The normal human primary bladder epithelial BdEC (ATCC^® PCS-420-010™) cell line was obtained from the ATCC (American Type Culture Collection, Manassas, VA, USA), and the human bladder transition cell carcinoma cell lines SW1738 and JMSU1 were kind gifts from CTY (Taipei Medical University - Shuang Ho Hospital, New Taipei City, Taiwan) and cultured in RPMI 1640 (Thermo Fisher Scientific Inc., Bartlesville, OK, USA). Culture medium was supplemented with 10% fetal bovine serum (FBS, #26140079, Thermo Fisher Scientific Inc., Bartlesville, OK, USA) and 100 U/ml of penicillin–streptomycin (Thermo Fisher Scientific Inc., Bartlesville, OK, USA). All cells used in the study were not greater than passage number 3 (≤P.3). Cells were subcultured at ≥98% confluence or culture media changed every 48 h. Stock solutions of 100 mM in 0.01% DMSO were stored at -20°C, until use.

Monoclonal antibodies against MED10 (C-2: #sc-393450), BAX (B-9: #sc-7480), BCL-xL (H-5: #sc-8392), Ki67 (#sc-23900), Vimentin (V9: #sc-6260), and SNAI1 (G-7: #sc-271977) were obtained from Santa Cruz Biotechnology (Dallas, TX, USA); OCT4A (#2840) and LIN28A (#8706) were purchased from Cell Signaling Technology, Inc. (CST, Beverly, MA, USA); and GAPDH (#sc-32233) was from Santa Cruz Biotechnology (Santa Cruz, CA, USA).

For cell proliferation, Invitrogen alamarBlue™ high-sensitivity cell viability reagent (#A50100, Thermo Fisher Scientific Inc., Bartlesville, OK, USA) was used strictly following the manufacturer’s instruction. Briefly, after 1 × 10³ wild-type (WT) or MED10-silenced (shMED10) SW1738 or JMSU1 cell lines were seeded per well in triplicates with three biological replica for each assay in 96-well microtiter plates containing supplemented growth media and incubated at 37°C in humidified 5% CO₂. After 24 h, the cells were incubated with alamarBlue™ for 2 h at 37°C. The number of dye-stained viable proliferating cells was read at 570-nm absorbance wavelength in the Molecular Devices SpectraMax M3 multimode microplate reader (Molecular Devices LLC., San Jose, CA, USA).

MED10-Human, 4 unique 29mer shRNA constructs in lentiviral GFP vector containing pGFP-C-shLenti (#TL303299; MED10 Human shRNA Plasmid Kit (Locus ID 84246; OriGene Technologies Inc., Rockville, MD, USA) were packaged and transfected into SW1738 or JMSU1 cells to silence MED10. Non-effective 29-mer scrambled shRNA cassette in pGFP-C-shLenti Vector, TR30021, served as negative control. Stably transfected monoclonal SW1738 or JMSU1 cells were selected using 2 μg/ml puromycin, as recommended by the manufacturer. The effect of the lentiviral infection was enhanced by adding Sigma-Aldrich^® polybrene (#TR-1003, Merck KGaA, Darmstadt, Germany). MED10 knockdown in the cells was verified by Western blotting and quantitative real-time PCR. The shRNA sequences for MED10 are as follows: shMED10#1 5′-GACAGCAGCTTCATGATATTA-3′, and shMED101#2 5′-ATCGACACCATGAAGAAATTT-3′.

