Silencer Activity of NFATc2 in the Interleukin-12 Receptor (cid:1) 2 Proximal Promoter in Human T Helper Cells*

Interleukin 12 (IL-12) is a potent enhancer of interferon (cid:2) production by activated T cells. The high-affinity IL-12 receptor (IL-12R) is a heterodimer of a (cid:1) 1 and a (cid:1) 2 subunit. Expression of the signaling IL-12R (cid:1) 2 chain is usually low, as compared with the more abundant (cid:1) 1 chain, and may be rate-limiting for IL-12 sensitivity. Little is known about the mechanisms controlling IL-12R (cid:1) 2 gene expression. Reporter gene assays in IL-12R (cid:1) 2-expressing Jurkat cells showed that truncation of the region from (cid:3) 151 to (cid:3) 61 abrogated promoter activity. The proximal promoter region does not contain a typical TATA box, suggesting a role for SP-1. Indeed, mutagenesis of the (cid:3) 63 SP-1 consensus site decreased transcription by 50%. Electrophoretic mobility shift experiments confirmed the binding of SP-1 and SP-3 at


Interleukin 12 (IL-12) is a potent enhancer of interferon ␥ production by activated T cells. The high-affinity IL-12 receptor (IL-12R) is a heterodimer of a ␤1 and a ␤2
subunit. Expression of the signaling IL-12R␤2 chain is usually low, as compared with the more abundant ␤1 chain, and may be rate-limiting for IL-12 sensitivity. Little is known about the mechanisms controlling IL-12R␤2 gene expression. Reporter gene assays in IL-12R␤2-expressing Jurkat cells showed that truncation of the region from ؊151 to ؊61 abrogated promoter activity. The proximal promoter region does not contain a typical TATA box, suggesting a role for SP-1. Indeed, mutagenesis of the ؊63 SP-1 consensus site decreased transcription by 50%. Electrophoretic mobility shift experiments confirmed the binding of SP-1 and SP-3 at this site. In contrast, truncation of ؊252 to ؊192 increased promoter activity. Likewise, mutagenesis of the consensus nuclear factor of activated T cells site at ؊206 increased promoter activity by 70%, suggesting silencer activity of this element. Electrophoretic mobility shift experiments with primary Th (T helper) cells showed the formation of a specific, T-cell receptor-inducible complex at this site that is sensitive to cyclosporin A and supershifted with anti-NFATc2 in both Th1 and Th2 cells. Accordingly, cyclosporin A dose-dependently increased IL-12R␤2 mRNA expression. These first data on IL-12R␤2 gene regulation indicate a TATA-less promoter, depending on SP-1/SP-3 transcription factors, and a negative regulatory NFAT element at ؊206. This element may contribute to the overall low level of IL-12R␤2 expression on Th cells.
T helper (Th) 1 cells can be categorized according to their cytokine expression profiles. The differential generation of Th cells expressing Th1 and/or Th2 cytokines is key to the outcome of both protective and pathologic immune responses (1,2). Th1 cells secrete high levels of IFN␥ and favor cellular immunity to intracellular pathogens, whereas Th2 cells secrete IL-4 and favor humoral immunity to extracellular pathogens (3). The polarization process of naive T cells is directed by cytokines that are present during initiation of the naive T-cell response. In this respect, IL-4 promotes Th2 cell development, whereas the antigen-presenting cell-derived cytokine IL-12 is a potent inducer of IFN␥ production and of the generation of Th1 cells (4,5). For Th cells to respond to these cytokines, they need functional receptors.
