INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

THE -ADRENERGIC RECEPTORS (ARs) are members of the seven-transmembrane domain, G protein-coupled receptor family. These include the receptors for a variety of neurotransmitters and hormones as well as light, smell, taste, and chemotactic agents. Members of this gene superfamily are coupled to different intracellular signal transduction mechanisms via interaction with heterotrimeric G proteins. Three AR subtypes, ₁AR, ₂AR, and ₃AR, have been characterized (9, 13, 19). The ₂AR was the first of the ARs cloned and has been the most thoroughly studied (19). The regulatory sequences and the 5' flanking region of the ₂ARs include a conventional TATA box and a number of well-characterized functional elements (5, 6, 10). In contrast, the ₁AR promoter does not contain a TATA box, an initiator element, or other common core promoter elements (36, 42).

We recently identified a novel transcription regulatory element in the ₁AR promoter (46), which includes a homeodomain region, an E-box for myc/max binding, and a glucocorticoid response element (GRE). Coregulation of ₁AR gene expression with glucocorticoids is a novel role for c-myc. In the myocardium, where ₁AR expression is highest, c-myc is expressed throughout the fetal period of cardiac development (20, 41). Several genes are expressed selectively during this period of proliferative cardiac growth including the -isoform of the myosin heavy chain (MHC), the major cardiac contractile protein, and atrial natriuretic factor (ANF) (16, 25). The protooncogenes c-myc, fos, and jun are also expressed during this period of cardiac development.

We undertook these studies to examine temporal changes in cardiac ₁AR gene expression. We used sensitive transcription-rate assays in heart tissue obtained from fetal and postnatal animals. We also examined the transcription rate of c-myc in these tissues. In addition, ₁AR regulation was studied in cell culture models, which allowed experimental manipulation of c-myc expression. Our results demonstrate that ₁AR gene expression is remarkably elevated in the fetal and early neonatal period compared to later life. We also showed in cell culture systems that ₁AR transcription activity was directly correlated with c-myc expression and blocked by disruption of c-myc binding sites in the promoter or by c-myc antisense. Thus the striking elevation in ₁AR transcription rate in late fetal cardiac development is due in part to concomitant changes in c-myc expression.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Nuclear preparation from cardiac tissue. All studies involving the use of animals were conducted in conformity with the "Guiding Principles for Research Involving Animals and Human Beings" of the American Physiological Society. All studies were reviewed in advance by the Institutional Advisory Committee for the Care and Use of Animals. Sprague-Dawley rats, obtained from Charles River Laboratory (Cambridge, MA), were used for cardiac tissue in all experiments. After rats were killed, heart tissue was obtained from proliferative, fetal rats of 19 days gestation (E19) and postproliferative rats of postnatal day 7-15 (P7-15). After administration of maternal anesthesia with pentobarbital (50 mg/kg ip), the fetal animals were delivered by laparotomy. The fetal rats were killed by decapitation and their hearts were removed. The P7-15 animals were anesthetized before death. The hearts were collected and transported in ice-cold PBS. Hearts were processed immediately after collection. Heart tissue was minced into tiny pieces for E19 hearts. For the P7-15 hearts, a single burst of a Polytron was used at setting 4. Tissue was then homogenized in buffer A (15 mM HEPES, 300 mM sucrose, 60 mM KCl, 15 mM NaCl, 2 mM EDTA, 0.5 mM EGTA, 14 mM 2-mercaptoethanol, 0.15 mM spermine, 0.5 mM spermidine, and 1 µl/ml of protease inhibitors). This was followed by centrifugation and removal of the supernatant. The pellet was resuspended in buffer B, which contained 10 µl/ml of Nonidet P-40 (NP-40) in addition to all of the constituents of buffer A. This resuspended pellet was layered onto a high-sucrose buffer (15 mM HEPES, 30% sucrose, 60 mM KCl, 15 mM NaCl, 2 mM EDTA, 0.5 mM EGTA, 14 mM 2-mercaptoethanol, 0.15 mM spermine, and 0.5 mM spermidine). Nuclei were then pelleted by centrifugation at 2,500 rpm in a hanging-bucket centrifuge (model RT6000B, Sorvall). The DNA content of the nuclear preparation was measured by spectrophotometry. The quality of nuclear preparations was assessed by fluorescence microscopy using 4,6-diamidino-2-phenylindole (DAPI) staining. Nuclei were stored for future use at 80°C in glycerol storage buffer (20 mM Tris · HCl, 50% glycerol, 75 mM NaCl, 0.5 mM EDTA, and 0.85 mM DTT).

