- Email: [email protected]
Cancer stem-like cells contribute to cisplatin resistance and progression in bladder cancer
Cancer Letters 322 (2012) 70–77
Contents lists available at SciVerse ScienceDirect
Cancer Letters journal homepage: www.elsevier.com/locate/canlet
Cancer stem-like cells contribute to cisplatin resistance and progression in bladder cancer Yi Zhang a, Zhi Wang a, Jin Yu a, Jia zhong Shi a, Chun Wang a, Wei hua Fu b, Zhi wen Chen b,c,⇑, Jin Yang a,⇑ a
Department of Cell Biology, The Third Military Medical University, Chongqing 400038, China Urology Institute of PLA, Southwest Hospital, The Third Military Medical University, Chongqing 400038, China c Southwest Cancer Center, Southwest Hospital, The Third Military Medical University, Chongqing 400038, China b
a r t i c l e
i n f o
Article history: Received 21 November 2011 Received in revised form 8 February 2012 Accepted 8 February 2012
Keywords: Cancer stem-like cell Bladder cancer Cisplatin resistance Epithelial–mesenchymal transition
a b s t r a c t A variety of cancer stem-like cells (CSCs) have been shown to be responsible for cancer tumorigenicity, relapse and metastasis. Despite several reports demonstrating the presence of CSCs in human bladder cancer, their identities are still under debate, and few studies have examined their roles in cisplatin resistance and related tumor progression. In this study, a subpopulation of CSCs was enriched following cisplatin selection from the bladder cell line T24. The cisplatin-resistant T24 cells displayed a greater self-renewal capacity as demonstrated by higher levels of sphere formation and stem cell marker expression, contained a larger proportion of side population cells and exhibited higher tumorigenicity. They also possessed epithelial–mesenchymal transition characteristics. Furthermore, a strong correlation between the levels of Bmi1 and Nanog expression and the degree of malignancy of urothelial cell carcinomas tissues was observed. We provide the first direct evidence that CSC-like cells exist in the population of cisplatin-resistant bladder cancer cells and may play a role in the progression and drug resistance of bladder cancer. Ó 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Bladder cancer is the fifth most common cancer in Western society and has a global incidence of more than 356,000 new cases a year and an estimated prevalence of 2.7 million cases [1]. After bladder cancer progresses to metastatic disease, the only treatment option is cisplatin-based combination chemotherapy [2]. Cisplatin is a platinum-based compound that forms both intra- and inter-strand DNA adducts. Cisplatin has a broad spectrum of antitumor activity and is widely used in the treatment of various malignant tumors [3]. However, intrinsic or acquired resistance to cisplatin severely limits the therapeutic potential of this drug. The molecular mechanisms that underlie this chemoresistance are still largely unknown. A more detailed understanding of the molecular mechanisms by which tumor cells survive chemotherapy is likely to lead to novel therapeutic targets with more successful outcomes. Emerging lines of evidence have associated chemoresistance with a pool of self-renewing malignant progenitors known as can⇑ Corresponding authors. Address: Department of Cell Biology, The Third Military Medical University, Chongqing 400038, China. Tel.: +86 23 68752260; fax: +86 23 65460268 (J. Yang), Urology Institute of PLA, Southwest Cancer Center, Southwest Hospital, The Third Military Medical University, Chongqing 400038, China. Tel.: +86 23 68765817; fax: +86 23 65460268 (Z. Chen). E-mail addresses: [email protected] (Z. Chen), [email protected] (J. Yang). 0304-3835/$ - see front matter Ó 2012 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2012.02.010
cer stem-like cells (CSCs) or cancer-initiating cells (CICs). Such cells share many properties with somatic stem cells, including self-renewal and multi-potent differentiation potential. Cancer stem cells have been proposed to be responsible for cancer tumorigenicity, relapse and metastasis [4,5]. CSCs appear to be highly resistant to various chemotherapies [6–8]. CSCs have also been found to exhibit a number of genetic and cellular adaptations that confer drug resistance. These adaptations include relative dormancy/slow cell cycle kinetics, efficient DNA repair, high expression of multidrugresistance-type membrane transporters and resistance to apoptosis [9,10]. The existence of CSCs has also recently been reported in human bladder cancer, and multiple potential markers have been identified. Oates et al. and Ning et al. identified SP (side-population) cells from primary bladder cancer and the T24 cell line [11,12]. In 2008, Yang and Chang showed that both the EMA (Epithelial Membrane Antigen) subset of cells and the CD44v6+ subset of cells are able to form colonies and could be bladder cancer-initiating cells [13]. Bentivegna et al. isolated CSCs from primary transitional cell carcinomas that had the ability to form spherical colonies [14]. Chan et al. and He et al. also described the isolation and characterization of a tumor-initiating cell subpopulation from primary human bladder cancer cells based on the CD44+CK5+CK20 or 67-kDaR+CK17+CEACAM6(CD66c) phenotypes [15,16]. To date, there have been the fewer than 10 reports about bladder cancer stem-like cells, and the localization and characteristics
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77
of human bladder stem cells are still under debate. Moreover, there is no direct evidence that bladder CSCs contribute to cisplatin resistance. It is becoming increasingly important to understand the molecular mechanisms of the self-renewal and differentiation of CSCs, which are likely related to tumor growth. The objectives of our current study are to molecularly characterize bladder CSCs, to investigate the role of CSCs in cisplatin resistance in bladder cancer, and to examine the clinical relevance of bladder CSCs. 2. Materials and methods
71
fibronectin (1:200; Santa Cruz Biotechnology). After the FITC-conjugated secondary antibodies had been applied, the nuclei were counterstained with DAPI. The staining was examined under a laser scanning confocal microscope (Leica, Germany).
2.7. Immunohistochemistry The paraffin sections were dewaxed with xylene and rehydrated in descending concentrations of ethanol. The endogenous peroxidase was inhibited, and the slides were incubated with antibodies against Nanog (1:200; Abcam, Inc., Cambridge, UK), Bmi1 (1:200; Abcam, Inc., Cambridge, UK), E-cadherin (1:100; Santa Cruz Biotechnology) or fibronectin (1:200; Santa Cruz Biotechnology). The expression levels of Nanog and Bmi1 were graded as described in Ref. [18].
2.1. Patients and tissue specimens Bladder urothelial carcinoma tissue samples were obtained from 51 patients (33 males and 18 females) by transurethral bladder tumor resection and radical cystectomy at Southwest Hospital, Third Military Medical University. The patients’ mean age was 65 years (ranging, 35–79). All samples were collected between 2006 and 2010. The tumor tissues were examined by one pathologist, and the grade was determined according to WHO criteria (2004). The research was approved by the ethics board of Southwest Hospital, Third Military Medical University.
2.8. Western blot analysis Equal amounts of protein lysates were subjected to SDS-PAGE and transferred to a PVDF membrane by electroblotting. The antibodies against Nanog (1:2000 dilution), Bmi1 (1:2000 dilution), Oct4 (1:2000 dilution), E-cadherin (1:1000 dilution) and fibronectin (1:5000 dilution) were used. A b-actin antibody at a 1:10000 dilution was used as a loading control. The protein bands were visualized using an imaging system (Bio-Rad, Hercules, CA).
2.2. Cell culture and drug selection The human bladder cancer cell line T24 was obtained from the American Type Culture Collection (ATCC, Rockville, MD) and maintained in RPMI-1640 with 10% fetal bovine serum (FBS) (Invitrogen, Carlsbad, CA) at 37 °C in a 5% CO2 atmosphere. T24 cells were treated with 30 lM cisplatin (Sigma-Aldrich Corp., St. Louis, MO) for 4 h and maintained in serum-free RPMI-1640 medium (SFM) supplemented with 5 lg/ml insulin (Sigma), 20 ng/ml EGF (Peprotech Inc., Rocky Hill, NJ) and 20 ng/ ml bFGF (Peprotech). Cisplatin (5 lM) was routinely added to the culture medium every other day. The cells were treated with the drug at least two times as described above. The surviving cells were considered drug-resistant cells (DRCs). The cells were dissociated and plated in triplicate at a density of 200 cells per well in 6-well plates. The total number of colonies (P20 cells) and the number of large colonies (P50 cells) were determined under a microscope. 2.3. Sphere formation and self-renewal assay The cells were inoculated at a density of 2 104 cells/well in ultra-low-attachment 6-well plates and grown in a serum-free medium supplemented with insulin, EGF and bFGF. After 5–10 days in culture, colonies that contained >20 cells were counted. In addition, to evaluate the self-renewal ability, cell spheres were enzymatically dissociated with 0.05% trypsin for 10 min at 37 °C and single-cell suspension was seeded. The secondary sphere formation was measured as described above. 2.4. Side population analysis The side population (SP) analysis was performed as described previously [17]. Briefly, the single-cell suspensions were stained with Hoechst 33342 (Sigma-Aldrich) at a final concentration of 5 lg/ml and incubated in the dark for 90 min at 37 °C with mixing at intervals. A negative control was incubated with 50 lM verapamil (Sigma-Aldrich) for 30 min at 37 °C before Hoechst 33342 was added. The FACS analysis was performed using 450 nm (Hoechst blue) and a 675 nm (Hoechst red) filters after excitation with a 350 nm UV light.
2.9. Cell migration The cells were cultured in a confluent monolayer and then scratched with a 1 mL pipette tip to create a cell-free area in 6-well culture plates. Mitomycin C (5 lg/mL) was added to inhibit cell proliferation, and the wound was photographed at 0 h, 12 h and 24 h.
2.10. In vivo xenograft experiments 1 105 Cells were resuspended in 200 ll PBS and injected subcutaneously into the dorsa of 4-week-old nude mice. The engrafted mice were observed every 2 days until tumors formed. The mice were sacrificed by cervical dislocation, and the xenograft tumors were resected, fixed in 10% phosphate-buffered formalin, embedded in paraffin for sectioning and processed for H&E staining and immunohistochemistry. The mice were maintained according to the guidelines for animal experiments of the Third Military Medical University.
