A Potential Respiratory Syncytial Virus (RSV) Infection

Respiratory Syncytial Virus (RSV) Infection Induces Cyclooxygenase 2: A Potential Target for RSV Therapy

This information is current as Joann Y. Richardson, Martin G. Ottolini, Lioubov Pletneva, of September 27, 2021. Marina Boukhvalova, Shuling Zhang, Stefanie N. Vogel, Gregory A. Prince and Jorge C. G. Blanco J Immunol 2005; 174:4356-4364; ; doi: 10.4049/jimmunol.174.7.4356

http://www.jimmunol.org/content/174/7/4356 Downloaded from

References This article cites 70 articles, 27 of which you can access for free at: http://www.jimmunol.org/content/174/7/4356.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 27, 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 © 2005 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Respiratory Syncytial Virus (RSV) Infection Induces Cyclooxygenase 2: A Potential Target for RSV Therapy1

Joann Y. Richardson,* Martin G. Ottolini,* Lioubov Pletneva,§ Marina Boukhvalova,§ Shuling Zhang,† Stefanie N. Vogel,‡ Gregory A. Prince,§ and Jorge C. G. Blanco2§

Cyclooxygenases (COXs) are rate-limiting enzymes that initiate the conversion of arachidonic acid to prostanoids. COX-2 is the inducible isoform that is up-regulated by proinflammatory agents, initiating many prostanoid-mediated pathological aspects of inflammation. The roles of cyclooxygenases and their products, PGs, have not been evaluated during respiratory syncytial virus (RSV) infection. In this study we demonstrate that COX-2 is induced by RSV infection of human lung alveolar epithelial cells with the concomitant production of PGs. COX-2 induction was dependent on the dose of virus and the time postinfection. PG production was inhibited preferentially by NS-398, a COX-2-specific inhibitor, and indomethacin, a pan-COX inhibitor, but not by

SC-560, a COX-1-specific inhibitor. In vivo, COX-2 mRNA expression and protein production were strongly induced in the lungs Downloaded from and cells derived from bronchioalveolar lavage of cotton rats infected with RSV. The pattern of COX-2 expression in vivo in lungs is cyclical, with a final peak on day 5 that correlates with maximal histopathology. Treatment of cotton rats with indomethacin significantly mitigated lung histopathology produced by RSV. The studies described in this study provide the first evidence that COX-2 is a potential therapeutic target in RSV-induced disease. The Journal of Immunology, 2005, 174: 4356–4364.

3 espiratory syncytial virus (RSV) is the leading viral 10). Many inflammatory mediators induced in the lung by RSV are http://www.jimmunol.org/ cause of death in children under 1 year of age and is an effective inducers of the cyclooxygenase 2 (COX-2) in lung epithelial R increasing cause of morbidity and mortality in transplant and inflammatory cells (11–15). The expressions of COX-2 and its patients and the elderly (1–3). RSV causes upper and lower respi- products, PGs and thromboxanes (TBx), have been correlated with ratory tract infections, occasionally leading to severe bronchiolitis the development of many inflammatory processes (16, 17), some of and pneumonia. In addition, RSV bronchiolitis has been associated which occur during viral infection (e.g., regulatory effects on the with the development of recurrent episodes of bronchiolar obstruc- immune response, vascular tone, platelets aggregation, airway remod- tion, specific IgE production, and establishment of asthma (4–6). eling, allergic processes, etc.). Moreover, RSV induces the production

It is unclear why children, the elderly, and the immunosuppressed of PGE2 in cultures of human monocytes and dendritic cells (18). The are at much higher risk for severe disease; however, an RSV- potent physiologic effects of PGs and TBx and their possible role in by guest on September 27, 2021 induced immune pathological mechanism has long been suspected. RSV infection suggest that these mediators should be considered as Yet there is no safe and effective vaccine against RSV. Passive potential important factors in the RSV disease process. The animal anti-RSV Ab, although effective in prophylactic settings, does not model of choice for RSV studies is the cotton rat (Sigmodon hispidus), provide any clinically beneficial outcome when applied therapeu- because it most closely resembles disease in humans (19). Preclinical tically, indicating that RSV-induced pathology is primarily the re- data from cotton rats have resulted in a successful prophylactic sult of the inflammatory response to infection, rather than a direct approach in humans (20–24). viral effect. Therefore, a combined antiviral and anti-inflammatory In this work we present data demonstrating the induction of therapy might represent the most safe and efficient treatment COX-2 expression during RSV infection in vitro, in human lung against RSV infection. alveolar epithelial cells and cotton rat macrophages, and in vivo, in During RSV infection, proinflammatory cytokines and chemo- the lungs of cotton rats infected with RSV. We present additional kines are detectable in isolates from bronchioalveolar lavage evidence that indicates the potential of nonsteroidal anti- (BAL) fluids of patients with lower respiratory tract infection (7– inflammatory drugs in the treatment of RSV-induced bronchiolitis. Materials and Methods Departments of *Pediatrics and †Microbiology and Immunology, Uniformed Services Animals ‡ University of the Health Sciences, Bethesda, MD 20814; Department of Microbiol- Inbred cotton rats (S. hispidus) were obtained from a colony maintained at ogy and Immunology, University of Maryland, Baltimore, MD 21201; and §Virion Systems, Rockville, MD 20850 Virion Systems. Cotton rats were housed in large polycarbonate rat cages with a bedding of pine shavings (Harlan Teklad) and were fed a diet of Received for publication October 9, 2003. Accepted for publication January 20, 2005. rodent chow and water. The cotton rat colony was monitored for Abs to The costs of publication of this article were defrayed in part by the payment of page paramyxoviruses, RSV, and rodent viruses; no such Abs were found. The charges. This article must therefore be hereby marked advertisement in accordance animals used for infection experiments were, on the average, 8–12 wk old with 18 U.S.C. Section 1734 solely to indicate this fact. and weighed 100 g at the time they were used. All animal experimentation 1 This work was supported by National Institutes of Health Grants RR13161-03 (to procedures were performed following National Institutes of Health and G.A.P.) and RO1AI057575-01 (to J.C.G.B. and S.N.V.). U.S. Department of Agriculture guidelines with institutional animal care 2 Address correspondence and reprint requests to Dr. Jorge C. G. Blanco, Virion and use committee approval. Systems, 9610 Medical Center Drive, Suite 100, Rockville, MD 20850. E-mail ad- Virus and tissue culture cells dress: [email protected] 3 Abbreviations used in this paper: RSV, respiratory syncytial virus; BAL, bronchoal- The Long strain (group A) of RSV was obtained from American Type veolar lavage; COX-2, cyclooxygenase 2; LT, leukotriene; m.o.i., multiplicity of in- Culture Collection. Virus stocks were prepared in HEp-2 cells and con- fection; rh, recombinant human; TBx, thromboxane. tained 1 ϫ 107.5 PFU/ml. Viral titers in stocks and lung homogenates were

