Immunohistochemical analysis was performed on samples from 114 patients with urothelial carcinoma, who were treated for bladder cancer by radical cystectomy at Osaka City University Hospital between January 1995 to December 2015. The clinicopathological characteristics of the patients are summarized in Table I. There were 89 male and 25 female patients, and the median age was 69 years (range, 33–84 years). Patients who were either incompletely resected, or histologically confirmed as small cell carcinoma of the bladder, or lost to follow-up, were excluded from the study. Pathologic staging was performed according to the 2009 Tumor, Node, Metastasis classification system (14), and grading was done according to the criteria by the World Health Organization, 2004 (15). The Institutional Review Board at Osaka City University Graduate School of Medicine approved the use of the specimens and clinical data in accordance with the Declaration of Helsinki and guidelines of Osaka City University Graduate School of Medicine (study approval no. 1955). All 114 patients included in the present study provided written informed consent for the collection and use of tissue samples and for the publication of their results.

Tissues obtained by radical cystectomy were fixed with 10% formalin for 24–48 h at room temperature and paraffin-embedded. Paraffin embedding was performed as follow: 100% ethanol (4×1.5 h at 37°C), 100% ethanol (2×2 h at 37°C), 100% ethanol (3 h at 37°C), 100% xylene (3×30 min at 35°C), paraffin wax (2×30 min at 58°C), paraffin wax (30 min at 59°C), paraffin wax (30 min at 60°C) and embedding of the tissues into paraffin blocks. Tissues were cut into 3-µm sections for histological analysis. Tissues were stained with Mayer's hematoxylin solution for 5 min and counterstained in eosin Y ethanol solution for 3–5 min both at room temperature. Formalin-fixed, paraffin-embedded tissues of bladder cancer were analyzed by immunohistochemical staining as described previously (4,16). Sections were incubated with rabbit polyclonal antibody to STS (HPA 002904; 1:100; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), rabbit polyclonal antibody to E-cadherin (sc-7870; 1:100; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) and rabbit monoclonal antibody to vimentin (D21H3; 1:100; Cell Signaling Technology, Inc., Danvers, MA, USA) at 4°C overnight. This was followed by incubation with biotinylated goat anti-rabbit IgG (BA-4000; 1:200; Vector Laboratories, Inc., Burlingame, CA, USA) for 30 min at room temperature. Immunoreactivity was detected using a VECSTAIN Elite ABC kit (PK-6101; Vector Laboratories, Inc.) and 3,3′-diaminobenzidine hydrochloride (Sigma-Aldrich, St Louis, MO, USA).

Immunohistochemical analysis was performed by three pathologists. Immunoreactivity of STS was observed in the cytoplasm of bladder cancer cells, but not in normal urothelium. A benign prostate tissue, which was simultaneously removed at the time of radical cystectomy, was used as a positive control. Tissues with >5% cancer cells immunoreactive for STS were defined as positive (17). Evaluation of staining for E-cadherin and vimentin were performed based on a staining index (18,19).

The human bladder cancer cell lines (5637, HT1376, UMUC3, TCCSUP and T24) and prostate cancer cell lines (LNCaP and PC-3) were purchased from the American Tissue Culture Collection (Manassas, VA, USA). Cells were authenticated by short tandem repeat analysis performed by Takara Bio, Inc. (Otsu, Japan) and tested to ensure that they were mycoplasma-free by DDC Medical (Thermo Fisher Scientific, Inc., Waltham, MA, USA) in November 2017. Cells were maintained as monolayer cultures at 37°C and 5% CO₂. The 5637 cell line was grown in RPMI-1640 (Sigma-Aldrich; Merck KGaA) supplemented with 10% fetal bovine serum (FBS) (Sigma-Aldrich; Merck KGaA), 1% HEPES, and 1% D-Glucose. HT-1376 was grown in Eagle's minimal essential medium (EMEM; Wako Pure Chemical Industries, Ltd., Osaka, Japan) supplemented with 10% FBS, 1% sodium pyruvate solution (Sigma-Aldrich; Merck KGaA) and 1% MEM non-essential amino acid solution (Thermo Fisher Scientific, Inc.). UMUC3 and TCCSUP were grown in EMEM, LNCaP and PC-3 in RPMI 1640, and T24 in McCoy's 5A medium (Thermo Fisher Scientific, Inc.), supplemented with 10% FBS.