A functional high-affinity IL-12R is composed of two protein subunits, the IL-12R␤1 and IL-12R␤2 chains. In the human the ␤1 and ␤2 subunits contribute equally to IL-12 binding (6). The ␤2 chain, however, appears to be rate-limiting for IL-12 responsiveness, as it is crucial for signal transduction (7) and, in contrast to the more abundant ␤1 chain, is expressed to a maximum of only a few hundred molecules per cell (6). We have previously shown that allergen-specific Th2 cell clones generated from atopic patients revealed a complete lack of signaling via the IL-12R, as indicated by their inability to phosphorylate STAT4 (8) and to secrete IFN␥ in response to IL-12. Rogge et al. (6) showed that development of naive T cells into Th2 cells is associated with IL-4-mediated suppression of IL-12R␤2 mRNA and protein expression leading to the loss of IL-12 responsiveness and, consequently, the inability of IL-12 to promote IFN␥ production (9). As IL-12 responsiveness is a major parameter in the regulation of specific immunity, we started to unravel the molecular mechanisms that govern the transcriptional regulation of the IL-12R␤2 gene in human Th cells.
To this aim, we cloned 0.6 kilobase of the 5Ј flanking region and, by serial truncation, tested for promoter activity applying a reporter gene assay in IL-12R␤2-expressing Jurkat T cells. In this report we provide the first experimental evidence that SP-1 family members are important for basal and inducible activity of the TATA-less core promoter of the IL-12R␤2 gene and that the inducible transcription factor NFATc2 binding at Ϫ206 has a suppressive role in IL-12R␤2 expression. This suppressive activity does not underlie the loss of IL-12R␤2 expression in Th2 cells.

EXPERIMENTAL PROCEDURES
Materials-Restriction enzymes and T4 ligase were purchased from Promega (Leiden, The Netherlands). All high pressure liquid chromatography-purified oligonucleotides were purchased from BIOSOURCE (Nivelles, Belgium). BstEII-digested DNA (New England Biolabs, Beverly, MA) was used as a molecular weight reference.
Plasmid Construction-Clone pAC188 containing IL-12R␤2-encoding genomic DNA was selected by screening a human genomic pAC * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM /EBI Data Bank with accession number(s) AF349574.
library (Genome Technology Center, Leiden University Medical Center, Leiden, The Netherlands) using IL-12R␤2 cDNA (ϩ781 to ϩ3229) as a probe. Starting from IL-12R␤2 exon 1, clone pAC188 was sequenced in the 5Ј direction. A fragment spanning Ϫ591 to ϩ54 (relative to the start of the reported cDNA sequence (10)), designated construct Ϫ591, was amplified by PCR using the pAC188 plasmid as a template. For cloning purposes, the 5Ј-sense primers were designed with an additional SacI restriction site and the 3Ј-antisense primer with a natural HincII site, resulting in PCR products spanning through ϩ54. Serial deletion fragments, designated Ϫ404 (Ϫ404 to ϩ54), Ϫ252 (Ϫ252 to ϩ54), Ϫ192 (Ϫ192 to ϩ54), Ϫ151 (Ϫ151 to ϩ54), Ϫ61 (Ϫ61 to ϩ54), and Ϫ36 (Ϫ36 to ϩ54), were generated by varying the 5Ј-sense primer ( Table I). The PCR products were subcloned into the pGEM®-T Easy plasmid (Promega) following the directions of the manufacturer. All constructs were checked by DNA sequencing using Thermo Sequenase (PerkinElmer Life Sciences) on an ABI Prism 310 Genetic Analyzer (PerkinElmer Life Sciences). For transfection studies pGL3-enhancer (pGL3e) vector (Promega) was used, which contains the Firefly luciferase gene. The cloned PCR products and the pGL3e vector (Promega) were digested with SacI and HincII or SacI and SmaI, respectively, agarose gel-purified (Qiagen, Hilden, Germany), and ligated with T4 DNA ligase (Promega). pGL3e constructs were checked by sequencing. Plasmid DNA was prepared from bacterial cultures using Qiagen Plasmid Midi Kits.
Mutagenesis-Site-directed mutagenesis of the IL-12R␤2 Ϫ591 to ϩ54 promoter construct was carried out using the Altered Sites II in vitro mutagenesis system from Promega. All reactions were carried out according to the manufacturer's instructions. The internal forward primers containing the mutated sites are shown in Table I. Products from this procedure were cloned into pGL3e and sequenced to confirm the introduction of the desired mutations.
Transient Transfection Studies-Jurkat cells (5 ϫ 10) (6) expressing both IL-12R␤1 and IL-12R␤2 were electroporated in the presence of 20 g of plasmid DNA in 0.5 ml of cytomix, as described previously (11), in a 0.4-cm gap electroporation cuvette (Bio-Rad) at 310 V, 900 microfarads using a Gene Pulser (Bio-Rad). To monitor transfection efficiency, 250 ng of pRL-CMV, an expression vector containing the Renilla luciferase gene under the control of a cytomegalovirus promoter (Promega), was added to each sample. To compensate for size differences of the constructs, empty pGL3e vector was added to obtain an equal amount of DNA in each sample. Immediately after transfection, 9.5 ml of complete culture medium was added (Iscove's modified Dulbecco's medium, Bio-Whittaker, Walkersville, MD) supplemented with 5% pooled, C-inactivated fetal calf serum (BioWhittaker) and gentamycin (80 g/ml; Duch-afa, Haarlem, The Netherlands). Cells were seeded in 2 wells of a 6-well culture plate (Costar, Cambridge, MA). After 1 day of culture at 37°C, cells were either left unstimulated or stimulated for 24 h with 1 ng/ml PMA and 1 g/ml ionomycin (Sigma-Aldrich) or with mouse mAbs to CD3 (1 g/ml; CLB-T3/4E) and CD28 (2 g/ml; CLB-CD28/1), both obtained from the Central Laboratory of the Netherlands Red Cross Blood Transfusion Service (CLB, Amsterdam, The Netherlands). Cells were harvested 48 h post-transfection. Using the Dual Luciferase reporter gene assay (Promega), cell lysates were prepared and both Firefly luciferase and Renilla luciferase were measured by a dual luminometer (Promega). Luciferase activities were normalized for transfection efficiency using Renilla luciferase activity. Transfections were performed in duplicate, and the results of at least three independent experiments were calculated as the mean Ϯ S.D. values for luciferase activity.
T-cell Isolation, Culture, and Stimulation-Peripheral blood mononuclear cells from healthy individuals were isolated by density gradient centrifugation on Lymphoprep (Nycomed, Torshov, Norway). Highly purified CD4 ϩ T cells (normally Ͼ98% as assessed by flow cytometry) were obtained from peripheral blood mononuclear cells with anti-CD4-coated Dynabeads (Dynal AS, Oslo, Norway) as described before (12). CD45RA ϩ CD45RO Ϫ naive Th cells were isolated from peripheral blood mononuclear cells through one-step high-affinity negative selection columns (R&D Systems, Abingdon, UK). Naive Th cells were stimulated under Th1 or Th2 driving conditions in IL-12 and IL-4, respectively, as described before (13) to generate highly polarized Th1 and Th2 cells. To test the effect of CsA on IL-12R␤2 mRNA expression, naive Th cells were stimulated for 3 days with immobilized CD3 mAb and soluble CD28 mAb (13) in 96-well culture plates (Costar; 10 (5) cells/well) with or without CsA at increasing concentrations (10 -1000 ng/ml). All T-cell cultures were grown in complete culture medium with rIL-2 (10 units/ml; Chiron, Emeryville, CA). Proliferative responses were assayed in parallel cultures of 2 ϫ 10 (4) cells/well after 3 days as described before (12).
