Human breast cancer cell lines, MCF-7 and MDA-MB-231, were obtained from the Shanghai Cell Bank (Shanghai, China). Cells were cultured in Dulbecco’s Modified Eagle’s medium (Gibco-BRL, Carlsbad, CA, USA) containing 10% fetal bovine serum in a humidified atmosphere with 5% CO₂ at 37°C. 2. The present study was approved by the ethics committee of Soochow University (Suzhou, Jiangsu, China).

In order to generate stable cell lines overexpressing BTG1, cDNAs for full-length human BTG1 were amplified by the polymerase chain reaction (PCR) with the following primer sequences: 5′-CACCATGCA TCCCTTCTACACCCGG-3′ (forward) and 5′-TTAACCTGA TACAGTCATCATATTG-3′ (reverse). The full-length cDNA was then cloned into a XhoI and BamHI linearized plasmid vector, pcDNA3 (Clontech Laboratories, Mountain View, CA, USA). The control pcDNA3 vector (pcDNA3-neo) or the human BTG1 expression vector, pcDNA3-BTG1, were transfected into MCF-7 or MDA-MB-231 cells using Lipofectamine 2000 (Invitrogen Life Technologies, Carlsbad, CA, USA). Stable clones were selected in medium containing 500 μg/ml G418 (Sigma-Aldrich, St. Louis, MO, USA). Individual clones were isolated and expanded for further characterization. The expression of BTG1 was determined using reverse transcription (RT)-PCR and western blot analyses.

A flat-bottomed 96-well plate was coated with 0.2 ml Matrigel™ (200 μg/ml) overnight at 4°C. Some wells were left uncoated as negative controls. The plate was washed twice with phosphate-buffered saline (PBS) and blocked with 1 mg/ml bovine serum albumin (BSA) for 2 h at 37°C, prior to the addition of 0.5 ml (5×10³) cells, which were transfected as described above. Following incubation at 0.5-, 1-, 1.5- and 2-h intervals at 37°C, unattached cells were removed from the plate by washing with PBS. MTT (Sigma-Aldrich) was added to each well and the percentage adhesion was calculated by dividing the absorbance values obtained from the coated wells by the absorbance values obtained from the uncoated wells (100%).

A wound-healing assay was used to detect cell migration in vitro. Cells were seeded in six-well plates at a density of 1×10⁶ cells/well and incubated at 37°C for 24 h. When cells were 90% confluent, a thin mark was drawn vertically with a pipette tip in the six-well plate. Cells were then washed three times with sterile PBS to remove any unattached cells. Following incubation for 24 or 48 h at 37°C with 5% CO₂, wound width was assessed using light microscopy.

Cell invasion was measured using Matrigel-coated Transwell^® inserts (Costar Inc., Cambridge, UK). Cells were initially cultured in serum-free medium for 12–24 h. A total of 5×10⁴ (100 μl) cells were then plated in serum-free medium with BSA on 24-well Transwell inserts coated with 100 ml Matrigel (250 mg/ml). The underside of the insert was precoated with 500 μl chemokine-containing cell supernatant. Following incubation for 48 h at 37°C with 5% CO₂, the inserts were fixed with 3.7% paraformaldehyde/PBS and stained with 2% crystal violet. The penetration of cells through the membrane was quantified by counting the number of cells that penetrated the membrane in 10 microscopic fields/filter (magnification, ×200). The experiment was repeated three times. Cell migration assays were performed in a similar manner, but without the Matrigel coating.

Each experimental group of cells was lysed using SDS sample buffer (80 mM Tris-HCl, 2% SDS, 300 mM NaCl and 1.6 mM EDTA). Cell extracts were separated using 10% SDS-PAGE, transferred onto nitrocellulose membranes and blocked with 5% skimmed milk. Following blocking, membranes were incubated with antibodies against β-actin, BTG1, matrix metalloproteinase (MMP)-2, MMP-9 and E-cadherin (E-cad) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), and then incubated with horseradish peroxidase (HRP)-conjugated anti-mouse or -rabbit immunoglobulin G antibodies (Santa Cruz Biotechnology, Inc.). Protein bands were visualized using enhanced chemiluminescence solution.

Four-week old female nude mice (specific pathogen-free BALB/c) were obtained from the Laboratory Animal Center of Soochow University (Suzhou, China) and housed at a constant temperature (23±2°C) and humidity (50–70%) with a 12-h light/dark cycle. MDA-MB-231 cells were either transfected with pcDNA3-BTG1 or pcDNA3-neo (negative control), or left untreated (blank), as described previously. Cells were grown to subconfluency, trypsinized, harvested, washed twice with PBS and resuspended in 0.2 ml PBS (5×10⁶ cells/0.2 ml), prior to subcutaneous injection into the three groups of mice. Each group contained six nude mice. Five weeks after injection, the presence of metastases in the lung and liver tissues from each mouse bearing a tumor mass on the back was evaluated by gross and microscopic examination. Tissue sections were deparaffinized, rehydrated and rinsed, prior to hematoxylin and eosin staining and examination for metastatic nodules and cell pathology.

