|Home | About | Journals | Submit | Contact Us | Français|
Little is known regarding the mechanisms that control the expression of G-protein α, β, and γ subtypes. We have previously shown that the G-protein γ3 gene is expressed in the heart, brain, lung, spleen, kidney, muscle, and testis in mice. We have also reported that the G-protein γ3 subunit is expressed in rat cardiac myocytes, but not in cardiac fibroblasts. Other studies have shown that the γ3 subunit couples to the angiotensin A1A receptor in portal vein myocytes, and has been shown to mediate β-adrenergic desensitization in cardiac myocytes treated with atorvastatin. In the present study, we evaluated G-protein γ3 promoter-luciferase reporter constructs in primary myocytes to identify key regulatory promoter regions. We identified two important regions of the promoter (upstream promoter region [UPR] and downstream promoter region [DPR]), which are required for expression in cardiac myocytes. We observed that removal of 48 bp in the UPR diminished gene transcription by 75%, and that the UPR contains consensus elements for myocyte-specific M-CAT and myocyte enhancer factor 1 (MEF-1) elements. The UPR and DPR share transcription factor elements for myocyte-specific M-CAT element. We observed that cardiac myocyte proteins bind to γ3 oligonucleotides containing transcription factor elements for myocyte-specific M-CAT and MEF-1. Myocyte-specific M-CAT proteins were supershifted with transcriptional enhancer factor-1 (TEF-1) antibodies binding to the γ3 M-CAT element, which is in agreement with reports showing that the M-CAT element binds the TEF-1 family of transcription factors. The 150 bp DPR contains three M-CAT elements, an INR element, an upstream stimulatory factor 1 element, and the transcription start site. We have shown that myocyte γ3 gene expression is regulated by myocyte-specific M-CAT and MEF-1 elements.
A family of heterotrimeric G-proteins transmits signals from heptahelical receptors to their respective effectors (Gilman, 1987; Neer, 1995). Relatively little is known about G-protein βγ subunits in the heart despite mounting evidence demonstrating their importance in receptor recognition (Gilman, 1987; Birnbaumer, 1992; Neer, 1995) and effector regulation (Kleuss et al., 1993; Wickman et al., 1994). The effector targets of G-protein βγ subunits include ion channels, phospholipase A2, phospholipase C, adenylyl cyclase, G-protein-coupled receptor kinases, PI3 kinase, Ras, Raf-1, Bruton tyrosine kinase, Tsk tyrosine kinase, and plasma membrane calcium pumps (Clapham and Neer, 1997). Both reconstitution and genetic approaches have shown that the nature of the γ subunit is an important determinant for interaction of the G-protein with the receptor (Kleuss et al., 1993; Richardson and Robishaw, 1999). Studies have revealed that the γ3 subunit couples to the somatostatin receptor in the brain (Kleuss et al., 1993), and the fluorescent protein-tagged γ3 subunit regulated Ca2+ channel inactivation (Zhou et al., 2003). Mice lacking the γ3 gene showed susceptibility to seizures and a defective immune response (Schwindinger et al., 2004; Dubeykovskiy et al., 2007). In the heart, studies have shown that the γ3 subunit couples the angiotensin AT1A receptor to increase cytoplasmic Ca+2 in rat portal vein myocytes (Macrez et al., 1997), and the overexpression of the γ3 subunit resulted in severe heart defects and upregulation of mitogen-activated protein kinase activity in zebrafish (Bell et al., 1999; Kelly et al., 2001). Other investigators have shown that β-adrenergic desensitization with atorvastatin involves reduced isoprenylation of γ3 proteins in cardiac myocytes (Muhlhauser et al., 2006). We have previously demonstrated that the γ3 gene is expressed in cardiac myocytes, but not in fibroblasts (Hansen et al., 1995). To investigate the regulation of the γ3 subunit in the heart, we have evaluated the expression levels of γ3 luciferase reporter constructs in cardiac myocytes.
