INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ENDOTHELIN-1 (ET-1), a peptide of 21 amino acid residues, is the most potent vasoconstrictive substance known. Originally isolated from porcine aortic endothelial cells (39), ET-1 is now known to be one of a family of three vasoactive peptides that also includes endothelin-2 (ET-2) and endothelin-3 (ET-3) (16).

Circulating ET-1 is produced by endothelial cells, and, physiologically, the vascular endothelium is the most abundant source of ET-1 in vivo (39). Preproendothelin-1 (preproET-1)-specific mRNA is constitutively expressed in endothelial cells (8, 39). There are also reports of ET-1 production under stimulation by a variety of nonendothelial cells. In contrast to endothelial cells, the constitutive expression of preproET-1 mRNA is low in vascular medial smooth muscle cells (SMC) in vivo. Nevertheless, medial smooth muscle expression of preproET-1 can be induced in vivo under certain conditions, such as air-induced chronic pulmonary hypertension (32). PreproET-1 transcript expression in cultured SMC is rapidly but transiently induced by growth factors such as platelet-derived growth factor-AA and transforming growth factor-, by the vasoconstrictor peptides ANG II and arginine-vasopressin, and by glucocorticoids (13, 25, 26, 40). Therefore, ET-1 expression appears to be inducible rather than constitutive in SMC.

Arterial wall vasoconstriction is a second stage in the mechanism of hemostasis following coagulation after hemorrhage. In some hemorrhagic vascular pathologies, the formation of an extravascular clot is associated with vasospasm, in which endothelin is involved (27). For example, perivascular clot formation associated with ruptured cerebral aneurysms may lead to a powerful, irreversible, and localized vasospasm. This vasospasm is not correlated with the level of diffuse, circulating endothelinemia but can be prevented by an endothelin antagonist (3), suggesting a role for a local, nonendothelial induction of the endothelin gene rather than a role for a diffuse constitutive expression by the endothelium. Thrombin, a procoagulant molecule involved in the early thrombotic stage of hemostasis, is also implicated in vascular cell signaling, including active and localized vasoconstriction (9). Thrombin has been shown to stimulate endothelin release by endothelial cells (24). However, the ability of thrombin to induce endothelin gene expression in nonendothelial cells and particularly in SMC, in which endothelin could play an autocrine function, is less well-known (13).

In the present study, we examined mechanisms that underlie the induction of endothelin expression by thrombin in cultured rat arterial SMC. To investigate these mechanisms, we have examined in parallel the rate of transcription and the stability of preproET-1 mRNA in response to thrombin, and we have investigated the intracellular signaling pathways involved in this induction. Because fibrate-activated peroxisome proliferator-activated receptors (PPARs) can repress gene induction associated with cell activation, we finally tested the ability of fibrate to repress thrombin-induced endothelin gene expression in SMC, as it does in endothelial cells (4).

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cell culture. Rat aortic SMC were isolated as previously described (1). Briefly, rat thoracic aortas were excised and rinsed, and the adventitia was mechanically dissociated from the media. The medial part was sliced into fine rings. Rat SMC were isolated by placing the aortic rings in an enzymatic bath containing collagenase (1,248 UI/ml) and elastase (17.5 UI/ml) (Eurobio) for 60 min at 37°C. The cell suspension was centrifuged, resuspended in culture medium, plated out in 25-cm² plastic culture flasks coated with 0.1% collagen, and incubated at 37°C in a humidified atmosphere (95% air-5% CO₂). Cells were cultured in a medium consisting of DMEM supplemented with 10% FCS (Biomedia), 20 mM HEPES (Boehringer Mannheim), 2 mM L-glutamine, 50 UI/ml penicillin, 50 µg/ml streptomycin, and 0.125 µg amphotericin B (Sigma, St. Louis, MO). The culture medium was removed and replaced with fresh medium three times weekly. At confluence, the cells were harvested for passage with trypsin (trypsin 0.05%-EDTA 0.02%; Boehringer Mannheim). SMC were characterized by immunohistological criteria. Cells stained positively to an -actin SMC antibody (Dako SA, Glostrup, Denmark) (12) and were negative for RECA, a specific rat endothelial cell antibody (7), and for anti-von Willebrand factor antibody (1). At passage 3, cells were cultured in six-well tissue culture plates. After the cells had grown to confluence, they were deprived of serum during 24 h. Human -thrombin (EC 3.4.21.5) (purified as previously described; see Ref. 2); TRAP^42-55, a synthetic 14-amino acid thrombin receptor activator peptide (Neosystem); phorbol 12-myristate 13-acetate (PMA; Sigma); fenofibrate (Fournier); staurosporin (Sigma); bisindolylmaleimide I; chelerythrine; genistein; and PD-98059 (2'-amino-3'-methoxyflavone; Calbiochem) were then added in serum-free medium. Unstimulated cells were considered controls.

