Regulatory Motifs for CREB-binding Protein and Nfe2l2 Transcription Factors in the Upstream Enhancer of the Mitochondrial Uncoupling Protein 1 Gene
2002; Elsevier BV; Volume: 277; Issue: 37 Linguagem: Inglês
10.1074/jbc.m108866200
ISSN1083-351X
Autores Tópico(s)Genomics, phytochemicals, and oxidative stress
ResumoThermogenesis against cold exposure in mammals occurs in brown adipose tissue (BAT) through mitochondrial uncoupling protein (UCP1). Expression of the Ucp1 gene is unique in brown adipocytes and is regulated tightly. The 5′-flanking region of the mouse Ucp1 gene contains cis-acting elements including PPRE, TRE, and four half-site cAMP-responsive elements (CRE) with BAT-specific enhancer elements. In the course of analyzing how these half-site CREs are involved in Ucp1 expression, we found that a DNA regulatory element for NF-E2 overlaps CRE2. Electrophoretic mobility shift assay and competition assays with the CRE2 element indicates that nuclear proteins from BAT, inguinal fat, and retroperitoneal fat tissue interact with the CRE2 motif (CGTCA) in a specific manner. A supershift assay using an antibody against the CRE-binding protein (CREB) shows specific affinity to the complex from CRE2 and nuclear extract of BAT. Additionally, Western blot analysis for phospho-CREB/ATF1 shows an increase in phosphorylation of CREB/ATF1 in HIB-1B cells after norepinephrine treatment. Transient transfection assay using luciferase reporter constructs also indicates that the two half-site CREs are involved in transcriptional regulation of Ucp1 in response to norepinephrine and cAMP. We also show that a second DNA regulatory element for NF-E2 is located upstream of the CRE2 region. This element, which is found in a similar location in the 5′-flanking region of the human and rodent Ucp1 genes, shows specific binding to rat and human NF-E2 by electrophoretic mobility shift assay with nuclear extracts from brown fat. Co-transfections with an Nfe2l2 expression vector and a luciferase reporter construct of the Ucp1enhancer region provide additional evidence that Nfe2l2 is involved in the regulation of Ucp1 by cAMP-mediated signaling. Thermogenesis against cold exposure in mammals occurs in brown adipose tissue (BAT) through mitochondrial uncoupling protein (UCP1). Expression of the Ucp1 gene is unique in brown adipocytes and is regulated tightly. The 5′-flanking region of the mouse Ucp1 gene contains cis-acting elements including PPRE, TRE, and four half-site cAMP-responsive elements (CRE) with BAT-specific enhancer elements. In the course of analyzing how these half-site CREs are involved in Ucp1 expression, we found that a DNA regulatory element for NF-E2 overlaps CRE2. Electrophoretic mobility shift assay and competition assays with the CRE2 element indicates that nuclear proteins from BAT, inguinal fat, and retroperitoneal fat tissue interact with the CRE2 motif (CGTCA) in a specific manner. A supershift assay using an antibody against the CRE-binding protein (CREB) shows specific affinity to the complex from CRE2 and nuclear extract of BAT. Additionally, Western blot analysis for phospho-CREB/ATF1 shows an increase in phosphorylation of CREB/ATF1 in HIB-1B cells after norepinephrine treatment. Transient transfection assay using luciferase reporter constructs also indicates that the two half-site CREs are involved in transcriptional regulation of Ucp1 in response to norepinephrine and cAMP. We also show that a second DNA regulatory element for NF-E2 is located upstream of the CRE2 region. This element, which is found in a similar location in the 5′-flanking region of the human and rodent Ucp1 genes, shows specific binding to rat and human NF-E2 by electrophoretic mobility shift assay with nuclear extracts from brown fat. Co-transfections with an Nfe2l2 expression vector and a luciferase reporter construct of the Ucp1enhancer region provide additional evidence that Nfe2l2 is involved in the regulation of Ucp1 by cAMP-mediated signaling. brown adipose tissue cAMP-responsive element CRE-binding protein peroxisomal proliferator activator receptor electrophoretic mobility shift assay Adaptive thermogenesis can be induced by cold exposure (1Rothwell N.J. Stock M.J. Nature. 