- Email: [email protected]
Isolation of cancer stem-like cells from a side population of a human glioblastoma cell line, SK-MG-1
Cancer Letters 291 (2010) 150–157
Contents lists available at ScienceDirect
Cancer Letters journal homepage: www.elsevier.com/locate/canlet
Isolation of cancer stem-like cells from a side population of a human glioblastoma cell line, SK-MG-1 Raita Fukaya a, Shigeki Ohta b, Masayuki Yamaguchi c, Hirofumi Fujii c, Yutaka Kawakami b, Takeshi Kawase a, Masahiro Toda a,* a
Department of Neurosurgery, Keio University School of Medicine, Tokyo, Japan Division of Cellular Signaling, Institute for Advanced Medical Research, Keio University School of Medicine, Tokyo Japan c Functional Imaging Division, Research Center for Innovative Oncology, National Cancer Center Hospital East, Kashiwa, Japan b
a r t i c l e
i n f o
Article history: Received 1 December 2008 Received in revised form 8 September 2009 Accepted 13 October 2009
Keywords: Brain cancer stem cell Side population Glioblastoma
a b s t r a c t Accumulating evidence suggests that in several types of brain tumors, including glioma, only a phenotypic subset of tumor cells called brain cancer stem cells (BCSCs) may be capable of initiating tumor growth. Recently, the isolation of side population (SP) cells using Hoechst dye has become a useful method for obtaining cancer stem cells in various tumors. In this study, we isolated cancer stem-like cells from human glioma cell lines using the SP technique. Flow cytometry analysis revealed that SK-MG-1, a human glioblastoma cell line, contained the largest number of SP cells among the five glioma cell lines that were analyzed. The SP cells had a self-renewal ability and were capable of forming spheres in a neurosphere culture medium containing EGF and FGF2. Spheres derived from the SP cells differentiated into three different lineage cells: neurons, astrocytes and oligodendrocytes. RT-PCR analysis revealed that the SP cells expressed a neural stem cell marker, Nestin. The SP cells generated tumors in the brains of NOD/SCID mice at 8 weeks after implantation, whereas the non-SP cells did not generate any tumors in the brain. These results indicate that SP cells isolated from SK-MG-1 possess the properties of cancer stem cells, including their self-renewal ability, multi-lineage differentiation, and tumorigenicity. Therefore, the SP cells from SK-MG-1 may be useful for analyzing BCSCs because of the ease with which they can be handled and their yield. Ó 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Gliomas are the most frequent primary tumors arising in the brain. The most malignant form of glioma is glioblastoma (WHO grade IV), with a median patient survival period of less than 1 year. Despite recent advances in surgery, radiology, and chemotherapy, the prognosis of patients Abbreviations: BCSC, brain cancer stem cell; SP, side population; EGF, epidermal growth factor; FGF2, fibroblast growth factor 2; WHO, World Health Organization; FBS, fetal bovine serum. * Corresponding author. Address: Department of Neurosurgery, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo 1608582, Japan. Tel.: +81 3 5363 3587; fax: +81 3 5363 3862. E-mail address: [email protected] (M. Toda). 0304-3835/$ - see front matter Ó 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2009.10.010
with glioblastoma has not improved substantially over the past few years [1–5]. Recently, several human cancers—including brain tumors—have been found to contain cancer stem cells (CSCs) [6–12]. Brain cancer stem cells (BCSCs) are characterized by their ability to form neurospheres, undergo self-renewal, differentiate into three cell lineages (neurons, astrocytes, and oligodendrocytes), and generate brain tumors [13]. Similar to most CSCs identified using neural stem cell markers, CD133 and Nestin are expressed in BCSC [7,8]. Since BCSCs not only have a potent tumor-initiating ability but are also resistant to chemotherapy and irradiation [14,15], new therapeutic strategies targeting BCSCs should be considered for patients with gliomas.
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157
So far, BCSCs have been isolated from brain tumors or glioma cell lines using several methods. First, BCSCs were isolated from glioma tissues as spheres cultured in serum-free medium containing EGF and FGF2, which is the same isolation method used to isolate neural stem cells from brain tissue [7,9,10,16–19]. Next, BCSCs were isolated from glioma tissues by sorting with anti-CD133 antibody (Ab), a neural stem cell marker [8,20–22]. Third, BCSCs were isolated using the side population (SP) technique on flow cytometry [23–25]. The flow cytometric isolation of SP cells is based on the cells’ ability to pump out drugs as well as the fluorescent dye Hoechst 33342 through the verapamil-sensitive ATP-binding cassette (ABC) transporter, ABCG2 [26]. In this study, we analyzed the SP cells in several human glioma cell lines in an attempt to isolate BCSCs. We found that SK-MG-1, a human glioblastoma cell line, had a relatively large SP cell fraction, compared with previously described fractions in other cell lines. Moreover, the SP cells isolated from SK-MG-1 were capable of self-renewal, multi-lineage differentiation, and tumorigenicity. 2. Materials and methods 2.1. Cell lines and cell culture Human glioma cell lines (SK-MG-1, U87MG, U373MG, KNS42 and U251) were obtained from the American Type Culture Collection (Rockville, MD, USA) and the Japanese Collection of Research Bioresourcesor (Osaka, Japan). All cell lines were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM) (Wako, Osaka, Japan) supplemented with 10% fetal bovine serum (FBS), 100 units/mL penicillin G, and 100 lg/mL streptomycin. To generate glioma spheres, cells were cultured in a neurosphere culture medium (NSP medium) consisting of Neurobasal medium (Invitrogen Corp., Carlsbad, CA) supplemented with human recombinant (h) EGF (20 ng/mL; PeproTech Inc., Rocky Hill, NJ), hrFGF2 (20 ng/mL; PeproTech), B27 (50; Invitrogen Corp.), heparin (10 ng/mL; Sigma–Aldrich Corp., St. Louis, MO) and hrLIF (10 ng/mL; Chemicon International Inc., Temecula, CA).
151
2.3. Cell proliferation and sphere formation assay SP and non-SP cells were sorted from bulk SK-MG-1 cells maintained in 10% FBS-DMEM. Then sorted SP and non-SP cells were seeded into 96-well plates at an initial cell density of 20 cells/lL and cultured in NSP medium. After 7 days of culture, the number of cells was measured using a CellTiter-Glo Luminescent Cell Viability Assay (Promega, Madison, WI); the viable cells were reflected in relative luminescence units. For the sphere formation assay, SP and non-SP cells sorted from SK-MG-1 were seeded at a low density of 20 cells/lL and the number of generated spheres (>100 lm) was counted after 7 days of culture.
2.4. Immunocytochemical analysis To examine the expression of neuronal, glial and oligodendrocyte markers in SP cells from SK-MG-1, the cells were cultured in NSP medium for 7 days. Then, the generated spheres were plated onto glass coverslips coated in poly-D-lysin and laminin (BD Biosciences) and cultured in a differentiation medium consisting of DMEM/F12 (Invitrogen Corp.), 1% FBS, 2 mM L-glutamine (Invitrogen Corp.), B27 (50; Invitrogen Corp.), 100 units/mL penicillin G, and 100 lg/mL streptomycin. After 3–4 h of culture, some of the spheres that had attached to the coverslips were fixed with 4% paraformaldehyde for 20 min at room temperature. After blocking with 10% normal goat serum (Vector Laboratories, Inc., Burlingame, CA), the cells were stained with anti-Nestin Ab (mouse monoclonal IgG1; 1:20; R&D Systems, Minneapolis, MN). The other spheres were cultured for 5 days, and the cells were stained with the following Abs: anti-GFAP Ab (rabbit polyclonal 1:400; Biomedical Technologies Inc., Stoughton, MA), anti-bIII tubulin Ab (mouse monoclonal IgG2b; 1:400; Sigma–Aldrich Corp.), and anti-O4 Ab (mouse monoclonal IgM; 1:20; Chemicon International Inc.). The primary antibodies were detected using Alexa 488-conjugated anti-rabbit IgG, Alexa 546-conjugated anti-mouse IgG, or Alexa 546-conjugated anti-mouse IgM (1:700; Molecular Probes, Eugene, Oregon). The cells were counterstained with To-Pro-3 iodide (1:300; Molecular Probes) to detect the nuclei.
2.2. Side population analysis 2.5. Analysis of tumorigenicity in vivo Glioma cells were incubated in DMEM supplemented with 10% FBS containing Hoechst 33342 (5 lg/mL; Sigma–Aldrich Corp., St Louis, MO) with or without verapamil (50 lM; Sigma–Aldrich Corp.), which is an inhibitor of some (verapamil-sensitive) ABC transporters, for 2 h at 37 °C. After staining, the cells were analyzed using a flow cytometer (EPIC Altra; Beckman Coulter). Live cells were gated using propidium iodide (PI) and analyzed using a dual-wavelength flow cytometer using a combination of a 450-nm (Hoechst blue) and a 675-nm (Hoechst red) filter after excitation with a 350-nm UV light. The SP cells were identified as a small, but distinct, cell population composed of unstained cells in the left lower quadrant of the flow cytometry profile. The major cell population, which was stained with Hoechst 33342, was identified as non-SP cells.
Nonobese diabetic (NOD)/severe combined immunodeficient (SCID) mice purchased from CLEA (Osaka, Japan) were maintained under standard conditions according to institutional guidelines. All procedures involving animal experiments were approved by the Animal Care and Use Committee of the Keio University School of Medicine. SP cells or non-SP cells (1 105 cells) suspended in 2 lL of phosphate buffered saline were inoculated into the right frontal lobe of NOD/SCID mouse brains. The brain tissues were removed surgically at 8 weeks after injection and were embedded in Tissue-Tek OCT compound (Sakura Finetechnical Corp., Tokyo, Japan) and then frozen. Cryostat sections (20 lm) were prepared for hematoxylin and eosin (H&E) staining.
152
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157
2.6. RT-PCR analysis Total RNAs of SP and non-SP cells from SK-MG-1 were isolated using TRIZOL reagent (Invitrogen Corp.). Total RNA (10 lg) was reverse transcribed using reverse transcriptase XL (AMV) (Takara, Tokyo, Japan), and a PCR reaction was performed using 0.2 mM dNTP (deoxyribonucleotide triphosphate mixture; Takara, Tokyo, Japan), 100 nM of primers, and 1U Ex Taq polymerase (Takara). The PCR conditions were as follows: 32 cycles of 94.0 °C for 1 min, 60.0 °C for 1 min, and 72.0 °C for 1 min. The PCR products were electrophoresed on 2% agarose gels and stained with ethidium bromide. The relative expression level for each gene was normalized to the endogenous control GAPDH. The primers used in this experiment were as follows: ABCG2 sense (5’-GGGTTCTCTTCTTCCTGACGACC-30 ) and antisense (50 -TGGTTGTGAGATTGACCAACAGACC -30 ); Nestin sense (50 -TGGCTCAGAGGAAGAGTCTGATable 1 Percentages of SP cell fractions in various human glioma cell lines. Cell line
SP cell fraction (%)
SK-MG-1 U87MG U373MG KNS42 U251
2.8 0.7 0.4 0.5 0.1
30 ) and antisense (50 -TCCCCCATTTACATGCTGTGA-30 ); Notch1 sense (50 -CAGGCAATCCGAGGACTATG-30 ) and antisense, (50 -CAGGCGTGTTGTTCTCACAG-30 ); and GAPDH sense (50 -TGAACGGGAAGCTCACTGG-30 ) and antisense, (50 -TCCACCACCCTGTTGCTGTA-30 ).
2.7. Statistical analysis Data are represented as the mean ± S.E. The means between groups were compared using the Student or Welch t-test. A value of p < 0.05 was regarded as indicating a statistically significant difference.
3. Results 3.1. Isolation of SP cells from a human glioma cell line, SK-MG-1 We analyzed the existence of SP cells in five human glioma cell lines using Hoechst 33342 dye. Since verapamil blocks the activity of the Hoechst 33342 transporter, the SP cell fraction was defined as the diminished region in the presence of verapamil [27]. Flow cytometry analysis revealed that all the glioma cell lines that were analyzed contained an SP cell fraction (Table 1). The SK-MG-1 cell line had the highest percentage (2.8%) of SP cells (Fig. 1), whereas the SP cell fractions in the other glioma cell lines were less than 1%.
50
p < 0.0001
Hoechst Blue (Intensity)
Verapamil (-)
Number of spheres per well
45
Verapamil (+)
2.8%
0.2%
40 35 30 25 20 15 10 5
Hoechst Red (Intensity)
0
Fig. 1. Identification of side population (SP) cells from SK-MG-1. Representative SP cell fraction from SK-MG-1. Cells labeled with Hoechst 33342 were analyzed using flow cytometry. The right panel shows the results when the cells were treated with 50 lM verapamil during the labeling procedure. The SP cells, which disappear in the presence of verapamil, are outlined and shown as a percentage of the total cell population.
Hoechst Blue (Intensity)
Bulk SK-MG-1
2.8%
SP
Non-SP
Fig. 3. Sphere formation capacity of SP cells from SK-MG-1. SP and nonSP cells were sorted and seeded into 96-well plates at low cell density (20 cells/lL) in neurosphere culture medium containing EGF and FGF2. The SP cells formed greater numbers of spheres than the non-SP cells 7 days after culture. Bars, ±S.E. p < 0.0001; Student t-test.
SP
4.4%
Non-SP
0.2%
Hoechst Red (Intensity) Fig. 2. SP cells from SK-MG-1 exhibit a self-renewal capacity. The SK-MG-1 cell line contains SP cells (left panel). After isolating both SP and non-SP fractions, the cells were cultured for 7 days and then reanalyzed using flow cytometry. Both SP and non-SP cell fractions are present in the cultured SP cells (middle panel), whereas no SP cells are apparent in the cultured non-SP cells (right panel).
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157 3.2. Self-renewal activity of SP cells from SK-MG-1 To compare the repopulation abilities of SP and non-SP cells from SKMG-1, sorted SP and non-SP cells were cultured in DMEM medium supplemented with 10% FBS for 1 week and then reanalyzed using flow cytometry and Hoechst 33342 dye. Sorted SP and non-SP cells were cultured in serum conditions to demonstrate the reproducibility of SP cells under long-term serum conditions. The cultured SP cells contained an SP cell fraction that was even larger than the fraction obtained from bulk SK-MG-1 cells, whereas the cultured non-SP cells contained very few SP cells (Fig. 2). These results suggest that SP cells, but not non-SP cells, from SK-MG-1 can generate SP cells in culture. 3.3. Ability of SP cells from SK-MG-1 to form spheres Sorted SP and non-SP cells from SK-MG-1 were cultured in NSP medium at low density (20 cells/mL) for 7 days, and the number of generated spheres was counted. The number of spheres generated from the SP cells was significantly greater than that generated from the non-SP cells (Fig. 3). 3.4. Ability of SP cells from SK-MG-1 to undergo multi-lineage differentiation Sorted SP cells from SK-MG-1 were cultured for 7 days in NSP medium and the generated spheres were stained for Nestin, a marker of neural stem cells. An immunocytochemical analysis showed that SP-derived spheres expressed Nestin (Fig. 4A). To explore the multi-lineage differentiation potential of SP cells from SK-MG-1, SP-derived spheres were further cultured for 5 days in differentiation medium without growth
153
factor on poly-D-lysin and laminin-coated dishes. The SP-derived spheres gave rise to cells expressing a neuronal marker, bIII tubulin; an astrocyte marker, GFAP; or an oligodendrocyte marker, O4 (Fig. 4B and C); however, the number of O4-positive cells was limited. Thus, SP cells from SK-MG-1 were capable of generating neuronal-lineage cells and glial-lineage (mainly astrocyte-lineage) cells in culture.
3.5. High proliferation activity of SP cells from SK-MG-1 To compare the proliferation abilities of SP and non-SP cells, the cells were cultured for 7 days and the cell numbers were counted. A luminescent cell viability assay revealed that the SP cells had a higher proliferation activity than the non-SP cells (Fig. 5).
3.6. Tumorigenic potential of SP cells from SK-MG-1 To compare the brain tumor-initiating property of SP cells with that of non-SP cells, sorted SP and non-SP cells from SK-MG-1 were implanted into the brains of NOD/SCID mice. H&E staining revealed that 4 out of 6 mice inoculated with SP cells developed tumors in the brain, whereas none of the mice (n = 5) inoculated with non-SP cells developed tumors in the brain (Fig. 6A). In a brain inoculated with SP cells, a tumor that had invaded the striatum and corpus callosum and that exhibited central necrosis was observed (Fig. 6B), and a high magnification view showed numerous high-cellularity tumor cells with large and pleomorphic nuclei (Fig. 6C). These histological findings were compatible with the typical histological features of a glioblastoma. In contrast, only one out of 5 brains inoculated with non-SP cells had a thin tumor outside of the brain that
Fig. 4. Immunocytochemical analysis of sphere and sphere-derived cells. (A) A sphere derived from SP cells from SK-MG-1 is positive for Nestin. Scale bar, 50 lm. (B) Spheres derived from SP cells from SK-MG-1 exhibit a multi-lineage differentiation capacity. Spheres generated from SP cells from SK-MG-1 were cultured in a differentiation medium without growth factor for 5 days. Scale bar, 100 lm. (C) The cells were immunolabeled for lineage-markers of glia (GFAP), neuron (bIII tubulin), and oligodendrocyte (O4) cells. Scale bar, 20 lm.
154
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157
Fig. 5. SP cells exhibited a greater proliferative ability than non-SP cells from SK-MG-1. Sorted SP and non-SP cells were seeded into 96-well plates and cultured at an initial cell density (20 cells/lL) in neurosphere culture medium containing EGF and FGF2. The cell number was measured using a CellTiter-Glo Luminescent Cell Viability assay 7 days after culture. Bars, ±S.E. p < 0.0001; Welch t-test.
did not resemble the characteristics of a glioblastoma (Fig. 6D). Thus, SP cells from SK-MG-1 had a greater tumorigenic potential than non-SP cells in vivo.
3.7. Gene expression analysis of SP cells from SK-MG-1 We analyzed the gene expressions of ABCG2, Nestin and Notch1 in SP and non-SP cells from SK-MG-1. The SK-MG-1 cells were cultured in 10% FBS-DMEM and were sorted using the SP technique. After sorting, the SP and non-SP cells were immediately treated with TRIZOL reagent to extract the total RNAs. ABC transporters are expressed in tumor cells [28] and are associated with drug resistance [29–31], since these transporters export a broad range of cytotoxic drugs [32]. In particular, ABCG2 has been implicated in the efflux capacity for Hoechst 33342 dye in SP cells [26,33]. Nestin, a marker of neural stem cells [34], was previously shown to be expressed in BCSCs [7,18,22]. And the Notch signaling pathways play important roles in normal stem cells [35,36] and various cancer cells [37], RT-PCR analysis revealed that ABCG2, Nestin, and Notch1 were more strongly expressed in SP cells than in non-SP cells (Fig. 7).
4. Discussion In the present study, we attempted to isolate cancer stem-like cells from human glioblastoma cell lines using the flow cytometry-based SP technique. Flow cytometry analysis revealed that SK-MG-1, a human glioblastoma cell line, contained the highest percentage (>2%) of SP cells among the five glioma cell lines that were analyzed. Previous reports have shown that SP cells in most glioblastoma cell lines account for up to 1% of the cells [38– 40]. We demonstrated that SP cells from SK-MG-1 are capable of self-renewal, generating both SP and non-SP cells, whereas few SP cells were generated from the non-SP cells.
The SP cells also showed a multi-lineage differentiation potential, producing glial and neuronal-lineage cells simultaneously under differentiation conditions. The SP cells formed spheres in NSP medium and had a significant capacity to proliferate in vitro as well as to grow into xenografted brain tumors in vivo. These results suggest that SP cells from SK-MG-1 possess BCSC properties: selfrenewal, a multi-lineage differentiation potential, and tumorigenicity. A large number of cells are constantly required to analyze the characteristics of BCSCs both in vivo and in vitro. BCSCs from glioma patients have been isolated using the sphere formation technique; however, obtaining large amounts of tissue samples from patients is sometimes difficult. Considering a report that under serum conditions, BCSCs change into differentiated cells immediately, similar to neural stem cells [7], glioma cell lines maintained under serum conditions are not suitable for analyses of BCSCs [18]. In contrast, some classical glioma cell lines reportedly contain a small number of cancer stem-like cells, even when they have been maintained under serum conditions for many years [23,41–43]. Considering the yield and ease of handling, cancer stem-like cells prepared from classical glioma cell lines have advantages over primary cultures for use as an experimental model. Tumors are thought to consist of heterogeneous cell populations that include a few tumor-initiating cells that have a high tumorigenicity, while most of the other cells have a lower tumorigenicity [44,45]. In this study, we divided a human glioma cell line into two types of cells using the SP technique and demonstrated that the SP cells exhibited the properties of cancer stem cells. Recently, some studies also support the idea that SP cells contain tumor-initiating cells in gliomas [24,25]. Thus, these cancer stem-like cells from the glioma cell line SK-MG-1may be useful for analyzing the characteristics of BCSCs for the development of new therapies. In this study, we observed higher gene expression levels of ABCG2, Nestin, and Notch1 in SP cells than in non-SP cells from SK-MG-1 using a RT-PCR analysis. ABCG2 is a molecular determinant of the SP cell phenotype and is a direct downstream target of Notch1, which regulates the expression of ABCG2 [46]. Notch1 also promotes the expression of Nestin, a neural stem cell marker, in glioma cells [47]. The Notch pathway is known to promote the survival and proliferation of neural stem cells and to inhibit their differentiation [36,48]. In addition, the Notch pathway is also associated with tumor development, potentially as an oncogene [49,50]. Although we did not analyze the functional roles of Notch signaling in the SP cells, the difference in the expressions of these genes between SP cells and non-SP cells suggests the involvement of Notch in the activities of BCSCs. In summary, we found that SK-MG-1, a human glioblastoma cell line, contained a high percentage of SP cells that possessed the properties of BCSCs, such as self-renewal, a multi-lineage differentiation potential, and tumorigenesis. These SP cells from SK-MG-1 may be useful for analyzing BCSCs, especially when used for in vivo analyses, because of the ease with which they can be handled and their yield.
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157
155
Fig. 6. Histological analysis of tumorigenicity in immunodeficient mouse brains inoculated with SP cells or non-SP cells from the SK-MG-1 cell line. (A) Four out of the six NOD/SCID mice inoculated with SP cells (1 105 cells) developed tumors in the brain, whereas none of the NOD/SCID mice (n = 5) inoculated with non-SP cells (1 105 cells) developed tumors in the brain at 8 weeks after inoculation. (B) H&E staining of a NOD/SCID mouse brain shows a large tumor in the right hemisphere at 8 weeks after the inoculation of SP cells (1 105 cells). The tumor was highly invasive and destroyed the neighboring corpus callosum and striatum, while exhibiting central necrosis (in the circle marked by the dotted line). Scale bar, 2 mm. (C) High power view of the H&E section shown in (B). Numerous high-cellularity tumor cells with large and pleomorphic nuclei are visible. Scale bar, 100 lm. (D) Only one out of five NOD/ SCID mouse implanted with non-SP cells (1 105 cells) had a thin tumor mass outside of the brain at 8 weeks after inoculation. Scale bar, 2 mm.
ABCG2
tential conflict of interest or financial dependence regarding this publication, as described in the instruction for authors. All authors have read and approved the manuscript.
Notch1 Acknowledgments
Nestin
Non-SP
SP
GAPDH
Fig. 7. Expression of stem cell-related genes in SP cells from SK-MG-1. The gene expressions of ABCG2, Notch1, and Nestin in SP and non-SP cells from SK-MG-1 were analyzed using RT-PCR. GAPDH was used as a control.
Conflict of interest We confirm that all authors fulfill all conditions required for authorship. We also confirm that there is no po-
This work was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan and a Grant-in-aid for the Global COE program to Keio University from MEXT. We thank the Collaborative Research Resources, Keio University School of Medicine (Keio-Med Open Access Facility) for technical assistance. We also thank Ms. Sayaka Teramoto (Department of Neurosurgery, Keio University School of Medicine) for technical assistance. References [1] E.C. Holland, Gliomagenesis: genetic alterations and mouse models, Nat. Rev. Genet. 2 (2001) 120–129. [2] P. Kleihues, D.N. Louis, B.W. Scheithauer, L.B. Rorke, G. Reifenberger, P.C. Burger, W.K. Cavenee, The WHO classification of tumors of the nervous system, J. Neuropathol. Exp. Neurol. 61 (2002) 215–225 (discussion 226–219). [3] E.A. Maher, F.B. Furnari, R.M. Bachoo, D.H. Rowitch, D.N. Louis, W.K. Cavenee, R.A. DePinho, Malignant glioma: genetics and biology of a grave matter, Genes Dev. 15 (2001) 1311–1333.
156
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157
[4] N. Sanai, A. Alvarez-Buylla, M.S. Berger, Neural stem cells and the origin of gliomas, New Engl. J. Med. 353 (2005) 811–822. [5] Y. Zhu, L.F. Parada, The molecular and genetic basis of neurological tumours, Nat. Rev. Cancer. 2 (2002) 616–626. [6] C. Hirschmann-Jax, A.E. Foster, G.G. Wulf, J.G. Nuchtern, T.W. Jax, U. Gobel, M.A. Goodell, M.K. Brenner, A distinct ‘‘side population” of cells with high drug efflux capacity in human tumor cells, Proc. Natl. Acad. Sci. USA 101 (2004) 14228–14233. [7] S.K. Singh, I.D. Clarke, M. Terasaki, V.E. Bonn, C. Hawkins, J. Squire, P.B. Dirks, Identification of a cancer stem cell in human brain tumors, Cancer Res. 63 (2003) 5821–5828. [8] S.K. Singh, C. Hawkins, I.D. Clarke, J.A. Squire, J. Bayani, T. Hide, R.M. Henkelman, M.D. Cusimano, P.B. Dirks, Identification of human brain tumour initiating cells, Nature 432 (2004) 396–401. [9] H.D. Hemmati, I. Nakano, J.A. Lazareff, M. Masterman-Smith, D.H. Geschwind, M. Bronner-Fraser, H.I. Kornblum, Cancerous stem cells can arise from pediatric brain tumors, Proc. Natl. Acad. Sci. USA 100 (2003) 15178–15183. [10] R. Galli, E. Binda, U. Orfanelli, B. Cipelletti, A. Gritti, S. De Vitis, R. Fiocco, C. Foroni, F. Dimeco, A. Vescovi, Isolation and characterization of tumorigenic, stem-like neural precursors from human glioblastoma, Cancer Res. 64 (2004) 7011–7021. [11] L. Ricci-Vitiani, D.G. Lombardi, E. Pilozzi, M. Biffoni, M. Todaro, C. Peschle, R. De Maria, Identification and expansion