The COMA complex is required for Sli15/INCENP-mediated correction ...

[Cell Cycle 8:16, 2570-2577; 15 August 2009]; ©2009 Landes Bioscience

Report

The COMA complex is required for Sli15/INCENP-mediated correction of defective kinetochore attachments James Knockleby1 and Jackie Vogel1,2,* 1Department

of Biology; and 2DBRI; McGill University; Montreal, QC Canada

Abbreviations: CFP, cyan fluorescent protein; MFI, mean fluorescent intensity; MT, microtubule; NZ, nocodazole; SAC, spindle assembly checkpoint; SC, sister chromatid; SPB, spindle pole body; VFP, venus yellow fluorescent protein; WT, wild-type Key words: kinetochore, tension, chromosome segregation INCENP, cell cycle, yeast

Before anaphase, kinetochores of sister chromatid (SC) pairs must be attached to microtubules emanating from opposing spindle poles. The conserved Aurora B kinase Ipl1 and its adaptor INCENP/Sli15 correct defective attachments and promote SC bi-orientation. Ame1 and Okp1 are essential components of the COMA sub-complex of the budding yeast central kinetochore, and Ame1 has been demonstrated to physically interact with Sli15. Here, we examine the significance of the Ame1-Sli15 interaction in vivo. We find Sli15 localization at kinetochores is reduced in the absence of Ame1. We demonstrate a role for Ame1 in the correction of defective attachments. While overexpression of OKP1 restores kinetochore localization of the mutant Ame1-4 protein, Sli15 localization is not restored and defective attachments are not corrected. Our findings reveal a new role for the central kinetochore in promoting Sli15 function, and suggest functional Ame1 is required for the correction of defective attachments by promoting the localization of Sli15/INCENP at kinetochores.

Introduction Accurate segregation of chromosomes during mitosis is required for the genomic stability and continued viability of cells. Microtubules (MTs) attach to SCs via the kinetochore, a multiprotein complex that binds to the centromere.1 Bi-orientation is achieved when the kinetochores of paired SCs are attached to the plus ends of MTs originating from opposite spindle poles. Bi-orientation of all chromosomes in metaphase ensures high fidelity segregation of SCs toward opposite spindle poles during anaphase (reviewed in ref. 2). *Correspondence to: Jackie Vogel; Department of Biology; Bellini Pavilion; McGill Life Sciences Complex; Room 269; McGill University; 3649 Promenade Sir William Osler; Montreal QC H3G 0B1 CA; Tel.: 514.398.5880; Fax: 514.398.5069; Email: [email protected] Submitted: 03/30/09; Revised: 06/07/09; Accepted: 06/12/09 Previously published online as a Cell Cycle E-publication: http://www.landesbioscience.com/journals/cc/article/9267

2570

Bi-orientation is an error-prone process and requires surveillance systems that detect and correct defects in kinetochore-MT attachments.3 The forces exerted by spindle MTs are sufficient to promote bi-oriented SCs in budding yeast.4 Application of force that produces tension between the two SCs is sufficient to silence the spindle assembly checkpoint (SAC).5 Defective attachment, e.g., syntelic attachment (both SCs attached to one pole) will not establish tension between the two SCs and this defect is reported locally at that kinetochore. The conserved Ipl1/Aurora B kinase and its adaptor/activator Sli15/INCENP are required to correct defective attachments.2 Phosphorylation of outer kinetochrore proteins by Ipl1/Sli15 releases MT plus ends and allows new attachments to form; this correction process ultimately achieves bi-orientation.6-8 Once all SCs are bi-oriented, tension between the SCs silences the spindle assembly checkpoint9 and anaphase is initiated through the activation of the cohesin protease Esp1/Separase which cleaves the cohesin subunit Scc1 thereby allowing SCs to be segregated to opposite poles.8,10,11 At the metaphase to anaphase transition, Ipl1 and Sli15 are translocated from the kinetochore to the spindle MTs in response to the release of Cdc14, possibly to prevent Sli15/Ipl1 from inappropriately destabilizing MT-kinetochore attachments during anaphase.12 Sli15 not only has a role in detaching erroneous MT-kinetochore attachments, but also acts in the surveillance mechanism that detects the presence of syntelic attachments. Ipl1/Sli15 together with the Survivin ortholog, Bir1, participate in the mechanicalchemical coupling required to translate the lack of centromere tension into activation of the spindle assembly checkpoint. Sli15 was shown to bind to the CBF3 inner kinetochore complex via Bir1/Survivin, and to MTs in vitro; MT binding was essential for sensing lack of tension at the kinetochores in vivo.13 Sli15 and Ipl1 have been shown to interact with components of the outer, central and inner kinetochore.14-16 Sli15 interaction with kinetochore components has generally been looked at only in the context of Ipl1 and its substrates.14,15 Therefore, the biological significance of the specific interaction between Sli15 and a building block of the central kinetochore, the COMA sub-complex (Ctf19,

Cell Cycle

2009; Vol. 8 Issue 16

Sli15 kinetochore function requires COMA complex

Okp1, Mcm21 and Ame1),16 has not been characterized. This interaction is primarily mediated through the Ame1 component of the COMA complex.16 Mutations in COMA proteins result in defects in chromosome segregation, suggesting an important role for the COMA complex in mediating proper kinetochoreMT attachments.17-20 Furthermore, loss of function ame1 mutant (ame1-4) results in defects in SC attachment and an unstable checkpoint response. Over-production of Okp1 in ame1-4 cells restores the kinetochore localization of the mutant Ame1-4 protein and a stable metaphase arrest.20 However, defective attachments persisted in these cells, raising the possibility that Sli15 coupling to functional Ame1 may be important for attachment correction. In this study, we provide evidence that Ame1 participates in the kinetochore localization of Sli15 in metaphase-arrested cells, and in the correction of defective MT-kinetochore attachments. We demonstrate that disruption of the COMA sub-complex by the ame1-4 mutation disrupts Sli15 but not Bir1 kinetochore localization. Cells bearing okp1-5 or ame1-4 mutations, which respectively partially or fully disrupt the COMA complex, exhibit monopolar attachments and un-attached SCs. Similar to ipl1 mutants, ame1-4 cells form syntelic attachments. Importantly, in the absence of the COMA complex, or functional Ame1, defective attachments remain un-corrected. Our findings are consistent with a role for Ame1 and the COMA complex as a critical scaffold required for Sli15 function.

Results Ame1 promotes the kinetochore localization of Sli15. The previously identified physical interaction between Ame1 and Sli15,16 is consistent with a functional relationship between Ame1 and Sli15 at kinetochores. To test this prediction in vivo, the relative mean fluorescent intensity (MFI) of Sli15-VFP foci was determined in cells arrested in metaphase with the microtubule destabilizing drug nocodazole (NZ) and incubated at 37°C, the restrictive temperature of the ame1-4 mutant.20 We chose to use NZ to overcome an Sli15 detection issue associated with this analysis. We found the intensity of Sli15-VFP signal to be extremely low and diffuse in the nucleus in wild-type (WT) metaphase cells incubated at 37°C. Low levels of Sli15-VFP (foci between or near the two SPBs) was frequently detected in okp1-5 cells raised at 37°C. Sli15-VFP was generally diffuse in ame1-4 cells, similar to WT cells (Suppl. Fig. 1). Moreover, the distinction between metaphase and anaphase is not straightforward; ame1-4 cells do not have a stable metaphase arrest20 and metaphase spindles can be abnormally long as a consequence of attachment defects in both ame1-4 and okp1-5 cells incubated at 37°C. We felt robust and steretotyped Sli15 localization was required in order to unambiguously and quantitatively compare the three populations; WT, ame1-4 and okp1-5. In NZ-treated cells, we determined that robust intensity of Sli15-VFP signal could be detected in metaphase arrested WT cells (Fig. 1). The metaphase arrest induced by NZ therefore provided a basis for quantitative measurement of Sli15 kinetochore localization in vivo. The SPB component Spc29-CFP was as a landmark for kinetochore localization in NZ-treated cells, as kinetochores cluster near the SPB when microtubules collapse. WT, okp1-5 and ame1-4 cells www.landesbioscience.com

were compared to determine if Ame1 alone or the entire COMA complex is specifically required for Sli15 localization. WT, ame1-4 and okp1-5 cells expressing Sli15-VFP and Spc29-CFP were arrested in metaphase with 20 μg/ml NZ, shifted into pre-warmed media and incubated at 37°C, collected at 3 hours post temperature shift and the mean fluorescent intensity (MFI) of Sli15-VFP foci adjacent to the spindle poles measured (see Materials and Methods). In WT cells (n = 312), Sli15-VFP (MFI 1.00, arbitrary units) localizes as one focus near the Spc29-CFP labelled SPBs (Fig. 1A). A similarly intense focus of Sli15-VFP (MFI 1.17; Fig. 1D) is detected adjacent to the Spc29-CFP labelled SPBs in okp1-5 cells (n = 313) after 3 hours (Fig. 1A). Although Sli15 intensities in ame1-4 cells are similar to levels in WT cells after 2 hours (MFI 1.05; data not shown), a significant reduction in Sli15-VFP MFI adjacent to the SPB was observed in ame1-4 cells (n = 307) after 3 hours (MFI 0.32; Fig. 1A and D). The loss of Sli15 localization was not due to the destabilization of the Sli15 protein, as a TAP epitope-tagged Sli15 remains stable and at WT levels in ame1-4 cells raised at restrictive temperature for 3 hrs (Fig. 1B). The observed decrease in Sli15-VFP intensity in ame1-4 cells relative to WT and okp1-5 cells suggests that Ame1 is specifically involved in the localization of Sli15, since Ame1 is the sole COMA component associated with kinetochores in okp1-5 cells.20 Sli15 has been found to be physically associated with Bir1 and Ipl1, and this interaction is required for proper Ipl1 function in activation of the spindle checkpoint.21 In agreement with a previous study,8 we were unable to detect Ipl1-VFP as a discrete focus in cells arrested in metaphase with NZ (data not shown). Therefore, we asked if Bir1 mis-localized from the kinetochore at metaphase in ame1-4 cells. We could detect a focus of Bir1-VFP adjacent to the SPBs in WT (n = 304) and ame1-4 (n = 300) cells arrested with NZ after 3 hours (Fig. 1C). There was no significant difference between the MFI of Bir1-VFP between WT and ame1-4 cells (Fig. 1E), suggesting that neither the COMA nor Ame1 is required to maintain Bir1 at the kinetochore. This result indicates that a pool of Bir1 localizes to the kinetochore independently of Sli15, presumably through the previously described interaction between CBF3 and Bir1 at the kinetochore.13 Overall, our data indicates that Ame1 is specifically required to maintain the kinetochore localization of Sli15, but not Bir1, in metaphasearrested cells. Correction of defective kinetochore attachments requires functional Ame1. Previously, we showed that a conditional loss of function mutation of Ame1, ame1-4, has a role in promoting the maintenance of the SAC.20 To further characterize the role of Ame1 in chromosome segregation, we examined the role of Ame1 and, by extension, the COMA complex, in forming bi-oriented kinetochore-MT attachments. To delineate a specific role for Ame1, we compared two alleles of COMA complex proteins, ame1-4 and okp1-5. okp1-5 and ame1-4 mutants disrupt the COMA complex but do not perturb other kinetochore complexes.16,20 However, okp1-5 and ame1-4 mutants differ in their impact on the kinetochore localization of the COMA components; Ame1 remains associated with kinetochore in okp1-5 cells while in ame1-4 cells all four COMA components are mis-localized.20 Thus,

