Journal of Cell Science partnership with Dryad

Journal of Cell Science makes data accessibility easy with Dryad

Mph1 kinetochore localization is crucial and upstream in the hierarchy of spindle assembly checkpoint protein recruitment to kinetochores
Stephanie Heinrich, Hanna Windecker, Nicole Hustedt, Silke Hauf

Summary

The spindle assembly checkpoint (SAC) blocks entry into anaphase until all chromosomes have stably attached to the mitotic spindle through their kinetochores. The checkpoint signal originates from unattached kinetochores, where there is an enrichment of SAC proteins. Whether the enrichment of all SAC proteins is crucial for SAC signaling is unclear. Here, we provide evidence that, in fission yeast, recruitment of the kinase Mph1 is of vital importance for a stable SAC arrest. An Mph1 mutant that eliminates kinetochore enrichment abolishes SAC signaling, whereas forced recruitment of this mutant to kinetochores restores SAC signaling. In bub3Δ cells, the SAC is functional when only Mph1 and the Aurora kinase Ark1, but no other SAC proteins, are enriched at kinetochores. We analyzed the network of dependencies for SAC protein localization to kinetochores and identify a three-layered hierarchy with Ark1 and Mph1 on top, Bub1 and Bub3 in the middle, and Mad3 as well as the Mad1–Mad2 complex at the lower end of the hierarchy. If Mph1 is artificially recruited to kinetochores, Ark1 becomes dispensable for SAC activity. Our results highlight the crucial role of Mph1 at kinetochores and suggest that the Mad1–Mad2 complex does not necessarily need to be enriched at kinetochores for functional SAC signaling.

Introduction

During cell division, chromosomes are segregated to the two daughter cells by microtubules of the mitotic spindle (Walczak et al., 2010). The spindle assembly checkpoint (SAC; also called mitotic checkpoint) is a signaling pathway that surveys chromosome attachment to the spindle and prevents anaphase as long as any of the chromosomes remains unattached (Musacchio and Salmon, 2007). The conserved SAC signaling network comprises the proteins Mps1 (Mph1 in fission yeast), Mad1, Mad2, Mad3 (BubR1 in some organisms), Bub1, and Bub3 (Musacchio and Salmon, 2007). In addition, the Aurora B kinase (Ark1 in fission yeast) is required for a functional SAC (Kallio et al., 2002; Petersen and Hagan, 2003; Maldonado and Kapoor, 2011; Santaguida et al., 2011; Saurin et al., 2011). The SAC prevents anaphase by inhibiting the activity of the anaphase-promoting complex or cyclosome (APC/C) (Peters, 2006) through binding of Mad2 and Mad3 to the APC/C-activator Cdc20 (Slp1 in fission yeast). The checkpoint signal originates from unattached kinetochores (Rieder et al., 1995) and functional kinetochores are essential for the production of a SAC signal (Fraschini et al., 2001; Gardner et al., 2001; Nabetani et al., 2001; McCleland et al., 2003; Meraldi et al., 2004). Most SAC proteins are enriched at unattached kinetochores (Musacchio and Hardwick, 2002; Burke and Stukenberg, 2008), but how they are recruited and whether the enrichment of all SAC proteins at unattached kinetochores is required for SAC signaling is largely unclear.

The dependencies among SAC proteins for their recruitment to kinetochores may reveal which proteins respond directly to the attachment state of the kinetochore. However, the available information is incomplete and often controversial, possibly as a result of varying degrees of protein depletion or inhibition in different experiments (Meraldi et al., 2004). We analyzed the network of kinetochore recruitment dependencies in the genetically amenable unicellular eukaryote Schizosaccharomyces pombe, where SAC genes can be entirely deleted to study the functional consequences. We find a hierarchical organization with the Aurora kinase Ark1 and the kinase Mph1 on top, and Mad2 at the bottom. Kinetochore enrichment of Mph1 is vital for SAC signaling and for kinetochore enrichment of other SAC proteins. Recruiting Mph1 to kinetochores seems the only crucial function of Ark1 in SAC signaling. In bub3-deleted fission yeast cells, SAC signaling is functional with Ark1 and Mph1 being the only SAC proteins that are visibly enriched at kinetochores. This demonstrates the central importance of Aurora and Mph1 as upstream factors at the kinetochore and suggests that activation of Mad2, which is ultimately required for inhibition of the APC/C, can happen away from kinetochores.

Results and Discussion

N-terminal truncation of Mph1 abolishes kinetochore enrichment and SAC signaling

It has been proposed that kinetochore localization of the SAC kinase Mps1 is essential for SAC signaling in vertebrates (Liu et al., 2003; Zhao and Chen, 2006; Hached et al., 2011), but more recent results have challenged this view (Maciejowski et al., 2010). To assess the importance of fission yeast Mph1 localization to unattached kinetochores, we truncated the Mph1 protein N-terminally, leaving the kinase domain intact (Fig. 1A). Expression of the truncated versions from the endogenous locus resulted in protein abundance similar to wild type Mph1 (Fig. 1B). The shorter truncation (Mph1-Δ1-150) maintained kinetochore localization and SAC signaling, whereas the longer truncation (Mph1-Δ1-302) abolished both kinetochore localization and SAC signaling (Fig. 1C,D), suggesting that kinetochore localization is crucial for SAC activity.

Fig. 1.

Mph1 localization to kinetochores is required for SAC activity. (A) Schematic of the N-terminal Mph1 truncations. (B) Extracts from asynchronous cultures of the indicated strains were analyzed by immunoblotting. Cdc2 serves as a loading control. (C) Cells expressing nda3-KM311, plo1+–mCherry and the indicated mph1 alleles were followed by live-cell imaging at the restrictive temperature for nda3-KM311 (Hiraoka et al., 1984), which prevents microtubule formation. Plo1 localization at spindle pole bodies (SPBs) was used as a marker for mitotic cells (Mulvihill et al., 1999). Representative cells and the percentage of mitotic cells with localized Mph1 signal are shown (n>30 cells). (D) Cells were followed by live-cell imaging at the restrictive temperature for nda3-KM311. The duration of prometaphase was determined by the presence of Plo1–mCherry at SPBs. Circles indicate cells in which the entire mitosis was recorded, triangles indicate cells in which entry into mitosis but not exit from mitosis was recorded, filled circles indicate cells that died in mitosis. (E) Expression of the indicated Mph1 constructs from the thiamine-regulatable nmt81 promoter (Basi et al., 1993) was induced by the depletion of thiamine. Mitotic cells were identified by the presence of Plo1–mCherry at SPBs and classified according to spindle length: <2 µm, prometaphase; between 2 and 2.5 µm, metaphase; >2.5 µm, anaphase; n>200 cells. (F) Serial dilutions of the strains used in E were grown on rich medium (with thiamine, nmt81 promoter repressed) or minimal medium lacking thiamine (without thiamine, nmt81 promoter induced). (G) Representative cells from the experiment shown in E. GFP signals were recorded under identical conditions. (H) The indicated mis12–mph1 fusions were weakly expressed from the thiamine-repressible nmt81 promoter and cells were followed by live-cell imaging at the restrictive temperature for nda3-KM311. The duration of prometaphase was determined by the presence of Plo1–mCherry at SPBs. Symbols are as in D.

To determine whether the SAC defect resulted from impaired kinetochore localization or whether the Mph1-Δ1-302 protein was generally dysfunctional, we artificially recruited the truncated protein to kinetochores by fusion to the kinetochore protein Mis12. Forced recruitment of wild-type Mph1 to kinetochores lead to a pronounced delay in mitosis and a growth defect (Fig. 1E,F), as has been seen before (Ito et al., 2012). The phenotypes were rescued by deletion of mad2, which indicates that forced recruitment of Mph1 artificially promoted SAC signaling and that the fusion to Mis12 did not impair kinetochore function. Forced recruitment of Mph1-Δ1-302 mimicked those of wild-type Mph1 (Fig. 1E,F). The slightly weaker phenotype when recruiting Mph1-Δ1-302 can be explained by the reduced presence of this protein at kinetochores compared to wild-type Mph1 (Fig. 1G). When both fusion proteins were expressed at very low levels, they were both able to support a mitotic delay in response to microtubule depolymerisation (Fig. 1H). These data indicate that Mph1-Δ1-302 retains the ability for SAC signaling and that the SAC failure in mph1-Δ1-302 cells is a consequence of eliminating kinetochore enrichment of Mph1. Hence, kinetochore enrichment of Mph1 is a crucial prerequisite for SAC signaling in fission yeast.

