A total of 142 cases of gastric adenocarcinoma patients who had radical gastrectomy at the 82nd hospital of the PLA between January 2008 and December 2009 were included in the present study. None of the patients received chemotherapy prior to surgery and were followed-up by a review of medical records and telephone calls. Data on gender, age, histological type, TNM stage, metastatic status and the Lauren classification were collected from the medical record library of the hospital. This study was approved by the Ethics Committee of the 82nd hospital of the PLA, and informed consent was obtained from the patients in accordance with the Declaration of Helsinki.

Immunohistochemical staining was carried out using a rabbit SP immunostaining kit (Zhongshan Goldenbridge Biotechnology, Beijing, China). Sections were dewaxed in xylene and dehydrated through a gradient concentration of alcohol. Then, sections were treated in a microwave for 10 min for antigen retrieval. After blocking the endogenous peroxidase and non-specific staining with 0.3% (v/v) hydrogen peroxide and normal goat serum, the sections were incubated with anti-DBC1 antibody (1:100 dilution, Bethyl Lab, Montgomery, TX, USA) overnight at 4°C. After washing with PBS, the sections were incubated with horseradish peroxidase (HRP)-conjugated biotinylated IgG for 30 min, followed by washing with phosphate-buffered saline (PBS). The sections were visualized by diaminobenzidine (DAB) solution and counterstained with hematoxylin (both from Zhongshan Goldenbridge Biotechnology). The sections incubated with rabbit IgG instead of primary antibody were used as negative controls.

Immunohistochemical staining was examined by two independent pathologists who were blinded to the clinicopathological information. Each case was evaluated to estimate the intensity of cell staining and the percentage of positive tumor cells. The intensity of cell staining was graded according to the following scale: 0 (negative), 1 (weak), 2 (moderate) and 3 (strong). The extent of staining was evaluated by the percentage of positive tumor cells: 0 (negative), 1 (1–25%), 2 (26–50%), 3 (51–75%) and 4 (76–100%). For each sample, the score for intensity was multiplied by the score for extent of staining to provide a final score. Therefore, the expression of DBC1 was defined as: 0 (negative, −), 1–4 (low expression, +), 5–8 (moderate expression, ++), and 9–12 (high expression, +++). For multiple lymph node metastases in a single case, an average immunohistochemical score was calculated to define the expression of DBC1 in metastatic lymph nodes.

Human MKN45 and MGC803 gastric cancer cell lines were cultured in modified Eagle’s medium (MEM) supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 µg/ml streptomycin. The cells were maintained at 37°C in a humidified 5% CO₂ incubator. For transient transfection, the gastric cancer cells were transfected with lentivirus-mediated DBC1 (Lent-DBC1) or DBC1 shRNA (Lent-sh-DBC1) at a multiplicity of infection of 10. The cells transfected with lentivirus vector encoding the green fluorescent protein were used as controls.

After a 48-h transfection, the gastric cancer cells were collected and continuously cultured in plates pre-coated with Poly-HEMA for 48 h (16,17). Subsequently, the cells were stained with Annexin V/PI and analyzed by flow cytometry.

The gastric cancer cells were seeded in 6-well plates at a concentration of 1.0×10⁵/well. After 24 h, the cells were infected with lentivirus and treated with Bay 11-7082 (10 µM) or PDTC (50 µM) 24 h later. After another 24 h, cells were co-transfected with 1 µg of NF-κB luciferase reporter pNF-κB-luc WT and 50 ng of Renilla luciferase plasmid phRL-TK (Promega, Madison, WI, USA) using Lipofectamine 2000 transfection reagent (Invitrogen, USA). Reporter activities were analyzed using the Promega dual luciferase assay kit (Promega) according to the manufacturer’s instructions. The luciferase activity was normalized to the Renilla luciferase activity.

Total cell lysate was prepared in 1X SDS buffer. Proteins were separated by SDS-PAGE and transferred onto PDVF membranes. The membranes were then blotted with antibodies to DBC1 (Bethyl Lab), c-FLIP, bcl-xl and β-actin (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA). Antigen-antibody complexes were visualized with enhanced chemiluminescence (Amersham Pharmacia Biotech, Piscataway, NJ, USA).

The TaqMan stem-loop RT-PCR method was used to assess the expression of c-FLIP, bcl-xl and GAPDH with kits from Applied Biosystems (Foster City, USA). For relative expression levels, the 2(−ΔCt) method was used as previously described (18). Experiments were carried out in triplicate for each data point, and a data analysis was performed using Bio-Rad IQ software (Hercules, CA, USA).

Data are presented as means ± SEM. The Chi-square test was used to compare the differences in the DBC1 expression level with various clinicopathological parameters in gastric cancer. Survival curves were plotted by the Kaplan-Meier method and the log-rank test was carried out to compare differences in survival. Statistical analyses were performed using SPSS 17.0 software (Chicago, IL, USA). P<0.05 was considered statistically significant.

Immunohistochemistry was performed to detect the expression of DBC1 in 142 cases of gastric adenocarcinoma tissues and the corresponding adjacent non-tumor tissues. DBC1 was detected primarily in the nuclei of cancer cells (Fig. 1). High expression (+++), moderate expression (++), low expression (+), and negative expression (−) of DBC1 were identified in 13, 32, 47 and 50 samples of the adenocarcinoma tissues, respectively, while in the adjacent non-tumor tissues, only a few cells with weak staining for DBC1 were observed in two cases. We also assessed the potential relationship between DBC1 expression and clinical characteristics of gastric adenocarcinomas. As shown in Table I, the expression of DBC1 was significantly correlated with TNM stage (P<0.05) and lymph node metastasis (P<0.05). However, there was no significant correlation between DBC1 expression and age, gender, distant metastasis, histological grade and the Lauren classification (P>0.05).

Given that DBC1 expression was significantly correlated with lymph node metastasis, we also detected the expression of DBC1 in lymph node metastasis and found that the expression level of DBC1 in the metastatic lymph nodes was concomitant with that in the primary gastric adenocarcinoma tissues (Linear correlation analysis for immunohistochemical scores, P<0.05). Moreover, in cases identified as low DBC1 expression, we found that the percentage of DBC1-positively expressed cancer cells in metastatic lymph nodes was markedly higher than that for the primary site (P<0.05, Fig. 1C).

The relationship between DBC1 expression and survival was also analyzed. The mean survival times in patients with high expression (+++), moderate expression (++), low expression (+) and negative expression (−) of DBC1 were 27.5, 38.0, 40.7 and 44.6 months, respectively. The Kaplan-Meier analysis (Fig. 2) revealed that DBC1 expression was significantly correlated with a shorter overall survival time (log-rank, Chi-square=8.551, P<0.05).

Resistance to anoikis has been considered a hallmark of metastatic cancer cells, as it is required for anchorage-independent growth during tumor dissemination. Previous studies have identified that DBC1 is an important co-factor for the control of the IKK-β/NF-κB signaling pathway that regulates anoikis (7). Therefore, we hypothesized that DBC1 is involved in the metastasis of gastric cancer cells by mediating resistance to anoikis via IKK-β/NF-κB signal. We initially examined the effect of DBC1 expression on anoikis in gastric cancer cells. To achieve this, lentivirus-mediated DBC1 or DBC1 shRNA were transfected into MKN45 and MGC803 cells. As shown by western blot analysis in Fig. 3A, DBC1 protein levels in the MKN45 and MGC803 cells following transfection with lentivirus-mediated DBC1 were significantly increased compared with that of controls. As expected, in the MKN45 and MGC803 cells, the DBC1 upregulation resulted in a significant inhibition of the cell detachment-induced anoikis (P<0.05, Fig. 3B). Concurrently, the specific down-regulation of DBC1 expression by lentivirus-mediated DBC1 shRNA induced an increased sensitivity to the anoikis in MKN45 and MGC803 cells (P<0.05, Fig. 3A and B).

In this study, Bay 11-7082, a highly specific inhibitor of IKK-β activity, and PDTC, a specific inhibitor of NF-κB activity, were used to assess the role of the IKK-β/NF-κB signaling pathway in the regulation of anoikis by DBC1 (19,20). Fig. 4A shows that in DBC1 upregulated gastric cancer cells (MKN45 and MGC803), the levels of NF-κB activation following detachment were significantly higher than that of the control cells (P<0.05). However, we also observed that in gastric cancer cells cultured in normal cell plates, DBC1 upregulation did not obviously influence the activation of NF-κB (P>0.05). Fig. 4B shows that NF-κB activation was significantly inhibited in gastric cancer cells following detachment after exposure of Bay 11-7082 or PDTC, and coincidently, the anoikis rates of gastric cancer cells were significantly increased. Fig. 4C shows that either Bay 11-7082 or PDTC reversed the increased anoikis resistance in DBC1-upregulated gastric cancer cells, which indicated that DBC1 contributed to the anoikis resistance by activating the IKK-β/NF-κB signaling pathway.

Among NF-κB target genes, c-FLIP and bcl-xl are particularly noted for their ability to regulate anoikis (7,21,22). We examined the mRNA and protein expression changes of c-FLIP and bcl-xl in DBC1 upregulated gastric cancer cells (MKN45 and MGC803). As shown in Fig. 5A and B, the mRNA and protein expression levels of c-FLIP and bcl-xl were significantly increased in DBC1 upregulated gastric cancer cells following detachment. We also examined the effects of Bay 11-7082 and PDTC on the expression of c-FLIP and bcl-xl. As expected, after exposure of Bay 11-7082 or PDTC, the improved mRNA and protein expression of c-FLIP and bcl-xl were reversed in the DBC1 upregulated gastric cancer cells, further indicating that DBC1 promotes anoikis resistance by regulating NF-κB-mediated transcription.