HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 25, 23054-23064, June 21, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/25/23054    most recent M200111200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Messenger, M. M. Articles by Litchfield, D. W. Search for Related Content PubMed PubMed Citation Articles by Messenger, M. M. Articles by Litchfield, D. W. Social Bookmarking                      What's this?

Interactions between Protein Kinase CK2 and Pin1

EVIDENCE FOR PHOSPHORYLATION-DEPENDENT INTERACTIONS^*

Moira M. Messenger, Ronald B. Saulnier, Andrew D. Gilchrist^§, Phaedra Diamond, Gary J. Gorbsky^¶, and David W. Litchfield

From the  Department of Biochemistry, University of Western Ontario, London, Ontario N6A 5C1, Canada and the ^¶ Department of Cell Biology, Biomedical Research Center, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma 73104

Received for publication, January 4, 2002, and in revised form, April 3, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The peptidyl-prolyl isomerase Pin1 interacts in a phosphorylation-dependent manner with several proteins involved in cell cycle events. In this study, we demonstrate that Pin1 interacts with protein kinase CK2, an enzyme that generally exists in tetrameric complexes composed of two catalytic CK2 and/or CK2' subunits together with two regulatory CK2 subunits. Our results indicate that Pin1 can interact with CK2 complexes that contain CK2. Furthermore, Pin1 can interact directly with the C-terminal domain of CK2 that contains residues that are phosphorylated in vitro by p34^Cdc2 and in mitotic cells. Substitution of the phosphorylation sites of CK2 with alanines resulted in decreased interactions between Pin1 and CK2. The other catalytic isoform of CK2, designated CK2', is not phosphorylated in mitotic cells and does not interact with Pin1, but a chimeric protein consisting of CK2' with the C terminus of CK2 was phosphorylated in mitotic cells and interacts with Pin1, further implicating the phosphorylation sites in the interaction. In vitro, Pin1 inhibits the phosphorylation of Thr-1342 on human topoisomerase II by CK2. Topoisomerase II also interacts with Pin1 suggesting that the effect of Pin1 on the phosphorylation of Thr-1342 could result from its interactions with CK2 and/or topoisomerase II. As compared with wild-type Pin1, isomerase-deficient and WW domain-deficient mutants of Pin1 are impaired in their ability to interact with CK2 and to inhibit the CK2-catalyzed phosphorylation of topoisomerase II. Collectively, these results indicate that Pin1 and CK2 interact and suggest a possible role for Pin1 in the regulation of topoisomerase II. Furthermore, these results provide new insights into the functional role of the mitotic phosphorylation of CK2 and provide a new mechanism for selectively regulating the ability of CK2 to phosphorylate one of its mitotic targets.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The human peptidyl-prolyl isomerase Pin1 was first isolated for its ability to interact with NIMA protein kinase in the two-hybrid system (1). Pin1 is an 18-kDA protein with orthologs that have been identified in yeast, Drosophila, Xenopus, fungi, mice, and plants (2-8). Pin1 belongs to the parvulin family of isomerases, which is distinct from the cyclophilin and FK506 binding protein peptidyl-prolyl isomerase families. Studies on Pin1 provide evidence for its involvement in the G₂/M transition and in mitosis. Pin1 can interact with many mitotic regulators including Myt1, Wee1, Plk1/Plx1, and Cdc25, although not with Cdc2 (8-10). Overexpression of Pin1 in HeLa cells resulted in a G₂ arrest, whereas depletion of Pin1 in HeLa cells and yeast led to a mitotic arrest (1). Two groups found that the addition of recombinant Pin1 to cycling Xenopus extracts in vitro stopped extracts from entering mitosis (8, 9). Additionally, Winkler et al. (11) observed that interphase Xenopus egg extracts depleted of the Xenopus Pin1 homolog, xPin1, entered mitosis more rapidly than controls. They also clearly demonstrated that xPin1 is required for the execution of the DNA replication checkpoint. Collectively, these results link Pin1 to the G₂/M transition and events regulating mitosis.

Another important protein implicated in cell cycle events and in cell viability is the highly conserved Ser/Thr protein kinase CK2,¹ a tetrameric enzyme that is composed of two catalytic (CK2 and/or CK2') subunits and two regulatory (CK2) subunits (reviewed in Refs. 12-16). Genetic studies in yeast suggest a role for CK2 at distinct stages during cell cycle progression including the G₂/M transition where it is required for the phosphorylation of topoisomerase II (12, 17). In mammalian cells, CK2 activity increases when quiescent cells are stimulated to proliferate (reviewed in Refs. 13-16). Studies performed using selective CK2 inhibitors, antisense down-regulation, or kinase-inactive mutants of CK2 also illustrate that CK2 is required during various stages of cell cycle progression including the G₂/M transition and in checkpoint control (18-20). Moreover, CK2 activity levels are elevated in a number of tumors and leukemic cells. Targeted overexpression of CK2 in transgenic mice results in the development of T cell lymphoma and mammary tumorigenesis (21, 22). Additionally, accelerated lymphomagenesis is observed when mice with elevated expression of CK2 in T cells are crossed with transgenic mice overexpressing c-Myc or Tal-1 or in mice that are deficient in p53 (21, 23, 24). Collectively, these studies implicate CK2 as an important component of signaling pathways involved in cell cycle progression and transformation.

Although the mechanism of CK2 regulation is not fully understood, one way in which it could be regulated is through phosphorylation. Several sites on CK2 are known to be phosphorylated in cells (25-28). In fact, in light of its predicted involvement in various aspects of cell cycle progression, it is notable that CK2 is phosphorylated in a cell cycle-dependent manner (26). In particular, one isoform of the CK2 catalytic subunit (i.e. CK2) is selectively phosphorylated in mitotic cells at 4 sites localized within its C-terminal 60 amino acids. Interestingly, the other isoform of the CK2 catalytic subunit (i.e. CK2') has a distinct C-terminal domain that lacks these phosphorylation sites and is not phosphorylated in mitotic cells (26). The mitotic phosphorylation sites on CK2 can all be phosphorylated in vitro by p34^Cdc2, suggesting that this enzyme is responsible for the mitotic phosphorylation of CK2 and that CK2 is a participant in a p34^Cdc2-mediated protein kinase cascade. Identification of the sites that are phosphorylated by p34^Cdc2 revealed that the mitotic phosphorylation sites on CK2 (Thr-344, Thr-360, Ser-362, and Ser-370) can all be classified as proline-directed phosphorylation sites because each of the phosphorylated residues is immediately N-terminal to a proline (28). The regulatory  subunit of CK2 is also phosphorylated at autophosphorylation sites and at a site that is maximally phosphorylated in mitotic cells (26). Despite the identification of these phosphorylation sites, no clear effect of phosphorylation of CK2 on CK2 activity has yet been observed (29). Because the mitotic phosphorylation sites on CK2 are all Ser-Pro or Thr-Pro sites that resemble the optimal binding motif for Pin1 (30), we hypothesized that CK2 may interact with Pin1 in a phosphorylation-dependent manner as a part of events regulating the cell cycle. Experiments described in this paper were undertaken to test for an interaction between Pin1 and CK2 and for the phosphorylation dependence of this interaction. Furthermore, having demonstrated interactions between CK2 and Pin1, we examined the effects of Pin1 on CK2 activity toward several substrates. Of particular interest was topoisomerase II, a protein that is important for cell cycle progression. At the non-permissive temperature, yeast with temperature-sensitive CK2 fail to enter mitosis (12, 27, 31). In these yeast, topoisomerase II is hypo-phosphorylated and inactive, suggesting that it is a bona fide physiological target for CK2. Additionally, it was recently demonstrated in mammalian cells that CK2 phosphorylates residues on topoisomerase II, including Thr-1342, which are residues that are maximally phosphorylated in mitotic cells (32-34). In this study, we demonstrated that Pin1 inhibits the in vitro phosphorylation of Thr-1342 on topoisomerase II by CK2 and that Pin1 interacts with topoisomerase II. Maximal inhibition of phosphorylation of Thr-1342 by CK2 requires full-length Pin1 with intact WW and isomerase domains.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

