[Frontiers in Bioscience E4, 205-213, January 1, 2012]
Gliomagenesis: a game played by few players or a team effort?
Stefania Mazzoleni1, 2, Rossella Galli1
1
TABLE OF CONTENTS
1. ABSTRACT
Glioblastoma multiforme (GBM) represents the most aggressive and deadliest brain tumor of adults. To date, cell heterogeneity within GBM has been explained by the "hierarchical" model of tumorigenesis, aka the "cancer stem cell" hypothesis. In agreement with this model, only rare tumor cells, namely the cancer stem cells (CSCs), are responsible for GBM initiation and, as such, are considered the favored target of therapy. However, multiple evidence has recently indicated that tumor-initiating cells (TICs) may not represent a restricted and infrequent GBM component; rather, they might constitute most of the cells within the tumor bulk. Here we review several studies that recently shed new light on the process of gliomagenesis. We critically analyze the methodological inconsistencies and drawbacks that are causing protracted controversy in the field. Finally, we discuss the clinical implications and the novel therapeutic scenarios that have been put forward by the presence of functionally and molecularly distinct subpopulations of GBM-initiating cells within the same tumor.
2. INTRODUCTION
Glioblastoma multiforme (GBM) is the most common and aggressive brain tumor of adults (1). Clinical presentation and molecular profiles subdivide GBM into primary and secondary subtypes, which comprise 90 and 10% of all cases, respectively. Primary GBM arise in an acute de novo manner with no evidence of prior symptoms or antecedent low-grade pathology. In contrast, secondary GBM derive from the progressive transformation of low-grade gliomas. GBM is characterized by rapid growth characteristics and extreme invasive behavior, thus being inherently resistant to conventional therapy. As a consequence, median patient survival is approximately 14 months. Although standard treatment with surgery, irradiation, and chemotherapy postpones progression and extends survival, these tumors invariably recur and result in death. Therefore, improvement of treatment options for patients with GBM is imperative.
As the adjective "multiforme" implies, GBM is a highly heterogeneous tumor. Tumor heterogeneity may be intended as the phenomenon by which, similar to other cancers, GBM comprise proper tumor cells and multiple non-tumor cell types (i.e. stromal cells, inflammatory cells, pro-angiogenic cells, etc.), which interact to promote and sustain tumorigenesis (2). On the other hand, tumor heterogeneity might also be intended as an additional level of cellular complexity in tumor organization, generated by intrinsic differences retrieved in proper tumor cells in terms of proliferation, invasion and specialization. This last definition of tumor heterogeneity has been explained through two main mechanistic models: the "cancer stem cell" (or "hierarchical" model) (3) and the "stochastic" (or "clonal evolution") model (4). According to the cancer stem cell model, tumor initiation and progression are sustained only by a rare subpopulation of tumor cells with "stem-like" properties and, thus, defined cancer stem cells (CSCs). These stem-like tumor cells are characterized by deregulated self-renewal, which leads to uncontrolled proliferation and aberrant differentiation. In this scenario, the majority of tumor cells, deriving by the differentiation of cancer stem cells, lacks self-renewal potential and, hence, does not contribute significantly to tumor perpetuation. Tumor heterogeneity is, therefore, generated by the coexistence of cancer stem cells (the real "players") and their not tumorigenic differentiated progeny within the same tumor. On the contrary, the stochastic model claims that most cells within a tumor are capable of self-renewal and contribute substantially to tumor initiation and maintenance. Tumor heterogeneity is ascribed mostly to genetic and epigenetic differences existing between different tumor clones. In this view, the distinct cell phenotypes retrievable within a tumor may reflect the presence of cellular clones at different stages of neoplastic transformation and malignancy. Most importantly, this mechanistic model endows all tumor cells with tumor-initiating ability, although to a different extent, thus resulting in the process of tumorigenesis being sustained by a cellular "team" effort.
3. CURRENT CONTROVERSIES IN THE CANCER STEM CELL FIELD
3.1. Tumor heterogeneity: cellular oligarchies or cellular confederations?
The field of CSC is currently animated by a vigorous debate concerning the applicability of the two different models to distinct type of tumors. In agreement with the CSC model, the cardinal property that allows defining a cancer cell as a bona fide cancer stem cell is the exclusive capability to generate tumors that recapitulate the main histo-pathological traits of the patient's tumor. This main characteristic has to be maintained also during serial transplantation and limiting dilution assay in vivo. Several evidences from various laboratories have shown that in many different types of cancer, such as hemopoietic malignancies, breast and colon cancer, only a rare subpopulation of tumor cells is endowed with the capacity to form new tumors, thus providing strong support to the "cancer stem cell" model. Historically, the first evidence of the existence of CSCs came from seminal studies in acute myeloid leukemia, in which a rare subset of cells, comprising 0.01-1% of the total cell population, was shown to induce leukemia upon experimental transplantation (5). Most recently, putative CSC populations have been identified in solid tumors, by exploiting either dye efflux assays or immuno-panning for distinct cell surface antigens. Among the latter, CD24, CD44 and ALDH1 have been used to identify CSC populations from breast tumors (6, 7), AC133 has been used to identify putative CSCs from brain (8, 9), colon (10,11), pancreatic (12) and lung (13) tumors, EpCAM and CD44 from colon tumors (14), and CD44, CD24 and ESA from pancreatic tumors (15).
However, very recent studies suggested that the cancer stem cell model might not fit to all types of cancers. In 2007, Kelly and colleagues demonstrated that, in contrast with previous findings, more than 10% of tumor cells isolated from three different mouse models of primary hemopoietic tumors readily induced tumor growth when injected into non-irradiated recipients (16). Likewise, in 2008, Shmelkov and colleagues reported that the AC133 negative cell fraction in mouse colon cancer was endowed with tumorigenic potential (17). Along this line, single cell transplantation assays indicated that the frequency of cancer-initiating cells in human melanoma could be higher than 25% (18). Similar conflicting results have been recently reported for brain tumors, in particular GBM. Indeed, multiple cohorts of tumor-initiating cells might be active in GBM, each identifiable based on the expression of specific markers and characterized by distinct functional (i.e. invasive, pro-angiogenic or proliferative) phenotypic features and molecular profiles (19, 20).
