Prostaglandin E2 Inhibits Group 2 Innate Lymphoid Cell Activation and Allergic Airway Inflammation Through E-Prostanoid 4-Cyclic Adenosine Monophosphate Signaling

Evidence is accumulating that group 2 innate lymphoid cells (ILC2) play an important role in allergic airway inflammation by producing a large amount of type 2 cytokines. But it remains poorly understood how its activities are properly controlled in vivo. Here, we demonstrated that prostaglandin E2 (PGE2) had a profound inhibitory effect on IL-33-induced ILC2 expansion and IL-5 and IL-13 production in vitro. This effect was mimicked by PGE1-alcohol but attenuated by ONO-AE3-208, indicating a selective action through the E-prostanoid 4 (EP4) receptor. In the IL-33-induced asthma model, coadministration of PGE2 or PGE1-alcohol resulted in diminished IL-5 and IL-13 production, reduced eosinophilia and alleviated lung pathology. In contrast, EP4-deficient mice displayed an exacerbated inflammatory response in another ILC2-mediated asthma model induced by Alternaria extract. Mechanistic studies demonstrated that the PGE2-mediated inhibition of ILC2 was dependent on cyclic adenosine monophosphate (cAMP) production. Further downstream, PGE2-EP4-cAMP signaling led to suppression of GATA3 and ST2 expression, which is known to be critical for ILC2 activation. These findings reveal a novel function of PGE2 as a negative regulator of ILC2 activation and highlight an endogenous counter-regulatory mechanism for the control of innate allergic inflammatory responses.

in lung tissues by enhancing the development of eosinophils in the bone marrow and their recruitment to the lung (1). And IL-13 is found to be necessary and sufficient for mounting airway hyperreactivity and for mucus overproduction (4).
T helper 2 cells have been long considered as the major source of type 2 cytokines in asthma and little is known about the role of innate immunity in the pathology of asthma until the recent discovery of innate lymphoid cells (ILC) (5)(6)(7). ILCs share a lymphoid morphology but lack specific antigen receptors and lineage markers. Three groups of ILC have been defined on the basis of their cytokine secretion profiles. Among them, the group 2 ILC (ILC2) is featured by the capacity to produce large amounts of IL-5 and IL-13 in response to epithelial cell-derived cytokines such as IL-33, IL-25, and thymic stromal lymphopoietin (TSLP) (8,9). In addition to a role in protective immunity against parasites, evidence is accumulating that ILC2 participates in the pathogenesis of allergic inflammation. Increased numbers of ILC2 have been observed in inflamed tissues, such as allergic lung tissues in mice (10), and nasal polyps and skin in human subjects (11)(12)(13). In fact, ILC2 represents more than half of the IL-5-and IL-13-producing cells in the lungs of mice subjected to ovalbumin (OVA)-or house dust mite (HDM)induced asthma. The implication of ILC2 in asthma is further demonstrated in various mouse models of asthma induced by Alternaria species, papain, HDM, OVA, and α-GalCer. In these models, ILC2 provokes hyperresponsiveness and goblet cell hyperplasia accompanied with increased eosinophil counts and type 2 cytokines independent of acquired immunity (10,(14)(15)(16)(17)(18).
Apart from IL-33, IL-25, and TSLP, ILC2 function is regulated by other cytokines or lipid mediators. TNF superfamily member TL1A, transforming growth factor β (TGF-β), and IL-1 have been shown to enhance the proliferation, type 2 cytokine secretion and/or chemoactivity of ILC2 (19)(20)(21). Mice deficient for DR3 (TL1A receptor) or lacking epithelial-derived TGF-β1 exhibit diminished airway inflammation after challenge with allergens. In contrast, both type I and type II interferon (IFN) and IL-27 are able to suppress ILC2 function in a manner dependent on the transcription factor STAT1. Accordingly, ILC2-mediated lung inflammation is enhanced in the absence of the IFN-γ receptor in vivo (22)(23)(24). More intriguingly, ILC2 appears to be functionally plastic, adapting to environmental cues. IL-12, for example, is able to convert IL-1-primed ILC2 into IFN-γ-producing ILC1 cells, and this transdifferentiation is reversed by IL-4 (25). As for lipid mediators, ILC2 expresses high levels of CRTH2, a receptor for prostaglandin (PG) D2, and CysLTR1, the high-affinity receptor for leukotriene D4 (LTD4). Both PGD2 and LTD4 provide activating signals for ILC2, which are nonredundant with IL-33. Defective ILC2-mediated type 2 immune responses are observed in the absence of either PGD2 or LT signaling (26)(27)(28)(29). Unlike PGD2, PGI2 serves as a negative regulator for ILC2 function. IL-5 and IL-13 production by IL-33-stimulated ILC2 is potently inhibited by PGI2, whereas deficiency in PGI2 receptor IP leads to exacerbated lung inflammatory response to the Alternaria extract (30). Lipoxin A4 (LXA4), another arachidonic acid metabolite, also appears to exert a suppressive effect on cytokine production by ILC2 through the unconventional ALX receptor. Notably, severe asthma is associated with decreased expression of LXA4 and its receptor ALX (31).