We overexpressed MED10 in BdEC cells by transfecting the human MED10 (NM_032286.2) cDNA sequence cloned into pCMV6-Entry vector (pCMV-MED10; #V0529, GeneCopoeia, Inc., Rockville, MD, USA) using Lipofectamine™ LTX with PLUS™ reagent (#15338100, Life Technologies, Thermo Fisher Scientific Inc., NY, USA). BdEC cells were seeded and cultured in 35-mm-diameter dishes till they attained 60% confluence. On transfection day, 1 mg of DNA diluted in 100 μl of serum-free medium and 6 μl of Lipofectamine™ LTX with PLUS™ reagents were then added. The DNA-PLUS mix was incubated at room temperature for 20 min, and 4 μl of Lipofectamine reagent was added, followed by an additional 20-min incubation. After incubation, the BdEC cells were carefully washed twice with serum-free media and 800 μl of serum-free transfection medium. The cells were then incubated with DNA-PLUS–Lipofectamine reagent mix in 5% humidified CO₂ incubator at 37°C for 3 h. Thereafter, recovery medium with 10% FBS was added to a final volume of 2 ml and incubated overnight. After this, the recovery medium was suctioned and fresh RPMI 1640 medium containing serum and antibiotics was added.

A total of 79 archived tissue specimens diagnosed with bladder urothelial carcinoma (UC) were retrieved from the Department of Pathology at Taipei Medical University–Shuang Ho Hospital. These samples were originally collected for routine clinical diagnostic purposes and subsequently preserved in the hospital’s pathology archives. As the research was retrospective and involved the analysis of previously collected, fully de−identified specimens, the Institutional Review Board (IRB) of Taipei Medical University reviewed the protocol. It addressed a related non−compliance (NC) event. Upon review, the IRB concluded that the non−compliance of this study was appropriately managed and that corrective measures were implemented in accordance with ethical and regulatory standards governing human subject research.

Immunohistochemical (IHC) staining was performed on formalin-fixed paraffin-embedded (FFPE) samples from our BLCA/UC cohort (n = 79) consisting benign (n = 21), T1 (n = 11), T2 (n = 17), T3 (n = 9), T4 (n = 5), and metastatic (M1, n = 16) bladder UC cases. Samples were probed with primary antibodies against MED10, MKI67/Ki67, OCT4, and LIN28A at 1:200 dilution following standard IHC protocol. Protein expression was scored by two independent pathologists using the quick-score (Q-score) formula Q = I × P, where I is staining intensity [0 (no staining), 1+ (weak), 2+ (moderate), and 3+ (strong)] and P represents percentage of stained cells. Maximum Q-score = 300. For immunofluorescence (IFC) staining, WT or shMED10 tumorspheres derived from corresponding SW1738 and JMSU1 cells were fixed with 4% paraformaldehyde after they were plated onto poly-L-lysine-coated glass coverslips, then they were washed carefully with cold PBS thrice, permeabilized with 0.1% Triton X-100/PBS solution for 10 min, and then incubated with primary antibodies against OCT4 and LIN28A at 1:400 dilution, followed by Cy5-labeled goat anti-mouse Alexa Fluor 488 secondary antibodies (#R37120, Thermo Fisher Scientific Inc.) for 1 h. DAPI (4′,6-diamidino-2-phenylindole; #D1306, Molecular Probes, Thermo Fisher Scientific Inc.) was used for nuclear staining. Cell visualization and imaging were performed using the Nikon E800 fluorescent microscope (Nikon Instruments Inc., Melville, NY, USA).

Protein blots derived from 20 µg of WT or shMED10 UC cell protein samples and separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) were transferred onto polyvinylidene fluoride (PVDF) membranes using the Bio-Rad Mini-Protein electro-transfer system (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The PVDF membranes were blocked for 1 h with 5% non-fat milk in Tris-buffered saline with Tween 20 (TBST) and then probed overnight at 4°C with primary monoclonal antibodies against MED10 (1:1,000, Santa Cruz), BAX (1:1,000, Santa Cruz), BCL-xL (1:1,000, Santa Cruz), MKI67/Ki67 (1:1,000, Santa Cruz), Vimentin (1:1,000, Santa Cruz), SNAI1 (1:1,000, Santa Cruz), OCT4 (1:1,000, Cell Signaling Technology), LIN28A (1:1,000, Cell Signaling Technology), and GAPDH (1:1,000, Santa Cruz). Thereafter, the membranes were incubated with secondary antibodies conjugated with horseradish peroxidase (HRP) at room temperature for 1 h and carefully washed thrice with cold 1× PBS. The protein bands were detected with the enhanced chemiluminescence detection system (Thermo Fisher Scientific Inc., Waltham, MA, USA), and protein band densitometry was done using ImageJ software version 1.49 (https://imagej.nih.gov/ij/).

