Leukotriene B4 (LTB4) plays a crucial role in the recruitment of neutrophils into the pleural space. We identified for the first time the mechanisms by which LTB4 interacts with mesothelial cells and recruits neutrophils in the pleural compartment. Primary pleural mesothelial cells express both the proinflammatory receptor for LTB4 BLT2, and the anti-inflammatory receptor for LTB4, PPARα. Parapneumonic pleural effusions highly increase BLT2 expression and, via BLT2 activation, increase the adhesion between mesothelial cells and neutrophils and the expression of ICAM-1 on mesothelial cells. The block of PPARα further increases both cell adhesion and ICAM-1 expression. BLT2 activation promotes the activation, on mesothelial cells, of STAT-1 but not the activation of NF-κB transcription factor. The increase of ICAM-1 expression is achieved via increased tyrosine phosphorylation activity since herbimycin, a tyrosine kinase inhibitor, reduces and since Na orthovanadate, a tyrosine phosphatase inhibitor, further increases ICAM-1 expression. This study demonstrates that pleural mesothelial cells, expressing both proinflammatory and anti-inflammatory LTB4 receptors, are able to mount an integrated response to LTB4 with a prevalence of BLT2 activities in the presence of an inflammatory milieu within the pleura.
Leukotrine B4 (LTB4)3 in the pleural space is an important proinflammatory agent, released by both resident and not resident pleural cells, which contributes to the directed migration of activated neutrophils into the pleural space during inflammatory events (1). LTB4 is able to interact with two cell-surface receptors (BLT1 and BLT2) and with a nuclear receptor called PPARα. The interaction of LTB4 with cell-surface receptors induces LTB4-mediated inflammatory activities while the interaction with PPARα induces the catabolism of LTB4 and, in turn, limits LTB4-correlated inflammation (2).
The pleural mesothelium actively promotes extravasation of leukocytes by participating in adhesive interactions and by secreting chemoactractant molecules (1, 3). The increased interactions between β-2 integrins on neutrophils (specifically CD11b or CD18), and ICAM-1 on pleural mesothelial cells may lead to the increase in the adhesive interaction between mesothelial cells and the neutrophils and, in turn, to the increased neutrophil recruitment into the pleural space (4, 5, 6).
Neutrophils are the predominant leukocytes that are recruited into the pleural space during the early steps of many inflammatory responses (3) and are the prevalent cell population of parapneumonic effusions (1, 7). The underlying cellular and molecular events involved in the movement of neutrophils across the mesothelium into the pleural cavity are not completely elucidated. Furthermore, it is not known whether LTB4, released during the early events of inflammation, contributes to the adhesiveness of mesothelial cells and whether mesothelial cells are responsive to LTB4 in a tightly controlled manner. The aims of the present study were to evaluate whether: 1) mesothelial cells express the cell-surface receptors for LTB4 (BLT1 and BLT2), as well as the nuclear receptor for LTB4 (PPARα); 2) LTB4, and parapneumonic pleural fluids (PF) modulate the adhesion between pleural mesothelial cells and neutrophils and whether this phenomenon is associated to the modulation of ICAM-1 expression on pleural mesothelial cells; and 3) the modulation of the ICAM-1 expression is related to the activation of specific transcription factors.
Materials and Methods
PFs were collected by thoracentesis from hospitalized patients with pneumonia (n = 8, age range 35 to 73 years). All subjects gave informed written consent and the study was approved by the institutional review board for human studies. The effusions were first classified as exudates by meeting at least one of the criteria described by Light (8). Standard clinical, laboratory, and radiological investigations were used to establish the diagnosis for each patient. Parapneumonic effusions were defined as exudates with a glucose concentration >60 mg/dl and pH >7.2 and no organism seen on Gram stain or found on PF culture in patients with pneumonia. No patients were undergoing anti-inflammatory or steroid therapies. The fluids were drawn into polypropylene bags containing heparin (10–20 IU/ml). The fluids were subsequently centrifuged at 400 × g for 10 min and cell-free fluids were immediately frozen at −70°C until they were used for subsequent analyses. The concentrations of LTB4 in parapneumonic pleural effusions were 236 ± 51 pg/ml.
Mesothelial cell cultures
Primary pleural mesothelial cells recovered from transudative pleural effusions, as previously described (9), were used in this study. The cells were cultured with Media 199 (Life Technologies) containing 10% FBS (Life Technologies), penicillin (100 U/ml) (Life Technologies), and streptomycin (100 μg/ml) (Life Technologies). Primary pleural mesothelial cells were used between four and eight passages.
Mesothelial cells were stimulated with synthetic LTB4 (0.2 ng/ml) (Sigma-Aldrich), with PPARα agonist (WY-14643;10 μM) (Sigma-Aldrich), with parapneumonic PFs, and with a BLT2 antagonist (LY255283) (1 μM) or with a PPARα antagonist (MK886) (1 μM) (Cayman Chemical). The BLT2 and PPARa antagonists were added 1 h before cell stimulation. In preliminary experiments, we identified 18 h of stimulation as optimal time point for evaluating adhesion molecules and 1 h of stimulation as optimal time point for evaluating the STAT-1 or NF-κB nuclear translocation or for evaluating phosphorylated STAT-1 expression.
In some experiments, ICAM-1 expression on mesothelial cells was evaluated after the preincubation with herbimycin (Sigma-Aldrich) (2 μM; 2 h before cell stimulation) and Na orthovanadate (Sigma-Aldrich) (10 μM; 2 h before cell stimulation).
BLT1, BLT2, and PPARα expression
The expression of BLT1 in pleural mesothelial cells and in neutrophils and the expression of BLT2 and PPARα in pleural mesothelial cells was evaluated by flow cytometry using a FACStarPlus (BD Biosciences) analyser using rabbit polyclonal Abs anti-BLT1, anti-BLT2 and anti-PPARα (Cayman Chemical) (1/250; for 1 h) followed by a FITC conjugated anti-rabbit IgG (DakoCytomation). To evaluate the expression of BLT1, BLT2, and of PPARα, before incubation with rabbit polyclonal Abs, cells were permeabilized using a commercial fix-perm cell permeabilization kit (Caltag Laboratories). Negative controls were performed using rabbit immunoglobulins negative control (DakoCytomation). Peripheral blood neutrophils, isolated as previously described (1), were used as positive control for BLT1 expression. Data are expressed as geomean fluorescence intensity.
Mesothelial cell-neutrophil adhesion assay
Cell-adhesion assay was performed by a method previously described with minor modifications (10). In brief, mesothelial cells were seeded in 96-well plates at half-confluency and kept for 18 h alone, with synthetic LTB4, with PF supernatants, and with PFs supernatants in the presence of anti-BLT2 receptor or of anti-PPARa receptor. Peripheral blood neutrophils were isolated as previously described (1), labeled for 30 min with 20 μM of fluorocromic dye CMFDA (Molecular Probes), and added at the concentration 1 × 104 to the mesothelial layer for 20 min to promote adherence. The plates were washed and the number of adhering neutrophils was determined by measuring fluorescence at 525 nm in a Wallac Victor multilabel counter (Wallac). Cell adhesion was expressed as percentage of baseline.
Evaluation of ICAM-1 expression
The expression of ICAM-1 on the surface of mesothelial cells was determined by flow cytometry with a FACStarPlus analyzer (BD Biosciences). Fluorescein (FITC)-conjugated mouse anti-human ICAM-1 Ab (anti-CD54; clone 6.5B5; Dakopats) was used. Negative controls were performed using a FITC-conjugated mouse IgG1 (from Dakopats). Data are expressed as percentage of baseline geomean fluorescence intensity.
Western blot analysis for STAT-1 and NF-κB nuclear translocation
Western blot analysis for STAT-1 and NF-κB (11) was performed as previously described with minor modifications. In brief, mesothelial cells were lysed (10 mM Tris-HCl (pH 7.4); 50 mM NaCl; 5 mM ethylenediaminetetraacetic acid; 1% Nonidet P-40; 10 μg/ml PMSF) and protein were extracted. To study STAT-1 and NF-κB nuclear translocation, the protein extracts were treated to separate the cytoplasmic and nuclear protein fractions by using a commercial kit following the manufacturer’s directions (Pierce). An amount of 50 μg of total proteins was subjected to SDS-PAGE on 4–12% gradient gels (Novex) and blotted onto nitrocellulose membranes. These were blocked with PBS containing 3% BSA, 0.1% Tween 20, and then probed with a polyclonal Ab directed against human STAT-1 (Santa Cruz Biotechnology) (1/200; 1 h), human phospho-STAT-1 (Tyr701) (Cell Signaling Technology) (1/100, overnight), and human NF-κB (Santa Cruz Biotechnology) (1/100; 1 h). Revelation was performed with an ECL system (NEN) followed by autoradiography. Negative controls were performed including an isotype-control Ab. During each experiment, the detection of β-actin as housekeeping protein was performed. Data underwent densitometric analysis and were expressed as densitometric arbitrary units by normalization with the density of the band obtained for β-actin.
Chromatin immunoprecipitation (ChiP) analysis
ChiP analysis was performed using the EZ-ChiP kit (Upstate-Millipore) following the manufacturer’s directions. The primary mesothelial cells stimulated with parapneumonic PFs were treated with formaldehyde and the crosslinked chromatin were sonicated to lengths spanning 200-1000 bp. The samples were precleared with 60 μl of protein A agarose and then incubated with a polyclonal Ab anti-human STAT1. Immunocomplexes were precipitated using protein A agarose. After washing, DNA fragments were isolated and purified with columns. PCR was performed using primers spanning the promoter region 3537 of ICAM-1 gene using the primers: 5′ CAC AGA GTG AGA CTC CAT C 3′ (forward) and 5′ TGT TGT CCA GGC TGG AGT A 3′ (reverse).
Data are expressed as mean counts ± SD. The comparison between different experimental conditions was evaluated by ANOVA corrected with the Bonferroni test. For cell adhesion assay, paired t test was also used. p < 0.05 was accepted as statistically significant.
Primary mesothelial cells express membrane and nuclear receptor for LTB4
Because it has been previously demonstrated that LTB4 is present in exudative pleural effusions and acts as a potent chemotactic agent for neutrophils during pleural inflammation (1), it was first evaluated whether primary mesothelial cells expressed proinflammatory cell-surface LTB4 receptors and/or anti-inflammatory nuclear LTB4 receptor. Mesothelial cells coexpressed the anti-inflammatory nuclear receptor for LTB4 PPARα and the proinflammatory membrane receptor for LTB4 BLT2, suggesting that mesothelial cells are responsive to pleural LTB4. As expected, mesothelial cells did not express the BLT1 receptor whose expression has been demonstrated to be limited to leukocytes (12) (Fig. 1,A–C). Peripheral blood neutrophils, used as positive controls for BLT1 expression, highly expressed BLT1 (Fig. 1 A).
Parapneumonic PFs increase the expression of BLT2 but not of PPARα receptors
The effect of parapneumonic PFs on the expression of BLT2 as well as of PPARα by primary pleural mesothelial cells was explored. Following the incubation with exudative parapneumonic PFs, pleural mesothelial cells showed a significantly higher increase of BLT2 expression in comparison to the increase observed for PPARα expression, suggesting that the presence of an inflammatory milieu induces more proinflammatory rather than anti-inflammatory receptors for LTB4 on normal pleural mesothelial cells (Fig. 2).
Parapneumonic PFs increase the adhesiveness between mesothelial cells and neutrophils
The recruitment of neutrophils into the pleural space is associated to the generation of a chemotactic gradient as well as to the increased adhesion between neutrophils and mesothelial cells (4). Because exudative parapneumonic PFs are able to increase BLT2 receptors and because LTB4 has been demonstrated to increase the adhesion between neutrophils and endothelial cells in other compartments (6), the influence of PFs containing high levels of LTB4 (i.e., parapneumonic PFs) on the adhesion between pleural mesothelial cells and neutrophils was assessed. Interestingly, the addition of parapneumonic PFs (1) to mesothelial cell-neutrophil cocultures significantly increased the adhesion between these two cell types, and this phenomenon was reverted by the presence of a specific antagonist of BLT2 receptor (Fig. 3) and by the activation of PPARα with a specific agonist. As expected, synthetic LTB4 also was able to increase the adhesiveness between these two cell types (Fig. 3). Differently, the presence of a specific antagonist of PPARα further increased the adhesion between mesothelial cells and neutrophils (Fig. 3). These findings demonstrate that LTB4 present in parapneumonic PFs contributes to the increased adhesiveness of mesothelium toward neutrophils during pleural inflammatory processes via a prevalent activation of the proinflammatory BLT2 receptor.
Parapneumonic PFs induce the increase of ICAM-1 expression on mesothelial cells
The next step was to determine whether the modulation of adhesion between mesothelial cells and neutrophils due to LTB4 was associated to the modulation of ICAM-1 on mesothelial cells. Interestingly, synthetic LTB4 and parapneumonic PFs were both able to increase the expression of ICAM-1 on mesothelial cells via the activation of BLT2 receptor because an antagonist of BLT2 receptor blocked this phenomenon. An antagonist of PPARα further increased while an agonist of PPARα decreased the expression of ICAM-1 on mesothelial cells, suggesting that the activation of this receptor may limit the proinflammatory activities promoted by LTB4 via the activation of BLT2 receptors (Fig. 4).
Parapneumonic PFs promote preferentially STAT-1 activation on mesothelial cells
Experiments were designed to better understand the mechanisms underlining the activity of BLT2 on ICAM-1 expression. Because the increase of ICAM-1 expression in other compartments is generated by the activation of specific transcription factors, the effect of parapneumonic PFs on NF-κB (Fig. 5) and on STAT-1 (Fig. 6) activation was explored. Synthetic LTB4 and parapneumonic PFs did not significantly modify the nuclear translocation of NF-κB (Fig. 5). Parapneumonic PFs promoted the nuclear translocation of STAT-1 (Fig. 6, A and C). The block of BLT2 by a specific antagonist abrogated this translocation, demonstrating that the activation of BLT2 leads to STAT-1 nuclear translocation (Fig. 6, A and C). A PPARα antagonist did not significantly modify STAT-1 activation (Fig. 6, A and C). Synthetic LTB4 increased STAT-1 nuclear translocation, while the block of BLT2 receptor prevented this event (data not shown). The activation of STAT-1 was further confirmed by p(Tyr701) STAT-1 evaluation (Fig. 6, B and C). Moreover, we conducted ChIP assays with Abs specific to Stat1 (Fig. 7) and the results of these experiments showed that STAT-1 could be detected on the promoter region of ICAM-1 in mesothelial cells after incubation with PF.
Increased tyrosine phosphorylation leads to the increased ICAM-1 expression
A first step in the activation of STAT-1 is the phosphorylation of tyrosine residues, and this event is tightly regulated by protein tyrosine kinases and protein tyrosine phosphatases (13, 14). The effect of herbimycin (a tyrosine kinases inhibitor) and of sodium ortho vanadate (a phosphotyrosine phosphatases inhibitor) on the expression of ICAM-1 was determined. Herbimycin reduced the increase of ICAM-1 expression (Fig. 8) due to synthetic LTB4 and to parapneumonic PFs, while sodium ortho vanadate further increased these phenomena (Fig. 8), suggesting that tyrosine phosphorylation is required for increasing ICAM-1 expression.
LTB4 in the pleural space is an important proinflammatory agent, released by both resident and not resident pleural cells, which contributes to the migration of activated neutrophils into the pleural space during pleural inflammation (1).
The present study demonstrates for the first time that the response of the pleural mesothelium to LTB4 is the result of a balance between the activation of receptors for LTB4 with a proinflammatory outcome (BLT2) and the activation of a different receptor for LTB4 with an anti-inflammatory outcome (PPARα). The prevalent activation of proinflammatory LTB4 receptors in the presence of an inflammatory milieu within the pleura may lead to an up-regulation of the adhesion between neutrophils and the mesothelium resulting in neutrophil retention on the surface of mesothelial cells.
Two types of G protein-coupled transmembrane domain receptors (GPCR) for LTB4 have been identified: BLT1 and BLT2 with a different localization and a different binding activity. While BLT1 receptors are highly expressed on leukocytes and have a high binding affinity for LTB4, BLT2 receptors are present, other than on leukocytes, on structural cells in a variety of body compartments and have a low binding affinity for LTB4 (12).
PPARα, an orphan nuclear receptor, has also been reported to bind and to catabolize LTB4 and to limit inflammation since PPARα-deficient mice exhibit a prolonged inflammatory response when challenged with LTB4 (15, 16). PPARα activation inhibits NF-κB activation (17) and down-regulates chemoattractant production and neutrophils infiltration (18). Taken together these findings suggest that at the site of inflammation, the level and degradation of LTB4 determine the extent and the duration of an inflammatory response (19, 20).
We previously showed that pleural mesothelial cells are able to express 5-lipoxygenase and to release LTB4 upon LPS stimulation (1). Now, we demonstrate for the first time that pleural mesothelial cells express both BLT2 and PPARα receptors but not BLT1 receptors showing that the structural cells of the pleura are also able to respond to LTB4. The expression of these receptors was evaluated in a primary mesothelial cell line isolated from transudative PFs to evaluate the expression of these receptors in an ex vivo model.
It is crucially important to gain insights into these mechanisms since the inflammation within the pleural space has unique features that distinguish it from inflammation in other organs. In solid organs, inflammatory cells gain access to the tissue via the vasculature, adhere to the endothelium, migrate through it, and then gain access to the tissue. Once through the basement membrane, they percolate within the tissue to the site of the inflammatory focus. By contrast, inflammatory cells that enter the pleural space must, in addition to crossing the endothelial barrier, cross the mesothelial barrier and enter a fluid filled compartment that is the pleural space. Pleural mesothelial cells actively contribute to the recruitment and to the accumulation of leukocytes within the pleural compartment modulating their adhesive properties and secreting soluble mediators with chemotactic activities (1, 3).
To clarify these aspects, we performed experiments using parapneumonic effusions because we wondered whether the presence of an inflammatory milieu might lead to a prevalence of the proinflammatory pathway. Parapneumonic effusions are ideal for this purpose, because they have an intense inflammatory milieu; they contain large numbers of neutrophils and high concentrations of biologically active LTB4 in comparison to other exudative PFs (1). Additionally, parapneumonic effusions avoid the confounding influences of the microbes, and consequently, their virulence products that might be present in empyema. In this study for the first time, it is demonstrated that parapneumonic effusions are able to induce a prominent increase of the proinflammatory receptor for LTB4 (BLT2) and the adhesion between pleural mesothelial cells and neutrophils while they induce a modest increase of anti-inflammatory (PPARα) receptors for LTB4.
This phenomenon is associated to an increase of the constitutive expression of ICAM-1 by pleural mesothelial cells. It is conceivable that also an increase of β-2 integrins on neutrophils may contribute to the increased adhesion between pleural mesothelial cells and neutrophils because LTB4 is able to induce the expression of MAC-1 on neutrophils (21).
To further gain insight into these mechanisms, experiments with antagonists for BLT2 and for PPARα were included to mimic a situation in which LTB4 activates exclusively PPARα or BLT2 receptor, respectively. The finding that the block of BLT2 abrogates the increase of ICAM-1 and the cell adhesion induced by parapneumonic PFs and by LTB4, strongly suggests that the increased adhesiveness of pleural mesothelial cells during inflammatory processes in the pleural space may be induced by the activation of BLT2 receptors. The concentrations of LTB4 used in in vitro experiments were within the range of LTB4 concentrations in parapneumonic pleural effusions to reproduce a model of LTB4 receptors (BLT2 and PPARα) stimulation as much close as possible to the in vivo stimulation of these receptors. The block of PPARα further increases, while a PPARα agonist reduces, the expression of ICAM-1 and the cell adhesion, suggesting that the function of PPARα during pleural inflammation is to limit the proinflammatory responses including uncontrolled increased adhesiveness of pleural mesothelial cells. Mesothelial cells express ICAM-1 (22) and this molecule is shed from the surface of cells in a regulated manner (23). Although it has been demonstrated that shedding of ICAM-1 is a function of matrix metalloproteinase-9 (24) that in turn is regulated by LTB4 (25), and although mice lacking PPARα have increased expression of ICAM-1 in the lungs (26), it is not known whether the activation of PPARα limits the expression of ICAM-1 in pleural mesothelial cells increasing the shedding of ICAM-1 from the surface of cells.
The regulation of ICAM-1 expression is promoted by the activation of specific transcription factors. Parapneumonic PFs and synthetic LTB4 are able to induce STAT-1 activation via the activation of BLT2 receptors since the block of BLT2 negatively interferes with this phenomenon. On the other hand, the block of PPARα does not affect STAT-1 activation, suggesting that the effect of PPARα activation in ICAM-1 expression and cell adhesion is mediated by the activation of other transcription factors.
BLT2 is a trasmembrane receptor belonging to the GPCR. The activation of GPCR by their ligands may activate STAT-1 via two different pathways. One of them is the activation of STAT-1 by Src family kinases, and the other one is the activation of STAT1 by JAK1/2 (27, 28, 29). Tyrosine phosphorylation is a crucial event in STAT-1 activation. In the present study, the modulation of ICAM-1 expression by parapneumonic fluids and by synthetic LTB4 is linked to tyrosine phosphorylation events. Herbimycin, a tyrosin kinase inhibitor, reduced, while sodium ortho vanadate, a phosphotyrosine phosphatases inhibitor, further increased the ICAM-1 expression. These data are consistent with a previous report demonstrating that ICAM-1 in human T cells is regulated by tyrosine phosphorylation activity through STAT-1-dependent signaling pathways (30). Moreover, in our experimental model, the modulation of ICAM-1 expression was not associated to NF-κB but it was associated to STAT-1 activation as confirmed by chip assay showing that STAT-1 was located at the promoter of ICAM-1 gene. Consistently, it has been previously demonstrated that vanadate and peroxovanadium derivatives induce ICAM-1 expression via STAT-1 but not of NF-κB (31).
In conclusion, pleural mesothelial cells, expressing both proinflammatory transmembrane BLT2 receptor and anti-inflammatory nuclear PPARα receptor, may provide an integrated response to pleural LTB4 present in parapneumonic effusions. A prolonged and prevalent activation of proinflammatory pathways inside the inflamed pleural cavity may lead to a persistence of pleural inflammation.
Other natural agonists (fatty acids, ecosanoids, metabolites generated by hydrolysis of lipoproteins) (32) may activate these receptors and further studies are needed to assess the content of these other natural ligands in PFs.
The authors have no financial conflict of interest.
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.
This work was supported by the Italian National Research Council and by the Alberta Lung Association.
Abbreviations used in this paper: LTB4, leukotriene B4; GPCR, G protein-coupled transmembrane domain receptor; PF, pleural fluid; ChiP, chromatin immunoprecipitation.