|
|
||||||||
Laboratory of Pneumology, Unit of Respiratory Pharmacology, Leuven, Belgium
CORRESPONDENCE: G.M. Verleden, Dept of Pneumology, UZ Gasthuisberg, 49 Herestraat, Leuven, B-3000, Belgium. Fax: 32 16346803. E-mail: geert.verleden@uz.kuleuven.ac.be
Keywords: cyclic adenosine monophosphate, cyclic adenosine monophosphate-dependent protein kinase, human airway smooth muscle cells, interleukin-1ß
Received: December 3, 2002
Accepted April 15, 2003
W.A. Wuyts is a research fellow of the Fonds voor wetenschappelijk onderzoek Vlaanderen. This study was funded by the Fonds voor wetenschappelijk onderzoek Vlaanderen G 0220.99. G.M. Verleden is a holder of the GSK Chair for pulmonary pharmacology at the Katholieke Universiteit Leuven.
| Abstract |
|---|
|
|
|---|
In this study, the authors investigated whether eotaxin and MCP-1 expression and release in human airway smooth muscle cells could be modulated by an increase in intracellular cyclic adenosine monophosphate (cAMP) concentration. The possible involvement of cAMP-dependent protein kinase A (PKA) was also studied.
Forskolin, a direct stimulator of adenylyl cyclase, decreased the interleukin (IL)-1ß-induced eotaxin and MCP-1 release by 73±8 and 65±6%, respectively. 8Bromo-cAMP, a cAMP analogue, similarly decreased the chemokine production by 58±9 and 63±8% for eotaxin and MCP-1, respectively. Prostaglandin E2, known as an activator of the prostanoid receptors EP2 and EP4, which are positively coupled to adenylyl cyclase, also decreased the IL-1ß-induced eotaxin and MCP-1 production by 57±17 and 53±4%, respectively. H-89, an inhibitor of PKA, was able to inhibit the decrease in eotaxin and MCP-1 protein release induced by forskolin. Using Western-blot analysis, no effect of cAMP was found on the IL-1ß-induced p38 mitogen-activated protein kinase, extracellular signal-related kinase or cJun N-terminal kinase activation.
This study shows that an increase in intracellular cyclic adenosine monophosphate concentration may decrease the interleukin-1ß-induced eotaxin and monocyte chemotactic protein-1 expression and production. This can be inhibited by addition of H-89, an inhibitor of cyclic adenosine monophosphate-dependent protein kinase. No decrease was observed in interleukin-1ß-induced p38 mitogen-activated protein kinase, extracellular signal-related kinase or cJun N-terminal kinase activation. These findings may be important for the further development of new anti-inflammatory drugs.
Asthma is a chronic inflammatory disease characterised by variable airflow obstruction, bronchial hyperresponsiveness and airway inflammation. Mast cells, lymphocytes and eosinophils are the most important inflammatory cells that are found in the airways of untreated asthmatics 1, 2. It is now widely recognised that eosinophils and their toxic derived proteins can cause bronchial mucosal damage in asthmatic airways, which may contribute to the symptoms of asthma 3.
Eosinophils are thought to be attracted by chemokines, such as eotaxin (which mainly attracts eosinophils) and monocyte chemotactic protein (MCP)-1 (which also attracts other types of inflammatory cells). The current authors have previously demonstrated that human airway smooth muscle cells (HASMC) in culture, when stimulated with interleukin (IL)-1ß, can express and release MCP-1, -2 and -3 4. IL-1ß is a potent pro-inflammatory cytokine that has a central role in inflammatory reactions as seen in asthma. These cytokines can be detected in bronchial tissue and in bronchoalveolar lavage fluid from asthmatic subjects 57.
The formation of cyclic adenosine monophosphate (cAMP) is performed by adenylyl cyclase (AC). Forskolin, a direct stimulator of AC increases intracellular cAMP concentration ([cAMP]i). 8Bromo (8Br)-cAMP is a known analogue of cAMP and prostaglandin (PG)E2 acts via the prostanoid receptors EP2 and EP4, which are positively coupled via a stimulatory G-protein to AC. Changes in [cAMP]i are able to modulate the expression of several genes in various cell types, such as Il-1ß-induced IL-6 release in human lung fibroblasts 8, IL-1
-induced IL-6 and IL-8 gene expression and release in human mesangial cells 9, and tumour necrosis factor (TNF)-
-induced MCP-1 release in mesangial cells 10. It is classically believed that cAMP acts through activation of the cAMP-dependent protein kinase A (PKA) 11, but there are some reports demonstrating that cAMP also induces some PKA-independent effects 12.
In a previous study, the present authors have already shown that p38 mitogen-activated protein kinase (MAPK), extracellular signal-related kinase (ERK) and cJun N-terminal kinase (JNK) are involved in the IL-1ß-induced eotaxin and MCP-1 expression and production in HASMC 13.
Therefore, in the present study the authors investigated whether the IL-1ß-induced chemokine release in HASMC could be modulated by agents that increase [cAMP]i and whether this modulation was PKA dependent. The authors also wanted to investigate whether this effect of cAMP was due to a modulation of p38 MAPK, ERK or JNK.
| Materials and methods |
|---|
|
|
|---|
All experiments were carried out between passage three and six. Cells were cultured from four different patients.
Experimental protocol
HASMC were pretreated with several [cAMP]i-increasing agents and 30 min later IL-1ß (10 ng·mL1) was added for another 4 h (ribonucleic acid (RNA)) or 24 h (protein). The mRNA expression and protein release were compared with those measured in HASMC stimulated with IL-1ß alone at the same time intervals.
HASMC were stimulated with 10 µM forskolin (Merck, Leuven, Belgium), a direct stimulator of AC, 1 mM of 8Br-cAMP (Merck), an analogue of cAMP, or 1 µM PGE2 (Sigma-Aldrich, Bornem, Belgium), an activator of the prostanoid receptors EP2 and EP4. In the second part of the study, indifferent experiments, HASMC were stimulated with IL-1ß, forskolin (10 µM) and different concentrations of H-89 (0.1, 0.3 and 3 µM), a specific PKA inhibitor (Calbiochem, Nottingham, UK).
In the third part of the study, HASMC were stimulated with IL-1ß (10 ng·mL1) and forskolin (10 µM) for several time points (5120 min) to investigate the effect of an increase in [cAMP]i on p38 MAPK, ERK or JNK activation.
HASMC stimulated with IL-1ß alone were used as controls and values measured were expressed as a percentage of the values measured in these control cells.
Northern-blot analysis
Total RNA was isolated by guandinium thiocyanated-phenol-chloroform extraction and isopropanol precipitation 14. The RNA samples were subjected to a 1% agarose/formaldehyde gel, containing 20 mM morpholinosulphonic acid, 5 mM sodium acetate and 1 mM EDTA (pH 7.0), and blotted onto a nylon membrane. Hybridisation was performed with a 227 base pair (bp) fragment specific to the human eotaxin complementary (c) deoxyribonucleic acid (DNA), a 170 bp fragment specific to the human MCP-1 cDNA and a 1,200 bp cDNA fragment specific to rat glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA (Clontec, Heidelberg, Germany). cDNA was labelled by a random primer labelling kit using [
-32P]dCTP (3,000 Ci·mM1) (MBI Fermentas, St Leon-Rot, Germany). After prehybridisation for 4 h at 42°C in a buffer containing 50% formamide, 4xstandard sodium citrate (SSC), 50 mM Tris-HCl (pH 7.5), 5xDenhardt's solution, 0.1% sodium dodecyl sulphate (SDS), 5 mM EDTA and 250 µg·mL1 denatured salmon sperm DNA, the blots were hybridised overnight at 42°C with the labelled probes (12x106 cpm·mL1). Following hybridisation, the blots were washed to a high stringency of 0.1xSSC, 0.1% SDS at 55°C before exposure to X-OMAT-S film. After the adequate exposure time, the autoradiographs were developed and analysed by a laser densitometer. The RNA levels were expressed as the ratio of chemokine messenger (m)RNA to GAPDH mRNA.
Measurement of total protein
Total protein was measured by the Bradford method 15 using a commercially available assay (Bio-RAD laboratories GmbH, Munich, Germany).
Measurement of secreted eotaxin and MCP-1 by use of ELISA
Eotaxin and MCP-1 were measured in the supernatant of cultured HASMC. This was carried out with a commercially available sandwich enzyme-linked immunosorbent assay (ELISA) kit (R&D Systems Europe Ltd, Abingdon, UK), as specified by the manufacturer. These ELISA kits are highly specific and there was no significant interference between any of the cytokines being investigated. The lower limit of detection was 5.0 pg·mL1 for both MCP-1 and eotaxin.
The ratio of measured protein to total protein in the same sample was calculated and the levels of protein were depicted as % of IL-1ß-induced protein production.
Immunoblot analysis of p38 MAPK, JNK and ERK
After extraction of cytosolic proteins, the threonine and tyrosine phosphorylation of p38 MAPK, JNK and ERK were analysed by Western-blot analysis, using phospho-p38 MAPK, JNK and ERK polyclonal antibodies, which only react with the phosphorylated form of the MAPKs studied. The assay was performed as described previously 16. Phospho-p38 MAPK, phospho-JNK, phospho-ERK, p38 MAPK, JNK and ERK polyclonal antibodies were diluted as recommended by the manufacturer (New England Biolabs, Beverly, MA, USA).
After the adequate exposure time, the autoradiographs were developed and analysed by a laser densitometer. The levels were expressed as the ratio of phosphorylated to nonphosphorylated MAPK.
Statistics
All data are presented as means±sem. Statistical analysis on absolute values was performed using the Mann-Whitney U and Kruskal-Wallis test.
| Results |
|---|
|
|
|---|
|
PGE2 (1 µM) decreased the eotaxin mRNA expression by 48±11% (p<0.05, n=4). These results were in accordance with the results of protein release. Addition of PGE2 induced a significant decrease in chemokine release. After stimulation for 24 h with IL-1ß and PGE2, the eotaxin release decreased by 57±17% (p<0.05, n=4).
Effect of [cAMP]i-elevating agents on IL-1ß-induced release and expression of MCP-1
These following results are shown in figure 2
. The actual value of secreted MCP-1 was 106±13 ng·mL1; when HASMC were not stimulated, the MCP-1 production was below the detection limit (5 pg·mL1). Forskolin (10 µM) decreased the MCP-1 mRNA expression by 57±8% (p<0.05, n=4) and the protein release by 65±6% (p<0.01, n=4).
|
PGE2 (1 µM) decreased the MCP-1 mRNA expression and protein release by 51±7% (p<0.05, n=4) and 53±4% (p<0.05, n=4), respectively.
Effect of blocking cAMP-dependent protein kinase in the presence of [cAMP]i-increasing agents
The IL-1ß-induced eotaxin and MCP-1 production was significantly decreased (73±8 and 65±6%, respectively) by using forskolin 10 µM in HASMC. When H-89, an inhibitor of PKA activation, was added, a concentration-dependent attenuation of the effect of forskolin on the IL-1ß-induced eotaxin and MCP-1 release was observed with a residual reduction of only 28±7 and 27±7% with 3 µM of H-89 (p<0.01, n=6) (fig. 3
).
|
|
| Discussion |
|---|
|
|
|---|
It is widely known that cAMP is able to regulate the expression of several genes. To investigate whether the IL-1ß-induced eotaxin and MCP-1 release could be modulated by cAMP, HASMC were stimulated with IL-1ß in combination with forskolin, a direct stimulator of AC, or with 8Br-cAMP, an analogue of cAMP, or PGE2, which is known to activate the prostanoid receptors EP2 and EP4. It is known that all these agents increase [cAMP]i 17, 18. This increase in [cAMP]i resulted in a decrease of eotaxin and MCP-1 mRNA expression. The release of eotaxin and MCP-1 protein was also decreased (after 4 h of stimulation). These data are consistent with the findings by Jocks et al. 19, who found that PGE2 reduces the glomerular mRNA expression of MCP-1 in antithymocyte antibody-induced glomerular injury. In HASMC it has been shown by Pang and Knox 20 that an increase in [cAMP]i significantly inhibits TNF-
-induced eotaxin release. Moreover, the same group has shown that salbutamol and forskolin, two [cAMP]i increasing agents, together with corticosteroids produce more inhibition than steroids alone on TNF-
-induced IL-8 release 21. In addition, in human and mouse mesangial cells incubated with TNF-
, it has been shown that an increase in [cAMP]i inhibits TNF-
-induced MCP-1 gene expression 10, 22. Ammit et al. 23 have demonstrated that RANTES (regulated upon activation, normal T-cell expressed and secreted) release is decreased after an increase of [cAMP]i in HASMC. The current results confirm those of Hallsworth et al. 24, demonstrating that cAMP is able to decrease IL-1ß-induced eotaxin production. However, the current study was broader and to the best of the authors' knowledge it is the first time that a decrease in eotaxin and MCP-1 mRNA expression and MCP-1 release has been observed after an increase in [cAMP]i in HASMC stimulated with IL-1ß. Furthermore, the authors have demonstrated that this effect is PKA dependent.
This current study enhances the present knowledge about the role of cAMP in IL-1ß-induced airway inflammation. IL-1ß is a pro-inflammatory cytokine in asthma, which may intensify inflammation by increasing the release of chemokines and, therefore, the attraction of inflammatory cells. At the same time, however, there seems to be a counterbalance, a mechanism that diminishes the inflammatory reaction, by activating cyclo-oxygenase-2 and hence intracellular PGE2 release, resulting in the activation of the prostanoid receptors EP2 and EP4. This increases [cAMP]i 25, which decreases the chemokine release, as is shown in this study. Therefore, an increase in [cAMP]i might be considered as a kind of auto-brake mechanism, which prevents inflammation from being excessive and resulting in manifest damage.
It is classically believed that the effects of cAMP are mediated by its receptor PKA, which can phosphorylate a diverse number of target proteins both in the cytoplasmic and the nuclear compartments 26. Some authors speculate that AC, responsible for the production of cAMP and PKA, is arranged in such a way that cAMP is "channelled" from the AC to PKA by a molecular mechanism 27. Conversely, in a few cell types, some effects of cAMP have been described to be PKA independent. In INS-1, a pancreatic islet beta cell line, it has been shown that inhibition of PKA with H-89 doesnot reverse the forskolin-induced suppression of insulin transcription 12. Bryce et al. 25 showed that elevation of cAMP activity by db-cAMP and forskolin induced significant effects on stimulated peripheral blood mononuclear cell proliferation. These antiproliferative effects persisted even after co-incubation with the selective PKA inhibitiors 28. As a consequence of this, the current authors investigated whether it was possible to block the forskolin-induced decrease in eotaxin and MCP-1 production with H-89, a specific inhibitor of PKA. A concentration-dependent inhibition of the decrease in MCP-1 production was observed following stimulation with forskolin. This suggests that PKA is involved in the decrease in eotaxin and MCP-1 production that is observed after increasing [cAMP]i. Recently, it has been observed in HASMC that treatment with H-89 could prevent the inhibitory effect of salbutamol on release of granulocyte-macrophage colony-stimulating factor, RANTES and eotaxin 24. It has been shown that blocking PKA results in a decrease of the synergistic inhibitory effect of adding salmeterol (which is know to increase [cAMP]i) to fluticasone 21, conversely, it has been shown by the same group that blocking PKA does not alter TNF-
-induced eotaxin release 20. These data show how complex the role of cAMP may be and how it may depend on the cytokines that are used to stimulate the cells.
To the present authors' knowledge, they are the first to describe that blocking PKA can prevent the decrease observed in IL-1ß-induced eotaxin and MCP-1 production in HASMC.
It has been shown previously that there is a modulation of p38 MAPK activation by cAMP in some cell types 29, 30. Ina previous study 13 the present authors showed that p38 MAPK, ERK and JNK are important pathways in IL-1ß-induced eotaxin and MCP-1 expression and production. Therefore, the effect on the IL-1ß-induced activation of these MAPKs of increasing [cAMP]i was investigated. However, when increasing [cAMP]i no effect was seen in IL-1ß-induced p38 MAPK, ERK or JNK activation. The mechanism of the observed cAMP-induced inhibition of eotaxin and MCP-1 expression is not clear and needs further investigation.
In conclusion, it has been demonstrated that the interleukin-1ß-induced protein release and messenger ribonucleic acid expression of eotaxin and monocyte chemotactic protein-1 in human airway smooth muscle cells are modulated by intracellular cyclic adenosine monophosphate levels; an increase in intracellular cyclic adenosine monophosphate decreases the release and messenger ribonucleic acid expression of these chemokines. Cyclic adenosine monophosphate protein kinase seems to be a key-element in the decrease in chemokine expression that is observed after an increase in intracellular cyclic adenosine monophosphate. These findings could support the importance of human airway smooth muscle cells, not only as contractile cells, but also as pro-inflammatory cells that might induce and enhance, and even control the airway inflammation in conditions such as asthma. These findings could be important with regards the future development of new anti-inflammatory drugs.
| Acknowledgements |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
-induced eotaxin release from cultured human airway smooth muscle cells by ß2-agonists and corticosteroids. FASEB J 2001;15:261269.
-induced interleukin-8 release from cultured human airway smooth-muscle cells. Am J Respir Cell Mol Biol 2000;23:7985.
-induced secretion of RANTES and interleukin-6 from human airway smooth-muscle cells. Modulation by cyclic adenosine monophosphate. Am J Respir Cell Mol Biol 2000;23:794802.This article has been cited by other articles:
![]() |
M. A. Punchard, A. O. Pozzi, T. P. de Prada, M. T. Coronado, P. Gonzalez, and P. Fantidis Letter to the Editor: Monocyte cAMP Content Is Decreased in Patients With Stable Angina Arterioscler. Thromb. Vasc. Biol., February 1, 2007; 27(2): 436 - 437. [Full Text] [PDF] |
||||
![]() |
W. Li, T. Wang, C. Ma, T. Xiong, Y. Zhu, and X. Wang Calcitonin gene-related peptide inhibits interleukin-1beta-induced endogenous monocyte chemoattractant protein-1 secretion in type II alveolar epithelial cells Am J Physiol Cell Physiol, September 1, 2006; 291(3): C456 - C465. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Kanefsky, M. Lenburg, and C.-M. Hai Cholinergic Receptor and Cyclic Stretch-Mediated Inflammatory Gene Expression in Intact ASM Am. J. Respir. Cell Mol. Biol., April 1, 2006; 34(4): 417 - 425. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Nie, L. Corbett, A. J. Knox, and L. Pang Differential Regulation of Chemokine Expression by Peroxisome Proliferator-activated Receptor {gamma} Agonists: INTERACTIONS WITH GLUCOCORTICOIDS AND {beta}2-AGONISTS J. Biol. Chem., January 28, 2005; 280(4): 2550 - 2561. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |