Platelet-driven leukotriene C₄-mediated airway inflammation in mice is aspirin-sensitive and depends on T prostanoid receptors

Abstract

Cysteinyl leukotrienes (cysLTs) are bronchoconstricting lipid mediators that amplify eosinophilic airway inflammation by incompletely understood mechanisms. We recently found that leukotriene C₄ (LTC₄), the parent cysLT, potently activates platelets in vitro and induces airway eosinophilia in allergen sensitized and challenged mice by a platelet- and type 2 cysLT receptor (CysLT₂R)-dependent pathway. We now demonstrate that this pathway requires production of thromboxane A₂ (TXA₂) and signaling through both hematopoietic and lung tissue-associated T prostanoid (TP) receptors. Intranasal administration of LTC₄ to ovalbumin (OVA) sensitized C57BL/6 mice markedly increased the numbers of eosinophils in the bronchoalveolar lavage (BAL) fluid, while simultaneously decreasing the percentages of eosinophils in the blood by a TP receptor-dependent mechanism. LTC₄ upregulated the expressions of intracellular adhesion molecule-1 (ICAM)-1 and vascular cell adhesion molecule (VCAM)-1 in an aspirin-sensitive and TP receptor-dependent manner. Both hematopoietic and non-hematopoietic TP receptors were essential for LTC₄ to induce eosinophil recruitment. Thus, the autocrine and paracrine functions of TXA₂ act downstream of LTC₄/CysLT₂R signaling on platelets to markedly amplify eosinophil recruitment through pulmonary vascular adhesion pathways. The findings suggest applications for TP receptor antagonists in cases of asthma with high levels of cysLT production.

Introduction

Cysteinyl leukotrienes (cysLTs) are a class of potent inflammatory mediators derived from arachidonic acid that is metabolized by myeloid cells expressing 5-lipoxygenase (5-LO) and LTC₄ synthase (LTC₄S) (1). LTC₄, the short-lived parent of the cysLTs, is transported across the plasma membrane, and converted extracellularly to LTD₄ by enzymatic removal of a glutamic acid residue (2). LTD₄, a potent smooth muscle spasmogen, is rapidly metabolized to LTE₄ by removal of glycine (3). Circulating eosinophils, basophils, and monocytes, as well as tissue-associated mast cells, dendritic cells and macrophages all generate LTC₄ in response to activation. In inflammatory foci, cysLTs are detectable with a relative abundance of LTE₄>LTC₄>>LTD₄, reflecting differences in biologic half-lives (4). CysLTs mediate smooth muscle constriction (5), vascular leakage (6), and inflammatory cell recruitment (7). They are established therapeutic targets in asthma (8) and potential targets in cardiovascular disease (9). While LTD₄ and LTE₄ are extracellular metabolites that diffuse through the fluid phase before binding their receptors, LTC₄ may function at close range between the cells of origin and its targets after its transport to the cell surface and before its conversion to LTD₄.

CysLTs bind and activate at least three G protein coupled receptors (GPCRs), termed the types 1 and 2 cysLT receptors (CysLT₁R (10) and CysLT₂R (11), respectively), and GPR99 (12). CysLT₁R preferentially binds LTD₄ over LTC₄; CysLT₂R binds LTC₄ and LTD₄ with equal affinity; and GPR99 exhibits preferential binding of LTE₄. Both hematopoietic and non-hematopoietic cells express these three receptors in various combinations. Mice lacking LTC₄S are markedly protected from the development of eosinophilic pulmonary inflammation (13,14), a hallmark of asthma, but not all of the relevant cell targets of cysLTs are known. Studies using receptor null mice reveal that each cysLT serves highly ligand- and cell target-specific functions in innate immunity and allergic inflammation (12,15,16). Drugs that block CysLT₁R are modestly efficacious in the treatment of asthma (17), but there are no data on the effects of blocking CysLT₂R or GPR99 in human disease. Preclinical models are essential to understand the contributions from these other receptors to disease pathophysiology in order to determine whether they are potential therapeutic targets.

In addition to their role in hemostasis, activated platelets bind via P-selectin to circulating granulocytes, particularly to eosinophils, in inflammatory disease states (18). This binding enhances the expression and avidity of both β1 and β2 integrins on eosinophils (18,19), which facilitates firm adhesion to intracellular adhesion molecule (ICAM)-1 and vascular cell adhesion molecule (VCAM)-1. Hence, platelets, like cysLTs, can amplify eosinophil recruitment to the lung in mouse models of allergen-induced pulmonary inflammation (20). Platelets also express CysLT₁R and CysLT₂R (21), suggesting that their functions may be influenced by cysLTs. We recently discovered that LTC₄, but not LTD₄ or LTE₄, potently induced platelet P-selectin expression, the release of the granule-associated chemokine CXCL7, and the generation of thromboxane A₂ (TXA₂) entirely by signaling through CysLT₂R (22). Furthermore, in a model of eosinophilic pulmonary inflammation induced by the inhalation of low-dose ovalbumin (OVA) by sensitized mice, intranasal administration of LTC₄ markedly (∼10-fold) increased the numbers and percentages of eosinophils recovered from the bronchoalveolar lavage fluid by a CysLT₂R and platelet-dependent pathway. Because of the potential pathobiologic significance of this pathway in eosinophilic inflammation, we sought to understand the mechanisms by which LTC₄ amplifies eosinophilic inflammation in this model. We now report that platelet-derived TXA₂ and its T prostanoid (TP) receptor are central to the amplification of pulmonary eosinophilia by LTC₄, revealing potential therapeutic targets.

Materials and Methods

Animals

All animal protocols were approved by the Dana Farber Cancer Institute Animal Care and Utilization Committee. Tbxar^-/- (TP receptor-null) mice were obtained from Dr. Thomas Coffman (Duke University, Durham, NC) (23). WT C57BL/6 mice were purchased from Charles River. Animals were housed for at least 2 weeks at the Smith Building before experiments were done.

Reagents

Ovalbumin and PBS were obtained from Sigma-Aldrich (St. Louis, MO). ELISA kits for thromboxane B₂ and prostaglandins were purchased from Cayman Chemical (Ann Arbor, MI). The ELISA kit for mouse CXCL7 was from Abcam (Cambridge, MA). Goat anti-mouse ICAM-1 and Goat anti-mouse VCAM-1 antibodies were purchased from R&D Systems (Minneapolis, MN). Rabbit anti-mouse GAPDH was from Cell Signaling (Danvers, MA).

Immunization and challenge

The mice were sensitized i.p. on days 0 and 5 with Alum-precipitated chicken egg OVA (10 µg). On days 16-18, the mice received intranasal challenge of 2.2 nmol LTC₄ or vehicle. On days 17-19, mice were challenged by inhalation of 0.1% OVA (22,24). In some experiments, the mice received 1 mg/ml lysine-aspirin (Lys-ASA) in drinking water from day 15 to day 20. Twenty-four hours after the final OVA aerosol challenge, the mice were euthanized and exsanguinated. The lungs were lavaged three times with 0.7 ml PBS/5 mM EDTA. Bronchoalveolar lavage (BAL) fluid cells were cytocentrifuged onto slides, stained with Diff-Quick (Fisher Diagnostics, Middletown, VA), and differentially counted. In some experiments, the selective TP receptor antagonist SQ 29,548 was injected to mice each day (2 mg/kg of body weight, intraperitoneal), or lysine aspirin (1 mg/ml) was added to the drinking water during the period of LTC₄ challenge.

Flow cytometry

Whole heparinized blood obtained by cardiac puncture was kept at room temperature, and assayed within 1 hour of collection. 10μL of unstimulated blood was incubated with directly conjugated antibodies specific for CD41 and CD45, or appropriate isotype controls (all antibodies from BD Biosciences, San Jose CA) for 20 minutes and fixed in 1% paraformaldehyde. At least 20,000 CD45⁺ cells were recorded for each sample on a BD FACSAria™ Flow Cytometer and were analyzed with FlowJo (Tree Star, Ashland, OR). CD45⁺ leukocytes were classified as eosinophils or neutrophils according to their known scatter characteristics, relative autofluorescence, and levels of CD45 expression. They were assessed for the presence of adherent platelets by relative expression of CD41.

SDS PAGE immunoblotting

Lung protein was extracted with T-PER tissue protein extraction reagent (Thermo Fisher Scientific, Rockford, IL), Lysates were subjected to 12% SDS-PAGE and transferred to PVDF membranes. The membranes were incubated with antibodies against ICAM-1, VCAM-1 and GAPDH in PBS, 5% dry milk, and 0.1% Tween-20 (1:1,000) overnight at 4°C on shaker and then with secondary antibody (peroxidase-conjugated anti–goat or anti–rabbit). Bands were visualized with enhanced chemiluminescence (Thermo Fisher Scientific).

Bone marrow transfer

Five-week old sex-matched WT and tbxar^-/- mice were lethally irradiated with 1200 Rads (12 Gy) in 2 split doses, 4 hours apart. Within 24 hours from the irradiation the bone marrow of WT and tbxar^-/- donors were collected and 1×10⁷ nucleated cells were infused via the tail vein into sex-matched irradiated mice in 200 µl of PBS. As a result of the bone marrow transfer, four groups of chimeric mice were generated: WT bone marrow into WT mice (WT → WT), WT → tbxar^-/-, tbxar^-/- → WT, and tbxar^-/- → tbxar^-/-. Ten weeks after the injection, mice were exposed to OVA and saline or LTC₄ sensitization/challenge according to the same protocol used above and euthanized 24 hours after the last challenge. Peripheral blood cell counts were performed before the beginning of treatment to ensure full engraftment.

Immunohistology

Lung tissue was fixed in 10% neutral buffered formalin and embedded in paraffin. Tissue sections were incubated with rabbit anti-VCAM-1 (Santa Cruz Biotechnology Ab H-276/sc8304) or non-immune rabbit IgG, and bound Abs were visualized with Dako EnVision+System-HRP and AEC chromogen (3-amino-9-ethylcarbazole). Other sections were incubated with rat anti-ICAM-1 mAb (Biolegend clone YN1/1.7.4) or isotype control, and bound Abs were visualized with the Rat on Mouse HRP-Polymer system (Biocare Medical) and diaminobenzidine. Platelets were visualized with rabbit IgG anti-mouse CD41 (Ab H-160; Santa Cruz Biotechnology) or nonimmune rabbit IgG for 1 h at room temperature. Ab binding was visualized with an avidin-biotin complex (Santa Cruz Biotechnology)(25). All sections were counterstained with Gill's hematoxylin #2.

Statistical analysis

Data are expressed as ±SEM from at least 10 mice from at least two experiments, except where otherwise indicated. Analyses were performed with Prism software (Graphpad). Differences between two treatment groups were assessed using Student t test, and differences among multiple groups were assessed using one-way ANOVA and Bonferroni post hoc test. P<0.05 was considered statistically significant. * p<0.05, ** p<0.01, *** p<0.001.

Results

Intrapulmonary LTC₄ challenges induce the release of platelet-derived products in vivo

To determine whether LTC₄ directly activated platelets in the lung, WT C57BL/6 mice were sensitized with OVA intraperitoneally on days 0 and 5, then challenged with 0.1% OVA on days 18-20 for 30 min each day. Twelve hours before each OVA challenge, mice received intranasal LTC₄ or a buffer control. Twenty four hours after the last OVA challenge, mice were euthanized, BAL fluid was collected, and the lungs were analyzed histologically. Compared with BAL fluid from the mice that received OVA + vehicle, the BAL fluid from the mice receiving OVA + LTC₄ contained significantly higher numbers and percentages of eosinophils (Fig. 1A), as well as increased numbers and percentages of neutrophils and lymphocytes (though at much lower levels than eosinophils, not shown). BAL fluid from LTC₄-treated mice also contained higher levels of TXB₂ (a stable metabolite of TXA₂) and the platelet-derived chemokine CXCL7 (Fig. 1B and 1C, respectively) than did BAL fluid from mice challenged with OVA alone. Histologic exam verified that LTC₄ treated mice had substantially more bronchovascular cellular infiltrates and goblet cell metaplasia than did mice challenged with OVA alone (not shown). Immunohistology revealed CD41+ platelets in the bronchial submucosa of the OVA-treated mice that appeared more prominent in the lungs of the LTC₄-treated mice than in those challenged with OVA alone (Supplemental Fig. 1). In the absence of OVA inhalaton, challenge of sensitized mice with LTC₄ did not elicit any cellular influx (Supplemental Fig. 2).

Figure 1.

Effect of LTC₄ on eosinophilic pulmonary inflammation and platelet activation. OVA-sensitized C57BL/6 were challenged with 0.1% OVA on three successive days, with or without inhalation of LTC₄ (2.2 nmol) 12 h before each challenge. BAL fluid was collected 24 h after the last OVA challenge. A. Total cell counts, total eosinophils, and percentage of eosinophils and in BAL fluids. B. Levels of TXB₂ in BAL fluid lipid fractions. C. Levels of BAL fluid CXCL7. Results in A-C are from ten mice/group from two separate experiments. * p<0.05, *** p<0.001.

Potentiation of lung eosinophilia by LTC₄ is sensitive to treatment with aspirin

To determine whether the potentiation of lung eosinophilia in response to LTC₄ involved contributions from cyclooxygenase (COX) pathway products, mice received aspirin in their drinking water from day 15 (one day before the first dose of LTC₄) today 21 (the day of euthanasia). BAL fluids were collected for cell counts, differentials, and mediator measurements. Compared with the mice not receiving aspirin, the aspirin-treated mice showed markedly diminished LTC₄-induced increases in total cells, eosinophils, neutrophils, and lymphocytes (Fig. 2A). The percentages of eosinophils and neutrophils remained higher in the LTC₄-challenged aspirin treated group than in the group challenged with OVA alone, although the percentages of eosinophils decreased by more than 50% (Fig. 2B). Treatment with aspirin eliminated LTC₄-induced increments in TXB₂, PGD₂, and PGE₂ elicited by LTC₄ (Fig. 2C). Treatment with aspirin also reduced the levels of CXCL7 detected in the BAL fluids of the LTC₄-treated mice (Fig. 2C).

Figure 2.

Effect of aspirin on LTC₄-induced pulmonary inflammation. Sensitized and challenged mice were treated with Lysine aspirin (1 mg/ml in drinking water) or water alone for 5 d before collection of BAL fluid. A. Numbers of total cells, eosinophils, neutrophils and lymphocytes in BAL fluids. B. Percentages of neutrophils and eosinophils in BAL fluid. C. Prostaglandins in BAL fluid lipid fractions and BAL fluid levels of CXCL7. Results in A-C are from ten mice/group from two separate experiments. * p<0.05, ** p<0.01, *** p<0.001.

Eosinophil recruitment in response to LTC₄ is eliminated by the absence of TP receptors

To determine whether TXA₂ and its receptor contributed directly to the recruitment of eosinophils to the lung in response to LTC₄, WT and tbxar^-/- mice were sensitized with OVA and challenged with OVA with or without LTC₄. We simultaneously monitored the BAL fluid for total and differential cell counts, and monitored the peripheral blood for both total and platelet-adherent eosinophils and neutrophils. WT mice and tbxar^-/- mice had similar BAL fluid cell counts and differentials following challenges with OVA alone (Fig. 3A). The percentages of peripheral blood CD45+ cells that were eosinophils (based on light scatter, as shown in Fig. 3A), and the fractions of the eosinophils with adherent platelets based on CD41 staining (not shown) were similar in the OVA-treated WT and tbxar^-/- mice, respectively (3B). LTC₄ challenge sharply decreased the percentages of CD45+ cells that were eosinophils (from 9.1 ± 2.0 to 3.1 ± 0.7%), but had no effect in the tbxar^-/- mice (from 9.7 ± 0.4 to 11.3 ± 1.7%) (Fig. 3B, as shown for representative samples, Fig. 3A). LTC₄ challenge also tended to decrease the percentages of blood eosinophils that were platelet-adherent (from 31.0 ± 7.4 to 19 ± 4.7%) in the WT mice, but not in the tbxar^-/- mice (from 26.6 ± 0.8 to 26.0 ± 2.0%, not shown). LTC₄-challenged tbxar^-/- mice were nearly completely protected from the increases in BAL fluid total cells, eosinophils, neutrophils, and lymphocytes induced by LTC₄ (Fig. 3C). The increases in BAL fluid levels of TXB₂ and of CXCL7 induced by LTC₄ were blunted in the tbxar^-/- mice (Fig. 3C). Treatment of the LTC₄-challenged mice with a single IP dose of the selective TP receptor antagonist SQ29.548 markedly suppressed the induction of pulmonary eosinophilia by LTC₄ (Supplemental Fig. 3).

Figure 3.

Effect of TP receptor deletion on LTC₄-mediated eosinophil recruitment to the lung. A. Cytofluorographic detection of eosinophils and neutrophils (based on light scatter characteristics within the CD45+ gate) in the blood of representative C57BL/6 WT and tbxar^-/- mice challenged with OVA alone or OVA + LTC₄. B. Percentages of blood granulocytes identified as eosinophils from the same groups of mice. C. Total BAL fluid cells, eosinophils, and lymphocytes from sensitized WT and tbxar^-/- mice challenged with OVA with or without LTC₄. BAL fluid levels of TXB₂ and CXCL7 are shown from the same mice. Results in A-C are from 10 mice/group from two separate experiments. * p<0.05, ** p<0.01.

Inductions of ICAM-1 and VCAM-1 by LTC₄ are sensitive to aspirin and depend on TP receptors

To determine whether the administration of LTC₄ induced the expressions of endothelial cell adhesion molecules in the lung, whole lung homogenates were used to generate western blots. Compared to the lungs of WT mice treated with OVA without LTC₄, the lungs of LTC₄-treated WT mice expressed higher levels of both ICAM-1 and VCAM-1 proteins (Fig. 4A, 4B). These LTC₄-mediated increases were completely abrogated in the lungs of mice treated with aspirin, and were absent in the lungs of LTC₄-treated tbxar^-/- mice. Neither treatment with aspirin nor deletion of TP receptors altered the baseline levels of ICAM-1 or VCAM-1 proteins expressed in response to OVA alone. Immunohistochemistry revealed stronger staining for both VCAM-1 and ICAM-1 proteins in the lungs of mice treated with OVA + LTC₄ than those treated with OVA alone. VCAM-1 staining localized to both venules and arterioles in the vicinity of the endothelium, whereas ICAM-1 staining localized to the alveolar walls as well as venules (Fig. 4C).

Figure 4.

LTC₄-mediated upregulation of pulmonary ICAM-1 and VCAM-1 proteins. Lysates were prepared from whole lung of the indicated strains of sensitized mice obtained 24 h after challenge with OVA with or without LTC₄. Some mice received aspirin in their drinking water. A. Western blots showing ICAM-1, VCAM-1, and GAPDH from the lungs of representative mice. B. Quantitative densitometry from the lungs of 5 mice/group. C. Immunohistochemical stains of lungs of representative OVA-sensitized mice challenged with low dose OVA either with vehicle or with LTC4. * p<0.05, ** p<0.01.

LTC₄-induced lung eosinophil recruitment requires both hematopoietic and non-hematopoietic TP receptor expression

To determine the most critical anatomic sites of TP receptor expression necessary for the amplifying effect of LTC₄ on eosinophil recruitment, lethally irradiated WT and tbxar^-/- mice were engrafted with WT or tbxar^-/- bone marrow. The resultant chimeric mice were then sensitized and challenged with OVA with or without the administration of intranasal LTC₄. Compared with WT hosts receiving WT marrow, tbxar^-/- mice receiving tbxar^-/- marrow demonstrated a markedly blunted response to LTC₄. Both the respective chimeric strains lacking only hematopoietic TP receptors and lacking only host tissue TP receptors also displayed dramatic reductions in BAL fluid total cell and eosinophil counts compared with the WT + WT mice (Fig. 5). Blood counts on the engrafted mice revealed complete restoration of all lineages (not shown).

Figure 5.

Contributions from hematopoietic and non-hematopoietic TP receptors to LTC₄-induced pulmonary inflammation. Lethally irradiated mice from the indicated recipient strains received bone marrow cells from the indicated donors. Ten weeks later, the mice were sensitized and challenged with OVA with or without LTC₄. A. Total BAL fluid cell counts and B. BAL eosinophil counts and percentages are displayed for 5 mice in each group. * p<0.05, ** p<0.01, *** p<0.001.

Discussion

Although cysLTs were initially recognized for their contractile effects on smooth muscle, they are also inducers and amplifiers of a range of immune and inflammatory responses. CysLT production is a consistent feature of the eosinophilic airway inflammation that is typical of asthma. This is especially true of aspirin exacerbated respiratory disease (AERD), a syndrome in which severe eosinophilic airway inflammation, asthma, sinonasal disease and respiratory reactions to nonselective cyclooxygenase inhibitors are accompanied by markedly elevated levels of cysLT generation (26). AERD is also associated with markedly increased numbers of platelet-adherent eosinophils in the blood and tissue (19). Although antagonists of CysLT₁R can blunt the severity of clinical reactions to aspirin in AERD (27) and reduce exacerbation frequency in asthma in general (17), their effects are modest and they do not affect airway inflammation, suggesting that additional cysLT receptors and downstream effectors are involved. We had previously reported that LTC₄, administered at a dose (2.2. nmol) equivalent to the maximally efficacious dose of LTE₄ in a similar model, far exceeded the potency of its metabolites for amplifying eosinophilic inflammation in C57BL/6 mice challenged with 0.1% OVA (a dose that elicits minimal inflammation (28) with minimal/negligible contributions from endogenous cysLTs (24,29)). These effects were strikingly platelet- and CysLT₂R-dependent (29). We therefore sought to better understand the mechanisms by which LTC₄, CysLT₂R, and platelets facilitated eosinophilic inflammation in this model.

In addition to priming leukocyte adhesion pathways by P selectin-dependent adherence to eosinophils and other leukocytes (20), platelets contain preformed mediators such as chemokines, and generate COX-1-derived lipid mediators that can amplify inflammation. The sharp increments in lung eosinophils induced by intrapulmonary LTC₄ in our model (Fig. 1A) were accompanied by increases in TXB₂ (a stable metabolite of TXA₂, the dominant COX product of platelets) (Fig. 1B), CXCL7 (a platelet-derived chemokine) (Fig. 1C). Intrapulmonary platelets were present in the bronchial submucosa of OVA-challenged mice, both with and without LTC₄ inhalation (Supplemental Fig. 1), suggesting that LTC₄ directly activated platelets that had been recruited to the lung in response to allergen, as demonstrated in previous studies (30). LTC₄ did not induce pulmonary eosinophilia on its own (Supplemental Fig. 2), indicating that chemotactic or viability-sustaining factors induced by the immune response to OVA are required for LTC₄ to amplify cell recruitment. Because platelets are both sources and responders to effector COX products, we next focused on the potential role of TXA₂ (and potentially other COX products) in our model.

Aspirin suppresses TXA₂ formation and platelet function sufficiently to induce vasoprotection in studies of cardiovascular disease (31). Aspirin also provokes clinical reactions in AERD that reflect depletion of bronchoprotective prostaglandins such as PGE₂ (32,33) due to a deficit in COX-2, a largely aspirin-resistant enzyme (34). Subsequently, aspirin induces desensitization and substantial clinical benefit by unknown mechanisms (35). We found that treatment of the mice with aspirin markedly suppressed LTC₄-induced potentiation of airway eosinophilia and neutrophilia (Fig. 2A), as well as the percentages of eosinophils but not of neutrophils (Fig. 2B). Concomitantly, aspirin markedly suppressed the formation of TXA₂ (as indicated by TXB₂ concentrations), PGD₂ (a product of both platelets and mast cells, Fig. 2C), and PGE₂ (a product of epithelial cells and macrophages) (Fig. 2C). The increases in all three prostaglandins by LTC₄ suggest that this ligand activates both platelets and additional cellular targets in this model. The suppression of CXCL7 generation by aspirin (Fig. 2C) likely reflects a loss of autocrine functions of TXA₂ on platelet activation. The capacity of TXA₂ and TP receptor signaling to facilitate lung inflammation in other models (36,37) led us to focus on whether TXA₂ might drive LTC₄-induced eosinophil accumulation using transgenic and pharmacologic approaches.

Eosinophils represent a small fraction of circulating granulocytes, but are recruited to sites of allergic inflammation in a highly selective manner. Platelets adhere to blood eosinophils with high frequency in asthma (18), and even higher in AERD (up to 80% of peripheral blood eosinophils with adherent platelets) (19), where they accumulate in large numbers in the respiratory tissues. Since it is plausible that platelet-derived TXA₂ could stimulate endothelial TP receptors to facilitate the entry of eosinophils (particularly those adherent to platelets) to the tissue, we sought to determine the role of TP receptors on LTC₄-induced eosinophil recruitment in our model. While the absence of TP receptors did not alter the blood eosinophil counts or the fraction of eosinophils with adherent platelets in the OVA-treated mice, the TP receptor deficient mice were completely protected from the LTC₄-induced decreases in blood eosinophils (Fig. 3A, 3B) and concomitant increases in BAL fluid eosinophil counts and platelet-derived mediators (Fig. 3C). Similarly, the TP receptor antagonist markedly decreased total BAL fluid cells and eosinophils when administered during the period of LTC₄ challenges (Supplemental Fig. 3). Thus, TP receptor signaling is not required for the intravascular formation of platelet-eosinophil complexes, but is absolutely required for these complexes to be recruited to the respiratory tissues and activated in response to LTC₄.

While TP receptors on platelets permit autocrine TXA₂-mediated amplification of platelet activation, TP receptors on endothelial cells mediate TXA₂-dependent upregulation of adhesion receptors (36,37). The lungs of WT mice treated with LTC₄ before OVA challenge expressed significantly higher levels of ICAM-1 and VCAM-1 proteins than did the lungs of mice treated with OVA alone. This induction was absent in both the aspirin-treated mice and in the tbxar^-/- mice (Fig. 4). The TXA₂/TP-dependent upregulation of ICAM-1 and VCAM-1 likely facilitates adhesion synergistically with platelet-dependent upregulation of β1 and β2 integrins on human and mouse eosinophils previously reported by our group and others (18,19). Whereas the ICAM-1/β₂ integrin pathway facilitates the adhesion of all leukocytes, the VCAM-1/β₁ pathway is selective for eosinophil recruitment (38), and may explain the dominance of the eosinophilic response to LTC₄ in our model. The fact that both non-hematopoietic and hematopoietic TP receptors are essential in this pathway (Fig. 5) suggests that both the autocrine and paracrine functions of platelet-derived TXA₂ are necessary for LTC₄ to elicit eosinophil recruitment. Since LTC₄ is the only cysLT that induces this pathway (29), it is tempting to speculate that platelet activation and TXA₂ formation may occur while the platelet contacts the granulocyte immediately after LTC₄ is released and before it is enzymatically converted to LTD₄. Given that LTC₄ and TXA₂ are both short-lived in vivo, rapid and successive activation of CysLT₂R (on intrapulmonary platelets) and TP receptors (on platelets and endothelial cells) is likely to be followed by gene induction events (including the induced expression of adhesion molecules) that permit subsequent leukocyte recruitment to occur over several hours.

While our previous study demonstrated that LTC₄ targeted CysLT₂R on platelets to induce inflammation, our present findings indicate that platelet-derived COX products (specifically TXA₂) are critical for this CysLT₂R-mediated pathway. TP receptor antagonists improved lung function and reduced the numbers of activated eosinophils in bronchial biopsies in controlled studies of subjects with mild asthma (39,40). It is possible that TP receptor antagonism may be especially efficacious in AERD, where platelet-dependent amplification of inflammation may be prominent due to the high levels of cysLT generation that is characteristic of this disease (19). The therapeutic effect of aspirin desensitization followed by high dose aspirin in AERD is not accompanied by a reduction in the high levels of cysLT generation, but rather by markedly suppressed TXA₂ (and PGD₂) production (41). We therefore suspect that aspirin therapy uncouples LTC₄ from its downstream COX-derived effectors as part of its therapeutic mechanism. Finally, the capacity for LTC₄ to activate platelets and elicit endothelial activation via TXA₂ may be relevant to the role of cysLTs in cardiovascular disease (42), in which animal models strongly support the role of CysLT₂R (43).