Enrichment of Mph1 at unattached kinetochores in bub3Δ cells is necessary for SAC signaling

Others and we previously reported that the SAC is functional in bub3-deleted fission yeast cells despite the failure of these cells to enrich Mad1, Mad2, Mad3, and Bub1 at unattached kinetochores (Millband and Hardwick, 2002; Vanoosthuyse et al., 2009; Windecker et al., 2009). Since the kinetochore localization of Mph1 seems obligatory for SAC signaling (Fig. 1), we analyzed whether the enrichment of this SAC protein at kinetochores is preserved in bub3Δ cells. Indeed, Mph1 localizes to unattached kinetochores in bub3Δ cells (Fig. 2A). The intensity of Mph1 at unattached kinetochores (Fig. 2A) and the fraction of cells, in which a signal could be detected (supplementary material Fig. S1), were similar between bub3+ and bub3Δ cells, as was the total abundance of Mph1 (supplementary material Fig. S1). Deletion of mph1 abolished the SAC-mediated delay in bub3Δ cells (Fig. 2B), confirming that Mph1 is needed in bub3Δ cells to generate the SAC signal. To test whether localization of Mph1 to kinetochores is required, we combined deletion of bub3 with the Mph1 N-terminal truncations. In the presence of Mph1-Δ1-150, the SAC was still functional in bub3Δ cells, although the mitotic delay was shorter than in mph1-Δ1-150 or bub3Δ cells (Fig. 2C). In the presence of Mph1-Δ1-302, the SAC response in bub3Δ cells was abrogated (Fig. 2C), demonstrating that recruitment of Mph1 to kinetochores is necessary for SAC function in bub3Δ cells. This raises the possibility that activation of Mph1 at unattached kinetochores in bub3Δ cells is sufficient to transmit the SAC signal to the nucleoplasm and challenges the idea that the enrichment of the Mad1–Mad2 complex at kinetochores is crucial (Kulukian et al., 2009).

Fig. 2.

Mph1 localization to kinetochores is required for SAC activity in bub3Δ cells. (A) Cells expressing the indicated markers were grown at the restrictive temperature for nda3-KM311 and fixed with methanol. Insets are 1.4-times magnified relative to the main picture. DIC, differential interference contrast. Mph1–GFP signal intensity was quantified (n>48 cells; box plot with whiskers from 5th to 95th percentile; data normalized to median of bub3+ cells). (B,C) The indicated strains were followed by live-cell imaging at the restrictive temperature for nda3-KM311. The duration of prometaphase was determined by the presence of Plo1–GFP at SPBs. Circles indicate cells in which the entire mitosis was recorded, triangles indicate cells in which entry into mitosis but not exit from mitosis was recorded. (D) Cells were followed by live-cell imaging with Sid4–mCherry as an SPB marker. Representative kymographs of the spindle region throughout mitosis are shown. The total signal intensity of Ark1–GFP on the spindle axis was determined for several time points in early mitosis (n>17 cells; box plot and normalization as in A).

A localization feedback loop between Ark1 and Mph1

In addition to preserving the localization of Mph1 (Fig. 2A), bub3Δ cells largely preserve localization of the Aurora kinase Ark1 (Fig. 2D; supplementary material Fig. S2). Hence, Ark1 and Mph1 are the only SAC proteins that still display enrichment at the centromere/kinetochore region in bub3Δ cells. Whereas in budding yeast Aurora (Ipl1) and Mps1 seem to localize independently (Maure et al., 2007), metazoan Mps1 localization to kinetochores has been shown to depend on Aurora B kinase activity (Vigneron et al., 2004; Santaguida et al., 2010; Saurin et al., 2011). We find that kinetochore localization of fission yeast Mph1 crucially depends on Ark1 kinase activity (Fig. 3A,B). When proper chromosome attachment was prevented by a conditional mutation in kinesin-5 (cut7-446), Mph1 localized to kinetochores, but the enrichment was abrogated by chemical genetic inhibition of Ark1 with the small molecule 1NM-PP1 (Fig. 3A). When we inhibited Ark1 in mitotic cells lacking microtubules, the localized Mph1 signals largely disappeared within 5 minutes of Ark1 inhibition (Fig. 3B), although the mitosis-specific slower migration of Mph1–GFP in SDS-PAGE was preserved and cells maintained high CDK1 activity, as indicated by the continued presence of Plo1 at SPBs (Fig. 3B) (Dischinger et al., 2008). The loss of Mph1 from kinetochores cannot be attributed to side effects of the inhibitor (supplementary material Figs S1, S3). Together this suggests that Ark1 is directly and continuously required to maintain Mph1 localization to kinetochores. When Mph1 was artificially recruited to the kinetochore, inhibition of Ark1 did not shorten the mitotic delay in the absence of microtubules (Fig. 3C; supplementary material Fig. S4). In agreement with data from human cells (Jelluma et al., 2010), this suggests that localization of Mph1 may be the only crucial function of Ark1 in the SAC.