5 × 10⁴ WT or shMED10 SW1738 and JMSU1 cells were seeded per well in ultra-low attachment 6-well plates (Corning, Corning, NY, USA) containing RPMI 1640 supplemented with Gibco™ B-27TM supplement (#17504044, Invitrogen, Carlsbad, CA, USA), 20 ng/ml basic fibroblast growth factor (bFGF; #13256029, Invitrogen), and 20 ng/ml epidermal growth factor (EGF; #PHG0311, Invitrogen). Cells were cultured at 37°C in a humidified 5% CO₂ incubator for 6 days. Cultivated primary tumorspheres ≥ 100 µm were counted under inverted phase-contrast microscope, and then secondary tumorspheres were generated from the primary tumorspheres by dissociating them and reseeding the dissociated cells as per the primary tumorspheres, from single-cell suspension acquired using a sterile 22-G needle.

We used the scratch wound-healing assay to assess cell migration. Briefly, WT or shMED10 bladder UC cells were seeded and allowed to grow in 6-well plates (Corning, Corning, NY, USA) containing complete growth media with 10% FBS. Media in wells were changed to low-serum (1% FBS) growth media when cells attained >98% confluence. The median axes of the monolayered adherent cells were scratched using sterile yellow pipette tips and carefully washed with low-serum media to rid detached cells. Cell migration based on scratch-wound healing was monitored over time, and images were captured at 0 and 12 h after denudation under a light microscope using a ×10 objective lens. Thereafter, the images were analyzed using the National Institutes of Health ImageJ software version 1.49 (https://imagej.nih.gov/ij/).

Invasion assay was performed using the Corning^® BioCoat™ Matrigel^® invasion chambers with a 8.0−μm PET membrane in a two-24-well plate system (#354480, Corning, Corning, NY, USA). 1 × 10⁵ WT or shMED10 JMSU1 cells were seeded per well in plates and incubated at 4°C overnight. The upper chambers contained low-serum (2% FBS) media while the lower chamber contained 600 ;μl high-serum (20% FCS) media. After 48-h incubation, the non-invaded cells in the upper chamber were carefully wiped off with sterile cotton swabs, while the invaded cells that penetrated through the membrane were fixed with ethanol, stained with crystal violet solution, and counted under light microscope from six random fields of vision.

2 × 10⁴ WT or shMED10 SW1738 and JMSU1 cells were seeded into 6-well culture plates (Corning Inc., Corning, NY, USA) and incubated for 13–15 days at 37°C. Cells were then washed three times with cold 1× PBS, fixed with ice-cold methanol, stained with 0.005% crystal violet, washed with 1× PBS, and dried at room temperature. The colonies formed were assessed and counted under microscope, as well as digitally using the National Institutes of Health ImageJ software version 1.49 (https://imagej.nih.gov/ij/). In each well, the colonies with diameter ≥ 100 μm were counted over six randomly selected fields in triplicate assays.

All data represent the mean ± standard deviation (SD) of assays performed at least 3 times in triplicates. The 2-sided Student’s t test was used for comparison between 2 groups, whereas one-way ANOVA with Tukey’s post-hoc test was used for comparison between 3 or more groups. Kaplan–Meier survival analyses aided comparison of survival rates between the control and test groups. All statistical analyses were performed using MS Excel (Microsoft Corporation. Microsoft Excel [Internet]. 2018. Available from: https://office.microsoft.com/excel) and GraphPad Prism version 8.0.0 for Windows (GraphPad Software, La Jolla, CA, USA). p-value < 0.05 was considered statistically significant.

Bioinformatics-aided analyses of the GPL570 platform [HG-U133_Plus_2] Affymetrix Human Genome U133 Plus 2.0 Array (n = 162,085) (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GPL570) revealed that the expression of MED10 transcripts was significantly higher in the BLCA patients than in their normal peers (1.66-fold, p = 0.002) ( Figure 1A ). This was corroborated by Cox PH modeling-based volcano plot visualization of differentially expressed genes in the TCGA BLCA cohort (n = 412), showing that the overexpression of the med10 gene was associated with increased hazard ratio (3.53-fold, p < 0.001) ( Figure 1B ). Survival analyses of the TCGA BLCA cohort showed that compared with their high MED10 counterparts (n = 333), patients with low MED10 expression (n = 56) enjoyed an 18% to 27% survival advantage from days 1,000 to 5,000 (concordance index = 48.67; hazard ratio HR = 1.78 (95% CI: 1.06–2.97), p = 0.028) ( Figure 1C ). Using the GPL570 platform [HG-U133_Plus_2], we also demonstrated that increased MED10 expression positively correlated with increased pathological tumor (pT) stage ( Figure 1D ). Our hazard ratio forest plot of BLCA/UC-relevant series in the GPL570 platform showed that high MED10 expression favors increased likelihood of disease-specific death (HR = 1.16 (95% CI: 0.85–1.58) and recurrence (HR = 1.07 (95% CI: 0.82–1.39) ( Figure 1E ). More so, MED10 expression was elevated in patients with high-grade BLCA compared to the low-grade peers, albeit statistically insignificant ( Figure 1F ). These data indicate that MED10 is aberrantly expressed in patients with BLCA and that this has an adverse prognostic implication.

Having demonstrated the aberrant expression of MED10 in patients with BLCA and its association with poor prognosis, we sort to gain further insight into the probable role of MED10 in disease progression and therapy response. Reanalyzing the TCGA BLCA cohort data, we demonstrated that compared to patients without metastasis (M0), MED10 expression is upregulated in metastatic disease (M1) ( Figure 2A ). Similarly, MED10 expression is higher in patients with recurred or progressed disease, compared with their disease-free counterparts ( Figure 2B ). Also, MED10 transcript expression is positively correlated with increased American Joint Committee on Cancer (AJCC) tumor stage (T1 < T2 < T3 < T4) ( Figure 2C ). Interestingly, we also observed a higher median expression of MED10 in patients with progressive (PD) and stable (SD) disease, compared with their peers who had complete response/remission (CR), while the partial responders had the lowest MED10 expression (PD > SD > CR > PR) ( Figure 2D ). More so, our systematical analysis of immune infiltrates in BLCA across diverse platforms using the TIMER 2.0 algorithm (19) showed that CD8+ cytotoxic T cells were largely inversely correlated with MED10 protein expression (Spearman’s rho_TIMER = 0.23, p < 0.05; Spearman’s rho_EPIC = -0.09, Spearman’s rho_CIBERSORT = -0.04, Spearman’s rho_XCELL = -0.15, p < 0.05), and infiltration of CD8+ effector memory T cells showed a mild positive correlation with MED10 expression (Spearman’s rho_XCELL = ~0.11, p < 0.05); however, MED10 expression was positively correlated with suppressors of immune response, namely, regulatory T cells, Treg (Spearman’s rho_CIBERSORT = 0.03; Spearman’s rho_QUANTISEQ = 0.13, p < 0.05) and myeloid-derived suppressor cells, and MDSC (Spearman’s rho_TIDE = 0.34, p < 0.05) ( Figure 2E ). Our correlative somatic copy number alteration (sCNA) analyses showed that compared to its ambivalent association with CD8+ T cells with Spearman’s rho ranging from -0.78 to 0.18, high amplification of med10 gene is positively correlated with Treg infiltration ( Figure 2F ). Of clinical relevance, we also showed that patients with concurrent high MED10 expression and low CD8+ T cell levels in the EPIC cohort (n = 1575) exhibited the worst cumulative survival rate (R² = 0.15, p = 4.21e-08) ( Figure 2G ). Using the TCGA BLCA cohort (n = 408), we observed that patients bearing high MED10 expression with or without high MDSC levels exhibited worse survival rates relative to those with concurrent low MED10 and low MDSC levels (R² = 0.13, p = 3.67e-07) ( Figure 2H ). Moreover, patients with concurrent high MED10 and low Treg levels in the TCGA BLCA cohort (n = 408) exhibited the worst cumulative survival rate, compared to those with concomitantly high MED10 and Treg levels or those with low MED10 expression regardless of Treg level (R² = 0.13, p = 7.86e-07) ( Figure 2I ). These data do indicate, at least in part, that MED10 expression is positively correlated with disease progression, immune suppression, and therapy failure.

