|Home | About | Journals | Submit | Contact Us | Français|
Increasing evidence indicates that Pulmonary Arterial Hypertension (PAH) is a vascular inflammatory disease. Prostacyclin (PGI2) is widely used to treat PAH and is believed to benefit patients largely through vasodilatory effects. PGI2 is also increasingly believed to have anti-inflammatory effects; including decreasing leukocyte cytokine production, yet few mechanistic details exist to explain how these effects are mediated at the transcriptional level. Since activated monocytes are critical sources of MCP-1 and other cytokines in cardiovascular inflammation, we examined the effects of iloprost on IFN-γ and IL-6 stimulated cytokine production in human monocytes. We found iloprost inhibited IFN-γ and IL-6-induced MCP-1, IL-8, RANTES, and TNF-α production in monocytes indicating wide-ranging anti-inflammatory action. We found that activation of STAT1 was critical for IFN-γ-induced MCP-1 production and demonstrated that iloprost inhibited STAT1 activation by several actions: 1) iloprost inhibited the phosphorylation of STAT1-S727 in the transactivation domain (TAD), thereby reducing recruitment of the histone acetylase and co-activator CBP/p300 to STAT1; 2) iloprost selectively inhibited activation of janus kinase 2 (JAK2), but not JAK1, both responsible for activation STAT1 via phosphorylation of STAT1-Y701, resulting in reduced nuclear recruitment and activation of STAT1; 3) SOCS-1, which normally terminates IFN-γ-signaling, was not involved in iloprost-mediated inhibition of STAT1, indicating divergence from the classical pathway for terminating IFN-γ-signaling. We conclude that PGI2 exerts anti-inflammatory action by inhibiting STAT1 induced cytokine production, in part by targeting the transactivation domain induced recruitment of the histone acetylase CBP/p300.
It has become increasingly clear that vascular inflammation plays a critical role in a wide variety of cardiovascular diseases including pulmonary arterial hypertension (PAH) (1–10). It thus stands to reason that anti-inflammatory strategies designed specifically to combat vascular inflammation could result in improved therapy for cardiovascular diseases. One route to this goal is to understand how endogenous anti-inflammatory mediators act to suppress vascular inflammation. Among such mediators is prostacyclin (PGI2), a potent endogenous vascular protective eicosanoid that activates a G-protein coupled receptor (GPCR), the IP receptor, which signals through Gs-adenylyl cyclase and cAMP-PKA (11). PGI2 was originally identified as a potent vasodilator and because several reports indicated a relative deficiency of PGI2 in the lung vasculature of PAH patients, it has been widely used a treatment for PAH (12–14). In addition to its vasodilator effects, there is now evidence suggesting PGI2 also benefits PAH patients via anti-inflammatory actions (15). Importantly, PGI2 has been demonstrated to exhibit anti-inflammatory actions in asthma models in vivo (16, 17), in diabetic patients with vascular inflammation (18), in critical limb ischemia (19), and bacterial inflammation in vitro (20, 21). Its use is reported to reduce the elevated circulating levels of the cytokine monocyte chemoattractant protein MCP-1 (CCL2) found in PAH patients (15). The potential anti-inflammatory effects of PGI2 thus represent an understudied feature of this eicosanoid. How it exerts these effects at the transcriptional level remains largely unknown.
In vascular remodeling (both pulmonary and systemic), many chemokines including MCP-1, RANTES (CCL5), fractalkine (CXCL1) and SDF-1 (CXCL12) are up regulated in the injured vasculature and cooperate in leukocyte recruitment and resident cell activation (3). Among these, MCP-1, which activatesthe receptor CCR2, plays a well-established critical role in many inflammatory vascular diseases (1–3, 7, 15, 22–25). MCP-1 recruits leukocytes, activates resident cells at sites of vascular inflammation (26, 27) and is a prognostic indicator of cardiovascular disease predicting increased likelihood of cardiovascular events in patients (2). In both PAH and atherosclerosis animal models, neutralizing MCP-1 reduces disease(1, 7, 24, 25). For example gene transfer of a CCR2 antagonist greatly reduced vein graft thickening where infiltrating macrophages express high levels of MCP-1(28). These observations indicate the need to determine how PGI2 inhibits transcription factors stimulating proinflammatory cytokine expression.
