Cysteinyl leukotriene E4 activates human group 2 innate lymphoid cells and enhances the effect of prostaglandin D2 and epithelial cytokines

Background Group 2 innate lymphoid cells (ILC2s) are a potential innate source of type 2 cytokines in the pathogenesis of allergic conditions. Epithelial cytokines (IL-33, IL-25, and thymic stromal lymphopoietin [TSLP]) and mast cell mediators (prostaglandin D2 [PGD2]) are critical activators of ILC2s. Cysteinyl leukotrienes (cysLTs), including leukotriene (LT) C4, LTD4, and LTE4, are metabolites of arachidonic acid and mediate inflammatory responses. Their role in human ILC2s is still poorly understood. Objectives We sought to determine the role of cysLTs and their relationship with other ILC2 stimulators in the activation of human ILC2s. Methods For ex vivo studies, fresh blood from patients with atopic dermatitis and healthy control subjects was analyzed with flow cytometry. For in vitro studies, ILC2s were isolated and cultured. The effects of cysLTs, PGD2, IL-33, IL-25, TSLP, and IL-2 alone or in combination on ILC2s were defined by using chemotaxis, apoptosis, ELISA, Luminex, quantitative RT-PCR, and flow cytometric assays. The effect of endogenous cysLTs was assessed by using human mast cell supernatants. Results Human ILC2s expressed the LT receptor CysLT1, levels of which were increased in atopic subjects. CysLTs, particularly LTE4, induced migration, reduced apoptosis, and promoted cytokine production in human ILC2s in vitro. LTE4 enhanced the effect of PGD2, IL-25, IL-33, and TSLP, resulting in increased production of type 2 and other proinflammatory cytokines. The effect of LTE4 was inhibited by montelukast, a CysLT1 antagonist. Interestingly, addition of IL-2 to LTE4 and epithelial cytokines significantly amplified ILC2 activation and upregulated expression of the receptors for IL-33 and IL-25. Conclusion CysLTs, particularly LTE4, are important contributors to the triggering of human ILC2s in inflammatory responses, particularly when combined with other ILC2 activators.

Group 2 innate lymphoid cells (ILC2s) are recognized as an innate source of type 2 cytokines in the pathogenesis of allergic conditions, such as asthma and atopic dermatitis. 1,2 ILC2s are known to be lymphoid effector cells that do not express rearranged antigen-specific receptors but express CD45, IL-7 receptor a, and chemoattractant receptor-homologous molecule expressed on T H 2 cells (CRTH2) while lacking lineage markers, including CD3, Tcell receptor (TCR) ab, TCRgd, CD14, CD19, CD56, CD11b, and CD11c. Although they can be found in various anatomic locations, their enrichment in mucosal surfaces of the lung, gut, and skin implies a local immunologic role. [2][3][4][5] ILC2s contribute to allergic and inflammatory conditions by producing IL-4, IL-5, IL-13, and GM-CSF. In the gut ILC2s are a source of type 2 cytokines for efficient expulsion of the helminth Nippostrongylus brasiliensis. 1 In the lungs they induce airway hyperresponsiveness and epithelial repair after influenza infection. 3,6 ILC2s not only respond to the epithelial cytokines IL-33, IL-25, and thymic stromal lymphopoietin (TSLP) but also respond to the mast cell lipid mediator prostaglandin D 2 (PGD 2 ). 2,7 It has also been reported that murine ILC2s respond to another group of lipid mediators, cysteinyl leukotrienes (cysLTs), to produce type 2 cytokines, although the effect of cysLTs in human ILC2s is still unclear. 8 CysLTs, including leukotriene (LT) C 4 , LTD 4 , and LTE 4 , are formed from arachidonic acid metabolism. 9 At the nuclear envelope, cytosolic phospholipase A 2 liberates membrane arachidonic acid, which binds to 5-lipoxygease-activating protein. 5-Lipoxygenase catalyzes the formation of LTA 4 by adding an oxygen moiety to arachidonic acid. Activated inflammatory cells, such as eosinophils, basophils, mast cells, and alveolar macrophages, possessing LTC 4 synthase can synthesize LTC 4 rapidly through conjugation of LTA 4 with reduced glutathione levels. 10 After extracellular export, LTC 4 is Abbreviations used CRTH2: Chemoattractant receptor-homologous molecule expressed on T H 2 cells CSF1: Macrophage colony-stimulating factor cysLT: Cysteinyl leukotriene CysLT 1 : Cysteinyl leukotriene receptor 1 CysLT 2 : Cysteinyl leukotriene receptor 2 ILC2: Group 2 innate lymphoid cell LT: Leukotriene PGD 2 : Prostaglandin D 2 TCR: T-cell receptor TSLP: Thymic stromal lymphopoietin converted to LTD 4 and then LTE 4 by means of sequential removal of the glutamic acid moiety, followed by the glycine moiety. LTD 4 is the most potent airway muscle contractant with the shortest half-life (in minutes); in contrast, LTE 4 is stable and the dominant LT detected in biological fluids. 11 Monitoring LTE 4 levels in urine, sputum, and exhaled air is an index of the activity of the cysLT synthesis pathway. 12 The surge in cysLT production associates with an increase in microvascular permeability, eosinophil recruitment, mucus hypersecretion, bronchoconstriction, and cell proliferation. 13 The role of cysLTs in the pathogenesis of allergic conditions, such as asthma, allergic rhinitis, urticaria, and other inflammatory conditions, has been well studied. 9 We reported recently that cysLTs potentiated the proinflammatory functions of T H 2 cells in response to PGD 2 . 14,15 Two G protein-coupled receptors for cysLTs have been characterized and designated as cysteinyl leukotriene receptor 1 (CysLT 1 ), with high binding affinity for LTD 4 , and cysteinyl leukotriene receptor 2 (CysLT 2 ), with similar affinity for LTD 4 and LTC 4 , but both receptors have low affinity for LTE 4 . 16,17 Other potential receptors for LTE 4 include adenosine diphosphate-reactive purinoceptor P2Y 12 , with the highest homology to CysLT 1 (32%) and GPR99. 18,19 CysLT 1 mediates bronchoconstriction and also a range of proinflammatory effects, including activation and migration of leukocytes. 20,21 CysLT 1 antagonists, most notably montelukast, are used to control asthma and allergic rhinitis. Definition of the role of cysLTs and their receptors in human ILC2s will improve our understanding of the pathogenic mechanisms of ILC2-mediated allergic inflammation and indicate potential novel therapeutic strategies.
In this study we explored the effect of cysLTs and their receptors on human ILC2s, particularly when combined with other ILC2 stimulators. We investigated whether human ILC2s express functional CysLT 1 and whether expression is increased by ILC2s from patients with atopic dermatitis. We assessed the effects of cysLTs on cytokine production, migration, and apoptosis of cells in the presence and absence of the CysLT 1 antagonist montelukast. Finally, we assessed whether cysLTs show a synergistic effect with PGD 2 and epithelial cytokines in activating the cells. Our study provides a broad understanding of the role of cysLTs in ILC2-mediated inflammatory responses, particularly in mixed inflammatory environments.

ILC2 cell preparation and culture
Human ILC2s were prepared from human blood from healthy donors and cultured by using a modified method, as described previously. 2 Briefly, PBMCs were isolated by using Lymphoprep gradients (Axis-Shield UK, Dundee, United Kingdom). CD3 1 T cells were predepleted with CD3 microbeads, and the remaining cells were labeled with an antibody mixture. Lineage (CD3, CD14, CD19, CD56, CD11b, CD11c, TCRab, TCRgd, FcεRI, and CD123)-negative, CD45 high , CD127 1 , and CRTH2 1 cells were sorted on a FACSAria III cell sorter (BD Biosciences, San Jose, Calif) and cultured for 5 to 6 weeks in RPMI 1640 containing 100 IU/mL IL-2, 10% heat-inactivated human serum, 2 mmol/L L-glutamine, 100 IU/mL penicillin, and 100 mg/mL streptomycin in the presence of gamma-irradiated PBMCs (from 3 healthy volunteers). Half of the medium was refreshed every 2 to 3 days. The cells were changed to fresh medium without IL-2 before treatment. Most ILC2s used in the study were derived from healthy donors, except where indicated specifically.
Adult (30-88 years) patients with atopic dermatitis received a diagnosis based on the United Kingdom refinements of the Hanifin and Rajka diagnostic criteria, and the disease severity score was defined by using SCORAD. None of the patients were receiving systemic therapy at the time of the sample acquisition. Use of human tissue samples was approved ethically by the Oxford Clinical Research Committee.

Human mast cell culture and activation
Human mast cells were cultured from CD34 1 progenitor cells and treated with human IgE and goat anti-human IgE in the presence or absence of MK886, as described previously. 14 Cell supernatants were collected and LTE 4 levels were measured with an ELISA, or the supernatants were stored at 2808C until used as mast cell supernatants for the treatment of ILC2s.

Chemotaxis assays
Cell migration assays were conducted, as described previously. 7 Briefly, ILC2s (approximately 5 3 10 4 cells/well) and treatment reagents were loaded into the upper and lower chambers, respectively, in a 5-mm pore-sized ChemoTx plate (Neuro Probe, Gaithersburg, Md). After incubation for 1 hour, migrated cells in the lower chambers were treated with a Cell Titer-Glo Luminescent Cell Viability Assay kit (Promega, Madison, Wis) and quantified by using a FLUOstar OPTIMA luminescence plate reader (BMG LabTech, Cary, NC).

Apoptosis assay
ILC2s (approximately 5 3 10 5 cells per condition) were harvested after different treatments and transferred to annexin-binding buffer, followed by incubation with phycoerythrin-Annexin V/propidium iodide at room temperature, according to the manufacturer's instructions (Invitrogen, Carlsbad, Calif). The stained cells were analyzed with an LSR Fortessa flow cytometer (BD Biosciences).

Multiplex bead array
After treatment for 4 hours, concentrations of selected cytokines in the supernatants of ILC2 (approximately 6

Flow cytometric analysis
PBMCs from patients or ILC2s from cultures were fluorescently labeled with antibodies and acquired by using Summit and FACSDiva software on a CyAn Flow Cytometer (Beckman Coulter, Fullerton, Calif) and LSR Fortessa, respectively. The flow cytometric data were analyzed by using FlowJo software (TreeStar, Ashland, Ore).

Statistics
Data were analyzed by using 1-way ANOVA, followed by the Newman-Keuls test or t test. P values of less than .05 were considered statistically significant. Data are presented as means 6 SEMs.

Human ILC2s express functional CysLT 1 , with higher expression in atopic subjects
To investigate the effect of cysLTs on human ILC2s, we first examined the expression of cysLT receptors in cells ex vivo. Human ILC2s were prepared from human blood (gating strategy is shown in Fig E1 in this article's Online Repository at www.jacionline.org). 2,7 Transcriptional levels of the established leukotriene receptors CysLT 1 and CysLT 2 and the putative receptors P2Y 12 and GPR99 in cells were measured (Fig 1, A). Similar to what has been previously observed in T H 2 cells, 22 the transcriptional level of CysLT 1 in ILC2s was high. Compared with CysLT 1 , the level of CysLT 2 was much lower in the cells. For P2Y 12 and GPR99, only trace levels of mRNA (Fig 1, A) but not protein (see Fig E2 in this article's Online Repository at www.jacionline.org) were detected. Expression of CysLT 1 on the cell surface was confirmed by using flow cytometry (Fig 1, B).
Because ILC2s play an important part in the initiation and maintenance of allergic inflammation in atopic lesions, 2,5 we further compared the level of CysLT 1 in ILC2s isolated ex vivo from patients with atopic dermatitis and healthy control subjects (Fig 1, C, and see Table E2 in this article's Online Repository at www.jacionline.org). CysLT 1 expression was significantly higher in cells from atopic subjects at both the protein (Fig 1, D) and mRNA (Fig 1, E) levels.

LTs mediate migration of human ILC2s
CysLTs are strong chemotactic agents for eosinophils, monocytes, T H 2 cells, and hematopoietic progenitor cells. 15,[22][23][24][25] To determine the role of cysLTs on ILC2 migration, we tested the effect of LTC 4 , LTD 4 , and LTE 4 using a transwell chemotaxis assay and compared them with IL-33, an established ILC2 chemokine. 2 All 3 cysLTs induced ILC2 chemotaxis in a dose-dependent manner, peaking at about 3 nmol/L for LTC 4 and LTD 4 and 10 nmol/L for LTE 4 (Fig 2, A). The maximum response achieved with LTE 4 was greater than that of the other 2 cysLTs. LTs, particularly LTE 4 , showed a significantly stronger effect than IL-33 on ILC2 chemotaxis. The combination of LTE 4 and PGD 2 enhanced the chemotaxis, but this additive effect was not observed when LTE 4 was combined with IL-25, IL-33, or TSLP (Fig 2, B). The chemotactic effect of cysLTs in ILC2s was inhibited by montelukast.

CysLTs promote survival of human ILC2s
LTs contribute to eosinophil and neutrophil survival during inflammatory responses. 26,27 We next examined the effect of cysLTs on ILC2 survival using an apoptosis assay. Removal of IL-2 led to a dramatic increase in the number of Annexin V-positive ILC2s (Fig 3, A). The increase in Annexin V binding was reduced by LTD 4 (50 nmol/L) or LTE 4 (50 nmol/L) within 20 hours. However, because of the degradation of early apoptotic cells, the precise potency of cysLTs on ILC2 apoptosis beyond 24 hours could not be determined. The prosurvival effect of cysLTs on ILC2s was concentration dependent, with a half maximal inhibitory concentration at 4, 36, and 1.8 nmol/L for LTC 4 , LTD 4 , and LTE 4 , respectively (Fig 3, B). LTE 4 was more potent in inhibiting ILC2 apoptosis. In contrast, no effect of cysLTs on ILC2 proliferation was detected (see Fig E3 in this article's Online Repository at www.jacionline.org). The antiapoptotic effect of cysLTs was abolished by montelukast (Fig 3, C).

CysLTs induce production of type 2 cytokines by human ILC2s
One of the important proinflammatory roles of cysLTs is enhancing type 2 cytokine production. 8,14,15 To investigate this effect in human ILC2s, we incubated ILC2s with increasing concentrations of LTC 4 , LTD 4 , and LTE 4 . All LTs enhanced IL-13 production in a dose-dependent manner with a similar half maximal effective concentration: 4 6 1.5, 4.1 6 1.9, and 3.9 6 1.1 nmol/L for LTC 4 , LTD 4 , and LTE 4 , respectively (Fig 4, A). Other type 2 cytokines (IL-4 and IL-5) were also induced by cysLTs (Fig 5, data not shown). Because LTE 4 induced greater responses than LTC 4 and LTD 4 by ILC2s, we subsequently focused on LTE 4 . The effect of montelukast was examined to investigate the role of CysLT 1 on the cytokine production induced by cysLTs. Montelukast exhibited dosedependent inhibition, with a half maximal inhibitory concentration at 0.24 nmol/L for 100 nmol/L LTE 4 treatment (Fig 4,  B). Ten to 30 nmol/L was required to completely abolish the effect of LTE 4 on cytokine production.

LTE 4 enhances the effect of PGD 2 and epithelial cytokines on activation of human ILC2s
Our previous report demonstrated that PGD 2 could stimulate the production of multiple cytokines (IL-3, IL-4, IL-5, IL-8, IL-13, GM-CSF, and macrophage colony-stimulating factor [CSF1]) by human ILC2s. 7 The cells were treated with these lipids alone or in combination to further understand the effect of combinations of cysLTs and PGD 2 on ILC2s. Compared with PGD 2 , LTE 4 only weakly increased production of IL-4, IL-5, IL-8, IL-13, GM-CSF, CSF1, and amphiregulin (Fig 5, A). The combination of LTE 4 and PGD 2 exhibited synergistic (IL-4, IL-8, GM-CSF, CSF1, and amphiregulin) or additive (IL-5 and IL-13) enhancement of cytokine production. The effect associated with LTE 4 was removed by montelukast. mRNA levels of type 2 cytokines were measured by using quantitative RT-PCR to confirm the synergistic effect of LTE 4 and PGD 2 at the gene level. The combination of LTE 4 and PGD 2 showed further enhancement in gene transcription levels compared with LTE 4 or PGD 2 alone (see Fig E4 in this article's Online Repository at www.jacionline.org).
Next, we measured the synergistic effect of LTE 4 and epithelial cytokines. Treatment with IL-33, IL-25, and TSLP alone for 4 hours showed only very weak induction of cytokine production by ILC2s (Fig 5, B). However, addition of IL-33, IL-25, or TSLP to LTE 4 significantly augmented cytokine production (IL-4, IL-5, IL-8, IL-13, GM-CSF, CSF1, and amphiregulin). The greatest enhancement was seen when IL-25, IL-33, and TSLP were used in combination with LTE 4 . Intracellular staining for type 2 cytokines in ILC2s ex vivo also confirmed this enhancing effect (see Fig E5 in  IL-2 is a critical cofactor in regulating ILC2 function in patients with lung inflammation. 28 To understand the role of IL-2 in human ILC2 cytokine production, we compared cell treatments in the presence or absence of IL-2 (Fig 6). IL-2 alone had no detectable effect on the production of most cytokines. However, 1 ILC2s detected in PBMCs from healthy subjects and patients with AD (Fig 1, C) and cohort mean fluorescence intensity CysLT 1 (Fig 1, D). FSC, Forward scatter. E, mRNA for CysLT 1 in blood ILC2s. *P < .05 (Fig 1, A,  To further investigate the potential mechanism involved in the synergistic enhancement of ILC2 activation by the combination of LTE 4 with other stimulators, we examined the mRNA levels of the relevant receptors CYSLT1, PTGDR2, IL1RL1, IL17RA, IL17RB, and IL2RA after treatment (Fig 6, B). Although PTGDR2 (CRTH2) was strongly downregulated and CYSLT1 and IL17RB were partially downregulated by the combined stimulations, IL17RA, IL2RA, and IL1RL1 were upregulated. Upregulation of IL1RL1 was particularly strong and showed strong correlation with cell activation detected based on cytokine production (see Fig E8 in this article's Online Repository at www.jacionline.org).

Endogenous cysLTs from activated human mast cells stimulate ILC2s
CysLTs are major lipid mediators produced by mast cells. 29 The effect of endogenously synthesized cysLTs from activated human mast cells on ILC2s was examined to confirm the role of cysLTs in ILC2 biology under physiologic conditions (Fig 7). Only low levels of LTE 4 (approximately 6 ng/2 3 10 6 cells/ mL) were detectable in supernatants from resting mast cells (Fig 7, A). Activation with IgE followed by anti-IgE antibody cross-linking of mast cells produced high LTE 4 levels (approximatley 86 ng/2 3 10 6 cells/mL). Cotreatment of IgE/anti-IgE-activated mast cells with MK886 (10 mmol/L), an inhibitor of 5-lipoxygease-activating protein, during the period of anti-IgE stimulation abolished LTE 4 production (approximately 7.9 ng/2 3 10 6 cells/mL). Using these supernatants to treat ILC2s revealed that the supernatant from IgE/anti-IgE-activated mast cells but not that of resting mast cells induced production of IL-5 and IL-13 (Fig 7, B). The type 2 cytokine production induced by the mast cell supernatant was partially but significantly inhibited by montelukast. Blocking cysLT synthesis in mast cells with MK866 also partially reduced the capacity of the supernatant to induce cytokine production by ILC2s. A similar effect was observed at the mRNA level (Fig 7, C).  (Fig 2, B, n 5 3).

DISCUSSION
Activation of ILC2s leads to production of type 2 cytokines, and therefore they represent a potential source of the dysregulated type 2 cytokine production seen in patients with eosinophilic inflammatory conditions, such as asthma and atopic dermatitis. 1 This group of cells is found in the blood, spleen, intestines, liver, skin, fat-associated lymphoid clusters, and lymph nodes and is enriched in inflamed skin, nasal polyps, and lungs (data unpublished). 1,2,4 Several previous studies have identified the innate epithelial cytokines IL-33, IL-25, and TSLP and the lipid mediator PGD 2 as important stimulators for ILC2s. 5,7 Although cysLTs have been investigated in mice, 8 in the current study we examined the previously unrecognized roles of cysLTs in human ILC2 activation. They promoted cytokine production, induced cell migration, and reduced apoptosis by ILC2s. These effects were also seen in response to endogenous cysLTs from human mast cells. LTE 4 showed the highest efficacy among cysLTs in ILC2 biology, and the effects were blocked by the CysLT 1 antagonist montelukast. The activation of ILC2s was significantly amplified when treated with cysLTs combined with PGD 2 and epithelial cytokines, particularly in the presence of IL-2, suggesting an important proinflammatory role of cysLTs in ILC2-mediated immune responses in the presence of other ILC2 stimulators.
LTs are potent proinflammatory lipids predominantly derived from activated mast cells and basophils. 29 Their biological functions are mediated through binding to a group of G protein-coupled receptors: CysLT 1 and CysLT 2 . 9,10 Because both receptors have only low affinity for LTE 4 based on ligand-binding assays, 16,17 the potent activity of LTE 4 in human ILC2s suggested to us a potential CysLT 1 /CysLT 2 -independent mechanism. P2Y 12 and GPR99 have been proposed as receptors with a preference for LTE 4 , 18,19 but we found that neither is expressed in ILC2s. A recent report demonstrated that CysLT 1 is critically important for responsiveness to LTE 4 within certain types of human cells. 30 This could be the case for human ILC2s because (1) human ILC2s express high levels of CysLT 1 , (2) LTE 4 showed higher efficacy than LTD 4 in human ILC2s, and (3) the activity of LTE 4 in human ILC2s was completely inhibited by montelukast. This phenomenon has also been observed previously in human T H 2 cells. 14,15 However, we still cannot rule out the possibility of involvement of another unknown LTE 4 receptor. If such a putative receptor exists, our data suggest that it should be responsive to montelukast. Mouse ILC2s also express CysLT 1 . 8 However, in contrast to the findings in human ILC2s, the effect of LTE 4 in mice seems weaker than that of LTD 4 . Furthermore, the IL-5-producing ILC2s induced by LTE 4 administered in vivo cannot be inhibited by montelukast in mouse models (although this interaction was not dissected in vitro). These between-species differences in the effects of LTE 4 on ILC2s might be consistent with involvement of a novel LTE 4 receptor.
The level of CysLT 1 in human ILC2s is upregulated in cells from patients with atopic dermatitis, suggesting a potential pathogenic role of cysLTs/CysLT 1 in patients with the disease. Several studies have shown an improvement of atopic dermatitis after treatment with CysLT 1 antagonists. [31][32][33] Furthermore, a significantly higher concentration of LTE 4 was detected in urine of patients with atopic dermatitis or asthma and in sputum from patients with asthma compared with healthy volunteers. [34][35][36] Collectively, these findings suggest that LTE 4 -mediated ILC2 activation could be a critical contributor to allergic inflammation. This role might be particularly important in patients with aspirin-exacerbated asthma, in whom LTE 4 overproduction and a clinical response to cysLT blockade are particularly obvious. 37 It has been well established that epithelial cytokines (IL-33, IL-25, and TSLP) are important stimulators for ILC2 responses, and the lipid mediator PGD 2 is also a strong regulator of ILC2s in type 2 immunity. 2,7,38 Here we have shown that cysLTs have multiple proinflammatory effects on ILC2 migration, apoptosis, and cytokine production. They not only elicited production of type 2 cytokines but also other proinflammatory cytokines, including IL-8, GM-CSF, CSF1, TNF-a, and amphiregulin. These cytokines could, orchestrating with type 2 cytokines, contribute to eosinophilic (GM-CSF and TNF-a), neutrophilic (IL-8, GM-CSF, and TNF-a), and tissue-remodeling (CSF1, TNF-a, and amphiregulin) effects during allergic inflammations. 3,[39][40][41] Such reactions can be confirmed in other types of cell systems inhibited by montelukast. 40 The efficacies of cysLTs were significantly higher than those of innate cytokines (IL-33, IL-25, and TSLP) but lower than that of PGD 2 in human ILC2s in in vitro assays when using them individually, indicating the important role of cysLTs in human ILC2s. In in vitro studies the ILC2 response to the lipid mediators is much faster (ie, hours) than that to the epithelial cytokines (ie, days). 2,4,7 However, the speed of these responses under physiologic conditions will also depend on the timing of enrichment of these stimulators at inflammatory sites. To confirm the role of cysLTs under arguably more physiologic conditions, we examined the effect on ILC2s of endogenously synthesized cysLTs from human mast cells. The ILC2 response to the mast cell supernatant was similar to that seen with exogenously synthesized cysLTs. The only difference was that the response to the mast cell supernatant could not be  (Fig 6, A) or quantitative RT-PCR (Fig 6, B). *P < .05 (n 5 3-5).
J ALLERGY CLIN IMMUNOL VOLUME 140, NUMBER 4 completely blocked by montelukast or by inhibition of cysLT synthesis. This could be caused by the presence of other active mediators released from activated mast cells, such as PGD 2 . 29 Thus cysLTs only partially deliver the stimulating signal from activated mast cells to ILC2s. CysLTs can also be generated by other cells of the innate immune system, such as basophils, eosinophils, and alveolar macrophages, after exposure to allergens, proinflammatory cytokines, and other stimuli during allergic inflammation and also transcellularly by plateletadherent leukocytes in patients with aspirin-exacerbated asthma. 9,29,37 Therefore cysLTs can contribute to IgEindependent innate responses of ILC2s. Given the association of tissue mast cells, basophils, and eosinophils with allergic dermatitis and asthma, the inhibition of cysLT-mediated ILC2 activation might provide a therapeutic opportunity for these diseases potentially when combined with other approaches.
Although cysLTs alone can activate ILC2s, the more pronounced effect was observed when they were used in combination with another lipid mediator (PGD 2 ) or epithelial cytokines (IL-33, IL-25, and TSLP). We suggest this effect is more representative of in vivo conditions in which, after allergen encounter, LTs and prostaglandins are simultaneously secreted by mast cells in parallel with IL-33, IL-25, and TSLP production by a compromised epithelium. Synergistic enhancement between LTE 4 and PGD 2 was also observed in human T H 2 cells. 14,15 Cross-enhancement of the effects of IL-25, IL-33, and TSLP has been reported in nasal epithelial and T H 2 cells. 42,43 However, the effect of cysLTs on the activity of epithelial cytokines is unknown. Here, for the first time, we reported that LTs markedly enhanced the effect of epithelial cytokines on ILC2 cytokine production. Our data demonstrated that the receptors for IL-33 and IL-25 were upregulated by the combination treatment and therefore might contribute to the enhanced effects. In keeping with this, we found a positive correlation between the degree of upregulation of IL1RL1 and enhancement of cytokine production under the same treatment. Further investigation is required to fully understand the mechanisms underlying this synergistic reaction. Another interesting observation is the promoting role of IL-2 on the ILC2 cytokine production in response to the combination of LTE 4 and epithelial cytokines. IL-2 greatly improved the stimulatory effect of LTE 4 in combination with IL-33, IL-25, and TSLP, whereas the effect of IL-2 on individual stimulator or on the combination of LTE 4 and PGD 2 was minimal. In vitro, IL-2 is required in the culture for ILC2 growth and proliferation. In vivo, it has been reported that IL-2 is a cofactor in regulating ILC2 function in pulmonary inflammation and coordinating type 2 cytokine expression in mouse models. 28,44 Increased levels of IL-2 were detected in the lungs of asthmatic patients, and inhaled IL-2 therapy has been shown to induce asthma-like airway inflammation. 45 Our data suggest a critical promoting effect of IL-2 on LTE 4 and epithelial cytokine-induced activation of human ILC2s, which might be relevant to the pathogenesis of asthma and atopic dermatitis.
In summary, the present investigation highlights the multifaceted role of cysLTs, particularly LTE 4 , and their receptors in the activation of human ILC2s. They initiate production of type 2 and other proinflammatory cytokines, induce cell migration, and suppress cell apoptosis. Most importantly, they significantly enhanced the ILC2 response to other established ILC2 activators. The effects of cysLTs can be blocked by the CysLT 1 inhibitor montelukast. Considering that increased levels of cysLTs (LTE 4 ) and CysLT 1 are associated with allergic diseases, such as atopic dermatitis and asthma, the findings in this study support the view that LTs play a pivotal role in ILC2-mediated allergic inflammation under physiologic conditions and represent an important drug target for related disorders, likely as part of therapeutic combinations.