Bruce McEwen Laboratory
Kinetochore Microtubule Dynamics
McEwen, B.F., Barnard, R.M., Portuese, T., and Hsieh, C-E. (2002)
Microtubule assembly requires binding but not hydrolysis of GTP. Rather GTP hydrolysis is required to destabilize the microtubule lattice prior to disassembly. Dynamic instability refers to the stochastic switching of microtubules between the growing and shrinking states. This transition is accompanied by a series of distinct conformational changes at the microtubule ends. During mitosis, microtubule plus-ends are embedded in the kinetochore and movement of a kinetochore towards and away from its attached spindle pole is coordinated with the shrinkage and growth of its bound microtubules. Thus, kinetochore movement must be coordinated with microtubule plus-end dynamics, implying that the kinetochore controls microtubule dynamics. Little is known about the mechanism of this control.
The distinct plus-end conformations associated with microtubule dynamics can be detected at the kinetochore using electron tomography. Currently we are developing high-throughput methods that will enable us to make statistical correlations between plus-end conformations and the stage of mitosis, or the direction of chromosome motion. These tools will then be used to determine the effects of pharmacological perturbations and depletion of select molecular components, and to test current hypotheses concerning how sites of kinetochore attachments modify microtubule dynamics during chromosome motion.
Figure 1: Movie illustration of mitosis. Chromosomes are represented in blue, spindle microtubules and spindle poles are green, and sister kinetochores appear as small red bars on opposite sides of each chromosome. Note the bioriented attachment of chromosomes and frequent changes in the direction of chromosome (kinetochore) movement. This movie is part of a mitosis demonstration created for the New York Hall of Science, a major science museum in Queens, NY.
Figure 2: Schematic representation of "typical" chromosome motions. M - monooriented, characterized by oscillatory motion; C - congression, characterized by movement away from the attached spindle pole by the trailing kinetochore and movement toward the attached spindle pole by the leading kinetochore; B - bioriented and congressed, characterized by oscillatory motion; A - anaphase, characterized by poleward movement of all kinetochores. G refers to a short rapid poleward glide that often occurs during initial monooriented attachment. Taken from Skibbens et. al. (1993) J. Cell Biol. 122:859-875.
A key feature of chromosome alignment is that individual kinetochores move both toward and away from their attached spindle poles. Thus, during anaphase all kinetochores primarily move towards the spindle pole, while during congression trailing kinetochores primarily move away from the attached spindle pole. When chromosomes are monooriented, or aligned at the spindle equator, individual kinetochores often oscillate between movement toward and away from the pole. Kinetochores also exhibit a pause state with little or no movement. Thus, a comprehensive molecular description of chromosome motion must explain three states of kinetochore movement: 1) towards the attached spindle pole; 2) away from the attached spindle pole; and 3) pause.
Figure 3: GTP-Cap Model for Microtubule Dynamic Instability. The polymerization phase is thought to be stabilized by a thin cap of tubulin dimers at the microtubule plus-end that have a GTP molecule associated with the beta-tubulin subunit. Stochastic loss of the GTP cap, due to hydrolysis or subunit loss, results in a transition to the depolymerizing phase known as catastrophe. Back transition (rescue) is also observed. Taken from Inoue and Salmon (1995) Mole. Biol. Cell 6:1619-1640.
Figure 4: Structural Cap Model for Microtubule Dynamic Instability. Cryo- and negative stain-electron microscopy of in vitro assembled microtubules have revealed that plus end conformational changes accompany transitions between the polymerization and depolymerization phases of dynamic instability. These studies have also identified a blunt end transitional state conformation that is capable of growth or transition into the shrinking phase. The structural cap model postulates that stochastic closing of the open sheet conformation leads to GTP hydrolysis and possible catastrophe transitions. Taken from Arnal, et. al., (2000) J. Cell Biol. 149:767-774.
Current Models for Kinetochore Attachment to Microtubule Plus-Ends. Most current models assume that the kinetochore attaches to the "tube" portion of the microtubule plus ends rather than to the very tip. This would allow conformational transitions to occur without interference from kinetochore attachments. Stabilization of kinetochore microtubules is thought to arise from the attachments preventing propagation of structural transitions associated with the depolymerization phase. Currently there is no direct evidence to support "tube portion" binding by the kinetochore.
Figure 5: Using electron tomography to investigate the plus-ends of kinetochore microtubules. (a) A 2D projection image from the tomogrphic tilt series. Section thickness is 250 nm. Protein and chromatin appear white in a-c. The chromosome, kinetochore, and a region containing microtubules are indicated. All specimens used for tomographic reconstruction were prepared by high-pressure freezing and freeze substitution (see Kinetochore Structure). (b) Single 2.0 nm thick slice from the tomographic reconstruction of the tilt data represented in (a). The plus-end of a single microtubule is outlined. Only a short segment of the microtubule is visible because its axis is not coplanar with the slice. (c) Slice from a 3D window containing the microtubule indicated in (b). The reconstruction volume was reoriented before the window was taken so that the microtubule cylindrical axis now lies vertically in the plane of the slice. As a result, a much longer segment of microtubule is visible. (d) Plus-end extraction. Contours of the microtubule boundary were manually traced in green on slices taken along the microtubule axis (i.e., orthogonal to the slice in c). Contours from successive slices were extracted from the reconstruction volume, stacked, and filled to produce the model. (e) Plus-end with connections. Kinetochore attachments were traced in red contours, extracted, and filled. Note that c and d are 3D models that can be viewed from all directions. The IMOD software package was used for all manual segmentation (Kremer, et. al. (1996) J. Structural Biol. 116:71-76).
Figure 6: Gallery of Segmented Kinetochore Microtubules. (a) Control Microtubules. For controls, segmentation was carried out on stretches of the main body of the microtubules, significantly removed from the plus-end. As expected, the microtubules are generally solid with occasional "holes" or "groves" in the sides. (b) Kinetochore Microtubule Plus-Ends. Note that both the blunt end and frayed end conformations were detected. A surprising feature is the high degree of fragmentation present in some of the plus-ends, even those in the blunt-end conformation. (c) Kinetochore Microtubule Plus-Ends with Attachments. Contrary to popular models, kinetochores frequently do bind to the tips of kinetochore microtubule plus-ends. Many of the attachments appear to "wrap around" microtubules, binding over a span of several subunits rather than binding at discrete points.
Conclusion
These characteristics indicate that the kinetochore is directly involved in modifying the conformational transitions of kinetochore microtubule plus-ends. Thus, current data imply that kinetochore exert some degree of control over dynamic instability.