Proceedings of The Physiological Society

King's College London (2011) Proc Physiol Soc 22, PC26

Poster Communications

Defining the spinal cord connectome: axon projection and dendritic fields in the developing Xenopus spinal cord

D. Conte1, S. R. Soffe1, R. Borisyuk2, A. Roberts1

1. Biological Sciences, University of Bristol, Bristol, United Kingdom. 2. Centre for Theoretical and Computational Neuroscience, University of Plymouth, Plymouth, United Kingdom.


How nervous systems develop with the remarkable specificity required for complex functions is a question that developmental neuroscientists have been chasing for many years. Are sophisticated recognition mechanisms required during early stages of vertebrate development, when simpler circuits assemble in the axial nervous system to generate first movements? In hatchling Xenopus laevis tadpoles, previous work has elucidated morphology, physiology and synaptic connections of neuron subclasses forming the spinal motor circuit that generates swimming. This puts us in a position to ask how these circuits self-assemble. Unexpectedly, paired neuron recordings showed at least some synapses between all neuron types raising the possibility of low specificity and that axons potentially form synapses with any dendrite they contact (Li et al 2007). Are connections in early circuits determined primarily by geographical location of axons and dendrites of different neuron types? To record 3D information on location of axons and dendrites in the tadpole CNS, we devised a measuring microscope. We obtain data on morphology of individual neurobiotin-filled neurons from electophysiological experiments where tadpoles were anaesthetised with 0.1% MS222 and then immobilised with alpha bungarotoxin. We define their relative position in 3D space in the nervous system. Neurons were viewed in whole mounts of the tadpole CNS under a Nikon Optiphot microscope and DeltaPix camera. Specimen position was controlled by a Scientifica PatchStar micromanipulator and LinLab software which allowed 3D co-ordinates describing neuronal morphology and relative location to be recorded. Excitatory descending interneurons (dINs) drive other swim circuit neurons during swimming (Dale & Roberts, 1985, Soffe et al 2009). Paired whole-cell recordings have shown that they synapse with each other (Li et al 2006). 3D co-ordinates of dIN axon and dendrite trajectories were recorded within the spinal cord and hindbrain, which is ~4mm rostro-caudally and ~100µm dorso-ventrally. We defined the rostral edge of the hindbrain, the midbrain border (MBB), as x=0 and the ventral edge of the spinal cord as y=0. Preliminary assessment of a dIN sample (n=17) whose somata ranged from 830-2278µm from MBB showed axon trajectories extending to 3340µm from MBB. The sample axon dorso-ventral occupation ranged from 1-71µm from the ventral edge; closely matched by dendritic arborisation dorso-ventral occupation, which ranged from 1-65µm from the ventral edge. These initial data show that the geography of dIN axons would allow contact with dIN dendrites to form synaptic connections. More detailed inspection may reveal possible relationships within dINs that may influence connectivity, e.g. do more rostral dINs have more ventral axons? Data on dINs provides an example of the method which we hope will predict the global network connectivity of the identified and recorded neuron types within the spinal cord.

Where applicable, experiments conform with Society ethical requirements