Single-molecule analysis of kinesin motility reveals regulation by the cargo-binding tail domain.

Conventional kinesin transports membranes along microtubules in vivo, but the majority of cellular kinesin is unattached to cargo. The motility of non-cargo-bound, soluble kinesin may be repressed by an interaction between the amino-terminal motor and carboxy-terminal cargo-binding tail domains, but neither bead nor microtubule-gliding assays have shown such inhibition. Here we use a single-molecule assay that measures the motility of kinesin unattached to a surface. We show that full-length kinesin binds microtubules and moves about ten times less frequently and exhibits discontinuous motion compared with a truncated kinesin lacking a tail. Mutation of either the stalk hinge or neck coiled-coil domain activates motility of full-length kinesin, indicating that these regions are important for tail-mediated repression. Our results suggest that the motility of soluble kinesin in the cell is inhibited and that the motor becomes activated by cargo binding.

Millennial musings on molecular motors.

In a universe that is dominated by increasing entropy, living organisms are a curious anomaly. The organization that distinguishes living organisms from their inanimate surroundings relies upon their ability to execute vectorial processes, such as directed movements and the assembly of macromolecules and organelle systems. Many of these phenomena are executed by molecular motors that harness chemical potential energy to perform mechanical work and unidirectional motion. This article explores how these remarkable protein machines might have evolved and what roles they could play in biological and medical research in the coming decades.

Formation of membrane networks in vitro by kinesin-driven microtubule movement.

Certain intracellular organelles such as the endoplasmic reticulum (Terasaki, M., L. B. Chen, and K. Fujiwara. 1986. J. Cell Biol. 103:1557-1568) and lysosomes (Swanson, J., A. Bushnell, and S. C. Silverstein. Proc. Natl. Acad. Sci. USA. 84:1921-1925) form tubular networks that are closely aligned with microtubules. Here we describe the formation of polygonal networks composed of interconnected membrane tubules that occurs when a preparation of microtubule affinity-purified squid kinesin is combined with microtubules and ATP on a glass surface. The membrane, which is a minor contaminant in the microtubule affinity-purified kinesin preparation, binds to microtubules translocating along kinesin-coated glass surfaces. Force exerted by kinesin upon the microtubule is transmitted to the membrane and a tubular extension of the membrane is produced. As the membrane tubule elongates, membrane tension exerts an opposing force upon the translocating microtubule that can alter its direction of movement by dissociating or partially dissociating the microtubule from the kinesin-coated surface. Membrane tubules that come in contact appear to fuse with one another, and thus give rise to two-dimensional polygonal networks of tubules that have similar features to endoplasmic reticulum networks in cells. Artificial liposomes composed of dimyristoylphosphatidylcholine and yolk phosphatidylglycerol also form stable tubular structures when subjected to shear forces, but do not interact with microtubules or form polygonal networks, suggesting that such phenomena may require membrane-associated proteins. These findings indicate that kinesin generates sufficient force to form tubular membrane extensions in vitro and suggest that this microtubule-based motility protein may also be responsible for creating tubular membrane networks within cells.

Controlling kinesin by reversible disulfide cross-linking. Identifying the motility-producing conformational change.

Conventional kinesin, a dimeric molecular motor, uses ATP-dependent conformational changes to move unidirectionally along a row of tubulin subunits on a microtubule. Two models have been advanced for the major structural change underlying kinesin motility: the first involves an unzippering/zippering of a small peptide (neck linker) from the motor catalytic core and the second proposes an unwinding/rewinding of the adjacent coiled-coil (neck coiled-coil). Here, we have tested these models using disulfide cross-linking of cysteines engineered into recombinant kinesin motors. When the neck linker motion was prevented by cross-linking, kinesin ceased unidirectional movement and only showed brief one-dimensional diffusion along microtubules. Motility fully recovered upon adding reducing agents to reverse the cross-link. When the neck linker motion was partially restrained, single kinesin motors showed biased diffusion towards the microtubule plus end but could not move effectively against a load imposed by an optical trap. Thus, partial movement of the neck linker suffices for directionality but not for normal processivity or force generation. In contrast, preventing neck coiled-coil unwinding by disulfide cross-linking had relatively little effect on motor activity, although the average run length of single kinesin molecules decreased by 30-50%. These studies indicate that conformational changes in the neck linker, not in the neck coiled-coil, drive processive movement by the kinesin motor.