3.2. GBM CSC markers: better surrogate or functional?
The progress in brain cancer stem cell/tumor-initiating cell field is currently flawed by the inconsistency of results reported by different laboratories. One of the most likely reasons for this discrepancy might relate to the different experimental procedures exploited for the isolation and enrichment of tumor cell subpopulations (21). For many years, the most frequently employed methodology relied on magnetic bead-based cell sorting. Recently, the specificity and accuracy of this procedure have been put into question, thus explaining the seeming lack of tumor initiation capacity of marker-negative cell fractions (22). Another bias in the interpretation of controversial findings is generated by the use of xenografts obtained from the implantation of patient's tumor samples rather than the exploitation of the human samples as such. Although xenografting bears the significant advantage of increasing the availability of human-derived biological material to be subjected to many different analyses and of preserving the expression of specific molecules, e.g. mutant variants of the EGFR (23,24), it is also known that, in general, in vivo passaging of tumor cells might result in relevant modifications in their phenotypic and functional properties (25-27). Likewise, the purification of tumor cell subpopulations only from in vitro established cancer stem cell lines without any concurrent cell isolation directly from patient's tumor specimens might restrict the significance and the breadth of the findings (20, 28). Finally, culturing of tumor cells before sorting might alter the outcome of the subsequent analyses (8, 25).
A variety of surface markers have been proposed to identify tumor-initiating cells (TICs) in GBM, the most commonly used being AC133, A2B5, L1CAM, SSEA-1/CD15 and, very recently, ITGA6 and EGFR (29) (Table 1). With the exception of ITGA6, L1CAM and EGFR, these markers do not exert a specific and clear biological function.
The widely used marker AC133 (also known in rodents as Prominin-1) is a pentaspan transmembrane glycoprotein, which specifically localizes to cellular membrane protrusions. This surface marker was used for the first time for the isolation of hematopoietic and neural stem cells (NSCs) (30, 31). Then, it has been employed for the identification of tumor-initiating cells with stem cell properties from different brain cancers, such as GBM and medulloblastoma (8, 32, 33). In the studies focused on GBM, AC133-positive cell fraction was characterized by exclusive tumorigenic potential after intracranial transplantation into NOD-SCID mice, whereas the AC133 negative cell fraction wasn't tumorigenic at all. However, several studies reported recently that AC133-negative GBM cells, either obtained directly from GBM specimens (26, 34) or from established GBM CSC lines (28) were endowed with tumor-initiating capacity, although to a lesser extent than AC133-positive cells. The finding that the AC133neg cell component isolated from fresh GBM specimens, without any in vitro culturing before and after cell sorting, could be tumorigenic, as shown in (34), might well be explained by the modality through which the distinct GBM subpopulations were purified, i.e. by fluorescence activating cell sorting (FACS). This method is different from that used in previous reports, in which cell enrichment was achieved by magnetic bead-based assays (Table 1). Notably, a very recent study, which compared the use of magnetic activated cell sorting with FACS in GBM fractionation, demonstrated that AC133-microbeads lacked of specificity, with no enrichment for AC133 expression in the isolated cell fractions (22). By this way, most tumor cells could have bound unspecifically to AC133-microbeads, thus explaining the apparently exclusive tumorigenic ability of AC133pos GBM cells. Another possible explanation for the contrasting results obtained by different laboratories might reside in the presence of contaminating AC133pos cells into the marker-negative cell fraction, although this possibility should be ruled out by standard assessment of cell purity or by the inclusion of internal controls, e.g. tumor tissues devoid of AC133pos cells in origin (19). Moreover, given that endothelial cells in the tumor stroma also express AC133, it might be hypothesized that the AC133pos cell fraction containing both tumor cells and stroma is endowed with a growing advantage over the AC133neg fraction. However, we recently reported that both FACS-purified AC133pos and AC133neg GBM cells were capable to promote tumor formation into nude mice even after exclusion of endothelial cells (19). Finally, also the type of anti-AC133 antibodies used for TIC isolation (e.g. anti-AC133/1 and anti-AC133/2, which recognize two different glycosylated epitopes of the transmembrane protein CD133), might affect the experimental consistency between laboratories (35). Notwithstanding the fact that AC133 could not be considered a universal marker, it remains helpful in association with other markers, given that its expression has been associated to fast proliferating cells (28, 36).
Another interesting marker is the cell surface ganglioside A2B5, known to be expressed in neural precursor cells in both the embryonic and adult human brain (37). In GBM, A2B5 expression identified a subpopulation of cells characterized by self-renewal, multipotency and tumor initiation capacity (34, 38). However, whereas the first study reported that both A2B5+CD133+ and A2B5-CD133- subpopulations were endowed with tumor formation capacity, the latter suggested that A2B5+ cells only were tumorigenic. As a matter of fact, these conflicting results might be due again to differences in the methodology employed for isolating GBM cells, with the use of FACS in the study documenting wider tumorigenic ability and the exploitation of magnetic bead-based sorting in the other study.
Very recently, a novel marker, CD15, has been proposed to specifically label tumor-initiating cell subpopulations both in medulloblastoma (MB) and in GBM (39, 40, 41). CD15 is a carbohydrate expressed by glycoproteins, glycolipids and proteoglycans. Also known as SSEA-1 (stage-specific embryonic antigen-1), CD15 represents a useful marker for the identification of ES-derived and embryonic NSCs (42, 43). Notably, GBM-derived CD15 positive cells are characterized by self-renewal, multipotency and enhanced expression of stem cell associated genes, such as Sox2 and Bmi1. CD15-positive GBM subpopulations only are highly tumorigenic, with occasional tumor formation observed after the transplantation of CD15-negative cells, ascribed by the Authors to a possible cell contamination after FACS. However, others and we recently demonstrated that FACS-purified CD15-negative cells from GBM could also give rise to experimental tumors with 100% efficiency, putting into question the exclusive tumor-initiating ability of CD15-positive GBM cells (44).
