In total, 49 pairs of BC tissues and corresponding adjacent normal bladder tissues were collected from Peking University Shenzhen Hospital (Shenzhen, China) from January 2010 to November 2011. Tissue specimens were collected from 49 patients (aged 30 to 80 years old) with BC who underwent cystectomy; 40 male patients and 9 female patients. All patients were diagnosed as transitional cell carcinoma clinically and pathologically. The exclusion criteria included patients with other tumors, patients with a history of other cancer treatments, or patients with bladder cancer who had received chemotherapy or radiation therapy before surgery. All human tissue samples were obtained with informed consent. The Ethics Committee of Peking University Shenzhen Hospital in China approved this study (approval no. 20090017).

All cell lines were obtained from the American Type Culture Collection and maintained using standard media and conditions. Human BC cells (T24, TCCSUP, UM-UC-3, J82 and 5637), human normal bladder epithelial cells (SV-HUC-1) and 293T cells were maintained in Roswell Park Memorial Institute (RPMI)-1640 medium, Dulbecco's modified Eagle's medium (DMEM) or F-12K supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (PS) (all from Gibco; Thermo Fisher Scientific Inc.). All cells were cultured at 37°C in a 5% CO₂ incubator.

Cells were transfected with 100 nM small interfering (si)RNA or mimics or inhibitors using Lipofectamine 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) for 24–48 h at 37°C. Then, the transfected cells were analyzed. All siRNAs, mimics and inhibitors were synthesized by Suzhou GenePharma Co., Ltd. The sequences of siRNAs were as follows: Negative control (NC), 5′-UUCUCCGAACGUGUCACGUTT-3′; siCASC9-1, 5′-CAACUGGAUUCCAACUUUAUU-3′; siCASC9-2, 5′-CAAGAAGUUUAGUAAACCAUU-3′; siCASC9-3, 5′-GAGAUCAUUAAGCCCAGAAUU-3′; mimics NC, 5′-UUGUACUACACAAAAGUACUG-3′; inhibitor NC, 5′-CAGUACUUUUGUGUAGUACAA-3′; mimics miR-758-3p, 5′-UUUGUGACCUGGUCCACUAACC-3′; inhibitor miR-758-3p, 5′-GGUUAGUGGACCAGGUCACAAA-3′.

Total RNA was extracted from cells or tissue specimens using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.). The cDNA was synthesized with random primers using PrimeScript RT reagent Kit (Takara Biotechnology, Co., Ltd.) or miScript II RT kit (Qiagen GmbH) according to the manufacturer's instructions. Quantitative RT-PCR was performed on a Roche Lightcycler 480 (Roche Diagnostics) using SYBR Premix Ex Taq kit (Takara Biotechnology, Co., Ltd.) according to the manufacturer's instructions. Quantitative RT-PCR amplification was performed according to the following thermocycling conditions: 30 sec at 95°C for initial denaturation; 40 cycles of 5 sec at 95°C for denaturation and 31 sec at 60°C for extension; and 10 min at 60°C for final extension. Human EMT RT² Profiler PCR Array (Qiagen GmbH) was used to analyze the expression of genes involved in EMT. The relative expression levels of candidate genes were analyzed using the 2^−ΔΔCq method (35). The primers for AHNAK, CTNNB1, EGFR, FN1, ITGAV, PDGFRB and SNAI3 were purchased from Qiagen, Inc. The primer sequences were as follows: AHNAK forward, 5′-CAGGCATTGGTGTTCAAGGC-3′ and reverse, 5′-TCTGCCCAGTTGGGAGTTTC-3′; CTNNB1 forward, 5′-TTGTGCGGCGCCATTTTAAG-3′ and reverse, 5′-TCCTCAGACCTTCCTCCGTC-3′; EGFR forward, 5′-AAGGCACGAGTAACAAGC-3′ and reverse, 5′-AGGGCAATGAGGACATAA-3′; FN1 forward, 5′-TGTGCCAAAGCTTTACTACTGT-3′ and reverse, 5′-TATTTCCCCCGAAGGTGTCT-3′; ITGAV forward, 5′-TCACTAAGCGGGATCTTGCC-3′ and reverse, 5′-AAGCACTGAGCAACTCCACA-3′; PDGFRB forward, 5′-GCTGTTACCCACTCTGGGAC-3′ and reverse, 5′-TGGTGTCCTTGCTGCTGATG-3′; SNAI3 forward, 5′-GCACAACTACCTCTCAGCCA-3′ and reverse, 5′-ATAGACGTGTGACATGGGGC-3′; CASC9 forward, 5′-CCAGACAGCAGCAAAGCAAT-3′ and reverse, 5′-GGAAGCAGCAAATGTGTCCAT-3′; TGF-β2 forward, 5′-CGACGAAGAGTACTACGCCA-3′ and reverse, 5′-GATGGCATTTTCGGAGGGGA-3′; GAPDH forward, 5′-CGCTCTCTGCTCCTCCTGTTC-3′ and reverse, 5′-ATCCGTTGACTCCGACCTTCAC-3′; U6 forward, 5′-CTCGCTTCGGCAGCACA-3′ and reverse, 5′-ACGCTTCACGAATTTGCGT-3′; miR-758-3p forward, 5′-ACACTCCAGCTGGGTTTGTGACCTGGTCCA-3′ and reverse, 5′-CTCAACTGGTGTCGTGGAGTCGGCAATTCAGTTGAGGGTTAGTG-3′.

