MCAK associates with EB1 - Nature

Oncogene (2008) 27, 2494–2500

& 2008 Nature Publishing Group All rights reserved 0950-9232/08 $30.00 www.nature.com/onc

SHORT COMMUNICATION

MCAK associates with EB1 T Lee, KJ Langford1, JM Askham, A Bru¨ning-Richardson and EE Morrison CRUK Clinical Centre at Leeds, Division of Cancer Medicine Research, St James’s University Hospital, Leeds, UK

The microtubule (MT)-associated protein EB1 localizes to and promotes growth at MT plus ends. The MT depolymerizing kinesin MCAK has also been reported to track growing MT plus ends. Here, we confirm that human MCAK colocalizes with EB1 at growing MT ends when expressed as a GFP fusion protein in transfected cells. We show that MCAK associates with the C-terminus of EB1 and EB3 but much less efficiently with RP1. EB1 associates with the N-terminal localization and regulatory domain in MCAK but not with the motor domain of the protein. The interaction is competitive with the binding of other EB1 ligands and does not require MTs. Knockdown of EB1 expression using siRNA impaired the ability of GFP-MCAK to localize to MT tips in transfected cells. We propose that MCAK is targeted to growing MT ends by EB1, that MCAK is held in an inactive conformation when associated with EB1 and that this could provide the basis for a mechanism that facilitates rapid switching between phases of MT growth and depolymerization. Oncogene (2008) 27, 2494–2500; doi:10.1038/sj.onc.1210867; published online 29 October 2007 Keywords: EB1; MCAK; APC; microtubule

Microtubules (MTs) are biological polymers of fundamental importance in eukaryotic cells. A defining characteristic of MTs is their dynamic behaviour, which is under the tight control of accessory proteins (Desai and Mitchison, 1997). Most MT dynamics within cells is driven by events at the plus end of the MT and microtubule-associated proteins (MAPs) that specifically act at this site have been identified (Lansbergen and Akhmanova, 2006; Morrison, 2007). Many of these, such as the EB1 protein family, also display a specific localization to growing MT ends (Mimori-Kiyosue et al., 2000). This property is shared by a number of proteins, collectively referred to as þ TIPs (Lansbergen and Akhmanova, 2006; Morrison, 2007). EB1 in vertebrate Correspondence: Dr E Morrison, Leeds Institute of Molecular Medicine, University of Leeds, St James’s University Hospital, Beckett St, Leeds, W Yorks LS9 7TF, UK. E-mail: [email protected] 1 Current address: Smith and Nephew Research Centre, York Science Park, York, UK. Received 14 February 2007; revised 14 September 2007; accepted 20 September 2007; published online 29 October 2007

cells suppresses MT catastrophe (Tirnauer et al., 2002). Conversely, members of the Kinesin-13 family of MT motor proteins induce catastrophe (Kinoshita et al., 2006; Moores and Milligan, 2006). The behaviour of MT ends within cells reflects the competing activities of such proteins. The present study was initiated following a report of an association between the Kinesin-13 protein MCAK and the EB1 binding partner APC, an important tumour suppressor protein (Banks and Heald, 2004). If EB1 also associated with MCAK, an interaction between these two functionally opposed proteins could have important implications for the regulation of MT dynamics. Our investigations were given further impetus by a report that MCAK localized to growing MT ends (Moore et al., 2005), since EB1 is known to promiscuously interact with a variety of þ TIPs (Morrison, 2007). The KIF2A and MCAK (KIF2C) cDNAs were obtained as IMAGE clones from the UK Medical Research Council GeneService. Both were cloned into the vector pEGFP-C1 (Clontech, St-Germain-en-Laye, France) allowing expression as GFP fusion proteins. Plasmids were transfected into COS-7 cells and cultures examined 12 h later by time-lapse fluorescence microscopy. As previously reported (Moore et al., 2005), GFP–MCAK localized to motile comet-like structures in the cytoplasm (Figure 1; Supplementary movie 1, Supplementary Information) whereas GFP-KIF2A did not (not shown). Coimmunostaining of transfected cell cultures revealed colocalization of GFP–MCAK with endogenous EB1 at MT plus ends (Figure 1, panels a and b), consistent with the possibility that MCAK might be associated with EB1 at this site. These data independently confirm previously reported findings and indicate that GFP–MCAK behaves normally in our experimental system. To test whether an association between EB1 and MCAK was possible, COS-7 cells were transfected with plasmids driving the expression of GFP, GFP–KIF2A or GFP-MCAK. After 48 h cells were extracted and precipitations performed using either GST or GST–EB1 bound to glutathione-agarose beads. SDS–PAGE and western blotting revealed the presence of both GFP–MCAK and endogenous MCAK in GST–EB1 precipitates (Figure 2a). However, GFP–KIF2A was not detected in GST–EB1 precipitates (Figure 2b). These data indicated that EB1 specifically associated with MCAK and that the addition of a GFP tag to the MCAK N-terminus had no effect upon this association.