In addition to the surrogate markers described so far, three functional markers have also been suggested as prospective GBM TIC markers. L1CAM, also known as L1 or CD171, is a transmembrane protein involved in neural cell growth, survival, migration, axonal outgrowth and neurite extension during central nervous system development (45). Based on these observations, L1CAM has been identified as a potential therapeutic target also in neuro-oncology, given that it is overexpressed in gliomas (46). In 2008, Bao and colleagues observed that L1CAM+ and CD133+ cells co-segregated in glioma cells (47). Suppressing L1CAM expression in CD133+ cells by RNAi negatively affected neurosphere formation and induced apoptosis, thus leading to GBM growth inhibition. The knockdown of L1CAM decreased Olig2 expression and upregulated the p21WAF1/CIP1 tumor suppressor in CD133+ glioma cells, thus indicating that L1CAM was required for maintaining the growth and survival of CD133+ glioma cells, both in vitro and in vivo (47).
Integrin alpha-6 (ITGA6) is a cell-surface protein that functions as receptor for the ECM protein laminin. Integrins are known to participate in cell adhesion as well as in cell-surface mediated signaling. Integrin a6 is highly expressed in embryonic, hematopoietic, and NSCs (48). Very recently, Lathia et al. proposed ITGA6 as an enrichment marker for GBM cancer stem cells (49). ITGA6 was coexpressed with the putative CSC marker AC133 and identified cancer cell with a stem cell phenotype and enhanced tumorigenic potential as compared to ITGA6-negative cells, which, however, could also give rise to experimental, tumors although at a very low frequency.
The third functional marker recently suggested playing a role in gliomagenesis, through the identification of highly malignant TICs is the epidermal growth factor (EGF) receptor (-R) (19). EGFR has been proposed as neural stem cells (NSC) marker (50) that plays a relevant role in regulating subevntricular zone (SVZ) NSC division and stemness maintenance (51,52). Most importantly, EGFR is one of the molecular markers that distinguish the two main GBM subtypes, being expressed in more than 60% of primary GBM and absent in secondary GBM (53). EGFR expression has both diagnostic and prognostic significance, although the prognostic value of the EGFR expression is still controversial (54). According to this study, EGFR is unevenly distributed among the cells of the same GBM patient's sample, thus proposing that EGFR might identify distinct subsets of GBM cells. Accordingly, different GBM cell subpopulations, FACS-purified from GBM patient's specimens and from cancer stem cell (CSC) lines based on the expression of EGFR and independent from the co-expression of AC133/CD15, were molecularly distinct and all endowed with tumorigenic potential (Figure 1). Notably, EGFR-expressing tumor-initiating cells (TICs) displayed the most malignant functional and molecular phenotype. Accordingly, modulation of EGFR expression by gain-of-function and loss-of-function strategies in CSC lines promotes and inhibits tumor formation, respectively. Remarkably, EGFR expression is also an obligated requirement for gliomagenesis. In fact, EGFRneg TICs re-express EGFR upon experimental tumorigenesis, thus implying that the expression of stem cell markers might be dinamically regulated (19). In agreement with this observation, the presence of dynamic subpopulations of tumor-maintaining cells has been recently proposed to explain tumor heterogeneity in melanoma (55) and glioblastomas as well (20).
In summary, it appears that a plethora of prospective TIC markers are already available, which can be exploited as enrichment markers, either alone or combined. However, the identification of markers that could simultaneously be used as enrichment marker and therapeutic target might be beneficial and highly desirable.
4. EXISTENCE OF MULTIPLE COHORTS OF TUMOR-INITIATING CELLS IN GBM.
4.1. Implications for therapy
Multiple evidence has convincingly demonstrated that, in solid tumors, CSCs may not represent a restricted and infrequent tumor component; rather, they might comprise the majority of the cells within the tumor bulk, all endowed with tumor-initiating ability, thus being better defined as tumor-initiating cells (TICs) (ref.17 eliminated). Accordingly, findings from different laboratories now indicate that a novel biological scenario, built upon the coexistence of different tumor-initiating cell (TIC) subpopulations within the same tumor and more in line with the stochastic model of tumorigenesis, might be in place also in GBM (19, 20).
Notably, the presence of functionally heterogeneous TIC subpopulations within the same tumor might affect clinical response to treatment and might be mainly responsible for treatment failure in GBM. In the case of EGFR, treatment by means of tyrosine kinase inhibitors (TKI), i.e. small-molecule compounds that inhibit the kinase domain of the EGFR, has been proposed as a possible therapeutic strategy in GBM (56). To date, several clinical studies have reported the only partial efficacy of single agent EGFR inhibition for GBM treatment (57). This failure has been ascribed primarily to the maintenance of a high level of Akt-dependent signalling in PTEN-mutated tumors (58, 59), to the co-activation in the same tumor cell of multiple RTKs (60), as well as to low drug concentrations in tumor tissue or drug inactivation mechanisms. In addition to these possible explanations, the unsatisfactory outcome of EGFR targeted therapies in GBM might also depend on the coexistence of distinct TICs within the same tumor. In fact, highly malignant EGFR-expressing GBM TICs did respond to tyrosine kinase inhibitors (TKIs), thus proving to be EGFR "oncogene addicted" and as such representing a valuable and rationale therapeutic target (19). On the other hand, EGFRneg GBM TICs, which were present within the same tumor together with EGFRpos cells, did not respond to treatment. Thus, although the EGFRpos cell component in EGFRpos GBM can be efficiently eliminated by TKI treatment (Figure 2A), the non-responsiveness of the EGFRneg TIC fraction to EGFR inhibitors might allow these cells to survive treatment, giving rise to tumor recurrence. Hence, to achieve complete tumor eradication, therapeutic strategies should take into consideration the GBM "minimal residual disease" constituted by non-responder GBM TICs, against which developing innovative and specifically targeted approaches (Figure 2B).
