This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services 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 Romieu-Mourez, R. Articles by Sonenshein, G. E. Search for Related Content PubMed PubMed Citation Articles by Romieu-Mourez, R. Articles by Sonenshein, G. E.

Protein Kinase CK2 Promotes Aberrant Activation of Nuclear Factor-B, Transformed Phenotype, and Survival of Breast Cancer Cells¹

Departments of Biochemistry [R. R-M., G. E. S.], and Medicine [E. L. B., D. C. S.], and the Program in Research on Women’s Health [R. R-M., E. L-B., D. C. S., G. E. S.], Boston University School of Medicine, Boston, Massachusetts 02118-2394

	ABSTRACT

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The Her-2/neu oncogene, the second member of the epidermal growth factor (EGF) receptor family, encodes a transmembrane tyrosine kinase receptor. Overexpression of Her-2/neu in 30% of breast cancers is associated with poor overall survival. Recently, we have found that Her-2/neu activates nuclear factor (NF)-B via a phosphatidylinositol 3 kinase (PI3-K)-Akt kinase signaling pathway in mouse mammary tumor virus (MMTV)-Her-2/neu NF639 mouse breast cancer cells. Surprisingly, the IB kinase (IKK) kinase complex, implicated in proteasome-mediated degradation of IB- and activation of NF-B via the canonical pathway, was not activated in these cells. Degradation of IB- was mediated via calpain, which in B cells is facilitated by phosphorylation of IB- by the protein kinase CK2. Here, we report that the inhibition of CK2 blocks Her-2/neu-mediated activation of NF-B. NF639 breast cancer cells, stably expressing CK2 or CK2' kinase-inactive mutants, displayed decreased NF-B binding and reduced ability to grow in soft agar, as well as increased sensitivity to tumor necrosis factor (TNF)- killing. Similarly, CK2 kinase-inactive subunits inhibited NF-B activity in Hs578T human breast cancer cells, which also display elevated CK2 activity. In NIH 3T3 fibroblasts, which express low basal NF-B and CK2 activities, overexpression of CK2 by retroviral gene delivery led to increased IB- turnover and the induction of classical NF-B (p50/RelA). Thus, CK2 plays an important role in Her-2/neu signaling, promoting IB- degradation and, thereby, NF-B activation. Furthermore, because ectopic CK2 activity appears sufficient to induce NF-B, the elevated CK2 activity observed in many primary human breast cancers likely plays a role in aberrant activation of NF-B and, therefore, represents a potential therapeutic target.

	INTRODUCTION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
NF-B³ /Rel is a family of dimeric transcription factors distinguished by the presence of a 300-amino-acid region, termed the Rel homology domain (1) . The NF-B family includes five known members in mammals: p50/p105, p52/p100, c-Rel, RelB, and RelA (p65). Classical NF-B complexes are composed of p50/RelA heterodimers. In most untransformed cells, other than B lymphocytes, NF-B/Rel proteins are sequestered in the cytoplasm bound to the specific IB-inhibitory proteins, of which IB- is the paradigm. A variety of agents that activate NF-B, e.g., TNF- and interleukin 1, mediate degradation of IB- via a canonical pathway involving phosphorylation of IB on two NH₂-terminal serine residues by a large multisubunit complex composed of the two IKKs IKK and IKKß (2 , 3) . This phosphorylation is followed by ubiquitination and rapid proteasome-mediated degradation of IB, allowing for translocation of free NF-B to the nucleus. In addition, phosphorylation of IB by protein kinase CK2 (formerly casein kinase II) has been implicated in basal IB- degradation (4, 5, 6, 7, 8) . CK2-mediated phosphorylation of IB- occurs preferentially at Ser-283, Ser-289, Thr-291, and Ser-293 within the COOH-terminal PEST domain, and mutations of these sites prolong IB- half-life (5 , 7 , 8) . In lymphoid cells, an alternative calpain-mediated IB- degradation pathway was described that contributed to the constitutive NF-B activation seen in these cells (9) . Moreover, we demonstrated that CK2 phosphorylation accelerates calpain-mediated degradation of IB in B cells (10) . Thus, two alternative phosphorylation signals and proteolytic systems regulate degradation of IB and activation of NF-B in what appears to be a signal-and cell type-specific manner.

CK2 is a highly conserved serine/threonine kinase that recognizes the general consensus sequence (S/T)XX(D/E). CK2 is a constitutively active kinase that is ubiquitously expressed in both the cytoplasm and nucleus of eukaryotic cells and exists in cells as a holoenzyme containing two catalytic (CK2 or CK2') and two regulatory (CK2ß) subunits. The two catalytic subunits are highly homologous, but CK2' has a unique required role in spermatogenesis (11) . The crystal structure of human CK2₂ß₂ was recently solved, showing flexible interdomain and intersubunit interactions in which each catalytic subunit makes no direct contact with the other and each interacts with both regulatory subunits, via the NH₂-terminal lobe of the catalytic subunit and an extended COOH-terminal tail of the regulatory subunit (12) . CK2 is essential for cell viability, and many of the >160 CK2 substrates identified thus far are growth- and cell cycle-related (13) . Overexpression or inhibition of CK2, CK2', or CK2ß subunits was shown to affect proliferation; however, results varied greatly with cell type (14, 15, 16, 17) . Overexpression of CK2 has been observed in many human cancers, including breast cancers (18, 19, 20, 21) . Furthermore, we have shown that enforced CK2 expression in transgenic mice is sufficient to induce T-cell lymphomas (22) and breast cancer (19) .