Against the background that microRNAs (miRs) are critical regulators of gene expression, transcription, and translation (21, 22), seeking to gain some mechanistic insight into the oncogenic and immune-suppressing function of MED10 in BLCA/UC, we performed a MED10–miR association probe. Our MED10–miR association plot showed that MED10 was strongly associated with several miRs including a relatively unknown hsa-miR-590-5p (F-stat = 2.11, p = 1.93e-05) ( Figure 3A ). More so, using the TCGA BLCA cohort (n = 412), we demonstrated that upregulated expression of the med10 gene is positively correlated with several oncogenic miRs (oncomiRs) including hsa-miR-590 ( Figure 3B ) but is inversely correlated with expression levels of tumor-suppressor miRs, including hsa-miR-483, hsa-miR-29c, hsa-miR-100, and hsa-miR-143 ( Figure 3C ). We also demonstrated a significant positive correlation between MED10 and hsa-miR-590-5p transcript expression (r = 0.24, p = 6.29e-07) ( Figure 3D ). Consistent with this, using the GISTIC module (https://www.genepattern.org/modules/docs/GISTIC_2.0), we found that unlike in patients with diploid hsa-miR-590-5p, hsa-miR-590-5p copy number gain or amplification was associated with upregulated MED10 transcript expression in patients with BLCA/UC ( Figure 3E ). These data indicate that MED10 is functionally co-expressed with hsa-miR-590 but is inversely associated with tumor-suppressor microRNAs.

