Targeting the OXE receptor as a potential novel therapy for asthma

Abstract

5-Oxo-6,8,11,14-eicosatetraenoic acid (5-oxo-ETE) is an arachidonic acid metabolite formed by oxidation of the 5-lipoxygenase (5-LO) product 5S-hydroxy-6,8,11,14-eicosatetraenoic acid (5S-HETE) by the NADP⁺-dependent enzyme 5-hydroxyeicosanoid dehydrogenase. It is the only 5-LO product with appreciable chemoattractant activity for human eosinophils. Its actions are mediated by the selective OXE receptor, which is highly expressed on eosinophils, basophils, neutrophils and monocytes. Orthologs of the OXER1 gene, which encodes this receptor, are found in many species except for rodents. Intradermal injection of 5-oxo-ETE into humans and monkeys elicits eosinophil infiltration into the skin, raising the possibility that it may play a pathophysiological role in eosinophilic diseases. To investigate this and possibly identify a novel therapy we sought to prepare synthetic antagonists that could selectively block the OXE receptor. We synthesized a series of indole-based compounds bearing substituents that mimic the regions of 5-oxo-ETE that are required for biological activity, which we modified to reduce metabolism. The most potent of these OXE receptor antagonists is S-Y048, which is a potent inhibitor of 5-oxo-ETE-induced calcium mobilization (IC₅₀, 20 pM) and has a long half-life following oral administration. S-Y048 inhibited allergen-induced eosinophil infiltration into the skin of rhesus monkeys that had been experimentally sensitized to house dust mite and inhibited pulmonary inflammation resulting from challenge with aerosolized allergen. These data provide the first evidence for a pathophysiological role for 5-oxo-ETE in mammals and suggest that potent and selective OXE receptor antagonists such as S-Y048 may be useful therapeutic agents in asthma and other eosinophilic diseases.

Graphical Abstract

1. Introduction

Arachidonic acid (AA) is a polyunsaturated fatty acid (PUFA) that is a key biological intermediate due to its conversion by a variety of enzymes to a large number of important mediators. The two major pathways of AA metabolism are initiated by the actions of cyclooxygenase, resulting in the formation of prostaglandins and thromboxanes, and 5-lipoxygenase (5-LO), which is responsible for the formation of leukotrienes. The 5-LO pathway was discovered in the late 1970s by Borgeat and Samuelsson, who showed that this enzyme oxygenates AA to 5-hydroperoxy-6,8,11,14-eicosatetraenoic acid (5S-HpETE) and then cyclizes 5S-HpETE to the unstable epoxide LTA₄ (Fig. 1) [1]. LTA₄ is then converted to either LTB₄ by LTA₄ hydrolase [2] or to LTC₄ following the addition of GSH by LTC₄ synthase [3]. Successive removal of γ-glutamate and glycine then leads to formation of LTD₄ and LTE₄ [4].

Figure 1.

Biosynthetic pathways and receptors for 5-LO products

LTB₄ is a potent proinflammatory mediator, the actions of which are mediated by the BLT₁ receptor, which is most highly expressed on neutrophils and to a lesser extent on other leukocytes, including monocytes, eosinophils, basophils, and lymphocytes [5]. Despite the expression of BLT₁ on human eosinophils and the stimulatory effects of LTB₄ on various eosinophil responses such as calcium mobilization and surface expression of CD11b [6], it is a very weak chemoattractant for these cells in humans [7–9], in contrast to its potent chemoattractant effects on rodent eosinophils [10]. Although LTB₄ can also activate the BLT₂ receptor, its main ligand is the cyclooxygenase product 12S-hydroxyheptadecatrienoic acid [11].

The first receptor recognized for cysLTs was the cysLT₁ receptor, which is a key receptor in allergy and asthma and mediates the proinflammatory and bronchoconstrictor effects of LTD₄ [12]. This receptor is a target for drugs used in the treatment of asthma, including montelukast, zafirlukast and pranlukast [5]. Subsequently, a second cysLT receptor, cysLT₂, was identified and shown to have equal affinity for LTs C₄ and D₄ [13]. This receptor has been reported to attenuate the activation of the cysLT₁ receptor by forming heterodimers [14]. GPR99 has been identified as a third cysLT receptor (cysLT₃) selective for LTE₄ [15], and has been reported to be involved in increased mucus production by pulmonary epithelial cells [16].

Another arm of the 5-LO pathway leads to the formation of lipoxin A₄ (LXA₄), which is formed in combination with 15-lipoxygenase (15-LO) and acts via the ALX receptor [17]. In contrast to all of the other 5-LO products, which are proinflammatory, LXA₄ promotes the resolution of inflammation and appears to play a protective role in asthma. Administration of a metabolically stable LXA₄ analog was found to block allergen-induced hyperresponsiveness and pulmonary inflammation in OVA-sensitized mice [18]. In a more recent study, LXA₄ was found to increase NK cell-mediated eosinophil apoptosis and to inhibit IL-13 release by ILC2s [19].

2. Formation of 5-oxo-6E,8Z,11Z,14Z-eicosatetraenoic acid (5-oxo-ETE)

2.1. 5-oxo-ETE is formed by a selective dehydrogenase

We obtained evidence for the presence of an additional 5-LO product following a study on the ω-oxidation of LTB₄ and its 6-trans isomers to 20-hydroxy metabolites by human neutrophils [20]. We found that in addition to hydroxylating 6-trans-LTB₄ and 12-epi-6-trans-LTB₄, neutrophils convert these compounds to dihydro metabolites via a 5-oxo intermediate, implicating a dehydrogenase as the initial step in their formation [21]. In contrast, LTB₄, which has a 6-cis double bond and is the preferred substrate for ω-hydroxylation, was not metabolized by this pathway. Since the above 6-trans isomers are nonenzymatic degradation products of LTA₄ and are biologically inactive, the existence of a selective pathway for their metabolism seemed unlikely. We therefore focused on 5S-HETE, which is an abundant 5-LO product containing a 6-trans double bond, as a potential substrate for this enzyme. We found that microsomal fractions from human neutrophils contain an NADPH-dependent dehydrogenase (5-HEDH; 5-hydroxyeicosanoid dehydrogenase) that is extremely selective for eicosanoids containing a hydroxyl group in the 5-position and in the S-configuration, followed by a 6-trans double bond [22]. The preferred substrate for this enzyme proved to be 5S-HETE, which is converted to 5-oxo-ETE.

In parallel with the above studies, Schröder and coworkers found that eosinophils convert AA to an eosinophil chemoattractant that they named “eosinophil chemotactic lipid” [8]. They subsequently showed that soybean lipoxygenase converts AA to a substance with identical properties, which they identified as 5-oxo-15-hydroxy-6E,8Z,11Z,13E-eicosatetraenoic acid [23]. At the same time, we found that the latter substance is also formed by the oxidation of the combined 5-LO/15-LO product 5S,15S-dihydroxy-6E,8Z,11Z,13E-eicosatetraenoic acid by neutrophil 5-HEDH [22]. These results suggested that 5-HEDH may be responsible for the formation of potent proinflammatory lipid mediators that may play important roles in inflammatory diseases such as asthma.