We and others have demonstrated aberrant activation of NF-B factors in breast cancer (23, 24, 25, 26, 27) . High levels of nuclear NF-B/Rel were found in human breast tumor cell lines and in the majority of primary human and rodent breast tumor tissue samples. In contrast, untransformed breast epithelial cells and normal mammary glands contain low basal levels of nuclear NF-B/Rel. In breast cancer cells, elevated levels of NF-B correlated with a decrease in the half-life of IB- protein (25) . We observed that many primary breast tumor tissue samples and human cancer cells display an increase in either CK2 or IKK activity (20) .

Recently, we have found that the Her-2/neu protein activates NF-B via a PI3-K- to Akt kinase-signaling pathway that can be inhibited via the tumor suppressor PTEN (28) . The Her-2/neu (or c-erbB-2) oncogene, the second member of the EGF receptor family (EGFR-2), encodes a transmembrane tyrosine kinase receptor. Overexpression of Her-2/neu, which is seen in 30% of breast cancers, is associated with poor overall survival, increased metastatic potential and resistance to chemotherapeutic agents (29) . Surprisingly, degradation of IB- in the MMTV-Her-2/neu-derived mammary tumor cells NF639 (30) did not appear to be mediated via the IKK complex or the proteasome but, rather, was blocked on addition of calpain inhibitors (28) . This raised the question of involvement of CK2 in activation of NF-B in NF639 cells, which we have addressed here. Expression of kinase-inactive CK2 catalytic subunit mutants decreased basal NF-B activity in NF639 cells, as well as in Hs578T human breast cancer cells, which were shown to overexpress CK2 (20) . Furthermore, the inhibition of CK2 enhanced susceptibility to cell death and inhibited transformed phenotype of NF639 cells. Conversely, ectopic CK2 was sufficient to induce NF-B activity in NIH 3T3 fibroblasts, which express low basal NF-B and CK2 activities. These studies indicate that CK2 directly controls NF-B activity, which, in turn, modulates the survival and transformed phenotype of breast cancers, including those in which Her-2/neu is amplified.

	MATERIALS AND METHODS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cell Culture and Treatment Conditions.
The NF639 MMTV-c-neu mouse mammary tumor cell line (kindly provided by P. Leder, Harvard Medical School, Boston, MA) was cultured as described previously (30) . NMuMG is an untransformed, immortalized mouse mammary epithelial cell line and was cultured as described previously (31) . The Hs578T human breast tumor cell line, which was derived from a carcinosarcoma and is epithelial in origin, was grown as described previously (24) . NIH 3T3 mouse fibroblast, 293T human embryonic kidney cell line, and Phoenix packaging cell lines were cultured in DMEM supplemented with 10% FCS and penicillin-streptomycin (Invitrogen Life Technologies, Inc., Carlsbad, CA). For protein synthesis inhibition, exponentially growing cells were treated with 30–50 µg/ml cycloheximide (Sigma-Genosys Inc., The Woodlands, TX).

Transfection Conditions.
For transfections, Fugene reagent was used according to the manufacturer’s directions (Roche Diagnostics Corporation, Indianapolis, IN). To evaluate transcriptional activity, cells were transfected in triplicate with an NF-B element-luciferase reporter vector, driven by three B elements from upstream of the MHC class I promoter, kindly provided by A. Chan (Mt. Sinai School of Medicine, New York City, NY; Ref. 32 ). The SV40-promoter ß-gal (pSV40-ß-gal) reporter vector was used to normalize transfection efficiency. Luciferase assays were performed as described previously (33) . Statistical analyses were performed using the Student’s t test. The pSVK3-human IB- WT, 2N (S32A and S36A), 3C (S283A, T291A and T299A), and 2N3C (S32A, S36A, S283A, T291A, and T299A) plasmid vectors were a kind gift from J. Hiscott (Institut Lady Davis de Recherches Medicales; Ref. 5 ). The pRc/CMV2 HA-tagged CK2 (pZW2), pRc/CMV2-HA-CK2 K68M (pGP8), pRc/CMV2-HA-CK2' (pGP3), and pRc/CMV2-HA-CK2' K69M (pGP18) vectors, and the backbone empty vector were provided by D. W. Litchfield (University of Western Ontario, Ontario, Canada; Ref. 34 ). The CK2 K68M or CK2' K69M mutants contain a single point mutation in the kinase domain of CK2 and CK2' catalytic subunits, respectively, and are devoid of kinase activity (34) . The pRc/CMV2 plasmid contains a neo-resistance gene driven by the CMV promoter (Invitrogen Life Technology). To establish stable transfectants, P100 dishes of cells were transfected with 20 µg of the appropriate pRc-CMV constructs. After 48 h, cells were selected with 600 µg/ml geneticin (Invitrogen Life Technology) for 10 days, and then grown in the presence of 100 µg/ml geneticin. Alternatively, cells were cotransfected with 1 µg of the pGKpuro selection plasmid, selected with 4 µg/ml puromycin (Sigma) for 4 days, and then grown in the presence of 1 µg/ml puromycin.

