Interleukin-5 (IL-5), which is produced by CD4+ T helper 2 (Th2) cells, but not by Th1 cells, plays a key role in the development of eosinophilia in asthma. Despite increasing evidence that the outcome of many diseases is determined by the ratio of the two subsets of CD4+ T helper cells, Th1 and Th2, the molecular basis for Th1- and Th2-specific gene expression remains to be elucidated. We previously established a critical role for the transcription factor GATA-3 in IL-5 promoter activation in EL-4 cells, which express both Th1- and Th2-type cytokines. Our studies reported here demonstrate that GATA-3 is critical for expression of the IL-5 gene in bona fide Th2 cells. Whereas mutations in the GATA-3 site abolished antigen- or cAMP-stimulated IL-5 promoter activation in Th2 cells, ectopic expression of GATA-3 in Th1 cells or in a non-lymphoid, non-IL-5-producing cell line activated the IL-5 promoter. During the differentiation of naive CD4+ T cells isolated from T cell receptor transgenic mice, GATA-3 gene expression was up-regulated in developing Th2 cells, but was down-regulated in Th1 cells, and antigen- or cAMP-activated Th2 cells (but not Th1 cells) expressed the GATA-3 protein. Thus, GATA-3 may play an important role in the balance between Th1 and Th2 subsets in immune responses. Inhibition of GATA-3 activity has therapeutic potential in the treatment of asthma and other hypereosinophilic diseases. Interleukin-5 (IL-5), which is produced by CD4+ T helper 2 (Th2) cells, but not by Th1 cells, plays a key role in the development of eosinophilia in asthma. Despite increasing evidence that the outcome of many diseases is determined by the ratio of the two subsets of CD4+ T helper cells, Th1 and Th2, the molecular basis for Th1- and Th2-specific gene expression remains to be elucidated. We previously established a critical role for the transcription factor GATA-3 in IL-5 promoter activation in EL-4 cells, which express both Th1- and Th2-type cytokines. Our studies reported here demonstrate that GATA-3 is critical for expression of the IL-5 gene in bona fide Th2 cells. Whereas mutations in the GATA-3 site abolished antigen- or cAMP-stimulated IL-5 promoter activation in Th2 cells, ectopic expression of GATA-3 in Th1 cells or in a non-lymphoid, non-IL-5-producing cell line activated the IL-5 promoter. During the differentiation of naive CD4+ T cells isolated from T cell receptor transgenic mice, GATA-3 gene expression was up-regulated in developing Th2 cells, but was down-regulated in Th1 cells, and antigen- or cAMP-activated Th2 cells (but not Th1 cells) expressed the GATA-3 protein. Thus, GATA-3 may play an important role in the balance between Th1 and Th2 subsets in immune responses. Inhibition of GATA-3 activity has therapeutic potential in the treatment of asthma and other hypereosinophilic diseases. Activated CD4+ T cells are subdivided into two subsets, T helper 1 (Th1) and Th2, based on their biological functions, which, in turn, depend on the cytokines they produce (1Bottomly K. Immunol. Today. 1988; 9: 268-273Abstract Full Text PDF PubMed Scopus (352) Google Scholar, 2Mossman T.R. Coffman R.L. Annu. Rev. Immunol. 1989; 7: 145-173Crossref PubMed Scopus (7139) Google Scholar, 3Janeway Jr., C.A. Bottomly K. Cell. 1994; 76: 275-285Abstract Full Text PDF PubMed Scopus (783) Google Scholar, 4Abbas A.K. Murphy K.M. Sher A. Nature. 1996; 383: 787-793Crossref PubMed Scopus (3917) Google Scholar). Th1 cells produce interleukin-2 (IL-2) 1The abbreviations used are: IL, interleukin; IFN-γ, interferon-γ; Bt2cAMP, dibutyryl cyclic AMP; EMSAs, electrophoretic mobility shift assays; bp, base pair; kb, kilobase; TCR, T cell receptor; APCs, antigen-presenting cells; Ag, antigen; ELISA, enzyme-linked immunosorbent assay; PMA, phorbol 12-myristate 13-acetate; Ab, antibody; TPA, 12-O-tetradecanoylphorbol-13-acetate.and interferon-γ (IFN-γ) and stimulate microbicidal activity in macrophages and promote cell-mediated immunity (1Bottomly K. Immunol. Today. 