Motile Systems: Tubulin-based motility races ahead

The study of biological motility has been one of the key areas of biophysics for the past 30 years. Understanding mechanico-chemical transduction-how macromol-ecules and macromolecular complexes convert chemical energy to the production of force-has been seen as one of the essential goals of those seeking a physical understanding of life at the molecular level. The main focus of motility studies for most of this time has been muscle, for a variety of reasons. Not only are muscle proteins , such as actin and myosin, highly abundant, but the ordered arrangements of the thick and thin filaments in striated muscle have made them particularly amenable to structural analysis by electron microscopy or X-ray dif-fraction. Within the past five years, atomic models for the actin monomer [1-3] and the globular head of myosin [4] have become available. However, the naive hope that these structures alone would tell us how muscle works has faded, as it has become apparent that the simplest structural schemes may not fit the data [5]. Thus, the publications that have appeared recently in the area of tubulin-based motility have generated great excitement, as it is possible that our first picture of how the hydrolysis of ATP can be converted to mechanical work may come first, not from the actin-myosin system, but rather from motor proteins that move along microtubules. Just as obtaining an atomic model for actin was a tremendous step in providing the foundation for an atomic model of actin-myosin motility, so an atomic model for tubulin will be essential in understanding how the motor molecules dynein or kinesin move along microtubules. Unfortunately, the helical filaments formed by actin (F-actin) and tubulin (the microtubule) have meant that these proteins cannot be directly crystallized for X-ray diffraction in their biologically important, fila-mentous form. The approach taken with actin has been to crystallize the monomer in complexes with actin-binding proteins (thus preventing polymer formation), to solve the monomer structure and then to use that to generate a model for the actin filament that is consistent with X-ray fiber diffraction data [6]. It now appears that an atomic structure for tubulin may come from an entirely different direction. One of the greatest accomplishments of electron micros-copy has been the determination of the three-dimensional structure of an integral membrane protein, bacteri-orhodopsin, by electron diffraction and electron imaging [7]. The solution of protein structures has previously been the exclusive province of X-ray …