Development of a Stabilized Form of the Regulatory CK2β Subunit That Inhibits Cell Proliferation*

A number of cancers are characterized by elevated expression of CK2 (formerly casein kinase II), which has been implicated as a key component in cell proliferation and transformation. Two lines of evidence, (a) deregulated expression of CK2 and (b) CK2β ubiquitination and degradation of these in a proteasome-dependent manner prompted further investigation of the regulation of CK2β protein stability. We demonstrate that mutating six surface-exposed lysine residues to arginine (6KR) to interfere with ubiquitin attachment can stabilize CK2β. Examination of 6KR expression in cells revealed increased stability over time and increased its steady-state expression level compared with CK2β. In cells, 6KR was no longer sensitive to proteasome inhibition but maintained an elevated expression level. In our studies, 6KR functioned as a normal CK2 regulatory subunit, because it participated in CK2β dimerization, associated with catalytic subunits, was autophosphorylated, and formed active, stable CK2 tetramers. The physiological role of CK2β stabilization was investigated in cell proliferation assays, which showed a significant decrease in proliferation in cells expressing 6KR compared with CK2β. Overall, our results indicate that a stabilized form of CK2β can be used to inhibit cell proliferation.

Fundamental cellular processes such as proliferation and survival involve regulation by CK2 (formerly casein kinase II), 3 a serine/threonine protein kinase that is ubiquitously expressed in eukaryotic cells (1). Further evidence for its critical role is revealed in the absolute requirement for CK2 for viability in yeast and slime mold and in the requirements for CK2 in the G 1 /S and G 2 /M cell cycle transitions in yeast and mammalian cells (2)(3)(4)(5)(6)(7).
Given the diverse, yet essential, roles of CK2 within the cell, it is important to understand the mechanisms regulating CK2, which are equally as diverse and critical. Furthermore, perturbations in expression or activity of CK2 are associated with human disease. Abnormally high levels of CK2 have been observed in cancers of the breast, prostate, lung, head and neck, and kidney (8 -12). Overexpression of catalytic CK2 subunits led to increased proliferation and transformation. By comparison, overexpression of the regulatory CK2 subunit has been associated with decreased proliferation in yeast and mammalian cells, although this inhibitory role has not been universally observed (13)(14)(15). Collectively, these results indicate that CK2 has a profound effect on cell proliferation and suggests that individual CK2 subunits may exert competing effects.
CK2␤ is phosphorylated at serine 209 in a cell-cycle-dependent manner by p34 cdc2 in vitro and in mammalian cells (27)(28)(29)(30). CK2␤ autophosphorylation at serines 2 and 3 is mediated by the catalytic subunits of CK2 (28,31). Although neither phosphorylation event is completely understood, there are indications that autophosphorylation of CK2␤ enhances its stability (27). Detailed investigation of CK2␤ protein stability and turnover revealed that CK2␤ exhibits a biphasic degradation pattern (32). More specifically, CK2␤ is normally expressed at a higher level than the catalytic subunits of CK2, allowing some CK2␤ to be incorporated into CK2 tetramers and stabilized, whereas the excess CK2␤ is rapidly degraded with a half-life of less than 1 h (32). Furthermore, the observed ubiquitination of CK2␤ and the accumulation of CK2␤ protein upon proteasome inhibition suggested that polyubiquitination of CK2␤ targets it for degradation (27).
We hypothesized that the stability of CK2␤ could be altered by mutations that disrupt its ubiquitination. Alterations in CK2␤ protein levels may influence the mechanisms that govern the affect of CK2 on cell proliferation, which may be important in understanding the onset of cancer. In this report we successfully generated a form of CK2␤, designated 6KR, that exhibits altered stability in cells. Further studies characterize the 6KR protein and examine its effect on cell proliferation.
Antibodies-Polyclonal antibodies raised against CK2␤, CK2␣, and CK2␣Ј have been previously described (28,35). Monoclonal antibodies directed against the HA epitope (12CA5) and the biotinylated anti-HA (3F10) were purchased from Roche Applied Science. Monoclonal biotinylated anti-Myc (9E10) was purchased from Sigma. Polyclonal antibodies directed against green fluorescent protein (GFP) were purchased from Molecular Probes. Anti-␤-tubulin antibodies were a generous gift from Lina Dagnino (Dept. of Physiology and Pharmacology, University of Western Ontario). Goat-anti-rabbit and Goat-anti-mouse secondary antibodies conjugated to horseradish peroxidase or alkaline phosphatase were pur-chased from Bio-Rad. Monoclonal anti-biotin secondary antibody conjugated to horseradish peroxidase was purchased from Jackson ImmunoResearch.