Cell Counting Kit-8 (CCK-8; US Everbright, Inc.) and EdU assay kit (Guangzhou RiboBio Co., Ltd.) were used to assess cell proliferation. Experiments were performed as previously described (36).

Cell migration was determined using wound healing assays. Transwell assays without or with Matrigel were used to assess BC cell migration and invasion abilities, respectively. Wound healing, Transwell and flow cytometric assays were performed as previously described (36). Serum-free medium was used in the wound healing experiment.

The CASC9 (or TGF-β2) fragment containing the predicted miR-758-3p binding site or a fragment with a mutated binding site were cloned into the psiCHECK-2 luciferase reporter vector (Wuhan GeneCreate Biological Engineering Co., Ltd.). Then, 1 µg/ml luciferase reporter vector psiCHECK-2-CASC9-WT (wild type) or psiCHECK-2-CASC9-MT (mutant type) or psiCHECK-2-TGF-β2-WT or psiCHECK-2-TGF-β2-MT and 100 nM miRNA mimic or mimic NC were co-transfected into 293T cells using Lipofectamine 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) for 24 h at 37°C. Renilla and firefly luciferase activities were detected using the Dual luciferase reporter assay system (Promega Corporation). miRNA mimic and mimic NC sequences were as follows: mimics NC, 5′-UUGUACUACACAAAAGUACUG-3′; and mimics miR-758-3p, 5′-UUUGUGACCUGGUCCACUAACC-3′.

Anti-E-cadherin (product code ab15148), anti-N-cadherin (product code ab18203) and anti-GAPDH (product code ab9485) were purchased from Abcam. Goat anti-mouse IgG-HRP (cat. no. sc-2005) and goat anti-rabbit IgG-HRP (cat. no. sc-2004) were purchased from Santa Cruz Biotechnologies. Western blotting was performed as previously described (36).

FISH assay was performed using the Ribo™ FISH Kit (Guangzhou RiboBio Co., Ltd.) according to the manufacturer's instructions. The lncRNA CASC9 probe was designed and synthesized by Guangzhou RiboBio Co., Ltd. and was labeled with Cy3 fluorescent dye. BC cells were seeded onto sterile coverslips until cells reached 30–60% confluence. The cells were washed with PBS, fixed with 4% paraformaldehyde for 10 min at 25°C, and then permeabilized with 0.5% Triton X-100 (PBS) for 10 min at 4°C. Next, the cells were blocked with prehybridization buffer for 30 min at 37°C and then incubated in 0.5 µM lncRNA CASC9 probe in hybridization buffer at 37°C overnight. The cells were then washed with saline sodium citrate (SSC) buffer solution and stained with DAPI for 10 min at 25°C. Finally, the cell slides were removed from the plate and fixed on a glass slide for detection by fluorescence microscopy (magnification, ×400).

