Casein kinase II-mediated phosphorylation of NF-kappaB p65 subunit enhances inducible nitric-oxide synthase gene transcription in vivo.

Nitric oxide (NO) produced by inducible nitric-oxide synthase (NOSII) is mainly regulated at the transcriptional level by the nuclear factor-kappaB (NF-kappaB). In the present study, we further analyzed the role of NF-kappaB in the in vivo transcriptional regulation of NOSII gene by comparing two clones isolated from the EMT-6 mouse mammary cancer cell line. In response to interleukin (IL)-1beta or lipopolysaccharide (LPS), EMT-6 clone J (EMT-6J) cells produce 3-fold more NO than EMT-6 clone H (EMT-6H) cells, an effect correlated with enhanced activation of NF-kappaB in EMT-6J cells. In response to IL-1beta, the kinetics of degradation of NF-kappaB inhibitors IkappaB-alpha and IkappaB-beta, the nucleo-cytoplasmic shuttling of the transcription factor and its binding to a specific DNA sequence were similar in both clones. In contrast, an IL-1beta-induced phosphorylation of serine residues in NF-kappaB p65 subunit was observed in EMT-6J, but not in EMT-6H, cells. This IL-1beta-induced phosphorylation of p65 was specifically prevented by pretreatment of EMT-6J cells with the casein kinase II inhibitor DRB. Small interfering RNA-mediated depletion of casein kinase II-alpha subunit also decreased NF-kappaB transcriptional activity and NOSII gene transcription in IL-1beta and LPS-stimulated EMT-6J cells to the levels observed in EMT-6H cells treated in the same conditions. Altogether, these data indicate that casein kinase II-mediated phosphorylation of p65 subunit can enhance the transcriptional activity of NF-kappaB in vivo. This post-translational modification of the transcription factor can be responsible for increased NOSII gene transcription and NO production in tumor cells exposed to either IL-1beta or LPS.

The free radical nitric oxide (NO), whose synthesis results from the oxidization of arginine to citrulline, plays a role in a variety of physiological and pathological processes such as immune response and autoimmune disease (1), neurotransmis-sion and neurotoxicity (2), vascular tone, and vascular disease (3). Of the three distinct genes that encode NO synthases, one encodes an inducible isoform known as NO synthase II (NOSII) 1 that was cloned in activated macrophages (4). Actually, this gene can be induced by a wide range of cytokines and bacterial products in a variety of cell types (reviewed in Refs. 5 and 6). NOSII-mediated production of NO is regulated at several levels that include gene transcription (7), alternative splicing (8), mRNA stability (9), enzyme dimerization (10), and substrate and co-factor availability (11). The gene is usually not expressed in non-stimulated cells. In mouse macrophages, the transcriptional activity of the NOSII gene promoter depends on a variety of transcription factors, including NF-B (12), IRF (13), HIF (14), NF-IL6 (15), OCT (16), Stat (17), and HSF (18). In other cell types, only NF-B is known to be a critical component of NOSII gene transcriptional activation in response to inflammatory stimuli (19 -22).
NF-B is a family of transcription factors composed of five gene products (RelA/p65, cRel, RelB, p50, p52) that combine to form active dimers (reviewed in Ref. 23). The predominant form is a heterodimer containing the p50 and p65 subunits. Each of these subunits is characterized by a Rel homology domain involved in dimerization, DNA binding, interaction with the inhibitory IB proteins, and nuclear localization. Unlike the p50 subunit, p65 also contains a transcriptional activator domain indispensable for NF-B transcriptional activity. In most resting cells, the p50/p65 heterodimer is sequestered in the cytoplasm by association with the inhibitor B (IB). NF-B is activated when signals from various stimuli are transduced to the IB kinase (IKK) complex, which phosphorylates IB-␣ and IB-␤ isoforms (24,25). Phosphorylated IBs are rapidly degraded by an ubiquitin/proteasome pathway, and the free NF-B complex is translocated into the nucleus to activate expression of its target genes. Then, IB-␣ is re-synthesized in an NF-B-dependent manner, enters in the nucleus, and removes the p50/p65 heterodimer from the DNA (26,27).
Although degradation of IBs is sufficient to cause nuclear translocation of NF-B, subsequent events such as p65 phosphorylation can potentiate the transcriptional activity of NF-B (reviewed in Refs. 28 -30). In B and T cells, a protein kinase A catalytic subunit (PKAc) mediates lipopolysaccharide (LPS)-induced phosphorylation of p65 on Ser-276 (31), whereas this Ser-276 phosphorylation of p65 involves mitogen-and stress-activated protein kinase 1 (MSK1) after TNF-␣ stimulation in other cell types (32). TNF-␣ can also trigger the phosphorylation of p65 Ser-311 through protein kinase C (PKC) isoform (33). These phosphorylation events in the p65 Rel homology domain enable the recruitment of co-factors bearing histone acetylase activity such as cAMP-response elementbinding protein-binding protein (p300/CBP) to facilitate the formation and subsequent activation of the preinitiation complex (32)(33)(34). The TA domain of p65 can also be phosphorylated on Ser-536 by IKKs (35) and Ser-529 by casein kinase II (CKII) (36). Although the phosphorylation of p65 TA domain was shown also to enhance the transcriptional activity of NF-B, genes whose expression could be modulated by these events in vivo have not been identified. The present study demonstrates that casein kinase II-mediated phosphorylation of p65 enhances the transcription of NOSII gene in a mammary cancer cell line in response to either interleukin-1␤ (IL-1␤) or LPS through modulation of NF-B transcriptional activation.
Determination of NO 2 Concentration-Cells were seeded in 96-well plates, and 24 h after stimulation with IL-1␤, nitrite concentrations were measured in the medium using the Green assay (38). The nitrite levels determined are representative of the amount of NO produced by cells (39). The results represent an average of three experiments in FIG. 1. IL-1␤ enhances NO production and NOSII expression in EMT-6 cells. A, nitrite accumulation, as measured by using the Griess reagent, in culture medium of indicated cell clones stimulated (black bars) or not (white bars) with 0.55 ng/ml IL-1␤ for 24 h. B and C, Western blot (B) and Northern blot (C) analysis of NOSII expression in the indicated cell clones stimulated or not with 0.55 ng/ml IL-1␤ for 6 h. Protein and RNA loading controls are heat shock protein 70 (HSC70) and ethidium bromide staining of ribosomal RNA, respectively. D, NOSII promoter activity as determined by using pNOSII-luc reporter gene transiently expressed in cells subsequently exposed to 0.55 ng/ml IL-1␤ for 15 h. Results are expressed as "-fold induction" relative to control, non-stimulated cells. *, n ϭ 5, p Յ 0.05 in comparison with EMT-6H cells.
FIG. 2. Differential activity of NF-B in IL-1␤ stimulated EMT-6 clones. A, indicated EMT-6 cell clones were transiently transfected with a NOSII promoter-luciferase reporter gene (pNOSII-luc) with either a wild type (WT) or a mutated (bDmut) downstream b site before exposure to 0.55 ng/ml IL-1␤ for 15 h. Luciferase activity: -fold increase relative to non-stimulated cells transfected with the wild type construct. *, n ϭ 3, p Յ 0.05 versus stimulated cells transfected with the wild type construct. B, indicated EMT-6 cell clones were transiently transfected with an NF-B-dependent-luciferase reporter gene p(NF-B) 4 -luc before exposure to 0.55 ng/ml IL-1␤ for 15 h. Luciferase activity, -fold increase relative to non-stimulated cells. *, n ϭ 4, p Յ 0.05 compared with stimulated EMT-6H cells.
which each measurement was performed in triplicate.
Immunoblotting-After appropriate stimulation, cells were washed with ice-cold PBS and proteins were extracted with RIPA buffer (150 mM NaCl, 50 mM Tris-HCl, pH 8, 0.1% SDS, 1% Nonidet P-40, 0.5% sodium deoxycholate) containing a mixture of protease inhibitors (Roche Diagnostics, Mannheim, Germany). Thirty g of total protein were separated by denaturing SDS-PAGE and transferred onto polyvinylidene difluoride membranes (Amersham Biosciences, Orsay, France). Antibody reactivity was monitored with anti-rabbit or antimouse IgG coupled to horseradish peroxidase (Jackson Immuno-Research Laboratoriess, West Grove, PA), using an enhanced chemiluminescence kit (Santa Cruz Biotechnology). HSC70 expression is indicated as internal control in each data. Results are representative of at least three independent experiments.
Transfection and Reporter Assay-Transfection of luciferase reporter gene and siRNA duplexes were performed with lipofectamine (Invitrogen) according to the manufacturer's protocol. Briefly, 5⅐10 4 EMT-6 cells were seeded in 24-well plates 24 h before transfection. Appropriate mixtures of LipofectAMINE with DNA or siRNA were added to each well containing 300 l of Dulbecco's modified Eagle's medium. Three hours latter, 500 l of complete Eagle's minimum essential medium was added. After 48 h of incubation, the medium was removed, and cells were stimulated with fresh complete Eagle's minimum essential medium containing IL-1␤ or LPS. Cells were routinely co-transfected with a TK-Renilla luciferase plasmid (Promega, Charbonniè res, France) to normalize for transfection efficacy. The dual luciferase reporter assay kit from Promega was used following the users manual. Luciferase and Renilla activities were measured with a luminometer Lumat LB9507 (EG&G Berthold). The values shown represent an average of three experiments in which each sample was performed in triplicate.
Northern Blot-Total RNA was prepared from EMT-6 cells and analyzed by Northern blot as described previously (40). The probe for detection of the murine NOSII was obtained by reverse transcriptase-PCR using one tube, two-enzyme Access reverse transcriptase-PCR system (Promega) with the following oligonucleotides, specific for the murine NOSII gene: 5Ј-CCAGTGTCTGGGAGCATCACCCCTG-3Ј (forward) and 5Ј-GAACTGAGGGTACATGCTGGAGCC-3Ј (reverse) amplifying a fragment of 498 bp. Data are representative of at least three independent experiments.
Site-directed Mutagenesis-The downstream b site in the promoter was mutated in the pNOSII-luc construct (Ϫ86 to Ϫ84, GGG to CTA) introducing a XbaI restriction site to verify the modified promoter. The mutation was performed using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) following the manufacturer's recommendations. The resulting construct was verified by enzymatic digestion and confirmed by automated sequencing.
Immunofluorescence Staining-Cells were seeded on glass cover- in EMT-6 cell clones exposed to 0.55 ng/ml IL-1␤ for the indicated times. HSC70, loading control. C, p65 cellular localization in the studied cell clones exposed to 0.55 ng/ml for indicated times. Calibration bar, 50 m.
FIG. 4. NF-B DNA binding activity in IL-1␤-stimulated EMT-6 cells. NF-B DNA binding activity was analyzed in the two EMT-6 clones exposed to 0.55 ng/ml IL-1␤ for indicated times. EMSA was performed by using the downstream b-site of the NOSII promoter (bD) as radiolabeled probe and three g of nuclear protein. A, time course of NF-B DNA binding activity. B, competition assay was performed by adding 50ϫ unlabeled specific (bD) and nonspecific (OCT) probe to nuclear extracts of EMT-6J cells exposed to IL-1␤ for 30 min. C, identification of EMSA complexes in EMT-6J cells exposed to IL-1␤ for 30 min. B and C, similar results were obtained in EMT-6H cells (not shown).
slips. After each stimulation, cells were rinsed twice in pre-warmed (37°C) PBS and were then fixed in cold (Ϫ20°C) methanol and soaked in cold acetone. Cells were re-hydrated with PBS, blocked with PBS containing 5% fetal calf serum and 0.1% bovine serum albumin (Sigma), and then incubated 60 min with a rabbit primary antibody anti-p65 at a 1:300 dilution in PBS with 0.1% bovine serum albumin. After several washes in PBS, the coverslips were incubated for 60 min with the secondary antibody Alexa Fluor 594-conjugated donkey anti-rabbit IgG (Molecular Probes, Eugene, OR) at a 1:1000 dilution in PBS with 0.1% bovine serum albumin. After several washes with PBS, the coverslips were aspirated dry and sealed on glass slides in a mounting medium (0.1 M Tris, pH 8.5, glycerol to 25% of final volume, and 0.1 g/ml of Mowiol (Calbiochem) containing DABCO ([1,4 diazabicyclo(2.2.2)octane]; Sigma) to prevent rapid photobleaching. The slides were examined with a Nikon E400 epifluorescence microscope equipped with a digital camera (DXM 1200, Nikon, Paris, France).
