Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

-Adrenergic receptors (-AR) are G_s-coupled receptors that transduce the binding of catecholamines into activation of adenylyl cyclase and elevation of cyclic AMP. The ₁-AR seems to be the dominant subtype of the -AR family for mediating the chronotropic, inotropic, and lusitropic actions of catecholamines in the heart. Targeted disruption of the ₁-AR gene produced mice that lacked chronotropic and inotropic responsiveness to catecholamines, further demonstrating the importance of ₁-AR for proper cardiac function (Rohrer et al., 1996). In addition to the ₁-AR, two additional subtypes of -AR have been cloned. The three subtypes of -AR are products of separate genes and are unusual in that they are intronless or, as in the case of the ₃-AR, contain a very small intron (Kobilka et al., 1987; Machida et al., 1990; Granneman et al., 1993; Searles et al., 1995). The 5'-flanking region of mammalian -AR reveal G+C-rich sequences which lack consensus TATA boxes to initiate transcription (Kobilka et al., 1987; Searles et al., 1995; Evanko et al., 1998). The major transcriptional start site (TSS) of the rat ₁-AR gene is at 253 relative to the first base of the initiation codon,¹ whereas that of the human ₂-AR gene is at 219, suggesting that similarities exist in their transcriptional initiation (Kobilka et al., 1987; Searles et al., 1995).

The expression of the ₁-AR is very low in tissues as well as in cultured primary ventricular myocytes (Bahouth et al., 1997a) and in cell lines that express the ₁-AR endogenously such as rat C6 glioma and human neuroepithelioma SK-N-MC cells (Bahouth et al., 1997b, 2001; Esbenshade et al., 1992). Promoter studies have shown that the 3-kb rat ₁-AR promoter has very low inherent ability to drive transcription compared with shorter promoter sequences (Searles et al., 1995, Bahouth et al., 1997b). It is likely that low basal expression of the ₁-AR is caused by inhibitory domains or to a weak basal promoter. The first possibility was addressed by deletion studies from the 5' end of the promoter. Shortening the 3-kb promoter to less than 1 kb substantially increased its activity in a variety of transiently transfected cell lines (Searles et al., 1995; Bahouth et al., 1997a,b). To characterize the basal promoter of the rat ₁-AR gene and to measure its relative transcriptional potency, we generated deletion mutants of the 5'-flanking sequence and then linked them to the expression of the gene for firefly luciferase. These constructs extended to bp 126 to avoid interference by regulatory sequences down-stream from bp 125 (Bahouth et al., 1997a,b). These analyses localized the sequence of the core promoter of the rat ₁-AR gene between 394 and 330 and identified two transcription factors that are involved in reciprocal regulation of the transcription of the ₁-AR gene.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Culture of Ventricular Myocytes. For each cell preparation, ventricles were isolated from 1- to 3-day-old Sprague-Dawley rats under aseptic conditions. Rat neonatal ventricular myocytes were dissociated with pancrelipase and isolated by discontinuous Percoll gradient centrifugation (Bahouth et al., 1997a). Using this procedure, highly enriched populations of ventricular myocytes (>95%) were obtained. Isolated myocytes were cultured on collagen-coated plates a density of 2 × 10⁶ cells per 60-mm plate in 68% DMEM, 17% medium-199, 10% horse serum, and 5% fetal bovine serum and antibiotics. The next day the medium was aspirated and replaced with 80% DMEM and 20% medium-199.

Construction of ₁-AR Promoter-Luciferase Chimera and Luciferase Assays. Progressive deletions within the KpnI-SacII genomic fragment of the rat ₁-AR encoding the sequences between 1251 to 126 were generated either by restriction enzymes or by PCR and then ligated into the promoterless pGL3basic luciferase expression plasmid. Each 60-mm plate was transfected with 5 µg of plasmid DNA composed of 3.8 µg of the smallest vector (pGL3basic), 1 µg of carrier pGEM7ZF⁺ DNA, and 0.2 µg of CMV-pRL (Renilla reniformis luciferase vector for correcting firefly luciferase activity) by the calcium phosphate precipitation technique (Bahouth et al., 1997a,b). In all transfections, the amount of each ₁-AR-luciferase construct was increased to an equivalent molar ratio of pGL3basic and the balance of the DNA was with pGEM7ZF⁺. In all experiments, the pGL3control vector, which is a luciferase vector driven by the SV40 promoter and enhancer sequences, was transfected in equimolar amounts to the pGL3basic vector. Luciferase assays were performed using the Dual Luciferase Assay System (Promega Corp., Madison, WI). The luminescence of each construct is expressed as a percentage of the expression of pGL3control after normalizing for transfection efficiency. Each construct was transfected into three 60-mm plates and these transfections were replicated for each cell type in a minimum of three separate experiments (n  9). The values from all experiments were combined and subjected to analysis of variance using Microsoft Excel (Microsoft Corp., Redmond, WA). Significance was determined by Student's t test (p = 0.05).

