Eosinophils Stimulated with IL-5 Or Igg Critically Involved in Effector

Home , Israpafant

Endogenous Platelet-Activating Factor Is Critically Involved in Effector Functions of Eosinophils Stimulated with IL-5 or IgG

This information is current as Kathleen R. Bartemes, Shannon McKinney, Gerald J. Gleich of October 3, 2021. and Hirohito Kita J Immunol 1999; 162:2982-2989; ; http://www.jimmunol.org/content/162/5/2982 Downloaded from References This article cites 40 articles, 14 of which you can access for free at: http://www.jimmunol.org/content/162/5/2982.full#ref-list-1

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

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

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on October 3, 2021 Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 1999 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Endogenous Platelet-Activating Factor Is Critically Involved in Effector Functions of Eosinophils Stimulated with IL-5 or IgG1

Kathleen R. Bartemes, Shannon McKinney, Gerald J. Gleich, and Hirohito Kita2

Eosinophil activation and subsequent release of inﬂammatory mediators are implicated in the pathophysiology of allergic diseases. Eosinophils are activated by various classes of secretagogues, such as cytokines (e.g., IL-5), lipid mediators (e.g., platelet-activating factor (PAF)), and Ig (e.g., immobilized IgG). However, do these agonists act directly on eosinophils or indirectly through the generation of intermediate active metabolites? We now report that endogenous PAF produced by activated eosinophils plays a critical role in eosinophil functions. Human eosinophils produced superoxide when stimulated with immobilized IgG, soluble IL-5, or PAF. Pretreating eosinophils with pertussis toxin abolished their responses to these stimuli, suggesting involvement of a metabolite(s) that acts on G proteins. Indeed, PAF was detected in supernatants from eosinophils stimulated with IgG or IL-5. Furthermore, structurally distinct PAF antagonists, including CV6209, hexanolamine PAF, and Y-24180 (israpafant), inhibited IgG- or IL-5-induced superoxide production and degranulation. Previous reports indicated that exogenous PAF stimulates eo- Downloaded from sinophil eicosanoid production through formation of lipid bodies. We found in this study that IgG or IL-5 also induces lipid body

formation and subsequent leukotriene C4 production mediated by endogenous PAF. Finally, inhibition of cytosolic phospholipase

A2, one of the key enzymes involved in PAF synthesis, attenuated both PAF production and effector functions of eosinophils. These findings suggest that endogenous PAF plays important roles in eosinophil functional responses to various exogenous stimuli, such as cytokines and Igs. Therefore, inhibition of PAF synthesis or action may be beneficial for the treatment of eosinophilic inflammation. The Journal of Immunology, 1999, 162: 2982–2989. http://www.jimmunol.org/

osinophils play important roles in the pathophysiology of voke eosinophil functions directly through independent signaling bronchial asthma and other allergic diseases (reviewed in pathways or act indirectly through the generation of common in- E Ref. 1). In such diseases, mediators released by T cells, termediate active metabolites. epithelial cells, and other inﬂammatory cells induce migration of PAF (1-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine) is a po- eosinophils from blood into the affected tissues (reviewed in Ref. tent phospholipid mediator synthesized by a range of cell types, 2). Subsequently, appropriate stimuli activate the eosinophils, re- including eosinophils (11, 12). PAF binds to a G protein-coupled

sulting in the local release of several inflammatory mediators, such seven-transmembrane receptor (13), and exerts various biological by guest on October 3, 2021 as leukotrienes (2, 3), superoxide anion (4), and toxic cationic activities, such as platelet activation, airway constriction, devel- granule proteins (5). Eosinophil infiltration and deposition of re- opment of bronchial hyperresponsiveness, and induction of micro- leased granule proteins are hallmarks of tissues from patients with vascular leakage and edema (reviewed in Refs. 14 and 15). Be- allergic diseases (6, 7). Although the mechanism(s) of eosinophil cause of these potent actions, PAF may play a role in the activation and mediator release in human diseases still needs to be pathophysiology of bronchial asthma (reviewed in Ref. 15). More- elucidated, a wide range of stimuli can induce eosinophil effector over, in eosinophils, PAF induces chemotaxis (16), degranulation functions in vitro. They include Ig (e.g., immobilized IgA and IgG) (17), and generation of superoxide anion (8, 18) and arachidonic 3 (5), lipid mediators (e.g., leukotriene B4 (LTB4) and platelet-ac- acid metabolites (19). Altogether, these observations as well as the tivating factor (PAF)) (8), cytokines (e.g., IL-5 and granulocyte- eosinophil’s capacity to produce PAF (3, 12) suggest that PAF macrophage CSF) (9), and complement fragments (e.g., C5a) (10). may act as an autocrine mediator promoting eosinophil functions. Structures of receptors for these secretagogues and intracellular In this study, we investigated the roles of endogenous PAF on signaling molecules associated with these receptors vary consid- eosinophil effector functions. Our findings suggest PAF is an es- erably. Therefore, the question exists whether these agonists pro- sential autocrine mediator in superoxide production, degranulation, and eicosanoid formation. Furthermore, endogenous PAF may amplify the activation signals provoked by various receptors, Departments of Immunology and Internal Medicine, Mayo Clinic and Mayo Foun- such as cytokine receptors and Fc receptors, culminating in the dation, Rochester, MN 55905 full-blown functions of eosinophils. Received for publication August 14, 1998. Accepted for publication November 25, 1998. Materials and Methods 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 Reagents with 18 U.S.C. Section 1734 solely to indicate this fact. PAF and CV6209 were purchased from BIOMOL Research Laboratories 1 This work was supported by grants from the National Institutes of Health (AI 34486 (Plymouth Meeting, PA), dissolved in absolute ethanol at 40 and 10 mM, and AI 34577) and by the Mayo Foundation. respectively, and stored at Ϫ20°C. 1-O-hexadecyl-2-acetyl-sn-glycero-3- 2 Address correspondence and reprint requests to Dr. Hirohito Kita, Department of phospho-(N,N,N-trimethyl)-hexanolamine (hexanolamine PAF) and PMA Immunology, Mayo Clinic, 200 First Street SW, Rochester, MN 55905. E-mail ad- were purchased from Calbiochem (La Jolla, CA), dissolved in water and dress: [email protected] DMSO at 20 mM and 5 mg/ml, respectively, and stored at Ϫ20°C. Per- ␮ 3 Abbreviations used in this paper: LT, leukotriene; EDN, eosinophil-derived neuro- tussis toxin (PTX) from Calbiochem was dissolved in water at 100 g/ml toxin; HSA, human serum albumin; IC50, 50% inhibiting concentrations; PAF, plate- and stored at 4°C. Y-24180 (israpafant) was a gift from Yoshitomi Phar- let-activating factor; PLA2, phospholipase A2; PTX, pertussis toxin. maceutical Industries (Fukuoka, Japan); it was dissolved in DMSO at 20

mM and stored at 20°C. Mepacrine was purchased from Sigma (St. Louis, (DuPont NEN, Boston, MA), following manufacturer’s protocol. The data MO). Human IL-5, a gift from Schering-Plough Research Institute (Ken- were expressed as pg of PAF produced/106 cells. The detection limit of the ilworth, NJ), was diluted in PBS containing 0.1% BSA at 100 ␮g/ml, and assay was 100 pg PAF produced/106 cells. stored at Ϫ70°C. All agonists and antagonists were diluted in reaction medium immediately before the experiments. Neither ethanol nor DMSO Eosinophil degranulation altered eosinophil functions at the solvent concentrations used in this study Ͻ Eosinophil degranulation was performed in 96-well flat-bottom tissue cul- ( 0.05%). Purified human serum IgG was purchased from Organon ture plates, as described previously (21, 22). Eosinophils were washed and Teknika-Cappel (Malvern, PA) and stored at 1 mg/ml in PBS at 4°C. For suspended in RPMI 1640 medium (Celox, Hopkins, MN), supplemented eosinophil activation, IgG was immobilized onto the wells of tissue culture with 10 mM HEPES and 0.02% HSA, at 5 ϫ 105 cells/ml. Serial dilutions plates, as described below. The other reagents, including catalase, super- of CV6209 (50 ␮l) were added to the wells, followed by 100 ␮l of eosin- oxide dismutase, human serum albumin (HSA), taurine, L-serine, glutathi- ophil suspension. Cells were stimulated with 50 ␮l of soluble agonists, and one, and cytochrome c, were purchased from Sigma. incubated for 180 min at 37°C and 5% CO2. Eosinophils incubated in the Eosinophil isolation wells coated with IgG received 50 ␮l of medium alone. After incubation, supernatants were collected and stored at Ϫ20°C until they were assayed Eosinophil isolation was performed by a magnetic cell separation system for eosinophil-derived neurotoxin (EDN). To quantitate eosinophil degran- (MACS; Becton Dickinson, San Jose, CA), as described previously, with ulation, the concentration of EDN in the sample supernatants was measured minor modifications (20). Briefly, venous blood anticoagulated with 50 by RIA. The RIA is a double-Ab competition assay, in which radioiodi- U/ml heparin was obtained from normal volunteers and diluted with PBS nated EDN, rabbit anti-EDN, and burro anti-rabbit IgG are used, as re- at a 1:1 ratio. Diluted blood was overlaid on isotonic Percoll solution (den- ported elsewhere (5). Total cellular EDN contents were measured simul- sity, 1.085 g/ml; Sigma) and centrifuged at 1000 ϫ g for 30 min at 4°C. taneously using supernatants from cells lysed with 0.5% Nonidet P-40 The supernatant and mononuclear cells at the interface were carefully re- detergent. All assays were conducted in duplicate. moved, and erythrocytes in sediment were lysed by two cycles of hypo- LTC production tonic water lysis. Isolated granulocytes were washed twice with PIPES 4 Downloaded from buffer (25 mM PIPES, 50 mM NaCl, 5 mM KCl, 25 mM NaOH, 5.4 mM Eosinophil production of LTC4 was performed in 96-well tissue culture glucose, pH 7.4) with 1% defined calf serum (HyClone Laboratories, Lo- plates, as described previously, with minor modifications (24). Eosinophils gan, UT). An approximately equal volume of anti-CD16-conjugated mag- were washed with HBSS supplemented with 10 mM HEPES, 20 mM L- netic beads (Miltenyi Biotec, Auburn, CA) was added to the cell pellet. serine, and 5 mM glutathione, and resuspended in the same medium at 1 ϫ After 60 min of incubation on ice, cells were loaded onto the separation 106 cells/ml. Serial dilutions of CV6209 or medium (50 ␮l) were added to column positioned in the strong magnetic field of the MACS. Cells were the wells of the tissue culture plate, followed by 100 ␮l of cell suspension. eluted with PIPES buffer containing defined calf serum. The purity of eo- ␮

Cells were stimulated by adding 50 l of soluble agonists. Eosinophils http://www.jimmunol.org/ sinophils counted by Randolph’s stain was regularly greater than 98%. incubated in the wells coated with IgG received 50 ␮l of medium alone. Purified eosinophils were washed and suspended in reaction medium, then After incubation for1hat37°C and 5% CO2, the supernatants from each used immediately. well were collected and frozen at Ϫ70°C or assayed immediately. Con- Superoxide anion production centrations of LTC4 in the sample supernatants were measured by ELISA using an LTC4 kit (Cayman Chemical, Ann Arbor, MI), following the Eosinophil superoxide production was induced by various stimuli in poly- procedure recommended by the manufacturer. The sensitivity of the assay styrene 96-well flat-bottom tissue culture plates (Corning CoStar, Cam- was 7.8 pg/ml. All experiments were conducted in duplicate. bridge, MA) and measured by reduction of cytochrome c, as previously described (21, 22). To immobilize human IgG onto the wells of tissue Lipid body formation ␮ ␮ culture plates, 50 l of human IgG diluted in PBS at 50 g/ml was added Formation of lipid bodies within the activated eosinophils was examined as to the wells and incubated overnight at 4°C. The solution was aspirated, by guest on October 3, 2021 ␮ described previously (25), with minor modifications. Briefly, wells of a and the wells were blocked with 50 l of 2.5% HSA in PBS for2hat Lab-Tek 16-well chamber slide (Nunc, Naperville, IL) were coated with or 37°C. After incubation, wells were washed twice with 0.9% NaCl and used without IgG and blocked with 2.5% HSA, as described above. Eosinophils immediately for the experiments. The tissue culture wells used for eosin- were washed in HBSS with 10 mM HEPES, and resuspended in the same ophil stimulation with soluble agonists were blocked with HSA without medium at 1 ϫ 106 cells/ml. Serial dilutions of CV6209 or medium (50 ␮l) IgG coating. Freshly isolated eosinophils were washed, and resuspended in ␮ ␮ ϫ 5 were added to the wells, followed by 100 l of cell suspension. Cells were HBSS with 10 mM HEPES and 200 M cytochrome c at 5 10 cells/ml. stimulated by adding 50 ␮l of soluble agonists. Eosinophils incubated in Fifty microliters of serial dilutions of inhibitors were added to the wells. the wells coated with IgG received 50 ␮l of medium alone. After incuba- One hundred microliters of cell suspension were dispensed onto the wells, tion for 60 min at 37°C and 5% CO2, the supernatants were removed by and the reactions were initiated by adding 50 ␮l of soluble stimuli, includ- 2ϩ 2ϩ ␮ aspiration. Eosinophils were fixed in 3.7% formaldehyde in Ca /Mg - ing IL-5, PMA, and PAF at 25 ng/ml, 1 ng/ml, and 0.3 M final concen- free HBSS, pH 7.4, and rinsed in 0.1 M cacodylate buffer, pH 7.4. Cells trations, respectively. Wells coated with immobilized Ig received 50 ␮lof were stained in 1.5% OsO4 (30 min), rinsed in water, immersed in 1.0% medium alone. In some experiments, cells were pretreated with an optimal thiocarbohydrazide (5 min), rinsed in 0.1 M cacodylate buffer, restained in concentration of PTX (100 ng/ml) in 15-ml conical tubes for2hat37°C 1.5% OsO4 (3 min), rinsed in water, and dried. Subsequently, chamber before addition to the wells. Immediately after addition of stimuli, the walls were removed and the slide was coverslipped. The numbers of lipid reaction wells were measured for absorbance at 550 nm in a microplate bodies were counted with phase-contrast microscope by consecutively autoreader (Thermomax, Molecular Devices, Menlo Park, CA), followed scanning 50 eosinophils with ϫ630 magnification. by repeated readings. Between absorbance measurements, the plate was incubated at 37°C. Superoxide anion generation was calculated with an Statistical analysis extinction coefficient of 21.1 ϫ 103 cmϪ1MϪ1 for reduced cytochrome c at 550 nm, and was expressed as nanomoles of superoxide production/105 Data are presented as means Ϯ SEM from the numbers of experiments cells. indicated. Statistical significance of the differences between various treatment groups (i.e., with or without inhibitor) was assessed using the Mann- PAF production Whitney U test or paired Student’s t test. Generation of PAF by eosinophils was performed in 96-well flat-bottom tissue culture plates, as described previously by van der Bruggen (23), with Results minor modifications. All incubations for PAF production were performed Effects of PTX on eosinophil superoxide production in enriched HBSS (HBSS supplemented with 1 mM CaCl , 5 mM glucose, 2 Previously, eosinophil effector functions, such as superoxide pro- and 0.5% (w/v) HSA and containing superoxide radical scavengers (2.5 mM taurine, 5000 U/ml catalase, and 380 U/ml superoxide dismutase)) to duction and degranulation, were shown to be induced by a wide prevent lipid degradation. Freshly isolated eosinophils were suspended in range of stimuli, such as cytokines (e.g., IL-5), lipid mediators enriched HBSS at 2 ϫ 106 cells/ml, and 100-␮l aliquots were added to the (e.g., PAF), and immobilized Ig (e.g., IgG) (8–10, 21). Consistent wells of tissue culture plates prepared as described above. Cells were stim- with these previous findings, Fig. 1 shows that PAF, human IgG ulated with 100 ␮l of IL-5, PMA, or medium alone. After incubation for 15 or 45 min at 37°C, the plates were briefly centrifuged at 400 ϫ g. Super- immobilized onto tissue culture plates and IL-5, as well as a pos- natants were collected and stored under nitrogen at Ϫ70°C. The concen- itive control, PMA, induced eosinophil superoxide production in a tration of PAF in the samples was determined by a commercial RIA kit time-dependent manner. As reviewed elsewhere, PAF is known to 2984 ENDOGENOUS PAF AND EOSINOPHIL ACTIVATION

FIGURE 1. Kinetics of agonist-induced eosinophil superoxide produc- FIGURE 3. Kinetics of agonist-induced eosinophil superoxide production in the presence and absence of PTX. Eosinophils were preincubated tion in the presence and absence of CV6209. Eosinophils were stimulated ␮ ␮ with (closed circles) or without (open circles) PTX for 2 h and stimulated with PAF (0.3 M; A), immobilized IgG (wells precoated with 50 g/ml with 0.3 ␮M PAF, immobilized IgG (wells precoated with 50 ␮g/ml hu- human IgG; B), soluble IL-5 (25 ng/ml; C), or PMA (1 ng/ml; D)inthe man IgG), 25 ng/ml IL-5, or 1 ng/ml PMA. Kinetics of superoxide pro- presence or absence of serial dilutions of a competitive PAF receptor an- Downloaded from duction was measured by reduction of cytochrome c, as described in Ma- tagonist, CV6209. The concentrations of CV6209 were 0 (squares), 0.1 ␮ terials and Methods. One representative experiment from a total of four (diamonds), 0.3 (circles), or 1 M (triangles). Kinetics of superoxide pro- experiments showing similar results is presented. duction was measured by reduction of cytochrome c, as described in Ma- terials and Methods. One representative experiment is shown; similar results were observed in ﬁve experiments. mediate its biological effects through activation of a G protein- coupled seven-transmembrane receptor (13). In addition, PTX cat- most likely provoked by generation of intermediate metabolite(s) http://www.jimmunol.org/ alyzes the ADP ribosylation of certain G protein ␣-subunits, and active on G proteins, rather than as a direct consequence of Fc␥R treatment of intact cells with PTX results in uncoupling of PTX- or IL-5R perturbation. sensitive G proteins from cell surface receptors (26). As shown in Fig. 1A, when cells were pretreated with PTX, eosinophil super- Production of PAF by eosinophils stimulated with IgG or IL-5 oxide production stimulated with PAF was abolished, suggesting Then, what is the intermediate metabolite(s) that stimulates eosin- that eosinophils’ PAF receptor is coupled to PTX-sensitive G pro- ophil functions in a G protein-dependent manner? Eosinophils are

tein(s). Surprisingly, as shown in Fig. 1, B and C, superoxide pro- able to generate several lipid mediators, such as PAF and LTC4,in duction induced by IgG immobilized onto tissue culture plates or

response to physiologic and pharmacologic stimuli (3, 12). Eosin- by guest on October 3, 2021 soluble IL-5 was markedly inhibited by PTX pretreatment. In con- ophils also contain fourfold higher levels of ether phospholipid, the trast, superoxide production induced by PMA (Fig. 2D) was not stored precursor of PAF, than do neutrophils, suggesting that the affected by pretreatment of cells with PTX. The receptors for IgG eosinophil is a good PAF producer (29). As shown in Fig. 2, we and IL-5 belong to the families of Fc␥R (reviewed in Ref. 27) and found that PAF was detectable in eosinophil supernatants 15 min hemopoietin receptor (reviewed in Ref. 28), respectively; there is after stimulation with immobilized IgG or soluble PMA; the no evidence for coupling of these receptors to PTX-sensitive G amounts of released PAF increased dramatically by 45 min. proteins. Furthermore, they are structurally distinct from typical G Smaller but signiﬁcant amounts of PAF (530.5 Ϯ 127.8 pg PAF/ protein-coupled seven-transmembrane receptors. Therefore, in eo- 106 cells) were also detectable in supernatants from eosinophils sinophils, the superoxide production stimulated by IgG or IL-5 is stimulated with soluble IL-5 for 45 min ( p Ͻ 0.05 compared with no stimulus, n ϭ 4). The kinetics and amounts of PAF released by eosinophils incubated with IgG immobilized onto the tissue culture plates (2908.8 Ϯ 428.3 pg PAF/106 cells) were comparable with those observed previously by eosinophils incubated with IgG immobilized to Sepharose beads (12). Eosinophils incubated in medium alone released minimal levels of PAF at both times. Thus, when stimulated with immobilized IgG or soluble IL-5, eosinophils produce and release PAF in less than 45 min.

Effects of PAF antagonists on eosinophil effector functions If PAF produced by activated eosinophils played major roles in effector functions of eosinophils themselves, then prevention of PAF binding to its receptor would dampen eosinophil effector functions. To examine this hypothesis, we used a potent and highly selective phospholipid analogue antagonist of PAF, CV6209; FIGURE 2. PAF release from eosinophils stimulated by IgG, IL-5, or CV6209 is a competitive antagonist of PAF receptor (30). In our PMA. Eosinophils were incubated with medium alone, immobilized IgG (wells precoated with 50 ␮g/ml human IgG), soluble IL-5 (25 ng/ml), or preliminary study, titration of CV6209 in an eosinophil superoxide PMA (1 ng/ml) for 15 or 45 min. PAF released into the supernatants was production assay using a series of concentrations of PAF as stimuli measured by RIA. Results are presented as means Ϯ SEM of three exper- showed that 50% inhibiting concentrations (IC50) of CV6209 for ,Signiﬁcant difference (p Ͻ 0.05) from values in the absence of 0.1, 0.3, 1, and 3 ␮M of PAF were 0.08, 0.23, 0.50, and 0.95 ␮M ,ء .iments stimuli. respectively. As shown in Fig. 3A, CV6209 inhibited eosinophil The Journal of Immunology 2985

FIGURE 5. Effects of structurally distinct PAF receptor antagonists on eosinophil superoxide production induced by IL-5. Eosinophils were in- FIGURE 4. Inhibition of eosinophil superoxide production by CV6209. cubated with 25 ng/ml IL-5 in the presence of serial dilutions of PAF Eosinophils were incubated with stimuli in the presence or absence of receptor antagonists, hexanolamine PAF, or Y-24180. Superoxide produc- serial dilutions of CV6209, as described in the legend for Fig. 3. Super- tion was measured by the reduction of cytochrome c at 1.5 h. Results are Significant ,ءء and ء .oxide production was measured by the reduction of cytochrome c at3h. presented as means Ϯ SEM of four experiments Ϯ p Ͻ 0.01) from values obtained with eosin- Downloaded from ,ءء ;p Ͻ 0.05 ,ء) Results are presented as means SEM of five experiments (IgG, PMA, differences and PAF) or three experiments (IL-5). In the absence of stimuli, superoxide ophils incubated without PAF antagonists. -Significant differ ,ء .(production was not detected (Ͻ0.04 nmol/105 cells ences (p Ͻ 0.05) from values obtained with eosinophils incubated without CV6209. lamine PAF and 10 ␮M Y-24180 completely inhibited the eo-

sinophil response to IL-5. http://www.jimmunol.org/ ␮ superoxide production stimulated with 0.3 M exogenous PAF in As described earlier, PAF is one of the strongest secretagogues a concentration-dependent manner. Interestingly, as shown in Fig. for eosinophils (9). To investigate the role of endogenous PAF in 3, B and C, eosinophil superoxide production induced by immo- eosinophil degranulation, puriﬁed cells were stimulated with im- bilized IgG or soluble IL-5 was also inhibited by CV6209, while mobilized IgG or soluble IL-5 and the effects of CV6209 were no exogenous PAF was added to the system. In contrast, super- examined. As shown in Fig. 6, both immobilized IgG and soluble oxide production by eosinophils stimulated with PMA (Fig. 3D) IL-5 induced eosinophil EDN release. CV6209, at 1 ␮M, signiﬁ- was not inhibited by CV6209, with the exception of a slight change cantly inhibited degranulation of eosinophils stimulated with im- in kinetics. mobilized IgG and IL-5 by 49% ( p Ͻ 0.0001, n ϭ 6) and 35%

The experiments were repeated using eosinophils from different ( p Ͻ 0.05, n ϭ 6), respectively. In the absence of stimuli, CV6209 by guest on October 3, 2021 donors, and the concentration-response curves of CV6209 are showed no effect on spontaneous release of EDN (Fig. 6), and it summarized in Fig. 4. Stimulation of eosinophils with PAF, im- did not inhibit PMA-induced degranulation of eosinophils (data mobilized IgG, IL-5, or PMA consistently induced substantial not shown). All in all, these findings indicate that endogenous PAF amounts of superoxide production. Unstimulated eosinophils did is involved in effector functions of eosinophils stimulated with Ͻ 5 not produce detectable levels of superoxide ( 0.04 nmol/10 immobilized IgG or soluble IL-5. However, this role of endoge- cells). When a PAF receptor antagonist, CV6209, was added to the nous PAF may be different among the functions; for example, incubation, the eosinophil responses to immobilized IgG and IL-5 superoxide production is likely to be more dependent on endoge- as well as those to exogenous PAF were inhibited in a concentra- nous PAF than is degranulation. tion-dependent manner; IC50 values for immobilized IgG, IL-5, or PAF were 0.8, 0.3, and 0.3 ␮M, respectively. Notably, CV6209 at 1 ␮M abolished the eosinophil response to IL-5 and significantly inhibited the response to immobilized IgG by 62% ( p Ͻ 0.01, n ϭ 5). In contrast, superoxide production induced by PMA was not affected by CV6209 at concentrations up to 1 ␮M. CV6209 showed no effect on eosinophil superoxide production in the absence of any stimuli (data not shown). Thus, eosinophil superoxide production induced by immobilized IgG, soluble IL-5, and exogenous PAF is sensitive to the PAF receptor antagonist, suggesting that endogenous PAF is involved in eosinophil responses to immobilized IgG or soluble IL-5. To confirm the observation described above, we examined the effects of other PAF receptor antagonists, structurally distinct from CV6209, on IL-5-induced superoxide production. Hexanolamine PAF is a structural analogue of PAF; therefore, it has an advantage FIGURE 6. Effects of CV6209 on eosinophil degranulation. Eosino- over other PAF receptor antagonists in that it may gain access phils were incubated with medium alone, immobilized IgG (wells pre- to otherwise unavailable action sites (31). Y-24180, a newly coated with 50 ␮g/ml), or soluble IL-5 (25 ng/ml) in the presence of in- discovered etizolam analogue, is a potent and specific PAF re- dicated concentrations of CV6209 for 3 h. The amounts of EDN released ceptor antagonist (32). As shown in Fig. 5, both hexanolamine in the supernatants were measured by RIA. Results are presented as (Significantly different (p Ͻ 0.05 ,ء .PAF and Y-24180 inhibited IL-5-induced superoxide produc- means Ϯ SEM of six experiments tion in a concentration-dependent manner; 30 ␮M hexano- from values obtained with eosinophils incubated without CV6209. 2986 ENDOGENOUS PAF AND EOSINOPHIL ACTIVATION

FIGURE 7. Effects of CV6029 on lipid body formation and LTC4 release by eosinophils stimulated with IL-5, immobilized IgG, or PAF. Eosinophils were incubated with medium alone, immobilized IgG (wells precoated with 50 ␮g/ml), soluble IL-5 (25 ng/ml), or PAF (1 ␮M) in the presence of indicated concentrations of CV6209 for 1 h. A, The numbers of lipid bodies were determined as described in Materials and Methods. Data are presented as means Ϯ signiﬁcant differences ,ء ;SEM from three independent experiments. †, Signiﬁcant differences (p Ͻ 0.05) from values obtained in the absence of stimuli

Ͻ Downloaded from (p 0.05) from values obtained with eosinophils incubated without CV6209 in the same stimulus group. B, The amounts of LTC4 released into the supernatants were measured by ELISA. Data are normalized to the values in the absence of CV6029 as 100% and presented as means Ϯ SEM from ﬁve .p Ͻ 0.01) from values obtained with eosinophils incubated without CV6209 ,ءء ;p Ͻ 0.05 ,ء) Signiﬁcant differences ,ءء and ء .independent experiments

Roles of endogenous PAF in eicosanoid metabolism of lipid bodies in eosinophils incubated with medium alone. Thus, eosinophils endogenous PAF, similarly to exogenous PAF, plays major roles http://www.jimmunol.org/ Recent studies suggest that intracellular lipid bodies are distinct, in eicosanoid metabolism of eosinophils. inducible, nonnuclear sites for eicosanoid synthesis in granulo- Blocking of phospholipase A2 inhibits PAF production and cytes (33, 34). Lipid body formation in eosinophils is rapidly in- eosinophil function duced by PAF stimulation, and PAF-induced lipid body production correlates strongly with increased production of eicosanoids (33). To confirm the involvement of the endogenous PAF pathway in Because endogenous PAF was involved in eosinophil functions eosinophil activation, we examined the effects of the blockade of stimulated with IL-5 or immobilized IgG, we examined whether PAF generation on eosinophil function. Phospholipase A2 (PLA2), these agonists may also stimulate lipid body formation and eico- the enzyme that catalyzes glycerol phospholipid to yield lysophos- sanoid generation in eosinophils. As shown in Fig. 7A, immobi- phatide and arachidonic acid, is one of the key enzymes involved by guest on October 3, 2021 lized IgG and soluble IL-5 increased the numbers of intracellular in PAF generation (reviewed in Ref. 35). Mepacrine competitively lipid bodies within 1 h, similarly to the effect of exogenous PAF. inhibits PLA2 activity by forming a stable complex of drug and Two and one-half times more lipid bodies were found in eosino- phospholipid substrate (36). As expected from its drug action, 100 ␮ phils stimulated with immobilized IgG, IL-5, or PAF, than in un- M mepacrine inhibited PAF release from eosinophils stimulated stimulated eosinophils. Increased numbers of lipid bodies were with IL-5 or immobilized IgG (Table II). As shown in Fig. 8, also accompanied by increased production of eicosanoids, as mepacrine also significantly inhibited eosinophil superoxide pro- Ͻ Ͻ shown by LTC release into the supernatants of activated eosino- duction induced by IL-5 or immobilized IgG ( p 0.05 and p 4 ␮ ϭ phils (Table I). As shown in Fig. 7, A and B, the PAF antagonist, 0.01 at 30 and 100 M, respectively, n 4). In contrast, PMA- induced superoxide production was not affected by up to 100 ␮M CV6209, blocked lipid body formation and LTC4 production induced by exogenous PAF. Similarly, CV6209, in concentrations as mepacrine. Thus, PLA2 is most likely involved in PAF generation low as 0.3 ␮M, significantly inhibited lipid body formation and by eosinophils stimulated with immobilized IgG or soluble IL-5, and blockade of this enzyme inhibits superoxide production re- LTC4 production by eosinophils stimulated with IL-5 or immobilized IgG ( p Ͻ 0.05, n ϭ 3 and 5, respectively). CV6209, at 1 ␮M, sponse to these stimuli.

inhibited lipid body formation by more than 70% and LTC4 production by more than 75%; CV6209 did not affect the number of Table II. Effects of mepacrine on PAF production by eosinophilsa

PAF Release (pg/106 cells) Table I. LTC4 production by eosinophils stimulated with various a agonists Stimuli Without mepacrine With mepacrine

Stimuli Concentration Released LTC (ng/106 cells) Expt. 1 4 None 10.6 15.6 None 14.3 Ϯ 6.9 IL-5 400.9 26.7 Immobilized IgG 843.5 Ϯ 301.4b IgG 922.6 43.4 IL-5 25 ng/ml 154.2 Ϯ 37.8b Expt. 2 PAF 1.0 ␮M 625.1 Ϯ 303.6b None 11.0 ND IL-5 112.7 42.8 a Eosinophils were incubated with the agonists as described above for1hat37°C IgG 56.0 12.1 and 5% CO2. Concentrations of LTC4 in the sample supernatants were measured by Ϯ a ELISA using an LTC4 kit. Data are presented as means SEM from ﬁve independent Eosinophils were incubated with medium alone, 25 ng/ml IL-5, or human IgG experiments. immobilized onto tissue culture plates in the absence or presence of 100 ␮M mepa- b Ͻ Denotes signiﬁcant difference (p 0.05) compared to the cells incubated with- crine for 45 min at 37°C and 5% CO2. Concentrations of PAF in the sample super- out stimuli. natants were measured by RIA using a PAF RIA kit. ND, not determined. The Journal of Immunology 2987

IgG-coated Sepharose beads stimulate eosinophils to produce PAF (12). We also found that eosinophils incubated with soluble IL-5 produced PAF (Fig. 2). Furthermore, PAF provokes various effector functions of eosinophils, including adhesion, degranulation, and generation of superoxide anion and arachidonic acid metabolites (8, 17–19). Altogether, the results suggest that endogenous PAF may play a role in inflammatory responses of eosinophils by acting as a central switch, which bridges activation signals from various cell surface receptors with various effector cell functions. The dis- ruption of the endogenous PAF pathway, either by inhibiting production or by preventing binding to a specific receptor, may pro- foundly impact eosinophil responses to a variety of stimuli. FIGURE 8. Effects of a PLA2 inhibitor, mepacrine, on superoxide production by eosinophils. Eosinophils were incubated with immobilized IgG The remaining question is how endogenous PAF stimulates ef- (squares, wells precoated with 50 ␮g/ml IgG), soluble IL-5 (circles, 25 fector functions of eosinophils. As shown in Fig. 2, we were able ng/ml), or PMA (triangles, 1 ng/ml) in the presence of serial dilutions of to detect PAF released extracellularly in the supernatants of eo- mepacrine. Superoxide production was measured by reduction of cyto- sinophils. Therefore, PAF released into the outside medium could chrome c at 1.5 h. Data were normalized to the values in the absence of stimulate eosinophils in a paracrine manner. However, a caveat for mepacrine. The amounts of superoxide produced by eosinophils stimulated this speculation is whether the concentration of PAF in the extra- with immobilized IgG, IL-5, and PMA without mepacrine were 6.1 Ϯ 0.9, 5 cellular space is high enough to induce eosinophil effector func- 4.2 Ϯ 0.2, and 22.5 Ϯ 1 nmol/10 cells, respectively (means Ϯ SEM, n ϭ Downloaded from tions. For example, if released PAF is uniformly distributed in the ,ءء and ء .Results are presented as means Ϯ SEM of four experiments .(4 p Ͻ 0.01) from values obtained incubation medium, we estimated 5 nM PAF in the eosinophil ,ءء ;p Ͻ 0.05 ,ء) Significant differences with eosinophils incubated without mepacrine. supernatants stimulated with immobilized IgG. This concentration may not be sufficient to fully activate eosinophils; it usually takes 100 nM or higher concentrations of exogenous PAF to stimulate Discussion eosinophil functions (22). However, much higher concentrations Activation of eosinophils and their subsequent release of inflam- can be expected in the microenvironment in close proximity to the http://www.jimmunol.org/ matory mediators are implicated in the pathophysiology of various cells. Therefore, another potential mechanism is that PAF gener- allergic and eosinophilic disorders (reviewed in Ref. 1). Although ated by eosinophils may remain within and activate the cells little is known regarding the mechanisms of eosinophil activation internally, similar to intracellular second messengers. Indeed, in in human diseases, previous studies suggest that the eosinophil is eosinophils stimulated with IgG-coated Sepharose beads, the activated in vitro by a wide range of stimuli, such as cytokines, kinetic study showed that intracellular concentrations of PAF lipid mediators, and immobilized Ig (8–10, 21). Our results indi- increased quickly after the stimulation, reached a plateau by 15 cate that endogenously generated PAF plays important roles in the min, and stayed at the similar levels for at least 60 min (12). In

functional activation of eosinophils stimulated with IL-5 or immo- addition, Ojima-Uchiyama et al. reported that more intracellular by guest on October 3, 2021 bilized IgG. This conclusion is based on several observations. 1) PAF remains than is released extracellularly (38). Furthermore, Although Fc␥R and IL-5R are not coupled to G proteins, eosino- the presence of intracellular PAF receptors has been suggested phil functional responses to IgG or IL-5 were inhibited by PTX, in nerve cells and granulocytes (39–41). Because intracellular suggesting that they are mediated by intermediate metabolite(s) PAF can be used immediately without involvement of secretory active on PTX-sensitive G proteins, rather than being a direct con- processes or without potential exposure to extracellular cata- sequence of Fc␥R or IL-5R ligation. 2) PAF itself was detected in lytic enzymes, this PAF hypothesis provides an economical cel- the supernatants of activated eosinophils. 3) Blockade of PAF in- lular strategy. Therefore, the traditional plasma membrane re- teraction with its receptor by three PAF receptor antagonists mark- ceptors may be primarily responsible for mediating paracrine edly inhibited superoxide production, degranulation, lipid body effects of PAF, while intracellular receptors may mediate the formation, and LTC4 release by eosinophils stimulated with IL-5 autocrine effects of PAF. or immobilized IgG. 4) The inhibitory effects of PAF antagonists Close examination of the kinetics of superoxide production pro- are unlikely due to nonspecific cytotoxicity because the IC50 and vides some insights regarding the roles of endogenous PAF in the slope of concentration-response curves of CV6209 for IL-5 or eosinophil functions. As shown in Fig. 3B, at early time points immobilized IgG were comparable with that for exogenous PAF (less than 30 min), IgG-induced superoxide production was min- (Fig. 4), and because each of three structurally distinct PAF an- imally affected by pretreatment of cells with a PAF receptor antagonists effectively inhibited IL-5 response (Figs. 4 and 5). In tagonist, CV6209. However, the inhibitory effect of CV6209 was addition, the eosinophil responses to PMA were not inhibited by pronounced when eosinophils were producing superoxide at these antagonists, indicating that they do not inhibit the enzymes later time points (i.e., 60 min). In contrast, the effects of exog- for cellular functions directly. 5) Finally, inhibition of PAF gen- enous PAF were completely inhibited by 1 ␮M CV6209 from eration by a PLA2 inhibitor attenuated eosinophil functional re- the beginning of the cellular response (Fig. 3A). The effect of sponses to IL-5 or IgG. PTX on PAF- and IgG-induced superoxide production also Our findings are compatible with a previous report, which dem- showed similar findings (Fig. 1, A and B). These delayed onsets onstrated an important role of PLA2 in eosinophil functions stimu- for the effects of a PAF receptor antagonist or PTX suggest that, lated by FMLP plus cytochalasin B (37). While these investigators at early time points, small amounts of superoxide may be gen- did not examine how PLA2 is involved in eosinophil functions, our erated by eosinophils through alternative mechanisms (e.g., sig- study suggests that PLA2 was involved in the previous study by nals from IgG Fc receptors) without the effects of endogenous inducing the production of endogenous PAF. A number of agonists PAF. Thus, at later time points, the dependency on endogenous are known to induce production of PAF by eosinophils. For ex- PAF may become greater when cells are vigorously producing ample, chemotactic factors, such as FMLP and C5a, stimulate superoxide. Therefore, the role of endogenous PAF may be to rapid (1–3 min) release of PAF from eosinophils (3). In addition, enhance the otherwise small responses of eosinophils triggered 2988 ENDOGENOUS PAF AND EOSINOPHIL ACTIVATION by cytokines or Ig, and to expand to full bloom the effector 9. Horie, S., G. J. Gleich, and H. Kita. 1996. Cytokines directly induce degranula- function of eosinophils. tion and superoxide production from human eosinophils. J. Allergy Clin. Immu- nol. 98:371. Eosinophil activation and subsequent release of inflammatory 10. Kernen, P., M. P. Wymann, V. von Tscharner, D. A. Deranleau, P.-C. Tai, mediators are implicated in the pathophysiology of allergic dis- C. J. Spry, C. A. Dahinden, and M. Baggiolini. 1991. Shape changes, exocytosis, eases, such as bronchial asthma. Our study suggests that endoge- and cytosolic free calcium changes in stimulated human eosinophils. J. Clin. nous PAF plays an important role in mediator release by eosino- Invest. 87:2012. 11. Prescott, S. M., G. A. Zimmerman, and T. M. McIntyre. 1990. Platelet-activating phils. Furthermore, in addition to its autocrine effects, exogenous factor. J. Biol. Chem. 265:17381. PAF exerts various biological activities relevant to bronchial 12. Cromwell, O., A. J. Wardlaw, A. Champion, R. Moqbel, D. Osei, and A. B. Kay. asthma, such as airway constriction, development of bronchial hy- 1990. IgG-dependent generation of platelet-activating factor by normal and low perresponsiveness, and induction of microvascular leakage and density human eosinophils. J. Immunol. 145:3862. 13. Izumi, T., and T. Shimizu. 1995. Platelet-activating factor receptor: gene expres- edema (reviewed in Refs. 14 and 15). Indeed, severe asthma has sion and signal transduction. Biochim. Biophys. Acta 1259:317. been positively correlated with a deficiency of PAF acetylhydro- 14. Hanahan, D. J. 1986. Platelet activating factor: a biologically active phospho- lase in a Japanese population (42). Despite the evidence for po- glyceride. Annu. Rev. Biochem. 55:483. tential roles of PAF in bronchial asthma, previous studies with 15. Barnes, P. J. 1991. Platelet activating factor and asthma. Ann. NY Acad. Sci. PAF receptor antagonists failed to show beneficial effects in pa- 629:193. 16. Wardlaw, A. J., R. Moqbel, O. Cromwell, and A. B. Kay. 1986. Platelet-activat- tients with asthma (43, 44). However, studies using newly devel- ing factor: a potent chemotactic and chemokinetic factor for human eosinophils. oped and more potent PAF receptor antagonists have shown the J. Clin. Invest. 78:1701. effects of these antagonists on bronchial asthma. For example, 17. Kroegel, C., T. Yukawa, G. Dent, P. Venge, K. F. Chung, and P. J. Barnes. 1989. Y-24180, which is also used in this study, inhibited allergen-in- Stimulation of degranulation from human eosinophils by platelet-activating factor. J. Immunol. 142:3518. Downloaded from duced eosinophilia and physiologic changes in guinea pig lungs 18. Zoratti, E. M., J. B. Sedgwick, R. R. Vrtis, and W. W. Busse. 1991. The effect of (45), and prevented bronchial hyperresponsiveness in patients with platelet-activating factor on the generation of superoxide anion in human eosin- asthma (46). Therefore, further studies are needed to elucidate the ophils and neutrophils. J. Allergy Clin. Immunol. 88:749. roles of PAF in bronchial asthma and other allergic diseases. Es- 19. Bruijnzeel, P. L. B., P. T. M. Kok, M. L. Hamelink, A. M. Kijne, and J. Verhagen. 1987. Platelet-activating factor induces leukotriene C4 synthesis by purified hu- pecially, as shown in this study, PAF antagonists inhibited eosin- man eosinophils. Prostaglandins 34:205. ophil functions induced by a range of physiologic agonists. This 20. Ide, M., D. Weiler, H. Kita, and G. J. Gleich. 1994. Ammonium chloride expo- http://www.jimmunol.org/ wide spectrum of PAF antagonist effects may be advantageous sure inhibits cytokine-mediated eosinophil survival. J. Immunol. Methods 168: from a therapeutic point of view because a number of eosinophil- 187. 21. Kaneko, M., S. Horie, M. Kato, G. J. Gleich, and H. Kita. 1995. A crucial role active mediators produced by various immunologic or interstitial ␤ for 2 integrin in the activation of eosinophils stimulated by IgG. J. Immunol. cells are likely to be involved in the pathophysiology of allergic 155:2631. diseases. The inhibition of only one mediator may be insufficient to 22. Horie, S., and H. Kita. 1994. CD11b/CD18 (Mac-1) is required for degranulation inhibit eosinophil functions in such circumstances. Our study sug- of human eosinophils induced by human recombinant granulocyte-macrophage colony-stimulating factor and platelet-activating factor. J. Immunol. 152:5457. gests that eosinophil activation and functions can be controlled 23. van der Bruggen, T., P. T. M. Kok, J. A. M. Raaijmakers, J. W. J. Lammers, and effectively by managing the endogenous PAF pathways, and this L. Koenderman. 1994. Cooperation between Fc␥ receptor II and complement

concept may open a new avenue for the treatment of patients with receptor type 3 during activation of platelet-activating factor release by cytokine- by guest on October 3, 2021 allergic or eosinophilic disorders. primed human eosinophils. J. Immunol. 153:2729. 24. Kita, H., R. I. Abu-Ghazaleh, S. Sur, and G. J. Gleich. 1995. Eosinophil major basic protein induces degranulation and IL-8 production by human eosinophils. J. Immunol. 154:4749. 25. Weller, P. F., S. J. Ackerman, A. Nicholson-Weller, and A. M. Dvorak. 1989. Cy- Acknowledgments toplasmic lipid bodies of human neutrophilic leukocytes. Am. J. Pathol. 135:947. We thank Ms. Cheryl Adolphson for editorial assistance and Ms. Linda 26. Moss, J., and M. Vaughan. 1988. ADP-ribosylation of guanyl nucleotide-binding H. Arneson for secretarial assistance. regulatory proteins by bacterial toxins. Adv. Enzymol. 61:303. 27. Dae¨ron, M. 1997. Fc receptor biology. Annu. Rev. Immunol. 15:203. 28. Devos, R., G. Plaetinck, S. Cornelis, Y. Guisez, J. Van der Heyden, and J. Tavernier. 1995. Interleukin-5 and its receptor: a drug target for eosinophilia References associated with chronic allergic disease. J. Leukocyte Biol. 57:813. 29. Sugiura, T., K. Mabuchi, A. Ojima-Uchiyama, Y. Masuzawa, N. N. Cheng, 1. Butterfield, J. H., K. L. Leiferman, and G. J. Gleich. 1995. Eosinophil-associated T. Fukuda, S. Makino, and K. Waku. 1992. Synthesis and action of PAF in diseases. In Samter’s Immunologic Diseases, Vol. 1. M. M. Frank, K. F. Austen, human eosinophils. J. Lipid Mediat. 5:151. H. N. Claman, and E. R. Unanue, eds. Little, Brown and Company, Boston, pp. 30. Terashita, Z. I., Y. Imura, M. Takatani, S. Tsushima, and K. Nishikawa. 1987. 501–527. CV-6209, a highly potent antagonist of platelet activating factor in vitro and in 2. Kita, H., C. Adolphson, and G. J. Gleich. Biology of eosinophils. 1998. In Al- vivo. J. Pharmacol. Exp. Ther. 242:263. lergy: Principles and Practice, 5th Ed. E. Middleton, Jr., C. E. Reed, E. F. Ellis, 31. Grigoriadis, G., and A. G. Stewart. 1991. 1-O-hexadecyl-2-acetyl-sn-glycero-3- N. F. Adkinson, Jr., J. W. Yunginger, and W. W. Busse. Mosby-Year Book, St. phospho (N,N,N-trimethyl)-hexanolamine: an analogue of platelet-activating fac- Louis, pp. 242–260. tor with partial agonist activity. Br. J. Pharmacol. 104:171. 3. Lee, T.-c., D. J. Lenihan, B. Malone, L. L. Roddy, and S. I. Wasserman. 1984. 32. Terasawa, M., H. Aratani, M. Setoguchi, and T. Tahara. 1990. Pharmacological Increased biosynthesis of platelet-activating factor in activated human eosino- actions of Y-24180. I. A potent and specific antagonist of platelet-activating phils. J. Biol. Chem. 259:5526. factor. Prostaglandins 40:553. 4. Sedgwick, J. B., R. F. Vrtis, M. F. Gourley, and W. W. Busse. 1988. Stimulus- 33. Bozza, P. T., W. Yu, J. F. Penrose, E. S. Morgan, A. M. Dvorak, and P. F. Weller. dependent differences in superoxide anion generation by normal human eosinophils and neutrophils. J. Allergy Clin. Immunol. 81:876. 1997. Eosinophil lipid bodies: specific, inducible intracellular sites for enhanced eicosanoid formation. J. Exp. Med. 186:909. 5. Abu-Ghazaleh, R. I., T. Fujisawa, J. Mestecky, R. A. Kyle, and G. J. Gleich. 1989. IgA-induced eosinophil degranulation. J. Immunol. 142:2393. 34. Weller, P. F., R. A. Monahan-Earley, H. F. Dvorak, and A. M. Dvorak. 1991. Cytoplasmic lipid bodies of human eosinophils: subcellular isolation and analysis 6. Filley, W. V., K. E. Holley, G. M. Kephart, and G. J. Gleich. 1982. Identification of arachidonate incorporation. Am. J. Pathol. 138:141. by immunofluorescence of eosinophil granule major basic protein in lung tissues of patients with bronchial asthma. Lancet 2:11. 35. Snyder, F.1994. Metabolic processing of PAF. Clin. Rev. Allergy 12:309. 7. Leiferman, K. M., S. J. Ackerman, H. A. Sampson, H. S. Haugen, P. Y. Venencie, 36. Jain, M. K., B. Z. Yu, J. Rogers, G. N. Ranadive, and O. G. Berg. 1991. Inter- G. J. Gleich. 1985. Dermal deposition of eosinophil-granule major basic protein facial catalysis by phospholipase A2: dissociation constants for calcium, sub- in atopic dermatitis: comparison with onchocerciasis. N. Engl. J. Med. 313:282. strate, products, and competitive inhibitors. Biochemistry 30:7306. 8. Krogel, C., T. Yukawa, G. Dent, P. Chanez, K. F. Chung, and P. J. Barnes. 1988. 37. White, S. R., M. E. Strek, G. V. P. Kulp, S. M. Spaethe, R. A. Burch, Platelet activating factor induces eosinophil peroxidase release from human eo- S. P. Neeley, and A. R. Leff. 1993. Regulation of human eosinophil degranulation sinophils. Immunology 64:559. and activation by endogenous phospholipase A2. J. Clin. Invest. 91:2118. The Journal of Immunology 2989

38. Ojima-Uchiyama, A., Y. Masuzawa, T. Sugiura, K. Waku, T. Fukuda, and (PAF) acetylhydrolase: correlation between deficiency of serum PAF acetyl- S. Makino. 1991. Production of platelet-activating factor by normodense and hydrolase and respiratory symptoms in asthmatic children. J. Clin. Invest. hypodense eosinophils. Lipids 26:1200. 82:1983. 39. Marcheselli, V. L., M. J. Rossowska, M.-T. Domingo, P. Braquet, and 43. Freitag, A., R. M. Watson, G. Matsos, C. Eastwood, and P. M. O’Byrne. 1993. N. G. Bazan. 1990. Distinct platelet-activating factor binding sites in synaptic Effect of a platelet activating factor antagonist, WEB 2086, on allergen induced endings and in intracellular membranes of rat cerebral cortex. J. Biol. Chem. asthmatic responses. Thorax 48:594. 265:9140. 44. Kuitert, L. M., K. P. Hui, S. Uthayarkumar, W. Burke, A. C. Newland, S. Uden, 40. Hwang, S. B., M. H. Lam, and S. S. Pong. 1986. Ionic and GTP regulation of and N. C. Barnes. 1993. Effect of the platelet-activating factor antagonist UK- binding of platelet-activating factor to receptors and platelet-activating factor- 74,505 on the early and late response to allergen. Am. Rev. Respir. Dis. 147:82. induced activation of GTPase in rabbit platelet membranes. J. Biol. Chem. 261: 45. Kagoshima, M., N. Tomomatsu, Y. Iwahisa, S. Yamaguchi, M. Matsuura, 532. Y. Kawakami, and M. Terasawa. 1997. Suppressive effects of Y-24180, a recep- 41. Svetlov, S., and S. Nigam. 1993. Evidence for the presence of specific high tor antagonist to platelet activating factor (PAF), on antigen-induced asthmatic affinity cytosolic binding sites for platelet-activating factor in human neutrophils. responses in guinea pigs. Inflamm. Res. 46:147. Biochem. Biophys. Res. Commun. 190:162. 46. Hozawa, S., Y. Haruta, S. Ishioka, and M. Yamakido. 1995. Effects of a PAF 42. Miwa, M., T. Miyake, T. Yamanaka, J. Sugatani, Y. Suzuki, S. Sakata, Y. Araki, antagonist, Y-24180, on bronchial hyperresponsiveness in patients with asthma. and M. Matsumoto. 1988. Characterization of serum platelet-activating factor Am. J. Respir. Crit. Care Med. 152:1198. Downloaded from http://www.jimmunol.org/ by guest on October 3, 2021