Cell cycle control of microtubule-based membrane transport and tubule formation in vitro.

When higher eukaryotic cells enter mitosis, membrane organization changes dramatically and traffic between membrane compartments is inhibited. Since membrane transport along microtubules is involved in secretion, endocytosis, and the positioning of organelles during interphase, we have explored whether the mitotic reorganization of membrane could involve a change in microtubule-based membrane transport. This question was examined by reconstituting organelle transport along microtubules in Xenopus egg extracts, which can be converted between interphase and metaphase states in vitro in the absence of protein synthesis. Interphase extracts support the microtubule-dependent formation of abundant polygonal networks of membrane tubules and the transport of small vesicles. In metaphase extracts, however, the plus end- and minus end-directed movements of vesicles along microtubules as well as the formation of tubular membrane networks are all reduced substantially. By fractionating the extracts into soluble and membrane components, we have shown that the cell cycle state of the supernatant determines the extent of microtubule-based membrane movement. Interphase but not metaphase Xenopus soluble factors also stimulate movement of membranes from a rat liver Golgi fraction. In contrast to above findings with organelle transport, the minus end-directed movements of microtubules on glass surfaces and of latex beads along microtubules are similar in interphase and metaphase extracts, suggesting that cytoplasmic dynein, the predominant soluble motor in frog extracts, retains its force-generating activity throughout the cell cycle. A change in the association of motors with membranes may therefore explain the differing levels of organelle transport activity in interphase and mitotic extracts. We propose that the regulation of organelle transport may contribute significantly to the changes in membrane structure and function observed during mitosis in living cells.

Microtubule translocation properties of intact and proteolytically digested dyneins from Tetrahymena cilia.

Tetrahymena cilia contain a three-headed 22S (outer arm) dynein and a single-headed 14S dynein. In this study, we have employed an in vitro assay of microtubule translocation along dynein-coated glass surfaces to characterize the motile properties of 14S dynein, 22S dynein, and proteolytic fragments of 22S dynein. Microtubule translocation produced by intact 22S dynein and 14S dynein differ in a number of respects including (a) the maximal velocities of movement; (b) the ability of 22S dynein but not 14S dynein to utilize ATP gamma S to induce movement; (c) the optimal pH and ionic conditions for movement; and (d) the effects of Triton X-100 on the velocity of movement. These results indicate that 22S and 14S dyneins have distinct microtubule translocating properties and suggest that these dyneins may have specialized roles in ciliary beating. We have also explored the function of the multiple ATPase heads of 22S dynein by preparing one- and two-headed proteolytic fragments of this three-headed molecule and examining their motile activity in vitro. Unlike the single-headed 14S dynein, the single-headed fragment of 22S dynein did not induce movement, even though it was capable of binding to microtubules. The two-headed fragment, on the other hand, translocated microtubules at velocities similar to those measured for intact 22S dynein (10 microns/sec). This finding indicates that the intact three-headed structure of 22S dynein is not essential for generating microtubule movement, which raises the possibility that multiple heads may serve some regulatory function or may be required for maximal force production in the beating cilium.

Directional instability of microtubule transport in the presence of kinesin and dynein, two opposite polarity motor proteins.

Kinesin and dynein are motor proteins that move in opposite directions along microtubules. In this study, we examine the consequences of having kinesin and dynein (ciliary outer arm or cytoplasmic) bound to glass surfaces interacting with the same microtubule in vitro. Although one might expect a balance of opposing forces to produce little or no net movement, we find instead that microtubules move unidirectionally for several microns (corresponding to hundreds of ATPase cycles by a motor) but continually switch between kinesin-directed and dynein-directed transport. The velocities in the plus-end (0.2-0.3 microns/s) and minus-end (3.5-4 microns/s) directions were approximately half those produced by kinesin (0.5 microns/s) and ciliary dynein (6.7 microns/s) alone, indicating that the motors not contributing to movement can interact with and impose a drag upon the microtubule. By comparing two dyneins with different duty ratios (percentage of time spent in a strongly bound state during the ATPase cycle) and varying the nucleotide conditions, we show that the microtubule attachment times of the two opposing motors as well as their relative numbers determine which motor predominates in this assay. Together, these findings are consistent with a model in which kinesin-induced movement of a microtubule induces a negative strain in attached dyneins which causes them to dissociate before entering a force-generating state (and vice versa); reversals in the direction of transport may require the temporary dissociation of the transporting motor from the microtubule. The bidirectional movements described here are also remarkably similar to the back-and-forth movements of chromosomes during mitosis and membrane vesicles in fibroblasts. These results suggest that the underlying mechanical properties of motor proteins, at least in part, may be responsible for reversals in microtubule-based transport observed in cells.

Role of the kinesin neck region in processive microtubule-based motility.

Kinesin is a dimeric motor protein that can move along a microtubule for several microns without releasing (termed processive movement). The two motor domains of the dimer are thought to move in a coordinated, hand-over-hand manner. A region adjacent to kinesin's motor catalytic domain (the neck) contains a coiled coil that is sufficient for motor dimerization and has been proposed to play an essential role in processive movement. Recent models have suggested that the neck enables head-to-head communication by creating a stiff connection between the two motor domains, but also may unwind during the mechanochemical cycle to allow movement to new tubulin binding sites. To test these ideas, we mutated the neck coiled coil in a 560-amino acid (aa) dimeric kinesin construct fused to green fluorescent protein (GFP), and then assayed processivity using a fluorescence microscope that can visualize single kinesin-GFP molecules moving along a microtubule. Our results show that replacing the kinesin neck coiled coil with a 28-aa residue peptide sequence that forms a highly stable coiled coil does not greatly reduce the processivity of the motor. This result argues against models in which extensive unwinding of the coiled coil is essential for movement. Furthermore, we show that deleting the neck coiled coil decreases processivity 10-fold, but surprisingly does not abolish it. We also demonstrate that processivity is increased by threefold when the neck helix is elongated by seven residues. These results indicate that structural features of the neck coiled coil, although not essential for processivity, can tune the efficiency of single molecule motility.

Alteration of binding properties and cytoskeletal attachment of nerve growth factor receptors in PC12 cells by wheat germ agglutinin.

Incubation of PC12 cells preloaded with 125I-nerve growth factor (NGF) reveals rapidly and slowly dissociating binding components indicative of a heterogeneous population of receptors. If the cells are previously exposed to wheat germ agglutinin (WGA) for 30 min, NGF now binds to an apparently homogeneous receptor population which exhibit slow dissociation kinetics. Total binding is also reduced by 50%. If WGA is added subsequent to 125I-NGF, total binding is not diminished, but rapidly dissociating receptors occupied with NGF are all converted to the slowly dissociating form. This conversion of receptors occurs rapidly, reaching completion within 2 min at 37 degrees or 4 degrees C, and is unaffected by metabolic energy poisons, suggesting that WGA-induced slowly dissociating receptors are not the product of internalization. The effects of the lectin are blocked by the sugar N-acetyl-D-glucosamine, and the lectin-induced slowly dissociating receptors are converted back to rapidly dissociating receptors by addition of this same sugar. WGA also affects the association of the NGF receptor with the Triton X-100 cytoskeleton. Greater than 90% of bound 125I-NGF becomes associated with Triton X-100 insoluble cytoskeletons in the presence of the lectin, compared with less than 20% before lectin addition. Cytoskeleton association of the NGF receptor by WGA shows similar kinetics as the conversion of rapidly to slowly dissociating receptors. This interaction may be involved in the alteration of NGF-receptor binding properties produced by this lectin.

Cell cycle regulation of dynein association with membranes modulates microtubule-based organelle transport.

Cytoplasmic dynein is a minus end-directed microtubule motor that performs distinct functions in interphase and mitosis. In interphase, dynein transports organelles along microtubules, whereas in metaphase this motor has been implicated in mitotic spindle formation and orientation as well as chromosome segregation. The manner in which dynein activity is regulated during the cell cycle, however, has not been resolved. In this study, we have examined the mechanism by which organelle transport is controlled by the cell cycle in extracts of Xenopus laevis eggs. Here, we show that photocleavage of the dynein heavy chain dramatically inhibits minus end-directed organelle transport and that purified dynein restores this motility, indicating that dynein is the predominant minus end-directed membrane motor in Xenopus egg extracts. By measuring the amount of dynein associated with isolated membranes, we find that cytoplasmic dynein and its activator dynactin detach from the membrane surface in metaphase extracts. The sevenfold decrease in membrane-associated dynein correlated well with the eightfold reduction in minus end-directed membrane transport observed in metaphase versus interphase extracts. Although dynein heavy or intermediate chain phosphorylation did not change in a cell cycle-dependent manner, the dynein light intermediate chain incorporated approximately 12-fold more radiolabeled phosphate in metaphase than in interphase extracts. These studies suggest that cell cycle-dependent phosphorylation of cytoplasmic dynein may regulate organelle transport by modulating the association of this motor with membranes.

Nucleotide specificities of anterograde and retrograde organelle transport in Reticulomyxa are indistinguishable.

Membrane-bound organelles move bidirectionally along microtubules in the freshwater ameba, Reticulomyxa. We have examined the nucleotide requirements for transport in a lysed cell model and compared them with kinesin and dynein-driven motility in other systems. Both anterograde and retrograde transport in Reticulomyxa show features characteristic of dynein but not of kinesin-powered movements: organelle transport is reactivated only by ATP and no other nucleoside triphosphates; the Km and Vmax of the ATP-driven movements are similar to values obtained for dynein rather than kinesin-driven movement; and of 15 ATP analogues tested for their ability to promote organelle transport, only 4 of them did. This narrow specificity resembles that of dynein-mediated in vitro transport and is dissimilar to the broad specificity of the kinesin motor (Shimizu, T., K. Furusawa, S. Ohashi, Y. Y. Toyoshima, M. Okuno, F. Malik, and R. D. Vale. 1991. J. Cell Biol. 112: 1189-1197). Remarkably, anterograde and retrograde organelle transport cannot be distinguished at all with respect to nucleotide specificity, kinetics of movement, and the ability to use the ATP analogues. Since the "kinetic fingerprints" of the motors driving transport in opposite directions are indistinguishable, the same type of motor(s) may be involved in the two directions of movement.

Engineering the processive run length of the kinesin motor.

Conventional kinesin is a highly processive molecular motor that takes several hundred steps per encounter with a microtubule. Processive motility is believed to result from the coordinated, hand-over-hand motion of the two heads of the kinesin dimer, but the specific factors that determine kinesin's run length (distance traveled per microtubule encounter) are not known. Here, we show that the neck coiled-coil, a structure adjacent to the motor domain, plays an important role in governing the run length. By adding positive charge to the neck coiled-coil, we have created ultra-processive kinesin mutants that have fourfold longer run lengths than the wild-type motor, but that have normal ATPase activity and motor velocity. Conversely, adding negative charge on the neck coiled-coil decreases the run length. The gain in processivity can be suppressed by either proteolytic cleavage of tubulin's negatively charged COOH terminus or by high salt concentrations. Therefore, modulation of processivity by the neck coiled-coil appears to involve an electrostatic tethering interaction with the COOH terminus of tubulin. The ability to readily increase kinesin processivity by mutation, taken together with the strong sequence conservation of the neck coiled-coil, suggests that evolutionary pressures may limit kinesin's run length to optimize its in vivo function.

Suppression of kinesin expression in cultured hippocampal neurons using antisense oligonucleotides.

Kinesin, a microtubule-based force-generating molecule, is thought to translocate organelles along microtubules. To examine the function of kinesin in neurons, we sought to suppress kinesin heavy chain (KHC) expression in cultured hippocampal neurons using antisense oligonucleotides and study the phenotype of these KHC "null" cells. Two different antisense oligonucleotides complementary to the KHC sequence reduced the protein levels of the heavy chain by greater than 95% within 24 h after application and produced identical phenotypes. After inhibition of KHC expression for 24 or 48 h, neurons extended an array of neurites often with one neurite longer than the others; however, the length of all these neurites was significantly reduced. Inhibition of KHC expression also altered the distribution of GAP-43 and synapsin I, two proteins thought to be transported in association with membranous organelles. These proteins, which are normally localized at the tips of growing neurites, were confined to the cell body in antisense-treated cells. Treatment of the cells with the corresponding sense oligonucleotides affected neither the distribution of GAP-43 and synapsin I, nor the length of neurites. A full recovery of neurite length occurred after removal of the antisense oligonucleotides from the medium. These data indicate that KHC plays a role in the anterograde translocation of vesicles containing GAP-43 and synapsin I. A deficiency in vesicle delivery may also explain the inhibition of neurite outgrowth. Despite the inhibition of KHC and the failure of GAP-43 and synapsin I to move out of the cell body, hippocampal neurons can extend processes and acquire as asymmetric morphology.

