Fate-mapping the mouse neural tube by Cre-loxP transgenesis
The mammalian central nervous system (CNS) develops from a thin layer of neuroepithelial stem cells that form the early neural tube. Initially neuroepithelial cells generate neurons, the identity of which depends on their position of origin along the embryonic axis. Glia are generated at the end of neuronogenesis and very little is known about their specification. One of the challenges of modern developmental biology is to understand the extent of the diversity of glial cell types found in the CNS and how this diversity is generated. In this thesis I describe work using Cre-loxP technology combined with PAC transgenesis that clarifies the origins of various glial cells in the CNS. CNS glial cell generation was studied by fate-mapping restricted pools of neural precursors and identifying their glial progeny. Using PAC transgenesis technology I generated mice expressing Cre recombinase in the same patterns as the transcription factors DbxJ, Emxl and MsxS. The resulting transgenic mice were crossed with Cre-dependant reporter mouse strains to visualise the progeny of the various neuroepithelial regions. Using these techniques I was able to demonstrate that the all the neuroepithelial domains studied produce neurons as well as the major glial cell types astrocytes and oligodendrocytes. Of particular interest was the demonstration of novel sources of oligodendrocytes in the dorsal spinal cord and cortex. This work helps to resolve much of the controversy surrounding the origins of cortical and spinal cord oligodendrocytes. Investigation of the mechanism of dorsal oligodendrocyte specification indicates that dorsally-derived oligodendrocytes may be specified by Sonic hedgehog (SHH) -independent pathways, unlike their ventrally derived counterparts. The implications of these results are discussed in the context of possible mechanisms of glial cell specification in the CNS.