MCAK associates with EB1 T Lee et al

2495

Figure 1 GFP–MCAK colocalizes with EB1 at growing MT plus ends. Top panel is a single frame from a movie of a COS-7 cell expressing GFP–MCAK, obtained using time-lapse fluorescence microscopy. Beneath this three panels show a detail from this movie, indicating that GFP–MCAK localizes to motile structures (arrows) in the cytoplasm. (a) and (b) are images of COS-7 cells expressing GFP–MCAK following fixation and coimmunostaining for GFP and EB1. Panel (a) shows GFP–MCAK (green), panel (b) shows a merged image overlaying GFP–MCAK and EB1 (red) immunostaining. GFP–MCAK colocalized with EB1 at microtubule ends in transfected cells. Bar for panels (a and b) ¼ 20 mm. Cell culture, transfections, immunostaining and imaging were performed as described previously (Morrison et al., 1998; Askham et al., 2002; Langford et al., 2006). Oncogene


2496

Figure 2 MCAK associates with EB1. (a) Beads coated with GST–EB1 or GST were incubated with the extracts of COS-7 cells overexpressing GFP–MCAK. Precipitates were subjected to SDS–PAGE and western blotting with an MCAK antibody. Both endogenous MCAK and GFP–MCAK were found in GST–EB1 precipitates. (b) GST–EB1-coated beads were used in precipitations from GFP, GFP–MCAK and GFP–KIF2A-transfected cell extracts. Precipitates and samples of cell extract were subjected to SDS–PAGE and western blotting with an anti-GFP antibody. Only GFP–MCAK was precipitated by GST–EB1. Note the ladder of GFP–MCAK degradation products in the cell extract. (c) GST–EB1, GST–EB1-C84 and GST–EB1-bZIP-coated beads were used in precipitations from GFP–MCAK-transfected cell extracts. Precipitates and samples of cell extract were subjected to SDS–PAGE and western blotting with an anti-GFP antibody. GFP–MCAK and the largest of the GFP-tagged MCAK degradation products (asterisk) were detected in the GST–EB1 and GST–-EB1-C84 precipitates. (d) GST–EB3-C89-coated beads were used in precipitations from GFP, GFP–MCAK and GFP–KIF2A-transfected cell extracts. Precipitates and samples of cell extract were subjected to SDS–PAGE and western blotting with an anti-GFP antibody. Only GFP-MCAK was precipitated by GST–EB3-C89. (e) GST–RP1-C89-coated beads were used in precipitations from GFP, GFP–MCAK and GFP–KIF2A-transfected cell extracts. Precipitates and samples of cell extract were subjected to SDS–PAGE and western blotting with an anti-GFP antibody. No GFP immunoreactive bands were seen in the GST–RP1-C89 precipitates. (f) Equal amounts of bead-bound GST–EB1-C84, GST–EB3-C89 and GST–RP1-C89 were used in precipitations from GFP–MCAK-transfected cell extracts. Precipitates were then subjected to SDS–PAGE and western blotting with an anti-GFP antibody. GST–RP1-C89 precipitated less GFP–MCAK than the other GST fusion proteins. In all panels P ¼ precipitate, Ext ¼ cell extract. GST fusion protein purification, precipitations and SDS–PAGE and western blotting were performed as previously described (Askham et al., 2000, 2002; Langford et al., 2006). (g) The EB1-specific mouse monoclonal antibody 1A11 and a non-specific mouse IgG were used in immunoprecipitations from COS-7 cell extracts. Precipitates were probed by western blotting with an MCAK antibody. MCAK was found in the EB1 but not the non-specific control immunoprecipitate. The MCAK antibody and a KIF2A antibody were also used in immunoprecipitations from extracts of COS-7 cells overexpressing EB1–GFP. Immunoblotting with a GFP antibody revealed the presence of the fusion protein in MCAK but not KIF2A precipitates.

