Differential regulation of fos and jun gene expression and AP-1 cis-element activity by endothelin isopeptides. Possible implications for mitogenic signaling by endothelin.

Endothelins (ET) are potent vasoconstrictor peptides that also function as mitogens for numerous cell types. Although regulation of second messenger pathways by ET peptides has been extensively investigated, little is known about the pathways of nuclear signaling by which ET controls gene expression. The present experiments investigated whether fos and jun contribute to nuclear signaling and gene regulation by ET isopeptides. ET isopeptides induced a subset of fos and jun mRNAs in mesangial cells, including c-fos, fra-1, c-jun, and JunB. fos and jun mRNAs were induced as members of the immediate-early gene response. Activation of the high affinity ET receptor moderately increased c-fos and fra-1 mRNA, whereas activation of the low affinity receptor markedly induced both fos and jun mRNAs. Thus, different ET receptor subtypes evoke distinct patterns of fos and jun induction. Prominent isopeptide- and cell-specific differences in the magnitude and kinetics of fos and jun expression were observed. Most striking was the marked elevation of c-fos steady-state mRNA and protein by ET-1, as compared with ET-3. In addition, ET-1, but not ET-3, increased transcriptional activity conferred by an AP-1 cis-element and directed collagenase gene expression. These results suggest that differential regulation of fos and jun expression and of AP-1 cis-element activity by ET isopeptides contributes to regulation of gene expression by ET. Furthermore, a role for AP-1 in mitogenic signaling by ET is suggested by the close correlation between AP-1 cis-element activity and cell growth.

exclusively by the vascular endothelium, it is now clear that numerous vascular and nonvascular cells secrete ET. ET secretion has been demonstrated in the central nervous system, lung, kidney, gut, pituitary, and hypothalmus, eye, and amnion (see Refs. 3 and 4 for review). Thus, it is not surprising that ET peptides regulate a wide range of biological actions, including contraction of vascular and nonvascular smooth muscle, neuromodulation and neurotransmission, secretion of biologically active compounds, such as prostaglandins and steroids, and control of cell growth. The pathways of transmembrane signaling that mediate short term actions of ET ( i e . contraction and secretion) have been extensively investigated (3). In contrast, the pathways of nuclear signaling that mediate long term actions of ET, such as mitogenesis, remain unclear.
The finding that ET-1 increases expression of c-fos mRNA levels in glomerular mesangial cells suggests that AP-1 transcription factors might contribute to nuclear signaling by ET peptides (5). AP-1 complexes are prototypes for transcription factors that couple receptor-generated second messengers to long term responses requiring differential regulation of gene expression (6-8). AP-1 consists of a mixture of homo-and heterodimers that bind to cis-acting elements (5'-TGAC/ GTCA-3') in promoter and enhancer regions of numerous cellular and viral genes (9,lO). The AP-1 proteins are encoded by two gene families: (i) fos family genes including c-fos, fosB, AfosB, fra-1, and fra-2; and (ii) the jun family including cjun, JunB, and junD. Dimerization of Fos and Jun proteins occurs through a leucine zipper structural motif that involves parallel association of a-helical domains in a structure similar to a coiled-coil (11,12). Dimerization is essential to align the basic DNA-binding regions and facilitate binding of AP-1 to its cognate cis-element. However, dimerization is tightly regulated and has important consequences for AP-1 activity. For example, Jun:Fos heterodimers are more stable than Jun:Jun dimers, and Fos proteins are unable to form stable heterodimers (13)(14)(15). Unlike the DNA-binding domains, which are highly conserved, the transactivator domains of AP-1 proteins are divergent. Thus, the dimer composition of AP-1 has been postulated to control transcriptional activity by increasing the combinatorial possibilities for protein:protein interactions with the RNA polymerase I1 preinitiation complex and with adjacently bound proteins on promoters and enhancers (7, 16). In at least one case (junB versus c-jun) AP-1 proteins have been reported to have different transcriptional properties (17,18). Expression of c-fos and c-jun mRNA is one of the earliest genomic responses to cell stimulation by phorbol esters, growth factors, and serum (19)(20)(21). Moreover, the distinct pattern of fos and jun expression has been proposed to account in part for the specific phenotypic response to AP-1 cis-Elements distinct stimuli. In PC12 pheochromocytoma cells, for example, mitogens, such as epidermal growth factor and nerve growth factor stimulate c-fos, c-jun, and junB expression, whereas only c-fos and junB are induced by membrane depolarization (22). Evidence from various hematopoietic cells also demonstrates agonist-specific patterns of fos and jun induction (23), but the biological consequences of agonist-specific patterns of fos and jun induction are not clear. We designed a series of experiments to directly test whether fos,jun, and AP-1 cis-elements contribute to nuclear signaling by ET. In the studies reported here, we demonstrate that ET-1 and ET-3 differentially regulate the expression of fos and jun genes. Only ET-1, however, causes a significant increase in transcriptional activity conferred by an AP-1 cis-element. Moreover, the ability of ET isopeptides to increase AP-1 activity correlates closely with their ability to stimulate cell growth. These data are consistent with a role for AP-1 factors in nuclear signaling and transcriptional regulation by ET peptides.

DISCUSSION
Evidence for fosljun and AP-1 cis-Element Activity in Nuclear Signaling by ET Peptides-Although the ability of ET peptides to control short term actions by generating second messenger cascades has been extensively investigated (see Ref. 3 for review), virtually nothing is known about the pathways of nuclear signaling by which ET regulates gene expression. Recent evidence indicates that ET-1 increases expression of genes for several trans-acting factors, including c-fos and c-myc (3). Differential regulation of gene expression by ET-1 is thought to contribute to the long term biological actions of ET, such as mitogenesis and vascular remodeling in disease (3,4,44).
In the present study, we have examined the role of fos and jun expression and AP-1 cis-element activity in nuclear signaling by ET. Several lines of evidence suggest that AP-1 contributes to regulation of gene expression by E T peptides.
First, ET isopeptides and serum induced expression of a subset of fos and jun mRNAs. c-fos, fra-1, c-jun, and junB were inducible in mesangial cells, whereas fosB and junD were never induced. Expression of fos and jun family genes regulates AP-1 activity by controlling the nuclear concentration of AP-1 protein and by directing the subunit composition of AP-1 dimers, which in turn determines in part the affinity for AP-1 cis-elements (6,8). For example, induction of the cfos gene by ET-1 would promote formation of stable c-Fos:Jun heterodimers with higher affinity for the AP-1 site than Jun:Jun homodimers (13)(14)(15). In addition, because AP-1 proteins contain unique trans-activating domains, the subunit composition of AP-1 dimers determines the combinatorial possibilities for gene regulation by protein:protein interactions with members of the RNA polymerase I1 preinitiation complex or with adjacently bound trans-acting factors (8,16). Thus, differential regulation of fos and jun expression is a potential mechanism for regulation of AP-1 activity by ET.
Our data also demonstrate significant cell-and ligandspecific differences in fos and jun induction. Although serum induces fosB mRNA in 3T3 fibroblasts (32), the present experiments show that fosB was not induced by serum, ET-1, * Portions of this paper (including "Experimental Procedures," "Results," Table 1, and Figs. 1-6) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are included in the microfilm edition of the Journal that is available from Waverly Press. or ET-3 in mesangial cells. Previous experiments demonstrate that fosB is not expressed in stimulated hepatoma cells despite the marked induction of c-fos and c-jun (45). Collectively, these data suggest that cell-specific differences in fosB expression exist. Similarly, fra-1 is differentially induced by serum in PC12 and 3T3 cells (30). Ligand-specific differences exist in both the dynamics and pattern of fos and jun induction by ET-1 and ET-3. The enhanced ability of ET-1 to increase both c-fos steady-state mRNA and protein was the most prominent isopeptide-specific difference. Collectively, these data are consistent with the hypothesis that AP-1 complexes consist of a continuously changing population of homo-and heterodimers (6,16) and that stimulation by ET peptides can shift the dynamic equilibrium of AP-1 formation and perhaps affect the transcriptional activity of AP-1. Our present data imply that in quiescent mesangial cells, AP-1 consists mostly of homo-and heterodimers without c-fos. After activation by ET-1, AP-1 complexes probably consist mostly of c-Fos:Jun heterodimers, and the proportion of c-Fos:Jun dimers declines until 4 h when c-Fos protein is no longer present. However, further experiments are necessary to characterize the population of AP-1 dimers following activation of ET receptors.
