According to the new restanding cancer stem cell (CSC) concept, only a small subset of cancer cells are thought to be tumorigenic and capable of self-renewing to generate additional CSCs as well as various nontumorigenic cells. Based on this school of thought, CSCs have high self-renewal capacity and can be generated via dysregulation of self-renewal process in normal stem or early progenitor cells.1, 2, 3, 4, 5
The POU homeodomain transcription factor, OCT-4 (also known as OCT-3, OCT-3/4 and POU5F1), is a key regulator of self-renewal and differentiation in embryonic stem cells.6, 7 Expression of OCT-4 is believed to be restricted to pluripotent cells, and its level decreases with the onset of differentiation and loss of pluripotency in these cells.8, 9OCT-4 is primarily expressed in early cleavage stage, inner cell mass, primitive ectoderm, primordial germ cells10 and also in embryonic stem (ES), embryonic germ and embryonic carcinoma cells.6 In 1992, a low level of expression of OCT-4 in human adult tissues was reported by Takeda et al.,11 but the significance of the finding remained unclear till a recent report by Tai et al.12 The latter report shows the expression of OCT-4 at mRNA and protein levels in several human tissue-specific adult stem cells.
Recently, it has been proposed that OCT-4 acts as a multifunctional factor in cancer and stem cell biology. Based on the reports that OCT-4 increases the malignant potential of ES cells in a dose-dependent manner,13 a possible oncogenic role was also attributed to OCT-4. On the other hand, ectopic expression of OCT-4 in epithelial tissues could lead to a dysplasic induction through inhibition of epithelial stem or progenitor cell differentiation. The effect seems very similar to the primary role of OCT-4 in embryonic cells.14
Although the expression of OCT-4 in germ cell tumors has been extensively studied,15 little is known about the expression in somatic cancers. Given the similarity between cancer cells and stem cells in one hand and the proposed role for dysregulated expression of self-renewal genes in cancer development on the other hand, the main aim of the present study was to explore the potential expression of OCT-4 in bladder cancer. Moreover, because OCT-4 plays an antidifferentiation role in normal stem cells, we also investigated whether OCT-4 expression has any correlation with the grade of tumors. To our knowledge, this study represents the first report of OCT-4 expression in bladder tissues, and our data suggest a strong correlation between the expression of OCT-4 and tumor state of the tissues.
Material and methods
Human clinical samples
Fresh tissue biopsies were obtained from patients who had been referred to Labbafi-Nejad Medical Center. The tissues were immediately snap-frozen in liquid nitrogen and categorized in 3 groups: 32 tumor samples prepared by transurethral resection from 32 patients with transitional cell carcinoma of the bladder (Group A), 13 nontumor tissues, which were taken from the margin of tumors (cystoscopically normal appearing, Group B) and 9 bladder samples from patients with no symptoms and signs of bladder cancer, receiving surgical treatments for benign prostatic hyperplasia (Group C). Histopathological parameters were evaluated according to WHO criteria for grade and TNM system for stage classification. The experimental design was approved by the Ethics Committees of Tarbiat Modares University and Urology–Nephrology Research Center, and the patients' written informed consent were collected prior to participation.
Human EC cells, NTERA2 cl.D1 (NT2/D1), were cultured in Dulbecco's modified Eagle medium supplemented with 10% fetal calf serum at 37°C and humidified 10% CO2 in air.
Total RNA was isolated from frozen tissues using the RNX plus™ solution (Cinnagen, Iran), according to the manufacturer's instructions and as described previously.16 The quality of RNA evaluated by gel electrophoresis and the concentration of RNA was measured by optical density at 260 nm.
One microgram of total RNA was treated with RNase-free DNase (Fermentase, Lithuania) and used for cDNA synthesis, using oligo(dT)18 primer (MWG, Germany) and RevertAid™ MMuLV Reverse Transcriptase (Fermentase) in a 20 μl reaction according to the manufacturer's instructions. For each sample, a No-RT control was used in parallel to detect any potential nonspecific amplification of contaminated genomic DNA.
