Dorsal-ventral patterning and the control of neural cell fate in the vertebrate ventral neural tube

During embryonic development organised structures are formed from homogeneous groups of undifferentiated cells. The specification of distinct cell types in these tissues is believed to be controlled by molecular cues that determine a cell’s fate according to the position of the cell within the tissue. In ventral regions of the vertebrate spinal cord the secreted protein Sonic Hedgehog (Shh) has been characterised as one such molecular cue. Shh appears to act at distance from its source, in a concentration dependent manner, to control the dorsal-ventral position in which distinct neuronal subtypes are generated. In this study I focus on the mechanisms of developmental patterning by graded Hedgehog (Hh) signalling that control the patterning of the spinal cord. Characterisation of the expression pattern of a series of molecular markers indicates that dorsal ventral patterning of the zebrafish spinal cord is similar to the mouse and the chick. Furthermore, use of the selective Hh signalling antagonist, cyclopamine, indicated that in zebrafish expression profiles of these molecular markers is dependent on Hh signalling. Using this approach, we provide evidence that ventral neural patterning depends on both the strength and duration of Hh signal in vivo. These data provide in vivo support for the idea that a gradient of Hh signalling is responsible for providing positional information to the ventral neural tube. To further understand how graded Hh signalling is interpreted by cells in the ventral neural tube we have analysed the spinal cord and hindbrain of mouse embryos lacking two genes, Nkx2.2 and Pax6, important for ventral vertebrate neural patterning. Previous analysis of each single mutant (Nkx2.2-/- and Pax6-/-) suggested a model in which Nkx2.2 is required for the generation of the most ventral interneuron type in the spinal cord, V3 neurons, while Pax6 is required to limit the expression of Nkx2.2. Analysis of mutant embryos lacking both Nkx2.2 and Pax6 has led us to modify this model. Our data indicate that Nkx2.2 is not directly required for V3 neuron generation but instead is required to repress Olig2 or similarly expressed gene. Moreover, our analysis revealed that Olig2 is regulated differently in the hindbrain and spinal cord and these data are consistent with the idea that Nkx2.2 is required for the generation of hindbrain visceral motor neurons. Finally, analysis of Ngn3 mutant mice indicated that Nkx2.2 acts upstream of Ngn3 and that Ngn3 is not required for V3 neuron generation