2.11. Statistical methods The correlation between the expression level of Nanog and Bmi1 and clinical grade was analyzed by either the chi-squared test or Fisher’s exact test. For the quantitative data, the difference between groups was assessed with Student’s t-test. The Kaplan-Meier estimate was used for the survival analysis. A difference was considered statistically significant at P < 0.05 or P < 0.01. All analyzes were performed with SPSS (Version 13.0).
3. Results 3.1. Stem-related genes are overexpressed in bladder cancer
2.5. Semi-quantitative reverse transcriptase-polymerase chain reaction (RT-PCR) The cells were harvested, and the total RNA was extracted using the RNeasy Total RNA system (Qiagen GmbH, Hilden, Germany) following the manufacturer’s protocol. The first-strand cDNA was synthesized using the Reverse Transcriptase System (Invitrogen Carlsbad, CA), and the target cDNA was amplified using primers for the Oct4, Nanog and Bmi1 genes. The sequences of the primers were as follows: Nanog, forward: 50 -CCGAAGAATAGCAATGGTG-30 and reverse: 50 GGAGAATTTGGCTGGAACT-30 ; Oct4, forward: 50 -TGGAGAAGGAGAAGCTGGAGCAAAA-30 and reverse: 50 -GGCAGATGGTCGTTTGGCTGAATA-30 ; Bmi1, forward: 50 ACTATCGTCCAATTTGCT-30 and reverse: 50 -AATGCCATCTGATTCTTAC-30 ; and actin, forward: 50 -CATCTCTTGCTCGAAGTCCA-30 and reverse: 50 ATCATGTTTGAGACCTTCAACA-30 . Aliquots (8 ll) of the amplification products were loaded onto 1.5% agarose gels, separated by electrophoresis and visualized by ethidium bromide staining. 2.6. Immunofluorescent staining The cells were fixed with 4% paraformaldehyde, blocked with 10% normal goat serum, and incubated with a primary antibody against CD44 (1:100; Santa Cruz Biotechnology, Santa Cruz, CA, USA), E-cadherin (1:100; Santa Cruz Biotechnology) or
CSCs are considered to share similar features with normal stem cells and are proposed to positively correlate with the malignant phenotype. To clarify the clinical role of bladder CSCs, we chose to analyze the expression of ‘‘stemness genes’’ such as nanog and bmi1 in different clinical grades of human bladder cancer tissues. Nanog and Bmi1 overexpression has been shown in some kinds of CSCs, suggesting that the function and regulation of these genes are essential for CSC self-renewal and tumorigenesis [19,20]. All parameters of immunohistochemical detection are summarized in Table 1 and Fig. 1. The statistical analysis showed that increased expression of Nanog and Bmi1 in bladder cancer was significantly associated with a high pathological grade (P = 0.001 and 0.029, respectively). Based on that Nanog and Bmi1 strongly expressed cells may represent a more stem-like population, our result suggested that CSCs may contribute to the tumorigenesis of human bladder cancer and Nanog and Bmi1 may be potential biological and prognostic markers in tumors.
72
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77
Table 1 The expression of Nanog and Bmi1 in bladder cancer tissues.
PUNLMP Low-grade High-grade
Low (%) Nanog/Bmi1
Moderate (%) Nanog/Bmi1
Intense (%) Nanog/Bmi1
N
P value Nanog/Bmi1
14(70.0)/12(60.0) 3(21.4)/4(28.6) 2(11.8)/3(17.7)
4(20.0)/5(25.0) 6(42.9)/6(42.8) 5(29.4)/4(23.5)
2(10.0)/3(15.0) 5(35.7)/4(28.6) 10(58.8)/10 (58.8)
20 14 17
0.001/0.029
Five random views were chosen and 100 cells per view were counted for analysis. The percentage <10 was defined as negative. It was reported as low (10–29%), moderate (30–60%) and intense (>60%) expression. PUNLMP: Papillary Urothelial Neoplasm of Low Malignant Potential; low-grade: low-grade papillary carcinoma; high-grade: highgrade papillary carcinoma.
Fig. 1. The immunochemical expression of Bmi1 and Nanog in human bladder cancer. (A–C) The expression of Nanog. (D–F) The expression of Bmi1. (A, D) PUNLMP; (B, E) Low-grade; (C, F) High-grade. Scale bar: 50 lm.
3.2. DR-T24 cells display self-renewal capacity A variety of CSCs has a drug-resistant phenotype in vitro and is proposed to play an important role in the drug resistance of tumors. To provide direct evidence that CSCs exist in bladder cancer, we first attempted to enrich CSCs from cisplatin-resistant bladder cancer cells. The cisplatin-resistant T24 cells (DR-T24) were selected as described in the Materials and Methods. The total and large colonies were counted to determine the clonogenic capacity of these cells. The cells gave rise to clones at an average frequency of 23 ± 2.9 colonies/well for parental T24 cells and 39 ± 5.8 colonies/well for DR-T24 cells at 7 days (P < 0.01), and the large clone-formation efficiencies of parental and cisplatin-resistant T24 cells were 6.4 ± 1.1 colonies/well and 10.8 ± 2.4 colonies/well (P < 0.01), respectively (Fig. 2A and B). Sphere formation has been well described as a typical characteristic of CSCs that reflects the potential for self-renewal [4]. We evaluated the ability of T24 and DR-T24 cells to generate spherical colonies in an ultra-low attachment and serum-starved culture system. After 7 days of culture, spherical colonies were counted. The DR-T24 cells exhibited a 3-fold higher frequency of spherical colonies than the parental T24 cells (12.64 ± 1.58% for the DRT24 cells versus 4.28 ± 0.98% for the parental T24 cells) (Fig. 2C and D). Furthermore, the self-renewing capacity of DR-T24 cells were assessed by dissociating primary spheres into single cells, which were incubated in anchorage-independent, stem cell culture conditions. After 5 days, these cells gave rise to secondary spheres (15.08 ± 2.44%) (Fig. 2E). Statistical analysis of the percentage of sphere formation is shown in Fig. 2F. Collectively, these data indi-
cate that the stem-like feature of DR-T24 cells can be maintained in vitro. 3.3. DR-T24 cells contain a larger side population The existence of side populations within different cancer cell lines bearing some stem cell-like features is well documented [21]. It has also been published that SP cells identified from primary bladder cancer and the T24 cell line harbor tumor-initiating cell characteristics [11,12]. We evaluated the SP fraction in both the T24 parental cells and the DR-T24 cells. A higher percentage of SP cells was observed in the DR-T24 cells compared with the T24 parental cells. The flow cytometric analysis showed that the percentage of SP cells increased from 2.5% in the parental T24 cells to 7.7% in the DR-T24 cells (P < 0.05) (Fig. 3). These results suggested that under drug selection, cancer stem-like cells might be enriched. 3.4. DR-T24 cells express stem cell markers To confirm the stem cell phenotype of the DR-T24 cells, the expression of ES cell marker genes was evaluated. The RT-PCR analysis showed that the DR-T24 cells expressed higher mRNA levels of Nanog, Bmi1 and Oct4 (Fig. 4A). Consistent with the RT-PCR data, the expression of these proteins was also stronger in DR-T24 cells, as demonstrated by immunoblotting (Fig. 4B). Because it has also been reported that CD44 may be a potential bladder CSC marker, immunofluorescent staining was performed to determine the level of protein expression of CD44. The DR-T24 cells also ex-
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77
73
Fig. 2. The higher proliferative potential and sphere formation of the DR-T24 population treated with cisplatin. (A) The total clone-forming capacity (>20 cells/clone). (B) The large clone-forming capacity (>50 cells/clone). (C) Primary tumor spheres originated from parental T24 cells. (D) Primary tumor spheres originated from DR-T24 cells. (E) Secondary spheres derived from single cells of primary tumor sphere. (F) The statistical analysis of the sphere formation efficiency. P < 0.01. Scale bars: (C–E) 100 lm.
Fig. 3. Flow cytometric analysis of side population in DR-T24 cells and parental T24 cells. As control, the SP fraction was less to 0.1% under the verapamil pre-incubated.
74
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77
lower than that of the parental T24-implanted mice (P < 0.01) (Fig. 5B). The H&E staining showed that the xenograft tumor cells originating from the DR-T24 cells contained numerous large and heterogeneous nuclei surrounded by a high density of microvessels, indicative of a more malignant phenotype (Fig. 5C). The isolated xenograft tumor cells were also able to form secondary spheres in stem cell medium, indicating their self-renewing capability (Fig. 5D). Immunofluorescent analysis of these xenografts revealed that CD44 was expressed at a significantly higher level in the tumors from the DR-T24 cells than those of the parental cells (Fig. 5E). These results indicated that the DR-T24 cells contained a larger population of bladder CSCs that promote tumorigenesis.
3.6. DR-T24 cells display features of EMT
Fig. 4. Expression of stem cell markers in T24 cells and DR-T24 cells. (A) The expression of Oct4, Nanog and Bmi1 mRNA detected by RT-PCR. (B) Western blot analysis of Oct4, Nanog, Bmi1 and CD44 protein expression. (C) Immunofluorescent staining of CD44 in cultured both T24 cells and DR-T24 cells. Scale bar: 25 lm.