Importantly, a wide-ranging therapeutic strategy should include also GBM subpopulations isolated by exploiting other putative TIC markers. As an example, a combination therapeutic paradigm might be hypothesized in GBM by simultaneously targeting ITGA6-expressing GBM TICs with inhibitors of integrin-dependent signaling, such as Cilengitide (49), and EGFR-expressing TICs with EGFR-targeted biological drugs (19).
4.2. Implications for drug discovery
Molecularly targeted therapies hold the promise of providing new anticancer treatments that are more effective and less toxic than traditional cytotoxic chemotherapy. Unfortunately, results of first generation targeted therapy trials for malignant gliomas have been disappointing. Indeed, single-agent drugs have been largely unsuccessful when tested in recurrent disease clinical trials. Thus, only the development of multi-targeted strategies, inhibiting multiple molecular (and cellular) targets simultaneously, might overcome GBM non-responsiveness. To this end, novel therapeutic molecular targets should be identified, which might complement current treatment options. Gene expression studies might be instrumental not only for the understanding of the functional differences between tumor (initiating) cells, but also for identifying novel therapeutic targets. As a matter of fact, clinically relevant distinct molecular subtypes of GBM have been already identified by transcriptional profiling of large collections of GBM patient specimens (61).
However, in spite of novel promising findings (62), neither consensus gene expression profiles have been conclusively associated to GBM patient outcome, nor specific genes have been convincingly interpreted as prognostic factors or promising "druggable" targets. To overcome these limitations, it might be useful to identify molecular mediators that might act both as predictors of survival and therapy choice as well as potential therapeutic targets by exploiting the distinct TIC subpopulations identified in GBM. The identification of minimal gene signatures associated to distinct TICs in GBM will help optimizing and individualizing GBM therapy by prospectively identifying those patients who will benefit most from a specific therapy and by pinpointing novel therapeutic targets for those patients refractory to standard therapy. In the end, a biomarker panel with predictive power to distinguish treatment-sensitive from treatment-refractory GBM patients might become available.
5. SUMMARY AND PERSPECTIVE
In conclusion, the biological landscape that many different studies are progressively building for GBM is complex and variegated.
Firstly, many evidences are suggesting that a stochastic model of tumorigenesis might be more appropriate than the CSC model to explain the intrinsic heterogeneity of GBM. These findings are in line with several recent studies supporting the same view for other types of cancer. However, this novel interpretation of gliomagenesis should not be envisioned as a dismissal of the CSC model. As a matter of fact, the CSC hypothesis proves to be well founded for many hemopoietic malignancies and for several solid tumors. It might be speculated that breast, colon, pancreatic cancers as well as medulloblastoma, all known to originate from tissues in which a rigorous defined cellular hierarchy is defined, i.e. the mammary gland, the intestinal crypts, the pancreatic glands and the external granular layer of the cerebellum, might maintain the same type of hierarchic cellular organization during tumorigenesis. On the contrary, solid cancers as melanomas and GBMs, which derive from tissues with a mesenchymal-like organization such as the neural crest and the mature brain, might be sustained by a stochastic and highly flexible model of tumorigenesis.
Secondly, the exploitation of a combination of different enrichment markers might be necessary to fully depict the complexity of GBM biology in terms of cell composition. In particular, the use of functional, rather than surrogate, markers might allow selective and efficient therapeutic targeting of the most malignant tumor-initiating cells. Advanced molecular studies are therefore required for singling out novel molecular targets that selectively identify each GBM subpopulation.
Finally, effective therapeutic strategy for GBM should take into consideration the presence of many different "actors" playing on the GBM stage. In this scenario, it should be reminded that the expression of stem cell markers in GBM might be dynamically regulated, given that stem cell marker-negative cells can become marker-positive. Hence, GBM heterogeneity should be considered a dynamic process mediated by temporarily distinct GBM cell subpopulations that need to be targeted timely and according to well-balanced treatment regimens and schedules.
6. REFERENCES
1. F. B. Furnari, T. Fenton, R. M. Bachoo, A. Mukasa, J. M. Stommel, A. Stegh, W. C. Hahn, K. L. Ligon, D. N. Louis, C. Brennan, L. Chin, R. A. DePinho and W. K. Cavenee: Malignant astrocytic glioma: genetics, biology, and paths to treatment. Genes Dev, 21(21), 2683-710 (2007)
doi:10.1101/gad.1596707
2. D. Hanahan and R. A. Weinberg: The hallmarks of cancer. Cell, 100(1), 57-70 (2000)
PMid:11689955
3. T. Reya, S. J. Morrison, M. F. Clarke and I. L. Weissman: Stem cells, cancer, and cancer stem cells. Nature, 414(6859), 105-11 (2001)
doi:10.1038/35102167
PMid:18519656
4. J. M. Adams and A. Strasser: Is tumor growth sustained by rare cancer stem cells or dominant clones? Cancer Res, 68(11), 4018-21 (2008)
doi:10.1158/0008-5472.CAN-07-6334
PMid:9212098
5. D. Bonnet and J. E. Dick: Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nature Medicine, 3(7), 730-7 (1997)
doi:10.1038/nm0797-730
PMid:12629218 PMCid:153034
6. M. Al-Hajj, M. S. Wicha, A. Benito-Hernandez, S. J. Morrison and M. F. Clarke: Prospective identification of tumorigenic breast cancer cells.(comment)(erratum appears in Proc Natl Acad Sci U S A. 2003 May 27;100(11):6890). Proceedings of the National Academy of Sciences of the United States of America, 100(7), 3983-8 (2003)
doi:10.1073/pnas.0530291100
PMid:18371393 PMCid:2423808
7. C. Ginestier, M. H. Hur, E. Charafe-Jauffret, F. Monville, J. Dutcher, M. Brown, J. Jacquemier, P. Viens, C. G. Kleer, S. Liu, A. Schott, D. Hayes, D. Birnbaum, M. S. Wicha and G. Dontu: ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell, 1(5), 555-67 (2007)
doi:10.1016/j.stem.2007.08.014
PMid:15549107
8. S. K. Singh, C. Hawkins, I. D. Clarke, J. A. Squire, J. Bayani, T. Hide, R. M. Henkelman, M. D. Cusimano and P. B. Dirks: Identification of human brain tumour initiating cells. Nature, 432(7015), 396-401 (2004)
doi:10.1038/nature03128
PMid:16226707
9. M. D. Taylor, H. Poppleton, C. Fuller, X. Su, Y. Liu, P. Jensen, S. Magdaleno, J. Dalton, C. Calabrese, J. Board, T. Macdonald, J. Rutka, A. Guha, A. Gajjar, T. Curran and R. J. Gilbertson: Radial glia cells are candidate stem cells of ependymoma. Cancer Cell, 8(4), 323-35 (2005)
doi:10.1016/j.ccr.2005.09.001
PMid:17122772
10. C. A. O'Brien, A. Pollett, S. Gallinger and J. E. Dick: A human colon cancer cell capable of initiating tumour growth in immunodeficient mice. Nature, 445(7123), 106-10 (2007)
doi:10.1038/nature05372
PMid:17122771
11. L. Ricci-Vitiani, D. G. Lombardi, E. Pilozzi, M. Biffoni, M. Todaro, C. Peschle and R. De Maria: Identification and expansion of human colon-cancer-initiating cells. Nature, 445(7123), 111-5 (2007)
doi:10.1038/nature05384
PMid:18371365
12. P. C. Hermann, S. L. Huber, T. Herrler, A. Aicher, J. W. Ellwart, M. Guba, C. J. Bruns and C. Heeschen: Distinct populations of cancer stem cells determine tumor growth and metastatic activity in human pancreatic cancer. Cell Stem Cell, 1(3), 313-23 (2007)
doi:10.1016/j.stem.2007.06.002
PMid:18049477
13. A. Eramo, F. Lotti, G. Sette, E. Pilozzi, M. Biffoni, A. Di Virgilio, C. Conticello, L. Ruco, C. Peschle and R. De Maria: Identification and expansion of the tumorigenic lung cancer stem cell population. Cell Death Differ, 15(3), 504-14 (2008)
doi:10.1038/sj.cdd.4402283
PMid:17548814 PMCid:1891215
14. P. Dalerba, S. J. Dylla, I. K. Park, R. Liu, X. Wang, R. W. Cho, T. Hoey, A. Gurney, E. H. Huang, D. M. Simeone, A. A. Shelton, G. Parmiani, C. Castelli and M. F. Clarke: Phenotypic characterization of human colorectal cancer stem cells. Proc Natl Acad Sci U S A, 104(24), 10158-63 (2007)
doi:10.1073/pnas.0703478104
PMid:17283135
15. C. Li, D. G. Heidt, P. Dalerba, C. F. Burant, L. Zhang, V. Adsay, M. Wicha, M. F. Clarke and D. M. Simeone: Identification of pancreatic cancer stem cells. Cancer Res, 67(3), 1030-7 (2007)
doi:10.1158/0008-5472.CAN-06-2030
PMid:17641192
16. P. N. Kelly, A. Dakic, J. M. Adams, S. L. Nutt and A. Strasser: Tumor growth need not be driven by rare cancer stem cells. Science, 317(5836), 337 (2007)
doi:10.1126/science.1142596
PMid:19052619 PMCid:2597380
17. S. V. Shmelkov, J. M. Butler, A. T. Hooper, A. Hormigo, J. Kushner, T. Milde, R. St Clair, M. Baljevic, I. White, D. K. Jin, A. Chadburn, A. J. Murphy, D. M. Valenzuela, N. W. Gale, G. Thurston, G. D. Yancopoulos, M. D'Angelica, N. Kemeny, D. Lyden and S. Rafii: CD133 expression is not restricted to stem cells, and both CD133 and CD133 metastatic colon cancer cells initiate tumors. J Clin Invest (2008)
18. E. Quintana, M. Shackleton, M. S. Sabel, D. R. Fullen, T. M. Johnson and S. J. Morrison: Efficient tumour formation by single human melanoma cells. Nature, 456(7222), 593-8 (2008)
