Prostaglandin E2 Prevents Hyperosmolar-Induced Human Mast Cell Activation through Prostanoid Receptors EP2 and EP4

Background Mast cells play a critical role in allergic and inflammatory diseases, including exercise-induced bronchoconstriction (EIB) in asthma. The mechanism underlying EIB is probably related to increased airway fluid osmolarity that activates mast cells to the release inflammatory mediators. These mediators then act on bronchial smooth muscle to cause bronchoconstriction. In parallel, protective substances such as prostaglandin E2 (PGE2) are probably also released and could explain the refractory period observed in patients with EIB. Objective This study aimed to evaluate the protective effect of PGE2 on osmotically activated mast cells, as a model of exercise-induced bronchoconstriction. Methods We used LAD2, HMC-1, CD34-positive, and human lung mast cell lines. Cells underwent a mannitol challenge, and the effects of PGE2 and prostanoid receptor (EP) antagonists for EP1–4 were assayed on the activated mast cells. Beta-hexosaminidase release, protein phosphorylation, and calcium mobilization were assessed. Results Mannitol both induced mast cell degranulation and activated phosphatidyl inositide 3-kinase and mitogen-activated protein kinase (MAPK) pathways, thereby causing de novo eicosanoid and cytokine synthesis. The addition of PGE2 significantly reduced mannitol-induced degranulation through EP2 and EP4 receptors, as measured by beta-hexosaminidase release, and consequently calcium influx. Extracellular-signal-regulated kinase 1/2, c-Jun N-terminal kinase, and p38 phosphorylation were diminished when compared with mannitol activation alone. Conclusions Our data show a protective role for the PGE2 receptors EP2 and EP4 following osmotic changes, through the reduction of human mast cell activity caused by calcium influx impairment and MAP kinase inhibition.


Introduction
Asthma is a complex chronic inflammatory disease of the airways that involves the activation of many inflammatory and structural cells. Each component releases inflammatory mediators that result in the pathophysiological changes of typical of the condition [1]. Human mast cells (HuMC) are recognized as the key effector cells of allergic and non-allergic inflammation in asthma [2]. In addition to allergens, many non-immunological stimuli activate complex signaling cascades in mast cells that lead to the secretion of a plethora of autacoid mediators, cytokines, and proteases [3].
Exercise-induced bronchoconstriction (EIB) is a condition in which vigorous physical activity triggers acute airway narrowing. EIB occurs in response to a loss of water from the airways caused by hyperventilation associated with exercise. The osmotic theory proposes that the primary effect of airway water loss is the induction of an increased osmolality in the airway surface liquid [4] that stimulates the release of various mediators via mast cell mechanisms. Both the epithelium and eosinophils may be involved in the generation of EIB-related mediators [5,6].
Experimental surrogates for exercise include the inhalation of hyperosmolar agents and mannitol drug powder [7]. The mannitol challenge is an indirect bronchial challenge [8], which exerts an osmotic effect on the airways and consequently has the potential to lead to mast cell activation [7], [9,10], [11]. Thus, it can mimic the effects of exercise on airway fluid osmolarity.
Prostaglandin E 2 (PGE 2 ) is a product of the cyclooxygenase pathway of arachidonic acid metabolism that is produced in mast cells, dendritic cells, epithelial cells, fibroblasts, and macrophages. Clinical studies have shown that experimental treatment with PGE 2 prevents allergen-, exercise-, and aspirin-induced airway obstruction [12], [13]. Furthermore, several studies have shown a link between asthmatic patients and low levels of PGE 2 in isolated airway cells [14], [15], [16], suggesting a homeostatic role for PGE 2 in the control of airway reactivity and/or inflammation. PGE 2 is a highly pluripotent prostanoid displaying a wide range of pro-inflammatory and anti-inflammatory effects in several tissues. Although PGE 2 is a potent pro-inflammatory mediator [17], its role as an anti-inflammatory mediator is now being studied [18,19]. In this context, it opposes the host inflammatory response, which potentially limits collateral damage to neighboring cells and tissues, thereby aiding the resolution of inflammation [20]. This dual effect appears to be dependent on the cell type, the tissue compartment, the state of cellular activation, and the expression pattern of four prostanoid (EP) receptor subtypes [21].
The EP receptors are members of the G protein-coupled receptor (GPCR) family. EP 1 signals through Gaq, which increases Ca 2+ levels. EP 2 and EP 4 signal through Gas to increase cyclic-AMP (cAMP) levels, while EP 3 primarily signals through Gai to decrease cAMP levels. Further diversity among EP receptors is generated in both the EP 1 and EP 3 receptors by alternatively spliced C-terminal variants, as discussed elsewhere [22]. The EP 2 receptor can downregulate antigen-mediated mast cell responses through Gas-dependent production of cAMP, whereas the EP 3 receptor can up-regulate antigen-mediated mast cell responses through enhanced calcium-dependent signaling [23,24]. It has been suggested that differences in EP 2 and EP 3 receptor expression in mast cells could dictate the upregulation or downregulation of antigen-mediated responses by PGE 2 . Thus, the distribution and relative expression of these four receptor subtypes provide a flexible system describing the ability of PGE 2 to evoke pleiotropic, sometimes opposing, tissue and cell actions [25]. Notably, the beneficial in vivo effects of PGE 2 in murine models of allergic asthma might be mediated through EP 2 receptors in airway mast cells [26,27].
This study aimed to evaluate how PGE 2 modulates the response to mannitol through prostanoid receptors as a model of exerciseinduced asthma in human mast cells, and to clarify the related signaling events.

Human Lung Mast Cells Purification and Culture
Mast cells were isolated from lung tissue obtained from patients undergoing lung resection for lung cancer. The study was approved by the Hospital Clinic Committee on Human Clinical Research, and by the Ethics Committee, (expedient number: 2012/7613, Hospital Clinic, Barcelona) and written informed consent was obtained from all patients. Using immunoaffinity magnetic selection, human lung mast cells (HLMCs) were dispersed from macroscopically normal lung obtained within 1 hour of resection from lung cancer patients, as described previously [31]. Final mast cell purity and viability were each 99%. HLMCs were cultured in Dulbecco's modified Eagles medium, 10% FCS, antibiotic/antimycotic solution, SCF (100 ng/ml), IL-6 (50 ng/ml), and IL-10 (10 ng/ml) [31].

Calcium mobilization
Calcium mobilization in LAD2 cells was followed by fluorimetric analysis of cytoplasmic-free calcium with Fluo-4 AM fluorescent dye (Molecular Probes, Invitrogen) as described elsewhere [36]. Briefly, 0.2610 6

RNA extraction and Real-time polymerase chain reaction
Total RNA was extracted with an RNAeasy Mini Kit (Qiagen, Hilden, Germany) from 2610 6 LAD2 cells. We generated complementary DNA from an mRNA High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems) according to the manufacturer's instructions. To amplify EP receptors from LAD2, the primer pairs detailed in Table 1 were used.

Real-time polymerase chain reaction
Real-time polymerase chain reaction (RT-PCR) for EP 1 , EP 2 , EP 3 , EP 4 , IL-8, and tumor necrosis factor (TNF) was performed using the TaqMan Gene Expression Assay (Applied Biosystems) on an ABI-Prism 7300 Sequence Detector (Applied Biosystems). 18S RNA amplification control was used for cycle normalization. Data were analyzed using the 7500 SDS Software (Applied Biosystems). All PCR reactions were set up in triplicate.

Statistical data analysis
All results are expressed as mean 6 standard deviation (SD). After confirming the normality of the sample distribution and performing variance analysis, we used the Student t test to determine significant differences (p value) between two experimental groups.

Mannitol-induced mast cell degranulation is a calcium dependent process
Mannitol was chosen as a hyperosmolar agent because of its ability to induce mast cell degranulation. To obtain the optimal concentration of mannitol for mast cell activation, LAD2 cells were incubated with a range of concentrations for 30 minutes ( Figure 1A). As shown, all mannitol concentrations caused degranulation; therefore, the intermediate concentration (10%) was chosen for further experiments. After mannitol treatment, the expressions of both b-hexosaminidase release and CD63 were measured by colorimetric assay and FACS, respectively, as markers of degranulation. FACS staining allowed us to distinguish degranulated cells and to discard dead cells with mannitol-induced toxic effects. Mannitol treated cells induced CD63 expression on the mast cell surface membrane ( Figure 1B).
Next, we extended the study to CD34+ derived HuMC and HLMC and obtained similar results ( Figure 1C). It has been described that degranulation induced by aggregation of highaffinity Immunoglobulin E receptor (FceRI) is dependent on the influx of extracellular calcium across the cell membrane. In contrast, non-immunological secretagogues can induce degranulation independently of extracellular calcium [37]. Using fluorimetric analysis, our data show that mannitol was able to release calcium from both the extracellular (Figure 2A) and the intracellular compartments ( Figure 2B). We next analyzed late mast cell responses by studying TNF and IL-8 production by RT-PCR after mannitol stimulation compared to PMA and ionomycin stimulation for 6 hours. We demonstrated that mannitol could trigger IL-8 and TNF production ( Figure 2C). Interestingly, mannitol was able to induce PGE 2 secretion as short as 30 minutes as we show in figure 2D. Collectively, the findings demonstrate that mannitol induced early and late events in mast cell activation.

PGE 2 down regulates mannitol-induced-degranulation in mast cells
Next, we studied the modulating effects of PGE 2 on mannitolinduced degranulation in human mast cells. First, we performed quantitative PCR to determine the pattern of expression of EP receptors by LAD2 cells, and then assayed the LAD2 cell lysates by western blot using specific antibodies against the EP receptors. Our data indicate that LAD2 cells express EP 2 , EP 3 , and EP 4 receptors, but not EP 1 receptors ( Figure 3A-B). Once the EP receptor pattern was established, we evaluated the effects of PGE 2 . EP receptor stimulation by PGE 2 has been reported to enhance FceRI-mediated mast cell degranulation via EP 3 in micromolar ranges [38], although the positive effect of PGE 2 seems to vary depending on the mast cell type [39]. Since we used an osmotic stimulus rather than an immunologic stimulus, both the mechanism and intensity may have differed. Therefore, we pre-incubated LAD2 cells with increasing doses of PGE 2 (0.1, 1, and 10 mM) for 10 minutes each, before activating them with 10% mannitol for 30 minutes. A b-hexosaminidase assay was conducted showing that PGE 2 significantly decreased mannitol-induced degranulation at lower doses ( Figure 3C).

PGE 2 exerts a protective effect through EP 2 and EP 4 receptors after mannitol activation
To identify the EP receptors involved in the protective effect, antagonists of prostanoid receptors were assayed. Our results reveal that b-hexosaminidase release was only slightly decreased when low doses of PGE 2 were used, but not at the highest concentration in EP 2 and EP 4 receptor antagonists pretreated cells ( Figure 4A). Thus, it is unlikely that the EP 3 receptor is responsible for the PGE 2 -induced reduction of mannitol-induced degranulation. Conversely, when the EP 2 ( Figure 4B) and EP 4 ( Figure 4C) receptors were free, and the EP 3 receptors were antagonized, mediator release was significantly decreased regardless of concentration. In parallel, the EP 2 and EP 4 receptors mediated increases in cAMP through activation of adenylyl cyclase, while EP 3 receptor has been shown to both inhibit and activate adenylyl cyclase as well as to drive calcium mobilization [21]. Consistent with these data, mannitol-induced calcium release was impaired when EP 3 receptors were antagonized ( Figure 4D). The data support a role for EP 3 receptors in calcium influx triggered by PGE 2 in mast cells.
We extended these studies by examining the role of PGE 2 after mannitol treatment in CD34+ derived mast cells and HLMC. As shown in Figure 5, HuMC and LAD2 cell lines responded similarly, suggesting significant protection when PGE 2 acted via the EP 2 and EP 4 receptors in CD34+ derived-mast cells  Table 1 (A). The EP receptors expression levels were normalized with the b-actin expression level, EP 1 expression was undetectable. Western blot analysis was carried out with specific antibodies against EP 1 , EP 2 , EP 3 , and EP 4 in whole cell lysates from CD34 + derived mast cells and LAD2 cells; blot against b-actin was performed as a loading control (B). PGE 2 titration was carried out before 10% mannitol stimulation in LAD2 cells (C). The experiments are representative of 3 independent assays. Statistical significance (*p#0.05) is relative to mannitol-stimulated cells. doi:10.1371/journal.pone.0110870.g003 ( Figure 5A), and that the EP 2 receptor appears responsible for PGE 2 -driven protection in HLMC ( Figure 5B).