PCR primers were designed using previously described human OCT-4 and β-2 microglobulin (hβ2m) sequences (GenBank accession numbers: NM_002701 and NM_004048, respectively). The appropriate PCR primers were designed using Genrunner software (version 3.02; Hastings Software) and were as follows:
PCR was performed using 2 μl of cDNA or No-RT sample with 1 U of Taq polymerase (Cinnagen), 1.5 mM MgCl2, 200 μM dNTPs and 0.4 μM of each primer in a 25 μl PCR reaction.
The PCR amplification was performed for either 35 (hOCT-4) or 26 (hβ2m) cycles, and the cycling conditions were as follows: 94°C for 30 sec, 57.5°C (hβ2m) or 62°C (hOCT-4) for 40 sec, 72°C for 45 sec, with a final extension at 72°C for 10 min. PCR primers amplified 470 and 191 bp fragments from OCT-4 and β2m cDNA, respectively. PCR products were separated on a 1% agarose gel, stained with ethidium bromide and visualized under the UV light. The intensity of bands was determined using Uvitec software (Uvitec, UK). The identity of PCR products confirmed by direct DNA sequencing (MWG, Germany).
All experiments were replicated 2 or 3 times, and the results were analyzed by performing analysis of variance test to determine the difference of OCT-4 expression among different biopsy groups. Also, Pearson's correlation coefficient was used to study the correlation of OCT-4 expression and tumor/nontumor state of the samples (SPSS software for windows, version 11, Chicago).
Frozen tissue samples and harvested cultured cells were homogenized and lysed in modified RIPA buffer (50 mM Tris-HCl, pH 7.4; 150 mM NaCl; 1 mM phenylmethylsulfonyl fluoride; 1 mM EDTA; 1% Triton X-100; 1% sodium deoxycholate; 0.1% SDS). The concentration of proteins in cell lysates was quantified by means of Bradford assay, and 25 μg of total protein was loaded in each lane. Samples were electrophoresed using SDS-PAGE (12.5%) and blotted for 2 hr onto Hyband-P PVDF membrane (Amersham Biosciences Europe GmbH, Germany). Membranes were stained with Ponceau-S to check loading and transfer, and then blocked for 2 hr in ECL advance blocking solution, according to the manufacturer's instructions (Amersham Biosciences Europe GmbH). Blots were incubated with the anti-OCT-4 antibodies for 3 hr and anti-β actin antibody for 1 hr, and then with the secondary antibodies for 1 hr in room temperature, before being visualized by ECL Advance Western Blotting detection kit (Amersham Biosciences Europe GmbH). Anti-OCT-4 antibodies (SC-5279 and SC-8629; Santa Cruz Biotechnology, CA) and anti-β actin antibody (Prosci, CA) were used at dilutions 1:2,000, and HRP-conjugated anti-goat IgG (Abcam, UK), HRP-conjugated anti-mouse IgG (Sigma, Germany) and HRP-conjugated anti-rabbit IgG (DakoCytomation, Denmark) were used at 1:50,000 dilution. All antibodies were diluted in ECL Advance Blocking solution, according to the manufacturer's instructions.
Formalin-fixed paraffin-embedded (FFPE) tissue sections (5 μm) were deparaffinized with xylene, rehydrated in descending concentrations of ethanol and boiled for 15 min in citrate buffer (10 mM, pH 6.0). Endogenous peroxidase activity was suppressed with 3% H2O2 for 20 min. Slides were serum-blocked (with normal goat serum) and incubated with Santa Cruz anti-OCT-4 polyclonal antibody (SC-8629; 1:50 dilution; Santa Cruz Biotechnology) for 2 hr at room temperature, and then stained with goat ABC staining system kit (Santa Cruz Biotechnology) according to the manufacturer's instructions. In negative control, all the conditions were kept the same, except that the first antibody was eliminated.
Expression of OCT-4 in malignant and nonmalignant human bladder tissues
We designed specific primers to amplify a segment of OCT-4, which is common in both spliced variants of the gene (GenBank accession numbers: NM_002701 and NM_203289). As expected, a 470 bp DNA fragment was amplified in the PCR reaction of OCT-4, and DNA sequencing further confirmed the accuracy of the amplified products. We detected the expression of OCT-4 in almost all (96%) of the examined tumor samples of bladder (31/32; Fig. 1a). The expression was also detected in 23% of Group B (3/13; Fig. 1b) as well as in 33% of Group C samples (3/9; Fig. 1c).