1988; 9: 268-273Abstract Full Text PDF PubMed Scopus (352) Google Scholar, 2Mossman T.R. Coffman R.L. Annu. Rev. Immunol. 1989; 7: 145-173Crossref PubMed Scopus (7139) Google Scholar, 3Janeway Jr., C.A. Bottomly K. Cell. 1994; 76: 275-285Abstract Full Text PDF PubMed Scopus (783) Google Scholar, 4Abbas A.K. Murphy K.M. Sher A. Nature. 1996; 383: 787-793Crossref PubMed Scopus (3917) Google Scholar). Th2 cells, on the other hand, produce IL-4 and IL-5, which stimulate IgE production and eosinophilic inflammation, respectively (1Bottomly K. Immunol. Today. 1988; 9: 268-273Abstract Full Text PDF PubMed Scopus (352) Google Scholar, 2Mossman T.R. Coffman R.L. Annu. Rev. Immunol. 1989; 7: 145-173Crossref PubMed Scopus (7139) Google Scholar, 3Janeway Jr., C.A. Bottomly K. Cell. 1994; 76: 275-285Abstract Full Text PDF PubMed Scopus (783) Google Scholar, 4Abbas A.K. Murphy K.M. Sher A. Nature. 1996; 383: 787-793Crossref PubMed Scopus (3917) Google Scholar). There is good evidence that in atopic asthmatics, a Th2-type response occurs in the airways (5Walker C. Virchow J. Bruijnzeel P.L.B. Blaser K. J. Immunol. 1991; 1990: 1829-1835Google Scholar, 6Hamid Q. Azzawi M. Ying S. Moqbel R. Wardlaw A.J. Corrigan C.J. Bradley B. Durham S.R. Collins J.V. Jeffery P.K. Quint D.J. Kay A.B. J. Clin. Invest. 1991; 87: 1541-1546Crossref PubMed Scopus (747) Google Scholar, 7Robinson D.S. Hamid Q. Ying S. Tsicopoulos A. Barkans J. Bentley A.M. Corrigan C. Durham S.R. Kay A.B. N. Engl. J. Med. 1992; 326: 298-304Crossref PubMed Scopus (2601) Google Scholar, 8Krishnaswamy G. Liu M.C. Su S. Kumai M. Xiao H. Marsh D.G. Huang S. Am. J. Respir. Cell Mol. Biol. 1993; 9: 279-286Crossref PubMed Scopus (79) Google Scholar). Although it appears that the outcome of many diseases such as asthma is determined by the ratio of Th1 to Th2 cells (4Abbas A.K. Murphy K.M. Sher A. Nature. 1996; 383: 787-793Crossref PubMed Scopus (3917) Google Scholar), the molecular basis for Th1- and Th2-specific gene expression remains to be elucidated. Asthma is a chronic obstructive disease of the small airways. Evolving evidence indicates that asthma is the result of an inflammatory process interacting with a susceptible airway best defined at present as airway hyperresponsiveness (9Martin L.B. Kita H. Leiferman K.M. Gleich G.J. Int. Arch. Allergy Immunol. 1996; 109: 207-215Crossref PubMed Scopus (174) Google Scholar). In asthma, the most striking and consistent pathophysiology is damage to the bronchial epithelium caused by cytotoxic cationic proteins released by infiltrating eosinophils (10Frick W.E. Sedgwick J.B. Busse W.W. Am. Rev. Respir. Dis. 1989; 139: 1401-1406Crossref PubMed Scopus (108) Google Scholar,11Bousquet J. Chanez P. Lacoste J.Y. Barneon G. Ghavanian N. Enander I. Venge P. Ahlstedt S. Simony-Lafontaine J. Godard P. Michel F.B. N. Engl. J. Med. 1990; 323: 1033-1039Crossref PubMed Scopus (2267) Google Scholar). Various lines of evidence indicate that secreted products of activated T cells, such as the cytokine IL-5, play a central role in orchestrating the unique inflammatory response seen in asthma. Since it was isolated and cloned in the mid-1980s, the intimate relationship between IL-5, eosinophils, and asthma has been extensively documented (12Sanderson C.J. Blood. 1992; 79: 3101-3109Crossref PubMed Google Scholar). IL-5 has multiple effects on the biology of eosinophils not limited to differentiation, proliferation, recruitment, and activation (12Sanderson C.J. Blood. 