Thirty-one urothelial carcinoma samples and 22 corresponding adjacent tissues were collected from patients who underwent total cystectomy. Nine patients were treated with transurethral resection of bladder carcinoma but no adjacent tissue was obtained. Patients were enrolled at the Department of Urology at the First Affiliated Hospital of Chongqing Medical University. All tissue specimens were confirmed to be bladder cancer or normal histologically, and histological grade and stage were determined according to UICC guidelines. Written informed consent was received from all participants. This study was approved by the Ethics Committee of Chongqing Medical University. All tissue specimens were stored at −80°C before the experiment.

The assay was performed as described previously (19). In brief, paraffin wax-embedded tissue sections were dewaxed, rehydrated and microwaved for 30 min in sodium citrate buffer to retrieve antigen epitopes. Endogenous peroxidase activity was suppressed by 3% H₂O₂ and blocked by goat serum 5% BSA. Diluted primary polyclonal rabbit antibodies against hepaCAM (ProteinTech, China), DNMT3A and DNMT3B (Immunoway Biotechnology, China) were added and left at 4°C overnight. As secondary reagents, we used biotin-labeled anti-IgG and avidin-biotin horseradish peroxidase complex, followed by staining with the chromogen diaminobenzidine (Zhongshan, China) until a brown color developed. Slides were counterstained with Mayer's hematoxylin and differentiated in a solution containing 1% hydrochloric acid and 99% ethanol. Cell nuclei were stained blue using lithium carbonate. Sections were dehydrated and a transparent coverslip was added to enable observation by microscopy.

Human bladder cancer cell lines (T24, EJ and BIU-87) were obtained from the American Type Culture Collection (ATCC; USA). All the cells were cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum (both from Gibco, USA) under standard culture conditions (5% CO₂ at 37°C).

To determine the optimal drug doses, bladder cancer cells (T24, EJ and BIU-87) were seeded in 96-well plates at a density of 1×10⁴ cells/well for 12 h and treated with 5-azacytidine (Sigma, USA) at different concentrations (0.5, 1, 2, 3, 4, 5, 6 and 7 µg/ml). Three plates were seeded, and the cells were cultured for 24, 48, 72 and 96 h, respectively. Before removal from the incubator, the cells were incubated with 20 µl 0.5 mg/ml MTT (Sigma) for an additional 4 h. Culture medium was discarded, and 150 µl dimethylsulfoxide (DMSO; Sigma) was added to dissolve the formazan crystals. The absorbance was measured in a microplate reader at an optical density (OD) of 492 nm. Inhibition rate was calculated using the formula: IR = (1 − experiment group/control group) × 100%.

To evaluate the cell viability, the CCK-8 Kit (Beyotime, China) was applied according to the manufacturer's instructions. The results were repeated in triplicate. The OD of the untreated controls was measured as 100% survival.

Bladder cancer cells were seeded into 6-well plates at a density of 300 cells/well. After a 48-h incubation, cells were treated with AZAC or DMSO. After 13 days, the cells were fixed with methanol for 20 min and stained with crystal violet. Colonies were observed and images were captured under a microscope.

Bladder cancer cells (1×10⁶) treated with AZAC or DMSO were cultured in 6-well plates for 48 h. The cells were collected in cold phosphate-buffered solution (PBS), fixed in 70% ethanol, and stored at 4°C for subsequent cell cycle analysis. Procedures for testing the cell cycle have been previously described (7).

Genomic DNA was extracted from bladder cancer cells according to the instructions provided in the TIANamp Genomic DNA kit (Tiangen Biotech, China). The bisulfite modification procedures were carried out using EZ DNA Methylation-Gold™ kit (Zymo Research, USA). Methylation-specific polymerase chain reaction (MSP) was performed to detect the methylation status of the hepaCAM gene (M, methylation status; U, unmethylation status). The PCR conditions were as follows: pre-denaturation at 98°C for 10 min, denaturation at 95°C for 10 sec; annealing at 58.4°C (U)/61.8°C (M) for 60 sec; extension at 72°C for 30 sec, final extension at 72°C for 5 min. Primer sequences were: hepaCAM(M) forward, AGAATTCGGTTTCGGAGTTTC and reverse, CTAAACGACGAC GAATATATCCG; and hepaCAM(U) forward, AGAATTTGGTTTTGGAGTTTTGA and reverse, CCTAAACAACAACAAATATATCCAAC. PCR products were separated on 3% agarose gel.