Because we have used a semiquantitative RT-PCR approach, a densitometric evaluation and comparison of relative expression of OCT-4/β2m between different tissue samples was feasible. The intensity of OCT-4 expression was significantly higher in neoplastic tissues (Group A) compared to the nonneoplastic (Group B and C) samples (p < 0.001; Fig. 2). There was a strong correlation of 0.6 between the expression of OCT-4 and the tumor/nontumor state of the samples (p < 0.001). However, none of the investigated clinicopathological variables (tumor grade, stage and size) showed a statistically significant correlation with the expression levels of OCT-4.
The pattern of OCT-4 expression was also examined at the protein level in the bladder samples. We also used the embryonic carcinoma cell line, NTERA2 (NT2) as a positive control to optimize the experiment. Using a polyclonal anti-OCT-4 antibody, we detected an expected 45 kDa band in NT2 cell line (Fig. 3a). We also detected a single strong band in tumor and a single weaker band in nontumor bladder tissues; however, the size of the latter bands was noticeably higher (∼52 kDa in size) than the expected size (Fig. 3a).
To ensure that the observed ∼52 kDa band is not a nonspecific protein detected by the antibody, we repeated the same experiment with a different monoclonal anti-OCT-4 antibody. The experiment produced a similar 45 kDa band in NT2 cell line and a single ∼52 kDa band in bladder tissues (Fig. 3b). In addition to this, further analysis revealed a weaker band in NT2 cells, which was similar in size to the band detected in bladder tissues (data not shown).
Tissue distribution and intracellular localization of OCT-4 protein in bladder tumors
Using the polyclonal anti-OCT-4 antibody, we examined the tissue distribution and subcellular localization of OCT-4 protein in bladder tissues by immunohistochemistry (IHC). As a positive control, we used FFPE sections of seminoma and embryonic carcinoma of testis, which are known to have a nuclear localization for OCT-4. Our results confirmed the same subcellular localization for OCT-4 in these cells (Fig. 4c).
Further, we examined the expression and subcellular localization of OCT-4 protein in FFPE sections of bladder carcinoma. OCT-4 was primarily localized in the nuclei of tumor cells, with no immunoreactivity in normal cells adjacent to the tumors (Fig. 4a). Interestingly, the intensity of immunoreactivity was variable among positive cells, suggesting that the cells within the tumors are heterogenous in term of OCT-4 expression. We also detected a cytoplasmic distribution of OCT-4 in some samples. (Fig. 4b). No immunoreactivity was observed in negative controls, which were incubated in the absence of primary antibody (Fig. 4d).
Tumor recurrence and multifocality are 2 common features of bladder tumors. Recently, several lines of evidence support a clonal nature for multifocal and recurrent urothelial carcinomas, suggesting that these tumors are derived from a primary transformed progenitor cell.17, 18, 19 Based on the new CSC hypothesis, this primary mutant cell is possibly a transformed stem/early progenitor cell (i.e. CSC), with a dysregulated self-renewal capacity generating more CSCs as well as phenotypically diverse cancer cells, with less tumorigenesis potential.1, 2 The new concept could change our understanding of tumor development and progression. It may also change our diagnostic and therapeutic approaches by allowing us to better identify and target CSCs.
Having the recent proposed role for CSCs in tumorigenesis and the fact that these cells are generated through uncontrolled self-renewal of normal stem or progenitor cells, it is very important to examine the expression and involvement of stem cell's self-renewal regulator genes in carcinogenesis. Here, we examined the expression of a well-known self-renewal regulatory factor, OCT-4, in human bladder cancer, and further determined the correlation between the expression of this gene with the tumor state of the samples.
