The immortalized human bladder epithelial cell line SV-HUC-1 and the human urinary bladder transitional carcinoma cell line T24 were both purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA). The human BC cell line UM-UC-3 was purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). SV-HUC-1 cells were cultured in F12K, T24 cells were cultured in RPMI-1640 medium, and UM-UC-3 cells were cultured in MEM. All media were supplemented with 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA), in a humidified air atmosphere of 5% CO₂ at 37°C. Trypsin (0.25% with 1 mM EDTA) (Invitrogen, Carlsbad, CA, USA) was also used to harvest the cells for further experiments.

One day prior to transfection, the cells were plated in growth medium without antibiotics at a density of 50–60%. miR-152 mimic and mimic-control were transfected into UM-UC-3 cells using Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocol. miR-152 inhibitor and inhibitor-control were transfected into SV-HUC-1 cells as previously described (23). Cells were harvested at 24, 48, 72 and 96 h post-transfection. For combined treatment of the miR-152 inhibitor and the DNMT inhibitor, 5-aza-2-deoxycytidine (DAC), SV-HUC-1 cells transfected with miR-152 inhibitor were further divided into two groups, in which one group was treated with 12.5 µM DAC (Sigma-Aldrich; Merck KGaA) for 72 h, and the other group was used as a negative control.

A set of 24 bladder tumor specimens was obtained from patients who underwent BC surgery between January 2012 and May 2015 at the First Affiliated Hospital of Guangzhou Medical University (Guangzhou, Guangdong, China). None of the patients received antitumor treatment prior to tumor sampling. A total of 24 adjacent tissues of bladder tumors from matched patients was collected as the control group. Details of the characteristics of the patients were described in our previous study (24). In order to further validate the relationship between miR-152 and DNMT1 mRNA expression, public data from both primary invasive and papillary BC tissues from ‘The Cancer Genome Atlas’ (TCGA) data portal (https://tcga-data.nci.nih.gov) were used, and explored through the cBio Cancer Genomics Portal (http://cbioportal.org). The data from 404 patients from Illumina HiSeq gene expression platforms (Illumina, Inc., San Diego, CA, USA) were extracted.

Total RNA from tissues or cell lines was extracted using TRIzol reagent (Invitrogen) according to the manufacturer's instructions. The quantity and quality of the RNA were determined by UV spectrophotometer. Bulge-loop™ miRNA qRT-PCR Primer sets (one RT primer and a pair of qPCR primers for each set) specific for miR-152 were designed by Guangzhou RiboBio Co., Ltd. (Guangzhou, China). The relative quantification value of the target, normalized to the control, was calculated by the the ΔΔCq method (26). PCR was carried out as follows: 95°C for 20 sec, 40 cycles of 10 sec at 95°C, 20 sec at 58°C and 30 sec at 72°C. Samples were analyzed in triplicate.

Protein was extracted from T24, UM-UC-3 and SV-HUC-1 cells for western blotting. Cultured cells were collected and washed three times with PBS. Following incubation with RIPA buffer (Sigma-Aldrich; Merck KGaA) containing 5 mM EDTA, PMSF, cocktail inhibitor and phosphatase inhibitor cocktail, cells were collected in a centrifuge tube. Cell lysates were centrifuged at 13,000 × g for 15 min at 4°C and insoluble debris was discarded. The proteins were determined by BCA quantitative method. Soluble proteins (30 µg) were subjected to 8% SDS-PAGE and transferred to a Hybond-P polyvinylidene difluoride (PVDF) membrane. Membranes were blocked in TPBS (PBS with 0.05% Tween-20) containing 5% (w/v) non-fat dry milk for 1 h at room temperature, washed in TPBS and then incubated with primary antibody. Immunoblottings were performed with 1:1,000 diluted anti-DNMT1 antibody (cat. no. D4692; Sigma-Aldrich; Merck KGaA) and 1:1,000 diluted anti-GAPDH antibody (cat. no. sc-166545; Santa Cruz Biotechnology, Santa Cruz, CA, USA). Blots were visualized using the Immobilon solutions (EMD Millipore, Billerica, MA, USA) under a chemiluminescence detection system, the ChemiDoc XRS (Bio-Rad Laboratories). Band area intensity was analyzed using Quantity One software (Bio-Rad Laboratories).