1992; 79: 3101-3109Crossref PubMed Google Scholar). Increasing evidence places IL-5 in a key role in the development of eosinophilia in asthma (5Walker C. Virchow J. Bruijnzeel P.L.B. Blaser K. J. Immunol. 1991; 1990: 1829-1835Google Scholar, 6Hamid Q. Azzawi M. Ying S. Moqbel R. Wardlaw A.J. Corrigan C.J. Bradley B. Durham S.R. Collins J.V. Jeffery P.K. Quint D.J. Kay A.B. J. Clin. Invest. 1991; 87: 1541-1546Crossref PubMed Scopus (747) Google Scholar). IL-5 mRNA was significantly enhanced in bronchoalveolar lavage cells obtained from asthmatics challenged with ragweed antigen (8Krishnaswamy G. Liu M.C. Su S. Kumai M. Xiao H. Marsh D.G. Huang S. Am. J. Respir. Cell Mol. Biol. 1993; 9: 279-286Crossref PubMed Scopus (79) Google Scholar). Again, peripheral blood T cells from asthmatics were found to secrete IL-5 in response to the common house dust mite (Dermatophagoides farinae) antigen (13Kamei T. Ozaki T. Kawaji K. Banno K. Sano T. Azuma M. Ogura T. Am. J. Respir. Cell Mol. Biol. 1993; 9: 378-385Crossref PubMed Scopus (13) Google Scholar). Most striking, in ovalbumin-sensitized guinea pigs and mice, monoclonal antibody to IL-5 decreased pulmonary eosinophilia and prevented the development of airway hyperresponsiveness (14Gulbenkian A.R. Egan R.W. Fernandez X. Jones H. Kreutner W. Kung T. Payvandi F. Sullivan L. Zurcher J.A. Watnick A.S. Am. Rev. Respir. Dis. 1991; 146: 263-265Crossref Scopus (130) Google Scholar, 15Nakajima H. Iwamoto I. Tomoe S. Matsumura R. Tomioka H. Takatsu K. Yoshida S. Am. Rev. Respir. Dis. 1992; 146: 374-377Crossref PubMed Scopus (377) Google Scholar). Also, in a mouse model of asthma, IL-5-deficient mice were found to lack eosinophilia, lung pathology (16Foster P.S. Hogan S.P. Ramsay A.J. Matthaei K.I. Young I.G. J. Exp. Med. 1996; 183: 195-201Crossref PubMed Scopus (1288) Google Scholar, 17Kopf M. Brombacher F. Hodgkin P.D. Ramsay A.J. Milbourne E.A. Dai W.J. Ovington K.S. Behm C.A. Kohler G. Young I.G. Matthaei K.I. Immunity. 1996; 4: 15-24Abstract Full Text Full Text PDF PubMed Scopus (545) Google Scholar), and airway hyperresponsiveness upon allergen challenge (16Foster P.S. Hogan S.P. Ramsay A.J. Matthaei K.I. Young I.G. J. Exp. Med. 1996; 183: 195-201Crossref PubMed Scopus (1288) Google Scholar). In both humans and mice, the production of IL-5 is restricted to a few cell types, which include T cells (7Robinson D.S. Hamid Q. Ying S. Tsicopoulos A. Barkans J. Bentley A.M. Corrigan C. Durham S.R. Kay A.B. N. Engl. J. Med. 1992; 326: 298-304Crossref PubMed Scopus (2601) Google Scholar), mast cells (18Plaut M. Pierce J.H. Watson C.J. Hanley-Hyde J. Nordan R.P. Paul W.E. Nature. 1989; 339: 64-67Crossref PubMed Scopus (1145) Google Scholar), and eosinophils (19Broide D.H. Paine M.M. Firestein G.S. J. Clin. Invest. 1992; 90: 1414-1424Crossref PubMed Scopus (332) Google Scholar), the predominant source being T cells of the Th2-type (7Robinson D.S. Hamid Q. Ying S. Tsicopoulos A. Barkans J. Bentley A.M. Corrigan C. Durham S.R. Kay A.B. N. Engl. J. Med. 1992; 326: 298-304Crossref PubMed Scopus (2601) Google Scholar). In general, IL-5 is not produced constitutively by Th2 cells. IL-5 gene expression has been shown to be stimulated by antigen, mitogens (concanavalin A), eicosanoid compounds (leukotriene B4 and prostaglandins), and cytokines (20Bohjanen P.R. Okajima M. Hodes R.J. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 5283-5287Crossref PubMed Scopus (91) Google Scholar, 21Lee H.J. Koyano-Nakagawa N. Naito Y. Nishida J. Arai N. Arai K. Yokota T. J. Immunol. 