We are performing an overexpression/misexpression screen for novel genes involved in guidance, target recognition, and synaptogenesis by motor neurons. The rationale for performing such a screen is that loss-of-function mutations in genes encoding surface molecules involved in guidance and synaptogenesis (e.g., Connectin and Fasciclin III) often do not produce dramatic phenotypes due to genetic redundancy, whereas overexpression or misexpression of the same molecules can cause strong phenotypes that are easily detectable in a screen. The neuromuscular system of larvae consists of 30-40 motorneurons that innervate 30 muscles per hemisegment. Motor neurons contact their target muscles during embryogenesis, and synaptogenesis continues through the larval stages. To perform the screen, we use the EP element developed by Pernille Rorth, which contains multiple copies of the UAS site recognized by the transcription factor GAL4, and thus drives expression of adjacent genes only in cells that contain GAL4. We crossed lines carrying single EP insertions to C155-GAL4; UAS-GFP, which which will drive gene expression in all postmitotic neurons. The GAL4 also drives high-level GFP expression in neurons, and the GFP diffuses efficiently into axons and presynatpic terminals. We can thus examine every synapse in the neuromuscular system in live larvae under a GFP dissecting scope. We can also examine larval behavior, and we initially selected for examination those lines for which the larvae generated by crosses to the driver line appeared sluggish or immobile. We have recovered several known genes in our screen thus far, including Fasciclin II, Ptp10D, exuperantia, Dfurin2, C/EBP, fat facets, and genes encoding a 14-3-3 epsilon isoform, a serine-threonine protein kinase, and a vacuolar sorting proteinThe next step in the screen is to determine whether the identified genes are endogenously expressed in neurons (since misexpression of genes normally used for development of other organs might have effects on the nervous system but this would not be of interest for understanding normal neural development). A number of the genes we isolated pass this test. We will then make loss-of-function mutations in the genes for which no mutant alleles currently exist, and examine their phenotypes in embryos and larvae.