Pattern formation in the Drosophila visual system

The Drosophila eye is a useful model system in which to study cell-cell signaling, cell fate determination, and neuronal connectivity. It is a highly ordered structure consisting of 800 identical subunits. The eight photoreceptor cells in each subunit connect to defined synaptic partners in the brain, allowing the fly to detect objects, color and motion.

1. Signaling and cell fate determination in the Drosophila eye disc
Differentiation of cells within the eye primordium, the eye imaginal disc, is a progressive process that moves across the disc from posterior to anterior. The secreted protein Hedgehog (Hh) induces photoreceptor differentiation and also indirectly activates its own expression, driving propagation of a wave of differentiation across the eye disc. Hh induces only the first photoreceptor to form within each cluster, R8; R8 then produces Spitz (Spi), a ligand for the Epidermal growth factor receptor (EGFR) that recruits additional photoreceptors to the cluster. These two conserved pathways also regulate growth and differentiation in vertebrates, and their misregulation can cause tumorigenesis.

Using a genetic screen for mutations affecting photoreceptor differentiation, we have discovered several novel components of the Hh and EGFR pathways and elucidated their mechanisms of action. One example is Rasp, a transmembrane acyltransferase that adds an essential palmitate modification to both Hh and Spi. We showed that palmitoylation of Spi tethers it to the plasma membrane of producing cells, restricting its diffusion and thus increasing its local concentration. We are currently searching for additional secreted signaling proteins that carry lipid modifications. mago nashi mutants specifically fail to express MAP kinase (MAPK), a critical downstream component of the EGFR pathway. Mago is a subunit of the exon junction complex (EJC), which is deposited onto all spliced mRNAs and affects their fate in the cytoplasm. We have found a novel role for the EJC in splicing mapk and other genes with large introns that are located in heterochromatin. Our screen also identified the endosomal protein Myopic, which enhances the activity of the EGFR by promoting its progression through endocytosis, and Vps4, which acts at a later stage of endocytosis to downregulate the EGFR. Endocytosis promotes cleavage of the EGFR to release the intracellular domain; we are investigating whether this domain can enter the nucleus and regulate transcription.

Another set of genes found in our screen encode both specific and general transcription factors. The GATA factor Pannier sets the dorsal boundary of the eye field, separating it from the head field, while the coactivator Chip and LIM-homeodomain proteins set the ventral boundary. MED12 and MED13, two subunits of an accessory submodule of the mediator complex, act as adaptors to recruit the mediator complex to Wingless and Notch target genes. Similarly, we have found specific developmental roles for subunits that distinguish two forms of the Brahma chromatin remodeling complex. Our screen also identified several cytoskeletal proteins that regulate cell shape, cell migration, cell survival and cell signaling in the eye disc and other tissues. Additional mutations from the screen remain to be characterized.

2. Axon targeting in the visual system
In addition to the pattern of photoreceptor differentiation, we are studying synaptic partner selection by photoreceptor axons. We found that the protein tyrosine phosphatase LAR and its interacting protein Liprin-a are specifically required for the axon of the UV-sensitive photoreceptor R7 to reach its normal termination layer. LAR promotes R7 targeting by a novel mechanism that requires specific structural features but not phosphatase activity. RPTP regulation and signaling are poorly understood; we are searching for ligands and effectors for LAR in order to clarify these processes in a well-defined system. We have shown that Liprin-a acts downstream of LAR, and has both positive and negative interactions with two other members of the Liprin family. Although LAR and the three Liprins are also involved in neuromuscular synapse growth, their molecular functions appear to be different in this system. We are beginning a new project to develop genetic methods to inducibly damage axons, which will allow us to use Drosophila as a model system for the study of axonal regeneration.