Interleukin-12 Induces Expression of Interferon Regulatory Factor-1 via Signal Transducer and Activator of Transcription-4 in Human T Helper Type 1 Cells*

IRF-1-deficient mice show a striking defect in the development of T helper 1 (Th1) cells. In the present report, we investigate the expression of IRF-1 during differentiation of human T helper cells. No significant differences of IRF-1 mRNA expression were found in established Th1 and Th2 cells; however, interleukin 12 (IL-12) induced a strong up-regulation of IRF-1 transcripts in Th1 but not in Th2 cells. We demonstrate that IL-12-induced up-regulation of IRF-1 is mediated by signal transducer and activator of transcription-4, which binds to the interferon (IFN)-γ-activated sequence present in the promoter of the IRF-1 gene. Strong IL-12-dependent activation of a reporter gene construct containing the IRF-1 IFN-γ-activated sequence element provides further evidence for the key role of signal transducer and activator of transcription-4 in the IL-12-induced up-regulation of IRF-1 transcripts in T cells. IRF-1 expression was strongly induced after stimulation of naive CD4+ T cells via the T cell receptor, irrespective of the cytokines present at priming, indicating that this transcription factor does not play a major role in initiating a Th1-specific transcriptional cascade in differentiating helper T cells. However, our finding that IRF-1 is a target gene of IL-12 suggests that some of the IL-12-induced effector functions of Th1 cells may be mediated by IRF-1.


IRF-1-deficient mice show a striking defect in the development of T helper 1 (Th1) cells. In the present report, we investigate the expression of IRF-1 during differentiation of human T helper cells. No significant differences of IRF-1 mRNA expression were found in established Th1 and Th2 cells; however, interleukin 12 (IL-12) induced a strong up-regulation of IRF-1 transcripts in Th1 but not in Th2 cells. We demonstrate that IL-12-induced up-regulation of IRF-1 is mediated by signal transducer and activator of transcription-4, which binds to the interferon (IFN)-␥-activated sequence present in the promoter of the IRF-1 gene. Strong IL-12-dependent activation of a reporter gene construct containing the IRF-1 IFN-␥-activated sequence element provides further evidence for the key role of signal transducer and activator of transcription-4 in the IL-12induced up-regulation of IRF-1 transcripts in T cells. IRF-1 expression was strongly induced after stimulation of naive CD4 ؉ T cells via the T cell receptor, irrespective
of the cytokines present at priming, indicating that this transcription factor does not play a major role in initiating a Th1-specific transcriptional cascade in differentiating helper T cells. However, our finding that IRF-1 is a target gene of IL-12 suggests that some of the IL-12induced effector functions of Th1 cells may be mediated by IRF-1.
The discovery of functionally distinct subsets of T helper cells has provided a better understanding for the heterogeneity of immune responses in normal and pathological situations (1)(2)(3)(4). However, the transcriptional programs that control the differentiation of naive CD4 ϩ T cells into polarized Th1 1 and Th2 cells are just beginning to be elucidated (5).
The differentiation process is initiated by stimulation of the TCR and directed by cytokines present during the initiation of a T cell response (6). IL-4 promotes Th2 development (7,8), whereas IL-12 produced by antigen presenting cells is a potent inducer of Th1 cells (9 -13). Signaling of these two cytokines is mediated through the activation of specific signal transducer and activator of transcription (STAT) proteins. STAT-6 is activated by IL-4 in Th2 cells (14 -16), and IL-12 activates STAT-4 in Th1 cells (17)(18)(19)(20)(21). Studies on knockout mice for these transcription factors clearly indicate the pivotal role of STAT-6 and STAT-4 in T helper cell differentiation. STAT-6-deficient T lymphocytes fail to differentiate into Th2 cells in response to IL-4 (22)(23)(24), and the analysis of STAT-4 Ϫ/Ϫ T cells has revealed an impaired production of IFN-␥ upon antigen receptor triggering, indicative of a defect in Th1 differentiation (25,26). More recently, additional transcription factors have been studied and characterized for their role in the Th2-specific expression of the IL-4 gene. The nuclear factor of activated T cells (NF-AT) has been shown to regulate IL-4 expression in Th2 cells through a cooperative binding with AP-1 family members (27). NF-AT functions are potentiated by the transcription factor NIP45 (for NF-ATp interacting protein) (28). The protooncogene c-maf is expressed in Th2 clones but not in Th1 clones and has also been identified as a potent transactivator of the IL-4 gene (29). More recently, it has been shown that the transcription factor GATA-3 is expressed at a high level in naive CD4 ϩ T cells and differentiating and effector Th2 cells, whereas its expression is suppressed in Th1 cells (30). Less is known about the regulation of the IFN-␥ gene in Th1 cells that appears to involve the interaction of numerous transcription factors, including NF-kB and NF-AT (31), or the cooperative binding of STAT-4 dimers (32).
Recently, it has been suggested that transcription factors belonging to the interferon regulatory factor (IRF) family are necessary for Th1 development (33,34). The IRF family includes at least 10 members: IRF-1, IRF-2, IRF-3, ISGF3␥/p48, ICSBP, Pip/ICSAT/IRF-4, IRF-5, IRF-6, IRF-7, and viral IRF, recently identified in the genome of HHV-8 (35). IRF expression is either constitutive and/or induced upon treatment with interferons (IFNs) or other cytokines or in response to viral infection. Moreover, some IRFs are specific for hematopoietic cells (ICSBP and IRF-4), whereas others are expressed in multiple tissues and cell lines. However, all members of this family share significant homology in the amino-terminal 115 amino acids, which make up the DNA binding domain that mediates the interaction with a specific consensus IRF binding sequence motif, termed IRF-E (36). Proteins of the IRF family have been shown to be involved in the regulation of the pleiotropic activities elicited by IFNs and other cytokines, including modulation of the immune response, inflammation, hematopoiesis, cell proliferation, and differentiation (37,38).
Studies of IRF-1-deficient mice have revealed that this factor is implicated in the regulation of several immune processes, such as T-cell selection and maturation (39,40), development of NK cells (41)(42)(43), and development of Th1 cells (33,34). The compromised Th1 differentiation was associated with defects in the expression of p40 subunit of IL-12 by cells of myeloid origin (33,34). Indeed, a potential IRF-1-responsive element was found in the promoter region of the IL-12 p40 gene (44,45). Together, these results suggest that IRF-1 may be a master gene, directly or indirectly controlling Th1 responses at multiple stages.
In the present report, we have analyzed the expression of IRF-1 in developing human Th1 and Th2 cells and its regulation in response to cytokines in polarized Th1 and Th2 cells. IRF-1 transcripts were strongly induced after stimulation of naive CD4 ϩ T cells via the T cell receptor, irrespective of the cytokines present at priming, and no differences in IRF-1 mRNA expression were detectable in established Th1 and Th2 cells. Interestingly, IL-12 treatment up-regulated IRF-1 transcripts only in established Th1 cells. This up-regulation does not depend on de novo protein synthesis. IL-12 induced binding of STAT-4 to the IFN-␥-activated sequence (GAS) of the IRF-1 gene promoter in Th1 cells but not in Th2 cells. Strong IL-12dependent activation of a reporter gene construct containing the IRF-1 GAS element further emphasizes the key role of STAT-4 for IL-12-induced up-regulation of IRF-1 transcripts in T cells.

Generation of Th1 and Th2 Lines from Cord Blood Leukocytes-
Human neonatal leukocytes were isolated from freshly collected, heparinized, neonatal blood by Ficoll-Paque (Amersham Pharmacia Biotech, Uppsala, Sweden) density gradient centrifugation. CD8 ϩ T cells were removed by positive selection with anti-CD8 microbeads and magnetic-activated cell sorting according to a protocol supplied by the manufacturer (Miltenyi Biotec, Bergisch Gladbach, Germany). Cells were stimulated with 2 g/ml phytohemagglutinin (Wellcome, Beckenham, United Kingdom) in the presence of 2.5 ng/ml IL-12 (Hoffmann-La Roche Inc., Nutley, NJ) and 200 ng/ml neutralizing anti-IL-4 antibodies (18500D, Pharmingen, San Diego, CA) for Th1 cultures or 1 ng/ml IL-4 (Pharmingen) and 2 g/ml neutralizing anti-IL-12 antibodies 17F7 and 20C2 (kindly provided by M. Gately, Hoffmann-La Roche Inc.) for Th2 cultures. Cells were washed on day 3 and expanded in complete RPMI 1640 medium (Life Technologies, Inc.) supplemented with 5% Fetal-Clone I (HyClone, Logan, UT), 2 mM L-glutamine, 1 mM sodium pyruvate, 100 units/ml penicillin-streptomycin) containing 100 units/ml IL-2 (Hoffmann-La Roche Inc.). The cells were treated with IL-12 (2.5 ng/ml) or IFN-␣ (1000 units/ml) followed by the extraction of total RNA. To determine whether de novo protein synthesis is required for IL-12induced up-regulation of IRF-1 transcripts, cells were treated with the protein synthesis inhibitor cycloheximide (Sigma, 10 g/ml) 45 min prior to the addition of IL-12 (2.5 ng/ml). RNA was extracted 16 h after the addition of IL-12.
Purification and Stimulation of Naive CD4 ϩ T Cells-CD4 ϩ / CD45RO Ϫ T cells were purified from cord blood leukocytes by negative selection using a CD4 ϩ T cell isolation kit and CD45RO microbeads according to a protocol supplied by the manufacturer (Miltenyi Biotec). The purity of the CD4 ϩ /CD45RO Ϫ T cells using this procedure was typically Ͼ98% as determined by flow cytometry. Purified naive T cells were stimulated with plate-bound anti-CD3 mAb (TR66 (46)) in the absence of exogenously added cytokines, in the presence of IL-12 (2.5 ng/ml) and neutralizing anti-IL-4 mAb (200 ng/ml), or in the presence of IL-4 (1 ng/ml).
Single Cell Analysis of Intracellular IFN-␥ and IL-4 Production-Single cell analysis of IFN-␥ and IL-4 production was performed as described previously (47). Briefly, T cell lines were collected 7 days after priming and washed, and 10 6 cells were restimulated with phorbol 12-myristate 13-acetate (50 ng/ml) (Sigma) and ionomycin (1 g/ml) (Sigma) for 2 h at 37°C in complete medium. Brefeldin A (10 g/ml) (Sigma) was added to the cultures and the cultures were incubated for an additional 2 h. Then, the cells were fixed with 4% paraformaldehyde and permeabilized with saponin. Fixed cells were stained with antihuman IFN-␥-FITC (Pharmingen) and anti-human IL-4-PE (Pharmingen) following a protocol provided by the manufacturer and analyzed with a FACScan flow cytometer (Becton Dickinson, Mountain View, CA).
Ribonuclease Protection Assays-To obtain the pBS IRF-1 construct, the plasmid pUC IRF-1 (a generous gift of T. Taniguchi) was digested with SmaI, and the 400-base pair-long fragment was cloned into the same sites of pBluescript/KS (Stratagene, La Jolla, CA). To generate the 32 P-labeled 280-base pair-long antisense IRF-1 RNA probe, the plasmid pBS IRF-1 was linearized with EcoRI and transcribed by T7 polymerase. A 327-base pair DNA fragment encompassing the cytoplasmic region of the human IL-12R␤2 subunit was subcloned in pGEM 3Z (21). This construct was linearized with EcoRI, and radiolabeled antisense transcripts were synthesized with SP6 polymerase and a commercial kit according to the manufacturer's protocol (Promega, Madison, WI). RNA was extracted from T cell lines using Ultraspec total RNA extraction reagent (Biotecx Laboratories Inc., Houston, TX). The antisense RNA probes were hybridized to 5 g of total RNA, and ribonuclease protection assays were performed with a commercial kit (Ambion Inc., Austin, TX) according to the company's protocol. Products were resolved on 6% denaturing polyacrylamide gels, and the protected fragments were visualized by autoradiography. The radioactivity present in the protected fragments was also quantitated using a Molecula-rImager (Bio-Rad). An 18 S RNA probe was used as a control for equal RNA loading.