Fig. 3.

Interdependent kinetochore localization of Mph1 and Ark1. (A) Cells were arrested in late G2 using the conditional cdc25-22 mutation (Moreno et al., 1989) and released into mitosis in the presence of 5 µM of the Ark1-as3 inhibitor 1NM-PP1 (Hauf et al., 2007) or an equivalent amount of the solvent DMSO. The cut7-446 allele leads to monopolar spindles (Hagan and Yanagida, 1990). Mitotic cells were fixed with methanol, and DNA was stained with DAPI. Insets are 2.5-times magnified relative to the main picture. The presence of a localized Plo1 signal indicates that cells are in mitosis. The percentage of cells with Plo1 and Mph1 signal was determined (n>100 cells). (B) Cells were grown at the restrictive temperature for nda3-KM311, treated with 1NM-PP1 and fixed with methanol after 0 min and 5 min. DNA was stained with DAPI. Insets are 2.5-times magnified relative to the main picture. Maximal Mph1–GFP signal intensity in mitotic cells was quantified (n>170 cells, box plot with whiskers from 5th to 95th percentile; data normalized to median of cells treated for 0 min). Protein abundance of Mph1–GFP in cycling cells (cyc) and mitotically arrested cells treated with (+) or without (−) 1NM-PP1 was determined by immunoblotting using anti-GFP and anti-Cdc2 (loading control) antibodies. (C) Cells expressing ark1-as3 and mph1+ or the mis12-mph1-Δ1-302 fusion construct were followed by live-cell imaging at the restrictive temperature for nda3-KM311 in the presence of 10 µM of the Ark1-as3 inhibitor 1NM-PP1 or an equivalent amount of the solvent DMSO. The duration of prometaphase was determined by the presence of Plo1–mCherry at SPBs. Symbols are as in Fig. 2B. (D) Cells were followed by live-cell imaging with Sid4–mCherry as the SPB marker. Representative kymographs of the spindle region are shown. The total signal intensity of Ark1–GFP on the spindle axis was determined for several time points in early mitosis (n>12 cells; box plot as in B; data normalized to median of mph1+ cells). (E) Cells were followed by live-cell imaging at the restrictive temperature for nda3-KM311. Representative cells in early mitosis are shown. The Ark1–GFP signal was measured over time as cells entered mitosis (n>23 cells; ±s.d.).

In contrast to the pronounced effect of Ark1 on Mph1 localization, Mph1 is only partially and largely indirectly required for localizing Ark1 (Fig. 3D,E; supplementary material Fig. S2). Ark1 concentration at centromeres is slightly reduced by deletion of mph1 both in an otherwise unperturbed mitosis (Fig. 3D) and when microtubule formation was prevented (Fig. 3E; supplementary material Fig. S2). Mph1 is required for the kinetochore localization of Bub1 (Vanoosthuyse et al., 2004) (supplementary material Fig. S5) and Bub1 is known to aid efficient recruitment of Ark1 to centromeres through phosphorylation of histone H2A-S121 (Kawashima et al., 2007; Kawashima et al., 2010; Tsukahara et al., 2010). This suggests that Mph1 is indirectly required for Ark1 localization through promoting the kinetochore enrichment of Bub1. Indeed, deletion of bub3, which also abrogates the kinetochore enrichment of Bub1 (Vanoosthuyse et al., 2004), leads to a similar defect in Ark1 localization as deletion of mph1 (supplementary material Fig. S2). No further enhancement is seen in the mph1Δ bub3Δ double deletion (supplementary material Fig. S2), indicating that Mph1 and Bub3 act in the same pathway. Deletion of bub1 impairs Ark1 localization more strongly than deletion of mph1 or bub3 (supplementary material Fig. S2), suggesting that Bub1 partially promotes Ark1 localization even when delocalized. Additional deletion of mph1 in bub1Δ cells further diminishes Ark1 centromere localization slightly, suggesting that Mph1 may also promote Ark1 localization independently of Bub1, but this effect is weak (supplementary material Fig. S2). Taken together, these results suggest that Ark1 is required for the kinetochore localization of Mph1, and recruitment of Mph1 reinforces localization of Ark1 to centromeres, presumably through localization of Bub1 and enhanced centromeric H2A-S121 phosphorylation. In bub3Δ cells, this feedback loop is broken, but some Ark1 still localizes to centromeres and is sufficient to recruit Mph1 to unattached kinetochores (Fig. 2).

