Millard Lab - Projects - School of Biomedical Sciences

We are always interested in recruiting enthusiastic scientists and welcome new ideas and techniques. Please contact Sean to enquire about Honours, Masters, PhD or Postdoctoral positions in the lab. Below is a list of ongoing projects.

Dscam2-related projects

The role of Dscam2 in learning and memory

In collaboration with Krystyna Keleman at QBI.

We are using a courtship conditioning assay to assess short-term memory in flies. Dscam2 mutants have a memory defect and we aim to figure out the role that it plays in this proces

**Dscam2 is required for short-term memory.** (A) Schematic of the courtship conditioning assay used in the lab. Naïve males learn that the previously mated female is non-receptive during the training session and they court a different mated female less in the testing session. (B) Control flies demonstrate learning in the testing session. (C) Dscam2 mutant flies have a learning defect.

Molecular mechanisms of Dscam2 repulsion

We have found that Dscam2 homophilic binding signals through the PI3K pathway to induce repulsion. We have an in vivo assay set up to measure this and we are using it to uncover the molecular players involved and how homophilic binding leads to their recruitment.

**Dissecting Dscam2 repulsion.** A Dscam2 pull-down experiment performed by Sarah Kerwin identified several proteins that interact with the Dscam2 cytoplasmic domain. Sarah setup an in vivo Dscam2 repulsion assay for validating these hits using da neurons. These neurons normally express Dscam2B, but it does not induce repulsion between them as can be observed from the tight longitudinal nerve bundles that form within neuromeres of the ventral nerve cord. Overexpressing Dscam2B in these neurons causes them to repel each other and increases the width of these bundles (Ds2^M). Knocking down some Dscam2-interacting proteins like Chickadee, suppresses repulsion whereas knocking down others like shaggy, increases it. Future work will try to determine how Dscam2 homophilic binding leads to the activation or inhibition of these downstream regulatory pathways.

How does Dscam2 inhibit synaptic strength at the larval NMJ?

Dscam2 suppresses synaptic strength in motor neurons and we are investigating the cellular mechanisms involved. Synaptic vesicle recycling appears to be critical.

**Dscam2 is trafficked to different intracellular compartments.** We have been studying how Dscam2 affects synaptic vesicle recycling using electrophysiology and electron microscopy. Different intracellular compartments accumulate in the Dscam2 mutant and PhD student Tarosi Senapati has been trying to identify them. We use an APEX-GBP labelling system developed by Rob Parton’s group to label specific intracellular structures. Shown in B is a GFP-tagged version of Dscam2. Labelling can be seen on the plasma membrane as well as on vesicles within this motor neuron terminal. We are trying to figure out how removing Dscam2 leads to an increased number of synaptic vesicles at active zones, resulting in an increase in synaptic strength.

Do different Dscam2 cytoplasmic isoforms have different functions?

Dscam2 has 5 alternatively spliced cytoplasmic isoforms that have very different protein compositions. We are investigating whether they have different functions at the larval NMJ.

**Alternative Dscam2 cytoplasmic domains may contribute to different Dscam2 functions.** Homophilic repulsion and the suppression of synaptic strength are two known functions of Dscam2, but the signalling mechanisms that lead to these different outcomes are uncharacterised. Five different alternatively spliced cytoplasmic domains have been identified at the RNA level. Each have unique protein sequences. Two of the cytoplasmic domains contain a PDZ binding domain, a protein motif commonly used to scaffold proteins at the synapse. Three of isoforms contain large stretches of low complexity prion-like domains that could participate in liquid liquid phase separation, as has been shown for other synaptic proteins. We are investigating how these domains affect Dscam2 function through rescue experiments and single molecule imaging.

ALS project

How the ALS risk factor, GGNBP2, contributes to autophagy

In collaboration with Naomi Wray’s group at IMB.

We have been studying how the gene, GGNBP2, contributes to ALS pathology using the fly. In humans, increased expression of this gene is correlated with increased risk of ALS. We have shown that it is required for synaptic development in fly motor neurons and that it is involved in autophagy, a process commonly defective in ALS.

**GGNBP2 regulates autophagy.** (A) Autophagy is a process that removes damaged organelles and aggregated proteins from cells by capturing them in vesicles and fusing with lysosomes. (I-L) Using a fluorescent version of an autophagic protein that gets degraded during this process (Ref(2)P in flies p62 in vertebrates) we observe that both loss of GGNBP2 function (K) and overexpression of GGNBP2 (L) causes defects in autophagic flux. (M) Quantification of the density of Ref(2)P puncta in motor neurons of different GGNBP2 genotypes.

Mitophagy project

Mechanisms of Fbxl4-mediated mitophagy in Drosophila

In collaboration with Julia Pagan’s lab at SBMS.

Mitophagy is the selective degradation of damaged mitochondria using autophagic machinery. The Pagan lab has identified a ubiquitin ligase (FbxL4) that limits mitophagy by degrading receptors on mitochondria required for this process. We are trying to understand the mechanisms involved using Drosophila.