pressed higher levels of CD44 than the parental T24 cells (Fig. 4B and C). Together, these results showed that potential cancer stem-like cells expressing ‘‘stemness’’ genes can be enriched by drug selection. 3.5. DR-T24 cells display highly tumorigenic behavior in vivo High tumorigenic potential is considered a hallmark of CSCs. To determine whether the DR-T24 cells exhibited higher tumorigenicity, the parental T24 cells and the DR-T24 cells were subcutaneously injected into nude mice for transplanted tumorigenicity analysis (n = 10, each). In the mice implanted with the parental T24 cells, tumors were visible after 32–52 days (average = 40.6 days). In contrast, the DR-T24 cells produced tumors within 18–30 days (average = 24.4 days) (Fig. 5A). The mean survival rate of the DR-T24-implanted mice was also significantly
Current data demonstrate that CSCs have a mesenchymal cell phenotype and are associated with tumor initiation, growth, metastasis, and therapeutic resistance [12]. To elucidate the role of epithelial–mesenchymal transition (EMT) in cancer stem cell generation, we compared the cell migration abilities of the DRT24 and control cells. In an in vitro wound-healing assay, the DRT24 cells migrated into the wound area more rapidly than control cells did (Fig. 6A). These data suggest that the DR-T24 cells have a stronger cell migration and invasion abilities and thus have a greater malignant potential. We next examined the expression levels of EMT markers in both the parental and DR-T24 cells. Mesenchymal and epithelial specific markers (fibronectin and E-cadherin, respectively) were chosen. Ecadherin expression is decreased both in the DR-T24 cells cultured in vitro and in the xenografts originating from cell transplantation, whereas fibronectin is expressed at a higher level in DR-T24 cells (Fig. 6C and D). Our results are consistent with the typical EMT features, including the loss of epithelial markers and the concomitant gain of mesenchymal markers. These observations suggest that the bladder CSCs underwent an EMT process. Upon Immunocytochemical analysis of E-cadherin localization, we were surprised to discover that E-cadherin appeared both in the cytoplasm and the nucleus of the DR-T24 cells, in contrast with
Fig. 5. The subcutaneous tumorigenicity in nude mice. (A) 1 105 cells were injected into nude mice. the left: T24 cells; the right: DR-T24 cells. Statistical graph was shown on the top right. P < 0.01. (B) The survival curves of mice injected with the parental T24 cells (dashed line) and DR-T24 cells (solid line). (C) The morphology of tumor developed from T24 cells (left) and DR-T24 cells (right) (H&E). (D) Secondary sphere formed after xenograft tumor cells re-seeded. (E) Immunofluorescent staining of CD44 in xenograft tumor derived from parental T24 cells (up) and DR-T24 cells (down). Scale bars: (C, E) 50 lm; (D) 100 lm;.
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77
75
Fig. 6. DR-T24 cells display EMT feature. (A) Micrographs of cell wound closure monitored by light microscopy and photographed every 12 h. (B) Immunofluorescent and (C) Western blot analysis of E-cadherin, Fibronectin expression in cultured T24 and DR-T24 cells. (D) Immunohistochemical staining of E-cadherin and Fibronectin in the tumor formed by implanting T24 cells and DR-T24 cells. Scale bars: (A) 50 lm; (B) 25 lm; (D) 50 lm .
the membrane localization in the parental T24 cells (Fig. 6B). This is the first report showing that cytoplasmic and nuclear E-cadherin appears present in both bladder cancers and cisplatin-resistant cells, suggesting a potentially important role for E-cadherin mislocalization in both processes. 4. Discussion To investigate the role of CSCs in cisplatin resistance and determine the molecular characteristics of bladder CSCs and their clinical relevance, a subpopulation of cancer stem-like cells from the bladder cell line T24 was enriched by cisplatin selection and was found to possess the features of both stem cells and cells that have undergone EMT. There was a high incidence of Bmi1 and Nanog overexpression in the bladder urothelial cell carcinoma tissues analyzed, and a strong correlation between the overexpression of Bmi1 and Nanog and the degree of malignancy of bladder tumors was observed. We provide the first direct evidence that CSC-like cells do exist among cisplatin-resistant cells in bladder cancer and may play a role in the progression and drug resistance of bladder cancer. The putative bladder cancer stem cells can be considered to be mediators of resistance to current therapies and represent strong candidate biological targets. To develop a therapy that targets bladder CSCs, the first key step is to identify and isolate CSCs. As we described in the introduction, the results of a few recent reports clearly support that a functionally distinct subpopulation of CSCs can be isolated from bladder cancer using either prospective surface marker-based FACS analysis or SP cell sorting. However, the relationship among the different CSC populations isolated using various markers or methods is unclear [11–16]. It is important that the bladder CSC phenotype be further defined before CSCs can be reliably identified and isolated. Among the methods of identification of CSCs, surface-marker-based FACS analysis and sorting are the most widespread methods. Unfortunately, the more controversial question arising from these recent studies is whether CSCs exhibit a common marker(s). It has been demonstrated that different cancer stem cells harboring different surface markers may exist in a single type of cancer. For example, in addition to CD133+ CSCs in glioma, CD133 tumor cells possess apparent stem cell-like properties but distinct molecular profiles and growth characteristics [22]. Other ways to identify CSCs are based on the phenotypic char-
acteristics, such as side population (SP) analysis. However, this technique can be hampered by UV exposure and the loss of cytoactivity by SP cells [21]. It should be noted that all of the isolation strategies used to date have their shortfalls, and perhaps a combination of different isolation methods will be required to enhance the purity of cancer stem cells. In this study, we found that CSCs could be enriched from T24 cells by cisplatin selection. The DR-T24 cells were observed to possess an increased ability to form tumor spheres and had a high clonogenic efficiency compared with the parental T24 cells (Fig. 2). The DR-T24 cells also express higher levels of ESC markers (e.g., Bmi1, Nanog, and Oct4) (Fig. 4). This result suggests that a selfrenewing population of CSCs is present in T24 cells. Immunohistochemical analysis of the DR-T24 cells also revealed a larger population of CD44+ and SP cells, which is consistent with results found in other reports (Fig. 3). To obtain in vivo evidence and provide evidence of differentiation, we implanted the DR-T24 cells and the parental T24 cells into nude mice and found that the DR-T24 cells have a higher tumorigenic potential. Immunostaining revealed an obvious increase in the level of CD44 in the tumors generated from the DR-T24 cells (Fig. 5E). Hence, our study provides the first direct evidence that CSC cells do exist among cisplatin-resistant bladder cancer cells and may thus play a role in the drug resistance of bladder cancer. This system may be a valuable model for the further study of both CSCs and chemoresistance. The preliminary drug selection of cancer cells followed by surface marker-based cell sorting may decrease the heterogeneity of the CSC population and afford a more efficient method to isolate the bladder cancer CSCs. Although anti-tumor drug resistance and cancer metastasis often occur at the same time clinically, there are no reports providing direct evidence of acquired cisplatin resistance in bladder cancer cells with an obvious invasive and metastasis phenotype. Based on the association of CSCs with cancer relapse and metastasis and our strong evidence that CSCs do exist among cisplatin-resistant bladder cancer cells, we had reason to raise the question of whether the DR-T24 cells are characterized by a more invasive phenotype. Indeed, an in vitro wound healing assay demonstrated that the DR-T24 cells migrate much more rapidly than the control cells, strongly suggesting that cisplatinresistant bladder cancer cells have a more powerful cell migration ability (Fig. 6A).
76
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77
The concept of epithelial–mesenchymal transition (EMT) initially developed in the field of embryology and has recently has been extended to cancer progression. It has been demonstrated that EMT is involved in carcinogenesis, invasion, metastasis, recurrence, and chemoresistance [23]. CSCs can also be induced by EMT [24]. To molecularly characterize the cisplatin-resistant cells, including the bladder CSCs, we further investigated the role of EMT and its involvement in the DR-T24 cellular phenotype. Compared with the level in parental T24 cells, the level of epithelial marker E-cadherin is decreased both in the DR-T24 cells cultured in vitro and in the tumor xenografts, In addition, the mesenchymal marker fibronectin is more highly expressed in DR-T24 cells. Our results are thus consistent with the typical EMT cellular features, including the concomitant loss of epithelial markers and gain of mesenchymal markers [23]. This first report of EMT properties in cisplatin-resistant bladder cancer cells may provide the initial clue to explain the high cell migration ability of the DR-T24 cells shown above. These observations may stimulate future interest in defining the molecular mechanisms underlying this epithelial-to-mesenchymal ‘‘switch’’ in the generation of the bladder CSCs and the acquisition of cisplatin resistance. To our surprise, in contrast to the membrane localization in the parental T24 cells, E-cadherin appeared in both the cytoplasm and the nucleus of the DR-T24 cells (Fig. 6). These findings are contrary to the traditional theory that E-cadherin mediates intercellular adhesion. Recent studies have noted nuclear E-cadherin expression in a variety of tumors [25]. However, the biological or pathological roles of cytoplasmic and nuclear E-cadherin are still unclear. Our discovery of the cytoplasmic and nuclear localization of E-cadherin in cisplatin-resistant bladder cancer cells implies that altered Ecadherin localization plays an important role in drug resistance and bladder CSC origination. Despite the identification of CSCs in many diseases and the potential of CSCs to explain the failure of many anticancer therapies to extend overall survival, few data actually exist to support that CSCs are clinically relevant. Based on the discovery that CSCs may contribute to cancer tumorigenicity, relapse and metastasis, it has been proposed that the expression level of stemness genes or key regulatory factors related to stem cell self-renewal may be associated with tumor progression. Among the stemness genes, the transcription factors Sox2, Oct3/4 and Nanog form a core regulatory network that coordinately determines the self-renewal and differentiation of embryonic stem cells [26–28]. Bmi1 has also emerged as a key regulator of self-renewal in several cell types, including hematopoietic, neural and mammary stem cells [29– 31]. Some studies have suggested that these factors may play a role in human malignancy. Nanog is overexpressed in many tumors, including carcinomas of the breast, lung, pancreas and ovary [20,32–34]. Ample evidence also exists linking Bmi1 expression levels to the self-renewal and maintenance of both normal and cancer stem cells [35,36]. Chang et al. and Atlasi et al. recently also discovered that Oct3/ 4 is highly expressed in human bladder cancer and is associated with disease progression, increased metastasis, and shorter cancer-related survival [37,38]. We therefore hypothesized that other embryonic stemness factors or key regulatory factors related to stem cell self-renewal may also contribute to the progression and drug resistance of bladder cancer. In our study, Nanog and Bmi1 were found for the first time to be highly expressed in bladder cancer tissues, and the expression of these genes was found to be positively correlated with clinical stage and grade (Fig. 1, Table 1). Together, these findings suggest not only that the progression and cisplatin resistance of bladder cancer may be related to the existence of cancer stem cells but that Nanog and Bmi1 may also be potential biological and prognostic markers in tumors.
