During normal embryonic development, wound healing and cancer metastasis, cells move collectively, making and breaking cell-cell contacts as they squeeze between other cells and tissues. Individual cells can leave an epithelium by undergoing the epithelial to mesenchymal transition. However, the program(s) enabling cells to escape as a group are less clear. Even less studied are the mechanisms by which groups of cells establish new connections at their destination. We established an in vivo model, the border cells in the Drosophila ovary, to study collective cell movements. We deploy the powerful Drosophila genetics toolkit and organ culture methods we developed to carry out high-resolution live imaging and optogenetics. During the last funding period, we investigated mechanisms by which border cells separate from the follicular epithelium in the process of delamination and how they make new connections when they reach the oocyte, a process we call neolamination. We identified multiple steps of each process and proteins that control them. In Aim 1, we will address key open questions in the septin field and build on our discoveries that the septin cytoskeleton is essential for border cell delamination and that the GTPase Rho recruits septins to the cortex. We propose to use novel in vivo analysis of cell and cluster geometry that we developed to compare and classify Rho, myosin, and septin phenotypes to clarify their roles and relationships in delamination. We will identify the GEFs and GAPs that regulate Rho upstream of septins. In Aim 2, we will follow up on our discovery that the nucleus plays an essential role in collective delamination. As border cells delaminate, they squeeze into tiny spaces much smaller than even a single nucleus. The nucleus can impede cell movement through rigid plastic pores, but we found that the nucleus plays a role in promoting border cell delamination. We propose a new “nuclear wedge” model in which nuclei push substrate cells apart. In Aim 2, we will decipher the mechanisms that move nuclei within the lead cell to promote its wedge function. We will test the function of the LIS1/NudE/Dynein pathway, which moves nuclei on microtubules (MTs) during the migration of neural progenitors in the developing neocortex. We will also study myosin, which transiently accumulates near sites of nuclear deformation and movement. In Aim 3, we will pursue our discovery that innexin (Inx) gap junction proteins promote both delamination and neolamination. Inxs function in a MT-dependent but channel-independent manner. We propose to probe the unexplored relationship between Inxs and MTs. We recently overcame a major technical hurdle and can now image MT dynamics in vivo, which we will use to characterize innexin phenotypes. We will decipher the relationship between MT post translational modifications and dynamics in vivo. We will also test our integrative working model for how nuclear, and cortical cytoskeletal systems interact to acco...