SEARCH

SEARCH BY CITATION

Keywords:

  • Cancer stem cell;
  • Uterine carcinosarcoma;
  • Malignant mixed Müllerian tumor;
  • CD133;
  • Müllerian duct

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Conclusion
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Cancer stem cells (CSCs) that display tumor-initiating properties have recently been identified. CD133, a surface glycoprotein linked to organ-specific stem cells, has been described as a marker of CSCs in different tumor types. We herein identify and characterize CSCs in human uterine carcinosarcoma (malignant mixed Müllerian tumor), which is one of the most aggressive and therapy-resistant gynecological malignancies and is considered to be of mesodermal origin. The CD133+ population was increased in uterine carcinosarcoma, and this population showed biphasic properties in the primary tumor. CD133+ cells predominantly formed spheres in culture and were able to differentiate into mesenchymal lineages. CD133+ cells were more resistant to cisplatin/paclitaxel-induced cytotoxicity in comparison with CD133 cells. A real-time polymerase chain reaction analysis of the genes implicated in stem cell maintenance revealed that CD133+ cells express significantly higher levels of Oct4, Nanog, Sox2, and Bmi1 than CD133 cells. Moreover, CD133+ cells showed a high expression level of Pax2 and Wnt4, which are genes essential for Müllerian duct formation. These CD133+ cells form serially transplantable tumors in vivo and the resulting CD133+ tumors replicated the EpCAM, vimentin, and estrogen and progesterone receptor expression of the parent tumor, indicating that CSCs likely differentiated into cells comprising the uterine carcinosarcoma tissue. Moreover, strong CD133 expression in both epithelial and mesenchymal elements in primary tumor demonstrated significant prognostic value. These findings suggest that CD133+ cells have the characteristics of CSCs and Müllerian mesenchymal progenitors. STEM CELLS 2011;29:1485–1495


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Conclusion
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Uterine carcinosarcoma (termed malignant mixed Müllerian tumor; MMMT) is a highly aggressive and primitive tumor composed of mixed malignant epithelial and mesenchymal components [1, 2]. This tumor is relatively uncommon, accounting for only 2%–5% of all uterine cancers, and mostly occurs in elder women. However, MMMT is associated with disproportionally higher mortality rates in comparison with other uterine corporeal malignancies [3, 4]. The mesenchymal components often show heterologous differentiation such as rhabdomyosarcoma, chondrosarcoma, and osteosarcoma [5], which is likely due in part to their Müllerian duct (MD) lineage, suggesting a stem cell tumor of the MD.

The cancer stem cells (CSCs) hypothesis holds that the cells composing a tumor are hierarchically organized with respect to their potential to initiate and sustain tumor growth. Only a rare subset of tumor cells, defined as CSCs, have the capacity to form tumors in serial xenotransplantation assays, and the ability to reestablish, at each in vivo passage, the hierarchical cell organization identified in leukemias [6], glioblastomas [7], melanomas [8], and different types of tumors [9], thus providing a biological basis for the design of rationally targeted therapy. Moreover, these CSCs express surface markers similar to those expressed by normal stem cells in each tissue [10].

CD133 (prominin-1/AC133) is a pentaspan membrane protein and was originally classified as a marker of primitive hematopoietic and neural stem cells [11]. The expression of CD133 has been recently associated with CSCs isolated from prostate [12], lung [13], brain [14], and ovarian cancers [15]. CD133+ cells with tumorigenic potential have also been identified in human endometrial cancer [16]. In soft tissues, CD133 has also been identified in various sarcomas, including Ewing's sarcoma [17] and osteosarcoma [18]. CD133+ Ewing's sarcoma cells express significantly higher levels of Oct4 and Nanog than their CD133 counterparts. It is presently unknown whether CD133+ cells with tumorigenic potential can be prospectively isolated from uterine carcinosarcoma.

