Urological Oncology 99Urology Journal Vol 5 No 2 Spring 2008 Overexpression of BMI1, a Polycomb Group Repressor Protein, in Bladder Tumors A Preliminary Report Afsaneh Malekzadeh Shafaroudi,1 Seyed Javad Mowla,1 Seyed Amir-Mohsen Ziaee,2 Ahmad-Reza Bahrami,3 Yaser Atlasi,1 Mahshid Malakootian1 Introduction: A Polycomb group repressor protein named BMI1 represses the genes that induce cellular senescence and cell death, and it can contribute to cancer when improperly expressed. We aimed to evaluate expression of BMI1 gene in bladder tumors. Materials and Methods: Tissue specimens containing bladder tumor were evaluated and compared with intact tissues from tumor margins and normal bladders. There were 40 tumor specimens of patients with transitional cell carcinoma of the bladder, 20 tumor-free tissues taken from the margin of the tumors, and 8 specimens from patients without tumor. Specific primers for BMI1 and B2M (as an internal control) were used for reverse transcript polymerase chain reaction technique. The production and distribution of BMI1 protein was also examined by western blotting and immunohistochemistry techniques. Results: Polymerase chain reaction generated a 683-bp product, corresponding to the expected size of BMI1 amplified region. The identity of the amplified fragment was then confirmed by direct DNA sequencing. The mean of expression of BMI1 detected in tumor tissues was significantly higher than that in intact tissues, and there was also a significant association between the mean of gene expression and the stage of malignancy (P < .001). The expression of BMI1 at protein level was further confirmed by western blotting and immunohistochemistry. Conclusion: BMI1 is a potent repressor of retinoblastoma and p53 pathways, and hence, elucidating its role in tumorigenesis is very important. We reported for the first time the expression of BMI1 and its correlation with incidence and progress of bladder tumors. Urol J. 2008;5:99-105. www.uj.unrc.ir Keywords: bladder neoplasms, tumor markers, BMI1, gene expression 1Department of Genetics, Faculty of Basic Science, Tarbiat Modares University, Tehran, Iran 2Urology and Nephrology Research Center and Shahid Labbafinejad Medical Center, Shahid Beheshti University (MC), Tehran, Iran 3Institute of Biotechnology, Ferdowsi University of Mashhad, Mashhad, Iran Corresponding Author: Seyed Javad Mowla, PhD Department of Genetics, Faculty of Basic Sciences, Tarbiat Modares University, PO Box: 1411-175 Tehran, Iran Tel: +98 21 8288 3464 Fax: +98 21 8288 3463 E-mail: sjmowla@modares.ac.ir Received December 2007 Accepted March 2008 INTRODUCTION Our understanding of the tumor biology has improved measures to prevent progression of superficial tumors to advanced and metastatic disease.(1) In bladder cancer for instance, we know that genetic alterations may occur at early stages and that they are retained in the recurrent tumors. Consequently, information on changes in the cell-cycle regulatory genes at early diagnosis may be useful for predicting tumor progression and recurrence. Today, cancer studies are focused on the identification of molecular markers that would predict which superficial bladder tumors will progress to invasive forms. However, our knowledge of predictive markers has not reached their reliable usage in practice.(1) The human BMI1 is located on BMI1 in Bladder Tumors—Malekzadeh Shafaroudi et al 100 Urology Journal Vol 5 No 2 Spring 2008 chromosome 10p13, a region known to be involved in translocations in various leukemias and rearrangements in malignant T-cell lymphomas.(2-4) It is responsible for expression of BMI1 protein, a member of the polycomb group (PcG) proteins which form multiprotein complexes that function as transcriptional repressors.(5) Bmi1 was first identified as an oncogene that cooperates with c-myc in the generation of mouse pre-B-cell lymphomas.(6,7) The oncogenic activity of BMI1 may be linked to its another fundamental function; several recent reports showed that BMI1 is essential for the self-renewal of both hematopoietic and neuronal stem cells, as well as cancer stem cells.(8-11) This function of BMI1 depends on its ability to repress the INK4A/ARF locus. Along with a role of BMI1 in stem-cell biology, BMI1 expression in the bone marrow is strong in undifferentiated precursor cells, but it gradually declines in the course of differentiation.