A Novel Prostaglandin E Receptor 4–Associated Protein Participates in Antiinflammatory Signaling
Prostaglandin E2 exerts an antiinflammatory action by ligation of the heptahelical receptor EP4 in human macrophages. Because the mechanism by which EP4 receptor stimulation suppresses inflammatory activation in macrophages remains undefined, we sought interactors with the carboxyl-terminal cytoplasmic domain of the EP4 receptor. Yeast 2-hybrid screening of the human bone marrow cDNA library with the EP4 receptor as a bait identified a cDNA clone encoding a 669-amino acid protein, designated here as EP4 receptor-associated protein (EPRAP), which contains 8 ankyrin motifs that might recruit other signaling molecules. EPRAP bound to the full-length EP4 receptor in HEK293 cells cotransfected with V5-tagged EPRAP and FLAG-tagged EP4 receptor cDNA, as anti-FLAG antibody coimmunoprecipitated EPRAP with the EP4 receptor from the lysates of cotransfected cells. Human macrophages derived from peripheral blood monocytes expressed an approximately 70-kDa protein detected by Western blotting with a polyclonal anti-EPRAP antibody. Fluorescence immunohistochemistry colocalized EPRAP with the EP4 receptor in human atheromata. Interference with EPRAP function by small interference RNA limited prostaglandin E2–mediated suppression of chemokine expression in macrophages activated with lipopolysaccharide and tumor necrosis factor α. In conclusion, the antiinflammatory action of prostaglandin E2 in macrophages involves EPRAP that associates directly with the cytoplasmic tail of EP4 receptor.
Prostanoids, including prostaglandin (PG) D2, PGE2, PGF2α, PGI2, and thromoboxane A2, regulate a broad range of physiological functions and modulate chronic inflammatory diseases such as rheumatoid arthritis, asthma, Crohn’s disease, and atherosclerosis. Nonsteroidal antiinflammatory drugs (NSAIDs) commonly used in the treatment of inflammatory conditions block the production of prostanoids by inhibiting cyclooygenase activity. However, recent clinical data have shown that selective cyclooygenase-2 inhibitors can increase coronary events in certain circumstances, perhaps by interfering with prostacyclin production.1
PGE2 can augment macrophage (MΦ) expression of proteolytic enzymes MMP-2 and -9, enzymes implicated in the vulnerability of atheroma to rupture and cause thrombosis.2 On the other hand, PGE2 relaxes vascular smooth muscle cells (SMC) by augmenting intracellular cAMP and can inhibit SMC proliferation induced by interleukin (IL)-1.3 PGE2 can also reduce inflammatory activation of human MΦ by suppressing the expression of cytokines including tumor necrosis factor (TNF)-α, IL-12, and interferon (IFN)γ,4,5 and chemokines such as monocyte chemoattractant protein (MCP)-1, macrophage inflammatory protein (MIP)-1β, MIP-1α, interferon-gamma–inducible protein (IP)-10, and IL-8, implicated in leukocyte recruitment to atheromata.6 Thus, PGE2 may function as an endogenous antiinflammatory mediator.
PGE2 exerts its biological functions mainly via 4 G protein–coupled 7 transmembrane receptors (EP1–4).7 Several lines of evidence suggest that the EP4 receptor may participate in the antiinflammatory action of PGE2 in human MΦ.6 Recent work demonstrated that targeted disruption of the gene for EP4 receptor in mice decreased PGE2-mediated inhibition of cytokine production in lipopolysaccharide (LPS)-activated MΦ8 and intensified chronic inflammatory diseases such as colitis9 and arthritis.10 These observations support a primary role for the EP4 receptor in this novel antiinflammatory pathway in MΦ.