The expression of OCT-4 has already been reported in germ cell tumors,15 a small number human kidney and lung cancer samples,15 human breast cancer and osteosarcoma biopsies,20, 21, 22 canine neoplasms,23 and also in some human cancer cell lines.12, 24 To our knowledge, the data presented here is a unique and comprehensive study of OCT-4 expression in tumor vs. nontumor tissues of a somatic cancer. According to our data, OCT-4 is highly expressed in bladder tumors. This might be interpreted as the expansion of bladder cells that intrinsically express OCT-4 or through acquisition of self-renewal capacity by other cancer cells, which led to a gain of OCT-4 expression. The sensitivity and specificity of OCT-4 expression as a molecular marker in detection of bladder tumors were determined as 96 and 66%, respectively. We found no significant correlation between the expression level of OCT-4 and tumor grade or stage. Further quantitative approaches are required in this regard to elucidate the significance of OCT-4 expression in the cancer. Work on archival collection of FFPE samples of bladder tumors is in progress in our lab to asses the correlation between OCT-4 expression and prognostic parameters, like tumor progression, tumor recurrence and cancer survival rate.
In addition to the tumor samples, we also observed a low expression of OCT-4 in some nontumor bladder tissues obtained from individuals with no symptoms and signs of bladder cancer (Group C). Also we have observed no significant difference in OCT-4 expression between Group B and Group C samples. Recently, Tai et al. have shown that OCT-4 is expressed in several human adult stem cells12 (e.g. breast, pancreas and liver stem cells) and Matthai et al. reported OCT-4 expression in normal human endometrium.25 The observed expression of OCT-4 in normal bladder tissues might reflect the presence of rare normal bladder stem cells in these samples. For that reason, we expect that evaluating the expression of OCT-4 in exfoliated cells of urine samples will potentially decrease the presence of normal bladder stem cells in the sample and increase the specificity of OCT-4 as a molecular marker for bladder cancer.
Both antibodies, used in our experiments, detected a slightly higher molecular weight of OCT-4 in bladder tumors compared to the one in NT2 cells. We speculate that a differential posttranslational modification of OCT-4 in 2 systems might be responsible for the observation. However, this possibility and its possible correlation with bladder carcinogenesis need to be further investigated.
According to our IHC results, OCT-4 positive cells were not distributed equally in different tumors, ranging from several scattered cells to aggregated clusters. Similar observation has been reported previously by Gibbs et al., who observed a variable OCT-4 expression in different bone tumors ranging from 1 to 25% of the cells. The mentioned variation was even higher for Nanog (another key regulator of stem cell self-renewal) positive cells, which comprised 1–50% of the cells in different samples.21 In contrast to nuclear staining, the cytoplasmic distribution of OCT-4 usually was restricted to the cells located adjacent to the basal lamina.
Using IHC analysis on tissue microarrays, Looijenga et al.15 have previously reported no expression of OCT-4 in a panel of somatic tissues such as bladder tumors. This finding is in contrast to our finding, and the reason for this inconsistency might be due to the heterogenous nature of tumors and the fact that tissue sampling for microarrays might not be a good representative of the whole tumor. Accordingly, using IHC analysis of the whole tissue sections, recently, several groups have reported the expression of OCT-4 protein in several somatic tissues.21, 25
In conclusion, our data is the first report on the expression of the ES cell marker, OCT-4, in bladder cancer, and would add more weight to the findings that candidate OCT-4 as a multifunctional factor involving in major biological processes such as embryonic development, control of differentiation and stem cell-based carcinogenesis. More specifically, OCT-4 can potentially be regarded as a new molecular marker of bladder tumors, in which its expression might indicate the existence of stem-like cancer cells in these tumors. This is also a further evidence to support the concept of stem cell origin of cancer. Accordingly, the data might provide valuable information on the nature and behavior of bladder tumors, leading to a new strategy for targeting the CSCs and perhaps one step closer to cure cancer recurrence and metastasis. However, further studies are required to isolate and characterize the putative CSCs from bladder tumors and to elucidate the role of OCT-4 in carcinogenesis of the tumor.
We are grateful to Dr. Hassan Inanloo for his valuable help in collecting clinical samples and Dr. Paul Gokhale for his critically reviewing and editing of the manuscript.