| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
,1
* Biology Graduate Group, University of Pennsylvania, Philadelphia, Pennsylvania 19104
Department of Genetics, University of Pennsylvania Medical School, Philadelphia, Pennsylvania 19104
Department of Molecular Biology, Princeton University, Princeton, New Jersey 08544
1 Corresponding author: Department of Molecular Biology, Lewis Thomas Lab, Princeton University, Princeton, NJ 08544.
E-mail: pmelloy{at}molbio.princeton.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
strain revealed a microtubule dependency of Cdc23-GFP spot localization, whether or not the checkpoint response was activated. On the basis of these data, we conclude that Cdc23p localization was dependent on microtubules and was affected by specific types of kinetochore disruption.
APC/C activity is regulated by two specificity factors, Cdc20p and Hct1p/Cdh1p (HARPER et al. 2002; IRNIGER 2002). In most organisms, including S. cerevisiae and humans, Cdc20p interacts with the APC/C to ubiquitinate the metaphase-anaphase inhibitor Pds1p, allowing sister chromatid separation and passage into anaphase, while Hct1p regulates APC/C activity during mitotic exit and G1 (COHEN-FIX et al. 1996; YAMAMOTO et al. 1996a,b; SCHWAB et al. 1997; VISINTIN et al. 1997; SHIRAYAMA et al. 1998, 1999). In S. cerevisiae, Cdc20p-dependent but not Hct1p-dependent APC/C activity is impaired upon mutation of Cdc28p consensus phosphorylation sites in Cdc23p and two other major APC/C subunits (RUDNER and MURRAY 2000).
The spindle checkpoint, a spindle damage surveillance system, blocks cell cycle progression by inhibiting the APC/C (MUSACCHIO and HARDWICK 2002; YU 2002; CLEVELAND et al. 2003). Spindle checkpoint components include the kinase Mps1p, and Mad1-3p and Bub1-3p, first discovered in budding yeast as mutants that failed to respond to microtubule-depolymerizing drugs (HOYT et al. 1991; LI and MURRAY 1991; WEISS and WINEY 1996). Cdc20p is known to interact with the spindle checkpoint pathway members Mad1p, Mad2p, Mad3p, and Bub3p (HWANG et al. 1998; HARDWICK et al. 2000). In budding yeast, the APC/C and Mad2p binding sites of Cdc20p are overlapping, suggesting a mutually exclusive relationship (ZHANG and LEES 2001). On the basis of extensive genetic and biochemical analysis, spindle checkpoint inhibition of the APC/C is mediated by Cdc20p.
In contrast, little is known about the spatial mechanism of APC/C inhibition by the spindle checkpoint pathway. In yeast and humans, several members of the spindle checkpoint pathway are at least transiently localized to the kinetochore (LEW and BURKE 2003; MCAINSH et al. 2003). In humans, these proteins include Cdc20, hBub1, hBubR1, Mad2, and Mad1 (LI and BENEZRA 1996; TAYLOR and MCKEON 1997; JABLONSKI et al. 1998; KALLIO et al. 1998). Recent studies in yeast have found similar results with Mps1p, Bub1p, Bub3p, Mad1p, and Mad2p (CASTILLO et al. 2002; IOUK et al. 2002; KERSCHER et al. 2003; KITAGAWA et al. 2003; GILLETT et al. 2004). We hypothesize that the APC/C has a dynamic localization during mitosis and the rest of the cell cycle. After observing APC/C localization in budding yeast during a normal cell cycle, we can then determine if its localization changes in response to spindle checkpoint activation or other perturbations.
Limited information about APC/C localization is available from studies performed in other organisms. Subunits of the mammalian APC/C have been localized to the kinetochore, centrosome, and mitotic spindle (TUGENDREICH et al. 1995; JORGENSEN et al. 1998; KURASAWA and TODOKORO 1999; TOPPER et al. 2002). Drosophila APC/C is found on the mitotic spindle, and the green fluorescent protein (GFP)-tagged APC/C subunit Cdc27p has been localized to mitotic chromosomes (HUANG and RAFF 1999, 2002). In S. cerevisiae, where immunolocalization is more difficult to resolve, the APC/C subunit Cdc23p has thus far been described as primarily nuclear (SIKORSKI et al. 1993; JACQUENOUD et al. 2002; HUH et al. 2003). On the basis of the localization of known APC/C substrates to the mitotic apparatus (such as Kip1p or Ase1p), one might predict that the S. cerevisiae APC/C would also be found on the mitotic spindle, the spindle poles, and/or the kinetochores (JUANG et al. 1997; GORDON and ROOF 2001).
In this work, we describe cell-cycle-dependent changes in the localization of the APC/C subunit Cdc23p. In addition, we focus on the relationship between the spindle checkpoint status and APC/C localization. Our comprehensive analysis not only confirms a similar localization pattern for the APC/C in S. cerevisiae and mammalian cells, but also extends these studies by identifying factors, such as microtubule depolymerization and defective kinetochore structure, that can affect this localization.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
|
|
Imaging and analysis:
To optimize the GFP signal, in particular the intensity of the nuclear signal, samples were taken from log-phase, well-aerated liquid YPD cultures (spun down and resuspended in minimal SD media supplemented with necessary amino acids for observations). Cultures were grown at 23° unless indicated otherwise. Supplemental adenine was added to cultures of ade2– strains.
