Mitotic Microtubule Crosslinkers: Insights from Mechanistic Studies

Current Biology 19, R1089–R1094, December 15, 2009 ª2009 Elsevier Ltd All rights reserved

Mitotic Microtubule Crosslinkers: Insights from Mechanistic Studies

Erwin J.G. Peterman1 and Jonathan M. Scholey2

Mitosis depends on the mitotic spindle, a subcellular protein machine that uses dynamic microtubules and mitotic motors to assemble itself and to coordinate chromosome movements. Spindle function depends critically on the interplay of microtubule polymer dynamics and the motor proteins and non-motor microtubule-associated proteins (MAPs) that crosslink adjacent microtubules. These microtubule crosslinkers can organize microtubules into bundles with specific polarity patterns and some of them can slide adjacent microtubules in relation to one another. Here, we discuss the functions and mechanisms of action of three such crosslinkers: the motors kinesin-5 and kinesin-14, and the non-motor MAPs of the Ase1p family. The major purpose of mitosis is to coordinate the accurate distribution of genetic instructions, packaged into chromosomes, to the daughter products of each cell division. Chromosome segregation depends upon the action of the mitotic spindle, a subcellular protein machine that uses kinesin and dynein motors together with microtubule dynamics to assemble itself and then to move separated chromatids polewards during anaphase A and to elongate the spindle during anaphase B. Microtubules are structurally polar polymers with distinct ‘plus’ and ‘minus’ ends that have different structures and kinetic properties. For many years it has been thought that spindle microtubules are organized into linear, parallel or antiparallel bundles by protein crosslinkers that form bridges visible by electron microscopy between the walls of adjacent microtubules [1]. For example, centrosome-dominant astral spindles are constructed from three sets of microtubules emanating from the poles with their plus ends distal (Figure 1) [2,3]. These three sets of microtubules comprise kinetochore microtubules (kMTs), astral microtubules and interpolar microtubules (ipMTs). The kMTs form parallel bundles, with their plus ends facing the kinetochores and their minus ends facing the poles, and their depolymerization contributes to chromosome-to-pole motion during anaphase A. Astral microtubules radiate outwards allowing their plus ends to interact with cortical force generators that exert pulling forces on spindle poles. The ipMTs have their plus ends overlapping at the central spindle, forming anti-parallel arrays that can slide inward or outward to exert forces on the spindle poles — during anaphase, ipMTs interdigitate and form the central spindle midzone, which plays an essential role in spindle elongation and the initiation of cytokinesis [3,4]. Within spindles, a variety of motors and non-motor

1Department

of Physics and Astronomy and Laser Centre, VU University, De Boelelaan 1081, 1081 HV Amsterdam, the Netherlands. 2Department of Molecular and Cell Biology, University of California at Davis, Davis, CA 95616, USA. E-mail: [email protected] (E.J.G.P.), [email protected] (J.M.S.)

DOI 10.1016/j.cub.2009.10.047

Minireview

MAPs are likely to form crosslinks between the microtubules of these three types of microtubule bundles [4–6]. In this minireview, we focus on a subset of such microtubule crosslinkers that have been shown to directly crosslink microtubules without the aid of additional factors, namely motors of the kinesin-5 and kinesin-14 families, and non-motor MAPs of the Ase1p family. We discuss several recent in vitro studies that illuminate their mitotic functions and mechanisms of action. Kinesin-5 and Kinesin-14 Motors in Mitosis Basic biochemical and cell biological studies support the view that members of the kinesin-5 and kinesin-14 motor protein families act primarily as microtubule–microtubule crosslinking and sliding motors. Kinesin-5 motors (Eg5 in Xenopus laevis and human, KLP61F in Drosophila melanogaster, bimC in Aspergillus nidulans) are thought to be ‘slow’ plus-end directed motors that move along microtubules at rates of z0.1–0.01 mm/s, characteristic of mitotic motility and about 10-fold slower than many intracellular transport motors, such as kinesin-1. Importantly, these kinesins assemble into bipolar homotetrameric complexes organized with pairs of amino-terminal motor domains located at opposite ends of a four-strand coiled-coil rod, that can crosslink adjacent spindle microtubules, displaying a preference for the anti-parallel orientation, and slide them in relation to one another [7,8]. Kinesin-14 motors (Ncd in Drosophila, Klp2p in Schizosaccharomyces pombe, Kar3 in Saccharomyces cerevisiae) are slow, minus-end-directed homodimers, consisting of two carboxy-terminal motor domains linked by coiled-coil rods to nucleotide-insensitive microtubule-binding tails, which are also capable of crosslinking and sliding adjacent spindle microtubules [9,10]. The mitotic functions of both motors appear to be regulated via the tail domains; the carboxy-terminal tail of kinesin-5s from Xenopus and Drosophila contains the Cdk-phosphorylatable ‘bimC-box’ and appears to control the motor’s microtubule binding, orientation preference and spindle association [7,8,11], whereas the amino-terminal tails of Xenopus kinesin-14 bind importin a/b, which blocks the motor’s microtubule–microtubule crosslinking activity and spindle association in a Ran-sensitive fashion [12]. Understanding the mitotic functions of kinesin-5 and kinesin-14 is complicated, as they appear to function differently in spindles of different design. These two motors are generally thought to function antagonistically to contribute to various aspects of spindle structure and function (Figure 1) [13,14]. For example, a plausible model posits that, in centrosome-controlled spindles, kinesin-5 and kinesin-14 may crosslink and slide anti-parallel ipMTs at the midzone outward and inward, respectively, allowing kinesin-5 to drive poleward flux and pole–pole separation and kinesin-14 to shorten the spindle via pole–pole collapse [15]. However, in anastral spindles, it is proposed that poleward flux and spindle length control may depend on kinesin-5 serving to transport microtubules along opposite polarity microtubule tracks from the chromosomes to the poles, where they are focused by minus-end motors [16]. Furthermore, in some

Current Biology Vol 19 No 23 R1090

A Metaphase

B Anaphase B

Figure 1. The functions of key microtubule crosslinking proteins in mitosis. (A) During metaphase the mitotic spindle has already assembled. The motor proteins kinesin-5 (green) and kinesin-14 (red) crosslink anti-parallel microtubules (plus ends marked) originating from opposite poles (grey circles) and slide them apart in opposite directions: kinesin-14 slides the microtubules inward (large red arrows on top), kinesin-5 outward (large green arrow on top), resulting in a force balance that contributes to spindle assembly and maintenance. In addition, kinesin-5 might be involved in bundling parallel microtubules close to the spindle poles. (B) After separation of the chromosomes (grey), microtubule crosslinkers of the Ase1p family of MAPs (blue) complement microtubule crosslinking motors, such as kinesin-5 or kinesin-6, to play key roles in organizing the central spindle midzone, which is essential for anaphase B spindle elongation and cytokinesis. The figures are simplified representations: bundles consisting of more than two microtubules will form, but are not shown for clarity.

polarity-marked microtubules showed that the static population comprised parallel bundles, while the sliding population comprised anti-parallel ones [22] (Figure 2). The experiments Kinesin-5 Kinesin-14 Ase1p demonstrated that the motor domains at opposite ends of the bipolar motors Current Biology move towards the plus ends of the microtubules that they crosslink. In later systems, kinesin-14 is proposed to transport parallel micro- experiments, the motility properties of individual kinesin-5 tubules poleward, thereby contributing to an increase in motors were addressed using optical tweezers and it was spindle length [12]. In some cell types, however, microtu- observed that truncated, dimeric human constructs are only bule–microtubule sliding by kinesin-5 and kinesin-14 moderately processive, taking, on average, eight steps appears to contribute relatively little to the control of spindle before dissociating from the microtubule [24]. Single-molelength [17], and in some cases these motors may use their cule fluorescence studies using tetrameric GFP-tagged microtubule crosslinking activities to focus microtubules at Xenopus Eg5 revealed more complex motility behaviour, spindle poles (e.g. [18]), or to serve as brakes that restrain consisting of episodes of directed, processive motion and the rate or extent of microtubule sliding driven by other episodes of diffusion along the microtubule [25]. The relative motors [19]. The complexity of kinesin-5 and kinesin-14 func- contribution of these two motility modes is sensitive to the tion is compounded by proposals that kinesin-5 and kinesin- ionic strength [26] and can be altered by the addition of the 14 contribute to microtubule polymer dynamics under some small-molecule Eg5 inhibitor monastrol [25]. Interestingly, circumstances, but if and how this relates to their microtu- at physiological salt concentrations, tetrameric Eg5 diffused bule crosslinking activity remains to be determined [20,21]. along individual microtubules but then moved in a much Clearly, further studies are required to illuminate the func- more directed fashion when it encountered the overlap tions and mechanisms of action of these key mitotic motors, zone of crosslinked microtubules [26]. On the basis of this and it is likely that mechanistic studies of the purified motors observation, a model was proposed in which Eg5 moves may illuminate this problem. passively and diffusionally in the absence of ATP hydrolysis when bound to individual microtubules, but switches to Biophysical Studies of Kinesin-5 active, vectorial, ATP-dependent motion when it crosslinks Direct evidence of kinesin-5’s capability to crosslink micro- two microtubules [26] (Figure 2). The molecular basis of tubules and slide them apart was provided by in vitro motility this activation mechanism remains to be uncovered. Finally, assays of purified full-length Xenopus Eg5 [22,23] and it was shown that KLP61F motors display a three-fold preferDrosophila KLP61F [8]. In these fluorescence microscopy- ence for anti-parallel versus parallel microtubule bundling based assays, one population of fluorescently tagged micro- [8]. Surprisingly, tetrameric KLP61F subfragments lacking tubules was firmly attached to the cover glass and a second the motor domains but containing the carboxy-terminal tails population was subsequently added, together with motors are able to crosslink microtubules with the same orientation and ATP. Bundling of microtubules could be readily ob- preference as the full-length protein [8,14]. It is thus likely served: a portion of the microtubule bundles remained that these microtubule-binding tail domains specify the static, while the remainder slid apart. Experiments with anti-parallel crosslinking preference. In addition, these