of human coloncancer-initiating cells, Nature 445 (2007) 111–115. [12] T. Schatton, G.F. Murphy, N.Y. Frank, K. Yamaura, A.M. WaagaGasser, M. Gasser, Q. Zhan, S. Jordan, L.M. Duncan, C. Weishaupt, R.C. Fuhlbrigge, T.S. Kupper, M.H. Sayegh, M.H. Frank, Identification of cells initiating human melanomas, Nature 451 (2008) 345–349. [13] T. Kondo, Brain cancer stem-like cells, Eur. J. Cancer 42 (2006) 1237– 1242. [14] S. Bao, Q. Wu, R.E. McLendon, Y. Hao, Q. Shi, A.B. Hjelmeland, M.W. Dewhirst, D.D. Bigner, J.N. Rich, Glioma stem cells promote radioresistance by preferential activation of the DNA damage response, Nature 444 (2006) 756–760. [15] M. Dean, T. Fojo, S. Bates, Tumour stem cells and drug resistance, Nat. Rev. Cancer 5 (2005) 275–284. [16] T.N. Ignatova, V.G. Kukekov, E.D. Laywell, O.N. Suslov, F.D. Vrionis, D.A. Steindler, Human cortical glial tumors contain neural stem-like cells expressing astroglial and neuronal markers in vitro, Glia 39 (2002) 193–206. [17] X. Yuan, J. Curtin, Y. Xiong, G. Liu, S. Waschsmann-Hogiu, D.L. Farkas, K.L. Black, J.S. Yu, Isolation of cancer stem cells from adult glioblastoma multiforme, Oncogene 23 (2004) 9392–9400. [18] J. Lee, S. Kotliarova, Y. Kotliarov, A. Li, Q. Su, N.M. Donin, S. Pastorino, B.W. Purow, N. Christopher, W. Zhang, J.K. Park, H.A. Fine, Tumor stem cells derived from glioblastomas cultured in bFGF and EGF more closely mirror the phenotype and genotype of primary tumors than do serum-cultured cell lines, Cancer Cell 9 (2006) 391–403. [19] J. Lee, M.J. Son, K. Woolard, N.M. Donin, A. Li, C.H. Cheng, S. Kotliarova, Y. Kotliarov, J. Walling, S. Ahn, M. Kim, M. Totonchy, T. Cusack, C. Ene, H. Ma, Q. Su, J.C. Zenklusen, W. Zhang, D. Maric, H.A. Fine, Epigenetic-mediated dysfunction of the bone morphogenetic protein pathway inhibits differentiation of glioblastoma-initiating cells, Cancer Cell 13 (2008) 69–80. [20] S.G. Piccirillo, B.A. Reynolds, N. Zanetti, G. Lamorte, E. Binda, G. Broggi, H. Brem, A. Olivi, F. Dimeco, A.L. Vescovi, Bone morphogenetic proteins inhibit the tumorigenic potential of human brain tumour-initiating cells, Nature 444 (2006) 761–765. [21] 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. [22] C. Calabrese, H. Poppleton, M. Kocak, T.L. Hogg, C. Fuller, B. Hamner, E.Y. Oh, M.W. Gaber, D. Finklestein, M. Allen, A. Frank, I.T. Bayazitov, S.S. Zakharenko, A. Gajjar, A. Davidoff, R.J. Gilbertson, A perivascular niche for brain tumor stem cells, Cancer Cell 11 (2007) 69–82. [23] T. Kondo, T. Setoguchi, T. Taga, Persistence of a small subpopulation of cancer stem-like cells in the C6 glioma cell line, Proc. Natl. Acad. Sci. USA 101 (2004) 781–786. [24] A.M. Bleau, D. Hambardzumyan, T. Ozawa, E.I. Fomchenko, J.T. Huse, C.W. Brennan, E.C. Holland, PTEN/PI3K/Akt pathway regulates the side population phenotype and ABCG2 activity in glioma tumor stem-like cells, Stem Cells 4 (2009) 226–235. [25] M.A. Harris, H. Yang, B.E. Low, J. Mukherje, A. Guha, R.T. Bronson, L.D. Shultz, M.A. Israel, K. Yun, Cancer stem cells are enriched in the side population cells in a mouse model of glioma, Cancer Res. 68 (2008) 10051–10059.
[26] C.W. Scharenberg, M.A. Harkey, B. Torok-Storb, The ABCG2 transporter is an efficient Hoechst 33342 efflux pump and is preferentially expressed by immature human hematopoietic progenitors, Blood 99 (2002) 507–512. [27] M.A. Goodell, K. Brose, G. Paradis, A.S. Conner, R.C. Mulligan, Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo, J. Exp. Med. 183 (1996) 1797–1806. [28] J.E. Diestra, G.L. Scheffer, I. Catala, M. Maliepaard, J.H. Schellens, R.J. Scheper, J.R. Germa-Lluch, M.A. Izquierdo, Frequent expression of the multi-drug resistance-associated protein BCRP/MXR/ABCP/ ABCG2 in human tumours detected by the BXP-21 monoclonal antibody in paraffin-embedded material, J. Pathol. 198 (2002) 213– 219. [29] C.J. Chen, J.E. Chin, K. Ueda, D.P. Clark, I. Pastan, M.M. Gottesman, I.B. Roninson, Internal duplication and homology with bacterial transport proteins in the mdr1 (P-glycoprotein) gene from multidrug-resistant human cells, Cell 47 (1986) 381–389. [30] S.P. Cole, G. Bhardwaj, J.H. Gerlach, J.E. Mackie, C.E. Grant, K.C. Almquist, A.J. Stewart, E.U. Kurz, A.M. Duncan, R.G. Deeley, Overexpression of a transporter gene in a multidrug-resistant human lung cancer cell line, Science 258 (1992) 1650–1654. [31] L.A. Doyle, W. Yang, L.V. Abruzzo, T. Krogmann, Y. Gao, A.K. Rishi, D.D. Ross, A multidrug resistance transporter from human MCF-7 breast cancer cells, Proc. Natl. Acad. Sci. USA 95 (1998) 15665– 15670. [32] K.D. Bunting, ABC transporters as phenotypic markers and functional regulators of stem cells, Stem Cells 20 (2002) 11–20. [33] M. Kim, H. Turnquist, J. Jackson, M. Sgagias, Y. Yan, M. Gong, M. Dean, J.G. Sharp, K. Cowan, The multidrug resistance transporter ABCG2 (breast cancer resistance protein 1) effluxes Hoechst 33342 and is overexpressed in hematopoietic stem cells, Clin. Cancer Res. 8 (2002) 22–28. [34] M.V. Lobo, M.I. Arenas, F.J. Alonso, G. Gomez, E. Bazan, C.L. Paino, E. Fernandez, B. Fraile, R. Paniagua, A. Moyano, E. Caso, Nestin, a neuroectodermal stem cell marker molecule is expressed in Leydig cells of the human testis and in some specific cell types from human testicular tumours, Cell Tissue Res. 316 (2004) 369–376. [35] K. Yoon, N. Gaiano, Notch signaling in the mammalian central nervous system: insights from mouse mutants, Nat. Neurosci. 8 (2005) 709–715. [36] N. Gaiano, G. Fishell, The role of notch in promoting glial and neural stem cell fates, Annu. Rev. Neurosci. 25 (2002) 471–490. [37] U. Koch, F. Radtke, Notch and cancer: a double-edged sword, Cell Mol. Life Sci. 64 (2007) 2746–2762. [38] L. Patrawala, T. Calhoun, R. Schneider-Broussard, J. Zhou, K. Claypool, D.G. Tang, Side population is enriched in tumorigenic, stem-like cancer cells whereas, ABCG2+ and ABCG2 cancer cells are similarly tumorigenic, Cancer Res. 65 (2005) 6207–6219. [39] E.E. Bar, A. Chaudhry, A. Lin, X. Fan, K. Schreck, W. Matsui, A.L. Vescovi, S. Piccirillo, F. Dimeco, A. Olivi, C.G. Eberhart, Cyclopaminemediated hedgehog pathway inhibition depletes stem-like cancer cells in glioblastoma, Stem Cells (2007). [40] X. Fan, W. Matsui, L. Khaki, D. Stearns, J. Chun, Y.M. Li, C.G. Eberhart, Notch pathway inhibition depletes stem-like cells and blocks engraftment in embryonal brain tumors, Cancer Res. 66 (2006) 7445–7452. [41] S.C. Yu, Y.F. Ping, L. Yi, Z.H. Zhou, J.H. Chen, X.H. Yao, L. Gao, J.M. Wang, X.W. Bian, Isolation and characterization of cancer stem cells from a human glioblastoma cell line U87, Cancer Lett. 265 (2008) 124–134. [42] L. Qiang, Y. Yang, Y.J. Ma, F.H. Chen, L.B. Zhang, W. Liu, Q. Qi, N. Lu, L. Tao, X.T. Wang, Q.D. You, Q.L. Guo, Isolation and characterization of cancer stem like cells in human glioblastoma cell lines, Cancer Lett. 279 (2009) 13–21. [43] X. Zheng, G. Shen, X. Yang, W. Liu, Most C6 cells are cancer stem cells: evidence from clonal and population analyses, Cancer Res. 67 (2007) 3691–3697. [44] P.B. Dirks, Brain tumour stem cells: the undercurrents of human brain cancer and their relationship to neural stem cells, Philos. Trans. Roy. Soc. Lond. B Biol. Sci. 363 (2008) 139–152. [45] A.L. Vescovi, R. Galli, B.A. Reynolds, Brain tumour stem cells, Nat. Rev. Cancer 6 (2006) 425–436. [46] S. Bhattacharya, A. Das, K. Mallya, I. Ahmad, Maintenance of retinal stem cells by Abcg2 is regulated by notch signaling, J. Cell Sci. 120 (2007) 2652–2662. [47] A.H. Shih, E.C. Holland, Notch signaling enhances Nestin expression in gliomas, Neoplasia 8 (2006) 1072–1082. [48] D.J. Solecki, X.L. Liu, T. Tomoda, Y. Fang, M.E. Hatten, Activated Notch2 signaling inhibits differentiation of cerebellar granule
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157 neuron precursors by maintaining proliferation, Neuron 31 (2001) 557–568. [49] F. Radtke, K. Raj, The role of notch in tumorigenesis: oncogene or tumour suppressor?, Nat Rev. Cancer 3 (2003) 756–767.
157
[50] A.P. Weng, J.C. Aster, Multiple niches for Notch in cancer: context is everything, Curr. Opin. Genet. Dev. 14 (2004) 48–54.
Contents lists available at ScienceDirect
Cancer Letters journal homepage: www.elsevier.com/locate/canlet
Isolation of cancer stem-like cells from a side population of a human glioblastoma cell line, SK-MG-1 Raita Fukaya a, Shigeki Ohta b, Masayuki Yamaguchi c, Hirofumi Fujii c, Yutaka Kawakami b, Takeshi Kawase a, Masahiro Toda a,* a
Department of Neurosurgery, Keio University School of Medicine, Tokyo, Japan Division of Cellular Signaling, Institute for Advanced Medical Research, Keio University School of Medicine, Tokyo Japan c Functional Imaging Division, Research Center for Innovative Oncology, National Cancer Center Hospital East, Kashiwa, Japan b
a r t i c l e
i n f o
Article history: Received 1 December 2008 Received in revised form 8 September 2009 Accepted 13 October 2009
Keywords: Brain cancer stem cell Side population Glioblastoma
a b s t r a c t Accumulating evidence suggests that in several types of brain tumors, including glioma, only a phenotypic subset of tumor cells called brain cancer stem cells (BCSCs) may be capable of initiating tumor growth. Recently, the isolation of side population (SP) cells using Hoechst dye has become a useful method for obtaining cancer stem cells in various tumors. In this study, we isolated cancer stem-like cells from human glioma cell lines using the SP technique. Flow cytometry analysis revealed that SK-MG-1, a human glioblastoma cell line, contained the largest number of SP cells among the five glioma cell lines that were analyzed. The SP cells had a self-renewal ability and were capable of forming spheres in a neurosphere culture medium containing EGF and FGF2. Spheres derived from the SP cells differentiated into three different lineage cells: neurons, astrocytes and oligodendrocytes. RT-PCR analysis revealed that the SP cells expressed a neural stem cell marker, Nestin. The SP cells generated tumors in the brains of NOD/SCID mice at 8 weeks after implantation, whereas the non-SP cells did not generate any tumors in the brain. These results indicate that SP cells isolated from SK-MG-1 possess the properties of cancer stem cells, including their self-renewal ability, multi-lineage differentiation, and tumorigenicity. Therefore, the SP cells from SK-MG-1 may be useful for analyzing BCSCs because of the ease with which they can be handled and their yield. Ó 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Gliomas are the most frequent primary tumors arising in the brain. The most malignant form of glioma is glioblastoma (WHO grade IV), with a median patient survival period of less than 1 year. Despite recent advances in surgery, radiology, and chemotherapy, the prognosis of patients Abbreviations: BCSC, brain cancer stem cell; SP, side population; EGF, epidermal growth factor; FGF2, fibroblast growth factor 2; WHO, World Health Organization; FBS, fetal bovine serum. * Corresponding author. Address: Department of Neurosurgery, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo 1608582, Japan. Tel.: +81 3 5363 3587; fax: +81 3 5363 3862. E-mail address: [email protected] (M. Toda). 0304-3835/$ - see front matter Ó 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2009.10.010
with glioblastoma has not improved substantially over the past few years [1–5]. Recently, several human cancers—including brain tumors—have been found to contain cancer stem cells (CSCs) [6–12]. Brain cancer stem cells (BCSCs) are characterized by their ability to form neurospheres, undergo self-renewal, differentiate into three cell lineages (neurons, astrocytes, and oligodendrocytes), and generate brain tumors [13]. Similar to most CSCs identified using neural stem cell markers, CD133 and Nestin are expressed in BCSC [7,8]. Since BCSCs not only have a potent tumor-initiating ability but are also resistant to chemotherapy and irradiation [14,15], new therapeutic strategies targeting BCSCs should be considered for patients with gliomas.
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157
So far, BCSCs have been isolated from brain tumors or glioma cell lines using several methods. First, BCSCs were isolated from glioma tissues as spheres cultured in serum-free medium containing EGF and FGF2, which is the same isolation method used to isolate neural stem cells from brain tissue [7,9,10,16–19]. Next, BCSCs were isolated from glioma tissues by sorting with anti-CD133 antibody (Ab), a neural stem cell marker [8,20–22]. Third, BCSCs were isolated using the side population (SP) technique on flow cytometry [23–25]. The flow cytometric isolation of SP cells is based on the cells’ ability to pump out drugs as well as the fluorescent dye Hoechst 33342 through the verapamil-sensitive ATP-binding cassette (ABC) transporter, ABCG2 [26]. In this study, we analyzed the SP cells in several human glioma cell lines in an attempt to isolate BCSCs. We found that SK-MG-1, a human glioblastoma cell line, had a relatively large SP cell fraction, compared with previously described fractions in other cell lines. Moreover, the SP cells isolated from SK-MG-1 were capable of self-renewal, multi-lineage differentiation, and tumorigenicity. 2. Materials and methods 2.1. Cell lines and cell culture Human glioma cell lines (SK-MG-1, U87MG, U373MG, KNS42 and U251) were obtained from the American Type Culture Collection (Rockville, MD, USA) and the Japanese Collection of Research Bioresourcesor (Osaka, Japan). All cell lines were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM) (Wako, Osaka, Japan) supplemented with 10% fetal bovine serum (FBS), 100 units/mL penicillin G, and 100 lg/mL streptomycin. To generate glioma spheres, cells were cultured in a neurosphere culture medium (NSP medium) consisting of Neurobasal medium (Invitrogen Corp., Carlsbad, CA) supplemented with human recombinant (h) EGF (20 ng/mL; PeproTech Inc., Rocky Hill, NJ), hrFGF2 (20 ng/mL; PeproTech), B27 (50; Invitrogen Corp.), heparin (10 ng/mL; Sigma–Aldrich Corp., St. Louis, MO) and hrLIF (10 ng/mL; Chemicon International Inc., Temecula, CA).
151
2.3. Cell proliferation and sphere formation assay SP and non-SP cells were sorted from bulk SK-MG-1 cells maintained in 10% FBS-DMEM. Then sorted SP and non-SP cells were seeded into 96-well plates at an initial cell density of 20 cells/lL and cultured in NSP medium. After 7 days of culture, the number of cells was measured using a CellTiter-Glo Luminescent Cell Viability Assay (Promega, Madison, WI); the viable cells were reflected in relative luminescence units. For the sphere formation assay, SP and non-SP cells sorted from SK-MG-1 were seeded at a low density of 20 cells/lL and the number of generated spheres (>100 lm) was counted after 7 days of culture.
2.4. Immunocytochemical analysis To examine the expression of neuronal, glial and oligodendrocyte markers in SP cells from SK-MG-1, the cells were cultured in NSP medium for 7 days. Then, the generated spheres were plated onto glass coverslips coated in poly-D-lysin and laminin (BD Biosciences) and cultured in a differentiation medium consisting of DMEM/F12 (Invitrogen Corp.), 1% FBS, 2 mM L-glutamine (Invitrogen Corp.), B27 (50; Invitrogen Corp.), 100 units/mL penicillin G, and 100 lg/mL streptomycin. After 3–4 h of culture, some of the spheres that had attached to the coverslips were fixed with 4% paraformaldehyde for 20 min at room temperature. After blocking with 10% normal goat serum (Vector Laboratories, Inc., Burlingame, CA), the cells were stained with anti-Nestin Ab (mouse monoclonal IgG1; 1:20; R&D Systems, Minneapolis, MN). The other spheres were cultured for 5 days, and the cells were stained with the following Abs: anti-GFAP Ab (rabbit polyclonal 1:400; Biomedical Technologies Inc., Stoughton, MA), anti-bIII tubulin Ab (mouse monoclonal IgG2b; 1:400; Sigma–Aldrich Corp.), and anti-O4 Ab (mouse monoclonal IgM; 1:20; Chemicon International Inc.). The primary antibodies were detected using Alexa 488-conjugated anti-rabbit IgG, Alexa 546-conjugated anti-mouse IgG, or Alexa 546-conjugated anti-mouse IgM (1:700; Molecular Probes, Eugene, Oregon). The cells were counterstained with To-Pro-3 iodide (1:300; Molecular Probes) to detect the nuclei.
2.2. Side population analysis 2.5. Analysis of tumorigenicity in vivo Glioma cells were incubated in DMEM supplemented with 10% FBS containing Hoechst 33342 (5 lg/mL; Sigma–Aldrich Corp., St Louis, MO) with or without verapamil (50 lM; Sigma–Aldrich Corp.), which is an inhibitor of some (verapamil-sensitive) ABC transporters, for 2 h at 37 °C. After staining, the cells were analyzed using a flow cytometer (EPIC Altra; Beckman Coulter). Live cells were gated using propidium iodide (PI) and analyzed using a dual-wavelength flow cytometer using a combination of a 450-nm (Hoechst blue) and a 675-nm (Hoechst red) filter after excitation with a 350-nm UV light. The SP cells were identified as a small, but distinct, cell population composed of unstained cells in the left lower quadrant of the flow cytometry profile. The major cell population, which was stained with Hoechst 33342, was identified as non-SP cells.
Nonobese diabetic (NOD)/severe combined immunodeficient (SCID) mice purchased from CLEA (Osaka, Japan) were maintained under standard conditions according to institutional guidelines. All procedures involving animal experiments were approved by the Animal Care and Use Committee of the Keio University School of Medicine. SP cells or non-SP cells (1 105 cells) suspended in 2 lL of phosphate buffered saline were inoculated into the right frontal lobe of NOD/SCID mouse brains. The brain tissues were removed surgically at 8 weeks after injection and were embedded in Tissue-Tek OCT compound (Sakura Finetechnical Corp., Tokyo, Japan) and then frozen. Cryostat sections (20 lm) were prepared for hematoxylin and eosin (H&E) staining.
152
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157
2.6. RT-PCR analysis Total RNAs of SP and non-SP cells from SK-MG-1 were isolated using TRIZOL reagent (Invitrogen Corp.). Total RNA (10 lg) was reverse transcribed using reverse transcriptase XL (AMV) (Takara, Tokyo, Japan), and a PCR reaction was performed using 0.2 mM dNTP (deoxyribonucleotide triphosphate mixture; Takara, Tokyo, Japan), 100 nM of primers, and 1U Ex Taq polymerase (Takara). The PCR conditions were as follows: 32 cycles of 94.0 °C for 1 min, 60.0 °C for 1 min, and 72.0 °C for 1 min. The PCR products were electrophoresed on 2% agarose gels and stained with ethidium bromide. The relative expression level for each gene was normalized to the endogenous control GAPDH. The primers used in this experiment were as follows: ABCG2 sense (5’-GGGTTCTCTTCTTCCTGACGACC-30 ) and antisense (50 -TGGTTGTGAGATTGACCAACAGACC -30 ); Nestin sense (50 -TGGCTCAGAGGAAGAGTCTGATable 1 Percentages of SP cell fractions in various human glioma cell lines. Cell line
SP cell fraction (%)
SK-MG-1 U87MG U373MG KNS42 U251
2.8 0.7 0.4 0.5 0.1
30 ) and antisense (50 -TCCCCCATTTACATGCTGTGA-30 ); Notch1 sense (50 -CAGGCAATCCGAGGACTATG-30 ) and antisense, (50 -CAGGCGTGTTGTTCTCACAG-30 ); and GAPDH sense (50 -TGAACGGGAAGCTCACTGG-30 ) and antisense, (50 -TCCACCACCCTGTTGCTGTA-30 ).
2.7. Statistical analysis Data are represented as the mean ± S.E. The means between groups were compared using the Student or Welch t-test. A value of p < 0.05 was regarded as indicating a statistically significant difference.
3. Results 3.1. Isolation of SP cells from a human glioma cell line, SK-MG-1 We analyzed the existence of SP cells in five human glioma cell lines using Hoechst 33342 dye. Since verapamil blocks the activity of the Hoechst 33342 transporter, the SP cell fraction was defined as the diminished region in the presence of verapamil [27]. Flow cytometry analysis revealed that all the glioma cell lines that were analyzed contained an SP cell fraction (Table 1). The SK-MG-1 cell line had the highest percentage (2.8%) of SP cells (Fig. 1), whereas the SP cell fractions in the other glioma cell lines were less than 1%.
50
p < 0.0001
Hoechst Blue (Intensity)
Verapamil (-)
Number of spheres per well
45
Verapamil (+)
2.8%
0.2%
40 35 30 25 20 15 10 5
Hoechst Red (Intensity)
0
Fig. 1. Identification of side population (SP) cells from SK-MG-1. Representative SP cell fraction from SK-MG-1. Cells labeled with Hoechst 33342 were analyzed using flow cytometry. The right panel shows the results when the cells were treated with 50 lM verapamil during the labeling procedure. The SP cells, which disappear in the presence of verapamil, are outlined and shown as a percentage of the total cell population.
Hoechst Blue (Intensity)
Bulk SK-MG-1
2.8%
SP
Non-SP
Fig. 3. Sphere formation capacity of SP cells from SK-MG-1. SP and nonSP cells were sorted and seeded into 96-well plates at low cell density (20 cells/lL) in neurosphere culture medium containing EGF and FGF2. The SP cells formed greater numbers of spheres than the non-SP cells 7 days after culture. Bars, ±S.E. p < 0.0001; Student t-test.
SP
4.4%
Non-SP
0.2%
Hoechst Red (Intensity) Fig. 2. SP cells from SK-MG-1 exhibit a self-renewal capacity. The SK-MG-1 cell line contains SP cells (left panel). After isolating both SP and non-SP fractions, the cells were cultured for 7 days and then reanalyzed using flow cytometry. Both SP and non-SP cell fractions are present in the cultured SP cells (middle panel), whereas no SP cells are apparent in the cultured non-SP cells (right panel).
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157 3.2. Self-renewal activity of SP cells from SK-MG-1 To compare the repopulation abilities of SP and non-SP cells from SKMG-1, sorted SP and non-SP cells were cultured in DMEM medium supplemented with 10% FBS for 1 week and then reanalyzed using flow cytometry and Hoechst 33342 dye. Sorted SP and non-SP cells were cultured in serum conditions to demonstrate the reproducibility of SP cells under long-term serum conditions. The cultured SP cells contained an SP cell fraction that was even larger than the fraction obtained from bulk SK-MG-1 cells, whereas the cultured non-SP cells contained very few SP cells (Fig. 2). These results suggest that SP cells, but not non-SP cells, from SK-MG-1 can generate SP cells in culture. 3.3. Ability of SP cells from SK-MG-1 to form spheres Sorted SP and non-SP cells from SK-MG-1 were cultured in NSP medium at low density (20 cells/mL) for 7 days, and the number of generated spheres was counted. The number of spheres generated from the SP cells was significantly greater than that generated from the non-SP cells (Fig. 3). 3.4. Ability of SP cells from SK-MG-1 to undergo multi-lineage differentiation Sorted SP cells from SK-MG-1 were cultured for 7 days in NSP medium and the generated spheres were stained for Nestin, a marker of neural stem cells. An immunocytochemical analysis showed that SP-derived spheres expressed Nestin (Fig. 4A). To explore the multi-lineage differentiation potential of SP cells from SK-MG-1, SP-derived spheres were further cultured for 5 days in differentiation medium without growth
153
factor on poly-D-lysin and laminin-coated dishes. The SP-derived spheres gave rise to cells expressing a neuronal marker, bIII tubulin; an astrocyte marker, GFAP; or an oligodendrocyte marker, O4 (Fig. 4B and C); however, the number of O4-positive cells was limited. Thus, SP cells from SK-MG-1 were capable of generating neuronal-lineage cells and glial-lineage (mainly astrocyte-lineage) cells in culture.
3.5. High proliferation activity of SP cells from SK-MG-1 To compare the proliferation abilities of SP and non-SP cells, the cells were cultured for 7 days and the cell numbers were counted. A luminescent cell viability assay revealed that the SP cells had a higher proliferation activity than the non-SP cells (Fig. 5).
3.6. Tumorigenic potential of SP cells from SK-MG-1 To compare the brain tumor-initiating property of SP cells with that of non-SP cells, sorted SP and non-SP cells from SK-MG-1 were implanted into the brains of NOD/SCID mice. H&E staining revealed that 4 out of 6 mice inoculated with SP cells developed tumors in the brain, whereas none of the mice (n = 5) inoculated with non-SP cells developed tumors in the brain (Fig. 6A). In a brain inoculated with SP cells, a tumor that had invaded the striatum and corpus callosum and that exhibited central necrosis was observed (Fig. 6B), and a high magnification view showed numerous high-cellularity tumor cells with large and pleomorphic nuclei (Fig. 6C). These histological findings were compatible with the typical histological features of a glioblastoma. In contrast, only one out of 5 brains inoculated with non-SP cells had a thin tumor outside of the brain that
Fig. 4. Immunocytochemical analysis of sphere and sphere-derived cells. (A) A sphere derived from SP cells from SK-MG-1 is positive for Nestin. Scale bar, 50 lm. (B) Spheres derived from SP cells from SK-MG-1 exhibit a multi-lineage differentiation capacity. Spheres generated from SP cells from SK-MG-1 were cultured in a differentiation medium without growth factor for 5 days. Scale bar, 100 lm. (C) The cells were immunolabeled for lineage-markers of glia (GFAP), neuron (bIII tubulin), and oligodendrocyte (O4) cells. Scale bar, 20 lm.
154
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157
Fig. 5. SP cells exhibited a greater proliferative ability than non-SP cells from SK-MG-1. Sorted SP and non-SP cells were seeded into 96-well plates and cultured at an initial cell density (20 cells/lL) in neurosphere culture medium containing EGF and FGF2. The cell number was measured using a CellTiter-Glo Luminescent Cell Viability assay 7 days after culture. Bars, ±S.E. p < 0.0001; Welch t-test.
did not resemble the characteristics of a glioblastoma (Fig. 6D). Thus, SP cells from SK-MG-1 had a greater tumorigenic potential than non-SP cells in vivo.
3.7. Gene expression analysis of SP cells from SK-MG-1 We analyzed the gene expressions of ABCG2, Nestin and Notch1 in SP and non-SP cells from SK-MG-1. The SK-MG-1 cells were cultured in 10% FBS-DMEM and were sorted using the SP technique. After sorting, the SP and non-SP cells were immediately treated with TRIZOL reagent to extract the total RNAs. ABC transporters are expressed in tumor cells [28] and are associated with drug resistance [29–31], since these transporters export a broad range of cytotoxic drugs [32]. In particular, ABCG2 has been implicated in the efflux capacity for Hoechst 33342 dye in SP cells [26,33]. Nestin, a marker of neural stem cells [34], was previously shown to be expressed in BCSCs [7,18,22]. And the Notch signaling pathways play important roles in normal stem cells [35,36] and various cancer cells [37], RT-PCR analysis revealed that ABCG2, Nestin, and Notch1 were more strongly expressed in SP cells than in non-SP cells (Fig. 7).
