Regulation of Tumor Necrosis Factor-α and Interleukin-1β Induced Adhesion Molecule Expression in Human Vascular Smooth Muscle Cells by cAMP
Abstract This study investigates the hypothesis that the elevation of intracellular cAMP may affect cytokine-induced expression of adhesion molecules on human vascular smooth muscle cells. In cultured human smooth muscle cells from coronary arteries and saphenous veins, tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) induced the expression of intercellular adhesion molecule (ICAM-1) and vascular cell adhesion molecule (VCAM-1), whereas interferon-γ (INF-γ) selectively stimulated the expression of ICAM-1. Adenylyl cyclase was stimulated either by the stable prostacyclin mimetic cicaprost or by forskolin. Adhesion molecules were detected by a cell surface enzyme immunoassay and the respective mRNA by reverse transcriptase polymerase chain reaction (rt-PCR). Cicaprost as well as forskolin significantly inhibited TNF-α- and IL-1β-induced cell surface expression of ICAM-1 and VCAM-1. Semiquantitative rt-PCR measurements showed a marked decrease of TNF-α- and IL-1β-induced mRNA levels of both adhesion molecules after preincubation with cicaprost. The stability of TNF-α-induced ICAM-1 and VCAM-1 expression at mRNA and protein level was not altered by cicaprost. The IFN-γ-induced increase of cell surface expression of ICAM-1 and the respective mRNA levels, however, were not significantly altered by elevation of intracellular cAMP. Basal and stimulated cAMP levels, measured by radioimmunoassay, did not differ in TNF-α- and IFNγ-treated cells. The present results demonstrate that the expression of adhesion molecules on human smooth muscle cells induced by cytokines is differentially modulated by activation of adenylyl cyclase.
- Received November 11, 1996.
- Accepted July 22, 1997.
In contrast with differentiated contractile SMCs in the media of an intact vessel, intimal SMCs in atherosclerotic plaques express ICAM-1 and VCAM-1 as shown by immunohistochemical analysis.1,2 The expression of these adhesion molecules can also be detected on intimal SMCs in experimental cardiac allograft vasculopathy.3 The inflammatory infiltration in the vessel wall in both diseases primarily consists of macrophages and lymphocytes. Both inflammatory cell types exhibit markers of activation and are a rich source of cytokines.3-6 Considerable amounts of TNF-α, IL-1β, and IFN-γ have been detected in the atherosclerotic vessel wall.4 These cytokines can influence SMC growth, differentiation, and gene expression.6 There is evidence that SMCs represent not only a target for the mediators released by inflammatory cells, but also play an active role in the development and maintenance of the immune response in the atherosclerotic vessel wall.2,7 In vitro studies showed a costimulation and cell-cycle arrest of preactivated T cells by SMCs.7,8 MHC class II expression on intimal SMCs suggests an active role of SMCs in antigen presentation to T lymphocytes.7,9 Interactions between ICAM-1 and VCAM-1 and their respective integrin ligands play a role in the stimulation of T cells by antigen presentation.10
Previous studies have shown that the proinflammatory cytokines TNF-α, IL-1β, and IFN-γ induce the expression of adhesion molecules on cultured human SMCs from different vascular origins.11–13 The signal transduction pathways involved in the modulation of adhesion molecule expression on vascular SMCs are largely unexplored. cAMP is a prominent regulator of gene expression by modulating signal transduction pathways and different transcription factors, eventually resulting in activation or inhibition of transcription.14 In vascular SMCs, cAMP reduces protooncogene expression15 and stimulates gene expression of nitric oxide synthase and soluble guanylyl cyclase.16,17
In this study, we addressed the question whether cAMP elevation by the stable prostacyclin mimetic cicaprost or by forskolin modulates cytokine effects on the expression of adhesion molecules, namely ICAM-1 and VCAM-1, on human vascular SMCs. In addition, the phosphodiesterase inhibitor IBMX and the cAMP analog db-cAMP were used. Adhesion molecule expression was determined at protein as well as at mRNA levels. We report here that cAMP elevation inhibits TNF-α- and IL-1β-induced expression of ICAM-1 and VCAM-1. IFN-γ-induced expression of ICAM-1, however, was not altered.
Recombinant human TNF-α and IL-1β were obtained from Biomol, and recombinant human IFN-γ from Genzyme. Forskolin, dideoxyforskolin, IBMX, and dithiothreitol were purchased from Sigma Chemical Co. For PCR the following primers were used: H-GAPDH-FF (5′-ATGACAACAGCCT CAAGATCATCAG-3′), H-GAPDH-RF (5′-CTGGTGGTC CAGGGGTCTTACTCCT-3′), ICAM-1-FF (5′-AACCGGA AGGTGTATGAACTG-3′), ICAM-1-RF (5′-CGAGGTGT TCTCAAACAGCTC-3′), VCAM-1-FF (5′-CCAGAATCTA GATATCTTGCTC-3′), and VCAM-1-RF (5′-CAGCCTGT CAAATGGGTA-TAC-3′). Cicaprost was kindly provided by Schering AG.
