Interferon-γ Stimulates the Expression of the Inducible cAMP Early Repressor in Macrophages through the Activation of Casein Kinase 2

Interferon-γ (IFN-γ) is a pleiotropic cytokine that modulates the immune function, cell proliferation, apoptosis, macrophage activation, and numerous other cellular responses. These biological actions of IFN-γ are characterized by both the activation and the inhibition of gene transcription. Unfortunately, in contrast to gene activation, the mechanisms through which the cytokine suppresses gene transcription remain largely unclear. We show here for the first time that exposure of macrophages to IFN-γ leads to a dramatic induction in the expression of the inducible cAMP early repressor (ICER), a potent inhibitor of gene transcription. In addition, a synergistic action of IFN-γ and calcium in the activation of ICER expression was identified. The IFN-γ-mediated activation of ICER expression was not blocked by H89, bisindoylmaleimide, SB202190, PD98059, W7, and AG490, which inhibit protein kinase A, protein kinase C, p38 mitogen-activated protein kinase, extracellular signal-regulated kinase, calcium-calmodulin-dependent protein kinase, and Janus kinase-2, respectively. In contrast, apigenin, a selective casein kinase 2 (CK2) inhibitor, was found to inhibit response. Consistent with this finding, IFN-γ stimulated CK2 activity and the level of phosphorylated cAMP response element-binding protein, which is known to induce ICER gene transcription, and this response was inhibited in the presence of apigenin. These studies, therefore, identify a previously uncharacterized pathway, involving the IFN-γ-mediated stimulation of CK2 activity, activation of cAMP response element-binding protein, and increased production of ICER, which may then play an important role in the inhibition of macrophage gene transcription by this cytokine.

Transcriptional regulation coupled to the cAMP-dependent signal transduction pathways is mediated through a related family of DNA-binding proteins comprising the cAMP-response element-binding protein (CREB), 1 cAMP-response element modulator protein (CREM), and activating transcription factors (ATF) (1,2). These proteins belong to the basic leucine zipper (bZIP) class of transcription factors and bind to the cAMP-response elements (CREs) in the promoter and enhancer regions of target genes (1,2). The increased level of cAMP induces the trans-activation potential of these proteins by a protein kinase A-mediated phosphorylation of a critical serine residue (Ser-133 in CREB) (1,2). However, activation can also be triggered via phosphorylation by other kinases, including protein kinase C, calcium-calmodulin-dependent protein kinases, casein kinase 2, and mitogen-activated protein kinases (1,2). In addition to phosphorylation, heterodimerization between the different family members has a profound effect on the regulation of cAMP-mediated gene transcription (1,2).
The CREM gene encodes multiple members by alternative splicing and internal transcriptional initiation through the use of two promoters (1)(2)(3). These proteins act as both activators and antagonists of cAMP-mediated gene transcription, with inducible cAMP early repressor (ICER) being the most potent repressor (1)(2)(3). ICER is produced from an internal CREM gene promoter (P 2 ), which, in contrast to the P 1 promoter that produces all CREM isoforms, contains CREs (1)(2)(3). The ICER protein consists mainly of the bZIP domain and, therefore, represses transcription either by heterodimerization with activating forms of CREB/CREM/ATF or other bZIP-containing transcription factors, or by competition with these proteins for DNA binding (1,2). The kinetics of cAMP-induced ICER expression is typical of early response genes, with maximal expression levels being attained within 3-6 h of stimulation (1)(2)(3). ICER then inhibits both its own transcription and the expression of other CRE-containing promoters (1)(2)(3). Indeed, the induced expression of ICER through the cAMP signaling pathway has been implicated in several responses, including the suppression of T lymphocyte function (4,5), attenuation of Fas ligand expression in T and NK lymphocytes (6), glucagonmediated suppression of insulin gene expression in pancreatic B cells (7), and the noradrenaline-triggered suppression of ␤2adrenoreceptor in brown adipocytes (8). The expression of ICER has also recently been shown to be induced by cAMPindependent pathways, and includes the action of nerve growth factor, epidermal growth factor, and the gastrointestinal peptides gastrin and cholecystokinin (9 -11).
Our laboratory is interested in understanding the mechanisms involved in the regulation of macrophage gene expression by cytokines, particularly interferon-␥ (IFN-␥) (e.g. Refs. [12][13][14][15][16]. This cytokine is a key modulator of the immune and inflammatory responses along with numerous other cellular actions during physiological and pathophysiological conditions (17,18). Such biological responses are characterized by the transcriptional activation and the inhibition of numerous genes (17,18). The mechanisms through which IFN-␥ induces gene transcription have been investigated over the last decade and identified a key role for the Janus kinase (JAK)-signal transducer and activator of transcription (STAT) pathway (18 -20). However, several recent studies, including gene expression profiling on STAT-1-deficient cells, have revealed the existence of STAT-1-independent pathways (20 -22), which remain to be deciphered in detail. In addition, although IFN-␥ suppresses the expression of numerous genes, little is understood of the corresponding mechanisms, which may be the result, at least in part, of limited research on this aspect that has been carried out to date. Further studies are required, especially in the light of the potential biological importance of such regulation. For example, IFN-␥ has been shown to inhibit macrophage foam cell formation, a critical early event in atherogenesis, at least in part by suppressing the expression of several genes that are involved in the uptake of lipoproteins, including lipoprotein lipase (LPL), very low density lipoprotein receptor, low density lipoprotein receptor-related protein, and the scavenger receptors A and CD136 (12,(23)(24)(25)(26)(27).
