J Biol Chem, Vol. 275, Issue 17, 12626-12632, April 28, 2000

CCAAT/Enhancer-binding Protein- Regulates Interferon-induced Transcription through a Novel Element^*

Sanjit K. Roy, S. James Wachira^§, Xiao Weihua^¶, Junbo Hu, and Dhananjaya V. Kalvakolanu

From the Marlene and Stewart Greenebaum Cancer Center, Department of Microbiology and Immunology, Molecular and Cellular Biology Program, University of Maryland School of Medicine, Baltimore, Maryland 21201

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

We have described previously a novel interferon (IFN)-responsive cis-acting enhancer element called -IFN-activated transcriptional element (GATE). GATE is distinct from the known IFN-stimulated elements and binds to novel transacting factors. To identify the -IFN-responsive transacting factors that interact with GATE, we have screened a cDNA expression library derived from IFN--stimulated murine macrophage cell line and isolated three different cDNAs. Among these is a gene coding for the pleiotropic transcription factor, CCAAT/enhancer-binding protein- (C/EBP-). We report here that the gene for C/EBP- binds to GATE and induces gene expression. A mutant C/EBP- interferes with the IFN--stimulated transcription of the ISGF3 (p48) promoter. Other members of the C/EBP family do not cause these effects. Interestingly, the expression of C/EBP-, not the other members of its family, is induced by IFN-. These studies thus identify a novel role for C/EBP- in the IFN-signaling pathways.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Members of the interferon (IFN)¹ family of cytokines regulate antiviral, antitumor, and immune responses in the vertebrates, inducing a number of cellular IFN-stimulated genes (1). IFN effects on the cells are, in part, may also be due to the repression of certain other genes (2). ISG induction has been largely due to activation of the well established Janus tyrosine kinase (JAK)-signal-transducing activators of transcription (STAT) pathway, wherein tyrosine-phosphorylated dual function factors (i.e. STATs) directly regulate the expression of down-stream genes (3, 4). In the IFN-/ induced response, STAT1 and STAT2, after their activation by JAK1 and TYK2, associate with ISGF3 (p48 or IRF-9), a member of the IFN-gene regulatory factor (IRF) family (4, 5). The resultant multimeric complex, ISGF3, migrates to nucleus prior to its binding to the IFN-stimulated response element (ISRE) and stimulation of target gene transcription (3-5). In a number of tumor- and viral oncogene-expressing cell lines, down-regulation of the physical levels or inactivation of the components of ISGF3 serves as a mechanism for evading the action of IFNs (6, 7). Thus, ISGF3 is central to the IFN response.

Although IFN- can alone induce ISGs, pretreatment with IFN- causes a robust induction of these and consequent biological response (8-10). This latter effect is achieved through an enhancement of the levels of ISGF3 (9-11). ISGF3 is an IFN-regulated gene, like certain other members of the IRF family (11). However, it is a slowly induced gene and requires new protein synthesis, unlike other IRFs (12). We have demonstrated earlier that GATE, a novel IFN- response element, and its cognate transacting factors regulate the murine p48 promoter (12). We have now identified the GATE binding factors (GBF) from a cDNA library prepared from an IFN--stimulated macrophage cell line. Here we show that one of the GBFs is transcription factor C/EBP-, a member of the CCAAT/enhancer-binding protein family (13). It regulates GATE-dependent gene expression, in an IFN-dependent manner. We also show that C/EBP- (NF-IL6, LAP, NF-M, IL6-DBP) is an IFN-stimulated gene. This factor has been previously shown to regulate the type I acute phase-responsive genes, under the control of another cytokine, IL-6 (14, 15). Our studies for the first time demonstrate a role for C/EBP- in the IFN signal transduction pathway and uncover a novel mechanism of IFN action.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Reagents-- Recombinant murine IFN- (Roche Molecular Biochemicals), IPTG and 5-bromo-4-chloro-3-indolyl -D-galactopyranoside (X-gal) (Life Technologies, Inc.), cDNA synthesis kits (Stratagene), mouse liver cDNA library (CLONTECH), restriction and modifying enzymes (New England Biolabs), and nitrocellulose membranes (Schleicher & Schuell) were used in these studies. Rabbit polyclonal antibodies specific for C/EBP-, C/EBP-, and actin were purchased from Santa Cruz Biotechnology. Murine ISGF3 promoter and its mutants were described previously (12).

Cell Culture and Plasmids-- Murine macrophage cell line RAW (RAW264.7) was grown in RPMI 1640 supplemented with 10% fetal bovine serum (12). Mammalian expression vectors for C/EBP- and C/EBP- were provided by Richard Hanson (Case Western Reserve University, Cleveland, OH) C/EBP- expression vector was a gift from Peter Johnson (National Cancer Institute, Frederick Cancer Research Facility, Frederick, MD). Mutant C/EBP- consisting of the DNA binding was generated by subcloning the region corresponding to the C-terminal 139 amino acids in the pcDNA 3.1 vector. Wild type cDNA (open reading frame) was also cloned into pcDNA3.1 and pCXN2 similarly. Both these constructs gave a comparable induction of the reporter genes in transient transfection (data not shown).

Gene Expression Analyses-- Southern, Northern, and Western blot analyses, transfection, -galactosidase and luciferase assays, electrophoretic mobility shift assays (EMSA), SDS-polyacrylamide gel electrophoresis, and sequence analysis were performed as described in our earlier publications (12). In vitro transcription and translation were performed using commercially available RiboMax system (Promega Inc.).

cDNA Libraries-- Poly(A)⁺ mRNA from RAW cells stimulated with murine IFN- (400 units/ml) for 0, 4, 8, and 12 h was pooled and cDNA was prepared using a commercially available kit. The cDNAs were cloned into -ZAPII vector between EcoRI and XhoI sites (Stratagene Inc.). The resultant library was packed in vitro and was used to infect Escherichia coli to obtain the final library. This library, consisting of more than 99% recombinants, was used for screening the proteins that bind to GATE. Induction of protein encoded by the cDNA is achieved by IPTG treatment. Three million plaques were screened using a ³²P-labeled, concatamerized GATE as described previously (16). Positive clones identified in the first round were subjected to two more rounds of screening with wild type (5'-CCCGAGGAGAATTGAAACTTAGGG-3') and mutant GATE probes (5'-CCCGAGGAGAATTGCTCGGCGAGGG-3'). The bases in the GATE sequence were mutated (shown in italics and underlined), primarily on the basis of our earlier observation that this sequence exhibited a partial homology (12) to the IFN-stimulated response element (16). In particular, AAACTT residues were altered. In each case a corresponding complimentary oligonucleotide was synthesized and annealed. These double-stranded oligonucleotides were concatamerized, labeled with ³²P, and used in the experiments. At the end, 13 independent phage clones expressing GATE-binding proteins were isolated. These were grouped into three, based on Southern blot analyses of the rescued inserts and partial sequence analysis. Among these, five clones expressing various sizes of the same cDNA were grouped as GBF-2. The others are being characterized currently. Inserts in the phage were rescued by in vivo excision of the inserts, which allowed the transfer of cDNA into pBluescript phagemid. Inserts were sequenced and used for further analysis. A commercially available mouse liver cDNA library (CLONTECH) was screened further to obtain full-length inserts.

Bacterial Expression-- Two different constructs, C/EBP-S and C/EBP-L, each expressing different sized products, were generated by polymerase chain reaction amplification (14 cycles) and subcloned into bacterial expression vector pET32A. For amplifying C/EBP-L, the forward and reverse primers were: 5'-TAGAATTCTACGGTTACGTGAGCCTC-3' and 5'-TTAAGCTTCTAGCAGTGGC CCGCCGAG-3', respectively. For amplifying C/EBP-S, the forward primer is 5'-TAGAATTCTTCGCCCTGCGCGCCTAC-3' and the reverse primer is same as described above. EcoRI and HindIII restriction sites (italicized and underlined) were included in these oligonucleotides to permit the cloning of the amplified inserts. C/EBP-S and C/EBP-L contained the C-terminal 139 and 191 amino acids, respectively. Proteins were induced with IPTG treatment, purified on nickel-nitrilotriacetic acid-agarose (Novagen) and analyzed by SDS-polyacrylamide gel electrophoresis. Where indicated, Western blotting was performed to detect the recombinant proteins.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Isolation of GATE-binding Factors-- To identify the GATE binding factors, we have screened (16) a -ZAPII cDNA expression library, prepared from an IFN--stimulated murine macrophage cell line RAW, using labeled wild type GATE as a probe (see "Materials and Methods"). Protein encoded by the cloned cDNA was induced by IPTG treatment. Phage plaques were overlaid with nitrocellulose membranes and the membranes were probed with a ³²P-labeled, concatamerized GATE. Clones identified, using the wild type GATE probe, in the first round were subjected to additional rounds of screening. Replica membranes from the same plate were probed with mutant or wild type GATE, separately. GATE is partially homologous (12) to ISRE. Mutant GATE was designed by altering residues that were homologous to ISRE. In particular the AAACTT residues at the 3' end of the sequence were changed to CTCGGC, because the AAA or TTT residues of ISREs are essential for IFN response (3, 4). Plaques that lit with wild type GATE were unable to interact with mutant GATE, indicating the specificity of the binding (Fig. 1). Positive phage clones were identified and purified. This approach yielded three distinct groups of clones (as assessed by Southern analysis and partial sequencing), which encoded proteins capable of binding to GATE. We named these proteins GBFs to reflect their function as GATE binding factors. From these, we have chosen GBF-2 for further characterization, for reasons described below.

View larger version (53K): [in this window] [in a new window]   Fig. 1.   Identification and isolation of a GBF-2. A representative phage clone (clone 3) corresponding to GBF-2 was plated on E. coli XL-1 blue MRF' host. Protein expression was induced upon overlaying IPTG (10 mM) impregnated nitrocellulose membranes. Duplicate membranes from the same plate were incubated with a concatamerized double-stranded DNA corresponding to mutant (Mut) or wild type GATE (see "Materials and Methods") in a binding buffer (50 mM Tris-HCl, pH 7.5, 50 mM KCl, 2 mM EDTA, 1 mM dithiothreitol, and 5% nonfat dry milk powder) at 4 °C for 12 h. The probes were labeled with 32P using a nick translation kit to a comparable specific activity (~109 cpm/µg). The filters were washed three times at room temperature with the binding buffer containing 0.25% nonfat dry milk powder, dried, and exposed to x-ray films to detect the clones expressing GATE-binding protein (16). Phages were plated at ~250 and 100 plaque-forming units in the second and third rounds, respectively. Note the enrichment of GATE binding clones in the third round. Only half of the filter is shown.

Identification of GBF-2 as C/EBP--- Five independent GBF-2 phage clones carrying different lengths of the same cDNA were isolated from the screen, all of which were capable of encoding GATE-binding proteins. A full-length cDNA was obtained by further screening of a mouse liver cDNA library. Sequence analysis of GBF-2 revealed that it was identical to a known transcription factor, C/EBP-. No other member of C/EBP family was found in the other GBF clones. The longest of these cDNAs was used as probe to isolate the corresponding full-length cDNA from a mouse cDNA library. In vitro translation of this cDNA yielded a protein with an apparent molecular mass (~35 kDa) similar to that of C/EBP- (Fig. 2A).

View larger version (52K): [in this window] [in a new window]   Fig. 2.   Expression and identification of GBF-2 as C/EBP-. A, in vitro translation of full length GBF-2 and the cloning vector pGEM7zf. Plasmid DNA (3 µg) linearized with XhoI was used as a template to generate RNA using T7 RNA polymerase. The resultant RNA was programmed into nuclease-treated rabbit reticulocyte lysates (Promega Inc.), and translation was carried out for 2 h at 30 °C in the presence of [35S]methionine (100 µCi, Amersham Pharmacia Biotech). Reaction components (20 µl) were separated on a 10% SDS-polyacrylamide gel and visualized after fluorography. Positions of the molecular mass standards (kDa) were indicated on the right. B, translation reactions were conducted with C/EBP RNA as described in panel A in the presence of [35S]methionine. Subsequently, the reaction products were immunoprecipitated with the indicated antibodies (3 µl) at 4 °C for 4 h. Protein G-agarose (200 µl) was added and incubated for additional 1 h at room temperature. Immunoprecipitates were recovered by centrifugation, washed with thrice with a buffer (0.5 M NaCl, 10 mM Tris, pH = 8.0, 0.1% Nonidet P-40) and separated on a 10% SDS-polyacrylamide gel. Proteins were visualized by fluorography.

To further prove that GBF-2 was C/EBP-, the rabbit reticulocyte translation reaction, programmed with in vitro transcribed RNA of GBF-2 cDNA, was subjected to immunoprecipitation using C/EBP-- and C/EBP--specific antibodies. A C/EBP--specific antibody specifically immunoprecipitated the protein encoded by GBF-2 (Fig. 2B). C/EBP--specific antibodies did not immunoprecipitate the protein. These data show that GBF-2 is C/EBP-. Hereafter, the term C/EBP- will be used to denote GBF-2.