Total RNA was extracted from cell lines using RNeasy Mini kit (Qiagen GmbH, Hilden, Germany) according to the manufacturer's instructions. RT-qPCR assay was performed as described previously (4). The catalogue numbers for the primers used for qPCR were as follows: STS, Hs00996676_m1; E-cadherin, Hs01023894_m1; Vimentin, Hs00185584_m1; GAPDH, Hs00266705_g1 (Applied Biosystems; Thermo Fisher Scientific, Inc.). The thermocycling conditions were: 20 sec at 95°C followed by 40 cycles of 3 sec at 95°C and 30 sec at 60°C. Data were then quantified using the 2^−ΔΔCq method for relative gene expression (20), compared to that of GAPDH as internal control.

Western blotting was performed as previously described (4). Primary antibodies for STS (1:1,000) and β-actin (ab8226; 1:1,000; Abcam, Cambridge, UK) were used for the present study. Goat anti-rabbit horseradish peroxidase (HRP)-conjugated IgG and goat anti-mouse HRP-conjugated IgG (nos. sc-2004 and sc-2005; 1:10,000; Santa Cruz Biotechnology) were used as secondary antibodies. Immunoreactive bands were visualized using ECL Prime Western Blotting Detection reagent (GE Healthcare, Chicago, IL, USA).

STS expression was transiently knocked down in T24 cells using Lipofectamine^™ RNAiMAX (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. STS-specific small interfering (si)RNAs (Silencer^® Select siRNAs) were obtained from Thermo Fisher Scientific, Inc. The sense sequences of siRNA for STS were as follows: si-STSA, 5′-CUAGCAACAUGGACAUAUUTT-3′; and si-STS B, 5′-GGACAUAUUUCCUACAGUATT-3′. Non-targeting control siRNA (cat. no. 4390844; Silencer^® Select Negative Control siRNA) was obtained from Thermo Fisher Scientific, Inc. T24 cells (1.5×10⁵) were transiently transfected with 10 nM si-STS A, si-STS B, or control siRNA in a 6-well plate. Following 48 h, cells were trypsinized and used in additional assays.

To investigate the effect of STS knockdown on cell proliferation, transfectants (1×10⁴/well) were seeded in a 96-well plate and grown in McCoy's 5A medium supplemented with 10% FBS at 37°C with 5% CO₂. After 48 h, cell proliferation was measured using a Cell Counting Kit-8 (Dojindo Molecular Technologies, Inc., Kumamoto, Japan) according to the manufacturer's instructions. The number of cells was measured using a microplate reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA) at 450 nm.

Migration assay was performed using a Cell Culture Insert with an 8.0-µm pore size PET filter (BD Biosciences, Franklin Lakes, NJ, USA) and invasion was assessed via a Matrigel invasion assay (BD Biosciences), according to the manufacturer's protocol. Briefly, 2–3×10⁴ cells in 500 µl serum-free McCoy's 5A medium were seeded in the upper chamber, whereas the lower chamber was loaded with medium containing 10% FBS. Following a 24-h incubation at 37°C, the cells that remained inside the inserts were removed with cotton swabs, and those that migrated or invaded the reverse side of the inserts were fixed with 5% glutaraldehyde for 15 min and stained with Giemsa for 30 min both at room temperature. The cells that had migrated or invaded through the membranes were counted by light microscopy (magnification, ×20).

