![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Jing Guo,
Qi Chen,
Chong Ding,
Juan Du, and
Xueliang Zhu*
Laboratory of Molecular Cell Biology, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, People's Republic of China
Received 3 September 2004/ Returned for modification 13 October 2004/ Accepted 8 February 2005
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Many proteins exhibit centromere localizations (25, 37). A few of them, for instance, CENP-B and CENP-C, associate with the centromere throughout the cell cycle. The rest, including cytoplasmic dynein, CENP-E, Mad2, Bub1, BubR1, and mitosin (CENP-F), only exhibit transient associations in mitosis. Both dynein and CENP-E are microtubule-based motors and are important for chromosome movement during congression and segregation (4, 8, 42). Dynein is a protein complex that powers movement towards microtubule minus ends, or spindle poles, with the help of another complex, dynactin. CENP-E, a kinesin motor-like protein, may direct motion in the opposite direction. Nevertheless, inactivation of either motor does not disrupt congression, a process for chromosomes to achieve alignment at the mitotic equator, possibly due to redundancy of their functions (4).
In Drosophila melanogaster, inactivation of dynein genetically or through dynein inhibitors causes multiple mitotic defects, including impaired centrosome positioning, kinetochore alignment, and poleward chromatid movement. Congression, however, still occurs (38, 41). Depleting mammalian CENP-E, on the other hand, results in monooriented chromosomes juxtaposed to spindle poles, decreased numbers of kinetochore microtubules, and markedly attenuated tension across sister kinetochores of fully aligned chromosomes (29, 33, 45, 47, 50). Mad2 and BubR1 are major components of the spindle checkpoint, a mechanism that guarantees anaphase onset only after the bipolar microtubule attachment for every pair of sister kinetochores. Their sensor-like functions are mainly attributed to their rapid depletions or attenuations from kinetochores upon microtubule attachment, so that the checkpoint is subsequently inactivated (4, 31, 40). Motor proteins also contribute to the checkpoint inactivation. Cytoplasmic dynein transports outer kinetochore proteins, including CENP-E, mitosin, and the aforementioned checkpoint proteins, to spindle poles to sequester them from kinetochores upon microtubule attachment (12, 13, 48, 49). CENP-E, on the other hand, activates BubR1 kinase and mediates silencing of BubR1 signaling (26).
Mitosin, also named CENP-F, is a protein of 3,113 amino acid residues located at the outer kinetochore plate in M phase (23, 34, 49, 52, 54). It is expressed in a cell cycle-dependent manner, with the protein levels low in G1 phase but elevated sharply from the G1/S boundary as a nuclear protein. It is hyperphosphorylated in M phase, exhibits localizations at kinetochores, spindle, and spindle poles, and is rapidly degraded at the end of mitosis (53, 54). Deletion analysis suggests that the "core region" located between residues 2792 and 2887 is responsible for kinetochore targeting, while other regions, including residues 2094 to 2487 and/or 2889 to 3113, stabilize the localization (51). In addition, its kinetochore targeting is also affected by Bub1, RanBP2, CENP-I, and Sgt1 (17, 18, 24, 39, 43). It is farnesylated at the C-terminal CAAX motif, and the modification is important for its kinetochore localization and degradation (1, 15). Moreover, overexpressing certain C-terminal portions of mitosin results in cell cycle delay, mainly in M phase (15, 54).
Although growing lines of evidence suggest a role of mitosin in M-phase progression, this issue has not been explored in details. In addition, a fragment of CENP-F (mitosin) has been shown to interact with CENP-E in a yeast two-hybrid screen (2). Coimmunoprecipitation of both proteins as well as reduction of kinetochore CENP-E in CENP-F-depleted cells further support their interaction (17, 50). Nevertheless, whether mitosin is functionally related to CENP-E remains unexplored. Here, we addressed these questions by studying phenotypes of mitosin depletion using vector-based RNA interference.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cell culture.
HeLa and HEK293T cells were maintained in Dulbecco's modified Eagle's medium (GIBCO) supplemented with 10% calf serum (Sijiqing Company, Hangzhou, People’s Republic of China) at 37°C in an atmosphere containing 5% CO2. Cells were transfected using the calcium phosphate method and assayed 48 h posttransfection unless otherwise described. After transfection for 
21 h, HeLa cells were initially treated with thymidine (2 mM) for additional 18 h for synchronization at the G1/S boundary and then cultured for another 9 h to enrich mitotic populations before being processed for immunostaining or live cell imaging. To disassemble microtubules (MTs), cells were released from thymidine block for 6 h and then cultured in the presence of nocodazole (5 µg/ml) for additional 3 h (22).
Antibodies.
Antibodies to mitosin, BubR1, p150glued (BD Transduction Laboratories), and 
-tubulin (Sigma) were monoclonal. Rabbit antibodies to survivin and activated caspase-3 were from Novus Biologicals and New England Biolabs, respectively. Chicken antimitosin was prepared as described (49). Rabbit antibodies to CENP-E, Mad2, and human Nuf2 were kindly provided by T. J. Yen (Fox Chase Cancer Center, Philadelphia, PA), E. D. Salmon (University of North Carolina at Chapel Hill, Chapel Hill, North Carolina), and J. Kilmartin (MRC Laboratory of Molecular Biology, Cambridge, United Kingdom), respectively. Sheep antibodies to Bub1 and Aurora B were gifts from S. S. Taylor (University of Manchester, Manchester, United Kingdom). Human anticentromere antibody from patients with CREST (calcinosis, Raynaud’s phenomenon, esophageal dysmotility, sclerodactyly, telangiectasia) syndrome was a gift from W. R. Brinkley (Baylor College of Medicine, Houston, TX). Secondary antibodies conjugated with Alexa 488, 546, or 633 were purchased from Molecular Probes.
Fluorescence microscopy, measurement, and quantitation. To stain for kinetochore proteins, cells growing on glass coverslips were rinsed twice with PHEM (60 mM piperazine ethanesulfonic acid, 45 mM HEPES, 10 mM EGTA, 2 mM MgCl2, pH 6.9) followed by permeabilization with 0.5% Triton X-100 in PHEM for 3 min, and then fixed with 3.5% paraformaldehyde in PHEM for 15 min. Cold treatment of cells was performed on ice for 10 min as described (5) before processing on ice for fixation. Indirect immunofluorescence labeling was performed as described previously (49). Fluorescence images were captured by using Leica TCS SP2 laser confocal microscope. Optical sections were scanned at 0.1- to 0.2-µm intervals. Z stack images were then formed by maximal projection. Statistic results were obtained in a blind fashion and presented as mean ± standard error of the mean.
To analyze kinetochore orientations, we first connected both outer edges of each sister kinetochore in an optical section with a straight line based on CREST staining to mark the kinetochore orientation. The angle included between this line and the spindle axis was then measured. Only kinetochores aligned at the spindle midzone were analyzed. Kinetochore pairs located in different optical sections were neglected due to technical difficulties in measuring their angles.
To quantify kinetochore immunofluorescence, an area with relatively homogenous background was selected in a Z-stack cell image. The mean brightness (Bm) of the area was measured for the CREST antigen and another protein by using Adobe Photoshop. One or several representative regions lacking centromere staining were then chosen within the same area to obtain the averaged background brightness (Bab). The relative kinetochore brightness BR = (Bm – Bab)/[Bm(CREST) – Bab(CREST)] to eliminate variations in staining and image acquisition (17). The number of kinetochores included in this area was estimated according to the CREST staining. Several separate areas might be chosen in each cell. The BR values from multiple measurements were averaged. The relative fluorescence intensity (Fig. 4C and Fig. 5C) was obtained by converting the average BR from control cells to 1 and then normalizing the value from mitosin-depleted cells accordingly.
|
|
|
Flow cytometry. HEK293T cells were seeded in petri dishes containing two glass coverslips. pEGFP-F (Clontech) was cotransfected into the cells with pBS/U6 or pBS/U6/Mi-1 at a ratio of 1:20; 48 h after transfection, cells were fixed in cold 70% ethanol for 1 h and then stained with propidium iodide (20 µg/ml) in the presence of RNase A (200 µg/ml). Samples were analyzed using a fluorescence-activated cell sorter (FACS, Becton Dickinson). The cell cycle distributions of green fluorescent protein (GFP)-positive cells were plotted. Cells growing on coverslips were processed for microscopic examinations.
Live cell imaging. A HeLa cell line expressing histone 2B-GFP (19) was kindly provided by G. M. Wahl (Salk Institute for Biological Studies, La Jolla, California). pDsRed-N1 (Clontech) was cotransfected with pBS/U6 or pBS/U6/Mi-1 at a ratio of 1:10 using Lipofectamine 2000 (Invitrogen). Cells growing on circular coverslips were mounted in Ludin chamber with Dulbecco's modified Eagle's medium (Gibco) containing 10% fetal calf serum (Sijiqing Company, Hangzhou, China) buffered with 20 mM HEPES, pH 7.2. DsRed-positive cells were recorded at 37°C for up to 240 min with exposure intervals of 2 to 3 min using a 60x objective. Cells were not exposed to light between exposure intervals. The Leica AS MDW (multidimensional workstation for live cell imaging) system with a CoolSNAP HQ Monochrome camera (Roper Scientific) (Fig. 6; see supplemental videos 1 to 3 in the supplemental material) and Olympus IX81 microscope with Evolution QEi FAST Monochrome camera (Media Cybernetics) (Fig. 8) gave comparable results with little bleaching.
|
|
Cotransfection efficiency was measured as follows. The HeLa cells transfected with pDsRed-N1 and pBS/U6/Mi-1 were fixed and immunostained for mitosin with Alexa 633. Images of DsRed-positive cells in prometaphase were recorded using a cooled charge-coupled device (SPOT II, Diagnostic Instruments) to allow examination of the mitosin staining because of the invisibility of Alexa 633 to the naked eye; 98.0 ± 1.0% DsRed-positive cells (total n = 200, two experiments) were found negative for mitosin. In contrast, mitosin staining was not affected in DsRed-postive mitotic cells cotransfected with pBS/U6 (n = 60).
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-tubulin, CENP-E, and p150glued, a dynactin subunit, were not affected (Fig. 1A). Quantitation of the immunoblots indicated 70.5 ± 5.5% reduction in mitosin levels at 24 h posttransfection and 96.5 ± 0.5% reduction at 48 h (Fig. 1A). Indirect immunofluorescence staining indicated that kinetochore mitosin levels were also significantly reduced. The average kinetochore intensity (n = 499) in prometaphase was decreased by 98.13 ± 0.02% compared with the value (n = 454) in surrounding control cells. Therefore, pBS/U6/Mi-1 was an appropriate construct for knocking down mitosin expression.
|
Despite high transfection efficiency, HEK293T cells were not a good choice for morphological analysis due to their small size and poor attachment to coverslips. We therefore switched to HeLa cells. Cell death characteristic of highly condensed mininuclei was also observed in GFP-positive HeLa cells cotransfected with pBS/U6/Mi-1 (Fig. 1C, panels 4 and 6), but not with pBS/U6 (Fig. 1C, panels 1 and 3) for 72 h. To further explore the nature of the cell death, activated caspase-3, which is essential for many types of apoptosis (32), was costained with DNA. The dying HeLa cells transfected with the RNA interference plasmid indeed contained activated caspase-3 (Fig. 1C, panels 4 to 6), while caspase-3-positive cells were rare in populations transfected with the vector (Fig. 1C, panels 1 to 3). There were 6.8-fold more apoptotic cells in mitosin-depleted populations on average than in the mock-transfected populations (Fig. 1D).
Mitosin depletion results in kinetochore malorientation and tension decrease.
Since accumulation of mitotic cells suggests defects in mitosis, we examined HeLa cells in detail. Transfectants of pBS/U6/Mi-1 were identified by their strongly declined centromere staining of mitosin compared with surrounding mitosin-positive cells, while the immunostaining by anticentromere antibodies from CREST patients was in either case unaffected (Fig. 2A and B, panels 1 to 5). Quantitation showed that kinetochore mitosin was reduced by 96% on average in prometaphase cells (see Fig. 4C).
Mitosin-depleted mitotic cells exhibited a high incidence of misaligned chromosomes (Fig. 2B, panels 1 to 5): metaphase-like cells with misaligned chromosomes, or pseudometaphase cells (47) accounted for 41.5 ± 2.5% (n = 200, two experiments) of total mitotic cells, while in surrounding mitosin-positive cells the frequency was only 7.0 ± 1.0% (Fig. 2E). Many fewer mitosin-depleted cells were in metaphase, anaphase, and telophase (Fig. 2E). On the other hand, the percentages of prophase and prometaphase cells were comparable (Fig. 2E). These results suggest a mitotic delay at metaphase in mitosin-depleted populations, consistent with the flow cytometry results (Fig. 1B).
Mitosin-depleted pseudometaphase cells were most frequently seen to contain two to five misaligned chromosomes (Fig. 2C), the majority of which lay juxtaposed to one or both spindle poles (Fig. 2B, panels 1 to 5, arrowheads). The polar chromosomes were not derived from premature sister chromatid segregation because the characteristic double-dot patterns of BubR1 were visible at their centromere regions (Fig. 2C, inset; data not shown). Moreover, their centromeres were frequently (17 of 26) seen to associate with aster microtubule bundles when the corresponding Z sections of confocal image sets were examined for 6 pseudometaphase cells (Fig. 2B, panels 6 to 14; data not shown), suggestive of monooriented attachments (Fig. 2B, panels 8, 11, and 14). Bipolar attachments were occasionally seen (2 of 26) (Fig. 2B, panels 12 to 14). The rest (7 of 26) was difficult to judge, mostly due to locations in areas with high microtubule density. Interestingly, many of the monooriented kinetochores (6 of 17) exhibited clear syntelic attachments, i.e., both sister kinetochores attached by microtubules from the same pole (Fig. 2B, panels 6 to 11). The actual incidence might be higher, considering difficulties in estimating microtubule attachment at other kinetochore pairs.
The kinetochore orientations of aligned chromosomes also exhibited abnormalities. In control metaphase cells, kinetochores were oriented roughly in the direction of the spindle long axis (Fig. 2A). However, kinetochores positioned in other directions were frequently noticed in mitosin-depleted cells (Fig. 2B, panels 1 to 5; Fig. 2D, inset). When the angles included between sister kinetochore orientations and the spindle long axis were measured (see Materials and Methods), a clear difference was seen: kinetochores oriented between 15 to 30 degrees and more than 30 degrees away from the long axis were, respectively, 2.5-fold and 8.3-fold more frequent in mitosin-depleted cells (Fig. 2D). Since kinetochore pairs located in different optical sections were not analyzed due to technical difficulties, these data might have underestimated the actual situations of kinetochore malorientation in mitosin-depleted cells. Despite this, the difference was already very significant.
We also compared tension across the kinetochore by measuring the distances between sister centromeres based on CREST serum staining. The longer the distance, the more tension across the kinetochore (29, 37, 50). In control metaphase cells, the average sister kinetochore separation was 1.59 ± 0.01 µm (n = 102). In mitosin-depleted pseudometaphase cells, the value was 1.26 ± 0.02 µm (n = 106) for chromosomes aligned at the spindle midzone. The difference was very significant between the two populations (P < 10–5). In contrast, the value was 0.97 ± 0.01 µm (n = 100) in nocodazole-treated cells. Given that the kinetochore distance in nocodazole-treated cells represented the neutral state, silencing mitosin decreased the net replacement of a kinetochore from its neutral position by 53% on average in fully aligned chromosomes, significantly attenuating tension across the sister centromere.
Mitosin depletion induces spindle abnormalities without impairing the stability of bipolar K-fibers.
The microtubule-kinetochore association is a result of complicated yet concerted coordination between dynamic microtubules and kinetochore activities (4, 8, 42). Proper spindle organization and microtubule-kinetochore interactions also contribute to normal mitosis. In mitosin-depleted HeLa cells, 16.0 ± 4.4% (n = 300) exhibited multipolar spindles (Fig. 3A, panels 1 to 4), while the incidence was 6.3 ± 3.1% (n = 300) in surrounding control cells. 
-Tubulin was found at each pole of multipolar spindles in mitosin-depleted cells (data not shown), suggesting that such spindles may result from extra centrosomes in polyploid cells due to failed or abnormal cytokinesis (27). In addition, 39.4 ± 5.9% (n = 300) of mitosin-depleted cells in late prometaphase or metaphase showed abnormal bipolar spindles that were distorted and/or contained microtubule bundles overgrown across the spindle equators (Fig. 3A, panels 5 to 8, arrows, and data not shown). Some of these long bundles were probably not kinetochore fibers, because they extended from one pole to the other (Fig. 3A, panels 5 to 8, arrows). Similar spindle abnormalities were found in only 8.5 ± 0.5% of control cells. Furthermore, the spindles were generally much longer in mitosin-depleted cells at pseudometaphase. The average spindle length was 13.5 ± 0.4 µm (n = 25), in contrast to 9.1 ± 0.2 µm (n = 15) in control cells from late prometaphase to metaphase.
|
Mitosin contributes to dynamics of kinetochore CENP-E and dynein/dynactin.
The malorientation and reduced tension for kinetochores of aligned chromosomes (Fig. 2) implied certain defects of the force-generating machinery, most likely kinetochore motors (4, 8, 42). CENP-E is a kinesin motor-like protein whose inactivation also leads to misaligned polar chromosomes and tension reduction (29, 33, 47, 50). Furthermore, its potential interaction with and partial dependency on kinetochore localization on CENP-F (mitosin) (2, 17, 50) imply a functional relationship between them. Consistently, we also observed reduction of kinetochore CENP-E in mitosin-depleted cells (Fig. 4A). In prometaphase cells, the average kinetochore intensity of CENP-E was reduced to 28% of that in control cells (Fig. 4C). A similar value (30%) was obtained from metaphase and pseudometaphase cells. In contrast, kinetochore BubR1 was not affected after silencing mitosin (Fig. 4B and C). Neither was kinetochore Mad2, Bub1, hNuf2, Aurora B, or survivin significantly altered (data not shown).
The partial influence appeared to preclude mitosin as the protein responsible for kinetochore targeting of CENP-E. Rather, it might stabilize the CENP-E-kinetochore association. In fact, many outer kinetochore proteins bind kinetochores dynamically in M phase. Their steady-state levels are thus the net results of dynamic association versus dissociation (12, 14). Moreover, their kinetochore levels often decline in accordance with the extent of microtubule-kinetochore attachment (10, 20).
To further understand how the kinetochore CENP-E was influenced by mitosin, we examined whether its reduction in mitosin-depleted cells was related to microtubule attachments. Indeed, after 3 h treatment with nocodazole, the kinetochore CENP-E levels increased to 75% of control values on average (Fig. 4C and D).
Due to the importance of dynein in spindle organization and chromosome movement, we also examined dynactin, which is essential for dynein's kinetochore targeting and functions in mitosis (4, 6, 9, 46). Dynactin was analyzed because its subunit p150glued showed stronger kinetochore staining than dynein (data not shown). We found that p150glued also displayed weaker kinetochore staining in mitosin-depleted prometaphase cells than in mitotic phase-matched control cells (Fig. 5A and data not shown). Its average centromere intensity was 22% of the control value (Fig. 5C). Phase match was important because kinetochore dynactin is sensitive to and is markedly reduced upon microtubule attachment (6). We also noticed that, in early prometaphase, when tensions across the kinetochore were not apparent according to the CREST staining, less p150glued was accumulated at spindle poles in control cells than in mitosin-depleted cells (Fig. 5A). Moreover, in contrast to the general low levels of kinetochore dynactin, bright staining was often seen at kinetochores away from the spindle in cells lacking mitosin (Fig. 5A, arrowhead; and data not shown), suggesting microtubule-dependent reduction of kinetochore dynactin. Indeed, in the presence of nocodazole, the centromere localization of p150glued in mitosin-depleted cells was completely restored to the levels in control cells (Fig. 5B and C). Therefore, mitosin depletion attenuated both kinetochore CENP-E (Fig. 4) and dynactin and dynein (Fig. 5) in the presence of microtubules.
Time-lapse microscopy reveals defective chromosome congression followed by cell death in mitosin-depleted cells. For global views of mitotic cell behaviors, we performed time-lapse microscopy using a HeLa cell line stably expressing histone 2B-GFP as a chromosome marker (19). When 15 live cells cotransfected with pBS/U6 and pDsRed-N1, a plasmid expressing the red fluorescence protein DsRed (Clontech) as a transfection marker, were monitored from prophase or early prometaphase, all of them started anaphase within 45 ± 2 min and completed mitosis within 58 ± 2 min. Congression was rapid; after nuclear envelope breakdown, all chromosomes moved quickly towards the metaphase plate to achieve full alignment in 18 ± 1 min. All these cells divided normally with no visible lagging chromosomes (Fig. 6A; see video S1 in the supplemental material; also data not shown).
The situation was quite different in DsRed expressors cotransfected with pBS/U6/Mi-1; 17 out of 18 cells that had been monitored for 240 min contained various numbers of chromosomes unable to migrate to the metaphase plate but remaining in the vicinity of spindle poles during congression. The remaining cell stayed in prometaphase. Three of the 17 cells exhibited only one unaligned chromosome at late stages of recording (Fig. 6C; see video S3 in the supplemental material; also data not shown). Nevertheless, anaphase onset was not seen in any of the 18 cells. From prophase or early prometaphase, the average time required for most chromosomes to achieve alignments was 32 ± 3 min (n = 17) (Fig. 6B and C; see videos S2 and S3 in the supplemental material), which was significantly longer than the time required for full chromosome alignment in control cells (18 ± 1 min). As time went on, the chromosomes became gradually decondensed in all cells and then, in 15 cells, aggregated to form mininuclei at 151 ± 7 min, a hallmark of cell death (Fig. 6B and C; see videos S2 and S3 in the supplemental material). Among the dying ones, four cells displayed plasma membrane blebbing reminiscent of apoptosis following chromosome deformation (30). These aberrations were not due to the experimental conditions, since normal mitoses of surrounding untransfected cells were seen in our raw image sets throughout the experiments (see video S3 in the supplemental material; also data not shown). Collectively, these results are in agreement with those from fixed cells (Fig. 1 and 2), indicating critical roles of mitosin in chromosome congression and mitotic cell survival. Moreover, they clarified that the cell death seen in fixed cells (Fig. 1) occurred in mitosis.
Abnormal chromosome movement was observed in mitosin-depleted cells. In vector-transfected cells, all chromosomes, even those close to poles, moved rapidly towards the metaphase plate (Fig. 6A; see video S1 in the supplemental material). In DsRed expressors cotransfected with pBS/U6/Mi-1, chromosomes near the metaphase equator congressed relatively rapidly, while those away from it often failed to move likewise and then tended to become misaligned (Fig. 6B to D; see videos S2 and S3 in the supplemental material). Occasionally, some unaligned chromosomes were seen to move away from the equator, possibly towards spindle poles (Fig. 6D). Some were also seen to move abruptly plateward after certain time points (Fig. 6D), indicating that congression of the polar chromosomes can occur gradually when their kinetochores are attached by proper microtubules.
Mitosin depletion-induced cell death coincides with premature chromosome decondensation. In an attempt to explore the cause of cell death, we carefully examined HeLa cells fixed after 13 h of release from thymidine block. About 75.5 ± 2.5% of dying or dead cells (n = 400, two experiments) contained highly condensed mininuclei of various numbers and sizes typical of apoptosis, some of which were accompanied by bright microtubule bundles reminiscent of spindle remains (Fig. 7, panels 21 to 25). Surprisingly, the remaining 24.5 ± 3.5% still exhibited brightly stained spindles and thus should be in earlier stages of cell death (Fig. 7, panels 1 to 20). Moreover, some cells were seen with several chromosomes decondensed into amorphous chunks of chromatin, while the remaining chromosomes still maintained typical morphologies (Fig. 7, panels 1 to 5). Such changes appeared to reflect initial stages of cell death, since cells with increasing extents of chromosome decondensation were also observed (Fig. 7, panels 6 to 15, arrows; data not shown) and the live cell recording showed progressive chromosome decondensation prior to chromatin condensation and membrane deformation (Fig. 6B and C; see videos S2 and S3 in the supplemental material; also data not shown).
|
Proteasome activity is dispensable in mitosin depletion-induced cell death. To confirm whether the mitosin-depleted cells died before anaphase onset, cells were treated with MG132, an inhibitor of the proteasome. Since proteasome activation is essential for chromatin segregation (4, 31, 40), its inhibition blocks cells in metaphase (3, 10). We found that 25 µM of MG132 (Sigma) was sufficient to arrest HeLa cells in metaphase for at least 4 h with few side effects regardless of cotransfection with pDsRed-N1 and pBS/U6 (Fig. 8A and data not shown). Of 12 DsRed-positive prometaphase/metaphase cells cotransfected with pBS/U6/Mi-1, however, eight still showed chromosome decondensation and aggregation characteristic of cell death during 4 h of recording (Fig. 8B and data not shown). Membrane deformation was also observed in many (5 of 8) of the dying cells (Fig. 8B and data not shown). Therefore, mitosin-depleted cells die before chromosome segregation or anaphase onset.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mitosin is required for full chromosome alignment, probably through interaction with CENP-E. The formation of pseudometaphase cells in the absence of mitosin (Fig. 2) closely resembles the phenotype of CENP-E depletion (27, 29, 33, 47, 50). First, 80 to 90% of living mitotic HeLa cells exhibits misaligned polar chromosomes upon repression of mitosin (Fig. 6 and data not shown) or CENP-E by RNA interference (45). Second, depletion of either protein results in increased incidence of spindle distortion and decreased tension across the sister kinetochore (Fig. 2 and 3) (29, 50). Third, kinetochores of the polar chromosomes in either type of cell lack stable microtubules, as demonstrated by electron microscopy studies (29, 33) or cold treatment (Fig. 3B). These similarities, together with the significant reduction of kinetochore CENP-E in mitosin-depleted cells (Fig. 4) and the potential interaction between the two proteins (2, 50), suggest that both proteins function in the same pathway for full chromosome alignment. Since neither protein is essential for congression of most chromosomes, they may be involved in initial processes of the kinetochore-microtubule interaction (29, 33).
Mitosin, however, is not functionally identical to CENP-E. For instance, mitosin-depleted cells hardly divided, even in the presence of only one misaligned chromosome (Fig. 6C; supplementary video 3; data not shown). Instead, they tended to undergo cell death after entering mitosis for 3 to 4 h. In contrast, the majority of CENP-E-depleted HeLa cells enter anaphase after a prolonged mitotic delay, the length of which, usually 2.5 to 8 h, correlates positively with the number of misaligned chromosomes (45). Only about 10% cells die during or after mitosis (45). Furthermore, CENP-E is an activator of BubR1 kinase, and its depletion suppresses the autophosphorylation of BubR1 (26, 45). In contrast, mitosin depletion did not affect BubR1 autophosphorylation (data not shown).
Mitosin facilitates proper microtubule-kinetochore attachments. Our current data suggest that mitosin helps to establish proper microtubule-kinetochore interactions. The high incidence of polar chromosome formation and transient microtubule association of polar chromosomes (Fig. 2 and 3) strongly suggest defects in efficient microtubule-kinetochore interactions and/or stabilization of such interactions. The slowed congression (Fig. 6) provided another support. In fact, CENP-E is involved in stabilization of microtubule attachment (29, 33). Mitosin may serve as a partner in this process. On the other hand, during early mitosis, errors such as syntelic and merotelic attachments can occur (3, 21). They must be corrected to allow normal congression and chromatid segregation, guaranteeing genome stability (3, 21). We did find syntelic attachment in at least 23% of polar chromosomes (Fig. 2B), which, if not destabilized, may hinder proper attachment. Mitosin depletion may thus lead to defects in a mechanism for the correction of such errors. Apparently, more work needs to be done to clarify this issue, since syntelic attachment may also be a result of polar localization of chromosomes where microtubules from the same pole exist in high density.
Mitosin stabilizes kinetochore CENP-E and cytoplasmic dynein and dynactin against microtubule-dependent stripping. Mitosin is apparently not required for kinetochore targeting of dynactin and dynein or CENP-E. Instead, it appears to facilitate their retentions on kinetochores. Kinetochore dynein and dynactin are sensitive to microtubule attachment and mainly dissociate in a microtubule-dependent manner (6, 10, 20). Microtubule disassembly even increases kinetochore dynein by 20- to 100-fold (10). Kinetochore CENP-E also exhibits similar but smaller microtubule-dependent oscillations (10). In mitosin-depleted cells, the steady-state levels of the kinetochore CENP-E and dynactin were further reduced (Fig. 4 and 5). Since their levels were largely rescued after microtubule disassembly (Fig. 4 and 5), the reduced intensities in the presence of microtubules are attributed to overstripping upon microtubule attachment. Such alterations might affect the normal functions of CENP-E and dynein and dynactin in the microtubule-kinetochore interaction and tension generation (4, 8, 42), contributing to decreased congression rates, the formation of polar chromosomes, and malorientation of aligned kinetochores.
Although kinetochore CENP-E can be transported to spindle poles by dynein/dynactin (13), its reduction in mitosin-depleted cells is unlikely due to increased dissociations of dynein/dynactin. Instead, these two events appear to be independent, since inactivation of dynein/dynactin does not influence kinetochore CENP-E and vice versa (6, 50). Furthermore, the kinetochores BubR1 and Mad2, which are also transported by dynein/dynactin (12, 13), were not affected in mitosin-depleted cells (Fig. 2C and data not shown). Mitosin may stabilize kinetochore CENP-E through direct interaction (2, 50). Transport of kinetochore mitosin by dynein/dynactin (49) suggests a certain association between them, either directly or indirectly. Mitosin is independently found to bind NudEl and NudE, two homologous regulators of dynein, in yeast two-hybrid screens (data not shown) (7, 22), which may provide clues to link mitosin to dynein/dynactin.
In addition to mitosin, Spc24 and Spc25, two components of the Ndc80/hNuf2 complex essential for stable microtubule-kinetochore attachment (5, 27), also affect retentions of kinetochore dynein/dynactin and CENP-E (28). Silencing either Spc24 or Spc25 in HeLa cells attenuates kinetochore dynactin and CENP-E only in the presence of microtubules without affecting CENP-F (mitosin) (28). Distinct from CENP-E and mitosin, which are not critical for maintenance of the kinetochore-microtubule attachment (Fig. 2 and 3) (29, 33), Spc24 and Spc25 are required for both establishment and maintenance of the attachment (28). Moreover, mitotic cells lacking any constituent of the Ndc80/Nuf2 complex manifested long microtubule bundles across the spindle and longer spindles (5, 27, 28). Similar phenotypes were also seen in mitosin-depleted cells (Fig. 3A), further supporting a role of mitosin in facilitating microtubule-kinetochore interactions. Moreover, these results provide interesting insights into possible relationships among the proteins involved in microtubule-kinetochore interactions.
Mitosin depletion induces mitotic cell death correlated with premature chromosome decondensation. The way in which mitosin-depleted cells die is also intriguing. First, the cells usually died before anaphase onset, according to live cell recordings in the absence and presence of MG132 and spindle morphologies found in dying cells (Fig. 6 to 8; supplemental videos 2 and 3). The dying cells contained activated caspase-3, indicating apoptosis (32). The incidence of cell death is high: the majority of mitotic cells showed signs of cell death within 4 h of monitoring (Fig. 6; Fig. 8; data not shown). Possibly, a portion of the cells may survive by entering G1 abnormally (27) and exhibit multipolar spindles in the next round of mitosis (Fig. 3). Second, cells in early stages of death showed several unique features. They contained various extent of decondensed chromosomes. The spindle was not disassembled in these cells because they were seen even in dying cells with mininuclei (Fig. 7). Moreover, centromeres, which were labeled by the CREST antigen together with spot-like chromatin DNA, were excluded from the decondensed chromatin (Fig. 7), suggesting that the decondensation occurs mainly at chromosome arms. These phenotypes, which have not been previously reported to our knowledge, may be unique to mitosin depletion. Alternatively, they might be common for mitotic cell death, a poorly studied type of cell death.
The aberrant chromosome decondensation appears to be the major cause of cell death upon mitosin depletion. The majority of CENP-E-depleted HeLa cells by RNA interference initiated anaphase after the pseudometaphase block for 2.5 to 8 h (45), while treatment with antisense RNA arrested cells with condensed chromosomes for more than 20 h (50). The relatively early onset of mitotic cell death (2 to 4 h) (Fig. 6) thus strongly suggests that death is a direct result of mitosin deficiency instead of pseudometaphase arrest. This is further corroborated by cell death in the presence of MG132, since control cells showed only metaphase arrest (Fig. 8). Moreover, death always occurred following progressive chromosome decondensation (Fig. 6 to 8; see videos S2 and S3 in the supplemental material). The relatively normal spindle morphologies and the spindle associations of centromeres (Fig. 7) suggest that the mitotic apparatus is still functional even in dying cells. Anaphase onset was not observed prior to cell death in live cells, suggesting that the spindle checkpoint is not inactivated. The MG132 block (Fig. 8) further confirmed that chromosome decondensation and cell death were not due to aberrant anaphase initiation. It is therefore possible that such premature chromosome decondensation accidentally interrupted mitotic cell homeostasis, leading to cell death by activating a checkpoint that monitors the orchestration of mitotic events.
Our results thus suggest a role of mitosin in chromosome decondensation, a process that is normally initiated from telophase for reassembly of nuclei in daughter cells (36). Silencing mitosin seems to decouple chromosome decondensation from M-phase progression and trigger this event oddly before anaphase (Fig. 6 to 8). In fact, in intact cells, the majority of mitosin is rapidly degraded at the end of mitosis (23, 52), coincident with the timing of nucleus reassembly. Whether mitosin degradation controls the timing of chromosome decondensation is also an interesting issue deserving further investigation.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by grants 30025021, 30330330, and 30421005 from the National Science Foundation of China, KSCX2-2-02 from the Chinese Academy of Sciences, and 2002CB713802 (the national key basic research and development program) from the Ministry of Science and Technology of China.
|
FOOTNOTES |
|---|
Supplemental material for this article may be found at http://mcb.asm.org.
These authors contributed equally to this work.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2. Chan, G. K., B. T. Schaar, and T. J. Yen. 1998. Characterization of the kinetochore binding domain of CENP-E reveals interactions with the kinetochore proteins CENP-F and hBUBR1. J. Cell Biol. 143:49-63.
3. Cimini, D., B. Moree, J. C. Canman, and E. D. Salmon. 2003. Merotelic kinetochore orientation occurs frequently during early mitosis in mammalian tissue cells and error correction is achieved by two different mechanisms. J. Cell Sci. 116:4213-4225.
4. 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]
5. DeLuca, J. G., B. Moree, J. M. Hickey, J. V. Kilmartin, and E. D. Salmon. 2002. hNuf2 inhibition blocks stable kinetochore-microtubule attachment and induces mitotic cell death in HeLa cells. J. Cell Biol. 159:549-555.
6. Echeverri, C. J., B. M. Paschal, K. T. Vaughan, and R. B. Vallee. 1996. Molecular characterization of the 50-kD subunit of dynactin reveals function for the complex in chromosome alignment and spindle organization during mitosis. J. Cell Biol. 132:617-633.
7. Feng, Y., E. C. Olson, P. T. Stukenberg, L. A. Flanagan, M. W. Kirschner, and C. A. Walsh. 2000. LIS1 regulates CNS lamination by interacting with mNudE, a central component of the centrosome. Neuron 28:665-679.[CrossRef][Medline]
8. Heald, R. 2000. Motor function in the mitotic spindle. Cell 102:399-402.[CrossRef][Medline]
9. Hirokawa, N. 1998. Kinesin and dynein superfamily proteins and the mechanism of organelle transport. Science 279:519-526.
10. Hoffman, D. B., C. G. Pearson, T. J. Yen, B. J. Howell, and E. D. Salmon. 2001. Microtubule-dependent changes in assembly of microtubule motor proteins and mitotic spindle checkpoint proteins at PtK1 kinetochores. Mol. Biol. Cell. 12:1995-2009.
11. Hong, S., H. Huo, J. Xu, and K. Liao. 2004. Insulin-like growth factor-1 receptor signaling in 3T3-L1 adipocyte differentiation requires lipid rafts but not caveolae. Cell Death Differ. 11:714-723.[CrossRef][Medline]
12. 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.
13. Howell, B. J., B. F. McEwen, J. C. Canman, D. B. Hoffman, E. M. Farrar, C. L. Rieder, and E. D. Salmon. 2001. Cytoplasmic dynein/dynactin drives kinetochore protein transport to the spindle poles and has a role in mitotic spindle checkpoint inactivation. J. Cell Biol. 155:1159-1172.
14. Howell, B. J., B. Moree, E. M. Farrar, S. Stewart, G. Fang, and E. D. Salmon. 2004. Spindle checkpoint protein dynamics at kinetochores in living cells. Curr. Biol. 14:953-964.[CrossRef][Medline]
15. Hussein, D., and S. S. Taylor. 2002. Farnesylation of Cenp-F is required for G2/M progression and degradation after mitosis. J. Cell Sci. 115:3403-3414.
16. Jiang, W., and T. Hunter. 1998. Analysis of cell-cycle profiles in transfected cells using a membrane-targeted GFP. BioTechniques 24:349-350. 352:354.
17. Johnson, V. L., M. I. Scott, S. V. Holt, D. Hussein, and S. S. Taylor. 2004. Bub1 is required for kinetochore localization of BubR1, Cenp-E, Cenp-F and Mad2, and chromosome congression. J. Cell Sci. 117:1577-1589.
18. Joseph, J., S. T. Liu, S. A. Jablonski, T. J. Yen, and M. Dasso. 2004. The RanGAP1-RanBP2 complex is essential for microtubule-kinetochore interactions in vivo. Curr. Biol. 14:611-617.[CrossRef][Medline]
19. Kanda, T., K. F. Sullivan, and G. M. Wahl. 1998. Histone-GFP fusion protein enables sensitive analysis of chromosome dynamics in living mammalian cells. Curr. Biol. 8:377-385.[CrossRef][Medline]
20. King, J. M., T. S. Hays, and R. B. Nicklas. 2000. Dynein is a transient kinetochore component whose binding is regulated by microtubule attachment, not tension. J. Cell Biol. 151:739-748.
21. Kline-Smith, S. L., A. Khodjakov, P. Hergert, and C. E. Walczak. 2004. Depletion of centromeric MCAK leads to chromosome congression and segregation defects due to improper kinetochore attachments. Mol. Biol. Cell. 15:1146-1159.
22. Liang, Y., W. Yu, Y. Li, Z. Yang, X. Yan, Q. Huang, and X. Zhu. 2004. Nudel functions in membrane traffic mainly through association with Lis1 and cytoplasmic dynein. J. Cell Biol. 164:557-566.
23. Liao, H., R. J. Winkfein, G. Mack, J. B. Rattner, and T. J. Yen. 1995. CENP-F is a protein of the nuclear matrix that assembles onto kinetochores at late G2 and is rapidly degraded after mitosis. J. Cell Biol. 130:507-518.
24. Liu, S. T., J. C. Hittle, S. A. Jablonski, M. S. Campbell, K. Yoda, and T. J. Yen. 2003. Hum. CENP-I specifies localization of CENP-F, MAD1 and MAD2 to kinetochores and is essential for mitosis. Nat. Cell Biol. 5:341-345.[CrossRef][Medline]
25. Maney, T., L. M. Ginkel, A. W. Hunter, and L. Wordeman. 2000. The kinetochore of higher eucaryotes: a molecular view. Int. Rev. Cytol. 194:67-131.[Medline]
26. Mao, Y., A. Abrieu, and D. W. Cleveland. 2003. Activating and silencing the mitotic checkpoint through CENP-E-dependent activation/inactivation of BubR1. Cell 114:87-98.[CrossRef][Medline]
27. Martin-Lluesma, S., V. M. Stucke, and E. A. Nigg. 2002. Role of Hec1 in spindle checkpoint signaling and kinetochore recruitment of Mad1/Mad2. Science 297:2267-2270.
28. McCleland, M. L., M. J. Kallio, G. A. Barrett-Wilt, C. A. Kestner, J. Shabanowitz, D. F. Hunt, G. J. Gorbsky, and P. T. Stukenberg. 2004. The vertebrate Ndc80 complex contains Spc24 and Spc25 homologs, which are required to establish and maintain kinetochore-microtubule attachment. Curr. Biol. 14:131-137.[CrossRef][Medline]
29. McEwen, B. F., G. K. Chan, B. Zubrowski, M. S. Savoian, M. T. Sauer, and T. J. Yen. 2001. CENP-E is essential for reliable bioriented spindle attachment, but chromosome alignment can be achieved via redundant mechanisms in mammalian cells. Mol. Biol. Cell. 12:2776-2789.
30. Mills, J. C., N. L. Stone, and R. N. Pittman. 1999. Extranuclear apoptosis. The role of the cytoplasm in the execution phase. J. Cell Biol. 146:703-708.
31. Musacchio, A., and K. G. Hardwick. 2002. The spindle checkpoint: structural insights into dynamic signalling. Nat. Rev. Mol. Cell. Biol. 3:731-741.[CrossRef][Medline]
32. Porter, A. G., and R. U. Janicke. 1999. Emerging roles of caspase-3 in apoptosis. Cell Death Differ. 6:99-104.[CrossRef][Medline]
33. Putkey, F. R., T. Cramer, M. K. Morphew, A. D. Silk, R. S. Johnson, J. R. McIntosh, and D. W. Cleveland. 2002. Unstable kinetochore-microtubule capture and chromosomal instability following deletion of CENP-E. Dev. Cell. 3:351-365.[CrossRef][Medline]
34. Rattner, J. B., A. Rao, M. J. Fritzler, D. W. Valencia, and T. J. Yen. 1993. CENP-F is a . ca 400 kDa kinetochore protein that exhibits a cell-cycle dependent localization. Cell Motil. Cytoskeleton 26:214-226.[CrossRef][Medline]
35. Rieder, C. L. 1981. The structure of the cold-stable kinetochore fiber in metaphase PtK1 cells. Chromosoma 84:145-158.[CrossRef][Medline]
36. Rieder, C. L., and A. Khodjakov. 2003. Mitosis through the microscope: advances in seeing inside live dividing cells. Science 300:91-96.
37. Rieder, C. L., and E. D. Salmon. 1998. The vertebrate cell kinetochore and its roles during mitosis. Trends Cell Biol. 8:310-318.[CrossRef][Medline]
38. Robinson, J. T., E. J. Wojcik, M. A. Sanders, M. McGrail, and T. S. Hays. 1999. Cytoplasmic dynein is required for the nuclear attachment and migration of centrosomes during mitosis in Drosophila. J. Cell Biol. 146:597-608.
39. Salina, D., P. Enarson, J. B. Rattner, and B. Burke. 2003. Nup358 integrates nuclear envelope breakdown with kinetochore assembly. J. Cell Biol. 162:991-1001.
40. Shah, J. V., and D. W. Cleveland. 2000. Waiting for anaphase: Mad2 and the spindle assembly checkpoint. Cell 103:997-1000.[CrossRef][Medline]
41. Sharp, D. J., G. C. Rogers, and J. M. Scholey. 2000. Cytoplasmic dynein is required for poleward chromosome movement during mitosis in Drosophila embryos. Nat. Cell Biol. 2:922-930.[CrossRef][Medline]
42. Sharp, D. J., G. C. Rogers, and J. M. Scholey. 2000. Microtubule motors in mitosis. Nature 407:41-47.[CrossRef][Medline]
43. Steensgaard, P., M. Garre, I. Muradore, P. Transidico, E. A. Nigg, K. Kitagawa, W. C. Earnshaw, M. Faretta, and A. Musacchio. 2004. Sgt1 is required for human kinetochore assembly. EMBO Rep. 5:626-631.[CrossRef][Medline]
44. Sui, G., C. Soohoo, B. Affar el, F. Gay, Y. Shi, and W. C. Forrester. 2002. A DNA vector-based RNAi technology to suppress gene expression in mammalian cells. Proc. Natl. Acad. Sci. USA 99:5515-5520.
45. Tanudji, M., J. Shoemaker, L. L'Italien, L. Russell, G. Chin, and X. M. Schebye. 2004. Gene silencing of CENP-E by siRNA in HeLa Cells leads to missegregation of chromosomes following a mitotic delay. Mol. Biol. Cell. 15:3771-3781.
46. Vallee, R. B., J. C. Williams, D. Varma, and L. E. Barnhart. 2004. Dynein: An ancient motor protein involved in multiple modes of transport. J. Neurobiol. 58:189-200.[CrossRef][Medline]
47. Weaver, B. A., Z. Q. Bonday, F. R. Putkey, G. J. Kops, A. D. Silk, and D. W. Cleveland. 2003. Centromere-associated protein-E is essential for the mammalian mitotic checkpoint to prevent aneuploidy due to single chromosome loss. J. Cell Biol. 162:551-563.
48. Yan, X., F. Li, Y. Liang, Y. Shen, X. Zhao, Q. Huang, and X. Zhu. 2003. Hum. Nudel and NudE as regulators of cytoplasmic dynein in poleward protein transport along the mitotic spindle. Mol. Cell. Biol. 23:1239-1250.
49. Yang, Z. Y., J. Guo, N. Li, M. Qian, S. N. Wang, and X. L. Zhu. 2003. Mitosin/CENP-F is a conserved kinetochore protein subjected to cytoplasmic dynein-mediated poleward transport. Cell Res. 13:275-283.[CrossRef][Medline]
50. Yao, X., A. Abrieu, Y. Zheng, K. F. Sullivan, and D. W. Cleveland. 2000. CENP-E forms a link between attachment of spindle microtubules to kinetochores and the mitotic checkpoint. Nat. Cell Biol. 2:484-491.[CrossRef][Medline]
51. Zhu, X. 1999. Structural requirements and dynamics of mitosin-kinetochore interaction in M phase. Mol. Cell. Biol. 19:1016-1024.
52. Zhu, X., K. H. Chang, D. He, M. A. Mancini, W. R. Brinkley, and W. H. Lee. 1995. The C terminus of mitosin is essential for its nuclear localization, centromere/kinetochore targeting, and dimerization. J. Biol. Chem. 270:19545-19550.
53. Zhu, X., L. Ding, and G. Pei. 1997. Carboxyl terminus of mitosin is sufficient to confer spindle pole localization. J. Cell Biochem. 66:441-449.[CrossRef][Medline]
54. Zhu, X., M. A. Mancini, K. H. Chang, C. Y. Liu, C. F. Chen, B. Shan, D. Jones, T. L. Yang-Feng, and W. H. Lee. 1995. Characterization of a novel 350-kilodalton nuclear phosphoprotein that is specifically involved in mitotic-phase progression. Mol. Cell. Biol. 15:5017-5029.[Abstract]
This article has been cited by other articles:
