IL (interleukin)-6 exerts pro- as well as anti-inflammatory activities. Beside many other activities, IL-6 is the major inducer of acute phase proteins in the liver, acts as a differentiation factor for blood cells, as migration factor for T-cells and is a potent inducer of the chemokine MCP-1 (monocyte chemoattractant protein-1). Recent studies have focused on the negative regulation of IL-6 signal transduction through the IL-6-induced feedback inhibitors SOCS (suppressor of cytokine signalling) 1 and SOCS3 or the protein tyrosine phosphatases SHP-2 (Src homology 2 domain-containing protein tyrosine phosphatase 2) and TcPTP (T-cell protein tyrosine phosphatase). Studies on the cross-talk between pro-inflammatory mediators (IL-1, tumour necrosis factor, lipopolysaccharide) and IL-6 elucidated further regulatory mechanisms. Less is known about the regulation of IL-6 signal transduction by hormone/cytokine signalling through G-protein-coupled receptors. This is particularly surprising since many of these hormones (such as prostaglandins and chemokines) play an important role in inflammatory processes. In the present study, we have investigated the inhibitory activity of PGE1 (prostaglandin E1) on IL-6-induced MCP-1 expression and have elucidated the underlying molecular mechanism. Surprisingly, PGE1 does not affect IL-6-induced STAT (signal transducer and activator of transcription) 3 activation, but does affect ERK (extracellular-signal-regulated kinase) 1/2 activation which is crucial for IL-6-dependent expression of MCP-1. In summary, we have discovered a specific cross-talk between the adenylate cyclase cascade and the IL-6-induced MAPK (mitogen-activated protein kinase) cascade and have investigated its impact on IL-6-dependent gene expression.
- interleukin-6 (IL-6)
- Janus kinase (JAK)
- prostaglandin (PG)
- signal transducer and activator of transcription (STAT)
- signal transduction
The immediate response of an organism towards inflammatory stimuli is the release of pro-inflammatory cytokines such as TNF (tumour necrosis factor) and IL (interleukin)-1. The activity of these cytokines is counteracted by anti-inflammatory cytokines such as IL-4 and IL-10. The members of the family of IL-6-type cytokines exert anti- as well as pro-inflammatory activities. Additionally, chemokines and PGs (prostaglandins) regulate inflammatory processes .
Further complexity results from the fact that the pro-inflammatory cytokine IL-1 induces other cytokines such as IL-6 or initiates the production of PGs by cyclo-oxygenase 2 induction, a key enzyme for the synthesis of PGs and thromboxanes. Furthermore, IL-1 blocks the functions of IL-6 by counteracting the synthesis of a set of acute-phase proteins in the liver . Less is known about the regulation of IL-6 signal transduction by PGs or chemokines.
IL-6 binds to a receptor complex composed of a ubiquitously expressed gp130 receptor chain and the more restrictively expressed IL-6Rα (IL-6 receptor α). Cells which do not express IL-6Rα respond to IL-6 in the presence of agonistically acting sIL-6Rα (soluble IL-6Rα) at the site of inflammation . Activation of the receptor complex by IL-6 leads to the initiation of the JAK/STAT (Janus kinase/signal transducer and activator of transcription), MAPK (mitogen-activated protein kinase) and PI3K (phosphoinositide 3-kinase) cascades (reviewed in [4,5]). The mechanisms leading to STAT1 and STAT3 activation have been studied in detail: after activation of JAK1, 2 and Tyk2 (tyrosine kinase 2), the signal transducer gp130 of the IL-6R complex becomes phosphorylated on tyrosine motifs within the cytoplasmic region. The four most distal membrane phosphotyrosine modules exhibit recruitment sites for STAT3 and to a lesser extend STAT1 [6,7]. Subsequently, these transcription factors are also tyrosine phosphorylated, dimerize, translocate into the nucleus and bind to specific response elements within promoters of IL-6-inducible target genes.