(12) The BMI1 gene is amplified in certain mantle-cell lymphomas and is overexpressed in a subset of non-small-cell lung cancer and colorectal carcinomas.(13-15) Moreover, it plays a role in the development of human breast cancer and increases in metastatic prostate cancer.(16-19) Despite many studies on the potential involvement of BMI1 in the oncogenesis of various lymphomas and leukemias, little is known about its role in the pathogenesis of solid tumors, including urothelial carcinomas. To our knowledge, the only available report is that of Glinsky and colleagues who employed a genomics approach to identify an 11-gene, including BMI1, signature that consistently displays a stem-cell- resembling expression profile in distant metastatic lesions of different cancers, including prostate and bladder cancers.(20) To investigate the potential involvement of BMI1 in human bladder cancer, we examined the expression of BMI1 in the bladder specimens with and without malignant lesions. MATERIALS AND METHODS Human Clinical Specimens Fresh tissue biopsies were obtained from patients who were referred to Shahid Labbafinejad Medical Center. The tissues were immediately snap-frozen in liquid nitrogen and categorized into 3 groups: 40 tumor specimens prepared by transurethral resection from patients with transitional cell carcinoma of the bladder (Table), 20 tumor-free tissues taken from the margin of tumors (cystoscopically normal appearance), and 8 specimens from patients with no symptoms and signs of bladder cancer who had undergone surgical treatment for benign prostatic hyperplasia. Histopathological parameters were evaluated according to the grading and TNM system for stage classification of the World Health Organization. The experimental design was approved by the Ethics Committees of Tarbiat Modares University and the Urology and Nephrology Research Center of Shahid Beheshti University (MC). The patients’ written informed consents were obtained prior to participation. RNA Extraction Total RNA was isolated from frozen tissues using the RNX plus solution (Cinnagen, Tehran, Iran) according to the manufacturer’s instructions. The quality of RNA was evaluated by gel electrophoresis, and the concentration of RNA was measured by optical density at 260 nm. Semiquantitative Polymerase Chain Reaction Two micrograms of the total RNA was used for cDNA synthesis using random hexamer primer (Fermentas, Vilnius, Lithuania) and RevertAid MMuLV Reverse Transcriptase (Fermentas, Vilnius, Lithuania) in a 20 μL reaction according Characteristics Values Patients number 40 Mean age, y 65.4 ± 11.1 Age range, y 33 to 84 Sex Male 36 (90.0) Female 4 (10.0) Stage Ta 7 (17.5) T1 24 (60.0 T2 9 (22.5) Grade Low 25 (62.5) High 15 (37.5) Characteristics of Patients With Bladder Cancer* *Values in parentheses are percents. BMI1 in Bladder Tumors—Malekzadeh Shafaroudi et al Urology Journal Vol 5 No 2 Spring 2008 101 to the manufacturer’s instructions. Reverse transcription polymerase chain reaction (PCR) primers were designed using previously described human BMI1 and β2-microglobulin (B2M) sequences (GenBank accession numbers: NM_005180 and NM_004048, respectively). The appropriate PCR primers were designed using Genrunner software (version 3.02; Hastings Software, New York, USA) as follows: BMI1F: 5’-GAG GGT ACT TCA TTG ATG CCA C-3’ BMI1R: 5’-CCA GTT CTC CAG CAT TTG TCA G-3’ B2MF: 5’-CTA CTC TCT CTT TCT GGC CTG-3’ B2MR: 5’-GAC AAG TCT GAA TGC TCC AC-3’ Polymerase chain reaction amplifications were performed using 2 μL of cDNA with 1 U of Taq polymerase (Cinnagen, Tehran, Iran), 1.5 mM of MgCl2, 200 μM of dNTPs, and 0.4 μM of each primer in a 25-μL PCR reaction. The PCR cycling conditions were as follows: initial denaturation at 94 °C for 4 minutes and following 38 (BMI1) or 32 (B2M) cycles of 94°C for 30 seconds, 59°C (for both BMI1 and B2M) for 40 seconds, 72°C for 45 seconds, with a final extension at 72°C for 10 minutes. The primers amplified 683-bp and 191-bp fragments for BMI1 and B2M cDNA, respectively. The PCR products were separated on a 1% agarose gel, stained with ethidium bromide, and visualized under the ultraviolet light. The intensity of bands was determined using Uvitec software (Uvitec, Cambridge, UK). The identity