C/EBP- Induces GATE-dependent Gene Expression-- Since C/EBP- cDNA has been isolated by virtue of its interaction with GATE, it is necessary to determine whether it is a transcriptional activator or repressor. Therefore, we have subcloned the C/EBP- cDNA under the control of a constitutive enhancer in mammalian expression vector pCXN2, and determined if it induces the luciferase reporter. A construct P4, driven by a 74-base pair enhancer element consisting of GATE region of p48, was promoter placed upstream of a SV40 promoter used in these studies. C/EBP- expression vector but not the control vector induced luciferase expression in a dose-dependent manner. Although at a higher molar ratio C/EBP- slightly inhibited the gene expression, luciferase activity was significantly more than the vector alone at that dose (Fig. 3A). This inhibitory effect may be due to a competition for limited amounts of general transcriptional co-activators such as histone acetylases available in the cell. A similar construct that lacked the GATE did not respond to C/EBP- (Fig. 3B). The effect of GBF-2 was also determined in the context of wild type promoter. Luciferase gene expression in the wild type as well as a mutant that bore a myc-stimulated element (MSE pm) was induced by GBF-2 (Fig. 3B). In contrast, the GATE mutant (GATE pm) was not induced under these conditions. GBF-2 did not induce the transcription of other mutant promoters that lacked GATE. Furthermore, luciferase expression from a reporter plasmid, carrying a wild type GATE (GATE-W) but not a mutant GATE (GATE-Mu), was induced by C/EBP-.

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Induction of gene expression by C/EBP-. Transfection was carried out using LipofectAMINE reagent (Life Technologies, Inc.) and normalized to an internal control, the -actin--galactosidase reporter gene (12). A, expression vector (pCXN2) or its derivative expressing C/EBP- were transfected along with the P4-luciferase construct (0.4 µg) into RAW cells and expression of luciferase was measured using 30 µg of total cell protein, 30 h after transfection. Numbers on the x axis refer to number of molecules of the expression vector or C/EBP- vector relative to that of P4 reporter. B, GATE-dependent induction of gene expression by C/EBP-. Cells were transfected with the indicated reporter genes (0.4 µg) and 0.1 µg of C/EBP expression vector. -Fold induction of luciferase gene was calculated, relative to that of vector (pCXN2) transfected in each case. Data represent mean of triplicate measurements. Construction of reporters was described elsewhere (12). A6, wild type p48 promoter; MSE pm, mutant of myc-stimulated element; GATE pm, mutant of GATE; P4 and P3 contain the p48 promoter sequences (thick black bar), placed upstream of a heterologous promoter (SV40). GATE-W and GATE-Mu (see "Materials and Methods") contain wild type and mutant GATE upstream of SV40 promoter. Cells were treated with IFN- (200 units/ml), where indicated. C, mutant C/EBP- inhibits IFN--inducible expression. Cells were transfected with expression vector (pcDNA3.1) or wild type C/EBP- or mutant C/EBP- (expresses the C-terminal 139 amino acids), along with P4 reporter gene and luciferase activity was measured. Each bar represents mean ± S.E. of triplicate measurements.

Because GATE was an IFN--inducible element and C/EBP- induced GATE-dependent gene expression, we next determined the influence of IFN- on GATE-dependent gene expression (Fig. 3B). RAW cells were transfected with various reporter genes along with C/EBP- expression vector and were treated with IFN-. Although C/EBP- alone was capable of inducing luciferase expression, IFN- treatment of cells further enhanced it strongly. These data may indicate that IFN--induced post-translational modifications such as phosphorylation play a role in the transcriptional regulation. Only those reporters that possessed a wild type GATE synergistically responded to IFN- (Fig. 3B). These results, combined with the binding data of Fig. 1, indicate that C/EBP- mediates both basal and inducible expression of ISGF3 promoter-dependent genes.

Since wild type C/EBP- induced gene expression, a mutant C/EBP- should fail to active GATE-dependent gene expression. To examine this, the P4 reporter was cotransfected with a mutant C/EBP- that expressed only the C-terminal DNA binding domain (139 amino acids) and luciferase activity was determined. Indeed, wild type C/EBP-, but not a mutant, induced gene expression (Fig. 3C). IFN- highly stimulated the expression of p48 in the presence of wild type C/EBP-. The mutant suppressed the inducible expression in a dominant manner.