The data were presented as the mean ± standard error of mean (SEM). Log-rank test, chi-square test, one-way ANOVA with Bonferroni post hoc test, paired and unpaired Student's t-test were employed for statistical analysis. Kaplan-Meier survival analysis from http://gepia.cancer-pku.cn (37) was used to reveal that the relationship between CASC9 expression and the prognosis of bladder cancer patients. Survival analysis was performed using log-rank test. Chi-square test was used to assess the association between CASC9 expression and clinicopathological characteristics of bladder cancer patients. When comparing the population means of only two groups, the Student's t-test was used, and when means of more than two groups were compared, ANOVA was selected. Paired Student's t-test was used to assess CASC9 expression in 49 pairs of BC tissues (Tumor) and matched adjacent normal tissues (Normal). The association of CASC9 expression and BC tumor invasion depth was calculated by one-way ANOVA with Bonferroni post hoc test. P<0.05 was considered to indicate a statistically significant difference. All statistical analyses were carried out with Graphpad Prism 6 (GraphPad Software, Inc.).

Bioinformatics tools (LncBase v2 and miRDB) were used to predict potential target miRNAs of CASC9 (38,39). Computational algorithms (TargetScan 7.1 and miRDB) were used to search for potential miR-758-3p target genes (39,40).

To investigate the role of CASC9 in BC, the expression levels of CASC9 in 49 BC tissues and adjacent normal bladder tissues from patients were first analyzed by RT-qPCR. Compared with paired adjacent normal tissues, 79.6% (39/49) of human BC tissues had upregulated CASC9 expression (Fig. 1A and B; P<0.001). In addition, the expression level of CASC9 in T3/T4 patients was higher than that in T1 patients, which may indicate that CASC9 is related to cell invasion ability (Fig. 1C; P<0.05).

The correlation between CASC9 expression levels and the tumor invasion depth in BC patients was further analyzed. As revealed in Table I, CASC9 upregulation was significantly associated with BC tumor invasion depth (n=49; P<0.05) and age (n=49, P<0.05), however sex and tumor size were not associated with CASC9 expression levels. These results indicated that CASC9 may play a carcinogenic role in BC.

In addition, Kaplan-Meier survival analysis revealed that the disease-free survival (DFS) of patients with high CASC9 expression was significantly decreased compared with patients with low CASC9 expression (Fig. 1D; analysis from GEPIA). Collectively, the present results revealed that CASC9 was upregulated in BC, and the expression level of CASC9 could serve as a predictor of prognosis in BC patients.

As revealed in Fig. 2A, CASC9 expression was significantly upregulated in BC cell lines (T24, TCCSUP, UM-UC-3, J82 and 5637) compared with normal human bladder epithelial cells (SV-HUC-1) (Fig. 2A). Notably, compared with SV-HUC-1 cells, the expression of CASC9 in J82 and 5637 cells was significantly increased by >40-fold (Fig. 2A). Therefore, 5637 and J82 cells were selected for further investigation.

The marked upregulated expression of CASC9 in BC tissues and various BC cell lines prompted us to further explore the role of CASC9 in tumorigenesis. Three siRNAs that specifically targeted CASC9 were first designed, and the knockdown efficiency of the siRNAs was quantified by RT-qPCR. siCASC9-1 and siCASC9-3 were efficient in depleting CASC9 compared with the negative control (NC treatment) or siCASC9-2 (Fig. 2B). Therefore, siCASC9-1 and siCASC9-3 were selected for further experiments. The effect of CASC9 on BC cell proliferation was further assessed using EdU and CCK-8 assays. EdU assay results revealed that CASC9 knockdown (siCASC9-1 or siCASC9-3 transfection) in 5637 and J82 cells significantly attenuated cell proliferation (Fig. 2C and D). Similar results were observed using CCK-8 assays (Fig. 2E). Collectively, the results demonstrated that CASC9 knockdown inhibited BC cell proliferation.

Flow cytometric assays were performed to assess whether CASC9 knockdown promotes BC cell apoptosis. However, no statistically significant differences in apoptotic rates were observed between 5637 cells transfected with siCASC9 and siNC (P>0.05, Fig. 2F and G). Similar results were observed in J82 cells (P>0.05; Fig. 3F and G).

Wound healing and Transwell assays (without Matrigel coating) were utilized to assess the effect of CASC9 on BC cell migration. In wound healing assays, the open wound area of siCASC9 (siCASC9-1 or siCASC9-3)-transfected cells was significantly increased compared with siNC-transfected cells 48 h after scratching (Fig. 3A). Transwell migration assays (without Matrigel) revealed that the number of migrated cells in siCASC9-treated groups was significantly decreased compared with siNC-treated groups (Fig. 3B and C). These results indicated that CASC9 promoted BC cell migration.

The effects of CASC9 on BC cell invasion were assessed using Transwell assays (with Matrigel). The number of invasive cells in the siCASC9 groups was reduced by ~50% in 5637 and J82 cells (Fig. 3B and C). These results indicated that CASC9 promoted BC cell invasion.

Increasing evidence indicates that lncRNAs may act as sponges for miRNAs, thereby regulating their downstream targets (19,21). Bioinformatics tools (LncBase v2 and miRDB) were used to predict potential target miRNAs of CASC9 (38,39). The predicted miRNAs were screened and the candidate to miR-758-3p was narrowed down. The results revealed that miR-758-3p expression levels were significantly decreased in BC cell lines (T24, TCCSUP, UM-UC-3, J82 and 5637) compared with normal human bladder epithelial cells (SV-HUC-1) (Fig. 4A).

Subsequent luciferase reporter assays revealed that miR-758-3p overexpression reduced the luciferase activity of a luciferase reporter harboring wild-type (WT) CASC9 but not the reporter carrying mutant (MT) CASC9 (Fig. 4B). Collectively, these results indicated that CASC9 acted as a sponge for miR-758-3p.

To understand the roles of miR-758-3p in BC, wound healing and Transwell assays were used to assess the effect of miR-758-3p on BC cell migration and invasion. Consistent with the theory that CASC9 functions as an oncogene, it was revealed that miR-758-3p mimics suppressed BC cell migration and invasion, while a miR-758-3p inhibitor promoted BC cell migration and invasion (Fig. 4C and D).

The effect of miR-758-3p on cell proliferation was further evaluated using EdU assays. As revealed in Fig. 5A-C, the proliferation of BC cells transfected with miR-758-3p mimics was significantly inhibited compared to the NC group (P<0.05), while BC cell proliferation in the miR-758-3p inhibitor group was increased compared with the NC group (P<0.05). The rate of BC cell apoptosis was also quantified using flow cytometric assays. However, no statistical significance was observed between any groups of BC cells (P>0.05; Fig. 5E).

An RNA FISH experiment was also performed to determine the cellular localization of CASC9. CASC9 was distributed in both the cytoplasm and nucleus (Fig. 5D), indicating that CASC9 may function in the nucleus and/or cytoplasm.

In order to explore the function of miR-758-3p, computational algorithms (TargetScan and miRDB) were used to search for potential miR-758-3p target genes (Fig. 6B). After knocking down the expression of CASC9, the expression changes of EMT-related genes were detected by the Human EMT RT² Profiler PCR Array (Fig. 6A). A miR-758-3p-binding site was revealed in the 3′-UTR of TGF-β2 (Fig. 6C). The TGF-β pathway plays a central role in inducing EMT in BC (41). To verify whether TGF-β2 is the direct target of miR-758-3p, the TGF-β2 3′-UTR sequence was subcloned into the luciferase reporter vector pSicheck-2. It was revealed that miR-758-3p overexpression suppressed the luciferase activity of the luciferase reporter harboring wild-type (WT) TGF-β2 3′-UTR but not the reporter carrying mutant (MT) TGF-β2 (Fig. 6C).

To assess the effect of CASC9 on TGF-β2 expression, CASC9 was knocked down in 5637 and J82 cells, and it was determined that CASC9 knockdown reduced TGF-β2 mRNA and protein levels (Fig. 6D-F). Subsequently, the protein levels of EMT markers were also determined by western blotting. The results revealed that after knockdown of CASC9, E-cadherin protein levels were significantly increased. After knocking down the expression of CASC9, 5637 and J82 cells exhibited stromal cell morphological characteristics (Fig. 6E), and N-cadherin protein expression was significantly decreased in 5637 and J82 cells (Fig. 6F and G).

Collectively, all of the aforementioned data indicated that CASC9 sponges miR-758-3p to promote EMT in BC by upregulating TGF-β2 (Fig. 7).