Scholarly article on topic 'An unconventional interaction between Dis1/TOG and Mal3/EB1 promotes the fidelity of chromosome segregation'

An unconventional interaction between Dis1/TOG and Mal3/EB1 promotes the fidelity of chromosome segregation Academic research paper on "Biological sciences"

CC BY
0
0
Share paper
Academic journal
Journal of Cell Science
OECD Field of science
Keywords
{""}

Academic research paper on topic "An unconventional interaction between Dis1/TOG and Mal3/EB1 promotes the fidelity of chromosome segregation"

© 2016. Published by The Company of Biologists Ltd.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.

An unconventional interaction between Dis1/TOG and Mal3/EB1 promotes

the fidelity of chromosome segregation

Yuzy Matsuo1,2, Sebastian P. Maurer1,3,4, Masashi Yukawa5, Silva Zakian6, Martin R Singleton6, Thomas Surrey1,7 and Takashi Toda2,5,7

1Synthetic and Systems Biochemistry of the Microtubule Cytoskeleton Laboratory, The Francis Crick Institute, 1 Midland Road, London, NW1 1AT, UK

2Cell Regulation Laboratory, The Francis Crick Institute, 44 Lincoln's Inn Fields, London, WC2A 3LY, UK

3Centre for Genomic Regulation (CRG), Barcelona Institute of Science and Technology (BIST), Dr. Aiguader 88, 08003 Barcelona, Spain 4Universitat Pompeu Fabra (UPF), Barcelona, Spain

5Hiroshima Research Center for Healthy Aging (HiHA), Department of Molecular Biotechnology, Graduate School of Advanced Science of Matter, Hiroshima University, 1-31 Kagamiyama, Higashi-Hiroshima, 739-8530, Japan

6Structural Biology of Chromosome Segregation Laboratory, The Francis Crick Institute, 1 Midland Road, London, NW1 1AT, UK

7 Corresponding authors: Thomas Surrey, Thomas.Surrey@crick.ac.uk; Takashi Toda, takashi-toda@hiroshima-u.ac.jp

Keywords: Crystallography/EB1/microtubule polymerase/TIRF microscopy/XMAP215-TOG

Abbreviations

CH, calponin-homology; EBH, EB homology; EBs, EB1 family proteins, HA, hemagglutinin; MACF, microtubule actin cross-linking factor; MAP, microtubule-associated protein; MT, microtubule; TBZ, thiabendazole; +TIPs, plus-end tracking proteins; TIRF-M, total internal reflection fluorescence microscopy

JCS Advance Online Article. Posted on 21 November 2016

ABSTRACT

Dynamic microtubule plus ends interact with various intracellular target regions such as the cell cortex and the kinetochore. Two conserved families of microtubule plus-end tracking proteins, XMAP215/TOG and EB1, play pivotal roles in regulating microtubule dynamics. Here we study the functional interplay between fission yeast Dis1/XMAP215 and Mal3/EB1. Using an in vitro microscopy assay, we find that purified Dis1 autonomously tracks growing microtubule ends and is a bona fide microtubule polymerase. Mal3 recruits additional Dis1 to microtubule ends, explaining the synergistic enhancement of microtubule dynamicity by these proteins. A non-canonical binding motif in Dis1 mediates the interaction with Mal3. X-ray crystallography shows that this novel motif interacts in an unconventional configuration with the conserved hydrophobic cavity formed within the Mal3 C-terminal region that typically interacts with the canonical SXIP motif. Selectively perturbing the Mal3-Dis1 interaction in living cells demonstrates that it is important for accurate chromosome segregation. Whereas in some metazoans the EB1-XMAP215/TOG interaction requires an additional binding partner, fission yeast relies on a direct interaction, indicating evolutionary plasticity of this critical interaction module.

INTRODUCTION

Microtubules (MTs) are structurally polar and highly dynamic tubulin polymers that undergo spontaneous transitions from growing to shrinking phases called dynamic instability (Mitchison and Kirschner, 1984). Such dynamic properties of MTs play an essential role in many cellular processes including intracellular transport, cell polarity and cell division (Desai and Mitchison, 1997). MT dynamics are regulated by a cohort of evolutionarily conserved microtubule-associated proteins (MAPs) (Akhmanova and Steinmetz, 2008). A subclass of MAPs, microtubules plus-end tracking proteins (+TIPs) have unique properties, as they can specifically interact with the dynamic ends of MTs, thereby playing a decisive role in determining the characteristics of MT plus ends in cells (Buey et al., 2012; Duellberg et al., 2013). In particular, EB1 and XMAP215 are important +TIPs because they can both accumulate autonomously at MT ends independent of other MAPs, however by different mechanisms (Akhmanova and Hoogenraad, 2005).

EB1 family proteins (EBs) bind in a nucleotide-dependent manner to an extended region at the growing MT end, resulting in a comet-like appearance at MT plus ends in cells (Bieling et al., 2008; Bieling et al., 2007; Maurer et al., 2011; Maurer et al., 2014; Maurer et al., 2012; Mohan et al., 2013; Zanic et al., 2009; Zhang et al., 2015). EBs can have direct and indirect effects on the MT dynamics (Galjart, 2010); several in vitro experiments suggest that purified EBs promote the MT growth rate and simultaneously increase the catastrophe frequency (Bieling et al., 2007; Li et al., 2012; Mohan et al., 2013; Vitre et al., 2008; Zanic et al., 2013).

In vivo EBs recruit several other MAPs to MT plus ends through direct proteinprotein interactions. EBs consist of four functional regions; the N-terminal calponin-homology (CH) domain required for MT binding (Hayashi and Ikura, 2003), the medial coiled-coil region involved in homo-dimerisation (De Groot et al., 2010) followed by the EB homology (EBH) domain and finally the C-terminal EEY/F motif (Duellberg et al., 2013). The EBH domain specifically binds to an SXIP motif found in a variety of +TIPs (Buey et al., 2012; Duellberg et al., 2014; Honnappa et al., 2009), whereas the EEY/F motif at the C-terminus of EBs binds to some CAP-Gly domains found in some MAPs (Duellberg et al., 2013; Honnappa et al., 2006; Weisbrich et al., 2007). MT plus end recruitment of other +TIPs by EBs is responsible for the indirect effects EBs can have on MT behaviour and hence on a variety of MT dependent cellular processes.

Mal3, the sole EB1 homologue in fission yeast Schizosaccharomycespombe, is nonessential for cell division, yet mal3 deletion mutants display a variety of defects derived

from abnormal MT architectures and dynamics. These include cell polarity defects during interphase (Beinhauer et al., 1997; Browning et al., 2003; Busch and Brunner, 2004; Busch et al., 2004) and chromosome segregation errors during mitosis (Asakawa et al., 2006; Asakawa and Toda, 2006; Asakawa et al., 2005; Beinhauer et al., 1997; Mana-Capelli et al., 2012). Mal3 has been shown to interact with the SXIP and CAP-Gly domain containing MAP Tip1, the fission yeast CLIP-170 orthologue and the Tea2 kinesin, thereby playing a critical role in regulation of interphase MT organisation and cell polarisation (Bieling et al., 2007; Browning et al., 2003; Busch et al., 2004). By contrast, our understanding of how Mal3 regulates mitotic progression remains poorly understood despite several earlier studies (Asakawa et al., 2006; Kerres et al., 2004). Work performed in vitro indicated that Mal3 alone has some impact on MT dynamics (Bieling et al., 2007; des Georges et al., 2008; Katsuki et al., 2009), however is likely that Mal3 cooperates with other +TIPs during mitosis through direct interactions as in interphase.

The XMAP215/TOG family comprises another class of +TIPs that play pivotal roles in many MT-mediated processes (Al-Bassam and Chang, 2011; Kinoshita et al., 2002; Ohkura et al., 2001). Members of this protein family contain N-terminal TOG domains that bind soluble tubulin and a separate microtubule binding site (Al-Bassam et al., 2006; Widlund et al., 2011) which in their combination allow them to act as microtubule polymerases accelerating microtubule growth (Al-Bassam et al., 2012; Ayaz et al., 2014; Ayaz et al., 2012; Brouhard et al., 2008; Li et al., 2012; Podolski et al., 2014; Reber et al., 2013; Takeshita et al., 2013). Consequently, these XMAP215/TOG proteins localise to the very MT end in contrast to EBs that bind to an extended region (Maurer et al., 2014). In the absence of tubulin, XMAP215/TOG has been shown to catalyse MT depolymerisation (Brouhard et al., 2008; Roostalu et al., 2015; Shirasu-Hiza et al., 2003).

Fission yeast contains two XMAP215/TOG orthologues, Alp14/Mtc1 and Dis1 (Garcia et al., 2001; Nakaseko et al., 2001; Ohkura et al., 2001). These two proteins share essential functions; each single deletion mutant is viable, while double deletions are inviable (Aoki et al., 2006; Garcia et al., 2002; Hsu and Toda, 2011; Kakui et al., 2013; Nabeshima et al., 1995; Nakaseko et al., 2001). Whereas Alp14 has been shown to be a microtubule polymerase like other family members (Al-Bassam et al., 2012; Hussmann et al., 2016), a biochemical characterisation of the catalytic properties of Dis1 at MT ends has not been performed yet.

In this study we explore the biochemical and physiological interplay between Dis1 and Mal3. Dis1 is a microtubule polymerase that, different from other XMAP215/TOG

family members also directly binds to Mal3. In combination, the two proteins promote MT dynamicity synergistically. Intriguingly, Disl does not have a canonical EB1 binding motif. Instead, crystallographic analysis has unveiled a non-canonical binding mode of a novel motif in Disl to the conserved hydrophobic cavity in the EBH domain that is also found in Mal3; these structural data demonstrate similarities and differences in the interaction between EB1 and the SXIP motif versus the unconventional EB binding motif in Disl. Genetic studies demonstrate that the interaction between Disl and Mal3 is of physiological significance.

RESULTS

Fission yeast Dis1 is a microtubule polymerase

We bacterially expressed and purified recombinant full-length Disl protein with a hemagglutinin (HA) tag or eGFP fused to its C-terminus (Fig. S1A). It has been demonstrated that the ability of XMAP215 to accelerate MT polymerisation is dependent on its interaction with both soluble tubulin and MTs (Widlund et al., 2011). Like XMAP215, purified Dis1-eGFP cosedimented with paclitaxel (taxol)-stabilised MTs (Fig. 1A) and also formed stable complexes with soluble tubulin dimers (Fig. 1B and Fig. S1B). These data indicate that Dis1 shares the binding characteristics of a microtubule polymerase.

To observe how Dis1-eGFP localises on dynamic MTs and how it affects their dynamic properties, we performed a total internal reflection fluorescence microscopy (TIRF-M)-based in vitro assay (Fig. 1C and Movie 1). MTs were grown from immobilised and GMPCPP-stabilised microtubule seeds (Materials and Methods) in the presence of Cy5-labelled tubulin and GTP (Bieling et al., 2010). Dis1-eGFP associated weakly with the MT lattice and accumulated on and tracked growing MT plus ends in a spot-like manner (Fig. 1C), similar to other XMAP215/TOG orthologues (Brouhard et al., 2008; Li et al., 2012; Podolski et al., 2014); by contrast, Dis1-eGFP was not observed on shrinking MT ends, similar to Alp14, the other XMAP215/TOG orthologue in fission yeast (Al-Bassam et al., 2012; Garcia et al., 2001; Hussmann et al., 2016; Nakaseko et al., 2001).

