Selective Prostacyclin Receptor Agonism Augments Glucocorticoid-Induced Gene Expression in Human Bronchial Epithelial Cells This information is current as of October 1, 2021. Sylvia M. Wilson, Pamela Shen, Christopher F. Rider, Suzanne L. Traves, David Proud, Robert Newton and Mark A. Giembycz J Immunol published online 30 October 2009

http://www.jimmunol.org/content/early/2009/10/30/jimmuno Downloaded from l.0902738

Why The JI? Submit online. http://www.jimmunol.org/

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on October 1, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2009 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published October 30, 2009, doi:10.4049/jimmunol.0902738 The Journal of Immunology

Selective Prostacyclin Receptor Agonism Augments Glucocorticoid-Induced Gene Expression in Human Bronchial Epithelial Cells1

Sylvia M. Wilson,* Pamela Shen,2* Christopher F. Rider,† Suzanne L. Traves,* David Proud,* Robert Newton,† and Mark A. Giembycz3*

Prostacyclin receptor (IP-receptor) agonists display anti-inflammatory and antiviral activity in cell-based assays and in preclinical models of asthma and chronic obstructive pulmonary disease. In this study, we have extended these observations by demonstrating that IP-receptor activation also can enhance the ability of glucocorticoids to induce genes with anti-inflammatory activity. BEAS-2B bron- chial epithelial cells stably transfected with a glucocorticoid response element (GRE) luciferase reporter were activated in a concen- tration-dependent manner by the glucocorticoid dexamethasone. An IP-receptor agonist, taprostene, increased cAMP in these cells and augmented luciferase expression at all concentrations of dexamethasone examined. Analysis of the concentration-response relationship Downloaded from that described this effect showed that taprostene increased the magnitude of transcription without affecting the potency of dexameth- asone and was, thus, steroid-sparing in this simple system. RO3244794, an IP-receptor antagonist, and oligonucleotides that selectively silenced the IP-receptor gene, PTGIR, abolished these effects of taprostene. Infection of BEAS-2B GRE reporter cells with an adenovirus vector encoding a highly selective inhibitor of cAMP-dependent protein kinase (PKA) also prevented taprostene from enhancing GRE- dependent transcription. In BEAS-2B cells and primary cultures of human airway epithelial cells, taprostene and dexamethasone interacted either additively or cooperatively in the expression of three glucocorticoid-inducible genes (GILZ, MKP-1, and p57kip2) that http://www.jimmunol.org/ have anti-inflammatory potential. Collectively, these data show that IP-receptor agonists can augment the ability of glucocorticoids to induce anti-inflammatory genes in human airway epithelial cells by activating a cAMP/PKA-dependent mechanism. This observation may have clinical relevance in the treatment of airway inflammatory diseases that are either refractory or respond suboptimally to glucocorticoids. The Journal of Immunology, 2009, 183: 6788–6799.

4 rostacyclin (PGI2) is a labile eicosanoid derived from ar- tor), which typically couples to Gs for the activation of adenylyl

achidonic acid following the sequential action of cycloox- cyclase (2). Although the role of PGI2 in the regulation of vascular

ygenase and PGI synthase (1). The biological actions of by guest on October 1, 2021 P 2 homeostasis has, for many years, been a primary research focus, it PGI2 are mediated primarily through the PGI2 receptor (IP-recep- is now appreciated that PGI2 and its cognate receptor may play important (patho)physiological roles in a variety of other pro- cesses, including airway inflammatory diseases (2). *Department of Physiology and Pharmacology and †Department of Cell Biology and Anatomy, Airways Inflammation Research Group, Institute of Infection, Im- The IP-receptor subtype is expressed on many immune and munity and Inflammation, Faculty of Medicine, University of Calgary, Calgary, proinflammatory cells, including the monocyte (3), T lymphocyte Alberta, Canada (4, 5), dendritic cell (6), epithelial cell (7), and airway (8) and Received for publication August 19, 2009. Accepted for publication September pulmonary vascular smooth muscle cells (9). Moreover, accumu- 21, 2009. lated evidence suggests that IP-receptor agonism in the lung may The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance exert anti-inflammatory and/or antiviral activity. For example, the with 18 U.S.C. Section 1734 solely to indicate this fact. IP-receptor agonists cicaprost and iloprost attenuate cytokine pro- ϩ 1 This study was funded in part by the Canadian Institutes of Health Research (CIHR) duction from anti-CD3-stimulated, murine CD4 T lymphocytes and the GlaxoSmithKline/Collaborative Innovation Research Fund. R.N. is a CIHR cultured under conditions that cause Th1 or Th2 cell polarization New Investigator and an Alberta Heritage Foundation for Medical Research (AHFMR) Scholar. M.A.G. is an AHFMR Senior Scholar. D.P. holds a Canadian (5). Cytokine secretion from murine bone marrow-derived den- Research Chair in Inflammatory Lung Diseases. dritic cells in response to LPS is also inhibited by iloprost (6, 10). 2 Current Address: Department of Pathology and Molecular Medicine, Centre for A central role for the IP-receptor in regulating allergic inflamma- Gene Therapeutics, McMaster University, Hamilton, Ontario, Canada. tory responses in vivo similarly has been documented. Thus, IP- 3 Address correspondence and reprint requests to Dr. Mark A. Giembycz, Department receptor gene deficiency in allergen-challenged, sensitized mice is of Physiology and Pharmacology, Airways Inflammation Research Group, Institute of Infection, Immunity, and Inflammation, 3280 Hospital Drive Northwest, Calgary, associated with a phenotype characterized by exaggerated pulmo- Alberta, Canada T2N 4N1. E-mail address: [email protected] nary inflammation and airway hyper-responsiveness when com- 4 Abbreviations used in this paper: PGI2, prostacyclin; COPD, chronic obstructive pared with wild-type animals (11, 12). These data are consistent pulmonary disease; GILZ, glucocorticoid-inducible leucine zipper; GR, glucocorti- coid receptor; GRE, glucocorticoid response element; E/[A], agonist concentration with the finding that PGI2 attenuates allergen-induced inflamma- ϩ effect; HEK, human embryonic kidney; HpAEC, human primary airway epithelial tion in mice by limiting pulmonary CD4 T cell recruitment (4, cell; ICS, inhaled glucocorticoid; IP-receptor, PGI2 receptor; MKP, MAPK phospha- 13), which may be due to an inhibitory effect on dendritic cell tase; MOI, multiplicity of infection; p57kip2, kinase inhibitor protein 2 of 57 kDa; PDE, phosphodiesterase; PKA, cAMP-dependent protein kinase A; PKI, PKA inhib- function (14). Activation of the IP-receptor may also be beneficial itor; PPAR, peroxisome proliferator-activated receptor; SFM, serum-free medium; in chronic obstructive pulmonary disease (COPD). Indeed, tapro- siRNA, small interfering RNA; TX, thromboxane. stene, an IP-receptor agonist, suppressed the generation of the ϩ Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00 CD8 T cell chemoattractants CXCL9 and CXCL10 from human

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0902738 The Journal of Immunology 6789

Table I. siRNA oligonucleotide sequences

Gene Oligonucleotides Accession No. Source

PTGIRa 5Ј-CATCCATCTCATTGTCTAAtt-3Ј NM_000960.3 Qiagen 3Ј-ggGTAGGTAGAGTAACAGATT-5Ј PTGIRb 5Ј-GGGCGACAGGAGCCAGAAAtt-3Ј NM_000960.3 Qiagen 3Ј-agCCCGCTGTCCTCGGTCTTT-5Ј GFP 5Ј-GGCAAGCTGACCCTGAAGTTCtt-3Ј U57609 Dharmacon 3Ј-ttCCGTTCGACTGGGACTTCAAG-5Ј GAPDH 5Ј-GAGCCACATCGCTCAGACAtt-3Ј NM_002046 Qiagen 3Ј-ggCTCGGTGTAGCGAGTCTGT-5Ј

a Refers to PTGIR-1 siRNA in the text. b Refers to PTGIR-2 siRNA in the text.

airway epithelial cells in vitro (7). Moreover, another IP-recep- Materials and Methods tor agonist, beraprost, protected rats against the development of Culture of BEAS-2B cells experimental cigarette smoke-induced emphysema, possibly by Cells were cultured for 2 days under a 5% CO2/air atmosphere at 37°C in means of a concerted inhibitory action on alveolar epithelial 6- or 24-well plastic plates containing DMEM/F12 (Invitrogen) supple- cell apoptosis, oxidative burden, matrix metalloproteinase ex- mented with 10% FBS (Invitrogen), L-glutamine (2.5 mM), and 0.15% Downloaded from pression, and proinflammatory cytokine generation (15). Fi- (v/v) sodium bicarbonate. The cells were then growth-arrested for 24 h in nally, there is evidence from animal studies that IP-receptor serum-free medium (SFM). At this time, cultures were tightly confluent agonists protect against the deleterious effects of respiratory and were processed for biochemical and functional measurements as de- scribed below. syncytial virus infection (16). Potentially, this is an important observation because viral infections are believed to precipitate Culture of human primary airway epithelial cells (HpAECs) exacerbation in many individuals with asthma and COPD (17). Cells were obtained by proteinase digestion of nontransplanted normal hu- http://www.jimmunol.org/ Collectively, therefore, these findings suggest that selective IP- man lung (International Institute for the Advancement of Medicine, Edi- receptor agonists could be exploited to therapeutic advantage in son, NJ), as previously described (25). Cells were seeded in 12-well plates the treatment of airway diseases where chronic inflammation (Corning Life Sciences) containing bronchial epithelial cell growth me- dium (Lonza) supplemented with penicillin (50 ␮g/ml) and streptomycin with or without associated parenchymal destruction is a defin- ␮ (10 g/ml), cultured under a 5% CO2/air atmosphere at 37°C until con- ing characteristic (11, 14). fluent (typically 14 days; medium was changed every 3 to 4 days), growth Unlike asthma, COPD is an example of a chronic inflammatory arrested for 24 h in supplement-free, bronchial epithelial cell basal medium disease that is relatively insensitive to inhaled glucocorticoids (Lonza), and processed for biochemical and functional measurements as (ICSs) (18). Clearly, therefore, a drug that can boost or reveal described below. Ethics approval for the use of human tissues has been latent sensitivity to ICSs while having intrinsic anti-inflammatory granted by the Conjoint Health Research Ethics Board of the University of Calgary (Calgary, Alberta, Canada). by guest on October 1, 2021 and/or antiviral activity could be a significant pharmacotherapeutic advance. Repression of inflammatory gene expression by ICSs is Culture of human embryonic kidney (HEK)-293 Epstein-Barr believed to occur by at least two general mechanisms. The clas- nuclear Ag cells sical repressive mode of glucocorticoid action is termed trans- HEK-293 Epstein-Barr nuclear Ag cells expressing the human recombinant

