As cells move, internal movements caused by their actin and microtubule cytoskeletons must be coordinated with events at their plasma membranes. During the cycle of events associated with cell translocation, the cells extend surface protrusions at their leading edges, make contact with the substratum at focal adhesions, and then disassemble the focal adhesions in the rear of the cell as the cell retracts its posterior to provide material for another round of surface protrusion. A lot is known about individual steps in this cycle, but much less is known about how the different steps are coordinated at the interface between the membrane and the cytoskeleton, a region of the cell known as the "membrane skeleton".

Our laboratory is interested in how the membrane skeleton controls the component events during very rapid cell movement, how it regulates cellular processes, such as chemotaxis and matrix invasion, and how these proteins are involved in organism-scale motile phenomena, including wound healing and immune function. Most of the same or similar proteins that we found in membrane skeletons from neutrophils (a type of white blood cell) also exist in human cervical and breast carcinoma cell lines and in smooth and striated muscle membranes. These proteins include cytoskeletal proteins (spectrin, actin, myosins I and II, a -actinin, supervillin), signaling proteins (Src family kinases, heterotrimeric G proteins), and proteins that organize cholesterol-enriched membrane domains (stomatin, flotillins).

Of these membrane skeleton proteins, supervillin is of special interest to us. Supervillin binds tightly to the neutrophil plasma membrane and also binds directly to at least six different cytoskeletal proteins, including actin- and microtubule-associated motors. Additional candidate supervillin interaction partners include a large number of oncogenes, tumor suppressors, and other proteins implicated in motile processes. Over- and under-expression of supervillin affects each step of the motility cycle. The gene encoding supervillin maps to a region of human chromosome 10p implicated in tumor cell motility and susceptibility to obesity and diabetes. Taken together, the working hypothesis is that supervillin is a membrane-associated adapter protein that works with a group of interacting proteins to regulate rapid motility and associated signaling processes in many cell types.

The supervillin-associated membrane skeleton localizes to specialized membrane structures, called podosomes in tissue culture cells and costameres in striated muscle. Targeting of supervillin structures to cell-substrate contact sites called focal adhesions results in focal adhesion disassembly. Reduced levels of supervillin disrupt the stimulus-mediated activation of the ERK1/2 family of mitogen-activated kinases, suggesting an important signaling role at either the plasma membrane or at internal sites of ERK1/2 activation. Live cell imaging of EGFP-tagged supervillin is consistent with associations with internal trafficking membranes, as well as at the plasma membrane.

Current projects include exploring the biochemical basis for supervillin-mediated activation of myosin contractility, identifying the mechanism(s) by which supervillin promotes reorganization of the actin cytoskeleton during invasion of extracellular matrices, and characterization of the role of the supervillin cytoskeleton during endosomal trafficking. We also have generated a supervillin-deficient mouse for studies of organ function; initial experiments on muscle contractility, immune function, and wound healing are in progress. The overall working hypothesis is that the supervillin-associated membrane skeleton coordinates cell motility and matrix invasion by promoting protein activation and trafficking at several steps in the motility cycle. The long-term goal is to understand how this coordination works during the motility and invasion associated with immune function, cancer metastasis, embryogenesis, and the formation of new blood vessels.