Preparation of Whole Cell and Nuclear Protein Extracts and Electrophoretic Mobility Shift Assays-Whole cell protein extracts were prepared from 5 ϫ 10 6 CD4 ϩ T cells. Nuclear protein extracts were prepared from 5 ϫ 10 6 cells CD4 ϩ , Th1, or Th2 cells, which were left unstimulated or were stimulated with anti-CD3/anti-CD28 for 30 min in the presence or absence of cyclosporin A (CsA; 1 g/ml, Sigma-Aldrich). Cells were washed with ice-cold phosphate-buffered saline. Nuclear and whole cell protein extracts were isolated essentially as described before (8), and the protein concentrations were determined by a Bradford microassay (Bio-Rad) using a calibrated solution of bovine a S, sense oligonucleotide; AS, anti-sense oligonucleotide. b Location, represents the location of the oligonucleotide in the IL-12R␤2 gene relative to the reported start of the mRNA (10). n.a., not applicable for these oligonucleotides.
c Mut, LC, oligonucleotide used, respectively, for mutagenesis or Light Cycler. d Appropriate transcription factor core-binding elements are underlined; dinucleotide substitutions in the mutant (Mut) relative to the corresponding wildtype (WT) probes are indicated in italics; bold nucleotides represent 5Ј overhang used to fill in by Klenow fragment the double-stranded oligo with [␣-32 P]dATP and d(C/G/T)TP. serum albumin (Sigma-Aldrich) as a reference. The samples were aliquoted and stored at Ϫ80°C. An electrophoretic mobility shift assay (EMSA) was performed as described before (8) with some minor modifications. The double-stranded DNA probe was [␣-32 P]dATP-labeled using the Prime-a-gene labeling system (Promega) and purified using Bio-Spin 6 chromatography columns (Bio-Rad). The binding reaction was incubated at 4°C for 45 min. Cold competitor oligonucleotides were added to the reaction mix prior to the protein extract. The SP-1 consensus sequence, binding SP-1 family members, is derived from human herpesvirus. The NF-B consensus oligonucleotide was obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The nomenclature and sequences of the oligonucleotides used in this study are summarized in Table I. For supershift experiments, 4 g of IgG 1 mAb NFATc1 (sc-7294, Santa Cruz), 0.5 g of IgG 1 mAb NFATc2 (BD Transduction Laboratories, San Diego, CA), 0.4 g of polyclonal IgG SP-1 antibody (sc-59-G, Santa Cruz), 4 g of IgG 1 mAb SP-3 (sc-644-G, Santa Cruz), or a nonrelevant IgG 1 isotype control were added to the binding reaction for an additional 45 min at 4°C. The whole sample was then loaded onto a 4% polyacrylamide gel in 0.5ϫ Tris/borate/EDTA buffer. The gels were prerun for 30 min and then run for 2 h at RT at 150 V. The gels were transferred to Whatman filter paper, dried, and exposed to x-ray film (Kodak XAR5 films, Rochester, NY) at Ϫ80°C.
Reverse Transcriptase (RT)-PCR and Real-time Quantitative RT-PCR Analysis of IL-12R␤1, IL-12R␤2, IL-2, and ␤2m mRNA Expression and IL-13 Measurement-RT-PCR was performed as described before (14) with IL-12R␤1-, IL-12R␤2-, and ␤2-microglobulin (␤2m)-specific primers (Table I). For quantitative analysis of IL-12R␤2 mRNA expression, naive Th cells were stimulated as described above and lysed for total RNA extraction at day 3 using a Nucleo Spin RNA isolation kit (Macherey-Nagel, Duren, Germany). First-strand cDNA was synthe-sized from total RNA using a cDNA synthesis kit (MBI-Fermentas, St. Leon-Rot, Germany). Real-time quantitative PCR was performed in a Light-Cycler (Roche Diagnostics, Almere, The Netherlands) based on specific primers and general fluorescence detection with SYBR green. ␤2m was used as a control. The primer sequences for IL-12R␤2 (melting temperature 60°C), IL-2 (melting temperature 58°C) and ␤2m (melting temperature 60°C) are listed in Table I. Measurement of IL-13 levels in culture supernatants were performed using the PeliKine compact human IL-13 enzyme-linked immunosorbent assay kit obtained from the CLB.

Identification of the IL-12R␤2 Core Promoter and Regulatory
Regions-To characterize the proximal promoter of the human IL-12R␤2 gene, we analyzed 0.6 kilobase of 5Ј flanking sequence obtained from a genomic pAC clone. This fragment, which is depicted in Fig. 1A, spans the region Ϫ591 through ϩ54 relative to the start of the IL-12R␤2 cDNA (10). In the immediate 5Ј region, no typical TATA box or CCAAT box motif was found. To identify the core promoter and regulatory elements, the transcriptional activity of the truncated constructs of the IL-12R␤2 5Ј upstream region ( Fig. 2A) were assayed. For that purpose, fulllength construct Ϫ591 and a series of 5Ј deletion fragments were subcloned into the pGL3-Enhancer vector, a promoterless luciferase reporter vector with an SV40 enhancer. All constructs were tested for promoter activity in unstimulated and PMA/ionomycin-stimulated conditions in Jurkat T cells, showing constitutive  (14) was tested for promoter activity. Numbering of the sequence is relative to the start of the reported cDNA sequence (10), and the gene is located on chromosome 1 at region p31.2 (33). B, sequence of the 5Ј flanking genomic DNA sequence as deposited under Gen-Bank TM accession number AF349574. The most distal 5Ј nucleotide of each oligonucleotide used to construct the different deletion fragments is indicated by a gray triangle and is named after this position accordingly. The most 3Ј ϩ54 nucleotide of all constructs is indicated by the black triangle. SP-1 family consensus binding elements (SP-1#1 and SP-1#2) and the NFAT consensus binding element are indicated in bold type above the consensus sequence (underlined). expression of the IL-12R␤1 chain and low but inducible expression of the IL-12R␤2 chain (Fig. 2B). The expression of the IL-12R␤2 chain in Jurkat cells has been described (15). Promoter activity was up-regulated in stimulated cells transfected with constructs Ϫ591, Ϫ404, Ϫ252, Ϫ192, and Ϫ151. Deletion from Ϫ591 to Ϫ404 and further to Ϫ252 reduced the reporter activity stepwise, suggesting multiple positive cis-acting elements between Ϫ591 and Ϫ252. Interestingly, deletion from Ϫ252 to Ϫ192 led to an increase in promoter activity, suggesting the presence of a suppressor element(s) in this region. Promoter activity was abrogated after truncation to Ϫ61 and Ϫ36, indicating that sequences in close upstream proximity of Ϫ61 are crucial for basal and inducible IL-12R␤2 promoter activity. Similar results were found upon stimulation with anti-CD3 and anti-CD28 (data not shown).
Identification of Functional Motifs in the Core Promoter and Negative Regulatory Region-Two potential binding sites for the SP-1 family of transcription factors (core sequence GGGCGG (16)) are located in the proximal promoter region at Ϫ3 and Ϫ63 (Fig. 1B). As SP-1 is frequently involved in transcription initiation in the absence of a TATA box (16), we analyzed whether these DNA motifs at Ϫ3 (SP-1#1) and Ϫ63 (SP-1#2) participate in the regulation of IL-12R␤2 promoter activity. To examine the relative roles of these two SP-1 sites for promoter activity, we mutated the GGGCGG motifs in these elements to GTTCGG in the context of the full-length promoter construct Ϫ591. We thus generated two IL-12R␤2 -591 to ϩ54 promoter-reporter gene constructs with either one of the SP-1 sites mutated. Promoter activity was tested after transient transfection of Jurkat cells and compared with the wild type full-length IL-12R␤2 promoter activity. As shown in Fig. 3, mutation of the SP-1#1 site at Ϫ3 does not result in a significant change of activity, whereas mutation of the SP-1#2 site at Ϫ63 results in a reduction of promoter activity by almost 50%, suggesting an important role of this cis-regulatory element. As determined by serial truncation, deletion of the promoter re- gion from Ϫ252 to Ϫ192 resulted in increased promoter activity. Within this region a reversed NFAT consensus binding site (TTTCC) is located at Ϫ206. To examine the functional significance of this NFAT consensus site, we mutated the TTTCC motif into TTTAA in the context of the full-length promoter construct Ϫ591. Promoter activity was tested after transient transfection of Jurkat cells and compared with the wild type construct Ϫ591. As shown in Fig. 3, a 70% increase of promoter activity was observed with the mutated construct, suggesting that the putative NFAT binding element at Ϫ206 has a negative cis-regulatory role in IL-12R␤2 transcription.
Identification of Nuclear Factors Binding to the Ϫ63 SP-1 Motif-To characterize transcription factor binding activities at the SP-1#2 element at Ϫ63, we performed EMSAs using whole cell extracts from CD4 ϩ T cells. EMSA with the doublestranded oligonucleotide SP-1#2 WT containing the intact Ϫ63 SP-1 element demonstrated the formation of two DNA-protein complexes C1 and C2 (Fig. 4, lane 1) not formed in the absence of protein extract (data not shown). The formation of these radioactive complexes was dose-dependently inhibited by competition with a 10-, 30-, or 90-fold molar excess of unlabeled SP-1#2 WT oligonucleotide (lanes 2-4) but was not affected by a 90-fold molar excess of the mutated SP-1#2 Mut oligonucleotide (lane 5). Similar to the SP-1#2 WT oligonucleotide, competition with a 10-, 30-, and 90-fold molar excess of a specific SP-1 consensus oligonucleotide dose-dependently competed the bands away (lane 6 -8), whereas a 90-fold molar excess of an oligonucleotide containing a nonrelevant NF-B consensus binding site had no effect (lane 9). SP-1 and SP-3 are known to bind to identical DNA elements (17). Therefore, binding reactions were performed in the presence of anti-SP-1, anti-SP-3, or both antibodies. The addition of anti-SP-1 antibody resulted in a supershift of most of complex C1 (Fig. 4, lane 11), whereas with anti-SP-3 antibody, a complete supershift was observed of the less abundant complex C2 (lane 12). The combination of both anti-SP-1 and anti-SP-3 antibodies did not result in additional shifts. These results suggest the binding of SP-1 and SP-3 at the Ϫ63 element.
Identification of Nuclear Factors Binding to the Ϫ206 NFAT Motif-We next tested whether NFAT could actually interact with the putative binding site at Ϫ206 in the IL-12R␤2 promoter. To this aim, we used the Ϫ220 to Ϫ180 DNA sequence as a double-stranded probe for EMSA. Nuclear extracts from anti-CD3/anti-CD28-stimulated CD4 ϩ T cells showed the inducible formation of complex A in addition to the increased intensity of a preexisting complex P (Fig. 5A, lanes 3 and 4). The formation of complexes A and P was abrogated by mutation of the probe (TTTCC to TTTAA; data not shown). Activation of the CD4 ϩ T cells in the presence of the immunosuppressant drug CsA, known to inhibit the nuclear translocation of NFAT, inhibited the formation of complex A but not P (Fig. 5B, lane 5). These results indeed suggest the involvement of NFAT in complex formation with the Ϫ206 element. Of the growing family of NFAT proteins, NFATc1 (NFATc, NFAT2) and NFATc2 (NFATp, NFAT1) are most prominent in peripheral T cells (18) and bind to the same DNA motif (19,20). To identify whether NFATc1 or NFATc2 is involved in complex A, binding reactions were performed in the presence of antibodies to NFATc1 or NFATc2. A supershifted   FIG. 3. The ؊63 SP-1 and ؊206 NFAT motifs are important for promoter activity. Jurkat cells were transiently transfected with luciferase constructs containing the full-length wild type promoter (construct Ϫ591) or the full-length promoter containing the mutated SP-1 motifs at Ϫ3 or Ϫ63 (GGGCGG to GTTCGG), or the mutated NFAT motif at Ϫ206 (TTTCC to TTTAA) as indicated by stars. Cells were stimulated with PMA and ionomycin for 24 h. Corrected luciferase activity was calculated, and promoter activity was expressed as the percentage of wild type promoter activity (top bar, 100%). In addition to deletion fragments, pGL3e (empty vector) is shown as a control. band (S) was obtained with NFATc2 antibody (Fig. 5A, lane 7) but not with antibody to NFATc1 (Fig. 5A, lane 6) or with the IgG 1 isotype control antibody (data not shown), indicating the inducible binding of NFATc2 at Ϫ206.
Because expression of the IL-12R␤2 gene is suppressed in Th2 cells and NFATc2 plays a role in the suppression of Th2-type cytokines (21,22), we next tested for differential binding activity of NFATc2 to the Ϫ206 element comparing nuclear extracts from Th1 and Th2 cells. However, neither in unstimulated nor in anti-CD3/anti-CD28 stimulated cells any difference was observed between Th1 and Th2 cell extracts (Fig. 5B). Preexisting complex P showed a similar increased intensity in Th1 and Th2 cells after TCR stimulation. Also, complex A was induced to the same extent in TCR-stimulated Th1 cells (lanes 1 and 2) and Th2 cells (lanes 5 and 6). Both in Th1 and Th2 extracts, complex A was supershifted with anti- NFATc2 (lanes 4 and 5 and 7 and 8).
The isotype control is shown in lane 9.
IL-12R␤2 mRNA Expression Is Up-regulated by CsA-The data so far suggested a general suppressive role of NFATc2 in the regulation of IL-12R␤2 expression, not discriminating between Th1 and Th2 cells. To test the role of NFAT in a more physiological system, naive T cells were stimulated for 3 days with anti-CD3/anti-CD28 in the presence of increasing concentrations of CsA. The mRNA expression of IL-12R␤2, IL-2, and ␤2m was measured by real-time PCR. The levels of IL-12R␤2 and IL-2 mRNA were normalized based on ␤2m mRNA levels in the same samples. Both IL-2 mRNA expression level, known to be inhibited by CsA (23), and IL-13 protein secretion, known to be enhanced by CsA (24), were used as controls. As expected from the EMSA data, CsA dose-dependently increased IL-12R␤2 mRNA expression (Fig. 6A), confirming the suppressive role of NFAT herein. Furthermore, the IL-2 mRNA expression was decreased (Fig.  6A), and the IL-13 protein secretion was increased (Fig. 6B) in the presence of increasing doses of CsA. DISCUSSION In this report, we have described the first data on the transcriptional regulation of the human IL-12R␤2 gene. The proximal promoter region was cloned and functionally characterized. The data indicate a TATA-less promoter, dependent on SP-1 family protein binding at Ϫ63, and a silencer NFAT element at Ϫ206, which binds NFATc2 and is involved in suppressing TCR-induced IL-12R␤2 expression. The pGL3e vector in which the promoter fragments were cloned contains an SV-40 enhancer located upstream of the luciferase gene. The enhancer element normally increases reporter gene expression provided that the promoter is active. Deletion from Ϫ151 to Ϫ61 resulted in fully abrogated transcription, even in the presence of the enhancer, underlining the critical role of this region in transcription initiation.
The core promoter, which drives TCR-induced transcription (construct Ϫ151), does not contain a TATA or CAAT box. In the absence of a TATA box, SP-1 binding motifs are frequently involved in alternative initiation of transcription (25,26). This seems to apply for the IL-12R␤2 gene as well, as it contains a functional SP-1/3 binding motif in its core promoter. Genes with TATA-less promoters, including many so-called "housekeeping" and receptor genes, are generally expressed at low levels (27). Indeed, even fully IL-12-responsive Th1 cells were shown to express only a few hundred IL-12R␤2 molecules on their membrane (6). The IL-12R␤2 gene contains a GC-rich (Ϯ75%) 5Ј noncoding region (10), which may, at least in part, explain the low rate of expression of these molecules, as GCrich 5Ј noncoding regions are known to hamper translation FIG. 5. Identification of ؊206-binding proteins. The radiolabeled probe NFAT Ϫ206 wild type was incubated with nuclear extracts from freshly isolated CD4 ϩ T cells (A) or from polarized Th1 or Th2 cell lines (B) that were left unstimulated or stimulated with anti-CD3/anti-CD28 for 30 min. Specific complexes (P, A, and S) are indicated by arrowheads. The binding reactions were carried out in the absence or presence of specific antibody to NFATc1 or NFATc2 or in the presence of IgG 1 isotype control antibody as indicated. A, nuclear extracts from CD4 ϩ T cells. Preexisting complex P showed increased binding activity upon T-cell stimulation. Complex A was induced after stimulation (lane 4) and was CsA-sensitive (lane 5). The addition of anti-NFATc2 but not anti-NFATc1 results in the supershift (S) of complex A but not complex P. As a control both antibodies are shown in the absence of nuclear extract (lanes 1 and 2). B, nuclear extracts from Th1 or Th2 cells showed no differences in the formation of complex P or A after TCR stimulation (lanes 1, 2 and 5, 6). Anti-NFATc2, but not anti-NFATc1, supershifts complex A equally well in Th1 and Th2 cells (lanes 3, 4 and 7, 8). The IgG 1 Isotype control is shown in lane 9. (27). However, data on the translational regulation of the IL-12R␤2 mRNA are not available yet.
We show here that IL-12R␤2 expression is inhibited at the transcriptional level. The NFAT element at Ϫ206 specifically binds NFATc2 and seems to be important for a general downregulation of TCR-inducible IL-12R␤2 gene expression. The availability and suppressive activity of NFATc2 does not discriminate between Th1 and Th2 cells and thus does not explain the loss of expression of the IL-12R␤2 chain in Th2 cells. The relative importance of the negative regulatory role of this element is underlined by the observation that CsA increases IL-12R␤2 mRNA expression in stimulated naive Th cells, i.e. in the context of fully intact regulatory regions, instead of the cloned proximal 591 base pairs of the promoter. The absence of further NFAT sites in the region, at least up to Ϫ1.2 kilobases (data not shown), further indicates this particular site as the mediator of the CsA-sensitive suppressive effect.
Similar negative regulatory effects of NFATc2 have been implicated in the regulation of several Th2 type cytokine genes in the mouse (21,22). Indirect data on the regulation of the human IL-13 gene points in the same direction (24). The present data are the first to suggest a suppressor function of NFATc2 in the regulation of a gene associated with Th1-type responses. Whether or not this site contributes to the polarization process of Th cells under Th1 or Th2 driving conditions is unclear thus far, as sites directly involved in the Th2-specific suppression of the IL-12R␤2 gene have not been identified yet. Our present data give no indication for differential NFATc2 activity at Ϫ206 in Th1 and Th2 cells. Instead, NFATc2 may play a role in the general low expression rate (6) or the kinetics of IL-12R␤2 gene expression, in particular in the shut-down of expression at later time points after TCR triggering as has been suggested in the regulation of IL-4 gene expression (28).
It is to be expected that several Th1-or Th2-specific transcription factors do play a role in IL-12R␤2 gene transcription either in a direct or an indirect way. For example, IL-12 strongly up-regulates IL-12R␤2 expression through phosphorylation of STAT4 (6). Therefore, STAT4 binding sites are expected to be located in the enhancer region further upstream as has been suggested before (29). Another transcription factor which may be directly involved in IL-12R␤2 gene regulation is the recently identified, Th1-specific T-box transcription factor, T-bet (30). T-bet is up-regulated by IL-12 and accounts for the Th1-specific expression of IFN␥ and repression of the opposing Th2 programs, at least in the mouse. In contrast, GATA-3, a Th2-specific and IL-4-induced transcription factor (31), may be involved in the direct suppression of the IL-12R␤2 gene. Indeed, Ouyang et al. (32) demonstrated a decrease in IL-12R␤2 mRNA expression after ectopic expression of GATA-3 into differentiated murine Th1 cells. We are currently investigating the role of Th1-and Th2-specific transcription factors in human IL-12R␤2 gene expression.