Cardiac myocytes were prepared from the hearts of 1–2-day-old Sprague-Dawley rats (Charles River Company, Wilmington, MD) as reported (McWhinney et al., 1997; McWhinney et al., 2000a,b). Briefly, the ventricles were removed and digested with a mixture of trypsin, chymotrypsin, and elastase in a Celstir apparatus at 37°C. Myocytes were suspended in DMEM containing 10% fetal bovine serum, 100 μM 5-bromo-2′-deoxyuridine, 50units/mL penicillin, and 50 μg/mL streptomycin and plated at a density of 10 × 106/100-mm dish on 0.1% gelatin-coated tissue culture dishes. After 24 h, the cardiac myocytes were serum-starved in DMEM with 0.1% fetal bovine serum for 72 h prior to experiments.
The p(EcoRI)-γ3-luciferase-reporter contains DNA from EcoRI to KpnI restriction enzyme sites (Dubeykovskiy et al., 2007), and was cloned into the KpnI site in the p-GL3-basic luciferase vector (Promega, Madison, WI). The p(SacI)-γ3-luciferase reporter was generated by using restriction enzymes SacI(−1550) and KpnI(−555). The SacI-KpnI fragment was subcloned into the pGL3-basic vector. The p(FspI)-γ3-luciferase reporter was generated using restriction enzymes FspI(−1502) and KpnI. The FspI/KpnI fragment was sub-cloned into the SmaI site in the pGL3-basic vector. The p(BspmI)-γ3-luciferase reporter was generated using restriction enzymes SacI and BspmI to remove it from the p(EcoRI)-γ3-luciferase reporter. The α-myosin heavy chain (α-MHC) promoter was removed with restriction enzyme digestion with HindIII (Gupta et al., 1994). The 4.2 kb α-MHC promoter was subcloned into the HindIII site of the basic pGL3-luciferase construct (Promega).
Cultured cells and/or cardiac myocytes were transfected with 1 μg of plasmid DNA. An aliquot of 1 μg of luciferase reporter plasmid was mixed with 7 μL of Lipofect-Amine Plus reagent (Invitrogen, Carlsbad, CA) and 100 μL of Opti-MEM medium (Invitrogen). After 5 h, the transfection medium was removed and the cells were maintained in Opti-MEM medium without antibiotics. Posttransfection 24 h, cells were transferred to DMEM medium (Invitrogen) with 10% fetal calf serum and antibiotics for 72 h (McWhinney et al., 2000a,b).
After 72 h of transfection, the cells were disrupted with 1× lysis buffer (Promega). Cell lysates were assayed for luciferase gene expression using a Bertholt luminometer and Promega substrate reagents (McWhinney et al., 2000a,b). For luciferase gene expression assays, the cells were lysed in 0.5 mL of 1× lysis buffer (Promega). The cell lysates were centrifuged at 15,000 g for 1 min at 4°C, and the supernatants were transferred to a new tube. Approximately 50 μL of the supernatant was assayed in triplicate for luciferase gene expression using a luciferase assay system with reporter lysis buffer (Promega). Relative light units were measured utilizing a Lumat LB9507 EG&G Berthold luminometer with cell lysates. Following luciferase assays, β-galactosidase enzyme assays were performed in triplicate samples. The cell lysates from the luciferase assays were assayed spectrophotometrically for β-galactosidase enzyme activity using the βgalactosidase enzyme assay system with reporter lysis buffer (Promega). The luciferase relative light unit numbers were divided by the β-galactosidase enzyme assay numbers to correct for transfection efficiency in cells. The luciferase activities were normalized to β-galactosidase enzyme activities to correct for differences in transfection efficiency, and the normalized luciferase activities were expressed as fold induction relative to the pGL3-basic luciferase reporters with no promoter (control). Both γ3-luciferase gene expression and α-MHC-pGL3-luciferase expression (Gupta et al., 1994) were measured in transiently transfected cardiac myocytes and/or cell lines (McWhinney et al., 2000a,b). Each experiment was repeated at least three times.