EB1 interactions with other proteins are normally dependent upon the C-terminal portion of the protein (Morrison, 2007). To investigate if this was also the case for the association between EB1 and MCAK two EB1 C-terminal fragments were used in GST fusion protein precipitations from GFP–MCAK transfected cell extracts. GST–EB1-C84 consists of the C-terminal 84 aa of EB1 fused to GST. It is known to interact with the EB1 ligands p150Glued and APC (Askham et al., 2002). GST–EB1-bZIP was derived from the same fragment of EB1 as GST–EB1-C84 but lacks the C-terminal 27 aa of the protein. It retains an ability to interact with APC (Askham et al., 2002). Only GST–EB1-C84 precipitated Oncogene

MCAK from cell extracts (Figure 2c). This confirmed that MCAK associated with the C-terminal region of EB1, that the extreme C-terminus of EB1 was essential for this and that the association was not dependent upon the previously reported APC–MCAK interaction (Banks and Heald, 2004). We next investigated whether MCAK associated with the other members of the human EB family, EB3 and RP1. C-terminal fragments consisting of the last 89 aa of each protein were made as GST fusion proteins and used in precipitations from extracts of cells transfected with either GFP–MCAK or GFP–KIF2A. GST–EB3C89 specifically precipitated GFP–MCAK from cell


2497

extracts (Figure 2d,f). GST–RP1-C89 also precipitated GFP-MCAK from cell extracts, but much less efficiently than GST-EB1, GST-EB1-C84 or GST-EB3-C89 (Figures 2e and f). We concluded that EB1 and EB3 shared an ability to associate with MCAK whereas RP1 was less likely to do so. RP1 is the most divergent member of the human EB1 protein family (Morrison, 2007) and this conclusion is consistent with the poor binding of EB1 and EB3 ligands to RP1 reported in previous studies (Bu and Su, 2003). Finally, we found that endogenous MCAK coimmunoprecipitated with endogenous EB1 (Figure 2g), indicating that the association between these proteins was not an artifact of overexpression. Confirming this using the MCAK antibody for precipitation was less successful, although this technique was capable of detecting an association between endogenous MCAK and overexpressed EB1-GFP. It is unclear whether this observation represents a limitation of the MCAK antibody used in this work or masking of the epitope recognized by this antibody when MCAK is associated with EB1. MCAK possesses a central motor domain flanked by N- and C-terminal domains with a smaller neck region located between the N-terminal and motor domains. All of these flanking regions are involved in the targeting of the protein and the regulation of motor domain activity (Moores and Milligan, 2006; Ems-McClung et al., 2007). Interestingly, when GFP–MCAK transfected cell extracts were immunoblotted using an anti-GFP antibody a reproducible pattern of degradation products was seen (Figures 2b–e). Since GFP is tightly folded and resistant to degradation in COS-7 cells (Riess et al., 2005), the observed fragments are likely to correspond to increasingly shorter fragments of the MCAK N-terminal domain fused to GFP. The largest of these fragments therefore represented approximately the first 20 kDa of MCAK fused to the 28 kDa GFP protein. We would predict that this fragment includes the entire MCAK N-terminal domain, though at present we cannot confirm whether it also includes any of the MCAK neck regions. This large fragment was precipitated by all of the GST fusion proteins that also precipitated full-length GFP–MCAK (Figures 2c and d, asterisk). We therefore conclude that EB1 associates with the MCAK N-terminus and possibly neck region, a statement supported by previously published work indicating that the ability of GFP–MCAK to tip-track is negatively regulated by phosphorylation of the N-terminal domain (Moore et al., 2005). Notably however, recombinant EB1 proteins appeared to possess a lower affinity for the GFP–MCAK fragment than for full-length GFP– MCAK in the same sample, suggesting that other regions in MCAK might contribute to the association with EB1. One possibility is that the MCAK tail, previously shown to contribute to the MT tip-tracking ability of the protein (Moore et al., 2005) contributes to the association with EB1. This is appealing since it suggests that an EB1 association with an inactive form of MCAK where the tail is bound to the neck region (Ems-McClung et al., 2007) might be plausible. Another possibility is that dimeric MCAK, a conformation