Comparative RT-PCR. Total RNA was extracted with Trizol (GIBCO, Belmont, CA) from confluent cultures of SMC. PreproET-1 mRNA expression was quantified in SMC by comparative RT-PCR. For the RT stage, 200 ng of total RNA were primed with 1 µg of oligo(dT) (Pharmacia Biotech) and were reverse transcribed in the presence of Moloney murine leukemia virus reverse transcriptase (GIBCO). The cDNA was amplified by the PCR with the use of specific oligonucleotide primers for rat preproET-1, endothelin-converting enzyme-1 (ECE-1), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The primers were designed on the basis of published rat cDNA sequences for preproET-1 (39), ECE-1 (29), and GAPDH (35) as follows: sense for preproET-1, 5'-TCTCTGCTGTTTGTGGCTTTC-3'; antisense for preproET-1, 5'-TCGGAGTTCTTTGTCTGCTTG-3'; sense for ECE-1, 5'-ACAAGCTCCTTTCTCGACCA-3'; antisense for ECE-1, 5'-TTACCCAGTTCTGGTAGGCC-3'; sense for GAPDH, 5'-GTGAAGGTCGGAGTCAACG-3'; and antisense for GAPDH, 5'-GGTGAAGACGCCAGTGGACTC-3'. Double-stranded cDNAs were synthesized and amplified using 1.25 U Taq polymerase (GIBCO), 20 mM Tris · HCl (pH 8.0), 50 mM KCl, 0.2 mM dNTP, 10 pmol of each of primers, 1.5 mM MgCl₂, and 4 × 10⁵ counts/min (cpm) of ³³P-labeled primer (Isotopchim) in a 25-µl reaction final volume. The amplification was carried out in a DNA thermal cycler (Techne) for 26 cycles for ET-1 (95°C, 30 s; 62°C, 1 min; and 72°C, 1 min). To permit semiquantitative analysis, RT-PCR of the gene GAPDH was used for 24 cycles (95°C, 30 s; 55°C, 1 min; and 72°C, 1 min).

Run-on assay. For each experiment, two confluent 225-cm² plates were treated for 60 min or 2 h with thrombin at the concentration of 4 U/ml, two other untreated plates being used as controls. Cells were washed twice with ice-cold PBS, scraped off, collected by centrifugation, and lysed in 10 mM Tris, pH 8.4, 1.5 mM MgCl₂, and 140 mM NaCl by addition of a 10% Nonidet P-40 solution to a final concentration of 1% (vol/vol). Nuclei (20 × 10⁶) were pelleted and resuspended in 100 µl of buffer (5 mM MgCl₂, 500 mM sorbitol, 2.5% Ficoll, 0.008% spermidine, 50% glycerol, 1 mM dithiothreitol, and 10 mM Tris · HCl, pH 7.5) and frozen immediately in liquid nitrogen until labeling. In vitro transcriptions were carried out in the presence of 20 × 10⁶ isolated nuclei and [-³²P ]CTP, as described by Mak et al. (21). Two plasmids containing fragments of preproET-1 and GAPDH cDNA (10 µg), respectively, were denatured and immobilized on a nylon filter (Hybond-N, Amersham). The empty plasmids (Bluescript KS plasmid and pGEM plasmid) were used as negative controls. Slot blots were then hybridized with equivalent amounts of radioactive RNA (7 × 10⁶ cpm) obtained from control or thrombin-treated cells. After hybridization, the blots were washed and exposed to Kodak Biomax MS film with an intensifying screen at 80°C for 7 days.

Determination of mRNA preproET-1 stability. Cells were preincubated for 2 h in the absence or presence of thrombin (4 U/ml) before incubation with 60 µM 5,6-dichloro-1--D-ribofuranosyl benzimidazole (Sigma), a specific RNA polymerase II inhibitor. RNA was extracted at different time points and assayed by RT-PCR as describe above.

Quantification of ET-1 peptide secretion. ET-1 levels were measured in 100-µl aliquots of culture medium using a commercial ELISA kit (R & D Systems). The ET-1 assay of R&D Systems utilizes a sandwich immunoassay technique for quantification of ET-1 in samples. This involves the simultaneous reaction of ET-1 present in the sample or standard with two antibodies directed against different epitopes of the ET-1 molecule. One antibody is coated onto the surface of the wells of a microtiter plate, and the other is conjugated to the enzyme horseradish peroxidase. Any ET-1 present forms a bridge between the two antibodies. After removal of unbound material by aspiration and washing, the amount of conjugate bound to the well is detected by reaction with a substrate specific for the enzyme, which yields a colored product proportional to the amount of conjugated ET-1. The colored product can be quantified photometrically (450 nm). By analyzing standards of known ET-1 concentration simultaneously with samples and plotting a curve of signal vs. concentration, the concentration of samples can then be determined. The anti-ET-1 antibody that was used showed 100% specificity for ET-1, <1% cross-reactivity to big ET-1, <2% cross-reactivity to sarafotoxin, 45% cross-reactivity to ET-2, and 14% cross-reactivity to ET-3.

Statistics. Results are expressed as means ± SE. Significant differences between means were computed using a two-factor ANOVA. For analysis of the effect of thrombin on preproET-1 mRNA stability, data were analyzed by the comparison of linear regression curves. P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Thrombin induces preproET-1 mRNA expression in rat aortic SMC via the thrombin receptor. Thrombin induced an increase in preproET-1 mRNA levels in SMC after stimulation at concentrations of 1-6 U/ml. Maximal induction occurred at a concentration of 4 U/ml after 2 h of stimulation in SMC (Fig. 1). Time-course experiments after thrombin (4 U/ml) stimulation were performed in SMC. Maximal induction was observed after 80 min of incubation.

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Effect of thrombin on levels of preproendothelin-1 (preproET-1) mRNA in cultured rat aortic smooth muscle cells (SMC). Total cell RNA was extracted and subjected to RT-PCR analysis with 33P-labeled oligonucleotides specific for rat preproET-1 sequence. Results are means ± SE of 3 experiments. ** P < 0.01 and ***P < 0.001 compared with control cells.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Effect of synthetic 14-amino acid thrombin receptor activator peptide (TRAP42-55) on levels of preproET-1 mRNA in cultured rat aortic SMC. Cells were incubated with 4 U/ml thrombin or with 10 µM TRAP for 2 h. Total cell RNA was extracted and subjected to RT-PCR analysis with 33P-labeled oligonucleotides specific for rat preproET-1 sequence. Results are means ± SE of 3 experiments. GAPDH, glyceraldehyde-3-phosphate dehydrogenase. ** P < 0.01 and *** P < 0.001 compared with control cells.

Release of ET-1 peptide by rat SMC. Measurement of ET-1 levels indicates that the induction of preproET-1 mRNA expression by thrombin (4 U/ml) in rat SMC was accompanied by the release of immunoreactive peptide ET-1 into the extracellular medium (F = 100, P < 0.0001) (Fig. 3). The amounts of ET-1 peptide released by stimulated SMC are 8.86 ± 1.61 vs. 4.58 ± 1.19 pg/ml ET-1 for 6 h (P < 0.001), 20.39 ± 2.49 vs. 11.24 ± 0.82 pg/ml ET-1 for 12 h (P < 0.001), and 39.54 ± 7.63 vs. 17.89 ± 2.91 pg/ml ET-1 for 24 h (P < 0.001). Because the ability of thrombin to promote ET-1 peptide secretion can depend on both an increase in preproET-1 transcript expression and the generation of ET-1 from big ET-1, we investigated the presence of the enzyme responsible for endothelin conversion, called ECE-1, by RT-PCR in rat arterial SMC. ECE-1 was indeed expressed in SMC, but thrombin did not affect ECE-1 expression (data not shown). The accumulation of ET-1 peptide did not occur when SMC were exposed to thrombin in the presence of 50 ng/ml cycloheximide, thus implying a de novo synthesis rather than release of preformed, intracellularly stored peptide.

View larger version (11K): [in this window] [in a new window]   Fig. 3.   Secretion of endothelin-1 (ET-1) peptide by rat SMC. Confluent rat SMC were exposed to thrombin (4 U/ml) for times indicated. ET-1 peptide in overlying medium was determined by use of ELISA. Results are means ± SE of 3 experiments performed in triplicate. *** P < 0.001 vs. control medium.

Effects of thrombin on preproET-1 gene transcription and mRNA stability. To determine whether the thrombin-induced stimulation of preproET-1 mRNA expression occurs via transcriptional or posttranscriptional mechanisms, we performed nuclear run-on transcriptional assay (Fig. 4). The rate of preproET-1 gene transcription was increased twofold above baseline levels by thrombin at 1 h of incubation, corresponding to the time-dependent increase in the steady-state levels of preproET-1 mRNA.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Nuclear run-on experiment. Nuclei were isolated from rat aortic SMC (controls and those stimulated by thrombin). In vitro transcription was allowed to resume in presence of [-32P]UTP. Equal amounts of 32P-labeled, in vitro transcribed RNA probes from each group were hybridized to denatured plasmid (10 µg/slot) containing either ET-1 cDNA or GAPDH cDNA or to plasmid lacking cDNA insert as control (pGEM and Bluescript KS). Results are representative of 3 experiments. Autoradiograms were analyzed by densitometry to quantify ET-1 transcription rate (ratio of transcription rate of ET-1 gene to that of GAPDH gene) (n = 3). ** P < 0.01 vs. control.

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Effect of thrombin on stability of preproET-1 mRNA in rat SMC. Cells were preincubated in absence (control; ) or presence (4 U/ml; ) of thrombin for 2 h and incubated for indicated times with 60 µM 5,6-dichloro-1--D-ribofuranosyl benzimidazole (DRB). Turnover of mRNA was measured by RT-PCR after addition of DRB to culture medium. mRNA levels were expressed relative to counts per minute (cpm) at time of addition of DRB. Half-lives (t1/2 values) of mRNAs were calculated from slope of best-fit straight line to a semilogarithmic plot of mRNA vs. time, with assumption of first-order kinetics (t1/2 = 0.693/slope). Intercepts of fitted lines for preproET-1 mRNA were as follows: untreated control cells, 7.23 ± 0.5; cells treated with thrombin (4 U/ml), 8.26 ± 0.05. By comparison of linear regression curves, intercepts between control cells and treated cells were significantly different. Slopes of the fitted lines for preproET-1 mRNA were as follows: untreated control cells, 0.039 ± 0.005; cells treated with thrombin (4 U/ml), 0.035 ± 0.005. By comparison of linear regression curves, slopes between control cells and treated cells were not significantly different.

PreproET-1 gene induction in response to thrombin is independent of protein kinase C. The seven-pass transmembrane PAR-1 has been demonstrated to be linked to the activation of protein kinase C (PKC), through the phospholipase C signaling pathway (31), and PKC activation. PreproET-1 gene induction has been linked to PKC activation (18, 38). Inhibition of PKC with PKC antagonists and activation of PKC by phorbol esters were used to determine the role of PKC in thrombin-mediated preproET-1 gene induction in aortic SMC. Activation with PMA (10 ng/ml) showed a significant increase in preproET-1 mRNA (Fig. 6). Furthermore, desensitization by 48-h pretreatment of SMC with PMA markedly reduced the subsequent ability of SMC to express ET-1 transcripts in response to thrombin. Staurosporin (50 nM) inhibited thrombin-induced preproET-1 gene expression. Inhibition of PKC with the selective inhibitors chelerythrine (10⁶ M) and bisindolylmaleimide I (5 µM) before thrombin stimulation failed to significantly inhibit preproET-1 gene expression. However, bisindolylmaleimide I inhibited the effect of phorbol esters on preproET-1 mRNA expression. Because of the fact that in other systems thrombin induces a response through mitogen-activated protein (MAP) kinase kinase and tyrosine kinase, these pathways were also tested (36). To determine the involvement of MAP kinase kinase activation and protein tyrosine kinase (PTK) activation in preproET-1 gene induction, we selectively inhibited MAP kinase kinase by using PD-98059 (30 µM) and PTK with the use of genistein (100 µM) before thrombin stimulation in SMC. Inhibition of MAP kinase kinase and PTK significantly decreased preproET-1 mRNA expression in thrombin-stimulated SMC to a level not different from controls. The results suggest that MAP kinase kinase and PTK are involved in preproET-1 gene induction after thrombin stimulation in SMC.

View larger version (24K): [in this window] [in a new window]   Fig. 6.   RT-PCR semiquantitative analysis of preproET-1 mRNA in rat aortic SMC exposed to thrombin (thr; 4 U/ml for 2 h), phorbol 12-myristate 13-acetate (PMA; 10 ng/ml for 2 h), PMA (10 ng/ml for 48 h before thrombin for 2 h), staurosporin (star; 50 nM; preincubation for 30 min before thrombin for 2 h), chelerythrine (chel; 106 M), bisindolylmaleimide I (GF; 5 µM), PD-98059 (PD; 30 µM), and genistein (genist; 100 µM). Data for inhibitor-treated conditions are not included in figure for clarity, but inhibitors alone did not affect basal preproET-1 mRNA expression. Results are means ± SE of 3 experiments. ctl, Control. ** P < 0.01 and *** P < 0.001 compared with thrombin-stimulated cells.

Addition of PPAR activators prevented preproET-1 gene induction in response to thrombin. PPARs have been reported to be capable of repressing the induction of genes due to cell activation (28). The presence of PPAR has been demonstrated in human aortic SMC (30). Interaction of fibrates with PPAR inhibited the induction of genes due to PMA activation of SMC (5, 11, 30).

View larger version (24K): [in this window] [in a new window]   Fig. 7.   RT-PCR semiquantitative analysis of preproET-1 mRNA in rat aortic SMC exposed to fenofibrate (100 µM for 2 h), thrombin (4U/ml for 2 h), and both (100 µM fenofibrate; preincubation for 30 min before thrombin for 2 h). Results are means ± SE of 3 experiments. *** P < 0.01 compared with control cells.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In the present study, we have shown that thrombin induces preproET-1 mRNA expression, leading to the production of detectable ET-1 peptide levels. Thrombin and TRAP^42-55 induce similar levels of preproET-1 mRNA, thus providing evidence for a receptor-mediated signaling event.

To investigate the mechanisms by which thrombin increases steady-state preproET-1 mRNA expression, we examined the rate of transcription and the stability of preproET-1 mRNA. Nuclear transcription run-on analyses revealed that thrombin induced transcription of the preproET-1 gene to 102% above baseline at 60 min after exposure to thrombin. Regulation of the transcriptional rate has been viewed as an overriding control of gene expression for a variety of genes. Analyses of the 5'-flanking regions of the preproET-1 gene suggest the presence of response elements that are linked to constitutive endothelium-specific transcription of preproET-1 mRNA (19), downregulation with retinoic acid (6), and binding of the fos/jun complex (20). In addition, more recent reports indicate that cis-acting elements, upstream to and distinct from the activator protein-1 and GATA sites, are involved in downregulation of the preproET-1 gene in response to shear stress (22). We can suppose the existence of cis-acting elements involved in the upregulation of preproET-1 gene in response to thrombin. Therefore, the promoter of the endothelin gene is a member of this promoter family, leading to constitutive expression by endothelial cells, but could also be easily induced in nonendothelial cells such as SMC, as recently demonstrated for angiotensin-converting enzyme (10) or cyclooxygenase-2 (COX-2) (30). Endothelin gene expression can also be induced in nonendothelial cells other than SMC such as cardiac myocytes, as recently demonstrated in the passage from compensated hypertrophy to congestive heart failure in rats (17).

Steady-state mRNA levels are also determined by the rate of mRNA degradation. We found that the half-life of preproET-1 mRNA in rat aortic SMC was ~20 min, consistent with previous reports in bovine aortic endothelial cells (14, 15, 23). This short lifespan of preproET-1 mRNA has been attributed to the presence of three AUUUA sequences in the 3'-untranslated region that are thought to mediate selective mRNA degradation (16). In cells treated with thrombin, the slope of preproET-1 mRNA decay was identical to that in non-treated control cells, demonstrating that thrombin did not influence the half-life of ET-1 mRNA. The present findings indicate that in rat arterial SMC, thrombin induces ET-1 expression by increasing the rate of transcription of the preproET-1 mRNA and not by influencing survival rate of ET-1 mRNA.

The specific inhibitors of PKC (chelerythrine and bisindolylmaleimide) failed to suppress thrombin-stimulated preproET-1 mRNA expression. These data suggest that thrombin induces the preproET-1 gene by a PKC-independent pathway in SMC. The MAP kinase kinase inhibitor (PD-98059) and the PTK inhibitor (genistein) blocked thrombin-stimulated preproET-1 mRNA. On the basis of our experiments, we conclude that preproET-1 gene induction by thrombin is mediated via a PKC-independent, PTK-dependent pathway in aortic SMC. These findings are in agreement with those for endothelial cells (34) and with the action of thrombin in another system (36).

In these SMC, we find that PPAR ligands inhibit the preproET-1 gene induction in response to thrombin. This inhibition occurred at fibrate concentrations required for the induction of positive PPAR-response genes (37). It has been recently demonstrated that the activation of vascular SMC could be inhibited by PPAR activators (30). In particular fibrates, by binding PPAR, are able to prevent PMA-induced COX-2 gene expression (30). The extension of these data to another gene inducible by PMA and receptor binding in SMC suggest that this mechanism of action of PPAR activators is a general pathway. This generalization has been recently confirmed in a similar stimulation of endothelin gene expression by thrombin in endothelial cells (4). Nevertheless, the exact action of PPAR activators on the repression of gene induction in SMC needs to be further documented.

In conclusion, we have shown that in aortic SMC, thrombin stimulates ET-1 gene expression, leading to peptide production via an increased transcription rate of the preproET-1 mRNA. We have also demonstrated that the observed induction of preproET-1 mRNA occurs through PAR-1 activation and implicates PTK activation. Therefore, endothelin gene expression can be induced in SMC by extracellular receptor ligands able to activate the cells, such as thrombin. Such a shift of gene expression and peptide release from endothelial cells to an inducible expression and release by SMC, on which endothelin could have an autocrine constrictor effect, could play a physiological role in the vascular stage of hemostasis and a pathological role in the vasospasm associated with internal hemorrhage. This phenomenon could also have some consequences on long-term structural remodeling of the arterial wall (32, 33).