MZT1 Regulates Microtubule Nucleation by Linking Γturc Assembly to Adapter-Mediated Targeting and Activation

RESEARCH ARTICLE MZT1 regulates microtubule nucleation by linking γTuRC assembly to adapter-mediated targeting and activation Rosa Ramıreź Cota1, Neus Teixidó-Travesa1, Artur Ezquerra1, Susana Eibes1, Cristina Lacasa1, Joan Roig1,2 and Jens Lüders1,*

ABSTRACT and GCP3, which form the smaller subcomplex γTuSC, are Regulation of the γ-tubulin ring complex (γTuRC) through targeting essential in flies and fungi, and are sufficient for microtubule and activation restricts nucleation of microtubules to microtubule- nucleation and targeting to MTOCs (Anders et al., 2006; Fujita organizing centers (MTOCs), aiding in the assembly of ordered et al., 2002; Lin et al., 2015; Venkatram et al., 2004; Vérollet et al., microtubule arrays. However, the mechanistic basis of this important 2006; Xiong and Oakley, 2009). These observations suggest that regulation remains poorly understood. Here, we show that, in human there are important organism-specific differences in the manner by cells, γTuRC integrity, determined by the presence of γ-tubulin which nucleation template assembly is linked to targeting and complex proteins (GCPs; also known as TUBGCPs) 2–6, is a activation. prerequisite for interaction with the targeting factor NEDD1, impacting In budding yeast, which lacks GCP4, GCP5 and GCP6, γ on essentially all γ-tubulin-dependent functions. Recognition of formation of higher-order -tubulin complexes seems to be γTuRC integrity is mediated by MZT1, which binds not only to the restricted to the spindle pole body (the yeast centrosome GCP3 subunit as previously shown, but cooperatively also to other equivalent). Here, interactions with adapter proteins containing γ GCPs through a conserved hydrophobic motif present in the N-termini CM1 and SPM motifs might trigger formation of TuSC oligomers γ of GCP2, GCP3, GCP5 and GCP6. MZT1 knockdown causes severe that are similar in structure and activity to TuRC (Kollman et al., cellular defects under conditions that leave γTuRC intact, suggesting 2010; Lin et al., 2015, 2014). The CM1 and SPM motifs of the that the essential function of MZT1 is not in γTuRC assembly. Instead, budding yeast adapter Spc110 have been shown to mediate γ MZT1 specifically binds fully assembled γTuRC to enable interaction interaction with TuSC, through direct interaction with the N- with NEDD1 for targeting, and with the CM1 domain of CDK5RAP2 terminus of GCP3 and possibly also GCP2 (Knop and Schiebel, for stimulating nucleation activity. Thus, MZT1 is a ‘priming factor’ for 1997; Lin et al., 2011; Nguyen et al., 1998). Phosphorylation of the γTuRC that allows spatial regulation of nucleation. adapter in the region between the two motifs further enhances the interaction. As a result, γTuSC oligomerizes and becomes a more KEY WORDS: Microtubule, Nucleation, γ-tubulin, Centrosome active nucleation template (Lin et al., 2014). Even in fission yeast, which contains GCP4, GCP5 and GCP6, formation of active higher- INTRODUCTION order γ-tubulin complexes requires interaction with the CM1- The γ-tubulin ring complex (γTuRC), composed of γ-tubulin and containing protein Mto1 that forms an adapter complex with Mto2 γ-tubulin complex proteins (GCPs, also known as TUBGCPs), at interphase MTOCs (Lynch et al., 2014). Adapter-mediated nucleates microtubules by providing a template for the assembly of γTuSC oligomerization at MTOCs might also explain how the α-tubulin–β-tubulin heterodimers (Kollman et al., 2011; Oakley experimental loss of the γTuRC-specific subunits GCP4, GCP5 and et al., 2015; Petry and Vale, 2015; Remy et al., 2013; Teixidó- GCP6 is compensated for in flies (Vérollet et al., 2006). Travesa et al., 2012). The GCPs that form the γTuRC core structure In contrast to fungi, in most animal cells γTuRC assembly, all have homologous regions known as Grip motifs (Gunawardane targeting and activation are not tightly coupled, and pre-assembled et al., 2000; Kollman et al., 2011) and in human cells are named but relatively inactive γTuRCs can be found in the cytosol, outside GCP2, GCP3, GCP4, GCP5 and GCP6 (Lüders and Stearns, 2007; of MTOCs. However, in human cells γTuRC pre-assembly, Murphy et al., 2001; Teixidó-Travesa et al., 2012). As a nucleator, targeting and activation are linked in the sense that all of these γTuRC is not only important for generating microtubules but also steps seem to involve the γTuRC-specific GCP4, GCP5 and GCP6 for controlling their position and orientation, which is fundamental (Bahtz et al., 2012; Choi et al., 2010; Izumi et al., 2008; Scheidecker to the formation of ordered microtubule arrays (Lüders and et al., 2015), but the underlying mechanism has not been revealed. Stearns, 2007). Thus, the activity of γTuRC is tightly controlled In fact, a systematic comparative analysis of the roles of all human by targeting and activation factors that spatially restrict nucleation to GCPs in γTuRC assembly and function has never been conducted. microtubule-organizing centers (MTOCs) such as the centrosome. Targeting of human γTuRC can be mediated by various different Surprisingly, whereas assembly and targeting of human γTuRC proteins. Pericentrin, AKAP9, CDK5RAP2 and myomegalin have seem to involve all GCPs (Bahtz et al., 2012; Choi et al., 2010; been implicated in γTuRC recruitment to the centrosome (Fong Izumi et al., 2008; Scheidecker et al., 2015), only γ-tubulin, GCP2 et al., 2008; Takahashi et al., 2002; Zimmerman et al., 2004) and to the Golgi, which functions as a non-centrosomal MTOC (Rivero et al., 2009; Roubin et al., 2013; Wang et al., 2010). All of these 1Institute for Research in Biomedicine (IRB Barcelona), Barcelona 08028, Spain. proteins contain a CM1 motif (Lin et al., 2014, 2015; Samejima 2 Molecular Biology Institute of Barcelona (IBMB-CSIC), Barcelona 08028, Spain. et al., 2008; Sawin et al., 2004), whereas an SPM motif is found *Author for correspondence ([email protected]) only in pericentrin and AKAP9 (Lin et al., 2014, 2015). Curiously, NEDD1 (also known as GCP-WD), which in human cells is the

Received 15 July 2016; Accepted 9 November 2016 most crucial factor for centrosomal targeting of γTuRC (Haren et al., Journal of Cell Science

406 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 406-419 doi:10.1242/jcs.195321

2006; Lüders et al., 2006), does not seem to contain any SPM or gradient fractionation, and its functionality, by analysis of cellular CM1 motifs. Regardless of the presence or absence of SPM or CM1 phenotypes using immunofluorescence microscopy. This allowed motifs, both CDK5RAP2 and NEDD1 seem to specifically us, in each case, to directly correlate γTuRC integrity with function. associate with γTuRC but not subcomplexes such as γTuSC (Choi Sucrose gradient fractionation of the extracts shown in Fig. 1A et al., 2010; Lüders et al., 2006), but how this specific recognition of and probing with antibodies against γTuRC subunits revealed two γTuRC is achieved is unclear. distinct peaks corresponding to γTuRC and to complexes of smaller Apart from mediating the interaction with γ-tubulin complexes, molecular mass for control extract (Fig. 1B,C; Fig. S2A–D). Given the CM1 motif seems to have a role in activating the nucleation that we had observed in previous work that the amplitude of the activity of γ-tubulin complexes. In yeast, activation might result γTuRC peak was variable between experiments but consistent from γTuSC oligomerization (Kollman et al., 2010; Lin et al., 2014; between samples that were processed in parallel in individual Lynch et al., 2014). In human cells, however, oligomerization per se experiments (Lüders et al., 2006; Teixidó-Travesa et al., 2010), we might not be sufficient for full nucleation activity: even though always analyzed knockdown samples side-by-side with control oligomeric complexes exist in the form of pre-assembled γTuRCs, extract. In GCP3-depleted extract, γ-tubulin in both peaks was the 50-amino-acid CM1 motif of human CDK5RAP2 alone is able shifted to fractions of very small molecular mass (<7S), consistent to strongly stimulate nucleation activity of these complexes in vitro with GCP3 being required for the assembly of both γTuRC and and in vivo (Choi et al., 2010). Indeed, it has been proposed that in smaller complexes (Fig. 1B,C). Depletion of any of the three addition to oligomerization the GCP3 subunit has to undergo a γTuRC-specific subunits (GCP4, GCP5 or GCP6) also reduced the conformational change to allow formation of an optimal nucleation amount of γTuRC present in the extract (Fig. 1B,C; Fig. S2A–D). template (Kollman et al., 2015). The strongest γTuRC disruption was observed with extracts Adding to the complexity of the regulatory network controlling depleted of GCP6, whereas depletion of GCP4 or GCP5 led to nucleation, the small ∼8-kDa protein MZT1, which is found in only partial γTuRC disruption, with a substantial amount of GCP6 animals, plants and most fungi, and is not related to any of the other still co-fractionating in a high-molecular-mass complex that seemed known adapters, is also required for targeting γTuRC to nucleation only slightly smaller than γTuRC (Fig. S2A–D). Knockdown sites (Dhani et al., 2013; Hutchins et al., 2010; Janski et al., 2012; cultures derived from the same transfections were also used for Masuda et al., 2013; Nakamura et al., 2012). However, whether analysis by immunofluorescence microscopy. Interestingly, MZT1 directly attaches γ-tubulin complexes to MTOCs and depletion of GCP6 caused a similar increase in the mitotic index whether this role is related to the function of other well- to that seen upon depletion of GCP3 (∼45% and ∼40%, established targeting factors such as NEDD1 or CM1-containing respectively). Depletion of GCP4 or GCP5 also increased the adapters, is unknown. It has also not been tested whether MZT1 has mitotic index but to a lesser extent (∼15% and 10%, respectively) any role in γTuRC or γTuSC oligomer assembly. Given that MZT1 (Fig. 1D). GCP6-depleted cells displayed mitotic defects that were interacts with GCP3 (Dhani et al., 2013; Janski et al., 2012; at least as severe as in cells depleted of GCP3. Approximately 80% Nakamura et al., 2012), the subunit that has been suggested to of GCP6-depleted mitotic cells had monopolar-like spindle undergo a conformational switch during the activation of γ-tubulin configurations with poorly organized microtubules, and few cells complexes (Kollman et al., 2015), and based on circumstantial at metaphase or in later mitotic stages were observed. Depletion of evidence, MZT1 has been proposed to stimulate nucleation activity GCP4 or GCP5 produced similar, but milder, mitotic spindle (Masuda and Toda, 2016; Nakamura et al., 2012). The relationship defects (Fig. 1E). with CM1-dependent activation, if any, remained unclear. We also analyzed the samples for possible centriole duplication In summary, whereas in some organisms a minimal machinery defects by counting centrin foci in mitotic cells. Again GCP6 involving adapters that interact with γTuSC might be sufficient to depletion produced defects that were comparable to those observed allow assembly, targeting and activation of higher-order γ-tubulin in GCP3-depleted cells. In both cases, depleted cells (∼60%) had complexes, in other systems, including in human cells, this process not duplicated their centrioles and contained a single centriole at is subject to more complex regulation and remains poorly each of their spindle poles (configuration ‘1+1’) compared to understood. Here, we have addressed this issue by systematic doublets in control cells (configuration ‘2+2’) (Fig. 1F). Depletion probing of the subunit requirements, in particular the role of MZT1, of GCP4 or GCP5 also impaired centriole duplication, but the in the spatial regulation of γTuRC in human cells. defects were milder than in GCP6-depleted cells. In 10–20% of the cells one of the centrioles had failed duplication (configuration RESULTS ‘2+1’) (Fig. 1F). Centriole duplication was recently shown to Impairment of γ-tubulin-dependent functions scales with require phosphorylation of GCP6 by Plk4 (Bahtz et al., 2012). Our compromised γTuRC integrity experiments additionally suggest that centriole duplication requires To analyze and compare the roles of the different GCPs in γTuRC an intact γTuRC. In summary, our data shows that impairment of assembly and function, we used RNA interference (RNAi)- γ-tubulin dependent functions is correlated with the degree of mediated depletion. Given that some depletion conditions, γTuRC disruption in human cells. including knockdown of GCP2, resulted in strong co-depletion of other γTuRC subunits (Fig. S1A), we first established RNAi Centrosomal targeting of γ-tubulin is directly correlated with conditions that allowed specific and efficient (>95%) depletion of γTuRC integrity only the target protein, as judged by western blotting of cell extracts To better understand how γTuRC disruption impairs spatial (Fig. 1A). With the exception of GCP3 knockdown, which always regulation of nucleation, we determined γ-tubulin recruitment to led to a slight co-depletion of GCP6, transfection of the selected centrosomes in control and siRNA-treated mitotic cells. Depletion small interfering RNAs (siRNAs) for 72 h specifically reduced the of GCPs reduced centrosomal γ-tubulin staining in all cases, levels of the targeted GCP with minimal effects on other GCP whereas pericentrin was not affected, indicating specific loss of subunits (Fig. 1A). Using these conditions, we then determined, for γ-tubulin (Fig. 2A). Simultaneous disruption of γTuSC and each transfection, the state of γTuRC, by performing sucrose γTuRC (upon depletion of GCP3) or specific disruption of only Journal of Cell Science

