Identification of the Functional Domain in the Transcription Factor RTEF-1 That Mediates α1-Adrenergic Signaling in Hypertrophied Cardiac Myocytes
2000; Elsevier BV; Volume: 275; Issue: 23 Linguagem: Inglês
10.1074/jbc.m001970200
ISSN1083-351X
AutoresTakahisa Ueyama, Chongxue Zhu, Yunuen M. Valenzuela, Joseph Suzow, Alexandre F.R. Stewart,
Tópico(s)Viral Infectious Diseases and Gene Expression in Insects
ResumoCardiac myocytes respond to α1-adrenergic receptor stimulation by a progressive hypertrophy accompanied by the activation of many fetal genes, including skeletal muscle α-actin. The skeletal muscle α-actin gene is activated by signaling through an MCAT element, the binding site of the transcription enhancer factor-1 (TEF-1) family of transcription factors. Previously, we showed that overexpression of the TEF-1-related factor (RTEF-1) increased the α1-adrenergic response of the skeletal muscle α-actin promoter, whereas TEF-1 overexpression did not. Here, we identified the functional domains and specific sequences in RTEF-1 that mediate the α1-adrenergic response. Chimeric TEF-1 and RTEF-1 expression constructs localized the region responsible for the α1-adrenergic response to the carboxyl-terminal domain of RTEF-1. Site-directed mutagenesis was used to inactivate eight serine residues of RTEF-1, not present in TEF-1, that are putative targets of α1-adrenergic-dependent kinases. Mutation of a single serine residue, Ser-322, reduced the α1-adrenergic activation of RTEF-1 by 70% without affecting protein stability, suggesting that phosphorylation at this serine residue accounts for most of the α1-adrenergic response. Thus, these results demonstrate that RTEF-1 is a direct target of α1-adrenergic signaling in hypertrophied cardiac myocytes. Cardiac myocytes respond to α1-adrenergic receptor stimulation by a progressive hypertrophy accompanied by the activation of many fetal genes, including skeletal muscle α-actin. The skeletal muscle α-actin gene is activated by signaling through an MCAT element, the binding site of the transcription enhancer factor-1 (TEF-1) family of transcription factors. Previously, we showed that overexpression of the TEF-1-related factor (RTEF-1) increased the α1-adrenergic response of the skeletal muscle α-actin promoter, whereas TEF-1 overexpression did not. Here, we identified the functional domains and specific sequences in RTEF-1 that mediate the α1-adrenergic response. Chimeric TEF-1 and RTEF-1 expression constructs localized the region responsible for the α1-adrenergic response to the carboxyl-terminal domain of RTEF-1. Site-directed mutagenesis was used to inactivate eight serine residues of RTEF-1, not present in TEF-1, that are putative targets of α1-adrenergic-dependent kinases. Mutation of a single serine residue, Ser-322, reduced the α1-adrenergic activation of RTEF-1 by 70% without affecting protein stability, suggesting that phosphorylation at this serine residue accounts for most of the α1-adrenergic response. Thus, these results demonstrate that RTEF-1 is a direct target of α1-adrenergic signaling in hypertrophied cardiac myocytes. skeletal muscle α-actin chloramphenicol acetyltransferase myocyte-specific CAT transcription enhancer factor-1 TEF-1-related factor cAMP-dependent protein kinase protein kinase C mitogen-activated protein kinase cytomegalovirus Cardiac myocytes respond to α1-adrenergic receptor stimulation by a progressive hypertrophy (1.Simpson P. J. Clin. Invest. 1983; 72: 732-738Crossref PubMed Scopus (546) Google Scholar), accompanied by a characteristic reactivation of many fetal genes, including β-myosin heavy chain (2.Waspe L.E. Ordahl C.P. Simpson P.C. J. Clin. Invest. 