PGE 2 interferes with Phosphatidyl Inositide 3-Kinase and Mitogen-Activated Protein Kinase Signaling after Mannitol Stimulation
We assessed the effects of PGE 2 in the phosphorylation of proteins after mast cell activation by osmotic changes. To do so, we examined the phosphatidyl inositide 3-kinase (PI3K) and mitogen-activated protein kinase (MAPK) pathways by western blot analysis in LAD2 ( Figure 6A) and HMC-1 cells ( Figure 6B). The phosphorylation status of AKT (Protein Kinase B) was assayed as a surrogate marker for PI3K activation. AKT phosphorylation was decreased under conditions where the EP 3 receptor was blocked in mannitol treated cells, indicating a decrease in PI3K activation/OK ( Figure 6A). AKT was constitutively increased in HMC-1 cell lines in which the KIT receptor was mutated and delivered signals independently of ligand engagement (data not shown). PGE 2 prevented mannitol-induced phosphorylation of ERK1/2, p38, and JNK in LAD2 and HMC-1 cell lines when the EP 3 receptor was blocked ( Figure 6A, B). Together, these results indicate that PGE 2 exhibits inhibitory effects on mannitol-induced osmotic activation by binding to EP 2 or EP 4 receptors in human mast cells.

Discussion
This study aimed to evaluate the protective effect of PGE 2 on mannitol-induced mast cell activation as a model of EIB in Asthma, where mannitol was used as a hyperosmolar stimulus. The use of a hypertonic agent stems from the theory that EIB is caused by increased osmolarity of the surface of the airways through the release of proinflammatory mediators [8]. Previous in vitro work on HLMCs showed that hyperosmolar stimulation induced histamine release, suggesting that hyperosmolar mediated release was a mechanism by which exercise-induced hyperventilation might induce asthma [40]. Our results show that mannitol induces mast cell signaling events that are possibly involved in the inflammatory response observed in asthma.
At early stages, mannitol increased degranulation in a calcium dependent manner before IL-8 and TNF alfa production occurred. Mannitol triggered the activation of PI3K and MAPK cascades, which enhanced ERK1/2, p38 and JNK phosphorylation. The MAPK pathway activates transcription factors such as AP-1 that in turn regulate cytokine and metalloprotease production [41]. Additionally ERK1/2 phosphorylates cytoplasmic phospholipase A2, which is involved in the production of the eicosanoid precursor arachidonic acid [42,43]. Interestingly, previous studies reported ERK phosphorylation in airway smooth muscle cells that cause increased production of both IL-1b and granulocyte-macrophage colony-stimulating factor, which are involved in the contractile response and remodeling of the airways in asthma [43]. The role of JNK in asthma is related to extracellular matrix deposition, with its activation causing the release of growth factors such as transforming growth factor beta, which may explain the phenotype transition from fibroblasts to myofibroblasts in the lung [44]. Moreover, p38 regulates the antigen-triggered migration of mast cells and mediates the production of IL-8. PGE 2 is a highly pluripotent prostanoid displaying a wide range of effects, including smooth muscle relaxation and contraction, and both pro-inflammatory and anti-inflammatory properties [21]. These opposing effects are possible due to the presence of at least four subclasses of EP receptors (EP 1-4 ) [45]. It has been reported that CD34+ derived mast cells express the PGE 2 receptors EP 2 , EP 3, and EP 4 [38]. Our data shows that the LAD2 cell line has a similar PGE 2 receptor pattern.
The aim of the study was to evaluate how PGE 2 modulates the response to mannitol through prostanoid receptors as a model of exercise-induced asthma in human mast cells. For that reason we used antagonist of the receptor instead of direct agonist of them. It has to be noted that AH6809, antagonist for EP 2, is also known to interact with DP1, and AH23848, a EP 4 antagonist, can interact also with the TP receptors. DP1 has been suggested to be expressed on murine mast cells having a role on murine mast cell maturation and differentiation [46]. In our experiments, we are dealing basically with mature human mast cell systems subject to a short term incubation with AH6809. No such maturation effect is expected under our circumstances/conditions. Regarding AH23848, there is very little information on the presence of TP receptors on the human mast cells surface. In fact, it has been reported that the TP agonits U-46619 has no effect on human mast cells [47].
We found that when PGE 2 triggers the EP 3 receptor, it exerts a limited protective effect on mannitol-induced mast cell degranulation. In contrast, when PGE 2 acts through EP 2 and EP 4 receptors, mannitol-induced mast cell degranulation and calcium influx are significantly nullified. Our data agree with other studies in which PGE 2 has been shown to work through EP 2 receptors to stabilize lung mast cells after IgE dependent activation [21,47] and with studies reporting that the EP 2 agonist butaprost exerts a protective effect in allergen-sensitized mice [27]. Additionally, a recent study using human bronchial smooth muscle proposes that PGE 2 -induced relaxation is mediated via the EP 4 receptor [48], which contrasts with reported role of the EP 3 receptor in the induction of PGE 2 airway irritability and cough [49].
Gas, the EP 2 and EP 4 receptor stimulation protein, results in adenylate cyclase activation and intracellular cAMP production. Conversely, EP 3 receptor signaling is predominantly coupled to protein Gai and produces reduced cAMP levels [50]. The accumulation of cAMP promoted by EP 2 and EP 4 receptors is associated with inhibition of cell function, whereas intracellular calcium increases induced by the EP 3 receptor are linked to cellular activation [51]. The evidence from this study, along with other reports, supports the notion that PGE 2 stabilizes mast cells through the EP 2 and/or EP 4 receptors, thereby providing control of the deleterious effects of mast cell degranulation in the airways. The presence of various EP 3 isoforms could explain the differential release of mediators in degranulation assays at different PGE 2 concentrations. It has been reported that, by interacting with the EP 3 receptor, higher doses of PGE 2 increase mediator release through IgE dependent mechanisms [52]. In addition, the presence of several EP 3 isoforms might explain the protective effects of EP 3 in suppressing allergic inflammation in mice [53]. Additionaly, it should be noted that the EP receptors expression pattern has been reported to be different in murine mast cells. EP 1 , EP 3 , and EP 4 transcripts have been found in IL-3-dependent murine mast cell line, MC/9 [54] and murine bone marrow derived mast cells [55]. but not EP 2 . Our data in LAD2 cells is supported by the data obtained in CD34+ derived cells and HLMCs where the decrease in mannitol-induced degranulation was significant when the EP 2 and EP 4 receptors were free to interact with PGE 2 .
The mannitol stimulus caused increased activation in both MAPK and PI3K signaling pathways in mast cells. PGE 2 modulated the mannitol phosphorylation profile of these pathways differently according to the receptor that was triggered. Thus, when the EP 3 receptor was involved, ERK1/2, p38, and JNK phosphorylation remained active, while their phosphorylation decreased with EP 2 or EP 4 receptor engagement. Our results suggest that PGE 2 is not only able to modulate early mast cell events through degranulation, but that it can regulate downstream Figure 5. Prostaglandin E 2 impairs mannitol-induced degranulation in human mast cells through the prostanoid receptors EP 2 and/or EP 4 . Human CD34 + derived-mast cell (A) and human lung mast cell (B) degranulation was induced by 10% mannitol and prostaglandin E 2 at 1 mM in the presence or absence of prostanoid receptor antagonists as indicated in the figure at doses indicated in material and methods section. Results are in triplicate and are the mean of three independent experiments expressed as mean 6 SD. Statistical significance (*p#0.05) is relative to mannitol-stimulated cells. doi:10.1371/journal.pone.0110870.g005 events that may perpetuate airway inflammation in diseases such as asthma.
Experimental treatment with PGE 2 prevents exercise-induced airway obstruction [13]. The preventive effects of exogenous effects of PGE 2 on EIB might suggest that an insufficient biosynthesis of endogenous PGE 2 during exercise in asthma patients can contribute to exercise-induced bronchoconstriction. Interestingly, exercise increases PGE 2 release in the airways of healthy subjects [56], but this increase is not detected in asthma patients [57].
In conclusion, we have provided functional in vitro evidence that EP 2 /EP 4 are potential therapeutical targets having a role in the regulation of MC degranulation that may prevent EIB in asthma by nullifying the hyperosmolar-induced degranulation of airway mast cells.