Retroviral Gene Delivery.
The murine CK2 cDNA (22) was excised from the pcDNA3.0 vector (Invitrogen Life Technologies, Inc.) by BamHI digestion and subcloned into the BamHI site of the pBabe-puro ecotropic retroviral vector (35) , yielding pBabe-puro-CK2. Phoenix packaging cells were used for the generation of retrovirus, which were selected with 500 µg/ml hygromycin to increase Gag and Pol viral protein expression. Briefly, P100 dishes of 80% confluent Phoenix cells were transfected with 15 µg of pBabe-puro-CK2 or pBabe-puro along with 5 µg of an Env-expressing vector. After 24 h, the medium was changed. and cells were incubated for another 24 h at 32°C to increase retrovirus half-life. Supernatants containing retrovirus were then harvested, filtered and transferred on NIH 3T3 target cells in the presence of 2 µg/ml polybrene (Sigma). After 24 h, infected cells were washed, selected with complete medium plus 4 µg/ml puromycin for 4 days, and expanded in medium containing 1 µg/ml puromycin. Single cell clones were isolated by limiting dilution. As a positive control, cells were infected with the pBabe-puro-GFP retrovirus, indicating more than 90% efficiency in retroviral infection of NIH 3T3 cells (data not shown).

EMSA.
The sequence of the WT URE NF-B-containing oligonucleotide from the c-myc gene is as follows: WT, 5'-AAGTCCGGGTTTTCCCCAACC-3' (36) . The core element is underlined. The mutant URE has a two G-to-C-bp conversion, indicated in bold, which blocks NF-B/Rel binding: 5'-AAGTCCGCCTTTTCCCCAACC-3'. The sequence of the Sp1 oligonucleotide is 5'-ATTCGATCGGGGCGGGGCGACC-3'. The Oct1 oligonucleotide has the following sequence: 5'-TGTCGAATGCAAATCACTAGAA-3'. Nuclear extracts from cell lines were prepared and samples (2.5–5 µg) subjected to EMSA as described (24) . For antibody supershift analysis, the binding reaction was performed in the absence of the probe, the appropriate antibody was added, and the mixture incubated for 16 h at 4°C. The probe was then added and the reaction incubated an additional 30 min at 25°C and the complexes resolved by gel electrophoresis, as above. Antibodies used included: anti-RelA (C-20): sc-372, anti-p50 (NLS): sc-114, anti-p52 (K-27): sc-298; and anti-c-Rel (C): sc-71 (all from Santa Cruz Biotechnology, Santa Cruz, CA). When indicated, 250 ng of either IB--GST fusion protein or GST alone or excess unlabelled WT or mutant oligonucleotide was added to the binding reaction just before addition of the probe. Data were quantified by densitometry using a Molecular Dynamics densitometer.

Immunoblotting.
To prepare WCEs, cells were washed with PBS, resuspended in cold PD buffer [40 mM Tris (pH 8.0), 500 mM NaCl, 1 mM EDTA, 1 mM EGTA, 10 mM ß-glycerophosphate, 10 mM NaF, 10 mM p-nitrophenylphosphate, 300 µM Na₃VO₄, 1 mM benzamidine, 2 µM phenylmethylsulfonyl fluoride, 10 µg/ml aprotinin, 1 µg/ml leupeptin, 1 µg/ml pepstatin, 1 mM DTT, and 0.1% NP40]. Cells were lysed by sonication, and debris was removed by centrifugation. Samples of WCEs or nuclear extracts, prepared as described above, were separated by electrophoresis in polyacrylamide-SDS gels, transferred to polyvinylidine difluoride membrane (Millipore, Bedford, MA) and subjected to immunoblotting. Monoclonal antibodies specific for HA-tag (F7) and ß-actin (AC-15) were purchased from Sigma. The rabbit polyclonal antibody specific for the CK2 subunit of human CK2 (residues 70–89) was from Stressgen (Victoria, British Columbia, Canada). Antibodies specific for mouse IB- (C-21), sc-371, and human IB- (C-15), sc-203, were from Santa Cruz Biotechnology. Antibodies specific for NF-B subunits were the same as those used for the supershift assays.

CK2 Kinase Assay.
Samples (2 µg) of WCEs were incubated with a 1 mM solution of the CK2-specific peptide substrate RRREEETEEE (Sigma-Genosys Inc.) in a CK2 kinase buffer {100 mM Tris (pH 8.0), 20 mM MgCl₂, 100 mM KCl, 0.1 µg/ml BSA, 100 µM Na₃VO₄, 5 µCi [-³²P]GTP} at 30°C for 10 min. The reaction was stopped by adding 25 µl of 100 mM ATP in 0.4 N HCl. Samples were spotted onto a P81 Whatmann filter and washed four times in 150 mM H₃PO₄ to remove unincorporated [-³²P]GTP; phosphorylated peptides were measured by scintillation counting. The samples were assayed in duplicate, and the background kinase activity in the absence of the peptide substrate was subtracted. For evaluation of phosphorylation of IB-, 10 to 20 µg of WCEs were diluted to 10-µl final volume in PD buffer. After the addition of 15 µl of buffer D {100 mM Tris (pH 8.0), 100 mM NaCl, 50 mM KCl, 20 mM MgCl₂, 100 µM Na₃VO₄, and 10 µCi [-³²P]GTP}, reactions were incubated at 30°C for 30 min in the presence of 200 ng GST-wtIB- as substrate. Alternatively, GST-3CIB-, with three point mutations at S283A, T291A, and T299A, kindly provided by J. Hiscott, was used as substrate (5) . The kinase reaction was stopped by the addition of 2x SDS-PAGE sample buffer, and the mixture subjected to SDS-PAGE analysis and visualized by autoradiography. To evaluate the kinase activity of transfected WT or mutant proteins HA-CK2 or HA-CK2' in cell extracts, samples (100 µg) of WCEs were precleared with protein G-Sepharose beads (Amersham Pharmacia Biotech AB, Piscataway, NJ) for 1 h at 4°C. The HA-tagged CK2 proteins were collected using 1 µg of the HA-tag F7 antibody. Controls included immunoprecipitations performed with normal polyclonal mouse IgG (Santa Cruz Biotechnology). After 16 h of incubation and extensive washing, immunoprecipitates were subjected to a CK2 kinase assay, as described above.