1990; 85: 1206-1214Crossref PubMed Scopus (155) Google Scholar), skeletal muscle α-actin (SKA)1 (3.Long C.S. Ordahl C.P. Simpson P.C. J. Clin. Invest. 1989; 83: 1078-1082Crossref PubMed Scopus (105) Google Scholar, 4.Karns L.R. Kariya K. Simpson P.C. J. Biol. Chem. 1995; 270: 410-417Abstract Full Text Full Text PDF PubMed Scopus (162) Google Scholar), and brain natriuretic factor (5.Thuerauf D.J. Glembotski C.C. J. Biol. Chem. 1997; 272: 7464-7472Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). The α1-adrenergic stimulation of these promoters requires an MCAT element (4.Karns L.R. Kariya K. Simpson P.C. J. Biol. Chem. 1995; 270: 410-417Abstract Full Text Full Text PDF PubMed Scopus (162) Google Scholar, 5.Thuerauf D.J. Glembotski C.C. J. Biol. Chem. 1997; 272: 7464-7472Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar), with the sequence CATN(T/C)(T/C) (6.Farrance I.K. Mar J.H. Ordahl C.P. J. Biol. Chem. 1992; 267: 17234-17240Abstract Full Text PDF PubMed Google Scholar). Transcription factors of the transcription enhancer factor-1 (TEF-1) multigene family bind to MCAT elements in the promoters of many genes expressed in cardiac and skeletal muscle cells (7.Larkin S.B. Farrance I.K.G. Ordahl C.P. Mol. Cell. Biol. 1996; 16: 3742-3755Crossref PubMed Scopus (72) Google Scholar). Thus, a role for TEF-1-related transcription factors in mediating the α1-adrenergic response has been proposed (4.Karns L.R. Kariya K. Simpson P.C. J. Biol. Chem. 1995; 270: 410-417Abstract Full Text Full Text PDF PubMed Scopus (162) Google Scholar, 8.Kariya K. Farrance I.K. Simpson P.C. J. Biol. Chem. 1993; 268: 26658-26662Abstract Full Text PDF PubMed Google Scholar, 9.Kariya K. Karns L.R. Simpson P.C. J. Biol. Chem. 1994; 269: 3775-3782Abstract Full Text PDF PubMed Google Scholar). Previously, we showed that the TEF-1-related factor RTEF-1 could potentiate the α1-adrenergic stimulation of the β-myosin heavy chain and SKA promoters, when overexpressed in cardiac myocytes. In contrast, TEF-1 did not affect their response to α1-adrenergic stimulation. Thus, we proposed a role for RTEF-1 in mediating the α1-adrenergic reactivation of fetal genes in cardiac myocytes (10.Stewart A.F.R. Suzow J. Kubota T. Ueyama T. Chen H.H. Circ Res. 1998; 83: 43-49Crossref PubMed Scopus (65) Google Scholar).The different effects of TEF-1 and RTEF-1 overexpression on the α1-adrenergic response of the SKA promoter must reflect differences in how TEF-1 and RTEF-1, or their associated co-factors, are modified by α1-adrenergic signaling. TEF-1 and RTEF-1 are identical in their DNA binding domains, are highly conserved in their carboxyl-terminal activation domains, but diverge in sequences flanking the DNA binding domain (see Fig. 1). To what extent these divergent sequences confer functional differences was not known.In the present study, we took advantage of the different response of a minimal α1-adrenergic-responsive SKA promoter to TEF-1 and RTEF-1 overexpression to examine the functional differences between TEF-1 and RTEF-1. By swapping functional domains between TEF-1 and RTEF-1 in chimeric expression constructs, the carboxyl-terminal domain of RTEF-1 was found to contain the α1-adrenergic-responsive sequences. Site-directed mutagenesis further mapped the serine residues targeted by α1-adrenergic signaling. Our results demonstrate that RTEF-1 is an authentic target of α1-adrenergic signaling, principally via a MAPK site at Ser-322 in the carboxyl-terminal domain.RESULTSSince TEF-1 and RTEF-1 are functionally different, we hypothesized that functional differences would reside in sequences that are divergent between TEF-1 and RTEF-1, as