1979; 281: 31-35Crossref PubMed Scopus (1205) Google Scholar, 2Smith R.E. Horwitz B.A. Physiol. Rev. 1969; 49: 330-425Crossref PubMed Scopus (525) Google Scholar) and/or a high fat diet (3Shibata H. Perusse F. Bukowiecki L.J. Can. J. Physiol. Pharmacol. 1987; 65: 152-158Crossref PubMed Scopus (25) Google Scholar, 4Levine J.A. Eberhardt N.L. Jensen M.D. Science. 1999; 283: 212-214Crossref PubMed Scopus (790) Google Scholar) in brown adipose tissue (BAT)1 through the mitochondrial uncoupling protein (UCP1). Although four homologues ofUcp1 have been identified (5Jacobsson A. Stadler U. Glotzer M.A. Kozak L.P. J. Biol. Chem. 1985; 260: 16250-16254Abstract Full Text PDF PubMed Google Scholar, 6Boss O. Samec S. Paoloni-Giacobino A. Rossier C. Dulloo A. Seydoux J. Muzzin P. Giacobino J.P. FEBS Lett. 1997; 408: 39-42Crossref PubMed Scopus (998) Google Scholar, 7Fleury C. Neverova M. Collins S. Raimbault S. Champigny O. Levi-Meyrueis C. Bouillaud F. Seldin M.F. Surwit R.S. Ricquier D. Warden C.H. Nat. Genet. 1997; 15: 269-272Crossref PubMed Scopus (1562) Google Scholar, 8Mao W., Yu, X.X. Zhong A., Li, W. Brush J. Sherwood S.W. Adams S.H. Pan G. FEBS Lett. 1999; 443: 326-330Crossref PubMed Scopus (322) Google Scholar, 9Sanchis D. Fleury C. Chomiki N. Goubern M. Huang Q. Neverova M. Gregoire F. Easlick J. Raimbault S. Levi-Meyrueis C. Miroux B. Collins S. Seldin M. Richard D. Warden C. Bouillaud F. Ricquier D. J. Biol. Chem. 1998; 273: 34611-34615Abstract Full Text Full Text PDF PubMed Scopus (271) Google Scholar), definitive proof establishing that uncoupling proteins are essential for thermogenesis has been shown only for UCP1 (10Enerback S. Jacobsson A. Simpson E.M. Guerra C. Yamashita H. Harper M.E. Kozak L.P. Nature. 1997; 387: 90-94Crossref PubMed Scopus (1098) Google Scholar). UCP1 is located in the inner membrane of mitochondria, where it reduces the mitochondrial membrane potential to generate heat instead of ATP synthesis during oxidative phosphorylation (11Nicholls D.G. Locke R.M. Physiol. Rev. 1984; 64: 1-64Crossref PubMed Scopus (1353) Google Scholar). Overexpression of Ucp1 can be achieved pharmacologically by administration of thermogenic β3 agonists (12Champigny O. Ricquier D. Blondel O. Mayers R.M. Briscoe M.G. Holloway B.R. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 10774-10777Crossref PubMed Scopus (120) Google Scholar,13Himms-Hagen J. Cui J. Danforth E., Jr. Taatjes D.J. Lang S.S. Waters B.L. Claus T.H. Am. J. Physiol. 1994; 266: R1371-R1382Crossref PubMed Google Scholar) or genetically by using tissue-specific gene promoters (14Kopecky J. Clarke G. Enerback S. Spiegelman B. Kozak L.P. J. Clin. Invest. 1995; 96: 2914-2923Crossref PubMed Scopus (489) Google Scholar, 15Li B. Nolte L.A., Ju, J.S. Han D.H. Coleman T. Holloszy J.O. Semenkovich C.F. Nat. Med. 2000; 6: 1115-1120Crossref PubMed Scopus (266) Google Scholar) to drive expression in transgenic mice or by the increase of UCP1 because of increased protein kinase A activity in protein kinase A RIIβ knockout mice (16Cummings D.E. Brandon E.P. Planas J.V. Motamed K. Idzerda R.L. McKnight G.S. Nature. 1996; 382: 622-626Crossref PubMed Scopus (367) Google Scholar). Each of these animals with increased UCP1 has increased brown fat activity, energy expenditure, and reduced adiposity. Accordingly, determining mechanisms to increase UCP1 has practical applications to the problem of obesity. There are two aspects of Ucp1 expression that require explanation, one is the molecular basis of its unique expression in BAT (5Jacobsson A. Stadler U. Glotzer M.A. Kozak L.P. J. Biol. Chem. 1985; 260: 16250-16254Abstract Full Text PDF PubMed Google Scholar) and the other is its tightly controlled regulation by the hypothalamus via the sympathetic nervous system (17Girardier L. Seydoux J. Trayhurn P. Nicholls D.G. Brown Adipose Tissue. Arnold Hodder Headline PLC, London1986: 122-151Google Scholar) in response to cold and possibly diet. A considerable body of information has accumulated showing that a 200-bp enhancer, located ∼2.5 kb upstream of the transcription start site (18Boyer B.B. Kozak L.P. Mol. Cell. Biol. 1991; 11: 4147-4156Crossref PubMed Scopus (54) Google Scholar, 19Kozak U.C. Kopecky J. Teisinger J. Enerback S. Boyer B. Kozak L.P. Mol. Cell. Biol. 1994; 14: 59-67Crossref PubMed Scopus (166) Google Scholar) that containscis-acting elements that play a critical role in the regulation of Ucp1. These elements include PPRE (20Sears I.B. MacGinnitie M.A. Kovacs L.G. Graves R.A. Mol. Cell. Biol. 1996; 16: 3410-3419Crossref PubMed Google Scholar), TRE/RARE (21Cassard-Doulcier A.M. Larose M. Matamala J.C. Champigny O. Bouillaud F. Ricquier D. J. Biol. Chem. 1994; 269: 24335-24342Abstract Full Text PDF PubMed Google Scholar, 22Rabelo R. Schifman A. Rubio A. Sheng X. Silva J.E. Endocrinology. 1995; 136: 1003-1013Crossref PubMed Scopus (91) Google Scholar), and cAMP responsive elements (CRE) (19Kozak U.C. Kopecky J. Teisinger J. Enerback S. Boyer B. Kozak L.P. Mol. Cell. Biol. 1994; 14: 59-67Crossref PubMed Scopus (166) Google Scholar). Recently, it has been shown that synergism between retinoids, isoproterenol, and thiazolidinedione regulate human Ucp1 transcription in an enhancer region located 3.5 kb upstream of the gene (23del Mar Gonzalez-Barroso M. Pecqueur C. Gelly C. Sanchis D. Alves-Guerra M.C. Bouillaud F. Ricquier D. Cassard-Doulcier A.M. J. Biol. Chem. 2000; 275: 31722-31732Abstract Full Text Full Text PDF PubMed Google Scholar). The brown adipocyte-specific expression of Ucp1 almost certainly involves the interaction of PPARγ, RXR, and PGC1 via the PPRE site. Additional regulatory elements and transcription factors are likely to be involved. The strong evidence that induction is initiated by norepinephrine action on G protein-coupled β1 and β3 adrenergic receptors (24Arch J.R.S. Ainsworth A.T. Cawthorne M.A. Piercy V. Senitt M.V. Thody V.E. Wilson C. Wilson S. Nature. 1984; 309: 163-165Crossref PubMed Scopus (651) Google Scholar, 25Bianco A.C. Sheng X. Silva J.E. J. Biol. Chem. 1988; 263: 18168-18175Abstract Full Text PDF PubMed Google Scholar, 26Kozak U.C. Held W. Kreutter D. Kozak L.P. Mol. Endocrinol. 1992; 6: 763-772PubMed Google Scholar) suggests that cyclic AMP (cAMP) directly regulates the expression of Ucp1 through the interaction of CREB with putative CREs in the 5′-flanking region of the Ucp1 gene. An alternative mechanism, suggested by Spiegelman and co-workers (27Puigserver P., Wu, Z. Park C.W. Graves R. Wright M. Spiegelman B.M. Cell. 1998; 92: 829-839Abstract Full Text Full Text PDF PubMed Scopus (3102) Google Scholar, 28Lowell B.B. Spiegelman B.M. Nature. 2000; 404: 652-660Crossref PubMed Scopus (1322) Google Scholar), postulates that the adrenergic regulation of Ucp1 does not involve CREB binding to Ucp1, rather CREB activatesPgc1 expression by the protein kinase A pathway andUcp1 is subsequently induced by the coactivation of PPARγ by PGC1. However, there is no evidence that CREB is involved in the activation of Pgc1. We only know that Pgc1mRNA levels are increased in BAT in response to cold exposure (27Puigserver P., Wu, Z. Park C.W. Graves R. Wright M. Spiegelman B.M. Cell. 1998; 92: 829-839Abstract Full Text Full Text PDF PubMed Scopus (3102) Google Scholar). It has also been reported that thyroid hormones (29Guerra C. Roncero C. Porras A. Fernandez M. Benito M. J. Biol. Chem. 1996; 271: 2076-2081Abstract Full Text Full Text PDF PubMed Scopus (69) Google Scholar), retinoids (30Alvarez R. de Andres J. Yubero P. Vinas O. Mampel T. Iglesias R. Giralt M. Villarroya F. J. Biol. Chem. 1995; 270: 5666-5673Abstract Full Text Full Text PDF PubMed Scopus (179) Google Scholar, 31Larose M. Cassard-Doulcier A.M. Fleury C. Serra F. Champigny O. Bouillaud F. Ricquier D. J. Biol. Chem. 1996; 271: 31533-31542Abstract Full Text Full Text PDF PubMed Scopus (71) Google Scholar, 32Puigserver P. Vazquez F. Bonet M.L. Pico C. Palou A. Biochem. J. 1996; 317: 827-833Crossref PubMed Scopus (119) Google Scholar), and thiazolidinediones (33Foellmi-Adams L.A. Wyse B.M. Herron D. Nedergaard J. Kletzien R.F. Biochem. Pharmacol. 1996; 52: 693-701Crossref PubMed Scopus (84) Google Scholar, 34Digby J.E. Montague C.T. Sewter C.P. Sanders L. Wilkison W.O. O'Rahilly S. Prins J.B. Diabetes. 1998; 47: 138-141Crossref PubMed Scopus (123) Google Scholar) increase transcription of the Ucp1 in rodent, in vivo and in vitro. Previous transient transfection analyses utilizing primary cell cultures from a SV40 t-antigen-induced brown adipocyte tumor showed that mutations to two of four half-site CREs in a chloramphenicol acetyltransferase-reporter construct carrying 3 kb of the 5′-flanking region almost completely abolished expression. These sites, CRE2 and CRE4, were located in the enhancer region and just 5′ of the TATA box region, respectively. Mutations to CRE1 and CRE3 showed only slight reductions in reporter activity. However, it was not established whether these essential half-site CREs bind homodimers of serine 133-phosphorylated CRE-binding protein (CREB) or whether they interacted with heterodimers formed between CREB and novel transcription factors. In this study, we have demonstrated that CRE2 in the enhancer region interacted with CREB using electrophoretic mobility shift assays (EMSA). Furthermore, transient transfection assays of luciferase reporter constructs and site-directed mutagenesis indicates that CREs are involved in transcriptional regulation of theUcp1 through interaction with phosphorylated CREB in response to cold exposure or administration of norepinephrine. We also show that two NF-E2 regulatory motifs, one of which overlaps with the CRE2 motif, bind to Nfe2l2 in a cAMP-dependent manner to control transcription of Ucp1. HIB-1B cells were maintained in Dulbecco's modified Eagle's medium (4,500 mg/literd-glucose, 584 mg/liter l-glutamine, and 15 mg/liter phenol red, Invitrogen) supplemented with 10% fetal bovine serum (Invitrogen) and 0.1 mm nonessential amino acids. Medium was changed every 2 days. Reporter constructs were transiently transfected into HIB-1B cells using LipofectAMINE PLUS reagent (Invitrogen) according to the manufacturer's protocol. The day before transfection, 2 × 105 cells were seeded into a 24-well cluster dish (Corning). Briefly, 0.5 μg of reporter construct was transfected with 50 ng of pRL/SV40 (Promega), a plasmid containingRenilla luciferase gene under control of SV40 promoter, in a mixture of PLUS and LipofectAMINE reagent. For the co-expression experiment, each 0.3 μg of reporter construct and expression vector were transfected with 50 ng of pRL/SV40. Transfected cells were cultured in the medium in the presence or absence 1 μmnorepinephrine (Sigma) or 0.5 mm 8-bromo-cAMP (Calbiochem) for 16 h. Cell extracts were prepared, and the activity ofPhotinus and Renilla luciferase were determined using the dual-luciferase reporter assay system (Promega). For each construct, the activity of the Photinus luciferase was divided by the activity of the Renilla luciferase to correct for transfection efficiency. Under each treatment, the corrected activity was again divided by activity from pGL3/basic (Promega), the empty vector, to estimate the -fold increase for each construct. The -fold increase for the overexpression experiment was obtained by dividing the corrected activity by the empty vector (pCMV/tag). Each experiment was performed in duplicate dishes. The 3.1 kb of the 5′-flanking region containing the four CRE and the 220-bp BAT-specific enhancer of the mouse Ucp1 gene were obtained by PCR amplification. The 3.85-kb BglI fragment in pGEM, which was previously used in our characterization of Ucp1 (19Kozak U.C. Kopecky J. Teisinger J. Enerback S. Boyer B. Kozak L.P. Mol. Cell. Biol. 1994; 14: 59-67Crossref PubMed Scopus (166) Google Scholar) (note that nucleotide positions correspond to those in the Ucp1 gene as described in GenBankTMU63418), was used as a template with forward and reverse primers, 5′-ggggtaCCGTGCACACTGCCAAATCATCTC (4379/4355, a new KpnI site is underlined) and 5′-gggagCTCCTGCAGAGCCACCTGGGCTAGG (7514/7538, a newSacI site is underlined), respectively, and subcloned into pGL3/basic using the KpnI and SacI restriction enzyme sites. To obtain the Ucp1 promoter with or without CRE4, forward primers 5′-ggggatccGAGTGACGCGCGGCTGGG (nucleotide sequences for CRE4 are shown as bold and a new BamHI site is underlined, 7261/7278) or 5′-ggggatcCGGCTGGGAGGCTTGCGCA (a new BamHI site is underlined, 7271/7289) and reverse primer 5′-gggaagcttGGGCTAGGTAGTGCCAG (a new HindIII site is underlined, 7504/7520) were used for PCR amplification and subcloned into pGL3/basic using BglII and HindIII restriction enzyme sites. For the 220 bp of BAT-specific enhancer region, the 3.85-kb BglI fragment was PCR amplified using primers 5′-ggggagCTCCTCTACAGCGTCACAGAGG (SacI site is underlined, 4841/4862) and 5′-gggctcgagAGTCTGAGGAAAGGGTTGA (a new XhoI site is underlined, 5025/5045) and subcloned into the luciferase reporter construct containing the Ucp1 promoter (give nucleotide sequences). For the rat Ucp1enhancer region, genomic DNA from rat liver was amplified by PCR using primers 5′-gtgaaccttgctgccgctcctttgc (forward primer, the putative NF-E2 site is underlined, −2519/−2494) and 5′-tgtgatgtcagctcaagacagggag (reverse primer, −2283/−2308) and subcloned into the luciferase reporter construct containing theUcp1 promoter. To generate the mutations in NF-E2 site, primer 5′-gtgaacctgtaggccgctcctttgc (forward primer, the putative NF-E2 site is underlined with mutations shown italic, −2519/−2494) and 5′-tgtgatgtcagctcaagacagggag (reverse primer, −2283/−2308) were used for PCR amplification. The structure of each fragment was verified by DNA sequencing. Nfe2l2 cDNA was kindly provided by Dr. Paul Ney (St. Jude Children's Research Hospital). A Nfe2l2 expression vector was made by cloning aNotI fragment into the pCMV/tag1 (Stratagene). CRE2 and CRE3 sequences in the 220 bp of the BAT-specific enhancer region were mutated using PCR and subcloned into the luciferase reporter plasmid, pGL3/basic. For CRE3 the forward primer was 5′-ggggagCTCCTCTACAGCtgaACAGAGG (CRE3 shown in bold with lowercase italic letters that represent mutations; a new SacI site is underlined, 4841/4862) and the reverse primer was 5′-gggctcgagAGTCTGAGGAAAGGGTTGA (a newXhoI site is underlined, 5025/5045). To mutate CRE2, two pairs of primers were required in separate amplifications. The first pair was 5′-ggggagCTCCTCTACAGCGTCACAGAGG (forward primer, a new SacI site is underlined, 4841/4862) and the 5′-AGTGGAAAGGTtcaGACTAGTTCAG (reverse primer, CRE2 is shown in bold with lowercase italic letters representing mutations, 4883/4907). The second pair was 5′-CTGAACTAGTCtgaACCTTTCCACT (forward primer, CRE2 is shown in bold with lowercase italic letters representing mutations, 4883/4907) and 5′-gggctcgagAGTCTGAGGAAAGGGTTGA (reverse primer, a newXhoI site is underlined, 5025/5045). To generate the 220-bp enhancer region with mutations in CRE2, aliquots (1 μl of each 50 μl PCR reactions) of the two PCR products were mixed and subjected to PCR amplification using primer pairs for intact the 220-bp BAT-specific enhancer region. The resulting mutations were confirmed by sequencing. To mutate both CRE2 and CRE3, the 220-bp fragment, which contains the mutation in CRE2, was subjected to PCR amplification using primer pairs 5′-ggggagCTCCTCTACAGCtgaACAGAGG (CRE3 shown in bold with lowercase italic letters that represent mutated sites; a new SacI site is underlined, 4841/4862) and 5′-gggctcgagAGTCTGAGGAAAGGGTTGA (a new XhoI site is underlined, 5025/5045). After the mutations were verified by sequencing, the DNA fragments containing the mutated sites in CRE2 and/or CRE3 were subcloned into luciferase reporter plasmid containingUcp1 promoter with or without CRE4. Nuclear extracts from various tissues of A/J or C57BL/6J (B6) mice, and HIB-1B cells were prepared as described (35Dignam J.D. Lebowitz R.M. Roeder R.G. Nucleic Acids Res. 1983; 11: 1475-1489Crossref PubMed Scopus (9164) Google Scholar), except that phosphatase inhibitor mixtures 1 and 2 (Sigma) were added. The protein concentration was determined by the Lowry method (36Lowry O.H. Passonneau J.V. Hasselberger F.X. Schulz D.W. J. Biol. Chem. 1964; 239: 18-30Abstract Full Text PDF PubMed Google Scholar) using bovine serum albumin as a standard. To prepare probes for EMSA, single-stranded oligonucleotides were synthesized and purified (Operon). 200 pmol each of the complementary oligonucleotides were annealed in 100 μl containing 100 mm NaCl to obtain a double-stranded probe. Five μg of nuclear extract (or in vitro translated CREB, Nfe2l2, and p18) were incubated initially for 10 min at room temperature in 29 μl containing 20 mm HEPES (pH 7.9), 100 mm KCl, 0.1 mm EDTA, 10% glycerol, 1 mmdithiothreitol, 1.5 μg of poly(dA-dT), and 5 mmMgCl2. The mixture was then incubated for an additional 20 min after adding 32P-labeled probe (4 × 105 cpm/μl) with or without an unlabeled competitor or antibody for supershift. The antibodies were purchased from Santa Cruz Biotechnology. The reaction was electrophoresed on a 6% polyacrylamide gel (Bio-Rad) in 0.5× TBE buffer. The gel was then dried and exposed to a phosphorimage screen. The radioactivity was visualized and quantified using PhosphorImager and ImageQuant software (AmershamBiosciences). Western blot analyses were performed as described by Laemmli (37Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207537) Google Scholar) and Towbin et al. (38Towbin H. Stachelin T. Gordon J. Proc. Natl. Acad. Sci. U. S. A. 1979; 76: 4350-4354Crossref PubMed Scopus (44939) Google Scholar) with little modification. Cell lysates from HIB-1B cells were prepared by adding SDS sample buffer containing 62.5 mmTris-Cl (pH 6.8), 2% (w/v) SDS, 10% glycerol, 50 mmdithiothreitol, 0.1% (w/v) bromphenol blue with 1% (v/v) phosphatase inhibitor mixtures 1 and 2 (Sigma). After cell lysates were separated on 8% SDS-polyacrylamide gels, protein was transferred onto nitrocellulose membrane (Millipore). The blots were then incubated with antibody against CREB (1:1,000 dilution, Santa Cruz), or phospho-CREB (Ser133, 1:1,000 dilution, New England Biolabs), overnight at 4 °C with gentle agitation, followed by incubation with anti-rabbit IgG as a secondary antibody (horseradish peroxide conjugated, Amersham Biosciences). Bands were visualized by using the enhanced chemiluminescence reagent (Amersham Biosciences) and exposed to X-Omat film (Kodak). In vitro protein translation was performed using the TnT coupled reticulocyte lysate system (Promega) according to the manufacturers protocol. The cDNA for CREB and Nfe2l2 was subcloned into pBluescript II (Stratagene) and pCITE4a (Novagen) vectors, respectively, then used for template. The cDNA for p18 was provided by Dr. Paul Ney (St. Jude Children's Research Hospital). To generate 32P-end-labeled probe for the DNase I footprinting assay, pGL3/3.1 kb (20 μg) was digested with theBstEII restriction enzyme. After cleaning up with phenol-chloroform extraction, the 5′ overhang was filled-in with [32P]dCTP (Amersham Biosciences) and Klenow fragment polymerase (New England Biolabs), and residual nucleotides were removed using the Qiagen nucleotide removal kit. A 286-bp of32P-end-labeled probe containing NF-E2 as well as CRE2 and CRE3 was isolated from DNA digested with XbaI and separated on the 0.8% agarose gel. Eluted DNA was subjected to further purification using Elutip-d (Schleicher & Schu¨ll). Nuclear extracts or in vitro translated CREB were incubated at room temperature for 1 h with a probe (15,000 cpm per reaction) in 180 μl of assay buffer containing 10 mm Tris-Cl (pH 8.0), 5 mm MgCl2, 1 mm CaCl2, 2 mm dithiothreitol, 50 μg/ml bovine serum albumin, 2 ng/ml calf thymus DNA, and 100 mm KCl. 0.05 unit of DNase I (Roche Molecular Biochemicals) was added, then incubated another 2 min at 37 °C for DNase digestion. DNA was precipitated by adding 700 μl of DNase I stop solution and separated on sequencing gel with the32P-end-labeled size marker. Four potential CRE sites were located in the 5′-flanking region ofUcp1 (Table I). All four CREs have half-site consensus sequences (CGTCA) and evidence that these sites are involved in the regulation of Ucp1 is limited to loss of chloramphenicol acetyltransferase reporter activity in transient expression assays in a BAT cell line (19Kozak U.C. Kopecky J. Teisinger J. Enerback S. Boyer B. Kozak L.P. Mol. Cell. Biol. 1994; 14: 59-67Crossref PubMed Scopus (166) Google Scholar). From this analysis CRE2 and CRE4 appeared to be essential; mutations to CRE1 showed no loss of expression and mutations to CRE3 only slightly reduced expression. This study will largely focus on evaluating the function of CRE2 located in the upstream enhancer (Fig.1). A CRE2 probe for EMSA was made with 5 bp of half-site CRE2 (CGTCA) flanked by 14 bp of 5′- and 3′-flanking sequences as shown in Table I. Nuclear extracts, prepared from BAT of newborn mice maintained at room temperature, and BAT, retroperitoneal fat tissue, inguinal fat tissue, and liver of mice kept in the cold (4 °C) overnight, showed a major retarded band that was eliminated by competition with a 20-fold excess of cold CRE2 (specific shifted bands are shown with the dark arrow in Fig.2A). However, probes prepared from the region just downstream of the CRE2 motif failed to form a similar retarded band (data not shown). The complex from liver was ∼10 times stronger than that of other fat tissues (loading for liver was 1/10th of the reaction as indicated in the legend; the second retarded band in liver is nonspecific and can be seen with other probes, data not shown). It is of great interest that the binding activity of nuclear extracts of newborn mice is much greater than that of adult mice (Fig. 2C).Table ISynthetic double-stranded CRE sequences used for electomobility shift assayNameLocationSequencesCRE14419/4437TTATAGTGCCGTCACTAACAATATCACGGCAGTGATTGCRE24884/4902TGAACTAGTCGTCACCTTTACTTGATCAGCAGTGGAAACRE34843/4861CCTCTACAGCGTCACAGAGGGAGATGTCGCAGTGTCTCCRE47258/7276TGGGAGTGACGCGCGGCTGACCCTCACTGCGCGCCGACCRE (Somatostatin)TTGGCTGACGTCAGAGAGAAACCGACTGCAGTCTCTCTmlCRE2AACTAGTCtgaACCTTTTTGATCAGactTGGAAAm2CRE2AACTAtgaGTCACCTTTTTGATactCAGTGGAAAEach DNA containing the half-site CRE motif (CGTCA) from mouseUcp1 or palindromic sequences from the somatostatin gene was annealed as described under "Experimental Procedures." Consensus sequences for CRE are underlined (lowercase letters represent mutations). Open table in a new tab Figure 2Binding of half-site CRE sequences to nuclear extracts from tissues of mice.A, autoradiogram of an EMSA using 32P-end-labeled CRE2 (0.1 pmol) from the mouseUcp1 gene. Each lane (except for liver which was 1/10th of the reaction) was loaded with a binding reaction containing 5 μg of nuclear extracts on 6% nondenaturing acrylamide gel. To verify specificity of binding, competitors included 2 pmol of cold probe or various antibodies (1 μl) as indicated. Slowly migrating bands representing CREB binding or free probes at the gel front are indicated by arrowheads. B, competitive binding activity of CRE2 with half-site CREs from the mouse Ucp1gene and a palindromic CRE from the somatostatin gene. Autoradiogram of an EMSA showing only the CREB bands. Each lane was loaded with a binding reaction containing 0.1 pmol of 32P-end-labeled CRE2, 5 μg of nuclear extracts from the BAT of A/J mice that were exposed to cold (4 °C, overnight), and 0.4 pmol of cold competitors as indicated on the top. Percent competition of32P-end-labeled CRE2 to CREB by CRE sequences from the mouse Ucp1 and somatostatin (named CRE) genes was calculated from the radioactivity of the slow migrating bands in the lane without (first lane) and the lane with the individual competitors as shown at the bottom. C, autoradiogram of an EMSA using 32P-end-labeled CRE2 from the mouse Ucp1gene. Each lane was loaded with a binding reaction containing 5 μg of nuclear extracts from adult (4-week-old) or 0-day-old B6 mice. Only the CREB bands are shown.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Each DNA containing the half-site CRE motif (CGTCA) from mouseUcp1 or palindromic sequences from the somatostatin gene was annealed as described under "Experimental Procedures." Consensus sequences for CRE are underlined (lowercase letters represent mutations). To further characterize the binding sites of CRE2, we have performed competitive binding assays with the same mutation, GTC to TGA, in two contiguous locations in the sequence. For the m1CRE2 probe, the mutation occurs in the middle of the half-site CRE motif, whereas the m2CRE2 mutation only overlaps the first C in the CRE motif (Table I). In competitive EMSA, the m1CRE2, but not the m2CRE2 mutant oligonucleotide, has lost the ability to compete with the labeled CRE2 probe (Fig. 2A). This suggests that the half-site CRE motif, but not the flanking 5′-region, is active in binding the specific factor(s). To identify the nuclear factor(s) that bind to CRE2, we applied specific antibodies against Fos, Jun, CBP, or CREB/ATF1 in an EMSA reaction. Because of the sequence similarity of CRE and AP-1-binding sites for the Jun/Fos heterodimer (palindromic CRE, TGACGTCA; palindromic AP-1, TGA(C/G)TCA; half-site CRE sequences are shown underlined) and the known interaction between CREB and CBP, we have tested their antibodies in the supershift assay. The data in Fig. 2A demonstrates that the factors that bind to CRE2 are part of the CREB/ATF1 family. It suggests that CREB/ATF1 does not interact with either jun andfos or CBP. To quantify binding of the four half-site CREs to CREB/ATF1, we measured the ability of each CRE to compete with the C
Referência(s)