Acknowledgments Supported by Grant (No. 30973385) from National Nature Science Foundation of China (NSFC) and Grant (No. 2010CB529402) from the National Basic Research Program of China (‘‘973’’ Program). References [1] B.W. van Rhijn, M. Burger, Y. Lotan, E. Solsona, C.G. Stief, R.J. Sylvester, J.A. Witjes, A.R. Zlotta, Recurrence and progression of disease in non-muscleinvasive bladder cancer: from epidemiology to treatment strategy, Eur. Urol. 56 (2009) 430–442. [2] J. Bellmunt, S. Albiol, C. Suarez, J. Albanell, Optimizing therapeutic strategies in advanced bladder cancer: update on chemotherapy and the role of targeted agents, Crit. Rev. Oncol. Hematol. 69 (2009) 211–222. [3] D.B. Zamble, S.J. Lippard, Cisplatin and DNA repair in cancer chemotherapy, Trends Biochem. Sci. 20 (1995) 435–439. [4] P. Dalerba, R.W. Cho, M.F. Clarke, Cancer stem cells: models and concepts, Annu. Rev. Med. 58 (2007) 267–284. [5] M.F. Clarke, M. Fuller, Stem cells and cancer: two faces of eve, Cell 124 (2006) 1111–1115. [6] C.S. O’Brien, G. Farnie, S.J. Howell, R.B. Clarke, Are stem-like cells responsible for resistance to therapy in breast cancer, Breast Dis. 29 (2008) 83–89. [7] G. Bertolini, L. Roz, P. Perego, M. Tortoreto, E. Fontanella, L. Gatti, G. Pratesi, A. Fabbri, F. Andriani, S. Tinelli, E. Roz, R. Caserini, S. Lo Vullo, T. Camerini, L. Mariani, D. Delia, E. Calabro, U. Pastorino, G. Sozzi, Highly tumorigenic lung cancer CD133+ cells display stem-like features and are spared by cisplatin treatment, Proc. Natl. Acad. Sci. USA 106 (2009) 16281–16286. [8] H. Fujii, K. Honoki, T. Tsujiuchi, A. Kido, K. Yoshitani, Y. Takakura, Sphereforming stem-like cell populations with drug resistance in human sarcoma cell lines, Int. J. Oncol. 34 (2009) 1381–1386. [9] M. Dean, Cancer stem cells: implications for cancer causation and therapy resistance, Discov. Med. 5 (2005) 278–282. [10] R. Morrison, S.M. Schleicher, Y. Sun, K.J. Niermann, S. Kim, D.E. Spratt, C.H. Chung, B. Lu, Targeting the mechanisms of resistance to chemotherapy and radiotherapy with the cancer stem cell hypothesis, J. Oncol. 2011 (2011) 941876. [11] Z.F. Ning, Y.J. Huang, T.X. Lin, Y.X. Zhou, C. Jiang, K.W. Xu, H. Huang, X.B. Yin, J. Huang, Subpopulations of stem-like cells in side population cells from the human bladder transitional cell cancer cell line T24, J. Int. Med. Res. 37 (2009) 621–630. [12] J.E. Oates, B.R. Grey, S.K. Addla, J.D. Samuel, C.A. Hart, V.A. Ramani, M.D. Brown, N.W. Clarke, Hoechst 33342 side population identification is a conserved and unified mechanism in urological cancers, Stem Cells Dev. 18 (2009) 1515– 1522. [13] Y.M. Yang, J.W. Chang, Bladder cancer initiating cells (BCICs) are among EMACD44v6+ subset: novel methods for isolating undetermined cancer stem (initiating) cells, Cancer Invest. 26 (2008) 725–733. [14] A. Bentivegna, D. Conconi, E. Panzeri, E. Sala, G. Bovo, P. Vigano, S. Brunelli, M. Bossi, G. Tredici, G. Strada, L. Dalpra, Biological heterogeneity of putative bladder cancer stem-like cell populations from human bladder transitional cell carcinoma samples, Cancer Sci. 101 (2010) 416–424. [15] K.S. Chan, I. Espinosa, M. Chao, D. Wong, L. Ailles, M. Diehn, H. Gill, J. Presti Jr., H.Y. Chang, M. van de Rijn, L. Shortliffe, I.L. Weissman, Identification, molecular characterization, clinical prognosis, and therapeutic targeting of human bladder tumor-initiating cells, Proc. Natl. Acad. Sci. USA 106 (2009) 14016–14021. [16] X. He, L. Marchionni, D.E. Hansel, W. Yu, A. Sood, J. Yang, G. Parmigiani, W. Matsui, D.M. Berman, Differentiation of a highly tumorigenic basal cell compartment in urothelial carcinoma, Stem Cells 27 (2009) 1487–1495. [17] J. Wang, L.P. Guo, L.Z. Chen, Y.X. Zeng, S.H. Lu, Identification of cancer stem cell-like side population cells in human nasopharyngeal carcinoma cell line, Cancer Res. 67 (2007) 3716–3724. [18] X. Jia, X. Li, Y. Xu, S. Zhang, W. Mou, Y. Liu, D. Lv, C.H. Liu, X. Tan, R. Xiang, N. Li, SOX2 promotes tumorigenesis and increases the anti-apoptotic property of human prostate cancer cell, J. Mol. Cell Biol. 3 (2011) 230–238. [19] L. Jiang, J. Li, L. Song, Bmi-1, stem cells and cancer, Acta. Biochim. Biophys. Sin. (Shanghai) 41 (2009) 527–534. [20] S.H. Chiou, M.L. Wang, Y.T. Chou, C.J. Chen, C.F. Hong, W.J. Hsieh, H.T. Chang, Y.S. Chen, T.W. Lin, H.S. Hsu, C.W. Wu, Coexpression of Oct4 and Nanog enhances malignancy in lung adenocarcinoma by inducing cancer stem celllike properties and epithelial–mesenchymal transdifferentiation, Cancer Res. 70 (2010) 10433–10444. [21] C. Wu, B.A. Alman, Side population cells in human cancers, Cancer Lett. 268 (2008) 1–9. [22] D. Beier, P. Hau, M. Proescholdt, A. Lohmeier, J. Wischhusen, P.J. Oefner, L. Aigner, A. Brawanski, U. Bogdahn, C.P. Beier, CD133(+) and CD133() glioblastoma-derived cancer stem cells show differential growth characteristics and molecular profiles, Cancer Res. 67 (2007) 4010–4015. [23] K. Polyak, R.A. Weinberg, Transitions between epithelial and mesenchymal states: acquisition of malignant and stem cell traits, Nat. Rev. Cancer 9 (2009) 265–273.
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77 [24] A.I. Neimark, E.A. Klyzhina, B.A. Neimark, M.A. Mel’nik, Effects of transcranial magnetotherapy on electroencephalographic parameters in females with overactive urinary bladder, Urologiia (2007) 40–44. [25] R. Chetty, S. Serra, S. Salahshor, Nuclear expression of E-cadherin, Am. J. Surg. Pathol. 32 (2008) 1269–1270. [26] L.A. Boyer, T.I. Lee, M.F. Cole, S.E. Johnstone, S.S. Levine, J.P. Zucker, M.G. Guenther, R.M. Kumar, H.L. Murray, R.G. Jenner, D.K. Gifford, D.A. Melton, R. Jaenisch, R.A. Young, Core transcriptional regulatory circuitry in human embryonic stem cells, Cell 122 (2005) 947–956. [27] M. Boiani, H.R. Scholer, Regulatory networks in embryo-derived pluripotent stem cells, Nat. Rev. Mol. Cell Biol. 6 (2005) 872–884. [28] F. Cavaleri, H.R. Scholer, Nanog: a new recruit to the embryonic stem cell orchestra, Cell 113 (2003) 551–552. [29] I.K. Park, D. Qian, M. Kiel, M.W. Becker, M. Pihalja, I.L. Weissman, S.J. Morrison, M.F. Clarke, Bmi-1 is required for maintenance of adult self-renewing haematopoietic stem cells, Nature 423 (2003) 302–305. [30] A.V. Molofsky, S. He, M. Bydon, S.J. Morrison, R. Pardal, Bmi-1 promotes neural stem cell self-renewal and neural development but not mouse growth and survival by repressing the p16Ink4a and p19Arf senescence pathways, Genes Dev. 19 (2005) 1432–1437. [31] S. Liu, G. Dontu, I.D. Mantle, S. Patel, N.S. Ahn, K.W. Jackson, P. Suri, M.S. Wicha, Hedgehog signaling and Bmi-1 regulate self-renewal of normal and malignant human mammary stem cells, Cancer Res. 66 (2006) 6063–6071.
77
[32] U.I. Ezeh, P.J. Turek, R.A. Reijo, A.T. Clark, Human embryonic stem cell genes OCT4, NANOG, STELLAR, and GDF3 are expressed in both seminoma and breast carcinoma, Cancer 104 (2005) 2255–2265. [33] J. Wen, J.Y. Park, K.H. Park, H.W. Chung, S. Bang, S.W. Park, S.Y. Song, Oct4 and Nanog expression is associated with early stages of pancreatic carcinogenesis, Pancreas 39 (2010) 622–626. [34] Y. Pan, J. Jiao, C. Zhou, Q. Cheng, Y. Hu, H. Chen, Nanog is highly expressed in ovarian serous cystadenocarcinoma and correlated with clinical stage and pathological grade, Pathobiology 77 (2010) 283–288. [35] S.W. Bruggeman, D. Hulsman, E. Tanger, T. Buckle, M. Blom, J. Zevenhoven, O. van Tellingen, M. van Lohuizen, Bmi1 controls tumor development in an Ink4a/ Arf-independent manner in a mouse model for glioma, Cancer Cell 12 (2007) 328–341. [36] J.J. Schuringa, E. Vellenga, Role of the polycomb group gene BMI1 in normal and leukemic hematopoietic stem and progenitor cells, Curr. Opin. Hematol. 17 (2010) 294–299. [37] C.C. Chang, G.S. Shieh, P. Wu, C.C. Lin, A.L. Shiau, C.L. Wu, Oct-3/4 expression reflects tumor progression and regulates motility of bladder cancer cells, Cancer Res. 68 (2008) 6281–6291. [38] Y. Atlasi, S.J. Mowla, S.A. Ziaee, A.R. Bahrami, OCT-4, an embryonic stem cell marker, is highly expressed in bladder cancer, Int. J. Cancer 120 (2007) 1598– 1602.