The goal of this study was to determine the signaling pathways through which PGI2 inhibits cytokine production in monocytes (15–17, 19–21). As an approach we investigated the effects of PGI2 on IFN-γ-stimulated MCP-1 production in monocytes, as this signaling cascade is fairly well characterized, and appears to be one of the principle means utilized to generate vascular inflammation sand MCP-1 in vivo (3, 29, 30). We also evaluated the effects of PGI2 on other cytokines produced by IFN-γ and IL-6. Our findings demonstrate that PGI2 exerts wide ranging anti-inflammatory effects. The inhibition of MCP-1 production is mediated by inhibition of STAT1 preventing phosphorylation of STAT1-S727 in the transactivation domain which controls recruitment of CBP/p300 and of STAT1-Y705 controlling nuclear recruitment and is independent of suppressor of cytokine signaling-1 (SOCS-1) activation (31, 32). Inhibiting phosphorylation within the transactivation domain (TAD) transcription factors targeted by PGI2 may be a means to achieve a coordinated widespread anti-inflammatory action. This information is useful in understanding how PGI2 limits production of the vascular inflammatory cytokine MCP-1 and potentially other cardiovascular inflammatory cytokines.
Rabbit polyclonal antibodies to STAT1, SOCS-1, p300 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA), Ab to CBP, phospho-STAT1 Y701, and phospho-STAT1-S727, activation state specific Ab phospho-JAK1-Y1022/1023, and phospho-JAK2-Y1007/1008 were purchased from Cell Signaling Technology (Danvers, MA). Horseradish peroxidase-labeled anti-rabbit antibodies, ECL reagents and were purchased from Amersham Pharmacia Biotech. HRP-labeled anti-rabbit Abs and chemiluminescence reagents were purchased from BioRad. PKA inhibitor H89, and forskolin were purchased from Calbiochem. The prostacyclin analogs Iloprost, treprostinil, carbaprostacyclin, and the prostacyclin receptor antagonist CAY10449 were purchased from Cayman Chemical, (Ann Arbor, MI). All other reagents were purchased from Sigma-Aldrich unless otherwise noted in the text.
THP-1 from American Type Tissue Culture Collection (ATCC, Manassas, VA) were maintained as recommended by supplier.
Human CD14+ monocytes were purified from PBMC fraction using CD14 Microbeads (Miltenyibiotech). Blood was drawn from healthy volunteers, under IRB protocol COMIRB 04-0758, and PBMC fraction was isolated by standard procedure using Ficoll-Plaque PLUS (GE healthcare).
SOCS-1 protein was knocked down using siRNA, Dharmacon RNA Technology, (Lafayette, CO, USA) a pool of four small interfering RNAs (siRNAs) provided by Dharmacon RNA Technology (ON-TARGETplus SMARTpool, L-011511–00-0005), whose sequences (Sense Sequences: 5′-GAGCCCCGACUGCCUCUUCUU; 5′-UCCGUUCGCACGCCGAUUAUU; 5′-GCAUCCGCGUGCACUUUCAUU; 5′-CCAGGUGGCAGCCGACAAUUU) annealed on different regions of human SOCS1 mRNA (NM_003745). In parallel, a pool of four nontargeting siRNAs was used as negative control (ON-TARGET plus siCONTROL, D-001810–10-05). THP-1 cells were transfected with SOCS1 or irrelevant siRNAs at a 50 nM final concentration using DharmaFECT transfection reagent (Lafayette, CO, USA), according to the manufacturer’s protocol. After 36 hours of transfection, the cells were used immediately for experimentation. For STAT1 siRNA mediated knock down we used Cell Signaling, SignalSilence Stat1 siRNA Kit #6545, at final concentration of 50 nM using cell signaling transfection reagent, after 36 hrs cells were utilized for experimentation.