Segments of human epicardial coronary arteries from explanted hearts were obtained by transplantation due to cardiomyopathy (Deutsches Herzzentrum). Specimens of saphenous veins were obtained during aortocoronary bypass surgery (Klinik für Herzchirurgie, Charité). Vascular SMCs were cultured from the media of the vessels by explant technique. The cells were cultured in Dulbecco’s modified Eagle’s medium with the following supplements: penicillin-streptomycin (48 U/mL-48 mg/mL), l-glutamine (19 mmol/L), sodium pyruvate (9.6 mmol/L; Biochrom), nonessential amino acids (minimal essential medium, 1×; Life Technologies) and 15% fetal calf serum (Life Technologies). Cells were identified as SMCs by specific growth pattern (“hill and valley”) and by detection of smooth muscle α-actin by immunofluorescence using a specific monoclonal antibody (Clone 1A4; Sigma Chemical Co).
For ELISA measurements, cells were grown on 96-well plates (5000 cells/well); for cAMP and mRNA measurements, cells were cultured on 6-well plates (2×105 cells/well). Cells from passages two to six were used for the experiments. Confluent cells were incubated with a medium containing 1% fetal calf serum for 48 hours for synchronization of cell cycle. Subsequently, the cells were treated with the respective test substances.
According to previous studies on the concentration-response relationships and the kinetics of cytokine effects on human coronary SMCs,12 the following concentrations were used: TNF-α (10 ng/mL), IL-1β (10 ng/mL), and IFN-γ (103 U/mL). An incubation period of 16 hours for achieving maximal effects was chosen. Similar concentration-response curves and kinetics of cytokine effects on ICAM-1 and VCAM-1 were observed on SMCs derived from human saphenous veins.
Cicaprost (100 nmol/L), forskolin (10 μmol/L), dideoxyforskolin (10 μmol/L), and db-cAMP (300 μmol/L) were added 15 minutes before the addition of cytokines. IBMX (0.1 mmol/L) was added 10 minutes before the addition of cicaprost.
For evaluation of the stability of ICAM-1 and VCAM-1 proteins, cells were stimulated with TNF-α (10 ng/mL) for 16 hours. Further protein synthesis was inhibited by cycloheximide (10 μg/mL), and cell surface expression of ICAM-1 and VCAM-1 was detected at different time-points from 30 minutes up to 24 hours by cell ELISA.
Detection of Cell Surface Adhesion Molecule Expression
Cell surface expression of ICAM-1 and VCAM-1 was detected by cell ELISA as previously described.12 Briefly, after fixation with glutaraldehyde (0.2%), cells were incubated with monoclonal antibodies against ICAM-1 or VCAM-1 (1:500; Dianova). After incubation with a second biotinylated antibody, streptavidin-biotinylated horseradish peroxidase complex (Amersham) was added. Substrate (orthophenylendiamine) turnover was measured photometrically (492/620 nm) using an ELISA reader (Anthos HT-III). All measurements were made in triplicate. Data were corrected for blank values obtained without use of the first antibody.
Detection of Adhesion Molecules by Immunofluorescence
Cells were cultured on multiwell glass slides until confluency was reached. Adhesion molecules on single cells were detected by indirect immunofluorescence using a confocal laser scanning microscope (Odyssey XL, Noran) as previously described.18 After treatment with ice-cold methanol, cells were incubated with monoclonal antibodies against ICAM-1 or VCAM-1, followed by incubation with a second antibody labeled with the fluorochrome Cy-3. All antibodies were diluted 1:50. Probes were excited with the 529 nm line of an argon-ion laser. Emitted light was detected at wavelengths >550 nm.
Determination of Intracellular cAMP Levels
If not otherwise mentioned, cells were incubated with the nonselective phosphodiesterase inhibitor IBMX (0.5 mmol/L) for 10 minutes at 37°C. If indicated, cells were incubated with cicaprost (100 nmol/L) or forskolin (10 μmol/L) for 15 minutes, followed by the addition of TNF-α (10 ng/mL) or IFN-γ (103 U/mL). After a subsequent incubation period of 15 minutes, the reaction was stopped by addition of ice-cold ethanol (96%). Ethanol was evaporated and Tris-buffered solution (50 mmol/L) was added. The dishes were stored at −70°C. cAMP content was measured by radioimmunoassay using a specific rabbit antibody developed in our laboratory.19 All measurements were performed in duplicate.
Determination of mRNA Levels
Total cellular RNA was prepared from confluent cell monolayers using RNeasyTM Total RNA kit (Qiagen). First-strand cDNA synthesis was performed by reverse transcription. Total RNA (200 ng) was heated at 70°C for 2 minutes and cooled on ice. RNA was added in a final 30-μL reaction mix containing 200 U of Moloney murine leukemia virus reverse transcriptase (Life Technologies), 32 U of recombinant RNase inhibitor rRNAsin (Promega), 0.033 mmol/L of oligo-dT, 0.033 mmol/L each of dNTP (Pharmacia Biotech), 10 mmol/L of dithiothreitol, 50 mmol/L of Tris-HCl (pH 8.3), 75 mmol/L of KCl, and 3 mmol/L of MgCl2 and incubated at 37°C for 30 minutes. cDNA-RNA hybrids were denatured for 2 minutes at 95°C and added to a final 50-μL reaction mixture containing 0.5 mmol/L each of dNTP, 1 U of Thermus aquaticus DNA polymerase (Pharmacia Biotech), 0.04 mmol/L of oligonucleotide primers for the housekeeping gene GAPDH and 0.2 mmol/L of oligonucleotide primers for one adhesion molecule, 10 mmol/L of Tris-HCl (pH 9), 50 mmol/L of KCl, and 1.5 mmol/L of MgCl2. We performed 25 cycles of PCR (Personal Cycler, Biometra) under the following conditions: 95°C for 50 seconds (denaturation), 59°C for 50 seconds (annealing) and 72°C for 1 minute (extension). The PCR products (5 μL) were separated on polyacrylamide gels (5%) at 800 V for 1 hour. Gels were stained with 0.1% silver nitrate. Densitometric quantification of the bands was performed with NIH Image 1.6 software (National Institutes of Health).
For semiquantitative analysis of the effects of cicaprost on mRNA levels of both adhesion molecules, we determined the ratios of the mRNA level of ICAM-1 or VCAM-1 to the signal of the housekeeping gene GAPDH.
For evaluation of the stability of ICAM-1 and VCAM-1 mRNA, cells were stimulated with TNF-α (10 ng/mL) for 4 hours. Further transcription was inhibited by actinomycin D (1 μg/mL). Levels of mRNA were determined at different time-points from 30 minutes up to 24 hours by reverse transcriptase-PCR.
Data are given as mean±SEM of n cell preparations from different individuals. Comparisons between different experimental groups were made by Mann-Whitney U test. P values <.05 were considered statistically significant.
Cicaprost and Forskolin Inhibit TNF-α- and IL-1β-Induced Expression of ICAM-1 and VCAM-1 on Coronary SMCs
There was a low basal expression of ICAM-1 and VCAM-1 on cultured human coronary SMCs. Incubation of the cells with TNF-α (10 ng/mL; 16 hours) elicited a 5-fold increase in basal ICAM-1 and VCAM-1 expression (n=10 to 11; Fig 1⇓). IL-1β (10 ng/mL) induced a 5-fold stimulation of basal ICAM-1 and a 3-fold stimulation of basal VCAM-1 expression (n=10 to 11; Fig 1⇓).
Incubation of coronary SMCs with cicaprost (100 nmol/L) did not alter basal expression of ICAM-1 and VCAM-1. However, cicaprost significantly decreased TNF-α- and IL-1β-induced expression of ICAM-1 (Fig 1⇑). The effects of cicaprost on TNF-α- and IL-1β-stimulated expression of VCAM-1 were even more pronounced (n=6 to 10; Fig 1⇑, a and b).
The inhibitory effects of cicaprost (100 nmol/L) on TNF-α (10 ng/mL; 16 hours)-induced expression of ICAM-1 on single coronary cells visualized by indirect immunofluorescence are exemplified in Fig 2⇓.
Incubation of SMCs with the direct activator of adenylyl cyclase forskolin (10 μmol/L) did not affect basal expression of either ICAM-1 and VCAM-1. After pretreatment with forskolin, TNF-α (10 ng/mL; 16 hours)- and IL-1β (10 ng/mL; 16 hours)-induced expression of ICAM-1 was significantly inhibited (P<.05; n=6 to 10; Fig 1⇑, c and d). Forskolin also markedly decreased TNF-α- and IL-1β-induced expression of VCAM-1 (P<.05; n=6 to 10; Fig 1⇑, c and d). Dideoxyforskolin (10 μmol/L), an inactive analog of forskolin, had no effects on TNF-α- and IL-1β-induced expression of ICAM-1 and VCAM-1 (n=3 to 4, data not shown).
Inhibition of TNF-α- and IL-1β-Induced Expression of Adhesion Molecules on Venous SMCs by cAMP
In order to study whether the observed effects were specific for human coronary SMCs, we also investigated cells from human saphenous veins. Similar to coronary cells, the expression of ICAM-1 and VCAM-1 was induced by TNF-α as well as by IL-1β. Preincubation of the cells with cicaprost significantly inhibited TNF-α (10 ng/mL; 16 hours)-induced expression of ICAM-1 and VCAM-1, by 44±5% and 65±5%, respectively (P<.05, n=6 to 7). IL-1β (10 ng/mL; 16 hours)-induced expression of ICAM-1 and VCAM-1 was inhibited by 46±6% and 65±10%, respectively (P<.05, n=5). In further experiments we investigated the kinetics of the inhibitory effects of cicaprost on the expression of ICAM-1 and VCAM-1. Cells were stimulated with TNF-α (10 ng/mL) for 4 to 48 hours. Cicaprost, given 15 minutes before TNF-α, significantly inhibited TNF-α-induced expression of ICAM-1 and VCAM-1 between 4 and 24 hours (P<.05; n=4; Fig 3⇓, a and b).
Preteatment of the cells with forskolin did not significantly alter basal adhesion molecule expression. However, forskolin significantly inhibited TNF-α- and IL-1β-induced expression of ICAM-1 (P<.05; n=6 to 8) and even more pronounced that of VCAM-1 (P<.05; n=6 to 8) (Table⇓). Pretreatment of venous SMCs with db-cAMP (300 μmol/L) also had no effect on basal expression of ICAM-1 and VCAM-1. However, db-cAMP (300 μmol/L) caused a weak inhibition of TNF-α- and IL-1β-induced expression of ICAM-1. The inhibitory potency of db-cAMP (300 μmol/L) on expression of VCAM-1 was more pronounced, resulting in a significant inhibition of both TNF-α- and IL-1β-induced stimulation of VCAM-1 (P<.05, n=4 to 6; Table⇓).
Effects of Cicaprost on ICAM-1 and VCAM-1 Protein Stability on Human SMCs
After stimulation with TNF-α (10 ng/mL; 16 hours) the ICAM-1 protein on the cell surface was very stable. Forty-eight hours after stimulation with the cytokine and inhibition of further protein synthesis with cycloheximide 84±3% of the photometric signal intensity still could be detected (n=8; Fig 4⇓). In contrast, the stability of VCAM-1 protein was markedly lower. The half-life was approximately 4 hours after stimulation with TNF-α (10 ng/mL, 16 hours) (n=5; Fig 4⇓.). Cicaprost did not alter the stability of ICAM-1 or of VCAM-1 protein (n=5; Fig 4⇓).
Cicaprost and Forskolin Do Not Affect IFN-γ-Induced Expression of ICAM-1 on Human SMCs
Incubation of the cells with IFN-γ (103 U/mL) resulted in 4- and 5-fold increases in basal ICAM-1 expression on SMCs from coronary arteries and saphenous veins, respectively (n=6 to 10; Fig 5⇓). IFN-γ had no significant effects on VCAM-1 expression on either SMC type (n=6). Preincubation of coronary SMCs with cicaprost (100 nmol/L) or forskolin (10 μmol/L) did not modify IFN-γ-induced expression of ICAM-1 (n=5 to 10) (Fig 5⇓). Similar results were obtained for SMCs derived from saphenous veins (n=5; data not shown).
Effects of Phosphodiesterase Inhibition on TNF-α- and IFN-γ-Induced Expression of Adhesion Molecules on Human SMCs
Phosphodiesterase inhibition by IBMX (0.1 mmol/L) alone did not significantly modify TNF-α- and IFN-γ-induced expression of ICAM-1 on SMC from human saphenous veins. Cicaprost significantly inhibited TNF-α-induced expression of ICAM-1, but did not affect IFN-γ-induced expression of ICAM-1 (n=6; Fig 6a⇓). Even the combination of cicaprost and IBMX did not significantly alter the effect of IFN-γ (Fig 6a⇓).
VCAM-1 expression induced by TNF-α was significantly inhibited by IBMX (P<.05; n=6). The combination of cicaprost and IBMX completely prevented VCAM-1 expression induced by TNF-α (n=6; Fig 6b⇑).
Elevation of cAMP Levels by Cicaprost and Forskolin in Human SMCs
To elucidate possible effects of the cytokines on cAMP generation, we investigated cAMP levels induced by cicaprost (100 nmol/L) or forskolin (10 μmol/L), using either TNF-α (10 ng/mL) or IFN-γ (103 U/mL) as a stimulating cytokine. The phosphodiesterase inhibitor IBMX (0.5 mmol/L) was used in these experiments to prevent degradation of cAMP. In the presence of IBMX alone, basal cAMP levels in SMCs cultured from coronary arteries and saphenous veins were low: 1.8±0.7 and 2±0.4 pmol/105 cells, respectively (n=3 to 4). TNF-α (10 ng/mL) and IFN-γ (103 U/mL) did not significantly change these values (n=3 to 4).
Incubation of the coronary cells with cicaprost (100 nmol/L) for 30 minutes significantly increased the cAMP level to 158±35 pmol/105 cells in the presence of TNF-α and 153±29 pmol/105 cells in the presence of IFNγ (n=3 to 4). Incubation with forskolin (10 μmol/L) resulted in an increase of cAMP levels to 41±8 pmol/105 cells in the presence of TNF-α and 46±9 pmol/105 cells in the presence of IFN-γ (n=3 to 4). Similar results were obtained on SMC from saphenous veins. There were no significant differences in cicaprost- and forskolin-induced cAMP levels using TNF-α or IFN-γ as a stimulating cytokine (n=3 to 4).
Cicaprost Decreases TNF-α-Induced mRNA Expression of ICAM-1 and VCAM-1 in Human SMCs
Stimulation of SMCs from coronary arteries as well as saphenous veins with TNF-α (10 ng/mL) caused a significant increase in ICAM-1 and VCAM-1 mRNA levels detected by reverse transcriptase-PCR. Maximal effects were observed between 4 and 16 hours of incubation. Afterward, the mRNA levels decreased, approaching basal values at 48 and 72 hours of incubation. Preincubation of the cells with cicaprost (100 nmol/L) markedly decreased maximal mRNA levels of ICAM-1 and even more pronounced those of VCAM-1. Fig 7⇓ demonstrates the effects of cicaprost on mRNA levels of both adhesion molecules induced by TNF-α at incubation times between 4 and 24 hours in one experiment in venous SMCs.
The effects of cicaprost on mRNA levels of both adhesion molecules on four different cell preparations from saphenous veins were determined semiquantitatively (Fig 8⇓). There was an increase in the basal ICAM-1/GAPDH ratio from 0.4 to 0.9 after stimulation with TNF-α for 4 hours. After stimulation for 24 hours, the signal decreased. Incubation of the cells with cicaprost (100 nmol/L) significantly reduced ICAM-1/GAPDH ratios obtained between 4 and 16 hours of incubation with TNF-α (P<.05; n=4; Fig 8a⇓). The kinetics of mRNA expression of VCAM-1 induced by TNF-α were similar. The VCAM-1/GAPDH ratio was increased from 0.3 to 0.9 at 4 hours of stimulation with TNF-α. Cicaprost markedly inhibited TNF-α-induced mRNA expression of VCAM-1 between 4 and 16 hours of stimulation, resulting in a significant decrease in the VCAM-1/GAPDH ratio (P<.05; n=4; Fig 8b⇓).
IFN-γ-Induced Expression of mRNA Levels of ICAM-1 on SMCs Is Not Affected by Cicaprost
The effects of cicaprost (100 nmol/L)-induced expression of ICAM-1 mRNA were also determined in four different experiments. Basal ICAM-1/GAPDH ratios were 0.3. Incubation of the cells with IFN-γ (103 U/mL) induced an increase up to 0.7 at 4 and 24 hours of stimulation, followed by a slow decrease of the signal. Pretreatment of cells with cicaprost (100 nmol/L) did not significantly change IFN-γ-induced mRNA expression of ICAM-1 (n=4; Fig 9⇓).
Effects of Cicaprost on Stability of ICAM-1 and VCAM-1 mRNA
ICAM-1 and VCAM-1 mRNA after stimulation with TNF-α (10 ng/mL; 4 hours) was stable. Even 24 hours after the inhibition of further transcription by actinomycin D, 86±7% of ICAM-1 and 94±10% of VCAM-1 mRNA expression levels were still present (n=4; data not shown).
The half-life of basal VCAM-1 mRNA was approximately 24 hours (n=4), whereas the half-life of basal ICAM-1 mRNA exceeded 24 hours (n=4; data not shown). Cicaprost (100 nmol/L) did not alter the stability of basal nor stimulated ICAM-1 and VCAM-1 mRNA.
In the present study, we have shown that intracellular cAMP differentially regulates cytokine-induced expression of adhesion molecules on human vascular SMCs. SMCs from coronary arteries and saphenous veins responded similarly to the addition of the different proinflammatory cytokines. TNF-α induced a potent stimulation of ICAM-1 and VCAM-1 expression on cells from both vascular regions. This is in accordance with previous data obtained from SMCs from human aorta and pulmonary arteries.11,12 IL-1β markedly upregulated ICAM-1 expression and stimulated significantly, but less effectively, the expression of VCAM-1. Significant stimulation of ICAM-1 expression by IL-1β has also been reported for SMCs from different vascular regions.11-13,20 Stimulation of VCAM-1 expression by IL-1β has been observed on SMCs from pulmonary arteries12,13,21 but could not be detected on SMCs from the human aorta.11,20 In the present study, IFN-γ selectively induced ICAM-1 expression but did not affect VCAM-1 expression. Failure of IFN-γ to induce VCAM-1 has also been reported on SMCs of human aorta.20 In contrast, potent stimulation of VCAM-1 expression was observed on human SMCs from aorta as well as saphenous veins in other studies.21,22 These discrepancies may be explained by different SMC sources, isolation, and culture procedures as well as different experimental conditions.
The receptor-dependent activation of adenylyl cyclase by cicaprost significantly reduced TNF-α- and IL-1β-induced cell surface expression of ICAM-1 and VCAM-1 on coronary as well as venous SMCs without affecting basal expression of either adhesion molecule. The elevation of intracellular cAMP as the underlying mechanism was confirmed by the actions of the direct activator of adenylyl cyclase forskolin, the cAMP analog db-cAMP, and the phosphodiesterase inhibitor IBMX. The stability of cell surface expression of ICAM-1 and VCAM-1 is quite different: Whereas ICAM-1 protein was very stable, the half-life of VCAM-1 was 4 hours. Cicaprost did not modify the stability of either protein. The observed decrease of surface expression of both adhesion proteins was due to marked inhibition of ICAM-1 and VCAM-1 mRNA levels. There was significant inhibition of ICAM-1 and VCAM-1 mRNA expression by cicaprost between 4 and 16 hours after stimulation with TNF-α. In accordance with the data on cell surface expression of both adhesion proteins, the decrease in mRNA of VCAM-1 was more prominent compared with the effects on ICAM-1. Besides effects on the transcription of both genes, effects on mRNA stability influence the level of mRNA expression. However, we have shown that cicaprost has no effects on basal and TNF-α-stimulated ICAM-1 and VCAM-1 expression.
Inhibition of TNF-α-induced cell surface expression of ICAM-1 and VCAM-1 by cAMP-generating agents has also been demonstrated on human airway SMCs.23 In these cells, db-cAMP and forskolin almost completely prevented TNF-α-induced ICAM-1 expression. The effects on VCAM-1 expression, however, were somewhat lower.23 In endothelial cells, TNF-α- and IL-1β-induced expression of E-selectin and VCAM-1 were inhibited by various cAMP-elevating agents,24,25 whereas ICAM-1 expression induced by TNF-α was not affected by cAMP elevation.24 IL-1β-induced expression of ICAM-1 was even increased by forskolin and IBMX.25 There is evidence for cell-specific regulation of adhesion molecule expression,26,27 which may account for these differences between SMCs and endothelial cells.
In the present study, activation of adenylyl cyclase failed to inhibit ICAM-1 mRNA and protein expression induced by IFN-γ on coronary as well as venous SMCs. There is one report on a glioma cell line demonstrating a biphasic effect of forskolin on IFN-γ-induced ICAM-1 expression, which results in stimulation after 1 hour and inhibition of expression after 24 hours.28 In murine bone marrow macrophages, IFN-γ has been shown to inhibit the activity of adenylyl cyclase in a concentration- and GTP-dependent manner.29 In vascular SMCs, however, we did not find any effects of IFN-γ on basal cAMP levels, and there were no differences between cicaprost- and forskolin-induced cAMP elevation using either TNF-α or IFN-γ as cytokine. Effects on gene expression by cAMP are mediated by the activation of transcriptions factors, such as CREB and CREM.14 However, respective binding sites have not been described in the promoter regions of either the ICAM-1 or the VCAM-1 gene.30 This is in accordance with our data that cAMP elevation does not nonselectively affect cytokine-induced expression of ICAM-1 and VCAM-1. The effects of TNF-α and IL-1β on gene expression are thought to be mediated by activation of the transcription factor NF-κB, whereas the effects of IFN-γ have been attributed to the activation of the JAK/STAT kinase cascade.30 Binding sites for the transcription factor NF-κB have been described in the promoter regions of the human ICAM-1 and VCAM-1 genes cloned from human endothelial cells, as well as other cell types.31-34 Combined mutagenesis and DNA-binding experiments on endothelial and hepatoma cell lines have shown that the TNF-α-induced transcription of ICAM-1 is mediated by a composite C/EBP and NF-κB binding site.32 There is evidence for physical interactions between C/EBPβ, a transcription factor which can be activated cAMP-dependently,35 and NF-κB. This may result in an reduced affinity of NF-κB to its binding site and its stimulatory effect on gene transcription.36 These interactions may contribute to the observed inhibition of ICAM-1 expression. The VCAM-1 promoter region has two tandem binding sites for NF-κB, which both are necessary for full response to TNF-α.30 This common transcription factor may be the target of cAMP. As NO has been shown to affect NF-κB activation in endothelial cells,37 an interference of NO production by cytokine-inducible NO-synthase might also be considered. However, we have not observed an alteration of cytokine-induced expression of adhesion molecules on SMCs by NO-synthase inhibitors (data not shown).
Our findings may be of physiologic relevance, as coculture experiments have shown that the presence of endothelial cells reduces the expression of VCAM-1 on SMC.21 Among other mechanisms, the release of prostacyclin by endothelial cells may modulate the expression of adhesion molecules by a cAMP-dependent mechanism. In addition, the decrease of TNF-α- and IL-1β-induced expression of ICAM-1 and even more pronounced that of VCAM-1 may also be relevant in atherosclerosis. ICAM-1 and VCAM-1 expression on intimal cells has been demonstrated in human atherosclerotic lesions38-41 and correlates with the inflammatory infiltration in the vessel wall.41 The expression of VCAM-1 on SMCs has also been detected after a feeding period of several weeks in the well established model of hypercholesterolemic rabbits.22 In addition to a role as an activation marker, the biologic function of the expression of ICAM-1 and VCAM-1 on SMCs remains to be established. However, by interfering with antigen presentation and influencing the activation of T cells they may contribute to the local inflammatory response in the atherosclerotic vessel wall.2,8-10 We have previously shown that the prostacyclin mimetic cicaprost significantly inhibits the development of atherosclerotic lesions in hypercholesterolemic rabbits.42 In addition to inhibitory effects on platelet and leukocyte activation42 as well as antiproliferative and antimigratory actions on vascular SMCs,43 the inhibition of cytokine-induced adhesion molecule gene expression on SMCs might contribute to an antiatherosclerotic potential of prostacyclin.
Selected Abbreviations and Acronyms
|ELISA||=||enzyme-linked immunosorbent assay|
|ICAM-1||=||intercellular adhesion molecule|
|PCR||=||polymerase chain reaction|
|SMC||=||smooth muscle cell|
|TNF-α||=||tumor necrosis factor-α|
|VCAM-1||=||vascular cell adhesion molecule|
The present study was supported by a grant from the German Federal Ministry for Education and Research (BMBF). The expert technical assistance of Friederike Steinle and Claudia Zernick is gratefully acknowledged.
Libby P, Li H. Vascular cell adhesion molecule-1 and smooth muscle cell activation during atherogenesis. J Clin Invest. 1993;92:538-539.
Ardehali A, Laks H, Drinkwater DC, Ziv E. Drake TA. Vascular cell adhesion molecule-1 is induced on vascular endothelial and medial smooth muscle cells in experimental cardiac allograft vasculopathy. Circulation.. 1995;92:450-456.
Hansson GK. Immune and inflammatory mechanisms in the development of atherosclerosis. Br Heart J. 1993;69(Suppl.):S38-S41.
Ross R. Cellular mechanisms of atherosclerosis. Atheroscler Rev. 1993;25:195-200.
Nilsson J. Cytokines and smooth muscle cells in atherosclerosis. Cardiovasc Res. 1993;27:1184-1190.
Murray AG, Libby P, Pober JS. Human vascular smooth muscle cells poorly co-stimulate and actively inhibit allogenic CD4+ T cell proliferation in vitro. J Immunol. 1995;154:151-161.
van Seventer G, Newman W, Shimizu Y, Nutman TB, Tanaka Y, Horgan KJ, Gopal TV, Ennis E, O’Sullivan D, Grey H, Shaw S. Analysis of T cell stimulation by superantigen plus major histocompatibility complex class II molecules or by CD3 monoclonal antibody: costimulation by purified adhesion ligands VCAM-1, ICAM-1 but not ELAM-1. J Exp Med. 1991;174:901-913.
Couffinhal T, Duplàa C, Moreau C, Lamazière J-MD, Bonnet J. Regulation of vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 in human vascular smooth muscle cells. Circ Res. 1994;74:225-234.
Imai T, Hirata Y, Kanno K, Marumo F. Induction of nitric oxide synthase by cyclic AMP in rat vascular smooth muscle cells. J Clin Invest. 1994;93:543-549.
Papapetropoulos A, Marczin N, Mora G, Milici A, Murad F, Catravas JD. Regulation of vascular smooth muscle soluble guanylate cyclase activity, mRNA, and protein levels by cAMP-elevating agents. Hypertension. 1995;26:696-704.
Pietsch P, Braun M, Felix SB, Baumann G. Visualization of cell adhesion molecules on cardiovascular cells by confocal microscopy. Zool Stud. 1995;34(suppl 1):237-238.
Schröder H, Strobach H, Schrör K. Nitric oxide but not prostacyclin is an autocrine endothelial mediator. Biochem Pharmacol. 1992;533-537.
Yang J, Xu Y, Zhu C, Hagan MK, Lawley T, Offermann MK. Regulation of adhesion molecule expression in Kaposi’s sarcoma cells. J Immunol. 1994;152:361-373.
Gamble JR, Bradley S, Noack L, Vadas MA. TGF-β and endothelial cells inhibit VCAM-1 expression on human vascular smooth muscle cells. Arterioscler Thromb Vasc Biol. 1995;15:949-955.
Panattieri RA, Lazaar AL, Puré E, Albelda SM. Activation of cAMP-dependent pathways in human airway smooth muscle cells inhibits TNF-α-induced ICAM-1 and VCAM-1 expression and T lymphocyte adhesion. J Immunol. 1995;154:2358-2365.
Pober JS, Slowik MR, DeLuca LG, Ritchie AJ. Elevated cyclic AMP inhibits endothelial cell synthesis and expression of TNF-induced endothelial leukocyte adhesion molecule-1, and vascular cell adhesion molecule-1, but not intercellular adhesion molecule-1. J Immunol. 1993;150:5114-5123.
Ghersa P, van Huijsduijnen RH, Whelan J, Cambet J, Pescini R, DeLamarter JF. Inhibition of E-selectin gene transcription through a cAMP-dependent protein kinase pathway. J Biol Chem. 1994;269:29129-29137.
Shu HB, Agranoff AB, Nabel EG, Leung K, Duckett CS, Neish AS, Collins T, Nabel GJ. Differential regulation of vascular cell adhesion molecule 1 gene expression and specific NF-κB subunits in endothelial and epithelial cells. Mol Cell Biol. 1993;13:6283-6289.
Iademarco MF, McQuillan JJ, Dean DC. Vascular cell adhesion molecule 1: contrasting transcriptional control mechanisms in muscle and endothelium. Proc Natl Acad Sci U S A.. 1993;90:3943-3947.
Celada A, Maki RA. IFN-γ induces the expression of the genes for MHC class II A-Aβ and tumor necrosis factor through a protein-kinase-C-independent pathway. J Immunol. 1991;146:114-120.
Collins T, Read MA, Neish AS, Whitley MZ, Thanos D, Maniatis T. Transcriptional regulation of endothelial cell adhesion molecules: NF-κB and cytokine-inducible enhancers. FASEB J. 1995;9:899-909.
Voraberger G, Schäfer R, Stratowa C. Cloning of the human gene for intercellular adhesion molecule 1 and analysis of its 5′-regulatory region. J Immunol. 1991;147:2777-2786.
Hou J, Baichwal V, Cao Z. Regulatory elements and transcription factors controlling basal and cytokine-induced expression of the gene encoding intercellular adhesion molecule 1. Proc. Natl Acad Sci U S A. 1994;91:11641-11645.
Neish AS, Williams AJ, Palmer HJ, Whitley MZ, Collins T. Functional analysis of the human vascular cell adhesion molecule 1 promoter. J Exp Med.. 1992;76:1583-1593.
Iadamarco MF, McQuillan JJ, Rosen GD, Dean DC. Characterization of the promoter for vascular cell adhesion molecule-1 (VCAM-1). J Biol Chem. 1992;267:16323-16329.
Metz R, Ziff E. cAMP stimulates C/EBP-related transcription factor rNFIL-6 to translocate to the nucleus and induce c-fos transcription. Genes Dev. 1991;5:1754-1766.
Stein B, Cogswell PC, Baldwin AS. Functional and physical associations between NF-κB and C/EBP family members: a rel domain-bZIP interaction. Mol Cell Biol. 1993;13:3964-3974.
Peng H-B, Libby P, Liao JK. Induction and stabilization of IκBα by nitric oxide mediates inhibition of NF-κB. J Biol Chem. 1995;270:14214-14219.
O’Brien KD, Allen MD, McDonald TO, Chait A, Harlan JM, Fishbein D, McCarty J, Ferguson M, Hudkins K, Benjamin CD, Lobb R, Alpers CE. Vascular cell adhesion molecule-1 is expressed in human coronary atherosclerotic plaques. J Clin Invest. 1993;92:945-951.
Schrör K. Prostacyclin (prostaglandin I2), and atherosclerosis. In: Rubanyi GM, Dzau V, eds. The Endothelium in Clinical Practice: Source and Target of Novel Concepts and Therapies. New York: Marcel Dekker; 1997:1-44.