Glucocorticoids Repress Transcription of Phosphoenolpyruvate Carboxykinase (GTP) Gene in Adipocytes by Inhibiting Its C/EBP-mediated Activation
2003; Elsevier BV; Volume: 278; Issue: 15 Linguagem: Inglês
10.1074/jbc.m300263200
ISSN1083-351X
AutoresYael Olswang, Barak Blum, Hanoch Cassuto, Hannah Cohen, Yael Biberman, Richard W. Hanson, Lea Reshef,
Tópico(s)Hormonal Regulation and Hypertension
ResumoThe cytosolic form of the phosphoenolpyruvate carboxykinase (PEPCK-C) gene is selectively expressed in several tissues, primarily in the liver, kidney, and adipose tissue. The transcription of the gene is reciprocally regulated by glucocorticoids in these tissues. It is induced in the liver and kidney but repressed in the white adipose tissue. To elucidate which adipocyte-specific transcription factors participate in the repression of the gene, DNase I footprinting analyses of nuclear proteins from 3T3-F442A adipocytes and transient transfection experiments in NIH3T3 cells were utilized. Glucocorticoid treatment slightly reduced the nuclear C/EBPα concentration but prominently diminished the binding of adipocyte-derived nuclear proteins to CCAAT/enhancer-binding protein (C/EBP) recognition sites, without affecting the binding to nuclear receptor sites in the PEPCK-C gene promoter. Of members of the C/EBP family of transcription factors, C/EBPα was the strongest trans-activator of the PEPCK-C gene promoter in the NIH3T3 cell line. The glucocorticoid receptor (GR), in the presence of its hormone ligand, inhibited the activation of the PEPCK-C gene promoter by C/EBPα or C/EBPβ but not by the adipocyte-specific peroxisome proliferator-activated receptor γ2. This inhibition effect was similar using the wild type or mutant GR and did not depend on GR binding to the DNA. The glucocorticoid response unit (GRU) in the PEPCK-C gene promoter (−2000 to +73) restrained C/EBPα-mediated trans-activation, because mutation of each single GRU element increased this activation by 3–4-fold. This series of GRU mutations were repressed by wild type GR to the same percent as was the nonmutated PEPCK-C gene promoter. In contrast, the repression by mutant GR depended on the intact AF1 site in the gene promoter, whereby mutation of the AF1 element abolished the repression. The cytosolic form of the phosphoenolpyruvate carboxykinase (PEPCK-C) gene is selectively expressed in several tissues, primarily in the liver, kidney, and adipose tissue. The transcription of the gene is reciprocally regulated by glucocorticoids in these tissues. It is induced in the liver and kidney but repressed in the white adipose tissue. To elucidate which adipocyte-specific transcription factors participate in the repression of the gene, DNase I footprinting analyses of nuclear proteins from 3T3-F442A adipocytes and transient transfection experiments in NIH3T3 cells were utilized. Glucocorticoid treatment slightly reduced the nuclear C/EBPα concentration but prominently diminished the binding of adipocyte-derived nuclear proteins to CCAAT/enhancer-binding protein (C/EBP) recognition sites, without affecting the binding to nuclear receptor sites in the PEPCK-C gene promoter. Of members of the C/EBP family of transcription factors, C/EBPα was the strongest trans-activator of the PEPCK-C gene promoter in the NIH3T3 cell line. The glucocorticoid receptor (GR), in the presence of its hormone ligand, inhibited the activation of the PEPCK-C gene promoter by C/EBPα or C/EBPβ but not by the adipocyte-specific peroxisome proliferator-activated receptor γ2. This inhibition effect was similar using the wild type or mutant GR and did not depend on GR binding to the DNA. The glucocorticoid response unit (GRU) in the PEPCK-C gene promoter (−2000 to +73) restrained C/EBPα-mediated trans-activation, because mutation of each single GRU element increased this activation by 3–4-fold. This series of GRU mutations were repressed by wild type GR to the same percent as was the nonmutated PEPCK-C gene promoter. In contrast, the repression by mutant GR depended on the intact AF1 site in the gene promoter, whereby mutation of the AF1 element abolished the repression. glucocorticoid receptor phosphoenolpyruvate carboxykinase-C CCAAT/enhancer-binding protein glucocorticoid-response element retinoid X receptor peroxisome proliferator-activated receptor cyclic AMP-response element PPAR-response element reverse transcriptase chloramphenicol acetyltransferase glucocorticoid response unit Dulbecco's modified Eagle's medium hepatocyte nuclear factor chicken ovalbumin upstream transcription factor Glucocorticoids play a fundamental role in the maintenance of homeostasis in mammals. Removal of the adrenals severely compromises the ability of animals to withstand fasting (for reviews see Refs. 1Baxter J.D. Rousseau G.G. Glucocorticoid Hormone Action: An Overview. Springer-Verlag Inc., New York1979: 10-24Google Scholar and 2Berne R.M. Berne R.M. Levy M.N. 4th Ed. Physiology. Mosby, Inc., St. Louis, MO1998: 942-944Google Scholar). Glucocorticoids exert their effects via the glucocorticoid receptor (GR),1 predominantly by modulating gene transcription (3Gustafsson J.A. Carlstedt-Duke J. Poellinger L. Okret S. Wikstrom A.C. Bronnegard M. Gillner M. Dong Y. Fuxe K. Cintra A. Endocr. Rev. 1987; 8: 185-234Crossref PubMed Scopus (350) Google Scholar, 4Tronche F. Kellendonk C. Reichardt H.M. Schutz G. Curr. Opin. Genet. & Dev. 1998; 8: 532-538Crossref PubMed Scopus (158) Google Scholar, 5Hu X. Lazar M.A. Trends Endocrinol. Metab. 2000; 11: 6-10Abstract Full Text Full Text PDF PubMed Scopus (238) Google Scholar). An attractive mode of regulation, especially in light of the coordinated effects of glucocorticoids in maintaining homeostasis, is the opposing control of the samegene in different tissues by GR. PEPCK-C gene provides an optimal model for studying this mode of regulation. The transcription of this gene is stimulated by glucocorticoids in the liver and kidney (6Meisner H. Loose D.S. Hanson R.W. Biochemistry. 1985; 24: 421-425Crossref PubMed Scopus (62) Google Scholar, 7Sasaki K. Crip T.P. Koch S.R. Andreone T.L. Petersen D.D. Beale E.G. Granner D.K. J. Biol. Chem. 1984; 259: 15242-15251Abstract Full Text PDF PubMed Google Scholar) but is repressed in the adipose tissue (8Nechushtan H. Benvenisty N. Brandeis R. Reshef L. Nucleic Acids Res. 1987; 15: 6405-6417Crossref PubMed Scopus (54) Google Scholar). PEPCK-C catalyzes a key reaction that determines the rates of gluconeogenesis in the liver and kidney and glyceroneogenesis in the adipose tissue and liver (9Hanson R.W. Reshef L. Annu. Rev. Biochem. 1997; 66: 581-611Crossref PubMed Scopus (634) Google Scholar). Glyceroneogenesis, the de novosynthesis of 3-glycerophosphate from pyruvate and amino acids (via an abbreviated version of gluconeogenesis), provides this precursor for the synthesis of triglycerides (10Ballard F.J. Hanson R.W. Leveille G.A. J. Biol. Chem. 1967; 242: 2746-2750Abstract Full Text PDF PubMed Google Scholar, 11Reshef L. Niv J. Shapiro B. J. Lipid Res. 1967; 8: 688-691Abstract Full Text PDF PubMed Google Scholar). Recently, we have performed a targeted mutation in the adipose tissue-specific enhancer of the PEPCK-C gene in embryonic stem cells. The mutation ablated PEPCK-C gene expression in white adipose tissue of mice homozygous for this mutation and caused a decrease in the storage of triglycerides, which in some mice developed into lipodystrophy (12Olswang Y. Cohen H. Papo O. Cassuto H. Croniger C.M. Hakimi P. Tilghman S.M. Hanson R. Reshef L. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 625-630Crossref PubMed Scopus (119) Google Scholar). This mutation therefore established the importance of PEPCK-C and glyceroneogenesis in the homeostasis of triglycerides in the adipose tissue. Because PEPCK-C is encoded by a unique copy gene, and is transcribed from a single promoter, it is likely that tissue-specific factors are involved in the reciprocal regulation that leads to stimulation (liver and kidney) or repression (adipose tissue) of the gene transcription in the presence of glucocorticoids. Yamamoto and colleagues (13Diamond M.L. Miner J.N. Yoshinaga S.K. Yamamoto K.R. Science. 1990; 249: 1266-1272Crossref PubMed Scopus (1070) Google Scholar) proposed the term composite GRE to describe a nonconsensus sequence that binds the GR with low affinity and, in turn, is capable of mediating either repression or activation of genes. The GRE identified in the PEPCK-C gene (14Imai E. Stromstedt P.E. Quinn P.G. Carlstedt-Duke J. Gustafsson J.A. Granner D.K. Mol. Cell. Biol. 1990; 10: 4712-4719Crossref PubMed Scopus (244) Google Scholar) is a nonconsensus sequence that binds GR at a very low affinity and is not able by itself to transmit a transcriptional response to glucocorticoids. In fact, PEPCK-C gene promoter harbors a GR unit (GRU) containing two low affinity, nonconsensus GR-binding sites and two accessory elements, AF1 and AF2. These elements do not bind steroid receptors, but their occupancy is required for the response of the PEPCK-C gene to glucocorticoids (see scheme in Fig. 2). The factors binding to the AF1 element are all nonsteroid nuclear receptors; these include hepatocyte nuclear factor (HNF) 4, chicken ovalbumin upstream transcription factor (COUPTF), retinoic acid receptor, retinoid X receptor (RXR), and members of the peroxisome proliferator-activated receptor (PPAR) family. The AF2 site (15Faber S. O'Brien R.M. Imai E. Granner D.K. Chalkley R. J. Biol. Chem. 1993; 268: 24976-24985Abstract Full Text PDF PubMed Google Scholar) binds HNF3β (16Wang J.C. Stromstedt P.E. Sugiyama T. Granner D.K. Mol. Endocrinol. 1999; 13: 604-618PubMed Google Scholar, 17Wang J.C. Stafford J.M. Scott D.K. Sutherland C. Granner D.K. J. Biol. Chem. 2000; 275: 14717-14721Abstract Full Text Full Text PDF PubMed Scopus (56) Google Scholar) and has been proposed to comprise an insulin-response element as well (18O'Brien R.M. Lucas P.C. Forest C.D. Magnuson M.A. Granner D.K. Science. 1990; 249: 533-537Crossref PubMed Scopus (289) Google Scholar). To identify tissue-specific factors that are involved in the glucocorticoid-mediated repression of PEPCK-C gene transcription in adipocytes, we have employed a systematic DNase I footprinting analysis of the PEPCK-C gene promoter, using adipocyte nuclear proteins. Their functional participation in the repression has been assessed using transient transfection experiments in PEPCK-nonexpressing NIH3T3 cells. The results from these two independent experimental systems consistently identified the involvement of members of the C/EBP family, but not those of PPAR, in the GR repression of the PEPCK-C gene promoter activity. Furthermore, experiments in NIH3T3 cells revealed a hierarchical constraint of PEPCK-C gene promoter trans-activation by the separate GRU elements. Both wild type and mutant GR (incapable of binding the DNA) repress the C/EBP-mediated trans-activation by 50–60%, regardless of whether the trans-activation is low or high. However, the repression by mutant GR critically requires an intact AF1 site in the PEPCK-C gene. Dulbecco's modified Eagle's medium (DMEM), F-12, and fetal and newborn calf serum were purchased from Biological Industries, Kibutz Beit Haemek, Israel. Biosynthetic human insulin was obtained from Novo Nordisk (Denmark). Dexamethasone, the synthetic glucocorticoid hormone, was purchased from Teva, Israel Pharmaceutical Industry. Ultraspec, the commercial reagent for the preparation of tissue RNA, was purchased from Biotecx Laboratories, Inc. (Austin, TX). Radioactive signals were quantified using a PhosphorImager (Fujix BAS 1000, Fuji, Japan). Reverse transcriptase was obtained from Invitrogen. Random hexanucleotide pd(N)6 was purchased from Amersham Biosciences, and the ribonuclease inhibitor, RNasin, was purchased from Promega (Madison, WI). The enzyme-linked immunosorbent assay kit for the determination of human somatotropin was purchased from Roche Molecular Biochemicals. 3T3-F442A cells, obtained from Dr. Howard Green (19Green H. Kehinde O. Cell. 1976; 7: 105-113Abstract Full Text PDF PubMed Scopus (616) Google Scholar), were grown to confluence in DMEM and supplemented with 10% newborn calf serum. For differentiation of the cells to adipocytes, newborn calf serum was replaced by 10% fetal calf serum; isobutylmethylxanthine was added at a final concentration of 0.2 mm, and the cells were incubated for 3 days. The cells were further incubated for at least 8 days in a medium containing 4 milliunits/ml insulin, until ∼80–90% of cells contained fat droplets. Dexamethasone (10−7m) was added to NIH3T3 cells no later than 20 h after transfection, when expression of the reporter gene was barely detected. Cells were harvested no later than 24 h after addition of dexamethasone to prevent cell lysis. 3T3-F442A adipocytes were similarly treated with 10−7m dexamethasone for 16–18 h. Total RNA was isolated from a single 100-mm cultured plate of 3T3-F442A adipocytes or from 30 mg of mouse liver using the commercial reagent Ultraspec according to the manufacturer's instructions. One μg of total RNA was reverse-transcribed according to manufacturer's instructions, using the Invitrogen reverse transcriptase kit in the presence of 7.5 units/ml random hexanucleotide pd(N)6 as primer and 1 unit/μl RNasin ribonuclease inhibitor, except that the incubation was for 30 min at 42 °C. PCR was performed using the PEPCK-C primers 5′-CTTGTCTACGAAGCTCTCAG from exon 9 and 3′-CGTCCGAACATCCACTC from exon 10. Primers for theaP2 gene were 5′-CCTGGAAGCTTGTCTCCAG from exon 1 and 3′-CTCTTGTGGAAGTCACGCC from exon 4. Primers for β-actin were the same as published previously (20Eden S. Hashimshony T. Keshet I. Cedar H. Thorne A.W. Nature. 1998; 394: 842Crossref PubMed Scopus (240) Google Scholar). PCR was performed in the presence of a trace of [32P]dCTP (0.5 × 106 dpm) to allow semi-quantification (20Eden S. Hashimshony T. Keshet I. Cedar H. Thorne A.W. Nature. 1998; 394: 842Crossref PubMed Scopus (240) Google Scholar). The PCR program included denaturation at 94 °C for 1 min, annealing at 55 °C for 1 min, and elongation at 72 °C for 2 min, 20 cycles each consisting of denaturation for 30 s, annealing for 1 min, and elongation for 1 min. The PCR product was separated by electrophoresis on 8% polyacrylamide gel and quantified using a PhosphorImager apparatus and visualized by its exposure to autoradiographic film. Nuclear proteins were extracted from rat liver according to Gorski et al. (21Gorski K. Carneiro M. Schibler U. Cell. 1986; 47: 767-776Abstract Full Text PDF PubMed Scopus (973) Google Scholar) as modified (22Trus M. Benvenisty N. Cohen H. Reshef L. Mol. Cell. Biol. 1990; 10: 2418-2422Crossref PubMed Scopus (32) Google Scholar). Nuclear protein extracts from adipocytes were prepared essentially as described previously (22Trus M. Benvenisty N. Cohen H. Reshef L. Mol. Cell. Biol. 1990; 10: 2418-2422Crossref PubMed Scopus (32) Google Scholar), except that the sucrose gradient step to further purify the nuclei was omitted. DNase I footprinting assays were performed as described previously (22Trus M. Benvenisty N. Cohen H. Reshef L. Mol. Cell. Biol. 1990; 10: 2418-2422Crossref PubMed Scopus (32) Google Scholar). The autoradiographic density signals of specific bands in the exposed film (see Figs. Figure 2, Figure 3, Figure 4) were quantified using Fluor-STMMultiImager with Multianalyst version 1.1 (Bio-Rad). 5 μg of nuclear proteins from adipocytes treated or untreated with dexamethasone (a synthetic glucocorticoid) were separated on 15% SDS-PAGE and transferred to nitrocellulose membrane (Protran BA 85, Schleicher & Schuell). C/EBPα was probed with rabbit polyclonal anti-C/EBPα antibody c100 diluted 1:1000 (a gift from Dr. Steven McKnight) and was detected using horseradish peroxidase-conjugated goat anti-rabbit antibody diluted 1:4000. Nuclear Y12 protein was probed with mouse anti-sm monoclonal antibody Y12 (23Raitskin O. Cho D.S. Sperling J. Nishikura K. Sperling R. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 6571-6576Crossref PubMed Scopus (80) Google Scholar) (a gift from Dr. Ruth Sperling), diluted 1:10, and detected using horseradish peroxidase conjugated to affinity-purified goat anti-mouse IgG F(ab)′ fragment diluted 1:3000 (23Raitskin O. Cho D.S. Sperling J. Nishikura K. Sperling R. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 6571-6576Crossref PubMed Scopus (80) Google Scholar). Secondary antibodies were visualized with SuperSignal West Pico chemiluminescent substrate (Pierce). NIH3T3 cells were grown on 100-mm plates in DMEM containing 10% fetal calf serum. For transfection, cells were transferred to DMEM containing 10% newborn calf serum. Transfection was performed essentially according to Chen and Okayama (24Chen C. Okayama H. Mol. Cell. Biol. 1987; 7: 2745-2752Crossref PubMed Scopus (4824) Google Scholar), as described previously (25Yanuka-Kashles O. Cohen H. Trus M. Aran A. Benvenisty N. Reshef L. Mol. Cell. Biol. 1994; 14: 7124-7133Crossref PubMed Scopus (42) Google Scholar), 1 or 2 days after the cells reached confluency. Supercoiled PEPCK-CAT plasmid (2 μg) and additional carrier pBlueScript DNA (Stratagene), to make a total of 12 μg, were used, and the transfection efficiency was monitored as described previously (25Yanuka-Kashles O. Cohen H. Trus M. Aran A. Benvenisty N. Reshef L. Mol. Cell. Biol. 1994; 14: 7124-7133Crossref PubMed Scopus (42) Google Scholar). Where indicated, 1 μg each of the expression vectors for C/EBPα and PPARγ2 together with RXRα or GR was used. The optimal quantity of C/EBPβ or C/EBPδ expression vectors added to the transfection mixture was 0.5 μg each. In all cases, titration test of various concentrations of the expression vectors allowed us to choose the amounts that yielded optimal effects. Assays for chloramphenicol acetyltransferase (CAT) activity were determined 44 h after transfection, as described (25Yanuka-Kashles O. Cohen H. Trus M. Aran A. Benvenisty N. Reshef L. Mol. Cell. Biol. 1994; 14: 7124-7133Crossref PubMed Scopus (42) Google Scholar), and quantified using a PhosphorImager apparatus. The S.E. was calculated as the S.D. divided by the square root of the number of experiments. The previously described plasmid 597-pck-CAT (PCK (600)-CAT) contains 597 bp of the rat PEPCK-C gene promoter region fused to the CAT reporter gene (26Benvenisty N. Nechushtan H. Cohen H. Reshef L. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 1118-1122Crossref PubMed Scopus (15) Google Scholar). The derived 2000-pck-CAT plasmid (PCK(2000)-CAT) contains 2000 bp of the PEPCK-C gene promoter (27Cassuto H. Aran A. Cohen H. Eisenberger C.L. Reshef L. FEBS Lett. 1999; 457: 441-444Crossref PubMed Scopus (16) Google Scholar). The mutated AF1 (AF1-mut), the combined mutation of GRE1 and GRE2 (mGRE1–2) (28Liu J. Roesler W.J. Hanson R.W. BioTechniques. 1990; 9: 738-741PubMed Google Scholar), and the mutated AF2 (AF2-mut) (29Lechner P.S. Croniger C.M. Hakimi P. Millward C. Fekter C. Yun J.S. Hanson R.W. J. Biol. Chem. 2001; 276: 22675-22679Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar) plasmids contain 2000 bp of the rat PEPCK-C gene promoter mutated at these sites. The mouse promoters PCK(840)-CAT and PCK(1500)-CAT contain 840 or 1500 bp upstream from the transcription start site of PEPCK-C gene promoter, derived from a mouse genomic clone, and fused to the CAT reporter gene as described previously (26Benvenisty N. Nechushtan H. Cohen H. Reshef L. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 1118-1122Crossref PubMed Scopus (15) Google Scholar). The PCK(1500-mut)-CAT plasmid contains a site-specific mutation of the PPARE sequence. The mutation was generated by inserting the restriction sites XhoI and SmaI 3′ and 5′, respectively, to the PPARE site. This enabled us to replace the PPARE site with 47 bp of the pBlueScript polylinker residing between XhoI and SmaI, except that the EcoRI site of the polylinker insert was deleted. The DNA concentration of the constructs containing the longer PEPCK-C gene promoters was corrected to achieve the same number of molecules as that of the shorter PEPCK-C gene promoters. Expression vectors encoding adipocyte-enriched transcription factors used in this work included the following: C/EBPα (30Friedman A.D. Landschultz W.H. McKnight S.L. Genes Dev. 1989; 3: 767-776Crossref Scopus (363) Google Scholar) obtained from Dr. Steven McKnight; C/EBPβ (31Ron D. Habener J.F. Genes Dev. 1992; 6: 439-453Crossref PubMed Scopus (986) Google Scholar) from Dr. David Ron; C/EBPδ (32MacDougald O.A. Cornelius P. Lin F.-T. Chen S.S. Lane M.D. J. Biol. Chem. 1994; 269: 19041-19047Abstract Full Text PDF PubMed Google Scholar) from Dr. Daniel Lane; PPARγ2 (33Tontonoz P. Hu E. Devine J. Beale E.G. Spiegelman B.M. Mol. Cell. Biol. 1995; 15: 351-357Crossref PubMed Google Scholar) from Dr. Bruce Spiegelman; RXRα (34Mangelsdorf D.J. Ong E.S. Dyck J.A. Evans R.M. Nature. 1990; 345: 224-229Crossref PubMed Scopus (1258) Google Scholar) and human GR (35Giguere V. Hollenberg S.M. Rosenfeld M.G. Evans R.M. Cell. 1986; 46: 645-652Abstract Full Text PDF PubMed Scopus (678) Google Scholar) from the laboratory of Dr. Ron Evans; and rat wild type and mutant GR were a gift from Dr. Keith Yamamoto (13Diamond M.L. Miner J.N. Yoshinaga S.K. Yamamoto K.R. Science. 1990; 249: 1266-1272Crossref PubMed Scopus (1070) Google Scholar). The addition of glucocorticoids to fully differentiated 3T3-F442A adipocytes for 16–18 h caused a strong repression of PEPCK-C gene expression as shown by the absence of its RT-PCR product (Fig.1). This repression was distinct because glucocorticoids failed to inhibit the expression of the adipocyte-specific aP2 gene (Fig. 1). In order to assess whether GR affects the binding of nuclear proteins to specific sites in the PEPCK-C gene promoter, we systematically footprinted the gene promoter, using nuclear proteins extracted from 3T3-F442A adipocyte cells that had been incubated without or with glucocorticoids. A scheme of PEPCK-C gene promoter depicting the binding sites of transcription factors is shown (Fig. 2a). The figure also includes the binding sites for nuclear proteins present in the liver, fetal liver (where the gene for PEPCK-C is not actively transcribed (22Trus M. Benvenisty N. Cohen H. Reshef L. Mol. Cell. Biol. 1990; 10: 2418-2422Crossref PubMed Scopus (32) Google Scholar, 36Benvenisty N. Reshef L. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 1132-1136Crossref PubMed Scopus (29) Google Scholar)), kidney, and adipocytes (Fig. 2b). Note that nuclear proteins from the liver and adipocytes bind to most sites included within positions −70 to −1200 of the transcription start site of the PEPCK-C gene promoter. Adipocyte nuclear proteins do not bind to the P2 site (HNF-1 recognition motif) in the PEPCK-C gene promoter (37Cassuto H. Olswang Y. Livoff A.F. Nechushtan H. Hanson R.W. Reshef L. FEBS Lett. 1997; 412: 597-602Crossref PubMed Scopus (32) Google Scholar), whereas nuclear proteins from the liver bind poorly to PPARE (the PPAR recognition site (see Fig.3b)). Unlike the kidney and fetal liver, nuclear proteins from adipocytes and liver bind all C/EBP recognition sites. These include the CRE-1 (cAMP-response element), P3I, P4, and CRE-2 sites (CRE-2 is a CRE-like sequence). Of these, P3I binds exclusively isoforms of the C/EBP family (38Park E.A. Gurney A.L. Nizielski S.E. Hakimi P. Cao Z. Moorman A. Hanson R.W. J. Biol. Chem. 1993; 268: 613-619Abstract Full Text PDF PubMed Google Scholar). CRE-1 and P4 sites bind AP1 as well, and CRE-1 also binds cAMP-response element-binding protein, ATF-2 (39Feife L.E. Obexer P. Andratsch M. Euler S. Taylor L. Tang A. Wei Y. Schramek H. Curthoys N.P. Gstraunthaler G. Am. J. Physiol. 2002; 283: F678-F688Google Scholar), and ATF-3 (40Allen-Jennings A.E. Hartman M.G. Kociba G.J. Hai T. J. Biol. Chem. 2002; 277: 20020-20025Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar). CRE-2 binds C/EBPα with very low affinity but binds a nonidentified temperature-labile protein in the liver nuclear proteins (22Trus M. Benvenisty N. Cohen H. Reshef L. Mol. Cell. Biol. 1990; 10: 2418-2422Crossref PubMed Scopus (32) Google Scholar). Our footprinting analysis revealed that nuclear proteins extracted from glucocorticoid-treated adipocytes failed to bind to all C/EBP recognition sites in the PEPCK-C gene promoter: CRE-1 (Fig.2c), P3 and P4 (Fig. 2d), and the single C/EBP-like CRE-2 site (Fig. 2c). The prominent inhibition of binding to all these sites occurred despite the wide spectrum of affinities of the C/EBP recognition sites to C/EBPα, which gradually decreases from the highest affinity (CRE-1 site) to the lowest (CRE-2 site) (22Trus M. Benvenisty N. Cohen H. Reshef L. Mol. Cell. Biol. 1990; 10: 2418-2422Crossref PubMed Scopus (32) Google Scholar, 41Park E.A. Roesler W.J. Liu J. Klemm D.J. Gurney A.L. Thatcher J.D. Shuman J. Friedman A. Hanson R.W. Mol. Cell. Biol. 1990; 10: 6264-6272Crossref PubMed Scopus (174) Google Scholar). In addition, binding to the P1 site (nuclear factor 1 recognition site) in the PEPCK-C gene promoter was also inhibited (Fig.2c), although this is not a C/EBP recognition site. In previous studies (38Park E.A. Gurney A.L. Nizielski S.E. Hakimi P. Cao Z. Moorman A. Hanson R.W. J. Biol. Chem. 1993; 268: 613-619Abstract Full Text PDF PubMed Google Scholar) the binding of C/EBPβ to CRE-1 has been shown to cooperate with the binding to the P1 site. Therefore, a reduced binding to CRE-1 site might have led to a secondary effect on the binding to the P1 site. The binding of adipocyte nuclear proteins to the recognition sites of nuclear receptors in the PEPCK-C gene promoter has not been affected by the glucocorticoid treatment. Thus, binding to AF1 site (also termed P6) remained similar whether using nuclear proteins from adipocytes not treated or treated with glucocorticoids (Fig. 3a). AF1 site is a nuclear receptor recognition site that interacts with a variety of nonsteroid nuclear receptors including HNF4, COUPTF, retinoic acid receptor, RXR, and members of the PPAR family. Similar to adipocytes, hepatic nuclear proteins also interact well with the AF1 site (Fig.3a). Another non-C/EBP-binding site, which likewise has not been affected by the glucocorticoid treatment, is PPARE (the recognition site of the heterodimer nuclear receptors PPARγ2/RXR) (Fig. 3b). In contrast to the adipocyte nuclear proteins, binding of hepatic nuclear proteins to the PPARE site could barely be detected (Fig.3b), unlike their efficient binding to the AF1 site (Fig.3a). Furthermore, the hypersensitive site (position −985) at the 3′ end of the protected region appeared only in the presence of adipocyte nuclear proteins. It was undetectable both in the absence of nuclear proteins and in the presence of liver nuclear proteins (Fig.3b). Therefore, PPAR isoforms seem less enriched in the liver (from nonfasted rat) than they are in adipocyte nuclear proteins. To gain quantitative estimation of the glucocorticoid effect, we measured the density signals of specific bands inside and outside the protected regions of the CRE-1 (Fig. 2a) and AF1 (Fig.3a) sites. The footprinting intensity of the CRE-1 site was quantified from the autoradiographic films by measuring the density signal of a band within the CRE-1 site (position −88 from the transcription start site of the PEPCK-C gene) and a band outside, at position −160 (both marked by arrows). Likewise, the footprinting intensity of the AF1 site was quantified by measuring the density signals of two bands inside (positions −445 and −446) and a band outside (position −457) the AF1 site. The ratios between the density signals inside and outside the protected region were computed. The ratios obtained from the footprinting done without nuclear proteins were arbitrarily set at 10 and used to normalize other ratios of the footprinting done in the presence of nuclear proteins. These measurements have clearly assessed that dexamethasone treatment interfered with the adipocyte nuclear protein footprinting of the CRE-1 site (Fig. 3c) but not with the footprinting of the AF1 site (Fig. 3d). Transcription of the gene for C/EBPα is also repressed by glucocorticoids but only for a few hours (32MacDougald O.A. Cornelius P. Lin F.-T. Chen S.S. Lane M.D. J. Biol. Chem. 1994; 269: 19041-19047Abstract Full Text PDF PubMed Google Scholar). Yet this repression might lead to a longer lasting reduction of C/EBPα protein concentration in adipocyte nuclei (32MacDougald O.A. Cornelius P. Lin F.-T. Chen S.S. Lane M.D. J. Biol. Chem. 1994; 269: 19041-19047Abstract Full Text PDF PubMed Google Scholar), resulting in an apparent, rather than real, interference of binding to its recognition sites. Therefore, we determined whether the concentration of C/EBPα in nuclei corresponded to the observed diminished footprinting of the PEPCK-C gene promoter. The concentration of C/EBPα in the adipocyte extracts of nuclear proteins used for footprinting was determined by Western blot assay. The hormonal treatment diminished the nuclear concentration of C/EBPα by 30% when normalized to the level of the Y12 nuclear protein (Fig. 4a). Whether this reduced concentration of C/EBPα accounted for the inhibited footprinting of nuclear proteins from glucocorticoid-treated adipocytes was assessed. Thus, we used lower amounts of nuclear proteins extracted from untreated adipocytes (10 (2/3) and 7.5 μg (1/2)), compared with the whole amount (15 μg), to footprint the gene promoter region containing the CRE-1, P1, and CRE-2 sites. This region enabled us to assay the site with the highest affinity for C/EBP binding (CRE-1) and the site with the lowest affinity for C/EBP binding (CRE-2). The analysis showed that 10 μg of nuclear proteins footprinted all three sites (Fig. 4b). Footprinting of the CRE-1 site was quantified as detailed above for Fig. 2c. Thus, the ratio in the absence of proteins was set at 10, relative to a ratio of 2.5 obtained with 10 μg of protein and a ratio of 8.5 obtained with 7.5 μg of protein (Fig. 4c). The dexamethasone effect on CRE-1 footprinting (Fig. 2c) that was likewise quantified yielded a ratio of 1.7 in the presence of 15 μg of nuclear proteins from untreated adipocytes, and 15.8 in the presence of 15 μg of nuc
Referência(s)