1993; 151: 6135-6142PubMed Google Scholar). Intracellular cAMP-increasing agents, such as IL-1α, prostaglandin E2, and the cAMP analogue dibutyryl cyclic AMP (Bt2cAMP), have been shown to differentially regulate cytokine production by Th1 and Th2 cells. Whereas the production of the Th1 response-inducing cytokine IL-12 and that of the Th1 cytokines IL-2 and IFN-γ are inhibited by cAMP-increasing agents, the production of IL-5 is strongly induced by the same agents, suggesting a possible immunoregulatory role for this second messenger (21Lee H.J. Koyano-Nakagawa N. Naito Y. Nishida J. Arai N. Arai K. Yokota T. J. Immunol. 1993; 151: 6135-6142PubMed Google Scholar, 22van der Pouw Kraan T.C.T.M. Boeije L.C.M. Smeenk R.J.T. Wijdenes J. Aarden L.A. J. Exp. Med. 1995; 181: 775-779Crossref PubMed Scopus (639) Google Scholar, 23Munoz E. Zubiaga A.M. Merrow M. Sauter N.P. Huber B.T. J. Exp. Med. 1990; 1990: 95-103Crossref Scopus (249) Google Scholar, 24Snijdewint F.G.M. Kalinski P. Wierenga E.A. Bos J.D. Kapsenberg M.L. J. Immunol. 1993; 150: 5321-5329PubMed Google Scholar). The molecular mechanisms underlying Th2 cell-specific IL-5 gene expression are unclear. In our previous studies of IL-5 promoter activation by cAMP in the murine cell line EL-4, which expresses both Th2- and Th1-type cytokines, we showed that deletion of the IL-5 promoter to −66, which disrupted a GATA site located between −70 and −60, abolished activation of the promoter (25Siegel M.D. Zhang D.-H. Ray P. Ray A. J. Biol. Chem. 1995; 270: 24548-24555Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar). Furthermore, in electrophoretic mobility shift assays (EMSAs), we demonstrated that the transcription factor GATA-3, but not GATA-4, binds to this GATA site (25Siegel M.D. Zhang D.-H. Ray P. Ray A. J. Biol. Chem. 1995; 270: 24548-24555Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar). This was the first description of the involvement of GATA-3 in the transcription of any cytokine gene (25Siegel M.D. Zhang D.-H. Ray P. Ray A. J. Biol. Chem. 1995; 270: 24548-24555Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar). Yamagata et al.(26Yamagata T. Nishida J. Sakai R. Tanaka T. Honda H. Hirano N. Mano H. Yazaki Y. Hirai H. Mol. Cell. Biol. 1995; 15: 3830-3839Crossref PubMed Scopus (80) Google Scholar), in their studies of transcription of the human IL-5 gene in the ATL-16T cell line, also demonstrated the importance of the GATA site in expression of the human IL-5 gene. However, two important points merit consideration in comparing these two studies. First, IL-5 gene expression in ATL-16T cells is largely constitutive (27Noma T. Nakakubo H. Sugita M. Kumagai S. Maeda M. Shimizu A. Honjo T. J. Exp. Med. 1989; 169: 1853-1858Crossref PubMed Scopus (69) Google Scholar). 2D.-H. Zhang and A. Ray, unpublished observations.However, in both humans and mice, the IL-5 gene is expressed in an inducible fashion, and therefore, the ATL-16T cells do not reflect the typical expression characteristics of the IL-5 gene either in humans or in mice. Second, GATA-4 is predominantly expressed in the heart, intestines, epithelium, and reproductive organs, and its expression is low or undetectable in both human and murine T cells (28Arceci R.J. King A.A.J. Simon M.C. Orkin S.H. Wilson D.B. Mol. Cell. Biol. 1993; 13: 2235-2246Crossref PubMed Google Scholar, 29Laverriere A.C. MacNeill C. Mueller C. Poelmann R.E. Burch J.B.E. Evans T. J. Biol. Chem. 1994; 269: 23177-23184Abstract Full Text PDF PubMed Google Scholar). Therefore, the atypical high level of GATA-4 activity in ATL-16T cells may contribute to the atypical (constitutive) nature of IL-5 gene expression in these cells. In another study, Prieschl et al.(30Prieschl E.E. Goouilleux-Gruart V. Walker C. Harrer N.E. Baumruker T. J. Immunol. 