Nuclear and/or whole-cell extracts were used in electrophoretic mobility shift analysis (EMSA) and were incubated for 20 min at 25°C with double-stranded oligonucleotides (McWhinney, 1996; McWhinney et al., 2000a,b). Protein extracts were prepared from cultured cells as reported (McWhinney, 1996; McWhinney et al., 1997). Competition experiments were performed with unlabeled oligonucleotides utilizing a 100-fold excess of competitor oligonucleotides to assess the specificity of the protein–DNA interactions in the EMSA. Protein complexes were electrophoresed on a 4.5% native polyacrylamide gel in 1× tris-glycine buffer (Sigma-Aldrich, Inc., St. Louis, MO) as reported (McWhinney et al., 2000a,b). Quantitations were made on a PhosphoImager ( Model SF; Molecular Dynamics, GE Healthcare, Piscataway, NJ) using ImageQuant Program.
Protein extracts were prepared from cultured cardiac myocytes using modified protocols for whole-cell extracts and nuclear extracts as reported (McWhinney et al., 1997). Briefly, cells were removed from tissue culture plates using Versene (phosphate-buffered saline with 0.2g/L EDTA; Invitrogen) and/or 0.25% trypsin (Invitrogen). Cells were washed in phosphate-buffered saline (9.0 g/L NaCl, 0.144 g/LKH2PO4, and 0.795 g/LNA2HPO4 7H2O; Invitrogen) and then with buffer A (20 mM HEPES, 1.5 mM MgCl2,10mM KCl, and 0.2mM EDTA). Cells were resuspended in buffer A with 0.625% NP40 detergent and allowed to swell on ice for 15min. Cell suspension was dounced 10 times to lyse cells, and nuclei were pelleted at 14,000 rpm for 30 s at 4°C. The pelleted nuclei were resuspended in 50 μL of buffer C (20 mM HEPES, 1.5mM MgCl2, 0.2 mM EDTA, and 375 mM KCl), and incubated on ice for 20 min to extract proteins. The sample was centrifuged for 5 min at 14,000 rpm at 4°C to pellet nuclear membranes. The supernatant was aliquoted and stored at −80°C.
The oligonucleotides were end-labeled with T4 polynucleotide kinase (Promega) and 32P-γ-ATP (Dupont, NEN, Boston, MA) utilizing 10× kinase buffer (Promega) for 15 min at 37°C (McWhinney, 1996). The unincorporated nucleotides were removed by filtration with MW 3000 membranes (Amicon, Beverly, MA). The oligonucleotides were resuspended in 100 μL of TE buffer (10 mM Tris, pH 7.5,1 mM EDTA, pH 8.0). Listed below are the γ3 DNA sequences from the 1.0 kb promoter for the double-stranded oligonucleotides used in EMSA experiments. The identified transcription factor elements are underlined.
Upstream promoter region (UPR) oligonucleotides: Myocyte enhancer factor 1 (MEF-1):
5′-GAGCCTTGCACCCTGCCAG-3′ M-CAT: 5′-TGCCAGAATCTGAATCTTGTG-3′
Downstream promoter region (DPR) oligonucleotides: M-CAT/DPR (+C/D):
5′-CCAGGCGGTTGGTATGCGCCAGAATC-3′ M-CAT/E: 5′-CTGTCCAGGCAGTTGGTATGCGCC-3′
Upstream stimulatory factor (USF): 5′-CCGGAGGCTGACTCCCTTGCTGTC-3′
Antibodies were purchased from Santa Cruz Biochemicals (USF-1 sc-229× [c-20]× and MyoD sc-760× [M-318]×), and were used in EMSA experiments.
The HL-1 myocyte cell line was obtained from Proctor and Gamble Company (Cincinnati, OH) and Dr. William Claycomb (McWhinney et al., 2000b). Briefly, the HL-1 cells were maintained in Ex-cell 320 medium (JRH Biosciences, Lenexa, KS) supplemented with 10% fetal bovine serum (Invitrogen), 10 μg/ mL endothelial cell growth supplement (Upstate Bio-technology, Lake Placid, NY), 1 μM retinoic acid (Sigma), 10 μM norepinephrine (Invitrogen), 100 units/mL penicillin, 100 μg/mL streptomycin (Invitrogen), and an additional 1× nonessential amino acids (Invitrogen). The HL-1 cells were plated at a density of 8.0×104 cells/cm2 on dishes precoated with 0.1% gelatin and 0.1% fibronectin on 100-mm dishes (106 cells). The cells were switched to medium without nor-epinephrine for 4 days prior to experiments.
Cloned γ3 DNA was used as templates for the following primers to generate the 150-bp γ3 DPR fragment (McWhinney et al., 2000b) using the Advantage Plus PCR kit (Clontech, Palo Alto, CA). Cycling: 25 cycles of 94°C30s,68°C 2min, final 68°C for 5min, PTC-100 Thermal Controller (MJ Research, Watertown, MA). PCR products or oligonucleotides were end-labeled with T4 polynucleotide kinase (Promega) and 32-P-γ-ATP (Dupont) utilizing 10× kinase buffer (Promega) for 15min at 25°C. Sense oligonucleotide: 5′-GCAGGTAGTGA GAGAGCAGC-3′; antisense oligonucleotide: 5′-CTGGTACCTGCTGAAGGATGC-3′.
We have previously shown that the γ3 gene is expressed in the heart, brain, lung, spleen, kidney, muscle, and testis in mice (Schwindinger et al., 2004; Dubeykovskiy et al., 2007), and in rat that the γ3 subunit is expressed in cardiac myocytes, but not cardiac fibroblasts (Hansen et al., 1995). To understand the regulation of the γ3 promoter in the heart, we analyzed the ability of the γ3 promoter to drive luciferase gene transcription in cardiac myocytes. We prepared a limited series of γ3 promoter luciferase reporters and evaluated their effectiveness in transfected cardiac myocytes (McWhinney et al., 2000b), an HL-1 myocyte cell line (McWhinney et al., 2000b), and a Swiss 3T3 fibroblast cell line (Kurokawa et al., 1994). The experiments in fibroblasts were used as negative controls. To assess promoter activity, we inserted 7.1 kb of γ3 DNA upstream of the gene encoding firefly luciferase gene in the pGL-3-basic-luciferase reporter construct generating the p(EcoRI)-γ3-luciferase reporter construct (Fig. 1A). The promoter activity of this construct was determined by transient transfection expression in cardiac myocytes (McWhinney et al., 2000a), myocyte HL-1 cells (McWhinney et al., 2000b), and Swiss 3T3 fibroblasts (Kurokawa et al., 1994). The luciferase activities were normalized to β-galactosidase enzyme activities to correct for differences in transfection efficiency, and the normalized luciferase activities were expressed as fold induction relative to the pGL3-basic luciferase reporters with no promoter (control). We demonstrate that the p(EcoRI)-γ3-luciferase reporter has 2.4-fold activity in cardiac myocytes and 3.3-fold activity in myocyte HL-1 cells, while displaying no activity in fibroblasts compared to vector (Fig. 1A). Thus, the γ3 luciferase reporter construct shows activity only in the cell types previously reported to express γ3 mRNA and does not show activity in cell types reported not to express γ3 mRNA (Hansen et al., 1995; Downes et al., 1998).
To map regulatory elements in the γ3 promoter region, we generated luciferase reporters with varying amounts of γ3 promoter from the p(EcoRI)-γ3-luciferase reporter construct. We analyzed the different γ3 luciferase reporters in transfected cardiac rat myocytes and fibroblasts (Fig. 1B). Fibroblasts are used as negative controls for the luciferase assays. The p(SacI)-γ3-luciferase construct showed enhanced 4.8-fold luciferase expression in cardiac myocytes with only control-level activity in fibroblasts (Fig. 1B). These data show that the 1.0 kb p(SacI)-γ3-luciferase construct is expressed at two-fold higher levels in cardiac myocytes than the original p(EcoRI)-γ3-luciferase construct, demonstrating that the 7.1 kb p(EcoRI)-γ3-luciferase construct contains negative regulatory elements. The p(FspI)-γ3-luciferase reporter construct, which has 48 bp removed from the 5′ end of the construct, demonstrated a 3.6-fold decrease in luciferase activity, which is 75% of the promoter activity in cardiac myocytes (Fig. 1B). The p(BspmI)-γ3-luciferase reporter construct, which has 150 bp removed between the BspMI and KpnI restriction enzyme sites at the 3′ end of the promoter, displayed a 4.8-fold decrease to control levels in cardiac myocytes. Neither the p(FspI)-γ3-luciferase reporter construct nor the p(BspMI)-γ3-luciferase reporter construct showed any activity in fibroblasts, which were used as negative controls (data not shown). These data demonstrate that both the 48 bp (UPR) and 150 bp (DPR) promoter sequences contain important regulatory elements for expression in cardiac myocytes.
The 1.0 kb p(SacI)-γ3-luciferase construct demonstrated the highest luciferase activity, showing that it contains important regulatory regions for expression in cardiac myocytes. Since removal of 48 bp from the p(SacI)-γ3-luciferase construct reduced gene transcription by 75% in cardiac myocytes, we created an oligonucleotide that covers the 48 bp UPR to determine if myocyte proteins would bind to this DNA.
To evaluate protein binding to individual transcription factor elements within the UPR, we synthesized two oligonucleotides, which cover 71% of the UPR, excluding a TG repetitive element at the end of the 48 bp region (Fig. 1C). In Figure 2A, we observed that proteins bind to the M-CAT oligonucleotides forming one protein complex. We demonstrate that a 100-fold excess of unlabeled self M-CAT oligonucleotides competed the protein complex, but that a nonspecific (+NS) oligonucleotide had no effect on the myocyte protein complex (Fig. 2A). These data show that the complex is specific for the M-CAT element. In addition, cardiac myocyte proteins binding to the M-CAT oligonucleotide were supershifted with transcriptional enhancer factor-1 (TEF-1) antibodies (Fig. 2B). The M-CAT element is reported to bind the TEF-1 family of transcription factors (Stewart et al., 1998).
We noticed that the UPR contained a 14 bp consensus element for myocyte-specific MEF-1 (Buskin and Hauschka, 1989; Horlick and Benfield, 1989; Amacher et al., 1993; Apone and Hauschka, 1995; Feo et al., 1995; Klamut et al., 1997), which is adjacent to two myocyte-specific M-CAT elements (Berberich et al., 1993; Gupta et al., 1994). In Figure 2C, we observed that proteins bind to the MEF-1 element forming a single complex. To investigate the specificity of protein complexes on these oligonucleotides, we performed competition experiments with unlabeled self (+S) oligonucleotides and nonspecific (+NS) oligonucleotides. We show that the myocyte protein complex formed on the labeled MEF-1 oligonucleotide is competed by a 100-fold excess of unlabeled self (+S) oligonucleotides, but is not competed by a 100-fold excess of nonspecific (+NS) oligonucleotide, demonstrating that the protein complex is specific (Fig. 2C). Investigators have shown that MyoD1 proteins bind to the MEF-1 element (Lassar et al., 1989). Therefore, we added MyoD1 antibodies to the protein complex on MEF-1 oligonucleotides (Fig. 2D). We did not observe a supershift with the addition of the MyoD antibody, but instead observed a reduction in cardiac myocyte proteins binding to the MEF-1 oligonucleotide (Fig. 2D).
With removal of 150 bp at the 3′ end of the promoter, γ3-luciferase gene transcription decreased to control levels. By EMSA, we observed the binding of several myocyte protein complexes to DPR oligonucleotides (data not shown). To measure myocyte protein binding to individual transcription factor binding elements in the DPR, we designed overlapping oligonucleotides, which span identified transcription factor elements in the DPR (Fig. 1C). Using an M-CAT/E oligonucleotide spanning overlapping E-box and two M-CAT elements from the DPR, we observed that proteins form three complexes, designated as A, B, and NS (Fig. 3A, lane 2). The A and B protein complexes were reduced when competed with unlabeled self (+S), while the bottom (NS) complex was enhanced (Fig. 3B, lane 3). The bottom protein (NS) complex was competed by nonspecific (+NS) oligonucleotides, demonstrating that it is a nonspecific protein complex (Fig. 3A, lane 6). These data demonstrate that the myocyte protein complexes A and B are specific for the overlapping E-box and two M-CAT elements (Fig. 3A). The M-CAT/R(+C/R) oligonucleotides, which contain two M-CAT elements and mutant E-box, were used for competition studies with the labeled M-CAT/E oligonucleotides. Competition with unlabeled + C/R oligonucleotides eliminated the B complex and reduced complex A, but did not affect the nonspecific (NS) complex, demonstrating that the two M-CAT elements are able to compete the myocyte complexes in complexes A and B (Fig. 3A, lane 4). To determine if proteins that specifically recognize only the M-CAT element were present in myocyte A and B complexes, we