Prostaglandin E2 is one of the most abundantly produced PGs. By acting on one or more receptor subtypes [E-prostanoid (EP) 1-4], PGE2 affects multiple aspects of immune and inflammatory responses (32). Nevertheless, little is known about its impact on ILC cells except for a recent report in which PGE2 has been shown to promote the proliferation and IL-22 production of ILC3 (33). The present study was focused on the role of PGE2 in ILC2 biology. Both IL-33-induced activation of ILC2 and ILC2-mediated lung inflammation were found to be profoundly inhibited by PGE2-EP4 signaling, most likely through downregulation of GATA3 and ST2.
resUlTs Pge 2 inhibits il-33-induced expansion and Type 2 cytokine Production of Pulmonary ilc2 cells To explore the potential impact of PGE2 on the activities of ILC2, murine lung ILC2 was isolated by FACS and cultured under various conditions. As previously reported (34), ILC2 cells stimulated with IL-33 underwent drastic expansion and produced a large amount of IL-5 and IL-13 (Figures 1A,C). While PGE2 displayed no apparent effect on resting ILC2, it markedly inhibited IL-33-induced ILC2 activation. Specifically, IL-33-induced cell expansion was almost completely abolished with the addition of 10-100 nM PGE2 ( Figure 1A). As Annexin V/7AAD staining revealed comparable levels of apoptotic cells in IL-33-stimulated cultures with or without PGE2 (Figure 1B), the reduced cell number in the presence of PGE2 most likely resulted from impaired cell proliferation rather than altered cell viability. When cytokine production was assessed, a similar inhibitory effect was observed with an 70% reduction in both IL-13 and IL-5 in the supernatant of PGE2-treated cultures ( Figure 1C). This decrease was not simply due to fewer ILC2 cells in the cultures. Intracellular staining showed that PGE2 caused a dose-dependent reduction of IL-5-and IL-13-producing cells in total ILC2 cells ( Figure 1D), suggesting that they were functionally inhibited as well. Collectively, these in vitro studies demonstrate that PGE2 acts as a negative regulator for ILC2 activation.
eP4 Mediates the inhibitory effect of Pge 2 on ilc2 Prostaglandin E2 signals through one of the four functionally distinct receptors, designated EP1, EP2, EP3, and EP4 (32). Their expression in ILC2 was examined by RT-PCR. While Ptger3 (EP3) mRNA was hardly detected in ILC2, various levels of expression were seen for the other three receptors, with relatively high abundance for Ptger1 (EP1) and Ptger4 (EP4) (Figure 2A). To determine which receptor was responsible for PGE2-induced inhibition of ILC2, we tested the impact of multiple receptorselective agonists and antagonists on IL-33-stimualted cultures. EP4 agonist PGE1-alcohol was capable of fully mimicking PGE2 in the assays. As such, similar inhibition was achieved with PGE2 Representative dot plots (left) and the percentage of IL-5 + IL-13 + cells (right) are shown. Each experiment was repeated at least three times with duplicates for each group. Data are presented as mean ± SD unless otherwise indicated. **p < 0.01 and ***p < 0.001. and PGE1-alcohol for IL-33-driven ILC2 expansion as shown by cell counts (Figure 2B) and for IL-5 and IL-13 production as determined by either ELISA (Figure 2C) or intracellular staining ( Figure 2D). Sulprostone (EP1/EP3 agonist) and butaprost (EP2 agonist), on the other hand, showed no effect (Figures 2B-D).
In further support of the key role of EP4, the inhibitory effect  (10 μM). The cultures were analyzed on day 6. Live cells were numerated by trypan blue exclusion (B). IL-5 and IL-13 levels in the supernatant were measured by ELISA (c). IL-5 and IL-13-secreting cells were detected by intracellular staining (D). Representative dot plots (left) and the percentage of IL-5 + IL-13 + cells (right) are shown. Each experiment was repeated at least three times with duplicates for each group. Data are presented as mean ± SD unless indicated otherwise. *p < 0.05; **p < 0.01; and ***p < 0.001.
of PGE2 was abrogated with EP4 antagonist ONO-AE3-208 but not with EP1/2/3 antagonist AH6809. Of note, neither the agonists nor the antagonists had significant impact on cell viability ( Figure S1 in Supplementary Material). These results suggest that PGE2 primarily signals through the EP4 receptor in ILC2.

Pge 2 suppresses il-33-induced Type 2 responses In Vivo
In view of its potent inhibitory function in vitro, we sought to determine whether PGE2 affected ILC2-mediated airway inflammation induced by IL-33 (22,23). Intranasal administration of IL-33 induced rapid and robust type 2 responses characterized by ILC2 accumulation, type 2 cytokine secretion, eosinophilia and goblet cell metaplasia. Such responses were largely diminished when PGE2 was coadministrated with IL-33. In comparison to mice treated with IL-33 alone, those receiving IL-33 + PGE2 showed approximately 50% reduction in total cells ( Figure 3A) and 70% reduction in Siglec-F + eosinophils ( Figure 3B) in the bronchoalveolar lavage fluid (BALF). In addition, coadministration of PGE2 led to approximately 60% reduction of both IL-5 and IL-13 in the lavage ( Figure 3C). As much as the ILC2 population was concerned, a much diminished accumulation was detected in the lung tissue of PGE2-treated mice ( Figure 3D). Moreover, ILC2 recovered from these mice contained a significantly lower percentage of IL-5-or IL-13-producing cells ( Figure 3E; Figure S2A in Supplementary Material). Furthermore, hemotoxylin and eosin (HE) and Periodic Acid-Schiff (PAS) staining of the lung tissue demonstrated that coadministration of PGE2 ameliorated IL-33induced lung pathology featured by intensive peribronchial lymphoid infiltration and goblet cell metaplasia ( Figure 3F).
CD4 + T cells represent another important source of type 2 cytokines. To formally exclude their contribution to the airway inflammation induced by short-term treatment with IL-33, flow cytometry was performed to analyze cytokine production by CD4 + T cells. In agreement with previous studies (30), IL-5-or IL-13-producing cells were barely detectable among CD4 + cells in the lung tissue following IL-33 administration, irrespective of the presence or absence of PGE2 ( Figure S2B in Supplementary Material). Therefore, PGE2 primarily exerts its anti-inflammatory effect by acting on ILC2 cells in such a scenario.
To verify the role of EP4 in PGE2-mediated inhibition in vivo, a group of mice received EP4 agonist PGE1-alcohol in conjunction with IL-33. PGE1-alcohol displayed equal potency to PGE2 in the inhibition of IL-33-indcued airway inflammation in all aspects examined, including exudation of eosinophils and other inflammatory cells, accumulation of ILC2, production of IL-5 and IL-13, and lung pathology (Figures 3A-E).
eP4-Deficiency leads to exaggerated asthmatic responses to Alternaria extract Next, we explored whether endogenous PGE2 signal participated in the control of ILC2-mediated airway inflammation. Given the key role of EP4 in PGE2 signaling in ILC2 cells, it was anticipated that its deficiency might resulted in aggravated inflammation. To test this, Ptger4 flox/flox mice were crossed with Vav-Cre mice to specifically delete Ptger4 in hematopoietic cells. The deletion efficiency was verified by the greatly reduced expression of Ptger4 mRNA in total bone marrow cells and CD4 + T cells in knockout versus wild-type mice ( Figure S3 in Supplementary Material). We first compared the wild-type and knockout mice in the IL-33-induced asthma model and found no overt difference between them. In consideration of the possibility that the overwhelming response to IL-33 might prevent the manifestation of potential aggravating effects of EP4 deficiency, we switched to a clinically more relevant model induced by the Alternaria extract, for which ILC2 is known to be a key player (35,36). In comparison to the rapid and potent inflammatory response induced by IL-33, a milder type 2 response was observed following the airway challenge with the Alternaria extract. The numbers of total inflammatory cells in the BALF and ILC2 cells in the lung tissue were reduced by half (Figures 4A,C in comparison with Figures 3A,D). An even dramatic difference was seen in terms of eosinophils in the BALF ( Figure 4B in comparison with Figure 3B). This allowed us to assess any inflammation-promoting effect introduced by EP4 deficiency. Indeed, the EP4-deficient mice showed increased numbers of ILC2, eosinophils and other inflammatory cells when compared to wild-type mice after challenge with the Alternaria extract (Figures 4A-C). Moreover, histological analyses demonstrated more intensive peribranchial lymphoid infiltration and increased mucus production ( Figure 4D). These results indicate that the endogenous PGE2-EP4 signal is actively involved in the regulation of ILC2-mediated airway inflammation.
Pge 2 -eP4-cyclic adenosine Monophosphate (caMP) signaling inhibits the expression of gaTa3 and sT2 E-prostanoid 4 has been shown to signal through two main pathways, either by triggering cAMP production and subsequent activation of PKA (37) or by engaging the PI3K-Akt pathway (38). To determine which one was involved in ILC2 inhibition, we tested small molecular compounds specifically targeting the two pathways. Dibutyryl cAMP (db-cAMP), a cell-permeable cAMP analog, was able to simulate PGE2 in ILC2 cultures, leading to  drastic inhibition of the IL-33-driven cell expansion ( Figure 5A) and IL-5/IL-13 production (Figures 5B,C). On the other hand, the suppressive activity of PGE2 remained unaltered in the presence of the PI3K inhibitor LY294002 (39) ( Figure S4 in Supplementary Material). Therefore, PGE2 primarily functions through the cAMP pathway instead of the PI3K pathway in ILC2 cells.
Similar to Th2 cells, ILC2 relies on the transcription factor GATA3 for its differentiation as well as the maintenance of its  Each experiment was repeated at least three times with duplicates for each group. Data are presented as mean ± SD unless indicated otherwise. *p < 0.05; **p < 0.01; and ***p < 0.001. functionality (40). In Th2 cells, IL-33 has been shown to enhance GATA3 expression, which in turn leads to elevated levels of ST2, a component of IL-33 receptor (41). We wondered whether IL-33 had a similar effect in ILC2 and how it was modulated by the PGE2-EP4-cAMP signaling. Purified ILC2 cells were thus stimulated with IL-33 in the presence or absence of PGE2, PGE1-alcohol, or db-cAMP and examined for GATA3 and ST2 expression by intracellular staining. As shown in Figure 5D, exposure to IL-33 resulted in an elevated level of GATA3 in ILC2. Further reminiscent of its effect in Th2 cells, IL-33 also induced the upregulation of ST2 expression by ILC2 ( Figure 5D). Intriguingly, the augmented expression of both GATA3 and ST2 was found to be abrogated with the addition of PGE2, PGE1-alcohol, or db-cAMP ( Figure 5D). Therefore, there appears to be a self-amplification mechanism in IL-33 signaling in ILC2 and PGE2-EP4-cAMP signaling imposes a break for the positive feedback loop.

DiscUssiOn
The present study revealed a novel function of PGE2 as a negative regulator of ILC2 activation. When added to ILC2 cultures, PGE2 profoundly inhibited IL-33-induced cell expansion and type 2 cytokine production in a dose-dependent manner. Moreover, intranasal administration of PGE2 prevented ILC2 accumulation in the lung and attenuated airway inflammation in an asthma model induced IL-33. As the inhibitory effect was fully mimicked by PGE1-alcohol but reverted by a EP4-selective antagonist, PGE2 appeared to signal exclusively through EP4 in ILC2. In agreement with this speculation, EP4-deficient mice displayed an exacerbated inflammatory response to the airway challenge of Alternaria extract. We went further to demonstrate that the PGE2-EP4 signaling was primarily mediated by cAMP, which in turn suppressed IL-33-induced upregulation of GATA3, a master transcriptional regulator for the ILC2 lineage. An increasing body of evidence indicates an important role of lipid mediators, especially those derived from arachidonic acid in orchestrating ILC2 functions under physiological and pathological conditions (42). Arachidonic acid is generated from cell membrane phospholipids by phospholipase A2 and subsequently converted into leukotrienes by 5-lipoxygenase, lipoxins by 15-lipoxygenase, or PGE2, PGI2, PGD2, PGF2α, and thromboxane A2 by cyclooxygenase (COX)-1/2 and specific PG synthase through two unstable intermediates PGG2 and PGH2 (43). PGD2 and LTD4 are potent activators of ILC2, triggering production of type 2 cytokines on their own or through synergism with IL-25/33 by inducing intracellular Ca 2+ mobilization (26)(27)(28)(29). PGI2 and LXA4, on the other hand, attenuate ILC2 responses to IL-33 via inducing cAMP production and ERK phosphorylation, respectively (30,31). Here, we reported the identification of PGE2, another metabolite of arachidonic acid, as a novel endogenous inhibitor for the activation of murine ILC2 cells. Of note, while this manuscript was in preparation, Maric et al. published their studies of PGE2 on human ILC2, in which PGE2 was shown to exert a similar inhibitory effect (44). Their studies were based exclusively on in vitro and pharmacological approaches. The present study, on the other hand, also provided in vivo and genetic evidence for PGE2-mediated inhibition of ILC2 and its implication in airway inflammation. Interestingly, While PGE2 primarily signals through the EP4 receptor in murine ILC2, both EP2 and EP4 are involved in the response of human ILC2 to PGE2. The species difference may partly result from the much lower expression of EP2 in murine ILC2.
Prostaglandin E2 is known for its diverse and sometimes paradoxical activities on different immune cells at different stages of immune responses (32). For example, while imposing an overall suppressive effect on T cell activation and production, it differentially regulates the lineage specification of CD4 + effector T cells, shifting the balance away from type 1 toward other forms of immunity, such as Th2, Th17, and Treg responses (39,(45)(46)(47). With respect to ILC cells, a recent study by Duffin et al. has demonstrated that PGE2-EP4 signaling promotes the homeostasis and IL-22 production of ILC3. Our finding that PGE2 antagonizes IL-33-induced ILC2 expansion and cytokine production indicates additional complexity of PGE2-mediated immunoregulation.
Prostaglandin E2 and PGI2 display similar yet distinct properties in their action on ILC2. While they both profoundly inhibit ILC2 proliferation and cytokine production, increased cell death is only observed in PGI2 but no in PGE2 cultures. Moreover, although a cAMP-dependent mechanism is shared by PGE2 and PGI2, further downstream targets appear to differ from each other. PGI2 causes a dramatic reduction of Id2 mRNA expression (30), whereas PGE2-mediated ILC2 inhibition is associated with a marked decrease in GATA3 at the protein level. This discrepancy may be partly explained by the different receptors engaged by PGI2 and PGE2. It is well established that GATA3 is essential for early ILC2 development. Besides, sustained GATA3 activity is required for the maintenance and functionality of ILC2 (40,48). Little is known, however, about the mechanisms governing GATA3 expression in ILC2. A recent study has reported that activation of resting ILC2 by TSLP increased the expression of GATA3 (49). We found that GATA3 expression was also enhanced in ILC2 after stimulation with IL-33. This is consistent with two previous reports in which IL-33 has been shown to upregulate GATA3 expression in Th2 and Treg cells, respectively (41,50). In the latter case, treatment with IL-33 even induces the conversion of Treg PGE2 Inhibits ILC2 Activation Frontiers in Immunology | www.frontiersin.org March 2018 | Volume 9 | Article 501 cells into Th2-like cells (50). Of note, IL-33 stimulation also leads to enhanced expression of its receptor component ST2 in Th2, Treg as well as ILC2 cells. The study by Guo et al. further reveals that GATA3 is directly involved in ST2 upregulation in IL-33 stimulated Th2 cells (41). Taken together, there appears to be a self-amplification mechanism in IL-33 signaling; IL-33 induces GATA3 expression, which in turn results in elevated levels of ST2 and increased responsiveness to IL-33. Despite its apparent beneficial effects, such a mechanism can be problematic if left unchecked. Data shown here indicate that the PGE2-EP4-cAMP signaling may serve as a break to prevent hyperactivation of ILC2. Antagonism against IL-33-induced GATA3 expression is centrally positioned in its action. A decrease in GATA3 activity not only directly affects the expression of effector cytokines including IL-5 and IL-13, but also disrupts the self-amplification loop of IL-33 signaling as a result of ST2 suppression. Nevertheless, the actual contribution of reduced GATA3 expression to inhibition of ILC activation remains to be experimentally verified, for example, by observing whether enforced GATA3 expression renders ILC2 resistant to PGE2-mediated inhibition, although it is technically challenging due to the limited number of ILC2 cells available.
Further studies are also warranted to unravel the mechanisms underlying cAMP-induced reduction of GATA3 expression in ILC2. One candidate could be NF-κB activation, which is known to be important for GATA3 expression and Th2 cell differentiation (51). In view of the striking similarities between the transcriptional programs controlling Th2 and ILC2 cells, it would not be surprising if NF-κB plays a similar role in ILC2. Prostagladins, especially PGD2 and PGE2, have been intensively studied for their implications in asthmatic inflammation. While it is generally agreed that PGD2 accelerates the development of asthma (52,53), data are controversial regarding the role of PGE2 in the pathogenesis of asthma. In contrast to the heightened inflammatory responses in COX-1-or COX-2deficient mice (54), mice lacking mPGES1, an enzyme specific for PGE2 synthesis, show reduced inflammation in OVA-induced asthma models (55). Another study, however, has demonstrated that mPGES1 deficiency has no significant impact on allergen sensitization and generation of effector T cells but induces marked remodeling of the pulmonary vasculature during chronic exposure to the HDM-derived allergen Der f, indicative of a beneficial effect of PGE2 in the late phase of asthmatic inflammation (56). Conflicting results have also been obtained in analyses of mice with selective deletion of genes encoding each of the four PGE2 receptors. Kunikata et al. have reported that, while mice lacking EP1, EP2, or EP4 display an allergic response comparable to that of the wild-type littermates, EP3 deficiency causes much more pronounced airway inflammation. In their model, the EP3 signal primarily influences the elicitation stage by suppressing the release of allergic mediators from mast cells and the production of chemokines by epithelial cells (57). Another study demonstrates that PGE2 may also exert a protective effect through EP2 by targeting CD4 + T cells at the sensitization stage (58). Still another study, however, reveals that the EP2 signal actually facilitates asthma development, possibly by enhancing IgE production by B cells (59). The discrepancy may be ascribed to the multifaceted activities of PGE2 on different cell types at different stages of asthma development. Alternatively, it may be reflective of the nature of asthma as an extraordinarily heterogeneous disease. Two major forms of asthma have been defined in clinic. The atopic asthma is characterized by Th2 cell-mediated responses and IgE production to various allergens. The nonatopic asthma, on the other hand, is independent on adaptive immunity (1). Instead, it is often associated with hyperactivation of ILC2. Consistent with the potent inhibitory activity of PGE2-EP4 signaling on ILC2 in vitro, intranasal administration of PGE2 or the EP4-selective agonist PGE1-alcohol attenuates IL-33-induced allergy in vivo as demonstrated by the reduced number of eosinophils and other inflammatory cells in BALF and improved histology of the lung tissue. On the other hand, EP4 deficiency results in a more severe airway inflammation when challenged with Alternaria extract. These data suggest a proresolving role of PGE2 in ILC2-mediated allergic inflammation. It would be interesting to explore any potential defects in the PGE2-EP4 signaling pathway in severe, non-atopic asthma induced, for example, by environmental pollutants, irritants, chronic airway mycosis, or repetitive viral infection.
In conclusion, the present study identifies new cellular targets and mechanisms for PGE2 in the regulation of asthma pathobiology. Its anti-inflammatory actions suggest a potential therapeutic strategy in asthma for which ILC2 plays a predominant role.

cell Preparation
Bronchoalveolar cells were prepared by lavaging lungs with 2 × 0.5 ml PBS. To obtain whole pulmonary cells, finely chopped lung tissues were incubated in RPMI 1640 containing 0.5 mg/ml DNase I (Sigma Aldrich), 1 mg/ml Collagenase D (Roche) for 1 h at 37°C with gentle shaking. Cell suspensions were obtained by passing through a 100 µm strainer. Mononuclear cells were isolated with a Percoll gradient (GE Healthcare) by harvesting cells at densities between 35 and 70% after centrifugation at 500 g for 20 min at room temperature.

airway inflammation Models and analytical Procedures
To induce airway inflammation, mice were anesthetized with pelltobarbitalum natricum and received intranasal treatment with IL-33 (1 µg per mouse) on day 0 and day 2. The animals were sacrificed 24 h after the last challenge. The Alternaria induced asthma model was generated as described (22,28). Briefly, mice were intranasaly challenged with Alternaria extract (100 µg per mouse, Greer Labs) on days 0, 3, and 6. The animals were sacrificed on day 8. BALF was collected after flushing with 2 × 500 μl PBS and analyzed for the presence of inflammatory cells and cytokine levels. The number of ILC2 cells in the lung was determined by flow cytometry. To detect IL-5/IL-13-secreting cells, lung cells were cultured in medium contain 20 ng/ml PMA and 10 µg/ml ionomycin with GolgiStop (BD Bioscience) for 6 h and then analyzed by flow cytometry after surface staining for lineage markers and intracellular staining of IL-5 and IL-13. For histology, lung tissues were fixed in 4% paraformaldehyde and stained following standard protocols for HE and PAS. In some experiments, PGE2 (300 mg/kg) and PGE1-alcohol (500 mg/kg) was administered concomitantly with IL-33 or Alternaria extract.

culture of ilc2s and Measurement of cytokines
Purified pulmonary ILC2s were seeded into 96-well round bottomed cell culture plates and allowed to rest for at least 24 h in RPMI 1640 supplemented with 10% FBS, 0.05 mM 2-mercaptoethanol, 100 µg/ml penicillin and streptomycin, 10 mM HEPES and 10 ng/ml IL-2 and IL-7. The cells were then replated at a density of 2,000-2,500/well and stimulated with 20 ng/ ml IL-33 in the absence of IL-2 and IL-7. When applicable, PGE2 was added at 10-100 nM, sulprostone, butaprost, PGE1alcohol at 1 μM, AH6809 and ONO-AE3-208 at 10 μM, and db-cAMP at 400 µM. After 6 days culture, the concentration of IL-5 and IL-13 in the supernatant was determined using ELISA Ready-SET-Go! Kits (eBioscience). Live cells were counted by trypan blue staining and apoptosis was analyzed by Annexin V and 7AAD staining. For analysis of cytokine production by individual cells, the culture was stimulated for another 6 h using 20 ng/ml PMA and 10 µg/ml Ionomycin with GolgiStop (BD Bioscience) prior to intracellular staining for IL-5 and IL-13.

eThics sTaTeMenT
This study was carried out in accordance with the recommendations of Ethics Committee of Peking University Health Science Center. The protocol was approved by the Ethics Committee of Peking University Health Science Center.

aUThOr cOnTriBUTiOns
YZhou, WW, YG, and YZhang designed the project. YZhou and WW did the experiment. YZhou, WW, and YZhang wrote the manuscript. YG applied the EP4flox/flox mice. CZ contributed to the pathology analysis. YW and HW contributed to establish the asthma model. XS contributed to ILC2s sorting. FigUre s1 | Unaltered cell viability following exposure to PGE2, PGE2 agonists, and/or PGE2 antagonists. ILC2 cell culture was established as described in Figure 2. Cell viability was examined on day 6 by Annexin V and 7AAD staining. Representative dot plots are shown.
FigUre s2 | IL-5 and IL-13 production by ILC2 versus CD4 + T cells in the IL-33-induced asthma models. Airway inflammation was induced as described in Figure 3. IL-5 and IL-13 production was analyzed by intracellular staining after gating on Lin-CD45 + ST2 + ILC2 (a) or TCRβ + (B) cells. Representative dot plots are shown.
FigUre s4 | Independence of PGE2-mediated inhibition on PI3K activation. ILC2 cells were stimulated with IL-33 (20 ng/ml) in the presence or absence of PGE2 (10 nM) with or without the addition of PI3K inhibitor LY294002 (10 µM). IL-5 and IL-13 levels in the supernatant were measured by ELISA.