Mouse embryonic stem cells as a discovery tool in neurobiology

Mouse ES cells can recapitulate, under suitable tissue culture conditions, early events in neurogenesis. As wildtype or genetically modified ES cells can be grown in unlimited quantities, their differentiation into neurons represents an attractive model for studying the function of genes involved in early development, such as those controlling neuronal specification and survival. A few years ago, our laboratory established a robust differentiation protocol leading to the generation of well-defined and virtually pure populations of Pax6-positive radial glial (RG) cells with a profile and developmental potential characteristic of Pax6-positive RG cells of the cortex. Like their in vivo counterparts, these progenitors generate homogeneous populations of glutamatergic neurons. In my thesis work, I first addressed the role of Pax6 in the generation, specification and developmental potential of RG cells, by analyzing the progeny of ES cells isolated from homozygote Pax6-mutant embryos. I found that while Pax6 is not required for the generation of neurogenic RG, it is both sufficient and necessary for specifying them into a glutamatergic lineage. RG cells lacking Pax6 express genes specifying an interneuron fate, like Mash1, and generate GABAergic inhibitory neurons. These cells die prematurely due to an aberrant over-expression of the neurotrophin receptor p75. I could verify these findings in the cortex of mutant embryos lacking Pax6. This work led to new insights as to the regulation of neuronal specification and survival during neurogenesis.
In the second part of my thesis, I used this ES cell-based differentiation system to test any potential instructive roles of the 3 neurotrophin tyrosine kinase receptors TrkA, TrkB and TrkC, after recombining them into the neuron-specific mapt locus. This approach led to the surprising observations that TrkA and TrkC cause neuronal death when not activated by their neurotrophin ligands, whereas TrkB does not. Both the death inducing activity of TrkA and TrkC and the lack of death-inducing activity of TrkB were explained by differential distribution of these receptors with p75. The TrkA and TrkC-induced death involves their segregation together with p75 in lipid rafts, and the subsequent proteolysis of the latter signals to the apoptotic machinery. By contrast, TrkB is not recruited to lipid rafts and it does not result in p75 proteolysis. Subsequent analyses of TrkA and ngf mutants, as well as of embryos lacking both TrkA and p75 receptors confirmed the relevance of this novel death triggering mechanism during the development of the peripheral nervous system. These findings also point to a major, and so far un-described, difference in the way growth factors regulate the survival of neurons in the developing peripheral versus central nervous system. It is the receptors themselves that cause neurons to become growth factor dependent in the peripheral, but not in the central nervous system.
Taken together, my results demonstrate that the differentiation of mouse embryonic stem cells into defined neuronal populations represents a useful tool allowing observations to be made that are relevant to the development of the nervous system.