5-aza C was purchased from Sigma-Aldrich (St. Louis, MO, USA). PD 98059 (PD) was purchased from Calbiochem (San Diego, CA, USA) and LY 294002 (LY) was purchased from Tocris Bioscience (Bristol, UK).

Human fibrosarcoma HT1080 cells (KCLB no. 10121; Korean Cell Line Bank, Seoul, Korea) were grown in RPMI-1640 medium (Gibco/Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS; Gibco-Invitrogen), penicillin (50 U/ml; Sigma-Aldrich) and streptomycin (50 μg/ml; Sigma-Aldrich), in a humid atmosphere, 5% CO₂ and 95% air at 37°C.

Proliferation of HT1080 cells was determined using colorimetric 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Cells were seeded into 96-well plates (0.5×10⁵ cells/well in 100 μl medium) and cultured for 24 h. After an overnight incubation, the cells were treated with 5-aza C concentration gradient (0, 5, 10, or 20 μM) for 24 h, or with 10 μM 5-aza C for the indicated time periods, and then washed with phosphate-buffered saline (PBS). Thereafter, the medium was replaced by fresh medium (200 μl) containing 0.5 mg/ml MTT, and the mixture was incubated for 4 h at 37°C. Following the incubation, 100 μl solubilization buffer (10% SDS, 0.01 N HCl) was added to each well to terminate the MTT reaction and dissolve formazan crystals. After agitation for 1 h at room temperature, optical density of the solution in each well was measured at 595 nm using a microplate reader (Molecular Devices, Sunnyvale, CA, USA). Proliferation inhibition rate was calculated as: [1-A₅₉₅ (experimental well)/A₅₉₅ (control well)] ×s 100%.

HT1080 cell migration was determined by wound-healing assay (15). Briefly, HT1080 cells (5×10⁵/well) were seeded into 35-mm culture dish and after reaching ~90% confluence, a scratch was made in the cells with a 10-μl pipette tip. The remaining cells were washed with medium and incubated with 5-aza C in the absence or presence of inhibitors for 24 h. The migration distance of cells was then captured and the images were quantitatively analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA). Distances between scratch edges were measured and statistically analyzed (see below).

HT1080 invasion was determined with Transwell Matrigel system (16). Cells (1×10⁵ cells/well) were cultured in the top chambers of 24-well Transwell plates (8.0 mm-pore; Corning Costar, Corning, NY, USA) and complete RPMI-1640 medium was added to the bottom chambers. After 24 h, the cells on the surface of the top chamber membrane were removed with a cotton swab. Cells that had migrated to the bottom surface of the top chamber membranes were fixed with 95% methanol, stained with 1 mg/ml crystal violet and hematoxylin solution, and counted (20 random fields) under a microscope (magnification, x200). Each experiment was performed in triplicate and repeated at least twice.

Cells (5×10⁵ cells) were grown in 35-mm culture dish to ~80% confluence and starved in serum-free medium at 37°C for 12 h. The cells were then treated with 5-aza C for 24 h and harvested in trypsin-EDTA (0.25%, pH 7.2) buffer. Cells were washed with PBS prior to suspension in cold propidium iodide (PI) solution (50 μg/ml) in PBS (pH 7.4) containing RNase A (0.1 mg/ml) for 30 min in the dark. After incubation, cell apoptosis and cell cycle distribution were analyzed with a flow cytometer (Partec GmbH, Munster, Germany).

Preparation of cell lysates and western blot analyses were performed as previously described (17). Briefly, equal amounts of proteins from whole cell lysates (20–30 μg) were resolved by 8–12% SDS-polyacrylamide gel electrophoresis and transferred to a nitrocellulose membrane. The membrane was then incubated in blocking buffer comprising 5% skimmed milk in Tris-buffered saline with Tween-20 (TBST), at room temperature for 1 h. The blocked membrane was next incubated with primary antibodies, at 4°C, overnight. The following primary antibodies were used: rabbit anti-DNMT-1 polyclonal antibody (1:1,000 dilution; Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA; sc-20701); mouse anti-phospho-Akt polyclonal antibody (1:1,000 dilution; Cell Signaling Technology Inc., Danvers, MA, USA; #9271); rabbit anti-Akt polyclonal antibody (1:1,000 dilution; Santa Cruz Biotechnology; #9272); rabbit anti-phospho-p44/42 MAP kinase (pERK1/2) polyclonal antibody (1:1,000 dilution; Cell Signaling Technology; #9101); rabbit anti-ERK-2 poly-clonal antibody (1:1,000 dilution; Santa Cruz Biotechnology; sc-154). The membrane was washed three times (10 min each) with TBST, and then incubated with secondary antibodies, at room temperature, for 2 h. The following secondary antibodies were used: anti-rabbit IgG antibody (1:3,000 dilution; Sigma-Aldrich; A0545); anti-mouse IgG antibody (1:3,000 dilution Enzo Life Sciences, Inc., Farmingdale, NY, USA; ADI-SAB-100). The antigen-antibody complex was detected with an enhanced chemiluminescence detection kit (Dogen, Seoul, Korea). Protein band intensities were quantified by the ImageJ and normalized to internal β-actin control. Normalized values were plotted and are presented.