Nuclear preparation from cell lines. Wild-type (WT) rat fibroblasts and fibroblast cell lines engineered to overexpress c-myc (myc^+/+) were used in transcription rate assays to assess the effects of alterations in c-myc on ₁AR expression. These cells have been extensively used for studies of c-myc expression and cell-cycle regulation (27, 28). Cells were grown to 75-80% confluence in high-glucose DMEM (GIBCO) with addition of 10% fetal bovine serum and penicillin-streptomycin. Cells were scraped after the plates were washed with PBS. Cells were centrifuged at 500 g for 10 min at 4°C. The cell pellet was then triturated in 5 ml of NP-40 lysis buffer (10 mM Tris · HCl, 10 mM NaCl, 3 mM MgCl₂, and 0.5% NP-40). After 5 min on ice, the solution was centrifuged at 300 g for 5 min. The pellet was resuspended and stored in 100-µl aliquots of the glycerol storage buffer. The DNA content of these preparations was measured by spectrophotometry at 280 nm, and the samples were stored at 80°C until transcription rate assays were performed.

Transcription rate assays. These were performed using intact nuclear preparations from the rat hearts and rat fibroblast cell lines. Nylon membranes (Nytran) with individual cDNA preparations immobilized by slot blotting were used for hybridization with in vitro transcription products. We excised 1.5- to 2-kb cDNA fragments of ₁AR (13), c-myc, actin, and GAPDH from the respective vectors. For transcription rate assays that involved heart-tissue preparations, the cDNAs for - and -MHC were also used. These fragments were prepared by PCR cloning according to the description of Sanchez et al. (40). All cDNA fragments were denatured with 1% NaOH for 30 min before slot blotting was performed onto membranes. The DNA was then cross-linked to nylon membrane by exposure to UV light for 3 min. Membranes were stored at 4°C for use in hybridization. For in vitro transcription, nuclear preparations were incubated with NTPs including [-³²P]UTP in transcription buffer (39). The reactions were allowed to progress for 30 min. DNA was then digested with RNase-free DNase, which was followed by extraction of newly formed RNA by TRIzol-chloroform. The resultant RNA was pelleted with isopropranolol and washed with 70% ethanol. These pellets were then dissolved in diethyl pyrocarbonate-water and hybridized to membranes with cDNAs of interest that had been prehybridized with Church buffer (1% BSA, 7% SDS, 0.5 M phosphate buffer, and 1 mM EDTA). After 72 h of hybridization, these membranes were washed in SSC-SDS solution and were then exposed to a phosphor imager. Radioactivity for different genes was expressed in light units per unit area and controlled for background intensity. The amount of radioactive [³²P]UTP incorporated over time is directly proportional to the number of "nascent" transcripts: those in which initiation had already begun in the intact, isolated nuclei (39). This approach has been referred to as nuclear runon, nuclear runoff, and/or transcription rate assays. The relative expression of ₁AR was then normalized to the expression of GAPDH observed in each assay.

Transient transfection experiments. A 2.3-kb fragment of the 5' flanking sequence of the ovine ₁AR region was subcloned into a luciferase reporter vector. This is the same construct used in prior studies (36, 45, 46). A 43-bp nucleotide that contained the glucocorticoid regulatory unit (GRU) (46) and a construct with a mutant E-box (E-box) were cloned into the pGL3C vector. This vector contains the simian virus (SV)-40 promoter and an SV-40 enhancer. These constructs were used for transient transfection experiments. A c-myc antisense expression vector was generated by using PCR to amplify exon II of human c-myc and subcloning it in antisense orientation into the pcDNA3 vector (7). Figure 1 shows a schematic of these constructs. All constructs were verified by direct sequencing. Transfection experiments to examine the effects of alterations in c-myc on ₁AR expression were carried out in myc^+/+ cell lines using Lipofectin reagent (GIBCO). Cells were subcultured into 24-well plates at ~80,000 cells/well and transfected with an empty plasmid vector, the 43-bp GRU construct, or the E-box. All experiments included cotransfection of alternate wells with the empty vector for each construct and were carried out in triplicate. Optimal transfection efficiency was obtained by using 1 µg of DNA with 4 µl of Lipofectin for each well. After transfection, the cells were allowed to express for 18 h in reduced serum medium (Opti-MEM I; GIBCO). The cells were then harvested in lysis buffer (0.1% SDS), and luciferase activity was measured in a luminometer. The effects of the c-myc antisense construct were also examined in myc^+/+ cells. These cells were maintained and transfected as described. In these studies, variable amounts of c-myc antisense and pcDNA3 vector were used to keep the total DNA quantity equal to 1.0 µg of DNA/well in all of the experiments. These cells were cotransfected with equal amounts of the GRU construct. All experiments were conducted in quadruplicate.

View larger version (8K): [in this window] [in a new window]   Fig. 1.   1-Adrenergic receptor (1AR) promoter. A: schematic for the 43-bp glucocorticoid regulatory unit (GRU) construct that includes a homeodomain region (HD), an E-box for myc/max binding, and a glucocorticoid response (GR) element. B: E-box, which is a GRU with a mutated E-box. C: a c-myc antisense construct.

Statistics. All data are shown as means ± SE. Significance of differences between age groups in the data from heart samples or between the different cell lines was compared by ANOVA. We accepted P < 0.05 as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Transcription rate assays. Transcription rate assays were carried out on intact, isolated nuclei from fetal hearts at E19 and from P7-15 animals. A representative blot from E19 and P15 is shown in Fig. 2A. As can be seen, the relative expression levels of both ₁AR and c-myc were substantially higher in E19 hearts compared with postnatal (P15) hearts (P < 0.01). As expected, the fetal -MHC isoform was more highly expressed in the E19 samples, whereas -MHC, the adult isoform, had a higher transcription rate in the P15 animals. GAPDH and -actin were included as "housekeeping genes" to normalize the relative level of gene expression in the different age groups. Similar experiments were carried out on hearts from four individual litters. For postproliferative hearts, experiments were carried out on P7 and P15 hearts. The results from P7 and P15 hearts did not differ and thus were analyzed collectively. The results for ₁AR and c-myc were normalized for GAPDH for each of the individual blots and expressed as the relative transcript level. There was a significantly higher level of both ₁AR and c-myc expression in the hearts of fetal animals compared with postnatal hearts (P < 0.01). Results are depicted in Fig. 2B. There were no significant differences in -actin or GAPDH between the fetal and postnatal hearts (P > 0.1).

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Transcription rate assays performed on fetal rat hearts of 19 days gestation (E19) and postproliferative rat hearts of postnatal day 7-15 (P7-15). A: representative blots from E19 and P15 experiments. B: relative expression of 1AR and c-myc as a ratio to GAPDH, pooled from four different experiments performed on E19 and P7 or P15 hearts. Data are means ± SE. *P < 0.05. MHC, myosin heavy chain.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Transcriptions rate assays performed on wild-type (WT) and myc+/+ rat fibroblast cell lines. A: representative blots from these experiments. B: relative expression of 1AR and c-myc, as a ratio to GAPDH, pooled from four different experiments. *P < 0.05.

Transient transfection studies. These results suggest that alterations in c-myc expression between fetal and postnatal hearts or alterations in myc expression in the two cell lines are closely correlated with corresponding changes in ₁AR expression. To verify the role of c-myc in ₁AR transcription regulation, we examined the activity of various ₁AR promoter constructs after transient transfection into myc^+/+ cells. We compared the activity of the 43-bp ₁AR promoter, which contained an intact GRU with an E-box for myc/max binding, with the activity of a mutant construct in which the E-box had been mutated (E-box). The presence of an intact E-box in the GRU sequence resulted in threefold greater luciferase activity than for cells transfected with the empty pGL3C vector alone (Fig. 4A). By contrast, transfection with the construct that contained the E-box (in which the myc binding site had been mutated) reduced expression to the baseline level observed with the empty vector (P < 0.05).

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Transient transfection assays. A: transfection of myc+/+ cells with the GRU construct and the E-box. Luciferase activity values are shown as means ± SE. Mean is from the control; pGL3C is set at 100%, and results of others are expressed as a percentage of this value. *P < 0.05. B: results after transfection with GRU (control) and cotransfection with varying amounts of c-myc antisense construct. Mean from GRU transfection is set at 100% and antisense transfection results are expressed as a percentage of this value. *P < 0.05.

View larger version (8K): [in this window] [in a new window]   Fig. 5.   Cotransfection of c-myc antisense reduces basal transcription of a 2.3-kb ovine 1AR promoter (control) in cultured rat primary cardiomyocytes. Results are means ± SE. Mean from the control is set at 100% and the results of c-myc antisense cotransfection are expressed as a percentage of this value. *P < 0.05.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

These results demonstrate striking developmental regulation of ₁AR transcription. The ₁AR transcription rate is fivefold higher in hearts of late fetal rats compared with postnatal animals. The high level of ₁AR transcription is associated with a similar elevation in the transcription rate of c-myc and is reduced along with reduced c-myc expression at the later stage of development. In cells that are genetically modified to demonstrate altered c-myc expression, there was a direct correlation between ₁AR and c-myc transcription rates. Mutagenesis of the c-myc binding site or blockade of c-myc expression by antisense blocked ₁AR transcription. We conclude that the developmental regulation of ₁AR is at least in part under the control of c-myc. This mode of regulation occurs at a developmental stage when the cardiac expression of ₁AR is critical to physiological homeostasis. This may also be critical during an important stage of cardiogenesis and regulation of cardiac growth (44).

Cardiomyocytes display two developmentally regulated modes of growth. During fetal and early neonatal life in rats, cardiomyocytes proliferate actively. The ability of cardiomyocytes to divide is subsequently lost as they exit the cell cycle (23). Cardiac growth subsequently occurs only by hypertrophy (29). The exact developmental stage at which mitotic activity is lost has been controversial. It was originally suggested that the change from hyperplasia to hypertrophy was a gradual transition that occurred during the third postnatal week in rats (2, 47). More recent evidence suggests that this occurs after postnatal day 4 or 5 in rats (23). The precise molecular mechanisms that control this switch are not understood.

During cardiac development, the switch from proliferative to hypertrophic growth of cardiomyocytes is temporally associated with a decrease in c-myc mRNA expression (41). Several investigations have shown c-myc to be expressed during the fetal and early neonatal periods (20, 41). One investigation suggested that c-myc expression was highest in E13 fetal heart and decreased continuously throughout the embryonic period (41). In another report, the highest levels of c-myc expression were seen in E12-E18 hearts and were similar at each stage (20). In both reports, c-myc expression decreased in newborn and adult hearts (20, 41). The expression of c-myc can be "reinitiated" in adult hearts following stimuli, which lead to hypertrophy (16, 20). C-myc has been implicated in the control of both proliferation and differentiation in various cell types (26). Transgenic mice that overexpress c-myc in the heart have an increase in cardiac mass that is secondary to an increase in cardiomyocyte number (17). This increase in cardiac myocytes is primarily observed during the fetal period of development. Despite enhanced early proliferative growth of the heart during fetal development in c-myc-overexpressing mice, there is also an accelerated switch to hypertrophic growth in the same animals (24). The mechanism for this precocious switch is not known.

The present results show that the switch from the proliferative to the hypertrophic phase is associated with a change in the transcription rate of ₁AR. Although the ₁AR gene is transcribed at a high rate during the fetal period, this wanes substantially during postnatal life. Our previous studies have identified a myc/max binding sequence in the promoter region of ₁AR (46). Mutations that involve this region abrogate the glucocorticoid-dependent expression of the ₁AR promoter (46). The results shown in Fig. 4A extend these studies in myc^+/+ cells by showing that the increased expression of the ₁AR promoter construct with an intact E-box is abolished when the E-box is mutated. Similarly, coexpression of a c-myc antisense construct abrogates ₁AR promoter activity in the same cells.

The ₁AR is the predominant -adrenergic receptor subtype in the heart and has more important roles in cardiac contractility and other cardiac functions compared with the ₂AR subtype (8, 14, 15). Transgenic mice with disruption of the ₁AR gene have high embryonic lethality (37, 38). We recently demonstrated significant inhibition of cardiomyocyte proliferation by AR blockade (44). Overexpression of ₁AR is characterized by a significant increase in cardiac cell number and cardiac mass (11). The role of AR agonists in the regulation of cardiac mass in other models of pathological growth has been studied extensively. Chronic infusion of subhypertensive doses of norepinephrine induces cardiac hypertrophy (18, 21). Sympathetic activation is observed in heart failure and is likewise associated with cardiac hypertrophy (1, 3, 4, 12). Hypertrophy-inducing stimuli, such as norepinephrine infusion into isolated perfused working rat hearts, induces a transient and sequential increase in the mRNA of c-fos and c-myc (48). Other pathological growth stimuli, such as glucocorticoid administration, result in cardiac hypertrophy, which may or may not be the result of an increase in the ₁AR transcription rate (30). The present results add insight into the unique mechanisms that govern both cardiac growth and catecholamine physiology during development. A high production rate of catecholamines and a high degree of AR ligand occupancy characterize fetal and early neonatal life (34, 35). This is critical to physiological homeostasis in utero. We suggest that the regulation of ₁AR transcription is integrated with the mechanism(s) that control cardiac growth during this developmental period. These mechanisms are important for physiological homeostasis, growth and development, and postnatal adaptation.

Our study does have some limitations. Although we demonstrated striking changes in the transcription rate of both ₁AR and c-myc, we did not examine steady-state mRNA levels or the stability of either transcript. However, other investigators have shown a decrease in c-myc mRNA levels that is synchronous with the switch from proliferative to hypertrophic growth (41). Coupled with the present results, changes in mRNA stability alone are not likely to account for our findings. Our cell culture studies were carried out in both primary cardiomyocytes and in c-myc^+/+-overexpressing cells. Both sets of experiments suggest that ₁AR gene expression is regulated by c-myc, although a direct effect was not demonstrated. An indirect mechanism that involves other proteins is possible. Overexpression studies should be interpreted with caution, because c-myc overexpression may lead to misregulation of genes that are not physiological targets of c-myc (32). However, this approach has also been used by many investigators to begin identification of myc targets (28, 31). Of note, the expression in the c-myc^+/+ cells is only three- to fourfold over that of WT cells (33).

We studied cardiac development only in fetal and newborn rats. Developing mice have a similar pattern of cardiac growth and development of the cardiac sympathetic axis to rats (25). In contrast, cardiac development in sheep and in particular, development of the cardiac sympathetic axis, occurs over a different developmental period (22). It has been suggested that the timing of human cardiac development more closely resembles that of sheep than rodents. Given the species differences in cardiac development, our results may not be directly generalized to other species. Nonetheless, although the timing of development of the cardiac sympathetic axis may differ, it is likely that our results offer important clues to the mechanism of regulation of cardiac growth and development.