Cell Culture and Transfections-COS7 (green monkey kidney) and HeLa (human cervical cancer) cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin, at 37°C in an atmosphere of 5% CO 2 . Cells were transiently transfected using the calcium phosphate precipitation method described previously (33). 16 -18 h after transfection, cells were washed and supplied with fresh medium. 24 -48 h later cells were harvested in cell lysis buffer (50 mM Tris-Cl, pH 7.5, 150 mM NaCl, 1% Nonidet P-40) supplemented with protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, 10 g/ml pepstatin A). Lysates were prepared and used directly or stored at Ϫ80°C.
Degradation Assays and Proteasome Inhibition Assays-Cells were co-transfected with HA-CK2␤ or HA-6KR and EGFP-C2 (transfection efficiency marker) as described above. After the DNA precipitate was washed from the cells, each plate of cells was used to seed the appropriate number of new plates so that protein levels could be compared under equivalent transfection efficiencies and allowed to recover for 24 h. For degradation assays, 10-cm plates of cells were treated with 95% ethanol (carrier) or with 150 g/ml cycloheximide, to inhibit protein synthesis. At various time intervals ranging from 0 to 3 h, cells were harvested and cell lysates prepared. For proteasome inhibition assays, cells were treated with Me 2 SO (carrier) or with final concentrations of N-carbobenzoxyl-Leu-Leu-leucinal (MG132), a proteasome inhibitor, ranging from 0.1 g/ml (1 M) to 10 g/ml (20 M). After 5 h, cells were harvested and cell lysates were prepared. Total protein concentrations of each lysate were determined, and 30 g of total protein was analyzed by SDS-PAGE and immunoblotting.
Immunoprecipitations and Binding Assays-Immunoprecipitations and binding assays were performed by transiently transfecting cells as indicated and preparing cell lysates as described above. Immunoprecipitations were preformed on equivalent amounts of total protein using protein-A-Sepharose and anti-HA(12CA5), anti-CK2␤, anti-CK2␣, or anti-CK2␣Ј antibodies as indicated and incubated for 1 h at 4°C with rotation. After washing, beads were used for CK2 kinase assays or proteins were eluted from the beads. Eluted proteins were analyzed by immunoblot as indicated.
For binding assays cell lysates with equivalent amounts of total protein were incubated with 100 l of nickel-Sepharose bead slurry (prepared according to the manufacturer's instructions) for 1 h at room temperature with rotation to specifically isolate histidine-tagged CK2␤ proteins. Nickel-Sepharose beads were collected, washed twice with 6 M guanidine-HCl, 12 mM imidazole, 0.1 M sodium phosphate, pH 8.0, twice with 8 M urea, 10 mM imidazole, 0.1 M sodium phosphate, pH 6.3, and twice with 50 mM sodium phosphate, pH 8.0, 100 mM KCl, 20% glycerol, 0.2% Nonidet P-40, 7 mM imidazole. Proteins were eluted by incubating the beads with 35 l of 2ϫ Laemmli sample buffer for 5 min. Eluted proteins were analyzed by immunoblot as indicated.
For ubiquitination experiments, lysates were prepared as described above in urea buffer (8 M urea, 0.1 M NaPO 4 , pH 8.0, 10 mM imidazole), and ubiquitinated complexes were isolated by incubating 1500 g of total protein with 40 l of Talon resin (BD Biosciences), equilibrated in urea buffer, for 2 h at 4°C. The Talon resin was washed thoroughly with urea buffer, and bound proteins were eluted by boiling with 40 l of 2ϫ Laemmli sample buffer. Eluted proteins were analyzed by immunoblot as indicated.
CK2 Kinase Assays-Immunocomplexes were prepared as described above. CK2 kinase assays were performed using enzyme immobilized on protein-A-Sepharose beads in 100 mM Tris-HCl, pH 7.6, 10 mM MgCl 2 , 150 mM NaCl, 0.1 mM ATP (specific activity 500 -1000 cpm/pmol) in a total volume of 30 l. The reaction was initiated by adding 2 mM ␣-casein to each sample. Control reactions contained 2 mM ␣-casein and were initiated by adding 2 g of GST-CK2␣Ј. Reactions were incubated at 30°C for 20 min and stopped by adding 30 l of 2ϫ Laemmli sample buffer. Samples were boiled and separated by SDS-PAGE, and dried gels were exposed to phosphor screens overnight and scanned. Incorporation of 32 P into ␣-casein was analyzed, which is directly proportional to the catalytic activity exhibited by the immunoprecipitated CK2 complexes.
Flow Cytometry-For cell cycle analysis, HeLa cells were transfected as described above with HA-CK2␤ or HA-6KR and EGFP-C2. At the indicated times (ranging from 1 to 5 days) cells were collected using PBS containing 5 mM EDTA, resuspended in PBS containing 2 mM EDTA, resuspended in PBS, filtered using a 40-m nylon filter, fixed in 1% paraformaldehyde, washed with PBS and permeabilized by dropwise addition into cold 70% ethanol and stored at Ϫ20°C until FACS analysis. Cells were stained with propidium iodide (50 g/ml), 0.1% sodium citrate, 0.1% Triton X-100, 0.1 mg/ml DNase-free RNase A by incubating for 30 min at 37°C in the dark and filtered again. 50,000 cells were analyzed on a BD Biosciences FACSCalibur cytometer at a flow rate of less than 200 cells/s. Cell cycle profiles were generated using CellQuest data acquisition and FlowJo cytometry analysis software.
Cell Proliferation Assays-HeLa cells maintained in Dulbecco's modified Eagle's medium were plated at a density of 2.5 ϫ 10 5 cells/6-cm plate. Cells were co-transfected, using ExGen500 (MBI Fermentas), with 4.5 g of p27, HA-CK2␤, HA-6KR, or pcDNA3.1(ϩ) and 0.5 g of pBABEpuro, a puromycin resistance plasmid. Cells transfected only with 5 g of pcDNA3.1(ϩ) were used as a negative control. Media was changed 6 h after transfection and 3 h later cells were selected for 16 h with 2 g/ml puromycin. Five days after transfection, plates were washed with PBS and stained in 0.2% (w/v) methylene blue in methanol at room temperature for 20 min followed by washing with ddH 2 O. Images of each plate were converted to grayscale using Adobe Photoshop 7.0. The number of gray pixels (defined as pixels whose levels range from 50 to 200) out of the total number of pixels per image was used to determine colony density. 35 S Labeling-HeLa cells were transiently transfected as described with HA-CK2␤, HA-6KR, pBI(CK2␣-HA/Myc-CK2␤), pBI(CK2␣-HA/Myc-6KR), or left untransfected. Cells were maintained in Dulbecco's modified Eagle's medium without tetracycline to induce protein expression from the bidirectional constructs. To label, cells were washed four times with PBS, once with Dulbecco's modified Eagle's medium without methionine, and incubated for 15 min at 37°C in Dulbecco's modified Eagle's medium without methionine. Cells were labeled with 200 Ci of [ 35 S]methionine per plate of cells for 18 h at 37°C. Lysates were prepared in 500 l of cell lysis buffer, and 150 g of total protein was used for each immunoprecipitation. Immunoprecipitations were performed as described above except that anti-HA (12CA5) was coupled to the protein-A-Sepharose beads using standard protocols. To visualize labeled proteins, eluted samples were run on an SDS-PAGE gel, fixed in 50% methanol, 10% glacial acetic acid, soaked in Enhance solution (PerkinElmer Life Sciences), soaked in cold ddH 2 O, dried and exposed to autoradiography film for 20 h or more.

Analysis of Potential Ubiquitination Sites in CK2␤-Pre-
vious studies demonstrated that CK2␤ is ubiquitinated and degraded by the proteasome (27). To further define the mechanisms controlling CK2 regulation, we were interested in determining whether CK2␤ protein stability could be altered by systematically removing potential ubiquitination sites. Alignment of the primary amino acid sequence of CK2␤ from various organisms revealed that each of the nine lysine residues present in human CK2␤ are highly conserved among species (Fig. 1A). Examination of the crystal structure of CK2␤ identified six lysines (including 33,139,177,191,208, and 212) that appear to be surface-exposed and prime targets for ubiquitination (Fig. 1B). Of these six lysines, 33 and 177 protruded further from the surface of CK2␤ than 139 and 191. Lysines 208 and 212, not present in the crystal structure, are likely surface-exposed because they are Stabilized CK2␤ Inhibits Proliferation OCTOBER 5, 2007 • VOLUME 282 • NUMBER 40 located on the extreme C-terminal tail. The remaining three lysines, 100, 134, and 147, are buried within the structure and are likely important for maintaining CK2␤ protein structure. Based on this analysis, the six surface-exposed lysines in CK2␤ were individually mutated to arginine to remove potential ubiquitin attachment sites while maintaining overall protein charge. Fig. 1C illustrates the location of each lysine in CK2␤ with respect to its functional domains. It is important to note that the HA tag used to facilitate detection of CK2␤ did not contain any additional lysine residues. Another important consideration is the low site specificity characteristic of the ubiquitination process (37). Based on this knowledge, we hypothesized that removal of multiple surface-accessible lysines and perhaps even all of the lysines may be required to abrogate ubiquitin-dependent degradation of CK2␤. To facilitate examination of this possibility a number of other multiple lysine mutants lacking 5, 6, 7, 8, or all 9 lysine residues were created (Fig. 1C).
Comparison of CK2␤ Single Lysine Mutant Expression and Degradation-To begin examining the importance of each lysine residue in the regulation of CK2␤ protein levels, the steady-state expression levels of the single lysine mutants were compared. As shown in Fig. 2A, each mutant was expressed at comparable levels to wild-type CK2␤, with the exception of HA-K212R, in COS7 cells. No protein was detected in mock transfected (control) or pcDNA3.1(ϩ) (vector) transfected cells. To ensure equal loading, membranes were stripped and reprobed with anti-␤tubulin as shown in the lower panel ( Fig. 2A). Similarly, mutation of single surface-exposed lysines did not significantly alter the sensitivity of CK2␤ to degradation by the proteasome, as indicated by the accumulation of each mutant upon treatment with MG132 (Fig. 2B). Collectively, these results suggest that substitu- showing the lysine residues. The catalytic subunits are shown in purple, whereas one regulatory subunit is shown in gray and the other are colored according to the accessibility of each amino acid to the protein surface (least accessible ϭ blue Ͻ cyan Ͻ green Ͻ yellow ϭ most accessible). Lysine residues are shown in red. Each surface-exposed lysine is denoted with an asterisk. Image was generated using Swiss PDB Viewer (1JWH) (39,40). C, schematic representation of N-terminally HA-tagged human CK2␤, indicating the zinc-finger region (Zinc-finger) and the positive regulatory region (P.R. region). Black arrows indicate the position of each lysine with respect to other functional domains in CK2␤. The lysines that were mutated in each of the multiple lysine mutants are summarized in the table, where the surface exposed lysines are underlined.
tion of a single lysine residue is insufficient to alter CK2␤ stability by preventing ubiquitin attachment.
Comparison of CK2␤ Multiple Lysine Mutant Expression and Degradation-Because substitution of single lysines was ineffective in stabilizing CK2␤, we investigated the effects of multiple lysine mutations in CK2␤ on steady-state expression levels. Each of the double lysine mutants, HA-K177R/K139R and HA-K177R/K191R, as well as HA-5KR, HA-6KR, and HA-7KR, exhibited increased steady-state expression compared with HA-CK2␤ (Fig. 3A). Interestingly, HA-5KR, HA-7KR, and particularly HA-6KR showed an increase in the level of autophosphorylated protein, as evidenced by a band with diminished electrophoretic mobility, compared with HA-CK2␤ and even the double lysine mutants. HA-8KR and HA-9KR, exhibited decreased protein expression, possibly due to improper protein folding and subsequent rapid degradation of these mutants (Fig. 3A).
Two or more lysine to arginine mutations resulted in little to no protein accumulation upon MG132 treatment (Fig. 3B) indicating that mutation of as few as two lysine residues could compromise the ability of CK2␤ to be degraded. Particularly striking was the complete abrogation of proteasome-dependent degradation of CK2␤, observed with mutation of 5, 6, or 7 lysine residues.
The HA-6KR mutant was selected for further investigation because it demonstrated the highest increase in expression compared with HA-CK2␤, and it showed the greatest accumulation of autophosphorylated protein. Furthermore, HA-6KR was representative of a group of stabilized forms of CK2␤ (HA-5KR, HA-6KR, and HA-7KR), because modification of its six surface-exposed lysines removed the most potential ubiquitination sites but would have the least impact on overall CK2␤ structure.
Examination of the Effect of Proteasome Inhibition on 6KR-Because HA-6KR protein was stabilized in cells, we hypothesized that it may no longer be ubiquitinated and therefore no longer sensitive to proteasome inhibition. To test this hypothesis, COS7 cells transfected with HA-CK2␤ or HA-6KR were treated with increasing concentrations of MG132. No exogenously expressed HA-CK2␤ or HA-6KR protein could be detected in mock transfected (control) or pcDNA3.1(ϩ) (vector) transfected cells. Whereas HA-CK2␤ protein accumulated with increasing concentrations of MG132, HA-6KR protein did not, confirming that HA-6KR was not affected by proteasome inhibition (Fig. 3C). Furthermore, increased levels of autophosphorylated HA-6KR compared with HA-CK2␤ suggested that HA-6KR is readily incorporated into CK2 tetramers and supramolecular complexes where it is subsequently autophosphorylated. This observation could also indicate that HA-6KR is a less desirable protein phosphatase substrate.
To examine the ubiquitination of CK2␤ and 6KR, cells were transfected with His-CK2␤ or His-6KR in the presence or absence of HA-ubiquitin (HA-Ub). Lysates were incubated with a metal affinity resin, to isolate His-tagged protein and any covalently bound proteins. No protein was detected in untransfected (not shown), His-CK2␤, or His-6KR transfected samples (Fig. 3D, left panel). However, when cells were transfected with both HA-Ub and His-CK2␤ a set of ubiquitinated His-CK2␤ bands were detected. Importantly, these bands increase in intensity upon MG132 treatment and are located at molecular weights that correspond to increasingly ubiquitinated forms of CK2␤. Co-transfection of His-Ub and HA-6KR resulted in a significantly lower amount of ubiquitinated HA-6KR compared with wild type in contrast to the higher levels of HA-6KR that are consistently observed (Fig. 3D). To perform the reciprocal experiment, cells were transfected with HA-CK2␤ or HA-6KR in the presence or absence of His-Ub and incubated with metal affinity resin (Fig. 3D, right panel). Again, an increase in ubiquitinated protein upon MG132 treatment was observed for HA-CK2␤ but not HA-6KR. Taken together, these results suggest that mutation of the six surface-exposed lysines of CK2␤ compromise the ability of CK2␤ to be ubiquitinated and degraded.
Examination of 6KR Protein Stabilization Over Time-To evaluate the stability of HA-6KR compared with HA-CK2␤ over time, COS7 cells transfected with HA-CK2␤ or HA-6KR were treated with cycloheximide to inhibit all subsequent protein synthesis. Endogenous CK2␤ did not significantly degrade upon cycloheximide treatment (Fig. 3E), which was expected, because equilibrium between levels of CK2 tetramer and free subunits would have been achieved prior to transfection. HA-CK2␤ protein significantly degraded after 1 h of cycloheximide treatment, whereas HA-6KR did not degrade even after 3 h, confirming that HA-6KR protein had indeed been stabilized in cells (Fig. 3E). To ensure equal loading, membranes were stripped and reprobed with anti-␤tubulin as shown in the lower panel (Fig. 3E).
Characterization of 6KR as a Regulatory Subunit of CK2-To assess the integrity of 6KR, we examined its ability to form complexes with other CK2 subunits. Using metal affinity binding experiments the ability of HA-6KR to bind CK2␤ was addressed. No protein was detected in untransfected, or cells transfected with only HA-CK2␤, His-CK2␤, or His-6KR (Fig. 4A). When both His-CK2␤ or His-6KR and HA-CK2␤ were present binding was detected. Efficient affinity purification of His-tagged CK2␤ or 6KR was demonstrated using anti-CK2␤ antibodies (lower panel). The ability of 6KR to bind another 6KR molecule was examined by immunoprecipitating HA-CK2␤ or HA-6KR from cell lysates expressing both HA and Myc tagged CK2␤ or 6KR. Similar to wild-type CK2␤, only when both Myc-6KR and HA-6KR were present could binding be detected (Fig. 4B). These results clearly indicate that 6KR is capable of forming CK2␤ dimers in cells, suggesting that it should be incorporated into tetrameric complexes.
To assess the ability of 6KR to bind the catalytic CK2 subunits several experiments were conducted. COS7 cells transfected with either HA-CK2␤ or HA-6KR in the presence or absence of GFP-CK2␣ were immunoprecipitated with anti-HA antibodies and immunoblotted with GFP and CK2␣ antibodies (Fig. 4C,  upper and middle panels). Distinct GFP-CK2␣ bands could only be detected when both GFP-CK2␣ and HA-CK2␤ or HA-6KR were present (Fig. 4C). Cell lysates were immunoprecipitated with anti-CK2␣ antibodies, isolating both endogenous CK2␣ and GFP-CK2␣ followed by immunoblotting with anti-HA (Fig.  4C, lower panel). Again, both HA-CK2␤ and HA-6KR were capable of forming complexes with CK2␣ with similar efficiency. Similarly, the ability of HA-CK2␤ or HA-6KR to form complexes with endogenous CK2␣ or CK2␣Ј was examined by immunoprecipitation. HA-CK2␤ and HA-6KR were able to form complexes with both endogenous CK2␣ and CK2␣Ј (Fig.  5A). Collectively these data suggest that 6KR is able to form functional CK2 tetramers, because there was a clear association between both exogenous and endogenous CK2␣ (or endogenous CK2␣Ј) and HA-6KR and because autophosphorylated forms of 6KR were also detected.
Immunokinase assays were used to examine the ability of 6KR to form active CK2 complexes in cells. Lysates from COS7 cells transfected with HA-CK2␤ or HA-6KR, were immunoprecipitated with anti-HA antibodies and immunocomplexes were utilized in kinase assays with ␣-casein as a substrate. No phosphorylated ␣-casein was detected in anti-HA immunoprecipitations from mock transfected (control) cells (Fig. 5B). However, immunoprecipitates from HA-CK2␤ or HA-6KR transfected cells, efficiently phosphorylated ␣-casein, indicating formation of active CK2 complexes. An anti-CK2␤ immunoblot shows that HA-CK2␤ or HA-6KR protein was present in each reaction and that a small amount of endogenous CK2␤ protein, presumably pulled down in heterogeneous CK2 complexes, was also present (lower panel). These results indicate not only that 6KR is able to form active CK2 complexes in cells, but also that 6KR does not inhibit CK2 kinase activity.
Effect of 6KR on Cell Cycle-Since some previous studies suggest that high levels of CK2␤ protein affects cell proliferation, and because we have successfully created a degradation resistant form of CK2␤ we began to investigate the effect of 6KR on the cell cycle. To facilitate these studies, HeLa cells were employed. As in COS7 cells, the steady-state level of 6KR and phosphorylated 6KR protein was elevated compared with CK2␤ (Fig. 6A). As a first step toward investigating the effect of 6KR on the cell cycle we hypothesized that a cell cycle arrest may be occurring. HeLa cells were transfected with HA-CK2␤ or HA-6KR and EGFP (transfection marker) and collected 1, 2, 3, and 5 days after transfection. After propidium iodide stain-   OCTOBER 5, 2007 • VOLUME 282 • NUMBER 40

Stabilized CK2␤ Inhibits Proliferation
ing, the cell cycle profile of enhanced GFP and PI-positive cells were analyzed by FACS for each time point. No significant difference in cell cycle progression could be detected (Fig. 6B).
Effect of 6KR on Cell Proliferation-To evaluate whether HA-6KR affects cell proliferation, HeLa cells were transfected with HA-CK2␤ or HA-6KR along with a puromycin resistance gene followed by selection with puromycin until colonies formed (7 days). Colonies were stained with methylene blue to assess the proliferative effects. Cells transfected with the puromycin resistance gene alone survived puromycin selection and formed colonies, while the cells lacking the puromycin resistance gene did not survive selection (Fig.  6C). Cells were also transfected with p27 KIP1 , a cell cycle inhibitor whose overexpression blocks proliferation through G 1 arrest, but does not affect viability (38). Cells remaining on this plate represent those that were transfected and thus were resistant to puromycin treatment but were unable to proliferate due to p27 KIP1 . These cells represent the experimental background. Transfection with HA-CK2␤ had no effect on cell proliferation while HA-6KR efficiently inhibited proliferation (Fig. 6C). Quantitation revealed that HA-6KR reduced proliferation to ϳ25% of control or HA-CK2␤ cells in comparison to the p27 KIP1 -expressing cells (Fig. 6D). Transfection with CK2␣-HA or HA-CK2␣Ј alone had no effect on cell proliferation. These results suggest that stabilization of CK2␤ leads to a significant decrease in cell proliferation that is not mediated by the catalytic subunits and may have implications in cellular processes such as the cell cycle, apoptosis, or in cancer.
Investigation of the Cell Proliferation Defect-Inhibition of cell proliferation upon stabilization of CK2␤ raises numerous questions relating to whether the defect is related to the CK2 dependent or independent functions of CK2␤. For example 6KR could be exerting its effect through interactions with other interactors such as A-Raf, Chk1, c-mos, or other proteins. Alternatively, the decrease in cell proliferation could be due to the effect that stabilization has on the CK2 complex.
To address both of these issues we conducted [ 35 S]methionine cell labeling experiments on HeLa cells transfected with HA-CK2␤, HA-6KR, CK2␣-HA and Myc-CK2␤, CK2␣-HA and Myc-6KR or left untransfected (control). After an anti-HA immunoprecipitation, the bound, labeled proteins were sepa- rated and visualized by autoradiography. As expected, there was a significant increase in the steady-state expression level of HA-6KR both in the presence and absence of CK2␣ (Fig. 7). Interestingly, there was a significant increase in the amount of phosphorylated HA-6KR but not HA-CK2␤ in the presence of CK2␣ confirming that HA-6KR readily forms active CK2 complexes consistent with our previous data. It is also interesting to note that there is no dramatic increase in the amount of CK2␣ in the presence of HA-6KR compared with HA-CK2␤ suggesting that there has not been an increase in overall CK2 complex expression due to the presence of HA-6KR. In addition we did not observe the presence of any obvious interaction partners by which 6KR might be mediating its effect on cell proliferation. Overall, these observations suggest that the effect of 6KR does not arise from stabilization of the CK2 complex or from major changes in interactions with other proteins but does not exclude the possibility that 6KR is influencing the actions of downstream targets or CK2 selectivity.

DISCUSSION
Previous studies indicated that CK2␤ was ubiquitinated in cells and degraded by the proteasome (27). Given this information, we hypothesized that removal of one or more lysine residues would stabilize CK2␤ protein in cells by preventing the covalent attachment of ubiquitin. Examination of the crystal structure of CK2␤ revealed that of the nine lysine residues in CK2␤, six were surface accessible, making them ideal sites for ubiquitin attachment (39,40).
Although mutation of each surface-exposed lysine on its own showed no significant alterations in steady-state levels compared with wild-type CK2␤, mutation of five or more lysines (with the exception of lysine 134 and 147) led to elevated steady-state expression levels and loss of sensitivity to proteasomal degradation. Because ubiquitin attachment is not highly site-specific, the fact that mutation of several lysine residues was required to prevent CK2␤ protein degradation by the proteasome is not surprising. The greatest enhancement in steadystate protein levels in cells occurred with the 6KR mutant, in accordance with the hypothesis that modification of the surface accessible ubiquitin attachment sites would prevent proteasomal degradation.
Previous studies demonstrated that only a portion of newly synthesized CK2␤ incorporates into CK2 tetramers and is stabilized, whereas free CK2␤ is rapidly degraded, establishing steady-state levels of CK2␤ within the cell (32). Given these findings it is postulated that overexpression of CK2␤ alone would result in the majority of new CK2␤ being rapidly degraded, while endogenous CK2␤ would remain stably associated with the CK2 catalytic subunits. In this context, the relative degradation rate of CK2␤ and 6KR was particularly interesting. Three hours after inhibition of protein synthesis, CK2␤ protein levels had significantly decreased while the 6KR protein level had not. Taken together, these results and the observation that 6KR was no longer sensitive to proteasome inhibition suggest that mutation of surface exposed lysines on CK2␤ (6KR) increases its half-life to at least 3 h and ablates its susceptibility to proteasomal degradation, resulting in a stabilized protein.
While 6KR results in stabilization of CK2␤, other characteristics of CK2␤ are not altered. The 6KR protein forms dimers with wild-type and 6KR molecules indicating that the zing-finger region is not disrupted by mutating lysine 139 to arginine within this region. Furthermore, 6KR could associate with GFP-CK2␣ as well as endogenous CK2␣ or CK2␣Ј, demonstrating that neither lysine 208 nor 212, in the positive regulatory region of CK2␤, are key residues mediating binding with the catalytic subunits. In addition, slightly slower migrating bands, representing autophosphorylated forms of CK2␤ were detected in lanes where CK2 complex formation was observed. Because auto-phosphorylation of CK2␤ most likely occurs within a supramolecular complex, this implies that CK2 tetramers formed are likely functional. Phosphorylation of ␣-casein in 6KR immunoprecipitates provides further evidence that active, stable CK2 tetramers containing 6KR had formed in cells.
Creation of a stabilized form of CK2␤ that appears to function normally as a regulatory subunit of CK2 is intriguing given the involvement of CK2␤ in cancer and its CK2-dependent and CK2independent roles. Although this achievement is significant on its own, we wondered what the physiological effects of CK2␤ stabilization might be. The traditional tetrameric view of CK2, that CK2␤ functions within the tetramer to modulate CK2 activity, has underscored the importance of understanding the function and regulation of CK2␤. Identification of catalytic subunit independent interactors of CK2␤ (A-Raf, c-Mos, and Chk1) as well as differences in subcellular localization of the subunits are just two of the lines of evidence supporting CK2 independent roles for CK2␤. This evidence suggests a need for investigating how CK2␤ protein levels are regulated and the cellular consequences of disrupting this regulation.
In past studies, the effect of CK2␤ overexpression on fundamental cellular processes such as cell proliferation yielded conflicting results. One study suggested that overexpression of CK2␤ led to inhibition of cell proliferation in Chinese hamster ovary (CHO) cells and that this defect was due to disruption of the G 1 phase of the cell cycle (13). By comparison, subsequent studies using human osteosarcoma (U2OS) cells stably expressing CK2␤ and mouse 3T3-L1 cells, found that CK2␤ overexpression had no effect on cell proliferation (14,15). In this report we show that overexpression of wild-type CK2␤ in HeLa cells does not affect cell proliferation. However, overexpression of 6KR, which is degradation resistant, leads to a significant decrease in cell proliferation. Because CK2 holoenzyme levels do not significantly change when 6KR is present it is suspected that the anti-proliferative effects are due to 6KR rather than the CK2 holoenzyme. This is supported by the proliferation assays where neither CK2␣-HA nor HA-CK2␣Ј affect proliferation on their own.
Elevation of CK2␤ protein levels alone however, may not influence cell proliferation rate, but rather, it may be a downstream consequence of CK2␤ protein level elevation. In our cell labeling experiments we did not observe any obvious interaction partners of 6KR that could explain the inhibition of cell proliferation. Similarly in large-scale immunoprecipitation experiments we did not observe any major bands that increased or decreased in the presence or absence of 6KR (data not shown) suggesting that rather than affecting one or two protein interactions, the effect may be more widespread in its consequences.
The involvement of CK2 at almost all stages of the cell cycle and in particular the G 0 /G 1 , G 1 /S, and the G 2 /M transitions raises the prospect that expression of 6KR could alter the rate of cell cycle progression (3,6,7,44,45). In light of these complex roles of CK2 in the control of proliferation, it is perhaps not surprising that we did not observe any specific changes in cell cycle profiles of cells transfected with 6KR as compared with cells transfected with CK2␤. In a similar respect, it is also intriguing that CK2 has both apoptotic and anti-apoptotic roles, both of which may be affected by stabilization of CK2␤ (46). Examination of FACS profiles (as illustrated in Fig. 6B) does not provide any indication of increased apoptosis in cells expressing 6KR as illustrated by the absence of cells exhibiting a sub G 0 /G 1 DNA content suggesting that apoptosis is not the sole event responsible for the attenuated proliferation observed with 6KR. Overall, given that CK2 has been implicated in a broad series of cellular events linked to proliferation, it is very likely that a number of distinct pathways and/or CK2 substrates contribute to the effects that we observe. Accordingly, elucidation of the precise mechanistic basis for the role of 6KR will require detailed systematic studies.
Based on the recent emergence of CK2-independent roles for CK2␤, stabilization of the CK2␤ subunit also has the potential to significantly affect a number of other signal transduction pathways through binding to A-Raf, c-Mos, or Chk1 (47)(48)(49)(50)(51). At this point it is not known whether 6KR will retain its ability to associate with each of these protein kinases, however, because 6KR retained binding to the catalytic subunits of CK2 it is anticipated that 6KR will also form complexes with A-Raf, c-Mos, and Chk1. Assuming that each kinase is able to bind 6KR, it is expected that expression of 6KR will elevate A-Raf activity toward MEK. Similarly 6KR expression is also expected to elevate Chk1 activity. Indeed Chk1 lacking its C-terminal region, which is hypothesized to have an autoinhibitory role, is 20-fold more active, and it is currently thought that binding of CK2␤ to full-length Chk1 alleviates this autoinhibition (50). Furthermore, studies by Chen and colleagues proposed that CK2␤ binds and inhibits c-Mos during its initial synthesis (48), and this inhibition is alleviated once c-Mos overcomes a certain threshold that is set by the amount of free CK2␤ available (48), thus allowing for activation of the MAPK pathway. Accordingly, this would make it possible that expression of stabilized CK2␤ would disrupt c-Mos signaling by significantly elevating the threshold level of free CK2␤ that must be overcome by c-Mos to activate the MAPK pathway.
In closing, our data demonstrate that a stabilized form of CK2␤ can be generated by mutating its six surface-accessible lysines to arginine. Evaluation of this mutant demonstrates that, unlike the catalytic subunits of CK2, which promote proliferation, expression of this stabilized form of CK2␤ can be used to inhibit proliferation. Given the evidence that CK2 is overexpressed in tumors and that the catalytic CK2 subunits exhibit oncogenic activity and promote transformation, our observation that a stabilized form of CK2␤ is more effective than wild-type CK2␤ in inhibiting proliferation raises interesting prospects for targeting CK2 for therapeutic intervention.