Anterior-posterior patterning of the avian neuraxis
Nieuwkoop and Nigtevecht's (1954) model of anterior-posterior patterning of the neuraxis states that the naive ectoderm first receives an 'activation' signal giving it an anterior, neural character and subsequently a 'transformation' signal that progressively caudalises the axis resulting in the full rostro-caudal pattern. Stern (2001) proposed a modification to this model that divides the first step into two: a transient 'activation' step and a subsequent 'stabilisation' step. The hypoblast has been implicated as the transient inducer of a pre-neural, pre-forebrain state. In this thesis, the molecular nature of this induction is investigated by grafting the hypoblast into the area opaca to look at the induction of the genes Sox3, Otx2, ERNI and Cyp26Al. It was found that FGFs recapitulate the induction of the first three markers and retinoic acid (RA) can induce Cyp26Al whilst loss of function experiments show that both FGF and RA are required for hypoblast- mediated induction. In the epiblast, these induced genes are maintained as the future forebrain develops. Potential stabilising signals were tested by combining hypoblast grafts with cells secreting various proteins. By antagonising Wnts and/or BMPs and/or Nodal, Sox3 and ERNI can be maintained, whilst Otx2 maintenance requires combined Wnt- and BMP-inhibition, but the definitive neural marker, Sox2, is never induced. This suggests that a further 'neuralising' step might be required. Unlike regions of the epiblast fated to form head structures, the cells that will contribute to the remainder of the neuraxis reside within a small population of progenitors near the node. This indicates that a different mechanism might be responsible for patterning more caudal regions mediated by a qualitative or quantitative mechanism. To test this, secondary axes were generated by grafting progressively older donor nodes but the patterning of these ectopic axes suggests that the node might caudalise in conjunction with the pre-somitic mesoderm (PSM). Indeed, homotopic PSM grafts between different staged embryos do affect the neural tube boundary of Hoxb9. PSM can caudalise the neurectoderm, an effect that increases with age of the donor and decreases with the age of the PSM cells. An interesting conclusion is that some of the same signals are responsible both for the initial activation stages and for the subsequent transformation steps. This highlights the importance of timing as to the response of a particular cell to a particular signal.