1995; 154: 6112-6119PubMed Google Scholar) showed that the GATA site located between −70 and −60 in the IL-5 promoter is also important for IL-5 gene expression in mast cells. In our previous study, we additionally demonstrated that activation of the IL-5 promoter also requires an intact AP-1 site within the CLE0 element (consensus lymphokineelement 0) located between −53 and −39 in the promoter; mutation of this site in the context of an ∼550-bp promoter totally abrogated promoter activity. In this report, we show that the transcription factor GATA-3 is crucial for IL-5 gene expression in bona fide Th2 cells and that ectopic expression of GATA-3 alone results in IL-5 promoter activation in a non-IL-5-producing cell line. We also show that GATA-3 activity is present only in Th2 cells and is undetectable in Th1 cells. Inhibition of GATA-3 activity may therefore be effective in the treatment of asthma and other hypereosinophilic diseases. Both D10 and C19 clones were maintained in Click's medium supplemented with 10% fetal bovine serum, 5 units/ml murine recombinant IL-2 (Boehringer Mannheim), 50 μmβ-mercaptoethanol, 2 mml-glutamine, and 50 μg/ml gentamycin at 37 °C with 5% CO2 (31Kaye J. Porcelli S. Tite J. Jones B. Janeway J.C.A. J. Exp. Med. 1983; 158: 836-856Crossref PubMed Scopus (629) Google Scholar, 32Baron J.L. Madri J.A. Ruddle N.H. Hashim G. Janeway J.C.A. J. Exp. Med. 1993; 177: 57-68Crossref PubMed Scopus (793) Google Scholar). The cells were stimulated every 2 weeks with the specific antigen (conalbumin for D10 cells, used at 100 μg/ml) and peptide AC1–16 (ASQKRPSQRHGSKYL; derived from myelin basic protein, used at 5 μg/ml) and mitomycin C-treated splenocytes from syngeneic mice (I-Ak for D10 cells and I-Au for C19 cells). Prior to use in experiments, dead cells were removed by density gradient fractionation using lymphocyte separation medium (Organon Teknika). DO11.10 mice, which are transgenic for the TCR recognizing the ovalbumin peptide 323–339 (pOVA323–339), were provided on BALB/c background by Dr. Ken Murphy (Washington University, St. Louis, MO). To generate Th1 or Th2 cells from DO11.10 mice, naive CD4+ T cells were first isolated from the spleens by negative selection using monoclonal antibodies to CD8, class II MHC I-Ad, and anti-Ig-coated magnetic beads (Collaborative Research). Cultures were set up in flasks containing equal numbers of CD4+ T cells and T cell-depleted APCs at a concentration of 2 × 106 cells/ml. To generate Th1 cells, cultures contained pOVA323–339 at 5 μg/ml, IL-12 at 5 ng/ml, IL-2 at 10 units/ml, and anti-IL-4 at inhibitory concentrations. To generate Th2 cells, cultures contained pOVA323–339 at 5 μg/ml, IL-4 at 200 units/ml, IL-2 at 10 units/ml, and anti-IFN-γ antibody. Cells were maintained in culture for 3 days, and at the end of this period, cells were further stimulated with fresh mitomycin C-treated and T cell-depleted APCs and Ag for 8 or 24 h (for making nuclear extracts) or for 48 h for cytokine assays. Culture supernatants were assayed for the presence of cytokines by ELISA using kits from Endogen, Inc. (sensitivity: IL-4, 6 pg/ml; IL-5, 0.1 ng/ml; and IFN-γ, 2 ng/ml). Total cellular RNA was prepared by using Trizol (Life Technologies, Inc.) according to the instructions of the manufacturer. 