In this study, we used the FU-MMT-1 cell line derived from a primary uterine carcinoscarcoma [19]. A cytogenetic analysis showed that the FU-MMT-1 cell line has only one chromosomal abnormality (trisomy of chromosome 8), thus suggesting a proliferation of mutated progenitor/stem cells [19]. Therefore, we thought that this cell line could be a good model to study CSCs in the human MD. To the best of our knowledge, there are only a few useful cell lines derived from human uterine carcinosarcomas, which have been fully characterized and used in vitro and in vivo to study this disease [19–21]; however, no study has so far examined CD133 expression in this primitive biphasic uterine tumor. According to a PubMed analysis, FU-MMT-1 was the only cell line that was used in both in vitro and in vivo studies, after the reported establishment of each particular cell line.

We herein report that, using FU-MMT-1, a human uterine carcinosarcoma cell line [19], we identified a population of carcinosarcoma cells expressing CD133 that fulfill the criteria of CSCs and display a Müllerian progenitor cell capacity both in vitro and in vivo.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Conclusion
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Cell Line

A human uterine carcinosarcoma cell line, FU-MMT-1, established previously from tissue isolated from a patient with uterine carcinosarcoma, was used in this study [19]. The FU-MMT-1 cells were grown in Ham's Dulbecco's modified Eagle's medium (DMEM)-12 with 10% fetal bovine serum and maintained at 37°C (humidified, 5% CO2); the medium was changed every 3–4 days and the cultures were trypsinized and recultured when they reached 85% confluence.

Magnetic and Cytofluorimetric Cell Separation by FACS

CD133+ and CD133 cells were isolated from cell culture by FACS or magnetic bead sorting using the MACS system (Miltenyi Biotech). For both separations, cells were incubated with the monoclonal CD133/1-PE (AC133 clone; Miltenyi Biotech, Auburn CA) for 15 minutes at 4°C. For magnetic separation, the cells were washed and incubated with PE-beads (Miltenyi Biotech) and selected using MS columns (Miltenyi Biotech), which retained positive cells linked by beads. For FACS separation, CD133/1-PE stained cells were sorted with a FACSAria II instrument (BD Pharmingen). In both instances, the purity of the CD133+ and CD133 cell populations was evaluated by a standard flow cytometry analysis. For detailed information refer to Supporting Information text.

For FACS analysis, the cells were labeled with a mouse anti-human CD29-PE (Integrin β1 chain), mouse anti-human CD44-FITC (Pgp-1, H-CAM, and Ly24), and mouse anti-human CD90-FITC (Thy-1) (all from BD Pharmingen, San Jose, CA). Cells were analyzed using a FACSCanto II (BD Pharmingen) and the FACSDiva analysis software program (BD Pharmingen) (Supporting Information text).

In Vitro Cell Characterization

After sorting the CD133+ and CD133 cells, each population's in vitro characteristics were determined by a serial analysis, including cell growth, sphere formation, and limiting dilution assay. In addition, the in vitro drug sensitivity assay and differentiation assay were performed as detailed in the Supporting Information text.

Immunohistochemistry

The immunohistochemical analyses were performed according to standard procedures. The antibody used were: anti-wide spectrum cytokeratin pAb (DBS, Pleasanton, CA), anti-EpCAM (OriGene, Rockville, MD), anti-vimentin mAb (Epitomics, Burlingame, CA), anti–smooth muscle actin (anti-SMA) mAb (Epitomics), anti-progesterone receptor (PGR)-α mAb (Dako, Tokyo, Japan; 1:800), anti-ER-β mAb (GeneTex, Irvine, CA), and anti-CD133 pAb (Abcam, Cambridge, MA). Technical details of immunostaining are reported in Supporting Information text.

CD133 staining was performed according to standard procedures using the paraffin-embedded sections from a total of 26 primary uterine carcinosarcomas. The quality of staining was judged according to the data in the literature about gene/protein expression of CD133 in various tissue types [11, 15, 16]. Depending on the expression level in each histological component, we (M.E. and B.C.) classified the samples into three groups; (a) strong expression of CD133 in both epithelial and mesenchymal components, (b) strong expression in the epithelial component but weak in the mesenchymal component, and (c) weak or no expression of CD133 in either component. Cases with different scoring were discussed to reach a consensus.

