Microtubule dynamics and microtubule caps: a time-resolved cryo- electron microscopy study

Microtubules display the unique property of dynamic instability characterized by phase changes between growth and shrinkage, even in constant environmental conditions. The phases can be synchronized, leading to bulk oscillations of microtubules. To study the structural basis of dynamic instability we have examined growing, shrinking, and oscillating microtubules by time-resolved cryo-EM. In particular we have addressed three questions which are currently a matter of debate: (a) What is the relationship between microtubules, tubulin subunits, and tubulin oligomers in microtubule dynamics?; (b) How do microtubules shrink? By release of subunits or via oligomers?; and (c) Is there a conformational change at microtubule ends during the transitions from growth to shrinkage and vice versa? The results show that (a) oscillating microtubules coexist with a substantial fraction of oligomers, even at a maximum of microtubule assembly; (b) microtubules disassemble primarily into oligomers; and (c) the ends of growing microtubules have straight protofilaments, shrinking microtubules have protofilaments coiled inside out. This is interpreted as a transition from a tense to a relaxed conformation which could be used to perform work, as suggested by some models of poleward chromosome movement during anaphase.     

Structural transitions at microtubule ends correlate with their dynamic properties in Xenopus egg extracts

Microtubules are dynamically unstable polymers that interconvert stochastically between growing and shrinking states by the addition and loss of subunits from their ends. However, there is little experimental data on the relationship between microtubule end structure and the regulation of dynamic instability. To investigate this relationship, we have modulated dynamic instability in Xenopus egg extracts by adding a catastrophe-promoting factor, Op18/stathmin. Using electron cryomicroscopy, we find that microtubules in cytoplasmic extracts grow by the extension of a two- dimensional sheet of protofilaments, which later closes into a tube. Increasing the catastrophe frequency by the addition of Op18/stathmin decreases both the length and frequency of the occurrence of sheets and increases the number of frayed ends. Interestingly, we also find that more dynamic populations contain more blunt ends, suggesting that these are a metastable intermediate between shrinking and growing microtubules. Our results demonstrate for the first time that microtubule assembly in physiological conditions is a two-dimensional process, and they suggest that the two-dimensional sheets stabilize microtubules against catastrophes. We present a model in which the frequency of catastrophes is directly correlated with the structural state of microtubule ends.     

Straight GDP-tubulin protofilaments form in the presence of taxol

Microtubules exist in dynamic equilibrium, growing and shrinking by the addition or loss of tubulin dimers from the ends of protofilaments. The hydrolysis of GTP in beta-tubulin destabilizes the microtubule lattice by increasing the curvature of protofilaments in the microtubule and putting strain on the lattice. The observation that protofilament curvature depends on GTP hydrolysis suggests that microtubule destabilizers and stabilizers work by modulating the curvature of the microtubule lattice itself. Indeed, the microtubule destabilizer MCAK has been shown to increase the curvature of protofilaments during depolymerization. Here, we show that the atomic force microscopy (AFM) of individual tubulin protofilaments provides sufficient resolution to allow the imaging of single protofilaments in their native environment. By using this assay, we confirm previous results for the effects of GTP hydrolysis and MCAK on the conformation of protofilaments. We go on to show that taxol stabilizes microtubules by straightening the GDP protofilament and slowing down the transition of protofilaments from straight to a curved configuration.     

Imaging microtubules and kinesin decorated microtubules using tapping mode atomic force microscopy in fluids

The atomic force microscope has been used to investigate microtubules and kinesin decorated microtubules in aqueous solution adsorbed onto a solid substrate. The netto negatively charged microtubules did not adsorb to negatively charged solid surfaces but to glass covalently coated with the highly positively charged silane trimethoxysilylpropyldiethylenetriamine (DETA) or a lipid bilayer of 1,2-dipalmitoyl-3-dimethylammoniumpropane. Using electron beam deposited tips for microtubules adsorbed on DETA, single protofilaments could be observed showing that the resolution is up to 5 nm. Under conditions where the silane coated surfaces are hydrophobic, microtubules opened, presumably at the seam, whose stability is lower than that of the bonds between the other protofilaments. This led to a “sheet” with a width of about 100 nm firmly attached to the surface. Microtubules decorated with a stoichiometric low amount of kinesin molecules in the presence of the non-hydrolyzable ATP-analog 5′-adenylylimidodiphosphate could also be adsorbed onto silane-coated glass. Imaging was very stable and the molecules did not show any scan-induced deformation even after hundreds of scans with a scan frequency of 100 Hz. Received: 23 February 1999 / Revised version: 19 July 1999 / Accepted: 17 August 1999     

Microtubules switch occasionally into unfavorable configurations during elongation

Tubulin assembles to form a range of structures that differ by their protofilament and monomer helix-start numbers. The microtubule lattice is believed to accommodate these different configurations by skewing the protofilaments so that the lateral interactions between tubulin subunits are maintained. Here, we present the characterization of 14 types of microtubules, including six novel ones, through an extensive analysis of microtubules assembled in vitro from pure tubulin. Although the six new types represented only 1 % of the total length of the population examined ( approximately 17 mm), they define the limits of microtubule structure and assembly. Protofilament skewing is restricted to within +/-2 degrees. Outside this range, the restoring force induced by the skewed protofilaments is compensated by a longitudinal shift (less than +/-0.2 nm) between adjacent protofilaments. Configurations with theoretical protofilament skew angles larger than +/-4 degrees or that necessitate larger modifications of the microtubule surface lattice were not observed. Analysis of the microtubule types distribution reveals that it is sharply peaked around the less skewed conformations. These results indicate that both the flexibility of the protofilaments and the strength of their lateral interactions restrict the range of structures assembled. They also demonstrate that growing microtubules can occasionally switch into energetically unfavorable configurations, a behavior that may account for the stochastic nature of catastrophes.     

The minimum GTP cap required to stabilize microtubules

Background: Microtubules polymerized from pure tubulin show the unusual property of dynamic instability, in which both growing and shrinking polymers coexist at steady state. Shortly after its addition to a microtubule end, a tubulin subunit hydrolyzes its bound GTP. Studies with non-hydrolyzable analogs have shown that GTP hydrolysis is not required for microtubule assembly, but is essential for generating a dynamic polymer, in which the subunits at the growing tip have bound GTP and those in the bulk of the polymer have bound GDP. It has been suggested that loss of the ‘GTP cap’ through dissociation or hydrolysis exposes the unstable GDP core, leading to rapid depolymerization. However, evidence for a stabilizing cap has been very difficult to obtain.Results We developed an assay to determine the minimum GTP cap necessary to stabilize a microtubule from shrinking. Assembly of a small number of subunits containing a slowly hydrolyzed GTP analog (GMPCPP) onto the end of dynamic microtubules stabilized the polymer to dilution. By labeling the subunits with rhodamine, we measured the size of the cap and found that as few as 40 subunits were sufficient to stabilize a microtubule.Conclusion On the basis of statistical arguments, in which the proportion of stabilized microtubules is compared to the probability that when 40 GMPCPP-tubulin subunits have polymerized onto a microtubule end, all protofilaments have added at least one GMPCPP-tubulin subunit, our measurements of cap size support a model in which a single GTP subunit at the end of each of the 13 protofilaments of a microtubule is sufficient for stabilization. Depolymerization of a microtubule may be initiated by an exposed tubulin–GDP subunit at even a single position. These results have implications for the structure of microtubules and their means of regulation.     

Conformational changes in tubulin in GMPCPP and GDP-taxol microtubules observed by cryoelectron microscopy

Microtubules are dynamic polymers that stochastically switch between growing and shrinking phases. Microtubule dynamics are regulated by guanosine triphosphate (GTP) hydrolysis by β-tubulin, but the mechanism of this regulation remains elusive because high-resolution microtubule structures have only been revealed for the guanosine diphosphate (GDP) state. In this paper, we solved the cryoelectron microscopy (cryo-EM) structure of microtubule stabilized with a GTP analogue, guanylyl 5′-α,β-methylenediphosphonate (GMPCPP), at 8.8-Å resolution by developing a novel cryo-EM image reconstruction algorithm. In contrast to the crystal structures of GTP-bound tubulin relatives such as γ-tubulin and bacterial tubulins, significant changes were detected between GMPCPP and GDP-taxol microtubules at the contacts between tubulins both along the protofilament and between neighboring protofilaments, contributing to the stability of the microtubule. These findings are consistent with the structural plasticity or lattice model and suggest the structural basis not only for the regulatory mechanism of microtubule dynamics but also for the recognition of the nucleotide state of the microtubule by several microtubule-binding proteins, such as EB1 or kinesin.     

Dynamic microtubules slow down during their shrinkage phase

Microtubules are dynamic polymers that undergo stochastic transitions between growing and shrinking phases. The structural and chemical properties of these phases remain poorly understood. The transition from growth to shrinkage, termed catastrophe, is not a first-order reaction but rather a multistep process whose frequency increases with the growth time: the microtubule ages as the older microtubule tip becomes more unstable. Aging shows that the growing phase is not a single state but comprises several substates of increasing instability. To investigate whether the shrinking phase is also multistate, we characterized the kinetics of microtubule shrinkage following catastrophe using an in vitro reconstitution assay with purified tubulins. We found that the shrinkage speed is highly variable across microtubules and that the shrinkage speed of individual microtubules slows down over time by as much as several fold. The shrinkage slowdown was observed in both fluorescently labeled and unlabeled microtubules as well as in microtubules polymerized from tubulin purified from different species, suggesting that the shrinkage slowdown is a general property of microtubules. These results indicate that microtubule shrinkage, like catastrophe, is time dependent and that the shrinking microtubule tip passes through a succession of states of increasing stability. We hypothesize that the shrinkage slowdown is due to destabilizing events that took place during growth, which led to multistep catastrophe. This suggests that the aging associated with growth is also manifested during shrinkage, with the older, more unstable growing tip being associated with a faster depolymerizing shrinking tip.     

S. pombe kinesins-8 promote both nucleation and catastrophe of microtubules

The kinesins-8 were originally thought to be microtubule depolymerases, but are now emerging as more versatile catalysts of microtubule dynamics. We show here that S. pombe Klp5-436 and Klp6-440 are non-processive plus-end-directed motors whose in vitro velocities on S. pombe microtubules at 7 and 23 nm s(-1) are too slow to keep pace with the growing tips of dynamic interphase microtubules in living S. pombe. In vitro, Klp5 and 6 dimers exhibit a hitherto-undescribed combination of strong enhancement of microtubule nucleation with no effect on growth rate or catastrophe frequency. By contrast in vivo, both Klp5 and Klp6 promote microtubule catastrophe at cell ends whilst Klp6 also increases the number of interphase microtubule arrays (IMAs). Our data support a model in which Klp5/6 bind tightly to free tubulin heterodimers, strongly promoting the nucleation of new microtubules, and then continue to land as a tubulin-motor complex on the tips of growing microtubules, with the motors then dissociating after a few seconds residence on the lattice. In vivo, we predict that only at cell ends, when growing microtubule tips become lodged and their growth slows down, will Klp5/6 motor activity succeed in tracking growing microtubule tips. This mechanism would allow Klp5/6 to detect the arrival of microtubule tips at cells ends and to amplify the intrinsic tendency for microtubules to catastrophise in compression at cell ends. Our evidence identifies Klp5 and 6 as spatial regulators of microtubule dynamics that enhance both microtubule nucleation at the cell centre and microtubule catastrophe at the cell ends.     

Model of Fission Yeast Cell Shape Driven by Membrane-Bound Growth Factors and the Cytoskeleton

Fission yeast serves as a model for how cellular polarization machinery consisting of signaling molecules and the actin and microtubule cytoskeleton regulates cell shape. In this work, we develop mathematical models to investigate how these cells maintain a tubular shape of approximately constant diameter. Many studies identify active Cdc42, found in a cap at the inner membrane of growing cell tips, as an important regulator of local cell wall remodeling, likely through control of exocyst tethering and the targeting of other polarity-enhancing structures. First, we show that a computational model with Cdc42-dependent local cell wall remodeling under turgor pressure predicts a relationship between spatial extent of growth signal and cell diameter that is in agreement with prior experiments. Second, we model the consequences of feedback between cell shape and distribution of Cdc42 growth signal at cell tips. We show that stability of cell diameter over successive cell divisions places restrictions on their mutual dependence. We argue that simple models where the spatial extent of the tip growth signal relies solely on geometrical alignment of confined microtubules might lead to unstable width regulation. Third, we study a computational model that combines a growth signal distributed over a characteristic length scale (as, for example, by a reaction-diffusion mechanism) with an axis-sensing microtubules system that places landmarks at positions where microtubule tips touch the cortex. A two-dimensional implementation of this model leads to stable cell diameter for a wide range of parameters. Changes to the parameters of this model reproduce straight, bent, and bulged cell shapes, and we discuss how this model is consistent with other observed cell shapes in mutants. Our work provides an initial quantitative framework for understanding the regulation of cell shape in fission yeast, and a scaffold for understanding this process on a more molecular level in the future.     

Characterization of microtubule protofilament numbers. How does the surface lattice accommodate?

Frozen-hydrated specimens of microtubules assembled in vitro were observed by cryoelectron microscopy. Specimens were of both pure tubulin, and of microtubule protein isolated by three cycles of assembly and disassembly. It is shown that the characteristic image contrast of individual microtubules allows the microtubule protofilament number to be determined unambiguously. Microtubules with 13, 14 and 15 protofilaments are observed to coexist in specimens prepared under various assembly conditions. Confirmation of these results is obtained by observations of thin sections of pelleted samples fixed and stained using the glutaraldehyde/tannic acid technique. Images of individual microtubules show both characteristic contrast profiles across their width and typical variations of these profiles along their length. The profiles across the images indicate the protofilament number of the microtubule. The lengthwise variations indicate how the protofilaments are aligned with respect to the microtubule axis giving what has previously been called a supertwist. In 13 protofilament microtubules the protofilaments are paraxial. In 14 and 15 protofilament microtubules, the protofilaments are skewed with respect to the microtubule axis. The skew is greater for the 15 protofilament case than for 14 protofilaments. The skew allows the extra protofilaments to be accommodated by the surface lattice. These results should also be relevant to situations in vivo.     

Lattice defects in microtubules: protofilament numbers vary within individual microtubules

We have used cryo-electron microscopy of vitrified specimens to study microtubules assembled both from three cycle purified tubulin (3x-tubulin) and in cell free extracts of Xenopus eggs. In vitro assembled 3x-tubulin samples have a majority of microtubules with 14 protofilaments whereas in cell extracts most microtubules have 13 protofilaments. Microtubule polymorphism was observed in both cases. The number of protofilaments can change abruptly along individual microtubules usually by single increments but double increments also occur. For 3x-tubulin, increasing the magnesium concentration decreases the proportion of 14 protofilament microtubules and decreases the average separation between transitions in these microtubules. Protofilament discontinuities may correspond to dislocation-like defects in the microtubule surface lattice.     

Insight into the GTPase activity of tubulin from complexes with stathmin-like domains

Microtubules are dynamically unstable tubulin polymers that interconvert stochastically between growing and shrinking states, a property central to their cellular functions. Following its incorporation in microtubules, tubulin hydrolyzes one GTP molecule. Microtubule dynamic instability depends on GTP hydrolysis so that this activity is crucial to the regulation of microtubule assembly. Tubulin also has a much lower GTPase activity in solution. We have used ternary complexes made of two tubulin molecules and one stathmin-like domain to investigate the mechanism of the tubulin GTPase activity in solution. We show that whereas stathmin-like domains and colchicine enhance this activity, it is inhibited by vinblastine and by the N-terminal part of stathmin-like domains. Taken together with the structures of the tubulin-colchicine-stathmin-like domain-vinblastine complex and of microtubules, our results lead to the conclusions that the tubulin-colchicine GTPase activity in solution is caused by tubulin-tubulin associations and that the residues involved in catalysis comprise the beta tubulin GTP binding site and alpha tubulin residues that participate in intermolecular interactions in protofilaments. This site resembles the one that has been proposed to give rise to GTP hydrolysis in microtubules. The widely different hydrolysis rates in these two sites result at least in part from the curved and straight tubulin assemblies in solution and in microtubules, respectively.     

New data on the microtubule surface lattice

The in vitro polymerisation of tubulin is a remarkable example of protein self-assembly in that several closely related microtubule structures coexist on the polymerisation plateau. Unfixed and unstained in vitro assembled microtubules were observed in vitreous ice by cryo-electron microscopy. New results are reported that considerably extend previous observations [47]. In ice, microtubule images have a distinctive contrast related to the number and skew of the protofilaments. The microtubules observed have from twelve to seventeen protofilaments. Comparison with thin sections of pelleted material allows a direct identification of images from microtubules with thirteen, fourteen and fifteen protofilaments. A surface lattice accommodation mechanism, previously proposed to explain how variable numbers of protofilaments can be incorporated into the basic thirteen protofilament structure, is described in detail. Our new experimental results are shown to be in overall agreement with the theoretical predictions. Only thirteen protofilament microtubules have unskewed protofilaments, this was confirmed by observations on axoneme fragments. The results imply that the microtubule surface lattice is based on a mixed packing which combines features of the standard A and B lattices.     

Structure of the tubulin dimer in zinc-induced sheets

The structure of tubulin has been studied in projection by minimum beam electron microscopy and image processing of negatively stained zinc-induced sheets. The reconstructed images include data to 15 Å resolution.We report here a clear and reproducible 82 Å repeat arising from the arrangement of heterodimers in sheet aggregates of tubulin. This repeat is only observed in diffraction patterns from images recorded by minimum beam methods (10 to 20 e/Ų) and arises from small, but consistent, structural differences between two similar subunits believed to represent the two chemical species of tubulin monomer (M_r, 55,000). At higher electron doses (100 to 200 e/Ų), the additional information is lost or very much reduced, and only a repeat of 41 Å is observed, owing to the loss of distinction between monomers in the tubulin heterodimer.The sheets are composed of 49 Å wide, polar protofilaments, similar to those observed in microtubules; however, the interprotofilament packing is completely different in the two structures. In these sheets, adjacent protofilaments point and face in opposite directions; i.e. they are related by dyad-screw axes normal to the protofilament axes and in the plane of the sheet. Thus, the zinc-induced sheets are crystals of space group P2₁, with cell dimensions of about 97 Å × 82 Å, containing one tubulin heterodimer per asymmetric unit.Reconstructed images of four individual sheets, and their average, show the arrangement and shapes of the two heterodimers contained in each unit cell. The structure and packing of heterodimers in sheets are compared to those in opened out microtubules where all protofilaments point and face in the same direction.     

Insights into allosteric control of microtubule dynamics from a buried β‐tubulin mutation that causes faster growth and slower shrinkage

αβ‐tubulin subunits cycle through a series of different conformations in the polymer lattice during microtubule growing and shrinking. How these allosteric responses to different tubulin:tubulin contacts contribute to microtubule dynamics, and whether the contributions are evolutionarily conserved, remains poorly understood. Here, we sought to determine whether the microtubule‐stabilizing effects (slower shrinking) of the β:T238A mutation we previously observed using yeast αβ‐tubulin would generalize to mammalian microtubules. Using recombinant human microtubules as a model, we found that the mutation caused slow microtubule shrinking, indicating that this effect of the mutation is indeed conserved. However, unlike in yeast, β:T238A human microtubules grew faster than wild‐type and the mutation did not appear to attenuate the conformational change associated with guanosine 5′‐triphosphate (GTP) hydrolysis in the lattice. We conclude that the assembly‐dependent conformational change in αβ‐tubulin can contribute to determine the rates of microtubule growing as well as shrinking. Our results also suggest that an allosteric perturbation like the β:T238A mutation can alter the behavior of terminal subunits without accompanying changes in the conformation of fully surrounded subunits in the body of the microtubule.     

On the stall force for growing microtubules

The assembly of microtubules generates forces that play a role in cellular motility processes such as the motion of chromosomes during mitosis. Recently, Mogilner and Oster proposed a model for the growth of microtubules that agrees quantitatively with the force-velocity relation measured for individual microtubules. In addition, the authors predicted that the stall force for any polymer consisting of N independently growing protofilaments should increase as the square root of N. We simulated this model and found that the stall force increases linearly with N, and is in fact consistent with the maximum force predicted by thermodynamic arguments. We show that this discrepancy can be explained by a more careful treatment of the “off-term” in the Mogilner-Oster model. Received: 27 September 1999 / Revised version: 12 December 1999 / Accepted: 20 December 1999     

Substructure of sidearms on squid axoplasmic vesicles and microtubules visualized by negative contrast electron microscopy

We present a high-resolution electron microscopic study of the sidearms on microtubules and vesicles that are suggested to form the crossbridges which produce the microtubule-based vesicle transport in squid axoplasm. The sidearms were found attached to the surfaces of the anterogradely transported vesicles in the presence of ATP. These sidearms were made of one to three filaments of uniform diameter. Each filament measured 5-6 nm in width and 30-35 nm in length. The filaments in some of the sidearms had splayed apart by pivoting at their base, thereby assuming a "V" shape. The spread configuration illustrated the independence of the individual filaments. The filaments in other sidearms were closely spaced and oriented parallel to each other, a pattern called the compact configuration. In axoplasmic buffer containing AMP-PNP, structures indistinguishable from the filaments of the sidearms on the vesicles were observed attached to microtubules. Pairs of filaments, thought to represent the basic functional unit, were observed attached to adjacent protofilaments of the microtubules by their distal tips. These data support a model of vesicle movement in which a pair of filaments within a sidearm forms two crossbridges and moves a vesicle by "walking" along the protofilaments of the microtubule.     

Assembly and three-dimensional image reconstruction of tubulin hoops

The three-dimensional structure of tubulin hoops has been determined by image reconstruction. The surface lattice of hoops is similar to that of microtubules, but in addition hoops possess a superstructure of protofilament triplets. The protofilaments differ mainly in their apparent volumes and lateral spacings. The volumes depend strongly on the orientation on the carbon support, while the spacings do not. The differences of appearance do not reflect changes of intrinsic subunit structure. They are explained by differential staining related to the orientation and packing of protofilament. Microtubule-associated proteins do not contribute to the average subunit structure. All apparent protofilament structures differ from that expected from X-ray patterns of microtubules in terms of subunit tilt and distribution of contrast. It is concluded that the negatively stained structure is a reliable representation of the arrangement of protein subunits, but not of their shape. Tubulin hoops occur in conditions of microtubule assembly near the critical concentration in a stabilizing buffer. Their formation depends on microtubule-associated proteins and on the initial presence of tubulin oligomers, which may associate into short protofilament triplets. If their elongation is rapid compared to lateral aggregation, they form closed hoops. The growth phase is followed by a redistribution phase, during which hoops disappear in favour of microtubules. This behaviour is explained by kinetic overshoot assembly. Each triplet resembles an incomplete microtubule wall so that the junction between two triplets may be compared to a junction between microtubule walls. Such junctions are formed by a closely spaced pair of protofilaments. They are analogous to junctions between microtubules and incomplete microtubule walls, and they have the same clockwise curvature when viewed at the growing end.     

Structural microtubule cap: stability,catastrophe, rescue,and third state

Microtubules polymerize from GTP-liganded tubulin dimers, but are essentially made of GDP-liganded tubulin. We investigate the tug-of-war resulting from the fact that GDP-liganded tubulin favors a curved configuration, but is forced to remain in a straight one when part of a microtubule. We point out that near the end of a microtubule, the proximity of the end shifts the balance in this tug-of-war, with some protofilament bending as result. This somewhat relaxes the microtubule lattice near its end, resulting in a structural cap. This structural cap thus is a simple mechanical consequence of two well-established facts: protofilaments made of GDP-liganded tubulin have intrinsic curvature, and microtubules are elastic, made from material that can yield to forces, in casu its own intrinsic forces. We explore possible properties of this structural cap, and demonstrate 1) how it allows both polymerization from GTP-liganded tubulin and rapid depolymerization in its absence; 2) how rescue can occur; 3) how a third, meta-stable intermediate state is possible and can explain some experimental results; and 4) how the tapered tips observed at polymerizing microtubule ends are stabilized during growth, though unable to accommodate a lateral cap. This scenario thus supports the widely accepted GTP-cap model by suggesting a stabilizing mechanism that explains the many aspects of dynamic instability.