Genomic DNA of tissues and cells was extracted using standard phenol/chloroform purification and ethanol precipitation. Bisulfite modification of the genomic DNA was performed using EZ DNA Methylation-Gold kit (Zymo Research Corp., Irvine, CA, USA). Methylation of each sample was evaluated in triplicate using MSP. PCR products were electrophoresed on a 2% agarose gel for analysis. Primer sequences for miR-152 are listed in Table I.

Chromatin immunoprecipitation was performed using EZ-Magna ChIP A/G kit (EMD Millipore) according to the manufacturer's instructions. Briefly, cells were grown on 100-mm plates to 85% confluence; formaldehyde was added to a final concentration of 1%, and the plates were incubated for 10 min at 37°C; the cross-linking reaction was stopped by the addition of 0.125 M glycine for 5 min at room temperature. Cells were sonicated and sheared to yield fragments between 200 and 1,000 bp. Anti-DNMT1 (5 µg; Abcam, Cambridge, UK) was added to the sonicated samples and incubated at 4°C overnight with rotation. Immune complexes were collected with Protein A/G agarose beads and washed with buffer to remove nonspecific binding. Protein/DNA complex was reverse crosslinked and DNA was purified using spin columns. DNA was detected with quantitative PCR. Primers for ChIP-qPCR are listed in Table I.

Cell proliferation analysis. The miR-152 mimic was transfected into the T24 and UM-UC-3 cells and the miR-152 inhibitor was transfected into the SV-HUC-1 cells lasting 6 h. Subsequently, the cells were plated into a new dish. Cells (2×10³) were plated in triplicate and measured at the indicated time-points: 24, 48, 72 and 96 h. The number of cells was determined using the CellTiter 96^® AQueous One Solution assay (Promega Corp., Madison, WI, USA). Triplicate plates were counted for each cell line.

The cells (1×10⁵ cells/well) were seeded in 24-well plates. When cells reached 70–80% confluence, 0.5 µg of reporter plasmids alone or co-transfected with or without microRNA mimics were transiently transfected using Lipofectamine 2000 according to the manufacturer's instructions (Invitrogen; Thermo Fisher Scientific, Inc.). Luciferase activity was determined as previously reported (23).

All analyses were performed with the SPSS 13 statistical software (SPSS, Inc., Chicago, IL, USA). Statistical significance was determined by Student's t-test and Mann-Whitney U test. The relationship between the expression level of miR-152 and clinicopathological parameters was analyzed using the Mann-Whitney U test. The correlations were analyzed using Spearman's correlation rank test. Differences in probability values (P-values) of <0.05 were considered to indicate a statistically significant difference.

To determine the expression of miR-152 in BC, real-time PCR was used to compare the expression levels of miR-152 in T24 cells, UM-UC-3 cells and SV-HUC-1 cells. A lower level of miR-152 was observed in T24 cells and UM-UC-3 cells compared with SV-HUC-1 cells (Fig. 1A). We further compared the expression of miR-152 between 24 BC tissue specimens and the matched adjacent normal tissues. Consistent with our observations in cells, compared to the matched adjacent tissues of BC, miR-152 had a relatively lower expression in BC tissues (Fig. 1B). Additionally, after transfection with the miR-152 mimic, the expression of miR-152 in T24 cells had an increased level in a time-dependent manner (Fig. 1C). Conversely, treatment with miR-152 inhibitor in SV-HUC-1 cells resulted in a reduction of miR-152 expression in a time-dependent manner (Fig. 1D). These results indicated that miR-152 was downregulated in BC cell and tumor tissues. Notably, the analysis of the clinicopathological features of BC patients revealed that a lower miR-152 expression level was significantly correlated with the stage and grade of BC patients, which was independent of patient sex and age (Table II).