407 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 406-419 doi:10.1242/jcs.195321

A B siRNA: (kDa) TuRC 250 GCP6 top 7S 19S bottom GCP5 -Tubulin 100

100 GCP3 GCP2 100 GCP2 GCP3 75 GCP4 GCP4

50 -Tubulin Control RNA GCP5 37 GAPDH * GCP6 C 50 -Tubulin -Tubulin 40 Control RNA GCP2 30 GCP3 siRNA GCP3 20 GCP4 10

GCP3 siRNA GCP3 siRNA GCP5 Band intensity [%] 0 GCP6 1 2 3 4 5 6 7 8 9 10 11 12 13 * fractions 50 -Tubulin -Tubulin 40 Control RNA GCP2 30 GCP6 siRNA GCP3 20 GCP4 10 GCP6 siRNA GCP6 siRNA Band intensity [%] GCP5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 * GCP6 fractions D E 50 **** **** 100 40 Control RNA **** **** GCP3 siRNA 30 80 GCP4 siRNA 20 **** 60 **** GCP5 siRNA

Mitotic cells [%] 10 *** GCP6 siRNA 40 ns 0 ns ns * ns Mitotic cells [%] siRNA: ns nsns Control GCP3 GCP4 GCP5 GCP6 20 ns ns ** F ns ns 100 *** *** *** *** 0 ** ** ** ** **** Control RNA 80 *** GCP3 siRNA phase taphase **** **** Prophase elo 60 GCP4 siRNA Me Anaphase T GCP5 siRNA Prometaphase 40 Monopolar Spindle **** GCP6 siRNA

Mitotic cells [%] **** **** **** 20 **** **** ns ns ns ns ** nsnsns ns 0 ** 2+2 2+1 1+1 1+0 >4 Centriole configuration

Fig. 1. The degree of γTuRC disruption correlates with the severity of cellular defects. (A) Western blot of inputs for the extracts subjected to sucrose gradient centrifugation that are shown in B and Fig. S2A,C, demonstrating specific depletion of individual GCPs. HeLa cells were transfected with siRNA and γTuRC proteins were detected with specific antibodies as indicated. GAPDH was detected as loading control. (B) Extracts from control-, GCP3- or GCP6-depleted HeLa cells shown in A were analyzed by sucrose gradient centrifugation. Fractions were probed with antibodies against the indicated proteins. Asterisks mark an unspecific band recognized by the anti-GCP6 antibody. The vertical box indicates the γTuRC peak fractions. Aldolase (158K, 7S) and thyroglobulin (669K, 19S) were used as size markers. (C) Fractionation profiles for γ-tubulin for the sucrose gradients in B. The γ-tubulin band intensities in western blots of gradient fractions were quantified and plotted for control and depleted extracts as indicated. Values represent percentages of the total signal for all fractions combined. (D) The percentage of mitotic cells was determined for HeLa cells depleted as in A. After staining cells with antibodies against pericentrin, γ-tubulin or α-tubulin, and DAPI, mitotic cells were counted by immunofluorescence microscopy; n=3 experiments (n=5 experiments for control cells). (E) Mitotic cells in D were scored according to the mitotic stage and the presence of a monopolar spindle configuration as indicated; n=3 experiments (n=5 experiments for control cells), 100 mitotic cells per condition. (F) The centriole configuration in mitotic HeLa cells transfected as in A was scored after staining with anti-centrin antibodies. For each of the two centrosomes, the number of centrin foci was counted; n=3 experiments (n=5 experiments for control cells), 100 mitotic cells per condition. All quantitative results are mean±s.e.m. ns, not significant; **P<0.01; ***P<0.001; ****P<0.0001 (unpaired t-test). Journal of Cell Science

408 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 406-419 doi:10.1242/jcs.195321

A B

C D siRNA: (kDa) ** ControlGCP2GCP3GCP4GCP5GCP6 Control GCP2 GCP3 GCP4 GCP5 GCP6 50 **** γ-Tubulin 100 **** GCP2 100 GCP3 **** **** **** 75 **** GCP4 **** 150 GCP5 250 GCP6 Input GST- C-NEDD1

GST Pull-down

Fig. 2. Disruption of γTuRC interferes with γ-tubulin centrosomal targeting by impairing interaction with NEDD1. (A) HeLa cells depleted as in Fig. 1A were imaged by immunofluorescence microscopy after staining with anti-γ-tubulin and anti-pericentrin antibodies, and DAPI to label DNA. Scale bar: 10 µm. (B) Microtubule depolymerization–repolymerization assay with cells shown in A. After 1 min of regrowth, cells were fixed and stained with anti-γ-tubulin and anti-α-tubulin antibodies. (C) The intensities for centrosomal γ-tubulin and microtubule asters surrounding the centrosome (α-tubulin) in cells as in B were quantified and plotted. Values were plotted as percentages of the values obtained for control cells (set to 100%); n>95 centrosomes per condition combined from two independent experiments. Results are mean±s.e.m. **P<0.01, ****P<0.0001 (unpaired t-test compared with control). (D) HeLa extracts depleted for individual GCPs were incubated with recombinant GST or GST–C-NEDD1 immobilized on glutathione–Sepharose. Beads were washed and the retained γTuRC subunits were detected by western blotting. Pulled down GST and GST–C-NEDD1 were detected on the blotted membranes by Ponceau staining.