doi:10.1038/nature07567
PMid:20858720
19. S. Mazzoleni, L. S. Politi, M. Pala, M. Cominelli, A. Franzin, L. Sergi Sergi, A. Falini, M. De Palma, A. Bulfone, P. L. Poliani and R. Galli: Epidermal growth factor receptor expression identifies functionally and molecularly distinct tumor-initiating cells in human glioblastoma multiforme and is required for gliomagenesis. Cancer Res, 70(19), 7500-13 (2010)
doi:10.1158/0008-5472.CAN-10-2353
PMid:20385361
20. R. Chen, M. C. Nishimura, S. M. Bumbaca, S. Kharbanda, W. F. Forrest, I. M. Kasman, J. M. Greve, R. H. Soriano, L. L. Gilmour, C. S. Rivers, Z. Modrusan, S. Nacu, S. Guerrero, K. A. Edgar, J. J. Wallin, K. Lamszus, M. Westphal, S. Heim, C. D. James, S. R. VandenBerg, J. F. Costello, S. Moorefield, C. J. Cowdrey, M. Prados and H. S. Phillips: A hierarchy of self-renewing tumor-initiating cell types in glioblastoma. Cancer Cell, 17(4), 362-75 (2010)
doi:10.1016/j.ccr.2009.12.049
PMid:19951686
21. C. M. Alexander, J. Puchalski, K. S. Klos, N. Badders, L. Ailles, C. F. Kim, P. Dirks and M. J. Smalley: Separating stem cells by flow cytometry: reducing variability for solid tissues. Cell Stem Cell, 5(6), 579-83 (2009)
doi:10.1016/j.stem.2009.11.008
PMid:19350631
22. V. Clement, V. Dutoit, D. Marino, P. Y. Dietrich and I. Radovanovic: Limits of CD133 as a marker of glioma self-renewing cells. Int J Cancer, 125(1), 244-8 (2009)
doi:10.1002/ijc.24352
PMid:15831234 PMCid:1871885
23. C. Giannini, J. N. Sarkaria, A. Saito, J. H. Uhm, E. Galanis, B. L. Carlson, M. A. Schroeder and C. D. James: Patient tumor EGFR and PDGFRA gene amplifications retained in an invasive intracranial xenograft model of glioblastoma multiforme. Neuro Oncol, 7(2), 164-76 (2005)
doi:10.1215/S1152851704000821
PMid:14603439
24. A. Pandita, K. D. Aldape, G. Zadeh, A. Guha and C. D. James: Contrasting in vivo and in vitro fates of glioblastoma cell subpopulations with amplified EGFR. Genes Chromosomes Cancer, 39(1), 29-36 (2004)
doi:10.1002/gcc.10300
PMid:18784658
25. J. E. Visvader and G. J. Lindeman: Cancer stem cells in solid tumours: accumulating evidence and unresolved questions. Nat Rev Cancer, 8(10), 755-68 (2008)
doi:10.1038/nrc2499
PMid:17955491
26. J. Wang, P. O. Sakariassen, O. Tsinkalovsky, H. Immervoll, S. O. Boe, A. Svendsen, L. Prestegarden, G. Rosland, F. Thorsen, L. Stuhr, A. Molven, R. Bjerkvig and P. O. Enger: CD133 negative glioma cells form tumors in nude rats and give rise to CD133 positive cells. Int J Cancer, 122(4), 761-8 (2008)
doi:10.1002/ijc.23130
PMid:17056721 PMCid:1618812
27. P. O. Sakariassen, L. Prestegarden, J. Wang, K. O. Skaftnesmo, R. Mahesparan, C. Molthoff, P. Sminia, E. Sundlisaeter, A. Misra, B. B. Tysnes, M. Chekenya, H. Peters, G. Lende, K. H. Kalland, A. M. Oyan, K. Petersen, I. Jonassen, A. van der Kogel, B. G. Feuerstein, A. J. Terzis, R. Bjerkvig and P. O. Enger: Angiogenesis-independent tumor growth mediated by stem-like cancer cells. Proc Natl Acad Sci U S A, 103(44), 16466-71 (2006)
doi:10.1073/pnas.0607668103
PMid:17483311
28. D. Beier, P. Hau, M. Proescholdt, A. Lohmeier, J. Wischhusen, P. J. Oefner, L. Aigner, A. Brawanski, U. Bogdahn and C. P. Beier: CD133+ and CD133- Glioblastoma-Derived Cancer Stem Cells Show Differential Growth Characteristics and Molecular Profiles. Cancer Res, 67(9), 4010-5 (2007)
doi:10.1158/0008-5472.CAN-06-4180
PMid:19760641 PMCid:2909872
29. C. A. Gilbert and A. H. Ross: Cancer stem cells: cell culture, markers, and targets for new therapies. J Cell Biochem, 108(5), 1031-8 (2009)
doi:10.1002/jcb.22350
PMid:9389721
30. S. Miraglia, W. Godfrey, A. H. Yin, K. Atkins, R. Warnke, J. T. Holden, R. A. Bray, E. K. Waller and D. W. Buck: A novel five-transmembrane hematopoietic stem cell antigen: isolation, characterization, and molecular cloning. Blood, 90(12), 5013-21 (1997)
PMid:11121071
31. N. Uchida, D. W. Buck, D. He, M. J. Reitsma, M. Masek, T. V. Phan, A. S. Tsukamoto, F. H. Gage and I. L. Weissman: Direct isolation of human central nervous system stem cells. Proceedings of the National Academy of Sciences of the United States of America, 97(26), 14720-5 (2000)
doi:10.1073/pnas.97.26.14720
PMid:17051156
32. S. Bao, Q. Wu, R. E. McLendon, Y. Hao, Q. Shi, A. B. Hjelmeland, M. W. Dewhirst, D. D. Bigner and J. N. Rich: Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. Nature, 444(7120), 756-60 (2006)
doi:10.1038/nature05236
PMid:19369008
33. J. X. Cheng, B. L. Liu and X. Zhang: How powerful is CD133 as a cancer stem cell marker in brain tumors? Cancer Treat Rev, 35(5), 403-8 (2009)