The reduction or absence of gene expression mediated by DNA hypermethylation plays an important role in cancer development (27). We hypothesized that downregulation of miR-152 in BC was due to DNA methylation. To verify our hypothesis, we used an MSP method to determine the methylation status of miR-152 promoter in T24 and SV-HUC-1 cells. We observed that miR-152 was methylated in T24 cells, but not in SV-HUC-1 cells (Fig. 2A). Similarly, 17 out of 24 (70.8%) samples had CpG island methylation of miR-152 promoter in BC tissues, while only 9 out of 24 (37.5%) samples had CpG island methylation of miR-152 promoter in the matched normal tissues adjacent to BC (Fig. 2B and C). Conversely, our previous study demonstrated that the expression level of DNMT1 was significantly higher in BC cells and tissues (24). To explore the potential regulatory circuit of DNMT1 and miR-152, we transfected miR-152 mimics in T24 cells and detected the expression of DNMT1 and the methylation level of miR-152 CpG island promoter. We determined that both DNMT1 mRNA expression and the DNA methylation of miR-152 promoter were decreased in T24 cells transfected with miR-152 mimics (Figs. 3B and 2D). Notably, the binding of DNMT1 to the miR-152 CpG island promoter was significantly decreased in T24 cells transfected with miR-152 mimics (Fig. 2E). In addition, DNMT1 binding to the miR-152 CpG island promoter was significantly increased in SV-HUC-1 cells transfected with the miR-152 inhibitor compared to the control group (Fig. 2F). To further determine the relationship between the expression of miR-152 and DNMT1, we performed Spearman's rank correlation analysis and found a significantly negative correlation between the expression levels of miR-152 and DNMT1 in BC specimens (r=−0.623, P=0.01) (Fig. 3A). In order to further confirm these findings, we determined DNMT1 and miR-152 expression in The Cancer Genome Atlas (TCGA) dataset which includes a total of 404 BC samples. Notably, we also identified a negative correlation (r=−0.145) in BC samples of stage IV, although there was no correlation of DNMT1 with miR-152 mRNA expression among BC samples of all stages. In addition, miR-152 levels may affect disease-free survival (P=0.026) (data not shown). Collectively, our results indicated that hypermethylation via DNMT1 is responsible for the silencing of miR-152 expression in BC.

DNMT1 is the key enzyme in DNA methylation which has been identified as one of the high-scoring candidate genes of miR-152 targets (28). To further determine whether the expression of DNMT1 is regulated by miR-152, we first determined the levels of DNMT1 mRNA in T24 and UM-UC-3 cells transfected miR-152 mimics by real-time PCR. We found that DNMT1 expression was decreased in both of T24 and UM-UC-3 cells transfected with the miR-152 mimic in a time-dependent manner (Fig. 3B and C). Conversely, the level of DNMT1 mRNA in SV-HUC-1 cells transfected the miR-152 inhibitor was increased in a time-dependent manner (Fig. 3D). These results were ascertained by western blot analysis (Fig. 3E-G). Notably, we revealed that miR-152 could directly target DNMT1 3′-UTR and suppress its expression in NiS-transformed cells (23). To further validate our results in T24 cells and SV-HUC-1 cells, a luciferase reporter assay was performed, in which miR-152 significantly reduced DNMT1 wild-type 3′-UTR luciferase activity, but not DNMT1 Mu 3′-UTR activity (Fig. 4), indicating that miR-152 directly bound to the DNMT1 3′-UTR and inhibited its expression in BC cells.

In order to clarify whether the expression of miR-152 plays a significant role in BC malignant phenotypes, we explored the effects of ectopically expressed miR-152 on the proliferation of T24, UM-UC-3 and SV-HUC-1 cells. Both T24 and UM-UC-3 cells transfected with miR-152 mimics exhibited significantly decreased cell proliferation at 48, 72 and 96 h, compared to the miR-control (Fig. 5A and B). Conversely, SV-HUC-1 cells transfected with miR-152 inhibitor had increased cell proliferation at 48, 72, and 96 h, compared to the miR-control (Fig. 5C). Notably, the combined treatment of the miR-152 inhibitor and the DNMT inhibitor, 5-aza-2-deoxycytidine (DAC), led to decreased cell proliferation compared to suppression of the miR-152 inhibitor alone, whereas the combined treatment resulted in a slight elevation of cell proliferation compared to that in control SV-HUC-1 cells at 48 and 72 h post treatment. These findings highlighted a reciprocal regulatory loop between miR-152 and DNMT1.