Advertisement

miR-143, miR-222, and miR-452 Are Useful as Tumor Stratification and Noninvasive Diagnostic Biomarkers for Bladder Cancer

      Altered microRNA (miRNA) expression may occur early in bladder cancer and may play a role in carcinogenesis and tumor behavior. We evaluated whether alterations in miRNA expression could improve disease stratification and outcome prognosis in bladder tumors and noninvasive diagnosis in urinary samples. miR-143, miR-222, and miR-452 expression levels were analyzed by quantitative RT-PCR (RT-qPCR) in paired urinary and matching tumors and in two independent prospective series of tumors and urinary specimens. Differential expression of miR-143, miR-222, and miR-452 in urine were verified by in situ hybridization in matching tumors. Tumor miRNA expression by RT-qPCR correlated with tumor grade, size, and presence of carcinoma in situ for miR-222, recurrence (miR-222 and miR-143), progression (miR-222 and miR-143), disease-specific survival (miR-222), and overall survival (miR-222). Protein expression patterns of potential miRNA targets, including vascular endothelial growth factor, BCL2, v-erb-b2 erythroblastic leukemia viral oncogene (ERBB) homolog 3, and ERBB4, were evaluated by IHC in tissue arrays containing tumors for which miRNAs were assessed by RT-qPCR. Target expression correlated with expression of their predicted regulatory miRNAs, recurrence (ERBB3), progression (ERBB4), disease-specific survival (ERBB3 and ERBB4), and overall survival (ERBB3 and ERBB4). Furthermore, RT-qPCR of miR-452 (area under the curve, 0.848) and miR-222 (area under the curve, 0.718) in urine provided high accuracies for bladder cancer diagnosis. Thus, bladder tumors were characterized by changes in miRNA expression that could aid in tumor stratification and clinical outcome prognosis, and miRNAs were detected in urinary specimens for noninvasive diagnosis.
      MicroRNAs (miRNAs) are small non–protein-coding RNAs that regulate gene expression post-transcriptionally by interacting with partially complementary target sites in mRNAs, either inducing their degradation or impairing their translation. miRNAs are implicated in several diseases and cellular functions, including apoptosis, differentiation, and proliferation, among others. Aberrant miRNA expression levels are associated with tumorigenesis, progression, and metastases, acting as oncogenes and/or tumor suppressors.
      • Gottardo F.
      • Liu C.G.
      • Ferracin M.
      • Calin G.A.
      • Fassan M.
      • Bassi P.
      • Sevignani C.
      • Byrne D.
      • Negrini M.
      • Pagano F.
      • Gomella L.G.
      • Croce C.M.
      • Baffa R.
      Micro-RNA profiling in kidney and bladder cancers.
      • Yang H.
      • Dinney C.P.
      • Ye Y.
      • Zhu Y.
      • Grossman H.B.
      • Wu X.
      Evaluation of genetic variants in microRNA-related genes and risk of bladder cancer.
      • Neely L.A.
      • Rieger-Christ K.M.
      • Neto B.S.
      • Eroshkin A.
      • Garver J.
      • Patel S.
      • Phung N.A.
      • McLaughlin S.
      • Libertino J.A.
      • Whitney D.
      • Summerhayes I.C.
      A microRNA expression ratio defining the invasive phenotype in bladder tumors.
      • Veerla S.
      • Lindgren D.
      • Kvist A.
      • Frigyesi A.
      • Staaf J.
      • Persson H.
      • Liedberg F.
      • Chebil G.
      • Gudjonsson S.
      • Borg A.
      • Månsson W.
      • Rovira C.
      • Höglund M.
      miRNA expression in urothelial carcinomas: important roles of miR-10a, miR-222, miR-125b, miR-7 and miR-452 for tumor stage and metastasis, and frequent homozygous losses of miR-31.
      • Lin T.
      • Dong W.
      • Huang J.
      • Pan Q.
      • Fan X.
      • Zhang C.
      • Huang L.
      MicroRNA-143 as a tumor suppressor for bladder cancer.
      • Hanke M.
      • Hoefing K.
      • Merz H.
      • Feller A.C.
      • Kausch I.
      • Jocham D.
      • Warnecke J.M.
      • Sczakiel G.
      A robust methodology to study urine microRNA as tumor marker: microRNA-126 and microRNA-182 are related to urinary bladder cancer.
      • Ichimi T.
      • Enokida H.
      • Okuno Y.
      • Kunimoto R.
      • Chiyomaru T.
      • Kawamoto K.
      • Kawahara K.
      • Toki K.
      • Kawakami K.
      • Nishiyama K.
      • Tsujimoto G.
      • Nakagawa M.
      • Seki N.
      Identification of novel miRNA targets based on microRNA signatures in bladder cancer.
      • Baffa R.
      • Fassan M.
      • Volinia S.
      • O'Hara B.
      • Liu C.G.
      • Palazzo J.P.
      • Gardiman M.
      • Rugge M.
      • Gomella L.G.
      • Croce C.M.
      • Rosenberg A.
      MicroRNA expression profiling of human metastatic cancers identifies cancer gene targets.
      • Wang G.
      • Zhang H.
      • He H.
      • Tong W.
      • Wang B.
      • Liao G.
      • Chen Z.
      • Du C.
      Up-regulation of microRNA in bladder tumor tissue is not common.
      • Dyrskjøt L.
      • Ostenfeld M.S.
      • Bramsen J.B.
      • Silahtaroglu A.N.
      • Lamy P.
      • Ramanathan R.
      • Fristrup N.
      • Jensen J.L.
      • Andersen C.L.
      • Zieger K.
      • Kauppinen S.
      • Ulhøi B.P.
      • Kjems J.
      • Borre M.
      • Orntoft T.F.
      Genomic profiling of miRNAs in bladder cancer: miR-129 is associated with poor outcome and promotes cell death in vitro.
      • Adam L.
      • Zhong M.
      • Choi W.
      • Qi W.
      • Nicoloso M.
      • Arora A.
      • Calin G.
      • Wang H.
      • Siefker-Radtke A.
      • McConkey D.
      • Bar-Eli M.
      • Dinney C.
      miR-200 expression regulates epithelial-to-mesenchymal transition in bladder cancer cells and reverses resistance to epidermal growth factor receptor therapy.
      • Catto J.F.
      • Miah S.
      • Owen H.C.
      • Bryant H.
      • Myers K.
      • Dudziec E.
      • Larré S.
      • Milo M.
      • Rehman I.
      • Rosario D.J.
      • Di Martino E.
      • Knowles M.A.
      • Meuth M.
      • Harris A.L.
      • Hamdy F.C.
      Distinct microRNA alterations characterize high- and low-grade bladder cancer.
      • Ostenfeld M.S.
      • Bramsen J.B.
      • Lamy P.
      • Villadsen S.B.
      • Fristrup N.
      • Sørensen K.D.
      • Ulhøi B.
      • Borre M.
      • Kjems J.
      • Dyrskjøt L.
      • Orntoft T.F.
      miR-145 induces caspase-dependant and -independent cell death in urothelial cancer cell lines with targeting of an expression signature present in Ta bladder tumors.
      • Chiyomaru T.
      • Enokida H.
      • Tatarano S.
      • Kawahara K.
      • Uchida Y.
      • Nishiyama K.
      • Fujimura L.
      • Kikkawa N.
      • Seki N.
      • Nakagawa M.
      miR-145 and miR-133a function as tumor suppressors and directly regulate FSCN1 expression in bladder cancer.
      • Wiklund E.D.
      • Bramsen J.B.
      • Hulf T.
      • Dyrskjøt L.
      • Ramanathan R.
      • Hansen T.B.
      • Villadsen S.B.
      • Gao S.
      • Ostenfeld M.S.
      • Borre M.
      • Peter M.E.
      • Ørntoft T.F.
      • Kjems J.
      • Clark S.J.
      Coordinated epigenetic repression of the miR-200 family and miR-205 in invasive bladder cancer.
      • Han Y.
      • Chen J.
      • Zhao X.
      • Liang C.
      • Wang Y.
      • Sun L.
      • Jiang Z.
      • Zhang Z.
      • Yang R.
      • Chen J.
      • Li Z.
      • Tang A.
      • Li X.
      • Ye J.
      • Guan Z.
      • Gui Y.
      • Cai Z.
      MicroRNA expression signatures of bladder cancer revealed by deep sequencing.
      • Noguchi S.
      • Mori T.
      • Hoshino Y.
      • Maruo K.
      • Yamada N.
      • Kitade Y.
      • Naoe T.
      • Akao Y.
      MicroRNA-143 functions as a tumor suppressor in human bladder cancer T24 cells.
      The tissue-specific nature of miRNA expression suggests that different tumors would express specific miRNA signatures. An increasing number of studies designed to decipher miRNAs specific to bladder cancer have attempted to understand regulatory effects of miRNAs on bladder tumorigenesis and progression and to search for potential roles of miRNAs as diagnostic and/or prognostic/predictive biomarkers.
      • Gottardo F.
      • Liu C.G.
      • Ferracin M.
      • Calin G.A.
      • Fassan M.
      • Bassi P.
      • Sevignani C.
      • Byrne D.
      • Negrini M.
      • Pagano F.
      • Gomella L.G.
      • Croce C.M.
      • Baffa R.
      Micro-RNA profiling in kidney and bladder cancers.
      • Yang H.
      • Dinney C.P.
      • Ye Y.
      • Zhu Y.
      • Grossman H.B.
      • Wu X.
      Evaluation of genetic variants in microRNA-related genes and risk of bladder cancer.
      • Neely L.A.
      • Rieger-Christ K.M.
      • Neto B.S.
      • Eroshkin A.
      • Garver J.
      • Patel S.
      • Phung N.A.
      • McLaughlin S.
      • Libertino J.A.
      • Whitney D.
      • Summerhayes I.C.
      A microRNA expression ratio defining the invasive phenotype in bladder tumors.
      • Veerla S.
      • Lindgren D.
      • Kvist A.
      • Frigyesi A.
      • Staaf J.
      • Persson H.
      • Liedberg F.
      • Chebil G.
      • Gudjonsson S.
      • Borg A.
      • Månsson W.
      • Rovira C.
      • Höglund M.
      miRNA expression in urothelial carcinomas: important roles of miR-10a, miR-222, miR-125b, miR-7 and miR-452 for tumor stage and metastasis, and frequent homozygous losses of miR-31.
      • Lin T.
      • Dong W.
      • Huang J.
      • Pan Q.
      • Fan X.
      • Zhang C.
      • Huang L.
      MicroRNA-143 as a tumor suppressor for bladder cancer.
      • Hanke M.
      • Hoefing K.
      • Merz H.
      • Feller A.C.
      • Kausch I.
      • Jocham D.
      • Warnecke J.M.
      • Sczakiel G.
      A robust methodology to study urine microRNA as tumor marker: microRNA-126 and microRNA-182 are related to urinary bladder cancer.
      • Ichimi T.
      • Enokida H.
      • Okuno Y.
      • Kunimoto R.
      • Chiyomaru T.
      • Kawamoto K.
      • Kawahara K.
      • Toki K.
      • Kawakami K.
      • Nishiyama K.
      • Tsujimoto G.
      • Nakagawa M.
      • Seki N.
      Identification of novel miRNA targets based on microRNA signatures in bladder cancer.
      • Baffa R.
      • Fassan M.
      • Volinia S.
      • O'Hara B.
      • Liu C.G.
      • Palazzo J.P.
      • Gardiman M.
      • Rugge M.
      • Gomella L.G.
      • Croce C.M.
      • Rosenberg A.
      MicroRNA expression profiling of human metastatic cancers identifies cancer gene targets.
      • Wang G.
      • Zhang H.
      • He H.
      • Tong W.
      • Wang B.
      • Liao G.
      • Chen Z.
      • Du C.
      Up-regulation of microRNA in bladder tumor tissue is not common.
      • Dyrskjøt L.
      • Ostenfeld M.S.
      • Bramsen J.B.
      • Silahtaroglu A.N.
      • Lamy P.
      • Ramanathan R.
      • Fristrup N.
      • Jensen J.L.
      • Andersen C.L.
      • Zieger K.
      • Kauppinen S.
      • Ulhøi B.P.
      • Kjems J.
      • Borre M.
      • Orntoft T.F.
      Genomic profiling of miRNAs in bladder cancer: miR-129 is associated with poor outcome and promotes cell death in vitro.
      • Adam L.
      • Zhong M.
      • Choi W.
      • Qi W.
      • Nicoloso M.
      • Arora A.
      • Calin G.
      • Wang H.
      • Siefker-Radtke A.
      • McConkey D.
      • Bar-Eli M.
      • Dinney C.
      miR-200 expression regulates epithelial-to-mesenchymal transition in bladder cancer cells and reverses resistance to epidermal growth factor receptor therapy.
      • Catto J.F.
      • Miah S.
      • Owen H.C.
      • Bryant H.
      • Myers K.
      • Dudziec E.
      • Larré S.
      • Milo M.
      • Rehman I.
      • Rosario D.J.
      • Di Martino E.
      • Knowles M.A.
      • Meuth M.
      • Harris A.L.
      • Hamdy F.C.
      Distinct microRNA alterations characterize high- and low-grade bladder cancer.
      • Ostenfeld M.S.
      • Bramsen J.B.
      • Lamy P.
      • Villadsen S.B.
      • Fristrup N.
      • Sørensen K.D.
      • Ulhøi B.
      • Borre M.
      • Kjems J.
      • Dyrskjøt L.
      • Orntoft T.F.
      miR-145 induces caspase-dependant and -independent cell death in urothelial cancer cell lines with targeting of an expression signature present in Ta bladder tumors.
      • Chiyomaru T.
      • Enokida H.
      • Tatarano S.
      • Kawahara K.
      • Uchida Y.
      • Nishiyama K.
      • Fujimura L.
      • Kikkawa N.
      • Seki N.
      • Nakagawa M.
      miR-145 and miR-133a function as tumor suppressors and directly regulate FSCN1 expression in bladder cancer.
      • Wiklund E.D.
      • Bramsen J.B.
      • Hulf T.
      • Dyrskjøt L.
      • Ramanathan R.
      • Hansen T.B.
      • Villadsen S.B.
      • Gao S.
      • Ostenfeld M.S.
      • Borre M.
      • Peter M.E.
      • Ørntoft T.F.
      • Kjems J.
      • Clark S.J.
      Coordinated epigenetic repression of the miR-200 family and miR-205 in invasive bladder cancer.
      • Han Y.
      • Chen J.
      • Zhao X.
      • Liang C.
      • Wang Y.
      • Sun L.
      • Jiang Z.
      • Zhang Z.
      • Yang R.
      • Chen J.
      • Li Z.
      • Tang A.
      • Li X.
      • Ye J.
      • Guan Z.
      • Gui Y.
      • Cai Z.
      MicroRNA expression signatures of bladder cancer revealed by deep sequencing.
      • Noguchi S.
      • Mori T.
      • Hoshino Y.
      • Maruo K.
      • Yamada N.
      • Kitade Y.
      • Naoe T.
      • Akao Y.
      MicroRNA-143 functions as a tumor suppressor in human bladder cancer T24 cells.
      Following the hypothesis that altered miRNA expression occurs early in bladder carcinogenesis and contributes to tumor behavior, we tested whether miRNAs previously shown to be differentially expressed in bladder tumors
      • Veerla S.
      • Lindgren D.
      • Kvist A.
      • Frigyesi A.
      • Staaf J.
      • Persson H.
      • Liedberg F.
      • Chebil G.
      • Gudjonsson S.
      • Borg A.
      • Månsson W.
      • Rovira C.
      • Höglund M.
      miRNA expression in urothelial carcinomas: important roles of miR-10a, miR-222, miR-125b, miR-7 and miR-452 for tumor stage and metastasis, and frequent homozygous losses of miR-31.
      • Noguchi S.
      • Mori T.
      • Hoshino Y.
      • Maruo K.
      • Yamada N.
      • Kitade Y.
      • Naoe T.
      • Akao Y.
      MicroRNA-143 functions as a tumor suppressor in human bladder cancer T24 cells.
      could be detected in urinary specimens and, thereby, useful in bladder cancer diagnosis. Specifically, miR-143, miR-222, and miR-452 were analyzed in matching urothelial tissues and independent sets of tumor specimens to determine their expression and correlation with pathological variables and outcome. Protein expression levels of potential miRNA targets by immunohistochemistry (IHC) were assessed to confirm their association with miRNA expression and clinicopathologic assessment, as shown in the experimental design (see Supplemental Figure S1, A–E, at http://ajp.amjpathol.org).

      Materials and Methods

      Tumor Samples

      Paraffin-embedded tumors were prospectively collected following Institutional Review Board–approved protocols. Non–muscle-invasive tumors underwent transurethral resection as primary treatment.
      • Kirkali Z.
      • Chan T.
      • Manoharan M.
      • Algaba F.
      • Busch C.
      • Cheng L.
      • Kiemeney L.
      • Kriegmair M.
      • Montironi R.
      • Murphy W.M.
      • Sesterhenn I.A.
      • Tachibana M.
      • Weider J.
      Bladder cancer: epidemiology, staging and grading, and diagnosis.
      Adjuvant intravesical chemotherapy was used for low-grade tumors, and maintenance Bacillus Calmette-Guerin was used for high-grade tumors. Radical cystectomy with pelvic lymphadenectomy was used for muscle-invasive tumors. Patients underwent surveillance stratified by their disease.
      • Kirkali Z.
      • Chan T.
      • Manoharan M.
      • Algaba F.
      • Busch C.
      • Cheng L.
      • Kiemeney L.
      • Kriegmair M.
      • Montironi R.
      • Murphy W.M.
      • Sesterhenn I.A.
      • Tachibana M.
      • Weider J.
      Bladder cancer: epidemiology, staging and grading, and diagnosis.
      Inclusion criteria of primary tumors were based on histopathological characteristics, requiring high-quality RNA for miRNA analyses. A first set included tumors with available matching urinary samples (n = 37). Demographic information indicated 31 males and 6 females (median age, 66 years; range, 51 to 86 years). A second set included tumors (n = 164) for which tissue arrays were constructed. High-quality RNA was obtained from 113 tumors. Demographic information indicated 101 males and 12 females (median age, 73 years; range, 35 to 94 years).

      Urinary Samples

      Individuals presenting microscopic hematuria under first bladder cancer suspicion provided urine specimens prospectively collected immediately before cystoscopy, following Institutional Review Board–approved protocols. Urinary specimens for which high-quality RNA was obtained in matching bladder tumors were used to assess the correlation of miRNA expression between matching pairs of specimens (n = 37). An independent urine set (n = 94) assessed the diagnostic utility of miRNAs discriminating patients with bladder cancer (n = 37) from controls (n = 57), including healthy individuals (n = 20) and patients with benign urological diseases (n = 37). Demographic information indicated 44 males and 50 females (median age, 50.3 years; range, 16 to 87 years). The histopathological tumor stage distribution was as follows: pTa, n = 3; pT1, n = 19; and pT2+, n = 15. Tumor grade distribution was as follows: low grade, n = 12; and high grade, n = 25.

      RNA Extraction

      Total RNA from tissue and urine specimens was extracted using TRIzol (Invitrogen, Carlsbad, CA).
      • Sanchez-Carbayo M.
      • Schwarz K.
      • Charytonowicz E.
      • Cordon-Cardo C.
      • Mundel P.
      Tumor suppressor role for myopodin in bladder cancer: loss of nuclear expression of myopodin is cell-cycle dependent and predicts clinical outcome.
      For paraffin-embedded tissues, tumor cell content was estimated to be >75% on consecutive sections after H&E staining, and corresponding pieces were digested overnight before RNA extraction using proteinase-K (Roche, Basel, Switzerland). The concentration and purity of RNA were determined with an ND-1000 spectrophotometer (NanoDrop, Wilmington, DE). RNA quality was evaluated based on 260/280 ratios of absorbances and gel electrophoresis using the 2100 Bioanalyzer (Agilent, Santa Clara, CA).

      RT-qPCR Data

      For quantitative determination of miR-16, miR-143, miR-222, and miR-452 expression, reverse transcription of RNA (300 ng) was performed in triplicate by quantitative RT-PCR (RT-qPCR) TaqMan miRNA assays (Applied Biosystems, Foster City, CA) using an ABI Prism 7900HT system (Applied Biosystems). miR-16 was used for internal normalization because of consistent nondifferential expression levels in tumor and nonneoplastic samples. The CT values were calculated with SDS software version 2.1 (Applied Biosystems). Fold differences were obtained using the following equation: 2−ΔCT, where CT is the threshold cycle, and ΔCT equals the mean CT of the sample gene minus the mean CT of miR-16.
      In situ hybridization was performed using formalin-fixed, paraffin-embedded normal urothelium biopsy specimens (5 μm thick) and bladder tumors belonging to patients with matching urinary samples analyzed by RT-qPCR. In situ hybridization was also performed using 10 pmol of locked nucleic acid–modified fluorescein isothiocyanate (FITC)–labeled oligonucleotide probes (Exiqon, Vedback, Denmark), complementary to miR-222, miR-143, and miR-452, overnight at 25°C lower than the predicted temperature of melting of the probe.
      • Pena J.T.G.
      • Sohn-Lee C.
      • Rouhanifard S.H.
      • Ludwig J.
      • Hafner M.
      • Mihailovic A.
      • Lim C.
      • Holoch D.
      • Berninger P.
      • Zavolan M.
      • Tuschl T.
      miRNA in situ hybridization in mammalian tissues fixed with formaldehyde and EDC.
      • Jørgensen S.
      • Baker A.
      • Møller S.
      • Nielsen B.S.
      Robust one-day in situ hybridization protocol for detection of microRNAs in paraffin samples using LNA probes.

      Prediction and Selection of miRNA Targets

      Potential miRNA targets were predicted using four algorithms: PicTar, TargetScan, miRanda, and DIANA microT. To decrease the number of false-positive results, a putative mRNA was only considered a target if found in at least two of the four algorithms. Targets were selected based on availability of antibodies to study their protein expression: BCL2, potential target of miR-143 according to TargetScan, Pictar, and DIANA microT, and of miR-222 according to DIANA microT; vascular endothelial growth factor (VEGF), potential target of miR-452 according to TargetScan and DIANA microT, and of miR-16 according to TargetScan, miRanda, and DIANA microT; v-erb-b2 erythroblastic leukemia viral oncogene (ERBB) homolog 3, potential target of miR-143 according to TargetScan, miRanda, and DIANA microT, and of miR-222 according to miRanda and DIANA microT; and ERBB4, potential target of miR-222 according to miRanda and DIANA microT, and of miR-452 according to TargetScan.

      TMA Data

      Tissue microarray (TMA) sections (n = 164, 5 μm thick) constructed at Centro Nacional de Investigaciones Oncológicas were used for IHC to evaluate protein expression patterns of potential miRNA targets. Demographic information indicated 141 males and 23 females (median age, 71 years; range, 32 to 94 years). The tumor stage distribution was as follows: pTa, n = 31; pT1, n = 69; pT2, n = 20; pT3, n = 1; and pT4, n = 1. The tumor grade distribution was as follows: low grade, n = 57; and high grade, n = 65.

      IHC Data

      Protein expression patterns were assessed using avidin-biotin immunoperoxidase procedures. Antigen retrieval (0.01% citric acid for 15 minutes using a microwave) was used before incubation with primary mouse monoclonal antibodies overnight at 4°C: ERBB3, clone RTJ.1 at 1:100 (Sigma-Aldrich, St. Louis, MO); ERBB4, clone 182803, at 1:200 (R&D Systems, Minneapolis, MN); BCL2 FLEX, undiluted (Dako, Glostrup, Denmark); and VEGF, undiluted (Abcam, Cambridge, England, UK). Staining conditions were optimized using controls for each antibody specified by manufacturers. The secondary biotinylated horse anti-mouse antibody (Vector Laboratories, Burlingame, CA) was used at 1:500 dilution. The absence of primary antibody was used as a negative control. Diaminobenzidine was used as a chromogen, and hematoxylin was used as a nuclear counterstain.
      • Sanchez-Carbayo M.
      • Schwarz K.
      • Charytonowicz E.
      • Cordon-Cardo C.
      • Mundel P.
      Tumor suppressor role for myopodin in bladder cancer: loss of nuclear expression of myopodin is cell-cycle dependent and predicts clinical outcome.

      Statistical Analysis

      Associations of miRNAs and protein levels with clinicopathologic variables were evaluated using nonparametric Wilcoxon-Mann-Whitney and Kruskal-Wallis tests. Associations among miRNAs and protein levels were analyzed using Kendall's τ-β test. Consensus values for protein expression of three representative cores from each tumor-arrayed sample were evaluated as continuous variables based on the number of cells expressing them in each subcellular compartment: ERBB3 and ERBB4 (membrane, cytoplasm, and nucleus) and VEGF and BCL2 (cytoplasm and nucleus). Expression cutoffs were selected based on median values of expression among groups analyzed. Associations of miRNAs and protein expression with outcome were evaluated using the log-rank test when follow-up data were available. Outcome end points included recurrence, progression, disease-specific survival, and overall survival. Tumor recurrence and progression were defined as the presence of pathological, radiological, or clinical evidence of similar stage (noninvasive tumor) or increased tumor stage (muscle invasive), respectively, and measured from the time of surgery to the time of proven event. Disease-specific survival and overall survival times were defined as the months elapsed between transurethral resection or cystectomy and death as a result of disease (or the last follow-up date) or of other causes, respectively. Patients alive at the last follow-up or lost to follow-up were censored. Outcome was plotted using Kaplan-Meier curves. Receiver operating characteristic curve analyses defined the diagnostic performance of urinary miRNAs, using areas under the curve (AUCs), 95% CIs, and two-sided P values. Statistical analyses were performed using the SPSS version 18.8 statistical package.

      Results

      miRNA Expression Profiles in Pairs of Bladder Tumors and Urinary Specimens

      RT-qPCR of miR-143, miR-222, and miR-452 was performed in pairs of urinary samples and matching tumors with high-quality RNA in both specimens (n = 37). Results showed a significant correlation in quantitative levels of expression between the urinary samples and matching tumors for miR-452 (τ-β = 0.320, P = 0.005) and miR-222 (τ-β = 0.305, P = 0.008), but not for miR-143. As expected, miR-16 levels in the urine also correlated to its expression in the matching tumors (τ-β = 0.227, P = 0.048). To investigate whether these miRNAs localized to neoplastic cells, in situ hybridization was performed in matching bladder tissues of urinary specimens with available mRNA (Figure 1). miR-143 staining was positive in urothelial hyperplasia (Figure 1A), whereas little to no expression was detected in tumor cells of a high-grade T1 tumor (Figure 1B). Strong signals were observed for both miR-222 (Figure 1C) and miR-452 (Figure 1D) in the same high-grade T1 tumor. Comparative quantitation of the candidate miRNAs through RT-qPCR in different tumor samples, including the tumor shown in Figure 1, is shown in Supplemental Figure S2 (available at http://ajp.amjpathol.org). Interestingly, little to no expression was observed for miR-452 and miR-222 in normal urothelium and surrounding stroma. Although miR-143 also stained negative in most normal urothelial cells, positive signals were observed in the surrounding stromal cells (see Supplemental Figure S3 at http://ajp.amjpathol.org). Illustrations of negative and positive controls are provided in Supplemental Figure S4 (available at http://ajp.amjpathol.org). These studies highlighted that the expression levels of urinary miRNAs correlated with those observed in matching tumors, and that cancer cells displayed low levels of expression for miR-143 and higher levels of expression for miR-222 and miR-452.
      Figure thumbnail gr1
      Figure 1Verification analyses by in situ hybridization of the differential expression of the identified miRNAs. Representative in situ patterns in urothelial hyperplasia for miR-143 (A) and, in a high-grade pT1 bladder tumor, for miR-143 (B), miR-222 (C), and miR-452 (D). In situ hybridization was performed on tissue specimens matching the urinary samples analyzed by RT-qPCR. The miRNA locked nucleic acid probes were labeled with FITC (green) and counterstained with DAPI (blue). Original magnification, ×400.

      Association of miRNA Profiles of miR-143, miR-222, and miR-452 in Bladder Tumors by RT-qPCR with Clinicopathologic Correlates

      miR-222 expression was significantly correlated with increasing tumor grade (P = 0.017), tumor size (P = 0.005), presence of carcinoma in situ (P = 0.035), and clinical outcome end points (recurrence, P = 0.006; progression, P = 0.003; disease specific, P = 0.034; and overall survival, P = 0.023) (Figure 2, A–D). Furthermore, miR-143 expression significantly correlated with the following clinical outcome end points: recurrence, P = 0.011; and progression, P = 0.039 (Figure 2, E and F).
      Figure thumbnail gr2
      Figure 2miRNA expression levels by RT-qPCR in tumor specimens are associated with clinical outcome end points in patients with bladder cancer (n = 113). miR-222 expression was associated with a high recurrence rate (log-rank P = 0.006) (A), progression into muscle-invasive disease (log-rank P = 0.003) (B), disease-specific survival (log-rank P = 0.034) (C), and overall survival (log-rank P = 0.023) (D). miR-143 expression was associated with a high recurrence rate (log-rank P = 0.011) (E) and progression into muscle-invasive disease (log-rank P = 0.039) (F).

      Association of Protein Expression Profiles of Potential Targets of miR-143, miR-222, and miR-452 in Bladder Tumors by IHC with miRNA Expression Profiles and Clinicopathologic Correlates

      Differential expression for BCL2, VEGF, ERBB3, and ERBB4 was observed among the bladder tumors tested (Figure 3). Significant associations were found with tumor stage for ERBB4 (P = 0.013), ERBB3 (P = 0.003), and BCL2 (P = 0.030); tumor grade for ERBB4 (P = 0.027), ERBB3 (P = 0.006), VEGF (P = 0.024), and BCL2 (P = 0.030); tumor size for ERBB4 (P = 0.025) and ERBB3 (P = 0.005); and pattern of growth for ERBB3 (P = 0.007), VEGF (P = 0.041), and BCL2 (P = 0.013). Moreover, ERBB4 and ERBB3 protein expression patterns correlated with clinical outcome (Figure 4). Specifically, ERBB4 nuclear expression correlated with progression (P = 0.001), disease-specific survival (P < 0.0005), and overall survival (P = 0.002). ERBB3 cytoplasmic expression correlated with recurrence (P = 0.044), disease-specific survival (P = 0.034), and overall survival (P = 0.012). Inverse correlations between miRNA expression and protein expression patterns were observed for the following pairs: VEGF–miR-452 (τ-β = −0.191, P = 0.010), VEGF–miR-16 (τ-β = −0.296, P < 0.0005), BCL2–miR-222 (τ-β = −0.216, P = 0.009), BCL2–miR-143 (τ-β = −0.195, P = 0.009), and ERBB3–miR-222 (τ-β = −0.214, P = 0.046).
      Figure thumbnail gr3
      Figure 3Verification analyses by IHC of the differential expression of the identified proteins in bladder tumors (n = 164). Representative protein expression patterns of the following proteins in bladder tumors contained in tissue arrays. For ERBB4 (A and B), ERBB3 (C and D), and BCL2 (E and F), protein expression was significantly different between noninvasive lesions (A, C, and E) and invasive tumors (B, D, and F). For VEGF, lower protein expression was observed in low-grade tumors (G) compared with high-grade tumors (H). Original magnification, ×400.
      Figure thumbnail gr4
      Figure 4Protein expression patterns of miRNA targets obtained by IHC in bladder tumors spotted on tissue arrays (n = 164) are associated with clinical outcome end points in patients with bladder cancer. ERBB4 nuclear expression was associated with progression into muscle-invasive disease (log-rank P = 0.001) (A), disease-specific survival (log-rank P < 0.0005) (B), and overall survival (log-rank P = 0.002) (C). ERBB3 cytoplasmic expression was associated with a high recurrence rate (log-rank P = 0.044) (D), disease-specific survival (log-rank P = 0.034) (E), and overall survival (log-rank P = 0.012) (F).

      Diagnostic Properties of miRNAs in Urinary Specimens

      The miRNA profiles by RT-qPCR were measured in an independent series of urinary specimens belonging to patients with bladder cancer (n = 37) and controls (n = 57). The receiver operating characteristic curve analyses indicated that urinary miR-452 alone (AUC = 0.772), and miR-452 (AUC = 0.848) or miR-222 (AUC = 0.718) normalized by miR-16 expression, provided significant accuracies for bladder cancer diagnosis (Figure 5).
      Figure thumbnail gr5
      Figure 5Utility of miRNA expression patterns for bladder cancer diagnostics in urinary specimens (n = 94). The receiver operating characteristic curve analyses indicated that the measurement of urinary miR-452 alone (AUC = 0.772), and those of miR-452 (AUC = 0.848) and miR-222 (AUC = 0.718), normalized by miR-16 expression, provided significant accuracies for the diagnosis of patients with bladder cancer from controls.

      Discussion

      Both RT-qPCR correlations of matching tumor and urinary specimens and in situ hybridization analyses showing differential expression of candidate miRNAs in bladder tumors supported the working hypothesis by which urinary miRNA measurement could mirror the miRNA expression in bladder tumors and be potentially used for noninvasive bladder cancer diagnostics. In situ hybridization probes detecting premature and mature miRNAs displayed cytoplasmic or nuclear staining, such as for miR-452 and miR-143.
      • Catto J.W.
      • Alcaraz A.
      • Bjartell A.S.
      • De Vere White R.
      • Evans C.P.
      • Fussel S.
      • Hamdy F.C.
      • Kallioniemi O.
      • Mengual L.
      • Schlomm T.
      • Visakorpi T.
      MicroRNA in prostate, bladder, and kidney cancer: a systematic review.
      • Tsai Z.Y.
      • Singh S.
      • Yu S.L.
      • Kao L.P.
      • Chen B.Z.
      • Ho B.C.
      • Yang P.C.
      • Li S.S.
      Identification of microRNAs regulated by activin A in human embryonic stem cells.
      In situ localization analyses served to address the source of miRNAs in the urine, from cancer urothelial cells and/or the stroma, as shown for miR-143.
      Previous miRNA array profiling reports indicated lower miR-143 expression in tumors.
      • Han Y.
      • Chen J.
      • Zhao X.
      • Liang C.
      • Wang Y.
      • Sun L.
      • Jiang Z.
      • Zhang Z.
      • Yang R.
      • Chen J.
      • Li Z.
      • Tang A.
      • Li X.
      • Ye J.
      • Guan Z.
      • Gui Y.
      • Cai Z.
      MicroRNA expression signatures of bladder cancer revealed by deep sequencing.
      • Noguchi S.
      • Mori T.
      • Hoshino Y.
      • Maruo K.
      • Yamada N.
      • Kitade Y.
      • Naoe T.
      • Akao Y.
      MicroRNA-143 functions as a tumor suppressor in human bladder cancer T24 cells.
      Higher miR-222 expression was found in muscle-invasive tumors,
      • Veerla S.
      • Lindgren D.
      • Kvist A.
      • Frigyesi A.
      • Staaf J.
      • Persson H.
      • Liedberg F.
      • Chebil G.
      • Gudjonsson S.
      • Borg A.
      • Månsson W.
      • Rovira C.
      • Höglund M.
      miRNA expression in urothelial carcinomas: important roles of miR-10a, miR-222, miR-125b, miR-7 and miR-452 for tumor stage and metastasis, and frequent homozygous losses of miR-31.
      and high miR-452 levels were found in tumors with lymph node metastases.
      • Veerla S.
      • Lindgren D.
      • Kvist A.
      • Frigyesi A.
      • Staaf J.
      • Persson H.
      • Liedberg F.
      • Chebil G.
      • Gudjonsson S.
      • Borg A.
      • Månsson W.
      • Rovira C.
      • Höglund M.
      miRNA expression in urothelial carcinomas: important roles of miR-10a, miR-222, miR-125b, miR-7 and miR-452 for tumor stage and metastasis, and frequent homozygous losses of miR-31.
      However, the three miRNAs were not previously described as useful in bladder cancer for diagnosis (miR-452) and clinical outcome prognosis (miR-143 and miR-222). None of these miRNAs were previously detected in urinary specimens. In support of our selection of miR-16 as the reference, this candidate also previously had consistent levels of expression in normal and bladder cancer–related specimens.
      • Hanke M.
      • Hoefing K.
      • Merz H.
      • Feller A.C.
      • Kausch I.
      • Jocham D.
      • Warnecke J.M.
      • Sczakiel G.
      A robust methodology to study urine microRNA as tumor marker: microRNA-126 and microRNA-182 are related to urinary bladder cancer.
      Our results suggested that miR-452 could contribute in tumorigenesis and aid in bladder cancer diagnostics, whereas miR-143 and miR-222 could be related to tumor progression and used for clinical outcome assessment. Consistent with previous reports, our findings suggested an oncogenic role for miR-222
      • Veerla S.
      • Lindgren D.
      • Kvist A.
      • Frigyesi A.
      • Staaf J.
      • Persson H.
      • Liedberg F.
      • Chebil G.
      • Gudjonsson S.
      • Borg A.
      • Månsson W.
      • Rovira C.
      • Höglund M.
      miRNA expression in urothelial carcinomas: important roles of miR-10a, miR-222, miR-125b, miR-7 and miR-452 for tumor stage and metastasis, and frequent homozygous losses of miR-31.
      and miR-452
      • Veerla S.
      • Lindgren D.
      • Kvist A.
      • Frigyesi A.
      • Staaf J.
      • Persson H.
      • Liedberg F.
      • Chebil G.
      • Gudjonsson S.
      • Borg A.
      • Månsson W.
      • Rovira C.
      • Höglund M.
      miRNA expression in urothelial carcinomas: important roles of miR-10a, miR-222, miR-125b, miR-7 and miR-452 for tumor stage and metastasis, and frequent homozygous losses of miR-31.
      and a tumor suppressor role for miR-143.
      • Han Y.
      • Chen J.
      • Zhao X.
      • Liang C.
      • Wang Y.
      • Sun L.
      • Jiang Z.
      • Zhang Z.
      • Yang R.
      • Chen J.
      • Li Z.
      • Tang A.
      • Li X.
      • Ye J.
      • Guan Z.
      • Gui Y.
      • Cai Z.
      MicroRNA expression signatures of bladder cancer revealed by deep sequencing.
      • Noguchi S.
      • Mori T.
      • Hoshino Y.
      • Maruo K.
      • Yamada N.
      • Kitade Y.
      • Naoe T.
      • Akao Y.
      MicroRNA-143 functions as a tumor suppressor in human bladder cancer T24 cells.
      Whether these miRNAs are involved in carcinogenesis and/or disease progression as tumor suppressors or oncogenic events remains to be determined in functional in vitro and in vivo analyses. Such experiments are out of the scope of this study, which aims to define the prognostic and diagnostic roles of these miRNAs.
      We observed inverse correlations between miRNAs and protein expression of their selected potential targets, which supported the validity of the target prediction.
      • Catto J.W.
      • Alcaraz A.
      • Bjartell A.S.
      • De Vere White R.
      • Evans C.P.
      • Fussel S.
      • Hamdy F.C.
      • Kallioniemi O.
      • Mengual L.
      • Schlomm T.
      • Visakorpi T.
      MicroRNA in prostate, bladder, and kidney cancer: a systematic review.
      • Tsai Z.Y.
      • Singh S.
      • Yu S.L.
      • Kao L.P.
      • Chen B.Z.
      • Ho B.C.
      • Yang P.C.
      • Li S.S.
      Identification of microRNAs regulated by activin A in human embryonic stem cells.
      • Schmidt W.M.
      • Spiel A.O.
      • Jilma B.
      • Wolzt M.
      • Müller M.
      In vivo profile of the human leukocyte microRNA response to endotoxemia.
      • Chen H.C.
      • Chen G.H.
      • Chen Y.H.
      • Liao W.L.
      • Liu C.Y.
      • Chang K.P.
      • Chang Y.S.
      • Chen S.J.
      MicroRNA deregulation and pathway alterations in nasopharyngeal carcinoma.
      Although the associations between the miRNAs and the targets evaluated in our series are novel in bladder cancer, the link between these miRNAs and the expression of these targets has been shown in other models: miR-222 and ERBB4 in embryonic stem cells,
      • Tsai Z.Y.
      • Singh S.
      • Yu S.L.
      • Kao L.P.
      • Chen B.Z.
      • Ho B.C.
      • Yang P.C.
      • Li S.S.
      Identification of microRNAs regulated by activin A in human embryonic stem cells.
      miR-143 and BCL2 in human leukocytes,
      • Schmidt W.M.
      • Spiel A.O.
      • Jilma B.
      • Wolzt M.
      • Müller M.
      In vivo profile of the human leukocyte microRNA response to endotoxemia.
      and miR-143 and VEGF in nasopharyngeal carcinoma.
      • Chen H.C.
      • Chen G.H.
      • Chen Y.H.
      • Liao W.L.
      • Liu C.Y.
      • Chang K.P.
      • Chang Y.S.
      • Chen S.J.
      MicroRNA deregulation and pathway alterations in nasopharyngeal carcinoma.
      ERBB4 expression inversely correlated with the expression of miR-222 and miR-452, which suggested that several miRNAs could simultaneously regulate the expression of one specific protein. Conversely, miR-222 correlated with the protein expression of BCL2 and ERBB3, supporting the notion that a single miRNA could regulate the expression of more than one protein.
      • Veerla S.
      • Lindgren D.
      • Kvist A.
      • Frigyesi A.
      • Staaf J.
      • Persson H.
      • Liedberg F.
      • Chebil G.
      • Gudjonsson S.
      • Borg A.
      • Månsson W.
      • Rovira C.
      • Höglund M.
      miRNA expression in urothelial carcinomas: important roles of miR-10a, miR-222, miR-125b, miR-7 and miR-452 for tumor stage and metastasis, and frequent homozygous losses of miR-31.
      • Noguchi S.
      • Mori T.
      • Hoshino Y.
      • Maruo K.
      • Yamada N.
      • Kitade Y.
      • Naoe T.
      • Akao Y.
      MicroRNA-143 functions as a tumor suppressor in human bladder cancer T24 cells.
      • Catto J.W.
      • Alcaraz A.
      • Bjartell A.S.
      • De Vere White R.
      • Evans C.P.
      • Fussel S.
      • Hamdy F.C.
      • Kallioniemi O.
      • Mengual L.
      • Schlomm T.
      • Visakorpi T.
      MicroRNA in prostate, bladder, and kidney cancer: a systematic review.
      Analyses of expression patterns of BCL2, VEGF, ERBB3, and ERBB4 served to associate the miRNA expression to their targets and were clinically relevant by being correlated with histopathological correlates of tumor progression and/or several outcome end points. ERBB4 has localized to several cellular counterparts: membrane,
      • Kassouf W.
      • Black P.C.
      • Tuziak T.
      • Bondaruk J.
      • Lee S.
      • Brown G.A.
      • Adam L.
      • Wei C.
      • Baggerly K.
      • Bar-Eli M.
      • McConkey D.
      • Czerniak B.
      • Dinney C.P.
      Distinctive expression pattern of ErbB family receptors signifies an aggressive variant of bladder cancer.
      • Chow N.H.
      • Liu H.S.
      • Yang H.B.
      • Chan S.H.
      • Su I.J.
      Expression patterns of ErbB receptor family in normal urothelium and transitional cell carcinoma: an immunohistochemical study.
      cytoplasm,
      • Kassouf W.
      • Black P.C.
      • Tuziak T.
      • Bondaruk J.
      • Lee S.
      • Brown G.A.
      • Adam L.
      • Wei C.
      • Baggerly K.
      • Bar-Eli M.
      • McConkey D.
      • Czerniak B.
      • Dinney C.P.
      Distinctive expression pattern of ErbB family receptors signifies an aggressive variant of bladder cancer.
      • De Boer W.I.
      • Houtsmuller A.B.
      • Izadifar V.
      • Muscatelli-Groux B.
      • Van der Kwast T.H.
      • Chopin D.K.
      Expression and functions of EGF, FGF and TGFbeta-growth-factor family members and their receptors in invasive human transitional-cell-carcinoma cells.
      and nucleus.
      • Junttila T.T.
      • Laato M.
      • Vahlberg T.
      • Söderström K.O.
      • Visakorpi T.
      • Isola J.
      • Elenius K.
      Identification of patients with transitional cell carcinoma of the bladder overexpressing ErbB2, ErbB3, or specific ErbB4 isoforms: real-time reverse transcription-PCR analysis in estimation of ErbB receptor status from cancer patients.
      Membrane loss was related to stage, grade, size, and poor survival,
      • Kassouf W.
      • Black P.C.
      • Tuziak T.
      • Bondaruk J.
      • Lee S.
      • Brown G.A.
      • Adam L.
      • Wei C.
      • Baggerly K.
      • Bar-Eli M.
      • McConkey D.
      • Czerniak B.
      • Dinney C.P.
      Distinctive expression pattern of ErbB family receptors signifies an aggressive variant of bladder cancer.
      whereas nuclear expression correlated with advanced disease.
      • Junttila T.T.
      • Laato M.
      • Vahlberg T.
      • Söderström K.O.
      • Visakorpi T.
      • Isola J.
      • Elenius K.
      Identification of patients with transitional cell carcinoma of the bladder overexpressing ErbB2, ErbB3, or specific ErbB4 isoforms: real-time reverse transcription-PCR analysis in estimation of ErbB receptor status from cancer patients.
      Similarly, our results revealed that increased nuclear staining was also linked to poor outcomes, suggesting a potential translocation of ERBB4 within the cell. Our studies also established novel associations of cytoplasmic ERBB3 with tumor stage, grade, size, growth pattern, and outcome (recurrence and disease-specific and overall survival). ERBB3 previously sublocalized to the membrane
      • Rajkumar T.
      • Stamp G.W.
      • Pandha H.S.
      • Waxman J.
      • Gullick W.J.
      Expression of the type 1 tyrosine kinase growth factor receptors EGF receptor, c-erbB2 and c-erbB3 in bladder cancer.
      and the cytoplasm.
      • Chow N.H.
      • Liu H.S.
      • Yang H.B.
      • Chan S.H.
      • Su I.J.
      Expression patterns of ErbB receptor family in normal urothelium and transitional cell carcinoma: an immunohistochemical study.
      • Rajkumar T.
      • Stamp G.W.
      • Pandha H.S.
      • Waxman J.
      • Gullick W.J.
      Expression of the type 1 tyrosine kinase growth factor receptors EGF receptor, c-erbB2 and c-erbB3 in bladder cancer.
      Lack of associations with membranous expression was also previously observed for ERBB3
      • Chow N.H.
      • Liu H.S.
      • Yang H.B.
      • Chan S.H.
      • Su I.J.
      Expression patterns of ErbB receptor family in normal urothelium and transitional cell carcinoma: an immunohistochemical study.
      • Rajkumar T.
      • Stamp G.W.
      • Pandha H.S.
      • Waxman J.
      • Gullick W.J.
      Expression of the type 1 tyrosine kinase growth factor receptors EGF receptor, c-erbB2 and c-erbB3 in bladder cancer.
      and ERBB4.
      • Chow N.H.
      • Liu H.S.
      • Yang H.B.
      • Chan S.H.
      • Su I.J.
      Expression patterns of ErbB receptor family in normal urothelium and transitional cell carcinoma: an immunohistochemical study.
      • De Boer W.I.
      • Houtsmuller A.B.
      • Izadifar V.
      • Muscatelli-Groux B.
      • Van der Kwast T.H.
      • Chopin D.K.
      Expression and functions of EGF, FGF and TGFbeta-growth-factor family members and their receptors in invasive human transitional-cell-carcinoma cells.
      • Junttila T.T.
      • Laato M.
      • Vahlberg T.
      • Söderström K.O.
      • Visakorpi T.
      • Isola J.
      • Elenius K.
      Identification of patients with transitional cell carcinoma of the bladder overexpressing ErbB2, ErbB3, or specific ErbB4 isoforms: real-time reverse transcription-PCR analysis in estimation of ErbB receptor status from cancer patients.
      Although translocation mechanisms of ERBB3 and ERBB4 in bladder cancer progression remain to be clarified, our results highlighted the clinical relevance of the proteins' cellular sublocalization for outcome prognosis.
      In the context of a prospective study, the follow-up ranged up to 3 years, as shown in the Kaplan-Meier curves. Based on the natural history of bladder cancer, many patients are expected to be alive and, thus, censored at the follow-up frame of this study. The distribution of censored patients varied depending on the miRNA and protein under analyses. To further highlight the statistical clinical outcome discrimination power of the measurements undertaken, these miRNAs and proteins served to identify patients with a poorer outcome, even in the presence of many censored alive patients. This discrimination would be expected to increase in the presence of fewer censored patients and a longer follow-up.
      In addition to the utility of miRNA expression assessment for tumor classification, our study suggested a potential diagnostic adjunct role of urinary miRNAs. The diagnostic accuracies of miR-452 (85%) and miR-222 (77%) should be considered relevant because cystoscopy as a gold standard remains invasive and relatively expensive for disease follow-up, and urinary cytological analysis provides overall diagnostic accuracies of 40% while failing in low-grade tumors. Our study suggests that, in the near future, novel miRNA-based panels might be explored for bladder cancer diagnostics. It is important to be aware that the amount of exfoliated cancer cells could be a limitation in the extraction of an adequate amount of mRNA from urinary specimens.
      In conclusion, urinary miRNAs were clinically useful for noninvasive bladder cancer diagnostics (miR-452 and miR-222) and tumor stratification and outcome assessment using tumor samples (miR-222 and miR-143). Protein expression profiles of targets (ERBB3 and ERBB4) potentially regulated by these miRNAs correlated with their expression, clinicopathologic correlates of tumor progression, and clinical outcome.

      Acknowledgments

      We thank all members of the Tumor Markers Group at the Centro Nacional de Investigaciones Oncológicas for their technical support and constructive suggestions in the preparation of the manuscript.

      Supplementary data

      • Supplemental Figure S1

        Experimental design. The present report was based on five major sets of experiments. A: miRNA expression levels in urinary samples (n = 37) belonging to patients with bladder cancer were analyzed using RT-qPCR. B: In situ hybridization of selected miRNAs was performed in matching tumor samples of those urinary specimens used for the miRNA arrays to evaluate the cell type localization of miRNAs. RT-qPCR was performed using an independent set of urinary samples (n = 37) and matching bladder tumors to evaluate whether urinary miRNA expression levels reflect and are associated with miRNA expression in tumor specimens. C: RT-qPCR of selected miRNAs was performed in a series of tissue arrays (n = 113) with available clinicopathologic information to evaluate the association of selected miRNA profiles with tumor staging and clinical outcome. D: Protein expression patterns of predicted miRNA targets were evaluated using IHC in a cohort of well-characterized and follow-up annotated cases (n = 164) from the TMAs in which miRNA profiles were assessed. These analyses served to evaluate the association of miRNA expression levels with the predicted target expression and the clinical significance of such potential targets at the microanatomical protein level. E: miRNA profiles, measured by RT-qPCR in an independent series of urinary specimens (n = 94) belonging to patients with bladder cancer and controls, were analyzed to determine utility in bladder cancer diagnostics.

      • Supplemental Figure S2

        Comparative quantitation of the candidate miRNAs through RT-qPCR in different tumor samples for the high-grade pT1 bladder tumor shown in Figure 1 (A and B) and for a patient with a T2+ bladder tumor (C and D). A and C: Variations of the number of cycles (x axis) for one specific run (y axis) showing each triplicate measurement for the four miRNAs under analysis (each miRNA is shown in different colors, and each triplicate is shown in the same color). B and D: Quantitative comparison of the mean expression of the CT raw data for each miRNA before normalization.

      • Supplemental Figure S3

        miRNA localization in normal urothelium by in situ hybridization. Representative staining in the muscularis stroma for miR-143 (A) and in the uroepithelium and lamina propria for miR-143 (B), miR-222 (C), and miR-452 (D). miRNA locked nucleic acid probes were labeled with FITC (green) and counterstained with DAPI (blue). Original magnification, ×10.

      • Supplemental Figure S4

        Negative and positive controls of miRNA in situ hybridization in various tissues. Representative staining of miR-143 in normal amygdala as a negative control (A) and in Sertoli cells in normal testes as a positive control (B and C), miR-222 in normal kidney as a negative control (D) and in tumor cells in hepatocellular carcinoma as a positive control (E and F), and miR-452 in normal pancreas as a negative control (G) and ductal and acinar cells in normal prostate as a positive control (H and I). miRNA locked nucleic acid probes were labeled with FITC (green) and counterstained with DAPI (blue). Original magnification: ×40 (A, B, D, E, G, and H); ×400 (C, F, and I).

      References

        • Gottardo F.
        • Liu C.G.
        • Ferracin M.
        • Calin G.A.
        • Fassan M.
        • Bassi P.
        • Sevignani C.
        • Byrne D.
        • Negrini M.
        • Pagano F.
        • Gomella L.G.
        • Croce C.M.
        • Baffa R.
        Micro-RNA profiling in kidney and bladder cancers.
        Urol Oncol. 2007; 25: 387-392
        • Yang H.
        • Dinney C.P.
        • Ye Y.
        • Zhu Y.
        • Grossman H.B.
        • Wu X.
        Evaluation of genetic variants in microRNA-related genes and risk of bladder cancer.
        Cancer Res. 2008; 68: 2530-2537
        • Neely L.A.
        • Rieger-Christ K.M.
        • Neto B.S.
        • Eroshkin A.
        • Garver J.
        • Patel S.
        • Phung N.A.
        • McLaughlin S.
        • Libertino J.A.
        • Whitney D.
        • Summerhayes I.C.
        A microRNA expression ratio defining the invasive phenotype in bladder tumors.
        Urol Oncol. 2010; 28: 39-48
        • Veerla S.
        • Lindgren D.
        • Kvist A.
        • Frigyesi A.
        • Staaf J.
        • Persson H.
        • Liedberg F.
        • Chebil G.
        • Gudjonsson S.
        • Borg A.
        • Månsson W.
        • Rovira C.
        • Höglund M.
        miRNA expression in urothelial carcinomas: important roles of miR-10a, miR-222, miR-125b, miR-7 and miR-452 for tumor stage and metastasis, and frequent homozygous losses of miR-31.
        Int J Cancer. 2009; 124: 2236-2242
        • Lin T.
        • Dong W.
        • Huang J.
        • Pan Q.
        • Fan X.
        • Zhang C.
        • Huang L.
        MicroRNA-143 as a tumor suppressor for bladder cancer.
        J Urol. 2009; 181: 1372-1380
        • Hanke M.
        • Hoefing K.
        • Merz H.
        • Feller A.C.
        • Kausch I.
        • Jocham D.
        • Warnecke J.M.
        • Sczakiel G.
        A robust methodology to study urine microRNA as tumor marker: microRNA-126 and microRNA-182 are related to urinary bladder cancer.
        Urol Oncol. 2010; 28: 655-661
        • Ichimi T.
        • Enokida H.
        • Okuno Y.
        • Kunimoto R.
        • Chiyomaru T.
        • Kawamoto K.
        • Kawahara K.
        • Toki K.
        • Kawakami K.
        • Nishiyama K.
        • Tsujimoto G.
        • Nakagawa M.
        • Seki N.
        Identification of novel miRNA targets based on microRNA signatures in bladder cancer.
        Int J Cancer. 2009; 125: 345-352
        • Baffa R.
        • Fassan M.
        • Volinia S.
        • O'Hara B.
        • Liu C.G.
        • Palazzo J.P.
        • Gardiman M.
        • Rugge M.
        • Gomella L.G.
        • Croce C.M.
        • Rosenberg A.
        MicroRNA expression profiling of human metastatic cancers identifies cancer gene targets.
        J Pathol. 2009; 219: 214-221
        • Wang G.
        • Zhang H.
        • He H.
        • Tong W.
        • Wang B.
        • Liao G.
        • Chen Z.
        • Du C.
        Up-regulation of microRNA in bladder tumor tissue is not common.
        Int Urol Nephrol. 2010; 42: 95-102
        • Dyrskjøt L.
        • Ostenfeld M.S.
        • Bramsen J.B.
        • Silahtaroglu A.N.
        • Lamy P.
        • Ramanathan R.
        • Fristrup N.
        • Jensen J.L.
        • Andersen C.L.
        • Zieger K.
        • Kauppinen S.
        • Ulhøi B.P.
        • Kjems J.
        • Borre M.
        • Orntoft T.F.
        Genomic profiling of miRNAs in bladder cancer: miR-129 is associated with poor outcome and promotes cell death in vitro.
        Cancer Res. 2009; 69: 4851-4860
        • Adam L.
        • Zhong M.
        • Choi W.
        • Qi W.
        • Nicoloso M.
        • Arora A.
        • Calin G.
        • Wang H.
        • Siefker-Radtke A.
        • McConkey D.
        • Bar-Eli M.
        • Dinney C.
        miR-200 expression regulates epithelial-to-mesenchymal transition in bladder cancer cells and reverses resistance to epidermal growth factor receptor therapy.
        Clin Cancer Res. 2009; 15: 5060-5072
        • Catto J.F.
        • Miah S.
        • Owen H.C.
        • Bryant H.
        • Myers K.
        • Dudziec E.
        • Larré S.
        • Milo M.
        • Rehman I.
        • Rosario D.J.
        • Di Martino E.
        • Knowles M.A.
        • Meuth M.
        • Harris A.L.
        • Hamdy F.C.
        Distinct microRNA alterations characterize high- and low-grade bladder cancer.
        Cancer Res. 2009; 69: 8472-8481
        • Ostenfeld M.S.
        • Bramsen J.B.
        • Lamy P.
        • Villadsen S.B.
        • Fristrup N.
        • Sørensen K.D.
        • Ulhøi B.
        • Borre M.
        • Kjems J.
        • Dyrskjøt L.
        • Orntoft T.F.
        miR-145 induces caspase-dependant and -independent cell death in urothelial cancer cell lines with targeting of an expression signature present in Ta bladder tumors.
        Oncogene. 2010; 29: 1073-1084
        • Chiyomaru T.
        • Enokida H.
        • Tatarano S.
        • Kawahara K.
        • Uchida Y.
        • Nishiyama K.
        • Fujimura L.
        • Kikkawa N.
        • Seki N.
        • Nakagawa M.
        miR-145 and miR-133a function as tumor suppressors and directly regulate FSCN1 expression in bladder cancer.
        Br J Cancer. 2010; 102: 883-891
        • Wiklund E.D.
        • Bramsen J.B.
        • Hulf T.
        • Dyrskjøt L.
        • Ramanathan R.
        • Hansen T.B.
        • Villadsen S.B.
        • Gao S.
        • Ostenfeld M.S.
        • Borre M.
        • Peter M.E.
        • Ørntoft T.F.
        • Kjems J.
        • Clark S.J.
        Coordinated epigenetic repression of the miR-200 family and miR-205 in invasive bladder cancer.
        Int J Cancer. 2011; 128: 1327-1334
        • Han Y.
        • Chen J.
        • Zhao X.
        • Liang C.
        • Wang Y.
        • Sun L.
        • Jiang Z.
        • Zhang Z.
        • Yang R.
        • Chen J.
        • Li Z.
        • Tang A.
        • Li X.
        • Ye J.
        • Guan Z.
        • Gui Y.
        • Cai Z.
        MicroRNA expression signatures of bladder cancer revealed by deep sequencing.
        PLoS One. 2011; 6: e18286
        • Noguchi S.
        • Mori T.
        • Hoshino Y.
        • Maruo K.
        • Yamada N.
        • Kitade Y.
        • Naoe T.
        • Akao Y.
        MicroRNA-143 functions as a tumor suppressor in human bladder cancer T24 cells.
        Cancer Lett. 2011; 307: 211-220
        • Kirkali Z.
        • Chan T.
        • Manoharan M.
        • Algaba F.
        • Busch C.
        • Cheng L.
        • Kiemeney L.
        • Kriegmair M.
        • Montironi R.
        • Murphy W.M.
        • Sesterhenn I.A.
        • Tachibana M.
        • Weider J.
        Bladder cancer: epidemiology, staging and grading, and diagnosis.
        Urology. 2005; 66: 4-34
        • Sanchez-Carbayo M.
        • Schwarz K.
        • Charytonowicz E.
        • Cordon-Cardo C.
        • Mundel P.
        Tumor suppressor role for myopodin in bladder cancer: loss of nuclear expression of myopodin is cell-cycle dependent and predicts clinical outcome.
        Oncogene. 2003; 22: 5298-5305
        • Pena J.T.G.
        • Sohn-Lee C.
        • Rouhanifard S.H.
        • Ludwig J.
        • Hafner M.
        • Mihailovic A.
        • Lim C.
        • Holoch D.
        • Berninger P.
        • Zavolan M.
        • Tuschl T.
        miRNA in situ hybridization in mammalian tissues fixed with formaldehyde and EDC.
        Nat Methods. 2009; 6: 139-141
        • Jørgensen S.
        • Baker A.
        • Møller S.
        • Nielsen B.S.
        Robust one-day in situ hybridization protocol for detection of microRNAs in paraffin samples using LNA probes.
        Methods. 2010; 52: 375-381
        • Catto J.W.
        • Alcaraz A.
        • Bjartell A.S.
        • De Vere White R.
        • Evans C.P.
        • Fussel S.
        • Hamdy F.C.
        • Kallioniemi O.
        • Mengual L.
        • Schlomm T.
        • Visakorpi T.
        MicroRNA in prostate, bladder, and kidney cancer: a systematic review.
        Eur Urol. 2011; 59: 671-681
        • Tsai Z.Y.
        • Singh S.
        • Yu S.L.
        • Kao L.P.
        • Chen B.Z.
        • Ho B.C.
        • Yang P.C.
        • Li S.S.
        Identification of microRNAs regulated by activin A in human embryonic stem cells.
        J Cell Biochem. 2010; 109: 93-102
        • Schmidt W.M.
        • Spiel A.O.
        • Jilma B.
        • Wolzt M.
        • Müller M.
        In vivo profile of the human leukocyte microRNA response to endotoxemia.
        Biochem Biophys Res Commun. 2009; 380: 437-441
        • Chen H.C.
        • Chen G.H.
        • Chen Y.H.
        • Liao W.L.
        • Liu C.Y.
        • Chang K.P.
        • Chang Y.S.
        • Chen S.J.
        MicroRNA deregulation and pathway alterations in nasopharyngeal carcinoma.
        Br J Cancer. 2009; 100: 1002-1011
        • Kassouf W.
        • Black P.C.
        • Tuziak T.
        • Bondaruk J.
        • Lee S.
        • Brown G.A.
        • Adam L.
        • Wei C.
        • Baggerly K.
        • Bar-Eli M.
        • McConkey D.
        • Czerniak B.
        • Dinney C.P.
        Distinctive expression pattern of ErbB family receptors signifies an aggressive variant of bladder cancer.
        J Urol. 2008; 179: 353-358
        • Chow N.H.
        • Liu H.S.
        • Yang H.B.
        • Chan S.H.
        • Su I.J.
        Expression patterns of ErbB receptor family in normal urothelium and transitional cell carcinoma: an immunohistochemical study.
        Virchows Arch. 1997; 430: 461-466
        • De Boer W.I.
        • Houtsmuller A.B.
        • Izadifar V.
        • Muscatelli-Groux B.
        • Van der Kwast T.H.
        • Chopin D.K.
        Expression and functions of EGF, FGF and TGFbeta-growth-factor family members and their receptors in invasive human transitional-cell-carcinoma cells.
        Int J Cancer. 1997; 71: 284-291
        • Junttila T.T.
        • Laato M.
        • Vahlberg T.
        • Söderström K.O.
        • Visakorpi T.
        • Isola J.
        • Elenius K.
        Identification of patients with transitional cell carcinoma of the bladder overexpressing ErbB2, ErbB3, or specific ErbB4 isoforms: real-time reverse transcription-PCR analysis in estimation of ErbB receptor status from cancer patients.
        Clin Cancer Res. 2003; 9: 5346-5357
        • Rajkumar T.
        • Stamp G.W.
        • Pandha H.S.
        • Waxman J.
        • Gullick W.J.
        Expression of the type 1 tyrosine kinase growth factor receptors EGF receptor, c-erbB2 and c-erbB3 in bladder cancer.
        J Pathol. 1996; 179: 381-385

      Linked Article

      • This Month in AJP
        The American Journal of PathologyVol. 180Issue 5
        • Preview
          The following highlights summarize research articles that are published in the current issue of The American Journal of Pathology.
        • Full-Text
        • PDF
        Open Access