Unfixed and DAPI-stained cells were observed on a microscope (Eclipse E800, Nikon, Tokyo) connected to a cooled, high-resolution charge-coupled device (CCD) camera (ORCA model C4742-95, Hammatsu Photogenics, Bridgewater, NJ). Images were analyzed using Image-Pro (Phase 3) Imaging software (Media Cybernetics, Silver Spring, MD). Differential interference contrast (DIC) images (exposure time, 0.6 sec), DAPI images (exposure time, 2 sec), and GFP images (exposure time, 5 sec) were taken as indicated. Contrast was enhanced using this Phase 3 Imaging program and Adobe Photoshop 4.0 (Adobe Systems, San Jose, CA).
APC/C localization in various mutant strains, as well as CFP/YFP colocalization studies, was conducted using a Leica DMRA2 microscope connected to a cooled, high-resolution CCD camera (ORCA extended range camera, Hammatsu Photogenics). Images were analyzed using OpenLab version 3.0.2 (Improvision). Filter sets for this microscope included cyan GFP v2 (D436/20 D480/40 455DCLP), yellow GFP BP (10C/Topaz—HQ500/20 HQ535/30 Q515LP), and a GFP filter set (E450/50D 480BP 510/50). On this microscope, exposure time for Cdc23-GFP was 4 sec, with 1x binning, while exposure time for Cdc23-YFP was 1.5 sec with 2x binning. Okp1-CFP and Slk19-CFP images were taken at 1 and 0.6 sec, respectively, with 2x binning and 20% gain. Spc42-CFP settings were the same as those for Slk19-CFP. Cdc20-GFP settings were the same as for Cdc23-YFP. Additional images were also collected using the DeltaVision Microscopy System.
For colocalization studies using CFP-tagged cellular markers and Cdc23-YFP, cells were put into three categories. For complete colocalization, the spindle pole body or kinetochore marker signal/signals had to overlap with the Cdc23p spot signal/signals. In the case of partial colocalization, this usually occurred when one spot signal was visible for one marker, but two were present for the second marker. For example, in some cases, two kinetochore-CFP spots were visible, but only one Cdc23-YFP spot was seen. Therefore, the overlap of the single Cdc23p spot with one of the kinetochore markers was defined as partial colocalization. Finally, no overlap of the spindle pole body or kinetochore signal with the Cdc23-YFP nuclear spot signal was defined as "no colocalization."
Indirect immunofluorescence was conducted to visualize microtubules in the okp1-5 mutant. We followed a standard protocol for this analysis, using the YOL1/34 rat anti-
-tubulin primary antibody and a FITC anti-rat secondary antibody (gift of S. Clark; PRINGLE et al. 1991).
Cell synchronization and treatment of temperature-sensitive strains:
For nocodazole synchronization, asynchronous cultures were treated with 15 µg/ml of the drug (M1404, Sigma, St. Louis) for 4 hr at 23°. At this concentration, typically 80% of the cells were arrested at metaphase. For temperature-sensitive strains, cells were grown overnight at 23° and then shifted to 37° for 4 hr. Cells were determined to be arrested when the percentage of cells at the arrest point matched previously published reports for that strain. Samples from representative experiments were fixed in 100% ethanol at this time to analyze the nuclear position by DAPI staining (1 µg/ml) to verify the arrest (data not shown).
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
As shown in Figure 1, Cdc23-GFP was localized to the nucleus and one or two distinct foci within the nucleus (Figure 1A, I, II, and III). A punctate signal was also visible along the length of the mitotic spindle in anaphase (Figure 1A, III). No signal was detected in the "no GFP" control (data not shown). The localization patterns of Cdc23-GFP and Apc9-GFP were very similar (data not shown). However, the Apc9-GFP signal was much weaker than that of Cdc23-GFP and therefore more difficult to analyze. Therefore, Cdc23-GFP was used as the APC-GFP tag for the remainder of the experiments.
|
At each cell cycle stage, the percentage of cells with a visible nuclear spot signal varied, with the highest percentage showing nuclear spots in metaphase (II) and the lowest in anaphase (III) (Figure 1A). This Cdc23-GFP localization pattern was observed in the S288C and W303 strain backgrounds as well, and this localization pattern was not background dependent (data not shown). Therefore, with the exception of the appearance of Cdc23-GFP onto the anaphase spindle, Cdc23-GFP localization did not change dramatically during the cell cycle.
The Cdc28 cyclin-dependent kinase phosphorylates Cdc23p and at least two other APC/C subunits during mitosis (RUDNER and MURRAY 2000). We looked at the localization pattern of Cdc23p in a strain with all Cdc28p consensus sites mutated in the APC/C subunits Cdc23p, Cdc27p, and Cdc16p (RUDNER and MURRAY 2000). This particular mutant was impaired in the Cdc20-dependent phase of APC/C activity, a defect potentially linked to a change in APC/C localization. As indicated in Figure 1B, cdc23A-YFP expression was not dependent on Cdc28p-mediated phosphorylation, since its pattern did not change in an APC/C phosphorylation mutant (n = 115).
