During endocytosis, the cell's plasma membrane is deformed and internalized to bring in extracellular cargo and transmembrane receptors. Nonclathrin/noncaveolar (CLIC/GEEC) endocytosis internalizes glycosylated receptors and extracellular fluid, and is connected to cell polarity development, blebbing, and the epithelial-to-mesenchymal transition during cancer metastasis. A molecular and mechanical understanding of CLIC is necessary to understand how CLIC coordinates membrane curvature and actin polymerization to internalize the plasma membrane against membrane tension. The primary limitation in studying CLIC has been the lack of unambiguous markers for the process. A machine learning approach used in Dr. Akamatsu's postdoctoral lab to classify bona fide endocytic events will be adapted to disambiguate markers for CLIC endocytosis relative to other types of endocytosis. This will define the first unique markers for CLIC and will reveal the order of protein assembly during CLIC membrane internalization, which is essential for understanding the function of each protein. To test the hypothesis that membrane tension controls CLIC progression, Dr. Akamatsu will combine lattice light-sheet microscopy with a calibration method he developed during his postdoctoral research to convert fluorescence intensity to numbers of molecules in live cells. With this new method, molecule-counting lattice light-sheet microscopy, he will measure the numbers of molecules of CLIC endocytic proteins at both the apical and basolateral surfaces of polarized iPS cells, which differ in their membrane tension. He will image the cells under osmotic stress to increase cellular membrane tension. Finally, to understand the feedback relationship between membrane curvature-sensing BAR proteins and actin polymerization during CLIC membrane internalization, he will incorporate membrane tubulation by BAR proteins and their reciprocal interactions with actin filament nucleation proteins into a multi-scale mathematical model developed during his postdoctoral work. Simulations of this model will predict the critical feedback relationships between plasma membrane curvature, tension and actin polymerization necessary for the timely completion of CLIC endocytosis. Predictions from the model will be tested in his own lab by imaging cells endogenously expressing protein domain truncations in the presence of inhibitors of actin nucleation and polymerization. Dr. Akamatsu has a longstanding interest in combining physical modeling with live-cell quantitative experiments. One to two years of additional postdoctoral training will allow him to fully develop both skills in order to effectively implement a highly synergistic feedback loop in his own lab. Co-advising in experimental approaches by David Drubin and in computational modeling by Padmini Rangamani at UCSD have given him the foundation for this integrated approach. Additional training in theory from Padmini Rangamani and Hernan Garcia, and in quantitative experimental methods from Eric Betzig, Matt Welch, and Dan Fletcher will fully prepare him to lead an integrated modeling and experimental lab of his own.
Cells respond to and control the tension of their plasma membrane by internalizing parts of their membrane through endocytosis, which regulates the number of receptors on the membrane. When this process is dysregulated, cells tend to lose the molecular identity of their membranes and become metastatic cancer cells, but the physical basis for this transition is not well-understood. This proposed project will combine live cell three-dimensional fluorescence microscopy with mathematical modeling to understand, on a molecular and mechanical level, how which cells regulate their membrane tension by bending and internalizing their plasma membranes during endocytosis.
