IDH-mutant gliomas (astrocytomas and oligodendrogliomas)
According to the latest classifications of the WHO, mutations in the IDH genes represent hallmarks of both astrocytomas and oligodendrogliomas. Already from early on, gliomas harboring IDH mutations were associated with a more favorable prognosis ( Parsons et al., 2008; Kim & Liau, 2012). Isocitrate dehydrogenases 1 and 2 (IDH1 and IDH2, encoded by the genes IDH1 and IDH2) are enzymes that catalyze the oxidative decarboxylation of isocitrate to produce α-ketoglutarate (αKG) in the cytoplasm and mitochondria, respectively, using NADP+ as a reducing co-factor. IDH1 and IDH2 are implicated in cellular metabolism and epigenetic regulation (reviewed in Molenaar, Maciejewski, Wilmink, & Noorden, 2018) (Figure 6). Missense mutations were identified in IDH1 and IDH2, at analogous arginine (R) residues of the active sites of the enzymes (R132 for IDH1 and R172 for IDH2, Yan et al., 2009). IDH mutations were found at a heterozygous state but seemed to inhibit the remaining wild type IDH activity, through the formation of catalytically inactive heterodimers (L. Dang et al., 2009; Zhao et al., 2009). Substitutions of IDH1 R132H represented the vast majority of the mutations found. IDH2 mutations were much less frequent than mutations in IDH1 (Hartmann et al., 2009). Mutant enzymes were demonstrated to acquire a neomorphic activity, leading to the conversion αKG to D-2-hydroxyglutarate (D-2HG). D-2HG has been regarded as an oncometabolite. Increased levels of D-2HG were shown to inhibit histone and DNA demethylases, triggering genome-wide epigenetic modifications (Xu et al., 2011). Hypermethylation of CpG islands in gliomas (termed « Glioma CpG island methylator phenotype » or G-CIMP), has been proven to highly depend on the presence of IDH mutations (Noushmehr et al., 2010; Turcan et al., 2012). In some cases, G-CIMP has been associated to a favorable clinical outcome. For instance, hypermethylation of the promoter of MGMT (O-6-Methylguanine-DNA Methyltransferase) has been shown to correlate with G-CIMP and has  been linked with a better overall survival and response to specific therapies (Mur et al., 2015). In the same line, shift of G-CIMP to a demethylated phenotype at recurrence cases, marked a worsened prognosis (Ceccarelli et al., 2016). On the other hand, G-CIMP has been associated with inhibition of differentiation and expression of putative oncogenes (Flavahan et al., 2016; C. Lu et al., 2012).
Other common molecular alterations in IDH-mutant gliomas
IDH-mutant astrocytomas and oligodendrogliomas were early segregated into distinct glioma groups, due to the existence of additional, frequently mutually exclusive molecular alterations. Co-deletion of chromosomal arms 1p and 19q (1p/19q co-deletion), mutation of the promoter of TERT (telomerase reverse transcriptase) and mutations in genes such as CIC (Capicua Transcriptional Repressor), and FUBP1 (Far Upstream Element Binding Protein 1) constituted hallmark alterations of oligodendrogliomas. Mutations in genes such as TP53 and ATRX (ATP-Dependent Helicase ATRX), are frequent events in astrocytomas. Oligodendrogliomas (IDH mutation, 1p/19q co-deletion ) As previously stated, the most defining among the molecular alterations harbored by oligodendrogliomas is the combined deletion of the chromosomal arms 1p and 19q. 1p/19q co deletion was early shown to co-occur with IDH mutations. This combined deletion results from an unbalanced translocation between chromosomes 1 and 19 (Jenkins et al., 2006). Mutations in the TERT promoter are also strongly associated with 1p/19q co-deleted, IDH-mutant oligodendrogliomas ( ~95% according to the analysis of the Cancer Genome Atlas Research Network et al., 2015). TERT, encoding for the catalytic subunit of telomerase, is typically suppressed in somatic cells but it can be reactivated in cancer upon mutations in its promoter (Heidenreich & Kumar, 2017). In addition, mutations in the CIC and FUBP1 constitute frequent alterations in oligodendrogliomas, found in 40-60% and 15-30% of the cases, respectively (Jiao et al., 2012, Cancer Genome Atlas Research Network et al., 2015). Interestingly, CIC maps on the chromosome 19q, while FUBP1 on the 1p; thus, these genes are biallelically inactivated in 1p/19q co-deleted tumors.
Defining “Glioma cells-of-origin” and “Glioma Stem Cells” (GSCs)
As it has become evident by now, deciphering the etiology and the development of gliomas is a necessity that requires scrutiny of their developmental origins and make-up. Two important notions which need to be highlighted are that of the glioma cells-of-origin and that of the glioma stem cells (GSCs). Lack of uniformity in the nomenclature used in the literature, can cause some confusion between GSCs and glioma-cells-of origin.
Similarly with other cancer types, the cells-of-origin of gliomas can be defined as the normal cells that are initially transformed by the acquisition of tumor promoting alterations. These cells correspond to physiologically existing cellular populations, in which developmental pathways go awry, giving rise to the gliomas.
GSCs are defined as the subset of cells within a glioma, which fuel tumor propagation, acting as a renewable reservoir. This is a “functional” definition, referring to the resemblance of a cellular subpopulation with cells with stem cell properties (self-renewal and multipotency) and can be assessed by the capacity of these cells to initiate new tumors. The views on GSCs have greatly evolved since their first definitions.
