Preventing the loss of competence for neural induction: HGF/SF, L5 and Sox-2

Abstract

The response to neural induction depends on the presence of inducing signals and on the state of competence of the responding tissue. The epiblast of the chick embryo loses its ability to respond to neural induction by the organizer (Hensen’s node) between stages 4 and 4+. We find that the pattern of expression of the L5220 antigen closely mirrors the changes in competence of the epiblast in time and in space. For the first time, we describe an experiment that can extend the period of neural competence: when L5220 expression is maintained beyond its normal time by implanting HGF/SF secreting cells, the competence to respond to Hensen’s node grafts is retained. The host epiblast forms a non-regionalized neural tube, which expresses the pan-neural marker SOX-2 (a Sry-related transcription factor) but not any region-specific markers for the forebrain, hindbrain or spinal cord. Although HGF/SF secreting cells can mimic signals from Hensen’s node that maintain L5 expression, they cannot rescue the ability of the node to induce anterior structures (which is normally lost after stage 4). The ectoderm may acquire stable neural characteristics during neural induction by going through a hierarchy of states: competence, neuralization and regionalization. Our findings allow us to start to define these different states at a molecular level, and show that the competence to respond to neural induction is not entirely autonomous to the responding cells, but can be regulated by extracellular signalling molecules.