The interaction between PGE2 and the EP2 and EP4 receptors transiently increases intracellular cAMP via Gα. cAMP activates protein kinase A (PKA), which subsequently phosphorylates downstream effector proteins such as cAMP response element–binding protein (CREB).11 This pathway appears involved in some PGE2 functions such as bone formation mediated by the EP4 receptor. However, our previous experiments using a PKA-selective inhibitor, H-89, did not support involvement of the PKA and CREB pathway in chemokine suppression by PGE2 in MΦ.6 Furthermore, the targeted disruption of the gene for EP2R, which is expressed by mouse MΦ and interacts with Gα, does not attenuate PGE2-mediated antiinflammatory activity.8
Interestingly, the EP4 receptor has an extended carboxy-terminal cytoplasmic domain compared with the other G protein–coupled receptors for prostanoids. These observations raised the possibility that EP4 receptor–mediated antiinflammatory action might involve a PKA-independent signal transduction pathway mediated by a molecule that interacts with the receptor carboxy-terminal tail. To investigate this hypothesis, we used yeast 2-hybrid screening and identified a novel protein, denoted here as EP4 receptor–associated protein (EPRAP), that binds to the EP4 receptor carboxy-terminal tail. EPRAP may function in transducing the action of PGE2 in MΦ.
Materials and Methods
Yeast Two-Hybrid Screening and Amino Acid Sequence Analysis
The cDNA encoding the carboxyl-terminal cytoplasmic domain of the human PGE2 receptor subtype EP4 (amino acids 421 to 488) was obtained by RT-PCR from phorbol myristate acetate–treated THP-1 mRNA and fused to the Gal4 DNA binding domain in a yeast bait vector pBD-GAL4 Cam (Stratagene, La Jolla, Calif). Yeast 2-hybrid screening was performed on a pretransformed human bone marrow library (Clontech Laboratories Inc, Mountain View, Calif), according to the instructions of the manufacturer of the MATCHMAKER Two-Hybrid System 3 (Clontech). The cDNA inserts were sequenced with an automated DNA sequencer (DNA Sequencing Facility, Brigham and Women’s Hospital, Boston, Mass). Homology search of amino acid and motif prediction was performed using BLAST programs (National Center for Biotechnology Information, Bethesda, Md).
The mammalian expression vector for the carboxyl-terminal V5 epitope-tagged EPRAP (pV5-EPRAP) was created by cloning a PCR-amplified fragment from the original isolated clone pACT-EPRAP into an expression vector pcDNA3.1D/V5-His-TOPO (Invitrogen Corp, Carlsbad, Calif). cDNA encoding the full-length EP4 receptor and EP2 receptor were isolated by PCR from human lung cDNA (Clontech) and kidney cDNA, respectively. The cDNA were cloned into a p3X FLAG-CMV-7.1 (Sigma Chemical Co, St Louis, Mo) to construct the NH2-terminal FLAG epitope-tagged EP4 receptor and EP2 receptor (pFLAG-EP4R and pFLAG-EP2R). The pV5-EPRAP was transiently cotransfected with pFLAG-EP4R, pFLAG-EP2R, or a mock vector into HEK 293 cells (American Type Culture Collection, Manassas, Va) with Superfect (Qiagen, Inc, Valencia, Calif), according to the instructions of the manufacturer. The ratio of EPRAP and the receptor DNA was 2:1. The cells were harvested 48 hours after transfection, washed with PBS, and then lysed on ice in 10 mmol/L Tris-HCl (pH 7.4), 150 mmol/L NaCl, 1 mmol/L EDTA, 0.2% (vol/vol) Triton X-100, and 5% glycerol supplemented with a proteinase inhibitor cocktail (Sigma) for 30 minutes. The lysate was centrifuged at 10 000g for 20 minutes. The cleared supernatants were incubated with anti-FLAG M2 Agarose (Sigma) for 16 hours. The agarose beads were then washed in the lysis buffer, resuspended in Laemmli buffer containing 2-mercaptoethanol, and boiled for 10 minutes. After centrifugation, a portion of the samples were separated by SDS-PAGE and transferred onto to a polyvinylidene fluoride membrane (PerkinElmer Life and Analytical Sciences Inc, Boston, Mass). The blot was blocked overnight in PBS containing 5% defatted dry milk and 0.1% Tween-20. After 2 hours of incubation with anti-V5 epitope, monoclonal antibody conjugated with peroxidase (Invitrogen) for V5 epitope-tagged EPRAP or anti-FLAG M2 monoclonal antibody conjugated with peroxidase (Sigma) for FLAG epitope-tagged EP4 and EP2 receptor. Finally, the blot was washed, and immunoreactive protein was visualized using chemiluminescence reagent (PerkinElmer).