Colocalization of Cdc23-YFP with spindle and kinetochore markers, but not a spindle pole body marker:
To narrow down the exact sites of Cdc23-GFP localization, we conducted colocalization studies using Cdc23-YFP and CFP-tagged spindle, spindle pole body, or kinetochore markers. In addition to colocalizing Cdc23-YFP with a spindle marker, we also analyzed the punctate linear Cdc23p pattern observed in anaphase cells in two other ways: comparing to a nuclear envelope GFP tag and using Cdc23-GFP in a cdc15-2 mutant (Figure 2). To rule out the possibility that the nuclear envelope was responsible for the signal, we compared the Cdc23-GFP signal to that of a nuclear envelope marker, ss-GFP-HDEL (GAMMIE and ROSE 2002). The Cdc23-GFP signal pattern clearly differed from that of the nuclear envelope marker pattern in the width and uniformity of the signal (Figure 2A). Colocalization studies (Figure 2B; Table 3) performed using CFP-Tub1 and Cdc23-YFP confirmed an overlapping signal. CFP-Tub1 colocalized with 90% of Cdc23-YFP cells displaying a spindle-like signal (Table 3). However, many cells in anaphase that showed the typical CFP-Tub1 signal did not display a Cdc23-YFP spindle-like signal. This could have been due to the general weakness of the Cdc23-YFP relative to Cdc23-GFP and/or to the transient nature of the Cdc23-YFP spindle localization. Finally, a cdc15-2 mutant strain containing Cdc23-GFP was used to enrich for late anaphase cells, a population expected to display a spindle localization of Cdc23-GFP (Figure 2C). This mutant arrested in late anaphase/telophase since the Cdc15p kinase was shown to be necessary for exit from mitosis (SCHWEITZER and PHILIPPSEN 1991; SURANA et al. 1993; JASPERSEN et al. 1998). As expected on the basis of earlier results, Cdc23-GFP was found along the anaphase spindle in the cdc15-2 mutant at nonpermissive temperature (Figure 2C, arrowheads). Therefore, by colocalization analysis and a comparison to a nuclear envelope fluorescent marker, as well as by observing Cdc23-GFP using the cdc15-2 mutant to enrich for anaphase cells, the linear Cdc23p anaphase signal did appear to colocalize with the anaphase spindle.
|
|
|
|
|
|
Microtubules are required for the proper localization of Cdc23-GFP:
We predicted that one possible mechanism of spindle checkpoint inhibition would be to mislocalize the APC/C. Conditions such as microtubule depolymerization and mutants to disrupt the spindle and kinetochore structure were used to test this hypothesis. First, cells expressing Cdc23-GFP or the GFP-tagged kinetochore component, Ndc10p, were treated with nocodazole. This drug depolymerized microtubules and resulted in spindle checkpoint activation (SHAH and CLEVELAND 2000). Localization of Ndc10-GFP was expected to be unaffected by microtubule depolymerization (GOH and KILMARTIN 1993). Both Cdc23-GFP and Ndc10-GFP localized to the kinetochore in medial nuclear division cells in the absence of nocodazole (Figure 4A, a and d; Figure 4B, g and i). However, in parallel cultures, the addition of nocodazole resulted in the delocalization of the punctate nuclear localization pattern of Cdc23-GFP but not that of Ndc10-GFP (Figure 4A, c and f; Figure 4B, h and j). For both Cdc23-GFP and Ndc10-GFP nocodazole-treated cells, visible spots often were slightly more diffuse or existed as a cluster of multiple spots (Figure 4A, b and e, Figure 4B, h and j).
|
mutant, Cdc23-GFP no longer retained its nuclear spot signal (Figure 4C, f and h). On the basis of these experiments, it appeared that Cdc23-GFP localization was physically dependent on microtubules.
The Cdc23-GFP nuclear spot signal is delocalized in the okp1-5 mutant:
Since nocodazole affected microtubule structure as well as the spindle checkpoint, we wanted to test APC/C localization under conditions activating the checkpoint but not affecting microtubules. For this purpose we examined Cdc23-GFP localization in an okp1-5 kinetochore mutant (ORTIZ et al. 1999).
At nonpermissive temperature, the okp1-5 mutant arrested at metaphase, but it was not previously known if this arrest was caused by spindle checkpoint activation. To make sure that the okp1-5 mutant activated the checkpoint, we deleted the MAD2 checkpoint gene in these cells to see if this relieved a checkpoint-activation-induced metaphase arrest. Deletion of MAD2 has been previously shown to cause a failure of arrest (BLOECHER and TATCHELL 1999; SHAH and CLEVELAND 2000; GARDNER et al. 2001). The okp1-5 mutant arrested in metaphase at nonpermissive temperature as expected, while the okp1-5mad2 mutants failed to arrest and cycled into anaphase and G1 (data not shown). From these results, it appeared that the okp1-5 metaphase arrest resulted from spindle checkpoint activation, and this arrest was relieved in the absence of MAD2. The okp1-5 mutant also had a synthetic phenotype with the mad2 spindle checkpoint mutant at 33° (data not shown). We tagged Cdc23p in the okp1-5mad2 mutant, but unfortunately the strain is too sickly to analyze the Cdc23-GFP kinetochore signal (data not shown).
After verifying spindle checkpoint activation in the okp1-5 strain, we went on to observe the localization of Cdc23-GFP and the Slk19-GFP control protein in this mutant. At nonpermissive temperature, a threefold decrease in the Cdc23-GFP nuclear spot signal in metaphase cells was seen when compared to wild-type cells in metaphase (Figure 5A, b, d, f, and h). However, the signal of the Slk19-GFP protein was unchanged at nonpermissive temperature in this mutant (Figure 5B, j and l). Indirect immunofluorescence analysis revealed that although okp1-5 spindle microtubules at 37° tended to be slightly longer than those at 23°, the spindles overall appeared normal and stable (see online supplemental Figure S1 at http://www.genetics.org/supplemental/). Therefore, disruption of the kinetochore and subsequent activation of the spindle checkpoint in an okp1-5 mutant resulted in the displacement of the Cdc23-GFP nuclear spot signal.
|
|
Cdc23-GFP localization is not changed in the checkpoint-sensor-defective ndc10-1 mutant:
Since we had previously shown that spindle checkpoint activation resulted in Cdc23-GFP delocalization from the kinetochore, we examined Cdc23-GFP expression in the ndc10-1 mutant (GOH and KILMARTIN 1993). Since ndc10-1 mutants do not arrest in nocodazole, this mutant was known to have a defective spindle checkpoint sensor (TAVORMINA and BURKE 1998; GARDNER et al. 2001). In addition, binding of the yeast CBF3 kinetochore complex to centromeric DNA was severely compromised in this mutant (SORGER et al. 1995). However, it was not known if this would affect Cdc23-GFP localization. In the case of at least five kinetochore-associated proteins, their kinetochore localization was lost in this mutant strain at nonpermissive temperature (HE et al. 2001). We predicted that Cdc23-GFP would remain on the kinetochores if the checkpoint sensor were defective, but did not rule out that the perturbed kinetochore structure in the ndc10-1 mutant may cause Cdc23-GFP to come off the kinetochore.
We localized Cdc23-GFP in the ndc10-1 strain, along with two other kinetochore proteins, Nuf2-GFP and Okp1-GFP. Nuf2p was known to be delocalized from the kinetochore in an ndc10-1 mutant, and Okp1p failed to bind centromeric DNA in a chromatin immunoprecipitation assay conducted using an ndc10-1 mutant (ORTIZ et al. 1999; HE et al. 2001). Thus both Nuf2-GFP and Okp1-GFP were expected to come off the kinetochore in an ndc10-1 mutant at nonpermissive temperature.
At permissive and nonpermissive temperature, a kinetochore signal for Cdc23-GFP was visible in the ndc10-1 strain (Figure 7A, a–d). At permissive temperature, Nuf2-GFP and Okp1-GFP spots were visible (data not shown), but these spots disappeared at nonpermissive temperature (Figure 7A, e–h). So, despite perturbations in the kinetochore structure, Cdc23-GFP was not delocalized in the ndc10-1 mutant. Since the ndc10-1 mutation is a complex mutation known to affect the checkpoint response to damage such as nocodazole treatment, the APC/C may not be removed from the kinetochore if the sensor were damaged (TAVORMINA and BURKE 1998).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
In addition to conservation of APC/C localization, the spindle checkpoint components themselves have been localized to the kinetochore in several different organisms (LEW and BURKE 2003). In the case of some budding yeast checkpoint proteins, their kinetochore localization was dependent on checkpoint activation (IOUK et al. 2002; KERSCHER et al. 2003). Extensive analysis done in mammalian cells indicated that mammalian Mad2 associated with unattached, phosphorylated kinetochores (WATERS et al. 1999; HOWELL et al. 2000). At least two of these phosphorylated proteins at the mammalian kinetochore were APC/C subunits (DAUM et al. 2000). However, during mitosis itself in mammalian cells the dephosphorylated form of Cdc27 was found on the chromosomes (TOPPER et al. 2002). These experiments indicated that at least in mammalian cells, the phosphorylation state of the APC/C affected its regulation and localization. Although our studies indicated that the APC/C localized properly in an APC/C mutant lacking consensus phosphorylation sites for Cdc28p, inhibitory phosphorylation at other sites by other kinases may have an effect on APC/C activity during spindle checkpoint activation. Future studies will address checkpoint protein localization changes in relationship to APC/C localization.
Colocalization of Cdc23-YFP with kinetochore components is not surprising, since many mitotic regulators are found at the kinetochore. These proteins include cyclin B1, human Cdc20, and Mad2 (KALLIO et al. 1998, 2002; CLUTE and PINES 1999; HOWELL et al. 2000; KERSCHER et al. 2003). However, colocalization of Cdc23-YFP with the kinetochores throughout the cell cycle is surprising, because the APC/C has no known ubiquitination function outside of mitosis and G1. Future studies may determine the significance of the APC/C nuclear spot localization during G2 and S phases.
It is important to note that although Cdc23-YFP colocalizes with kinetochore components, we cannot rule out that the nuclear spot localization actually represents the microtubule ends. It is also possible that Cdc23p may lie at the interface between the kinetochore and the microtubule. Certain proteins known to associate with the outer kinetochore, such as the DASH complex member Ask1p, associate with the kinetochore in a microtubule-dependent manner, just like Cdc23p (LI et al. 2002).
