The actin cytoskeleton is important in the machinery of motile and non-motile cells. Different cytoskeletal assemblies fulfill different roles, and their different functions are reflected in their different designs. Cytoskeletal design is mediated by the proportion, kind and organization of the actin-binding proteins present. We have been studying the organization and assembly of the cytoskeletons of specialized cells because their cytoskeletons are relatively simple in composition and structure. Several recent advances have opened a pathway to obtaining detailed models for in vivo cytoskeletal structures, in particular, actin bundles. The discovery that many of the proteins are modular in design permits us to dissect the cytoskeleton into parts. We propose in particular to study the fimbrin and fimbrin/villin bundles found respectively, in the specialized hair cells of the inner ear and in the brush border cells of the intestine. We propose to study rafts of actin crosslinked by EF1alpha, fascin, and aldolase. The results will tell us how actin is crossbridged by different bundling proteins and how the structure is deformed to accommodate them. We will also obtain maps of 2-D arrays of F-actin crosslinked by intact fimbrin or villin or both. We proposed to dock the atomic models for the components (all are known) into the maps obtained from the 2-D arrays and thereby derive a 3-D model for a row of crossbridged filaments. This model in turn will be used to construct a 3-D model of an in vivo 3-D bundle (e.g., the actin-villin bundle). In this way, we can begin the long task of constructing atomic models for the cytoskeleton. No one technique nor any one lab can accomplish all of this, but as evidenced in the recent atomic model for the acto-S1 I complex, it can be done by a combination of techniques and labs. We also propose to study the structural states of actomyosin. We plan to produce 3-D maps of actomyosin in the rigor, ADP and ATP states. We will use constructs of smooth muscle myosin and myosin V. We will also obtain maps of smooth muscle HMM in order to understand why double but not single headed myosin is regulated. We will also look at myosin V (actually a construct having 2 IQ domains) in the presence of calcium in order to visualize the changes associated with regulation.
Showing the most recent 10 out of 33 publications