diagrammed in Fig.1. The most divergent sequence of RTEF-1 is located in the amino-terminal domain that flanks the DNA binding domain. A synthetic peptide from Ile-8 to Gln-29 was used to develop an antibody specific to human RTEF-1. As shown in Fig.2 A, antigenic differences may account for the specificity of the anti-human RTEF-1 antibody. Since the rat RTEF-1 sequence remains unreported, the mouse TEFR-1 amino-terminal sequence (14.Yockey C.E. Smith G. Izumo S. Shimizu N. J. Biol. Chem. 1996; 271: 3727-3736Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar) is shown for comparison. Note that the mouse sequence has an aspartic acid residue at position 17 (Fig.2 A, arrow) in addition to Ser-23/Ile-24 instead of Thr-23/Ala-24. These changes would alter the antigenicity of the peptide and may account for the absence of signal in neonatal rat myocyte nuclear extract, even though RTEF-1 RNA could be detected by Northern blot analysis (data not shown). In Fig. 2 B, the anti-RTEF-1 peptide-specific antibody detected human RTEF-1 only in nuclear extracts from transfected neonatal rat cardiac myocytes. Moreover, there was a 3-fold increase in RTEF-1 signal when the nuclear extract was pretreated with calf intestinal alkaline phosphatase, demonstrating that the amino-terminal domain of RTEF-1 is phosphorylated in cardiac myocytes under basal conditions (in the absence of α1-adrenergic receptor stimulation).Figure 2Phosphorylation of the amino-terminal domain of RTEF-1 revealed with a peptide-specific antibody. A, potential phosphorylation sequences in the human RTEF-1 amino-terminal peptide used for antibody production. CKII, casein kinase II (33.Russo G. Vandenberg M., Yu, I. Bae Y. Franza Jr., B.R. Marshak D. J. Biol. Chem. 1992; 267: 20317-20325Abstract Full Text PDF PubMed Google Scholar); MAPK, mitogen-activated protein kinase (25.Davis R. J. Biol. Chem. 1993; 268: 14553-14556Abstract Full Text PDF PubMed Google Scholar). The mouse TEFR-1 sequence (14.Yockey C.E. Smith G. Izumo S. Shimizu N. J. Biol. Chem. 1996; 271: 3727-3736Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar) has an aspartic acid residue at position 17 (arrow). B, Western blot of nuclear extracts from neonatal rat cardiac myocytes, transfected either with the empty CMV expression vector or the human RTEF-1 expression vector. Extracts were either treated with calf intestinal alkaline phosphatase (PPase) (+) or not (−) prior to loading on the SDS-polyacrylamide gel. 100 μg of nuclear extract were loaded per lane. The position of protein size markers is indicated.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Polymerase chain reaction-based mutagenesis was used to introduce convenient restriction enzyme sites in TEF-1 and RTEF-1 that allowed different domains of RTEF-1 and TEF-1 to be swapped in the expression constructs and cotransfected with the SKA reporter. Since the amino-terminal domain of RTEF-1 appeared to be phosphorylated even under basal conditions, it was exchanged with that of TEF-1 to produce two new chimeric proteins, RT and TR (Fig.3). Surprisingly, the amino-terminal domain of RTEF-1 contributed little to the transactivation and α1-adrenergic response. TR retained all the transactivation function and most of the α1-adrenergic-responsive sequences of RTEF-1. An additional chimeric protein was made, TLR, that contains the amino-terminal, DNA-binding, and proline-rich domains of TEF-1 and the carboxyl-terminal 200 amino acids of RTEF-1. This construct lost some of the transactivation function of RTEF-1, but retained most of the α1-adrenergic-responsive sequences. These results suggested that the transactivation function of RTEF-1 