of PCR products was confirmed by direct DNA sequencing (Millegen, Toulouse, France). Western Blotting Frozen tissue samples were homogenized and lysed in modified RIPA buffer (Tris-HCl, 50 mM, pH 7.4; NaCl, 150 mM; phenylmethylsulfonyl fluoride, 1 mM; ethylenediamine tetra-acetic acid, 1 mM; Triton X-100, 1%; sodium deoxycholate, 1%; sodium dodecyl sulfate, 0.1%; trypsin inhibitor, 10 μg/mL). The concentration of proteins in cell lysates was quantified by means of Bradford assay, and 20 μg of total protein was loaded in each lane. Samples were electrophoresed using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (12.5%) and blotted for 2 hours onto Hyband-P polyvinylidene difluoride membrane (Amersham Biosciences Europe GmbH, Freiburg, Germany). Membranes were then blocked for 2 hours in ECL advance blocking solution (Amersham Biosciences, Piscataway, New Jersey, USA), according to the manufacturer’s instructions. Blots were incubated with the anti-Bmi1 antibody for 3 hours and anti-β-actin antibody for 1 hour, and then with the secondary antibodies for 1 hour in room temperature, before being visualized by ECL Advance Western Blotting detection kit (Amersham Biosciences, Piscataway, New Jersey, USA). Anti-Bmi1 antibody (mouse monoclonal antibody to Bmi1 (ab14389) and anti-β-actin antibody (Prosci, Poway, California, USA) were used at dilutions 1:1000, and horse raddish peroxidase-conjugated antimouse IgG A4416 (Dako, Glostrup, Denmark) and horse- radish peroxidase-conjugated antirabbit IgG (Sigma, St Louis, Missouri, USA) were used at 1:42000 dilution. All antibodies were diluted in ECL Advance Blocking solution (Amersham Biosciences, Piscataway, New Jersey, USA), according to the manufacturer’s instructions. Immunohistochemistry Formalin-fixed paraffin-embedded tissue sections (5 μm) were deparaffinized with xylene, rehydrated in descending concentrations of ethanol, and boiled for 15 minutes in citrate buffer (10 mM, pH 6.0). Endogenous peroxidase activity was suppressed with 3% hydrogen peroxidase for 20 minutes. Slides were serum blocked and incubated with mouse monoclonal antibody to Bmi1 (ab14389, 1:200 dilution) for 2 hours at room temperature followed by staining with secondary horse raddish peroxidase- conjugated antimouse antibody (A4416, 1:200 dilution). In negative controls, all the conditions were kept the same, except that the first antibody was eliminated. Statistical Analyses All experiments were replicated 2 or 3 times, and the results were analyzed by performing analysis of variance test to determine the relative intensity of BMI1 expression among different biopsy groups. Also, the least significant difference test BMI1 in Bladder Tumors—Malekzadeh Shafaroudi et al 102 Urology Journal Vol 5 No 2 Spring 2008 was used to study the differences between pairs of tumor stages. The SPSS software (Statistical Package for the Social Sciences, version 15.0, SPSS Inc, Chicago, Illinois, USA) was used for statistical analyses. RESULTS Expression of BMI1 We collected a total number of 68 specimens of tumoral tissue, margin of tumor, and apparently normal bladder tissue. To make sure of using the same amount of RNA for each PCR reaction, B2M gene expression was employed as an internal control. For each sample, the reverse transcript PCR reaction was carried out for both B2M and BMI1 genes, in separate tubes and under similar conditions (except for the number of cycles). Furthermore, in each reaction, a tube with no cDNA was used as a negative control. We determined the relative expression of BMI1 in 40 tumors, 20 tumor margins from the same patients, and 8 apparently normal tissue samples of the bladder. An expected 191-bp PCR product corresponding to amplified B2M segment was visualized in all of the examined samples (Figure 1). The designed primers for BMI1 amplified an expected 683-bp segment in most of the samples. The accuracy of the amplified products was further confirmed by DNA sequencing. We detected the expression of BMI1 in 21 of 40 (52.5%) examined tumoral specimens of the bladder. The expression was also detected in 5 of 20 (25.0%) tumor margin specimens, as well as in 2 of 8 (25.0%) apparently normal specimens. However, the intensity of expression was much higher in tumoral tissues compared to the other ones (Figure 1). Because we have