Dscam2 is a cell surface protein expressed on neurons and required for neurodevelopment in Drosophila. It mediates boundary formation between different brain regions through homophilic repulsion and promotes contacts between neurons through homophilic adhesion. These functions occur at developmental times when axons and dendrites are actively growing and searching for synaptic partners. Interestingly, we found that a specific isoform of Dscam2 is expressed in larval motor neurons after growth has been completed and synaptic connections have been established.
Aims:To gain insight into post-developmental functions of Dscam2, we analysed Dscam2 mutant phenotypes at the larval neuromuscular junction.
Methods:We used a combination of neurogenetics, electrophysiology, light and electron microscopy (EM) to characterise phenotypes in Dscam2mutants.
Results:We found that Dscam2 suppresses the strength of motor neuron synapses. Evoked postsynaptic potentials were increased in Dscam2null animals compared to controls. We determined that loss of Dscam2 increased the number of synaptic vesicles released in response to an action potential. Mutants that suppress synaptic strength are rare so we screened through a number of candidates to identify signalling molecules in this pathway. We found a potential genetic interaction between the PI3K enhancer, Centaurinγ1A, and Dscam2. Consistent with this, Dscam2null motor neurons exhibited a striking increase in levels of the PI3K substrate, PIP2. PI3K regulates synaptic vesicle recycling, which in turn regulates synaptic release, so we next investigated whether Dscam2 might be involved in this process. A genetic marker for early endosomes was reduced in Dscam2 mutants and changes in endosomal structures were observed by EM. EM also revealed an increase in docked synaptic vesicles compared to controls, providing evidence that Dscam2 negatively regulates a necessary step in synaptic vesicle release.
Conclusion:These data argue that Dscam2 contributes to the maintenance of synapses, a strikingly different role than it plays during development. This function appears to be independent of homophilic interactions as Dscam2is not expressed in postsynaptic muscle. Our data indicate that Dscam2 interacts with the synaptic vesicle recycling machinery to prevent excess vesicles from tethering to synaptic release sites. This ‘brake’ on synaptic release may have important implications for neuronal plasticity.