γTuRC (upon depletion of GCP6) reduced the amount of centriole duplication defects, impaired centrosomal targeting of centrosomal γ-tubulin by ∼55% compared to control cells, γ-tubulin scales with the degree of γTuRC disruption. whereas depletion of GCP4 or GCP5 caused milder effects (∼30% and ∼15% reduction, respectively) (Fig. 2A–C). The γTuRC integrity is crucial for binding to the targeting subunit reduction in centrosomal γ-tubulin correlated with a proportional NEDD1 reduction in centrosomal nucleation activity in a microtubule The phenotypes that we observed in cells with disrupted γTuRC depolymerization–repolymerization assay (Fig. 2B,C). We note have striking similarity with the phenotypes previously reported for that the measured reduction in centrosomal γ-tubulin is likely an knockdown of the targeting factor NEDD1 (Haren et al., 2006; underestimation, as many severely affected mitotic cells, in Lüders et al., 2006). To investigate this, we determined the protein particular after depletion of GCP3 or GCP6, detached from the levels of NEDD1 in depleted extracts. Surprisingly, with the dish or displayed fragmented centrosomes and thus were not used exception of GCP5 depletion, all siRNA treatments caused a strong for quantification. We conclude that, similar to the mitotic and co-depletion of NEDD1 (82%, 69% and 51% reduction in cells Journal of Cell Science

409 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 406-419 doi:10.1242/jcs.195321 depleted of GCP6, GCP3 and GCP4, respectively) (Fig. S3A), centrosomes (Hutchins et al., 2010). However, it remained unclear suggesting that an intact γTuRC is required for NEDD1 stability and whether MZT1 mediates targeting directly or indirectly, by affecting supporting the notion that NEDD1 is a core subunit of γTuRC γTuRC integrity. To address this we generated an antibody against (Lüders et al., 2006; Teixidó-Travesa et al., 2010). To confirm that human MZT1 that recognized a protein of the expected size by the defects that we have described above for depletion of GCPs were western blotting. The specificity of the band was confirmed by a direct consequence of γTuRC disruption and not a secondary siRNA-mediated depletion of MZT1 (Fig. 3F,G). We then effect caused by NEDD1 co-depletion, we performed rescue confirmed that MZT1-depleted cells arrested in mitosis with experiments by expressing RNAi-resistant recombinant NEDD1– monopolar spindles and displayed a specific reduction in EGFP in siRNA-treated cells. In cells depleted of NEDD1 centrosomal γ-tubulin as described previously (Hutchins et al., exogenously expressed NEDD1–EGFP efficiently rescued loss of 2010) (Fig. S4A–F). Interestingly, loss of centrosomal γ-tubulin centrosomal γ-tubulin and mitotic arrest (Fig. S3B–E). However, was also observed in interphase cells, indicating that MZT1 is not a expression of NEDD1–EGFP did not restore centrosomal γ-tubulin mitosis-specific factor but is required for γTuRC targeting and mitotic progression in GCP3- or GCP6-depleted cells throughout the cell cycle (Fig. 3A,B). In agreement with this, we (Fig. S3C–E). This result demonstrates that the disruption of found that MZT1 depletion also impaired centriole duplication γTuRC and not the loss of NEDD1 is the primary cause of the (Fig. 3E). Compared to γ-tubulin, centrosomal localization of mitotic defects in cells depleted of GCP3 or GCP6. NEDD1 was not sensitive or less sensitive to MZT1 depletion Our finding that centrosomal targeting of human γ-tubulin (Fig. 3C,D; Fig. S4E,F), suggesting that MZT1 does not control depends on γTuRC integrity, and that depletion of GCP6 produces centrosomal targeting of NEDD1 but rather links NEDD1 to phenotypes indistinguishable from those observed after depletion of γTuRC. To test this more directly, we immunoprecipitated NEDD1 GCP3, contrasts with results in other organisms. In Drosophila and from control and MZT1-knockdown cells and probed for co- fungi, only γTuSC subunits were found to be essential (Anders precipitation of γ-tubulin and GCP6. Indeed, in the absence of et al., 2006; Fujita et al., 2002; Venkatram et al., 2004; Vérollet MZT1, interaction of NEDD1 with these γTuRC subunits was et al., 2006; Xiong and Oakley, 2009). A key to this discrepancy strongly diminished (Fig. 3F). Importantly, when we analyzed might be the fact that γ-tubulin targeting depends strongly on MZT1-depleted extracts on sucrose gradients, γ-tubulin and GCPs NEDD1 in human cells, whereas this protein seems to be less were still detected in fractions corresponding to the size of γTuRC, important in flies and is not present in fungi. Given that NEDD1 was despite the reduced levels of MZT1 in these fractions (Fig. 3G,H). also destabilized under conditions that disrupted γTuRC but left Thus, under MZT1 depletion conditions that cause severe cellular γTuSC intact (Fig. S3A), we speculated that only intact γTuRC, but defects, γTuRC is still present, but contains a reduced amount of not γTuSC, interacts with NEDD1. To test this, we studied the MZT1 and is defective in binding to the targeting factor NEDD1. interaction between NEDD1 and γTuRC subunits in control extract and extracts depleted of individual GCPs. To make the assay MZT1 directly binds to the N-termini of multiple GCPs insensitive to the destabilization of endogenous NEDD1, we Given that MZT1 was required for NEDD1 to interact with γTuRC, supplemented the extract with an excess of a recombinant GST we speculated that MZT1 might also be responsible for the fusion of the C-terminal half of NEDD1 (GST–C-NEDD1), which preferential recognition of γTuRC over subcomplexes. However, it is sufficient for γTuRC binding (Haren et al., 2006; Lüders et al., was previously shown that MZT1 itself directly binds to the N- 2006; Manning et al., 2010). After affinity purification on terminus of GCP3, a subunit that is not specific to γTuRC (Dhani glutathione beads, GST-C–NEDD1-associated GCPs and γ- et al., 2013; Janski et al., 2012; Nakamura et al., 2012). We decided tubulin were detected by western blotting with specific antibodies. to re-investigate the interactions of MZT1 with γTuRC subunits. We Interestingly, depletion of any of the GCPs reduced the amounts of generated N-terminal fragments of each of the GCPs, comprising γ-tubulin and of all remaining GCPs that were bound to GST–C- the first Grip domain and any N-terminal extensions (NTEs) NEDD1 (Fig. 2D). The strongest impairment was observed after (Fig. 4A). We then performed co-immunoprecipitation experiments knockdown of GCP2, GCP3 or GCP6. Importantly, in depleted in human cells overexpressing FLAG-tagged N-terminal GCP extracts, none of the remaining GCPs retained the strong binding fragments together with EGFP–MZT1. Surprisingly, in this assay observed in control extract (Fig. 2D). As a result, despite the we observed interaction of MZT1 not only with the N-terminal differences in the total amount of γ-tubulin and GCPs that were fragment of GCP3, as previously described, but also with the bound to GST–C-NEDD1, the relative amounts of the retained corresponding fragments of GCP5 and GCP6 (Fig. 4B). N-terminal subunits were always similar, strongly suggesting that NEDD1 fragments of GCP2 and GCP4 failed to coprecipitate with MZT1. In interacts preferentially with intact γTuRCs composed of the the case of GCP3, the MZT1-binding site was mapped to the NTE complete set of subunits. Individual subunits or subcomplexes, that precedes the conserved first Grip domain (Dhani et al., 2013; such as γTuSC, cannot interact efficiently with NEDD1, explaining Janski et al., 2012). Indeed, deletion of the corresponding regions in why only γTuRC can target γ-tubulin to centrosomes in human the N-terminal fragments of GCP3, GCP5 and GCP6 diminished cells. the interaction with EGFP–MZT1 in the coprecipitation assay in each case (Fig. 4B). To test whether these newly discovered γTuRC interaction with NEDD1 is promoted by MZT1 interactions were direct, we employed the yeast two-hybrid assay. How does NEDD1 selectively associate with γTuRC? Previous This revealed that MZT1 strongly interacted with the N-terminal work suggested that NEDD1 directly binds to γ-tubulin (Manning fragments of GCP3, GCP5 and GCP6, and also with GCP2, et al., 2010), but subcomplexes of γTuRC including γTuSC, despite although somewhat less strongly (Fig. 4C), whereas no interaction the presence of γ-tubulin, do not associate with NEDD1 (Fig. 2D) was observed with the N-terminus of GCP4. Serving as controls, (Lüders et al., 2006). Thus interaction with γ-tubulin cannot explain γ-tubulin and GCP8, two subunits of the γTuRC that are unrelated to the preferential binding of NEDD1 to γTuRC. Interestingly, similar GCPs 2–6 (Teixidó-Travesa et al., 2010), also failed to interact with to NEDD1 the recently discovered γTuRC subunit MZT1 also MZT1 (Fig. 4C). Despite the overall very low sequence preferentially binds to γTuRC and is required for its recruitment to conservation in the NTEs of GCP2, GCP3, GCP5 and GCP6, Journal of Cell Science

410 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 406-419 doi:10.1242/jcs.195321

A B Fig. 3. Interaction of γTuRC with the targeting  factor NEDD1 requires MZT1. (A) Interphase Pericentrin -Tubulin Merge/DNA 120 Control siRNA control and MZT1-depleted cells were stained with 100 MZT1 siRNA anti-γ-tubulin and anti-pericentrin antibodies as indicated. DNA was stained with DAPI. Scale bar: 80 10 µm. (B) The intensities for centrosomal 60 γ Control siRNA -tubulin and pericentrin staining in A were **** quantified and plotted as percentages of the signal 40 in control cells (set to 100%); n>100 centrosomes 20 per condition combined from three independent

Fluorescence intensity [%] experiments. (C) Interphase control and MZT1- 0 Pericentrin -Tubulin

MZT1 siRNA depleted cells were stained with anti-NEDD1 and anti-γ-tubulin antibodies as indicated. DNA was C  D NEDD1 -Tubulin Merge/DNA stained with DAPI. Scale bar: 10 µm. (D) The Control siRNA 120 intensities for centrosomal NEDD1 and γ-tubulin MZT1 siRNA 100 staining in C were quantified and plotted as percentages of the signal in control cells (set to 80 100%); n>100 centrosomes per condition Control siRNA 60 **** combined from three independent experiments. (E) The centriole configurations in control and 40 MZT1-depleted mitotic HeLa cells were quantified 20 as in Fig. 1F; n=3 experiments, 100 mitotic cells per condition. (F) Endogenous NEDD1 was Fluorescence intensity [%] 0 MZT1 siRNA NEDD1 -Tubulin immunoprecipitated from control and MZT1- depleted extracts. Precipitation with unspecific IgG E 90 F served as a control. The precipitates were probed 80 with antibodies against the indicated proteins by western blotting. Probing for GAPDH served as a 70 Control siRNA siRNA ControlMZT1 siRNA siRNAControlMZT1 siRNAControlMZT1 siRNA siRNA loading control. (G) Control and MZT-depleted 60 MZT1 siRNA (kDa) Input IgG IP NEDD1 IP HeLa cell extracts were analyzed by western 50 *** blotting with the indicated antibodies. (H) The 75 NEDD1 control and MZT1-depleted cell extracts shown 40 *** 10 MZT1 30 in G were analyzed by sucrose gradient Mitotic cells [%] *** 50 -Tubulin centrifugation as in Fig. 1B. Asterisks mark an 20 250 GCP6 unspecific band recognized by the anti-GCP6 10 γ 37 GAPDH antibody. The vertical box indicates the TuRC 0 peak fractions. Aldolase (158K, 7S) and 2+2 2+1 2+0 1+1 1+0 1+3 >4 thyroglobulin (669K, 19S) were used as size Centriole configuration markers. All quantitative results are mean±s.e.m. G H  ***P<0.001; ****P<0.0001 (unpaired t-test Input top7S 19S TuRC bottom compared with control).  siRNA -Tubulin (kDa) ntrol siRNA Co MZT1 GCP2 250 GCP6 GCP3 100 GCP5 GCP4 100 GCP3 GCP5 100 GCP2 Control siRNA GCP6 75 * GCP4 MZT1 50 -Tubulin 10 -Tubulin MZT1 GCP2 37 GAPDH GCP3 GCP4 GCP5

