The cortical cytoskeleton of animal cells not only provides cells with their shape, but also is involved in such fundamental processes as cell motility, cytokinesis, phagocytosis, endocytosis, secretion and transmembrane signaling. The ERM (ezrin-radixin-moesin) family of closely related cortical proteins are believed to play an important role in linking the plasma membrane to the cytoskeleton. Currently, most is known about ezrin, which is the focus of this proposal. Ezrin was originally isolated as a component of the intestinal microvillus cytoskeleton. It is a substrate for certain protein tyrosine kinases, and a correlation between ezrin phosphorylation and changes in cell surface structures has been found in several systems. Ezrin is expressed at high levels in polarized epithelial cells where it specifically associates with the microvillar plasma membrane on the apical aspect of the cell. We have made the surprising finding that purified ezrin exists as stable monomers and dimers in solution, suggesting that two functionally different pools of ezrin exist in vivo. Two sites in ezrin have been implicated in the formation of the dimer, one of which appears to be masked in the monomer. We propose to map these sites in the protein sequence and elucidate the structural differences between monomers and dimers. Since different cell types have different monomer/dimer ratios and this may be correlated with the number of cell surface structures, we propose to determine whether the monomer/dimer ratio changes as cells are induced to form more surface structures and whether they are functionally segregated. A search will also be made for an activity that converts monomers to dimers. To understand the function of ezrin, cells will be microinjected with defined fragments of ezrin that might interfere with its function in vivo. A number of approaches will be used to identify cytoskeletal and membrane proteins that interact with monomeric or dimeric ezrin. These include binding to F-actin, biochemical dissection of placental microvilli, which are rich in ezrin, and affinity chromatography on monomeric, dimeric and defined domains of ezrin. To begin to explore the functional differences between members of the ERM family, a search for moesin-binding proteins will be undertaken. This will employ a moesin-containing cell line lacking ezrin, and also lung tissue, which is a rich source of moesin. For a complete understanding of the ERM family, it will be necessary to determine the three-dimensional structure of ezrin. This project will be initiated by crystallizing the putative N-terminal membrane-binding domain for structural analysis by X-ray crystallography. Since the members of the ERM family are very closely related, the results with ezrin should apply to other members of the family. The results should also extend to other members of the band 4.1 superfamily which includes the product of the neurofibromatosis 2 tumor suppressor gene, merlin, protein tyrosine phosphatases and the membrane-cytoskeletal linking protein talin.
Showing the most recent 10 out of 60 publications