Statistical analyses were performed with GraphPad Prism 7 (GraphPad Software, Inc., La Jolla, CA, USA). Fisher's exact test was used to evaluate the differences in incidence of STS expression patterns among clinical and pathological parameters. The recurrence-free survival was defined as the time between the date of surgery and the last date of follow-up or the date of recurrence. The curves were analyzed using the Kaplan-Meier method with the log-rank test to assess statistical significance. For multiple analyses, Cox proportional hazards analysis was used to determine the relative contribution of various factors to the risk of progression. One-way ANOVA followed by a Sidak's post hoc test was used to assess the difference between the in vitro assays. P<0.05 was considered to indicate a statistically significant difference.

Western blot analysis demonstrated a strong expression of STS in T24 cells, and a weak expression in PC-3 and 5637 cells, compared with those in other cell lines (Fig. 1A). The mRNA expression levels of STS were analyzed by RT-qPCR in three bladder cancer cell lines with invasive characteristics, T24, TCCSUP and UMUC3 cells (Fig. 1B). The expression level of STS was significantly higher in T24 cells than in TCCSUP and UMUC3 cells; therefore, T24 was selected for an in vitro knockdown assay.

Immunohistochemical analysis of STS was performed using 114 formalin-fixed, paraffin-embedded tissues from patients with bladder cancers (Fig. 2). Normal urothelium (Fig. 2A and D), non-invasive bladder cancer (Fig. 2B and E) and invasive bladder cancer (Fig. 2C and F) were compared with benign prostate tissue (Fig. 2G) as a positive control of STS. Immunoreactivity of STS was observed in the cytoplasm of bladder cancer cells in invasive areas of cancer (Fig. 2E), but not in the surface regions, especially in papillary tumors (Fig. 2F). As summarized in Table I, the incidence of STS-positive cancers was significantly higher in muscle invasive bladder cancers (MIBCs; pT2 + pT3 + pT4; 37.1%) than in non-muscle invasive bladder cancers (NMIBCs; pTa + pT1 + pTis; 11.5%). No significant association was demonstrated with age or sex. Immunohistochemical analyses of E-cadherin and vimentin were performed using 10 NMIBC and 10 MIBC tissues to analyze the association between STS and these EMT-related factors (Fig. 3). The expression of E-cadherin was positive in 85% (17/20) of tissues, while vimentin was negative in 90% (18/20). There was no statistical significance in the positivity of E-cadherin and vimentin between NMIBCs and MIBCs. Also, no significant association was observed in the expression between STS and these EMT-related markers (Table II).

Correlation analysis of STS expression with clinical outcomes in 114 patients with bladder cancer revealed worse survival rates in patients with STS-positive cancer. Patients with STS-positive cancer exhibited shorter recurrence-free survival [RFS; Fig. 4A; hazard ratio (HR)=3.037; P=0.0027] and cancer-specific survival (CSS; Fig. 4B; HR=3.209; P=0.003) than those with STS-negative cancers. For patients treated with radical cystectomy, univariate analyses of clinicopathological parameters and RFS or CSS revealed the pathological stage (NMIBC vs. MIBC), lymph node metastasis and STS expression as the major risk factors (Table III). Multivariate analysis demonstrated that stage (MIBC vs. NMIBC) and positive lymph node involvement were independent risk factors for RFS, and stage (MIBC vs. NMIBC) was the only independent risk factor for CSS (Table IV).

STS-specific siRNA reduced STS mRNA expression levels by 86–90% compared with that in the negative control cells (Fig. 5A). Knockdown of STS did not significantly affect cell proliferation (Fig. 5B), but significantly inhibited cell migration (35–80%; Fig. 5C) and invasion (66–73%; Fig. 5D).

Total RNA was extracted from T24 cells transfected with negative control siRNA, si-STS A or si-STS B. RT-qPCR demonstrated that expression of E-cadherin was significantly upregulated (fold change, 6.75–6.77) and that of vimentin was significantly downregulated (fold change, 0.44–0.60) by STS knockdown (Fig. 5E and F).