the control mice at the end of the experiment (Fig. 6B and C). Western blotting was performed in order to detect the expression of E-cadherin, N-cadherin and p-c-Met in the tumor tissue. As shown in Fig. 6D, treatment with GA decreased the expression of HGF and the phosphorylation level of c-Met. Levels of E-cadherin were markedly increased, while levels of N-cadherin were markedly decreased in tumor tissue from the GA-treated mice. We did not observe any metastases both in the vehicle and GA-treated group at the end of the experiment. These in vivo findings demonstrated that GA effectively suppressed tumor growth, and inhibition of the activation of c-Met in tumor cells and prevention of EMT may be the underlying mechanism of GA.