TNF--induced Apoptosis Assay.
Cells were plated at 2 x 10⁴/ml in 96-well plates. After 24 h, cells were treated with recombinant human TNF- (PeproTech Inc., Rocky Hill, NJ) in the presence of 30 µg/ml cycloheximide (Sigma). After 16 h, cell viability was evaluated by the nonradioactive MTS cell proliferation assay (Promega, Madison, WI).

Focus Formation Assay.
Cells were plated at 3 x 10³/ml in top plugs consisting of complete medium and 0.4% SeaPlaque agarose (FMC Bioproducts, Rockland, ME). Plates were subsequently incubated for 18 days in humidified incubator at 37°C. Cells were stained with 0.5 ml of 0.0005% crystal violet, and colonies were counted using a dissecting microscope.

	RESULTS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Expression of Protein Kinase-inactive CK2 or CK2' Inhibits CK2 Activity in NF639 MMTV-Her-2/neu Mammary Tumor Cells.
Using the NF639 cell line, which was established from a mammary tumor that arose in a MMTV-Her-2/neu transgenic mouse (30) , we demonstrated that Her-2/neu activates classical NF-B via a noncanonical pathway, i.e., involving calpain- rather than IKK-induced proteasome-mediated degradation of IB- (28) . Given the recent evidence implicating phosphorylation by CK2 in degradation of IB- via calpain (9 , 10) , we compared expression levels of CK2 catalytic subunit in NF639 cells with levels in untransformed NMuMG mouse mammary epithelial cells (Fig. 1A) . As control for equal loading, samples were similarly analyzed for levels of ß-actin (Fig. 1A) . Immunoblot analysis revealed an increase in the level of CK2 protein in NF639 cells compared with NMuMG cells. Scanning of this and a duplicate blot analysis indicated that NF639 cells display a 2.8 ± 0.3-fold increase in total level of CK2 compared with NMuMG cells.

View larger version (28K): [in this window] [in a new window]   Fig. 1. Expression of kinase-inactive CK2 or CK2' subunit inhibits CK2 activity in Her-2/neu-induced NF639 breast cancer cells. A, CK2 expression. WCEs were prepared from two independent cultures of NF639 and NMuMG untransformed mouse breast epithelial cells at 70% confluence. Samples (80 µg) were subjected to immunoblot analysis for CK2 and ß-actin levels. B, expression of transfected CK2 constructs in NF639 cells. Cells were transfected with parental pRc/CMV2, HA-CK2 K68M, HA-CK2' K69M, HA-CK2 WT, or HA-CK2' WT plasmid expression vectors. Stably transfected cells were selected with puromycin. WCEs were prepared, and equal amounts (300 µg) were immunoprecipitated with either the F7 antibody against HA-tag (HA) or normal mouse IgG (Ig). Immunoprecipitates were subjected to a CK2 kinase assay in the presence of [-32P]GTP using either GST-wtIB- or, as a negative control, GST-3CIB- (3C, with three point mutations at S283A, T291A, and T299A in the COOH-terminal PEST domain). Proteins were resolved by PAGE-SDS and visualized by autoradiography. Positions of molecular weight protein standards of Mr 64,900 (64.9) and 52,800 (52.8) and of the GST-wtIB- (wt) and GST-3CIB- (3C) proteins are as indicated. C, total CK2 activity. NF639 cells were transfected with the indicated constructs and selected with geneticin. Samples of WCEs (2 µg) were assayed in duplicate for CK2 phosphorylation using a specific CK2 substrate peptide. Data are expressed as the means with background (cpm in the absence of peptide) subtracted.

Inhibition of CK2 Reduces Basal NF-B Binding Activity and Stabilizes IB- Degradation in Her-2/neu-induced NF639 Breast Cancer Cells.
We next measured the effects of CK2 inhibition on the Her-2/neu-induced NF-B activity in NF639 cells, which we described previously (28) . To assess NF-B binding levels, nuclear extracts were prepared from NF639 cells expressing HA-CK2 K68M, HA-CK2' K69M, or both kinase-inactive variants combined, and were subjected to EMSA. As probe, an oligonucleotide containing the NF-B element upstream of the c-myc promoter was used and Oct1 served as a control for equal loading (Fig. 2A) . Two major NF-B complexes were observed with extracts from the NF639 cell line, which have been identified previously as p50/RelA and p50 homodimer complexes (28) . Expression of kinase-inactive CK2 or kinase-inactive CK2' caused a significant reduction in both p50/RelA and p50/p50 binding levels compared with transfection with the pRc/CMV2 plasmid. Coexpression of CK2 K68M and CK2' K69M mutants did not have an additive effect.

View larger version (37K): [in this window] [in a new window]   Fig. 2. Expression of kinase-inactive CK2 or CK2' subunits inhibits constitutive NF-B binding activity in NF639 breast cancer cells. Cells were transfected and mixed populations of stably transfected cells selected with puromycin. A, NF-B binding activity. Nuclear extracts were prepared and samples (5 µg) subjected to EMSA; probes were oligonucleotides containing either NF-B- or Oct1-binding elements. The position of the p50/p50 and p50/RelA complexes, identified previously (28) , are as indicated. Similar results were observed with two additional independent sets of NF639 cells stably transfected with CK2 mutant constructs (data not shown). B, NF-B subunits: total and nuclear levels. WCEs and nuclear extracts were prepared, and samples (50 µg of WCEs and 5 µg of nuclear extracts) were subjected to SDS-PAGE and immunoblot analysis of p50, RelA, p52, and ß-actin, as loading control. *, positions of nonspecific bands. C, IB- half-life. Exponentially growing cells were incubated with 50 µg/ml cycloheximide (CHX), and WCEs were prepared at 0, 2, 4, and 6 h. Samples (50 µg) were subjected to immunoblot analysis for IB-, and ß-actin, as loading control. *, the position of a nonspecific band.

