Human bladder cancer cell lines EJ and UM-UC-3 were obtained from American Type Culture Collection and maintained at 37°C in incubator containing 5% CO₂. Both cell lines were grown in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (all from Gibco; Thermo Fisher Scientific, Inc.). EGCG was obtained from Sigma-Aldrich (Merck KGaA). Epidermal growth factor (EGF), basic fibroblast growth factor (bFGF) and insulin were obtained from Peprotech, Inc., and B27 was obtained from Gibco (Thermo Fisher Scientific, Inc.). MTT and dimethyl sulfoxide (DMSO) were obtained from Sigma-Aldrich (Merck KGaA). Purmorphamine was obtained from Selleckchem; it was dissolved in DMSO to make a stock solution at a concentration of 10 mM. Antibodies against CD44 (cat. no. 15675), CD133 (cat. no. 18470), Oct4 (cat. no. 60242), ALDH1A1 (cat. no. 15910), Nanog (cat. no. 14295), Bcl2 (cat. no. 12789), Bax (cat. no. 50599), cleaved-caspase-3 (cat. no. 25546), caspase-8 (cat. no. 13423), Cyclin D1 (cat. no. 60186), proliferating cell nuclear antigen (PCNA; cat. no. 24036), Shh (cat. no. 20697), Smo (cat. no. 20787) and GAPDH (cat. no. 10494) were obtained from ProteinTech Group, Inc. Antibodies against cleaved-caspase-9 (cat. no. 9509) were obtained from Cell Signaling Technology, Inc. Antibodies against Gli1 (cat. no. DF7523) and Gli2 (cat. no. DF7541) were obtained from Affinity Biosciences, Inc. The primers of CD44, CD133, Oct4, ALDH1A1, Nanog and GAPDH were acquired from BGI Genomics Co., Ltd. The EJ cell line has been authenticated by short tandem repeat profiling.

The bladder EJ and UM-UC-3 cells were cultured in serum-supplemented medium (SSM), which consisted of RPMI 1640 medium with 10% FBS and 1% penicillin/streptomycin. Cells were maintained in the incubator with 5% CO₂ at 37°C for 48 h. In order to achieve tumorsphere formation, bladder cancer cells were seeded into 24-well plates at a density of 5,000 cells per well and were cultured in serum-free medium (SFM) consisting of Dulbecco's modified Eagle's medium: Nutrient Mixture F-12 (DMEM/F12; Gibco; Thermo Fisher Scientific, Inc.) with 20 ng/ml bFGF, 20 ng/ml EGF, 5 µg/ml insulin and 2% B27 for 3 days, when the tumorspheres began to form. Tumorspheres were maintained in the incubator with 5% CO₂ at 37°C for 7 days. Tumorspheres were imaged by an inverted microscope at a magnification of ×100. For investigating the effect of EGCG and purmorphamine on the tumorsphere formation capacity, various concentrations of EGCG (0, 30, 60 or 90 µM) or purmorphamine (1 µM) were added to each well. After 7 days of treatment in the incubator with 5% CO₂ at 37°C, the number of EJ and UM-UC-3 tumorspheres >50 µm in diameter that had formed was photographed and counted with the inverted microscope.

UM-UC-3 and EJ cells were incubated in SSM, and tumorspheres were incubated in SFM in the incubator with 5% CO₂ at 37°C. EGCG- and purmorphamine-treated tumorspheres, and cells were washed with PBS, followed by lysing in Radio Immunoprecipitation assay buffer (Beyotime Institute of Biotechnology). Bicinchoninic Acid Protein assay (Pierce; Thermo Fisher Scientific, Inc.) were utilized to detect the concentration of each protein. Proteins (50 µg per lane) were then separated by SDS-PAGE on a 10% gel and transferred to nitrocellulose membranes (EMD Millipore). After blocking in 5% non-fat dry milk at room temperature for 1 h, the membranes were incubated with primary antibodies against Cyclin D1, Shh, Nanog (all 1:500 dilution), CD44, CD133, Oct4, ALDH1A1, Bcl2, Bax, cleaved-caspase-3, cleaved-caspase-9, caspase-8, PCNA, Smo, Gli1, Gli2 and GAPDH (all 1:1,000 dilution) at 4°C overnight. Then they were incubated with horseradish peroxidase-conjugated goat anti-rabbit (cat. no. S0001) or anti-mouse (cat. no. S0002; both 1:10,000 dilution; Affinity Biosciences) immunoglobin G secondary antibodies for 1 h at room temperature. GAPDH was used as the loading control. The signals were detected with enhanced chemiluminescence detection system (cat. no. WBKLS0500; EMD Millipore). The target bands were quantified using ImageJ version 1.46 software (National Institutes of Health).