In Vivo Studies of Tumorigenicity

Single cells were resuspended in serum-free Ham's DMEM-12 (1% penicillin/streptomycin) with matrigel (BD Matrigel, Bedford, MA). A 200-μL suspension containing between 102 and 106 cells was injected orthotopically or s.c. into the flanks of 6- to 8-week-old female BALB/cA Jcl-nu athymic nude mice (Clea, Tokyo, Japan). All procedures are performed on animals were approved by the Institute of Experimental Animal Science, Fukuoka University Medical School (No.0910350). In vivo studies were performed as described in Supporting Information text.

Statistical Analysis

The results of the colony formation assay, drug resistance assay, real-time polymerase chain reaction (PCR) analysis, and in vivo tumorigenicity assay were analyzed by unpaired t-test. Overall survival (OS) was calculated with the Kaplan-Meier method and compared by using the log-rank test. A p value of <.05 was regarded as statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Conclusion
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Identification of CD133+ Cells in a Human Uterine Carcinosarcoma-Derived Cell Line

The FU-MMT-1 cells are composed exclusively of mesenchymal (sarcomatous) cells; including spindle, polygonal, and strap cells. These cells grew haphazardly in multiple layers and occasionally formed interlacing bundles chiefly showing rhabdomyoblastic differentiation, with expression of both epithelial and mesenchymal markers [19]. To determine whether the FU-MMT-1 cell line contains a distinct population of CSCs, we examined the expression of cell surface markers previously assigned to human CSCs, namely CD133 (Supporting Information Fig. S1A), CD29 (Supporting Information Fig. S1B), CD44 (Supporting Information Fig. S1C), and CD90 (Supporting Information Fig. S1D). The FU-MMT-1 cell line showed high expressions of all antigens: CD133 (65.9%), CD29 (>96.3%), CD44 (22.6%), and CD90 (39.4%). There was 100%, 43.6%, and 28.8% overlap between the CD133+/CD29+, CD133+/CD44+ (Supporting Information Fig. S1E), and CD133+/CD90+ (Supporting Information Fig. S1F) cell populations. To confirm that the FU-MMT-1 cells are not derived from endothelial progenitors, we analyzed the expression of endothelial markers CD31, CD34, CXCR-4, and VEGFR. FU-MMT-1 and its CD133+ population did not express any of these markers (Supporting Information Fig. S1G–S1N).

Identification and Prognostic Value of CD133+ Cells in Uterine Carcinosarcoma

The primary tumor from which the FU-MMT-1 cells were derived contained both malignant epithelial and mesenchymal elements (Supporting Information Fig. S2A). Both components of the primary FU-MMT-1 tumor (Supporting Information Fig. S2B) and the metastatic tumor (Supporting Information Fig. S3C) were highly immunoreactive to CD133, supporting that the FU-MMT-1 cell line maintains the characteristics of the primary tumor.

In an attempt to further examine the immunoreactivity and possible prognostic value of CD133 for uterine carcinosarcomas, we retrospectively analyzed its expression by immunohistochemistry (IHC) in a total of 25 primary uterine carcinosarcomas, including the primary tumor from which the FU-MMT-1 cells were derived (Supporting Information Table S1; Fig. 1A–1C). Patients with tumors strongly expressing CD133 in both malignant epithelial and mesenchymal components showed significantly shorter survival in comparison with other patients (those with strong epithelial but weak mesenchymal staining: p = .011, and those with weak or no expression in either component: p = .008). Patients expressing a weak CD133 expression or none at all showed a significantly longer survival in comparison with those demonstrating a strong dual expression of CD133 (p = .008), and they also tended to show a longer survival in comparison with the strong epithelial and weak mesenchymal CD133 expression groups (p = .054) (Fig. 1D)

thumbnail image

Figure 1. CD133 expression of primary human uterine carcinosarcomas. Representative picture of three different cases: (A) strong expression of CD133 in both epithelial and mesenchymal components (group 1), (B) strong expression in epithelial component and weak in the mesenchymal component (group 2), and (C) weak or no expression of CD133 in either component, (group 3). (D): The cumulative proportion survival (Kaplan-Meier) plot for 25 patients with uterine carcinosarcoma according to their CD133 expression. The median survival time was 10, 38, and 122 months for groups 1, 2, and 3, respectively. The observed survival times are indicated by circles (complete) or crosses (censored observations).

Download figure to PowerPoint

Functional Characterization of CD133+ Cells

We next analyzed the ability of the cells to grow in the absence of attachment, which is one of the important characteristics of stem cells, using microwell chips. For more than 2–3 weeks, 11 of 100 wells of the CD133+ population formed actively growing spheroids (Fig. 2A), where CD133 cells hardly showed any proliferation and displayed spindle features showing tight attachment (Fig. 2B). In the colony formation assay, the CD133+ populations exhibited significantly more colonies in comparison with CD133 cells, even when plated as low as four to six cells per well (Fig. 2C).

thumbnail image

Figure 2. In vitro characterization of CD133+ cells. CD133+ cells (A) grew in the absence of attachment, while CD133 cells (B) were spindle-shaped and could not proliferate. The CD133+ cells showed significantly higher colony formation ability than the CD133 cells (C). CD133+ cells grew as floating spheres in serum-free medium (D). When these spheres were seeded in serum containing medium, they showed a spindle-like structure replicating the primary cell culture (E). (±SEM; *, p < .05)

Download figure to PowerPoint

The downregulation of CD133 expression was observed in adherent cultures (n = 5) of the CD133+ population, grown in serum-supplemented medium, compared with the 100% sorted population. However, 34.6%–38.0% of the cells continued to express CD133 in long-term culture (Supporting Information Fig. S3, right). Moreover, only the CD133+ population grew haphazardly in multiple layers and occasionally formed interlacing bundles like the original cell line. In contrast, the CD133 population continued to be CD133 negative during the entire observation period (Supporting Information Fig. S3, left).

Both normal and CSCs from neural and epithelial organs can be expanded as sphere-like aggregates in serum-free epidermal growth factor/basic fibroblast growth factor–supplemented medium [22, 23] that favors the proliferation of undifferentiated cells. Under serum-free conditions, the CD133+ population proliferated as floating spheres (Fig. 2D), whereas the CD133 population could not survive. When floating spheres of CD133+ cells were seeded into serum containing medium, these cells showed spindle-like feature replicating the original cell culture (Fig. 2E).

Immunohistochemical Profile and Differentiation Ability of CD133+ Cells

Both CD133+ and CD133 cells were immunostained with both epithelial and mesenchymal markers, because uterine carcinosarcoma shows biphasic features. Both CD133+ and CD133 cells expressed epithelial markers (Fig. 3A). However, with regard to mesenchymal markers, the CD133+ cells expressed vimentin, which is an anaplastic cell marker of mesodermal-derived tissues, whereas CD133 cells expressed α-SMA alone, which is mostly expressed in differentiated striated cells (Fig. 3A). However, both markers can be expressed in either less or more highly differentiated stromal/mesenchymal cells, depending on the tissue. However, after a 14-day culture, a flow cytometric analysis of the CD133+ population showed that 35.7% of CD133+ cells continued to express vimentin (Supporting Information Fig. S4A), while 30.1%–39.6% of the cells became CD133α-SMA+ cells (Supporting Information Fig. S4B).

thumbnail image

Figure 3. Differentiation capacity of CD133+ cells. (A): Both CD133+ and CD133 cells expressed epithelial markers. The CD133+ cells differentiated into CD133 cells expressing α-SMA. The CD133+ cells showed chondrogenic (B), osteogenic (C), and adipogenic (D) differentiation. Abbreviations: DAPI, 4′,6-diamidino-2-phenylindole; α-SMA, α smooth muscle actin.

Download figure to PowerPoint

In contrast, after a 14-day culture, the CD133 cells continued to express α-SMA (Supporting Information Fig. S4B) and were negative for vimentin and CD133 staining (Supporting Information Fig. S4A). We also examined the ability of the CD133+ population to differentiate into multiple lineages by IHC staining. The CD133+ population differentiated into chondrocytes (Fig. 3B), osteocytes (Fig. 3C), and adipocytes (Fig. 3D) when cultured under appropriate culture conditions.