doi:10.1016/j.ctrv.2009.03.002
PMid:18382330
34. A. T. Ogden, A. E. Waziri, R. A. Lochhead, D. Fusco, K. Lopez, J. A. Ellis, J. Kang, M. Assanah, G. M. McKhann, M. B. Sisti, P. C. McCormick, P. Canoll and J. N. Bruce: Identification of A2B5+CD133- tumor-initiating cells in adult human gliomas. Neurosurgery, 62(2), 505-14; discussion 514-5 (2008)
doi:10.1227/01.neu.0000316019.28421.95
PMid:18535813 PMCid:2585385
35. S. Bidlingmaier, X. Zhu and B. Liu: The utility and limitations of glycosylated human CD133 epitopes in defining cancer stem cells. J Mol Med, 86(9), 1025-32 (2008)
doi:10.1007/s00109-008-0357-8
PMid:18829544 PMCid:2561223
36. M. Jaksch, J. Munera, R. Bajpai, A. Terskikh and R. G. Oshima: Cell cycle-dependent variation of a CD133 epitope in human embryonic stem cell, colon cancer, and melanoma cell lines. Cancer Res, 68(19), 7882-6 (2008)
doi:10.1158/0008-5472.CAN-08-0723
PMid:12627226
37. M. C. Nunes, N. S. Roy, H. M. Keyoung, R. R. Goodman, G. McKhann, 2nd, L. Jiang, J. Kang, M. Nedergaard and S. A. Goldman: Identification and isolation of multipotential neural progenitor cells from the subcortical white matter of the adult human brain. Nat Med, 9(4), 439-47 (2003)
doi:10.1038/nm837
PMid:19243384
38. A. Tchoghandjian, N. Baeza, C. Colin, M. Cayre, P. Metellus, C. Beclin, L. Ouafik and D. Figarella-Branger: A2B5 cells from human glioblastoma have cancer stem cell properties. Brain Pathol, 20(1), 211-21 (2010)
doi:10.1111/j.1750-3639.2009.00269.x
PMid:19185848 PMCid:2664097
39. T. A. Read, M. P. Fogarty, S. L. Markant, R. E. McLendon, Z. Wei, D. W. Ellison, P. G. Febbo and R. J. Wechsler-Reya: Identification of CD15 as a marker for tumor-propagating cells in a mouse model of medulloblastoma. Cancer Cell, 15(2), 135-47 (2009)
doi:10.1016/j.ccr.2008.12.016
PMid:19487286
40. R. J. Ward, L. Lee, K. Graham, T. Satkunendran, K. Yoshikawa, E. Ling, L. Harper, R. Austin, E. Nieuwenhuis, I. D. Clarke, C. C. Hui and P. B. Dirks: Multipotent CD15+ cancer stem cells in patched-1-deficient mouse medulloblastoma. Cancer Res, 69(11), 4682-90 (2009)
doi:10.1158/0008-5472.CAN-09-0342
PMid:19427293
41. M. J. Son, K. Woolard, D. H. Nam, J. Lee and H. A. Fine: SSEA-1 is an enrichment marker for tumor-initiating cells in human glioblastoma. Cell Stem Cell, 4(5), 440-52 (2009)
doi:10.1016/j.stem.2009.03.003
PMid:17131412
42. P. Barraud, S. Stott, K. Mollgard, M. Parmar and A. Bjorklund: In vitro characterization of a human neural progenitor cell coexpressing SSEA4 and CD133. J Neurosci Res, 85(2), 250-9 (2007)
doi:10.1002/jnr.21116
PMid:19725119
43. J. Pruszak, W. Ludwig, A. Blak, K. Alavian and O. Isacson: CD15, CD24, and CD29 define a surface biomarker code for neural lineage differentiation of stem cells. Stem Cells, 27(12), 2928-40 (2009)
PMid:20181261 PMCid:2841664
44. C. Patru, L. Romao, P. Varlet, L. Coulombel, E. Raponi, J. Cadusseau, F. Renault-Mihara, C. Thirant, N. Leonard, A. Berhneim, M. Mihalescu-Maingot, J. Haiech, I. Bieche, V. Moura-Neto, C. Daumas-Duport, M. P. Junier and H. Chneiweiss: CD133, CD15/SSEA-1, CD34 or side populations do not resume tumor-initiating properties of long-term cultured cancer stem cells from human malignant glio-neuronal tumors. BMC Cancer, 10, 66 (2010)
doi:10.1186/1471-2407-10-66
PMid:17189949
45. P. F. Maness and M. Schachner: Neural recognition molecules of the immunoglobulin superfamily: signaling transducers of axon guidance and neuronal migration. Nat Neurosci, 10(1), 19-26 (2007)
doi:10.1038/nn1827
PMid:15677537 PMCid:1770574
46. T. Suzuki, S. Izumoto, Y. Fujimoto, M. Maruno, Y. Ito and T. Yoshimine: Clinicopathological study of cellular proliferation and invasion in gliomatosis cerebri: important role of neural cell adhesion molecule L1 in tumour invasion. J Clin Pathol, 58(2), 166-71 (2005)
doi:10.1136/jcp.2004.020909
PMid:18676824 PMCid:2739001
47. S. Bao, Q. Wu, Z. Li, S. Sathornsumetee, H. Wang, R. E. McLendon, A. B. Hjelmeland and J. N. Rich: Targeting cancer stem cells through L1CAM suppresses glioma growth. Cancer Res, 68(15), 6043-8 (2008)
doi:10.1158/0008-5472.CAN-08-1079
PMid:14563990
48. N. O. Fortunel, H. H. Otu, H. H. Ng, J. Chen, X. Mu, T. Chevassut, X. Li, M. Joseph, C. Bailey, J. A. Hatzfeld, A. Hatzfeld, F. Usta, V. B. Vega, P. M. Long, T. A. Libermann and B. Lim: Comment on " 'Stemness': transcriptional profiling of embryonic and adult stem cells" and "a stem cell molecular signature". Science, 302(5644), 393; author reply 393 (2003)
doi:10.1126/science.1086384
PMid:20452317
