Voltar
Artigo Acesso aberto Revisado por pares
BIBTEX RIS

Sp1 and CREB Mediate Gastrin-dependent Regulation of Chromogranin A Promoter Activity in Gastric Carcinoma Cells

1998; Elsevier BV; Volume: 273; Issue: 51 Linguagem: Inglês

10.1074/jbc.273.51.34000

ISSN

1083-351X

Autores

Michael Höcker, Raktima Raychowdhury, Thomas Plath, Hongjang Wu, Daniel T. O’Connor, Bertram Wiedenmann, Stefan Rosewicz, Timothy C. Wang,

Tópico(s)

Digestive system and related health

Resumo

Chromogranin A (CgA) is a multifunctional acidic protein that in the stomach is expressed predominantly in enterochromaffin-like cells (ECL cells) where it is regulated by gastrin. In order to investigate the transcriptional response of the mouse CgA (mCgA) promoter to gastrin stimulation, we studied a 4.8-kilobase mCgA promoter-luciferase reporter gene construct in transiently transfected AGS-B cells. 5′-Deletion analysis and scanning mutagenesis of mCgA 5′-flanking DNA showed that a Sp1/Egr-1 site spanning −88 to −77 base pairs (bp) and a cyclic AMP-responsive element (CRE) at −71 to −64 bp are essential for gastrin-dependent mCgA transactivation. Gastrin stimulation increased cellular Sp1 protein levels and Sp1-binding to the mCgA −88 to −77 bp element, as well as binding of CREB to its consensus motif at −71 to −64 bp. Gastrin also stimulated CREB Ser-133 phosphorylation, and abundance of cellular CREB protein levels. Overexpression of either Sp1 or phosphorylated CREB transactivated the mCgA promoter dose dependently, while coexpression of both transcription factors resulted in an additive mCgA promoter response. mCgA −92 to −64 bp, comprising the Sp1/Egr-1 site and the CRE motif, conferred gastrin responsiveness to a heterologous thymidine kinase promoter system, and therefore functions as a "true" enhancer element. This report demonstrates that Sp1 and CREB mediate CCK-B/gastrin receptor-dependent gene regulation, and that the effect of gastrin on the CgA gene is brought about by cooperative action of both transcription factors. Chromogranin A (CgA) is a multifunctional acidic protein that in the stomach is expressed predominantly in enterochromaffin-like cells (ECL cells) where it is regulated by gastrin. In order to investigate the transcriptional response of the mouse CgA (mCgA) promoter to gastrin stimulation, we studied a 4.8-kilobase mCgA promoter-luciferase reporter gene construct in transiently transfected AGS-B cells. 5′-Deletion analysis and scanning mutagenesis of mCgA 5′-flanking DNA showed that a Sp1/Egr-1 site spanning −88 to −77 base pairs (bp) and a cyclic AMP-responsive element (CRE) at −71 to −64 bp are essential for gastrin-dependent mCgA transactivation. Gastrin stimulation increased cellular Sp1 protein levels and Sp1-binding to the mCgA −88 to −77 bp element, as well as binding of CREB to its consensus motif at −71 to −64 bp. Gastrin also stimulated CREB Ser-133 phosphorylation, and abundance of cellular CREB protein levels. Overexpression of either Sp1 or phosphorylated CREB transactivated the mCgA promoter dose dependently, while coexpression of both transcription factors resulted in an additive mCgA promoter response. mCgA −92 to −64 bp, comprising the Sp1/Egr-1 site and the CRE motif, conferred gastrin responsiveness to a heterologous thymidine kinase promoter system, and therefore functions as a "true" enhancer element. This report demonstrates that Sp1 and CREB mediate CCK-B/gastrin receptor-dependent gene regulation, and that the effect of gastrin on the CgA gene is brought about by cooperative action of both transcription factors. chromogranin A enterochromaffin-like cells histidine decarboxylase extracellular-signal regulated kinase cAMP-responsive element CRE-binding protein base