During the course of our studies on the regulation of macrophage gene expression by IFN-␥, we were intrigued by the presence of ICER binding sites in the promoter regions of many IFN-␥ inhibited genes, 2 the co-regulation of some of these genes by cAMP (e.g. LPL) (28), and a study demonstrating the ability of IFN-␥ to suppress CRE-or the bZIP-containing AP1-regulated promoters (29), and wondered whether this cytokine induces ICER expression. We present here studies that investigate the IFN-␥-ICER link in detail. We show for the first time that IFN-␥ produces a dramatic increase in ICER expression in macrophages, and acts in a synergistic manner with agents that induce intracellular calcium levels. In addition, we identify the mechanisms through which IFN-␥ induces ICER expression. The studies, therefore, indicate the existence of a potentially novel pathway for IFN-␥-mediated suppression of macrophage gene transcription.
Cell Culture-The J774.2 cell line was maintained in Dulbecco's modified Eagle's medium, which was supplemented with 10% (v/v) heat-inactivated fetal calf serum (56°C, 45 min), 2 mM L-glutamine, 100 units/ml penicillin, and 100 g/ml streptomycin (designated as complete medium). The cultures were maintained at 37°C in a humidified atmosphere containing 5% (v/v) CO 2 in air. For human monocyte-derived macrophages, blood was collected from healthy volunteers into a syringe containing an equal volume of sterile 2% (w/v) dextran in 1ϫ PBS containing 0.8% (w/v) trisodium citrate and was allowed to stand for 30 min to permit erythrocyte sedimentation. The upper layer was collected, underlayered with Lymphoprep TM (Nycomed Pharmaceuticals) (2:1 (v/v) supernatant:Lymphoprep TM ) and centrifuged at 800 ϫ g for 20 min. The resultant interface was collected and washed six to eight times with PBS, 0.4% (w/v) trisodium citrate to remove contaminating platelets. Monocytes were plated out in six-well plates (1 ϫ 10 6 cells/well) in complete medium, as above except containing 10% (v/v) human serum, and were allowed to adhere to the surface for 6 h in a humid incubator at 37°C containing 5% (v/v) CO 2 . The cells were then washed twice with PBS to remove any other mononuclear cells, and fresh medium was added. The homogeneity of macrophages was verified by morphological analysis.
The PCR products were size-fractionated on a 2% (w/v) agarose gel, photographed using a Syngene gel documentation system (GRI), converted to an uncompressed TIFF file format, and quantified using the Quantiscan computer package (Biosoft).
Cloning of ICER I and I␥ into the pcDNA3 Vector-J774.2 macrophages were stimulated for 12 h with 8-Br-cAMP and, following isolation of RNA and cDNA synthesis as above, the ICER I/I␥ isoforms were amplified using primers containing BamHI and EcoRI restriction sites to facilitate the cloning into the pcDNA3 vector (Invitrogen) in the correct orientation (5Ј-ACTTTAGGATCCACTGTGGTACGGCCAAC-3Ј and 5Ј-GAGCTCGAATTCCCAATTCACACTCTACAGCAG-3Ј, with the restriction endonuclease sites shown in bold) (39). The amplification was carried out using the high fidelity Pwo DNA polymerase to minimize PCR-generated mutations (40,41), and this was confirmed by sequencing. The amplification product (ϳ530 bp) was purified, digested with BamHI and EcoRI, and subcloned into the pcDNA3 vector that had been previously digested with the same enzymes.
In Vitro Transcription/Translation Reactions and Preparation of Whole Cell Extracts-In vitro transcription and translation was carried out using the TNT TM coupled reticulocyte lysate system using the protocol recommended by the manufacturer (Promega). The recombinant protein was labeled with [ 35 S]methionine (40 Ci) to aid detection. Nuclear and whole cell extracts were prepared essentially as described previously (12,15,16,40,42). Protease inhibitors (0.5 mM phenylmethylsulfonyl fluoride, 1 g/ml pepstatin A, 10 g/ml aprotinin, 10 g/ml I-S soybean trypsin inhibitor) and DTT (0.5 mM) were added to all buffers before use. 2 J. R. Mead and D. P. Ramji, unpublished observations.
Where the maintenance of proteins in the phosphorylated state was desired (e.g. immunodetection of phosphorylated protein), whole cell extracts were made in a buffer containing both phosphatase and protease inhibitors (43). Briefly, the cells were washed twice with ice-cold PBS, containing 10 mM sodium fluoride and 100 M sodium orthovanadate, and then scraped off in 500 l of phosphatase-free whole cell extraction buffer (10 mM Tris-HCl, pH 7.05, 50 mM NaCl, 50 mM sodium fluoride, 1% (v/v) Triton X-100, 30 mM sodium pyrophosphate, 5 M zinc chloride, 100 M sodium orthovanadate, 1 mM DTT, 2.8 g/ml aprotinin, 2.5 g/ml each of leupeptin and pepstatin, 0.5 mM benzamidine, and 0.5 mM phenylmethylsulfonyl fluoride). After vortexing for 45 s at 4°C, the lysates were cleared by centrifugation at 10,000 ϫ g for 10 min at 4°C. Aliquots were immediately frozen at Ϫ70°C until use.
The concentration of proteins in extracts was determined using the MicroBCA protein assay kit as described by the manufacturer (Pierce).