The network of SAC protein localization dependencies

The preserved localization of Ark1 and Mph1 in bub3Δ cells indicates that these two proteins can be recruited independently of all other SAC proteins and are on top of the kinetochore localization hierarchy. Indeed, Ark1 and Mph1 are fully or partially required for the kinetochore enrichment of all other SAC proteins (Fig. 4A; supplementary material Figs S5–S10). For Mph1, its kinetochore localization is important, because recruitment of Bub1, Mad1 (Fig. 4B,C) and Ark1 (supplementary material Fig. S10) is equally defective upon deletion of mph1 or expression of mph1-Δ1-302.

Fig. 4.

The hierarchy of SAC protein localization dependencies. (A) Schematic of SAC protein localization dependencies. Black arrows indicate a dependency, dashed arrows a partial dependency and gray arrows the absence of dependency. Dependencies were determined in this study (Figs 2, 3; supplementary material Figs S2, S5–S9), except for those indicated by symbols (‡, Windecker et al., 2009; §, Vanoosthuyse et al., 2004; #, Kadura et al., 2005; *, Millband and Hardwick, 2002). n.d., not determined. (B,C) The indicated strains were followed by live-cell imaging at the restrictive temperature for nda3-KM311. The GFP signal was measured over time as cells entered mitosis (n>7 cells; ±s.d.). Representative nuclei in early mitosis are shown. (DF) General scheme for SAC protein localization dependencies at unattached kinetochores (D), kinetochores not under tension (E) or kinetochores of bub3Δ cells (F). See text for details. P, phosphorylation, O-Mad2, Mad2 in the ‘open’ conformation; C-Mad2, Mad2 in the ‘closed’ conformation (Mapelli and Musacchio, 2007).

We systematically analyzed the kinetochore localization dependencies for all other SAC proteins and find that Bub1 and Bub3, which depend on each other for their kinetochore enrichment, form the second layer of the hierarchy (Fig. 4A). Deletion of bub1 or bub3 abolishes kinetochore enrichment of Mad1, Mad2, and Mad3 (supplementary material Figs S7, S8) (Millband and Hardwick, 2002; Windecker et al., 2009), but deletion of bub3 leaves localization of Mph1 and Ark1 largely intact (Fig. 2; supplementary material Figs S1, S2). In the third layer of the hierarchy, Mad3 has a very slight effect on the localization of Mad1 and Mad2 (supplementary material Figs S7, S8), whereas neither deletion of mad1 nor mad2 affects the localization of Mad3 (supplementary material Fig. S9) (Millband and Hardwick, 2002). Kinetochore enrichment of Mad2 depends on Mad1, but not vice versa (supplementary material Figs S7, S8), putting Mad2 lowest in the kinetochore localization hierarchy.

Although the data on SAC protein kinetochore localization dependencies from other organisms are fragmentary and sometimes contradictory (supplementary material Fig. S11), the principal hierarchy seems conserved. Vertebrate Aurora B is required for efficient localization of Mps1 (Vigneron et al., 2004; Santaguida et al., 2010; Saurin et al., 2011), and Aurora B and Mps1 are required for the localization of other SAC proteins. In all organisms examined, Bub1 is upstream of Mad1 and Mad2 (Sharp-Baker and Chen, 2001; Gillett et al., 2004; Meraldi et al., 2004; Essex et al., 2009). Most variable in the hierarchy are Mad3 and BubR1, which coincides with the stronger evolutionary divergence compared to other SAC proteins (Musacchio and Salmon, 2007).

Our experiments and published results now suggest a general model for SAC protein recruitment to the kinetochore (Fig. 4D–F). Ark1 (Aurora B), whose localization to centromeres is largely independent of other SAC proteins, is the most upstream component. Aurora B phosphorylates kinetochore proteins dependent on the chromosome attachment state (Liu et al., 2009), which may provide a binding platform for the Mph1 (Mps1) kinase. Mph1 (Mps1) can then recruit additional SAC proteins through phosphorylation of kinetochore components (Kemmler et al., 2009; London et al., 2012; Shepperd et al., 2012; Yamagishi et al., 2012) or SAC proteins (Hardwick et al., 1996). Recruitment of Bub1 and Bub3, probably to the outer kinetochore protein Spc7 (KNL-1, Blinkin) (Kiyomitsu et al., 2007; Bolanos-Garcia et al., 2011; Kiyomitsu et al., 2011; Krenn et al., 2012), may initiate a feedback loop with Ark1 (Aurora B) to strengthen the interactions. Bub1 and Bub3 may provide a direct physical recruitment platform for Mad1 at the kinetochore. In addition, recruitment of Mad1 may require CDK1 activity (Vázquez-Novelle and Petronczki, 2010; Ito et al., 2012) or conformational changes at the kinetochore that signal the absence of microtubule attachment (Garcia et al., 2002). Mad1 co-recruits Mad2 and the Mad1–Mad2 complex activates additional Mad2 molecules, which then inhibit Cdc20 (De Antoni et al., 2005) (Fig. 4D).