2.2. Regulation of 5-oxo-ETE synthesis

Although neutrophil microsomes display high 5-HEDH activity, intact neutrophils release relatively small amounts of 5-oxo-ETE when stimulated with calcium ionophore [24]. This is due to the high selectivity of 5-HEDH for NADP⁺, which is present within cells primarily as its reduced form, NADPH [25]. Thus, in addition to activation of cytosolic phospholipase A₂ (cPLA₂) and 5-LO by calcium and phosphorylation, the synthesis of 5-oxo-ETE requires the elevation of intracellular NADP⁺, which occurs in neutrophils following activation of the respiratory burst [26] as well as during the process of cell death [27]. Resting neutrophils convert 5S-HETE principally to 20-hydroxy-5S-HETE by the action of the cytochrome P450 enzyme LTB₄ 20-hydroxylase. In contrast, 5-oxo-ETE is the major metabolite of 5S-HETE in neutrophils in which NADPH oxidase has been activated by phorbol myristate acetate [26]. Activation of the respiratory burst also increases 5-oxo-ETE synthesis in monocytes [28] and eosinophils [9] (Table 1). High levels of 5-HEDH activity are also present in B lymphocytes [29], but in this case the synthesis of 5-oxo-ETE is driven by oxidative stress, which augments intracellular NADP⁺ levels through the GSH redox cycle. Oxidative stress also strongly promotes 5-oxo-ETE synthesis by monocytes [30].

Table 1.

Synthesis of 5-oxo-ETE by different types of cells

	AA ↓ 5oETE	5-HETE ↓ 5oETE	Regulation of [NADP+]
Cell type			NOX	Ox stress	Cell death	Ref
Neutrophils	✓	✓	✓		✓	[27, 30]
Monocytes	✓	✓	✓	✓		[30]
Eosinophils	✓	✓	✓			[9]
B lymphocytes	✓	✓		✓		[29]
Platelets		✓		✓		[30]
Airway epithelial cells		✓		✓		[31]
Airway smooth muscle cells		✓		✓		[31]
Endothelial cells		✓		✓		[32]
Keratinocytes*		✓		✓		[87]
Dendritic cells**		✓	✓			[124]
Tumor cells***		✓		✓	✓	[33]

Many types of inflammatory cells can convert AA to 5-oxo-ETE because they possess both 5-LO and 5-HEDH, whereas other types of cells have been shown to convert 5S-HETE to 5-oxo-ETE because they contain 5-HEDH but not necessarily 5-LO. Depending on the cell type, NADP⁺ levels can be regulated by activation of NADPH oxidase (NOX), oxidative stress (OX stress), or cell death

^*The substrate was 5S-HODE rather than 5S-HETE

^**Derived from monocytes

^***PC3 (prostate), MCF7 (breast), and A-427 (lung) tumor cells

The inflammatory cells discussed above possess both 5-LO and 5-HEDH and can therefore synthesize 5-oxo-ETE from endogenous arachidonic acid. However, a number of other cell types that contain little or no 5-LO activity, including platelets [30], airway epithelial and smooth muscle cells [31], and vascular endothelial cells [32] also possess high 5-HEDH activity and can convert 5S-HETE (but not arachidonic acid) to 5-oxo-ETE (Table 1). These cells could therefore utilize inflammatory cell-derived 5S-HETE to synthesize 5-oxo-ETE by transcellular biosynthesis, as we have shown to occur when PC3 prostate cancer cells are cocultured with neutrophils under conditions of oxidative stress [33]. Leukotrienes can also be formed by transcellular biosynthesis [34], including the synthesis LTC₄ by both vascular endothelial cells and platelets from neutrophil-derived LTA₄ and the synthesis of LTB₄ by erythrocytes, also from neutrophil-derived LTA₄.

The optimal sites for the production of 5-oxo-ETE are likely to be inflammatory foci, where activated phagocytes having high levels of intracellular NADP⁺ due to activation of the respiratory burst could convert endogenous AA to 5-oxo-ETE. These cells could then release both 5-oxo-ETE and its precursor 5S-HETE, which could subsequently be converted to additional 5-oxo-ETE by nearby structural cells, including airway epithelial and smooth muscle cells as well as vascular endothelial cells. Exposure of these cells to oxidants released by inflammatory cells could increase their NADP⁺ levels, thus favoring the synthesis of 5-oxo-ETE. Furthermore, as infiltrating inflammatory cells begin to undergo apoptosis, their ability to synthesize 5-oxo-ETE increases dramatically, as we have shown with neutrophils [27]. This raises the possibility that 5-oxo-ETE could play an important role in chronic inflammation, in contrast to some other lipid mediators like prostaglandin D₂ (PGD₂), which is rapidly released from mast cells following antigen stimulation [35] and may therefore be more important in acute inflammatory responses.

2.3. Formation of 5-oxo-ETE as a result of lipid peroxidation

Although the primary route for the synthesis of 5-oxo-ETE appears to be enzymatic, like most other eicosanoids, it can also be formed as a product of lipid peroxidation. Unlike isoprostanes and H(p)ETEs, 5-oxo-ETE does not have any chiral centers, so it will be biologically active irrespective of the mechanism of its formation, as long as the double bonds are in the correct configuration. Substantial amounts of 5-oxo-ETE were formed after exposure of erythrocytes to t-butyl hydroperoxide [36] as well as in model systems of chemically-induced lipid peroxidation of phospholipids [37]. 5-Oxo-ETE can also be formed by murine macrophages as a result of nonenzymatic heme-catalyzed dehydration of 5S-HpETE [38].

3. In vitro effects of 5-oxo-ETE

3.1. Effects of 5-oxo-ETE on neutrophils and eosinophils

We first investigated the effects of 5-oxo-ETE on neutrophils, in part because these were the cells in which we first identified the pathway for its formation, but also because its precursor 5S-HETE had previously been shown by O’Flaherty’s group to weakly activate these cells via what appeared to be a novel receptor [39]. We postulated that 5-oxo-ETE is the true ligand for this receptor and found it to be a potent activator of calcium mobilization and chemotaxis in neutrophils, with an EC₅₀ that was about 100 times lower than that of 5S-HETE and 10 times more potent than 5-oxo-15S-HETE [40]. It also stimulates actin polymerization, surface expression of CD11b and adherence of these cells [41]. Although 5-oxo-ETE by itself has little [42] or no [43] effect on the production of superoxide by neutrophils, in the presence of GM-CSF [43], TNF-α or platelet-activating factor (PAF) [44] it is a potent stimulator of this response. Similarly, 5-oxo-ETE induces a modest degree of degranulation on its own, but this is dramatically enhanced in the presence of TNF-α or PAF.

