The formation of a healthy nervous system involves the regulated production of neurons during the course of embryonic development. We have used a combination of genetic, cellular, molecular and computational approaches to identify mechanisms that divide the nervous system into functional compartments like the forebrain, midbrain, hindbrain and spinal cord. We also study how neurons are made in the correct number and location within each of these compartments. Differential exposure to signaling factors secreted from the caudal margin of the neural plate helps divide the neural plate into discrete compartments along the anterior-posterior axis. Within these compartments the expression of proneural genes defines neurogenic domains where cells acquire the potential to become neurons. Within each neurogenic domain cells compete to become neurons by inhibiting their neighbors from adopting a similar fate, a process known as lateral inhibition, which is mediated by Notch signaling. Neural induction leads to the specification of neural tissue that initially has a forebrain-like identity. Exposure to posteriorizing factors like FGFs, Wnts and Retinoids subsequently leads to specification of progressively posterior compartments like the midbrain, hindbrain and spinal cord. Our investigation of the zebrafish headless (hdl) mutant revealed the essential role of this gene in the determination of anterior parts of the nervous system like the forebrain and eyes. Headless encodes a homologue of T-Cell Factor?3 (TCF3), a transcriptional factor that has an essential role as a repressor in inhibiting a response to posteriorizing factors. In the absence of this repressor, forebrain tissue is lost because posteriorizing factors force the prospective forebrain tissue to adopt a midbrain or hindbrain-like fate. There is a progressive loss of anterior brain structures and expansion of posterior brain structures when function of headless and additional TCF3 homologues is lost. Our observations provided support for a theoretical model that suggests a gradient of posteriorizing activity is interpreted to define discrete compartments of the neural plate that eventually form forebrain, midbrain, hindbrain and spinal cord. In the past year we have used computer simulations to better understand some specific changes in the shape of gene expression domains that are observed in mutant zebrafish with a progressive loss of TCF3 function. The simulations allowed us to visualize how posteriorizing signals in the blastoderm margin could, in principle, account for the initial shape of early compartments in the prospective neural plate. The simulations also showed how signals from the prechordal plate, which lies under the neural plate and is the source of Wnt antagonists, could modify the shape and size of gene expression domains established earlier in gastrulation. The strengths and weaknesses of these models was determined by comparing their predictions with changes in the shape of gene expression domains in posteriorized mutants and in a mutant that lacks the prechordal plate. The expression of zebrafish ?proneural? achaete-scute and atonal homologues (see Park et al 2003) helps define ?neurogenic? domains in the neural plate where cells acquire the potential to become neurons. However, the mechanisms that prevent neurogenesis in specific parts of the neural plate are poorly defined. Previous studies in Xenopus had suggested a key role for zic2 in defining non-neurogenic domains. We examined the role of zic2.1, zic2.2 and zic3 in patterning early neurogenesis. Stable anti-sense oligo nucleotides, called morpholinos, directed against individual or combinations of these zics, were injected at the one cell stage to inhibit translation of these genes. Our analysis revealed partially redundant but distinct roles for these genes in different parts of the nervous system. In the trigeminal ganglion losss of zic2 and zic3 function was associated with an increase in the apparent size of the ganglion and the number of differentiating neurons. At the anterior edge of the neuroectoderm there was an increase in the expression of some neurogenic genes. However, there was no increase in the number of differentiating neurons, suggesting that additional factors might be capable of delaying differentiation of neurons in this potentially neurogenic domain. In contrast to the inhibitory role for zic2.1, zic2.2 and zic3 at the boundary of the anterior neural plate, knock-down these genes led to a loss of neurogenesis in the hindbrain. Together these studies show that zic2 homologues and zic3 have a complex role: They influence the pattern of neurogenesis by inhibiting neurogenesis in trigeminal cells adjacent to the midbrain-hindbrain boundary (MHB) domain and the anterior neural plate, while promoting neurogenesis within the CNS in the prospective hindbrain of the zebrafish embryo. Lateral inhibition mediated by Notch signaling leads to the selection of cells that are permitted to differentiate in the developing neural tube. By inhibiting neuronal determination Notch signaling helps generate diversity of neurons and ensures that a large enough pool of progenitors persists to contribute to the growth of the nervous system as different types of neurons and glia are added to the nervous system. In the absence of Notch signaling there is a clear increase in ?Primary? neurons, whose fate is determined very early, and a reduction in some ?secondary? neurons whose fate is determined relatively late. However, secondary neurons are a diverse population generated at different times and loss of Notch signaling reduces the number of some late neurons while increasing others. To explore the role of Notch activation in different spinal cord lineages we created a zebrafish transgenic line that would allow us to examine the dynamic pattern of Notch activation in live embryos. Notch signaling leads to the activation of Hairy Enhancer of split related genes that are referred to as HER genes in zebrafish or HES genes in other vertebrates. Previous studies in zebrafish have shown that her4 is induced by Notch in the nervous system. We have taken advantage of this observation and used regulatory elements included in a genomic fragment 3.4 kilobases (Kb) upstream of the her4 gene to create a transgenic zebrafish line where this genomic fragment drives expression of GFP or destabilized monomeric RFP. We have shown that it drives expression of GFP transcripts in a pattern that is similar to endogenous her4 in the CNS, however, it lacks elements required to suppress expression in the rostral neural tube. Analysis of endogenous her4 expression and expression of fluorescent proteins driven by the 3.4 Kb genomic fragment shows that while early activation of her4 in the CNS is dependent on Notch 1a and Notch5, by one day its expression becomes dependent on an additional homologue, Notch1b. We are now using the transgenic line to examine the role of Notch signaling in individual spinal cord lineages.
Showing the most recent 10 out of 17 publications