is located carboxyl-terminal to the DNA-binding domain and that the α1-adrenergic-responsive sequences reside in the most conserved region between TEF-1 and RTEF-1 in the carboxyl-terminal 200 amino acid residues.Figure 3Domain swap between TEF-1 and RTEF-1 maps the α1-adrenergic-responsive sequences of RTEF-1. Chimeric expression constructs were cotransfected with the SKA reporter and tested for their transactivation function under basal conditions (white bars) and in the presence of 100 μmphenylephrine (black bars). The amino-terminal domain of RTEF-1 (in the chimeric RT construct) contributes little to the α1-adrenergic response. Most of the transactivation and the α1-adrenergic-responsive sequences of RTEF-1 lie in the carboxyl-terminal 200 residues (see TR andTLR). The activity of the SKA promoter is expressed as a -fold of the activity when cotransfected with the empty CMV expression vector in the presence of vehicle (100 μm ascorbic acid) on the x axis (dashed line) (mean ± S.E., n = 6 experiments). The effect of each chimeric construct was compared with the ability of wild type RTEF-1 to potentiate the α1-adrenergic response of the SKA promoter (set at 100% potentiation).View Large Image Figure ViewerDownload Hi-res image Download (PPT)To identify the residues of RTEF-1 that are important in mediating the α1-adrenergic response, an exhaustive site-directed mutagenesis of all serine residues that are putative targets of phosphorylation was conducted (see TableI and Fig.4). The effect of these mutations was tested on the ability of RTEF-1 to augment the α1-adrenergic response of the SKA promoter. The activity of the SKA promoter in the presence of 100 μmphenylephrine was set at 0% potentiation when cotransfected with the empty CMV expression plasmid, whereas the activity was set at 100% potentiation of the α1-adrenergic response when cotransfected with the wild type RTEF-1 expression construct (Fig. 4). Thus, mutations that reduced the level of potentiation were considered to be targets of α1-adrenergic receptor dependent phosphorylation.Figure 4Site-directed mutagenesis of RTEF-1 identifies the carboxyl-terminal Ser-322 residue as the major target of α1-adrenergic signaling.Serines at eight positions in RTEF-1 that are not found in TEF-1 were mutated to alanine residues that cannot be phosphorylated. Two double mutants (254/322 and 290/358) and a construct with all four carboxyl-terminal serines mutated (C4) were also made. The activity of the SKA promoter is expressed as a -fold of the activity when cotransfected with the empty CMV expression vector in the presence of vehicle (100 μm ascorbic acid) on the x axis (dashed line) (Mean ± S.E.,n = 7 experiments). The effect of each mutant construct was compared with the ability of wild type RTEF-1 to potentiate the α1-adrenergic response of the SKA promoter (set at 100% potentiation).View Large Image Figure ViewerDownload Hi-res image Download (PPT)None of the four serine residues in the amino-terminal half of RTEF-1, when mutated to an alanine residue, significantly lowered the level of potentiation. These results agree with the chimeric study (Fig. 3) that confined most of the α1-adrenergic response to the carboxyl-terminal domain of RTEF-1. Mutation at each of four carboxyl-terminal serine residues tended to lower the α1-adrenergic potentiating effect of RTEF-1, with a significant effect seen for constructs carrying the Ser-322 → Ala mutation (p < 0.05). The Ser-322 → Ala mutation reduced the level of potentiation by 70% relative to the wild type RTEF-1 protein (Fig. 4).The Ser-322 → Ala mutation did not affect protein stability, as shown by a gel mobility