New insights into the regulation of growth cone motility by cell adhesion molecules (CAMs) come from a study published in the Journal of Cell Biology. Nakai and Kamiguchi show that certain CAMs are localized to specialized microdomains within the cell membrane, known as lipid rafts, and that these rafts are important for growth cone migration.

Lipid rafts are regions of membrane into which cholesterol and sphingolipids are preferentially packaged. Specific proteins can attach to these areas, allowing regional concentration of signalling molecules. By separating out the lipid rafts, or detergent-resistant membranes (DRMs), from mouse cerebellar neurons, the authors found that two specific CAMS — L1 and N-cadherin (N-cad) — were expressed in DRMs (as well as elsewhere in the membrane). Another CAM, β1 integrin, was found only in non-DRM fractions of cell membrane.

Is this localization of L1 and N-cad to DRMs essential for their function in mediating growth cone motility? Nakai and Kamiguchi used various methods of disrupting DRMs in living neurons to show that such disruption prevented growth cones from moving on a substrate of L1 or N-cad (both CAMs show homophilic binding, so that growth on an L1 substrate, for example, is mediated by L1 in the growth cone). Growth cone motility on a laminin substrate, to which β1 integrin binds, was unaffected by DRM disruption.

To look at regional requirements for DRMs within the growth cone, the authors used a technique known as micro-scale chromophore-assisted laser inactivation (micro-CALI). A target molecule, in this case GM1 gangliosides (a marker for lipid rafts), is labelled with a dye-conjugated ligand. When a part of the cell is then irradiated with a laser, the labelled molecules and their immediate neighbours are selectively perturbed. Disruption of DRMs in growth cones by micro-CALI impaired growth cone motility mediated by L1 or N-cad, but not by β1 integrin. By irradiating just part of the growth cone, Nakai and Kamiguchi showed that DRMs were needed in the periphery of the growth cone, but not the central domain, for growth cone motility.

These results fit well with previous studies showing that these two domains within the growth cone probably have distinct functions in growth cone guidance, with the peripheral domain generating instructive signals and the central domain being responsible for more permissive signals. This elegant study constitutes one of the best proofs of a functional role for lipid rafts in neurons.