Cell shape and polarity during neurogenesis and brain morphogenesis

We believe that understanding brain development will depend on uncovering how individual cells contribute to brain morphogenesis and how the subcellular organisation of cells drives their behaviours. For these reasons we use the transparency and accessibility of the living zebrafish embryo brain together with advanced in vivo microscopy to reveal cell and sub-cellular behaviours. We recently developed an optogenetic approach that uses far red laser light to manipulate sub-cellular protein localisation with high spatial and temporal precision in the intact embryo. This should allow more precise dissection of cell polarity and cellular asymmetries in vivo.

We have previously shown that two distinct behaviours can be responsible for the cellular rearrangements required at the neural midline for lumen formation and neuroepithelium production. One is the specialised mirror-symmetric division that generates daughters with mirror-image apicobasal polarity and the second is the resolution of cell interdigitation by retraction (i.e. de-interdigitation) that occurs at the midline if the mirror-symmetric division is experimentally inhibited. Our data suggests both of these processes will remodel cadherin-based adhesions from a relatively coarse distribution of adhesions across the tissue midline to the more precisely midline distribution that is required for adherens junction formation at the nascent lumen surfaces. In addition to refining this general distribution of adhesions, the cells also have to do something perhaps more challenging and that is specifically lose adhesions between contralateral neighbours while retaining and strengthening adhesions between ipsilateral neighbours. Without the specific loss of contralateral adhesions the lumen will not be able to form. In addition to dynamic molecular and physical interactions between neuroepithelial cells, critical signals that regulate brain morphogenesis and polarity will also come from the surrounding extracellular matrix and mesoderm. How cell polarity is generated through the integration of these multiple intrinsic and extrinsic signals will be fundamental to establishing the inside-out axis of the brain and relevant to the morphogenesis of many other body organs.

The emergence of neuronal morphology in vivo is poorly understood, yet neuron polarity and dendritic and axonal morphology are likely to be key determinants of neuronal connectivity. A lot of data is available on the process of neuronal differentiation in vitro, but very little addresses this process in vivo. In vitro experiments suggest neurons transition through a multipolar phase and one of the multiple processes is selected to mature as the axon. Time-lapse analysis of new-neurons in cultured slices of mammalian cortex also indicate that a multipolar phase precedes radial migration to the appropriate cortical layer. Whether this transient multipolar phase is a behaviour conserved in other regions of the neural tube is uncertain and how much of the emerging morphology of the dendritic and axonal projections are determined cell-intrinsically rather than through extrinsic signals or cell-cell interactions is also unknown. We also wish to understand how neuronal differentiation is coordinated in space and time to achieve the choreographed temporal and spatial distribution of multiple neuron subtypes along the spinal cord.

We will extend our observations on neuronal differentiation beyond the initiation of axogenesis to encompass the emergence of dendritic arbors. We will use neuron type specific drivers to visualise dendrite development. The fine details of dendritic arborisation are likely to be regulated by activity but regulation of their initial growth and shape is less well understood for most neurons. Different spinal neuron subtypes have dendrites that occupy distinct dorsoventral domains that will have a significant influence on the neuron’s connectivity and potential to participate in functional circuits. A recent analysis of the early connections between spinal neurons suggests early connectivity may be driven solely by cell morphologies rather than having to evoke molecular specificities. How the specific spatial domain of dendrites is achieved is not certain, but work on dendritic domains in fly CNS suggest they may be regulated by midline signalling. We will quantify the spatial domains of spinal neurons, and then manipulate the neurons’ ability to read potential neurite guidance cues to test their role in dendritic arborisation. If we can understand the mechanisms that control the spatial distribution of dendrites we should be able to manipulate this and test the idea that anatomical proximity is a significant regulator of connectivity.

Some parts of the vertebrate nervous system are composed of dramatically different types of epithelia. For example the dorsal aspect of the rhombencephalon generates a very thin simple squamous epithelium that forms the expanded roof of the 4th ventricle and contributes to choroid plexus formation. How the growth of this pavement-like epithelium is integrated with the growth of the adjacent pseudostratified epithelium of the rhombic lip is not known, but this will be critical to brain morphology as well as choroid plexus development and therefore CSF production. Together with our colleagues in Richard Wingate’s lab, we are investigating the role of a novel neural cell type that sits at the interface of the rhombic lip and the epithelial roof of the 4th ventricle. We hypothesise these cells respond to biophysical forces from neighbouring tissues to form the stem zone that generates the rhombencephalic roof, as well as being a morphological corner-stone to integrate the two diverse epithelial morphologies.