In Vitro Resistance of CD133+ Cells to Chemotherapeutic Agents

The CD133+ and CD133 populations were exposed to increasing concentrations of the chemotherapeutic agents, cisplatin and paclitaxel, which are currently in use in the clinical setting for uterine carcinosarcoma. As shown in Figure 4, the CD133+ population was significantly resistant to both cisplatin (A) and paclitaxel (B) in comparison with the CD133 population. To better characterize how the stem cell properties of the CD133+ cell population confer resistance to therapy, we examined the difference between CD133+ spheroids in serum-free media and CD133+ adherent cells grown in serum containing media. The spheroids were more resistant to both chemotherapeutics in comparison with adherent cells (Fig. 4C, 4D).

thumbnail image

Figure 4. Drug resistance of CD133+ cells. The CD133+ cells were highly resistant to cisplatin (A) and paclitaxel (B) in comparison with the CD133 cells. The spheroids were significantly resistant to cisplatin (C) and paclitaxel (D) in comparison with the adherent CD133+ cells. (±SEM; *, p < .05; **, p < .01). Abbreviation: NS, not significant.

Download figure to PowerPoint

Microarray Analysis of CD133+ Cells

We also analyzed the genomic traits of uterine carcinosarcoma CD133+ and CD133 cells using a gene expression microarray. We found that 4,853 genes were upregulated more than equal to twofold and 4,818 genes were downregulated more than equal to twofold in the CD133+ cells compared with the CD133 cells (p < .05). The expression levels of stemness-related genes are presented in Supporting Information Table S2. The CD133+ cells expressed significantly higher levels of stemness genes, including Bmi1, Bmp4, Nanog, Sox4, and Vim, in comparison with CD133 cells. Additionally, tumor-associated genes, including Myc, Aurka, Fzd2, Fzd7, Melk, and Tmsl8, were significantly expressed at a higher level in CD133+ compared with CD133 cells. Among the observed tumor suppressor genes, Bax, p18-INK6, p14ARF, Ext1, Tusc5, Bcl11b, and Lypd1 were remarkably downregulated in CD133+ cells in comparison with CD133 cells. In addition, the tumor suppressor P53 expression demonstrated no significant difference between the two populations; however, its expression was low in both populations.

To shed light on the CD133+-specific signaling pathways, all genes identified in the microarray that were differentially expressed in CD133+ versus CD133 cells were subjected to a pathway analysis using a canonical pathway analysis. The predominant processes upregulated in CD133+ cells include those pertaining to the RNA transcription reactome, mRNA processing reactome, translation factors, metabolism, and cell cycle regulation (Supporting Information Table S3a). In contrast, the genes upregulated in CD133 cells included those related to focal adhesion, integrin-mediated cell adhesion, striated muscle contraction, biogenic amine synthesis, and striated muscle-related genes, suggesting that these cells has more differentiated profile (Supporting Information Table S3b).

Stemness and Müllerian Gene Expression Profile

To verify the microarray results, the expression level of the stemness genes were examined by real-time PCR and Western blotting (Fig. 5A, 5C). The expression level of the genes involved in the maintenance of stemness, markedly Bmi1 and Nanog, was found to be consistently higher in the CD133+ population compared with their CD133 counterparts, thereby supporting the microarray data. Moreover, the expression levels of the Oct4, Klf4, Myc, and Sox2 genes significantly increased in the CD133+ population in comparison with the CD133 population, by real-time PCR. C-Myc was also upregulated in CD133+ cells in comparison with CD133 cells, and Oct4 and Sox2 expression was almost the same in the two populations according to the Western blotting findings. To better define the molecular features of cancer-initiating cells with respect to Müllerian lineage, the expression of the MD formation-related genes Pax2, Wnt4, estrogen receptor 2 (ESR2), and PGR was analyzed (Fig. 5B). Both Pax2 and Wnt4 were expressed at a significantly higher level in the CD133+ population in comparison with the CD133 population. PGR expression was higher in the CD133 population. Interestingly, ESR2, which reported to be increased in carcinosarcoma stroma compared with normal endometrial stroma [24], was significantly increased in the CD133+ population in comparison with the CD133 cells.

thumbnail image

Figure 5. The gene expression profile of CD133+ cells. The real-time polymerase chain reaction (PCR) analysis of stemness-related gene expression in the CD133+ and CD133 populations (A). The real-time PCR analysis of Müllerian duct formation-related genes (B). A Western blotting analysis of the stemness-related genes; GAPDH was used as a loading control (C). (±SEM; *, p < .05; **, p < .01). Abbreviation: RQ, relative quantity; MD, Müllerian duct.

