MCB
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental material
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Liu, X.
Right arrow Articles by Erikson, R. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Liu, X.
Right arrow Articles by Erikson, R. L.
Molecular and Cellular Biology, June 2005, p. 4993-5010, Vol. 25, No. 12
0270-7306/05/$08.00+0     doi:10.1128/MCB.25.12.4993-5010.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

CCT Chaperonin Complex Is Required for the Biogenesis of Functional Plk1{dagger}

Xiaoqi Liu,* Chin-Yo Lin,{ddagger} Ming Lei, Shi Yan, Tianhua Zhou, and Raymond L. Erikson

Department of Molecular and Cellular Biology, Harvard University, Cambridge, Massachusetts 02138

Received 3 January 2005/ Returned for modification 1 February 2005/ Accepted 15 March 2005


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Experiments from several different organisms have demonstrated that polo-like kinases are involved in many aspects of mitosis and cytokinesis. Here, we provide evidence to show that Plk1 associates with chaperonin-containing TCP1 complex (CCT) both in vitro and in vivo. Silencing of CCT by use of RNA interference (RNAi) in mammalian cells inhibits cell proliferation, decreases cell viability, causes cell cycle arrest with 4N DNA content, and leads to apoptosis. Depletion of CCT in well-synchronized HeLa cells causes cell cycle arrest at G2, as demonstrated by a low mitotic index and Cdc2 activity. Complete depletion of Plk1 in well-synchronized cells also leads to G2 block, suggesting that misfolded Plk1 might be responsible for the failure of CCT-depleted cells to enter mitosis. Moreover, partial depletion of CCT or Plk1 leads to mitotic arrest. Finally, the CCT-depleted cells reenter the cell cycle upon reintroduction of the purified constitutively active form of Plk1, indicating that Plk1 might be a CCT substrate.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
It is now widely accepted that cancer arises at least partly due to perturbation of normal cell cycle progression, in which reversible protein phosphorylation plays an important regulatory role. Central to this regulation are the cyclin-dependent kinases and their activating proteins, the cyclins (22). The mitotic cyclin-dependent kinase complex consists of a Cdc2 catalytic subunit and a cyclin B regulatory subunit. The kinase activity of Cdc2/cyclin B is regulated both by the stability of cyclin B and by phosphorylation/dephosphorylation of the catalytic subunit. Cyclin B-associated Cdc2 undergoes inhibitory phosphorylation at Thr14 and Tyr15 by Wee1 and Myt1 protein kinases during interphase; the inhibitory phosphates are removed by the activating phosphatase Cdc25C at entry into mitosis (22, 23).

In addition to cyclin-dependent kinases, the polo-like kinase (plk) family has also emerged as a key player in many cell cycle-related events. Genetic and biochemical experiments with several different organisms have documented that plk's are involved in many aspects of mitosis. In addition to the N-terminal kinase domain, all plk family members, including mammalian Plk1, Snk, and Fnk/Prk; Xenopus laevis Plx1; Drosophila melanogaster polo; fission yeast Plo1; and budding yeast Cdc5, have a distinctive highly conserved region in the C-terminal noncatalytic domain, designated the polo box (3, 10, 22). At the onset of mitosis, Xenopus Plx1 phosphorylates and activates Cdc25C, which subsequently activates Cdc2. Thus, Plk1 was proposed to be a trigger kinase for cells to enter mitosis (14). This finding was further supported by a series of experiments with the Xenopus oocyte system (1, 12, 17, 27, 28, 29). Microinjection of mRNA encoding constitutively active Plx1 (S128D/T201D) into Xenopus oocytes directly induced the activation of both Cdc25C and Cdc2/cyclin B activity. Conversely, either microinjection of anti-Plx1 antibody into oocytes or immunodepletion of Plx1 in cycling egg extracts prevented the activation of Cdc25C and Cdc2/cyclin B. Reversal of the antibody-mediated inhibition in vitro by introduction of active Cdc25C suggests that Plx1 acts upstream of Cdc25C. Moreover, injection of active Cdc25C, which directly activates Cdc2/cyclin B, also caused the activation of Plx1. Thus, it was proposed that Plx1 is capable of activating Cdc25C and being a component of the positive-feedback loop that activates Cdc2/cyclin B at the G2/M-phase transition. Recently, it was proposed that phosphorylation of Cdc25C by Cdc2/cyclin B might generate a docking site for the Plk1 polo-box domain to facilitate subsequent activation of Cdc25C by Plk1 (8, 9). However, the idea that Plk1 is a trigger kinase to activate Cdc25C prior to the activation of Cdc2/cyclin B was also challenged by the identification of the upstream regulators of Plk1 in starfish meiotic and early embryonic cycles (25). Distinct kinases, Cdc2/cyclin B and mitogen-activated protein kinase, along with Cdc2/cyclin B or cyclin A, and Cdc2/cyclin A, are unique upstream components that contribute to Plk1 activation at meiosis I, meiosis II, and embryonic M phase, respectively, suggesting that Plk1 is not the trigger kinase at reinitiation of meiosis I. At entry into meiosis II and M phase of the first cleavage cycle, Plk1 activity is required for Cdc2/cyclin B activation, but primarily through suppression of Myt1, rather than through activation of Cdc25. In line with the above observation, Myt1 was recently shown to be a Plk1 substrate both in vitro and in vivo (21).

Analyses of both Drosophila polo1 and fission yeast Plo1 mutants show that they display a phenotype of monopolar spindles, indicating a role for plk in centrosome assembly and separation during the formation of bipolar mitotic spindles (24, 38). More recent data derived from two strongly hypomorphic mutations of Drosophila polo (polo9 and polo10) revealed a requirement for polo in the metaphase-anaphase transition and also provided additional insight into the requirement for plk to organize microtubule nucleating centers (6). The weakly hypomorphic polo1 mutant is able to progress through multiple cell cycles, whereas the strongly hypomorphic mutations polo9 and polo10 completely block the proliferation of diploid tissues. This is an example of different phenotypes observed based on the different degrees of polo deficiency. An essential function of polo-like kinase in late mitosis has also been implicated in different organisms. In budding yeast, ectopic expression of Cdc5 causes the formation of additional septin ring structures (16), and an obvious cytokinetic defect was observed in cells overexpressing the polo-box domain of Cdc5, presumably due to a dominant-negative effect (34). Drosophila polo is also required for cytokinesis (5). Cytokinetic defects in the premeiotic divisions are revealed by reduced numbers of ring canals and by enlarged cells in cysts of primary spermatocytes.

Compared to other systems, functional study of mammalian Plk1 has been limited, presumably due to technical difficulties. In an early study, it was shown that microinjection of anti-Plk1 antibody into HeLa cells caused mitotic arrest (15). These cells had abnormal distributions of condensed chromatin and monoastral microtubule arrays that were nucleated from duplicated but unseparated centrosomes. The drastic size reduction of centrosomes and impairement of {gamma}-tubulin and MPM-2 immunoreactivity in Plk1 antibody-injected cells led the authors to conclude that Plk1 activity is required for the functional maturation of centrosomes in late G2/early prophase and consequently for the establishment of a bipolar spindle. We might point out that, although Plk1 antibody microinjection into HeLa cells caused mitotic arrest, presumably due to inhibition of anaphase-promoting complex (APC), a fraction of injected cells arrested at G2, similar to the phenotypes described below for partial ablation of Plk1 by RNA interference (RNAi) and probably reflecting the stage of the cell cycle in which the cell was at the time the antibody was injected. To investigate the function of Plk1 in mammalian cells in more detail, we and others recently reported the phenotypes of cultured cells after Plk1 depletion, by using either direct transfection of 21-nucleotide double-stranded RNA or vector-based RNA interference (18, 19, 35, 37, 42). We found that short-term Plk1 depletion in unsynchronized HeLa cells (48 h posttransfection) results in the stabilization of cyclin B and thus elevated Cdc2 activity. This indicates that Plk1 might be dispensable for mitotic entry of HeLa cells after short-term depletion conditions but essential for the activation of anaphase-promoting complex, whose activation leads to the degradation of cyclin B (13). The fact that Plk1-depleted HeLa cells show the formation of a dumbbell-like DNA organization suggests that sister chromatids are not completely separated in these cells, in agreement with the requirement of Cdc5 activity for cohesin cleavage at the onset of anaphase in budding yeast (2). Phenotype analysis based on the vector-based RNAi approach reveals the essential function of Plk1 for cell proliferation and viability. Plk1 depletion in HeLa cells over an extended time period induced apoptosis, as shown by the appearance of subgenomic DNA in fluorescence-activated cell sorting (FACS) profiles, the activation of caspase 3, and the formation of fragmented nuclei. We further showed that DNA damage occurred and p53 was stabilized in Plk1-depleted HeLa cells (19). More recent analysis using RNAi targeting Plk1 indicated that Plk1 depletion activates the spindle checkpoint (37, 42).