Tumor xenografts were removed, deparaffinized, rehydrated, washed and subjected to antigen retrieval. Following washing in distilled water and PBS, xenografts were treated with 0.03% hydrogen peroxide for 5 min to block endogenous peroxidase activity. Xenografts were then incubated with mouse anti-human cluster of differentiation (CD) 31, BTG1 and vascular endothelial growth factor (VEGF) antibodies (Santa Cruz Biotechnology Inc.) diluted 1:100 for 60 min at room temperature. Following washing in PBS, the sections were incubated with labeled HRP-conjugated anti-mouse antibodies, for 30 min at room temperature, prior to washing twice in PBS and incubating with diaminobenzene for 10 min. Following washing, the sections were counterstained with hematoxylin, washed and dipped briefly in a water bath containing drops of ammonia, prior to dehydration and mounting in Diatex. The stained sections were analyzed and scored using a Nikon microscope (Nikon Corp., Tokyo, Japan).

Results are presented as the mean ± standard deviation. A value of P<0.05 was considered to indicate a statistically significant difference. Statistical analyses were performed using SPSS 17.0 (SPSS Inc., Chicago, IL, USA). Each experiment was repeated three times.

RT-PCR was used to assess the expression of BTG1 in human breast tissues from primary tumors, lymph node metastases, benign breast tumors and normal human breast tissue (Fig. 1A). BTG1 mRNA levels were observed to be significantly higher in normal human breast tissue than those in benign breast tumors and lymph node metastases (P<0.05). Compared with lymph node metastases, BTG1 mRNA levels were found to be significantly higher in benign breast tumors (P<0.05).

To further explore the role of BTG1 in breast cancer, MCF-7 and MDA-MB-231 cells were used to stably overexpress BTG1 (MDA-MB-231/BTG1 or MCF-7/BTG1). As shown in Fig. 1B, BTG1 mRNA and protein levels were examined in the MCF-7 and MDA-MB-231 cells using RT-PCR and western blot analysis. When compared with the blank or pcDNA3-neo cells, BTG1 expression was significantly increased in the pcDN3-BTG1-transfected cells.

To assess whether BTG1 expression is associated with adhesion in breast cancer cells, an in vitro adhesion assay was performed to evaluate the adhesive capacity of the blank, pcDNA3-neo and pcDNA3-BTG1 cells. As shown in Fig. 2A, the rate of cell adhesion in the pcDNA3-BTG1 group was significantly decreased compared with that in the blank and pcDNA3-neo groups of cells (P<0.05).

To examine whether overexpression of BTG1 decreases migratory and invasive capacity in breast cancer cell lines, wound healing, migration and invasion assays were performed in MDA-MB-231 and MCF-7 cells. The wound healing assay revealed that blank and pcDNA3-neo cells demonstrated more rapid wound closure than pcDNA3-BTG1 cells (Fig. 2B). Furthermore, as shown in Fig. 3A, cell invasion assay revealed that the number of invading cells significantly decreased with BTG1 overexpression (P<0.05). It was also observed that the BTG1-overexpressing pcDNA3-BTG1-transfected cells exhibited significantly reduced motility towards chemokines compared with that in blank and pcDNA3-neo breast cancer cells (Fig. 3B; P<0.05).

Western blot analysis was performed for detection of the metastasis-associated proteins MMP-2 and -9 and the cell-cell adhesion-associated protein E-cad. As shown in Fig. 3C, overexpression of BTG1 was observed to significantly inhibit MMP-2 and -9 expression (P<0.05) and significantly induce E-cad expression (P<0.05).

In summary, these results suggest that BTG1 expression is inversely associated with the invasiveness of breast cancer cells in vitro. The inverse correlation between BTG1 expression and invasive capacity in breast cancer cells in vitro indicates that BTG1 may be a metastasis suppressor gene in breast cancer cells.

Three groups of MDA-MB-231 cells were subcutaneously injected into nude mice. Five weeks after injection, the effect of BTG1 overexpression on the promotion of distant metastases was assessed in a mouse tumor model. Lung and liver tissue samples were obtained from the blank (n=6), pcDNA3-neo (n=6) and pcDNA3-BTG1 (n=6) groups of mice. Hepatic metastases were observed in the liver tissue in one out of six mice from the blank and pcDNA3-neo groups (Fig. 4A). However, no significant hepatic metastases were found in the mice in the pcDNA3-BTG1 group, and no pulmonary metastases were observed in any of the three groups.

The effect of BTG1 overexpression on angiogenesis in tumors in vivo was then assessed using immunohistochemistry. Anti-CD31 antibodies were utilized to detect CD31 levels, which were used to reflect the level of vascular endothelial proliferation. As shown in Fig. 4B, the number of tumor microvessels in the blank and pcDNA3-neo tumor groups, was 28.33±1.53 and 29.67±2.08, respectively. By contrast, the number of tumor microvessels in the pcDNA3-BTG1 group was 14.00±1.00. Furthermore, in the tumors in the pcDNA3-BTG1 group, BTG1 expression was observed to be higher (Fig. 4C) and the level of VEGF expression was observed to be lower than that in the blank and pcDNA3-neo groups. These results suggest that overexpression of BTG1 may be capable of inhibiting angiogenesis through the downregulation of VEGF expression, thereby inhibiting tumor metastasis.