Next, we measured the effect of Dis1 on the dynamics of individual MTs (Fig. 1D). MT dynamic instability parameters, such as the average growth rate, shrinkage rate and catastrophe frequency, were quantified for different Dis1 concentrations (Table S1 and Fig. 1E). The strongest polymerase effect was measured for a concentration of 20 nM Dis1, where it accelerated the growth of MTs by -threefold, from 0.54 ± 0.02 |m/min (mean ± S.E.M, n=50) to 1.58 ± 0.09 |m/min (mean ± S.E.M, n=50: ^<0.0001 by a student-t test) (Fig. 1E). By contrast, Dis1 did not make strong impact on shrinkage rate or catastrophe frequency

(Fig. S1C). Interestingly, similarly to XMAP215 and ch-TOG (Brouhard et al., 2008; Roostalu et al., 2015), Dis1 also facilitated the depolymerisation of GMPCPP-stabilised MTs in the absence of free tubulin (Fig. 1F and Fig. S1D). These data show that Dis1 is a bona fide microtubule polymerase.

Dis1 directly binds to Mal3 and synergistically they alter the dynamic properties of microtubules

Previous studies have shown evidence for a functional interplay between members of the XMAP215/TOG and EB1 protein families (Li et al., 2012; van der Vaart et al., 2011; Zanic et al., 2013), although there are no reports showing a direct interaction between these two +TIPs. In fission yeast, the Alp14/XMAP215 localisation to MT plus ends is independent of Mal3/EB1 (Al-Bassam et al., 2012); whether this is also the case for Dis1 has not been tested. Hence, we bacterially expressed and purified recombinant full-length Mal3 proteins with or without eGFP or mCherry fused to its C-terminus (Fig. S2A). We found that GST tagged Mal3 (GST-Mal3), but not GST alone, pulled down Dis1-eGFP (Fig. 2A). Furthermore, analytical gel filtration demonstrated that recombinant Dis1 with an HA tag (Dis1-HA) and Mal3 form a stable complex in solution (Fig. 2B and Fig. S2B). These data show that Dis1 directly binds to Mal3.

Next, we examined the effect of Mal3 on the localisation of Dis1-eGFP on dynamic MTs using our TIRF-M assay (Fig. 2C). Whereas 20 nM Dis1-eGFP localised to growing MT ends again in a spot-like manner like XMAP215 (Brouhard et al., 2008; Li et al., 2012; Podolski et al., 2014) (Fig. 2C and Movie 2), upon the addition of 200 nM Mal3, the amount of Dis1-eGFP at MT ends was strongly increased; Dis1-eGFP localised now to a much more elongated region at growing MT ends, rather reminiscent of the appearance of end tracking Mal3-GFP (Fig. S2C and Movie 3) (Maurer et al., 2011). In the presence of the higher concentrations of 1 |iM Mal3, Dis1-eGFP bound to the entire MT lattice (except for the GMPCPP seeds) (Fig. 2C and Movie 4), again similar in appearance to the localisation of Mal3-GFP at this high concentration (Fig. S2C).

To simultaneously visualise the localisation of Dis1, Mal3 and MTs, we used triple colour TIRF-M. We observed that Dis1-eGFP and Mal3-mCherry colocalised to the MT end region (Fig. 2D,E and Movie 5). In control experiments, we observed that Mal3 does not directly bind to Alp14 (Fig. S2D). These results demonstrate that Mal3 recruits Dis1, but not Alp14 to the extended EB binding region at growing MT ends.

We then examined the effect of Mal3-dependent Dis1 MT end recruitment on the dynamic properties of MTs. Compared to tubulin alone, 20 nM Mal3 mildly accelerated the

growth rate by 1.6 fold (0.54 ± 0.02 |m/min vs 0.87 ± 0.04 |m/min; mean ± S.E.M, n=50: ^<0.0001 by a student-t test). 10 nM Dis1 accelerated the growth rate by 2.3 fold (1.22 ± 0.05 |m/min; mean ± S.E.M, n=50: ^<0.0001). In the presence of both 20 nM Mal3 and 10 nM Dis1, the growth rate was even more strongly increased by 4.6 fold (to 2.48 ± 0.08 |m/min; mean ± S.E.M, n=50: ^<0.0001), suggesting synergistic action of the two proteins (Table S2, Fig. 2F and Fig. S2E), reminiscent of what has been reported for the vertebrate homologues of these +TIPs (Li et al., 2012; Zanic et al., 2013). In addition, we found that Mal3 and Dis1 together substantially promoted the MT catastrophe frequency (-3.8 fold); Mal3 alone had modest impact (-1.9 fold), while Dis1 had no effect on its own (Fig. S2F). It is of note that no synergistic effect was observed for shrinkage rate (Fig. S2F). Therefore, the increased accumulation of Dis1 at MT ends by Mal3 together with Mal3's own presence strongly enhanced the dynamicity of MTs, in particular increased the growth rate and the catastrophe frequency. Taking all these data together, we conclude that Mal3 specifically binds to Dis1 but not to Alp14, thereby enhancing MT dynamicity. The C-terminal tail region of Dis1 is the primary site for Mal3 binding Although Dis1 directly binds to Mal3 (Fig. 2), there is no SXIP motif or CAP-Gly domain within the Dis1 protein. Therefore, we decided to experimentally identify the interaction site of Dis1 for Mal3 binding. Using GST pull-down assays with several truncated Dis1 constructs (Fig. 3A), we found that the C-terminal region of Dis1 (Dis1 C2: 690-882 a.a.) is the minimal region that interacts with Mal3 as efficiently as full-length Dis1 (Dis1 FL); the preceding coiled-coil region (Dis1 C3: 690-784 a.a.) alone did not bind to Mal3 (Fig. 3A,B). This data indicated that the C-terminal tail of Dis1 (784-882 a.a.) is necessary for Mal3 binding. To test this possibility, we constructed the C-terminally truncated Dis1 that lacked this region (DislAtail) and compared the binding to Mal3 between Disl FL and DislAtail. We found that the binding of DislAtail to Mal3 was significantly reduced compared to that of Dis1 FL (Fig. 3A,C).

Consistent with these pull-down assays, TIRF-M assays demonstrated that Dis1 constructs containing the C-terminal tail could track the MT plus end in the presence of Mal3 (Fig. S3A,B). Importantly, while Dis1 FL could track the MT end regardless of the presence or absence of Mal3, Dis1 C2 (690-882 a.a), which lacked the N-terminal part of Dis1 required for its autonomous end tracking ability, tracked ends only in the presence of Mal3 (Fig. S3B). Notably, Dis1 C2 in the presence of Mal3 tracked a more extended region at MT ends compared to Dis1-FL alone, which showed a spot-like appearance on the MT ends.

These data strongly suggest that the C-terminal tail region is the primary site for Mal3 binding.

Three amino acid residues in the C-terminus of Dis1 are critical for binding to Mal3

To identify the Mal3 binding site within the C-terminal tail region of Dis1, we analysed the Dis1-Mal3 interaction by means of a tiling peptide array. A membrane was spotted with 20-residue peptides covering the C-terminal region of Dis1 (505-882 a.a.) with a 2-residue start increment per spot, and then probed with Mal3-HA protein followed by immunoblotting. Remarkably, a single region consisting of 21 amino acid residues (833-853 a.a.) bound to Mal3 (Fig. 3D). To verify this interaction, we synthesised the corresponding peptide (Dis1 peptide) and conducted a competition experiment in the GST pull-down assay. Confirming the importance of the identified peptide for the Dis1-Mal3 interaction, an excess amount of the Dis1 peptide effectively abolished the binding between Mal3 and Dis1 C2 (690-882 a.a.) (Fig. 3E).

To identify the amino acid residues that are most critical for the interaction between Dis1 and Mal3, we further performed alanine scanning mutagenesis against each amino acid and found that the Dis1 residues L841, P844 and F847 are indispensable residues for Mal3 binding (Fig. 3F). To further test the importance of these key residues, we mutated them in full-length Dis1; we constructed two types of Dis1 mutants with substitutions (LAPA: L841A and P844A; LAPAFA: L841A, P844A and F847A) and compared their binding to Mal3. Both Dis1 mutant proteins bound more weakly to Mal3 in comparison to wild type Dis1 (Fig. 3G), confirming that the three amino acid residues L841, P844 and F847 in the Dis1 C-terminal region are critical for the interaction with Mal3. Since the identified binding motif differs from the canonical SXIP motif, Dis1 likely binds to Mal3 in a novel, unconventional manner.

The coiled-coil and EBH domains of Mal3 are necessary and sufficient for binding to Dis1

To identify in turn the Dis1-binding region within Mal3, several truncated Mal3 constructs were made and used for Dis1 binding assays by GST pull-down. We found that the C-terminal half (Mal3 C1: 144-308 a.a.) is responsible for binding to Dis1 (Fig. 4A,B). This C-terminal region consists of the parallel coiled-coil region (144-194 a.a.) followed by the EBH domain (197-247 a.a.) (Slep et al., 2005) (Fig. 4A). We produced fragments corresponding to these two regions (Mal3 C2 and Mal3 C3 respectively) and found that neither bound to Dis1 (Fig. 4C). However, a Mal3 construct that consists of the C-terminal part of the coiled-coil and the EBH domains (Mal3 C4: 174-247 a.a.) was sufficient to bind to Dis1 (Fig. 4A,D).

Introduction of the LAPA mutations (L841A and P844) to the Dis1 C2 construct abrogated the binding of Mal3 C4 (Fig. 4D). Taken together, the dimerised EBH domain within Mal3 is necessary and sufficient for the binding to Dis1.

Direct interaction between Dis1 and Mal3 is essential for synergistic impact on microtubule dynamics

We next examined the localisation of 10 nM Dis1-LAPA-eGFP in the presence or absence of 20 nM Mal3 on dynamic MTs using TIRF-M assays. As shown in Fig. 5A, the localisation patterns of Dis1-LAPA to the MT tip were indistinguishable regardless of the presence or absence of Mal3; they displayed thin spot-like appearances rather than thick comets seen in wild type Dis1-eGFP in the presence of Mal3. This firmly established that Dis1-LAPA fails to interact with Mal3 not only in solution but also on MTs.

We then asked whether Dis1 and Mal3 in combination are capable to alter the dynamic properties of MTs without their direct interaction. To this end, we quantified various parameters of the MT dynamics. Remarkably, no additive or synergistic impact was observed between Dis1-LAPA and Mal3 (Fig. 5B and Table S3); growth rate was 1.08 ± 0.09 |m/min (mean ± S.E.M, n=50), which was very similar to that of Dis1-LAPA alone (1.02 ± 0.04 |m/min, mean ± S.E.M), while catastrophe frequency was almost identical to that of Mal3 alone (0.22 ± 0.02 |m/min vs 0.23 ± 0.01 |m/min; mean ± S.E.M, n=50). It is of note that authentic activities of Dis1-LAPA on its own towards the MT dynamics are indistinguishable compared to wild type Dis1 (Fig. 5B and Table S3). Collectively, these results have illuminated the marked differences of XMAP215/TOG and EB1 families between fission yeast and vertebrates; while fission yeast Dis1/TOG and Mal3/EB1 need to interact directly in order to exert synergistic impacts on the MT dynamics, vertebrate orthologues are poised to do so without physical binding.