competed with unlabeled + C/D oligonucleotides that contain an M-CAT element, but do not contain an E-box. Competition with excess + C/D oligonucleotides eliminated complex B and reduced complex A to an extent similar to self M-CAT/E oligonucleotides (Fig. 3A, lane 5). Both M-CAT-containing oligonucleotides (+C/R and + C/D) abolish complex B and reduce complex A on the M-CAT/E oligonucleotides, demonstrating that M-CAT binding proteins are present in the myocyte protein complexes A and B (Fig. 3A, lanes 4 and 5). The + C/D oligonucleotides contain one M-CAT element, but do not contain an E-box element, demonstrating that the M-CAT element is responsible for the competition of complex A and B. These data provide evidence that M-CAT binding proteins are present in myocyte complexes A and B. To further evaluate myocyte protein binding to M-CAT elements within the promoter, we labeled M-CAT/DPR oligonucleotides, which contain an M-CAT element from the DPR, which does not contain an E-box (Berberich et al., 1993; Gupta et al., 1994). In Figure 3B, we observed one major myocyte complex binding to the M-CAT/DPR oligonucleotides (lane 2). The protein complex on the M-CAT/DPR oligonucleotides was competed by un-labeled self (+S) oligonucleotides, but were unaffected by nonspecific (+NS) oligonucleotides (Fig. 3B, lanes 3 and 4). These studies confirm that proteins bind to M-CAT elements in the promoter (Gupta et al., 1994).
In the DPR, there is an exact consensus transcription factor binding element for USF1, which is immediately upstream of the transcription start site (USF1; Xiao et al., 2002). In Figure 4A, we observed that one major myocyte protein complex binds to USF oligonucleotides. The protein complex was abolished when competed by self USF oligonucleotides (Fig. 4A, lane 3), but the complex was unaffected by nonspecific (+NS) oligonucleotides, showing that the myocyte complex is specific for its sequence (Fig. 4A, lane 4). In Figure 4B, we explored whether the addition of a USF1 antibody to the EMSA protein complex would cause a supershift or whether it would cause inhibition of the USF proteins binding to the USF oligonucleotide. We observed that addition of the USF1 antibody caused a twofold reduction in the binding of USF proteins to the USF oligonucleotides (Fig. 4B).
To compare the γ3 promoter strength with a known cardiac-specific promoter, we transfected the α-MHC-luciferase reporter construct into cardiac myocytes and fibroblasts.
In Figure 5, the α-MHC-promoter-luciferase reporter demonstrated a 5.0-fold ± 0.40 increase in luciferase gene expression compared to control and no expression in fibroblasts, demonstrating cardiac-specific expression. The p(SacI)-γ3-luciferase reporter construct showed 4.8-fold ± 0.40 in cardiac myocytes and no expression in fibroblasts above control levels, demonstrating that γ3-luciferase reporter activity is comparable to α-MHC-luciferase expression in cardiac myocytes (Fig. 5). The p(EcoRI)-γ3-luciferase reporter construct also shows enhanced luciferase activity in cardiac myocytes, but no activity in fibroblasts. These data demonstrate that γ3-luciferase constructs display strong gene expression in cardiac myocytes.
We have identified two important regulatory regions in the promoter, which are required for γ3 expression in cardiac myocytes. The removal of the 48 bp UPR from the 1.0 kb promoter resulted in a 75% reduction in γ3 luciferase gene expression in cardiac myocytes. The UPR contains two myocyte-specific transcription factor elements, that is, M-CAT and MEF-1. The 150 bp DPR contains the reported transcription start site for the γ3 gene in mouse and human species (Schwindinger et al., 2004), and its removal abolished all transcription activity, which demonstrates that the DPR contains the core promoter. Using the reported transcription start site (Schwindinger et al., 2004), the DPR contains 88 bp upstream and 43 bp downstream of the transcription start site, which overlaps a consensus INR element (Kraus et al., 1996) (Table 1).The DPR has 60% G + C content, does not contain a TATA-box, but does contain a USF1 element 24 bp upstream of the INR element and the transcription start site (Table 1).