DNA Constructs-- CK2 constructs had HA or Myc tags to allow the identification or isolation of transfected CK2 subunits from endogenous CK2 subunits. As described previously (35), the HA tag consists of three repeats of the influenza hemagglutinin epitope YPYDVPDY, and the Myc tag consists of the Myc epitope MASMEQKLISEEDLNN. Constructs encoding CK2 with a C-terminal HA tag were generated using pRc/CMV (Invitrogen). The following CK2 constructs were employed: wild-type -HA, -4A-HA, -4D-HA, and -4E-HA. In these latter three constructs, the four mitotic phosphorylation sites on CK2, (i.e. Thr-344, Thr-360, Ser-362, and Ser-370) were mutated to alanines, aspartic acids, and glutamic acids, respectively, by using sequential PCR. All constructs were verified by sequencing.

The CK2' subunit was HA-tagged at its N terminus and was also expressed using pRc/CMV (Invitrogen). In addition to wild-type CK2', a construct encoding a chimera consisting of the N-terminal domain of CK2' with the C-terminal domain of CK2, as described previously (35), was utilized. The latter chimeric construct encodes residues 1-296 of CK2' together with the C-terminal fragment of CK2 (i.e. residues 296-391 of CK2) instead of the natural C terminus of CK2' (i.e. residues 297-350 of CK2'). Generation of the chimeric construct was described previously (35) and was achieved using a Bsu36I restriction site that is conserved between CK2 and CK2'. Importantly, the four mitotic phosphorylation sites on CK2 (i.e. Thr-344, Thr-360, Ser-362, and Ser-370) are all located with the C-terminal portion of CK2 that was transferred to CK2' to generate the chimera. Constructs encoding CK2 subunits had N-terminal Myc tags and were also in pRc/CMV. In addition to wild-type Myc-CK2, constructs were generated to encode Myc-CK2 with mutations at its mitotic phosphorylation site at Ser-209. For these constructs, Ser-209 was mutated to alanine (i.e. Myc-S209A) or to aspartic acid (i.e. Myc-S209D). All constructs were verified by sequencing.

Several DNA constructs were made for use in the production of recombinant GST-Pin1 fusion proteins. GST fusion proteins were generated using the pGEX-KG (36) vector. The Pin1 cDNA was obtained from American Type Culture Collection (ATCC 928227). It was amplified by PCR to introduce NcoI and HindIII restriction sites to the 5' and 3' end, respectively, of the Pin1 coding region to facilitate subcloning. Primers for PCR were as follows: GGA TCC CCA TGG CGG ACG AGG AGA AGC TG (forward primer designated p1) and GGA TCC AAG CTT CAC TCA GTG CGG AGG ATG ATG (reverse primer designated p2). The PCR product (~490 bp) was ligated into PCR-Blunt (Invitrogen) and sequenced. The modified Pin1 cDNA was then subcloned into pGEX-KG using the NcoI and HindIII sites to generate pGEX-KG-Pin1.

Four mutant GST-Pin1 constructs were also generated. GST-Pin1^Y23A and GST-Pin1^R68A,R69A are full-length GST-Pin1 fusion proteins with point mutations resulting in the loss of WW binding ability (37) and a dramatic decrease in isomerase activity (9, 30), respectively. GST-Pin1-(1-54) encodes a GST fusion protein with the WW domain of Pin1 and GST-Pin1-(47-163) is a GST fusion protein encoding only the peptide prolyl isomerase domain of Pin1 (37). Pin1^Y23A and Pin1^R68A,R69A were also made by sequential PCR. In the first round for Pin1^Y23A, primer pairs p-1 with p-3 (TTG AAG TAG GCC ACT CGG CCT GAG CT) and p-2 with p-4 (CCG AGT GGC CTA CTT CAA CCA CAT) were used. Likewise, for Pin1^R68A,R69A primers p-1 with p-5 (CGA GGG TGC TGC TGA CTG GCT GTG C) and p-2 with p-6 (AAG CAC AGC CAG TCA GCA GCA CCC TCG TCC T) were used. For the second round of PCR, primers p-1 and p-2 were used for both fragments. Pin1-(1-54) encompassing the WW domain of Pin1 was created using primers p-1 and p-8 (GGA TCC AAG CTT CCT GGC AGG CTC CC) and Pin1-(47-163) with primers p-7 (TGG CCT TGG CTG AGC TGC AGT) and p-9 (GGA TCCATG GCAAAC GGG CAG GGG GAG). All of these final PCR products were then separately ligated into PCR-Blunt. Fragments encoding Pin1-(47-163), Pin1^Y23A, and Pin1^R68A,R69A in PCR-Blunt were digested with NcoI and HindIII yielding the desired fragments of ~340, 490, and 490 bp, respectively. These NcoI/HindIII fragments were then used to replace NcoI/HindIII fragments from the Pin1 in pGEX-KG. For Pin1-(1-54), a 165-bp NcoI/BlpI fragment was isolated from the pCR-Blunt construct and subcloned into pGEX-KG vector that had been similarly digested. For the constructs described above in detail, all PCR amplifications were done using Pfu DNA polymerase (Stratagene) and sequenced by dideoxy sequencing (38) using T7 polymerase (Amersham Biosciences kit) or by ABI Prism BigDye Terminator method (PerkinElmer Life Sciences) at the Robarts Research Institute (Ontario, Canada). Sequencing and restriction digests were used to verify all constructs.

Plasmids encoding GST-C92, a GST fusion protein encoding the C-terminal 92 amino acids of c-Myc (39), GST-Max, a GST fusion protein encoding Max (40), and GST-631, a GST fusion protein encoding the N-terminal 198 amino acids of c-Myb (41), were generous gifts of Dr. B. Luscher (Hannover, Germany).

Protein Production and Purification-- GST, GST-Pin1, and GST-Pin1 mutant proteins were expressed in BL21 or DH5 bacteria grown in 2× YT with 100 µg/ml ampicillin (Roche Molecular Biochemicals) at 37 °C. After cultures reached an absorbance of greater than 0.6 at 600 nm, isopropyl--D-thiogalactoside (Indofine Chemical Corp.) was added to a final concentration of 0.1 mM. After 2-4 h, cells were pelleted, washed with PBS (130 mM NaCl, 3 mM KCl, 10 mM NaHPO₄, 2 mM KH₂PO₄, pH 7.4), and then resuspended in PBS containing 1.5 µg/ml aprotinin, 10 µg/ml leupeptin, 100 µM phenylmethylsulfonyl fluoride (PMSF) (Sigma) before being lysed using a French press or by sonication. Triton X-100 was added to this bacterial lysate to a final concentration of 1%, and the solution was incubated with continual mixing for 15-25 min at 4 °C. Cell debris was pelleted by centrifugation at 6600 × g. The supernatant was incubated with glutathione immobilized on agarose beads (Sigma) for 20-60 min at 4 °C or room temperature with continuous mixing. After thorough washing with PBS, the beads were used directly, or the protein was eluted using free glutathione (10 mM reduced glutathione (Sigma), 50 mM Tris-HCl, pH 8.0, 1 mM DTT). Eluted protein was dialyzed against 50 mM Tris-HCl, pH 8.0, 150 mM NaCl, with or without 1 mM dithiothreitol (DTT) and stored at 80 °C. Protein concentrations were determined using Coomassie protein assay (Pierce) and using bovine serum albumin as a standard.

Other GST fusion proteins that were used in kinase assays were similarly isolated. These included GST-C92 (39), a GST fusion protein encoding the C-terminal 92 amino acids of c-Myc, GST-Max (40), a GST fusion protein encoding Max, and GST-631 (41), a GST fusion protein encoding the N-terminal 198 amino acids of c-Myb. GST-CK2-C126 was expressed, purified, and phosphorylated with purified p34^Cdc2 using [-³²P]ATP as described previously (28).

Cleavage of GST-Pin1 was achieved by incubating with thrombin (1:100) in 10 mM Tris, pH 7.4, 150 mM NaCl, 2.5 mM CaCl₂, 2 mM DTT at 37 °C. Pin1 was subsequently separated from the GST by FPLC using a Mono S (HR5/5) column. Briefly, thrombin cleaved GST, and Pin1 fractions were diluted with an equal volume of 50 mM Hepes, pH 7.5, 0.1 mM EDTA and applied to the Mono S column. GST was not retained in the column and appeared in the flow-through. Pin1 was retained on the column and eluted using a 0-1 M NaCl gradient in 50 mM Hepes, pH 7.5, 0.1 mM EDTA. Fractions containing Pin1 were identified by SDS-PAGE and stored at 20 °C.

Y2C, a 205-amino acid portion of the topoisomerase II C terminus with a His tag to facilitate purification, and a mutant form of Y2C with a threonine to alanine substitution at residue 1342 were described previously (33). The Y2C fragment encodes residues 1158-1362 of topoisomerase II and has a pI of 8.5 indicating an overall basic character. These proteins were expressed in BL21 bacteria grown in LB or 2× YT with 50 µg/ml kanamycin monosulfate (Sigma) at 37 °C. After cultures reached an absorbance of greater than 0.4 at 600 nm, isopropyl--D-thiogalactoside was added to a final concentration of 1 mM, and 3 h later the cells were pelleted. Cells were used directly or stored at 20 °C. Proteins were harvested using a modified version of the manufacturer's recommendations using Ni-NTA Superflow beads (Qiagen). Briefly, pellets were thawed on ice for 15 min, resuspended in 8 M urea, 0.1 M NaH₂PO₄, 0.01 M Tris-Cl, pH 8.0, and incubated at room temperature for 15-60 min with gentle mixing. After centrifugation at 6,600 × g for 20-30 min, the supernatant was incubated with Ni-NTA Superflow beads for ~1 h with vigorous mixing. Beads were washed repeatedly with 8 M urea, 0.1 M NaH₂PO₄, 0.01 M Tris-Cl, pH 6.3. After removing the last wash, sample buffer (108 mM Tris, pH 6.8, 3.6% SDS, 18% glycerol, 0.01% bromphenol blue, 10% -mercaptoethanol) was added to beads to release the Y2C proteins.

Pin1 with an N-terminal His tag and a fusion protein designated N70 encoding residues 1-610 of SecA also with an N-terminal His tag were obtained from Dr. K. Hamilton and Dr. B. Shilton, respectively (London, Ontario, Canada).

Antibodies-- Polyclonal antipeptide antibodies directed against the N terminus of CK2 (residues 2-19), the C terminus of CK2 (residues 376-391), the C terminus of CK2' (residues 333-350), and the C terminus of CK2 (residues 198-215) have been described elsewhere (19, 42). Polyclonal rabbit anti-mouse antibodies were the IgG fraction of antiserum purchased from Sigma. Monoclonal antibody 12CA5 (Babco) recognizes the HA epitope (43, 44), and monoclonal antibody 9E10 recognizes the Myc epitope (45). The anti-phosphoepitope monoclonal antibody 3F3/2 was described previously (33). Anti-topoisomerase II Ab-1 (Calbiochem) is a monoclonal antibody directed against the C-terminal of human topoisomerase II. Biotin-conjugated anti-HA monoclonal (3F10) antibodies were purchased from Roche Molecular Biochemicals, and peroxidase-conjugated anti-biotin monoclonal antibodies were purchased from Jackson ImmunoResearch. Goat anti-rabbit and goat anti-mouse horseradish peroxidase and alkaline phosphatase-conjugated secondary antibodies were obtained from Bio-Rad.

Cell Culture and Transfections-- U20S/UTA6 cells, a human osteosarcoma cell line with a tetracycline-regulated transcriptional activator fusion protein (generous gift from Dr. Christoph Englert, Froschungszentrum Karlsruhe, Germany (46)), were grown in Dulbecco's modified Eagle's medium with 10% fetal calf serum, 0.1 mg/ml streptomycin, and 100 units/ml penicillin (Invitrogen). Standard calcium phosphate transfections (47) were performed using 34-40 µg of DNA per 10-cm plate. Specifically 17 µg of  or ' constructs, 17 µg of  construct, and ~3.4 µg of pEGFP-C2 (CLONTECH), pEGFP-C3 (CLONTECH), or -galactosidase in pRc/CMV were used per 10-cm plate. The pEGFP or -galactosidase plasmids were transfected into cells to monitor transfection efficiency by fluorescence microscopy or by 5-bromo-4-chloro-3-indolyl--D-galactopyranoside (X-gal) staining, respectively. The following sequence of events was then performed: 14-16 h after the DNA was added to the cells, the precipitate was washed and fresh media were added; 18 h later new media with a final concentration of 0.1 µg/ml nocodazole (Sigma) was added to the plates to arrest cells in mitosis; a further 16-18 h later okadaic acid (Calbiochem) was added to media to a final concentration of 1 µM; and 2 h later the cells were harvested. For interphase cell populations the nocodazole and okadaic acid were not added.

Extact Preparation for GST and GST-Pin1 Binding Assays-- After removing media from the tissue culture plates and washing twice with PBS, interphase cells were harvested by scraping in 500 µl per 10-cm plate of Tris Lysis Buffer (50 mM Tris, pH 8.0, 1% Nonidet P-40, 150 mM NaCl) with 1 mM DTT, 1.5 µg/ml aprotinin, 10 µg/ml leupeptin, 100 µM PMSF, and 1 µM microcystin LR (Sigma). Mitotic cells were shaken off the plates by repeated pipetting of the media over the plate surface. The media were collected; the cells were pelleted by centrifugation, and the pellet was washed once with PBS. The cell pellet was then resuspended in a volume equivalent to 500 µl of Tris Lysis Buffer per 10-cm plate harvested. Both interphase and mitotic cells were then sonicated for 30 s and then centrifuged at 107,700 × g for 20 min at 4 °C. The cell extract supernatants were then stored until use at 80 °C. Protein determinations were performed using the BCA protein assay (Pierce) using bovine serum albumin as standard.

Interphase nuclear extracts and mitotic chromosome preparations to be used as a source of topoisomerase II for GST and GST-Pin1 binding assays were isolated as described (33). Prior to use in the binding assays, these preparations were subjected to DNase (Roche Molecular Biochemicals) treatment in the presence of 10 mM MgCl₂ and 5 mM CaCl₂ for 10-20 min at room temperature.

Chromatographic Fractionation of Cell Extracts-- Untransfected cells were used to prepare interphase and mitotic fractions containing CK2 to be used in kinase assays. To obtain mitotic extracts, cells were grown in media with 0.1 µg/ml nocodazole for 16-21 h. Interphase and mitotic cells were harvested as described above except using Buffer A (50 mM Tris-Cl, pH 7.5, 2 mM EDTA, 1 mM DTT, 50 mM NaCl) with 1.5 µg/ml aprotinin, 10 µg/ml leupeptin, 100 µM PMSF, 1 µM microcystin, and 10 mM NaF instead of Tris Lysis Buffer. Equal amounts of protein, as determined by Coomassie protein assay (Pierce), were loaded on an FPLC column. Chromatographic fractionation was performed with either a Amersham Biosciences Mono Q column 5 × 50 mm on a Amersham Biosciences FPLC system or a Waters Protein Pak Q-8HR column on a Waters 625LC system. Protein was eluted using a 25-ml salt gradient starting with Buffer A and ending with Buffer B (50 mM Tris-Cl, pH 7.5, 2 mM EDTA, 1 mM DTT, 1 M NaCl) at a flow rate ranging from 0.25 to 0.5 ml/min. Eluted fractions were analyzed by SDS-PAGE followed by immunoblotting for the presence of CK2 and by kinase assays using a synthetic peptide to determine the amount of activity present in each fraction (as described below) (48). Immunoblots also allowed for the examination of the phosphorylation state of CK2 because phosphorylation of CK2 results in a shift in electrophoretic mobility (26, 28).

GST-Pin1 and His-Pin1 Binding Assays-- For binding assays with GST fusion proteins, equal amounts of GST-Pin1 or GST were incubated with 20 µl of a 1:1 slurry of glutathione-immobilized agarose beads to PBS for 20 min at 4 °C with continual mixing. The beads were then washed 3 or 4 times with PBS and/or Tris Lysis Buffer. An aliquot of the cell extract (prepared as in described in section 2.5) was mixed 1:1 with sample buffer and set aside for the gel. In each binding assay, 20 µl of 1:1 beads was incubated with 350-600 µl of the cell extract (typically 0.1-2 mg/ml) for 1 h at 4 °C with continuous mixing. The beads were then washed three times with Tris Lysis Buffer. After the removal of the last wash, a 50-µl sample buffer was added to the beads to elute bound proteins.

Binding assays with His-Pin1 were similarly performed by incubating His-Pin1 or N70 with Ni-NTA resin (Qiagen) according to the manufacturer's recommendations.

SDS-PAGE and Immunoblots-- Proteins were separated on 6 or 12% SDS-PAGE gels and transferred to polyvinylidene difluoride membrane (Roche Molecular Biochemicals) as described by Towbin et al. (49), over 1 h at 100 V in Blotting Buffer (25 mM Tris-Cl, pH 7.5, 190 mM glycine, 20% methanol). Prestained molecular weight markers (broad range from New England Biolabs) were used for reference. Markers and their molecular masses are as follows: 175-kDa MBP--galactosidase, 83-kDa MBP-paramyosin, 62-kDa glutamic dehydrogenase, 47.5-kDa aldolase, 32.5-kDa triose-phosphate isomerase, 25-kDa -lactoglobulin A, 16.5-kDa lysozyme, and 6.5-kDa aprotinin.

Immunoblots for colorimetric detection were treated in the following way: after washing in TBS (20 mM Tris-Cl, pH 7.5, 500 mM NaCl) the membranes were blocked for 30-60 min with 3% gelatin in TBS, followed by a 1-h incubation in the primary antibody solution and then 30-60 min in the secondary antibody solution. TBST (TBS with 0.05% Tween 20) was used to wash blots between incubations. Both primary and secondary antibodies were diluted in 1% gelatin in TBST. The dilutions of primary antibody used were as follows: 1:1000 for anti-, 1:500 for anti-, anti-HA, and anti-Myc. Alkaline phosphatase-conjugated goat anti-rabbit or goat anti-mouse antibodies (1:3000) were utilized as secondary antibodies. After the final TBST wash, the blots were washed with TBS and then incubated in a solution of 5-bromo-4-chloro-3-indoyl phosphate p-toluidine salt and p-nitro blue tetrazolium chloride (Bio-Rad) in AP Buffer (100 mM Tris-Cl, pH 9.5, 0.5 mM MgCl₂).

Immunoblots were similarly performed for chemiluminescent detection of HA-tagged CK2 with the following exceptions: blots were blocked with 2% bovine serum albumin (BSA) in PBS; all antibody solutions were made up in 1% BSA in PBST (PBS with 0.1% Tween 20); the primary antibody biotin-conjugated anti-HA was diluted 1:500; horseradish peroxidase-conjugated anti-biotin was diluted 1:20,000; washes were done with PBST; and in the last step blots were incubated in SuperSignal West Pico peroxide and luminal/enhancer solution (Pierce) and then exposed to x-ray film (Eastman Kodak). Immunoblots for the detection of topoisomerase II from binding assays were processed similarly except that gelatin was used in the place of BSA and anti-topoisomerase II Ab-1 at 6.3 µg/ml in 1% gelatin in TBST was used as the primary antibody.

Immunoblots with radioactive detection were performed as described above for colorimetric blots except that the secondary antibody used was the rabbit anti-mouse antibody diluted 1:11,000, and there was a third incubation after this with protein A-¹²⁵I (specific activity 70-100 µCi/µg, ICN) at 50,000-100,000 cpm/ml in 1% gelatin in TBST for 1-4 h. After this the blot was washed extensively, air-dried, and visualized using a PhosphorImager (Molecular Dynamics).

Topoisomerase II Phosphorylation Assay-- The topoisomerase II phosphorylation assay was performed as described in Daum and Gorbsky (33). Briefly, the Y2C fragments of topoisomerase II were run on SDS-PAGE gel and blotted as described above. The blots were cut into strips along the edge of the lanes. These blot strips were blocked in 3% gelatin in TBS for 15-30 min. Blots were washed once in TBST and again in TBST or TEM (50 mM Tris-HCl, pH 7.5, 10 mM EGTA, 4 mM MgSO₄). The blots were equilibrated in TEM with inhibitors (1.5 µg/ml aprotinin, 10 µg/ml leupeptin, 100 µM PMSF, 1 mM DTT, and 200 or 400 nM microcystin), and then the kinase assay was performed. Blot strips were incubated in small trays with 1 mM ATP, column fractions containing purified CK2, and with GST or GST-Pin1 diluted in TEM with inhibitors for 20-30 min at 30-38 °C. To halt the reaction, blot strips were washed four times with TBST. Blot strips were incubated with a 1:5000 dilution of 3F3/2 antibody in 1% BSA in TBST for 1 h, washed, incubated with goat anti-mouse horseradish peroxidase 1:15,000 in 1% BSA in TBST for 1 h, and finally incubated in the SuperSignal solution (Pierce). The activity of CK2 from FPLC fractions used in these assays ranged from 7 to 34 pmol/min/ml. The range of GST and GST-Pin1 concentrations used in these experiments was from 0 to 2 µM.

CK2 Kinase Assays-- CK2 kinase assays were performed using a synthetic peptide substrate, RRRDDDSDDD, as described previously (19, 48). In brief, small aliquots (6-12 µl) of CK2 from the FPLC fractions were incubated with and without 0.1 mM peptide substrate in 20 mM Tris-Cl, pH 7.5, 60 mM NaCl, 10 mM MgCl₂, 1 mM DTT, and 100 µM [-³²P]ATP (specific activity 50-100 cpm/pmol) for 5 and 10 min. The total reaction volume was 30 µl. An aliquot of the mix was then spotted on P81 phosphocellulose paper, washed extensively with 1% phosphoric acid, washed once with 95% ethanol, and quantified using a scintillation counter or PhosphorImager. To examine the effects of Pin1 on CK2 activity, peptide phosphorylation assays were similarly performed in the presence of the indicated amount of GST or GST-Pin1 using FPLC fractions as the source of CK2. Alternatively, these kinase assays were performed using casein or GST fusion proteins encoding known protein substrates of CK2 in the presence of the indicated amounts of Pin1. For the latter assays, kinase assays were performed in buffer containing 50 mM Tris, pH 7.5, 10 mM MgCl₂, 100 mM NaCl, 0.2 mM DTT, 100 µM [-³²P]ATP (specific activity 400-1000 cpm/pmol) using each of the indicated proteins as substrate. Following incubation for 10 min at 30 °C, kinase reactions were terminated by the additional SDS-PAGE sample buffer and boiling. Following SDS-PAGE, phosphorylated proteins were detected using a PhosphorImager.

Incorporation of ³²P into Topoisomerase II-- The wild-type Y2C fragment of topoisomerase II and the mutant form with the Thr-1342 to alanine substitution T1342A were produced in bacteria and then purified and blotted to polyvinylidene difluoride membrane as described above (33). The membranes were blocked in 5% BSA in TBST. The blots were then treated in the following manner: washed twice with TBST, once with TEM, once with TEM with 0.5% Triton X-100, once with TEM with 25 µM ATP and 200 nM microcystin LR, and then again with TEM. The blots were then incubated for 35 min at 37 °C with mixing in TEM with 25 µM ATP, 5 µCi [-³²P]ATP, 80 µl of CK2-enriched FPLC fraction, 1.5 µg/ml aprotinin, 10 µg/ml leupeptin, 100 µM PMSF, 1 mM DTT, and 200 nM microcystin with or without 50 µg/ml heparin in a total volume of 3 ml. To terminate the reaction, the blots were washed three times with TEM with 10 mM EDTA, 20 mM NaF, and 0.1% Triton X-100 and once in TEM with 10 mM EDTA, 20 mM NaF. The blots were air-dried and visualized using a PhosphorImager.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Pin1 Interacts with Protein Kinase CK2-- As a first step toward testing whether Pin1 could interact with CK2, recombinant Pin1 proteins were generated. GST and GST-Pin fusion proteins were produced in bacteria and purified using glutathione-agarose beads (not shown). GST-Pin1 or GST binding assays (i.e. pulldowns) were performed using U2OS cell lysates. To test whether CK2 interacts with Pin1, the proteins that bound to GST-Pin1 or GST were examined by immunoblotting with antibodies against CK2 (Fig. 1A). CK2 was readily detected in both interphase and mitotic extracts (Fig. 1A). In the mitotic extract, multiple -bands are present with the bands of reduced electrophoretic mobility representing the phosphorylated forms of  that have been characterized previously (26, 28, 29). Because CK2 can be phosphorylated at up to 4 sites in mitotic cells, individual bands represent different phosphorylated forms of CK2 with the bands exhibiting the slowest electrophoretic mobility representing the most heavily phosphorylated forms of CK2. In binding assays, GST-Pin1 interacted strongly with mitotic CK2 but not with interphase CK2 (Fig. 1A). It is also apparent that the most heavily phosphorylated form of CK2 exhibits the greatest interactions (Fig. 1A, last lane). CK2 was not detected in the GST binding assays (Fig. 1A) indicating that the Pin1 portion of the GST-Pin1 proteins is that which is required for interactions with CK2. Overall, these results indicate that Pin1 interacts with CK2.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Pin1 interacts with protein kinase CK2. A, extracts from unsynchronized U2OS cells (marked I) and from cells arrested in mitosis (marked M) were used in GST and GST-Pin1 binding assays as described under "Materials and Methods." Extracts as well as the proteins bound to GST or to GST-Pin1 (designated as Pin1) were separated by 12% SDS-PAGE, transferred to membranes, and analyzed on immunoblots using anti-CK2 antibodies with colorimetric detection. The position of endogenous CK2 is indicated () as is the phosphorylated form of endogenous CK2 (p). B, to test whether interactions between Pin1 and CK2 can be observed with transfected CK2, GST and GST-Pin1 binding assays were performed as in A using extracts from U2OS cells that had been transfected with epitope-tagged CK2 subunits. The immunoblot was probed with anti-HA antibodies to detect the HA-tagged CK2 subunits. As in A, lanes I are derived from unsynchronized cells, and lanes M are derived from cells arrested in mitosis with nocodazole. The position of CK2 with a C-terminal HA tag is indicated (-HA) as is the phosphorylated form of HA-tagged CK2 (p-HA). The positions of molecular weight markers are also illustrated to the right of each panel.

Interactions between CK2 and Pin1 were also examined using transfected CK2. Cells were transfected with HA-tagged CK2 (designated -HA) together with Myc-tagged wild-type CK2 and extracts prepared for examination of interactions with Pin1 (Fig. 1B). As seen with CK2 from untransfected cells, Pin1 interacts with transfected CK2-HA with the greatest interaction observed with the CK2-HA from mitotic cells (Fig. 1B, last lane). However, in contrast to the results observed with the endogenous CK2, transfected CK2-HA exhibits the appearance of multiple -bands in interphase extracts in the transfected cell extracts suggesting that the CK2-HA is phosphorylated even without mitotic synchronization (Fig. 1B, 1st lane). We do not know the precise reason for the presence of these phosphorylated forms of CK2-HA, but we speculate that these extra interphase bands may be the result of aberrant cell cycle events induced by overexpression of CK2 or by the transfection procedure. Because some of the CK2-HA that is present in interphase extracts is phosphorylated, it is therefore not surprising that some CK2-HA is observed in the GST-Pin1 pulldown from interphase extracts (Fig. 1B, 2nd last lane). In this pulldown, there is an obvious enrichment of the mostly heavily phosphorylated form of CK2-HA. Overall, the transfection experiments reinforce the suggestion that Pin1 preferentially interacts with phosphorylated CK2.

Comparison of Interactions between Pin1 and CK2 Versus CK2'-- Significant differences between CK2 and CK2' are found only within their C-terminal regions (50), and the mitotic phosphorylation sites on  all lie within this region (26, 28). CK2' does not have any of the mitotic phosphorylation sites and is not phosphorylated in mitotic cells. To determine whether CK2' could be transformed into a Pin1 interactor if it gained the mitotic phosphorylation sites of CK2, a chimeric construct of CK2' (designated '/) with the C terminus of ' replaced by that of  was used in GST-Pin1 binding assays (Fig. 2). The addition of the C-terminal domain of CK2 to CK2' resulted in the appearance of multiple bands as seen with phosphorylated  (Fig. 2) indicating that the '/ chimera undergoes mitotic phosphorylation reminiscent of that seen with CK2. In pulldown assays, the CK2'/ chimera clearly exhibits interactions with GST-Pin1, whereas negligible interactions between GST-Pin1 and CK2' are observed under these conditions (Fig. 2B, compare last 2 lanes). As well, no CK2 subunits were retained in GST binding assays (Fig. 2). Clearly, the addition of the C-terminal domain of CK2 is capable of converting CK2' into a Pin1 interactor. It is also noteworthy that the transfections of CK2, CK2', and CK2'/ were all performed in the presence of Myc-CK2. Consequently, because CK2' was not retained by GST-Pin1, it would appear that the ability of CK2 to interact with Pin1 resides in CK2 and not CK2.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   The C-terminal domain of CK2 mediates interactions with Pin1. A, constructs encoding HA-tagged CK2 (CK2), HA-tagged CK2' (CK2'), and an HA-tagged chimera comprising CK2' with the C-terminal domain of CK2 (CK2'/) were generated. As described under "Materials and Methods," CK2'/ was generated by replacing residues 297-351 of CK2' with residues 296-391 of CK2. The mitotic phosphorylation sites of CK2 (i.e. Thr-344, Thr-360, Ser-362, and Ser-370) that are all located within this C-terminal region of CK2 are indicated (P). B, these constructs were transfected into U2-OS cells together with Myc-CK2 and examined for their ability to interact with GST or with GST-Pin1 (designated Pin1) as in Fig. 1. Extracts were prepared from transfected cells that had been arrested in mitosis with nocodazole. Extracts as well as the proteins bound to GST or to GST-Pin1 (designated as Pin1) were separated by 12% SDS-PAGE, transferred to membranes, and analyzed on immunoblots using anti-HA antibodies. Molecular weight markers are also indicated (lane M).

Pin1 Interacts with a GST Fusion Protein Encoding the C-terminal 126 Amino Acids of CK2-- Results shown in Figs. 1 and 2 demonstrate that phosphorylated forms of CK2 can be isolated from cell extracts using GST-Pin1. However, because the source of CK2 was soluble cell extracts, it was not possible to determine whether CK2 interacts directly with GST-Pin1 or whether the interaction between CK2 and Pin1 is mediated by other proteins that are present in cell extracts. Consequently, we performed experiments to determine whether CK2 can interact directly with Pin1 and to determine whether the C-terminal domain of CK2 is sufficient for interactions with Pin1. Pulldown assays were performed using a bacterially expressed GST fusion protein encoding the C-terminal 126 amino acids of CK2. This bacterially expressed fusion protein was purified and phosphorylated at the mitotic phosphorylation sites of CK2 by p34^Cdc2 using [-³²P]ATP (28). As illustrated in Fig. 3, this phosphorylated fusion protein exhibits interactions in pulldown assays with His-tagged Pin1 that had also been expressed and purified from bacteria. Notably, the amount of ³²P-labeled fusion protein that is retained by His-tagged Pin1 immobilized on Ni-NTA resin is ~4-5-fold above the background binding of this fusion protein that is observed using Ni-NTA resin alone or an unrelated His-tagged protein immobilized on Ni-NTA (Fig. 3, C and D). These results demonstrate that the C-terminal domain of CK2 is sufficient for interactions with Pin1 and indicate that CK2 can interact directly with Pin1.

View larger version (34K): [in this window] [in a new window]   Fig. 3.   A GST fusion protein encoding the C-terminal domain of CK2 interacts with Pin1. A GST fusion protein encoding the C-terminal 126 amino acids of CK2, designated GST-CK2-C126, was phosphorylated in vitro with purified p34Cdc2 using [32P]ATP. A and B, 32P-labeled GST-CK2-C126 was examined for interactions with His-tagged Pin1 that had been immobilized on Ni-NTA. 32P-Labeled GST fusion protein that was retained on His-Pin1 was visualized with a PhosphorImager following electrophoretic separation. The input lanes represent the phosphorylated 32P-labeled GST-CK2-C126 fusion proteins that were utilized for these interactions. As controls, the 32P-labeled GST-CK2-C126 fusion protein was also examined for interactions with Ni-NTA resin alone (A) or with N70, an unrelated His-tagged fusion protein that represents an N-terminal fragment of SecA that had been immobilized on Ni-NTA (B). C and D, graphical representation of relative amounts of 32P-labeled GST-CK2-C126 that were retained on each resin as determined by PhosphorImager analysis of scans illustrated in A and B, respectively.

Examination of Interactions between Pin1 and Phosphorylation Site Mutants of CK2-- By having demonstrated that the C-terminal domain of CK2 is sufficient for interactions with Pin1, further experiments were undertaken to examine directly the importance of phosphorylation for interactions of CK2 with Pin1. Cells were transfected with wild-type CK2 (-HA) or a mutant CK2 (4A-HA). CK24A-HA is a mutant where each of the four mitotic phosphorylation sites of CK2 have been substituted with non-phosphorylatable alanines. Although wild-type CK2-HA appears as several bands exhibiting shifts in electrophoretic mobility resulting from phosphorylation, the 4A-HA mutant appears as a single band that co-migrates with non-phosphorylated CK2. Consistent with the prediction that phosphorylation is important for the interaction, 4A-HA was minimally retained on GST-Pin beads in comparison with mitotic wild-type  in binding assays (Fig. 4A, compare last 2 lanes). Neither -HA nor 4A-HA were retained by GST beads (Fig. 4A). In other experiments, this interaction was quantified using iodinated protein A as described under "Materials and Methods." There was 5-40-fold more wild-type -HA than 4A-HA retained on GST-Pin1 beads. These results indicated that at least one of the mitotic phosphorylation sites of CK2 is important for its interactions with Pin1.

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Examination of Pin1 interactions with phosphorylation site mutants of CK2. A, wild-type CK2 (WT) and a mutant of CK2 with each of its four mitotic phosphorylation sites mutated to alanine (4A) were examined for interactions with GST or GST-Pin1 as in Figs. 1 and 2 using extracts derived from nocodazole-arrested cells. Extracts as well as the proteins bound to GST or to GST-Pin1 (designated as Pin1) were separated by 12% SDS-PAGE, transferred to membranes, and analyzed on immunoblots using anti-HA to detect either wild-type CK2 or the 4A mutant of CK2. The positions of phosphorylated wild-type CK2 (p--HA) as well as non-phosphorylated wild-type CK2 (-HA) and the 4A mutant of CK2 (-4A-HA) are indicated. Non-phosphorylated wild-type CK2 (-HA) and the 4A mutant of CK2 (-4A-HA) exhibit similar electrophoretic mobility. B, the four mitotic phosphorylation sites on CK2 were mutated to glutamic acid (4E) or to aspartic acid (4D) to determine whether negatively charged amino acids could mimic phosphorylation in terms of interactions with Pin1. Extracts prepared from transfected cells arrested in mitosis as well as the proteins bound to GST or to GST-Pin1 (designated as Pin1) were separated by 12% SDS-PAGE, transferred to membranes, and analyzed on immunoblots using anti-HA to detect the wild-type and mutant forms of CK2. Phosphorylated wild-type CK2 exhibits similar electrophoretic mobility to that of either the 4E or the 4D mutants. C, Ser-209, the mitotic phosphorylation site on CK2 was mutated to either alanine (designated A209) or to aspartic acid (designated D209) to determine whether alteration in the phosphorylation of CK2 would affect interactions between CK2 and Pin1. Extracts as well as the proteins bound to GST or to GST-Pin1 (designated as Pin1) were separated by 12% SDS-PAGE, transferred to membranes, and analyzed on immunoblots using anti-HA to detect CK2HA and anti-Myc to detect wild-type (WT) or phosphorylation site mutants of CK2. The positions of CK2-HA (-HA) and its phosphorylated form (p--HA) as well as Myc-CK2 (Myc-) and its autophosphorylated form (p-Myc-) are also indicated. Note that mutation of Ser-209 does not affect the electrophoretic mobility of Myc-.

Next, the effect of replacing the phosphorylation sites of CK2 with glutamic acid or aspartic acid was examined. It was hypothesized that the negative charge of these amino acids might mimic the negative charge of phosphate groups and allow for interactions with Pin1. Cells were transfected with 4E-HA or 4D-HA together with Myc-, and again GST and GST-Pin1 binding assays were performed (Fig. 4B). These two phosphorylation site mutants of CK2 bound far less effectively to the Pin1 beads relative to wild-type mitotic CK2 (Fig. 4B). None of the proteins bound to GST beads (Fig. 4B, lanes 5-7). Therefore, the negative charges provided by the glutamic and aspartic acids were not sufficient to mimic completely the phosphorylated serines and threonines on CK2 for Pin1 recognition.

The regulatory CK2 subunit of CK2 is also phosphorylated in mitotic cells at a Ser/Pro site (25, 27). Therefore, to test whether phosphorylation of CK2 has an impact on interactions with Pin1, a similar strategy was used utilizing cells transfected with Myc-tagged wild-type , or mutants of  where serine 209 had been replaced with either alanine or with aspartic acid (i.e.  S209A and  S209D, respectively). Each of these constructs was transfected into cells along with CK2-HA, and pulldowns were performed. As seen in Fig. 4C (last 3 lanes), there was no striking difference in the amount of  and  detected when GST-Pin1 binding assays were performed using these three different  constructs, suggesting that phosphorylation of  does not affect interactions of CK2 with GST-Pin1. Of note is the predominance of the -band with the higher mobility in GST-Pin binding assays. This band is autophosphorylated  and is indicative of  that is part of a tetrameric CK2 complex (19, 51, 52), suggesting that  is being pulled down as a part of a complex with . Collectively, these results demonstrate that it is unlikely that the mitotic phosphorylation site on  is important for interactions with Pin1.

In summary, the importance of phosphorylation of CK2 for Pin1 interaction was demonstrated. Elimination of mitotic phosphorylation sites on  diminished its interactions with Pin1, and the elimination of the mitotic phosphorylation site on  did not have a significant effect. Furthermore, ' could be transformed into a Pin1-interacting protein with the addition of the C-terminal domain of  containing the mitotic phosphorylation sites.

Examination of Interactions between CK2 and Pin1 Using Pin1 Mutants-- Pin1 is composed of a WW domain that exhibits phosphorylation-dependent interactions with target proteins as well as a peptidylprolyl isomerase domain that exhibits phosphorylation-dependent catalytic activity (30, 37). To examine the role of each of these domains for binding CK2, binding assays were performed with various Pin1 mutants (Fig. 5). By itself, the isomerase domain of Pin1 designated GST-Pin1^47-163 did not bind CK2-HA (Fig. 5, 4th lane). By comparison, GST-Pin1^1-54 encoding only the WW domain of Pin1 did exhibit interactions with CK2-HA (Fig. 5, 5th lane). Similarly, GST-Pin1^Y23A, a mutant that abolishes the binding activity of the WW domain (37) did not bind CK2-HA, whereas GST-Pin1^R68A,R69A, a mutant reported to diminish significantly isomerase activity (9, 31) did bind CK2-HA (Fig. 5). Overall, these results suggest that the WW domain is most important for binding CK2-HA. However, the interactions between the WW domain and CK2-HA appeared to be much weaker than those observed with wild-type full-length Pin1 with CK2-HA. Although we have not rigorously excluded the possibility that other regions of Pin1 are important for interactions with CK2, the importance of the WW domain is consistent with evidence indicating that this protein interaction module is essential for the cellular functions of Pin1.

View larger version (35K): [in this window] [in a new window]   Fig. 5.   Examination of interactions between CK2 and Pin1 mutants. To identify the region of Pin1 that interacts with CK2, several GST-Pin1 mutant proteins were used in pulldown assays. Extracts as well as the proteins bound to GST or to the various GST-Pin1 fusion proteins were separated by 12% SDS-PAGE, transferred to membranes, and analyzed on immunoblots using anti-HA antibodies to detect epitope-tagged CK2 (-HA). In addition to GST-Pin1, the following mutant proteins were utilized: GST-Pin147-163 which encodes the isomerase domain of Pin1, GST-Pin11-54 which represents the WW domain of Pin1, GST-Pin1Y23A which encodes full-length Pin1 with an inactivating mutation within its WW domain, and GST-Pin1R68A,R69A which encodes full-length Pin1 with an inactivating mutation within its isomerase domain. Pulldowns were performed as in previous figures using extracts of cells that had been transfected with HA-tagged CK2 and arrested in mitosis with nocodazole. HA-tagged CK2 was detected using chemiluminescence. The pulldown with GST-Pin1 represents half as much volume as was loaded for each of the other pulldowns.

Examination of the Effects of Pin1 on CK2 Activity-- To determine whether Pin1 affects the catalytic activity of CK2, we examined the effect of Pin1 on the ability of CK2 to phosphorylate peptide and protein substrates. As seen in Fig. 6, Pin1 at various concentrations does not exert any dramatic effect on the in vitro activity of CK2 toward an optimized peptide substrate (Fig. 6A) nor toward known CK2 substrates such as casein (Fig. 6B), GST-Max (Fig. 6C), or a GST fusion protein encoding a portion of c-Myb that contains residues known to be phosphorylated by CK2 (Fig. 6D) (53). Similarly, Pin1 had no effect on the phosphorylation of a GST fusion protein encoding a portion of c-Myc that contains known CK2 phosphorylation sites (data not shown). Collectively, these results indicate that Pin1 is not a general inhibitor of CK2.

View larger version (43K): [in this window] [in a new window]   Fig. 6.   Pin1 is not a general inhibitor of CK2. CK2 fractions that had been isolated by FPLC from mitotic cell extracts as described under "Materials and Methods" were examined for their ability to phosphorylate known CK2 substrates including the synthetic peptide RRRDDDSDDD (A), casein (B), GST-Max (C), and a GST fusion protein designated GST-Myb that encodes the N-terminal 198 amino acids of c-Myb (D). Phosphorylation of the synthetic peptide was performed using a P81 filter assay as described under "Materials and Methods" in the presence of the indicated concentrations of GST (open bars) or GST-Pin1 (solid bars). Phosphorylation of casein, GST-Max, and GST-Myb was similarly performed in the presence of the indicated concentrations of Pin1. 32P incorporation into casein, GST-Max, and GST-Myb was detected using a PhosphorImager.

Although CK2 is active at all stages in the cell cycle (29), there is evidence that the activity of CK2 toward specific substrates could be modulated at specific stages during the cell cycle. For example, CK2 was recently shown to phosphorylate residues on topoisomerase II that are maximally phosphorylated in mitotic cells (33, 34). These observations strongly suggest that topoisomerase II is a mitotic target of CK2 in mammalian cells. Studies with temperature-sensitive mutants of CK2 in yeast provide similar indications that topoisomerase II is also a mitotic target for CK2 in yeast (31). Because we have previously shown that CK2 is phosphorylated in mitotic cells (26, 28) and because Pin1 interacts preferentially with mitotic CK2 and has been implicated in the control of mitotic events, we were thus interested in examining the effects of Pin1 on the CK2-catalyzed phosphorylation of the mitotic phosphorylation sites on topoisomerase II. To achieve this objective, we utilized a recombinant fragment of topoisomerase II, designated Y2C, as a substrate for CK2. The Y2C fragment of topoisomerase II used in this study is the fragment of topoisomerase II encoding residues 1158-1362. This fragment was previously utilized for the identification of Thr-1342 as the phosphorylated residue on topoisomerase II that is recognized by the 3F3/2 monoclonal antibodies (33). Furthermore, topoisomerase II can be immunoprecipitated from mitotic cells using 3F3/2 monoclonal antibodies, and CK2 phosphorylation can generate 3F3/2 reactivity on topoisomerase II that is associated with isolated chromosomes (33). Consequently, it appears that Thr-1342 is indeed phosphorylated in mitotic cells and that CK2 can phosphorylate this residue in intact topoisomerase II and in the Y2C fragment.

The 3F3/2 monoclonal antibody was utilized to assay for phosphorylation of one of the mitotic sites (i.e. Thr-1342) of topoisomerase II by partially purified mitotic CK2 (Fig. 7). The CK2 utilized in these studies was partially purified from mitotic extracts utilizing ion exchange chromatography as described under "Materials and Methods." Partial purification was performed to separate CK2 from many of the other protein kinase or phosphatase activities in extracts that could confound the examination of CK2 activity. It is evident that the wild-type Y2C fragment of topoisomerase II was readily phosphorylated by CK2, whereas the same topoisomerase II fragment with Thr-1342 mutated to alanine was negligibly detected (Fig. 7A). This result is consistent with previous findings (33). Additionally, assays were performed in the absence of CK2 or ATP. In either case, no bands were detected with the 3F3/2 antibody (data not shown). These control assays further confirmed that the antibody was sensitive to phosphorylation.

View larger version (45K): [in this window] [in a new window]   Fig. 7.   Pin1 inhibits the phosphorylation of Thr-1342 on topoisomerase II by CK2. Phosphorylation of Thr-1342 within the Y2C fragment of topoisomerase II was monitored using the phospho-specific 3F3/2 antibody as described under "Materials and Methods." A, wild-type Y2C fragment of topoisomerase II (wt) as well as a mutant form of the Y2C fragment that harbors a substitution of Thr-1342 with alanine (i.e. T1342A) were incubated with CK2 and ATP prior to detection on immunoblots with 3F3/2 antibodies. For this experiment, CK2 was partially purified by FPLC from extracts derived from asynchronously growing cells as described under "Materials and Methods." B, CK2 was partially purified by FPLC from extracts derived from cells arrested in mitosis as described under "Materials and Methods" and used to phosphorylate wild-type Y2C or T1342A mutant Y2C fragment of topoisomerase II using [32P]ATP in the presence or absence of the CK2 inhibitor heparin as indicated. Phosphate incorporation into Y2C fragments was detected using a PhosphorImager. C, phosphorylation of Thr-1342 within the wild-type Y2C fragment of topoisomerase II by partially purified CK2 was performed as in A in the presence of increasing amounts of GST or GST-Pin1 (as indicated) and detected on an immunoblot using the 3F3/2 monoclonal antibody. Partially purified mitotic CK2 was prepared as in B.

Because topoisomerase II is multiply phosphorylated and a potential target for kinases distinct from CK2 (31, 54-57), we also performed experiments to determine whether the Y2C fragment was phosphorylated by kinases other than CK2 present in the FPLC fraction. As seen in Fig. 7B, the CK2 inhibitor heparin completely abolished the incorporation of ³²P into both the wild-type and Thr-1342/Ala mutant Y2C fragments. This result suggests that no other kinases were present and phosphorylating the topoisomerase II fragment to any appreciable extent. Comparison of the wild-type Y2C topoisomerase II fragment with the Thr-1342/Ala mutant topoisomerase II fragment indicates that residues in addition to Thr-1342 are phosphorylated by CK2 (Fig. 7B). By utilizing the 3F3/2 antibody to examine the phosphorylation of Thr-1342, we examined the effect of GST-Pin1 on the ability of CK2 to phosphorylate this residue. A dose-dependent inhibition of Thr-1342 phosphorylation by CK2 was clearly observed (Fig. 7C). By comparison, the addition of GST generated no inhibition of Thr-1342 phosphorylation. Thus, Pin1 can inhibit the phosphorylation of Thr-1342 on topoisomerase II by CK2.

Investigation of the Mechanism by Which Pin1 Inhibits Phosphorylation of Topoisomerase II by CK2-- The peptidylprolyl isomerase activity of Pin1 is essential for its in vivo functions (9, 30). Consequently, there has been considerable speculation that Pin1 does indeed catalyze conformational changes in at least some of its target proteins. However, for the most part, direct evidence for Pin1-catalyzed conformational changes in its target proteins has not been obtained. One exception is the cell cycle regulatory phosphatase Cdc25 (58, 59). Lines of evidence demonstrating that Cdc25 undergoes conformational changes include the fact that its dephosphorylation by protein phosph