Copyright © 2005 by The American Association of Immunologists, Inc. 0022-1767/05/$02.00 The Journal of Immunology 4357 determined by plaque assay on HEp-2 cells (19). A549 cells were obtained the RSV F protein gene was amplified using the following primers: forward from American Type Culture Collection and were grown in DMEM sup- primer, AATCCTCAAAGCAAATGCAATTACC; and reverse primer, AT- plemented with 2 mM L-glutamine, 100 IU/ml penicillin, 100 ␮g/ml strep- GGCTCCTAGAGATGTGATAACGGAGC, using 32 (macrophages) or 25 tomycin, and 10% FBS in a humidified 37°C incubator with 5% CO2. For (A549 cells) amplification cycles and 55°C for the annealing temperature. The RSV infection, cells were seeded in a six-well plate at a density of 4 ϫ 105 product of the reaction (a 1.2-kb band) was visualized by direct ethidium cells/well 1 day before infection. The cells were generally 70–80% con- bromide staining of agarose gel after electrophoresis. fluent on the day after seeding. Purified stocks of RSV were diluted in DMEM to yield multiplicities of infection (m.o.i.) of 1, 0.5, and 0.1. The COX-2 Western blot and immunoprecipitation medium was removed from the cells and replaced with medium containing virus dilutions for 1 h, then replaced with complete DMEM. Uninfected A549 cells were homogenized in lysis buffer containing 20 mM Tris-HCl HEp-2 cell culture supernatant served as the control (mock inoculum). (pH 8), 100 mM NaCl, 1% Nonidet P-40, 4 mM DTT, 0.5 mM PMSF, 7.5 Cells were harvested at various times postinfection, and the supernatants mM sodium fluoride, 2 mM EDTA, and a mixture of protease inhibitors were stored at Ϫ70°C for quantification of PGE by ELISA (Cayman (Complete; Roche). Cell lysates were normalized by protein concentration, 2 ␮ Chemicals). For inhibition experiments of COX enzymes in A549 cells, and 40 g/lane was subjected to electrophoresis in 10% SDS-PAGE gels. infected cells were incubated with different concentrations of indomethacin Western blot analysis was used to detect COX-2 protein using a polyclonal (in saline), Indocin, a pan-COX inhibitor (Merck); NS-398 (in DMSO), a rabbit Ab (catalogue no. 160116; Cayman Chemical) directed against the COX-2-specific inhibitor (Cayman Chemicals); SC-560 (in DMSO), a epitope located between aa 584 and 598 (SHSRLDDINPTVLIK) of the COX-1-specific inhibitor (Cayman Chemicals); or the vehicles as controls. mouse COX-2 protein, which is 100% conserved in the cotton rat COX-2, For analysis of the supernatants of infected A549 cells for the presence of and detected via secondary HRP-conjugated anti-rabbit Abs (made in the soluble factors that cause COX-2 activation, the supernatants were filtered donkey; Amersham Biosciences; catalogue no. NA934) and the ECL de- through Centricon columns (Centricon Centrifugal Filter Devices, YM- tection system (Amersham Biosciences). 100, 100,000 kDa cutoff; catalogue no. 4212; Amicon Bioseparations) and For detection of COX-2 protein in the lungs of RSV-infected cotton rats, incubated with 50 ␮g/ml of the following Abs: anti-human IL-1␣ (R&D the left lung of each rat was homogenized in 2 ml of lysis buffer using a Downloaded from Systems; catalogue no. MAB200), anti-human IL-1␤ (R&D Systems; cat- microhomogenizer (Omni International). The homogenates were centri- ϫ alogue no. MAB201), anti-human TNF-␣ (R&D Systems; catalogue no. fuged at 4,000 g for 15 min at 4°C to remove cell debris. Each super- MAB210), or control mouse IgG before adding them to the cells. Recom- natant was sonicated (pulses of 10 s with microtip at 4°C) and finally ϫ binant human (rh) IL-1␣, rhIL-1␤, rhTNF-␣ (1 ng/ml; BioSource Interna- centrifuged at 16,000 g at 4°C for 10 min. Four hundred microliters of tional; catalogue no. PHC0014, PHC0814, and PHC3015, respectively) each lung homogenate was incubated with 800 ng of anti-COX-2 Ab (sc- were used as positive controls in the indicated concentrations. 1746; Santa Cruz Biotechnology) and mixed by rotation (30 rpm) for 3 h Peritoneal macrophages were prepared from cotton rats by injecting 6–8 at 4°C. Immunocomplexes were precipitated with ImmunePure-immobi- http://www.jimmunol.org/ ml of thioglycolate broth i.p. into each 8- to 16-wk-old S. hispidus. Four lized protein G-Sepharose (50% slurry in lysis buffer; Pierce) and washed days postinjection, rats were killed by CO inhalation, and the peritoneal three times with lysis buffer, and the pellet was electrophoresed and ana- 2 ␤ cavity was washed twice with 30 ml of cold saline solution. Peritoneal lyzed by Western blot for COX-2. -Actin Abs (sc-8432; Santa Cruz, exudates (ϳ2 ϫ 107 cell/animal) were washed with PBS, counted, and Biotechnology) were used to normalize the amount of protein extract used resuspended to 1 ϫ 106 cells/ml in RPMI 1640 medium containing 3% for immunoprecipitation.

FCS. Cells were plated and incubated overnight at 37°C and 5% CO2. The next day, cell monolayers were washed to remove nonadherent cells, and Animal protocol adherent cells were treated with protein-free Escherichia coli K235 LPS Cotton rats were inoculated intranasally under isoflurane (Abbott Labora- (25) or infected with RSV as indicated. Cells were lysed, and protein ly- tories) anesthesia with 100 ␮l of RSV/Long A suspension containing 1 ϫ sates were subjected to Western blot analysis using anti-COX-2 Ab as 106.5 PFU. Controls included uninfected cotton rats and cotton rats inoc- previously described (26). ulated intranasally with mock inoculum. Animals were killed, a midline by guest on September 27, 2021 Cloning of cotton rat cox-2 gene vertical chest incision was made to expose the thoracic contents, and the lungs were removed en bloc. The right lung was ligated and submitted for A fragment encoding part of the cotton rat COX-2-coding sequence (1540 viral titration (19). BAL was performed by inserting a blunted 20-gauge bp long) was amplified using cDNA obtained from LPS-activated cotton needle into the trachea, which was secured in place by ligature. Three rat macrophages and the following primers obtained from comparison of milliliters of cold PBS was introduced, and the pulmonary tree was la- human, mouse, and rat sequences: forward, 5Ј-CAGCAAATCCTTGCT vaged. The recovered sample was centrifuged at 800 ϫ g, and the cell GTTCCMAYCCATGYCARAAYCGWGGKGWATGTATGAGYRYRG pellet was resuspended in lysis buffer after counting and volume normal- GATTYGACCAR-3Ј; and reverse, 5Ј-GTATTGAGGAGAACAGATGG ization. Cell lysates were stored at Ϫ70°C until analyzed. Intraperitoneal GATTACCC-3Ј. Amplification was performed using standard PCR injection of indomethacin (Indocin; Merck; 0.3 mg/kg in saline solution) reagents and cycle conditions. PCR-amplified samples were subjected to was administered 12 h before inoculation with RSV/Long and subse- agarose gel electrophoresis, and a single DNA band of the predicted m.w. quently at 12 and 36 h postinfection. Control groups consisted of unin- was isolated from the gel by QIAEX II gel extraction kit (Qiagen), quan- fected animals, mock-infected animals, and RSV-infected, sham-treated tified by UV spectrophotometry, sequenced, and labeled using the Dig animals (i.p injection of saline). Animals were killed on day 4 postinfec- DNA Labeling and Detection Kit (Roche). A cotton rat cDNA library tion, and lungs were removed en bloc for viral titration and histopathology. (derived from LPS-activated peritoneal macrophages) was constructed using the plasmid vector pSPORT 1 (Invitrogen Life Technologies) and Inhibition of COX-2 activity in an ex vivo-stimulated whole transformed into E. coli MAX Efficiency DH5␣ cells (Invitrogen Life blood assay Technologies). A total of 5000 colonies were screened, positive clones were isolated, and a full-length clone encoding the cotton rat COX-2 was Blood (ϳ1 ml) was collected by cardiac puncture in heparinized tubes for entirely sequenced. TxB2 analysis from animals treated i.p. with indomethacin (0.3 mg/kg) or saline three times at 24-h intervals. To determine the extent to which in-

RT-PCR analysis domethacin inhibited COX activity in vivo, TxB2 (a subproduct of TxA2) was measured from whole blood platelets after stimulation with a calcium Semiquantitative analysis of COX-2 mRNA expression in lungs of RSV- ionophore (A23187) as previously described (28). infected cotton rats by RT-PCR and Southern blot analysis was performed similarly to that previously described for other cotton rat genes (27). The Histopathology following primers and conditions were used: forward primer, 5Ј-CATT GACCAGAGCAGGCAGATGAAATAC-3Ј; reverse primer, 5Ј-GTAGT Formalin-fixed lung tissues were embedded in paraffin, and 4-␮m sections ACTGTGGGATTGATATCATCTAGTC-3Ј; and probe, 5Ј-AAGCCCTAT were obtained and stained with H&E (Histoserv). One H&E-stained slide GAATCATTTGAAGAACTTAC-3Ј, using 22 amplification cycles and was prepared per animal killed. Three histopathologic parameters were 65°C for the annealing temperature. For analysis of COX-2 expression in examined on each H&E-stained slide: peribronchiolitis, alveolitis, and in- human A549 cells, RNA isolation was performed as previously described (26), terstitial pneumonitis. Peribronchiolitis is defined as inflammatory cells, but using the following primers and conditions: forward primer, 5Ј- mostly lymphocytes and macrophages, that accumulate around the periph- CATTGACCAGAGCAGGCAGATGAAATAC-3Ј; reverse, 5Ј-GTAG ery of small airways. Alveolitis is defined as inflammatory cells, mostly TACTGTGGGATTGATATCATCTAGTC-3Ј; and probe, 5Ј-AAGC macrophages and neutrophils, within the air space. Interstitial pneumonitis CCTATGAATCATTTGAAGAACTTAC-3Ј, using 20 amplification cycles is defined as thickening of the alveolar wall, associated with an influx of and 65°C for the annealing temperature. For analysis of viral gene expression, inflammatory cells of various types. Each parameter was scored separately 4358 RSV AND COX-2 under blinded conditions and was assigned a score based on a scale of 0 (no inflammation) to 4 (maximum inflammation) as previously described (29). Statistics Arithmetic means were calculated for each of the three histopathologic parameters obtained for each group of cotton rats. The RSV-infected/indomethacin treated group was compared with the RSV-infected/untreated group using the two-tailed Student’s t test. Results

Induction of COX-2 expression and PGE2 production by RSV in vitro The human type II pneumocyte cell line, A549, has been used extensively in RSV studies in vitro. A549 cells are highly permis- sive to RSV infection and produce large syncytia and high levels of cytokines and chemokines, such as IL-1 (30), TNF-␣ (30), IL-6 (31), and IL-8 (31, 32). To investigate whether COX-2 was also inducible by RSV in these cells, A549 cell monolayers were infected with RSV at various m.o.i. and harvested at different times postinfection. Downloaded from Infected cell lysates were evaluated for the presence of COX-2 and viral mRNA by RT-PCR (Fig. 1A) and for protein by Western analysis (Fig. 1B). Induction of steady-state levels of COX-2 mRNA was detected in cells infected with higher m.o.i. (1 and 0.5) by 40 h postinfection. When an m.o.i. of 0.1 was used, a signiﬁcant

increase in the level of COX-2 mRNA expression was seen only http://www.jimmunol.org/ after 44 h. At 48 h, the steady state levels of COX-2 mRNA peaked at all m.o.i. of virus (Fig. 1A, graph), and the expression of ␤-actin, a housekeeping gene, was slightly reduced, consistent with the observed cytopathic effect of the virus on the cell monolayer. The steady state levels of expression of the gene that encodes the RSV F protein were constant throughout the period monitored in infected cells. COX-2 protein expression was undetectable during the ﬁrst day of infection, but peaked during the second day postin- by guest on September 27, 2021 fection, with the highest m.o.i. giving the strongest COX-2 signal FIGURE 1. COX-2 mRNA and protein expression in the human lung (Fig. 1B), consistent with the mRNA expression data. type II alveolar epithelial cell line A549 infected with different m.o.i. of Induction of COX-2 protein was also dependent on viral repli- RSV. A, Southern blot showing COX-2 and ␤-actin mRNA expression cation in these cells, because treatment of virus with UV light, or assessed by RT-PCR. A549 cells were treated with a mock preparation of palivizumab, an anti-RSV F protein mAb (Synagis; MedImmune) virus (Mock) or with RSV for 24–48 h. No changes in the levels of COX-2 did not lead to COX-2 mRNA (Fig. 2A) and protein expression mRNA expression were detected before 24 h. The expression of RSV F (Fig. 2B). At the time, the expression of RSV F protein mRNA was protein was measured by RT-PCR, followed by agarose gel electrophore- undetectable, and no replicative virus was isolated. Together, these sis. Ethidium bromide-stained gels are shown. The accompanying graph represents the expression of COX-2 mRNA relative to ␤-actin mRNA. B, data demonstrate that COX-2 is induced during RSV infection in Western blot showing COX-2 expression in cell lysates after exposure to human lung epithelial cells and requires infection with replicative different RSV inocula. Cϩ, cell lysates from a positive control, LPS- viral particles. treated cotton rat macrophages. Western blots for ␤-actin were included as Because mRNA expression of COX-2 in these cells was de- a loading control. The data shown are representative of four independent tected at ϳ40 h postinfection, its induction is more likely to be experiments. secondary to the action of a secreted factor(s) induced by viral infection in these cells. To test this hypothesis, we incubated A549 COX-2 mRNA expression (compare lanes 4 and 7 with lane 10). cells for only 6 h with several dilutions of ﬁltered (virus-free) Taken collectively, these data demonstrate that COX-2 mRNA and supernatants obtained from cultures of A549 cells that were in- protein expression is strongly induced in human lung epithelial fected for 48 h with different m.o.i. of RSV (Fig. 3A). In contrast cells by RSV. The induction is dependent on viral replication and to undiluted control supernatants, these supernatants from virus- requires the action of soluble factors, including IL-1␣, IL-1␤, and infected cells strongly activated the expression of COX-2 in a TNF-␣, that act on the cells to stimulate COX-2 expression. dose-dependent manner. IL-1 (33, 34) and TNF (14) are known to To determine whether COX-2 expression was associated with be potent inducers of COX-2. They are secreted in A549 cells after PG production in A549 cells, supernatants of cells infected with ␣ ␤ RSV infection (30). We used blocking Abs against IL-1 , IL-1 , RSV were assayed for the presence of PGE2. As depicted in Fig. ␣ and TNF- to determine whether these cytokines contribute to 4A, PGE2 was not detected on day 1, but was detected 2 days COX-2 mRNA expression when released into the supernatants by postinfection with levels that correlated well with the expressed

RSV-infected cells. Incubation of culture supernatants of RSV- COX-2 mRNA and protein seen by Western blot (Fig. 1). PGE2 infected A549 cells with Abs against these cytokines produced a levels reached maximum levels on day 3 for all m.o.i., when the marked reduction of COX-2 mRNA expression (Fig. 3B). Block- cell monolayer was almost completely destroyed by the infection. ing of IL-1␣ had the strongest inhibitory effect on the activation of Treatment of A549 cells with the selective COX-2 inhibitor NS- COX-2 mRNA expression by supernatants (compare lane 1 with 398 (Ͼ50-fold selective for COX-2 over COX-1) (35) or indo- lane 10), whereas blocking TNF-␣ and IL-1␤ partially inhibited methacin (a pan-COX inhibitor) resulted in stronger inhibition of The Journal of Immunology 4359

FIGURE 2. Inactivation of RSV by UV or pretreatment with palivizumab (an anti-RSV F protein Ab) completely abolishes the expression of COX-2 mRNA and protein. A, Analysis of COX-2 mRNA expression. For the UV-inactivated RSV study, cells were harvested at 24–48 h. Palivi- zumab-treated RSV (400 ␮g/ml palivizumab, 1 h incubation at room temperature) or control, untreated, RSV was incubated with the cells for 48 h. RSV F protein mRNA expression is shown as a control for viral replication. B, Western blot analysis of the expression of COX-2 protein. The Downloaded from virus preparations used were RSV alone (RSV), RSV inactivated by UV (UV-RSV), RSV incubated for 1 h with palivizumab (400 ␮g/ml), or RSV treated with a control unrelated mAb (C mAb RSV). RSV was used at an equivalent of 0.5 m.o.i. in all cases. Data shown are representative of four independent experiments. http://www.jimmunol.org/

FIGURE 3. COX-2 expression is induced by soluble factors secreted by PGE2 synthesis than SC-560, a highly selective COX-1 inhibitor A549 cells in response to RSV infection. A, Cell culture supernatants of (28) (Fig. 4B), confirming that the PGE2 produced was the result of the enzymatic activity of COX-2, not COX-1. None of the in- A549 cells infected with RSV for 48 h at m.o.i. of 1 (Sup 1), 0.5 (Sup 2), hibitors significantly affected RSV replication even when used at and 0.1 (Sup 3) were filtered through a Centricon column (100,000 m.w. cutoff) to remove infective particles and transferred in different concentra- concentrations as high as 100 nM (data not shown). Moreover, tions (undiluted and diluted 1/5 and 1/10 with complete medium) to new PGE2 production was dependent on virus replication because it cultures of A549 cells for 6 h (see flow chart). Southern blots of COX-2 was not detected when the virus used for infection was UV-inac- and ␤-actin mRNAs are shown. Ϫ, Filtered supernatant of uninfected A549 tivated or preincubated with palivizumab (data not shown). cells. B, Secreted IL-1␣, IL-1␤, and TNF-␣ are responsible for COX-2 by guest on September 27, 2021 induction in A549 cells infected with RSV. Supernatants of RSV-infected Cloning and characterization of cotton rat COX-2 A549 cells (48 h; m.o.i., 1) were preincubated with mAb (50 ␮g/ml) The cotton rat is a powerful model for studying RSV infection in against IL-1␣, TNF-␣, or IL-1␤. The ability of specific mAb to reduce vivo. To analyze the pattern of gene expression of several pro- and COX-2 expression was compared with that of an IgG1 isotype control Ab ␣ ␣ ␤ anti-inflammatory mediators during RSV infection, we developed (C). rIL-1 , rTNF- , and rIL1 (1 ng/ml) were used in the indicated lanes with or without the respective blocking Abs as controls for induction and cotton rat reagents (27) that have been used to demonstrate that blocking reactions. RSV infection induced the expression of a number of important cytokines and chemokines in the lungs of primary infected and reinfected cotton rats. Our present interest in COX-2 led us to clone the cotton rat COX-2 gene. Cotton rat COX-2 cDNA (Gen- phages and bronchioalveolar macrophages were isolated from cot- Bank accession no. AY065644) is 90.5, 90.3, and 85.4% identical ton rats and infected with RSV at m.o.i. of 0.5 and 1, as indicated with the rat (accession no. S67722), mouse (accession no. (Fig. 5). We found an increase in the expression of COX-2 protein NM_011198), and human (accession no. L15326) homologues, re- 1 day postvirus infection, which persisted on day 2 with levels of spectively, with corresponding amino acid sequence homologies of COX-2 protein that approach those obtained after stimulation of 94, 93, and 87%. macrophages with LPS. Virus replicates very inefficiently in these LPS is one of the most powerful inducers of COX-2 in macro- cells, with only marginal levels of infective virus obtained 2 days phages (26, 36, 37). To determine whether COX-2 expression is postinfection (between 2 and 3 log10 less that what was seen in induced and can be detected in cotton rat cells, peritoneal exudate A549 cells), but the expression of the RSV F protein gene was macrophages were treated with LPS (Fig. 5A, left panel; Fig. 5B, detectable by RT-PCR. Taken together, these data indicate that LPS lanes). A strong induction of a protein that migrated at the both lung epithelial cells and macrophages (peritoneal and bron- expected Mr of COX-2 protein (72 kDa) was detected with an Ab chioalveolar) are potential sources of COX-2 and PG during in- that recognizes the conserved C-terminal amino acid sequence of fection with RSV. the COX-2 protein. Previous studies demonstrated the infection of peritoneal mac- RSV induction of COX-2 in vivo rophages and neonatal monocytes with RSV in vitro, which re- To investigate in vivo the expression of COX-2 during RSV in- sulted in the expression of several cytokines, including IL-1, fection, cotton rats were infected intranasally with 106.5 PFU. On TNF-␣, and IL-6 (38, 39). Moreover, Bartz et al. (18) showed that different days after infection, animals were killed, and total lung

PGE2 is produced by RSV-infected human macrophages and den- RNA was isolated. Steady state levels of COX-2 mRNA were dritic cells. To investigate whether COX-2 is induced in cotton rat analyzed by RT-PCR (Fig. 6A). COX-2 mRNA expression was macrophages during RSV infection, puriﬁed peritoneal macro- detected at very low levels in uninfected animals, whereas a small 4360 RSV AND COX-2

FIGURE 5. RSV induction of COX-2 protein expression in cotton rat peritoneal (A) and alveolar (B) macrophages. Macrophages were infected

FIGURE 4. Production of PGE2 in RSV-infected A549 cells. A, PGE2 with RSV (m.o.i., 0.5 and 1) and analyzed at the indicated times by West- Downloaded from levels measured in supernatants collected from RSV-infected A549 cells at ern blot for COX-2 and ␤-actin protein levels. Macrophages stimulated the indicated times post-RSV infection. Bars represent accumulated PGE2 with 100 ng/ml LPS (LPS lanes) were included as a positive control. For concentrations (nanograms per milliliter of tissue culture) Ϯ SD. B, Inhi- negative controls (Ϫ), mock-treated macrophages were used. Detection of bition of PGE2 accumulation in culture supernatants of A549 cells infected RSV F protein gene expression was assessed by RT-PCR and ethidium with RSV by NS-398 (a COX-2-selective inhibitor), indomethacin (a pan bromide staining of agarose gels. COX inhibitor), and SC-560 (a COX-1-selective inhibitor). Cells were infected with RSV (m.o.i., 0.1) and 6 h later were treated with the indicated http://www.jimmunol.org/ concentrations of inhibitors. Cell supernatants were harvested on day 3 slightly after 12 h, but it remained detectable throughout day 3 postin- postinfection, and PGE2 was measured by ELISA. The total amount of fection (on days 1 and 2 this was evidenced after ␤-actin normaliza- PGE2 produced on day 3 of RSV infection in the absence of COX inhib- tion). A second peak was seen on day 5 postinfection and correlated itors (vehicles) was 10.9 Ϯ 1.5 ng/ml for saline and 16 Ϯ 0.8 ng/ml for with the peak of lung pathology during primary RSV infection in DMSO-treated cells (indicated as 100%). cotton rats (29). These data clearly demonstrate that COX-2 is strongly induced in vivo during primary RSV infection in the lung. BAL washes in humans have been used to analyze the expres- increase was seen in animals inoculated with a mock preparation sion of cytokines and chemokines after RSV infection. We ana- of the virus. However, the presence of RSV in the inoculum gen- lyzed the expression of COX-2 in cells contained in the BAL of by guest on September 27, 2021 erated a triphasic pattern of COX-2 gene expression. In the first RSV-infected cotton rats by Western analysis. We did not detect 6 h, the expression peaked at its maximum (Ͼ8-fold compared any immunoreactive COX-2 in samples of BAL from mock-in- with uninfected animals and ϳ5-fold compared with mock-infected animals (Fig. 6C). We detected an increase in COX-2 ex- fected controls) and then strikingly diminished at 12 h postinfec- pression in BAL from three of five animals killed at 6 h postin- tion. COX-2 mRNA levels slowly climbed to a second peak on day fection. At 12 h and 1 day postinfection, the samples taken from all 1, again declining on day 2, and finally climbing again to a third animals tested demonstrated a strong induction of COX-2 protein. peak on day 5. Interestingly, levels of ␤-actin mRNA expression Cells in BAL of healthy (uninfected) animals were mostly mac- were reduced on days 1 and 2, just before the viral titers in lungs rophages (68 Ϯ 6.27%; n ϭ 10) with some granulocytes (ϳ30%) reached maximum levels (data not shown). COX-2 mRNA slowly and almost no lymphocytes. The number of granulocytes in the decreased after day 5, and on days 7 and 10 postinfection, its levels lung increased during the first day of RSV infection, whereas the were indistinguishable from control values. These data indicate number of lymphocytes increased starting on day 3 post-RSV in- that RSV induces the expression of COX-2 mRNA in lungs of fection (40). The expression of COX-2 persisted in some animals infected cotton rats. until day 5 postinfection (Fig. 6C), although levels were barely To determine whether the change in steady state levels of COX-2 detectable. Taken together, these data clearly demonstrate the in- mRNA correlates with a change in the expression of COX-2 protein duction of COX-2 mRNA and protein expression in the lungs and in the lung, we analyzed lung homogenates of RSV-infected cotton cells of the BAL after RSV infection of cotton rats. rats. Groups of animals were infected and killed at different times postinfection. Because primary infection in cotton rats usually affects Effect of inhibition of COX activity on RSV-induced lung Ͻ2% of the lung tissue (19), Western analysis of whole lung homog- histopathology enates was not sensitive enough to detect local mediators of inflam- To determine the role of COX in RSV pathogenesis, the effect of mation. To increase the signal-to-noise ratio, lung homogenates were indomethacin treatment on RSV-induced lung pathology was eval- used as substrate for immunoprecipitation of COX-2 protein. By us- uated. Cotton rats were treated with indomethacin i.p. at a dose 20 ing this method, a large volume of tissue can be analyzed at once. ␮g/100 g rat given 12 h before inoculation with RSV. This dose, COX-2 protein was not detected in immunoprecipitates from lungs of administered i.v., is equivalent to that used to promote closure of mock-infected cotton rats (Fig. 6B). We concurrently performed the ductus arteriosus in infants. Indomethacin was readministered Western blot for the detection of ␤-actin protein as evidence of the at 12 and 36 h postinfection. Control groups consisted of unin- total protein in the original lung homogenates. COX-2 was immuno- fected animals, mock-infected animals, and animals infected with precipitated in all infected animals during the first 7 days postinfection RSV and treated i.p. with saline. All animals were killed 4 days with RSV. There were two peaks of immunoreactive COX-2. The postinfection, when peak pathology scores for RSV have been first was detected as early as 6 h postinfection, which decreased shown (41). Three histopathologic parameters were assessed on The Journal of Immunology 4361

FIGURE 6. RSV induction of COX-2 in vivo. Cotton rats were inocu- 6.5 lated intranasally with 100 ␮l of RSV suspension containing 1 ϫ 10 Downloaded from PFU. Animals were killed at the indicated intervals. Both uninfected (Ϫ) cotton rats and cotton rats inoculated with uninfected tissue culture supernatant (MOCK) were used as controls. A, RT-PCR densitometry for quantification of steady state levels of COX-2 mRNA normalized to ␤-actin mRNA. Determinations were made in triplicate. Each bar represents data collected from five or more rats, and the results shown are representative of three independent experiments. B, Western blot analysis of COX-2 pro- http://www.jimmunol.org/ tein levels in whole lung homogenates after immunoprecipitation with an anti-COX-2-specific Ab (Santa Cruz Biotechnology; sc-1746). Western blot was performed using a different anti-COX-2 Ab (Cayman Chemical; catalogue no. 160116). Western blot for ␤-actin was performed as a control for protein loading consistency. The m and M lanes represent cotton rat macrophage lysates precipitated with a control normal rabbit sera or polyclonal rabbit anti-COX-2, respectively. C, COX-2 expression in BAL of RSV-infected cotton rats. Total cell extracts of BAL were immunoprecipi- tated and analyzed by Western blot for COX-2 as described in B. A West- by guest on September 27, 2021 ern blot for ␤-actin was performed as a control for protein loading consistency. each slide: peribronchiolitis, alveolitis, and interstitial pneumonitis. Fig. 7A illustrates mean histopathology scores. Inhibition of FIGURE 7. Inhibition of COX activity decreases RSV-induced bron- cyclooxygenase activity by indomethacin resulted in a significant chiolitis, alveolitis, and interstitial pneumonitis in cotton rats. A, Histopa- thology scores of bronchiolitis, interstitial pneumonia, and alveolitis. Each reduction in all three histopathological score parameters measured -Val ,ء .bar represents the arithmetic mean of eight to 10 animals Ϯ SEM when compared with untreated or to mock-treated control groups ues in indomethacin-treated animals that are significantly decreased (p Ͻ (Fig. 7C shows representative views of lung histology from each 0.05) from those in infected, saline-treated groups. B, Groups of animals group). There were no statistical differences in viral replication were treated with indomethacin (indo) or saline solution (saline) i.p. three measured in the lungs of infected animals treated with vehicle or times at 24-h intervals for each treatment reproducing a similar regimen as indomethacin on day 4 postinfection (104.6Ϯ 0.09 vs 104.5Ϯ 0.15 that in A. No virus infection was performed in these animals. Blood sam- PFU/g), indicating that indomethacin treatment did not affect viral ples were collected 24 h after the last treatment and induced with the replication in vivo. The same regimen of indomethacin was tested calcium ionophore (A23187) in vitro, and TxB2 production was measured ء in a separate group of animal to determine its bioavailability. In- by ELISA. , TxB2 production was significantly decreased by indometh- Ͻ domethacin treatment diminished TxB production (a nonenzy- acin treatment (p 0.05). C, Photomicrograph of H&E-stained lung tissue 2 (ϫ64). Normal lung architecture is seen in panel a (uninfected cotton rat). matic metabolite of TxA2) from whole blood platelets stimulated ϳ In panel b, lungs of RSV-infected cotton rats killed on day 4 postinfection with calcium ionophore (A23187) by 40% (Fig. 7B) compared are shown. Significant alveolar infiltration with inflammatory cells, peri- with animals treated i.p. with saline. These data indicate that in- bronchiolar thickening with lymphocytic infiltration, and coarse interstitial hibition of COX activity by indomethacin is more likely the cause thickening are seen. Panel c, RSV-infected cotton rats were treated i.p. of the reduction in pathology scores observed in infected and with three doses of indomethacin. Significant reductions in alveolitis, bron- treated animals. chiolitis, and interstitial pneumonitis are seen.

Discussion In the present study we have demonstrated that RSV infection results in induction of COX-2 gene and protein expression both in Induction of COX-2 expression has been demonstrated for a vitro, in human lung epithelial cells and cotton rat macrophages, handful of other viruses, including human CMV, human herpes- and in vivo, in whole lung and BAL cells of cotton rats infected virus, EBV, encephalomyocarditis, and Theiler’s virus (42–46). In with RSV. To our knowledge, this is the ﬁrst report that analyzes an in vivo study of bronchial biopsies from human volunteers ex- the expression of COX-2 during RSV infection. perimentally infected with human rhinovirus, the levels of COX-2 4362 RSV AND COX-2 immunostaining were increased, particularly in macrophages, eo- acting with TLR4 (57–59). Moreover, TLR4-deﬁcient mice exhibit sinophils, and mast cells (47). an impaired capacity to clear RSV from the lungs (57, 59). In Production of eicosanoids has been documented in the cases of addition, early activation of NF-␬B in the lungs of infected mice

RSV and other related viruses. RSV induction of PGE2 was re- was shown to be both alveolar macrophage- and TLR4-dependent cently demonstrated in RSV-infected human cord blood macro- (59). This initial interaction between the pathogen and host would phages and dendritic cells (18). Moreover, three studies, two using be predicted to elicit a cascade of inflammatory mediators that, in cattle that were experimentally infected with bovine RSV (48, 49) turn, would potentiate the inflammation response associated with and one in infants who were positive for RSV infection (50), RSV. Taken together, these data suggest that a different mecha- showed increased levels of plasma PGE2 and 6-keto-PGF1␣. These nism of induction of COX-2 is used by macrophages (or dendritic data indicate that viral infection produces fluctuations in the del- cells) and lung epithelial cells; in macrophages, COX-2 induction icate balance of local as well as systemic PGs and suggest that is dependent upon TLR4 engagement and is independent of viral these mediators must be considered as potentially important fac- replication, whereas in epithelial cells, COX-2 induction requires tors in the natural disease process. viral replication.

The effect of PGE2 on viral replication has been studied for Our results indicate that COX-2 mRNA and protein expression several viruses with mixed results (reviewed in Ref. 51). In our showed three peaks in lungs of cotton rats during primary infection study, inhibition of COX enzyme did not produce measurable with RSV (Fig. 6). The ﬁrst was detected as soon as 6 h postin- changes in viral replication in A549 cell or in lungs of animals fection and coincided with the induction of mRNA for cytokines treated with indomethacin, suggesting that inhibition of COX en- such as IL-1␤ and TNF-␣ and chemokines such as IFN-inducible zyme(s) has an effect on the host inﬂammatory response and not on protein-10 and GRO (M. Boukhvalova and J. C. G. Blanco, un- Downloaded from the pathology caused by virus replication. published observations). Although we cannot rule out a possible Epithelial cells are the main target for RSV replication in the role of these factors in the induction of COX-2 in vivo, their co- lung. We found that RSV infection was an effective inducer of expression at this early time point most likely indicates the use of both COX-2 and PGE2 in A549 lung epithelial cells and that the shared mechanisms of gene induction triggered during innate re- expression of COX-2 requires viral replication. We detected the sponses to the virus, rather than an induction through secondary

induction of COX-2 mRNA 40 h postinfection, whereas no induc- mediators. The second peak on days 1 and 2 and the third peak on http://www.jimmunol.org/ tion was seen at earlier times (e.g., 6, 12, 24 h; data not shown). day 5 correlate with viral replication and pathology scores, respec- The levels of COX-2 mRNA expression paralleled those of the tively (29). Although the final eicosanoid products that are syn- protein, which peaked on day 2 postinfection as well. Longer time thesized during each of these instances are potentially different and periods were not studied due to the sharp decrease in cell viability may have different impacts on lung physiology, our data indicate which precluded accurate measurement of COX-2 mRNA and pro- that partial inhibition of eicosanoid production during infection is tein expression. Airway epithelial cells respond to proinflamma- beneficial for the pathological outcome. Eicosanoid function in the tory cytokines such as TNF-␣ (14) and IL-1 (52) by induction of airways has been subdivided into stimulatory PGs (bronchocon-

COX-2 and PGE2 release. These mediators are induced in A549 strictors) such as PGD2, PGF2 ␣ and TBx, on the one hand, and by guest on September 27, 2021 cells infected with RSV (30, 32) and may have some effect on inhibitory PGs (bronchodilators) such as PGE2, on the other (60). RSV-induced COX-2 at later times postinfection, because their PGE2 is the most abundant PG in lungs (60) and is considered to presence in tissue culture of these cells after RSV infection is low be a potent immunosuppressor. In vitro, PGE2 activates the Th2 or undetectable before 30 h postinfection (30). Together, these data subset of T cells while suppressing the Th1 subset (61). PGE2 suggested that COX-2 induction in A549 cells was more likely the suppression of Th1 responses is due to the selective and dose- result of viral replication, acting directly or through combined ac- dependent inhibition of IFN-␥ production by stimulated human T tion with virus-induced cytokines in an autocrine loop. First, we cells (62). PGE2 also induces apoptosis of thymocytes and inhibits have demonstrated that a soluble factor present in supernatants of T cell function, such as the production of IL-2 (63). In vivo, PGE2 A549 cells infected for 48 h with RSV was responsible for the was shown to suppress the production of Th1 cytokines such as induction of COX-2 within a 6-h period (Fig. 3A). Second, we IL-12 (64). A recent study demonstrated that treatment with NS- demonstrated that the activity of these supernatants was com- 398, the speciﬁc COX-2 inhibitor, promoted a Th1 immune re- pletely blocked by Abs against IL-1␣ and, to a lesser extent, by sponse in BALB/c mice (65). In the same model, RSV infection Abs against TNF-␣ and IL-1␤ (Fig. 3B). These results correlate increased Th2-type cytokines and increased CD4ϩ T cells in BAL with the amount of these cytokines contained in supernatants of ﬂuid from challenged animals immunized with formalin-inacti- A549 cells infected with RSV (IL-1␣ Ͼ TNF-␣ Ͼ IL-1␤) (30). vated RSV vaccine (1). Taken together, these data suggest that

Macrophages are also targets of RSV infection (53). Although excessive COX-2-mediated production of PGE2 during RSV in- less efficient at replicating virus, these cells are an important fection could potentially harm the host by repressing a desirable source of inflammatory mediators and have also been involved in Th1 response while promoting an undesirable Th2 response. the expression of COX-2 during inflammatory processes. Our data Because both PGs and leukotrienes (LT) are clinically important demonstrate that COX-2 was induced in RSV-infected cotton rat proinflammatory molecules in the lung, and they both are the prod- alveolar and peritoneal macrophages. These data complement a ucts of arachidonic acid, inhibiting one pathway (in this case, PGs) previous study using macrophages and dendritic cells (18), in would be anticipated to result in changes in the other pathway (LT) which the authors reported that secretion of PGE2 by these cells in the host. RSV infection induces the expression of 5-lipoxygen- was independent of viral infectivity (UV-irradiated RSV induced ase, a key enzyme in the production of LTs from arachidonic acid levels of COX-2 expression comparable to wild-type virus). To- (66). Determination of changes induced by the expression of gether, these data suggest that receptor-triggered signal transduc- COX-2 and 5-lipoxygenase during RSV infection will be impor- tion mechanisms might also be involved in the induction of tant for understanding the dynamic interplay between PG and LT COX-2. It has been demonstrated in some inflammatory settings pathways and for the optimization of therapies that target these that COX-2 activation is dependent on the TLR4 signal transduc- pathways. tion pathway (54–56). Recently, it was shown that RSV F protein Vaccine programs spanning 4 decades have yet to result in a is able to activate the innate immune response directly by inter- safe and effective RSV vaccine. A major breakthrough in RSV The Journal of Immunology 4363 prophylaxis was the development and licensure of both polyclonal 10. Bont, L., C. J. Heijnen, A. Kavelaars, W. M. van Aalderen, F. Brus, and monoclonal Ig products for use in high risk infants (RespiGam J. M. Draaisma, M. Pekelharing-Berghuis, R. A. van Diemen-Steenvoorde, and J. L. Kimpen. 2001. Local interferon-␥ levels during respiratory syncytial virus (RSV Ig) and Synagis (palivizumab); MedImmune). Although lower respiratory tract infection are associated with disease severity. J. Infect. highly effective in a preventive setting, reducing the hospitaliza- Dis. 184:355. tion of high risk infants by up to 80% (22), Ig has not achieved 11. Laporte, J. D., P. E. Moore, J. H. Abraham, G. N. Maksym, B. Fabry, R. A. Panettieri, Jr., and S. A. Shore. 1999. Role of ERK MAP kinases in re- success as a therapeutic agent. Despite the fact that Ig-treated pa- sponses of cultured human airway smooth muscle cells to IL-1␤. Am. J. Physiol. tients showed a reduction in virus titers (67, 68), there was no 277:L943. clinically beneficial outcome from the treatment (67–70). Similar 12. Laporte, J. D., P. E. Moore, T. Lahiri, I. N. Schwartzman, R. A. Panettieri, Jr., and S. A. Shore. 2000. p38 MAP kinase regulates IL-1␤ responses in cultured airway observations were made in the cotton rat, in which treatment of smooth muscle cells. Am. J. Physiol. 279:L932. infected animals with palivizumab exclusively eliminated RSV 13. Fong, C. Y., L. Pang, E. Holland, and A. J. Knox. 2000. TGF-␤1 stimulates IL-8 from the host, but had no significant effect on pulmonary histopa- release, COX-2 expression, and PGE2 release in human airway smooth muscle cells. Am. J. Physiol. 279:L201. thology (41). Successful treatment of RSV bronchiolitis and pneu- 14. Chen, C. C., Y. T. Sun, J. J. Chen, and K. T. Chiu. 2000. TNF-␣-induced cy- monia, however, was possible when a combination of antiviral clooxygenase-2 expression in human lung epithelial cells: involvement of the (palivizumab) and anti-inflammatory (glucocorticoids) drugs was phospholipase C-␥ 2, protein kinase C-␣, tyrosine kinase, NF-␬B-inducing kinase, and I-␬B kinase 1/2 pathway. J. Immunol. 165:2719. used in cotton rats (41). Thus, the histopathology induced by RSV 15. Pang, L. 2001. COX-2 expression in asthmatic airways: the story so far. Thorax infection is now largely attributed to the inflammatory response to 56:335. infection. Glucocorticoids, however, are nonspecific immunosup- 16. Smith, W. L. 1992. Prostanoid biosynthesis and mechanisms of action. pressors whose use is generally discouraged during infectious dis- Am. J. Physiol. 263:F18. 17. Dubois, R. N., S. B. Abramson, L. Crofford, R. A. Gupta, L. S. Simon, eases, particularly in children. Study of the roles of the multiple L. B. Van De Putte, and P. E. Lipsky. 1998. Cyclooxygenase in biology and Downloaded from inflammatory components regulated during RSV infection will po- disease. FASEB J. 12:1063. tentially open the door to new and possibly more specific thera- 18. Bartz, H., F. Buning-Pfaue, O. Turkel, and U. Schauer. 2002. Respiratory syncytial virus induces prostaglandin E2, IL-10 and IL-11 generation in antigen peutic strategies. One of these components, COX-2, has not been presenting cells. Clin. Exp. Immunol. 129:438. studied in association with RSV infection before the work reported 19. Prince, G. A., A. B. Jenson, R. L. Horswood, E. Camargo, and R. M. Chanock. in this paper. Full treatment experiments with administration of 1978. The pathogenesis of respiratory syncytial virus infection in cotton rats. Am. J. Pathol. 93:771. high concentrations of the nonsteroidal anti-inflammatory drug, 20. Prince, G. A., R. L. Horswood, and R. M. Chanock. 1985. Quantitative aspects http://www.jimmunol.org/ indomethacin, shortly before and after infection indicated that of passive immunity to respiratory syncytial virus infection in infant cotton rats. COX inhibitors significantly decrease lung pathology associated J. Virol. 55:517. 21. Prince, G. A., R. L. Horswood, D. W. Koenig, and R. M. Chanock. 1985. An- with RSV infection in the cotton rat. Furnished with this informa- tigenic analysis of a putative new strain of respiratory syncytial virus. J. Infect. tion, our final goal is to modify the current treatment protocol from Dis. 151:634. a prophylactic one to one that can be applicable therapeutically. 22. Meissner, H. C., R. C. Welliver, S. A. Chartrand, B. J. Law, L. E. Weisman, H. L. Dorkin, and W. J. Rodriguez. 1999. Immunoprophylaxis with palivizumab, a humanized respiratory syncytial virus monoclonal antibody, for prevention of Acknowledgments respiratory syncytial virus infection in high risk infants: a consensus opinion. We thank Dr. Mike Flora for his sequencing services, Drs. Maryna Eichel- Pediatr. Infect. Dis. J. 18:223. berger and Bettina Richter, for helpful discussions, and Kevin Yim, Lorraine 23. Prevent Study Group. 1997. Reduction of respiratory syncytial virus hospitaliza- by guest on September 27, 2021 tion among premature infants and infants with bronchopulmonary dysplasia using Ward, Victor Tineo, Layla Soroush, and Charles Smith for assistance with the respiratory syncytial virus immune globulin prophylaxis. Pediatrics 99:93. animals and viral titrations. 24. IMpact-RSV Study Group. 1998. Palivizumab, a humanized respiratory syncytial virus monoclonal antibody, reduces hospitalization from respiratory syncytial Disclosures virus infection in high-risk infants. Pediatrics 102:531. The authors have no financial conflict of interest. 25. McIntire, F. C., H. W. Sievert, G. H. Barlow, R. A. Finley, and A. Y. Lee. 1967. Chemical, physical, biological properties of a lipopolysaccharide from Esche- richia coli K-235. Biochemistry 6:2363. References 26. Blanco, J. C., C. Contursi, C. A. Salkowski, D. L. DeWitt, K. Ozato, and 1. Graham, B. S., J. A. Rutigliano, and T. R. Johnson. 2002. Respiratory syncytial S. N. Vogel. 2000. Interferon regulatory factor (IRF)-1 and IRF-2 regulate in- ␥ virus immunobiology and pathogenesis. Virology 297:1. terferon -dependent cyclooxygenase 2 expression. J. Exp. Med. 191:2131. 2. McCarthy, A. J., H. M. Kingman, C. Kelly, G. S. Taylor, E. O. Caul, D. Grier, 27. Blanco, J. C., J. Y. Richardson, M. E. Darnell, A. Rowzee, L. Pletneva, J. Moppett, A. B. Foot, J. M. Cornish, A. Oakhill, et al. 1999. The outcome of 26 D. D. Porter, and G. A. Prince. 2002. Cytokine and chemokine gene expression patients with respiratory syncytial virus infection following allogeneic stem cell after primary and secondary respiratory syncytial virus infection in cotton rats. transplantation. Bone Marrow Transplant. 24:1315. J. Infect. Dis. 185:1780. 3. Thompson, W. W., D. K. Shay, E. Weintraub, L. Brammer, N. Cox, 28. Smith, C. J., Y. Zhang, C. M. Koboldt, J. Muhammad, B. S. Zweifel, A. Shaffer, L. J. Anderson, and K. Fukuda. 2003. Mortality associated with influenza and J. J. Talley, J. L. Masferrer, K. Seibert, and P. C. Isakson. 1998. Pharmacological respiratory syncytial virus in the United States. J. Am. Med. Assoc. 289:179. analysis of cyclooxygenase-1 in inflammation. Proc. Natl. Acad. Sci. USA 4. Welliver, R. C., D. T. Wong, M. Sun, E. Middleton, Jr., R. S. Vaughan, and 95:13313. P. L. Ogra. 1981. The development of respiratory syncytial virus-specific IgE and 29. Prince, G. A., J. P. Prieels, M. Slaoui, and D. D. Porter. 1999. Pulmonary lesions the release of histamine in nasopharyngeal secretions after infection. N. Engl. in primary respiratory syncytial virus infection, reinfection, and vaccine-en- J. Med. 305:841. hanced disease in the cotton rat (Sigmodon hispidus). Lab. Invest. 79:1385. 5. Sigurs, N., R. Bjarnason, and F. Sigurbergsson. 1994. Eosinophil cationic protein 30. Patel, J. A., M. Kunimoto, T. C. Sim, R. Garofalo, T. Eliott, S. Baron, in nasal secretion and in serum and myeloperoxidase in serum in respiratory O. Ruuskanen, T. Chonmaitree, P. L. Ogra, and F. Schmalstieg. 1995. Interleu- syncytial virus bronchiolitis: relation to asthma and atopy. Acta Paediatr. kin-1␣ mediates the enhanced expression of intercellular adhesion molecule-1 in 83:1151. pulmonary epithelial cells infected with respiratory syncytial virus. Am. J. Respir. 6. Sigurs, N., R. Bjarnason, F. Sigurbergsson, B. Kjellman, and B. Bjorksten. 1995. Cell Mol. Biol. 13:602. Asthma and immunoglobulin E antibodies after respiratory syncytial virus bron- 31. Arnold, R., B. Konig, H. Galatti, H. Werchau, and W. Konig. 1995. Cytokine chiolitis: a prospective cohort study with matched controls. Pediatrics 95:500. (IL-8, IL-6, TNF-␣) and soluble TNF receptor-I release from human peripheral 7. Hornsleth, A., L. Loland, and L. B. Larsen. 2001. Cytokines and chemokines in blood mononuclear cells after respiratory syncytial virus infection. Immunology respiratory secretion and severity of disease in infants with respiratory syncytial 85:364. virus (RSV) infection. J. Clin. Virol. 21:163. 8. Sheeran, P., H. Jafri, C. Carubelli, J. Saavedra, C. Johnson, K. Krisher, 32. Patel, J. A., Z. Jiang, N. Nakajima, and M. Kunimoto. 1998. Autocrine regulation ␣ P. J. Sanchez, and O. Ramilo. 1999. Elevated cytokine concentrations in the of interleukin-8 by interleukin-1 in respiratory syncytial virus-infected pulmo- nasopharyngeal and tracheal secretions of children with respiratory syncytial vi- nary epithelial cells in vitro. Immunology 95:501. rus disease. Pediatr. Infect. Dis. J. 18:115. 33. O’Banion, M. K., V. D. Winn, and D. A. Young. 1992. cDNA cloning and 9. Bont, L., C. J. Heijnen, A. Kavelaars, W. M. van Aalderen, F. Brus, functional activity of a glucocorticoid-regulated inflammatory cyclooxygenase. J. T. Draaisma, S. M. Geelen, and J. L. Kimpen. 2000. Monocyte IL-10 produc- Proc. Natl. Acad. Sci. USA 89:4888. tion during respiratory syncytial virus bronchiolitis is associated with recurrent 34. Ristimaki, A., S. Garfinkel, J. Wessendorf, T. Maciag, and T. Hla. 1994. Induc- wheezing in a one-year follow-up study. Am. J. Respir. Crit. Care Med. tion of cyclooxygenase-2 by interleukin-1␣: evidence for post-transcriptional reg- 161:1518. ulation. J. Biol. Chem. 269:11769. 4364 RSV AND COX-2

35. Warner, T. D., F. Giuliano, I. Vojnovic, A. Bukasa, J. A. Mitchell, and J. R. Vane. cytokines in human pulmonary epithelial cells: regulation by dexamethasone. 1999. Nonsteroid drug selectivities for cyclo-oxygenase-1 rather than cyclo-ox- Br. J. Pharmacol. 113:1008. ygenase-2 are associated with human gastrointestinal toxicity: a full in vitro anal- 53. Panuska, J. R., N. M. Cirino, F. Midulla, J. E. Despot, E. R. McFadden, Jr., and ysis. Proc. Natl. Acad. Sci. USA 96:7563. Y. T. Huang. 1990. Productive infection of isolated human alveolar macrophages 36. O’Sullivan, M. G., F. H. Chilton, E. M. Huggins, Jr., and C. E. McCall. 1992. by respiratory syncytial virus. J. Clin. Invest. 86:113. Lipopolysaccharide priming of alveolar macrophages for enhanced synthesis of 54. Rhee, S. H., and D. Hwang. 2000. Murine Toll-like receptor 4 confers lipopoly- prostanoids involves induction of a novel prostaglandin H synthase. J. Biol. saccharide responsiveness as determined by activation of NF␬B and expression Chem. 267:14547. of the inducible cyclooxygenase. J. Biol. Chem. 275:34035. 37. Zhang, S., K. Thomas, J. C. Blanco, C. A. Salkowski, and S. N. Vogel. 2002. The 55. Perera, P. Y., T. N. Mayadas, O. Takeuchi, S. Akira, M. Zaks-Zilberman, role of interferon regulatory factors, IRF-1 and IRF-2, in LPS induced cycloox- S. M. Goyert, and S. N. Vogel. 2001. CD11b/CD18 acts in concert with CD14 ygenase-2 (COX-2) expression in vivo and in vitro. J. Endotoxin Res. 8:379. and Toll-like receptor (TLR) 4 to elicit full lipopolysaccharide and taxol-induc- 38. Tsutsumi, H., K. Matsuda, S. Sone, R. Takeuchi, and S. Chiba. 1996. Respiratory ible gene expression. J. Immunol. 166:574. syncytial virus-induced cytokine production by neonatal macrophages. Clin. Exp. 56. Medvedev, A. E., A. Lentschat, D. B. Kuhns, J. C. Blanco, C. Salkowski, Immunol. 106:442. S. Zhang, M. Arditi, J. I. Gallin, and S. N. Vogel. 2003. Distinct mutations in 39. Takeuchi, R., H. Tsutsumi, M. Osaki, S. Sone, S. Imai, and S. Chiba. 1998. IRAK-4 confer hyporesponsiveness to lipopolysaccharide and interleukin-1 in a Respiratory syncytial virus infection of neonatal monocytes stimulates synthesis patient with recurrent bacterial infections. J. Exp. Med. 198:521. of interferon regulatory factor 1 and interleukin-1␤ (IL-1␤)-converting enzyme 57. Kurt-Jones, E. A., L. Popova, L. Kwinn, L. M. Haynes, L. P. Jones, R. A. Tripp, and secretion of IL-1␤. J. Virol. 72:837. E. E. Walsh, M. W. Freeman, D. T. Golenbock, L. J. Anderson, et al. 2000. 40. Prince, G. A., A. B. Jenson, V. G. Hemming, B. R. Murphy, E. E. Walsh, Pattern recognition receptors TLR4 and CD14 mediate response to respiratory R. L. Horswood, and R. M. Chanock. 1986. Enhancement of respiratory syncytial syncytial virus. Nat. Immunol. 1:398. virus pulmonary pathology in cotton rats by prior intramuscular inoculation of formalin-inactivated virus. J. Virol. 57:721. 58. Haynes, L. M., D. D. Moore, E. A. Kurt-Jones, R. W. Finberg, L. J. Anderson, 41. Prince, G. A., A. Mathews, S. J. Curtis, and D. D. Porter. 2000. Treatment of and R. A. Tripp. 2001. Involvement of Toll-like receptor 4 in innate immunity to respiratory syncytial virus bronchiolitis and pneumonia in a cotton rat model with respiratory syncytial virus. J. Virol. 75:10730. systemically administered monoclonal antibody (Palivizumab) and glucocortico- 59. Haeberle, H. A., R. Takizawa, A. Casola, A. R. Brasier, H. J. Dieterich, steroid. J. Infect. Dis. 182:1326. N. Van Rooijen, Z. Gatalica, and R. P. Garofalo. 2002. Respiratory syncytial Downloaded from 42. Zhu, H., J. P. Cong, D. Yu, W. A. Bresnahan, and T. E. Shenk. 2002. Inhibition virus-induced activation of nuclear factor-␬B in the lung involves alveolar mac- of cyclooxygenase 2 blocks human cytomegalovirus replication. Proc. Natl. rophages and Toll-like receptor 4-dependent pathways. J. Infect. Dis. 186:1199. Acad. Sci. USA 99:3932. 60. Hanley, S. P. 1986. Prostaglandins and the lung. Lung 164:65. 43. Janelle, M. E., A. Gravel, J. Gosselin, M. J. Tremblay, and L. Flamand. 2002. 61. Phipps, R. P., S. H. Stein, and R. L. Roper. 1991. A new view of prostaglandin Activation of monocyte cyclooxygenase-2 gene expression by human herpesvirus E regulation of the immune response. Immunol. Today 12:349. 6: role for cyclic AMP-responsive element-binding protein and activator pro- 62. Snijdewint, F. G., P. Kalinski, E. A. Wierenga, J. D. Bos, and M. L. Kapsenberg. tein-1. J. Biol. Chem. 277:30665. 1993. Prostaglandin E2 differentially modulates cytokine secretion proﬁles of

44. Murono, S., H. Inoue, T. Tanabe, I. Joab, T. Yoshizaki, M. Furukawa, and human T helper lymphocytes. J. Immunol. 150:5321. http://www.jimmunol.org/ J. S. Pagano. 2001. Induction of cyclooxygenase-2 by Epstein-Barr virus latent 63. McConkey, D. J., S. Orrenius, and M. Jondal. 1990. Cellular signalling in pro- membrane protein 1 is involved in vascular endothelial growth factor production grammed cell death (apoptosis). Immunol. Today 11:120. in nasopharyngeal carcinoma cells. Proc. Natl. Acad. Sci. USA 98:6905. 64. Newberry, R. D., W. F. Stenson, and R. G. Lorenz. 1999. Cyclooxygenase-2- 45. Steer, S. A., J. M. Moran, L. B. Maggi, Jr., R. M. Buller, H. Perlman, and dependent arachidonic acid metabolites are essential modulators of the intestinal J. A. Corbett. 2003. Regulation of cyclooxygenase-2 expression by macrophages immune response to dietary antigen. Nat. Med. 5:900. in response to double-stranded RNA and viral infection. J. Immunol. 170:1070. 65. Kuroda, E., T. Sugiura, K. Zeki, Y. Yoshida, and U. Yamashita. 2000. Sensitivity 46. Molina-Holgado, E., A. Arevalo-Martin, S. Ortiz, J. M. Vela, and C. Guaza. difference to the suppressive effect of prostaglandin E2 among mouse strains: a 2002. Theiler’s virus infection induces the expression of cyclooxygenase-2 in possible mechanism to polarize Th2 type response in BALB/c mice. J. Immunol. murine astrocytes: inhibition by the anti-inﬂammatory cytokines interleukin-4 164:2386. and interleukin-10. Neurosci. Lett. 324:237. 66. Behera, A. K., M. Kumar, H. Matsuse, R. F. Lockey, and S. S. Mohapatra. 1998. 47. Seymour, M. L., N. Gilby, P. G. Bardin, D. J. Fraenkel, G. Sanderson, Respiratory syncytial virus induces the expression of 5-lipoxygenase and endo- J. F. Penrose, S. T. Holgate, S. L. Johnston, and A. P. Sampson. 2002. Rhinovirus by guest on September 27, 2021 thelin-1 in bronchial epithelial cells. Biochem. Biophys. Res. Commun. 251:704. infection increases 5-lipoxygenase and cyclooxygenase-2 in bronchial biopsy specimens from nonatopic subjects. J. Infect. Dis. 185:540. 67. Hemming, V. G., W. Rodriguez, H. W. Kim, C. D. Brandt, R. H. Parrott, 48. Gershwin, L. J., S. N. Giri, R. S. Stewart, and J. Chen. 1989. Prostaglandin and B. Burch, G. A. Prince, P. A. Baron, R. J. Fink, and G. Reaman. 1987. Intrave- thromboxane concentrations in plasma and lung lavage ﬂuids during sequential nous immunoglobulin treatment of respiratory syncytial virus infections in infants infection of vaccinated and nonvaccinated calves with bovine respiratory syncy- and young children. Antimicrob. Agents Chemother. 31:1882. tial virus. Am. J. Vet. Res. 50:1254. 68. Rodriguez, W. J., W. C. Gruber, J. R. Groothuis, E. A. Simoes, A. J. Rosas, 49. Gershwin, L. J., and S. N. Giri. 1992. Effects of allergen challenge on plasma M. Lepow, A. Kramer, and V. Hemming. 1997. Respiratory syncytial virus im- concentrations of prostaglandins, thromboxane B2, and histamine in calves in- mune globulin treatment of RSV lower respiratory tract infection in previously fected with bovine respiratory syncytial virus. Am J Vet Res. 53:1670. healthy children. Pediatrics 100:937. 50. Skoner, D. P., P. Fireman, H. W. Davis, R. Wall, and L. A. Caliguiri. 1989. 69. Rimensberger, P. C., and U. B. Schaad. 1994. Clinical experience with aerosol- Increases in plasma concentrations of a prostaglandin metabolite in acute airway ized immunoglobulin treatment of respiratory syncytial virus infection in infants. obstruction. Arch. Dis. Child. 64:1112. Pediatr. Infect. Dis J. 13:328. 51. Steer, S. A., and J. A. Corbett. 2003. The role and regulation of COX-2 during 70. Rimensberger, P. C., A. Burek-Kozlowska, A. Morell, D. Germann, viral infection. Viral Immunol. 16:447. A. K. Eigenmann, F. Steiner, R. Burger, M. Kuenzli, and U. B. Schaad. 1996. 52. Mitchell, J. A., M. G. Belvisi, P. Akarasereenont, R. A. Robbins, O. J. Kwon, Aerosolized immunoglobulin treatment of respiratory syncytial virus infection in J. Croxtall, P. J. Barnes, and J. R. Vane. 1994. Induction of cyclo-oxygenase-2 by infants. Pediatr. Infect. Dis. J. 15:209.