Download figure to PowerPoint

In Vivo Tumorigenic Potential of Isolated CD133+ Cells

Because the definition of cancer-initiating cells relies mainly on their functional properties, such as their ability to sustain tumor growth recapitulating the original cellular heterogeneity, we investigated the tumorigenic capacity of FU-MMT-1-derived CD133+ and CD133 cell populations. Palpable tumors were visible after a variable time interval in the majority of CD133+-injected mice (Supporting Information Table S4). CD133 cells also developed into tumors when they were injected into mice at a concentration higher than 104 (Supporting Information Table S4). However, the tumor consistency was different, the CD133+ tumors showed high population of cells (Fig. 6A), whereas the CD133 tumors showed more myogenic differentiation, with degenerated features representing massive hyalinization (Fig. 6B). To investigate whether the initial tumors originating from CD133+ cells possessed a higher long-term tumorigenic potential compared with such tumors originating from the CD133 fraction, we performed serial transplantation assays in nude mice. The CD133+ tumors were able to generate tumors maintaining the primary morphology and proportion of CD133+ cells in serial transplantation, whereas the CD133 tumors lost tumorigenic potential after the initial passage (Supporting Information Table S4).

thumbnail image

Figure 6. In vivo tumorigenicity of CD133+ cells. The CD133+-derived tumor (A) showed a high density of tumor cells in comparison with their CD133 counterparts, which mostly consisted of nonproliferative hyalinized differentiation (B). An orthotopic tumor derived from CD133+ cells (C). High intratumoral vascularity in the orthotopic tumor detected by contrast-enhanced ultrasound (D). H&E staining in an orthotopic tumor growing in the murine uterus (E) most tumor cells were positive for CD133 (F).

Download figure to PowerPoint

Moreover, we orthotopically transplanted CD133+ and CD133 cells into a murine MD, and found that the CD133+ cells aggressively formed tumors (Fig. 6C) with ascites in all four mice. The orthotopic CD133+ tumors displayed highly angiogenic features by color Doppler US (Fig. 6D). The histological features in the MD tumors were similar to the subcutaneous CD133+ tumors (Fig. 6E), and the tumor was highly positive for CD133 (Fig. 6F).

The IHC analyses revealed that tumors derived from CD133+ cells faithfully reproduced the primary tumor, thus suggesting that the CD133+ cell population is enriched in cells capable of initiating uterine carcinosarcoma in nude mice (Fig. 7). EpCAM, vimentin, α-SMA, and ESR2 were highly positive in both the CD133+ tumors and primary FU-MMT-1 tumors. The PGR expression was also seen in the mesenchymal cells in both the primary tumor and CD133 tumors.

thumbnail image

Figure 7. The immunohistochemical profile of CD133+ and CD133 xenografts and the primary FU-MMT-1 tumor. CD133+ tumors showed poorly differentiated features, in comparison with the primary FU-MMT-1 tumor and the immunohistochemical findings resembled those of the primary tumors, in comparison with the CD133 tumors. Abbreviations: ESR2, estrogen receptor 2; PGR, progesterone receptor; α-SMA, α smooth muscle actin.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Conclusion
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Uterine carcinosarcomas grow aggressively and often show heterologous mesenchymal differentiation, which suggested the probability that they were “a stem cell tumor” of the MD. Many of these tumors occur in postmenopausal females, and the average age of patients affected with uterine carcinosarcomas is significantly older than those with endometrial cancers. There have so far been a few research studies, including our own, in which the investigators were able to establish and fully characterize cell strains of this uncommon biphasic tumor [19–21, 25]. The FU-MMT-1 line was near diploid and any observed abnormalities were simple and stable (trisomy 8) [19]. Although a chromosomal analysis has not yet been previously used to examine previous CSCs from solid tumors, the results in human or mouse embryonic stem cells (ESCs) include aneuploidy with a gain of chromosomes (trisomy) 8, 11, 12, 17, and X in long-term culture [26]. Therefore, further chromosomal analyses of CSC from solid tumors are needed.