Cell Cycle

2571


Figure 1. Ame1 is required for Sli15 localization to the kinetochore. (A) Representative images of Sli15-VFP and Spc29-CFP in NZ-treated WT, ame1-4 and okp1-5 cells 3 hours after the shift to restrictive temperature (37°C). Insets shown are a 3X magnification of the area (surrounding the SPB) used for intensity measurements. (B) Quantification of Bir1-VFP MFI in WT and ame1-4 cells (measured as in D). Bar = 2 μm, insets = 0.6 μm. (C) Representative images of Bir1-VFP localization in WT and ame1-4 cells. Bir1-VFP localization is not significantly reduced in ame1-4 cells. (D) Sli15-TAP protein remains stable in NZ-treated WT, ame1-4 and okp1-5 cells raised at restrictive temperature as in (A). (E) Relative fluorescence intensity (MFI: corrected and normalized to WT values, see methods) of Sli15-VFP 3 hours at 37°C. After 3 hours, Sli15-VFP is significantly reduced in ame1-4 cells, relative to WT and okp1-4.

the comparison of kinetochore-MT attachments in ame1-4 and okp1-5 mutants would allow the investigation of the requirement for Ame1 in the correction of defective attachments. Kinetochore attachment in ame1-4 and okp1-5 cells was visualized by proxy using the LacO/LacI-GFP reporter system for the centromere of chromosome XV as previously described.22 The position of the poles and the axis of the mitotic spindle was determined using a SPB marker, Spc29-CFP. GFP-labelled centromeres reside along the axis of the spindle when bi-oriented. In contrast, defective monopolar attachments result in SCs that are positioned adjacent to one of the two spindle poles.23 Un-attached SCs reside outside the axis of the spindle or “off-axis”.23 Defective attachments were defined as SCs positioned adjacent to a SPB (adjacent: within 25% of the pole-to-pole length, in μm). SCs were considered unattached when positioned off-axis relative to the axis defined by the spindle poles (see Fig. 2). WT, ame1-4 and okp1-5 cells expressing LacI-GFP and Spc29-CFP were arrested in G1 with α-factor and shifted into pre-warmed media at 37°C. After two hours, the majority of WT 2572

cells had completed one cell cycle and re-entered a second, and cells that were in metaphase exhibited exclusively on-axis SCs (Fig. 2). In contrast, ame1-4 and okp1-5 cells arrested at G2/M and in the case of both mutants, mixed populations of cells with on-axis (attached), off-axis (unattached) and monopolar SCs were observed, the latter comprising more than 50% of all SCs (Fig. 2). Thus, after 2 hours at the restrictive temperature, both ame1-4 and okp1-5 mutants exhibited a substantial number of bi-orientation defects. Although both ame1-4 and okp1-5 cells have attachment defects, a large difference in attachment correction in the two populations of mutant cells emerged after 4 hours. The majority of okp1-5 cells (n = 312) have on-axis SCs at 4 hours post-release (58%, Fig. 2). Furthermore, the population of cells with monopolar attachments decreased in okp1-5; suggesting that correction of defective attachments had occurred. In contrast, the majority of ame1-4 cells at 4 hours (n = 309) contained off-axis SCs (59%) or monopolar attachments (36%; Fig. 2). The number of on-axis SCs was reduced (Fig. 2), indicating that attachments (incorrect or correct)

Cell Cycle

2009; Vol. 8 Issue 16


are lost over time in ame1-4 cells. Since the sole component of the COMA that remains localized in okp1-5 cells is Ame1, and the entire COMA is mis-localized in ame1-4 cells,20 this data strongly suggests that Ame1 is required for both correcting defective kinetochore-MT attachments and maintaining attachments. Mis-segregation of centromeres into the bud is a phenotype is observed in ipl1 mutant allele cells and is attributed to the formation of syntelic attachments that are not corrected.8 If new attachments can indeed form in the absence of the COMA complex, but cannot be corrected, ame1-4 cells would be expected to exhibit syntelic attachments. We detected syntelic attachments in WT cells and ame1-4, ipl1-321 and ame1-4 ipl1-321 mutants using a previously described assay,13,23 based in the bud-directed segregation of SC pairs. After 4 hours at restrictive temperature, a Figure 2. Loss of Ame1 at kinetochores leads to uncorrected monopolar attachments and un-attached proportion of ame1-4 cells enter anaphase SCs. The position of GFP-labeled centromeres (acting as proxy reporters for the SCs) was measured in with separated SCs,20 thus co-segregation relation to the spindle axis and to the position of the SPBs in WT, ame1-4 and okp1-5 cells. Defective of paired SCs into the bud can be scored at attachments are observed in both ame1-4 and okp1-5 cells after 2 hours at 37°C. After 4 hours, 37°C. Cells were synchronized in G1 with defective attachments are corrected in okp1-5 cells, but are not corrected in ame1-4 cells. α-factor, released and incubated at 37°C for 4 hours. Less than 2% of WT (n = 312) cells exhibited syntelic attachments (Fig. 3). Importantly, we observed bud-directed SCs in ame1-4 (36%, n = 314) and ipl1-321 cells (47%, n = 302; Fig. 3). The observed difference between ame1-4 and ipl1-321 cells in the frequency of syntelic attachments formed is likely due to an additional defect in maintaining attachments associated with the loss of the COMA complex in ame1-4 cells, since we observed ipl1-321 ame1-4 double mutant cells have similar numbers of bud-directed SCs compared to ame1-4 cells (35%, n = 301; Fig. 3). Together, these data suggest that new kinetochore attachments can form in the absence of Ame1 and the COMA complex. However, as in ipl1 cells, the mechanism promoting bi-orientation fails in ame1-4 cells, resulting in an increased number of uncorrected syntelic attachments. Overexpression of OKP1 does not reinstate Sli15 localization or function. Overproduction of Okp1 restores the kinetochore localization of the mutant Ame1-4 protein, and also stabilizes metaphase arrest of ame1-4 cells.20 We asked if Sli15 localization at the kinetochore requires a functional Ame1 protein. We next asked if over-production of Okp1 might influence Sli15 localizaFigure 3. ame1-4 cells form syntelic attachments. Syntelic attachments are tion to kinetochores in ame1-4 cells. To test this, WT and ame1-4 reported by pairs of SCs that co-segregated into the bud in anaphase. cells expressing Spc29-CFP and either 2u-OKP1 or control Quantification of WT, ame1-4, ipl1-321 and ipl1-321 ame1-4 cells that plasmid were arrested in metaphase with NZ, and incubated for formed syntelic attachments that remained uncorrected. 3 hours at 37°C. Sli15 localization is decreased at kinetochores in ame1-4 cells bearing 2μ-OKP1 plasmid or control plasmid rela- of the plasmid control (Fig. 4B; WT n = 302 MFI: 1.00; ame1-4 tive to WT (Fig. 4A). Okp1 overproduction has no significant pRS425 n = 305 MFI: 0.24; ame1-4 2μ-OKP1 n = 301 MFI: effect on Sli15 kinetochore localization; the MFI of Sli15-VFP 0.35). Taken together, these results suggest that functional Ame1 in the 2u-OKP1 condition is not significantly different than that is required for Sli15 kinetochore localization. www.landesbioscience.com

Cell Cycle

2573


populations have a majority of cells with monopolar attachments or unattached SCs (ame1-4 pRS425, n = 315: 66% monopolar, 21% offaxis; ame1-4 2μ-OKP1, n = 304: 70% monopolar, 12% off axis; Fig. 5). After four hours, the 2μ-OKP1 cell population is similar in respect to the plasmid control population, with the majority of cells having monopolar attachments or unattached SCs (ame1-4 pRS425, n = 313: 51% monopolar, 42% off-axis; ame1-4 2μ-OKP1, n = 305: 49% monopolar, 42% off axis; Fig. 5). Thus in cells where the checkpoint has been stabilized and the kineFigure 4. Overexpression of OKP1 does not reinstate Sli15 localization in ame1-4 cells. (A) Representative tochore localization of the mutant images of metaphase arrested WT, ame1-4 2μ-OKP1 and ame1-4 PRS425 cells, after 3 hours at 37°C. Insets ame1-4 protein has been restored, shown are a 3X magnification of the area (surrounding the SPB) used for intensity measurements. (B) Levels defective attachments remain uncorof relative MFI of Sli15-VFP from (A). Bar = 2 μm, insets = 0.6 μm. rected. Taken together, these data indicate that the Sli15-dependent repair mechanism is defective due to a specific perturbation of Ame1 function.

Discussion Over the past decade, the S. cerevisiae kinetochore has been extensively characterized. Most of the protein components that make up the structure have been elucidated, and have been assigned to discrete functional subcomplexes. The function of the COMA complex, however, remains relatively uncharacterized. Subunits of the COMA complex have been shown to be required for side-on kinetochore attachments3 and maintenance of a stable checkpoint response, but the COMA complex is not required for the assembly of the kinetochore.16,20 Our data suggests an additional novel role for the COMA complex in the correction of defective MT-attachments. We found that the kinetochore localization of Sli15 Figure 5. Overexpression of OKP1 does not reinstate Sli15 function in ame1-4 cells. The position is dependent on the proper function of Ame1. of GFP-labeled centromeres (acting as proxy reporters for the SCs) was measured in relation to As a consequence of the loss of Sli15 function, the spindle axis and to the position of the SPBs in WT cells and ame1-4 cells containing either ame1-4 cells are unable to correct defective a control 2μ plasmid (pRS425) or 2μ pOKP1. The over-production of Okp1, which reinstates Ame1-4 protein localization and a stable metaphase arrest, does not promote the correction of kinetochore-microtubule attachments. We have now shown that the interaction between Ame1 attachment defects. and Sli15 is important for Sli15’s function in the correction of defective attachments. We We next assayed the effect of over-production of Okp1 on SC propose that the major Sli15-interacting partner in the COMA attachments in ame1-4 cells bearing a 2u-OKP1 plasmid or pRS425 complex is Ame1, based on the differences we have described plasmid control. WT and ame1-4 cells expressing LacI-GFP and between ame1-4 and okp1-5. We have previously shown that Ame1 Spc29-CFP and bearing either 2μ-OKP1 or pRS425 were arrested remains localized to the kinetochore in okp1-5 cells raised at restricin G1 with α-factor and released into pre-warmed media and tive temperature,20 while Ctf19 and Mcm21 are mislocalized.9 In incubated at 37°C. As expected, after two hours, both mutant contrast, none of the COMA complex proteins remain localized at 2574

Cell Cycle

2009; Vol. 8 Issue 16


the kinetochore in ame1-4 cells at restrictive temperature, including Ame1.20 Even though Okp1, Mcm21 and Ctf19 are mislocalized from the kinetochore in okp1-5 cells, Sli15 remains localized to the kinetochore at WT levels (Fig. 1). However, if the entire COMA complex is mislocalized, as in ame1-4 cells, Sli15 mislocalizes from the kinetochore. Even if a mutant form of ame1-4p is returned to the kinetochore, as is the case in cells with overexpressed OKP1, Sli15 does not remain localized to the kinetochore. We cannot conclude that Sli15 solely interacts with COMA via Ame1, as Okp1, Mcm21 and/or Ctf19 may play subtle but nonessential roles. However our data indicate that Ame1 plays a primary role within the COMA complex in facilitating the kinetochore localization of Sli15. Further work will be required to characterize the biochemical nature of the Sli15-Ame1 interaction, as the ame1-4 allele contains >20 non-silent amino acid changes throughout the primary sequence, preventing conclusions regarding structurefunction relationships (Suppl. Table 1). The COMA complex as a regulatory scaffold. The COMA complex was first identified as a tight heterotetramer that formed the core of the Ctf19 complex.16,24 The larger Ctf19 complex consists of loosely bound accessory proteins, of which the function of many remains unclear. What is striking is the number of regulatory proteins that potentially interact with sub-units of the COMA complex. For example, Bub3, a member of the SAC, was found to co-purify with Okp1, linking the COMA complex to the SAC. Additionally, Sli15 was shown to co-purify with Ame1.16 We have previously shown that the COMA complex plays a role in maintaining a checkpoint response to defective attachments.20 We have shown that Sli15 remains mis-localized in ame1-4 cells even with OKP1 overexpression. This result suggests that the checkpoint maintenance observed in ame1-4 2m-OKP1 cells does not require the tension sensing mechanism of Sli15/Ipl1. Instead, the checkpoint maintenance observed in ame1-4 2m-OKP1 cells is likely due to the reinstatement of the SAC by stabilizing Ame1 and/ or Okp1 at the kinetochore. With the checkpoint re-established, the SAC can respond to the persistent monopolar/unattached kinetochores remaining in ame1-4 2m-OKP1 cells. One possibility is that Ame1 or Okp1 are required to anchor Bub3 to the kinetochore until bi-orientation is achieved. The reinstatement of the SAC in ame1-4 2m-OKP1 cells would allow for Bub3 to remain localized to the kinetochore via Okp1. Overall, based on the other co-purification and protein-protein interacting data, it is tempting to speculate that the COMA complex could act as a scaffold for regulatory proteins acting in both attachment and checkpoint control. The binding of regulatory proteins to the COMA complex may also be dynamic and depend on multiple factors, including phosphorylation state of both the scaffold and the associated regulatory protein. All four sub-units of the COMA complex have consensus sequences for at least one of the major mitotic kinases, Cdc28/ Cdk1, Cdc5/Polo and Ipl1/Aurora B (data not shown). Therefore, the COMA complex may play an important role in regulating central kinetochore functions in attachment and checkpoint function in relation to cell cycle signals. www.landesbioscience.com

The COMA complex facilitates the Sli15-dependent sensing of tension between sister chromatids. The mechanism of how a cell recognizes and corrects syntelic attachments has garnered much attention as reviewed in ref. 26. One notion is that tension may be sensed by Sli15,13 and that this signal is then transmitted to the SAC via phosphorylation of Mad3 by Ipl1.9 It is likely other proteins are also required for the sensing of tension. For example, Shugoshin has also been shown to be required for tension sensing,27 possibly by loading Aurora B to the kinetochore in metaphase, as suggested by studies in fission yeast.28 The results presented in this study indicate that the kinetochore COMA complex is also required for Sli15 kinetochore localization and thus functions in correcting defective attachments. We propose that the COMA complex provides a primary attachment point for Sli15 in the central kinetochore. This particular localization lends itself to a model, first suggest by Tanaka and colleagues8 whereby tension between chromosomes creates a spatial gap that is sensed by INCENP/Aurora B. We extend the model here by placing Sli15 attached to the COMA complex rather than to chromatin, as has been previously suggested.8 In our model, Sli15 and Ipl1 are anchored at the inner kinetochore by the COMA complex. Since Ipl1 requires Sli15 to be active, and since tension as a result of bi-orientation likely results in the spatial separation of outer and inner kinetochore, bi-orientation would interrupt the interaction of Ipl1 with Sli15 and consequently lead to the inactivation of Ipl1. Conversely, syntelic attachments do not generate spindle forces that separate inner and outer kinetochore and therefore allow Sli15 and Ipl1 to interact the outer kinetochore. Ipl1 then phosphorylates outer kinetochore targets such as Dam1 leading to the disassociation of the MT from the kinetochore and to the activation of the SAC. A similar model involving the spatial regulation of Aurora B and kinetochore subunits has recently been proposed for mammalian cells,29 suggesting that this mechanism may be conserved. Finally, our model does not preclude a role for Bir1 in sensing tension. In fact Bir1 is important in sensing tension by promoting the localization of Ipl1 at the metaphase kinetochore, but not Sli15.32 Our data complements this finding by showing that the COMA complex is required for Sli15 localization to the kinetochore, but not Bir1. As well, our data further confirms that the presence of Bir1 at the kinetochore is not sufficient for Sli15 localization to the kinetochore, as Bir1 remains localized at the kinetochore in ame1-4 cells yet Sli15 is absent. All together, our findings indicate that the COMA complex is required in concert with other kinetochore components to present a scaffold for Sli15/INCENP and Ipl1/Aurora B, the key components in the establishment of bi-orientation in metaphase.

Materials and Methods Yeast strains, plasmids and media. Strains used in this study are listed in Table 1. Media and general yeast methods including α-factor arrests are as previously described.20 Epitope tagging was made directly for all genes at their endogenous loci, as described previously.33 Electrophoresis and immunoblotting. Standard SDS-PAGE using 10% gels was carried out as described previously.20 Mouse

Cell Cycle

2575


Table 1 Strains used in this study Strain

Genotype

Source

MATa ura3-52 lys2-801 ade2-101 his3∆200 leu2∆1 trp1∆63

P. Hieter

MATa ame1-4:TRP1

20

YJK301

MATa his3-11,15::LacI:GFP(pAFS144, thermostable):HIS3 CEN15(1.8 kb):LacO:URA3 ame1-4:TRP1 SPC29:CFP:kanMX6

This study

YJK302

MATa his3-11,15::LacI:GFP(pAFS144, thermostable):HIS3 CEN15(1.8 kb):LacO:URA3 okp1-5:TRP1 SPC29:CFP:kanMX6

This study

YJK303

MATa his3-11,15::LacI:GFP(pAFS144, thermostable):HIS3 CEN15(1.8 kb):LacO:URA3 SPC29:CFP:kanMX6 ipl1-321

This study

YJK304

MATa his3-11,15::LacI:GFP(pAFS144, thermostable):HIS3 CEN15(1.8 kb):LacO:URA3 SPC29:CFP:kanMX6 ipl1-321 ame1-4:TRP1

This study

YJK2839

MATa SLI15-TAP:HIS3

This study

YJK2841

MATa SLI15-TAP:HIS3 ame1-4:TRP1

This study

YPH499 YPH1676

YJK2843

MATa SLI15-TAP:HIS3 okp1-5:TRP1

This study

YJK305

MATa SLI15-VFP:kanMX6 SPC29:CFP:hphMX4

This study

YJK306

MATa SLI15-VFP:kanMX6 SPC29:CFP:hphMX4 ame1-4:TRP1

This study

YJK307

MATa SLI15-VFP:kanMX6 SPC29:CFP:hphMX4 okp1-5:TRP1

This study

YJK308

MATa BIR1-VFP:kanMX6 SPC29:CFP:hphMX4

This study

YJK309

MATa BIR1-VFP:kanMX6 SPC29:CFP:hphMX4 ame1-4:TRP1

This study

YJK310

MATa BIR1-VFP:kanMX6 SPC29:CFP:hphMX4 okp1-5:TRP1

This study

YJK317

MATa SLI15-VFP:kanMX6 SPC29:CFP:hphMX4 ame1-4:TRP1 PRS425-OKP1

This study

YJK318

MATa SLI15-VFP:kanMX6 SPC29:CFP:hphMX4 ame1-4:TRP1 PRS425

This study

YJK319

MATa his3-11,15::LacI:GFP(pAFS144, thermostable):HIS3 CEN15(1.8 kb):LacO:URA3 ame1-4:TRP1 SPC29:CFP:kanMX6 PRS425-OKP1

This study

YJK320

MATa his3-11,15::LacI:GFP(pAFS144, thermostable):HIS3 CEN15(1.8 kb):LacO:URA3 ame1-4:TRP1 SPC29:CFP:kanMX6 PRS425

This study

YJK317

MATa SLI15-VFP:kanMX6 SPC29:CFP:hphMX4 ame1-4:TRP1 PRS425-OKP1

This study

anti-actin (MP Biomedicals, Irvine, CA) and rabbit anti-TAP (Openbiosystems, Huntsville, AL) were used at 1:10,000. Horseradish peroxidase-conjugated secondary antibodies were used at 1:10,000 (GE Healthcare, Little Chalfont, Buckinghamshire, United Kingdom) and were detected using Supersignal Dura West chemiluminescence reagent (Pierce, Rockford, IL). Fluorescence microscopy. Strains used for microscopy were grown as described previously.20 Mean fluorescent intensity (MFI) was calculated as described previously,20 with the following modifications. WT, ame1-4 and okp1-5 cells containing Sli15-VFP and Spc29-CFP were arrested in 20 μg/ml nocodazole (NZ) and released into media at 37°C containing NZ and collected every hour for 3 hours. At each time point, the mean fluorescent intensity (background corrected) of VFP in a 3-dimensional volume of 225 voxels surrounding the SPB was measured for >100 cells and normalized against the mean fluorescent intensity (background corrected) of CFP. The ratio was then compared to WT levels and expressed as a percentage of WT MFI. The error bars represent the standard error of the mean for three independent experiments. A Student’s t-test was performed between the MFI of each mutant condition and WT (p < 0.05). Images used for MFI calculations were captured using a Nikon TE2000-U inverted microscope equipped with a 63x 1.4 NA objective mounted on a PE piezo z-drive/Improvision Orbit controller, and Hamamatsu ORCA-ERG camera. Image stacks (0.5 μm optical sections with 10–12 stacks per image) were acquired using Volocity 3.7. All measurements were made using 3D 2576

re-constructions with Volocity 3.7 Classification. Images shown are the extended focus of a z-series of 0.5 μm optical sections. Image processing was performed with Adobe Photoshop CS. Acknowledgements

The authors thank Susi Kaitna, Damien D’Amours, Vivien Measday and Craig Mandato for their comments on the manuscript, and members of the Vogel lab for numerous fruitful discussions. JV is supported by a New Investigator Award from the Canadian Institutes of Health Research (MSH 69117). This research was supported by an operating grant from the Natural Science and Engineering Research Council (262246-03) and infrastructure grants from the Canada Foundation for Innovation (CFI 7395). Note

Supplementary materials can be found at: www.landesbioscience.com/supplement/KnocklebyCC8-16Sup.pdf References 1. McAinsh AD, Tytell JD, Sorger PK. Structure, function, and regulation of budding yeast kinetochores. Annu Rev Cell Dev Biol 2003; 19:519-39. 2. Tanaka TU. Chromosome bi-orientation on the mitotic spindle. Philos Trans R Soc Lond B Biol Sci 2005; 360:581-9. 3. Tanaka K, Mukae N, Dewar H, van Breugel M, James EK, Prescott AR, et al. Molecular mechanisms of kinetochore capture by spindle microtubules. Nature 2005; 434:987-94. 4. Dewar H, Tanaka K, Nasmyth K, Tanaka TU. Tension between two kinetochores suffices for their bi-orientation on the mitotic spindle. Nature 2004; 428:93-7. 5. Nicklas RB. How cells get the right chromosomes. Science 1997; 275:632-7.

Cell Cycle

2009; Vol. 8 Issue 16

Sli15 kinetochore function requires COMA complex 6. Biggins S, Severin FF, Bhalla N, Sassoon I, Hyman AA, Murray AW. The conserved protein kinase Ipl1 regulates microtubule binding to kinetochores in budding yeast. Genes Dev 1999; 13:532-44. 7. He X, Rines DR, Espelin CW, Sorger PK. Molecular analysis of kinetochore-microtubule attachment in budding yeast. Cell 2001; 106:195-206. 8. Tanaka TU, Rachidi N, Janke C, Pereira G, Galova M, Schiebel E, et al. Evidence that the Ipl1-Sli15 (Aurora kinase-INCENP) complex promotes chromosome bi-orientation by altering kinetochore-spindle pole connections. Cell 2002; 108:317-29. 9. King EM, Rachidi N, Morrice N, Hardwick KG, Stark MJ. Ipl1p-dependent phosphorylation of Mad3p is required for the spindle checkpoint response to lack of tension at kinetochores. Genes Dev 2007; 21:1163-8. 10. Ciosk R, Zachariae W, Michaelis C, Shevchenko A, Mann M, Nasmyth K. An ESP1/ PDS1 complex regulates loss of sister chromatid cohesion at the metaphase to anaphase transition in yeast. Cell 1998; 93:1067-76. 11. Uhlmann F, Lottspeich F, Nasmyth K. Sister-chromatid separation at anaphase onset is promoted by cleavage of the cohesin subunit Scc1. Nature 1999; 400:37-42. 12. Pereira G, Schiebel E. Separase regulates INCENP-Aurora B anaphase spindle function through Cdc14. Science 2003; 302:2120-4. 13. Sandall S, Severin F, McLeod IX, Yates JR, 3rd, Oegema K, Hyman A, et al. A Bir1-Sli15 complex connects centromeres to microtubules and is required to sense kinetochore tension. Cell 2006; 127:1179-91. 14. Kang J, Cheeseman IM, Kallstrom G, Velmurugan S, Barnes G, Chan CS. Functional cooperation of Dam1, Ipl1, and the inner centromere protein (INCENP)-related protein Sli15 during chromosome segregation. J Cell Biol 2001; 155:763-74. 15. Shang C, Hazbun TR, Cheeseman IM, Aranda J, Fields S, Drubin DG, et al. Kinetochore protein interactions and their regulation by the Aurora kinase Ipl1p. Mol Biol Cell 2003; 14:3342-55. 16. De Wulf P, McAinsh AD, Sorger PK. Hierarchical assembly of the budding yeast kinetochore from multiple subcomplexes. Genes Dev 2003; 17:2902-21. 17. Hyland KM, Kingsbury J, Koshland D, Hieter P. Ctf19p: A novel kinetochore protein in Saccharomyces cerevisiae and a potential link between the kinetochore and mitotic spindle. J Cell Biol 1999; 145:15-28. 18. Ortiz J, Stemmann O, Rank S, Lechner J. A putative protein complex consisting of Ctf19, Mcm21, and Okp1 represents a missing link in the budding yeast kinetochore. Genes Dev 1999; 13:1140-55. 19. Poddar A, Roy N, Sinha P. MCM21 and MCM22, two novel genes of the yeast Saccharomyces cerevisiae are required for chromosome transmission. Mol Microbiol 1999; 31:349-60. 20. Pot I, Knockleby J, Aneliunas V, Nguyen T, Ah-Kye S, Liszt G, et al. Spindle checkpoint maintenance requires Ame1 and Okp1. Cell Cycle 2005; 4:1448-56. 21. Kim JH, Kang JS, Chan CS. Sli15 associates with the ipl1 protein kinase to promote proper chromosome segregation in Saccharomyces cerevisiae. J Cell Biol 1999; 145:1381-94. 22. Goshima G, Yanagida M. Establishing biorientation occurs with precocious separation of the sister kinetochores, but not the arms, in the early spindle of budding yeast. Cell 2000; 100:619-33. 23. Pinsky BA, Tatsutani SY, Collins KA, Biggins S. An Mtw1 complex promotes kinetochore biorientation that is monitored by the Ipl1/Aurora protein kinase. Dev Cell 2003; 5:735-45. 24. Cheeseman IM, Anderson S, Jwa M, Green EM, Kang J, Yates JR, 3rd, et al. Phosphoregulation of kinetochore-microtubule attachments by the Aurora kinase Ipl1p. Cell 2002; 111:163-72. 25. Joglekar AP, Bouck DC, Molk JN, Bloom KS, Salmon ED. Molecular architecture of a kinetochore-microtubule attachment site. Nat Cell Biol 2006; 8:581-5. 26. Pinsky BA, Biggins S. The spindle checkpoint: tension versus attachment. Trends Cell Biol 2005; 15:486-93. 27. Indjeian VB, Stern BM, Murray AW. The centromeric protein Sgo1 is required to sense lack of tension on mitotic chromosomes. Science 2005; 307:130-3. 28. Kawashima SA, Tsukahara T, Langegger M, Hauf S, Kitajima TS, Watanabe Y. Shugoshin enables tension-generating attachment of kinetochores by loading Aurora to centromeres. Genes Dev 2007; 21:420-35. 29. Liu D, Vader G, Vromans MJ, Lampson MA, Lens SM. Sensing chromosome biorientation by spatial separation of aurora B kinase from kinetochore substrates. Science 2009; 323:1350-3. 30. McCleland ML, Gardner RD, Kallio MJ, Daum JR, Gorbsky GJ, Burke DJ, et al. The highly conserved Ndc80 complex is required for kinetochore assembly, chromosome congression and spindle checkpoint activity. Genes Dev 2003; 17:101-14. 31. Gillett ES, Espelin CW, Sorger PK. Spindle checkpoint proteins and chromosomemicrotubule attachment in budding yeast. J Cell Biol 2004; 164:535-46. 32. Shimogawa MM, Widlund PO, Riffle M, Ess M, Davis TN. Bir1 is required for the tension checkpoint. Mol Biol Cell 2009; 20:915-23. 33. Longtine MS, McKenzie A, 3rd, Demarini DJ, Shah NG, Wach A, Brachat A, et al. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast 1998; 14:953-61.

www.landesbioscience.com

Cell Cycle

2577