Bladder cancer adherent cells were cultured in SSM, which consisted of RPMI-1640 medium with 10% FBS and 1% penicillin/streptomycin, and tumorspheres were cultured in SFM. Bladder cancer adherent cells and tumorspheres were maintained at 37°C in an incubator containing 5% CO₂ for 7 days. Total RNA from the cells was isolated with TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) and transcribed into cDNA using 5X All-In-One RT MasterMix (Applied Biological Materials, Inc.) according to the manufacturers' protocol with 1 µg total RNA. The RT-qPCR analysis was performed using the EvaGreen 2X qPCR MasterMix (Applied Biological Materials, Inc.) and primers of CD44, CD133, Oct4, ALDH1A1, Nanog and GAPDH on the LightCycler^® 96 real-time PCR detection system (Roche Molecular Systems, Inc.). The mRNA expression for each gene was normalized by GAPDH. The amplification reactions were as follows: An initial hold step (95°C for 5 min) and 45 cycles of a two-step PCR (95°C for 15 sec, 54°C for 30 sec and 72°C for 30 sec). Fold changes in the expression of each gene were calculated by the 2^−ΔΔCq method (21). The primer sequences are shown in Table I.

Both bladder cancer cells (1×10⁴/well) were seeded into 96-well plates (Corning Incorporated), which were grown in SSM and incubated with various concentrations of EGCG (0, 30, 60, 90, 120, 150 and 180 µM) in an incubator with 5% CO₂ at 37°C for 48 h. After incubation, 20 µl MTT solution (5 mg/ml in PBS) was added to each well, and the plates were further incubated for 4 h at 37°C. Then, the medium containing MTT was discarded and 150 µl DMSO was added to dissolve the formazan crystals. Absorbance was measured at 490 nm using a microplate reader. All measurements were performed in triplicate.

Bladder cancer adherent cells, which cultured in SSM, and tumorspheres, which cultured in SFM, were maintained at 37°C in an incubator containing 5% CO₂ for 7 days. The cells were collected and then washed twice with ice-cold PBS. Then the cells were fixed in 70% cold ethanol at 4°C overnight. After that, cells were treated with 200 µg/ml PureLinkRNase A, 20 µg/ml propidium iodide (PI) and 0.1% Triton X-100 (all from Invitrogen; Thermo Fisher Scientific, Inc.) at 37°C for 15 min. Next, the cell cycle was analyzed in FACSCalibur^™ (BD Biosciences).

The apoptosis assay was performed using a fluorescein isothiocyanate (FITC)-Annexin V Apoptosis Detection kit purchased from BD Biosciences. EJ and UM-UC-3 tumorspheres, which cultured in SFM, were seeded into 6-well plates at a density of 1×10⁶/well and treated with various concentrations of EGCG (0, 30, 60 and 90 µM) for 7 days. Cells were centrifuged at 1,000 × g for 5 min at 4°C and washed twice with PBS. Then the cells were resuspended in 100 µl cold binding buffer, and incubated with 5 µl FITC-Annexin-V and 5 µl PI. After 1 min at room temperature in dark, the apoptotic cells were detected by flow cytometry within 1 h. The data were analyzed using FlowJo version 10.0.7 software (FlowJo LLC).