Immunoblots for phosphorylated and total kinases were performed by standard techniques as described previously(33).
immunoprecipitation of STAT1, CBP, p300 was performed as described previously using 1 mg Ab per 0.5 ml lysate per 5 million cells (33).
Immunoreactive TNF-α, IL-8, RANTRES, IL-6, VEGF, and MCP-1 were quantified using commercially available ELISA kits (PeproTech, NJ), according to the manufacturer’s instructions and as described previously (33).
Nuclear extracts were prepared and assayed by EMSA as described previously (33, 34). For the analysis of STAT1, the gamma interferon activated sequence GAS-DNA consensus sequence was used. Synthetic double-stranded sequences (with enhancer motifs underlined) were filled in and labeled with [γ-32P]dATP using Sequenase DNA polymerase as follows: Gel shift oligonucleotides were purchased from Santa Cruz, 5′-CAT-GTT-ATG-CAT-ATT-CCT-GTA-AGT-3′ consensus binding motif for STAT1, negative control mutant 5″-CAT-GTT-ATG-CAT-ATT GGA GTA AGT-G3′ to which no specific binding (removable by pre-incubation with STAT1 Ab) was observed.
Janus kinase (JAK) assay was performed as described previously (34) except the activity was determined by WB with activation state specific Ab phospho-JAK1-Y1022/1023, and phospho-JAK2-Y1007/1008.
Data are presented as mean ± SD for each experimental group. One-way ANOVA and Student’s t test (for comparisons between two groups) was used, with values of p < 0.05 considered significant.
To define the anti-inflammatory effects of PGI2, we evaluated the effects of iloprost, a clinically used stable analog of PGI2, on IFN-γ and IL-6 stimulated cytokine expression in the widely employed monocyte cell line THP-1. Iloprost (50 nM) inhibited IFN-γ-induced MCP-1 production in a dose response experiment (Figure 1A), demonstrating at least a 50% inhibition of MCP-1 production at the highest concentration of IFN-γ used. Measuring the dose response to iloprost, at maximal concentration of IFN-γ (20 ng/ml), we observed an even greater inhibitory effect on MCP-1 production (greater than 90% inhibition) with an IC50 of 3 nM, as the dose of iloprost was increased (Figure 1B). This response is consistent with activation of the high affinity (low nM EC50) IP receptor for PGI2. It should be noted that iloprost does activate the related eicosanoid receptor EP1 with equal affinity, the natural ligand for which is PGE2 (35). However previous studies have positively identified the iloprost response in leukocytes as being due to the prostacyclin IP receptor, based on loss of response in leukocytes derived from IP−/− mice (21). Iloprost also profoundly inhibited IFN-γ-stimulated TNF-α (Figure 1C), IL-8 (Figure 1D) and RANTES (Figure 1E) synthesis, all implicated in both PAH and other vascular inflammatory disorders (3, 29, 36). We also investigated the capacity of iloprost to inhibit IL-6 induced MCP-1 and TNF-α production as IL-6 has been repeatedly implicated in PAH pathogenesis (5, 9, 37). We found that iloprost significantly decreased the ability of IL-6 to stimulate MCP-1 production and TNF-α (data not shown) in monocytes. The prostacyclin analogs treprostinil and carbaprostacyclin, like iloprost, also inhibited IFN-γ-stimulated MCP-1 production, and the response to all agents was abolished in the presence of the prostacyclin receptor antagonist CAY10449 (Figure 1F). This data strengthens the notion that the effects of iloprost are predominantly due to activation of the prostacyclin receptor.
These data suggest relatively broad anti-inflammatory effects of PGI2 in monocytes. Based on previous studies demonstrating the importance of MCP-1 in cardiovascular disease including PAH (1–3, 7, 15, 22, 23, 25, 28), we will focus on this chemokine in experiments shown below evaluating the inflammatory signaling pathways targeted by PGI2.