C/EBP- but Not Other Members of Its Family Induce GATE-dependent Gene Expression-- To further demonstrate the specificity of C/EBP- in regulating GATE-dependent gene expression, we have tested whether two other members of its family, C/EBP- and C/EBP- also induced gene expression. In these studies, the P4 reporter plasmid was co-transfected with C/EBP- and C/EBP- expression vectors and measured the luciferase activity. Both C/EBP- and C/EBP- did not significantly induce GATE-dependent gene expression, compared with the vector-transfected cells (Fig. 4). C/EBP-, as expected, caused the highest increase in luciferase gene expression. IFN-, as shown in Fig. 3, strongly induced the gene expression. In contrast C/EBP- suppressed the IFN--inducible expression (compare it to vector-transfected samples). These data show that C/EBP- is a specifically induced GATE-driven gene expression.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Effect of various members of the C/EBP family on gene induction. Cells were transfected with pMSV expression vector or the same vectors carrying C/EBP-, C/EBP-, and C/EBP- cDNAs (0.25× molar concentration with respect to the reporter), along with the P4 construct (0.4 µg) and luciferase activity was measured as described above.

C/EBP- Binds to GATE-- Although C/EBP- was isolated by its ability to bind to GATE in a Southwestern type of screening, it was important to determine whether full-length GBF bound to GATE. For conducting these experiments, an in vitro translated C/EBP- protein was employed (Fig. 2A). A control reaction programmed with vector-derived RNAs was also transcribed and translated similarly. Protein translation was performed in the presence of unlabeled methionine. The reaction components were then incubated with ³²P-labeled GATE and EMSA was performed to detect DNA binding. Reticulocyte lysates expressing C/EBP- protein, but not the controls, formed a band with GATE in EMSA (Fig. 5A). A similar experiment was conducted with a labeled mutant GATE (see "Materials and Methods") (Fig. 5B). Neither the lysates programmed with C/EBP- RNA nor those programmed with vector-derived RNA formed a complex with the mutant GATE. These data thus indicate that C/EBP- specifically binds to GATE.

View larger version (44K): [in this window] [in a new window]   Fig. 5.   C/EBP- bonds to GATE. In vitro translation was carried out essentially same as in Fig. 2, except that unlabeled methionine was used. Binding of in vitro translated full-length C/EBP- to GATE. Ten µl of reticulocyte lysate was incubated with 32P-labeled GATE (65,000 cpm), and EMSA was performed (A). Vector indicates reticulocyte lysates programmed with pGEM7Zf-derived RNA. C/EBP- reticulocyte lysate translated in the presence of C/EBP--specific mRNA expressed from the same vector. B, an EMSA with a labeled mutant GATE (80,000 cpm). None, no lysate. Other notations are similar to panel A. C, C/EBP- protein in complex with wild type GATE. EMSA was performed as in panel A. C/EBP-, reticulocyte translation reaction (20 µl) programmed with C/EBP- RNA; None: probe alone; ab', antibody specific to C/EBP-; C/EBP- + ab', same as C/EBP- lane, except C/EBP--specific antibody was included reaction. Arrow indicates the supershifted complex. D, comparison of wild type and mutant GATE sequences with the consensus C/EBP- binding site (C/EBP Cons). A line between the nucleotides indicates homology. N can be any base. Note the closer homology between wild type GATE and the consensus C/EBP- binding site.

The formation of C/EBP-·GATE complex was confirmed by using antibodies specific to C/EBP- (Fig. 5C). As expected, the reticulocyte lysate containing C/EBP- formed a complex with GATE. Incubation of this complex with an antibody specific to C/EBP- caused a "supershifting" of the complex (shown with an arrow). Antibody alone did not form any detectable complexes. Thus, full-length C/EBP- was capable of binding to GATE. The sequence relationships between C/EBP- consensus site and wild type GATE are shown in Fig. 5D. This sequence has only a partial homology to the consensus C/EBP- binding site. Mutant GATE used in these studies lacks the residues necessary for C/EBP- binding in the right half. Thus, the AAACTT nucleotides of the GATE are necessary for C/EBP- binding to GATE. Our preliminary DNase I footprinting data are consistent with this observation (data not shown).

To demonstrate that the C terminus of C/EBP- without the transactivatng domain was sufficient to bind GATE, two different C/EBP- inserts, C/EBP-S (139 amino acids) and C/EBP-L (191 amino acids), were expressed as histidine-tagged fusion proteins in E. coli. The resultant recombinant proteins were purified (Fig. 6A) and were analyzed by Western blot using C/EBP--specific antibodies (Fig. 6B) to confirm the expression of expected protein. A control sample containing Tag alone was used in these experiments (lane 1). This antibody specifically detected both fusion proteins, which contained various lengths of C/EBP- (lanes 2 and 3), but not the Tag protein (Fig. 6B). These proteins were then incubated with a ³²P-labeled GATE in separate reactions and EMSA was performed. Recombinant C/EBP-S and C/EBP-L protein, but not the Tag, bound to GATE (Fig. 6C). Incubation of these reactions with C/EBP--specific antibodies, but not a control antibody, caused a supershifting of the complex (Fig. 6D).

View larger version (32K): [in this window] [in a new window]   Fig. 6.   The C terminus of C/EBP- is sufficient for GATE binding. A, silver staining of bacterially expressed C/EBP-, separated on 12% SDS-polyacrylamide gel electrophoresis. C/EBP-S (139 amino acids) and C/EBP-L (191 amino acids) represent two different His-tagged fusion proteins representing the C terminus of the C/EBP- cDNA. Tag indicates the purified His tag protein from pET32A-expressing cells. Protein concentrations are 500, 200, and 500 ng/lane in lanes 1, 2, and 3, respectively. B, Western blot of proteins shown in panel A, using rabbit polyclonal antibodies specific for C/EBP- (1/5000 dilution). Open and filled arrowheads indicate the positions of C/EBP- proteins. The C terminus of C/EBP- is sufficient for GATE binding. C, 200 ng of C/EBP-S or C/EBP-L or Tag protein (Fig. 2B) was incubated with a 32P-labeled GATE (70,000 cpm) and EMSA was performed. None, probe alone. B, super shifting of complexes by C/EBP--specific antibodies. EMSA was conducted as in panel A with GBF-2L protein. Where indicated, 2 µl of specific antibody was added to the reactions and incubated for 10 min, prior to EMSA. Minus () and plus (+) indicate deletion addition of the reagent in the reaction. Control antibody was normal rabbit serum.

GBF-2 (C/EBP-) Is an IFN-stimulated Gene-- Given the fact that C/EBP- induced IFN- stimulated gene expression and GATE binding factors are synthesized in response to IFN- (12), we next determined whether IFN- induced the expression of C/EBP-. RAW cells were treated with IFN-, and the expression of C/EBP- mRNA was monitored by Northern blotting. As shown in Fig. 6A, the mRNA of C/EBP- was strongly induced by IFN-. Under these conditions, there was no change in the expression of -actin mRNA. These blots were also probed with C/EBP- probe to detect the changes in its expression. No change in the levels of C/EBP- occurred with IFN- treatment (data not shown and see below). To provide further evidence that GBF-2 (C/EBP-) is an IFN--inducible gene, we have determined its induction in vivo. BALB/c mice were injected with IFN- (50,000 units) via tail vein, and total RNA from various tissues was extracted. These RNAs were Northern blotted and the blots probed with ³²P-labeled GBF-2 cDNA. As shown in Fig. 6B, GBF-2 mRNA was readily induced by IFN- treatment while a basal level of its expression was seen in most tissues.

We next determined whether induction of C/EBP- mRNA also caused a corresponding increase in its protein levels. Western blot analysis of IFN--stimulated RAW cell extracts was performed to detect changes in C/EBP- protein. Indeed, IFN- caused a time-dependent increase in the levels of C/EBP- protein (Fig. 7C). Presence of comparable amount of protein in these lanes was ensured with probing of these blots with an antibody raised against actin (Fig. 6C, lower panel). A similar analysis of these samples with C/EBP--specific antibodies did not suggest an evidence for its induction (Fig. 7D). C/EBP- protein was also induced in the human fibrosarcoma cell line 2fTGH and mouse kidney and spleen and tissues after IFN- treatment (data not shown).

