This investigation was conducted in accordance with ethical standards, the Declaration of Helsinki, and national and international guidelines. Ethical approval for using human material in the present study was obtained from the Internal Review Board of the University Hospital of Bonn (IRB nos. 036/08 and 093/12). The participants of the study agreed in writing beforehand and were anonymised before inclusion in the study cohort.

The database cBioPortal for cancer genomics (http://cbioportal.org; version 1.4.0) was utilized in January 2017 to investigate the mRNA expression of MED15 in BCa extended by the use of survival analysis.

cBioPortal provides a Web resource for exploring, visualising, and analysing multidimensional cancer genomics data. The portal reduces molecular profiling data from cancer tissues and cell lines into readily understandable genetic, epigenetic, gene expression and proteomic events (16,17). For the analysis, RNA sequencing data from The Cancer Genome Atlas (TCGA) (18) were used (RNA Seq V2 RSEM) with two SDs from the mean and a threshold of ±2.0. The mRNA z-scores were precomputed from the expression values and were compared to the expression distribution of each gene associated with the tumors that are diploid for this gene (16). Two mRNA analyses for BCa, named ‘Bladder Urothelial Carcinoma (TCGA, Provisional)’ and ‘Bladder Urothelial Carcinoma (TCGA, Nature 2014)’ are available on cBioPortal with 408 and 129 samples, respectively.

Additionally, a mutation analysis was performed in both data sets using cBioPortal [Copy number alterations, mRNA expression (RNA Seq V RSEM)].

A total of 182 specimens were used for this study. The urothelial BCa cohort, provided by the Clinic for Urology of the University Hospital Bonn, contained a total of 18 benign samples, 126 BCa samples, and 38 metastases derived from BCa (Table I).

TMAs were performed as described previously (19,20). Briefly, formalin-fixed paraffin-embedded (FFPE) tissues were cut into 4-µm-thick sections and mounted on slides. After staining with haematoxylin and eosin (H&E), relevant areas of benign tissue and primary tumor were identified and circled by a pathologist. Each tumor and the corresponding benign region were represented with up to three cores measuring 0.6 mm in diameter on a TMA recipient block using a semiautomatic tissue arrayer (Beecher Instruments, Sun Prairie, WI, USA). H&E TMA sections were assessed again to confirm the histology.

Prior to performing IHC analyses in the selected TMAs, the specificity of the antibody was confirmed according to the manufacturer's instructions by using non-seminomatous germ cell tumor (NSGCT) as a positive control. IHC was performed using the Ventana Benchmark automated staining system (Ventana Medical System, Tuscon, AZ, USA). In short, slides were incubated with the primary antibody according to the manufacturer's instructions: Anti-MED15 rabbit polyclonal (1:50, clone 11566-1-AP; ProteinTech Group, Inc., Chicago, IL, USA) at room temperature; antibody dilution was conducted using a Ventana diluent. For signal detection, the ultraView Universal DAB detection kit (Ventana Medical System) was used. Finally, slides were counterstained with haematoxylin and bluing reagent, dehydrated, and mounted.

After IHC was performed, the staining was evaluated independently by two pathologists (GK and TM). Only cases with at least one assessable TMA core that had sufficient tumor tissue were included in the analysis. Quantification of nuclear protein expression was evaluated according to a score of 0 (=no expression) to 3 (=high expression). The cut-off used to determine a nuclear MED15 overexpression (≥2) was consistently set above the expression of benign samples.

Urothelial BCa cell lines T24 and TCCSUP were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA) and were grown in a 5% CO₂ incubator at 37°C and 85% humidity. Monolayer cultures were maintained in the RPMI-1640 medium (Biochrom AG, Berlin, Germany) that was supplemented with 10% heat-inactivated foetal calf serum (FCS; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), 1% streptomycin-penicillin antibiotics (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA), and 1% glutamine (Thermo Fisher Scientific, Inc., Darmstadt, Germany).

The cell transfections in the cell lines T24 and TCCSUP were carried out with 100 nmol/l siRNA using Screenfect A (Genaxxon Bioscience GmbH, Ulm, Germany) for 48 (for PCR) and 72 h (for western blotting), respectively. MED15 siRNA (sc-75767; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) is a pool of three target-specific siRNAs designed to knockdown gene expression. A non-targeting scrambled siRNA was used as negative control (Exiqon Life Sciences, Copenhagen, Denmark).

Post-transfected cells were washed with ice-cold phosphate-buffered saline (PBS) and lysed in a protein extraction buffer for 30 min. In addition, samples were sonicated for 2 min and subsequently centrifuged at 16,000 × g for 30 min at 4°C. Protein concentration was measured using bicinchoninic acid (BCA) Protein Assay kit (Thermo Fisher Scientific, Inc., Waltham, MA, USA), fractionated by SDS-PAGE and transferred to a nitrocellulose membrane. After incubation with 5% non-fat milk in TBST (50 mM Tris, 150 mM NaCl, 0.05% Tween-20, pH 7.5) for 60 min, the membrane was incubated with anti-MED15 rabbit polyclonal (1:500, clone 11566-1-AP; ProteinTech Group, Inc.) or anti-GAPDH rabbit monoclonal (1:2,000, clone 14C10; Cell Signaling Technology, Inc., Danvers, MA, USA) primary antibodies at 4°C overnight. Then, the membrane is washed three times for 5 min with TBST and incubated with horseradish peroxidase conjugated anti-rabbit antibody for 1 h at room temperature. Staining was detected using SuperSignal West Femto Maximum Sensitivity Substrate, an ultra-sensitive enhanced chemiluminescence (ECL) substrate (Thermo Fisher Scientific, Inc.).

RNA was isolated from cell line pellets using the Total RNA Purification Mini Spin Column Kit (Genaxxon Bioscience GmbH, Ulm, Germany). RNA quantity and quality were analysed using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Inc., Wilmington, DE, USA). cDNA was synthesised using 200 ng total RNA and the PrimeScript RT Reagent kit with gDNA Eraser (Takara Bio, Saint-Germain-en-Laye, France). qPCR was performed using 5 ng/µl cDNA, Takara Bio SYBR Premix Ex Taq II with ROX Plus, and 10 pmol/µl forward and reverse primer. The following primer sequences were used: MED15, forward: 5′-TTGAGGATGATGAGCGGCAG-3′, reverse: 5′-GGAGGTCCTTGTCATCCAGC-3′; and β-actin, Forward: 5′-CCAACCGCGAGAAGATGA-3′, reverse: 5′-CCAGAGGCGTACAGGGATAG-3′. PCR was performed on an ABIPrism 7900 HT Fast Real-Time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.). Data were analysed using Qbase+ (Biogazelle, Ghent, Belgium) with β-actin as reference gene, applying the 2^−∆∆Cq algorithm.

The EZ4U cell proliferation assay kit was used according to the manufacturer's protocol (EZ4U; Biomedica Group, Vienna, Austria). The siRNA transfections for proliferation assays were performed in 96-well plates. In each well of a flat-bottom 96-well plate, either 1.2×10⁴ cells (T24) or 2.0×10⁴ cells (TCCSUP) were seeded in 200 µl cell culture medium. siRNA-mediated MED15 knockdown was then performed, and cells were incubated to adhere and grow for 72 h. After incubation, 20 µl of EZ4U substrate solutions were added before incubating the resultant solution for 3 h (T24) and 4 h (TCCSUP) until the color of the solution changed from yellow to orange. The absorbance was measured using a microplate reader (Tecan Spectra Rainbow microplate reader; Tecan Deutschland GmbH, Crailsheim, Germany) at 450 nm wavelength. Each experiment was repeated at least three times.

The siRNA transfections for migration assays were performed in 6-well plates. 48 h post transfection, cells were trypsinised and seeded into migration Boyden chambers. 3×10⁴ (T24) or 5×10⁴ cells (TCCSUP) were plated in the upper chamber of migration inserts (VWR, Darmstadt, Germany) containing 0% FCS medium. The lower chamber was filled with medium containing 10% FCS for chemotactic attraction. After 24 h of incubation, cells on the upper surface were removed with the help of a cotton swab. BCa cells invading the lower surface of membrane were fixed with 4% paraformaldehyde (Merck KGaA, Darmstadt, Germany), stained with haematoxylin (Waldeck, Münster, Germany), and washed with water. Membranes were scanned and manually evaluated in four randomly selected fields by counting. Each experiment was repeated at least three times.

The invasion analysis was carried out in a similar as the migration except for the use of Matrigel invasion chambers (VWR) and a cell number of 7.5×10⁴ per well.

