Lipid Mediators of Allergic Disease: Pathways, Treatments, and Emerging Therapeutic Targets

Abstract

Bioactive lipids are critical regulators of inflammation. Over the last 75 years, these diverse compounds have emerged as clinically-relevant mediators of allergic disease pathophysiology. Animal and human studies have demonstrated the importance of lipid mediators in the development of asthma, allergic rhinitis, urticaria, anaphylaxis, atopic dermatitis, and food allergy. Lipids are critical participants in cell signaling events which influence key physiologic (bronchoconstriction) and immune phenomena (degranulation, chemotaxis, sensitization). Lipid-mediated cellular mechanisms including: (1) formation of structural support platforms (lipid rafts) for receptor signaling complexes, (2) activation of a diverse family of G-protein coupled receptors, and (3) mediating intracellular signaling cascades by acting as second messengers. Here, we review four classes of bioactive lipids (platelet activating factor, the leukotrienes, the prostanoids, and the sphingolipids) with special emphasis on lipid synthesis pathways and signaling, atopic disease pathology, and the ongoing development of atopy treatments targeting lipid mediator pathways.

Introduction

Approximately 5 % of eukaryotic genes encode proteins which synthesize and remodel the cellular lipid repertoire [1]. In addition to energy storage and membrane formation, certain lipids mediate cell signaling events. Lipids influence signaling via multiple mechanisms including (1) by concentrating signal transduction complexes into topologically-constrained “lipid rafts,” (2) by transducing signals as primary/secondary messengers, and (3) by acting as kinase/ phosphatase co-factors [2]. Beginning in the 1940s, the pioneering studies of Kellaway and Trethewie first recognized the bronchoconstrictive effects of substances later identified as leukotrienes (LTs) [3]. Over time, the role of lipids in the pathogenesis of allergic disease has continued to expand with each generation of investigators, driven by ever more sophisticated techniques capable of identifying and quantifying diverse lipid mediators. New systems biology-based technologies, including mass spectrometry-based lipidomics, have facilitated a growing appreciation for the dynamic nature of lipid metabolism during immune activation and the linkage of lipid regulating genes with immune-mediated human disease [4–6]. Herein, we review the basic biology and clinical relevance of four classes of immune-modulating lipids (platelet activating factor, the leukotrienes, prostanoids, and the sphingolipids) with a focus on how these molecules influence the pathogenesis of allergic disease.

Platelet-Activating Factor

Platelet-activating factor (PAF, 1-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine) (Fig. 1) is a potent pro-inflammatory and coagulation-inducing phospholipid mediator first identified in basophils by Jacques Benveniste in 1972 [7]. PAF enhances inflammation via immune cell chemotaxis, triggering of de-granulation, and immune cell adhesion to the vascular endothelium [8].

Fig. 1.

Select bioactive lipids—PAF (purple, representative hexadecyl (16:0) species); sphingolipids (pink, representative Sph (d18:1)/C16 Cer and phosphorylated forms); prostanoids (green); leukotrienes (yellow). Cer, ceramide; C1P, ceramide-1-phosphate; LTB4, leukotriene B4; LTC4, leukotriene C4; LTD4, leukotriene D4; LTE4, leukotriene E4; PAF, platelet activating factor; PGD2, prostaglandin D2; PGE2, prostaglandin E2; PGI2, prostaglandin I2/prostacyclin; Sph, sphingosine; S1P, sphingosine-1-phosphate; TXA2, thromboxane A2

Synthesis, Metabolism, and Signaling

Although PAF can be produced via a de novo pathway, the majority of cellular PAF is generated via remodeling of membrane phospholipids (most commonly phosphatidylcholine, PC) by phospholipase A₂ (PLA₂), generating arachidonic acid (AA) and lysophosphatidylcholine (LPC) (Fig. 2). LPC is subsequently converted to PAF by the activity of LPC acetyl-transferase (LPCAT) [9]. PAF is inactivated by PAF-acetylhydrolase (PAF-AH) enzymes (plasma and cytoplasmic PAF-AH) and has a short half-life of ~3–13 min [10]. PAF signals via the G-protein coupled receptor (GPCR) PAF receptor (PAFR) which initiates a signaling cascade resulting in Ca2⁺-induced protein kinase C (PKC) activation. Most cells that produce PAF also express PAFR, suggesting autocrine signaling may mediate many of PAF’s effects [11].

Fig. 2.

Integrated bioactive lipid synthesis and signaling pathways. Sphingolipids (blue), PAF (yellow), prostanoids (pink), leukotrienes (green). Substrates (boxes), regulatory proteins/enzymes (italics), receptors (arches, bold italics), enzymatic activity (solid lines/arrow), regulatory activity (dotted lines). 5-LO, 5-lipoxygenase enzyme; BLT1,-2, LTB4 receptor-1,-2; CDase, ceramidase; Cer, ceramide; CERK, ceramide kinase; C1P, ceramide-1-phosphate; CerS1-6, ceramide synthases1–6; CoA, coenzyme A; cPLA2, cytosolic phospholipase A₂; COX-1/2, cyclooxygenase-1/-2; CystLT1/2, Cysteinyl leukotriene receptors-1/-2; D1-2, PGD2 receptor-1,-2; EP1-4, PGE2 receptor-1,-2,-3,-4; FP, PGF2 receptor; GCase, glucosylceramidase; GCS, glucosylceramide synthase; GluCer, glucoceramide; GlycoSLs, Glycosphingolipids; GPR99, G-protein receptor-99; IP, PGI2 receptor; KDS, 3-Ketodihydrosphingosine Reductase; LPC, lyso-phosphatidylcholine; LPCAT, LPC acetyltransferase; LTB4, leukotriene B4; LTC4, leukotriene C4; LTD4, leukotriene D4; LTE4, leukotriene E4; LTA4H, LTA4 hydroxylase; LTC4S, LTC4 synthase; ORMDLs, ORM1-like proteins; PAF, platelet activating factor; PAFR, PAF receptor; Pase, phosphatase; PGD2, prostaglandin D2; PGE2, prostaglandin E2; PGH2, prostaglandin H2; PGI2, prostaglandin I2/prostacyclin; PGDS, PGD2 synthase; PGES, PGE2 synthase; PGFS, PG2F synthase; PGIS, PGI2 prostaglandin synthase; SM, sphingomyelin; Sph, sphingosine; SK1,2, sphingosine kinases-1,-2; SMase, sphingomyelinase; SMS, sphingomyelin synthase; SPPase, sphingosine phosphate phosphatase; S1P, sphingosine-1-phosphate; S1P_1-5, S1P receptors1-5; SPT, serine palmitoyltransferase; TP, TXA2 receptor; TXAS, TxA2 synthase; TXA2, thromboxane A2

Effects on Cellular Mediators of Allergic Disease

PAF is produced and released by a variety of cells including neutrophils, eosinophils, mast cells (MCs), endothelial cells, fibroblasts, epithelial cells, and endothelial cells [8]. PAF induces platelet aggregation and leukocyte degranulation and adhesion. In the airways, PAF has multiple effects, serving as a potent chemoattractant for neutrophils and eosinophils [12•, 13], promoting vascular permeability and edema [8] and, inducing bronchoconstriction via effects on airway smooth muscle [14, 15].

Eosinophils

Eosinophils are central effector cells in the majority of Th2-mediated diseases, releasing a broad range of cytotoxic inflammatory mediators that can damage infiltrated end-organs [16]. PAF influences eosinophil chemotaxis, vascular adhesion, and activation. Eosinophils produce PAF upon activation while also responding to PAF release via their PAFRs. PAF-mediated eosinophil chemotaxis occurs by two distinct signaling pathways (monomeric and dimeric PAFR signaling). Monomeric PAFR signaling drives eosinophil chemotaxis whereas the dimeric receptor induces degranulation [17]. PAF also increases production of LTC₄, and induces eosinophil release of multiple cytokines including interleukin 13 (IL-13), eotaxin-1, basic fibroblast growth factor, CCL5/RANTES, and platelet-derived growth factor (PDGF) [18].

Mast Cells

MCs are myeloid-derived, tissue-resident immune effector cells that mediate IgE-driven immune responses in most allergic disease processes [19]. All MCs subtypes appear to produce PAF upon activation; however, MC PAFR expression varies in a tissue-specific manner. Lung and peripheral blood MCs express PAFR while dermal MCs do not, however, dermal MCs do express the MRGX2 neuropeptide receptor. Furthermore, in vitro studies have shown PAF induces histamine release from cultured lung and peripheral blood-derived MCs, but not from dermal MCs [12•]. Intriguingly, cutaneous microdialysis studies show cutaneous histamine release following intradermal PAF injection, but this effect was significantly reduced by nerve blockade [20], suggesting that in vivo PAF may indirectly trigger MC degranulation via peripheral nerve release of MC-activating neuropeptides.

Association with Allergic Disease

Asthma

PAF mediates airway hyperresponsiveness (AHR), inflammation, and remodeling. Lung MCs released histamine in response to PAF in a dose-dependent manner [12•]. PAF has been demonstrated to increase airway hyperactivity with blockade of PAF preventing responsiveness of the airway smooth muscle [15]. PAF also drives airway inflammation during both infection and allergen exposure by increasing LTB₄ production. In addition, PAF has a proposed role in airway remodeling including specific effects on smooth muscle proliferation. A provocative recent report indicated that short acting beta 2 agonist (SABA) bronchodilators induce PAF release, possibly contributing to long-term airway inflammation and smooth muscle changes [21].

Anaphylaxis

PAF is a mediator in the pathophysiology of anaphylaxis and is found at significantly higher concentration in patients post-anaphylaxis than in healthy controls [22••]. Plasma PAF-AH activity varies between individuals. Lower activity of PAF-AH was associated with peanut allergy-induced severe anaphylaxis [22••]. Comparing serum histamine, tryptase and PAF levels after anaphylaxis shows that serum PAF is the most specific indicator of the three mediators, as it correlates most accurately with severity of anaphylactic reaction. PAF is elevated in 100 % of patients with severe anaphylaxis; histamine and tryptase were 61 and 75 % respectively [23]. Approximately 70 % of serum PAF-AH is bound to low-density lipoprotein (LDL) and 30 % is bound to high-density lipoprotein (HDL). Decreasing levels of LDL are associated with prolonged PAF half-life. Theoretically, medical therapy to reduce LDL levels could increase the risk of anaphylaxis [24].

Urticaria and Chronic Rhinitis

MC-mediated diseases such as urticaria and rhinitis are also affected by PAF. [25] As noted earlier, while dermal MCs do not appear to directly respond to PAF in vitro, in vivo PAF indirectly activates dermal MCs via neurogenic activation [12•, 20]. In addition, PAF may amplify skin and mucosal inflammation via its chemotactic properties. In individuals with allergic disease, PAFR is significantly upregulated in epithelial and immune system cells. In allergic rhinitis, rhinorrhea and mucous secretion are associated with the increased vascular permeability caused by PAF. In addition, PAF promotes the rapid translocation of inflammatory cells into nasal tissues [17].

Therapeutics

During the 1990s, multiple PAF antagonists (modipafant, WEB2086, SR27417, UK74,505) were evaluated in asthma clinical trials, but none demonstrated clinical efficacy [26]. In contrast, rupatadine, a dual second-generation H1 antihistamine and PAFR blocker has proven clinically efficacious in urticaria, allergic rhinitis, and rhinoconjunctivitis [17, 27]. Randomized trials comparing rupatadine and levocetirizine demonstrate that rupatadine is better tolerated and more effective for chronic urticaria (Table 1) [41, 42]. Rupatadine currently is not available in the USA.

Table 1.

Lipid pathway modulators/therapeutics

Compound	Target(s)	Mechanism of action	Clinical Indications (or disease model)	Status	Effect	References
Rupatadine	PAFR, H1 receptor	Dual affinity receptor competitive inhibitor	AR, CIU	In clinical use, Approved (E.U.)	Reduces rhinorrhea, sneezing, nasal and ocular pruritis and nasal congestion. Also >75 % reduction in hives in CIU pts at 4 weeks.	[17, 28]
Montelukast	CsytL1	Competitive inhibitor of LTC4, LTD4	Asthma, EIB, AR	In clinical use, approved	Improves FEV1, reduces exacerbations and blood eosinophil counts.	[29, 30, 31•]
Zileuton	5-LO	Competitive inhibitor	Asthma, AERD	In clinical use, approved	Improves FEV1, reduced bronchodilator use and blood eosinophil counts.	[32]
OC000459	DP2(CRTH2)	Competitive antagonist	Asthma, eosinophilic esophagitis (EoE)	Clinical trials underway	Reduced late airway response, reduced drop in FEV1, decreased eosinophil count after bronchial allergen challenge; Reduced esophageal eosinophil infiltration.	[33, 34]
Figolimod (FTY720)	S1P1, S1P3, S1P4, S1P5	Pro-drug, functional S1P receptor antagonist	RRMS (asthma, AD, AR)	In clinical use, approved (MS only). (Pre-clinical)	In RRMS, reduced, relapse rate by ~50 %. (In asthma mouse models, IT-treatment suppressed AHR, inflammation. In mouse AD, topical-treatment reduced epidermal hypertrophy, MCs. In mouse model of AR, IN-treatment reduced inflammation, Th2 cytokines.)	[35, 36, 37•, 38]
SK1-1 (BML-258)	SphK1	Competitive inhibitor	(Asthma)	(Preclinical)	(In asthma mouse models, IN-treatment suppressed AHR, inflammation, Th2 cytokine.)	[39]
AAL-R	S1P1, S1P3, S1P4, S1P5	Functional S1P receptor antagonist	(Asthma)	(Preclinical)	(In asthma mouse models, IN-treatment suppressed AHR, inflammation, Th2 cytokine.)	[40]

For preclinical testing, the murine disease model in which the compound was evaluated is bracketed in parentheses and italicized

5-LO 5-lipoxygenase, AD atopic dermatitis, AERD Aspirin-exacerbated respiratory disease, AHR airway hyper-responsiveness, AR allergic rhinitis, CIU chronic idiopathic urticarial, EIB exercise-induced bronchoconstriction, EU European Union, IT intratracheal, MC mast cells, RRMS relapsing remitting multiple sclerosis

Leukotrienes

Leukotrienes (LTs) (Fig. 1), also known as “slow-reacting substance of anaphylaxis”, are a class of immune-modulating eicosanoids that have emerged as useful clinical targets for the treatment of allergic disease [17, 43, 44, 45•]. Like PAF, LTs are not preformed, but rather are rapidly synthesized in response to various stimuli [46–48]. As these molecules were first detected in leukocytes and they share a carbon backbone containing three covalent double bonds (a triene), the substances were dubbed “leukotrienes”—a term credited to Swedish biochemist Bengt Samuelsson [49]. There are two distinct classes of LTs based on structure: (1) dihydroxyl LTs and (2) cysteinyl LTs (cysLTs). LTs play a key role in the pathogenesis of allergic rhinitis [48, 50, 51], asthma [43, 52, 53], and aspirin-exacerbated respiratory disease (AERD) [54, 55].

Synthesis

LTs are synthesized (Fig. 2) de novo from AA in activated leukocytes including eosinophils, MCs, tissue macrophages, and basophils. Activation-mediated calcium transients induce the translocation of the 5-lipoxygenase enzyme (5-LO) to the perinuclear membrane, where it associates with the 5(five)-lipoxygenase-activating protein (FLAP). FLAP, a perinuclear membrane protein, transfers free AA to 5-LO, which converts the fatty acid into the short-lived intermediate LTA₄. LTA₄ is the common precursor for both classes of LTs. LTB₄ is generated via LTA₄ hydroxylase (LTA₄H), and LTC₄ is produced through the addition of a reduced glutathione to LTA₄, a reaction catalyzed by LTC₄ synthase (LTC₄S). LTA₄H is expressed by macrophages, MCs, and neutrophils, whereas LTC₄S is expressed in eosinophils, basophils, MCs, and macrophages. The additional cystLTs are then generated via sequential cleavage of residues from LTC₄. LTC₄ is converted to LTD₄ by the extracellular enzyme γ-glutamyl transpeptidase, and the most stable and abundant cystLT, LTE₄, is produced when a dipeptidase removes the terminal glycine residue from LTD₄ [56].

Cysteinyl Leukotrienes

The cystLTs (LTC₄, LTD₄, LTE₄) (Fig. 1) contain the amino acid cysteine in their structure. They are the most prevalent class of LTs synthesized in eosinophils and MCs and are highly relevant to pathogenesis of allergic disease [48, 57]. The cystLTs and their metabolites can be found in plasma, urine, sputum, bronchoalveolar lavage [58–60]. They exert their effects by binding to the GPCRs CysLT1 and CysLT2, with LTD₄ and LTC₄ having the greatest binding affinity for these receptors (LTE₄ has relatively low affinity) [57]. CystLT1 is found on bronchial smooth muscle and myeloid cells including MCs and macrophages, whereas CystLT2 is found on these cells plus endothelial cells, adrenal medulla, brain and cardiac Purkinje cells. When activated by its ligand, the CystLT1 receptor induces bronchoconstriction, mucus secretion, and edema [57]. In contrast, CystLT2 does not participate in bronchoconstriction but appears to drive inflammation and edema by acting upon platelets, leukocytes, and vascular endothelium [61, 62]. Since LTE₄ is both a bronchoconstrictor and proinflammatory agent, yet has a low affinity for either CystLT1 or CystLT2, investigators have long hypothesized about the potential existence of additional unidentified LTE₄-sensitive receptor(s) [57, 63•]. Recent studies have suggested that LT signaling is far more complex than previously envisioned, with an emerging immune modulatory role of P2Y purinergic receptors (P2Y₂, P2Y₆, P2Y₁₂) and the identification of GPR99/OXGR1 (oxoglutarate receptor 1) as a direct LTE4 receptor in vitro (mouse, human) and in vivo (mouse) [57, 63•, 64]. As LTE₄ is the chief LT detected in inflamed tissues and biological fluids, the emergence of a new putative receptor that could be targeted pharmacologically has excited significant interest in the field [63•].

Dihydroxyl Leukotrienes

LTB₄, the only known member of the dihydroxyl LTs, is primarily synthesized by neutrophils and macrophages (Figs. 1 and 2) [65, 66]. The biological effects of LTB₄ are mediated through the BLT1 and BLT2 receptors. The BLT1 receptor is differentially upregulated in response to stimuli such as LPS and TNFα [67]. Binding of this receptor results in chemotaxis of eosinophils, neutrophils, MC progenitors, CD4⁺, and CD8⁺ T lymphocytes [68]. It has been hypothesized that LTB₄ may have a role in neutrophilic variant asthma, which is resistant to conventional glucocorticoids therapy [65, 69]; however, a recent trial of the potent FLAP inhibitor GSK2190915 failed to affect sputum neutrophils despite significantly reducing LTB₄ levels [69, 70].

Association with Allergic Disease

Asthma

LTs play multiple roles in the pathophysiology of asthma. They induce bronchoconstriction, recruit inflammatory cells, induce plasma extravasation, and drive tissue edema [71]. At a cellular level, LTs enhance allergen and IL-13-dependent allergic lung disease by amplifying levels of Th2-specific cytokines, CCL7 and CCL17, and increasing Th2 cell recruitment to the lungs [72]. They induce smooth muscle contraction, leading to bronchoconstriction [52] and have been found to stimulate airway remodeling [73]. CystLTs have also been implicated in mucous gland secretion and bronchovascular leakage [74], chemotaxis of leukocytes, and increased pro-inflammatory cytokine production [48]. Sputum cystLT metabolite levels directly correlate with asthma severity and are increased in patients during exacerbations, following exercise, and during allergen challenge [53, 58].

Aspirin-Exacerbated Respiratory Disease (AERD)

AERD is classically described as a triad of asthma, chronic rhinosinusitis disease, and nasal polyps with worsening of symptoms with ingestion of non-steroidal anti-inflammatory drugs (NSAIDs). This disease is considered to be non-immunoglobulin E (IgE)-mediated (pseudoallergy). The mechanisms driving AERD pathology include: dysregulation of AA metabolism with increased production of LTs (specifically LTC₄) from bronchial MCs [75], overexpression of CystLT1 receptor on leukocytes in nasal mucosa, and reduced levels of prostaglandin E₂ (PGE₂), which suppresses LT production [76].

Therapeutics

Pharmacologic targeting of the LT pathway has led to the successful development of two classes of clinically useful LT modifiers [43]. Montelukast, zafirlukast, and pranlukast (available only in Japan) are orally bioavailable cystLT1 receptor antagonists (leukotriene receptor anatgonists, LTRA) indicated for mild persistent asthma [43], allergic rhinitis, and persistent urticaria [29]. Zileuton (available only in US) inhibits the catalytic activity of 5-LO and decreases both cystLTs and LTB₄ levels. Pretreatment of people with asthma with CystLT1 antagonists or 5-LO inhibitors has been shown to decrease airflow obstruction provoked by allergen, aspirin [77], exercise [45•], and sulfur dioxide exposure [78] (Table 1).

In patients with asthma, LTRAs are a first-line controller medication, and in clinical trials, they have proven superior to placebo for multiple clinical outcomes (lung function/spirometry, symptomology, quality of life, beta agonist rescue medication, and frequency of asthma exacerbations) [30, 31•, 79]. When compared to inhaled glucocorticoids as a controller, however, clinical competitiveness studies show inhaled glucocorticoids are superior or equivalent depending on the study [80, 81]. In contrast, adherence with an oral, once-a-day medication (like montelukast) is superior to inhaled steroids [82], and there are potential additive or synergistic benefits to using both inhaled steroids and LT-modifying medications [83, 84]. Of note, mouse studies have shown that extended LTRA use reversed airway remodeling (smooth muscle hypertrophy, subepithelial collagen deposition), an effect not observed with dexamethasone use [85]. This finding underscores the heterogeneity of asthma pathogenesis and the potential benefits of multiple treatment modalities.

In patients with AERD, LT modifiers are effective in blocking or blunting the response to NSAIDs and in improving rhinosinusitis and asthma. Due to ease of treatment, LT receptor antagonists (montelukast) are considered first-line therapy. Zileuton therapy has shown to be highly effective therapy, although it requires liver monitoring and more frequent dosing and has more potential drug interactions [77].

These medications are well tolerated but a variety of adverse effects have been reported. Side effects of montelukast use include suicidal thinking/ideation, and behavioral changes that have resulted in a black-box warning, but these reports were not supported in reviews of the clinical trial data or separate nested, case-control study of insurance claims. [86, 87]

Prostanoids

Prostanoids (PNs; prostaglandins and thromboxanes) are lipid-soluble eicosanoids with hormone-like physiological functions whose structure was first determined by Sune K. Bergström and colleagues in the 1960–1970s, for which they were awarded the Nobel Prize in Medicine in 1982. In addition to regulating blood pressure, coagulation, pain, and fever, PNs regulate the physiology of inflammatory and allergic responses [88, 89].

Synthesis and Signaling

PNs are synthesized by almost every cell type, and due to their chemical instability, they mostly act locally in a paracrine or autocrine way. Similar to LTs, PN synthesis (Fig. 2) also begins with cPLA2 activation, liberating AA from cell membrane phospholipids by hydrolysis of their sn2 ester bonds, thus providing a substrate for cyclooxygenases (COX-1 and COX-2). COX-1 and COX-2 convert AA first into prostaglandin G₂ (PGG₂) and second into prostaglandin H₂ (PGH₂). Specific prostaglandin synthases transform the unstable intermediate PGH₂ into five main derivatives, prostacyclin (PGI₂), prostaglandin D₂ (PGD₂), prostaglandin E₂ (PGE₂), prostaglandin F_2α (PGF_2α), and thromboxane A₂ (TxA₂).

Each of these molecules exerts its effects by binding to one or more PN-specific GPCRs [88]. To date, nine different PN GPCRs have been identified, and their gene expression patterns and the synthesis of their cognate PNs are cell-type specific and can be altered under inflammatory conditions, thus imparting specificity to the cellular responses induced by each prostaglandin [90]. For this reason and due to the instability of PNs in vivo, the PN signaling network has been difficult to analyze in detail with regard to its multiple physiological roles. Below each PN is discussed individually with a focus on their specific properties which are most relevant to atopic disease. Given PGF_2α minimal relevance to atopic diseases, however, this PN will not be discussed further (PGF_2α roles in reproduction, renal physiology, and modulating intraocular pressure are reviewed elsewhere) [91, 92].

PGD₂

PGD₂ (Figs. 1 and 2) is the major PN produced by allergen-specific IgE-coated MCs [93, 94]. It is found in substantial amounts in tissues affected by allergic reactions (lung [95], skin [95, 96], and esophagus [97]) and binds to two structurally distinct GPCRs with a similar affinity [98]. Prostaglandin D2 receptor 1 (DP1), widely expressed in many tissues (brain [99], vasculature [100], eosinophils [101], basophils, DCs [102], T cells [103]), causes vasodilation and smooth muscle relaxation via the increase of cAMP [104]. The role of DP1 in allergic responses remains controversial as both pro- and anti-inflammatory functions (e.g., inhibiting dendritic cell migration [102], promoting T regulatory cells, decreased development of airway hyperreactivity in DP-deficient mice [105], increased mucus production induced by DP activation in vitro [106]) have been reported. In contrast, activation of prostaglandin D2 receptor 2/ chemokine receptor homologous molecule expressed on Th2 lymphocytes (DP2/CRTH2) by PGD₂ induces eosinophil and Th2 T cell chemotaxis to the site of allergic inflammation [107]. Notably, DP2/CRTH2 is the only PN receptor with homology to chemoattractant receptors like the formyl peptide (FMLP) receptor 1 in its function and structure [108].

Both innate lymphoid type 2 cells (ILC2s) and pathogenic effector Th2 cells (peTh2s) are phenotypically defined by the expression of DP2/CRTH2 on their surface. These cell populations release potent Th2 cytokines (IL-5, IL-13, IL-4) that mediate allergic disease pathology. Counts of peTh2 cells (CRTH2⁺, hPGDS⁺) correlate with blood eosinophil counts in patients with atopic dermatitis or eosinophilic gastrointestinal disease (EGID) [109, 110••]. ILC2s are elevated in nasal polyps of patients with chronic rhinosinusitis and regulate eosinophil homeostasis at basal conditions by providing a constant source of IL-5 [111].

Several PGD₂ receptor antagonists are currently under evaluation for their potential beneficial effects in allergic inflammation. Among them are the CRTH2 antagonists OC000459 [33, 34] and Bl671800 [112], AMG 853 (a dual DP1/CRTH2 antagonist) [113], and Ramatroban (a dual thromboxane receptor (TP)/CRTH2 antagonist effective in allergic rhinitis) (Table 1).

PGE₂

PGE₂ (Figs. 1 and 2) modulates cellular activity by binding one of the four prostaglandin E₂ receptors (EP1, EP2, EP3, and EP4). These receptor subtypes activate distinct second messenger molecules—EP2 and EP4 increase intracellular cAMP concentrations; EP1 and EP3 increase intracellular Ca²⁺ concentrations [114]. Depending on the receptor subtype expressed on a given target cell and the receptor’s binding affinity [115], PGE₂ can either enhance the endothelial barrier function (via EP4) [116•], promote tumor angiogenesis (via EP2) [117], inhibit apoptosis of tumor cells [118], or increase survival of eosinophils [119]. PGE₂ exerts both anti-inflammatory and bronchodilator activity in the lung [120]; therefore, EP4 receptor agonists are being considered as a potential treatment strategy in asthma and COPD [121]. In sputum from patients with asthma, PGE₂ levels are increased when compared to those of healthy individuals, and these levels correlate with disease severity [59]. As an endogenous counterpart to pro-inflammatory mediators, PGE₂ might protect from allergic responses and airway inflammation by inhibiting eosinophil [122, 123] and macrophage-functions [124].

PGE₂ has a complex, bimodal effect on human MCs, which is linked to the ratio of EP2:EP3 receptor expression. EP2 activates MCs, whereas EP3 blocks cytokine transcription in human cord blood-derived MCs [125]. Hence, the EP2:EP3 ratio appears to fine-tune the positive or negative effect of PGE₂ on MC degranulation in vivo [126].

PGI₂

Produced mainly by the vascular endothelium [127, 128], PGI₂ (prostacyclin) (Figs. 1 and 2) induces vasodilation [129], smooth muscle relaxation, and inhibition of platelet aggregation by binding the prostacyclin receptor (IP). Activated IP increases intracellular cAMP and activates protein kinase A [130, 131]. Both PGI₂ and PGD₂ are produced in the lung during acute antigen-induced anaphylactic reactions, and in mice, IP deficiency increases allergic airway inflammation [132, 133]. Due to their potent vasodilation capacity, prostacyclin analogues are used as a treatment for pulmonary hypertension [134]. Intriguingly, the prostacyclin analogue Cicaprost was recently shown to inhibit human ILC2 function by decreasing IL-33-induced IL-5 and IL-13 release. Similarly, Cicaprost treatment reduced IL-5⁺ and IL-13⁺ ILC2s in a fungal murine model in which mice were challenged with Alternaria alternata for four consecutive days [135].

TxA2

By binding the TP receptor on endothelial cells and platelets, TxA2 (thromboxane A₂) (Figs. 1 and 2) triggers vasoconstriction, platelet aggregation, and bronchoconstriction [136•] and hence can promote the pathology of allergic asthma. TP receptor activation has been implicated in the interaction between CD4⁺ cells and DCs, and both TP deficiency and the TP antagonist S-145 (administered during sensitization) enhance the inflammatory response in a murine model of contact hypersensitivity [137].

Therapeutics

The relevance of the AA metabolism is evidenced by the fact that the most common analgesic and anti-inflammatory drugs—NSAIDs, aspirin, and specific COX-2 inhibitors (coxibs)—inhibit PN synthesis. Corticosteroids, highly effective in symptom-based treatment of severe allergies, interfere with AA metabolism by the transcriptional downregulation of COX-2 and by enhancing the expression of annexin A1, which suppresses PLA₂. PNs can act as either pro- or anti-inflammatory agents, and their function is highly dependent on the type and condition of the target cell. This complexity can lead to the undesirable off-target effects of certain COX inhibitors but also has fueled the development of a new generation of selective PN receptor agonists/antagonists for treating allergic diseases; several of these agonists/ antagonists are currently being tested in clinical trials (Table 1).

Sphingolipids

Sphingolipids, a ubiquitous and diverse class of cellular lipids defined by their aliphatic amino alcohol backbones (Fig. 1), were first identified by the German-born neurochemist J.L.W. Thudichum in 1884. Given their enigmatic function, they were named after the mysterious Egyptian Sphinx [138, 139]. Over the last two decades, sphingolipids have emerged as critical structural and signaling molecules that regulate a wide array of cellular activities including cell growth, survival, signal transduction, immune cell trafficking, and inflammation [140–142]. The sphingolipid class encompasses a large array of molecules, ranging from the simple sphingoid bases (single-carbon chain) to the ceramides (two-carbon chain) to the complex sphingolipids (such as sphingomyelin (SM), cerebrosides, gangliosides, and sulfatides), which are differentiated by their hundreds of different known head groups [143].

Synthesis and Metabolism

Distinct sphingolipid species are asymmetrically distributed across the intracellular compartments (endoplasmic reticulum (ER), Golgi apparatus, lysosomes, plasma membrane) and can self-assemble into detergent-resistant signaling structures termed lipid rafts [139]. Fluctuations in the distribution and concentrations of specific sphingolipids are associated with changes in cellular morphology (polarization), protein trafficking, and activation state [144, 145]. Rapid interconversion of sphingolipid metabolites mediate signaling cascades with myriad cell-specific effects. Mammalian cells and tissues contain very low concentrations of the two major free sphingoid bases (sphingosine (4-sphingenine), 10–20 pmol/mg; dihydrosphingosine (sphinganine), 1–5 pmol/mg), whereas their phosphorylated forms (sphingosine 1-phosphate (S1P), 0.31 μM in serum; dihydrosphingosine 1-phosphate (dhS1P), 0.04 μM in serum) are more abundant, especially in the blood/ lymph [146–148]. Sphingoid bases are absorbed from various dietary sources, but in mammals, the majority of spingolipids appear to be synthesized endogenously (Fig. 2) [149]. Sphingolipid de novo synthesis occurs in the ER via the condensation of serine and palmitoyl CoA by serine palmitoyltransferase (SPT), a process regulated by the asthma-linked ORMDL protein family [150, 151]. The resulting intermediate (3-ketosphinganine) is rapidly reduced to dihydrosphingosine and subsequently N-acylated by one of the six ceramide synthases (CerS1-CerS6) to dihydroceramide. Following dehydrogenation, the substrate is converted to the prototypical sphingolipid ceramide (N-acyl-sphingosine) [141]. Ceramides are at the center of sphingolipid metabolism. Once synthesized, de novo ceramides are exported throughout the cell, being elaborated into complex glycosphingolipids in the Golgi apparatus or deacylated to sphingosine by ceramidases at the plasma membrane or in lysosomes. Ceramides can also be generated at the plasma membrane via inducible sphingomyelinase activity, leading to direct effects on cell signaling (activation of PP2A phosphatase and PKCζ kinase), apoptosis, and Nlrp3 inflammmasome formation [140]. Ceramides and the sphingoid bases can also be phosphorylated by various kinases (sphingosine kinase-1,2 SphK1 SphK2; ceramide kinase, CerK) to generate the bioactive lipids S1P and ceramide-1-phosphate (C1P) [141]. Extracellular S1P, generated by erythrocytes and vascular endothelial cells and preferentially carried by apoM/HDL, is present at a high concentration in the blood and lymph but kept at low levels in the tissue by the activity of S1P lyase, thus creating a gradient that has a critical role in immune cell trafficking [152•, 153, 154]. Extracellular and membrane S1P acts a ligand for the five known GPCR S1P receptors (S1P_1–5) with each receptor exerting unique effects via their distinct signaling cascades and cellular expression patterns (extensively reviewed in [155, 156]). Receptor-independent intracellular S1P also is bioactive and has an emerging role in IL-1/IRF1-driven autoinflammatory mechanisms. [157] C1P is generated intra-cellularly within the trans-golgi network by CerK and directly binds and activates cPLA2α, thus driving eicosanoid production [158•, 159]. The broad diversity of sphingolipid-driven cellular processes is beyond the scope of this review; therefore, we will focus on the specific role of sphingolipids as they relate to allergic disease.

Effects on Cellular Mediators of Allergic Disease

Mast cells

Studies have highlighted a key role for multiple sphingolipids in regulating MC function [160]. Crosslinking of the IgE high-affinity receptor (FcεR1) on MCs induces activation and de-granulation; however, multiple counter-regulator pathways can blunt this process, including the activity of leukocyte mono-immunoglobulin-like receptor 3 (LMIR3/CD300F/ CLM-1). Both mouse and human LMIR3 are highly expressed on MCs, and upon binding extracellular sphingolipids (mouse LMIR3 binds ceramide; human LMIR300 binds ceramide and SM), these receptors suppress FcεR1-driven degranulation in multiple allergic model systems (passive systemic anaphylaxis, ovalbumin (OVA)-sensitized asthma, and house dust mite eczema) [161, 162]. MC activation is also influenced by the recruitment of the FcεR1 signaling complex to ganglioside-enriched lipid rafts, and exogenous GM, GM3 and GD1a gangliosides are capable of enhancing IgE-mediated histamine and LT release in human MCs [163, 164]. The SRC family kinase LYN, which transduces the activation signal from crosslinked FcεR1 receptors, coprecipitates with GD1b ganglioside in MCs, thus linking sphingolipids from lipid rafts with the IgE signal apparatus. Activated MCs also generate a burst of both intracellular and extracellular S1P following IgE/antigen activation, likely via colocalization of SphK1 and SphK2 to the FcεR1 complex driven by interactions with FYN and LYN kinases [165, 166]. The released S1P acts in an autocrine fashion on MC S1P1 receptors, enhancing chemotaxis and degranulation [167]; however, a role for MC S1P2 receptors remains controversial, with conflicting reports debating a role for S1P2 in degranulation [160, 168]. Also, additional studies have revealed that while mouse MCs require SphK2 for degranulation, calcium mobilization, and cytokine and leukotriene production, human MCs only require SphK1 activity for full functionality [169]. Elevated tissue and serum S1P levels skew both human and mouse MCs towards a hyper-reactive phenotype that may lower the threshold for triggering systemic anaphylaxis [170, 171].

Eosinophils

Human eosinophils express multiple S1P receptors, migrate towards S1P gradients, and upregulate the eotaxin receptor CCR3 following S1P exposure [172]. Peripheral blood eosinophils expressed higher levels of S1P receptor mRNA and protein when isolated from patients with allergic rhinitis than non-allergic controls, and S1P receptors levels increased fol-lowing nasal allergen challenge [173]. Similar to MCs, the sphingolipid-binding inhibitory receptor LMIR3/CD300F/ CLM-1 is also highly expressed on mouse and human eosinophils, is upregulated on the eosinophils of patients with allergic rhinitis, and a murine knock-out of this gene increased tissue eosinophil levels [174]. In addition, orosomucoid-like-3 (ORMDL3), the asthma-associated inhibitor of de novo ceramide synthesis, is expressed by eosinophils and is reported to regulate eosinophil chemotaxis and degranulation via CD48 activation [175].

Association with Allergic Disease and Therapeutics

Asthma

S1P is elevated in the airways of patients with asthma after sub-segmental allergen challenge and modulates airway smooth muscle contraction and cytokine production [176]. In murine asthma models, S1P has a central role in mediating AHR (via smooth muscle S1P1 activation) and enhancing inflammation via effects on MCs, eosinophils, and DCs [140, 177]. Nebulized delivery of FTY720 (figolimod), an S1P1 functional antagonist, and multiple SphK1 inhibitors (SK1-I, N,N-dimethylsphingosine (DMS), AAL-R) have demonstrated that S1P mediates both the sensitization and effector phases of murine allergic asthma [35, 39, 40, 178].

Beyond S1P’s pleiotropic cellular effects, multiple, well-powered genome-wide association studies (GWAS) and cis-expression quantitative trait loci (cis-eQTL) investigations have linked overexpression of the gene ORMDL3 with pediatric-onset asthma [179••, 180–182]. Overexpression of ORMDL3 in a transgenic mouse model led to increases in AHR, airway inflammation, goblet cell metaplasia, and basal IgE levels [183]; however, the cellular mechanisms by which ORMDL3 overexpression enhances asthma pathogenesis remains unclear [184]. Increased airway ceramide levels have been implicated in the pathogenesis of pulmonary diseases such as emphysema and cystic fibrosis, but little is known about the specific roles of ceramide signaling in asthma pathogenesis [185, 186]. We have recently shown that intratracheal delivery of a pharmacologic inhibitor of SPT, (myriocin) enhances AHR and Th2-mediated inflammation in a house dust mite-mediated model of asthma [187]. Intriguingly, ceramide synthase-2 (CerS2) knockout mice, which have reduced pulmonary long-chain ceramide levels, also have exaggerated airway inflammation and increased baseline AHR. [188]

Food Allergy

Allergic responses to food allergens involve lymphocyte sensitization (CD4⁺ T cells, IgE-producing B cells) and enhanced activity of MCs, frequently in the colon. SphK1 and SphK2 knockout mice had reduced food allergen IgE levels, OVA-primed CD4⁺ T cells, and colonic MC counts compared to wild-type control mice in an intragastric OVA food allergy model [189]. Coadministration of FTY720 in an intraperitoneal-primed OVA food allergy model blunted allergic diarrhea by reducing pathogenic CD4⁺ T cell induction and diminishing colonic MC recruitment but had no effect on colonic eosinophil counts [190]. A clinical trial is currently underway to assess serum S1P levels in pediatric patients with food allergy undergoing oral challenge (NCT01776489).

Atopic Dermatitis

In human skin, the epidermal stratum corneum (SC) acts as an air–liquid interface, blocking desiccation of the underlying cellular structure. Ceramide is the most abundant lipid component of the SC [191] and higher ceramide:cholesterol (CH) ratios correlate with improved barrier function [192]. Lesional and non-lesional atopic dermatitis skin is characterized by reduced ceramide-1 (Cer-EOS) and ceramide-3 (Cer-NP) content and a reduced Cer-CH ratio, resulting in impaired barrier function [193, 194]. Emollient treatments containing exogenous ceramides, however, have not been superior to petroleum-based or glycyrrhetinic acid-containing emollients [195]. Topical steroids and calcineurin inhibitors are both reported to increase the Cer-CH ratio in the skin of healthy control patients, but in patients with atopic dermatitis, topical steroids caused ultrastructural disordering of the SC and skin atrophy, effects not noted in calcineurin inhibitor-treated skin [192, 196]. A pre-clinical evaluation of topical FTY720 in the NC/Nga mouse model of dust mite-induced atopic dermatitis revealed reduced epidermal hypertrophy, MC accumulation, and CD3⁺ T cell infiltration in treated animals [36].

Conclusion

Lipid mediators of allergic inflammation gained increasing clinical attention following the approval of drugs targeting the LT pathway in the late 1990s. Multiple clinical trials are currently underway to test the efficacy of novel PN receptor inhibitors for multiple atopic indications [197]. Drugs targeting S1P signaling also are emerging as potent anti-inflammatory/anti-autoimmune compounds [198]. Unfortunately, we are aware of no clinical trials currently testing the efficacy of anti-S1P compounds for allergic indications despite accumulating evidence from animal studies of possible utility. The genetic linkage of ORMDL3 to asthma has also intensified interest in the role of sphingolipid metabolism in allergic airway disease, but unanswered mechanistic questions regarding how ORMDL3 locus polymorphisms contribute to asthma pathogenesis remain [184]. Future clinical and basic science studies, especially those utilizing the unbiased lipidomics approach, are required to more completely elucidate the mechanisms by which lipid mediators contribute to allergic disease. Increased understanding of these critical bioactive molecules will drive future innovations in treating atopy in its many forms.