CK2-mediated phosphorylation of IB- in the COOH-terminal PEST domain has been implicated in the basal and signal-dependent turnover of free and NF-B-bound IB- (4, 5, 6, 7, 8) . We next tested the involvement of CK2-mediated phosphorylation in the rate of IB- turnover in NF639 cells. Exponentially growing NF639 cells, stably transfected with HA-CK2 K68M or parental vectors, were treated with the protein synthesis inhibitor cycloheximide and WCEs prepared at the times indicated and subjected to immunoblotting for IB- expression (Fig. 2C) . The stability of IB- protein was substantially increased in NF639 cells upon inhibition of CK2. In two independent sets of stably transfected NF639 cells, the half-life of IB- decay was between 2 and 4 h in pRc/CMV2-transfected control cells and increased to more than 6 h in cells expressing CK2 K68M (Fig. 2C , and data not shown). Overall, these results demonstrate the ability of protein kinase CK2 to affect NF-B levels in Her-2/neu-induced NF639 breast cancer cells via the regulation of IB- phosphorylation and degradation.

Inhibition of Protein Kinase CK2 Sensitizes Her-2/neu-induced NF639 Mouse Breast Cancer Cells to Apoptosis and Loss of Anchorage-independent Growth.
Work from many laboratories, including our own, have highlighted the importance of constitutive NF-B activity in protecting breast cancer cells from apoptosis (24) . We, therefore, assessed the effect of the inhibition of CK2 on TNF--induced cell death. NF639 cells, stably transfected with HA-CK2 K68M expression vector, were treated with cycloheximide and stimulated with an increasing dose of TNF- (25–400 ng/ml). Cell viability was assessed by MTS assay 24 h after stimulation. As a control, NF639 cells, stably expressing the IB- 2N3C (S32A, S36A, S283A, T291A, and T299A) super-repressor mutant, which cannot be phosphorylated by IKK or CK2 kinases and is, therefore, resistant to degradation, were similarly assessed. NF639 cells that expressed IB- 2N3C displayed a profound decrease in p50/RelA binding and NF-B transcriptional activity, as expected (data not shown). Minimal cell death was observed in parental pRc/CMV2-transfected NF639 cells, which is consistent with the elevated NF-B constitutive levels observed in this cell line (Ref. 28 ; Fig. 3A ). In contrast, cells expressing CK2 K68M or IB- 2N3C displayed increased susceptibility to TNF--induced cell death, as judged by MTS assay. Results obtained with this and a second independent set of stably transfected cells were quantified. At a concentration of 400 ng/ml TNF-, the percentage of viable pRc/CMV2-, CK2 K68M-, or IB- 2N3C-expressing NF639 cells was 85.0 ± 5.6%, 52.5 ± 4.9% (P < 0.001), and 58.5 ± 12.1% (P < 0.03), respectively. Therefore, the inhibition of CK2 activity sensitizes NF639 cells to TNF--induced cell death.

View larger version (56K): [in this window] [in a new window]   Fig. 3. Expression of kinase-inactive CK2 subunit increases susceptibility to TNF--induced apoptosis and reduces anchorage-independent growth of NF639 breast cancer cells. A, TNF--induced cell death. Mixed population of NF639 cells stably transfected with HA-CK2 K68M, parental pRc/CMV2, or IB- 2N3C super-repressor plasmid expression vectors were plated at 2 x 103 in 96-well plates. After 24 h, cells were treated with recombinant human TNF- in the presence of 30 µg/ml cycloheximide. After 16 h, cell viability was evaluated by the MTS cell proliferation assay. Results are expressed as percentage of viable cells (absorbance at 490 nm normalized to absorbance of cultures treated with cycloheximide alone). B, soft agar assay. Cells were plated, in triplicate, at 3 x 103/ml in top plugs consisting of complete medium and 0.4% agarose. After 18 days, the numbers of foci were scored and pictures taken with a Nikon camera (x1.5). Results were obtained with two independent sets of transfected cells. Below the images, values of colony number per high power field.

Expression of Kinase-inactive CK2 Inhibits NF-B Activity in Hs578T Human Breast Cancer Cells and 293T Human Embryonic Kidney Cells.
We next asked whether the inhibition of CK2 activity in other cell lines would similarly decrease NF-B activity. The Hs578T human breast tumor cell line, which overexpresses CK2 protein and kinase activity (20) , was selected for study. Hs578T cells were identified previously to express predominantly activated p50/RelA and p50/p50 NF-B complexes (20 , 24) . For these experiments, mixed cell populations of Hs578T cells stably expressing HA-tagged mutant CK2 subunit were obtained by geneticin selection. Nuclear extracts and WCE were prepared and analyzed for NF-B binding levels and expression of HA-tagged CK2, respectively (Fig. 4A) . Results obtained with this and a second set of stably transfected cells showed an 50% drop in NF-B binding in HA-CK2 K68M-expressing Hs578T cells compared with parental pRc/CMV2-transfected cells. Expression of CK2 K68M mutant was confirmed by immunoblotting for the HA tag. Thus, the inhibition of CK2 similarly reduces NF-B binding in Hs578T human breast cancer cells.

