Switch-based mechanism of kinesin motors

Kinesin motors are specialized enzymes that use hydrolysis of ATP to generate force and movement along their cellular tracks, the microtubules. Although numerous biochemical and biophysical studies have accumulated much data that link microtubule-assisted ATP hydrolysis to kinesin motion, the structural view of kinesin movement remains unclear. This study of the monomeric kinesin motor KIF1A combines X-ray crystallography and cryo-electron microscopy, and allows analysis of force-generating conformational changes at atomic resolution. The motor is revealed in its two functionally critical states-complexed with ADP and with a non-hydrolysable analogue of ATP. The conformational change observed between the ADP-bound and the ATP-like structures of the KIF1A catalytic core is modular, extends to all kinesins and is similar to the conformational change used by myosin motors and G proteins. Docking of the ADP-bound and ATP-like crystallographic models of KIF1A into the corresponding cryo-electron microscopy maps suggests a rationale for the plus-end directional bias associated with the kinesin catalytic core.     

How kinesin waits between steps

Kinesin-1 (conventional kinesin) is a dimeric motor protein that carries cellular cargoes along microtubules by hydrolysing ATP and moving processively in 8-nm steps. The mechanism of processive motility involves the hand-over-hand motion of the two motor domains ('heads'), a process driven by a conformational change in the neck-linker domain of kinesin. However, the 'waiting conformation' of kinesin between steps remains controversial-some models propose that kinesin adopts a one-head-bound intermediate, whereas others suggest that both the kinesin heads are bound to adjacent tubulin subunits. Addressing this question has proved challenging, in part because of a lack of tools to measure structural states of the kinesin dimer as it moves along a microtubule. Here we develop two different single-molecule fluorescence resonance energy transfer (smFRET) sensors to detect whether kinesin is bound to its microtubule track by one or two heads. Our FRET results indicate that, while moving in the presence of saturating ATP, kinesin spends most of its time bound to the microtubule with both heads. However, when nucleotide binding becomes rate-limiting at low ATP concentrations, kinesin waits for ATP in a one-head-bound state and makes brief transitions to a two-head-bound intermediate as it walks along the microtubule. On the basis of these results, we suggest a model for how transitions in the ATPase cycle position the two kinesin heads and drive their hand-over-hand motion.     

Movement of microtubules by single kinesin molecules

Kinesin is a motor protein that uses energy derived from ATP hydrolysis to move organelles along microtubules. Using a new technique for measuring the movement produced in vitro by individual kinesin molecules, it is shown that a single kinesin molecule can move a microtubule for several micrometers. New information about the mechanism of force generation by kinesin is presented.     

Single kinesin molecules studied with a molecular force clamp.

Kinesin is a two-headed, ATP-driven motor protein that moves processively along microtubules in discrete steps of 8 nm, probably by advancing each of its heads alternately in sequence. Molecular details of how the chemical energy stored in ATP is coupled to mechanical displacement remain obscure. To shed light on this question, a force clamp was constructed, based on a feedback-driven optical trap capable of maintaining constant loads on single kinesin motors. The instrument provides unprecedented resolution of molecular motion and permits mechanochemical studies under controlled external loads. Analysis of records of kinesin motion under variable ATP concentrations and loads revealed several new features. First, kinesin stepping appears to be tightly coupled to ATP hydrolysis over a wide range of forces, with a single hydrolysis per 8-nm mechanical advance. Second, the kinesin stall force depends on the ATP concentration. Third, increased loads reduce the maximum velocity as expected, but also raise the apparent Michaelis-Menten constant. The kinesin cycle therefore contains at least one load-dependent transition affecting the rate at which ATP molecules bind and subsequently commit to hydrolysis. It is likely that at least one other load-dependent rate exists, affecting turnover number. Together, these findings will necessitate revisions to our understanding of how kinesin motors function.     

Myosin VI is an actin-based motor that moves backwards.

Myosins and kinesins are molecular motors that hydrolyse ATP to track along actin filaments and microtubules, respectively. Although the kinesin family includes motors that move towards either the plus or minus ends of microtubules, all characterized myosin motors move towards the barbed (+) end of actin filaments. Crystal structures of myosin II (refs 3-6) have shown that small movements within the myosin motor core are transmitted through the 'converter domain' to a 'lever arm' consisting of a light-chain-binding helix and associated light chains. The lever arm further amplifies the motions of the converter domain into large directed movements. Here we report that myosin VI, an unconventional myosin, moves towards the pointed (-) end of actin. We visualized the myosin VI construct bound to actin using cryo-electron microscopy and image analysis, and found that an ADP-mediated conformational change in the domain distal to the motor, a structure likely to be the effective lever arm, is in the opposite direction to that observed for other myosins. Thus, it appears that myosin VI achieves reverse-direction movement by rotating its lever arm in the opposite direction to conventional myosin lever arm movement.     

Bead movement by single kinesin molecules studied with optical tweezers

Kinesin, a mechanoenzyme that couples ATP hydrolysis to movement along microtubules, is thought to power vesicle transport and other forms of microtubule-based motility. Here, microscopic silica beads were precoated with carrier protein, exposed to low concentrations of kinesin, and individually manipulated with a single-beam gradient-force optical particle trap ('optical tweezers') directly onto microtubules. Optical tweezers greatly improved the efficiency of the bead assay, particularly at the lowest kinesin concentrations (corresponding to approximately 1 molecule per bead). Beads incubated with excess kinesin moved smoothly along a microtubule for many micrometres, but beads carrying from 0.17-3 kinesin molecules per bead, moved, on average, only about 1.4 microns and then spontaneously released from the microtubule. Application of the optical trap directly behind such moving beads often pulled them off the microtubule and back into the centre of the trap. This did not occur when a bead was bound by an AMP.PNP-induced rigor linkage, or when beads were propelled by several kinesin molecules. Our results are consistent with a model in which kinesin detaches briefly from the microtubule during a part of each mechanochemical cycle, rather than a model in which kinesin remains bound at all times.     

Correlation between the ATPase and microtubule translocating activities of sea urchin egg kinesin

Coupling between ATP hydrolysis and microtubule movement was demonstrated several years ago in flagellar axonemes and subsequent studies suggest that the relevant microtubule motor, dynein, uses ATP to drive microtubule sliding by a cross-bridge mechanism analogous to that of myosin in muscles. Kinesin, a microtubule-based motility protein which may participate in organelle transport and mitosis, binds microtubules in a nucleotide-sensitive manner, and requires hydrolysable nucleotides to translocate microtubules over a glass surface. Recently, neuronal kinesin was shown to possess microtubule-activated ATPase activity although coupling between ATP hydrolysis and motility was not demonstrated. Here we report that sea urchin egg kinesin, prepared either with or without a 5'-adenylyl imidodiphosphate(AMPPNP)-induced microtubule binding step, also possesses significant microtubule-activated ATPase activity when Mg-ATP is used as a substrate. This ATPase activity is inhibited in a dose-dependent manner by addition of Mg-free ATP, by chelation of Mg2+ with EDTA, by addition of Na3VO4, or by addition of AMPPNP with or without Mg2+. Addition of these same reagents also inhibits the microtubule-translocating activities of sea urchin egg kinesin in a dose-dependent manner, supporting the hypothesis that kinesin-driven motility is coupled to the microtubule-activated Mg2+-ATPase activity.     

Cytoplasmic dynein functions as a gear in response to load

Cytoskeletal molecular motors belonging to the kinesin and dynein families transport cargos (for example, messenger RNA, endosomes, virus) on polymerized linear structures called microtubules in the cell. These 'nanomachines' use energy obtained from ATP hydrolysis to generate force, and move in a step-like manner on microtubules. Dynein has a complex and fundamentally different structure from other motor families. Thus, understanding dynein's force generation can yield new insight into the architecture and function of nanomachines. Here, we use an optical trap to quantify motion of polystyrene beads driven along microtubules by single cytoplasmic dynein motors. Under no load, dynein moves predominantly with a mixture of 24-nm and 32-nm steps. When moving against load applied by an optical trap, dynein can decrease step size to 8 nm and produce force up to 1.1 pN. This correlation between step size and force production is consistent with a molecular gear mechanism. The ability to take smaller but more powerful strokes under load--that is, to shift gears--depends on the availability of ATP. We propose a model whereby the gear is downshifted through load-induced binding of ATP at secondary sites in the dynein head.     

Quantized velocities at low myosin densities in an in vitro motility assay.

An in vitro motility assay has been developed in which single actin filaments move on one or a few heavy meromyosin (HMM) molecules. This movement is slower than when many HMM molecules are involved, in contrast to analogous experiments with microtubules and kinesin. Frequency analysis shows that sliding speeds distribute around integral multiples of a unitary velocity. This discreteness may be due to differences in the numbers of HMM molecules interacting with each actin filament, where the unitary velocity reflects the activity of one HMM molecule. The value of the unitary velocity predicts a step size of 5-20 nm per ATP, which is consistent with the conventional swinging crossbridge model for myosin function.     

Coalignment of vimentin intermediate filaments with microtubules depends on kinesin.

Intermediate filaments in most types of cultured cells coalign with microtubules. Depolymerization of microtubules results in collapse of vimentin and desmin intermediate filaments to the nucleus where they form a perinuclear cap. Collapse can also be induced by microinjection of antibodies against intermediate filament or microtubule proteins. Thus, two filament systems interact with each other. But the molecules mediating this interaction are unknown. One of the candidates for this role is a microtubule motor kinesin. Recent data showed that kinesin is involved in the plus end-directed movement of the membranous organelles along microtubules such as radial extension of lysosomes in macrophages and centrifugal movement of pigment in melanophores. Here we report that injection of the anti-kinesin antibody into human fibroblasts results in the redistribution of intermediate filaments to a tight perinuclear aggregate but had no effect on the distribution of microtubules. Thus, kinesin is involved not only in organelle movement but also in interaction of the two major cytoskeletal systems, intermediate filaments and microtubules.     

Two mitotic kinesins cooperate to drive sister chromatid separation during anaphase

During anaphase identical sister chromatids separate and move towards opposite poles of the mitotic spindle. In the spindle, kinetochore microtubules have their plus ends embedded in the kinetochore and their minus ends at the spindle pole. Two models have been proposed to account for the movement of chromatids during anaphase. In the 'Pac-Man' model, kinetochores induce the depolymerization of kinetochore microtubules at their plus ends, which allows chromatids to move towards the pole by 'chewing up' microtubule tracks. In the 'poleward flux' model, kinetochores anchor kinetochore microtubules and chromatids are pulled towards the poles through the depolymerization of kinetochore microtubules at the minus ends. Here, we show that two functionally distinct microtubule-destabilizing KinI kinesin enzymes (so named because they possess a kinesin-like ATPase domain positioned internally within the polypeptide) are responsible for normal chromatid-to-pole motion in Drosophila. One of them, KLP59C, is required to depolymerize kinetochore microtubules at their kinetochore-associated plus ends, thereby contributing to chromatid motility through a Pac-Man-based mechanism. The other, KLP10A, is required to depolymerize microtubules at their pole-associated minus ends, thereby moving chromatids by means of poleward flux.     

Mediation of meiotic and early mitotic chromosome segregation in Drosophila by a protein related to kinesin.

Contrary to the traditional view that microtubules pull chromosomes polewards during the anaphase stage of meiotic and mitotic cell divisions, new evidence suggests that the chromosome movements are driven by a motor located at the kinetochore. The process of chromosome segregation involves proper arrangement of kinetochores for spindle attachment, followed by spindle attachment and chromosome movement. Mechanisms in Drosophila for chromosome segregation in meiosis differ in males and females, implying the action of different gene products in the two sexes. A product encoded at the claret locus in Drosophila is required for normal chromosome segregation in meiosis in females and in early mitotic divisions of the embryo. Here we show that the predicted amino-acid sequence of this product is related to the heavy chain of kinesin. The conserved region corresponds to the kinesin motor domain and includes the ATP-binding site and a region that can bind microtubules. A second region contains a leucine repeat motif which may mediate protein-subunit interactions necessary for attachment of chromosomes to the spindle. The mutant phenotype of chromosome nondisjunction and loss, and its similarity to the kinesin ATP-binding domain, suggest that the product encoded at claret not only stabilizes chromosome attachments to the spindle, but may also be a motor that drives chromosome segregation in female meiosis.     

Regular steps in bending cilia during the effective stroke.

A ciliary beat cycle consists of an effective stroke in which the extended cilium makes an oar-like movement towards one side, and a recovery stroke in which the cilium moves back by propagating a bend from base to tip (Fig. 1A). In the sliding microtubule model of ciliary and flagellar movement, which is now supported by substantial evidence1-3, the sliding displacement of microtubules in any region of the cilium is related linearly to the angular change in the direction of that region (Fig. 1B). Thus, during the effective stroke, microtubule sliding is not confined to the region near the base of the cilium but involves the whole length of the extended organelle, and the relative speed of sliding can be measured as the angular velocity of the ciliary motion. I report here that in molluscan cilia the effective stroke consists of regularly alternating rapid and slow phases of angular movement. This suggests that the microtubules slide in quantal steps.     

Identification of globular mechanochemical heads of kinesin

Kinesin is a mechanoenzyme which uses energy liberated from ATP hydrolysis to transport particles towards the 'plus ends' of microtubules. The enzyme consists of two polypeptide heavy chains of relative molecular mass (Mr) approximately 110,000-140,000 (110K-140K) plus copurifying light chains; these polypeptides are arranged in a structure consisting of two globular heads attached to a fibrous stalk which terminates in a 'feathered' tail. Here we report that a function-disrupting monoclonal antikinesin, which binds to the 45K fragment of the kinesin heavy chain, recognizes an epitope located towards the N-terminal end of the heavy chain, and decorates the two globular heads lying at one end of the intact molecules (one antibody per head). The results show that the two heavy chains of native kinesin are arranged in parallel, and that the 45K fragments, which display nucleotide-sensitive interactions with microtubules, represent mechanochemical 'heads' located at the N-terminal regions of the heavy chains. Thus, it is likely that the kinesin heads are analogous to the subfragment-1 domains of myosin.     

Retrograde transport by the microtubule-associated protein MAP 1C

Microtubules are involved in several forms of intracellular motility, including mitosis and organelle movement. Fast axonal transport is a highly ordered form of organelle motility that operates in both the anterograde (outwards from the cell body) and retrograde (from the periphery towards the cell body) direction. Similar microtubule-associated movement is observed in non-neuronal cells, and might be involved in secretion, endocytosis and the positioning of organelles within the cell. Kinesin is a mechanochemical protein that produces force along microtubules in an anterograde direction. We recently found that the brain microtubule-associated protein MAP 1C (ref. 7) is a microtubule-activated ATPase and, like kinesin, can translocate microtubules in an in vitro assay for microtubule-associated motility. MAP 1C seemed to be related to the ciliary and flagellar ATPase, dynein, which is thought to produce force in a direction opposite to that observed for kinesin. Here we report that MAP 1C, in fact, acts in a direction opposite to kinesin, and has the properties of a retrograde translocator.     

Four ATP-binding sites in the midregion of the beta heavy chain of dynein.

The 'motor' proteins of eukaryotic cells contain specialized domains that hydrolyse ATP to produce force and movement along a cytoskeletal polymer (actin in the case of the myosin family; microtubules in the case of the kinesin family and dyneins). There are motor-protein superfamilies in which each member has a conserved force-generating domain joined to a different 'tail' which conveys specific attachment properties. The minus-end-directed microtubule motors, the dyneins, may also constitute a superfamily of force-generating proteins with distinct attachment domains. Axonemal outer-arm dynein from sea urchin spermatozoa is a multimeric protein consisting of two heavy chains (alpha and beta) with ATPase activity, three intermediate chains and several light chains. Here I report the sequence of cloned complementary DNA encoding the beta heavy chain of a dynein motor molecule. The predicted amino-acid sequence reveals four ATP-binding consensus sequences in the central domain. The dynein beta heavy chain is thought to associate transiently with a microtubule during ATP hydrolysis, but the ATP-dependent microtubule-binding sequence common to the kinesin superfamily is not found in the dynein beta heavy chain. These unique features distinguish the dynein beta heavy chain from other motor protein superfamilies and may be characteristic of the dynein superfamily.     

Interaction of brain cytoplasmic dynein and MAP2 with a common sequence at the C terminus of tubulin

Two main types of microtubule-associated proteins (MAPs) have been identified in neuronal cells. The fibrous MAPs, including MAP2 and tau, serve to organize and regulate the assembly of microtubules. A second distinct class of force-producing MAPs, including kinesin, dynein and dynamin, are involved in microtubule-based movement. These proteins are mechanochemical ATPases which seem to be responsible for the bidirectional transport of organelles and perhaps also the movement of chromosomes. Here we report that MAP2 inhibits microtubule gliding on dynein-coated coverslips, as well as the microtubule-activated ATPase of dynein, indicating that MAP2 and other fibrous MAPs could be important modulators of microtubule-based motility in vivo. By proteolytic modification of tubulin, we found that dynein interacts with microtubules at the C termini of alpha- and beta-tubulin, the regions previously reported to be the sites for the interaction of MAP2. The use of site-directed antibodies implicates a small region of alpha- and beta-tubulin, containing the sequence Glu-Gly-Glu-Glu, as the site of the interaction of dynein and MAP2 with the microtubule.     

Myosin head movements are synchronous with the elementary force-generating process in muscle.

Motor proteins such as myosin, dynein and kinesin use the free energy of ATP hydrolysis to produce force or motion, but despite recent progress their molecular mechanism is unknown. The best characterized system is the myosin motor which moves actin filaments in muscle. When an active muscle fibre is rapidly shortened the force first decreases, then partially recovers over the next few milliseconds. This elementary force-generating process is thought to be due to a structural 'working stroke' in the myosin head domain, although structural studies have not provided definitive support for this. X-ray diffraction has shown that shortening steps produce a large decrease in the intensity of the 14.5 nm reflection arising from the axial repeat of the myosin heads along the filaments. This was interpreted as a structural change at the end of the working stroke, but the techniques then available did not allow temporal resolution of the elementary force-generating process itself. Using improved measurement techniques, we show here that myosin heads move by about 10 nm with the same time course as the elementary force-generating process.     

The Drosophila claret segregation protein is a minus-end directed motor molecule

A product encoded at the claret locus in Drosophila is needed for normal chromosome segregation in meiosis in females and in early mitotic divisions of the embryo. The predicted amino-acid sequence of the segregation protein was shown recently to be strikingly similar to Drosophila kinesin heavy chain. We have expressed the claret segregation protein in bacteria and have found that the bacterially expressed protein has motor activity in vitro with several novel features. The claret motor is slow (4 microns min-1), unlike either kinesin or dyneins. It has the directionality, the ability to generate torque and the sensitivity to inhibitors reported previously for dyneins. The finding of minus-end directed motor activity for a protein with sequence similarity to kinesin suggests that the dynein and kinesin motor domains are ancestrally related. The minus-end directed motor activity of the claret motor is consistent with a role for this protein in producing chromosome movement along spindle microtubules during prometaphase and/or anaphase.     

Actin-dependent organelle movement in squid axoplasm.

Studies of organelle movement in axoplasm extruded from the squid giant axon have led to the basic discoveries of microtubule-dependent organelle motility and the characterization of the microtubule-based motor proteins kinesin and cytoplasmic dynein. Rapid organelle movement in higher animal cells, especially in neurons, is considered to be microtubule-based. The role of actin filaments, which are also abundant in axonal cytoplasm, has remained unclear. The inhibition of organelle movement in axoplasm by actin-binding proteins such as DNase I, gelsolin and synapsin I has been attributed to their ability to disorganize the microtubule domains where most of the actin-filaments are located. Here we provide evidence of a new type of organelle movement in squid axoplasm which is independent of both microtubules and microtubule-based motors. This movement is ATP-dependent, unidirectional, actin-dependent, and probably generated by a myosin-like motor. These results demonstrate that an actomyosin-like mechanism can be directly involved in the generation of rapid organelle transport in nerve cells.