SDS-PAGE and Western Blot Analysis-Samples (10 g of whole cell extracts or 5 l of a 50-l in vitro transcription/translation reaction) were size-fractionated under reducing conditions using 10 -15% (w/v) polyacrylamide gels containing SDS (12,13,42). The in vitro translated protein was detected by fluorography (31,44). For Western blot analysis, the proteins on the gel were transferred to Immobilon-P polyvinylidene difluoride membranes (Millipore) by blotting (12,42). The blotted membranes were incubated first for 1 h at room temperature in blocking solution (1ϫ TBS containing 5% (w/v) nonfat milk powder and 0.05% (w/v) Tween 20) to reduce any nonspecific interaction of the antiserum with the membrane. Following three washes of 5 min each with 1ϫ TBS containing 0.05% (v/v) Tween 20, the membrane was incubated with the primary antibody for 1 h in 1ϫ TBS containing 5% (w/v) nonfat milk powder and 0.1% (v/v) Tween 20. After washes as above, the membrane was incubated with secondary horseradish peroxidase-conjugated antibodies (goat anti-rabbit IgG or mouse anti-goat IgG) for 30 min at room temperature in 1ϫ TBS containing 5% (w/v) nonfat milk powder and 0.05% (v/v) Tween 20. After washing once more as above, the membranes were developed using an enhanced chemiluminescence detection kit (Amersham Biosciences) and XAR sensitive film (Eastman Kodak Co.). The sizes of the proteins were determined by comparison with Rainbow molecular weight markers (Amersham Biosciences) that had been subjected to electrophoresis and blotting on the same gel as the test samples.
The oligonucleotides were radiolabeled by "fill-in" reactions using [␣-32 P]dCTP and Klenow DNA polymerase. Whole cell extracts (2.5-30 g) or in vitro translated protein (3 l of a 50-l reaction) were incubated in a total reaction volume of 20 l containing 34 mM potassium chloride, 5 mM magnesium chloride, 0.1 mM DTT, and 3 g of poly(dI-dC). After 10 min on ice, 32 P-labeled probes (50,000 cpm) were added and the incubation continued for 30 min at room temperature. Following the addition of 5 l of a 20% (w/v) Ficoll solution to each sample, the free probe and the DNA-protein complexes were resolved on 4 -6% (w/v) polyacrylamide gels in 0.5ϫ TBE (45 mM Tris base, 45 mM boric acid, 1 mM EDTA). The gels were then dried under vacuum and exposed to x-ray film. For antibody supershift assays, samples of nuclear or whole cell extracts were incubated with the appropriate antiserum for 30 min on ice prior to the addition of the radiolabeled probe (12,15,16,30,32,40,42). For competition assays, a 250 -500-fold molar excess of the double-stranded oligonucleotides was added to the samples of proteins prior to the addition of the radiolabeled probe (12, 15, 16, 30 -32, 40, 42).
Measurement of Nitrite Formation Using the Greiss Reagent-Determination of nitrite levels, an indicator of nitric oxide (NO) production by inducible nitric-oxide synthase, was carried out as described by Ruetten and Thiemermann (45) with minor modifications. J774.2 cells were grown in 24-well plates with 500 l of culture medium and exposed for 24 h with the respective mediators (see "Results"). The nitrite levels were then determined by adding 100 l of Greiss reagent (Sigma) to 100 l of conditioned medium in a 96-well plate. Optical density at 560 nm was measured in a MRX microplate reader (Dynex Technologies), and the nitrite concentration was calculated by comparison with the A 560 of sodium nitrite standards prepared in the cell culture media.
Casein Kinase 2 Assay-The method used was a combination of those employed by Sung et al. (46) and Lodie et al. (47) with minor modifications. The cells were incubated with the mediators for the requisite time, and whole cell extracts were prepared using phosphatase-free buffer as detailed above. CK2 was then immunoprecipitated as follows. The extracts were first precleared by the addition of the protein A/Gagarose beads (20 l; Santa Cruz) with incubation and gentle rolling at 4°C for 1 h, followed by centrifugation at 13,000 ϫ g for 3 min to remove the beads. The precleared supernatant was mixed with the anti-CK2␣chain antiserum (2 g/ml) and incubated overnight at 4°C with gentle rolling. The resulting protein-antibody complex was isolated by the addition of protein A/G-agarose (20 l) with gentle rotary mixing for 2 h at 4°C. The beads, containing the immunocomplex, were collected by centrifugation (13,000 ϫ g for 3 min at 4°C) and washed once with phosphatase-free cell extraction buffer (see above). The pellet was then resuspended in 25 l of kinase buffer (100 M sodium orthovanadate, 100 mM Tris-HCl (pH 8), 100 mM sodium chloride, 20 mM magnesium chloride, 50 mM potassium chloride) containing 5 Ci of [␥-32 P]ATP and 5 mg/ml ␤-casein. Kinase reactions were incubated for 15 min at 37°C and stopped by the addition of 10 l of reducing solubilizing buffer (50 mM Tris-HCl (pH 6.8), 100 mM DTT, 2% (w/v) SDS, 0.1% (w/v) bromphenol blue, 10% (v/v) glycerol). Samples were boiled for 10 min and subjected to SDS-PAGE using 15% (w/v) acrylamide gels. After electrophoresis, the gels were fixed for 20 min in a solution containing 40% (v/v) methanol and 10% (v/v) acetic acid. These were washed once with distilled water before being dried under vacuum and visualized by autoradiography.

ICER Gene Expression in the Macrophage J774.2 Cell Line-
The expression of ICER has been shown to be induced rapidly in several cell types in response to increase in intracellular cAMP levels (e.g. Refs. 3, 7, 9, and 10). However to our knowledge, the expression of ICER isoforms in macrophages has, as yet, not been determined. To examine ICER expression, murine J774.2 macrophages were incubated with the membrane-permeable, non-hydrolyzable cAMP anologue, 8-Br-cAMP, and forskolin, a cell-permeable activator of adenylyl cyclase, for various periods of time. Total RNA was isolated and subjected to RT-PCR using primers specific for ICER and the control ␤-actin gene. For mouse ICER, four functionally indistinct isoforms, ICER I, ICER I␥, ICER II, and ICER II␥, have been identified that arise as a result of alternative splicing of a transcript produced from the CREM P 2 promoter (see Fig. 1A for a schematic representation of the different ICER isoforms). Our initial RT-PCR was carried out using representative time points (0, 6, 12, and 24 h) from 8-Br-cAMP-treated cells using primer sets that allow amplification of all four ICER transcripts (primers 1 and 2 in Fig. 1A). As shown in Fig. 1B, the mRNA levels of all ICER isoforms was increased by 8-Br-cAMP treatment, with the induction of ICER I/I␥ being more dramatic than that for ICER II/II␥. As a result, all further studies on J774.2 macrophages were restricted to the examination of ICER I/I␥ expression (primers 1 and 3 in Fig. 1A). A detailed time course experiment was carried out with cells exposed to 8-Br-cAMP or forskolin. The expression of ICER I/I␥ RNA was induced immediately following incubation of the cells with 8-Br-cAMP, peaking at ϳ3 h and being maintained at similar or slightly reduced levels until 24 h (Fig. 1C). On the other hand, the levels of ICERI/I␥ were induced immediately following addition of forskolin to the cells, peaking at 2 h, and declining thereafter to reach almost undetectable levels at 6 h (Fig. 1D). The induction in these experiments was specific to the addition of 8-Br-cAMP or forskolin and not the result of any alterations in the status of the cells over time because no ICER mRNA expression was observed in untreated cells at 4, 12, and 24 h in the experiment.
ICER levels in the rat pancreatic cell line AR42H have been shown recently to be modulated by the action of PMA (10), a protein kinase C agonist. We decided to perform a similar experiment on the J774.2 cell line because the effect of this signaling cascade on ICER expression in macrophages had not previously been carried out. As shown in Fig. 1E, the addition of PMA to the cells also resulted in an immediate increase in ICERI/I␥ mRNA levels that peak at 1 h and then decline gradually to reach levels at 6 h that were similar to those seen in untreated cells.
To confirm that the amplification products shown in Fig. 1 were indeed derived from the ICER transcripts, the DNA was subcloned into the pcDNA3 expression vector as described under "Experimental Procedures." Sequence analysis confirmed that the products corresponded to ICER I and ICER I␥; the sequences have been deposited in the EMBL nucleotide sequence data base under the accession numbers AJ311667 and AJ292222, respectively. In addition, in vitro transcription and translation of the recombinant plasmid, in the presence of [ 35 S]methionine, followed by SDS-PAGE of the protein and fluorography showed a protein with a relative molecular mass (ϳ13.5 kDa) that was consistent with the size expected from the cDNA insert (data not shown). Furthermore, this in vitro translated protein was demonstrated to interact specifically with a CRE (see below). Moreover, transient transfection assays showed that the LPL promoter, which contains a number of CRE-like sequences and expression of which in J774.2 macrophages is suppressed by both 8-Br-cAMP 3 and IFN-␥ (12)(13)(14), was inhibited in a dose-dependent manner by an ICER expression plasmid (data not shown).
We next examined whether the induction of ICER I/I␥ mRNA seen in the experiments shown in Fig. 1 was also accompanied by an increase in the expression of the ICER protein. Whole cell protein extracts from J774.2 macrophages were, therefore, subjected to Western blot analysis using an anti-CREM-1 primary antibody, which recognizes ICER. Anti-CREM antiserum has been used in a number of studies to investigate ICER expression (3, 4, 6, 11, 33, 39, 48 -50). As shown in Fig. 2A, the levels of the 13.5-kDa ICER protein were induced following exposure of the cells to 8-Br-cAMP. The level of the protein at 3 h following treatment of the cells with 8-Br-cAMP was substantially greater than untreated cells, and this is likely to be of major functional significance given that ICER represses the action of the activating members of the CREB/CREM/ATF family at substoichiometric levels (1-3). The maximal levels were attained at 12-24 h and were, therefore, delayed compared with that for the mRNA (see Fig. 1C). A delay in the induction of ICER protein compared with the corresponding RNA has also been reported previously for forskolin-treated AtT20 cells (3).
Finally, the effect of 8-Br-cAMP on ICER DNA binding was investigated using extracts that were employed for Western blot analysis (Fig. 2A). The ICER protein binds, either as homodimers or heterodimers with other CREB/CREM/ATF family members, to CRE recognition sites in the regulatory regions of target genes to exert the transcriptional repressive effects. A CRE binding site oligonucleotide, such as that from the somatostatin gene promoter has been, therefore, used extensively to study the DNA binding properties of ICER, and also other members of the CREB/CREM/ATF family, from different cell types (3,10,51). The binding of the ICER homodimers is identified by their fastest migration during gel electrophoresis because ICER is the smallest member of the CREB/CREM/ATF family (consisting essentially of the CREM bZIP region), similar migration to that obtained using in vitro or bacterially produced ICER, and the inhibition of binding by an anti-CREM antiserum, which recognizes the protein. As shown in Fig. 2B, at least three DNA-protein complexes were obtained when extracts from J774.2 macrophages were used. None of these complexes were present when only the radiolabeled probe was used (data not shown). All these three complexes could be competed out using a 250-or 500-fold molar excess of the unlabeled somatostatin CRE element but not with the same excesses of unrelated, unlabeled C/EBP or Sp1 binding site oligonucleotides (Fig. 2C). This indicates that all three complexes represent specific interaction of proteins with the CRE binding site oligonucleotide (i.e. belong to the CREB/CREM/ ATF family). From these, the signal from the fastest migrating complex was judged to be composed of ICER because its migration was consistent with that expected from the smallest CREbinding protein. Indeed, this complex migrated only marginally more slowly than the single complex produced by in vitro translated ICER protein (Fig. 2B)  could be caused by some form of post-translational modification such as phosphorylation; see, e.g., Ref. 52). In addition, the use of the CREM-1 antiserum, which also recognizes ICER, in EMSA could completely inhibit the formation of this complex and that formed by the in vitro translated protein (Fig. 2B). The kinetics of the formation of this complex in response to 8-Br-cAMP were similar to that seen at the level of the ICER protein ( Fig. 2A), being absent with extracts from untreated cells but increasing dramatically in a time-dependent manner following addition of 8-Br-cAMP to the cells, with maximal binding seen at 24 h (Fig. 2B).
IFN-␥ Induces ICER Expression in Macrophages-Having established that ICERs are expressed in macrophages, we next determined whether their expression could be regulated by IFN-␥. Time-course RT-PCR showed that exposure of J774.2 macrophages to IFN-␥ results in a dramatic induction of ICER I/I␥ mRNA expression, peaking between 1 and 4 h, and declining gradually thereafter (Fig. 3A). Western blot analysis using an anti-CREM-1 antiserum showed that the activation of ICER I/I␥ mRNA expression was followed by the corresponding protein (Fig. 3B). However, similar to the case with cells exposed to 8-Br-cAMP ( Fig. 2A), there was a lag between mRNA and protein expression, and this may be peculiar to macrophages. EMSA showed that IFN-␥ induced the binding of the fastest migrating ICER-CRE complexes in a time-dependent manner, and that the formation of these complexes was inhibited by the inclusion of the anti-CREM antiserum but not the nonimmune serum (Fig. 3C). The presence of two closely migrating ICER-CRE complexes is consistent with the possibility of a number of similar sized ICER proteins being produced from the various transcripts generated by alternative splicing (see Fig. 1A). The kinetics of the induction of the ICER-CRE complexes in response to IFN-␥ treatment of the cells was similar to that seen at the level of the ICER protein (Fig. 3B).
The IFN-␥-mediated induction of ICER mRNA expression detailed above was shown in the macrophage J774.2 cell line. To rule out that this was not a peculiar property of this cell line, the experiment was repeated using primary cultures of human monocyte-derived macrophages. Thus, RNA from untreated macrophages and those exposed to IFN-␥ were subjected to RT-PCR using primers that have been used previously to study ICER expression in pituitary adenomas and hyperfunctioning thyroid adenomas (34,35). These primers have been shown to amplify products of 657 and 257 bp, which correspond to ICER I and II, respectively, along with additional product(s) that contain partial sequences of the 657-bp product, which have been postulated to be the result of alternative splicing (35). As shown in Fig. 4, the expression of ICER-I and ICER-II was increased when the cells were treated with IFN-␥.
Having established that IFN-␥ induces the expression of ICER in both the murine J774.2 cell line and primary cultures of human monocyte-derived macrophages, our further studies focused on understanding the potential mechanisms that are involved in this action of the cytokine.
Signal Transduction Pathway(s) Involved in the IFN-␥-mediated Induction of ICER Expression-As described in the Introduction, the activation of CREB by phosphorylation at Ser-133 is a critical step in the production of ICER by virtue of its ability to transactivate the CREM gene P 2 promoter (1-3). It was, therefore, possible that IFN-␥ also induces the cellular levels of CREB phosphorylated at this residue. This possibility was investigated by time-course Western blot analysis using antisera specific for Ser-133-phosphorylated CREB. As shown in Fig. 5, IFN-␥ treatment of J774.2 macrophages does indeed lead to an immediate and marked increase in the level of phosphorylated CREB. Fig. 1 showed that forskolin, an activator of adenylyl cyclase, and thus protein kinase A-mediated phosphorylation of CREB (1-2), could induce ICER expression. We, therefore, investigated whether IFN-␥ also acted through this pathway in the activation of ICER expression using H89, an inhibitor of cyclicnucleotide dependent protein kinase. The action of all the pharmacological agents employed in this study was demonstrated not to be the result of any cytotoxic effect because they had no effect on both the viability of the cells, as judged by the trypan blue exclusion assay (data not shown), and the expression of the mRNA for the constitutively expressed marker (see Figs. 6-9). As shown in Fig. 6A, incubation of the cells with H89 did not inhibit the IFN-␥-mediated induction of ICER mRNA expression, indicating that the protein kinase A signaling cascade was not involved. As expected, H89 did prevent the forskolinmediated induction of ICER mRNA levels, thereby proving that the inhibition of ICER gene expression by H89 can be measured using this experimental technique (Fig. 6A). In the light of this finding, the potential role of the other signaling pathways that are involved in the phosphorylation, and thereby activation of CREB, was investigated using a panel of inhibitors. These included: (i) SB202190, an inhibitor of p38 MAP kinase (55); (ii) PD98059, an inhibitor of the extracellular signal-regulated kinase MAP kinase pathway (56); (iii) W7, an inhibitor of calcium-calmodulin-dependent protein kinase (36); (iv) bisindoylmaleimide, an inhibitor of protein kinase C (58); and (v) apigenin, an inhibitor of CK2 (59 -63). In addition, a JAK2 inhibitor, AG490 (64,65), was included because its activation is required for the expression of numerous IFN-␥-regulated genes (17)(18)(19)(20). The concentration of all the inhibitors used was based on an extensive literature survey on their action in macrophages. Overall, the IFN-␥-induced ICER mRNA expression was not affected by incubation of the cells with SB202190, PD98059, W7, bisindoylmaleimide, and AG490 (Fig. 6, B-E). The functionality of these inhibitors was confirmed by investigation of their ability to affect a previously reported response (Fig. 7, A-D). Thus, W7 could inhibit the previously reported LPS-induced expression of IL-1␤ in macrophages (36) (Fig. 7A). In addition, bisindoylmaleimide could inhibit the PMA-medi-ated induction of LPL in the macrophage THP-1 cell line, which is mediated through protein kinase C (66) (Fig. 7B). Similarly, AG490 could successfully inhibit the LPS-induced nitrite production in J774.2 macrophages (45) or the IFN-␥-mediated stimulation of SOCS1 mRNA, which is known to require the JAK-STAT pathway (20) (Fig. 7, C and D). Furthermore, PD98059 and SB202190 were found to inhibit the PMA-induced differentiation of THP-1 monocytes and the reduced serum concentration-mediated activation of LPL mRNA expression. 4 In contrast to all these pharmacological agents, incubation of macrophages with the CK2 inhibitor apigenin prevented the IFN-␥-induced expression of ICER (Fig. 6F). This suggests that CK2 plays an important role in this action of the cytokine and was, therefore, investigated in detail.
The Role of CK2 in the IFN-␥-mediated Induction of ICER Expression-A single concentration of apigenin was used in the experiments shown in Fig. 6F. A dose-response experiment was, therefore, carried out and showed that apigenin inhibited the IFN-␥ induced activation of ICER mRNA expression in a concentration-dependent manner (Fig. 8A). These results, therefore, confirmed that the activation of CK2 was a key event in the IFN-␥-induced ICER expression. However, to our knowl- edge, no studies have been published with respect to IFN-␥ and increased CK2 activity and was thus investigated. For this, macrophages were stimulated with IFN-␥ for 2 h, a time point when the major increase in ICER expression were seen (see Fig. 3A), in the absence or the presence of apigenin. After preparation of whole cell extracts using a buffer containing phosphatase and protease inhibitors, CK2 was immunoprecipitated using an antibody raised against the ␣-subunit and subjected to the in vitro kinase assay as described under "Experimental Procedures." As shown in Fig. 8B, a dramatic increase in the phosphorylation of the ␤-casein substrate was seen with extracts from cells treated with IFN-␥ and this was inhibited in the presence of apigenin. The increased phosphorylation of ␤-casein in this experiment could be caused by a cytokinemediated increase in either the activity or the steady state levels of the protein. Western blot analysis was carried out to distinguish between these two possibilities and showed that IFN-␥ induced CK2 activity in macrophages (Fig. 8C).
Because the level of phosphorylated CREB, an inducer of ICER transcription, was found to increase when macrophages were incubated with IFN-␥ (Fig. 5), it was possible that the activation of CK2 played an important role. We, therefore, investigated whether the IFN-␥-triggered increase in the level of phospho-CREB was inhibited in the presence of apigenin. Western blot analysis showed that this was indeed the case (Fig. 8D).
The Action of Intracellular Ca 2ϩ Levels on the IFN-␥-induced Expression of ICER in Macrophages-Krueger and co-workers (67) have shown recently that elevation of intracellular calcium levels can prevent the forskolin-mediated induction of ICER mRNA expression in the murine WEHI7.2 thymocyte cell line. Furthermore, they demonstrated that this inhibition was mediated by a 83-kDa Ca 2ϩ -activated repressor protein that bound to the ICER promoter region (67). We therefore investigated whether a similar mechanism for the prevention of IFN-␥-induced ICER mRNA expression also existed in macro- FIG. 4. The action of IFN-␥ on ICER mRNA expression in primary cultures of human monocyte-derived macrophages. The cells were either left untreated or stimulated with IFN-␥ for the indicated period of time, and total RNA was subjected to RT-PCR using primers that recognize human ICERs and GAPDH. ϪRT denotes a reaction where no reverse transcriptase was added during the cDNA synthesis step (cDNA from the 2-h sample was used). The products were analyzed by agarose gel electrophoresis along with molecular size markers (M). The positions of the ICER I and ICER II products of approximately 657 and 257 bp, respectively, are shown. * indicates a smaller product that has been observed previously with these primers and contains partial sequences of ICER I (35). Panel A also shows the positive control for H89, the previously noted inhibition of forskolin-mediated ICERI/I␥ expression. ϪRT denotes a reaction where no reverse transcriptase was added during the cDNA synthesis step (cDNA from untreated control cells was used). The products were analyzed by electrophoresis on a 2% (w/v) agarose gel along with molecular size markers (M). The results shown are representative of two to three independent experiments. phages. For this, the cells were treated with IFN-␥ for 2 h in the absence or the presence of two different agents that raise intracellular Ca 2ϩ levels: thapsigargin, which releases endoplasmic reticulum-stored Ca 2ϩ ; or the ionophore A23187, which passively transfers extracellular calcium as well as triggering release from intracellular stores. Total RNA was then isolated and subjected to RT-PCR with ICERI/I␥-and ␤-actin-specific primers. Fig. 9  (panels A and B) shows that, instead of inhibiting the IFN-␥mediated ICER mRNA expression, the presence of A23187 or thapsigargin enhanced it in a synergistic manner. This was particularly apparent in cells treated with IFN-␥ and A23187, where the increase in ICER mRNA expression in the presence of both factors was approximately 2-fold greater than that expected from a simple additive action. These studies, therefore, reveal a novel synergistic action of IFN-␥ and Ca 2ϩ in the activation of ICER mRNA expression in macrophages.