The UPR and DPR contain multiple M-CAT transcription factor elements (Gupta et al., 1994). The M-CAT transcription factor elements are highly conserved between mouse and human species, suggesting that the myocyte-specific M-CAT elements are important for γ3 gene expression in cardiac myocytes. Studies have shown that myocyte-specific M-CAT elements are required for the expression of cardiac genes in the heart (Gupta et al., 1994; Molkentin and Markham, 1994). Myocyte-specific M-CAT elements have been shown to regulate the cardiac α-MHC promoter (Gupta et al., 1994), the cardiac β-myosin heavy chain (β-MHC) promoter (Giger et al., 2002), human cardiac phospholamban promoter (McTiernan et al., 1999), cardiac B-type natriuretic peptide promoter (Thuerauf and Glembotski, 1997), the cardiac troponin T promoter (Butler and Ordahl, 1999), and the cardiac troponin C promoter (Larkin et al., 1996). The cardiac troponin T promoter contains two M-CAT elements in tandem (Butler and Ordahl, 1999). In the UPR, the two M-CAT elements are tandem to one another; in the DPR, two of the M-CAT elements are separated by only 4 bp. Two of the three M-CAT elements in the DPR are identical to the M-CAT elements in the cardiac β-MHC and mouse cardiac Troponin C promoters (Larkin et al., 1996). TEF-1 transcription factor binds to M-CAT elements in the promoters of many genes expressed in cardiac muscle cells (Larkin et al., 1996; Stewart et al., 1998).
The UPR contains a 14 bp myocyte-specific MEF-1 consensus element, which binds cardiac myocyte proteins (Buskin and Hauschka, 1989; Horlick and Benfield, 1989; Amacher et al., 1993; Apone and Hauschka, 1995; Feo et al., 1995; Klamut et al., 1997). The MEF-1 has been shown to positively regulate transcription from muscle and/or cardiac promoters, that is, muscle creatine kinase (Buskin and Hauschka, 1989; Amacher et al., 1993), β enolase gene (Feo et al., 1995), and muscle/cardiac-specific form of the dystrophin gene (Klamut et al., 1997). Investigators have reported that MyoD binds the 14 bp MEF-1 element and that MyoD antibodies recognize MEF-1 proteins (Lassar et al., 1989). Thus, the 48 bp UPR, which accounts for 75% of the promoter activity in cardiac myocytes, contains two myocyte-specific transcription factor elements.
In addition to the myocyte-specific transcription factor elements identified, the 1.0 kb promoter contains one USF element that is immediately upstream of the transcription start site and INR element. The USF1 elements have been shown to be important regulatory elements in cardiac promoters (Makaula et al., 2006). The USF1 transcription factor has been shown to regulate the cardiac isoform of the acetyl-CoA carboxlyase promoter (Makaula et al., 2006), enhance the contractile-mediated activity of the cardiac α-MHC promoter (Xiao et al., 2002), and control the expression of the cardiac ventricular myosin light-chain 2 promoter (Navankasattusas et al., 1994). It is reported that USF1 proteins can act as a docking proteins to recruit basal transcription factors to facilitate formation of the transcription preinitiation complex (Roy et al., 1991), and have been shown to interact with TFIID to stimulate gene transcription (Kokubo et al., 1993). The USF element in the DPR is 24 bp upstream of the INR element and the TFIID element (Table 1), suggesting that it may play a similar role in the γ3 promoter.
In summary, we have shown that the γ3 luciferase constructs are expressed at high levels in cardiac myocytes, while showing no activity in fibroblasts. The γ3 promoter shares myocyte-specific transcription factor elements with the α-MHC promoter. The UPR, which accounts for 75% of the promoter activity in cardiac myocytes, contains myocyte-specific M-CAT and MEF-1 elements. The DPR contains USF1, INR elements, and the transcription start site. We have shown that myocyte proteins specifically bind to MEF-1, M-CAT, and USF elements to regulate γ3 gene expression in cardiac myocytes.
We thank Anna Kempski and Lois Carl for providing expert technical assistance. This work was supported by National Institutes of Health Research Grant #HL066776-01A2 awarded to C.D.M.