The recent identification and functional studies of glioma stem cells (GSCs) have provided strong evidence to support their critical roles in the initiation and propagation of glioblastoma.1 GSCs are defined as cells within glioblastoma that possess the capacity to self-renew and to produce the heterogeneous lineages of glioma cells that comprise the tumor.1 The GSC hypothesis can help to explain certain clinical features such as tumor hierarchy, recurrence, metastasis, and therapy resistance,1-4 and may affect the way in which we diagnose and treat tumors. The therapeutic aim would change from eliminating the bulk of rapidly dividing but terminally differentiated components of the tumor, to refocusing on the minority stem cell population that fuels tumor growth.1 Such a transition highlights the need for laboratory work to provide information on the genetic and epigenetic “blueprints” of GSCs, and to identify true signatures of “stemness”. The work primarily requires the derivation and identification of bona fide GSCs populations.
GSCs cultures have been derived from surgical specimen and established cell lines (C6, U87, U251, et al) by various in vitro methods,5-10 amongst which neurosphere assay is currently widely used as growing cells in serum-free medium supplemented with mitogens and on non-adherent plates, where stem-like cells are able to continually divide and form multipotent clonal spheres called neurospheres. Although used as the standard in vitro method for identifying the presence of stem cells,2,11-13 the neurosphere assay is associated with some limitations, e.g., the heterogeneous cell composition in the neurospheres14,15 and low efficiency for producing GSC lines. Interestingly, Pollard et al16 recently reported a more effective adherent culture of GSCs by direct plating glioma cells onto a laminin-coated flask in serum-free medium, however, the study provided no quantitative assessment of the in vitro stemness phenotypes, and no comparison of in vivo tumorigenicity between cells from the adherent and neurosphere tumor cultures.17
C6 glioma cells (Shanghai Institute of Biochemistry and Cell Biology, China) were cultured under three different growth conditions, i.e., adherent growth in conventional 10% serum medium (C6-Adh), non-adherent spheres growth in serum-free medium (C6-SC-Sph), as well as adherent growth on laminin-coated flask in serum-free medium (C6-SC-Adh).
The 10% serum medium consisted of Dulbecco’s modification of Eagle’s Medium (DMEM) plus 10% fetal bovine serum (FBS, Hyclone, Thermo Fisher Scientific Inc., USA), and the serum-free medium consisted of DMEM/F12 supplemented with BIT (bovine serum albumin, insulin and transferrin; 20%, Stemcell Technologies Inc., Canada), epidermal growth factor (EGF; 20 ng/ml, PeproTech Inc., USA) and basic fibroblast growth factor (bFGF; 20 ng/ml, PeproTech Inc.).18
The laminin-coated flask was freshly prepared 2 hours before cells were planted. Adherent cells were harvested with 0.25% trypsin (Sigma-Aldrich, USA), and neurospheres were dissociated by mechanical cellular filter into single cells for passaging. All cells were incubated at 37°C with 5% CO2
, and 100% humidity. The cells were cultured for one and three months respectively and then harvested for in vitro
and in vivo
Cell proliferation assay
Standard growth curves
were plotted to compare cell proliferation rate under respective culture conditions. Cells were transferred to fresh culture flask (25 cm2
) and in 5-ml medium according to the three different culture conditions at a density of 2×105
cells/ml. For cell proliferation analysis, cells were passaged every three days. Growth rate of the different cell populations was analysed by counting the cell number obtained at each passage. Growth curves were drawn according to the cell number at each passage.
Cell cycle analysis
Cell cycle analysis was carried out using BD FACS Aria (Becton, Dickinson and Company, USA). Cells with synchronization were maintained under different culture conditions for 24 hours and then all the cells were harvested, dissociated into single cells, resuspended in 70% ethanol and stored at –20°C overnight. Before flow cytometry, cells were washed with PBS, centrifuged and resuspended in 0.5-µl buffer containing propidium iodide (50 µg/ml, Sigma) and RNAse (100 µg/ml, Sigma). Staining for 30 minutes, cell cycle was detected by flow cytometry (FCM) analysis.
