The human HCC cell lines with highly invasive capacities (HCCLM3 and MHCC-97H) were obtained from the Chinese Academy of Sciences Committee Type Culture Collection cell bank. The cells were grown in Dulbecco’s modified Eagle’s medium (DMEM; Life Technologies, Ann Arbor, MI, USA) supplemented with 10% fetal bovine serum (FBS; Life Technologies) and both penicillin and streptomycin (100 mg/ml each) at 37°C in a humidified atmosphere of 5% CO₂.

Cells (5×10⁴) were seeded in the upper chamber of Transwell plates with 8-μm pores (Costar, Cambridge, MA, USA). The lower chambers of the Transwell plates were filled with 500 μl medium containing 10% FBS as a chemoattractant. The plates were incubated at 37°C for 12 h. Cell invasion assays were performed using the same method. The Transwell chambers were covered with 50 μl 1:2 Matrigel and phosphate-buffered Saline mixture, and the cells were cultivated for 24 h. Cells that migrated or invaded to the lower surface were stained with Giemsa solution and quantified by counting five randomly selected microscopic fields at ×200 magnification.

The supernatant was collected from the HCCLM3 cells cultured in a serum-free medium with or without Salin treatment. The Matrigel angiogenesis assay was performed as previously described (12,13). In brief, BD Matrigel (BD Biosciences San Jose, CA, USA) matrix was plated in 96-well flat-bottom cell culture cluster plates. After incubation for 30 min, 10,000 HUVEC cells/well and 50 μl of the HCCLM3 supernatant with or without Salin treatment were placed on the Matrigel. The plate was incubated at 37°C for 16–18 h. Following incubation, the wells were photographed, and the results were quantified by measuring the length of the tube-like structures using Nikon NIS-Elements computer software.

To measure the concentrations of MMP2 and MMP9 secreted from the cultured tumor cell lines, the supernatants were assessed using ELISA. Cells (5×10⁴/well) in 24-well plates were incubated at 37°C in a 5% CO₂ atmosphere in DMEM containing 10% FBS. After 24 h, the cells were washed and incubated for 24 h in serum-free medium without/with Salin (5–20 μmol/l). The culture-conditioned medium was collected, centrifuged, and the concentrations of MMP2 and MMP9 were determined by quantitative ELISA (R&D Systems, Minneapolis, MN, USA), according to the manufacturer’s instructions.

To clone the putative promoter and/or enhancer region of the MMP9 gene, a PCR-based method was used, and specific primers were designed from the 5′-end of the known MMP9 promoter sequence from a previous study (14). The amplified DNA fragment of 2302 bp was cloned into the pGL3-Basic vector (Promega, Madison, WI, USA) to construct pGL3-MMP9-WT containing the potential enhancer element. AP-1 site-mutated MMP9 (pGL3-MMP9-mAP1-1, pGL3-MMP9-mAP1-2 and pGL3-MMP9-mAP1-1+2), NF-κB site-mutated MMP9 and Sp-1 site-mutated MMP9 (pGL3-MMP9-mSp1) luciferase promoters were used in the transient transfection assays as previously described (15,16). These vectors contain the luciferase gene driven by the SV40 promoter. The composition of the constructs was confirmed by restriction endonuclease digestion and DNA sequencing.

Cells (5×10⁴/well) in 24-well plates were incubated at 37°C in a 5% CO₂ atmosphere in serum-free DMEM. After 24 h, the plasmids were transfected into cells using Lipofectamine-2000 reagent according to the manufacturer’s instructions (Invitrogen, Carlsbad, CA, USA). After transfection, the cells were cultured in FBS without or with Salin (5–20 μmol/l) for 24 h. The cell lysate was used to detect luciferase activity according to the manufacturer’s protocol (Promega). Luciferase activity was normalized to the β-galactosidase activity in the cell lysates and expressed as the average of three independent experiments.

Total cellular proteins were extracted using RIPA buffer and quantified by the BCA method. The sample proteins were separated by electrophoresis on SDS-PAGE and transferred onto a PVDF membrane. After blocking, the membranes were incubated overnight at 4°C with various primary antibodies and α-tubulin at 1:500–1:1000 dilution (Santa Cruz Biotechnology, Santa Cruz, CA, USA). The secondary HRP-conjugated antibodies were diluted 1:2000. The immunocomplexes were detected using the ECL system (Beyotime Biotechnology, China). A Li-COR Odyssey scanner (LICOR) was used to detect the antigens on the blots.

All statistical analyses were performed using SPSS 17.0 software. Data are reported as mean ± SD, and mean values were compared using the Student’s t-test and ANOVA. Values of p<0.05 and p<0.01 were considered statistically significant.

It is known that Salin has anticancer activity, and previous studies have shown that it can induce cell apoptosis and suppress cancer cell proliferation. We determined the effects of Salin on the migration and invasion of HCC cells using the Transwell assay and tube formation assay. Salin (5–20 μmol/l) significantly inhibited HCCLM3 cell migration and invasion in a concentration-dependent manner (Fig. 1A). In addition, significantly fewer integrated capillary-like structures were found in the HCCLM3 cells treated with Salin (5–20 μmol/l), indicating that Salin affected the release of pro-angiogenic proteins in the HCCLM3 cells (Fig. 1B). Similar results were confirmed in the other invasive HCC cell line, MHCC-97H (data not shown), suggesting that the anti-invasive or anti-migratory effect of Salin in the HCC cell lines is universal.

To elucidate the anti-invasive and anti-migratory mechanisms of Salin, we examined MMP9 and MMP2 expression. HCCLM3 cells were treated with Salin at concentrations of 5–20 μmol/l. The results showed that Salin at 5 μmol/l or higher concentrations attenuated MMP9 expression in the HCCLM3 cells, but did not significantly affect the expression of MMP2 (Fig. 2A). In addition, Salin inhibited MMP9 expression at the protein level in a concentration-dependent and time-dependent manner (Fig. 2B). To determine the importance of MMP9 in the anti-invasive effect of Salin in HCCLM3 cells, we established stable overexpression of MMP9 in the HCCLM3 transfectants by infecting the cells with a lentivirus encoding MMP9. The results showed that overexpression of MMP9 was significantly blocked by 10–39% in the Salin-treated group compared with the group without Salin treatment (p<0.01; Fig. 2C). The results also showed that Salin decreased MMP9 expression by 14–36% compared with the random group (Ver transfectants) (p<0.05; Fig. 2C). These results suggest that MMP9 may play a role in the anti-invasive and anti-migratory effects of Salin.

The MMP9 promoter contains two AP-1 sites (located at −79 bp and −533 bp), a Sp1 site (located at −560 bp) and an NF-κB site (located at −600 bp). To determine whether regulation of MMP9 is related to these cis-acting regulatory elements for transcription factors, several constructs with deletions or mutations were used and have been described in Materials and methods. Experimental cells were transfected with reporter vectors that included the tandem repeat of AP-1, NF-κB or Sp-1 binding sites. Noteworthy, luciferase activity in the cells with the AP-1 construct was significantly reduced by treatment with Salin at 10–20 μmol/l, whereas luciferase activity containing the NF-κB or Sp-1 binding site construct showed no statistically significant changes in the cells treated with Salin (Fig. 3A). These results showed that both AP-1 sites in the MMP9 promoter were essential for Salin-regulated MMP9 enhancer region activity.

The AP-1 complex is composed of several protein components, including c-Fos, JunD, JunB and Fra-1. We aimed to determine which members play a major role in the downregulation of MMP9 by Salin. The results demonstrated that only JunD was significantly downregulated by Salin in the HCCLM3 cells (Fig. 3B). JunB showed a slightly increasing trend, particularly following treatment with 20 μmol/l Salin, whereas Fra-1 and c-Fos were not significantly affected by Salin at concentrations ranging from 5 to 20 μmol/l (Fig. 3B).

To evaluate the role of JunD in Salin-regulated MMP9 activity, we investigated JunD and MMP9 expression at different times and at different Salin concentrations. The data showed that Salin simultaneously reduced JunD and MMP9 protein expression (Fig. 4A). It was also shown that Salin affected protein expression at 10 μmol/l or higher concentrations, and this was concentration-dependent (Fig. 4A). In addition, as shown in Fig. 4B, the target proteins were inhibited by Salin in a time-dependent manner. The dynamic effects of JunD and MMP9 expression occurred after 12 h.

To further confirm the effects of Salin on these signaling cascades, anisomycin, an activator of c-Jun N-terminal kinase (JNK), was added to the HCCLM3 cells with or without Salin. The results showed that Salin blocked anisomycin-induced JunD expression (Fig. 4A). Antibodies against the phosphorylated forms of JNK were then used to determine the changes in the phosphorylation level following Salin intervention. The data showed that the phosphorylated forms of JNK1 and JNK2, p-JNK1 and p-JNK2, were significantly impeded by Salin in a time-dependent manner (Fig. 4C). These results showed that JNK phosphorylation and JunD may be involved in the Salin-regulated MMP9 signaling pathway in HCCLM3 cells and may mediate HCC cell biological characteristics.