“Glioma-cells-of-origin” : putative candidates
Cells that proliferate in physiological conditions represent tempting candidates for cells from which gliomas originate. As described below, cumulative evidence from murine experimental models as well clinical studies, have pinpointed the neural stem cells (NSCs), as well as the oligodendrocyte precursor cells (OPCs), as the origin of gliomas. There are a few studies which have implicated more differentiated cell types (neurons and astrocytes) as glioma cells of origin; however, these findings are debatable (Figure 8).
Oligodendrocyte precursors cells (OPCs) as glioma cells-of-origin
Although OPCs seemingly lack in pluripotency compared to NSCs, they represent a very abundant cell population and constitute the major proliferating pool in the adult brain (Dawson et al., 2003; Geha et al., 2010). There are several lines of evidence in the literature supporting the notion that OPCs also serve as cellular origin of different glioma types. Gene transfer of Pdgf-b to CNP-expressing cells in neonatal mouse brains resulted in formation of tumors which harbored features of oligodendrogliomas (CNP+ cells represent primarily OPCs at the neonatal stage, Lindberg, Kastemar, Olofsson, Smits, & Uhrbom, 2009). Persson and colleagues also suggested a non-stem cell origin for oligodendroglioma (Persson et al., 2010). In their study, expression of activated Egfr (v-erbB) under control of the human S100b promoter (a broad marker of glial and committed progenitor cells) resulted in oligodendroglioma-like tumors in mice, which had a worsened latency in a Trp53-deficient background. In pre-symptomatic mice, BrdU labeling in SVZ and white matter (WM) regions showed that proliferation was unaffected in the NSCs of the SVZ, whereas WM regions exhibited considerable proliferation. Moreover, the numbers of NG2+OLIG2+ OPCs were increased in the mutant animals, suggesting that the development of those tumors relied on the expansion of OPCs.
Table of contents :
Résumé extensif en français
Table of contents
Table of Figures
List of Acronyms
Part I: Development of Oligodendroglial cells
1. Stem cells in the neocortex: the origins of neurons and glia
2. Defining “oligodendrocytes” (OLs) and “oligodendrogenesis”
3. Defining “oligodendrocyte precursor cells” (OPCs)
3.1 Embryonic origin of OPCs
3.2 OPCs in the postnatal and adult brain
3.3 Differentiation of OPCs to OLs
3.4 Heterogeneity of OPCs and OLs
4. Regulation of oligodendrogenesis
4.1 Transcriptional regulation
4.2 Regulation by developmental signaling pathways
4.3 Epigenetic regulation
Part II: Gliomas
1. Defining “gliomas”
2. Glioma epidemiology and classification
2.1 Epidemiological data and standard of care
2.2 Glioma classification by the World Health Organization (WHO)
3. Molecular alterations in adult diffuse gliomas
3.1 IDH-mutant gliomas (astrocytomas and oligodendrogliomas)
3.1.a IDH mutations
3.1.b Other common molecular alterations in IDH-mutant gliomas
3.2 IDH-wild type gliomas (GBM)
4. Defining “Glioma cells-of-origin” and “Glioma Stem Cells” (GSCs)
4.1 “Glioma-cells-of-origin” : putative candidates
4.1.a Neural stem cells as glioma cells-of-origin
4.1.b Oligodendrocyte precursors cells (OPCs) as glioma cells-of-origin
4.1.c Differentiated cells as glioma cells-of-origin
4.2 Glioma Stem Cells (GSCs): from subpopulations to functional states
4.2.a Early studies of GSCs
4.2.b GSCs in the era of single cell transcriptomics
5. Tumoral heterogeneity in gliomas
6. The glioma “ecosystem” and its microenvironment
Part III: TCF12 in the CNS and in gliomas
1. TCF12 is mutated in oligodendrogliomas
2. TCF12 belongs to the E protein class of the bHLH family
3. TCF12 roles in development
4. TCF12 roles in the CNS
5. TCF12 roles in cancer
Part I: Submitted manuscript
Figure 1: TCF12 is altered in gliomas.
Supplementary Figure 1 (related to Figure 1)
Figure 2: TCF12 is expressed in oligodendroglial cells
Supplementary Figure 2 (related to Figure 2)
Figure 3: TCF12 mainly occupies active promoter regions in oligodendroglial cells.
Supplementary Figure 3 (related to Figure 3)
Figure 4: TCF12 inactivation in OPCs in vivo results in proliferation defects
Supplementary Figure 4 (related to Figure 4)
Figure 5: Transcriptomic analyses of Tcf12 inactivated cells highlight defects in OPC proliferation, differentiation, and cancer-related pathways, that are conserved in hum
Supplementary Figure 5 (related to Figure 5)
Part II: Additional results
Tcf12 inactivation in a glioma mouse model
1. TCF12 expression in oligodendroglial cells
2. Identification of TCF12 targets in oligodendroglial cells
3. Consequences of Tcf12 inactivation in OPCs in vivo
3.a. Histological analyses of Tcf12 knock-out mice
3.b.Transcriptomic analyses of Tcf12-deficient oligodendroglial cells
4. Parallels between TCF12 roles in oligodendrogenesis and gliomagenesis .