49. J. D. Lathia, J. Gallagher, J. M. Heddleston, J. Wang, C. E. Eyler, J. Macswords, Q. Wu, A. Vasanji, R. E. McLendon, A. B. Hjelmeland and J. N. Rich: Integrin alpha 6 regulates glioblastoma stem cells. Cell Stem Cell, 6(5), 421-32 (2010)
doi:10.1016/j.stem.2010.02.018
PMid:19332781 PMCid:2669396
50. E. Pastrana, L. C. Cheng and F. Doetsch: Simultaneous prospective purification of adult subventricular zone neural stem cells and their progeny. Proc Natl Acad Sci U S A, 106(15), 6387-92 (2009)
PMid:15797549
51. Y. Sun, S. K. Goderie and S. Temple: Asymmetric distribution of EGFR receptor during mitosis generates diverse CNS progenitor cells. Neuron, 45(6), 873-86 (2005)
doi:10.1016/j.neuron.2005.01.045
PMid:20804972
52. S. R. Ferron, N. Pozo, A. Laguna, S. Aranda, E. Porlan, M. Moreno, C. Fillat, S. de la Luna, P. Sanchez, M. L. Arbones and I. Farinas: Regulated segregation of kinase Dyrk1A during asymmetric neural stem cell division is critical for EGFR-mediated biased signaling. Cell Stem Cell, 7(3), 367-79 (2010)
doi:10.1016/j.stem.2010.06.021
PMid:17189397
53. M. K. Nicholas, R. V. Lukas, N. F. Jafri, L. Faoro and R. Salgia: Epidermal growth factor receptor - mediated signal transduction in the development and therapy of gliomas. Clin Cancer Res, 12(24), 7261-70 (2006)
doi:10.1158/1078-0432.CCR-06-0874
PMid:15977639
54. H. Ohgaki and P. Kleihues: Population-based studies on incidence, survival rates, and genetic alterations in astrocytic and oligodendroglial gliomas. J Neuropathol Exp Neurol, 64(6), 479-89 (2005)
PMid:16282176
55. A. Roesch, M. Fukunaga-Kalabis, E. C. Schmidt, S. E. Zabierowski, P. A. Brafford, A. Vultur, D. Basu, P. Gimotty, T. Vogt and M. Herlyn: A temporarily distinct subpopulation of slow-cycling melanoma cells is required for continuous tumor growth. Cell, 141(4), 583-94 (2010)
PMid:20167819
56. I. K. Mellinghoff, M. Y. Wang, I. Vivanco, D. A. Haas-Kogan, S. Zhu, E. Q. Dia, K. V. Lu, K. Yoshimoto, J. H. Huang, D. J. Chute, B. L. Riggs, S. Horvath, L. M. Liau, W. K. Cavenee, P. N. Rao, R. Beroukhim, T. C. Peck, J. C. Lee, W. R. Sellers, D. Stokoe, M. Prados, T. F. Cloughesy, C. L. Sawyers and P. S. Mischel: Molecular determinants of the response of glioblastomas to EGFR kinase inhibitors. N Engl J Med, 353(19), 2012-24 (2005)
doi:10.1056/NEJMoa051918
PMid:15956649
57. P. C. De Witt Hamer: Small molecule kinase inhibitors in glioblastoma: a systematic review of clinical studies. Neuro Oncol, 12(3), 304-16 (2010)
PMid:19147502 PMCid:2652334
58. D. A. Haas-Kogan, M. D. Prados, T. Tihan, D. A. Eberhard, N. Jelluma, N. D. Arvold, R. Baumber, K. R. Lamborn, A. Kapadia, M. Malec, M. S. Berger and D. Stokoe: Epidermal growth factor receptor, protein kinase B/Akt, and glioma response to erlotinib. J Natl Cancer Inst, 97(12), 880-7 (2005)
doi:10.1093/jnci/dji161
PMid:10647931
59. F. Griffero, A. Daga, D. Marubbi, M. C. Capra, A. Melotti, A. Pattarozzi, M. Gatti, A. Bajetto, C. Porcile, F. Barbieri, R. E. Favoni, M. Lo Casto, G. Zona, R. Spaziante, T. Florio and G. Corte: Different response of human glioma tumor-initiating cells to epidermal growth factor receptor kinase inhibitors. J Biol Chem, 284(11), 7138-48 (2009)
doi:10.1074/jbc.M807111200
PMid:20478252
60. J. M. Stommel, A. C. Kimmelman, H. Ying, R. Nabioullin, A. H. Ponugoti, R. Wiedemeyer, A. H. Stegh, J. E. Bradner, K. L. Ligon, C. Brennan, L. Chin and R. A. Depinho: Coactivation of Receptor Tyrosine Kinases Affects the Response of Tumor Cells to Targeted Therapies. Science (2007)
61. H. S. Phillips, S. Kharbanda, R. Chen, W. F. Forrest, R. H. Soriano, T. D. Wu, A. Misra, J. M. Nigro, H. Colman, L. Soroceanu, P. M. Williams, Z. Modrusan, B. G. Feuerstein and K. Aldape: Molecular subclasses of high-grade glioma predict prognosis, delineate a pattern of disease progression, and resemble stages in neurogenesis. Cancer Cell, 9(3), 157-73 (2006)
doi:10.1016/j.ccr.2006.02.019
PMid:16530701
62. H. Colman, L. Zhang, E. P. Sulman, J. M. McDonald, N. L. Shooshtari, A. Rivera, S. Popoff, C. L. Nutt, D. N. Louis, J. G. Cairncross, M. R. Gilbert, H. S. Phillips, M. P. Mehta, A. Chakravarti, C. E. Pelloski, K. Bhat, B. G. Feuerstein, R. B. Jenkins and K. Aldape: A multigene predictor of outcome in glioblastoma. Neuro Oncol, 12(1), 49-57 (2010)
PMid:20150367 PMCid:2940562
Key Words: Gliomagenesis, Glioblastoma Multiforme, Tumor-Initiating Cells, Tumor Heterogeneity, Review
Send correspondence to: Rossella Galli, Neural Stem Cell Biology Unit, Division of Regenerative Medicine, Stem Cells and Gene Therapy, San Raffaele Scientific Institute, Via Olgettina 58, 20132 Milan, Italy, Tel: 39022643 4626, Fax: 390226434621, E-mail:galli.rossella@hsr.it