Cancer Stem Cells in Brain Tumors and Their Lineage Hierarchy

Abstract

Despite recent advances in the development of novel targeted chemotherapies, the prognosis of malignant glioma remains dismal. The chemo-resistance of this tumor is attributed to tumor heterogeneity. To explain this unique chemo- resistance, the concept of cancer stem cells has been evoked. Cancer stem cells, a subpopulation of whole tumor cells, are now regarded as candidate therapeutic targets. Here, the author reviews and discusses the cancer stem cell concept.

Introduction

Glioma is the most common primary brain neoplasm. In particular, glioblastoma (GBM), is the most malignant type of glioma, and is highly aggressive and has a dismal prognosis with a median survival of about 1 year, despite the use of multidisciplinary treatment approaches. Like to other malignant cancers, GBM shows marked heterogeneity in terms of its cellular morphology, multiple genetic alterations, such as, amplification or activation of epidermal growth factor receptor (EGFR), and in terms of its clinical course (1, 2). This heterogeneity allows tumors to avoid novel chemotherapeutic agents and resist radiotherapy, and even leads to chemo-resistance to several drugs that function via different mechanisms. Therefore, many scientists working in the cancer biology field have studied ways of overcoming this tumor heterogeneity. In particular, radiation- or chemo-resistance may be due to a small subpopulation of cancer stem cells or tumor-initiating cells residing among tumor cells. Cancer stem cell and tumor-initiating cell concepts have been used recently in the cancer biology field, and recent studies have suggested that self-renewing stem cells and cancer cells may be involved in brain tumor development (3-6). However, the differentiation of the cells of origin and cancer stem cells has not been achieved to date and many authors have used these two terms interchangeably (7).

Cancer stem cells of brain

These cancer stem cells provide a reservoir of cells with self-renewal capabilities and maintain tumors by generating differentiated tumor cells. They are also regarded to play an essential role in recurrence after chemo-radiotherapy. Therefore, the identification of cancer stem cells and understanding the mechanisms by which cancer stem cells act appear to be crucial for the development of treatments for these intractable tumors. Cancer stem cells have been shown to exhibit resistance to radiation and chemotherapeutic drugs (3, 8), and to share many similarities with normal stem cells with respect to self-renewal, maintenance of an organ or tumor, and the generation of differentiated cells. Thus, normal tissue stem cells could be converted by the accumulation of genetic mutations, to cells that form cancers. Tumorigenic cells exhibit properties that mirror those of normal stem cells, albeit in a dysregulated fashion (9).

Highly tumorigenic subpopulations of cells in malignant gliomas have been identified by flow cytometry to express the hematopoietic stem cell marker CD133 (5). These tumor stem cells can be cultured as self-renewing ‘ ‘ spheres’ ’ with multipotent characteristics (3). Furthermore, the neural stem cell marker CD133 has been reported to identify cells within GBMs that can initiate neurosphere growth and tumor formation, and CD133(+) cells have been found to become radio-resistant by up-regulation of DNA repair mechanisms (3).

However, in some GBMs, the proportion of CD133(+) cells is immeasurable or very low at ＜1% of all tumor cells, and this finding is at odds with the concept that cancer stem cells persist at a certain level and repetitively produce more differentiated cells. However, these apparently contradictory results may be due to shortcomings of cell-sorting techniques. In the recent studies, CD133(-) cells were found to exhibit properties similar to those of CD133(+) cells. In a xenograft experimental model, larger cell numbers with non-cancer stem cells have been injected to reproduce growth of the tumor, although the rate of tumor formation are lower than that in CD133(+) cells (10, 11). Furthermore, in PTEN-deficient GBM tumors, CD133(-) cells were reported to initiate neurosphere growth (12), which indicates that PTEN status is closely associated with cell lineage and the capabilities of GBM cells. As a result, several authors have questioned the validity of CD133 as a candidate cancer stem cell marker (13). To purify cancer stem cells, multiple markers are required than a single surface marker (14), and thus, attempts to identify more optimal candidate markers continue. Transplantation experiments in a medulloblastoma model demonstrated that cells expressing that progenitors cells expressing CD15 are capable of tumor propagation (15) and other studies suggest that Olig2, a committed neural stem cell progenitor, is required for glioma formation and that a common Olig2-dependent mechanism for cell cycle control regulates the growth of normal and malignant neural progenitor cells (16). The accumulation of mutations in progenitor cells can cause tumor formation, and furthermore tumor-initiating cells have the capacity for multilineage differentiation in situ.