In a search for proteins that interact with Plk1 using a yeast two-hybrid system, we identified a protein component of the 900-kDa complex, the chaperonin-containing TCP1, or CCT. Chaperonins are a family of proteins involved in the assistance of the folding of other proteins. The best-characterized chaperonins are those present in eubacteria (GroEL/GroES) and in mitochondria and chloroplasts of eukaryotes. GroEL consists of two homoheptameric rings and works with a cochaperonin, GroES. The small conformational rearrangement of GroEL upon ATP binding allows binding of GroES, which caps the chaperonin cavity after the large structural changes (31). Multiple folding pathways have been proposed in the eukaryotic cytosol. Newly translated proteins may require the assistance of hsp70, hsp90, or CCT. CCT has a cylinder-like architecture, two back-to-back stacked rings, each one enclosing a cavity where protein folding takes place. Unlike the processes observed with GroEL, ATP binding to CCT generates large conformational changes that close the ring cavity (11). Genetic and biochemical studies show that CCT is required for the folding of the cytoskeletal proteins actin and tubulin. Recent reports show that CCT is also required for folding of other proteins (7). Indeed, as many as 15% of newly translated proteins may need the assistance of CCT for correct folding (41). Most recently, the CCT chaperonin was shown to promote activation of the APC through the generation of functional Cdc20 (4). Cdc20 forms a stable complex with CCT, and the release of Cdc20 from the CCT depends on ATP hydrolysis. The folding of the Cdc20 family of APC activators was proposed as an essential function of the CCT chaperonin. Previous reports show that the stability of Plk1 decreases upon treatment with geldanamycin, a specific inhibitor of hsp90. Furthermore, hsp90 regulates Plk1 stability through its interaction with the polo-box domain (32). Whether Plk1 folding also involves CCT is the key question we will address in this paper.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Vector construction. To specifically deplete endogenous CCT{alpha} in HeLa cells, plasmid pBS/U6-CCT{alpha} was constructed as described previously (36). The targeting sequence of human CCT{alpha} (accession no. NM_030752) was GGGAGAAGTCAAATGGAGAGT, corresponding to the coding region 604 to 624 relative to the first nucleotide of the start codon. Plasmid pBS/U6-CCT{alpha}-1st half (sense strand) was used as a control vector. This control vector produces RNA that cannot form a hairpin structure to generate interfering RNA. Plasmid pBS/U6-Plk1 was described previously (19). Plasmids pBS/U6-GFP-CCT{alpha} and pBS/U6-GFP-Plk1, targeting the same sequences as those of pBS/U6-CCT{alpha} and pBS/U6-Plk1, express green fluorescent protein (GFP) independently of hairpin RNAs, so only GFP-positive cells express small interfering RNAs (siRNAs).

Cell culture and synchronization. HeLa cells were maintained in Dulbecco modified Eagle medium (DMEM) supplemented with 10% (vol/vol) fetal bovine serum, 100 units/ml penicillin, and 100 units/ml streptomycin at 37°C in 8% CO2. FT210 cells were maintained in RPMI medium supplemented with 10% (vol/vol) fetal bovine serum, 100 units/ml penicillin, and 100 units/ml streptomycin at 33°C in 8% CO2. To synchronize HeLa cells, cells were treated with 2.5 mM thymidine for 16 h, released for 8 h, and then treated with thymidine a second time for 16 h. After two washes with phosphate-buffered saline (PBS), cells were cultured for different times as indicated in each experiment and harvested. Based on our experience, cells accumulate at G2 phase after 8 h of release into normal medium, most cells arrest at mitosis after 12 h of release in the presence of 100 ng/ml of nocodazole, and 13.5 h of release in the absence of nocodazole results in at least 50% of cells at telophase/cytokinesis.

DNA transfections. For phenotype analysis of gene depletion in randomly growing cells, HeLa cells were cotransfected with pBS/U6-CCT{alpha} or -Plk1 and pBabe-puro at a ratio of 9:1 using GenePorter reagents. After 2 days of selection of transfection-positive cells with 2 µg/ml puromycin, floating cells were washed away with PBS, and the attached cells were incubated until harvesting for phenotype analysis. To deplete CCT in well-synchronized cells, HeLa cells were cotransfected with pBS/U6-CCT{alpha} and pBabe-puro first, incubated for about 30 h, and subjected to double thymidine block in the presence of puromycin. After synchronization/selection, the floating cells were removed, and the remaining cells were released into fresh medium containing 100 ng/ml nocodazole for different times. A similar protocol was used to deplete Plk1 in well-synchronized cells: HeLa cells were cotransfected with pBS/U6-Plk1 and pBabe-puro first, incubated for about 5 h, and treated with thymidine-containing medium for 16 h. Puromycin was included during the 8-h release interval, and cells were blocked with a second thymidine for 16 h. After synchronization/selection, the surviving cells were released into fresh medium for different times.

In vitro translation. Hemagglutinin (HA)- or Flag-tagged Plk1 constructs were translated in a reticulocyte lysate system (Promega) for 30 min at 30°C in the presence of [35S]methionine. After the reactions were terminated with cycloheximide and RNase A, the products were immunoprecipitated with anti-CCT{alpha} antibody and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

Immunoprecipitation (IP) and immunoblotting. HeLa or FT210 cells were lysed in TBSN buffer (20 mM Tris, pH 8.0, 150 mM NaCl, 1.5 mM EDTA, 5 mM EGTA, 0.5% Nonidet P-40, 0.5 mM Na3VO4) supplemented with phosphatase and proteinase inhibitors (20 mM p-nitrophenyl phosphate, 1 mM Pefabloc, 10 µg/ml pepstatin A, 10 µg/ml leupeptin, 5 µg/ml aprotinin), and the lysates were clarified by centrifugation at 15,000 x g for 30 min. Cell lysates were incubated with either anti-Plk1 or anti-CCT antibody for 1.5 h at 4°C, followed by 1 h of incubation with protein A-Sepharose beads. Immunocomplexes were resolved by SDS-PAGE, and coimmunoprecipitated proteins were detected by Western blotting using antibodies indicated in the specific experiment.

Immunofluorescence staining. HeLa cells were fixed with paraformaldehyde and permeabilized with methanol. After three washes with 0.1% Triton X-100-PBS, cells were incubated with anti-phospho-histone H3 antibody, followed by incubation with Cy3-conjugated secondary antibody. Finally, DNA was stained with DAPI (4',6'-diamidino-2-phenylindole).

Kinase assay. Cdc2 or Plk1 was immunoprecipitated from cell lysates with the relevant antibody and resuspended in TBMD buffer (50 mM Tris, pH 7.5, 10 mM MgCl2, 5 mM dithiothreitol, 2 mM EGTA, 0.5 mM sodium vanadate, 20 mM p-nitrophenyl phosphate) supplemented with 25 µM ATP and 50 µCi of [{gamma}-32P]ATP. The reaction mixtures were incubated at 30°C for 30 min in the presence of appropriate substrates and resolved by SDS-PAGE. The gels were stained with Coomassie brilliant blue, dried, and subjected to autoradiography.

Protein delivery into CCT-depleted cells. About 0.5 µg of glutathione S-transferase (GST)-Plk1-TD (constitutively active form, purified from baculovirus-infected Hi5 cells) was diluted to 40 µl with PBS and added to dried BioPORTER reagent (Gene Therapy Systems, San Diego, CA). After incubation for 5 min at room temperature, the BioPORTER-protein mixture was diluted to 0.5 ml with serum-free DMEM and transferred to cells covered with 0.5 ml serum-free DMEM. After 4 h of incubation at 37°C, cells were supplied with 1 ml of 20% fetal bovine serum-containing DMEM and incubated until harvesting.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Plk1/CCT interaction. After the identification of CCT as a potential Plk1-interacting protein complex in a yeast two-hybrid screen, we next analyzed the association between Plk1 and CCT in vitro (Fig. 1). HA-tagged Plk1 was translated in a rabbit reticulocyte lysate system in the presence of [35S]methionine, and the reaction products were subjected to anti-CCT{alpha} antibody immunoprecipitation. Only Plk1, but not luciferase, another kinase, Krs, or a Golgi protein, Grasp65, was coimmunoprecipitated with anti-CCT{alpha} antibody, indicating the specificity of association between Plk1 and CCT (Fig. 1A). The association between Plk1 and CCT was independent of Plk1 kinase activity, and furthermore, even a mutant with a frameshift in its polo-box domain still could be coimmunoprecipitated by anti-CCT{alpha} antibody, suggesting that the N-terminal domain of Plk1 interacts with CCT (Fig. 1B). Anti-CCT{alpha} antibody did not recognize GST-Plk1 purified from baculovirus-infected insect cells (Fig. 1C), indicating that CCT{alpha} antibody does not cross-react with Plk1. The specificity of the CCT/Plk1 interaction was also confirmed by using anti-Erk2 antibody for immunoprecipitation (Fig. 1D). Since CCT was initially identified as a Plk1-interacting partner using the C-terminal polo-box domain as a bait in the yeast two-hybrid system (data not shown), we performed more experiments to test whether both the N-terminal kinase domain and the C-terminal polo-box domain are responsible for CCT interaction. The Flag-tagged N-terminal kinase domain or the C-terminal polo-box domain was translated in vitro, and the translation products were subjected to anti-CCT antibody immunoprecipitation. The results indicated that both domains interact with CCT (Fig. 1E). Moreover, careful inspection of anti-CCT coimmunoprecipitation (co-IP) efficiency of translated Plk1 fragments revealed that the N-terminal domain showed higher binding affinity than the C-terminal domain and that the extreme N terminus (the fragment from amino acids 1 to 137) has the highest binding affinity for CCT.