used a semiquantitative reverse transcript PCR approach, a densitometric evaluation and comparison of the relative expression of BMI1/B2M between different tissue samples was feasible. The intensity of BMI1 expression was significantly higher in neoplastic tissues compared to the specimens with no neoplastic changes (P = .004; Figure 2). Then, we examined the potential correlation between BMI1 expression and the clinicopathological features (eg, tumor grade and stage) of the patients. Despite the fact that the expression level of BMI1 in high-grade tumors was more than that of the low-grade ones, the difference was not significant (Figure 2). Furthermore, the intensity of BMI1 expression was significantly lower in stage Ta malignant tumors compared to that in tumors with higher stages of T1 or T2 (P = .03; Figure 2). Expression of BMI1 at Protein Level We employed western blotting and immunohistochemical techniques to confirm the expression of BMI1 protein in bladder tumors and also to determine its tissue distribution and subcellular localization. The western blot data showed a single band of ~45 kDa in tumor specimens, corresponding to the expected size of the protein (Figure 3). We also used the embryonic carcinoma cell line, NTERA2 (NT2), as a negative control. Based on our reverse transcript PCR results, there was no expression of BMI1 in the cell line (data not shown). As it is evident in Figure 3, there was no signal relevant to BMI1 protein expression in the cells. To determine the tissue distribution and Figure 1. Reverse transcript polymerase chain reaction analysis of the expression of BMI1 and B2M (as an internal control) in the bladder tissues obtained from tumors (T) and the margin of tumors (M) from the same patients (numbers). BMI1 in Bladder Tumors—Malekzadeh Shafaroudi et al Urology Journal Vol 5 No 2 Spring 2008 103 subcellular localization of BMI1 in bladder tumors, formalin-fixed paraffin-embedded blocks were collected from archival collections of Shahid Labbafinejad Medical Center and 5-μm tissue sections prepared from the samples. The immunohistochemical data revealed that the BMI1 protein is primarily localized in the nuclei of the tumor cells (Figure 4). To asses the specificity of the antibody, a negative control slide was accompanied each immunohistochemical processing (Figure 4). Similar conditions were used for both positive and negative slides, except for the omission of the first antibody in the negative slides. DISCUSSION The cell cycle is a complex process in which many molecules are involved. Among these molecules, inhibitors of cyclin-dependent kinase p16INK4a and p14ARF (coded from CDKN2a locus) correlates inversely with the markers of eukaryotic cell proliferation pRB and p53, respectively.(21) The unique genomic structure and compact organization of these genes, which have common reading frames, may be essential for maintaining a balanced Rb and p53 pathway function.(22) In terms of cancer, the INK4A/ARF locus which is negatively regulated by BMI1 is also a frequent target for mutations, deletions, and epigenetic silencing in a wide spectrum of human tumors.(23-25) This raises the possibility that transcriptional regulators of the locus may also be involved in cancer progression. In line with this prediction, it has recently been shown that Figure 2. Relative (B2M-normalized) intensities of BMI1 expression (mean + standard error). Top, Tumoral and intact bladder tissues. Middle, Low-grade and high-grade bladder tumors. Bottom, Different stages of bladder tumors. Relative band intensities for BMI1 for each sample were quantitated by densitometry, normalized to B2M expression. Statistical analysis revealed that the expression of BMI1 is significantly higher in tumor specimens compared to that in normal tissues and also in stage T1 and T2 compared to Ta (P < .001). Figure 3. Western blot analysis of BMI1 protein expression in the bladder tissues. Total proteins were isolated from NT2 cell line (that showed no expression in reverse transcript polymerase chain reaction) and 2 representative tumor (T1 and T2) bladder biopsies. The experiments confirmed the expression of an approximately 45 kDa form of BMI1 protein in the bladder tissues. The expression of β-actin was used as a loading control. BMI1 in Bladder Tumors—Malekzadeh Shafaroudi et al 104 Urology Journal Vol 5 No 2 Spring 2008 BMI1 is amplified in some hematologic disorders, such as mantle-cell and non-Hodgkin lymphomas, and is also overexpressed in solid tumors such as non-small-cell lung cancer, medulloblastoma, colorectal cancer, breast cancer, and prostate cancer.(13-19) In this study, the suitability of the BMI1 gene expression was evaluated as a potential molecular marker in diagnosis and molecular classification of bladder tumors. BMI1 is a transcriptional repressor that involves in many cellular mechanisms including tumorigenesis. Despite several reports demonstrating overexpression of BMI1 in a series of cancers,(13-19) to date, there is no evidence on potential involvement and causative role of BMI1 in bladder cancer. The main goal of this study was to determine and compare the relative expression of BMI1 gene in neoplastic tissue versus intact tissue of bladder cancer and its potential involvement in induction and progression of bladder cancer. Our data revealed a differential expression of BMI1 in tumoral tissues versus apparently normal tissues and tumor margins of bladder cancer. Furthermore, the data suggests that BMI1 may play a role in bladder tumor progression rather than initiating the process of tumorigenesis. Based on our data, the expression of the gene is not a good indicator for early detection of bladder tumors (eg, at Ta stage). However, it could be employed to distinguish Ta lesions, which are papillary in nature and limited to the mucosa from T1 lesions, which invade the submucosa or the lamina propria, and hence, are more aggressive.(26) Despite overexpression of BMI1 in high-grade tumors compared to the low-grade ones, the difference was not statistically significant in our series. This is probably due to the small size of the patients’ population. The data is also in accordance with our previous report on the expression of OCT4 in bladder cancer,(27) where we found no significant correlation between the expression level of OCT4 and the grade of the tumors. CONCLUSION The present study suggests that alteration of the BMI1 might play a role in the development and progression of bladder cancer. Furthermore, the current study provides some novel data that further elucidate the complex biology of bladder tumor cells and could potentially be used in diagnosis, prognosis, and treatment of bladder cancer. In summary our data has revealed for the first time that (1) the relative expression of BMI1 in tumor tissues is much higher than the nontumor samples (~5 times more), and (2) there is a statistically significant correlation between the level of BMI1 expression and the stage of the bladder tumors. Accordingly, the relative amount of BMI1 expression in stage Ta is significantly lower than that in stages T1 and T2. In other words, the BMI1 overexpression is not a primary event in the genetics of bladder cancer, and probably, the gene is involved in the progression of the tumor. This finding is in accordance with previous reports showing a gain and amplification of 10 p (the BMI1 locus) in stages T1 and T2, but not in stage Ta of bladder cancer.(16) Figure 4. Representative immunohistochemical data of BMI1 protein expression in formalin-fixed, paraffin-embedded sections of tumor tissues. Bladder sections were deparaffinized and subsequently incubated with BMI1 antibody and avidin- horse-radish peroxidase before being visualized with diaminobenzidine. Top, BMI1 expression was primarily localized in the nuclei (in brown color, white arrows) of the cells. Bottom, The negative control, sections from the same specimen was identically processed, except for the omission of the first BMI1 antibody. BMI1 in Bladder Tumors—Malekzadeh Shafaroudi et al Urology Journal Vol 5 No 2 Spring 2008 105 CONFLICT OF INTEREST None declared. ACKNOWLEDGEMENT We are grateful to Ms Nasim Hatefi for her valuable help and assistance. This work was financially supported by a research grant from Iranian Stem Cell Network. All the samples were provided by Shahid Labbafinejad Medical Center and the experiments were performed in Tarbiat Modares University. REFERENCES 1. Hussain SA, James ND. Molecular markers in bladder cancer. Semin Radiat Oncol. 2005;15:3-9. 2. Alkema MJ, Wiegant J, Raap AK, Berns A, van Lohuizen M. Characterization and chromosomal localization of the human proto-oncogene BMI-1. Hum Mol Genet. 1993;2:1597-603. 3. Pui CH, Raimondi SC, Murphy SB, et al. An analysis of leukemic cell chromosomal features in infants. Blood. 1987;69:1289-93. 4. Berger R, Baranger L, Bernheim A, Valensi F, Flandrin G. Cytogenetics of T-cell malignant lymphoma. Report of 17 cases and review of the chromosomal breakpoints. Cancer Genet Cytogenet. 1988;36:123- 30. 5. Orlando V. Polycomb, epigenomes, and control of cell identity. Cell. 2003;112:599-606. 6. Haupt Y, Alexander WS, Barri G, Klinken SP, Adams JM. Novel zinc finger gene implicated as myc collaborator by retrovirally accelerated lymphomagenesis in E mu-myc transgenic mice. Cell. 1991;65:753-63. 7. van Lohuizen M, Frasch M, Wientjens E, Berns A. Sequence similarity between the mammalian bmi-1 proto-oncogene and the Drosophila regulatory genes Psc and Su(z)2. Nature. 1991;353:353-5. 8. Molofsky AV, Pardal R, Iwashita T, Park IK, Clarke MF, Morrison SJ. Bmi-1 dependence distinguishes neural stem cell self-renewal from progenitor proliferation. Nature. 2003;425:962-7. 9. Lessard J, Sauvageau G. Bmi-1 determines the proliferative capacity of normal and leukaemic stem cells. Nature. 2003;423:255-60. 10. Park IK, Qian D, Kiel M, et al. Bmi-1 is required for maintenance of adult self-renewing haematopoietic stem cells. Nature. 2003;423:3011. Valk- Lingbeek ME, Bruggeman SW, van Lohuizen M. Stem cells and cancer; the polycomb connection. Cell. 2004;118:409-18. 12. Lessard J, Baban S, Sauvageau G. Stage-specific expression of polycomb group genes in human bone marrow cells. Blood. 1998;91:1216-24. 13. Bea S, Tort F, Pinyol M, et al. BMI-1 gene amplification and overexpression in hematological malignancies occur mainly in mantle cell lymphomas. Cancer Res. 2001;61:2409-12. 14. Vonlanthen S, Heighway J, Altermatt HJ, et al. The bmi-1 oncoprotein is differentially expressed in non- small cell lung cancer and correlates with INK4A-ARF locus expression. Br J Cancer. 2001;84:1372-6. 15. Kim JH, Yoon SY, Kim CN, et al. The Bmi-1 oncoprotein is overexpressed in human colorectal cancer and correlates with the reduced p16INK4a/ p14ARF proteins. Cancer Lett. 2004;203:217-24. 16. Dimri GP, Martinez JL, Jacobs JJ, et al. The Bmi-1 oncogene induces telomerase activity and immortalizes human mammary epithelial cells. Cancer Res. 2002;62:4736-45. 17. Kim JH, Yoon SY, Jeong SH, et al. Overexpression of Bmi-1 oncoprotein correlates with axillary lymph node metastases in invasive ductal breast cancer. Breast. 2004;13:383-8. 18. Berezovska OP, Glinskii AB, Yang Z, Li XM, Hoffman RM, Glinsky GV. Essential role for activation of the Polycomb group (PcG) protein chromatin silencing pathway in metastatic prostate cancer. Cell Cycle. 2006;5:1886-901. 19. van Leenders GJ, Dukers D, Hessels D, et al. Polycomb-group oncogenes EZH2, BMI1, and RING1 are overexpressed in prostate cancer with adverse pathologic and clinical features. Eur Urol. 2007;52:455-63. 20. Glinsky GV, Berezovska O, Glinskii AB. Microarray analysis identifies a death-from-cancer signature predicting therapy failure in patients with multiple types of cancer. J Clin Invest. 2005;115:1503-21. 21. Dai CY, Furth EE, Mick R, et al. p16(INK4a) expression begins early in human colon neoplasia and correlates inversely with markers of cell proliferation. Gastroenterology. 2000;119:929-42. 22. Kawada Y, Nakamura M, Ishida E, et al. Aberrations of the p14(ARF) and p16(INK4a) genes in renal cell carcinomas. Jpn J Cancer Res. 2001;92:1293-9. 23. Jacobs JJ, Scheijen B, Voncken JW, Kieboom K, Berns A, van Lohuizen M. Bmi-1 collaborates with c-Myc in tumorigenesis by inhibiting c-Myc-induced apoptosis via INK4a/ARF. Genes Dev. 1999;13:2678- 90. 24. Jacobs JJ, Kieboom K, Marino S, DePinho RA, van Lohuizen M. The oncogene and Polycomb-group gene bmi-1 regulates cell proliferation and senescence through the ink4a locus. Nature. 1999;397:164-8. 25. Lowe SW, Sherr CJ. Tumor suppression by Ink4a- Arf: progress and puzzles. Curr Opin Genet Dev. 2003;13:77-83. 26. Knowles MA. Molecular subtypes of bladder cancer: Jekyll and Hyde or chalk and cheese? Carcinogenesis. 2006;27:361-73. 27. Atlasi Y, Mowla SJ, Ziaee SA, Bahrami AR. OCT-4, an embryonic stem cell marker, is highly expressed in bladder cancer. Int J Cancer. 2007;120:1598-602.