Lineage hierarchy

Regardless of the types of surface markers that best represent brain cancer stem cells, brain tumorigenic cells similar to normal neural stem cells be regulated or dysregulated by normal signaling pathways. Understanding the normal cellular hierarchy from stem cells, that is, from committed progenitor cells to more differentiated cells, is essential for the identification of cancer cells. Furthermore highly tumorigenic cellular subpopulations in adult primary tumors share some similarities with normal stem cells and progenitor cells within the same lineage (17). According to the cell of origin model, different tumor subtypes arise from a lineage hierarchy originating from primitive stem cells (7), which implies that various molecular types can represent multiple genetic mutations of normal lineage hierarchy cells at each step, and that tumor cells are composed of mutated variants from the normal lineage hierarchy. These relations in turn imply that various types of cellular subpopulations in the lineage hierarchy and surrounding microenvironment can contribute to tumorigenesis. This theory evolved from the conventional cancer stem cell concept that only one type of cell is responsible for tumorigenesis and tumor destiny. Simple mutations of neural stem cells without accompanying microenvironmental factors do not induce tumorigenesis, for example, in intestinal cancer, cancer stem cells are closely associated with dysregulation of the WNT signaling pathway.

To preserve the stemness of cancer cells, specialized cellular microenvironments, so-called niches, are required (9). Stem cells within the brain are localized to specific microenvironments in which they are maintained (18). In addition, to the subventricular zone (SVZ), which is known to be representative of the neural stem cell niche, a perivascular location has been shown for brain tumor stem cells (19). The tumor perivascular niche is a complex microenvironment in which multiple cell types and signaling factors are involved in crosstalk between endothelial cells and stem cells (20). In this stem cell niche, SVZ, relatively quiescent ‘ ‘ type B’ ’ neural stem cells give rise to rapidly dividing ‘ ‘ type C’ ’ progenitor cells (21). Furthermore, epidermal growth factor receptor activation increases the proliferation of these type C transit amplifying cells (22, 23), and the activation of platelet-derived growth factor receptor by receptor tyrosine kinases promotes the self-renewal and differentiation of quiescent neural stem cells. Neural stem cells and their progeny are tightly regulated by a variety of intrinsic and extrinsic factors, and this microenvironment modulates cell numbers, self-renewal, and the fate of stem cells. Many efforts have been made to identify the mechanisms that specifically regulate the biology of stem cells rather than whole tumors. The maintenance of balance between neural stem cells and neural progenitors in the subventricular zone is regulated by notch signaling, wnt signaling, bone morphogenic protein (BMP), and the sonic hedgehog pathway (21). Notch signaling participates in the mediation of cell-to-cell communication and promotes neural stem cell expansion without differentiation (16, 24). In addition, perivascular nitric oxide is known to activate notch signaling and promote stem cell features in platelet derived growth factor (PDGF)- induced glioma cells (20). On the other hand, sonic hedgehog signaling is implicated in the maintenance of cells in an undifferentiated state (25, 26), and the inhibition of this signaling was found to reduce tumor sphere and especially the sphere-forming ability of CD133(+) cells (27). Similarly, the Wnt signaling pathway also regulates the survival of neural stem cells (28-30).

Transgenic or specific gene targeting technologies can be used to explore the effects of oncogenes and tumor suppressors in different cellular contexts to clarify the being of cancer stem cells and progenitor cells. The targeting of only one cell subpopulation is expected to reveal tumors that recapitulate the phenotype of the human cancer. In addition to this approach, lineage tracing of cells could also contribute to the elucidation of the origins of brain tumors, and has been suggested to be suitable for identifying cells targeted for transformation.

Conclusion

The prospective isolation of cancer stem cells provides better targets for the development of new therapies and of new ways of measuring treatment efficacy. In addition, more carefully designed experiments are required on the lineage tracing of neural stem cells to clarify the role played by cancer stem cells in brain tumors.

Potential conflict of interest

The authors have no conflicting financial interest.