Since co-expression does not necessarily translate into molecular interaction, having shown that MED10 is functionally co-expressed with hsa-miR-590, we probed for a probable interaction between MED10 and hsa-miR-590. Using the mutual exclusivity test for MED10 and hsa-miR-590 in a pooled BLCA cohort (n = 786 patients, 806 samples) consisting of bladder cancer (MSK/TCGA, 2020 n = 476), bladder cancer (MSKCC, Eur Urol 2014 n = 109), bladder urothelial carcinoma (BGI, Nat Genet 2013 n = 99), bladder urothelial carcinoma (DFCI.MSKCC, Cancer Discov 2014 n = 50), and urothelial carcinoma (Cornell/Trento, Nat Gen 2016 n = 72), we confirmed the spatiotemporal association between the dyad (co-occurrence: log2 odds ratio >3, p = 0.02) ( Figure 4A ). To better understand the MED10/hsa-miR-590-5p relationship and for visualization of our hypothesized molecular interaction between MED10 and hsa-miR-590-5p in BLCA/UC cells, we employed a bioinformatics approach to generate the tertiary (3D) structure of MED10 based on sequence homology modeling (NCBI Reference Sequence: NP_115662.2) (https://www.ncbi.nlm.nih.gov/protein/NP_115662.2; Figure 4B ). We also generated the 3D structure of hsa-miR-590 from its sequence (NCBI Reference Sequence: NR_030321.1)-derived “brackets and dots” linear structure and the minimum free energy (MFE) secondary (2D) structure ( Figure 4C ). Using the Schrödinger PyMOL 2.5 molecular interaction and visualization software (https://pymol.org/2/), we observed a high interaction propensity, broad conservation, and good complementarity between the 5′ end of hsa-miR-590 and the C-terminal RNA-binding motif of MED10, demonstrated by a shape complementarity/docking score of 14,146, atomic contact energy (ACE) of −802.59 kcal/mol, approximate MED10/hsa-miR-590-5p complex interface area of 1,990.70 Ų, root-mean-square deviation (RMSD) of atomic positions of 4.0 Å, and a p-value of 3.11e−04, with the 3D transformation data, consisting of three rotational angles (2.42°, 0.38°, 120.70°) and three translational parameters (−1.18, −53.27, −102.28) applied on the ligand molecule, hsa-miR-590 ( Figure 4D ). Furthermore, to confirm the functional and/or modulatory nature of the demonstrated MED10/hsa-miR-590-5p interaction, using agarose gel electrophoresis, we showed that the ectopic expression of MED10 in normal human primary bladder epithelial BdEC cells significantly upregulated the expression of hsa-miR-590, compared to the WT cells; more so, upon silencing MED10 (shMED10) in metastatic human bladder transition cell carcinoma SW1738 cells, hsa-miR-590 expression was markedly suppressed ( Figure 4E ). These findings indicate that MED10 interacts directly with and upregulates hsa-miR-590 expression in BLCA/UC cells. For functional insights, using the TIMER 2.0 algorithm (23), our correlative sCNA analyses showed that akin to MED10, high amplification of hsa-miR-590 was largely correlated with suppressed CD8+ T cell infiltration level but is positively correlated with Treg infiltration (Spearman rho_XCELL = 0.31) ( Figure 4F ). Moreover, survival analysis of the TCGA BLCA cohort (n = 454) showed that high hsa-miR-590 expression conferred survival disadvantage compared to low expression (t-stat = 0.85, p = 0.04) ( Figure 4G ). These data indicate, at least in part, that MED10 interacts directly with hsa-miR-590, a modulator of immune infiltration and survival.

Reaffirming previous data, we demonstrated similarity in the cancer-normal differential expression profile of MED10 (fold change, FC = 1.76, p = 5.7e-06) and hsa-miR-590-5p (FC = 5.78, p = 1.4e-30) in the TCGA BLCA cohort ( Figures 5A, B ). To rule out multifactorial inter-sample inconsistencies/variation and establish replicability of the observed expression profile, our gene detectability power analysis of the GSE81157 aggressive BLCA cohort (n = 9) showed that variation in the true abundance of MED10 (biological coefficient of variation, BCOV = 0.17, power = 0.99) or hsa-miR-590 (BCOV = 0.45, power = 0.02) between replicate RNA samples was apparently insignificant ( Figure 5C ). Bioinformatics-aided evaluation of CRISPR-induced loss of MED10 function (criMED10) in 29 BLCA cell lines showed that increased suppression of med10 elicited increased downregulation of hsa-miR-590 copy number in metastatic cell lines 253J, UMUC1, UMUC13, UMUC14, and JMSU1 (Spearman r = 0.30) compared with the apparent non-effect in primary tumor cell lines (Spearman r = -0.003) ( Figure 5D ). Furthermore, in a non-stratified CCLE–Broad–MIT pool of BLCA cells, criMED10 had no apparent effect (Pearson r = 0.03); however, upon extraction and probe of only transitional cell carcinoma/UC cells, we found that criMED10 elicits suppressed expression of hsa-miR-590-5p in the BC3C, UMUC1, RT112, TCCSUP, BFTC905, UMUC3, VMCUB1, KU1919, 639V, and CAL29 UC cell lines ( Figure 5E ). These data indicate that targeting MED10 elicits downregulation of hsa-miR-590-5p expression preferentially in metastatic, transitional cell (urothelial) carcinoma cells.

Having shown that targeting MED10 elicits downregulation of hsa-miR-590-5p expression preferentially in metastatic, transitional cell (urothelial) carcinoma cells, for functional characterization of the effect of altered MED10/hsa-miR-590-5p signaling in UC cells, we performed several functional assays. We observed that shMED10 significantly suppressed the proliferation of SW1738 (4.47-fold, p < 0.001) and JMSU1 (4.88-fold, p < 0.01) cells ( Figure 6A ). Results of our migration assays demonstrate that shMED10 markedly attenuated the migration of JMSU1 cells (2.11-fold, p < 0.01) ( Figure 6B ). shMED10 also significantly inhibited the invasive capability of JMSU1 cells (4.35-fold, p < 0.01) ( Figure 6C ). Similarly, we demonstrated that shMED10 profoundly suppressed the ability of the SW1738 (9.58-fold, p < 0.001) and JMSU1 (3.28-fold, p < 0.01) cells to form colonies ( Figure 6D ). In parallel assays, we found that shMED10 significantly downregulated the expression levels of MED10, anti-apoptosis BCL-xL, proliferation marker MKI67/Ki67, biomarkers of metastasis VIM and SNAI1, and stemness/pluripotency markers OCT4 and LIN28A, while concomitantly upregulating pro-apoptosis BAX protein in both the SW1738 and JMSU1 cells ( Figure 6E ). Because of the association of onco-aggression and disease recurrence with cancer stemness (21), we assessed for probable effects of altered MED10/hsa-miR-590 signaling on cancer stem cell activities in UC cells. Reanalysis of the AFFY_HG_U133_PLUS_2, GSE31684 UC cohort (n = 93) showed that MED10 is co-overexpressed with stemness markers CD44, SOX2, PROM1/CD133, KLF4, NANOG, and LIN28A in patients with urothelial recurrence and is concomitantly upregulated with NANOG and CD44 in patients with local recurrence, but profoundly suppressed, akin to the other stemness/pluripotency markers, in patients with non-recurrent disease ( Figure 7A ). Interestingly, using our in-house bladder UC tissue samples (n = 79), we showed that compared with the benign tissue samples, MED10 expression, alongside Ki67, OCT4, and LIN28A, is disease progression from non-muscle invasive bladder cancer (NMIBC: pT1) to muscle-invasive bladder cancer (MIBC: pT2–pT4) and more so in patients with distant metastatic disease (M1) (benign < T1 < T2 < T3/4 < M1) ( Figure 7B ). Reminiscent of the loss of self-renewal capability, we demonstrated that shMED10 significantly suppressed the ability of the SW1738 and JMSU1 cells to form primary and subsequently secondary tumorspheres, quantitatively and qualitatively ( Figure 7C ), with concomitant inhibition of the nuclear translocation and co-localization of the pluripotency/stemness markers OCT4 and LIN28A ( Figure 7D ). This is further corroborated by our biological function gene ontology (GO) enrichment analysis showing that MED10 networks with hsa-miR-590, KLF4, SOX2, OCT4/POU5F1, LIN28A, and NANOG and is implicated in “stem cell population maintenance” (GO:0019827: FDR-adjusted p = 5.0e-10), “somatic stem cell population maintenance” (GO:0035019: FDR-adjusted p = 7.2e-09), “maintenance of cell number” (GO:0098727: FDR-adjusted p = 5.0e-10), and “cell fate commitment” (GO:0045165: FDR-adjusted p = 3.3e-04) ( Supplementary Figure 1 ). These data demonstrate that targeting MED10 in bladder UC cells significantly attenuates their oncogenicity and metastatic and cancer stemness phenotypes.