A total of 86 patients were enrolled in the present study. The patients did not receive any preoperative cancer treatment and their follow-up data were available. They were followed-up after surgical treatment until May 2011, with a median follow-up of 29 months (range 2–73. 2 months). During the follow-up, the patients were monitored every 2–3 months as described previously (7). CT scanning or MRI was performed when tumor recurrence was suspected. The recurrent tumors were treated as described previously (8). Clinical samples were collected from these patients after obtaining informed consent according to an established protocol approved by the Ethics Committee of Fudan University (Shanghai, China).

Immunohistochemical staining was performed to detect the expression of ARHGDIA in HCC and matched non-cancer tissue. The primary antibody against ARHGDIA was obtained from Epitomics (1:50). Intensity of staining was scored as 0 (negative), 1 (weak), 2 (moderate) or 3 (strong). The extent of staining was based on the percentage of positive tumor cells: 0 (negative), 1 (1–25%), 2 (26–50%), 3 (51–75%) and 4 (76–100%). The final score of each sample was assessed by summarization of the results of the intensity and extent of staining. Therefore, each case was considered negative if the final score was 0–1 (−) or 2–3 (±) and positive if the final score was 4–5 (+) or 6–7 (++), respectively. These scores were determined independently by two senior pathologists.

Huh-7, SMMC-7721 and MHCC-97H cells were cultured in DMEM medium with 10% FBS, maintained at 37°C in a humidified air atmosphere containing 5% carbon dioxide.

The coding sequence of human ARHGDIA was cloned into the expression vector pCDH-CMV-MCS-EF1-Puro (System Biosciences, Mountain View, CA, USA). The siRNA against ARHGDIA were synthesized by Ribobio and inserted into the pLKO.1-TRC cloning vector (Invitrogen, Carlsbad, CA, USA). All constructs were verified by sequencing. A mixture of pCDH-ARHGDIA or pLKO.1-siARHGDIA cloning vector, and adjuvant vectors psPAX2 and pMDG2 were transfected into HEK293T cells using Lipofectamine 2000 reagent to generate lentiviruses. Huh-7, SMMC-7721 and MHCC-97H cells were infected with the recombinant lentivirus-transducing units plus 8 mg/ml polybrene (Sigma).

Cell proliferation was measured with the Cell Counting Kit-8 (CCK-8) assay kit (Dojindo Corp.); 5,000 cells were plated into each well of a 96-well plate, in which 10 μl CCK-8 was added to 90 μl of culture medium. The cells were subsequently incubated for 1 h at 37°C and the attenuance was measured at 450 nm. Three independent experiments were performed.

The 500 cells were plated into 6-well culture-plates and cultured for 14 days to allow colony formation. Colonies were stained with 0.1% crystal violet (Amersco, Solon, OH, USA) in 50% methanol and 10% glacial acetic acid for counting.

For the migration assays, 2×10⁴ cells were added into the upper chamber of the insert with the non-coated membrane (Millipore, 8-mm pore size). For the invasion assays, each well insert was layered with 50 μl of a 1:4 mixture of Matrigel/Dulbecco’s minimal essential medium (BD Bioscience). Cells (1×10⁵) were added into the upper chamber of the insert. In both assays, cells were plated in medium without serum, and medium containing 10% FBS in the lower chamber served as chemoattractant. After several hours of incubation, the cells that did not migrate or invade through the pores were carefully wiped out with cotton swab. Cells on the lower surface of the membrane were fixed with methanol and stained with Giemsa and counted. Each experiment was performed in triplicates.

For in vivo metastasis assays, SMMC-7721 cells infected with either the ARHGDIA-siRNAs or the vector were transplanted into nude mice (5-week-old BALB/c-nu/nu, 6 per group, 2×10⁶ cells for each mouse) through the tail vein. After 6 weeks, mice were sacrificed. The lungs were removed, fixed in formalin, and embedded in paraffin. Consecutive sections of the whole lung were subjected to hematoxylin and eosin staining. All of the metastatic foci in lung were calculated microscopically to evaluate the development of pulmonary metastasis. The lung metastases were calculated and evaluated independently by two pathologists.

Equal amounts of protein were resolved by 10% SDS-polyacrylamide gel electrophoresis and transblotted onto nitrocellulose membrane (Bio-Rad). After blocking in 5% non-fat milk, the membranes were incubated with rabbit anti-ARHGDIA antibody (mAb; 1:1,000; Epitomics), rabbit anti-Rac1 antibody (mAb; 1:1,000; Epitomics), rabbit anti-RhoA antibody (mAb; 1:1,000; Epitomics), rabbit anti-Cdc42 antibody (mAb; 1:1,000; Epitomics) or rabbit anti-GAPDH mAb (1:5,000; Epitomics). The proteins were detected with enhanced chemiluminescence reagents (Pierce).

The protocol used was based on the availability of a mouse monoclonal antibody directed against the active form of Cdc42, RhoA and Rac1 commercially through NewEast Biosciences (Malvern, PA, USA). Cells were lysed in 1 ml of ice-cold lysis buffer for 10 min. Aliquots of each cell lysate were added to two microcentrifuge tubes, one for analysis of the active and the other for the analysis of total potein content. Then 1 μl of anti-active Cdc42, RhoA or Rac1 monoclonal antibody was added, as well as 20 μl of Dynabead Protein G added, and samples were incubated overnight with rotation at 4°C. Beads were pelleted by centrifugation for 1 min at 5,000 g, then washed three times with 0.5 ml of lysis buffer, resuspended in 20 μl of 2× reducing SDS-PAGE sample buffer, heated at 100°C for 5 min, then separated on 12% polyacrylamide gels and processed for western blotting after transferring to PVDF membranes. Rabbit polyclonal antibody against total Cdc42, RhoA and Rac1 (mAb; 1:1,000; Epitomics) was used for western blotting.

Statistical analysis was performed with SPSS 15.0 (SPSS Inc, Chicago, IL, USA) and values are expressed as the mean ± standard deviation. The differences between groups were analyzed using Student’s t-test (only two groups), or one-way analysis of variance (more than two groups were compared). P<0.05 was considered statistically significant.

The protein levels of ARHGDIA in 86 cases of HCC patient samples and corresponding non-cancer liver tissues (40 cases) were measured by immunohistochemical staining. Strong staining of ARHGDIA was observed in adjacent non-cancer liver tissue, but weaker in more than half (65%) of the HCC tissues (Fig. 1A). The expression level is significantly downregulated in HCC compared with non-cancer liver tissues (P<0.001) (Fig. 1B). At various regions of tumor within the same slide, it appeared that ARHGDIA expression remarkably decreased at invasive cancer in situ such as tumor embolus. Next, we analysed the relationships between ARHGDIA and clinical pathological features of HCC. Significant correlations were observed between ARHGDIA and vascular invasion (tumor invasion in blood vessel or bile duct) (P=0.0216). Low level of ARHGDIA expression was observated in 79.07% of vascular invasion group (Table I). The result indicated that ARHGDIA might correlated with HCC metastasis. Then, the ARHGDIA level was analyzed in a panel of human HCC cell lines with different metastatic potential. The level of ARHGDIA in the high-metastatic HCC cell lines (MHCC-97H) was much lower than that in the less-metastatic HCC cell lines (Huh-7, SMMC-7721) (Fig. 3C), indicating that the downregulation of ARHGDIA was related to the metastatic ability of HCC.

In the Kaplan-Meier analyses, the expression level of ARHGDIA was significantly associated with OS and TTR. The patients with low level of ARHGDIA exhibited a decreased postoperative OS and a shorter TTR compared those with high level (Fig. 2). The 1-, 3- and 5-year OS rates of the patients with low level were 74.8, 42.2 and 36.6%, respectively, which were significantly lower than those with high level group (86.9, 65.4 and 51.5%, respectively; P=0.017). The 1-, 3- and 5-year cumulative recurrence rates of low level group were 37.8, 63.9 and 75.6%, respectively, which were significantly higher than those of the high level group (20.5, 47.2 and 60.4%, respectively; P=0.012).

To explore the functions of ARHGDIA in HCC, specific siRNAs against ARHGDIA were exploited to knockdown expression in SMMC-7721, and Huh-7 cell lines. As shown in Fig. 3D, siRNA significantly reduced the expression of ARHGDIA protein. We also constructed a lentivirus vector expressing ARHGDIA and established the stable cell line MHCC-97H, which has low basal levels of ARHGDIA (Fig. 3E). In cell proliferation assays, knocking down ARHGDIA showed no obvious impact on the proliferation of SMMC-7721 and Huh-7 cells (Fig. 3A). Similarly, overexpression of ARHGDIA did not affect MHCC-97H cell growth either (Fig. 3B). Next, the colony formation assays were performed to observe the effects of ARHGDIA on the anchoring growth ability of HCC cells. No obvious effects were observed on the colony formation ability of HCC cells after infection with ARHGDIA-siRNAs or lenti-ARHGDIA (Fig. 4).

Given that expression of ARHGDIA is highly associated with the metastatic property of HCC, we wondered whether ARHGDIA could play an important role in HCC cell invasion and metastasis. Transwell assays without Matrigel demonstrated that downregulation of ARHGDIA could significantly promote migration of Huh-7 and SMMC-7721 cells when compared with vector groups (Fig. 5A). Transwell assays with Matrigel showed that the invasive capacities were dramatically enhanced in these two stable cell lines when compared with the control cells (Fig. 5B). However, the migration and invasion of MHCC-97H cells decreased when ARHGDIA was upregulated (Fig. 5). These results indicated that loss of ARHGDIA could significantly enhance HCC cell migration and invasion in vitro. To further explore the role of ARHGDIA in tumor metastasis in vivo, SMMC-7721 cells infected with si-ARHGDIA were transplanted into nude mice through the tail vein. Interestingly, the number of the metastatic nodules in the lung were dramatically increased in si-ARHGDIA groups compared with vector control (P=0.0377) (Fig. 6). Taken together, these observations suggested that ARHGDIA is a negative metastatic regulator for HCC.

Regulation of the cytosol-membrane cycling of the Rho GTPase by ARHGDIs has a major role in controlling Rho GTPase activity and function. Given the important role of ARHGDIA in HCC cell migration and invasion, we conducted immunoprecipitation assays to determine the status of Rho GTPases in HCC cells. The mouse monoclonal antibody directed against the active form of Cdc42, RhoA and Rac1 were used in the immunoprecipitation assays. The results indicated that loss of ARHGDIA significantly induced RhoA and Rac1 activation in SMMC-7721 cells (Fig. 7A), particularly the activity of RHOA increased nearly 4-fold compared to the control. The Cdc42 activity was also slightly increased, but did not reach statistical significance (Fig. 7B). Numerous studies have confirmed that activation of signaling of Rho GTPases plays an important role in cancer progression and metastasis (9). Therefore, the activation of Rho GTPase proteins induced by silencing ARHGDIA might contribute to tumor invasion and metastasis of HCC.