Contents lists available at SciVerse ScienceDirect
Cancer Letters journal homepage: www.elsevier.com/locate/canlet
Cancer stem-like cells contribute to cisplatin resistance and progression in bladder cancer Yi Zhang a, Zhi Wang a, Jin Yu a, Jia zhong Shi a, Chun Wang a, Wei hua Fu b, Zhi wen Chen b,c,⇑, Jin Yang a,⇑ a
Department of Cell Biology, The Third Military Medical University, Chongqing 400038, China Urology Institute of PLA, Southwest Hospital, The Third Military Medical University, Chongqing 400038, China c Southwest Cancer Center, Southwest Hospital, The Third Military Medical University, Chongqing 400038, China b
a r t i c l e
i n f o
Article history: Received 21 November 2011 Received in revised form 8 February 2012 Accepted 8 February 2012
Keywords: Cancer stem-like cell Bladder cancer Cisplatin resistance Epithelial–mesenchymal transition
a b s t r a c t A variety of cancer stem-like cells (CSCs) have been shown to be responsible for cancer tumorigenicity, relapse and metastasis. Despite several reports demonstrating the presence of CSCs in human bladder cancer, their identities are still under debate, and few studies have examined their roles in cisplatin resistance and related tumor progression. In this study, a subpopulation of CSCs was enriched following cisplatin selection from the bladder cell line T24. The cisplatin-resistant T24 cells displayed a greater self-renewal capacity as demonstrated by higher levels of sphere formation and stem cell marker expression, contained a larger proportion of side population cells and exhibited higher tumorigenicity. They also possessed epithelial–mesenchymal transition characteristics. Furthermore, a strong correlation between the levels of Bmi1 and Nanog expression and the degree of malignancy of urothelial cell carcinomas tissues was observed. We provide the first direct evidence that CSC-like cells exist in the population of cisplatin-resistant bladder cancer cells and may play a role in the progression and drug resistance of bladder cancer. Ó 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Bladder cancer is the fifth most common cancer in Western society and has a global incidence of more than 356,000 new cases a year and an estimated prevalence of 2.7 million cases [1]. After bladder cancer progresses to metastatic disease, the only treatment option is cisplatin-based combination chemotherapy [2]. Cisplatin is a platinum-based compound that forms both intra- and inter-strand DNA adducts. Cisplatin has a broad spectrum of antitumor activity and is widely used in the treatment of various malignant tumors [3]. However, intrinsic or acquired resistance to cisplatin severely limits the therapeutic potential of this drug. The molecular mechanisms that underlie this chemoresistance are still largely unknown. A more detailed understanding of the molecular mechanisms by which tumor cells survive chemotherapy is likely to lead to novel therapeutic targets with more successful outcomes. Emerging lines of evidence have associated chemoresistance with a pool of self-renewing malignant progenitors known as can⇑ Corresponding authors. Address: Department of Cell Biology, The Third Military Medical University, Chongqing 400038, China. Tel.: +86 23 68752260; fax: +86 23 65460268 (J. Yang), Urology Institute of PLA, Southwest Cancer Center, Southwest Hospital, The Third Military Medical University, Chongqing 400038, China. Tel.: +86 23 68765817; fax: +86 23 65460268 (Z. Chen). E-mail addresses: [email protected] (Z. Chen), [email protected] (J. Yang). 0304-3835/$ - see front matter Ó 2012 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2012.02.010
cer stem-like cells (CSCs) or cancer-initiating cells (CICs). Such cells share many properties with somatic stem cells, including self-renewal and multi-potent differentiation potential. Cancer stem cells have been proposed to be responsible for cancer tumorigenicity, relapse and metastasis [4,5]. CSCs appear to be highly resistant to various chemotherapies [6–8]. CSCs have also been found to exhibit a number of genetic and cellular adaptations that confer drug resistance. These adaptations include relative dormancy/slow cell cycle kinetics, efficient DNA repair, high expression of multidrugresistance-type membrane transporters and resistance to apoptosis [9,10]. The existence of CSCs has also recently been reported in human bladder cancer, and multiple potential markers have been identified. Oates et al. and Ning et al. identified SP (side-population) cells from primary bladder cancer and the T24 cell line [11,12]. In 2008, Yang and Chang showed that both the EMA (Epithelial Membrane Antigen) subset of cells and the CD44v6+ subset of cells are able to form colonies and could be bladder cancer-initiating cells [13]. Bentivegna et al. isolated CSCs from primary transitional cell carcinomas that had the ability to form spherical colonies [14]. Chan et al. and He et al. also described the isolation and characterization of a tumor-initiating cell subpopulation from primary human bladder cancer cells based on the CD44+CK5+CK20 or 67-kDaR+CK17+CEACAM6(CD66c) phenotypes [15,16]. To date, there have been the fewer than 10 reports about bladder cancer stem-like cells, and the localization and characteristics
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77
of human bladder stem cells are still under debate. Moreover, there is no direct evidence that bladder CSCs contribute to cisplatin resistance. It is becoming increasingly important to understand the molecular mechanisms of the self-renewal and differentiation of CSCs, which are likely related to tumor growth. The objectives of our current study are to molecularly characterize bladder CSCs, to investigate the role of CSCs in cisplatin resistance in bladder cancer, and to examine the clinical relevance of bladder CSCs. 2. Materials and methods
71
fibronectin (1:200; Santa Cruz Biotechnology). After the FITC-conjugated secondary antibodies had been applied, the nuclei were counterstained with DAPI. The staining was examined under a laser scanning confocal microscope (Leica, Germany).
2.7. Immunohistochemistry The paraffin sections were dewaxed with xylene and rehydrated in descending concentrations of ethanol. The endogenous peroxidase was inhibited, and the slides were incubated with antibodies against Nanog (1:200; Abcam, Inc., Cambridge, UK), Bmi1 (1:200; Abcam, Inc., Cambridge, UK), E-cadherin (1:100; Santa Cruz Biotechnology) or fibronectin (1:200; Santa Cruz Biotechnology). The expression levels of Nanog and Bmi1 were graded as described in Ref. [18].
2.1. Patients and tissue specimens Bladder urothelial carcinoma tissue samples were obtained from 51 patients (33 males and 18 females) by transurethral bladder tumor resection and radical cystectomy at Southwest Hospital, Third Military Medical University. The patients’ mean age was 65 years (ranging, 35–79). All samples were collected between 2006 and 2010. The tumor tissues were examined by one pathologist, and the grade was determined according to WHO criteria (2004). The research was approved by the ethics board of Southwest Hospital, Third Military Medical University.
2.8. Western blot analysis Equal amounts of protein lysates were subjected to SDS-PAGE and transferred to a PVDF membrane by electroblotting. The antibodies against Nanog (1:2000 dilution), Bmi1 (1:2000 dilution), Oct4 (1:2000 dilution), E-cadherin (1:1000 dilution) and fibronectin (1:5000 dilution) were used. A b-actin antibody at a 1:10000 dilution was used as a loading control. The protein bands were visualized using an imaging system (Bio-Rad, Hercules, CA).
2.2. Cell culture and drug selection The human bladder cancer cell line T24 was obtained from the American Type Culture Collection (ATCC, Rockville, MD) and maintained in RPMI-1640 with 10% fetal bovine serum (FBS) (Invitrogen, Carlsbad, CA) at 37 °C in a 5% CO2 atmosphere. T24 cells were treated with 30 lM cisplatin (Sigma-Aldrich Corp., St. Louis, MO) for 4 h and maintained in serum-free RPMI-1640 medium (SFM) supplemented with 5 lg/ml insulin (Sigma), 20 ng/ml EGF (Peprotech Inc., Rocky Hill, NJ) and 20 ng/ ml bFGF (Peprotech). Cisplatin (5 lM) was routinely added to the culture medium every other day. The cells were treated with the drug at least two times as described above. The surviving cells were considered drug-resistant cells (DRCs). The cells were dissociated and plated in triplicate at a density of 200 cells per well in 6-well plates. The total number of colonies (P20 cells) and the number of large colonies (P50 cells) were determined under a microscope. 2.3. Sphere formation and self-renewal assay The cells were inoculated at a density of 2 104 cells/well in ultra-low-attachment 6-well plates and grown in a serum-free medium supplemented with insulin, EGF and bFGF. After 5–10 days in culture, colonies that contained >20 cells were counted. In addition, to evaluate the self-renewal ability, cell spheres were enzymatically dissociated with 0.05% trypsin for 10 min at 37 °C and single-cell suspension was seeded. The secondary sphere formation was measured as described above. 2.4. Side population analysis The side population (SP) analysis was performed as described previously [17]. Briefly, the single-cell suspensions were stained with Hoechst 33342 (Sigma-Aldrich) at a final concentration of 5 lg/ml and incubated in the dark for 90 min at 37 °C with mixing at intervals. A negative control was incubated with 50 lM verapamil (Sigma-Aldrich) for 30 min at 37 °C before Hoechst 33342 was added. The FACS analysis was performed using 450 nm (Hoechst blue) and a 675 nm (Hoechst red) filters after excitation with a 350 nm UV light.
2.9. Cell migration The cells were cultured in a confluent monolayer and then scratched with a 1 mL pipette tip to create a cell-free area in 6-well culture plates. Mitomycin C (5 lg/mL) was added to inhibit cell proliferation, and the wound was photographed at 0 h, 12 h and 24 h.
2.10. In vivo xenograft experiments 1 105 Cells were resuspended in 200 ll PBS and injected subcutaneously into the dorsa of 4-week-old nude mice. The engrafted mice were observed every 2 days until tumors formed. The mice were sacrificed by cervical dislocation, and the xenograft tumors were resected, fixed in 10% phosphate-buffered formalin, embedded in paraffin for sectioning and processed for H&E staining and immunohistochemistry. The mice were maintained according to the guidelines for animal experiments of the Third Military Medical University.
2.11. Statistical methods The correlation between the expression level of Nanog and Bmi1 and clinical grade was analyzed by either the chi-squared test or Fisher’s exact test. For the quantitative data, the difference between groups was assessed with Student’s t-test. The Kaplan-Meier estimate was used for the survival analysis. A difference was considered statistically significant at P < 0.05 or P < 0.01. All analyzes were performed with SPSS (Version 13.0).