dependent upon the C-terminal domain, might associate more efficiently with EB1. This is an attractive concept since EB1 itself can exist as a dimer in vivo (Morrison, 2007). We have previously identified an APC fragment, APC-C1, which precipitated EB1 from cell extracts (Askham et al., 2000). We have also shown that this fragment competed for binding to EB1 with p150Glued (Askham et al., 2002). In the present work, although GST–APC-C1 precipitated EB1 from extracts of COS-7 cells (Figure 3c), no GFP–MCAK was detected in the precipitates (Figure 3a). This suggested that the association between MCAK and EB1 competed with the binding of EB1 to other ligands. To confirm this we immobilized GST–EB1 on glutathione-sepharose beads at concentrations ensuring the saturation of bead binding capacity. These beads were then pre-incubated for 5 min at 4 1C with either GST or GST–APC-C1, washed briefly to remove unbound protein, then used in precipitations from the extracts of cells transfected with GFP–MCAK. Pre-incubation with GST had no effect upon the ability of GST–EB1 to precipitate GFP–MCAK from cell extracts. However, pre-incubation with GST–APC-C1 inhibited the precipitation of GFP–MCAK in a concentration-dependent manner (Figure 3b). Non-specific protein staining confirmed that GST–APC-C1 bound to the bead-immobilized GST–EB1 rather than displacing it from the beads (data not shown). We conclude that the MCAK association with EB1 is directly competitive with that of APC and hence is likely to be competitive with other EB1 ligands. To further define the EB1 binding region in MCAK we obtained a commercially available active fragment of the protein, consisting primarily of the neck and motor domains, in the form of a GST fusion protein. This was used in precipitation experiments from COS-7 cell extracts. No EB1 was detected in these precipitates (Figure 3c). We conclude that the motor region of MCAK is not involved in mediating the association with EB1, while the neck region is also incapable of mediating the association in the absence of the N-terminal domain. We next examined whether the association between EB1 and MCAK required the presence of intact MTs. GST–EB1 was used to precipitate GFP–MCAK from cells treated with the MT depolymerizing drug nocodazole under conditions known to destroy essentially all cytoplasmic MTs (Morrison et al., 1998), or cells treated with the MT stabilizing drug taxol that has previously been shown to antagonize the polarized plus-end distribution of EB1 (Morrison et al., 1998) and to induce a non-polarized association of GFP–MCAK with MTs in transfected cells (Maney et al., 2001). Treatment with either MT poison had no effect upon the precipitation of GFP–MCAK (Figure 3d), indicating that the EB1–MCAK association is not dependent upon either intact MTs or MT plus-end dynamics. Finally, in order to determine whether EB1 plays a role in targeting MCAK to growing MT ends we used siRNA technology to knock down EB1 expression in HeLa cells (Figure 4). This cell line was used in preference to the COS-7 cells used elsewhere in this study Oncogene


