Kinetics of the spontaneous organization of microtubules in solution

Optically anisotropic zones occur spontaneously in solutions of microtubules. These tactoids, in which microtubules are arranged in parallel arrays, can be visualized by their birefringence. With microtubules assembled in the presence of associated proteins (MAPs), birefringence appears immediately after nucleation of polymerization, even at relatively low protein concentrations. It is not dependent on whether the assembly is initiated by temperature jump or by isothermal addition of GTP. With pure tubulin, assembled in buffers containing 25% glycerol or 4% dimethylsulfoxide and/or taxol, birefringence appears within a few hours, but it can be speeded up by gentle agitation. With tubulin assembled in the presence of MAPs, spontaneous orientation occurs simultaneously with polymerization. This may be due to the existence of more pronounced repulsive forces between microtubules when they are covered with MAPs. A simple calculation of the covolume, suggests that tactoid formation is expected for microtubules of lengths of 5 to 10 m at protein concentrations in the range 1 to 3 mg/ml (as observed), and that repulsive forces will promote tactoid formation at even lower protein concentrations. Offprint requests to: Y. Engelborghs     

Assembly of pure tubulin in the absence of free GTP: effect of magnesium, glycerol, ATP, and the nonhydrolyzable GTP analogues

We describe in vitro microtubule assembly that exhibits, in bulk solution, behavior consistent with the GTP cap model of dynamic instability. Microtubules assembled from pure tubulin in the absence of free nucleotides could undergo one cycle of assembly, but could not sustain an assembly plateau. After the initial peak of assembly was reached and bound E-site GTP hydrolyzed to GDP, the microtubules gradually disassembled. We studied buffer conditions that maximized this disassembly while still allowing robust assembly to take place. While both glycerol and glutamate increased the rate of initial assembly and then slowed disassembly, magnesium promoted initial assembly and, surprisingly, enhanced disassembly. After cooling, a second cycle of assembly was unsuccessful unless GTP or the hydrolyzable GTP analogue GMPCPOP was readded. The nonhydrolyzable GTP analogues GMPPNP and GMPPCP could not support the second assembly cycle in the absence of E-site GTP. Analysis using HPLC found no evidence that GMPPNP, GMPPCP, or ATP could bind to free tubulin, and these nucleotides did not compete with GTP for the E-site. We have, however, demonstrated that the nonhydrolyzable GTP analogues and ATP do have an important effect on microtubule assembly. GMPPNP, GMPPCP, and ATP could each enhance the rate of assembly and stabilize the plateau of assembled microtubules against disassembly, while not binding appreciably to free tubulin. We conclude that these nucleotides, as well as GTP itself, enhance assembly by binding to a site on microtubules that is not present on free, unpolymerized tubulin. We estimate the affinity (KD) of the polymeric site for nucleotide triphosphates to be approximately 10(-4)M.     

Regulation of the microtubule steady state in vitro by ATP.

ATP increases microtubule steady state assembly and disassembly rates in vitro in a concentration-dependent manner. Bovine brain microtubules, composed of 75% tubulin and 25% high molecular weight microtubule-associated proteins (MAPs), were purified by three cycles of assembly and disassembly in the absence of ATP. When assembled to steady state, these microtubules add dimers at one end and lose them at the other in a unidirectional assembly-disassembly process. In the presence of 1.0 mM ATP the unidirectional flow of tubulin from one end of the microtubules to the other increases as much as 20 fold, as revealed by loss of 3H-GTP from uniformly labeled microtubules under GTP chase conditions and by the rate of disassembly following addition of 50 microM podophyllotoxin. UTP, CTP and 5' adenylylimidodiphosphate (AMP-PNP) cannot substitute for ATP in producing this effect. Furthermore, the increase in steady state flow rate persists afer ATP is removed. Thus microtubules assembled in ATP and centrifuged through sucrose cushions to separate them from nucleotides continue to exhibit increased rates in the next assembly cycle in the absence of ATP. It is possible that an ATP-dependent microtubule protein kinase is responsible for the observed increase in tubulin flow rate. A kinase activity associated with brain MAPs has been reported to be cAMP-dependent (Sloboda et al., 1975). We have found an adenylate cyclase activity associated with these microtubules. Whether the adenylate cyclase is a contaminant or due to a specific microtubules-associated protein, and whether its activity is functionally linked to the increased rate of assembly and disassembly in the presence of ATP, remain to be determined.     

Stabilization of microtubules by GTP analogues

We recently demonstrated that the nonhydrolyzable analogues of GTP (GMPPCP and GMPPNP) and ATP support the elongation phase of tubulin assembly and are incorporated into the E-site of polymerized tubulin. In this report we studied the stability of microtubules containing GTP analogues by examining length redistributions after shearing at polymer steady state. The mean length of a population of microtubules containing GMPPCP increased only by 37% over a 150 min time period after shearing. Microtubules which contained 70% ATP and 30% GDP at the E-site increased in length by 88%. In contrast, the mean length of microtubules assembled in the presence of GTP increased by 410% in the same time period. These results suggest that microtubules containing GMPPCP or ATP at their ends are stabilized from depolymerization.     

Effects of ATP and cyclic AMP on the in vitro assembly and stability of mammalian brain microtubules

The relevance of protein phosphorylation, transphosphorylation and binding phenomena in the kinetics of the ATP-induced assembly of cycle-purified microtubule protein from mammalian brain were studied. ATP was able to induce the polymerization of microtubules of normal appearance. However, the assembled structures, were unstable and microtubules depolymerized after achievement of a transitory maximum. Cyclic AMP reduced the amplitude of the polymerization maximum in a concentration-dependent manner, correlating with the stimulation of the endogenous phosphorylation reaction. When microtubule assembly was induced by GTP, in the presence of various concentrations of ATP, the slope of the depolymerization phase was found to depend on the concentration of ATP. Fluoride ion inhibited the endogenous phosphorylation reaction and reduced the disassembly rate, in a concentration-dependent manner. Evidence is also presented indicating that ATP did not bind to phosphocellulose-purified tubulin. These results further contribute to indicate that ATP and cyclic AMP, acting coordinately to control the phosphorylation extent of microtubule proteins are important factors to determine microtubule stability within the cell. Some implications of this mechanism for the regulation by cAMP of the initiation of DNA synthesis and mitosis are considered.Abbreviations MAPs microtubule-associated proteins - MAP2 microtubule-associated protein 2, Mes-4-morpholinoethanesulfonic acid - EGTA ethylene-glycol bis (-aminoethyl ether)N,N,N,N-tetraacetic acid - PMSF phenylmethylsulfonyl fluoride - PEI polyethyleneimine - PC phosphocellulose - DEAE Diethylaminoethyl - SDS-PAGE polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate     

Effect of cibacron blue on tubulin assembly in the absence and presence of microtubule-associated proteins

Cibacron blue was found to inhibit assembly and increase the critical concentration of microtubule proteins. In the presence of 4 mol Cibacron blue/mol tubulin, assembly was completely inhibited and pre-formed microtubules disassembled. Addition of 8% (v/v) dimethylsulfoxide to Cibacron blue-inhibited samples induced assembly of normal microtubules in addition to sheets of protofilaments. Disassembly was induced upon addition of 1 mM colchicine or 2mM Ca²⁺. No obvious difference was seen in the protein composition of these microtubules compared with controls. GTP exchange was not affected by the presence of Cibacron blue nor was GTP able to counteract its effect. This indicates that the exchangeable GTP site is not involved. The extent of assembly of phosphocellulose purified tubulin in the presence of 8% (v/v) dimethylsulfoxide was only slightly less in the presence of Cibacron blue, although the assembly rate was decreased. These results suggest that Cibacron blue might alter the binding of one or more of the associated proteins stimulating assembly.     

Chromium (III)-nucleotide complexes as probes of the guanosine 5′-triphosphate-induced microtubule assembly

Cr(III)GTP is shown to promote assembly of microtubules that are indistinguishable from those assembled using MgGTP. The rate of assembly using Cr(III)GTP is faster than the rate of assembly using MgGTP. The action of Cr(III)GTP is not due to dissociation of GTP from Cr(III)GTP. Microtubules assembled using [8-¹⁴C]Cr(III)GTP are shown to bind 0.55 mol of ¹⁴C label per mol of tubulin dimer. This is comparable to [8-¹⁴C]GTP binding under identical conditions. The distribution of ¹⁴C label is shown to be 55% Cr(III)GTP, 30% GDP, and 15% Cr(III)GDP. Microtubules assembled using Cr(III)GTP have marked resistance to calcium depolymerization but are readily depolymerized by exposure to cold. Two possible models for calcium-induced depolymerization are discussed. Neither Cr(III)ATP nor Cr(III)GDP was found to support assembly. Cr(III)ATP was found to inhibit ATP- and UTP-induced assembly but not GTP- or ITP-induced assembly. This is discussed in terms of the nucleoside diphosphokinase postulated to shuttle phosphoryl groups from nucleoside-5′-triphosphates to GDP, thereby supporting assembly indirectly.     

The association of heterotrimeric GTP-binding protein (Go) with microtubules

The heterotrimeric GTP-binding regulatory proteins (G proteins) play an important role in the regulation of membrane signal transduction. Recently, we identified the association of Go protein with mitotic spindles. Here we have investigated the relationship between Go protein and microtubules. We used temperature-dependent reversible assembly and taxol methods to purify microtubules from bovine brains. Goalpha and Gbeta proteins were identified in the microtubular fraction by both methods. The Goalpha subunit in the microtubular fraction could be ADP ribosylated by pertussis toxin. Co-immunoprecipitation data also revealed that Go protein can interact with microtubules. Exogenous Go protein could be incorporated into the assembled microtubular fraction, and 5 microg/ml (60 nM) of Go protein inhibited 40% of microtubule assembly. Western blot analysis of Goalpha-1 and Goalpha-2 in microtubular fractions showed that only Goalpha-1 is associated with microtubules. We conclude that the Goalpha-1betagamma proteins are associated with microtubules and may play some role in regulating the assembly and disassembly of microtubules.     

An oscillatory mode for microtubule assembly.

Depending upon the conditions under which polymerization takes place, pure tubulin can assemble into microtubules following either the usual monotonic kinetics or a more complex oscillatory mechanism. When present, these oscillations involve large cyclic changes in the extent of polymer formed before a steady-state is reached. Analysis of the microtubules formed at different times shows that these oscillations involve marked redistribution in both the length and number of microtubules. No significant difference is found between two populations of microtubules corresponding to the same level of assembly, one for which the extent of polymerization will remain stable with time and one for which it will decrease by as much as 90% in the next oscillation. The amplitude of these oscillations is sensitive to changes in the concentrations of protein, nucleotide (GTP, GDP or GMPpNp), magnesium ion or GTP regenerating system. A complete shift from an oscillatory to a monotonic polymerization can be induced by a minor increase in the concentration of free nucleotide, GTP or GDP.     

GTP analogues interact with the tubulin exchangeable site during assembly and upon binding

The question of whether nonhydrolyzable nucleotide analogues and other nucleoside triphosphates support tubulin assembly was addressed. Tubulin which contained residual GTP at the exchangeable site polymerized in the absence of added GTP in the presence of DMSO or glycerol. After maximum absorbance was reached, disassembly occurred at a slow rate. When 0.5 mM GMPPCP, GMPPNP, or ATP was included in the assembly reaction, disassembly did not occur, and about 0.1 mol of these nucleotides per mole of tubulin was incorporated into the protein. When 5 mM nucleotide was used or alkaline phosphatase was included in the case of the nonhydrolyzable analogues, a greater amount of assembly occurred and about 0.7-0.8 mol of analogue was incorporated. The products of the assembly reaction were cold-labile microtubules and protofilament ribbons. After cold-depolymerization of the microtubules and ribbons, a second cycle of assembly produced some microtubules, but cold-stable amorphous polymers were the major product. In addition, when GTP at the exchangeable site was first removed by a cycle of assembly, followed by depolymerization, assembly in the presence of GMPPCP, GMPPNP, or ATP produced a mixture of microtubules and cold-stable polymers, both of which contained bound analogue. Incorporation of GMPPCP, GMPPNP, or ATP into polymerized tubulin always occurred at the expense of GDP at the exchangeable site, the content of which decreased correspondingly. Incubation of tubulin with 5 mM GMPPCP, GMPPNP, or ATP under nonassembly conditions also displaced GDP.(ABSTRACT TRUNCATED AT 250 WORDS)     

The Balance Between τ Protein's Microtubule Growth and Nucleation Activities: Implications for the Formation of Axonal Microtubules

Abstract: The microtubule-associated protein τ is found primarily in neuronal tissues and is highly enriched in the axon. It promotes microtubule assembly in vitro and stabilizes microtubules in cells. To study how τ protein might be involved in the unique features of axonal microtubules, we have analyzed the effect of E. coli -synthesized τ protein using an in vitro centrosome-mediated microtubule regrowth assay over a wide range of τ/tubulin ratios. We report that microtubule assembly promoted by τ protein exhibits characteristic changes dependent on the τ/tubulin ratio. Above a threshold level, nucleation of new microtubules is favored over growth of existing ones, τ isoform variation does not change this phase transition in microtubule assembly. We discuss how τ might participate in the elaboration of axonal morphology based on our results and present evidence that the phase transition from microtubule growth to nucleation is critical for axonal development.     

Promotion of tubulin assembly by poorly soluble taxol analogs

Promotion or inhibition of tubulin assembly into microtubules is the standard in vitro assay for evaluating potential antimicrotubule agents. Many agents to be tested are poorly soluble in aqueous solution and require a cosolvent such as dimethyl sulfoxide (DMSO). However, DMSO itself can promote tubulin assembly, and its inclusion in assays for compounds that induce tubulin assembly complicates interpretation of the results. Substituting GDP for GTP in the exchangeable nucleotide binding site of tubulin produces a less active form of the protein, tubulin-GDP. Here it is shown that tubulin-GDP can be assembled into normal microtubules in DMSO concentrations up to 15% (v/v), and polymerization assays performed under these conditions can be compared with assays run under more standard conditions. Assays for measuring the effective concentration of a ligand for promotion of tubulin assembly (EC(50)), measuring the concentration for inhibition of tubulin assembly (IC(50)) by a colchicine site ligand, and measuring tubulin critical concentrations in the presence of poorly soluble taxol derivatives are illustrated.     

Association of microtubules and neurofilaments in vitro is not mediated by ATP

Runge et al. [Runge, M.S., Laue, T.M., Yphantis, D.A., Lifsics, M.R., Saito, A., Altin, M., Reinke, K., & Williams, R.C., Jr. (1981) Proc. Natl. Acad. Sci. U.S.A. 78, 1431-1435] found that mixtures of microtubules and neurofilaments formed a viscous, sedimentable complex when incubated at 37 degrees C for 20 min in the presence of ATP. They did not observe the high viscosities associated with the complex when the incubation was carried out in the absence of ATP. This paper reports an investigation of the roles of time and ATP in the formation of the complex. Microtubules assembled in a mixture containing GTP and neurofilaments prepared from bovine brain remained assembled for a shorter period of time than they did in similar solutions containing no neurofilaments. Adding ATP to the neurofilament-containing solutions, or doubling their GTP concentration, extended the time during which the microtubules remained assembled. These mixtures then became highly viscous. These phenomena resulted from the action of at least two enzymes present in the neurofilament preparation. A GTPase raised the GDP/GTP ratio, in the mixtures in which ATP was absent, to levels sufficient to cause disassembly of the microtubules. When ATP was present, a nucleotide diphosphokinase catalyzed regeneration of GTP from GDP while converting ATP to ADP. This process kept the GDP/GTP ratio low and delayed the disassembly of the microtubules. These results show that the apparent ATP dependence of formation of the microtubule-neurofilament complex observed by Runge et al. is attributable to a GDP-induced disassembly of microtubules rather than to a disruption of microtubule-neurofilament contacts. Those contacts can form in the absence of ATP.     

Mechanism of endogenous phosphorylation of microtubule proteins during GTP-induced microtubule assembly and implications for stability of the assembled structures

Cycle-purified microtubule protein from mammalian brain incorporated [32P]Pi upon incubation with [gamma-32P]GTP under the conditions used to promote assembly. This phosphorylation also occurred in the same proteins when phosphorylated with [gamma-32P]ATP and was only slightly stimulated by cAMP. GTP was a much less effective substrate than ATP. The transfer of phosphoryl groups from [gamma-32P]GTP to endogenous proteins followed a linear time-course and was stimulated by low concentrations of ATP and, more efficiently, by ADP. These data are in agreement with the predictions derived from a mechanism of phosphorylation by which [gamma-32P]GTP does not act as a phosphoryl donor for the protein kinase activity but, instead, only as a repository of high group transfer potential phosphoryl groups used to make [gamma-32P]ATP, from contaminating ADP, by means of the nucleoside diphosphate kinase activity. Using 100 mM fluoride, which suppressed protein phosphorylation without inhibiting the nucleoside diphosphate kinase activity, formation of [gamma-32P]ATP was detected. Fluoride was also able to protect microtubules from a slow depolymerization which was found to occur during long-term incubation of microtubules. This indicates that the phosphorylation observed in the presence of GTP is sufficient to destabilize microtubules.     

Neurotubule assembly at substoichiometric nucleotide levels using a GTP regenerating system

Tubulin derived from cold depolymerized bovine microtubules has been gel filtered to obtain a tubulin preparation with only 3% of the tubulin dimers containing exchangeable [³H]-guanine nucleotide. In the presence of acetyl-P and bacterial acetate kinase, this preparation polymerizes to form microtubules which are morphologically indistinguishable from microtubules formed in the presence of excess GTP. The extent of microtubule formation at substoichiometric nucleotide levels using the GTP regenerating system exceeds the extent of assembly obtained with excess GTP. It is concluded that the exchangeable guanine nucleotide site can be virtually unoccupied in intact neurotubules and this finding indicates that GDP can “catalyze” tubule assembly in the presence of a GTP regenerating system.     

Influence of Mg2+ and inorganic phosphates on the assembly of tubulin depleted of its exchangeable guanine nucleotide.

After the removal of the exchangeable guanine nucleotides by chromatography on phenyl-Sepharose [Hanssens, I., Baert, J., and Van Cauwelaert, F. (1990) Biochemistry 29, 5160-5165] tubulin polymerizations with GTP, GDP, tripolyphosphate, pyrophosphate or orthophosphate as possible stimulants are compared. It is demonstrated that, besides GTP and pyrophosphate, also tripolyphosphate stimulates the assembly into microtubules at high concentrations (4.65 mM) of Mg2+. The influence of Mg2+ is more pronounced in combination with pyrophosphate and tripolyphosphate than with GTP. The microtubules assembled in combination with Mg2+ and tripolyphosphate or pyrophosphate are short, suggesting that especially the nucleation step of microtubule assembly is favoured.     

The effects of various GTP analogues on microtubule assembly.

We synthesized 27 GTP analogues with modification or substitution at positions C2, C6, C8 and ribose moiety to investigate their effect on microtubule (Mt) assembly. It was found that C2 and C6 are both functional for the analogues supporting Mt assembly. It was surprising to find that 2-amino- ATP (n2ATP) substantially supports assembly, and that the appearance of the assembled Mts was indistinguishable from those assembled in the standard GTP assembly buffer solution. Furthermore, 2-amino dATP and dGTP are even more potent than GTP in supporting assembly. The substitution of oxo group at C6 with reactive thiol largely reduced the activity of the analogue to support assembly. When free rotation of the glycosidic linkage of GTP was blocked by the introduction of sulfur atom between C8 and C2' of ribose moiety, it resulted in total suppression of assembly. Purine nucleoside triphosphate was found to support assembly better than GTP, and even more efficient was 2-amino purine nucleoside triphosphate. Interestingly, their deoxy-type analogues were totally inhibitory. Although 2-amino 8-hydroxy ATP and other analogues supported assembly much better than did GTP, their diphosphate analogues were totally incapable of supporting assembly. Finally, bulky fluorescent probes were introduced at C3' of ribose moiety (Mant-8-Br-GTP or Mant-GTP) to visualize the fluorescent signal in assembled Mts. Even in this case, the number of most protofilaments was found to be 14, consistent with that found in Mts assembled in GTP standard buffer solution.     

Taxol assembles tubulin in the absence of exogenous guanosine 5'-triphosphate or microtubule-associated proteins

Taxol increases the rate and extent of microtubule assembly in vitro and stabilizes microtubules in vitro and in cells [Schiff, P. B., Fant, J., & Horwitz, S. B. (1979) Nature (London) 277, 665-667; Schiff, P. B., & Horwitz, S. B. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 1561-1565]. We report herein that taxol has the ability to promote microtubule assembly in the absence of microtubule-associated proteins, rings, and added guanosine 5'-triphosphate (GTP or organic buffer. The drug enhances additional microtubule assembly when added to microtubules at apparent steady state. This additional assembly can be attributed to both elongation of existing microtubules and spontaneous nucleation of new microtubules. Taxol-treated microtubules have depressed dissociation reactions as determined by dilution experiments. The drug does not inhibit the binding of GTP or the hydrolysis of GTP or guanosine 5'-diphosphate (GDP) in our microtubule protein preparations. Taxol does not competitively inhibit the binding of colchicine to tubulin.     

Mammalian brain microtubules are sensitive to cyclic AMP in vitro

Microtubules assembled in vitro with ATP were depolymerized by the addition of cyclic AMP, which correlates with a stimulation of the endogeneous phosphorylation reaction. When assembled with GTP, however, microtubules were only sensitive to cyclic AMP when ATP was present. This nucleoside triphosphate induced the disassembly of microtubules in a concentration-dependent, cyclic nucleotide-stimulated manner. Since UTP, CTP and the nonhydrolyzable ATP analog adenosine-5'-(beta, gamma-methylene)triphosphate were without comparable effect, it was assumed that phosphorylation of the microtubule-associated proteins may represent a physiological mechanism by which microtubules in the living cell respond to external stimuli.     

GTP hydrolysis during microtubule assembly

The GTP cap model of dynamic instability [Mitchison, T., & Kirschner, M.W. (1984) Nature (London) 312, 237] postulates that a GTP cap at the end of most microtubules stabilizes the polymer and allows continuing assembly of GTP-tubulin subunits while microtubules without a cap rapidly disassemble. This attractive explanation for observed microtubule behavior is based on the suggestion that hydrolysis of GTP is not coupled to assembly but rather takes place as a first-order reaction after a subunit is assembled onto a polymer end. Carlier and Pantaloni [Carlier, M., & Pantaloni, D. (1981) Biochemistry 20, 1918] reported a lag of hydrolysis behind microtubule assembly and a first-order rate constant for hydrolysis (kh) of 0.25/min. A lag has not been demonstrated by other investigators, and a kh value that specifies such a slow rate of hydrolysis is difficult to reconcile with reported steady-state microtubule growth rates and frequencies of disassembly. We have looked for a lag using tubulin free of microtubule-associated protein at concentrations of 18.5-74 microM, assembly with and without glycerol, and two independent assays of GTP hydrolysis. No lag was observed under any of the conditions employed, with initial rates of hydrolysis increasing in proportion to rates of assembly. If hydrolysis is uncoupled from assembly, we estimate that kh must be at least 2.5/min and could be much greater, a result that we argue may be advantageous to the GTP cap model. We also describe a preliminary model of assembly coupled to hydrolysis that specifies formation and loss of a GTP cap, thus allowing dynamic instability.