shift assay using nuclear extracts from cardiac myocytes transfected with wild type and Ser-322 → Ala RTEF-1 constructs (Fig. 5). Increasing amounts of an antibody specific to TEF-1 was used to supershift, and effectively remove, endogenous TEF-1 from the MCAT shift (Fig.5 A), revealing the overexpressed RTEF-1 proteins. Approximately 85% of the MCAT binding complex was supershifted by the TEF-1 antibody, suggesting that the residual 15% is composed of other TEF-1-related factors (like RTEF-1, DTEF-1, or ETF) present in neonatal rat cardiac myocyte nuclear extracts. Phosphorimage analysis showed that the intensity of the Ser-322 → Ala RTEF-1 band was 93% that of the wild type RTEF-1 (Fig. 5 B, n = 4 experiments), demonstrating that the loss of function was not due to loss of protein (assuming DNA binding activity is not reduced by the mutation). The construct with all four carboxyl-terminal serine residues mutated (C4) produced a gel shift complex that was 76% as strong as that of wild type RTEF-1, suggesting that loss of protein can only explain part of the loss of function for this construct. Thus, these results suggest that most of the α1-adrenergic-responsive sequences of RTEF-1 lie in the carboxyl-terminal 200 residues as shown by the domain swap experiments and that Ser-322 is the principal target of α1-adrenergic signaling.Figure 5Gel mobility shift assay reveals equivalent levels of wild type RTEF-1 and Ser-322 mutant RTEF-1 in nuclear extracts of transfected cardiac myocytes. A, an antibody specific to TEF-1 (residues 86–199) was used in increasing amounts in a gel mobility shift assay with a radiolabeled oligonucleotide carrying the MCAT element of the SKA promoter and 10 μg of nuclear protein extracts from cardiac myocytes. At the optimal dilution (0.2), the antibody supershifted 85% of the MCAT binding factor (MCBF). B, gel shift assay with nuclear extracts from cardiac myocytes transfected with the empty CMV (C), the wild type RTEF-1 (WT), the mutant Ser-322 RTEF-1, or the mutant C4 RTEF-1 expression vectors. Normalizing to the signal for the complex supershifted with the TEF-1-specific antibody, the mutant Ser-322 and C4 RTEF-1 proteins were expressed at 93% and 76% of the wild type RTEF-1 protein, respectively.View Large Image Figure ViewerDownload Hi-res image Download (PPT)DISCUSSIONCatecholamines are elevated in patients with heart failure and are associated with a worsened prognosis (for review, see Ref. 15.Esler M. Kaye D. Lambert G. Esler D. Jennings G. Am. J. Cardiol. 1997; 80: 7L-14LAbstract Full Text Full Text PDF PubMed Scopus (196) Google Scholar). The myocardium of heart failure patients shows a down-regulation (16.Bristow M.R. Minobe W.A. Raynolds M.V. Port J.D. Rasmussen R. Ray P.E. Feldman A.M. J. Clin. Invest. 1993; 92: 2737-2745Crossref PubMed Scopus (212) Google Scholar) and uncoupling (17.Ungerer M. Bohm M. Elce J.S. Erdmann E. Lohse M.J. Circulation. 1993; 87: 454-463Crossref PubMed Scopus (757) Google Scholar) of β-adrenergic receptors, but no change in the levels of α1-adrenergic receptors (18.Bristow M.R. Minove W. Rasmussen R. Hershberger R.E. Hoffman B.B. J. Pharmacol. Exp. Ther. 1988; 247: 1039-1045PubMed Google Scholar). Thus, the α1-adrenergic signaling pathway in the myocardium of patients with heart failure becomes more prevalent. Rat cardiac myocytes respond to α1-adrenergic receptor stimulation by a progressive hypertrophy via the α1A-adrenergic receptor subtype (19.Knowlton K.U. Michel M.C. Itani M. Shubeita H.E. Ishihara K. Brown J.H. Chien K.R. J. Biol. Chem. 1993; 268: 15374-15380Abstract Full Text PDF PubMed Google Scholar). This α1-adrenergic receptor is induced in