Analysis of tubulin oligomers by analytical ultracentrifugation

This chapter describes the use of analytical ultracentrifugation in a Beckman XLA to study the self-association properties of tubulin and the interaction of tubulin with antimitotic drugs. Procedures for sample preparation, operation of the ultracentrifuge, and collection of data conform to standard modern methods. Analysis of sedimentation velocity data initially includes generation of g(s) sedimentation coefficient distributions with DCDT(+2) and determination of weight average sedimentation coefficients S(w). S(w) versus concentration data are then fit to isodesmic or indefinite assembly models to extract K(iso) values, the association constant for each successive assembly step. Alternatively the raw data can also be analyzed by direct boundary analysis methods using the program Sedanal. Direct boundary analysis also extracts the K(iso) value by fitting to the shape of the sedimentation boundary as a function of total concentration. While the fitting of weight average data as a function of protein or drug concentration to indefinite assembly models has been shown to be equivalent to direct boundary fitting of multiple data sets with Sedanal, direct boundary fitting is preferred because it robustly identifies the presence of irreversible aggregation or mechanisms that are more complex.