Cdc23-GFP delocalization from the kinetochore or microtubule ends as a mechanism of the spindle checkpoint response:
Our data suggest that the spindle checkpoint may halt the cell cycle not only by inhibiting APC/C activity, but also by changing its localization. The method of spindle checkpoint activation and the sensing of damage may be critical for APC/C displacement.
We activated the spindle checkpoint by several different mechanisms resulting in the delocalization of the Cdc23-GFP nuclear spot signal. These mechanisms included nocodazole treatment and the use of the okp1-5 mutant. Because depolymerizing microtubules resulted in Cdc23-GFP localization regardless of spindle checkpoint status, several different methods of checkpoint activation were applied for this study. In each of our experimental conditions, the spindle or kinetochore damage itself did not appear to cause general protein displacement from the kinetochore, since control proteins such as Slk19-GFP and Ndc10-GFP consistently remained at the kinetochore during our experiments. In addition to these methods, we tested APC/C localization using the cdc20-1 mutant. This mutant was known to activate the BUB2-dependent portion of the spindle checkpoint pathway, as well as to display defects in the organization of microtubules within the mitotic spindle (SETHI et al. 1991; O'TOOLE et al. 1997; TAVORMINA and BURKE 1998). Delocalization of the Cdc23-GFP nuclear spot signal in this mutant may have been caused by a number of factors, including checkpoint activation, spindle structural changes, or a requirement of Cdc20p itself to maintain the APC/C at the kinetochore. In all cases of spindle checkpoint activation, delocalization of the nuclear spot Cdc23-GFP signal did not necessarily mean complete loss of APC/C activity, since the APC/C retained its general nuclear localization.
In addition to the previously mentioned methods of spindle checkpoint activation, we analyzed the localization of Cdc23-GFP and two kinetochore control proteins in response to MPS1 overexpression (data not shown). Previous studies indicated that overexpression of this key regulatory kinase resulted in phosphorylation of Mad1p and spindle checkpoint activation even without spindle damage (HARDWICK et al. 1996; WEISS and WINEY 1996). The Cdc23-GFP nuclear spot signal was still visible upon MPS1 overexpression, but in some cases seemed to move from a typical kinetochore nuclear position, yet remained as foci (data not shown). Therefore, these results were inconclusive. One possibility was that these foci represented movement of the APC/C along the microtubules to the spindle poles. Perhaps different methods of spindle checkpoint activation may have had slightly different effects on APC/C localization.
Recently, several articles showing the enrichment of certain checkpoint proteins at the kinetochore, such as Mad2p, upon spindle checkpoint activation have been published (KERSCHER et al. 2003; KITAGAWA et al. 2003; GILLETT et al. 2004). However, little is known about how the amount of APC/C at the kinetochore or microtubule ends changes as these checkpoint proteins move to the kinetochore. Studies done in mammalian cells reveal that the APC/C and Mad2 may have overlapping Cdc20 binding sites (ZHANG and LEES 2001). One might predict that as Mad2 moves to the kinetochore upon spindle checkpoint activation, the APC/C is eventually excluded from Cdc20p and is forced off the kinetochore or microtubule ends.
The role of the checkpoint sensor in APC/C localization:
The checkpoint sensor appeared to be critical for APC/C regulation, since Cdc23-GFP was not displaced in the perturbed yet checkpoint-sensor-defective ndc10-1 kinetochore mutant. However, it was unexpected that Cdc23-GFP remained at the kinetochore in an ndc10-1 mutant, given the many examples of kinetochore proteins displaced in this strain (HE et al. 2001).
It is surprising that Cdc23-GFP does not localize to the kinetochore/microtubule ends in an okp1-5 mutant, but does in an ndc10-1 mutant. This means that Cdc23-GFP must not be physically dependent on Okp1p to be bound to the kinetochore, since Okp1p itself does not bind the kinetochore in an ndc10-1 mutant (ORTIZ et al. 1999). The explanation for these data is most likely related to the activation of the spindle checkpoint sensor and to the inactivation of Cdc20p. Since Cdc20p may form a complex with Mad2p brought to the kinetochore upon spindle checkpoint activation, the APC/C may now come off the kinetochore in response to this activation. To address this issue, one could look at Cdc23-GFP localization in double mutants containing mad2 and another mutation activating the spindle checkpoint such as cse4-1, ctf13-30, or dad1-1 (BIGGINS and WALCZAK 2003). In addition, Cdc23-GFP could be observed in other types of mutants with a failure of checkpoint activation, such as spc24-1, to see if this failure is directly responsible for the lack of Cdc23-GFP delocalization (BIGGINS and WALCZAK 2003).
It is also possible that if Cdc23-GFP is associated with the microtubule ends and not the kinetochore, or that if Cdc23-GFP moves onto the microtubule ends in an ndc10-1 mutant, Cdc23-GFP could retain a nuclear spot localization at nonpermissive temperature. This is consistent with previous studies indicating that certain proteins associating with the kinetochore and the spindle can retain their spindle localization in an ndc10-1 mutant (HE et al. 2001; JANKE et al. 2002; MCAINSH et al. 2003).