Clonal formation assay
Single cell suspensions were calibrated to 5000 cells/ml in serum-free supplemented medium, then diluted into gradient cell titers at 1000, 500, 200, 100, 50, 20, and 10 cells per 200 µl respectively, and further transferred into the wells of a 96-well microplate. Each well was fed 50-µl serum-free medium every three days, and cell spheres (nonadherent, tight and spherical masses >75 µm in diameter) were counted under microscope at the end of two weeks. To confirm the gradient dilution results described above, we further underwent more stringent clonal assay by plating single cells into the 96-well plate, i.e., one viable cell per well, and at the end of two weeks wells containing clonal spheres derived from the single cells were calculated.
Side population (SP) assay
For SP analysis, cells were prepared into single cell suspensions and resuspended at 106 cells/ml in prewarmed DMEM containing 2% FBS. Cells were incubated with Hoechst 33342 (Sigma) at 5 µg/ml either alone or as control in combination with verapamil (Sigma) at 50 µmol/L for 90 minutes at 37°C with intermittent mixing. At the end of incubation, cells were centrifuged and resuspended in cold PBS containing 2% FBS for flow cytometry analysis. Before the assay for distinguishing dead cells, 1 µg/ml propidium iodide (PI, Sigma) was added. Cells were subjected to flow cytometry analysis using BD FACS Aria (Becton, Dickinson and Company).
Cells were fixed with 4% paraformaldehyde and then blocked by goat serum. Primary antibodies used were rabbit polyclonal to CD133 (1/500, Abcam, England), rabbit polyclonal to glial fibrillary acidic protein (GFAP) (1/500, Abcam) and chicken polyclonal to beta III Tubulin (1/500, Abcam). Primary antibodies were incubated for 16 hours at 4°C followed by detection with corresponding fluorescent secondary antibodies. Nuclei were counter-stained with diamidino-phenyl-indole (DAPI) (Beyotime, China). Samples were subjected to evaluation under fluorescence microscope (NIKON TE2000S, Japan).
Reverse transcription (RT)-PCR
Total RNA was extracted from cells using AxyPrep total RNA preparation kit (Axygen
, USA) according to the manufacturer’s instructions. Complimentary DNA (cDNA) templates from each sample were prepared from 1 µg of total RNA primed with oligo dT primers using First Strand cDNA Synthesis Kit (TOYOBO, Japan). PCR was performed with gene-specific primers as shown in Table, followed by 30 PCR amplification cycles (94°C for 30 seconds, annealing
at correlative tempreture for 30 seconds, and extension at 72°C for 60 seconds). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a reaction standard; 10 µl of each PCR product was analyzed by 1.5% agarose gel electrophoresis. Authentic bands were detected by GelDocXR imaging system (Bio-Rad, USA) and determined by Quantity One Software (Bio-Rad
|view in a new window |
Table. Primers used in this study and annealing temperature (AT)
Cells were rinsed with ice-cold PBS and resuspended with cold RIPA Lysis Buffer (Beyotime). After 5 minutes on ice, lysates were harvested and centrifuged at 12 000 ×g for 5 minutes. Total protein concentration of lysates was measured using BCA Protein Assay Kit (Beyotime). Soluble protein (80 μg/lane) was separated on 8% polyacrylamide gels and blotted onto nitrocellulose membrane by standard procedures. Primary antibodies used were rabbit polyclonal to CD133 (1/500, Abcam), rabbit polyclonal to GFAP (1/500, Abcam) and GAPDH (1/1000, Boster, China). Nitrocellulose membranes were incubated with the primary antibody at 4°C overnight, washed for three times with Tris Buffered Saline with Tween (TBST) and incubated with the alkaline phosphatase (AP)-labeled secondary antibody for 1 hour at room temperature. The membranes were visualized using diaminobenzidine (DAB) detection system. Densitometry was used to quantify the bands by Quantity One Software (Bio-Rad).
Flow cytometry analysis of CD133
CD133 expression of glioma cells under different culture condition were detected through flow cytometry analysis. Up to 106
cells were resuspended in recommended buffer (containing phosphate buffered saline pH 7.2, 0.5% bovine serum albumin, and 2 mmol/L EDTA), and incubated for 30 minutes at 4°C with CD133 antibody (1/300, Abcam) or with Rabbit (DA1E) mAb IgG XPTM
Isotype Control (Cell Signaling, USA). Anti-Rabbit IgG (H+L) Alexa Fluor®
488 (1/500, Cell Signaling) was applied as the secondary antibody. CD133 detection and analysis were performed on BD FACS Aria
Female BALB/c nude mice, 6 to 8 weeks of age, were housed under specific pathogen-free conditions. All animal experimental protocols were approved by Institutional Animal Care and Use Committee, Huazhong University of Science and Technology. Briefly, 1×105 cells in 5-µl PBS were implanted stererotactically into the right basal ganglia of the nude mice brain (coordinates: 1.0 mm forward from bregma, 2.0 mm lateral, and 3.0 mm ventral from the dura) using a 10-µl Hamilton syringe at a speed of 1 µl/min. To determine the tumorigenicity, survival and general performance of mice were monitored daily. Mice with the development of weight loss greater than 10% or neurological signs were inspected through magnetic resonance imaging (MRI) and then the mice brains were immediately fixed by formalin and imbedded by paraffin following hematoxylin & eosin (HE) staining.
Statistics were analysed using SPSS 17.0 software (SPSS Inc., USA). Comparisons among the groups were performed with one-way analysis of variance (ANOVA). Statistical significance was accepted at P <0.05.
The cells cultured under different conditions presented different growth characteristics. C6-Adh cells
showed firm adherence and had elongated branches (Figure 1A and 1D); Trypsinization
for passaging took about 4–5 minutes at 37°C. C6-SC-Sph
cells grew as floating spheres and proliferated to 100–200 µm in diameter in 3–4 days. They could be mechanically filtered and dissociated into single cell without trypsinization, and then formed secondary spheres for serial passage (Figure 1B and 1E)
. C6-SC-Adh cells grew as an adherent monolayer in laminin-coated flask while exhibited shorter cellular branches (Figure 1C and 1F), and took only 2–3 minutes at 37°C for trypsinization.
|view in a new window |
Figure 1. C6 glioma cells cultured under three different growth conditions after one month (A–C) and three months (D–F). A and D showed adherent growth in conventional 10% serum medium (C6-Adh). B and E showed non-adherent spheres growth in serum-free medium (C6-SC-Sph), and C and F showed adherent growth on laminin-coated flask in serum-free medium (C6-SC-Adh).
The proliferation rate varied between the cells as determined by growth curve plot (Figure 2). C6-Adh
and C6-SC-Adh cells
showed a higher rate of proliferation than C6-SC-Sph cells
either for a short-term culture (one month) or a long-term culture (three months) (P <0.05).
Though both were cultured adherently, C6-Adh cells showed a marginally higher propagation rate than C6-SC-Adh cells at one-month culture, while after prolonged culture under respective conditions for three months, the difference turned out to be significant (P
<0.05) (Figure 2
B). Correspondingly, cell cycle
analysis revealed that both adherent cells displayed an increased population of cells in S and G2-M phase than C6-SC-Sph cells cultured for three months (Figure 2C and 2D).
|view in a new window |
Figure 2. Growth curves of C6 glioma cells under respective culture conditions after one month (A) and three months (B). The horizontal axis represents numbers of cell passages and the vertical axis represents ln values, which were derived from ln (numbers of cells/106). C showed cell cycle analysis. Flow cytometry analysis of propidium iodide-stained C6 cells under respective culture conditions after three months. D showed rate of cell population in S and G2-M phase. *P <0.05.
