REGULATION OF MICROTUBULE ASSEMBLY AND STEADY-STATE DYNAMICS

Polymerization of microtubular protein is promoted by GTP but not GDP. GDP is shown to inhibit both the rate and extent of GTP-induced microtubule assembly. The critical protein concentration for assembly was increased from 0.3 mg/ml with 200 (mu)M GTP to 0.8 mg/ml with 200 (mu)M GTP and 600 (mu)M GDP. The relative binding constants for GDP and GTP to the tubulin exchangeable binding site (E-site) are used to quantitatively analyze the mechanism by which GDP raises the critical protein concentration and thereby reduces the extent of microtubule assembly.

Evidence that guanine nucleotides also interact with microtubular protein at a previously unknown, low affinity binding site is supported by the following observations: (1) At a constant ratio of GDP/GTP, a greater degree of inhibition of the extent of microtubule assembly was observed at higher concentrations of total nucleotide. This result cannot be accounted for by binding at only the E-site. (2) Inhibition of microtubule assembly by a high concentration of GDP (3 mM) is not reversed by increasing the GTP concentration to 7 mM. Based upon the dissociation constants for GDP and GTP at the E-site, tubulin would be largely saturated with GTP under these conditions and should polymerize in the absence of additional nucleotide interactions. (3) GTP specifically inhibits microtubule assembly at concentrations greater than 3 mM. Inhibition of assembly by high concentrations (7 mM) of ATP and CTP is shown to involve a different mechanism. (4) Microtubules are depolymerized by high concentrations of GTP (> 5 mM) or GDP (> 2 mM).

('32)P labeling of microtubular protein by endogenous protein kinase activity is shown to result from a net increase in protein-bound phosphate and is not the result of a phosphate exchange reaction between ATP and phosphoprotein. Protein phosphorylation is maximal in the presence of 0.5 mM Mg('2+) and 0.25 and mM ATP, resulting in approximately 2.8 nmol of phosphate/mg of protein. However, phosphorylation can be increased two to threefold by cAMP. The protein substrates for phosphorylation in either the absence of presence of cAMP are the microtubule-associated proteins which copurify with tubulin and promote microtubule assembly. Phosphorylation of microtubule-associated proteins inhibits both the rate and extent of microtubule assembly when the protein is exposed to conditions which result in dissociation of rings. The extent of microtubule assembly is reduced as a result of a decrease in the fraction of tubulin polymerized, without a significant change in the critical protein concentration. The decreased rate of microtubule assembly using phosphorylated MAPs reflects a reduction in microtubule nucleation resulting in fewer, but 2-fold longer, microtubules at steady-state. Analysis of microtubule dynamics at steady-state reveals that the rate of directional incorporation of subunits (flux) is 22 subunits(.)MT('-1)(.)sec('-1) using phosphorylated MAPs compared to 10 subunits MT('-1)(.)sec('-1) using unphosphorylated MAPs. The initial rate of disassembly determined by isothermal dilution of 232 subunits(.)MT('-1)(.)sec('-1) for microtubules assembled using phosphorylated MAPs compared to 102 subunits(.)MT('-1)(.)sec('-1) for microtubules assembled using unphosphorylated MAPs. Using these results, the directionality (the number of successful subunit additions relative to the total number of association events per unit time) for subunit addition is found to be 0.1 for microtubules assembled with either phosphorylated or unphosphorylated MAPs. These observations are interpreted in terms of a mechanism in which phosphorylation of MAPs increases the rate of steady-state subunit flux by an equivalent enhancement of the rates of subunit association and dissociation, such that the critical protein concentration and directionality remain unchanged.