The Cyclic AMP Response Element Modulator Family Regulates the Insulin Gene Transcription by Interacting with Transcription Factor IID

Abstract

We analyzed a mechanism of transcriptional regulation of the human insulin gene by cyclic AMP response element modulator (CREM) through four cyclic AMP response elements (CREs). We isolated two novel CREM isoforms (CREMΔQ1 and CREMΔQ2), which lack one of the glutamine-rich domains, Q1 and Q2 respectively, and six known isoforms (CREMτα, CREMα, inducible cyclic AMP early repressor (ICER) I, ICER Iγ, CREM-17X, and CREM-17) from rat pancreatic islets and the RINm5F pancreatic β-cell line. CREM isoforms functioned as efficient transcriptional activators or repressors to modulate insulin promoter activity by binding to all of the insulin CREs. The binding activity of repressors is higher than that of activators and suppressed not only basal activity but also activator-induced activities. Furthermore, CREM activator interacted directly with the transcription factor IID components hTAFII130 and TATA box-binding protein (TBP). These results suggest that the activation of the insulin gene transcription by CREM activator is mediated by not only direct binding to the CREs but also by recruiting transcription factor IID to the insulin promoter via its interaction with hTAFII130 and TBP. On the other hand, the CREM repressor ICER competitively interrupts the binding of the activators to CREs and does not interact with either TBP or hTAFII130; therefore, it might fail to stabilize the basal transcriptional machinery and repress transactivation. We analyzed a mechanism of transcriptional regulation of the human insulin gene by cyclic AMP response element modulator (CREM) through four cyclic AMP response elements (CREs). We isolated two novel CREM isoforms (CREMΔQ1 and CREMΔQ2), which lack one of the glutamine-rich domains, Q1 and Q2 respectively, and six known isoforms (CREMτα, CREMα, inducible cyclic AMP early repressor (ICER) I, ICER Iγ, CREM-17X, and CREM-17) from rat pancreatic islets and the RINm5F pancreatic β-cell line. CREM isoforms functioned as efficient transcriptional activators or repressors to modulate insulin promoter activity by binding to all of the insulin CREs. The binding activity of repressors is higher than that of activators and suppressed not only basal activity but also activator-induced activities. Furthermore, CREM activator interacted directly with the transcription factor IID components hTAFII130 and TATA box-binding protein (TBP). These results suggest that the activation of the insulin gene transcription by CREM activator is mediated by not only direct binding to the CREs but also by recruiting transcription factor IID to the insulin promoter via its interaction with hTAFII130 and TBP. On the other hand, the CREM repressor ICER competitively interrupts the binding of the activators to CREs and does not interact with either TBP or hTAFII130; therefore, it might fail to stabilize the basal transcriptional machinery and repress transactivation. Insulin gene transcription is regulated throughcis-acting elements in response to stimulation, such as glucose concentration or increased cAMP level (1Sander M. German M.S. J. Mol. Med. 1997; 75: 327-340Crossref PubMed Scopus (285) Google Scholar). One of thecis-acting elements, cAMP response element (CRE), 1The abbreviations used are: CRE, cyclic AMP response element; CREM, cyclic AMP response element modulator; CREB, cyclic AMP response element-binding protein; ICER, inducible cyclic AMP early repressor; nt, nucleotide(s); TBP, TATA box-binding protein; hTBP, human TBP; TAF, TBP-associated factor; hTAF, human TAF; GST, glutathione S-transferase; PAGE, polyacrylamide gel electrophoresis was first identified as an inducible enhancer of genes that can be transcribed in response to increased cAMP levels (2Comb M. Birnberg C. Seasholtz A. Herbert E. Goodman H.M. Nature. 1986; 323: 353-356Crossref PubMed Scopus (525) Google Scholar, 3Montminy M.R. Sevarino K.A. Wangner J.A. Mandel G. Goodman R.H Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 6682-6686Crossref PubMed Scopus (1056) Google Scholar). In a previous study, we identified four functional CREs in the human insulin gene and demonstrated that each of these CREs has different sequences from the others and also from the consensus CRE motif (TGACGTCA) (4Inagaki N. Maekawa T. Sudo T. Ishii S. Seino Y. Imura H. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 1045-1049Crossref PubMed Scopus (88) Google Scholar). Since a remarkable reduction of insulin gene transcription is observed when CREs are mutated (4Inagaki N. Maekawa T. Sudo T. Ishii S. Seino Y. Imura H. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 1045-1049Crossref PubMed Scopus (88) Google Scholar), transcription factors that bind to this regulatory element might play an important role in the regulation of insulin gene transcription. However, which transcription factors are primarily involved in the regulation of insulin gene transcription through CREs is unclear. Transcription factors such as CREB, CRE-BP1, and ATF1, which belong to the ATF/CREB family, bind to CRE (5Hoeffler J.P. Meyer T.E. Yun Y. Jameson J.L. Habener J.F. Science. 1988; 242: 1430-1433Crossref PubMed Scopus (526) Google Scholar, 6Gonzalez G.A. Yamamoto K.K. Fischer W.H. Karr K. Menzel P. Biggs III, W. Vale W.W. Montminy M.R. Nature. 1989; 337: 749-752Crossref PubMed Scopus (650) Google Scholar, 7Maekawa T. Sakura H. Kanei-Ishii C. Sudo T. Yoshimura T. Fujisawa J. Yoshida M. Ishii S. EMBO J. 1989; 8: 2023-2028Crossref PubMed Scopus (292) Google Scholar, 8Hai T. Liu F. Coukos W.J. Green M.R. Genes Dev. 1989; 3: 2083-2090Crossref PubMed Scopus (760) Google Scholar). CRE modulator (CREM) is a member of this family, and the CREM gene generates both transcriptional activators (CREMτ, -τα, -τ1, and -τ2) and repressors (CREMα, CREMβ, CREMγ, S-CREM, CREM-17X, and CREM-17) by alternative splicing (9Sassone-Corsi P. EMBO J. 1994; 13: 4717-4728Crossref PubMed Scopus (68) Google Scholar, 10Foulkes N.S. Sassone-Corsi P. Cell. 1992; 68: 411-414Abstract Full Text PDF PubMed Scopus (222) Google Scholar, 11Foulkes N.S. Borrelli E. Sassone-Corsi P. Cell. 1991; 64: 739-749Abstract Full Text PDF PubMed Scopus (534) Google Scholar, 12Laoide B.M. Foulkes N.S. Schlotter F. Sassone-Corsi P. EMBO J. 1993; 12: 1179-1191Crossref PubMed Scopus (188) Google Scholar, 13Inada A. Yamada Y. Someya Y. Kubota A. Yasuda K. Ihara Y. Kagimoto S. Kuroe A. Tsuda K. Seino Y. Biochem. Biophys. Res. Commun. 1998; 253: 712-718Crossref PubMed Scopus (32) Google Scholar, 14Gellersen B. Kempf R. Telgmann R. Mol. Endocrinol. 1997; 11: 97-113Crossref PubMed Scopus (62) Google Scholar). In addition, use of an intronic promoter (P2) generates inducible cAMP early repressors (ICERs) (15Stehle J.H. Foulkes N.S. Molina C.A. Simonneaux V. Pévet P. Sassone-and Corsi P. Nature. 1993; 365: 314-320Crossref PubMed Scopus (356) Google Scholar, 16Molina C.A. Foulkes N.S. Lalli E. Sassone-Corsi P. Cell. 1993; 75: 875-886Abstract Full Text PDF PubMed Scopus (527) Google Scholar). This property is unique to CREM within this family of transcription factors. CREMτ and CREMτα contain the phosphorylation domain (P-box) and two glutamine-rich domains (Q1 and Q2), which may be essential for gene activation and function as transcriptional activators. The CREM repressors, CREMα, -β, and -γ, contain the P-box but lack both Q1 and Q2. Thus, by a shuffling of exons, various combinations of functional domains are generated, and a large number of functionally different CREM isoforms are produced (12Laoide B.M. Foulkes N.S. Schlotter F. Sassone-Corsi P. EMBO J. 1993; 12: 1179-1191Crossref PubMed Scopus (188) Google Scholar). These isoforms have been well studied in the hypothalamic-pituitary-gonadal axis, where they play key physiological and developmental roles by regulating gene transcription through CREs (17Foulkes N.S. Mellström B. Benusiglio E. Sassone-Corsi P. Nature. 1992; 355: 80-84Crossref PubMed Scopus (409) Google Scholar, 18Delmas V. Van Der Hoon F. Mellström B. Jégou B. Sassone-Corsi P. Mol. Endocrinol. 1993; 7: 1502-1514Crossref PubMed Scopus (187) Google Scholar, 19Foulkes N.S. Borjigin J. Snyder S.H. Sassone-Corsi P. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 14140-14145Crossref PubMed Scopus (136) Google Scholar, 20Fitzgerald L.R. Li Z. Machida C.A. Fishman P.H. Duman R.S. J. Neurochem. 1996; 67: 490-497Crossref PubMed Scopus (27) Google Scholar, 21Monaco L. Foulkes N.S. Sassone-Corsi P. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 10673-10677Crossref PubMed Scopus (102) Google Scholar, 22Deadouets C. Matesic G. Molina C.A. Foulkes N.S. Sassone-Corsi P. Brechot C. Sobczak-Thepot J. Mol. Cell. Biol. 1995; 15: 3301-3309Crossref PubMed Scopus (218) Google Scholar). Since the human insulin gene also possesses CREs, CREM could be a candidate for promoting insulin gene transcription. Recent studies have demonstrated direct interactions of site-specific transcriptional factors and TFIID (23Chiang C-M. Roeder R.G. Science. 1995; 267: 531-536Crossref PubMed Scopus (352) Google Scholar, 24Tanese N. Pugh B.F. Tjian R. Genes Dev. 1991; 5: 2212-2224Crossref PubMed Scopus (244) Google Scholar, 25Tanese N. Saluja D. Vassallo M. Chen J-L. Admon A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13611-13616Crossref PubMed Scopus (118) Google Scholar), a multiprotein of the TATA box-binding protein and TBP-associated factors R. Sci. 1995; PubMed Scopus Google Scholar, J.A. G. Tjian R. Cell. 1996; Full Text Full Text PDF PubMed Scopus Google Scholar, Roeder R.G. Biol. 1996; Scholar), and that it the of promoter binding and of K. Cell. 1996; Full Text Full Text PDF PubMed Scopus Google Scholar). we have analyzed the mechanism of transcriptional regulation of the insulin gene by We identified two novel CREM and and six known isoforms in pancreatic islets and the pancreatic β-cell line. We demonstrated that these isoforms regulated insulin gene transcription by binding to CREs and that CREM and its phosphorylation involved in insulin promoter We the direct interaction CREM isoforms and the TFIID hTAFII130 or TBP, a mechanism by which CREM isoforms might insulin gene transcription. rat in was the with was used as N. H. J. 75: PubMed Scopus Google with a to for the of binding of the to or the to to and to was by of rat was used as a for was CREMτ in each islets isolated from the of by M. T. T. S. PubMed Scopus Google Scholar), by was isolated from isolated islets by the W.J. PubMed Scopus Google and used as a for pancreatic and the pancreatic β-cell RINm5F B. PubMed Scopus Google by the rat from that could the transcription was as a The sequences of the for the of CREM and ICER as T.E. Habener J.F. Res. 1992; PubMed Scopus (27) Google J.H. Foulkes N.S. Molina C.A. Simonneaux V. Pévet P. Sassone-and Corsi P. Nature. 1993; 365: 314-320Crossref PubMed Scopus (356) Google Scholar), The of the was to the of each The sequences of the for CREM as of the as and to B. was the of and the sequences by the The sequences of the CREM CREMα, ICER I, and ICER Iγ, the of CREMτα to was by an in and the the The sequences of the activation domains as and to of and of CREMτα to in to the of the domain in The was by the for the domain from of the promoter in activity of the activation the function of the domain and Q2 domain Q2 domain and the of and Q2 domain to the of the domain in the phosphorylation is was mutated by The the of the and to the phosphorylation and to the glutamine-rich Q2. The in a which is the for the with the and the the than the for the gene by R. A. pancreatic β-cell P. 1986; PubMed Scopus Google in with and in an of in by the the to The with a of of the which human insulin gene promoter to N. Maekawa T. Sudo T. Ishii S. Seino Y. Imura H. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 1045-1049Crossref PubMed Scopus (88) Google or which of each CREM or CREM in which domain and of F. Van Der A. PubMed Scopus Google Scholar). and human and from human and the The of the was T. Fujisawa J. Yoshida M. EMBO J. PubMed Scopus Google Scholar), and the of the was of CREB, and T. Sakura H. Kanei-Ishii C. Sudo T. Yoshimura T. Fujisawa J. Yoshida M. Ishii S. EMBO J. 1989; 8: 2023-2028Crossref PubMed Scopus (292) Google promoter activity by with various concentration of the and for and for than to binding activity of and ICER to the human insulin CREs with of the of and ICER and E. of protein with the to the the of the by by and for and the binding activity was analyzed an are to to to and to of of the and used in that to CRE insulin CRE sequences are in The consensus CRE is but the sequences of and the the and it the in a The CRE sequences are in The consensus CRE is but the sequences of and the the and it the and in was in of by and to with in and with to for with for of these was by for in was as in the The human was a of N. TBP was from human The of the was N. Pugh B.F. Tjian R. Science. PubMed Scopus Google Scholar), and the of the was and hTAFII130 the of and the of and hTAFII130 in a in in the of The by a and as Y. M. E. T. S. R. Yamamoto T. M. Roeder R.G. J. Biol. 1997; Full Text Full Text PDF PubMed Scopus Google Scholar, J.A. S. Tjian R. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: PubMed Scopus Google Scholar). protein or in or hTAFII130 was with GST, in binding for the with of binding The by and by or by an as CREM is in pancreatic studies of rat pancreatic the of the CREM a was used as a are not In with previous (9Sassone-Corsi P. EMBO J. 1994; 13: 4717-4728Crossref PubMed Scopus (68) Google Scholar, 18Delmas V. Van Der Hoon F. Mellström B. Jégou B. Sassone-Corsi P. Mol. Endocrinol. 1993; 7: 1502-1514Crossref PubMed Scopus (187) Google Scholar, F. L. Foulkes N.S. D. M. K. A. M. Sassone-Corsi P. Nature. 1996; PubMed Scopus Google Scholar), the of CREM protein was to be to of when the to be the We used two rat pancreatic one for the and the other for the with the The CREM both in the and of the pancreatic in the pancreatic which isoforms in the pancreatic and the pancreatic β-cell we with rat pancreatic and the RINm5F β-cell We isolated CREM of which two are novel isoforms and the other six are known of two novel isoforms (CREMΔQ1 and CREMΔQ2), repressors (CREMα, ICER I, ICER Iγ, CREM-17X, and and an activator is in B. and lack one of the glutamine-rich domains (Q1 and and contain an binding domain These are to of and (9Sassone-Corsi P. EMBO J. 1994; 13: 4717-4728Crossref PubMed Scopus (68) Google Scholar, 12Laoide B.M. Foulkes N.S. Schlotter F. Sassone-Corsi P. EMBO J. 1993; 12: 1179-1191Crossref PubMed Scopus (188) Google Scholar), which also lack one of the glutamine-rich domains (Q1 and and contain an we have that and functioned as repressors A. Yamada Y. Someya Y. Kubota A. Yasuda K. Ihara Y. Kagimoto S. Kuroe A. Tsuda K. Seino Y. Biochem. Biophys. Res. Commun. 1998; 253: 712-718Crossref PubMed Scopus (32) Google Scholar). the function of the other CREM isoforms in insulin gene the isolated isoforms the and with the the human insulin promoter and insulin promoter and On the other hand, CREMα, ICER I, and ICER the activity to and of We the of the repressors, and ICER, to repress not only basal transcriptional activity but also activator-induced transcriptional CREM activator transcriptional and ICER this to of and and ICER also and transcriptional activity to of the function of the phosphorylation and glutamine-rich domains in the transcriptional activity of CREM, the domain the Q2 domain and the of and Q2 domains in to the of the domain in the phosphorylation is involved in CREM we mutated the to in the domain which is the for the phosphorylation (9Sassone-Corsi P. EMBO J. 1994; 13: 4717-4728Crossref PubMed Scopus (68) Google Scholar, R. J. Sassone-Corsi P. Mol. Endocrinol. 1993; Scholar, J. Sassone-Corsi P. EMBO J. 1993; 12: PubMed Scopus Google Scholar). of is with the and the activity increased and in the of Q2 domain or domains, respectively, but in the of Q2 domain The to domain has been to insulin P. R. PubMed Scopus Google Scholar, P. PubMed Scopus Google Scholar). We CREM and are involved in insulin promoter the of the phosphorylation in CREM, in the domain of CREMτα was mutated by CREB, and with and the with various of insulin promoter activity was increased by glucose in a in the of CREMτα However, in the of and CREB, in insulin promoter activity was observed by glucose stimulation, that CREM but not is involved in insulin promoter and that the is We glucose domain the and domains with and the with various of The activity of the domain increased in a but the activity not this is that glucose might the activation of CREM by We have four sequences in the human insulin gene that are different from each other and from the consensus CRE motif (4Inagaki N. Maekawa T. Sudo T. Ishii S. Seino Y. Imura H. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 1045-1049Crossref PubMed Scopus (88) Google Scholar). CREM isoforms bind to these sequences of the human insulin we produced and ICER and human insulin gene CRE The binding activity was analyzed an The binding of produced and ICER to each CRE to as the of from and We ICER with CREMτα or for in ICER with CREMτα in a to bind to results the of ICER and CREMτα of the insulin gene to ICER than On the other hand, the binding activity of ICER to was the as that of CREM might interact directly with the TFIID components TBP and hTAFII130 we in interaction The CREM isoforms and and in a in in the of isoforms with or protein was by and by in to and However, not bind only to ICER the interaction TBP and the CREM we from and the interaction of TBP and produced CREM to with a TBP to CREMτα and and but not bind to ICER We the interaction of hTAFII130 and the CREM hTAFII130 only to CREMτα and and The binding of hTAFII130 to was hTAFII130 not bind to either or ICER and The insulin gene has elements, and its transcription is to be regulated by the interaction of various and factors (1Sander M. German M.S. J. Mol. Med. 1997; 75: 327-340Crossref PubMed Scopus (285) Google Scholar). In addition, factors such as transcription factors are involved in transcription by with each other and the N.S. Borjigin J. Snyder S.H. Sassone-Corsi P. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 14140-14145Crossref PubMed Scopus (136) Google Scholar, 20Fitzgerald L.R. Li Z. Machida C.A. Fishman P.H. Duman R.S. J. Neurochem. 1996; 67: 490-497Crossref PubMed Scopus (27) Google Scholar, 21Monaco L. Foulkes N.S. Sassone-Corsi P. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 10673-10677Crossref PubMed Scopus (102) Google Scholar, 22Deadouets C. Matesic G. Molina C.A. Foulkes N.S. Sassone-Corsi P. Brechot C. Sobczak-Thepot J. Mol. Cell. Biol. 1995; 15: 3301-3309Crossref PubMed Scopus (218) Google Scholar, C-M. Roeder R.G. Science. 1995; 267: 531-536Crossref PubMed Scopus (352) Google Scholar, 24Tanese N. Pugh B.F. Tjian R. Genes Dev. 1991; 5: 2212-2224Crossref PubMed Scopus (244) Google Scholar, 25Tanese N. Saluja D. Vassallo M. Chen J-L. Admon A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13611-13616Crossref PubMed Scopus (118) Google Scholar). In the study, the mechanism of transcriptional regulation of the human insulin gene by CREM through CREs was we demonstrated by and by that CREM is in pancreatic the levels of and We isolated two novel CREM isoforms (CREMΔQ1 and and the six known isoforms (CREMτα, CREMα, ICER I, ICER Iγ, CREM-17X, and CREM-17) from rat pancreatic islets and the RINm5F pancreatic β-cell line. has been that all isoforms contain one of the domains, Q1 and Q2 (12Laoide B.M. Foulkes N.S. Schlotter F. Sassone-Corsi P. EMBO J. 1993; 12: 1179-1191Crossref PubMed Scopus (188) Google Scholar). that and also contain one domain and that are to that of the and activators (9Sassone-Corsi P. EMBO J. 1994; 13: 4717-4728Crossref PubMed Scopus (68) Google Scholar), and to function as activators. and in the to function as activators of insulin In addition, we that CREMτα as an efficient CREMα, ICER I, and ICER as repressors of insulin In the previous study, we have that and function as efficient repressors of insulin promoter A. Yamada Y. Someya Y. Kubota A. Yasuda K. Ihara Y. Kagimoto S. Kuroe A. Tsuda K. Seino Y. Biochem. Biophys. Res. Commun. 1998; 253: 712-718Crossref PubMed Scopus (32) Google Scholar). We also by that CREM are both in the and of pancreatic as the of G. Habener J.F. Mol. Endocrinol. 1991; 5: PubMed Scopus (62) Google is also in the of CREM isoforms B. Kempf R. Telgmann R. Mol. Endocrinol. 1997; 11: 97-113Crossref PubMed Scopus (62) Google Scholar). previous a of the the in the when in G. Habener J.F. Mol. Endocrinol. 1991; 5: PubMed Scopus (62) Google Scholar), that the CREM isoforms may in the and that the isoforms that lack the in the In a previous study, we have identified four sequences to in the human insulin gene (4Inagaki N. Maekawa T. Sudo T. Ishii S. Seino Y. Imura H. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 1045-1049Crossref PubMed Scopus (88) Google Scholar). These CREs are the the consensus CRE is but the sequences of and the the and it the the for the binding of CREM to these sequences was In the study, we demonstrated that the CREM CREMτα and bind to each This that CREM activator insulin promoter activity by binding directly to CREs. In addition, we demonstrated that the CREM ICER Iγ, also to each CRE and that the binding activity of repressor to insulin CRE is higher than that of that CREM repressor insulin promoter activity by competitively the binding of the activators to CREs. We the of the of the family to the insulin We in the that could not insulin with it is that the of the family, CREM a role in insulin promoter activity in pancreatic We the of CREM and to insulin promoter in the of is well known that glucose insulin promoter but the mechanism is P. R. PubMed Scopus Google Scholar, P. PubMed Scopus Google Scholar). we for the first that CREM but not is involved in insulin promoter is that CREM and a in the to insulin promoter not only in the but also in the of This that CREM a role different from that of in promoting insulin promoter in pancreatic Since of CREMτ has been to be the for phosphorylation by protein protein and and such phosphorylation CREM activity (9Sassone-Corsi P. EMBO J. 1994; 13: 4717-4728Crossref PubMed Scopus (68) Google Scholar, R. J. Sassone-Corsi P. Mol. Endocrinol. 1993; Scholar, J. Sassone-Corsi P. EMBO J. 1993; 12: PubMed Scopus Google Scholar), we mutated to the of this in insulin promoter We by that the phosphorylation of CREM is for insulin promoter it is that glucose might an to the phosphorylation of which function to insulin promoter has been that TFIID interact with each other Y. M. E. T. S. R. Yamamoto T. M. Roeder R.G. J. Biol. 1997; Full Text Full Text PDF PubMed Scopus Google Scholar, K. T. R. M. M. Y. Roeder R.G. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: PubMed Scopus Google and that can function as by direct interaction with site-specific transcription factors N. Pugh B.F. Tjian R. Genes Dev. 1991; 5: 2212-2224Crossref PubMed Scopus (244) Google Scholar, R. Sci. 1995; PubMed Scopus Google J.L. K. Tjian R. Cell. 1994; Full Text PDF PubMed Scopus Google Scholar, T. G. Chen J-L. Tjian R. Cell. 1993; Full Text PDF PubMed Scopus Google therefore, it was of to CREM directly with In this study, we that CREM activators interact with both TBP and hTAFII130 and its and have been to with or directly through the glutamine-rich domain of or N. Pugh B.F. Tjian R. Genes Dev. 1991; 5: 2212-2224Crossref PubMed Scopus (244) Google Scholar, 25Tanese N. Saluja D. Vassallo M. Chen J-L. Admon A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13611-13616Crossref PubMed Scopus (118) Google Scholar, J.L. K. Tjian R. Cell. 1994; Full Text PDF PubMed Scopus Google Scholar, T. G. Chen J-L. Tjian R. Cell. 1993; Full Text PDF PubMed Scopus Google Scholar, K. G. Montminy M. Proc. Natl. Acad. Sci. U. S. A. 1994; PubMed Scopus Google Scholar, T. C. Montminy M. Genes Dev. 1997; 11: PubMed Scopus Google Scholar). that all CREM activators one or two glutamine-rich domains (9Sassone-Corsi P. EMBO J. 1994; 13: 4717-4728Crossref PubMed Scopus (68) Google Scholar), Q1 and Q2 to be for CREM to interact with hTAFII130 and to function as an the interaction of Q1 and Q2 with hTAFII130 may CREM to function as an results that interaction and interaction might be mediated by the glutamine-rich domains and by the phosphorylation domain of CREM, Since it has been that the interaction of TFIID with site-specific transcriptional factors the of binding of TFIID to the promoter or K. Cell. 1996; Full Text Full Text PDF PubMed Scopus Google Scholar), it is that an interaction of CREM with TBP and hTAFII130 may and stabilize the binding of TFIID to the insulin In other CREM activators could the levels of insulin gene transcription by both binding directly to the CREs and recruiting TFIID to the insulin promoter via its interaction with and hTAFII130 On the other hand, CREM repressors that interact with both hTAFII130 and TBP, that CREM repressor might fail to stabilize it is that CREM repressor might insulin promoter activity by competitively the binding of the activators to CREs by to stabilize the basal transcriptional Since the only interactions of CREM with and hTAFII130 in this study, studies of the interactions CREM and the components of the TFIID such as K. S. R. Y. M. Roeder R.G. Nature. 1993; PubMed Scopus Google Scholar), N. Saluja D. Vassallo M. Chen J-L. Admon A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13611-13616Crossref PubMed Scopus (118) Google Scholar, Y. M. E. T. S. R. Yamamoto T. M. Roeder R.G. J. Biol. 1997; Full Text Full Text PDF PubMed Scopus Google Scholar), and K. T. R. M. M. Y. Roeder R.G. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: PubMed Scopus Google be to the of the the insulin gene transcription in In we in this that CREM activators and repressors are in pancreatic and that they play important roles in the insulin gene transcription in response to important from is that the activation of insulin gene transcription by CREM activator is mediated by not only direct binding to the CREs but also by recruiting TFIID to the insulin promoter via its interaction with and Furthermore, the CREM repressor competitively interrupts the binding of the activators to CREs and does not interact with both TBP and hTAFII130; it might fail to stabilize the basal transcriptional machinery and repress transactivation. We N. for human and R. A. for We S. Ishii for of the and S. of and of for