cultured cardiac myocytes by chronic exposure to catecholamines and in the myocardium of rats with pressure overload caused by aortic banding (20.Rokosh D.G. Stewart A.F.R. Chang K.C. Bailey B.A. Karliner J.S. Camacho S.A. Long C.S. Simpson P.C. J. Biol. Chem. 1996; 271: 5839-5843Abstract Full Text Full Text PDF PubMed Scopus (142) Google Scholar). Hypertrophy is accompanied by the activation of fetal genes, including β-myosin heavy chain and SKA. The switch to a fetal phenotype was once thought to be relevant only to the rodent model of hypertrophy and heart failure, since healthy human hearts already express considerable levels of β-myosin heavy chain. However, two recent reports have demonstrated the same transition to higher levels of β-myosin heavy chain expression at the expense of α-myosin heavy chain in the myocardium of heart failure patients (21.Lowes B.D. Minobe W. Abraham W.T. Rizeq M.N. Bohlmeyer T.J. Quaife R.A. Roden R.L. Dutcher D.L. Robertson A.D. Voelkel N.F. Badesch D.B. Groves B.M. Gilbert E.M. Bristow M.R. J. Clin. Invest. 1997; 100: 2315-2324Crossref PubMed Scopus (418) Google Scholar, 22.Nakao K. Minobe W. Roden R. Bristow M.R. Leinwand L.A. J. Clin. Invest. 1997; 100: 2362-2370Crossref PubMed Scopus (329) Google Scholar). Thus, similar signaling and transcriptional mechanisms are operating in failing rat and human hearts.We have identified the TEF-1-related transcription factor RTEF-1 as abone fide target of α1-adrenergic signaling in cardiac myocytes. Although we anticipated that the most divergent sequences between TEF-1 and RTEF-1 would be most likely to confer the functional difference in the α1-adrenergic response, instead we found that this difference lies in the carboxyl terminus of RTEF-1, a highly conserved region among member of the TEF-1 family. Of the four serine residues in the carboxyl terminus of RTEF-1, two are potential PKC targets (Ser-290 and Ser-358) and two are potential MAPK sites (Ser-254 and Ser-322). PKC and MAPK have been implicated in the α1-adrenergic induction of fetal genes and hypertrophy in cardiac myocytes (23.Kariya K. Karns L.R. Simpson P.C. J. Biol. Chem. 1991; 266: 10023-10026Abstract Full Text PDF PubMed Google Scholar, 24.Gillespie-Brown J. Fuller S.J. Bogoyevitch M.A. Cowley S. Sugden P.H. J. Biol. Chem. 1995; 270: 28092-28096Abstract Full Text Full Text PDF PubMed Scopus (149) Google Scholar). Site-directed mutagenesis identified Ser-322 as the major target of α1-adrenergic signaling. However, the effects of mutating the PKC sites either alone or together, or mutating Ser-254, did not reach statistical significance in more than seven experiments. Whereas these sequences may also participate, Ser-322 appears to be most important in mediating the α1-adrenergic response. Ser-322 is located in the STY domain of the carboxyl terminus, a region rich in serine, threonine, and tyrosine residues (11.Xiao J.H. Davidson I. Matthes H. Garnier J.M. Chambon P. Cell. 1991; 65: 551-568Abstract Full Text PDF PubMed Scopus (301) Google Scholar). Ser-322 conforms to the consensus of a MAPK site (25.Davis R. J. Biol. Chem. 1993; 268: 14553-14556Abstract Full Text PDF PubMed Google Scholar), and although we have not proven by phosphopeptide mapping that Ser-322 is phosphorylated in vivo, it seems likely that phosphorylation by a MAPK activates RTEF-1 in cardiac myocytes. This MAPK site is conserved in the chicken (26.Stewart A.F.R. Larkin S.B. Farrance I.K.G. Mar J.H. Hall D.E. Ordahl C.P. J. Biol. Chem. 1994; 269: 3147-3150Abstract Full Text PDF PubMed Google Scholar) and mouse RTEF-1 (14.Yockey C.E. Smith G. Izumo S. Shimizu N. J. Biol. Chem. 1996; 271: 