We next investigated whether the synergism between IFN-␥ and intracellular Ca 2ϩ elevating agents seen with ICER mRNA expression was also manifested at the level of ICER-CRE DNA binding. Because of a longer incubation period required in such experiments as a result of the noted delay between ICER mRNA and protein expression (e.g. Figs. 2A and 3B), such studies could only be carried out for IFN-␥ and thapsigargin (A23187 was found to be toxic to the cells at 12 h). As shown in Fig. 9C, the signal from the fastest migrating ICER-CRE-binding complex, which was inhibited by the inclusion of the anti-CREM-1 antiserum, was substantially greater when extracts from cells treated with both IFN-␥ and thapsigargin were used compared with those obtained with either agent alone. In these experiments, we also examined whether the IFN-␥-mediated induction of ICER-CRE complex was inhibited by apigenin, as found at the level of ICER mRNA expression (Figs. 6F and 8A). As shown in Fig. 9C, this was indeed found to be the case. DISCUSSION The studies presented in this paper show for the first time several findings that are relevant to the regulation of macro- ϪRT and M denote a reaction where no reverse transcriptase was added during the cDNA synthesis step (cDNA from the control sample was used) and markers, respectively. B, the action of IFN-␥ and apigenin on CK2 activity of J774.2 macrophages. The cells were either untreated or exposed to IFN-␥ for 2 h in the absence or the presence of apigenin (40 M). CK2␣ was immunoprecipitated and subjected to in vitro kinase assay as described under "Experimental Procedures." Also shown are a positive control (ϩve) with purified CK2 enzyme and a negative control (-ve) without any enzyme. The signal from phosphorylated ␤-casein substrate (24 kDa) is shown. C, Western blot analysis of CK2-␣ from cells that were either untreated or stimulated with IFN-␥ for 2 h in the absence or the presence of apigenin. The position of the 42-kDa CK2-␣ polypeptide is shown. D, effect of apigenin on IFN-␥-mediated phosphorylation of CREB. Western blot analysis was carried out on extracts from cells that were either untreated or stimulated with IFN-␥ for 2 h, in the absence or the presence of apigenin (40 M), using the phospho-CREB antiserum as described under "Experimental Procedures." All the results are representative of two independent experiments. phage gene transcription: (i) ICER is expressed in macrophages; (ii) the levels of ICER mRNA and proteins in macrophages along with their binding to CRE are induced dramatically when the cells are stimulated with IFN-␥, and this is inhibited by the CK2 inhibitor apigenin; (iii) this cytokine stimulates CK2 activity and increases the level of activated CREB phosphorylated at Ser-133; and (iv) IFN-␥ synergizes with Ca 2ϩ in the induction of ICER expression. These results, therefore, reveal the existence of a potentially novel pathway for the IFN-␥-mediated suppression of macrophage gene transcription that involves increased activation of CK2 and CREB leading to transcriptional stimulation from the CREM P 2 promoter, thereby culminating in increased ICER production.
IFN-␥ has been shown to suppress the expression of numerous genes; for example, analysis of the published literature shows at least 30 genes for which expression is suppressed by the cytokine (LPL; very low density lipoprotein receptor; scavenger receptor A; CD136; low density lipoprotein receptor-re-lated protein; C/EBP␣; matrix metalloproteinase 2, 9, and 13; cathepsin K; histidine decarboxylase; neu/Her-2; collagenase-1; angiotensin AT1a receptor; cyclin A; CXCR4 (fusin); ␣1(I)and ␣1(III)-procollagen; IL-4 receptor; stromelysin; chemokine receptor CCR2; long pentraxin PTX3; fibronectin; ␤-amyloid precursor; renin; IL-8; epithelial neutrophil activating protein 78; thyrotrophin receptor; IL-1␤; and c-Myc) (12, 16, 23-27, 68 -89). The expression of some of these genes has been investigated in monocytes/macrophages (12, 16, 23-27, 77, 81, 82, 86, 88). In addition, another 12 IFN-␥-inhibited genes have been identified recently from a limited expression profiling on the human fibrosarcoma HT1080 cell line treated for 6 h with the cytokine (90). However, studies aimed at delineating the mechanisms through which IFN-␥ suppresses gene expression have been limited and restricted to a few genes: (i) matrix metalloproteinase-9 and -13 (does not occur in the absence of STAT-1) (70, 71); (ii) stromelysin (requires an AP1 binding site) (80); (iii) scavenger receptor A (mediated by competition between STAT1 and AP1/ets transcription factors for a limiting amount of shared co-activator) (25); (iv) ␣1(I) procollagen (requires a region containing overlapping binding sites for Sp1, Sp3, and nuclear factor-1) (78); and LPL (our recent studies showing a novel role for Sp1 and Sp3) (12). Thus, disparate mechanisms have so far been identified for the IFN-␥-mediated regulation of such genes. However, it is possible that regulation through ICER may turn out to be a more common mechanism, given that it is a potent repressor that acts at substoichiometric levels (1)(2)(3). Detailed characterization of the promoter regions of a large number of genes for which transcription is inhibited by IFN-␥ in the future will clarify this issue. It is also possible that, besides acting as a direct repressor of genes that are expressed constitutively at high levels and inhibited by IFN-␥, ICER may also play an important role in the transcriptional suppression of immediate early genes that have been induced transiently by this cytokine through other pathways. For example, the NGF-mediated expression of ICER has been proposed to be involved in the inhibition of c-fos gene transcription once it has been induced immediately by the mediator (9).
It is interesting to note that there is a differential expression of the different ICER transcripts in macrophages (e.g. Fig. 1B). Such differential expression has been noted previously (39,51,91), but its significance remains unclear as all four isoforms are thought to function identically as potent transcriptional repressors (1-3). The ICER protein levels do not peak until 12-24 h, at least 6 h after the peak for ICER I mRNA expression (Figs.  1-3). Only a few previous studies on ICER gene expression have analyzed both mRNA and protein expression (3,7,9). In most of these studies, a close correlation was observed between mRNA and protein expression except for forskolin-treated AtT20 cells (3) and, as shown in this study, in macrophages. However, such an expression profile correlates well with the kinetics of IFN-␥-inhibited genes, such as LPL (12), with maximum decreases seen between 12 and 24 h. Several extracellular mediators have been shown to increase the intracellular ICER levels, including NGF, thyroid-stimulating factor, follicle-stimulating hormone, glucagons, noradrenaline, gastrin, and cholecystokinin (7-10, 48, 51, 92). Although most of these factors act by increasing intracellular cAMP levels, the use of pharmacological agents has shown that the action of NGF and gastrin was not mediated through this route (9,10). Similarly, the use of a range of pharmacological agents showed that the IFN-␥-induced ICER expression was not affected by inhibitors of cyclic nucleotide-dependent protein kinase, MAP kinases, Ca 2ϩ /calmodulin-dependent protein kinase, protein kinase C, and JAK2 but required activation of CK2 (Fig. 6). The lack of the effect of the JAK2 inhibitor AG490 Total RNA was isolated and subjected to RT-PCR using primers against ICER-I/I␥ and ␤-actin as described under "Experimental Procedures." ϪRT denotes a reaction where no reverse transcriptase was added during the cDNA synthesis step (cDNA from untreated control cells was used). C, EMSA was carried out using extracts from J774.2 macrophages that were either left untreated or exposed for 12 h to IFN-␥, thapsigargin, or apigenin in the absence or presence of anti-CREM-1 antiserum (anti-CREM-1) or nonimmune serum (NIS), as indicated. The vertical line indicates the position of the fastest migrating ICER-CRE (ICER) complex. All the results are representative of two independent experiments. may be of interest in light of several recent studies indicating the potential existence of JAK-STAT-independent pathways in IFN-␥ action (20 -22). For example, although IFN-␥ induces major histocompatibility complex class II and intracellular adhesion molecule-I (ICAM-I) gene expression in human corneal epithelial cells, incubation of the cells with a global protein tyrosine kinase inhibitor prior to IFN-␥ prevents the expression of major histocompatibility complex class II but not ICAM-1, thus inferring that ICAM-1 transcription is activated in this instance through a JAK-independent pathway (93).
The importance of CREB phosphorylation at Ser-133 for the initiation of transcription from the CREM P 2 promoter is well documented (1)(2)(3). It should, however, be noted that CREB phosphorylation does not necessarily constrain a cell to ICER production (9). For example, treatment of PC12 cells with either epidermal growth factor or NGF leads to increased Ser-133 phosphorylation of CREB, but only NGF treatment leads to an increase in ICER expression (9). As IFN-␥ was shown to modulate ICER mRNA expression, we also decided to investigate whether there was a concominant CREB phosphorylation. Indeed, an increased level of Ser-133-phosphorylated CREB was found when the cells were incubated with IFN-␥ and this was inhibited in the presence of apigenin (Figs. 5 and 8). However, CK2 is not currently thought to phosphorylate CREB at Ser-133 but instead at five closely spaced sites (Ser-108, Ser-111, Ser-114, Ser-117, and Thr-119) (1,2,94). Nevertheless, given the general paucity of research in this particular area, it is entirely possible that this situation does not occur in macrophages, and that CK2 can indeed directly phosphorylate CREB at Ser-133. Alternatively, because of co-operation between the numerous consensus protein kinase sites that exist in the "phosphorylation P-box" domain of CREB (1, 2), it is possible that a CK2-mediated phosphorylation may initiate a processive phosphorylation cascade, catalyzed by kinases other than CK2, that potentially results in Ser-133 phosphorylation. Although the precise "CREB kinase" responsible for Ser-133 phosphorylation remains to be identified, classical kinases, such as cAMP-dependent protein kinase, protein kinase C , and Ca 2ϩ -calmodulin-dependent protein kinases, can be excluded (Fig. 6). In this respect, it is interesting to note that basic fibroblast growth factor has recently been shown to activate a novel 120-kDa CREB kinase during differentiation of immortalized hippocampal cells, which then triggers the phosphorylation of Ser-133 in CREB (46).
Another interesting finding to emerge from studies presented in this manuscript is the synergism between intracellular Ca 2ϩ elevating agents and IFN-␥ (Fig. 9). A synergism between Ca 2ϩ and the ICER inducer cAMP has been seen previously in the regulation of lysozyme gene expression in chicken myelomonocytic HD11 cell line, c-fos in the corticotroph cell line AtT20, and apoptosis of WEH17.2 cells (53,54,57). Interestingly, the synergism in relation to the apoptosis of WEH17.2 cells and the induction of c-fos expression has been demonstrated to be the result of the convergence of the protein kinase A and Ca 2ϩ signaling pathways, via activation of calcium-calmodulin kinase IV, at CREB phosphorylation on Ser-133 (54,57). Extrapolating from these findings, it is thus likely that the synergism between IFN-␥ and Ca 2ϩ is the result of increased CREB phosphorylation modulated by the convergence of IFN-␥-induced CK2 activity and thapsigargin/A23187induced calcium-calmodulin kinase activity.
In conclusion, we have identified a novel induction of ICER expression in macrophages by IFN-␥ that is mediated through the activation of CK2 and CREB. Future studies should seek to identify all the components from the interaction of IFN-␥ to its cell-surface receptors to the activation of ICER gene transcrip-tion, and the potential role of this potent inhibitor in the suppression of downstream genes by this cytokine.