MZT1 siRNA * GCP6 MZT1 careful analysis revealed a conserved sequence motif composed similar to the wild-type versions. However, in contrast to the mostly of hydrophobic residues (Fig. 4D). To test whether this wild-type proteins, the 3A mutants did not efficiently coprecipitate motif may be involved in MZT1 binding, we mutated three of the endogenous MZT1 and NEDD1 (Fig. 5A). Consistent with this conserved residues to alanine (‘3A’ mutants). Indeed, N-terminal finding, immunofluorescence analysis showed that GCP 3A mutants GCP fragments carrying these mutations did not bind MZT1 in the were defective in targeting to centrosomes (Fig. 5B,C). We also yeast two-hybrid assay (Fig. 4E). Together the results suggest that determined the functional consequences for γTuRC containing a MZT1 directly binds to the N-termini of multiple GCPs and that a mutant GCP in the absence of the corresponding endogenous conserved motif present in the NTEs is required for this interaction. protein. We expressed FLAG-tagged wild-type GCP3 or the 3A We next asked how mutation of this motif would affect MZT1 mutant in cells depleted of endogenous GCP3 (Fig. 5D). Whereas binding in the context of full-length GCPs. GCPs with mutated loss of centrosomal γ-tubulin and the increased mitotic index after

MZT1 binding motif coprecipitated with other GCPs and γ-tubulin GCP3 depletion were rescued in cells expressing recombinant Journal of Cell Science

411 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 406-419 doi:10.1242/jcs.195321

A NTE Grip domain 1 Grip domain 2 GCP2 GCP3 GCP4 GCP5 GCP6 “N-GCP” “N-GCPNTE”

B C pGAD-MZT1 GFP-MZT1 + negative pGBK- 

E E E E E E E E control -Tubulin

T T T T T T T T

N N N N N N N N

       

3 5 5 2

3 4 5 6 6 3 3 4 5 6 6

2 2 positive

P P P P P P P P P P P P P P

P P P P pGBK-

C C C C C C C C C C C C C C

C C C

C control P P N-GCP2

G G G G G G G G G G G G G G

G G

F F

------

- -

N N N N

N N N N N N N N N N

N N

- - -

------

- -

G G G G G G G G G G G G G G

G G

A A A A

A A A A A A A A A A

A A

A A A pGBK- pGBK-

L L L L L

L L L L L L L L L

L L

F F F F F F F F F F F F F

F F

F F F N-GCP4 N-GCP3 Input FLAG IP (kDa) negative pGBK- 75 control N-GCP5

50 FLAG 37 positive pGBK- control N-GCP6 37 GFP-MZT1 GFP pGBK- pGBK- MZT1 GCP8 D H.s. GCP3 1-249 A.t. GCP3 1-188 D.m. GCP3 1-229 A.n. GCP3 1-265 S.p. GCP3 1-192 H.s. GCP2 1-209 H.s. GCP5 1-269 H.s. GCP6 1-316

H.s. GCP3 92-101 A.t. GCP3 104-113 E D.m. GCP3 110-119 pGAD-MZT1 + A.n. GCP3 139-148 S.p. GCP3 95-104 pGBK-N-GCP3 pGBK-N-GCP5 pGBK-N-GCP6 H.s. GCP2 80-89  WT  WT WT H.s. GCP5 109-118 H.s. GCP6 109-118    mut mut mut MZT1 bdg. motif mutant: A A A

Fig. 4. MZT1 binds to a conserved motif present in the N-termini of GCP2, GCP3, GCP5 and GCP6. (A) Schematic representation of the domain structure of the human GCPs. The positions of the N- and C-terminal Grip domains are indicated by black boxes. The N-terminal extensions (NTEs) that precede the N-terminal Grip domains in GCP2, GCP3, GCP5 and GCP6 are colored in beige. The domains comprised by the constructs used in the interaction assays are indicated by black lines. (B) Co-immunoprecipitation (IP) assay to reveal interactions between MZT1 and the indicated GCP fragments. GFP–MZT1 and FLAG-tagged GCP fragments were coexpressed in HEK293 cells. After FLAG immunoprecipitation samples were analyzed by western blotting with anti-FLAG and anti-GFP antibodies. (C) Yeast two-hybrid interactions between MZT1 and N-terminal fragments of the GCPs. Positive interaction is revealed by growth and blue color. (D) Alignment of the amino acid sequences of the NTEs of human GCP2, GCP3, GCP5, and GCP6 with the NTEs of GCP3 orthologs from various species (A.t., Arabidopsis thaliana; D.m., Drosophila melanogaster; A.n., Aspergillus nidulans; S.p., Schizosaccharomyces pombe). The numbers indicate the amino acid positions. The magnified region shows a conserved hydrophobic motif. Alanine replacement mutations used in MZT1-binding mutants are indicated. The alignment was made with the software Geneious using the MUSCLE algorithm (Edgar, 2004). (E) Yeast two-hybrid assay revealing the loss of interaction between MZT1 and N-terminal GCP fragments (wild type, WT) upon mutation of the MZT1-binding motif (mut). Growth and blue color indicates positive interaction. +, positive control; –, negative control. wild-type GCP3, these defects were still observed in cells expressing to multiple GCPs, and that mutation of the MZT1-binding motif in the GCP3 3A mutant (Fig. 5D–F). Taken together, these results only one of the GCPs was sufficient to reduce MZT1 and NEDD1 suggest that MZT1 specifically recognizes γTuRC through binding binding, and interfere with centrosomal targeting. Journal of Cell Science

412 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 406-419 doi:10.1242/jcs.195321

MZT1 promotes not only γTuRC targeting but also activation importance of GCP6 might be related to its central role in γTuRC Only a relatively small percentage of total γTuRC localizes to the assembly, as suggested by previous work in fungi in which GCP6 centrosome and the large majority is present in the soluble fraction was shown to be required for the incorporation of GCP4 and GCP5 of the cytoplasm (Bauer et al., 2016; Moudjou et al., 1996). Given into complexes with γTuSC (Anders et al., 2006) and targeting of that microtubule nucleation occurs predominantly at the Aspergillus GCP4 and GCP5 to the spindle pole body was found to centrosome, the cytosolic γTuRC is presumably relatively require GCP6, but GCP6 spindle pole body localization was found inactive. Therefore, apart from targeting, spatial regulation of to be independent of GCP4 and GCP5 (Xiong and Oakley, 2009). γTuRC should also involve activation of its nucleation activity. However, whereas in control extracts GCP6 co-fractionates almost Indeed, expression of a 50-amino-acid fragment of CDK5RAP2 exclusively with γTuRC, GCP4 and GCP5 are also present in containing the conserved CM1 motif (CDK5RAP2 51–100) complexes smaller than γTuRC. If GCP4 or GCP5 knockdown strongly stimulates the nucleation activity of cytosolic γTuRC preferentially affect this non-γTuRC-associated pool of the proteins, (Choi et al., 2010). Using the same assay, we asked whether the this might leave some γTuRC intact and reduce the severity of the activation of cytosolic γTuRC by CDK5RAP2 51–100 required knockdown phenotypes compared to cells depleted of GCP6. Thus, MZT1. Cells transfected with plasmid encoding EGFP or EGFP– to unequivocally determine the relative importance of GCP4, GCP5 CDK5RAP2 51–100 and with siRNA to knockdown γTuRC and GCP6 for γTuRC integrity and activity, reconstitution of subunits were subjected to cold to completely depolymerize γTuRC from recombinant proteins might be required. microtubules and then incubated at 37°C for 10 s to allow Why do animal cells contain pre-assembled cytosolic γTuRCs? microtubule nucleation to occur. After fixation, cytoplasmic Limiting the assembly of such higher-order γ-tubulin complexes to microtubules were analyzed by performing immunofluorescence MTOCs might satisfy the needs of relatively simple microtubule microscopy and then quantified. As expected, the basal nucleation networks consisting of only a few microtubules, such as those in activity of γTuRC in control cells expressing EGFP was low, but yeast cells. In more complex animal cells that contain hundreds of could be stimulated by expression of EGFP–CDK5RAP2 51-100, microtubules, pre-assembled γTuRCs might be more suitable for as described previously (Fig. 6A,B) (Choi et al., 2010). Strikingly, responding quickly to sudden changes in the demand for nucleators, this stimulatory effect was eliminated by simultaneous knockdown for example, when cells initiate mitotic spindle assembly. Spatial of MZT1. Complete γTuRC disruption upon depletion of GCP2 had control over nucleation requires γTuRC pre-assembly to be a similar effect (Fig. 6A,B). Interestingly, the basal activity in separated from targeting and activation, to ensure that EGFP-expressing control cells was also reduced by depletion of incompletely assembled γTuRC or its subcomplexes are not MZT1, Again, this was similar to cells in which γTuRC was recruited to MTOCs, and fully assembled γTuRCs become active disrupted by GCP2 knockdown (Fig. 6A,B). As an additional only upon interaction with MTOCs. We found that a fundamental control, we also tested whether MZT1 knockdown affected the element in this regulation is the small subunit MZT1, which, on the levels of endogenous CDK5RAP2, but this was not the case one hand allows specific recognition of intact γTuRC through a (Fig. 6D). Taken together, these results suggested that activation of multi-subunit binding mode, and, on the other hand, enables γTuRC γTuRC nucleation activity required MZT1. to interact with NEDD1 and CDK5RAP2, to allow targeting and Similar to NEDD1, CDK5RAP2 also binds specifically to activation, respectively. Owing to its central role in γTuRC γTuRC rather than to smaller subcomplexes, and the CM1 domain regulation, one can speculate that MZT1 activity (controlled at the of CDK5RAP2 is sufficient for this interaction (Choi et al., 2010). level of available MZT1 protein or by post-translational We speculated that MZT1 might not only promote interaction of modification) will determine the amount of γTuRC that can be γTuRC with the C-terminus of NEDD1, but also with the CM1 targeted and activated to nucleate microtubules. Other γTuRC domain of CDK5RAP2. To test this, we immunoprecipitated regulators could also be subject to this type of regulation. Consistent transiently expressed EGFP–CDK5RAP2 51-100 from control cells with this notion, the regulatory factors NEDD1 and CDK5RAP2 and cells depleted of MZT1. Supporting our hypothesis, EGFP– have a significantly higher turnover rate compared to the γTuRC CDK5RAP2 51-100 efficiently coprecipitated γTuRC subunits structural core subunits γ-tubulin, GCP2, GCP3, GCP4, GCP5 and from control extracts, whereas in MZT1-depleted extract this GCP6 (Jakobsen et al., 2011). Moreover, NEDD1 is phosphorylated interaction was diminished (Fig. 6C). This result strongly suggests at multiple sites to specifically control γTuRC function in distinct that the inability of EGFP–CDK5RAP2 51-100 to stimulate γTuRC mitotic nucleation pathways (Gomez-Ferreria et al., 2012; Haren nucleation activity in the absence of MZT1 (Fig. 6A,B) was due to et al., 2009; Johmura et al., 2011; Lüders et al., 2006; Pinyol et al., impaired γTuRC binding. 2013; Scrofani et al., 2015; Sdelci et al., 2012; Teixidó-Travesa et al., 2010, 2012). Future work will show whether human MZT1 is DISCUSSION subject to similar regulation. Here, we have elucidated the spatial regulation of microtubule Based on genetic evidence, it has been recently shown that GCP6 nucleation in human cells. Side-by-side analysis of GCP4-, GCP5- and MZT1 in fission yeast act synergistically in mitotic spindle and GCP6-knockdown cells revealed differences in the requirement assembly, but the underlying mechanism remained unclear of these GCPs for γTuRC integrity. In all cases, the degree of (Masuda and Toda, 2016). Our findings now allow a mechanistic γTuRC disruption was correlated with impairment of all tested γ- interpretation of this synergistic effect: GCP6 is not only crucial for tubulin-dependent functions, including centrosomal nucleation, γTuRC integrity but also interacts directly with MZT1. In fact, our centriole duplication, assembly of the mitotic spindle and mitotic data indicates that MZT1 binds to all GCPs but GCP4, and not only progression. Cells were particularly sensitive to knockdown of to GCP3, as previously suggested. MZT1 binding is mediated by a GCP6 with phenotypes indistinguishable from defects after GCP3 conserved motif present in the NTEs of GCP2, GCP3, GCP5 and knockdown, suggesting that in human cells γTuSC alone is not able GCP6. The individual binary interactions appear to be relatively to support γ-tubulin-dependent processes. Mechanistically, this weak and in some cases were barely detectable in yeast two-hybrid dependence on γTuRC integrity is also rooted in the fact that only or co-immunoprecipitation assays, which might explain why they intact γTuRC can interact with the targeting factor NEDD1. The were not revealed in previous studies. In this context, it is interesting Journal of Cell Science

