The JI PBL Intereron Source
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Muñoz, N. M.
Right arrow Articles by Leff, A. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Muñoz, N. M.
Right arrow Articles by Leff, A. R.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*Nucleotide
*Protein*UniGene
*Compound via MeSH
*Substance via MeSH
The Journal of Immunology, 2006, 177: 574-582.
Copyright © 2006 by The American Association of Immunologists

Transcellular Secretion of Group V Phospholipase A2 from Epithelium Induces beta2-Integrin-Mediated Adhesion and Synthesis of Leukotriene C4 in Eosinophils1

Nilda M. Muñoz*, Angelo Y. Meliton*, Anissa Lambertino*, Evan Boetticher*, Jonathan Learoyd*, Faraz Sultan*, Xiangdong Zhu*, Wonhwa Cho{ddagger} and Alan R. Leff2,*,{dagger}

* Section of Pulmonary and Critical Care Medicine, Department of Medicine, and {dagger} Department of Neurobiology Pharmacology and Physiology and Committees on Molecular Medicine, Clinical Pharmacology, and Cell Physiology, The University of Chicago, Chicago, Illinois 60637; and {ddagger} Department of Chemistry, University of Illinois, Chicago, IL 60607


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
We examined the mechanism by which secretory group V phospholipase A2 (gVPLA2) secreted from stimulated epithelial cells activates eosinophil adhesion to ICAM-1 surrogate protein and secretion of leukotriene (LT)C4. Exogenous human group V PLA2 (hVPLA2) caused an increase in surface CD11b expression and focal clustering of this integrin, which corresponded to increased beta2 integrin-mediated adhesion. Human IIaPLA2, a close homolog of hVPLA2, or W31A, an inactive mutant of hVPLA2, did not affect these responses. Exogenous lysophosphatidylcholine but not arachidonic acid mimicked the beta2 integrin-mediated adhesion caused by hVPLA2 activation. Inhibition of hVPLA2 with MCL-3G1, a mAb against gVPLA2, or with LY311727, a global secretory phospholipase A2 (PLA2) inhibitor, attenuated the activity of hVPLA2; trifluoromethylketone, an inhibitor of cytosolic group IVA PLA2 (gIVA-PLA2), had no inhibitory effect on hVPLA2-mediated adhesion. Activation of beta2 integrin-dependent adhesion by hVPLA2 did not cause ERK1/2 activation and was independent of gIVA-PLA2 phosphorylation. In other studies, eosinophils cocultured with epithelial cells were stimulated with FMLP/cytochalasin B (FMLP/B) and/or endothelin-1 (ET-1) before LTC4 assay. FMLP/B alone caused release of LTC4 from eosinophils, which was augmented by coculture with epithelial cells activated with ET-1. Addition of MCL-3G1 to cocultured cells caused ~50% inhibition of LTC4 secretion elicited by ET-1, which was blocked further by trifluoromethylketone. Our data indicate that hVPLA2 causes focal clustering of CD11b and beta2 integrin adhesion by a novel mechanism that is independent of arachidonic acid synthesis and gIVA-PLA2 activation. We also demonstrate that gVPLA2, endogenously secreted from activated epithelial cells, promotes secretion of LTC4 in cocultured eosinophils.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Phospholipase A2 (PLA2)3 enzymes are a heterogeneous family that is generally subdivided into three subclasses: Ca2+-dependent secretory PLA2 (1, 2, 3, 4), Ca2+-dependent cytosolic group IVA PLA2 (gIVA-PLA2) (5, 6, 7, 8), and Ca2+-independent PLA2 (9, 10). The 85-kDa gIVA-PLA2 has been associated with 1) secretion of eicosanoid metabolites in inflammatory cells (11, 12, 13), 2) cell migration (14, 15, 16), and 3) airway hyperresponsiveness in allergen-sensitized mice and guinea pigs (8, 17, 18). Blockade of gIVA-PLA2 activity by the pharmacological inhibitor, trifluoromethylketone (TFMK), selectively attenuates these responses (11, 15, 16, 17). Deletion of the gIVA-PLA2 gene in mice has confirmed that this enzyme is essential for the generation of lipid mediators in inflammatory cells and fibroblasts (8, 11, 12).

Recently, we have demonstrated that human gVPLA2 (hVPLA2) (1, 19) causes the release of eicosanoid metabolites from inflammatory cells (neutrophils and eosinophils) (20, 21, 22). This enzyme binds and hydrolyzes phosphatidylcholine (PC), which is abundant in the outer plasma membrane of mammalian cells (20, 21, 22, 23). Exogenous hVPLA2 causes leukotriene (LT)B4 secretion through ERK1/2-mediated phosphorylation of gIVA-PLA2 in isolated human neutrophils (21, 22). This response also corresponds to an increase in intracellular Ca2+ concentration. In human eosinophils, hVPLA2 also is internalized to cause lipid hydrolysis from perinuclear membranes (23); this results in LTC4 synthesis and secretion (20). However, hVPLA2-mediated synthesis of LTC4 in human eosinophils does not require either gIVA-PLA2 activation or intracellular Ca2+ concentration release (20).

Prior reports have implicated the phosphorylation of gIVA-PLA2 by either ERK1/2 (14, 15, 16) or PI3K (17) as an essential step in beta2 integrin receptor clustering in human eosinophils, which precedes their adhesion to ICAM-1. However, there has been no prior report of beta2 integrin-mediated adhesion caused by a secretory PLA2 or of integrin adhesion occurring independently of gIVA-PLA2 activation.

Secretory gVPLA2 is found in high concentrations in macrophages and epithelium but not in granulocytes (24, 25). In this study, we hypothesized that gVPLA2 might be a transcellular messenger protein for activation of eosinophil adhesion and eicosanoid synthesis. To test this hypothesis, we first assessed the activity of hVPLA2 on beta2 integrin-mediated adhesion to ICAM-1-surrogate protein in human eosinophils. We found that hVPLA2-induces adhesion independent of MAPK or gIVA-PLA2 activation. Adhesion corresponded to the increased surface CD11b expression and focal clustering of this integrin, all of which was inhibited by MCL-3G1, a specific mAb directed against gVPLA2 (18, 19). Lysophosphatidylcholine (LysoPC), a PLA2 reaction product of PC that contains relatively long sn-1 fatty acyl chains (1, 21, 23), mimicked the beta2 integrin-mediated adhesion caused by hVPLA2. We also demonstrated translocation of gVPLA2 from stimulated epithelial cells to eosinophils in coculture, which corresponded to up-regulation of eosinophil adhesion to ICAM-1 and augmented eicosanoid synthesis. Finally, we demonstrated that gIIaPLA2, previously postulated to be an enzyme of inflammation in human asthma, has no effect in up-regulation of eosinophil adhesion or secretion.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Expression and purification of secretory PLA2 enzymes

Purified hIIaPLA2, hVPLA2, and the hVPLA2 mutant W31A were expressed in Escherichia coli as previously described (21, 22, 23, 26). A structure-function analysis on the putative substrate-binding site of gVPLA2 identified that tryptophan 31 (W31) residue is essential for its unique ability to bind and hydrolyze PC molecules (21, 22, 23, 26). Thus, mutation of W31 to alanine (A) lowers membrane affinity and inactivates the hydrolytic activity of hVPLA2 (21, 22, 23). The purity of secretory PLA2 enzymes as assessed by SDS-PAGE was typically ≥90%.

Purification of human peripheral blood eosinophils

Eosinophils were isolated from peripheral blood obtained from mildly atopic, nonsmoking volunteers (14, 15, 16, 20, 27, 28). Atopy was defined by criteria used in The University of Chicago Asthma Research Center for the National Heart, Lung, and Blood Institute Human Cooperative Asthma Genetics projects. Briefly, blood was anticoagulated with 1/1000 heparin solutions, diluted with calcium-free HBSS, and layered onto 1.089 g/ml Percoll solution. The cell pellet containing granulocytes and erythrocytes was lysed and washed twice with HBSS containing 0.2% BSA. The total number of cells was counted by a Coulter counter and anti-CD16 beads (0.65 µl/106 neutrophils) were added to selectively extract neutrophils. Thirty minutes later, the cell suspension was transferred to a type C magnetic separation column held within a 0.6 Testa MACS magnet. The purity of eosinophils was determined by differential counts of H&E-stained cytospin preparations. Eosinophils viability was assessed by trypan blue exclusion dye (14, 15, 16, 20, 27, 28).

Epithelial cell culture

Studies were performed using monolayer of the human epithelial cell line (American Type Culture Collection), and were maintained in DMEM supplemented with 0.6 µg/ml penicillin, 60 µg/ml streptomycin, 2 mM L-glutamine, and 10% heat-inactivated FCS in ventilated tissue culture flasks at 37°C and 5% CO2. Passage number was kept to fewer than five passages from original stocks (29, 30, 31, 32). To minimize effects of exogenous growth factors in our system, we reduced the supplemented serum concentration to 0.5% 24 h before and during all experimental protocols. This serum supplement concentration slowed, but did not stop, cell division. For experiments, cells were detached from the plastic by incubation in 0.05% trypsin and 0.01% EDTA for 15 min and seeded at 200,000 cells/sterilized 12-microplate wells on the day before experimental procedure. The cells were generally 80–90% confluent on the day after seeding.

Flow cytometric analysis

Eosinophils were activated with buffer control, 10–9–10–7 M hVPLA2, 10–7 M hIIaPLA2 (close homolog of gVPLA2), 10–7 M W31A (inactive hVPLA2 mutant), or 10–7 M FMLP for 15 min at 4°C in PBS solution containing 2% BSA. Treated cells were washed and incubated with the mAb against CD11b (clone Bear I) or equivalent concentration of isotype-matched control Ab (negative control) for another 30 min at 4°C. Cells were washed and counterstained with 1/50 FITC-conjugated secondary goat anti-mouse Ig for additional 30 min. Thereafter, the treated cells were fixed with 1% paraformaldehyde in PBS, and mean fluorescence intensity (MFI) was determined on at least 5000 cells from each sample as acquired using a FACScan flow cytometry (BD Biosciences). For quantitative evaluation, the CD11b populations were gated manually, and the percentage of CD11b-positive cells was determined using CellQuest software (BD Immunocytometry Systems).

To confirm the release of gVPLA2 from epithelial cells, 5 x 105 quiescent cells were activated with 10–6 M endothelin-1 (ET-1) at different time intervals, and surface membrane expression of gVPLA2 was analyzed as previously described. Briefly, treated cells were incubated with 2 µg/ml MCL-3G1, a specific mAb directed against gVPLA2, or isotype-matched control Ab (negative control) for 30 min and washed before addition of FITC-conjugated goat anti-mouse Ig. After 30 min, cells were again washed and fixed with 1% paraformaldehyde solution, and the surface expression of gVPLA2 was analyzed as for CD11b expression.

Adhesion assay and measurement of residual eosinophil peroxidase (EPO) activity

We have established previously that plated BSA is a full surrogate for ICAM-1 for integrin binding in vitro (15, 27). Briefly, eosinophils (104 cells) were added to BSA-coated microplate wells and activated with buffer control, 10–9–10–7 M hVPLA2, 10–7 M hIIaPLA2, 10–7 M W31A, 10–7 M FMLP, 10–7 M hIbPLA2, 10–7 M hXPLA2, or 10–7 M Naja naja naja PLA2 for 15 min at 37°C. At the end of the activation period, nonadherent cells were washed three times with HBSS buffer/0.1% gelatin, and 100 µl of HBSS/0.1% gelatin was added to the reaction wells. To generate a standard curve, serial dilutions of the original cell suspension (106 cells/10 ml) were added to the noncoated microplate wells, and experiments were conducted as follows. A mixture of 1 mM hydrogen peroxide, 1 mM o-phenylenediamine, and 0.1% Triton X-100 in Tris buffer (pH 8.0) was used as a substrate to analyze cell adhesion, and the reaction mixture was stopped by addition of 4 M sulfuric acid. All assays were performed in duplicate. The detection of EPO by this method was linear between concentrations of 103–104 cells/well as measured by a standard curve. No residual EPO was detected in a cell-free reaction supernatant indicating that EPO was not released due to spontaneous cell degranulation. The absorbance was measured at 490 nm in a Thermomax microplate reader (Molecular Devices).

In another set of experiments, eosinophils also were pretreated with 1) 10 µg/ml MCL-3G1, a specific mAb directed against gVPLA2 (24, 25); 2) 10 µM LY311727, a global secretory PLA2 inhibitor (15, 27); 3) 10 µM TFMK, a selective gIVA-PLA2 inhibitor (14, 15, 20, 21, 27); or 4) 10 µg/ml anti-CD11b neutralizing mAb (15, 27) for 25 min before activation with 10–7 M hVPLA2, and adhesion assay was performed as described.

To understand the mechanism by which gVPLA2 induces beta2 integrin adhesion, the effect of two by-products of outer plasma membrane hydrolysis by hVPLA2, LysoPC and arachidonic acid, on cell adhesion was examined. Eosinophils were activated with increasing concentrations of LysoPC (3, 10, or 30 µM) or arachidonic acid (0.1–10 µM) for 15 min before adhesion assay performed as described. LysoPC was selected as a representative lysophospholipid because the major component of the outer plasma membrane of mammalian cells is PC (1, 23).

Western blot analysis of phosphorylation of ERK1/2 and gIVA-PLA2

To determine whether eosinophil adhesion elicited by hVPLA2 was mediated through activation of ERK1/2, which causes subsequent phosphorylation and activation of gIVA-PLA2, eosinophils were stimulated with buffer control, 10–7 M hVPLA2 alone, or 10–7 M FMLP alone and phosphorylation of ERK1/2 and gIVA-PLA2 were analyzed as follows. The cell pellet was lysed in a disruption buffer (20 mM Tris-HCl, 30 mM Na4P2O7, 50 mM NaF, 40 mM NaCl, 5 mM EDTA, 1% Nonidet-40, protease inhibitors tablet) and centrifuged at 12,000 x g for 30 s to remove nuclear and cellular debris. SDS-PAGE loading buffer was added to the collected supernatant, boiled for 5 min, and then stored at –70°C. Prepared samples were subjected to 10% SDS-PAGE for ERK1/2 and a 7.5% gel for gIVA-PLA2 under reducing conditions as previously described (14, 15, 16, 21, 22, 23, 33). The resolved proteins were transferred onto polyvinylidene fluoride membrane using a semidry apparatus system. After blockade with 1% BSA, the membrane was incubated with Abs specific for phosphorylated ERK1/2 or Ser505 phosphorylation-specific gIVA-PLA2 (Cell Signaling Technology) overnight at 4°C. The protein of interest was visualized by an ECL system (Amersham).

Focal clustering of CD11b

To determine the mechanism by which hVPLA2 up-regulates beta2 integrin-mediated adhesion, eosinophils were stained with mAb directed against CD11b and focal clustering of CD11b expression was visualized by confocal microscopy (28). In these experiments, eosinophils were activated with buffer control, 10–7 M hVPLA2, 10–7 M hIIaPLA2, and 10–7 M FMLP for 15 min, and cytoslides were prepared in duplicate (Cytospin 2; Shandon). In another set of experiments using an aliquot from the same eosinophil isolation, the specificity of focal clustering of CD11b in response to hVPLA2 stimulation was assessed. Cells first were treated with MCL-3G1, a mAb directed against hVPLA2 (20, 24, 25), for 30 min before hVPLA2 stimulation and cytoslides were prepared as earlier described. The slides containing treated cells were stained with FITC-conjugated CD11b (clone Bear I) mAb, and fluorescence was analyzed by using an Axiovert confocal microscope (Zeiss) equipped with an external argon-krypton laser at 488 nm.

Imaging of secreted gVPLA2 from epithelial cells

To examine the translocation/migration of endogenously secreted gVPLA2 from activated epithelial cells to adhering eosinophils, we performed a real-time confocal microscopic imaging in which fluorescent phospholipid, PED6 (N-(6-(2,4-dinitrophenyl)amino)hexanoyl)-1-hexadecanoyl-2-(4,4-difluro-5,7-dmethyl-4-bora-3a, 4a-diaza-s-indacene-3-pentanoyl) sn-glycerol-3-phosphoethanolamine triethylammonium salt) vesicle solution (0.75 mM 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoserine (POPS)/cholesterol/1-palmitoly-2-oleoyl-sn-glycero-3-phosphoglycerol (POPG)/PED6 at a 107:31:20:1 molecular ratio) was primarily labeled on the plasma membrane of eosinophils (21, 23). Briefly, eosinophils first were overlaid with 10 µl of PED6 in HBSS and incubated for 50 min at 37°C with 5% CO2 (21, 23). The labeled cells were washed with HBSS containing 2 mM CaCl2 to remove the excess dye and transferred to wells containing resting epithelial cells. ET-1 at 10–6 M concentration was added to the reaction wells and imaging was done with a Zeiss 510 laser scanning confocal microscope with the detector gain adjusted to eliminate the background auto fluorescence. The signal from hydrolyzed PED6 by migrating gVPLA2 was visualized with a 488 nm argon/krypton laser and a 530-nm line-pass filter. A 63x H2O immersion lens was used for monitoring the MFI of gVPLA2 activity (21, 23).

Determination of LTC4 secretion by ELISA

The concentration of LTC4 secreted during eosinophil activation was assayed by competitive enzyme immunoassay (Cayman Chemicals) using a method that we have reported previously (20, 33, 34). A total of 21 eosinophil isolations from different donors were used in this protocol. The effect of exogenous 1) 10–6 M ET-1, 2) 10–7 M FMLP plus 5 µg/ml cytochalasin B (FMLP/B), or 3) FMLP/B with or without ET-1 on LTC4 secretion was first examined in naive eosinophils or epithelial cells. All samples were run in duplicate, and LTC4 secretion was expressed in picograms per 106 cells.

Using aliquots from the same isolations, eosinophils (2.5 x 105 cells) were cocultured with epithelial cells and stimulated with 1) vehicle control, 2) 10–7 M FMLP/B, 3) 10–6 M ET-1, or 4) FMLP/B with or without ET-1 at 37°C in a final volume of 250 µl of HBSS plus Ca2+ buffer. Fifteen minutes later, the reaction mixture was terminated by centrifugation and activated supernatants were collected for later analysis of eosinophil LTC4 secretion by ELISA.

To determine the selectivity of inhibitors used in this protocol, two additional control experiments were conducted for epithelial cells alone or eosinophils alone pretreated with 10 µM TFMK and/or 10 µg/ml MCL-3G1 for 30 min before cell contact and activation of cocultured cells with 10–7 M FMLP/B or 10–6 M ET-1. Fifteen minutes later, supernatant was collected from treated cells after centrifugation and secretion of LTC4 was analyzed as described.

In a final series, four additional experiments were conducted to determine whether secretion of LTC4 caused by gVPLA2 was further augmented through independent activation of gIVA-PLA2: 1) to assess the effect of gIVA-PLA2, eosinophils were pretreated with 10 µM TFMK for 25 min before addition to microplate wells coated with epithelial cells; 2) to assess the effect of gVPLA2, epithelial cells were preincubated with 10 µg/ml MCL-3G1 for 25 min before addition of eosinophils; 3) to further confirm the simultaneous activation of gVPLA2 and gIVA-PLA2, cocultured cells were pretreated with MCL-3G1 plus TFMK before adhesion and activation with ET-1 plus FMLP/B; and 4) to determine whether LTC4 production caused by epithelial secretion of gVPLA2 was augmented by independent activation of gIVA-PLA2 by FMLP/B, nonstimulated eosinophils were allowed to adhere on treated epithelial cells for 10 min on ice before addition of ET-1 plus FMLP/B for an additional 15 min. The reaction was terminated by centrifugation, and supernatant was collected for later analysis of eosinophil LTC4 secretion by ELISA. Density measurements were made on a microplate absorbance spectrophotometer at 405 nm. The final concentrations were calculated from standard curves fitted by four-parameter analysis (Softmax v2.01 software; Molecular Devices), and data are expressed as picograms per 106 cells.

Statistical analysis

All data are shown as mean ± SEM. Statistical analyses among groups were performed using ANOVA. Where differences between groups were detected, statistical significance was analyzed by a Fisher’s least protected difference test. Statistical significance was determined at values of p < 0.05.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Effect of secretory PLA2 isozymes on surface expression of CD11b

We first determined the effect of purified secretory PLA2 enzymes on the surface expression of the CD11b adhesion molecules in eosinophils. Exogenous hVPLA2 selectively up-regulated CD11b surface expression in a concentration-dependent manner as determined by flow cytometric analysis (n = 4) (Fig. 1). In resting eosinophils (positive control), MFI for surface CD11b expression was 57.2 ± 10.8 (p < 0.05 vs 6.2 ± 0.31 MFI, negative isotype control) and progressively increased to 81.3 ± 0.44 MFI after 10–8 M hVPLA2 activation (p = NS vs positive control) and to 120.2 ± 12.7 MFI after activation with 10–7 M hVPLA2 (p < 0.05 vs positive control). At 10–7 M, hIIaPLA2 or W31A did not up-regulate the surface expression of CD11b (p = NS vs positive control).


Figure 1
View larger version (13K):
[in this window]
[in a new window]
  		FIGURE 1. Composite data of up-regulated surface CD11b expression in isolated human eosinophils. Cells activated with 10–9–10–7 M hVPLA2 (hV), 10–7 M hIIaPLA2 (hIIa), and 10–7 M W31A and were stained with mAb against CD11b for 30 min and incubated with FITC-labeled goat anti-mouse Ig before analysis. Intracellular FITC intensity was analyzed by flow cytometry. Negative control is stained with isotype-match control, and positive control is baseline surface CD11b expression for unstimulated cells. Data represent the mean ± SEM from four different donors. *, p < 0.05 for 10–7 M hVPLA2 vs hIIaPLA2, or W31A.

 
Effect of purified secretory PLA2 isozymes on cell adhesion

We next examined whether the increased surface expression of CD11b elicited by 10–9–10–7 M hVPLA2 caused increased adhesion of eosinophils to microplate wells-coated ICAM surrogate protein (BSA; see Materials and Methods) (15, 27). Human gVPLA2 increased eosinophil beta2 integrin adhesion of eosinophils in a concentration-dependent manner (Fig. 2A) and corresponded to up-regulated surface expression of CD11b (see Fig. 1). Adhesion was comparable for buffer-activated cells and cells activated with 10–9 M hVPLA2 (5.9 ± 0.09 vs 7.8 ± 3.3%; p = NS). Thereafter, an incremental increase in adhesion was demonstrated to 13.63 ± 1.5% with 10–8 M hVPLA2 (p < 0.05 vs buffer-stimulated cells or 10–9 M hVPLA2) and further to 19.8 ± 3.2% for cells activated with 10–7 M hVPLA2 (p < 0.01 vs buffer-stimulated cells or 10–9 M hVPLA2) (Fig. 2A). Neither 10–7 M W31A (5.83 ± 0.58%) nor 10–7 M hIIaPLA2 (4.5 ± 0.44%) caused increased adhesion above control values. As a positive control, we used FMLP, which has been shown previously to cause an increase in beta1/beta2-mediated adhesion of eosinophils in a time- and concentration-dependent manner (14, 15, 27). At 10–7 M FMLP, adhesion increased to 15.35 ± 1.56% (p < 0.05 vs buffer control).


Figure 2
View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 2. Effect of purified secretory PLA2 enzymes on cell adhesion. A, Eosinophils were activated with two highly homologous 14 kDa secretory PLA2 (10–9–10–7 M hVPLA2 (hV), 10–7 M hIIaPLA2 (hIIa)), 10–7 M FMLP, and 10–7 M W31A, an inactive mutant of hVPLA2, for 15 min at 37°C before adhesion assay. The adherence of eosinophils was assessed as described in Materials and Methods. B, Effect of purified 14–19 kDa secretory PLA2 enzymes on cell adhesion. Cells were activated with purified 10–7 M hIb, 10–7 M hX, and 10–7 M Naja naja naja PLA2 for 15 min and adhesion assay was assessed as described in Materials and Methods. Data represent the average of duplicate wells and represent the mean ± SEM (n = 5 experiments). *, p < 0.05 for 10–8 M hVPLA2 or 10–7 M FMLP vs buffer-stimulated, W31A, or hIIaPLA2; **, p < 0.01 for 10–7 M hVPLA2 vs buffer-stimulated or hIIaPLA2, or W31A.

 
We further assessed the specificity of hVPLA2 in causing beta2 integrin-mediated adhesion of eosinophils. Eosinophils were activated with three types of secretory PLA2 enzymes, 10–7 M human group Ib PLA2 (hIbPLA2), 10–7 M human group X PLA2 (hXPLA2), and 10–7 M Naja naja naja PLA2 (Fig. 2B). Although hIbPLA2 has no hydrolytic activity for inflammatory cells (3, 31), hXPLA2 and Naja naja naja PLA2 have been shown to bind and hydrolyze the outer PC plasma membrane of granulocytes. These isoforms, however, have no capacity for uptake into mammalian cells due to their low affinity for cell surface heparan sulfate proteoglycan (HSPG) (3, 7, 23, 26). Accordingly, secretory hIbPLA2, hXPLA2, or Naja naja naja PLA2 did not induce cell adhesion (Fig. 2B).

Effect of selective blockade of stimulated cell adhesion

We next determined whether gIVA-PLA2 is required for the up-regulation of beta2 integrin adhesion caused by hVPLA2 in eosinophils. Cells first were treated with 10 µg/ml MCL-3G1 (a specific mAb directed against gVPLA2), 10 µM LY311727 (a global secretory PLA2 inhibitor), 10 µM TFMK (a selective gIVA-PLA2 inhibitor), or 10 µg/ml anti-CD11b mAb (clone Bear I) before activation with 10–7 M hVPLA2. MCL-3G1, LY311727, and anti-CD11b mAb substantially attenuated the effect of 10–7 M hVPLA2 on eosinophil beta2 integrin binding to ICAM-1 surrogate (Fig. 3); hVPLA2-mediated cell adhesion decreased from 22.7 ± 1.2 to 4.7 ± 0.73% after pretreatment with 10 µg/ml MCL-3G1 (p < 0.001), to 7.5 ± 1.1% for eosinophils treated with 10 µM LY311727, and to 8.65 ± 0.81% for eosinophils pretreated with anti-CD11b mAb (p < 0.01) before hVPLA2 activation. By contrast, 10 µM TFMK, a concentration that blocks beta2 integrin adhesion caused by 10–6 M FMLP (14, 15, 16), did not inhibit hVPLA2-mediated adhesion of eosinophils, indicating that gIVA-PLA2 activation is not required for eosinophil adhesion elicited by 10–7 M hVPLA2.


Figure 3
View larger version (11K):
[in this window]
[in a new window]
  		FIGURE 3. Selective blockade of hVPLA2-induced cell adhesion. Cells were preincubated with optimal concentration of MCL-3G1 (a specific mAb against gVPLA2), LY311727 (a global secretory PLA2 inhibitor), TFMK (a selective gIVA-PLA2 inhibitor), or anti-CD11b mAb (clone Bear I) before 10–7 M hVPLA2 stimulation. Eosinophil adhesion was assessed as described in Materials and Methods in the presence of these inhibitors. Results are shown as mean ± SEM (n = 4 experiments). *, p < 0.001 vs buffer-activated or MCL-3G1; **, p < 0.01 vs LY311727 or anti-CD11b mAb.

 
Effect of LysoPC and arachidonic acid on beta2 integrin-mediated adhesion

LysoPC caused up-regulation of eosinophil adhesion to ICAM-1 surrogate protein at a concentration of 10 µM, which was comparable to that caused by 10 nM hVPLA2 (Fig. 2A). At less concentration (3 µM), LysoPC did not induce cell adhesion (4.8 ± 0.51%) and was comparable to that induced by buffer-activated cells (5.9 ± 0.12%; p = NS). However, adhesion increased to 12.91 ± 1.63% in response to 10 µM LysoPC (p < 0.05 vs buffer-stimulated cells or 3 µM LysoPC-activated cells). At 30 µM, LysoPC caused no greater up-regulation of adhesion than at 10 µM concentration. Arachidonic acid (0.1–10 µM) had no effect in up-regulating eosinophil adhesion (Fig. 4). These data indicate that LysoPC, a direct hydrolysis product of hVPLA2, causes augmented adhesion of eosinophil to ICAM-1 surrogate protein.


Figure 4
View larger version (11K):
[in this window]
[in a new window]
  		FIGURE 4. Effect of LysoPC and arachidonic acid (AA) on cell adhesion. Cells were activated with increasing concentrations of LysoPC ranging from 3 to 30 µM and from 0.1 to 10 µM arachidonic acid, and adhesion assay was performed as described in Materials and Methods. Each point is the average of duplicate wells and represents the mean ± SEM (n = 4 experiments). *, p < 0.05 vs buffer-activated or 3 µM LysoPC- or arachidonic acid-treated cells.

 
Mechanism hVPLA2-induced cellular adhesion

Effect of hVPLA2 on ERK1/2 and gIVA-PLA2 phosphorylation. We next investigated whether hVPLA2-mediated adhesion caused by beta2 integrin involves activation of ERK1/2 and gIVA-PLA2 pathways. It has been shown that phosphorylated ERK1/2 caused gIVA-PLA2 phosphorylation in various mammalian cells (14, 21, 32, 33). Immunoblotting demonstrated that 10–7 M hVPLA2 did not cause phosphorylation of either ERK1/2 (Fig. 5A) or gIVA-PLA2 (Fig. 5B). By contrast, 10–7 M FMLP, a G protein-mediated activator of eicosanoid synthesis and adhesion in eosinophils, caused phosphorylation of both ERK1/2 and gIVA-PLA2 (Fig. 5).


Figure 5
View larger version (31K):
[in this window]
[in a new window]
  		FIGURE 5. ERK1/2 and gIVA-PLA2 phosphorylation during stimulated eosinophil adhesion to ICAM-1 surrogate protein. A, ERK1/2 was detected by polyclonal Ab, which identified only the phosphorylated forms of ERK1/2 (top). Equivalent loading is demonstrated by stained total ERK expression (bottom). B, gIVA-PLA2 phosphorylation was identified by Ser505 gIVA-PLA2 phosphorylated mAb as described in Materials and Methods. Immunoblotting demonstrated that 10–7 M hVPLA2 (hV) did not cause phosphorylation of either ERK1/2 (A) or gIVA-PLA2 (B). By contrast, 10–7 M FMLP caused phosphorylation of both ERK1/2 and gIVA-PLA2. Experiments were conducted three times with comparable results, and representative immunoblots are shown.

 
Effect of hVPLA2 on focal clustering of CD11b. In cells stained with the mAb against CD11b, confocal microscopy revealed that 10–7 M hVPLA2, which does not activate gIVA-PLA2 (Fig. 5B) caused crescent stain of CD11b, indicated by yellow-green color clustering (Fig. 6). This response was specific; pretreatment of eosinophils with MCL-3G1 blocked completely the focally clustered appearance of CD11b caused by 10–7 M hVPLA2. Cytosolic distribution of CD11b diffused after MCL-3G1 pretreatment was nonclustered as observed for cells receiving buffer alone or cells activated with 10–7 M hIIaPLA2. At 10–7 M, FMLP-activated cells caused comparable CD11b clustering when compared with hVPLA2-treated cells.


Figure 6
View larger version (31K):
[in this window]
[in a new window]
  		FIGURE 6. Focal clustering of CD11b: representative photomicrograph of stained eosinophils. Eosinophils were activated with buffer control (negative control), 10–7 M hIIaPLA2, 10–7 M hVPLA2± 10 µg/ml MCL-3G1, and 10–7 M FMLP before addition of mAb directed against CD11b. Clustering of CD11b (distinct crescent yellow-green stain) was assessed by confocal microscopy. Experiments were performed three times with identical results.

 
Translocation of endogenously secreted gVPLA2 from epithelial cells

Stimulated secretion of gVPLA2. Cultured epithelial cells were activated with 10–6 M ET-1 at different times, and secreted gVPLA2 was analyzed by flow cytometric analysis (Fig. 7A). Baseline (positive control) surface expression of gVPLA2 from epithelial cells was 56.13 ± 11.3 MFI; this increased to 97.2 ± 40.7 MFI 30 s after stimulation with ET-1 (p < 0.05). Thereafter, a gradual decrease in gVPLA2 expression was observed.


Figure 7
View larger version (61K):
[in this window]
[in a new window]
  		FIGURE 7. Expression of gVPLA2 secreted from activated epithelial cells. A, Endogenous hVPLA2 was measured by flow cytometric analysis for epithelial cells in response to 10–6 M ET-1 activation. Data represent the mean ± SEM (n = 4 experiments). *, p < 0.05 vs positive control. B, Transcellular communication: representative images in real-time. PED6-labeled eosinophils were overlaid onto serum-starved epithelial cells before activation with ET-1. Translocation of gVPLA2 from epithelial cells (red arrow) was analyzed by confocal microscopy (magnification, x63). Inset shows enlarged view of transport protein.

 
Transcellular localization of gVPLA2 to eosinophils. This study was designed to visualize the actual transfer/translocation of endogenously secreted gVPLA2 in real-time by confocal microscopy from stimulated epithelial cells to adherent eosinophils in coculture. Naive eosinophils first were labeled with a fluorogenic phospholipid PED6 containing vesicles for 20 min before addition to epithelial cell-coated microplate wells. Fig. 7B shows illustrative photomicrographs obtained by confocal microscopy of eosinophils stained with POPS/cholesterol/POPG/PED6 in coculture with epithelial cells before and after activation with 10–6 M ET-1 corresponding to the collated data in Fig. 7A. Translocation of gVPLA2 from epithelial cells to eosinophils is demonstrated by increasing fluorescence intensity (Fig. 7B, inset, green dye) on the outer plasma membrane of labeled eosinophils. Images were generated during the hydrolysis of PED6 incorporated into endogenously secreted gVPLA2.

Effect of endogenously secreted gVPLA2 from epithelial cells on eosinophil LTC4 secretion

To determine the effect of endogenously secreted gVPLA2 from activated epithelial cells for eosinophil activation, synthesis of cysteinyl LTC4 was measured in treated supernatant by ELISA. We first tested the specificity of each activating agents on epithelial cells alone and eosinophils alone for LTC4 secretion. Neither 10–7 M FMLP plus 5 µg/ml cytochalasin B (FMLP/B) nor 10–6 M ET-1 caused secretion of cysteinyl LTC4 from epithelial cells alone (Fig. 8). Activation of eosinophils with FMLP/B caused release of LTC4 from 23.6 ± 8.3 pg/106 cells (buffer alone) to 130 ± 44.2 pg/106 cells (p < 0.05); ET-1 did not cause LTC4 secretion in eosinophils. Activation of eosinophils alone with 10–7 M FMLP/B did not differ for eosinophils alone stimulated with ET-1 ± FMLP/B (130 ± 44.2 pg/106 cells vs 143 ± 34 pg/106 cells; p = NS).


Figure 8
View larger version (15K):
[in this window]
[in a new window]
  		FIGURE 8. Demonstration of secreted gVPLA2 from stimulated epithelial cells on LTC4 secretion from cocultured eosinophils. Epithelial cells and eosinophils were activated with ET-1 and/or FMLP, and secretion of LTC4 was assessed by ELISA. Each point represents the mean ± SEM (n = 5 experiments). p = NS for epithelial cells alone for all comparisons. *, p < 0.05 for eosinophils alone (EOS alone) activated by ET-1 vs FMLP/B or ET-1 + FMLP/B; p = NS for FMLP/B vs ET-1 + FMLP/B or for ET-1 vs buffer-activated cells. p < 0.05 for epithelial cells plus eosinophils (+ EOS) activated by ET-1 + FMLP/B vs FMLP, ET-1 vs FMLP/B, or ET-1 vs buffer-activated cells; **, p < 0.01 for ET-1 + FMLP/B vs ET-1.

 
We next examined the effect of adhesive ligation of eosinophils to epithelial cells on release of LTC4 after treatment with either ET-1 or FMLP/B or with the combination of these agents. Baseline secretion (before activation) of LTC4 was insignificant in all treatment protocols (Fig. 8). For epithelial cells alone, LTC4 concentration in the supernatant was 22.2 ± 5.1 pg/106 cells, 23.6 ± 8.3 pg/106 cells for eosinophils alone, and 31.2 ± 3.9 pg/106 cells for cocultured eosinophils plus epithelial cells (p = NS vs baseline for all comparisons). ET-1-induced LTC4 secretion in cocultured cells was 147 ± 24.7 pg/106 cells, which is comparable to those eosinophils activated solely with FMLP/B or ET-1 with or without FMLP/B (p = NS). Activation of cocultured cells with FMLP/B alone further enhanced the release of LTC4 to 288 ± 41.9 pg/106 cells (p < 0.05 vs ET-1-activated cocultured cells). The addition of ET-1 with or without FMLP/B caused further augmentation of LTC4 secretion to 466.1 ± 65.3 pg/106 cells (p < 0.05 vs FMLP/B alone and p < 0.01 vs ET-1 alone). Our data suggest that two pathways, gVPLA2-dependent and gIVA-PLA2-dependent, both cause secretion of LTC4 in human eosinophils through independent mechanisms.

Specificity of LTC4 secretion by endogenously secreted gVPLA2 from epithelial cells

Selectivity of inhibitors first was tested in eosinophils alone or cultured epithelial cells alone pretreated with TFMK and/or MCL-3G1 before cell contact and activation with FMLP/B alone (Fig. 9A) or ET-1 alone (Fig. 9B). Synthesis of LTC4 in cocultured cells was comparable for all buffer-stimulated groups. Release of LTC4 was 251.7 ± 39.5 pg/106 cells for cocultured cells stimulated with FMLP/B alone (Fig. 9A) and 192.2 ± 32.1 pg/106 cells after ET-1 activation (p = NS) (Fig. 9B). Pretreatment with MCL-3G1 alone, which completely inhibited LTC4 secretion caused by ET-1 activation (75.25 ± 26.8 pg/106 cells) (Fig. 9B), had no inhibitory effect on FMLP/B-activated cells (290 ± 56 pg/106 cells) (Fig. 9A). By contrast, secretion of LTC4 decreased to 96.82 ± 26.22 pg/106 cells for cells exposed to TFMK alone (p < 0.05 vs FMLP/B-activated, no ET-1) (Fig. 9A); TFMK did not affect the ET-1-induced LTC4 secretion (Fig. 9B). Coincubation of eosinophils with TFMK plus MCL-3G1 before epithelial cells contact blocked the secretion of LTC4 caused by FMLP/B activation to 67.02 ± 17.05 pg/106 cells (p < 0.05) (Fig. 9A). Blockade of epithelial cells with TFMK plus MCL-3G1 also inhibited the LTC4 secretion to 92.33 ± 19.6 pg/106 cells in response to ET-1 activation (p < 0.05) with no FMLP/B (Fig. 9B).


Figure 9
View larger version (24K):
[in this window]
[in a new window]
  		FIGURE 9. Specificity of endogenously secreted gVPLA2 from epithelial cells in activating eosinophil secretion. A, Eosinophils were exposed to MCL-3G1 and/or TFMK before contact with resting epithelial cells and activation with FMLP/B (n = 3 experiments). *, p < 0.05 for FMLP/B or MCL-3G1 vs TFMK or MCL-3G1; p = NS for FMLP/B vs MCL-3G1. B, Epithelial cells were exposed to MCL-3G1 and/or TFMK before addition of eosinophils and activation with 10–6 M ET-1 (n = 3 experiments). *, p < 0.05 for ET-1 vs MCL-3G1 or MCL-3G1 + TFMK; p = NS for ET-1 vs TFMK. C, Epithelial cells were pretreated with MCL-3G1 (n = 6 experiments), or eosinophils were pretreated with TFMK (n = 6 experiments) or cocultured cells were preincubated with MCL-3G1 + TFMK (n = 3 experiments) before cell-cell contact and activation with ET-1 + FMLP/B. Data represent the mean ± SEM. *, p < 0.05 for ET-1 + FMLP vs MCL-3G1; **, p < 0.01 for ET-1 + FMLP/B vs TFMK or MCL-3G1 + TFMK.

 
Selective inhibition of gVPLA2 with MCL-3G1, a neutralizing mAb directed against gVPLA2, caused a significant blockade of augmented eosinophil LTC4 secretion (Fig. 9C). Treatment with MCL-3G1 blocked augmented LTC4 secretion caused by ET-1 with or without FMLP/B from 419.33 ± 32.7 pg/106 cells to 194.2 ± 23.3 pg/106 cells (p < 0.01; n = 6 experiments). Blockade of gIVA-PLA2 by TFMK also inhibited the LTC4 secretion caused by ET-1 plus FMLP/B to 106.4 ± 15.34 pg/106 cells (p < 0.001 vs ET-1 plus FMLP/B-activated cells; n = 6 experiments). Finally, LTC4 secretion caused by ET-1 plus FMLP/B was completely inhibited for cells treated with MCL-3G1 plus TFMK.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
This study demonstrates that 14 kDa secretory gVPLA2 regulates integrin binding of eosinophils to ICAM-1 surrogate and augments stimulated secretion of LTC4 by a mechanism that does not rely upon activation of gIVA-PLA2. Prior investigations have implicated the role of epithelial cells in initiating asthma (31, 35) and atopic responsiveness (32). Although cytokines (IL-1beta, GM-CSF), chemokines (RANTES, IL-8) and other soluble factors (platelet-activating factor, platelet-derived growth factor) are secreted from epithelial cells, these compounds have a modest effect, if any, on eosinophil secretion of LTC4. To date, there have been no prior reports on the identity of a specific physiological trigger originating from epithelial cells for direct eosinophil activation. Secretory gVPLA2 has been associated with the secretion of LTB4 in neutrophils (21, 22) and synthesis of LTC4 in eosinophils (20), but the role of gVPLA2 in up-regulation of adhesion is not known. In these studies, we examined the effect of purified hVPLA2 and endogenously secreted gVPLA2 from activated epithelial cells on adhesion and secretion of LTC4 in human eosinophils. Our results demonstrate that hVPLA2 up-regulates cell surface CD11b expression (Fig. 1) and enhances binding of CD11b/CD18 to ICAM-1 (Fig. 2A) by a mechanism that does not require activation of gIVA-PLA2 (Fig. 5). The increase in beta2 integrin-mediated adhesion was specific for hVPLA2 because 1) other major forms of human secretory PLA2 including Ib, IIaPLA2, X, or Naja naja naja (Fig. 2B) and 2) W31A, an inactive putative mutant of hVPLA2, did not elicit integrin adhesion. The effect of hVPLA2 on cellular adhesion was blocked by the addition of mAb that is highly specific for gVPLA2 (MCL-3G1) (24, 25), or surface CD11b expression (clone Bear I) that is specific for beta2 integrin binding to ICAM-1 surrogate, or by a global secretory PLA2 inhibitor (LY311727).

We have established previously that Trp (W) (31) is the active binding site of hVPLA2 (1, 20, 21, 22, 23). Mutation of Trp to alanine (A) at site 31 (W31A) attenuated the surface CD11b expression and adhesion (Figs. 1 and 2), suggesting that interfacial membrane binding and hydrolytic activities of hVPLA2 are required for hVPLA2-mediated eosinophil adhesion.

We have shown previously that hVPLA2 acts on both the outer plasma membrane and the perinuclear membranes of human eosinophils after intercellular uptake by HSPG (20, 21, 22, 23). We have also reported previously that the internalization of exogenous hVPLA2 is a rapid process (5 min) and that internalized hVPLA2 is then translocated to the perinuclear membrane of cells (21, 22). This unique activity of hVPLA2 makes it difficult to analyze in the supernatant of activated cells containing gVPLA2. By contrast, gIIaPLA2 is measurable during cell activation because the preferential substrates for this secretory PLA2 are phosphatidylserine and phosphatidylethanolamine, which are located at the inner leaflet of cell membrane (2, 3, 4). In this study, we also demonstrated that Naja naja naja PLA2 and hXPLA2 (37, 38), both of which have greater PC-hydrolyzing activity than hVPLA2 (36, 37), did not induce eosinophil adhesion. This is likely the result of the low affinity of each secretory PLA2 for cell surface HSPG that mediate the internalization of secretory PLA2. Thus, it would seem that the main site of hVPLA2 activation in eosinophil adhesion is intracellular, most likely, perinuclear membranes (1, 20, 21, 22, 23, 36).

Lysophospholipids and free fatty acids are the two by-products of outer cell membrane hydrolysis caused by hVPLA2 activation in mammalian cells (20, 21, 22, 23). We chose LysoPC as a representative lysophospholipid because the major component of the outer leaflet of mammalian cell membrane is PC (1, 20, 21, 22, 36). Given exogenously, LysoPC caused increased beta2 integrin-mediated adhesion induced by hVPLA2 in eosinophils (Fig. 4). Although LysoPC was less efficacious than hVPLA2, an incremental increase in eosinophil binding to ICAM-1 surrogate was demonstrated as assessed by measurement of residual EPO activity. By contrast, arachidonic acid, which constitutes only 5% of total fatty acids incorporated into the phospholipids in the outer plasma membrane, was ineffective to elicit cell adhesion. Our data indicate that beta2 integrin adhesion caused by hVPLA2 could be mediated through release of LysoPC from membrane hydrolysis independent of arachidonic acid synthesis. However, we are unable at this time to establish LysoPC as the sole activator.

If hVPLA2 mainly acts on outer or intracellular membranes to produce fatty acids and on lysophospholipids to mediate eosinophil adhesion, it may not need to activate intercellular gIVA-PLA2 to initiate integrin adhesion or cysteinyl LT synthesis. Our current data are consistent with this supposition. A selective gIVA-PLA2 inhibitor, TFMK, which blocks FMLP-induced (15, 20, 21), IL-5-induced (14, 16), or eotaxin-induced (16) adhesion of eosinophils, did not inhibit hVPLA2-mediated eosinophil adhesion. Furthermore, exogenous hVPLA2 did not cause phosphorylation of gIVA-PLA2 in human eosinophils. ERK1/2 is an activator of gIVA-PLA2 and has been implicated in the control of cell migration (14, 15); p38 MAPK is constitutively expressed in eosinophils but does not regulate integrin adhesion (14). gIVA-PLA2 can be phosphorylated by ERK1/2 (14, 15, 17) or by protein kinase C (39), but ERK1/2 has a much larger effect on gIVA-PLA2 activity and phosphorylation. Activation of ERK1/2 has been reported after beta1/beta2 integrin engagement with ICAM-1, VCAM-1, or fibronectin in a variety of inflammatory cells (14, 15, 33). However, our Western blot analysis demonstrated that stimulation of eosinophils by hVPLA2 induced phosphorylation of neither ERK1/2 (Fig. 5A) nor gIVA-PLA2 (Fig. 5B).

It is important to note some limitations of our findings. The signaling and regulating mechanisms controlling movement of eosinophils play an important role in the inflammatory process. Focal clustering of CD11b plays a key role in firm adhesion (28). Our results demonstrate a distinct clustering of CD11b (Fig. 6) in response to hVPLA2 activation, suggesting that adhesion is mediated through up-regulation of this integrin.

In these studies, expression of gVPLA2 in epithelial cells first was established by Western blot (25) and flow cytometric analysis (Fig. 7A). Translocation of endogenously secreted gVPLA2 from activated epithelial cells to adhering eosinophils was confirmed further by labeling the eosinophils with fluorogenic phospholipid PED6. This procedure is used to assess hydrolysis of PED6 incorporated with secreted gVPLA2 from epithelial cells (Fig. 7B). We chose this dye because it has been reported that secreted PLA2 have relatively high activity on PED6 in in vitro vesicle assay (21, 23, 30). The relative fluorescence intensities caused by incorporated endogenously secreted gVPLA2 from stimulated epithelial cells migrating to eosinophil surface membrane was visualized by confocal microscopy. Our subsequent studies also showed selective stimulation of epithelial cells causes release of gVPLA2 in concentrations sufficient to activate eosinophils in vitro.

Further studies demonstrated that gVPLA2 secreted from epithelial cells also causes up-regulation of cysteinyl LT synthesis by a mechanism that does not depend upon gIVA-PLA2 activation (Fig. 8). Simultaneous activation of gVPLA2 and gIVA-PLA2 causes augmented synthesis of LTC4 that is blocked partially by TFMK (a selective gIVA-PLA2 inhibitor) or MCL-3G1 (a specific mAb directed against gVPLA2). Combined administration of both MCL-3G1 and TFMK caused complete attenuation of LTC4 synthesis (Fig. 9C).

In summary, the present study has identified the epithelial cell as a natural source of gVPLA2, which may serve as an intercellular messenger protein to regulate the beta2 integrin-mediated adhesion of eosinophils to surface ICAM-1 adhesion and to cause synthesis of cysteinyl LTC4. This transcellular signaling pathway mediated by secreted gVPLA2 demonstrates that epithelial cells may communicate signals to eosinophils that up-regulate beta2 integrin-mediated adhesion and synthesis of cysteinyl LTC4. This novel signaling pathway does not require the activation of gIVA-PLA2, which previously has been reported to be essential for integrin adhesion (14, 15) and generation of lipid mediators (11, 13, 17, 21, 22). It is likely that eosinophil beta2 integrin binding to ICAM-1 surrogate protein caused by hVPLA2 is largely mediated through generation of lysophospholipid, particularly LysoPC. Our data also suggest that gIIaPLA2 is not likely to be a mediator of airway inflammation as proposed previously.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by Grant HL-46368 and Specialized Center of Research Grant HL-56399 from the National Heart, Lung, and Blood Institute (to A.R.L.), by GlaxoSmithKline Center of Excellence (to A.R.L.), by Grant AI-5209 from the National Institute of Allergy and Infectious Diseases (to X.Z.), and by Grant GM52598 from the National Institutes of Health (to W.C.). Back

2 Address correspondence and reprint requests to Dr. Alan R. Leff, Section of Pulmonary and Critical Care Medicine, M6076, Department of Medicine, The University of Chicago, 5841 South Maryland Avenue, Chicago, IL 60637. E-mail address: aleff{at}medicine.bsc.uchicago.edu Back

3 Abbreviations used in this paper: PLA2, phospholipase A2; gVPLA2, group V PLA2; hVPLA2, human group V PLA2; gIVA-PLA2, cytosolic group IVA PLA2; EPO, eosinophil peroxidase; PC, phosphatidylcholine; TFMK, trifluoromethylketone; ET-1, endothelin-1; LT, leukotriene; LysoPC, lysophosphatidylcholine; HSPG, heparan sulfate proteoglycan; MFI, mean fluorescence intensity. Back

Received for publication November 10, 2005. Accepted for publication April 10, 2006.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Cho, W.. 2000. Structure, function, and regulation of group V phospholipase A2. Biochim. Biophys. Acta 1488: 48-58. [Medline]
  2. Janssen, M. J., L. Vermeulen, H. A. Van der Helm, A. J. Aarsman, A. J. Slotboom, M. R. Egmond. 1999. Enzymatic properties of rat group IIA and V phospholipase A2 compared. Biochim. Biophys. Acta 1440: 59-72. [Medline]
  3. Singer, A. G., F. Ghomashchi, C. Le Calvez, J. Bollinger, S. Bezzine, M. Rouault, M. Sadilek, E. Nguyen, M. Lazdunski, G. Lambeau, M. H. Gelb. 2002. Interfacial kinetic and binding properties of the complete set of human and mouse groups I, II, V, X, and XII secreted phospholipase A2. J. Biol. Chem. 277: 48535-48549. [Abstract/Free Full Text]
  4. Gesquiere, L., W. Cho, P. V. Subbaiah. 2002. Role of group IIa and group V secretory phospholipase A2 in the metabolism of lipoproteins: substrate specificities of the enzymes and the regulation of their activities by sphingomyelin. Biochemistry 41: 4911-4920. [Medline]
  5. Stewart, A., M. Gosh, D. M. Spencer, C. C. Leslie. 2002. Enzymatic properties of human cytosolic phospholipase A2{gamma}. J. Biol. Chem. 277: 29526-29536. [Abstract/Free Full Text]
  6. Yang, H. C., M. Mosior, C. A. Johnson, Y. Chen, E. A. Dennis. 1999. Group-specific assay that distinguish between the four major types of mammalian phospholipase A2. Anal. Biochem. 269: 278-288. [Medline]
  7. Kudo, I., M. Murakami. 2002. Phospholipase A2 enzymes. Prostaglandins Other Lipid Mediat. 68: 3-58.
  8. Nagase, T., N. Uozumi, S. Ishii, K. Kume, T. Izumi, Y. Ouchi, T. Shimizu. 2000. Acute lung injury by sepsis and acid aspiration: a key role for cytosolic phospholipase A2. Nat. Immunol. 1: 42-46. [Medline]
  9. Jenkins, C. M., D. J. Mancuso, W. Yan, H. F. Sims, B. Gibson, R. W. Gross. 2004. Identification, cloning, expression, and purification of three novel human calcium-independent phospholipase A2 family members possessing triacylglycerol lipase and acylglycerol transacylase activities. J. Biol. Chem. 279: 48968-48975. [Abstract/Free Full Text]
  10. Manguikian, A. D., S. E. Barbour. 2004. Cell cycle dependence of group VIA calcium-independent phospholipase A2 activity. J. Biol. Chem. 279: 52881-52892. [Abstract/Free Full Text]
  11. Ghosh, M., A. Stewart, D. E. Tucker, J. V. Bonventre, R. C. Murphy, C. C. Leslie. 2004. Role of cytosolic phospholipase A2 in PGE2 production in lung fibroblast. Am. J. Respir. Cell Mol. Biol. 30: 91-100. [Abstract/Free Full Text]
  12. Bonventre, J. V., Z. Huang, M. R. Takeri, E. O’Leary, E. Li, M. A. Moskowitz, A. Sapristein. 1997. Reduced fertility and post ischaemic brain injury in mice deficient in cytosolic phospholipase A2. Nature 390: 622-625. [Medline]
  13. Fonteh, A. N., G. Atsumi, T. LaPorte, F. H. Chilton. 2000. Secretory phospholipase A2 receptor-mediated activation of cytosolic phospholipase A2 in murine bone marrow-derived mast cells. J. Immunol. 165: 2773-2782. [Abstract/Free Full Text]
  14. Zhu, X., B. Jacobs, E. Boetticher, S. Myou, A. Meliton, H. Sano, A. T. Lambertino, N. M. Muñoz, A. R. Leff. 2002. IL-5-induced integrin adhesion of human eosinophils caused by ERK1/2-mediated activation of cPLA2. J. Leukocyte Biol. 72: 1046-1053. [Abstract/Free Full Text]
  15. Zhu, X., N. M. Muñoz, K. P. Kim, H. Sano, W. Cho, A. R. Leff. 1999. Cytosolic phospholipase A2 activation is essential for beta1/beta2-integrin-dependent adhesion of human eosinophils. J. Immunol. 163: 3423-3429. [Abstract/Free Full Text]
  16. Liu, J., N. M. Muñoz, A. Y. Meliton, X. Zhu, A. T. Lambertino, C. Xu, S. Myo, S. Myou, E. Boetticher, M. Johnson, et al 2004. beta2-integrin adhesion caused by eotaxin but not IL-5 is blocked by PDE-4 inhibition and beta2-adrenoceptor activation in human eosinophils. Pulm. Pharmacol. Ther. 17: 73-79. [Medline]
  17. Myou, S., A. R. Leff, S. Myo, E. Boetticher, J. Tong, A. Y. Meliton, J. Liu, N. M. Muñoz, X. Zhu. 2003. Blockade of inflammation and airway hyperresponsiveness in immune-sensitized mice by dominant-negative phosphoinositide 3-kinase-TAT. J. Exp. Med. 198: 1573-1582. [Abstract/Free Full Text]
  18. Sapirstein, A., J. V. Bonventre. 2000. Specific physiological roles of cytosolic phospholipase A2 as defined by gene knockouts. Biochim. Biophys. Acta 1488: 139-148. [Medline]
  19. Chen, Y., E. A. Dennis. 1998. Expression and characterization of human group V phospholipase A2. Biochim. Biophys. Acta 1394: 57-64. [Medline]
  20. Muñoz, N. M., Y. J. Kim, A. Y. Meliton, K. P. Kim, S.-K. Han, E. Boetticher, E. O’Leary, S. Myou, X. Zhu, J. V. Bonventre, et al 2003. Human group V phospholipase A2 induces group IVA phospholipase A2-independent cysteinyl leukotriene synthesis in human eosinophils. J. Biol. Chem. 278: 38813-38820. [Abstract/Free Full Text]
  21. Kim, Y. J., K. P. Kim, S. K. Han, N. M. Muñoz, X. Zhu, H. Sano, A. R. Leff, W. Cho. 2002. Group V phospholipase A2 induces leukotriene biosynthesis in human neutrophils through the activation of group IVA phospholipase A2. J. Biol. Chem. 277: 36479-36488. [Abstract/Free Full Text]
  22. Kim, K. P., J. D. Rafter, L. Bittova, S. K. Han, Y. Snitko, N. M. Muñoz, A. R. Leff, W. Cho. 2001. Mechanism of human group V phospholipase A2 (PLA2)-induced leukotriene biosynthesis in human neutrophils: a potential role of heparan sulfate binding in PLA2 internalization and degradation. J. Biol. Chem. 276: 11126-11134. [Abstract/Free Full Text]
  23. Kim, K. P., S. K. Han, M. Hong, W. Cho. 2000. The molecular basis of phosphatidylcholine preference of human group-V phospholipase A2. Biochem. J. 348: 643-647.
  24. Muñoz, N. M., K. P. Kim, S. K. Han, E. Boetticher, A. I. Sperling, H. Sano, X. Zhu, W. Cho, A. R. Leff. 2000. Characterization of monoclonal antibodies specific for 14-kDa human group V secretory phospholipase A2 (hVPLA2). Hybridoma 19: 171-176. [Medline]
  25. Muñoz, N. M., E. Boetticher, A. I. Sperling, K.-P. Kim, A. Y. Meliton, X. Zhu, A. Lambertino, W. Cho, A. R. Leff. 2002. Quantitation of secretory group V phospholipase A2 in human tissues by sandwich enzyme-linked immunosorbent assay. J. Immunol. Methods 262: 41-51. [Medline]
  26. Koduri, R. S., S. F. Baker, Y. Snitko, S. K. Han, W. Cho, D. C. Wilton, M. H. Gelb. 1998. Action of human group IIa secreted phospholipase A2 on cell membrane. J. Biol. Chem. 273: 32142-32153. [Abstract/Free Full Text]
  27. Zhu, X., R. Subbaraman, H. Sano, B. Jacobs, A. Sano, E. Boetticher, N. M. Muñoz, A. R. Leff. 2000. A surrogate method for assessment of beta2-integrin-dependent adhesion of human eosinophils to ICAM-1. J. Immunol. Methods 240: 157-164. [Medline]
  28. Myou, S., X. Zhu, E. Boetticher, S. Myo, A. Y. Meliton, A. Lambertino, N. M. Muñoz, A. R. Leff. 2003. Blockade of focal clustering and active conformation in beta2-integrin-mediated adhesion of eosinophils to intercellular adhesion molecule-1 caused by transduction of HIV TAT-dominant negative Ras. J. Immunol. 169: 2670-2676.
  29. Panja, A., S. Goldberg, L. Eckmann, P. Krishen, L. Mayer. 1998. The regulation and functional sequence of proinflammatory cytokine binding on human intestinal epithelial cells. J. Immunol. 161: 3675-3684. [Abstract/Free Full Text]
  30. Markova, M., R. A. Koratkar, K. A. Silverman, V. E. Sollars, M. M. Pellini, R. Walters, J. P. Palazzo, A. M. Buchberg, L. D. Siracusa, S. A. Faber. 2005. Diversity in secreted PLA2-IIA activity in inbred mouse strains that are resistant or susceptible to ApcMin/+ tumorigenesis. Oncogene 24: 6450-6458. [Medline]
  31. Shebani, E., S. Shahana, C. Janson, G. M. Roomans. 2005. Attachment of columnar airway epithelial cells in asthma. Tissue Cell 37: 145-152. [Medline]
  32. Wark, P. A., S. L. Johnston, F. Bucchieri, R. Powell, S. Puddicombe, V. Laza-Stanca, S. T. Holgate, D. E. Davies. 2005. Asthmatic bronchial epithelial cells have a deficient innate immune response to infection with rhinovirus. J. Exp. Med. 201: 937-947. [Abstract/Free Full Text]
  33. Sano, H., X. Zhu, A. Sano, E. Boetticher, T. Shioya, B. Jacobs, N. M. Muñoz, A. R. Leff. 2001. Extracellular signal-regulated kinase 1/2-mediated phosphorylation of cytosolic phospholipase A2 is essential for human eosinophils to fibronectin. J. Immunol. 166: 3515-3521. [Abstract/Free Full Text]
  34. Stake, Y., B. L. Diaz, B. Balestrieri, B. K. Lam, Y. Kanaoka, M. J. Grusby, J. P. Arm. 2004. Role of group V phospholipase A2 in zymosan-induced eicosanoid generation and vascular permeability revealed by targeted gene disruption. J. Biol. Chem. 279: 16488-16494. [Abstract/Free Full Text]
  35. Frieri, M.. 2004. Airway epithelial cells release of cytokines: modulation by various therapeutic agents. Allergy Asthma Proc. 25: 387-393. [Medline]
  36. Kim, Y. J., K. P. Kim, H. J. Rhee, S. Das, J. D. Rafter, Y. S. Oh, W. Cho. 2002. Internalized group V secretory phospholipase A2 acts on the perinuclear membranes. J. Biol. Chem. 277: 9358-9365. [Abstract/Free Full Text]
  37. Pan, Y. H., B. Z. Yu, A. G. Singer, F. Ghomashchi, G. Lambeau, M. H. Gelb, J. Bollinger, M. K. Jain, B. J. Bahnson. 2002. Crystal structure of human group X secreted phospholipase A2: electrostatically neutral interfacial binding surface targets zwitterionic membranes. J. Biol. Chem. 277: 29086-29093.