Monocytes were from leukopacs of healthy donors isolated by density-gradient centrifugation, using Lymphocyte Separation Medium (ICN Biomedicals Inc, Irvine, Calif) and subsequent adherence to cell culture dishes. Monocytes were cultured in M199 medium (BioWhittaker Inc, Walkersville, Md) containing 5% human serum (ICN Biomedicals) for 7 to 9 days to obtain MΦ. Forty-eight hours before and during the experiments, MΦ were incubated in M199 medium containing 1% human serum.
Anti-EPRAP Serum Generation and Western Blot Analysis
Rabbit antisera against a 16-amino acid peptide (SSSPEEPLNGESYESC), corresponding to predicted amino acid residues 274 to 289 of EPRAP, were raised by the Zymed proprietary procedure (PolyQuik; Zymed Laboratories Inc, South San Francisco, Calif). The IgG fraction was purified from antisera with ImmunoPure (A) IgG Purification Kit (Pierce Biotechnology Inc, Rockford, Ill). HEK 293 cells transfected with pV5-EPRAP and MΦ were washed with PBS and lysed on ice in 10 mmol/L Tris-HCl (pH 7.4), 150 mmol/L NaCl, 1 mmol/L EDTA, 1% (vol/vol) Triton X-100, and 0.5% sodium dodecyl sulfate supplemented with proteinase inhibitor cocktail (Sigma). Thirty micrograms of protein were separated by 10% SDS-PAGE under reducing conditions. Western blot analysis was performed as described above using anti-human EPRAP antibody.
Human carotid plaques and fragments of nonatherosclerotic arteries were obtained from endarterectomies and heart transplantation donors according to protocols approved by the Human Investigation Review Committee at Brigham and Women’s Hospital. Air-dried serial cryostat sections (6 μm) were fixed in acetone for 5 minutes at −20°C, treated with PBS containing 0.3% H2O2, and then incubated with anti-EPRAP antibody. Subsequent processing was performed using a Universal DAKO LSAB2 peroxidase kit (DAKO Corp, Carpenteria, Calif), according to the recommendations of the manufacturer. Antibody binding was visualized with 3-amino-9-ethyl carbazole (Sigma). For colocalization by immunofluorescence double labeling, rabbit anti-human EP4 receptor antibody was applied for 90 minutes, followed by biotinylated secondary antibody (45 minutes) and Texas red–conjugated streptavidin (20 minutes; Amersham Pharmacia Biotech, Piscataway, NJ). After application of the avidin/biotin blocking kit (Vector Laboratories, Burlingame, Calif), biotinylated anti-human EPRAP was added (overnight, 4°C). Finally, the appropriate secondary antibodies were applied for 45 minutes, followed by streptavidin–fluorescein isothiocyanate (Amersham Pharmacia Biotech) for 20 minutes.
Small Interference RNA Construction and Cell Treatment
Small interference RNAs (siRNA) were generated using SilencerTM siRNA Construction Kit (Ambion, Austin, Tex), according to the manual provided. A target sequence is 5′-AAC CAU GGA CCU CCG CAC CGC-3′ corresponding to EPRAP mRNA including the AUG start codon. Silencer negative control #1 siRNA (Ambion) with no significant homology to any known human genes was used as negative control siRNA. For transfection to human primary MΦ with siRNA, a JetPEI-Man MΦ transfection reagent (PolyPlus-Transfection, Illkirch, France) was used. Human primary MΦ were incubated with M199 medium containing 1% human serum in the presence of 50 nmol/L siRNA at final concentration 48 hours before PGE2 treatment. The culture medium was replaced with the fresh medium containing 50 nmol/L PGE2 or vehicle. After a 30 minutes treatment, the cells were stimulated with 5 ng/mL LPS or 25 ng/mL TNF for 12 hours, and then the culture medium was harvested. MIP-1β, TNF, and MCP-1 protein release was measured by ELISA.6 Intracellular cAMP level was quantified by cAMP EIA Kit (Amersham Bioscience), according to the instructions of the manufacturer. Expression of EPRAP was analyzed by Western blotting with anti-EPRAP polyclonal antibodies. Densitometric analysis of immunoreactive bands used ImagePro software (Media Cybernetics, Silver Spring, Md) applied to digital images of the Western blots.
