Characterization of rat brain crude extract microtubule assembly: correlation of cold stability with the phosphorylation state of a microtubule-associated 64K protein.

We have conducted preliminary investigations into the control of microtubule assembly in rat brain crude extract supernatants. The rationale for these experiments is that microtubules interact with many proteins and are undoubtedly subject to physiological control mechanisms that are lost during tubulin purification. A more complete understanding of the cellular regulation of microtubules must include the physiology of these proteins. Assembly can be monitored in rat brain crude extract high-speed supernatants by measuring the increase in solution turbidity. We find that assembly is maximal in both rate and extent in the absence of added nucleotide. Increasing concentrations of either adenosine 5'-triphosphate (ATP) or guanosine 5'-triphosphate (GTP) inhibit both initiation and elongation of microtubules. GTP appears necessary for assembly and is apparently replenished from an intrinsic energy source during the time course of the assembly reaction. Inhibition of GTP production prevents microtubule assembly, and addition of exogenous GTP will reverse the blockage. Enzymatic removal of GTP at steady state causes a rapid depolymerization to the cold-stable microtubule level. Both GTP production and microtubule assembly display periodic oscillatory maxima. Cold-stable microtubules, which are always present in rat brain crude extract preparations, are rapidly made labile by addition of ATP. Analysis of proteins in cold-stable and cold-labile microtubule fractions shows changes in protein phosphorylation but not in the microtubule-associated protein composition. The tentative conclusion is that the state of phosphorylation of a 64K protein, designated the "switch protein", determines the cold stability or lability, and therefore the dimer association and dissociation rates, of crude extract microtubules.