Structure and function of scabrous, a secreted glycoprotein in Drosophila neurogenesis

Abstract

Cell-cell interactions mediated by Notch signaling are implicated in a wide variety of cell fate decisions during development of invertebrates and vertebrates. In Drosophila, the Notch (N) gene encodes a surface receptor with two known ligands, Delta (Dl) and Serrate (Ser). It has been suggested that the scabrous (sca) gene might be a another Notch ligand acting in the peripheral nervous system because the sca gene sequence predicts a secreted protein and sca interacts genetically with N and Dl. Alternatively Sca could bind to other receptors and work parallel to Notch signaling.;Using expression in Drosophila Schnieder cells, I showed that sca encodes a secreted, homodimeric glycoprotein and this protein is similar to protein detected in embryos. To determine whether sca encodes a N ligand, the secreted Sca was purified and a cell line expressing 18,000 N molecules per cell surface was used to test sca binding by cell adhesion assays and cell binding with labeled Sca. No interaction was detected between Sca and N or Dl, suggesting that Sca is not a high affinity Notch ligand.;The sca gene sequence predicts two domains, one being amino-terminal coiled-coil domain and the other the carboxy-terminal fibrinogen-related domain (FReD). To identify the receptor-binding domain, and help clone the Sca receptor in the future, these two domains were respectively expressed in flies using the GAL4/UAS expression system. The coiled-coil domain rescued the sca mutant extrabristle phenotype, whereas the FReD enhanced it, suggesting that both domains are potentially involved in protein-protein interactions and the coiled-coil domain contains a potential receptor-binding site.;N signaling is required to organize the dorsal-ventral axis in the developing wing margin. Ectopic expression of full length or coiled-coil domain of Sca in decapentaplegic-expressing region leads to notches of the wing margin. I provide evidence that ectopic expression of Sca blocks expression of N-downstream genes including vestigial, wingless, cut, and Enhancer of split-m 8. Moreover, ectopic Sca suppresses ectopic Cut expression induced by Dl, and affects the ectopic Cut expression pattern induced by Ser. This effect caused by Sca is through downregulation of N activity, because loss of N activity causes similar results and increasing N activity suppresses Sca-induced wing notches. I propose that ectopic Sca antagonizes N signaling through downregulation of N activity to lead to a wing-notch phenotype.