Total RNA was isolated from bladder cancer cells using RNAiso Plus (Takara, Japan) according to the manufacturer's instructions. Complementary DNA was synthesized using a reverse transcription kit (Takara) according to the manufacturer's protocol. The PCR conditions consisted of predenaturing at 95°C for 5 min, denaturing at 95°C for 10 sec, annealing for 50 sec, extension at 72°C for 1 min, a total of 35 cycles from denaturing, and a final extension at 72°C for 5 min. Primer sequences were as follows: hepaCAM forward, TACTGTAGATGTGCCCATTTCG and reverse, CTTCTGGTTTCAGGCGGTC; DNMT3A forward, GGCGTTAGTGACAAGAGGG and reverse, TGGACTGGG AAACCAAATA; DNMT3B forward, CAAAGAGTTGGGC ATAAAGG and reverse, GCGTGAGTAATTCAGCAGGT; and β-actin forward, CACCACACCTTCTACAATGAGC and reverse, GTGATCTCCTTCTGCATCCTGT. The β-actin gene was used as an internal standard. All RT-PCR reactions were executed in triplicate. Each PCR product was electrophoretically tested on 1.5% agarose gel.

Total protein extraction from cells and tissues was performed using RIPA buffer supplemented with protease inhibitor (PMSF) and phosphatase inhibitors (NaF and Na₃VO₄) (Roche, Switzerland). The BCA Protein Assay kit (Beyotime) was used to detect the quantity of each sample according to the manufacturer's instructions. Proteins (100 µg) were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Invitrogen, USA). Then the proteins were electrotransferred onto polyvinylidene difluoride membranes (Millipore, USA). The membranes were blocked with 5% skim milk at room temperature for 1 h and then incubated with the primary antibody against hepaCAM, DNMT3A, DNMT3B and β-actin (Zhongshan, China) overnight at 4°C. Membranes were washed twice with TBST and once with TBS, and then incubated with the HRP-conjugated secondary antibody for 1 h at room temperature. Proteins were visualized using a chemiluminescence (ECL) reagent (Millipore, Billerica, MA, USA), and the densities of the bands were quantified and normalized to that of β-actin by Quantity One software. The experiments were performed as described previously (20).

Nude mice (4–6 weeks of age) were purchased from the Animal Institute of the Chinese Medical Academy (Beijing, China). Viable EJ cells (3×10⁶) resuspended in PBS were injected into the right flanks of 8 male nude mice. The mice were randomized into two groups: group A (DMSO group) and group B (AZAC group) after 2 weeks. Subsequently, AZAC was injected into mice in group B and DMSO in group A every three days for 6 times. The weight of each nude mice was determined daily for 3 weeks. After 3 weeks, all mice were sacrificed. The xenograft tumor weight was measured at the terminal time, and tumor tissues were resected for hematoxylin and eosin (H&E) staining and immunohistochemical examination of hepaCAM, DNMT3A and DNMT3B. All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Chongqing Medical University.

All statistical analyses were performed with SPSS 19.0 software using paired-samples t-test. Data are shown as mean ± SD, and P<0.05 served as the criterion for statistical significance.

To investigate whether there was any correlation between DNMT3A/3B and hepaCAM expression, we used anti-DNMT3A/3B and anti-hepaCAM antibodies to detect the expression of DNMT3A/3B and hepaCAM in the bladder cancer and adjacent tissues. The results showed that DNMT3A/3B was strongly expressed in the bladder cancer tissues, but weakly expressed in the adjacent tissues. However, in regards to hepaCAM, the expression levels in the adjacent tissues were higher than levels in the cancer tissues (Fig. 1A and B). We made use of the mean density for estimating the protein expression levels of DNMT3A/3B and hepaCAM. The results revealed a negative linear correlation between DNMT3A/3B and hepaCAM expression in the same patients according to Pearson's analysis (r=−0.7176/−0.7127, P<0.05, Fig. 1C). However, there was no significant difference in expression levels in regards to age, gender, and disease recurrence (Tables I and II).

