Mitotic Phosphorylation of DNA Topoisomerase II α by Protein Kinase CK2 Creates the MPM-2 Phosphoepitope on Ser-1469*

DNA topoisomerase IIα is required for chromatin condensation during prophase. This process is temporally linked with the appearance of mitosis-specific phosphorylation sites on topoisomerase IIα including one recognized by the MPM-2 monoclonal antibody. We now report that the ability of mitotic extracts to create the MPM-2 epitope on human topoisomerase IIα is abolished by immunodepletion of protein kinase CK2. Furthermore, the MPM-2 phosphoepitope on topoisomerase IIα can be generated by purified CK2. Phosphorylation of C-truncated topoisomerase IIα mutant proteins conclusively shows, that the MPM-2 epitope is present in the last 163 amino acids. Use of peptides containing all conserved CK2 consensus sites in this region indicates that only the peptide containing Arg-1466 to Ala-1485 is able to compete with topoisomerase IIα for binding of the MPM-2 antibody. Replacement of Ser-1469 with Ala abolishes the ability of the phosphorylated peptide to bind to the MPM-2 antibody while a peptide containing phosphorylated Ser-1469 binds tightly. Surprisingly, the MPM-2 phosphoepitope influences neither the catalytic activity of topoisomerase IIα nor its ability to form molecular complexes with CK2 in vitro. In conclusion, we have identified protein kinase CK2 as a new MPM-2 kinase able to phosphorylate an important mitotic protein, topoisomerase IIα, on Ser-1469.

Precise coordination of cell cycle progression is critical not only for normal cell division but also under conditions of stress leading to DNA damage or incomplete DNA synthesis. Deregulation of cell cycle control has been shown to be a leading cause of genetic instability in human cancers (1)(2)(3) for which reason considerable effort is invested toward the identification and characterization of the surveillance mechanisms that control cell cycle progression ("check points"). To prevent damaged cells to divide, the G 2 checkpoint is activated in response to DNA damage or incomplete DNA synthesis leading to cell cycle arrest at the G 2 /M interphase (2,4,5). In addition, vertebrate cells can activate a checkpoint during early prophase in response to DNA damage resulting in return of damaged cells to G 2 (6). Traditionally defined, the prophase stage of mitosis starts with the first visible sign of chromosome condensation and ends at nuclear envelope breakdown. While the exact biochemical mechanisms controlling the onset of prophase are incompletely understood (for recent review, see Ref. 7), chromosome condensation is associated with extensive phosphorylation of proteins involved in the regulation of chromatin structure. For example, the nuclear enzyme DNA topoisomerase II␣ which is known to play an important role in chromosome condensation (8 -12) is subject to complex phosphorylation during mitosis including phosphorylation by the mitotic kinase cdc2 (also known as p34 cdc2 -cyclin B or CDK1). These events lead to the generation of mitosis-specific phosphorylation sites which are recognized by the monoclonal MPM-2 and 3F3/2 antibodies (13)(14)(15). Among these phosphorylation sites, the MPM-2 epitope appears to be particularly important, since its presence on mitotic chromosomes is closely associated with the condensed state (6).
The MPM-2 monoclonal antibody was originally raised against mitotic HeLa cells. Subsequent studies show, that it specifically recognizes a cell cycle-regulated phosphoepitope present in mitotic and meiotic proteins from a wide variety of species (16). These proteins become phosphorylated at the G 2 /M transition and are dephosphorylated at the end of mitosis (17). In addition to topoisomerase II␣, more than 50 other phosphorylated proteins are recognized by the MPM-2 antibody including microtubule-associated proteins, components of the anaphase-promoting complex, phosphatases, and a number of protein kinases including protein kinase CK2 (16, 18 -26). Multiple kinases are able to generate MPM-2 epitopes including mitotic kinases such as cdc2 kinase as well as kinases which are also active during interphase such as MAP 1 kinase (22,(27)(28)(29)(30)(31)(32). Interestingly, some MPM-2 kinases as, for example, NIMA are themselves activated by other MPM-2 kinases indicating the complexity of the signaling pathways which regulate mitotic entry (22,33).
Protein kinase CK2 is a serine/threonine kinase which has shown to be dramatically phosphorylated in mitotic cells (34,35). CK2 is the major kinase phosphorylating topoisomerase II in yeast (36) and a stable topoisomerase II-CK2 molecular complex has been demonstrated (12). Interestingly, CK2 differentially phosphorylates topoisomerase II in a cell cycle-dependent manner: some phosphoacceptor sites are preferentially phosphorylated in G 1 , while others are preferentially phosphorylated in mitosis (36).
In the present study, we have identified CK2 as a topoisomerase II-directed MPM-2 kinase and characterized the phosphorylation site which leads to generation of the MPM-2 epitope on human topoisomerase II␣. We have also investigated the influence of this phosphorylation on the catalytic activity of topoisomerase II as well as on the ability of topoisomerase II to form molecular complexes with CK2.

EXPERIMENTAL PROCEDURES
Materials-Nocodazole, leupeptin, pepstatin A, CHAPS, EGTA, Trizma, HEPES, ATP, GTP, heparin, Tween 20, IGEPAL CA-630, penicillin G, streptomycin, ␤-mercaptoethanol, and soluble peroxidase substrate tablets were purchased from Sigma-Aldrich. DNase I, protein A-Sepharose, and Pefabloc SC were obtained from Roche Molecular Biochemicals. Immunoblot polyvinylidene difluoride membranes, Tris-Tricine 10 -20% linear gradient ready gel, Tris-Tricine sample buffer, 10 ϫ Tris/Tricine/SDS, and peptide molecular weight markers were supplied from Bio-Rad. -Phosphatase and protein molecular weight markers were from New England Biolabs, Inc. Protein Phosphatase 2A was purchased from Upstate Biotechnology, PIPES was obtained from Research Organics Inc. and protein G-Sepharose was supplied by Zymed Laboratories Inc. Immunoplate maxisorp surface (96 well) were supplied by Nalge Nunc International. Microcon centrifugal filters were purchased from Millipore while Western blot detection ECL reagents were obtained from Amersham Pharmacia Biotech. Peptides were synthesized, purified, and characterized by mass spectrometry by Neosystem Laboratories.
