Co-operative signalling through DP₁ and DP₂ prostanoid receptors is required to enhance leukotriene C₄ synthesis induced by prostaglandin D₂ in eosinophils

Abstract

BACKGROUND AND PURPOSE

Prostaglandin (PG) D₂ has emerged as a key mediator of allergic inflammatory pathologies and, particularly, PGD₂ induces leukotriene (LT) C₄ secretion from eosinophils. Here, we have characterized how PGD₂ signals to induce LTC₄ synthesis in eosinophils.

EXPERIMENTAL APPROACH

Antagonists and agonists of DP₁ and DP₂ prostanoid receptors were used in a model of PGD₂-induced eosinophilic inflammation in vivo and with PGD₂-stimulated human eosinophils in vitro, to identify PGD₂ receptor(s) mediating LTC₄ secretion. The signalling pathways involved were also investigated.

KEY RESULTS

In vivo and in vitro assays with receptor antagonists showed that PGD₂-triggered cysteinyl-LT (cysLT) secretion depends on the activation of both DP₁ and DP₂ receptors. DP₁ and DP₂ receptor agonists elicited cysLTs production only after simultaneous activation of both receptors. In eosinophils, LTC₄ synthesis, but not LTC₄ transport/export, was activated by PGD₂ receptor stimulation, and lipid bodies (lipid droplets) were the intracellular compartments of DP₁/DP₂ receptor-driven LTC₄ synthesis. Although not sufficient to trigger LTC₄ synthesis by itself, DP₁ receptor activation, signalling through protein kinase A, did activate the biogenesis of eosinophil lipid bodies, a process crucial for PGD₂-induced LTC₄ synthesis. Similarly, concurrent DP₂ receptor activation used Pertussis toxin-sensitive and calcium-dependent signalling pathways to achieve effective PGD₂-induced LTC₄ synthesis.

CONCLUSIONS AND IMPLICATIONS

Based on pivotal roles of cysLTs in allergic inflammatory pathogenesis and the collaborative interaction between PGD₂ receptors described here, our data suggest that both DP₁ and DP₂ receptor antagonists might be attractive candidates for anti-allergic therapies.

LINKED ARTICLE

This article is commented on by Mackay and Stewart, pp. 1671–1673 of this issue. To view this commentary visit http://dx.doi.org/10.1111/j.1476-5381.2011.01236.x

Introduction

Eosinophil accumulation and subsequent activation at sites of allergic inflammation control the generation and release of diverse lipid and protein mediators critical to the development and perpetuation of allergic processes (Busse and Lemanske, 2001; Rothenberg and Hogan, 2006). Indeed, eosinophils represent a major source of leukotriene (LT) C₄, the intracellular parent of the cysteinyl LTs (cysLTs; LTC₄/D₄/E₄) (Weller et al., 1983; Cowburn et al., 1998; Bandeira-Melo and Weller, 2003). Central to the pathogenesis of allergic diseases, cysLTs cause bronchoconstriction, mucus hypersecretion, increased microvascular permeability, bronchial hyperresponsiveness, eosinophil infiltration and airway remodelling (Drazen and Austen, 1987). LTC₄, as the main eosinophil-derived product of the 5-lipoxygenase pathway of arachidonic acid metabolism in allergic diseases, is formed by the conjugation of LTA₄ with reduced glutathione by LTC₄ synthase. After active transport to the extracellular space, LTC₄ is converted to LTD₄ and LTE₄ through sequential enzymatic removal of glutamic acid and glycine. Even though the eosinophil LTC₄-synthesizing enzymatic pathways are well known (Bandeira-Melo et al., 2002a), it remains of particular interest to fully characterize endogenous allergy-relevant stimuli, receptors and signalling pathways, as well as to understand the intracellular compartmentalization mechanisms that control allergen-induced, eosinophil-driven, LTC₄ synthesis.

Prostaglandin D₂ (PG) D₂ is another key lipid mediator of allergic airway inflammation that is produced following allergen exposure in patients with asthma, atopic dermatitis or allergic rhinitis. Similar to cysLTs, PGD₂ can mimic a number of important features of allergic processes (Pettipher, 2008). In regard to the sensitization phase of an allergic response, it has been postulated that PGD₂ modulates cytokine production by dendritic cells, leading to the polarization of T helper type 2 cell (Th2) cells (Theiner et al., 2006; Hammad et al., 2007). In addition to its immuno-regulatory role, PGD₂ is a highly effective trigger of blood flow changes, promoting oedema formation and therefore contributing to the nasal congestion symptom of allergic rhinitis (Doyle et al., 1990; Widdicombe, 1990). PGD₂ also mediates the typical cell accumulation of late phase allergic responses by functioning as a selective chemoattractant for Th2 cells, basophils and eosinophils (Hirai et al., 2001; Monneret et al., 2001).

The immuno-modulatory and inflammatory functions of PGD₂ are mediated by high-affinity interaction with two receptors, the prostanoid DP₁ receptor and the chemoattractant receptor-homologous molecule expressed on Th2 cells receptor, now referred to as the DP₂ receptor (receptor nomenclature follows Alexander et al., 2009). Although DP₁ and DP₂ receptors bind the same ligand, there is very little homology between the two receptors; DP₁ receptors being members of the prostanoid receptor family that includes EP_1-4, FP, IP and TP, whereas the DP₂ receptors are more closely related to other chemotactic receptors such as the LTB₄ receptors, BLT1 and BLT2, and the C5a receptor (Pettipher, 2008). Furthermore, while DP₁ receptors are coupled to Gα_s proteins and signal through elevation of intracellular levels of cAMP, the DP₂ receptors are coupled to Gα_i and their activation leads to the elevation of intracellular calcium, reduction in cAMP (Sawyer et al., 2002) and downstream activation of phosphatidylinositol-3-kinase (PI3K) (Xue et al., 2007).

Eosinophils co-express the classical DP₁ receptors coupled to adenylyl cyclase, as well as the Pertussis toxin (PTX)-sensitive DP₂ receptors (Monneret et al., 2001). Although the ability of PGD₂ to activate eosinophils while concurrently elevating cAMP levels seems paradoxical, DP₁ receptors appear to be able to co-mediate, with DP₂ receptors, the mobilization of eosinophils from bone marrow as well as chemotaxis (Schratl et al., 2007). Alternatively, it has also been shown that the final chemotaxis-related response of eosinophils to PGD₂ may be, ultimately, determined by a balance between two opposing downstream signalling pathways: the cAMP-dependent, inhibitory pathway activated via DP₁ receptors and the prevailing, stimulatory pathway activated via DP₂ receptors (Monneret et al., 2001; Kostenis and Ulven, 2006; Sandig et al., 2007).

Recently, we have shown that, in addition to its chemotactic activity towards eosinophils, PGD₂ controls allergy-relevant eosinophil activation, particularly the increased LTC₄ synthesizing capacity of these cells (Mesquita-Santos et al., 2006). Indeed, other eosinophil chemoattractants, including eotaxin (CCL11), RANTES (CCL5) and platelet activating factor (PAF) are capable of triggering LTC₄ synthesis within eosinophils through the activation of their cognate Gα_i -coupled chemotactic receptors, such as CCR3 (Bozza et al., 1996; Bandeira-Melo et al., 2001). Therefore, we had initially hypothesized that PGD₂-induced LTC₄ synthesis could be mediated by the stimulatory activation of DP₂ receptors while being counter-balanced by a parallel inhibitory cAMP-dependent DP₁ receptor activation. However, here by employing pharmacological strategies, we have uncovered a novel kind of interaction between the PGD₂ receptor types expressed on eosinophils. In contrast to the PGD₂-driven opposing signalling related to eosinophil chemotactic activities, eosinophil LTC₄ synthesis triggered by PGD₂ appeared to be controlled by the complementary stimulatory events between DP₁ receptor-activated, PKA-driven, lipid bodies and concurrent DP₂ receptor signalling. While PGD₂ emerges as a potent inflammatory mediator of allergic disorders and as an interesting therapeutic target, because of the mandatory dual activation of DP₁ and DP₂ receptors for increasing eosinophil LTC₄ synthesis, either DP₁ or DP₂ receptor antagonists might be highly effective candidates as anti-allergic tools to control cysLTs production regulated by the activation of eosinophils at sites of allergic reactions.

Methods

Animals

All animal care and experimental protocols were approved by Oswaldo Cruz Foundation Animal Welfare Committee. For in vivo experiments, male Swiss mice of 16–20 g were obtained from Oswaldo Cruz Foundation breeding unit (Rio de Janeiro, Brazil).

PGD₂-induced and allergic pleurisy in sensitized mice

As previously described (Mesquita-Santos et al., 2006), mice were sensitized with an s.c. injection (0.2 mL) of ovalbumin (OVA; 50 µg; Sigma, St. Louis, MO, USA) and Al(OH)₃ (5 mg) in 0.9% NaCl solution (saline) at days 1 and 7. Allergic challenge was performed at day 14 by means of an intrapleural injection of OVA (12 µg per cavity; 0.1 mL). Alternatively, sensitized mice were challenged with PGD₂ (35 pmol per cavity), BW245C (35 pmol per cavity), DK-PGD₂ (35 pmol per cavity) or a combination of the latter two agonists (both at 35 pmol per cavity) (all from Cayman Chemicals, Ann Arbor, MI, USA). All stimuli were diluted in sterile saline immediately before use. Control animals were injected with the same volume (0.1 mL) of vehicle. Mice were killed by CO₂ inhalation 24 h after challenge. Pleural fluid was obtained by rinsing cavities with 1 mL of phosphate-buffered saline containing BSA (0.1%). After samples were taken for lipid body analysis, pleural fluid was centrifuged and cell free supernatants were used for the quantification of cysLTs.

Isolation and in vitro stimulation of human eosinophils

Peripheral blood was obtained with informed consent from healthy donors under protocols approved by the ethical review boards of both the Federal University of Rio de Janeiro and the Oswaldo Cruz Foundation (Rio de Janeiro, Brazil). Eosinophils were isolated by negative selection using EasySep™ system (StemCell Technologies Inc., Vancouver, Canada) (purity ∼98%; viability ∼95%) (Bezerra-Santos et al., 2006). Purified eosinophils (2 × 10⁶ cells·mL⁻¹) in Hank's balanced salt solution without Ca²⁺ and Mg²⁺ (HBSS^−/−) were incubated for 1 h in a water bath (37°C) with PGD₂ (25 nM), BW245C (5–625 nM), DK-PGD₂ (5–625 nM), a combination of BW245C and DK-PGD₂ (both at 25 nM) or PAF (10 µM; from Cayman Chemicals). After samples were taken for lipid body analyses, eosinophils were resuspended in Hank's balanced salt solution with Ca²⁺ and Mg²⁺ (HBSS^+/+) and stimulated with 0.1 µM A23187 (Sigma) for another 15 min (37°C). Cell-free supernatants were then collected for cysLT quantification. Each in vitro experiment was repeated at least three times, with eosinophils purified from different donors.

Treatments

Using the pleurisy models, animals were pretreated with i.p. injections of the DP₁ receptor antagonist BWA868C (1 mg·kg⁻¹; Cayman Chemicals), the dual DP₂/TP receptor antagonist ramatroban (also known as Bay-u3405; 1 mg·kg⁻¹; Cayman Chemicals) or the selective DP₂ receptor antagonist Cay10471 (1 mg·kg⁻¹; Cayman Chemicals) 30 min before either PGD₂ or allergic challenges.

For in vitro studies, eosinophils in HBSS^−/− were pretreated for 30 min with the DP₁ receptor antagonist BWA868C (200 nM), the dual DP₂/TP receptor antagonist ramatroban (200 nM), the selective DP2 antagonist Cay10471 (200 nM), two PKA inhibitors H-89 and PKI (both at 10 µM; Calbiochem, La Jolla, CA, USA); PTX (1 µg·mL⁻¹; Calbiochem) or cell-permeable calcium chelator BAPTA-AM (25 µg·mL⁻¹; Sigma). Of note, these pretreatments did not modify basal lipid body content found within cytoplasm of non-stimulated eosinophils or affected eosinophil viability (∼90%) (data not shown).

Quantification of cysLTs

The amount of cysLTs found in cell-free pleural fluid and eosinophil supernatants was measured by the Cysteinyl Leukotriene EIA kit (catalog number 520501; from Cayman Chemicals), according to the manufacturer's instructions.

EicosaCell for intracellular LTC₄ immuno-detection

As previously described (Bandeira-Melo et al., 2001) to localize LTC₄ at its sites of synthesis, in vitro stimulated eosinophils were mixed with a solution of 1-ethyl-3-(3-dimethylamino-propyl) carbodiimide (EDAC; 0.1% final concentration with cells in HBSS^−/−), used to crosslink eicosanoid carboxyl groups to amines in adjacent proteins. After 15 min incubation at room temperature with EDAC to promote both cell fixation and permeabilization, eosinophils were then washed with HBSS^−/−, cytospun onto glass slides and blocked with HBSS^−/− containing 1% BSA for 30 min. Cells were incubated with rabbit anti-LTC₄ antibodies (Cayman Chemicals) or non-immune rabbit IgG (Jackson ImmunoResearch Inc., West Grove, PA, USA) overnight. Routinely, cells were co-incubated with guinea pig anti-adipose differentiation-related protein (ADRP) antibody (1:300 dilution; Fitzgerald Industries, North Acton, MA, USA) to distinguish cytoplasmic lipid bodies within eosinophils. The cells were washed 3× from 10 min with HBSS^−/− containing 1% BSA and incubated with Alexa488-labelled anti-rabbit IgG and Alexa546-labelled anti-guinea pig secondary antibodies for 1 h.

The specificity of the LTC₄ immuno-labelling was confirmed by the: (i) lack of immunofluorescence within PGD₂-stimulated human eosinophils incubated with irrelevant IgG (data not shown); and (ii) lack of LTC₄ immuno-labelling within non-stimulated human eosinophils that were incubated with anti-LTC₄ antibody.

Images were obtained using an Olympus BX51 fluorescence microscope at 100× magnification and photographs were taken with the Olympus 72 digital camera (Olympus Optical Co., Tokyo, Japan) in conjunction with Cell^F Imaging Software for Life Science Microscopy (Olympus Life Science Europa GMBH, Hamburg, Germany). The images were edited using Adobe Photoshop 5.5 software (Adobe Systems, San Jose, CA, USA).

Lipid body staining and enumeration

To enumerate lipid bodies, cells recovered from pleural cavities or human eosinophils were cytocentrifuged (450 rpm, 5 min) onto glass slides. Cells, while still moist, were fixed in 3.7% formaldehyde (in HBSS^−/−; pH 7.4), rinsed in 0.1 M cacodylate buffer (pH 7.4), stained with 1.5% OsO₄ (Sigma) for 30 min, rinsed in distilled H₂O, immersed in 1.0% thiocarbohydrazide for 5 min, rinsed in 0.1 M cacodylate buffer, restained with 1.5% OsO₄ for 3 min, rinsed in distilled water, and then air dried and analysed. Cell morphology was observed and lipid bodies were enumerated by light microscopy. Fifty consecutively scanned eosinophils were evaluated by more than one individual and results were expressed as the number of lipid bodies per eosinophil.

Statistical analysis

Results are expressed as mean ± SEM and were analysed statistically by means of analysis of variance followed by the Newman–Keuls Student test, with the level of significance set at P < 0.05.

Results

Both DP₁ and DP₂ receptors contribute to PGD₂-induced enhanced LTC₄ secretion from eosinophils

We have reported that PGD₂, formerly recognized only as an eosinophil chemoattractant, is also able to trigger eosinophil activation, characterized by enhanced LTC₄ synthesis (Mesquita-Santos et al., 2006). Here, by employing a mouse model of PGD₂-induced eosinophil activation in vivo (Mesquita-Santos et al., 2006), as well as human-purified eosinophils stimulated in vitro with PGD₂, we investigated the molecular and cellular mechanisms involved in PGD₂-induced LTC₄ synthesis within eosinophils. Our initial goal was to identify the specific PGD₂ receptor, either DP₁ or DP₂, the two well recognized, cloned and eosinophil-expressed receptors for PGD₂ that was involved in this response. In vivo, as illustrated in Figure 1, both PGD₂ receptors appeared to contribute to the LTC₄ secretion from eosinophils, inasmuch as pretreatment with either the DP₁ receptor antagonist BWA868C or the dual DP₂/TP receptor antagonist ramatroban displayed high levels of inhibition. BWA868C and ramatroban significantly reduced LTC₄ secretion levels found increased in sites of PGD₂-triggered eosinophilic inflammation by 63 and 81%, respectively (Figure 1A). Similarly, Figure 1B shows that increased amounts of LTC₄ detected in the supernatants of in vitro PGD₂-stimulated human eosinophils were also inhibited by either the DP₁ receptor antagonist BWA868C, the dual DP₂/TP receptor antagonist ramatroban or by the selective DP₂ antagonist Cay10471 (83, 96 and 89% inhibition, respectively), indicating essential roles for each receptor type, which in turn, perhaps by acting synergistically, evoke PGD₂-driven secretion of LTC₄ from eosinophils. Of note, by showing that (i) pretreatments with BWA868C and Cay10471 did not interfere with PAF-induced LTC₄ synthesis by human eosinophils (Table 1), or (ii) the pretreatment with the selective TP antagonist SQ29548 failed to alter LTC₄ synthesis triggered by in vitro stimulation of eosinophils (Figures 1B; 10% inhibition), respectively, we excluded potential non-specific effects of PGD₂ receptor antagonists and demonstrated that the inhibitory effects of ramatroban were dependent on DP₂, rather than TP receptor antagonism.

Figure 1.

Both DP₁ and DP₂ receptors control cysLTs production triggered by PGD₂. In A, sensitized mice were pretreated with BWA868C (1 mg·kg⁻¹) or ramatroban (1 mg·kg⁻¹) and then stimulated with an i.pl. injection of PGD₂ (35 pmol per cavity). Analysis of cysLTs synthesis was performed 24 h after PGD₂ administration. Results are expressed as the means ± SEM from at least six animals. ^†P≤ 0.05 compared with control animals and *P≤ 0.05 compared with PGD₂-injected mice. In B, for in vitro analysis of LTC₄ synthesis, human eosinophils were pretreated for 30 min with BWA868C (200 nM), ramatroban (200 nM), Cay10471 (200 nM) or SQ29548 (200 nM), and then stimulated for 1 h with PGD₂ (25 nM). In vitro results are expressed as the means ± SEM from at least three independent experiments with eosinophils purified from different donors. ^†P≤ 0.05 compared with control. *P≤ 0.05 compared with PGD₂-stimulated eosinophils. cysLT, cysteinyl leukotriene; DP₁, D prostanoid receptor 1; DP₂, D prostanoid receptor 2; i.pl., intrapleural; PGD₂, prostaglandin D₂.

Table 1.

PAF-induced lipid body biogenesis and LTC₄ synthesis by human eosinophils are dependent on PTX-sensitive protein Gα_i and on intracellular calcium mobilization, but not on activation of DP₂ receptors or PKA activity

Conditions	Treatments	Lipid bodies/eosinophil	cysLTs (ng/2 × 106 cells)
Control		7.7 ± 0.4	1.1 ± 0.6
PAF		16.4 ± 1.4†	4.4 ± 1.4†
	+BWA868C	16.9 ± 1.7	4.2 ± 1.3
	+CAY10471	16.7 ± 1.6	4.5 ± 1.2
	+H-89	13.6 ± 2.3	4.5 ± 1.6
	+PTX	10.5 ± 1.4*	1.6 ± 0.8*
	+BAPTA-AM	8.6 ± 0.5*	1.5 ± 0.8*

^†P≤ 0.05 compared with control group.

^*P≤ 0.05 compared with PAF-stimulated eosinophils.

Human eosinophils were pretreated for 30 min with BWA868C (200 nM), CAY10471 (200 nM), H-89 (10 µM), Pertussis toxin (PTX; 1 µg·mL⁻¹) or BAPTA-AM (25 µM), and stimulated with PAF (10 µM). Analysis of lipid body biogenesis and LTC₄ synthesis were performed 1 h after PAF stimulation. Results were expressed as the means ± SEM from at least three different experiments.

cysLT, cysteinyl leukotriene; DP₂, D prostanoid receptor 2; PAF, platelet activating factor.

DP₁ and DP₂ receptors cooperate to trigger PGD₂-driven enhanced LTC₄ secretion from eosinophils

To test the hypothesis that a synergistic mechanism of action between the PGD₂ receptors, DP₁ and DP₂, controls PGD₂-driven induction of LTC₄ secretion from eosinophils, selective agonists of DP₁ receptors (BW245C) or of DP₂ receptors (DK-PGD₂) were applied to in vivo and in vitro systems of eosinophil activation. Figure 2 shows that, alone, neither BW245C nor DK-PGD₂ was able to elicit LTC₄ release from eosinophils at concentrations that PGD₂ by itself is able to trigger LTC₄ secretion both in vivo (35 pmol per cavity; Figure 2A) and in vitro (25 nM; Figure 2B). In contrast, by co-stimulating eosinophils with a mixture of DP₁ and DP₂ agonists, increased cysLTs levels were detected at eosinophilic inflammatory sites (Figure 2A) and human eosinophil-free supernatants (Figure 2B), with comparable magnitude to those found in vivo (Figure 2A) or in vitro (Figure 2B) stimulation with PGD2, reinforcing the possibility of a synergism between DP₁ and DP₂ receptors.

Figure 2.

DP₁ and DP₂ receptors cooperate to trigger cysLTs production. In A, sensitized mice received an i.pl. injection of PGD₂ (35 pmol per cavity), BW245C (35 pmol per cavity), DK-PGD₂ (35 pmol per cavity) or BW245C plus DK-PGD₂ (both at 35 pmol per cavity). Analysis of cysLTs production within pleural fluids was performed 24 h after i.pl. administration. Results are expressed as the means ± SEM from at least six animals. ^†P≤ 0.05 compared with control animals. In B, for in vitro analyses of LTC₄ production in cell-free supernatants, human eosinophils were stimulated for 1 h with PGD₂ (25 nM), BW245C (25 nM), DK-PGD₂ (25 nM) or with a combination of BW245C plus DK-PGD₂ (both at 25 nM). In vitro results are expressed as the means ± SEM from at least three independent experiments with eosinophils purified from different donors. ^†P≤ 0.05 compared with control. P≤ 0.05 compared with PGD₂-stimulated eosinophils. cysLT, cysteinyl leukotriene; DP₁, D prostanoid receptor 1; DP₂, D prostanoid receptor 2; i.pl., intrapleural; PGD₂, prostaglandin D₂.

Cooperation between DP₁ and DP₂ receptors controls LTC₄ synthesis within eosinophil cytoplasmic lipid bodies

Stimulus-dependent extracellular detection of cysLTs depends on sequential intracellular events, including: (i) an initial step of activation and proper compartmentalization of both substrate and enzymatic machinery, culminating in intracellular LTC₄ synthesis; (ii) intracellular transport to direct newly synthesized LTC₄ to secretory pathways; that is followed by (iii) an active carrier-dependent LTC₄ release through the plasma membrane (see Bandeira-Melo et al., 2002a). Therefore, as virtually no cysLTs can be detected in cell-free supernatants of BW245C- or DK-PGD₂-stimulated eosinophils, one can argue that, under these stimulatory conditions, the intracellular step of LTC₄ synthesis actually occurs, without the subsequent LTC4 transport/release events, which would only follow after the co-stimulation of both PGD₂ receptors. By employing EicosaCell technology, a microscopic method that immobilizes and immuno-detects newly synthesized lipid mediators at their sites of synthesis, we have excluded this hypothesis by showing that LTC₄ synthesis itself is elicited in vitro within human eosinophils only if both DP₁ and DP₂ receptors are simultaneously engaged by either PGD₂ itself (Figure 3; top panel) or by the combination of both PGD₂ receptor agonists BW245C and DK-PGD₂ (Figure 3; botton panel). Eosinophils activated only with DP1 or DP2 agonists show no immuno-fluorescent LTC₄ (Figure 3; middle panels) in EicosaCell preparations, indicating that the LTC₄ synthesizing machinery was indeed not activated under these conditions of stimulation of a single PGD₂ receptor.

Figure 3.

LTC₄ synthesis is triggered within eosinophil cytoplasmic lipid bodies by simultaneous activation of DP₁ and DP₂ receptors by either PGD₂ or the combination of BW245C and DK-PGD₂ stimulation of human eosinophils in vitro. EicosaCell images illustrate intracellular immuno-detection of newly formed LTC₄ (green) and of ADRP (red) in PGD₂-stimulated, BW245-stimulated, DK-PGD₂-stimulated or BW245C/DK-PGD₂ co-stimulated human eosinophils (as indicated). Overlay images of identical fields are shown in the right column. Arrows indicate co-localization of immunolabelled synthesized LTC₄ with ADRP-bearing lipid bodies. For EicosaCell analyses, cells were fixed and permeabilized with EDAC and sequentially incubated with anti-LTC₄ and anti-ADRP antibodies and Alexa488-labelled anti-rabbit IgG plus Alexa546-labelled anti-guinea pig secondary antibodies. Images are representative of three independent experiments. ADRP, adipose-differentiation-related protein; DP₁, D prostanoid receptor 1; DP₂, D prostanoid receptor 2; PGD₂, prostaglandin D₂.

A more detailed analysis of these preparations revealed that the intracellular LTC₄-synthesizing compartment within either PGD₂- or BW245C/DK-PGD₂-stimulated eosinophils was in a punctate cytoplasmic pattern, proximate to, but separate from the nucleus, and fully consistent in size and form with eosinophil lipid bodies. In fact, the compartmentalization of newly formed LTC₄ to eosinophil lipid bodies was confirmed by their co-localization with ADRP (Figure 3; top and bottom panels), a lipid body marker protein. Virtually no LTC₄ was immuno-localized within non-stimulated eosinophils (not shown), thus showing that the newly formed lipid bodies of in vitro DP₁/DP₂-engaged eosinophils are the enzymatically fully equipped organelles responsible for the effective LTC₄ synthesis. Of note, eosinophils stimulated solely with BW245C, but not with DK-PGD₂, display a lipid body-enriched cytoplasm as detected by the punctate cytoplasmic ADRP immuno-labelling (Figure 3; middle panel as indicated), suggesting that the initially hypothesized synergistic effect at receptor level between DP₁ and DP₂ does not take place. Instead, PGD₂-induced LTC₄ synthesis by eosinophils appears to be due to distinct mechanistic roles of DP₁ and DP₂ receptors with complementary functional features.

DP₁ activation, but not that of DP₂, evokes biogenesis of eosinophil lipid bodies

To study whether the induction of lipid body assembly driven by DP₁-receptor pathways contributed to PGD₂-induced LTC₄ synthesis, we employed an osmium-based staining methodology that allows the enumeration of these organelles to check for lipid body biogenesis, under either single or combined receptor stimulation. In agreement with EicosaCell images shown previously (Figure 3), Figure 4 shows that selective activation of DP₁ receptors by BW245C triggers rapid (within 1 h) assembly of new lipid bodies within human eosinophils in a dose-dependent manner. DP₂ receptor activation by DK-PGD₂, on the other hand, even when used in high concentrations (Figure 4B) failed to induce eosinophil lipid body biogenesis. No alteration of DP₁ receptor-induced lipid body biogenesis was observed when eosinophils were co-stimulated with both DP₁ and DP₂ receptor agonists (Figure 4A), suggesting that only the DP₁ receptor controls the formation of these LTC₄-synthesizing organelles.

Figure 4.

DP₁, but not DP₂, activation triggers lipid body biogenesis within human eosinophils in vitro. In A, human eosinophils were stimulated with PGD₂ (25 nM), BW245C (25 nM), DK-PGD₂ (25 nM) or with a combination of BW245C plus DK-PGD₂ (both at 25 nM). B shows a dose-response effect of PGD₂ (25 nM), BW245C (25 nM) or DK-PGD₂ (25 nM) on lipid body biogenesis after stimulation of human eosinophils. Analysis of lipid body biogenesis was performed 1 h after stimulation in osmium-stained cells. Results are expressed as means ± SEM from at least three different experiments with eosinophils purified from distinct donors. ^†P≤ 0.05 compared with control. DP₁, D prostanoid receptor 1; DP₂, D prostanoid receptor 2; PGD₂, prostaglandin D₂.

To further investigate the role of the DP₁ receptor on lipid body biogenesis, we evaluated the participation of each receptor on PGD₂-induced lipid body formation (osmium-stained cells) by pretreating PGD₂-challenged sensitized mice (Figure 5A) and PGD₂-stimulated human eosinophils (Figure 5B) with either DP₁ or DP₂ receptor antagonists. In vivo, while the pretreatment with the DP₂ receptor antagonist ramatroban did not affect the number of cytoplasmic lipid bodies found within infiltrating eosinophils of PGD₂-elicited inflammatory reaction, pretreatment with the DP₁ antagonist BWA868C nearly abolished this in vivo biogenic process. Similarly, in vitro pre-treatment of human eosinophils with the DP₁ receptor antagonist BWA868C significantly inhibited PGD₂-induced eosinophil lipid body biogenesis, while two different DP₂ receptor antagonists, ramatroban and Cay10471, failed to modify the lipid body assembly triggered by PGD₂. Thus, it appears reasonable to postulate that the main role of DP₁ receptors in PGD₂-induced LTC₄ synthesis is to initiate the intracellular signalling pathway that leads to the biogenesis of LTC₄ synthesizing organelles in eosinophils.

Figure 5.

DP₁, but not DP₂ receptors, control eosinophil lipid body biogenesis triggered by PGD₂ either in vivo or in vitro. In A, sensitized mice were pretreated with BW868c (1 mg·kg⁻¹) or ramatroban (1 mg·kg⁻¹), and then stimulated with an i.pl. injection of PGD₂ (35 pmol/cavity). Analysis of lipid body biogenesis was performed 24 h after PGD₂ administration in osmium-stained cells. Results are expressed as means ± SEM from at least six animals. ^†P≤ 0.05 compared with control animals and *P≤ 0.05 compared with PGD₂-injected mice. In B, for in vitro analysis of lipid body biogenesis, human eosinophils were pretreated for 30 min with BW868c (200 nM), ramatroban (200 nM), Cay10471 (200 nM) or SQ29548 (200 nM), stimulated for 1 h with PGD₂ (25 nM) and subsequently stained with osmium. In vitro results are expressed as the means ± SEM from at least three different experiments with eosinophils purified from distinct donors. ^†P≤ 0.05 compared with control. *P≤ 0.05 compared with PGD₂-stimulated eosinophils. DP₁, D prostanoid receptor 1; DP₂, D prostanoid receptor 2; i.pl., intrapleural; PGD₂, prostaglandin D₂.

DP₁ receptors signal via PKA activation to trigger eosinophil lipid body biogenesis: a requirement for PGD₂-induced LTC₄ synthesis

To further evaluate the role of DP₁ receptors in PGD₂-induced lipid body-driven LTC₄ synthesis, we have studied the contribution of DP₁ receptor-related cAMP-dependent signalling on PGD₂-induced assembly of new lipid bodies within eosinophils. As shown in Figure 6, in vitro lipid body assembly triggered by stimulation with either PGD₂ (Figure 6A) or BW245C (Figure 7) was consistent with the central role of DP₁ receptors, as pretreatment with H-89 or PKI (10 µM), two non-structurally related inhibitors of PKA activation, decreased the numbers of cytoplasmic lipid bodies found within PGD₂- (Figure 6A; bottom panel) and BW245C-stimulated human eosinophils (Figure 7). While the specificity of PKA involvement in DP₁ receptor-driven effect was strengthened by the lack of effect of H-89 on PAF-induced eosinophil lipid body biogenesis (Table 1), the ability of forskolin, a well-known activator of adenylate cyclase, to trigger rapid formation of lipid bodies within human eosinophils substantiates the role of cAMP/PKA signalling pathway on the regulation of lipid body biogenic process (C. Bandeira-Melo, unpubl. data).

Figure 6.

DP₁ receptor-driven PKA activation cooperates with DP₂-driven Gα_i protein activation and calcium influx to mediate lipid body-driven LTC₄ sytnhesis within human eosinophils triggered by in vitro PGD₂. Human eosinophils were pretreated for 30 min with H-89 (10 µM) and PKI (10 µM) in A, or with PTX (1 µg·mL⁻¹) or BAPTA-AM (25 µg·mL⁻¹) in B and then stimulated with PGD₂ (25 nM). In vitro analysis of LTC₄ production in cell-free supernatants and lipid body biogenesis were analysed 1 h after PGD₂. Results are expressed as the means ± SEM from at least three different experiments with eosinophils purified from different donors. ^†P≤ 0.05 compared with control group. *P≤ 0.05 compared with PGD₂-stimulated eosinophils. DP₁, D prostanoid receptor 1; DP₂, D prostanoid receptor 2; PGD₂, prostaglandin D₂; PTX, Pertussis toxin.

Figure 7.

PKA activation, but not Gα_i protein and calcium influx, mediates lipid body biogenesis within human eosinophils triggered by BW245C in vitro. Human eosinophils were pretreated for 30 min with PTX (1 µg·mL⁻¹), BAPTA-AM (25 µg·mL⁻¹), H-89 (10 µM) or PKI (10 µM), and then stimulated with BW245C (25 nM). Lipid body biogenesis was analysed 1 h after BW245C stimulation. Results are expressed as the means ± SEM from at least three different experiments with eosinophils purified from different donors. ^†P≤ 0.05 compared with control group. *P≤ 0.05 compared with PGD₂-stimulated eosinophils. PTX, Pertussis toxin.

In agreement with DP₁-driven induction of new lipid bodies as a prerequisite to concurrent DP₂-elicited signalling required for successful LTC₄ synthesis, PKA inhibition by either H-89 or PKI pretreatment also consequently reduced LTC₄ synthesis triggered in vitro by the PGD₂ stimulation of human eosinophils (Figure 6A; upper panel).

DP₂ receptor activation signals via PTX-sensitive Gα_i protein and calcium mobilization to prompt DP₁ receptor-driven, newly formed lipid bodies to synthesize LTC₄

To establish how the concurrent activation of DP₂ receptors contributes to PGD₂-induced lipid body-driven LTC₄ synthesis, we have studied the potential role of Gα_i activation and cytoplasmic calcium mobilization elicited by PGD₂. Lipid body assembly triggered in vitro by stimulation with either PGD₂ (Figure 6B) or BW245C (Figure 7) was not modified by pretreatments with inhibitors of either Gα_i activation or cytoplasmic calcium influx, PTX (1 µg·mL⁻¹) and BAPTA-AM (25 µg·mL⁻¹), respectively, ruling out once more the involvement of DP₂ receptors in lipid body formation triggered by PGD₂. As shown in Table 1, PTX and BAPTA-AM were also able to reduce lipid body biogenesis induced by PAF within human eosinophils.

Although playing no role in PGD₂-induced lipid body assembly, the inhibition of Gα_i protein and calcium mobilization by, respectively, PTX and BAPTA-AM pretreatment, reduced LTC₄ synthesis triggered in vitro by PGD₂ stimulation of human eosinophils (Figure 6B; upper panel), indicating that DP₂ receptors, by activating calcium-dependent signalling, converted DP₁ receptor-induced lipid bodies into enzymatically active organelles capable of LTC₄ synthesis.

Eosinophil lipid body-driven LTC₄ synthesis elicited in vivo by allergic inflammation is also mediated by a complementary signalling between DP₁ and DP₂ receptors

To verify whether the cooperative signalling between DP₁ and DP₂ receptors also operates under allergic inflammatory conditions in vivo, we employed a mouse model of allergic inflammation characterized by eosinophil accumulation and activation. As shown in Figure 8, while the DP₁ antagonist BWA868C reduced both eosinophil lipid body biogenesis and increased levels of cysLTs without significantly affecting eosinophil infiltration found in allergic reaction sites, the DP₂ receptor antagonists ramatroban and Cay10471 decreased the numbers of recruited eosinophils and inhibited allergic cysLT production but failed to alter eosinophil lipid body biogenesis; thus, yet again, illustrating the mandatory cooperation between the two PGD₂ receptors to evoke LTC₄ synthesis, as well establishing the relevance of such cooperation to the molecular mechanisms underlying allergy

Figure 8.

Cooperation between DP₁ and DP₂ receptors to trigger lipid body-driven LTC₄ synthesis within human eosinophils also takes place in allergic inflammatory response in vivo. Sensitized mice were pretreated with BWA868C (1 mg·kg⁻¹), ramatroban (1 mg·kg⁻¹) or Cay10471 (1 mg·kg⁻¹), and then challenged with an i.pl. injection of ovalbumin (12 µg per cavity). Analyses of lipid body biogenesis and cysLTs production were performed 24 h after allergic challenge. Results are expressed as means ± SEM from at least six animals. ^†P≤ 0.05 compared with saline-challenged mice and *P≤ 0.05 compared with ovalbumin-challenged mice. cysLT, cysteinyl leukotriene; DP₁, D prostanoid receptor 1; DP₂, D prostanoid receptor 2; i.pl., intrapleural.

Discussion

PGD₂ is now emerging as a potential mediator of allergic inflammatory pathologies, because it modulates the polarization of Th2 cells, oedema formation and eosinophil recruitment (Pettipher, 2008). In addition, PGD₂ is able to directly activate recruited eosinophils, particularly by eliciting the capacity of eosinophils to synthesize LTC₄ at sites of allergic inflammation. While the enzymic pathway by which eosinophils synthesize LTC₄ is well characterized, the pathophysiological stimuli and intracellular signalling cascades that govern such activity remain to be fully elucidated. Seeking such characterization, it is important to consider the evolving understanding of the potential roles that some eosinophil chemoattractants, which participate in the recruitment of eosinophils to sites of allergic inflammation, have as priming stimuli on eosinophil LTC₄ synthesis.

Our previous studies have demonstrated that, besides PGD₂, other eosinophil chemoattractants, such as PAF, CCL11 and CCL5, are secreted in response to allergic challenge and, by acting on their specific receptors expressed on eosinophil membranes, initiate an intracellular cascade leading to enhanced LTC₄ synthesis (Bozza et al., 1996; Bandeira-Melo et al., 2001). Our attempts to characterize the intracellular signalling pathways committed to chemoattractant-induced enhanced LTC₄ synthesis revealed that diverse stimuli-specific intracellular signalling events control LTC₄ synthesis within eosinophils. For instance, while PAF, acting via its PTX-sensitive Gα_i-protein-linked receptor appears to induce LTC₄ synthesis via a downstream signalling involving PKC and phospholipase C (PLC) activation (Bozza et al., 1996; 1997; 1998;), CCL11 and CCL5, acting via Gα_i protein-linked CCR3 receptors, signal via the activation of mitogen-activated proteins kinases and phosphatidylinositide 3-kinase, but not PKC or PLC (Bandeira-Melo et al., 2001). Even though diverse downstream cascades for LTC₄ synthesis can be engaged, common upstream steps triggered by the activation of CCR3 and PAF chemotactic receptors share Gα_i protein- and calcium influx-regulated cellular activities that, besides culminating in cell polarization/migration, also leads to LTC₄ synthesis. Moreover, the well-documented role of PGD₂ in eosinophil trafficking as well as a variety of migration-related cellular responses, including actin polymerization and increased expression of adhesion molecules, depends on the activation of Gα_i proteins and calcium influx (Monneret et al., 2001; Sawyer et al., 2002). Accordingly, we initially hypothesized that the PGD₂ chemotactic DP₂ receptor expressed on eosinophils, signalling via the activation of Gα_i proteins and calcium influx, would be responsible for PGD₂-induced LTC₄ synthesis. However, our data showed that specific DP₂ receptor stimulation, by itself, was not sufficient to trigger LTC₄ synthesis.

Another shared intracellular event triggered by eosinophil chemoattractants that is essential for the successful LTC₄ synthesis is the rapid assembly of new cytoplasmic lipid bodies – a biological process recognized as an acute, highly regulated cellular event that is stimulus- and cell-specific (Bozza et al., 2007). As multifunctional organelles, lipid bodies are a hallmark of leukocyte activation and, together with perinuclear envelope (Bandeira-Melo et al., 2001; Tedla et al., 2003) and phagosomes (Balestrieri et al., 2006), represent a potential intracellular domain for LTC₄ synthesis. Within eosinophils, compartmentalized LTC₄ synthesis triggered by eosinophil chemotactic agents, including PGD₂, has been located specifically within lipid bodies, thereby explaining why the biogenesis of lipid bodies critically affects the biosynthesis of LTC₄ (Bandeira-Melo et al., 2001; 2002b; Mesquita-Santos et al., 2006). Again, different from the other eosinophil chemotactic receptors, DP₂ receptor stimulation alone did not promote PGD₂-induced lipid body biogenesis, which was also not dependent on DP₂ receptor-related Gα_i and calcium signalling. Inasmuch as the lipid body biogenic process is mandatory for LTC₄ synthesis, the inability of DP₂ receptors to trigger lipid body biogenesis in part explains its inability to promote enhanced LTC₄ synthesis within eosinophils.

We found that the discrepancy between DP₂ receptors and the other eosinophil chemotactic receptors, in terms of eliciting LTC₄ synthesis, relies on the more complex PGD₂ receptor system expressed on eosinophils. Besides DP₂ receptors, eosinophils also express the non-chemotactic DP₁ receptor. By dissecting the specific contributions that each receptor makes to PGD₂-induced LTC₄ synthesis, we demonstrated that the PGD₂-elicited rapid de novo assembly of lipid bodies was insensitive to PTX or a calcium chelator, but was largely dependent on the activation of DP₁ receptor-elicited PKA signalling. Of note, the DP₁ receptor activation by PGD₂ signalling through Gα_s proteins leads to increased cAMP and PKA activity – an intracellular signal transducing cascade that is classically related to the inhibition of chemoattractant-induced eosinophil motility (Hirai et al., 2001; Monneret et al., 2001), and consistent with the idea of cAMP elevating agents as powerful anti-inflammatory (Teixeira et al., 1995; Diaz et al., 1996) or pro-resolution (Sousa et al., 2009) agents for the treatment of diseases in which eosinophil accumulation is thought to play a relevant role (Sousa et al., 2009). In this context, it was noticeable that DP₁ receptor-driven, PKA-dependent, newly formed lipid bodies were not able to synthesize LTC₄, as shown in the EicosaCell preparations. In contrast to PAF- or CCL11-induced LTC₄-synthesizing lipid bodies, we demonstrated that under PGD₂ stimulation, compartmentalized LTC₄ synthesis within DP₁ receptor-driven eosinophil lipid bodies, which were assembled under cAMP/PKA regulation, demands concurrent DP₂ receptor stimulation, inasmuch as: (i) antagonists for either receptor were equally able to reduce PGD₂-induced LTC₄ synthesis; (ii) DP₁, but not DP₂ receptor antagonists, inhibited PGD₂-induced lipid body biogenesis; (iii) only by co-stimulating eosinophils with both DP₁ and DP₂ receptor agonists, was PGD₂-induced LTC₄ synthesis mimicked; (iv) inhibition of PKA inhibited PGD₂-induced DP₁ receptor-driven lipid body biogenesis and subsequent lipid body-compartmentalized LTC₄ synthesis; and (v) PTX and BAPTA-AM, while failing to interfere with DP₁ receptor-dependent lipid body biogenesis, inhibited PGD₂-induced lipid body-driven LTC₄ synthesis. Therefore, the molecular mechanisms orchestrating how DP₂ receptor activation converts DP₁ receptor-driven lipid bodies into enzymatically active organelles capable of effective LTC₄ synthesis depend on the coordinated Gα_i activation and calcium mobilization.

Recently, focusing on eosinophil-driven allergic pathologies, we have found that, alongside CCL11, CCL5, PAF and macrophage migration inhibitory factor (Bandeira-Melo et al., 2001; Vieira-de-Abreu et al., 2005; 2010;), PGD₂ is as an endogenous and potent biogenic stimulus of enzymatically active lipid bodies, organelles involved in LTC₄ synthesis by eosinophils (Mesquita-Santos et al., 2006). Here, despite the evidence showing that PGD₂-driven eosinophil migration-related activities are mediated by a balance of opposing intracellular signalling cascades downstream of DP₁ and DP₂ receptor activation within eosinophils (Monneret et al., 2001), we uncovered that the intracellular mechanisms of receptor-mediated PGD₂-induced LTC₄ synthesis rely on the collaborative signalling between both PGD₂ receptors. The PGD₂-elicited LTC₄ synthesis is dependent on the activation of DP₁ receptor-elicited PKA-regulated lipid bodies, in addition to an equally important and concomitant DP₂ receptor-elicited Gα_i/calcium-regulated signaling pathway, which prompts DP₁ receptor-driven, newly formed lipid bodies to synthesize LTC₄.

Collectively, our findings indicate that PGD₂ binding to DP₁ receptors triggers PKA-driven biogenesis of cytoplasmic lipid bodies, but is incapable of activating the LTC₄-synthesizing machinery, which is switched on by concurrent DP₂ receptor activation. Furthermore, by using PGD₂ receptor antagonists, we also demonstrated that during allergen-elicited eosinophilic inflammatory reactions, cysLTs production is also regulated by DP₁/DP₂-orchestrated eosinophil activation, thus indicating that either DP₁ or DP₂ antagonists might be highly effective at controlling eosinophil activation-regulated LTC₄ synthesis at sites of allergic reactions. However, considering the disappointing clinical trial results of the DP₁ receptor antagonist laropiprant in asthmatics and allergic rhinitis patients (Philip et al., 2009), therapies based on dual blockade of DP₁ and DP₂ receptors or PGD₂ synthesis inhibition may display increased beneficial outcome.

Glossary

Abbreviations

5-LO

5-lipoxygenase

ADRP

adipose differentiation-related protein

CRTH2

chemoattractant receptor-homologous molecule expressed on T helper type 2 cell (Th2) cells

cysLTs

cysteinyl leukotrienes

DP₁

D prostanoid receptor 1

DP₂

D prostanoid receptor 2

EDAC

1-ethyl-3-(3-dimethylamino-propyl) carbodiimide

LTC₄

leukotreine C₄

OVA

ovalbumin

PGD₂

prostaglandin D₂

Conflict of interest

The authors have declared that no competing interests exist.

Supporting Information

Teaching Materials; Figs 1–8 as PowerPoint slide.