The findings of Schroder’s group referred to above [23], demonstrating the powerful chemoattractant effects of 5-oxo-15S-HETE on eosinophils, suggested that the eosinophil may be a major target for 5-oxo-ETE. This proved to be the case, as 5-oxo-ETE was more potent than both 5-oxo-15S-HETE and PAF as an eosinophil chemoattractant [9] and has a variety of other effects on these cells [6, 45, 46]. In contrast, other 5-LO products, including both LTD₄ and LTB₄, which, despite its well documented effects on guinea pig eosinophils [7], had barely detectible chemoattractant effects on human eosinophils [9]. Interestingly, 5-oxo-ETE and LTB₄ had similar potencies in stimulating surface expression of CD11b and intracellular calcium mobilization in these cells, but 5-oxo-ETE was a much better stimulator of actin polymerization and shedding of L-selectin [6]. It seems likely that the potent chemoattractant effects of 5-oxo-ETE on eosinophils are closely related to its powerful effects on actin polymerization. 5-Oxo-ETE also activates the respiratory burst in eosinophils [46] and mediates the stimulatory effects of PAF and IL-5 on the surface expression of the leukocyte activation marker CD69, whereas LTB₄ has little effect on this response [47].

3.2. Synergistic interactions between 5-oxo-ETE and other eosinophil chemoattractants

A variety of mediators would be present at inflammatory sites and synergistic interactions among them are likely to be important in regulating the course of inflammation. For example, low concentrations of 5-oxo-ETE shift the concentration-response curve for PAF-induced eosinophil migration to the left [9]. Similarly, threshold concentrations of the CC chemokines eotaxin-1 (CCL11) and RANTES (CCL5) increase the chemotactic potency of 5-oxo-ETE [48]. PGD₂ has also been shown to enhance 5-oxo-ETE-induced eosinophil migration [49]. Although 5-oxo-ETE alone has a relatively weak effect on eosinophil degranulation, in the presence of PAF it induced a very strong response [50]. Moreover, 5-oxo-ETE strongly enhances eosinophil degranulation induced by LTB₄, C5a and fMLP.

3.3. Interactions between 5-oxo-ETE and GM-CSF

5-Oxo-ETE and GM-CSF can potentially interact at multiple levels. GM-CSF stimulates the formation of 5-LO products by increasing the release of arachidonic from inflammatory cells [51] as well as by increasing the expression of both 5-LO [52] and FLAP [53], and is therefore likely to increase the formation of 5-oxo-ETE. Conversely, we have shown that 5-oxo-ETE increases the production of GM-CSF by human monocytes and by this mechanism can increase the survival of eosinophils [54]. GM-CSF has also been shown to increase the responsiveness of eosinophils to 5-oxo-ETE as exemplified by O’Flaherty’s finding that priming with this cytokine results in a strongly enhanced degranulation response [50].

3.4. 5-Oxo-ETE promotes the transmigration of eosinophils across the basement membrane and endothelial cell monolayers

Although stimulation of eosinophil migration is critical for the tissue infiltration of these cells, they must also cross the vascular endothelium and the underlying basement membrane, a process that is dependent on secreted proteases. Laviolette’s group found that 5-oxo-ETE is a potent stimulator of the passage of eosinophils through Matrigel basement membranes [55], being more effective than both PAF and eotaxin [56]. The response to 5-oxo-ETE was enhanced by IL-5, which by itself had little effect on eosinophil transmigration [55]. The effect of 5-oxo-ETE was not due solely to its chemoattractant properties, but was also mediated by the release of MMP-9 from eosinophils as well as by the urokinase plasminogen activator receptor (uPAR) and was inhibited by antibodies against either MMP-9 or uPAR. As observed for Matrigel-coated membranes, 5-oxo-ETE also strongly promoted the passage of eosinophils through monolayers of human endothelial cells [57]. These results indicate that 5-oxo-ETE can promote the transendothelial migration of eosinophils by a combination of its chemoattractant effects and its stimulatory effects on the release or activation of tissue proteases, resulting in degradation of the extracellular matrix.

4. The 5-oxo-ETE receptor

4.1. The OXE receptor

Activation of neutrophils by 5-oxo-ETE shows a high degree of structural specificity, suggesting that it acts by a highly selective receptor [40, 44]. As noted above, 5-oxo-ETE is about 100 times more potent than its precursor 5S-HETE in activating neutrophils [40]. Moreover, pretreatment of neutrophils with 5-oxo-ETE selectively desensitizes these cells so that they no longer respond to subsequent addition of 5-oxo-ETE, but respond normally to other agonists such as LTB₄. The stimulatory effect of 5-oxo-ETE on neutrophils was blocked by pertussis toxin, indicating that its actions are mediated by a G protein-coupled receptor (GPCR) [42, 43, 58]. O’Flaherty’s group subsequently identified binding sites on neutrophil plasma membranes that were highly selective for 5-oxo-ETE and sensitive to GTP, which reduced ligand binding, as expected for a GPCR [59]. The binding interaction was complicated by esterification of 5-oxo-ETE into cellular lipids, but this could be prevented by pretreatment with the acyl-CoA synthetase inhibitor triacsin C. The affinity of the neutrophil membrane binding sites for 5-oxo-ETE (K_d, 4 nM) was close to the EC₅₀ value (2 nM) found for its effect on calcium mobilization in these cells [40].

The receptor for 5-oxo-ETE was cloned by three separate groups. With the aid of a BLAST search Hosoi and coworkers identified an intronless nucleotide sequence with homology to GPCRs in chromosome 2p21 [60]. This sequence, which they tentatively named TG1019, was predicted to code for a protein containing 423 amino acids. They then used constructs of TG1019 fused to Gα_i, Gα_s, or Gα_q to screen a library of potential ligands for orphan receptors based on their ability to promote the binding of GTPγS. In the same year, Takeda and coworkers independently identified the same sequence, codenamed hGPCR48, as the receptor for 5-oxo-ETE [61]. In both cases, 5-oxo-ETE was by far the most potent ligand among those tested, with an EC₅₀ of about 6 nM for stimulation of GTPγS binding to the fusion protein. The following year Jones et al identified a sequence, termed R527, that had high homology to the hydroxycarboxylic acid receptor HCAR3 (then known as HM74) [62]. R527 was nearly identical to TG1019/hGPCR44, except that it lacked the initial 39 amino acids at the amino terminus, and had 5-oxo-ETE as its preferred ligand. The receptor for 5-oxo-ETE (TG1019/hGPCR48/R527) has been designated as the OXE receptor by the IUPHAR Nomenclature Committee for Leukotriene and Lipoxin Receptors with the corresponding gene being OXER1 [63].

In humans OXER1 mRNA is highly expressed in peripheral blood leukocytes [60], particularly in eosinophils [62] and basophils [64, 65] and to a lesser extent in neutrophils, monocytes and macrophages. Among human GPCRs the OXE receptor is most closely related to the hydroxycarboxylic acid receptors HCA₁, HCA₂, and HCA₃, with which it bears about 40% sequence identity. These receptors are encoded by the HCAR1, HCAR2 and HCAR3 genes and have as their principal ligands lactic acid, 3-hydroxybutyric acid, and 3-hydroxyoctanoic acid, respectively [66]. The HCA₂ receptor is also activated by nicotinic acid. The OXE receptor is less related to other eicosanoid receptors, bearing 34% sequence identity at the protein level with the receptor for 12S-HETE (GPR31/HETER1) and 23% identity with the CysLT₁ receptor for LTD₄.