Review R1091

Figure 2. Mechanisms of action of kinesin-5 (green), kinesin-14 (red), and Ase1p (blue) as revealed by in vitro experiments. (A) Kinesin-5 diffuses along individual microtubules (dotted, pale green arrow); parallel microtubules (left) are statically crosslinked, while the motors move towards the microtubule plus ends (green arrows) and do not generate microtubule-sliding forces; antiparallel microtubules (right) are crosslinked and slid apart (black arrow), while the motors move directionally towards the plus ends of both crosslinked microtubules (green arrows). (B) Kinesin-14 diffuses randomly along individual microtubules (dotted, pale red arrow), while being bound with its amino-terminal nucleotide-insensitive microtubule-binding domains; parallel microtubules (left) are statically crosslinked; anti-parallel microtubules (right) are crosslinked and slid apart (black arrow), while the motors move directionally towards the minus ends of both microtubules (red arrows; note that the direction of relative sliding is opposite to kinesin-5). (C) Ase1p diffuses randomly along individual microtubules (dotted pale blue arrows); diffusion is substantially reduced between anti-parallel microtubules (shorter arrow) and static Ase1p multimers are formed.

A Kinesin-5

B Kinesin-14

C Ase1p

domains might be responsible for the diffusive interaction with microtubules, since no diffusion was observed for truncated, dimeric constructs lacking them [26]. Biophysical Studies of Kinesin-14 Interaction of the non-motor microtubule-binding site at the carboxyl terminus of Drosophila Ncd or S. pombe Klp2p with a microtubule leads to 1D diffusion along the microtubule polymer lattice [9,27,28]. Recent in vitro experiments, similar to the ones discussed above for kinesin-5, have shown that dimeric kinesin-14 constructs containing both its motor domains and its amino-terminal tail domains can crosslink microtubules [9,29], but, in contrast to kinesin-5, no clear preference for the parallel or anti-parallel orientation was observed [9]. Parallel microtubule pairs are statically locked together by Ncd [9] or Klp2p [29], while anti-parallel microtubule pairs are slid apart in the opposite direction to kinesin-5 (Figure 2). In vitro, over time, this results in the ‘sorting’ of microtubules into bundles of parallel polarity, since antiparallel microtubules are driven out of the bundle leaving behind parallel ones [29]. Although the effect of kinesin-14 action on microtubule bundles appears superficially to resemble that of kinesin-5, the underlying mechanism is quite different. Kinesin-5 interacts via its motor domains with both of the microtubules that it crosslinks [22,26], while kinesin-14 interacts with one microtubule via its motor domains and with the other via its nucleotide-insensitive microtubule-binding site in the tail. In addition, under physiological conditions, kinesin-14 moves non-processively [9,27,28,30,31], indicating that many of them must cooperate to drive persistent sliding. As mentioned above, models have been proposed in which a force balance between kinesin-5and kinesin-14-driven microtubule motility is required for proper spindle formation. It would be very interesting to see how these two motor proteins act together on a single microtubule bundle. So far, this question has only been addressed in surface-gliding assays using surface-attached

Current Biology

KLP61F and Ncd in different ratios [14]. In this study, it was found that the microtubule gliding velocity remained unimodal and changed gradually as a function of the molar ratio of the two motors, eventually reversing direction. At the ‘balance point’ ratio where the average velocity was zero, microtubules were observed to undertake submicron oscillations that are probably too small to be detected by light microscopy of mitotic spindle poles. It will now be interesting to extend such experiments with purified motors to microtubule bundles and to measure and apply forces to them using optical tweezers. MAPs of the Ase1p Family in Mitosis The spindle midzone, formed during anaphase, consists of a dense network of overlapping anti-parallel microtubules containing a plethora of motor proteins, kinases and non-motor MAPs [4]. In all eukaryotes, one of the key proteins is a member of the Ase1p family of microtubule–microtubule crosslinking MAPs (PRC1 in humans [32]; the products of the genes fascetto (Feo) and sofe in Drosophila [33]; Ase1p in S. cerevisiae [34] and S. pombe [35]; MAP65 in plants [36]). Ase1p family members have been shown to localize to the spindle midzone [32,35,37,38] and perturbation of their function leads to failure of spindle midzone formation resulting in two disconnected half spindles [33,35,37,39]. Overexpression of these proteins in interphase results in excessive microtubule bundling [39]. In vertebrate cells, PRC1 forms complexes with various kinesins (KIF4, MKLP1, CENP-E, KIF14) and other proteins like the microtubule plus-end binding protein CLASP1 [40– 43]. In particular, PRC1 and KIF4 depend on each other’s activity for proper localization to the central spindle midzone [41,43]. A model has emerged in which KIF4 moves PRC1 to the plus ends of interdigitating microtubules [41,43]. In Drosophila, something similar might be happening: the kinesin-4 KLP3A is required for central spindle organization and cytokinesis [44], possibly in conjunction with Feo and Sofe.

Current Biology Vol 19 No 23 R1092

In addition, KLP3A appears to play a poorly understood role in the suppression of ipMT depolymerization at spindle poles, which triggers anaphase B spindle elongation [45]. In yeast, which does not have a kinesin-4 homologue, Ase1p localization most likely does not require motor protein activity. Ase1p is essential for localization of Klp9p (kinesin6) to the midzone, a motor that is essential for spindle elongation [46]. In humans, the microtubule-bundling activity of PRC1 is regulated by Cdk (Cdc2/cyclin B) phosphorylation: in early mitosis phosphorylated PRC1 is bound to spindle microtubules but is unable to crosslink them, whereas in anaphase the protein gets dephosphorylated and is then able to bundle anti-parallel microtubules [47]. Similar phosphorylation mechanisms appear to also regulate Ase1p function in S. pombe [46] and S. cerevisiae [48]. Biophysical Studies of MAPs of the Ase1p Family In vitro, the microtubule-bundling activity of several Ase1p family members has been confirmed with optical and electron microscopy [36–39]. To date, detailed in vitro studies have only been performed on Ase1p from S. pombe using single-molecule fluorescence microscopy [49]. At low concentrations, dimeric Ase1p–GFP fusion proteins were observed to bind to microtubules and to diffuse along them (Figure 2), a result that was confirmed in vivo by the expression of Ase1p–GFP in mammalian cells. At higher concentrations, Ase1p–GFP dimers were incorporated into multimers that bound in a relatively immobile state to the microtubules. Ase1p was observed to crosslink microtubules, with an approximately 3:1 preference for anti-parallel versus parallel bundling [38,49]. Within microtubule bundles, Ase1p diffusion was substantially slower than on single microtubules and multimerization occurred at much lower Ase1p concentrations (Figure 2). These results indicate that the specific localization of Ase1p to regions of anti-parallel microtubule overlap is driven by a mechanism involving diffusion of Ase1p dimers along the microtubule lattice and incorporation of dimers into immobile multimers that form with high preference between overlapping anti-parallel microtubules [49]. These multimers could act as a stable scaffold in the spindle midzone to which partner proteins can bind, in accordance with the hypothesis that Ase1p may form a central spindle ‘matrix’ [37]. It remains to be determined whether other proteins of the Ase1p family, which differ substantially in sequence outside the highly conserved central microtubule-binding domain, show similar diffusion and multimerization behaviour. In addition, it would be of great interest to determine how the phosphorylation state of the MAPs and their interactions with motor proteins (especially kinesin-4 motors) affect translocation along microtubules and multimerization. Concluding Remarks Thus, recent cell biological and biophysical studies have established the importance of the microtubule-crosslinking activity of kinesin-5, kinesin-14 and Ase1p family MAPs in vivo and in vitro, but there remain many questions about their mechanisms of action, biological functions, regulation and functional cooperation with each other and with other microtubule crosslinkers, both in the spindle and elsewhere. For example, what is the molecular mechanism by which kinesin-5 and Ase1p preferentially crosslink microtubules in the anti-parallel orientation? In the case of kinesin-5, this preference appears to be mediated by the non-motor