Antibodies-Mouse monoclonal anti-MPM-2 antibodies were purchased from Upstate Biotechnology. Rabbit polyclonal anti-CK2 ␣ and anti-CK2 ␤ antibodies were prepared as described (37,38). Anti-topoisomerase II␣ mouse monoclonal antibodies SWT3D1 and SWR1C2, herein designated T3D1 and R1C2, were generous gifts from Gary Gorbsky (University of Oklahoma, Oklahoma City, OK). Peroxidaseconjugated goat anti-rabbit and donkey anti-mouse antibodies were supplied by Jackson Immunoresearch Laboratories, Inc.
Enzymes-The YepWOB6 plasmid containing hTopoII␣ cDNA under the Gal1 promoter (39) was kindly provided by James C. Wang (Harvard University, MA). The plasmid was overexpressed in Saccharomyces cerevisiae DBY 745 strain and purified as described (40). Purified enzyme preparations contained no detectable DNA topoisomerase I activity as determined by relaxation of supercoiled plasmid DNA in the absence of ATP. C-terminal truncated forms of hTopoII␣ were constructed from the YepWOB6 plasmid. Recombinant CK2 was expressed in baculovirus-infected insect cells and purified as described (41). Cdc2 kinase was generously provided by Laurent Meijer (Station Biologique, Roscoff, France).
Cell Culture-HeLa S3 cells were grown in 0.5-liter spinner flasks in Dulbecco's modified Eagle's medium (Life Technologies, Inc.) supplemented with 10% fetal bovine serum (Life Technologies, Inc.), 60 g/ml penicillin G, and 100 g/ml streptomycin sulfate. To arrest cells in mitosis, cells were incubated for 14 h in the presence of 75 ng/ml nocodazole. The mitotic index was determined by microscopic analysis of propidium iodide-stained cells and ranged from 70 to 90% for nocodazole-blocked cells. Mitotic chromosomes were isolated as described previously (15).
Gel Electrophoresis and Immunoblotting-Proteins were separated by electrophoresis using 5-20% gradient SDS-polyacrylamide gels and transferred to Immunoblot polyvinylidene difluoride membranes. For Western blot analysis, membranes were first blocked for 45 min at room temperature with 20 mM Tris-HCl, pH 7.9, 137 mM NaCl (TBS) containing 5% bovine serum albumin (Sigma-Aldrich). For immunoblotting, the MPM-2 mouse monoclonal antibody was diluted to 0.5 g/ml, the rabbit polyclonal anti-CK2 ␣ antibody diluted 1000 times and the anti-htopoII␣ mouse monoclonal T3D1 antibody 400 times in TBS containing 0.1% Tween 20 (TBST). Incubation with the first antibody was carried out for 90 min at room temperature. Membranes were then washed two times for 10 min with TBST followed by 1 h incubation with the peroxidase-conjugated secondary antibodies. The goat anti-rabbit antibodies were diluted 80,000 times in TBST whereas the donkey anti-mouse antibodies were diluted 40,000 times in the same buffer. Membranes were washed again with TBST and the results were revealed with the ECL chemiluminescence kit.
Reactions were stopped by addition of 15 l of 2 ϫ SDS-PAGE loading buffer.
Cell Extract Preparation for MPM-2 Kinase Assay-Nocodazoleblocked HeLa S3 cells (5 ϫ 10 6 ) were centrifuged for 5 min at room temperature at 200 ϫ g. Cell pellets were washed three times with phosphate-buffered saline (PBS) and 1 ml of lysis buffer containing 50 mM Tris-HCl, pH 7.5, 10 mM EGTA, 4 mM MgSO 4 (TEM), 1% CHAPS, 200 nM microcystin-LR (Calbiochem), 200 nM okadaic acid (Sigma) and a mixture of protease inhibitors (5 g/ml of pepstatin A, leupeptin, and Pefabloc SC) was added. Extraction mixtures were incubated 20 min at 4°C with intermittent vortexing followed by centrifugation at 4°C for 15 min at 20,000 ϫ g. Supernatants were saved and protein concentrations determined. For CK2 immunodepletion or heparin depletion experiments, protein concentrations were adjusted to 500 g/ml prior to addition of anti-CK2 antibodies or heparin-Sepharose. In addition, the lysis buffer also contained 300 mM NaCl.
CK2 Immunodepletion-Mitotic cell extracts were prepared as described and incubated for 3 h at 4°C under rotation in the absence or presence of a 1/100 dilution of a polyclonal antibody directed against the regulatory ␤ subunit. Extracts were incubated with protein A-Sepharose beads for 30 min at 4°C. Supernatants were saved and the precipitation repeated overnight with or without the CK2 ␤-directed antibodies. After incubation with protein A-Sepharose, supernatants were saved and new precipitations were carried out for 3 h at 4°C in the absence or presence of a 1/100 dilution of a polyclonal antibody directed toward the catalytic ␣-subunit of CK2 in order to eliminate CK2 activity which is not associated with the ␤ subunit. Supernatants were collected after incubation with protein A-Sepharose and their remaining MPM-2 kinase activity was determined Depletion of Heparin-binding Proteins-Mitotic cell extracts were prepared as described above followed by incubation for 3 h at 4°C in the presence or absence of Sepharose or heparin-Sepharose beads. The incubation was then repeated overnight and the remaining MPM-2 kinase activity of supernatants was determined.
Determination of MPM-2 Kinase Activity-Depleted and non-depleted mitotic extracts were diluted to 20 ng/l with TEM (20 l final volume), containing phosphatase and protease inhibitors. The reaction buffer was supplemented with 1 mM DTT, 0.5 mM ATP or GTP as indicated. Three hundred ng of purified mutant or wild type htopoII␣ were incubated with diluted extracts for 25 min at 37°C. Heparin at 25 g/ml was included in some assays to determine if this would inhibit the MPM-2 kinase. Reactions were stopped with 2 ϫ SDS-PAGE loading buffer (30% glycerol, 4% sodium dodecyl sulfate, 187 mM Tris-HCl, pH 6.8, and bromphenol blue).
Preparation of Extracts from Isolated Chromosomes and Mitotic Cells-Nocodazole-blocked HeLa S3 cells (1 ϫ 10 6 ) were centrifuged 5 min at room temperature at 200 ϫ g. Pellets were washed three times with PBS prior to extraction with 200 l of a lysis buffer containing 50 mM Tris-HCl, pH 7.5, 750 mM NaCl, 2 mM EGTA, 0.75% IGEPAL CA-630, 200 nM microcystin-LR, 200 nM okadaic acid and protease inhibitor mixture. The salt concentration of this buffer is elevated in order to extract tightly bound chromosomal proteins such as hTopoII␣. Extraction mixtures were then incubated with DNase I for 5 min at 37°C followed by 20 min incubation at 4°C with intermittent vortexing and centrifugation at 4°C for 15 min at 20,000 ϫ g.
Conjugation of T3D1 and R1C2 Antibodies to Protein G-Sepharose Bead-Protein G-Sepharose beads were washed three times with 20 mM Tris-HCl, pH 7.5, 1 mM EDTA, 150 mM NaCl, and 0.5% Tween 20 (immunoprecipitation buffer). Two hundred l of protein G-Sepharose beads were then combined with 300 l of each antibody in a total volume of 5 ml of immunoprecipitation buffer and incubated overnight at 4°C. Antibody-conjugated beads were washed three times in the corresponding immunoprecipitation buffer prior to immunoprecipitation.
hTopoII␣ Immunodepletion-One hundred l of mitotic extracts prepared by high salt extraction were incubated with 15 l of antibodyconjugated or non-conjugated protein G-Sepharose beads for 2 h at 4°C and supernatants were collected.
CK2-mediated Phosphorylation of Extracts from Mitotic Cells-Two l of control or TopoII immunodepleted mitotic extracts were dephosphorylated by 80 units of -phosphatase in a 10-l final volume reaction mixture for 30 min at 30°C. Dephosphorylated extracts were then incubated with 200 ng of purified CK2 for 25 min at 30°C in 20 l final volume containing 50 mM Tris-HCl, pH 7.5, 10 mM MgCl 2 , 1 mM DTT, 40 M GTP, 100 mM NaCl, 2 mM vanadate, and a mixture of protease inhibitors. Reactions were stopped by addition of 20 l of 2 ϫ SDS-PAGE loading buffer.
Chromosome Extract Preparation and Rephosphorylation Assay-Isolated mitotic chromosomes (prepared from 2.5 ϫ 10 6 HeLa S3 cells) were suspended in 25 l of 60 mM PIPES, 25 mM HEPES, pH 7.5, 10 mM EGTA, 4 mM MgSO 4 , 1 mM DTT, 1% CHAPS, supplemented with protease and phosphatase inhibitors and incubated for 5 min at 37°C with 2.5 l of DNase I. To determine the overall presence of MPM-2 phosphoepitopes, chromosomes were diluted 1:1 with 2 ϫ SDS-PAGE loading buffer and analyzed by gel electrophoresis and Western blot analysis with the MPM-2 antibody. For rephosphorylation experiments, 2 l of DNase-treated chromosomes were dephosphorylated by 80 units of -phosphatase in a 10-l final volume. Then, chromosomes were incubated with 200 ng of purified CK2 for 25 min at 30°C in a 20-l final volume containing 50 mM Tris-HCl, pH 7.5, 10 mM MgCl 2 , 1 mM DTT, 40 M GTP, 100 mM NaCl, 2 mM vanadate, and protease inhibitor mixture.
Peptide Phosphorylation by CK2-Peptides (250 M) were incubated for 25 min at 30°C with 100 ng of purified CK2 in a 10-l final volume containing 50 mM Tris-HCl, pH 7.5, 10 mM MgCl 2 , 1 mM DTT, 40 M ATP in the presence or absence of 1 Ci of [␥-32 P]ATP and 100 mM NaCl. The ability of the peptides to compete with hTopoII␣ for binding of the MPM-2 antibody was determined as described below.
Competition of Peptides for Binding of the MPM-2 Antibody to hTo-poII␣-Two hundred ng of purified hTopoII␣ was incubated either overnight at 4°C or alternatively, for 2 h at 37°C in 96-well immunoplates followed by blocking with 200 l of 2% BSA/PBS for 1 h at 37°C. CK2-phosphorylated peptides were diluted in PBS containing 50 g/ml heparin and 0.5 g/ml anti-MPM-2 antibody to 100 l final volume. Mixtures were incubated for 1 h at room temperature, added to hTopoII␣Ϫcoated wells, washed three times with 0.1% Tween 20-PBS (PBST), and incubated at 37°C for 1.5 h. Wells were washed 5 times with PBST prior to incubation with secondary antibodies (1/2000 dilution in 2% BSA/PBST) for 1 h at 37°C. Wells were then washed 5 times with PBST and MPM-2 binding revealed with soluble peroxidase substrate. Reactions were stopped after 15 min at room temperature by addition of H 2 SO 4 followed by measurement of the optical density at 490 nM with a Dynex MRX microtiter plate reader (Dynex Technologies, Inc.).
Enzyme-linked Immunosorbent Assay Test on CK2-phosphorylated Peptides-Peptides were phosphorylated by CK2 as described above. Reaction mixtures were loaded on Microcon centrifugal filters (molecular weight cut-off: 10 kDa) and centrifuged for 12 min at 10,000 ϫ g. Eluates, containing only peptides, were saved and the final concentration adjusted to 15 g/ml with PBS. One hundred l of diluted peptides were loaded in 96-well immunoplates and kept 2 h at 37°C. After three washes with 200 l of PBS, each well was incubated with 200 l of 2% BSA/PBS for 1 h at 37°C. Wells were washed 5 times with PBST prior to incubation for 2 h at 37°C with the anti-MPM-2 antibody. MPM-2 reactivity was then determined as described above.
Co-immunoprecipitation of hTopoII␣ and CK2-Purified hTopoII␣ (250 ng) was incubated for 25 min at 30°C with 40 ng of purified CK2 in 15 l final volume of phosphorylation buffer in the absence or presence of ATP. Then, 10 l of anti-topoisomerase II␣ antibody-conjuguated protein G beads were added, the mixture was diluted in immunoprecipitation buffer followed by incubation for 2 h at 4°C. The beads were washed three times with 500 l of immunoprecipitation buffer and resuspended in 50 l of 1.5 ϫ SDS-PAGE loading buffer. Samples were then subjected to electrophoresis and Western blot analysis.
Relaxation Assay-Purified topoisomerase II (400 ng) was first preincubated in the presence or absence of 100 units of phosphatase in a dephosphorylation buffer containing 50 mM Tris-HCl, pH 7.5, 0.1 mM Na 2 EDTA, 5 mM DTT, 0.01% Brij 35, and 2 mM MnCl 2 for 30 min at 30°C. Alternatively, dephosphorylation was carried out with 0.1 unit of protein phosphatase 2A in a buffer containing 20 mM Tris-HCl, pH 7.5, 5 mM MgCl 2 , 0.2 mM MnCl 2 , 0.5 mM DTT, and 50 mM KCl. These conditions result in complete dephosphorylation of topoisomerase II. Different amounts of topoisomerase II (1-5 ng/l), preincubated with or without phosphatase, were then added to relaxation buffer (150 ng of pBR322 DNA, 125 mM KCl, 7.5 mM MgCl 2, 0.75 mM ATP, 20 mM Tris-HCl, pH 7.5) and reaction mixtures incubated for 10 min at 37°C. Samples were subjected to electrophoresis in 0.8% agarose gels with 1 ϫ TBE buffer (2 mM EDTA, 90 mM Tris borate, pH 8.3) for 4 h at 6 V/cm at room temperature followed by staining with 0.5 g/ml ethidium bromide.

Protein Kinase CK2 Can Generate the MPM-2 Epitope on
Human Topoisomerase II␣-Both protein kinase CK2 and cdc2 kinase are able to phosphorylate human topoisomerase II␣ during mitosis (14,15). We therefore wished to determine if one of these two protein kinases was also able to generate the MPM-2 epitope. The results show that the MPM-2 epitope is created on purified, recombinant human topoisomerase II␣ after phosphorylation with CK2, but not after phosphorylation with cdc2 kinase (Fig. 1A, lanes 2 and 3). Since CK2 is extensively phosphorylated by cdc2 kinase during mitosis (34,35), the influence of the simultaneous presence of both CK2 and cdc2 kinase was also determined. The results show that the presence of cdc2 kinase greatly stimulates the ability of CK2 to generate the MPM-2 phosphoepitope on topoisomerase II␣ (Fig. 1A, lane 4).
Mitotic Cell Extracts Can Create the MPM-2 Epitope on Topoisomerase II␣ Using GTP and Is Inhibited by Heparin-The ability of extracts from mitotic cells to generate the MPM-2 phophoepitope was determined. The results show that the MPM-2 epitope can be created both in the presence of ATP and GTP (Fig. 1B, lanes 2 and 3). In contrast, no MPM-2 epitope is created if low concentrations of heparin (25 ng/l) are included in the incubation mixture (Fig. 1B, lanes 4 and 5). These data are consistent with a role for CK2 as a topoisomerase II-directed MPM-2 kinase.
CK2 Immunodepletion or Treatment with Immobilized Heparin Abolish the Ability of Mitotic Cell Extracts to Create the MPM-2 Epitope-To further confirm the role of CK2 as the kinase creating the MPM-2 phosphoepitope on topoisomerase II␣, mitotic extracts were either depleted with heparin-Sepharose beads, which selectively remove proteins with strong affinity to heparin such as CK2 or immunodepleted with a CK2directed antibody. Whereas treatments with Sepharose or protein A-Sepharose beads by themselves have little effect on formation of the MPM-2 epitope (Fig. 1C, lanes 2 and 4), both depletion of heparin-binding proteins and CK2 immunodepletion result in complete loss of MPM-2 kinase activity (Fig. 1C,  lanes 3 and 5). Together, these results strongly suggest that the mitotic MPM-2 epitope on topoisomerase II is generated by CK2.  2 and 4) or GTP (lanes 3 and 5), in the absence (lanes 2 and 3) or presence (lanes 4 and 5) of heparin followed by Western blot analysis with a monoclonal antibody directed against the MPM-2 phosphoepitope. C, topoisomerase II␣ was incubated with mitotic extracts in the presence of ATP followed by Western blot analysis with a monoclonal antibody directed against the MPM-2 phosphoepitope (lane 1). Preincubation of the extracts with Sepharose beads or with protein A-Sepharose beads had little influence on the formation of the MPM-2 epitope (lanes 2 and 4). In contrast, treatment with heparin-Sepharose beads or immunodepletion of CK2 lead to complete loss of the MPM-2 phosphoepitope (lanes 3 and 5).
additional interaction domains may be required for the specific phosphorylation of MPM-2 proteins by a given MPM-2 kinase. We therefore wished to determine how many of the MPM-2 reactive proteins present in isolated mitotic chromosomes might be substrates for CK2. Immunoblot analysis of isolated human prometaphase chromosomes with the MPM-2 antibody (Fig. 2A, lane 1) reveals the presence of seven major bands in agreement with previous findings (32,33). Dephosphorylation of isolated chromosomes leads to almost complete loss of MPM-2 epitopes (Fig. 2A, lane 2), which are not restored by reincubation of the dephosphorylated chromosomes with ATP or GTP ( Fig. 2A, lanes 4 and 6). However, rephosphorylation of chromosomes by purified CK2 in the presence of either ATP or GTP results in recreation of a single MPM-2 epitope with a molecular size of 170 kDa ( Fig. 2A, lanes 3 and 5). This band has previously been shown to correspond to topoisomerase II␣ (13).
CK2 Creates a Limited Number of MPM-2 Reactive Bands in Total Mitotic Cell Extracts-Next, we wanted to establish how many mitotic proteins are substrates for the MPM-2 kinase activity of CK2. Total mitotic extracts were dephosphorylated (Fig. 2B, lane 1) followed by rephosphorylation with purified CK2 in the presence of GTP. This is accompanied by the creation of a limited number of MPM-2 reactive epitopes. At the most, six MPM-2 reactive bands can be distinguished, corresponding to proteins with molecular masses of ϳ250, 210, 170, 110, 60, and 55 kDa (Fig. 2B, lane 2). Among these, the two bands at 110 and 170 kDa are apparent in all preparations examined, whereas the presence of the other bands is more variable. To confirm that the MPM-2 reactive band of 170 kDa corresponds to topoisomerase II␣, dephosphorylated mitotic cell extracts (Fig. 2C, lane 1) were immunodepleted with a topoisomerase II-directed antibody (Fig. 2C, lane 2) followed by rephosphorylation with purified CK2. The results show that topoisomerase II immunodepletion leads to selective disappearance of the 170-kDa MPM-2 reactive band confirming its identification as topoisomerase II␣ (Fig. 2C, compare lanes 3  and 4).
The MPM-2 Epitope Is Present in the Last 163 Amino Acids of Topoisomerase II␣-It has previously been reported that CK2-mediated phosphorylation of topoisomerase II␣ is directed exclusively toward the C-terminal part of the molecule, with the major phosphorylation site at Ser-1525 and a second site of phosphorylation at Ser-1377 (43). Residue numbers are taken from Medline files. A third phosphorylation site has been described for Thr-1343 (44). Recent studies report that the MPM-2 epitope on topoisomerase II␣ is present on Thr-666 whereas the 3F3/2 epitope corresponds to Thr-1343 (15,45). A possible explanation for the apparent contradiction with regard to the MPM-2 phosphorylation site is that this site was identified by comparative sequence analysis of sequence motifs present in different MPM-2 reactive proteins in contrast to the other phosphorylation sites which were determined experimentally by use of topoisomerase II fragments. We therefore constructed and purified a series of C-truncated topoisomerase II␣ mutants including T1 (amino residues 1-1368), T2 (amino acid residues 1-1256), and T3 (amino acid residues 1-1195) (Fig. 3,  A and B, first panel). Autoradiograms of full-length and truncated forms of topoisomerase II phosphorylated by CK2 in the presence of ␥-32 P-labeled ATP clearly show that only fulllength enzyme and the T1 mutant can be phosphorylated by CK2 (Fig. 3B, middle panel). This is consistent with the reported phosphorylation sites as well as with the 3F3/2 site, but not with the assignment of the MPM-2 site to the central part of the molecule. Immunoblot analysis with the MPM-2 antibody reveals that the MPM-2 epitope is present on full-length topoisomerase II but undetectable on any of the truncated forms (Fig. 3B, last panel). Since phosphorylation of CK2 by cdc2 kinase greatly stimulates the ability of CK2 to generate the MPM-2 epitope on topoisomerase II␣ (Fig. 1A), we wished to establish if CK2 activity present in mitotic extracts would generate the same phosphorylation sites on topoisomerase II␣ as being observed for CK2 purified from non-synchronized cells. To answer this question, full-length and C-truncated forms of topoisomerase II␣ were phosphorylated by mitotic extracts in the presence of ATP. This resulted in the creation of the MPM-2 epitope on full-length, but not on C-truncated forms of topoisomerase II (Fig. 3C) a picture similar to what was observed for purified CK2 (Fig. 3B, last panel). Together, these results clearly show that the MPM-2 site which is targeted by CK2 on human topoisomerase II␣ is present in the last 163 amino acids of the molecule.
Identification of Regions Likely to Contain the MPM-2 Phosphorylation Site-To identify motifs likely to contain the MPM-2 phosphoepitope, we first determined which serine and threonine residues conformed to the CK2 consensus motif. It is generally held, that the minimum consensus for CK2 is (S/ T)XX(E/D), where the carboxylic determinants, Glu and Asp, may be replaced by phosphorylated Ser or Tyr. Additional positive determinants include multiple acidic residues surrounding the Ser/Thr residues at position Ϫ3 to ϩ7 (46 -50). Next, sequence analysis between different mammalian topoi-  6). Rephosphorylation of chromosomes by purified CK2 in the presence of ATP or GTP (lanes 3 and 5) results in the recreation of a single MPM-2 reactive band with a molecular size of 170 kDa. The migration of molecular size markers is indicated on the left. B, total mitotic extracts were dephosphorylated (lane 1) followed by rephosphorylation with purified CK2 in the presence of GTP (lane 2). This is accompanied by the creation of a limited number of MPM-2 reactive bands with molecular weights as indicated to the right. C, dephosphorylated mitotic extracts (lane 1) were immunodepleted with a topoisomerase II-directed antibody (lane 2) followed by rephosphorylation with purified CK2 of native and immunodepleted extracts (lanes 3 and 4). The results show that topoisomerase II immunodepletion leads to the selective disappearance of the 170-kDa MPM-2 reactive band (indicated by a closed arrow) confirming its identification as topoisomerase II␣. Open arrowheads, heavy and light chain of the antibody used for immunodepletion of topoisomerase II.
somerase II␣ was carried out to determine which of these residues were evolutionally conserved. This resulted in identification of 6 potential phosphorylation sites comprising Ser-1374, Ser-1377, Ser-1469, Thr-1470, Ser-1476, and Ser-1525 (Fig. 4). Therefore, further studies were carried out with three synthetic peptides, K14D containing Lys-1370 to Asp-1383, R20A containing Arg-1466 to Ala-1485, and K15F containing Lys-1517 to Phe-1531. As a negative control, the peptide K24K containing Lys-1411 to Lys-1434 was included, since this peptide contains a high proportion of evolutionally conserved Ser and Thr residues but no CK2 consensus motif.
K14D, K15F, and R20A are Substrates for CK2-The ability of CK2 to phosphorylate the four previously identified peptides, K14D, K15F, R20A, and K24K was evaluated by autoradiography after phosphorylation of the peptides by CK2 in the presence of radiolabeled ATP. For comparison, the decapeptide RRREEETEEE, a classical CK2 substrate containing an optimized CK2 consensus motif was also included. The results show that K14D, K15F, and R20A are all good substrates for CK2 in contrast to K24K which is not phosphorylated (Fig. 5A). Comparison of the relative degree of phosphorylation shows that K15F is phosphorylated to the same extent as the reference substrate, whereas K14D and R20A although both good substrates are phosphorylated to a lesser degree (Fig. 5A).
Phosphorylated R20A Is Able to Inhibit Recognition of Topoisomerase II␣ by the MPM-2 Antibody-To determine which of the topoisomerase II peptides carry the MPM-2 phosphorylation site(s), a competition assay was carried out, in which the ability of phophorylated peptides to compete with the recognition of topoisomerase II␣ by the MPM-2 antibody was established. The results (Fig. 5B) show that neither K14D, K15F, nor the reference CK2 substrate are able to compete with topoisomerase II␣ for binding by the MPM-2 antibody. In contrast, R20A clearly inhibits the binding of the MPM-2 antibody to topoisomerase II␣.
Ser-1469 Is Needed for Recognition of R20A by the MPM-2 Antibody-To establish which residue in the R20A peptide carries the MPM-2 phosphoepitope, Ser-1469 or Thr-1470 were replaced by an alanine. The results (Fig. 5C) show that after phosphorylation both phosphorylated R20A and phosphorylated R20A with a T1470A substitution were able to compete with topoisomerase II␣ for MPM-2 binding, whereas phosphorylated R20A containing a S1469A substitution was not, sug-gesting that phosphorylated Ser-1469 is an absolute requirement for recognition by the MPM-2 antibody. Next, a direct binding assay was used to further confirm the assignment of the MPM-2 site to Ser-1469. In this assay R20A and R20A where Ser-1469 has been replaced by Ala were incubated with CK2 in the absence or presence of ATP. Following purification of the peptides, an enzyme-linked immunosorbent assay with the MPM-2 antibody was carried out. The results (Fig. 5D) clearly show that the ability of the R20A polypeptide to bind to the MPM-2 antibody is abolished if Ser-1469 is replaced with an alanine. Finally, a R20A peptide containing phosphorylated Ser-1469 was synthesized, and the ability of this peptide to compete with topoisomerase II␣ for binding of the MPM-2 antibody was determined. The results (Fig. 5E) show that R20A containing a phosphorylated Ser-1469 is able to compete with topoisomerase II␣ for binding of the MPM-2 antibody. In contrast, neither unphosphorylated R20A nor R20A where Ser-1469 is replaced with an Ala were able to compete with topoisomerase II␣. Together, these results clearly show that phosphorylation of Ser-1469 represents a major determinant for generation of the MPM-2 phosphoepitope on topoisomerase II␣.
Phosphorylation Has No Effect on the Catalytic Activity of Topoisomerase II␣-To determine if creation of the MPM-2 site may influence the catalytic activity of topoisomerase II␣, topoisomerase II was extensively phosphorylated with CK2. This did not affect the catalytic activity (results not shown), in agreement with results reported by others (51,52). More surprisingly, dephosphorylation of topoisomerase II␣ by phosphatase (Fig. 6A) or protein phosphatase 2A (results not shown) also had no effect on the catalytic activity of topoisomerase II as measured by relaxation of supercoiled DNA. These results suggest that the catalytic activity of human topoisomerase II␣ is not regulated by phosphorylation, whether mediated by CK2 or by other protein kinases.
Phosphorylation Has No Effect on the Ability of Topoisomerase II␣ to Form Molecular Complexes with CK2-We have previously reported that topoisomerase II from yeast is able to form stable molecular complexes with CK2 which are independent of the phosphorylation status of topoisomerase II (12). To determine if this is also the case for the human enzyme, topoisomerase II␣ and CK2 were incubated in the presence or absence of ATP followed by immunoprecipitation with an an-  2-4). Only wild type and the T1 mutant are phosphorylated by CK2 as revealed by autoradiography after incubation with labeled ATP (middle) while only wild type topoisomerase II␣ is recognized by a monoclonal antibody directed against the MPM-2 epitope (bottom). C, wild type topoisomerase II␣ (lane 1) and the three truncated forms of topoisomerase II (lanes 2-4) were incubated with mitotic extracts in the presence of ATP and the generation of the MPM-2 phosphoepitope revealed by Western blot analysis. tibody directed toward human topoisomerase II␣. Western blot analysis with an antibody directed toward the catalytic subunit of CK2 shows that despite clear differences in the level of the MPM-2 phosphoepitope on topoisomerase II, equal amounts of CK2 were recovered in topoisomerase II␣ immunoprecipitates (Fig. 6B). Therefore, like previously described for the yeast enzyme, human topoisomerase II␣ forms stable molecular complexes with CK2 which are not affected by the phosphorylation state of the topoisomerase.

DISCUSSION
Formation of MPM-2 reactive epitopes is a biochemical hallmark of mitosis in a wide variety of animal species ranging from nematodes to human (16). The distribution of MPM-2 reactive epitopes during mitosis displays a dynamic localization pattern which parallels that of the ongoing mitotic process. Furthermore, microinjection of MPM-2 antibodies into mitotic or meiotic cells leads to growth arrest, strongly suggesting that were incubated with radiolabeled ATP in the absence (a) or presence (b) of CK2 followed by electrophoresis and autoradiography. B, four different peptides including the standard CK2 substrate mentioned above, K14D, K15F, and R20A were phosphorylated by CK2 and the ability of the phosphorylated peptides to compete with topoisomerase II␣ for binding of the MPM-2 monoclonal antibody was determined. C, to establish, which residue in peptide R20A competes with topoisomerase II␣ for binding of the MPM-2 antibody, Ser-1469 or Thr-1470 were replaced with alanine, and the resulting peptides were phosphorylated by CK2. The ability of the modified peptides to compete with topoisomerase II␣ for binding of the MPM-2 antibody was determined. D, peptide R20A and peptide R20A where Ser-1469 has been replaced with Ala, were incubated with CK2 in the absence or presence of ATP followed by purification of the peptides, and an enzyme-linked immunosorbent assay was carried out with the MPM-2 antibody. E, a R20A peptide containing phosphorylated Ser-1469 was synthesized, and the ability of this peptide to compete with topoisomerase II␣ for binding of the MPM-2 monoclonal antibody was determined.
Previous studies have identified six different MPM-2 kinases including cdc2 kinase, NIMA, Polo-like kinase, a poorly identified mitotic kinase called ME kinase-H, MAP kinase, and the two isoforms of MAP kinase kinase (22,27,28,32,33,54). We now report that CK2 also has MPM-2 kinase activity and that this activity results in the generation of a specific MPM-2 epitope on topoisomerase II␣, an important mitotic protein required for chromosome condensation as well as for segregation of intertwined sister chromatids (for review, see Ref. 55).
CK2 is a ubiquitous messenger-independent serine/threonine protein kinase present in both the cell nucleus and the cytoplasm (56,57). Comparison with other major families of protein kinases shows that the catalytic subunit of CK2 displays greatest similarity to the CDC28 family of cyclin-dependent protein kinases, suggesting a possible role for CK2 in cell cycle regulation (58). Despite a large number of substrates, the exact physiological role of this protein kinase is not clear (59,60). However, it has unambiguously been shown that CK2 is essential for viability in S. cerevisiae, Schizosaccharomyces pombe, and Dictyostelium discoideum (61)(62)(63)(64). Interestingly, depletion of CK2 activity in S. cerevisiae is accompanied by formation of large, budded cells which seem to be arrested in mitosis (61,65). In addition, both CK2 subunits are dramatically phosphorylated by cdc2 kinase in cells that are arrested in mitosis (34). The high stoichiometry of phosphorylation suggests that phosphorylation may regulate certain functional properties of CK2 and that this might contribute to the burst of phosphorylation that accompanies the activation of cdc2 kinase at the G 2 /M transition. Consistent with this observation, it has recently been reported that the mitotic 3F3/2 phosphoepitope on topoisomerase II␣ is created by CK2 (15). We now show that CK2 also generates a second mitotic phosphoepitope on topoisomerase II␣, which is recognized by the MPM-2 antibody.
Despite the large number of cellular substrates for CK2, we have observed that the number of proteins which are substrates for the MPM-2 kinase activity of CK2 seems to be quite limited since phosphorylation of isolated mitotic chromosomes only leads to the formation of a single MPM-2 reactive protein while less than 10 substrates are present in total mitotic extracts. The two proteins which most consistently are substrates for the MPM-2 kinase activity of CK2 are topoisomerase II␣ and a protein with a molecular mass of ϳ110 kDa. Although the identity of this protein is not known, its molecular weight corresponds to another major CK2 substrate, nucleolin, which like topoisomerase II␣ also forms stable molecular complexes with CK2 (66) and undergoes mitosis-specific phosphorylation (67).
It is not known how the different MPM-2 kinases recognize their specific substrates. It is likely that at least two factors are involved, the motif around the MPM-2 phophorylation site and the presence of additional sequence motifs (42). The second factor may not play an important role in the case of CK2, since the R20A polypeptide by itself is a good substrate for the MPM-2 kinase activity of CK2. Rather, the choice of substrates may be due to the unique sequence requirements of CK2. In contrast to most Ser/Thr protein kinases, CK2 is extremely acidophilic (46 -48, 50). Furthermore, in contrast to prolinedirected protein kinases such as cdc2 kinase, CDK2, and MAP kinases, a proline at the ϩ1 position is a strong negative determinant for CK2-mediated phosphorylation (50). This may be especially important since both MAP kinases and cdc2 kinase also have MPM-2 kinase activities. Thus, the generation of the MPM-2 epitope on a given substrate by a specific MPM-2 kinase may, at least in part, be due to differences in consensus requirement among different MPM-2 kinases.
Further analysis of the MPM-2 epitope on topoisomerase II␣ reveals that this sequence motif shows some unusual features both with respect to most MPM-2 phosphoepitopes and with respect to typical CK2 phosphorylation sites. While the residues downstream from Ser-1469 are highly acidic and typical for CK2 sites, the presence of a proline in the Ϫ1 position as well as the cluster of basic residues further upstream is quite unusual (50). The MPM-2 site on topoisomerase II␣ is also unusual compared with most other MPM-2 sites since the proline residue is N-terminal rather than C-terminal to the phosphorylated Ser/Thr residue (26,29). While this variation clearly does not prevent the MPM-2 antibody from recognizing phosphorylated Ser-1469, it is not yet clear if this motif is functionally similar to more classical MPM-2 sites in other proteins.
It is puzzling that some of the MPM-2 kinases also are active during interphase, raising the question how they are able to generate mitosis-specific phosphorylation sites. The simplest explanation would be that substrate and enzyme are present in different cellular subcompartments during interphase. However, this is clearly not the case for CK2 and topoisomerase II␣, since CK2 is the major kinase targeting topoisomerase II during interphase in a variety of organisms ranging from yeast to man (36,43,51,68). A possible clue is that both subunits of CK2 are extensively phosphorylated by cdc2 kinase during mitosis (34). This is particularly dramatic for the catalytic ␣ subunit where 4 residues in the C-terminal domain are phosphorylated by cdc2 kinase resulting in substantial conformational modifications (35). This is consistent with our in vitro findings that the ability of CK2 to create the MPM-2 epitope is greatly stimulated in the presence of cdc2 kinase, although this FIG. 6. Phosphorylation of human topoisomerase II␣ has no influence on its catalytic activity nor on its ability to associate with CK2. A, dephosphorylation of human topoisomerase II␣ (left) with phosphatase (right) has no influence on its catalytic activity as measured by relaxation of supercoiled plasmid DNA. B, topoisomerase II␣ was incubated with CK2 in the presence or absence of ATP followed by immunoprecipitation with a topoisomerase II␣-directed antibody and Western blot analysis with an antibody directed toward the ␣ subunit of CK2. The results show that the presence or absence of the MPM-2 epitope has no influence on the molecular interaction between CK2 and topoisomerase II␣ . kinase by itself has no MPM-2 kinase activity toward topoisomerase II. Cdc2-dependent phosphorylation of topoisomerase II could enhance the accessibility of the MPM-2 epitope for its CK2-catalyzed phosphorylation. Alternatively, CK2 activity could be stimulated upon phosphorylation by cdc2 or, as previously suggested, by a mechanism that does not imply the phosphorylation of CK2 (69). Further experiments will be required to discriminate between these different mechanisms.
It is widely believed that formation of MPM-2 epitopes is functionally important for orderly mitotic progression. However, despite almost two decades of active research, the mechanism by which phosphorylation of the MPM-2 phosphoepitope affects its various substrates remains unclear. Our results indicate that the MPM-2 activity of CK2 has no effect on the catalytical activity of topoisomerase II␣. More surprisingly, the observation that dephosphorylation of topoisomerase II␣ by two different phosphatases also have no influence on the catalytic activity of the enzyme suggests that the catalytic activity of human topoisomerase II␣ is, at least in vitro, not regulated by phosphorylation. This may not be restricted to the human enzyme since similar results have been reported for topoisomerase II from fission yeast (70).
Even without changing the catalytic activity of topoisomerase II␣, phosphorylation of Ser-1469 may still alter other properties of the enzyme such as its intracellular localization or its interactions with certain molecular partners. Protein phosphorylation in the vicinity of nuclear localization signal sequences may influence the cellular localization of proteins (71). While multiple potential bipartite nuclear localization signals have been identified in the C-terminal part of human topoisomerase II␣, only one sequence corresponding to amino acids 1454 to 1497 was shown to confer strong nuclear localization to a reporter protein (72). This suggests that the formation of the MPM-2 phosphoepitope on Ser-1469 may influence the localization of topoisomerase II␣ during mitosis. Interestingly, while the MPM-2 motif is highly conserved among topoisomerase II␣ from different mammalian species, it is not present on topoisomerase II␤. The two different topoisomerase II isoforms show different cellular localization during mitosis, since topoisomerase II␣ remains tightly associated with the mitotic chromosomes whereas the ␤ isoform dissociates from the chromatin during this step of the cell cycle (73).
Another possibility is that the MPM-2 phosphoepitope may influence the association of topoisomerase II␣ with other cellular proteins. A particular attractive candidate is the novel mitotic regulator, Pin1. This protein is an essential peptidylprolyl-cis-trans isomerase, which is able to catalyze rotation around the peptide bond adjacent to a proline residue thereby influencing the conformation of certain proteins (74). Pin1 binding is dependent on mitosis-specific phosphorylation of target proteins and shows almost the same substrate specificity as the MPM-2 antibody (75). Preliminary results in our laboratory suggest that topoisomerase II␣ forms molecular complexes with Pin1 in a phosphorylation-dependent manner, as has been described for other MPM-2 epitopes. 2 In conclusion, these data suggest that CK2 in addition to its numerous functions during interphase may play an important role in mitosis. Our results show that CK2 has MPM-2 kinase activity toward Ser-1469 of topoisomerase II␣, an important mitotic protein required for chromosome condensation and for segregation of intertwined sister chromatids. These findings provide a framework for further investigation into the role of the MPM-2 phosphoepitope on the mitotic functions of topoisomerase II␣. In addition, identification of other CK2 sub-strates at the G 2 -M transition will undoubtedly facilitate efforts to define the role of CK2 during cell division.