View larger version (55K):
[in this window]
[in a new window]
  		FIG. 1. Association of newly translated Plk1 with CCT. (A) Four different genes, whose expression is under the control of the T7 promoter, were translated in vitro in a rabbit reticulocyte lysate system (Promega) for 30 min at 30°C in the presence of [35S]methionine. The translation products were immunoprecipitated with anti-CCT{alpha} antibody, and the direct translation mixtures (–) or anti-CCT{alpha} IP (+) products were resolved by SDS-PAGE and detected by autoradiography. (B) Four different Plk1 constructs (WT; KM, kinase defective; TD, constitutively active; FS, frameshift at the polo box) were translated in vitro and subjected to anti-CCT{alpha} antibody IP. (C) CCT{alpha} antibody does not cross-react with Plk1.Purified GST-Plk1 was incubated with CCT{alpha} antibody, followed by incubation with protein G-agarose beads, and IP pellets were subjected to anti-GST antibody Western blotting. (D) After Plk1 was translated in vitro as in panel A, the translation products were subjected to either Erk2 or CCT antibody immunoprecipitation, and the IP products were analyzed by SDS-PAGE. (E) Different domains of Plk1 were translated in vitro and subjected to anti-CCT{alpha} antibody IP. (F) HA-Plk1 (either WT or KM) was translated in vitro in the presence of cold methionine and immunoprecipitated with anti-HA antibody. The IP pellets were subjected to either autophosphorylation (top panel) or kinase reactions using casein as a substrate (bottom panel). (G to J) Rabbit reticulocyte lysates were immunodepleted with CCT{alpha} antibody on ice for 1.5 h and subsequently used to translate HA-Plk1 in the presence of [35S]methionine. (G) The lysates were subjected to anti-CCT Western blotting to analyze the depletion efficiency. (H to J) The translation products were either directly separated by SDS-PAGE and detected by autoradiography (H) or immunoprecipitated with anti-HA antibody and subjected to either autophosphorylation (I) or phosphorylation using casein as a substrate (J). (K) Association of Plk1 and CCT in vivo. FT210 cells were either grown randomly at 33°C or treated with 200 ng/ml of nocodazole for 14 h to arrest at mitosis. Lysates from mitotic (M) or interphase (I) FT210 cells were subjected to anti-Plk1 antibody immunoprecipitation first and then analyzed by Western blotting with the indicated antibodies.

 
Next, we assayed the kinase activity of HA-Plk1 translated in vitro. Either wild-type (WT) or kinase-dead mutant (KM) Plk1 was translated in reticulocyte lysates, immunoprecipitated with HA antibody, and subjected to autophosphorylation or kinase reactions using casein as a substrate (Fig. 1F). Both autophosphorylation and casein phosphorylation activities of WT HA-Plk1 were significantly higher than those of the KM mutant, indicating the correcting folding of Plk1 translated in vitro. To test whether the correct folding of newly translated Plk1 requires CCT activity, we treated the reticulocyte lysates with CCT{alpha} antibody before the translation reaction (Fig. 1G). Quantification indicates that immunodepletion of the lysate with CCT antibody only slightly decreased the Plk1 translation efficiency (Fig. 1H). Both autophosphorylation and the activity toward casein of the Plk1 translated in the lysate depleted with CCT antibody were about 25% of control (Fig. 1I and J), suggesting that correct folding of Plk1 might require the assistance of CCT. The residual Plk1 activity in the CCT antibody-treated lysate appears to be due to incomplete depletion (Fig. 1G).

The interaction between Plk1 and CCT was also confirmed in vivo (Fig. 1K). Mouse mammary gland carcinoma FT210 cells, which encode a temperature-sensitive Cdc2 protein kinase (40), have a higher Plk1 expression level. Cells were treated with nocodazole to arrest at mitosis (M), cell lysates were subjected to anti-Plk1 immunoprecipitation, and coimmunoprecipitation of CCT was clearly detected by Western blotting. The failure to detect the association between CCT and Plk1 in interphase lysate is apparently due to the low Plk1 level during interphase.

CCT depletion induces G2/M block and apoptosis. To examine the function of endogenous CCT, we took advantage of the recently developed vector-based RNAi technology to specifically deplete CCT{alpha} in HeLa cells (36). The targeting sequence of human CCT{alpha} is the coding region 604 to 624 relative to the first nucleotide of the start codon. As described previously (19), we have developed a protocol to select the transfection-positive cells, to circumvent the variation of transfection efficiency in different experiments. Briefly, pBabe-puro, which expresses a puromycin resistance gene, was cotransfected with pBS/U6-CCT{alpha} to permit selection of the transfected cells. After 2 days of selection with puromycin, the floating untransfected cells were washed away with PBS, and the attached cells were used for phenotype analysis. As indicated in Fig. 2A, CCT{alpha} was efficiently depleted by siRNA, whereas the level of Erk2 was unchanged. The level of CCT{alpha} protein was reduced by at least 95% 72 h posttransfection, indicating that the vector-based RNAi approach can efficiently deplete CCT{alpha} in mammalian cells. Interestingly, the level of Plk1 was obviously increased in CCT-depleted cells, indicating possible G2/M arrest caused by CCT depletion. The requirement of CCT for cell proliferation and viability was then determined. Transfection with the control vector did not affect the growth rate of HeLa cells, whereas transfection with pBS/U6-CCT{alpha} and pBabe-puro strongly inhibited cell proliferation and decreased cell viability (Fig. 2B and C). To characterize the inhibition of cell growth by CCT{alpha} depletion, cell cycle progression was analyzed by FACS. As shown in Fig. 2D, the transfection of control vectors did not affect the cell cycle profile, whereas CCT{alpha} depletion induced an obvious increase in the percentage of cells with 4N DNA content, suggesting that CCT{alpha}-depleted cells block at G2/M phase. Starting from 4 days posttransfection, CCT-depleted cells showed a significant sub-G1 DNA population, indicating that these cells were undergoing apoptosis. To further characterize this phenotype in CCT-depleted cells, anti-active caspase 3 immunofluorescence staining was performed (Fig. 2E). Caspase 3, the executioner caspase in apoptosis, was clearly activated in CCT-depleted cells.



View larger version (20K):
[in this window]
[in a new window]
  		FIG. 2. Phenotypes of CCT{alpha} depletion. (A) HeLa cells were cotransfected with pBS/U6-CCT{alpha} and pBabe-puro at a ratio of 9:1. After 1 day of incubation, puromycin was added for an additional 2 days to select the transfection-positive cells. Floating cells were removed, attached cells were harvested, and cell lysates were subjected to direct Western blotting using antibodies indicated on the left. (B and C) HeLa cells were transfected and selected with puromycin as in panel A. Cells were harvested at the times indicated, and cell proliferation (B) and viability (C) were monitored. (D) FACS profiles of CCT-depleted cells harvested at the times indicated on the left. The positions of G1, G2/M, and sub-G1 populations are labeled (2N, 4N, and 1N, respectively). (E) Four days posttransfection, CCT-depleted cells were subjected to anti-active caspase 3 antibody immunofluorescence staining.

 
To examine whether CCT depletion-induced apoptosis is due to cell cycle arrest at G2/M phase, we tested the effect of mimosine treatment on CCT-depletion-induced cell death. Mimosine, an iron chelator that inhibits DNA replication, causes cell cycle arrest at G1 (26). HeLa cells were transfected and selected with puromycin as described above, mimosine was added 2 days posttransfection, and cell viability was determined after further incubation. At 4 days posttransfection, 50% of CCT-depleted cells were attached to the plates, whereas 90% of mimosine-treated CCT-depleted cells still attached to the plates. At 5 days posttransfection, 20% of CCT-depleted cells were viable, in striking contrast to 80% viability in the presence of mimosine (see Fig. S1A in the supplemental material). Therefore, CCT depletion-induced apoptosis is due to G2/M arrest.

The increase in the population of cells with subgenomic DNA upon CCT depletion indicates DNA damage. Considering the important role of the kinase ATM in the cellular response to DNA damage, we analyzed its role in these events by using two ATM inhibitors. Caffeine inhibits the catalytic activity of both ATM and ATR (ATM and Rad3 related), whereas wortmannin potently inhibits the function of ATM, far less efficiently than ATR (30). As illustrated in Fig. S1B and C in the supplemental material, treatment of cells with either of these ATM inhibitors strongly potentiated the lethality of CCT depletion. After HeLa cells were transfected and selected with puromycin as described above, either 2 mM caffeine or 5 µM wortmannin was added 2 days posttransfection. At 3 days and 4 days posttransfection, CCT depletion-induced cell death was not very evident, and caffeine or wortmannin treatment alone at these concentrations did not cause significant cell death. However, the presence of caffeine or wortmannin in CCT-depleted cells for 1 day caused 30 to 40% of the cells to undergo apoptosis, and about 50 to 70% of CCT-depleted cells underwent apoptosis after 2 days of incubation with the drug (see Fig. S1 in the supplemental material).

CCT is required for mitotic entry. To distinguish whether CCT depletion induces G2- or M-phase arrest, we established a protocol to couple vector-based RNAi with cell synchronization (Fig. 3A). Briefly, HeLa cells were cotransfected with pBS/U6-CCT{alpha} and pBabe-puro at a 9:1 ratio, incubated for about 30 h, and then subjected to double thymidine block (16 h of thymidine treatment and 8 h of release, followed by a second 16-h thymidine incubation) in the presence of puromycin. After synchronization/selection, the floating cells were washed away with PBS, and the attached cells were then released into fresh medium containing 100 ng/ml nocodazole up to 12 h. Harvested cells were first subjected to Western blotting to examine the depletion efficiency. As shown in Fig. 3B, CCT{alpha} was effectively depleted by this protocol, whereas the level of Erk2 was not significantly affected. Mitotic entry was confirmed by a high level of Plk1, Wee1 degradation, and hyperphosphorylation of Cdc25C, as indicated in the three middle panels of Fig. 3B (compare 12 h to 0 h of control samples). However, CCT depletion significantly decreased the level of Plk1, stabilized Wee1, and inhibited the hyperphosphorylation of Cdc25C (compare CCT-depleted lanes to control lanes at the same time point). In addition, we immunoprecipitated Cdc2/cyclin B from the cell lysates with anti-Cdc2 antibody and measured its kinase activity toward histone H1 (Fig. 3C). The low Cdc2 activity in CCT{alpha}-depleted cells suggested that CCT depletion induces cell cycle arrest at G2. Similar experiments were performed with the cells growing on coverslips, and well-synchronized, CCT{alpha}-depleted cells were subjected to immunofluorescence staining with phospho-H3 antibody, a mitosis marker (Fig. 3D). In line with the stabilization of Wee1, the decreased phosphorylation state of Cdc25C, and reduced Cdc2 activities, only 8% of CCT-depleted cells showed phospho-H3-positive staining whereas up to 50% of control cells were phospho-H3 positive after 12 h of release from double thymidine block (Fig. 3E). Taken together, we conclude that CCT depletion causes cell cycle arrest at G2, not at M phase. As a control experiment, we also measured DNA synthesis in CCT-depleted cells by bromodeoxyuridine (BrdU) labeling after the release from double thymidine block for 4 h (Fig. 3F). CCT-depleted cells still incorporated BrdU efficiently, suggesting that CCT depletion does not affect early phases of the cell cycle. Finally, HeLa cells growing on coverslips were depleted of CCT{alpha} by use of the protocol described in Fig. 3A, released into fresh medium without nocodazole for different times, and subjected to anti-phospho-H3 antibody staining. For control cells, mitotic index reached a peak after 11 h of release from the double thymidine block and then gradually decreased to that of interphase level after 15 h of release. In contrast, less than 10% of CCT-depleted cells showed phospho-H3-positive staining, indicating that the majority of CCT-depleted cells never enter mitosis over the entire release period (Fig. 3G).



View larger version (40K):
[in this window]
[in a new window]
  		FIG. 3. CCT depletion induces G2 arrest. (A) Protocol used to deplete CCT{alpha} in well-synchronized HeLa cells. (B and C) CCT{alpha} was depleted in HeLa cells using the protocol in panel A, and the attached cells were harvested. Cell lysates were prepared and subjected to either direct Western blotting analysis using the antibodies indicated on the left (B) or an anti-Cdc2 immunoprecipitation/kinase assay using histone H1 as a substrate (C). Stars on the right indicate the hyperphosphorylated forms of Cdc25C in panel B. (D to F) HeLa cells growing on coverslips were depleted of CCT{alpha} using the protocol in panel A. (D) Representative images of phospho-H3 immunofluorescent staining 12 h after release. (E) Histograms quantifying the results. (F) Representative images of BrdU labeling after 4 h of release from the double thymidine block. Bars, 10 µm. (G) HeLa cells growing on coverslips were depleted of CCT{alpha} by use of the protocol described in panel A, released into fresh medium without nocodazole for different times, and subjected to anti-phospho-H3 antibody staining.

 
Plk1 is required for mitotic entry. We reasoned that, if CCT is required for Plk1 folding, Plk1-depleted cells would be expected to show phenotypes similar to those of CCT-depleted cells. To test this, we depleted Plk1 in well-synchronized HeLa cells using the protocol shown in Fig. 4A. Briefly, transfected cells were subjected to a first thymidine block at 8 h posttransfection, and puromycin selection started during the 8-h interval. Compared to CCT, the cellular Plk1 level is much lower, and this shortened depletion protocol is sufficient to almost completely (>95%) deplete Plk1 (Fig. 4B). Cdc2 activity of Plk1-depleted cells was significantly lower than that of control cells at the same time points (Fig. 4C). That the majority of Plk1-depleted cells showed 4N DNA content after 12 h of release from the double thymidine block indicates that Plk1 is not required for cell cycle progression through G1/S (Fig. 4D). Moreover, BrdU labeling showed that DNA synthesis was also not affected by Plk1 depletion (data not shown). Finally, the percentage of phospho-H3-positive cells was significantly lower than that of control cells at the same times (Fig. 4E and F). To confirm this observation, HeLa cells stably expressing GFP-histone H2B were also depleted of Plk1 using the same protocol described in Fig. 4A and subjected to anti-lamin A/C antibody staining 11 h after the release from the double thymidine block (Fig. 4G). Both chromosome condensation and nuclear envelope breakdown (NEBD) were clearly detected for control cells, while chromosomes remained decondensed and nuclear envelopes were still intact for Plk1-depleted cells. Thus, under this condition Plk1 depletion prevents cells from entering mitosis. Similar to the experiment described in Fig. 3G, HeLa cells growing on coverslips were depleted of Plk1 by use of the protocol described in Fig. 4A, released into fresh medium without nocodazole for different times, and subjected to anti-phospho-H3 antibody staining. The very low percentage of mitotic index of Plk1-depleted cells over the entire release period further supports the contention that cells do not enter mitosis when Plk1 is completely depleted (Fig. 4H).



View larger version (41K):
[in this window]
[in a new window]
  		FIG. 4. Plk1 is required for mitotic entry. (A) Protocol used to deplete Plk1 in well-synchronized HeLa cells. (B to D) Plk1 was depleted in HeLa cells using the protocol in panel A, and the attached cells were harvested. Cell lysates were prepared and subjected to either direct Western blotting analysis using the antibodies indicated on the left (B) or an anti-Cdc2 immunoprecipitation/kinase assay using histone H1 as a substrate (C). (D) FACS profiles of synchronized cells. (E and F) HeLa cells growing on coverslips were depleted of Plk1 using the protocol in panel A. (E) Representative images of phospho-H3 immunofluorescent staining 12 h after release. (F) Histograms quantifying the results. (G) HeLa cells stably expressing GFP-H2B were depleted of Plk1 using the protocol in panel A and subjected to anti-lamin A/C immunofluorescent staining 11 h after release from the double thymidine block. (H) HeLa cells growing on coverslips were depleted of Plk1 by use of the protocol described in panel A, released into fresh medium without nocodazole for different times, and subjected to anti-phospho-H3 antibody staining. Bars, 10 µm.

 
To confirm that the complete depletion of Plk1 leads to G2 block, we developed different synchronization protocols to achieve different degrees of silencing of Plk1 with vector pBS/U6-GFP-Plk1 (Fig. 5A). This vector, which expresses GFP independently of the shRNA to target Plk1, allows us to follow the behavior of transfection-positive cells without puromycin selection. A GFP signal is easily detected as early as 12 h posttransfection. As shown in Fig. 5B, Plk1 was almost completely depleted after 60 h posttransfection whereas about 50% of Plk1 was still detectable at 40 h posttransfection (Fig. 5B), and the level of Plk1 was only slightly decreased at 30 h posttransfection (data not shown). Figure 5C shows the phospho-H3 immunostaining results of these various protocols. After 60 h posttransfection, only 5% of GFP-positive cells had entered mitosis, compared to 30% of control cells. In contrast, 18% of Plk1-depleted cells were in mitosis after 30 h of depletion (compared to 20% of control cells), and 9% of Plk1-depleted cells were in mitosis after 40 h of depletion (compared to 22% of control cells), indicating a correlation between the level of Plk1 and the ability of cells to enter mitosis.



View larger version (44K):
[in this window]
[in a new window]
  		FIG. 5. Mitotic entry when Plk1 is partially depleted. (A) Three protocols used to deplete Plk1 using vector pBS/U6-GFP-Plk1, which expresses GFP independently of the short hairpin RNA. (B) Western blotting analysis to examine Plk1 depletion efficiency using these protocols. Lane 1: cells were double thymidine blocked and then released for 12 h. Lane 2: Plk1 was depleted as described in Fig. 4A, which is similar to protocol c of panel A. Lane 3: Plk1 was depleted using protocol b of panel A. Briefly, cells were transfected with pBS/U6-Plk1 and pBabe-puro at a ratio of 9:1. After 12 h of thymidine block, puromycin was added for an additional 16 h to select the transfection-positive cells. Cells were released for 12 h before harvest. (C) HeLa cells growing on coverslips were subjected to Plk1 depletion using protocols described in panel A and stained with phospho-H3 antibody. (D to G) Plk1 depletion in unsynchronized HeLa cells. (D) Protocol used to deplete Plk1 in randomly growing cells. (E) Phospho-H3 antibody staining of Plk1-depleted cells using the protocol in panel D. (F and G) Plk1 was depleted in randomly growing HeLa cells using the protocol in panel D and selected with puromycin for 2 days, and the attached cells were harvested for FACS analysis (F) and cyclin B Western blotting and anti-Cdc2 IP/kinase assay (G).

 
Given the essential roles of Plk1 during mitotic progression, it is possible that Plk1 depletion in a heterogeneous population of cells may cause partial mitotic arrest. Thus, the phenotypes of Plk1 depletion in randomly growing cells were examined using the protocol illustrated in Fig. 5D. FACS analysis indicated that 40% of Plk1-depleted cells had 4N DNA content after 3 days of transfection of an unsynchronized population (Fig. 5F). Among them, 15% were phospho-H3 positive, indicating arrest in mitosis, whereas only 3% of control cells were mitotic (Fig. 5E). Moreover, cyclin B was stabilized and Cdc2 activity was increased about twofold in Plk1-deficient cells compared to control cells (Fig. 5G).

To further confirm the observation that Plk1 depletion for a short time using vector-based RNAi did cause cells to enter mitosis (Fig. 5C), we next examined the involvement of Plk1 in mitotic entry using direct transfection of siRNA into well-synchronized HeLa cells stably expressing GFP-tubulin (Fig. 6A). About 30 h posttransfection, cells were analyzed by anti-phospho-H3 staining. As shown in Fig. 6B, Plk1 siRNA-treated cells entered mitosis 2 to 3 h later than control cells and were arrested in mitosis eventually. Most of the control cells reached telophase/cytokinesis as indicated by the tubulin pattern after 11 h of release from the double thymidine block, whereas Plk1 siRNA-treated cells were still in G2 phase (Fig. 6C). After 14 h of release, Plk1 siRNA-treated cells accumulated in mitosis with rosette-like condensed chromosomes (Fig. 6D). Moreover, most of the Plk1 siRNA-treated cells had monopolar spindles (Fig. 6E).



View larger version (27K):
[in this window]
[in a new window]
  		FIG. 6. Normal Plk1 level is not required for mitotic entry. (A) Protocol used to downregulate Plk1 with siRNA in well-synchronized HeLa cells stably expressing GFP-tubulin. (B) After release from the double thymidine block for different times, cells were analyzed by anti-phospho-H3 staining. (C) Representative images of cells after release from the double thymidine block for 11 h. (D) Representativeimages of Plk1-depleted cells after release from the block for 14 h. (E) Typical images of Plk1-depleted cells indicate the formation of monopolar spindles.

 
Mitotic entry when CCT is partially depleted. We reasoned that, if there is a threshold of active Plk1 for mitotic entry, then the partial CCT depletion should allow the mitotic entry. To examine this hypothesis, we depleted CCT in a short version as indicated in Fig. 7A. The Western blot indicated that CCT was only partially depleted using this protocol (Fig. 7B). Following the release from the double thymidine block, CCT partially depleted cells were stained with phospho-H3 antibody. As expected, two populations of CCT-depleted cells were detected. One population never entered mitosis, presumably due to a high degree of CCT depletion. The other population entered mitosis at the same time as control cells but remained to be arrested at mitosis afterwards, suggesting the difficulty for mitotic progression (Fig. 7C). The different depletion degree is common for the RNAi-based gene knockdown approach, due to the different transfection efficiencies of different cells.



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 7. Normal CCT level is not required for mitotic entry. (A) Protocol used to partially downregulate CCT with RNAi in well-synchronized HeLa cells stably expressing GFP-tubulin. (B and C) After release from the double thymidine block for different times, cells were analyzed by Western blotting using antibodies indicated (B) or immunofluorescence using anti-phospho-H3 and DAPI staining (C).

 
Plk1 is inactive in CCT-depleted cells. We next determined the effect of CCT depletion on Plk1 activity. Toward that end, CCT{alpha} was depleted in well-synchronized cells using the protocol outlined in Fig. 3A. The cells were then released into fresh medium without nocodazole for 8 h, harvested, and subjected to anti-Plk1 IP/kinase assay using purified GST-TCTP as a substrate. As shown in Fig. 8, the level of Plk1 in CCT{alpha}-depleted cells was comparable to that of control cells at the same time point, and the percentage of cells with 4N DNA content at the given time point was also similar, indicating that CCT depletion does not affect the cell cycle through S phase. Furthermore, CCT-depleted cells showed a level of Cdc2 activity similar to that of control cells after 8 h of release, both significantly lower than that of control cells after 12 h of release in the presence of nocodazole (Fig. 8C). Thus, we compared Plk1 activities after 8 h of release, the point when both control and CCT-depleted cells are at G2 phase. The dramatic decrease of Plk1 activity of CCT-depleted cells strongly suggests that CCT might be required for the biogenesis of functional Plk1 (Fig. 8D).



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 8. CCT depletion inhibits Plk1 activity. HeLa cells were depleted of CCT{alpha} by the protocol outlined in Fig. 3A and harvested at the indicated times. (A) Direct Western blotting analysis using antibodies indicated on the left. (B) FACS profiles of well-synchronized cells. (C) Cdc2 IP/kinase assay using histone H1 as a substrate. (D) Plk1 IP/kinase assay using GST-TCTP as a substrate, whereas Plk1 protein level is indicated in the top panel. (E and F) Time course of CCT depletion-induced DNA damage formation and Plk1 inactivation. Randomly growing HeLa cells were transfected with pBS/U6-CCT{alpha} and pBabe-Puro as described in Fig. 2A. After different times as indicated, cells were harvested and subjected to anti-phospho-H2AX staining (E) or Cdc2 and Plk1 IP/kinase assays (F). VP16 treatment was used as a positive control to introduce DNA damage formation. Bars, 20 µm.

 
Since DNA damage also inhibits the activation of Plk1 (33), there is a possibility that CCT depletion might be causing the G2/M arrest mainly by inducing DNA damage. Thus, we performed a time course experiment to examine both DNA damage formation using phospho-H2AX staining and Plk1 inactivation after different periods of CCT depletion. While Plk1 inactivation was obvious after 2 days posttransfection, obvious DNA damage was detected only at a later stage of CCT depletion (Fig. 8E and F), indicating that CCT depletion-induced Plk1 inactivation is not likely due to DNA damage formation.

The phenotype of unsynchronized cells after CCT depletion was also examined using the protocol illustrated in Fig. 9A. Randomly growing cells were depleted of CCT for 3 days, harvested, and subjected to Cdc2 and Plk1 immunocomplex kinase assays (Fig. 9B). CCT depletion increased Cdc2 activity about two- to threefold compared to that of control cells, indicating a partial mitotic arrest. Consistent with the Cdc2 activity, the percentage of phospho-H3-positive cells after CCT depletion increased to 19% compared to 3% of control cells. Importantly, Plk1 activity of CCT-depleted cells was comparable to that of control interphase cells (Fig. 9C), albeit the level of Plk1 was significantly higher than that of control (Fig. 2A), supporting the notion that Plk1-specific activity is reduced in CCT-depleted cells.



View larger version (49K):
[in this window]
[in a new window]
  		FIG. 9. Cell cycle reentry of CCT-depleted cells upon introduction of purified Plk1-TD. (A to C) CCT depletion in randomly growing HeLa cells. HeLa cells were transfected with pBS/U6-CCT using the protocol illustrated in panel A. After 2 days of selection with puromycin, the attached cells were harvested and subjected to anti-Cdc2 IP/kinase and anti-Plk1 IP/kinase assays (B). Similar experiments were performed with cells growing on coverslips, and percentages of phospho-H3 positive cells were determined (C). (D and E) An 0.5-µg quantity of purified GST-Plk1-T210D protein was directly delivered into HeLa cells growing on coverslips by use of BioPORTER reagents. Cells were stained with anti-GST antibody 16 h after protein delivery (D), and percentage of the protein uptake was determined (E). Bars, 20 µm. (F and G) After CCT depletion using the protocol in panel A, the attached cells were harvested, replated onto six-well plates, and incubated for 12 h. Cells were translocated with purified GST-Plk1-TD, incubated for different times as indicated, and harvested for anti-phospho-H3 immunostaining (F) or FACS analysis to determine the relative ratio of cell populations at different phases (G). (H) CCT was depleted in well-synchronized HeLa cells using the protocol described in Fig. 3A. Upon release from the double thymidine block, cells were translocated with purified GST-Plk1-TD, incubated for the times indicated, and stained with phospho-H3 antibody.

 
Active Plk1 rescues CCT depletion phenotypes. We attempted to reverse the phenotype of CCT depletion by introduction of active Plk1. For that purpose, we first determined the efficiency of a protein delivery reagent, BioPORTER (GTS). The BioPORTER reagent was incubated with 0.5 µg of GST-Plk1-T210D (constitutively active, purified from baculovirus-infected Hi5 cells), and the mixture was added to HeLa cells growing on coverslips. After 16 h of incubation, cells were stained with anti-GST antibody (Fig. 9D), and quantification indicated that the protein uptake efficiency was about 85% (Fig. 9E). This approach was subsequently utilized for CCT-depleted cells. Briefly, after CCT was depleted in randomly growing HeLa cells using the protocol in Fig. 9A, cells were replated onto six-well plates and incubated for 12 h. BioPORTER protein delivery reagent (GTS) was then used to translocate GST-Plk1-TD into CCT-depleted cells. After further incubation for the times indicated, cells were harvested and analyzed by phospho-H3 staining (Fig. 9F) and FACS (Fig. 9G). As shown in Fig. 9F, about 20% of CCT-depleted cells remained arrested at mitosis during the 24-h incubation period. In contrast, delivery of Plk1-TD decreased the mitotic index to 14% after 12 h of incubation and to 6% after 24 h of incubation, indicating efficient mitotic exit. As a control, we delivered purified GST-Plk1-KM (kinase dead) into CCT-deficient cells, and no obvious effect on mitotic index was observed (data not shown). FACS was used to follow cell cycle profiles after delivery of active Plk1 into CCT-deficient cells. About 45% of CCT-depleted cells had a 4N DNA content during the 24-h incubation period, and delivery of GST-Plk1-TD decreased the cell population with 4N DNA to 25%. Interestingly, rather than a significant increase of cell population with 2N DNA, the cell population with sub-G1 DNA increased from 7% of control cells to 25%, indicating that the cells had undergone apoptosis under this condition. A similar experiment was performed with well-synchronized cells using the protocol illustrated in Fig. 3A. Upon the release of double thymidine block, purified GST-Plk1-TD was delivered into CCT-depleted cells. After incubation for 13.5 h, cells were immunostained with phospho-H3 antibody. The mitotic index increased from 7% for control to 24% after Plk1 delivery for 13.5 h, indicating that CCT depletion-induced G2 block is at least partly due to the malfunction of Plk1 (Fig. 9H).

We were concerned that the purified Plk1-TD mutant protein was causing too much cell death. The reduction in mitotic index may simply reflect the fact that these cells are dying rather than exiting mitosis. To address this issue more directly, we performed the experiments outlined in Fig. 10A. Compared to the protocol used to fully deplete CCT{alpha} as described in Fig. 9 (80 h posttransfection), CCT{alpha} was partially depleted here after about 60 h posttransfection (Fig. 10E). We reasoned that the residual CCT should be able to generate enough functional Plk1 to drive mitotic exit if Plk1-TD was overexpressed. As shown in Fig. 10B and C, overexpression of GFP-Plk1-TD efficiently rescued the partial CCT depletion-induced mitotic arrest. Moreover, FACS analysis indicated that the cell death under this condition was not serious (Fig. 10D). Since Cdc20 is another known positive regulator of APC, we tested whether overexpression of Cdc20 also rescues the phenotype. As efficient expression of Myc-tagged Cdc20 in partial CCT-depleted cells was indicated by both Western blotting and immunofluorescence staining, overexpression of Cdc20 caused cells to exit mitosis, presumably due to APC activation (Fig. 10E to G). These data indicated that Plk1 and Cdc20, both under the control of CCT, regulate APC function independently.



View larger version (62K):
[in this window]
[in a new window]
  		FIG. 10. Rescue of partial CCT depletion-induced mitotic arrest in unsynchronized cells by overexpression of GFP-Plk1-T210D and Cdc20. (A) Protocol used in this set of experiments. (B to D) CCT{alpha} was depleted in HeLa cells using the protocol in panel A, and the remaining attached cells were harvested and reseeded on the coverslips. After 10 h of incubation, the cells were transfected with GFP-Plk1-T210D and subjected to anti-phospho-H3 staining 24 h later (B and C) or FACS analysis (D). In the FACS profiles, GFP-Plk1-T210D-transfected cells were presented as total population (left panel), GFP-negative cells (middle panel), and GFP-positive cells (right panel) after gating. (E to G) HeLa cells were depleted with CCT{alpha}, harvested, and reseeded on the coverslips as in panel A. After 10 h of incubation, the cells were transfected with Myc-tagged Cdc20 and subjected to either Western blotting using antibodies indicated (E) or anti-phospho-H3 staining 24 h later (F). (G) Histograms quantifying the results. Bars, 20 µm.

 
Finally, to make sure that the purified Plk1-TD mutant causes fully CCT-depleted, G2 blocked cells as described in Fig. 9H to enter bona fide mitosis, a series of more thorough analyses was performed (Fig. 11). CCT was first depleted in well-synchronized HeLa cells stably expressing GFP-tubulin or GFP-H2B using the protocol described in Fig. 3A. Upon release from the double thymidine block, cells were translocated with purified GST-Plk1-TD, incubated for different times, and stained with either phospho-H3 or lamin A/C antibody to follow the bipolar spindle formation and NEBD, respectively. The bipolar spindle formation was rare in CCT-depleted cells but was clearly detected after the delivery of GST-Plk1-TD (Fig. 11A). Compared to about 5% of fully CCT-depleted cells with NEBD, at 13 h post-delivery of Plk1-TD the percentage of NEBD increased to 20% (Fig. 11B and C).



View larger version (61K):
[in this window]
[in a new window]
  		FIG. 11. CCT was depleted in well-synchronized HeLa cells stably expressing GFP-tubulin (A) or GFP-H2B (B and C) using the protocol described in Fig. 3A. Upon release from the double thymidine block, cells were translocated with purified GST-Plk1-TD, incubated for the times indicated, and stained with phospho-H3 (A) or lamin A/C (B) antibody. (A and B) Representative images after 13 h of release. (C) Quantification data to indicate the percentage of cells with NEBD. Bars, 20 µm.

 

   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In budding yeast, a CCT mutant fails to separate sister chromatids and exit from mitosis due to inactivation of anaphase-promoting complex (4). Using the RNAi approach, we analyzed the phenotypes of CCT depletion in mammalian cells. CCT depletion in randomly growing cells leads to arrest at mitosis of about 20% of the cell population, whereas complete CCT depletion in well-synchronized cells mainly causes a G2 block. Long-term (at least 4 days posttransfection) CCT depletion induces apoptosis. Interestingly, cell death can be prevented by mimosine, a drug that blocks the cell cycle at G1, indicating that it is very critical for cells to go through normal mitosis for survival. Moreover, we found that ATM inhibition significantly potentiates the lethality of CCT depletion, suggesting involvement of the DNA damage response pathway in CCT depletion-induced cell death.

Multiple functions of Plk1 in mitosis have been well documented. However, whether Plk1 is a trigger kinase for mitotic entry is still controversial. In Xenopus laevis, it was shown that partial (90%) Plk1 inhibition by Plx1 antibody microinjection only delays M-phase entry (27), whereas complete immunodepletion of prophase extracts totally blocks it (29). Here we provide another example showing that different degrees of Plk1 silencing lead to different phenotypes in mammalian cells. Complete Plk1 depletion is necessary to block cells in G2 and for failure to activate Cdc25C, whereas partial Plk1 depletion apparently arrests cells at mitosis. To be consistent with the feedback loop model for Cdc2 activation using Plk1 as a trigger, only a minimum level of Plk1 activity would be required to activate the amplification loop to drive the cells to enter mitosis in mammalian cells. In contrast, maximum Plk1 activity is required for cells to proceed through normal mitosis. Both our data and the data from the Xenopus system (27, 28, 29) are in contrast to the report that Plk1 inhibition did not block Cdc25C activation and G2/M progression in starfish oocytes (25). However, it should be noted that starfish Plk1 activity was inhibited by only 90% after antibody injection in those experiments (25). Therefore, we feel the current work resolves to a great extent this discrepancy about the role of Plk1 as a trigger kinase in Cdc25C activation by demonstrating that, in mammalian cells as in Xenopus, a substantial silencing of Plk1 that causes a phenotype in one stage of mitosis may not be sufficient to block Plk1 action at another stage. Moreover, the ability of Plk1 to promote Cdc25C activation even at low levels of expression is probably due to the known association of Plk1 with Cdc25C (14) that is mediated via the polo-box domain (17).

During the preparation of the manuscript, two papers describing the phenotype of Plk1 depletion in well-synchronized cultures appeared (37, 42). The first paper, using the vector-based RNAi approach in U2OS cells, reported that Plk1 is not required for mitotic entry (42). In that paper, U2OS cells were transfected with an RNAi vector targeting Plk1 and incubated for 18 h. After treatment with thymidine for 24 h, cells were released for different times and analyzed by FACS to follow cell cycle progression and by phospho-H3 staining to monitor mitotic entry. In the FACS profiles, control cells seemed to accumulate at mitosis after 12 h of release from the thymidine block. However, only 3% of control cells were phospho-H3 positive at that time, and a lower percentage of control cells were phospho-H3 positive at other times, suggesting that U2OS cells were not well synchronized in these experiments (42). Thus, the results obtained under these circumstances are expected to be analogous to what we report here using randomly growing cells (Fig. 5E to H). Indeed, Plk1 depletion using the protocol described above in U2OS cells by those authors led to partial mitotic arrest with about 15% to 20% phospho-H3-positive staining after 50 h posttransfection, consistent with the percentage of mitotic cells after Plk1 depletion using unsynchronized HeLa cells. In the other paper, HeLa cells were synchronized at late G1 by double thymidine treatment. The authors transfected synthetic double-stranded RNA targeting Plk1 at the time of second thymidine addition and concluded that normal Plk1 level is not required for mitotic entry (37), consistent with what we observed in Fig. 6. Compared to the protocol used to deplete Plk1 described here (~60 h in Fig. 4), the depletion time was much shorter (~30 h) in their protocol. Thus, it is possible that residual Plk1 is high enough for cells to enter mitosis.

CCT chaperonin was initially identified as a molecular chaperone specifically required for the folding of the cytoskeletal proteins tubulin and actin (39, 44). In a more recent study to search for CCT substrates, up to 15% of cellular proteins were proposed to require the assistance of CCT for correct folding upon translation (41). Thus, the substrate spectrum of CCT appears to be much broader than what was originally proposed. Among them, two proteins that are involved in cell cycle progression have been demonstrated as CCT substrates. In a yeast-based screen designed to identify proteins that interact with human cyclin E, CCT was shown to be required for the maturation of cyclin E (43). CCT also proved to be essential for the correct folding of Cdc20, since Cdc20 from CCT mutant yeast cells lost its ability to bind to APC and to the checkpoint proteins (4). These results raise the intriguing model that the CCT chaperonin might play a crucial role in cell cycle progression through regulating the folding of key cell cycle players.

Several lines of evidence support the notion that Plk1 is a CCT substrate in mammalian cells. First, the phenotype of CCT depletion is reminiscent of that of Plk1 depletion, consistent with the notion that the correct folding of Plk1 requires CCT. Both CCT and Plk1 depletion induce partial mitotic arrest in randomly growing cells and block the cell cycle at G2 in well-synchronized cells. As described previously, long-term Plk1 depletion (4 days posttransfection) also induces apoptosis in cultured cells and ATM inhibition strongly potentiates the lethality of Plk1 depletion (19). Moreover, mimosine treatment also prevents Plk1 depletion-induced apoptosis (X. Liu and R. L. Erikson, unpublished data). Second, using both an in vitro translation system and cultured cell lysates, we have shown that Plk1 associates with CCT. Third, Plk1 activity was very low in CCT-depleted cells, both in randomly growing and in well-synchronized cells. Fourth, direct delivery of purified active Plk1 partially reverses the cell cycle arrest in CCT-depleted cells, suggesting that misfolding of Plk1 is at least partially responsible for the CCT depletion-induced phenotype. It is of interest that, in fully CCT-depleted, randomly growing cells, reintroduction of purified Plk1-TD protein did not increase the cell population with 2N DNA content; instead, a significant increase of cell population with subgenomic DNA was observed (Fig. 9G). This might be due to the misfolding of other CCT substrates due to CCT depletion. Moreover, overexpression of both Plk1-TD and Cdc20 constructs rescued the partial CCT depletion-induced mitotic arrest, indicating that Plk1 and Cdc20, both under the control of CCT, regulate APC function independently.

A photo-cross-linking method involving incorporation of photoactivatable probes into ribosome-bound newly translated peptides revealed that CCT binds to nascent chains cotranslationally (20). In agreement with this model, newly translated Plk1 with a frameshift mutation in the polo-box domain still associated with CCT, and the extreme N terminus of Plk1 has higher binding affinity for CCT. Furthermore, other members of the chaperone family such as hsp70 or hsp90 might facilitate CCT binding to newly translated chains. Indeed, hsp90 regulates the stability of Plk1 through its interaction with the polo-box domain (32). Thus, it seems that the maturation of Plk1 involves the association of both CCT and hsp90.

Taken together, the data presented here lead us to conclude that CCT is required for the biogenesis of functional Plk1. The nascent Plk1 polypeptide chain most likely associates with CCT cotranslationally and achieves its native state with the help of hsp90. Mitotic entry in mammalian cells does require some level of Plk1 activity to initiate the Cdc2/cyclin B feedback loop, and the full activation of Plk1 is necessary for cells to progress through mitosis.


   ACKNOWLEDGMENTS
 
We thank Eleanor Erikson and Xiaoyi Zhang for critical reading and comments on the manuscript and Ronald Melki for providing anti-CCT{alpha} antibody. We thank Kyung Lee and Toru Miki for providing HeLa cells stably expressing GFP-tubulin or H2B. We thank Hongtao Yu and Chou-Zen Giam for the Cdc20 construct and two anonymous reviewers for critical comments that improved the quality of the manuscript.

This work was supported by National Institutes of Health grant GM59172 to R.L.E. R.L.E. is the John F. Drum American Cancer Society Research Professor.


   FOOTNOTES
 
* Corresponding author. Mailing address: Department of Molecular and Cellular Biology, Harvard University, 16 Divinity Ave., Cambridge, MA 02138. Phone: (617) 495-9686. Fax: (617) 495-0681. E-mail: liu13{at}fas.harvard.edu. Back

{dagger} Supplemental material for this article may be found at http://mcb.asm.org/. Back

{ddagger} Present address: Genome Institute of Singapore, Genome Building, #02-01, 60 Biopolis Street, Singapore 138672, Singapore. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 
1. Abrieu, A., T. Brassac, S. Galas, D. Fisher, J. C. Labbe, and M. Doree. 1998. The Polo-like kinase Plx1 is a component of the MPF amplification loop at the G2/M-phase transition of the cell cycle in Xenopus eggs. J. Cell Sci. 111:1751-1757.[Abstract]

2. Alexandru, G., F. Uhlmann, K. Mechtler, M. A. Poupart, and K. Nasmyth. 2001. Phosphorylation of the cohesin subunit Scc1 by Polo/Cdc5 kinase regulates sister chromatid separation in yeast. Cell 105:459-472.[CrossRef][Medline]

3. Barr, F. A., H. H. W. Sillje, and E. A. Nigg. 2004. Polo-like kinases and the orchestration of cell division. Nat. Rev. Mol. Cell Biol. 5:429-440.[CrossRef][Medline]

4. Camasses, A., A. Bogdanova, A. Shevchenko, and W. Zachariae. 2003. The CCT chaperonin promotes activation of the anaphase-promoting complex through the generation of functional Cdc20. Mol. Cell 12:87-100.[CrossRef][Medline]

5. Carmena, M., M. Riparbelli, G. Minestrini, A. Tavares, R. Adams, G. Callaini, and D. M. Glover. 1998. Drosophila polo kinase is required for cytokinesis. J. Cell Biol. 143:659-671.[Abstract/Free Full Text]

6. Donaldson, M. M., A. A. Tavares, H. Ohkura, P. Deak, and D. M. Glover. 2001. Metaphase arrest with centromere separation in polo mutants of Drosophila. J. Cell Biol. 153:663-676.[Abstract/Free Full Text]

7. Dunn, A. Y., M. W. Melville, and J. Frydman. 2001. Review: cellular substrates of the eukaryotic chaperonin TRiC/CCT. J. Struct. Biol. 135:176-184.[CrossRef][Medline]

8. Elia, A. E., L. C. Cantley, and M. B. Yaffe. 2003. Proteomic screen finds pSer/pThr-binding domain localizing Plk1 to mitotic substrates. Science 299:1228-1231.[Abstract/Free Full Text]

9. Elia, A. E., P. Rellos, L. F. Haire, J. W. Chao, F. J. Ivins, K. Hoepker, D. Mohammad, L. C. Cantley, S. J. Smerdon, and M. B. Yaffe. 2003. The molecular basis for phosphodependent substrate targeting and regulation of Plks by the Polo-box domain. Cell 115:83-95.[CrossRef][Medline]

10. Glover, D. M., I. M. Hagan, and A. A. M. Tavares. 1998. Polo-like kinases: a team that plays throughout mitosis. Genes Dev. 12:3777-3787.[Free Full Text]

11. Gutsche, I., L. O. Essen, and W. Baumeister. 1999. Group II chaperonins: new TRiC(k)s and turns of a protein folding machine. J. Mol. Biol. 293:295-312.[CrossRef][Medline]

12. Karaiskou, A., X. Cayla, O. Haccard, C. Jessus, and R. Ozon. 1998. MPF amplification in Xenopus oocyte extracts depends on a two-step activation of cdc25 phosphatase. Exp. Cell Res. 244:491-500.[CrossRef][Medline]

13. Kotani, S., S. Tugendreich, M. Fujii, P. Jorgensen, N. Watanabe, C. Hoog, P. Hieter, and K. Todokoro. 1998. PKA and MPF-activated polo-like kinase regulate anaphase-promoting complex activity and mitosis progression. Mol. Cell 1:371-380.[CrossRef][Medline]

14. Kumagai, A., and W. Dunphy. 1996. Purification and molecular cloning of Plx1, a Cdc25-regulatory kinase from Xenopus egg extracts. Science 273:1377-1380.[Abstract]

15. Lane, H. A., and E. A. Nigg. 1996. Antibody microinjection reveals an essential role for human polo-like kinase 1 (Plk1) in the functional maturation of mitotic centrosomes. J. Cell Biol. 135:1701-1713.[Abstract/Free Full Text]

16. Lee, K. S., and R. L. Erikson. 1997. Plk is a functional homolog of Saccharomyces cerevisiae Cdc5, and elevated Plk activity induces multiple septation structures. Mol. Cell. Biol. 17:3408-3417.[Abstract]

17. Liu, J., A. L. Lewellyn, L. G. Chen, and J. L. Maller. 2004. The polo box is required for multiple functions of Plx1 in mitosis. J. Biol. Chem. 279:21367-21373.[Abstract/Free Full Text]

18. Liu, X., and R. L. Erikson. 2002. Activation of Cdc2/cyclin B and inhibition of centrosome amplification in cells depleted of Plk1 by siRNA. Proc. Natl. Acad. Sci. USA 99:8672-8676.[Abstract/Free Full Text]

19. Liu, X., and R. L. Erikson. 2003. Polo-like kinase (Plk)1 depletion induces apoptosis in cancer cells. Proc. Natl. Acad. Sci. USA 100:5789-5794.[Abstract/Free Full Text]

20. McCallum, C. D., H. Do, A. E. Johnson, and J. Frydman. 2000. The interaction of the chaperonin tailless complex polypeptide 1 (TCP1) ring complex (TRiC) with ribosome-bound nascent chains examined using photo-cross-linking. J. Cell Biol. 149:591-602.[Abstract/Free Full Text]

21. Nakajima, H., F. Toyoshima-Morimoto, E. Taniguchi, and E. Nishida. 2003. Identification of a consensus motif for Plk (Polo-like kinase) phosphorylation reveals Myt1 as a Plk1 substrate. J. Biol. Chem. 278:25277-25280.[Abstract/Free Full Text]

22. Nigg, E. A. 2001. Mitotic kinases as regulators of cell division and its checkpoints. Nat. Rev. Mol. Cell Biol. 2:21-32.[CrossRef][Medline]

23. O'Farrell, P. H. 2001. Triggering the all-or-nothing switch into mitosis. Trends Cell Biol. 11:512-519.[CrossRef][Medline]

24. Ohkura, H., I. M. Hagan, and D. M. Glover. 1995. The conserved Schizosaccharomyces pombe kinase plo1, required to form a bipolar spindle, the actin ring, and septum, can drive septum formation in G1 and G2 cells. Genes Dev. 9:1059-1073.[Abstract/Free Full Text]

25. Okano-Uchida, T., E. Okumura, M. Iwashita, H. Yoshida, K. Tachibana, and T. Kishimoto. 2003. Distinct regulators for Plk1 activation in starfish meiotic and early embryonic cycles. EMBO J. 22:5633-5642.[CrossRef][Medline]

26. Oppenheim, E. W., I. M. Nasrallah, M. G. Mastri, and P. J. Stover. 2000. Mimosine is a cell-specific antagonist of folate metabolism. J. Biol. Chem. 275:19268-19274.[Abstract/Free Full Text]

27. Qian, Y. W., E. Erikson, C. Li, and J. L. Maller. 1998. Activated polo-like kinase Plx1 is required at multiple points during mitosis in Xenopus laevis. Mol. Cell. Biol. 18:4262-4271.[Abstract/Free Full Text]

28. Qian, Y. W., E. Erikson, and J. L. Maller. 1999. Mitotic effects of a constitutively active mutant of the Xenopus polo-like kinase Plx1. Mol. Cell. Biol. 19:8625-8632.[Abstract/Free Full Text]

29. Qian, Y. W., E. Erikson, F. E. Taieb, and J. L. Maller. 2001. The polo-like kinase Plx1 is required for activation of the phosphatase Cdc25C and cyclin B-Cdc2 in Xenopus oocytes. Mol. Biol. Cell 12:1791-1799.[Abstract/Free Full Text]

30. Sarkaria, J. N., E. C. Busby, R. S. Tibbetts, P. Roos, Y. Taya, L. M. Karnitz, and R. T. Abraham. 1999. Inhibition of ATM and ATR kinase activities by the radiosensitizing agent, caffeine. Cancer Res. 59:4375-4382.[Abstract/Free Full Text]

31. Sigler, P. B., Z. Xu, H. S. Rye, G. Burston, W. A. Fenton, and A. L. Horwich. 1998. Structure and function in GroEL-mediated protein folding. Annu. Rev. Biochem. 67:581-608.[CrossRef][Medline]

32. Simizu, S., and H. Osada. 2000. Mutations in the Plk gene lead to instability of Plk protein in human tumour cell lines. Nat. Cell Biol. 2:852-854.[CrossRef][Medline]

33. Smits, V. A. J., R. Klompmaker, L. Arnaud, G. Rijksen, E. A. Niggs, and R. H. Medema. 2000. Polo-like kinase-1 is a target of the DNA damage checkpoint. Nat. Cell Biol. 2:672-676.[CrossRef][Medline]

34. Song, S., and K. S. Lee. 2001. A novel function of Saccharomyces cerevisiae cdc5 in cytokinesis. J. Cell Biol. 152:451-469.[Abstract/Free Full Text]

35. Spankuch-Schmitt, B., J. Bereiter-Hahn, M. Kaufmann, and K. Strebhardt. 2002. Effect of RNA silencing of polo-like kinase-1 (PLK1) on apoptosis and spindle formation in human cancer cells. J. Natl. Cancer Inst. 94: