| During development, tissues undergo precise and controlled changes in shape and size. Various signaling pathways regulate these changes, temporally and spatially, by altering cytoskeleton dynamics to alter cell shape. One such pathway involves the Shroom and Rock proteins that reorganize the actomyosin cytoskeleton in epithelial cells to alter tissue morphology. The Shroom family of proteins are multi-domain, actin-binding proteins required for many developmental processes such as neural tube formation, and retinal morphogenesis. Shroom proteins interact with Rho-kinase (Rock), another conserved cytoskeleton regulator, to activate non-muscle Myosin II and assemble a contractile actomyosin network. All Shroom proteins contain a highly conserved C-terminal domain called Shroom Domain 2 (SD2) that interacts with the Rock Shroom binding domain (SBD) and is required for Shroom-mediated apical constriction. In the Shroom-Rock system it is unclear how this interaction activates the kinase activity of Rock. The goal of this dissertation is to understand the mechanistic details of the Shroom-Rock interaction. Using structural studies I have started to dissect the SD2-SBD interaction. I first determined the crystal structure of the Drosophila Shroom SD2 domain at 2.7 A resolution to be a novel fold composed of a three-segmented, anti-parallel, coiled-coil dimer. Using mutational analysis and a combination of in vivo and in vitro assays we identified surfaces within the central coiled-coil segment of the SD2 domain that mediate Rock binding. The anti-parallel nature of the SD2 domain introduces internal symmetry into the SD2 domain such that there are two identical binding sites for Rock on opposite sides of the molecules suggesting interesting implications for the Shroom-Rock interaction. We also determined the crystal structure of the Rock SBD to 2.5 A resolution and saw that it is a parallel coiled-coil dimer. Using mutational analysis combined with biochemical assays I have identified two conserved patches on opposite ends of Rock SBD that are required for Shroom interaction. These patches, unlike the patches on the SD2 domain, are not identical. Biochemical characterization of the SD2-SBD complex suggests that molar ratio of this complex is 1:1. Based on these results we can start to suggest models for how Shroom and Rock interact. v. |
