GATA Factors Are Essential for Transcription of the Survival Gene E4bp4 and the Viability Response of Interleukin-3 in Ba/F3 Hematopoietic Cells
2002; Elsevier BV; Volume: 277; Issue: 30 Linguagem: Inglês
10.1074/jbc.m200924200
ISSN1083-351X
AutoresYung‐Luen Yu, Yun-Jung Chiang, J. J. Yen,
Tópico(s)interferon and immune responses
ResumoE4bp4, a member of the basic region/leucine zipper transcription factor superfamily, is up-regulated by the interleukin-3 (IL-3) signaling pathway and plays an important role in the anti-apoptotic response of IL-3. In this study, we demonstrated that E4bp4 is regulated by IL-3 mainly at the transcriptional level. Promoter analysis revealed that a GATA motif downstream of a major transcription initiation site is essential forE4bp4 expression in the IL-3-dependent Ba/F3 cell line. Gel shift assays demonstrated that both GATA-1 and GATA-2 proteins bind to the E4bp4 GATA site in vitro, and the chromatin immunoprecipitation assay further confirmed thein vivo binding of GATA-1 to the E4bp4promoter. Overexpression of GATA-1 alone transactivates theE4bp4 reporter, whereas transactivation of theE4bp4 reporter by GATA-2 is dependent on the stimulation of IL-3. Last, we demonstrated that alteration of GATA-1 binding to the GATA site by stably overexpressing GATA-1 or a GATA-1 mutant containing only the DNA-binding domain not only modulates the expression of theE4bp4 gene but also influences apoptosis induced by IL-3 removal. Taken together, our results suggest that the GATA factors play an important role in transducing the survival signal of IL-3, and one of their cellular targets is E4bp4. E4bp4, a member of the basic region/leucine zipper transcription factor superfamily, is up-regulated by the interleukin-3 (IL-3) signaling pathway and plays an important role in the anti-apoptotic response of IL-3. In this study, we demonstrated that E4bp4 is regulated by IL-3 mainly at the transcriptional level. Promoter analysis revealed that a GATA motif downstream of a major transcription initiation site is essential forE4bp4 expression in the IL-3-dependent Ba/F3 cell line. Gel shift assays demonstrated that both GATA-1 and GATA-2 proteins bind to the E4bp4 GATA site in vitro, and the chromatin immunoprecipitation assay further confirmed thein vivo binding of GATA-1 to the E4bp4promoter. Overexpression of GATA-1 alone transactivates theE4bp4 reporter, whereas transactivation of theE4bp4 reporter by GATA-2 is dependent on the stimulation of IL-3. Last, we demonstrated that alteration of GATA-1 binding to the GATA site by stably overexpressing GATA-1 or a GATA-1 mutant containing only the DNA-binding domain not only modulates the expression of theE4bp4 gene but also influences apoptosis induced by IL-3 removal. Taken together, our results suggest that the GATA factors play an important role in transducing the survival signal of IL-3, and one of their cellular targets is E4bp4. interleukin electrophoretic mobility shift assay hemagglutinin chromatin immunoprecipitation GATA family proteins are a group of transcription factors containing two related zinc fingers that mediate DNA binding (1Martin D.I. Orkin S.H. Genes Dev. 1990; 4: 1886-1898Crossref PubMed Scopus (330) Google Scholar). Among the six known members, GATA-1, GATA-2, and GATA-3 are preferentially expressed in hematopoietic cells, whereas the other GATA factors are expressed exclusively in nonhematopoietic tissues. GATA-1 is highly expressed in mature erythroid cells, mast cells, and megakaryocytes (2Yamamoto M., Ko, L.J. Leonard M.W. Beug H. Orkin S.H. Engel J.D. Genes Dev. 1990; 10: 1650-1662Crossref Scopus (451) Google Scholar, 3Martin D.I. Zon L.I. Mutter G. Orkin S.H. Nature. 1990; 344: 4444-4447Crossref Scopus (343) Google Scholar, 4Romeo P.H. Prandini M.H. Joulin V. Mignotte V. Prenant M. Vainchenker W. Marguerie G. Uzan G. Nature. 1990; 344: 447-449Crossref PubMed Scopus (321) Google Scholar) and is expressed at a lower level in progenitor cells (5Sposi N.M. Zon L.I. Care A. Valtieri M. Testa U. Gabbianelli M. Mariani G. Bottero L. Mather C. Orkin S.H. Peschle C. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6353-6357Crossref PubMed Scopus (122) Google Scholar, 6Leonard M. Brice M. Engel J.D. Papayannopoulou T. Blood. 1993; 82: 1071-1079Crossref PubMed Google Scholar) and Sertoli cells of testis (7Ito E. Toki T. Ishihara H. Ohtani H., Gu, L. Yokoyama M. Engel J.D. Yamamoto M. Nature. 1993; 362: 466-468Crossref PubMed Scopus (257) Google Scholar, 8Yomogida K. Ohtani H. Harigae H. Ito E. Nishimune Y. Engel J.D. Yamamoto M. Development. 1994; 120: 1759-1766Crossref PubMed Google Scholar). GATA-2 is preferentially expressed in the stem and progenitor cell population (5Sposi N.M. Zon L.I. Care A. Valtieri M. Testa U. Gabbianelli M. Mariani G. Bottero L. Mather C. Orkin S.H. Peschle C. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6353-6357Crossref PubMed Scopus (122) Google Scholar, 6Leonard M. Brice M. Engel J.D. Papayannopoulou T. Blood. 1993; 82: 1071-1079Crossref PubMed Google Scholar) and in mast cells and megakaryocytes (9Zon L.I. Yamaguchi Y. Yee K. Albee E.A. Kimura A. Bennett J.C. Orkin S.H. Ackerman S.J. Blood. 1993; 81: 3234-3241Crossref PubMed Google Scholar). Expression of GATA-2 is also reported in endothelial cells (10Dorfman D.M. Wilson D.B. Bruns G.A. Orkin S.H. J. Biol. Chem. 1992; 267: 1279-1285Abstract Full Text PDF PubMed Google Scholar, 11Wilson D.B. Dorfman D.M. Orkin S.H. Mol. Cell. Biol. 1990; 10: 4854-4862Crossref PubMed Scopus (131) Google Scholar) and the nervous system (2Yamamoto M., Ko, L.J. Leonard M.W. Beug H. Orkin S.H. Engel J.D. Genes Dev. 1990; 10: 1650-1662Crossref Scopus (451) Google Scholar). GATA-3 is expressed in chicken erythroid cells, T lymphocytes, and neuronal cells (2Yamamoto M., Ko, L.J. Leonard M.W. Beug H. Orkin S.H. Engel J.D. Genes Dev. 1990; 10: 1650-1662Crossref Scopus (451) Google Scholar, 12George K.M. Leonard M.W. Roth M.E. Lieuw K.H. Kioussis D. Grosveld F. Engel J.D. Development. 1994; 120: 2673-2686Crossref PubMed Google Scholar, 13Kornhauser J.M. Leonard M.W. Yamamoto M. LaVail J.H. Mayo K.E. Engel J.D. Mol. Brain Res. 1994; 23: 100-110Crossref PubMed Scopus (37) Google Scholar). The essential role of GATA factors in the development of hematopoietic lineage was first established by the fact that the functionally important GATA motifs were identified in virtually all erythroid-specific genes (14Orkin S.H. Blood. 1992; 80: 575-581Crossref PubMed Google Scholar, 15Weiss M.J. Orkin S.H. Exp. Hematol. 1995; 23: 99-107PubMed Google Scholar). Further characterization of the GATA-1 and GATA-2-null embryonic stem cells through targeted gene disruption reconfirmed their essential role in hematopoietic differentiation phenotype (16Pevny L. Simon M.C. Robertson E. Klein W.H. Tsai S.F. D'Agati V. Orkin S.H. Costantini F. Nature. 1991; 349: 257-260Crossref PubMed Scopus (1028) Google Scholar, 17Tsai F.Y. Keller G. Kuo F.C. Weiss M. Chen J. Rosenblatt M. Alt F.W. Orkin S.H. Nature. 1994; 371: 221-226Crossref PubMed Scopus (1171) Google Scholar). However, these studies also revealed that GATA-1 and GATA-2 were essential to support the viability of red cell precursors and early progenitors, respectively, by suppressing apoptosis (16Pevny L. Simon M.C. Robertson E. Klein W.H. Tsai S.F. D'Agati V. Orkin S.H. Costantini F. Nature. 1991; 349: 257-260Crossref PubMed Scopus (1028) Google Scholar, 17Tsai F.Y. Keller G. Kuo F.C. Weiss M. Chen J. Rosenblatt M. Alt F.W. Orkin S.H. Nature. 1994; 371: 221-226Crossref PubMed Scopus (1171) Google Scholar). Although several GATA-binding motif-containing genes, including erythropoietin receptor, SCL/tal-1, and the cytosolic glutathione peroxidase genes, possess anti-apoptotic activity, their expression appears to be unaffected in GATA-1-deficient proerythroblasts (18Weiss M.J. Keller G. Orkin S.H. Genes Dev. 1994; 8: 1184-1197Crossref PubMed Scopus (480) Google Scholar). Therefore, the direct anti-apoptotic target(s) of GATA-1 in proerythroblasts remains elusive. The E4BP4 (adenovirus E4 promoter-binding protein) protein was initially identified by its ability to recognize and repress the adenovirus E4 promoter (19Cowell I.G. Skinner A. Hurst H.C. Mol. Cell. Biol. 1992; 12: 3070-3077Crossref PubMed Scopus (153) Google Scholar, 20Cowell I.G. Hurst H.C. Nucleic Acids Res. 1994; 22: 59-65Crossref PubMed Scopus (76) Google Scholar) and was subsequently identified as NF-IL3A in T cells to be capable of binding and activating the human IL-31 promoter (21Zhang W. Zhang J. Kornuc M. Kwan K. Frank R. Nimer S.D. Mol. Cell. Biol. 1995; 15: 6055-6063Crossref PubMed Scopus (104) Google Scholar). E4BP4 is a protein of 462 amino acids and is a member of the basic region/leucine zipper transcription factor superfamily (19Cowell I.G. Skinner A. Hurst H.C. Mol. Cell. Biol. 1992; 12: 3070-3077Crossref PubMed Scopus (153) Google Scholar). The basic region/leucine zipper domain in E4BP4 is highly related to the basic region/leucine zipper domains of Caenorhabditis elegans cell death specification protein CES-2 (22Metzstein M.M. Hengartner M.O. Tsung N. Ellis R.E. Horvitz H.R. Nature. 1996; 382: 545-547Crossref PubMed Scopus (138) Google Scholar) and acute lymphoblastic leukemia oncoprotein E2A-HLF (23Hunger S.P. Ohyashiki K. Toyama K. Cleary M.L. Genes Dev. 1992; 6: 1608-1620Crossref PubMed Scopus (184) Google Scholar, 24Inaba T. Roberts W.M. Shapiro L.H. Jolly K.W. Raimondi S.C. Smith S.D. Look A.T. Science. 1992; 257: 531-534Crossref PubMed Scopus (245) Google Scholar). The consensus E4BP4-binding DNA sequence, i.e. (G/A)T(G/T)A(C/T)GTAA(C/T) (19Cowell I.G. Skinner A. Hurst H.C. Mol. Cell. Biol. 1992; 12: 3070-3077Crossref PubMed Scopus (153) Google Scholar), is also similar to the consensus binding sequences for CES-2 and E2A-HLF proteins (22Metzstein M.M. Hengartner M.O. Tsung N. Ellis R.E. Horvitz H.R. Nature. 1996; 382: 545-547Crossref PubMed Scopus (138) Google Scholar). Because both CES-2 and E2A-HLF were involved in cell death regulation, it was predicted that E4BP4 might be related to death control in hematopoietic cells. Further experiments indeed demonstrated that this is the case. In an IL-3-dependent pro-B cell line, Ba/F3, E4bp4 expression was activated by the IL-3 signaling pathway, and enforced expression of E4BP4 delayed apoptosis caused by IL-3 deprivation without promoting cell division (25Ikushima S. Inukai T. Inaba T. Nimer S.D. Cleveland J.L. Look A.T. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 2609-2614Crossref PubMed Scopus (117) Google Scholar). Furthermore, overexpression of the dominant-negative form of E4BP4 attenuated the survival response of IL-3 (26Kuribara R. Kinoshita T. Miyajima A. Shinjyo T. Yoshihara T. Inukai T. Ozawa K. Look A.T. Inaba T. Mol. Cell. Biol. 1999; 19: 2754-2762Crossref PubMed Scopus (73) Google Scholar). These findings clearly suggest that E4bp4 is one component of the transcription network that contributes to the survival activity of IL-3. In this study, we tried to determine the mechanism responsible for IL-3 induction of the survival gene E4bp4. We have demonstrated that IL-3 activation of the E4bp4 gene is regulated at the transcriptional level. Through promoter analysis we identified that a GATA motif downstream of a major transcription initiation site is essential for E4bp4 expression in the IL-3-dependent Ba/F3 cell line. We also demonstrated that GATA-binding factors, most likely GATA-1 and GATA-2, play an important role in the transcriptional activation of E4bp4 and are involved in the anti-apoptotic signaling of IL-3. Murine IL-3-dependent pro-B cell line Ba/F3 was maintained in medium containing murine IL-3, as described previously (27Lee S.F. Huang H.M. Chao J.R. Lin S. Yang-Yen H.F. Yen J.J. Mol. Cell. Biol. 1999; 19: 7399-7409Crossref PubMed Google Scholar). For restimulation experiments, 100 units/ml of recombinant IL-3 (R & D Systems, Minneapolis, MN) was added back to cells that had been previously cultivated in low serum medium containing no cytokine. For culturing Ba/F3 derivatives stably overexpressing GATA factors, the regular growth medium supplemented with 200 μg/ml of G418 was used. Bone marrow-derived IL-3-dependent primary cells were isolated and cultured as described previously by Rodriguez-Tarduchy et al.(28Rodriguez-Tarduchy G. Collins M. Lopez-Rivas A. EMBO J. 1990; 9: 2997-3002Crossref PubMed Scopus (206) Google Scholar). Total cellular RNA was isolated from Ba/F3 cells and bone marrow-derived IL-3-dependent primary cells with the Trizol reagent kit (Invitrogen) according to a procedure recommended by the manufacturer. Equal amounts of total RNA (30 μg) from each treatment were then subjected to the standard Northern blot analysis using probes specific to the E4bp4 or the G3pdh(glyceraldehyde-3-phosphate dehydrogenase) gene. Specific signals on the Northern blot were quantified either with Instant Imager (Packard Instrument Co., Meriden, CT) or with a densitometer (Molecular Dynamics, Sunnyvale, CA). A nuclear run-on assay to determine the rate of transcription of the E4bp4 gene was performed essentially as described (29Sambrook J. Russell D.W. Molecular Cloning: A Laboratory Manual. 3rd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY2001: 17.23-17.29Google Scholar), with nuclei prepared as described by Dignam et al. (30Dignam J.D. Lebovitz R.M. Roeder R.G. Nucleic Acids Res. 1983; 11: 1475-1489Crossref PubMed Scopus (9132) Google Scholar). The murine E4bp45′-flanking region was obtained by PCR amplification of a murine genomic library (Genome Walker kits; CLONTECH, Palo Alto, CA) using two sets of nested primers as recommended by the manufacturer. One set of the nested primers contains sequences complementary to the flanking adaptors and was provided by the vendor. The other set of the nested primers with the following sequences was designed according to the coding sequence of E4bp4: E1 primer, 5′-GCCACCTCAGCTAAGGCAGAGTTCAGC-3′, and E2 primer, 5′-AGCATCTTGTCTGAGCTGCTGGTAGGA-3′. The sequencing of this DNA fragment confirmed that it contains 1,139 bp of the 5′-flanking sequence (GenBankTM accession number AF512511) and the first 77 bp of the open reading frame of the murine E4bp4cDNA. The transcriptional initiation site of the E4bp4 gene was determined by both primer extension analysis and S1 nuclease protection as described (29Sambrook J. Russell D.W. Molecular Cloning: A Laboratory Manual. 3rd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY2001: 17.23-17.29Google Scholar). The PE1 primer (5′-AGACCGGATGGAGGAGACAAATCACTTCCCCAGTCTTC-3′; see also Fig.2A) was used both for the primer extension as well as for the sequencing reaction. For the S1 nuclease protection assay, a32P-labeled single-stranded DNA probe spanning the region between −274 and −24 (see Fig. 2A) was used to form a specific DNA-RNA hybrid prior to being digested with nuclease S1. This probe was synthesized with the 32P-labeled PE1 primer and the denatured plasmid DNA template p(−1139/+1) e4bp4-luc (see below) using a procedure essentially as described in Ref. 29Sambrook J. Russell D.W. Molecular Cloning: A Laboratory Manual. 3rd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY2001: 17.23-17.29Google Scholar. The 1140-bp DNA fragment of theE4bp4 gene spanning the region between −1139 and +1 (relative to the translational start site) was subcloned into theSmaI and HindIII sites of a promoter-less luciferase reporter vector pGL2-basic (Promega, Madison, WI) to generate the reporter construct p(−1139/+1) e4bp4-luc. The reporter constructs of p(−821/+1)e4bp4-luc, p(−451/+1)e4bp4-luc, p(−205/+1)e4bp4-luc, and p(−1139/-205)e4bp4-luc were generated by isolating corresponding inserts from p(−1139/+1)e4bp4-luc via digestion with appropriate restriction enzymes and subcloning them into the SmaI/HindIII orSmaI/NheI sites of pGL2-basic. The reporter plasmids of p(−194/+1)e4bp4-luc, p(−166/+1)e4bp4-luc, p(−135/+1)e4bp4-luc, and p(−205/-138)e4bp4-luc were generated by PCR synthesizing the corresponding fragments with an appropriate pair of primers as described below and subsequently cloning these PCR fragments into the multiple cloning sites of the pGL-2-basic vector. The 5′ primers for making these constructs are as follows: mE4P5(−194)NheI (5′-GCTAGCTGCCCAAGGGACTCACTG-3′), mE4P5(−166)NheI (5′-GCTAGCTTTATTGCAGATAACCCA-3′), mE4P5(−135)NheI (5′-GCTAGCGACAGATTTACCCTGTGC-3′), and mE4P5(−205)NheI (5′-GCTAGCACACAGCTGCCCAAGGGA-3′). The 3′ primer for the first three constructs is mE4P3(+1)HindIII (5′-AAGCTTCAGAAAGGACCTCCTCGT-3′), and the 3′ primer for the last construct is mE4P3(−138) BglII (5′-AGATCTGAGCCTTTCATGGGTTAT-3′). Plasmids p(−205/+1mE) e4bp4-luc, p(−205/+1mG)e4bp4-luc, and p(−205/+1mEmG)e4bp4-luc were derived from p(−205/+1)e4bp4-luc by PCR-assisted, site-directed mutagenesis of each individual sequence element as indicated in the schemes shown in Fig.4A. Plasmid p(−1139/+1mG)e4bp4-luc was generated by replacing the region between −205 and +1 of p(−1139/+1)e4bp4-luc with a corresponding fragment from p(−205/+1mG)e4bp4-luc. All of the constructs generated via methods involving the PCR step were confirmed by sequencing. Ba/F3 cells were transiently transfected with plasmids by electroporation using a Bio-Rad Gene Pulser II RF module system and assayed for luciferase activity as described previously (31Chen W., Yu, Y.L. Lee S.F. Chiang Y.J. Chao J.R. Huang J.H. Chiong J.H. Huang C.J. Lai M.Z. Yang-Yen H.F. Yen J.J. Mol. Cell. Biol. 2001; 21: 4636-4646Crossref PubMed Scopus (35) Google Scholar). To analyze the effect of the dominant-negative mutant of GATA-1 in the reporter gene assays, the electroporated cells were recovered in mIL-3-containing medium for 12 h and then deprived of mIL-3 for 8 h before restimulation of cells with mIL-3 was initiated. Three hours after mIL-3 restimulation, the cell lysates were prepared and assayed for luciferase activity. Nuclear extracts were prepared according to the method described by Dignamet al. (30Dignam J.D. Lebovitz R.M. Roeder R.G. Nucleic Acids Res. 1983; 11: 1475-1489Crossref PubMed Scopus (9132) Google Scholar). Binding reactions were performed as described by Wadman et al. (32Wadman I.A. Osada H. Grutz G.G. Agulnick A.D. Westphal H. Forster A. Rabbitts T.H. EMBO J. 1997; 16: 3145-3157Crossref PubMed Scopus (719) Google Scholar). An E4bp4 GATA element (GATA probe, 5′-TTTATTGCAGATAACCCATGAAAGGCTC-3′) or a 4-bp mismatched MutGATA element (MutGATA probe, 5′-TTTATTGCAagccACCCATGAAAGGCTC-3′) was used in the EMSA. The competition and supershift experiments were performed as described previously (31Chen W., Yu, Y.L. Lee S.F. Chiang Y.J. Chao J.R. Huang J.H. Chiong J.H. Huang C.J. Lai M.Z. Yang-Yen H.F. Yen J.J. Mol. Cell. Biol. 2001; 21: 4636-4646Crossref PubMed Scopus (35) Google Scholar). The GATA-1-containing complex was supershifted with an anti-GATA-1 antibody (N6; Santa Cruz Biotechnology Inc.; Santa Cruz, CA), and the GATA-2-containing complex was supershifted with a rabbit polyclonal anti-GATA-2 antibody (H-116; Santa Cruz Biotechnology Inc.). The murine GATA-1 cDNA encompassing the entire coding region (∼1.2 kb) was synthesized by PCR amplification of a murine cDNA library (CLONTECH) with the following two primers: sense, 5′-GGATCCGATTTTCCTGGTCTAGGG-3′, and antisense, 5′-CTTAAGTCAAGAACTGAGTGGGGC-3′ (33Tsai S.F. Martin D.I. Zon L.I. D'Andrea A.D. Wong G.G. Orkin S.H. Nature. 1989; 339: 446-451Crossref PubMed Scopus (663) Google Scholar). The PCR-amplified murine GATA-1 cDNA fragment was digested with restriction enzymesBamHI and EcoRI and cloned into theBamHI and EcoRI sites of the pCMVflag vector to generate the expression vector pCMVflag-mGATA-1. The expression vector pcDNA3-HA-mGATA-2 was generated by PCR amplifying the murine GATA-2 cDNA with the following two primers: sense, 5′-GAATTCGAGGTGGCTCCTGAGCAG-3′, and antisense, 5′-CTCGAGCTAGCCCATGGCAGTCAC-3′; and the resultant PCR product was subcloned into the EcoRI and XhoI sites of the pcDNA3 vector. The dominant-negative mutant of mGATA-1 (34Visvader J.E. Crossley M. Hill J. Orkin S.H. Adams J.M. Mol. Cell. Biol. 1995; 15: 634-641Crossref PubMed Google Scholar) containing only the C-terminal zinc finger (amino acids 230–336) was generated by PCR amplification of the murine GATA-1 cDNA template with the following two primers (N-terminal primer GATA230P5, 5′-GGATCCTTGTATCACAAGATGAAT-3′; C-terminal primer GATA336P3, 5′-TCACACCATGAAGCCACCTGC-3′). The resultant DNA fragment was subcloned into the pcDNA3-HA vector to yield the expression vector pcDNA3-HA-mGATA-1-dn. All constructs generated via methods involving the PCR step were confirmed by sequencing to be free of base mutations in the amplified region. Antibody specific for GATA-1 was purchased from Santa Cruz Biotechnology (N6; Santa Cruz, CA). Antibody for FLAG tag was purchased from Sigma (M2), and antibody for hemagglutinin (HA) tag was purchased from Roche Molecular Biochemicals (monoclonal antibody 12CA5; Mannheim, Germany). The cell lysates were prepared and analyzed by Western blotting as described previously (31Chen W., Yu, Y.L. Lee S.F. Chiang Y.J. Chao J.R. Huang J.H. Chiong J.H. Huang C.J. Lai M.Z. Yang-Yen H.F. Yen J.J. Mol. Cell. Biol. 2001; 21: 4636-4646Crossref PubMed Scopus (35) Google Scholar). The specific signals were visualized by an ECL Western blot system (Amersham Biosciences). The release of nucleosome into the cytosolic fraction was used to determine the degree of cells undergoing apoptosis and was performed as described previously (35Huang H.M., Li, J.C. Hsieh Y.C. Yang-Yen H.F. Yen J.J. Blood. 1999; 93: 2569-2577Crossref PubMed Google Scholar) with an enzyme-linked immunosorbent assay kit (Roche Molecular Biochemicals). The viability of cells was determined by Trypan blue exclusion staining. The ChIP assay was carried out according to a published protocol (36Saccani S. Pantano S. Natoli G. J. Exp. Med. 2001; 193: 1351-1359Crossref PubMed Scopus (335) Google Scholar). Briefly, cross-linked chromatin prepared from rapid growing cells cultured in IL-3-containing medium was sonicated to an average size of 400 bp prior to being immunoprecipitated with either anti-GATA-1 antibody or control rat IgG (anti-murine CD44) at 4 °C overnight. The immunoprecipitated chromatin, after reversal of cross-linking, was PCR-amplified with various sets of primers that either specific to the E4bp4 or a negative control gene (mouse mcl-1) (see below). The amplified DNA product was then resolved by agarose gel electrophoresis or further transferred onto membrane and subjected to Southern blot analysis using 32P-labeled E4bp4promoter-specific oligonucleotide containing the GATA motif (5′-CTACTTTATTGCAGATAACCCATGAAAGGC-3′). For PCR amplification of a specific region (F1–F5; see Fig. 6A) of theE4bp4 genomic locus, the following sets of primers were used: F1 (nucleotides −1020 to −724), mE4P5–1020 (5′-ATCTTAACTTTCAAGAGAGCTGTGTTTTAA-3′) and mE4P3–724 (5′-TGTTAACCTGCTGTCCCACCTCTGAGGGCC-3′); F2 (nucleotides −663 to −414), mE4P5–663 (5′-GGGCAGCACTCGTGTTCTGGTCACCATTGT-3′) and mE4P3–414 (5′-ACTTGGACCCATGCAATCCTTTCGTCTAGT-3′); F3 (nucleotides −360 to +1), mE4P5–360 (5′-TCTGCTGGACCACATAGTCCAAGGCAAAGA-3′) and mE4P3+1 (5′-CAGAAAGGACCTCCTCGTCCTACAGACCGG-3′); F4 (nucleotides +122 to +372), mE4P5+122 (5′-CAGGTGAAGATTTGCTCCTGAACGAAGGGA-3′) and mE4P3+372 (5′-TAATTTCAGGGAGAGCAGCTCAGCTTTTAA-3′); F5, (nucleotides +421 to +734), mE4P5+421 (5′-CTCAGTAATTCCACAGCTGTCTACTTTCAG-3′) and mE4P3+734 (5′-TGGTAGATGGAGGTGGAATACGTGCCCCGC-3′). The primers used to amplify themcl-1 gene promoter (53Chao J.R. Wang J.M. Lee S.F. Peng H.W. Lin Y.H. Chou C.H., Li, J.C. Huang H.M. Chou C.K. Kuo M.L. Yen J.J.Y. Yang-Yen H.F. Mol. Cell. Biol. 1998; 18: 4883-4898Crossref PubMed Google Scholar) are as follows: mMclP5–420 (5′-ACGAGAAAGGCTAAGGCAGGACTGC-3′) and mMclP3+20 (5′-CGCCGCAGGCTGAGGGGAAGGAGCG-3′). E4bp4 is a survival gene known to be activated by IL-3 in a murine pro-B cell line Ba/F3 (25Ikushima S. Inukai T. Inaba T. Nimer S.D. Cleveland J.L. Look A.T. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 2609-2614Crossref PubMed Scopus (117) Google Scholar). In this study, our initial goal was to understand the mechanism by which IL-3 regulates E4bp4 gene expression. We first examined whether IL-3 stimulation of the E4bp4mRNA expression could also be observed in bone marrow-derived primary IL-3-dependent cells. To address this issue, a Northern blot analysis was carried out, and the results are shown in Fig. 1A. As in the case of Ba/F3 cells (lanes 1–5), IL-3 stimulation of the primary cells resulted in an increase of the E4bp4 mRNA levels (Fig. 1A, lanes 6–8). The induction fold (∼8-fold) and kinetics (peaks at 1 h) observed with the primary cells were also very similar to those observed with the Ba/F3 cells. To determine whether IL-3 stimulation of E4bp4 mRNA expression was regulated at the transcriptional level, the nuclear run-on assay was carried out with nascent RNA probes prepared from Ba/F3 cells with or without IL-3 stimulation. As shown in Fig.1B, the transcription activity of E4bp4, like that of the positive control JunB, was markedly increased upon IL-3 stimulation, whereas transcription of either G3pdhor β-actin was unaffected by IL-3 (Fig. 1B, compare lanes 1 and 2). We next determined whether the stability of E4bp4 transcripts in response to IL-3 stimulation might also be affected. To address this issue, we measured the half-life of the E4bp4 mRNA in cells with or without IL-3 stimulation. As illustrated in Fig. 1C, the levels of E4bp4 mRNA in cells with or without IL-3 stimulation both decreased dramatically within 1 h after treatment of cells with a potent transcription inhibitor, actinomycin-D (Fig.1C). Under the same conditions no significant decline was observed with the levels of G3pdh mRNA (data not shown). The estimated half-life of E4bp4 mRNA was around 30 min regardless of the presence or absence of IL-3 (Fig. 1C). These results together with those observed with the nuclear run-on assay suggest that IL-3 stimulation of the E4bp4 mRNA expression is mainly regulated at the transcriptional level. To characterize the IL-3 signaling pathway that leads to transcriptional activation of the E4bp4 gene, the genomic DNA spanning theE4bp4 promoter region was PCR-amplified from theSspI-cut, adapter-tagged murine genomic DNA using a pair of nested primers (Fig. 2A, E1 and E2) as described under “Experimental Procedures.” To further confirm the identity of the cloned E4bp4 promoter locus, other restriction enzyme-digested adaptor-tagged murine genomic DNA was used as a template to amplify the E4bp4 promoter region using the same set of E4bp4 nested primers. All of the resulting DNA fragments were found to overlap with each other, and the sequence of the longest clone is shown in Fig. 2A. To map potential transcription start site(s) in the 5′-flanking region of E4bp4, both primer extension and S1 nuclease protection analyses were performed. In primer extension analysis (Fig.2B), multiple potential transcriptional initiation sites were reproducibly observed with RNA isolated from IL-3-stimulated cells, with one major site positioned at −167 (relative to the translation start site) and a minor one at −136 (Fig. 2B,lanes 2 and 3). The minor site at −136 was likely to be an artifact generated because of a secondary structure formed at the 5′ end of the E4bp4 transcript, because this band became more obscure when an identical experiment was repeated at 42 °C (data not shown). On the other hand, a single major transcriptional initiation site at −158 was mapped when the S1 nuclease protection assay was carried out with RNA isolated from IL-3-stimulated cells (Fig. 2C, lane 1). Under the same conditions, neither RNA from IL-3-deprived cells nor yeast control RNA gave any extended products (Fig. 2B, lanes 1 and 2) or protected bands (Fig. 2C,lanes 2 and 3). An A/T-rich sequence between −167 and −158, which tends to cause breathing of double-stranded DNA and shortening of the protected fragments in the exonuclease reaction, might have accounted for the discrepancy in mapping the transcription initiation site by these two methods. The sequence surrounding the transcription initiation site at −167 (5′-C-T-A+1-C-T-T-T-A-3′) is highly homologous to the consensus of the initiator element of eukaryotic genes (5′-Y-Y-A+1-N-T/A-Y-Y-Y-3′) (37Smale S.T. Jain A. Kaufmann J. Emami K.H., Lo, K. Garraway I.P. Cold Spring Harb. Symp. Quant. Biol. 1998; 63: 21-31Crossref PubMed Scopus (77) Google Scholar), suggesting that the nucleotide at −167 is the major transcription initiation site utilizedin vivo. To further demarcate the critical promoter sequence in the 5′-flanking region ofE4bp4, luciferase reporters driven by various regions of theE4bp4 promoter were transiently transfected into Ba/F3 cells, and their responses to IL-3 were analyzed. As shown in Fig.3, the 5′-flanking DNA fragment between −1139 to +1, when placed in front of the promoter-less pGL2-basic luciferase vector, exhibited some basal promoter activity without the stimulation of IL-3 (Fig. 3B, compare pGL2-basicand −1139/+1). Exposure of cells transfected with this reporter to IL-3 resulted in an ∼4-fold increase in luciferase activity (Fig. 3), suggesting that there is a basal promoter as well as an IL-3 response element present in this region. Reporters containing various 5′ deletions between −1139 and −205 still retained nearly full inducibility by IL-3. However, further 5′ deletions into nucleotide −194 resulted into the decrease of induction fold to ∼2.6 (Fig. 3A). Furthermore, the −166/+1 and the −205/−138 constructs that lack the potential initiator consensus A at −167 or the sequence downstream of −138, respectively, exerted a basal promoter activity very similar to that of the other constructs mentioned above (Fig. 3B). Both of the constructs still responded to IL-3 induction, although their induction responses were decreased to ∼2-fold. On the other hand, the DNA fragment spanning either between −135 and +1 or between −1139 and −205 exerted a much weaker basal promoter activity as compared with others mentioned above. Furthermore, neither of these latter two reporters responded to IL-3
Referência(s)