3. Results 3.1. Stem-related genes are overexpressed in bladder cancer
2.5. Semi-quantitative reverse transcriptase-polymerase chain reaction (RT-PCR) The cells were harvested, and the total RNA was extracted using the RNeasy Total RNA system (Qiagen GmbH, Hilden, Germany) following the manufacturer’s protocol. The first-strand cDNA was synthesized using the Reverse Transcriptase System (Invitrogen Carlsbad, CA), and the target cDNA was amplified using primers for the Oct4, Nanog and Bmi1 genes. The sequences of the primers were as follows: Nanog, forward: 50 -CCGAAGAATAGCAATGGTG-30 and reverse: 50 GGAGAATTTGGCTGGAACT-30 ; Oct4, forward: 50 -TGGAGAAGGAGAAGCTGGAGCAAAA-30 and reverse: 50 -GGCAGATGGTCGTTTGGCTGAATA-30 ; Bmi1, forward: 50 ACTATCGTCCAATTTGCT-30 and reverse: 50 -AATGCCATCTGATTCTTAC-30 ; and actin, forward: 50 -CATCTCTTGCTCGAAGTCCA-30 and reverse: 50 ATCATGTTTGAGACCTTCAACA-30 . Aliquots (8 ll) of the amplification products were loaded onto 1.5% agarose gels, separated by electrophoresis and visualized by ethidium bromide staining. 2.6. Immunofluorescent staining The cells were fixed with 4% paraformaldehyde, blocked with 10% normal goat serum, and incubated with a primary antibody against CD44 (1:100; Santa Cruz Biotechnology, Santa Cruz, CA, USA), E-cadherin (1:100; Santa Cruz Biotechnology) or
CSCs are considered to share similar features with normal stem cells and are proposed to positively correlate with the malignant phenotype. To clarify the clinical role of bladder CSCs, we chose to analyze the expression of ‘‘stemness genes’’ such as nanog and bmi1 in different clinical grades of human bladder cancer tissues. Nanog and Bmi1 overexpression has been shown in some kinds of CSCs, suggesting that the function and regulation of these genes are essential for CSC self-renewal and tumorigenesis [19,20]. All parameters of immunohistochemical detection are summarized in Table 1 and Fig. 1. The statistical analysis showed that increased expression of Nanog and Bmi1 in bladder cancer was significantly associated with a high pathological grade (P = 0.001 and 0.029, respectively). Based on that Nanog and Bmi1 strongly expressed cells may represent a more stem-like population, our result suggested that CSCs may contribute to the tumorigenesis of human bladder cancer and Nanog and Bmi1 may be potential biological and prognostic markers in tumors.
72
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77
Table 1 The expression of Nanog and Bmi1 in bladder cancer tissues.
PUNLMP Low-grade High-grade
Low (%) Nanog/Bmi1
Moderate (%) Nanog/Bmi1
Intense (%) Nanog/Bmi1
N
P value Nanog/Bmi1
14(70.0)/12(60.0) 3(21.4)/4(28.6) 2(11.8)/3(17.7)
4(20.0)/5(25.0) 6(42.9)/6(42.8) 5(29.4)/4(23.5)
2(10.0)/3(15.0) 5(35.7)/4(28.6) 10(58.8)/10 (58.8)
20 14 17
0.001/0.029
Five random views were chosen and 100 cells per view were counted for analysis. The percentage <10 was defined as negative. It was reported as low (10–29%), moderate (30–60%) and intense (>60%) expression. PUNLMP: Papillary Urothelial Neoplasm of Low Malignant Potential; low-grade: low-grade papillary carcinoma; high-grade: highgrade papillary carcinoma.
Fig. 1. The immunochemical expression of Bmi1 and Nanog in human bladder cancer. (A–C) The expression of Nanog. (D–F) The expression of Bmi1. (A, D) PUNLMP; (B, E) Low-grade; (C, F) High-grade. Scale bar: 50 lm.
3.2. DR-T24 cells display self-renewal capacity A variety of CSCs has a drug-resistant phenotype in vitro and is proposed to play an important role in the drug resistance of tumors. To provide direct evidence that CSCs exist in bladder cancer, we first attempted to enrich CSCs from cisplatin-resistant bladder cancer cells. The cisplatin-resistant T24 cells (DR-T24) were selected as described in the Materials and Methods. The total and large colonies were counted to determine the clonogenic capacity of these cells. The cells gave rise to clones at an average frequency of 23 ± 2.9 colonies/well for parental T24 cells and 39 ± 5.8 colonies/well for DR-T24 cells at 7 days (P < 0.01), and the large clone-formation efficiencies of parental and cisplatin-resistant T24 cells were 6.4 ± 1.1 colonies/well and 10.8 ± 2.4 colonies/well (P < 0.01), respectively (Fig. 2A and B). Sphere formation has been well described as a typical characteristic of CSCs that reflects the potential for self-renewal [4]. We evaluated the ability of T24 and DR-T24 cells to generate spherical colonies in an ultra-low attachment and serum-starved culture system. After 7 days of culture, spherical colonies were counted. The DR-T24 cells exhibited a 3-fold higher frequency of spherical colonies than the parental T24 cells (12.64 ± 1.58% for the DRT24 cells versus 4.28 ± 0.98% for the parental T24 cells) (Fig. 2C and D). Furthermore, the self-renewing capacity of DR-T24 cells were assessed by dissociating primary spheres into single cells, which were incubated in anchorage-independent, stem cell culture conditions. After 5 days, these cells gave rise to secondary spheres (15.08 ± 2.44%) (Fig. 2E). Statistical analysis of the percentage of sphere formation is shown in Fig. 2F. Collectively, these data indi-
cate that the stem-like feature of DR-T24 cells can be maintained in vitro. 3.3. DR-T24 cells contain a larger side population The existence of side populations within different cancer cell lines bearing some stem cell-like features is well documented [21]. It has also been published that SP cells identified from primary bladder cancer and the T24 cell line harbor tumor-initiating cell characteristics [11,12]. We evaluated the SP fraction in both the T24 parental cells and the DR-T24 cells. A higher percentage of SP cells was observed in the DR-T24 cells compared with the T24 parental cells. The flow cytometric analysis showed that the percentage of SP cells increased from 2.5% in the parental T24 cells to 7.7% in the DR-T24 cells (P < 0.05) (Fig. 3). These results suggested that under drug selection, cancer stem-like cells might be enriched. 3.4. DR-T24 cells express stem cell markers To confirm the stem cell phenotype of the DR-T24 cells, the expression of ES cell marker genes was evaluated. The RT-PCR analysis showed that the DR-T24 cells expressed higher mRNA levels of Nanog, Bmi1 and Oct4 (Fig. 4A). Consistent with the RT-PCR data, the expression of these proteins was also stronger in DR-T24 cells, as demonstrated by immunoblotting (Fig. 4B). Because it has also been reported that CD44 may be a potential bladder CSC marker, immunofluorescent staining was performed to determine the level of protein expression of CD44. The DR-T24 cells also ex-
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77
73
Fig. 2. The higher proliferative potential and sphere formation of the DR-T24 population treated with cisplatin. (A) The total clone-forming capacity (>20 cells/clone). (B) The large clone-forming capacity (>50 cells/clone). (C) Primary tumor spheres originated from parental T24 cells. (D) Primary tumor spheres originated from DR-T24 cells. (E) Secondary spheres derived from single cells of primary tumor sphere. (F) The statistical analysis of the sphere formation efficiency. P < 0.01. Scale bars: (C–E) 100 lm.
Fig. 3. Flow cytometric analysis of side population in DR-T24 cells and parental T24 cells. As control, the SP fraction was less to 0.1% under the verapamil pre-incubated.
74
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77
lower than that of the parental T24-implanted mice (P < 0.01) (Fig. 5B). The H&E staining showed that the xenograft tumor cells originating from the DR-T24 cells contained numerous large and heterogeneous nuclei surrounded by a high density of microvessels, indicative of a more malignant phenotype (Fig. 5C). The isolated xenograft tumor cells were also able to form secondary spheres in stem cell medium, indicating their self-renewing capability (Fig. 5D). Immunofluorescent analysis of these xenografts revealed that CD44 was expressed at a significantly higher level in the tumors from the DR-T24 cells than those of the parental cells (Fig. 5E). These results indicated that the DR-T24 cells contained a larger population of bladder CSCs that promote tumorigenesis.
3.6. DR-T24 cells display features of EMT
Fig. 4. Expression of stem cell markers in T24 cells and DR-T24 cells. (A) The expression of Oct4, Nanog and Bmi1 mRNA detected by RT-PCR. (B) Western blot analysis of Oct4, Nanog, Bmi1 and CD44 protein expression. (C) Immunofluorescent staining of CD44 in cultured both T24 cells and DR-T24 cells. Scale bar: 25 lm.
pressed higher levels of CD44 than the parental T24 cells (Fig. 4B and C). Together, these results showed that potential cancer stem-like cells expressing ‘‘stemness’’ genes can be enriched by drug selection. 3.5. DR-T24 cells display highly tumorigenic behavior in vivo High tumorigenic potential is considered a hallmark of CSCs. To determine whether the DR-T24 cells exhibited higher tumorigenicity, the parental T24 cells and the DR-T24 cells were subcutaneously injected into nude mice for transplanted tumorigenicity analysis (n = 10, each). In the mice implanted with the parental T24 cells, tumors were visible after 32–52 days (average = 40.6 days). In contrast, the DR-T24 cells produced tumors within 18–30 days (average = 24.4 days) (Fig. 5A). The mean survival rate of the DR-T24-implanted mice was also significantly
Current data demonstrate that CSCs have a mesenchymal cell phenotype and are associated with tumor initiation, growth, metastasis, and therapeutic resistance [12]. To elucidate the role of epithelial–mesenchymal transition (EMT) in cancer stem cell generation, we compared the cell migration abilities of the DRT24 and control cells. In an in vitro wound-healing assay, the DRT24 cells migrated into the wound area more rapidly than control cells did (Fig. 6A). These data suggest that the DR-T24 cells have a stronger cell migration and invasion abilities and thus have a greater malignant potential. We next examined the expression levels of EMT markers in both the parental and DR-T24 cells. Mesenchymal and epithelial specific markers (fibronectin and E-cadherin, respectively) were chosen. Ecadherin expression is decreased both in the DR-T24 cells cultured in vitro and in the xenografts originating from cell transplantation, whereas fibronectin is expressed at a higher level in DR-T24 cells (Fig. 6C and D). Our results are consistent with the typical EMT features, including the loss of epithelial markers and the concomitant gain of mesenchymal markers. These observations suggest that the bladder CSCs underwent an EMT process. Upon Immunocytochemical analysis of E-cadherin localization, we were surprised to discover that E-cadherin appeared both in the cytoplasm and the nucleus of the DR-T24 cells, in contrast with
Fig. 5. The subcutaneous tumorigenicity in nude mice. (A) 1 105 cells were injected into nude mice. the left: T24 cells; the right: DR-T24 cells. Statistical graph was shown on the top right. P < 0.01. (B) The survival curves of mice injected with the parental T24 cells (dashed line) and DR-T24 cells (solid line). (C) The morphology of tumor developed from T24 cells (left) and DR-T24 cells (right) (H&E). (D) Secondary sphere formed after xenograft tumor cells re-seeded. (E) Immunofluorescent staining of CD44 in xenograft tumor derived from parental T24 cells (up) and DR-T24 cells (down). Scale bars: (C, E) 50 lm; (D) 100 lm;.
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77
75
Fig. 6. DR-T24 cells display EMT feature. (A) Micrographs of cell wound closure monitored by light microscopy and photographed every 12 h. (B) Immunofluorescent and (C) Western blot analysis of E-cadherin, Fibronectin expression in cultured T24 and DR-T24 cells. (D) Immunohistochemical staining of E-cadherin and Fibronectin in the tumor formed by implanting T24 cells and DR-T24 cells. Scale bars: (A) 50 lm; (B) 25 lm; (D) 50 lm .
the membrane localization in the parental T24 cells (Fig. 6B). This is the first report showing that cytoplasmic and nuclear E-cadherin appears present in both bladder cancers and cisplatin-resistant cells, suggesting a potentially important role for E-cadherin mislocalization in both processes. 4. Discussion To investigate the role of CSCs in cisplatin resistance and determine the molecular characteristics of bladder CSCs and their clinical relevance, a subpopulation of cancer stem-like cells from the bladder cell line T24 was enriched by cisplatin selection and was found to possess the features of both stem cells and cells that have undergone EMT. There was a high incidence of Bmi1 and Nanog overexpression in the bladder urothelial cell carcinoma tissues analyzed, and a strong correlation between the overexpression of Bmi1 and Nanog and the degree of malignancy of bladder tumors was observed. We provide the first direct evidence that CSC-like cells do exist among cisplatin-resistant cells in bladder cancer and may play a role in the progression and drug resistance of bladder cancer. The putative bladder cancer stem cells can be considered to be mediators of resistance to current therapies and represent strong candidate biological targets. To develop a therapy that targets bladder CSCs, the first key step is to identify and isolate CSCs. As we described in the introduction, the results of a few recent reports clearly support that a functionally distinct subpopulation of CSCs can be isolated from bladder cancer using either prospective surface marker-based FACS analysis or SP cell sorting. However, the relationship among the different CSC populations isolated using various markers or methods is unclear [11–16]. It is important that the bladder CSC phenotype be further defined before CSCs can be reliably identified and isolated. Among the methods of identification of CSCs, surface-marker-based FACS analysis and sorting are the most widespread methods. Unfortunately, the more controversial question arising from these recent studies is whether CSCs exhibit a common marker(s). It has been demonstrated that different cancer stem cells harboring different surface markers may exist in a single type of cancer. For example, in addition to CD133+ CSCs in glioma, CD133 tumor cells possess apparent stem cell-like properties but distinct molecular profiles and growth characteristics [22]. Other ways to identify CSCs are based on the phenotypic char-
acteristics, such as side population (SP) analysis. However, this technique can be hampered by UV exposure and the loss of cytoactivity by SP cells [21]. It should be noted that all of the isolation strategies used to date have their shortfalls, and perhaps a combination of different isolation methods will be required to enhance the purity of cancer stem cells. In this study, we found that CSCs could be enriched from T24 cells by cisplatin selection. The DR-T24 cells were observed to possess an increased ability to form tumor spheres and had a high clonogenic efficiency compared with the parental T24 cells (Fig. 2). The DR-T24 cells also express higher levels of ESC markers (e.g., Bmi1, Nanog, and Oct4) (Fig. 4). This result suggests that a selfrenewing population of CSCs is present in T24 cells. Immunohistochemical analysis of the DR-T24 cells also revealed a larger population of CD44+ and SP cells, which is consistent with results found in other reports (Fig. 3). To obtain in vivo evidence and provide evidence of differentiation, we implanted the DR-T24 cells and the parental T24 cells into nude mice and found that the DR-T24 cells have a higher tumorigenic potential. Immunostaining revealed an obvious increase in the level of CD44 in the tumors generated from the DR-T24 cells (Fig. 5E). Hence, our study provides the first direct evidence that CSC cells do exist among cisplatin-resistant bladder cancer cells and may thus play a role in the drug resistance of bladder cancer. This system may be a valuable model for the further study of both CSCs and chemoresistance. The preliminary drug selection of cancer cells followed by surface marker-based cell sorting may decrease the heterogeneity of the CSC population and afford a more efficient method to isolate the bladder cancer CSCs. Although anti-tumor drug resistance and cancer metastasis often occur at the same time clinically, there are no reports providing direct evidence of acquired cisplatin resistance in bladder cancer cells with an obvious invasive and metastasis phenotype. Based on the association of CSCs with cancer relapse and metastasis and our strong evidence that CSCs do exist among cisplatin-resistant bladder cancer cells, we had reason to raise the question of whether the DR-T24 cells are characterized by a more invasive phenotype. Indeed, an in vitro wound healing assay demonstrated that the DR-T24 cells migrate much more rapidly than the control cells, strongly suggesting that cisplatinresistant bladder cancer cells have a more powerful cell migration ability (Fig. 6A).
76
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77
The concept of epithelial–mesenchymal transition (EMT) initially developed in the field of embryology and has recently has been extended to cancer progression. It has been demonstrated that EMT is involved in carcinogenesis, invasion, metastasis, recurrence, and chemoresistance [23]. CSCs can also be induced by EMT [24]. To molecularly characterize the cisplatin-resistant cells, including the bladder CSCs, we further investigated the role of EMT and its involvement in the DR-T24 cellular phenotype. Compared with the level in parental T24 cells, the level of epithelial marker E-cadherin is decreased both in the DR-T24 cells cultured in vitro and in the tumor xenografts, In addition, the mesenchymal marker fibronectin is more highly expressed in DR-T24 cells. Our results are thus consistent with the typical EMT cellular features, including the concomitant loss of epithelial markers and gain of mesenchymal markers [23]. This first report of EMT properties in cisplatin-resistant bladder cancer cells may provide the initial clue to explain the high cell migration ability of the DR-T24 cells shown above. These observations may stimulate future interest in defining the molecular mechanisms underlying this epithelial-to-mesenchymal ‘‘switch’’ in the generation of the bladder CSCs and the acquisition of cisplatin resistance. To our surprise, in contrast to the membrane localization in the parental T24 cells, E-cadherin appeared in both the cytoplasm and the nucleus of the DR-T24 cells (Fig. 6). These findings are contrary to the traditional theory that E-cadherin mediates intercellular adhesion. Recent studies have noted nuclear E-cadherin expression in a variety of tumors [25]. However, the biological or pathological roles of cytoplasmic and nuclear E-cadherin are still unclear. Our discovery of the cytoplasmic and nuclear localization of E-cadherin in cisplatin-resistant bladder cancer cells implies that altered Ecadherin localization plays an important role in drug resistance and bladder CSC origination. Despite the identification of CSCs in many diseases and the potential of CSCs to explain the failure of many anticancer therapies to extend overall survival, few data actually exist to support that CSCs are clinically relevant. Based on the discovery that CSCs may contribute to cancer tumorigenicity, relapse and metastasis, it has been proposed that the expression level of stemness genes or key regulatory factors related to stem cell self-renewal may be associated with tumor progression. Among the stemness genes, the transcription factors Sox2, Oct3/4 and Nanog form a core regulatory network that coordinately determines the self-renewal and differentiation of embryonic stem cells [26–28]. Bmi1 has also emerged as a key regulator of self-renewal in several cell types, including hematopoietic, neural and mammary stem cells [29– 31]. Some studies have suggested that these factors may play a role in human malignancy. Nanog is overexpressed in many tumors, including carcinomas of the breast, lung, pancreas and ovary [20,32–34]. Ample evidence also exists linking Bmi1 expression levels to the self-renewal and maintenance of both normal and cancer stem cells [35,36]. Chang et al. and Atlasi et al. recently also discovered that Oct3/ 4 is highly expressed in human bladder cancer and is associated with disease progression, increased metastasis, and shorter cancer-related survival [37,38]. We therefore hypothesized that other embryonic stemness factors or key regulatory factors related to stem cell self-renewal may also contribute to the progression and drug resistance of bladder cancer. In our study, Nanog and Bmi1 were found for the first time to be highly expressed in bladder cancer tissues, and the expression of these genes was found to be positively correlated with clinical stage and grade (Fig. 1, Table 1). Together, these findings suggest not only that the progression and cisplatin resistance of bladder cancer may be related to the existence of cancer stem cells but that Nanog and Bmi1 may also be potential biological and prognostic markers in tumors.