For IHC, associations with clinical-pathological parameters were performed. Survival analysis was evaluated with the help of the Kaplan-Meier estimator and log rank tests. Statistical evaluation was performed in Microsoft Excel and SPSS. The IHC data were not normal distributed, so we performed a non-parametric test. When comparing more than two sets of numerical data, a multiple group comparison test such as Kruskal-Wallis test was followed by a post hoc test such as Dunn-Bonferroni test. When comparing two or less sets of numerical data, the Mann-Whitney U test was applied. The data were analysed anonymously.

For the statistical analysis of the functional investigations in the cell lines, we compared the results of scrambled RNA and target RNA on the one hand in TCCSUP and on the other hand in the T24. The respective cell line had its own scrambled control. The respective ‘scrambled’ control was set to 1. Due to the fact that the data were distributed normally, the parametric test was performed and because of two and less groups the unpaired t-test was used.

The cell lines with each other have not been compared statistically.

Using the cBioPortal database, we analysed the mRNA expression of MED15 in BCa samples. In the two analyses, 408 and 129 samples of BCa were investigated. The results of the expression analyses are shown in Fig. 1.

In both analyses, named ‘Bladder Urothelial Carcinoma (TCGA, Provisional)’ and ‘Bladder Urothelial Carcinoma (TCGA, Nature 2014)’, an alteration of MED15 is described in 2.9% (12/408) and 2.3% (3/129), respectively. All of the alterations are upregulations of MED15 mRNA expression (Fig. 1).

Due to the small number of alterations and events in both the analyses, there are no significant differences in overall- and progression-free-survival. Additionally, the analyses cannot really determine a tendency (data not shown).

As shown in Fig. 2, in the TCGA 2014 analysis (n=130), the following mutation profiles were found: Shallow deletions (n=38), diploid (n=51), gain-of-function (n=34) and amplification (n=4). However, the amplification logically led to the highest mRNA expression. Missense mutations are found four times; no in-frame mutation.

The TCGA Provisional analysis (n=130) shows a similar profile: Deep deletion (n=1), shallow deletion (n=39), diploid (n=49), gain of function (n=34), amplification (n=4). Again, the amplification leads to the highest mRNA expression. This analysis revealed three missense and one in-frame mutations.

In order to further illuminate the expression and role of MED15 in BCa, we performed IHC staining on a TMA cohort with the available clinical information and functional analysis in cell lines of essential tumor parameters.

IHC staining shows MED15 nuclear expression in benign, tumor and metastasis tissue (Fig. 3). A MED15 cytoplasmic expression seemed not to be specific, since almost every sample shows a weak cytoplasmic signal. There are no differences here. The expression of MED15 in BCa was just found in epithelial cells. In the stroma, only a few fibroblasts displayed a weak non-specific expression, comparable to the cytoplasm. For IHC, we found MED15 to have a higher expression in NMIBC (staining score=2.0) compared to benign (staining score=1.5, P=0.028) and MIBC (staining score=1.0, P≤0.001; Fig. 4A and B). The overexpression of MED15 in BCa did not associate with positive lymphnode- or distant-metastasis stage [N: P=0.252 (data not shown), M: P=0.434; Fig. 4C] or the grading status of the tumor (Fig. 4D). Unfortunately, no significant differences between samples with or without overexpression (staining score >2) of MED15 have been found for CSS (P=0.770) and progression-free survival (PFS, P=0.503) (data not shown) for survival analysis of the IHC. An evaluation of the pure MED15 expression (division into positive and negative, without definition of an intensity or overexpression) on the basis of the clinicopathological data can be found in Table II.

To investigate the functional role of the subunit MED15 in malignant progression and metastasis, we performed in vitro siRNA-mediated knockdown of MED15 in the BCa cell lines T24 (P≤0.001) and TCCSUP (P≤0.001; Fig. 5A-C). Thereafter, we compared the measured proliferative activity of scrambled control and MED15 knockdown cells and observed a significant decrease in T24 (P=≤0.001) and TCCSUP (P≤0.001; Fig. 5D). Further, migration and invasion were significantly reduced in T24 (migration P=0.003, invasion P≤0.001) and TCCSUP (migration P≤0.001, invasion P≤0.001) cells following the transient MED15 knockdown, compared to scrambled control cells (Fig. 5E-H). In conclusion, MED15 knockdown led to reduced malignant behaviour in the BCa cell lines T24 and TCCSUP.