413 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 406-419 doi:10.1242/jcs.195321

A -3

AG-GCP3 3A AG-GCP6 3AAG-GCP-6 FLAG-GFPFLAG-GCP-3FLAG-GCP3FLAG-GCP5 FLAG-GCP53A FLAG-GFP 3AFLAG-GCPFL FLAG-GCP5FLAG-GCP5 3A FLAG-GCP-6FL FL FLAG-GCP6 3A (kDa) Input FLAG IP (kDa) Input FLAG IP 100 250 FLAG 150 FLAG 50 γ-Tubulin

37 100 GCP5 10 MZT1 50 γ-Tubulin 37 GAPDH 100 GCP2 75 GCP4 150 GCP5 75 NEDD1 10 MZT1 37 GAPDH B GCP3 GCP3 3A GCP5 GCP5 3A GCP6 GCP6 3A FLAG Pericentrin Merge/DNA

C E Pericentrin 120 FLAG 150 **** ****

80 **** 100 **** **** 40 -tubulin intensity [%] γ 50

Fluorescence intensity [%] 0

GCP3 GCP5 GCP6 Centrosomal GCP3 3A GCP5 3A GCP6 3A 0 Ctrl siRNA GCP3 siRNA D GCP3 siRNA + GCP3 GCP3 siRNA + GCP3 3A +EGFP +EGFP +GCP3 +GCP3 3A interphase mitosisinterphase mitosis F *** *** 25 FLAG

15 NEDD1 10 Mitotic Index [%] 5

0 -Tubulin

γ Ctrl siRNA GCP3 siRNA +EGFP +EGFP +GCP3 +GCP3 3A

Fig. 5. See next page for legend. Journal of Cell Science

414 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 406-419 doi:10.1242/jcs.195321

Fig. 5. GCPs carrying mutations in the MZT1-binding motif are defective at MTOCs. Although we cannot rule out that MZT1 also directly in binding to NEDD1 and targeting to centrosomes. (A) FLAG-tagged wild- regulates γTuRC, our data suggests that MZT1 functions as a ‘ ’ type or MZT1-binding mutant ( 3A ) of full-length GCP3, GCP5 and GCP6 were ‘priming’ factor that prepares γTuRC for interaction with other expressed in HEK293 cells and immunoprecipitated (IP). The associated proteins were identified by western blotting with specific antibodies as proteins to control targeting and activation. indicated. GAPDH was detected as a loading control and to confirm the specificity of the immunoprecipitation. (B) FLAG-tagged constructs as in A MATERIALS AND METHODS were expressed in U2OS cells. After staining with anti-FLAG and anti- Molecular biology pericentrin antibodies, and DAPI to label DNA, cells were analyzed by Full-length MZT1 cDNA (Human MGC Verified FL clone, ID: 40122127, immunofluorescence microscopy. Scale bar: 10 µm. (C) Centrosomal FLAG accession: BC125183) was amplified by PCR and inserted into the plasmid and pericentrin signal intensities of cells stained as in B were quantified and pEGFP-C1 (Clontech) containing a modified multiple clone site with FseI plotted as percentages relative to control cells expressing wild-type proteins and AscI restriction sites. To express GCP2, GCP3, GCP4, GCP5 and GCP6 (set to 100%); n>55 centrosomes combined from three independent full-length proteins and fragments tagged with FLAG, the respective experiments. (D) Immunofluorescence images of HeLa cells depleted of GCP3 cDNAs (Murphy et al., 2001, 1998) were PCR amplified and cloned into a and expressing RNAi-resistant FLAG-tagged wild-type or 3A mutant GCP3. pCS2+-based vector encoding an N-terminal 3xFLAG tag and carrying a Cells were stained with anti-FLAG, anti-NEDD1 and anti-γ-tubulin antibodies. modified cloning site with FseI and AscI restriction sites. For the expression γ Scale bar: 10 µm. (E) Centrosomal -tubulin staining was quantified in of His- or GST-tagged proteins in Escherichia coli, cDNAs encoding interphase HeLa cells transfected with control or GCP3 siRNA, and with MZT1, GCP2 1–506, GCP3 1–552, GCP4 1–347, GCP5 1–713 and GCP6 – – – plasmid expressing FLA EGFP, and FLAG GCP3 or FLAG GCP3 3A 1–710 were inserted into pET28a with a modified cloning site using FseI mutant. Mean intensities were plotted relative to control cells (set to 100%); and AscI or into pGEX-4X-1 (GE Healthcare, Piscataway, NJ) using EcoRI n=60 centrosomes per condition combined from three independent and XhoI restriction sites. Previously described plasmids were used for experiments. (F) The mitotic index was scored in cells transfected with control – – or GCP3 siRNA, and with plasmid expressing FLAG–EGFP, and FLAG–GCP3 expression of GST C-NEDD1 (amino acids 344 667) (Lüders et al., 2006), – – or FLAG-GCP3 3A mutant; n=3 experiments, >200 cells per condition in each EGFP CDK5RAP2 51-100 (Choi et al., 2010) and NEDD1 EGFP (Haren experiment. All quantitative results are mean±s.e.m. ***P<0.001; ****P<0.0001 et al., 2009). For expression in yeast, MZT1 and the N-terminal fragments of (unpaired t-test compared with wild type or as indicated). GCP2–GCP6 where inserted into plasmids pGBKT7 and pGADT7 (Clontech) using modified cloning sites with FseI and AscI restriction sites. GCP3 I93A/L94A/L97A, GCP5 I110A/L111A/L114A, and GCP6 to note that one study showed weak interaction of the Arabidopsis V110A/L111A/L114A mutants, and RNAi-resistant mutants of NEDD1 MZT1 homologs GIP1a and GIP1b with GCP2 (Nakamura et al., and of GCPs generated by introduction of multiple silent mutations in the 2012). However, the authors did not interpret this as a positive target sequences where made by following the QuikChange Site-Directed interaction, possibly due to a much stronger interaction with GCP3 Mutagenesis protocol (Agilent Technologies). Sequence analysis and that was observed in the same assay. alignments were performed with Geneious software (Biomatters). Although binary interactions between MZT1 and GCPs appear to For RNAi-mediated depletion of GCP2, GCP3, GCP4, GCP5, GCP6 and γ MZT1, we used RNA oligonucleotides with the following targeting be weak, binding of MZT1 to TuRC is robust: in sucrose gradients sequences: GCP2 #1, 5′-GAGCUAUGCCUGUACCUAA-3′ (#s21284, the majority of total MZT1 is consistently found to be associated Ambion/Thermo Fisher); GCP2 #2: 5′-GGCUUGACUUCAAUGGUUU- with γTuRC and only a small amount is present in lower molecular 3′ (#s21286, Ambion/Thermo Fisher); GCP3, 5′-GGACUUGCUAAAA mass fractions. Our data is consistent with a cooperative binding CCAGAA-3′ (Silencer Select Pre-designed siRNA, #s20395, Ambion/ mode that allows strong and specific interaction of MZT1 with Thermo Fisher); GCP4, 5′-GCAAUCAAGUGGCGCCUAA-3′ (Choi γTuRC by binding to multiple GCPs. This is in agreement with our et al., 2010); GCP5, 5′-GGAACAUCAUGUGGUCCAUCA-3′ (Izumi observation that mutation of the MZT1-binding motif in only a et al., 2008); GCP6, 5′-AAACGAGACUACUUCCUUA-3′ (Silencer single GCP impairs the interaction of γTuRC with MZT1 and, as a Select Pre-designed siRNA, #s2249992, Ambion/Thermo Fisher); and ′ ′ result, NEDD1-dependent centrosomal targeting. Considering its MZT1, 5 -GCUUUAUCAUCGGUUAUUA-3 (Ambion ID s54042). small size (∼8 kDa), binding of MZT1 to multiple GCPs likely RNAi-mediated depletion of NEDD1 was performed as previously described (Lüders et al., 2006). involves multiple MZT1 molecules, possibly in an oligomeric form. Indeed, purified recombinant MZT1 has been previously shown to form oligomers in solution (Dhani et al., 2013). Antibodies and reagents Based on our results, we propose that the function of MZT1 does Anti-MZT1, anti-GCP2, anti-GCP3, anti-GCP5 and anti-GCP6 rabbit not require interaction with specific GCPs but rather with the polyclonal antibodies were generated against human His–MZT1 or His- γ tagged N-terminal fragments of the human GCPs (GCP2 1–506, GCP3 1– common conserved motif in the NTEs, either in the form of TuRC – – or, in organisms that contain MZT1 but lack GCP4, GCP5, and 552, GCP5 1 713 and GCP6 1 710) expressed in ArticExpress cells (Agilent Technologies), solubilized in 8 M urea and affinity-purified under GCP6 (Lin et al., 2015), in the form of γTuSC oligomers (Fig. 7). γ denaturing conditions using Ni-Sepharose beads (GE Healthcare). The The essential role of MZT1 does not seem to be TuRC assembly, proteins were used for immunization of rabbits (Antibody Production given that MZT1 depletion conditions that produced severe cellular Service, Facultat de Farmàcia, Universitat de Barcelona, Spain). Anti-MZT1 defects, still allowed detection of γTuRC by sucrose gradient specific antibodies were purified by acid elution after binding to GST– fractionation. Instead MZT1 seems to ‘prime’ γTuRC for interaction MZT1 immobilized on Affi-Gel 15 resin (Biorad), the anti-GCP2-6 with other regulatory factors. An open question is how MZT1 antibodies were affinity-purified using the antigens subjected to PAGE promotes these interactions. We envision two scenarios that are not and blotted on membranes. mutually exclusive. MZT1 might directly interact with targeting and Other antibodies used in this study were: mouse anti-γ-tubulin (GTU-88; γ Sigma-Aldrich; western blotting, 1:10,000), mouse anti-γ-tubulin (TU-30, activation factors to promote their binding to TuRC or may act γ indirectly, by inducing a conformational change in the γTuRC Exbio; immunofluorescence, 1:500), rabbit anti- -tubulin (Sigma-Aldrich: immunofluorescence, 1:250), mouse anti-α-tubulin (DM1A, Sigma-Aldrich; structure that increases affinity or exposes a binding site for these immunofluorescence, 1:2000), mouse anti-NEDD1 (7D10, Abnova; western proteins (Fig. 7). blotting, 1:2000), rabbit anti-NEDD1 (immunofluorescence, 1:500; western In summary, our work demonstrates that human MZT1 is central blotting, 1:2500; Lüders et al., 2006), rabbit anti-GCP3 (15719-1-AP, to the regulation of microtubule nucleation by linking specific Proteintech; western blotting, 1:1000), rabbit anti-GCP4 (GTX115949, recognition of fully assembled γTuRC to attachment and activation Genetex; western blotting, 1:500), rabbit anti-GCP5 (BS3303, Bioworld; Journal of Cell Science