Previous studies demonstrated STAT1 is probably the major transcription factor involved in IFN-γinduced synthesis of MCP-1 (38). This being correct, then analysis can be restricted to STAT1, rather than including all IFN-γ activated transcription factors (STAT2, STAT3, STAT5, AP1, Pu.1, IRF-1, and IRF-4/8(30)). We down regulated STAT1 using specific siRNA and measured MCP-1 in response to IFN-γ. Figure 2A shows the effectiveness of our siRNA STAT1 in monocytes. Knockdown of STAT1 blocked IFN-γ induced MCP-1 production in THP-1 cells compared to those transfected with control siRNA (Figure 2B). Thus as previously asserted in literature, IFN-γ-stimulated MCP-1 production depends predominantly on STAT1 activity and therefore it is sufficient to consider only STAT1.
Activation of STAT1 is entirely determined by phosphorylation at Y701 and S727, separate events each required for full activation (39). The activity of STAT1 is significantly enhanced by phosphorylation of STAT1-S727 which activates the transcativation domain (TAD)-dependent recruitment of CBP/p300 (39). Transactivation domains are found in many transcription factors and usually increase transcription by by serving as docking sites for CBP/p300 which facilitates the unwinding of DNA thereby allowing assembly of the large multiprotein transcriptional complex. To determine effect of iloprost on STAT1-S727 phosphorylation, we stimulated cells with sub-maximal concentration IFN-γ, 1 ng/ml, and increasing concentrations of iloprost 0.1–100 nM for 4 hours. Using submaximal concentration of IFN-γ allows for the detection of changes in EC50 for IFN-γ-stimulated responses. Iloprost inhibited phosphorylation of STAT1-S727 in a dose dependant manner (Figure 3A) at concentrations in the low nM range consistent with inhibition of MCP-1 (Fig. 1B). Iloprost dramatically decreased the phosphosphorylation of STAT1-S727 over time (Figure 3B). The dose response to IFN-γ in the presence of iloprost shows a dramatic reduction in IFN-γ-stimulated phosphorylation of STAT1-S727 (Figure 3C).
PGI2 acts via the IP receptor, stimulating adenylyl cyclase thereby increasing cAMP-PKA, an action responsible for many biological effects of PGI2 including anti-inflammatory action (12, 40, 41). Therefore we tested effects of the widely used selective PKA inhibitor H89 on iloprost-induced inhibition of MCP-1 production, and found that the ability of iloprost to inhibit MCP-1 synthesis was attenuated demonstrating that the inhibitory action of PGI2 is mediated largely by cAMP-PKA signaling (Figure 3D). Activation of PKA with forskolin, widely employed to directly activate adenylyl cyclase hence increasing cAMP-PKA, resulted in a similar inhibition of IFN-γ induced MCP-1 synthesis suggesting a role for PKA in this pathway (Figure 3D). Since the cAMP-PKA system clearly participates, it suggested cAMP-PKA could mediate the effects of iloprost on phospho-STAT1-S727. Forskolin reduced the phosphorylation of STAT1-S727 confirming the suspicion that cAMP-PKA mediates the effects of iloprost on this phosphorylation (Figure 3E).
Since phosphorylation of STAT1-S727 is required for the recruitment of CBP and p300, we measured levels of CBP associated with STAT1 in cells treated with iloprost (Figure 4A). There was a decreased level of CBP associated with STAT1 in monocytes treated with iloprost compared to control in a cAMP-PKA dependent manner (Figure 4A). Iloprost treatment also reduced recruitment of p300, the functional homolog of CBP, to activated STAT1 (Figure 4B). These results confirm that iloprost inhibits MCP-1 production by inhibition of STAT1-S727 phosphorylation resulting in decreased recruitment of CBP/p300 histone acetylase to the MCP-1 gene.
Since phosphorylation of Y701 is essential for nuclear accumulation of STAT1 and binding to the GAS (Gamma interferon activated sequence) elements of the promoter, we evaluated the effects of iloprost on this event. It is well known that phosphorylation of STAT1-Y701 and –S727 are separate molecular events with the former controling initiation of STAT1-dependent gene transcription and the latter controling the extent of gene transcription (39). To determine if there was an EC50 shift, THP-1 monocytes were challenged with sub-maximal concentration IFN-γ, 1 ng/ml, and increasing concentrations of iloprost 0.1–100 nM for 4 hours, immunoprecipitated for STAT1 and probed with antibody to phospho-STAT1-Y701. Iloprost treatment inhibited phospho-STAT1-Y701, at sub-maximal IFN-γ concentrations similar to those effective at blocking production MCP-1 (Figure 5A). The dose response curve to IFN-γ-stimulated phosphorylation of STAT1-Y701 demonstrated that iloprost shifted the EC50 to the right, such that at low concentrations of IFN-γ there was an inhibitory effect of iloprost (Figure 5B). At low concentrations of IFN-γ, iloprost also reduced the accumulation of STAT1-Y701 in nuclear fraction (Figure 5C). At low concentrations of IFN-γ there was reduced binding of STAT1 to the the GAS element as measured by EMSA (Figure 5D). Iloprost clearly differentially regulates phosphorylation of STAT1 at –Y701 and –S727, such that the predominant effects are to diminish phosphorylation of the latter, and more moderate effects on the former. Therefore the effects of iloprost seem to be associated with reduced magnitude of MCP-1 gene expression, a reflection of CBP/p300 recruitment, rather than effecting the initiation of MCP-1 gene expression. These data do not rule out the possibility of direct inhibitory effects on IFN-γ receptor, or other STAT1 independent effects.
IFN-γ-stimulated responses activate JAK1 and JAK2 via IFN-γ-R1, or -R2, which phosphorylate STAT at Y701, resulting in its nuclear translocation and binding to the GAS element thereby stimulating IFN-γ-activated genes (29, 30). To determine effects of iloprost on JAK kinases, we immunoprecipitated JAK1 and JAK2 from iloprost treated cells and probed for phosphorylation using specific antibodies phospho-JAK1-Y1022/1023, and phospho-JAK2-Y1007/1008. Iloprost selectively reduced activation of JAK2 and not JAK1 (Figure 6A & 6B), which could explain, reduced phosphorylation of STAT1-Y701. The selective loss of JAK2 activation indicates how STAT1, can be partially inactivated by reduced (but not absent) phosphorylation of STAT1-Y701 as in (Figure 5B).
SOCS proteins are negative feedback proteins terminating cytokine signaling by multiple methods including direct inhibition of JAK1 and JAK2 kinases via a kinase inhibitory domain (31, 32). SOCS-1 is known to be the principal SOCS protein attenuating IFN-γ-stimulated STAT1 activation (31, 32, 34). Therefore PGI2, like some other anti-inflammatory agents could exert this effect through up regulation of SOCS to terminate MCP-1 production. We measured levels of SOCS-1 from cells treated with IFNγ and iloprost by immunopreciptation of SOCS-1 followed by immunoblot (Figure 7A). IFN-γ increased SOCS-1, as previously described (31, 32, 42), but iloprost did not further increase SOCS-1, as one would predict if iloprost used the SOCS-1 pathway to terminate IFN-γ-stimulated MCP-1 production (Figure 7A). To determine whether inhibition of MCP-1 by iloprost requires SOCS-1 we treated cells with SOCS-1 siRNA and studied the effects of iloprost on IFN-γ induced MCP-1 production. Knockout of SOCS-1 increased IFN-γ-stimulated MCP-1 production as expected (Figure 7B) given the negative feedback role of SOCS-1 on IFN-γ-stimulated responses described in literature. SOCS-1 knock down by siRNA did not prevent iloprost from dramatically inhibiting IFN-γ-stimulated MCP-1 production (Figure 7B) indicating SOCS-1 is not involved in the inhibitory effect of iloprost exerts on MCP-1 production.
Because the above experiments were performed in a human monocyte cell line (THP-1), we wanted to determine if iloprost could inhibit IFN-γ stimulated MCP-1 production in primary human monocytes. We found that iloprost significantly attenuated the IFN-γ-stimulated MCP-1 production in primary human mononuclear cells (Figure 8A). As with data in THP-1 monocytes, iloprost significantly attenuated the phosphorylation of STAT1-Y701, and –S727 in these cells. These data suggest that the results with THP-1 monocytes accurately reflect the regulation of IFN-γ-stimulated MCP-1 and STAT1 responses in primary human monocytes.
PGI2 has been traditionally thought of as a vasodilator and is used as such to treat PAH patients. However, recent studies demonstrate significant anti-inflammatory actions of PGI2 (15–21, 43, 44). For instance, iloprost inhalation suppressed asthma in the mouse ovalbumin-induced model by reducing migration of dendritic cells to sites of inflammation and the Th2 response (reduced IL-4, IL-5, and IL-13)(16). Independent studies corroborated these findings by demonstrating that iloprost suppresses CCL17, CCL11, and lung eosinophil accumulation, and recruitment of CD4+ Th2 cells to the airways in asthma models (17, 43). Further, PGI2 counteracts the capacity of T cells to produce proinflammatory cytokines, IL-12p70, TNF-α, MIP1α, MCP-1, IL-6 (21) (44). Effective anti-inflammatory actions against leukocytes stimulated by the bacterial toxin LPS has also been demonstrated, an action connected to inhibition of NF-κB (20). Anti-inflammatory action is also evident in vascular inflammatory diseases, such as in diabetic vascular inflammation (18). In patients with type 2 diabetes mellitus, PGI2 therapy reduced circulating levels of the vascular inflammation marker VCAM-1, and arterial intima-media thickness (18). In vitro studies demonstrated beraprost, a PGI2 analog, reduced TNF-α-induced expression of VCAM1 in human vascular endothelial cells (18). In PAH, increased circulating MCP-1 appears to be a pathogenic factor, and PGI2 therapy clearly decreased circulating MCP-1 (15). Therefore anti-inflammatory effects of PGI2 may be an additional, hitherto unrecognized benefit of PGI2-therapy in PAH patients. The present studies extend previous published studies by demonstrating that PGI2 exerts wide spread anti-inflammatory actions in mononuclear cells and provides insight into the specific pro-inflammatory signaling pathways abrogated by prostacyclin therapy.
Understanding how endogenous anti-inflammatory agents like PGI2 work could provide insight into mechanisms potentially valuable for the development of better therapeutics to combat vascular inflammation (11, 15, 16, 18, 20, 21, 45). Our data, and that of others, indicating broad anti-inflammatory action of PGI2, suggests that many transcription factors are likely to be targets of PGI2, rather than the few currently described in the literature (46–50). To date PGI2 is known to activate CREB (50), inhibit NF-κB (20), and activate the nuclear receptor family members PPAR-δ (47), PPAR-γ (48). However other than these few limited characterizations, there are few mechanistic details indicating how PGI2 inhibits transcription factors driving proinflammatory gene expression. Our data provides new insights into how PGI2 controls expression of another important pro-inflammatory transcription factor, STAT1.
The IFN-γ-STAT1 pathway, critical in cardiovascular inflammation, activates STAT1 by two essential but independent phosphorylations; 1) Y701 which controls binding to γ-interferon stimulated genes, and 2) S727, within the TAD, which controls recruitment of CBP/p300 (39). We found that PGI2 inhibits the phosphorylation-dependent activation of the TAD (phospho-STAT1-S727), thereby reducing the recruitment of CBP/p300 and thus MCP-1 production. The histone acetylase CBP, and its homolog p300 activate transcription by opening the DNA, which is normally tightly wound in the nucleosome. Essentially CBP acetylates lysine residues in histones, thereby neutralizing the positive charge which in turn weakens the interaction of positively charged histones with negatively charged phosphate backbone of DNA (51). Recruitment of CBP/p300, to the TAD of transcription factors is a common mechanism for activating transcription. Notable examples where phosphorylation within the TAD of a transcription factor leads to activation by recruitment of CBP/p300 include CREB (phospho-CREB-S133), NF-κB (phospho-p65-S536), the tumor suppressor p53 (phospho-p53-T18/S20), and c-Fos of AP1 (phospho-c-Fos-S63) (52). It will be important in the future experiments to determine if PGI2 inhibits other proinflammatory transcription factors by a coordinated action of inhibiting phosphorylation dependent TAD directed mechanisms, thereby reducing recruitment of CBP/p300. Our experiments also suggest an additional mechanism by which PGI2 can inhibit MCP-1 production; i.e. by attenuating nuclear accumulation of phospho-STAT1-Y701, via selective inhibition of JAK2, but not JAK1. This action reduces but does not eliminate binding of STAT1 to the γ-interferon activated sequence (GAS) element which is critical for transcription of STAT1 driven genes (30).
Importantly we show that SOCS-1, the SOCS protein known to terminate IFN-γ signaling (31, 32), is not involved in the actions of PGI2. SOCS proteins terminate cytokine signaling, controlling the extent and duration of cytokine responses (31, 32). Several GPCR signaling systems involved in tissue protective endogenous anti-inflammatory mechanisms, including vasoactive intestinal peptide (VIP) (34, 53), adenosine (54), lipoxin A4 (55) have been shown to exert anti-inflammatory effects by reducing STAT transcription factor activation, which may involve SOCS proteins. Some evidence has been presented to suggest that the anti-inflammatory effects of cAMP may be in part due to increased expression of SOCS-3, known to terminate IL-6-induced STAT3 activation (41). Other studies have shown VIP, which like PGI2-IP activates Gs-adenylyl cyclase thereby increasing cAMP decreases IFN-γ-induced STAT1 activation, by inhibiting both JAK1, and JAK2 activation, but the involvement of SOCS proteins was not determined (5, 34). On the other hand there are indications that anti-inflammatory action of GPCRs does not need to involve SOCS proteins. For example PGE2 has been shown to attenuate IL-6 production induced MCP-1 by reducing activation of ERK1/2 without any inhibitory action on STAT3 per se (53). Therefore it appears anti-inflammatory agents including adenosine, PGE2, PGI2 may utilize distinct mechanisms to terminate STAT-signaling opening the possibility for combined strategies to combat inflammation in cardiovascular disease.
PGI2 is now known to be a powerful cardiovascular protective agent with wide-ranging anti-inflammatory actions (11, 15–17, 19, 21, 43–45), yet few details are known as to how this agent combats proinflammatory cytokine generation at the transcriptional level (20, 46, 47, 49). For the first time our data define detailed transcriptional mechanisms by which PGI2 inhibits MCP-1 production via the critical cardiovascular inflammatory axis of IFN-γ-STAT1 in monocytes. Iloprost inhibited STAT1 by preventing phosphorylation of STAT1-S727 in the TAD, reducing recruitment of the histone acetylase CBP. Secondly iloprost selectively inhibited activation of JAK2, but not JAK1, inhibiting phosphorylation of STAT1-Y701, resulting in reduced nuclear recruitment and activation of STAT1. Thirdly SOCS-1, which normally terminates IFN-γ-signaling, was not involved in iloprost-mediated inhibition of STAT1, indicating divergence from the classical pathway terminating IFN-γ-signaling. Our data suggest that one major mechanisms by which PGI2 limits proinflammatory cytokine production may be by reducing TAD dependent recruitment of histone acetylase CBP to transcription factors driving proinflammatory gene transcription. It remains to be seen if this is a general mechanism predicting how endogenous cardiovascular protective agents limit vascular inflammation. Defining such mechanisms will provide invaluable insight to the molecular actions required to combat vascular inflammation.
Sources of funding: National Institutes of Health (HL64917 to M.D., HL57144-09, HL14985-33, P01 HL014985-35, P50 HL084923-03 to K.R.S.