EMSAs-Cells were seeded in flasks at 2.5 10 6 cells/flask 24 h before each experiment. After appropriate stimulation, cells were washed with ice-cold PBS, scraped, and pelleted by centrifugation. Nuclear extraction and EMSAs procedure were carried out as described (18). The oligonucleotide corresponding to the downstream B site in the NOSII promoter was the following: (Ϫ89) ACTGGG-GACTCTCCCTT (Ϫ73). It was hybridized with the corresponding complementary oligonucleotide, labeled, and purified on Quick Spin columns for radiolabeled DNA purification (Roche Diagnostics) to eliminate unincorporated radioactivity. Data are representative of at least three independent experiments.
Immunoprecipitation-Immunoprecipitation of phosphoserine proteins and identification of p65 in the immunocomplexes were performed by Western blot as described previously (41). Briefly, 1 mg of total protein was incubated with an anti-phosphoserine antibody for 2 h and then precipitated with protein A-Sepharose (Amersham Biosciences) for 1 h. The immunocomplexes were washed, boiled in Laemmli buffer, separated by SDS-PAGE, and transferred onto polyvinylidene difluoride membranes (Amersham Biosciences). Data are representative of three independent experiments.
Casein Kinase Assay-The kinase activity was measured by a casein kinase II assay kit (Upstate Biotechnology) following the manufactur-er's instructions. Briefly, we measured the phosphorylation of the synthetic peptide RRRDDDSDDD by EMT-6 cell lysates (25 g) (36) in the presence of a protein kinase A inhibitor.

IL-1␤ Induces NOSII Expression and Nitric
Oxide Production in EMT-6 Cells-By limited dilution of the EMT-6 murine mammary cancer cell line, we isolated two cell clones with distinct NO production upon stimulation. By measuring nitrite accumulation in culture medium in response to IL-1␤ stimulation, EMT-6 clone J (EMT-6J) was observed to produce at least 3-fold more NO than clone H (EMT-6H) (Fig. 1A). Western blot analysis failed to detect any NOSII protein in these two clones when studied in the absence of stimulation. Upon IL-1␤ stimulation, NOSII protein could be detected in both clones but reached a higher level in EMT-6J than EMT-6H cells (Fig.  1B). Northern blot analysis indicated that IL-1␤ increased NOSII mRNA level in both cell clones. Again, this effect was more important in EMT-6J compared with EMT-6H cells (Fig. 1C). Using a luciferase reporter gene under control of a 1.7-kb NOSII mouse promoter, we measured a 3-fold higher induction in EMT-6J compared with EMT-6H cells in response to IL-1␤. We concluded that the differential response to IL-1␤ stimulation may be related to distinct expression or regulation of the transcription factors involved in NOSII gene transcription. Fig. 2A, mutation of the downstream b site in the mouse NOSII promoter abolished the response to IL-1␤ in both EMT-6J and EMT-6H clones. Using another construct in which the luciferase reporter gene was  4 -luc and either a siRNA control or a CKII-␣ subunit-specific siRNA before stimulation with 0.55 ng/ml IL-1␤ for 15 h. Activity: -fold induction relative to control non-stimulated cells transfected with a control siRNA. *, n ϭ 3, p Յ 0.05 versus EMT-6J cells co-transfected with a control siRNA and exposed to IL-1␤. driven by four b sites in tandem, we observed that NF-B transcriptional activity induced by exposure to IL-1␤ was 2-3fold higher in EMT-6J compared with EMT-6H cells (Fig. 2B).

NF-B Activation in Response to IL-1␤ Appears Similar in Both EMT-6
Clones-To understand why the response to NF-B was more important in EMT-6J that in EMT-6H cells, we first analyzed the kinetics of degradation of its inhibitor. We observed that the two IB isoforms, namely IB-␣ and IB-␤, were degraded in the two EMT-6 clones within 30 min of IL-1␤ stimulation. Sixty minutes after stimulation, the level of IB-␣ increased again, suggesting new synthesis (Fig. 3A). IB-␣ phosphorylation was observed in both EMT-6 clones with the same kinetics, i.e. this event occurred 10 min after IL-1␤ stimulation (Fig. 3B). In addition, when studied by indirect immunofluorescence using a polyclonal antibody (Fig. 3C), the p65 protein was identified in the cytoplasm of resting cells and transiently accumulated in the nucleus of IL-1␤-stimulated cells. The kinetics of p65 cellular redistribution correlated with IB degradation and synthesis, i.e. the protein accumulated in the nucleus 15 min after IL-1␤ stimulation and re-appeared in the cytoplasm within 60 min.
Then, we analyzed the DNA binding activity of NF-B by using the downstream B site (bD) in NOSII gene promoter as a probe for an EMSA. NF-B complexes bound to this site with a similar kinetic in both EMT-6 clones (Fig. 4A). Two complexes were highly induced 15 min after IL-1␤ stimulation and rapidly decrease 45 min later. Their specificity was checked by competition with a 50-fold excess of unlabeled specific (bD) and nonspecific (OCT) oligonucleotides (Fig. 4B), and their composition was determined using specific antibodies (Fig. 4C); the slow migrating complex was shown to be the p50/p65 heterodimer, whereas the fast migrating complex was identified as the p50 homodimer in both clones (Fig. 4C, data not shown for EMT-6H). CKII Phosphorylates p65 in EMT-6J Cells-Recent reports (28 -30) have demonstrated that the transactivating potential of p65 could be enhanced by serine phosphorylation. Serinephosphorylated proteins were immunoprecipitated from EMT-6H and EMT-6J cells by using a specific anti-phosphoserine antibody, and the presence of p65 in these immune complexes was detected by Western blot (Fig. 5A). While no phosphorylation of p65 could be detected in EMT-6H clone at any time point after IL-1␤ treatment, a phosphorylated protein was identified in EMT-6J 10 min after IL-1␤ stimulation. This  1␤, 20 min). B, EMT-6 clones were transiently transfected with control or CKII-␣ subunit-specific siRNA and lysed 24 h later. Upper panel, Western blot analysis of CKII-␣ and HSC70 expression (NOSII was never detected in the studied conditions, not shown). Lower panel, CKII activity measured in cell lysates is expressed as a percentage of EMT-6H cells transfected with the control siRNA (100%). C, Western blot analysis of NOSII and CKII-␣ in indicated clones transfected with either a control or a CKII-␣ subunit-specific siRNA, then stimulated with IL-1␤ for 6 h. D, luciferase activity in indicated clones co-transfected with a pNOSII-luc reporter gene and either control or CKII-␣ subunit specific siRNA, then stimulated or not with IL-1␤ for 15 h. Activity is expressed as -fold induction relative to control non-stimulated cells co-transfected with a control siRNA. n ϭ 3, p Յ 0.05 versus EMT-6J cells co-transfected with a control siRNA, then stimulated with IL-1␤. ns, non-stimulated. phosphorylation appeared to be maximal 20 min after stimulation, then decreased to basal level 40 min after cytokine treatment.
To identify the kinase(s) mediating p65 phosphorylation in EMT-6J cells, we tested a variety of protein kinase inhibitors. As shown in Fig. 5B, pretreatment of EMT-6J cells with either H89, an inhibitor of PKA and MSK1, or LY-294002, an inhibitor of phosphatidylinositol 3-kinase, had no effect on IL-1␤induced p65 phosphorylation. In contrast, cell pretreatment with DRB, an inhibitor of CKII, prevented IL-1␤-induced p65 phosphorylation in these cells. In addition, siRNA-mediated depletion of CKII-␣ subunit strongly decreased IL-1␤-induced NF-B activity in EMT-6J cells, as demonstrated by co-transfection with the p(NF-B) 4 -luc reporter gene. In these conditions, IL-1␤-induced NF-B activity returned to the level observed in EMT-6H cells (Fig. 5C). Conversely, siRNA-mediated depletion in CKII-␣ subunit did not influence NF-B activity in EMT-6H cells.
