A HISTORICAL OVERVIEW OF TUBULIN AND MICROTUBULE STRUCTURE

Eva Nogales
Molecular and Cell Biology, UC Berkeley
and Lawrence Berkeley National Laboratory
Berkeley, CA 94720-3200

Microtubules were first observed in the 1950's in electron microscopy studies of cell sections (Manton and Clarke, 1952; Fawcet and Porter, 1954). During the next 20 years microtubule preservation and contrast were improved with the use of different fixatives and embedding media (Sabatini et al., 1963; Tilney et al., 1973), showing that these polymers are ubiquitous and that they are composed of 13 protofilaments (Tilney et al., 1973). Denaturation studies of purified microtubules showed them to be composed of αβ tubulin heterodimers (Bryan and Wilson, 1971). With the first in vitro assembly of purified tubulin (Weisenberg, 1972) came the opportunity to study the ultrastructure of microtubules. Electron microscopy studies and helical image analysis of negatively stained flagellar microtubule doublets (Amos and Klug, 1974), and reconstituted microtubules from brain homogenates (Erickson, 1974), showed that the microtubule was made of 13 parallel protofilaments, giving rise to a polar polymer, and that the tubulin subunits were organized in a helical 3-start lattice within the microtubule wall. Two different arrangements, the A lattice (where lateral contacts between tubulin subunits are heterologous αa-β and β-α and the B lattice (where the contacts are homologous, α-α, β-β), were proposed for the A and B tubules in the flagellar doublets, respectively (Amos and Klug, 1974). The A lattice warranted a helical lattice of dimers and was also assumed to be the correct model for reconstituted single microtubules (Erickson, 1974). Using sections of glutaraldehyde and tannin fixed reconstituted microtubules, it was shown that microtubules could have different number of protofilament and that their distribution depended on the polymerizing conditions (Pierson et al., 1978). Similar diversity has also been observed in vivo (Eichenlaub-Ritter and Tucker, 1984; Dallai and Afzelius, 1990).

With the discovery of a large sheet polymer of tubulin obtained in vitro in the presence of zinc ions (Larsson et al., 1976), it became possible to study tubulin using the newly developed method of Unwin and Henderson for 2-dimentional crystal lattices (Unwin and Henderson, 1975). The hope for an increase in resolution from that obtained from microtubules (~20-25 Å) did not materialized due to the limited size of the sheets and the shortcomings of negative stain (Amos and Baker, 1979; Tamm et al., 1979). A different approach to characterize the structure of the microtubule was X-ray fiber diffraction. Using gels of oriented microtubules, a 25 Å resolution reconstruction of the microtubule showed that the outside surface was subdivided by vertical grooves separating the 13 protofilaments, while the inside surface was dominated by 10-start grooves (Mandelkow et al., 1977). A 18-Å resolution structure of the microtubule obtained by combining X-ray diffraction data with phase information from electron micrographs proposed a tubulin subunit divided into three, well separated domains (Beese et al., 1987).

cyto_figure A big step forward in the study of microtubules came with the use of cryo-electron microscopy. The use of frozen-hydrated microtubules proved very successful in preserving the integrity of the microtubule wall, and the potential of cryo-techniques to characterize microtubule assembly and disassembly (Mandelkow and Mandelkow, 1985; Murray, 1986). The exquisite preservation of the microtubule lattice allowed the determination of the number of protofilaments from the moiré pattern originated by the superposition of the two sites of the microtubule wall in projection. This led to the proposition of the lattice accommodation theory for microtubule closure (Wade et al., 1990): different protofilament numbers result in a skew with respect to the microtubule, while the surface lattice remains unchanged. Detailed analysis of microtubule images revealed that the protofilament number can change even within a single microtubule (Chrétien et al., 1992). Cryo-electron microscopy of depolymerizing microtubules showed that these polymers come apart by curling and peeling of protofilaments (Mandelkow et al., 1991). Studies of fast growth from centrosomes, on the other hand, showed that microtubules grow by elongation of two-dimensional sheets that close into a cylinder at variable rates (Chrétien and Wade, 1991).

With the expression of recombinant motor proteins, an explosion of new structural studies on microtubules became possible. Recombinant kinesin motor domain was use to decorate microtubules and doublets showing that all microtubules have a B lattice (Song and Mandelkow, 1993). This meant that microtubules with 13 protofilaments, the most abundant in vivo, must have a seam where lateral contacts involve heterologous subunits. Such seam has been directly visualized, both in vivo and in vitro, using freeze-fracture replicas (Kikkawa et al., 1994). As kinesin binds preferentially to the β subunit, observation of the binding pattern of kinesin head at the end of opened up microtubules of known polarity contains information on which subunit crowns each end of the microtubule. After some controversy, the general agreement is that β-tubulin crowns the plus end, while α-tubulin crowns the minus end (Hirose et al., 1995; Hoenger et al., 1995; Kikkawa et al., 1995; Song and Mandelkow, 1995; Hoenger and Milligan, 1996). This was further supported by the finding that antibodies specific for α-tubulin decorated only the minus end of microtubules (Fan et al., 1996). During the last three years cryo-electron microscopy studies have concentrated on the reconstruction of microtubules decorated with different motors, both monomeric and dimeric, and in different nucleotide states (Hirose et al., 1995; Hoenger et al., 1995; Kikkawa et al., 1995; Arnal et al., 1996; Hirose et al., 1996; Sosa and Milligan, 1996; Hoenger and Milligan, 1997; Sosa et al., 1997; Arnal and Wade, 1998; Hirose et al., 1998; Hoenger et al., 1998).

Additional structural information on microtubules has come from X-ray solution scattering studies. Time-resolved studies following polymerization and depolymerization of microtubules have resulted in models for the structural pathways of these processes (Mandelkow et al., 1980; Mandelkow et al., 1988; Diaz et al., 1996). Models both for the structure of microtubules and GDP-tubulin rings, to about 30 Å resolution, have also been obtained by this method (Andreu et al., 1992; Andreu et al., 1994; Díaz et al., 1994)

In addition to more direct methods, biochemistry approaches also lead to structural models of the tubulin structure. Between them is the two-domain model for the contacts at intra and interdimer interfaces obtained using proteolysis and crosslinking (Kirchner and Mandelkow, 1985). Very extensive proteolysis used to map the tubulin dimer and taxol-stabilized microtubules, identified regions in the molecule that become protected after polymerization (de Pereda and Andreu, 1996). Site directed antibodies have also been used to identify regions of the structure exposed to the solvent in dimers and microtubules (Andreu, 1993).

In the last 5 years a continued progress towards a high resolution structure of tubulin using cryo-electron crystallography of zinc-induced sheets (Downing and Jontes, 1992; Nogales et al., 1995), has finally lead to the first atomic model of tubulin (Nogales et al., 1998a). The 3.7 Å model ( Fig. 1) shows that the protein is made of three sequential domains: an N-terminal, nucleotide-binding domain; a smaller second domain containing the binding site of taxol; and two C-terminal helices that form the crest of the protofilament on the outside surface of the microtubule (Nogales et al., 1998c). The fold and nucleotide binding motif of tubulin differ very clearly from those of classical GTPases while resembling that of dinucleotide binding proteins with a Rossmann fold (Nogales et al., 1998b). Both the N-terminal and second domain of tubulin are however very similar to those of the bacterial protein FtsZ (Löwe and Amos, 1998). Tubulin and FtsZ form a new family of GTP hydrolyzing proteins that contain their own GTPase activating motif (Nogales et al., 1998b).

Note: The present article in no way pretends to be a comprehensive review on the subject. We apologize to all those who have made important contributions to our understanding of microtubule and tubulin structure but are not mentioned here due to lack of space.

  • Kikkawa, M., Ishikawa, T., Nakata, T., Wakabayashi, T. and Hirokawa, N. (1994). "Direct visualization of the microtubule lattice seam both in vitro and in vivo." J. Cell Biol. 127, 1965-1971.
  • Kikkawa, M., Ishikawa, T., Wakabayashi, T. and Hirokawa, N. (1995). "Three-dimensional structure of the kinesin head-microtubule complex." Nature 376, 274-277.
  • Kirchner, K. and Mandelkow, E.-M. (1985). "Tubulin domains responsible for assembly of dimers and protofilaments." EMBO J. 4(9), 2397-2402.
  • Larsson, M., Wallin, M. and Edstrom, A. (1976). "Induction of a sheets polymer of tubulin by Zn2+." Exp. Cell Res.
  • Löwe, J. and Amos, L. A. (1998). "Crystal structure of the bacterial cell division protein FtsZ." Nature 391, 203-206.
  • Mandelkow, E., Thomas, J. and Cohen, C. (1977). "Microtubule structure at low resolution by x-ray diffraction." Proc. Natl. Acad. Sci. USA 74(8), 3370-3374.
  • Mandelkow, E.-M., Harmsen, A., Mandelkow, E. and Bordas, J. (1980). "X-ray kinetic studies of microtubule assembly using synchroteon radiation." Nature 287, 595-599.
  • Mandelkow, E.-M., Lange, G., Jangla, A., Spann, U. and Mandelkow, E. (1988). "Dynamics of the microtubule oscillator: role of nucleotides and tubulin-MAP interactions." EMBO J. 7(2), 357-365.
  • Mandelkow, E.-M. and Mandelkow, E. (1985). "Unstained microtubules studied by cryoelectron microscopy. Substructure, supertwist and diassembly." J. Mol. Biol. 181, 123-135.
  • Mandelkow, E.-M., Mandelkow, E. and Milligan, R. A. (1991). "Microtubules dynamics and microtubules caps: a time-resolved cryo-electron microscopy study." J. Cell Biol. 114(5), 977-991.
  • Manton, I. and Clarke, R. (1952). "An electron microscopy study of the spermatozoid of sphagnum." J. Exp. Bot. 3, 265-275.
  • Murray, J. M. (1986). "Electron microscopy of frozen hydrated eukariotic flagella." J. Ultrastruct. Mol. Struct. Res.(196-209).
  • Nogales, E., Wolf, S. G. and Downing, K. H. (1998a). "Structure of the ab tubulin dimer by electron crystallography." Nature 391, 199-203.
  • Nogales, E., Downing, K. H., Amos, L. A. and Löwe, J. (1998b). "Tubulin and FtsZ form a distinct family of GTPases." Nature Struc. Biol. 5, 451-458.
  • Nogales, E., Whittaker, M., Milligan, R. A. and Downing, K. H. (1998c). "High resolution structure of the microtubule." submitted.
  • Nogales, E., Wolf, S. G., Khan, I. A., Ludue&#241a, R. F. and Downing, K. H. (1995). "Structure of tubulin at 6.5 Âand location of the taxol-binding site." Nature 375, 424-427.
  • Pierson, G. B., Burton, P. R. and Himes, R. H. (1978). "Alterations in number of protofilaments in microtubules assembled in vitro." J. Cell Biol. 76, 223-228.
  • Sabatini, D. D., Bensch, K. and Barrnett, R. J. (1963). "Cytochemistry and electron microscopy: the preservation of cellular ultrastructure and enzymatic activity by aldehyde fixation." J. Cell Biol. 17, 19-58.
  • Song, Y.-H. and Mandelkow, E. (1993). "Recombinant kinesin motor domain binds to b-tubulin and decorates microtubules with a B surface lattice." Proc. Natl. Acad. Sci. USA 99, 1671-1675.
  • Song, Y. H. and Mandlekow, E. (1995). "The anatomy of flagellar microtubules: polarity, seams, junctions, and lattice." J. Cell Biol. 128, 81-94.
  • Sosa, H., Dias, D. P., Hoenger, A., Whittaker, M., Wilson-Kubalek, E., Sablin, E., Fletterick, R. J., Vale, R. D. and Milligan, R. A. (1997). "A model for the microtubule-Ncd motor protein complex obtained by cryo-electron microscopy and image analysis." Cell 90, 217-224.
  • Sosa, H. and Milligan, R. A. (1996). "Three-dimensional structure of ncd-decorated microtubules obtained by a back-projection method." J. Mol. Biol. 260, 743-755.
  • Tamm, L. K., Crepeau, R. H. and Edelstein, S. J. (1979). "Three-dimensional reconstruction of tublin in zinc-induced sheets." J. Mol. Biol. 130, 473-492.
  • Tilney, L. G., Bryan, J., Bush, D. J., Fujiwara, K., Mooseker, M. S., Murphy, D. B. and Snyder, D. H. (1973). "Microtubules: evidence for 13 protofilaments." J. Cell Biol. 59, 267-275.
  • Unwin, P. N. T. and Henderson, R. (1975). "Molecular structure determination by electron microscopy of unstained crystalline specimens." J. Mol. Biol. 94, 425-440.
  • Wade, R. H., Chr&#233tien, D. and Job, D. (1990). "Characterization of microtubule protofilament numbers. How does the surface lattice accomodate?" J. Mol. Biol. 212, 775-786.
  • Weisenberg, R. C. (1972). "Microtubule formation in vitro in solutions containing low calcium concentrations." Science 177, 1104-1105.