Little is known about the initiation of IL-6-induced STAT-independent signalling pathways. Activation of these pathways is crucial for the cell-type-specific activities of IL-6; it has been shown that IL-6-induced STAT3 activation elicits an anti-apoptotic signal, whereas activation of the MAPK cascade generates a pro-mitotic signal in cells responding with proliferation . In other cells, STAT3 activation is crucial for IL-6-induced differentiation . Finally, IL-6-induced neurite outgrowth depends on the activation of the MAPK cascade . It has been suggested that binding of the adapter protein and protein-tyrosine phosphatase SHP-2 (Src homology 2 domain-containing protein tyrosine phosphatase 2) to the phosphorylated Tyr759 motif in gp130 is the initial step for activating the MAPK cascade in response to IL-6 [8,11]. Later, we identified this motif within gp130 to also be responsible for negative regulation of the JAK/STAT pathway  by recruiting both the phosphatase SHP-2  as well as the IL-6-induced feedback inhibitor SOCS (suppressor of cytokine signalling) 3 . Obviously, Tyr759 does not simply mediate activation of the MAPK cascade but rather regulates the balance between activation of the JAK/STAT and MAPK cascades in response to IL-6 .
The requirement for a tight regulation of IL-6 signal transduction is obligatory. Several mechanisms counteracting IL-6-induced STAT activation and consequently STAT-dependent gene induction have been described. Most importantly, the IL-6-induced feedback inhibitors of the family of SOCS proteins, SOCS1 and SOCS3, inhibit STAT activation by binding to JAKs or to gp130 respectively, as already mentioned above [14,16–20]. The phosphatase activity of SHP-2 is responsible for tyrosine dephosphorylation of JAKs, gp130 and STATs in the cytoplasm , whereas TcPTP (T-cell protein tyrosine phosphatase) dephosphorylates STAT factors in the nucleus . Furthermore, pro-inflammatory mediators such as LPS (lipopolysaccharide), TNF and IL-1β counteract IL-6 signalling by inducing SOCS3 expression or by stabilizing SOCS3 mRNA [22,23], by activating NF-κB (nuclear factor κB), which competes with STAT3 for binding to promoters of IL-6-inducible genes [24,25] or by blocking STAT activation at the receptor, independent of SOCS induction [26,27].
Less is known about the regulation of IL-6 signal transduction by mediators signalling through G-protein-coupled receptors such as chemokines and PGs. The E-type PGs (PGEs) signal through the G-protein-coupled EP receptors which initiate different signalling pathways, dependent on the EP receptor subtype expressed by a specific cell type. Ligand-bound EP1 activates the phospholipase Cβ pathway through Gαq, whereas EP2 and EP4 activate the adenylate cyclase pathway through Gαs. Cells expressing EP3 activate Gαi and inhibit adenylate cyclase activity. Thus the specific outcome of PGE signalling is determined by the receptor type expressed at the cell surface.
PGE1 exerts anti-inflammatory as well as vasodilatory activities. Synthetic analogues of PGE1 ameliorate methotrexate-induced enterocolitis and counteract ROS (reactive oxygen species) production . Recently, PGE1 has been shown to protect against ischaemia/reperfusion-induced liver and lung damage [29,30]. In vivo studies demonstrated that PGE1 in the blood efficiently decreases the levels of the CCL chemokines MCP-1 (monocyte chemoattractant protein 1; CCL2) and is therefore benefical in peripheral arterial obstructive disease . On the other hand IL-6, together with its soluble receptor, is a potent inducer of MCP-1 in fibroblasts . Up to now, no results on a cross-talk between PGE1 and IL-6 signal transduction and its impact on MCP-1 expression are available.
In the present study we describe a specific regulatory function of PGE1 for IL-6 signal transduction. We show a series of initiators of cAMP signalling, including PGE1, that specifically inhibit IL-6-dependent ERK (extracellular-signal-regulated kinase), but not STAT, activation through a PKA (protein kinase A)-dependent pathway which finally inhibits IL-6-induced MCP-1 expression. The present study shows a new and specific mechanism of negative regulation of IL-6-induced MAPK activity and its consequence on IL-6-induced MCP-1 gene expression.
