Michael F. Schmid, Ph.D.
Department of Biochemistry and Molecular Biology
Co-Director, National Center for Macromolecular Imaging
- Ph.D., Biochemistry, 1974, University of Washington, Seattle
- Postdoctoral, 1974-1977, Rosensteil Basic Medical Sciences Research Center, Brandeis University
Structural Studies of Crystals and Macromolecular Complexes
Actin is a key protein in cell structure and motility. We are studying an actin-based cytoskeletal structure, by applying the powerful methods of electron crystallography. This structure is the acrosomal bundle from Limulus (horseshoe crab) sperm, a natural crystal containing up to 100 actin filaments cross-linked side by side to form a rigid hexagonal array. Our efforts to understand this assembly started with a view of an individual filament averaged from all the filaments in the array (left figure below). This view revealed the architecture of a single filament to a resolution of 13 Å and showed the two-domain scruin cross-linker binding to and wrapped around a central core of actin. We were able to assign an actin-binding site for scruin based on the atomic model of actin. We are pursuing the three-dimensional structure of the bundle itself to discover the interactions between the filaments that produce the bundle.
One of the challenges in electron microscopy is to make ordered arrays of specimens of interest. A thin specimen covering a large area yields the best images. Lipid layers meet this requirement. Phospholipid layers incorporating a specific ligand or a compatible surface charge have been used to induce ordered arrays of proteins. We have extended this method to high resolution (3 Å in the case of streptavidin on biotinylated lipid). We have crystallized the 50S ribosomal subunit on a monolayer. We have probed the cellular function of botulinum toxin, which binds at low pH to the lysosomal vesicle wall and creates small channels (right figure below). We are also studying a capsid protein from Moloney murine leukemia virus that is expressed with a histidine tag and crystallized on a nickel-lipid monolayer. This technology promises to be of general applicability for making these protein arrays.
We are also developing new software to make image processing easier and more reliable. Several advances have been made recently that have been released to and used by electron microscopists worldwide.