The dose-response relationship of fos and jun mRNA induction suggests that activation of distinct ET receptor subtypes gives rise to different patterns of AP-1 gene induction. We note that activation of the high affinity receptor stimulated an increase in only c-fos and fra-1 steady-state mRNA. Activation of this receptor in mesangial cells stimulates a ligand-operated Ca2+ channel and does not activate phospholipase C (5,36). Thus, stimulation of c-fos mRNA is likely to occur via the CRE/CaRE element (CAMP and Ca2+ response element) in the c-fos promoter, which is sensitive to increments in cytosolic free [Ca"] (46). Consistent with these results, the c-jun promoter is not known to have a Ca'+sensitive element. On the other hand, activation of the low affinity ET receptor stimulates even greater increases in c-fos and fra-1 while also inducing c-jun and junB. The low affinity ET receptor in mesangial cells activates numerous effectors in mesangial cells, including phospholipase C, protein kinase C, and phospholipase A2 (see Ref. 3). We surmise, therefore, that activation of different signaling cascades by ET receptor subtypes leads to unique patterns of fos and jun induction. As discussed below, these differences might be reflected in the ability of the low affinity, but not the high affinity receptor, to stimulate mitogenesis in mesangial cells.
One of the unexpected results of our experiments is that ET-3 fails to increase transcriptional activity conferred by an AP-1 cis-element, despite increasing mRNA levels for fra-1, c-jun, and junB. A significant increase in CAT expression from the -73/+63 Coll-CAT constructs was observed only with ET-1. Significant induction of the native collagenase gene was also observed only with ET-1. These results not only confirm the regulatory data observed with the -73/+63 Coll-CAT constructs but also demonstrate that ET-1 directs expression of the collagenase gene. Expression of the collagenase gene by ET-1 in mesangial cells might have important implications for the role of ET-1 as a proinflammatory mediator (4,44). Several hypotheses might explain why ET-1, but not ET-3, increases AP-1 activity. First, AP-1 activity in vivo appears to be regulated by poorly characterized accessory proteins (47,48) that might be subject to positive regulation by ET-1 but not ET-3. Post-translational modification by reversible phosphorylation/dephosphorylation is also known to be an important mechanism regulating AP-1 activity (49,50). Perhaps ET-1 stimulates post-translational modification of AP-1 proteins to a greater extent than ET-3. Consistent with this hypothesis, ET-1 is a much stronger stimulus for phospholipase C activation and Ca2+ signaling than ET-3 (3,36). The importance of ligand-specific post-translational regulation of AP-1 is supported by recent studies in U937 monoblastic leukemic cells, where single doses of diacylglycerol, as opposed to phorbol ester, induced c-fos and c-jun mRNA but failed to elevate AP-1 cis-element activity (51). Thus, similar to our results in mesangial cells with ET isopeptides, simple induction of fos and jun mRNA does not always increase AP-1 activity and suggests that ligand-dependent post-transcriptional mechanisms might play an important role. Finally, part of the explanation for isopeptide-specific regulation might lie in the potency of ET-1 to increase c-fos expression. Induction of c-fos gives rise to AP-1 dimers with enhanced stability; moreover, the transactivation domain of c-fos could contribute to unique protein:protein interactions that somehow enhance the initiation rate of RNA polymerase 11. If this interpretation is correct, it points to regulation of c-fos expression as a critical step in nuclear signaling by ET-1. Further experiments are necessary to distinguish between these possible explanations.
Taken together, the data discussed above demonstrate a role for AP-1 in nuclear signaling by ET peptides and support the hypothesis that isopeptide-specific activation via AP-1 cis-elements can differentially regulate transcriptional responses.
Possible Role for AP-I in Mitogenic Signaling by ET-1-AP-1 transcription factors are thought to play an important role in mitogenic signaling. First, v-fos and v-jun are potent transforming oncogenes (42, 52), which suggests that the cellular homologs of these proteins play a fundamental role in normal growth control. Moreover, transient transfection with a variety of oncogenes (i.e. src and H-ras) increases AP-1 transcriptional activity (43,50). AP-1 genes are induced by a wide variety of growth factors and mitogenic stimuli (6,8).
Rapidly cycling cells express higher levels of c-jun mRNA than serum-starved cells (53). In addition, blockade of c-fos expression by antisense c-fos RNA (54, 55) or microinjection of anti-fos antibodies (56) attenuates the mitogenic response of fibroblasts. The finding that ET-1, but not ET-3, stimulates AP-1 cis-element activity in mesangial cells provided us with a useful system to examine the role of AP-1 in mitogenic signaling by ET. We surmised that if AP-1 activation contributes to cell growth, then ET-1 but not ET-3 should stimulate mitogenesis. In fact, ET-1 caused a mitogenic response at concentrations that activate the low affinity ET receptor, whereas ET-3 was largely inactive at equivalent concentrations. ET-3 induced a miminal increase in cell growth only at 100 nM. In addition, the ability of ET-1, ET-3, and serum to increase transcription conferred by an AP-1 element (as measured by the Coll-CAT constructs) correlated with their mitogenic potential. Thus, these results lend further support to a role for AP-1 in mitogenic signaling by ET-1 and point to the importance of isopeptide-specific induction of c-fos by ET-1. Further experiments employing antisense c-fos RNA will be necessary to test the requirement for c-fos in the mitogenic response to ET.
Implications for AP-1 and Nuclear Signaling by ET Peptides-AP-1 transcription factors are implicated in diverse phenotypic responses such as differentiation, development, and cell growth (6,8). In addition, AP-1 induction follows stimulation by diverse ligands that often direct divergent phenotypic changes. These observations have led to an apparent paradox. How would AP-1 induction contribute to a specific genomic response as directed by extracellular ligands? The data presented here suggest that isopeptide-specific dif-ferences in fos and jun gene expression and in AP-1 ciselement activation play an important role in determining the specificity of the mesangial cell response to ET. The finding that fosB is not induced in mesangial cells also suggests that cell-specific differences in the AP-1 response might also confer specificity. However, there are other potential mechanisms by which AP-1 might participate in a specific transcriptional response to ET. For example, AP-1 complexes activated by ET peptides might require interactions with cell-or tissuespecific transcription factors to render a gene inducible. Recent evidence demonstrates that growth factors and phorbol esters phosphorylate histone H3, which is thought to cause specific reorganization of chromatin structure in transcriptionally active genes (57). Perhaps ET-dependent reorganization of chromatin also contributes to transcriptional specificity. Finally, it is well established that the developmental program of a cell renders certain genes transcriptionally inactive. This would argue that the specificity of the transcriptional response to ET would depend in part on the target cell to which ET binds. Thus, rather than viewing AP-1 as the unique mediator of transcriptional responses by ET peptides, we propose that AP-1 plays a general role in the transcriptional response of target genes following activation of ET receptors. However, the exact outcome of the phenotypic response to ET isopeptides most likely depends on other concurrent events that have yet to be identified.
Acknowledgments-We gratefully acknowledge Dr. John Nilson (Department of Pharmacology, Case Western Reserve University) for his stimulating discussions and for sharing his protocol for measurements of CAT activity. We thank Drs. Michael Karin and Angela Yang-Yen (Department of Pharmacology, University of California, San Diego) for their generous gift of the Coll-CAT constructs. We are also grateful to Dr. Tom Curran (Department of Molecular Oncology and Virology, Roche Institute of Molecular Biology) for providing the rat c-fos cDNA clone. We also thank Dr. Ken Walsh (Department of Physiology and Biophysics, Case Western Reserve University) for his critical reading of this manuscript and insightful comments. Arif Nawaz provided excellent technical assistance.

. 9 k b
GAPDH by guest on March 24, 2020