In our retrospective analysis of uterine carcinosarcoma patients, the strong expression of CD133 in both elements (epithelial and mesenchymal) was associated with a significantly poorer prognosis, thus suggesting that an increased number of cancer stem and progenitor cells may negatively affect the disease progression. On the other hand, the absence or weak expression of CD133 was correlated with a better outcome to treatment and prolonged survival. These data support the previous studies in non-small-cell lung cancers and triple-negative breast cancer, where high CD133 expression was related to poorer OS and disease-free survival (DFS) [27, 28]. Moreover, the presence of circulating tumor cells that were positive for CEA+CK+CD133+ was a significant prognostic factor for OS and DFS in colorectal cancer [29] and CD133+Ki67+ cells were also a prognostic factor of DFS and poor clinical outcome in glioblastoma multiforme [30]. Therefore, a further analysis including a larger number of patients with uterine carcinosarcomas is needed.

Previous studies of stromal or epithelial stem-like cells in the endometrium demonstrated that clonogenic cells derived from the human endometrium are able to differentiate into mesenchymal lineages, which are similar to the characteristic capacities of bone marrow and adipose tissue stem cells [31, 32]. Ono et al. [33] demonstrated that rare populations of stromal side population (SP) cells display phenotypic and functional characteristics of uterine myometrial stem cells. In addition, a small proportion of human uterine stromal cells expressed mesenchymal stem cell markers, such as CD29, CD44, CD73, CD90, CD105, and CD146 [34]. The CD133+ population of the FU-MMT-1 cells was highly double positive for CD29/CD133, CD44/CD133, and CD90/CD133.

The CD133+ cells had a higher rate of colony formation and sphere formation, and were more resistant to the cytotoxic effects of chemotherapeutic agents than the CD133 cells. The CD133+ cell derived spheres showed more drug resistance than the CD133+ adherent cells. The CD133+ cells were also able to differentiate into mesenchymal lineages. The FU-MMT-1-derived CD133+ cells continued to express cytokeratin, EpCAM, and vimentin, resembling the epithelial component of the primary tumor. In contrast, the CD133 cells expressed cytokeratin and α-SMA, like the mesenchymal component of primary tumor. Moreover, during the long-term culture of CD133+ population regenerated CD133α-SMA+ cells, which replicating the original FU-MMT-1 cell line.

Recent evidence suggests that there is a shared genomic fingerprint between ESCs, adult tissue stem cells, and CSCs. Therefore, we assessed the expression of genes that play a prominent role in stem cell maintenance and nuclear reprogramming, including C-Myc, Sox2, Oct-4, and Nanog in the CD133+ and CD133 populations [13]. All four genes were consistently and significantly overexpressed in the CD133+ fractions in each of the cell isolates by real-time PCR.

Especially, regarding such self-renewal related proteins as Bmi-1 and Nanog expression was high in the CD133+ population. Thus, these results are considered to reflect the consistently high CD133 expression that was observed during the long-term culture of the CD133+ population.

C-Myc is recognized as a dominant-acting oncogene, but encodes a 62-kDa transcription factor thought to regulate the cell cycle transition from G0 to G1. In uterine carcinosarcoma, a few studies have shown the amplification of this gene [19, 20, 35] and this study showed that this amplification was significantly higher in CD133+ cells, which suggests that there are oncogenic differences between CD133+ and CD133 cells. Moreover, the gene expression profile of the CD133+ population in FU-MMT-1 cells preferentially expressed a set of 70 stem cell related genes compared with the CD133 population.

Bmi-1 is an epigenetic regulator required to maintain the transcriptionally repressed state of many genes through the methylation and acetylation of chromatin and histones [36]. Therefore, it maintains the primitive state of tissue stem cells, while also maintaining stem cell self-renewal [37–39]. Thus, Bmi-1 has emerged as a Myc-cooperating oncogene in the generation of mouse pre-B-cell lymphomas [40, 41]. Bmi-1 desensitizes cells to apoptosis and facilitates both cell proliferation and tumorigenesis through the deregulation of p16INK4a and p14ARF [42, 43], which is reflected by the strong oncogenic collaboration between Bmi1 and c-Myc or Ras in MEFs [44]. In our study, CD133+ cells showed the upregulation of c-Myc and Bmi-1 with the suppression of p18-INK6 and p14ARF. As both genes play an important role in the maintenance of stem cells, there might be a correlation between self renewal and anti-apoptotic mechanism regarding the Bmi-1 and c-Myc expression of CSCs.