APC/C localization in relation to its substrates and substrate-specificity factors:
Cdc23-GFP appears to maintain its nuclear and nuclear spot localization during most of the cell cycle, but is visible on the spindle in anaphase. This spindle localization would position the APC/C for a "wave of proteolysis" as seen with cyclin B in Drosophila embryos (HUANG and RAFF 1999). It is possible that certain substrates are specifically shuttled to the spindle for ubiquitination, as in the case of mammalian cyclin B1 (CLUTE and PINES 1999). However, elimination of the spindle association of the motor protein Kip1p does not appear to prevent its ubiquitination (GORDON and ROOF 2001). Although the association of an APC/C substrate with the spindle may not be essential for ubiquitination, the timing and specificity of ubiquitination may be altered if substrates are not properly targeted to the APC/C.
Cdc23-GFP is localized to many of the same locations as known APC/C substrates. However, Hsl1p, an APC/C substrate, is found at the bud neck and not at any of the observed APC/C locations (BURTON and SOLOMON 2000, 2001). This particular substrate may need to be shuttled into the nucleus for degradation or ubiquitinated during nuclear migration through the neck. Kip1p and Cin8p are not degraded without a nuclear localization sequence, and certain nuclear transport proteins are necessary for Pds1p proteolysis (BAUMER et al. 2000; GORDON and ROOF 2001; HILDEBRANDT and HOYT 2001). These observations support our localization data, suggesting that the primary site of APC/C activity is the nucleus. Future studies should address the significance of the specific localization of the APC/C within the nucleus, determining if certain substrates are stabilized when the APC/C is displaced.
In addition to the importance of APC/C localization in determining APC/C activity, the localization of substrate-specificity factors such as Cdc20p and Hct1p may also be critical for APC/C regulation. Our studies indicate that Cdc20-GFP is localized to the nucleus and presumptive kinetochore from S phase to the end of mitosis. This localization pattern is very similar to studies conducted in mammalian cells, with a slight difference in the time period of kinetochore localization (WEINSTEIN 1997; KALLIO et al. 1998, 2002). Our findings also build on previous overexpression studies conducted in S. cerevisiae where Cdc20-GFP was described as nuclear during these same stages (JACQUENOUD et al. 2002). Since the duration of Cdc20p localization to the nucleus and presumptive kinetochore is more restricted than that of the APC/C, the expression of Cdc20p in the nucleus and at the kinetochore may be a critical step in APC/C activation. Later stages of APC/C activity may also be tied closely to localization of a second specificity factor, Hct1p/Cdh1p. Although Hct1p is expressed throughout the cell cycle, the phosphorylated form remains in the cytoplasm during much of the cell cycle, with the dephosphorylated form present in the nucleus mainly during G1 (JACQUENOUD et al. 2002; ZHOU et al. 2003). Clb2p proteolysis is impaired if Hct1p fails to enter the nucleus (JACQUENOUD et al. 2002). Clearly localization of these specificity factors plays a critical role in APC/C activation and regulation.
Conclusions:
In these studies, we have established that the S. cerevisiae APC/C subunit, Cdc23p, is localized to the nucleus, mitotic spindle, and either the kinetochore or the microtubule ends. These localization data are consistent with studies conducted in mammalian cells. Delocalization of the Cdc23-GFP nuclear spot signal upon spindle checkpoint activation also establishes a possible universal mechanism for APC/C inhibition.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
BAUMER, M., M. KUNZLER, P. STEIGEMANN, G. H. BRAUS and S. IRNIGER, 2000 Yeast ran-binding protein Yrb1p is required for efficient proteolysis of cell cycle regulatory proteins Pds1p and Sic1p. J. Biol. Chem. 275: 38929–38937.
BI, E., and J. R. PRINGLE, 1996 ZDS1 and ZDS2, genes whose products may regulate Cdc42p in Saccharomyces cerevisiae. Mol. Cell. Biol. 16: 5264–5275.[Abstract]
BIGGINS, S., and C. E. WALCZAK, 2003 Captivating capture: how microtubules attach to kinetochores. Curr. Biol. 13: R449–R460.[CrossRef][Medline]
BLOECHER, A., and K. TATCHELL, 1999 Defects in Saccharomyces cerevisiae protein phosphatase type I activate the spindle/kinetochore checkpoint. Genes Dev. 13: 517–522.
BURTON, J. L., and M. J. SOLOMON, 2000 Hsl1p, a Swe1p inhibitor, is degraded via the anaphase-promoting complex. Mol. Cell. Biol. 20: 4614–4625.
BURTON, J. L., and M. J. SOLOMON, 2001 D box and KEN box motifs in budding yeast Hsl1p are required for APC-mediated degradation and direct binding to Cdc20p and Cdh1p. Genes Dev. 15: 2381–2395.
CASTILLO, A. R., J. B. MEEHL, G. MORGAN, A. SHUTZ-GESCHWENDER and M. WINEY, 2002 The yeast protein kinase Mps1p is required for assembly of the integral spindle pole body component Spc42p. J. Cell Biol. 156: 453–465.
CHEESEMAN, I. M., S. ANDERSON, M. JWA, E. M. GREEN, J. KANG et al., 2002 Phospho-regulation of kinetochore-microtubule attachments by the aurora kinase Ipl1p. Cell 111: 163–172.[CrossRef][Medline]
CHEN, R.-H., D. M. BRADY, D. SMITH, A. W. MURRAY and K. G. HARDWICK, 1999 The spindle checkpoint of budding yeast depends on a tight complex between the Mad1 and Mad2 proteins. Mol. Biol. Cell 10: 2607–2618.