Self-renewal capacity is one of the most important stem cell phenotypes. As manifested by the limiting dilution clone assay and further supported by the more stringent single-cell sphere formation assay (Figure 3), both of C6-SC-Sph and C6-SC-Adh
cells grown in serum-free medium possessed increased self-renewal potentiality compared with C6-Adh cells grown in 10% serum medium. The increased self-renewal capability was even more prominent between C6-SC-Adh and C6-Adh cells as well as after prolonged culture for three months (Figure 3B and 3D).
|view in a new window |
Figure 3. Capability of tumor sphere formation of C6 glioma cells under respective culture conditions after one month (A, B) and three months (C, D). A and C showed capability of tumor sphere formation from gradient cell titers at 1000, 500, 200, 100, 50, 20, and 10 cells per 96-well repectively. B and D showed capability of neurosphere formation from single cell per 96-well. *P <0.05.
C6 glioma cells under different culture conditions showed discriminatory expression of cell makers
To determine the immunophenotype of the three kinds of cells, we detected their immunoreactivity for makers of stem cells and differentiated cells. CD133 is a well evaluated immunophenotype of neural stem cells, as well as GSCs despite of the controversy. For the analysis of CD133 expression, immunofluorescence, relative quantitative RT-PCR, Western blotting and flow cytometry were applied. However, all the three kinds of cells exhibited lack of immunoreactivity for CD133 (Figure 4, negative data not all shown)
. Contrarily, immunoreactivity
for Nestin and βIII tubulin was detected positive in all the three kinds of cells, but the differences fell short of statistical significance (Figure 4). GFAP is a marker of differentiated neural cell type for astrocytes. Strikingly, C6 glioma cells under three kinds of culture conditions exhibited distinguishing immuno- reactivity for GFAP. Positive expression
of GFAP was only detected in C6-Adh cells by immumofluorescence and RT-PCR (Figure 4A–4D). Compared with C6-SC-Sph and C6-SC-Adh, C6-Adh showed statistically significant immunoreactivity for GFAP at protein level determined by Western blotting, while there was no statistical difference of immunoreactivity for GFAP between C6-SC-Sph and C6-SC-Adh (Figure 4E).
|view in a new window |
Figure 4. C6 cells under different culture conditions showed discriminatory expression of cell makers. A–C: Immunofluorescence analysis of CD133, GFAP and βIII tubulin expression. D: RT-PCR analysis of CD133, Nestin, GFAP and βIII tubulin expression. E: Western blotting analysis of GFAP expression.
Distinct SP ratios of the three populations of cells
SP is considered as a stem cell phenotype and GSC can be enriched in the SP cells. To determine where the C6 cells under different culture conditions have different SP ratios, we detected their capacity to extrude Hoechst 33342 dye by flow cytometry. As shown in Figure
5, SP cells existed in all the three kinds of populations, while only a small proportion of cells had this ability to extrude Hoechst 33342 dye. Distinct SP ratios were found among these three kinds of cells. Cells of C6-SC-Sph and C6-SC-Adh cultured under serum-free supplemented medium
showed enhanced SP ratios, compared with C6-ADH under serum-containing medium (Figure 5G).
|view in a new window |
Figure 5. Side population (SP) analysis. A, C and E showed flow cytometry analysis of SP cells under respective culture conditions after three months. The cells were treated with 50 µmol/L verapamil during the labeling procedure (B, D and F). G showed that C6 cells cultured under serum-free supplemented medium possessed enhanced SP ratios. *P <0.05.
Compared to C6-SC-Sph, C6-SC-Adh had an increased SP ratio, but without statistical significance. The above results suggest SP cells can be enriched under serum-free supplemented medium.
Difference of tumorigenicity
To determine the tumorigenicity of C6 glioma cells under different culture conditions, three kinds of cells were implanted stererotactically into the right basal ganglia of the nude mice brain. Tumor burdened mice with the development of weight loss greater than 10% or neurological signs were analyzed by MRI and then evaluated by HE staining. C6-Adh, C6-SC-Sph and C6-SC-Adh were all displayed in situ oncogenicity. Cells of C6-SC-Adh were more oncogenic with shortened survival. However, statistically significant differences of median survival time were not confirmed (Figure 6).
|view in a new window |
Figure 6. Difference of tumorigenicity. A–C: MRI coronal sections of the brains of nude mice injected with C6 cells cultured under respective culture conditions after three months. The present MRI was done at day 27 after the injection of C6 cells. Abnormal signal hyper- intensities were detectable on the T2-weighted image. D–F: HE staining of mice brains bearing tumors. G: Kaplan-Meier blot showing survival of mice. H: Mean survival time of mice after injected with C6 cells under respective culture conditions.
Despite advances in surgery, radiation and chemotherapy, malignant gliomas are still highly lethal tumors. Traditional treatments that rely on nonspecific, cytotoxic approaches have a marginal impact on patient survival. Most current glioma researches are focused on the molecular and cellular analysis of the bulk tumor mass. However, recent advances in the cancer biology underlying glioma pathogenesis have revealed that GSCs play critical roles in the initiation and propagation of gliomas. The GSCs hypothesis posits that a subpopulation of cells within gliomas has true clonogenic and tumorigenic potential.19 The identification of GSCs provides a powerful tool to investigate the tumorigenic process and to develop therapies targeted at the GSCs. Specific genetic and molecular analyses of the GSCs will further our understanding of the mechanisms of glioma growth, reinforcing parallels between normal neurogenesis and brain tumorigenesis. Researches for GSCs primarily require the derivation and identification of bona fide GSCs populations.
GSCs cultures have been derived from surgical specimen and established cell lines by various in vitro methods,5-10 amongst which neurosphere assay is widely used. In neurosphere culturing, cells are plated on non-adherent surface and in serum-free medium supplemented with mitogens, where stem-like cells are able to continually divide and form multipotent clonal spheres called neurospheres. Although used as the standard in vitro method for identifying the presence of stem cells,2,11-13 the neurosphere assay is now subjected to challenges since some authors recently reported a more effective adherent culture of GSCs.16 However, the report provided no quantitative comparison of the stemness phenotypes of the cells cultured as neurospheres and adherently.17
In the present study, C6 glioma cells were used to clarify the validity of using established cell lines as GSCs source of culture, and to critically evaluate and quantitatively compare the stemness characteristics of the cells grown under distinctive culture conditions. We firstly found that although both C6-Adh and C6-SC-Adh cells grew adherently, C6-SC-Adh cells had shorter branches and looser adherence to the plate even though grown on laminin-coated surface. Thus we speculated that cells grown in the so called “stem cell” medium tend to aggregate non-adherently although they could be “forced” to an adherent growth when put onto the laminin surface. It is then followed by a question how the adherent or sphere growth pattern is relevant to the stemness characteristics.
Self-renewal capacity is a cardinal characteristic of stem cell biology. Interestingly, in our cells grown under different conditions, the dominant determinant of cells self-renewal capability was the stem cell medium, i.e., serum-free medium supplemented with mitogens, but not adherent or sphere growth patterns. While it is notable that the increased self-renewal capability was even more remarkable between C6-SC-Adh and C6-Adh cells than that between C6-SC-Sph and C6-Adh cells, and even more so after prolonged culture in the stem cell medium. In other words, the adherent growth in stem cell medium provided the most favorable conditions for cells self-renewal. In line with the results, it has been conceived that the physical and geometric constraints of a growing sphere limit the diffusion of culture media, and the presence of internal and external cells results in the formation of gradients of nutrients, oxygen, and growth factors within the spheres, all of which could act as important determinants of stem cell phenotypes.20-22 Our results that the adherent cells showed higher proliferation rate also supported this conjecture.
CD133 is the most exploited surface marker of GSCs, while there are emerging paradoxes, e.g., CD133– GSCs may exist. Surprisingly, in the current study, CD133 immunophenotyping was complete negative in all the three cell types, and the results were further confirmed by Western blotting and PCR. On the other hand, Nestin was positive among all the cells while there was no differential expression between the cells. Remarkably, GFAP expression was significantly downregulated in cells cultured in the stem cell medium regardless growing adherently or as spheres, while the marker for neuronal lineage cells – βIII tubulin – remained unchanged. Taken together, of the tentative markers used for GSCs and differentiated cells, only the astrocytic lineage marker GFAP showed a significant decrease after C6 cells being transferred into the stem cell medium. In line with other reports, the cells showed an aberrant co-expression of GFAP and βIII tubulin.23 Contrary to the few reports that had detected a low expression of CD133 in C6 cells,24-26 we failed to confirm it even when the cells were grown under stem cell culture conditions for three months. Taking into account the varied CD133 antibodies commercially available against its distinctive glycosylated epitopes (i.e., AC133 and AC141) and the fact that these glycosylated epitopes may be not synonymous with CD133 itself,10 further validation using different antibodies may be needed.
SP assay is a valid marker-independent method of identifying cancer stem cells. We detected a very low fraction of SP cells in the conventionally cultured C6 cells and the SP fraction increased in cells cultured in stem cell medium, which is accordant with the report by others.23 However, there are ongoing controversies regarding the toxicity of Hoechst 33342 used in SP assay and there are also other intrinsic limitations of the assay itself.10 Final identification of the GSCs still rely on the in vivo tumorigenicity assay.
Surprisingly, in vivo assays on the three cell types revealed no significant difference in their tumorigenicity. Thus it is deducible that the in vitro stemness alterations seen in the cells cultured in serum-free serum might be only reflective of epigenetic phenomena resulting from the artificial manipulation of in vitro growth conditions, and not sustainable and reproducible in vivo. Given the fact that the C6 itself is a cell line maintained in vitro for numerous passages in serum-containing medium, we suggest precautions in using the cell as the GSCs source of culture. Furthermore, although the cells grown adherently in stem cell medium were endowed with increased stemness characteristics compared with those grown as spheres, it could only be quantified in vitro but not in vivo, which might supplement the data reported by Pollard et al16 and contradict the notion that adherent culture in stem cell medium provided a growth condition in favor of GSCs expanding. Nevertheless, further assays, i.e., limiting dilution assay of tumorigenicity and similar work on other established glioma cell lines, are indicated for an undisputed conclusion.
In summary, the present studies critically evaluated and compared the stemness phenotypes of C6 cells grown under different culture conditions. The findings suggested C6 glioma cell line is endowed with some GSC phenotypes that can be moderately enriched in vitro when transferred into stem cell culture condition, which is however not sustainable and reproducible in vivo. Therefore, precautions have to be taken when using the cell as the GSCs source of culture. Meanwhile, adherent culture in stem cell medium did not prove to be a growth condition in favor of GSCs expanding in vivo assays, contradicting some recently proposed notions. Further work in the field is needed for an undisputed conclusion.
1. Clarke MF, Dick JE, Dirks PB, Eaves CJ, Jamieson CH, Jones DL, et al. Cancer stem cells – perspectives on current status and future directions: AACR Workshop on cancer stem cells. Cancer Res 2006; 66: 9339-9344.
2. Vescovi AL, Galli R, Reynolds BA. Brain tumour stem cells. Nat Rev Cancer 2006; 6: 425-436.
3. Liu Q, Nguyen DH, Dong Q, Shitaku P, Chung K, Liu OY, et al. Molecular properties of CD133+ glioblastoma stem cells derived from treatment-refractory recurrent brain tumors. J Neurooncol 2009; 94: 1-19.
4. Fu J, Liu ZG, Liu XM, Chen FR, Shi HL, Pangjesse CS, et al. Glioblastoma stem cells resistant to temozolomide-induced autophagy. Chin Med J 2009; 122: 1255-1259.
5. Hua W, Yao Y, Chu Y, Zhong P, Sheng X, Xiao B, et al. The CD133+ tumor stem-like cell-associated antigen may elicit highly intense immune responses against human malignant glioma. J Neurooncol 2011; Epub ahead of print.
6. Niu CS, Li DX, Liu YH, Fu XM, Tang SF, Li J. Expression of NANOG in human gliomas and its relationship with undifferentiated glioma cells. Oncol Rep 2011; 26: 593-601.
7. Fan H, Guo H, Zhang IY, Liu B, Luan L, Xu S, et al. The different HMGA1 expression of total population of glioblastoma cell line U251 and glioma stem cells isolated from U251. Brain Res 2011; 1384: 9-14.
8. Inoue A, Takahashi H, Harada H, Kohno S, Ohue S, Kobayashi K, et al. Cancer stem-like cells of glioblastoma characteristically express MMP-13 and display highly invasive activity. Int J Oncol 2010; 37: 1121-1131.
9. Zheng X, Shen G, Yang X, Liu W. Most C6 cells are cancer stem cells: evidence from clonal and population analyses. Cancer Res 2007; 67: 3691-3697.
10. Wan F, Zhang S, Xie R, Gao B, Campos B, Herold-Mende C, et al. The utility and limitations of neurosphere assay, CD133 immunophenotyping and side population assay in glioma stem cell research. Brain Pathol 2010; 20: 877-889.
11. Reynolds BA, Rietze RL. Neural stem cells and neurospheres — re-evaluating the relationship. Nat Methods 2005; 2: 333-336.
12. Chaichana K, Zamora-Berridi G, Camara-Quintana J, Quinones-Hinojosa A. Neurosphere assays: growth factors and hormone differences in tumor and nontumor studies. Stem Cells 2006; 24: 2851-2857.
13. Chaichana KL, Guerrero-Cazares H, Capilla-Gonzalez V, Zamora-Berridi G, Achanta P, Gonzalez-Perez O, et al. Intra-operatively obtained human tissue: protocols and techniques for the study of neural stem cells. J Neurosci Methods 2009; 180: 116-125.
14. Reynolds BA, Weiss S. Clonal and population analyses demonstrate that an EGF-responsive mammalian embryonic CNS precursor is a stem cell. Dev Biol 1996; 175: 1-13.
15. Suslov ON, Kukekov VG, Ignatova TN, Steindler DA. Neural stem cell heterogeneity demonstrated by molecular phenotyping of clonal neurospheres. Proc Natl Acad Sci U S A 2002; 99: 14506-14511.
16. Pollard SM, Yoshikawa K, Clarke ID, Danovi D, Stricker S, Russell R, et al. Glioma stem cell lines expanded in adherent culture have tumor-specific phenotypes and are suitable for chemical and genetic screens. Cell Stem Cell 2009; 4: 568-580.
17. Reynolds BA, Vescovi AL. Brain cancer stem cells: think twice before going flat. Cell Stem Cell 2009; 5: 466-467; author reply 468-469.
18. Campos B, Wan F, Farhadi M, Ernst A, Zeppernick F, Tagscherer KE, et al. Differentiation therapy exerts antitumor effects on stem-like glioma cells. Clin Cancer Res 2010; 16: 2715-2728.
19. Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T, et al. Identification of human brain tumour initiating cells. Nature 2004; 432: 396-401.
20. Carlsson J, Acker H. Relations between pH, oxygen partial pressure and growth in cultured cell spheroids. Int J Cancer 1988; 42: 715-720.
21. Bell HS, Whittle IR, Walker M, Leaver HA, Wharton SB. The development of necrosis and apoptosis in glioma: experimental findings using spheroid culture systems. Neuropathol Appl Neurobiol 2001; 27: 291-304.
22. Meeson AP, Hawke TJ, Graham S, Jiang N, Elterman J, Hutcheson K, et al. Cellular and molecular regulation of skeletal muscle side population cells. Stem Cells 2004; 22: 1305-1320.
23. Kondo T, Setoguchi T, Taga T. Persistence of a small subpopulation of cancer stem-like cells in the C6 glioma cell line. Proc Natl Acad Sci U S A 2004; 101: 781-786.
24. Angelastro JM, Lame MW. Overexpression of CD133 promotes drug resistance in C6 glioma cells. Mol Cancer Res 2010; 8: 1105-1115.
25. Zhou XD, Wang XY, Qu FJ, Zhong YH, Lu XD, Zhao P, et al. Detection of cancer stem cells from the C6 glioma cell line. J Int Med Res 2009; 37: 503-510.
26. Shen G, Shen F, Shi Z, Liu W, Hu W, Zheng X, et al. Identification of cancer stem-like cells in the C6 glioma cell line and the limitation of current identification methods. In Vitro Cell Dev Biol Anim 2008; 44: 280-289.