In contrast, protein lysates were extracted from 10 bladder cancer patient samples. Importantly, western blotting with antibodies against DNMT3A/3B showed that 9 of the 10 bladder cancer tissues had elevated levels of DNMT3A/3B, respectively, whereas only 1 of the 10 adjacent normal tissues had detectable expression of DNMT3A/3B (P<0.05, Fig. 1D and E).

To confirm the mode of action of AZAC in bladder cancer cells, we determined the absorption at a variety of concentrations and times. We calculated the IR with the above-documented equation. The results indicated that bladder cancer cell proliferation was markedly inhibited by AZAC in a concentration- and dose-dependent manner (P<0.05, Table III). According to Table III, 4 µg/ml (BIU-87) and 5 µg/ml (T24 and EJ) was confirmed as the most appropriate concentrations. The dose-effect curve indicated that the bladder cancer cell IR gradually increased with prolongation of treatment time at the same concentration compared with the controls (P<0.05, Fig. 2A). Based on these results, concentrations of 4 µg/ml (BIU-87 cells) and 5 µg/ml (T24 and EJ cells) AZAC were selected as the optimal treatment conditions for the following experiments.

Hypermethylation of the hepaCAM promoter was apparently observed in the T24, EJ and BIU-87 cell lines by use of MSP. In addition, the hepaCAM promoter was obviously demethylated by 5 µg/ml of AZAC in the T24 and EJ cell lines, and 4 µg/ml of AZAC in the BIU-87 cell line at 48 h (Fig. 2B).

Based on our findings, we analyzed the mRNA and protein expression of DNMT3A/3B and hepaCAM by RT-PCR and immunoblotting. The results revealed that DNMT inhibitor AZAC inhibited both the mRNA and protein levels of DNMT3A/3B in the bladder cancer cell lines. AZAC not only reversed the expression of hepaCAM mRNA (Fig. 2C and D) but also its protein level (Fig. 2E and F). These results suggest that the inactivation of DNMT3A/3B by AZAC treatment may participate in restoring the expression of the tumor-suppressor gene hepaCAM.

To investigate the effect of AZAC against bladder cancer cells, cell proliferation was detected by CCK-8 assay, colony formation assay and flow cytometry. The results showed that the growth of bladder cancer cells was inhibited after exposure to AZAC for 48 h compared with the control groups (P<0.05, Fig. 3A). This result was further supported by a colony formation assay. As shown in Fig. 3B and C, there was a significantly lower colony formation potential in the AZAC-treated cells compared to that in the DMSO treatment groups (P<0.05).

It has been reported that the growth inhibition of AZAC is always associated with cell cycle arrest (21). To evaluate the exact cell cycle phase, bladder cancer cells were inspected at 48 h, under stimulation by 4 µg/ml (BIU-87) and 5 µg/ml (T24 and EJ) AZAC. The results revealed that bladder cancer cells treated with AZAC had a significant accumulation in the G0/G1 phase compared to the controls (P<0.05, Fig. 3D and E). These data indicate that the growth of bladder cancer cells was markedly inhibited by AZAC.

To further investigate the tumor-suppressive function of AZAC and its mechanisms in vivo, the antitumorigenicity of AZAC was tested in nude mice. Thirty days after injection, tumors were excised from the tested mice for further analysis. The volume and weight of the tumors in the AZAC treatment groups were significantly decreased, compared with the control groups (P<0.05, Fig. 4A–D). Immunohistochemistry was further performed to analyze the expression of DNMT3A/3B and hepaCAM in the xenograft tumors. Numerous tumor cells with high nuclear fragmentation were found in H&E-stained sections from the DMSO treatment groups compared with the AZAC groups. Furthermore, low expression of DNMT3A/3B but high levels of hepaCAM immunostaining were observed in the AZAC treatment-derived tumor tissues (Fig. 4E). Taken together, these data indicate that AZAC acts as an antitumor agent, suppressing the proliferation of bladder cancer and enhancing the expression of hepaCAM via downregulation of the levels of DNMT3A/3B.