The assembly and disassembly of microtubules at particular times are essential steps in the cell cycle. These processes are closely regulated, and interference with the regulatory mechanisms can lead to cell death. These properties have made tubulin both a fascinating specimen for biophysical studies and a useful target for anti-cancer drugs. It is important to understand how tubulin molecules interact with each other as well as with a large number of other proteins and ligands in these activities in order to have a full understanding of the life of the cell. As a first step in this direction we have determined the structure of tubulin by electron crystallography. In further work we are extending our understanding of the structure and learning more about the processes that give tubulin its unique properties. We study the interaction of tubulin with drugs that stabilize microtubules and the interactions with some of the proteins that regulate the microtubule cytoskeleton. Through studies of tubulin in complexes with ligands such as aluminum fluoride and pentalysine, and using genetic manipulation, we are investigating factors that give microtubules their particular metastable character. This work is aimed at developing a rational understanding of the functional mechanisms of microtubule dynamics and may reveal the underlying mechanism of microtubule stabilization, eventually allowing development of new, more effective drugs targeted to tubulin.

We are also developing instrumentation and technology to improve data collection in electron microscopy, especially applied to protein structure determination. One major project is the development of a new charge-coupled device (CCD) camera for intermediate voltage electron microscopes (IVEMs). CCDs have found wide application in certain aspects of electron microscopy. They provide much faster availability of image and diffraction data than photographic film, and this improvement in turnaround time has allowed a tremendous increase in the efficiency of specimen evaluation, microscope operation and automation, and data recording. However, the CCD performance is seriously compromised when the microscope is operated much above 100 kV, due to the decrease in efficiency of the scintillator, which decreases the signal level, and increased lateral scattering of the electrons within the scintillator, which decreases the resolution. Together these effects can produce a detective quantum efficiency (DQE) which is far below that of film, and indeed too low for work such as low-dose imaging of proteins. The use of IVEMs operating at 300 - 400 kV is increasing as the many advantages of the higher accelerating voltage become more widely recognized, so the CCD performance on these microscopes is becoming more of a limitation. We plan to overcome the poor performance of CCD cameras on IVEMs by decelerating the electrons before they reach the camera. In our design the CCD is mounted so that it can be floated to around 200 kV. When the microscope is operated at 300 kV we will then obtain the advantages of the IVEM but the camera will perform as well as on a 100 kV microscope. Further improvements will be made to the scintillator to produce even better signal and resolution characteristics than currently available. These adaptations will produce substantial benefits in work ranging from high-resolution structural studies of proteins and viruses to the three dimensional study of cell ultrastructure.

On a larger scale, we are studying the molecular architecture of some simple but intact cells. Our goal is to apply the same frozen-hydrated specimen preparation methodology used for molecular studies to map out the locations of the major protein complexes in small bacteria. Recent advances suggest that whole microbial cells could be imaged by electron tomography to "molecular" resolution, sufficient to locate and identify large macromolecular complexes in the native state within their cellular contexts. We plan to develop and test this technique on two microbes, one chosen for its ideal imaging characteristics, and the other for its potential Department of Energy mission relevance. Information gained by this technique will be critical to four stages of the DOE Microbial Cell Project: (1) complete functional characterization of each gene product, (2) identification of complexes and pathways, (3) fine localization and quantification of cell components, and (4) understanding the effects of microbe customizations. These stages represent some of the basic goals of all of cell biology, so that developing the technology in the context of the MCP will have very broad implications for the study of other organisms.