View larger version (37K): [in this window] [in a new window]   Fig. 7.   IFN--induced expression of C/EBP-. Panels A and B, Northern blot analysis of total RNAs (30 µg) from RAW cells stimulated with IFN- (200 units/ml) (panel A) or the indicated mouse tissues, using C/EBP- probe. Numbers above panel A indicate the length of IFN treatment. Panel B, in vivo induction of C/EBP- mRNA in BALB/c mice. Where indicated with a plus (+) sign, mice were treated with 50,000 units/ml IFN- for 12 h by tail vein injection. A minus () sign indicates saline treatment. Each blot has been reprobed with -actin to ensure the presence of a comparable amount of RNA in the lanes. Panels C and D, Western blot analysis of IFN--treated RAW cell extracts (150 µg) with indicated antibodies. Numbers above the lanes indicate the hours after IFN- treatment.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Immunoregulatory cytokine IFN- is critical for mediating a wide array of biological responses in the vertebrates, including antiviral and antitumor functions (17, 18). A number of ISGs have been identified using a variety of techniques (1, 2). The IFN--stimulated genes are induced with variable kinetics unlike those stimulated by IFN-/ (1). All IFN responses are attributed to the activation of JAK-STAT signaling pathway, a rapidly activated and deactivated process (3, 4). STAT activation often lasts for no longer than 1 h after the ligand engagement with receptor, despite the presence of IFN in the extracellular environment (3, 4). Although the activation of STATs and subsequent expression of certain ISGs is independent of de novo RNA or protein synthesis, several IFN--regulated genes require the synthesis of other protein factors (1, 2). Furthermore, the induction of certain IFN-stimulated genes does not correspond to the STAT activation and deactivation kinetics (1, 2). Thus, it appears that transcription factors other than STATs regulate the expression of these "late induced genes." Indeed a number of factors, such as hXBP1, RF-X, interferon gene regulatory factor-1 (IRF-1), class II transactivator, and IFN consensus sequence-binding protein represent the secondary mediators of signals emanated from the initial burst of STAT activation (19-22). However, these factors can only partially account for the pleiotropic nature of IFN response, and additional undefined IFN-induced regulators may exist. That said, IRFs also regulate gene expression independent of STAT proteins (20).

We have shown previously that the induction of murine ISGF3 (p48) gene, an IFN- induced component of the ISGF3 complex, occurs at a slower rate compared with other ISGs (12). Our studies have also identified a novel IFN- response element, GATE, which is preferentially regulated by IFN- (12). To define the transacting factors that interact with GATE, we have screened a cDNA expression library derived from a mouse macrophage cell line, stimulated with IFN- (Fig. 1), and identified the candidate cDNAs. One of these is a known transcription factor, C/EBP-. In this study, we have shown that transcription factor C/EBP- acts as an IFN-regulated factor. The fact that five independent clones of C/EBP- bind to GATE, but not to a mutant GATE, indicates a specific interaction of C/EBP- with GATE (Figs. 1 and 5). Sequence and immunoprecipitation analyses (Fig. 2B) have established that GBF-2 is indeed C/EBP-. This library also included the cDNAs corresponding to C/EBP- (data not shown). However, they did not seem to bind to DNA. Consistent with this, C/EBP-a did not cause significant gene induction (Fig. 4). These observations indicate a specific binding of C/EBP- to GATE.

Although C/EBP- binding to wild type GATE, but not to mutant GATE, can itself provide evidence for its binding specificity, it is important to note that the protocol employs a concatamerized GATE, not a monomeric GATE. It cannot distinguish whether C/EBP- binds to GATE or junctions of tandem GATE sequences. However, EMSA with monomeric GATE provide a direct proof that full-length C/EBP- binds to GATE (Fig. 5). The C terminus of C/EBP- contains a b-ZIP domain and is sufficient for GATE binding (Fig. 6). Thus, interaction of C/EBP- with GATE involves the previously identified DNA binding domain (23). The antibody supershift experiments (Figs. 5C and 6D) proved that C/EBP- indeed forms a complex with GATE. C/EBP- consensus sequence TTNNGNAAT (14, 15) has a partial homology to GATE. Six of the nine nucleotides in this region of GATE match with consensus C/EBP- binding site. Mutation of the AAACTT nucleotides of the wild type GATE resulted in the loss of C/EBP- binding (Figs. 1 and 5). Thus, these nucleotides seem to be essential for C/EBP- binding to GATE. Preliminary DNase I footprinting analysis suggests that the GAAACTT region of the GATE is protected by C/EBP- (data not presented).

