Minus-end-directed motion of kinesin-coated microspheres driven by microtubule depolymerization.

Dynamic changes in microtubule (MT) length have long been thought to contribute to intracellular motility. Both the polymerization and depolymerization of tubulin have been shown to do work in vitro, but the biochemical complexity of objects moved, such as chromosomes, has complicated the identification of proteins that couple MT dynamics with motility. Work with MTs grown from and tethered to pellicles of lysed Tetrahymena has shown that disassembly-dependent movement of chromosomes in vitro can be inhibited with antibodies against the motor domain of kinesin. To study proteins that can function in disassembly-dependent motion, we have refined this motility assay, replacing chromosomes with protein-coated latex microspheres. We report here the ability of several enzymes, including kinesin, to support in vitro motility of latex microspheres on disassembling MTs (Fig. 1a). The polarity of kinesin's motor activity can be reversed by MT disassembly and interactions between a motor and a MT end can either slow or speed the rate of tubulin depolymerization.

Antibodies to the kinesin motor domain and CENP-E inhibit microtubule depolymerization-dependent motion of chromosomes in vitro.

Chromosomes can move with the ends of depolymerizing microtubules (MTs) in vitro, even in the absence of nucleotide triphosphates (Coue, M., V. A. Lombillo, and J. R. McIntosh. 1991. J. Cell Biol. 112:1165-1175.) Here, we describe an immunological investigation of the proteins important for this form of motility. Affinity-purified polyclonal antibodies to kinesin exert a severe inhibitory effect on depolymerization-dependent chromosome motion. These antibodies predominantly recognize a polypeptide of M(r) approximately 250 kD on immunoblots of CHO chromosomes and stain kinetochores as well as some vesicles that are in the chromosome preparation. Antibodies to CENP-E, a kinetochore-associated kinesin-like protein, also recognize a 250-kD electrophoretic component, but they stain only the kinetochroe region of isolated chromosomes. Polyclonal antibodies that recognize specific domains of the CENP-E polypeptide affect MT disassembly-dependent chromosome motion in different ways; antibodies to the head or tail portions slow motility threefold, while those raised against the neck region stop motion completely. Analogous antibodies that block conventional, ATP-dependent motility of cytoplasmic dynein (Vaisberg, G., M. P. Koonce, and J. R. McIntosh. 1993. J. Cell Biol. 123:849-858) have no effect on disassembly-dependent chromosome motion, even though they bind to kinetochores. These observations suggest that CENP-E helps couple chromosomes to depolymerizing MTs. A similar coupling activity may allow spindle MTs to remain kinetochore-bound while their lengths change during both prometaphase and anaphase A.

A plus-end-directed motor enzyme that moves antiparallel microtubules in vitro localizes to the interzone of mitotic spindles.

Mitosis comprises a complex set of overlapping motile events, many of which involve microtubule-dependent motor enzymes. Here we describe a new member of the kinesin superfamily. The protein was originally identified as a spindle antigen by the CHO1 monoclonal antibody and shown to be required for mitotic progression. We have cloned the gene that encodes this antigen and found that its sequence contains a domain with strong sequence similarity to the motor domain of kinesin-like proteins. The product of this gene, expressed in bacteria, can cross-bridge antiparallel microtubules in vitro, and in the presence of Mg-ATP, microtubules slide over one another in a fashion reminiscent of microtubule movements during spindle elongation.

Microtubule depolymerization promotes particle and chromosome movement in vitro.

We have developed a system for studying the motions of cellular objects attached to depolymerizing microtubules in vitro. Radial arrays of microtubules were grown from lysed and extracted Tetrahymena cells attached to a glass coverslip that formed the top of a light microscope perfusion chamber. A preparation of chromosomes, which also contained vesicles, was then perfused into the chamber and allowed to bind to the microtubule array. The concentration of tubulin was then reduced by perfusing buffer that lacked both tubulin and nucleotide triphosphates, and the resulting microtubule depolymerization was observed by light microscopy. A fraction of the bound objects detached in the flow and washed away, while others stabilized the microtubules to which they were bound. Some of the particles and chromosomes, however, moved in toward the Tetrahymena ghost as their associated microtubules shortened. The mean speeds for particles and chromosomes were 26 +/- 20 and 15 +/- 12 microns/min, respectively. These motions occurred when nucleotide triphosphate levels were very low, as a result of either dilution or by the action of apyrase. Furthermore, the motions were unaffected by 100 microM sodium orthovanadate, suggesting that these forces are not the result of ATP hydrolysis by a minus end-directed mechanoenzyme. We conclude that microtubule depolymerization provided the free energy for the motions observed. All the objects that we studied in detail moved against a stream of buffer flowing at approximately 100 microns/s, so that the force being developed was at least 10(-7) dynes. This force is large enough to contribute to some forms of motility in living cells.