It remains controversial whether kinetochores that contact microtubules but do not come under tension create a SAC signal (Nezi and Musacchio, 2009; Maresca and Salmon, 2010). We note that the upper two layers of the SAC protein localization hierarchy are recruited to ‘tension-less’ kinetochores (Skoufias et al., 2001; Garcia et al., 2002; Gillett et al., 2004; Howell et al., 2004) (Fig. 4E). In contrast, Mad1 and Mad2 are only enriched at kinetochores lacking attachment (Skoufias et al., 2001; Garcia et al., 2002; Gillett et al., 2004) (Fig. 4D). Because the SAC is functional in bub3-deleted fission yeast cells, although Mad1 and Mad2 are undetectable at kinetochores, Mph1 may in principle have the capacity to activate the Mad1–Mad2 complex even if the latter is away from kinetochores (Fig. 4F). Whether this is also possible in wild-type cells with kinetochores that are attached but not under tension is an open question (Fig. 4E). Additional recruitment of Mad1–Mad2 may additionally boost SAC signaling at unattached kinetochores (Fig. 4D). Mph1 localization to kinetochores is central for all aspects of SAC signaling at kinetochores, and it will be interesting to determine how Mph1 is recruited.

Materials and Methods

S. pombe strains

Strains are listed in supplementary material Table S1. GFP tagging was performed by PCR-based gene targeting (Bähler et al., 1998). To generate Mph1 truncation mutants, the bases corresponding to amino acids 2 to 150 (SKRN…NKTP) and 2 to 302 (SKRN…TPIP) were deleted by PCR and the modified mph1 genes were integrated into the endogenous locus by replacing ura4+ in an mph1Δ::ura4+ strain. For fusion of Mph1 to Mis12, the hygR << Pnmt81 and mis12+-(GGSG)2 fragments were connected by PCR and integrated into the genomic locus of the mph1-S(GGGGS)3-GFP << kanR strain.

Culture conditions

Strains harbouring the cdc25-22 cut7-446 alleles were grown in EMM (Moreno et al., 1991) at 25°C until log phase. Cells were shifted to 36°C for 4.5 hours. DMSO or 5 µM 1NM-PP1 {(4-amino-1-tertbutyl-3-(1′-naphthylmethyl)pyrazolo[3,4-d]pyrimidine; Toronto Research Chemicals} was added and cells were incubated for another 30 minutes at 36°C. Cells were released by shifting to 25°C and were harvested after 10 min. Strains with the nda3-KM311 mutation were grown at 30°C in YE (Moreno et al., 1991) supplemented with Adenine, and shifted to 16°C for 6 hours.

Microscopy

Live-cell imaging was performed on a DeltaVision microscope (Applied Precision) as previously described (Windecker et al., 2009). Imaging of fixed cells was performed on a Zeiss AxioImager microscope with a charged-coupled device camera and MetaMorph software (Molecular Devices Corporation). Typically, a Z-stack of 3 µm thickness with single planes spaced by 0.3 µm was acquired and subsequently projected to a single image.

Immunoblotting

Synchronization of cells (Windecker et al., 2009) and protein extraction (Koch et al., 2012) were performed as previously described. Mouse anti-GFP (Roche, 11814460001), mouse anti-Cdc13 (GeneTex/Acris, GTX10873) or rabbit anti-Cdc2 (Santa Cruz, SC-53) were used as primary antibodies.

Acknowledgments

We thank Julia Binder, Eva-Maria Illgen, Maria Langegger and Eva-Maria Schwoerzer for excellent technical support, Julia Kamenz, André Koch, and Yoshinori Watanabe for comments on the manuscript, and Yoshinori Watanabe for communicating unpublished results. We are grateful to Mitsuhiro Yanagida, Masayuki Yamamoto and the Yeast Genetic Resource Center (YGRC) for yeast strains.

Footnotes

  • Accepted June 7, 2012.

References

View Abstract