2498

Figure 3 The EB1–MCAK association is competitive with other EB1 ligands. (a) GST–APC-C1-coated beads were used in precipitations from GFP, GFP–MCAK and GFP–KIF2A-transfected cell extracts. Precipitates and samples of cell extract were subjected to SDS–PAGE and western blotting with an anti-GFP antibody. No GFP immunoreactive bands were seen. P ¼ precipitate, Ext ¼ cell extract. (b) Beads saturated with GST–EB1 were pre-incubated with a 5-fold molar excess of GST, with varying quantities of the EB1-binding protein GST–APC-C1, or were left untouched. The beads were then used in precipitations from GFP–MCAKtransfected cell extracts. Precipitates were subjected to SDS–PAGE and western blotting with an anti-GFP antibody. Precipitations using untreated beads or beads pre-incubated with GST contained GFP–MCAK. Beads pre-incubated with GST–APC-C1 precipitated less GFP–MCAK in a concentration-dependent manner. (c) A GST–MCAK neck/motor protein was immobilized on beads and used in precipitations from COS-7 cell extracts. GST and GST–APC-C1 were used as negative and positive controls respectively. Precipitates were subjected to SDS–PAGE and western blotting with an anti-EB1 antibody. Only the GST–APC-C1 precipitate contained EB1. (d) Bead-bound GST–EB1 was used to precipitate GFP–MCAK from untreated cell extracts, extracts of cells treated with 20 mg ml1 nocodazole for 1 h and extracts of cells treated with 10 mm taxol for 1 h. Precipitates were subjected to SDS–PAGE and western blotting with an anti-GFP antibody. GFP–MCAK was precipitated equally well in all conditions tested.

because they displayed a more consistent GFP–MCAK expression level following transient transfection, an important prerequisite for our analyses. Furthermore, it was recently demonstrated that the suppression of EB1 expression in these cells was sufficient to remove other tip-tracking EB1 binding partners from MT ends (Watson and Stephens, 2006). MT tip-associated GFP– MCAK was readily observed in transfected HeLa cells (Figure 4a). Fluorescence microscopy was used to define whether GFP–MCAK was localized at MT ends in the individual transfected cells and immunofluorescence was used to examine how this correlated with the expression level of EB1 in that cell. We observed that EB1 was expressed at a consistent expression level in control HeLa cell cultures while our experimental protocol (see Figure 4) knocked down this expression in a variable manner. Close examination revealed that EB1 was still present at MT ends in the cells expressing low levels of the protein, confirming that growing MT ends were still present (not shown). We identified 74 cells expressing GFP–MCAK in EB1 siRNA-treated cultures that met our criteria for further analysis. Briefly, cells with a grossly abnormal morphology or that were judged to be over- or underexpressing GFP–MCAK on the basis of fluorescence intensity were excluded. We found that GFP–MCAK was observed at MT ends in only 5 of 30 (17%) transfected cells where EB1 expression was suppressed compared to 42 of 44 (95%) transfected cells where EB1 expression was normal, supporting the hypothesis that EB1 is required for MCAK targeting to the end of growing MTs. Oncogene

Studies on the interactions between EB1 and other þ TIPs have indicated that ligand-binding activates the MT growth-promoting activity of EB1 (Nakamura et al., 2001; Hayashi et al., 2005). It is therefore possible that the MCAK association also serves to activate EB1. This paradoxical function for an MT depolymerizing kinesin implies that the EB1 association plays an important part in the normal function of MCAK. The most likely role for the association is the targeting of MCAK to the MT plus end, as suggested by the results of our EB1 siRNA studies. This would efficiently locate MCAK at its primary site of action, albeit presumably in an inactive form. If the association between EB1 and MCAK were regulated by MCAK phosphorylation as has been demonstrated for other EB1 binding partners (Askham et al., 2000) and as implied by the observation that MCAK tip-tracking is inhibited by N-terminal domain phosphorylation (Moore et al., 2005), then a simple regulatory mechanism for the control of MT dynamicity would exist. This model could easily incorporate the findings of previous investigators who have shown that MCAK can display onedimensional lattice diffusion along with MTs in vitro (Helenius et al., 2006), since a means of maintaining an association with the MT following MCAK release from EB1 would remain advantageous to the function of the protein. The elegance of such a mechanism would be enhanced if MCAK activity were directly suppressed by binding to EB1. Consistent with this, previous investigators have shown that overexpression of EB1 can antagonize the


2499

Figure 4 EB1 is required for the GFP–MCAK MT-tip association in transfected cells. HeLa cells were transfected with individual EB1 siRNAs (Silencer pre-designed siRNAs specific for target gene MAPRE1; siRNA ID numbers 3891, 136499 and 136501), pooled EB1 siRNAs, or a control siRNA (Silencer negative control number 1) using the siPORT NeoFX reagent. All siRNAs were used at a final concentration of 100 nM and according to protocols specified by supplier (Ambion, Foster City, CA, USA). 36 h after the cells were transfected with the GFP–MCAK expression vector and 18 h after this they were fixed and immunostained for EB1. Parameters for EB1 and GFP–MCAK fluorescence imaging were optimized using untreated control cells and settings were left unchanged for the imaging of siRNA-treated cultures. EB1 expression was classified as normal or low on the basis of fluorescence intensity. (a) In cells with normal EB1 expression levels GFP–MCAK colocalized with EB1 at MT ends (arrows), while in cells with low EB1 expression levels (arrowhead), GFP–MCAK was diffusely distributed. Bar ¼ 20 mm. (b) Summary of the data obtained by fluorescence microscopy. Results shown are pooled data derived from two independent experiments, both of which gave similar results. All three siRNA sequences used in this study knocked down EB1 expression while the control siRNA did not (not shown).

Oncogene


2500

MT depolymerizing activity of overexpressed MCAK in co-transfected cells (Moore et al., 2005). Furthermore, although we found no evidence for a direct interaction between EB1 and the motor domain of MCAK our observations show that an EB1 association site lies in the N-terminal domain recognized to be involved in both the intracellular targeting of the protein (Moores and Milligan, 2006) and the regulation of its activity (Ems-McClung et al., 2007). We therefore propose that EB1 binding at this site holds MCAK in an inactive conformation. Given recent progress in our understanding of the roles played by EB1 and MCAK in mitosis (Kline-Smith et al., 2004; Draviam et al., 2006; Knowlton et al., 2006; Ohi et al., 2007), the possibility that these proteins could directly influence one another’s respective activities at MT plus ends in mitotic cells is intriguing.

In conclusion, we have confirmed that MCAK and EB1 colocalize at growing MT plus ends in mammalian cells show that they associate in cell extracts through the domains known to be responsible for the regulation of their competing activities at the MT tip and observed that the knockdown of EB1 expression inhibits the MT tip association of GFP–MCAK in transfected cells. These observations provide the basis for a mechanism that could play an important role in controlling MT dynamics in mammalian cells.

Acknowledgements This study was supported by Cancer Research UK and Yorkshire Cancer Research.

References Askham JM, Moncur PM, Markham AF, Morrison EE. (2000). Regulation and function of the interaction between the APC tumour suppressor protein and EB1. Oncogene 19: 1950–1958. Askham JM, Vaughan KT, Goodson HV, Morrison EE. (2002). Evidence that an interaction between EB1 and p150(Glued) is required for the formation and maintenance of a radial microtubule array anchored at the centrosome. Mol Biol Cell 13: 3627–3645. Banks JD, Heald R. (2004). Adenomatous polyposis coli associates with the microtubule-destabilizing protein XMCAK. Curr Biol 14: 2033–2038. Bu W, Su MK. (2003). Characterisation of functional domains of human EB1 family proteins. J Biol Chem 278: 49721–49731. Desai A, Mitchison TJ. (1997). Microtubule polymerization dynamics. Annu Rev Cell Dev Biol 13: 83–117. Draviam VM, Shapiro I, Aldridge B, Sorger PK. (2006). Misorientation and reduced stretching of aligned sister kinetochores promote chromosome missegregation in EB1- or APC-depeleted cells. EMBO J 25: 2814–2827. Ems-McClung SC, Hertzer KM, Zhang X, Miller MW, Walczak CE. (2007). The interplay of the N- and C-terminal domains of MCAK control microtubule depolymerization activity and spindle assembly. Mol Biol Cell 18: 282–294. Hayashi I, Wilde A, Mal TK, Ikura M. (2005). Structural basis for the activation of microtubule assembly by the EB1 and p150(Glued) complex. Mol Cell 19: 449–460. Helenius J, Brouhard G, Kalaidzidis Y, Diez S, Howard J. (2006). The depolymerizing kinesin MCAK uses lattice diffusion to rapidly target microtubule ends. Nature 441: 115–119. Kinoshita K, Noetzel TL, Arnal I, Dreschel DN, Hyman AA. (2006). Global and local control of microtubule destabilization promoted by a catastrophe kinesin MCAK/XKCM1. J Musc Res Cell Mot 27: 107–114. Kline-Smith SL, Khodjakov A, Hergert P, Walczak CE. (2004). Depletion of centromeric MCAK leads to chromosome congression and segregation defects due to improper kinetochore attachments. Mol Biol Cell 15: 1146–1159. Knowlton AL, Lan W, Stukenberg PT. (2006). Aurora B is enriched at merotelic attachment sites, where it regulates MCAK. Curr Biol 16: 1705–1710.

Langford KJ, Askham JM, Lee T, Adams M, Morrison EE. (2006). Examination of actin and microtubule dependent APC localisations in living mammalian cells. BMC Cell Biol 7: 3. Lansbergen G, Akhmanova A. (2006). Microtubule plus end: a hub of cellular activities. Traffic 7: 499–507. Maney T, Wagenbach M, Wordeman L. (2001). Molecular dissection of the microtubule depolymerizing activity of mitotic centromere-associated kinesin. J Biol Chem 276: 34753–34758. Mimori-Kiyosue Y, Shiina N, Tsukita S. (2000). The dynamic behaviour of the APC-binding protein EB1 on the distal ends of microtubules. Curr Biol 10: 865–868. Moore AT, Rankin KE, von Dassow G, Peris L, Wagenbach M, Ovechkina Y et al. (2005). MCAK associates with the tips of polymerizing microtubules. J Cell Biol 169: 391–397. Moores CA, Milligan RA. (2006). Lucky 13-microtubule depolymerisation by kinesin-13 motors. J Cell Sci 119: 3905–3913. Morrison EE. (2007). Action and interactions at microtubule ends. Cell Mol Life Sci 64: 307–317. Morrison EE, Wardleworth B, Askham J, Markham AF, Meredith DM. (1998). EB1, a protein which interacts with the APC tumour suppressor, is associated with the microtubule cytoskeleton throughout the cell cycle. Oncogene 17: 3471–3477. Nakamura M, Zhou XZ, Lu KP. (2001). Critical role for the EB1 and APC interaction in the regulation of microtubule polymerization. Curr Biol 11: 1062–1067. Ohi R, Burbank K, Liu Q, Mitchison TJ. (2007). Nonredundant functions of Kinesin-13s during meiotic spindle assembly. Curr Biol 17: 953–959. Riess NP, Milward K, Lee T, Adams M, Askham JM, Morrison EE. (2005). Trapping of normal EB1 ligands in aggresomes formed by an EB1 deletion mutant. BMC Cell Biol 6: 17. Tirnauer JS, Grego S, Salmon ED, Mitchison TJ. (2002). EB1-microtubule interactions in Xenopus egg extracts: role of EB1 in microtubule stabilization and mechanisms of targeting to microtubules. Mol Biol Cell 13: 3614–3626. Watson P, Stephens DJ. (2006). Microtubule plus-end loading of p150(Glued) is mediated by EB1 and CLIP-170 but is not required for intracellular membrane traffic in mammalian cells. J Cell Sci 119: 2758–2767.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc). Oncogene