EJ and UM-UC-3 tumorspheres, which cultured in SFM, were seeded into 6-well plates at a density of 1×10⁶/well and treated with various concentrations of EGCG (0, 30, 60 and 90 µM) for 7 days. The bladder tumorspheres were washed in PBS Tween-20 (PBST), fixed with 4% paraformaldehyde for 15 min at room temperature, and then were washed three times with PBST. After that, plasma membranes were ruptured with 0.1% Triton X-100 (Sigma-Aldrich; Merck KGaA) for 30 min. The tumorspheres were then blocked with 5% bovine serum albumin (Sigma-Aldrich; Merck KGaA) for 1 h at room temperature and stained with rabbit anti-CD133 antibodies (1:100 dilution) at 4°C overnight. They were then stained with FITC-conjugated goat anti-rabbit antibodies (cat. no. S0008; 1:200 dilution; Affinity Biosciences, Inc.) for 2 h at room temperature. Next, 4′,6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich; Merck KGaA) was used to stain the nuclei for 15 min at room temperature and mounted in aqueous mounting medium (cat. no. ab128982; Abcam). The fluorescent images were imaged using a fluorescence microscope at a magnification of ×100.

EJ and UM-UC-3 cells were seeded into 6-well plates at a density of 2×10⁵ cells in RPMI-1640 medium containing 10% FBS without antibiotics. After 12 h of incubation at 37°C, cells were then transiently transfected with Gli1-overexpressing vectors (EX-NEG-M29-Gli1) or control vectors (EX-NEG-M29; both 2 µg; GeneCopoeia, Inc.) with Lipofectamine Reagent 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocols. After 12 h of transfection, cells were trypsinized and seeded into 24-well plates at a density of 5,000 cells per well containing SFM. Another 12 h later, cells were treated with or without 60 µM EGCG for 4 days. Finally, the western blot analysis was used to detect the protein expression levels.

Data are presented as mean ± standard deviation. Student's t-test was used for comparison between two groups. One-way analysis of variance with Dunnett post hoc test and Tukey test was used for comparison between multiple groups. Statistical tests were performed using Graph-Pad Prism 7.0 software (GraphPad Software, Inc.). P<0.05 indicated that the difference between groups was statistically significant.

SFM culture is commonly performed to enrich CSCs properties, such as the capability of forming tumorspheres in vitro (22,23). The expression of bladder CSCs markers were examined including CD44, CD133, Oct4, ALDH1A1 and Nanog. As shown in Fig. 1A, bladder adherent cancer cells were grown in SSM and tumorspheres were formed in EJ and UM-UC-3 bladder cancer cells cultured for 7 days in SFM. The protein and mRNA expression levels of bladder CSCs markers were significantly higher in cells grown in SFM compared with those cultured in SSM (Fig. 1B and C). Increasingly, reports have indicated that CSCs were more likely to be quiescent than cancer cells (24,25). Cell cycle analysis showed that the percentage of bladder CSCs in the G1 phase of the cell cycle was greater than that of cells in the S phase (Fig. 1D). It also showed that the percentage of tumorspheres, which were cultured in SFM, in the G1 phase was significantly greater than that of bladder adherent cells, which were cultured in SSM. Compared with adherent cells, the percentage of tumorspheres in the G1 phase greater by ~20%, and the percentage of tumorspheres in the S phase was significantly lower by 17%. This result is consistent with the theory of CSCs. CSCs have been shown to always have a slower growth rate compared with cancer cells as they are resistant to anticancer drugs (26). These results suggested that EJ and UM-UC-3 tumorsphere-forming cells cultured in SFM showed the characteristics of bladder CSCs.

To investigate the effect of EGCG on bladder cancer, cells were treated with EGCG for 48 h, which inhibited cell proliferation in what appeared to be a dose-dependent manner (Fig. 2A). EGCG inhibited the growth of EJ and UM-UC-3 cells with an IC₅₀ of 100.9 and 84.5 µM at 48 h, respectively. Treatment of EJ and UM-UC-3 tumorspheres with increasing concentrations of EGCG (0, 30, 60 and 90 µM) decreased the size and number of bladder tumorspheres in what appeared to be a dose-dependent manner (Fig. 2B and C). The number of tumorspheres in cells treated with 30, 60 and 90 µM EGCG was significantly lower compared with untreated cells. Western blot analysis showed that the protein expression of bladder CSCs markers was significantly downregulated in the tumorspheres of both cell lines compared with untreated cells (Fig. 2D). In EJ cells, CD44, Oct4 and Nanog were significantly downregulated at 60 and 90 µM EGCG, while CD133 and ALDH1A1 were significantly downregulated at 30, 60 and 90 µM EGCG compared with untreated cells. In UM-UC-3 cells, CD44 and ALDH1A1 were significantly downregulated at 60 and 90 µM EGCG, CD133 and Nanog at 30, 60 and 90 µM EGCG, and Oct4 at 90 µM EGCG compared with untreated cells. Immunofluorescent staining showed the seemingly dose-dependent decrease of CD133-positive sphere-forming cells (Fig. 2E). These results indicateed that EGCG inhibited bladder CSCs properties in EJ and UM-UC-3 tumorspheres.

Next, the authors examined the effects of EGCG on cell apoptosis and protein expression in bladder CSCs. As shown in Fig. 3A, EGCG treatment significantly downregulated Bcl-2, and significantly upregulated Bax, caspase-8, cleaved caspase-9 and cleaved caspase-3 compared with untreated cells. In EJ cells, Bcl-2 was significantly downregulated at 30, 60 and 90 µM EGCG. In these cells, Bax and cleaved caspase-3 were significantly upregulated at 30, 60 and 90 µM EGCG, while caspase-8 was significantly upregulated at 90 µM EGCG and cleaved caspase-9 was significantly upregulated at 60 and 90 µM EGCG. In UM-UC-3 cells, Bcl-2 was significantly downregulated at 90 µM EGCG, while Bax, caspase-8, cleaved caspase-9 and cleaved caspase-3 were significantly upregulated at 30, 60 and 90 µM EGCG.

EGCG treatment also downregulated the expression of cell proliferation-related proteins PCNA and Cyclin D1 (17) (Fig. 3B). In UM-UC-3 cells, PCNA was significantly downregulated at 90 µM EGCG, while Cyclin D1 was significantly downregulated at 30, 60 and 90 µM EGCG compared with untreated cells. In EJ cells, PCNA and Cyclin D1 were significantly downregulated at 60 and 90 µM EGCG, while only PCNA was significantly downregulated at 30 µM EGCG compared with untreated cells. Meanwhile, flow cytometry analysis also showed that EGCG induced apoptosis of sphere-forming cells in both EJ and UM-UC-3 cell lines as the rates of cell apoptosis increased with increasing EGCG doses in what appeared to be a dose-dependent manner (Fig. 3C). These results suggested that EGCG decreased cell proliferation and induced apoptosis in bladder CSCs.

To investigate whether EGCG could modulate the sonic hedgehog signaling pathway on bladder CSCs, the expression of Shh-related markers after the treatment of EGCG was analysed. Western blot analysis showed that 30, 60 and 90 µM EGCG significantly decreased the expression levels of Smo, Shh, Gli1 and Gli2 compared with untreated cells (Fig. 3D). However, no significant differences were observed in Smo expression in EJ cells and Shh expression in UM-UC-3 cells treated with 30 µm EGCG.

To further investigate the involvement of the Shh pathway in the effect of EGCG, cells were co-treated with purmorphamine, a specific activator of the Shh pathway (27), and EGCG. As shown in Fig. 4A and 4B, purmorphamine treatment promoted tumorsphere formation and abolished the effect of EGCG on bladder CSCs. EGCG significantly decreased and purmorphamine significantly increased tumorsphere formation compared with untreated cells (Fig. 4B). Tumorsphere formation in EGCG and purmorphamine co-treated cells was similar to untreated cells, and significantly increased when compared with EGCG-treated cells. These results were confirmed by western blot analysis, which showed that the effects of EGCG on the expression of bladder CSCs protein markers (Fig. 4C), cell proliferation-related proteins (Fig. 4D) and apoptosis-related proteins (Fig. 4E) were significantly reversed by purmorphamine-induced Shh signaling pathway activation compared with EGCG-treated cells. In addition, purmorphamine significantly promoted the activation of Shh signaling pathway (Fig. 4F).

To confirm the involvement of the Shh signaling pathway, cells were transfected with Gli1-overexpressing plasmids. Overexpression of Gli1 significantly reversed the EGCG-induced downregulation of bladder CSCs markers compared with EGCG-treated cells (Fig. 4G). However, Gli1 expression in EJ cells treated with EGCG and EX-NEG-M29-Gli1 was not significantly different from EJ cells treated with EGCG alone. These results suggested that EGCG inhibited bladder CSCs by suppressing the activity of the Shh signaling pathway.