Abstract Previous studies suggested that individual components of the activator protein 1 (AP-1) complex behave in a highly idiosyncratic fashion at the level of the human atrial natriuretic peptide (ANP) gene promoter. ANP gene transcription is activated by c-jun and is generally suppressed by c-fos. In the present study, fra-1, a close relative of the c-fos gene product in terms of its structure and functional activity, behaved like fos in cardiac atriocytes, effecting an approximately 50% reduction in c-jun–activatable expression of a human ANP chloramphenicol acetyltransferase (CAT) reporter. In cardiac ventriculocytes, however, fra-1 effected a synergistic amplification of the c-jun response (a 2.5-fold increase over c-jun alone). In atrial cells, fos-like proteins were not uniformly inhibitory in that a carboxy terminal deletion mutant of c-fos activated a human ANP-CAT reporter in the atriocyte cultures. Finally, using a series of domain-swap mutations in the fos/fra structural sequences, we showed that sequences at both the amino and the carboxy termini are required to realize the full fra-1–dependent stimulatory effect as well as the c-fos–dependent inhibition of ANP gene transcription. These findings suggest considerable heterogeneity in the response of the ANP promoter to different components of the AP-1 complex. Such heterogeneity may serve to broaden the range of biological responses available to this promoter as the cardiac cell attempts to adapt to perturbations in the extracellular environment.
Ventricular expression of the atrial natriuretic peptide (ANP) gene has come to be regarded as one of the earliest and most specific markers of myocardial hypertrophy. As a member of the so-called “embryonic repertoire” (or fetal gene program), the ANP gene is expressed shortly after the activation of the immediate early gene (ie, c-jun, c-fos, egr-1, and c-myc) response and prior to activation of the constitutively expressed genes (eg, cardiac actin and myosin light chain–2) that contribute to increased myocardial mass.1 Previously, we established a linkage between the immediate early gene response and ANP gene expression by demonstrating that transient overexpression of c-jun leads to a significant increase in the activity of a cotransfected human ANP gene promoter.2 A similar, though smaller, increase in promoter activity is effected by jun B, and when used with c-jun, jun B leads to a synergistic activation of transcription (B.K.-M. and D.G.G., unpublished data, 1995). At low levels of expression, c-fos acts synergistically with c-jun in amplifying human ANP gene transcription; however, at higher levels of expression, c-fos effects a significant inhibition of promoter activity. Unlike the activating function, which appears to traffic through a consensus activator protein 1 site located between nucleotides −241 and −235, the inhibitory activity of c-fos is dependent on a sequence between nucleotide −212 and the transcription start site.2
Sharing extensive sequence homology at the nucleotide and amino acid levels, fra-1 and c-fos are members of the same gene family3 ; fra-1 is activated by many of the same stimuli that effect increments in c-fos gene expression,3 4 and it has activity similar to that of c-fos in a number of biological systems.5 6 We sought to determine whether fra-1 would share the relatively unique biphasic activity of the c-fos prototype in the cardiac myocyte. Our findings suggest that the effects of fra-1 are dependent on cellular context. In the atrial myocyte fra-1 mimics the suppressant activity of c-fos, and in the ventricular myocyte it behaves as a typical transcriptional activator.
Restriction and DNA modification enzymes were purchased from Boehringer Mannheim, and [3H]acetyl coenzyme A was purchased from DuPont–New England Nuclear. Other reagents were obtained from standard commercial suppliers.
Fusion plasmids linking the human ANP gene promoter sequence to the bacterial chloramphenicol acetyltransferase (CAT) reporter have been described elsewhere.7 There are 410 bp of the human ANP gene 5′ flanking sequence harbored by −410hANPCAT upstream from the CAT reporter. The fra-1 expression vector (RSV-fra) was provided to us by J. Miner,8 the RSV–c-jun by R. Tjian,9 and the BK28 (FBJ LTR c-fos) by I. Verma.10 Chimeric domain-swap mutations of rat fra-1 and c-fos were constructed8 and provided to us by J. Miner. Each mutant, as well as the wild-type proto-onocogenes, were expressed from the same RSV promoter. The Δfos mutant, harboring a 70–amino acid deletion at its carboxy terminus, was derived from the parent BK28 as described previously.2
Cell Isolation and Transfection
Primary cultures of neonatal rat atrial and ventricular cardiac myocytes were generated by use of established techniques.7 On the day of isolation, cells were mixed with DNA and electroporated (Bio-Rad Gene Pulser) at 250 μF and 280 V at a cell density of 107 cells per 0.4 mL PBS containing 0.1% glucose.2 Additional carrier plasmid (pUC 18) was added to each culture to keep transfected DNA concentration constant at 40 μg. Cells were cultured in DME-H21 medium containing 10% enriched calf serum. Medium was changed 24 hours after transfection and again 24 hours later, at which point 1.5 mmol/L exogenous CaCl2 was added. Cells were harvested and lysed 72 hours after transfection, and CAT activity was measured using the method of Neumann et al.11 CAT activity is expressed as [3H]acetylchloramphenicol-produced counts per 100 μg protein. A mock reaction containing no protein was included in each CAT assay to determine background activity, which was subtracted from each experimental value. Each experiment was carried out at least two to three times with similar results. Representative experiments are presented. Where appropriate, statistical analyses were carried out using one-way ANOVA and the Newman-Keuls test for significance (P<.05). Treatment of animals conformed to institutional guidelines.
We have previously demonstrated that overexpression of c-fos in atrial, as well as ventricular, cardiac myocytes effects an inhibition of both basal and c-jun–activated promoter activity.2 A closely related member of the c-fos family, fra-1, is able to mimic c-fos activity in a variety of systems.3 4 5 6 Like c-fos, fra-1 regulates transcription through heterodimerization with a member of the jun family and subsequent association with TPA response elements in the target gene’s promoter. It cannot dimerize, bind DNA, or enhance transcription in the absence of jun.5 6
In ventricular cardiocytes, transfection of an fra-1 expression vector alone had a minimal effect (approximately 10% inhibition) on −410hANPCAT reporter activity (Fig 1A⇓). However, in contrast to the repression of c-jun–dependent activity seen with c-fos (see below), the c-jun/fra-1 combination specified strong enhancement of human ANP–CAT reporter activity (Fig 1A⇓). Cotransfection with c-jun produced a fivefold stimulation, and in the presence of fra-1, activity increased 16-fold over that with −410hANPCAT alone. Thus, fra-1 had neutral or modestly inhibitory properties when used alone yet displayed a synergistic activation in the presence of c-jun. At the same time, fra-1 did not interact to a significant degree with jun B (Fig 1B⇓).
In atrial myocytes the situation was quite different; fra-1 promoted a significant decrease in c-jun–dependent human ANP gene transcription (Fig 2⇓). Data compiled from several experiments are presented in composite form in Table 1⇓ and suggest that there are fundamental differences in the ways that atrial and ventricular myocytes recognize and respond to this proto-oncogene product.
We conjectured that this discrepant regulatory activity might reflect a property intrinsic to the atrial myocyte that, in some fashion, renders all fos activity inhibitory at the level of the ANP promoter. To address this issue, we used a Δfos mutant that harbors a 70–amino acid deletion at its carboxy terminus. This deletion has proved effective at removing a putative inhibitory domain from this portion of the fos molecule and rendering the residual molecule stimulatory at the level of the ANP promoter in cardiac ventricular myocytes2 (Fig 3A⇓). As shown in Fig 3B⇓, cotransfection with Δfos led to a “dose”-dependent increment in c-jun–activatable human ANP–CAT reporter activity in the atrial myocyte population. This argues against the intrinsic negativity of c-fos and fos-like gene products in the cardiac atriocyte and suggests that c-fos, as well as fra-1, possesses two independent activities: one inhibitory, which appears to dominate in the experimental paradigms used here, and the other stimulatory, which becomes manifest only when the inhibitory activity is suppressed by permutations in the level of proto-oncogene expression,2 cellular context (Figs 1⇑ and 2⇑), or structural mutation2 (Fig 3⇓).
Finally, to characterize further the stimulatory response of fra-1 in ventricular myocytes, we attempted to localize the structural region of the fra-1 protein responsible for the activity. To accomplish this, we used a series of fra-1/c-fos domain swap mutants constructed and characterized previously by Miner and Yamamoto.8 As shown in the schematic depicted in Fig 4⇓, the fra-1 and c-fos cDNAs were divided into four domains on the basis of regions of shared functional and structural homology. Domain 2 contains the basic DNA binding region and domain 3 harbors the leucine zipper motif thought to be important for heterodimerization.8 Domains 1 and 4 contain the residual amino terminal and carboxy terminal domains, respectively. The mutant constructs have exchanged one or more structural domains into the heterologous proto-oncogene backbone. As shown in Table 2⇓, each of the chimeric constructions displayed a phenotype intermediate between that of fra-1 (ie, stimulatory) and c-fos (ie, inhibitory). Of the group, RRFF, which harbors the fra-1 sequence in the amino terminal and DNA binding domains, gave the highest level of expression in the presence of c-jun (50% increase in activity), while c-jun+FRRR and c-jun+FRFF were functionally neutral with regard to reporter activity.
The present work demonstrates that fra-1, like other members of the extended c-jun/c-fos family, has important effects at the level of ANP gene transcription. As does c-fos, it suppresses transcription in the cardiac atriocyte; however, in the ventricular myocyte it potently activates human ANP gene transcription in the presence of c-jun while c-fos retains its predominantly inhibitory profile.2 Such behavior could reflect the weaker fos-like properties of fra-1.5 In the atriocyte, which is exquisitely sensitive to c-fos activity,12 even reduced agonist activity might prove sufficient to support the observed inhibitory activity. Retention of this negative regulatory activity would not necessarily extend to the ventriculocyte, which appears to be less sensitive, in general, to the effects of c-fos.2 12 If the intrinsic inhibitory activity of fra-1 is insufficient to trigger the fos-like suppression of human ANP promoter activity, one would anticipate that the stimulatory activity, usually seen only with lower levels of c-fos expression, would operate unopposed, resulting in substantial promoter activation.
Differences in intrinsic inhibitory activity could be due to structural differences between the two proteins; c-fos and fra-1 share a high degree of sequence homology in the DNA binding/leucine zipper and extreme carboxy terminal domains, but these domains are separated by regions that possess less sequence conservation. Abate et al13 have localized the inhibitory regions of c-fos to residues 58 through 116 and 270 through 380. The importance of the latter domain for transcriptional suppression has been confirmed by several groups,14 including our own2 (Fig 3⇑). Sequence homology between c-fos and fra-1 across the carboxy terminal 18 amino acids, harboring key serine residues known to be important for inhibitory activity, is on the order of 80% and conserves all but a single serine residue. Sequence homology in the amino terminal region of the two proteins is more limited. It is conceivable that differences in the primary sequence or conformation of this latter domain could account for the differences in their respective capacities for promoter suppression or activation. The chimeric mutation data offer some support for this model. Although the data suggest that most, if not all, of the individual structural domains are required to demonstrate full activating (ie, fra-1) or inhibitory (c-fos) activity, two of the domains (the amino terminal and DNA binding domains) appear to be particularly important. Retention of fra-1 amino terminal sequence RRFF provided the only phenotype that approached that of fra-1 alone, while replacement of the amino terminal domain with c-fos sequence FRFF resulted in a neutral, though not frankly inhibitory, response. Furthermore, substitution of c-fos for fra-1 in the amino terminal domain (FRRR) was sufficient to eliminate the stimulatory activity of fra-1. The fact that FRFF effected a neutral phenotype implies that the DNA binding domain also plays a role in determining the direction of the fos/fra-1 effect. This latter domain has recently been shown to play a similarly important role in establishing the direction of glucocorticoid-regulated transcription of a composite element in the proliferin gene promoter.8
Taken together, these findings suggest variability in the biological effects even among highly homologous members of the same transcription factor gene family. Such variability would be predicted to broaden the repertoire of biological responses available to the ANP gene promoter and thereby render it more sensitive to the environmental cues that serve to regulate it.
This work was supported by grant HL-35753 from the National Institutes of Health.
Chien KR, Knowlton KU, Zhu H, Chien S. Regulation of cardiac gene expression during myocardial growth and hypertrophy: molecular studies of an adaptive physiologic response. FASEB J. 1991;5:3037-3046.
Kovacic-Milivojevic B, Gardner DG. Divergent regulation of the human atrial natriuretic peptide gene by c-jun and c-fos. Mol Cell Biol. 1992;12:292-301.
Cohen DR, Curran T. Fra-1: a serum-inducible, cellular immediate-early gene that encodes a fos-related antigen. Mol Cell Biol. 1988;8:2063-2069.
Cohen DR, Ferreira PCP, Gentz R, Franza BR, Curran T. The product of a fos-related gene, fra-1, binds cooperatively to the AP-1 site with jun: transcription factor AP-1 is comprised of multiple protein complexes. Genes Dev. 1989;3:173-184.
Miner J, Yamamoto KR. The basic region of AP-1 specifies glucocorticoid receptor activity at a composite response element. Genes Dev. 1992;6:2491-2501.
Turner R, Tjian T. Leucine repeats and adjacent DNA binding domain mediate the formation of functional c-fos-c-jun heterodimers. Science. 1989;243:1689-1694.
Neumann JR, Morency CA, Russian KO. A novel rapid assay for chloramphenicol acetyltransferase gene expression. Biotechniques. 1987;5:444-447.
Kovacic-Milivojevic B, Gardner DG. Regulation of the human atrial natriuretic peptide gene in atrial cardiocytes by the transcription factor AP-1. Am J Hypertens. 1993;6:258-263.
Abate C, Luk D, Curran T. Transcriptional regulation of fos and jun in vitro: interaction among multiple activator and regulatory domains. Mol Cell Biol. 1991;11:3624-3632.