10 μg of total RNA from each sample was fractionated on a formaldehyde-agarose gel and transferred to a nylon membrane. DNA fragments derived from murineGATA-3 cDNA (∼60-bp BglI-ClaI fragment not containing any part of the zinc finger domain) were labeled with [α-32P]dCTP using a random primer DNA labeling kit (Boehringer Mannheim). Hybridization was performed using QuikHyb (Stratagene) according to the instructions of the manufacturer. Rested D10 or A.E7 cells were washed once in serum-free RPMI 1640 medium and resuspended in the same medium. Cells (5 × 106) were incubated with 15 μg of DNA (5 μg of reporter plasmid, 2 μg of cytomegalovirus-β-galactosidase plasmid as a monitor for transfection efficiency, and carrier plasmid pGEM7Z to make up to 15 μg of total DNA) for 10 min at room temperature, and electroporation was carried out using a GenePulser (Bio-Rad) at 0.27 kV and 960 microfarads. The cells were left on ice for 10–30 min, diluted to 5 ml with fresh medium, and incubated at 37 °C with or without Bt2cAMP + PMA. For antigen stimulation, rested cells were first stimulated with conalbumin and mitomycin C-treated and T cell-depleted APCs in complete medium containing 5 units/ml IL-2 for 72 h and then subjected to electroporation. Cells were harvested for reporter gene assays as described previously (25Siegel M.D. Zhang D.-H. Ray P. Ray A. J. Biol. Chem. 1995; 270: 24548-24555Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar). HeLa cells were transfected as described previously (33Ray A. LaForge K.S. Sehgal P.B. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 7086-7090Crossref PubMed Scopus (34) Google Scholar, 34Ray A. Prefontaine K.E. Proc. Natl. Acad. Sci. U. S. A. 1994; 86: 752-756Crossref Scopus (937) Google Scholar). Cells were left unstimulated or were stimulated as described above. All APCs were mitomycin C-treated and depleted of T cells. Nuclear extracts were prepared as described previously (25Siegel M.D. Zhang D.-H. Ray P. Ray A. J. Biol. Chem. 1995; 270: 24548-24555Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar). The probes in the EMSAs were two double-stranded oligonucleotides containing sequences between −57 and −34 (containing the CLE0 element) and between −73 and −54 (containing the GATA element) in the IL-5 gene, and 22-bp oligonucleotides containing the consensus CREB element (from Stratagene). The oligonucleotides for the mutant CREB element were purchased from Santa Cruz Biotechnology. The sequences of the oligonucleotides used in the EMSAs were as follows:−73CCTCTATCTGATTGTTAGCA−54 (wild-type GATA), CCTCgcgaTGATTGTTAGCA (GATA mutant 1), CCTCTATCTGAaaccTAGCA (GATA mutant 2), CCTCTATCcttTTGTTAGCA (GATA mutant 3), and−57AGCAATTATTCATTTCCTCAGAGA−34 (CLE0). Complementary oligonucleotides were annealed before use in EMSAs. The antibodies to the c-Jun, JunB, JunD, and GATA-3 proteins were purchased from Santa Cruz Biotechnology. The anti-GATA-3 antibody was a mouse monoclonal IgG1 that does not cross-react with GATA-1, GATA-2, or GATA-4. The anti-Fos antibody was purchased from Oncogene Science Inc. The anti-GATA-4 antibody was kindly provided by Dr. David Wilson. The competitor oligonucleotides were added at a 100-fold molar excess. The binding reactions were analyzed by electrophoresis on 6% native polyacrylamide gels (acrylamide/bisacrylamide = 30:1). Electrophoresis was carried out at 200 V in 0.5 × TBE (1 × TBE = 0.05 m Tris base, 0.05 m boric acid, and 1.0 mm EDTA) at 4 °C. Gels were dried and subjected to autoradiography. We have used the established nontransformed murine T cell clones D10.G4.1 (Th2) (31Kaye J. Porcelli S. Tite J. Jones B. Janeway J.C.A. J. Exp. Med. 1983; 158: 836-856Crossref PubMed Scopus (629) Google Scholar) and C19 (Th1) (32Baron J.L. Madri J.A. Ruddle N.H. Hashim G. Janeway J.C.A. J. Exp. Med. 1993; 177: 57-68Crossref PubMed Scopus (793) Google Scholar) and Th1 and Th2 cells obtained by differentiation of naive CD4+ T cells from DO11.10 TCR transgenic mice to gain insight into mechanisms that permit IL-5 gene expression in Th2 cells but limit its expression in Th1 cells. In our previous studies of cAMP-induced IL-5 promoter activation using an ∼550-bp promoter fragment and the murine T cell line EL-4, we had identified two regions in the IL-5 5′-flanking region that were critical for induction of the IL-5 promoter: one was the AP-1-binding site within the CLE0 element, whereas the other was a region between −70 and −60 containing two overlapping GATA sites, deletion of which abrogated activation of the promoter (25Siegel M.D. Zhang D.-H. Ray P. Ray A. J. Biol. Chem. 1995; 270: 24548-24555Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar). To elucidate the molecular mechanisms underlying transcriptional activation of the IL-5 gene by Ag in Th2 cells, we transfected the murine Th2 clone D10.G4.1 (31Kaye J. Porcelli S. Tite J. Jones B. Janeway J.C.A. J. Exp. Med. 1983; 158: 836-856Crossref PubMed Scopus (629) Google Scholar) with a reporter gene (firefly luciferase) construct containing a 1.7-kb promoter fragment from the 5′-flanking region immediately upstream of the transcriptional start site of the IL-5 gene. Both Ag and Bt2cAMP + PMA caused a 10–20-fold activation of the wild-type IL-5 promoter, and mutations in the AP-1 site or the GATA site in the context of the 1.7-kb promoter completely abolished activation of the IL-5 promoter (Fig. 1 A). Mutations in the NF-AT site, on the other hand, had no effect on IL-5 promoter activity. We next investigated the binding of nuclear proteins to the GATA element and the CLE0 element using nuclear extracts from Th2 (D10) cells. As shown in Fig. 2 A (lane 1), two complexes were detected using nuclear extracts from unstimulated D10 cells. The binding intensity of both complexes was augmented upon stimulation of the cells with Bt2cAMP + PMA (lane 2). Although both complexes were competed for by an excess of the unlabeled wild-type oligonucleotide, competition for complex I formation was incomplete even with a 100-fold molar excess of the unlabeled competitor, suggesting that complex I binds with a lower affinity to the GATA site than complex II (lane 3). Oligonucleotides containing specific mutations in three different regions of the double GATA site were also used as competitors. Mutant 1 contained mutations in the distal GATA sequence, and mutant 2 in the proximal sequence, whereas mutant 3 was mutated in both sequences. None of these mutants was able to compete for formation of the complexes, suggesting the involvement of the entire sequence between −70 and −60 in the formation of the complexes (lanes 4–6). The complexes (especially complex II) were supershifted by an anti-GATA-3 antibody (Ab) (lane 7), but not by an anti-GATA-4 antiserum (lane 9). To compare GATA-3 DNA binding activity between Th1 and Th2 cells, we performed a similar analysis with nuclear extracts from both D10 and C19 (Th1) cells prepared in the same experiment under identical conditions. As shown in Fig. 2 B (lower panel), nuclear proteins from induced D10 cells generated two complexes that were supershifted by the anti-GATA-3 Ab (lane 3), but not by the anti-GATA-4 Ab (lane 4). Using identical protein amounts of nuclear extracts prepared from C19 cells, we detected a very low level of binding activity in resting cells (Fig. 2 B, upper panel, lane 1). Upon treatment of the C19 cells with Bt2cAMP and PMA, whether alone (data not shown) or in combination, the intensity of the complexes did not increase, but consistently diminished (Fig. 2 B, upper panel, compare lane 2 withlane 1). We then tested the same extracts for binding to the AP-1 site within the CLE0 element. As shown in Fig. 2 B, used at only half the amounts used in the GATA-3 binding assays, robust inducible binding activity was detected with both D10 and C19 nuclear extracts. The anti-c-Jun and anti-c-Fos (reactive to all Fos family proteins) Abs did not affect the formation of the complex with either extract (lanes 7 and 10, respectively), whereas the anti-JunB and anti-JunD Abs supershifted/inhibited complex formation (lanes 8 and 9, respectively). Taken together, these results demonstrated that D10 and C19 cells contain similar levels of AP-1 binding activity. However, D10 cells constitutively contain some GATA-3 DNA binding activity that is augmented upon stimulation of the cells. In contrast, unstimulated C19 cells contain very little GATA-3 DNA binding activity that decreases upon stimulation of the cells. We investigated whether the observed difference in GATA-3 binding activity between the Th1 (C19) and Th2 (D10) clones was also true in Th1 and Th2 cells obtained by differentiation of naive splenic CD4+ T cells from DO11.10 TCR transgenic mice. Nuclear extracts were prepared from the Th1 and Th2 cells stimulated with Ag for 8 or 24 h, and GATA and AP-1 binding activities were determined with the extracts. The Th1 cell extract from both time points generated two complexes with the GATA probe (Fig. 3 A, lanes 1–6), whereas the Th2 extract generated three complexes (lanes 7–12). Complexes I and II were of the same mobility and reactivity to antisera as complexes I and II illustrated in Fig. 2. Complex II was not formed with the Th1 extract. A new complex (III), specific for Ag stimulation and of mobility intermediate between complexes I and II, was formed with both Th1 and Th2 extracts. Complex III was more distinct with the 24-h extracts (Fig. 3 A,left panel, compare lane 4 with lane 1 and lane 10 with lane 7). Complex III, formed with both the Th1 extracts (lanes 3 and 6) and the Th2 extracts (lanes 9 and 12), was inhibited by the anti-GATA-4 antiserum. However, the anti-GATA-3 Ab did not react with any of the complexes formed with the Th1 extracts (lanes 2and 5). By contrast, in our EMSAs with the Th2 extracts, complex I was slightly inhibited by the anti-GATA-3 Ab, and complex II, only formed with Th2 nuclear proteins, was supershifted by the anti-GATA-3 Ab (lanes 8 and 11). Essentially identical data were obtained in EMSAs with nuclear extracts from Ag-stimulated D10 and C19 cells (data not shown). We have also explored GATA-3 binding activity in another Th1-type clone, A.E7 (35Beverly B. Kang S.M. Lenardo M.J. Schwartz R.H. Int. Immunol. 1992; 4: 661-671Crossref PubMed Scopus (336) Google Scholar). No IL-5 mRNA was detected in Northern blot analyses of RNA prepared from A.E7 cells that were stimulated for 24 h with Ag (data not shown). Also, nuclear extracts prepared from Ag-induced A.E7 cells had no detectable GATA-3 binding activity (data not shown). We do not know the exact composition of the different complexes that are formed with Th2 nuclear extracts and the IL-5 GATA site. Complex II could represent a higher order form (dimer or tetramer) of complex I that may contain a monomer or dimer of GATA-3. The oligonucleotide competition experiments suggest that complex I binds to the GATA site with a lower affinity than complex II. Also, formation of complex II was consistently more sensitive to the anti-GATA-3 antibody than formation of complex I, suggesting that the epitope recognized by the anti-GATA-3 monoclonal antibody in the GATA-3 protein is more accessible in complex II. Taken together, it appears that activated Th1 cells lack GATA-3 DNA binding activity. The significance of the reactivity of complex III to the anti-GATA-4 antiserum is unclear at the present time since the expression of GATA-4 has only been described in the heart, intestines, and gonads (28Arceci R.J. King A.A.J. Simon M.C. Orkin S.H. Wilson D.B. Mol. Cell. Biol. 1993; 13: 2235-2246Crossref PubMed Google Scholar, 29Laverriere A.C. MacNeill C. Mueller C. Poelmann R.E. Burch J.B.E. Evans T. J. Biol. Chem. 1994; 269: 23177-2318