415 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 406-419 doi:10.1242/jcs.195321

A GFP GFP-CDK5RAP2 (51-100) siRNA: Control GCP2 MZT1 Control GCP2 MZT1 GFP -Tubulin α

B C Input IP 700 Control siRNA + - + - + - + - MZT1 siRNA - + - + - + - + 600 GFP + + - - + + - - GFP-C5R2 (51-100) - - + + - - + + 500 (kDa) 250 GCP6 400 100 GCP2 10 300 MZT1 GFP 25 200 **** **** D 100 siRNA: *** *** ControlMZT1 (kDa) 250 CDK5RAP2 Cytoplasmic Fluorescence Intensity [%] 0 siRNA: Control GCP2 MZT1 Control GCP2 MZT1 10 MZT1 GFP GFP-CDK5RAP2 (51-100) 37 GAPDH

Fig. 6. MZT1 is required for the activation of γTuRC nucleation activity by mediating interaction with the CDK5RAP2 51–100 fragment. (A) Immunofluorescence images of U2OS cells were transfected with siRNA and with GFP or GFP–CDK5RAP2 51-100, as indicated. Microtubules were depolymerized by being subjected to cold and allowed to regrow for 10 s before fixation. Cells were stained with anti-GFP and anti-α-tubulin antibodies. The magnified regions show nucleated cytoplasmic microtubules for each condition. (B) The amount of nucleated microtubules in the entire non-centrosomal cytoplasmic area of cells as in A was quantified and plotted as percentages of the amount of microtubules measured in cells transfected with control siRNA and GFP (set to 100%); n>76 cells per condition combined from three independent experiments. Results are mean±s.e.m. ***P<0.001; ****P<0.0001 (unpaired t-test). (C) HEK293 cells transfected as in A were subjected to immunoprecipitation (IP) with anti-GFP antibodies. Samples were analyzed by western blotting with antibodies against the indicated proteins. (D) Western blot of MZT1-depleted extract probed with the indicated antibodies. western blotting, 1:500), rabbit anti-GCP6 (AB95172, Abcam; western Cell culture, transfection and microtubule regrowth blotting, 1:2000), rabbit anti-pericentrin (1:500; Lüders et al., 2006), mouse Hek293, U2OS and HeLa (all ATCC) cells were grown in Dulbecco’s anti-GFP (3E6, Thermo Fisher; western blotting, 1:2000; modified Eagle’s medium (DMEM; Invitrogen) supplemented with 10% immunofluorescence, 1:500), rabbit anti-GFP (TP401, Torrey Pines fetal calf serum, at 37°C with 5% CO2, and were routinely confirmed to be Biolabs; western blotting, 1:2000; immunofluorescence, 1:500), mouse free of contamination. Cells were transfected with plasmid using anti-Centrin 3 (H00001070-M01, Abnova; immunofluorescence, 1:500), Lipofectamine 2000 or, in the case of Hek293 cells, with calcium mouse anti-FLAG (F3165, Sigma-Aldrich; immunofluorescence, 1:1000), phosphate, and siRNA using Lipofectamine RNAiMAX (Thermo Fisher). rabbit anti-FLAG (F7425, SigmaAldrich; immunofluorescence, 1:200), For rescue experiments, cells were first transfected with siRNA and 48 h mouse anti-GAPDH (sc-47724, Santa Cruz Biotechnology; western later with plasmid. At 24 h after the second transfection cells were fixed and blotting, 1:10,000). Dilutions of affinity-purified, custom-made antibodies analyzed. varied from preparation to preparation, but were typically used 1:500 for For mitotic microtubule regrowth experiments, dishes containing immunofluorescence and 1:2000 for western blotting. coverslips with transfected HeLa cells were incubated with 100 ng/ml Alexa-Fluor-488, Alexa-Fluor-568-, and Alexa-Fluor-350-conjugated nocodazole for ∼16 h. Nocodazole was then washed out and cells were secondary antibodies used for immunofluorescence microscopy were from incubated for 30 min in an ice-water bath to depolymerize microtubules. For Thermo Fisher (1:200). For co-labeling with two mouse primary antibodies, regrowth experiments in interphase, transfected U2OS cells without drug isotype-specific secondary antibodies were used. Horseradish-peroxidase- treatment were incubated 30 min in an ice-water bath. To allow microtubule coupled secondary antibodies for western blotting were from Jackson regrowth, coverslips with cells were moved to medium at 37°C, followed by

ImmunoResearch Laboratories. methanol fixation. Journal of Cell Science

416 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 406-419 doi:10.1242/jcs.195321

γTuSC Immunoprecipitation and GST pulldown γTuSC-like? Transfected cells were washed in PBS and lysed in buffer (50 mM HEPES, pH 7.5, 150 mM NaCl, 1 mM MgCl2, 1 mM EGTA, 0.5% NP-40 and protease inhibitors) for 10 min on ice. Cleared lysates were obtained by centrifugation for 15 min at 16,000 g at 4°C. For immunoprecipitation, inactive γTuRC lysates were incubated with antibodies for 1 h at 4°C. Sepharose–protein-G beads (GE Healthcare) were added and the mixture was incubated for an additional hour at 4°C. For GST pulldown, purified recombinant GST or MZT1 GST–C-NEDD1 (Lüders et al., 2006) pre-bound to glutathione–Sepharose beads (GE Healthcare) were added to cleared lysates and incubated for 2 h. ‘priming’ The beads were pelleted and washed three times with lysis buffer. Samples were prepared for SDS-PAGE by boiling in sample buffer.

