The mechanisms by which cells move within a three-dimensional tissue are poorly understood. To gain some insights into this process, we use time- lapse 3D imaging to visualize and quantify the trajectories of cells moving within a cell aggregate. The imaging technique, namely computational optical-sectioning microscopy, permits observation of a living specimen at very low light levels and high spatial resolution in x,y and z over extended time periods. We have been applying these methods to the slime mold Dictyostelium discoideum, during the first multicellular stage of its development, when it forms a hemispherical mound of about one million cells. This organism offers many advantages for an analysis of motility and development, including a wide range of molecularly generated mutants. At the mound stage of development, cell movement is especially critical for the segregation of different cell types to particular regions within the structure. We have now visualized and quantified the movements of individual cells within the mound, and found a striking diversity of motile behaviors. This assortment of behaviors suggests that there could be a number of factors influencing cell motion within this tissue, including, but not limited to, both chemoattractant signals and differential adhesiveness of various cell types. To gain an understanding of the molecular mechanisms underlying the assortment of motile behaviors in the mound, we propose to continue an analysis of mutants in cytoskeletal, signal-transduction and adhesion proteins. For cytoskeletal proteins, we have already found that the myosin II heavy chain is essential for virtually all of the motile behaviors present in the mound. We now propose to examine motility in cells with various deletions and site-directed alterations in the myosin II molecule. These studies should help determine whether parts of the molecule are involved in particular motile behaviors. We also plan to screen a panel of cells defective in various myosin I molecules to look for any defects in their 3D motility. To address the role of signal transduction, we will examine cells lacking receptors for cAMP, as well as a strain lacking a G protein subunit (G- alpha4) potentially involved in sensitivity to folate. To investigate a role for adhesion, we will continue analysis of a strain lacking an adhesion molecule, gp80. This analysis will be augmented by development of several new techniques to visualize neighbor-neighbor relations among a cluster of cells in the mound. To accomplish this, we will track the movement of adjacent cells within a cluster by visualizing their nuclei, and we will also use photoactivation of a caged fluorescent probe to generate fluorescence in a cluster of cells, and then observe over time how the fluorescent cells disperse. Many of the analyses of mutant strains will be aided by the ability to visualize specific cell types using the Green Fluorescent Protein (GFP) and related analogs. The significance of these studies is two fold. First, we have already found several cases in which these 3D imaging methods uncovered significant defects in cell locomotion that had gone undetected by less stringent assays. Second, and more generally, our studies should help identify some of the key molecules and mechanisms involved in orchestrating morphogenesis in a simple developmental system.