It was shown that phosphorylation of IB-␣ by CKII regulates its intrinsic stability in non-stimulated cells but does not affect its TNF-␣-induced degradation (42). Western blot analysis indicated that the CKII inhibitor did not alter the basal expression of IB-␣ and IB-␤ in either clone nor did it influence the modulation of their expression induced by IL-1␤ (Fig.  5D, EMT-6H not shown) CKII Is Responsible for the Increased NOSII Expression Observed in IL-1␤-stimulated EMT-6J Cells-CKII activity can be dramatically decreased in the studied cells by a 1-h pre-treatment with 30 M DRB (Fig. 6A). Immunoblot analysis of NOSII expression demonstrated that pretreatment with 30 M DRB also decreases NOSII protein level in IL-1␤-stimulated EMT-6J cells without having any affect on NOSII expression in EMT-6H cells treated in the same conditions (Fig. 6A). To confirm the results obtained with DRB pretreatment, cells were transiently transfected with CKII-␣ subunit-specific siRNA, which reduced both CKII-␣ expression and CKII basal activity in both clones (Fig. 6B). Immunoblot analysis of NOSII expression showed that depletion of the CKII-␣ subunit decreased NOSII protein level in IL-1␤-stimulated EMT-6J cells without having any affect on NOSII expression in EMT-6H cells treated in the same conditions (Fig. 6C). In addition, co-transfection of siRNA with the pNOSII-luc reporter gene indicated that the CKII-␣ subunit was involved in NOSII gene stimulation in EMT-6J cells but not in EMT-6H cells (Fig. 6D). In all these experiments, CKII depletion or inhibition decreased the level of NOSII in IL-1␤-stimulated EMT-6J cells to that observed in EMT-6H cells treated in the same conditions. CKII Is Responsible for the Increased NOSII Expression Observed in LPS-stimulated EMT-6J Cells-To determine whether CKII played a role in the transcriptional regulation of NOSII induced by another stimulus, EMT-6 cells were either pretreated or not with DRB 30 M before stimulation with 50 ng/ml of LPS. As shown in Fig. 7A, DRB pretreatment prevented NOSII protein level increase observed in LPS-stimulated EMT-6J cells without modifying NOSII protein level increase in LPS-stimulated EMT-6H cells. In addition, DRB pretreatment prevented LPS-induced phosphorylation of p65 in EMT-6J cells (not shown). We also observed that siRNAmediated decrease in CKII-␣ expression prevented the overactivation of NF-B-driven reporter genes in LPS-stimulated EMT-6J cells without affecting the activation of NF-Bdriven reporter genes in LPS-stimulated EMT-6H cells (Fig. 7,  B and C). DISCUSSION The present study demonstrates that CKII induces the phosphorylation of p65 and enhances NF-B-dependent transcription of the NOSII gene in the EMT-6 mammary cancer cell line.

FIG. 7. Effects of casein kinase II on NOSII expression in EMT-6 clones stimulated with 50 ng/ml LPS.
A, Western blot analysis of NOSII in indicated clones exposed to 30 M DRB for 1 h, then stimulated with LPS for 6 h (NOSII was never detected in control cells, see Fig. 6A). B and C, luciferase activity in indicated clones co-transfected with a pNOSII-luc reporter gene and either control or CKII-␣ subunit specific siRNA (B) or with an NF-B-dependent-luciferase reporter gene p(NF-B) 4 -luc and either a siRNA control or a CKII-␣ subunit specific siRNA (C), then stimulated or not with LPS for 15 h. Activity is expressed as -fold induction relative to control non-stimulated cells co-transfected with a control siRNA. n ϭ 3, p Յ 0.05 versus EMT-6J cells co-transfected with a control siRNA, then stimulated with LPS. ns, non-stimulated. This phosphorylating event accounts for the heterogeneity of EMT-6 cells with regards to NO synthesis in response to IL-1␤ or LPS. As far as we know, this is the first report demonstrating that casein kinase II efficiently modulates the expression of an NF-B-dependent gene in vivo.
EMT-6 cells have been shown to produce NO in response to various stimuli including cytokines (43), LPS (44), and chemotherapeutic drugs (45). We show here that these cells are heterogeneous with regards to NO synthesis. Such a heterogeneity has been previously described in K1735 murine melanoma cells (46). The decreased NO synthesis observed in some of these latter cells has been related to decreased NOSII transcription through unidentified mechanisms (47). Mutation of the proximal b site in the NOSII promoter abolishes NOSII transcription induced by IL-1␤ in EMT-6 cells, in agreement with the role ascribed to this site in previous studies (12,21). By cloning EMT-6 cells with different NO production in response to stimuli, we demonstrate that the absence of CKII-mediated phosphorylation of p65 is responsible for the decreased NF-Bmediated activation of NOSII gene in some of these cells.
Various mechanisms could account for the enhanced NF-B activity in EMT-6J compared with EMT-6H cells. We did not observe any differential phosphorylation of IB-␣ by IKKs, and both IB-␣ and IB-␤ were degraded with similar kinetics in both clones. The DNA binding properties of NF-B, the subunits forming the DNA binding dimers, and the nuclear export of p50/p65 heterodimers were also observed to be similar in the two clones. In contrast, we identified a differential phosphorylation status of p65 at serine residues when the two clones are stimulated with IL-1␤ or LPS. The p65 phosphorylation identified in EMT-6J cells appears to be transient, lasting 30 min after IL-1␤ stimulation, which could be related to the interaction of p65 with the protein phosphatase 2Ac as described in human melanocytes (48). Phosphorylation of several p65 serine residues facilitates the docking of proteins such as the transcriptional co-activator p300/CBP (33,34), which enhances NF-B transcriptional activity, e.g. by recruiting BRCA1 and PARP-1 within the enhanceosome (49,50). Several protein kinases have been shown to directly phosphorylate the p65 subunit of NF-B at serine residues, including the catalytic subunit of PKAc, IKK complexes, phosphatidylinositol 3-kinase, casein kinase II (31,35,36,51), and more recently MSK1 and PKC (32,33). In EMT-6J cells, the casein kinase II inhibitor (DRB) was the only tested protein kinase inhibitor to affect p65 phosphorylation induced by stimuli. CKII had been previously identified to associate with NF-B complexes in the cytoplasm of IL-1␤-stimulated HepG2 cells, which correlated with p65 phosphorylation (52). However, this report did not show whether this phosphorylation contributed to NF-B transcriptional potential. More recently, a pretreatment with a casein kinase II inhibitor (PD144795) of HeLa cells stimulated with TNF-␣ was shown to prevent transactivation properties of a NF-B-dependent reporter gene without affecting its DNA binding activity (36). While our results confirm these earlier findings, they furthermore demonstrate that CKII efficiently modulates the expression of an NF-B-dependent gene in vivo. We demonstrate that siRNA-mediated depletion of the CKII-␣ subunit completely abolishes the additional activation of NF-B, identified in EMT-6J cells when compared with EMT-6H cells. Thus, although casein kinase II-mediated phosphorylation of p65 is not an absolute requirement for NF-B activity in stimulated-EMT-6 cells, this event is responsible for the strong enhancement in the transcriptional activity of NF-B, without affecting its DNA binding activity, and accounts for the enhanced NOSII transcription in some of these cells. CKII-mediated phosphorylation of p65 may facilitate the formation of a more efficient enhanceosome by forming docking sites for co-activators or adjacent transcription factors. The basal level of CKII activity and the increased activity triggered by IL-1␤ stimulation were similar in EMT-6H and EMT-6J cell lysates. Thus, the difference between the two cell lines is either located downstream of CKII or related to other parameters that modulate the enzyme activity, e.g. the subcellular distribution of its subunits (53).
CKII is a highly conserved and ubiquitous serine/threonine kinase present as a holoenzyme composed of two catalytic subunits (␣␣, ␣Ј␣Ј, or ␣␣Ј) and two ␤ regulatory subunits (reviewed in Ref. 54) or in a dissociated form in different cellular compartments (53,55,56). Its regulation and function are not well understood. The currently identified targets of this enzyme include signaling viral and structural proteins, metabolic enzymes, and transcription factors (reviewed in Ref. 57). CKII seems thus to be a critical component of the NF-B signaling pathway, regulating the intrinsic stability of IB-␣ by constitutive phosphorylation at C-terminal sites (42,58), leading to UV-induced IB degradation by inducible phosphorylation in the same domain (59) and enhancing transactivation properties by phosphorylation of the p65 subunit (36).
Our study identifies NF-B as a transcription factor whose activity can be modulated in vivo by casein CKII through the phosphorylation of p65. In a cancer cell line exposed to either IL-1␤ or LPS, this phosphorylating event is shown to account for an increase in NOSII gene expression and NO synthesis. Thus, casein kinase II appears as a potential target to modulate NF-B activity and NO synthesis in tumor cells.