Western blotting Cells were washed in PBS and lysed in buffer (50 mM HEPES pH 7.5, 150 mM NaCl, 1 mM MgCl2, 1 mM EGTA, 0.5% NP-40 and protease adapters inhibitors) on ice. Cleared extracts were prepared by centrifugation and, after determination of protein concentration by Bradford assay (Biorad), subjected to SDS-PAGE. Proteins were transferred to PVDF or targeting & activation nitrocellulose membranes by tank blotting with transfer buffer containing 20% methanol and 0.1% SDS for 90 min and probed with antibodies. For detection of MZT1 transfer buffer without SDS and PVDF membrane with inactive 0.2 µm pore size was used, and transfer time was reduced to 60 min. active

Sucrose gradient centrifugation 300 µl of cell extract prepared as above was loaded on a 4.2 ml 10–40% sucrose gradient prepared in 50 mM HEPES, pH 7.5, 150 mM NaCl, 1 mM MgCl2 and 1 mM EGTA and centrifuged in a SW-55Ti rotor (Beckman) for 4h at ∼366,000 g at 4°C. Fractions were collected from the top of the active gradient by pipetting and analyzed by western blotting. As molecular mass standards, aldolase (158K, 7S) and thyroglobulin (669K, 19S) (GE Healthcare) were analyzed in gradients prepared and run under identical Fig. 7. Model for the role of MZT1 in linking γTuRC assembly to interaction conditions. with adapters that mediate targeting and activation. Cytosolic γTuRC and subcomplexes, such as γTuSC and hypothetical γTuSC-like complexes Fluorescence microscopy (Farache et al., 2016), are inactive as nucleators. Interaction of γTuRC HeLa and U2OS cells grown on coverslips were fixed in methanol at −20°C subcomplexes with MZT1 is weak (dashed arrows). Only fully assembled for at least 5 min and processed for immunofluorescence. Fixed cells were γTuRC interacts efficiently with MZT1 owing to a cooperative binding mode blocked in PBS-BT (1× PBS, 0.1% Triton X-100 and 3% BSA) and involving multiple MZT1 molecules and oligomeric GCPs. The positions of incubated with antibodies in the same buffer. Images were acquired with an GCP4, GCP5 and GCP6 in the γTuRC are unknown and speculatively shown Orca AG camera (Hamamatsu) on a Leica DMI6000B microscope equipped ‘ ’ to be at the seam. Binding of MZT to the N-termini of the GCPs is a priming with 1.4 NA 63× and 100× oil immersion objectives. AF6000 software γ step that allows TuRC to interact with different regulatory adapter proteins. (Leica) was used for image acquisition. Image processing and quantification MZT1 mediates the interaction with adapters either directly or indirectly, for of fluorescence intensities was performed with ImageJ software. Intensities example, by inducing a conformational change that reveals or increases the were measured in images acquired with constant exposure settings and affinity of a binding site for the adapter (bright spot; position hypothetical). The were background-corrected. To quantify non-centrosomal, cytoplasmic adapters might also interact as oligomers, but for simplicity only monomeric adapters are shown. The adapters serve as targeting and activation factors that microtubules in regrowth experiments, the cell boundaries and the area attach γTuRC at MTOCs, such as the centrosome, and activate its nucleation around the centrosome containing astral microtubules were outlined in activity. Targeting and activation might be mediated by the same adapter or by ImageJ. The intensity value obtained for astral microtubules around the distinct adapters as recently suggested (Muroyama et al., 2016). centrosome was subtracted from the intensity value obtained for total microtubule staining within the cell boundaries. Yeast two-hybrid assay To test the interactions between GCPs and MZT1 the Matchmaker two- Statistical analysis hybrid system was used according to the manufacturer’s protocol Statistical analysis was performed using Prism 6 software. Two-tailed, (Clontech). Briefly, plasmids containing a GAL4 DNA-binding domain unpaired t-tests were performed to compare experimental groups. The (pGBKT7 backbone) and plasmids containing a GAL4 activation domain results are reported in the figures and figure legends. (pGADT7 backbone) carrying the desired cDNAs were transformed into yeast strains AH109 and Y187, respectively. Yeast mating was performed Acknowledgements in 2× YPDA medium for 16–20 h, and diploids were selected by leucine We are grateful to visiting student Eva Domenjo for her help with and tryptophan prototrophy. Protein interactions were identified selecting immunoprecipitation experiments. for histidine and adenine prototrophy. Mel1 activity of yeast strains with positive protein interactions was assayed by adding the substrate Competing interests X-α-Gal (5-bromo- 4-chloro- 3-indolyl-α-D-galactopyranoside, Clontech) The authors declare no competing or financial interests. directly to the plates. Yeast strains carrying pGADT7-T (SV40 – large T-antigen 84-708) and pGBKT7-53 (encoding murine p53 72 390), Author contributions and pGADT7-T (encoding SV40 large T-antigen 84-708) and pGBKT7- Conceptualization: R.R.C., N.T.-T., J.R. and J.L.; Methodology, formal analysis, Lam (encoding human lamin C 66–230) served as positive and negative investigation: R.R.C., N.T.-T., A.E., S.E. and C.L.; Resources: R.R.C., N.T.-T., C.L.; controls, respectively. Writing – original draft preparation: J.L.; Writing – review and editing: R.R.C., N.T.-T., Journal of Cell Science

417 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 406-419 doi:10.1242/jcs.195321

A.E., J.R. and J.L.; Visualization: R.R.C., N.T.-T., A.E. and J.L.; Supervision, project microtubule-based microtubule nucleation by mammalian polo-like kinase 1. administration, funding acquisition: J.R. and J.L. Proc. Natl. Acad. Sci. USA 108, 11446-11451. Knop, M. and Schiebel, E. (1997). Spc98p and Spc97p of the yeast gamma-tubulin Funding complex mediate binding to the spindle pole body via their interaction with This study was supported by grants from the Ministerio de Economıaý Spc110p. EMBO J. 16, 6985-6995. Competitividad (MINECO) (BFU2009-08522, BFU2012-33960 and BFU2015- Kollman, J. M., Polka, J. K., Zelter, A., Davis, T. N. and Agard, D. A. (2010). 69275-P to J.L. and BFU2014-58422-P to J.R.), and IRB Barcelona intramural Microtubule nucleating gamma-TuSC assembles structures with 13-fold funds. J.L. acknowledges additional support from the Ramón y Cajal Programme microtubule-like symmetry. Nature 466, 879-882. Kollman, J. M., Merdes, A., Mourey, L. and Agard, D. A. (2011). Microtubule (RYC-2010-07108 MINECO, Spain). R.R.C. was supported by a PhD fellowship nucleation by γ-tubulin complexes. Nat. Rev. Mol. Cell Biol. 12, 709-721. from the Consejo Nacional de Ciencia y Tecnologıá (Mexican National Council for Kollman, J. M., Greenberg, C. H., Li, S., Moritz, M., Zelter, A., Fong, K. K., Science and Technology; CONACYT) (CVU 269229). Fernandez, J.-J., Sali, A., Kilmartin, J., Davis, T. N. et al. (2015). Ring closure activates yeast γTuRC for species-specific microtubule nucleation. Nat. Struct. Supplementary information Mol. Biol. 22, 132-137. Supplementary information available online at Lin, T.-C., Gombos, L., Neuner, A., Sebastian, D., Olsen, J. V., Hrle, A., Benda, http://jcs.biologists.org/lookup/doi/10.1242/jcs.195321.supplemental C. and Schiebel, E. (2011). Phosphorylation of the yeast γ-tubulin Tub4 regulates microtubule function. PLoS ONE 6, e19700. References Lin, T.-C., Neuner, A., Schlosser, Y. T., Scharf, A. N. D., Weber, L. and Schiebel, Anders, A., Lourenço, P. C. C. and Sawin, K. E. (2006). Noncore components of E. (2014). Cell-cycle dependent phosphorylation of yeast pericentrin regulates the fission yeast gamma-tubulin complex. Mol. Biol. Cell 17, 5075-5093. γ-TuSC-mediated microtubule nucleation. eLife 3, e02208. Bahtz, R., Seidler, J., Arnold, M., Haselmann-Weiss, U., Antony, C., Lehmann, Lin, T.-C., Neuner, A. and Schiebel, E. (2015). Targeting of γ-tubulin complexes to W. D. and Hoffmann, I. (2012). GCP6 is a substrate of Plk4 and required for microtubule organizing centers: conservation and divergence. Trends Cell Biol. centriole duplication. J. Cell Sci. 125, 486-496. 25, 296-307. Bauer, M., Cubizolles, F., Schmidt, A. and Nigg, E. A. (2016). Quantitative Lüders, J. and Stearns, T. (2007). Microtubule-organizing centres: a re-evaluation. analysis of human centrosome architecture by targeted proteomics and Nat. Rev. Mol. Cell Biol. 8, 161-167. fluorescence imaging. EMBO J. 35, 2152-2166. Lüders, J., Patel, U. K. and Stearns, T. (2006). GCP-WD is a gamma-tubulin Choi, Y.-K., Liu, P., Sze, S. K., Dai, C. and Qi, R. Z. (2010). CDK5RAP2 stimulates targeting factor required for centrosomal and chromatin-mediated microtubule microtubule nucleation by the gamma-tubulin ring complex. J. Cell Biol. 191, nucleation. Nat. Cell Biol. 8, 137-147. 1089-1095. Lynch, E. M., Groocock, L. M., Borek, W. E. and Sawin, K. E. (2014). Activation of Dhani, D. K., Goult, B. T., George, G. M., Rogerson, D. T., Bitton, D. A., Miller, the γ-Tubulin Complex by the Mto1/2 Complex. Curr. Biol. 24, 896-903. C. J., Schwabe, J. W. R. and Tanaka, K. (2013). Mzt1/Tam4, a fission yeast Manning, J. A., Shalini, S., Risk, J. M., Day, C. L. and Kumar, S. (2010). A direct MOZART1 homologue, is an essential component of the -tubulin complex and interaction with NEDD1 regulates gamma-tubulin recruitment to the centrosome. directly interacts with GCP3Alp6. Mol. Biol. Cell 24, 3337-3349. PLoS ONE 5, e9618. Edgar, R. C. (2004). MUSCLE: multiple sequence alignment with high accuracy and Masuda, H. and Toda, T. (2016). Synergistic role of fission yeast Alp16GCP6 and high throughput. Nucleic Acids Res. 32, 1792-1797. Mzt1MOZART1 in γ-tubulin complex recruitment to mitotic spindle pole bodies Farache, D., Jauneau, A., Chemin, C., Chartrain, M., Rémy, M.-H., Merdes, A. and spindle assembly. Mol. Biol. Cell 27, 1753-1763. and Haren, L. (2016). Functional analysis of gamma-tubulin complex proteins Masuda, H., Mori, R., Yukawa, M. and Toda, T. (2013). Fission yeast MOZART1/ indicates specific lateral association via their N-terminal domains. J. Biol. Chem. Mzt1 is an essential -tubulin complex component required for complex recruitment 291, 23112-23125. to the microtubule organizing center, but not its assembly. Mol. Biol. Cell 24, Fong, K.-W., Choi, Y.-K., Rattner, J. B. and Qi, R. Z. (2008). CDK5RAP2 is a 2894-2906. pericentriolar protein that functions in centrosomal attachment of the gamma- Moudjou, M., Bordes, N., Paintrand, M. and Bornens, M. (1996). Gamma-Tubulin tubulin ring complex. Mol. Biol. Cell 19, 115-125. in mammalian cells: the centrosomal and the cytosolic forms. J Cell Sci 109, Fujita, A., Vardy, L., Garcia, M. A. and Toda, T. (2002). A fourth component of the 875-887. fission yeast gamma-tubulin complex, Alp16, is required for cytoplasmic Muroyama, A., Seldin, L. and Lechler, T. (2016). Divergent regulation of microtubule integrity and becomes indispensable when gamma-tubulin function functionally distinct γ-tubulin complexes during differentiation. J. Cell Biol. 213, is compromised. Mol. Biol. Cell 13, 2360-2373. 679-692. Gomez-Ferreria, M. A., Bashkurov, M., Helbig, A. O., Larsen, B., Pawson, T., Murphy, S. M., Urbani, L. and Stearns, T. (1998). The mammalian gamma-tubulin Gingras, A.-C. and Pelletier, L. (2012). Novel NEDD1 phosphorylation sites complex contains homologues of the yeast spindle pole body components spc97p regulate γ-tubulin binding and mitotic spindle assembly. J. Cell Sci. 125, and spc98p. J. Cell Biol. 141, 663-674. 3745-3751. Murphy, S. M., Preble, A. M., Patel, U. K., O’Connell, K. L., Dias, D. P., Moritz, M., Gunawardane, R. N., Martin, O. C., Cao, K., Zhang, L., Dej, K., Iwamatsu, A. and Agard, D., Stults, J. T. and Stearns, T. (2001). GCP5 and GCP6: two new Zheng, Y. (2000). Characterization and reconstitution of Drosophila gamma- members of the human gamma-tubulin complex. Mol. Biol. Cell 12, 3340-3352. tubulin ring complex subunits. J. Cell Biol. 151, 1513-1524. Nakamura, M., Yagi, N., Kato, T., Fujita, S., Kawashima, N., Ehrhardt, D. W. and Haren, L., Remy, M.-H., Bazin, I., Callebaut, I., Wright, M. and Merdes, A. (2006). Hashimoto, T. (2012). Arabidopsis GCP3-interacting protein 1/MOZART 1 is an NEDD1-dependent recruitment of the gamma-tubulin ring complex to the integral component of the γ-tubulin-containing microtubule nucleating complex. centrosome is necessary for centriole duplication and spindle assembly. J. Cell Plant J 71, 216-225. Biol. 172, 505-515. Nguyen, T., Vinh, D. B., Crawford, D. K. and Davis, T. N. (1998). A genetic analysis Haren, L., Stearns, T. and Lüders, J. (2009). Plk1-dependent recruitment of of interactions with Spc110p reveals distinct functions of Spc97p and Spc98p, gamma-tubulin complexes to mitotic centrosomes involves multiple PCM components of the yeast gamma-tubulin complex. Mol. Biol. Cell 9, 2201-2216. components. PLoS ONE 4, e5976. Oakley, B. R., Paolillo, V. and Zheng, Y. (2015). γ-Tubulin complexes in Hutchins, J. R. A., Toyoda, Y., Hegemann, B., Poser, I., Hériché, J.-K., Sykora, microtubule nucleation and beyond. Mol. Biol. Cell 26, 2957-2962. M. M., Augsburg, M., Hudecz, O., Buschhorn, B. A., Bulkescher, J. et al. Petry, S. and Vale, R. D. (2015). Microtubule nucleation at the centrosome and (2010). Systematic analysis of human protein complexes identifies chromosome beyond. Nat. Cell Biol. 17, 1089-1093. segregation proteins. Science 328, 593-599. Pinyol, R., Scrofani, J. and Vernos, I. (2013). The role of NEDD1 phosphorylation Izumi, N., Fumoto, K., Izumi, S. and Kikuchi, A. (2008). GSK-3beta regulates by Aurora A in chromosomal microtubule nucleation and spindle function. Curr. proper mitotic spindle formation in cooperation with a component of the gamma- Biol. 23, 143-149. tubulin ring complex, GCP5. J. Biol. Chem. 283, 12981-12991. Remy, M.-H., Merdes, A. and Gregory-Pauron, L. (2013). Assembly of gamma- Jakobsen, L., Vanselow, K., Skogs, M., Toyoda, Y., Lundberg, E., Poser, I., tubulin ring complexes: implications for cell biology and disease. Prog. Mol. Biol. Falkenby, L. G., Bennetzen, M., Westendorf, J., Nigg, E. A. et al. (2011). Novel Transl. Sci. 117, 511-530. asymmetrically localizing components of human centrosomes identified by Rivero, S., Cardenas, J., Bornens, M. and Rios, R. M. (2009). Microtubule complementary proteomics methods. EMBO J. 30, 1520-1535. nucleation at the cis-side of the Golgi apparatus requires AKAP450 and GM130. Janski, N., Masoud, K., Batzenschlager, M., Herzog, E., Evrard, J.-L., Houlné, EMBO J. 28, 1016-1028. G., Bourge, M., Chabouté, M.-E. and Schmit, A.-C. (2012). The GCP3- Roubin, R., Acquaviva, C., Chevrier, V., Sedjaï, F., Zyss, D., Birnbaum, D. and interacting proteins GIP1 and GIP2 are required for γ-tubulin complex protein Rosnet, O. (2013). Myomegalin is necessary for the formation of centrosomal and localization, spindle integrity, and chromosomal stability. Plant Cell 24, Golgi-derived microtubules. Biol. Open 2, 238-250. 1171-1187. Samejima, I., Miller, V. J., Groocock, L. M. and Sawin, K. E. (2008). Two distinct Johmura, Y., Soung, N.-K., Park, J.-E., Yu, L.-R., Zhou, M., Bang, J. K., Kim, regions of Mto1 are required for normal microtubule nucleation and efficient

B.-Y., Veenstra, T. D., Erikson, R. L. and Lee, K. S. (2011). Regulation of association with the gamma-tubulin complex in vivo. J. Cell Sci. 121, 3971-3980. Journal of Cell Science

418 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 406-419 doi:10.1242/jcs.195321

Sawin, K. E., Lourenço, P. C. C. and Snaith, H. A. (2004). Microtubule nucleation set of core components and identifies the novel subunit GCP8. Mol. Biol. Cell 21, at non-spindle pole body microtubule-organizing centers requires fission yeast 3963-3972. centrosomin-related protein mod20p. Curr. Biol. 14, 763-775. Teixidó-Travesa, N., Roig, J. and Lüders, J. (2012). The where, when and how of Scheidecker, S., Etard, C., Haren, L., Stoetzel, C., Hull, S., Arno, G., Plagnol, V., microtubule nucleation - one ring to rule them all. J. Cell Sci. 125, 4445-4456. Drunat, S., Passemard, S., Toutain, A. et al. (2015). Mutations in TUBGCP4 Venkatram, S., Tasto, J. J., Feoktistova, A., Jennings, J. L., Link, A. J. and alter microtubule organization via the γ-tubulin ring complex in autosomal- Gould, K. L. (2004). Identification and characterization of two novel proteins recessive microcephaly with chorioretinopathy. Am. J. Hum. Genet. 96, 666-674. affecting fission yeast gamma-tubulin complex function. Mol. Biol. Cell 15, Scrofani, J., Sardon, T., Meunier, S. and Vernos, I. (2015). Microtubule nucleation 2287-2301. Vérollet, C., Colombié, N., Daubon, T., Bourbon, H.-M., Wright, M. and in mitosis by a RanGTP-dependent protein complex. Curr. Biol. 25, 131-140. Raynaud-Messina, B. (2006). Drosophila melanogaster gamma-TuRC is Sdelci, S., Schütz, M., Pinyol, R., Bertran, M. T., Regué, L., Caelles, C., Vernos, I. dispensable for targeting gamma-tubulin to the centrosome and microtubule and Roig, J. (2012). Nek9 phosphorylation of NEDD1/GCP-WD contributes to nucleation. J. Cell Biol. 172, 517-528. Plk1 control of gamma-tubulin recruitment to the mitotic centrosome. Curr. Biol. Wang, Z., Wu, T., Shi, L., Zhang, L., Zheng, W., Qu, J. Y., Niu, R. and Qi, R. Z. 22, 1516-1523. (2010). Conserved motif of CDK5RAP2 mediates its localization to centrosomes Takahashi, M., Yamagiwa, A., Nishimura, T., Mukai, H. and Ono, Y. (2002). and the Golgi complex. J. Biol. Chem. 285, 22658-22665. Centrosomal proteins CG-NAP and kendrin provide microtubule nucleation sites Xiong, Y. and Oakley, B. R. (2009). In vivo analysis of the functions of gamma- by anchoring gamma-tubulin ring complex. Mol. Biol. Cell 13, 3235-3245. tubulin-complex proteins. J. Cell Sci. 122, 4218-4227. Teixidó-Travesa, N., Villén, J., Lacasa, C., Bertran, M. T., Archinti, M., Gygi, Zimmerman, W. C., Sillibourne, J., Rosa, J. and Doxsey, S. J. (2004). Mitosis- S. P., Caelles, C., Roig, J. and Lüders, J. (2010). The gammaTuRC revisited: a specific anchoring of gamma tubulin complexes by pericentrin controls spindle comparative analysis of interphase and mitotic human gammaTuRC redefines the organization and mitotic entry. Mol. Biol. Cell 15, 3642-3657. Journal of Cell Science

419