3727-3736Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar) sequences but is absent from other TEF-1 family members (Fig.6). The absence of this MAPK site in TEF-1 may reflect a difference in the regulation of TEF-1 and RTEF-1 by phosphorylation. Phosphorylation might alter RTEF-1 protein conformation or an interaction of RTEF-1 with a specific cofactor.Figure 6Aligned sequences of TEF-1 family members in the STY region. The sequences are from human TEF-1 (hTEF-1; Ref. 11.Xiao J.H. Davidson I. Matthes H. Garnier J.M. Chambon P. Cell. 1991; 65: 551-568Abstract Full Text PDF PubMed Scopus (301) Google Scholar), human RTEF-1 (hRTEF-1; Ref. 12.Stewart A.F.R. Richard 3rd, C.W. Suzow J. Stephan D. Weremowicz S. Morton C.C. Adra C.N. Genomics. 1996; 37: 68-76Crossref PubMed Scopus (35) Google Scholar), human DTEF-1 (HDTEF-1; Ref.34.Jacquemin P. Martial J.A. Davidson I. J. Biol. Chem. 1997; 272: 12928-12937Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar), and mouse ETF (mETF; Ref. 35.Yasunami M. Suzuki K. Houtani T. Sugimoto T. Ohkubo H. J. Biol. Chem. 1995; 270: 18649-18654Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar). The MAPK site at Ser-322 in human RTEF-1 is boxed.View Large Image Figure ViewerDownload Hi-res image Download (PPT)In TEF-1, phosphorylation sites appear to cluster around the DNA binding domain (see Fig. 1). Two of the sites that flank the DNA binding domain in TEF-1 conform to the consensus for protein kinase A (PKA) (27.Pearson R.B. Kemp B.E. Methods Enzymol. 1991; 200: 62-81Crossref PubMed Scopus (866) Google Scholar). There is no consensus PKA site in human RTEF-1. PKA signaling is important in maintaining the expression of the α-myosin heavy chain and other genes in the healthy myocardium. Expression of the α-myosin heavy chain gene is regulated by PKA (28.Gupta M.P. Gupta M. Stewart A. Zak R. Biochem. Biophys. Res. Commun. 1991; 174: 1196-1203Crossref PubMed Scopus (33) Google Scholar), and its basal and PKA-dependent expression requires a hybrid MCAT/E-boxcis-element (29.Gupta M.P. Gupta M. Zak R. J. Biol. Chem. 1994; 269: 29677-29687Abstract Full Text PDF PubMed Google Scholar) to which TEF-1 and the basic helix-loop-helix leucine zipper protein Max bind cooperatively (30.Gupta M.P. Amin C.S. Gupta M. Hay N. Zak R. Mol. Cell. Biol. 1997; 17: 3924-3936Crossref PubMed Google Scholar). Thus, TEF-1 and/or its cofactor Max appear to be the target of PKA signaling. Mice homozygous for a TEF-1 null allele die during embryonic development with a thin-walled myocardium reminiscent of a dilated cardiomyopathy (31.Chen Z. Friedrich G.A. Soriano P. Genes Dev. 1994; 8: 2293-2301Crossref PubMed Scopus (259) Google Scholar). More recently, transgenic mice expressing a dominant-negative form of the CREB transcription factor in the myocardium, one that cannot be phosphorylated by PKA (CREBA133), developed dilated cardiomyopathy that closely resembled many of the anatomical, physiological, and clinical features of human idiopathic-dilated cardiomyopathy (32.Fentzke R.C. Korcarz C.E. Lang R.M. Lin H. Leiden J.M. J. Clin. Invest. 1998; 101: 2415-2426Crossref PubMed Scopus (215) Google Scholar). Thus, removing components of PKA signaling in the myocardium results in heart failure, perhaps by allowing PKC and MAPK signaling to go unchecked.From these studies and the results reported here, it appears that TEF-1 and RTEF-1 may serve two different functions, where TEF-1 activates myocardial genes under basal conditions and in response to PKA, whereas RTEF-1 activates fetal genes under conditions that mimic heart failure. Therapeutic strategies that target RTEF-1 may be of benefit in limiting the progression to heart failure. Cardiac myocytes respond to α1-adrenergic receptor stimulation by a progressive hypertrophy (1.Simpson P. J. Clin. Invest. 1983; 72: 732-738Crossref PubMed Scopus (546) Google Scholar), accompanied by a characteristic reactivation of many fetal genes, including β-myosin heavy chain (2.Waspe L.E. Ordahl C.P. Simpson P.C. J. Clin. Invest. 1990; 85: 1206-1214Crossref PubMed Scopus (155) Google Scholar), skeletal muscle α-actin (SKA)1 (3.Long C.S. Ordahl C.P. Simpson P.C. J. Clin. Invest. 1989; 83: 1078-1082Crossref PubMed Scopus (105) Google Scholar, 4.Karns L.R. Kariya K. Simpson P.C. J. Biol. Chem. 1995; 270: 410-417Abstract Full Text Full Text PDF PubMed Scopus (162) Google Scholar), and brain natriuretic factor (5.Thuerauf D.J. Glembotski C.C. J. Biol. Chem. 1997; 272: 7464-7472Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). The α1-adrenergic stimulation of these promoters requires an MCAT element (4.Karns L.R. Kariya K. Simpson P.C. J. Biol. Chem. 1995; 270: 410-417Abstract Full Text Full Text PDF PubMed Scopus (162) Google Scholar, 5.Thuerauf D.J. Glembotski C.C. J. Biol. Chem. 1997; 272: 7464-7472Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar), with the sequence CATN(T/C)(T/C) (6.Farrance I.K. Mar J.H. Ordahl C.P. J. Biol. Chem. 1992; 267: 17234-17240Abstract Full Text PDF PubMed Google Scholar). Transcription factors of the transcription enhancer factor-1 (TEF-1) multigene family bind to MCAT elements in the promoters of many genes expressed in cardiac and skeletal muscle cells (7.Larkin S.B. Farrance I.K.G. Ordahl C.P. Mol. Cell. Biol. 1996; 16: 3742-3755Crossref PubMed Scopus (72) Google Scholar). Thus, a role for TEF-1-related transcription factors in mediating the α1-adrenergic response has been proposed (4.Karns L.R. Kariya K. Simpson P.C. J. Biol. Chem. 1995; 270: 410-417Abstract Full Text Full Text PDF PubMed Scopus (162) Google Scholar, 8.Kariya K. Farrance I.K. Simpson P.C. J. Biol. Chem. 1993; 268: 26658-26662Abstract Full Text PDF PubMed Google Scholar, 9.Kariya K. Karns L.R. Simpson P.C. J. Biol. Chem. 1994; 269: 3775-3782Abstract Full Text PDF PubMed Google Scholar). Previously, we showed that the TEF-1-related factor RTEF-1 could potentiate the α1-adrenergic stimulation of the β-myosin heavy chain and SKA promoters, when overexpressed in cardiac myocytes. In contrast, TEF-1 did not affect their response to α1-adrenergic stimulation. Thus, we proposed a role for RTEF-1 in mediating the α1-adrenergic reactivation of fetal genes in cardiac myocytes (10.Stewart A.F.R. Suzow J. Kubota T. Ueyama T. Chen H.H. Circ Res. 1998; 83: 43-49Crossref PubMed Scopus (65) Google Scholar). The different effects of TEF-1 and RTEF-1 overexpression on the α1-adrenergic response of the SKA promoter must reflect differences in how TEF-1 and RTEF-1, or their associated co-factors, are modified by α1-adrenergic signaling. TEF-1 and RTEF-1 are identical in their DNA binding domains, are highly conserved in their carboxyl-terminal activation domains, but diverge in sequences flanking the DNA binding domain (see Fig. 1). To what extent these divergent sequences confer functional differences was not known. In the present study, we took advantage of the different response of a minimal α1-adrenergic-responsive SKA promoter to TEF-1 and RTEF-1 overexpression to examine the functional differences between TEF-1 and RTEF-1. By swapping functional domains between TEF-1 and RTEF-1 in chimeric expression constructs, the carboxyl-terminal domain of RTEF-1 was found to contain the α1-adrenergic-responsive sequences. Site-directed mutagenesis further mapped the serine residues targeted by α1-adrenergi
Referência(s)