4. Discussion In the present study, we attempted to isolate cancer stem-like cells from human glioblastoma cell lines using the flow cytometry-based SP technique. Flow cytometry analysis revealed that SK-MG-1, a human glioblastoma cell line, contained the highest percentage (>2%) of SP cells among the five glioma cell lines that were analyzed. Previous reports have shown that SP cells in most glioblastoma cell lines account for up to 1% of the cells [38– 40]. We demonstrated that SP cells from SK-MG-1 are capable of self-renewal, generating both SP and non-SP cells, whereas few SP cells were generated from the non-SP cells.
The SP cells also showed a multi-lineage differentiation potential, producing glial and neuronal-lineage cells simultaneously under differentiation conditions. The SP cells formed spheres in NSP medium and had a significant capacity to proliferate in vitro as well as to grow into xenografted brain tumors in vivo. These results suggest that SP cells from SK-MG-1 possess BCSC properties: selfrenewal, a multi-lineage differentiation potential, and tumorigenicity. A large number of cells are constantly required to analyze the characteristics of BCSCs both in vivo and in vitro. BCSCs from glioma patients have been isolated using the sphere formation technique; however, obtaining large amounts of tissue samples from patients is sometimes difficult. Considering a report that under serum conditions, BCSCs change into differentiated cells immediately, similar to neural stem cells [7], glioma cell lines maintained under serum conditions are not suitable for analyses of BCSCs [18]. In contrast, some classical glioma cell lines reportedly contain a small number of cancer stem-like cells, even when they have been maintained under serum conditions for many years [23,41–43]. Considering the yield and ease of handling, cancer stem-like cells prepared from classical glioma cell lines have advantages over primary cultures for use as an experimental model. Tumors are thought to consist of heterogeneous cell populations that include a few tumor-initiating cells that have a high tumorigenicity, while most of the other cells have a lower tumorigenicity [44,45]. In this study, we divided a human glioma cell line into two types of cells using the SP technique and demonstrated that the SP cells exhibited the properties of cancer stem cells. Recently, some studies also support the idea that SP cells contain tumor-initiating cells in gliomas [24,25]. Thus, these cancer stem-like cells from the glioma cell line SK-MG-1may be useful for analyzing the characteristics of BCSCs for the development of new therapies. In this study, we observed higher gene expression levels of ABCG2, Nestin, and Notch1 in SP cells than in non-SP cells from SK-MG-1 using a RT-PCR analysis. ABCG2 is a molecular determinant of the SP cell phenotype and is a direct downstream target of Notch1, which regulates the expression of ABCG2 [46]. Notch1 also promotes the expression of Nestin, a neural stem cell marker, in glioma cells [47]. The Notch pathway is known to promote the survival and proliferation of neural stem cells and to inhibit their differentiation [36,48]. In addition, the Notch pathway is also associated with tumor development, potentially as an oncogene [49,50]. Although we did not analyze the functional roles of Notch signaling in the SP cells, the difference in the expressions of these genes between SP cells and non-SP cells suggests the involvement of Notch in the activities of BCSCs. In summary, we found that SK-MG-1, a human glioblastoma cell line, contained a high percentage of SP cells that possessed the properties of BCSCs, such as self-renewal, a multi-lineage differentiation potential, and tumorigenesis. These SP cells from SK-MG-1 may be useful for analyzing BCSCs, especially when used for in vivo analyses, because of the ease with which they can be handled and their yield.
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157
155
Fig. 6. Histological analysis of tumorigenicity in immunodeficient mouse brains inoculated with SP cells or non-SP cells from the SK-MG-1 cell line. (A) Four out of the six NOD/SCID mice inoculated with SP cells (1 105 cells) developed tumors in the brain, whereas none of the NOD/SCID mice (n = 5) inoculated with non-SP cells (1 105 cells) developed tumors in the brain at 8 weeks after inoculation. (B) H&E staining of a NOD/SCID mouse brain shows a large tumor in the right hemisphere at 8 weeks after the inoculation of SP cells (1 105 cells). The tumor was highly invasive and destroyed the neighboring corpus callosum and striatum, while exhibiting central necrosis (in the circle marked by the dotted line). Scale bar, 2 mm. (C) High power view of the H&E section shown in (B). Numerous high-cellularity tumor cells with large and pleomorphic nuclei are visible. Scale bar, 100 lm. (D) Only one out of five NOD/ SCID mouse implanted with non-SP cells (1 105 cells) had a thin tumor mass outside of the brain at 8 weeks after inoculation. Scale bar, 2 mm.
ABCG2
tential conflict of interest or financial dependence regarding this publication, as described in the instruction for authors. All authors have read and approved the manuscript.
Notch1 Acknowledgments
Nestin
Non-SP
SP
GAPDH
Fig. 7. Expression of stem cell-related genes in SP cells from SK-MG-1. The gene expressions of ABCG2, Notch1, and Nestin in SP and non-SP cells from SK-MG-1 were analyzed using RT-PCR. GAPDH was used as a control.
Conflict of interest We confirm that all authors fulfill all conditions required for authorship. We also confirm that there is no po-
This work was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan and a Grant-in-aid for the Global COE program to Keio University from MEXT. We thank the Collaborative Research Resources, Keio University School of Medicine (Keio-Med Open Access Facility) for technical assistance. We also thank Ms. Sayaka Teramoto (Department of Neurosurgery, Keio University School of Medicine) for technical assistance. References [1] E.C. Holland, Gliomagenesis: genetic alterations and mouse models, Nat. Rev. Genet. 2 (2001) 120–129. [2] P. Kleihues, D.N. Louis, B.W. Scheithauer, L.B. Rorke, G. Reifenberger, P.C. Burger, W.K. Cavenee, The WHO classification of tumors of the nervous system, J. Neuropathol. Exp. Neurol. 61 (2002) 215–225 (discussion 226–219). [3] E.A. Maher, F.B. Furnari, R.M. Bachoo, D.H. Rowitch, D.N. Louis, W.K. Cavenee, R.A. DePinho, Malignant glioma: genetics and biology of a grave matter, Genes Dev. 15 (2001) 1311–1333.
156
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157
[4] N. Sanai, A. Alvarez-Buylla, M.S. Berger, Neural stem cells and the origin of gliomas, New Engl. J. Med. 353 (2005) 811–822. [5] Y. Zhu, L.F. Parada, The molecular and genetic basis of neurological tumours, Nat. Rev. Cancer. 2 (2002) 616–626. [6] C. Hirschmann-Jax, A.E. Foster, G.G. Wulf, J.G. Nuchtern, T.W. Jax, U. Gobel, M.A. Goodell, M.K. Brenner, A distinct ‘‘side population” of cells with high drug efflux capacity in human tumor cells, Proc. Natl. Acad. Sci. USA 101 (2004) 14228–14233. [7] S.K. Singh, I.D. Clarke, M. Terasaki, V.E. Bonn, C. Hawkins, J. Squire, P.B. Dirks, Identification of a cancer stem cell in human brain tumors, Cancer Res. 63 (2003) 5821–5828. [8] S.K. Singh, C. Hawkins, I.D. Clarke, J.A. Squire, J. Bayani, T. Hide, R.M. Henkelman, M.D. Cusimano, P.B. Dirks, Identification of human brain tumour initiating cells, Nature 432 (2004) 396–401. [9] H.D. Hemmati, I. Nakano, J.A. Lazareff, M. Masterman-Smith, D.H. Geschwind, M. Bronner-Fraser, H.I. Kornblum, Cancerous stem cells can arise from pediatric brain tumors, Proc. Natl. Acad. Sci. USA 100 (2003) 15178–15183. [10] R. Galli, E. Binda, U. Orfanelli, B. Cipelletti, A. Gritti, S. De Vitis, R. Fiocco, C. Foroni, F. Dimeco, A. Vescovi, Isolation and characterization of tumorigenic, stem-like neural precursors from human glioblastoma, Cancer Res. 64 (2004) 7011–7021. [11] L. Ricci-Vitiani, D.G. Lombardi, E. Pilozzi, M. Biffoni, M. Todaro, C. Peschle, R. De Maria, Identification and expansion