The specificity of C/EBP- in inducing GATE-driven gene expression was demonstrated using multiple approaches. 1) Reporter genes that contained a functional GATE were inducible by it (Fig. 3B). 2) A C/EBP- mutant that possessed only the DNA binding domain failed to induce gene expression (Fig. 5). 3) Other members of the C/EBP family did not induce IFN--stimulated gene expression (Fig. 4). C/EBP- represses gene expression by possibly interfering with the function of endogenous C/EBP- (Fig. 4). Such repression may occur via a heterodimerization between C/EBP- and C/EBP- (24). Enhancement of C/EBP--inducible gene expression by IFN- suggests the participation of a protein kinase in this pathway. This observation is consistent with our earlier studies, which showed that inhibitors of protein kinases block IFN--induced expression from GATE (12). Our recent studies show that inhibitors of the ERK pathway suppress IFN-induced gene expression (data not presented). Consistent with our original suggestion that GATE binding factors are distinct from ISRE-binding proteins, we have now identified the C/EBP-, a factor that has not been implicated in IFN response previously.

In addition to being a regulator of IFN-response, C/EBP- is also an IFN-inducible gene (Fig. 7). In RAW macrophage cell line, the mRNA and the protein levels of C/EBP- are induced in response to IFN-. Under these conditions C/EBP- protein levels have not changed, suggesting the specific effect of IFN-. The induction of C/EBP- gene is not a cell line-specific effect, because it is also induced in various tissues by IFN-, in vivo (Fig. 7B). IFN- failed to induce C/EBP- protein in mutant cell lines, which lacked STAT1 and JAK1 genes (data not presented). Analysis of the C/EBP- published promoter sequence reveals a putative GAS-like element, TTCCAGGGAA (25). The relevance of this sequence to IFN- response needs to be established. Based on these results, we suggest that IFN- induces the expression of C/EBP- gene, whose transcriptional activity is enhanced further by the posttranscriptional modifications stimulated by IFN-.

To our knowledge, this is the first report that suggests a novel role for C/EBP- in the IFN signal transduction pathway. Our data are also consistent with a previous report that identified C/EBP- as an IFN-inducible gene (2). We have now provided a direct evidence for C/EBP- in the IFN signaling pathway. Previously, this factor was implicated in the regulation of gluconeogenesis, acute phase responses induced by interleukin-6, tumor necrosis factor-, and lipopolysaccharide (14, 15). C/EBP- also mediates its effects through a cyclic AMP-responsive element and ATF binding sites (26-29). It also interacts with other transcription factors such as the retinoblastoma tumor suppressing protein and NF-B (30). Although C/EBP- induces IL-6 gene expression, its activity is repressed in the presence of wild type p53 (31). Indeed, IL-6 can cross-regulate p48 gene expression through C/EBP- (data not shown). The participation of C/EBP- in disparate signaling pathways may be a testimony for its versatility as opposed to the promiscuity. Such plasticity may be necessary for C/EBP- to mediate distinct biological outcomes. Indeed, mice lacking C/EBP- gene have defects in the macrophage-dependent antibacterial and antitumor defenses (32), glucose homeostasis (33), and develop lymphoproliferative disorders (34, 35). C/EBP- may be entailed by specific cues in each of these processes. IFN- signaling is critical for mediating anti-infectious pathogen defenses (36, 37) and suppression of neoplastic cell growth in vivo (18). The fact that C/EBP- is regulated by IFN- (this report) is consistent with a hypothesis that C/EBP- contributes partly to the in vivo effects of IFN- (32, 38). Furthermore, IFNs have also been implicated in the cellular differentiation of certain hematopoietic cell lines (39, 40). C/EBP-, a known regulator of differentiation (41, 42), may now provide a basis for such observations.

	ACKNOWLEDGEMENTS

We thank Richard Hanson and Peter Johnson for providing C/EBP expression plasmids.

	FOOTNOTES

^* This work was supported by National Institutes of Health NCI Grants CA71401 and CA 78282 (to D. V. K.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

These authors contributed equally to this study and should be considered as first authors.

^§ Current address: Morgan State University, Baltimore, MD 21251.

^¶ Current address: NCI-ABL Basic Research Program, Frederick, MD 21702.

To whom correspondence should be addressed.

	ABBREVIATIONS

The abbreviations used are: IFN, interferon; C/EBP, CCAAT enhancer-binding protein; EMSA, electrophoretic mobility shift assay; GATE, -interferon-activated transcriptional element; GBF, GATE binding factor; IRF, interferon gene regulatory factor; ISG, interferon-stimulated gene; ISGF, interferon-stimulated gene factor; ISRE, interferon-stimulated response element; JAK, Janus tyrosine kinase; STAT, signal transducing activator of transcription; IL, interleukin; IPTG, isopropyl-1-thio--D-galactopyranoside.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES