Prostaglandin D Synthase in Host Immunity Induction and Function Of

Home , Cyclooxygenase, PGDS

Induction and Function of Lipocalin Prostaglandin D Synthase in Host Immunity Myungsoo Joo, Minjae Kwon, Ruxana T. Sadikot, Philip J. Kingsley, Lawrence J. Marnett, Timothy S. Blackwell, R. This information is current as Stokes Peebles, Jr, Yoshihiro Urade and John W. Christman of September 28, 2021. J Immunol 2007; 179:2565-2575; ; doi: 10.4049/jimmunol.179.4.2565 http://www.jimmunol.org/content/179/4/2565 Downloaded from

References This article cites 71 articles, 27 of which you can access for free at: http://www.jimmunol.org/content/179/4/2565.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• 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

by guest on September 28, 2021 *average

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 © 2007 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Induction and Function of Lipocalin Prostaglandin D Synthase in Host Immunity1

Myungsoo Joo,2* Minjae Kwon,* Ruxana T. Sadikot,† Philip J. Kingsley,‡ Lawrence J. Marnett,‡ Timothy S. Blackwell,* R. Stokes Peebles, Jr.,* Yoshihiro Urade,§ and John W. Christman†

Although mainly expressed in neuronal cells, lipocalin-type PGD synthase (L-PGDS) is detected in the macrophages infiltrated to atherosclerotic plaques. However, the regulation and significance of L-PGDS expression in macrophages are unknown. Here, we found that treatment of macrophages with bacterial endotoxin (LPS) or Pseudomonas induced L-PGDS expression. Epigenetic suppression of L-PGDS expression in macrophages blunted a majority of PGD2 produced after LPS treatment. Chromatin immunoprecipitation assays show that L-PGDS induction was regulated positively by AP-1, but negatively by p53. L-PGDS expression was detected in whole lung and alveolar macrophages treated with LPS or Pseudomonas. L-PGDS overexpressing Downloaded from transgenic mice improved clearance of Pseudomonas from the lung compared with nontransgenic mice. Similarly, intratracheal instillation of PGD2 enhanced removal of Pseudomonas from the lung in mice. In contrast, L-PGDS knockout mice were impaired in their ability to remove Pseudomonas from the lung. Together, our results identify induction of L-PGDS expression by inflammatory stimuli or bacterial infection, the regulatory mechanism of L-PGDS induction, and the protective role of L-PGDS expression in host immune response. Our study suggests a potential therapeutic usage of L-PGDS or PGD2 against Pseudomonas http://www.jimmunol.org/ pneumonia. The Journal of Immunology, 2007, 179: 2565–2575.

acrophages are key players in innate immunity. When resulting in activation of canonical I␬B kinase (IKK)3 and activated by endotoxin (LPS) or live bacteria, mac- MAPKs, including ERK, JNK, and p38 kinase (17). The activated M rophages in the lung trigger a series of inflammatory protein kinases increase transcriptional activities of NF-␬B and responses by producing cytokines, thereby recruiting neutro- AP-1, resulting in expression of inflammatory genes (17). On the phils (1). The inflammatory responses normally lead to clear- other hand, MyD88-independent Toll/IL-1R signaling activates ance of invading bacteria. Therefore, depletion of macrophages noncanonical the IKK complex, which activates NF-␬B and IFN by treating mice with liposomal clodronate impairs inflamma- regulatory factor-3 (18, 19). It has been shown that the MyD88- by guest on September 28, 2021 tory responses and clearance of bacteria including Pseudomo- dependent pathway regulates many classic markers of inflamma- nas from the lung (2–4). tion including cyclooxygenase (COX)-2 (20). Macrophages are responsible for the toxicity caused by LPS (5), LPS treatment induces COX-2 expression in macrophages, and one of the major bacterial factors that trigger inflammation, which NF-␬B activation is sufficient for COX-2 induction (21–23). Along is in part attributed to the fact that although expressed in various with the constitutively expressed isoenzyme COX-1, COX-2 con- cell types, TLR4 is highly expressed in macrophages (6–9). TLR4 verts arachidonic acid to PGH2, which serves as a precursor for is a receptor for LPS and is crucial for host innate immunity other prostanoids including PGD2 (24). PGH2 is converted to against bacterial infection (10–16). Engaged by LPS, TLR4 trig- PGD2 by two tissue-specific enzymes, lipocalin-type PGD syn- gers Toll/IL-1R-mediated signaling via MyD88 adaptor protein, thase (L-PGDS) and hemopoietic isotype (H-PGDS). However, these two enzymes share no homology in DNA and polypeptide sequences (25). Northern blot analyses also show that the expression profiles of these enzymes are distinctive, because L-PGDS is *Division of Allergy, Pulmonary and Critical Care Medicine, Department of Medi- mainly detected in CNS and related organs (26), whereas H-PGDS cine, Vanderbilt University Medical Center, Nashville, TN 37232; †Section of Pul- is in hemopoietic cells including macrophages (27). monary, Critical Care and Sleep Medicine, University of Illinois and the Jesse Brown PGD is involved in lung inflammation. PGD exacerbates Veterans Affairs Medical Center, Chicago IL 60612; ‡A. B. Hancock, Jr., Memorial 2 2 Laboratory for Cancer Research, Departments of Biochemistry and Chemistry, asthma (28–30), suggesting that PGD2 is proinflammatory. On the Vanderbilt Institute of Chemical Biology, Center in Molecular Toxicology, and the other hand, injection of PGD2 or retroviral delivery of PGD syn- Vanderbilt-Ingram Cancer Center, Vanderbilt University School of Medicine, Nash- ville, TN 37232; and §Department of Molecular Behavioral Biology, Osaka Bio- thase to the lung suppresses inflammatory processes elicited by science Institute, Osaka, Japan bleomycin and monosodium urate monohydrate crystal challenges

Received for publication October 2, 2006. Accepted for publication May 31, 2007. (31–33). In addition, PGD2 in a murine model of pleuritis caused 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 with 18 U.S.C. Section 1734 solely to indicate this fact. 3 Abbreviations used in this paper: IKK, I␬B kinase; COX, cyclooxygenase; L-PGDS, 1 lipocalin-type PGD synthase; H-PGDS, hemopoietic-type PGD synthase; BMDM, This work was supported by the Department of Veterans Affairs and by National bone marrow-derived macrophage; KO, knockout; i.t., intratracheal; PA103, P. Institutes of Health Grants HL061419, HL075557, HL066196, AI054660, and aeruginosa 103; BAL, bronchoalveolar lavage; siRNA, small interfering RNA; 15d- HL069449. ⌬12,14 PGJ2, 15-deoxy- -PGJ2; ChIP, chromatin immunoprecipitation; RT, reverse 2 Address correspondence and reprint requests to Dr. Myungsoo Joo, Allergy, Pul- transcription; moi, multiplicity of infection; DP, PGD2 receptor; CRTH2, chemoat- monary and Critical Care Medicine, Vanderbilt University School of Medicine, 1161 tractant receptor-homologous molecule expressed on Th2 cells. 21st Avenue South, MCN B1222, Nashville, TN 37232-2650. E-mail address: [email protected] Copyright © 2007 by The American Association of Immunologists, Inc. 0022-1767/07/$2.00 www.jimmunol.org 2566 LIPOCALIN PGD SYNTHASE EXPRESSION IN BACTERIAL IMMUNITY

FIGURE 1. Induction of COX-2 and production of

PGD2 by BMDM. COX-1 and COX-2 expression were measured in BMDM and RAW 264.7 cells. A, BMDM (1 ϫ 106 cells) were treated with LPS (1 ␮g/ml) for 4 h (lane 2) or with PA103 by moi 1 for the indicated time periods (lanes 4, 5, and 6) before cell harvest. An equal amount of total cell lysate was analyzed by SDS-PGDS and Western blotting (WB) for COX-2 expression (top) and COX-1 expression (bottom). B, RAW 264.7 cells (1 ϫ 106) were treated with LPS (lane 3) or PA103 (lanes 5 and 6), similar to A. To verify COX-2, 293 cells were transfected with a plasmid encoding a murine COX-2 (lane 1). C, BMDM (1 ϫ 106 cells) were treated with LPS (1 ␮g/ml) for 12 h in the presence or absence of CAY10404 (100 nM), a COX-2 speciﬁc inhibitor, and the cell culture supernatant was collected for PGD2 measurement. Each bar represents the mean PGD2 (nM) Ϯ SEM (for n ϭ 3 wells per treatment group), and p Ͻ 0.05 ,ء) the experiment was repeated three times

compared with other group). Downloaded from

by carrageenan is associated with resolution of lung inflammation, FVB/N background were described previously (37). Male C57BL/6 mice harboring homozygous deletion of L-PGDS gene, L-PGDS knockout (KO) suggesting that PGD2 exerts anti-inflammatory functions. These results clearly show that the outcome of PGD production depends mice, were described previously (38). Animal experiments were performed 2 per protocol approved by the Vanderbilt University Institutional Animal on the inflammatory milieu. Care and Use Committee (Nashville, TN). We made every effort to min- http://www.jimmunol.org/ Although not detectable in macrophages normally, L-PGDS ex- imize pain, discomfort, and the number of animals for the study. pression can be detected in macrophages infiltrated to the atherosclerotic plaques (34), suggesting that L-PGDS is aberrantly ex- LPS and PGD2 administration pressed in macrophages in a pathological environment. However, After sedation, mice were treated with intratracheal (i.t.) administration of the regulation of L-PGDS expression in macrophages and the sig- PGD2. Mouse tracheas were directly exposed by surgical resection, pierced with a 26-gauge needle, and injected with 50 ␮lofPGD (0.1 ␮g/g) diluted nificance of L-PGDS expression in a pathological condition re- 2 in sterile PBS. The dose of PGD2 was chosen based on the published dose main unknown. In this study, we attempted to identify the mech- ⌬12,14 of 15-deoxy- -PGJ2 (15d-PGJ2) administered i.t. (39). The neck anism that controls L-PGDS expression in macrophages. Given the wound was closed with sterile sutures under aseptic conditions. For i.p.

␮ by guest on September 28, 2021 role of PGD2 in lung inflammation, we examined the effect of injections, a single dose of 3 g of LPS per g of body weight was admin- L-PGDS expression on lung inflammation caused by Pseudomo- istered (40). nas lung infection, a common cause of nosocomial pneumonia and Bronchoalveolar lavage (BAL) fluid and total and differential the most serious respiratory pathogen in cystic fibrosis patients cell counts (35). Our findings revealed a novel regulatory mechanism of L- PGDS expression in macrophages and the protective function of After mice were asphyxiated with CO2, tracheas were cannulated, and lungs were lavaged in situ with sterile pyrogen-free physiological saline L-PGDS expression or PGD2 against bacterial lung infection. that was instilled in four 1-ml aliquots and gently withdrawn with a 1-ml tuberculin syringe. Lung lavage fluid was centrifuged at 400 ϫ g for 10 Materials and Methods min. Supernatant was kept at Ϫ70°C, the cell pellet was suspended in Reagents serum-free RPMI 1640, and total cell counts were determined on a grid hemocytometer. Differential cell counts were determined by staining cy-

PGD2, PGE2, and COX-2 inhibitor CAY10404 were purchased from Cay- tocentrifuge slides with a modified Wright stain (Diff-Quik; Baxter) and man Chemical. TLR4-specific Escherichia coli LPS was obtained from counting 400–600 cells in complete cross-sections. Alexis Biochemical, and doxorubicin was from Sigma-Aldrich. Various MAPK inhibitors including SB 220025 (p38 inhibitor), JNK inhibitor II, Bacterial infection and colony counting and U0126 (ERK1/2 inhibitor) were purchased from Calbiochem. Pseudomonas aeruginosa 103 (PA103) was cultured in a dialysate of trip- Cell culture ticase soy broth supplemented with 10 mM nitrilotriacetic acid (Sigma- Aldrich), 1% glycerol, and 100 mM monosodium glutamate. Macrophages Bone marrow-derived macrophages (BMDM) were obtained as described were incubated with Pseudomonas with 5 ϫ 104 CFU for up to 7 h. elsewhere (36). In short, cellular material from femurs of mice ranging Unless specified, 1 million bacteria in 100 ␮l of PBS were instilled by from 8 to 16 wk of age was flushed out with PBS. One million cells were i.t. injection with a 26-gauge needle to surgically exposed mouse tracheas. cultured at 37°C and 5% CO2 in DMEM (Cellgro) containing 4 mM L- The neck wound was closed with sterile sutures under aseptic conditions. glutamine (Invitrogen Life Technologies), 100 U of penicillin (Invitrogen Before the lung was harvested, the right ventricle was infused with 1 ml of Life Technologies) per ml, 0.1 mg of streptomycin (Invitrogen Life Tech- sterile PBS to remove blood from the lung tissue, and then the lungs were nologies) per ml, 10% FBS (HyClone), and 10% L929 cell culture medium removed aseptically and homogenized in 3 ml of sterile PBS. Lung ho- for 5 days to obtain mature macrophage. L929 cell culture medium con- mogenate was cultured overnight on soy base blood agar plate for bacterial tains GM-CSF necessary for macrophage differentiation and maturation. colony counting. BMDM were maintained in DMEM containing 10% FBS before experiment. A murine macrophage cell line, RAW 264.7 (American Type Culture Protein isolation and Western blot analysis Collection) were similarly cultured in DMEM containing 10% FBS. Total cell lysate was prepared with radioimmunoprecipitation assay cell Animal model lysis buffer (50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 2 mM EDTA, 1% sodium orthovanadate, 1% Triton X-100, 0.5% deoxycholate, 0.1% SDS) Male and female mice weighing 20–28 g were used in this study. Trans- supplemented with protease inhibitors (Roche). To obtain proteins from genic mice encoding human L-PGDS cDNA under the control of the tissue, harvested organs were quickly frozen in liquid nitrogen, and 200 mg chicken ␤-actin promoter and the CMV immediate early enhancer on a of tissue were suspended in radioimmunoprecipitation assay buffer. Tissue The Journal of Immunology 2567

transferred to polyvinylidene difluoride (Bio-Rad), which was incubated with appropriate Abs. Immune complex was detected by enhanced che- moluminescence (ECL plus; Amersham). COX-1 and -2 Abs were purchased from Cayman Chemical, and other Abs used in this study were purchased from Santa Cruz Biotechnology. DNA constructs and L-PGDS small interfering RNA (siRNA) cell line To construct L-PGDS siRNA plasmids, a 19-nt-long candidate sequence was located by using a software (OligoEngine) from 528 to 546 nt (5Ј- GGACGAGCTGAAGGAGAAA-3Ј) in the open reading frame of murine L-PGDS. The candidate sequence was further analyzed by BLAST search for potential homology. Those sequences were cloned into pSUPER.retro.puro (OligoEngine), and stably transfected into RAW 264.7 cells. Transfected cells were cultured in DMEM containing 2 ␮g/ml puromycin (Sigma-Aldrich) for selecting stable transfectants. Western blotting of L-PGDS was performed to confirm successful siRNA FIGURE 2. Induction of L-PGDS expression in macrophages. A, After colonies. A p53-expressing vector plasmid was a gift from Dr. Philip Hinds ϫ 5 ␮ BMDM (5 10 cells) were treated with LPS (1 g/ml) for the indicated (Tufts University School of Medicine, Boston, MA). For transfection of time points, an equal amount of total cell lysate was analyzed by SDS- macrophages, GenePORTER 2 (Gene Therapy Systems) was used per the PGDS and Western blotting (WB) for H-PGDS (top), L-PGDS (middle), suggested protocol of the manufacturer. 5 and ␣-tubulin (bottom) as internal controls. B, BMDM (5 ϫ 10 cells) were Downloaded from incubated with PA103 (moi 1) for the indicated periods. The total cell Prostanoid measurement lysate of differentially treated cells was prepared and analyzed by Western PGD2 and other prostanoids were measured by a liquid chromatographic- blotting for H-PGDS (top) and for L-PGDS (bottom). The experiment was electrospray ionization-mass spectrometry-mass spectrometry as previ- performed at least three times independently. ously described (41). Liquid chromatographic separation was per- ␮ ϫ formed isocratically on a Phenomenex Luna 3- mC18 5.0- 0.2-cm column. The mass spectrometer was operated in positive-ion electros- in the buffer was homogenized and incubated on ice for 15 min with oc- pray ionization mode. Detection of the analytes was accomplished by http://www.jimmunol.org/ casional vortexing. Cell debris was removed by centrifugation at 1000 ϫ selected reaction monitoring, using the following reactions: 3703317 3 g for 10 min at 4°C. Protein content was quantified by the Bradford assay (PGE2 and PGD2), 374 321 (PGD2-d4). The quantum was set to the (Bio-Rad) as specified by manufacturer. After SDS-PAGE, proteins were following parameters: capillary V ϭ 35 V; spray voltage ϭ 4.3 kV; by guest on September 28, 2021

FIGURE 3. Induction of L-PGDS in macrophages is responsible for a maximal production of PGD2. A, Epigenetic suppression of L-PGDS expression in RAW 264.7 cells. RAW 264.7 cells were stably transfected with a vector plasmid expressing L-PGDS siRNA to suppress L-PGDS expression. To analyze L- and H-PGDS expressions, the selected L-PGDS siRNA cell line (denoted as silenced) was treated with LPS (1 ␮g/ml) for the indicated periods, and Western blotting (WB) was performed for L-PGDS expression (top), H-PGDS expression (middle), and ␣-tubulin (bottom) for equal loading. Normal indicates a RAW 264.7 cell line transfected stably with an empty vector plasmid. LPS treatment induced L-PGDS expression (top). Expressions of H-PGDS (middle) and ␣-tubulin (bottom) were similarly analyzed by Western blotting. The experiment was performed three times, and additional L-PGDS siRNA cell lines similarly obtained were also examined. B, The normal and silenced cell lines were treated with LPS for 16 h, and total cell lysate of two cell lines was fractionated by SDS-PAGE and analyzed by Western blotting for L-PGDS (top). The membrane was striped and reblotted with ␣-tubulin Ab to ensure ϫ 5 equal loading (bottom). C, PGD2 production in the silenced cell line. The silenced and normal cell lines (5 10 cells, respectively) were treated with LPS OO for the indicated periods, and the cell culture supernatant of treated cells was collected for PGD2 measurement. , PGD2 produced in culture medium Ϯ ϭ of the normal; ––––,that of the silenced. Each point represents the mean PGD2 (nM) SEM (for n 3 wells per treatment group), and the experiment p Ͻ 0.05 compared with silenced). Three more siRNA cell lines obtained from RAW 264.7 cells were tested similarly, yielding ,ء) was repeated three times similar results. D, COX-2 expression in the silenced cell line. The parental cell line, RAW 264.7 cells, and the silenced cell line (1 ϫ 106 cells, respectively) were treated with LPS (1 ␮g/ml) for4htoinduce COX-2. The total cell lysate of variously treated cells was prepared and analyzed by Western blotting for COX-2 and ␤-actin. The experiment was performed three times. 2568 LIPOCALIN PGD SYNTHASE EXPRESSION IN BACTERIAL IMMUNITY Downloaded from

FIGURE 4. MAPKs are involved in induction of L-PGDS. A, RAW 264.7 cells (1 ϫ 106 cells) were pretreated with either vehicle, DMSO (lanes 1 and 2) or various MAPK inhibitors (10 ␮M each) such as JNK inhibitor (lane 3), SB 220025 (p38 inhibitor; lane 4), and U0126 (ERK1/2 inhibitor; lane 5), as recommended by the manufacturer. Pretreated cells were subsequently treated with LPS (1 ␮g/ml) for 16 h to induce L-PGDS (lanes 2–5). Total cell lysate was prepared from variously treated RAW 264.7 cells for Western blotting (WB) analysis of L-PGDS and ␣-tubulin for equal loading. L-PGDS expression induced by LPS treatment was used as a positive control (lane 2). The experiment was repeated at least three times. B, Before virus infection, RAW 264.7 cells were transfected with an NF-␬B-luciferase reporter construct, along with a tk-Renilla luciferase reporter construct. The transfected cells http://www.jimmunol.org/ were infected with recombinant adenoviruses encoding either eGFP (Ad-eGFP; lanes 2 and 3)orI␬B␣S32,36A (Ad-I␬BDN; lanes 4 and 5) with an moi of 0.5 for 1 h and subsequently treated with LPS (1 ␮g/ml) for 16 h to induce L-PGDS (lanes 3 and 5). Total cell lysate was analyzed by Western blotting for L-PGDS expression and ␣-tubulin for equal loading. The experiment was repeated at least three times. C, To verify the efﬁcacy of Ad-I␬BDN in suppressing NF-␬B activity, dual luciferase assay was performed. Each bar represents the mean relative light units Ϯ SEM (for n ϭ 3 wells per treatment .p Ͻ 0.05 compared with other group ,ء .group), and the experiment was repeated three times

capillary temperature ϭ 300°C; tube lens V ϭ 137 V; sheath gas ϭ 49 area, generating a 225-bp PCR product. Abs used for immunoprecipitation ␺; auxiliary gas ϭ 25 (no units); collisionally induced dissociation pres- were purchased from Santa Cruz Biotechnology. To exclude nonspecific sure ϭ 1.0 mTorr. These values were observed to maximize the response precipitation of DNA bound protein, isotypic IgG (Santa Cruz Biotechnol- by guest on September 28, 2021 of the simplified risk model transitions used. Collision energy was set to 13 ogy) was used in parallel. eV for both reactions. Quantitation was accomplished by stable isotope dilution. Semiquantitative RT-PCR Chromatin immunoprecipitation (ChIP) assay Total RNA was prepared with a RNeasy kit (Qiagen) per the manufacturer’s manual. Reverse transcription (RT) of 2 ␮g of total RNA was per- Reagents were obtained from Upstate Biotechnology, and assay was performed with murine leukemia virus reverse transcriptase and a random formed as described previously (42). Briefly, we grew cells to 90% con- hexamer primer (PerkinElmer) to generate cDNA. Actin cDNA level from 7 fluence with 1 ϫ 10 cells for each experiment. After treated with 1% each sample was used to normalize the samples for differences in PCR formaldehyde for 5 min, cells were harvested, suspended in SDS-lysis efficiency. L-PGDS mRNA quantity was determined by using end-point buffer (50 mM Tris-HCl, pH 8.1, 10 mM EDTA, 1% SDS, protease in- dilution PCR, including three serial 1/5 dilutions (1/5, 1/25, and 1/125) of hibitors) and underwent sonication (4 times for 12 s each at one-fifth of the RT products for PCR amplification. To eliminate genomic DNA contam- maximum potency). After centrifugation at 4°C for 10 min, supernatants ination, equal amounts of total RNA from each sample were PCR amplified were diluted 1/10 with dilution buffer (16.7 mM Tris-HCl (pH 8.1), 1.2 without RT reaction. A portion of the cDNA was amplified with 0.5 U of mM EDTA, 167 mM NaCl, 0.01% SDS, 1.1% Triton X-100) and added Taq polymerase (PerkinElmer) and appropriate oligonucleotides at 94°C with salmon sperm-saturated protein A (Zymed) for2hat4°Ctoremove for 40 s, 60°C for 30 s, and 72°C for 40 s for 35 cycles with an initial 4 min nonspecific Ig. Immunoprecipitation was performed by adding 1 ␮gof denaturation at 95°C and final 10 min of extension at 72°C. The oligonu- specific Abs to the cell lysate overnight at 4°C. Immune complex was cleotides used in this study were as follows: L-PGDS forward primer, captured with 40 ␮l of salmon sperm-saturated protein A and washed twice 5Ј-GGTTCCGGGAGAAGAAAGCT-3Ј; L-PGDS reverse primer, 5Ј-CA (5 min each at 4°C) with low-salt buffer (0.1% SDS, 1% Triton X-100, 2 CTGACACGGAGTGGATGC-3Ј; ␤-actin forward primer, 5Ј-AGAGGG mM EDTA, 20 mM Tris-HCl (pH 8.1), and 150 mM NaCl), once with AAATCGTGCGTGAC-3Ј; and ␤-actin reverse primer: 5Ј-CAA high-salt buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris- TAGTGATGACCTGGCCGT-3Ј. HCl (pH 8.1), and 500 mM NaCl), once with LiCl buffer (0.25 M LiCl, 1% Nonidet P-40, 1% deoxycholate, 1 mM EDTA, and 10 mM Tris-HCl, pH 8.1), and twice with Tris-EDTA buffer for 5 min. The immune complex Recombinant adenovirus and luciferase assay was extracted three times with 200 ␮l of elution buffer (1% SDS, 0.1M ␬ ␣S32,36A NaHCO3). Elutes were heated at 65°C for at least4htoreverse formal- Recombinant adenoviruses encoding either enhanced GFP or I B dehyde cross-linking. The samples were treated with 10 ␮g of proteinase and the NF-␬B luciferase reporter construct were described previously K at 45°C for 1 h. The recovered DNA was purified with a DNA cleanup (23). DNA was transfected to RAW 264.7 cells by GenePORTER 2 kit (Qiagen), and samples of input DNA were also prepared in the same (Gene Therapy Systems) for 24 h. Transfected cells were infected with way. PCR conditions were as follows: 94°C for 240 s; 30–32 cycles at the viruses with multiplicity of infection (moi) of 0.5 for 1 h. After 94°C for 40 s; 54°C for 40 s; 72°C for 60 s; and final elongation at 72°C removing viral inoculum, we added to the infected cells a fresh culture for 10 min. PCR for the input was performed with 100 ng of genomic medium with LPS (1 ␮g/ml) to induce L-PGDS. Luciferase assay was DNA. The PCR products ran on either 1% agarose or 8% polyacrylamide performed with a dual luciferase kit and the manufacturer’s manual gel. Primers were 5Ј-GTTCAAGGCACAATGGTGCTT-3Ј and 5Ј-TCCA (Promega). NF-␬B-driven luciferase activity was normalized with tk- GAGGCAGAACTGGCTCAG-3Ј. This primer set covers the AP-1 site Renilla luciferase activity. The Journal of Immunology 2569 Downloaded from http://www.jimmunol.org/ FIGURE 5. AP-1 is involved in induction of L-PGDS. A, A schematic of a murine L-PGDS promoter indicates various putative binding sites of transcription factors. A putative AP-1 binding site, underlined, was located between Ϫ292 and Ϫ302 nt from a transcription initiation site (bold, A). The AP-1 site is partially overlapped with a putative p53 binding site located between Ϫ296 and Ϫ305 nt (bold italics). Translation starting codon, ATG, is indicated at the end of sequence. B, Binding of c-Jun, a subunit of AP-1, to the putative AP-1 site of L-PGDS promoter in vivo was monitored by ChIP assays with an anti (␣-)-cJun Ab. Before LPS (1 ␮g/ml) treatment for the indicated periods, RAW 264.7 cells were treated with DMSO (lanes 1–4), JNK inhibitor (lanes 5 and 6), or p38 inhibitor (lanes 7 and 8) for 1h. RAW 264.7 cells treated with LPS (1 ␮g/ml) for 4 h were used for ChIP assay with isotypic IgG to exclude nonspecific DNA precipitation (lane 9). DNA coprecipitated with c-Jun was amplified by PCR with the primers flanking the AP-1 site (denoted as AP-1). As input controls for PCR, 100 ng of genomic DNA were used (denoted as Input). To ensure the fidelity of PCR reaction, some of PCR products were subcloned and sequenced. The experiment was repeated at least three times. by guest on September 28, 2021

Statistical analysis indicating that COX-2 activity is responsible for a majority of For comparison among groups, paired or unpaired t tests and one-way PGD2 produced by macrophages after LPS treatment. ANOVA tests were used (with the assistance of InStat, Graphpad Software; p values Ͻ.05 are considered significant. L-PGDS expression is induced by LPS or Pseudomonas treatment of macrophages Results Because H-PGDS is expressed in macrophages (43), we examined COX-2 expression is responsible for PGD2 production in macrophages whether PGD2 produced by LPS treatment correlates with an increase of H-PGDS expression. BMDM were treated with LPS for Because COX-1 and COX-2 synthesize PGH2, the precursor of various periods, and the total cell lysate of the treated cells was PGD2 (24), we first examined the expression profiles of COX-1 analyzed by Western blotting. As shown in Fig. 2A, H-PGDS was and of COX-2 in macrophages. BMDM were prepared from mice expressed constitutively, which was not altered by LPS treatment and treated with LPS or P. aeruginosa 103 (PA103). The treated (Fig. 2, top). Rather, LPS treatment induced L-PGDS expression in cells were analyzed by Western blotting for COX-1 and COX-2 macrophages (Fig. 2, middle). expression. As shown in Fig. 1A, LPS treatment of BMDM in- To examine whether live bacteria induce L-PGDS expression, duced COX-2 expression (Fig. 1A, top, lanes 1 and 2). Similarly, we performed a similar experiment with PA103. As shown in Fig. PA103 treatment of BMDM induced COX-2 expression (Fig. 1A, 2B, PA103 treatment also induced L-PGDS expression, without top, lanes 3–6). However, neither LPS nor PA103 affected the altering H-PGDS expression in macrophages. Together, these re- constitutive expression of COX-1 (Fig. 1A, bottom). We also per- sults reveal that L-PGDS expression is inducible by LPS or bac- formed a similar experiment with a murine macrophage cell line, terial infection. RAW 264.7 cells (Fig. 1B). Treatment of RAW 264.7 cells with LPS or PA103 induced COX-2 expression (Fig. 1B, top), but not affected the constitutive expression of COX-1 (Fig. 1B, bottom). Induction of L-PGDS expression accounts for a majority of PGD2 produced by macrophages Next, to determine the impact of COX-2 expression on PGD2 production, we treated BMDM with LPS in the presence or ab- Next, to determine the impact of L-PGDS expression on PGD2 sence of a COX-2-specific inhibitor, CAY10404. At 12 h after LPS production, we generated RAW 264.7-derived cell lines that en- treatment, the cell culture supernatant was collected for measure- code a siRNA specific for L-PGDS mRNA to suppress L-PGDS ment of PGD2 released in the culture medium. As shown in Fig. expression epigenetically. As shown in Fig. 3, A and B,anL- 1C, the COX-2 inhibitor abolished PGD2production by BMDM, PGDS siRNA cell line (denoted as silenced) blunted the induction 2570 LIPOCALIN PGD SYNTHASE EXPRESSION IN BACTERIAL IMMUNITY

of L-PGDS expression elicited by LPS without altering constitutive H-PGDS expression. With the L-PGDS siRNA cell line, we

determined the production proﬁle of PGD2 (Fig. 3C). Compared with the control cell line, RAW264.7 cells stably transfected with an empty vector plasmid (denoted as normal), the silenced cell line

produced significantly lower amounts of PGD2. To confirm that the decrease of PGD2 production was a specific effect of L-PGDS silencing, we tested three additional L-PGDS siRNA cell lines derived from RAW 264.7 cells and obtained similar results (data

not shown). Because PGD2 production correlated with COX-2 expression, as shown in Fig. 1C, we examined the possibility that low

PGD2 production results from impaired COX-2 expression in the silenced cell line. As shown in Fig. 3D, the silenced cell line expressed COX-2 to a similar level as the parental cell line, indicat-

ing that impaired COX-2 expression did not cause the low PGD2 production in the silenced cell line. Together, these results suggest FIGURE 6. Suppression of L-PGDS expression by p53. A, RAW 264.7 that L-PGDS is a key enzyme for the maximal production of PGD2 cells were transfected with either an empty (lanes 1 and 2) or a p53- in LPS-treated macrophages. expressing vector plasmid (lanes 3 and 4). Transfected cells were treated

with LPS (1 ␮g/ml) for 16 h to induce L-PGDS (lanes 2 and 4), which was Downloaded from JNK and p38 kinase are involved in L-PGDS expression measured by Western blot (WB). To ensure equal loading, ␤-actin was Because LPS binding to TLR4 activates MAPKs and NF-␬B (44, measured by Western blotting after stripping the membrane. B, RAW 264.7 ␮ 45), we tested the potential role of MAPKs in L-PGDS expression. cells were treated with LPS (1 g/ml; lanes 2 and 4) and subsequently with ␮ Before LPS treatment, RAW 264.7 cells were treated with various doxorubicin (Doxo; 1 M)for6h(lanes 3 and 4) before cell harvest at 16 h after LPS treatment. Total cell lysate was prepared for Western blot MAPK inhibitors (10 ␮M) for 1 h. At 16 h after LPS treatment, analysis of L-PGDS (top), p53 (middle), and ␤-actin (bottom). LPS induced total cell lysate was analyzed by Western blotting for L-PGDS. As L-PGDS (lane 2), and doxorubicin induced p53 (lanes 3 and 4). C, Tem- http://www.jimmunol.org/ shown in Fig. 4A, the inhibitor of either JNK or p38 kinase, but not poral bindings of cJun and p53 to their cognate sites were examined by that of ERK1/2, suppressed L-PGDS induction, indicating that ChIP assay. After treatment with LPS for indicated periods, RAW 264.7 JNK and p38 kinase activities led to L-PGDS induction. cells (5 ϫ 105 cells per experimental group) were treated and harvested for Next, to examine the potential role of NF-␬B in L-PGDS ex- ChIP assay. The cell lysate in each time point was equally divided for pression, we tested whether inhibition of NF-␬B activity affects immunoprecipitation of c-Jun and p53, respectively. DNA bound to c-Jun L-PGDS expression. Before LPS treatment, RAW 264.7 cells were (top) and DNA bound to p53 (middle) were ampliﬁed by PCR with the transfected with NF-␬B reporter construct for 24 h and subse- primer ﬂanking both AP-1 and p53 binding sites. The cell lysate of 16 h of quently infected with a recombinant adenovirus encoding LPS treatment was used for ChIP assay with isotypic IgG to exclude non-

specific DNA precipitation (lane 6), and 100 ng of genomic DNA were by guest on September 28, 2021 I␬B␣S32,36A, a dominant negative inhibitor of NF-␬B (23). After used as input control for PCR (Input). The experiment was repeated three 16 h of LPS treatment, total cell lysate was prepared for Western times. blot analysis of L-PGDS and for luciferase assay. As shown in Fig. 4B, suppressed NF-␬B activity did not alter L-PGDS expression induced by LPS (Fig. 4B, lanes 3 and 5), suggesting that NF-␬Bis not involved in L-PGDS expression. To verify the efficacy of the p53 induced by LPS suppresses L-PGDS expression recombinant adenovirus in suppressing NF-␬B activity, we mea- Promoter analysis in Fig. 5A reveals that the AP-1 site was par- sured NF-␬B-driven luciferase activity, showing effective suppres- tially overlapped with a p53 binding site. Given the late induction sion of NF-␬B-driven luciferase activity by the recombinant ade- of p53 in LPS-treated macrophages (47), we tested whether p53 novirus (Fig. 4C). Together, these results suggest that L-PGDS functions as a competitor of AP-1. First, to examine whether p53 expression is mediated by JNK and p38 kinase, but not by NF-␬B. affects L-PGDS expression, we transfected RAW 264.7 cells with a p53-expressing plasmid, and the transfected cells were subse- AP-1 is involved in L-PGDS expression quently treated with LPS. Western blot analysis in Fig. 6A shows Our finding that JNK and p38 MAPK were associated with L- that ectopic expression of p53 suppressed L-PGDS expression PGDS expression suggests that a transcription factor activated by (Fig. 6, lanes 2 and 4). Because it is possible that ectopic expres- both JNK and p38 kinase is involved. To identify the transcription sion of p53 induces apoptosis of the transfected cells, resulting in factor, we analyzed the promoter sequence of murine L-PGDS a low expression of L-PGDS, we performed similar experiment by gene and located an AP-1 binding site from Ϫ302 to Ϫ292 nt of treating RAW 264.7 cells with doxorubicin for6htoinduce en- the promoter (Fig. 5A). Given that AP-1 is activated by both JNK dogenous p53, in which significant cell death did not occur (48). and p38 (46), we performed ChIP assay to examine whether the As shown in Fig. 6B, p53 expression induced by doxorubicin treat- putative AP-1 site is involved in L-PGDS expression. RAW 264.7 ment blunted L-PGDS induction elicited by LPS. These results cells were treated with LPS in the absence or presence of 10 ␮M show that p53 functioned as an inhibitory factor in L-PGDS concentrations of either JNK kinase or p38 kinase inhibitor. Nu- expression. clear fraction was prepared, and DNA cross-linked to c-Jun, a To investigate whether the overlapping binding sites of AP-1 subunit of AP-1, was immunoprecipitated by a c-Jun-specific Ab and p53 are associated with the inhibitory function of p53 in L- and amplified by PCR with the primers flanking the AP-1 site. As PGDS expression, we performed ChIP assay for the bindings of shown in Fig. 5B, LPS treatment induced c-Jun binding to the p53 and c-Jun to their cognate sites. From RAW 264.7 cells treated cognate site (lanes 1–4), which was blocked by either JNK inhib- with LPS, DNA bound to each protein was coprecipitated by the itor (Fig. 5B, comparing lanes 3 and 6) or p38 kinase inhibitor corresponding Abs and amplified by PCR with a set of primers (Fig. 5B, comparing lanes 3 and 8). Together, our results indicate encompassing both the AP-1 and p53 sites. As shown in Fig. 6C, that AP-1 is involved in inducing L-PGDS expression. LPS treatment induced c-Jun binding to the cognate site (Fig. 6C, The Journal of Immunology 2571

sions. As shown in Fig. 7A, LPS challenge induced L-PGDS mRNA expression in the lung, without affecting constitutive expression of H-PGDS expression. To examine whether alveolar macrophages are capable of expressing L-PGDS, we prepared alveolar macrophages from BAL ﬂuid, treated them with LPS or PA103 for 16 h, and performed Western blotting to detect L-PGDS expression. As shown in Fig. 7B, either LPS or PA103 treatment induced L-PGDS expression in alveolar macrophages. Together, these results suggest that L- PGDS expression is induced during lung inﬂammation.

L-PGDS or PGD2 enhances removal of Pseudomonas infected to the lung

FIGURE 7. Induction of L-PGDS in the inflamed lung and activated Next, to examine the effect of L-PGDS on Pseudomonas lung in- alveolar macrophages. A, C57BL/6 mice (n ϭ 4) were injected i.p. with fection, transgenic mice overexpressing L-PGDS and nontrans- LPS (3 ␮g/g of mouse) for indicated periods. After various treatments, total genic littermate control mice received i.t. injection of PA103 (1 ϫ 6 RNA was extracted from the lung of C57BL/6 mice and analyzed by semi- 10 CFU/mouse) without significant mortality. At 24 h after bac- quantitative RT-PCR for L-PGDS mRNA expression (top), H-PGDS (mid- terial infection, lung was harvested and the bacteria in the lung dle), and ␤-actin as internal controls (bottom). To exclude a false positive were counted. As shown in Fig. 8A, the number of bacteria in the Downloaded from by carryover genomic DNA, PCR was performed without RT (ϪRT; lane L-PGDS-transgenic mice was lower than in control mice, suggest- 10). The fidelity of PCR was verified by performing PCR with cDNAs of ing that the L-PGDS-transgenic mice are more effective in clearing ␤ L- and H-PGDS and -actin (plasmid; lane 9). Two representative results Pseudomonas than are the control mice. from each experimental group were shown. B, Alveolar macrophages from Because the L-PGDS is expressed in the lung of transgenic mice BAL fluid of mice were treated with LPS (1 ␮g/ml) or PA103 (moi 1) for B 16 h. Total cell lysate was analyzed by Western blotting (WB) for L-PGDS (Fig. 8 ) and produce PGD2 (37), we tested whether PGD2 instil- and ␤-actin. The experiment was repeated three times. lation to the lung also enhances bacterial clearance. C57BL/6 mice http://www.jimmunol.org/ received i.t. injections of PA103, along with either PGD2 or vehicle (DMSO). At 24 h after bacterial injection, the bacteria in the lung

lanes 2 and 3), which was replaced by p53 binding, as the cells were counted. As shown in Fig. 8C, PGD2 i.t. injection reduced the were incubated with LPS for longer periods (Fig. 6C, lanes 3–5). bacterial titer in the lung compared with the vehicle-treated mice.

Together, these results show that AP-1 and p53 regulated L-PGDS Together, these results suggest that L-PGDS expression or PGD2 expression by competing with each other for their overlapped contributes to removal of bacteria from the lung. binding sites, suggesting a self-regulatory mechanism of L-PGDS expression induced by LPS. L-PGDS or PGD2 contributes to neutrophil influx to the lung by guest on September 28, 2021 In an effort to determine the mechanism by which L-PGDS ex- L-PGDS expression is induced in the inflamed lung and in pression or PGD contributes to host defense against Pseudomo- alveolar macrophages 2 nas, we tested whether L-PGDS expression or PGD2 instillation Because the lung harbors abundant macrophages and PGD2 is im- affects inflammatory cell influx to the lung. Because macrophages plicated in lung inflammation (31–33), we examined whether L- and neutrophils are predominant immune cells in the lung after PGDS is inducible in inflamed lung. A single i.p. injection of mice PA103 challenge, neutrophils are critical in bacterial clearance, with LPS elicits lung inflammation (40). Thus, we injected and delayed neutrophil infiltration is related with impaired clear- C57BL/6 mice i.p. with LPS, and after various time points total ance of Pseudomonas from the lung (4), we measured neutrophil RNA was extracted from the lungs of the treated mice for semi- influx to the lung. The L-PGDS-expressing transgenic mice and quantitative RT-PCR analyses of L- and H-PGDS mRNA expres- nontransgenic control mice received i.t. injection of PA103, and

FIGURE 8. L-PGDS or PGD2 enhances clearance of PA103 from the mouse lung. A, Either transgenic mice (n ϭ 4; L-PGDS T/G) or wild-type littermate (n ϭ 3; WT) were injected i.t. 1 ϫ 106 CFU of PA103 per mouse. At 24 h after bacteria injection, total lung was harvested, and bacterial colonies were counted. Each p Ͻ ,ء .bar represents the mean CFU (log 4/ml) Ϯ SEM 0.05 compared with wild type. B, Lung and liver of transgenic (T/G) and nontransgenic littermate control mice (WT) were obtained, and an equal amount of proteins was loaded for Western blot (WB) analysis of L- PGDS. Results of mice from three different experimental settings are shown. C, Experiment similar to A was performed by using C57BL/6 mice (n ϭ 6). The mice ␮ were injected i.t. with either PGD2 (0.1 g/g) or DMSO, along with 1 ϫ 106 CFU of PA103. At 24 h after i.t. injection of bacteria, bacterial colonies were counted from the lung. Bacterial colony number of

PGD2-treated mice was compared with vehicle- .p Ͻ 0.05 ,ء .treated mice 2572 LIPOCALIN PGD SYNTHASE EXPRESSION IN BACTERIAL IMMUNITY Downloaded from

FIGURE 10. L-PGDS KO mice are less effective in clearing PA103

FIGURE 9. L-PGDS or PGD2 is associated with neutrophil recruitment from the lung. A, BMDM of wild-type control and L-PGDS KO mice were to the lung. A, Either transgenic mice (n ϭ 4; L-PGDS T/G) or wild-type prepared, treated with LPS for 16 h, and analyzed by Western blotting littermates (n ϭ 4; WT) were injected i.t. with 1 ϫ 106 CFU of PA103 per (WB) for L-PGDS expression (top) and H-PGDS (bottom). B, Bacterial mouse. At 3 h after bacteria injection, BAL fluid was collected for counting colonies were counted from the lungs of L-PGDS KO mice (n ϭ 4) or http://www.jimmunol.org/ p Ͻ 0.05 wild-type littermate control mice (n ϭ 4) that were injected i.t. with 5 ϫ ,ء .neutrophils. Each bar represents the mean cell number Ϯ SEM compared with wild type. B, Similar experiment was performed. At 24 h 105 CFU of PA103 per mouse. At 24 h after bacteria injection, total lung after bacteria injection, when no significant mortality was observed, neu- was harvested for counting bacteria. Each bar represents the mean CFU p Ͻ 0.05 compared with wild type. C, Neutrophils ,ء .trophils in BAL fluid were counted. Each bar represents the mean cell (log 2/ml) Ϯ SEM number Ϯ SEM and there was no statistical difference between two groups. were counted in BAL fluid of L-PGDS KO and wild-type littermate mice C, C57BL/6 mice (n ϭ 4 per experimental group) were received i.t. injec- (n ϭ 4 each) at 3 h after receiving i.t. injection of PA103 (1 ϫ 106 CFU Ͻ ء ␮ ϫ 6 Ϯ tion of either DMSO or PGD2 (0.1 g/g), along with PA103 (1 10 CFU per mouse). Each bar represents the mean cell number SEM. , p 0.05 per mouse). At 3 h after i.t. injection, neutrophils in BAL fluid were compared with wild type. D, A similar experiment was performed with p Ͻ 0.05 L-PGDS KO and wild-type littermate mice (n ϭ 4 each) at 24 h after ,ء .counted. Each bar represents the mean cell number Ϯ SEM by guest on September 28, 2021 compared with DMSO-injected mouse group. D, Similar experiment with receiving i.t. injection of PA103 (5 ϫ 105 CFU per mouse). Each bar p Ͻ 0.05 compared with wild ,ء .C57BL/6 mice (n ϭ 4 per experimental group) was performed. At 24 h represents the mean cell number Ϯ SEM after i.t. injection, neutrophils in BAL fluid were counted. Each bar rep- type. resents the mean cell number Ϯ SEM, and there was no statistical difference between two groups.

received the similar dose used for transgenic mice. At 24 h after BAL fluid of the treated mice was obtained for neutrophil counting PA103 i.t. injection, the lung was harvested, and bacterial colonies at 3 h after PA103 instillation. As shown in Fig. 9A, the L-PGDS were counted. As shown in Fig. 10B, L-PGDS KO mice were less expressing transgenic mice recruited more neutrophils than the effective in clearing PA103 from the lung. wild-type littermate in a given time. A similar experiment was To examine whether the impaired bacterial clearance is associ- performed by i.t. injection of C57BL/6 mice with PGD2 or vehicle, ated with disturbed neutrophil recruitment, we measured neutro- along with PA103. As shown in Fig. 9C, similar to L-PGDS-trans- phils in the lung after PA103 challenge. The L-PGDS KO mice genic mice, PGD2 injection resulted in greater neutrophil influx in and wild-type littermate control mice received i.t. injection of the early time point than in vehicle-treated mice. Alternatively, PA103 (1 ϫ 106 CFU/mouse), and BAL fluid of the treated mice neutrophil influx to the lung at 24 h after PA103 injection in the was obtained for neutrophil counting at 3 h after PA103 instilla- transgenic mice (Fig. 9B) or in C57BL/6 mice that received i.t. tion. As shown in Fig. 10C, the number of neutrophils in the lung injection of PGD2 was not statistically different from results in of L-PGDS KO mice was lower than in wild-type controls. We control mice (Fig. 9D). There was no neutrophil in the lung with- performed similar experiment by injecting PA103 (5 ϫ 105 CFU/ out PA103 challenge in both transgenic and PGD2-instilled mice. mouse) for 24h. In this experiment, we lowered the dose of

Together, our results suggest that L-PGDS expression or PGD2 Pseudomonas because of lethality (data not shown). As shown in facilitates neutrophil recruitment to the lung. Fig. 10D, the number of neutrophils in KO mice was higher than in control mice at 24 h after PA103 challenge. There was no neu- Genetic deletion of L-PGDS impairs effective removal of trophil in the lung without PA103 injection. These results suggest Pseudomonas from the lung that ineffective clearance of bacteria from the lung of L-PGDS KO

Because L-PGDS expression or PGD2 instillation to the lung en- mice is associated with impaired recruitment of neutrophils to hances clearing Pseudomonas, we examined whether lack of L- the lung. PGDS impairs clearance of the bacteria from the lung. L-PGDS KO mice received i.t. injection of PA103 (Fig. 10A). For the ex- Discussion periment, we used a low dose of PA103 (5 ϫ 105 CFU/mouse) L-PGDS is constitutively expressed in the CNS and related organs because of a high mortality of L-PGDS KO mice when the mice (26) but is also detectable in the macrophages of atherosclerosis The Journal of Immunology 2573

plaques (34). These results suggest induction of L-PGDS expres- Pseudomonas challenge. Neutrophils are professional phagocytic sion in macrophages in a pathological environment. However, the cells that clear bacterial pathogens including Pseudomonas from induction mechanism and the role of L-PGDS expression in mac- the lung, and thus impaired recruitment of neutrophils undermines rophages are unknown. Given the noncanonical expression of L- the capability of a host to remove Pseudomonas from the lung (4). PGDS in macrophages (34), we used various macrophages to study Therefore, it is possible that L-PGDS expression and thereby in-

the mechanism of L-PGDS expression. Because PGD2 is closely creased PGD2 production during lung inﬂammatory process facil- associated with lung inﬂammation, we used Pseudomonas lung itate neutrophil recruitment, contributing in part to effective re- infection as a model for studying the function of L-PGDS expres- moval of bacterial pathogens.

sion. Here, we found that LPS or Pseudomonas treatment induced The effect of PGD2 on lung inflammation is complex because L-PGDS expression in macrophages and in mice, which involved PGD2 either promotes or suppresses inflammation depending on AP-1 and p53. In mice, L-PGDS expression or PGD2 treatment the inflammatory milieu (24, 51–54). The net effect of PGD2 could facilitated removal of Pseudomonas from the mouse lung. To- be more complicated because of the fact that PGD2 undergoes gether, our study suggests that L-PGDS expression is induced by nonenzymatic processes to generate 15d-PGJ2, an anti- inflammatory stimuli and contributes to host defense against bac- inflammatory lipid (55, 56). In RAW 264.7 cells, 15d-PGJ2 was terial lung infection. composed of approximately one-tenth of PGD2 in cell culture me- ␬ Our results show induction of L-PGDS expression by inflam- dium (data not shown). 15d-PGJ2 suppresses NF- B activity by matory stimuli. Given that atherosclerosis involves inflammation forming an adduct with NF-␬B and IKK (57, 58) and by activating ␥ (49), our results suggest that L-PGDS expression in the macro- peroxisome proliferator-activated receptor- (59, 60). PGD2 exerts Downloaded from phages of the plaques is likely due to inflammatory stimuli in the its effect by binding to its receptors, PGD2 receptor (DP; Ref. 61) atherosclerotic plaques. Because our results show that COX-2 ex- and chemoattractant receptor-homologous molecule expressed on

pression directly affects the amount of PGD2, it is conceivable that Th2 cells (CRTH2) receptor (62). 15d-PGJ2 also binds to these induced L-PGDS plays a role for coping with a surge of PGH2 due receptors with a similar avidity (63). Although associated with to induction of COX-2 expression in macrophages activated by anti-inﬂammatory function of PGD2 in in vitro studies, DP recep-

inﬂammatory stimuli. tor is related with the proinﬂammatory function of PGD2 in many

COX-2 expression in macrophages occurs early after LPS treat- in vivo experimental settings (64). Alternatively, CRTH2 is mostly http://www.jimmunol.org/ ␬ ment and involves NF- B that is rapidly activated by LPS (23). associated with proinﬂammatory function of PGD2 (64). In mice, L-PGDS expression in macrophages was detectable 8 h after LPS DP receptor is expressed in various tissues including lung (64), treatment and involved AP-1 rather than NF-␬B. Therefore, it is whereas CRTH2 expression is mostly conﬁned to hemopoietic plausible that macrophages express COX-2 before L-PGDS ex- cells such as Th2 cells, eosinophils, basophils, mast cells, and a pression by differentially using key transcription factors, ensuring subset of monocytes (64). Given that Pseudomonas lung infection

that a sufﬁcient amount of PGH2 is available for L-PGDS to pro- predominantly elicits an inﬂux of neutrophils in the lung (4) and duce PGD2. We also show that p53 suppresses L-PGDS expression that alveolar macrophages are critical in removing bacteria includ- by competing with AP-1 for the partially overlapping AP-1 and ing Pseudomonas (2–4), it is possible that, in conjunction with by guest on September 28, 2021 p53 binding sites, to which AP-1 and p53 bound sequentially after Pseudomonas, PGD2 delivered to the lung acts on the DP receptor LPS treatment. Thus, our results uncovered a unique self-regula- of alveolar macrophages, increasing innate immune activity. In tory mechanism of L-PGDS expression. light of the role of lung epithelial cells against Pseudomonas lung

It is not clear why neuronal cells express constitutively infection (65), it is also plausible that PGD2 binds to the DP re- L-PGDS, whereas macrophages did so in a signal-dependent man- ceptor of both alveolar macrophages and lung epithelial cells, co- ner. Given that AP-1 is expressed in various cell types and acti- operatively clearing Pseudomonas from the lung. Our results did

vated by various stimuli, it is possible that neuronal cells express not exclude a possible involvement of anti-inﬂammatory 15d-PGJ2 AP-1 abundantly and constitutively, resulting in the constitutive by directly inactivating NF-␬B activity. Given the fact that a sig-

expression of L-PGDS. In RAW 264.7 cells, AP-1 expression was niﬁcantly high amount of 15d-PGJ2 is required to execute these

low but increased signiﬁcantly by LPS treatment (data not shown), activities (57, 66), it is likely that 15d-PGJ2 generated from PGD2 which may explain, at least in part, the signal dependent expres- could bind to DP receptor, contributing to PGD2 effect in our ex- sion of L-PGDS in macrophages. Alternatively, unlike macro- perimental settings. phages, a signaling or stimulation peculiar to neuronal cells causes Given the induction of L-PGDS expression in macrophages by a constant binding of AP-1 to L-PGDS promoter. Pseudomonas treatment, it is likely that Pseudomonas infection As shown in Fig. 7, L-PGDS mRNA expression was induced in induces L-PGDS expression in the lung as well. However, Pseudo- mouse lung challenged with LPS. In addition, alveolar macro- monas lung infection also induces COX-2 expression in mice (67)

phages extracted from mouse lung also expressed L-PGDS when and, as a result, produces not only PGD2 but also other prosta- treated with either LPS or Pseudomonas. Considering that macro- glandins including PGE2 (67, 68). Increase of PGE2 production in phages are a major responder to LPS (5), it is likely that alveolar the lung is associated with bacterial pneumonia (69). A seminal

macrophages are largely responsible for L-PGDS expression in the study shows that PGE2 inhibits the phagocytic activity of alveolar LPS-treated lung. Although it is possible that neutrophils inﬁl- macrophage (70). In addition, PGE2 suppresses production of re-

trated to the lung also contribute to L-PGDS expression, it is no- active oxygen species (67). These results suggest that PGE2 exac- table that neutrophils activated by various inﬂammatory stimuli erbates Pseudomonas infection. Given that inhibition of COX-2

including LPS do not produce a detectable level of PGD2 (50). protects mice from Pseudomonas infection (67, 71), it is possible

Currently, we are investigating whether or not other lung cells that PGE2 effects prevail over that of PGD2. However, in light of including epithelial cells express L-PGDS. our results showing the protective effect of PGD2 against Pseudo- Effective bacterial clearance by L-PGDS-overexpressing trans- monas lung infection, it is conceivable that PGD2 generated during

genic mice and PGD2-treated mice seemed to be associated with bacterial pneumonia counterbalances the otherwise detrimental ef- neutrophil recruitment. Supportive to this notion, L-PGDS KO fect of PGE2, contributing to successful clearance of bacteria. mice that were less effective in clearing bacteria than control mice The purpose of our study is to elucidate the mechanism and the showed impaired neutrophil recruitment at an early time point after role of L-PGDS expression in macrophages. Our results revealed 2574 LIPOCALIN PGD SYNTHASE EXPRESSION IN BACTERIAL IMMUNITY

the self-regulatory mechanism of L-PGDS induction by AP-1 and 21. D’Acquisto, F., T. Iuvone, L. Rombola, L. Sautebin, M. Di Rosa, and ␬ p53 in inﬂammatory environment and the protective role of L- R. Carnuccio. 1997. Involvement of NF- B in the regulation of cyclooxygenase-2 protein expression in LPS-stimulated J774 macrophages. FEBS Lett. 418: PGDS in inﬂammation and host defense against bacterial lung in- 175–178.