Crystallographic analysis of the Mal3-Dis1 binding interface

Previous crystal structures of the EB1 C-terminal domain showed that the highly conserved surface of the EBH domain comprises a hydrophobic cavity, which is important for the binding to SXIP containing +TIPs (Honnappa et al., 2009; Honnappa et al., 2005; Slep et al., 2005). A crystal structure of the corresponding Mal3 EBH domain has not been reported yet. To obtain structural insight into the basis of Mal3-Dis1 interaction, we crystallised Mal3 C4 (174-247 a.a.) alone and in complex with the minimal Mal3-binding Dis1 peptide (833-852 a.a.) (Fig. 3E), and solved their structures (Fig. 6A, Fig. S4A,B and Table S4). The homodimerised EBH domain of Mal3 also displayed a hydrophobic cavity, demonstrating that this structural feature is conserved (Fig. 6B). The Dis1 peptide bound to this

homodimerised EBH domain (Fig. 6A and Fig. S4A,B), and its binding was driven by predominantly by hydrophobic interactions between L841, P844 and F847 in the Disl peptide and the conserved hydrophobic cavity on the outside edge of the EBH domain of Mal3 (Fig. 6B). These structural data are fully consistent with the results of alanine scanning and pull-down assay using the mutant Disl protein with the disrupted binding motif (Fig. 3F,G).

Next, we carried out a detailed comparison of the interaction interface between Mal3-Disl and EBl-MACF (microtubule actin cross-linking factor), an EBl interactor containing the canonical SXIP motif (Honnappa et al., 2009) (Fig. 6C, D and Fig. S4C,D). We found that the 839SMLQKP844 residues in the Disl peptide are functionally equivalent to the 5477SLIPTP5482 motif in the MACF peptide, especially the L84l residue in the Disl peptide and the corresponding I5479 residue in the MACF peptide fit into the conserved hydrophobic cavity formed by the EBH domain. Intriguingly, however, we noted a clear difference between the individual interaction modes of the two peptides: while the binding interface for EBl-MCAF is provided predominantly by one EBl monomer, the interface for the Disl peptide spans two Mal3 monomers (Fig. 6E). This structural data explains the necessity of dimerisation of the EBH domain for Disl binding via the preceding coiled-coil of Mal3 (Fig. 4A,C). These findings indicate that the binding between Disl and Mal3 is indeed non-canonical, although the hydrophobic cavity is conserved in Mal3 and EBl. Physiological significance of the binding between Dis1 and Mal3

Given that Disl and Mal3 directly interact in vitro, we next examined their in vivo interaction using pull-down by GFP trap from fission yeast lysates. We found that Disl-2GFP and Mal3-5FLAG indeed formed a complex (Fig. 7A). We then created a mutant strain in which wild type dis1+-2GFP was replaced by dis1-LAPA-2GFP (L84lA and P844A). Consistent with the in vitro data (Fig. 3F,G), the pull-down assay clearly indicated that Disl-LAPA-2GFP did not bind to Mal3-5FLAG (Fig. 7A), demonstrating that the identified Mal3 binding site in Disl is relevant in the context of living cells.

In order to explore the functional significance of the direct binding between Disl and Mal3, we examined genetic interactions. It was previously shown that the double mutant between Adis1 and Aalp14 was synthetically lethal (Garcia et al., 2002) (Nakaseko et al., 200l). Interestingly, while the Amal3 mutant displayed synthetic lethality with Aalp14, Ama!3Adis1 was viable (Kerres et al., 2004). This result implies that Disl and Mal3 act in the same genetic pathway, which shares essentiality with Alpl4 (Fig. 7B). Moreover, dis1-LAPA mutants also showed synthetic lethal interaction with Aalp14 (Fig. 7B,C). It is noteworthy

that unlike Adisl, which displayed cold sensitivity (Ohkura et al., 1988), disl-LAPA cells were capable of forming colonies at low temperature (Fig. 7D). This result indicates that failure of interaction between Disl and Mal3 did not abolish all the cellular functions of Disl (Garcia et al., 2002; George and Walworth, 2015; Sanchez-Perez et al., 2005); instead by forming a complex with Mal3, Disl plays an essential role in cell division, which is compensated for by Alp14.

Although a strain containing disl-LAPA cells grew apparently normally, these cells displayed hyper-sensitivity to the microtubule-destabilising drug thiabendazole (TBZ) (Fig. 7D), which slows down growth in mutants with defects in microtubule regulation and chromosome segregation (Garcia et al., 2001; Toyoda et al., 2002; Umesono et al., 1983); note that Amal3 and Adisl cells were also individually hypersensitive to TBZ (Fig. 7D) as previously shown (Aoki et al., 2006; Beinhauer et al., 1997). We then examined MT integrity and mitotic progression in the disl-LAPA mutant. No noticeable differences in microtubule structures and intensities were observed; however, a modest yet reproducible increase in the rate of MT elongation during anaphase B (phase III, Nabeshima et al., 1998) was detected (Fig. 7E). Currently the reason for this faster anaphase B progression is unknown; nonetheless, this result uncovered that a direct interaction between Dis1 and Mal3 plays a role in the proper MT dynamics in late mitosis.

Mutations in genes required for chromosome transmission fidelity can be identified by a minichromosome loss assay (Fleig et al., 1996; Niwa et al., 1989; Takahashi et al., 1994). It was previously observed that both mal3 and disl mutants displayed a minichromosome loss phenotype (Aoki et al., 2006; Beinhauer et al., 1997). Although it was not as severe as Adisl cells (6.9%), a strain containing Dis1-LAPA lost minichromosomes at an elevated rate compared to wild type cells (2.0% vs <0.1%) (Fig. 7F). Taken together, these results indicate that the physical interaction between Dis1 and Mal3 is indeed necessary for accurate chromosome segregation.

DISCUSSION

Our study of the biochemical and physiological interplay between fission yeast Dis1/XMAP215 and Mal3/EB1 has provided five major results. First, Dis1 is a bona fide microtubule polymerase like other XMAP215/TOG family members. Second, Dis1 and Mal3 synergistically control MT dynamics. Third, this is explained by a direct interaction between Dis1 and Mal3 involving a novel EB1 binding motif. Fourth, the binding mode of this motif

in Disl to the conserved hydrophobic pocket in the EBH domain of Mal3 is unconventional. Fifth and finally, the Dis1-Mal3 interaction is of physiological significance.

Our in vitro reconstitution experiments showed that Disl is a microtubule polymerase that autonomously localises to the growing MT plus ends, thereby accelerating the MT growth rate by -threefold. This behaviour is very similar to that of Alp14, the other XMAP215/TOG orthologue in fission yeast (Al-Bassam et al., 2012; Garcia et al., 2001; Nakaseko et al., 2001). As in other members of XMAP215/TOG such as frog XMAP215, budding yeast Stu2 or fission yeast Alp14 (Al-Bassam et al., 2006; 2012; Hussmann et al., 2016; Widlund et al., 2011), the N-terminal TOG domains of Dis1 are likely to bind tubulin dimers, while the central Ser-Lys-rich (SK-rich) domain of Dis1 is probably responsible for binding to the MT lattice, which is consistent with our results with truncated Dis1-eGFP constructs (Fig. S3B) and also with the detailed domain analysis previously performed (Nakaseko et al., 1996).

The vertebrate proteins Xenopus XMAP215 and human EB1 have previously been reported to synergistically increase the MT polymerisation rate in vitro, although these two proteins do not interact directly (Zanic et al., 2013). Synergy was proposed to result from the ability of EB1 to potentially alter the curvature of the MT end structure allosterically, allowing XMAP215 to promote MT growth more efficiently (Zanic et al., 2013). Here we have also observed a synergistic enhancement of the microtubule growth rate by fission yeast Mal3 and Dis1, even if to a lesser extent. For the fission yeast proteins, synergy is likely to arise from the enhanced accumulation of Dis1 at microtubule ends as a consequence of its recruitment by Mal3 via direct interaction, as our TIRF microscopy experiments have demonstrated. Consistent with this notion, the Dis1 mutant protein with specific defects in binding to Mal3 (Dis1-LAPA) failed to exhibit additive or synergistic effects on the MT dynamics in combination with Mal3, although Dis1-LAPA on its own still retains normal enzymatic activities to a similar extent to those of wild type Dis1 (Fig. 5B). In this context it is interesting to note that in budding yeast, an interaction between Bim1/EB1 and Stu2/XMAP215 is known; however their mode of interaction is unclear (Wolyniak et al., 2006). This suggests that fungi and vertebrates have developed similar yet distinct regulatory systems by which to regulate the MT dynamics exerted by XMAP215/TOG and EB1 families.

In human cells the protein SLAIN2 acts as a linker mediating an indirect interaction between ch-TOG and EB1 (van der Vaart et al., 2011). The effect of all three human proteins together on microtubule dynamics has not been investigated yet in vitro. However, the

indirect interaction between XMAP215 and EB1 proteins is probably conserved in metazoa, as the unrelated Drosophila protein Sentin plays a similar role for Drosophila Msps/XMAP215 and EB1, mediating their indirect interaction (Li et al., 2011; Li et al., 2012). A mild synergistic effect on the growth rate of microtubules exerted by a combination of these three Drosophila proteins was also observed, possibly as a consequence of their tripartite interaction (Li et al., 2012). Taken together, it is possible that members of the XMAP215/TOG and EB1 families act collaboratively at the MT plus ends in many, if not all, eukaryotes, having a synergistic impact on MT dynamics. Their interaction can either be indirect as in metazoans, or direct as in fission yeast, demonstrating evolutionary plasticity of this regulatory module (Fig. 8A,B).

Here we have solved the first X-ray structure of the EBH domain of Mal3, the protein originally used to reconstitute autonomous microtubule end tracking by an EB protein and its recruitment of other +TIPs to growing microtubule ends (Bieling et al., 2007). Our crystallographic analysis shows the two hydrophobic pockets that are formed by the C-terminal part of the Mal3 EBH domains and to which typically recruited +TIPs bind via their canonical SxIP motif. This pocket is conserved, in agreement with the observation of SXIP motif containing proteins being recruited by Mal3 in vitro (Bieling et al., 2007). The polypeptide fold however is unconventional, because a hydrophobic pocket is formed by helices of two Mal3 polypeptides (trans-configuration), whereas in EB1 a hydrophobic pocket is formed by the corresponding helices of a single Mal3 polypeptide (cis-configuration).

The interaction of the non-canonical Dis1 peptide with the hydrophobic pocket is also unconventional compared to the binding of SXIP motif-containing peptides. Although both peptides interact strongly with the hydrophobic cavity of the EBH domain, their exact binding interface varies (Fig. 6E). Due to these differences, the binding interface in Mal3-Dis1 is at 90° to the linker between the two helices in the EB monomer, while in EB1-MACF (containing the canonical SXIP motif) it is on the same side (Fig. 6C-E and Fig. S4C,D). Hence, the SXIP motif is not be the sole motif that can interact with the EBH domain of EB 1. In fact, previous comprehensive proteomic analysis of EB1-binding proteins identified a fairly large number of interactors that do not contain the canonical SXIP sequence (Jiang et al., 2012; Tamura et al., 2015). Therefore, it is likely that non-canonical EB1 interaction motifs similar to the one we discovered in Dis1 will exist in proteins of other organisms than fission yeast, likely including also humans.

Previous work showed that Dis1 localises to the mitotic kinetochore through interaction with the Ndc80/Hec1 outer kinetochore component and plays a critical role in proper kinetochore-microtubule attachment (Hsu and Toda, 2011). It is, therefore, likely that Dis1 and Mal3 interact at the kinetochore-MT interface upon kinetochore capture by the MT plus ends. Interestingly, Alp14/TOG also localises to the mitotic kinetochore, yet in a different manner; Alp7/TACC binds to Ndc80, thereby recruiting the Alp7-Alp14 complex to the kinetochore (Sato et al., 2004; Tang et al., 2015). The result that Aalp14 and Mal3-binding defective dis1-LAPA display synthetic lethality indicates that these two fission yeast XMAP215/TOG members in concert establish and maintain kinetochore-microtubule attachment by exploiting distinct binding partners, Mal3/EB1 for Dis1/TOG and Alp7/TACC for Alp14/TOG.

Intriguingly, the interaction between XMAP215/TOG and the Ndc80 complex is conserved in budding yeast and humans (Miller et al., 2016). Importantly, XMAP215/TOG orthologues of these two organisms (Stu2 and ch-TOG) regulate kinetochore-microtubule attachment in a tension-sensitive manner. Intriguingly, in vitro work showed that the XMAP215/TOG family members play this critical role without altering MT dynamics (Miller et al., 2016). However, in cells undergoing mitosis, the fine-tuning of MT dynamics and the state of tension should temporally be coupled. We envisage that the EB1 proteins might be important proteins promoting proper kinetochore-MT attachment in response to the state of tension. Further in vitro studies by adding the EB1 proteins (and SLAIN2) to XMAP215/TOG and the Ndc80 complex would be of great interest to explore this possibility.

In conclusion, our work provides novel insight into the unconventional binding mode of Mal3/EB1 with Dis1/TOG, which is likely to indicate the possibility of further modes of interactions for other uncharacterised EB1 interactors. From an evolutionary point of view, the interaction between EB1-XMAP215 family members is probably ubiquitous; intriguingly however, the mechanism by which they interact varies among different organisms, illuminating diversification and remarkable adaptability through evolution.

MATERIALS AND METHODS Protein expression and purification

dis1 cDNA optimised for E. coli codon was synthesised (GeneArt Gene Synthesis) and subcloned into the pETM-11 vector (Maurer et al, 2011). The purification of the various Dis1-eGFP constructs was performed as described previously (Maurer et al, 2011) except that Dis1 buffer (50 mM Tris-HCl pH7.5, 300 mM KCl, 10% (v/v) glycerol, 1 mM EDTA and 1 mM DTT) containing EDTA-free protease inhibitors (Roche) and IgG Sepharose 6 Fast Flow (GE Healthcare) were used. In addition, TEV protease (for Dis1-eGFP) or Precision protease (for Dis1-HA) was used. The purification of the Mal3, Mal3-eGFP and Mal3-mCherry was performed as described previously (Maurer et al, 2011).

The EBH domain (174-247a.a.) of Mal3 was purified by the same procedures described above except that buffer A (50 mM Tris-HCl pH7.5, 500 mM NaCl, 10 mM ß-mercaptoethanol) containing 30 mM imidazole was used for cell suspension and columns washing. In addition, His-tag purification resin (Roche) and buffer A supplemented with 500 mM imidazole were used for affinity purification and elution respectively. The cleavage of oligo-histidine tag was performed as described earlier. Mal3 proteins were further purified by gel filtration using Superose200 10/300 columns equilibrated with buffer B (20 mM Tris-HCl (pH7.5), 100 mM NaCl, 2 mM DTT).

Porcine brain tubulin was purified and cycled tubulin fractions were labelled with Cy5 (Lumiprobe) or EZ-link NHS Biotin (Thermo Scientific) by standard methods as described previously (Bieling et al, 2007). Protein concentration was determined by Bradford assay (Bio-Rad).

Microtubule cosedimentation and binding assays

The microtubule cosedimentation assay was performed as described previously (Maurer et al, 2011). 0.2 |M Dis1-eGFP was incubated with taxol-stabilised microtubules at concentrations from 0 to 4 |M in BRB80 supplemented with 85 mM KCl and 85 mM CH3COOK in the presence of 20 |M taxol. Analytical gel-filtration

5 |M Dis1-eGFP and/or 10 |M tubulin (or 5 |M Dis1-HA and/or 20 |M Mal3) was incubated in gel-filtration buffer (25 mM K-HEPES pH7.5, 200 mM KCl, 1 mM EGTA, 1 mM MgCl2) before loading on a Superose 6 10/300 (GE Healthcare) equilibrated with gelfiltration buffer. The absorbance of the eluted protein was measure at 280 nm.

TIRF microscopy and image analysis

TIRF microscopy-based dynamic microtubule assays were performed as previously described (Bieling et al., 2010). TIRF assay buffer consisted of BRB80 supplemented with 85 mM KCl, 1 mM GTP, 10 mM P-mercaptoethanol, 0.1% Brij-35, 0.1% methylcellulose (Sigma-Aldrich) and an oxygen scavenger system (glucose, glucose oxidase (Serva) and catalase (Sigma-Aldrich)). For simultaneous dual (or triple)-colour time-lapse imaging of the Cy5 and GFP (and mCherry) channel, imaging was performed at 1s intervals with 100 ms exposure time, using a magnification of x100 at 30 ± 1°C on a custom TIRF microscope equipped with a Cascade II, cooled charge-coupled device camera (Photometrics), illuminating the sample with 488 nm and 640 nm (and 561nm) lasers. Image analysis was performed as described previously (Duellberg et al., 2014). GST pull-down assay

For domain analysis of the Dis1 protein, 1|M of GST or GST-Mal3 and 0.5|M of several truncated Dis1-eGFP proteins were incubated in pull-down buffer (BRB80 supplemented with 85 mM KCl, 0.01% Brij-35 and 1mM DTT). Then, 20 |l of glutathione sepharose beads (GE Healthcare) were added and after incubation for 1 h at 4°C, beads were washed and eluted with the buffer containing 10mM reduced glutathione. For domain analysis of Mal3, 0.5 |M Dis1 C1-eGFP and 1 |M of several truncated GST-Mal3 proteins were used and the same procedures were followed.

For peptide competition assay, 1 |M of purified GST-Mal3 bound to the glutathione beads were pre-incubated with various concentration of the Dis1 peptide (833-852a.a) (0, 0.5, 2.5, 5.0, 50, 100 |M) in pull-down buffer for 30 min, followed by further incubation with 0.5 |M Dis1 C2-eGFP for 1 h. Peptide array assay

Peptide array assays were performed as previously described (Hsu and Toda, 2011). The blocked membrane was incubated with 1 |g/ml Mal3-HA protein and bound Mal3-HA was detected by immunoblotting with an anti-HA antibody (diluted 1:1000, 16B12, Covance).

X-ray crystallography

X-ray data were collected on beamlines I04 and I03 of the Diamond Light Source. Data were integrated and scaled using the Xia2 pipeline. The unliganded Mal3 EBH (hereon Mal3) domain was solved using ab initio molecular replacement as implemented in the ARCIMBOLDO-LIGHT package (Sammito et al., 2015). 20-residue a-helical segments were used as a search model. A strong (30.2% origin) native Patterson peak at 0.155, 0.5, 0.5 in the

native data suggested the presence of translational pseudo-symmetry, and appropriate correction was applied in Phaser. The preliminary poly-alanine model was then assigned sequence in Buccaneer, and manually rebuilt and refined with Coot (Emsley and Cowtan, 2004) and Refmac5 (Murshudov et al., 2011). The final model was refined with anisotropic temperature factors applied to all atoms. The Mal3-Dis1 structure was solved by molecular replacement in Phaser (McCoy et al., 2007) using the apo-Mal3 structure as a search model. Model building and refinement were carried out as for the apo-structure. Clear density could be seen for the bound Dis1 peptides in an unbiased difference map at both homodimer interfaces and were built in the final stages of refinement. All data collection and refinement statistics are provided in Table S4.

Crystallisation and structure solution

Crystals were grown by sitting-drop vapour diffusion. The Mal3174-247 proteins at 7 mg/ml were mixed with crystallisation solution comprising 0.2 M NaCl, 30% MPD and 0.1 M sodium acetate at pH4.6. Crystals grew to full size after 5 d and were flash-cooled in liquid nitrogen with no additional cryoprotection. To obtain the Mal3 EBH domain-Dis1 complex, The Mal3174-247 protein was incubated with the Dis1 peptide (residues 833-852) in a 1:5 molar ration for 1 hour at 4°C before crystallisation. Crystals were grown by sitting-drop vapour diffusion. The protein complex at 7 mg/ml was mixed with the crystallisation solution consisting of 0.2 M MgCl2, 0.1 M HEPES pH 7, 20% PEG 6K. Crystals grew to full size after a week and were flash-cooled in liquid nitrogen.

Immunoprecipitation

Immunoprecipitation was performed as previously described (Hsu and Toda, 2011) except that IP buffer (50 mM Na-HEPES, pH7.5, 150 mM NaCl, 1 mM EDTA, 0.1% Tween20, 1 mM DTT, protease inhibitor cocktail (Sigma-Aldrich), GFP-Trap A and an anti-FLAG antibody (diluted 1:2000, Anti-FLAG M2, Sigma-Aldrich) or an anti-GFP antibody (diluted 1:2000, Anti-eGFP 632569, Clontech) was used.

Fission yeast strains, media, genetic methods, minichromosome loss assay and cell biology

Fission yeast strains used in this study are listed in Table S5. Standard fission yeast methodologies were followed (Moreno et al., 1991). Spot assays were carried out by spotting 5-10 pi of cells at a concentration of 2*106 cells/ml after 10-fold serial dilutions onto rich YE5S plates with or without a drug (10 |g/ml TBZ). C-terminal tagging and gene disruption were performed using PCR generated fragments as described (Bahler et al., 1998; Sato et al.,

2005). Minichromosome loss assay was carried out as described (Niwa et al., 1989; Tange et al., 2016).

Fluorescence microscope images were obtained using the DeltaVision microscope system (Applied Precision, Inc.) with a cooled CCD camera CoolSNAP.HQ (Photometrics). Live cells were imaged in a glass-bottomed culture dish (MatTek Corporation) coated with soybean lectin. Statistical data analysis

In all the experiments in which statistical data are presented, the number of experiments and sample size are described in the main text, tables and figure legends. Results are expressed as the mean ± standard error of the mean (S.E.M) or standard deviation (SD). Statistical evaluation was performed using Prism (GraphPad). All p-values are from two-tailed unpaired student t-tests. We followed this key for asterisk placeholders for p-values in the figures: *p<0.01, n.s. not significant.

Protein Data Bank

Atomic coordinates and structure factors of Mal3174-247 and the Mal3174-247:Dis1 peptide have been deposited in the Protein Data Bank under accession codes 5M97 and 5M9E respectively.

ACKNOWLEDGEMENTS

We are grateful to Hirofumi Takada for his contribution to the initial part of this study, N. O'Reilly for peptide synthesis, A. Purkiss for assistance with X-ray data collection, Yasuto Murayama for plasmid and Yasutaka Kakui, all members of the Cell Regulation Laboratory and Microtubule Cytoskeleton Laboratory for discussion and technical advice.

COMPETING INTERESTS

The authors declare that they have no conflict of interest.

AUTHOR CONTRIBUTIONS

Y.M., S.P.M., MRS., T.M. and T.T. designed the experiments. M.Y. performed most experiments. S.P.M. performed initial protein purifications and TIRF-microscopy experiments, and S.Z. and M.R.S. performed crystallisation and X-ray data acquisition and analysis. M.Y analysed dis1-LAPA mutants under microscopy. Y.M., M.R.S., T.S. and T.T. wrote the manuscript helped by input from S.P.M., M.Y. and S.Z.

FUNDING

This work was supported by Cancer Research UK and the Francis Crick Institute which receives its core funding from Cancer Research UK (FC001155, FC001163, FC001184), the UK Medical Research Council (FC001155, FC001163, FC001184), and the Wellcome Trust (FC001155, FC001163, FC001184), the Japan Society for the Promotion of Science KAKENHI Scientific Research (A) (16H02503) (T.T.), Challenging Exploratory Research (16K14672) (T.T.), Scientific Research (C) (16K07694) (M.Y.), the Naito Foundation (T.T.) and a Marie Curie fellowship (PIEF-GA-2009-253043) (S.P.M.).

REFERENCES

Akhmanova, A. and Hoogenraad, C. C. (2005). Microtubule plus-end-tracking proteins: mechanisms and functions. Curr. Opin. Cell Biol. 17, 47-54.

Akhmanova, A. and Steinmetz, M. O. (2008). Tracking the ends: a dynamic protein network controls the fate of microtubule tips. Nat. Rev. Mol. Cell Biol. 9, 309-322. Al-Bassam, J. and Chang, F. (2011). Regulation of microtubule dynamics by TOG-domain proteins XMAP215/Dis1 and CLASP. Trends Cell Biol. 21, 604-614. Al-Bassam, J., Kim, H., Flor-Parra, I., Lal, N., Velji, H. and Chang, F. (2012). Fission yeast Alp14 is a dose dependent plus end tracking microtubule polymerase. Mol. Biol. Cell 23, 2878-2890.

Al-Bassam, J., van Breugel, M., Harrison, S. C. and Hyman, A. (2006). Stu2p binds tubulin and undergoes an open-to-closed conformational change. J. Cell Biol. 172, 10091022.

Aoki, K., Nakaseko, Y., Kinoshita, K., Goshima, G. and Yanagida, M. (2006). Cdc2 phosphorylation of the fission yeast Dis1 ensures accurate chromosome segregation. Curr. Biol. 16, 1627-1635.

Asakawa, K. and Toda, T. (2006). Cooperation of EB1-Mal3 and the Bub1 spindle checkpoint. Cell Cycle 5, 27-30.

Asakawa, K., Kume, K., Kanai, M., Goshima, T., Miyahara, K., Dhut, S., Tee, W. W., Hirata, D. and Toda, T. (2006). The V260I mutation in fission yeast a-tubulin Atb2 affects microtubule dynamics and EB1-Mal3 localization and activates the Bub1 branch of the spindle checkpoint. Mol. Biol. Cell 17, 1422-1436.

Asakawa, K., Toya, M., Sato, M., Kanai, M., Kume, M., Goshima, T., Garcia, M. A., Hirata, D. and Toda, T. (2005). Mal3, the fission yeast EB1 homolog. cooperates with Bub1 spindle checkpoint to prevent monopolar attachment, EMBO Rep. 6, 1194-1120. Ayaz, P., Ye, X., Huddleston, P., Brautigam, C. A. and Rice, L. M. (2012). A TOG aß-tubulin complex structure reveals conformation-based mechanisms for a microtubule polymerase. Science 337, 857-860.

Ayaz, P., Munyoki, S., Geyer, E. A., Piedra, F. A., Vu, E. S., Bromberg, R., Otwinowski, Z., Grishin, N. V., Brautigam, C. A. and Rice, L. M. (2014). A tethered delivery mechanism explains the catalytic action of a microtubule polymerase. eLife 3, e03069. Bähler, J., Wu, J., Longtine, M. S., Shah, N. G., McKenzie III, A., Steever, A. B., Wach, A., Philippsen, P. and Pringle, J. R. (1998). Heterologous modules for efficient and versatile PCR-based gene targeting in Schizosaccharomycespombe. Yeast 14, 943-951.

Beinhauer, J. D., Hagan, I. M., Hegemann, J. H. and Fleig, U. (1997). Mal3, the fission yeast homologue of the human APC-interacting protein EB-1 is required for microtubule integrity and the maintenance of cell form. J. Cell Biol. 139, 717-728.

Bieling, P., Kandels-Lewis, S., Telley, I. A., van Dijk, J., Janke, C. and Surrey, T. (2008). CLIP-170 tracks growing microtubule ends by dynamically recognizing composite EB1/tubulin-binding sites. J. Cell Biol. 183, 1223-1233.

Bieling, P., Laan, L., Schek, H., Munteanu, E. L., Sandblad, L., Dogterom, M., Brunner, D. and Surrey, T. (2007). Reconstitution of a microtubule plus-end tracking system in vitro. Nature 450, 1100-1105.

Bieling, P., Telley, I. A., Hentrich, C., Piehler, J. and Surrey, T. (2010). Fluorescence microscopy assays on chemically functionalized surfaces for quantitative imaging of microtubule, motor, and +TIP dynamics. Methods Cell Biol. 95, 555-580. Brouhard, G. J., Stear, J. H., Noetzel, T. L., Al-Bassam, J., Kinoshita, K., Harrison, S. C., Howard, J. and Hyman, A. A. (2008). XMAP215 is a processive microtubule polymerase. Cell 132, 79-88.

Browning, H., Hackney, D. D. and Nurse, P. (2003). Targeted movement of cell end factors in fission yeast. Nat. Cell Biol. 5, 812-818.

Buey, R. M., Sen, I., Kortt, O., Mohan, R., Gfeller, D., Veprintsev, D., Kretzschmar, I., Scheuermann, J., Neri, D., Zoete, V., Michielin, O., de Pereda, J. M., Akhmanova, A., Volkmer, R. and Steinmetz, M. O. (2012). Sequence determinants of a microtubule tip localization signal (MtLS). J. Biol. Chem. 287, 28227-28242.

Busch, K. E. and Brunner, D. (2004). The microtubule plus end-tracking proteins mal3p and tip1p cooperate for cell-end targeting of interphase microtubules. Curr. Biol. 14, 548559.

Busch, K. E., Hayles, J., Nurse, P. and Brunner, D. (2004). Tea2p kinesin is involved in spatial microtubule organization by transporting tip1p on microtubules, Dev. Cell6. 831-843. De Groot, C. O., Jelesarov, I., Damberger, F. F., Bjelic, S., Scharer, M. A., Bhavesh, N. S., Grigoriev, I., Buey, R. M., Wuthrich, K., Capitani, G., Akhmanova, A. and Steinmetz, M. O. (2010). Molecular insights into mammalian end-binding protein heterodimerization. J. Biol. Chem. 285, 5802-5814.

des Georges, A., Katsuki, M., Drummond, D. R., Osei, M., Cross, R. A. and Amos, L. A.

(2008). Mal3, the Schizosaccharomycespombe homolog of EB1, changes the microtubule lattice. Nat. Struct. Mol. Biol. 15, 1102-1108.

Desai, A. and Mitchison, T. J. (1997). Microtubule polymerization dynamics. Annu. Rev. Cell Dev. Biol. 13, 83-117.

Duellberg, C., Fourniol, F. J., Maurer, S. P., Roostalu, J. and Surrey, T. (2013). End-binding proteins and Ase1/PRC1 define local functionality of structurally distinct parts of the microtubule cytoskeleton. Trends Cell Biol. 23, 54-63.

Duellberg, C., Trokter, M., Jha, R., Sen, I., Steinmetz, M. O. and Surrey, T. (2014). Reconstitution of a hierarchical +TIP interaction network controlling microtubule end tracking of dynein. Nat. Cell Biol. 16, 804-811.

Emsley, P. and Cowtan, K. (2004). Coot: model-building tools for molecular graphics. Acta Crystallogr. D. Biol. Crystallogr. 60, 2126-2132.

Fleig, U., Sen-Gupta, M. and Hegemann, J. H. (1996). Fission yeast mal2+ is required for chromosome segregation. Mol. Cell. Biol. 16, 6169-6177.

Galjart, N. (2010). Plus-end-tracking proteins and their interactions at microtubule ends. Curr. Biol. 20, R528-R537.

Garcia, M. A., Koonrugsa, N. and Toda, T. (2002). Two kinesin-like Kin I family proteins in fission yeast regulate the establishment of metaphase and the onset of anaphase A. Curr. Biol. 12, 610-621.

Garcia, M. A., Vardy, L., Koonrugsa, N. and Toda, T. (2001). Fission yeast ch-TOG/XMAP215 homologue Alp14 connects mitotic spindles with the kinetochore and is a component of the Mad2-dependent spindle checkpoint. EMBO J. 20, 3389-3401. George, A. A. and Walworth, N. C. (2015). Escape from mitotic arrest: an unexpected connection between microtubule dynamics and epigenetic regulation of centromeric chromatin in Schizosaccharomycespombe. Genetics 201, 1467-1478. Gluszek, A. A., Cullen, C. F., Li, W., Battaglia, R. A., Radford, S. J., Costa, M. F., McKim, K. S., Goshima, G. and Ohkura, H. (2015). The microtubule catastrophe promoter Sentin delays stable kinetochore-microtubule attachment in oocytes. J. Cell Biol. 211, 11131120.

Li, W., Moriwaki, T., Tani, T., Watanabe, T., Kaibuchi, K. and Goshima, G. (2012). Reconstitution of dynamic microtubules with Drosophila XMAP215, EB1, and Sentin. J. Cell Biol. 199, 849-862.

Li, W., Miki, T., Watanabe, T., Kakeno, M., Sugiyama, I., Kaibuchi, K. and Goshima,

G. (2011). EB1 promotes microtubule dynamics by recruiting Sentin in Drosophila cells. J. Cell Biol. 193, 973-983.

Hayashi, I. and Ikura, M. (2003). Crystal structure of the amino-terminal microtubule-binding domain of end-binding protein 1 (EB1). J. Biol. Chem. 278, 36430-36434. Honnappa, S., Gouveia, S. M., Weisbrich, A., Damberger, F. F., Bhavesh, N. S., Jawhari, H., Grigoriev, I., van Rijssel, F. J., Buey, R. M., Lawera, A., Jelesarov, I., Winkler, F. K., Wuthrich, K., Akhmanova, A. and Steinmetz, M. O. (2009). An EB1-binding motif acts as a microtubule tip localization signal. Cell 138, 366-376. Honnappa, S., John, C. M., Kostrewa, D., Winkler, F. K. and Steinmetz, M. O. (2005). Structural insights into the EB1-APC interaction. EMBO J. 24, 261-269. Honnappa, S., Okhrimenko, O., Jaussi, R., Jawhari, H., Jelesarov, I., Winkler, F. K. and Steinmetz, M. O. (2006). Key interaction modes of dynamic +TIP networks. Mol. Cell 23, 663-571.

Hsu, K. S. and Toda, T. (2011). Ndc80 internal loop interacts with Dis1/TOG to ensure proper kinetochore-spindle attachment in fission yeast. Curr. Biol. 21, 214-220. Hussmann, F., Drummond, D. R., Peet, D. R., Martin, D. S. and Cross, R. A. (2016). Alp7/TACC-Alp14/TOG generates long-lived, fast-growing MTs by an unconventional mechanism. Sci. Rep. 6, 20653.

Jiang, K., Toedt, G., Montenegro Gouveia, S., Davey, N. E., Hua, S., van der Vaart, B., Grigoriev, I., Larsen, J., Pedersen, L. B., Bezstarosti, K., Lince-Faria, M., Demmers, J., Steinmetz, M. O., Gibson, T. J. and Akhmanova, A. (2012). A proteome-wide screen for mammalian SxIP motif-containing microtubule plus-end tracking proteins. Curr. Biol. 22, 1800-1807.

Kakui, Y., Sato, M., Okada, N., Toda, T. and Yamamoto, M. (2013). Microtubules and Alp7-Alp14 (TACC-TOG) reposition chromosomes before meiotic segregation. Nat. Cell Biol. 15, 786-796.

Katsuki, M., Drummond, D. R., Osei, M. and Cross, R. A. (2009). MAL3 masks catastrophe events in Schizosaccharomyces pombe microtubules by inhibiting shrinkage and promoting rescue. J. Biol. Chem. 284, 29246-29250.

Kerres, A., Vietmeier-Decker, C., Ortiz, J., Karig, I., Beuter, C., Hegemann, J., Lechner, J. and Fleig, U. (2004). The fission yeast kinetochore component Spc7 associates with the EB1 family member Mal3 and is required for kinetochore-spindle association. Mol. Biol. Cell 15, 5255-5267.

Kinoshita, K., Habermann, B. and Hyman, A. A. (2002). XMAP215: a key component of the dynamic microtubule cytoskeleton. Trends Cell Biol. 6, 267-273.

Mana-Capelli, S., McLean, J. R., Chen, C. T., Gould, K. L. and McCollum, D. (2012). The kinesin-14 Klp2 is negatively regulated by the SIN for proper spindle elongation and telophase nuclear positioning. Mol. Biol. Cell 23, 4592-4600. Maurer, S. P., Bieling, P., Cope, J., Hoenger, A. and Surrey, T. (2011). GTPyS microtubules mimic the growing microtubule end structure recognized by end-binding proteins (EBs). Proc. Natl. Acad. Sci. USA 108, 3988-3993.

Maurer, S. P., Cade, N. I., Bohner, G., Gustafsson, N., Boutant, E. and Surrey, T.

(2014). EB1 accelerates two conformational transitions important for microtubule maturation and dynamics. Curr. Biol. 24, 372-384.

Maurer, S. P., Fourniol, F. J., Bohner, G., Moores, C. A. and Surrey, T. (2012). EBs recognize a nucleotide-dependent structural cap at growing microtubule ends. Cell 149, 371382.

McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C. and Read, R. J. (2007). Phaser crystallographic software. J. Appl. Crystallogr. 40, 658-674. Miller, M. P., Asbury, C. L. and Biggins, S. (2016). A TOG protein confers tension sensitivity to kinetochore-mcrotubule attachments. Cell 165, 1428-1439. Mitchison, T. and Kirschner, M. W. (1984). Dynamic instability of microtubule growth. Nature 312, 237-242.

Moreno, S., Klar, A. and Nurse, P. (1991). Molecular genetic analysis of fission yeast Schizosaccharomycespombe. Methods Enzymol. 194, 795-823.

Murshudov, G. N., Skubak, P., Lebedev, A. A., Pannu, N. S., Steiner, R. A., Nicholls, R. A., Winn, M. D., Long, F. and Vagin, A. A. (2011). REFMAC5 for the refinement of macromolecular crystal structures. Acta Crystallogr. D. Biol. Crystallogr. 67, 355-67. Nabeshima, K., Kurooka, H., Takeuchi, M., Kinoshita, K., Nakaseko, Y. and Yanagida,

M. (1995). p93dis1, which is required for sister chromatid separation, is a novel microtubule and spindle pole body-associating protein phosphorylated at the Cdc2 target sites. Genes Dev. 9, 1572-1585.

Nabeshima, K., Nakagawa, T., Straight, A.F., Murray, A., Chikashige, Y., Yamashita, Y.M., Hiraoka, Y., and Yanagida, M. (1998). Dynamics of centromeres during metaphase-anaphase transition in fission yeast: Dis1 is implicated in force balance in metaphase bipolar spindle. Mol. Biol. Cell 9, 3211-3225.

Nakaseko, Y., Goshima, G., Morishita, J. and Yanagida, M. (2001). M phase-specific kinetochore proteins in fission yeast microtubule-associating Dis1 and Mtc1 display rapid separation and segregation during anaphase. Curr. Biol. 11, 537-549.

Nakaseko, Y., Nabeshima, K., Kinoshita, K. and Yanagida, M. (1996). Dissection of fission yeast microtubule associating protein p93dis1: regions implicated in regulated localization and microtubule interaction. Genes Cells 1, 633-644.

Niwa, O., Matsumoto, T., Chikashige, Y. and Yanagida, M. (1989). Characterization of Schizosaccharomycespombe minichromosome deletion derivatives and a functional allocation of their centromere. EMBO J. 8, 3045-3052.

Ohkura, H., Adachi, Y., Kinoshita, N., Niwa, O., Toda, T. and Yanagida, M. (1988). Cold-sensitive and caffeine supersensitive mutants of the Schizosaccharomyces pombe dis genes implicated in sister chromatid separation during mitosis. EMBO J. 7, 1465-1473. Ohkura, H., Garcia, M. A. and Toda, T. (2001). Dis1/TOG universal microtubule adaptors - one MAP for all?. J. CellSci. 114, 3805-3812.

Podolski, M., Mahamdeh, M. and Howard, J. (2014). Stu2, the budding yeast XMAP215/Dis1 homolog, promotes assembly of yeast microtubules by increasing growth rate and decreasing catastrophe frequency. J. Biol. Chem. 289, 28087-28093. Reber, S. B., Baumgart, J., Widlund, P. O., Pozniakovsky, A., Howard, J., Hyman, A. A. and Julicher, F. (2013). XMAP215 activity sets spindle length by controlling the total mass of spindle microtubules. Nat. Cell Biol. 15, 1116-1122.

Roostalu, J., Cade, N. I. and Surrey, T. (2015). Complementary activities of TPX2 and chTOG constitute an efficient importin-regulated microtubule nucleation module. Nat. Cell Biol. 17, 1422-1434.

Sammito, M., Millan, C., Frieske, D., Rodriguez-Freire, E., Borges, R. J. and Uson, I.

(2015). ARCIMBOLDO_LITE: single-workstation implementation and use. Acta Crystallogr. D. Biol. Crystallogr. 71, 1921-1930.

Sanchez-Perez, I., Renwick, S. J., Crawley, K., Karig, I., Buck, V., Meadows, J. C., Franco-Sanchez, A., Fleig, U., Toda, T. and Millar, J. B. (2005). The DASH complex and Klp5/Klp6 kinesin coordinate bipolar chromosome attachment in fission yeast. EMBO J. 24, 2931-2943.

Sato, M., Dhut, S. and Toda, T. (2005). New drug-resistant cassettes for gene disruption and epitope tagging in Schizosaccharomyces pombe. Yeast 22, 583-591. Sato, M., Vardy, L., Garcia, M. A., Koonrugsa, N. and Toda, T. (2004). Interdependency of fission yeast Alp14/TOG and coiled coil protein Alp7 in microtubule localization and bipolar spindle formation. Mol. Biol. Cell 15, 1609-1622.

Shirasu-Hiza, M., Coughlin, P. and Mitchison, T. (2003). Identification of XMAP215 as a microtubule-destabilizing factor in Xenopus egg extract by biochemical purification. J. Cell Biol. 161, 349-358.

Slep, K. C., Rogers, S. L., Elliott, S. L., Ohkura, H., Kolodziej, P. A. and Vale, R. D.

(2005). Structural determinants for EB1-mediated recruitment of APC and spectraplakins to the microtubule plus end. J. Cell Biol. 168, 587-598.

Takahashi, K., Yamada, H. and Yanagida, M. (1994). Fission yeast minichromosome loss mutants mis cause lethal aneuploidy and replication abnormality. Mol. Biol. Cell 5, 11451158.

Takeshita, N., Mania, D., Herrero, S., Ishitsuka, Y., Nienhaus, G. U., Podolski, M., Howard, J. and Fischer, R. (2013). The cell-end marker TeaA and the microtubule polymerase AlpA contribute to microtubule guidance at the hyphal tip cortex of Aspergillus nidulans to provide polarity maintenance. J. Cell Sci. 126, 5400-5411. Tamura, N., Simon, J. E., Nayak, A., Shenoy, R., Hiroi, N., Boilot, V., Funahashi, A. and Draviam, V. M. (2015). A proteomic study of mitotic phase-specific interactors of EB1 reveals a role for SXIP-mediated protein interactions in anaphase onset. Biol. Open 4, 155169.

Tange, Y., Chikashige, Y., Takahata, S., Kawakami, K., Higashi, M., Mori, C., Kojidani, T., Hirano, Y., Asakawa, H., Murakami, Y., Haraguchi, T. and Hiraoka, Y.

(2016). Inner nuclear membrane protein Lem2 augments heterochromatin formation in response to nutritional conditions. Genes Cells 21, 812-832.

Toyoda, Y., Furuya, K., Goshima, G., Nagao, K., Takahashi, K. and Yanagida, M.

(2002). Requirement of chromatid cohesion proteins Rad21/Scc1 and Mis4/Scc2 for normal spindle-kinetochore interaction in fission yeast. Curr. Biol. 12, 347-358. Umesono, K., Toda, T., Hayashi, S. and Yanagida, M. (1983). Two cell division cycle genes NDA2 and NDA3 of the fission yeast Schizosaccharomycespombe control microtubular organization and sensitivity of antimitotic benzimidazole compounds. J. Mol. Biol. 168, 271-284.

van der Vaart, B., Manatschal, C., Grigoriev, I., Olieric, V., Gouveia, S. M., Bjelic, S., Demmers, J., Vorobjev, I., Hoogenraad, C. C., Steinmetz, M. O. and Akhmanova, A.

(2011). SLAIN2 links microtubule plus end-tracking proteins and controls microtubule growth in interphase. J. Cell Biol. 193, 1083-1099.

Vitre, B., Coquelle, F. M., Heichette, C., Garnier, C., Chretien, D. and Arnal, I. (2008). EB1 regulates microtubule dynamics and tubulin sheet closure in vitro. Nat. Cell Biol. 10, 415-421.

Weisbrich, A., Honnappa, S., Jaussi, R., Okhrimenko, O., Frey, D., Jelesarov, I., Akhmanova, A. and Steinmetz, M. O. (2007). Structure-function relationship of CAP-Gly domains. Nat. Struct. Mol. Biol. 14, 959-967.

Widlund, P. O., Stear, J. H., Pozniakovsky, A., Zanic, M., Reber, S., Brouhard, G. J., Hyman, A. A. and Howard, J. (2011). XMAP215 polymerase activity is built by combining multiple tubulin-binding TOG domains and a basic lattice-binding region. Proc. Natl. Acad. Sci. USA 108, 2741-2746.

Wolyniak, M. J., Blake-Hodek, K., Kosco, K., Hwang, E., You, L. and Huffaker, T. C.

(2006). The regulation of microtubule dynamics in Saccharomyces cerevisiae by three

interacting plus-end tracking proteins. Mol. Biol. Cell 17, 2789-2798.

Zanic, M., Stear, J. H., Hyman, A. A. and Howard, J. (2009). EB1 recognizes the

nucleotide state of tubulin in the microtubule lattice. PLoS ONE. 4, e7585.

Zanic, M., Widlund, P. O., Hyman, A. A. and Howard, J. (2013). Synergy between

XMAP215 and EB1 increases microtubule growth rates to physiological levels. Nat. Cell

Biol. 15, 688-693.

Zhang, R., Alushin, G. M., Brown, A. and Nogales, E. (2015). Mechanistic origin of microtubule dynamic instability and Its modulation by EB proteins. Cell 162, 849-859.

"rö c

Figures

Fig. 1. Fission yeast Dis1 is a microtubule polymerase.

(A) Microtubule binding activity of Dis1-eGFP. Dis1-eGFP concentrations were kept constant at 0.2 |iM, while the tubulin concentration was varied (from left to right, 0, 1, 2, 4 |iM). Both supernatant (S) and pellet (P) were stained with Coomassie Brilliant Blue.

(B) Tubulin binding activity of Dis1-eGFP. 5 |iM Dis1-eGFP (green), 10 |iM tubulin dimers (blue) and their mixture (red) were fractionated by gel filtration column.

(C) Schematic of the assay (top) and dual colour TIRF-M kymographs showing binding events of 10 nM Dis1-eGFP (green in merge) to the ends of growing microtubules.

(D) Dual-colour TIRF-M kymographs showing Cy5-labelled microtubules growing in the absence or presence of Dis1-eGFP (green).

(E) Plot of the mean growth rate. Cy5 labelled tubulin concentration was 8 |iM (C-E). Data points, black: error bars are S.E.M. n=50.

(F) Kymographs showing GMPCPP-stabilised Cy5-labelled microtubules. Scale bars, 5 |im (horizontal) and 1 min (vertical) (C, D and F).

Fig. 2. Disl directly binds to Mal3.

(A) Binding between Disl and Mal3. GST or GST-Mal3 bound to glutathione beads was mixed with Disl-eGFP.

(B) Analytical gel filtration chromatography. 5 |iM Dis1-HA (green), 20 |iM Mal3 (blue) or the mixture of both proteins (red) was analysed.

(C) TIRF-M kymographs showing binding events of 20 nM Dis1-eGFP (green) to the ends of growing Cy5-labelled microtubules (red) in the absence or presence of unlabelled Mal3. Cy5-labelled tubulin concentration was 20 |iM. Snap shots are shown at the bottom.

(D) Schematic of the assay (top) and triple colour TIRF-M kymographs showing binding events of 10 nM Dis1-eGFP (green in merge) to the ends of growing Cy5-labelled microtubules (blue in merge) in the presence of 200 nM Mal3-mCherry (red in merge) (bottom).

(E) 10 nM Dis1-eGFP accumulates at the ends of a microtubule in a similar manner to 200 nM Mal3-mCherry. Cy5 labelled tubulin concentration was 10 |iM (D and E). Scale bars, 5 |im (horizontal) and 1 min (vertical) (C-E).

(F) Plot of the mean growth rate. 10 nM Dis1 and/or 20 nM Mal3 were added in the presence of 8 |iM tubulin. Data points, black: error bars are S.E.M. n=50.

Fig. 3. The C-terminal tail region of Disl is the primary binding site for Mal3.

(A) A schematic representation and a summary of their binding to Mal3. TOG1, TOG2, TOG homology domains; SK rich, rich in serines and lysines; CC, coiled-coil.

(B) Binding between various truncated Disl proteins and Mal3. GST-Mal3 bound to glutathione beads was mixed with truncated Disl-eGFP.

(C) Binding between the full-length or tail-less Disl protein and Mal3.

(D) Peptide array analysis. The arrays cover from 505th to 882nd amino acid residues of Disl. The arrays were incubated with a solution containing Mal3-HA protein, followed by immunoblotting against an anti-HA antibody. Amino acid sequences corresponding to bound residues (red) are shown on the bottom.

(E) Binding between the tail region of Disl and Mal3. GST-Mal3 bound to the glutathione beads were pre-incubated with various concentration of the Disl peptide (833-852 a.a.) (from left to right, 0, 0.5, 2.5, 5.0, 50, l00 |M) for 30 min before incubation with 0.5 |M Disl C2-eGFP for an additional hour.

(F) Alanine scanning mutagenesis analysis. Each position in the Disl peptide was substituted with alanine. Amino acid residues shown in red were sensitive to alternations.

(G) Binding between Mal3 and the wild type or mutated Disl protein. Disl-eGFP proteins (LAPA: L84lA and P844A or LAPAFA: L84lA, P844A and F847A).

O it-o

Fig. 4. The coiled-coil and EBH domains of Mal3 are necessary and sufficient for binding to Dis1

(A) A schematic representation of various truncated Mal3 proteins and a summary of their binding to Dis1. CH, calpoin-homology domain; CC, coiled-coil; EBH, EB homology. (B, C) Binding between various truncated Mal3 proteins and Dis1. Dis1 C1 (520-882 a.a.)-eGFP proteins were mixed with various GST tagged truncated Mal3 proteins (B, FL, N, C1 or C, C2, C3), which were bound to the glutathione beads.

(D) Binding between the dimerised EBH domain of Mal3 protein and Dis1. EBH domain-containing Mal3 protein (174-247a.a.) was mixed with wild type or mutated Dis1 C1-eGFP (2A: L841A, P844A) and pulled down by GFP trap.

Fig. 5. Physical binding between Disl and Mal3 is required for synergetic impact on microtubule dynamics

(A) TIRF-M kymographs showing Disl-eGFP (10 nM, green in merge) in the presence or absence of 20 nM Mal3. Cy5-labelled tubulin (red) concentration was 8 pM. Scale bars, 5 pm (horizontal) and 1 min (vertical).

(B, C) Plot of the mean growth rate (B) or catastrophe frequency (C). 10 nM Dis1-eGFP (WT or LAPA) and/or 20 nM Mal3 were added in the presence of 8 |iM tubulin. Data points, black: error bars are S.E.M. n=50.

U it-o

Unconserved

Conserved

MACF-EB1

MACF-EB1

Fig. 6. Molecular details of the Mal3-Dis1 interaction

(A) Ribbon diagram of the complex between the coiled-coil and EBH domains of Mal3 (residues l74-247 forming a homodimer as depicted in blue and pink) and the interacting Disl peptide (residues 833-852 as depicted in green). A second peptide binds on the other side of the dimer, but has been omitted from the figure for the sake of clarity.

(B) Details of the key interacting residues. Surface conservation plot of the EBH domain of Mal3 from low (blue) to high (purple) similarity.

(C, D) Close-up view of the interaction seen in the Mal3-EBH dimer (residues l74-247 as depicted in blue and pink) and Disl peptide (residues 833-852 as depicted in green) (C) or in the human EBl-EBH dimer (residues l9l-260 as depicted in yellow and green) and the SXIP motif containing MACF peptide (residues 5468-5497 as depicted in orange) (D). (E) Cartoon showing the comparison of the interaction interfaces between Mal3-Disl (left) and EBl-MCAF (right). A (A') and B (B') represent each EBH domain derived of two dimerised Mal3 molecules.

Fig. 7. Dis1 interacts with Mal3 in vivo and its interaction is critical for Dis1 function.

(A) Interaction between Dis1 and Mal3. Whole cell extracts (1.0 mg) were pulled down with

GFP trap. Immunoblotting was performed by anti-GFP and anti-FLAG antibodies. Input: 15 ^g.

(B) Summary of genetic interaction. Solid lines, synthetically lethal: dotted line, viable.

(C) Tetrad dissection. Five representative tetrads are shown (parental ditype (1), tetratype (24) and non-parental ditype (5)). Double mutants were inviable (red circles).

(D) Spot assay. Indicated strains were serially diluted (10 fold each), spotted on YES and incubated for 4 d or at 26°C for 4 d in the presence of 10 |ig/ml TBZ (right).

(E) Mitotic progression and MT dynamics in dis1-LAPA cells. Mitotic progression of wild type (left, n=38) and dis1-LAPA cells (right, n=31) containing a tubulin marker (mCherry-Atb2) was followed in live under fluorescence microscopy at the room temperature (23°C). Phase I (red), the initial stage of spindle elongation; Phase II (yellow), the pre-anaphase stage with constant spindle length; Phase III (blue), anaphase B (Nabeshima et al., 1998). All p-values are from two-tailed unpaired student t-tests (>30 cells). n.s., not significant; p-value, *p<0.01.

(F) Minichromosome loss assay. Indicated strains carrying the minichromosome Ch16 (Niwa et al., 1989) were grown on rich YE plates (lacking adenine) and incubated at 25°C for 6 days. The percentage of red and/or sectored colonies is shown at the bottom (n>1, 000).

Fig. 8. Interplay between Disl and Mal3 and comparison to metazoans

(A) In fission yeast, Disl (orange) binds directly to Mal3 (blue), thereby enhancing the MT growth rate and the catastrophe frequency in a synergistic manner (left). By contrast, in metazoans including humans and Drosophila, ch-TOG and Msps (orange) interact with EB1 (blue) via SLAIN2 and Sentin (yellow) respectively (right), thereby accelerating MT dynamics. Dimer formation of EB1/Mal3 is not shown for simplicity.

(B) Summary of individual parameters representing MT dynamics in the presence of Disl (wild type or LAPA) and/or Mal3. -, not altered compared to control samples (tubulin only); thin arrow, increased by up to twofold; intermediate arrow, increased by -threefold; thick arrow, increased by more than fourfold.

Table SI. Kinetic parameters of MT dynamics in vitro in the presence of various concentrations of Disl (0-40 nM) at 8 jiM tubulin

Disl N Growth rate Shrinkage rate Catastrophe frequency

nM (im/min (n) (im/min (n) min"1

0 2 0.54 ±0.02 (50) 15.79 ±0.59 (50) 0.11 ±0.01

5 2 1.14 ±0.08 (50) 14.93 ±0.76 (50) 0.16 ±0.01

10 2 1.22 ±0.05 (50) 12.02 ±0.78 (50) 0.13 ±0.01

20 2 1.58 ±0.09 (50) 11.28 ±0.57 (50) 0.17±0.01

40 2 1.18 ±0.07 (50) 12.19 ±0.68 (50) 0.13 ±0.01

'N' represents the number of experiments, while 'n' represents the number of events calculated for growth rate and shrinkage rate. As for catastrophe frequency, at least 30 microtubules were observed per each experiment. Each value is shown in mean±S.E.M. Rescue events were not observed under this condition.

Table S2. Kinetic parameters of MT dynamics in vitro in the presence or absence of 10 nM Disl and/or 20 nM Mal3 at 8 ¿iM tubulin

nM N Growth rate

(im/min (n) None 2 0.54 ± 0.02 (50)

Shrinkage rate

(im/min (n) 15.79 ±0.59 (50)

Catastrophe frequency

0.11 ±0.01

Disl 2 1.22 ±0.05 (50)

12.02 ±0.78 (50)

0.13 ±0.01

Mal3 2 0.87 ± 0.04 (50) Disl

+ 2 Mal3

2.48 ±0.08 (50)

20.09 ±0.93 (50) 10.40 ±0.94 (50)

0.22 ±0.01

0.42 ±0.02

'N' represents the number of experiments, while 'n' represents the number of events calculated for growth rate and shrinkage rate. As for catastrophe frequency, at least 30 microtubules were observed per each experiment. Each value is shown in mean±S.E.M. Rescue events were not observed under this condition.

