Cell of origin of glioma: biological and clinical implications

Abstract

The cellular origin of gliomas remains a topic of controversy in cancer research. Advances in neurobiology, molecular genetics, and functional genomics have ushered new insights through exploiting the development of more sophisticated tools to address this question. Diverse distinct cell populations in the adult brain have been reported to give rise to gliomas, although how these studies relate physiologically to mechanisms of spontaneous tumour formation via accumulation of tumour-initiating mutations within a single cell are less well developed. Recent studies in animal models indicate that the lineage of the tumour-initiating cell may contribute to the biological and genomic phenotype of glioblastoma. These results suggest that the cell of origin may not only serve as a source of diversity for these tumours, but may also provide new avenues for improved diagnostics and therapeutic targeting that may prolong the lives of patients.

Figure 1

Lineage hierarchy of tumour-initiating cells in GBM. Neural stem cells (NSC) are activated to give rise to progenitor cells that exhibit varying levels of potentiality: multipotent progenitors (MPPs) that give rise to all CNS cell types; bipotential progenitors (BPP), such as Ascl1 progenitors, that identify both adult oligodendrocyte progenitor cells (OPCs) and neural progenitor cells (NPCs); and unipotent progenitors (UPP) that differentiate into oligodendrocytes, neurons, or astrocytes. Dashed lines indicate that the hierarchical relationship between different progenitors is still unclear. Experiments using different inducible cre driver lines in conjunction with known driver mutations employed to target different adult brain cell types are shown. Models using Nestin-creER^T2, which is expressed in neural stem cells as well as multipotent progenitors, can give rise to GBM. The Gfap-creER^TM, which is expressed in SVZ neural stem cells as well as a subset of mature astrocytes, develop tumours near the neurogenic niches. The Ascl1-creER^TM model, which targets NPCs and OPCs, gives rise to two GBM subtypes. One of these subtypes appears similar to GBM that develop using NG2-creER^TM. Such animal models in which cre recombination is temporally controlled in the adult stem and progenitor populations have been shown to induce GBM formation, although with different molecular subtypes.

Figure 2

Lineage-based functional subtyping of GBM. (Upper left panel) Cell of origin model of intertumoural diversity. Mouse models developed from targeting of tumour-initiating mutations into different cells of origin give rise to different GBM subtypes (represented by different colours). The GBM subtypes from different mouse models are molecularly separable, as hypothetically shown via locally linear embedding (LLE) dimension reduction analysis (upper right panel), and gene expression profiling (lower right panel). Studies on the molecular subtypes of GBMs that express distinct biomarkers, activate or turn off specific signalling pathways, and exhibit differential response to different treatment modalities may pave the way for improved diagnostics and therapeutics.