Reporter Gene Assays-Jurkat cells were co-transfected with the following constructs: (i) pBOS-IL-12R␤1, pBOS-IL-12R␤2 (48) (kindly provided by Dr. U. Gubler, Hoffmann-La Roche Inc.) and pBOS-Stat4 (kindly provided by Dr. R. Chizzonite, Hoffmann-La Roche Inc.), (ii) pIRF-luc, a luciferase reporter plasmid containing three copies of an oligonucleotide representing the GAS of the IRF-1 promoter (5Ј-CT-GATTTCCCCGAAATGAC-3Ј) inserted upstream of the SV40 promoter in pGL3-promoter (Promega), and (iii) pCMV-␤-gal, a mammalian expression vector containing the ␤-galactosidase gene driven by the cytomegalovirus promoter. 30 g of plasmid DNA (7 g each of pIRF-luc, pBOS-IL-12R␤1, pBOS-IL-12R␤2, and pBOS-Stat4 and 2 g of pCMV-␤-gal) were used to electroporate 8 ϫ 10 6 cells in 400 l of RPMI medium with a 0.4-cm gap electroporation cuvette at 960 microfarads and 300 V using a Gene Pulser (Bio-Rad). Four ml of complete medium were added immediately after transfection, and the cells were seeded in 2 wells of a 24-well plate. Four h after transfection, 4 ng/ml IL-12 was added to half of the cultures, and the cells were incubated for additional 20 h. Luciferase activity was measured using a luciferase reporter gene assay (Boehringer Mannheim), and ␤-galactosidase activity was determined with the ␤-galactosidase reporter gene assay (Boehringer Mannheim). Both activities were measured in a LB 9507 Lumat luminometer (EG & G Berthold, Milano, Italy). Luciferase activities were normalized using ␤-galactosidase activity.

IRF-1 Expression in Human Th1 and Th2 Cells-Recent
reports have demonstrated that IRF-1-deficient mice lack the capacity to develop a Th1 response (33,34). To study whether the transcription factor IRF-1 may also play a role in human T helper cell development, we have analyzed expression of IRF-1 transcripts in human Th1 and Th2 cells generated from cord blood (21,50). The T helper phenotype has been determined by the analysis of intracellular IFN-␥ and IL-4 production after restimulation of T cells with phorbol 12-myristate 13-acetate and ionomycin (Fig. 1).
We examined the expression of IRF-1 transcripts in Th1 and Th2 cells treated with or without IL-12 or IFN-␣ by RNase protection assays. We could not detect significant differences of the IRF-1 mRNA levels in human Th1 and Th2 cells 2 weeks after priming ( Fig. 2A, lanes 1 and 7). Interestingly, we found that IRF-1 expression is clearly induced by IL-12 in Th1 but not in Th2 cells ( Fig. 2A, lanes 2, 3, 8, and 9). The IL-12-induced up-regulation of IRF-1 transcripts is not blocked by cycloheximide indicating that de novo protein synthesis is not required for IL-12 action (Fig. 2A, lane 4). Treatment of Th1 cells with cycloheximide in the absence of IL-12 does not result in an increased IRF-1 mRNA expression (data not shown). IFN-␣ induced IRF-1 transcripts in both T helper subsets ( Fig. 2A,  lanes 5, 6, 10, and 11). A quantitative analysis of the results is shown in Fig. 2B. Treating Th1 cells for 4 h with IL-12 resulted in a 5-fold up-regulation of IRF-1 transcripts; a 9-fold upregulation was detected after 16 h. Consistent with previously published data (51), we observed that IFN-␥ up-regulates IRF-1 in human Th2 but not Th1 cells (data not shown). These results demonstrate that IRF-1 is a target gene of IL-12 in human Th1 cells.
STAT-4 Binds to a GAS Element within the IRF-1 Promoter-To analyze whether the IL-12-induced expression of IRF-1 is mediated by a specific STAT protein, we performed gel-shift experiments using a specific oligonucleotide corresponding to the GAS within the IRF-1 promoter (49,52). Total cell extracts were prepared from untreated Th1 and Th2 cells or from cells incubated with IL-12, IFN-␣, IFN-␥, and IL-4 and were used in gel-shift experiments shown in Fig. 3. Consistent with our previous findings (21) and those of other laboratories (19,20,53), IL-12 activated a STAT complex only in Th1 cells but not in Th2 cells. IFN-␣ strongly induced two STAT complexes with different electrophoretic mobility both in Th1 and Th2 cells, whereas only Th2 cells were sensitive to IFN-␥. The activation of a STAT molecule by IFN-␥ in Th2 but not in Th1 cells correlates with the selective expression of the IFN-␥ receptor ␤ chain on Th2 cells (51,54,55).
In order to define which STAT molecules bind to the IRF-1 GAS element following IL-12 treatment, we performed supershift assays using specific antibodies against STAT proteins.
As shown in Fig. 4, anti-STAT-1 antibodies supershifted the fast migrating complex, whereas the anti-STAT-4 antibodies abrogated the slow migrating complex induced by IFN-␣ both in Th1 and Th2 cells. These findings are consistent with the previously described observation that IFN-␣ activates STAT-4 in human mitogen-activated lymphoblasts and in human Th1 and Th2 cells (50,56). The complex induced in IL-12-treated Th1 cells is abrogated with STAT-4 specific antibodies, confirming previously published results (17,20,21). Addition of anti-STAT-3 antibodies did not significantly change the DNAbinding complexes, suggesting that STAT-3 does not play a major role in activating the IRF-1 promoter in response to the cytokines tested in this study. These results suggest that STAT-4 can induce IRF-1 expression in Th1 cells in response to IL-12.
Mechanisms Underlying the IRF-1 Expression by IL-12-We next analyzed the mechanism by which IL-12 induces IRF-1 expression. Three copies of the GAS element present in the FIG. 1. Cytokine profiles of human Th1 and Th2 cell lines generated from cord blood leukocytes. Human Th1 and Th2 cell lines were generated by stimulating cord blood leukocytes that had been depleted of CD8 ϩ cells with mitogen in the presence of IL-12 and anti-IL-4 mAbs and in the presence of IL-4 and anti-IL-12 mAbs, respectively. Cells were harvested 7 days after stimulation, and the intracellular production of IL-4 and IFN-␥ was analyzed by flow cytometry as described under "Experimental Procedures."

FIG. 2. IL-12 induces IRF-1 transcripts in human Th1 cells. A,
Th1 and Th2 lines generated from cord blood were harvested 2 weeks after stimulation. The cells were treated as indicated with IL-12 (2.5 ng/ml) or IFN-␣ (1000 units/ml) followed by the extraction of total RNA. To determine whether de novo protein synthesis is required for IL-12induced up-regulation of IRF-1 transcripts, cells were treated with the protein synthesis inhibitor cycloheximide (CHX) (10 g/ml) 45 min prior to the addition of IL-12 (2.5 ng/ml). RNA was extracted 16 h after the addition of IL-12. Transcripts encoding IRF-1 (top panel) and 18 S RNA as loading control (bottom panel) were analyzed by ribonuclease protection assays as described under "Experimental Procedures." B, the radioactivity present in the protected fragments was quantitated using a MolecularImager. The black bars correspond to the relative expression levels of IRF-1 mRNA after normalization with an 18 S RNA probe. IRF-1 promoter were inserted upstream of a luciferase reporter gene. The resulting construct, termed IRF-1-luc, was transiently transfected in Jurkat cells. Because Jurkat cells do not respond to human IL-12, we cotransfected expression vectors encoding human IL-12R␤1, IL-12R␤2, and STAT-4. The cells were treated for 20 h with or without IL-12 prior to the preparation of cell extracts. Strong IL-12-dependent activation of the IRF-1-luc construct was detected when the components of the IL-12 signaling machinery, i.e. IL-12R␤1, IL-12R␤2, and STAT-4, were simultaneously expressed in Jurkat cells (Fig. 5). In contrast, no IL-12-dependent induction of luciferase activity was observed when the cDNAs encoding IL-12R␤2 or STAT-4 were omitted. Treatment of transfected Jurkat cells for 6 h with IL-12 also resulted in increased luciferase activity (data not shown). To exclude the possibility that the up-regulation of IRF-1 is secondary to IFN-␥ production induced in response to IL-12 we determined IFN-␥ in the supernatant of the transfected cells by enzyme-linked immunosorbent assays. There was no detectable IFN-␥-production by the transfected cells, indicating that IL-12-dependent induction of luciferase activity is not mediated by IFN-␥ (data not shown). These results provide strong evidence that IL-12-induced activation of the IRF-1 gene is mediated by STAT-4 and depends on the presence of functional IL-12 receptors on the cell surface. Moreover, our data suggest that IRF-1 expression in human Th1 cells is regulated by IL-12 through STAT-4 activation.
Regulation of IRF-1 Expression in Differentiating T Helper Cells-Two recent studies have demonstrated that IRF-1-deficient mice show a striking defect in the development of Th1 cells (33,34). Interestingly, these two reports differ in their conclusions regarding the relevance of IRF-1 in CD4 ϩ T cells.
To determine the role of IRF-1 in the development of human T helper cells, we analyzed the expression of IRF-1 along the differentiation of naive T cells under neutral, Th1-polarizing, or Th2-polarizing conditions. Purified CD4 ϩ , CD45RO Ϫ T cells isolated from cord blood were stimulated with plate-bound anti-CD3 mAb in the absence of exogenously added cytokines, in the presence of IL-12 and neutralizing anti-IL-4 mAb to induce Th1 development, or in the presence of IL-4 to promote Th2 development. The expression of transcripts encoding IL-12R␤2 was used to monitor T helper cell differentiation (Fig.  6A, middle panel). Consistent with our previous findings, IL-12R␤2 transcripts were not detectable in purified naive T cells but were induced by T cell receptor triggering. Transcripts encoding the IL-12R␤2 subunit were expressed at significantly higher levels in T cells stimulated in the presence of IL-12 than in T cells stimulated in the presence of IL-4 or without exogenously added cytokines (21). Next, we analyzed IRF-1 expression in differentiating Th1 and Th2 cells (Fig. 6A, top panel). IRF-1 transcripts were detectable in naive CD4 ϩ T cells (Fig.  6A, lane 1) but were strongly up-regulated by TCR triggering (lane 2). IRF-1 mRNA was slightly more abundant in cells stimulated in the presence of Th1-inducing conditions than in cells stimulated in the presence of IL-4 or with anti-CD3 mAb FIG. 3. Cytokines induce binding of transcription factors to the GAS element in the IRF-1 promoter. Th1 and Th2 cells were harvested 6 days after priming, washed, and resuspended in complete medium. 10 7 cells were incubated at 37°C in 1 ml of complete medium with or without IL-12 (2.5 ng/ml) for 1 h or IFN-␣ (1000 units/ml), IFN-␥ (1000 units/ml), or IL-4 (1 ng/ml) for 15 min, followed by the preparation of whole cell extracts. Gel shift assays were performed with a 32 P end-labeled double stranded oligonucleotide corresponding to the GAS element present within the promoter of the IRF-1 gene, as described under "Experimental Procedures. "   FIG. 4. IL-12 induces binding of STAT-4 to the IRF-1 GAS element in human Th1 cells. Supershift assays were performed as described in Fig. 3; however, extracts were preincubated with antibodies specific for STAT molecules prior to the addition of the 32 P endlabeled double stranded oligonucleotide corresponding to the IRF-1 GAS element. Somewhat reduced band intensities were also observed in extracts preincubated with unrelated control antibodies (data not shown). The weaker signal in the samples pretreated with anti-STAT-1 and anti-STAT-3 antibodies therefore does not appear to be a specific effect of these antibodies.

FIG. 5. The IL-12-induced activation of an IRF-1 GAS reporter gene construct is mediated by STAT-4.
Jurkat T cells were transiently transfected with a reporter gene construct containing three copies of the GAS element present in the IRF-1 gene promoter. Where indicated, cells were co-transfected with expression vectors encoding IL-12R␤1, IL-12R␤2, and/or STAT-4. Cells were left untreated (unstim.) or were treated for 20 h with 2.5 ng/ml IL-12 (IL-12) prior to the preparation of lysates. Luciferase assays and normalization of results were performed as described under "Experimental Procedures." alone ( Fig. 6A, lanes 2-10). IL-4 does not suppress expression of IRF-1 transcripts to the same extent as it does with IL-12R␤2 transcripts. Only a slight down-regulation of IRF-1 mRNA is detectable when comparing cultures stimulated in the presence of IL-4 or with anti-CD3 mAb alone (Fig. 6A, compare lane 2 with 4 and lane 5 with 7). A quantitation of the data is shown in Fig. 6B. These results show that T cell receptor triggering strongly induces expression of IRF-1 irrespective of whether the stimulation was performed in the presence of Th1-or Th2inducing cytokines. DISCUSSION In this study, we analyzed whether the transcription factor IRF-1 plays a role in the differentiation of human naive CD4 ϩ T cells into polarized T helper cell subsets. Previous results obtained with IRF-1-deficient mice revealed a striking deficiency of Th1 cell development. Compromised Th1 development was shown to be associated with an impaired production of IL-12 by macrophages (33,34) and defective development of natural killer cells (41)(42)(43). It remained controversial whether compromised Th1 development also resulted from a defect of CD4 ϩ T cells from IRF-1 Ϫ/Ϫ mice to develop into Th1 effector cells. Cell transfer experiments indicated that IRF-1 Ϫ/Ϫ CD4 ϩ T cells could develop into Th1 cells when transferred to recombination activating gene-1-deficient (IRF-1 ϩ/ϩ ) mice, lacking T and B cells (34). However, the analysis of CD4 ϩ T cells purified from IRF-1 Ϫ/Ϫ TCR transgenic mice showed a deficiency in Th1 development when stimulated in vitro with peptide presented by wild-type antigen presenting cells (33). Here, we have examined IRF-1 expression and regulation along human T helper cell differentiation.
Analysis of IRF-1 transcripts in human Th1 and Th2 cells generated from cord blood leukocytes 2 weeks after stimulation revealed no significant differences in the two populations. However, after treatment with IL-12, IRF-1 transcripts were strongly induced in Th1 but not in Th2 cells. This finding demonstrates that IRF-1 is a target gene of IL-12. The IL-12induced up-regulation of IRF-1 transcripts could already be detected after 4 h and did not depend on de novo protein synthesis, arguing against a potential secondary effect mediated by IFN-␥ induced in response to IL-12. In addition, previous results have shown that Th1 cells do not respond to IFN-␥, because they do not express the IFN-␥R ␤ subunit (Fig. 3 and Refs. 51, 54, and 55).
Previous studies have demonstrated that induction of the IRF-1 gene in response to interferons and IL-6 is mediated by the binding of activated STAT-1 and STAT-3 to the palindromic GAS element present in the promoter of the IRF-1 gene (49,52,(57)(58)(59). In this report, we demonstrate that IL-12 induces IRF-1 gene transcription via activated STAT-4. We observed strong binding of an IL-12-induced factor to the GAS element of the IRF-1 promoter in extracts prepared from Th1 cells; conversely, IFN-␥ induced a gel shift only in Th2 cells. These results are consistent with the previously described selective expression of the IL-12 receptor ␤2 (IL-12R␤2) subunit and of the IFN-␥ receptor ␤ (IFN-␥R␤)-chain on Th1 and Th2 cells, respectively (21,51,(53)(54)(55). The IL-12-and IFN-␥-induced protein-DNA complexes migrated with a different mobility, indicating the binding of different STAT molecules to the IRF-1 GAS element. To determine which STAT molecules activated in response to IL-12 and IFN-␣ bound to the IRF-1 GAS element, we used specific antisera against STAT-1, STAT-3, and STAT-4 to supershift the protein-DNA complexes. These experiments revealed that the faster migrating complex formed in response to IFN-␣ contains mainly STAT-1, whereas the slower migrating band induced by IL-12 contains STAT-4. Binding of STAT-4 to the IRF-1 GAS element is also induced by IFN-␣, confirming the recently demonstrated IFN-␣-induced activation of STAT-4 in human mitogen-stimulated lymphoblasts (56) and in human Th1 and Th2 cells (50). Taken together, the gel shift experiments demonstrated that activated STAT-4 binds to the GAS element in the IRF-1 promoter. It is also of interest to note that the optimal recognition sequences for STAT-4, TTCCGGGAA (32) and (T/A)TTCC(C/G)GGAA(T/A) (60), as determined by binding site selection, are very similar to the DNA sequence of the GAS element in the human IRF-1 promoter, TTTC-CCCGAAA (49,52).
Transient transfection assays with a reporter gene construct containing three copies of the IRF-1 GAS element in Jurkat cells provided strong evidence that the transcriptional activation of the IRF-1 gene in response to IL-12 is mediated by STAT-4. Because Jurkat cells do not express detectable levels of functional IL-12 receptors and STAT-4 (data not shown), we co-transfected expression constructs encoding IL-12R␤1, IL-12R␤2, and STAT-4. A strong IL-12-dependent activation of the IRF-1 GAS reporter gene construct was observed when the components of the IL-12 signaling machinery are co-expressed in T cells.
To determine the role of IRF-1 in the differentiation of T helper subsets, we analyzed IRF-1 expression at different time intervals early after the stimulation of naive T cells under neutral, Th1-inducing, or Th2-inducing conditions. We com-FIG. 6. Antigen receptor triggering induces expression of IRF-1 transcripts in naive CD4 ؉ T cells. CD4 ϩ /CD45RO Ϫ T cells were purified by negative selection from cord blood as described under "Experimental Procedures." Purified CD4 ϩ /CD45RO Ϫ T cells (3 ϫ 10 6 ) were stimulated with plate-bound anti-CD3 mAb with or without the addition of IL-12 (2.5 ng/ml) or IL-4 (1 ng/ml). RNA was extracted from unstimulated CD4 ϩ /CD45RO Ϫ T cells (lane 1) or at the indicated time after CD3-stimulation. Transcripts encoding IRF-1 (top panel), IL-12R␤2 (middle panel), and 18 S RNA as loading control (bottom panel) were quantitated in RNase protection assays. B, the radioactivity present in the protected fragments was quantitated using a MolecularImager. The open and filled bars correspond to the relative expression levels of IL-12R␤2 and IRF-1 mRNAs, respectively. Results were normalized using an 18 S RNA fragment. pared IRF-1 expression to the expression of the IL-12R␤2 subunit, which is induced during differentiation of human naive cells along the Th1 but not the Th2 pathway (21,53). In contrast to IL-12R␤2, IRF-1 transcripts are detectable in naive CD4 ϩ T cells. Cross-linking of the TCR strongly induces IRF-1 mRNA even in the absence of exogenously added cytokines. At later time intervals after T cell stimulation, IRF-1 transcripts were slightly more abundant in cultures that were stimulated in the presence of IL-12. Consistent with the lack of functional IL-12 receptors, we did not observe any IL-12-induced upregulation of IRF-1 transcripts in naive T cells (data not shown). Taken together, these data show that activated STAT-4 is able to transactivate IRF-1 expression in differentiating Th1 cells accounting for a sustained expression of IRF-1.
In conclusion, our findings point to a potential hierarchy of transcriptional events during the differentiation of naive T cells into polarized Th1 cells; TCR ligation is an essential prerequisite for the initial expression of functional IL-12 receptors on developing Th1 cells, although the mechanisms and factors regulating this phase have not yet been uncovered. In the second phase, IL-12 induces further up-regulation of the IL-12R␤2, IRF-1, and other, yet to be identified, IL-12-regulated genes. The rapid tyrosine phosphorylation and activation of STAT-4 by IL-12 and the phenotype of the STAT-4 Ϫ/Ϫ mice predict that this phase is regulated by STAT-4. In a third phase, transcription factors, such as IRF-1, and other, so far unknown, regulatory proteins induced by STAT-4 may themselves regulate target genes that are important for the effector functions of Th1 cells. IRF-1 appears to be the first member of a probably large family of regulatory proteins induced by IL-12. It will be of great interest to identify additional members of this family and to analyze their functions with respect to the differentiation and effector functions of T helper cells.