An expanded Materials and Methods section is available in the online data supplement at http://circres.ahajournals.org.
To seek EP4 receptor–specific signal transduction pathways that may mediate the PKA-independent antiinflammatory action of PGE2, we used 2-hybrid screening with the carboxy-terminal cytoplasmic tail of the human EP4 receptor as a bait. Screening of 5.0 ×106 transformants from a human bone marrow library resulted in the isolation of 2 independent clones that interacted with EP4 receptor. Sequence analysis reveal that both plasmids contained the same open reading frame encoding a 669-amino acid protein (Figure 1), which included a potential initiation ATG that matched well the consensus sequence for translation initiation and a stop codon. We designated this protein EP4 receptor–associated protein (EPRAP). BLAST protein homology search analysis revealed that EPRAP is identical to a protein of unknown function registered in GenBank (accession no. XM_035635) and has a high degree of homology with murine Fem1a, a reported homolog of the Caenorhabditis elegans FEM-1 protein.12 EPRAP contains 8 ankyrin motifs but no predicted enzyme catalytic domain.
To verify the binding of EPRAP to full-length EP4 receptors in mammalian cells, we transfected HEK293 cells with pFLAG-EP4R and pV5-EPRAP and performed coimmunoprecipitation with an anti-FLAG tag antibody that recognizes EP4 receptor. The sample immunoprecipitated from the cell lysate contained FLAG-tagged EP4 receptor and V5-tagged EPRAP, as demonstrated by Western blotting with anti-FLAG and anti-V5 antibodies (Figure 2). This procedure did not yield EPRAP in either the sample derived from the cell transfected with pFLAG control vector and pV5-EPRAP or in the sample transfected with pFLAG-EP4R and pcDNA3.1D/V5 control vector. The EP2 receptor did not associate with EPRAP in HEK293 cells tested by coimmunoprecipitation (Figure 2). Treatment of HEK293 expressing EP4 receptor and EPRAP with PGE2 (50 nmol/L) did not alter substantially the interaction between EP4 receptor and EPRAP.
To demonstrate the expression of EPRAP in primary human MΦ, we generated a specific polyclonal antibody against EPRAP and performed Western blotting. We detected the V5-tagged EPRAP specifically in the HEK 293 cells transfected with pV5-EPRAP using the anti-EPRAP antibody (Figure 3A-1). The antibody reacted mainly with an approximately 70-kDa band, the predicted molecular weight of EPRAP, and weakly with an approximately 50-kDa band in primary human MΦ (Figure 3A-2). Treatment with LPS (5 ng/mL) or PGE2 (50 nmol/L) for 12 hours did not change the protein expression of EPRAP in human MΦ significantly (data not shown).
We previously demonstrated the prominent expression of EP4 receptor by CD68-positive MΦ in atheroma.6 We next tested whether macrophages coexpress EP4 receptor and EPRAP in atheroma by immunofluorescence double labeling using anti-human EP4 receptor antibody and biotinylated anti-human EPRAP. The EPRAP localized mainly in MΦ and intimal SMC in human carotid atheroma. In contrast, nonatherosclerotic arterial tissues express no immunodetectable EPRAP (n=3, Figure 3B). EPRAP colocalized with the EP4 receptor in the MΦ-rich shoulder area in human atheroma (n=3, Figure 3C), consistent with a role of the EP4 receptor -EPRAP complex in modulating inflammation in atherogenesis.
We previously showed that PGE2 regulates chemokine mRNA expression induced by pro-inflammatory stimuli in MΦ.6 Finally, to examine the involvement of EPRAP in the antiinflammatory effect of PGE2 in human primary MΦ, we used siRNA13 to reduce EPRAP function in the cells. A 2-day exposure of EPRAP-specific siRNA (50 nmol/L) significantly reduced 70-kDa EPRAP expression by 43.7±4.8% (mean±SD of the 3 independent experiments), as shown by Western blotting (Figure 4A) and densitometric analysis. We evaluated the antiinflammatory action elicited by PGE2 after the same treatment. In the presence of EPRAP-specific siRNA, LPS stimulated MIP-1β expression in human primary MΦ. However, the EPRAP siRNA treatment impaired the inhibitory action of PGE2 on MIP-1β expression (Figure 4B). We observed that the siRNA treatment attenuated PGE2-dependent suppression of TNF (control siRNA: 95.1±2.1%; EPRAP siRNA: 63.6±6.5% inhibition by 50 nmol/L PGE2 treatment±SD of the 3 independent experiments) and MCP-1 (control siRNA: 91.8±1.0%; EPRAP siRNA: 49.3±6.3%) as well as MIP-1β in LPS-activated human MΦ significantly (P<0.05). Moreover, siRNA treatment significantly reduced PGE2-dependent suppression of MCP-1 (control siRNA: 91.2±3.5%; EPRAP siRNA: 51.3±3.8% inhibition by 50 nmol/L PGE2 treatment±SD of the 3 independent experiments) in TNF-activated human MΦ (P<0.01). Taken together, these data indicate that EPRAP participates in the signaling cascade of PGE2-mediated attenuation of inflammation in MΦ. The interaction between PGE2 and EP4 receptors transiently increases intracellular cAMP via Gα. We examined the elevation of intracellular cAMP level by PGE2 in human MΦ treated with EPRAP siRNA, as described above. The siRNA treatment had no significant effect on cAMP elevation by the physiological concentration range of PGE2 (Figure 4C).
Although usually considered a proinflammatory mediator, recent work demonstrated that PGE2 can limit inflammation by suppression of proinflammatory mediators such as cytokines, chemokines, and costimulatory molecules.4–6 Several lines of data from our own and other laboratories implicate EP4 receptor, among 4 human subtypes of PGE2 receptor, as the transducer of PGE2 antiinflammatory effects. EP4 receptor belongs to the 7 transmembrane receptor superfamily, whose action classically depend on transiently increased intracellular cAMP via G protein Gα.
The EP4 receptor possesses a relatively longer carboxy-terminal cytoplasmic domain (156 amino acids) than the other G protein–coupled receptors for prostanoids. The cytoplasmic domain of EP4 receptor lacks substantial sequence similarity with those of other prostanoid receptors. Several studies indicate that the cytoplasmic domain of the EP4 receptor interacts with arrestins, causing receptor internalization.14,15 However, the specific signaling mechanism by which EP4 receptor stimulation suppresses inflammatory activation and the interaction of downstream signaling molecules with the receptor remains undefined.
Using of yeast 2-hybrid methodology and coimmunoprecipitation in HEK293 cells transfected with cDNA for EP4 receptor and EPRAP, this study demonstrates that the cytoplasmic tail (amino acids 421 to 488) of human EP4 receptor interacts with a novel protein, designated here as EPRAP. EPRAP contains 8 sequential ankyrin repeats and shows the highest degree of homology with Murine Fem1a, a reported homolog of the C elegans FEM-1 protein. The FEM-1 protein participates in signal transduction in C elegans by regulating the activity of transcription factors in the sex-determination cascade. Although the biological mechanism of action of murine Fem1a and EPRAP remain undefined, the existence of a sequential ankyrin repeat motif suggests that its function involves protein–protein interaction.
Various ankyrin motif–containing proteins such as the Rel-NFκB/IκB and Notch function directly and indirectly in transcriptional regulation.16,17 Thus, PGE2 may modulate the expression of pro-inflammatory genes by activating the EP4 receptor–EPRAP complex. Compared with other Gα-coupled receptors for prostanoids such as EP2 for PGE2 and IP for PGI2, the specific interaction between EP4 receptor and EPRAP may explain the potent antiinflammatory activity of EP4 receptor agonists. Interestingly, our EPRAP gene–silencing experiment suggests that EPRAP may have no directly involvement in Gα-mediated cAMP elevation in human MΦ treated with optimum concentration of PGE2 (10 to 100 nmol/L). At least in macrophages, the cAMP elevation may not be necessary in PGE2-mediated antiinflammation. Some proprietary EP4 receptor agonists have antiinflammatory activity via EP4 receptor without elevation of cAMP (K.T., unpublished results). Thus, EPRAP may mediate a novel nonclassical G protein–coupled and cAMP-independent antiinflammatory pathway triggered by EP4 receptor ligation. Additionally, consistent with the participation of a nonclassical G protein–coupled and cAMP-independent pathway in EP4 receptor signaling cascades, a recent study demonstrated that T-cell factor (Tcf) activation mediated by EP4 receptor occurs primarily through a phosphatidlyinositol 3-kinase pathway independent of cAMP elevation.18 Elucidation of the downstream molecular mechanism by which the EP4-EPRAP pathway suppresses inflammatory molecules in MΦ will require further investigation.
In conclusion, these results identify a new mechanism for EP4 receptor signal transduction that involves a novel interactor. These findings have important implications for the pathogenesis of atherosclerosis and other chronic inflammatory diseases. This study raises the possibility that specific modulation of the EP4-EPRAP signaling cascade might provide a new therapeutic strategy for management of these diseases.
This work was supported by the Leducq Foundation and a grant from the National Heart, Lung, and Blood Institute (HL-34636) (to P.L.). We thank Eugenia Shvartz and Elissa Simon-Morrissey for skillful technical assistance and Karen Williams for editorial assistance.
Original received June 15, 2005; resubmission received December 16, 2005; accepted January 5, 2006.
Cipollone F, Prontera C, Pini B, Marini M, Fazia M, De Cesare D, Iezzi A, Ucchino S, Boccoli G, Saba V, Chiarelli F, Cuccurullo F, Mezzetti A. Overexpression of functionally coupled cyclooxygenase-2 and prostaglandin E synthase in symptomatic atherosclerotic plaques as a basis of prostaglandin E(2)-dependent plaque instability. Circulation. 2001; 104: 921–927.
van der Pouw Kraan TC, Boeije LC, Smeenk RJ, Wijdenes J, Aarden LA. Prostaglandin-E2 is a potent inhibitor of human interleukin 12 production. J Exp Med. 1995; 181: 775–779.
Takayama K, Garcia-Cardena G, Sukhova GK, Comander J, Gimbrone MA Jr, Libby P. Prostaglandin E2 suppresses chemokine production in human macrophages through the EP4 receptor. J Biol Chem. 2002; 277: 44147–44154.
Yoshida K, Oida H, Kobayashi T, Maruyama T, Tanaka M, Katayama T, Yamaguchi K, Segi E, Tsuboyama T, Matsushita M, Ito K, Ito Y, Sugimoto Y, Ushikubi F, Ohuchida S, Kondo K, Nakamura T, Narumiya S. Stimulation of bone formation and prevention of bone loss by prostaglandin E EP4 receptor activation. Proc Natl Acad Sci U S A. 2002; 99: 4580–4585.
Penn RB, Pascual RM, Kim YM, Mundell SJ, Krymskaya VP, Panettieri RA Jr, Benovic JL. Arrestin specificity for G protein-coupled receptors in human airway smooth muscle. J Biol Chem. 2001; 276: 32648–32656.
Malek S, Huxford T, Ghosh G. Ikappa Balpha functions through direct contacts with the nuclear localization signals and the DNA binding sequences of NF-kappaB. J Biol Chem. 1998; 273: 25427–25435.