pair(s) phorbol 12-myristate 13-acetate kilobase pair(s) thymidine kinase electrophoretic mobility shift assay mitogen-activated protein. Chromogranin A (CgA)1 is a 48-kDa acidic protein which was initially identified as the major soluble protein co-stored and released with neurotransmitters and secretory peptides in the neuroendocrine system (1O'Connor D.T. Wu H. Gill B.M. Rozansky D.J. Tang K. Mahata S.K. Mahata M. Eskeland N.L. Videen J.S. Zhang X. Takiyuddin M.A. Parmer R.J. Ann. N. Y. Acad. Sci. 1993; 729: 36-45Google Scholar, 2Bauerfeind R. Ohashi M. Hutttner W.B. Ann. N. Y. Acad. Sci. 1994; 733: 385-390Crossref Scopus (21) Google Scholar, 3Iacangelo A. Eiden L. Regul. Pept. 1995; 58: 65-88Crossref PubMed Scopus (154) Google Scholar). In neuroendocrine cells, CgA stabilizes secretory granules, influences pro-hormone processing, and regulates peptide sorting into the regulated secretory pathway (1O'Connor D.T. Wu H. Gill B.M. Rozansky D.J. Tang K. Mahata S.K. Mahata M. Eskeland N.L. Videen J.S. Zhang X. Takiyuddin M.A. Parmer R.J. Ann. N. Y. Acad. Sci. 1993; 729: 36-45Google Scholar, 2Bauerfeind R. Ohashi M. Hutttner W.B. Ann. N. Y. Acad. Sci. 1994; 733: 385-390Crossref Scopus (21) Google Scholar, 3Iacangelo A. Eiden L. Regul. Pept. 1995; 58: 65-88Crossref PubMed Scopus (154) Google Scholar). Cleavage products of CgA are able to regulate endocrine and exocrine secretory functions after release into the bloodstream (1O'Connor D.T. Wu H. Gill B.M. Rozansky D.J. Tang K. Mahata S.K. Mahata M. Eskeland N.L. Videen J.S. Zhang X. Takiyuddin M.A. Parmer R.J. Ann. N. Y. Acad. Sci. 1993; 729: 36-45Google Scholar, 2Bauerfeind R. Ohashi M. Hutttner W.B. Ann. N. Y. Acad. Sci. 1994; 733: 385-390Crossref Scopus (21) Google Scholar, 3Iacangelo A. Eiden L. Regul. Pept. 1995; 58: 65-88Crossref PubMed Scopus (154) Google Scholar). In the stomach, enterochromaffin-like cells (ECL cells) of the corpus mucosa, which are responsible for synthesis, storage, and release of histamine controlling gastric acid secretion, have been identified as the main source for CgA expression (4Cetin Y. Grube D. Cell Tissue Res. 1991; 264: 231-241Crossref PubMed Scopus (41) Google Scholar, 5Stabile B.E. Howard T.J. Passaro E. O'Connor D.T. Arch. Surg. 1990; 125: 451-453Crossref PubMed Scopus (49) Google Scholar). In vivo experiments in rodents as well as observations made in patients with gastrinoma have shown that hypergastrinemia-induced degranulation of ECL cells is accompanied by a parallel secretion of histamine and CgA, followed by enhanced production of both molecules (4Cetin Y. Grube D. Cell Tissue Res. 1991; 264: 231-241Crossref PubMed Scopus (41) Google Scholar, 5Stabile B.E. Howard T.J. Passaro E. O'Connor D.T. Arch. Surg. 1990; 125: 451-453Crossref PubMed Scopus (49) Google Scholar, 6Syversen U. Mignon M. Bonfils S. Kristensen A. Waldum H.L. Acta Oncol. 1993; 32: 161-165Crossref PubMed Scopus (67) Google Scholar, 7Dimaline R. Evans D. Forster H.L. Sandvik A.K. Dockray G.J. Am. J. Physiol. 1993; 264: G583-G588Crossref PubMed Google Scholar, 8Watkinson A. Dockray G.J. Regul. Pept. 1992; 40: 51-61Crossref PubMed Scopus (36) Google Scholar, 9Hakanson R. Ding X.-Q. Norlen P. Chen D. Gastroenterology. 1995; 108: 1445-1452Abstract Full Text PDF PubMed Scopus (96) Google Scholar). Furthermore, CgA up-regulation is also accompanied by increased synthesis of histidine decarboxylase (HDC), the key enzyme for gastric histamine synthesis (10Chen D. Monstein H.J. Nylander A.G. Zhao C.M. Sundler F. Hakanson R. Gastroenterology. 