Rapid movements of vimentin on microtubule tracks: kinesin-dependent assembly of intermediate filament networks.

The assembly and maintenance of an extended intermediate filament (IF) network in fibroblasts requires microtubule (MT) integrity. Using a green fluorescent protein-vimentin construct, and spreading BHK-21 cells as a model system to study IF-MT interactions, we have discovered a novel mechanism involved in the assembly of the vimentin IF cytoskeleton. This entails the rapid, discontinuous, and MT-dependent movement of IF precursors towards the peripheral regions of the cytoplasm where they appear to assemble into short fibrils. These precursors, or vimentin dots, move at speeds averaging 0.55 +/- 0.24 micrometer/s. The vimentin dots colocalize with MT and their motility is inhibited after treatment with nocodazole. Our studies further implicate a conventional kinesin in the movement of the vimentin dots. The dots colocalize with conventional kinesin as shown by indirect immunofluorescence, and IF preparations from spreading cells are enriched in kinesin. Furthermore, microinjection of kinesin antibodies into spreading cells prevents the assembly of an extended IF network. These studies provide insights into the interactions between the IF and MT systems. They also suggest a role for conventional kinesin in the distribution of non-membranous protein cargo, and the local regulation of IF assembly.

Reconstitution of membrane transport powered by a novel dimeric kinesin motor of the Unc104/KIF1A family purified from Dictyostelium.

Motor-powered movement along microtubule tracks is important for membrane organization and trafficking. However, the molecular basis for membrane transport is poorly understood, in part because of the difficulty in reconstituting this process from purified components. Using video microscopic observation of organelle transport in vitro as an assay, we have purified two polypeptides (245 and 170 kD) from Dictyostelium extracts that independently reconstitute plus-end-directed membrane movement at in vivo velocities. Both polypeptides were found to be kinesin motors, and the 245-kD protein (DdUnc104) is a close relative of Caenorhabditis elegans Unc104 and mouse KIF1A, neuron-specific motors that deliver synaptic vesicle precursors to nerve terminals. A knockout of the DdUnc104 gene produces a pronounced defect in organelle transport in vivo and in the reconstituted assay. Interestingly, DdUnc104 functions as a dimeric motor, in contrast to other members of this kinesin subfamily, which are monomeric.

Purified plasma membranes inhibit polypeptide growth factor-induced DNA synthesis in subconfluent 3T3 cells.

Plasma membranes derived from NR-6 cells, a variant line of Swiss mouse 3T3 cells that does not have cell surface receptors for epidermal growth factor (EGF), inhibited EGF-induced stimulation of DNA synthesis by 50% in serum-starved, subconfluent 3T3 cells. Membranes derived from SV3T3 cells were much less effective in inhibiting EGF-induced DNA synthesis. This inhibition on DNA synthesis by NR-6 membranes was not a direct effect of membranes on EGF, nor could it be overcome by high concentrations of EGF. NR-6 membranes were most effective when added 3 h before EGF addition and had little effect when added 2 h or more after EGF. NR-6 membranes also reduced the stimulation of DNA synthesis induced by platelet-derived growth factor or fibroblast growth factor in serum-starved 3T3 cells. These findings indicate that membrane-membrane interactions between nontransformed cells may diminish their ability to proliferate in response to serum polypeptide growth factors.

Cloning and expression of a human kinesin heavy chain gene: interaction of the COOH-terminal domain with cytoplasmic microtubules in transfected CV-1 cells.

To understand the interactions between the microtubule-based motor protein kinesin and intracellular components, we have expressed the kinesin heavy chain and its different domains in CV-1 monkey kidney epithelial cells and examined their distributions by immunofluorescence microscopy. For this study, we cloned and sequenced cDNAs encoding a kinesin heavy chain from a human placental library. The human kinesin heavy chain exhibits a high level of sequence identity to the previously cloned invertebrate kinesin heavy chains; homologies between the COOH-terminal domain of human and invertebrate kinesins and the nonmotor domain of the Aspergillus kinesin-like protein bimC were also found. The gene encoding the human kinesin heavy chain also contains a small upstream open reading frame in a G-C rich 5' untranslated region, features that are associated with translational regulation in certain mRNAs. After transient expression in CV-1 cells, the kinesin heavy chain showed both a diffuse distribution and a filamentous staining pattern that coaligned with microtubules but not vimentin intermediate filaments. Altering the number and distribution of microtubules with taxol or nocodazole produced corresponding changes in the localization of the expressed kinesin heavy chain. The expressed NH2-terminal motor and the COOH-terminal tail domains, but not the alpha-helical coiled coil rod domain, also colocalized with microtubules. The finding that both the kinesin motor and tail domains can interact with cytoplasmic microtubules raises the possibility that kinesin could crossbridge and induce sliding between microtubules under certain circumstances.

Isolation of a ribonucleoprotein complex involved in mRNA localization in Drosophila oocytes.

Localization of bicoid (bcd) mRNA to the anterior and oskar (osk) mRNA to the posterior of the Drosophila oocyte is critical for embryonic patterning. Previous genetic studies implicated exuperantia (exu) in bcd mRNA localization, but its role in this process is not understood. We have biochemically isolated Exu and show that it is part of a large RNase-sensitive complex that contains at least seven other proteins. One of these proteins was identified as the cold shock domain RNA-binding protein Ypsilon Schachtel (Yps), which we show binds directly to Exu and colocalizes with Exu in both the oocyte and nurse cells of the Drosophila egg chamber. Surprisingly, the Exu-Yps complex contains osk mRNA. This biochemical result led us to reexamine the role of Exu in the localization of osk mRNA. We discovered that exu-null mutants are defective in osk mRNA localization in both nurse cells and the oocyte. Furthermore, both Exu/Yps particles and osk mRNA follow a similar temporal pattern of localization in which they transiently accumulate at the oocyte anterior and subsequently localize to the posterior pole. We propose that Exu is a core component of a large protein complex involved in localizing mRNAs both within nurse cells and the developing oocyte.

Nucleotide specificity of the enzymatic and motile activities of dynein, kinesin, and heavy meromyosin.

The substrate specificities of dynein, kinesin, and myosin substrate turnover activity and cytoskeletal filament-driven translocation were examined using 15 ATP analogues. The dyneins were more selective in their substrate utilization than bovine brain kinesin or muscle heavy meromyosin, and even different types of dyneins, such as 14S and 22S dynein from Tetrahymena cilia and the beta-heavy chain-containing particle from the outer-arm dynein of sea urchin flagella, could be distinguished by their substrate specificities. Although bovine brain kinesin and muscle heavy meromyosin both exhibited broad substrate specificities, kinesin-induced microtubule translocation varied over a 50-fold range in speed among the various substrates, whereas heavy meromyosin-induced actin translocation varied only by fourfold. With both kinesin and heavy meromyosin, the relative velocities of filament translocation did not correlate well with the relative filament-activated substrate turnover rates. Furthermore, some ATP analogues that did not support the filament translocation exhibited filament-activated substrate turnover rates. Filament-activated substrate turnover and power production, therefore, appear to become uncoupled with certain substrates. In conclusion, the substrate specificities and coupling to motility are distinct for different types of molecular motor proteins. Such nucleotide "fingerprints" of enzymatic activities of motor proteins may prove useful as a tool for identifying what type of motor is involved in powering a motility-related event that can be reconstituted in vitro.

The bipolar kinesin, KLP61F, cross-links microtubules within interpolar microtubule bundles of Drosophila embryonic mitotic spindles.

Previous genetic and biochemical studies have led to the hypothesis that the essential mitotic bipolar kinesin, KLP61F, cross-links and slides microtubules (MTs) during spindle assembly and function. Here, we have tested this hypothesis by immunofluorescence and immunoelectron microscopy (immunoEM). We show that Drosophila embryonic spindles at metaphase and anaphase contain abundant bundles of MTs running between the spindle poles. These interpolar MT bundles are parallel near the poles and antiparallel in the midzone. We have observed that KLP61F motors, phosphorylated at a cdk1/cyclin B consensus domain within the BimC box (BCB), localize along the length of these interpolar MT bundles, being concentrated in the midzone region. Nonphosphorylated KLP61F motors, in contrast, are excluded from the spindle and display a cytoplasmic localization. Immunoelectron microscopy further suggested that phospho-KLP61F motors form cross-links between MTs within interpolar MT bundles. These bipolar KLP61F MT-MT cross-links should be capable of organizing parallel MTs into bundles within half spindles and sliding antiparallel MTs apart in the spindle midzone. Thus we propose that bipolar kinesin motors and MTs interact by a "sliding filament mechanism" during the formation and function of the mitotic spindle.