fection. Our results suggest that L-PGDS or PGD2 may be used for 22. Hwang, D., B. C. Jang, G. Yu, and M. Boudreau. 1997. Expression of mitogen- a treatment of infection by bacteria highly resistant to antibiotics inducible cyclooxygenase induced by lipopolysaccharide: mediation through both mitogen-activated protein kinase and NF-␬B signaling pathways in macrophages. such as P. aeruginosa. Biochem. Pharmacol. 54: 87–96. 23. Joo, M., Y. S. Hahn, M. Kwon, R. T. Sadikot, T. S. Blackwell, and Acknowledgments J. W. Christman. 2005. Hepatitis C virus core protein suppresses NF-␬B activation and cyclooxygenase-2 expression by direct interaction with I␬B kinase beta. We thank Dr. Jeffrey Wright for his initial work. We also thank Ning Ning J. Virol. 79: 7648–7657. Hu, Heng Zeng, Taylor Sherrill, and Wei Han for their technical help. 24. Dubois, R. N., S. B. Abramson, L. Crofford, R. A. Gupta, L. S. Simon, L. B. Van De Putte, and P. E. Lipsky. 1998. Cyclooxygenase in biology and Disclosures disease. FASEB J. 12: 1063–1073. 25. Kanaoka, Y., H. Ago, E. Inagaki, T. Nanayama, M. Miyano, R. Kikuno, Y. Fujii, The authors have no financial conflict of interest. N. Eguchi, H. Toh, Y. Urade, and O. Hayaishi. 1997. Cloning and crystal struc- ture of hematopoietic prostaglandin D synthase. Cell 90: 1085–1095. References 26. Urade, Y., and O. Hayaishi. 2000. Biochemical, structural, genetic, physiological, 1. Gilroy, D. W., P. R. Colville-Nash, S. McMaster, D. A. Sawatzky, and pathophysiological features of lipocalin-type prostaglandin D synthase. Bio- D. A. Willoughby, and T. Lawrence. 2003. Inducible cyclooxygenase-derived chim. Biophys. Acta 1482: 259–271. ´ 15-deoxy(A)12–14PGJ2 brings about acute inflammatory resolution in rat pleu- 27. Urade, Y., and N. Eguchi. 2002. Lipocalin-type and hematopoietic prostaglandin risy by inducing neutrophil and macrophage apoptosis. FASEB J. 17: 2269–2271. D synthases as a novel example of functional convergence. Prostaglandins Other 2. Berg, J. T., S. T. Lee, T. Thepen, C. Y. Lee, and M. F. Tsan. 1993. Depletion of Lipid Mediat. 68–69: 375–382.

alveolar macrophages by liposome-encapsulated dichloromethylene diphospho- 28. Matsuoka, T., M. Hirata, H. Tanaka, Y. Takahashi, T. Murata, K. Kabashima, Downloaded from nate. J. Appl. Physiol. 74: 2812–2819. Y. Sugimoto, T. Kobayashi, F. Ushikubi, Y. Aze, N. Eguchi, Y. Urade, 3. Broug-Holub, E., G. B. Toews, J. F. van Iwaarden, R. M. Strieter, S. L. Kunkel, N. Yoshida, K. Kimura, A. Mizoguchi, Y. Honda, H. Nagai, and S. Narumiya. R. Paine, III, and T. J. Standiford. 1997. Alveolar macrophages are required for 2000. Prostaglandin D2 as a mediator of allergic asthma. Science 287: protective pulmonary defenses in murine Klebsiella pneumonia: elimination of 2013–2017. alveolar macrophages increases neutrophil recruitment but decreases bacterial 29. Fujitani, Y., Y. Kanaoka, K. Aritake, N. Uodome, K. Okazaki-Hatake, and clearance and survival. Infect. Immun. 65: 1139–1146. Y. Urade. 2002. Pronounced eosinophilic lung inflammation and Th2 cytokine 4. Kooguchi, K., S. Hashimoto, A. Kobayashi, Y. Kitamura, I. Kudoh, J. Wiener- release in human lipocalin-type prostaglandin D synthase transgenic mice. J. Im- Kronish, and T. Sawa. 1998. Role of alveolar macrophages in initiation and munol. 168: 443–449. http://www.jimmunol.org/ regulation of inflammation in Pseudomonas aeruginosa pneumonia. Infect. 30. Spik, I., C. Brenuchon, V. Angeli, D. Staumont, S. Fleury, M. Capron, Immun. 66: 3164–3169. F. Trottein, and D. Dombrowicz. 2005. Activation of the prostaglandin D2 re- 5. Beutler, B., X. Du, and A. Poltorak. 2001. Identification of Toll-like receptor 4 ceptor DP2/CRTH2 increases allergic inflammation in mouse. J. Immunol. 174: (Tlr4) as the sole conduit for LPS signal transduction: genetic and evolutionary 3703–3708. studies. J. Endotoxin Res. 7: 277–280. 31. Ianaro, A., A. Ialenti, P. Maffia, B. Pisano, and M. Di Rosa. 2001. Role of 6. Medzhitov, R., P. Preston-Hurlburt, and C. A. Janeway, Jr. 1997. A human ho- cyclopentenone prostaglandins in rat carrageenin pleurisy. FEBS Lett. 508: mologue of the Drosophila Toll protein signals activation of adaptive immunity. 61–66. Nature 388: 394–397. 7. Muzio, M., G. Natoli, S. Saccani, M. Levrero, and A. Mantovani. 1998. The 32. Ando, M., Y. Murakami, F. Kojima, H. Endo, H. Kitasato, A. Hashimoto, human toll signaling pathway: divergence of nuclear factor ␬B and JNK/SAPK H. Kobayashi, M. Majima, M. Inoue, H. Kondo, S. Kawai, and I. Hayashi. 2003. activation upstream of tumor necrosis factor receptor-associated factor 6 Retrovirally introduced prostaglandin D2 synthase suppresses lung injury induced by bleomycin. Am. J. Respir. Cell Mol. Biol. 28: 582–591.

(TRAF6). J. Exp. Med. 187: 2097–2101. by guest on September 28, 2021 8. Rock, F. L., G. Hardiman, J. C. Timans, R. A. Kastelein, and J. F. Bazan. 1998. 33. Murakami, Y., T. Akahoshi, I. Hayashi, H. Endo, A. Hashimoto, S. Kono, A family of human receptors structurally related to Drosophila Toll. Proc. Natl. H. Kondo, S. Kawai, M. Inoue, and H. Kitasato. 2003. Inhibition of monosodium Acad. Sci. USA 95: 588–593. urate monohydrate crystal-induced acute inflammation by retrovirally transfected 9. Zhang, F. X., C. J. Kirschning, R. Mancinelli, X. P. Xu, Y. Jin, E. Faure, prostaglandin D synthase. Arthritis Rheum. 48: 2931–2941. A. Mantovani, M. Rothe, M. Muzio, and M. Arditi. 1999. Bacterial lipopolysac- 34. Cipollone, F., M. Fazia, A. Iezzi, G. Ciabattoni, B. Pini, C. Cuccurullo, charide activates nuclear factor-␬B through interleukin-1 signaling mediators in S. Ucchino, F. Spigonardo, L. M. De, C. Prontera, et al. 2004. Balance between cultured human dermal endothelial cells and mononuclear phagocytes. J. Biol. PGD synthase and PGE synthase is a major determinant of atherosclerotic plaque Chem. 274: 7611–7614. instability in humans. Arterioscler. Thromb. Vasc. Biol. 24: 1259–1265. 10. Sultzer, B. M. 1968. Genetic control of leucocyte responses to endotoxin. Nature 35. Sadikot, R. T., T. S. Blackwell, J. W. Christman, and A. S. Prince. 2005. Patho- 219: 1253–1254. gen-host interactions in Pseudomonas aeruginosa pneumonia. Am. J. Respir. 11. Coutinho, A., L. Forni, F. Melchers, and T. Watanabe. 1977. Genetic defect in Crit. Care Med. 171: 1209–1223. responsiveness to the B cell mitogen lipopolysaccharide. Eur. J. Immunol. 7: 36. Sadikot, R. T., H. Zeng, F. E. Yull, B. Li, D. S. Cheng, D. S. Kernodle, 325–328. E. D. Jansen, C. H. Contag, B. H. Segal, S. M. Holland, T. S. Blackwell, and 12. da Silva, C. J., K. Soldau, U. Christen, P. S. Tobias, and R. J. Ulevitch. 2001. J. W. Christman. 2004. p47phox deficiency impairs NF-␬B activation and host Lipopolysaccharide is in close proximity to each of the proteins in its membrane defense in Pseudomonas pneumonia. J. Immunol. 172: 1801–1808. receptor complex. transfer from CD14 to TLR4 and MD-2. J. Biol. Chem. 276: 37. Pinzar, E., Y. Kanaoka, T. Inui, N. Eguchi, Y. Urade, and O. Hayaishi. 2000. 21129–21135. Prostaglandin D synthase gene is involved in the regulation of non-rapid eye 13. Lien, E., T. K. Means, H. Heine, A. Yoshimura, S. Kusumoto, K. Fukase, movement sleep. Proc. Natl. Acad. Sci. USA 97: 4903–4907. M. J. Fenton, M. Oikawa, N. Qureshi, B. Monks, et al. 2000. Toll-like receptor 38. Eguchi, N., T. Minami, N. Shirafuji, Y. Kanaoka, T. Tanaka, A. Nagata, 4 imparts ligand-specific recognition of bacterial lipopolysaccharide. J. Clin. In- N. Yoshida, Y. Urade, S. Ito, and O. Hayaishi. 1999. Lack of tactile pain (allo- vest. 105: 497–504. dynia) in lipocalin-type prostaglandin D synthase-deficient mice. Proc. Natl. 14. Miller, S. I., R. K. Ernst, and M. W. Bader. 2005. LPS, TLR4 and infectious Acad. Sci. USA 96: 726–730. disease diversity. Nat. Rev. Microbiol. 3: 36–46. 39. Mochizuki, M., Y. Ishii, K. Itoh, T. Iizuka, Y. Morishima, T. Kimura, 15. Poltorak, A., X. He, I. Smirnova, M. Y. Liu, C. Van Huffel, X. Du, D. Birdwell, T. Kiwamoto, Y. Matsuno, A. E. Hegab, A. Nomura, et al. 2005. Role of 15- E. Alejos, M. Silva, C. Galanos, et al. 1998. Defective LPS signaling in C3H/HeJ deoxy-A´ (12,14)prostaglandin J and Nrf2 pathways in protection against acute and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282: 2085–2088. 2 lung injury. Am. J. Respir. Crit. Care Med. 171: 1260–1266. 16. Poltorak, A., P. Ricciardi-Castagnoli, S. Citterio, and B. Beutler. 2000. Physical contact between lipopolysaccharide and Toll-like receptor 4 revealed by genetic 40. Everhart, M. B., W. Han, T. P. Sherrill, M. Arutiunov, V. V. Polosukhin, Proc. Natl. Acad. Sci. USA J. R. Burke, R. T. Sadikot, J. W. Christman, F. E. Yull, and T. S. Blackwell. 2006. complementation. 97: 2163–2167. ␬ 17. Guha, M., and N. Mackman. 2001. LPS induction of gene expression in human Duration and intensity of NF- B activity determine the severity of endotoxin- monocytes. Cell Signal. 13: 85–94. induced acute lung injury. J. Immunol. 176: 4995–5005. 18. Sharma, S., B. R. tenOever, N. Grandvaux, G. P. Zhou, R. Lin, and J. Hiscott. 41. Kingsley, P. J., C. A. Rouzer, S. Saleh, and L. J. Marnett. 2005. Simultaneous 2003. Triggering the interferon antiviral response through an IKK-related path- analysis of prostaglandin glyceryl esters and prostaglandins by electrospray tan- way. Science 300: 1148–1151. dem mass spectrometry. Anal. Biochem. 343: 203–211. 19. Fitzgerald, K. A., S. M. McWhirter, K. L. Faia, D. C. Rowe, E. Latz, 42. Joo, M., G. Y. Park, J. G. Wright, T. S. Blackwell, M. L. Atchison, and D. T. Golenbock, A. J. Coyle, S. M. Liao, and T. Maniatis. 2003. IKK⑀ and TBK1 J. W. Christman. 2004. Transcriptional regulation of the cyclooxygenase-2 gene are essential components of the IRF3 signaling pathway. Nat. Immunol. 4: in macrophages by PU. 1. J. Biol. Chem. 279: 6658–6665. 491–496. 43. Urade, Y., M. Ujihara, Y. Horiguchi, K. Ikai, and O. Hayaishi. 1989. The major 20. Bjorkbacka, H., K. A. Fitzgerald, F. Huet, X. Li, J. A. Gregory, M. A. Lee, source of endogenous prostaglandin D2 production is likely antigen-presenting C. M. Ordija, N. E. Dowley, D. T. Golenbock, and M. W. Freeman. 2004. The cells: localization of glutathione-requiring prostaglandin D synthetase in histio- induction of macrophage gene expression by LPS predominantly utilizes Myd88- cytes, dendritic, and Kupffer cells in various rat tissues. J. Immunol. 143: independent signaling cascades. Physiol. Genom. 19: 319–330. 2982–2989. The Journal of Immunology 2575

44. Fitzgerald, K. A., D. C. Rowe, and D. T. Golenbock. 2004. Endotoxin recognition 59. Forman, B. M., P. Tontonoz, J. Chen, R. P. Brun, B. M. Spiegelman, and ´ and signal transduction by the TLR4/MD2-complex. Microbes. Infect. 6: R. M. Evans. 1995. 15-Deoxy-A12,14-prostaglandin J2 is a ligand for the adi- 1361–1367. pocyte determination factor PPAR␥. Cell 83: 803–812. 45. Takeda, K., and S. Akira. 2004. TLR signaling pathways. Semin. Immunol. 60. Kliewer, S. A., J. M. Lenhard, T. M. Willson, I. Patel, D. C. Morris, and 16: 3–9. J. M. Lehmann. 1995. A prostaglandin J2 metabolite binds peroxisome prolif- 46. Baud, V., and M. Karin. 2001. Signal transduction by tumor necrosis factor and erator-activated receptor gamma and promotes adipocyte differentiation. Cell 83: its relatives. Trends Cell Biol. 11: 372–377. 813–819. 47. Xaus, J., M. Comalada, A. F. Valledor, J. Lloberas, F. Lopez-Soriano, 61. Hayaishi, O., and Y. Urade. 2002. Prostaglandin D2 in sleep-wake regulation: J. M. Argiles, C. Bogdan, and A. Celada. 2000. LPS induces apoptosis in mac- recent progress and perspectives. Neuroscientist 8: 12–15. rophages mostly through the autocrine production of TNF-␣. Blood 95: 62. Hirai, H., K. Tanaka, O. Yoshie, K. Ogawa, K. Kenmotsu, Y. Takamori, 3823–3831. M. Ichimasa, K. Sugamura, M. Nakamura, S. Takano, and K. Nagata. 2001. 48. Hassan, F., S. Islam, M. M. Mu, H. Ito, N. Koide, I. Mori, T. Yoshida, and Prostaglandin D2 selectively induces chemotaxis in T helper type 2 cells, eosin- T. Yokochi. 2005. Lipopolysaccharide prevents doxorubicin-induced apoptosis in ophils, and basophils via seven-transmembrane receptor CRTH2. J. Exp. Med. RAW 264.7 macrophage cells by inhibiting p53 activation. Mol. Cancer Res. 3: 193: 255–261. 373–379. 63. Hata, A. N., and R. M. Breyer. 2004. Pharmacology and signaling of prostaglan- 49. Libby, P., P. M. Ridker, and A. Maseri. 2002. Inflammation and atherosclerosis. din receptors: multiple roles in inflammation and immune modulation. Pharma- Circulation 105: 1135–1143. col. Ther. 103: 147–166. 50. Pouliot, M., C. Gilbert, P. Borgeat, P. E. Poubelle, S. Bourgoin, C. Creminon, 64. Kostenis, E., and T. Ulven. 2006. Emerging roles of DP and CRTH2 in allergic J. Maclouf, S. R. McColl, and P. H. Naccache. 1998. Expression and activity of inflammation. Trends Mol. Med. 12: 148–158. prostaglandin endoperoxide synthase-2 in agonist-activated human neutrophils. 65. Prince, A. 2006. Flagellar activation of epithelial signaling. Am. J. Respir. Cell FASEB J. 12: 1109–1123. Mol. Biol. 34: 548–551. 51. Williams, T. J., and J. Morley. 1973. Prostaglandins as potentiators of increased vascular permeability in inflammation. Nature 246: 215–217. 66. Narumiya, S., and G. A. FitzGerald. 2001. Genetic and pharmacological analysis 52. Gualde, N., and H. Harizi. 2004. Prostanoids and their receptors that modulate of prostanoid receptor function. J. Clin. Invest. 108: 25–30. dendritic cell-mediated immunity. Immunol. Cell Biol. 82: 353–360. 67. Sadikot, R. T., H. Zeng, A. C. Azim, M. Joo, S. K. Dey, R. M. Breyer, 53. Harris, S. G., J. Padilla, L. Koumas, D. Ray, and R. P. Phipps. 2002. Prostaglan- R. S. Peebles, T. S. Blackwell, and J. W. Christman. 2007. Bacterial clearance of Downloaded from dins as modulators of immunity. Trends Immunol. 23: 144–150. Pseudomonas aeruginosa is enhanced by the inhibition of COX-2. Eur. J. Im- 54. Hata, A. N., and R. M. Breyer. 2004. Pharmacology and signaling of prostaglan- munol. 37: 1001–1009. din receptors: multiple roles in inflammation and immune modulation. Pharma- 68. Sorrell, T. C., C. P. Rochester, F. N. Breen, and M. Muller. 1989. Eicosanoids col. Ther. 103: 147–166. produced during interactions between Pseudomonas aeruginosa and alveolar 55. Fitzpatrick, F. A., and M. A. Wynalda. 1983. Albumin-catalyzed metabolism of macrophages are species-dependent. Immunol. Cell Biol. 67(Pt 3): 169–176. 69. Asai, T., M. Okada, Y. Yokomizo, S. Sato, and Y. Mori. 1996. Suppressive effect prostaglandin D2: identification of products formed in vitro. J. Biol. Chem. 258: 11713–11718. of bronchoalveolar lavage fluid from pigs infected with Mycoplasma hyopneu- moniae on chemiluminescence of porcine peripheral neutrophils. Vet. Immunol. 56. FitzGerald, G. A. 2003. COX-2 and beyond: approaches to prostaglandin inhi- http://www.jimmunol.org/ bition in human disease. Nat. Rev. Drug Discov. 2: 879–890. Immunopathol. 51: 325–331. 57. Straus, D. S., G. Pascual, M. Li, J. S. Welch, M. Ricote, C. H. Hsiang, 70. Aronoff, D. M., C. Canetti, and M. Peters-Golden. 2004. Prostaglandin E2 inhibits L. L. Sengchanthalangsy, G. Ghosh, and C. K. Glass. 2000. 15-Deoxy-A´ 12,14- alveolar macrophage phagocytosis through an E-prostanoid 2 receptor-mediated ␬ prostaglandin J2 inhibits multiple steps in the NF- B signaling pathway. Proc. increase in intracellular cyclic AMP. J. Immunol. 173: 559–565. Natl. Acad. Sci. USA 97: 4844–4849. 71. Campanile, F., A. Giampietri, U. Grohmann, L. Binaglia, M. C. Fioretti, and 58. Rossi, A., P. Kapahi, G. Natoli, T. Takahashi, Y. Chen, M. Karin, and P. Puccetti. 1993. Accelerated hematopoietic recovery and protective effect of the M. G. Santoro. 2000. Anti-inflammatory cyclopentenone prostaglandins are di- cyclooxygenase inhibitor indomethacin in bacterial infection of neutropenic rect inhibitors of I␬B kinase. Nature 403: 103–108. mice. Cell. Immunol. 147: 341–352. by guest on September 28, 2021