of human coloncancer-initiating cells, Nature 445 (2007) 111–115. [12] T. Schatton, G.F. Murphy, N.Y. Frank, K. Yamaura, A.M. WaagaGasser, M. Gasser, Q. Zhan, S. Jordan, L.M. Duncan, C. Weishaupt, R.C. Fuhlbrigge, T.S. Kupper, M.H. Sayegh, M.H. Frank, Identification of cells initiating human melanomas, Nature 451 (2008) 345–349. [13] T. Kondo, Brain cancer stem-like cells, Eur. J. Cancer 42 (2006) 1237– 1242. [14] S. Bao, Q. Wu, R.E. McLendon, Y. Hao, Q. Shi, A.B. Hjelmeland, M.W. Dewhirst, D.D. Bigner, J.N. Rich, Glioma stem cells promote radioresistance by preferential activation of the DNA damage response, Nature 444 (2006) 756–760. [15] M. Dean, T. Fojo, S. Bates, Tumour stem cells and drug resistance, Nat. Rev. Cancer 5 (2005) 275–284. [16] T.N. Ignatova, V.G. Kukekov, E.D. Laywell, O.N. Suslov, F.D. Vrionis, D.A. Steindler, Human cortical glial tumors contain neural stem-like cells expressing astroglial and neuronal markers in vitro, Glia 39 (2002) 193–206. [17] X. Yuan, J. Curtin, Y. Xiong, G. Liu, S. Waschsmann-Hogiu, D.L. Farkas, K.L. Black, J.S. Yu, Isolation of cancer stem cells from adult glioblastoma multiforme, Oncogene 23 (2004) 9392–9400. [18] J. Lee, S. Kotliarova, Y. Kotliarov, A. Li, Q. Su, N.M. Donin, S. Pastorino, B.W. Purow, N. Christopher, W. Zhang, J.K. Park, H.A. Fine, Tumor stem cells derived from glioblastomas cultured in bFGF and EGF more closely mirror the phenotype and genotype of primary tumors than do serum-cultured cell lines, Cancer Cell 9 (2006) 391–403. [19] J. Lee, M.J. Son, K. Woolard, N.M. Donin, A. Li, C.H. Cheng, S. Kotliarova, Y. Kotliarov, J. Walling, S. Ahn, M. Kim, M. Totonchy, T. Cusack, C. Ene, H. Ma, Q. Su, J.C. Zenklusen, W. Zhang, D. Maric, H.A. Fine, Epigenetic-mediated dysfunction of the bone morphogenetic protein pathway inhibits differentiation of glioblastoma-initiating cells, Cancer Cell 13 (2008) 69–80. [20] S.G. Piccirillo, B.A. Reynolds, N. Zanetti, G. Lamorte, E. Binda, G. Broggi, H. Brem, A. Olivi, F. Dimeco, A.L. Vescovi, Bone morphogenetic proteins inhibit the tumorigenic potential of human brain tumour-initiating cells, Nature 444 (2006) 761–765. [21] 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. [22] C. Calabrese, H. Poppleton, M. Kocak, T.L. Hogg, C. Fuller, B. Hamner, E.Y. Oh, M.W. Gaber, D. Finklestein, M. Allen, A. Frank, I.T. Bayazitov, S.S. Zakharenko, A. Gajjar, A. Davidoff, R.J. Gilbertson, A perivascular niche for brain tumor stem cells, Cancer Cell 11 (2007) 69–82. [23] T. Kondo, T. Setoguchi, T. Taga, Persistence of a small subpopulation of cancer stem-like cells in the C6 glioma cell line, Proc. Natl. Acad. Sci. USA 101 (2004) 781–786. [24] A.M. Bleau, D. Hambardzumyan, T. Ozawa, E.I. Fomchenko, J.T. Huse, C.W. Brennan, E.C. Holland, PTEN/PI3K/Akt pathway regulates the side population phenotype and ABCG2 activity in glioma tumor stem-like cells, Stem Cells 4 (2009) 226–235. [25] M.A. Harris, H. Yang, B.E. Low, J. Mukherje, A. Guha, R.T. Bronson, L.D. Shultz, M.A. Israel, K. Yun, Cancer stem cells are enriched in the side population cells in a mouse model of glioma, Cancer Res. 68 (2008) 10051–10059.
[26] C.W. Scharenberg, M.A. Harkey, B. Torok-Storb, The ABCG2 transporter is an efficient Hoechst 33342 efflux pump and is preferentially expressed by immature human hematopoietic progenitors, Blood 99 (2002) 507–512. [27] M.A. Goodell, K. Brose, G. Paradis, A.S. Conner, R.C. Mulligan, Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo, J. Exp. Med. 183 (1996) 1797–1806. [28] J.E. Diestra, G.L. Scheffer, I. Catala, M. Maliepaard, J.H. Schellens, R.J. Scheper, J.R. Germa-Lluch, M.A. Izquierdo, Frequent expression of the multi-drug resistance-associated protein BCRP/MXR/ABCP/ ABCG2 in human tumours detected by the BXP-21 monoclonal antibody in paraffin-embedded material, J. Pathol. 198 (2002) 213– 219. [29] C.J. Chen, J.E. Chin, K. Ueda, D.P. Clark, I. Pastan, M.M. Gottesman, I.B. Roninson, Internal duplication and homology with bacterial transport proteins in the mdr1 (P-glycoprotein) gene from multidrug-resistant human cells, Cell 47 (1986) 381–389. [30] S.P. Cole, G. Bhardwaj, J.H. Gerlach, J.E. Mackie, C.E. Grant, K.C. Almquist, A.J. Stewart, E.U. Kurz, A.M. Duncan, R.G. Deeley, Overexpression of a transporter gene in a multidrug-resistant human lung cancer cell line, Science 258 (1992) 1650–1654. [31] L.A. Doyle, W. Yang, L.V. Abruzzo, T. Krogmann, Y. Gao, A.K. Rishi, D.D. Ross, A multidrug resistance transporter from human MCF-7 breast cancer cells, Proc. Natl. Acad. Sci. USA 95 (1998) 15665– 15670. [32] K.D. Bunting, ABC transporters as phenotypic markers and functional regulators of stem cells, Stem Cells 20 (2002) 11–20. [33] M. Kim, H. Turnquist, J. Jackson, M. Sgagias, Y. Yan, M. Gong, M. Dean, J.G. Sharp, K. Cowan, The multidrug resistance transporter ABCG2 (breast cancer resistance protein 1) effluxes Hoechst 33342 and is overexpressed in hematopoietic stem cells, Clin. Cancer Res. 8 (2002) 22–28. [34] M.V. Lobo, M.I. Arenas, F.J. Alonso, G. Gomez, E. Bazan, C.L. Paino, E. Fernandez, B. Fraile, R. Paniagua, A. Moyano, E. Caso, Nestin, a neuroectodermal stem cell marker molecule is expressed in Leydig cells of the human testis and in some specific cell types from human testicular tumours, Cell Tissue Res. 316 (2004) 369–376. [35] K. Yoon, N. Gaiano, Notch signaling in the mammalian central nervous system: insights from mouse mutants, Nat. Neurosci. 8 (2005) 709–715. [36] N. Gaiano, G. Fishell, The role of notch in promoting glial and neural stem cell fates, Annu. Rev. Neurosci. 25 (2002) 471–490. [37] U. Koch, F. Radtke, Notch and cancer: a double-edged sword, Cell Mol. Life Sci. 64 (2007) 2746–2762. [38] L. Patrawala, T. Calhoun, R. Schneider-Broussard, J. Zhou, K. Claypool, D.G. Tang, Side population is enriched in tumorigenic, stem-like cancer cells whereas, ABCG2+ and ABCG2 cancer cells are similarly tumorigenic, Cancer Res. 65 (2005) 6207–6219. [39] E.E. Bar, A. Chaudhry, A. Lin, X. Fan, K. Schreck, W. Matsui, A.L. Vescovi, S. Piccirillo, F. Dimeco, A. Olivi, C.G. Eberhart, Cyclopaminemediated hedgehog pathway inhibition depletes stem-like cancer cells in glioblastoma, Stem Cells (2007). [40] X. Fan, W. Matsui, L. Khaki, D. Stearns, J. Chun, Y.M. Li, C.G. Eberhart, Notch pathway inhibition depletes stem-like cells and blocks engraftment in embryonal brain tumors, Cancer Res. 66 (2006) 7445–7452. [41] S.C. Yu, Y.F. Ping, L. Yi, Z.H. Zhou, J.H. Chen, X.H. Yao, L. Gao, J.M. Wang, X.W. Bian, Isolation and characterization of cancer stem cells from a human glioblastoma cell line U87, Cancer Lett. 265 (2008) 124–134. [42] L. Qiang, Y. Yang, Y.J. Ma, F.H. Chen, L.B. Zhang, W. Liu, Q. Qi, N. Lu, L. Tao, X.T. Wang, Q.D. You, Q.L. Guo, Isolation and characterization of cancer stem like cells in human glioblastoma cell lines, Cancer Lett. 279 (2009) 13–21. [43] X. Zheng, G. Shen, X. Yang, W. Liu, Most C6 cells are cancer stem cells: evidence from clonal and population analyses, Cancer Res. 67 (2007) 3691–3697. [44] P.B. Dirks, Brain tumour stem cells: the undercurrents of human brain cancer and their relationship to neural stem cells, Philos. Trans. Roy. Soc. Lond. B Biol. Sci. 363 (2008) 139–152. [45] A.L. Vescovi, R. Galli, B.A. Reynolds, Brain tumour stem cells, Nat. Rev. Cancer 6 (2006) 425–436. [46] S. Bhattacharya, A. Das, K. Mallya, I. Ahmad, Maintenance of retinal stem cells by Abcg2 is regulated by notch signaling, J. Cell Sci. 120 (2007) 2652–2662. [47] A.H. Shih, E.C. Holland, Notch signaling enhances Nestin expression in gliomas, Neoplasia 8 (2006) 1072–1082. [48] D.J. Solecki, X.L. Liu, T. Tomoda, Y. Fang, M.E. Hatten, Activated Notch2 signaling inhibits differentiation of cerebellar granule
R. Fukaya et al. / Cancer Letters 291 (2010) 150–157 neuron precursors by maintaining proliferation, Neuron 31 (2001) 557–568. [49] F. Radtke, K. Raj, The role of notch in tumorigenesis: oncogene or tumour suppressor?, Nat Rev. Cancer 3 (2003) 756–767.
157
[50] A.P. Weng, J.C. Aster, Multiple niches for Notch in cancer: context is everything, Curr. Opin. Genet. Dev. 14 (2004) 48–54.