Table S3. Kinetic parameters of MT dynamics in vitro in the presence or absence of 10 nM Disl (WT or Disl-LAPA) and/or 20 nM Mal3 at 8 ¿iM tubulin

N Growth rate

Shrinkage rate

Catastrophe frequency

(im/min (n) 0.54 ±0.02 (50)

(im/min (n) 9.16 ±0.30 (50)

0.11 ±0.01

Disl WT

Disl LAPA

Disl WT +

Disl LAPA +

1.09 ±0.06 (50) 1.02 ±0.01 (50) 0.82 ±0.02 (50) 2.35 ±0.12 (50)

1.08 ±0.09 (50)

8.97 ±0.41 (50) 8.85 ±0.53 (50) 8.87 ±0.68 (50) 7.76 ±0.38 (50)

8.39 ±0.63 (50)

0.16±0.01

0.14 ± 0.01

0.23 ±0.01

0.43 ±0.02

0.22 ±0.02

'N' represents the number of experiments, while 'n' represents the number of events calculated for growth rate and shrinkage rate. As for catastrophe frequency, at least 30 microtubules were observed per each experiment. Each value is shown in mean±S.E.M. Rescue events were not observed under this condition.

Table S4. Crystallographic data, phasing, and refinement

Mal3174"247 Mal3174"247 +Dis1833 852

Wavelength (A) 0.97949 0.97625

Resolution range (A) 35.52 - 1.33 (1.378 - 1.33) 80.8 -2.83(2.931 -2.83)

Space group P 2, 21 21 P 65 2 2

Unit cell (A,°) 35.38 37.91 101.64 90 90 90 93.305 93.305 196.545

90 120

Total reflections 401356 (33134) 232502 (23332)

Unique reflections 32149 (3130) 12733 (1186)

Multiplicity 12.5 (10.6) 18.3 (19.0)

Completeness (%) 1.00 (0.99) 0.97 (1.00)

Mean l/sigma(l) 16.19 (1.15) 9.13 (1.30)

Wilson B-factor 14.18 76.71

R-merge 0.09253 (1.963) 0.1766 (2.937)

R-measured 0.09652 (2.063) 0.1819 (3.019)

CC1/2 1 (0.571) 0.997 (0.58)

CC* 1 (0.852) 0.999 (0.857)

Reflections used in refinement 32148 (3128) 12370 (1186)

Reflections used for R-free 1570 (168) 603 (58)

R-work 0.1509 (0.3349) 0.3024 (0.4630)

R-free 0.2037 (0.3444) 0.3399 (0.5033)

CC(work) 0.958 (0.606) 0.936 (0.552)

CC(free) 0.908 (0.658) 0.911 (0.391)

Number of non-hydrogen atoms 1320 2704

macromolecules 1223 2704

Protein residues 138 330

RMS(bonds) 0.030 0.031

RMS(angles) 2.44 1.99

Ramachandran favored (%) 99 97

Ramachandran allowed (%) 1.4 2.5

Ramachandran outliers (%) 0 0.64

Rotamer outliers (%) 1.4 7.5

Clashscore 7.76 2.22

Average B-factor 22.33 90.11

macromolecules 21.09 90.11

solvent 37.97 None

Table S5. S. pombe strain list used in this study

Strain Genotype Figure used

513 h leul-32 ura4-D18 7A

D3 h+ leul-32 his2 ura4-D18 7D

YMP31 h+ leul-32 his2 ura4-D18 mal3+-5FL4G-hphAlX6 7A

YMP34 h leul-32 ura4-D18 disl+-2GFP-kcmKIX6 7A

YMP35 h leul-32 ura4-D18 disl-L4PA-2GFP-kcmALX6 7A-C

YMP53 h leul-32 ura4-D18 disl+-2GFP-kcmKIX6 mal3-5FLAG-hphMX6 7A

YMP54 h leul-32 ura4-D18 disl-L4PA-2GFP-kcmAIX6 mal3-5FL4G-hphAlX6 7A

CP021 h leul-32 ura4-D18 disl::hphMX6 7D

KSH475 h+ leul-32 his2 ura4-D18 ctlpl4::urct4+ 7C

FM023 H leul-32 ura4-D18 mal3::kcmAIX6 7D

YMP46 h+ leul-32 his2 ura4-D18 mal3::urct4+ 7D

YMP38 h leul-32 ura4-D18 disl*-3GFP-karAlX6 aurlR-Pnda3-mCherry-atb2+ 7E

YMP44 h leul-32 ura4-D18 disl-L4PA-2GFP-kanA fX6 aurlR-Pnda3-mCherry-atb2+ 7E

TPR20-1 h+ leul-32 ade6-M210 Chl6 7F

YMP56 h leul-32 ade6-M210 Chl6 disl::hphMX6 7F

YMP57 h leul-32 ade6-M210 Chi 6 disl-LAPA-2GFP-kar&tX6 7F

100-75_

150100-

- Dis1 -HA

kOa 250-

-Disl-WT-eGFP

100755037-

-Dis1 -LAPA-eGFP

5 10 20 Dis1 concentration [nM]

kDa 5037150-

100150-

0.20 T

0.15 -

0.05 -

Elution volume (ml)

10 _l_

-Tubulin

Disl-eGFP

■ Disl-eGFP

-Tubulin

■ ■

5 10 20 40 Disl concentration [nM]

3. | 0.10

0 5 40

Dis1 concentration [nM] Fig. SI. Fission yeast Disl is a microtubule polymerase.

(A) Coomassie Brilliant Blue-stained SDS-PAGE gels of the purified recombinant Dis1 proteins as indicated.

(B) Tubulin binding activity of Disl-eGFP was measured by analytical gel filtration chromatography. 5 ^M purified full-length Disl-eGFP (green), 10 |iM unpolymerised tubulin dimers (blue) and the mixture of 5 ^M Disl-eGFP and 10 ^M tubulin dimers (red) were fractionated by gel filtration column (as shown Fig. 1B), followed bv Coomassie Brilliant Blue staining of each fraction. This data corresponds to Fig. 1B.

(C) Shrinkage rate and catastrophe frequency in the presence of 8 |iM tubulin and various concentrations of Disl. The mean shrinkage rate and catastrophe frequency were plotted against individual Dis1 concentrations. Data points, black: error bars are S.E.M. n=50.

(D) Plot of the mean shrinkage rate as a function of the Dis1 concentration. Data points, black: error bars are S.E.M. n=50. Corresponding kymograph images are shown in Fig.lF.

Fig. S2. Disl directly binds to Mal3.

(A) Coomassie Brilliant Blue-stained SDS-PAGE gels of the purified recombinant Mal3 proteins as indicated.

(B) Analytical gel filtration chromatography for 5 |iM purified full-length Dis1-HA (green), 20 ^M purified full-length Mal3 (blue) and the mixture of 5^M Dis1-HA and 20 ^M Mal3 (red) were fractionated, followed by Coomassie Brilliant Blue staining of each fraction. This data corresponds to Fig. 2B

(C) Dual-colour TIRF-M kymographs showing Cy5-labelled microtubules (red) growing in the presence of 200 nM or 1 ^M Mal3-eGFP (green). Cy5 labelled tubulin concentration was 10 ^M. Scale bar, 5 ^m

(D) Binding between Alp14 and Mal3 was tested by GST pull-down assay. Purified GST or GST-Alp14 bound to the glutathione beads was mixed with purified Mal3. Each fraction was stained with Coomassie Brilliant Blue.

(E) Kymographs showing MT polymerisation with combinations of Mal3 (20 nM) and/or Disl (10 nM) in the presence of 8 ^M tubulin. Scale bars, 5 |im (horizontal) and 1 min (vertical) (C and E).

(F) Parameters representing MT dynamics in the presence of 8 |iM tubulin and various concentrations of Dis1 (10 nM) and/or Mal3 (20 nM). The mean shrinkage rate and catastrophe frequency were plotted. Data points, black: error bars are S.E.M. n=50.

No Ма13

Werg« Dts1-«GFP CyS-tutXiin

Merge Osl-eGFP CyStubuln

Dis1 FL

Dîs1 N

Dis1 C1

Dis1 C2

Dis1 C3

Fig. S3. The C-terminal tail region of Disl binds to Mal3.

(A) A schematic representation of various truncated Disl proteins and a summary of MT plus end tracking and binding

to the MT lattice and Mal3.

(B) TIRF-M kymographs showing binding events of several truncated Dis1-eGFP constructs (20 nM, green in merge) to

the plus ends of growing Cy5-labelled MTs (red in merge) in the absence (left) or presence (right) of 200 nM Mal3. Cy5

labelled tubulin concentration was 10 ^M. Scale bars, 5 ^m (horizontal) and 1 min (vertical).

E 0 a d

0 u c 0 'u CO

Fig. S4. Details of the experimental electron density for the Dis1 ligand and the interaction interface between Mal3-Dis1 and EB1-MACF..

(A) Ribbon diagram of the Mal3 EB1 domain dimer (residues 174-247) with an unbiased omit (Fo-Fc) map for the Disl peptide (red) contoured at 3a.

(B) Ribbon diagram of the Mal3 EB1 domain dimer (residues 174-247) with the final refined (2Fo-Fc) map for the Disl peptide (blue) contoured at 1.5a.

(C) 3D-image showing the interaction interface between the Mal3-EBH dimers and the Disl peptide. Three amino acid residues of the Dis1 peptide (S839, L841 and P844) are shown. Individual EBH domains are distinguished by blue and pink colours. Snap shot image is shown in Fig. 5C.

(D) 3D-image showing the interaction interface between the EB1-EBH dimers and the MACF peptide. Three amino acid residues of the MACF peptide (S5477, I5479 and P5480) are shown. Individual EBH domains are distinguished by yellow and green colours. Snap shot image is shown in Fig. 5D.

► • # Q 197533-filel.mov

00:00 |

« ► ►►

-01:40

Movie 1: Movie showing the MT plus end tracking of Dis1-eGFP (10 nM, green) in the presence of Cy5-tubulin (8 ^M, red).

Corresponding Kymographs are shown in Fig 1C. The time stamp (upper left corner) is in seconds. Scale bar, 5pm.

• • • B 197533-file2.mov

44 ► ►►

00:00 | -00:50 |

Movie 2: Movie showing the MT plus end tracking of Disl-eGFP (20 nM, green) in the presence of Cy5-tubulin (20 ^M, red).

Corresponding Kymograph is shown in Fig 2C. The time stamp (upper left corner) is in seconds. Scale bar, 5^m.

• • • B 197533-file3.mov

44 ► ►►

00:00 | -00:50

Movie 3: Movie showing the MT plus end tracking of Dis1-eGFP (20 nM, green) in the presence of 200 nM Mal3 and Cy5-tubulin (20 ^M, red). Corresponding Kymograph is shown in Fig 2C. The time stamp (upper left corner) is in seconds. Scale bar, 5 pm.

► • •

Q 197533-file4.mov

Movie 4: Movie showing the MT plus end tracking of Dis1-eGFP (20 nM, green) in the presence of 1 ^M Mal3 and Cy5-tubulin (20 ^M, red). Corresponding Kymograph is shown in Fig 2C. The time stamp (upper left corner) is in seconds. Scale bar, 5^m.

• • • & 197533-file5.mov

00:00 |

44 ► ►►

-00:50

Movie 5: Movie showing colocalisation between Disl-eGFP (20 nM, green) and Mal3-mCherry (200 nM, red) in the presence of Cy5-tubulin (10 ^M, blue). Corresponding Kymographs are shown in Fig 2D. The time stamp (upper left corner) is in seconds. Scale bar, 5 pm.