Antibodies to (p)ERK1 and (p)ERK3 (phosphorylated ERK1 and ERK2) as well as to activated STAT3 [(p)Tyr705-STAT3] and the PKA substrate-specific antibody (100G7) were obtained from Cell Signaling Technology. Antibodies to ERK1, ERK2 and SHP-2 were purchased from Santa Cruz Biotechnology. Pertussis toxin, forskolin, PGE1, the EP2 receptor agonist (R)-butaprost, the phosphodiesterase inhibitor IBMX (3-isobutyl-1-methylxanthine), aprotinin, pepstatin and gentamycin were from Sigma–Aldrich. Leupeptin was from MP Biomedical. Pefabloc was purchased from Roth. The PKA inhibitor H89 was from Calbiochem. The Src inhibitor PP1 (protein phosphatase 1) was from Biomol. The PKA and Epac (exchange protein directly activated by cAMP) agonists N-6Phe-cAMP and 8-pCTP2′-O-Me-cAMP were from Axxora. DMEM (Dulbecco's modified Eagle's medium) was from Invitrogen-Gibco and fetal calf serum was from PAA. Recombinant IL-6 and soluble IL-6R were prepared as described previously . The specific activity of IL-6 was 2×106 B cell-stimulatory factor-2 units/mg of protein. Oligonucleotides were synthesized by Eurogenetec.
Western blot analysis
For the isolation of cellular proteins, confluent cell cultures were lysed in 500 μl of lysis buffer [50 mM Tris/HCl (pH 7.5), 150 mM NaCl, 1% Nonidet P40, 1 mM NaF and 1 mM Na3VO4] supplemented with 10 μg/ml each of aprotinin, pepstatin and leupeptin as well as 1 mM pefabloc. Proteins were separated by SDS/PAGE and transferred on to a PVDF membrane (PALL). Antigens were detected by incubation with specific primary antibodies (1:1000) and HRP (horseradish-peroxidase)-coupled secondary antibodies (1:2000; DAKO). The membranes were developed with an ECL (enhanced chemoluminescence) kit (GE Healthcare).
Primary NHDFs (normal human dermal fibroblasts) and MEFs (murine embryonic fibroblasts) were grown in DMEM supplemented with 10% fetal calf serum and gentamycin (50 mg/l).
RT (reverse transcription)
Total RNA isolation was performed according to the manufacturer's protocol using the RNAeasy kit provided by Qiagen. The RT reaction was performed using the first strand cDNA synthesis kit (Roche). RNA (4 μg) was transcribed into cDNA and used for determining the PG receptor profile in NHDFs according to manufacturer's protocol. The first-strand cDNA synthesis (RT) was performed at 42 °C for 70 min (extension) followed by 5 min at 99 °C to inactivate reverse transcriptase. Subsequently, a PCR reaction with primers specific for the PG receptors EP1–EP4 was performed using the following programme (for 35 cycles): 120 s at 94 °C for the first denaturation, 30 s at 94 °C for further denaturation, 30 s at 55 °C for primer annealing, 40 s at 72 °C for extension and 10 min at 72 °C for the final extension. The amplicons were analysed by agarose gel gelectrophoresis.
The primers used  were: EP1fw, 5′-TCTACCTCCCTGCAGCGGCCACTG-3′; EP1rev, 5′-GAAGTGGCTGAGGCCGCTGTGCCGGGA-3′; EP2fw, 5′-TTCATCCGGCACGGGCGGACCGC-3′; EP2rev, 5′-GTCAGCCTGTTTACTGGCATCTG-3′; EP3fw, 5′-GAGCACTGCAAGACACACACGGAG-3′; EP3rev, 5′-GATCTCCCATGGGTATTACTGACAA-3′; EP4fw, 5′-CCTCCTGAGAAAGACAGTGTC-3′; and EP4rev, 5′-AGGACTCAGAGAGTGTCTT-3′.
Determination of MCP-1 expression
MCP-1 in the supernatant of NHDFs was determined with a quantitative sandwich ELISA according to the manufacturer's protocol (R&D). Briefly, MCP-1 antibodies were immobilized in microtitre plates. MCP-1 standard solutions and the supernatant from stimulated NHDFs were applied to the pre-coated wells and incubated for 2 h at room temperature (27 °C). Unbound proteins were washed away and HRP-conjugated MCP-1-antibody was added to each well and incubated for 1 h at room temperature. Following the washing steps, HRP substrate was added and further incubated at room temperature for 20 min. The reaction was stopped and the absorbance was measured in a microtitre plate reader (Molecular Dynamics) at 450 nm and 570 nm for λ-correction. MCP-1 concentrations in the analysed samples were re-calculated using the MCP-1 standard curve.
PGE1 inhibits IL-6-induced MCP-1 chemokine gene expression
MCP-1 is induced during inflammation by a number of inflammatory agents, including IL-6. Since PGE1 is known to be anti-inflammatory, in the present study we asked whether IL-6-induced MCP-1 induction is influenced by PGE1 (Figure 1). Prior to the stimulation with IL-6, PGE1 or IL-6 and PGE1, NHDFs were pre-incubated with the phosphodiesterase inhibitor IBMX to block the breakdown of intracellular cAMP generated by PGE1. It is essential to also add the phosphodiesterase inhibitors to those samples without PGE1 to consider potential effects of basal cAMP in the absence of PGE1 (Figure 1, 1st and 2nd bars). Although stimulation with IL-6/sIL-R resulted in a strong increase in MCP-1 gene expression (Figure 1, 2nd bar) when compared with the non-stimulated control cells (Figure 1, 1st bar), the IL-6-dependent MCP-1 expression was reduced in the presence of PGE1 (Figure 1, 4th bar). These results further demonstrate the physiological impact of PGE1 on IL-6-induced gene expression.
IL-6-induced MAPK activation, but not STAT activation, is inhibited in the presence of forskolin
PGE1 signals through G-protein-coupled receptors. To check whether signalling through G-protein-coupled receptors affects IL-6-induced activation of the JAK/STAT or MAPK cascades we stimulated NHDFs with IL-6 and the sIL-6R for up to 60 min in the presence or absence of the adenylate cyclase activator forskolin. ERK1/2 phosphorylation as well as STAT3 phosphorylation in whole cell lysates was analysed by Western blotting (Figures 2A and 2B). As a loading control we re-stained the blot for ERK1/2 and STAT3 (Figures 2C and 2D). Figure 2 demonstrates that ERK1/2 phosphorylation is detecTable 15 min after stimulation with IL-6/sIL-6R and fades out after 1 h (Figure 2A, first five lanes). Remarkably, ERK phosphorylation is severely impaired in cells treated with forskolin (Figure 2A, right-hand side), whereas STAT3 phosphorylation (Figure 2B) is hardly affected by treatment with forskolin. Inhibition of IL-6-dependent ERK activation by forskolin treatment could be confirmed in MEFs (see Supplementary Figure 1 at http://www.BiochemJ.org/bj/412/bj4120065add.htm) suggesting that our observation is not only specific for human fibroblasts. These results indicate that stimuli inducing cAMP-dependent signalling pathways specifically counteract IL-6-induced ERK activation but not STAT3 phosphorylation.
Pertussis toxin counteracts IL-6-induced ERK activation
Pertussis toxin inhibits signalling through Gi-protein-coupled receptors by halting Gi in its GDP-bound form. Consequently, the inhibition of target molecules such as the adenylate cyclase is impaired and the cellular cAMP concentration increases. Therefore we tested whether pertussis toxin leads to the same effect on IL-6-induced ERK activation as observed in response to forskolin. Figure 3 demonstrates reduced IL-6-dependent ERK activation in pertussis-toxin-treated NHDFs (Figure 3A, right-hand side), whereas STAT3 activation is hardly affected by the toxin (Figure 3B). Again, inhibited ERK activation but unchanged STAT3 activation in response to IL-6 could be confirmed in pertussis-toxin-treated MEFs (see Supplementary Figure 2 at http://www.BiochemJ.org/bj/412/bj4120065add.htm). These results suggest that activation of adenylate cyclase specifically counteracts IL-6-induced ERK phosphorylation but not activation of STAT3.
PGE1 blocks IL-6-induced ERK activation, which is crucial for IL-6-induced MCP-1 expression
To test whether PGE1 influences IL-6-induced ERK activation in a similar manner, we monitored IL-6-dependent ERK activation in whole cell lysates of NHDFs in the presence or absence of PGE1. Figure 4 shows that PGE1 blocked ERK activation after IL-6 treatment (Figure 4A) but not IL-6-dependent STAT3 phosphorylation (Figure 4B). To test whether inhibition of the MAPK cascade has any relevance for IL-6-induced MCP-1 expression, we compared MCP-1 induction by IL-6 in the absence or presence of the MEK inhibitor UO126 (Figure 5). Although stimulation with IL-6/sIL-R induces MCP-1 gene expression in the absence of UO126 (Figure 5, 2nd bar), the presence of UO126 strongly reduces the potential of IL-6 to induce MCP-1 expression (Figure 5, 4th bar). These results indicate a crucial role of the IL-6-intiated MAPK cascade for the induction of MCP-1.
The PGE1 receptor EP2 is expressed on NHDFs, and its activation counteracts IL-6-induced ERK activation and MCP-1 expression
Next we elucidated the molecular mechanisms involved in PGE1-dependent inhibition of IL-6-induced MAPK activation. First, we clarified which EP receptor was involved in PGE1-mediated repression of ERK activation in IL-6-stimulated fibroblasts. We analysed the expression of EP1 to EP4 using RT–PCR (Figure 6A) and found that EP1, EP2 and EP3 were expressed in NHDFs. Since EP2 and EP4 are known to activate adenylate cyclase, we focused on EP2 and used (R)-butaprost as a specific EP2 agonist. Similar to forskolin and PGE1, (R)-butaprost inhibited IL-6-dependent ERK activation (Figure 6B, top panel) without affecting STAT3 phosphorylation (Figure 6B, 2nd panel). These results strongly suggest that PGE1 suppresses IL-6-dependent ERK activation through the EP2 receptor.
In line with these observations, we tested whether butaprost could inhibit IL-6-dependent MCP-1 expression. Figure 6(C) shows that MCP-1 protein expression induced by IL-6/sIL-6R was reduced in the presence of the EP2 agonist butaprost (Figure 6C, compare the 2nd and 4th bars). These results further indicate that IL-6-dependent MCP-1 expression can be inhibited by EP2 activation.
Src does not contribute to PGE1-mediated inhibition of ERK activation
Previous reports have shown that Src is required for cAMP-mediated inhibition of growth-factor-induced ERK phosphorylation by activating Rap1 (Ras-proximate 1) . Thus we tested whether the Src-kinase inhibitor PP1 restored IL-6-induced ERK activation in the presence of PGE1 (Figure 7A). Surprisingly, the inhibitory potential of PGE1 was even more pronounced in the presence of PP1, clearly indicating that Src kinase activity was not crucial for the inhibitory activity of PGE1 on IL-6 signalling.
PGE1 acts through cAMP-activated PKA and not through cAMP-activated Epac
cAMP is known to activate directly PKA as well as the guaninenucleotide-exchange factor Epac. Our next aim was to investigate which of these cAMP targets has the potential to inhibit IL-6-induced ERK activation. We treated NHDFs with either the specific agonist for PKA (N-6Phe-cAMP) or the specific agonist for Epac (8-pCTP2′-O-Me-cAMP) and monitored ERK phosphorylation after stimulation with IL-6 (Figure 8A). STAT3 phosphorylation was not affected by either agonist (Figure 8A, 2nd panel). In contrast, the PKA agonist N-6Phe-cAMP inhibited IL-6-initiated ERK phosphorylation, whereas the Epac agonist 8-pCTP2′-O-Me-cAMP did not affect ERK activation after IL-6 stimulation (Figure 8A, top panel). These observations suggest that not the activation of Epac but the activation of PKA by cAMP is crucial for the inhibition of IL-6-induced ERK activation by PGE1.
To elaborate further whether PGE1 acts as an inhibitor by activating PKA, we treated NHDFs with the PKA inhibitor H89 and monitored whether blocking PKA by H89 renders these cells resistant to PGE1-mediated inhibition of IL-6-dependent ERK activation. Figure 8(B) shows the recovery of ERK phosphorylation when the cells were treated with PGE1 in the presence of the PKA inhibitor H89 (Figure 8B, top panel). Obviously, H89 overrides basal inhibition of ERK activation in the absence of IL-6. STAT3 activation was unaltered by H89 treatment (Figure 8B, 2nd panel). These results further indicate that PGE1 inhibits ERK phosphorylation in response to IL-6 by acting through PKA.
Previous studies on growth-factor signalling indicated that PKA phosphorylates and thereby inhibits the MAPKKK (MAPK kinase kinase) c-Raf-1. Thus we tested whether PGE1 induces phosphorylation of c-Raf-1 in NHDFs (Figure 8C). c-Raf-1 protein was precipitated from cellular extracts of NHDFs treated with PGE1 for 5 min or from untreated NHDFs. Phosphorylation of c-Raf-1 was monitored by Western blotting and subsequent staining with an antibody specific for phosphorylated PKA consensus sides. The right-hand lane in the top panel of Figure 8(C) indicates that PGE1 induced phosphorylation of c-Raf-1 within a protein motif representing a substrate for PKA. In summary these results suggest that PGE1 counteracts IL-6-dependent ERK activation by activating PKA which leads to c-Raf-1 phosphorylation to block the initiation of the MAPK cascade.
Inflammation is a response of an organism to cope with infections, sterile injuries and other trauma. The extent of inflammation is controlled by a set of pro- and anti-inflammatory cytokines as well as chemokines and non-protein mediators such as PGs, NO and ROS. Although much information is available with respect to the signal transduction of the individual mediators, only recently attention has been put on the mutual regulation of the signalling pathways. Regulating the cellular cAMP concentration is a crucial event for the signal transduction of chemokines and PGs. Previous studies have focused on the induction of cytokine expression by cAMP [36–39], whereas the present study investigates the influence of cAMP signalling on IL-6 signal transduction and IL-6-induced MCP-1 gene induction.
We analysed the influence of cellular cAMP on IL-6 signal transduction in fibroblasts and demonstrate specifically that IL-6-initiated ERK activation is counteracted by cAMP whereas STAT3 activation is not affected. Furthermore, we show that cAMP acts through PKA and c-Raf phosphorylation to inhibit the MAPK cascade, whereas activation of Epac by cAMP is not involved. This detail is important, since recently Sands and co-workers  reported that cAMP is able to induce SOCS3 expression through an Epac-dependent pathway in vascular endothelial cells. Obviously our observations reflect a different mechanism, since we did not detect reduced STAT3 activation in response to Epac agonists but specifically a reduction of IL-6-dependent ERK activation by PKA agonists (Figure 8A). Very probably, the induction of SOCS3 by Epac agonists and subsequent inhibition of STAT3 activation is specific for vascular endothelial cells.
It has been shown previously and also confirmed in the present study (results not shown) that the intracellular concentration of cAMP in fibroblasts increases drastically (100-fold) in response to PGE1 [41–43]. Accordingly, we observed a strong reduction of IL-6-dependent ERK activation in response to PGE1 in primary NHDFs. This observation indicates that not only pharmaceuticals which increase the cAMP concentration in the cell but also natural inflammatory regulators are potent regulators of IL-6 signal transduction.
Recently, Cheon and co-workers  described that PGE2 augments IL-10-mediated STAT3 and STAT1 activation in THP-1 cells. In contrast, PGE2 suppresses IL-6-induced STAT3 and STAT1 phosphorylation through mechanisms requiring de novo protein synthesis, probably SOCS3 expression . Corroborating this hypothesis, and in line with the results of the present study, cAMP alone did not affect STAT activation after IL-6 treatment. In addition to these observations, we show in the present study for the first time that PGE1 suppresses IL-6-induced ERK activation and MCP-1 gene induction.
We focused on the mechanism of how PGE1 affects IL-6-induced ERK activation. The fact that STAT3 activation by IL-6 is not affected by PGE1 argues the case to look for targets downstream of the activated receptor and JAKs. Although de Silva and co-workers  demonstrated PGE2-induced down-regulation of IL-6Rα expression in NFS-60 cells, we could exclude an effect of PGE1 on receptor expression because of ongoing STAT activation in the presence of PGE1 and the stimulation with the agonistically acting sIL-6Rα. Furthermore, the cell-surface expression of gp130 was controlled by FACS analyses (results not shown).
From previous studies by Schmitt and Stork [46–48] we know that cAMP also antagonizes ERK activation by growth factors. In this context PKA, activated through elevated intracellular cAMP concentrations, phosphorylates and activates Src kinase, which in turn leads to Rap1 activation . Rap1 counteracts Ras function in cells such as fibroblasts, which do not express B-Raf but activates ERK in cells expressing B-Raf [47,48]. In the course of preparing the present manuscript Stork and co-workers  further elucidated the mechanism of how cAMP activates the MAPK cascade through B-Raf. The authors demonstrate that activation of B-Raf through PKA but not through the cAMP-dependent guanine-nucleotide-exchange factor Epac is crucial for the initiation of the MAPK cascade in B-Raf-expressing cells. In summary, in both cases PKA accounts for the induction, as well as for the repression, of the growth-factor-induced MAPK cascade, dependent on the presence or absence of B-Raf.
In the present study, we found no evidence for a contribution of Epac (Figure 8A) or Src (Figure 7) to the inhibition of IL-6-dependent ERK activation. Src suppresses the inhibitory activity of PGE1 rather than mediating its inhibitory function on IL-6-dependent ERK activation (Figure 7). Instead, we demonstrated a crucial role of PKA (Figures 8A and 8B) for PGE1-mediated inhibition of ERK activation by IL-6. PKA is also known to inhibit c-Raf-1 by phosphorylation . Indeed, we could demonstrate PGE1-dependent phosphorylation of PKA-target sites within c-RAF-1 (Figure 8C) and a repression of IL-6-induced MCP-1 expression by PGE1 (Figure 1) which is in line with a crucial role of the IL-6-initiated MAPK cascade for MCP-1 expression as shown in Figure 5.
In summary our results show adenylate cyclase, PKA and c-Raf-1 to be involved in the inhibition of the MAPK cascade by PGE1 and its consequence on IL-6-induced MCP-1 gene expression. Understanding this cross-talk will help to critically judge the outcome of pharmaceutical approaches targeting PGE1 and IL-6.
We thank Hans F. Merk and Yvonne Marquardt from the Department of Dermatology (University of Aachen, Aachen, Germany) for the supply of primary dermal fibroblasts. This work was supported by the Deutsche Forschungsgemeinschaft (DGF-SFB 542) to F. S.
Abbreviations: DMEM, Dulbecco's modified Eagle's medium; EP, prostaglandin E receptor; Epac, exchange protein directly activated by cAMP; ERK, extracellular-signal-regulated kinase; HRP, horseradish peroxidase; IBMX, 3-isobutyl-1-methylxanthine; IL, interleukin; IL-6R, IL-6 receptor; JAK, Janus kinase; MAPK, mitogen-activated protein kinase; MCP-1, monocyte chemoattractant protein-1; MEF, murine embryonal fibroblast; NHDF, normal human dermal fibroblast; PG, prostaglandin; PGE, prostaglandin E; PKA, protein kinase A; Rap1, Ras-proximate 1; ROS, reactive oxygen species; RT, reverse transcription; SHP-2, Src homology 2 domain-containing protein tyrosine phosphatase 2; sIL-6R, soluble IL-6R; SOCS, suppressor of cytokine signalling; STAT, signal transducer and activator of transcription; TNF, tumour necrosis factor
- © The Authors Journal compilation © 2008 Biochemical Society