Role of the Ink4a/Arf tumor suppressors in cerebellar development, stem cells and cancer

In order to take proper cell fate decisions, cells have to guide their biochemical machinery through the appropriate decisions in both differentiation and proliferation. Especially for stem cells such decisions are critical as they have the capacity to self-renew, i.e. give rise to new daughter stem cells and differentiated progeny. Important to mention here is at least a subset of cancers is now thought of as to derive from cancer stem cells. The current view is that either stem cells aberrantly keep on or more differentiated progenitors switch on a developmental expression programme that is enabling them with the dangerous capacity to self-renew. A locus that emerges as central in the regulation of both self-renewal and proliferation in stem cells, progenitors and cancer is the Ink4a/Arf locus. The Ink4a/Arf locus encodes for two proteins, p16Ink4a and p19Arf, that are upstream of the two most important and well-known tumor suppressors, Rb and p53 respectively. In that respect, Ink4a/Arf is important in regulating cell cycle progression and loss of Ink4a/Arf function leads to tumor development. In contrast, Ink4a/Arf accumulation can lead to a special form of cell cycle arrest, termed senescence. Some have suggested that such accumulation also occurs in stem cells, partly to prevent them from oncogenic transformation, and that possibly stem cells senescence contributes to aging of the organism. In this thesis, we discussed a role for Inka4a/Arf under the control of Bmi1 and TBX2/TBX3 in stem cells, progenitors and cancer. TBX2 and TBX3 are members of the T box gene family of DNA binding transcription factors, which have both been found as negative regulators of the Ink4a/Arf locus thereby leading to immortilization of mouse embryonic fibroblasts. We established that TBX2/3 are direct regulators of Arf expression via new DNA-binding T site in the Arf promoter, that is a variant of the archetypical and consensus T site and is essential for TBX2/3 mediated regulation. Furthermore, we found that the variant T site is specific for TBX2/3, potentially explaining how in vivo T box genes confer specificity in regulating target genes. Our analysis contributes to the connection between T box transcription factors and tumorigenesis via Arf regulation. Another negative regulator of the Ink4a/Arf locus is Bmi1, a member of the Polycomb group gene family of epigenetic chromatin repressors, which have been implicated in cell fate decisions. Bmi1 deficient mice suffer from progressive neurological and hematopoietic defects that can all be traced back to an intrinsic defect in the self-renewal of stem cells in those compartments. We demonstrate that the Bmi1 deficient mice have significantly reduced cerebellar size resulting from hypoproliferation in the cerebellar granule neuron progenitors. Interestingly, we show that Bmi1 is a target of Shh, the morphogen that is driving proliferation of the CGNPs. Furthermore, we demonstrate Bmi1 overexpression in a substantial amount of primary medulloblastomas, tumors derived from aberrant Shh signalling in CGNPs. We proposed that multiple pathways downstream of Shh are regulating cerebellar granule neuron progenitor proliferation, one of which is via Bmi1. The morphogen Shh is known to signal, at least partially, via three Gli transcription factors. Corroborating Bmi1 as part of the Shh pathway, its expression is also induced by Gli transcription factors. Corroborating earlier studies in skin tumorigenesis, we demonstrate oncogenic transformation by Gli1 and Gli2, but not Gli3. We also show that overexpression of Gli1 and Gli2 induces L-myc expression and selects for N-myc but not c-Myc overexpression. As in vivo tumors with activated Shh signaling almost always show N-myc overexpression, our results suggest a special preference for tumorigenesis induced by the combination of N-myc/L-myc and Shh pathway activation. Finally, we demonstrated that in cerebellum Shh is regulating Ink4a/Arf through Bmi1. As Ink4a/Arf is coding for two genes, we determined the different requirements for Ink4a versus Arf regulation in multiple defects of Bmi1 deficient mice. We observed that Arf deregulation is affecting almost all defects of Bmi1 deficient mice. Furthermore, we suggest Ink4a deregulation is more profoundly affecting undifferentiated cells. This suggests differential self-renewal and cell cycle regulation in stem cells and progenitors versus differentiated cells under the control of PcG regulation. In conclusion, research in this thesis extends the knowledge about the mode and function of Ink4a/Arf regulation by Bmi1 and TBX2 in stem cells, progenitors and cancer. Understanding stem cells is of increasing importance, as some have suggested therapeutic potential in stem cells and benefits may be expected in the treatment of cancer by eradication of the cancer stem cells