The other gene highly expressed in CD133+ cells was Nanog, which is a hallmark of pluripotent cells in vivo and in vitro [45, 46]. In culture, the constitutive expression of Nanog can enhance the self-renewal of ESCs and also maintain an uncommitted state [47].

Uterine carcinosarcoma is also called a malignant mixed müllerian (mesodermal) tumor because this tumor histologically mimics mesodermal development. Among the factors required for female reproductive tract (FRT) development, Lim1, Pax2, Emx2, Tcf, Dach, Wnt4, and Wnt9c seem to play an essential role in the formation and elongation of the MD [48]. The real-time PCR analysis indicated that the Pax2 and Wnt4 genes were highly expressed in theCD133+ population. Pax2, a member of the paired-box gene family encoding transcription factors, is also expressed in the epithelium of the MD, and is essential for the development of the epithelial components derived from the intermediate mesoderm [49]. Wnt4 is expressed in the mesenchymal cells surrounding the newly formed MD and is required for the initial steps of MD formation in both sexes, but also for the suppression of the male differentiation pathway in the female gonad [50]. These results suggest that the CD133+ population might be derived from a progenitor cell of the MD.

In contrast, another important mechanism for the development of the FRT is the regulation of steroid hormones, particularly by estrogen acting through two different receptors, ESR1 and ESR2 [48]. Early studies demonstrated that ESR1 and PGR were present at very low levels in myometrial SP cells compared with myometrial non-SP cells, thus suggesting their immaturity and undifferentiated nature [33]. In this study, ESR1 and PGR were present at very low levels in the CD133+ population compared with CD133 cells. Moreover, α-SMA was expressed at a significantly higher level in the CD133 cells compared with the CD133+ cells. These findings confirm that the CD133 population had a more differentiated profile than the CD133+ population. However, another estrogen receptor, ESR2, which may play a crucial role in carcinosarcoma progression [24], was significantly upregulated in the CD133+ population.

Conclusion

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Conclusion
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

The isolation of a CD133+ tumor cell subpopulation in a uterine carcinosarcoma cell line that displays Müllerian progenitor and CSC properties provides new insight into the biology of this tumor, and may constitute the first identification of CSCs in a human uterine carcinosarcoma (MMMT). Therefore, the CD133+ tumor cell subpopulation may provide a valuable target for the design of therapeutic strategies aimed toward improving the prognosis of this highly aggressive biphasic uterine tumor.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Conclusion
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

We thank Yumiko Hirose for her expertise on cell sorting, and Hiromi Yamada for her valuable technical assistance.This work was supported by Central Research Institute of Fukuoka University, Fukuoka, Japan (Fund No:106001).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Conclusion
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Conclusion
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
STEM_711_sm_SuppFig1A-F.tif599KSupporting Information Figures 1 A to F
STEM_711_sm_SuppFig1G-N.tif337KSupporting Information Figures 1 G to N
STEM_711_sm_SuppFig2.tif2376KSupporting Information Figure 2
STEM_711_sm_SuppFig3.tif4583KSupporting Information Figure 3
STEM_711_sm_SuppFig4.tif246KSupporting Information Figure 4
STEM_711_sm_SuppTable1.doc46KTable S.1. CD133 Expression as Prognostic Marker in Uterine Carcinosarcoma Patients; Patient Characteristics
STEM_711_sm_SuppTable2.doc126KTable S.2. Stemness- and ES Cell-Related Genes Showing More Than 2-fold Changes in Expression in CD133+ Cells Compared to CD133- cells Table S.2.
STEM_711_sm_SuppTable3.doc81KTable 3a. Gene Signaling Pathways Upregulated in CD133− Cells p<0.05
STEM_711_sm_SuppTable4.doc33KTable S.4. In Vivo Tumorigenicity of CD133+ and CD133− Cells
STEM_711_sm_SuppInfo.doc53KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.