1994; 107: 18-27Abstract Full Text PDF PubMed Google Scholar). The functional importance of gastrin for the parallel expression pattern of HDC and CgA genes was supported by observations in homozygous CCK-B/gastrin receptor- and gastrin-deficient mice, as these knockout mice showed dramatic reductions in HDC and CgA mRNA expression in comparison to wild type littermates (11Nagata A. Ito M. Iwata M. Kuno J. Takano H. Minova O. Chihara K. Matsui T. Noda T. Proc. Natl. Acad. Sci. U. S. A. 1997; 93: 11825-11830Crossref Scopus (219) Google Scholar, 12Koh T.J. Goldenring J.R. Ito S. Mashimo H. Kopin A.S. Varro A. Dockray G.J. Wang T.C. Gastroenterology. 1997; 113: 1015-1025Abstract Full Text PDF PubMed Scopus (215) Google Scholar, 13Schubert M. Curr. Opin. Gastroenterol. 1996; 12: 493-502Crossref Scopus (7) Google Scholar). These observations supported the concept that replenishment of ECL cell histamine stores after a secretory challenge requires the parallel de novosynthesis of the granule matrix protein CgA in order to provide the basis for appropriate packaging and release of histamine. Histamine production in ECL cells depends on the activity of the rate-limiting enzyme of histamine synthesis, HDC (14Modlin I.V. Tang L.H. Gastroenterology. 1996; 111: 783-810Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar, 15Zhang Z. Höcker M. Koh T.J. Wang T.C. J. Biol. Chem. 1996; 271: 14188-14197Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). In a previous study we identified a downstream (+2 to +24)cis-acting element in the human HDC (hHDC) promoter through which gastrin exerts its transactivating effect (15Zhang Z. Höcker M. Koh T.J. Wang T.C. J. Biol. Chem. 1996; 271: 14188-14197Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). This HDC element shows no homology to any previously reported consensus motif, and apparently binds (a) novel transcription factor(s) (15Zhang Z. Höcker M. Koh T.J. Wang T.C. J. Biol. Chem. 1996; 271: 14188-14197Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). Analysis of signal transduction pathways revealed that MAP kinase/ERK signaling pathways are crucial for the regulation of the hHDC promoter (16Höcker M. Zhang Z. Henihan R. Rosewicz S. Riecken E.O. Koh T.J. Wang T.C. J. Biol. Chem. 1997; 272: 27015-27024Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar, 17Höcker M. Rosenberg I. Xavier R. Henihan R.J. Wiedenmann B. Rosewicz S. Podolsky D.K. Wang T.C. J. Biol. Chem. 1998; 273: 23046-23054Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar). While previous studies analyzed the structural basis for nicotine-stimulated mCgA transcriptional activity in pheocromocytoma cells (18Tang K. Wu H. Mahata S.K. Taupenot L. Rozansky D.J. Parmer R.J. O'Connor D.T. J. Biol. Chem. 1997; 271: 28382-28390Abstract Full Text Full Text PDF Scopus (55) Google Scholar, 19Tang K. Wu H. Mahata S.K. Gill B.M. Parmer R.J. O'Connor D.T. J. Clin. Invest. 1997; 100: 1180-1192Crossref PubMed Scopus (41) Google Scholar), the molecular mechanisms of gastrin-stimulated CgA expression are still unclear. Therefore, to better understand the molecular mechanisms involved in gastrin-dependent CgA gene expression, we have analyzed the regulatory elements and transcription factors involved in the control of the CgA promoter by gastrin. We find that gastrin transactivates the mCgA promoter dose and time dependently in AGS-B gastric carcinoma cells and GH3 pituitary cells. Conventional 5′-deletion analysis, in combination with scanning mutagenesis of the mCgA core promoter region, demonstrate that a DNA-stretch spanning mCgA −93 to −62 bp represents the gastrin-responsive region of the mCgA promoter. Furthermore, we show that two elements located in this region are necessary for full gastrin responsiveness: an Sp1/Egr-1 motif located at −88 to −77 bp and a CRE-like element at −71 to −64 bp. Our study demonstrates that the mCgA gene is regulated by gastrin on a transcriptional level through binding of two separate transcription factors to their consensuscis-acting elements. Oligonucleotides were provided by the Center for the Study of Inflammatory Bowel Diseases, Massachusetts General Hospital, Boston. Sp1 and CREB consensus oligonucleotides as well as anti-Sp1, anti-CREB/ATF1 antibodies were purchased from Santa Cruz Biotechnology, Santa Cruz, CA. Anti-phospho-CREB antibody was obtained from Upstate Biotechnology. The Gal4/CREB transactivator plasmid and the protein kinase A expression construct were kindly provided by Dr. J. Habener (20Uhler M.D. McKnight G.S. J. Biol. Chem. 1987; 262: 15202-15207Abstract Full Text PDF PubMed Google Scholar, 21Lee C.Q. Yun Y. Hoeffler J.P. Habener J.P. EMBO J. 1990; 9: 4455-4465Crossref PubMed Scopus (183) Google Scholar). The Gal4/CREB/A construct contains a mutation at Ser-133 so that it cannot be transactivated (20Uhler M.D. McKnight G.S. J. Biol. Chem. 1987; 262: 15202-15207Abstract Full Text PDF PubMed Google Scholar, 21Lee C.Q. Yun Y. Hoeffler J.P. Habener J.P. EMBO J. 1990; 9: 4455-4465Crossref PubMed Scopus (183) Google Scholar). The Gal4/Sp1 transactivator plasmid and the Sp1 overexpression construct were kindly provided by Dr. G. Suske (Marburg, Germany) (22Hagen G. Müller S. Beato M. Suske G. EMBO J. 1994; 13: 3843-3851Crossref PubMed Scopus (654) Google Scholar, 23Hagen G. Dennig J. Preiss A. Beato M. Suske G. J. Biol. Chem. 1995; 270: 24989-24994Abstract Full Text Full Text PDF PubMed Scopus (186) Google Scholar). The 5xGal4/Luciferase reporter construct was provided by Dr. Anil Rustgi, Massachusetts General Hospital, Boston, MA. The Sp1 overexpression plasmid has been obtained from Dr. J. Merchant (24Mortensen E.R. Marks P.A. Shiotani A. Merchant J.L. J. Biol. Chem. 1997; 272: 16540-16547Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar). Phorbol ester 12-myristate 13-acetate (PMA) was purchased from Biomol, Plymouth Meeting, PA. AGS-B gastric cancer and GH3-B neuroendocrine cells were derived from parent cells (AGS and GH3, ATCC) through stable transfection of the expression vector CCKB-pcDNAINeo, containing the full-length coding region of the human CCK-B/gastrin receptor and the neomycin gene, and have been previously described (25Höcker M. Zhang Z. Fenstermacher D.A. Tågerud S. Chulak M. Joseph D. Wang T.C. Am. J. Physiol. 1996; 270: G619-G633Crossref PubMed Google Scholar). AGS-B cells were grown in Dulbecco's modified Eagle's medium containing 10% bovine calf serum, 100 IU/ml penicillin, and 100 IU/ml streptomycin in a humidified atmosphere (5% CO2/95% air). Transient transfections of cultured AGS-B and GH3-B cells were carried out using the calcium-phosphate precipitation technique (DNA Transfection Kit, 5 Prime-3 Prime Inc.). Cells were plated at a density of 1 × 106 cells/35-mm well and transfected the next day. In general, cells were harvested and luciferase assays done at 48 h. Luciferase (Luc) assays were performed using luciferin, ATP, and coenzyme A (Promega system) with a Monolight Luminometer (Analytic Luminescence Laboratory) as described previously (16Höcker M. Zhang Z. Henihan R. Rosewicz S. Riecken E.O. Koh T.J. Wang T.C. J. Biol. Chem. 1997; 272: 27015-27024Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar, 17Höcker M. Rosenberg I. Xavier R. Henihan R.J. Wiedenmann B. Rosewicz S. Podolsky D.K. Wang T.C. J. Biol. Chem. 1998; 273: 23046-23054Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar, 25Höcker M. Zhang Z. Fenstermacher D.A. Tågerud S. Chulak M. Joseph D. Wang T.C. Am. J. Physiol. 1996; 270: G619-G633Crossref PubMed Google Scholar). Incubations were performed in triplicates or quadruplicates and results calculated as mean ± S.E. Values for mCgA-Luc activity were expressed as fold increase in luciferase activity compared with untreated controls. The empty pTK-Luc construct served as an additional control (16Höcker M. Zhang Z. Henihan R. Rosewicz S. Riecken E.O. Koh T.J. Wang T.C. J. Biol. Chem. 1997; 272: 27015-27024Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar, 17Höcker M. Rosenberg I. Xavier R. Henihan R.J. Wiedenmann B. Rosewicz S. Podolsky D.K. Wang T.C. J. Biol. Chem. 1998; 273: 23046-23054Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar, 25Höcker M. Zhang Z. Fenstermacher D.A. Tågerud S. Chulak M. Joseph D. Wang T.C. Am. J. Physiol. 1996; 270: G619-G633Crossref PubMed Google Scholar). Activities varied less than 15% between transfection experiments. Expression of human growth hormone from the plasmid vector pXGH5, containing the human growth hormone gene under the control of the metallothionein-I promoter, was used as an internal control for transfection efficiency (16Höcker M. Zhang Z. Henihan R. Rosewicz S. Riecken E.O. Koh T.J. Wang T.C. J. Biol. Chem. 1997; 272: 27015-27024Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar, 17Höcker M. Rosenberg I. Xavier R. Henihan R.J. Wiedenmann B. Rosewicz S. Podolsky D.K. Wang T.C. J. Biol. Chem. 1998; 273: 23046-23054Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar, 25Höcker M. Zhang Z. Fenstermacher D.A. Tågerud S. Chulak M. Joseph D. Wang T.C. Am. J. Physiol. 1996; 270: G619-G633Crossref PubMed Google Scholar). The mCgA 5′-deletion constructs, which are all based on the promoterless luciferase reporter gene vector pXP1, have been previously reported (26Wu H. Mahata S.K. Rozansky D.J. Webster N.J.G. Parmer R.J. O'Connor D.T. J. Clin. Invest. 1995; 96: 568-578Crossref PubMed Scopus (47) Google Scholar). A series of scanning mutants between mCgA −100 and −49 bp were prepared by polymerase chain reaction amplification using the pXP100 construct as a template, employing different mutated 5′-oligonucleotide primers and a common 3′ primer (26Wu H. Mahata S.K. Rozansky D.J. Webster N.J.G. Parmer R.J. O'Connor D.T. J. Clin. Invest. 1995; 96: 568-578Crossref PubMed Scopus (47) Google Scholar). The mutant fragments were subcloned intoXhoI and HindIII sites of pXP1 and mutant sequences were confirmed by dideoxy sequencing (26Wu H. Mahata S.K. Rozansky D.J. Webster N.J.G. Parmer R.J. O'Connor D.T. J. Clin. Invest. 1995; 96: 568-578Crossref PubMed Scopus (47) Google Scholar). In order to study the characteristics of potential mCgA regulatory elements in a heterologous promoter system, the region mCgA −93 to −62 bp was subcloned either adjacent or 1.2 kb away from the enhancerless herpes simplex thymidine kinase (TK) viral promoter into the plasmid pTK-Luc (26Wu H. Mahata S.K. Rozansky D.J. Webster N.J.G. Parmer R.J. O'Connor D.T. J. Clin. Invest. 1995; 96: 568-578Crossref PubMed Scopus (47) Google Scholar). Nuclear extracts from AGS-B cells were prepared as described (15Zhang Z. Höcker M. Koh T.J. Wang T.C. J. Biol. Chem. 1996; 271: 14188-14197Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). In brief, double-stranded oligonucleotides were radiolabeled with [α-32P]dCTP and EMSAs were performed with 5 μg of nuclear extracts in a final volume of 20 μl of binding buffer containing 10 mm Tris-HCl (pH 7.5), 50 mm NaCl, 5 mm MgCl2, 1 mm dithiothreitol, 1 mm EDTA, 1 μg of poly(dA-dT), and 10% glycerol. Mixtures were incubated with 10 fmol of double stranded oligonucleotide probes for 20 min at room temperature. DNA-protein complexes were electrophoresed on a 6% nondenaturing polyacrylamide gel containing 0.25 × TBE at a constant current of 15 mA. Gels were dried and exposed to Kodak X-AR films at room temperature. Gel shift assays required 10 μmZnSO4 for complete Sp1 binding. For competition experiments, nuclear extracts were incubated with a 100-fold excess of double-stranded competitor oligos at room temperature for 10 min before addition of radiolabeled probes. For supershift experiments, nuclear extracts were incubated with 1 μl of anti-Sp1 or anti-CREB antibodies for 10 min at room temperature, followed by an incubation period of 20 min at 4 °C prior to addition of radiolabeled probes. Preparation of cell lysates and SDS-polyacrylamide gel electrophoresis analysis was performed as described previously (16Höcker M. Zhang Z. Henihan R. Rosewicz S. Riecken E.O. Koh T.J. Wang T.C. J. Biol. Chem. 1997; 272: 27015-27024Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar, 17Höcker M. Rosenberg I. Xavier R. Henihan R.J. Wiedenmann B. Rosewicz S. Podolsky D.K. Wang T.C. J. Biol. Chem. 1998; 273: 23046-23054Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar). In brief, AGS-B cells were washed with ice-cold phosphate-buffered saline twice and lysed with ice-cold lysis buffer containing 10 mm HEPES (pH 7.4), 30 mmNaCl, 2% glycerol, 0.2% Triton X-100, 0.3 mmMgCl2, 0.2 mm EDTA, 1 mmphenylmethylsulfonyl fluoride, 1 μg/ml leupeptin, 1 mmsodium orthovanadate, 1 mm sodium fluoride. 50 μg of the lysates were electrophoresed on 8 or 10% SDS-polyacrylamide gels for detection of Sp1 or CREB proteins, respectively. After gel separation, proteins were blotted on nitrocellulose membranes and visualized by the enhanced chemiluminescence (ECL, Amersham) method according to the instructions of the manufacturer using a 1:1000 dilution of antibodies. In AGS-B cells, gastrin stimulated mCgA promoter activity time and dose dependently (Fig. 1), whereas activity of pTK-Luc was not significantly influenced (not shown). Essentially identical kinetics were found for gastrin activation of the mCgA4.8kb-Luc construct in pituitary GH3-B cells (data not shown). Since activation of the mCgA promoter by cAMP-mediated pathways has been established in bovine medullary cells, we investigated the interaction of this signaling route with gastrin- and PMA-activated pathways. In AGS-B cells, dibutyryl-cAMP stimulated mCgA promoter activity 2–3-fold, and co-stimulation with gastrin or PMA and db-cAMP showed additive responses (Fig. 2). 5′-Deletion analysis of mCgA 5′-flanking DNA revealed that 100 bp upstream of the mCgA cap site (+1) are able to confer full responsiveness to gastrin, and that further 5′-deletion to mCgA −77 abrogated gastrin responsiveness (Fig. 3 A). To further characterize the region downstream of mCgA −77 functionally, a dense series of scanning mutants was employed in transient transfection experiments. Analysis of scanning mutants demonstrated that two mCgA promoter regions (mutants M6 and M13) are essential for gastrin-dependent transactivation: a DNA-stretch spanning −88 to −77 bp, comprising putative binding sites for Sp1 and Egr-1, and the region between −71 and −64, bp which represents a consensus CRE element (Fig. 3 B).Figure 2mCgA promoter activity after stimulation of different intracellular signaling pathways. AGS-B cells transiently transfected with the mCgA4.8kb-Luc construct (0.5 μg/well) were incubated with or without 10−8m forskolin (Fors.), 10−5m dibutyryl (db)-cAMP, 10−8m gastrin, 10−8m PMA, or a combination of these compounds. Luciferase activity is expressed as a fold increase relative to unstimulated controls and results represent the mean ± of four separate experiments. All stimulated groups showed a statistically significant difference (p < 0.05) compared with unstimulated controls using Student's ttest.View Large Image Figure ViewerDownload (PPT)Figure 3Analysis of mCgA 5′-deletion mutants and "scannining mutagenesis" of the mCgA core promoter in AGS-B cells. A, AGS-B cells were transiently transfected with mCgA 5′-deletion mutants of different length (0.5–1.0 μg/well) and incubated with or without gastrin (10−8m) for 24 h. B, a dense series of scanning mutants of the core promoter region (−100 to −49 bp) was generated by an polymerase chain reaction-based approach (using the construct PxP100-Luc as a template) and transiently transfected into AGS-B cells (0.5–1.0 μg/well). After transfection, gastrin stimulation was performed for 24 h and cell lysates were subsequently analyzed for luciferase activity. Luciferase activity is expressed as a fold increase relative to unstimulated controls and represents the mean ± of four separate experiments. Please note: the mutation in mCgA-M14 converts the mCgA-CRE element to a consensus CRE site.View Large Image Figure ViewerDownload (PPT) To investigate whether the region mCgA −93 to −62 bp is able to confer gastrin responsiveness to a gastrin-insensitive heterologous promoter system, we employed the pTK-Luc construct. In this construct, expression of the luciferase reporter gene is under control of the enhancerless herpes simplex virus thymidine kinase minimal promoter. When constructs in which the mCgA −93 to −62 bp region was subcloned either adjacent or 1.2 kb away from the TK promoter were transfected into AGS-B cells, gastrin stimulation elicited a 2.5–4.5-fold increase in reporter gene activity (Fig. 4). EMSA analysis revealed that in AGS-B cells two nuclear complexes appear to bind to mCgA −93 to −62 bp, the complete element able to confer gastrin responsiveness to a heterologous promoter system. The "A" probe, which represents the complete sequence of this element, is bound by 2 major complexes (Fig.5 B). In experiments using an upstream ("B" probe, mCgA −93 to −73 bp) or downstream probe("C" probe, mCgA −76 to −62 bp), the transcription factor in complex A was identified as Sp1 (Fig. 5 C), whereas in complex B the probe is bound by CREB (Fig. 5 D). EMSA analysis of nuclear extracts obtained from AGS-B cells after 10 min of gastrin or PMA stimulation revealed that this treatment increased the binding of Sp1 to its upstream binding site (Fig. 5 C). Since CREB-dependent transactivation is generally brought about by phosphorylation of the transcription factor at residue serine 133, we performed EMSA supershift experiments employing an anti-phospho-CREB antibody, which is directed against the serine 133 epitope and exclusively recognizes phosphorylated CREB. We found enhanced binding of phosphorylated CREB in response to gastrin and PMA after 10 min of treatment (Fig. 5 D). A similar picture emerged when supershift experiments were performed with an antibody identifying CREB proteins also in the unphosphorylated state (Fig. 5 D). To investigate whether increased binding of Sp1 and CREB to the CgA promoter is accompanied by elevated cellular levels of both factors, we analyzed basal and gastrin-stimulated CREB and Sp1 abundance in AGS-B cells by Western blot analysis. We found that gastrin dose dependently stimulated Sp1 abundance, having a pronounced effect as early as 5–10 min after stimulation (Fig.6 A). Similar results were obtained when total lysates from gastrin-stimulated AGS-B cells were analyzed for the abundance of phophorylated CREB (Fig. 6 B). Maximal elevation in the abundance of phophorylated CREB was observed after 10 min of gastrin stimulation. To demonstrate that elevation of cellular Sp1 and CREB levels has functional impact on the transcriptional activity of the CgA promoter, we transiently overexpressed Sp1 or phosphorylated CREB by transfection of appropriate expression vectors into AGS-B cells. Phosphorylation of transfected CREB was achieved by co-transfection of pCMV-CREB with an expression construct for the catalytic subunit (C) of protein kinase A. This strategy has been shown to result in effective phosphorylation of recombinant CREB proteins encoded by the CREB expression construct (20Uhler M.D. McKnight G.S. J. Biol. Chem. 1987; 262: 15202-15207Abstract Full Text PDF PubMed Google Scholar, 21Lee C.Q. Yun Y. Hoeffler J.P. Habener J.P. EMBO J. 1990; 9: 4455-4465Crossref PubMed Scopus (183) Google Scholar). When increasing amounts of these constructs were co-transfected with mCgA100-Luc into AGS-B cells, reporter gene activity was dose dependently stimulated (Fig.7 B). Similarily, mCgA reporter gene activity was dose dependently stimulated by transfecting increasing amounts of a full-length human Sp1 expression construct (23Hagen G. Dennig J. Preiss A. Beato M. Suske G. J. Biol. Chem. 1995; 270: 24989-24994Abstract Full Text Full Text PDF PubMed Scopus (186) Google Scholar) together with mCgA4100-Luc (Fig. 7 A). When Sp1 and CREB overexpression was performed in cells transfected with the mCgA −93 to −62 bp TK-Luc construct, coexpression of both factors was more effective than expression of either factor alone, indicating positive cooperative action of Sp1 and CREB in the regulation of the mCgA promoter (Fig. 7 C). After EMSA and Western blot analysis demonstrated that gastrin stimulation of AGS-B cells results in increased abundance of Sp1 and phosphorylated CREB, we next investigated the effect of gastrin on CREB- or Sp1-dependent transactivation in the Gal4/Gal4–5xLuciferase system. The Gal4-CREB transactivator plasmid contains the regulatory sequence of CREB, which comprises the serine 133 residue, critical for its phosphorylation. In the Gal4-CREB/A construct, the serine residue at position 133 is replaced by an arginine which cannot be phosphorylated. The Gal4/Sp1 transactivator plasmid has previously been described (20Uhler M.D. McKnight G.S. J. Biol. Chem. 1987; 262: 15202-15207Abstract Full Text PDF PubMed Google Scholar, 21Lee C.Q. Yun Y. Hoeffler J.P. Habener J.P. EMBO J. 1990; 9: 4455-4465Crossref PubMed Scopus (183) Google Scholar). After co-transfection of Gal4-transactivator and Gal4-reporter plasmids, binding of Gal4-CREB or Gal4-Sp1 fusion proteins encoded by the transactivator plasmids to the Gal4-binding domain of the reporter plasmid depends on the phosphorylation state of subcloned CREB or Sp1 sequences. In the reporter plasmid 5xGal4-Luciferase, expression of the luciferase reporter gene is under control of a multimer (five copies) of the Gal4-binding domain. Stimulation of AGS-B cells transfected with Gal4-CREB or Gal4-Sp1 transactivator constructs together with the reporter construct Gal4/5xGal4-Luc resulted in a 2.5- or 1.6-fold increase in reporter gene activity, respectively (Fig.8). Gastrin had no effect on Gal4-CREB/A-dependent transactivation. These results demonstrate that in AGS-B cells, gastrin stimulates CREB-dependent transactivation through enhanced phosphorylation of CREB. Using the Gal4-Sp1 system, we found that gastrin also had a modest stimulatory effect on Sp1-dependent transactivation, raising the possibility of post-translational modification of Sp1 in response to gastrin. The purpose of this study was to characterize the regulatory effect of gastrin on the mCgA promoter, along with thecis-regulatory elements and transcription factors mediating the actions of gastrin. We found that gastrin transactivates the mCgA promoter dose and time dependently in both AGS-B gastric carcinoma cells and GH3 pituitary cells. Conventional 5′-deletion analysis, in combination with scanning mutagenesis of the mCgA proximal core promoter region, demonstrated that a DNA-stretch spanning −93 to −62 bp represents the gastrin-responsive region of the mCgA promoter, and that two regulatory elements located in this

Referência(s)