repression, in which the activity of key proinflammatory transcrip- D prostanoid 1-receptor (DP1R-HEK), E prostanoid 2-receptor (EP2R- tion factors, such as NF-␬B and AP-1, is inhibited via direct HEK), E prostanoid 4-receptor (EP4R-HEK), or IP-receptor (IPR-HEK) interactions with the ligand-bound glucocorticoid receptor (GR) subtype were cultured for 2 days under a 5% CO2/air atmosphere at 37°C (19). However, in simple model systems, glucocorticoids often in 24-well plastic plates containing DMEM supplemented with 10% (v/v) FBS, geneticin (200 ␮g/ml), and hygromycin B (200 ␮g/ml). Confluent are relatively weak (partial) inhibitors of inflammatory gene cells were then growth-arrested in DMEM for 24 h under a 5% CO2 at- transcription, implying that processes in addition to trans-re- mosphere at 37°C in the absence of serum and antibiotics before processing pression must be operative to account for the anti-inflammatory for cAMP experiments as described below. effects seen in bona fide models of inflammation (20–23). In Generation of a glucocorticoid response element (GRE) this respect, the induction (trans-activation) by glucocorticoids reporter of anti-inflammatory genes that then repress proinflammatory processes is now believed to be a major mechanism of glu- Stable transfection was used to generate a GRE reporter cell line as de- cocorticoid action (20–23). Moreover, in the context of the scribed previously (26). The construct, pGL3.neo.TATA.2GRE, contains two copies of a consensus simple GRE site (sense strand, 5Ј-TGTACAG present study, cAMP-elevating agents can, in gene expression GATGTTCT-3Ј) positioned upstream of a minimal ␤-globin promoter studies, interact positively with glucocorticoids (20, 24). Ac- driving a luciferase gene and a separate neomycin gene to confer resistance cordingly, we have tested the hypothesis, using human bron- to geneticin (26). BEAS-2B cells at ϳ70% confluence in T162 flasks were ␮ ␮ chial epithelial cells as a model system, that selective agonism transfected with 8 g of plasmid DNA and 20 l of Lipofectamine 2000 (Invitrogen). After 24 h, geneticin (100 ␮g/ml) was added until foci of of the IP-receptor will enhance the anti-inflammatory activity of stable transfectants appeared that were harvested to create heterogeneous a glucocorticoid to a level that cannot be achieved by the glu- populations of cells in which the site of integration was randomized. cocorticoid alone. We submit that the demonstration of such a phenomenon could have clinical utility in the treatment of air- Transfection of BEAS-2B GRE reporter cells with siRNAs way inflammatory diseases, including COPD, that are either RNAiMax (Invitrogen) and the small interfering RNA (siRNA) of interest refractory or respond suboptimally to glucocorticoids. Such a (20 nM) (see Table I for siRNA oligonucleotide sequences) were diluted to finding would also provide a precedent for the idea that novel 2ϫ the desired final concentration with antibiotic-free, serum-free DMEM: F12 (Invitrogen) combined in a 1:1 ratio and left at room temperature for glucocorticoid combination therapies could be developed in 30 min. Subconfluent (ϳ70%) BEAS-2B GRE reporter cells in 24-well which the cAMP-elevating drug is tailored to the inflammatory plates were exposed to the siRNA/lipid mix for6hat37°C under a 5%

disease of interest. CO2/air atmosphere. The medium was then replaced with fresh DMEM/ 6790 TAPROSTENE AUGMENTS GRE-DEPENDENT TRANSCRIPTION

Table II. Primer pairs for real-time qPCRa titative PCR analysis of cDNA using the primer sequences shown in Table II (designed using Primer Express software; Applied Biosystems) encoding glucocorticoid-induced leucine zipper (GILZ; also known as TGF-␤-stim- Gene Oligonucleotide Accession No. ulated clone 22, domain family member 3 or TSC22D3), MAPK phospha- MKP-1 (DUSP1) tase (MKP)-1 (also known as dual-specificity phosphatase 1 or DUSP-1), Forward 5Ј-GCTCAGCCTTCCCCTGAGTA-3Ј NM_004417 and kinase inhibitor protein 2 of 57 kDa (p57kip2; also known as cyclin- Reverse 5Ј-GATACGCACTGCCCAGGTACA-3Ј dependent kinase inhibitor 1C or CDKN1C) was performed using an ABI kip2 p57 (CDKN1C) 7900HT instrument (Applied Biosystems) on 2.5 ␮l of cDNA in 20-␮l Ј Ј Forward 5 -CGGCGATCAAGAAGCTGTC-3 NM_000076 reactions using SYBR GreenER chemistry (Invitrogen) according to the Reverse 5Ј-GGCTCTAAATTGGCTCACCG-3Ј GILZ (TSC22D3) manufacturer’s guidelines. Relative cDNA concentrations were determined Forward 5Ј-GGCCATAGACAACAAGATCG-3Ј NM_001015881 from a cDNA standard curve that was analyzed simultaneously with the Reverse 5Ј-ACTTACACCGCAGAACCACCA-3Ј test samples. Amplification conditions were as follows: 50°C for 2 min and GAPDH 95°C for 10 min followed by 40 cycles of 95°C for 15 s and 60°C for 1 Forward 5Ј-TTCACCACCATGGAGAAGGC-3Ј NM_002046 min. Dissociation (melt) curves (95°C for 15 s and 60°C for 20 s with Ј Ј Reverse 5 -AGGAGGCATTGCTGATGATCT-3 ramping to 95°C over 20 min and then 95°C for 15 s) were constructed to a Forward and reverse primers for each gene are listed. Common genes names are confirm primer specificity. shown and gene symbols appear in brackets. Curve fitting Monophasic agonist concentration effect (E/[A]) curves were fitted by F12 supplemented with 1% FBS (v/v) and left for 42 h. Cells were then least-squares, nonlinear iterative regression to the following form of the growth arrested in SFM for a further 48 h before beginning the experiment. Hill equation (Prism 4; GraphPad Software) shown in Equation 1,

͑Emax Ϫ Emin͒ ϭ ϩ Downloaded from Treatment of GRE BEAS-2B reporter cells and measurement of E Emin ͑ ͓ ͔ Ϫ ͓ ͔͒n (1) 1 ϩ 10 p A 50 p A luciferase

where E is the effect, Emin and Emax are the lower and upper asymptotes Confluent, growth-arrested cells were treated with dexamethasone, tapro- (i.e., the basal response and maximum agonist-induced response, respec- stene, PGE2, and the IP-receptor antagonist RO3244794 (27) as indicated tively), p[A] the log molar concentration of agonist, p[A] a location pa- in the text and figure legends. Unless stated otherwise, cells were incubated 50 ϫ rameter equal to the log molar concentration of agonist producing Emax/2, at 37°C under a 5% CO2 atmosphere and harvested 5 h later in 1 reporter and n the gradient of the E/[A] curve at the p[A] level. lysis buffer (Promega). Luciferase activity was then measured using a 50

The antagonism of taprostene-induced responses by RO3244794 was http://www.jimmunol.org/ Monolight luminometer (BD Biosciences) according to the manufacturer’s evaluated by least-squares, nonlinear regression using a modification of the instructions. Data are expressed as the fold induction of luciferase relative Hill and Gaddum/Schild equations derived by Waud et al. (30). Each fam- to that of unstimulated cells. ily of E/[A] curves (i.e., the control E/[A] curve and E/[A] curves con- Infection of BEAS-2B Cells with Ad5.CMV.PKI␣ structed in the presence of increasing concentrations of RO3244794) were fitted simultaneously to Equation 2. Thus, Subconfluent (ϳ70%) BEAS-2B GRE reporter cells were infected (multi- Ϫ Ϫ ͑E Ϫ E ͒ plicity of infection (MOI) ϭ 40) with an E1 /E3 replication-deficient ϭ ϩ max min E Emin ͓ ͔ Ϫ n (2) adenovirus vector (Ad5.CMV.PKI␣) containing a 251-base pair DNA frag- 10p A 50͓1 ϩ ͓͑B͔/10 pA2͒S͔ ΂1 ϩ ͩ ͪ΃ ment encoding the complete amino acid sequence of the ␣-isoform of ͓A͔ cAMP-dependent protein kinase (PKA) inhibitor ␣ (PKI␣) downstream of the constitutively active CMV immediate early promoter (28). After 48 h, where [A] and [B] are the molar concentrations of taprostene and by guest on October 1, 2021 cells were processed for the assessment of GRE-dependent transcription as RO3244794 respectively, S is the Schild slope factor, which indicates the ϭ described above. To control for biological effects of the virus per se, the nature of antagonism, and pA2 is the affinity of the antagonist when S 1, vector Ad5.CMV.Null, expressing no transgene, was used in parallel. Us- which is equivalent to the pKb. To determine whether S deviated signifi- ing this experimental protocol, we have previously reported that Ͼ90% of cantly from unity, the entire family of E/[A] curves that made up an indi- BEAS-2B cells are infected with Ad5.CMV.PKI␣, resulting in the expres- vidual experiment was fitted globally to Equation 2 under two conditions: sion of a completely functional transgene with no adverse effect on cell one where S was constrained to a constant equal to 1 and the other where viability (28). it was a shared value for all data sets. The F test was applied to determine which equation gave the best fit, which was then used for the analysis. cAMP mass determination Assessment of cell viability BEAS-2B and HEK-293 cells in 24-well plates were growth arrested for 24 h and then pretreated (30 min) with the phosphodiesterase (PDE) in- Cell viability was evaluated colorimetrically by measuring the reduction of hibitors rolipram (10 ␮M) and siguazodan (10 ␮M). Prostanoid receptor the tetrazolium salt, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium agonists and antagonists were then added at the concentrations indicated in bromide, to formazan by mitochondrial dehydrogenases. the text and figure legends in the continued presence of PDE inhibitors. Drugs, Abs, and analytical reagents After 45 min of incubation at 37°C under 5% CO2, cells were lysed with HCl (100 mM) and cAMP mass was measured by enzyme immunoassay ONO-AE1–259 ((Z)-7-[(1R,2S,3R,5R)-5-chloro-3-hydroxy-2-[(E,4S)- (Cayman Chemical) according to the manufacturer’s instructions. 4-hydroxy-4-(1-prop-2-enylcyclo butyl)but-1-enyl]cyclopentyl]hept-5-enoic acid) and ONO-AE1–329 (2-[3-[(1R,2S,3R)-3-hydroxy-2-[(E,3S)-3-hydroxy- Measurement of CXCL10 5-[2-(methoxymethyl)phenyl]pent-1-enyl]-5-oxo-cyclopentyl]sulfanylpropyl sulfanyl]acetic acid) were from ONO Pharmaceuticals. RO3244794 ((R-3-(4- Growth-arrested BEAS-2B cells were pretreated (30 min) with GW9662 (1 ␮ ␥ fluoro phenyl)-2-[5-(4-fluorophenyl)-benzofuran-2-ylmethoxycarbonylami- M), a peroxisome proliferator-activated receptor (PPAR) antagonist no]propionic acid) was provided by Roche Pharmaceuticals. Iloprost, PGD , (29), or its vehicle followed by a PAR␥ agonist, rosiglitazone (1 ␮M), for 2 ␥ PGE2, BW245C (7-[3-(3-cyclohexyl-3-hydroxy-propyl)-2,5-dioxo-imidazoli- a further 30 min. IFN- (100 ng/ml; p[A]90) was then added and the cells din-4-yl]heptanoic acid), rosiglitazone, and GW9662 (2-chloro-5-nitrobenza- incubated at 37°C under a 5% CO2 atmosphere for 24 h. The amount of nilide) were obtained from Cayman Chemical and rolipram and siguazodan CXCL10 released into the culture supernatant was quantified by sandwich were purchased from Calbiochem (EMD Biosciences). Goat anti-human PKI␣ ELISA (Human DuoSet development system; R&D Systems) according to (code 1944) and goat anti-human ␤-actin (code 1615) were from Autogen the manufacturer’s instructions (see Ref. 7 for details). Bioclear. Taprostene, forskolin, and all other reagents were from Sigma- RNA isolation, reverse transcription, and real-time quantitative Aldrich. Drugs were dissolved in DMSO and diluted to the desired working concentration in the appropriate culture medium. PCR Definitions and statistics Total RNA was extracted from BEAS-2B cells and HpAECs that had been treated with dexamethasone (1 ␮M) and taprostene (1 ␮M) or forskolin (10 In the text, the term “additivity” refers to two drugs that, when combined, ␮M) alone or in combination using RNeasy mini kits (Qiagen) and was produce an effect that is the sum of their individual components. In con- reverse transcribed to cDNA as described previously (26). Real-time quan- trast, the term “positive cooperativity” is used when the biological response The Journal of Immunology 6791

FIGURE 1. Chemical structure of PGI2, taprostene, and RO3244794. .denotes chiral center. Optically active isomers are indicated (ء) Asterisk

of two drugs given in combination is greater than the sum of their indi- vidual effects. Data points and values in the text and figure legends represent the mean Ϯ SEM of N independent determinations. Data were analyzed by Student’s t test or ANOVA (one-way or two-way as indicated) followed, when appropriate, by Bonferroni’s multiple comparison test. The null hy- pothesis was rejected when p Ͻ 0.05. Results None of the compounds or their vehicles used in the experiments Downloaded from described herein significantly affected cell viability.

Selection of taprostene as an IP-receptor agonist The explicit classification of responses mediated by IP-receptors is hindered by a paucity of suitable pharmacological tools. In the http://www.jimmunol.org/ present study, we established that a synthetic PGI2 analog, tapro- stene (31), is a highly selectively agonist suitable for examining the (patho)physiological role of the IP-receptor subtype (Fig. 1). Thus, in IPR-HEK cells, taprostene increased the cAMP content in Ϯ a concentration-dependent manner with a p[A]50 of 12.7 0.2 (Fig. 2a). Relative to iloprost, taprostene was a full agonist in these cells (intrinsic activity (␣) ϭ 1.01; Fig. 2a). In contrast, taprostene failed to increase cAMP mass in HEK-293 cells expressing the

recombinant DP1-, EP2-, and EP4-receptor subtypes (all adenylyl by guest on October 1, 2021 cyclase coupled) at a concentration (1 nM) that maximally in- creased the cAMP level in IPR-HEK cells (Fig. 2, b–d). Indeed, even at a concentration of 10 ␮M, taprostene was weak, eliciting ␳ a cAMP response ( ) that was 13% of that produced by PGD2 in DP1R-HEK cells and 4.9 and 10% of that produced by PGE2

in EP2R-HEK and EP4R-HEK cells, respectively (Fig. 2, b–d). In contrast, at a concentration of 100 nM, BW245C (DP1-selective),

ONO-AE1–259 (EP2-selective), and ONO-AE1–329 (EP4-selec- tive) were full agonists in DP1R-HEK, EP2R-HEK, and EP4R- HEK cells, respectively (Fig. 2, b–d).

Effect of taprostene on GRE-dependent transcription Treatment of BEAS-2B reporter cells with dexamethasone (0.1 nM to 1 ␮M) for 5 h induced GRE-dependent transcription in a FIGURE 2. Effect of taprostene on cAMP accumulation in HEK-293 concentration-dependent manner with a p[A] and E of cells stably expressing adenylyl cyclase-coupled prostanoid receptor sub- 50 max types. Cells expressing the human recombinant IP-(IPR-HEK) (a), DP - 7.96 Ϯ 0.11 and 19.3 Ϯ 3.8-fold, respectively (Fig. 3a). Tap- 1 (DP R-HEK) (b), EP -(EP R-HEK) (c), and EP -receptor (EP R-HEK) (d) ␮ 1 2 2 4 4 rostene (1 M) alone was without effect on the GRE reporter were pretreated (30 min) with the PDE3 and PDE4 inhibitors and sigua- but, when added to cells concurrently with dexamethasone, sig- zodan (S; 10 ␮M) and rolipram (R; 10 ␮M), respectively, to prevent cAMP nificantly augmented GRE-dependent transcription above that hydrolysis. E/[A] curves to taprostene were then constructed (IP: from 100 ␮ produced by the glucocorticoid alone (Fig. 3a). Thus, the in- aM to 10 nM; DP1,EP2, and EP4: from 100 pM to 10 M) using cAMP teraction of taprostene and dexamethasone was one of positive and measured after 45 min as a biochemical output. The open and filled/ cooperativity. Analysis of the E/[A] relationship that described hatched bars in each panel represent the cAMP response in cells treated this effect showed that taprostene enhanced the magnitude with the PDE inhibitors in the absence or presence of the natural prosta- (E ) of transcription (from 19.3 Ϯ 3.8- to 37.3 Ϯ 5.5-fold) glandin and/or synthetic agonist, respectively, as indicated (each at 100 max ␣ without affecting the potency of dexamethasone (p[A] ϭ nM). The intrinsic activity ( ) of taprostene in IPR-HEK relative to iloprost 50 is indicated in a.Inb–d, the cAMP response (␳) of taprostene (10 ␮M) is 7.73 Ϯ 0.2) (Fig. 3a). Further studies established that in the shown relative to that of PGD2 (DP1) and PGE2 (EP2 and EP4), which are presence of a maximally effective concentration of dexameth- used as reference full agonists. The dashed line in each panel indicates the ␮ asone (1 M), the ability of taprostene to augment GRE-de- baseline cAMP level. Data points and bars represent the mean Ϯ SEM of pendent transcription was concentration-dependent, with a N independent determinations. Ϯ p[A]50 of 6.82 0.12 (Fig. 3b). 6792 TAPROSTENE AUGMENTS GRE-DEPENDENT TRANSCRIPTION

FIGURE 5. Schild analysis of the antagonism by RO3244794 of tap- rostene-induced responses in BEAS-2B cells. BEAS-2B GRE cells were pretreated (30 min) with RO3244794 at the concentrations indi- cated in each panel. Taprostene E/[A] curves were then constructed in the continued presence of RO3244794 for the enhancement of dexa- methasone (1 ␮M)-induced GRE-dependent transcription. Modified

Schild analysis was then performed (30) that yielded a pKb of 9.21. The filled bar and dashed line indicate the effect on luciferase expression of dexamethasone alone. Data points and bars represent the mean Ϯ SEM of five independent determinations. Downloaded from

(inactive by itself) augmented luciferase expression from 6- to 14- FIGURE 3. Effect of taprostene on GRE-dependent transcription. fold and from 10- to 18.5-fold, respectively. BEAS-2B GRE reporter cells were treated respectively with either dexa- methasone (0.1 to 1 ␮M) in the absence and presence of taprostene (1 ␮M) Affinity of RO3244794 for antagonizing taprostene-induced http://www.jimmunol.org/ ␮ (a) or with taprostene (1 nM to 10 M) in the presence of dexamethasone responses in BEAS-2B GRE reporter cells (1 ␮M) (b). After 5 h cells were lysed and luciferase activity was deter- mined. The filled bar and dashed line in b indicate the effect on luciferase Pretreatment of BEAS-2B GRE reporter cells with the selective expression of dexamethasone alone. Data points and bars represent the IP-receptor antagonist RO3244794 (10 and 30 nM; Ref. 27), had mean Ϯ SEM of N independent determinations. no effect on the expression of luciferase measured at5horonthe ability of dexamethasone to promote GRE-dependent transcription (data not shown). These data are thus consistent with the inability

Kinetics of the enhancement by taprostene of GRE-dependent of BEAS-2B cells to synthesize PGI2 (32). However, in dexameth- transcription asone (1 ␮M)-treated cells RO3244794 produced a graded, parallel by guest on October 1, 2021 Treatment of the BEAS-2B reporter with dexamethasone (1 ␮M) rightwards displacement (12- and 36-fold at 10 and 30 nM, re- induced the luciferase gene in a time-dependent manner (Fig. 4). spectively) of the taprostene E/[A] curve (Fig. 5). Enumeration of This effect reached a maximum at 5 h and was maintained for a the Schild slope factor S (which indicates the nature of the antag- further 11 h. Thereafter, the luciferase signal declined. In the pres- onism) by simultaneously fitting to Equation 2 each RO3244974 ence of taprostene (1 ␮M), which did not activate the reporter, the and agonist E/[A] curve indicated that this parameter did not de- induction by dexamethasone of GRE-dependent transcription was viate significantly from unity. Thus, RO3244794 behaved in a augmented at all time points (Fig. 4). The greatest positive coop- manner that was consistent with surmountable competitive antag- erativity was seen at the 2 and 5 h time points when taprostene onism (33). Accordingly, S was constrained to a value of 1 from Ϯ which a pKb value of 9.21 0.25 was derived. Effect of “silencing” PTGIR on the enhancement by taprostene of GRE-dependent transcription Lipid-mediated transfection of BEAS-2B GRE reporter cells with siRNAs (each 20 nM) directed against the gene of interest PTGIR, the gene in the jellyfish Aequorea victoria that encodes GFP or a universal negative control oligonucleotide (AllStars; Qiagen) that is reported to not recognize any human mRNA, affected neither basal nor dexamethasone (1 ␮M)-induced luciferase expression (Fig. 6). The transfection lipid RNAiMax was also inactive. Treatment of BEAS-2B reporter cells with taprostene (1 ␮M) or

PGE2 (100 nM) did not promote GRE-dependent transcription but FIGURE 4. Kinetics of GRE-dependent transcription in BEAS-2B GRE significantly augmented (from 13.0 Ϯ 0.4- to 24.9 Ϯ 2.9- and ␮ reporter cells. Cells were treated with dexamethasone (Dex; 1 M), tap- 25.4 Ϯ 1.9-fold, respectively) the transcriptional response pro- ␮ ␮ rostene (Tap; 1 M), a combination of both drugs (each 1 M), or vehicle duced by a maximally effective concentration of dexamethasone (1 (Control). Cells were then incubated for 1–19 h and harvested for luciferase ␮M). In cells transfected with siRNAs (PTGIR-1 and PTGIR-2) activity. Data points represent the mean Ϯ SEM of three independent de- -p Ͻ 0.05, significant enhancement of transcription relative that target PTGIR, the ability of taprostene to enhance GRE-de ,ء .terminations to time-matched, unstimulated cells; ϩ, p Ͻ 0.05, significant enhancement pendent transcription was abolished whereas RNAiMax alone and of transcription relative to time-matched, dexamethasone-treated cells. in the presence of the “control” siRNAs was without effect (Fig. 6). Data were analyzed by using two-way ANOVA followed by Bonferroni’s In contrast, the augmentation of dexamethasone-induced reporter

multiple comparison test. activity by PGE2 was not affected by PTGIR-1, PTGIR-2, the The Journal of Immunology 6793

FIGURE 6. Effect of silencing PTGIR by using siRNA on the enhance- ment by taprostene of GRE-dependent transcription. Subconfluent (70%) BEAS-2B GRE reporter cells were transfected (6 h) using RNAiMax with siRNAs (each 20 nM) directed against PTGIR, GFP, and a universal neg- ative control oligonucleotide. Cells were washed, incubated for 18 h in DMEM/F12 supplemented with 1% (v/v) FBS and then growth arrested for 48 h in SFM. The effect of dexamethasone (Dex: 1 ␮M) alone and in the ␮ presence of taprostene (Tap; 1 M) and PGE2 (100 nM) on GRE-de- pendent luciferase expression at 5 h was then determined. The dashed line indicates the effect on luciferase expression of dexamethasone alone. Bars represent the mean Ϯ SEM of seven independent determi- Downloaded from nations. See Materials and Methods for further details. NS, No stimu- p Ͻ 0.05, significant ,ء ;lation representing basal luciferase expression inhibition of the enhancement by taprostene of GRE-dependent tran- scription relative to cells transfected with a universal control oligonu- cleotide. Data were analyzed by using one-way ANOVA followed by FIGURE 7. Effect of taprostene on the cAMP/PKA cascade in BEAS-2B Bonferroni’s multiple comparison test. cells. a, BEAS-2B cells were pretreated (30 min) with the PDE3 and PDE4 ␮ ␮

inhibitors and siguazodan (S; 10 M) and rolipram (R; 10 M), re- http://www.jimmunol.org/ spectively, to prevent cAMP hydrolysis and then exposed to taprostene (0.1 nM to 10 ␮M) for 45 min. Cells were lysed and cAMP mass was “control” siRNAs, or RNAiMax (Fig. 6), thereby confirming the measured by enzyme immunoassay according to the manufacturer’s in- selectivity of the knockdown. structions (Cayman Chemical). b, Subconfluent cells were infected with Ad5.CMV.PKI␣, Ad5.CMV.Null (both at MOI ϭ 40), or left untreated Role of the cAMP/PKA pathway in the enhancement by for 48 h. Cells were then treated with dexamethasone (Dex; 1 ␮M) taprostene of GRE-dependent transcription alone (filled bar) or taprostene (1 nM to 10 ␮M) in the presence of dexamethasone (1 ␮M). After 5 h cells were lysed and luciferase ac- The IP-receptor couples typically to Gs for the stimulation of ad- enylyl cyclase, although in some systems it can also promote phos- tivity was determined. c, Western blot, representative of three experi- ␣ by guest on October 1, 2021 pholipase C-dependent inositol phosphate production (34, 35). To ments, of the expression of the PKI transgene in BEAS-2B cells 48 h after infection with Ad5.CMV.PKI␣. The filled bar and dashed line in establish the mechanism by which activation of the IP-receptor b indicate the effect on luciferase expression of dexamethasone alone. enhanced GRE-dependent transcription, BEAS-3B cells were ex- Data points and bars represent the mean Ϯ SEM of N independent posed to taprostene and cAMP mass was determined. As shown in determinations. Lane 1, Naive cells; lane 2, Ad5.CMV.Null-infected Fig. 7a, taprostene increased cAMP in a concentration-dependent cells; lane 3, Ad5.CMV.PKI␣-infected cells. Ϯ manner with a p[A]50 of 8.01 0.01. In subsequent experiments, BEAS-2B GRE reporter cells were infected with an adenovirus vector, Ad5.CMV.PKI␣, which directs overexpression of PKI␣,a GW9662 as a selective antagonist (29), studies were performed highly selective inhibitor of PKA (28). In uninfected cells, PKI␣ to establish whether PPAR␥ mediated the augmentation by tap- was not detected by Western blotting in any experiment. However, rostene of dexamethasone-induced GRE-dependent transcrip- 48 h after infection of BEAS-2B cells with Ad5.CMV.PKI␣ tion. GW9662 (1 ␮M) had no effect on basal or dexamethasone- (MOI ϭ 40), a single peptide was labeled by the anti-PKI␣ Ab that induced luciferase expression (Fig. 8a). Similarly, GW9662 did migrated as a 12-kDa band on SDS-polyacrylamide gels (Fig. 7c). not affect the E/[A] relationship that described the enhancement As shown in Fig. 7b, taprostene enhanced dexamethasone (1 ␮M)- by taprostene of GRE-dependent transcription (Fig. 8a). Indeed, Ϯ Ϯ induced GRE-dependent transcription in a concentration-depen- neither the p[A]50 (6.68 0.07) nor Emax (38.2 5.5-fold) ϭ Ϯ ϭ Ϯ ϭ Ϯ ϭ dent manner (p[A]50 6.91 0.14; Emax 24.0 0.9-fold), were altered by GW9662 (p[A]50 6.81 0.19; Emax which was prevented in cells expressing the PKI␣ transgene (Fig. 39.3 Ϯ 6.3-fold). In contrast, GW9662 (1 ␮M) abolished the 7b). In contrast, cells infected with a null virus, Ad5.CMV.Null, inhibitory effect of rosiglitazone, a PPAR␥ agonist, on IFN-␥- responded to taprostene in a manner that was not significantly induced CXCL10 output from BEAS-2B cells under similar ex- ϭ Ϯ ϭ different from uninfected cells (p[A]50 6.86 0.08; Emax perimental conditions (Fig. 8b). 26.1 Ϯ 1.6-fold; Fig. 8b). Forskolin, a direct activator of adenylyl cyclase, also augmented GRE-dependent transcription by 2.78- Effect of taprostene and forskolin on the expression of fold (from 25.0 Ϯ 5.8 to 69.5 Ϯ 11.3; n ϭ 5), and this effect also glucocorticoid-inducible anti-inflammatory genes in BEAS-2B was abolished in cells expressing the PKI␣ transgene (data not cells shown). The data presented in the preceding sections indicate that tapro- stene enhanced dexamethasone-induced transcription from a con- ␥ Role of PPAR in the enhancement by taprostene of ventional simple GRE reporter. To determine whether this finding GRE-dependent transcription is applicable to real genes, we took advantage of data derived from Agonism of the human IP-receptor subtype is reported to activate a prior microarray analysis in which dexamethasone-inducible the nuclear hormone receptor PPAR␥ (36). Accordingly, using genes were identified in pulmonary type II A549 cells (see Ref. 20 6794 TAPROSTENE AUGMENTS GRE-DEPENDENT TRANSCRIPTION

FIGURE 8. Effect of the PPAR␥ antagonist GW9662 on the enhance- ment by taprostene of GRE-dependent transcription and on the suppression by rosiglitazone of IFN-␥-induced CXCL10 release. a, BEAS-2B GRE reporter cells were pretreated with GW9662 (1 ␮M; 30 min) or vehicle. Dexamethasone (1 ␮M) was then added alone (bars) or in combination with taprostene (1 nM to 10 ␮M). After 5 h cells were lysed and luciferase activity was determined. b, BEAS-2B cells were pretreated with GW9662

(1 ␮M; 30 min) or vehicle and exposed to rosiglitazone (100 nM) for a Downloaded from further 30 min. IFN-␥ (100 ng/ml) was then added and the amount of CXCL10 released into the supernatant at 24 h was measured by ELISA. The bars and dashed line in panel a indicate the effect on the luciferase expression of dexamethasone (Ϯ GW9662) alone. Data points and bars ,p Ͻ 0.05 ,ء .represent the mean Ϯ SEM of N independent determinations p Ͻ 0.05, significant reversal ,ءء ;significant inhibition of CXCL10 release of CXCL10 release by GW9662. Data were analyzed by using one-way FIGURE 9. Effect of taprostene and forskolin on the expression of glu- http://www.jimmunol.org/ ANOVA followed by the Bonferroni multiple comparison test. cocorticoid-inducible genes in BEAS-2B cells. Cells were treated with dexamethasone (Dex; 1 ␮M), taprostene (Tap; 1 ␮M), both agents simul- taneously (Dex ϩ Tap), or vehicle (control), and RNA was harvested at 1, 2,and6h.(a, c, and e) A similar experiment was also performed in which for details). Several of these genes have anti-inflammatory poten- taprostene was replaced with forskolin (Forsk; 10 ␮M) (b, d, and f). After tial, including GILZ, MKP-1, and p57kip2 (22, 37–40), and were cDNA synthesis, real-time RT-PCR was performed for GILZ (a and b), selected to examine the interaction between dexamethasone and p57kip2 (c and d), and MKP-1 (e and f) and normalized to GAPDH. Data are taprostene (or forskolin). expressed as fold stimulation at each time point relative to GAPDH and are The data derived from the GRE reporter showed substantial pos- plotted as means Ϯ SEM of six independent observations. The horizontal by guest on October 1, 2021 itive cooperativity between dexamethasone (1 ␮M) and taprostene dashed line in each panel indicates basal gene expression at the 0-h time (1 ␮M) at 1 h, 2 h and 5 h. Accordingly, the expression of GILZ, point, which has been normalized to a value of 1. The nature of the inter- kip2 action between dexamethasone and the cAMP-elevating agent at one or Ͻ ء -MKP-1 and p57 were examined over the same time-frame. Rel ␮ more time points is indicated in each panel. , p 0.05, significant en- ative to unstimulated cells, dexamethasone (1 M) strongly in- hancement of transcription relative to time-matched, unstimulated cells; ϩ, duced GILZ in a time-dependent manner; the mRNA level peaked p Ͻ 0.05, significant enhancement of transcription relative to time- at2h(ϳ75-fold induction) and remained elevated up to 6 h (Fig. matched, dexamethasone-treated cells. Data were analyzed by using two- 9a). In contrast, taprostene (1 ␮M) was inactive over the same time way ANOVA followed by Bonferroni’s multiple comparison test. frame but augmented the transcriptional activity of dexamethasone at1and2hinamanner reminiscent of that produced in GRE reporter cells (Fig. 9a). Similarly, forskolin, a direct activator of and forskolin (10 ␮M) also induced MKP-1 (maximally by 2.1- adenylyl cyclase, failed to induce GILZ at any time point but, like and 4.1-fold, respectively, at 1 h), but this effect also was short- taprostene, enhanced the effect of dexamethasone (Fig. 9b). Thus, lived and was essentially lost by 2 h (Fig. 9, e and f). When used taprostene and forskolin interacted with dexamethasone in the reg- in combination, dexamethasone (1 ␮M) and taprostene (1 ␮M) or ulation of GILZ in a positive cooperative fashion. dexamethasone (1 ␮M) and forskolin (10 ␮M) at 1 h interacted in p57kip2 also was rapidly induced by dexamethasone (1 ␮M) in a purely additive fashion (Fig. 9, e and f). At later time points (2 a time-dependent manner, although the inducibility and kinetics of and 6 h) the expression of MKP-1 induced by both drug combi- the response were different from those for GILZ. Thus, mRNA nations was not different from the effect produced by dexametha- levels increased almost linearly over the course of the experiment sone alone (Fig. 9, e and f). (ϳ12-fold at 6 h) and were significantly lower than the GILZ tran- scripts matched for time (Fig. 9c). Taprostene (1 ␮M) alone had a Effect of taprostene and forskolin on the expression of negligible effect on the expression of p57kip2 (ϳ1.4-fold induction) glucocorticoid-inducible anti-inflammatory genes in HpAECs but augmented the transcriptional activity of dexamethasone at all Additional studies were performed to determine whether the drug time points. Qualitatively comparable data were obtained when interactions found in BEAS-2B cells also occurred in HpAECs. To forskolin was used as a cAMP-elevating stimulus (Fig. 9d), but the this end, GILZ and MKP-1 were selected as genes that were reg- magnitude of this positive cooperative interaction was far more ulated by glucocorticoid and a cAMP-elevating agent in a positive pronounced and greatest at the 6-h time point (ϳ43- vs ϳ17-fold cooperative and additive manner, respectively. As shown in Fig. for taprostene; compare GILZ, Fig. 9b). 10, a and b, dexamethasone (1 ␮M) induced GILZ in a time-de- The third gene examined, MKP-1, was induced by dexametha- pendent manner. Consistent with the data obtained with BEAS-2B sone (1 ␮M) in a transient manner; mRNA levels peaked at the 2-h cells (Fig. 9, a and b), both taprostene (1 ␮M) and forskolin (10 time point and then slowly decayed (Fig. 9e). Taprostene (1 ␮M) ␮M) were inactive but, in the presence of dexamethasone (1 ␮M), The Journal of Immunology 6795

of certain glucocorticoid-inducible genes in BEAS-2B cells and HpAECs, including GILZ and MKP-1, which have potential anti- inflammatory activity. These findings support and extend the existing body of knowledge, by demonstrating that IP-receptor agonists may have utility in treating chronic airway inflamma- tion by boosting the clinical efficacy of glucocorticoids as part of an IP-receptor agonist/ICS combination therapy. Selection of taprostene as an IP-receptor agonist A major problem in classifying responses mediated by IP-recep- tors has been a paucity of suitable pharmacological tools. Indeed,

PGI2 and many PGI2 analogs and mimetics, including iloprost, carbacyclin, AFP-07, and TEI-9063, are not sufficiently selective for their biological actions to be diagnostic of IP-receptor agonism (41, 42). Even cicaprost, which is often the agonist of choice in studies examining IP-receptor pharmacology, must be used with

caution at it does not definitively discriminate IP-, EP4-, and, to a lesser extent, EP3-receptor-mediated responses (41, 42). To over- FIGURE 10. Effect of taprostene (Tap) and forskolin (Forsk) on the come these problems, we used, in the present study, a synthetic expression of glucocorticoid-inducible genes in HpAECs. Cells were PGI analog, taprostene (31). Two sets of data led us to select this Downloaded from treated and cDNA prepared as described in the legend to Fig. 9. Real-time 2 ligand. First, taprostene showed high selectivity for the human RT-PCR was performed for GILZ (a and b) and MKP-1 (c and d) and normalized to GAPDH. Data are expressed as fold stimulation at each time recombinant IP-receptor expressed in HEK-293 cells relative to point relative to GAPDH and are plotted as means Ϯ SEM of three inde- the other prostanoid receptors that couple positively to adenylyl pendent observations. The horizontal dashed line in each panel indicates cyclase. Although absolute selectivity ratios cannot be calculated ϳ basal gene expression at the 0-h time point, which has been normalized to from these data, taprostene at its p[A]95 (i.e., 100 pM) was in- http://www.jimmunol.org/ a value of 1. The nature of the interaction between dexamethasone (Dex) active at the DP1-, EP2-, and EP4-subtypes (Fig. 2). Second, the and a cAMP-elevating agent at one or more time points is indicated in each ability of taprostene to enhance GRE-dependent transcription in p Ͻ 0.05, significant enhancement of transcription relative to BEAS-2B reporter cells was antagonized in a competitive manner ,ء .panel ϩ Ͻ time-matched, unstimulated cells; , p 0.05, significant enhancement of by RO3244794, an IP-receptor-blocking drug, with an affinity transcription relative to time-matched, dexamethasone-treated cells. Data (pK ϭ 9.21) consistent with an interaction at IP-receptors (27, were analyzed by using two-way ANOVA followed by Bonferroni’s mul- b 43). Collectively, these results strongly suggest that in BEAS-2B tiple comparison test. cells, taprostene augments GRE-dependent transcription by an IP- receptor-mediated mechanism. This conclusion was confirmed in gene silencing studies where, under stringently controlled condi- by guest on October 1, 2021 augmented (markedly at 6 h) the expression of GILZ. However, tions, two siRNAs that target PTGIR abolished the ability of tap- several notable differences were apparent. First, GILZ was at least rostene to enhance GRE-dependent transcription, whereas the an order of magnitude less inducible in HpAECs at all time points same response evoked by PGE2 was preserved. (compare Fig. 10a with Fig. 9a). Second, although taprostene and dexamethasone interacted cooperatively in HpAECs, this was seen Interaction of taprostene and dexamethasone only at the 6-h time point whereas cooperativity was seen at all Using a simple GRE-reporter construct stably transfected into time points when forskolin and dexamethasone were combined BEAS-2B cells, taprostene augmented the ability of dexameth- (compare Fig. 10, a and b with Fig. 9, a and b). Third, the kinetics asone to promote the induction of the luciferase gene in a time- of GILZ induction in response to dexamethasone in combination and concentration-dependent manner. Inspection of the E/[A] with taprostene or forskolin appeared to be slower in HpAECs curves showed that taprostene increased the ability of dexa- when compared with BEAS-2B cells (compare Fig. 10, a and b methasone to “drive” the luciferase gene (i.e., Emax was in- with Fig. 9, a and b). creased) without significantly affecting its potency (i.e., p[A]50 In HpAECs the MKP-1 gene was also induced by dexametha- was unchanged). Of potential clinical significance is that this sone (1 ␮M) and, additionally, by taprostene (1 ␮M) and forskolin interaction is one of positive cooperativity, because taprostene (10 ␮M; Fig. 10, c and d). For all three stimuli, MKP-1 mRNA alone did not promote GRE-dependent transcription (see be- appeared very transiently (more so than in BEAS-2B cells); low). Moreover, taprostene was “steroid-sparing” in this model peak levels were achieved at 1 h and had declined to baseline at system. For example, dexamethasone, at its p[A]95, produced a 2 h. In combination, taprostene and dexamethasone, or forsko- 17.4-fold induction of the luciferase gene. However, in the pres- lin and dexamethasone interacted additively on MKP-1 expres- ence of taprostene (1 ␮M), which was inactive by itself, the sion with similarly transient kinetics (Fig. 10, c and d). In con- same degree of gene induction was achieved at a concentration trast to GILZ, the inducibility of MKP-1 in HpAECs was similar of dexamethasone that was significantly lower (7.3-fold if the to that found in BEAS-2B cells (compare Fig. 10, c and d with p[A]95 of the glucocorticoid response alone is taken as refer- Fig. 9, e and f). ence) (see Fig. 11). Prior microarray analysis of genes induced by dexamethasone in Discussion A549 cells identified several candidates for subsequent analysis The results of the present study show that the selective agonism that are predicted to exert anti-inflammatory activity. The three of the IP-receptor on the human bronchial epithelial cell line genes studied here were found to be regulated in an additive or BEAS-2B with taprostene enhanced dexamethasone-induced positive cooperative manner by glucocorticoid and taprostene in transcription of a simple GRE reporter by activating the cAMP/ combination. Thus, GILZ and p57kip2 were induced in a positive PKA cascade. Similarly, taprostene augmented the expression cooperative fashion in BEAS-2B, and a similar effect was seen 6796 TAPROSTENE AUGMENTS GRE-DEPENDENT TRANSCRIPTION

an IP-receptor agonist, thereby limiting the clinical efficacy of such a combination therapy. For example, the expression of metabolic genes such as phosphoenol pyruvate carboxykinase and glucose- 6-phosphastase, which regulate gluconeogenesis, are induced by cAMP-elevating stimuli and glucocorticoids and, in combination, act at least additively (49–51). Clearly, therefore, the pharmaco- kinetics of a glucocorticoid will dictate the propensity to which such adverse effect genes may be induced. In respiratory diseases, a glucocorticoid should be delivered directly to the lung and have low oral absorption, high plasma protein binding to limit systemic exposure, and high first-pass hepatic metabolism. Under these con- FIGURE 11. Interaction between taprostene and dexamethasone on ditions, the ability of an IP-receptor agonist to enhance the tran- GRE-dependent transcription. The graph in Fig. 3a has been redrawn to scriptional activity of that glucocorticoid should be retained pre- illustrate that taprostene augments dexamethasone-induced GRE reporter dominantly in the lung, allowing the superior clinical efficacy of activity in a steroid-sparing manner. BEAS-2B cells stably expressing a the combination therapy to be achieved. GRE-reporter were treated with taprostene (1 ␮M) in the absence otr pres- ence of dexamethasone (100 pM to 1 ␮M). After 6 h, cells were harvested Mechanism of action for luciferase assay. The data show that the effect of taprostene and dexa- It has been reported that PGI2 analogs, including carbacyclin and methasone in combination promotes GRE-dependent transcription above treprostinil, are PPAR␥ agonists (36). However, no evidence for the maximum effect achieved by the glucocorticoid alone (1.93-fold at the Downloaded from such a mechanism was obtained in the present study based on the E ). Taprostene was also glucocorticoid sparing in this model. Thus, max ␥ dexamethasone at a concentration that evoked 95% of the maximum re- inability of GW9662, a PPAR antagonist (29), to inhibit tapro- sponse produced a 19.3-fold induction of the luciferase gene. However, in stene-induced transcriptional responses. Indeed, our data suggest the presence of taprostene (1 ␮M), which was inactive by itself, the same that the mechanistic basis for the enhancement by taprostene of degree of gene induction was achieved at a concentration of glucocorticoid GRE-reporter activity involves the activation of the classical that was significantly (7.3-fold) lower. cAMP/PKA pathway. Thus, this effect was mimicked by forskolin and abolished in cells infected with Ad5.CMV.PKI␣. Similarly, http://www.jimmunol.org/ the ability of forskolin to interact either additively or cooperatively with GILZ in HpAECs. Although there were differences in the with dexamethasone on the expression of GILZ, p57kip2, and magnitude and kinetics of gene induction between BEAS-2B cells MKP-1 in BEAS-2B cells and HpAECs confirms that cAMP can, and HpAECs, both cell types responded at certain time points in a in some way, augment the transcription of real glucocorticoid- manner that resembled activation of the GRE reporter. Indeed, a inducible genes. kip2 ␤ functional conventional simple GRE is present in the p57 and We have reported previously that salmeterol, a 2-adreno- GILZ promoters (44, 45), which implies that taprostene and dexa- ceptor agonist, and forskolin also augment GRE-dependent methasone in combination can induce in a positive cooperative transcription in BEAS-2B cells, but only a certain, undefined by guest on October 1, 2021 manner real anti-inflammatory genes. GILZ has clear anti- population of glucocorticoid-inducible genes are affected (20). inflammatory potential through its ability to attenuate the activa- Indeed, genes that encode tristetraprolin, aminopeptidase N, tion of NF-␬B and/or AP-1, which are critical transcription factors and plasminogen activator inhibitor-1 are not regulated in an that control the expression of many proinflammatory genes (37, additive or positive cooperative manner by either of these kip2 ␤ 38). Similarly, p57 encodes a cell cycle kinase inhibitor that cAMP-elevating agents (20). Current data indicate that 2-ad- regulates the anti-proliferative effects of glucocorticoids (39). renoceptor agonists and forskolin display an identical, qualita- Given that airway remodeling in COPD could be explained, in tive profile of activity on a panel of glucocorticoid-inducible part, by the increase in airway wall volume occupied by bronchial genes (20). Whether taprostene shares this defined activity or epithelial cells as well as smooth muscle (46), it is possible that an displays a distinct or overlapping profile is unknown. Although IP-receptor agonist/ICS combination therapy could be disease the present study was undertaken to establish the concept that modifying through its ability to promote the expression of p57kip2. an IP-receptor agonist can augment GRE-dependent trans-acti- The p57kip2 gene product also blocks JNK (47) suggesting that it vation, this is nevertheless an important objective for future may suppress proinflammatory events known to be regulated by research. Indeed, it is conceivable that different cAMP-elevat- this MAPK. ing drugs may enhance the transcription of distinct populations The third gene examined was MKP-1. This gene encodes an of glucocorticoid-inducible genes. Clearly, this raises the pos- enzyme that dephosphorylates (inactivates) the three core mam- sibility of selectively regulating the expression of certain genes, malian MAPKs (p38 MAPK, ERK, and JNK) that are central to which would have clear therapeutic potential. the induction of many proinflammatory genes (e.g., growth-related The identity of the targets downstream of PKA that more oncogene-␣) (Ref. 40). As shown in Figs. 9 and 10, the regulation directly up-regulate GRE-dependent transcription is vague. It of MKP-1 was distinct from that of p57kip2 and GILZ. Thus, has been proposed that cAMP-elevating agents enhance the MKP-1 was induced by both taprostene and dexamethasone, and in translocation of the GR from the cytosol to the nucleus and that combination the interaction of these stimuli was transient and this boosts transcription (52–54). Indeed, the cAMP-elevating purely additive. These data indicate that that unlike p57kip2 and drug formoterol augments the ability of budesonide to promote GILZ, the glucocorticoid-dependent regulation of MKP-1 is not the binding of the GR to GREs on target genes above the level primarily mediated via simple GREs, which is consistent with the produced by the glucocorticoid alone (55). Such an effect would finding that the promoter of this and many other glucocorticoid- be consistent with the enhanced GR:DNA binding seen in cells inducible genes do not feature a conventional GRE consensus se- overexpressing PKA (56). Another possibility is that cAMP- quence (i.e., sense strand, 5Ј-TGTACAGGATGTTCT-3Ј) (48). elevating agents increase the expression of functional GR (57) It is noteworthy that the transcription of glucocorticoid-induc- and/or stabilize the interaction of glucocorticoid-bound GR ible genes that mediate adverse effects may also be augmented by with DNA through activation of PKA (58). Although these are The Journal of Immunology 6797 plausible ideas, both mechanisms necessarily imply that the event that IP-receptor agonists are shown to be efficacious in transcription of all glucocorticoid-inducible genes would be en- treating the cardiovascular abnormalities in COPD, a drug ex- hanced by a cAMP-elevating drug, which is not the case (20). emplified by NS-304 could be suitable for twice a day dosing To account for this, we propose that taprostene and other with an existing ICS. cAMP-elevating agents may act predominantly within the nu- Conclusions cleus by regulating the activity and/or recruitment of specific cofactors that affect only the transcription of a specific subset of In the present study we report that a selective IP-receptor ag- glucocorticoid-inducible genes (59–61). Although this pro- onist, taprostene, enhances the ability of a glucocorticoid to posal remains to be investigated, it would confer promoter spec- transcribe genes, in airway epithelial cells, with potential anti- ificity and explain why the expression of some glucocorticoid- inflammatory activity. This novel mechanism of action has clear inducible genes is augmented in a positive cooperative fashion clinical relevance in diseases, such as COPD, that are generally by cAMP-elevating agents whereas others are not. refractory to glucocorticoids as a monotherapy. Moreover, these

data complement a growing body of literature in which PGI2 Targeting the IP-receptor in COPD analogs are reported to have both anti-inflammatory and anti- COPD is a multifaceted disorder. Typically, neutrophilic inflam- viral activity in preclinical models of airway inflammation (see mation of the small airways and lungs is a defining pathological Introduction). We submit that because activation of the IP-re- feature that is associated with chronic airflow limitation (60). ceptor also elicits beneficial effects on platelet reactivity, pul- Many individuals with COPD also have systemic inflammation monary vascular smooth muscle tone. and, indirectly, cardiac (62) that is positively related to disease severity (63). Pulmonary function, ICS/IP-receptor agonist combination therapy could vascular remodeling leading to impaired gas exchange and/or pul- provide an effective treatment option in COPD that may be Downloaded from monary hypertension is also frequently seen. In many cases, such particularly effective in subjects with hematological and car- abnormalities occur in conjunction with right-side heart failure diovascular dysfunction. (64, 65) and platelet hyper-reactivity, rendering afflicted individ- uals at increased risk of developing thromboses (66–68). Unlike Disclosures asthma, stable COPD is an example of a chronic inflammatory The authors have no financial conflict of interest. disease that is relatively insensitive to ICSs (18). Conceivably, an http://www.jimmunol.org/ “add-on” therapy that is known to effectively target the cardiovas- References cular pathologies and concurrently augment the anti-inflammatory 1. Vane, J. R., and R. M. Botting. 1995. Pharmacodynamic profile of prostacyclin. Am. J. Cardiol. 75: 3A–10A. effect of an ICS could provide an effective and potentially disease- 2. Hata, A. N., and R. M. Breyer. 2004. Pharmacology and signaling of prostaglan- modifying treatment. We submit that a selective IP-receptor ago- din receptors: multiple roles in inflammation and immune modulation. Pharma- col. Ther. 103: 147–166. nist could fulfill this role for several tangible reasons. In particular, 3. Meja, K. K., P. J. Barnes, and M. A. Giembycz. 1997. Characterization of the the expression of PGI2 synthase and the level of 6-keto-PGF1␣, the prostanoid receptor(s) on human blood monocytes at which prostaglandin E2 primary metabolite of PGI , are reduced in whole lung lysates inhibits lipopolysaccharide-induced tumour necrosis factor-␣ generation. 2 Br. J. Pharmacol. 122: 149–157. prepared from subjects with emphysema (69). Conversely, the 4. Jaffar, Z., K. S. Wan, and K. Roberts. 2002. A key role for prostaglandin I2 in by guest on October 1, 2021 J. Immu- urinary excretion of 11-dehydro-thromboxane (TX) B2, the ma- limiting lung mucosal Th2, but not Th1, responses to inhaled allergen. jor enzymatic metabolite of TXA in subjects with COPD, is nol. 169: 5997–6004. 2 5. Zhou, W., T. S. Blackwell, K. Goleniewska, J. F. O’Neal, G. A. FitzGerald, significantly enhanced relative to that in healthy individuals M. Lucitt, R. M. Breyer, and R. S. Peebles, Jr. 2006. Prostaglandin I2 analogs matched for age and gender (70). Thus, there appears to be a inhibit Th1 and Th2 effector cytokine production by CD4 T cells. J. Leukocyte Biol. 81: 809–887. PGI2 deficiency in emphysematous COPD and a reciprocal in- 6. Zhou, W., K. Hashimoto, K. Goleniewska, J. F. O’Neal, S. Ji, T. S. Blackwell, crease in TXA2 generation with predicted undesirable effects on G. A. FitzGerald, K. M. Egan, M. W. Geraci, and R. S. Peebles. 2007. Prosta- platelet reactivity and pulmonary vascular smooth muscle func- glandin I2 analogs inhibit proinflammatory cytokine production and T cell stim- ulatory function of dendritic cells. J. Immunol. 178: 702–710. tion. Clearly, a selective IP-receptor agonist could restore this 7. Ayer, L. M., S. M. Wilson, S. L. Traves, D. Proud, and M. A. Giembycz. H PGI2/TXA2 imbalance and, thereby, reduce the threshold for 2008. (4,5-Dihydro-1 -imidazol-2-yl)-[4-(4-isopropoxy-benzyl)-phenyl]- platelet aggregation (71), lower pulmonary vascular resistance amine (RO1138452) is a selective, pseudo-irreversible orthosteric antagonist at the prostacyclin (IP)-receptor expressed by human airway epithelial cells: (72) (through smooth muscle relaxation and, long term, by in- negative IP-receptor-mediated regulation of CXCL9 and CXCL10 release. hibiting mitogenesis), and even promote bronchodilatation (73). J. Pharmacol. Exp. Ther. 324: 815–826. 8. Clarke, D. L., M. G. Belvisi, S. J. Smith, E. Hardaker, M. H. Yacoub, K. K. Meja, The results of this study indicate that an IP-receptor agonist R. Newton, D. M. Slater, and M. A. Giembycz. 2005. Prostanoid receptor ex- could also suppress airway and pulmonary vascular inflamma- pression by human airway smooth muscle cells and regulation of the secretion of tion by enhancing the clinical efficacy of an ICS. It is notewor- granulocyte colony-stimulating factor. Am. J. Physiol. 288: L238–L250. 9. Walch, L., C. Labat, J. P. Gascard, M. de, V., C. Brink, and X. Norel. 1999. thy that IP-receptor agonists are effective in the treatment of Prostanoid receptors involved in the relaxation of human pulmonary vessels. pulmonary arterial hypertension (74). However, there have been Br. J. Pharmacol. 126: 859–866. few controlled trials in pulmonary hypertension associated with 10. Jozefowski, S., M. Bobek, and J. Marcinkiewicz. 2003. Exogenous but not en- dogenous prostanoids regulate cytokine secretion from murine bone marrow den- COPD (65), and the suitability of these drugs for treating car- dritic cells: EP2, DP, and IP but not EP1,EP3, and FP prostanoid receptors are diovascular defects in this pathological setting would need to be involved. Int. Immunopharmacol. 3: 865–878. 11. Takahashi, Y., S. Tokuoka, T. Masuda, Y. Hirano, M. Nagao, H. Tanaka, tested empirically in well-designed clinical trials. N. Inagaki, S. Narumiya, and H. Nagai. 2002. Augmentation of allergic inflam- Recently, there has been a considerable effort in medicinal mation in prostanoid IP-receptor deficient mice. Br. J. Pharmacol. 137: 315–322. 12. Nagao, K., H. Tanaka, M. Komai, T. Masuda, S. Narumiya, and H. Nagai. 2003. chemistry to develop new PGI2 mimetics for pulmonary arterial Role of prostaglandin I2 in airway remodeling induced by repeated allergen chal- hypertension that are both stable and selective for the IP sub- lenge in mice. Am. J. Respir. Cell Mol. Biol. 29: 314–320. type. For example, chemists at Nippon Shinyaku have discov- 13. Jaffar, Z., M. E. Ferrini, M. C. Buford, G. A. FitzGerald, and K. Roberts. 2007. Prostaglandin I2-IP signaling blocks allergic pulmonary inflammation by pre- ered a number of nonprostanoid pro-drugs that are metabolized ϩ venting recruitment of CD4 Th2 cells into the airways in a mouse model of in vivo into potent and selective IP-receptor agonists (75). One asthma. J. Immunol. 179: 6193–6203. of these, NS-304, is an N-acylsulphonamide that is slowly hy- 14. Idzko, M., H. Hammad, M. van Nimwegen, M. Kool, N. Vos, H. C. Hoogsteden, and B. N. Lambrecht. 2007. Inhaled iloprost suppresses the cardinal features of drolyzed by the liver to produce the corresponding carboxylic asthma via inhibition of airway dendritic cell function. J. Clin. Invest. 117: acid MRE-269, which has a t1/2 in blood of 7.9 h (75). In the 464–472. 6798 TAPROSTENE AUGMENTS GRE-DEPENDENT TRANSCRIPTION

15. Chen, Y., M. Hanaoka, P. Chen, Y. Droma, N. F. Voelkel, and K. Kubo. 2009. 42. Wise, H., and R. L. Jones. 2000. Prostacyclin and Its Receptors. Kluwer Aca- Protective effect of beraprost sodium, a stable prostacyclin analog, in the devel- demic Publishers, Hingham, MA. opment of cigarette smoke extract-induced emphysema. Am. J. Physiol. 296: 43. Jones, R. L., M. A. Giembycz, and D. F. Woodward. 2009. Prostanoid receptor L648–L656. antagonists: development and therapeutic applications. Br. J. Pharmacol. 158: 16. Hashimoto, K., B. S. Graham, M. W. Geraci, G. A. FitzGerald, K. Egan, 104–145. W. Zhou, K. Goleniewska, J. F. O’Neal, J. D. Morrow, R. K. Durbin, et al. 2004. 44. Alheim, K., J. Corness, M. K. Samuelsson, L. G. Bladh, T. Murata, T. Nilsson, Signaling through the prostaglandin I2 receptor IP protects against respiratory and S. Okret. 2003. Identification of a functional glucocorticoid response element syncytial virus-induced illness. J. Virol. 78: 10303–10309. in the promoter of the cyclin-dependent kinase inhibitor p57Kip2. J. Mol. Endo- 17. Traves, S. L., and D. Proud. 2007. Viral-associated exacerbations of asthma and crinol. 30: 359–368. COPD. Curr. Opin. Pharmacol. 7: 252–258. 45. van der Laan, S., R. A. Sarabdjitsingh, M. F. Van Batenburg, S. B. Lachize, H. Li, 18. Suissa, S., R. McGhan, D. Niewoehner, and B. Make. 2007. Inhaled corticoste- T. F. Dijkmans, E. Vreugdenhil, E. R. de Kloet, and O. C. Meijer. 2008. Chro- roids in chronic obstructive pulmonary disease. Proc. Am. Thor. Soc. 4: 535–542. matin immunoprecipitation scanning identifies glucocorticoid receptor binding 19. De Bosscher, K., W. Vanden Berghe, and G. Haegeman. 2003. The interplay regions in the proximal promoter of a ubiquitously expressed glucocorticoid tar- between the glucocorticoid receptor and nuclear factor-␬B or activator protein-1: get gene in brain. J. Neurochem. 106: 2515–2523. molecular mechanisms for gene repression. Endocr. Rev. 24: 488–522. 46. Hogg, J. C., F. Chu, S. Utokaparch, R. Woods, W. M. Elliott, L. Buzatu, 20. Kaur, M., J. E. Chivers, M. A. Giembycz, and R. Newton. 2008. Long-acting R. M. Cherniack, R. M. Rogers, F. C. Sciurba, H. O. Coxson, and P. D. Pare. ␤ 2-adrenoceptor agonists synergistically enhance glucocorticoid-dependent tran- 2004. The nature of small-airway obstruction in chronic obstructive pulmonary scription in human airway epithelial and smooth muscle cells. Mol. Pharmacol. disease. N. Engl. J. Med. 350: 2645–2653. 73: 201–214. 47. Chang, T. S., M. J. Kim, K. Ryoo, J. Park, S. J. Eom, J. Shim, K. I. Nakayama, 21. Newton, R., and N. S. Holden. 2007. Separating transrepression and transacti- K. Nakayama, M. Tomita, K. Takahashi, et al. 2003. p57KIP2 modulates stress- vation: A distressing divorce for the glucocorticoid receptor? Mol. Pharmacol. activated signaling by inhibiting c-Jun NH2-terminal kinase/stress-activated pro- 72: 799–809. tein Kinase. J. Biol. Chem. 278: 48092–48098. 22. Clark, A. R., J. R. S. Martins, and C. R. Tchen. 2008. Role of dual specificity 48. So, A. Y., C. Chaivorapol, E. C. Bolton, H. Li, and K. R. Yamamoto. 2007. phosphatases in biological responses to glucocorticoids. J. Biol. Chem. 283: Determinants of cell- and gene-specific transcriptional regulation by the glu- 25765–25769. cocorticoid receptor. PLoS. Genet. 3: e94.

23. Clark, A. R. 2007. Anti-inflammatory functions of glucocorticoid-induced genes. 49. Yeagley, D., and P. G. Quinn. 2005. 3Ј,5Ј-Cyclic adenosine monophosphate re- Downloaded from Mol. Cell. Endocrinol. 275: 79–97. sponse element-binding protein and CCAAT enhancer-binding protein are dis- 24. Nordeen, S. K., M. L. Moyer, and B. J. Bona. 1994. The coupling of multiple pensable for insulin inhibition of phosphoenolpyruvate carboxykinase transcrip- signal transduction pathways with steroid response mechanisms. Endocrinology tion and for its synergistic induction by protein kinase A and glucocorticoids. 134: 1723–1732. Mol. Endocrinol. 19: 913–924. 25. Churchill, L., F. H. Chilton, J. H. Resau, R. Bascom, W. C. Hubbard, and 50. Schmoll, D., C. Wasner, C. J. Hinds, B. B. Allan, R. Walther, and A. Burchell. D. Proud. 1989. Cyclooxygenase metabolism of endogenous arachidonic acid by 1999. Identification of a cAMP response element within the glucose-6-phospha- cultured human tracheal epithelial cells. Am. Rev. Respir. Dis. 140: 449–459. tase hydrolytic subunit gene promoter which is involved in the transcriptional 26. Chivers, J. E., L. M. Cambridge, M. C. Catley, J. C. Mak, L. E. Donnelly, regulation by cAMP and glucocorticoids in H4IIE hepatoma cells. Biochem. J. http://www.jimmunol.org/ P. J. Barnes, and R. Newton. 2004. Differential effects of RU486 reveal distinct 338: 457–463. mechanisms for glucocorticoid repression of prostaglandin E release. Eur. J. Bio- 51. van Schaftingen, E., and I. Gerin. 2002. The glucose-6-phosphatase system. Bio- chem. 271: 4042–4052. chem. J. 362: 513–532. 27. Bley, K. R., A. Bhattacharya, D. V. Daniels, J. Gever, A. Jahangir, C. O’Yang, 52. Eickelberg, O., M. Roth, R. Lorx, V. Bruce, J. Rudiger, M. Johnson, and S. Smith, D. Srinivasan, A. P. Ford, and M. F. Jett. 2006. RO1138452 and L. H. Block. 1999. Ligand-independent activation of the glucocorticoid receptor RO3244794: characterization of structurally distinct, potent and selective IP ␤ by 2-adrenergic receptor agonists in primary human lung fibroblasts and vas- (prostacyclin) receptor antagonists. Br. J. Pharmacol. 147: 335–345. cular smooth muscle cells. J. Biol. Chem. 274: 1005–1010. 28. Meja, K., M. C. Catley, L. M. Cambridge, P. J. Barnes, H. Lum, R. Newton, and 53. Usmani, O. S., K. Ito, K. Maneechotesuwan, M. Ito, M. Johnson, P. J. Barnes, M. A. Giembycz. 2004. Adenovirus-mediated delivery and expression of a and I. M. Adcock. 2005. Glucocorticoid receptor nuclear translocation in airway cAMP-dependent protein kinase inhibitor gene to BEAS-2B epithelial cells abol- cells after inhaled combination therapy. Am. J. Respir. Crit Care Med. 172: ishes the anti-inflammatory effects of rolipram, salbutamol and prostaglandin E : 2 704–712. a comparison with H-89. J. Pharmacol. Exp. Ther. 309: 833–844. 54. Profita, M., R. Gagliardo, R. Di Giorgi, F. Pompeo, M. Gjomarkaj, G. Nicolini, by guest on October 1, 2021 29. Leesnitzer, L. M., D. J. Parks, R. K. Bledsoe, J. E. Cobb, J. L. Collins, J. Bousquet, and A. M. Vignola. 2005. Biochemical interaction between effects of T. G. Consler, R. G. Davis, E. A. Hull-Ryde, J. M. Lenhard, L. Patel, et al. 2002. beclomethasone dipropionate and salbutamol or formoterol in sputum cells from Functional consequences of cysteine modification in the ligand binding sites of mild to moderate asthmatics. Allergy 60: 323–329. peroxisome proliferator activated receptors by GW9662. Biochemistry 41: 6640–6650. 55. Roth, M., P. R. Johnson, J. J. Rudiger, G. G. King, Q. Ge, J. K. Burgess, G. Anderson, M. Tamm, and J. L. Black. 2002. Interaction between glucocorti- 30. Waud, D. R., S. L. Son, and B. E. Waud. 1978. Kinetic and empirical analysis of coids and ␤ agonists on bronchial airway smooth muscle cells through synchro- dose-response curves illustrated with a cardiac example. Life Sci. 22: 1275–1285. 2 nised cellular signalling. Lancet 360: 1293–1299. 31. Schneider, J., E. Friderichs, B. Kogel, U. Seipp, H.-J. Stahlberg, R. Terlinden, and K. Heintze. 1993. Taprostene sodium. Cardiovasc. Drug Rev. 11: 479–500. 56. Rangarajan, P. N., K. Umesono, and R. M. Evans. 1992. Modulation of glu- 32. Aksoy, M. O., X.-x. Li, M. Borenstein, Y. Yi, and S. G. Kelsen. 1999. Effects of cocorticoid receptor function by protein kinase A. Mol. Endocrinol. 6: topical corticosteroids on inflammatory mediator-induced eicosanoid release by 1451–1457. human airway epithelial cells. J. Allergy. Clin. Immunol. 103: 1081–1091. 57. Oikarinen, J., L. Hamalainen, and A. Oikarinen. 1984. Modulation of glucocor- 33. Neubig, R. R., M. Spedding, T. Kenakin, and A. Christopoulos. 2003. Interna- ticoid receptor activity by cyclic nucleotides and its implications on the regulation tional Union of Pharmacology Committee on Receptor Nomenclature and Drug of human skin fibroblast growth and protein synthesis. Biochim. Biophys. Acta Classification. XXXVIII. Update on terms and symbols in quantitative pharma- 799: 158–165. cology. Pharmacol. Rev. 55: 597–606. 58. Espinas, M. L., J. Roux, R. Pictet, and T. Grange. 1995. Glucocorticoids and 34. Wise, H., and R. L. Jones. 1996. Focus on prostacyclin and its novel mimetics. protein kinase A coordinately modulate transcription factor recruitment at a glu- Trends Pharmacol. Sci. 17: 17–21. cocorticoid-responsive unit. Mol. Cell. Biol. 15: 5346–5354. 35. Chow, K. B., R. L. Jones, and H. Wise. 2004. Agonists can discriminate between 59. Rowan, B. G., N. Garrison, N. L. Weigel, and B. W. O’Malley. 2000. 8-Bromo- cloned human and mouse prostacyclin receptors. Prostaglandins Leukot. Essent. cyclic AMP induces phosphorylation of two sites in SRC-1 that facilitate ligand- Fatty Acids 70: 423–429. independent activation of the chicken progesterone receptor and are critical for 36. Falcetti, E., D. M. Flavell, B. Staels, A. Tinker, S. G. Haworth, and L. H. Clapp. functional cooperation between SRC-1 and CREB binding protein. Mol. Cel.l 2007. IP-receptor-dependent activation of PPAR␥ by stable prostacyclin ana- Biol. 20: 8720–8730. logues. Biochem. Biophys. Res. Commun. 360: 821–827. 60. Fenne, I. S., T. Hoang, M. Hauglid, J. V. Sagen, E. A. Lien, and G. Mellgren. 37. Mittelstadt, P. R., and J. D. Ashwell. 2001. Inhibition of AP-1 by the glucocor- 2008. Recruitment of coactivator glucocorticoid receptor interacting protein 1 to Ј Ј ticoid-inducible protein GILZ. J. Biol. Chem. 276: 29603–29610. an estrogen receptor transcription complex is regulated by the 3 ,5 -cyclic aden- Ј 38. Eddleston, J., J. Herschbach, A. L. Wagelie-Steffen, S. C. Christiansen, and osine 5 -monophosphate-dependent protein kinase. Endocrinology 149: B. L. Zuraw. 2007. The anti-inflammatory effect of glucocorticoids is mediated 4336–4345. by glucocorticoid-induced leucine zipper in epithelial cells. J. Allergy Clin. Im- 61. Hoang, T., I. S. Fenne, C. Cook, B. Borud, M. Bakke, E. A. Lien, and munol. 119: 115–122. G. Mellgren. 2004. cAMP-dependent protein kinase regulates ubiquitin-protea- 39. Samuelsson, M. K., A. Pazirandeh, B. Davani, and S. Okret. 1999. p57Kip2,a some-mediated degradation and subcellular localization of the nuclear receptor glucocorticoid-induced inhibitor of cell cycle progression in HeLa cells. Mol. coactivator GRIP1. J. Biol. Chem. 279: 49120–49130. Endocrinol. 13: 1811–1822. 62. Barnes, P. J. 2008. Immunology of asthma and chronic obstructive pulmonary 40. Issa, R., S. Xie, N. Khorasani, M. Sukkar, I. M. Adcock, K. Y. Lee, and disease. Nat. Rev. Immunol. 8: 183–192. K. F. Chung. 2007. Corticosteroid inhibition of growth-related oncogene pro- 63. Sin, D. D., and S. F. Man. 2007. Do chronic inhaled steroids alone or in com- tein-␣ via mitogen-activated kinase phosphatase-1 in airway smooth muscle cells. bination with a bronchodilator prolong life in chronic obstructive pulmonary J. Immunol. 178: 7366–7375. disease patients? Curr. Opin. Pulm. Med. 13: 90–97. 41. Abramovitz, M., M. Adam, Y. Boie, M. Carriere, D. Denis, C. Godbout, 64. Peinado, V. I., S. Pizarro, and J. A. Barbera. 2008. Pulmonary vascular involve- S. Lamontagne, C. Rochette, N. Sawyer, N. M. Tremblay, et al. 2000. The use of ment in COPD. Chest 134: 808–814. recombinant prostanoid receptors to determine the affinities and selectivities of 65. Chaouat, A., R. Naeije, and E. Weitzenblum. 2008. Pulmonary hypertension in prostaglandins and related analogs. Biochim. Biophys. Acta. 1483: 285–293. COPD. Eur. Respir. J. 32: 1371–1385. The Journal of Immunology 6799

66. Mitchell, R. S., G. W. Silvers, G. A. Dart, T. L. Petty, T. N. Vincent, S. F. Ryan, structive Bronchitis and Haemostasis Study Group. Am. J. Respir. Crit Care Med. and G. F. Filley. 1968. Clinical and morphologic correlations in chronic airway 156: 1794–1799. obstruction. Aspen. Emphysema. Conf. 9: 109–123. 71. Beghetti, M., G. Reber, M. P. de, L. Vadas, A. Chiappe, I. Spahr-Schopfer, and 67. Ferroni, P., S. Basili, F. Martini, M. Vieri, G. Labbadia, C. Cordova, P. C. Rimensberger. 2002. Aerosolized iloprost induces a mild but sustained C. Alessandri, and P. P. Gazzaniga. 2000. Soluble P-selectin as a marker of inhibition of platelet aggregation. Eur. Respir. J. 19: 518–524. platelet hyperactivity in patients with chronic obstructive pulmonary disease. 72. McLaughlin, V. V., A. Shillington, and S. Rich. 2002. Survival in primary pul- J. Investig. Med. 48: 21–27. monary hypertension: the impact of epoprostenol therapy. Circulation 106: 1477–1482. 68. Cordova, C., A. Musca, F. Violi, C. Alessandri, A. Perrone, and F. Balsano. 1985. 73. Norel, X., L. Walch, C. Labat, J. P. Gascard, E. Dulmet, and C. Brink. 1999. Platelet hyperfunction in patients with chronic airways obstruction. Eur. Prostanoid receptors involved in the relaxation of human bronchial preparations. J. Respir. Dis. 66: 9–12. Br. J. Pharmacol. 126: 867–872. 69. Nana-Sinkam, S. P., J. D. Lee, S. Sotto-Santiago, R. S. Stearman, R. L. Keith, 74. Oudiz, R. J., and H. W. Farber. 2009. Dosing considerations in the use of intra- Q. Choudhury, C. Cool, J. Parr, M. D. Moore, T. M. Bull, et al. 2007. Prosta- venous prostanoids in pulmonary arterial hypertension: an experience-based re- cyclin prevents pulmonary endothelial cell apoptosis induced by cigarette smoke. view. Am. Heart J. 157: 625–635. Am. J. Respir. Crit Care Med. 175: 676–685. 75. Kuwano, K., A. Hashino, T. Asaki, T. Hamamoto, T. Yamada, K. Okubo, and 70. Davi, G., S. Basili, M. Vieri, F. Cipollone, S. Santarone, C. Alessandri, K. Kuwabara. 2007. 2-[4-[(5,6-diphenylpyrazin-2-yl)(isopropyl)amino]butoxy]- P. Gazzaniga, C. Cordova, and F. Violi. 1997. Enhanced thromboxane biosyn- N-(methylsulfonyl)acetamide (NS-304), an orally available and long-acting pros- thesis in patients with chronic obstructive pulmonary disease. The Chronic Ob- tacyclin receptor agonist prodrug. J. Pharmacol. Exp. Ther. 322: 1181–1188. Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021