View larger version (27K): [in this window] [in a new window]   Fig. 4. Expression of kinase-inactive CK2 or CK2' subunit inhibits NF-B binding activity in Hs578T human breast cancer cells and 293T human embryonic kidney cells. A, Hs578T cells. Mixed population of cells, stably transfected with pRc/CMV2 or HA-CK2 K68M plasmid vectors, were selected in geneticin, and nuclear extracts and WCEs were prepared. Top panel, samples of nuclear extracts (5 µg) were subjected to EMSA for NF-B or Oct1. The position of the p50/p50 and p50/RelA complexes identified previously (20) are as indicated. Bottom panel, samples of WCE (80 µg) were subjected to immunoblot analysis of exogenous CK2, using an HA tag-specific antibody. *, the position of a nonspecific band. The position of molecular weight protein standards of Mr 61,000 (61) and Mr 49,000 (49) are as indicated. B, 293T cells. Cells were transiently transfected with parental pRc/CMV2, HA-CK2 K68M, or HA-CK2' K69M plasmid vectors. Cells were harvested after 6 days and analyzed as described in A. Band 1 and band 2, the two major NF-B complexes.

CK2 Overexpression in NIH 3T3 Cells by Retroviral Gene Delivery Leads to Increased Nuclear NF-B Expression.
To determine whether increased CK2 expression is sufficient to induce NF-B levels, we next attempted to increase CK2 activity through overexpression of the CK2 catalytic subunit of CK2. We turned to the NIH 3T3 fibroblast cell line, which has lower levels of endogenous CK2 than the NF639 or Hs578T cell lines (data not shown). Cells were retrovirally infected with a vector expressing an untagged murine CK2, and a mixed population of infected cells as well as single clones were selected in puromycin and were screened for total CK2 protein levels. Two stable CK2-overexpressing NIH 3T3 clones, designated Clone 4 and Clone 6, were chosen for this study. To monitor the relative levels of CK2 expression in the selected cells, immunoblot analysis was performed with a CK2 polyclonal antibody using samples of WCEs and nuclear protein extracts (Fig. 5A) . Two bands were detected in the WCEs. The top band corresponded to the full length CK2 (M_r 45,000), and the bottom one to a protein of M_r 40,000. Immunoblot analysis performed with antibodies specific for either the NH₂-terminal or COOH-terminal part of CK2 indicated that the M_r 40,000 protein likely resulted from COOH-terminal clipping of CK2 (data not shown), as described previously during the in vitro purification of human recombinant CK2 (12) . The WCEs of the Babe-CK2 mixed population, Clone 4, and Clone 6 demonstrated elevated levels of CK2 protein compared with the parental vector control (Babe)-infected NIH 3T3 cells, which showed only low basal expression. When the blots were scanned and normalized to ß-actin levels, a 2.6-, 15.1-, and 9.5-fold increase in CK2 expression was observed in NIH 3T3 Babe-CK2 mixed-population cells, Clone 4, and Clone 6, respectively, compared with the parental Babe cells. No significant increase in CK2' was observed in Babe-CK2 cells (data not shown). Higher levels of CK2 expression were also detected in the nuclei of Babe-CK2 cells (Fig. 5A) . These results were confirmed by immunofluorescent staining of CK2, which showed a strong accumulation of CK2 in both the cytoplasm and the nuclei of Clone 4 and Clone 6 compared with parental NIH 3T3 Babe cells (data not shown). To confirm that increased CK2 expression led to increased CK2 enzymatic activity, a CK2 kinase assay was performed with GST-wtIB- or GST-3CIB- as substrates. WCEs, prepared from NIH 3T3 Babe-CK2 mixed population and clones, were used directly in in vitro CK2 phosphorylation assays (Fig. 5B) . Kinase assays demonstrated strong preferential phosphorylation of GST-wtIB- compared with GST-3CIB-, consistent with the assay measuring CK2 activity. The mixed population, Clone 4, and Clone 6 displayed an increase in CK2 kinase activity of 1.2-, 2.3-, and 1.6-fold, respectively, compared with the parental NIH 3T3 Babe cells. Although the magnitude of increase in activity is less than the increase in protein, these results confirm that the two clones and the mixed population of cells display elevated CK2 activity.

View larger version (50K): [in this window] [in a new window]   Fig. 5. Overexpression of the CK2 subunit in NIH 3T3 cells. NIH 3T3 cells were infected with recombinant retrovirus for pBabe (Babe) or pBabe-CK2 vectors. A mixed population of Babe-CK2 (MP) cells and two clones (Clone 4 and Clone 6) were selected in puromycin. A, CK2 expression. Samples of WCEs (80 µg) or nuclear extracts (15 µg) from the indicated cells were separated by SDS-PAGE and subjected to immunoblot analysis for CK2 and ß-actin levels. Two bands are seen in the WCEs with the CK2 antibody. Top band, the full-length CK2 protein; bottom band, (likely) a COOH-terminal clipped form of CK2. B, CK2 activity. WCEs were prepared from the indicated cell populations, and samples (10 µg) were subjected to a CK2 kinase assay using as substrate either GST-wtIB- (wt) or GST-3CIB- (3C), as a negative control.

View larger version (58K): [in this window] [in a new window]   Fig. 6. Overexpression of CK2 subunit results in increased NF-B binding in NIH 3T3 cells. A, NF-B binding activity. Nuclear extracts were prepared from the indicated cells, and samples (5 µg) were subjected to EMSA with oligonucleotides specific for NF-B and Sp1. Band 1 and band 2, the positions of the two major NF-B complexes. B, supershift analysis. Nuclear extracts from NIH 3T3 Babe-CK2 Clone 4 cells were incubated in the absence (none), or presence of antibody () against p50, p52, RelA, or c-Rel NF-B subunits, or 1 µg GST-wtIB- or GST protein, and were subjected to EMSA with a NF-B oligonucleotide. To test for binding specificity, the binding reaction was performed in the presence of 20x excess unlabelled WT (20X NF-B) or mutant (20X mNF-B) NF-B oligonucleotide.

CK2 Overexpression in NIH 3T3 Cells Leads to Increased Turnover of IB- and NF-B Transcriptional Activity.

View larger version (27K): [in this window] [in a new window]   Fig. 7. Overexpression of CK2 subunit results in a higher rate of IB- degradation and NF-B transcriptional activity in NIH 3T3 cells. A, IB- half-life. Left panel, the indicated cells, in exponential growth, were treated with 50 µg/ml cycloheximide (CHX), and WCEs were prepared at 0, 4, 8, or 12 h. Samples (50 µg) were subjected to immunoblot analysis for IB-. Right panel, the immunoblots in the left panel were subjected to densitometry, and the results were presented as the percentage relative to the control. B, NF-B transcriptional activity. Lower panel, cultures of parental Babe or Clone 4 cells were transiently transfected, in triplicate, with 1 µg of NF-B element-driven luciferase reporter construct and 0.5 µg of pSV40-ß-gal in the presence of 0.5 or 1 µg of pSV/Neo empty parental vector or of the human IB- construct vectors: WT (WT), 2N mutant (S32A, S36A), or 3C mutant (S283A, T291A, and T299A in the COOH-terminal PEST domain). After 48 h, cultures were harvested, normalized for ß-gal activity, and assayed for luciferase activity. Values in the bar graph, luciferase normalized for ß-gal activity. Top panel, Babe-CK2 Clone 4 cells were transfected with equal amounts of pSV/Neo empty parental vector or human IB- construct vectors: WT, 2N mutant, or 3C mutant. After 18 h, WCEs were prepared, and samples (50 µg) were subjected to immunoblot analysis for human IB- (hIB-) using a human-specific antibody. *, the position of a nonspecific band. (All lanes are from the same gel.)

	DISCUSSION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Here we show that the inhibition of elevated CK2 activity in cancer cells reduces constitutive NF-B activity, whereas ectopic expression of CK2 is sufficient to induce NF-B activity in NIH 3T3 fibroblasts. Thus, CK2 plays a pivotal role in the regulation of constitutive NF-B activity. Importantly, we show for the first time that the inhibition of CK2 activity decreases Her-2/neu-induced NF-B activity. Thus, kinase-inactive CK2 subunit decreased nuclear NF-B and transformed phenotype, whereas it increased sensitivity to TNF- induced death in MMTV-Her-2/neu-derived NF639 mouse mammary carcinoma cells. Similarly, the inhibition of CK2 in Hs578T cells, which also display elevated basal CK2 activity (20) , decreased NF-B binding. Conversely, CK2 overexpression in NIH 3T3 cells was sufficient to increase IB- turnover and basal NF-B activity. Previously, we and others demonstrated that primary breast cancer samples from patients or from a carcinogen-induced rodent model as well as breast cancer cell lines display increased CK2 activity (18, 19, 20, 21) and aberrant activation of NF-B (23 , 24 , 26 , 27) , whereas only low levels of CK2 and NF-B activation were detected in normal breast epithelial cells. Our findings here demonstrate a direct link between overexpression of CK2 and NF-B in these cancers. Furthermore, they suggest that CK2 kinase is a downstream mediator of Her-2/neu signaling and, thus, represents a potential new therapeutic target for the treatment of these malignancies.

In our studies, we made use of CK2 K68M or CK2' K69M mutants, which display a single point mutation in the kinase domain of these catalytic subunits and which are devoid of kinase activity (34) . In the breast cancer cells, these inactive subunits were capable of reducing CK2 activity, i.e., acting as dominant negatives, although they do not act this way in all cells (Ref. 16 ; as discussed below). In particular, we observed a 30–40% inhibition of total CK2 activity in NF639 cells expressing CK2 K68M or CK2' K69M as compared with parental cells. The inhibition of CK2 resulted in a drop in NF-B binding in both NF639 and Hs578T breast tumor cells as well as in 293T human embryonic kidney cells. Thus, CK2 inhibition had a direct affect on NF-B activity in various cell types. Consistent with these observations, we observed previously that treatment with the selective pharmacological inhibitors of CK2, apigenin or emodin, inhibited NF-B activity in human breast cancer cell lines (20) , and in mouse B-cell lymphomas (10) . Interestingly, whereas ectopic expression of WT mouse CK2 catalytic subunit in NIH 3T3 fibroblast cells, which display low basal CK2 activity, led to a substantial increase in CK2 activity and NF-B binding and activity, we were unable to similarly increase levels of CK2 protein or activity in NF639 cells (data not shown). Interestingly, the NF-B activity in CK2-overexpressing NIH 3T3 cells consisted predominantly of classical p50 and RelA-containing complexes, similar to breast tumor tissue from transgenic mice overexpressing CK2 in the mammary gland (19) .

We observed previously that Her-2/neu activates NF-B via a PI3-K-to-Akt-kinase signaling pathway that can be inhibited via antibody against the receptor or by the tumor suppressor PTEN in NF639 breast cancer cells (Ref. 28 ; see Fig. 8 ). Different mechanisms may be involved in the regulation of NF-B activity by CK2. Previous work has indicated that basal and signal-dependent turnover of free and NF-B-bound IB- is controlled by phosphorylation of residues in the COOH-terminal PEST domain by CK2, and, thus, mutation of these sites results in longer half-life of the IB protein (5 , 7 , 8) . Our evidence indicates that the level of CK2 activity affects IB- stability in NF639 cells (Fig. 8) . Previously, we demonstrated that the dominant proteolytic pathway for IB- degradation in NF639 cells is mediated via calpain (28) ; and, because in B cells, we observed that CK2 phosphorylation accelerates degradation of IB by calpain (10) , it is possible that a similar mechanism occurs in breast cancer cells. The mechanisms leading to enhanced CK2 activity and the potential role of any additional kinases in the signaling pathway remain to be determined (Fig. 8) . In addition, CK2 has been proposed to control NF-B transcriptional activity by direct phosphorylation of the RelA subunit in response to TNF- stimulation (38 , 39) . Here, we observed that NF-B in CK2-overexpressing NIH 3T3 cells is clearly nuclear and transcriptionally active in a reporter assay; however, it remains to be determined whether CK2 has a similar affect on basal activity of RelA in breast cancer cells. In addition, CK2 phosphorylates PTEN directly (40, 41, 42) . Phospho-PTEN has increased stability but reduced lipid phosphatase activity (40) , thereby promoting Akt phosphorylation and activation (42) . However, no significant change in the activity of Her-2/neu receptor or in the levels of phosphorylated Akt was evident in NF639 breast cancer cells transduced with the kinase-inactive CK2 subunits (data not shown), which suggests that either PTEN expression itself is down-regulated or PTEN phosphorylation cannot be reduced by the kinase-inactive CK2 constructs.

View larger version (22K): [in this window] [in a new window]   Fig. 8. Scheme of CK2 involvement in Her-2/neu-mediated NF-B activation in breast cancer cells. Overexpression of Her-2/neu and heterodimerization with other EGF receptor family members (EGFR, c-erbB-3, c-erbB-4) activate NF-B via a PI3-K-to-Akt-kinase signaling pathway that can be inhibited via treatment with an anti-Her-2/neu antibody or the tumor suppressor PTEN (28) . Increased cytoplasmic and nuclear CK2 expression and activity have been observed in NF639 cells and many human breast cancers, although the mechanisms controlling this increase remain to be elucidated. Overexpression of CK2 leads to increased IB- phosphorylation and rate of degradation, which we have shown to be mediated via a calpain proteolytic pathway (28) . The possibility that elevated levels of CK2 promotes NF-B transcriptional activity by direct phosphorylation of its RelA subunit, as seen previously on TNF- treatment (38 , 39) , is under investigation in breast cancer cells.

Aberrant nuclear NF-B activity has been reported in many cancers (44) . Products of several oncogenes such as Her-2/neu (28 , 45) , the EGF receptor signaling pathway (27 , 46) , and the oncogenic Raf and Ras proteins (47 , 48) induce NF-B activity in various cell types. CK2 is markedly elevated in a number of hematopoietic and solid tumors (49) and, given our findings that demonstrate the ability of CK2 to directly affect NF-B activation, it would be interesting to determine whether NF-B and CK2 can cooperate to induce transformation. Experiments are in progress with bitransgenic mice that overexpress the c-Rel NF-B family member and CK2 subunit in mammary epithelial cells. Lastly, our findings suggest that combinations of proteasome and calpain inhibitors or IKK and CK2 kinase inhibitors could be more effective than the use of single inhibitors in blocking IB degradation and NF-B activation, in promoting tumor cell apoptosis, and in sensitizing cancer cells to the proapoptotic effects of radiation or chemotherapy.

	ACKNOWLEDGMENTS

 
This article is dedicated to the memory of Darin Sloneker, former Administrative Assistant of the Program in Research on Women’s Health, Boston University School of Medicine, Boston, MA. We thank John Hiscott, David W. Litchfield, Alan Chan, and Philip Leder for generously providing cloned DNAs and cells, Judith Foster for use of the densitometer, and Muriel Jeay for assistance in preparation of the manuscript.

	FOOTNOTES

 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

¹ Supported by grants from the Massachusetts Department of Public Health Breast Cancer Program (to E. L-B., R. R-M.), by Department of Army DAMD 17-01 10158 (to R. R-M.), and by NIH Grants RO1 CA82742 and PO1 ES11624 (to G. E. S., D. C. S.).

² To whom requests for reprints should be addressed, at Department of Biochemistry, Boston University Medical School, 715 Albany Street, Boston, MA 02118-2394. E-mail: gsonensh{at}bu.edu

³ The abbreviations used are: NF-B, nuclear factor-B; ß-gal, ß-galactosidase; EGF, epidermal growth factor; EMSA, electrophoretic mobility shift analysis; GST, glutathione-S-transferase; PI3-K, phosphatidylinositol 3-kinase; WCE, whole cell extract; WT, wild type; MMTV, mouse mammary tumor virus; IKK, IB kinase; TNF, tumor necrosis factor; CMV, cytomegalovirus; Oct1, Octomer-1 (oligonucleotide); PEST, proline-glutamic-serine-threonine rich region.

Received 7/10/02. Accepted 9/20/02.

	REFERENCES

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION