Leukotriene D₄ paradoxically limits LTC₄-driven platelet activation and lung immunopathology

Abstract

Background:

The 3 cysteinyl leukotrienes (cysLTs), leukotriene (LT) C₄ (LTC₄), LTD₄, and LTE₄, have different biologic half-lives, cellular targets, and receptor specificities. CysLT₂R binds LTC₄ and LTD₄ in vitro with similar affinities, but it displays a marked selectivity for LTC₄ in vivo. LTC₄, but not LTD₄, strongly potentiates allergen-induced pulmonary eosinophilia in mice through a CysLT₂R-mediated, platelet-and IL-33–dependent pathway.

Objective:

We sought to determine whether LTD₄ functionally antagonizes LTC₄ signaling at CysLT₂R.

Methods:

We used 2 different in vivo models of CysLT₂R-dependent immunopathology, as well as ex vivo activation of mouse and human platelets.

Results:

LTC₄-induced CD62P expression; HMGB1 release; and secretions of thromboxane A₂, CXCL7, and IL-33 by mouse platelets were all were blocked by a selective CysLT₂R antagonist and inhibited by LTD₄. These effects did not depend on CysLT₁R. Inhaled LTD₄ blocked LTC₄-mediated potentiation of ovalbumin-induced eosinophilic inflammation; recruitment of platelet-adherent eosinophils; and increases in IL-33, IL-4, IL-5, and IL-13 levels in lung tissue. In contrast, the effect of administration of LTE₄, the preferred ligand for CysLT₃R, was additive with LTC₄. The administration of LTD₄ to Ptges^−/− mice, which display enhanced LTC₄ synthesis similar to that in aspirin-exacerbated respiratory disease, completely blocked the physiologic response to subsequent lysine-aspirin inhalation challenges, as well as increases in levels of IL-33, type 2 cytokines, and biochemical markers of mast cell and platelet activation.

Conclusion:

The conversion of LTC₄ to LTD₄ may limit the duration and extent of potentially deleterious signaling through CysLT₂R, and it may contribute to the therapeutic properties of desensitization to aspirin in aspirin-exacerbated respiratory disease.

Graphical Abstract

Cysteinyl leukotrienes (cysLTs) are potent lipid mediators of inflammation that are prominent in asthma and mucosal type 2 immune responses. Inhaled cysLTs cause bronchoconstriction¹ and airway mucous secretion,² and they potentiate airway eosinophilia when administered by inahalation to human subjects.^3–5 They are derived from arachidonic acid oxidized by 5-lipoxygenase (5-LO) to generate the unstable precursor leukotriene (LT)A₄.⁶ LTA₄ is then conjugated to reduced glutathione by leukotriene C₄ synthase (LTC₄S),⁷ forming LTC₄, the parent cysLT. The principal sources of LTC₄ are mast cells (MCs), eosinophils, basophils, macrophages, and platelet-adherent granulocytes.⁸ Following release from these cell types, LTC₄ is converted extracellularly to LTD₄ by gamma-glutamyl transferase 5 (GGT5), removing glutamic acid from the glutathione adduct.⁹ LTD₄ is converted to the stable metabolite LTE₄ by di-peptidases that remove glycine.¹⁰ The conversion of LTC₄ to LTD₄ occurs rapidly, and the conversion of LTD₄ to LTE₄ occurs even more rapidly, such that the typical ratio of the 3 molecules detected in biologic fluids is as follows: LTE₄ is greater than LTC₄ is much greater than LTD₄.⁸ 5-LO activity and consequent synthesis of LTC₄ increase during exacerbations of asthma, as is reflected by increases in the urinary levels of LTE_4.¹¹ Steady-state LTC₄ synthesis rates are especially high in aspirin-exacerbated respiratory disease (AERD),¹² a phenotypically distinct clinical syndrome characterized by severe eosinophilic asthma and nasal polyps. Administration of aspirin or other COX-1–active drugs to subjects with AERD precipitates pathognomonic clinical reactions typified by changes in lung function and sinonasal obstruction, accompanied by a level of LTC₄ synthesis that is markedly increased above the high baseline.¹² Pharmacologic inhibition of 5-LO¹³ or blockade of the type 1 receptor for cysLTs (CysLT₁R)¹⁴ can reduce the severity of these reactions; it can also improve baseline lung function and reduce the frequency of asthma exacerbations in aspirin-tolerant patients.^15,16 These findings verify the role of cysLTs in asthma exacerbations, as well as the idiosyncratic immunopathology of AERD.

The 3 cysLT-selective receptors, termed CysLT₁R, CysLT₂R, and CysLT₃R,^17–19 differ in their ligand binding preferences and sites of expression. CysLT₁R binds LTD₄ with higher affinity (~1 nM) than it binds LTC₄ and LTE₄ (~1 and 2 log-fold lower affinity than LTD₄, respectively). CysLT₁R is expressed by vascular and airway smooth muscle, several leukocyte subsets, and platelets. CysLT₂R binds LTC₄ and LTD₄ with nearly identical affinity (~20 nM), but it exhibits weak or negligible activity for LTE₄.¹⁸ It is expressed by endothelial cells, cardiac Purkinje cells, leukocytes, and platelets. CysLT₃R binds LTE preferentially (~2 nM), but it can also respond to LTC₄ and LTD₄ in vivo.¹⁹ It is expressed principally by respiratory mucosal epithelial cells²⁰ and proximal convoluted tubular cells. Mouse and human receptors have very similar ligand affinities and preferences, and their sequences are highly conserved.^17,21 LTC₄ and LTD₄ are exceedingly potent bronchoconstrictors in human subjects^1,22 and guinea pigs.²³ These responses are sensitive to CysLT₁R antagonists.²⁴ Although LTE₄ is a weaker constrictor than its precursors,^25,26 subjects with asthma are approximately 10-fold more sensitive to the contractile effects of LTE₄ than are subjects without asthma.²⁶ Like LTC₄- and LTD₄-medicated bronchoconstrction, LTE₄-mediated bronchoconstriction in human subjects is sensitive to blockade by CysLT₁R antagonists.^4,27 Thus, cysLT-induced contractile effects in human airways are mediated largely, if not exclusively, by CysLT₁R.

Whereas CysLT₁R mediates direct contractile responses to cysLTs in vivo,²⁷ the expression of cysLT receptors by epithelial,²⁸ endothelial,²⁹ and hematopoietic cells^30,31 suggest additional noncontractile functions of cysLTs relevant to asthma and inflammation. Inhalation of LTE₄, but not inhalation of LTD₄, by human subjects with mild asthma elicits airway wall eosinophilia, basophilia, and neutrophilia when the 2 ligands are administered at doses that elicit equivalent degrees of contraction.^3,5 Inhalation challenge of naive mice with LTE₄ elicits CysLT₃R-dependent activation of group 2 innate lymphoid cells and consequent eosinophil recruitment by a mechanism involving IL-25 and epithelial brush cells.³² Inhaled LTE₄ also activates airway MCs in both humans and mice by a mechanism that is blocked by CysLT₁R antagonists.^27,33 In mice, repetitively inhaled LTC₄ upregulates expression of the type 2 cytokine IL-33 by alveolar type 2 (AT2) cells³⁴ and also elicits a substantial additional rapid, transient increase in lung IL-33 that requires recruited platelets.³⁵ CysLT₂R deletion or blockade eliminates both of these LTC₄-inducible pools. The LTC₄/CysLT₂R-driven increment in IL-33 synergizes with direct LTC₄/CysLT₁R-dependent signaling on group 2 innate lymphoid cells (ILC2s) to promote ILC2 expansion and IL-5 and IL-13 generation, resulting in markedly potentiated airway eosinophilia.³⁴ Curiously, although LTC₄ and LTD₄ bind to CysLT₂R with nearly equal affinity, LTD₄ potentiates eosinophilia only weakly and does not reproduce the effects of LTC₄ on platelet activation in vivo or ex vivo; nor does it induce expression of IL-33 or type 2 cytokines in lung tissue.³⁴ These observations suggest that LTC₄ and LTD₄ elicit qualitatively different signaling at CysLT₂R in vivo despite equivalent binding affinities for this receptor in vitro.

In the current study, we have demonstrated that rather than activating CysLT₂R, LTD₄ functionally antagonizes the effects of LTC₄ at this receptor. LTC₄, but not LTD₄, strongly activates p38 mitogen-activated protein kinase (MAPK) in platelets, which is blocked by LTD₄. LTD₄ blocks ex vivo platelet activation; production of thromboxane A₂ (TXA₂); and release of CXCL7, preformed IL-33, and high mobility box 1 (HMGB1) in response to either exogenous or endogenous LTC₄ in a dose-dependent manner, all of which require p38 activation. The inhibitory effects of LTD₄ neither require the presence of CysLT₁R nor are blocked by montelukast, a CysLT₁R-selective antagonist. The intransal administration of LTD₄ to wild-type (WT) mice markedly suppresses LTC₄-elicited potentiation of allergen-induced pulmonary inflammation, upregulation of IL-33 expression, increases in platelet activation markers, recruitment of platelet-adherent eosinophils to the lung, and increases in lung type 2 cytokines. Moreover, LTD₄ administration to C57BL/6 mice lacking prostaglandin E₂ synthase (Ptges^−/− mice) blocks the AERD-like physiologic response to inhalation challenge with lysine-aspirin (Lys-ASA), including the characteristic CysLT₂R-and platelet-dependent increase in lung IL-33, type 2 cytokine expression, and MC activation. Although the conversion of LTC₄ to LTD₄ generates a short-lived potent contractile agonist in vivo, it may also limit potential pathology induced by excessive endogenous CysLT₂R receptor signaling and facilitate clinical desensitization in AERD.

METHODS

Reagents

Extract from Dermatophagoides farinae (Df) was obtained from Greer Laboratories (XPB81D3A25; Lenoir, NC). Ovalbumin (OVA) and PBS were obtained from Sigma-Aldrich (St Louis, Mo). The mMCP-1 EIA kit was purchased from eBiosciences (San Diego, Calif). LTA₄, LTC₄, LTD₄, LTE₄, MK571, and HAMI3379 were obtained from Cayman Chemical (Ann Arbor, Mich). Histamine, thromboxane receptor B₂, PGD₂, and cysLT EIA kits were obtained from Cayman. IL-4, IL-5, IL-13, ICAM-1, and VCAM-1 EIA kits were from R&D Systems (Minneapolis, Minn). The CXCL7 EIA kit was purchased from Abcam (Cambridge, Mass). The HMGB1 EIA kit was from LifeSpan (Providence, RI). The monoclonal goat anti-mouse IL-33 was purchased from R&D Systems (Minneapolis, Minn), and the rat anti-mouse IgG (H1L) secondary antibody, fluorescein isothiocyanate (FITC) anti-mouse CD11c, FITC anti-mouse/human CD11b, FITC anti-mouse IgE, FITC anti-mouse CD3ε, FITC anti-mouse CD19, FITC anti-mouse CD8a, FITC anti-mouse NK-1.1, FITC anti-mouse Ly-6G/Ly-6C (Gr-1), allophycocyanin (APC) anti-mouse CD45, APC/cyanine 7 (Cy7) anti-mouse/human CD44, PerCP/Cy5.5 anti-mouse CD90.2, phycoerythrin (PE) anti-mouse CD278 (inducible costimulatory molecule), APC anti-mouse CD41, PE/Cy7 anti-mouse CD62P, APC anti-human CD61, anti-mouse CD16/32, PE/Cy7 anti-mouse CD45, and PE anti-mouse Siglec F were all obtained from BioLegend (San Diego, Calif).

Mice

The C57BL/6 mice lacking mPGES-1 (Ptges^−/− mice) were a gift from Dr Shizuo Akira (Osaka University, Japan).³⁶ All of the mice, including the WT C57BL/6 controls, were housed at Brigham and Women’s Hospital’s Hale Building for Transformative Medicine (Boston, Mass). Six-to 8-week-old male mice were used. All animal studies were approved by the Animal Care and Use Committee of Brigham and Women’s Hospital (protocol 2016N000294).

Immunization and challenge

To study potentiation of airway inflammation by exogenous cysLTs, mice were sensitized intraperitoneally on days 0 and 5 with alum-precipitated chicken egg OVA (10 µg). On days 16 to 18, the mice received intranasal challenge of 2.2 nmol LTC₄ or vehicle. On days 17 to 19, the mice were challenged by inhalation of 0.1% OVA.³⁷ Twenty-four hours after the final OVA aerosol challenge, the mice were humanely killed and exsanguinated. The lungs were lavaged 3 times with 0.7 mL of PBS and 5 mM EDTA. Bronchoalveolar lavage (BAL) fluid cells were cytocentrifuged onto slides, stained with Diff-Quick (Fisher Diagnostics, Middletown, Va), and differentially counted.

In the Ptges^−/− mice, airway inflammation was induced by intranasal administration of Df (Greer, 3 µg) as described elsewhere.³⁸ The mice were challenged with Lys-ASA 24 hours after the last treatment.

Flow cytometry

The mouse lungs (right lobes) were transferred into a 6-well dish, and the tissue was teased apart with forceps. The tissue was then digested at room temperature for 45 minutes in 2 mL of dispase (2 U/mL), after which 0.5 mg of DNAse per mouse was added to the mixture, which was then incubated for 10 minutes at RT with gentle rocking on a shaker to 200 rpm. The cells were filtered through 70-µm nylon mesh and pelleted by centrifugation for 10 minutes at 350 g at 4°C. Red blood cell lysis was performed by resuspending the pellet in 2 mL of 1× red blood cell lysis buffer (Biolegend) and incubating it on ice for 4 minutes, after which the reaction was terminated by addition of 13 mL of Dulbecco modified Eagle medium. The cells were centrifuged for 10 minutes at 350 g at 4°C and then washed twice with fluorescence-activated cell sorting buffer (0.5% BSA in PBS). Next, 1 × 10⁶ cells were stained with antibodies in 100 µL of fluorescence-activated cell sorting buffer for 20 minutes on ice in the dark. The cells were washed and resuspended in 300 µL of 1% paraformaldehyde in PBS before analysis on a FACSCanto flow cytometer (BD Biosciences). ILC2s were quantitated as Lin^–CD45⁺CD44⁺CD278⁺CD90.2⁺ cells in the lymphocyte gate.

For analysis of the platelet-adherent eosinophils, lung or BAL cells were stained with PE anti-mouse Siglec F and PE/Cy7 anti-mouse CD45 (to identify eosinophils), as well as with APC anti-mouse CD41 (to identify the platelets), after which the eosinophils were analyzed by flow cytometry. The percentages of eosinophils that were platelet-adherent (CD41⁺) and the total numbers of platelet-adherent eosinophils were quantified.

Platelets in platelet-rich plasma (PRP) were identified on the basis of size and presence of CD41. The CD62P⁺ baseline of the CD41⁺ events was set at 5% for the vehicle control treatment and compared with the agonist treatment. For human samples, healthy volunteer subjects were recruited from the Brigham and Women’s Hospital primary care practice for blood donations. The local institutional review board approved the study, and all subjects provided written informed consent. PRP was stimulated with LTA₄, LTC₄, LTD₄, or an equal volume of ethanol (vehicle control) and processed for flow cytometry for surface CD62P⁺ on the CD61⁺ gate. In some of the experiments, supernatants and/or pellets were collected after stimulation to analyze CXCL7 and thromboxane receptor B₂ release by ELISA, or the IL-33 release by Western blot.

MAPK activation

Lysates of washed platelets stimulated with LTC₄ or vehicle in the absence or presence of LTD₄ were used to generate Western blots. Membranes were probed with antibodies specific for phosphorylated and total c-Jun terminal kinase (JNK), p38, and extracellular signal–regulated kinase (ERK) (Cell Signaling Technologies). In some experiments, platelets were stimulated in the presence of selective inhibitors of JNK (SP600125, Sigma), p38 (SB203580, InVivogen), or ERK (UO126 InVivogen) (10 µM each) for 30 minutes before activation and measurement of mediator release or flow cytometry assays.

Measurement of airway resistance

Airway resistance (R_L) in response to Lys-ASA was assessed with an Invasive Pulmonary Function Device (Buxco, Sharon, Conn). Briefly, the mice were anesthetized 24 hours after the last Df challenge, and tracheotomy was performed. After allowing for R_L to reach a stable baseline, Lys-ASA (12 µL of 100 mg/mL) was delivered to the lung via a nebulizer, and R_L was recorded for 45 minutes. The results were expressed as the percentage of change in R_L from baseline.

Statistical analysis

Data are expressed as means plus or minus SEMs from at least 10 mice from at least 2 experiments, except where otherwise indicated. Analyses were performed with Prism software (GraphPad Software, La Jolla, Calif). Differences between 2 treatment groups were assessed by using the Student t test, and differences among multiple groups were assessed by using 1-way ANOVA and the Bonferroni post hoc test. P values less than .05 were considered statistically significant.

RESULTS

LTD₄ blocks LTC₄-induced CysLT₂R-mediated platelet activation ex vivo

Although mouse and human platelets express both CysLT₁R and CysLT₂R,^37,39 their ex vivo activation in response to LTC₄ is entirely CysLT₂R-dependent and is not reproduced by either LTD₄ or LTE₄.³⁷ To determine whether LTD₄ or LTE₄ interfered with CysLT₂R-dependent platelet activation, mouse PRP was stimulated with LTC₄ in the absence or presence of various concentrations of LTD₄. Some samples were stimulated with LTA₄, which platelets convert to LTC₄ that can activate platelets in an autocrine manner.³⁵ CD62P surface expression was monitored on CD41⁺ platelets as an index of activation. LTD₄ inhibited LTC₄-induced CD62P expression in a dose-dependent manner (Fig 1, A [representative plots shown in Fig 1, E]), completely blocking its effect when provided at an equimolar dose, and it also blocked LTA₄-induced CD62P (Fig 1, D). This effect was not reproduced by LTE₄ (not shown). To determine whether the inhibitory effect of LTD₄ on LTC₄-induced CD62P expression required the presence of its preferred receptor CysLT₁R, we conducted studies using the selective CysLT₁R antagonist MK571 and performed experiments using PRP from Cysltr1^−/− mice. The inhibitory effect of LTD₄ on LTC₄-mediated CD62P expression resisted CysLT₁R blockade of WT platelets (Fig 1, B), and it was equally efficacious on platelets lacking CysLT₁R and on WT platelets (Fig 1, C). LTD₄ also blocked platelet activation by LTA₄, but it did not alter activation induced by the thromboxane receptor (TP receptor) agonist U-46619 (Fig 1, D).

FIG 1.

LTD₄ blocks LTC₄/CysLT₂R-dependent platelet activation. PRP from the indicated mouse strains was stimulated with LTC₄ (250 nM) for 30 minutes with or without the simultaneous addition of the indicated agonists and antagonists. Surface expression of CD62P was monitored by flow cytometry. A, Dose-dependent effect of LTD₄ on LTC₄-elicited expression of CD62P. B, Lack of effect of the CysLT₁R-selective antagonist MK571 on LTC₄-elicited CD62P expression and its suppression by LTD₄. C, Lack of effect of Cysltr1 deletion on LTC₄-elicited CD62P expression. D, Effect of LTD₄ on CD62P expression elicited by the indicated agonists. E, A representative plot showing the effect of LTC₄ on CD62P expression and the blocking effect of LTD₄. Results in (A) to (D) are the means ± SDs from 3 separate experiments. ***P < .001. APC-A, Allaphycocyanin; SSC-A, side angle light scatter.

To determine whether LTD₄ blocked the secretion of soluble inflammatory mediators by LTC₄-stimulated platelets, we measured thromboxane receptor B₂ (as a surrogate for TXA₂ generation) and for the platelet-associated chemokine CXCL7 in supernatants of washed LTC₄-activated mouse platelets. LTC₄ elicited the release of these products, whereas LTD₄ blocked their LTC₄-mediated release (Fig 2, A). LTC₄-induced TXA₂ production and CXCL7 release by mouse platelets were blocked by the CysLT₂R-selective antagonist HAMI-3379, but not by MK571. MK571 did not reverse the LTD₄-induced inhibition of LTC₄-induced TXA₂ synthesis and CXCL7 release. To verify that LTD₄ could also interfere with LTC₄-dependent activation of human platelets, we stimulated washed platelets from healthy human donors with LTC₄ in the absence or presence of LTD₄. Neither LTC₄ nor LTD₄ elicited CD62 expression by human platelets (not shown). LTC₄, but not LTD₄, induced the release of CXCL7 and TXA₂ production from human platelets. LTD₄ completely blocked these responses of human platelets to LTC₄ without inducing platelet activation (Fig 2, B). In contrast to the activation response of mouse platelets, both MK571 and HAMI-3379 blocked the activation response of human platelets to LTC₄.

FIG 2.

LTD₄ blocks the release of soluble mediators from mouse and human platelets activated by LTC₄. Washed platelets from WT mice (A) or healthy human volunteers (B) were stimulated for 30 minutes with LTC₄ in the absence or presence of the indicated agonists or antagonists. CXCL7 and thromboxane receptor B₂ (TXB₂) (as a surrogate for TXA₂ production) were measured in the supernatants by ELISA. The data in (A) are means 6 SDs from 3 experiments. The data in (B) are from 10 experiments using platelets from different donors. ***P < .001; **P < .01; *P < .05.

LTC₄ elicits the release of preformed IL-33 from mouse platelets by a CysLT₂R-dependent mechanism that depends on signal amplification from endogenous HMGB1 and receptor for advanced glycation end products (RAGE).³⁵ To determine whether LTD₄ blocked this response to LTC₄, we monitored the surface expression of HMGB1 and performed Western blotting for IL-33 on supernatants and pellets from platelets activated by LTC₄ in the absence or presence of LTD₄. LTC₄ induced both HMGB1 surface expression (see Fig E1, A in this article’s Online Repository at www.jacionline.org) and release of IL-33 into the supernatants, accompanied by a decrement in IL-33 in the platelet lysate (see Fig E1, B). LTD₄ blocked both of these responses to LTC₄ (see Fig E1, A and B).

FIG E1.

Effect of LTD₄ on LTC₄-elicited HMGB1 and IL-33 release by platelets. A, PRP from WT mice was stimulated with the indicated combinations of agonists and antagonists for 30 minutes. HMGB1 release was detected on the basis of its surface expression. Results are from 3 independent experiments. B, Washed platelets were stimulated with the indicated agonists and antagonists for 30 minutes. Supernatants and pellets were collected, and the proteins were resolved by SDS-PAGE. The resultant Western blot (top) is representative of 3 separate experiments. Quantitative densitometry showing percentage of release of IL-33 (bottom). Veh, Vehicle.

To identify mechanisms responsible for the differential responses of CysLT₂R to LTC₄ versus to LTD₄, we performed signaling assays on platelets. Neither LTC₄ nor LTD₄ caused calcium flux in platelets (not shown). LTC₄, but not LTD₄, elicited phosphorylation of JNK, extracellular signal regulated kinase (ERK), and p38 kinase in platelets (see Fig E2, A in this article’s Online Repository at www.jacionline.org). The concomitant administration of LTD₄ blocked these responses (see Fig E2, A and B). A selective inhibitor of p38 (SB203580), but not inhibitors of ERK (uo126) or JNK (SP600125), blocked LTC₄-induced release of IL-33 (see Fig E2, C), secretion of TXA₂ and CXCL7 (see Fig E2, D), and surface induction of CD62P and HMGB1 (see Fig E2, E) in response to LTC₄.

FIG E2.

Differential effects of LTC₄ on and LTD₄ on MAPK activation. A, PRP from WT mice was stimulated with the indicated cysLTs for 10 minutes. Western blots were probed for phospho-specific and total JNK, p38, and ERK. B, Quantitative densitometry. Effect of specific inhibitors of the indicated MAPK on IL-33 release (C), soluble mediator release (D), and platelet surface inductions of CD62P and HMGB1 (E). Data in (B), (D), and (E) are from 3 experiments. ***P < .001; **P < .01; *P < .05. p, Phosphorylated; WB, Western blot.

LTD₄ inhibits CysLT₂R-dependent immunopathology induced by LTC₄, but not by CysLT₃R-dependent responses to LTE₄

Given its ability to block LTC₄-induced platelet activation ex vivo via CysLT₂R, we sought to determine whether LTD₄ blocked the induction of eosinophilic inflammation, platelet activation markers, and type 2 cytokine expression by exogenous LTC₄ or LTE₄ in vivo, which depend on CysLT₂R and CysLT₃R, respectively.³⁴ LTD₄ (2.2 nM) was administered intranasally on 3 successive days to WT mice that had been sensitized to OVA; it was administered concomitantly with equal quantities of either LTC₄ or LTE₄. In some experiments, the LTD₄ was administered 1 hour after the LTC₄ or LTE₄ to determine whether it dampened inflammation even after the initial receptor stimulations. Twenty-four hours after each dose of cysLTs, the mice were challenged with low-dose OVA (0.1% for 30 minutes) (Fig 3, A). Twenty-four hours after the last challenge, the mice were humanely killed, BAL fluid and lung lysates were collected, and single-cell suspensions were prepared for analysis by flow cytometry. As expected, intranasal cysLTs increased the numbers of total BAL fluid cells and eosinophils with a rank order of LTC₄ greater than LTE₄ much greater than LTD₄ (Fig 3, B). Each cysLT also increased the numbers and percentages of ILC2s in single–lung cell suspensions over control mice not receiving cysLTs, with a rank order of potency that paralleled that of the BAL fluid cell counts (Fig 3, C). LTD₄ sharply decreased the numbers of BAL fluid total cells and eosinophils induced by LTC₄ when the 2 ligands were administered together (Fig 3, B), reaching the levels observed with LTD₄ alone. In contrast, LTD₄ did not reduce the numbers of lung ILC2s induced by LTC₄ inhalation (Fig 3, C). Delayed administration of LTD₄ had no effect on LTC₄-mediated amplification of BAL fluid eosinophilia (see Fig E3 in this article’s Online Repository at www.jacionline.org). LTD₄ had no effect on LTE₄/CysLT₃R-mediated increases in BAL fluid cellularity or eosinophilia. In contrast to the suppressive effects of LTD₄, the effects of LTE₄ on total BAL fluid cell count and eosinophil count (Fig 3, B) were additive to those of LTC₄, and LTE₄ did not suppress LTC₄-elicited ILC2 expansion (Fig 3, C). LTC₄ increased the extent of bronchovascular cellular infiltrates and markedly increased the numbers of PAS⁺ goblet cells. These increases were significantly attenuated by treatment of the mice with LTD₄ (Fig 3, D and E). Total serum IgE measurements in all OVA-treated mouse groups increased significantly over those in the unsensitized controls, and they were modestly increased by inhaled LTC₄ (see Fig E4 in this article’s Online Repository at www.jacionline.org).

FIG 3.

LTD₄ blocks amplification of OVA-induced pulmonary inflammation induced by LTC₄ but not by LTE₄. A, Time line of the experiments. WT mice were sensitized on days 0 and 7 with OVA/alum intraperitoneally. On days 14 to 16, the mice received single intranasal doses (2.2 nmol) of the indicated cysLTs, followed 24 hours later by inhaled OVA (0.1% for 30 minutes). BAL fluid and lung tissue were collected 24 hours after the last dose of OVA. B, Total BAL fluid cell counts (left) and eosinophil counts (right). C, Numbers of ILC2s in dispersed lung tissue from the indicated groups. D, Hematoxylin and eosin (H&E) and periodic acid–Schiff (PAS) stains from the lungs of the indicated groups of mice. E, Quantification of inflammation score and PAS⁺ goblet cells. Results in (B-D) are from 3 separate experiments with 10 to 15 mice/group.

FIG E3.

Delayed administration of LTD₄ does not reverse of the effects of LTC₄. A. Time line for the experiments. B. Total BAL fluid cell counts (left) and eosinophils (right). C. Lung ILC2s. Results are from 5 mice/group. ***P < .001; **P < .01. i.n., Intranasal; i.p., intraperitoneal.

FIG E4.

LTC₄ amplifies lung inflammation without altering IgE sensitization. A, Time line for the experiments. B, BAL fluid total cells (left) and eosinophils (right). C, Total serum IgE for the indicated groups of mice. Results are from 5 mice/group. ***P < .001. i.n., Intranasal; i.p., intraperitoneal.

LTD₄ blocks LTC₄-induced upregulation of lung IL-33 expression, platelet activation markers, and adhesion receptors

The amplification of OVA-induced immunopathology by exogenous LTC₄ requires platelet-dependent upregulation of the endothelial adhesion receptor expression⁴⁰ that is necessary for recruitment of eosinophils, as well as for the release of platelet HMGB1 and signaling through RAGE.³⁵ LTC₄ also elicits an increase in intranuclear IL-33 protein associated with AT2 cells.³⁴ LTC₄/CysLT₂R-elicited IL-33 synergizes with direct CysLT₁R-mediated stimulation to expand ILC2s and induce their production of IL-5 and IL-13.³⁴ Depletion of platelets or blockade of CysLT₂R in this model eliminates LTC₄-induced eosinophilia, platelet activation markers, and increments in IL-5 and IL-13, the latter of which are also eliminated by ILC2 depletion. To determine whether the CysLT₂R-dependent processes were specifically blocked by LTD₄ in vivo, we examined whole lung lysates from OVA plus LTC₄–challenged mice for IL-33 protein and downstream type 2 cytokines (IL-4, IL-5, and IL-13), as well as from BAL fluid for markers of platelet activation (CXCL7 and HMGB1) and endothelial activation (VCAM-1, ICAM-1). As expected, intranasal LTC₄ increased the content of IL-33, IL-4, IL-5, and IL-13 proteins in the lung lysates (Fig 4, A) 24 hours after the final dose. LTC₄ increased the BAL fluid content of CXCL7 and HMGB1 (Fig 4, B), as well as the content of soluble ICAM-1 and VCAM-1 (Fig 4, C). All of these were blocked by the concomitant administration of LTD₄. LTE₄ only modestly increased lung IL-33 levels (Fig 4, A), substantially potentiated IL-5 but not IL-4 or IL-13 levels, and had no effect on platelet or endothelial activation markers (Fig 4, B and C). None of the LTE₄ effects were blocked by the concomitant administration of LTD₄.

FIG 4.

Effect of LTD₄ on lung lysate cytokines and BAL fluid mediators induced by LTC₄ and LTE₄. A, ELISA measurements of IL-33, IL-4, IL-5, and IL-13 in whole lung lysates from WT OVA-sensitized and challenged mice receiving the indicated cysLTs. Measurements of BAL fluid HMGB1 and CXCL7 (B) and soluble adhesion receptors (C) from the same mice as in (A). Results are from 10–15 mice/group. ***P < .001; **P < .01; *P < .05.

Although LTC₄-induced increases in lung IL-33 principally reflect its expression by AT2 cells at 24 hours after the final challenge, LTC₄ inhalation also elicits rapid and transient platelet recruitment to the lungs of sensitized mice, resulting in an additional platelet-dependent increment in BAL fluid eosinophils and a marked additional increment in lung IL-33 protein via CysLT₂R.³⁵ To determine the potential impact of LTD₄ on these rapid LTC₄/CysLT₂R-induced responses, we administered a fourth dose of LTC₄ with or without LTD₄ to OVA-challenged mice and humanely killed the mice 1 hour later (Fig 5, A). As expected, the additional dose of LTC₄ caused rapid, additional increases in lung levels of BAL fluid total cells (a 50% increase) and eosinophils (a 2-fold increase) (Fig 5, B). LTC₄ also sharply increased the percentages and numbers of platelet-adherent (CD41⁺) eosinophils that are present in the BAL fluid (Fig 5, C [as shown for a representative experiment]) and dispersed lung tissue (Fig 5, D), as well as the increases in ILC2s (Fig 5, E). Compared with control mice that received vehicle, the mice that received the additional dose of LTC₄ also displayed significant increases in lung lysate levels of IL-33 and BAL fluid concentrations of HMGB1, CXCL7, and soluble endothelial adhesion receptors (ICAM-1 and VCAM-1) (Fig 5, F).^34,35 The concomitant administration of LTD₄ nearly completely blocked the rapid LTC₄-induced increases in total BAL fluid cells, total eosinophils, and platelet-adherent eosinophils in the BAL fluid and lung tissue, and it also eliminated the rapid additional increases in ILC2s, IL-33, HMGB1, and platelet and endothelial activation markers (Fig 5, B–F).

FIG 5.

LTC₄ rapidly elicits incremental lung recruitment of platelet-adherent eosinophils and IL-33, which are blocked by LTD₄. A, Time line for the experiments. WT OVA-sensitized mice that were challenged with 3 doses of LTC₄ and OVA were challenged 24 hours later with either LTC₄ (2.2 nmol) or PBS. BAL and lung tissue were collected 30 minutes afterward. B, Total numbers of cells (left) and eosinophils (right) in BAL fluids collected from the indicated groups of mice. C, Percentages (left) and total numbers (right) of platelet-adherent (CD41¹) eosinophils identified in BAL fluid cells by flow cytometry. D, Percentages (left) and numbers (right) of platelet-adherent eosinophils identified in single–lung cell suspensions. Representative plots showing platelet-adherent eosinophils in the BAL fluid from mice of the indicated treatment groups. E, Numbers of ILC2s in dispersed lung tissues from the indicated groups. F, Whole lung levels of IL-33 and platelet and endothelial activation markers from the same mice as in (B). Results are from 10 to 15 mice/group. ***P < .001; **P < .01; *P < .05. Comp., Complete; i.n., intranasal; SSC, side scatter.

LTD₄ prevents AERD-like reactions to Lys-ASA in Ptges^−/− mice

We sought to determine whether exogenous LTD₄ could interfere with AERD-like responses to inhaled Lys-ASA challenges of Ptges^−/− mice, an acute physiologic event in the airways that depends on endogenously generated LTC₄.³⁸ We administered LTD₄ to Df-primed Ptges^−/− mice before inhalation challenge with Lys-ASA. These mice exhibited sharply increased BAL fluid levels of cysLTs, accompanied by cysLT-dependent recruitment and activation of platelets, release of platelet-derived IL-33, IL-33-dependent activation of MCs and ILC2s, and increases in airway resistance (R_L). Df-primed Ptges^−/− mice received intranasal LTD₄ before Lys-ASA challenge (Fig 6, A). Compared with vehicle-challenged controls, the Lys-ASA–challenged mice treated with LTD₄ displayed a lower peak R_L value (Fig 6, B); exhibited sharply reduced BAL fluid levels of the MC-derived mediators mMCP-1, PGD₂, and histamine (Fig 6, C); and reduced BAL fluid levels of CXCL7 and HMGB1 and IL-33 in the lung (Fig 6, D). Lys-ASA challenges also rapidly increased the levels of IL-4, IL-5, and IL-13 in the lung lysates (Fig 6, E). These increases were eliminated by the administration of exogenous LTD₄.

FIG 6.

Inhaled LTD₄ blocks AERD-like reactions to lysine aspirin challenges of Ptges^−/− mice. A, Time line for the experiments. Df-primed Ptges^−/− mice were challenged by inhalation of PBS or Lys-ASA. Some mice received a single inhaled dose of LTD₄ 30 minutes before challenge. B, Maximum percentage of change in R_L monitored continuously for 45 minutes after the administration of Lys-ASA or PBS. C, Levels of MC activation markers (mMCP-1, histamine, and PGD₂). D, Platelet activation markers (CXCL7 and HMGB1) in BAL fluids from the indicated groups of mice. Whole lung levels of IL-33 from the same mice are shown. E, Whole lung levels of IL-4, IL-5, and IL-13 from the indicated groups. Results are from 10 mice per group. ***P < .001; **P < .01; *P < .05.

The absence of CysLT₂R converts LTD₄ into a potent inducer of eosinophilic inflammation

To verify that the ability of LTD₄ to block LTC₄-elicited increases in indices of lung inflammation were due to blockade of CysLT₂R, we examined the effects of each of the 3 cysLTs on potentiation of OVA-induced eosinophilia and associated production of mediators in WT mice and Cysltr2^−/− mice (see Fig E5, A in this article’s Online Repository at www.jacionline.org). The absence of CysLT₂R eliminated potentiation of OVA-induced BAL eosinophilia and all associated features of inflammation (ILC2 expansion and increases in IL-33 and platelet activation markers) (see Fig E5, B–D in this article’s Online Repository at www.jacionline.org) induced by LTC₄, while only minimally affecting the responses to LTE₄. Notably, LTD₄ behaved as a stronger agonist for BAL fluid eosinophilia, ILC2 expansion, and IL-33 induction in Cysltr2^−/− mice than in the WT controls.

FIG E5.

Deletion of CysLT₂R eliminates potentiation of OVA-induced inflammation by LTC₄ but converts LTD₄ into a potent agonist. WT and Cysltr2^−/− mice were sensitized with OVA/alum and challenged with the indicated inhaled cysLTs on 3 successive days, with each challenge followed 24 hours later by inhaled OVA. Samples were collected 24 hours after the final OVA challenge. A, Time line for the experiments. B, BAL fluid total cell and eosinophil counts and lung ILC2s in the indicated groups. C, Lung IL-33 levels. D, Levels of platelet (CXCL7 and HMGB1) and endothelial (VCAM-1 and ICAM-1) activation markers in BAL fluid from the indicated groups. Results are from 10 mice/group. i.n., Intranasal; i.p., intraperitoneal.

DISCUSSION

The synthesis and release of LTC₄ is followed by successive metabolism to LTD₄ and LTE₄, effectively providing 3 distinct extracellular ligands with different respective functional properties in inflammation. Whereas some of these functions differ between mice and humans (eg, the direct CysLT₁R-dependent bronchoconstriction elicited by cysLT inhalation in humans^1,22 is not displayed by naive mice⁴¹), some are conserved (eg, airway eosinophilia^3–5,32,42 and MC activation elicited by inhaled LTE₄^27,33,43). Although CysLT₂R-dependent upregulation of lung IL-33 expression by LTC₄ has not been examined in humans, LTC₄ can directly induce IL-33 release from freshly surgically excised human sinonasal tissue.³⁵ Restriction of LTC₄ synthesis by cyclic AMP and protein kinase A–induced 5-LO phosphorylation,⁴⁴ ligand-dependent internalization/desensitization of CysLT₁R,^45,46 restriction of CysLT₁R membrane expression, signaling, and/or ligand binding by dimerization with CysLT₂R⁴⁷ or GPR17,⁴⁸ and inhibition of CysLT₃R signaling by coexpression of CysLT₁R and CysLT₂R^19,49 may all limit the potentially deleterious effects of excessive cysLT-induced signaling in vivo. To date, no described mechanism has limited signaling through CysLT₂R, which is frequently expressed by the same cells that express LTC₄S,³⁰ implying a potential autocrine role for LTC₄/CysLT₂R signaling. The curious observations that the potency of exogenous LTC₄, acting at CysLT₂R, far exceeded the potency of LTD₄ for inducing type 2 immunopathology and that LTC₄, but not LTD₄, could activate mouse platelets through CysLT₂R³⁷ are counterintuitive for a receptor known to bind LTC₄ and LTD₄ with equal affinity in vitro. These observations prompted this study.

Exogenous LTC₄ upregulates CD62P on mouse platelets in a CysLT₂R-dependent manner, permitting their binding to granulocytes and other PSGL-1–expressing leukocytes.³⁷ Because platelets also express LTC₄S, they can convert granulocyte-derived LTA₄ to LTC₄, an event that also elicits CD62P via an autocrine, CysLT₂R-dependent circuit.³⁵ Although platelets also express CysLT₁R, these responses in mice require only CysLT₂R, are unaffected by deletion of Cysltr3, and cannot be reproduced by either LTD₄ or LTE₄.³⁷ Several lines of evidence are consistent with LTD₄ functionally antagonizing platelet CysLT₂R. First, the complete blockade of LTC₄-induced CD62P by an equimolar dose of LTD₄ reflects the nearly identical affinities of LTC₄ and LTD₄ for CysLT₂R.²¹ Second, LTD₄-mediated inhibition of LTC₄-elicited platelet activation is both resistant to the CysLT₁R antagonist MK571 (Fig 1, B) and unaffected by Cysltr1 deletion (Fig 1, C). Third, LTD₄ also blocked the response to exogenous LTA₄ (Fig 1, D), which depends on CysLT₂R signaling following conversion of LTA₄ to LTC₄ by platelet LTC₄S.³⁵ Finally, LTD₄ did not affect CD62P expression elicited by the selective agonist of the TP receptor (Fig 1, D), suggesting a CysLT receptor–specific action. LTD₄ also blocked the release of soluble mediators (TXA₂ and CXCL7) (Fig 2, A) and prevented release of the alarmins HMGB1 and IL-33 (see Fig E1). Moreover, LTD₄ blocked the phosphorylation of all 3 MAPKs elicited by an equimolar dose of LTC₄ (see Fig E2), including p38, and inhibition of p38 (like LTD₄) abrogated all indices of LTC₄-induced platelet activation. Thus, although a non–receptor-dependent effect cannot be completely excluded, it seems likely that LTD₄ competes with LTC₄ for binding to CysLT₂R.

Although the selective agonist effect of LTC₄ and the blocking effect of LTD₄ observed in mouse platelets was similar for human platelets (Fig 2, B), there were significant interspecies differences. First, LTC₄ induced CXCL7 release and TXA₂ generation by human platelets in quantities nearly identical to the quantities induced in mouse platelets despite its lack of ability to induce CD62P expression. Second, both HAMI-3379 and MK571 blocked LTC₄-induced mediator release by human platelets, whereas only HAMI-3379 blocked the response of mouse platelets (Fig 2, B). A previous study reported that the CysLT₁R antagonist pranlukast blocked cysLT-induced secretion of the chemokine RANTES by human platelets³⁹; CysLT₂R function was not assessed. It is possible CysLT₁R displays different degrees of surface expression on human and mouse platelets, or that both receptors are regulated functionally by divergent postreceptor signaling pathways in the 2 species. Nevertheless, whereas the role of CysLT₁R differs, the contrasting effects of LTC₄ and LTD₄, the requirement for CysLT₂R for LTC₄-elicited secretory function, and the functional antagonism exerted by LTD₄ are common to the platelets of both species. This may explain the curious observation that CysLT₂R displays a sharp preference for LTC₄ in certain in vivo models,^29,50 and it highlights the fact that G protein–coupled receptor functions are highly modified by the cellular context in which they are expressed.

Both LTC₄ (through CysLT₂R) and LTE₄ (through CysLT₃R) elicit mild airway eosinophilia when administered to naive mice by inhalation,^32,34 and both significantly amplify eosinophilia and ILC2 expansion when administered to OVA-sensitized mice before OVA challenge with a potency of LTC₄ greater than that of LTE₄. This amplification of the effector phase is not due to a requirement for LTC₄ to induce IgE sensitization (see Fig E4). Although LTD₄ and LTC₄ can both elicit CysLT₁R-dependent expansion of lung ILC2s in vivo,³⁴ LTD₄ potentiates allergen-induced pulmonary inflammation only weakly when compared with LTC₄ and LTE₄,³⁴ and unlike LTC₄, it does not elicit CysLT₂R-dependent IL-33 expression and platelet activation. Several lines of evidence suggest that LTD₄ instead actively suppresses CysLT₂R signaling to block inflammation elicited in response to LTC₄ in vivo. First, LTD₄ suppressed all features of this model that depend exclusively on signaling through CysLT₂R (potentiation of BAL fluid eosinophilia [Fig 3, B], induction of bronchovascular inflammation [Fig 3, D and E], increases in IL-33 protein content and level of downstream type 2 cytokines [Fig 4, A], increases in HMGB1 and CXCL7 [Fig 4, B], and upregulation of endothelial adhesion receptors [Fig 4, C]). Second, the CysLT₁R-dependent expansion of ILC2s was modestly induced, rather than suppressed, by LTD₄ (Fig 3, C). Third, LTD₄ failed to alter the activity of LTE₄, which induces eosinophilic pathology through epithelium-associated CysLT₃R and the platelet-associated purinergic receptor P2Y₁₂^34,42 (Fig 3, B and C). These findings support a surprising degree of functional segregation of cysLT-driven immunologic effects despite predicted crossover of these ligands on the same receptors based on studies using the recombinant proteins. The fact that the antagonism of LTC₄ effects by LTD₄ in vivo required that the 2 ligands be administered simultaneously (see Fig E3) implies that the mechanism likely involves direct competition for the same binding site on platelets or other relevant effectors.

Activated platelets adhere avidly to circulating eosinophils in asthma⁵¹ and especially in AERD,⁵² likely priming them for adhesion and migration to the respiratory tissue. Platelets store IL-33 in their cytosol,⁵³ and they release IL-33 in response to LTC₄ by a mechanism involving CysLT₂R-driven mobilization of cytosolic HMGB1 and autocrine signaling through RAGE.³⁵ Whereas LTC₄-potentiated IL-33 protein localizes principally to AT2 cells 24 hours after LTC₄ challenge,³⁴ intransal LTC₄ also elicits rapid additional increments in IL-33, type 2 cytokine generation, and eosinophil recruitment, all of which depend on rapid CysLT₂R-dependent intrapulmonary platelet recruitment and activation. LTD₄ completely blocked these rapid platelet-and CysLT₂R-dependent events (Fig 5, B–D and F), which is consistent with its ability to prevent LTC₄-induced release of IL-33 from platelets ex vivo (see Fig E1). Notably, virtually the entire increment in eosinophils observed in this short time frame was due to the platelet-adherent fraction, suggesting another platelet-related effect of LTC₄. It is possible that eosinophils provide a vehicle to transport platelet-associated IL-33 to the lung tissue to elicit a rapid amplification of type 2 cytokine generation from ILC2s and potentially additional cell types. Notably, LTC₄ potently synergizes with IL-33 to induce ILC2 activation ex vivo,⁵⁴ and platelet/eosinophil complexes could provide both of these ligands following their recruitment. Additionally, platelet-associated IL-33 could also directly activate eosinophils.⁵⁵

CysLT-dependent respiratory reactions to aspirin and other COX-1–active drugs are the pathognomonic feature of AERD,¹³ a disease that involves both cysLToverproduction and selective hyperresponsiveness to LTE₄-induced bronchoconstriction.^12,56 Ptges^−/− mice display both of these AERD-like properties,^33,38 whereas WT mice show negligible cysLT-specific airway responses.^33,41 Our previous studies demonstrated that several features of AERD-like responses of Df-primed Ptges^−/− mice to inhaled Lys-ASA were blocked by HAMI-3379, including the rapid recruitment and activation of platelets in the lung, IL-33–driven MC activation, and changes in airway physiology,³⁵ as well as rapid increases in IL-5 and IL-13 protein levels. Accordingly, LTD₄ administered by inhalation before Lys-ASA challenge prevented all HAMI-3379–sensitive features of the reaction to Lys-ASA (Fig 6). We cannot exclude an additional significant contribution from ligand-induced downregulation of CysLT₁R expression in vivo, a function that limits signaling in response to LTD₄ in transfected cells,^45,57 and this may account for the sharp reduction in CysLT₁R expression by leukocytes in nasal tissue biopsy specimens from subjects with AERD after therapeutic desensitization to aspirin.⁴⁶ Nonetheless, our findings suggest that LTD₄ in vivo likely terminates signaling through CysLT₂R as an adaptive feature leading to a loss of end-organ responsiveness to endogenous cysLTs. This process may be especially relevant to AERD, in which aspirin-induced reactions are followed by refractoriness to cysLT-induced bronchoconstriction⁵⁸ and in which high-dose aspirin produces a clinical benefit despite increasing the levels of cysLT generation.⁵⁹

Although CysLT₂R is essential to the agonistic effects of LTC₄ on platelets³⁷ and dermal fibroblasts,⁵⁰ it also suppresses CysLT₁R-dependent signaling responses to LTD₄ in dendritic cells^60,61 and MCs.⁴⁷ As expected, deletion of CysLT₂R eliminated the potentiation of OVA-induced BAL fluid eosinophilia elicited by LTC₄, but it markedly potentiated the response to LTD₄, suggesting a loss of CysLT₂R-mediated restraint of CysLT₁R signaling on 1 or more key cell types. Because responses to LTE₄ were not altered, it seems likely that CysLT₂R does not regulate the LTE₄-CysLT₃R pathway that elicits airway inflammation. Notably, whereas the absence of CysLT₂R markedly inhibited the induced expression of IL-33 in response LTC₄, it sharply increased this response to LTD₄, implying that at least 1 intrapulmonary cell type is competent to drive IL-33 induction when the restraining effects of CysLT₂R on CysLT₁R signaling are removed. Thus, depending on the cellular context, CysLT₂R may act to suppress or induce functions relevant to type 2 inflammation. Notably, variants of CysLT₂R that affect expression or ligand binding are associated with asthma and allergic sensitization, and they could be important determinants of whether LTD₄ acts to induce or suppress inflammation and cellular functions.^62,63

Key messages.

• Leukotriene D₄ paradoxically blocks leukotriene C₄–induced platelet activation and accumulation of platelet-adherent eosinophils.

• Leukotriene D₄ may act to terminate reactions to aspirin by preventing pathogenic leukotriene C₄–induced signaling at CysLT₂R.

Abbreviations used

AERD

Aspirin-exacerbated respiratory disease

APC

Allophycocyanin

AT2

Alveolar type 2

BAL

Bronchoalveolar lavage

CysLT

Cysteinyl leukotriene

CysLT₁R

Type 1 receptor for cysLT

CysLT₂R

Type 2 receptor for cysLT

CysLT₃R

Type 3 receptor for cysLT

Df

Extract from Dermatophagoides farina

ERK

Extracellular signal–regulated kinase

FITC

Fluorescein isothiocyanate

5- ILC2

Group 2 innate lymphoid cell

HMGB1

High-mobility box 1

JNK c

Jun terminal kinase

LO

5-Lipoxygenase

LT

Leukotriene

LTC₄S

Leukotriene C₄ synthase

Lys

-ASA Lysine-aspirin

MAPK

Mitogen-activated protein kinase

MC

Mast cell

NFAT

Nuclear factor of activated T

OVA

Ovalbumin

PE

Phycoerythrin

PRP

Platelet-rich plasma

PTGES

Prostaglandin E₂ synthase

RAGE

Receptor for advanced glycation end products

R_L

Lung resistance

TXA₂

Thromboxane A₂

WT

Wild-type