CHEN, Y., R. E. BAKER, K. C. KEITH, K. HARRIS, S. STOLER et al., 2000 The N terminus of the centromere H3-like protein Cse4p performs an essential function distinct from that of the histone fold domain. Mol. Cell. Biol. 20: 7037–7048.
CLEVELAND, D. W., Y. MAO and K. F. SULLIVAN, 2003 Centromeres and kinetochores: from epigenetics to mitotic checkpoint signaling. Cell 112: 407–421.[CrossRef][Medline]
CLUTE, P., and J. PINES, 1999 Temporal and spatial control of cyclin B1 destruction in metaphase. Nat. Cell Biol. 1: 82–87.[CrossRef][Medline]
COHEN-FIX, O., J.-M. PETERS, M. KIRSCHNER and D. KOSHLAND, 1996 Anaphase initiation in Saccharomyces cerevisiae is controlled by the APC-dependent degradation of the anaphase inhibitor Pds1p. Genes Dev. 10: 3081–3093.
DAUM, J. R., S. TUGENDREICH, L. M. TOPPER, P. M. JORGENSEN, C. HOOG et al., 2000 The 3F3/2 anti-phosphoepitope antibody binds the mitotically phosphorylated anaphase-promoting complex/cyclosome. Curr. Biol. 10: R850–R852.[CrossRef][Medline]
DONALDSON, A. D., and J. V. KILMARTIN, 1996 Spc42p: a phosphorylated component of the S. cerevisiae spindle pole body (SPB) with an essential function during SPB duplication. J. Cell Biol. 132: 887–901.
GAMMIE, A. E., and M. D. ROSE, 2002 Assays of cell and nuclear fusion. Methods Enzymol. 351: 477–498.[Medline]
GARDNER, R. D., A. PODDAR, C. YELLMAN, P. A. TAVORMINA, M. C. MONTEAGUDO et al., 2001 The spindle checkpoint of the yeast Saccharomyces cerevisiae requires kinetochore function and maps to the CBF3 domain. Genetics 157: 1493–1502.
GILLETT, E. S., C. W. ESPELIN and P. K. SORGER, 2004 Spindle checkpoint proteins and chromosome-microtubule attachment in budding yeast. J. Cell Biol. 164: 1–12.
GOH, P. Y., and J. V. KILMARTIN, 1993 NDC10: a gene involved in chromosome segregation in Saccharomyces cerevisiae. J. Cell Biol. 121: 503–512.
GORDON, D. M., and D. M. ROOF, 2001 Degradation of the kinesin Kip1p at anaphase onset is mediated by the anaphase-promoting complex and Cdc20p. Proc. Natl. Acad. Sci. USA 98: 12515–12520.
GOSHIMA, G., and M. YANAGIDA, 2000 Establishing biorientation occurs with precocious separation of the sister kinetochores, but not the arms, in the early spindle of budding yeast. Cell 100: 619–633.[CrossRef][Medline]
HALL, M. C., M. P. TORRES, G. K. SCHROEDER and C. H. BORCHERS, 2003 Mnd2 and Swm1 are core subunits of the Saccharomyces cerevisiae anaphase-promoting complex. J. Biol. Chem. 278: 16698–16705.
HARDWICK, K. G., E. WEISS, F. C. LUCA, M. WINEY and A. W. MURRAY, 1996 Activation of the budding yeast spindle assembly checkpoint without mitotic spindle disruption. Science 273: 953–956.[Abstract]
HARDWICK, K. G., R. C. JOHNSTON, D. L. SMITH and A. W. MURRAY, 2000 MAD3 encodes a novel component of the spindle checkpoint which interacts with Bub3p, Cdc20p, and Mad2p. J. Cell Biol. 148: 871–882.
HARPER, J. W., J. L. BURTON and M. J. SOLOMON, 2002 The anaphase-promoting complex: it's not just for mitosis any more. Genes Dev. 16: 2179–2206.
HARTWELL, L. H., J. CULOTTI, J. R. PRINGLE and B. J. REID, 1974 Genetic control of the cell division cycle in yeast. Science 183: 46–51.
HE, X., S. ASTHANA and P. K. SORGER, 2000 Transient sister chromatid separation and elastic deformation of chromosomes during mitosis in budding yeast. Cell 101: 763–775.[CrossRef][Medline]
HE, X., D. R. RINES, C. W. ESPELIN and P. K. SORGER, 2001 Molecular analysis of kinetochore-microtubule attachment in budding yeast. Cell 106: 195–206.[CrossRef][Medline]
HILDEBRANDT, E. R., and M. A. HOYT, 2001 Cell cycle-dependent degradation of the Saccharomyces cerevisiae spindle motor Cin8p requires APCCdh1 and a bipartite destruction sequence. Mol. Biol. Cell 12: 3402–3416.
HOLLOWAY, S. L., M. GLOTZER, R. W. KING and A. W. MURRAY, 1993 Anaphase is initiated by proteolysis rather than by the inactivation of maturation-promoting factor. Cell 73: 1393–1402.[CrossRef][Medline]
HOWELL, B. J., D. B. HOFFMAN, G. FANG, A. W. MURRAY and E. D. SALMON, 2000 Visualization of Mad2 dynamics at kinetochores, along spindle fibers, and at spindle poles in living cells. J. Cell Biol. 150: 1233–1250.
HOYT, M. A., L. TOTIS and B. T. ROBERTS, 1991 S. cerevisiae genes required for cell cycle arrest in response to loss of microtubule function. Cell 66: 507–517.[CrossRef][Medline]
HUANG, J., and J. W. RAFF, 1999 The disappearance of cyclin B at the end of mitosis is regulated spatially in Drosophila cells. EMBO J. 18: 2184–2195.[CrossRef][Medline]
HUANG, J.-Y., and J. W. RAFF, 2002 The dynamic localization of the Drosophila APC/C: evidence for the existence of multiple complexes that perform distinct functions and are differentially localised. J. Cell Sci. 115: 2847–2856.
HUH, W.-K., J. V. FALVO, L. C. GERKE, A. S. CARROLL, R. W. HOWSON et al., 2003 Global analysis of protein localization in budding yeast. Nature 425: 686–691.[CrossRef][Medline]
HWANG, L. H., L. F. LAU, D. L. SMITH, C. A. MISTROT, K. G. HARDWICK et al., 1998 Budding yeast Cdc20: a target of the spindle checkpoint. Science 279: 1041–1044.
IOUK, T., O. KERSCHER, R. J. SCOTT, M. A. BASRAI and R. W. WOZNIAK, 2002 The yeast nuclear pore complex functionally interacts with components of the spindle assembly checkpoint. J. Cell Biol. 159: 807–819.
IRNIGER, S., 2002 Cyclin destruction in mitosis: a crucial task of Cdc20. FEBS Lett. 532: 7–11.[CrossRef][Medline]
JABLONSKI, S. A., G. K. T. CHAN, C. A. COOKE, W. EARNSHAW and T. J. YEN, 1998 The hBUB1 and nBUBR1 kinases sequentially assemble onto kinetochores during prophase with hBUBR1 concentrating at the kinetochore plates in mitosis. Chromosoma 107: 386–396.[CrossRef][Medline]
JACQUENOUD, M., F. VAN DROGEN and M. PETER, 2002 Cell cycle-dependent nuclear export of Cdh1p may contribute to the inactivation of APC/CCdh1. EMBO J. 21: 6515–6526.[CrossRef][Medline]
JANKE, C., J. ORTIZ, T. U. TANAKA, J. LECHNER and E. SCHIEBEL, 2002 Four new subunits of the Dam1-Duo1 complex reveal novel functions in sister kinetochore biorientation. EMBO J. 21: 181–193.[CrossRef][Medline]
JASPERSEN, S. L., J. F. CHARLES, R. L. TINKER-KULBERG and D. O. MORGAN, 1998 A late mitotic regulatory network controlling cyclin destruction in Saccharomyces cerevisiae. Mol. Biol. Cell 9: 2803–2817.
JENSEN, S., M. SEGAL, D. J. CLARKE and S. I. REED, 2001 A novel role of the budding yeast separin Esp1 in anaphase spindle elongation: evidence that proper spindle association of Esp1 is regulated by Pds1. J. Cell Biol. 152: 27–40.
JORGENSEN, P.-M., E. BRUNDELL, M. STARBORG and C. HOOG, 1998 A subunit of the anaphase-promoting complex is a centromere-associated protein in mammalian cells. Mol. Cell. Biol. 18: 468–476.
JUANG, Y.-L., J. HUANG, J.-M. PETERS, M. E. MCLAUGHLIN, C.-Y. TAI et al., 1997 APC-mediated proteolysis of Ase1 and the morphogenesis of the mitotic spindle. Science 275: 1311–1314.
KAHANA, J., B. SCHNAPP and P. SILVER, 1995 Kinetics of spindle pole body separation in budding yeast. Proc. Natl. Acad. Sci. USA 92: 9707–9711.
KALLIO, M., J. WEINSTEIN, J. R. DAUM, D. J. BURKE and G. J. GORBSKY, 1998 Mammalian p55CDC mediates association of the spindle checkpoint protein Mad2 with the cyclosome/anaphase-promoting complex, and is involved in regulating anaphase onset and late mitotic events. J. Cell Biol. 141: 1393–1406.
KALLIO, M. J., V. A. BEARDMORE, J. WEINSTEIN and G. J. GORBSKY, 2002 Rapid microtubule-independent dynamics of Cdc20 at kinetochores and centrosomes in mammalian cells. J. Cell Biol. 158: 841–847.
KERSCHER, O., L. B. CROTTI and M. A. BASRAI, 2003 Recognizing chromosomes in trouble: association of the spindle checkpoint protein Bub3p with altered kinetochores and a unique defective centromere. Mol. Cell. Biol. 23: 6406–6418.
KITAGAWA, K., R. ABDULLE, P. K. BANSAL, G. CAGNEY, S. FIELDS et al., 2003 Requirement of Skp1-Bub1 interaction for kinetochore-mediated activation of the spindle checkpoint. Mol. Cell 11: 1201–1213.[CrossRef][Medline]
KURASAWA, Y., and K. TODOKORO, 1999 Identification of human APC10/Doc1 as a subunit of anaphase promoting complex. Oncogene 18: 5131–5137.[CrossRef]