microtubule-binding sites in the Cdk1-phosphorylatable carboxy-terminal tails, which also control its microtubulebinding properties and spindle targeting, raising the question of whether this phosphorylation also controls the orientation preference. Further understanding of the mechanism of the orientation preference of both proteins will require more information on their overall structures and especially the torsional rigidity of the motors within their rod domains. In vitro experiments have shown that kinesin-5, kinesin-14 and Ase1p proteins can diffuse along the microtubule polymer lattice matrix, but what is the mechanism of this diffusive motion, and how is this motility mode influenced by the various structural and regulatory domains of these crosslinkers? In the case of Ase1p it has been shown that diffusion is relevant in vivo [49], but this remains to be tested for the other proteins. It could be that this diffusive motion is important to prevent the accumulation of crosslinkers at the microtubule ends, or to act as a sort of ‘clutch’ that facilitates the release of excess tension that could accumulate within the crosslinked microtubule bundles. How do these crosslinkers cooperate with each other and with other spindle components to mediate accurate mitosis? In vivo studies indicate that some Ase1p family members require the assistance of the motor protein kinesin-4, not just diffusive motility, for proper localization to the spindle midzone, but further biochemical and biophysical experiments will be required to resolve the mechanism of this functional interaction. Furthermore, it is of great interest to improve the understanding of the kinesin-5 versus kinesin14 force balance, which is proposed to contribute to the assembly and maintenance of many spindles, but the precise role and mechanism of which remain unclear. For example, motility assays have provided some important insights into how these proteins might work individually as microtubule crosslinking and sliding motors, but what happens when kinesin-5 and kinesin-14 act competitively in opposing directions on the same microtubule pairs? In addition, how does Ase1p activity modulate the action of these crosslinking motors? Will it act as a brake or allow the crosslinked microtubules to be slid apart unrestrained by the motor-generated forces [38]? To address these questions, more advanced in vitro microtubule crosslinking assays involving the measurement and application of forces will be required. Finally, it is likely that we are looking at the tip of the iceberg and that many more motor proteins and non-motor MAPs with microtubule crosslinking functions, possibly acting by distinct mechanisms, will be found both in the spindle and elsewhere. For example, the loss of function of some motors leads to disorganization of microtubule bundles, but how these proteins contribute to microtubule bundling is unclear. Among such motors, some may act as direct microtubule crosslinkers, like kinesin-5 and kinesin14, whereas others may act indirectly by targeting microtubule crosslinkers to their site of action. For example, the dynein–dynactin complex localizes NuMA to the spindle poles where it crosslinks and focuses microtubule minus ends, in a manner reminiscent of Ase1p targeting by kinesin-4. Indeed, the deployment of such crosslinkers may vary in a system-specific manner, as exemplified by observations that anaphase B spindle elongation uses kinesin-5 in some spindles and a kinesin-6–Ase1p complex in others [46]. Microtubule bundling by such crosslinkers is not restricted to mitotic spindles, as these crosslinkers are also

Review R1093

important elsewhere. Ase1p and Klp2p are involved in interphase nuclear positioning in S. pombe [35] and in neurons, microtubule crosslinkers, including motors like kinesin-5 and MAPs such as Tau and MAP-2, are thought to contribute to the microtubule-polarity patterns of axons and dendrites [50]. Genome-wide screens might be needed to uncover the full spectrum of crosslinking proteins involved in these processes, and further in vitro studies will be required to determine their mechanism of action, whether they include crosslinkers with a preference for parallel microtubules, whether all crosslinkers display diffusive movement along microtubules, and how they are regulated e.g. by phosphorylation or interaction with binding partners. In the years to come, an extensive combination of studies carried out in vivo, in vitro and in silico should greatly expand our understanding of the mechanisms and functions of these fascinating proteins. Acknowledgments We thank our lab members and collaborators for insightful discussions, and Siet van den Wildenberg, Bram Prevo (VU University, Amsterdam), Marcel Janson (Wageningen University) and Claire Walczak (Indiana University) for comments on this manuscript. References

18.

Goshima, G., Nedelec, F., and Vale, R.D. (2005). Mechanisms for focusing mitotic spindle poles by minus end-directed motor proteins. J. Cell Biol. 171, 229–240.

19.

Saunders, A.M., Powers, J., Strome, S., and Saxton, W.M. (2007). Kinesin-5 acts as a brake in anaphase spindle elongation. Curr. Biol. 17, R453–R454.

20.

Gardner, M.K., Bouck, D.C., Paliulis, L.V., Meehl, J.B., O’Toole, E.T., Haase, J., Soubry, A., Joglekar, A.P., Winey, M., Salmon, E.D., et al. (2008). Chromosome congression by Kinesin-5 motor-mediated disassembly of longer kinetochore microtubules. Cell 135, 894–906.

21.

Sproul, L.R., Anderson, D.J., Mackey, A.T., Saunders, W.S., and Gilbert, S.P. (2005). Cik1 targets the minus-end kinesin depolymerase kar3 to microtubule plus ends. Curr. Biol. 15, 1420–1427.

22.

Kapitein, L.C., Peterman, E.J.G., Kwok, B.H., Kim, J.H., Kapoor, T.M., and Schmidt, C.F. (2005). The bipolar mitotic kinesin Eg5 moves on both microtubules that it crosslinks. Nature 435, 114–118.

23.

Cahu, J., and Surrey, T. (2009). Motile microtubule crosslinkers require distinct dynamic properties for correct functioning during spindle organization in Xenopus egg extract. J. Cell Sci. 122, 1295–1300.

24.

Valentine, M.T., Fordyce, P.M., Krzysiak, T.C., Gilbert, S.P., and Block, S.M. (2006). Individual dimers of the mitotic kinesin motor Eg5 step processively and support substantial loads in vitro. Nat. Cell Biol. 8, 470–479.

25.

Kwok, B.H., Kapitein, L.C., Kim, J.H., Peterman, E.J.G., Schmidt, C.F., and Kapoor, T.M. (2006). Allosteric inhibition of kinesin-5 modulates its processive directional motility. Nat. Chem. Biol. 2, 480–485.

26.

Kapitein, L.C., Kwok, B.H., Weinger, J.S., Schmidt, C.F., Kapoor, T.M., and Peterman, E.J.G. (2008). Microtubule cross-linking triggers the directional motility of kinesin-5. J. Cell Biol. 182, 421–428.

27.

Furuta, K., Edamatsu, M., Maeda, Y., and Toyoshima, Y.Y. (2008). Diffusion and directed movement in vitro motile properties of fission yeast kinesin14 Pkl1. J. Biol. Chem. 283, 36465–36473.

28.

Furuta, K., and Toyoshima, Y.Y. (2008). Minus-end-directed motor ncd exhibits processive movement that is enhanced by microtubule bundling in vitro. Curr. Biol. 18, 152–157.

1.

Hepler, P.K., McIntosh, J.R., and Cleland, S. (1970). Intermicrotubule bridges in mitotic spindle apparatus. J. Cell Biol. 45, 438–444.

2.

McDonald, K.L., O’Toole, E.T., Mastronarde, D.N., and McIntosh, J.R. (1992). Kinetochore microtubules in PTK cells. J. Cell Biol. 118, 369–383.

29.

3.

McIntosh, J.R., McDonald, K.L., Edwards, M.K., and Ross, B.M. (1979). Three-dimensional structure of the central mitotic spindle of Diatoma vulgare. J. Cell Biol. 83, 428–442.

Braun, M., Drummond, D.R., Cross, R.A., and McAinsh, A.D. (2009). The kinesin-14 Klp2 organizes microtubules into parallel bundles by an ATP-dependent sorting mechanism. Nat. Cell Biol. 11, 724–755.

30.

Case, R.B., Pierce, D.W., HomBooher, N., Hart, C.L., and Vale, R.D. (1997). The directional preference of kinesin motors is specified by an element outside of the motor catalytic domain. Cell 90, 959–966.

31.

deCastro, M.J., Fondecave, R.M., Clarke, L.A., Schmidt, C.F., and Stewart, R.J. (2000). Working strokes by single molecules of the kinesin-related microtubule motor ncd. Nat. Cell Biol. 2, 724–729.

32.

Jiang, W., Jimenez, G., Wells, N.J., Hope, T.J., Wahl, G.M., Hunter, T., and Fukunaga, R. (1998). PRC1: A human mitotic spindle-associated CDK substrate protein required for cytokinesis. Mol. Cell 2, 877–885.

4.

Glotzer, M. (2009). The 3Ms of central spindle assembly: microtubules, motors and MAPs. Nat. Rev. Mol. Cell Biol. 10, 9–20.

5.

Walczak, C.E., and Heald, R. (2008). Mechanisms of mitotic spindle assembly and function. Int. Rev. Cytol 265, 111–158.

6.

Manning, A.L., and Compton, D.A. (2008). Structural and regulatory roles of nonmotor spindle proteins. Curr. Opin. Cell Biol. 20, 101–106.

7.

Sharp, D.J., McDonald, K.L., Brown, H.M., Matthies, H.J., Walczak, C., Vale, R.D., Mitchison, T.J., and Scholey, J.M. (1999). The bipolar kinesin, KLP61F, cross-links microtubules within interpolar microtubule bundles of Drosophila embryonic mitotic spindles. J. Cell Biol. 144, 125–138.

33.

Verni, F., Somma, M.P., Gunsalus, K.C., Bonaccorsi, S., Belloni, G., Goldberg, M.L., and Gatti, M. (2004). Feo, the Drosophila homolog of PRC1, is required for central-spindle formation and cytokinesis. Curr. Biol. 14, 1569–1575.

8.

van den Wildenberg, S., Tao, L., Kapitein, L.C., Schmidt, C.F., Scholey, J.M., and Peterman, E.J.G. (2008). The homotetrameric kinesin-5 KLP61F preferentially crosslinks microtubules into antiparallel orientations. Curr. Biol. 18, 1860–1864.

34.

Pellman, D., Bagget, M., Tu, H., and Fink, G.R. (1995). Microtubule-associated proteins required for anaphase spindle movement in Saccharomyces cerevisiae. J. Cell Biol. 130, 1373–1385.

35.

9.

Fink, G., Hajdo, L., Skowronek, K.J., Reuther, C., Kasprzak, A.A., and Diez, S. (2009). The mitotic kinesin-14 Ncd drives directional microtubule-microtubule sliding. Nat. Cell Biol. 11, 717–723.

Loiodice, I., Staub, J., Setty, T.G., Nguyen, N.P.T., Paoletti, A., and Tran, P.T. (2005). Ase1p organizes antiparallel microtubule arrays during interphase and mitosis in fission yeast. Mol. Biol. Cell 16, 1756–1768.

36.

10.

Mountain, V., Simerly, C., Howard, L., Ando, A., Schatten, G., and Compton, D.A. (1999). The kinesin-related protein, HSET, opposes the activity of Eg5 and cross-links microtubules in the mammalian mitotic spindle. J. Cell Biol. 147, 351–366.

Chan, J., Jensen, C.G., Jensen, L.C.W., Bush, M., and Lloyd, C.W. (1999). The 65-kDa carrot microtubule-associated protein forms regularly arranged filamentous cross-bridges between microtubules. Proc. Natl. Acad. Sci USA 96, 14931–14936.

37.

11.

Cahu, J., Olichon, A., Hentrich, C., Schek, H., Drinjakovic, J., Zhang, C., Doherty-Kirby, A., Lajoie, G., and Surrey, T. (2008). Phosphorylation by Cdk1 increases the binding of Eg5 to microtubules in vitro and in Xenopus egg extract spindles. PLoS One 3, e3936.

Schuyler, S.C., Liu, J.Y., and Pellman, D. (2003). The molecular function of Ase1p: evidence for a MAP-dependent midzone-specific spindle matrix. J. Cell Biol. 160, 517–528.

38.

Janson, M.E., Loughlin, R., Loiodice, I., Fu, C.H., Brunner, D., Nedelec, F.J., and Tran, P.T. (2007). Crosslinkers and motors organize dynamic microtubules to form stable bipolar arrays in fission yeast. Cell 128, 357–368.

39.

Mollinari, C., Kleman, J.P., Jiang, W., Schoehn, G., Hunter, T., and Margolis, R.L. (2002). PRC1 is a microtubule binding and bundling protein essential to maintain the mitotic spindle midzone. J. Cell Biol. 157, 1175–1186.

40.

Gruneberg, U., Neef, R., Li, X., Chan, E.H., Chalamalasetty, R.B., Nigg, E.A., and Barr, F.A. (2006). KIF14 and citron kinase act together to promote efficient cytokinesis. J. Cell Biol. 172, 363–372.

41.

Kurasawa, Y., Earnshaw, W.C., Mochizuki, Y., Dohmae, N., and Todokoro, K. (2004). Essential roles of KIF4 and its binding partner PRC1 in organized central spindle midzone formation. EMBO J. 23, 3237–3248.

42.

Liu, J., Wang, Z., Jiang, K., Zhang, L., Zhao, L., Hua, S., Yan, F., Yang, Y., Wang, D., Fu, C., et al. (2009). PRC1 cooperates with CLASP1 to organize central spindle plasticity in mitosis. J. Biol. Chem. 284, 23059–23071.

43.

Zhu, C.J., and Jiang, W. (2005). Cell cycle-dependent translocation of PRC1 on the spindle by Kif4 is essential for midzone formation and cytokinesis. Proc. Natl. Acad. Sci. USA 102, 343–348.

12.

Cai, S., Weaver, L.N., Ems-McClung, S.C., and Walczak, C.E. (2009). Kinesin14 family proteins HSET/XCTK2 control spindle length by cross-linking and sliding microtubules. Mol. Biol. Cell 20, 1348–1359.

13.

Saunders, W.S., and Hoyt, M.A. (1992). Kinesin-related proteins required for structural integrity of the mitotic spindle. Cell 70, 451–458.

14.

Tao, L., Mogilner, A., Civelekogiu-Scholey, G., Wollman, R., Evans, J., Stahlberg, H., and Scholey, J.M. (2006). A homotetrameric kinesin-5, KLP61F, bundles microtubles and antagonizes Ncd in motility assays. Curr. Biol. 16, 2293–2302.

15.

Brust-Mascher, I., Sommi, P., Cheerambathur, D.K., and Scholey, J.M. (2009). Kinesin-5-dependent poleward flux and spindle length control in Drosophila embryo mitosis. Mol. Biol. Cell 20, 1749–1762.

16.

Burbank, K.S., Mitchison, T.J., and Fisher, D.S. (2007). Slide-and-cluster models for spindle assembly. Curr. Biol. 17, 1373–1383.

17.

Goshima, G., Wollman, R., Stuurman, N., Scholey, J.M., and Vale, R.D. (2005). Length control of the metaphase spindle. Curr. Biol. 15, 1979–1988.

Current Biology Vol 19 No 23 R1094

44.

45.

46.

47.

48.

49.

50.

Williams, B.C., Riedy, M.F., Williams, E.V., Gatti, M., and Goldberg, M.L. (1995). The Drosophila kinesin-like protein KLP3A is a midbody component required for central spindle assembly and initiation of cytokinesis. J. Cell Biol. 129, 709–723. Brust-Mascher, I., Civelekoglu-Scholey, G., Kwon, M., Mogilner, A., and Scholey, J.M. (2004). Model for anaphase B: role of three mitotic motors in a switch from poleward flux to spindle elongation. Proc. Natl. Acad. Sci. USA 101, 15938–15943. Fu, C., Ward, J.J., Loiodice, I., Velve-Casquillas, G., Nedelec, F.J., and Tran, P.T. (2009). Phospho-regulated interaction between kinesin-6 Klp9p and microtubule bundler Ase1p promotes spindle elongation. Dev. Cell 17, 257–267. Zhu, C.J., Lau, E., Schwarzenbacher, R., Bossy-Wetzel, E., and Jiang, W. (2006). Spatiotemporal control of spindle midzone formation by PRC1 in human cells. Proc. Natl. Acad. Sci. USA 103, 6196–6201. Khmelinskii, A., Roostalu, J., Roque, H., Antony, C., and Schiebel, E. (2009). Phosphorylation-dependent protein interactions at the spindle midzone mediate cell cycle regulation of spindle elongation. Dev. Cell 17, 244–256. Kapitein, L.C., Janson, M.E., van den Wildenberg, S., Hoogenraad, C.C., Schmidt, C.F., and Peterman, E.J.G. (2008). Microtubule-driven multimerization recruits Ase1p onto overlapping microtubules. Curr. Biol. 18, 1713–1717. Baas, P.W., and Myers, K.A. (2007). Forces generated by kinesin-5 are a major regulator of microtubule transport and the growth properties of the axon. J. Neurochem. 102 Suppl. 1, 168.