MMP-9 gelatinase activity in conditioned media was performed using SDS-PAGE zymography, as previously described (18). Briefly, HT1080 cells were split into 35-mm culture dish. After 24 h, cells were serum-starved in RPMI-1640 medium (Gibco/Invitrogen) for additional 20 h, and then treated with increasing concentrations of 5-aza C (0, 5, 10 and 20 μM) for 24 h. Subsequently, 30 μl conditioned medium was mixed with reducing agent-devoid of 4X SDS sample buffer (62.5 mM Tris, 4% SDS, 25% glycerol, 0.01% bromophenol blue, pH 6.8) and subjected to electrophoresis in 7.5% SDS-PAGE gels containing 1 mg/ml gelatin (Sigma-Aldrich). After electrophoresis, gels were washed twice with 2.5% (v/v) Triton X-100 for 60 min at room temperature to remove SDS and subsequently incubated overnight at 37°C in a developing buffer containing 10 mM CaCl₂, 150 mM NaCl, and 50 mM Tris-HCl, pH 8.0, for 24 h. The gels were stained with 0.5% Coomassie brilliant blue R-250 in 10% acetic acid (v/v) and 50% methanol, for ~1 h, and then destained in 50% methanol and 10% acetic acid solution at room temperature to clearly visualize the digested bands. MMP-9 proteolytic activities were visualized as clear bands against blue background of stained gelatin.

Data are presented as means ± SEM from three independent experiments and compared using one-way ANOVA. For P<0.05, differences were considered statistically significant and are indicated by an asterisk in the figures.

To evaluate the effect of 5-aza C on cellular proliferation, cells were cultured with increasing 5-aza C concentrations (5, 10 and 20 μM), as described in Materials and methods, and their viability was assessed by MTT assay. 5-Aza C inhibited proliferation of cells in a dose-dependent manner (Fig. 1A). We next performed FACS analysis to investigate the effect of 5-aza C on the cell cycle (Fig. 1B). FACS analysis revealed that 5-aza C induced G₂/M phase arrest (Fig. 1B). Only 32.2% untreated control HT1080 cells were in G₂/M phase. However, 24 h of treatment with 5, 10 and 20 μM 5-aza C, the proportion of HT1080 cells arrested in the G₂/M phase increased to 34.8, 42.9 and 62.8%, respectively (Fig. 1B). These results indicated that 5-aza C inhibits proliferation in a dose-dependent manner (Fig. 1).

Metastatic HT1080 cells possess high migration ability and therefore they are suitable for migration and invasion experiments verifying the effects of 5-aza C on tumor metastasis. HT1080 24-h exposure to 5-aza C resulted in significant increase in HT1080 cell migration area and induced the spread of HT1080 cells along wound edges, compared with the untreated control cells (Fig. 2A and B). Next, Matrigel invasion assays were performed with 5-aza C-treated HT1080 cells (Fig. 2C). HT1080 cell invasion capacity was significantly induced by 5-aza C, and the number of invasive cells increased when 5-aza C concentration was increased to 10 μM (Fig. 2C). Approximately 29 and 54% induction in invasion was observed after treatment with 5 and 10 μM 5-aza C, respectively, compared with the control (Fig. 2C). Both HT1080 migration and invasion were unaffected by 20 μM 5-aza C treatment (Fig. 2B and C), most possibly due to a cytotoxic, cell-cycle arresting effect of the compound (Fig. 1B).

No cytotoxic effect was observed during HT1080 cell treatment with 5-aza C concentrations <10 μM and, therefore, 2–10 μM 5-aza C concentrations were used in subsequent experiments. To elucidate the potential mechanisms underlying the meta-static effects of 5-aza C on HT1080 cells, we assessed MMP-9 expression via zymography analysis. As shown in Fig. 3, 5-aza C treatment significantly induced MMP-9 expression in a dose- and time-dependent manner when compared with the control (Fig. 3). These findings suggested that induction of MMP-9 expression might be involved in the increased migration and invasion of HT1080 cells upon 5-aza C treatment.

To determine whether the MMP-9 activity and metastatic capacities of HT1080 cells induced by 5-aza C were associated with PI3-kinase and mitogen-activated protein kinase (MAPK) signaling pathways, we examined Akt, ERK1/2, p38 kinase and JNK1/2 phosphorylation. As shown in Fig. 4, 5-aza C caused a significant increase in phosphorylated Akt and ERK1/2 levels (Fig. 4). It also activated the p38 and JNK1/2 signaling pathways (data not shown). 5-aza C-induced MMP-9 activity and HT1080 cell metastatic capacities were significantly attenuated after blocking of ERK1/2 and PI3-kinase/Akt with PD and LY inhibitors, respectively (Fig. 5). In contrast, p38 kinase inhibition by SB203580 and JNK1/2 inhibition by SP600125 exerted no apparent effects (data not shown). Collectively, the data indicated that PI3-kinase/Akt and ERK1/2 may be the major pathways mediating 5-aza C-induced MMP-9 activity and metastatic capacities, migration and invasion, of HT1080 cells.