Many species possess orthologs of human OXER1, with the notable exception of rodents. 5-Oxo-ETE is a potent activator of eosinophils and neutrophils from both cats [67] and monkeys [68], which have OXE receptors that are about 79 and 96% identical, respectively, in sequence to the human receptor. 5-Oxo-ETE also activates zebrafish leukocytes, acting by a receptor encoded by oxer1/hcar1–4 [69]. The human protein most closely related to the protein encoded in zebrafish by the mRNA for this gene is the OXE receptor (34% sequence identity), followed by the human HCA₁, HCA₂ and HCA₃ receptors, which have about 31% sequence identity.

4.2. Other potential receptors for 5-oxo-ETE

Although rodents lack orthologs of OXER1, both in vitro and in vivo studies with rats, mice and guinea pigs suggest that it may act in these species, presumably via a different receptor. For example, in two separate studies we found that 5-oxo-ETE induces infiltration of eosinophils into the lungs of Brown Norway rats [70, 71]. 5-Oxo-ETE has also been reported to activate certain mouse cell lines, stimulating cAMP-dependent steroidogenesis in murine MA-10 Leydig cells [72] and Gα_i-dependent migration of MG5 murine microglial cells [73], although in the latter case micromolar concentrations were required. 5-Oxo-ETE was recently implicated in causing pain in some cases of irritable bowel syndrome in both humans and a mouse animal model, although the concentrations required for these effects were in the micromolar range [74]. At least in the mouse, these effects appear to be mediated by Mrgprd, a receptor found on dorsal root ganglion afferents that is also activated by β-alanine and has about 23% identity with the human OXE receptor.

5-Oxo-ETE also has some activity in the guinea pig. It contracts guinea pig bronchial smooth muscle, by a mechanism that is dependent on TXA₂, since its effects are blocked by the cyclooxygenase inhibitor indomethacin as well as by the selective TP receptor antagonist SQ-29548 [75]. Interestingly, it appears to have the opposite effect on human bronchi, inducing relaxation of bronchial smooth muscle precontracted with methacholine [76]. This effect appeared to be due to activation of BK_Ca channels, since it was blocked by iberiotoxin, but was probably not related to the OXE receptor, since it was not blocked by pertussis toxin. Moreover, OXER1 mRNA was undetectable in human airway smooth muscle cells [62]. 5-Oxo-ETE also evokes isotonic volume reduction of guinea pig intestinal crypt epithelial cells, which is dependent on chloride secretion and involves activation of protein kinase C [77]. In contrast, jejunal villus cell volume was unaffected.

With the exception of Mrgprd in the mouse, no information is available on the identities of the receptors mediating the actions of 5-oxo-ETE in rodents. All three of the above species have genes for hydroxycarboxylic acid receptors (Hcar1 and Hcar2), with the corresponding proteins having approximately 40% sequence identity with the human OXE receptor. Therefore, it is possible that one or both of these receptors might mediate the effects of 5-oxo-ETE in rodents. So far, there is no evidence for the existence of 5-oxo-ETE receptors other than the OXE receptor in humans or other primates. As discussed below (Section 7.2), selective OXE receptor antagonists completely block the activation of neutrophils and eosinophils by 5-oxo-ETE, suggesting that at least in these cases, no other 5-oxo-ETE receptors are involved.

5. Intracellular signaling mechanisms for the OXE receptor

The actions of 5-oxo-ETE, in particular its chemoattractant effects, appear to be mediated primarily through G_i/o proteins, as most of its actions can be blocked by pertussis toxin [42, 43, 58]. Unlike the receptors for many other chemoattractant receptors, the OXE receptor does not appear to be appreciably coupled to G_q [60, 78]. The most rapid responses of granulocytes to 5-oxo-ETE are calcium mobilization and actin polymerization [6, 45, 46], which occur within the first few seconds of exposure and are blocked by pertussis toxin. Calcium mobilization is also blocked by the phospholipase C (PLC) inhibitor U73122 [79], indicating that the initial step in this process is activation of PLC, leading to the formation of inositol trisphosphate (IP₃) and diacylglycerol (DAG), resulting respectively in intracellular calcium release and activation of both conventional and novel protein kinase C (PKC) isoforms (Fig. 2). The novel PKC isoform PKCδ has been implicated in the effects of 5-oxo-ETE on eosinophil motility and the release of the metalloproteinase MMP-9, which is involved in promoting the transmigration of granulocytes from the circulation into tissues [80]. There is also evidence for the involvement of the atypical PKC isoform PKCζ in eosinophil migration [80]. In addition, the novel PKC isoform, PKCε, has been reported to mediate the proliferative effect of 5-oxo-ETE on prostate cancer cells and this effect can be blocked by U73122, confirming the role of PLC in initiating this response [81].

Figure 2.

Intracellular signaling pathways coupled to the OXE receptor and cellular responses induced in granulocytes by 5-oxo-ETE.

5-Oxo-ETE has also been shown to activate phosphatidylinositol 3-kinase (PI3K), resulting in the phosphorylation of Akt [42, 79]. Inhibitors of both PI3K (LY294002) and PLC (U73122), as well as pertussis toxin, strongly inhibit 5-oxo-ETE-induced cell migration, indicating that both the PLC and the PI3K pathway are involved in this response [79]. 5-Oxo-ETE also activates the MAP kinase pathway, inducing phosphorylation of ERK-1 and ERK-2. However, in this case, there is disagreement in the literature on the sensitivity of this response to pertussis toxin [43, 79, 82], raising the possibility of additional signaling mechanisms, possibly involving the recruitment of β-arrestin [82]. 5-Oxo-ETE also activates cPLA₂, inducing its phosphorylation, and this is thought to be mediated by MAP kinases [43]. Studies with selective inhibitors suggest that both ERK-1/2 and p38 are involved in the stimulatory effect of 5-oxo-ETE on CD11b expression in eosinophils, whereas ERK1/2, (along with PKCδ, as mentioned above) mediate its effect on MMP-9 secretion [80].

Consistent with the role of a G_i protein in mediating the response to activation of the OXE receptor, 5-oxo-ETE inhibits cAMP formation by adenylyl cyclase in both transfected cells [60] and neutrophils [83]. However, the functional significance of this response is unclear, as Kostenis’ group identified a biased OXE receptor antagonist, Gue1654, that blocked nearly all of the responses to 5-oxo-ETE that were tested, including cell migration, actin polymerization, calcium mobilization, and CD11b expression, but did not inhibit forskolin-stimulated cAMP formation [83]. This is presumably because Gue1654 forms a stable complex with the OXE receptor in which the α_i-mediated inhibition of adenylyl cyclase is unaffected, whereas responses mediated by the βγ subunits and β-arrestin are inhibited. Thus, the OXE receptor-mediated functional responses of granulocytes to 5-oxo-ETE are mediated principally by βγ and possibly β-arrestin rather than by the α_i subunit of G_i.

6. Structure-activity relationships for activation of the OXE receptor

We conducted extensive structure-activity studies to define the regions of 5-oxo-ETE that are essential for activation of the OXE receptor. In addition to the 5-oxo group, a free carboxyl group is required for potent biological activity [58]. The 5-oxo group must be conjugated with at least two double bonds (5-oxo-Δ^6E,8Z) and the total chain length must be at least 18 carbons [84]. The ω-end of the molecule must be hydrophobic, as addition of a hydroxyl group at C₂₀ by LTB₄ 20-hydroxylase results in dramatically reduced agonist activity [58]. Although still reducing activity by about 10-fold [85], a hydroxyl group at C₁₅ as in 5-oxo-15S-HETE, is better tolerated. This product is formed by eosinophils by the combined actions of 5-LO, 15-LO and 5-HEDH [9, 45]. In contrast, hydroxylation of 5-oxo-ETE at C₁₂ by platelet 12-LO, resulting in the formation of 5-oxo-12S-HETE, abolishes agonist activity, and instead results in the formation of an OXE receptor antagonist with an IC₅₀ of about 500 nM [86].

Certain 5-oxo metabolites of PUFA other than AA are also potent OXE receptor agonists. Sebaleic acid (5Z,8Z-octadecadienoic acid), which is the major PUFA in human sebum and skin surface lipids, is converted to 5-oxo-6E,8Z-octadecadienoic acid (5-oxo-ODE) by human neutrophils [87], whereas its 8-trans isomer, 5-oxo-8-trans-ODE, is a constituent of the tubers of the South American plant Lepidium meyenii [88]. These two C₁₈ compounds have potencies almost identical to 5-oxo-ETE in activating both human neutrophils and eosinophils [84, 87]. 5-Oxo-ODE could thus potentially contribute to granulocyte infiltration in inflammatory diseases of the skin, especially considering that sebaleic acid cannot be converted to leukotrienes. Similarly, Mead acid (5Z,8Z,11Z-eicosatrienoic acid), which can accumulate under conditions of essential fatty acid deficiency [89], is converted by neutrophils to 5-oxo-6E,8Z,11Z-eicosatrienoic acid (5-oxo-ETrE), which activates both neutrophils and eosinophils with potencies similar to 5-oxo-ETE [84]. 5-Oxo-ETrE would be the most important granulocyte chemoattractant among 5-LO products derived from Mead acid, since this PUFA is not converted to significant amounts of LTB₃ due to the inhibition of LTA₄ hydrolase by LTA₃ [90]. Neutrophils convert the ω3 PUFA 5Z,8Z,11Z,14Z,17Z-eicosapentaenoic acid (EPA) to 5-oxo-6E,8Z,11Z,14Z,17Z-eicosapentaenoic acid (5-oxo-EPE), which is approximately 5 times less potent than 5-oxo-ETE in activating neutrophils and eosinophils [84, 91].

7. OXE receptor antagonists

The absence of OXER1 in rodents has impeded investigation into the pathophysiological role of 5-oxo-ETE because of the unavailability of gene knockout models. For this reason, the availability of selective OXE receptor antagonists is critical to establish the potential roles of this mediator in inflammatory diseases. Since the eosinophil is a major target of 5-oxo-ETE, such antagonists may be useful therapeutically as anti-inflammatory agents in allergic diseases such as asthma.

The effects of 5-oxo-ETE could also be prevented by using a 5-LO inhibitor or a FLAP antagonist, which would have the additional advantage of blocking the formation of cysLTs (Fig. 3). However, these agents would also block the formation of LXA₄, which has been shown to have a protective effect in asthma and other inflammatory conditions [18]. They would also inhibit LTB₄ formation, which might have somewhat unpredictable effects. There is some evidence that LTB₄ plays a role in asthma, principally through effects on dendritic cells and lymphocyte subsets, rather than its classic target cell, the neutrophil [92]. Although animal experiments lend some support for such a role, clinical studies in asthmatics using a BLT₁ antagonist and an LTA₄ hydrolase inhibitor have been unsuccessful [93]. On the other hand, there is evidence for a beneficial role for LTB₄ in host defense against both bacterial [94] and viral [95, 96] infections. Thus, an OXE receptor antagonist, either on its own or combined with a cysLT₁ antagonist, might offer a more selective way to block the effects of proinflammatory 5-LO products involved in asthma without affecting the potentially beneficial effects of LXA₄ and LTB₄.

Figure 3. Comparison of OXE antagonists and 5-LO/FLAP inhibitors.

An OXE antagonist will block the actions of 5-oxo-ETE irrespective of its source but will not affect the beneficial actions of LXA₄ and LTB₄.

7.1. Naturally occurring OXE receptor antagonists

We found that stimulation of mixtures of neutrophils and platelets with calcium ionophore in the presence of phorbol myristate acetate led to the formation of 5-oxo-12S-HETE (Fig. 4) by transcellular metabolism due to the combined actions of 5-LO and 5-HEDH from neutrophils and 12-LO from platelets [86]. As noted above, this metabolite of 5-oxo-ETE was devoid of OXE agonist activity, but instead blocked neutrophil activation by 5-oxo-ETE. Although this was the first report of an OXE receptor antagonist, this compound was unsuitable for in vivo studies because of its instability and susceptibility to metabolism. Various PUFA, including 4Z,7Z,10Z,13Z,16Z,19Z-docosahexaenoic acid (DHA), EPA, dihomo-γ-linolenic acid, and Mead acid were reported to antagonize the activation of the OXE receptor in transfected cells [60]. However, these fatty acids were not very potent (IC₅₀ ~2 to 6 μM) and it is unlikely that they would be selective for the OXE receptor. Moreover, they can be converted to a variety of biologically active metabolites, including 5-oxo-ETE analogs as noted above and the DHA metabolite protectin D1 (PD1) [97]. PD1 has been reported to reduce eosinophil activation and migration in response to 5-oxo-ETE, although this effect was not selective for the OXE receptor, as the responses to other eosinophil agonists were also reduced [98].

Figure 4. Potencies of some OXE receptor antagonists.

Concentration-response curves for the inhibition of calcium mobilization in human neutrophils stimulated with 5-oxo-ETE (10 nM), added 2 min after the addition of the antagonist: initial lead compound (Δ; ref. [103]), 5-oxo-12S-HETE (♦; ref. [86]), Gue1654 (∇; ref. [103]), S-230 (▲), S-C025 (○; ref. [107]), and S-Y048 (●; ref. [108]). The effects of Gue1654 on 5-oxo-ETE-induced calcium mobilization were determined as described previously [103].

A recent study suggests that testosterone may be an endogenous OXE receptor antagonist. Although steroids classically act via nuclear receptors, there is considerable interest in extranuclear receptors located in the plasma membrane and elsewhere in the cell [99]. In an attempt to identify membrane receptors for testosterone, Kalyvianaki et al. incubated DU145 prostate cancer cells with a membrane impermeable form of testosterone (testosterone-BSA-FITC) and searched for testosterone binding proteins having the characteristics of membrane receptors [100]. They found only one such protein, which they identified as the OXE receptor, and showed that testosterone could inhibit 5-oxo-ETE-induced actin redistribution, cell migration, and inhibition of cAMP formation in DU145 cells. Testosterone could impact the actions of 5-oxo-ETE in other ways as well, as it has been shown to inhibit 5-LO activity, thereby explaining the reduced ability of human neutrophils from males to synthesize 5-LO products compared to neutrophils from females [101]. These effects of testosterone may contribute to the reduced incidence of asthma in men compared to women [102]. If 5-oxo-ETE is an important proinflammatory mediator in asthma as we suspect, its role may be diminished in men compared to women, and drugs targeting the OXE receptor may therefore be more effective in women.

7.2. Synthetic OXE receptor antagonists

To identify OXE receptor antagonists Kostenis’ group used a combination of pharmacological and in silico approaches. They screened a library of small molecules using HEK293 cells transfected with OXER1 and the promiscuous Gα₁₆ protein and then employed computer modeling and further testing to identify Gue1654 (Fig. 4) as a biased antagonist that blocked all of the actions of 5-oxo-ETE on neutrophils and eosinophils except for inhibition of adenylyl cyclase (see Section 5).

We used a different approach, based on structure-activity relationships, to design synthetic OXE receptor antagonists. From the structure-activity studies described in Section 6 we knew that the regions of 5-oxo-ETE that are critical for engagement of the OXE receptor included the initial 5-oxovalerate grouping (C₁-C₅), at least 2 double bonds conjugated with the keto group, and the terminal hydrophobic region of the molecule. Our strategy was to incorporate these groups into a rigid molecule using indole as the backbone to give conformationally-restricted molecules that would bind strongly to the receptor and fix it in an inactive conformation. Examination of a series of indoles containing 5-oxovalerate and alkyl groups in different positions led to the identification of two lead compounds containing a hexyl group in the 2-position and a 5-oxovalerate group in either the 1-position (IC₅₀, 1.1 μM; not shown) or the 3-position (IC₅₀, 9.8 μM) (Fig. 4) [103].

We modified the above two lead compounds with a view to improve their metabolic stability and increase their potency. Because of concern that β-oxidation of the acyl side chain might be a problem, we examined various modifications of this group that would block this metabolic pathway. Addition of a methyl group at C₃ of the 5-oxovalerate group not only served this purpose, but also dramatically increased potency, but only when it was in the S-configuration [104, 105]. Further increases in potency were achieved by the addition of a chloro group at C₅ of the indole and methylation of the indole nitrogen to give compound S-230, which is a potent inhibitor of 5-oxo-ETE-induced calcium mobilization (IC₅₀, 6 nM) (Fig. 4). Similar modification of the second lead compound referred to above led to compound S-264, which had about the same antagonist potency (IC₅₀, 7 nM). Relatively high concentrations of S-230 appeared in the blood shortly after oral administration to monkeys, but its concentration declined rapidly thereafter [68], probably due to the susceptibility of its hexyl side chain to ω2-oxidation, resulting in the formation of ω2-hydroxy-S-230 (Fig. 4), which is its major microsomal and plasma metabolite [68]. S-264 was similarly metabolized by ω2-oxidation, but also underwent partial loss of the N-acyl group, resulting in a pharmacokinetic profile that was inferior to that of S-230 [106]. For this reason, we focused on the S-230 series in further studies.

To reduce oxidation of the alkyl side chain of S-230, the hexyl side chain was replaced with a phenyl group attached to the 2-position of the indole via a polymethylene spacer, the optimal length of which was found to be 6 carbons as in compound S-C025 [107]. In addition to reducing the rate of metabolism, the presence of the phenyl group dramatically increased potency (IC₅₀, 100 pM) (Fig. 4) and increased half-life in plasma following oral administration to monkeys. The plasma levels of S-C025 4 h after administration of a dose of 5 mg/kg were substantially higher than those of S-230, 4 h after administration of a 6-fold higher dose of the latter. In contrast to S-230, the major plasma metabolite of S-C025 (i.e. αS-hydroxy-S-C025) was formed by stereoselective hydroxylation of the methylene of the alkylphenyl group α to the indole (Fig. 4). Interestingly, this biologically-formed metabolite is itself a potent OXE receptor antagonist (IC₅₀, 400 pM), whereas the corresponding αR-hydroxy derivative of S-C025 is over 500 times less potent [107].

Further improvements in the OXE antagonist were achieved through modification of the phenyl group. The most potent of these compounds is the meta-chloro derivative of S-C025 (i.e. S-Y048, Fig. 4), which has an IC₅₀ of 20 pM in inhibiting calcium mobilization in neutrophils elicited by a 500-fold higher concentration of 5-oxo-ETE [108]. Although the maximal plasma level of S-Y048 following a dose of 5 mg/kg is a little less than that for S-C025, it has a longer half-life, and its concentrations are substantially higher than those of S-C025 at time points of 4 h and later. Like S-C025, the principal plasma metabolite of S-Y048 is the corresponding αS-hydroxy derivative, which is equipotent with its precursor and has a prolonged lifetime in the blood, with its levels remaining fairly constant between 4 and 24 h. This suggests that αS-hydroxy-S-Y048 could contribute significantly to the in vivo antagonist effects of S-Y048 at longer time points.

8. Pathophysiological role of 5-oxo-ETE

8.1. In vivo effects of 5-oxo-ETE

The powerful in vitro effects of 5-oxo-ETE on eosinophil migration and extracellular matrix degradation as discussed in Sections and 3.1 and 3.4 suggest that this mediator could elicit the tissue infiltration of eosinophils in vivo. To test this hypothesis, 5-oxo-ETE was administered intradermally in the forearms of both healthy and asthmatic human subjects and the numbers of inflammatory cells in the skin were counted 24 h later [109]. 5-Oxo-ETE significantly increased the numbers of macrophages (CD68⁺), neutrophils (elastase⁺), and eosinophils (MBP⁺) in the skin in both healthy and asthmatic subjects as well as the numbers of mast cells in the skin of asthmatic subjects, but did not significantly affect the numbers of lymphocytes (Fig. 5). Furthermore, the numbers of both eosinophils and mast cells following administration of 5-oxo-ETE were significantly higher in asthmatic subjects compared to healthy controls. Among all of the cell types evaluated, the most pronounced effect was on eosinophils in asthmatic subjects, which were present at numbers about 50-fold higher than in vehicle-treated control subjects. In subsequent studies we found that intradermal administration of 5-oxo-ETE also induces infiltration of eosinophils into both cynomolgus and rhesus monkey skin [110].

Figure 5. Recruitment of inflammatory cells into human skin in response to 5-oxo-ETE.

5-Oxo-ETE was injected intradermally into the forearms of subjects with mild atopic asthma (n=11) and healthy control subjects (n=10). After 24 h punch biopsy samples were obtained, and the numbers of inflammatory cells counted following immunostaining for markers specific for different cell types: T-lymphocytes (Lymph; anti-human CD3), macrophages (Macr, anti-human CD68), neutrophils (Neutr, anti-human neutrophil elastase), eosinophils (anti-human major basic protein), and mast cells (anti-human tryptase). V, vehicle; 5o, 5-oxo-ETE; Con, healthy subjects; Asth, asthmatic subjects. *, p < 0.05; **, p < 0.01; ***, p < 0.001. The data were taken from [109].

8.2. 5-Oxo-ETE levels in body fluids and tissues

There is limited information available about the levels of 5-oxo-ETE in body fluids, possibly partly because it is metabolized, principally by ω-oxidation [58] and incorporation into cellular lipids [59], and also because it is not as well known as other 5-LO products such as the leukotrienes. It has been reported that the levels of 5-oxo-ETE in the brain are increased by nearly 50-fold following experimentally-induced cerebral ischemia in rats [111]. Increased levels of 5-oxo-ETE were also recently observed in extracellular vesicles isolated from the plasma of subjects with the autoimmune disease granulomatosis with polyangiitis and this was thought to be associated with increased production of reactive oxygen species and release of double-stranded DNA from neutrophils from these patients [112].

There is also evidence of elevated 5-oxo-ETE levels in various pulmonary diseases, including asthma. Murphy’s group found that the levels of 5-oxo-ETE in lung tissue from subjects with severe pulmonary hypertension were about three times higher than in control subjects, but lower than controls in hypertensive patients who had been treated with prostacyclin [113]. 5-Oxo-ETE has also been reported to be elevated in BAL fluid from female smokers with COPD but not in male smokers with COPD [114].

Kowal et al. recently measured the levels of eicosanoids in exhaled breath condensate from a group of asthmatic subjects who were sensitive to house dust mite allergen (HDM) [115]. HDM challenge resulted in increased levels of 5-oxo-ETE > LTD₄ > 8-iso-PGF2α in exhaled breath condensate of those subjects who exhibited a bronchoconstrictor response to allergen. Furthermore, the relative change in 5-oxo-ETE levels was positively correlated with the severity of the subsequent late asthmatic response. Increased levels of 5-oxo-ETE were also observed in BAL fluid following allergen challenge of HDM-sensitized mice [116]. 5-Oxo-ETE may also play a role in the development of nasal polyps, as it is formed by epithelial cells from this tissue and increases the levels of eosinophil cationic protein in organ cultures derived from nasal polyps [117]. These studies raise the possibility that 5-oxo-ETE may be an important mediator in eosinophilic diseases such as atopic dermatitis, allergic rhinitis, and asthma.

8.3. Pathophysiological roles of 5-oxo-ETE in animal disease models

Although elevated levels of 5-oxo-ETE have been detected in body fluids or tissues in various diseases or disease models as discussed above, such studies do not provide conclusive evidence for the involvement of 5-oxo-ETE or the OXE receptor in these conditions. To provide stronger evidence for a pathophysiological role for 5-oxo-ETE the use of receptor or enzyme knockdown or selective antagonists or enzyme inhibitors would be required. As noted above, this has been hampered by the lack of an ortholog of the OXE receptor in rodents. To date, only two published studies have addressed the direct involvement of 5-oxo-ETE in animal models of disease.

Enyedi et al. discovered an important role for 5-oxo-ETE in wound healing in zebrafish, which have a receptor, encoded by the oxer1/hcar1–4 gene, which has about 40% identity with the human OXE receptor [69]. Wounding of the tail fin of larval zebrafish in hypotonic medium results in the rapid recruitment of leukocytes as part of the wound healing process. This response was inhibited by depletion of cPLA₂ and restored by addition of AA, 5S-HETE or 5-oxo-ETE to the medium. Leukocyte recruitment was also inhibited by knocking down oxer1/hcar1–4 with a translation-blocking morpholino, which also blocked the response to 5-oxo-ETE added to the medium. In contrast, wound-induced leukocyte infiltration was not affected by inhibition or knockdown of LTA₄ hydrolase, the enzyme required for the formation of LTB₄. These results suggest that exposure of zebrafish cells to hypotonic medium as a result of wounding results in cell swelling and the cPLA₂-mediated release of AA, followed by its conversion to 5-oxo-ETE, resulting in the recruitment of leukocytes to the wound site.

Because of our interest in the potential role of 5-oxo-ETE in asthma and other allergic diseases we sought to test our OXE receptor antagonists in animal models. As noted above, rodent models were not possible due to the absence of the targeted receptor. We therefore initially investigated the possibility of using the cat, as this species possesses an ortholog of the OXE receptor and is particularly susceptible to asthma [118]. We found that BAL cells from asthmatic cats synthesize 5-oxo-ETE, which we also detected in BAL fluid [67]. 5-Oxo-ETE is a potent stimulator of actin polymerization in feline eosinophils and also induces leukocyte migration. However, compound 230 was several hundred-fold less potent in blocking activation of the feline OXE receptor compared to the human receptor. Similar results were subsequently obtained with the more potent antagonist S-C025. This is presumably due to differences between the two receptors, which differ by about 21% in their sequences.

8.4. Effects of OXE receptor antagonists on allergen-induced tissue eosinophilia in monkeys

Because of the low potency of our OXE receptor antagonists in the cat we decided to focus on the monkey, which has an OXE receptor with about 96% sequence identity with the human receptor. Monkey leukocytes synthesize 5-oxo-ETE, which is a potent stimulator of calcium mobilization, leukocyte migration, and actin polymerization in eosinophils from this species. Most importantly, OXE receptor antagonists block the activation of monkey eosinophils by 5-oxo-ETE with a potency almost identical to that observed with human cells [68].

To determine whether OXE receptor antagonists can inhibit allergen-induced eosinophilia we initially conducted a pilot experiment using racemic 230, one of our earlier antagonists, and wild-caught cynomolgus monkeys which had an acquired sensitivity to Ascaris suum through prior environmental exposure. Intradermal injection of Ascaris extract elicited eosinophil infiltration into the skin in all 3 animals tested, although the magnitude of the response was quite variable among individual monkeys. Administration of 230 (3 × 30 mg/kg; 8 h apart) by oral gavage reduced the numbers of eosinophils in the skin in all three animals tested, but this difference was not statistically significant because of the small number of animals and the high degree of inter-animal variability.

In an attempt to reduce the variability among monkeys, subsequent in vivo studies were carried out using captive-bred rhesus monkeys that had been experimentally sensitized to HDM, a well-established model developed at the California National Primate Research Center at the University of California, Davis [119]. These animals were sensitized by weekly subcutaneous injections of HDM followed by challenge with intradermal HDM after treatment with either vehicle or S-Y048 (2 x either 5 or 10 mg/kg), administered by nasogastric intubation [110]. The two doses of S-Y048 gave nearly identical results, inhibiting eosinophil infiltration into the skin in 5 out of the 6 animals in the study, with an overall inhibition of about 50%. This is the first evidence supporting a role for 5-oxo-ETE as a proinflammatory mediator in allergic responses.

To determine whether 5-oxo-ETE could also be involved in allergen-induced pulmonary inflammation we conducted a subsequent study using rhesus monkeys sensitized to HDM by a combination of subcutaneous injection as described above followed by biweekly exposure to aerosolized HDM (Cossette et al., manuscript in preparation). Following sensitization, animals were challenged with aerosolized HDM after administration of either S-Y048 or vehicle. S-Y048 significantly reduced the numbers of eosinophils, neutrophils and macrophages in BAL fluid 24 h after challenge as well as the numbers of eosinophils and neutrophils in lung tissue. Interestingly, the numbers of mucus-containing cells in the bronchi were also significantly reduced by S-Y048 as evaluated by staining with Alcian blue/periodic acid-Schiff and antibodies to MUC2 and MUC5AC.

9. Conclusions

5-Oxo-ETE is a 5-LO product formed from AA along with cys-LTs and LTB₄. However, its synthesis is more tightly controlled than leukotrienes, as it requires not only activation of cPLA₂ and 5-LO, but also elevation of the intracellular level of the obligatory 5-HEDH cofactor NADP⁺, which is normally present at very low concentrations in comparison to its reduced counterpart NADPH. Thus, conditions that promote the oxidation of NADPH to NADP⁺, including activation of the respiratory burst in phagocytic cells [26], oxidative stress [30], and cell death [27], all of which are associated with inflammation, favor the formation of 5-oxo-ETE (Fig. 6). This suggests that the synthesis of 5-oxo-ETE may be more prominent in prolonged inflammation, prior to resolution, rather than in the immediate acute response to cell stimulation, where other mediators such as PGD₂ (rapidly released from mast cells) and LTs may predominate. For example, we found that when neutrophils are stimulated with calcium ionophore they initially produce more LTB₄ than 5-oxo-ETE, but at longer time points (up to 24 h) the amounts of 5-oxo-ETE are much greater than those of LTB₄ [27]. 5-Oxo-ETE can stimulate GM-CSF synthesis by monocytes [54] and this may result in synergy between these two mediators, as GM-CSF can potentially increase the synthesis of 5-oxo-ETE by activating cPLA₂ [51] and increasing the expression of 5-LO and FLAP [52, 53], enhance the responsiveness of granulocytes to 5-oxo-ETE [50], and prolong eosinophil survival [120].

Figure 6. Regulation of synthesis and actions of 5-oxo-ETE.

The synthesis of 5-oxo-ETE requires activation of 5-LO as well as increased availability of the obligatory 5-HEDH cofactor NADP⁺ under conditions that are found in inflammation. 5-Oxo-ETE can promote inflammation by eliciting the tissue infiltration of inflammatory cells and also by stimulating the release of GM-CSF from monocytes, which could potentially result in positive feedback loops to increase the synthesis and response to 5-oxo-ETE.

Our recent studies with OXE receptor antagonists in monkeys provide the first evidence for a pathophysiological role for 5-oxo-ETE in mammals. The use of the monkey as an animal model has the advantage that the results of these studies are highly relevant to human disease. However, the high cost of experiments with monkeys severely restricts their scope, making it impossible to test a wide range of conditions. Thus, the range of antagonist doses that we were able to test was very limited, and it is possible that lower or higher doses might have been more effective. In the skin and lung studies described above the animals were exposed acutely to a single dose of allergen and the numbers of inflammatory cells were determined 24 h later. It would be interesting to examine the effects of repeated dosing with antagonist over a period of days or weeks in a chronic model of allergic inflammation.

Although the initial skin study discussed above focused on allergen-induced eosinophil infiltration, our unpublished findings on allergen-induced pulmonary inflammation suggest that 5-oxo-ETE may also target neutrophils and monocytes/macrophages, raising the possibility that it could play a role in endotypes of severe asthma displaying high numbers of neutrophils [121]. The significant inhibitory effects of S-Y048 on the numbers of mucin-containing cells in the bronchi were unanticipated and are quite interesting. Further work is clearly required to explain the mechanism for these effects and to determine whether 5-oxo-ETE acts directly on epithelial cells or indirectly, for example by stimulating the release of a cytokines or growth factor from inflammatory cells. For example, coculture of eosinophils with epithelial cells in the presence of GM-CSF and IL-3 has been reported to lead to an increase in the synthesis of MUC5AC by epithelial cells [122].

Targeting the OXE receptor with S-Y048 or related selective antagonists would be a novel therapeutic approach for the treatment of asthma and other allergic diseases. The effects of 5-oxo-ETE could also be prevented by using 5-LO or FLAP inhibitors, but this approach would not be specific. Although it would benefit from the concomitant inhibition of cysLT synthesis, the formation of LTB₄ and LXA₄ would also be prevented, which may not be beneficial because of the role of the former in host defense [94] and the latter in the resolution of inflammation [123]. OXE receptor antagonists could be used either on their own or in combination with current asthma medication such as glucocorticoids or cysLT₁ antagonists, which act by different mechanisms, thus potentially resulting in synergy between these drugs.

Abbreviations

S-C025

(S)-5-(5-chloro-1-methyl-2-(6-phenylhexyl)-1H-indol-3-yl)-3-methyl-5-oxopentanoic acid

5(S)-HETE

5S-hydroxy-6E,8Z,11Z,14Z-eicosatetraenoic acid

5(S)-HpETE

5S-hydroperoxy-6E,8Z,11Z,14Z-eicosatetraenoic acid

PGD₂

prostaglandin D₂

5-oxo-15S-HETE

5-oxo-15S-hydroxy-6E,8Z,11Z,13E-eicosatetraenoic acid

5-oxo-ETE

5-oxo-6E,8Z,11Z,14Z-eicosatetraenoic acid

LT

Leukotriene

S-230

(S)-5-(5-chloro-2-hexyl-1-methyl-1H-indol-3-yl)-3-methyl-5-oxopentanoic acid

S-Y048

(S)-5-(5-chloro-2-(6-(3-chlorophenyl)hexyl)-1-methyl-1H-indol-3-yl)-3-methyl-5-oxopentanoic acid

AA

arachidonic acid

OXE receptor

oxoeicosanoid receptor

LXA₄

lipoxin A₄

5-HEDH

5-hydroxyeicosanoid dehydrogenase

PAF

platelet-activating factor

12S-HETE

12S-hydroxy-5Z,8Z,10E,14Z-eicosatetraenoic acid

PKC

protein kinase C

5-oxo-12S-HETE

5-oxo-12S-hydroxy-6E,8Z,10E,14Z-eicosatetraenoic acid

5-oxo-ETrE

5-oxo-6E,8Z,11Z-eicosatrienoic acid

EPA

5Z,8Z,11Z,14Z,17Z-eicosapentaenoic acid

DHA

4Z,7Z,10Z,13Z,16Z,19Z-docosahexaenoic acid

HDM

house dust mite allergen

BAL

bronchoalveolar lavage

S-264

(S)-5-(6-chloro-2-hexyl-1H-indol-1-yl)-3-methyl-5-oxopentanoic acid