Culture and Transient Transfection of SL2 Cells. Drosophila melanogaster SL2 cells were cultured at room temperature in Schneider's Drosophila medium (SDM) supplemented with 10% FBS. DNA (1.5 µg per 35-mm plate) was mixed with 100 µl SDM and 9 µl of CellFECTIN solution (Invitrogen, Carlsbad, CA) for 15 min. DNA consisted of 0.5 µg of ₁-AR promoter luciferase vector, 0.1 µg pRL-CMV R. reniformis luciferase vector and the balance was composed of either carrier DNA or the appropriate Sp expression vector as described in the legend of Fig. 3. To the DNA-CellFECTIN mixture, 0.8 ml of SDM with 1.5% FBS was added, and the DNA/liposome complexes were layered over the cells for 5 h. After transfection, the DNA containing medium was replaced with 2 ml of SDM with 10% FBS for 36 h. Thereafter, the cells were harvested and luminescence was determined. The Sp1 and SP0 vectors under the control of the actin 5C promoter were obtained from R. Tjian (Kadonga et al., 1987; Courey and Tjian, 1988) and the other Sp vectors were provided by Guntram Suske and are described in Dennig et al. (1995).

Site-Directed Mutagenesis of ₁-AR Genomic Fragments. Mutagenesis of the KpnI-SacII genomic fragment of the ₁-AR gene was performed based on the methods described in the Transformer site-directed mutagenesis manual (CLONTECH, Palo Alto, CA). The sequence of the mutagenic primer for site directed mutagenesis of the triplet between 380 and 378 was 5'-pGGGACACCATTGTAAAGGGGCGTGCCTTG-3'; for mutating the sequence between 371 and 369 was 5'-pTTGTTCGGGGGCGAAACTTGGCGACGATTG-3'; and for mutating the sequence between 374 and 372 was 5'-pCCATTGTTCGGGGAAATGCCTTGGCGACG-3'. The underlined oligonucleotides indicate the mutated sequence. Plasmid DNA was sequenced by automated dye-termination sequencing using a primer 5'-TCTGGAAGAAGCCTGAGCAG corresponding to the sequence between 519 and 500 in 5'-flanking region the ₁-AR gene. Using the mutagenized KpnI-SacII fragment as template, the 3311 to 126 and 484 and 126 genomic fragments were prepared and subcloned into pGL3basic to generate the desired vector.

Gel Electromobility-Shift Assays (EMSA). Three double-stranded oligomers with CTAG overhangs representing the wild-type and mutated sequences between 385 and 365 in the ₁-AR promoter were synthesized. These were the wild type sequence 5'-ATTGTTCGGGGGCGTGCCTTG-3'; the mut-Egr-1 oligomer 5'-ATTGTaaaGGGGCGTGCCTTG-3', in which the Egr-1 site was mutated; and the mut-Sp1 oligomer 5'-ATTGTTCGGGGGCGaaaCTTG-3', in which the Sp1-binding site was mutated. These oligomers were labeled with Klenow enzyme and [-³²P]dCTP and combined with nuclear extracts for 20 min in a binding buffer composed of 80 mM KCl, 10 mM HEPES, pH 7.1, 1 or 2 µg of poly(dI-dC), and 10% glycerol (Bahouth et al., 1997a). The resulting complexes were resolved on 5% nondenaturing acrylamide gels in 25 mM Tris, 200 mM glycine at 4°C (Bahouth et al., 1997a). Nuclear extracts prepared from rat neonatal ventricular myocytes were prepared as described in Bahouth et al. (1997a). Sp1 was purchased from Promega Corp. Egr-1 cDNA cloned in pRSET-A (Cui et al., 1996), was obtained from N. Mackman (Scripps Research Institute, La Jolla, CA). Hexahistidine-tagged Egr-1 expressed in bacteria was partially-purified by affinity chromatography with nickel nitriloacetic acid resin (Cui et al., 1996).

DNase Footprinting Assay. The PstI Genomic fragment between 484 and +269 was ³²P-labeled on one end as described in Bahouth et al. (1997b). Each ³²P-labeled DNA fragment (20,000 cpm) was incubated with purified transcription factor (10 ng) in binding buffer composed of 20 mM HEPES, pH 7.6, 0.1 mM EDTA, 1 mM dithiothreitol, 10% glycerol, 50 mM NaCl, and 1 µg of poly(dI-dC) for 30 min at 0°C (Bahouth et al., 1997b). Digestion with 0.03-0.1 units of DNase I was allowed to proceed for 45 s, followed by adding 150 mM NaCl, 0.7% SDS, 15 mM EDTA, and 30 µg of yeast tRNA. The samples were extracted and subjected to electrophoresis on 6% acrylamide in 8 M urea gels. The protected DNA sequences were identified by running a separate lane containing a G sequence ladder generated by cleaving the ³²P-DNA fragment with piperidine as described previously (Bahouth et al., 1997b).

Generation of HeLa Cells with Tetracycline-Regulated Expression of Egr-1. HeLa S3 cells harboring a stably integrated copy of the regulator plasmid pTet-Off, were purchased from CLONTECH. pTet-Off contains a fusion of the Tet repressor and the carboxyl-terminal 130 amino acids of VP16 as well as a G418 resistance cassette (Gossen and Bujard, 1992; Yin et al., 1996). Transcription of the gene under the control of the tetracycline responsive element is inhibited by tetracycline or its analog doxycycline. Conversely, transcription of the gene of interest is active and maintained so long as tetracycline is absent (Gossen and Bujard, 1992; Yin et al., 1996). pTet-Off-HeLa S3 cells were cultured in DMEM plus 10% FBS supplemented with 600 µg/ml G418 and 5 ng/ml doxycycline.

Culture and Transient Transfection of SK-N-MC Cells. SK-N-MC cells are human neuroepithelioma cells that endogenously express the ₁-AR (Esbenshade et al., 1992; Bahouth et al., 2001). A mammalian expression vector for Sp1 was generated from pPac Sp1 as follows: a forward primer, 5'-GCAATGAACTCGTACTTTGGAACAGGC-3', and reverse primer, 5'-TCAGAAGCCATTGCCACTGATATTAAT-3', were used to amplify the 2.1-kb Sp1 cDNA fragment (Kadonga et al., 1987). The PCR-generated cDNA was cloned into the PCR 2.1 TOPO vector (Invitrogen), excised with KpnI and XbaI and cloned into the mammalian expression vector pcDNA 3.1(Invitrogen). The mammalian expression vectors for Egr-1, pCMV-Egr-1 and for ETTL, pCMV-ETTL were described previously (Sukhatme et al., 1988). SK-N-MC cells were cultured in DMEM supplemented with 10% FBS. For transfection, SK-N-MC cells were cultured in DMEM and transfected with 5 µg of each vector per 150-mm plate (10 µg DNA per plate) by the LipofectAMINE method for 6 h. After 6 h, the medium was aspirated, and the cells were cultured in DMEM supplemented with 10% FBS for 48 h. Total cellular RNA was extracted by the RNA-STAT 60 method and 25 µg or RNA was subjected to Northern blotting and transferred to Nytran filters as described previously (Bahouth et al., 2001). The blot was prehybridized for 6 h in Ultrahybrid solution (Ambion, Austin, TX), then incubated in Ultrahybrid solution containing 2 × 10⁶ cpm/ml of radiolabeled human ₁-AR probe for 16 h at 42°C. The ₁-AR probe was a 227-bp fragment from human ₁-AR cDNA corresponding to the sequences +522 to +748 (Bahouth et al., 2001). To control for the variability in RNA loading, the blot was stripped and reprobed with a 103-bp human cyclophilin cDNA probe corresponding to the sequences between +38 and + 140 (Ambion). The cpm in each band was counted by electronic autoradiography in the InstantImager (Packard Bioscience, Meriden, CT) and the data for each condition subtracted from background.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Transcriptional Regulation of the Rat ₁-AR Gene by Sp1. To define the region within the ₁-AR gene that is involved in regulating basal expression, deletion mutants of the 5'-flanking sequences between 1251 to 126 were constructed and linked to the expression of the firefly luciferase reporter gene. Progressive deletions from the 5' end indicated that the minimal promoter of the ₁-AR gene lies within a 65-bp region between 394 and 330 (Fig. 1A). The first 26-bp of the minimal promoter between 394 and 368 accounted for about 45 ± 5% of the basal activity. Deleting this construct down to 349 resulted in an additional 35 ± 5% decline in activity, and the remaining 20% of the activity was lost when the promoter was truncated to 330.

View larger version (40K): [in this window] [in a new window]   Fig. 1.   Characterization of the early promoter region of the rat 1-adrenergic gene by transient transfections and gel-shift assays. A, the sequences in the 1-AR promoter shown in the left side were ligated into the multiple cloning region of the promoterless firefly luciferase vector pGL3basic and transfected into ventricular myocytes. The activity of luciferase relative to the SV40-driven firefly luciferase vector pGL3control was determined. Transfections were performed in triplicate and replicated five times, n = 15. The sequence of the core promoter of the rat 1-AR gene and the percentile activity of its segments are shown. B, a 32P-labeled oligomer corresponding to the sequence between 385 and 365 in the rat 1-AR promoter was incubated with 1 µl of nuclear extract prepared from rat neonatal ventricular myocytes that were cultured in serum-free medium (lanes 2-10) and with IgG or immunoglobulins against Sp family members as indicated for 20 min. The complexes were resolved on 5% nondenaturing polyacrylamide gels in Tris/glycine and visualized by autoradiography. * and **, major and minor binding complexes of Sp1 and Sp3, respectively. ***, faster migrating band 3.

View larger version (70K): [in this window] [in a new window]   Fig. 2.   DNase I footprinting of the 1-adrenergic receptor promoter with Sp1 and Egr-1. The 32P-labeled PstI 484 to +268 1-AR genomic fragment (20,000 cpm) was incubated alone (lanes 2 and 6), with 30 µunits DNase I (lanes 3 and 7), or with 1 µl of either Sp1 (lane 4) or histidine-tagged Egr-1 (lane 5) and subjected to DNase I footprinting as described under Materials and Methods. The localization of the DNase I footprints was determined using Maxam-Gilbert G ladders (lanes 1 and 5). The sequences for protected bands I and II are outlined along the figure.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Effect of transient expression of Sp family members in Schneider D. melanogaster SL2 cells on reporter gene expression by 1-AR-luciferase constructs. A total of 1.5 µg DNA per 35-mm culture plate consisting of 1 µg 1-AR-luciferase expression vector alone or with the indicated amounts of pPacSp1, pPacSp3 or pPac0 with 0.1 µg of pRL-CMV R. reniformis luciferase vector were transiently transfected into D. melanogaster SL2 cells. The expression of the 1-AR construct in response to Sp1 or Sp3 relative to the expression of pPac0 are provided. The data represent the mean ± S.E. for the combined results of four transfections. Each transfection used triplicate samples.

Characterization of Egr-1 Binding to the Early Promoter of the Rat ₁-AR Gene. Another transcription factor that binds to G+C rich elements is the early growth response gene 1 (Egr-1) factor (Sukhatme et al., 1988; Rauscher et al., 1990), also known as Zif268, NGF1-A, krox24, and TIS8. Egr-1 is a zinc-finger transcription factor that binds to a consensus sequence GCGGGGGCG (Lim et al., 1989). We observed that the sequence between 385 and 365 contained overlapping binding sites for Sp1 and Egr-1 (Fig. 1). In this region, the Sp1-footprint is between nucleotides 377 and 365, and the putative Egr-1 binding motif is between 380 and 372 (Figs. 1 and 4). Each site is imperfect by one nucleotide from the consensus Sp1 and Egr-1 binding sites.

View larger version (54K): [in this window] [in a new window]   Fig. 4.   Electromobility shift assay between nuclear extracts prepared from phorbol ester-treated ventricular myocytes and wild-type or point mutants of the sequence between 385 and 365. A, nuclear extracts were prepared from rat neonatal ventricular myocytes that were cultured in serum-free medium for 2 days and then exposed either to DMSO or 50 ng/ml PMA for 1 h. Nuclear extracts (1 µl) were combined with 20,000 cpm of 32P-labeled oligomer representing the sequence between 385/365. Hexa-histidine human Egr-1 purified from Escherichia coli was added in lane 3 to mark the migration of Egr-1. B, the sequences of oligonucleotides representing either the wild-type sequence between 365/365 or those with a 3-bp mutation between 371 and 361 (mut Sp1) or with a 3-bp mutation between 380 and 378 (mut-Egr-1) are shown. C, represents competition between increasing amounts of unlabeled wild-type, mut-Sp1, and mut-Egr-1 oligonucleotides versus the 32P-labeled 385/365 oligomer in binding to nuclear extracts prepared from PMA-treated ventricular myocytes. As indicated above each lane, a 100-fold molar excess (100×), 10-fold molar excess (10×), or equal amounts of unlabeled oligomer (1×) was added to each binding reaction mixture.

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Comparison of the Sp1 and Egr-1 binding region in 5'-flanking region of mammalian 1-AR. A sequence comparison of the Sp1 and Egr-1 binding region in promoters of the human, rat, mouse, monkey, and pig 1-AR genes. The numbers are with respect to the translation initiation site. The accession numbers of the genes are as follows: X75540, rhesus; AF042454, porcine; D00634 and J05561, rat; X69168, human, and mouse (Cohen et al., 1993).

Transcriptional Regulation of the Rat ₁-AR Gene by Transient Expression of Egr-1. The induction of Egr-1 mRNA in neonatal rat ventricular myocytes in response to PMA or other hypertrophic stimuli is transient in nature (Knowlton et al., 1993). As illustrated in Fig. 6A, Egr-1 mRNA was undetectable in neonatal rat ventricular myocytes that were cultured in serum-free medium. Exposing these cells to PMA, maximally induced Egr-1 mRNA within 30 min and this effect was attenuated in 3 h even though the concentration of PMA was sustained through out the incubation period. These data indicated to us that proper kinetics of induction and suppression of the Egr-1 gene are required to appropriately assess its transcriptional regulatory effects.

View larger version (35K): [in this window] [in a new window]   Fig. 6.   Comparison between the kinetics of transient Egr-1 induction in ventricular myocytes by phorbol-12-myristate-13-acetate (PMA) and regulated Egr-1 expression in HeLa cells by the Tet-Off system. A, rat neonatal ventricular myocytes were cultured in DMEM/medium-199 (80:20) for 2 days and then exposed to 50 ng/ml of PMA. B, HeLa S3 cells expressing pTet-Off and human Egr-1 cDNA that were cultured for 2 days in medium containing 0.5% FBS and doxycycline to suppress Egr-1 expression. Then doxycycline was removed and Egr-1 was induced for 90 min, followed by the re-addition of 5 ng/ml doxycycline (arrow) to suppress Egr-1. RNA was prepared before Egr-1 induction, during Egr-1 induction, and 1, 2, 3, and 4 h after Egr-1 was suppressed. RNA were subjected to Northern blotting and probed with 32P-labeled human Egr-1 cDNA probe. C, plasmid DNA (2 µg) consisting of 1.9 µg of 1-AR-luciferase expression vector and 0.1 µg of pRL CMV R. reniformis luciferase vector were transiently transfected into 60-mm plates containing the double-mutant HeLa S3 cells Eg-15 or Et-32. Two vectors were used, 394, 126-Luc with the Egr-1 binding site and 368, 126 Luc without the Egr-1 binding site. After transfection, the cells were cultured overnight in the presence of 5 ng/ml doxycycline to suppress Egr-1 or ETTL. The next day, Egr-1 or ETTL were induced for 90 min, followed by the readdition of 5 ng/ml doxycycline. After 24 h, the cells were lysed for the measurement of luciferase activities. The corrected luciferase activity in light units/10 µg of protein ± S.E. for the combined results of four transfections, each in triplicate were calculated.

View this table: [in this window] [in a new window]   TABLE 1 Effect of PMA on the responsiveness of the wild-type and mutated 1-AR promoter in neonatal ventricular myocytes The wild-type and mutagenized 1-AR promoters indicated below were ligated to luciferase in pGL3basic. A total of 5 µg of DNA per 60-mm culture plate consisting of 4.8 µg of 1-AR-luciferase expression vector with carrier DNA and 0.2 µg of pRL-CMV R. reniformis luciferase vector were transiently transfected into neonatal rat ventricular myocytes. The next day, the cells were washed with phosphate-buffered saline and cultured for 48 h in 80:20 DMEM/medium-199 to suppress Egr-1 expression. PMA (50 ng/ml) was added and its effect on luciferase activity was determined after 24 h. The corrected luciferase activity in light units/10 µg of protein ± S.E. in the presence or absence of PMA were calculated and divided by the luciferase activity in the absence of PMA to determine the percentage change in activity in response to PMA. The data represent the combined results of four transfections where each transfection was in triplicate.

Reciprocal Regulation of ₁-AR mRNA Expression by Sp1 and Egr-1 in SK-N-MC Cells. To test the functional relevance of Sp1 and Egr-1 in regulating the transcriptional activity of the intact ₁-AR gene, we determined the effect of transient expression of these transcription factors on ₁-AR mRNA levels in SK-N-MC cells. SK-N-MC cells are human neuroepithelioma cells that express moderate amounts of cell-surface ₁-AR (Esbenshade et al., 1992; Bahouth et al., 2001). Transient transfection of a CMV-driven Sp1 expression vector increased the levels of ₁-AR mRNA by 235 ± 50% compared with cells that were transfected with the control inactive vectors (Fig. 7). The expression of CMV-Egr-1 on the other hand reduced ₁-AR mRNA levels by 40%. Coexpression of equal amounts of Sp1 and Egr-1 vectors reduced Sp1-mediated induction of ₁-AR mRNA by 60%. Therefore, under these conditions the transcriptional activity of the ₁-AR gene was reciprocally regulated by Sp1 and Egr-1.

View larger version (23K): [in this window] [in a new window]   Fig. 7.   Effect of transient expression of Sp1 and Egr-1 on 1-AR mRNA levels in SK-N-MC cells. SK-N-MC cells in DMEM were transiently transfected for 6 h with mammalian expression vectors harboring pcDNA, Sp1, Egr-1, or its inactive derivative ETTL by the LipofectAMINE method. After 6 h, the medium was aspirated and the cells were cultured in DMEM + 10% FBS for 48 h. RNA was extracted and 1-AR mRNA levels were determined by Northern blotting as described under Materials and Methods. The values of 1-AR mRNA are the means ± S.E.M. of 3 separate determinations. The percentage of 1-AR mRNA in control (ETTL + pCDNA) was 100%; in Sp1, 235% ± 50; in Egr-1 alone, 60% ± 20; in Egr-1 + Sp1, 140% ± 30. **p < 0.001 for 1-AR mRNA in control versus Sp1, *p < 0.01 for 1-AR mRNA in control versus Egr-1.

A Shared Overlapping Site Is Involved in Reciprocal Regulation of ₁-AR Expression by Sp1 and Egr-1. To determine the mechanism by which Sp1 and Egr-1 bind to their individual sites in their common binding region between 380 and 369, the ³²P 385/365 oligonucleotide was incubated with increasing amounts of Egr-1 in the presence of a fixed amount of Sp1 (Fig. 8A). As the concentration of Egr-1 was raised, the amount of Egr-1 complexed to ³²P 385/365 increased, whereas the Sp1 complexed to ³²P 383/365 decreased. In the next set of experiments, we performed EMSA using the ³²P-labeled mut-Egr-1 oligomer to determine whether the Egr-1 binding site is necessary for the competitive interaction of Egr-1 with Sp1 (Fig. 8B). The binding of Sp1 to the ³²P-mut-Egr-1 was not affected by this mutation. However, the ability of Egr-1 to competitively displace Sp1 was severely compromised. These data indicate that Sp1 and Egr-1 do not simultaneously bind to 385/365 and that these two transcription factors compete for binding.

View larger version (97K): [in this window] [in a new window]   Fig. 8.   Competitive displacement of Sp1 binding of the 385 and 365 G+C-rich region in the 1-AR promoter by Egr-1. EMSA was performed on 32P-labeled wild-type 385/365 oligonucleotide (left) or 32P-mut-Egr-1-385/365 oligonucleotide (right), in which the Egr-1 binding site was mutated. The oligomers were incubate with a constant amount of Sp1 (0.05 footprinting unit/incubation) and increasing amounts of (His)6-Egr-1 fusion protein (2-50 ng/lane) for 20 min at room temperature. The resulting complexes were the resolved by electrophoresis and visualized by autoradiography.

Functional Relevance of the Egr-1 Binding Site in the ₁-AR Promoter. To determine whether reciprocal regulation of the ₁-AR gene described earlier was caused by the interaction of Sp1 or Egr-1 with their respective binding sites, the nonoverlapping Sp1 and Egr-1 binding sites were mutated by site directed mutagenesis. Two rat genomic 1-AR constructs were mutagenized, a long ₁-AR promoter construct extending from 3311 and 126 and a shorter version extending from 494 and 126. To disrupt Egr-1 binding, the sequences between 380/378 that are involved in Egr-1 binding were mutated. To disrupt Sp1 binding, the sequences between 371/369 that are involved in Sp1 binding were mutated. In addition we disrupted the Sp1 and Egr-1 binding sites simultaneously by mutagenizing the GCG sequence in the core Sp1/Egr-1 binding region between 374/372 (Table 2). Transient transfection of the long ₁-AR promoter constructs into ventricular myocytes indicated that mutating the Egr-1 site between 380 and 378 did not inhibit the activity of the ₁-AR promoter (Table 2). However, mutating the Sp1-binding site between 371 and 369 or the core sequence between 374 and 372, caused in both instances a 38% drop in the activity of these constructs. Therefore, the Sp1-binding site seems to be the dominant site in stimulating the expression of the ₁-AR promoter from this region.

View this table: [in this window] [in a new window]   TABLE 2 Effect of disrupting the Sp1 or Egr-1 binding sites by site-directed mutagenesis on the activity of the 1-AR promoter-luciferase chimera in neonatal rat ventricular myocytes Vectors and carrier DNA (4.8 µg) were transiently transfected into neonatal ventricular myocytes with 0.2 µg pRL-CMV by calcium phosphate precipitation. After 2 days, the activity of firefly luciferase were measured and corrected for minor changes in transfection efficiency. The relative expression of luciferase by the mutated 1-AR promoter constructs relative to the expression of luciferase by the intact 1-AR promoter are provided. The data represent the mean ± S.E. for the combined results of four transfections. Each transfection utilized triplicate samples.

View larger version (13K): [in this window] [in a new window]   Fig. 9.   Effect of transient expression of Sp1 in D. melanogaster SL2 cells on reporter gene expression by mutated 1-AR genomic-luciferase constructs. 5'flanking sequences containing point mutations within the rat 1-AR promoter were subcloned into the luciferase expression vector and transfected into SL2 cells as described in the legend of Fig. 3. The effect of Sp1 on the expression of the wild type 1-AR construct and the constructs in which the Sp1-binding site (494-mut-Sp1), the Egr-1-binding site (494-mut-Egr-1) or the Sp1/Egr-1 binding sites (494-mutSp1/Egr-1) were mutated are provided. The data represent the expression of each vector in response to Sp1 relative to its expression in response to pPacO. The data represent the mean ± S. E. for the combined results of three transfections. Each transfection used triplicate samples.

View larger version (23K): [in this window] [in a new window]   Fig. 10.   Effect of transient expression of Egr-1 or ETTL on the expression of mutated 1-AR genomic-luciferase constructs in HeLa cells. A total of 2 µg DNA per 60-mm culture plate consisting of 1.9 µg of the 1-AR-luciferase expression vectors (described in the legend of Fig. 9) and 0.1 µg of pRL CMV renilla luciferase vectors were transiently transfected into the double-mutant HeLa S3 cells Eg-15 or Et-32 cells as described under Materials and Methods. After transfection, Egr-1 or ETTL were induced for 90 min and the cells were processed as described in the legend of Fig. 6. The corrected luciferase activity in light units/10 µg of protein ± S. E. for the combined results of three transfections each were calculated in triplicate.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In this work, we identified the core promoter region of the rat ₁-AR gene and characterized a G+C-rich region that has a consensus Sp1-binding site. An intact Sp1 site was required for the full activity of the ₁-AR promoter in heart cells. The conservation of this binding site across mammalian ₁-AR genes suggests that this element is crucial to the regulated expression of the ₁-AR gene (Fig. 5). Sp1 is a potent activator of a wide variety of G protein-coupled receptor promoters, such as the _1B- and _1D-adrenergic receptor genes (Chen et al., 1997; Arai et al., 1999), the - and µ-opioid receptor genes (Ko et al., 1998; Liu et al., 1999), and the dopamine D₂-receptor gene (Yajima et al., 1998). The migration patterns of nuclear extracts from SH-SY5Y cells bound to a 40-mer fragment containing the Sp1 site of the µ-opioid receptor (Liu et al., 1999) were similar to those shown in Fig. 1B. In both instances, the more slowly migrating bands were supershifted by anti-Sp antibodies, whereas the faster migrating band was not affected. In this study, we inhibited the binding of nuclear factors to the faster migrating band by increasing the concentration of poly(dI-dC) in the EMSA. This result suggests that the binding of nuclear factors to the faster migrating band had a lower affinity than their binding to the more slowly migrating complex.

In the µ-opioid and the D₂ dopamine receptor genes, Sp1 and Sp3 were bound to the Sp-binding sites of these GPCR promoters. The effect of Sp3 on Sp1-mediated activation of transcription of these GPCR genes was promoter dependent. In the µ-opioid receptor gene, Sp3 trans-activated the promoter and its activity was additive with Sp1 (Liu et al., 1999). Whereas in the D₂ dopamine receptor gene, Sp3 alone failed to affect transcription and repressed Sp1-induced trans-activation of transcription. The mechanism by which Sp3 inhibits Sp1-mediated trans-activation of transcription is unknown. In the collagen 2A1 promoter, Sp3 did not influence gene transcription, but concomitant overexpression of Sp1 and Sp3 inhibited the transcription of the collagen 2A1 promoter via Sp3-mediated blockade of Sp1-mediated induction of the collagen 2A promoter activity (Ghayor et al., 2001). These results indicate that the behavior of Sp3 over the ₁-AR promoter resembles its activity over the promoters for the D₂ dopamine receptor and collagen 2A1.

The binding of Sp1 to G+C boxes is often critical to achieving significant levels of transcription from promoters that lack TATA and CCAAT elements, such as the ₁-AR gene (Araki et al., 1991; Boisclair et al., 1993). The activity of the basal promoter of the ₁-AR gene relative to the SV40-driven pGL3control vector was about 60%, indicating that the ₁-AR gene contained a strong minimal promoter. However, the expression of ₁-AR mRNA is very low in tissues and cell lines that endogenously express ₁-AR (Bahouth et al., 1997b, 2001; Searles et al., 1995). Promoter studies have shown that the complete 3.3-kb rat ₁-AR promoter has very low inherent ability to drive transcription compared with the minimal promoter (Searles et al., 1995; Bahouth et al., 1997b; Evanko et al., 1998). Therefore, it is likely that the activity of the minimal promoter is modulated by sequence specific enhancer and/or silencer binding proteins that produce low-level as well as tissue-specific expression. Progressive deletions and site-directed mutagenesis of the ₁-AR gene revealed positive and negative regulatory domains flanking both ends of the core promoter between 394 to 330 that played a critical role in regulating the expression of this gene. A strong negative regulatory domain between the sequences from 2870 to 2740 was found in the rat and human ₁-AR promoters (Bahouth et al., 1997b; Evanko et al., 1998). In addition, two hormonally responsive domains in the 5'-flanking region were identified. An element that inhibited the transcription of the ₁-AR gene in response to glucocorticoids was localized in the sequence between 950 and 926 and a cyclic AMP-responsive element that inhibited the transcription of the ₁-AR gene in C6 glioma cells in response to -agonists was localized within the sequences between 1258 and 1250 (Bahouth et al., 1996; Fitzgerald et al., 1996). The mechanism of -agonist-induced repression of ₁-AR gene transcription is interesting. -Agonists rapidly stimulate the expression of the inducible cyclic AMP early repressor which is a member of the cyclic AMP response element modulator family of transcription factors (Fitzgerald et al., 1996). Increased expression of inducible cyclic AMP early repressor and cyclic AMP response element modulator inhibited the expression of ₁-AR-luciferase constructs that contained the cyclic AMP-responsive element.

Kirigiti et al. (2000) and Searles et al. (1995) identified two additional domains within the rat ₁-AR promoter that lie 5' and 3' to the TSS. The first domain situated upstream from the TSS at 389 was necessary for expression and possessed AP-2 like consensus elements which bound the recombinant AP-2 protein (Kirgiti et al., 2000). The domain 3' to the TSS consisted of two clusters between 1 to 159 and 186 to 211 that when deleted increased or decreased transcription, respectively (Searles et al., 1995). In addition, Bahouth et al. (1997b) identified two additional domains that lie 3' to the TSS. One domain was a thyroid hormone responsive element between nucleotides 101 and 117 that mediated transcriptional activation of the ₁-AR gene in response to thyroid hormones and the other was an adjacent domain between nucleotides 118 and 125 that suppressed the expression of ₁-AR-luciferase chimera in a tissue-specific manner (Bahouth et al., 1997a,b). These data suggest that low expression levels of the ₁-AR gene were not primarily caused by a weak promoter but more probably by inhibitory sequences in the promoter of the ₁-AR gene (Machida et al., 1990; Searles et al., 1995; Fitzgerald et al., 1996; Bahouth et al., 1997a,b; Evanko et al., 1998).

Our studies revealed that the protooncogene Egr-1 binds to a sequence that overlaps with the Sp1 binding site. The binding of Egr-1 to its consensus element can either activate or suppress the transcription of associated genes depending on the respective promoter (Cao et al., 1993; Khachigian et al., 1995; Cui et al., 1996; Thottassery et al., 1999). The data in Figs. 4 and 8 show that Sp1 and Egr-1 do not bind simultaneously to their respective binding sites and that elevated amounts of Egr-1 can displace prebound Sp1 from the Sp1-binding site. Functionally, Egr-1-mediated inhibition of basal transcription was dependent on the Egr-1 binding site as well as on the Sp1-binding site.

What is the physiological role of the Egr-1 site in the ₁-AR promoter? As described in Fig. 7, transient expression of Egr-1 alone reduced the transcription of the endogenous ₁-AR gene and reversed Sp1-mediated activation of ₁-AR gene transcription. Therefore, it is conceivable that the Egr-1 site might mediate a novel inhibitory feedback loop between GPCR that induce transient Egr-1 expression and the ₁-AR. Egr-1 is induced in response to activation of phospholipase C-coupled receptors such as ₁-adrenergic receptors (Chien et al., 1991; Knowlton et al., 1993), angiotensin II type₁-receptors (Day et al., 1999), and muscarinic acetyl choline-receptors (von der Kammer et al., 1998). Activation of each of these GPCRs invariably diminishes the density or the response of membranous ₁-AR. For example, ₁-AR and -AR are inversely cross-regulated and sustained activation of myocardial ₁-AR reduces the actions that are mediated via the ₁-AR (Kunos and Ishac, 1987; Molderings and Schummann, 1989; von der Kammer et al., 1998). Similarly, infusion of angiotensin II onto primary cultures of ventricular myocytes produced 38 and 55% decreases in the density of ₁-AR after 3 and 8 days, respectively (Henegar et al., 1998). Regulation of ₁-AR by angiotensin II might be physiologically relevant, because activation of ₁-AR increases the release of renin from the kidney, which ultimately increases the concentration of angiotensin II. In this regard, increased activation of the angiotensin II type₁-receptor by angiotensin II activates protein kinase C that in turn diminishes the density and responsiveness of ₁-AR (Schwartz and Naff, 1997). The antagonism between muscarinic cholinergic and ₁-AR is well documented. In primary cultures of rat neonatal myocytes, the muscarinic receptor agonist carbachol reduced by 30% the number of cell-surface ₁-AR after 20 h of exposure (Paraschos and Karliner, 1994). These data provide evidence for a physiologically relevant inhibitory feedback loop between phospholipase C-coupled receptors and the ₁-AR. Consequently, ₁-AR expression might be transcriptionally repressed through a novel receptor cross talk pathway involving transient expression of Egr-1.

Heterologous induction Egr-1 and its potential role in ₁-AR regulation provides another dimension to ₁-AR regulation in addition to the phenomena described previously for homologous and heterologous GPCR desensitization, phosphorylation, and mRNA destabilization that are intimately involved in long- and short-term receptor regulation.