Acknowledgments Supported by Grant (No. 30973385) from National Nature Science Foundation of China (NSFC) and Grant (No. 2010CB529402) from the National Basic Research Program of China (‘‘973’’ Program). References [1] B.W. van Rhijn, M. Burger, Y. Lotan, E. Solsona, C.G. Stief, R.J. Sylvester, J.A. Witjes, A.R. Zlotta, Recurrence and progression of disease in non-muscleinvasive bladder cancer: from epidemiology to treatment strategy, Eur. Urol. 56 (2009) 430–442. [2] J. Bellmunt, S. Albiol, C. Suarez, J. Albanell, Optimizing therapeutic strategies in advanced bladder cancer: update on chemotherapy and the role of targeted agents, Crit. Rev. Oncol. Hematol. 69 (2009) 211–222. [3] D.B. Zamble, S.J. Lippard, Cisplatin and DNA repair in cancer chemotherapy, Trends Biochem. Sci. 20 (1995) 435–439. [4] P. Dalerba, R.W. Cho, M.F. Clarke, Cancer stem cells: models and concepts, Annu. Rev. Med. 58 (2007) 267–284. [5] M.F. Clarke, M. Fuller, Stem cells and cancer: two faces of eve, Cell 124 (2006) 1111–1115. [6] C.S. O’Brien, G. Farnie, S.J. Howell, R.B. Clarke, Are stem-like cells responsible for resistance to therapy in breast cancer, Breast Dis. 29 (2008) 83–89. [7] G. Bertolini, L. Roz, P. Perego, M. Tortoreto, E. Fontanella, L. Gatti, G. Pratesi, A. Fabbri, F. Andriani, S. Tinelli, E. Roz, R. Caserini, S. Lo Vullo, T. Camerini, L. Mariani, D. Delia, E. Calabro, U. Pastorino, G. Sozzi, Highly tumorigenic lung cancer CD133+ cells display stem-like features and are spared by cisplatin treatment, Proc. Natl. Acad. Sci. USA 106 (2009) 16281–16286. [8] H. Fujii, K. Honoki, T. Tsujiuchi, A. Kido, K. Yoshitani, Y. Takakura, Sphereforming stem-like cell populations with drug resistance in human sarcoma cell lines, Int. J. Oncol. 34 (2009) 1381–1386. [9] M. Dean, Cancer stem cells: implications for cancer causation and therapy resistance, Discov. Med. 5 (2005) 278–282. [10] R. Morrison, S.M. Schleicher, Y. Sun, K.J. Niermann, S. Kim, D.E. Spratt, C.H. Chung, B. Lu, Targeting the mechanisms of resistance to chemotherapy and radiotherapy with the cancer stem cell hypothesis, J. Oncol. 2011 (2011) 941876. [11] Z.F. Ning, Y.J. Huang, T.X. Lin, Y.X. Zhou, C. Jiang, K.W. Xu, H. Huang, X.B. Yin, J. Huang, Subpopulations of stem-like cells in side population cells from the human bladder transitional cell cancer cell line T24, J. Int. Med. Res. 37 (2009) 621–630. [12] J.E. Oates, B.R. Grey, S.K. Addla, J.D. Samuel, C.A. Hart, V.A. Ramani, M.D. Brown, N.W. Clarke, Hoechst 33342 side population identification is a conserved and unified mechanism in urological cancers, Stem Cells Dev. 18 (2009) 1515– 1522. [13] Y.M. Yang, J.W. Chang, Bladder cancer initiating cells (BCICs) are among EMACD44v6+ subset: novel methods for isolating undetermined cancer stem (initiating) cells, Cancer Invest. 26 (2008) 725–733. [14] A. Bentivegna, D. Conconi, E. Panzeri, E. Sala, G. Bovo, P. Vigano, S. Brunelli, M. Bossi, G. Tredici, G. Strada, L. Dalpra, Biological heterogeneity of putative bladder cancer stem-like cell populations from human bladder transitional cell carcinoma samples, Cancer Sci. 101 (2010) 416–424. [15] K.S. Chan, I. Espinosa, M. Chao, D. Wong, L. Ailles, M. Diehn, H. Gill, J. Presti Jr., H.Y. Chang, M. van de Rijn, L. Shortliffe, I.L. Weissman, Identification, molecular characterization, clinical prognosis, and therapeutic targeting of human bladder tumor-initiating cells, Proc. Natl. Acad. Sci. USA 106 (2009) 14016–14021. [16] X. He, L. Marchionni, D.E. Hansel, W. Yu, A. Sood, J. Yang, G. Parmigiani, W. Matsui, D.M. Berman, Differentiation of a highly tumorigenic basal cell compartment in urothelial carcinoma, Stem Cells 27 (2009) 1487–1495. [17] J. Wang, L.P. Guo, L.Z. Chen, Y.X. Zeng, S.H. Lu, Identification of cancer stem cell-like side population cells in human nasopharyngeal carcinoma cell line, Cancer Res. 67 (2007) 3716–3724. [18] X. Jia, X. Li, Y. Xu, S. Zhang, W. Mou, Y. Liu, D. Lv, C.H. Liu, X. Tan, R. Xiang, N. Li, SOX2 promotes tumorigenesis and increases the anti-apoptotic property of human prostate cancer cell, J. Mol. Cell Biol. 3 (2011) 230–238. [19] L. Jiang, J. Li, L. Song, Bmi-1, stem cells and cancer, Acta. Biochim. Biophys. Sin. (Shanghai) 41 (2009) 527–534. [20] S.H. Chiou, M.L. Wang, Y.T. Chou, C.J. Chen, C.F. Hong, W.J. Hsieh, H.T. Chang, Y.S. Chen, T.W. Lin, H.S. Hsu, C.W. Wu, Coexpression of Oct4 and Nanog enhances malignancy in lung adenocarcinoma by inducing cancer stem celllike properties and epithelial–mesenchymal transdifferentiation, Cancer Res. 70 (2010) 10433–10444. [21] C. Wu, B.A. Alman, Side population cells in human cancers, Cancer Lett. 268 (2008) 1–9. [22] D. Beier, P. Hau, M. Proescholdt, A. Lohmeier, J. Wischhusen, P.J. Oefner, L. Aigner, A. Brawanski, U. Bogdahn, C.P. Beier, CD133(+) and CD133() glioblastoma-derived cancer stem cells show differential growth characteristics and molecular profiles, Cancer Res. 67 (2007) 4010–4015. [23] K. Polyak, R.A. Weinberg, Transitions between epithelial and mesenchymal states: acquisition of malignant and stem cell traits, Nat. Rev. Cancer 9 (2009) 265–273.
Y. Zhang et al. / Cancer Letters 322 (2012) 70–77 [24] A.I. Neimark, E.A. Klyzhina, B.A. Neimark, M.A. Mel’nik, Effects of transcranial magnetotherapy on electroencephalographic parameters in females with overactive urinary bladder, Urologiia (2007) 40–44. [25] R. Chetty, S. Serra, S. Salahshor, Nuclear expression of E-cadherin, Am. J. Surg. Pathol. 32 (2008) 1269–1270. [26] L.A. Boyer, T.I. Lee, M.F. Cole, S.E. Johnstone, S.S. Levine, J.P. Zucker, M.G. Guenther, R.M. Kumar, H.L. Murray, R.G. Jenner, D.K. Gifford, D.A. Melton, R. Jaenisch, R.A. Young, Core transcriptional regulatory circuitry in human embryonic stem cells, Cell 122 (2005) 947–956. [27] M. Boiani, H.R. Scholer, Regulatory networks in embryo-derived pluripotent stem cells, Nat. Rev. Mol. Cell Biol. 6 (2005) 872–884. [28] F. Cavaleri, H.R. Scholer, Nanog: a new recruit to the embryonic stem cell orchestra, Cell 113 (2003) 551–552. [29] I.K. Park, D. Qian, M. Kiel, M.W. Becker, M. Pihalja, I.L. Weissman, S.J. Morrison, M.F. Clarke, Bmi-1 is required for maintenance of adult self-renewing haematopoietic stem cells, Nature 423 (2003) 302–305. [30] A.V. Molofsky, S. He, M. Bydon, S.J. Morrison, R. Pardal, Bmi-1 promotes neural stem cell self-renewal and neural development but not mouse growth and survival by repressing the p16Ink4a and p19Arf senescence pathways, Genes Dev. 19 (2005) 1432–1437. [31] S. Liu, G. Dontu, I.D. Mantle, S. Patel, N.S. Ahn, K.W. Jackson, P. Suri, M.S. Wicha, Hedgehog signaling and Bmi-1 regulate self-renewal of normal and malignant human mammary stem cells, Cancer Res. 66 (2006) 6063–6071.
77
[32] U.I. Ezeh, P.J. Turek, R.A. Reijo, A.T. Clark, Human embryonic stem cell genes OCT4, NANOG, STELLAR, and GDF3 are expressed in both seminoma and breast carcinoma, Cancer 104 (2005) 2255–2265. [33] J. Wen, J.Y. Park, K.H. Park, H.W. Chung, S. Bang, S.W. Park, S.Y. Song, Oct4 and Nanog expression is associated with early stages of pancreatic carcinogenesis, Pancreas 39 (2010) 622–626. [34] Y. Pan, J. Jiao, C. Zhou, Q. Cheng, Y. Hu, H. Chen, Nanog is highly expressed in ovarian serous cystadenocarcinoma and correlated with clinical stage and pathological grade, Pathobiology 77 (2010) 283–288. [35] S.W. Bruggeman, D. Hulsman, E. Tanger, T. Buckle, M. Blom, J. Zevenhoven, O. van Tellingen, M. van Lohuizen, Bmi1 controls tumor development in an Ink4a/ Arf-independent manner in a mouse model for glioma, Cancer Cell 12 (2007) 328–341. [36] J.J. Schuringa, E. Vellenga, Role of the polycomb group gene BMI1 in normal and leukemic hematopoietic stem and progenitor cells, Curr. Opin. Hematol. 17 (2010) 294–299. [37] C.C. Chang, G.S. Shieh, P. Wu, C.C. Lin, A.L. Shiau, C.L. Wu, Oct-3/4 expression reflects tumor progression and regulates motility of bladder cancer cells, Cancer Res. 68 (2008) 6281–6291. [38] Y. Atlasi, S.J. Mowla, S.A. Ziaee, A.R. Bahrami, OCT-4, an embryonic stem cell marker, is highly expressed in bladder cancer, Int. J. Cancer 120 (2007) 1598– 1602.